PROCEEDINGS OF THE THIRTY'FOURTH ANNUAL CONGKESS...proceedings of the thirty'fourth annual congkess i960 of the south african sugar technologists' association . the south african ...

PROCEEDINGS OF THE

THIRTY'FOURTH ANNUAL CONGKESS

i 9 6 0 OF THE

SOUTH AFRICAN SUGAR TECHNOLOGISTS'

ASSOCIATION

THE SOUTH AFRICAN SUGAR TECHNOLOGISTS' A S S O C I A T I O N

The South African Sugar Technologists' Association was

founded in 1926. It is an organisation of technical

workers and others directly interested in the technical

aspect of the South African Sugar Industry. It operates

under the aegis of the South African Sugar Association,

but is governed under its own constitution by a Council

elected by its members.

The office of the Association is situated on premises

kindly made available to it by the South African Sugar

Association at the tatter's Experiment Station at Mount

Edgecombe.

in

PRINCIPAL CONTENTS THE SOUTH AFRICAN SUGAR TECHNOLOGISTS' ASSOCIATION iii

OFFICERS OF THE SOUTH AFRICAN SUGAR TECHNOLOGISTS' ASSOCIATION . . v

LIST OF MEMBERS AND GUESTS vi

OPENING ADDRESS BY MR. C. J. SAUNDERS, Chairman of the S.A. Sugar Association 1 PRESIDENT'S ADDRESS . . . . . . . . . . . 4 THIRTY-FIFTH ANNUAL SUMMARY OF CHEMICAL LABORATORY REPORTS, by Chs.

G. M. PERK 7

WEATHER REPORT FOR THE YEAR 1ST JUNE 1959 TO 31ST MAY 1960, by J. L. du Toit 25 ANNUAL SUMMARY OF AGRICULTURAL DATA FOR THE SUGARCANE CROP, 1957-58,

by.J. L . d u Toit and K . E . F . Alexander . . . . . . . 3 0 RIVER WATERS, by D. W. W. Hendry 39 MILLING AND OVERALL PERFORMANCES, by Th. Fourmond . . . . . 41

A DESCRIPTION OF THE CONVEYOR BELT SYSTEM AT UMFOLOSI MILL, by G. G. Ashe 45 MILLING CONTROL DATA WITH REFERENCE TO A MORE INTENSIVE METHOD OF

SAMPLING AND ANALYSIS, compiled by the Technical Staff at Z.S.M. & P. Ltd., and presented b y E . H . Phipson . . . . . . . . 5 0

MINIMISATION OF THE HUMAN ELEMENT IN MILLING, by D. J. L. Hulett. . . 57 A RECORDING ROLLER LIFT INDICATOR, by A. van Hengel . . . . 63 DRAWING A STRAIGHT LINE THROUGH POINTS ON A GRAPH, by W. 0. Christianson . 67 MIXED JUICE SCREENING AT DOORNKOP, by V. S. Winterton . . . . 70

A NEW ION EXCHANGE PROCESS FOR REMOVING COLOUR BODIES AND COLLOIDAL

IMPURITIES FROM CANE SUGAR, by R. A. Grant . . . . . . 74 SOME OBSERVATIONS MADE ON THE REHEATING OF MASSECUITE IN A CRYSTALLIZER,

by W. F. Davies 78 DROP IN PURITY BETWEEN MASSECUITE AND MOLASSES, by A. D. Elysee and

L. E. Turner 82 STARCH IN THE MANUFACTURE OF RAW SUGAR, by P. N. Boyes . . . . 91 SOME NOTES ON THE MELT CARBONATATION REFINERY OF REYNOLDS BROS. LTD.,

AT SEZELA, by W. G. Galbraith. and E. Dedekind . . . . . . 98 THE REFINING PROCESS AT ILLOVO, by E. Beesley . . . . . . 104 SOME NOTES ON THE NEW MELT CLARIFICATION AT HULETT'S REFINERIES, by

J. B. Alexander 107 A SHORT DESCRIPTION OF GLEDHOW REFINERY, by L. F. Chiazzari . . . 112

NOTES REGARDING THE SULPHITATION-REFINING PROCESS AND PLANT AT UMFOLOSI,

b y J . D . Thumann . . . . . . . . . . . 114 THE JUICE CARBONATATION PROCESS AND REPERCUSSIONS OF ECONOMICS ON

TECHNOLOGY, by J. Rault 120 SYSTEMATIC PLANNING AND SCIENTIFIC CONTROL OF FIELD OPERATIONS, by G. C.

Sheppard . . . . . . . . . . . . . 128 NATAL SUGARCANE VARIETIES—SOME OBSERVATIONS AND STATISTICS, by Exley

Steward 134 YIELD DATA FOR EXPERIMENTS HARVESTED AT ILLOVO, 1958-59, by C G. Halse

and G . D . Thompson . . . . . . . . . . 141 SOIL CONSERVATION IN SUGARCANE FIELDS, by T. G. Archibald . . . . 151

THE COMPACTION OF SUGAR-BELT SOILS AT VARIOUS MOISTURE LEVELS, by R. R.

Maud . . . . . . . . . . . . . 154 IRRIGATION CONTROL AND EXPERIMENTATION AT ILLOVO, by G. D. Thompson . 161

SOME NOTES ON IRRIGATION, by T. G. Cleasby 170

VARIETY AND ENVIRONMENT, by P. G. C. Brett . . . . . . 1 7 6 SOME ITEMS OF ECONOMIC IMPORTANCE IN SUGARCANE PRODUCTION, by J. L. du Toit 183 THE DEVELOPMENT OF THE MECHANISATION COMMITTEE OF THE SOUTH AFRICAN

SUGAR ASSOCIATION, by George S. Bartlett 189 INSTRUCTIONS TO AUTHORS . . . . . . . . . . 1 9 6

iv

OFFICERS 1960-1961

Life Patron

W. A. CAMPBELL

President Vice-President

J. L. DU TOIT J. DICK

Hon. Secretary Hon. Technical Secretary

(Mrs.) N. BOYD-SMITH W. 0. CHRISTIANSON

Former Presidents

1926-27 M. MCMASTER

1927-28 M. MCMASTER

1928-29 H. H. DODDS

1929-30 H. H. DODDS

1930-31 G. S. MOBERLY

1931-32 G. C. DYMOND


1933-34 B. E. D. PEARCE

1934-35 E. CAMDEN-SMITH

1935-36 G. C. WILSON

1936-37 G. C. WILSON

1926-27 L. E. ROUILLARD

1927-28 H. H. DODDS

1928-29 G. S. MOBERLEY

1929-30 G. S. MOBERLEY


1931-32 A. C. WATSON

1932-33 A. C. WATSON

1934-35 B. E. D. PEARCE

1935-36 E. CAMDEN-SMITH

1936-37 J. RAULT

1937-38 J. RAULT

1938-39 P. MURRAY

1949-40 P. MURRAY

1940-41 E. P. HEDLEY

1941-42 F. W. HAYES

1942-43 A. MCMARTIN

1943-44 G. BOOTH

1944-45 G. S.MOBERLY

1945-46 G. S. MOBERLY

1946-47 W. BUCHANAN

1947-48 W. BUCHANAN

1948-49 J. L. DU TOIT

1949-50 H. H. DODDS

1950-51 A. MCMARTIN





1955-56 J. B. GRANT

1956-57 J. B. GRANT

1957-58 J. P. N. BENTLEY



Former Vice-Presidents

1937-38 P. MURRAY



1940-41 F. W. HAYES

1941-42 A. MCMARTIN

1942-43 G. BOOTH

1943-44 F. B. MACBETH

1944-45 G. BOOTH

1945-46 W. BUCHANAN




1949-50 J. L. DU TOIT

1950-51 O. W. M. PEARCE

1951-52 O. W. M. PEARCE

1952-53 K. DOUWES-DEKKER

1953-54 J. B. GRANT

1954-55 K. DOUWES-DEKKER

1956-57 W. G. GALBRAITH

1957-58 J. L. DU TOIT

1958-59 J. L. DU TOIT

1959-60 J. L. Du TOIT

J . B. ALEXANDER

J. P. N. BENTLEY

W. J. G. BARNES

L. F. CHIAZZARI

W. O. CHRISTIANSON

A. DE BROGLIO

J. DICK

K. DOUWES-DEKKER

J. L. DU TOIT

W. G. GALBRAITI

J. B. GRANT

J. R. GUNN

J. RAULT

C VAN DER POL

J. WILSON

v

Council of the Association

South African Sugar Technologists' Association

Thirty-fourth Annual Conference

The Thirty-fourth Annual Congress of the South African Sugar Technologists' Association was held at the M.O.T.H. Memorial Centre, Old Fort Road, Durban, on the 28th March, and at the Shell Cinema on the 29th, 30th, 31st March, and

1st April, 1960.

The following Members and Visitors were present:

J. P. N. BENTLEY (President) was in the Chair.

ADDISON, L. P. AITKEN, E.

ALEXANDER, H. P. Al-EXANDKR, J. B. ALEXANDER, J. T.

ALEXANDER, K. E. F.

ALLSOPP, J. G.

ALLSOPP, L.

ALMOND, F. L.

ANDERSON, J. R.

ANDERSON, R. V.

ANTONOWITZ, A.

ARCHIBALD, T. G.

ARMSTRONG, M. G.

ASHE, G. G.

BAILEY, E. G.

BAIRD, Mrs. J.

BARNES, A. C.

BARNES, Mrs. M. M.

BARNES, W. J. G.

BARTLETT, G. S.

BATCHELOR, W. W.

BEATER, B. E.

BEESLEY, E.

BENTLEY, J. P. N.

BLACK, W. B.

BLENNER-HASSETT, D. N.

BONFA, N. J. A.

BOOTH, C.

BORCHERS, 0.

BOULE, P. F.

BOURNE, J. H.

BOWLES, Miss R. BOYES, P. N.

BOYD-SMITH, Mrs. N.

BRADFORD, I). \V.

BRASSEY, T. B.

BRETT, P. G. C.

BROGLIO, de A.

BROUARD, T.

BROOK, F. E. W.

BROWN, J. \V.

BUCHANAN, E. J.

BUDDENBROOK, R. von

CAMDEN-SMITH, E.

CAMPBELL, W. A.

CARGILL, J. M.

CARTER, R. A.

CHIAZZART, L. F.

CHIKHOLM, P. R.

CHRLSTIANSON, W. O.

CLEASBY, T. G.

CLUTTERBUCK, N. S.

COLLINS, A. P.

Cox, R. B. CURRIN, R. L.

DAMANT, E. L.

DAVIES, W. F.

DEDEKIND, E. T. J.

DELGADO, N.

DENT, C E.

DICK, Mrs. B.

DICK, J.

DICK, J. McD.

DOUGLAS, W. E. O.

DODDS, Mrs. E. A.

DODDS, H. H.

DOUWES DEKKER, K.

DUCASSE, F.

DUNTON, D. C.

DUNN, G. A.

DU TOIT, J. L.

DYMOND, K.

EASTMAN, Miss R. J. EDDINGS, E.

EDWARD, J. C.

ENGLISH, K. R.

ELYSEE, A. D.

FABIEN, R.

FARQUHARSON, J. ('.

FENWICK, j . H .

I 'ERGUSON, D.

FLEWELLEN, G. C.

FOURMOND, Th.

FRANCIS, V.

FRANGS, G. B.

FROBERVILLE, P. de

FROESLER, H. P.

FROST, R. K.

GAINLEY, I.

GARZ, F. C.

GIRDLER, J.

GRAHAM, W. S.

GRANT, J. B.

GRANT, R. A.

GRICE, L. D. C.

GROOM, W. A.

GUNN, J. R.

HALSE, C. G.

HALSE, R. H.

HAMLYN, D. M. A.

HARCOURT-BALDWIN, J. L.

HARRIS, V. G.

HEDGCOCK, G. W.

HEDLEY, E. P.

HEMPSON, P.

HEMPSON, W. J.

HENDRY, D. W. W.

HESLOP, W. B.

HOWSON, G. W.

HULETT, D. HULETT, R. L.

HUMFRYES, A. G.

HUNTER, J. W.

JOHNSTON, F. S.

JOHNSTONE, P.

KARLSON, M. A.

KING, N. C.

KRAMER, F. A.

LAAPER, J. N.

LANGE, E. de

LARGE, S. M.

LARSEN, A.

LAX, R. LEE, G. L E E , Mrs. S. LINDEMANN, W. C.

LINCOLN, M. A.

LINTON, G.

LLOYD, A. A.

LOUDON, T. R.

MACGREGOR, W. I.

MAIN, J. W.

MANDY, H. G.

MARAIS, Miss C. P. MASSON, F.

MAUD, R. R.

MAYNE, P.

MCGRATH, G. D.

MCIVER, N. F.

McKlNLAY, D. MCMARTIN, A.

MOORE, J. B.

MORRISON, E.

MUNGLE, J.

MUNRO, D. B. B.

MUNRO, D. J.

MURRAY, L. C.

NAPIER, J.

NEILSON, W.

NICKSON, G.

NOEL, R.

ODENDAAL, G. A.

OOSTHUIS, L. L.

PALMER, H.

PATERSON, A.

PEARSON, C. H. O.

PERK, C. G. M.

PFOTENHAUER, V. O.

PHIPSON, E. H.

PILGRIM, A. K.

PICCIONE, J. PORTEOUS, J.

POUGNET, J. H. POUSSON, J. B.

PRINGLE, D. H.

vi

RABE, A. E.

RAMSEOTTOM, A. T. S.

RAULT, J.

RAUTENBACH, S.

RENAUD, C. L.

RlCQUEBOURG, P. M.

RISHWORTH, A. G.

RIVIERE, M. P.

ROBARTS, W. E.

ROBBINS, W. T.

ROBERTSON, J. M.

ROBILLARD, M. de

Roos, R. de

Ross, D. C.

ROSTRON, P. M. F.

ROUTLEDGE, D. A.

RUDD, C. M.

RUSSELL, G. D. N.

SANDSTROM, G. R. A.

SAUNDERS, C. J.

SAUNDERS, Mrs. P.

SARGENT, N. V.

SCHLEISS, D.

SEYMOUR, G. E.

SEYMOUR, R. W.

SEXTON, T. A. F.

SHARVELL, N. A. D.

SHEPPARD, G. C.

SHERRARD, C.. D.

SHERRELL, R.

SHORTEN, C. J.

SIMPSON, R. M. O.

SINGERY, L. C.

SMEATON, I. G. B.

SMITH, L. L.

SOUCHON, A. C. H.

SOUCHON, C. L. A.

SOUCHON. L. A. M.

SOUCHON, R. M. T.

STEPHENSON, R. A.

STEWARD, E.

STEWARD, K. M.

STEWARD, Mrs. J.

STEWART. M. J.

STEYN, C. L.

STRONG, S. R.

TAYFIELD, D. J.

TEDDER, G. W.

TERNENT, M. J.

THOMPSON, G. D.

THOMSON, G. M.

THUMANN, J. D.

TODD, W.

TORR, Mias M.

TOURNOIS, P.

TURNER, L. E.

TURNER, Q. A.

UDAL, W. N.

VAN DONGEN, D. H.

VAN HENGEL, A.

VAN DER POL, A. VAN NlEKERK, G. J.

vii

WAGNER, C. L.

WALSH, G. H.

WALSH, W. H.

WARD, J. M. J.

WARNE, D. E.

WARREN, F. L.

WEIGHTMAN, E. B.

WELLS, Mrs. M.

WESSELS, M. H.

WHEELER, F. D.

WILLIAMS, A. R.

WILSON, G. C.

WILSON, J.

WILSTON, B. T.

WOOD, G. B.

W H I T E , T. O W E N

WRIGHT, Mrs. B. F.

WEBSTER, I. McC.

YOUNG, C. M.

YOUNG-THOMPSON, I. C.

OPENING CEREMONY The President: Ladies and Gentlemen—I have

great pleasure in asking Mr. C. J. Saunders, Chairman of the South African Sugar Association, to open our Forty-fourth Congress.

Mr. Saunders: Ladies and Gentlemen—This is an auspicious occasion, for I realise that gathered here today is an association of people whose knowledge of problems relative to sugar totals an enormous sum—a store of experience, perhaps unequalled by any single industrial enterprise in this country. I am proud and honoured to speak at this Conference, and in addressing you I fully realise and appreciate that I am speaking to those who hold the key to the future success of the Sugar Industry in South Africa. Certainly you, the leaders of technological and industrial advancement, have demonstrated time and again the qualities necessary for successful battle. In the face of perplexing difficulties you have risen to undreamed-of heights, and achieved all tha t has been asked of you by the leaders of the Industry.

Time without number in the past, you have been exhorted to produce more sugar, and I want to congratulate the technologists on the way that they have listened to those honourable gentlemen who have asked for an annual production of more than 1,000,000 tons, who have demanded more sugar in the bag. I earnestly hope that you will listen and pay as much attention to what I have to say! My subject, however, is nowhere near as palatable as some of my immediate predecessors, for my theme is governed by the imminent threat of controlled production.

Improved Methods Under restricted conditions of manufacture and

production it is more essential than at any stage in the past 15 years that you must now turn your attention to reducing unit costs of production and obtaining the highest possible return from a limited market. This the Industry can and must do by improving its methods and making better use of its

resources. The warning which I sound and shall repeat is that the future success of the Industry, both for itself and the country, depends now primarily on the extent to which it can improve its profitability by increased internal efficiencies. No longer can standards be measured by volume of output alone.

In the world today where the production of sugar both from cane and beet is on the increase, the marketing of the final product becomes not only more and more difficult, but of tremendous importance. I do not believe it is necessary to emphasise the obvious vital factors of quality and saleability of output. Rather would I discuss" our markets and eventually return to matters concerning future domestic policy.

Of the total world annual production of nearly 50,000,000 long tons of sugar only some 5,000,000 to 6,000,000 tons is bought and sold at the so-called "world free market price". Approximately nine-tenths of the world sugar, therefore, is sold either in domestic markets or at specially negotiated prices under international contracts of a long-term nature. Cuba sells a large portion of her crop to the United States of America at a price in the vicinity of the negotiated price paid by the United Kingdom to Commonwealth countries. The balance of the Cuban production is sold on the world free market at whatever price it will fetch, and this exercises a tremendous and frightening influence on what is known as the free market price of sugar—a price that is in no way related to the costs of production.

There are few, if any, sugar industries which could exist on the world free market price of sugar alone. This is a fact, Mr. President, which seems not only to be misunderstood by the consumer, but misconstrued by the Press and misinterpreted by higher authorities in this country.

The bulk of South Africa's exports are sold to the United Kingdom under the Commonwealth Sugar Agreement. This is an agreement between the Commonwealth producing countries and the United

THIRTY-FOURTH ANNUAL CONGRESS

Proceedings of the Thirty-fourth Annual Congress of the South African Sugar Technologists' Association, held at the M.O.T.H. Memorial Centre, Old Fort Road, Durban, on the 28th March, and at the Shell Cinema on the 29th, 30th,

31st March and 1st April, 1960.

J. P. N. BENTLEY (President) was in the Chair

Kingdom Government. Further, these Commonwealth quotas are irreducible and are entrenched in the International Sugar Agreement to which the South African Government is a party.

At this stage I wish to dwell on the actual make-up of the Commonwealth Sugar Agreement which gives South Africa an overall quota of 224,000 short tons, of which this year 177,240 short tons fall under what is known as the negotiated price quota. The price for 1960 is £39 Us. 7d. per short ton, and this, of course, is for raw sugar on a 96° polarisation basis. It is a price which is intended to give a reasonable return to efficient producers. This price represents approximately £12 above the local market price which we receive in South Africa. Without the solid foundations which the Sugar Industry has established over 100 years and which have been cemented by firm Government policy and sound agreements this difference in the level of price could not exist.

The negotiated price is fixed in London each year at discussions between officials of the United Kingdom Ministry of Agriculture, Fisheries and Food and representatives of the Commonwealth sugar producing countries. The remainder or balance of the Commonwealth quota—that is the difference between the 224,000 short tons and 177,240 short t o n s -carries a preference of £3 15s. Od. a ton over and above the free market price when sales are made to either the United Kingdom or Canada.

Under the International Sugar Agreement the Commonwealth has an additional allocation of quota of 224,000 short tons for the years 1960 and 1961, which may be sold only in non-preferential markets. South Africa's pro rata share of this quota is 18,931 short tons.

By a matter of simple addition, South Africa's total exports covered by Commonwealth and International Agreements amounts to 242,931 short tons, which you will appreciate is a significant proportion of our total production. These quotas amount to approximately £9,000,000 in revenue, nearly all of which comes from the United Kingdom. We in South Africa have tended in the early post-war years of short crops to regard our domestic market as the foundation stone of our economy. Now in the years of abundance and expanding production we must look to our International and Commonwealth Quota Agreements as the cornerstone of our whole economic, structure.

Definite Obligations Our overseas markets are of inestimable value,

but, and I wish to emphasise this point, the benefits derived from the Commonwealth Sugar Agreement, naturally impose very definite obligations on its members, the chief of which is to ensure that sugar

is produced at the cheapest possible cost—in other words, maximum efficiency.

Perhaps I might illustrate my point specifically by referring to one of our most important critics— the beet farmers of the United Kingdom, and let me remind you that then; are 40,000 of them. Taking full advantage of the scientific march of progress which has characterised agriculture and industry since the war, their industry has emerged from a state of eonomic poverty and embarrassment to become the strongest challenger to the cane producers of the Commonwealth. It is only natural that the owners of a factory such as Wissington, which because of automation, is able to produce 340 tons of sugar per day with a total labour force in all departments of 184 units and a shift force of 27 units, must look with natural envy at an arrangement which guarantees a profit margin, formerly fixed at 29 per cent, of costs, and now adjusted to nearly 21 per cent—but with depreciation provided on all field and factory assets at full replacement value.

Another factor is that United Kingdom farmers are expected by their own Government to increase in annual efficiency at the rate of two per cent in gross turnover. Therefore, we as a Commonwealth producer must ensure that sugar is produced at the cheapest possible cost. It is not sufficient to compare our technical efficiency with other world sugar industries. We must strive for economic efficiency and this is the duty of industrial management and farmers alike. Technical efficiency expressed in terms of sugar per acre harvested, and which covers field and factory practices, must now only be viewed in the light of falling costs per ton. If this argument applies to the cornerstone of our economic structure, what of our foundation stone—our domestic market?

The Industry's output has continued to expand. The future, therefore, must be assessed in the light of prospective demands for sugar. If present trends continue demand is likely to increase at a lesser rate than the supply of the product, for there has been a definite levelling off in consumption over the last three years, and at present the increased offtake of sugar in the domestic market is in no greater proportion than the increase in population of the country, further, it would appear that consumption levels per capita in South Africa have reached very near saturation point, particularly when it is considered that as a consuming nation on an individual basis we are one of the highest in the world, in spite of a large proportion of the people who fall into the very low income category.

High Consumption South Africa's per capita consumption figure is

extremely difficult to assess, but it is believed that the European per capita consumption is 121 lb.

per annum and the non-European 84.5. This is certainly surprisingly high in comparison with almost all Western European nations, where taking France as the example, the per capita consumption is only 58 lb. per annum and countries such as Portugal and Italy fall between 40 and 50 lb. per annum. Australia, which has the highest sugar consuming public in the world, has a per capita figure of 121 lb.

It is imperative that technologists turn their attention to reducing unit costs of production, otherwise as the tendency for supplies to outstrip demand continues, the maintenance of reasonable standards for those engaged in the Industry will not be possible without increasing the burden on the consumer.

What is required is a steady and continuing improvement not only in farm methods but also in management and operational control. In general, there is room to better results by the more economical use of labour and machinery, better equipped management and a more thoroughly trained and, therefore more competent, labour force. I do not propose to tell you how to accomplish the task, but merely ask you to examine the position.

I repeat that the future success of the Industry now depends primarily on the extent to which it can improve its profitability by reducing unit costs of production. Success cannot in the future be measured by the volume of output alone.

I do not wish to detain you any longer, Mr. President, and I hope that your deliberations will prove of great benefit to the Industry and as always be not only of extreme interest but of much importance to all concerned. I now declare your conference open.

Mr. J. Rault: Mr. President, Ladies and Gentlemen—when at the last council meeting of our Technologists' Association I informed you, Mr. President, that I proposed to read at the coming Congress a paper which would be my swansong, in the nature of a funeral oration praising the achievements of the Juice Carbonatation Process, and deploring the repercussions of Economics on Technology, you entrusted me with the further honourable task of proposing a vote of thanks to the guest speaker opening our Congress.

Little did I realise at that time, and even last night, that I should have this morning the privilege of listening to an amplified version of my original theme, orchestrated and supported by inside information, but elevated to the height of a discourse on the economics of the whole South African sugar industry, in its relations to the national and international situation.

It is my pleasant duty before pronouncing my "Nuc dimittis", to answer this illuminating, though grave address, to our Technologists' Association by a

brilliant, if young, member of a family known for its long and devoted association with the history and progress of the sugar industry.

Mr. Saunders has fittingly spoken today as the chairman of the S.A. Sugar Association, but he shares amongst his other responsibilities the honour of being the Managing Director of a firm reputed for its efficiency, its successful innovations, and its three years successive records of bulk production amongst all South African factories.

Mr. Saunders' appeal to the Technologists today is to think not so much and only on maximum volume of production, but in terms of minimum cost per unit of production, by directing our activities to economy of labour and material, and a better use of them.

This line of thought may sound somewhat of a paradox to our ears, attuned to the ancient dogma accepted as an axiom, that increased or massed production is one of the surest means of reducing per cent cost per unit, at least in what concerns overhead charges.

The agriculturist attempting to grow two blades of the giant grass, sugarcane, on the soil where previously grew only one; the engineer with a standard equipment and a stable labour force determined to crush consistently at higher capacities and higher mill extraction; the production manager by a superior technique putting in bags 1 or 2 per cent more brilliant crystals from the same amount of sucrose given by the mills, all feel alike, that they have contributed to reduce the cost of production without having to put limitations on the wealth extractible from mother Earth, or still worse for the stability of society by having recourse to labour dismissals and the curse of demoralising unemployment.

In placing before us this unpleasant dilemma of a brake on the impetus of production, and a reduction in cost per unit, you have, Sir, paid a great compliment to our Association, and perhaps unwittingly raised the status of the Technologist to the position of a trustee, on whom devolves the uneasy task of reconciling in a harmonious synthesis, some awkward and conflicting elements.

He is expected to solve the problems of helping capital to find a profitable return on its accumulated wealth, guarantee to the farmer a fair market for his agricultural products, labour in its widest sense an adequate wage to maintain a high standard of living, the consuming public a food-stuff of standard purity, abundant and reasonably cheap enough to be accessible to the masses.

We Technologists, will cheerfully accept your trust, which is also a challenge for further efforts, at this moment of overhanging clouds on the South African horizon amidst the rumblings of social and

3

racial adjustments. We pray God that this "wind of change" may not blow with the cyclonic violence that has lately laid low the flourishing industry of our Mauritian neighbours, but rather turn out to be the expected monsoon bringing in abundance the gentle rain from heaven, benefiting a whole continent.

In the anxious years ahead of us, our confidence will be strengthened by the thought that we can rely on leaders of your calibre.

In common with a famous statesman of the Napoleonic era, William Pitt the second, he shares "the atrocious crime of being a young man", and a charming one at that. This is why in a spirit of affection, more than familiarity we all call him, not Mr., but Chris, Saunders and I ask you all to accord him a hearty vote of thanks.

THE PRESIDENT'S ADDRESS Mr. J. P. N. Bentley (President): Ladies and

Gentlemen—in the past the work done by our Association has often formed the subject of the Presidential Address, but I make no apology for speaking to you again on this theme, as recent Press statements have spotlighted an aspect of our organisation which requires our earnest attention if we are to avoid drifting into a position which becomes more and more difficult to retrieve as the years go by.

At the outset I would like to say that I will endeavour to present to you only the facts of the situation as I see them, and if at any time it appears that my remarks are critical of those in authority I trust tha t they will be accepted as a sincere effort to rectify an aspect of our Association within the world organisation which, at present, prevents us from playing our full part in the advance of sugar technology.

May I crave your indulgence for a few moments while we review the past history of our Association. Many of you will recall that the first congress of this nature was held in 1923 under the auspices of the South African Sugar Association. This sugar week continued until 1927, when a separate technical society was formed as the result of efforts of such men as Mr. Patrick, Mr. David Fowler, Mr. Douglas Saunders, Dr. H. H. Dodds, Mr. G. S. Moberly, the late Mr. G. C. Dymond and Mr. W. A. Campbell, who still takes a most active interest in the welfare of our Association. The International Society of Sugar Cane Technologists was formed in 1924, only three years before the formation of the South African Sugar Technologists' Association.

Aware of Problems Since 1927 the South African Sugar Technologists'

Association has held its annual Congress in Durban and technical problems connected with the growing

of cane and production of sugar have been freely discussed and carefully recorded. To the outside public it may appear that Saccharum Spontaneum can hardly justify such lengthy debate, but those within the Industry are well aware of the infinite number of problems that arise in an industrial undertaking such as ours, where a continual increase in efficiency is necessary to meet overseas competition and rising production costs.

The circulation of weekly and monthly factory results and the publication of the annual summary of factory results highlights the work done at each factory and engenders a healthy competitive spirit among factory personnel. Papers read at our congress and the discussions that follow are published in the Annual Proceedings which form a valuable reference for sugar technologists, and copies of these proceedings are sent to interested organisations of a similar nature in many parts of the world.

In addition, the Council of our Association controls a number of standing committees, which from time to time devote their attention to particular problems that arise. The importance of these committees can be gauged by the outstanding success achieved by the Chemical Control Committee, whose recommended methods for factory control are accepted throughout the Industry as a yardstick for the measurement of factory efficiency, and the work done by the Education Committee, which has once again blueprinted a training scheme for sugar technologists. Other committees such as the Editorial Committee work quietly in the background ensuring the maintenance of the high standards that have been set in the past.

Furthermore, field and factory days are arranged at regular intervals, so that members may have the opportunity of studying new developments in the Industry and discuss among themselves the various problems with which they are confronted.

Our Association forms a regional division of the International Society of Sugar Cane Technologists and under the constitution of the International Society membership is open to individuals only through the regional division in that area. Where no regional division exists, membership may be obtained by direct application, but this is usually only a temporary means of obtaining membership until such time as a regional division is formed. The International Society holds a congress of three weeks duration every three years. The first International meeting took place in Hawaii in 1924 and 35 years later, in 1959, Hawaii was again the host country for the Tenth International Congress.

In between, the International Society met in Cuba in 1927, Java 1929, Puerto Rico 1932, Australia 1935, Louisiana 1938, Australia 1950, the British West Indies 1953 and India 1956. The Tenth Inter-

national Congress in Hawaii was attended by over 300 delegates, including ten from South Africa. Others came from all corners of the globe, including Jamaica, Mauritius, the Phillipines, Venezuela, Queensland, Brazil, Cuba, Uganda, Kenya, Fiji, New South Wales, Puerto Rico, India, Reunion, Dominican Republic, Egypt, Nicaragua, Taiwan, France, the United Kingdom, the United States of America and Mexico. It was hoped that Russia would send a delegation, but though they contri buted to the technical papers read at the Congress, no Russians arrived. Commenting on the Congress, the Editor of the Sugar Journal of America says "The most significant thing about the entire con ference was the congeniality and friendliness that existed between delegates, who represented all races and all religions . . . delegates were quick to respond to the common bond and harmony existing between sugar men the world over. The cordiality which rapidly developed between men of different social, racial and religious backgrounds was proof that people are the same the world over, when given the opportunity to show it."

As your Chairman, I was privileged to attend the meetings of the Executive Committee of the International Society whose function was, among others, to decide where the congress in 1962 should be held. Four countries competed for the privilege of being hosts to the International Society and Mauritius finally secured the majority vote.

Many Factors Involved An invitation to act as hosts to the International

Society cannot be issued without serious consideration of all the factors involved. The costs of such a visit can be considerable, suitable accommodation and transport facilities must be fully investigated, the publishing of the proceedings is the responsibility of the host country and even this apparently simple task can be a nightmare for typists trying to transcribe from tape recordings discussions on technical papers spoken in what passes for English in different parts of the world.

Nevertheless, we are pleased that our near neighbour should be thus honoured in 1962, but the very proximity of this International gathering must surely make us think again of the possibility of some such meeting in Natal. I have been fortunate in that I have twice had the opportunity of world tours of the cane sugar producing areas and I can assure you, ladies and gentlemen, that there are many aspects of our Industry from which an International delegation could learn some, lessons and could copy to their advantage. For example our self-loading trailers and other ingenious cane handling equipment would surprise many delegates, including those from highly mechanised countries; our irrigation schemes and their control must rank amongst

the best; our management methods, our research stations, our scientists and our technologists could all meet the test of an International visit.

Reply to Press Report

At this stage I must say how greatly we appreciate the assistance that the South African Sugar Association has always given us. Without their generous financial support it would not have been possible for our members to play the active part tha t they have done in International sugar affairs. I must , however, in all fairness deny a recent Press s ta te ment that our Association has at any time been given a firm "no" by the Government to our request to hold an International congress here. We have in fact never made such a request to the Government, and I am convinced that the Government is fully aware of the need to foster as much as possible the progress of scientific societies in South Africa.

In the continent of Africa our Association is at the moment in a position where it can take the lead in all matters of sugar technology. Our reputat ion stands firmly on three main foundations, firstly the part played by our members at International meetings and the standard of scientific papers published in our annual proceedings, secondly the work done by the Experiment Station and the active part the Experiment Station staff have taken in International sugar affairs and thirdly the growing influence t h a t the Sugar Milling Research Institute is exerting, both inside and outside South Africa.

Seven Enterprises Affiliated Already, with encouragement and approval at

Government level, seven sugar enterprises in Africa have become affiiliated to our Research Inst i tute . These include Wonji Sugar Estate in Addis Ababa, Societa Agricola near Mogadiscio, Italian Somali-land, Sena Sugar Estates and Colonial do Buzi in Portuguese East Africa, Arusha Chini Estates in Tanganyika and Ubombo Ranches and Mhlume in Swaziland. From time to time our scientists are called upon to visit these territories and advise their technical personnel; and their technologists are in turn showing a growing interest in a n d a desire to see for themselves how our fields and factories are operated. It is essential that the interchange of technical personnel be devoid of all restrictions if we are to continue playing the leading role in the advance of sugar technology in Africa.

In this brief review I have tried to show how in 33 years our Association has grown in stature from the time when only the Director of the Experiment Station was sent to international meetings to the present day when a delegation of ten, representing all phases of the Industry is sent to benefit from and make its contribution towards a better understanding of the inctricacies of cane sugar manufacture.

5

6

I have also endeavoured to show how in the conti-nent of Africa our Association is in a position where it can take the lead in all matters connected with sugar technology and that the Experiment Station and the Sugar Milling Research Institute are the only two research organisations of any importance that the developing sugar industry of Africa can turn to for advice and assistance.

The very fact that we hold this position in Africa and that our industry can be justifiably proud of its achievements during the past two decades makes me feel that the time is approaching when we should take the opportunity to show the world what we can do. Not only would an International visit do a great deal to further stimulate our technological advance, but it would also help towards eliminating some of the misconceptions that exist abroad. As I said earlier, I am sure that we have a great deal in our fields and our factories that would be of interest to an International gathering and I feel confident that if we can provide the venue for such a gathering delegates from all parts of the world would enjoy our renowned South African hospitality.

Mr. Exley Steward: Ladies and Gentlemen—to-day, for the third consecutive year, we have been privileged to be addressed from the Presidential chair by Mr. Bentley who has been President of this Association since 1957. I am sure that I express the feelings of everyone here when I say that today's address has been as informative and thought-provoking as his previous ones, and that the constructive ideas he has placed before us in all three speeches have already, or will in the future, become established facts of great benefit to our Industry.

Last year Mr. Bentley stressed the importance of the economic utilisation of the by-products of the sugar industry and, you may remember, con-cluded his address by quoting Dr. H. B. Hass as having said that "If we have our way, one day you will wear sugar, wash with sugar, use sugar in plastics and paints, spray it on your plants, feed it to your animals and, well—eat sugar too". As we now make too much sugar to eat, and restricted production is upon us, the great importance of its other possible uses is very apparent.

In 1958 Mr. Bentley dealt with the need for organised facilities for training young men who had chosen the Sugar Industry as a career, and made

particular mention of the processing and agricultural sides of the Industry. A previous course in Sugar Technology had been discontinued in 1952 due to lack of support. That course was started in 1928 and it is of interest to recall that Dr. B. E. Beater who has recently published a book on the soils of the sugar belt, and other people present here today, including myself, were among the first students in 1928. Due largely to Mr. Bentley's efforts the course was restarted in February 1959. It is of three years duration and at present there are sixteen students in their second year. The work is controlled by the Educational Sub-Committee of this Association under the Chairmanship of Dr. C. van der Pol. Although no course on Sugar Cane Agriculture has yet been established, the idea has not been dropped and the Educational. Committee is working on the by no means easy task of devising ways and means to get a course started. It is possible that agricultural students may receive initial instruction at the Cedara School of Agriculture, being later taken over by the Educational Committee for specialisation in sugar.

In the address to which we have just listened Mr. Bentley has stressed the importance and the mutual benefits derived from the holding of the three weeks congress, every three years, by the International Society of Sugarcane Technologists, in the various countries of the sugar producing world. It is encouraging to learn that the delegates from South Africa have increased from one in the early days to the ten who attended last year's congress in Hawaii, thus making the weight of South African influence in the sugar world more apparent.

The confidence expressed by the President that our Industry is in a position where other countries would benefit by the study of some of our methods, as we have benefited from many an example in other countries, is very acceptable and gratifying.

It is hoped that we may in the not too distant future, have the honour of being the host country to the International Society.

In conclusion I would like to express the appreciation of all for the sterling work done by Mr. Bentley during his three years of office and to express the hope that he will, at some future date, again accept nomination to the presidential chair.

7

THIRTY-FIFTH ANNUAL SUMMARY OF CHEMICAL LABORATORY REPORTS OF SOUTH AFRICAN

SUGAR FACTORIES

SEASON 1959-1960

By Chs. G. M. PERK

A.—SUGAR PRODUCTION DATA OF SOUTH AFRICAN SUGAR FACTORIES

(Short Tons of 2,000 lbs.)

The above data do not include the sugar produced

by the Swaziland factories. The overall production

for 1958-59 inclusive of Big Bend was 1,135,423 tons

of sugar (tel quel) and for the 1959-60 season inclusive

of Big Bend and Mhlume the overall production

amounts to 1,062,051 tons of sugar.

B.—TONNAGES OF CANE CRUSHED BY SOUTH AFRICAN SUGAR FACTORIES

(Short Tons of 2,000 lbs.)

All tons mentioned in this summary are short

tons of 2,000 lbs. However, to comply with the

publications of the Internat ional Sugar Council

and in connection with the resolution adopted at

the Third Congress of the I.S.S.C.T., t he following

table shows cane and sugar productions in met r ic

tons for recent years:

Cane and Sugar Production of South African Factories

(in metr ic tons)

In order not to confuse comparison with results

of the previous seasons, we will restrict the d a t a

recorded in this summary to those of the eighteen

South African factories. This holds in par t icular

to those tables of results which have been totalized

and which show the average values. In the first

columns of the Tables I, II and I I I at the e n d of

the summary, the da ta of Ubombo Ranches (Big

Bend) are shown; however, these d a t a have not

been taken into account when total izing a n d

averaging the results of all factories.

8

C—CANE QUALITY Comparison of Results between:

Cane harvested during the July-November Periods and

Cane harvested during Earlier and Later Months of the same harvesting season.

During the last decade there have been four optimum periods with a higher sucrose content and five with a lower than this year's sucrose content

of the cane during the optimum period. The fibre content of the cane during this year's optimum period is below the average of the last decade, viz. 15.76 per cent against an average of 15.88 per cent fibre. The purity of the mixed juice has been seven times higher than this year and twice it has been lower. The average cane to sugar ratio of this decade amounts to 8.39 against this year's ratio of 8.47 for the optimum period. The latter ratio could have been better than the average if only the overall recovery had been more satisfactory this year.

D.—THE CHANGING VARIETAL SCENE

The table shows how the varietal scene has completely changed during the last decade. In 1949-50 0.39 per cent Uba; 2.90 per cent P.O.J.; 0.71 per cent Co.290 and 47.30 per cent Co.281 were crushed. These varieties were replaced first by Co.301 and Co.331 and later by N:Co.310. The latter variety seems to have reached its biggest extension in 1957-58 when the newer varieties appeared on the scene. First it was N:Co.293 and N:Co.339, later also N:Co.292 and N:Co.376 appeared. Small quantities of N:Co.382 were also crushed during the last two seasons. New varieties will be recorded as soon as more than one per cent of all cane crushed consists of this variety. Old varieties will be recorded as long as more than one per cent of all cane crushed consists of such a variety. In this way Co.281 disappeared after 1954-55 and we see that Co.301 is well on its way out.

E.- DURATION OF SEASON AND TIME EFFICIENCY

9

As the expansion programme of sugar production was launched on 1st May, 1948, we may assume that 1950-51 was one of the last seasons before the crushing capacities of the factories were materially extended. It is therefore that we compare, in the next table, the present crashing rate of each factory with the rate as recorded in the 1950-51 season.

Individual Crushing Rates of 1959-60, compared with those of 1950-51

This table shows that the Mean Crushing Rate has increased by 31 per cent since 1950-51. The tonnage of cane to be crushed, however, has in the same period increased by 61 per cent. Hence the season was extended by 30 per cent.

F.—AVAILABLE B.T.U.'s IN BAGASSE AND REQUIRED B.T.U.'s FOR PROCESSING

The available b.t.u.'s in our natural fuel are directly proportional to the fibre percentage of the cane and the calorific value of the bagasse. On the other hand the b.t.u. 's required for processing are directly proportional to the brix to be processed. (It being tacitly understood that other conditions remain unchanged.) In the following table are accordingly compared:

(i) brix in mixed juice per 100 cane

(ii) fibre per 100 cane

(iii) lower calorific value of bagasse (L.C.V.).

The table shows that there are three factories, to wit: Ubombo Ranches, Pongola and Umfolozi, where the quantity of brix in mixed juice which has to be processed, is more than the amount of fibre in cane. All other factories are in a more favourable position; fibre being more than brix to be processed.

An unfavourable ratio between brix and fibre will be aggravated when the calorific value of the bagasse is low. With this regard, we review here the average calorific values of the Natal mills from 1925 onward:

The review shows that due to the high moisture content of the bagasse the average L.C.V. of the Natal Bagasse has at no time been very high. However, what is worse, it has not improved since 1925, but has deteriorated owing to further increase in moisture content of the bagasse. We refer to Table VI at the end of this Summary for L.C.V.'s of bagasses from other countries. A country not shown is Queensland; bagasse analysis being 47.48 per cent moisture and 3.02 per cent pol correspond-ing with a L.C.V. of 3,462 b.t.u./lb. according to the formula: L.C.V. = 7,650—18 pol—86.4 moisture (P.O.J.) or equivalent to 3,525 b.t.u./lb. according to Queensland's formula of:

L.C.V. = 7,783-22.1 pol—33.27 moisture.

Drier bagasse would render another advantage. As drier bagasse screens better than wet bagasse, less complaints about shortage of bagacillo for the rotary vacuum filter would be heard and the sucrose content of the filter cake would be reduced. It must not be overlooked that the rule of thumb that 30 lbs. of bagacillo is required per ton of cane crushed applies to bagasse of 50 per cent moisture; more bagacillo being required when the bagasse is wetter.

When a factory starts crushing faster, the screening area for bagacillo should be increased accordingly and if a higher moisture content of the bagasse is expected too, it should not be overlooked that more bagacillo per ton of cane will be needed and that the bagasse will screen less easily.

G. - MILLING TRAIN PERFORMANCE We should always mention to what extent each

milling train is pulling its weight, when we compare milling results. Some milling tandems have a much higher feed rate than other tandems of the same composition and same roller size; this higher feed rate affecting their results. We therefore show in the following table the feed rate (lbs. of fibre per hour) per cubic feet of total roller volume in addition to the percentage of lost juice and the imbibition ratio.

We are fully aware that "feed rate per cubic foot of total roller volume" is not ideal. The same size rollers may have 21" diameter by 30" journals or 19" diameter by 28" journals resulting in a 15 per cent lower maximum allowable hydraulic load for the latter. The percentage of true fibre may be higher in one case than in the other where a greater part of the calculated fibre consists of trash and tops. Another question is that one tandem may be

preceded by a couple of wide pitched knives and the other by a shredder. However, "lbs/hr./cu.ft. total roller volume" is a clear and simple indication of the feed rate. We all know that Gledhow's present 18-roller tandem operates far above Royston's capacity rating. In the table we see that the feed rate has been pushed up to 65 lbs. of fibre per cu. ft. T.R.V. Other hard pushed tandems, according to Natal standards are: Big Bend with a feed rate of 59.3 lbs/hr., Melville with 56 lbs/hr. and Glendale with a feed rate of 52.9 lbs/hr. Here, we meet another shortcoming of our feed rate ratio. As Melville has four mills after its first mill, Melville can supply at four places imbibition of diluted juice and water. Glendale has only three opportunities for imbibition because the first mill is followed by only three other mills. In addition, Melville's first mill is preceded by a Searby shredder and Glendale's is not. Big Bend is even worse off as the tandem comprises eleven rollers only and only 138 per cent imbibition can be applied.

Also showing a high feed rate is Z.S.M. with 54 lbs/hr./cu. ft. T.R.V. Since Z.S.M.'s tandem has been designed for a high feed rate, we will be following with interest the performance of this 18-roller tandem when the crushing rate has been increased further. Though the average figure for the lost juice has been high this season owing to very high lost juice figures at the beginning, we must not forget that at the end of the season when the teething troubles were overcome, a lost juice percentage of 43 was obtained.

It is, of course, not only the number of times imbibition of diluted juices and water can be applied to the intermediate bagasses, but also the amount of water applied which affects the final results. As the imbibition rate increases more mills go over to the old practice of dividing the quantity of imbibition water over two mills, i.e. in front of the last and of the penultimate mill. As a departure from the counterflow leaching system it is not to be applauded, but as a means to reduce the moisture content of the final bagasse in the case of a high imbibition ratio, it is appreciated.

So far we have discussed the rate at which the mills were fed. We will now peruse the table from the point of view of losses of absolute juice, per cent fibre. However, before we do this, we want to stress once more that: "although sucrose extraction may be a suitable gauge for financial comparisons, as the mills squeeze out juice (and not sugar as such), lost juice per cent fibre evaluates better the results obtained by the repeated, combined action of squeezing and diluting.

The table reveals that there are six factories which were able to reduce lost juice to ratios below 40 per cent. They are: Renishaw (35.2 per cent); Natal

10

11

Estates (36.1 per cent); Tongaat (36.7 per cent); Darnall 37.3 per cent); Amatikulu (38.2 per cent) and Sezela (39.9 per cent lost absolute juice, on fibre). Regarding the latter factory, the ratio would have been lower if the imbibition had not had to be reduced to an average; of 175 per cent on fibre. This measure; was required in connection with the increased crushing rate and in order to maintain the syrup density above 60 per cent brix.

NOTE: The: rate of mixing of the imhibition water with the residual juices present in the intermediate bagasses a.s indicated by "Inhibi t ion Efficiency" is shown below the inhibition ratio in Table 11 at the end of the summary.

H.—SUCROSE LOST IN BOILING HOUSE It has been an unsatisfactory season with regard to

recovery; the average percentages of sucrose lost in filter cake and molasses being higher than we reasonably may have expected. The average purity of mixed juice (as far as it may be used to evaluate the quality of the juice) does not give any infor-mation on the reason for this season's poor recovery. However, since all factories suffered, to a lesser or greater extent, of a low recovery, it is allowable to assume that this season's peculiar juice qualities caused difficulties during processing which could not be successfully mastered either by the available clarification plant or by the available equipment for C-massecuites.

A review of boiling house performance, boiling house recovery, overall recovery, mixed juice purity and reducing sugars/sucrose ratios in mixed juice from 1925 onward can be found in Table VII. This table reveals that the average B.H.P. over the period from 1925 to 1934 inclusive, has been 91 per cent, over the 1935-1944 decade 95 per cent, over 1945-1954 the average was 97 per cent. Since then the average B.H.P. figure has always been more than 97 per cent; the peak being 98.5 per cent obtained in season 1957.

More recent figures are shown in the following table covering only the last five years:

Sucrose Losses in filter cake—The more we sweeten off, the more the ratio decreases between sucrose (apparently) recovered and impurities extracted simultaneously from the filter cake. In addition, the amount of sucrose extracted per gallon of water used, diminishes rapidly after a certain amount of sucrose has been extracted. It is therefore that we have never been advocates of sweetening off to the

extreme. We all know that the purity of the diluted juices discharged from the filters drops more and more, the further the sweetening off process is carried on, but what is not generally known is the fact that the lime salt content of the filter juices increases even more rapidly. It is not known to what extent these impurities and the lime salts extracted from filter cake will be precipitated again when the filter juices are returned and pass the clarification process once more. The generally accepted rule seems to be that the purity of the filter returns should not be more than 1-1 units lower than the purity of clarified juice.

However, this season's general rise in the figure for sucrose per cent filter cake is not a result of an intentional reduction of the degree of sweetening off with the above in mind, but it has been brought about by other causes. One may be lack of sufficient bagacillo of the required degree of fineness. We have previously drawn attention to the fact t ha t an increase in crushing rate combined with an increase in moisture content of the bagasse requires an increase in screening area far above an increase in proportion to the increase in crushing rate . The wetter the bagasse, the more difficult the bagasse screens and the less will be the percentage of fines in the bagacillo. Hence, the more pounds of bagacillo per ton of cane are required, the less easily the bagasse screens.

The percentage of fines in bagacillo should be checked regularly; 90 per cent of the bagacillo sample should pass through a 20 Mesh Tyler sieve.

Sucrose Losses in final molasses—These losses have been higher than corresponds with the mixed juice purity, because of the general rise in final molasses purity. This year's molasses purities take us right back to the level of exhaustion as before 1952, when molasses purities were consistently above 40 according to table VII. Table I I I (at the end of the summary) reveals that there are only a few-factories which obtained an exhaustion of more than 60 per cent from their C-massecuite (Exhaus t ion= crystallized sucrose in massecuite as a percentage ratio of all available sucrose in the same massecuite). However, since there have always been only a few factories which have achieved this in recent years, it does not explain the average rise in molasses puri ty this season. This is not to say that if more factories had been able to achieve more than 60 per cent exhaustion in their C-massecuites, the average pur i ty of the molasses would not have been materially lower.

What are the reasons that not all the factories obtain an exhaustion of more than 60 per cent from their C-massecuite? Proper exhaustion requires in the first place sufficient crystal surface area, viz. a very finely grained C-strike. Secondly, t he grain should be of uniform size and of a sound crystal

iSPiflj

form. Generally speaking, the grain of the required specification can be more easily obtained by graining in a low medium (molasses) than in a high purity medium (syrup). Thirdly, the purity of the C-masse-cuite must be sufficiently low; in general it must be lower than 60 per cent purity and 58 per cent purity seems to be a suitable figure. The required low purity can be more easily obtained when starting with a low purity footing (graining in molasses) than when the footing has been grained on syrup. As the crystallizers have only a supplementary action with respect to exhaustion, most of the exhaustion should be obtained in the pan. Boiling slowly, maintaining the masse tight during the whole boiling process is a requirement for proper exhaustion. The exhaustion achieved by the pan should be checked regularly, by taking a Nutsch from the massecuite, directly after striking. Another Nutsch taken just before curing will reveal the part played by the crystallizer in the total exhaustion. Moreover, a comparison of the result of the second Nutsch with the purity of the molasses spun off by the factory centrifugals gives a check on what is taking place there. Finally, the strike should be tightened up heavily, but slowly to prevent the formation of microscopic fine grain at the last moment. However, in nearly all factories, the tightening up process is partly undone shortly afterwards by mingling the tightened up massecuite with the pan steamings. Sometimes it is lack of proper re-heating equipment which forces the factory to take recourse to dilution of the massecuite. At other times, it is the C-masse-cuite pump which makes the use of water a necessity. The possibility that water or steam was applied within the basket is of course remote.

Sometimes "stickiness" of the massecuite is brought forward as the cause of unsatisfactory exhaustion. When this stickiness is not caused by the presence of false grain (or air bubbles), stickiness will not necessarily lead to a higher molasses purity, if only the capacity of the C-centrifugals is big enough. In this respect, we draw attention to conditions in the Hawaiian sugar industry where the viscosity of the final molasses is such that the C-massecuites have to be spun from three-quarters to more than one hour. In this respect, we must not overlook the fact that in general the centrifugal force exerted by their 40" diameter baskets driven by 60 cycles/second A.C. electric motors is higher by 37 per cent compared with our 42" diameter baskets running on 50 Herz systems. Hawaii installed 550 sq. in. for low-grade massecuites and 275 sq. in. screen area for high-grades, per short ton of cane per hour.

A downward trend of the purity of the Natal molasses became possible when and where bigger capacities for C-massecuite centrifugals were in-stalled. Sezela pioneered in this respect by installing

seven 42" X 24" centrifugals (1,460 r.p.m.) for a crushing capacity of 125 t.c.h., which is equivalent to 177 sq. in. screen area per t.c.h. and 1,272-times "g". Illovo followed suit with four 40" X 30" centri-fugals (1,383 r.p.m.) for 70 t.c.h., which corresponds with 215 sq. in. per t.c.h. and 1,100-times "g" . We see that these screen areas are still far below Hawaii's ratio of 550 sq. in. per low-grade curing.

A generous centrifugal capacity for C-massecuite is needed in order to be able to maintain a low final molasses purity even under less favourable conditions as were experienced this year.

Returning to the question of "stickiness" as the cause of unsatisfactory exhaustion, we want to point out that too much lime in the case of defecation can be a reason for a too high molasses purity. Our lime dose should be adjusted to achieve the highest possible recovery, i.e. lowest molasses purity. This should be our aim and to achieve this aim, we should try different pH values for limed juice until we get the best results. Liming just to a certain pH of clarified juice or with the purpose of preventing the pH of the syrup dropping below pH7 is not aiming at the lowest molasses purity.

We mention all these facts in order to bring to the fore the fact that it is not a regional question of different juice qualities, which cause only a small number of factories to obtain satisfactory exhaustions.

Undetermined Sucrose Losses—This third source of losses consists mainly of mechanical losses (including inaccuracies and mistakes in weighing, sampling and compositing, analysis and calculation) and only a very small part is due to chemical reactions, i.e. inversion of sucrose. Sucrose is a rather stable chemical substance and Stadler's table confirms that under normal operational conditions in properly designed apparatuses the loss due to inversion is very small indeed. Entrainment is the main source of sucrose losses and dunder water another. As there is a relation between crushing rate and entrainment, an increase in the former can result in a higher undetermined loss. The sucrose losses in dunder water are often much higher than realized. Here again, we can assume a relation between losses in dunder water and length of season and crushing rate. Shorter off-seasons hamper maintenance and repairs, higher crushing rates can lead to increased spillage.

Individual Boiling House Results—The highest boiling house performance figures have been recorded by the rawhouse departments of the factories-cum-refineries Sezela and Gledhow (respectively 98.7 per cent and 98.5 per cent), and by Melville with 98.5 per cent. Their undetermined sucrose losses are (in the same sequence): 0.18 per cent, 0.74 per cent and 1.21 per cent, while their molasses purities are

12

13

respectively 36.5 per cent, 39.8 per cent and 40.6 per cent. Felixton and Umfolozi had also undetermined losses of less than 1 per cent, viz. 0.69 per cent and 0.84 per cent on sucrose in cane respectively. The lowest final molasses purity was at Illovo, viz. 35.3 per cent, followed by Sezela (36.5 per cent) and Renishaw (37.4 per cent). There are four more factories which were able to keep their final molasses purities below 40 per cent; they are Pongola (39.1 per cent); Gledhow (39.8 per cent) and Felixton (39.9 per cent), while Ubombo Ranches recorded an apparent purity of 35.5 per cent (Gravity Purity approx. 38.5 per cent).

I.—SUCROSE AND BRIX LOSSES BASED ON CLARIFIED JUICE ANALYSIS, NON-SUCROSE REMOVAL BY CLARIFICATION AND FINAL

MOLASSES OBTAINED

To this section belong three tables. Firstly, the table showing the non-sucrose present in total final molasses as a ratio of the non-sucrose present in mixed juice, while the second table shows data regarding removal of non-sucrose, brix and sucrose losses based on clarified juice analysis, etc. Thirdly, another new table showing the non-sucrose circulation inside the system "C-massecuite—C-sugar— Final Molasses".

Regarding the first table, we repeat here the warning that the ratio between non-sucrose in total final molasses and non-sucrose in mixed juice cannot be considered as an absolute gauge for non-sucrose removed and the non-sucrose increase during processing because the non-sucrose as present in mixed juice differs in composition from the non-sucrose as calculated to be present in total final molasses.

TABLE I RATIO BETWEEN NON-SUCROSE IN MIXED JUICE AND NON-SUCROSE IN TOTAL FINAL MOLASSES

The second table comprises five columns. The first column (i) shows the non-sucrose removal by-clarification as a percentage ratio of the non-sucrose originally present, i.e. in mixed juice. The non-sucrose present in clarified juice is calculated with the aid of the purity of the clarified juice, because this is the official method of calculating. When, however, the purity of syrup was used, the non-sucrose removal should have been bigger because many factories show a higher purity of syrup than that of clarified juice. (Also the brix per cent clarified juice is sometimes higher than the brix per cent mixed juice.)

TABLE II

NON-SUCROSE REMOVAL, PERCENTAGE of FINAL MOLASSES OBTAINED and UNDETERMINED SUCROSE and BRIX LOSSES based on CLARIFIED JUICE PURITY*

The second column (ii) shows the actual quantity obtained of total final molasses as a percentage ratio of the calculated quantity based on clarified juice purity. As the percentage ratios as shown in column (ii) are—in general—rather low, we drew sucrose and brix balances for all those factories which weighed their final molasses; the balance being based on sucrose and brix present in clarified juice. The results, i.e. the undetermined sucrose. and brix losses are shown in columns iii and iv respectively. Finally, we divided the sucrose loss by the brix loss; the results being shown under the title of "Puri ty of the Loss" in column (v).

In accordance with the—in general—low ratios as shown in column (ii), the undetermined brix losses are—again in general—much higher than the sucrose losses, resulting in low purity figures (column v). These low "purities" can only partially be caused by a too low purity of the clarified juice and/or a too low weight of final molasses.

*As sugar has been remelted in the juice, no figures for NE could be calculated.

TABLE III

BATIO BETWEEN NON-SUCROSE IN C-MASSECUITE AND NON-SUCROSE IN FINAL MOLASSES

In the third table, we compare the lbs. of non-sucrose in C-massecuite with lbs. of non-sucrose present in weighed final molasses; both quantities per ton of cane crushed. In the last column of the table, the percentage ratios between the two are shown; a high ratio pointing to a greater circulation of non-sucrose than a low ratio figure.

Also, in all three cases, the analysis of the C-masse-cuite has been: 98.9 per cent Brix and 58 per cent purity; while the lbs of N.S. in final molasses per ton of cane has been assumed to be equal to 35.00 lbs in all three cases. It shows clearly how the volume of C-massecuites per ton of cane increases as a result of increased circulation of N.S. when the purity of the ('-sugar drops.

In order to illustrate the. effect of a higher purity of the C-massecuite on the quantity of C-massecuite, we have calculated a similar table, but this time based on a C-massecuite analysis of 60 per cent purity and 98.9 per cent Brix, instead of on 58 per cent purity and 98.9 per cent Brix.

The quantity of C-massecuite per ton of cane crushed depends on different factors such as juice purity, massecuite purity and last but not least on the purity of the (pre-cured) C-sugar. Umfolozi, which has to process a mixed juice with the lowest purity of all factories, shows accordingly one of the highest lbs. of C-massecuite. per ton of cane crushed. The quanti ty would have been lower if the purity of the C-massecuite could have been lowered. The high figure is not caused by an excessive non-sucrose circulation in the system: "C-massecuite—C-sugar— Weighed Final Molasses" as the ratio of 117 per cent in the last column shows.

To illustrate the relation between purity of (pre-) cured C-sugar and the ratio "N.S. in C-massecuite per cent N.S. in final molasses", we have calculated the latter ratios for three cases, viz. for a puri ty of C-sugar equal to 92 per cent, to 85 per cent and to 78 per cent. In all three cases the final molasses has a purity of 38 per cent.

78 1:27.26) 50..34 35. 00 144

Comparing the results it shows that there is a higher ratio of non-sucrose circulation in the case of 60 per cent purity than when the purity of the C-masse-cuite is 58 per cent. This is caused by the higher yield of C-sugar when the C-massecuite. has a purity of 60 per cent, viz. the yield being 41 per cent of C-sugar of 92 per cent purity against only 37 per cent, sugar of 92 per cent purity when the purity of the C-massecuite is 58 per cent.

Both the tables show too that the quantity of C-massecuite is controlled by three factors:

(i) the quantity of non-sucrose which has to be expelled in the form of final molasses,

(ii) the purity of the C-massecuite, and

(iii) the degree of circulation of non-sucrose inside the system of C-massecuite-final molasses-C-sugar.

14

15

Finally, the quanti ty of non-sucrose to be expelled as final molasses (i) depends on the input of non-sucrose into the boiling house (purity of the mixed juice) and the degree at which non-sucrose is removed and non-sucrose is formed during the clarification process. Hence, the quantity of non-sucrose in final molasses depends on the quality of the juice and the clarification method (carbonatation or defecation and sulphitation).

We have enlarged on this subject in order to explain how more C-massecuite can be "made".

J.—LIME, SULPHUR AND PHOSPHORIC CONSUMPTIONS

Defecation -has been practised during the whole, crushing season by Illovo, Tongaat, Darnall, Amati-kulu, Felixton, Umfolozi, Melville, Umzimkulu and Pongola; Illovo and Pongola also use this process for producing millwhites. For part of the season Entumeni also used defecation when turning out millwhite.

AVERAGE LIME CONSUMPTION OF THE DEFECATION FACTORIES FOR RECENT YEARS

Sulphitation—has been exclusively used by Doorn-kop, Glendale, Chaka's Kraal, Renishaw and Sezela. Ubombo Ranches used sulphur and lime during those periods when millwhite was produced. En tumeni used sulphitation for part of the season when portion of the production was millwhite.

AVERAGE CONSUMPTION OF THE SULPHITATION FACTORIES

(lbs. per ton)

When perusing the individual consumptions of chemicals of the different factories, we see tha t it is in particular the consumption of phosphoric which varies mostly. Two factories did not use phosphoric at all, i.e. Sezela and Ubombo Ranches; the latter using tank sulphitation. Some factories use an amount of sulphur in accordance with their consumption of lime and phosphoric. Others do not, even when we take the time efficiency into account with a view to sulphur losses during stoppages. One of the factories where the lime-sulphur-phosphoric accounts tally is Chaka's Kraal. Chaka's Kraal used 10.82 lbs of sulphur per ton of Brix, which is equivalent to 0.99 x 10.82 X 56/32=18.74 lbs of CaO; assuming 99 per cent S in the sulphur. The phosphoric consumption was 1.34 lbs which is roughly equivalent to 1/2 X .34 or 0.67 lbs of CaO. The total requirements for neutralization of S0 2

and phosphoric are therefore 19.41 lbs of CaO or 20.43 lbs of lime of 95 per cent CaO. The actual consumption was 29.46 lbs, leaving 9.03 lbs per ton of Brix for the "defecational" part of the defecation-sulphitation process. A consumption of 9.03 lbs lime per ton of Brix in mixed juice tallies with what pure defecation factories use.

Miscellaneous Mixed juice carbonatation has been applied by

Natal Estates, while Sezela and Gledhow applied the carbonatation process to the melt liquor of their raw sugars in order to produce refined sugar. Umfolozi applied the sulphitation process to the melt liquor.

Table I.—CANE CRUSHED, SUGAR MADE, CANE VARIETIES AND THROUGHPUT

Table II.—SUCROSE BALANCE, RECOVERIES, ANALYSIS OF BAGASSE, JUICES, CAKE AND SYRUP

TORY UR PC UF ZM FX EN AK DK GD DL GL MV CK TS NE IL RN SZ UK Averages

Table III.—MASSECUITES AND MOLASSES, LIME, SULPHUR AND PHOSPHORIC PASTE

Table IV.—COMPACTION OF FINAL RESULTS FOR S.A. SUGAR FACTORIES

(Season 1950 to Season 1959 inclusive)

Table V.—AVERAGE MANUFACTURING RESULTS BY MONTHLY PERIODS FOR S.A. SUGAR FACTORIES,

SEASON 1959—1960

Table VI.—COMPARATIVE RESULTS FROM OTHER COUNTRIES FOR RECENT YEARS

22

23

Mr. Bentley (President) said the Report was drawn up with all the thoroughness Mr. Perk was noted for. He asked Mr. Perk for more information on how he suggested the juice should be limed to give maximum recovery rather than liming, as at present, to a pre-determined pH.

Mr. Perk replied that he was often told that factory operatives use so much lime to get a certain pH, but nobody mentioned that he adjusted the amount of lime to obtain the highest recovery. Usually too much lime was used and the molasses became viscous as a. consequence. The important point was that lime should be added to get a pH which would render the highest possible recovery. In this case he was talking of defecation factories.

Mr. Elysee asked Mr. Perk how low a purity he would like the C massecuite to be kept to, because this increased in pH with lower purity and there was a danger of inversion. lie sa.id. that his aim was to get C massecuitc of about 7 pH and he wanted to know how far down he could go with the purity when the pH might reach as low as 5.

Mr. Boyes asked how Mr. Elysee got such a low pH value.

Mr. Elysee replied that molasses was diluted down 50 per cent and the reading taken on a potentiometer, but he considered the pH was even lower than that shown by that instrument.

Mr. Boyes said ho considered the pH was probably about 6.5 based on a determination diluting 50:50 with water.

Mr. Rault said that diluting massecuite or molasses did change the pH value very slightly. There was no difficulty in reading the pH of a molasses in its original thick state. The pH reading of a massecuite was really that of the mother liquor, but there was a danger of breaking the glass electrode when forcing it into the thick massecuite.

Mr. Thumann asked Mr. Perk how it would be possible to estimate from day to day what pH one would have to work to to get the highest recovery. The cane varied enormously from day to day and it was hard to judge whether to use a lower or higher dose of lime for these varying canes. He considered that the best criterion was to get: a good settling, bearing in mind that one could not go below a. certain pH because of the possibility of inversion. At the same time if one went very high in pH it might have a serious effect on the recovery by increasing the viscosity of the final boilings.

Mr. Perk replied that liming just for the sake of reaching a certain pH was not correct. The point was that the amount of lime to be used was that which would give the highest recovery.

Mr. Thumann asked how could it be determined that higher recovery was due to the lime applied.

He said the B molasses purity was important, but if one did not lime well one would be faced with a very difficult molasses to work with. There were several factors in recovery and one could not consider lime alone, nor could one go by the purity of the C massecuite only.

Mr. Antonowilz considered that the biological aspect should be considered. If there were insanitary conditions round the filter station the amount of lime going into solution was very much higher than usual and as a. consequence extra molasses was formed. He agreed with Mr. Perk that one should waich the liming but he considered the biological aspect should also be stressed.

Dr. Douwes-Dckker considered that some confusion was noticeable in the discussions. Insufficient liming of juice, resulting in a. very low pH created the clanger of inversion, hence a low recovery. Over-liming, on the other hand, as indicated by a high pH, would result in an. excessive amount of soluble lime salts being formed, hence rather viscous C-massecuites and also a low recovery. A juice was limed correctly if it corresponded to maximum recovery. To judge the correctness of the amount of lime added it was not possible to compare the actual amount with a standard quantity, for different juices required varying amounts of lime. Neither could we always use a standard pH. The pH was, however, a much better criterion than any other, but we should never forget that liming to a certain pH was not a purpose in itself; the correct aim was to lime in such a way that maximum recovery was achieved. The pH was only used as a convenient expedient, and saying that the juices were limed to such and such a pH did not always mean that the juice was correctly limed.

Mr. Rault said that Dr. Douwes-Dekker, in Honig's book, had contributed an interesting study of the bad effect of lime salts on molasses formation and loss of sucrose therein. Residual lime salts were often a result of the composition of the raw juice, and not only due to too high a final pH, and for that reason somewhat outside human control. He wished that all factories would determine the lime salts in the clarified juice as a. routine: control, but in the case: of the defecation and sulphitation process, the presence of magnesia interfered with the determination of lime salts by the usual methods.

Mr. Thumann said he might get a better recovery by giving a. correct dose of lime but there was always the possibility that we might impair the capacity of the clarification and boiling house stations. We had been told today that we must try and reduce the: number of man-hours to produce a ton of sugar. To some extent the question of increase in recovery by 1 per cent might result in extending the season and this had to be considered.

24

Mr. Rault wished to congratulate Mr. Perk on the high standard of his annual summary of laboratory reports, which this year was a searching enquiry on the efficiency of the industry and especially on the factors influencing the exhaustion of final molasses and the advice given for improvement along this line. Special stress is laid on the necessity of centrifugal power for the C massecuite although his calcu

lations are quite correct on the larger a m o u n t of recirculated non-sugars, by a bad curing of a 60° purity against a 58° purity massecuite, he had found that there was a greater risk of h a v i n g a badly cured sugar from larger circulation of molasses when at tempting to lower too much the p u r i t y of the final massecuite, rather than by the s l igh t ly richer one above 58°.

WEATHER REPORT FOR THE YEAR 1st JUNE, 1959, TO 31st MAY, I960

By J. L. DU TOIT

General

In this report the same procedure will be followed as in recent years. The climatic conditions during the year 1st June, 1959 to the 31st May, 1960 will be discussed in some detail but conditions during the previous twelve months will also have an effect on the crop to be cut in 1960-61. and these will, therefore, also be referred to.

Meterological data other than rainfall data which are collected from 04 recording stations, arc: from the Experiment Station, Mount Edgecombe. This station is, however, fairly centrally situated and the records may, therefore, be considered fairly representative of the weather conditions prevailing in the Industry as a whole.

Rainfall Returns from 54 Centres

The centres from which rainfall data are obtained are well scattered and representative of the whole sugar belt. The Industry is divided into the normal geographic divisions, i.e. South Coast, North Coast and Zululand and further sub-divided into magisterial districts.

Table I gives the annual rainfall for the past five years for each of the 54 recording centres.

Table II gives the rainfall by magisterial districts, and also for the three main divisions for each month of the year from June 1959 to May 1960.

Table III gives the calculated mean rainfall for the past 36 years and the monthly percentage distribution. The actual rainfall for the year now under review is also given as are the evaporation data taken at the Experiment Station, Mount Edgecombe.

Table IV gives the rainfall distribution according to growing periods for the past two years for all magisterial districts and the three main sub-divisions of the Industry.

Table V gives the monthly rainfall for the 54 centres for the past 4 years, the evaporation from an open water tank at the Experiment Station for the same period, and the amount by which evaporation exceeded rainfall each month.

TABLE I

Rainfall for Fifty-four Centres

25

26

TABLE II

Rainfall in Inches by Districts for the Months of June, 1959, to May, 1960, inclusive

TABLE III

Rainfall and Evaporation Data

27

TABLE IV

Rainfall in Inches by Districts for the Two-year Period June, 1959, to May, 1960, inclusive

TABLE V

Rainfall and Evaporation in Inches for the Past Four Years

Comments on Rainfall The rainfall for the year ending 31st May, 1960

was 35.66 inches, which was somewhat below our mean annual rainfall of 37.93 inches. The Industry has, therefore, gone through two successive years of rainfall below normal and although these adverse conditions were by no means extreme they had a depressing effect on crop yields.

The lowest rainfall during any month of the year was recorded during June 1959 when on the average only 0.07 inches fell in the sugar belt and with the exception of Dumisa no rain was recorded at all on the South Coast. July was quite warm with the temperature 2.2°F above normal but again the rainfall was deficient with an average fall of only 0.66 inches and particularly in Zululand the drought position became quite serious. Conditions deteriorated during August and cane started to die in Zululand when during the last few days of August excellent rains fell throughout the Industry and an average of 2.35 inches was recorded. The rainfall for September was about normal amounting to 2.63 inches but during October the rainfall was well above normal at 4.58 inches and the crop was making good progress. Cold cloudy conditions and a relatively low rainfall of 3.03 inches during November, however, somewhat retarded growth. Similarly the rainfall during December, 3.99 inches, was below normal.

This was followed by an exceptionally dry January with only 2.10 inches of rain. Zululand had only 1.65 inches of rain and was the worst affected by this drought. Fortunately, however, this area had the best rains in February when the whole Industry recorded the above average total of 5.05 inches. March had a somewhat lower rainfall with 4.68 inches but excellent rains totalling 5.12 inches fell during April with the result that the crop was not so badly affected by the rather low rainfall of 1.40 inches during May.

Summarising the rainfall over the past two years, the following conditions prevailed. The crop was adversely affected by a rather severe winter drought from June to August 1958. From mid-September to the end of 1958, however, the crop made good growth. A severe summer to autumn drought, however, set in with the new year which was not ended until the middle of May when devastating floods occurred on the South Coast. The floods were

in turn again followed by a dry winter spell but normal spring rains improved the crop which again suffered a set-back with the January drought of I960. Good rains from February to April were followed by a rather dry May but the condition of the crop at the end of May was still fair.

Temperatures The mean screen temperature at the Experiment

Station for the year ending 31st May, 1960 was 68.7°F or 0.1° below normal. The winter months, June to August, were warm particularly July which had a mean temperature 2.2°F above normal. March also had a mean temperature above normal and October was normal. The temperature for all other months was low however, and April had a mean screen temperature of 68.0"F or 2.3° below normal.

Evaporation The evaporation from an open water surface

totalled 48.63 inches or 2.09 inches above normal. There were only two months, February and April, with rainfalls above evaporation and these were probably the only two months when cane would not have benefited from irrigation. The accumulated rainfall deficiency for the year, i.e. the monthly totals of evaporation in excess of rainfall amounted to 14.92 inches.

Hours of Sunshine The hours of sunshine for the year now under

review were 101.0 per cent of normal. The hours of sunshine for the period March to June were also slightly above normal. Hours of sunshine for this period were found to correlate with sucrose per cent cane and consequently a sucrose per cent cane slightly above normal can be expected this year.

Summary and Conclusions The Industry has now gone through its second

successive year with a rainfall somewhat below average. The winter drought and a January drought have been partly off-set by fairly good rainfall conditions in the spring and from February to April and the crop is in fair condition.

Slightly lower cane per acre yields than in recent years is anticipated but it is hoped tha t the sucrose per cent cane may be somewhat better.

28

S.A.S.A. Experiment Station, MOUNT EDGECOMBE.

4th July, 1960.

29

TABLE VI

The following are the Screen Temperatures by Months in Degrees Fahrenheit at the Experiment Station for the year June, 1959, to May, 1960, compared with the Means for the Period 1928 to 1959

TABLE VII

The following Table gives the Mean Monthly Earth Temperatures

ANNUAL SUMMARY OF AGRICULTURAL DATA FOR THE SUGARCANE CROP 1957/58 By J. L. DU TOIT and K. E. F. ALEXANDER

As in the past this report is mainly based on the Special Census of Sugarcane Plantations compiled by the Government Department of Census and Statistics. Other sources of information are:

(a) Survey of Cane Production by the Sugar Industry Central Board.

(b) Annual Summary of Laboratory Reports by Chs. G. M. Perk.

(c) Annual weather reports from the Experiment Station.

(d) Fertilizer Traders' Association data submitted to the Experiment Station.

Rainfall and Yield The average crop yield for Europeans for the

season 1957-58 was 36.1 tons cane per acre, which up to that date, was an all-time record. The average rainfall for the sugar belt for the year ending 31st May 1957, which would largely affect the subsequent crop of the 1957-58 season was 48.88 inches. This figure is more than 10 inches above our normal rainfall. Undoubtedly the crop profited from this excellent rain and with better varieties grown and more fertilizer being used the record crop was produced. Where our crop grows for more than one season before being cut it is natural that the rainfall of the previous year i.e. the year ending 31st May 1956 would also have had some influence on the 1957-58 season's yield. The rainfall that year was 38.33 inches or slightly above normal.

Table I gives the average yield for the past 18 years as well as the average rainfall for the years ending 31st May. All yields refer to Europeans only.

Table I once again shows how extremely dependent our crop is on rainfall. The 1956 set-back in production compared with 1955 was simply due to the lower rainfall of that year and with the good rains of 1957 a new record in yield was established— a record that was again bettered according to Central Board figures in 1958 when the rainfall was even better than in 1957. At the same time the basic yield of the Industry, independent of rainfall, has improved considerably during recent years and this is undoubtedly due to improved varieties, the increased amounts and better balanced fertilizers being used, and to improved agricultural methods in general. The rainfall in 1956 was comparable with that of 1944 but the yield 2 1/2 tons better at 36.1 tons cane per acre in spite of the fact that the 1944 benefited from the better rains in 1943. Similarly the rainfall in 1957 can be compared with t ha t of 1943 but the yield in 1957 was more than 5 tons better than in 1943. The best varieties grown in 1943 and 1944 were however Co.301 and Co.281. Co.281 has since disappeared and 22,000 acres of Co.301 harvested, admittedly mostly as old ratoons, during 1957 only averaged 26.9 tons cane per acre compared with the industrial average for all varieties of 36.1 tons cane per acre.

Rainfall distribution is of course also of prime importance in cane production and in Table II the rainfall distribution for the year 1st J une 1956 to 31st May 1957 is given.

TABLE II

In addition to the excellent total rainfall for the year, the rainfall distribution was also on the whole very good, but there were severe floods during December at Mtunzini and on the Umfolozi flats.

The winter of 1956 was relatively warm bu t lack of heat during the period from September to December rather retarded cane growth. January and Feb ruary 1957 had temperatures well above norma l and this heat and ample moisture made for excellent conditions of cane growth.

30

TABLE I

ANNUAL SUMMARY OF AGRICULTURAL DATA FOR THE SUGARCANE CROP 1957/58 By J. L. DU TOIT and K. E. F. ALEXANDER

As in the past this report is mainly based on the Special Census of Sugarcane Plantations compiled by the Government Department of Census and Statistics. Other sources of information are:

[a) Survey of Cane Production by the Sugar Industry Central Board.

(A) Annual Summary of Laboratory Reports by Chs. G. M. Perk."

(c) Annual weather reports from the Experiment Station.

(d) Fertilizer Traders' Association data submitted to the Experiment Station.

Rainfall and Yield The average crop yield for Europeans for the

season 1957-58 was 36.1 tons cane per acre, which up to that date, was an all-time record. The average rainfall for the sugar belt for the year ending 31st May 1957, which would largely affect the subsequent crop of the 1957-58 season was 48.88 inches. This figure is more than 10 inches above our normal rainfall. Undoubtedly the crop profited from this excellent rain and with better varieties grown and more fertilizer being used the record crop was produced. Where our crop grows for more than one season before being cut it is natural that the rainfall of the previous year i.e. the year ending 31st May 1956 would also have had some influence on the 1957-58 season's yield. The rainfall that year was 38.33 inches or slightly above normal.

TABLE I

Table I gives the average yield for t h e pa . s t 18 years as well as the average rainfall for t h e years ending 31st May. All yields refer to E u r o p e a n s only.

Table I once again shows how extremely d e p e n d e n t our crop is on rainfall. The 1950 set-back in p r o d u c tion compared with 1955 was simply d u e to t he lower rainfall of that year and with the g o o d ra ins of 1957 a new record in yield was e s t a b l i s h e d -a record that was again bettered according to C e n t r a l Board figures in 1958 when the rainfall w a s even better than in 1957. At the same time t h e basic yield of the Industry, independent of r a i n f a l l , has improved considerably during recent years a n d this is undoubtedly due to improved v a r i e t i e s , the increased amounts and better balanced f e r t i l i z e r s being used, and to improved agricultural m e t h o d s in general. The rainfall in 1956 was c o m p a r a b l e with that of 1944 but the yield 2 1/2 tons b e t t e r at 30.1 tons cane per acre in spite of the fact that t h e 1944 benefited from the better rains in 1943. S i m i l a r l y the rainfall in 1957 can be compared w i t h t h a t of 1943 but the yield in 1957 was more t h a n 5 tons better than in 1943. The best varieties g r o w n in 1943 and 1944 were however Co.301 a n d C o . 2 8 1 . Co.281 has since disappeared and 22,000 a c r e s of Co.301 harvested, admittedly mostly as old r a t o o n s , during 1957 only averaged 26.9 tons cane p e r acre compared with the industrial average for all v a r i e t i e s of 36.1 tons cane per acre.

Rainfall distribution is of course also of p r i m e importance in cane production and in T a b l e II the rainfall distribution for the year 1st J u n e 1 95(5 to 31st May 1957 is given.

TABLE II

In addition to the excellent total rainfall f o r t he year, the rainfall distribution was also on t h e whole; very good, but there were severe floods d u r i n g December at Mtunzini and on the Umfoloz i flats.

The winter of 1956 was relatively warm b u t lack of heat during the period from September to D e c e m ber rather retarded cane growth. January a n d F e b ruary 1957 had temperatures well above n o r m a l and this heat and ample moisture made for e x c e l l e n t conditions of cane growth.

30

Total Yields and Areas During the season 1957-58 the Industry crushed

a total of 8,594,618 tons of cane to make 959,872 tons of sugar. The Special Census of Sugarcane Plantations 1957-58 deals with 982 returns totalling 7,58(5,21)7 tons of cane or 88.3 per cent of the total crop, and according to the Central Board 96.2 per cent of the European production. The Central Board survey is in that respect more complete and deals with the whole crop covering European and non-European production. The latter, however, do not give the varieties grown or a plant and ratoon analysis.

In Table III yields and areas for the 1957-58 season as given by our two sources of information are recorded.

TABLE III

The Central Board figures may further be broken down to give the average yield and percentage production for the various groups comprising the Industry and the results for the 1957-58 crop are given in Table IV.

With the exception of N:Co.339 which gave the top performance with the excellent yield of nearly 45 tons cane per acre, the yields seem to increase from the oldest to the newest variety. One has to take into account that the old varieties are predominantly grown as old ratoons and newer varieties as plant cane, but is there not also further evidence of varietal decline? There are a number of complicating factors which make an answer difficult but in the following table the performance of a number of varieties is given over a period of time and the rainfall is given as a guide to what the yield should have been.

At first glance the yields obtained during 1950 seem to compare more than favourably with those of 1956 and 1957 when climatic conditions were much better and consequently a deterioration in the varieties Co.301, Co.331 and N:Co.310 may be suspected; but that this is largely ilusionary and simply the effect of older ratoons depressing the yields of the latter years is shown in Table VII where a comparison is based on plant cane only. This table also shows the high yields being obtained from the new varieties N:Co.376 and N:Co.339.

TABEL VII

31

TABLE IV

TABLE VIII

TABLE V

Variety and Yield The 1957-58 Census returns list N:Co.37G and

N:Co.292 for the first time and both gave excellent yields (over 40 tons cane per acre) but these yields which will be mostly plant cane are from such small areas that no definite conclusion can be drawn.

TABLE VI

Table VIII gives the percentage areas cut and the percentage areas under plant cane for the different varieties.

N:Co.310 has probably now reached its peak as percentage area cut or area under cane, because the area under plant cane of N:Co.310 has steadily fallen from 55.4 per cent on the 30th April 195(5 to 41.7 per cent on the 30th April 1958. Area under plant cane has fallen during the same period from 19.5 to 12.8 cent in the case of Co.33.1 and 10.1 to 7.0 per cent for N:Co.339. Co.301 is now virtually disappearing as a commercial variety and the spectacular rise in area under N:Co.37(5 seems well warranted and is likely to continue for some time.

Yield from Different Areas The following districts had yields in excess of

40 tons cane per acre: Piet Retief 43.5, Lower Umfolozi 42.2 and Inanda 40.2 tons cane per acre. The increased yield of 1957 over 1955 which was also a good year was most marked in the case of the South Coast, but north of the Tugela it was only Eshowe that increased its yield appreciably. The following table gives the yields for the Industry and its main sub-divisions.

somewhat better but that is largely due to t h e fact that only the better fields arc left for older r a t o o n s . Table XI shows the yields obtained from p l a n t cane and ratoons.

Fertilizer Used During 1958 the Industry used the fol lowing

amounts of fertilizer:

The 1958 figures reflect substantial i n c r e a s e s in nitrogen and potash and also an increase in phosphate used.

Irrigation According to Central Hoard data there w e r e on

the 31st May 1959, 62,519 acres under i r r i g a t i o n out of a total area under cane on the s a m e dale 591,598 acres. This means that during t h e y e a r an additional area of 3,817 acres were put u n d e r irrigation.

At Pongola there were 13,687 acres out of a to t a l of 14,199 acres under irrigation and at N k w a l e n i 8,787 acres under irrigation out of a to t a l of 9,039 acres under cane.

The following table reveals the extent of i r r i g a t i o n in the Industry on the 31st May, 1959.

32

TABLE IX

Plant Cane and Ratoons Of the total area of 444,014 acres under cane on

the 30th April 1958 as given by the Special Census the following areas were under plant cane and ratoons.

TABLE X

Once again there is an increase in the area under third, fourth and older ratoons which averaged respectively on the 30th April 1956, only 11.2 2.8 and 0.9 per cent. Still the table shows, as does the average age of cane at ploughing out 6.4 years, that cane is most commonly ploughed out after the second ratoon. The Special Census returns for the period 1st May 1957 to 30th April 1958 still show a marked fall in yield from plant cane to first and second ratoons. The yield from older ratoons is

TABLE XI

Fertilizer usage during recent years is r e f l e c t e d by the following tables.

TABLE X I I

TABLE X I I I

TABLE XIV The increase in irrigation compared with the year

before, is clue almost entirely to more cane having

been put under irrigation by European planters.

33

S.A.S.A. Experiment Station, MOUNT EDGECOMBE.

20th February, 1960.

AREA OF CANE HARVESTED A N D YIELDS FOR DIFFERENT VARIETIES A N D RATOONS

(EUROPEAN PLANTERS ONLY) 1957-1958

Compiled f rom Union Department of Census Returns

YIELDS OF CANE HARVESTED BY DISTRICTS (EUROPEAN PLANTERS ONLY)

Compiled from Union Department of Census Returns

AREA OF CANE HARVESTED A N D YIELDS BY DISTRICTS (EUROPEAN PLANTERS ONLY) 1957-1958






RIVER WATERS By D. W. W. HENDRY

During July 1956 there was announced in a Government Gazette conditions under which river water may be used industrially.

One of these conditions is that water so extracted must be returned to its source in a condition approaching to that prior to its abstraction from the river, or, of such a standard of purity as may be set by the competent authorities.

We in the Sugar Industry, and in particular the South Coast are very conscious of this, as most of our rivers run into the sea at holiday resorts, and woe betide the factory should the river be polluted, accidentally or otherwise.

In a sugar factory, the evaporation station is the greatest user of water for condensing purposes, and this water is returned to its source. A fair amount is used for washing down floors, etc. and this is kept separate from the condenser water, as it usually contains sugar.

As an introduction, we shall describe the method of use and disposal at the Renishaw Mill of Messrs. Crookes' Bros. Water is conserved in the m'Pambinyoni River by a wall of sand, from which the pan condensers are served. Arrangements are made for part of the condensate which returns to the river below the weir to be pumped if necessary to a spray pond, from whence it returns by gravity to the dam. Surplus water and condensate flow downstream, and at a point some two hundred yards below the dam, water is extracted through well points for the quadruple. The hot water from the quadruple returning to the river with the pan condensate. It will be appreciated that there is always a certain amount of recirculation. Arrangements are possible to extract water from the river about a mile below the factory, but this only becomes necessary in the case of a severe drought.

As the result of an investigation by Mr. H. Bayers, a Senior Inspector of The Union Health Department, samples of water were taken, and analysed by Messrs. A. Harding-Kloot and Martin, Durban, as follows:

Colour all good

Sediment all slight

Turbidity all very slight

Results, expressed as parts per million

39

The points of sampling, were as follows:

1. About half a mile above/ fac to ry .

2. Overflow al dam.

3. Sixty feet below outlet of condenser water to river.

4. Below well points.

5. Old wooden bridge leading to Scottburgh (washed away).

6. At top of lagoon, near National Road.

Rainfall for month 0.24 inches.

The explanation for the increased dissolved solids in sample No. 6, is the fact that the river at times is tidal to a point above the point of sampling. The mill washings are pumped to a dam, and from there pumped on to a sandy field, whence it must eventually seep back to the river.

There is, however, a fact which the chemical analysis does not disclose, and this is an algal growth which is observed growing for a distance of several hundred feet down stream from the point of discharge of the condenser water to the river. There is no doubt that the temperature of the water stimulates the growth. As a matter of interest we observed the following temperatures along the growth area:—

Above condenser water discharge 22°C.

Condenser water discharge 34°C.

25 yards below 28°C.

100 „ „ 2G°C.

200 ,, „ 25°C.

The nature of the river bed is sandy. Water flows over a large area and is very shallow. There is an absence of rapids and turbulence to increase oxidation. The growth consists of greyish white tufts, resembling cotton wool, and becomes attached to stones or other objects below the surface of the. water. That it is a complex compound, there is little doubt. On exposure to the air it darkens, and if dried in an oven the residue is of a brown clay colour.

gg

A. G. Southgate, in his excellent book on Industrial Waste Waters, quotes:

"An important factor in the growth of fungus is the temperature of the water. The optimum temperature of a fungus called Sphaerotilus has been given by some workers as 20-25°C., and by others as 30°C. (It would appear as if we confirm this). It is quite evident that this fungus grows more rapidly in warm water than cold. This is an important point since under certain conditions growth of fungus in a polluted river may be stimulated bv the discharge of comparatively clean but warm waste water, such as condenser water.

In spite of the fact that increase in temperature stimulates growth it is often found that the length of a belt of fungus in a river below the point of discharge of a polluting liquid is greater in winter than summer. This, however, is due to the fact that in winter the organic matter in which the organism depends is not rapidly decomposing, and is therefore available over a long distance."

Fortunately the pollution (if we may use the word) is of a minor nature. Young mullet have been observed from time to time, which is surely a good

40

test of the pollution. However, we are c o n c e r n e d with a method of preventing the algal or b a c t e r i a l growth (if in quantity) from being c a r r i e d d o w n stream. The most obvious method is to let t h e condensate run along a furrow of say 8 0 0 y a r d s when the limit of the growth will be e x c e e d e d due to the drop in temperature, and a r educ t ion in t h e sulphur compounds, etc. effected. Long g r a s s in t h e furrow would be an advantage to t a k e up t h e products of activity.

The writer is indebted to Mr. Bayers , of T h e Union Health Department, for his a s s i s t ance , a n d hopes that the bringing forward at this t i m e of a paper such as this, will show that the Sugar I n d u s t r y is fully aware of its responsibilities.

41

MILLING AND OVERALL PERFORMANCES By Th. FOURMOND

There are 2 factors which govern the extraction of sucrose and its recovery:

(a) The fibre content of the cane

(b) The purity of the juice.

Some sucrose will be immobilised by the fibre of the cane and the impurities of the juice, however efficient our milling and processing techniques may be. It, therefore, stands to reason that efficiencies of sugar mills ought to be expressed in terms of crystallisable sucrose rather than in terms of total sucrose.

At present, the only formula available for expressing the total work performed by the sugar mills is the Overall recovery. However, this formula is merely a quantitative expression, as mill Extraction and Boiling House Recovery are closely related to the fibre content of the cane and to the purity of the juice. In other words, the Overall Recovery does not express the efficiency of the work as cane of low fibre content and of high purity juice is bound to yield a higher Overall Recovery than cane of high fibre content and of low purity juice.

In order to judge the quality or efficiency of the work performed by the milling and processing departments, we have to resort to the "Lost absolute juice per cent Fibre" and to the "Boiling House Performance" respectively. As those two figures cannot be linked up, we are at a loss to express the Overall Performance of sugar mills by a formula which would express both quality and quantity at the same time and directly as a percentage of the maximum obtainable in practice.

If to calculate the efficiency of the Boiling House, we adopt the arbitrary principle (Winter & Carp's hypothesis based on practical results) tha t one part of non-sucrose will immobilise so much sucrose, there is no reason why we should not apply the same principle to the milling process by assuming that so much absolute juice shall be retained by 100 fibre of the cane. We could, therefore, determine the quantity of crystallisable sucrose available in the cane, according to the fibre content of the cane and to the purity of the juice, and from which, the milling and overall performances could be calculated.

Obviously, we will have to assume a reasonable target for milling efficiency, which shall have to be based on practical results. For instance, we do know that some Natal mills have achieved a performance equivalent to some 30 lost absolute juice per cent fibre in the 1.957-58 crushing season. We could, therefore, fix the target, at present, at 30 lost absolute juice (until such time when this efficiency will

be surpassed) and endeavour to calculate the corresponding crystallisable sucrose available at such an efficiency.

It would be rather difficult, if not impossible, to determine or calculate theoretically, the sucrose extraction corresponding to an efficiency of 30 lost absolute juice per cent fibre. However, from practical results, we do find that there is a rather close correlation, irrespective of milling efficiency, between sucrose and juice extractions as clearly shown by the following figures which represent the averages of all Natal mills.

To prove that the correlation is close, irrespective of milling efficiency, we shall quote the figures of some mills with milling performances differing, widely.

As can be seen, the difference between sucrose and juice extraction is more or less constant, irrespective of milling efficiency.

We can, therefore, assume that, at the present standard of our milling technique, sucrose extraction is always higher than juice or brix extraction and can be approximated at +0.80 per cent. Should our milling technique improve and show a bigger difference, this present factor of 0.80 could be adjusted.

If this school of thought is correct, then it is easy to calculate the expected sucrose extraction (corresponding to an efficiency expressed as lost absolute juice per cent fibre) according to the fibre content of the cane. The purity of the juice being known, the crystallisable sucrose can be determined and from which the milling and overall performances would be calculated.

out if they arc of any value to enlighten us about efficiencies of sugar mills.

Let us apply these formulae to Empangeni's Tongaat's, Natal Estates' and Renishaw's figures for 1957-58 crushing season with a view to finding

In the light of such figures, we find t h a t Natal Estates has achieved the best efficiency in both milling and processing, thus ensuring the h o s t overall performance although Tongaat shows a. higher boiling house and overall recoveries. R e n i s b a w comes second best although Tongaat shows a higher mill extraction and overall recovery. J u d g i n g by the recovery standard, one would think t h a t . Natal Estates has''recovered (86.29 80.36) = 5.93 p e r c e n t more sucrose than Z.S.M. However, the p e r f o r m a n c e standard shows that Natal Estates actual ly r e c o v e r e d (99.20 - 91.80) = 7.40 per cent more s u c r o s e than Z.S.M.

Applying these formulae for the w h o l e sugar industry since 1954 we find tha t :

The efficiency of the Natal sugar mills h a s been progressing steadily, reaching its peak in 1956-57. Judging by the recovery standard, one w o u l d conclude that the 1958-59 season was t h e lowest . However, the performance standard shows t h a t it is second best and that + 0.88 per cent more sucrose was recovered than in 1954-55.

This could be quite misleading to the l a y m a n if he had to correlate mill extraction with fibre c o n t e n t of the cane and boiling house recovery with t h e pu r i t y of the juice. With efficiencies expressed as mi l l ing , boiling house and overall performance, e v e n the ordinary layman will understand tha t t h e h igher the figure, the higher is the efficiency as t h e y e x p r e s s efficiency directly as a per cent of the m a x i m u m obtainable in practice.

Is there any necessity, then, to f o r m u l a t e such expressions as milling, boiling house a n d overa l l performance? Some sugar technologists m a y c o n s i d e r that "lost absolute juice per cent fibre" and " b o i l i n g house performance" are good enough da ta to e x p r e s s the efficiencies of the sugar mills. H o w e v e r , if we consider that the entire staff of the sugar m i l l s and

42

Juice extraction, equivalent to a performance of 30 absolute juice per cent fibre can be theoretically calculated in the following way:

those connected with the sugar industry, in some way or other, are not necessarily sugar technologists, then we must admit the milling, boiling house and overall performances, being figures which express efficiency directly as a percentage of the maximum obtainable in practice, are far more logical and easier to understand as they express quality and quantity at the same time.

Furthermore, we all expect a clearer definition of millable cane in the near future. When a practical method for the determination of the fibre content is available for the analysis of individual cane consignments, the value of cane can then only be defined as "crystallisable sucrose available in the cane" and shall be determined according to the fibre percentage of the cane and the purity of the juice.

Therefore, for all these reasons, it would appear that there is some justification in introducing such expressions as milling and overall performances in our chemical control.

Summary The author suggests that formulae such as milling

and overall performances be introduced in our chemical control to express the efficiencies of sugar mills.

Such formulae, based on crystallisable sucrose, have an absolute value as opposed to the mill extraction and overall recovery, which, being based on total sucrose, have only a relative value as the fibre of the cane and the impurities of the juice will immobilise some of the sucrose of the cane.

From practical results, he proves that there is a close correlation, irrespective of milling efficiency, between sucrose and juice extractions, and the expected sucrose extraction, calculated from such a school of thought, makes the determination of crystallisable sucrose available quite easy and accurate.

REFERENCES Summaries of Chemical Laboratory Reports of South African

Sugar Factories - l954 to 1959.

Mr. Perk pointed out that sucrose extraction and juice extraction are not directly related. He said that the chief engineer of a factory was interested in the performance of his mills and therefore required the lost absolute juice per cent fibre figure, while the process manager was interested in the boiling house efficiency. He could not see that producing this overall picture was therefore of much value.

Mr. Fourmond replied that while the engineer might not be interested in overall performance, nevertheless mill extraction, being closely related to fibre per cent cane, he thought that it would

be valuable to convert lost absolute juice to mill extraction. He said we now used the figure overall recovery and therefore he could not see why we should not use an overall performance figure, which has an absolute value. The overall figure is the most valuable one to all concerned in the Industry as it represents the true efficiency of a sugar mill.

Mr. Beesley said that we were dealing with two different processes. In the boiling house we were endeavouring to separate sucrose from non-sucrose and therefore the Boiling House Performance figure which is based on these two quantities, has a certain validity. However in milling we were endeavouring to separate sucrose from fibre and our endeavours inevitably resulted in the extraction of non-sucrose as well but not to a controllable extent. Hence he objected to the use of a fixed difference between brix and sucrose extraction when the relationship between the two was as far as he knew, wholly beyond the control of the man in charge.

Mr. Fourmond said that brix-sucrose relationship depended on the purity of the juice. The steps leading to the calculation of the milling performance are confusing as brix is taken into consideration. It was, nevertheless, a true reflection of milling efficiency, even if expressed in terms of total sucrose. To prove it, one has to divide the actual mill extraction by the expected sucrose extraction as per suggested formula and it will be found that milling efficiency, thus expressed in terms of total sucrose, is the same as the milling performance's figure.

Mr. Beesley repeated that he did not think the difference between brix and sucrose extraction was a controllable quantity.

Mr. Fourmond said that sucrose extraction had to be correlated with fibre and the more sucrose one extracted the more brix one would extract as well. He said that in judging the performance of the mill we must look at how much sucrose was recovered from that which was available actually in the cane. The determination of crystallisable sucrose available is necessary for the calculation of the overall performance which, as already explained, is the most important figure to express efficiency. This is most valuable for both technical and financial control.

Mr. Beesley repeated that he could not agree to the use of a performance figure (for milling) which was based on a relationship that was uncontrollable. In support of this statement he quoted the difference between brix and sucrose extraction for four mills for the 1957 season. All these mills gave figures very close to 30 Lost Absolute Juice per cent Fibre yet the differences were 0.5(5, 0.92, 0.50 and 0.52.

Mr. Fourmond said that such differences as Mr. Beesley had mentioned could be explained by the

43

nature of fibre. Whereas trash would increase the fibre, where a lot of mud was concerned, then this was a different matter. One found the boiling-house efficiency was higher on the South Coast than it was on the "North Coast. Was this due to more efficient working or was it due to the cane? Comparing yearly figures one found that the boiling house efficiency varied without any change in technique. This must be due to the nature of the impurities in the juice. He had worked on various figures showing a milling loss of 30 per cent. In general, studying the figures, he had found an average of 0.8 per cent between brix extraction and sucrose extraction.

Mr. van Hengel doubted that the milling performance figure suggested, was a more practical formula than Lost Absolute Juice per cent fibre.

Mr. Antonowitz said that possibly only one sugar factory in South Africa would be interested in calculating crystallisable sugar in cane and that was Umfolozi, which used a similar formula (or modification), to evaluate cane. He asked Dr. DouwesDekker if his formula modifying the Umfolozi formula was in any way similar to that outlined by the author.

Dr. Douwes-Dekker replied in the negative.

Mr. Fourmond said the millers were interested in buying crystaUisable sucrose and not total sucrose therefore the value of cane should take into consideration both fibre and purity of the juice. It cane value was based on crystaUisable sucrose , the quality of cane delivered at the mills w o u l d improve automatically with, the result: that efficiency and throughput would go up, corresponding to a shorter crushing season. The farmers would a l s o score as they would be carting less fibre and impurities. It would definitely contribute to br ingdown the cost of sugar production.

Mr. Thumann said the comparison from mill to mill was in some countries made by the R e d u c e d Extraction figure. The fibre per cent of the c a n e was something which was eliminated in the a c c o u n t in such a formula.. If one had a hard cane with h a r d internodes extra power would be required as compared with low fibre cane. He supported Mr. F o u r -mond's idea. that there should be an overall f i gu re such as suggested by Mr. Fourmond. He h a d never known in any other countries boiling h o u s e recoveries to vary from week to week as they d i d in South Africa. He thought that a paper such as this might lead to further papers which would l e a d to a better system than the one we have now.

44

A DESCRIPTION OF THE CONVEYOR BELT INSTALLATION AT UMFOLOZI MILL

By G. G. ASHE

This paper is not intended to be a technical one, but more of a descriptive nature.

In these days of rising costs and labour shortage, cheap labour that is, one has had to study each department to see if the jobs that are being done manually could be substituted by the implementation of automatics, instruments or mechanical handling.

Instruments and Automatics have already made their appearance in the sugar industry in Natal and have been a great success, but this is a subject on its own and will not be covered in this paper.

Mechanical handling of materials is nothing new, and through the years the methods employed have: l)e:en improved upon from time to time and one of the methods being widely used these days is the rubber conveyor belt.

In modern sugar mills you will find carriers for cane, inter-carriers, bagasse, bagacillo, ash, coal, mud, sugar and bags, all being converted to endless rubber converyor belts.

In the Umfolozi Co-operative Planters Limited factory at Riverview there is a total of 5,181 feet of rubber belt conveyors installed.

Bagasse Carriers The bagasse is carried from the 84" mill on a 36"

wide belt running at 350 feet per minute and this handles 46 tons of bagasse an hour. This bagasse, together with the bagasse from the 66" mill are fed on to a 42" wide conveyor running at 350 feet per minute and running up at an angle of 20° and handling 66 tons of bagasse per hour. This belt discharges on to a chain and slat conveyor which feeds the individual boilers, the excess bagasse discharges from the slat conveyor on to a 24" wide belt conveyor, which carries the excess bagasse out of the boiler house on to a dump.

When there is a shortage; of bagasse: for any particular reason the excess bagasse: on the dump is shovelled on to a 24" wide conveyor belt, which runs underground under the dump, and this is conveyed back into the factory and on to the 36" wide belt coming from the 84" mill, and is fed to the boilers on the normal system of conveyors.

An added feature on this system is the installation of a photo-electric cell across the: 36" wide conveyor and immediately this belt should run empty due to a mill stoppage, a relay is started and the coal

feeders on the boilers, fitted with Spreader Stokers, are started automatically and coal is fired into the furnaces before the supply of bagasse: to the: boilers is finished. On the. resumption of supply, the coal feeders are stopped automatically. There is a time lag before the starting and stopping of coal feeders. An alarm is also sounded.

Bagasse is then fed to the boilers which are not fitted with spreaders, by means of the return belt from the stock pile.

Bagacillo Conveyor The collection of bagacillo is done by means of

wire mesh screens placed in the; chutes feeding the furnaces; the bagacillo so collected falls on to a 14" wide conveyor belt running at 370 feet per minute; this belt discharges the bagacillo into a chute feeding the mud mixer.

Mud Conveyors The filter station has five vacuum filters and

these discharge the mud or cake on to a short cross conveyor 14" wide. These in turn discharge on to a wide collecting belt running at 250 feet per minute. This discharges on to another conveyor which runs through the factory and is 14" wide and runs at 312 feet per minute, discharging on to a conveyor 1,440 feet long, the last 700 feet being 25 feet above the ground, and along this length the mud is ploughed off on to the ground below, forming a pyramid-shaped dump. This belt runs at 410 feet per minute and during the flood periods approximately 20-22 tons of mud is conveyed per hour. A second branch conveyor 40 feet high is at present being constructed to help handle the season's collection of mud, as very little of the mud produced is carted away during the season.

Coal Conveyors The boilers fitted with spreader stokers are also

provided with coal feeders and these feeders are supplied by a system of coal conveyors 1.4" wide and running at 270 feet per minute. This belt is fed from a coal storage bin and can handle up to 40 tons per hour.

Bag Conveyors The handling of bagged sugar has been completely

mechanised and no bag is lifted higher than two feet or carried further than 10 yards, and this is only done by the stackers. As Umfolozi is a Refinery

45

there are no less than five types of packages to be handled, namely Refined in 100-lb. pockets, 50-lb. paper sacks, and 48-lb. cartons; Mill White in 100-lb. pockets and 50-lb. paper sacks; Brown Sugar in 210-lb. and 100-lb. bags. During the past season there were times when all five packages were being conveyed at the same time.

In the bagging store the bags are fed on to the main 36" wide rubber conveyor, running at 150 feet per minute, by short slat conveyors which are controlled by an operator who keeps the main belt filled with bags by stopping and starting the slat feeder conveyors from a central control panel.

The main sugar store is some little distance from the factory and the bags are conveyed via an inclined covered belt which has to cross several train tracks before it enters the main sugar store just under the roof trusses in the one corner of the building.

The sugar store is 420 feet long by 180 feet wide and 30 feet from floor to underside of roof trusses. The inclined belt discharges the bags on to a 36" wide belt which runs along the 420 feet side of the store at a height of 28 feet. There are five 36" wide conveyors running at right angles to this long conveyor and they are spaced at. 70 feet centres and arc 26 feet above the floor. The general layout of the conveyors in the store is shown in Figure 1.

At each transfer point along the long conveyor is fitted a mechanical plough. This plough is nothing else but a vertically mounted conveyor belt with rubber strips bolted to the belt. This plough can be swung into position rapidly by means of an airoperated cylinder. When ploughing, it is at an angle of about 40° to the main belt. The vertical belt runs at about 300 feet per minute. When not in operation the plough is lying parallel with the main belt and bags continue on to the next plough point. When these bags get onto the cross conveyor they can be taken off at one of five discharge points along the length of the cross conveyor. Here again the bag is ploughed off with a vertically-driven belt, b u t this plough is mounted on top of a steel tower which is mounted on t ram wheels and runs on rails parallel with the cross conveyor. This tower, which is six feet square, has a spiral at 4' 6" pitch consisting of 3/4" diameter rods at 1 1/2" centres, and J" mild steel plate 6" high is welded round the edge of the spiral to prevent the bags from falling off. One. of these towers can be seen in the picture. On two sides of the tower there is a steel framework for supporting a platform.

The bags are ploughed off the cross conveyor down a chute, which is a t tached to the tower, and on to the spiral rods, they then gravitate to t he bottom where they are ploughed off by means of a round bar chute and this chute drops the bag on to a spiraveyor, which runs at right angles to the cross conveyor and tower, on the floor. The spiraveyor runs as far as the next cross conveyor. The s tacking boys take the bags off the spiraveyor and stack them on the floor equidistant on each side of the spiraveyor. As the stack grows in height the spiraveyor, which has universal joints, is merely lifted on to the next row of bags and rests on top of the bags. After several rows the platform on the side of the tower is raised to the next level together with the motor of the spiraveyor, and the round bar chute is moved up one turn of the spiral. The bags are then s tacked for several more rows, when the platform is moved up again, and so on until the stack is up to the underside of the roof trusses.

The mechanical plough is swung out of the cross conveyor and the whole tower is moved to the next position and is ready to start a new stack. Direct ly below the five top cross conveyors there are five 36" wide conveyors on the ground. These r u n in

46

47

48

the reverse direction to the top conveyors and the drive end of these conveyors discharges through a small hole in the wall on to a short chute which discharges into an S.A.R. truck standing in a covered siding. The rail level is 12 feet below the floor level of the sugar store.

When bags are being loaded directly into S.A.R. trucks and are not being stacked, the bags travel right to the bottom of the spiral tower and discharge directly on to the belt conveyor at the bottom. The bags travel along this belt directly into the S.A.R. truck and are only handled by the boys stacking in the truck itself. When a truck has to be loaded from a stack, the spiraveyor is placed on top of the stack and this time the motor is on top of the stack and the spiraveyor discharges into the spiral tower. The bag then travels to the bottom on to the belt conveyor below, and into the truck.

Flap gates with counters on are mounted on all conveyors to count the number of bags as they are conveyed along.

From the above it will be seen that a lot of conveying is done at Umfolozi and a lot of experience, some hard-earned, has been gained on belt conveyors.

Some of the advantages of belt conveyors are:

1. Low horse power compared to slat and other conveyors.

2. Properly trained belts can last many years in comparison to chains.

3. Maintenance is low.

The primary aim when designing, operating and maintaining a conveyor is to increase the life of the belting used on the conveyor as this represents from 40 to 60 per cent of the total cost of the conveyor. A point to remember is that a conveyor can be constructed from almost any kind of equipment and although they are abused and neglected, they will still continue to run over idlers that have not turned for years and are buried in the material it is supposed to carry, with resultant damage to the surface and edges of the belt, also overloading of the driving motor and tensioning the belt beyond the tension it is designed for. If a belt is mis-aligned it can cut its way through wood or even the steel framework of the conveyor and the belt will still go on running during the process and at the same time destroy the edges of the belt.

If a belt is treated properly it will respond with years of increased life and service.

The author wishes to thank the Umfolozi Cooperative Sugar Planters Limited for permission and co-operation in presenting this paper.

The President commented that Mr. Ashe had mentioned a belt conveyor system being changed to a slat-conveyor system. He thought: this a retrograde step as bagasse could be conveyed very easily by a belt conveyor at about one third of t he H . P . required for slat conveyors. One disadvantage was that the belt conveyor system was not as clean as the slat conveyor. He. noticed tha t the bagaeillo conveyor was running at ,'310 ft. per minute a n d he asked if any precautions were made to avoid windage . Mr. Ashe did not mention if the belt used for conveying filter mud was the ordinary standard design or if special material were used. He asked how high the bags of sugar in the store were; stacked, how many bags were stacked above each other find if he had any trouble with the breaking of bags.

Mr. Ashe replied that a slat, conveyor was not put i n t o replace a conveyor belt. This slat conveyor had always been there. The rubber belt was installed so that surplus bagasse could be stored, to fee l back when necessary on to the main bagasse; carr iers . It was founel that loading of bagasse on to the conveyor in front of the boilers was never used. The inclined belt conveyor now discharges directly on to the slat conveyor. The bagaeillo conveyor belt operates in a light metal cover, the belt being completely enclosed, and the only windage would be caused by the belt itself. When mud conveyors were first installed at Z.S.M. just ordinary belting w a s used but this ran for about 5 years without any t roub le . It was then replaced by another ordinary be l t but this began to give trouble because it was being bent in one way all the time. Oil resisting belts were then used but they also suffered from the same thing. At Umfolozi ordinary belts were put o n ; t hey ran for about three years and when they wore out they were replaced by ordinary belting again. Periodically these belts had to be turned over. The present belting was oil resisting and had g iven no further trouble.

Mr. Munro said that the oil resisting belt now installed had not been entirely successful. After this had been replaced recently by a more su i tab le belt this had run for two or three months wi thout any further trouble. Time would tell whe the r it could last better than the standard belt.

Mr. Ashe, in reply to the President, said, t ha t stacking of sugar bags started off at s tacking to about 25 ft. high. Because of the shortage of S.A.R. trucks they were made up to 30 ft. high a n d h a d no trouble at all with bags bursting. He h a d soon very much higher stacking than this elsewhere without trouble due to bags bursting. This app l ied to new bags, but probably trouble would rise wi th second-hand bags.

Mr. Hulett said that he was interested in t h e surplus bagasse storage arrangement and w a n t e d

49

to know if the bagasse was stored in the open or in a shed and roughly how many boys were required to feed the mill from the dump.

Mr. Ashe said the bagasse was now stored in an area with walls but no roof. When there was a long mill stop all the boys from the carrier went along to the dump and they fed the excess bagasse back into the system. A quantity of bagasse was required when the mill was stopped, but not as much when the mill was crushing. He said that about 18 boys went out on to the dump to feed the boilers with bagasse.

Mr. Munro pointed out that bagasse was not the only fuel relied upon during the mill stoppage. Some boilers were fed by coal. In the case of mill stoppage coal was first of all used. Bagasse was however required from the dump for the B. & W. boiler furnaces.

Mr. Davies said that he had seen belting such as that described running at Mhlume, and while he was impressed by the cleanliness of this system he was told that the cost was more or less the same as slat conveyors. When a belt conveyor was installed at Felixton to handle mud there seemed very little wear.

Mr. Ashe said that once the edge of the mud conveyor belt was rubbed off the belt seemed to

deteriorate very quickly. It was essential that it should be kept strictly in alignment.

Mr. Rault inquired how much mud per cent cane was handled. Could it ever rise as high as JO per cent on cane under the usual conditions of sulphi-tation or defecation?

Mr. Ashe said that at Umfolozi after floods the mud lying underneath the mill had to be shovelled out by labourers so at times it was probably more than 10 per cent.

The President asked, on the question of aligning belts if in the sugar store the action of ploughers did not tend to shift the belt over and if he used any self aligning guide pulleys to prevent the belt from running out of line.

Mr. Ashe said that when they first started the conveyor belt to the sugar store they had no self aligning idlers, but he had used these idlers at. 2 ft. centres. At the plow points the bags tended to move the belt over about an inch, but immediately the bags were removed the belt went back straight away. There was nothing at that point to cut the sides of the belt. At Umfolozi no aligning idlers were now used in the conveying of sugar bags. He said that the stacking towers illustrated were most easily handled, as being mounted on wheels, they could be shifted from point to point very easily.

50

MILLING CONTROL DATA WITH REFERENCE TO A MORE INTENSIVE METHOD OF SAMPLING AND ANALYSIS

Compiled by the Technical staff at Z.S.M. & P. Ltd., and presented by E. H. PHIPSON

Milling control data with reference to the extraction of individual units was introduced to the South African Sugar Industry in 1931.

A brief description of the methods of sampling is as follows:

1. The bagasse and juice are sampled each hour from only one mill, starting with the crusher and ending with the fourth mill after five hours.

2. One mill is sampled hourly over a period of four to eight hours.

3. One mill is sampled hourly over a period of 24 hours.

4. All the mills are sampled simultaneously over a short period of say one hour.

From the above methods it can be observed that each unit is only sampled during a portion of the available time during a week's run, in order not to unduly increase the amount of work beyond the limitations of the laboratory. Although mill work analysis figures as supplied by the laboratory are used to a great extent by the mechanical staff, at the best they are only produced weekly, and only give general indications of the work performed.

It must also be remembered that however accurately the laboratory staff do the necessary analyses and compute the results to two decimal places, the final result is nevertheless only as accurate as the degree to which the samples taken represent the cane being crushed.

A recent survey at Z.S.M. into the sampling and analysis done in our laboratorv, showed that 126 determinations were done daily to control the efficiency of the milling plant, compared with 428 other determinations necessary for factory control as apart from milling, with the exception of the last mill which is sampled for the hourly determinations of polarisation and moisture of the bagasse, and the analysis of the last expressed juice.

When it is considered that an 84-inch milling train complete with carriers, cranes, etc., costs approximately £500,000, and for such a milling unit to produce 100,000 tons of sugar from 119,000 tons of sucrose in cane, purchased from growers for approximately £1,700,000, the inevitable conclusion must be reached that the work done in our mill laboratory is not commensurate with the huge capital expenditure tor the purchase of sucrose and maintenance of the milling train involved.

We believe that the correct answer to t h i s is to establish a chemico-cngincering laboratory, quite apart from the ordinary routine control l abora to ry , in which all the time and equipment will be d e v o t e d solely to all classes of milling problems.

As a commencement to this scheme, it was dec ided to introduce a more intensive system of mill work analysis so that accurate results could be. p r o d u c e d daily, or even if so desired., for 12-hourly per iods . Such a ehemico-engineering laboratory shou ld be, for proper control, quite apart from the rout ine testing laboratory.

Building and Equipment required for the Bagasse Laboratory

A room 20 f t . x ! 2 ft. was fitted with su i tab le concrete benches to house the necessary e q u i p m e n t .

The equipment amounted to:

1. Two Gallenkamp drying ovens for m o i s t u r e determinations of the bagasse.

2. Ten hotplates.

3. Ten bagasse digestors and condenser l ids.

4. One hot water geyser of 30 gallons total c a p a c i t y .

5. One balance.

6. One Facit calculator.

7. The necessary accessories such as b e a k e r s , funnels, hydrometer jars, Brix hydromete r s and buckets, etc.

The cost of the above equipment and necessa ry innovations to the laboratory, except for one oven and the Facit calculator, amounted to £590.

Staff required 3 testers, 3 G.2 assistants, 3 labourers for s a m p l i n g

bagasse, 3 labourers for sampling juice and 3 l abou re r s which were shared with the mill laboratory.

Facilities provided for Sampling The imbibition trays were so arranged t h a t t hey

faced towards the last mill in the opposite d i r ec t ion from the travel of the bagasse leaving the d i scha rge rollers, in order not to upset the continuity of the imbibition system during sampling, and {giving ample space to enable the sampling of the bagasse during normal crushing conditions without in ter fering with the milling operations in any way.

Suitable platforms were provided near t h e discharge of each mill to enable the samples of b a g a s s e

51

to be taken as conveniently and. quickly as possible by means of long-handled tongs.

Period and method of Sampling

The bagasse and juice were sampled simultaneously every ten minutes from each unit, starting with the crusher and ending with the fourth mill.

At the end of each hour all the samples of bagasse from the respective mills were sub-sampled and taken to the laboratory for analysis. The amount of the bagasse sampled hourly across the discharge of each mill is of the order of 2.25 cu. ft. approximately, and the samples were taken up to a depth of 6 inches below the surface of the blanket of bagasse in the carrier.

The juice was sampled across the discharge roller of each mill.

In Table 1 the mill work analysis is shown daily for the month of November, together with the average results for each week.

In an attempt to show the difference between sampling and testing all the mills every hour and the old method of sampling and testing only one mill hourly, the analyses for the month of November have been averaged out in the same order that the sampling would have been done if the old method had been used, and the weekly results tabulated for comparison in Table I I .

A perusal of this Table shows that the more intensive system of sampling and testing gives more consistent results than the old system, where only about one-fifth of the samples were taken and tested, especially where fluctuations in the results can be shown, which are not due to poor milling but to other causes, such as improper sampling and fluctuations in the sucrose content and quality of the cane at the time of sampling.

In the comparison it must be borne in mind that all the sampling was done every 10 minutes (six samples per hour). If, however, only one sample was taken per hour, the fluctuations in the results might have boon more marked.

A similar comparison is given in Table III for a longer period of a month, when it will be observed that the differences will be less marked. In fact, the results will be seen to agree fairly closely, and if a comparison could be made over a season, the differences would probably be negligible.

However, waiting longer periods for results would defeat the whole object of milling tests, in so much as up to date information could not be given when required to be of any practical use.

We can quote an example where the individual milling data were of assistance in helping to rectify a sudden drop in the extraction.

During the night of the "28th September, 1959, an 18-lb. hammer went through the mills and the extraction dropped from 92.58 for the previous day, to 91.95 for the 12-hour night run mentioned above.

On instructions from the General Manager, the data for the extraction of the individual mill units were calculated for the twelve hours' run during the night, and showed that the total extraction up to and including the fourth mill was 88.99 compared with 88.59 the previous day. However, on comparing the results of the last or fifth mill, the extraction for the unit showed a drop from 3.99 to 2.96 and the efficiency (sucrose per cent sucrose entering unit) had dropped from 35.02 to 26.84.

The mill was shut down and investigations disclosed that the openings of the mills had been spread on one side due to the adjusting bolts having been forced into the bearings, and that damage had been done to the last mill trashplate.

The necessary repairs and adjustments were carried out and the mill was crushing again within two hours.

The following day the extraction was back to normal and showed 92.81 for the 24-hour run.

The data for the mill work analysis during the above mentioned periods are given in Table IV.

Conclusion This new method at Z.S.M. has proved itself to

be of the greatest assistance to engineers and technicians.

.ft

T A B L E I

THE DAILY AND WEEKLY AVERAGE WORK OF THE INDIVIDUAL UNITS OF THE MILLING TRAIN

for the Month of November, 1959

1

TABLE II

A WEEKLY COMPARISON OF THE RESULTS OF THE NEW AND THE OLD METHODS OF SAMPLING

CO

?*>'rsix3-!&v&'S' ***'*, H I

51

TABLE III

COMPARISON OF THE RESULTS OF THE NEW AND OLD METHODS OF SAMPLING

for the Month of November, 1959

TABLE IV

A COMPARISON OF THE WORK OF THE INDIVIDUAL MILLS DURING THE PERIODS WHEN THE HAMMER WENT THROUGH THE MILLS

55

The President in thanking Mr. Phipson for his paper said that the imbibition system had been altered to enable samples to be taken without interfering with the flow of bagasse. He had always been of the opinion that imbibition should be applied immediately after a mill so as to allow better absorption. Another point in the paper was that the extraction after the hammer went through was better than before.

Mr. Phipson agreed that the maceration should be applied as soon as possible after the bagasse emerged from the mill.

Dr. Douwes-Dekker explained that the point at which imbibition was applied would make very little difference to its absorption by the bagasse. He complimented Mr. Phipson on the work earned out at Z.S.M. He felt that not sufficient figures were usually obtained on the milling work, and that if Mr. Phipson's contribution brought about more intensive testing of bagasse and juices in the mills, this would lead to better understanding of the performance of the mills. The sampling of juice was not very difficult but to get a representative result from the bagasse was difficult. Bagasse-should be sampled over the total depth of the layer. Usually imbibition was applied to the bagasse as near as possible to the discharge opening with the idea that a better mixing of the imbibition liquid with the residual juice in the bagasse occurred. He was sceptical that this was so. When the determinations done in the past in various countries were examined, it was found that little liquid was absorbed except in the top few inches wherever the imbibition liquid was applied. He had seen figures showing that when imbibition was applied immediately before the next mill there was no difference in absorption. We can understand this if we realise that the imbibition liquid does not penetrate right through the bagasse. He said that a major cause of poor extraction was the poor mixing of imbibition and the residual juice in the bagasse.

Dr. Kerr on his recent visit here stressed that if we want to study the performance of the mills separately there should be adequate means for sampling and that meant the imbibition liquid had to be applied at such a distance from the discharge opening that we could sample bagasse. If we stopped the imbibition for a couple of minutes to try and get a sample of bagasse we could not be sure that we would get corresponding samples from each mill. This statement was made by a man from a country where much research was done in milling.

Mr. Beesley said that although imbibition only penetrated the top few inches of bagasse in the carrier, tumbling the bagasse off the end of the carrier into the next mill greatly improved the distribution of the imbibition. He agreed with Mr. Phipson that it was better not to stop the imbi

bition to (say) the fourth mill when sampling the third, as he had found that this could lead to sampling under dry milling conditions.

Mr. Rault wished to congratulate Mr. Phipson on the thoroughness of his control, and his Company for supporting him in providing the labour and expense, necessary for this valuable work. He had sampled a bagasse layer and found that the top few inches only were saturated by the usual water spray system, while at the bottom such was not the case, and so he had tried to split the water application on top as well as the bottom, but the final results were no better. His experience had been that the application of imbibition water immediately before entering the mill was paradoxically just as good as at any other place. The mixing of the imbibition with the bagasse seemed to take place at the beginning of the crushing of the next mill, under the special conditions of over-capacity crushing rates, where a layer of juice was always present on top of the bagasse blanket pulled in by the top and feeding roller.

Dr. Douwes-Dekker said that he was convinced that the proper mixing of residual juice with imbibition liquid would increase extraction very considerably. Mixing of the imbibition and residual juice was our main problem. As Mr. van Hengel had indicated, in Australia the use of pushers in the mills helped with the mixing of the imbibition and residual juice. He had tried experiments in the laboratory mixing bagasse by hand but unfortunately the tests were crude and would have to be done on a better basis.

Mr. Noel said that two factories in Mauritius experimented to try to effect better mixing of bagasse with imbibition and they installed a mixer about three feet after the water was applied to the bagasse, a second addition of water being made at that point. These two mills got better milling results as a consequence. He asked if it would not be better to use a new type bagasse sucrose extractor which would enable tests to be done in a few minutes instead of the large plant employed in the laboratory at Z.S.M.

Mr. Phipson agreed that that type of apparatus would be very useful but unfortunately he did not have this at the time. He had no difficulty in using the hot water digestors because he had ample staff. He thought that more than one bagasse sucrose extractor would be required with cold water extraction. Accumulating samples might lead to drying out.

Mr. Carter said that at Tongaat where there was a lengthy chute into the mills there would be ample space for both imbibition and bagasse sampling. He asked if it were possible to make the imbibition go right through the blanket of bagasse, would that

56

not lead to a better mixing of the imbibition and bagasse residual juice?

The Chairman said that some experiments of this nature could be carried out at Tongaat. Now they had installed an extra mill and it would be interesting to see if they could apply more imbibition than before.

Dr. Graham said that the sampling of the bagasse was the factor limiting the accuracy of the determination of extraction achieved by a mill. This error will generally be greater for the first mills in the tandem because the bagasse is not well broken up. For this reason, the method reported by Steward (Q.S.S.C.T. 1955 p. 273) for the determination of 1st mill extraction should be valuable because only the pol of first mill juice, second mill juice and mixed juice is required to calculate the 1st mill extraction.

Mr. Phipson thought that this would be an idea, but would not be practical.

Mr. Rault said that in his paper wr i t ten on his experience of 25 years data on this subject, he found that the last mill unit receiving water only a n d not thin juice, usually put up a higher ind iv idua l performance than the immediately proceeding ones. This had also been the experience- at T o n g a a t . He could not see a similar trend in Mr. Phipson's figures.

Mr. Phipson pointed out that at Z.S.M. t h e r e was no shredder. This would probably improve ext raction considerably if it were installed.

Mr. Noel said in Mauritius the cane was m o r e easy to mill than in Natal and their low fibre c a n e gave them a better opportunity to get high ex t rac t ions and also the mills ran more slowly than here . There was no common school of thought about milling in Mauritius where adjustment was done according to individual ideas. Many mills in Maur i t ius were working under capacity, but even at the; factory where the roller speed was up to 55 feet pe r minute and the mill was pushed to its maximum crushing rate, the moisture was still down to 48 per c e n t .

57

MINIMISATION OF THE HUMAN ELEMENT IN MILLING By D. J. L. HULETT

At Darnall it has long been our continuous problem The "Killer Plate"" Idea as applied to the to ensure that the milling train is kep well supplied Darnall Tandem with cane and that each individual milling unit is Sketch No. I. properly loaded so that a maximum throughput of This apparattus consists of a float over the feed to cane is obtained, consistent with the highest possible each mill which actuates through an adj ustable link extraction. mechanism, a standard Ford truck master cylinder.

The various mill engine: drivers found that if they A hydraulic tube connects this to the equivalent of continually maintained their engines at their maxi- whcel cylinder situated in the governor rod of the mum r.p.m., few chokes occurred and their work Bellis & Morcom Engine. This cylinder, inciden-became comparatively easy. However, this practice- tally was made in a few hours in. the mill workshop of course, caused a fall-off in extraction and it has and incorporates standard wheel cylinder "U" been management's problem for some years to elimin- rubbers. ate this practice. ,,,, , , , .. . r ,,

The mode of operating is as follows: The obvious answer to this problem seemed to be As the mill fills with bagasse, the float on the

the Australian "Killer Plate" idea and so a simple bagasse, blanket rises, forces oil from the master method of incorporating this with our Belliss & cylinder to the engine governor cylinder which in Morcom steam engine driving units was devised. turn collapses a return spring and so shortens the

effective length of the governor rod. This of course Then, to ensure that the milling train was fed increases the speed of the engine and the level of

with a continuous supply of cane it was considered bagasse is controlled. An interesting feature of the essential to automate the difficult task of feeding the system is the negative feed-back afforded by the cane knives. The operator at this post had to keep engine governor itself, for as the signal from the the main cane carrier supplied with cane to a certain mill opens the governor valve, the increase in speed depth by varying the speed of the auxiliary carrier of the governor weight closes the valve to a new feeding cane to the revolving cane knives. This he balance position. This makes the control extremely accomplished with a liquid controller in the rotor stable and no hunting at all occurs. circuit of the auxiliary cane carrier driving motor. However, this was not the only problem facing this

Ihe Control Mechanism tor the operator. He had to keep a continual watch on the Feeding ol the Lane Knives cane knife motor ammeter and regulate the cane feed accordingly, for an overload of this unit caused an Sketch No. 2. electrical trip and a ten minute delay while the motor The system used for the control of the feed to the was re-started. Furthermore, should the main cane cane knives is shown diagramatically in Sketch No. 2. carrier stop, then he immediately had to stop the It consists primarily of a liquid controller in the rotor auxiliary carrier for fear of a "pile-up" of cane in circuit of the auxiliary cane carrier motor which is the cane knife house. actuated by a standard Hagan power cylinder.

This Hagan unit is a device which varies the With the aid of a few relays, second-hand spares position of its plunger according to the air signal

from the centrifugal machines and a Hagan applied to it, i.e. with a signal of 3 p.s.i. the piston boiler control unit, a device was engineered to per- in remain at one end of its stroke but, with a signal form the duty of the cane knife feeding operator. f 15 p.s.i., it will remain at the other end; with After a few teething troubles had been overcome, signals" of between 3 and 15 p.s.i. it will take up this device turned out to be an extremely satisfactory relative positions between these two extremes. arrangement.

The remaining problem to the feeding of the tan- The air signal to the Hagan is derived in the dem lay in the diligent loading of the: auxiliary cane following manner: carrier and this has been solved temporarily by A 15 p.s.i. constant supply of air is bled through employing a more intelligent type of Indian to an orifice to a manifold to which are connected the supervise the cane yard. Ultimately it is hoped to Hagan controller diaphragm, a pressure gauge and operate the various carriers feeding the auxiliary the atmosphere through a Martenair solenoid cane carrier by remote: control from a central control operated valve. The manifold is also vented to the tower. This, it is felt must result in a better co- atmosphere through a needle valve' actuated by a ordination of the cane supply to the cane knives. float on the main cane carrier.

The method of operation of the control is simple and is principally an "on-off" control, in the cane knife motor circuit there is a. current transformer energising a relay which in turn supplies power to the Martenair valve. This relay is arranged so that in the event of the cane knife motor current reaching a certain high value, the Martenair valve will become energised and open, venting the manifold to the atmosphere and reducing the pressure on the Hagan control to less than 3 p.s.i. This causes the plunger to travel to its extreme position and lift the dippers of the liquid controller out of the electrolyte. This of course, stops the auxiliary cane carrier motor and the feed to the cane knives. The relay automatically opens when the load on the cane knife motor reaches the low limit setting and the pressure on the Hagan builds up, restarting the feed.

The interlock for the main cane carrier consists of a centrifugal switch connected directly to the drive motor shaft. This switch closes when the motor ceases to revolve and energises the Martenair valve.

The float on the carrier acts as an overrider to the control so that should the main cane carrier become fully loaded, the float opens the needle valve, reduces the pressure in the manifold and so slows down the feed. This control mechanism was in operation throughout the last two months of the crushing season and appeared to control the feeding of the cane knives far better than the operator had done prior to its installation.

Unfortunately, as a sugar mill is a practical concern, it is not always feasible to try out one experiment at a time in order to determine exactly what benefit is derived from each arrangement. The performance of Darnall Mill did in fact improve after fitting the various controls but it is difficult to attribute any improvement directly to the installation of these devices. However, it is felt that, provided the arrangement seems reasonable as an improvement and is no disadvantage to the process, it should be incorporated.

A direct saving can be attributed to the control set-up and that is the possible saving of six units of iabour on each shift, namely—two top roll boys, three engine drivers and one cane carrier driver.

The President, Mr. Bentley (in the Chair) said that the type of individual one had to operate a milling train tended to somewhat over-control it. He thought it very noticeable that most mills seemed to run far better during the late hours of the night and early hours of the morning than during the day. The reason for this was that when most of our engine drivers and mill boys saw a senior official walking around they pretended to be doing something and

were inclined to get the mills running f a s t e r or slower, whereas at night when, there was little super-vision, they set the engines and sat down and everything ran much better. Control of t h e sort described, he felt, would assist greatly in e l i m i n a t i n g the human error in milling that did, without doubt occur. He said his only query was that on t h i s par-ticular type of engine drive he had noticed t h a t in spite of setting the remote control in any given position except when running at full speed there was a tendency to hunt over a wide range.

Mr. D. J. L. Hulett said he did not find t h i s . He had more horsepower in his engines, but wi th Ton-gaat engines, which normally have a top s p e e d of 420, if one tried to control them they will h u n t 20 r.p.m. either way, and that was a little unsatis-factory. In his case, where no hunting at all o c c u r r e d , it was probably due to the control, system. The engine speed did vary before he had this control. You set the engine speed at 350, as the load varied the engine must slow down before t h e governor could open and let in more steam. W h a t amount you did get in cycle hunting through a d a p tion of this control was clue to the fact that at t i m e s there was a tremendous amount of feed p r e s s u r e . He said they had 650 h.p. engines against T o n g a a t ' s 450.

Mr. van Hengel said he would like to ask Mr. H u l e t t why it was that when he took different m e a s u r e s at the same time he could not just show any i m p r o v e ment or slight improvement in extraction and o t h e r milling data. Did the introduction of this s y s t e m lead to a better mechanical efficiency of his m i l l i n g tandem? Did this lead to less breakage, etc. on h i s mill?

Mr. D. J. L. Hulett said that during the last m o n t h they had one of the highest mechanical efficiencies at Darnall, except for the last week when one o p e r a tor forgot to oil the pump on the engine and t h e y had also ran on one boiler, and again were r u n n i n g slowly and stopping for cane. Although they h a d a fine mechanical efficiency he could not say it h a d anything to do with the control. He said. the extraction went up. He said they put on 55 p e r cent imbibition and that also improved the e x t r a c t i o n . He said he could not find strictly that the c o n t r o l had improved efficiency, but he thought it h e l p e d , and it cost very little.

Mr. Beesley said that he had noticed that c e r t a i n canes tended to fluff-up and others to c o m p a c t , when knifed or shredded and asked Mr. H u l e t t if he had experienced similar canes at Darnall. If so, whether he had found that the float operat ing the needle valve working the Hagan controller ( S k e t c h No. 2) tended to give variations in crushing r a t e ?

Mr. D. J. L. Hulett said that when he first thought about this whole scheme he visualised this float

58

59

being the actual control. As you could see it was arranged to work a needle valve. He said he had to abandon the use of the float and adjust on a purely overriding control. It only worked when the carrier filled up too high. Speed control was mostly by the cane knife current which was set to shut off at 350 amps and start again at 250 amps. He found that he was not getting enough cane through and had to increase to up to 450. Average cane knife current is 350 amps. A big bundle of cane sent the current higher than 350 amps. If the cane is loaded properly on the auxiliary cane carrier the drive would reach full speed. When the feed was regular the cane knife current would hover at 250 to 350 amps until one got a big bundle coming along and then it stopped the carrier. When the cane is fed into the auxiliary cane carrier in big bundles the control does not work well and one got a few humps in the main cane carrier. This would pile one bundle on top of another.

Mr. Sharvcll asked for some idea of the range of speed over which the mill engines operated under the influence of the "Killer Plate" controls .

Mr. D. J. L. Hulett said each individual mill did not vary enormously. No. 5 mill speed did not vary from say about 340 to 375 or 400 r.p.m. This depended on how the mill was set. It would keep about that speed and if not pulling its weight it was then tightened up to bring it into the operating

range. He could not operate below 280 r.p.m. The governor at that speed had no control. The governor weights were right in and had no control then or at a very high speed when they were right out. Mr. Hulett said he had seen a mill operating under 280 on the hydraulic alone without any help from the governor but it was very unstable and hunted badly.

Mr. W. H. Walsh asked whether the limitations of the present governor affected the control. He said he felt the control would be more effective if another type of governor, which was more stable, was used.

Mr. D. J. L. Hulett said they had governors driven by weights but the oil controlled type was such a nuisance that they went back to the old governor.

Mr. Gunn said he believed that Darnall had put their shredder in front of the 1st mill and he wondered whether Mr. Hulett had any ideas of regulating his main cane carrier and controlling it by the shredder motor.

Mr. D. J. L. Hulett said the problem was to keep the feed to the shredder constant. They had a control through the cane knife motor. The shredder motor was a synchronous motor but the ammeter movement was so much that it was impossible to employ this for control.

63

A RECORDING ROLLER LIFT INDICATOR By A. VAN HENGEL

Nowadays, most of the factories use lift indicators to guide the mill operators in adjusting their roller speeds. However, it is not only important to keep the roller floating all the time, but as pointed out earlier,1 great importance should be attached to the fact that the rollers should remain as closely as possible to a predetermined position, as any movement of the top roller affects the work ratio.

The necessity of keeping the top roller at its predetermined position prompted the S.M.R.I. to develop a reliable means of continuously recording the lift of a top roller. Further important considerations were the advantage of having an instrument that enables the staff to carry out a "from minute to minute" control, the fact that the S.M.R.I. intends to propose the initiation of a system of "Mutual Milling Control" and the fact that some investigations concerning moisture in final bagasse were being conducted at Illovo.

As direct recording was regarded as being impossible, due to the constant vibrations of a mill, an electrical system was selected.

The "electrical" readjustment was accomplished by placing potentiometer R4 in position 2 and moving R3 so that the mA meter just gave a full scale reading. It will be clear that the reading with the potentiometer in position I will be zero.

Since it was considered essential to make measurements both at pinion and at pintle sides of the mill, a six-point recording mA meter was transformed into a two-point recorder by short circuiting the. terminals.

Basically, the linear movement of the roller plus bearing is transformed into a rotating one by means of a chain, sprocket wheel and contraweight (all very cheap Meccano parts). A movement of 0.5" of the roller causes a rotation of 29° of the sprocket wheel, and as a potentiometer normally turns 290-300° between its extreme positions, the potentiometer was linked with the sprocket wheel by means of a 1:10 gear ratio (also Meccano parts). Hence, a movement, of 0.5" of the top roller caused the potentiometer to rotate over its full range.

A rectifier with voltage stabilisation was built in order to make the reading independent of possible variations in the voltage of the mains, and the circuit decided upon is shown in Fig. I (see also Appendix).

The "mechanical" readjustment of the chain over the sprocket wheel was accomplished by means of a rigging screw. The degree of linearity of a normal production potentiometer is always within range of 10 per cent and as the average lift would

64

be 0.25", the possible error is 0.1x0.25=0.025" either side, a very small error.

The apparatus was installed at Illovo on the last mill. Ideally, a mill should be set in such a way that the roller lifts just 0.25" all the time. The scale of the instrument was calibrated from 0" (0—mark) to 0.5" (100—mark). It was felt that it was of some, advantage to install the potentiometer R4 in such a way that at the pintle side the potentiometer was in position 2 and at the pinion side in position 1 when the top roller was at rest. This was simply achieved by reversing the direction of rotation of the sprocket wheels.

By doing this, only one line in the middle of the graph paper would result if the roller was lifting 0.-25", any deviation being shown as two different lines and the further the roller moved from its ideal position the greater would be the distance between the lines.

The apparatus worked for a few months at Illovo and enabled us to ascertain the work opening of the last mill with a very high degree of accuracy. This accuracy made it possible to draw certain conclusions that will perhaps lead to a better understanding of the actual milling process.

Acknowledgements I would like to express my gratitude to Mr. J.

Bruijn of the S.M.R.I. who assisted me with the design of a suitable circuit for the recorder. Further I thank Mr. E. E. Beeslev of Illovo for his careful handling of and keen interest in the equipment. Finally, I thank the Management of Illovo for giving me the opportunity of trying the recorder on their mill.

APPENDIX A characteristic of the mA meter is that it requires a maximum

current of 0.1 mA for full scale reading and its internal resistance is 260 Ohms.

The internal resistance was increased by R5 to 10K Ohms and hence a voltage of =1 V is necessary between point 2

and earth. R4 should be as small as possible to keep the influence of the internal resistance of the mA meter itself down to a minimum and therefore is 50 Ohms. The substitution resistance for (R4 + R5) is therefore:

This deviation is so small that it is negligible.

The total drop in voltage over R4 and R5 must be 85V and as R4 is 50 Ohms, R3 must be:

(R4 + R3):85 = R4:l 85 R4 = R4 + R3,

R3 = 84 R4 = 84 X 50 = 4,200 Ohms

Hence an adjustable potentiometer of 10K Ohms will be very suitable.

R1 is preferably 2,000 Ohms and therefore R2 can be calculated from (R1 + R2):R2 = 250:85 as the tension at the "85 A1" voltage stabilizer should be 85V.

R2 = 2,000 X 85 = 1,030 Ohms. A readjustable resistor of 2,000 165

Ohms will therefore be necessary.

REFERENCE 1 A. van Hengel and K. Douwes Dekker: Some Notes on the

Setting and Operation of Mills, Proc. S.A. Sugar Technologists' Congress, 32, 1958.

The President, Mr. Bentley (in the Chair) stated that Mr. van Hengel had given a very clear exposition of apparatus which would be advantageous to use in most sugar factories. The equipment could easily be made. His objection was that t he equipment was made up with a view to minimum expense while he thought a more robust piece of appara tus would be more suitable.

Mr. Beesley wished to comment on Mr. v a n Hen-gel's paper. He said that the instrument as described, was in fact a prototype and that a lot of development still had to be done to it. For instance the recording part of it was not cheap, costing about £700, but there were other instruments which could be used instead of this very expensive i t em. In any case, he knew that Mr. van Hengel had a lot of ideas he intended developing in the future. Commenting on the usefulness of the instrument , he said that he thought it was basically a research tool, but that it had a lot to recommend it to the mill engineer, as for instance, he could get information on the correctness of his engine speed vs. mill settings much quicker than normally. It would be particularly useful where mills were coupled together and it was desired to run the second mill at slightly higher lift than the first. F inal ly he thought this type of instrument could be used to control the speed of the next mill in the t andem.

65

Mr. W. H. Walsh asked if any of the older hands at Illovo could tell us of an apparatus which was developed some years ago by Mr. Wheeler. He asked how the apparatus now described compared with that put up by Mr. Wheeler. He realised a recorder would be rather a big problem, but Mr. Wheeler certainly accomplished something with his previous apparatus at Illovo. With the advent of the turbine driven mills and varying roller speeds, the piece of apparatus, such as described by Mr. van Hengel, would be most useful. In the case of a turbine drive it was important to keep speed at its most economical pitch.

Mr. Beesley said that all the mills at Illovo were equipped with lift indicators and he thought that these were what Mr. Walsh had referred to.

Mr. W. H. Walsh said this was not what he meant. It was somewhat similar to the present instrument in that it was connected to a potentiometer. Mr. Wheeler's apparatus was similar to the one described by Mr. van Hengel except that it did not have a recorder.

Mr. van Hengel said that in reply to a second question by Mr. Walsh there was no improvement in the moisture content of bagasse or extraction. The recorder, however, enabled them to determine that the maximum possible pressure was placed on the rollers. He indicated that pressure by itself was not the only thing to be considered.

In connection with turbine drives, he said that in this case it would be more necessary to know how to set the mill so that the top roller could run in a predetermined position.

Mr. Rault said that when Mr. Wheeler went to Illovo there was an immediate increase in extraction. He did not know exactly what Mr. Wheeler had done to accomplish this, but he did show results.

Mr. Beesley said that the best extractions achieved at Illovo were those obtained in recent years.

Mr. van Hengel explained that it was not possible to print the dots shown on the recorder. These however showed that the movement of the top roller was most irregular. He said the recorder also indicated that on the pinion side the lift was consistently much higher than the pintle side and the roller was never really level.

Mr. D. J. L. Hulett said we could tell with this machine that the roller was riding on the bagasse. On every mill they had at Darnall there was a recording pressure gauge. This was worked from the Munson hydraulic pressure accumulator. He said it had been mentioned that a recorder could be used to govern the engine speed. He said this has been done at Darnall from the hydraulic system. He said this worked, but perhaps not very satisfactorily.

Mr. Rault said that a mill, being essentially a machine for crushing a bagasse dry, the moisture content was a sure criterion for judging its mechanical performance. He would accordingly like to know the variation in moisture content of different parts of the bagasse, when it was found that the roller was not level.

Mr. Beesley said that for the short period during which the recorder had worked on the last mill at Illovo, the gear side had shown more lift than the pintle side, however tests on the bagasse from both sides had shown no difference in moisture.

Mr. van Hengel said that over 50 lbs. per cu. ft. escribed volume was the ideal volume to be aimed at but at Illovo they only achieved some 44 lbs.

Dr. Graham said that if the recorder showed that the top roller was continuously moving up and down as was found at Illovo, this would mean that the feed to the mill was very uneven. This could contribute towards the low extraction obtained.

Mr. van Hengel said the amount of fibre could alter considerably in spite of the fact that the amount of cane being fed into the first mill might appear constant. So, constant feed of the milling train is not the same as keeping the rollers in their predetermined position. In order to ascertain how great the variations could be, a recorder would be of great assistance.

Dr. Douwes-Dekker said that a statistical interpretation of J ava data had shown that best results were obtained when a first mill was fed at a rate of 35 lbs. of fibre/cu. ft. e.v. and a last mill at 55 lbs. Obviously these figures had to be checked for Natal conditions. To calculate the e.v. over a period of say one week, it was necessary to know accurately the average lift of the top roller over that period. The lift indicator described in Mr. van Hengel's paper was a first a t tempt to find the average lift. It had been applied at the last mill at Illovo and it was found that the fibre content of the final bagasse did not reach the expected value. But neither had it been possible to reach the target figure of 55 lbs. of fibre/cu. ft. e.v. At 40 lbs. the top roller was already lifting freely and pushing more fibre into the discharge opening would only result in a higher lift, not in a higher figure for lbs. of fibre/cu. ft. e.v.

Why 55 lbs. could not be attained was not clear, it might be due to a too large mill ratio, or to the grooving of the rollers. Obviously more data are required and the S.M.R.I, had approached the milling companies asking whether they would be prepared to co-operate in the introduction of a mutual milling control system. The replies from the factories had been quite satisfactory.

66

The main points of the new system would be the determination of the essential data indicating the performance of each unit of the tandems of the participating factories.

As such we had to see in the first place the regular and accurate determination of lbs. of fibre/cu. ft. e.v. and of the fibre percentage of the bagasse of each mill. This required recording instruments.

The one used at Illovo was too expensive to be used in large numbers and the S.M.R.I, was now trying to build a much cheaper type. Until a satisfactory instrument had been developed, we h a d to ask the participating mills to be patient. It was, however, the intention to convene a meeting soon of the chief engineers and chief chemists to discuss the whole problem.

67

DRAWING A STRAIGHT LINE THROUGH POINTS ON A GRAPH

By W. O. CHRISTIANSON

It is common practice for many people to feel that they can best illustrate the correlation between two sets of figures by plotting points on graph paper and then drawing a straight line through these, points to express the average relationship between the two sets of data.

There is of course more than one way of drawing this line, the most popular being, apparently, by guessing where it should go. Unfortunately this guesswork can be most inaccurate and so the writer now presents a simple mathematical way of determining such a line with precision. The calculations involved, may appear rather formidable but with a calculating machine they do not take very many minutes.

This mathematical process is called the method of least squares, the line so determined being such that the sum of the squared deviations between this line and the plotted points is at a minimum. The straight line so determined is called a 1st Order Polynomial Curve and the method of calculating this curve is conveniently illustrated by taking an example.

For such an example let us first of all examine the relationship between the Sucrose per cent Cane, and the Sucrose per cent Bagasse figures shewn in the 34th Annual Summary of Chemical Laboratory Reports, Table V(1), from which the following data are taken.

Similarly for Sucrose per Cent Bagasse we have:

For our purpose we, require to calculate for each set of the figures the total and mean (shewn above), sum of the squares of the deviations from the mean, and lastly the sum of the products of the corresponding deviations from the means.

Thus for Sucrose per cent Cane we calculate:

Total sum of products of deviations from means

From the above we can now calculate a Regression Co-efficient and a Regression Formula to express the average relationship between our two sets of data.

The Regression Formula giving the average Sucrose per cent Bagasse for any value of Sucrose per cent Cane is calculated thus:

Sucyose Sucrose per cent per cent

Cane Bagasse Extraction

May 11.60 2.16 93.33 June 12.48 2.34 93.33 July 13.26 2.53 93.25

August 14.03 2.66 93.21 September 14.12 2.06 93.21 October 14.07 2.71 92.93

November 13.43 2.67 92.58 December 12. 80 2.56 92.38 January 12.73 2.51 92.01

Total 118.52 22.80 836.23 Mean 13.1689 2.5333 92.9144

Hence for 11.00 Sucrose per cent Cane we calculate:

and for 14.00 Sucrose per cent Cane we calculate:

and so on.

We can now draw our straight line through points so calculated, as illustrated in the "graph" or more properly, the "scatter diagram", shewn in Figure 1.

68

But, and this is very important, has this line any meaning or in other words, does it reveal any real correlation between Sucrose per cent Cane and Sucrose per cent Bagasse? This question we can answer by calculating the Correlation Coefficient, (r), which is given by the formula:

By referring now to Prof. K. A. Fisher's table of r (2) we find that for 9 pairs of observations:

our Correlation Co-efficient of .9312 is therefore highly significant and we must therefore conclude that there is a definite association between Sucrose per cent Cane and Sucrose per cent Bagasse in the data we have dealt with.

A similar examination of the Sucrose per cent Cane and Extraction figures in Perk's table1 leads to the Regression Equation:

The line drawn from this formula is shewn in the scatter difigram illustrated in Figure 2.

The Correlation Co-efficient for the Sucrose per cent Cane and Extraction figures is however only 0.057, which from Fisher's table of "r" is not significant. We therefore must conclude that there is no significant association between Sucrose per cent Cane and Extraction in the data examined, even though the line slopes. (This slope has been accentuated, of course, in Figure 2 by suitably spacing the ordinates).

In the above the writer has taken all legitimate short cuts and presents a simple mechanical process as simply as possible. This has been done in the hope that the use of this mathematical method will find more general use, and also for simplicity, he has purposely left out a lot of detail and discussion which would only tend to confuse.

REFERENCES

(1) Perk, C. G. M., 34th Annual Summary of Chemical Laboratory Reports. Proc. S.A. Sugar Tech. Association, 33, 1959.

(2) R. A. Fisher: Statistical Methods for Research Workers.

Dr. Douwes-Dekker asked if the figures extracted from Perk's table were those of one factory or the averages from all factories.

Mr. Christianson replied that Perk's Table V showed the monthly averages for all factories in South Africa.

Mr. Beesley said that he was glad tha t Mr. Christianson had brought the method of least squares to notice, as it was applicable to many problems in a sugar factory, for instance, he had applied it to find molasses purity, and to Nutch puri ty vs. C massecuite purity, to determine the average C. massecuite puri ty that would give best molasses extraction for a given C massecuite station.

He said that while appreciating that Fig. II was meant mainly as an illustration of lack of correlation, he felt that the choice of variables (extraction vs. sucrose per cent cane) was most unfortunate, as he believed that sucrose per cent cane did have an effect on extraction. However in investigating the relationship it was necessary to s tudy the effect of both sucrose per cent cane and fibre per cent cane at the same time, and preferable to follow the course of individual mills rather than the whole industry. He had developed a formula along these lines and it appeared to apply quite well to Tongaat, Natal Estates and Renishaw.

Mr. Christianson said that as sucrose per cent cane and sucrose per cent bagasse were so intimately associated one could not expect to obtain an increase in extraction with increase in sucrose per cent cane.

Mr. du Toit stated that the method outlined in the paper should be applied more generally than it was at present. He agreed with the results of the two correlations shown in the paper and said that Mr. Beesley's multiple correlation should be checked by calculating partial correlation coefficients. After eliminating the effect of fibre, he was convinced that no correlation would be found between sucrose per cent cane and extraction.

69

Mr. Rault enquired if it was accepted that each part of fibre would have the same effect on extraction, and if in the comparison of sucrose per cent cane with extraction, any other factors were taken into consideration.

Mr. Christianson replied that quality of fibre as well as the quantity of fibre must be taken into con

sideration but there was no doubt however that on

the average the quantity of fibre was inversely asso

ciated with extraction. The comparison of sucrose

with extraction shown, in Figure 2 was a simple

straightforward correlation and no other factors

were taken into account.

70

MIXED JUICE SCREENING AT DOORNKOP By V. S. WINTERTON

The mixed juice from the mills is first screened through a cush cush screen (35 sq. ft. with 169 X 1/32" and 81 X 1/16" round punched holes to the square inch) and the fine bagasse particles are removed afterwards by means of Hummer juice screens.

Although our method of treating mixed juice is quite conventional, the Hummer type of juice screen is still fairly new to this country and some remarks on our experience with this type of screen may be of interest.

The Hummer screen differs from the types of screen more usually encountered in that vibration is confined to the screening surface only. The vibrator has two major assemblies, the magnet and the armature. The armature is "floating" but positioned in relation to the magnet by means of

two counteracting coil springs and a steel guide strip. The magnet receives electric impulses of a fixed frequency and alternatively attracts and releases the armature. When attracted by the powerful magnetic field the armature snaps upward unti l the motion is stopped by metal to metal contact of the armature striking plates and s ta t ionary wearing plates mounted on the vibrator head. A " U " shaped push bracket transmits the force of the upper spring to the armature, returning to its lower position. The gap between the upper and lower positions determines the amplitude or s t roke of vibration, which can be adjusted by a. hand-wheel on top of the vibrator. The armature post, which is coupled to the armature, carries the vibration through the screen bracket to the vibrating str ips which hold the wire screen cloth.

Cross section of Hummer Vibrator

71

The vibrations resulting from the up and down movement of the armature are in the vertical plane only and unlike vibrations produced by eccentric circular motion, there is no tendency for the screen to act as a conveyor. It is hence necessary to rely on gravity for the discharge of solids from the screen and for this reason the screen has to be mounted at an angle of about 30° to the horizontal.

Placing a screening surface at an angle reduces the effective diameter of the screening aperture, and hence the capacity of the screen. Particularly with very fine screens, where the diameter of the aperture is only slightly larger than the diameter of the wire, a 30° angle will reduce the effective screening area to about 70 per cent of the actual open area. [Apparent aperture = (0.866 X width opening)—(0.134 X diameter of wire) at 30°].

To overcome this decreased capacity effect, without the risk of reduced screening efficiency associated with a larger aperture, the manufacturers have designed a special feed box from which the juice should spray on the screen at right angles to it. This pressure feed box, fitted with tapered holes in the bottom, covers the first half of the screening area, the latter half of the screen acting as drainage area on which the separated bagacillo is allowed to drain.

The first Hummer screen 5 ft. by 4 ft. was supplied with a 60 X 40 mesh woven screen 0.009" wire thickness and without the pressure feed box. The 60 x 40 mesh screen was chosen merely because our existing horizontal screen was equipped with such a screen. The feed to the screen was by means of a weir covering the full width of the screen.

As could have been expected for the reasons outlined above, the capacity of the Hummer screen was far below that expected, although the screening efficiency, judged qualitatively, was greatly superior to the old horizontal screen, equipped with an identical screen.

In an effort to increase the capacity of the screen, we made up a feed box to the manufacturers design, containing 84 holes tapered from ¼" to 1¾". The capacity of this feed box by far exceeded the capacity of the screen and even after blocking half the number of holes the screen capacity was still no larger than when using the weir type of feeder, which was not surprising as the pressure jet feeder box was unable to give jets of juice and resulted in a less efficient feed than the weir feeder.

The next step to increase the capacity of the screen was to fit a 40 X 40 mesh screen with 0.0076" wire thickness. The weir juice feeder was put back as it was felt that the jet feed box would have to be completely rebuilt to reduce its capacity within the

range of the capacity of the screen. The capacity of the screen improved to about 1¼ to 2 tons juice per sq. ft. screen area per hour, without impaired screening efficiency.

At this stage a second Hummer screen was obtained to which a 30 X 30 mesh screen was fitted with 0.0076" wire thickness. With this type of screen the capacity of the screen improved to 2½ to 3 tons juice per sq. ft. screen area per hour using the weir juice feeder, at the expense of a very slight reduction in screening efficiency. With the 30 X 30 mesh screen the mixed juice sent to process contains less than 100 p.p.m. bagacillo, determined by filtration through a filter paper. We are satisfied with the screen capacity at this efficiency and do not propose to experiment further with the jet feed box as we suspect, from theoretical consideration, that any increased capacity which may be achieved will be at the expense of screening efficiency.

Comparing the Hummer screen with the horizontal screen previously used by us, we found the following points in favour of the Hummer screen:

1. The life of a screen by far exceeds that of those used on the horizontal screen.

2. Due to the inclined position of the screens, steam cleaning can be carried out much more effectively.

3. The whole vibration mechanism is completely enclosed, needs no lubrication and can be easily adjusted to suit various conditions whilst the screen is in operation.

The only disadvantage of the Hummer screen is its somewhat noisy method of operation.

The President (in the Chair) asked Mr. Carter to give more details about the quantity of juice being screened.

Mr. Carter replied that they started off with the 60 x 40 mesh, which choked up very quickly and they could pass not even 30 tons of juice per hour through the screen. When they went to 30 x 30 mesh they could take 55/60 tons of juice per hour.

Mr. Thumann asked if there was any truth in the statement that the screening was more efficient near the vibrating element and that further away it was not so effective.

Mr. Carter pointed out that the screen vibrated over its entire area. As mentioned in the paper most of the bagacillo was removed in the first foot or so of the screen. There was a centre strip down the middle of the screen which tended to throw the juice out towards the side. A baffle was therefore placed in the middle of the screen to avoid splashing.

72

Mr. Elysee asked how many machines were required to operate with at 50 to 60 tons of juice per hour.

Mr. Carter replied they had two machines which could be changed over every four hours.

Mr. Elysee said that the Peck strainer appeared to be a very dirty element in the factory, but it was a very efficient strainer as determined by taking a sample of juice after passing through the screen and then this was passed through an 80 mesh screen, and the total amount measured by filtration through paper. When a reduction was made from 180 mesh down to 80 mesh there was a definite increase in the amount of bagacillo passed. The present strainer used at Amatikulu was not nearly as efficient as the Peck strainer, although it looks much cleaner. He asked if tests had been carried out on the various mesh sizes and he wanted to know how much more bagacillo was now passed through the bigger mesh screen.

Mr. Carter said that as the screen was inclined at an angle of 30° the full apperture was not available so that not so very much bagacillo could be passed through as might be expected.

Mr. Ross said that he had used three Link Belt strainers in tandem, and that a Russell Separator had been installed for experimental purposes. This separator proved unsuccessful on mixed juice but was installed in the supply line from the supply tanks to the vacuum pans. It was surprising the amount of fine bagacillo which was removed by this Russell separator. It had quite often been found that sugars deteriorated in polarisation after lying in the warehouse for some months and it was highly probable that this was due to the amount of fines which were passing to the boiling house.

Mr. Carter asked if it was possible that some of the bagacillo could not have been returned from the filters.

Mr. Alexander asked if the amount quoted of 100 parts per million of bagacillo collected on the filter was the total amount of total insoluble matter.

Dr. Douwes-Dekker said it was true that if bagacillo was left in the juice the sugar deteriorated quicker. He recalled a discussion with John Payne in Hawaii about the use of vibrating screens on clarified juice. Mr. Payne's reply was that if bagacillo was left in the juice this meant that the sugar became more hygroscopic and deteriorated quickly.

Mr. Elysee said he did not know if it was due to the Link Belt system used, but in recent years he had found large quantities of sand collecting in the bottom of tanks. He wondered if something similar had been found at Doornkop using the Hummer screen.

Mr. Carter said that owing to the arrangements of the factory he could not say if this was the case or not.

The Chairman (Mr. Bentley) said that in his experience vibrating screens, especially Link Belts, removed quite a lot of bagacillo but do not seem to take out much of the sand in the. juice.

Dr. Douwes-Dekker said he had done some work on Hydro-cyclones to see if they could be more efficient than vibrating screens, especially as regards the removal of sand. He found that, in the juice which came from the Link Belt screens there was not very much sand. Similar tests were carried out at Glendale where again it appeared that a considerable amount of sand was removed, by the Link Belt screen.

Mr. van Hengel bore out what. Dr. Douwes-Dekker had said, saying that a considerable amount of sand was trapped in the bagacillo removed by the vibrating screen.

Mr. D. J. L. Hulett asked if the Hummer screen lasted longer than the Link Belt or other type of vibrating screens and he therefore enquired how long the screens lasted—whether one, two or three screens were used per season.

Mr. Carter replied that the original screens were still in use and he imagined they would last for a very long time.

Mr. Thumann asked if a different angle had been considered as compared with a 33° angle, wliich was the inclination applied to now sugar screens now being installed at Umfolosi.

Mr. Carter replied that if one did not use a steep angle one would not be able to get rid of the bagacillo. As yet no other angle had been tried.

Mr. Keus said that tests had shown that the 30° angle was most suitable. He quoted figures about the solids left after passing through the Hummer screen. Analyses of the Cuban juices showed that 66 per cent of the solids (bagacillo) were retained on a 50 mesh testing sieve and, therefore, the use of 40 or 50 mesh screens on the Hummer removed the greatest, percentage of the objectionable fibre while maintaining satisfactory capacities. Screening raw juices at finer meshes such as 80 or 100, would not recover sufficient; additional solids to compensate for greater cost, of installation and operation, shorter screen life and greater tendency of the finer meshes to clog with gum. The percentage by weight of washed and dried bagacillo in the screened juices from this mill, showed that screening at 40 mesh removed 49 per cent: of the solids and that screening at 80 mesh removed 80 per cent of the solids. In Australia, the amount: of solids removed was as follows: At 40 mesh—80 per

73

cent removed, 50 mesh—86 per cent removed, GO mesh—94 per cent removed. The juices sent to the screens in Australia contained very much more solids than that at the Cuban mills which had already been passed through a coarse screen.

Mr. Rault said that our attitude to the action of bagacillo, in its bad effects on juice clarification, seemed very illogical. After liaving gone to the trouble of screening and successfully reducing bagacillo content in mixed juice down to 100 ppm., at a later stage when working the Oliver filters, we close our eyes to the fact that we deliberately

reintroduce three or four times this amount of bagacillo as a filter medium, which is partly washed back in the recirculated cloudy filtrate, thus eventually finding its way into the boiling house.

Mr. Carter considered that bagacillo was put in the mud as a filter-aid and not in the juice.

Mr. Noel mentioned the case of a Mauritian factory which did not use any bagacillo in the mud and which still had a very good performance at the filters. This was not done on purpose but the screening of the mixed juice, although poor, had no adverse effect on clarification.

74

A NEW ION EXCHANGE PROCESS FOR REMOVING COLOUR BODIES AND COLLOIDAL IMPURITIES FROM

CANE SUGAR By R. A. GRANT

Ion exchange resins have been employed in the sugar refining industry to a limited extent as decolourizing agents and for absorbing ash constituents. However, in spite of much research in this field, an ion excliange process capable of replacing established techniques such as carbonatation, sulphi-tation and bone char treatment, has not been developed for application in the cane sugar industry. Ion exchange resins are used mainly for the removal of ash and residual colour remaining after bone char treatment and find their principal use in the production of high grade liquid sugars.

The work reported in this communication was initiated with the object of developing a comprehensive ion exchange process which could be applied to cane sugar solutions containing high concentrations of colour and other impurities, e.g. solutions of third crop crystals and liquors from the early stages of the refining process. The colour bodies present in raw cane sugar appear to be comprised of several molecular species of widely different properties. In order to simplify the problem, it was assumed that these fell into two main groups. A high molecular weight type giving the solution a brown colour and a low molecular weight species behaving as a weak organic acid and imparting a yellow colour to the solution at neutral or alkaline pH values. Although the chemical properties of these colour bodies have not been studied in detail, the foregoing division has proved convenient. It was found that available ion exchange resins were virtually useless in dealing with the high M.W. type of colour body. The reason for this appears to be connected with the porosity of the resin. For an ion to be easily adsorbed and subsequently eluted from an ion exchange resin it must be small enough to diffuse readily into and out of the resin particles. For ions above a certain limiting size, the capacity of the resin is extremely low. This has been known for some time and resins with low degrees of cross linking have been developed for decolourizing sugar solutions. However, even these special resins are strictly limited in their application.

In order to eliminate diffusion within the resin particle as a limiting factor, a new type of resin has been developed in which the active groups are situated on or near the surface of the resin particles. Preparative details of this type of resin will be published elsewhere. Resins prepared in accordance with this principle have proved very effective in removing high M.W. colour bodies and colloidal

impurities from cane sugar solutions, and are stable enough to withstand several hundred cycles of operation. Regeneration is very rapid, probably because all the active sites are readily accessible to the regenerating ions. Figure 1 shows a typical operating curve for this type of resin (curve A). The test solution contained 40 per cent sucrose and was highly coloured (optical density 1.4 at 420 mµ) The solution was passed through a 2 gram bed of the resin having a depth of 3 cm. and the effluent was collected in 250 ml. fractions. The optical density was determined in a Zeiss spectrophotometer type PMQ II at a wavelength of 420 mµ using 1 cm. cells. A total of 1500 ml. was passed through the bed, this being equivalent to GOO g. sucrose giving a sugar/ resin ratio of 300:1. This bed was considered to be exhausted when the optical density of the effluent reached 50 per cent of the initial value. The original solution was dark brown and almost opaque, the effluent was a clear yellow. By this simple treatment more than half of the total colour was removed from the solution.

The yellow effluent was then treated with a new-type of porous anion exchange resin (Figure 1, curve B). This type of resin has proved effective in removing the low M.W. yellow colour bodies. Again, a high uptake of colour was achieved, the final effluent having a pale straw colour.

Tests have also been made on refinery liquors (Table I). Since the colour of these solutions is known to vary with pH, this was adjusted to be within the range 6.5-7.0 prior to and after treatment with the resin. Liquor 1, or "settled juice" (O.D. 2.0 at pH7) was passed through a 2 g. bed of the surface active resin. The effluent had an optical density of 0.27, giving a colour uptake of 86 per cent. Liquor 2 (remelted raw sugar), having an initial colour density of 0.61, gave an effluent having a colour density of 0.10 corresponding to a colour uptake of 83 per cent. Liquor 3 had been subjected to the carbonatation process and had an initial colour density of 0.52 at pH 7.9, adjustment of the pH to 6.5 with hydrochloric acid resulted in a drop in colour to 0.185. This may be compared with the colour change produced by sulphitation to 0.17 at pH 6.7 which indicates that the decrease in colour following sulphitation results mainly from the pH change. After passing through the resin, this liquor gave an effluent with a colour density of 0.03 corresponding to a colour uptake of 84 per cent. Liquor 4

7 5

(after sulphitation) was also greatly impoved by the resin treatment.

A concentrated clarified syrup having a sugar content of 65 per cent and a colour density of 0.3 (deep yellow) could not be successfully treated with this type of resin. However, passage through the porous anion exchanger removed virtually all the colour giving an almost water clear effluent (O.D. 0.01-0.02). With a load of 50 bed volumes, the resin appeared to be only partly exhausted (Figure 2). In this case the resin bed was operated at a temperature of 60°C maintained by pumping thermostat water through a water jacket.

Between them, these two types of resin appear capable of removing virtually all the colour bodies found in the cane sugar liquors examined. A detailed analysis of the economic factors has not been made,

but it would appear that the ion exchange syst could compete favourably with the existing processes.

T A B L E 1

250 ML. FRACTIONS

FIGURE 1. A.. Surface active resin (2 g. dry weight) operating with 40 per

cent sucrose (colour density 1.4). Effluent volume 1,500 ml. equivalent to 600 g. sugar. Sugar/resin ratio 300:1, bed depth 3 cm.

B. Porous anion exchange resin (10 g.) operating with effluent from A. Bed depth 5 cm. working temperature 60°C..

FIGURE 2.

Porous anion exchange resin (10 g.) operating with concentrated clarified syrup containing 65 per cent sugar, colour density 0.3, working temperature 60°C, load 1,000 ml., bed depth 5 cm. Sugar/resin ratio 65:1.

76

Mr. Bentley asked Dr. Grant whether he could give the probable costs of using resins of these types on a practical scale.

Dr. Grant replied that it was difficult to give any exact idea of the economics of the process. Naturally, the more colour in the sugar, the more expensive it was to remove it. For a colour density of 1.4 at 40 per cent sucrose concentration he had worked out that the cost of treating one pound of sugar would be about O.ld. He said he had done some tests at Hulett 's refinery and had got quite good results with greens and third crop crystals.

Mr. Boyes said that on a practical scale one would have to use many tons of resin and quite a lot of sucrose would be held in the resin bed, so a considerable quantity of water would have to be used to remove the sugar before one could tackle the colour bodies.

Dr. Grant said that when the resin was exhausted one could rinse with one or two bed volumes of water to remove the sugar. The colour bodies would not be removed until the regenerating solution was used.

Mr. Boyes said that the amount of water used to rinse off the regenerative substance would mean a considerable amount of evaporation in the factory.

Dr. Grant said that he had worked with a sugar to resin ratio of 300-1 with a solution colour density of 1.4. Some resins, such as Amberlite 401 were recommended only for taking up residual colour. Supposing in the case of a 2 gram bed with a volume of 20 ml., one used one bed volume of w a t e r for rinsing to remove the sugar, then with a 50 bed volume load of syrup the increase in total volume would be only 2 per cent.

Dr. van der Pol asked whether the surface absorption resin was commercially available. He suggested that either bone char or vegetable carbon could be used for the second step instead of the resin used by the author, as only the lower molecular weight type of coloured material was left at this stage.

Dr. Grant said that there was no commercially available resin of the type he had described for dealing with high M.W. colour bodies. He agreed that bone char could be used for the second stage of colour removal, but he had endeavoured to develop an ion exchange process which could, in principle, replace bone char.

Dr. Douwes-Dekker said he found the properties of the surface reacting resins particularly interesting. He asked if Dr. Grant could give, in simple terms, the composition of the resin, give some information on the removal of ash from the sugar liquor and, thirdly, if Dr. Grant could give some information on the poisoning of resins by those impurities in the sugar liquors which were not easily removed from the resins by regeration.

Dr. Grant said that the resins could be prepared in a large number of ways, for example, they could be deposited on substances like keiselguhr having large surface areas. The actual preparation and manufacture of these resins was, however, still in the experimental stage. With regard to ash removal, the number of active groups per unit weight of resin was not very high and these would not greatly affect the ash constituents, such as would the conventional ion exchange resins. There was a certain amount of ash removal however, and phosphate and sulphate would be taken up, but not chloride to any marked extent. With regard to poisoning of the resins he could not easily see how poisoning could occur. In the case of commercial porous resins, the resins could be blocked by large ions and this could be said to be a type of poisoning. With surface active type resins this was unlikely. He had not encountered in his experience anything which could be called poisoning of the resins.

Dr. Graham said that in uranium plants it took a very long time for resins to be poisoned. They could be used 20 or 30 times before poisoning was apparent. He would like to know if Dr. Grant had tried to use the resins as a mixed bed. If it were necessary to use different regenerative agents the resins could be separated by an airstream, for instance. He asked what pH was used, because this could be very important, especially at high temperatures. Evaporating the effluent after resin treatment would be dangerous if the acidity was high. He asked what the composition was of the impurities left after the second resin treatment and wondered if using a slower flow rate would improve removal of colour.

Dr. Grant said, in his experiments, he had studied substances such as egg albumen as well as colour bodies but poisoning did not occur. He thought that in the uranium industry they probably used very drastic chemicals, this might account for the resin poisoning in this industry.

Dr. Graham said that the poisoning in the uranium process was mostly due to cobalt.

Dr. Grant said that the type of resin used would depend upon the substance one wished to remove. For the removal of different types of colour bodies one required different types of ion exchange resins. He had not tried to use mixed beds, but this could be conceivably done, because both resins would be basic. In regard to the pH of the effluent, it could not be, as far as the bulk was concerned, very different from that of the intake material. With regard to the apparent residual colour remaining after the second stage, he thought that this was in part due to a slight haze in this particular solution which was measured as colour by the spectrophotometer. In general, varying the flow rate did not greatly affect the colour uptake.

77

Mr. Dedekind asked if any attempt was made to crystallize the liquor. In tests at Sezela they got the liquor absolutely colourless but could not get this water-white liquor to grain.

Mr. Alexander said the refinery had used these resins for t reat ing greens and boiled sugar from the effluent. Instead of a greyish product, the treated liquor gave an at t ract ive yellow-coloured one.

Dr. Grant pointed out that the experiments Mr. Alexander was referring to, only went as far as the removal of the higher molecular weight colour bodies and the sugar made from the resulting yellow-

coloured liquor was yellow in consequence. He could not see why it should be harder to crystallise the liquor after coming from the ion exchange beds, than that coming from bone char, provided there was no marked change in the pH.

Mr. Walsh asked Dr. Grant if he could give the time of the cycle.

Dr. Grant replied that that would depend upon the volume of effluent and the size of the resin bed and also on the amount of colour in the material to be treated.

78

SOME OBSERVATIONS MADE ON THE REHEATING OF MASSECUITE IN A CRYSTALLIZER

By W. F. DAVIES

The writer apologises for the scantiness of data contained in this paper. However, had more data been available it is unlikely that this article would have gone into print. At the outset let it be said that the object of this paper is not to show how a massecuite should be reheated, rather an attempt is made to point out the snags encountered when massecuite is heated in plant of inadequate design and more or less devoid of instruments.

In 1955 a new set of high speed centrifugals was put into operation at Felixton. It soon became evident that the massecuite reheater supplied with these machines was not satisfactory. The blowing of steam through a Blanchard type element resulted in poor heat distribution and local overheating of the massecuite. A temperature difference of 35°F (120-155°F) was found when taken at either end of the reheater mixer. Final molasses purities were always higher by 2-3° when the massecuite was reheated. Final molasses purities with no heat applied were 38-39 and 40-43 when the massecuite was heated. Minor alterations were made to the reheater but with little effect. Finally the use of the reheater was abandoned and only in cases of extreme urgency was its use permitted. The old method of diluting massecuite with water in the crystallizers and at the massecuite pump was again adopted.

During June and July 1959 great difficulty was experienced with the purging of C massecuites and the resulting recirculation of molasses caused the B massecuites to fall off in quality, to purge less freely and to increase in quantity.

At this stage the advice of the S.M.R.I, was sought. An officer of the S.M.R.I, and a technologist of the Huletts Group visited Felixton in regard to the foregoing, and the following points resulted from the discussion:

(a) It was considered good practice to change the method of graining for C massecuites. Whereas previously syrup grain was used as a footing for C massecuites, it was proposed to grain on A molasses with a view to improving grain regularity.

(b) Due to the sticky nature of C massecuites it was suggested to end the heavying up at about 97° brix and to consider this brix as acceptable, in light of the difficulties being encountered.

(c) It was stressed that the heating of final massecuite can contribute greatly to the ease of curing. However, as pointed out previously, the reheating

device supplied with the centrifugals was not considered of adequate design, so other means of reheating were investigated.

As Felixton is fitted out with six 1,200 cu. ft. crystallizers with Blanchard cooling elements through which hot water can also be passed, it was attempted to heat massecuite in these crystallizers.

The necessary attention was given to the three above mentioned points. The results in practice were:

SUB. A Unfortunately full advantage could not be made of this recommendation as Felixton was not equipped with a Guitometer and it was felt that graining on A molasses, without the help of such an instrument would be rather risky. To achieve as much similarity to the proposal as possible, the following graining method was used. Instead of slurry graining on syrup, C massecuite footings were slurry grained on a 50/50 blend of syrup and A molasses giving a graining charge purity between 73°-75°. After establishing the grain it was worked up on A molasses to 900 cu. ft. and used as footings for two C massecuite strikes.

It must be said that hardly any improvement in grain regularity was noticed, but it should be taken into account that a large quantity of syrup was still used for graining.

SUB. B During the period that the massecuites were being discharged at a lower brix than normal the massecuites were also being heated, so it is very difficult to say how this practice affected either curing or the final molasses purity.

SUB. C First of all it should be stated that the application of heat to the massecuite in the crystallizer greatly contributed to easing the tension at the final massecuite station. The necessary amount of final molasses could be discarded from the factory and hence the crushing rate of the mill could be maintained. Also, due to less recirculation of non-sugars in final molasses an immediate improvement in the quality of the B massecuites was noticed. Before reheating of massecuites, magma purities were between 77° and 79° and after heating these rose to 83°-85°. However, the author feels that it should be investigated as to what costs are involved to gain the above

79

advantages, which in nature, are only normal in a smooth running factory. (On the basis of Nutsch molasses obtained and assuming a final molasses of 37° at the machines, a rough estimate is that something in the range of £14,000 could be saved in the course of an 80,000 ton sugar crop).

We only became aware of the loss incurred after the position in the factory had eased up so much that this emergency way of heating could be discontinued and the normal way of dealing with the final massecuite, namely slight dilution at the pump and in the crystallizers was again practised.

Table 1 gives final molasses purities when massecuites were, heated and Table II when the heating was stopped.

The brix figures in the table are inclined to be misleading. During October the brix of final molasses was actually higher than that shown in the table, due to the. fact that dilution was necessary in order to be able to pump the molasses to the scale tank where sampling takes place. Quite confidently it can be said that the actual brix of the final molasses must have been about 96°.

In an endeavour to find out what happens when massecuite is heated in this type of crystallizer under the existing set-up, Nutsch tests were carried out and Table III shows that at a brix of 95-96° molasses purities of 35-36° regularly occur.

Hence it can be said that heating the. massecuite caused a molasses purity rise of approximately 5-6° whilst feeding with water caused only a 3-4° rise.

As such, it is clear that unless better means of heating are available than those which exist at Felixton, preference should be given to diluting with water.

Let it be pointed out, however, that the above is certainly not an answer to the question of whether reheating should be applied or not. Figures in Table I and II merely indicate that if reheating is seen as a simple process that can be done by any available type of heater, disappointment can be expected.

It is worthwhile to take into account that the "bet ter" method of feeding with water still causes a loss of sucrose due to the purity rise of 3 4°.

Conclusion The author feels that it must be profitable to

see reheating as a potential means of reducing the losses in final molasses, only if suitable plant is available to do this job. The advantages will not only be a. better purging of the massecuite and a lower purity of the final molasses but an increase in C sugar purity and a resulting decrease of non-sugar recirculation.

To achieve the above advantages it would seem that a plant of the following characteristics should be available:

1. A heat exchanger of the continuous flow type where heating and the heating medium are run counter-currently.

2. A heat exchanger which has a large heating surface ratio to volume of massecuite.

3. A mixer above the centrifugals with a small heating element to maintain the temperature of the massecuite.

4. The centrifugal baskets should be covered and conditioned air should be blown in to prevent cooling of the massecuite during purging.

Additionally, the laboratory should be sufficiently staffed and well equipped to be able to maintain a continuous control on the saturation temperature of each boiling, this again to prevent undue resolution of sucrose.

http://massecuit.es

80

TABLE 1

TABLE 2

TABLE 3

M A S S E C U I T E A

M A S S E C U I T E B

81

The President, Mr. Bentley (in the Chair) said Mr. Davies' paper threw a lot of light on some of the difficulties experienced in sugar factories. Operators cannot get the equipment they want when they want it and they always have to try and make do with what is available and try to get the best results from that.

Mr. Turner asked Mr. Davies when he attempted to reheat the massecuite in the crystallizer, to what temperature did he manage to reheat the massecuite?

Mr. Davies said that they aimed to heat the massecuite to a temperature of 1250F. One interesting point was that they found that in a Blanchard type of crystallizer they were getting a temperature of 114° on one side and 127° to 128° on the other side, which seemed to indicate that they were getting little or no circulation whatever in places. Temperatures wen; all taken at the surface of the massecuite.

Dr. Graham drew attention to the point made in the paper that: £14,000 could be saved. This should provide the necessary equipment.

Mr. Carter asked the author what was the purity of the 1st molasses when he attempted to grain on first molasses only, and at what purity it was brought up to when he mixed it with syrup.

Mr. Davies said that purity of the first molasses varied between 65° and 68° and the blend of syrup and molasses was about 75° purity. It was slurry grained first of all without the use of a cuitometer. Now that they had a cuitometer they had better control.

Mr. Thumann asked about the Brix of the 3rd massecuite at 97°. He said that would mean a rather low crystal content of massecuite. He would like to enquire the capacity of the Blanchard crystallizers and the capacity of the 3rd massecuite centrifugals. He also asked if Felixton had tried to grain on A molasses. He thought that the tendency nowadays was to bring the purity of the third massecuite down by lowering the purity of the blend used for graining. He had found at times it was possible to grain on 60° purity whereas on other

occasions it was difficult to grain on 70°. Through the low reducing sugar ratio in Natal it was difficult to grain on low purity molasses. He also wanted to know if the material was very viscous. He thought that Felixton. might have been more successful had they used hot water instead of steam for reheating the massccuites. This might account for why they found differences in temperature in the crystallizer.

Mr. Davies said that he could see. the point that while the massecuite Brix was 98° the Nutsch Brix was only 97° it would indicate a low crystal content. As far as centrifugal capacity was concerned at 150 tons per hour they had six 24 x 42 in. machines.

With regard to graining on A molasses they found at Felixton that they had. not been successful in graining on A molasses although this has been done in other countries.

The viscosity of the molasses was very high indeed. He did not think this was due solely to the low crystal content of the massecuite. He said that water was used instead of steam through the heater and in the crystallizers. The temperature of the water was too high, but if the temperature was lowered one would not have enough crystallizers for they were used as heaters as well as crystallizing machines.

Mr. Thumann asked if there was a long time-lag between the massecuite leaving the crystallizer and entering the centrifugal mixer. If this was the case the heating effect might be lost.

Mr. Davies said the massecuite had to be pumped about 40 ft. from the crystallizer to the mixer. He had pointed out in the conclusion that the mixer above the centrifugals is with a small heating element which was used to maintain the temperature of the massecuite.

Mr. Beesley congratulated the author for pointing out so clearly that the correct treatment for C massecuites prior to centrifuging was re-heating in a specially designed piece of plant, and if such plant was not available, it was preferable to dilute the massecuite rather than re-heat in the crystallizers.

82

DROP IN PURITY BETWEEN MASSECUITE AND MOLASSES

By A. D. ELYSEE and L. E. TURNER

The main objective in sugar pan boiling is to achieve the highest yield of crystals from massecuite and correlativcly the lowest drop in purity of the mother liquor boiled.

In all sugar factories routine analysis of masse-cuites and molasses is considered essential for the control of pan boiling. The usual data reported daily from the Laboratory comprise Brix and purity of massecuites boiled, and Brix and purity of molasses separated at the centrifugals, indicating an overall drop in purity.

From this information the Process Manager can visualise the general, trend of results, to plan, proportions and massecuite cycles.

In factories where pans are not fully equipped with instruments and automatic control, doubts may arise as to whether irregular work may be due to neglect when pan operation is left entirely to the human element, and to inadequate and inefficient plant. It has been ascertained in practice that irregularities may occur at the centrifugal station, consequently undesirable results could not with any degree of certainty be specifically attributed to bad work at the pan, over-feeding of crystallisers or irregular work at the centrifugals.

The authors of this paper have collaborated during the past three years to determine definitely the drop of purity achieved in the pans and further reduction of purity whilst massecuites are cooled before curing. Initially, a few weekly tests were done which revealed such a wide variation of results that it was decided to have the mother liquor of every B. and C. massecuite analysed at striking. This duplication of analysis of pan products increased considerably the routine work of the laboratory. A large chart, Figs. 1 and 2, was placed at the pan station and the drop in purity of mother liquor from every B. and C. massecuite was plotted daily, indicating to all pan boilers the results of their respective massecuites boiled. This information was found most valuable, as subsequently pan work was stabilised.

Careful observations of variations of the work led to investigations of efficiency of vacuum pumps, condenser water supply, steam supply, temperatures of boilings, boiling time, quantity and quality of grain, early establishment, of metastable super-saturation in every boiling, highest concentration of Brix possible without causing loss of time in striking, and even the steaming of pan and also feeding of massecuites cooling in crystallisers. At the curing

station: routine check of speed of centrifugals, quanti ty of steam and water used, check of possible perforated screens, proper separation of molasses and sugar washings.

Amatikulu Factory has not deviated from a. near four massecuite pan cycle during the last three years, viz:

Virgin massecuite and " A " massecuite, 82 purity syrup grain. Two to three hours cooling time.

" B " Massecuite, footing with "C'" sugar magma, 4 hours boiling and 20-24 hours, cooling time.

"C'" Massecuites, syrup blended grain, six hours boiling and 36 -40 hours, cooling time.

All graining is done with a predetermined quantity of slurry.

Five pans are in operation Nos. l , 3 a n d 5 700 cu. ft., calandria 1,450 H.S.,

Ht. 7.25 ft. above tube, plate.

No. 2 850 cu. ft., calandria 1,450 H.S., Ht. 8.1 ft. above tube plate.

No. 4 1,400 cu. ft., calandria 1,800 H.S., Ht. 9.5 ft. above tube plate.

" A " Massecuites, single cured in twenty-three 30" water driven centrifugals, 1,200 r.p.m.

" B " Massecuites, cured in eight 42" belt driven machines, 1,000 r.p.m.

"C" Massecuites, cured in six 36" belt driven machines, 900- 1,000 r.p.m.

From the above description of the plant it may be concluded that this equipment is not of approved modern standards.

This factory in the past has experienced an overall drop in purity of 27° for " B " massecuites and 21° drop for "C" massecuites.

The monthly data tabulated for the three years do not indicate the maximum and minimum drops, nevertheless the averages leave much to be desired. It may be mentioned that the throughput of this factory has been inc.rea.sed '20 per cent in recent years with no additional factory plant.

It is generally accepted that during the last: two seasons great difficulty was experienced in the industry in maintaining a. satisfactory standard of sucrose recovery and boiling house performance. Consensus of opinion is that the composition of non-sugars in cane juices is not consistent, influences of climatic conditions have been noted to change the

http://ma.sscc.uite

http://inc.rea.sed

83

Fig. 1

84

nature of factory products although the quotient of purity may only indicate very slight variations.

The decreasing overall drop of purity and the corresponding decrease in sugar yields indicated by the tabulated figures during the last two years leaves no doubt that there is room for improvement.

In conclusion it may be stated that revealing this

practical information on pan boiling, it would be enlightening to all process personnel if similar work was done, particularly where modern pans, crystal-Users and centrifugals are in use under different conditions. From practical data and further research, a definite standard of Pan work may be set and maintained to the benefit of the Sugar Industry.

88

The President said that regarding the graining of A massecuite he asked why molasses was not used for graining? Was this due to the small number of centrifugals they had? He wanted to know what cycle time six 36" machines would give them with the throughput they handled at present.

Mr. Turner said they had not tried molasses graining because their pan capacity was very low and graining on molasses took a long time. Owing to the limited capacity of C centrifugals a large grain had to be boiled. Twelve water-driven machines had to be used as well for about 20 per cent of the available time, and the cycle was about 15 to 20 minutes. This is a very short cycle for an old-fashioned machine.

Mr. Johnston remarking on Chart 1 pointed to the very low drops shown in certain cases between purity of massecuite and molasses. He wanted to know what this could be attributed to.

Mr. Elysee pointed out that it was not a question of a drop of 12° or 24° Nutsch purity. As far as the variation in a particular individual's work was concerned, they had ample steam and water and vacuum, and it was merely the operation of the pan boiler that counted. All the work was not good and that is why he would like to see other factories try out these tests. It might be that their plant was not of sufficient size or modern design at Amatikulu. Further work of this nature would indicate what capacity was required and what drops in purity could be expected. In season 1957, the recovery was 91.22. In the case of second massecuite the drop in purity was 21.9 and only 2.4 in the crystal-lizer. In May of every year they had the highest drops. This might indicate that the nature of the product was not the same throughout the season. Throughput had not changed and the purity of syrup was more or less the same over the past three seasons. Throughput had been the same, equipment had been the same, but the results have deteriorated particularly during this last year. Some of the other mills which had achieved good results in the past had not been able to repeat them this past season, so he thought he was justified in claiming that the products were not of the same nature as before.

Dr. Douwes-Dekker stated that the paper was useful because it gave more data than was usually obtained in the monthly data. The authors were to be congratulated on having determined all these Nutsch purities. He thought more factories should carry out these tests to guide them in their boiling and cooling process. He had carried out a large scale investigation in Java to find out how the various factories operated their C-pans, crystallizers and centrifugals. It appeared that hardly two factories were working in the same way. Some achieved a much greater drop in purity in the pan than others,

the latter doing their utmost to obtain the required exhaustion in the crystallizers. Factories with a generous pan capacity usually got the highest drop in the pans, and when a generous crystallizer volume was available, the. factory often did not seem to bother to reach maximum exhaustion in the pan. In the case of one factory complete exhaustion was obtained in the pan. Purity of molasses after centri-fuging was no different from that obtained from Nutsch test when the pan was struck, and the molasses was satisfactorily exhausted when spun off.

Mr. Turner said that when it was decided to make exhaustive study of B and C massecuites figures, these tests were immediately given to the pan, boilers--finally in the form of a graph. The pan-boilers apparently took great interest in the graph for they soon corrected any errors they spotted in them.

Mr. Beesley said that Mr. Elysce had asked for figures from other factories. In this respect he could say that under average conditions, Illovo expected to get 29° apparent purity crystallizer Nutsch from 58° purity massecuite. He said that Illovo boiled C massecuites very slowly and tight all the way. It took three to four hours from starting a pan to having the grain ready for feed and then 10½ hours before striking the pan.

Mr. Elysee said that whereas previously the drop had been 29° degrees at Amatikulu they had not obtained anything like this recently. This probably explained the poor results in recent years.

Mr. Beesley pointed out that the grain size of C sugars at Illovo was 0.28 mm. long.

Mr. Turner said that this was a most important point, where time and centrifugal capacity was suitable, to form grain of that size was very desirable.

The Chairman asked Mr. Beesley how long at Illovo could they spin third massecuites.

Mr. Beesley replied that they cured 100 cubic feet of C massecuite per hour in a start to stop cycle of about 22 minutes. The machines were 40" x 30" baskets running at about 1400 rev. per minute. They were going to be speeded up to 1700 revs. this year.

Mr. Dedekind asked if they could get any circulation in the pans at Amatikulu when there was over nine feet of massecuite above the tube plate.

Mr. Elysee said that the discharge gutter from a pan did not have a steep slope so the massecuites could not be concentrated to a higher Brix. On boiling to get a 99° Brix massecuite, molasses was introduced to reduce the brix to 97° Brix for discharge. However when the massecuite was discharged it was followed by the molasses indicating that there was no circulation in the pan.

89

Mr. Dedekind enquired if any mechanical stirring device had been considered.

Mr. Turner said not, not as yet, but he hoped to investigate the possibility.

Mr. Ducasse said there he had a similar height of massecuite above the tube plate. It took nine hours at Renishaw to boil a pan so he thought perhaps at Amatikulu they were boiling too fast. He enquired the amount of footing used in the pan. At times even seventeen hours had been taken to boil a 3rd massecuite.

Mr. Turner said that boiling time was limited by capacity. As far as footing was concerned, his opinion was that one should not use more than 35 or 40 per cent of the capacity of the pan.

Mr. Davies felt that; with the old type of pan when boiling a second or third massecuite you finished with a "jelly" on the top. At Felixton they had comparatively new pans with only about 5' 6" of massecuite above the tube plate, when full. He considered that after about 3' above the tube plate- circulation ceased. Pan research in America had suggested the; same thing.

Mr. Rault said that one could not always go by drops in purity. This depended also on the purity level of the massecuite, a high purity first massecuite giving a better yield with a small drop between massecuite and molasses, than a third massecuite. He endorsed the remark that equipment was generally inadequate with its 20° to 23° drop in purity and through the increased throughput of recent years, many factories were; no longer balanced. He enquired from Mr. Beesley if any increase in recovery had materialised in his factory as a result of the successful reduction of molasses purity to a further stage than other factories. He said that in spite of Amatikulu getting lower drops, recovery figures were still comparatively fair at that mill.

Mr. Beesley said that one had to look at the non-sucrose in mixed juice—some of which went out of the factory in filter cake, but most of which went out in final molasses. Hence as he was convinced that destruction of sucrose into non-sucrose in a raw-house was negligible, one inevitably must get higher recoveries from lowe;r molasses purities. However, he suspected Mr. Rault was having a quiet "dig" at him about the low recovery at Illovo in 1959-60 and in this respect he could say: The clarification plant at Illovo was under-capacity and resulted in mud going over in clear juice at times, this gave; increased losses in final molasses and probably increased the; refinery losses due to poor filtrability, etc. Further it was very difficult to wash the filter cake to a low pol, and even if this was achieved, the filtrate returns dropped in purity, which in turn probably affected the final molasses.

Finally, it. must not be forgotten that Illovo's Boiling House Recovery included the Refinery.

Referring to an earlier remark about circulation, he' pointed out that the C massecuite pan at Illovo was a floating calandria type, with the feed straight into the side; of the pan above; the calandria, hence the feed went straight to the top of the massecuite and did not help circulation at all. In fact he had noticed that during the' last 100 to 150 cu. ft. of each strike, there was a layer of froth resting on the seemingly still massecuite. However the strike took feed quite normally and judging by the conductivity (if left on while striking) the strikes were always homogeneous throughout. Conversely he said that some pans he had observed showed a marked increase in conductivity as the last of the; massecuite passed the electrodes, which definitely indicated poor circulation and a slack layer on top.

Mr. Davies said that the advantage at Illovo was that it took a long time to build up their pan.

Mr. Rabc encjuired about the effectiveness of louvres in the pan.

Mr. Johnston said that at Maidstone they had installed louvres and he thought that to work correctly they should be installed in a pan with a rather shallow base. They would be more effective in a pan with proper circulation.

Mr. Boyes said that the louvres installed did not shew any great improvement. He had studied performance from the point of view of evaporation at different levels in the pan. At the low levels (up to 3 feet above the calandria) there appeared to be a slightly better rate of evaporation. He pointed our however tha t the steam coil originally in the pan spoken of by Mr. Johnstone had been removed, thereby reducing the effective heating surface by about 45 square feet.

Mr. Phipson asked if anybody had used perforated steel steam coils in the pan to encourage circulation. This had been tried at Empangeni with great advantage.

Mr. Rault said he had seen this in beet factories in American and had noted its marked success in increasing circulation. This system was installed in Natal Estates but it was abused. In other words, the pan boilers had used this to assist the massecuite to be discharged from the pan.

Mr. Beesley said that to his mind there was no substitute for slow boiling to get good exhaustion and even with excellent circulation, the strikes should be boiled slowly. However, he did point out, that assistance to circulation was sometimes very necessary and mentioned the first C massecuite pan he worked on, in Australia. It was a tall narrow coil pan and was fitted with four ¼" steam inlets

90

through the bottom cone. The pan just would not boil unless the steam inlets were slightly open.

Mr. Rabe said that a 3" perforated steam pipe was introduced into a pan with poor circulation. Sugar however eventually got into this pipe and it choked up.

Mr. Thumann said in regard to massecuite above the tube plate the performance of the pan could not be stepped up when there was a high level of massecuite above the tube plate. There would be no boiling in the top portion of the pan. It would appear that the pan with a nine foot height above the. tube plate had been altered since it was first installed.

Mr. Turner said that the pan in question was a fairly new one designed to boil 1,000 cu. ft., but when sucrose was high it had to be pushed up to 1,400 ft.

Mr. Johnston said that the efficiency of a pan could be much impaired by neglecting the cleaning of pans from week to week and the efficiency was thereby impaired.

Mr. Davies said when he gave information of his pan capacity to a factory manager from East Africa he was told that his capacity was ample, but he should put in steam traps rather than " U " tubes for discharging the condensate from the calandria.

Mr. Thumann said that it was obvious that evaporation was fastest when the pan was first started, and maybe the " U " tubes did not have the capacity to take away the condensate from the calandria in the initial stages.

Mr. Phipson said that Gcsta steam traps had been installed on all the pans at Z.S.M. and they had found them most satisfactory.

STARCH IN THE MANUFACTURE OF RAW SUGAR By P. N. BOYES

Starch has formed the basis for a considerable amount of discussion and a number of interesting papers. It has been proved1,2,3,4 that starch slows down the filtration of sugar liquor in the refining of sugar. It is claimed by Balch5 that it exerts a considerable influence on the viscosity of massecuites and the data of Buchanan1 suggest that it might double the viscosity of the low grade massecuites. Further it is well known that starch retards the rate of crystallisation of sucrose as witnessed by its use for this purpose in the confectionary trade. It is well known that it co-crystallises with sucrose and is to be found within the crystal. In cases where 3rd sugars are single-cured6 and used as a footing for 1st and 2nd sugars large amounts of starch can be occluded. Finally the recent work of Bennett and Schmidt7,8 has shown that starch—in particular potato starch added in viscous solution—acts as an aid to flocculation.

The question of where the starch exists in the cane is well illustrated by Balch5 who shows that it is deposited principally in the nodes of cane stalks. It is usually located in one or more layers of cells immediately surrounding the fibro-vascular bundles as they emerge from the root band of each node. However, in some varieties Balch observed a certain amount of "diffused" starch where the starch was located throughout all tissues. Chiu6 found that starch was distributed throughout the cane stalk for certain Formosan varieties including N:Co.310, but was much more concentrated in the tops and nodes.

Various authors5,9 have shown that the starch content of juice extracted from different cane varieties is largely characteristic of the variety. In order to obtain more information on the quantity squeezed out from different N:Co. varieties normally crushed, a large number of tests were carried out on a milling tandem with the following results. (See Table I).

T A B L E I

Starch in Juice from different N:Co. Varieties

Sta rch expressed in m g m s per 1,(100 gins juice

CKUSHE-K JUICE. MIXED JVICK

Avg. Aug. Variety Mitt. Max. Mian Brix A/in. Max. Mean Brix

The tests were carried out during October/November which is past the peak maturity period. By taking a number of samples from each variety it was hoped that variations due to ratoon, age, soil, rainfall, etc. would be smoothed out. The results indicated that there is a definite varietal trend but considerable fluctuations occurred within each variety. N:Co.310 showed an outstandingly high starch content while a new variety N:Co.334 was extremely low by comparison.

The considerable variation that occurred within each variety led the author to look at the origin of the various samples and this disclosed an interesting correlation with soil types.10

TABU-: n

Influence of Soil Type on Starch Content

Sta rch expressed as m g m s per 1,000 gins juice

It was observed that cane grown from a large inland area of sandstone gave consistently lower starch figures irrespective of variety. The correlation here is good but in other cases there is insufficient spread of different varieties for a particular soil type. It is also significant that some of the highest figures were obtained on Dwyka soil. It is quite possible that the water holding capacities and drought resistance of different soil types are controlling factors.

It has been suggested that heavy milling is the cause of large quantities of starch being squeezed into the mill juices. In order to investigate this problem more thoroughly a suitable milling tandem was selected. This tandem normally crushed at 70 TCH and consisted of two sets of knives before a 66"x35" Krajewski Crusher with a Searby Shredder before the 1st Mill. There were 5 sets of mills with maceration fed onto the last. Imbibition from the 3rd mill was split evenly to feed the 1st and 2nd mills. The mixed juice to the scales therefore consisted of a combination of crusher, 1st mill and 2nd mill juices.

Throughout a period of two weeks samples of juice were taken from all mills and analysed for starch and brix. The weekly results were averaged and are presented in Table III.

91

9'2

Using the weekly figures for the tandem for the respective weeks concerned it was possible to calculate the tons of juice being squeezed out at each set of rollers. It was then a simple matter using the starch figures given in Table III to calculate the amount of starch contained in the juice. The results of this calculation are given in Table IV and for simplicity the starch figures are given in lbs. per week.

It will be seen that in each case the crusher and 1st mill together extract about 70 per cent of the total starch. The extraction for the 1st mill is higher than the crusher. All subsequent mills extract at a reasonably constant figure. It is therefore clear that the cane preparation before the first two units has resulted in the squeezing out of considerable quantities of starch. The inference that heavy milling facilitates the squeezing out of starch granules is therefore given a quantitative basis.

Consideration will now be given to the development and installation of a method for removing starch from juice.

Certain factories operating in the same rural area showed widely different starch contents in their raw sugars. Investigation showed that the starch contents in the mixed juice of three factories under revue were similar yet the sugars contained the following starch contents. (See Table VI).

This balance can be checked by comparing the cumulative starch figures for crusher, 1st mill and 2nd mill which together constitute the mixed juice with the figure obtained separately for mixed juice. Thus using the figures given in Table IV it will be seen that the cumulative figures for the two weeks are 7,063 and 8,751 pounds compared with the separately determined figures of -7,924 and 8,047 pounds for mixed juice. The accuracy is considered sufficient for broad conclusions to be drawn and by subtracting the starch content in the juice from a mill from the starch in the maceration to the mill it is possible to calculate the actual quantity of starch squeezed into the juice at that mill. This has been done and the results are given in Table V below.

Clearly A and B were eliminating more starch in their process than C and subsequent investigation gave the following results (Table VII). The analyses were carried out on snatch samples only and results are therefore only indicative of trends.

93

The elimination of starch was taking place in the primary tanks of factories A and B where the procedure was to heat raw juice to 160-180°F and then allow this juice to remain for 7-20 minutes. Factory A, which is a sulphitation mill, appeared to have better control and consequently gave better results than factory B, a. defecation mill. In the case of factory C raw juice was partially limed and then heated to 215°F in 2-4 minutes. It is interesting to observe therefore that the two mills were unknowingly making use. of enzyme action in starch removal. 11,12

Lab experiments revealed that gelatinisation of starch took place at 100°F (7()°C) but enzyme was de-activated at 180°F (80°C). Further, sucrose destruction being dependent on time, temperature and pH, it was essential to strike a practical balance between stan'li removal and these factors. It was found that raw juice at 5.2 pH heated to 170°F (70°C) for 10 minutes lost 0.3 per cent sucrose. Liming the cold juice to 7.0 pH eliminated sucrose loss but no starch removal was observed. The process adopted was therefore to add Oliver filtrate to the raw juice and in this way increase the pH to 5.8-0.0 thereby reducing the sucrose loss to a figure that varied from zero to 0.17 per cent for 15 minutes heating at 170°F. In this way the starch removal obtained was 50-60 per cent.

The final process therefore consisted of mixing Oliver filtrate with raw juice thereby increasing alkalinity and also allowing reaction between enzymes and the gelatinised starch in the filtrate. The juice was then heated to 100-180°F and allowed to settle in tanks for 8-12 minutes. The primary lime was added after the settling tanks.

The subsequent investigation at factory C on a factory scale was divided into a study of starch removal and sucrose removal. The results of starch removal are given in Tables VIII and IX. It will be seen that it took 10 days before the system had been cleared to establish the true picture. Subsequent data are given of daily raws and it will be seen that the results are comparable with factories A and B. The result has been a reduction in starch content of sugar of 50-00 per cent.

Sucrose losses were studied by collecting composite samples of juice before the primary heaters and juice emerging from the settling tanks. The apparent sucrose was determined in each case because results were purely comparative. The glucose ratio was also used as a. guide to destruction. It will he observed from the figures given in Table X that a

small amount of evaporation takes place in passage through the settlers and during sampling. The average sucrose figures have therefore to be corrected for this small difference in brix and average out at 12.58 per cent before and 12.56 per cent after the settlers. It could be argued that this small difference has resulted in a loss of sucrose of 0.16 per cent b u t the analytical method cannot be considered so sensitive. The glucose ratio gives an average figure of 4.77 before and 4.64 after, indicating a small loss in glucose.

In connection with sucrose loss it is of interest to note that factory B employing the defecation process had the second highest boiling house recovery in the industry (90.9 per cent). It would therefore be interesting to know how much the small loss of sucrose possibly occurring in the process is compensated for by increased recovery in the boiling house in the light of the known influence of starch.

To date the investigation has been purely to establish a process and get it working in a satisfactory manner. The factory operation occurred during the last four weeks of crushing and liquidation of stock masked any possible differences in recovery. Certain impressions were gained with regard to settling, sugar colour and boiling, but these aspects would have to be studied carefully over an extended period.

Analytical Method for Starch Determination The analytical technique employed in all t h e

starch tests given above is that of Balch5 and Alexander.9

Acknowledgements The author would like to thank Mr. W. Paterson

for his assistance in carrying out the sucrose loss investigation.

94

TABLE VIII

Starch Readings during Settling Down Period of Starch Removal Process

Date

27.11.59

28.11.59

30.11.59

1.12.59

2.12.59

4.12,59

7.12.59

Average readings before process

Mixed

Juice

—

—

595

415

340

235

360

Clear

Juice

148

135

75

223

—

155

155

143

310

A

Sugar

560

335

420

450

575

330

320

710

B

Sugar

._.

750

540

700

375

620

425

Comments

Start up of process.

Juice heaters off 7 hours for

cleaning.

Treated juice now considered to be right through process.

Starch in juices expressed as mgms litre juice. Starch in sugars expressed as mgms 1,000 gms sugar. Results are for snatch samples only.

TABLE IX

Daily Starch Content of Raw Sugar during normal running of Process

Monday Tuesday Wednesday ... Thursday ... . Friday Saturday ... . Sunday Monday Tuesday Wednesday ... Thursday ... . Friday Average reading before process .

Resu Its a ire fo r dai

Date

. ... 7.12.59

ly co

8.12.59 9.12.59

10.12.59 11.12.59 12.12.59 13.12.59 14.12.59 15.12.59 16.12.59 17.12.59 18.12.59

mposited samples.

Starch ppm

250 350 350 300 385 290 275 290 325 400 385 275 710

Table X—STUDY OF SUCROSEAND REDUCING SUGARS IN STARCH REMOVAL

DATE

Friday

Saturday

Monday

Tuesday

Thursday

Tuesday

Wednesday

Thursday

Saturday

Monday

Tuesday

Thursday

Averages

11.30 a.m.

..

..

..

,.

,,

,,

,.

.,

,,

,.

Corrected for brix ...

27.11.59

28.11.59

30.11.59

3.12.59

4.12.59

8.12.59

9.12.59

10.12.59

12.12.59

14.12.59

15.12.59

17.12.59

Days without cleaning

digest tanks ;

0

1

3

-

1

-

-

0

-

1

2

-

-1

0

-

0

2

-_

RAW JUICE + OLIVER FILTRATE TO HEATERS

Brix

15.50

15.37

15.22

—

15.32

14.51

14.89

14.34

14.04

14.24

14.24

14.21

14.77

14.74

{14.64

{15.27

14.74

14.36

14.79

14.70

—

Sue.

13.24

13.29

13.0(5

—

13.47

12.42

12.73

12.25

12.13

12.27

12.00

12. 2S

12.74

12.02

12.42

13.04

12.64

12.05

12.12

12.58

—

Ply.

85.4

86.4

85. 8

—

87.9

85.6

85.5

85.4

86.4

86.2

84.3

86.4

86.2

85.6

84.8

85.4

85.7

83.9

81.9

—

—

Glue.

0.49

—

0.56

—

0.49

0.50

0.42

0.47

0.77

0.51

0.51

0.06

0.81

0.64

0.66

0.61

0.76

0.74

0.61

0.60

—

Glue. Ratio

3.69

—

4.31

—

3.60

4.02

3.03

3.84

6.35

4. 16

4.25

5.37

6.36

4.34

5.31

4.68

6.01

6.14

5.03

4.77

—

pH

—

—

—

—

• -

—

—

—

....

—

5.5

5.5

—

6.0

5.9

—

—

JUICE AFTER HEATING TO 170cF AND DIGESTING S-12 MINUTES

Brix.

15.40

15.27

14.80

—

15.34

14.57

15.05

14.47

14.61

14.64

14.64

14.71

15.14

14.84

15.04

15.74

15.11

14.59

15.09

14.89

—

Sue.

13. 14

13.20

12.60

13.47

12.65

12.98

12.27

12.42

12.57

12.32

12.61

13.04

12.74

12.67

13.27

13.04

12.25

12.15

12.72

12.56

Ply.

85. 3

86. 4

85. 1

—

87.8

86.2

86.3

84.8

84.9

85.9

84.2

85. 7

86. 1

85.8

84.2

84.3

86.3

84.0

80.5

—

—

.Glue.

0.46

—

0.51

—

0.50

0.50

0.38

0.47

0.79

0. 50

0.49

0.67

0.80

0.65

0.63

0.62

0.78

0.74

0.60

0.59

—

Glue. Ratio

3. 52

—

4.30

—

3.69

3.90

3.42

3.83

6.37

3. 98

3.98

5.31

0.13

4.38

4.97

3.94

5.98

6.04

4.94

4.64

—

pH

—

—

—

—

— •

—

—

6.0

5.9

0.0

5.2

5.8

5.7

5.8}

5.7}

—

6.1

5.8

5.8

—

COMMENTS

Started process.

Cleaned tanks.

Cleaned tanks.

Tanks dirty—not taken

in average.

Cleaned tanks.

95

96

The President, Mr. Bentley (in the Chair) stated it was obvious that a great deal of work had been done by the author on this subject.

Dr. Douwes- Dekker congratulated Mr. Boyes on usine the Australian process of Nicholson & Horsley." He asked if the author had found here, as they did in Australia, that starch content and phosphate content of mixed juice were related. Mr. Boyes had discussed the effect of heavy milling. He asked what was meant by this and if the extrac-tion was increased from 92 to say 94 would the starch content thereby be increased? Studying the figures shown in Table VII he had noticed that in .factory C starch removal had been 14 percent only. Factory C, applying the defecation process, one might be temp-ted to conclude that the normal defecation process was not capable of removing more than 14 per cent of the starch. The S.M.R.I., when conducting a similar investigation at two factories, had however found a much higher figure. He now wanted to know, in view of the fact that the paper states "The analyses were carried out on snatch samples only and results are therefore only indicative of trends" what accuracy Mr. Boyes attached to the figure of 14 per cent?" In Table VIII he noticed that the average amount of starch in the B sugar was only 710 parts per million while similarly in Table IX the average before the process was applied still remained at 710, although this figure apparently now applied to all the sugar.

Mr. Boyes said he had no phosphate figures which could be correlated with the individual tests shown in the paper here. By heavy milling he meant heavy cane preparation such as two sets of knives and a Searby Shredder. With regard to the removal of starch it is known that juice contains a certain amount of starch in the form of granules. It is also known that juice contains enzymes which are capable of breaking down the starch particles. Unfortunately in its granular form enzymes are unable to attack starch effectively. Sufficient tem-perature is required whereupon these granules swell and burst, enabling the enzymes to come into better contact with the starch. The temperature at which this takes place is about TOT. It is therefore the first pre-requisite that at some stage of the process a temperature of TOT is reached. On the other hand the enzymes themselves are inactivated above a certain temperature. This temperature, has been determined to be about 80T. Therefore, taking the case of Factory C, a period is reached during the heating by the juice heaters where a temperature of TOT is reached but very rapidly after that a temperature of 100T is also reached. In other words the time factor plays its part in that the enzymes have not got sufficient time to do their work. There is also a second great important feature about Factory C, i.e. it was adopting the process

of adding some of its lime to the cold juice. This heavy liming has also been found to have the effect of de-activating the enzyme. These two factors combined have resulted in Factory C not achieving a very high removal. Now he could not talk about Factory A because he was not familiar with the process, but he could speak about Factory R. In this case when the process of defecation was started the procedure adopted by the process manager there was to heat his juices to between 1G0-180°F. He also had available three tanks. He decided instead of by-passing these tanks to make use of them as surge tanks, which he was still doing to this day. He therefore had the ideal conditions for removing a certain amount of starch and, as Dr. Douwes-Decker pointed out, he obtained a 42 per cent removal. He said he subsequently understood from Dr. Douwes-Dekker's remarks thai the S.M.K.I, carried out an investigation whereby they made fuller use of the tank capacity giving about '20 minutes standing, whereas in. fact the factory manager had only been using about 7, and thereby managed to increase the removal figures to something like 60 per cent. However this long period of resi-dence would result in a loss of sucrose by inversion.

Dr. Douwes-Dekker explained that, the S.M.K.I, had conducted two investigations, one at Z.S.M. and the other at Melville (Factory B). At Z.S.M. weekly periods of normal work were alternated with weeks where the juice was heated to 70-75T and kept at that temperature for 20-25 minutes, in open tanks. Starch was determined in hourly-taken samples of mixed and clarified juice, care being taken that after sampling, starch decomposition was interrupted by pouring the sample into absolute alcohol. Composite daily samples of sugar were also tested for starch. It was found that under "normal" conditions, approximately 43 per cent of the starch in mixed juice was removed in the clarification process, which figure increased to approximately 71 per cent when the Nicholson-Horsley process was applied. Simultaneously the starch content of the raw sugar dropped from approximately 417 to 240 ppm., i.e., by 42 per cent. In the tests at Melville the retention time of the juice heated to 70-75°C was only 10-14 minutes. Here it was found that in two periods of one week that when the factory was working "normally", 48.5 and 55.3 per cent respectively of the starch in mixed juice was removed during clarification, whilst in the week that the juice was retained at 70-75T., the removal was 65 per cent. Simultaneously, the starch content of the sugar dropped by 21 per cent. The lesser effect of the Nicholson-Horsley process at Melville was thought to be due to the shorter retention time. The interesting point was, however, that whilst the S.M.R.I, investigations indicated that normal defecation was capable of removing

97

45-55 per cent of the starch present in mixed juice, Mr. Boyes had found only 14 per cent at factory C. It would seem that further work on the question was desirable.

Mr. Boyes said Dr. Douwes-Dekker mentioned the fact that the samples were snatch samples only. He thought that in determining starch it was pro-bably advisable to stick to snatch sampling rather than try and collect composite, samples over a long period. The results were actually determined from a great number of analyses and not just the result of a couple of determinations. From the results given for different varieties and over a short space of time there was a variation in mixed juice between 175 parts per million up to a maximum of (>2~> parts per million, so he agreed with Dr. Dekker that figures of removal based on mixed juice and clear juice are rather difficult to call accurate. At Factory A it was quite possible that during the testing of mixed juice, the starch fluctuated very considerably

and when compared with the sample of clear juice taken 3 to 4 hours later the comparison would not be strictly true. In comparing removals he had tried to use sugar figures to indicate removal. Referring to Dr. Dekker's fourth question, Mr. Boyes said he was afraid the Table VIII was in error. The figure of 710 was not meant to indicate starch in B sugar. It was meant to indicate starch in total sugars. This was the S.M.R.I, figure and the same figure appeared in Table IX.

Mr. Alexander said that in a number of tests done at Darnall in 1955 when sulphitation was being used, the percentage of the removal of starch was about 17 per cent. Quite a number of samples contri-buted to obtaining this average.

Mr. A. C. Barnes said that the varieties from which these figures were obtained were more closely related to wild varieties than those grown in the West Indies where no process difficulties caused by starch were experienced.

98

SOME NOTES ON THE MELT CARBONATATION REFINERY OF REYNOLDS BROS. LTD. AT SEZELA

By W. G. GALBRAITH and E. DEDEKIND

Historical Prior to 1953 the question of sugar quality was

becoming more and more a serious problem. At that time mill white sugar was produced mainly by the C. G. Smith Group of factories. Refined sugar was produced by the Central Refinery, Illovo Sugar Estates and Natal Estates. Due to the varying qualities of the mill white, an effort was made to sub-divide these various types of mill white into three groups namely: "Special", "Superfine" and plain "Mill White". To encourage the production of the best quality sugar, "Special" white was given the maximum price, whilst the other two grades were penalised to the extent of 3d. per 100 lbs. In spite of this the position was still not very satisfactory from a consumer's point of view, as even the "Spe-cials" varied too widely in quality and appearance.

As a result of the position at that time, the directors of Reynolds Bros, decided to investigate the possi-bilities of refining sugar at the Sezela Factory, and in that event, to discontinue the manufacture of mill white sugar. Apart from the quality of sugar, the economies of refining sugar under the same roof as the raw sugar factory were economically attractive.

After preliminary investigations, it was decided to send two members of their technical staff overseas to study the question, both from the process and engineering aspects. Visits were made to the U.S.A., U.K. and Holland, and the principal processes in use in those countries were studied. These processes included the vegetable carbon, carbonatation and bone char.

As a result of the above investigation, a report was submitted in which it was recommended that, of all the processes studied, the carbonatation process was considered to be the most desirable. The reason this was recommended was:

1. The carbonatation process presented no serious filtration problems, which has always been the bug-bear as far as overseas refineries are concerned when dealing with our sugars.

2. Economically—undoubtedly the most economical.

3. Supply of raw materials viz: Limestone, which is obtainable in South Africa and at an economical landed cost at the factory.

In 1956 when the decision was made to proceed with the refining project, two members of the techni-cal staff were sent to England, where by the courtesy of the British Sugar Corporation, they were allowed

to spend sufficient time at the Kidderminster Factory in Worcestershire to obtain first hand practical experience, in the running of the process which at that time was refining cane sugar raws. This visit confirmed that the recommendations made in the 1953 report were correct.

The Refinery has just completed its third refining season and the results obtained have more than justified the faith put into the adoption of this process.

Raw House The sulphitation process is practised and all raw

sugar produced, viz: "A" and " B " single-cured and "C" double cured, is refined. The sugar is weighed and analysed, before it is melted.

Refinery A surge bin is installed before the automatic raw

sugar scale from where the sugar is fed into a con-tinuous melter. The melter has two compartments; each compartment has a stirrer.

Temperature of 150°F is maintained by injecting low pressure steam into each compartment. A con-stant Brix of melt of 53°-54° is aimed at.

The melt is pumped through a heater, heated to 160°F and stored in buffer tanks from which it flows by gravity into three batch system 1st Carbona-tation tanks. Constant Be. milk of lime is added into each tank in five stages. Gassing with C02 takes place continuously to an end point of 8.1 pH. This procedure takes 12-15 minutes per tank. Carbonated liquor is discharged into a liquor mixing tank equipped with a slow stirrer, which acts as a buffer tank for the filters. The carbonated liquor temper-ature is corrected to 160°F and filtered in "Auto-Filters", which are precoated with " H Y F L O " filter aid. The filtering cycle is 8 hours.

"Sweetening-off" is done in the filter and the sludge is passed through a "Dynocone" centrifuge which separates the carbonated lime from the sludge.

The refinery is equipped for the double carbona-tation process but it has not been found necessary to bring the continuous, 2nd Carbonatation tank into use. First carb. filtrate is passed through the "Quarez" continuous sulphitation tank and gassed down with S02 to (>.8-7.0 pH.

Sulphured liquor is filtered in modern "Fas-Flo" filters (precoated), and heated to 215°F before

99

entering the sealed downtake triple effect evaporator. Vapour from the first vessel is bled off for all liquor heating.

Evaporator thick liquor of 65°-67° Brix is heated to 170°F and then follows the final filtration in "Fas-Flo" filters, (precoated), and pumped to pan supply tanks.

Four or more strkes, depending on colour, are boiled. Graining is done on melt and the intermediate strikes are a mixture of melt, "wash" and "greens". The final strike is boiled on "greens" only. Run-offs from the final boiling are analysed, quanti ty determined and returned to the raw house " A " boilings.

The pans are of modern "Centre Mow" design with a large discharge door. Each pan lias receivers below them from where the massecuite is fed by gravity into the mixer of the centrifugal battery. The sugar is dried in a rotary drier and stored in a four-com-partment blending bin of 120 ton capacity. Blending of the sugar is carried out to put as near as possible a constant quality of sugar on the market.

"Sezela" sugar is bagged either in 50 lb. "Valve Pack Multiwall" paper pockets or in 1.00 lb. Hessian pockets. The sugar moves in a closed circuit and is neither touched by hand nor exposed to surround-ing atmosphere to obviate bacterial contamination.

The pockets of sugar from the bagging station are conveyed by means of conveyor belts into S.A.R. trucks which are consigned either direct to customers or to Durban for storage.

Lime Kiln A "Cocksedge" kiln, as commonly operated

successfully by the British Sugar Corporation pro-vides burnt lime for both raw house and refinery processes. It is not automatic but produces burnt lime of excellent quality and has not given us any anxiety.

Boiler Plant It was calculated that 240,000 lbs. of steam per

hour would be required for the Raw House and to refine 100 per cent of the raw sugar output.

The boiler plant consisted of six Stirling type boilers of 30,000 lbs. per hour rating each, with the horse-shoe type of furnace.

To increase efficiency, the boilers were converted to "Spreader Stoker" system of firing to boost the output to 40,000 lbs. per hour each.

Two were equipped with coal feeding hoppers to be put on coal firing should bagasse be in short supply clue to cane shortage, or similar difficulty.

Instrumentation installed at Carb. Station. 1. Heckman pll Meter—recorders for 1st Carb.,

2nd Carb., and Sulphured liquors (Figs. 1 and 2).

2. Co^ Recorder (Mono).

.1. C.V;3 Temperature Recorder (Thermograph).

4. Rotameter Controller—Recorder for milk of lime; density control (Fig. 3).

Laboratory Control By means of hourly sampling and compositing,

purities, colour and other data of liquors, etc. in all stages of process are determined.

A "Beckman" colorimeter with a 10 cm. cell is used for colour determination and has proved itself a reliable and accurate instrument. It compares favourably with the S.M.R.I, instrument.

Sugars from each shift are composited. Colour index and conductivity ash, and specific grain size is determined and the results made available to shift staff who take a keen interest in their respective shifts analyses.

The "Greens" returned to the raw house are strictly controlled and results summarized for recovery calculations.

A careful check on drains, entrainment sampling devices and sweetening off of each individual carbonatation filter is maintained throughout to ensure that no sugar losses occur.

Daily Analysis Summary Sheet The following is a summary of daily analyses:

REYNOLDS BROS. LTD. —SEZELA

DAILY REFINERY LABORATORY REPORT

D a y N o . 86 D a t e : 11th Augus t , 1959

Raw Melt 1st Carbonated Lkpmr Sulphited Liquor Evaporated Thick Lit]nor " A " Refined Massecuiles 1st Greens " B " Refined Masseeuites 2nd Greens

Brix

53.51 40. 93 50.40 60.50 90. 63 69.59 91.05

. . .

Sucrose

52.36 48. 56

. 48. 95 65. 20

87. 79 66.50 87.21

1

Purity

97.85 97. 25 97.12

98.04 96.86 95.56

95. 78

Glucose percent

. 172

.127

. 118

. 205

pH

7. 12 7 . 00 0.83 0.75

— 6.02

—

Colour

418 117 79 75

285. 588

1114

Colour removed

72.0% 32.5 %

5.1%

N u m b e r o f T a n k s C a r b o n a t e d T o n s Melt to Refinery T o n s Sucrose to Refinery T o n s Re lined s u g a r bagged Refined Suga r : Colour I n d e x

Per cen t Mois tu re Per cen t Ash So., p p m Per cen t Reduc ing Suga r s

S ludge : P e r cen t Sucrose Pol. R a w Suga r R e t u r n e d t o R a w s . . .

71 4 0 3 . 250 4 8 6 . 0 0 0 420.0110

14 0 . 0 4 0.0011 1.9

— 0 . 0 4

98. 07 2151 c u b . ft.

Maximum allowed

50.0 0 . 0 6 % 0 . 0 3 ° , , 25 p p n i . 0 . 0 2 ° , ,

— — —

Remarks

R a w Melt 1st C a r b o n a t e d S u l p h u r e d Th ick L iquor

CaO

0.031. 0 . 0 1 4 0 . 0 1 4 0 . 0 1 0

100% Brix

0 . 0 5 8 0 . 0 2 8 0 . 0 2 8 0 . 0 2 9

S.G.S. Refined 0.4'9 m m .

T o t a l colour r emoved 82 . 1".,

100

J01

102

103

For discussion on this paper see page 125

104

THE REFINING PROCESS AT ILLOVO By E. BEESLEY

Introduction and History The present method of refining sugar at Illovo has

evolved during the past five years in the following manner.

In 1920-28 an activated carbon method of refining was installed at Illovo and operated until the end of 1954. The method was not highly satisfactory in as much as filtorability was for the most part very poor, even with low Brix melts and carbon usage was very high, which necessitated revivification, which in turn precluded the addition of filter aid. However, new carbon addition under these conditions was quite low (see Table I). Also sucrose losses were rather high which is believed to be due to the low Brix of the, melt. The sugar quality was good.

At the beginning of .1954, sugar production was reviewed and it was decided to produce only quota requirements of refined sugar instead of 100 per cent refined as made previously. It was hoped that this step would mean improved refinery operation due to only A sugar being treated. However, the improve-ment still did not compensate for (a) the difference in price received for mill white and refined (£1); (b) the demand for mill white, and (c) the fact that test strikes indicated that very good quality mill white sugar could be made by simply omitting the carbon treatment and filtration from the normal refining process.

During the next few years this remelt production of mill white was standard practice, however, it was obvious that the demand for refined sugar was increasing. Hence the problem of refining by some method other than activated carbon or by some drastic modification of this method had to be solved. The first thought was to put mill white sugar over carbon and it was found that the carbon required to decolourise the liquor was much less than the old method and also that the filter rate at 65° Bx was much higher than the old method at 43° Bx.

Table I gives a comparison of the two.

T A B L E I I

ac* 420

ac* 720

Colour Index

Br ightness

Dom.. Wave leng th

Col. Pur i ty

Ash (calc.)

1st Massecuite

Sugar

. 10

1 8

8 7 . 0

none

zero

0 . 0 0 3 " , ,

2nd Massecuite.

Sugar

40

12

22

8 2 . 2

5 7 4 . 8

1.76

0.000", ,

T A B L E 1

Melt Br ix

lbs. Bx / sq . f t . /hr

I m p . gal l . /sq. f t /hr

N'ew c a r b o n add i t i on , lbs . / ton sugar

To ta l c a rbon add i t i on , lbs . / ton sugar

Old Method

43

... 1.5

. . . 0 . 2 9

... 3 . 2 3

.. . 3 6 . 2 1

New Method

05

4 . 0

0.53

4

12

The refined sugar produced was tested by the S.M.R.I. (13th September, 1957) and gave the following figures:

T A B L E I I I

1. 2

3 .

4.

5. (i. 7.

8.

10.

11.

12.

13.

14.

5.

Rel ined Sugar

Ref. Sugar and L u m p s before dr ie r .

Rel ined Liquor after evapor a t o r

Rel ined Liquor before evapor a t o r

Mill W h i t e mel t

Mill W h i t e sugar

Affiliated Sugar melt

Affiliated Sugar

All W h i t e Massecuites

All W h i t e Kun-Offs

S w e e t w a t e r to Melt

S w e e t w a t e r to Raws

Refinery t a n k b o t t o m s

Mill W h i t e t a n k b o t t o m s

L u m p s

Bx

90 .97

9 9 . 0 0

7 0 . 0 0

0 4 . 0 0

65. 00

98. 50

7 0 . 0 0

9 8 . 50

9 0 . 0 0

7 0 . 0 0

6.70 5 . 50

4 5 . 0 0

4 5 . 0 0 9 9 . 9 0

Brix yield of all W h i t e Massixui te t

Pol.

0 9 . 9 4

9 8 . 1 1

9 7 . 6 1

0 . 4 7

5 . 25

4 2 . 08

4 2 . 08

9 0 . 4 0

Ply.

9 0 . 9 7

9 9 . 9 7

9 9 . 60

9 9 . 10

9 0 . 5 0

9 5 . 5 0

9 3 . 50

0 3 . 5 0

9 9 . 5 0

48 pe r cen t

lbs. Bx.

100,000

600

600

1,000

1,000

500

These figures were so good that it was decided to go into limited production and sell the total output in pre-packed form.

Subsequent experience has indicated that it is worthwhile to install, the necessary extra equipment to produce all our refined sugar for 1960-6l by this method.

The procedures mentioned above- may be: sum-marized by referring to Figure I which is a Brix diagram of the present method starting from the affiliated raw sugar stage. Splitting the diagram at A shows the old refining method to the right and the method of mill white production to the left (provided the return of 3rd Refined Run-Off to the 2nd Mill White massecuite is eliminated).

Again referring to Figure I it will be seen that the mill white section of the diagram can be regarded as a physical clarification process as against the normal chemical methods (Williamson, etc.) used in conjunction with activated carbon.

The Flow Diagram The Brix Flow Diagram mentioned above (Figure I)

is calculated to give a general representation of the method; the basic criteria used are as follows:

Plant Requirements It is estimated that for the 1960-61 crop, the

following plant will yield 800 to 825 tons refined sugar per week.

3 X Batch Melters. 1X 1100 cu. ft. Vacuum pan (exhaust steam) for Mill White boiling 1 X 800 cu. ft. Vacuum pan (exhaust steam) for Helmed boiling 4 X 900 sq. ft. Auto-Suchar filters (over capacity). 3 X Plate and Frame presses for sweetening off (over capacity) f X 3,500 sq. ft. Triple evaporator. 5 X 30" X18" X 1,100 r.p.rn. B.D. centrifugals for Mill White curing 2 X 30" X 18" X 1,100 r.p.rn. B.D. centrifugals plus. 1X 42" X 27" X 900 r.p.m. B.D. centrifugal for Refined sugar curing

ILLOVO SUGAR ESTATES March, 1960

It is also estimated that sufficient steam will be

generated from bagasse alone to operate the process

in 1960.

Acknowledgments

The author wishes to thank the Management of

Illovo Sugar Estates for permission to publish the

paper, especially Mr. K. W. Pearce, who as Factory

Manager, originated and developed the process

through its logical steps.

105


106

SOME NOTES ON THE NEW MELT CLARIFICATION AT HULETT'S S.A. REFINERIES

By J. 1:5. ALEXANDER

Previous to March 1959, Hulett's S.A. Refineries employed various modifications of the phosphodefecation process coupled with paper-pulp and occasionally kieselguhr filtration, followed by single charring of liquors. In March 1.959, flue gas carbonatation alar Tate and Lyle was introduced. It is difficult to make an accurate comparison of the new and old processes as many of the properties now used to measure clarification efficiency were not: in. vogue prior to July 1959. However, there is no doubt that both the clarity of the liquors and the colour of the refined sugar produced by the carbonatation process are vastly superior to those produced immediately prior to its introduction.

Briefly, the clarification process as practised at. present is as follows:

After being weighed by a Servo-Balms the raws to be melted are cold mingled with 85-92 purity affination syrup in a scroll agitator giving a retention time of about 10 minutes at 50 t.p.h. melt. The mingled sugar is spun on 10 E.D. 48 X 30-inch Watson Laidlaw machines with maximum speed of 960 r.p.m. The washed sugar is melted with Johnson sweetwater to an automatically controlled brix. The liquor is again brixed by a polishing brixer before being fed to carbonatation.

Carbonatation is conducted in 3 saturators in series, giving a retention of some 90 minutes with a melt rate of 50 t.p.h. All the milk of lime (Density 1.085) is added to "A" saturator and control of lime and liquor is effected by means of variable speed lime and liquor wheels. Lime per cent solids has varied between 0.75 and 1.5 during the season and has averaged 1.21.

Gassing-out is achieved with scrubbed flue gas. The flue gas runs at a steady 12 per cent CO2 concentration leaving the boilers but is reduced to 10 per cent after having passed the scrubbing units which consist of a Musgrave aerodynamic dust collector followed by a water scrubbing tower and finally a soda tower. The main object of the gas-scrubbing is to remove corrosive sulphur dioxide. About 80 per cent gassing out is achieved in " A " saturator corresponding to a pH of approximately 10.5. In " B " saturator gassing-out by a further 15 per cent causes the pH to fall to around about 9.6. The gassing-out in "C" saturator finally reduces the pH to 9.

The filter-station consists of 9 Auto filters (1,260 sq. ft each) with 4 Johnson plate and frame presses

(1,500 sq. ft. each) for sweetening-oil. The clear filtered liquor (63° Brix) is passed over bone char twice. It first comes into contact with char which has already been used for partly decolourising liquor before coming into contact with freshly burnt char.

As in most sugar refineries the efficiency of the clarification process at H.U.L.S.A.K. is measured indirectly by the removal or non-removal of the impurities rather than gain in purity as such. The fates of the following classes of impurity are used as a yardstick to gauge the thoroughness of clarification:

(a) All types of coloured bodies.

(b) Suspended and colloidal impurities.

(c) Both organic and inorganic anions.

(d) Cations.

(e) Reducing sugars.

The concentration of coloured bodies in sugar liquors is measured by the attenuation index at 420 millimicrons using a Beckman model DU. spectrophotometer. In the case of turbid liquors such as carbonatation supply liquor, a double filtration over kieselguhr is made prior to colour measurement.

The concentration of suspended and colloidal impurities is still judged visually in the plant by examination of liquors for brilliance. Although A*c 720 readings are made on many liquors their interpretation as turbidity is often suspect and thus no A*c 720 figures are shown in Table I.

It has been customary to regard either gravimetric or conductiometric ash as giving a suitable measure of the total inorganic matter in liquors. Although conductiometric ash is still regarded as a very rapid and desirable guide as to what is happening in respect of ionised matter, the use of the determination of "total cations" would appear to have many-advantages especially where a higher degree of accuracy is required. It is less time consuming and requires less skill than a gravimetric ash determination and by expressing the results in terms of micro-equivalents a very simple balance of ionic constituents results. The determination is carried out in a manner similar to that described by Domingues.1 If the sulphate and chloride determinations are made on the same liquor an approximate measure of the organic acids may be obtained by difference, since other anions do not normally occur in refinery liquors to any extent. Ca + Mg and sulphate have been determined by titrations with

107

E.D.T.A. solution and chloride has been determined by conductiometric titration with silver nitrate.

All reducing sugar determinations have been made using the Lane and Eynon method as attempts to introduce the more rapid T.T.C. method have not yet produced the degree of reproducibility desired.

Table I shows the average monthly analysis of materials through clarification over seven months. A general increase in reducing sugars is noticeable from November when the refinery started drawing on stored sugars. Table II shows the reducing sugar "destroyed" by carbonatation and the percentage colour and Ca+Mg removed by carbonatation. The percentage of colour and Ca+Mg (based on filtered liquor) remaining after double charring are also shown.

Although only seven sets of figures are available the possibility of an inverse relationship between reducing sugars destroyed and percentage colour removed and percentage Ca + Mg removed appears not unlikely.

The drop in efficiency of char in removing both Ca + Mg and colour can probably be attributed to the fact that a relatively large amount of new char was added to the stock in the July period. Apparently regeneration of char as practised at H.U.L.S.A.R. is unable to restore char to the level of its original efficiency.

In an attempt to measure the amounts of organic acids throughout the clarification process composite daily samples were analysed over the period 11th— 16th February. Both the washed raw sugar and the carbonatation supply liquor were filtered over kieselguhr at 50° Brix before analysis to prevent the interference of insoluble salts which would be removed in practise by filtration in any case.

The decrease in total cations shown in Table I I I is due almost entirely to removal of calcium and magnesium as can be seen by the almost constant level held by cations other than Ca+Mg. The sulphate ion is rather efficiently recovered by carbonatation and is still further reduced by char. Even

the initial laboratory kieselguhr filtration of the carbonatation supply liquor at 50° Brix was able to remove considerable sulphate. The increase in chloride content from washed raw sugar to carbonatation supply liquor is undoubtedly due to the water and Johnson sweet water used for melting.

The increase in organic acids from washed raw sugar to carbonatation supply is also apparently due to organic acids in the Johnson sweetwater. The increase across carbonatation and first charring is probably due to the destruction of reducing sugars already mentioned. Organic acids appear to be the main anion after carbonatation and even after charring the organic acid content is still nearly double that of the original washed raw sugar. This increase in organic acids not only represents a loss of sugars but is also no doubt responsible for the less efficient removal of impurities over char. The mechanism of formation of organic acids during carbonatation and first charring is not known. However, previous laboratory tests carried out by the author at the Sugar Milling Research Institute were able to demonstrate that manganese acts as a catalyst in the oxidation of sugars to acids, by atmospheric oxygen. Even calcium saccharate solutions containing small quantities of manganese ions produced considerable quantities of organic calcium precipitates when oxygen was passed through the solution at 40°C. Since the lime used for carbonatation contains considerable amounts of manganese (0.5-2 per cent) it seems not unlikely that one of the sources of organic acids may be the oxidation of sugars (sucrose included) by the oxygen contained in flue-gases.

Acknowledgement

The author wishes to thank the staff of the refinery laboratory and in particular J. Bourne, P. de Froberville and G. Frangs for their co-operation in carrying out of the numerous analysis called for in this investigation.

Reference 1 L. P. Domingues: Determination of Total Cations in Com

mercial Sugars. Proc. of the Fourth Session of Bone Char, 1955.

108

TABLE I

AVERAGE MONTHLY ANALYSIS OF MATERIALS THROUGH CLARIFICATION

TABLE III

CONCENTRATION OF SOME IMPURITIES IN MICRO-EQUIVALENTS

PER GRAM SOLIDS THROUGH CLARIFICATION IN PERIOD

llth-16th FEBRUARY 1960

110

TABLE II

111


112

A SHORT DESCRIPTION OF GLEDHOW REFINERY By L. F. CHIAZZARI

The affinated raw sugar is stored in two 50-ton bins thus affording up to about four hours buffer period to smoothe out any fluctuations in the production rate from the raw house as well as the flow to the refinery. It is also useful to allow the refinery to stop before the raw house and "boil off" for the week-end shut downs and at the same time gives a useful stock on hand for the refinery to restart again early.

From the bins the raw sugar passes through a Servo Balans weigher prior to entering the melter. The density of the melt is controlled by a Micro switch on the scale operating an electronic timer so that each dump of the scale opens the water valve for a given time.

Continuous single carbonatation is practised so automatic proportioning of lime to melt is necessary. This is effected through a lime and liquor wheel, motor driven through a variable speed gear so as to give a wide range of throughput and at the same time variable between lime and liquor so as to vary the proportion of lime to melt. As the whole operation is a volumetric one it is most important tha t the density of the melt and the milk of lime be consistent. Under normal conditions the melt is 55° brix and the milk of lime 20° brix. As temperature has quite an effect on the absorption of CO2 it is desirable to hold the melt at 50°C to 55°C, hence the necessity for an ample capacity melter, but it is usually nearer 60°C.

Gassing is done in three similar saturators, with a 9 ft. head of melt. The last one, however, is fitted with a calandria, the temperature of which is thermostatically controlled for optimum filtration which is usually about 85°C. The bulk of the gas is admitted in the first saturator to bring the pH down to about 10.0, the second one down to 9.0 pH and finally corrected to 7.5-7.8 pH in the last one.

The first filtration is accomplished in four Auto Filters of 1,260 square feet area each. No filter aid is used but precoating is done by gravity from an overhead tank. The cloudy filtrate is returned to No. 1 saturator where it serves as a nucleus for floe formation, the clear filtrate goes to the sulphitation plant, and the sludge to plate and frame filter presses for de-sweetening. The water for de-sluicing and de-sweetening is 85 p.s.i.g. and 85°C.

The sulphitation plant comprises a small rotary burner of the Glen Falls type and a wooden absorption tower. The draught is created by a Keebush fan which so far appears to be quite impervious to SO2. Very little sulphur is required as the liquor is only brought to a pH of 6.8-7.0. The sulphited

liquor is then heated again for the second time in a vertical heater thermostatically controlled to about 90°C before the final filtering in plate and frame filter presses again. The plate and frame presses for sulphited liquor and de-sweetening are all standard 42-inch, have a filtering area of 970 square feet each and are hydraulically operated.

This completes the clarification but to assist the pans and economise in steam, the liquor, which has by now dropped approximately 4° in Brix, is evaporated in a triple of 4,000 square feet heating surface to raise the density to about 65° Brix.

Instrumentation is quite extensively installed and is mostly centralised at two points—the saturator floor and the filter floor. In brief, it comprises:

1. Tachometer indicator and recorder reflecting refinery intake of melt.

2. Visible and audible alarms for all prime movers.

3. Indication of levels of all tanks as well as visible and audible alarms for both high and low level.

4. Automatic shutting down of lime and liquor wheels in the event of the flow of the melt or milk of lime failing.

5. Indication and recording of CO2 content of the gas.

5. Indicating and recording flow meters of the total CO2 being used as well as that going into each saturator.

It is intended to operate automatic pH control of the refinery sulphitation plant and the 2nd and 3rd saturators during the coming season.

The other instrumentation is mentioned elsewhere.

On the pan floor, a straight three boiling system is employed in three pans. The first and second boilings are done in two 1500-cu. ft. calandria pans with a heating surface of 2,250 square feet. The third boiling, and, if desired, an occasional seed for the first or second, is done in an 850 cu. ft. pan, also calandria heated, of 1,275 square feet heating surface. All the discharge valves are hydraulically operated and can be controlled either from the pan floor or near the strike receivers.

The centrifugal station consists of eight machines, 42" x 24" at 1,000 r.p.m. Refinements installed are greens and rich runnings separation as well as automatic dilution of same and tip chutes on the sugar discharge to avoid drippings entering the main stream of sugar.

The lime kiln is of the most modern design and is capable of producing up to 25 tons of lime per day,

although, up till now, the maximum demand has been in the vicinity of only 15 tons per day. The dimensions of the kiln proper are 52'6" high x 6'6" diameter, but. as it is elevated 33 feet above ground level to give storage space for burnt lime and thence gravity feed through the slaker, screens, storage tanks and milk of lime pumps, its overall height is over 85 feet.

The complete operation of the plant is from the ground floor. The skip is first loaded with the requisite quantity of coke from a pre-determined scale via a vibrating conveyor and then filled with limestone in a similar way. A turntable brings the, skip into its correct position for hoisting up the shaft.

It is impossible to overcharge the kiln because the operator must discharge sufficient burnt lime to make space to receive a new charge before being able to operate the hoisting motor switch. From then until the return of the empty skip the whole operation is automatic and the seal off at the kiln throat is so effective that no perceivable drop in the CO2 content of the gas is apparent. There is a counter and recorder to keep a tally of the number of skips and a Rotameter takes care of the density of the milk of lime.

Coke per cent limestone for the past season was 9 per cent and the CO2 is usually between 36 per cent and 40 per cent.

113


NOTES REGARDING THE SULPH1TATION-REFINING PROCESS AND PLANT AT UMFOLOZI

By J. I). THUMANN

Introduction During 1959 the Umfolozi Co-operative Sugar

Planters Limited started their refinery to produce a refined sugar by means of the melt-sulphitation-process.

The most important advantages of the application of sulphitation of raw-liquor are simplicity of equipment and low consumption of chemicals.

The known difficulty of sufficient ash- and colour-removal from the juice and evaporator-syrup forms a limiting factor for the production of a millwhite sugar of sufficiently high quality to compete successfully with refined sugar. This is especially so in the field of industrial consumption, but also more and more in the retail markets.

Ever-increasing production costs have been met in part for some time, by increased plant capacity; and South Africa has in no way lagged behind in the extension of the individual factory unit. But this has led to the necessity for further activities to provide a finished product on the spot, which saves a good deal of handling and transport as well as fuel expenses. With the introduction of the process under reference further savings have become possible due to the reasons stated above, and the quantity of chemicals, such as lime, sulphur, as also filteraids, etc. can be kept within reasonable limits, when compared with other processes.

Following the usual procedure for the production of raw sugar by defecation (affination being included in the raw-house centrifugal operation), the process is perhaps the shortest way to produce refined sugar.

Process Control Observations of the behaviour of juices and syrups,

as well as those of operations from day to day, show how easy it is to get variations in the quality of the end-product, unless constant conditions can be maintained in respect of concentration, temperature and pH. With the progress made in the application of industrial control equipment and automation, the risk of producing a product not always up to standard, has been eliminated almost completely. Personally I would go as far as to say that it is doubtful if a good quality refined sugar could be made by this process if the factory had to depend upon the more elementary systems of control and the human element, as was the case a few years ago.

The temperature and brix-control need not present great obstacles to the operator, but alertness to sudden changes can never replace the instrument.

This is the case with pH, the value of which cannot be sensed and the only means for sufficiently accurate measurement is the pH meter. Earlier types often presented difficulties by quick fouling of the electrodes, but. this has been overcome. Instruments of more recent construction are very accurate and so sensitive that they will show immediately the faintest difference and combined with suitable amplifiers and relays will open or shut a valve for the admission of the required reagent, in our case, milk of lime.

Milk of lime is added through a so-called Splitterbox, which is operated by a small electric motor. This motor is started, stopped and reversed by a. relay, activated by the amplified e.m.f. of (he cell formed by the electrodes of the pH meter. The pH meter-cum-controller, which is automatic, maintains the pH within very narrow limits to the pre-set value by passing on the impulses, caused by the changing pH to the motor of the Splitterbox. Thereby more or less lime is added to the liquor.

Therefore, we do not only have the instrument to show the prevailing pH continuously, but it also makes the necessary adjustments, all the time, to maintain a constant value at a pre-determined level, called the Setpoint. There are deviations, no doubt, but these are not so much due to the lack of sensitivity of the instrument, and can be traced back to time-lag in the complete response of the chemicals, the dosing apparatus, and the lack of suitable and immediate inter-mixing of the materials. It goes without saying that these conditions can be improved upon. In our case we have not found any insurmountable obstacles in this respect and the necessary adjustments to the plant have been and are being made in due course.

Production of SO, Gas

The sulphur burner forms an important part of the installation. It is worked with molten sulphur, contained in pits in which steam coils melt, the rock-sulphur and keep it at a temperature of about I35°-140°C. At this temperature the sulphur is least viscous, and may be pumped without any difficulty into the burner. Steam-jacketed piping is essential to prevent re-solidifying of the sulphur in the pipelines. The sulphur enters the burner through a nozzle, by which it is sprayed on to a splashtile and flows down a checker-work of firebricks, the heat of which ignites the sulphur. When starting up, the burner is preheated to a sufficiently high

114

temperature by means of an oilburner. The gas produced is cooled in two stages, first in stainless-steel piping, atmospherically, and then by passing through a water-cooled Karbate cooler. The purpose of the cooling is two-fold, viz. to prevent hot gas from destroying the sugar in the sulphitation, and to remove any SO3 formed in the combustion. This second point presents a peculiar problem, as the gas, while cooling, tends to be even further converted to SO3 and thence to H2SO4, because of the presence of moisture in the air and in the sulphur. Only a very rapid cooling to below the so-called critical zone can check this continued oxidation.

Process The process is as follows:

Raw-sugar of 99.1-99.3 is melted continuously in a horizontal cylindrical inciter of the Prins-Steuer-wald type to a solution of 68° Brix at 78°C.

A Thermo-Controller maintains the temperature by opening and closing a pneumatically operated steam valve. Steam is admitted directly into the raw-liquor.

A Density-Controller does the same to a water-valve and admits or shuts off the water required to maintain the specified degree of Brix.

Part of the raw-melt thus made, circulates back to the mingler into which the raw sugar is fed from

an overhead bin, via a vibrator feeder. Thereby fluctuations in Brix and temperature are largely eliminated and levelled off, and the task of the automatic controllers is diminished, as they have now to make adjustments between much narrower limits.

The raw-liquor is pumped into the sulphitation system, which consists of two tanks provided with gas distributors near the conical bottom and an external circulation system. The circulation system supplies the means of getting intensified reaction contact, but experience has shown that it does not create the desired degree of inter-mixing. Correct pH is maintained by a controller after each vessel, adding milk of lime through Splitter boxes.

The flow of SO2 is not altered on purpose. Any fluctuations occurring therein are caused incidentally, and the required adjustments of pH are made exclusively by the changes in the flow of milk of lime. There remains, of course, the possibility of increasing or reducing the flow of gas, but once the best operational conditions have been established, no alterations to the flow of gas are normally required. The distribution of the gas to the individual sulphitation vessels is in the ratio of approximately 70 per cent to the first vessel and 30 per cent to the second vessel. This is achieved by orifice-plates of corresponding apertures in each of the two SO2

inlet pipes.

115

It is obvious that the quantity of calcium sulphite formed in the processing is important for the extent to which the clarification, i.e. the removal of undesirable matter from the liquor, is required. Therefore, a certain quantity of gas must be admitted per ton of Brix, and for a higher or a lower throughput the gas required will vary. This is adjusted by setting the valves for the admission of air to the sulphur burner with the aid of a differential-pressure gauge, and changing the speed of the pump for the supply of sulphur to the burner.

The sulphited liquor, leaving the sulphitation vessels, is still subject to a certain completion of reaction, and/or a completion of the required structure or formation of calcium sulphite crystals, and therefore the liquor is sent to a retention tank, where an agitator keeps the material in motion, so as to sponsor the right crystal-formation and the adsorption of more of the impurities.

At the same time the temperature, which was 78°C during the melting and can be taken to have dropped by a few degrees only while the liquor was

under treatment in the sulphitation tanks, is being raised to 90°C, the danger of inversion now having been eliminated, as the pH is in the region of 7.1-7.2. This rise is necessary in order to precipitate as much CaSO3 as possible, which substance is less soluble at elevated temperatures.

in the original process, allowance was made for the incorporation of vegetable carbon in the clarification and a tank has been installed between the second sulphitation vessel and. the retention tank.

Vegetable carbon is to be added to the liquor in order to improve its colour, and consequently the colour of the sugar made from this liquor. It also reduces the ash-content to some extent. No data are available from Umfolozi regarding this subject, as carbon has not yet been applied to the clarification. It is thought, however, that it may be advisable to add the carbon at a later stage, i.e. after filtration. Adding it to the mixture of a sugar solution and calcium sulphite precipitate might cause a fair percentage of the carbon to be occluded by the precipitate, before it can adsorb colour from the

116

liquor proper. This will have to be seen in practical operation, and should it prove to be the case, certain alterations in the lay-out will have to be made.

From the retention tank the liquor is pumped to the filter-feed tank in which it is kept at the right temperature to suit speedy filtration, i.e. about 85°C. Filtration is done in three Sparkler Leaf Filters over cotton duck or twill which is precoated with Hyflo-supercel and the filtrate is then passed over Stellar Candle Filters for a "polishing" treatment of the liquor.

The filtercake from the leaf filters, following sweetening off with hot water, is sluiced down by hot water and dropped into so-called "pulping" tanks from where it is pumped into the sump of a string filter. Water is sprayed on to the cake formed on the string filter and the total filtrate is returned to the sweetwater tank. From this tank it is piped to the melter for raw-melt. The cake from the string filter is dropped into a tank with stirring gear and pulped to a slurry, after which it is pumped out of the factory as waste. Originally filtration was only done over the leaf filters, but this was considered insufficient and two Stellar filters were installed, which have given satisfaction and have proved themselves already, although they have been in operation for only a few weeks. The Stellar filter cake is returned to the leaf filter feed tank.

The filtrate, leaving the Stellar filters, is pumped to the panfioor and worked off to sugar. Boiling is done under reduced vacuum (about 22-23 inches of vacuum) and a two or two-and-a-half massecuite system is applied. Final run-off is returned to the raw-house and boiled away into A-massecuites. The massecuites are spun in 1450 r.p.m. centrifugals of diameter 40 inches.

Sugar is dried in a rotary sugar-drier and coarse grain and lumps sent to a reject-melter. A fine crystal fraction is separated and elevated into a seed-bin on the panfioor. When available, this seed is drawn into the pans into a footing of concentrated liquor, forming the footing for the massecuite. Thereby the necessity for graining all the time is avoided, and a much more regular crystal is produced in the massecuite. This facilitates spinning, improved drying, yields more crystal in the bag and saves a good deal of time and steam. It also makes the factory less dependent on the skill of the individual panboiler and there is less difference in the appearance of the sugar boiled by different panboilers. Graining is reverted to only when insufficient seed is available, but usually this brings forth a good supply of seed for the following cycles, especially if the grain has been kept sufficiently small, so that most of the sugar produced from a grained strike returns as seed.

This will not affect the throughput to any great extent, as liquor drawing is proceeding as usual or even faster than when working on seed, because more grain can be formed and except for the period of actual graining, boiling can be kept very fierce.

When working with seed, the panboiler need not worry so much about secondary grain, should this appear, especially in the first stages of the boiling, as it will develop into fine crystals which will be sieved out in the screening of the sugar and return as seed.

There should not be too many secondary crystals though, as this would affect the centrifugal work and hence the quality of sugar for bagging.

Operating Experience Operating experience was gained during the past

season and several problems had to be dealt with and have gradually been overcome.

The first problem arose from the SO2-production plant. The nozzle through which the molten sulphur is pumped into the combustion chamber kept choking up with a substance so far undertermined, but suggested by experts to be mainly Carsulph, mixed with impurities from dust and dirt contained in and deposited on the rocksulphur in storage. Insufficient and unsteady supply of sulphur to the burner caused the temperature to drop below the minimum of 550°-600°C resulting in sublimation in the cooling system, choking of the gas pipes and lack of gas to the sulphitation vessels. Very irregular operation was caused by this, and since one of the most important conditions for a successful operation is a steady flow with a very constant pH, the processing was not at all satisfactory. Changing the type of nozzle to a very simple one with a much larger aperture has remedied the trouble.

Another difficulty was that part of the sulphur did not burn up on its way down from the splash-title to the bottom of the burner. This also caused sublimation and it was found that some channelling of the liquid sulphur took place. It was overcome by placing additional checker-brickwork inside the burner, thereby increasing the total burning area and better combustion was achieved. This is not necessarily a fault of the plant as such, but partly caused by the fact that the plant had to work below guaranteed minimum SO2 production for a successful operation. About 150-200 lbs. of sulphur per hour average was burnt to produce the SO2, for the actual throughput, whereas the burner, which is of the Monsanto-cascade type has been designed for about 300 lb. per hour minimum, this being the smallest possible capacity for a guaranteed good operation. The difficulty seemed to be the contact space and burning area to ensure complete combustion of

117

118

119

sulphur by the air. To improve this the above mentioned extra bricks were placed in the burner, while for the next season cooler-piping will be jacketed and the compressed air for the burner will be passing through the jacket, before entering the burner, so that a more spontaneous and a more complete combustion of the sulphur can be expected, with the higher air-inlet temperature, and the consequently higher combustion-temperature in the burner.

The cooling caused by the air flow through the jackets of the atmospheric cooler section will assist to overcome the increase in gas-temperature by the preheated air and possibly also give a more rapid cooling of the gas to below the point of SO3 formation, so that this can be kept at a minimum.

Early in the season it was found that the formation of filter-cake was such that enormous volumes were experienced. This caused the filters to be filled up with cake the thickness of which was far in excess of the expected size. This could be explained to some extent as being due to very erratic operations in the first few weeks. Sometimes liquor was being circulated for a long while through the sulphitation vessels, and the gassing was not always interrupted, causing an excess of calcium sulphite formation. But after the initial difficulties had been overcome and more or less normal operation could be maintained, it was found that the cake was still generally more voluminous than expected. The filters were supposed to run on stream for up to 12 hours or thereabouts for the formation of filtercakes of up to 1 or 1.25 inch thickness, but it was found that after a six hour run they had to be taken off, the thickness of the cake being as desired and not too much to cause difficulty in sluicing down, as experienced when a thicker cake had been formed.

This would indicate that the specific volume of the produced calcium sulphite filtercake is fairly large and therefore of a porous structure which should facilitate the filtration and also sweetening off. For the coming season an extra filter will be installed to have capacity under all circumstances, e.g. in case it is desired to change the filtersheaths for a clean set.

The formation of sufficiently-sized crystals of calcium sulphite appears to be influenced by the pH of the liquor and fluctuations seem to have some effect, so that filtrates were found to be less sparkling

from time to time. With the installation of polishing filters, following the primary filtration, fine liquor of a high quality could be had, resulting in an improved quality of the sugar.

Some of the cloudiness observed from time to time in the primary filtrate can be attributed to a sudden drop in pressure in a filter, when a clean filter was put on stream along with it. The newly started filter would offer less resistance to the liquor so that the pressure of the feedpump would suddenly be reduced and the cake might crack, resulting in some liquor passing through the cracked surface and causing cloudy filtrate for a while. To overcome this each filter will be fed by its own pump in the future.

Originally the cake as produced from the filter was conveyed on a belt, but the material proved to be too sticky for this method and was abandoned. The next move was the pulping of the cake with muddy juice from the clarifiers of the raw-house, the mixture being returned to the raw-house filters so that there was no need for the separate disposal of the refinery-filtercake. Unfortunately this caused the formation of a very fine and hard scale in the 3rd and 4th vessels of the evaporator plant, which impaired the operation of this section of the factory, so that it had to be abandoned too.

Further consideration to this problem may be given in the future, together with certain contemplated alterations in the raw-house filtration-plant.

It was found that corrosion of the mild-steel surfaces in the various tanks of the plant, which were not protected, caused a slight colouring of the liquor. To stop this effect all tanks will now be treated with a corrosion-resistant coating.

Conclusion In spite of various problems that had to be tackled

during the first season of operations with the new refinery the quality of the sugar was as good as any produced elsewhere and with forthcoming alterations and improvements the Company is confident of an even better quality.

The writer wishes to express his thanks to the Management of the Umfolozi Co-operative Sugar Planters, Limited, for the permission to present this paper.

For discussion on this paper see page 125.

THE JUICE CARBONATATION PROCESS AND REPERCUSSIONS OF ECONOMICS ON TECHNOLOGY

By J. RAULT

Having started my sugar career by dealing with the free working juices of tropical Mauritius, the following seven years' struggle with the Natal sulfodefecation process—also an immigrant brought over in the years preceeding the 1914 war—painfully taught me, that filtration through cloth of the muddy settlings, in spite of a powerful battery of plate and frame machines, was one of the most difficult problems of the South African sugar mills.

All the craft of the "secretive juice temperer", the "cracking point" or various heating temperatures, the ceaseless determinations of SO2 in juice, before and after liming, the addition of superphosphate, the fancy end points by litmus paper and acidi-metric titrations and later the novelty of pH control improved but little, the filtrability of the muds, which were the seat of heavy losses, frequent factory stoppages and poor results in the subsequent operations of evaporation, crystallisation, curing, with a final poor recovery of a raw sugar causing endless troubles to the refinery and particularly at the filtration stages.

Such losses were aggravated if the finished product was to be placed on the market as a consumption white usually sold on "samples" on account of its colour variations, much to the delight and profit of the speculator buyer.

It was then an accepted statement from the practical miller, that it required 1 ton of cane more to make a ton of sugar when turning from a raw sugar to a white sugar, or in more technical language, a fall in recovery exceeding 5 per cent was to be expected.

Farnell1 of Uba days, in his investigations on the colloids of cane juices in Natal and Mauritius opened up a line of fruitful research, on the causes of refactoriness which are now being followed up by Alexander and co-workers of the S.M.R.I.2 and Boyes3.

For a short while some South African factories adopted a process of Australian origin, attempting to eliminate the filter station by sending the settlings back to the milling train, using the moving bagasse blanket as a huge filtering medium.

This prima facie economical method of cutting costs by short circuiting difficulties, apart from creating milling and fuel troubles, had a major objectionable feature, i.e. the unreliability of chemical control in separating milling and boiling house efficiency as well as the incorrect determination of the sucrose present in the raw material.

Nevertheless a survival of this hyphenated process, is the now successful types of continuous settler, whilst the moving bagasse filter idea has materialised in the large unit rotary vacuum filter which has displaced the plate and frame press, both forming a valuable combination for labour saving and continuous work.

The vacuum filters both in beet and cane carbonatation factories, are operating as real filters. The advent of this type of filter has moreover been a trump card in the successful handling of muddy settlings from cane juices, treated by the very small amount of lime now used in the rehabilitated simple lime defecation process which in Uba days had. failed at the filtration stage. As used on cane juices, this filtration is in reality a straining through metallic sieves covered with a layer of bagacillo, delivering a more or less cloudy filtrate, seldom good enough for the evaporator, and for that reason recirculated, increasing the duty on the settling equipment, for a three hours' mill juice capacity.

Refiners' adverse comments on the filtrability of South African raws—a legacy from Uba days— are not without some foundation, notwithstanding the introduction of non-Uba varieties, more modern equipment, and sharper chemical control.

It is not surprising that in 1920, after a few months ' operation during the previous season using improvised and inadequate equipment from the sulphitation factory, gifted with a monumental stone lime kiln, a very bold decision was taken to go ahead with the double carbonatation process of juice clarification.4

A comparatively old one for beet juice clarification, and the only one still used at the present time, it was somewhat foreign to the cane industry although the Javan technologists had also successfully adapted it for the treatment of tropical cane juices.

The following advantages over the established methods, had influenced this optimism: (a) Free running of the filter station and in con

sequence uninterrupted run of the milling plant.

(b) A 100 per cent filtered clarified juice, sparkling, lighter coloured to the eye, and moreover, showing by analysis an exceptionally high purity rise.

(c) An improved speed of the evaporation, crystallisation and curing plant through handling a lower viscosity product, capable of increased turnover from the same size equipment.

120

(d) A better quality consumption white, offering possibilities of better prices and ultimate competition with the bone char product of that time.

(e) Last but not least, in virtue of its lower non-sugar content and lower molasses formation, a boiling house recovery 1 or 2 points ahead of the other processes, when dealing with raw juices of the same initial purity.

Pioneered by the Natal Estates, and adjusted to the imperatives of local conditions, it has stood the test of 40 years, varying its production from 50 to 100 per cent white sugars, according to the vicissitudes of the market.

Starting at 11 to 13,000 tons in 1920, the factory progressively stepped up its annual production to a level surpassing the 100,000 tons of refined sugar made from canes, in addition to the refining of 12,000 odd tons from the purchase of outside raws, in the course of the same crushing season.

A number of factors have contributed to this industrial progress, but the original claims of the carbonatation process have been the one reliable stabilising element allowing the other forces to operate, whilst smoothing the to be expected conflict between "more sugar and a better quality sugar", with a minimum of capital expenditure.

From its inception one serious drawback, accepted as a very evident debit against the process has offset to a certain extent, the achievement realised by the technologist, not readily expressible in terms of money and for that reason not fully appreciated by the cost accountant drawing a balance sheet.

This drawback is the heavier clarification lime requirements, and the magnitude of the chemical bill artificially swollen by the transport charges from the S.A. Railways.

Here it should be placed on record that since adopting this process, using a local raw material and not an imported one, our national economy has benefited by the transport of no less a figure than one million tons of limestone together with its 10 per cent additional fuel requirements under the form of coke.

It is to be deplored that the spiralling railway transport charges have steadily inflated the cost of limestone off-loaded at the factory, to the absurd level of three times its true value at the Northern Transvaal quarry.

Under such irrational conditions, common sense has reluctantly forced our management in its economy campaign, to go back to a technically less efficient, but definitely cheaper method of cane juice clarification, than the one that has served so long and so well.

Carbonatation will not be entirely dispensed with, but will be given a softer job, namely the comparatively easy duty of polishing further, a material much purer than cane juice, i.e. a remelted raw sugar solution requiring a small percentage of the lime previously used. This explains the subtitle of this communication, but we believe that some notes on our trials and failures, the findings on some of our local conditions, the studies of some problem, the details of which have too long been kept in our files, may prove of interest to our fellow technologists.

Lime burning The cost per ton stone, would have been con

siderably reduced and the necessity for a change of process averted, by the finding of a suitable limestone in our province to replace the Northern Transvaal material, but we failed in our efforts at burning the calcitic Umzimkulu stones, in the standard shaft kiln of carbonatation factories. The suggested experimentation with rotary kilns was too much of an expensive gamble, where no guarantee was given as to the gas production side.

Physical structure and not so much its chemical composition was the obstacle.

We have experimented on a factory scale, with dolomitic limestone, containing as much as 30 per cent of Magnesium Carbonate.

Other very pure, but softer stones, powdering in the kilns, and at other times harder structured but highly siliceous limestones unduly raised the limestone consumption.

A rather unattractive bluish slate-coloured material burning to a light chocolate quick lime, due to manganese and iron, but endowed with excellent burning qualities, after exhaustive laboratory and factory scale testing, is now constituting one of our main sources of supply.

Not quite as good a filtering medium as other white limes, through the finer state of dispersion of its particles, it is nevertheless a more energetic purifier, and a higher lime cream yielder per ton stone burnt, and its brown colour has no adverse effect on the colour of the refined sugar.

The function of a lime kiln is as much that of a continuous CO2 gas producer, as of a quick lime for clarification purposes. The factory week-end stoppage is a cause of CO2 waste and temperature disturbance controlled in our case by using fans at the week-end.

It has been found that a kiln gas of lower CO2

concentration than 30 per cent, slows down the neutralisation speed at the CO2 tanks.

The running of a limekiln for refining operations, in a sulphitation-defecation factory, is a source of superabundance of gas, where the carbonatation stage should be a very easy problem.

121

The Carbonatation Station

Much has already been written on this subject in a previous paper.5 For the subsequent 14 years, with a growing market for refined sugars, and a colour quality complex created by the Bureau of Standards, more stress has been laid on this aspect of manufacture, sometimes at the expense of recovery through lack of equipment proportionate to the speed of expansion.

Our constant problem has always been, the efficiency of lime neutralisation at the first car-bonatation stage—a crucial operation influencing the settling and filtration rates of large volumes of mud, the immediate concern of the running staff, and also the chemical composition of the filtered juice, vital for the ultimate quality and quantity of sugar recovered.

The same equipment has had to deal with the juice from a milling train increasing its crushing rate from a level of 140 to 180 tons per hour, limiting the carbonatation cycle to barely seven minutes per tank, the continuous settling to 1 1/4 hours forcing the vacuum filters to a duty of 80 to 85 lbs. of wet cake per sq. ft. per hour.

This liming and gassing stage was further overloaded by the treatment of outside raws, and our own last 3rd sugars, remelted, and worked in the raw juice circuit.

The very justified criticism of our results, is the excessive elimination of reducing sugars to the extent of 35 to 50 per cent of the original present in mixed juice, i.e. a start from a glucose ratio 3.0 finishing at 1.5 to 2.0 in clarified juice.

We have found that out of the factors causing this destruction, namely high temperatures, high pH, and time of contact, it is the second one which is the most potent.

More damage is done by liming and gassing at a pH 11.3 instead of 10.5, than by raising the juice from the accepted 55°C to the high level of 70°C, when working at the lower pH range.

Maintaining the higher pH range is, however, t he one positive means of increasing the speed of CO2

absorbtion, which the workman will always resort t o , when pressed for time in gassing the amount of lime necessary for safe filtration.

Numerous investigations spread over a n u m b e r of years have invariably confirmed the fact tha t a better preservation of reducing sugars, and a remarkable lowering of residual lime salts, have resulted by keeping the pH low during the simultaneous addition of lime and gas.

The enforcement of this very desirable procedure, was always defeated by the impossibility of keeping pace with the mill.

We must admit our failure in solving satisfactorily, the physico-chemical phenomenon of high CO2 gas absorbtion, and out of many designs, the best of a bad job, is the system of perforated baffles that we have used since 1934, notwithstanding its known obstacles to circulation and homogeneous reaction.

Whilst fighting a losing battle at the carbonata t ion tanks, a mass of data have been accumulated on the ever changing conditions.

This work does not claim to have been conducted with the rigorous exactitude and unselfish academic outlook of pure research, but was necessitated through the day by day problems confronting t h e technologist, responsible for controlling process operations, and solving in a realistic way, the conflict between high production capacity and qual i ty work, with an eye on economics.

These investigations cover a wide field, viz. gas absorbtion, glucose destruction, pH range dur ing carbonatation, lime salt formation, colour, filtration and settling rates, filter cake washings, etc. t h e details of which would unduly lengthen this communication.

A few notes and figures are nevertheless presented. The progressive action of carbonatation during

the treatment of a first carbonatation t ank lasting approximately 8 minutes follows a pattern expressed in the following table:

122

Final Limesalts after 2nd Carbonatation In spite of a very accurate end-point control at

the 2nd carbonatation, by factory pH meters, variations of 8.3 to 8.7 have very little influence on the final limesalt content of clarified juice.

This is governed by the nature of the raw juice, and especially the pH reaction during the lime addition at the first CO2 stage.

The use of soda ash as a delimer at the second carbonatation has helped to reduce by 25 to 30 per cent the level of soluble lime salts, and in its turn definitely lowered the ash content of our refined sugar.

Action of Carbonatation on the Mineral Impurities of Juice

In the course of the past 20 years the average a s h content has been 3.48 per cent sucrose in raw j u i c e and 2.39 per cent sucrose in clarified juice s h o w i n g an inorganic impurity elimination of 31.1 per c e n t .

It is interesting to study the nature and pe rcen tage of the elements composing the mineral cons t i tuents of mixed juice and the changes that occur by clarif ication, on a representative sample collected for t h e whole crushing season.

For comparison purposes the same determinations had also been carried out on the molasses ash of the CO2 factory and two other sulphitation factories in the course of the previous season, 1944.

The relationship between ash in raw and clarified juice does not form part of the control of most factories, but one Zululand factory working by sulphitation had this information available and showed practically no drop and more often an increase after clarification.

The above table shows that carbonatation elimin

ates completely magnesia, nearly all the silica, m o s t of the iron alumina and phosphate.

It removes also a high percentage of the s u l p h a t e radicle, whilst increasing the percentage lime c o n t e n t mostly under the form of organic lime salts, a p p e a r ing as carbonates, through incineration for ana ly t i c a l purposes.

The influence of cane maturity and the per iod of the crushing season on exhaustion of final molasses and the variations of some of its const i tuents is noted hereafter:

123

Wax Content of Filler Cake

A reference to old records on Uba days shows the remarkable change for the better that has taken place during recent years,6 in the percentage of this refractory substance present in one factory refuse.

An industry for recovering cane wax from filter cake was a payable proposition at that time.

It is strange to find that worked on the tonnage of Filter cake, the total elimination of wax was not much higher in the carbonatation factory, being in the region of 3,140 lbs. of Wax per 1,000 tons of canes, against 3,214 lbs. for the sulphitation factories, which were experiencing great difficulty in filtering this refuse.

Further experiments at the carbonatation stage, showed that during this operation most of the wax had been saponified by the first small dose of lime, which formed a precipitate unfilterable at that stage—whilst the subsequent additions of lime cream and its carbonating were snow-balling this viscous mud, until it became filterable at the end of the operation.

Here are two examples of this investigation, carried out on factory products, collected during the normal progressive action of carbonatation, with lime cream added in 4 successive charges up to a total of 12 per cent on juice.

These same tests showed that in what concerns nitrogen removal, there was a positive increase by the progressive lime treatment.

Although the nitrogen content of sulphitation cake is 1.25 percent dry substance against 0.47 per cent in the more bulky carbonatation one, the total pounds of nitrogen removed per 1,000 tons of canes is 250 lbs. against nearly double that amount i.e. 498 lbs. in the case of the carbonatation factory.

Gum Content

An objectionable viscous constituent of sugar liquors is "gum"—a complex substance more correctly labelled "alcogel" after the analyt ical procedure for its determination. It is probably related to that other villain "starch", the scapegoat lately loaded with so many sins.

A fairly close inverse relationship exists between filtrability and alcogel content.

The comparatively smaller percentage of th is impurity in carbonatation products, whether th ick syrups, molasses or sugars is the reason for thei r superior working quality.

The progress of its removal during the "refining" of a raw juice at the carbonatation stage is demonstrated in the following 4 examples:

Purity Rise and Boiling House Recovery This incursion in the behaviour of some non-

sugars, very seldom studied in the usual factory control, is a confirmation of the statement tha t t h e unusual purity rise between raw juice and carbonatation clarified cane juice, must be accepted as a positive removal of non-sugars, and not as sometimes suggested a camouflaged destruction of reducing sugars, obscuring the nature and relationship of non-sugars, which are replaced by worse products of decomposition, from high t empera tu re carbonatation.

This substantial purity rise—one of the highest in the world sugar industry—has seldom been under five degrees and is more often over.

The average of South African factories not using carbonatation is seldom over 1.5 and more often barely 1°.

The annual S.M.R.I. Summary of factory figures, brings definite evidence on this subject by the computation of non-sugars, where it is shown tha t 80 per cent of the original amount present in initial r a w juice, still finds its way into molasses, against only 55 per cent for the carbonatation factory, through its more energetic removal at the filtration stage.

Unless this very reduced amount of non-sugars, has the power of carrying away a very much larger proportion of sucrose, and like beet juices cannot be brought down to a low purity molasses, it s tands to reason that the carbonatation factory is in a more favourable position to recover a high percentage of crystals, than the others burdened with larger volume of molasses.

124

1 25

Notwithstanding the unfavourable glucose to ash ratio of our liquors, improved crystallisation tech-nique has lowered the purity of final molasses— nearer that of other factories and it is definitely the shortage of pan and specially low grade centrifugals, that has been the obstacle to further progress.

The boiling house recoveries of most South African factories refer to the raw sugar production and do not divulge the ultimate results obtained on 100 per cent refined, as shown in our case.

In working out this tabic we have liberally conceded a 1.5 purity rise for the other process and limited the carbonatation to the bare 5°.

Within the usual range of South African mixed juice purities say 85, we see that it is possible to exhaust the final molasses with a purity as high as 44 in carbonatation, and still be slightly better off than a defecation sulphitation factory exhausting down to 36—a low level reached by two mills, whilst the rest are nearer 40 this season.

On the other hand, the same carbonatation factory, if provided with well balanced boiling house equipment, starting from an 86 purity mixed juice and exhausting its molasses to 41 purity would still be 1.7 per cent better off in recovery than the others, after a major effort to lower their final molasses to 36 purity from an 86 juice purity start.

This is much in accordance with previous Javan experience, where the carbonatation factories were usually 1 to 2 per cent ahead of the others in their boiling house recovery.

The money value of this improved recovery on a 100,000 tons sugar crop at the average selling price of sugar, represents a substantial source of revenue, that goes far to compensate the heavy chemical bill incumbent to the process.

In going back to the cheaper juice clarification methods, after spending the capital necessary for the altered type of equipment, there is a risk of sacrificing the potential extra boiling house recovery unless the purity of molasses is lowered a further seven degrees on the present range (42 to 45).

The following table based on the elementary mathematics of the S.J.M. formula clearly shows the position for a 100 per cent white sugar manufacture, and the influence of juice and molasses purity on the boiling house recovery.

In practice the real boiling house recovery, will be found on. a somewhat lower range, by the con-stant deduction of 1.5 per cent for losses in filter cake and undetermined.

Other imponderables, together with the very evident saving in the limestone bill have been the decisive factor, in this triumph of economics over technique.

REFERENCES

Mr. Boyes asked Mr. Dedekind if the carbonatated lime was used again after being passed through the Dynocone centrifuge.

Mr. Dedekind replied that carbonatated lime was not used again but sent into the fields. The Dynocone centrifuge was an ingenious piece of machinery which seemed to be most satisfactory in its operation.

Mr. Eddings asked Mr. Chiazzari—

(1) What methods were used to measure and control pH at the saturators?

(2) Why was it felt desirable to apply accurate automatic pH control?

(3) What are the effects subsequently observed on the process when the pH of liquors in the saturators is allowed to fluctuate?

Mr. Chiazzari replied that for Nos. 1 and 2 saturators, straight titration with N/28 HC1 was used, and in No. 3, sometimes phenol red solution and sometimes titration with N/100 NaOH, depending upon filtration qualities and the desired pH. Owing to the high pH prevailing in No. 1 saturator, requir-

ing a large volume of C02 and also the necessity to quickly reduce the high alkalinity, this meant a rather onerous task would be imposed on any pH controller. Nos. 2 and 3 saturators and sulphitation are very different propositions, the control being within comparatively narrow limits and so more readily adapted to instrumentation.

Mr. Perk asked Mr. Chiazzari did he intend adjusting the pH by controlling the gassing or by adjusting the lime, and might not any attempt to adjust it by gassing interfere with the proper working of the lime kiln?

Mr. Chiazzari replied that the gas from the lime kiln was practically constant and the control would be by gassing. He said they always had an excess of gas which was blown off into the atmosphere. The gas in the saturator went in at constant pressure and because lime was also burnt for the raw house there was always a surplus of gas.

Mr. Rault said Mr. Chiazzari was wise in using the gas from a lime kiln for conducting his carbonatation of melt. Those refineries who had to rely on the low concentration of C02 in flue gas were always limited in the possibility of additional lime treatment for better filtration. The gas absorption efficiency was so low with a 10 per cent C02 gas content that this had to be balanced by a huge pumping plant and size of carbonatation tanks and prolonged time of gassing. The worst feature was the high alkalinity level at the first liming tank in order to promote speed of absorption. He felt that the long gassing time, sometimes presented as a help for nitration, was really a necessity called a virtue.

Mr. Noel was surprised at the number of techniques used to produce the same quality of refined sugar. He considered there must be one method which was more economical than the others.

The Chairman (Mr. Bentley) said that at the moment all methods had their own supporters.

Dr. Douwes-Dekker had hoped that some sort of an answer to that question would have been put forward today. He was sure that in the course of time, with the five different processes in use in this country, it would be possible to determine which was the most economical.

Mr. Alexander asked Mr. Thumann if a special type of vegetable carbon was to be used at Umfolosi to facilitate ash removal.

Mr. Thumann in reply said he did not expect any ash removal at all from treatment of the liquor with activated carbon.

Mr. Young asked Mr. Thumann how fouling of the electrodes in the pH meter had been overcome.

Mr. Thumann replied that this was due to the type of glass now used. A certain amount of fouling,

however, still took place, but it was not of much importance, because electrodes could normally be used for several hours.

Mr. Antonowitz enquired whether Mr. Thumann meant that it was the quality of the glass electrode which kept it clean, and that no special procedure was necessary to prevent fouling.

Mr. Thumann said in reply there were several different types of instruments, some fouled up quicker than others. Nowadays with the better quality of glass used, excessive fouling need not be experienced.

Mr. Beesley, in answer to a remark about the poor ash removal of vegetable carbon, pointed out that Norit carbon was used at Illovo and that the refined sugar had a very low ash content.

Mr. Alexander said he was referring to t he removal of ash by vegetable carbon and not the amount of ash left in the sugar as such.

Mr. Dedekind asked about the cleaning procedure at Umfolozi and at Hulsar as far as the cleaning of tubes was concerned.

Mr. Thumann said the quality of the liquor was such that there was very little deposit on the pan tubes.

Mr. Alexander explained that they had some scaling of their calandrias. He thought this was mostly due to calcium sulphate left on the strips from the previous process. He thought the answer to that problem was by boiling out with caustic soda.

Mr. Dedekind asked if it was a mild steel ribbon that was in use at Hulsar.

Mr. Young said that the pans used did not require a lot of cleaning as there was very little scale built up.

Mr. du Casse said he noticed that a 22-23" of vacuum was used at Umfolozi and he asked at what temperature the pans were boiling and was there no risk of entrainment at what was probably a very high temperature ?

Mr. Thumann replied at 150-155°F. He considered there was no risk of entrainment by boiling under a vacuum of 22-33". Boiling can be done under any vacuum and entrainment will not take; place if the vacuum is not altered suddenly.

Mr. du Casse said that in his experience boiling at 22/23" of vacuum, the temperature would be over 150°F and one would have trouble with water getting into the vacuum pump. If the pan was boiling at a high temperature it had to be eventually cooled down and he said that by dropping the temperature one was bound to have entrainment.

126

Mr. Thumann said that at Umfolozi they had no trouble with water getting into the pump as they had an air extractor. The level of the massecuite in the pan was only about 4/4|-" above the tube plate. He did not consider the temperatures used at Umfolozi unduly high, in the first place because of the low level of massecuite and secondly because of the high grade of the massecuite. In raw sugar C massecuites when the Brix was 99/100° one could anticipate a temperature of about 170°F. The higher the level of the pan the more the risk of entrainment.

Dr. Douwes-Dekker agreed with Mr. Rault that it was a pity that this more or less ideal method of juice clarification is now being abandoned. He said this was also happening in Java for the simple reason that the carbonatation process was too expensive to use in sugar manufacture. We could be pleased that the carbonatation process was still

being retained, not in application to mixed juice, but to refinery melts. This was because the carbonatation process was more efficient in a high density medium than in the low density and that is why its use was continued in refinery processes.

Mr. Carter said on behalf of his collegaues who had joined the Industry in 1922 he would like to pay a tribute to Mr. Rault for his tuition at the Technical College which laid the foundation on which they had built over the years.

The Chairman (Mr. Bentley) also paid tribute to Mr. Rault on his retirement, saying that not only had he been an active member of the Council but also an active member of the Association. He said that he was sure Mr. Rault would continue to attend our meetings, so this, as far as we are concerned, was not really his swan song.

We hope to sec him for many years to come.

127

128

SYSTEMATIC PLANNING AND SCIENTIFIC CONTROL OF FIELD OPERATIONS

By G. C. SHEPPARD

There is evidence that in many cane growing countries field practices have not kept pace with scientific progress, and in fact may be lagging far behind. One of the reasons for this is that there are no yardsticks for measuring productivity in the fields comparable with extraction and other efficiencies that are continuously and accurately calculated in sugar mills in relation to the measured weight of cane processed. The weighbridge at the mill is therefore the dividing line between scientifically controlled processes in the mill and rule of thumb operations in the fields.

There are so many variables in agricultural conditions that efforts to make comparisons between zones, estates and even adjacent fields are complicated to such a degree that they often become meaningless. In such circumstances there is a natural tendency to lower budgetted field efficiency targets to a point only slightly above current performances, whereas it has been proved in practice that improvements in efficiencies of 50 to 100 per cent can be obtained simply by making fuller use of existing facilities.

The purpose of this paper therefore is to describe planning and control methods that can be used profitably to increase the utilization of labour, machinery, materials and fields in cane growing and in other branches of agriculture. Needless to say any direct improvement in the physical features of agricultural work will in the long run reinforce the development of scientific practices and thereby improve yield.

It is desirable at the start of the programme to be described, to form a small and suitably trained planning staff to work under the direction of management. The planning and control procedures to be introduced, while similar to those employed in highly organized engineering and manufacturing organisations, are adapted to the peculiar conditions found in agriculture. They can be applied effectively to any reasonably well run agricultural undertaking and do not rely for their success on the presence of ideal conditions. Their function is rather to measure the effectiveness of existing conditions and systematically to find ways for improving them.

Good timing is more critical in importance in agriculture than in any other industry because of the effect of the seasons on plant growth. Failure to carry out any one operation in the life cycle of a plant at the right time and in the proper sequence will either result in permanent damage to the plant

or destroy the effects of previous work which will have to be done over again.

The first step in developing control through plan-ning is for management to make formal estimates in complete detail of every phase of agricultural work to be carried out during each month of the year ahead. This involves laying down a programme for every field in respect of land preparation, planting, weeding, fertilizing, cultivation, reaping, cane transport, irrigation and every other conceivable activity. Priorities must be set and quantities measured or estimated in terms of tons, acres or miles.

Labour and equipment requirements are estimated for every project on the programme, for each month of the year. These requirements are then plotted graphically, in the manner shown in Graph A, to ascertain whether or not the programme can be carried out by the anticipated amount of labour and equipment available. It invariably occurs that the first attempt at programming fails to make the most economic use of the available resources or to balance them in each month of the year. Any lack of balance imposes too great a load on the available resources at some times and too small a load at other times, so that the cost of production is increased either way.

The final objective is of course a programme that flattens out the peaks and fills in the valleys to achieve the fullest utilization of the available resources.

Experience has proved that capable agriculturists can programme and estimate with a highly satisfactory degree of accuracy. While some flexibility must be allowed for the uncertainty of weather conditions, the targets of a well conceived programme can be achieved. Contingencies are dealt with as they arise.

To plan and to progress is to set targets and to establish controls which measure actual performances against the targets. Most agriculturists plan informally or have a mental picture: of work they intend to do in its sequence. Unfortunately planning in this form cannot be displayed for those whose job it is to follow the progress of work. Further, the longer the period of operations becomes, the more difficult and less accurate are the mental calculations involved, and the harder they are to remember.

While the compilation of a yearly programme in the complete detail necessary is an arduous task on

the first occasion, it develops into routine in sub-sequent years, with ever increasing results. It provides the only reliable basis for determining policy with regard to labour recruiting and tends to reduce overall labour requirements. It is a positive brake on the costly tendency to engage excess labour when it is in plentiful supply as insurance against the possible shortage of labour later in the season.

Opportunities are presented to minimize equipment requirements by centralising control. For example; where it is customary to allocate one crawler tractor to each section for land preparation, regardless of the task in hand, it is invariably possible to reduce the number of crawlers in use by pooling them and allocating machines to sections according to programmed requirements. The management at a local sugar estate, successfully used this approach to reduce the number of crawlers in service for land preparation from nine to three and halved the crawler cost per acre planted.

The next stop in planning is to break down the approved annual or long term programme into short term monthly or fortnightly periods which are issued to section managers and overseers as they fall due. The short term programmes constitute a directive from management specifying how and when each phase of planned work should be carried out. These programmes are plotted graphically so as to illustrate visually the necessary phasing of operations and the results achieved in relation to time.

A specimen field work programme is shown in Table B. It lays down the work to be done in a four week period together with labour standards and budget costs.

TABLE B

FIELD WORK PROGRAMME

Section: Example Period: May (4 weeks)

Management is directly responsible for planning the supply of labour, equipment and materials outside the control of the man on the spot.

Follow-up is provided by normal supervision who account for work completed in relation to programme. Supervision also imposes quality controls by system-atically rating the cleanliness of fields in respect of weed growth, cane left behind after cutting, bad stumping and trash spreading. Management meetings are held weekly in order to measure progress, to revise programmes where work is not up to date and to discuss future projects.

The "Job Request" system is employed for modifying programmes. Job requests are numbered orders contained in triplicate books issued to every official. They are used to provide written notification from one level of management to another when a change from the original programme must be made or to issue a new instruction. The original order is given to the official concerned and a copy is routed to the field manager to keep him always in the picture. The second copy remains in the book.

Job requests are followed to completion systematically. The issued copies are placed on special wall boards in the offices of the receiving official and the field manager in order of priority and under the headings "In Hand", "Hold Over" and "Com pleted". When a request has been completed it is suitably endorsed and returned to the originator via the field manager. Adherence to this routine provides: better understanding by all concerned of the work in hand; improvement in the standard of instructions because they are given officially and in writing; better follow-up and a reduction of the misunderstandings and recriminations that so often occur where verbal orders are given.

The basic programmes with all modifications are plotted graphically on a calendar basis against achievements plotted in a contrasting colour. These graphs provide the agriculturists concerned with a simple, up to date and visual means of control over every aspect of their work.

Once the broad principles of planning and control have been established there is wide scope for systematically improving productivity by introducing various scientific management techniques. Among these are:

1. Labour Control for fixing labour complements, keeping them up to date with changing policies and ensuring that actual labour employed does not exceed the complement.

2. Time study to measure work, to reveal the causes of lost time and to set more accurate production targets for labourers and machines.

3. Methods study to improve on existing work methods and to ascertain accurately and conclusively which is the best and most economic of two or more alternative methods to be used under given conditions.

129

1. Motor Transport

Tons cane handled Running and maintenance costs Cost per ton of cane

130

Labour used on weeding and weedicide

Acres weeded Unit days worked Unit days per acre

Total direct labour

Unit days worked Measured productivity index

Indirect Held labour

Unit (lays worked Productivity Index

S. Expenditure on controllable field stores ('That is all stores used in the fields excluding "pol icy" commodities such as fertilizer and weedicide).

Expenditure Reduction

Most cane planters rely for the results they achieve on personal experience, knowledge gained, from their predecessors, advice received from working colleagues or agronomists and instinct. New methods giving spectacular and therefore easily identifiable results, are invariably passed on for the good of an estate or the industry as a whole, bu t there can be no doubt that many relatively good practices developed by individuals are neither identified nor adopted by others.

It is not difficult economically to justify the introduction of formal effective planning in any-sugarcane estate. It systematically harnesses the skills of many individuals to a common cause, provides a plan and a follow-up that clearly illustrates performance trends, which, although they may be insignificant in the first instance, can be fostered and improved to gain great results. It will most assuredly improve timing and good timing saves money in a big way. It will achieve optimum labour complements and improved utilization of equipment together with increased growth of cane.

4. Job descriptions that outline and formally record the selected methods of carrying out every aspect of agricultural work.

5. Aptitude testing and labour training based on job descriptions.

6. Incentive pay to encourage higher productivity on those jobs where it is possible to measure accurately the quantity and quality of work output in relation to pre-determined standards.

7. Budgetary control based on measured standards of work output and of expenditure.

A selection of results achieved by a sugar estate which has applied these principles of scientific management in their fields is tabulated below. It is noteworthy that the labour budget figures for the current year are based on an improvement of 12 per cent in utilization for direct labour.

2. Crawlers and planting

wheel tractors used on land preparation and

Acres planted Running and Maintenance costs Cost per acre

3. Labour used on Land preparation and planting (Plowing, land preparation, clearing, stumping, planting maila application, mix fertilizer, roll cane and re-plant).

Year I Year 2 Year .5

Acres planted Unit days worked Unit days per acre

4. Labour used for reaping and sundry jobs (Reaping, spread trash, second load, long carry, lay tramlines, pre-trash and cultivate).

Year I Year 2 Year '1

Tons cane reaped Unit days worked Unit days per 100 ton...

132

Mr. Sheppard said, before commencing the discussion on his paper, that Mr. C. J. Saunders when opening our Congress stressed that it was more essential now than at any stage in the past 15 years, that sugar technologists should turn their attention to reducing production costs and obtaining the highest possible return from a limited market.

The Chairman, Dr. McMartin, said this was the first paper of this nature presented to our Association. He felt that the results achieved would no doubt be of interest particularly to those in the administrative side of field operations.

Dr. Dick said that obviously the scheme described was designed for a big estate and he wanted to know if this system could also be adapted to the needs of small individual growers.

Mr. Sheppard said that discussions had taken place with a view to putting the system across to smaller growers. This is being done in the Federation in the case of tobacco growers. After an exploratory period of six months, which proved satisfactory, these tobacco growers are now carrying out a comprehensive study of the use of labour and machinery throughout the industry. A strong team of consultants is working in conjunction with employees of the Tobacco Producers' Association. A manual of the most efficient methods employed in the various operations in all tobacco areas will be made available to the 4,000 odd big and small growers. He was confident that this would result in greatly-improved productivity. There is no doubt that farmers generally would have to keep more comprehensive records in order to measure the effect of introducing new working methods and systems. In general the method outlined in the paper could be applied to the smaller cane growers.

Mr. Boulle asked for a short description of aptitude testing and labour training.

Mr. Sheppard said facilities for carrying out such tests were provided by a Government sponsored body, "The National Institute of Personnel Research''. The aptitude testing was based on requirements from each type of labour and was directed to engaging the best type of labour for any particular occupation. One estate in this Industry had established such a testing system.

Mr. Pearson said that in the graph at the end of the paper it was apparent that reaping was not the only great requirement from labour. Weeding also was most important and greater economies could be effected in this field than by mechanical harvesting. Another important point was the great reduction in units of labour required. He wondered if this was due to the policy of weeding more or less before one saw the weeds, or was it due to

greater efficiency in the management of weeding gangs ?

Mr. Sheppard replied that it was a combination of both. This depended upon better timing and better utilisation of labour later on. On the estate mentioned in the paper, weeding constituted, the biggest portion of the labour force. In more Northern climates, however, the reaping force might-outnumber the weeding force.

Mr. Pearson said that we had possibly greater-chances of improving our weeding by mechanisation than we had in reaping.

Mr. Sheppard thought that in general there was more prospect of using machines for weeding than for harvesting.

Dr. Cleasby said that the author had stated that the crop of 1958 could compare less favourably with the 1959 figures. He wanted to know what the crop figures were for the season, 1959.

Mr. Sheppard gave these figures: Tons cane handled by Estate transport: 1957 -154,000; 1958—213,000; 1959—151,000. He said that the final productivity results for the year 1959, beat the budget figures shewn in the paper.

Mr. Thompson said that the figures given included estate and planters' fields dealt with.

Mr. Bentley said that by systematic control the efficiency of labour had been improved. He asked if this had been brought about by incentive payment or by better supervision? If by incentive payment had the total cost been reduced? His experience had not always shewn this to be the case.

Mr. Sheppard replied that the improvements in this particular example were not due to incentive bonuses to any extent. He quoted figures showing that the biggest reduction was in indirect labour— there was no increase in cost involved there. Generally there had been very little increase in bonus payment. Apart from some small incentive payment for boys weeding, there had been virtually no change in incentive payments to the natives, so that most of this reduction in labour affected direct saving in labour costs. Increased bonuses were paid to overseers and supervisors who were largely respon-sible for achieving the results shown. Their bonuses had been more than paid for by the result in saving of stores expenditure alone. Bonuses could always be useful provided the quality of work was ade-quately controlled.

Mr. Brook asked if the saving in harvesting costs was not due to the introduction of self-loading transport—thus saving loading costs.

Mr. Sheppard replied that there was some saving due to extra mechanisation but this amounted to only a small part of the total saving.

Mr. Thompson said that the merit of pre-planning in weeding for instance, lay in the fairly accurate estimation of labour requirements before the operation was commenced. Field management thus

played a direct part in the control of individual operations.

The Chairman said that Mr. Saunders in his opening address to Congress, had stressed the necessity of reducing unit costs and Mr. Sheppard had pointed out one way in which this improvement could be brought about.

133

NATAL SUGARCANE VARIETIES SOME OBSERVATIONS AND STATISTICS

By EXLEY STEWARD

Since the time when sugarcane was first grown commercially in Natal, over a hundred years ago, perhaps the most important question that has had to be decided by growers is which varieties of cane to cultivate to give the best returns under local conditions. It may not be generally realised that many hundreds of varieties exist in the sugar world to-day. McMartin1 states that from the early days up till 1940, 180 varieties had been introduced into Natal alone, although only a few of these ever established themselves as commercial varieties. The first of the early varieties were imported from Mauritius about 1850, and Uba, the one time saviour of the industry, was brought into the country in 1883. Uba proved to be immune to mosaic and smut disease which were attacking the other canes to such a degree that some varieties were almost completely destroyed. In an attempt to rid the industry of these and other diseases the cultivation of any variety other than Uba, after June 30th 1927, was made illegal by proclamation of the Government.

It is the object of this paper to review the variety position since that time and to give some indication of the behaviour and results obtained from the canes that have subsequently been made available to the industry. The observations and statistics which follow refer to the 21,000 acres of cane lands which are under control of the author who does not presume to speak for the entire sugar belt. It is probable that different results have been obtained in other areas but it is hoped that the facts given here will prove of some value and lead, at least, to beneficial discussion. The figures, given in the various tables below have been taken over four seasons, from 1956-57 to 1959-60, and represent the findings from the harvesting of 30,163 acres.

Since 1930, 16 varieties have been released by the South African Sugar Experiment Station, at Mount Edgecombe, for commercial planting. All of these will be briefly discussed. In the paragraphs below the varieties are numbered in order of release, the year of release and the parentage, in parenthesis, given.

The immediate performance was very good and planters became enthusiastic about this variety which gave the best results in heavy red soil. After about twelve years Co.281 declined very rapidly. The last planting was done in 1949 when only 1.4 per cent of the total area planted was put under this variety (see Table IV). Even selected seed, planted out in rich virgin soil, failed to produce a satisfactory crop after 1949. The reason for the dying out of this, and other varieties, is unknown.

1.34

The same remarks apply as for Nos. 1 and 2 except that Co.290 lasted longer than the P.O.J. although Co.290 proved susceptible to red rot, especially at the higher altitudes.

These were successful to varying degrees but were subsequently replaced by more suitable varieties and little, if any, still remain in cultivation.

The history of Co.301 is similar to that of Co.281. After producing good crops for fifteen years or so it also failed. Being very susceptible to smut disease was one of the major reasons for this failure. No plantings of any significance have been made since 1952.

The following is an example of the collapse of Co.301. In 1951 a 23-acre field of Co.331 was planted, in the middle of which a three-acre strip was, at the same time, put under Co.301. In 1959 the Co.33I, as 29 month old 4th ratoon, gave 43.9 tons per acre while all that was left of the 301 were a few struggling leaves.

A most prolific grower which is producing as well today as ever it did. In very sandy soils it will give a satisfactory crop where other varieties fail. Unfortunately the sucrose content is low and the yield of sucrose per acre poor, as shown in Table I. Another disadvantage is that it is difficult to get good weights on trucks or lorries due to the sticks being light and springy and thus difficult: to pack. Consequently the output per labourer is less than in other varieties with an increase in harvesting costs. Co.331 will rapidly dry out, and a lot of sticks will die, if it is allowed to go beyond the peak of growth, therefore it should not be left to stand over in times of surplus.

The following figures from a 50 acre field show that there is no decline in the yield of Co.331. Planted in 1948 the yields, in per tons acre, from subsequent crops were:

Naturally the different ages at which the crops were cut had a bearing on the result, furthermore fertilisation has been heavier in recent years; however, it is interesting to note that the 2nd ratoon, cut. at 22 months, gave 30.9 tons per acre while the 5th ratoon, also cut at 22 months, gave 38.0. Nevertheless, mainly because of the low yield of sucrose per acre, Co.331 will not be planted as extensively in the future as it was in the past.

Reference to Table I will show that this variety has given the best results of all every year since it was first harvested in 1957-58. It is rapidly gaining great popularity throughout the industry and is the one cane that may shortly exceed N:Co.310 in the percentage under cultivation. It is the next; variety after N:Co.292 to show drying of the leaves under drought conditions, but quickly recovers. An easy cane to handle, the sticks are straight and the crop does not lodge. A bright future is anticipated for this variety.

135

Of the same parentage and year as N:Co.292 it has proved the more popular especially at the high altitudes. It is a slow grower the first year.

The first variety to show a browning of the leaves in dry spells it quickly recovers after rain. A straight cane, it packs well and good truck weights are obtained.

The most successful variety ever released. The phenomonal increase in the production of the Natal Industry during the last decade is largely due to the cultivation of this variety which has also been favour-ably received in other parts of the world. In times of surplus it has been found that N:Co.310) can be left standing three years or more without deterior-ating. Although it sometimes flowers profusely this does not appear to have any serious ill effects. This variety is very susceptible to gumming disease especially in the high altitude mist belt areas. It has been found, however, that infected stools throw off all visible effects of this disease when adequately fertilised in good growing conditions.

Some growers maintain that N:Co.310 is now showing signs of decline but in the author's experience this is not the case. Table I shows that in 1956-57 the yield of sucrose per acre was 4.59 tons while in 1959-60 the figure was 5.38 tons.

This is giving satisfactory results under all soil conditions including light sands where it is second only to Co.331. It is an erratic germinator and for the best germination results the setts should be very lightly covered with soil, from 1" to 2", at time of planting. Although very susceptible to mosaic disease it is very tolerant. The danger exists, however, in the possible spread of mosaic to varieties that are intolerant. Withstands dry conditions well.

Not much is known as yet about these recent releases although both appear to be growing well. 334 does well in low lying swamp areas. Recently some 382 over two years old was used as seed, as there was no other seed available. This was planted with some trepidation but the subsequent germination was excellent.

No information can be given on this the latest release. According to the Experiment Station2 the following are the characteristics:

Not superior in yielding ability to the best of the present commercial varieties.

Resistant to the major sugarcane diseases.

Not recommended for high altitudes or coastal

sands.

Sheds trash freely and flowers profusely.

A medium to thin cane it grows to a good length, It is inclined to lodge late in the growth cycle.

T A B U : I

VARIETY YIELDS IN TONS SUCROSE PER ACRE 1956-57 to 1959-60

The above figures represent the findings from a total of 3O, 163 acres of cane harvested for the four years under review.

Taking the present average price of sucrose at per ton the difference in yields of 0.l ton of sucrose represents a difference in income of

TABLE II

Comparison of variety yields in tons cane per acre, tons sucrose per acre, sucrose per cent cane and area of variety reaped per cent total crop reaped, for four years 1956-57 to 1959-60.

136

TABLE III

Average performance of varieties over four years, 1956-57 to 1959-60, listed in order of yield of tons sucrose per acre. (Excluding N2Co.334 and N:Co.382).

TABLE IV

The following figures show the percentage of the total yearly planting planted to the various varieties for the eleven years 1948-49 to 1958-59 and indicates the trend of popularity.

REFERENCES 1 McMartin, A. (1940). The Sugarcane Varieties of Natal.

Proc. S.A. Sugar Technologists' Association.

2S.A.S.A. Experiment Station (1959). S.A. Sugar Journal, Volume 43, No. 7, page 597.

137

TABLE 11- -continued.

Mr. du Toit (in the Chair), said he was pleased to hear that N:Co.310 was the most successful variety to date and that it was still good today although later it would probably be superseded by other varieties. Mr. Steward had already corrected himself in stating that for many of these varieties, such as N:Co.376, the comparison was between plant cane or younger ratoons in the case of the newly released varieties and rather older ratoons in the case of the older released varieties. The comparison between these varieties and the old canes must always be very difficult because one has got such a predominance of plant cane in the newer releases. The only true comparison one could make was between yields of the same ratoon, but then of course the areas under cane were small and the comparison might also be unreliable.

Mr. Steward said that although he had said N :Co. 293 was a slow grower in the first year he agreed with Mr. du Toit that it was in fact a good grower the first year, but he had noticed that, in spite of formation of the cane stick during the first twelve months, the crop put on a lot of weight in the second year, more so than that of other varieties in their second year. Regarding the possible decline of N:Co.310 he said that the variety had served us well for 15 years and, as far as his figures showed, was still giving satisfactory results, whereas varieties Co.281 and Co.301 had completely disappeared after 12 years or so. As far as Uba was concerned he admitted that he could not say positively that it was immune from smut disease, but from his experience he had never observed smut in Uba even in the days when the whole industry was planted to that variety. The first smut he had ever seen was found in Co.301.

Mr. Roy Halse commented that he had found smut in Uba but it was never really very serious.

Dr. McMartin said that Uba was planted because of its immunity to mosaic. It was susceptible to smut but smut was not one of our problems when Uba was grown, being in a quiescent state. In other countries, as in Portuguese East Africa, Uba was riddled with smut, at the same time as it was still being fairly largely cropped in Natal. Concerning the statement that N:Co.293 was a slow grower there are two types of varieties—those which tiller early and produce the stool and then form cane, and others which form cane earlier instead of tillering profusely, and he was of the opinion that this variety was of the latter type. Mr. Steward said that some growers maintained that N:Co.310 was deteriorating, whereas in the author's experience this was not the case. Looking back over the years it seems to have been the experience that the decline of any variety has been preceded by a lot of argument as to whether it was declining or not, and the fact that the argument had already begun suggested that

N:Co.310 was not going to be a permanency any more than any other variety.

Mr. Steward said regarding the decline of N:Co.31.O, he had made a point of saying earlier on in the paper that he was only concerned with the areas over which he was in charge and also said it was likely that in other places different opinions would be found. There was no smoke without fire and, as Dr. McMartin said, the very fact that deterioration of N:Co.310 was being discussed up and down the Sugar Belt, pointed to there being something in it. He felt however, it should still have its champions as it was the one cane that did put us right on top. It may be superseded later on but it had lasted us well for 15 years and was still going strong.

Dr. Cleasby, commenting on N:Co.3IO, said he considered that it was going out and it was the Tongaat Sugar Co.'s policy to plant limited amounts only in very good areas. In the last few years, weekly factory sucrose figures for N:(Co.3lO had shewn decreasing superiority over other varieties and one of the first things that happened when a. variety started to deteriorate was that the sucrose content fell off. With regard to N:Co.334, it did well in wet areas and under irrigation. From this cane over 70 tons per acre had been reaped in 16 months. He considered that under these moist conditions ratoon stunting disease should not be a major factor with this variety. A disturbing feature at the moment was the way in which N:Co.376 was becoming infected with mosaic disease. The new variety No.50/211 looked most promising wherever it had been planted at Tongaat.

Mr. Steward said he thought that people would benefit from the remarks and observations which had been passed.

Mr. Hyde Palmer noticed there was no reference to ratoon stunting in N:Co.310. In his experience he found this to be quite bad and felt that ratoon stunting disease was the primary cause of deterioration in various varieties. Cane that had been heat-treated showed a remarkable improvement.

Mr. Steward said he remembered some years ago Mr. N. C. King came out round the fields with him and they went from variety to variety. Mr. King examined dozens and dozens of sticks of all varieties and although he was careful not to commit himself definitely, he said as far as he could see every stick of every variety had ratoon stunting.

Mr. G. M. Thomson stated that with ratoon stunting disease and the running out of varieties he did consider that ratoon stunting disease was only one of the factors which caused running out of varieties. It was not the only factor to be considered but the running out of varieties could at least be delayed by heat-treatment. He considered that N:Co. was deteriorating, and he felt the

138

139

other varieties should be closely watched, especially N:Co.376. Dr. Cleasby rightly mentioned mosaic disease in N:Co.376 and he considered the dangerous cane in this connection was N:Co.339. Last season was a bad one for mosaic disease. N:Co.376 was fairly resistant but could be badly damaged by disease. It must be remembered too that different strains of the various diseases could make themselves apparent. With large amounts of N:Co.339 in the industry there was the chance of mosaic disease rapidly developing.

Mr. Wilson said that an attempt had been made by Dr. Brett to assess the relative merits of our current varieties in the environment to which they were best adapted and it would appear that a good deal too much has been expected of varieties like N:Co.310 which had been widely planted throughout this Industry in areas to which they were not suited. Tongaat had now arrived at the correct siting of N:Co.3I0 on rich soils and varieties better suited to the poorer soils were now going to be planted in such areas. With regard to N:Co.339, he had been under pressure from certain quarters to recommend its withdrawal. The withdrawal of this variety would justifiably cause a certain section of the Industry to claim that its whole economy would be upset. The answer would appear to be, from Dr. Brett's assessment anyhow, that we have a substitute for N:Co.339 now in the variety, N:Co.382, and although N:Co.382 is still in its infancy, and while it may not as yet meet with the general approval of a lot of people, there is every indication that it will find a place, certainly to the extent that N :Co. 339 has occupied, and there might be some justification for seriously considering the withdrawal of N:Co.339, at least from certain areas. There was a tendency, as he had said, with a good variety to plant it everywhere and hope for the best. He did not think this was sound practice. It was essential to get the right varieties in the right place and anything we could get in the way of information to achieve this end was extremely valuable. He was extremely interested in the Chairman's and Dr. Cleasby's views on the recent release of N.50/211. Dr. Brett, he knew, issued this variety with some reluctance. He felt it had been insufficiently tested, a not uncommon difficulty at the Experiment Station, which is constantly under pressure to speed up the release of new varieties. N:Co.310 was one variety he believed, which was only tested for 8 years instead of the desirable 12 to 13 years. He hoped that N.50/211 was another one of those successful varieties and that it would become as useful as N:Co.310.

Dr. Dodds said that the point raised about Uba and smut was the fact that it was not infected much in this country. He had seen thousands of acres of Uba in P.E.A. totally destroyed by smut. He did

not know if this was the same species of fungus as that affecting the canes in Natal. Natal was fortu-nate in that at one time they had received a large number of untested seeds of the cross Co.421 and Co.312. It was one of the most promising crosses one could imagine. Neither parent was any good in this country but each had been outstanding in its own sphere in India, Co.421 in the tropical zone, and Co.312 in the sub-tropical zone. All our N:Co. varieties came from this source.

Mr. Grice said that The Natal Estates Limited had found the variety N :Co.37(5 to be most promising. The planting of Co.301 had been discontinued because it was extensively affected by smut disease, and it was feared that this disease might spread to other more valuable varieties. N:Co.293 had proved to be an excellent cane under irrigation, where it was possible to harvest two crops in three years. Unfortunately, it was now evident that this variety had lost its resistance to smut and as a result was no longer being planted. The question was asked as to whether anyone present had found that this valuable variety was no longer resistant to smut disease?

Mr. Udal said that at Sezela they found that N:Co.293 was very susceptible to smut.

Mr. Pearson said that referring back to the original issue of N:Co.293, when it was advocated for high altitudes, the Experiment Station was well aware that it was subject to smut. Where it was planted in a field that contained Co.301 they got endless smut. In the more recent variety releases the cane had been heat-treated. He wondered how much attention was given by the planter to heat-treated cane when he was cutting up setts, and if the knives used were properly sterilized before the setts were cut.

Mr. P. F. Boulle asked how one could differentiate between coastal and non-coastal soils.

Mr. Wilson said that from Dr. Brett's table N:Co.293 came top only on better soils at high altitudes. N:Co.293 was highly sensitive to environment and unless planted in the right area it could be an unsatisfactory variety.

Dr. McMartin said that he recollected when N:Co. 293 was dropped out of trials at the Experiment Station. It was a poor yielder in the lower coastal areas, including the Experiment Station, and it was only because of an outbreak of red rot at Eshowe that this variety, and others, were planted there to see how they would do under the conditions in which red rot was destroying the commercial varieties. Under these conditions N:Co.293 was found to be outstanding.

Mr. de Robillard referring to Table III said that in Natal Estates they had similar results. He agreed with Dr. Cleasby that the sign of N::Co.310 deteriorating was that there was a drop in sucrose. The figure for Co.331 did not quite agree with this table. He said that it used to only do well on coastal sands but was now doing well on inland soils. If one calculated the growth per acre per month it seemed that N:Co.293 was one of the best.

Dr. Cleasby supported Mr. de Robillard that Co.331 had responded very well indeed to irrigation. It

was a variety which undoubtedly should be kept going.

Dr. McMartin said, referring to smut disease, that mosaic had spread all over the world with the spread of noble canes, and the suggestion has been made that with the increased use of wild cane in breeding everywhere smut may be expected to have a wider range. He himself felt t h a t this was a disease which could slip through present quarantine pro-cedure.

140

YIELD DATA FOR EXPERIMENTS, HARVESTED AT ILLOVO, 1958—1959

By C. G. HALSE and G. D. THOMPSON

Introduction

The plant cane stages of a considerable number of experiments have been harvested at Illovo in the p a s t two years. This paper deals with the results of these experiments, pertinent information concerning them, and the translation of the responses to treatment into improved field practice. Thirteen experiments are summarized, each illustrating different interesting features which may benefit the testate's subsequent development. It is possible t h a t similar experiments have been carried out elsewhere in the sugar belt, and these data are presented to swell the fund of general information.

The paper is sub-divided into three sections:

(a) Nutritional Experiments.

(h) Variety Trials,

(c) Miscellaneous Experiments.

It should be noted that throughout the paper "significance" refers to statistical significance at t h e five per cent level and "high significance" to s tat is t ical significance at the one per cent level. In all discussions "tons sucrose per acre" is used to compare yields, although "tons cane per acre" is a l so given in the summaries of results.

P A R T I

NUTRITIONAL EXPERIMENTS

(i) 4 x 2 x 3 NPK Experiments

Three experiments of this type have been harves ted . The designs are the same as those of the industry-wide "Regional Fertilizer Trials", and the resul ts are therefore suitable for integration with those from similar climates and soil types. The experiments are intended to define the optimum levels of the three major nutrients for the conditions u n d e r which the trials are conducted.

EXPERIMENT No. 1 This experiment was carried out in the Dalton

a r e a , which is approximately 3,000 feet above sea level . The land had been under continuous cultiva t i on for many years prior to the planting of sugarcane, the previous crop having been maize. D u e to the sandy nature of the soil, variety Co.331 w a s selected for the experiment.

TABLE 1 Planted: 26/9/56 Harvested: 13/10/58

TONS CANE PER ACRE

TONS SUCROSE PER ACRE

Treatments/Acre

Analysis and Discussion The only significant response obtained in this

experiment was to nitrogen at the N2, N3 and N levels over the N1 level.

The apparent responses to phosphate and potash treatments were not statistically significant. The response to nitrogen is shown more dramatically in the following summary:

TABLE II

141

EXPERIMENT No. 2

This experiment is at approximately the same altitude as Experiment No. 1, but the land had been under wattle for numerous years prior to sugarcane cultivation. The experiment is in the Wartburg area on a deep red lateritic soil, for which the variety N:Co.293 was deemed suitable.

EXPERIMENT No. 3 This experiment was conducted in the Powers-

court area, some 25 miles inland from Illovo at an altitude of approximately 2,500 ft. The area has been under continuous sugarcane production for many years. Variety N:Co.293 was planted on a Mist Belt T.M.S. soil.

TABLE III Table IV Planted: 27/9/56 Harvested: 29-31/10/58 Planted: 1/11/56 Harvested: 17/11/58

TONS CANE PER ACRE TONS CANE PER ACRE

TOMS SUCROSE PER ACRE

Analysis and Discussion

This experiment showed no statistically significant treatment effects. It is perhaps unfortunate that a zero phosphate treatment was not included as there appears to be a beneficial trend with increasing phosphate nutrition. Visually apparent responses to high phosphate treatment have been observed on a neighbouring farm with similar soils. This might be expected due to the high sesqui-oxide content of these lateritic soil types

TONS SUCROSE PER ACRE

Treatments/Acre

Analysis and Discussion

The data for both tons cane per acre and tons sucrose per acre show a highly significant linear response to potash treatment, and there remain, the possibility of even greater responses at a higher level of treatment. The least significant difference; at the five and one per cent levels were 0.72 and 1.00 tons sucrose per acre respectively. This experi-ment probably shows one of the most remarkable responses to potash since this element was realized to be a limiting nutritional factor in numerous areas on the sugar belt some ten years ago. The economics of the response are summarized in Table V.

142

Treatments/Acre

148

Acknowledgements

The authors wish to thank the S.A S A Experi ment Station staff for conducting the sucrose analyses and the subsequent statistical analyses of the majority of these experiments, and for the interest they have shown in our research programme

Mr. Thompson said that he agreed with Mr. du Toit's remarks about the non-inclusion of zero levels of nitrogen in the Regional Fertiliser Trials and considered it a great pity that we did not have these treatments. As far as responses to nutrients other than nitrogen at Wartburg and Dalton were concerned it should be pointed out that these might well develop in the ratoon stages. It was true that potash responses were not significant at. either Wartburg or Dalton but it might be pointed out that in those two areas cane had been planted for the first time, the previous crops being in the one instance, maize and in the other, wattle. At Powers-court it was cane after cane and cane had been grown for 50 years, so it was not surprising tha t responses to potash were obtained there. In regard to the economics of the responses to t reatments at Dalton he had worked strictly to tons of sucrose per acre and not to tons of cane per acre. This was the criterion which was their accepted yardstick to judge results by. There had been many discussions about whether or not the sucrose data of small plot experiments could be strictly accepted but that was the present policy. The average sucrose per cent cane figures for the molasses treatments were: zero treatment, 16.32; 500 gallons per acre, 10.41; 1,000 gallons, 15.90; 1,500 gallons, 15.51; and for 2,000 gallons, 16.12.

Dr. Cleasby considered that in the Powerscourt experiments, a most interesting result was that of Experiment V. There was a tendency in the industry to swing away from superphosphate and to use rock phosphate instead and also to use increasing quantities of lime. The experiment shewed that superphosphate was equivalent to rock phosphate and that applications of lime had not produced a significant response even though the soil pH was of the order of 4.8. There seemed to be little response to phosphate in the nutritional experiments reported and he asked for further information on the level of phosphate in the soil.

Mr. Thompson said the levels applied at Wartburg and Powerscourt were made on the basis of the Experiment Station recommendations and it was therefore decided to use the two treatments, 100 and 200 lbs. P205 per acre, and no zero treatment.

Mr. Pearson pointed out that in the comparison between superphosphates and other phosphates and the filter cake treatment it should be remembered that filter cake does carry a certain amount o f nitrogen. He wondered if this was instrumental in raising the yield from filter cake. Most trials were carried out during a high, rainfall period, 1958-59. In a paper produced by himself regarding .spacing it appeared that water available was most important. With narrow spacing more water was required.

Mr. Thompson said that he had deliberately tried to mask the effect of nitrogen in filter cake in the

149

experiment shown in Table VII by putting on 300 lbs. ammonium nitrate per acre on all treatments. He thought that this met the requirements of a p lant crop on a soil of that type so that he could only hope that they did not have the effect of nitrogen in filter cake there. As far as rainfall in this area was concerned, the population and fertiliser tr ial was in the same area as the regional fertiliser trial , which was in the previous cycle. The yields for N:Co.293 in 1958 were actually higher under maximum fertilisation than he got in the experiment with 3 ft. 6 in. rows in 1959 so that he thought 1958-59 was not an optimum rainfall cycle. The area was in the mist belt where the rainfall was good and the soil very deep. He had never seen any sign of drought there.

Dr. McMartin thought that as far as the row spacing was concerned the factor really affecting the yield was not necessarily so much the amount of cane planted as the number of stalks that the field could support. That was limited by factors such as fertility and moisture, particularly at the time when these factors were limiting, such as during winter . If the winter check was followed by good growing conditions in spring one had a big flush of bull shoots and the position righted itself again. He thought that the correct amount of cane to plant, was the minimum amount that would give the maximum leaf development at an early age over the field, and closing in of rows to absorb the radiant energy.

The age at which the rows close in was another factor affecting the distance one should space the plants . In Java it was found that when they replaced some of the older varieties with P.O.J.2878, which was a plant with upright leaves, they could plant closer. He though Dr. Dodds would remember some of the older experiments at Umfolozi where it was found that one could profitably reduce the row width of Co.281, but not with some of the other varieties.

Mr. Wilson said he wished to express gratitude to Illovo Sugar Estates, Ltd., for allowing the authors to record these experiments for our benefit. It was not often that management allowed these things to be published. His particular interest was in the plant population experiment which he thought pointed out possibilities of following out the recommendations of the chairman of the South African Sugar Association's opening address. He intended following up this question of spacing as he was not convinced t h a t moisture at the rate of 35" per annum on an average was insufficient to supply a much greater p lant population than was considered normal at the moment. He asked a question following the very last sentence of the paper on page 147, which stated it was doubtful if the result was due to "a beneficial effect from hand cultivation, or the depressed

yields due to the harmful effects of weedicide". His own experience with weedicides had been confined to maize and although weedicides have always been quite effective they were uneconomic at the price in relation to the wages of labour. There was always a benefit from cultivation per se as distinct from weeding. This raised the point that in any intensive form of production it may be an advantage to ensure that conditions were such that one could cultivate. It might be necessary to get rid of trash in order to do so under certain circumstances.

Mr. Thompson replied that the nature of the experiment did not permit of a detailed analysis to be made of possible bad effect of weedicide, but from observation he would say that for the most part the poorer yields were due to the harmful effect of the chemicals applied. The weeded cane gained something from cultivation. With regard to cultivation, was it not true that there were higher yields from narrow row widths under irrigation? This would lead one to the conclusion that available moisture could be the deciding factor in the selection of the best row width for dry land conditions.

Mr. Pearson stated that when he had an experiment on 1'6" spacing it was always obvious that some other factor was coming into the picture. Where he had used this 1' 6" spacing of rows only one hand weeding was done. In the narrow planting it was noticeable that there was less hardening of the crust of the soil after the application of water.

Dr. Dodds asked if the authors had any plans for continuing to the ratoon stage especially with regard to varieties. He was interested to see that liming of the soil had no beneficial effect. This was done in the early days of the Experiment Station on acid soils in Canelands and Eshowe where it was found by adding lime even to a pH of 7.0 (neutralisation), no increase in yield resulted. He would like to know if anybody present had found any increase in yield from liming.

Mr. du Toit said that Dr. Dodds had raised an interesting point. That was liming. Liming had increased enormously in the sugar industry during the last year and the Experiment Station had at the moment a bulletin in preparation on that subject. The experiments which Dr. Dodds referred to were old experiments carried out on quite acid soils which gave no positive results and it was also interesting to note that the authors had limed a very acid soil. We were finding that our soils were getting progressively more acid, and that was perhaps very largely a result of the increased quantity of nitrogen now being used, with by far the worst offender, ammonium sulphate. We should know how far we could allow the soil to get more acid before we would get into real trouble, because although sugar cane undoubtedly would stand a very wide range of pH we could get to a position where the soil would be

150

too acid for cane to grow normally. We have had indications in experiments that too low a pH would lead to very slow growth. The position that we have taken up is that we need further liming experiments at this very moment and the Experiment Station is going to carry out such tests on a fairly large scale. If again we find no response, then we would have to continue for a while longer. He did not believe that we could apply lime with no research, but at the same time it was felt that as \vre had not had any lime experiments for a long time, in the interim, until further tests were made, we must at least ensure that our soils do not get more acid. This could be done by applying small quantities of lime to those soils which are already acid, so as to stabilize the pH. Lime may be necessary in certain circumstances. Another important factor may be the calcium content of the soil. In other words the Bulletin being prepared would advise that during the interim period, until we have had more data, applications of smallish quantities of lime to acid soils should be made so as to guard against our soils becoming more acid with the application of nitrogen.

Dr. McMartin said that recent plant physiological research had shewn that between the pH range of 4 and 9 no major agricultural crop was affected by pH per se, unless the intake of plant foods was affected. It was not mentioned in the paper, but it occurred to him when on the subject of acidity caused by sulphate of ammonia, to mention the feeling in some quarters that urea should take the

place of sulphate of ammonia. He had heard that in some countries they were getting disappointing results from urea, and wondered if in Natal there was any data here to confirm or dispute that?

Mr. du Toil said that in an experiment carried out at the Experiment Station there appeared to be some loss of nitrogen even when ammonium sulphate was being used.

Mr. Main asked if the Experiment Station could throw more light on the subject. He thought quite a lot might depend on the current methods of application, and urea might be most susceptible if applied during dry conditions.

Mr. Steward, about planting 1/ 9" rows, asked how the cane could be weeded and cultivated when it reached the ratoon crops, which would become a solid mass of interwoven cane?

Mr. Thompson replied to questions on the application of various phosphate forms and said that they were not repeating the phosphate treatments in the ratoons. They were putt ing on adequate nitrogen and potash and attempting to measure the residual effects of the phosphate treatments so that records could be kept of the performance without further phosphate treatment. The amount of lime used in the experiment was limited to 2,000 lbs. per acre, although a higher t reatment might have been preferable, simply to keep the total cost of the treatment down to the equivalent of 1,000 lbs. of superphosphate at the standard price.

151

SOIL CONSERVATION IN SUGARCANE FIELDS By T. D. ARCHIBALD

With the coming of the white man to South Africa with his ability to cultivate large tracts of land, which were formerly covered with thick vegetation, bush or grass, has come a tremendous acceleration in nature ' s pace of erosion.

Today the amount of soil which is being washed from our lands annually has reached astronomical figures. It has been computed that the Union of South Africa is losing something like 800,000 tons of soil each year and that if this figure were expressed in terms of acreage of average soil depth, say two feet, over 90,000 acres would be eroded down to rock or their non-productive, sub-soils. Each year, in other words, South Africa is losing this huge acreage and to that extent its productive potential is shrinking.

To bring this tragic state of affairs into even more striking reality, let us come nearer home and consider what goes clown the Tugela River each year in the form of silt. 1 understand that from tests carried out and observations made on the Tugela a few years ago, the authorities have come to the conclusion that , so great is the amount of silt passing out to sea in these waters each year that to load it into 3-ton trucks to carry it would take a string of trucks standing bumper to bumper running one and three quarter times round the world. So bad is our erosion that it is not surprising that Jacks and Whyte, authors of The Rape of the Earth should have s tated that a national catastrophe due to soil erosion is perhaps more imminent in the Union of South Africa than in any other country visited by them; t ha t the tragedy of South Africa has been the appalling rapidity with which its fertility reserves have been depleted and its thin soil covering washed away; and that in no other country have the disastrous consequences of erosion followed so quickly after i ts commencement.

These you must all agree are most staggering findings and an appalling waste of our soils and their fertility. Erosion is one of the major causes of our small tonnage per acre and consequently a high cost per ton of cane.

Wha t are we who are making a living, or should I say trying to make a living from growing cane doing to prevent or curtail this waste of our soil capital? If wc: are honest with ourselves I am afraid t h a t we shall have to admit that while we are trying to do something, we are not doing enough and could do more:.

Fortunately for us cane is a crop that is only planted or replanted once every six to eight years and for this reason can be considered a fairly good

crop to be grown on our hilly lands. Nevertheless, it is surprising how quickly a six to eight year cycle comes round again and one finds oneself back to planting or replanting a particular field. It is at th i s stage in the cycle of cane growing that most of t h e damage from erosion is done.

For many years it has been, and still is, common practice to plough out a whole field, or face of a hill, from the bottom to the top, regardless of how far the distance might be; it is usually anything from 300 to 400 or more yards. The result is that w h e n rains come, what starts as a mere trickle of water at the top of the hill breaks through to the next furrow below, gathering more and more water from each successive furrow down the hill, until what s t a r t ed at the top as a tiny trickle becomes a veritable r iver by the time it nears the bottom taking with it m a n y tons of soil, and leaving what was once a well p re pared field a mass of scars from where the soil h a s been washed. Once a field has been badly scarred one can make the field look normal again but t h e r e is nothing that one can do to bring back to that field the fertility and the soil which has been lost. Once lost it is lost for ever.

The question which arises from this and one t h a t should be exercising our minds is how do we go on growing cane on our hillsides and at the same t ime reduce this hazard of erosion over the planting period? One of the most effective methods to counter erosion is to break up one's fields into contour bands or strips. We have used this method over the past s ix years and have found the results to be highly encouraging. While I cannot state the increase in tonnages from this method, I can say that on o u r own farm, which is on the South Coast, 1,800 feet above sea level, with a granite type of soil over i t s greater portion and a small area of Table Mountain Sandstone soil, we have increased our crop average over the past three years. Where it was formerly somewhere between 25 and 30 tons per acre it h a s now gone up to 42.5 in 1957-58; 40.3 in 1958-59 and 38.8 in 1959-60. The drop here I attribute to t h e damage done by terrific wind during the May floods when nearly all, or over 90 per cent of our plant cane was blown down. I would go so far as to say t h a t this increase is very largely due to our changed methods as our fertilizer costs have definitely dropped over the same period. I believe, too, that 75 per cent of our former rate of erosion has been stopped by the strip replanting method.

On our farm which is very hilly a graded system of roads has been built up over the course of years . These admittedly cost a little more to build in t h e initial stages than a haphazard road with its ups a n d downs, but over a period of time have proved to be

152

far cheaper from a point of view of maintenance and more particularly from the ease of operating vehicles over them.

These graded roads give natural divisions for the strips or bands of cane. When ploughing out fields, only cane between one road and the next is taken out, leaving a belt of cane with its trash blanket between that road and the next one above, so dividing up the field into alternate bands or strips of prepared soil for planting and trash blanketed cane. In this way each planted strip is protected from above by the roots of the ratoon cane and its trash blanket which reduces the run-off to an absolute minimum, while the ratoon cane below, with its trash blanket, acts as a filter for any soil which might in exceptionally heavy rain be washed from the planted strips. Furthermore, as roads are usually not more than 100 yards apart the amount of water caught up from one furrow to the next, even in fairly heavy rain, does not have sufficient distance to run to accumulate enough to do serious damage.

This method does not in any way inconvenience the working of a field as a whole. Even though one might have to use small trucks and tramlines there will still be mature or millable cane but not, of course, of the same age, on each side of the line provided one does not fallow. Where fallowing is practised, the division can easily be made on the natural division of the "carry" to the trucks, with the tramline more or less one third from the bottom of the strip and two thirds down from the top. With those who use lorry or tractor-trailer transport there is absolutely no inconvenience whatsoever, whether one fallows or cuts, ploughs, and plants in the same season. The little inconvenience that might, here and there, be caused to users of tramlines is far more than offset by the saving in soils and fertility.

Mentioning fallowing as I have just done brings to mind another aspect of erosion which should claim our serious thoughts. In this hilly country of ours should we practise fallowing at all? The more I come to think of it the more inclined I am to believe that we are actually doing more harm than good by fallowing. There can surely be no argument about the fact that by fallowing we are keeping our soils in their most vulnerable stage for twice as long as we need do. I believe we should try as far as possible to cut, plough, and plant in the same season or, at most, to leave only the last field cut in the season to be planted first in the next planting season or for a fallow period of roughly not more than six to seven months. There is no doubt about it that fallowed land, in which there are no traces of trash, will erode far more easily than lands which still show signs of decomposing trash, as each little particle of trash, however soft it might be, will help to hold back and to bind the soil together.

Tramlines and roads running through fields are a great source of danger from an erosion point of view, for, with their hard surfaces it is the first place on which water will start to accumulate and run off. These roads where they run through or immediately above freshly planted cane should be covered with a fairly thick mat of trash. Even the peelings from the cane setts at planting are, in themselves, enough and it is surprising how much run-off can be stopped by this simple method. When it comes to weeding the hoed out vegetation should be left spread over these roads which will still further redact: the ra te of any run-off.

Another effective step against erosion is to keep as much trash as possible on the fields by not burning, even though a field is to be ploughed out it should not be burnt, but trashed and as much of this t rash as possible ploughed in. A heavy Rome disc will make a good job of this. For those who cannot afford this heavy equipment, an ordinary single-furrow reversible plough followed by three labourers, pushing into the furrow just ploughed as much trash as they can, will do an equally good, if not better, job. The trash pushed into the furrow will be almost completely covered by the next furrow ploughed and in this way almost 80 per cent of the trash can be incorporated into the soil. That which remains above the ground after a discing should be burnt off in favourable weather, leaving the buried trash to decompose, which it will do sufficiently in wet weather in three to four months, to allow normal further working of the soil in preparation for planting.

There are some who will say tha t the trash in decomposing will rob the soil of the nitrogen the young cane will need. This may be true, but it is very easily overcome by broadcasting a light dressing of nitrogenous fertilizer at the time of ploughing-in.

I would like to summarise by saying that t he most effective steps one can take against erosion in cane fields are: 1. Break up the fields into strips or bands to suit

the form of transport. 2. Don't go in for long fallows, i.e. two years. On

steep hills I would say don' t fallow at all. 3. Keep as much trash on the fields as possible by

not burning. 4. When ploughing out a field incorporate as much

trash as possible into the soil. 5. Keep the roads and tramlines when: they run

through or above freshly planted cane covered with trash and dead weeds.

I trust that you have found something of interest in this paper and that, at least, it will set some of you on the path of "Str ip" replanting, which I regard as the most easily achieved and yet most effective measure against soil erosion in our cant: fields.

153

Mr. N. C. King wanted to know if any implement had been devised to turn the. soil uphill rather than plough it downhill. With the. present method of ploughing it seemed that the top soil was gradually brought downhill to the valley below.

Mr. Archibald replied that he did not know of a n y ins t rument at the moment but a disc plough could tu rn the soil uphill by keeping the tractor wheels in the furrow which reduces the angle of t i l t . A wheel tractor, however, could not work under ex t reme conditions of slope.

Mr. A. C. H. Souchon said that in his experience the Ferguson reversible single furrow mouldboard plough d id the job of turning the soil up the hill ve ry well indeed. By using the simple extension on the moulclboard and by starting to plough on the t o p of t he hill it was possible to level the tractor a n d plough sufficiently for the long mouldboard and i ts extension to throw a sod right over. Disc ploughs had a tendency, through their design, to kick the sod s t ra ight up instead of rolling it over and, thereby, stopping it from falling hack into the furrow. In this sort of work it is also advisable to have a plough of the three-point linkage type and easily manoeuvrable .

Mr. Tedder said he could not agree with Mr. Archibald abou t ploughing in the trash. He found in dry years in M a y or June, after ploughing in trash, the t r a s h c a m e out of the ground as it went in. He asked if na tu r e ever provided for trash in its raw s t a t e to go into the ground. He preferred rather t h a n bu rn ing the trash to cart it off the field, onto d r y hil l tops where humus was lacking.

Mr. Archibald replying to Mr. Tedder said that n a t u r e neve r intended us to plough it at all. The field tha t was cut in May was usually planted about October. Under dry conditions one would find small pa tches of trash, which would come up to the top again b u t those could be pushed across into the furrow.

Mr. Maclver said there was only one effective way of dealing with trash and that was to chop it up. He knew of only one implement that could do that a n d tha t was the rotary hoe. The system of turning t r a sh in to the soil in winter presented great difficulties in t h a t it would not decompose.

The Chairman (Dr. McMartin) said that all conservat ion practices aimed at the minimum of tillage. T h e sugar cane grower had an advantage over other forms of agriculture in that he did not require a very fine seed b e d for establishing his crop so that a much more c rude job could be tolerated. He said that ni t rogen assimilation by pieces of trash and stubble could a lways be dealt with by the application of ni trogen. Where Mr. Archibald referred to a drop in fertilizer used, he asked if this was per ton of cane?

Mr. Archibald said that they were now using a smaller total amount of money on fertilizer because they are not now losing their fertilizer by it being washed away down the river.

Mr. Pearson referring to Mr. Archibald's idea of piling trash and weeds on the tramline, said he thought that perhaps a better idea was to allow grass to grow in the roadway.

Mr. Archibald agreed, and said that that would apply only to ratoon crops but he was referring particularly to the planting time when roads were usually widened or some road repair done. In those cases trash should be imported and spread over the road surface to stop the accumulation and subsequent run-off of water. He agreed that one should encourage grass and weeds as much as possible in the roadways.

Dr. Cleasby said that a new practice at Tongaat was to burn the last crop and replant cane immediately so that the ground was left unprotected for a minimum period. He wondered if this was a possible approach on the steep hillsides of the South Coast which, in conjunction with strip planting, would justify burning the last crop and allow easier preparation of the land.

Mr. Archibald said anything that would reduce the period of land being bare was to be desired. He did not know how much trash would be left after burning but he tried to get as much organic matter back into his soil as he possibly could by not burning at all.

Mr. Tedder said it was not easy to plough on the steep hillsides as it was on the flat. His idea was to cut the cane and burn the residual trash after rain, which got rid of the bulky portion of the trash, leaving the partly rotted trash to be ploughed in. This old trash rotted quickly.

Mr. Pearson said that with the introduction of the self-loading trailer should not roads be spaced in accordance with the length of carry of the loader rather than the contour fall of the land? In this way the tractor and trailer might be kept off the stools of cane and compaction restricted to the road area where it was required.

Dr. McMartin said that one good reason why a crop should be kept in as pure a stand as possible was to aid the plant pathologists in disease surveys.

Mr. Pearson said he appreciated the fact that N:Co.310 had got a bad reputation at Pongola when actually the cane which had smut was Co.301.

The Chairman (Dr. McMartin) referred to the opening paragraph of the paper where soil erosion was blamed on the coming of the white man, whereas in the Native reserves the erosion was even worse.

154

"HE COMPACTION OF SUGAR-BELT SOILS AT VARIOUS MOISTURE LEVELS

Bv R. R. MAUD)

Introduction Along with other forms of organised agriculture

elsewhere, the South African Sugar Industry has become increasingly aware of the importance, and indeed, the ultimate necessity for mechanisation in various agricultural practices. With the rapid progress of engineering of recent years, it is now common to observe numerous loading appliances and. cane transporting vehicles distributed throughout the field of the industry whereas in former times, loading was almost entirely manual, and cane transportation restricted to the cane line and 'gollovan', or small rail truck.

Mechanisation committees have been established within the industry to investigate mainly those aspects of field mechanisation concerned with labour efficiency. A futher aspect not being neglected, is the deleterious effect of the passage of heavy machinery over the soil, compacting and destroying its structure, thereby inducing conditions less favourable for plant growth, with resultant loss in crop yield.

The Effect of Compaction on Plant Growth With the use of heavy machinery in fields, soils are

compacted by the reduction of pore space between the soil particles. Restricted soil aeration and moisture movement are an important limiting factor in the development of an extensive root system such as sugarcane. It impairs the respiration of an established root system, thereby retarding water and consequently nutrient uptake, also preventing normal biological processes. Comprehensive field examinations of compaction by tillage and transportation machinery in Hawaii, for example, showed that where roots had grown in a compacted soil they were flattened and distorted, while those in uncom-pacted soil were rounded in cross-section. In addition to the above ill-effects resulting from compaction, a soil that has been compressed may-cause impedence to roots because of increased mechanical strength, the roots having to exert greater forces to penetrate the soil mass. In Hawaii, in a basalt soil, it was found that a soil with a bulk density of above 1.55 is impenetrable to roots while the critical density affecting root penetration is 1.4G.-

The Effect of Pressure on Soil Density The process of compaction in a soil is of course

dependent upon a number of factors, amongst which may be included the mechanical and mineral-ogical composition, the moisture content and the

natural consolidation of the soil itself. Thus compaction resulting from pressure on the soil is not directly related to pressure alone, but modified by these other factors.

The most important single modifying factor in soil compaction is the moisture content. The effect; of varying amounts of moisture in soil during compaction of Sugar-Belt soils is the subject of this paper and will be discussed below.

A further modifying factor in soil compaction is its textural or inecha.nic.al, and its iniiieralogical composition. Thus a pure sand will not undergo compaction to the same degree as a. loam in which sand and clay minerals are present in about equal amounts. On compaction of a, loam, espec.ia.lly in the presence of some moisture, the platy clay minerals are compressed into the voids between I lie sand grains yielding a concrete-like structure. In a clayey soil, with little sand, compaction is not so great as these soils are fairly coherent and little, deformation can take place. A complicating factor in this case, however, is the tendency for some clay minerals to swell on hydration yielding a hard dense structure on dehydration. It is in its effect on soil clay minerals that the degree of moisture present has its most profound effect.

Initial consolidation affects compaction in so far as an already fairly compact soil will not compact much further or to the same degree as one of low initial consolidation.

The Effect of Moisture on the Compaction of Soils The relationship between water content and degree

of compaction, when the soil is pounded by a falling hammer in a standardised manner is illustrated in the Figure I. It will be seen that increasing water content at first assists the rearrangement and closer packing of the soil particles, but above; a certain optimum water content for maximum compaction, additional water prevents further close packing by occupying the remaining pore space. As stated previously, for a given method, of compaction, the highest densities are attained in those soils that have a wide particle size range as in sandy loams. Thus from the figure it will be seen that the light Red Sand, T.M.S., Dwyka, Middle ICcca, Granite and heavy Red Sand soils fall within this group, as their particle size range is wide and they are all classed as coarse and fine sandy loams. High densities are not reached in soils that have coarse or fine; particles only. Thus the Grey Sand, as will be seen from the figure, is somewhat below the sandy loams, it being (-hissed as

http://inecha.nic.al

http://espec.ia.lly

155

a sand on particle size analysis. Similarly predominantly fine grained clayey soils do not attain high densities as can be seen from the figure in the case of the red and black dolerites. The Lower Ecca and Tugela Schist soils although containing predominantly fine grained constituents do also have larger sized particles. This accounts for their somewhat higher maximum densities as can also be seen from Figure I.

In the case of the Mist Belt T.M.S. soil, compaction is considerably less than in any other type of soil on account of the high organic matter content (±10 per cent). With increase in organic matter content compactibility decreases as has been proved by Free1 in the United States. The behaviour of the Mist Belt soil is in agreement with this.

The method on which Figure I is ..based is that developed by Proctor," and is commonly used by civil engineers in roadway construction. It consists of ramming a soil at varying moisture contents in a standard manner in a standard cylinder. This is not ideal for compaction studies involving machinery in motion, as the pressure in the latter case is applied in a different manner.

It was found by Weaver and Jamieson,7 however, that compaction under tractor wheels yields a water-content-compaction curve similar to the falling hammer method of Proctor.

The moisture contents for maximum compaction of the common Sugar-Belt soils as found by experiment are given below in Table I.

TABLE I Per cent Moisture

la,

Per cent Porosity

(a; Maximum Maximum Maximum

Soil Type Density Density Density

Tugela Schist ... Granite T.M.S T.M.S. Mist Belt Uwyka Lower Ecca Middle Ecca ... Red Dolerite ... Black Dolerite Light Red Sand Heavy Red Sand Grev Sand

1.82 1.84 1.93 1.21 1.92 1.64

90 50

1.49 1.95 1.85 1.75

24.5 13.0 11.0 41.5 U . o 21.5 12.5 30.5 28.0

6.5 16.0 10.0

32. 25.

42.5 22.9 35.9 21.8 33.9 35.2 27.2 31.7 33.5

In general the sandy loams and sands undergo their maximum compaction at about 10 per cent moisture, intermediate soils at about 20 per cent, while the more clayey soils undergo maximum compaction at about 30 per cent moisture. Mist Belt soil maximum compaction moisture is about 40 per cent.

These figures are in agreement with what is known of the water holding and field capacities of these soils. Indeed with the possible exception of the Mist Belt soil with its peculiar properties, maximum com

paction in the other soils takes place at a moisture content very near their field capacity.

Density of the soil is not the only factor to be taken into account in the study of the compaction of soils. Another important factor is the percentage porosity which gives an indication of the degree of "openness" of the soil. The porosity governs the circulation of both air and water in the soil as well as providing channels for root growth. From Table I it will be seen that the Middle Ecca and Dwyka soils suffer the most reduction of porosity during compaction. Anyone who has had experience with these soils will bear this out. Compaction affects the porosity of the other soils to varying degrees, the Mist Belt soil being again the least affected.

It is interesting to observe to what degree the porosities of the soils can be reduced by extreme compaction with adequate moisture under laboratory conditions. These: are given in Table II .

FIG. II 5.0

•4.5

=P= -+ ^m

f*r+

|SP|L. " N D C T R A S H

-* T j - i i - -|

5 6 7 8 NUMBER OF TRIPS OF TRACTOR AND TRAILER

AND 3 TON LOAD OVER FIELD

i r>8

can be seen, compaction increases very rapidly initially but falls off after about the third trip whereafter further compaction is slow. Much the same condition was obtained by running the loaded tractor and trailer across trash covered soil, the difference being that the initial compaction was not as rapid, although in both cases after about the third trip further compaction falls off. Thus there is no essential difference between compaction of bare soil and that under trash. Indeed the trash by conserving soil moisture may induce increased compaction by maintaining a moisture content near that required for maximum compaction, whereas a bare soil would have dried out to considerably below this value in the same time. The incorporation of trash in a wet soil would yield a brick-like structure on ultimate complete drying.

In the case illustrated in the figure as mentioned previously, the bare soil was at a moisture content conducive to maximum compaction, whereas that under trash was considerably more moist even yielding by plastic flow age. With lighter loads also, the initial compaction was not as rapid. From this it is evident that the number of times a soil is subjected to compactive forces should be reduced to a minimum if deleterious effects to the soil are to be avoided.

Soils compacted and subsequently broken up by subsoiling and similar operations do not give the same yield as one not subjected to compacting conditions. In Hawaii, it was found recently, that a reconditioned roadway plot yielded 79 tons per acre and 10.9 tons sucrose per acre while non compacted control plots yielded 108 tons cane per acre and 14.4 tons sucrose per acre.5 Thus it is better to minimise the risk of compaction rather than to have to recondition soils at a later stage.

Summary

The common Sugar-Belt soils when subjected to compaction, attain their maximum density at differing moisture contents. Sands and sandy loams compact most at about 10 per cent moisture, intermediate sandy clay loams at about 20 per cent while for clay loams the figure is about 30 percent moisture. Mist Belt soils are affected the most at about 40 per cent moisture. These moisture values are very close to the respective field capacities of these soils. Compaction proceeds very rapidly initially but drops off subsequently, reconditioned compacted soils having reduced yields compared with uncom-pacted soils. It is better to minimise the risk of compaction than to attempt rectification at a later date.

The Chairman, Dr. McMartin said that this was an initial step in opening up an important subject. A change seemed to be taking place recently in our ideas of the function of the soil. Now that fertiliser was applied in greater quantity the grower could look at the soil not so much as a source of minerals, but rather as the carrying agent for the minerals applied to it. Hence the physical properties of the soil should receive more attention than in the past.

Mr. Maud said it was not possible to put down hard and fast rules, such as to say that after 2.5" of rain not to put a tractor on Middle Ecca soil, because there might have been a long period of drought prior to that or conversely it might have just experienced a heavy rain. It was better in this case to express the water in the soil as a percentage rather than in terms of the amount of rainfall that it received.

Mr. Bartlett considered that Mr. Maud's work was a start in obtaining some of the necessary basic information which was required to ensure that local designers produce machinery best suited to Natal conditions. He stated that this paper, carried a step further, would be of assistance to manufacturers in tackling the problem of compaction caused by cane haulage units.

Mr. Black asked if Mr. Maud could give a pointer as to possible field management with regard to use of vehicles and other implements, particularly cane harvesting trailers. He wanted to know if it might not be better to sacrifice acreage of cane to establish roads which could remain there for the life of the farm.

Mr. Maud replied that he thought that permanent roadways were desirable. He considered that the more vehicles were confined to permanent roadways, the better. The matter required further study.

Mr. Bartlett said that the mechanisation committee had considered this problem, but had decided not to proceed any further until they had received more information such as the figures given in the paper.

159

Mr. A. C. Barnes said that this problem loomed as a major factor in sugar cane agriculture, but he did not consider that soils were necessarily damaged by carrying heavy loads, provided the loads were moved when the soil was in a good condition to carry them. Last year this Congress discussed removal of cane during harvest under the worst conditions, with particular reference to Umfolozi. Various suggestions were made, but he was not sure with what result. It was fortunate that the reaping season in this country corresponded with the driest part of the year. It was obvious that a planter could not stop moving his cane because of a shower of rain, so that equipment must be so constructed so as to impose the least compaction on the soil. He agreed that vehicles should run in well-defined permanent tracks, especially from the point of view that cane soils are required to bear ratoon crops. It should be possible to determine actual loadings on different soils so that the grower could be guided upon the equipment he should use.

Mr. Boulle said that he did. not agree with indiscriminate running over the field with the tractor, but this might be better than confining them to particular paths. It was not possible to have permanent roads when a field was planted.

Mr. Maud agreed that the more the load could be distributed on the field the less compaction could be expected, but rather than risk even a small amount of compaction over a wide area, he thought it would be better to establish permanent roadways on which cane would never be planted.

Mr. Barnes said that one question of importance was how the equipment was moved over the field.

Mr. Main said that at Umfolozi some growers tried trashing but had to give it up. This applied particularly to the fiats. On the flats he thought that compaction was a problem, but if one cultivated deeply after removing the crop, there was no sign of compaction. On silty coils conditions could be different from hillside soils.

Mr. Pearson asked Mr. Bartlett to consider a system of roads not based on the slope of the ground, but on the carrying distances of harvested cane.

Mr. Bartlett said that there was a continual demand for further mechanisation and it was desired to eliminate, wherever possible, the necessity for carrying cane manually. When further mechanisation was applied one would have to go into the question of equipment best suited for carrying cane from the field. He appreciated that on extreme slopes the cane would have to be removed manually, and that the carrying distance should be considered when laying out a system of roads.

Mr. Pearson enquired about the use of smaller vehicles carrying a smaller load but using wide tracks.

Mr. Bartlett replied that there were a number of farmers in Zululand who were1 building systems which used smaller trailers for infield work. The' merits of system would be studied this coming season.

Mr. Boulle asked if the compaction was caused by the front wheel of the tractor or the back wheel of the tractor.

Mr. Maud replied that tractor front wheels cut into the soil, but the rear wheels, which were driving wheels, actually exerted as much pressure as the front wheels as they were usually carrying a greater load. In addition much of this load was applied to the soil from the relatively small area of the tyre-lugs in contact with the soil.

The Chairman said that a. certain amount of criticism of late had been directed against the bulk density method of determining compaction. The reason for this was that in the bulk density measurement the range of figures was very narrow. It was possible that compaction could affect the root development long before the obvious effect of heavy implements was felt.

Mr. Maud replied that that was why he included soil porosity as well as maximum density. Referring to Table I, he quoted that whereas in this case the T.M.S. soil had a density 1.93 and a porosity 25.8 per cent, the natural soil in an uncompacted condition had a density of 1.70 and porosity of 33 per cent. Likewise uncompacted Mist Belt soil had a density 1.13 and a porosity of 50 per cent, Dwyka a density of 1.65 and porosity 33 per cent, Middle Ecca, density 1.65, porosity 35 per cent, Dolerite density 1.43, porosity 38 per cent, and Red Sand density 1.61 and porosity 36 per cent. The porosity value was independent of the actual densities of the soil minerals. Thus the density of the quartz particles which predominate in a sandy soil is about 2.6 while the density of the clay particles which predominate in clayey soils is usually somewhat less.

The Chairman said that compaction was spoken of as something new but he asked what the compaction might be, with a large gang of Natives cutting cane in small areas, together with animals and implements. He understood that the pressure by animals' hooves was very great, greater than that caused by wheeled implements.

Mr. Maud agreed that the pressure by animals was great. This principle is incorporated in the "sheep's-foot" roller used in roadmaking to compact, fills and unconsolidated material.

160

Mr. Palmer asked about the effect of soil compaction caused by large sized tyres as compared with track tractors.

Mr. Maud thought that compaction with track tractors would be much less than with wheel tractors. The larger the area over which a given load could be distributed, the smaller would be the compaction resulting from it.

Mr. Palmer asked about the organic matter in T.M.S. mist belt soils, and wondered what the effect of burning would be on such soils, and would not ploughing-in of trash prevent some compaction?

Mr. Maud replied that the organic matter in the T.M.S. mist belt soils as it was in the form of humus was in a different category as compared with trash. The soils that he mentioned were very springy, and were not usually very compactable.

Mr. Sexton said he had soon a shallow strip of land on which a. number of veliicles travelled. The effect of compaction on this strip was most noticeable in that the germination was better.

Mr. Maud replied that in a well formed crumb structured soil there were much larger voids than in finer sandy soils. He said slight compaction might aid germination but if one went further it would, affect germination detrimentally.

161

IRRIGATION CONTROL AND EXPERIMENTATION AT ILLOVO

By G. D. THOMPSON

Introduction Evapotranspiration or the consumptive use of

water by a sugarcane crop varies both with climate and the stage of development of the crop. Summer usage is higher than that in winter and the loss of moisture from the soil immediately after planting is limited to evaporation, whilst the loss from a full cover of green leaves is almost entirely due to transpiration. In occasional climatic cycles on the Natal South ("oast the total rainfall may possibly exceed the total moisture requirements of the crop, but the maldistribution of the natural precipitation invariably precludes maximum crop production without considerable irrigation. The control of overhead spray irrigation under these variable climatic conditions constitutes a problem which has received attention over the past live years at Illovo, where the soils range within fields from Recent Sands to heavy Lower Ecca Shales, and in depth from two or three inches to several feet.

Theory Irrigation Control Maximum production can only be realized when

the vegetatively growing crop is subjected to no moisture stress, and this condition obtains essentially between the limits of Field Capacity and Wilting Point, being respectively the water holding capacity of the soil against gravity and the moisture content when plants wilt permanently. The purpose of irrigation, therefore, is to supplement rainfall to the extent that the soil does not dry out to Wilting Point until it is advantageous for it to do so during the deliberate "ripening-off" period at the end of the crop.

Some knowledge of the progressive soil moisture content variations is essential to intelligent irrigation control. In areas where rainfall is predictably very low and may be disregarded as a source of moisture for the crop, irrigation may proceed in regular time cycles. The frequency of the cycles may be based on experience with a particular crop, its growth rate or moisture content being related seasonally to irrigation intensity, or alternatively on such soil moisture measuring devices as Bouyoucos blocks or soil tensiometers. Where rainfall is significant in amount and distribution however, irrigation becomes supplementary, no matter how essential it may be, and regular cycles of irrigation can only be practised between falls of significant rain. After every rain sufficiently great to saturate all irrigated lands to Field Capacity, the process of soil dessication

proceeds at essentially the same rate wherever there is a full cover of green transpiring leaf, and hence all such areas reach the stage of requiring further precipitation at the same time.

If a certain amount of overhead spray irrigation equipment is to serve a maximum area, it is therefore necessary that irrigation operations commence as soon as the first point to be irrigated can absorb an efficient cycle of irrigation, and any delay simply reduces tin: area irrigable or the length of drought insurance on the total area to be irrigated.

The control of established overhead spray irrigation under these conditions of climate; thus hinges very largely on being able to predict when a cycle of irrigation may earliest begin, and when irrigation should cease because the total effect of irrigation and rainfall exceeds the current soil moisture deficit. To these ends instruments were found to serve little purpose and attention has therefore been concentrated on the consumptive use of water by the sugarcane crop.

The amounts of water transpired by a full cover of sugarcane leaf, and the small amounts of evaporation from the soil beneath are functions of the prevailing climatic conditions. Meteorological data such as atmospheric temperature and humidity, solar radiation and duration of sunshine have been integrated in various formulae by Blaney and Criddle,1 Thornthwaite,2 and Penman3 to give reasonable estimates of daily consumptive use of water by a crop. However, the far simpler daily measurement of evaporation from a free water surface has been found to bear an approximate relationship which could well serve the purposes of practical irrigation control where rainfall periodically saturates the soil and thus eliminates all cumulative errors in the calculation of a soil moisture deficit.

In order to apply this method of control in practice under local conditions, certain assumptions have to be made to reduce complications without introducing very significant errors.

(a) It is assumed that all the moisture requirements of the crop are met from the top foot of soil only. This is not necessarily so, but shallower soils would require an uneconomically higher frequency of irrigation, and although deeper soils could take more irrigation applied less frequently the depth of soil on Illovo Sugar Estates irrigated land varies so greatly that an estimated one foot depth has been used and found to serve well enough in practice.

162

{b) It is assumed that Field Capacity in this top foot of soil constitutes 4.70 inches of water. This is an average of the available data for the soil types being irrigated, these being Lower Ecca Shales, Dwyka, Dolerite, Red Recent Sand, Grey Recent Sand and Alluvium.

(c) It is assumed that Wilting Point in the. top foot of soil constitutes 2.00 inches of water, and this again is an average for the various soil types. The available water supply from the soil is thus assumed to be 2.70 inches.

(d) It is assumed that the consumptive use of water from a full cover of irrigated sugarcane leaf is 85 per cent of evaporation from a free water surface. This is the approximate value calculated from the data of Fuhriman and Smith'1 in Puerto Rico for the period of maximum use prior to ripening-off, and actually refers to evaporation from a U.S. Weather Bureau pan. Cowan and Innes5 in Jamaica established a factor of 0.58 relating evapotranspiration from a full cover of cane and evaporation from a free water surface. (Factors for sugarcane from other sources range as high as 1.20, and in view of these considerable variations it is well that a co-operative experiment is now being conducted in Natal to determine the relationship under local conditions).

Partial cover by young cane leaves is estimated as a percentage of total cover each month, and a bare soil is assumed to lose 0.05 inches of moisture per day by evaporation.

(e) It is assumed that overhead spray is 75 per cent efficient, this figure having been determined by Fuhriman and Smith4 in Puerto Rico.

(/) It is assumed that rainfall less than 0.30 inches per day, unless continuous with a previous or following day's fall, may be disregarded. Other investigators are inclined to regard all measured rainfall as being efficient but the ignoring of small amounts of precipitation may be an advantage in practice to ensure the adequacy of the theoretical system of control. Rainfall in excess of that required to raise the soil moisture to Field Capacity is regarded either as runoff or percolate, and not used by the crop.

Practice The translation of the theory of consumptive

use of moisture by the crop into practice is effected very simply by conducting a day to day soil moisture profit and loss account for the first points to be irrigated in each field. The only reliable initial reference point is immediately following a rainfall more than sufficient to saturate the top foot of soil, even if a wilting condition has previously obtained. Thereafter, the estimated evapotranspiration per day is deducted each day from an initial figure representing Field Capacity and as soon as the soil moisture

deficit reaches a level equivalent to the amount of water applied in one efficient irrigation cycle, irrigation may commence.

Any rainfall exceeding 0.30 inches is added as profit" to the soil moisture reservoir, and irrigation operations are suspended only when the combined effects of rainfall and irrigation exceed the current soil moisture deficit. Should the effect of rainfall alone subsequently raise the soil moisture level to Field Capacity, then irrigation is recommenced from the first points to be irrigated in the field, and the field as a whole is credited with an irrigation cycle only if more than half of the field had already been irrigated. If, however, the effect of rainfall alone has been insufficient to raise the soil moisture level to Field Capacity, then irrigation is recommenced on the same points at which it was suspended, and this when evapotranspiration lias theoretically accounted for the amount of rainfall exceeding one efficient irrigation cycle less than Field Capacity. The field is credited with a single irrigation cycle.

A sample irrigation control sheet is shown in Appendix I.

Since rainfall is reasonably frequent during parts of the year, irrigation operations are planned only for a week in advance. In order to do this satisfactorily it is necessary to predict the approximate evapotranspiration per day for the week. At Illovo, average S.A.S.A. Experiment Station evaporation data from Mt. Edgecombe have been used in the past, the average daily evaporation for each month being multiplied by the factor 0.85 to give the estimated daily evapotranspiration. At the end of each week a correction is made for any significant difference between these data and those measured during the week in the local evaporation tank.

The data used for predicting the evapotranspiration from a full cover of cane are shown in Table I:

T A B I . K I

When the cane crop is deliberately dried off prior to harvest in order to raise the sucrose content, the soil moistun- level falls to Wilting Point. The control of irrigation during this period would normally be based on sheath moisture values, these values being

163

required to fall gradually and steadily from approximately 80 per cent to 73 per cent over the drying-off period. The only reliable drying-off period under local climatic conditions is from May to August, and it has been found by experience that irrigation can safely be suspended towards the end of May if the cane is to be harvested in the August-September period.

After harvesting a field, the estimated daily consumptive use is reduced to 0.05 inches per day, and thereafter increased monthly as the proportion of canopy increases until full green leaf cover is once1

again obtained.

Irrigation Experimentation In accordance with the general policy at Illovo

of instituting experiments to study problems of sugarcane agriculture, under local conditions, an experiment was designed and laid down to determine the adequacy of the system of irrigation control whilst also attempting to measure the value of irrigation in terms of response to water application.

Design of Experiment The site selected for the experiment was a very

gently sloping area of Lower Ecca Shale with 8-12 inches of top soil overlying a well-weathered subsoil down to three feet. The shape of the available land, bounded on one side by a canal and on the other by a road, was surveyed and the possible disposition of the 54 plots superimposed on the plan.

It was decided that the plots should be square with a quarter-circle rainer located on each corner so that the overlap design would simulate a perfectly representative area in a field irrigated with a square system of rainer points. Low pressure rainers with -& inch diameter nozzles were selected. The exact plot size was 0.02278 acres, being nine rows 3 ft. 6 ins. apart and 31 ft. 6 ins. long. The rows were planted 1 ft. 9 ins. short at each end in an attempt to reduce end effect. The breaks between plots were 8ft. 6 ins. wide.

Three levels of water were included in the experiment, the first W0, being dry land conditions relying entirely on natural precipitation. The second level, Wj, was designed to represent average irrigated field conditions, which involved irrigating so that the mid-point between Wilting Point and Field Capacity was reached half way through the application of water to the W: plots. The third level of water, W2, involved the application of water to the plots as soon as the soil could theoretically absorb one efficient cycle of irrigation to raise it to Field Capacity. This level represented the equivalent of the first point to be irrigated in a field scale scheme, and was intended in effect to test the adequacy of average field practice. The estimated consumptive use of water by the crop was calculated from average evaporation data, using a factor of 0.85,

and irrigation was based on soil moisture profit and loss accounts as in field practice. A sample irrigation control sheet for the experiment is shown in Appendix IT.

The fertilizer levels compounded factorially with the three levels of water were:

Nitrogen 3 levels: 100, 200 and 300 lbs. N. per acre. Phosphate 2 levels: 100 and 200 lbs. I\,05 per acre. Potash 3 levels: 0, 200 and 400 lbs. K2() per acre.

There were no replications, the experiment, comprising 54 plots only.

Experimental Procedure The Field Capacity of the soil on the experimental

site: was determined at live different points for the top three feet of soil, each foot of depth separately by the method described by Piper." The results are shown in Table II . The Wilting Points were not determined but assumed to be 2.50 inches for the top foot of soil.

The plots were planted on 6th September, 1957, with N:Co.310 seed cane. A complete lattice of drains 18-24 inches deep was subsequently dug to separate all plots. All of the phosphate and one-third of the nitrogen and potash fertilizers was placed in the furrow at the time of planting. A first top dressing of a further one third of the total nitrogen and potash applications was made on 20th January, 1958, and the final amounts were applied on 24th September, 1958.

Irrigation was carried out exclusively at night when there was no wind. This proved to be the severest problem in the conducting of the experiment, delays due to wind in one instance causing the W t

plots to fall to Wilting Point for a short period. The rainer standpipes were initially 3 ft. 9 ins. high, and subsequently 8 ft. 9 ins. high when the cane had

developed in height. Four plots were irrigated simultaneously, 16 rainers thus being in operation at any one time. Each application amounted to 685 gallons per plot of 0.0228 acres. This represented a total application of 1.33 inches, or 1.00 inches of efficient water per application.

The W, plots were thus irrigated when the estimated soil moisture level had fallen to 4.32 inches in the top foot of soil, and the W1 plots when the estimated level on these plots had fallen to 3.41 inches. Actual rainfall, efficient rainfall, and irrigation data are shown in Table I I I .

The plots were deliberately dried off from the end of May 1959, preparatory to harvesting at the end of August, 1959. Flowering occurred extensively during the first winter's growth.

The experiment was burnt before harvesting on 31st August, 1959. The outside row on both sides of each plot was discarded to eliminate border effect, and because of the very pronounced end effect on the remaining seven rows per plot, 5 ft. 3 ins. of row was measured off exactly at each end, cut and discarded. The area actually harvested was therefore 0.01181 acres per plot, which areas appeared to be free of edge effects entirely.

Discussion of Results A study of the precipitation data in Table I I I

reveals the following:

It will be observed that the total rainfall (79.81") for the crop cycle exceeded the total estimated crop requirements (65.64") by a considerable margin but that the poor distribution of the natural precipitation was such that 42 per cent of it was ineffective even in the dry land plots. The necessity for irrigation under such climatic conditions is thus revealed as compensation for maldistribution of rain rather than inadequate rain. The dry land plots were theoretically at Wilting Point for 1.86 days prior to deliberate drying-off of the crop before harvesting, and it is the effects of these periods of inadequate available moisture which must constitute the reason for any differences in yield due. to water treatments. The fact that 73 of these days occurred during the second peak season of growth, December 1.958 to May 1959 shows the absolute necessity for summer irrigation. Failure to appreciate the difference between total and effective rainfall during summer could lead to a false sense of security, with irrigation being suspended when its effects are most profitable.

The mean yields in terms of tons cane per acre are shown in the following table:

There is a very highly significant increase due to the first level of water, but the increase of W2 over W1 is only slight.

The linear response to potash is very highly significant.

There is a highly significant W—K interaction, showing tha t there is a far greater response to water in the presence of K than in its absence.

There is a significant P—K interaction indicating tha t the higher level of P depresses the yield in the absence of K, but seems to increase it in the presence of K.

The average sucrose per cent cane data are shown on page 166.

165

166

t reatments, in the order of 20 tons of cane per acre can be ant icipated in a plant crop, there must be some hesitation in predicting that the highly significant increases in sucrose per cent, cane due "to water application in th i s experiment are likely to be obtained in field practice. It is almost certain that the generally high sucrose levels wore due largely to the; dessicating effect of the extremely hot fire which burn t the cane. T h e moisture contents of the cane samples have been calculated, and if the averages for water t r e a t m e n t s are converted to a standard 70 per cent, the effects on sucrose per cent cane; are as follows:

A great propor t ion of the difference in sucrose per cent cane between treatments is thus shown to be due to moisture content. Since Estate harvest data do not indicate t h a t irrigated cane has a significant ly higher sucrose, content than dry land cane, it is quite possible that this response to irrigation in field practice may be obtained in terms of tons cane per acre ra ther t h a n sucrose per cent cane.

The increasing sucrose per cent cane at a standard moisture content is difficult to explain, and until sucrose studies have been conducted for the different t reatments in b o t h the first and second years of growth of the first ratoon stage, and it is attempted to relate the resul ts to climatic conditions, no full interpretation of the sucrose data can be offered, ft is important , nevertheless, to know how safety the experimental data may be used to study the economics of irrigation, and it is interesting to note the following harvest data for a hillside field of 46 acres which w a s harvested as a first ratoon, N:Co.310, both in 1954 and 1958:

The increased yield due to irrigation at the Wx

level in the experiment was 3.77 tons sucrose per acre, and for purposes of an economic evaluation a response of 3.50 tons sucrose per acre over 24 months has been assumed and found to give a profit entirely warranting the expenditure for irrigation.

Regarding the experiment as a test of the adequacy of the theoretical method of irrigation control, it is unfortunate that the factor relating evaporation from a free water surface and evapotranspiration from a full cover of cane had not been studied locally prior to the institution of this experiment. If the indications that the factor in summer considerably exceeds 0.85 prove to be valid, then the higher sucrose yield at the W2 level might be obtained even in field practice. It is extremely doubtful that such a high frequency of overhead irrigation would be economically warranted however, and it would then be necessary in practice to assume a soil depth of 18 inches or more to contain the available soil moisture if existing equipment were to serve the same area.

Conclusions The conclusions which may be reached following

the harvest of the plant stage of this experiment are: (a) that 100 lbs. of nitrogen per acre is adequate for

the plant stage of either irrigated or dry land cane on this type of soil.

(b) that 100 lbs. of P205 per acre is also adequate under both irrigated and dry land conditions.

(c) that highly economical responses to 400 lbs. of K20 per acre may be obtained on irrigated land, but the small returns in relation to outlay on dry lands probably warrants no more than 200 lbs. K20 per acre for these conditions.

(d) that adequate balanced nutrition is imperative if the value of irrigation is to be realized.

(e) that the average level of irrigation, W1 is the most economical water treatment although higher yields of sucrose per acre may be obtained at higher levels of water.

(/) that the response to irrigation at the rate of 28 inches of water per two year crop on hillside land is approximately 3.5 tons sucrose per acre where the cane suffers from no other limitation.

Acknowledgements The author wishes to thank the S.A.S.A. Experi

ment Station staff for the co-operation received in conducting the Irrigation Experiment. Both the statistical analysis of the results and the sucrose tests were carried out at the Experiment Station.

167

For discussion on this paper see page 174.

168

APPENDIX I

IRRIGATION CONTROL SOIL MOISTURE LEVELS -MARCH, 1958

169

APPENDIX II

SOIL MOISTURE LEVELS—SEPTEMBER, 1958

SOME NOTES ON IRRIGATION By T. G. CLEASBY

This paper attempts to discuss certain aspects of irrigation on the basis of limited information which has been gained at Tongaat during the past few years. The ideas put forward may or may not be agreed with, but if they stimulate interest and discussion on the numerous problems and unknowns, which arise from the application of water to sugar cane, then this paper has served its purpose.

The word irrigation has come to mean the controlled application of water to the land, that is, specific quantities of water at planned times, and any application of water which falls short of these objectives requires another word to describe it. There are two methods of irrigation—surface or furrow irrigation and overhead, spray or sprinkler irrigation. It is important to realise that either can be the right way to irrigate. Where the land is gently sloping and water abundant and cheap, furrow irrigation must be considered first and foremost. When the terrain is difficult and, in particular, water scarce or expensive, then overhead irrigation comes into its own. It is for precisely these reasons that overhead irrigation has been developed at Tongaat.

Although the following conclusions relate to overhead irrigation, the principles involved apply equally well to furrow irrigation governed only by the practical difficulties of applying precise quantities of water at each application.

Response to Irrigation

Up to the present time only a few experiments have been carried out to study the response to the overhead irrigation of sugar cane in Natal. It is interesting to examine the results which are available if, at the same time, the limited confidence which can be placed on them is also borne in mind.

In addition to the experiments at Tongaat, the Experiment Station has results from one unrepli-cated trial over plant cane and four ratoons and at Illovo Sugar Estates the results from the plant cane crop of a verv critical trial which was harvested in August 1959.

These experiments have all been carried out on shale soils, mainly shallow, with above average levels of fertiliser and irrigated to ensure that the cane suffers a minimum of moisture stress. Other soils would almost certainly give different responses to irrigation, but irrigation must first be considered for soils which are most prone to droughts and these

undoubtedly are the shallow shales. It would require a relatively low duty of water to enable; the experimental applications of water to be applied in the field. In the Tongaat experiments, water applications were based on evaporation da ta and would have required no less than 1 cusec to irrigate 200 acres. How much this can be increased without a reduction in response still remains to be determined.

* Considered the most reliable experiments.

With this in mind the results of these experiments have been tabulated in Table I and the responses expressed as the increase in tons cane per acre over control (no irrigation) per inch of water applied. The results are published in this form solely clue to the degree of consistency they show. The more reliable results have been marked and a conservative mean taken as 0.8 tons cane per acre per inch of water applied.

Effect of Irrigation on Sucrose per cent Cane

Experiments at Tongaat have shown that irrigation does not necessarily mean a reduction in sucrose per cent cane if an at tempt is made to d r y off the cane prior to harvest. If controlled drying off is practised it would appear tha t irrigation on shallow soils stands a very good chance of improving

170

the sucrose. The results are- expressed in the table below.

In their above form the results cannot be treated quanti tat ively, as irrigated refers to the mean of several different irrigation treatments. Qualitatively, however, they illustrate the point which is being made. The explanation for the increase in sucrose per cent is probably that the cane is kept in good condition during drought when it is irrigated and the green leaves are therefore able to manufacture sucrose more efficiently.

Row Spacing Trial under Irrigation

With the introduction of irrigation one of the first questions to be asked is, does one close up or open out the cane row? The same question has been asked under dry land conditions and experiments carried out indicate that varying row widths and spacing within the row did not significantly affect the yield. The explanation given by Pearson1 was that the yield was primarily limited by moisture and therefore plant population was not so important. In the same paper Pearson shows results which indicate under irrigation continuous planting in 4 ft. (i in. lines yielded better than when the setts were spaced.

An irrigated experiment at Tongaat appears to show that by decreasing the row widths yields of cane per acre can be increased. The experiment consisted of two replications of a factorial 4 spacings x 3 levels of nitrogen x 2 levels of potash. The levels of nitrogen were 100, 200 and 300 lbs. N. per acre, and potash 120 and 240 lbs. K.20 per acre. The results obtained so far have been tabulated below.

No valid reason can be given for the lower response to closer planting in the 1st ratoon crop, unless the

greater rainfall the plant cane crop received was not completely compensated for by the irrigation. Three other points are worthy of mention. Firstly, at three months the first ratoon crop spaced 3 ft. 0 in. was showing signs of moisture stress, while the other cane appeared to have adequate moisture. Secondly, the reduced weed control problem at 3 ft. 0 in. spacing compared with 5 ft. 3 in. due to the rapid closing in of the cane and consequent smother effect. Thirdly, there has been no apparent interaction between the heavier application of fertiliser and the closer spacings.

Response to Nitrogen under Irrigation

In both the row spacing trial and Irrigation experiments II and III three levels of nitrogen were used. In neither case has an economic response been observed to a level of nitrogen greater than 200 lbs./ acre and this only in the first ratoon crop. The results have been tabulated below.

On the basis of this information on the soils on which these experiments were carried out 100 lbs . N per acre appears to be adequate for the p lant cane crop and something in the region of 200 lbs./acre for the 1st ratoon crop. These indications are followed in the fertiliser policy applied at Tongaat.

Rate of Precipitation by Overhead Irrigation

The rate at which irrigation water should be applied has been given much thought at Tongaat during recent months. By far the most popular sprays used on sugar cane are large ones which g ive a precipitation rate of approximately 0—1 in. p e r hour and these have been used in the Tongaat i r r i gation schemes at present in operation. Two d i s advantages have been observed. Firstly, on hi l l sides, which consist chiefly of middle ecca shales a n d , to a lesser extent, lower ecca shales, a considerable quantity of water has been running off the surface. It has not been possible to give a percentage figure to this run-off but it is appreciable even in t h e presence of trash. Secondly, extensive puddling of the surface soil takes place under large rainers and in

171.

certain areas it has been felt that resulting soil compaction has undoubtedly retarded cane growth, tending to undo the benefit which should be possible through the introduction of irrigation. A more abstract disadvantage of large rainers has been stated in California, in that a plant can suffer physical damage which will retard growth through the impact of water and that this is a factor with large rainers. This is actually not as far fetched as it may seem, for the effect of wind in retarding the growth of plants has been demonstrated in the presence of adequate moisture at the National Vegetable Research Institute in England.

On the basis of experience in California, admittedly not with sugar cane, Gray'- of National Rain Bird Sales and Engineering Corporation, states that the infiltration rates which should not be exceeded are:

Sandy soil ... 0.3 per hour

Loams 0.2 in. per hour

Clayey soils ... 0.15 in. per hour

It is worth recalling that a rain storm which deposits 1 in. in one hour is very heavy and definitely leads to considerable run-off and washing. The same precipitation from a rainer is not deposited continuously but in 12-15 bursts. How, therefore, can the soil on a hillside possibly absorb these bursts which are equivalent to tens, or even hundreds of inches per hour? Trash helps to stem the impact but once the trash is saturated the quantity of water involved is the same.

In the infiltration process the size of the water drops is as important as the rate of application. Again, at the National Vegetable Research Institute2, lettuce plants have been successfully boosted by irrigation applied as a mist at the rate of 4 in. per hour when applications from conventional rainers of 1 in. per hour have been too heavy and retarded growth. A similar thing has been demonstrated at Tongaat in a crude way. Small sprinklers which produce small drops in the spray have been spaced close together to give a precipitation rate equivalent to a larger rainer throwing much Irager drops of water. At the end of the test which was carried out on shale soil, the better infiltration from the smaller rainers was obvious.

Recently rainers have been tested over the following ranges:

1. Large rainers supplying 260 g.p.m. at 75 p.s.i. spaced 180' x 160', precipitating 1" per hour.

2. Large rainers supplying 200 g.p.m. at 75 p.s.i. spaced 160' x 160', precipitating 0.75" per hour.

3. Medium rainers supplying 30 g.p.m. at 50-60 p.s.i. spaced 80' x 100' precipitating 0.4" per hour.

4. Small rainers supplying 0 g.p.m. at. 40 p.s.i. spaced 40' x 60' precipitating 0.2" per hour.

The following advantages have been observed with the small rainers:

1. Puddling and run-off is reduced to nothing. In addition, the jet of water which occurs on opening and closing large rainers is avoided.

2. Less power is required for irrigation due to the lower operating pressure required for small rainers which work satisfactorily from 30-40 p.s.i. (see comparative horse power requirement at end of paper.) This is very important where the cost of power is a major factor in the development of irrigation.

3. Water distribution is improved under windy conditions as 100 per cent overlap is allowed for in the spacing, i.e. one rainer throws at least to its neighbour. This cannot be done with large; rainers where 20 per cent is normal.

4. Easier control is possible over when' the rainers are working. Fewer moves are involved, per day as the rainers stay in one position from 4-6 hours.

5. Small rainers need not be any more expensive than large ones in labour. If capital is available to buy sufficient portable piping and rainers, operating labour can be substantially reduced. The ultimate conclusion of this is the solid set systems used in America where labour is very expensive. The whole area to be irrigated is covered with permanent pipes so that only rainers have to be set up, even this is avoided in some cases, and irrigation becomes a matter of turning taps on and off.

6. Finally, an important point which must be mentioned although it is not an advantage of small rainers over large rainers, is that very few stoppages have taken place due to blockage or breakdown during the six months the small rainers have been under test at Tongaat.

The disadvantage of the small rainers is that they require large amounts of portable piping to distribute the amount of water which can be sprayed by one large rainer and this is expensive. A considerable saving can be effected by using pipes of decreasing diameter as the flow decreases.

In conclusion it is interesting to compare capital cost of two 100 acre schemes, one using large and the other small rainers, when the water consumed is 240 g.p.m. and the portable lines are 800' long. If it is presumed that the scheme is close to the water source and that there is a nominal lift of only 20 ' , then the breakdown of costs is as follows:

172

Labour to construct the scheme and erect a pump house will be extra, but the same in each case;.

The comparative capital installation costs therefore differ only very slightly. It may be possible to reduce the cost of the portable pipes for small rainers by using galvanised or even plastic pipes in place of aluminium. This is being investigated.

In conclusion it must be emphasised that small rainers and slower precipitation rates are still in the experimental stage at Tongaat. In the near future, 500 acres of land at present under large rainers will be converted to small ones. The soil in this area is a heavy clay and slower infiltration should be of benefit. An experiment will be incorporated in this change-over and two adjacent areas will be observed, one irrigated by small rainers and the other by a large rainer. This is not an ideal test but over a period of time should provide some information.

The Chairman, Mr. Du Toit, thought it most appropriate to have two papers on overhead irrigation, particularly following the two seasons of drought we had experienced. The two papers were supplementary. He thought that they would be read with great interest in every sugar country where irrigation was necessary. The rainfall in the sugar growing areas of South Africa was one of the lowest and it was also true that water was not very plentiful for irrigation purposes. No doubt, due to this, overhead irrigation had been more developed in this country than in other parts of the world. Dr. Cleasby, referring to the effect of irrigation on sucrose, stated correctly that provided there was a reasonable drying-off period one might

expect an increase in sucrose content. This was not altogether surprising because irrigation kept the leaves green when solar radiation was high, whereas without irrigation in drought, the leaves could not work so efficiently.

Mr. Thompson's results showed a most outstanding increase in sucrose content, through irrigation. Mr. Thompson explained this by saying that the irrigated plots bearing heavily suffered a greater drying out due to the experiment being burnt, and that this might not. always obtain in practice. While these results could not be guaranteed in field' practice, it was likely that some increase in sucrose content would be obtained. The figures shown by Mr. Thompson were perhaps accentuated by the severe heat of the accidental fire. An interesting point which he could not explain was that Mr. Thompson had indicated a much better response to potash under irrigation than under dry land conditions. Naturally with the heavier yields from irrigation one would obtain a heavy response to potash, but: potash has been said to be a drought insurance.

Mr. Dymond said that with regard to the spacing experiments the fertilizer was quoted as being put on at a rate of so much per acre. Would it not be better to express this application as rate per line?

Dr. Cleasby said that this could be done but the experiments had compared different levels of nitrogen up to quite high applications and did not give any significant interaction with closer spacing.

Mr. Pearson said that in 1950 he had made enquiries as to the increases which could be attributed to irrigation and had found little more than good guesses. The papers indicated the intensive study of irrigation in the cane belt that had taken place in the last ten years. Due to queries by people who had used irrigation he had installed irrigation plants at the Experiment Station farm at Chaka's Kraal in order to find just how much gain could be attributed to irrigation. The figures so obtained were perhaps not very reliable. In the Experiment Station experiment they had endeavoured to cut a crop of about 50 tons per acre and he wondered if this might have led to the high yield per inch of water shown in Dr. Cleasby's paper. At Chaka's Kraal in the early stages irrigation was applied only during the winter period and this might have had a greater effect on the amount of water used per ton of cane. From information gained here and overseas it would appear that a better control of irrigation should have been used at Chaka's Kraal. He was pleased to see the trend away from the big sprinkler. Tests which he had carried out between big and small sprinklers were now confirmed by the tests at Tongaat. He had been using one set of galvanised pipes for just on nine years, and asked if rumours could be confirmed that aluminium pipes had been giving a certain amount of trouble over recent years.

173

Dr. Cleasby said one important point remarked upon by Mr. Pearson was the frequency of cutting of cane". He felt that one had to be cautious, for if the annual rate of growth was linear and one got 30 tons in 12 months and 60 tons in two years, then cutting twice in two years incurred additional expenses such as fertilizing and weeding, and would therefore be less economical. With regard to aluminium and galvanised pipes, when large rainers were used portable pipes of 4 and 5 inches diameter were required. Aluminium had to be used as galvanised piping was too heavy. With small rainers however, one could use 3 and 2 inch diameter pipes and therefore galvanised iron was possible. One big disadvantage of aluminium pipes was that once they were broken they were very often a write-off. Some plastic materials and fibre glass had been tried to repair them, as yet without real success, and if one tried welding, the join only lasted for a short while, the pipe soon breaking again near the join.

Mr. Thompson said that his criterion with regard to age of cutting of cane was that cane before being cut should be preceded by a summer period of good growth. If one did not have a good cover by the 15th January one would be advised to leave the crop until the next year.

He said with one year old cane on irrigated flat-lands he got 40 tons per acre with a reasonably high sucrose due to an adequate drying-off period. Cutting at a year old they spent more money on fertilizer, but if they ran the crop for two years they reaped about 63 tons per acre, as compared with a 40 ton per acre crop for 12 months. He said that at Illovo they tended to cut cane at two years old on hillsides and one year old on the flats.

Dr. Brett asked whether any experiments were carried out, as in Hawaii, to control flowering by withholding irrigation at certain times of the year.

Mr. Thompson said he was not encouraged to try this after discussing the matter with Hawaiian technologists because here we would have to cut off the water in February, when the cane needed it most, and when fairly frequent rain could be expected anyway.

Dr. McMarrin said that on the question of vield of cane per inch of water, he had averaged out pot tests on the transpiration ratio and had got a figure of 1.2" of water per ton of cane. This figure however should be treated with caution because vield is not directly proportional to the amount of water used. As there was a different water requirement for each tonnage, he did not know if such figures meant very much. He also was not happy about the use the wilting point was put, to determine the water requirements for cane. There was not one point onlv at which the cane was affected by lack of water. There

was one point at which cane would wilt but quite a lot of cane growth would be lost before it actually showed wilting. While this might not be so important in areas such as those discussed, where supplemental irrigation only was practised, it was of vital importance in certain areas which were something like 100 per cent dependent on irrigation.

Mr. Thompson said that the wilting point data was obtained by experiments on young sunflower plants. In practice, the difference between wilting point and the slowing up of cane growth was not so important. Using an average figure of 2.7" for available water from one foot of soil might not satisfy Dr. McMartin's requirements but in practical working one had to use some definite figure.

Dr. Cleasby agreed that the figures shown by himself and Mr. Thompson were not accurate, but it was important to have a system to work to and later, with further information, one; might be able to use more, correct figures. He thought Mr. Thompson was justified in taking 2.7" as available moisture. If one changed from a heavy soil to a sandy soil the plant would adjust itself by spreading its roots much further in the sandy soil and so obtain the necessary amount of water. He thought that for response per inch or irrigation, the figure .8, while rough, would be of use to people contemplating irrigation as an indication to what return they might gain. It implied that scientific irrigation was practised and control based on an evaporation approach as outlined by Mr. Thompson in his paper. This would require a water duty of the order of one cusec to 200-250 acres.

Dr. McMartin asked if the figures of 1' for root development and 4.7" for field capacity were correct. Could not 4.7" be too high?

Dr. Cleasby replied that 4.7" was field capacity and 2.5" was the available water. He said it was most amazing how this actually worked in practice. He said laboratory tests indicated tha t when cane showed signs of drought, the laboratory records showed the moisture deficiency, based on evaporation, to be of the order of 2.5 to 3.0 inches.

Mr. Tedder asked what could be considered an economical tonnage to cut at 12 months.

Mr. Thompson said he would, rather cut two 40 ton crops per two years than have to irrigate through winter and wait two years to get a 63 ton crop. The financial yield, of course, depended also upon distance from the mill.

Mr. Pearson asked where a person was compelled to cut irrigated cane, say in June or in May, whether Mr. Thompson had got any experience of whether or not there is an increase in growth due to water application for the young cane in the winter period.

174

Mr. Thompson in replying said that they had much experience in the effects of irrigation during winter. This was judged by crop logging results and particularly the sheath moisture tests. Additional irrigation was not found to prevent desiccation of the plants during winter when sheath moisture levels invariably fell below 75 per cent. In fact, soil temperatures seemed to be a limiting factor during winter. The crop log growth measurements shewed no increase during winter.

Mr. Pearson said it would seem that while temperatures were low irrigation would be unprofitable to apply , and that there was a critical point where one should take advantage of this temperature rise.

Mr. Thompson said he considered one should irrigate during winter, but if the amount of water put on was excessive, the effect could be harmful due to an unnecessary lowering of the soil temperatu re . It was obvious one had to relate these two factors to find out just how much water it was profitable to apply.

Mr. Du Toit (thc Chairman) referred to the experience of a Hawaiian technologist who found that he could not quite make out when he got to Iran a n d started irrigating, that there was really very li t t le response, or irrigation. The air temperature was high enough, but he just could not get the cane to grow. Then he found that the water he had used came from high peaks covered in snow and the wa te r temperature was so low that it depressed the soil temperature and therefore he got very poor growth .

Dr. Cleasby said that one should apply irrigation in the winter months. Evaporation figures were ve ry low in June, July and August, but one should still irrigate at a rate determined by the evaporation.

Mr. M. J. Stewart said that regarding fertilizer application, he noticed that the cane yields were 40-45 tons per acre. The ratio between N-P-K was 1:1:2 and if he was right in assuming 40-45 tons per acre then a ton of cane took up to 2.0 lbs. of ni t rogen and phosphate. Could one therefore estim a t e tha t the potential multiplied by this fertilizer rat io? There was not much response because they s t a r t e d at a fairly high level. In the case of potash it was 4.0 lbs. of K20 per ton of cane. He asked if t h a t being so, can we just multiply our respective crop by 4.0 lbs. and see how much potash is necessary?

Mr. Thompson said he did not think this could be done because it depended upon the actual soil ferti l i ty as shown by soil analyses or plant analyses. The actual whole cane analyses had shown a much higher K 20 removal by the crop than was indicated in t he literature, where 150 lbs. K 2 0 per acre for a

40 ton cane crop was a figure he remembered. He would certainly be hesitant about fertilizer requirements calculated as Mr. Stewart suggested.

Mr. M. J. Stewart said that results of soil analyses should be reduced to a standard but applications could be modified according to actual requirements after analysis and experiments had indicated these.

Mr. Main asked if Mr. Thompson had established a relationship between the application of overhead irrigation and the absorption of water through the leaves of the plant. During the last drought he was amazed to see very heavy dew on leaves that were drying out. While he had been in the past convinced that a certain amount of absorption did take place, he now could find no evidence of this at all. He thought the relative merits of overhead and furrow methods should be further investigated.

Dr. Cleasby said that the terrain and the availability of water really determined which system should be used. As far as any evidence of water being taken up by the leaves was concerned, he did not know the answer and thought that it would be very difficult to investigate this point.

Dr. McMartin said that radio-active tracer studies have shown that both water and minerals can be taken up by the plant through the leaves and transported to the point where they are needed.

Mr. Main said that it might be that low temperature might have been a factor in moisture absorption. He thought further experiments might indicate whether there was any positive absorption through the leaves. He said that in the Chaka's Kraal experiment a gain of about Hi tons cane per acre was achieved by overhead as against furrow irrigation. There, humidity was determined outside the cane and below the canopy. This showed that the humidity outside was about 77 and under the canopy about 99 per cent and this was important from the point of view of soil temperatures.

Mr. Wilson noted from Dr. Cleasby's paper that a statement was made that irrigation was mostly used on soil that was subject to drought, such as shallow shales. He thought that prospects for irrigation were very considerable in this industry, but that it could be used more effectively and efficiently on the deeper soils for intensive production rather than on the shallow soils for drought insurance.

Mr. Maclver asked which was the best time to irrigate—at night time or during the day.

Dr. Cleasby said that one should, if possible, irrigate day and night considering the capital expenditure involved, but if sufficient money was available it would be better to irrigate during the day as control of irrigation was very difficult at night.

175

Introduction Within the genus Saccharum there exists great

diversity of form. At one extreme are the so-called "noble" varieties, representatives of the species S.officinarum. These varieties are found only in cultivation; they are essentially equatorial in origin; they produce relatively few stalks, which are thick, soft and rich in sugar; their leaves are broad, with loosely-clinging sheaths; they do not flower freely, particularly when grown outside the tropics; they are susceptible to mosaic disease; generally they are weak ratooners. At the other extreme are the representatives of the species S.spontaneun. These are wild sugarcanes, some of which are found far from the tropics; usually they produce a large number of thin, hard stalks containing very little sugar; their leaves are thin, and usually have tightly-clasping sheaths; sub-tropical forms generally flower freely under the conditions to which they are adapted; they are resistant to mosaic disease; generally they are capable of ratooning strongly.

Nearly all the varieties in commercial cultivation today have been derived from hybridization between representatives of the above-mentioned species. From the "noble" ancestors the hybrids derive their ability to produce high yields of sugar under good conditions, and from their wild ancestors, their hardiness. The degrees to which different varieties possess these two characteristics is thought to play a large part in determining their relative value under different conditions.

Factors Influencing the Choice of Variety It is obvious that the best variety to plant in a

particular field is, in nearly all cases, the one that will give the most profitable returns over the whole crop cycle in that field. (That this does not apply in all cases is because a variety—superior in all other respects—may act as a carrier and spread a disease to surrounding fields of other, and less tolerant, varieties.) The returns given by a variety depend upon, firstly, its yield in terms of sucrose per acre per month and, secondly, upon the costs involved in its cultivation and handling. These costs in turn are affected by varietal characteristics such as the ability to produce a good canopy at an early stage of growth, the thickness and straightness of stalks, the tightness with which the trash clings, and—as a factor affecting particularly transport costs—sucrose content. Generally speaking, these

characteristics are of less importance than yield in determining the choice of variety.

The yield of a variety depends largely on its growth rate under the particular environmental conditions. Less directly, the environment affects yield by its influence on flowering and the prevalence of pathogens. The more unfavourable the environment, the fewer will be the number of varieties capable of giving reasonable returns, and under very adverse conditions the choice of variety is inevitably very restricted. On the other hand, though under very favourable: conditions a large number of varieties may be found capable of giving fair yields, cultivation usually becomes restricted to the one or two showing some superiority to the rest.

Basis for Assessing the Reaction of Varieties to Environment

The basis to be used here for assessing the reaction of varieties to environment rests on the assumptions set out below.

1. The optimum conditions for sugarcane growth are the same for all varieties. Under these optimum conditions, the yield of a variety depends solely upon its production potential.

2. As conditions depart from the optimum all varieties decline in yield. Some, however, react less intolerantly than others, and these are said to possess hardiness.

3. It is not necessary to consider separately all the factors which affect the growth of cane. Such factors as rainfall and nutrient status of the soil can be ignored if the sugarcane plant itself is used as the basis for assessing the general fertility of a particular field or area. The average yield over a period of years, in comparison with the average of the Industry, provides an indication whether field-fertility is high, medium or low. As field fertility is partly dependent upon agronomic practice, the status of a particular field may vary with time.

The poorer the field, the greater is the necessity for hardiness in the varieties planted.

4. Though climate and soil type need not themselves be taken into account, they may, in addition to their effect upon field fertility, also exert an influence upon (i) flowering, (ii) the prevalence of diseases such as red rot, smut and gumming, and (iii) sucrose content.

176

VARIETY AND ENVIRONMENT By P. G. C. BRETT

In any sub-tropical area, as the plant succession advances, the vegetation becomes more and more tropical.—(Bews, ] 925)

Rating the Present Commercial Varieties In Table I points are given to a number of varie

ties in accordance with the basis of assessment outlined above. The points system, while making possible a ready comparison of the varieties in regard to the characters considered of importance, introduces certain errors of simplification. Thus, by approximation, N.50/211 has been given a sucrose rat ing of 2, whereas its sucrose content appears in fact to lie somewhere between that of Co.331, given a rat ing of 1, and that of N:Co.376, given a rating of 2.

The points for production potential are allotted to the varieties on the assumption that healthy seed cane is available and that they have not degenera ted through virus infection or any other cause.

Resistance to mosaic has been omitted from consideration here. For this reason, a mosaic-susceptible var ie ty such as N:Co.339 escapes a penalty that, by rights, it should incur.

In order to estimate the probable performance of a variety, it is first necessary to assess the expected demands of the environment for the area in which the variety is to be grown. A very poor field would demand a hardiness rating of 3, whereas in a very good field a hardiness rating of 1 would be sufficient.

It is the failure of a variety to satisfy some of t he demands of the environment that is of importance in determining its adaptation. If a variety is able to meet these demands, it would derive no further benefit even if the ability to exceed them were conferred upon it. In a country where Fiji disease is a serious problem, a variety may succeed largely because of its resistance, but in South Africa, where this disease does not occur, such resistance would confer no added advantage upon a variety. Hence, in estimating the probable performance of a variety in a particular area, points should be deducted for failure to satisfy any of the assessed demands of the area, but no points should be awarded for surpassing these demands. For example, in a field having an assessed demand of 3 for freedom from

the tendency to flower, N:Co.339—with a rating of I—would lose two points, whereas Co.331—with a rating of 3—would lose none. In a field where the assessed demand is only 1, neither variety would be penalized.

Assessment of the Demands made by Different Regions upon Cane Varieties

The cane-growing areas of South Africa m a y conveniently be grouped into the four main regions described below. Though a particular field-fertility may be characteristic of a region, areas of high, medium or low fertility may be found in all regions.

1. Mistbelt. Within this region the diseases of red rot and gumming are particularly troublesome, and varieties must possess adequate resistance to prove successful. On the credit side, flowering does not present a problem, and smut disease is usually not quite so prevalent as at the lower altitudes. Field-fertility is favourably influenced by the good rainfall of the region.

2. Main Coastal Belt. This is by far the largest of the four regions. Within it, the demands made by the environment vary considerably from place to place, but do not normally reach the extremes to be found in other parts. As a rule, field-fertility is medium. In Table IV, a rating of 2 has been set as the demand of this region for freedom from the tendency to flower. However, in some parts—particularly the low-lying areas—a ra t ing of 1 would be more accurate.

3. Littoral region. In this region flowering is usually profuse and smut disease very prevalent. Field-fertility over large portions is low.

4. Deltas. Typically the fertility of deltas is high and generally these areas show a close correspondence with the irrigated, high-fertility port ions of the main coastal belt. The presence of a high water-table in the deltas, however, makes it particularly important that varieties grown there should have a good sucrose content.

For each of the above regions, the assessed importance of different varietal characters is given in Table II . The arbitrary scale of points used corresponds to that of Table I.

177

Estimated Degree of Adaptation of Certain Varieties to Particular Areas

From Tables I and II it is possible to calculate the number of points lost by the different varieties for lack of adaptation to areas of high, medium and low fertility in each of the four main regions. The results of these calculations are given in Table III.

From these results the varieties can be grouped, for each area, according to the number of points they have lost. A further sub-division can be made on the closeness of the correspondence between the hardiness rating of a variety and the hardiness requirement of the area. In the case of the medium fertility areas, varieties with a hardiness rating of 2 are given preference over others; then, as a convention, a rating of I is preferred to a rating of 3.

Thus determined, the relative values of the different varieties for each sub-division of the four main regions are given in Table IV. Solid lines are used to indicate point differences between varieties, dotted lines to indicate differences in hardiness. Where such differences are indicated, the higher-placed variety, or group of varieties, is considered better adapted than the lower.

Discussion and Conclusions

The ranking of the varieties in Table IV appears to show general agreement with that found in experimental trials and field practice. However, the accuracy of the placing of N.50/211 must be in doubt until more is known about the characteristics of this recently-distributed variety. Some apparent discrepancies between the theoretical and actual values may occur in certain areas with some of the older varieties, if these have begun to decline in yield through infection with ratoon-stunting disease or other, unknown, causes. However, the fact that the agreement is generally good indicates that it is possible to obtain a reasonable guide to the performance of most varieties in most areas bv con

sidering only the characters of production potential, hardiness in relation to field fertility, sucrose content, resistance to certain diseases, and the tendency to flower: soil type and climate need not themselves be taken into consideration.

It is interesting to note tha t even under the highly artificial conditions of sugarcane agriculture there appears to exist a similar trend to that which occurs in natural vegetation, and referred to in the extract from Bew's "Plant Forms" at the beginning of this paper.

With good agronomic practices the fertility of a field may be raised. As this is brought about, the field's requirements for hardiness in a variety grow less. Now, hardy varieties in comparison with non-

178

TABLE IV

Theoretical relative values of certain varieties in different regions of the Natal cane belt.

(For explanation see text)

hardy types, usually show greater resemblance in form to the "wild" canes—from which this characteristic appears to have been inherited—and less resemblance to the "noble", tropical varieties. It may therefore happen that, as the demands of the environment for hardiness are lessened, varieties showing more resemblance to the tropical, noble varieties may be grown. Will, then, varieties closely resembling the attractive-looking noble types finally come to be grown in South Africa? Probably not; the best conditions for growth are, inevitably, uniform conditions for any departure, from the optimum implies a change for the worse. In the sub-tropics, where greater climatic changes take place during the year than in the tropics, the demand for hardiness is likely to be correspondingly greater. If, as seems likely, this necessitates a larger proportion of "wild" characteristics in the varieties grown in the sub-tropics, the sight of varieties bred for the tropics, growing in the tropics, will always arouse feelings of envy in a South African planter.

Summary It appears that a fairly reliable guide to the

relative values of different varieties for a particular area can be obtained by:

(1) rating the varieties for the degree to which they possess the characteristics of (i) a high "production potential" (ii) hardiness, (iii) a high sucrose content, (iv) resistance to the diseases of smut, red rot and gumming, and (v) freedom from the tendency to flower,

(2) rating similarly the particular importance of each of the above characteristics in the area being considered.

(3) Comparing the ratings of (1) and (2) and finding the number of points lost by each variety for inability to satisfy the particular demands of the area.

The Chairman, Mr. du Toil, said that the paper was involved and somewhat difficult to discuss on its merits as it had arrived so late. Dr. Brett had given numerical values to desirable and undesirable features of canes. He thought the. system outlined was excellent although over-simplification might be a possible danger. The rating range of 1 to 3 seemed to be rather small, and Dr. Brett might find it necessary later on to have a wider range. If the spontaneum variety showed all desirable aspects except one, it would only lose a few points and yet might prove, because of one defect, unsuitable. Thus if smut was sufficiently bad in N:Co.293 it would

be thrown out. He found it strange that the diseases mentioned were confined to red rot, ratoon stunting disease and gumming. He asked what would happen to a variety which was not tolerant to mosaic. On Dr. Brett 's system on productivity and hardiness he asked if the potential should be determined only under very favourable conditions, would one or two experiments be sufficient, and on the other hand, how did he get his classification of hardiness? Was that done by comparing the variety under favourable and unfavourable conditions?

Dr. Brett said that the production potential of a particular variety should be estimated from its relative performance in a number of trials under very good conditions. A comparison of these results with those obtained under unfavourable conditions would then provide an indication of its hardiness. As far as N:Co.293 and smut disease was concerned, it was in fact susceptibility to smut disease that was largely responsible for this variety's low placing in the table, for areas where smut was prevalent. The. reason why mosaic disease had been omitted from consideration was that the full implications of this disease were not universally agreed upon. Some might consider that a very susceptible variety like N:Co.339 should not be too severely penalized because it was capable of giving good yields even when infected; others would disagree because of the risk of mosaic spreading from N:Co.339 to other varieties.

Mr. W. J. G. Barnes said that, some years ago, he had pointed out that, as far as variety selection in respect of disease was concerned, the final decision lay with the grower. He considered the paper a very useful one in that it suggested a whole lot of factors which the grower could take into account when considering which varieties he was going to plant on his farm. However, according to Dr. Brett, N:Co.293 was one of the worst varieties to select; yet, in practice, on his farm it was one of the best. Its susceptibility to smut floored it badly in Dr. Brett 's tables; yet the risk of smut was one which a grower could accept in the light of this variety's other strong points. Further, the item "sucrose content" was one of Dr. Brett 's factors, whereas the grower was more concerned with production of tons of sucrose per acre per month as a variety selection factor.

Dr. McMartin said he was pleased Dr. Brett had emphasised the importance of environment but was unable to agree with the general statement that the. optimum conditions were the same for all varieties. Some varieties might have optimum conditions in common, but he had formed the impression that some attempts to introduce varieties from other countries failed because, varieties were geographically displaced. He knew that some canes which were poor in hardiness here, for example the

179

P.O.J. varieties, could crop very well under plentiful supply of water and nutrients, but on the other hand there were some varieties which would never do well here because of environmental unsuitability of other factors such as length of day.

Dr. Brett said that varieties such as N:Co.310 which do very well under our best conditions might give even higher yields under tropical conditions, but then be surpassed there by varieties having a high proportion of noble blood. It might be that there was an inverse relationship between production potential and hardiness, and that a greater amount of hardiness was always required under subtropical conditions. If this were so, the very best varieties for tropical conditions would inevitably lack sufficient hardiness to be of value under even the best of our own, sub-tropical, conditions. Referring to Mr. Barnes's remarks, he said that production potential was related to production in terms of tons sucrose per acre per month; however, sucrose content was also taken into account independently of production potential as, by influencing harvesting and transport costs, this factor had an effect upon the economic value of a variety.

Dr. Dodds said that Dr. Brett's efforts in recent years must have been of far reaching benefit to the industry. This paper was perhaps the best because it summarized all the work that he had done in recent years. He found a very important factor in the selection of varieties in his time was the personal preference of the planters. When the Experiment Station was first started, Uba was the only cane under cultivation and it was very difficult to get any growers to take up any other variety at all. It was then considered that a cure should be found for streak disease in Uba cane, rather than introduce other varieties. In East Africa he had found one estate which was wedded to one particular variety though it was not the best variety that could have been grown. It was with difficulty that he got them to experiment with N:Co.310 and N:Co.376, but they did find the change well worth while. With regard to the lack of permanency of these new varieties he wanted to know if further light could be thrown on the subject. Some of the older varieties could last for centuries without appreciable deterioration. The tendency of the modern hybrids was to ran out, apparently due to unknown disease factors. He would like more information on the subject, if obtainable.

The Chairman thought that there was such a big demand for new varieties in this country that the conservatism mentioned by Dr. Dodds did no longer apply in this country.

Dr. Brett said that the running-out of varieties was a question for the pathologist. He was inclined to think that one could not explain everything on

the basis of ratoon stunting disease; there were probably other diseases, as yet unidentified, which were also responsible for varietal deterioration.

Mr. du Toit felt that the variety classification given in the table should not be final, as varieties changed with time. A variety may get more susceptible to disease. He further agreed with Dr. Brett 's reply on the matter of the running out of varieties. He said that those who attended the International Conference in Hawaii, heard an interesting symposium on the subject of running out of varieties, but he doubted if there was anything really final in the symposium. Many things were suggested but nothing was definitely known. As Dr. Brett said, certainly ratoon stunting was playing a part in the running out of some varieties, but he did not think there was anyone bold enough to say that that was the whole story, and we in South Africa would certainly not subscribe to ratoon stunting being the only factor. We know that in the case of Co.281 this was not so, as the hot water treatment so far had been quite unable to get Co.281 back to its old position. One of the things suggested in Hawaii was that probably one of the reasons for a variety running out was the change in the amount of organisms in the soil. Certain micro-organisms will build up as a result of the continued planting of a certain variety and harmful organisms may be the downfall of that variety. Once a new variety was planted these may not affect that variety so much as it would take some time again before harmful organisms would build up.

Mr. Hempson said that he thought one subject which had not been considered in the running out of varieties was bad husbandry. He though in this industry with better treatment the varieties would last longer. In the case of poor farmers where bacteria in the soil caused reduction in yield it might even spread to other farms. He thought tha t seed selection was very important.

The Chairman could not see why husbandry methods other than seed selection should cause one variety to run out and not another.

Mr. Boulle said there were no two points which could be correlated that affected flowering. He asked Dr. Brett how in practice the assessment on flowering could be worked out. The aspects of the farm determined whether the variety would flower or not. He asked Dr. Brett where one should plant a variety which was susceptible to flowering.

Dr. Brett said that past experience was the best guide as to whether or not a particular field was prone to flowering. As a generalization, it could be said that flowering usually decreased with increasing distance from the sea and increasing altitude; however, within a particular area, flowering was usually more profuse on the hillsides than in the valleys.

180

Probably these effects resulted from low night, temperatures discouraging flowering.

Dr. McMartin said that varietal deterioration was not confined to sugarcane varieties alone, but applied to all crops. The suggestion has been made that the sugarcane varieties which lasted longest were those most closely related to wild types. The breeder may be faced with the choice of producing high yielding, highly bred varieties which soon run out, or producing semi-wild types which last longer.

Mr. Tedder felt that usually one introduced a new variety in good soil and then compared it with the old varieties growing in poorer soil, and so he agreed with Mr. Hempson that all varieties should be looked after as well as possible. This process of always planting the older varieties on poorer fields he was sure was the reason for their speedy running out.

Mr. Turner asked Dr. Brett to enlarge upon the importance of the choice of variety in the case of mosaic, whether one should live with the variety which had mosaic or was it necessary to throw it out.

Dr. Brett said it was sometimes difficult to assess the importance of a particular disease in a particular variety. N:Co.339 could become 100 per cent infected with mosaic without apparently falling off in yield; nevertheless, it was to be expected that such a source of infection would constitute a danger to other varieties. It did not appear necessary nowadays to take the risk of growing N:Co.339—there were other varieties which grew as well and were far more resistant to mosaic disease.

Mr. J. Wilson agreed with Dr. Brett that we had now reached the situation where we have a substitute for N:Co.339 and we should now give serious consideration to withdrawing N:Co.339. There was a strong case for suspecting that N:Co.376 was being affected in areas where N:Co.339 was grown.

Mr. Maclver said that he was advised years ago not to plant mealies in cane fields because they were the host of mosaic disease and therefore he agreed that it would be a good idea to get rid of N:Co.339 which would also act as a host for mosaic.

Dr. Dick was surprised that Mr. Thompson had not mentioned another point in connection with mosaic in N:Co.339. It was said that N:Co.339 was tolerant and it did not suffer from mosaic disease, but there was a danger of this tolerance breaking down and that variety might also suffer.

Mr. Main asked for more, information about the susceptibility of N:Co.376 to mosaic. He always understood it was resistant. There was a variety N:Co.349 which gave better results than N:Co.310 but which was not released because of its suscepti

bility to mosaic. When N:Co.376 was first released it was said to be tolerant and not very susceptible to mosaic.

Mr. King said a lot of blame seemed to be attached to N:Co.339 as a source of infection for N:Co.376, Was it not possible that the mosaic in N:Co.339 was due to its being planted in areas where mosaic was more prevalent? For instance in a field on the South Coast N:Co.339 was 100 per cent infected. He had also seen a field in Zululand where it was not affected at all. He felt that rather than being an indication of its being a danger to N:Co.376 it was more an indication of the susceptibility shown in a particular environment. On the South Coast where valleys were heavily wooded there were a number of grasses which were heavily infected with mosaic. The possibility was that the infection spread from these grasses to N:Co.376 Mosaic did not readily spread from one variety to another, but rather through grasses.

Mr. Hyde Palmer said that N:Co.339 was excellent as a yielder and he would like to know, because it flowered, what it should be replaced by?

Dr. Brett said that on the poorer soils N:Co.382, and on the better soils, N:Co.376 could be used to replace N:Co.339.

Dr. McMartin said that the planting of maize amongst sugarcane was not the same problem as whether N:Co.339 should be planted near other varieties. The insect that spread mosaic bred on maize but not on sugarcane. He thought the possibility might exist, as once suggested by Dr. Dick, that N:Co.339 was more attractive to the insect which carried the disease. The argument might thus be made that if N:Co.339 was more attractive to the insects, would it not be better to plant more of this variety so that the insects would not feed on the other varieties? However, if N:Co.382 was as good as N:Co.339 then the areas where the latter was affected could be planted to N:Co.382.

Dr. Dick said that he did not have any direct evidence that N:Co.339 was more attractive to the insects. He had tried to determine this by keeping different varieties in the same cage and releasing aphis maidis. He could find no significant difference in the relative attraction, although the insect lived longer on N:Co.339.

Dr. Cleasby said that the disease must be taken seriously. He confirmed the fact that mosaic was spreading fast, and he thought that the first essential was to remove the bulk of the infection which he considered was N:Co.339, and then to use seed from beds which were free from the disease. Furthermore it would be necessary to rogue out plants from young cane which were infected with mosaic.

181

Mr. Steyn said that he had grown excellent crops of N:Co.339 up to 5 or C ratoons where mosaic was rife. But seed cane apparently free from the disease planted in new areas now was found to be 80 per cent infected.

Mr. Borchards asked if mosaic was more, or less, prevalent in high altitudes. The cane inspector had inspected his fields just recently and found almost no mosaic infection on his farm in the N:Co.339 which was grown at a high altitude.

Mr. B. T. Wilson asked if N:Co.339 could throw off mosaic sufficiently to consider it absolutely free.

Dr. Dodds said that there was apparently only one vector, aphis maidis, known to convey mosaic in this country, as yet. On visiting other sugar-growing countries he had found that the tendency was to rely on cane varieties that were either tolerant or resistant to mosaic. He thought that the Experiment Station should be authorised to instruct planters to withdraw replanting of varieties that were infected with mosaic and were not tolerant to it.

Mr. J. Wilson replied that he considered the only thing the Experiment Station could do was to recommend that when the variety N:Co.339 was ploughed out it should not be replanted.

182

183

SOME ITEMS OF ECONOMIC IMPORTANCE IN SUGARCANE PRODUCTION

By J. L. DU TOIT

The purpose of this paper is to examine some items of cane production which are common to all cane growers and which will have an effect on the whole sugarcane Industry and not to analyse the economic performance of one type of implement and compare it with another or even one method of agriculture or process of manufacture with another.

It is sometimes considered that the research worker should avoid economic consideration because it may well distract him from his own possible; contribution and that by doing so he is entering a field in which he is most unlikely to be an expert. This contention is not altogether devoid of substance, but. at the same time, and this applies particularly to applied research, it would be wrong and indefensible for the research worker to blind himself to all economic considerations and to ignore economic implications which result from an intimate knowledge of his own field of work.

The profit margin in the production of an essential food such as sugar must of necessity always be somewhat limited and consequently any contribution that will lead to a truer assessment of value, that will accomplish a saving or that will avoid loss, will be worthy of consideration.

Sucrose Content Sugarcane is grown exclusively for the sugar that

can be processed from it. In fact it can be stated that it is only the sugar or sucrose which the cane contains which is an asset and the rest of its weight is a liability. Sucrose content is therefore, as an economic factor, all-important. A good stand of cane is admired by the farmer not so much in keen anticipation of a huge tonnage costing a lot of money to harvest and transport but rather on account of the anticipated average sucrose content which when multiplied by the large tonnage of cane is likely to result in a good profit. The days when cane was paid for on a cane tonnage basis are long past and consequently one often hears that experimental results and even field yields should be given in tons sucrose per acre or tons sucrose per acre per unit time basis. Undoubtedly tons sucrose per acre is an improvement on tons cane per acre but it is an over-simplification and not ideal as will be shown.

Let us assume the price per ton of sucrose to the planter is £14.3 which means that the price per one per cent sucrose in cane is 2.86s. and that the cost of cutting, loading and transport per ton of cane for this individual amounts to 12.87s. If he

have been required to make it better than variety A.

This method of assessing the relative values of two varieties is certainly a great improvement on a T.S.A. basis. The cost of cutting, loading and transport will however, vary from planter to planter and for milling companies, but the cost and value of cane will be known and each individual can make his own assessment, but what about experimental results from the Experiment Station dealing with the Industry as a whole? Transport cost, unlike cutting and loading costs, is most variable but any reasonable assumed average cost for obtaining profitable sucrose per acre must be more correct than T.S.A. which presupposes no such cost at all. The assumed cost of 12.87s. per ton may therefore be retained until a truer average becomes available.

It is further known that our laboratory tests almost invariably result in a sucrose per cent cane higher than that obtained in a mill return. Thus a laboratory analysis of 15 per cent sucrose may nor-

On the basis of tons sucrose per acre and also on total income variety B will be favoured but its cost of cutting, loading and transport will amount to £32.175 compared with £25.74 for variety A leaving a nett return per acre from variety A of £48.62 compared with £46.475 for variety 11 Obviously it is more economical to grow variety A and sucrose per acre by itself is not altogether a sound basis to determine which variety is to be grown. For this

planter a sucrose per cent cane equal to

will always be needed simply to pay for cutting and transport. This means that 4.5 should be subtracted from the sucrose per cent cane before multiplying by tons per acre in order to get the profitable tons sucrose per acre and a just method of comparison.

Thus:

now has two varieties yielding respectively 40 tons cane at 13.0 per cent sucrose and 50 tons cane at 11.0 per cent sucrose, then on a sucrose per acre basis we have:

=»

mally result in a factory return of 13.5 per cent sucrose. This discrepancy is largely due to the fact that in the laboratory clean, fresh mature sticks are tested but in the factory trash and immature as well as dead sticks form part of the cane supply which is often not fresh either. Whatever the reason it is on the factory result that payment is made and it is consequently on this that the assessment is to be made. This means that we have a further 1.5 to be added to the 4.5 before subtraction from the laboratory sucrose per cent cane.

If now variety C yields 40 tons cane per acre at a laboratory sucrose of 15 per cent what must the yield of variety D be at 12 per cent sucrose to compare favourably with C?

On a sucrose per acre basis:

But if we take the abovementioned factor into consideration to get profitable sucrose from our experimental result the comparison is as follows:

It will be seen that on T.S.A. variety D needs only to yield 50 T.C.A. to compare favourably with variety C whereas in reality it should yield at least GO T.C.A. to compare economically with C under average conditions.

It is felt that these considerations have to be borne in mind in the release of varieties and in all experimental work on fertilizer, etc. where a sucrose difference is obtained and by doing so greater importance will be justly attached to sucrose content, for the lower sucrose cane will have to outyield the higher sucrose cane by a greater amount than just to equal it in T.S.A.

Recoverable Sucrose The above considerations take no account of a

further premium which high sucrose canes should get, i.e. the fact that it must cost less to recover the sucrose from a high sucrose cane than from a low-sucrose cane—a fact which is taken into account in some cane payment systems but not in ours. Neither does our payment system take into account the recoverabihty of the sucrose in the cane. Both fibre and purity are known to effect recovery. It is impossible with the same input of energy to extract as much sucrose from high fibre cane as from a low-fibre cane and it is similarly impossible to recover the same percentage sucrose from a low purity juice as from a high purity juice. Thus although the same

is paid at present for cane with 14 per cent sucrose, 18 per cent fibre and 85 purity crusher juice as for cane with 14 per cent sucrose, 14 per cent fibre and 90 purity crusher juice, it is known that the latter cane will result in more sugar in the bag than the first and it is therefore worth more. By paying the same for good and bad cane per unit of sucrose we are subsidising the bad at the expense of the good to the detriment of the Industry as a whole, for it is only recoverable sugar that ensures a monetary return to the Industry.

In addition to the adverse effect of high fibre on extraction, which should determine the value of sucrose to the extent that it reduces its availability, fibre also leads to the mal-distribution of the sucrose for payment purposes.

Not only is cane, with 14 per cent sucrose associated with 18 per cent fibre not as valuable to the Industry as a whole as the same sucrose per cent cane associated with only 14 per cent fibre, but even worse, because of a common Java Ratio the first consignment will be paid for more than the second. Using the Queensland formula of pol. per cent cane = pol.

of 1st expressed juice and accepting

100 — (F-f 5) as the real operating Java Ratio, then the factor or Java Ratio to be applied to the first lot of cane is 77 and to the second 81, but under our system using the average Java Ratio 79, the apparent sucrose for the poor cane is 14.36 per cent and for the good 13.65, resulting in the payment of 41s. per ton of bad and 39s. per ton of superior cane! If the effect of fibre on extraction and purity on recovery are taken into consideration then the values of the two lots of cane are almost exactly the reverse of the above! Surely under these circumstances there is little inducement to get the best cane to the mills and the Industry as a whole must surfer economically.

It is felt that purity and fibre content should also be taken into account in experimental work and particularly in determining the release of varieties, but however necessary these may be, such an assessment will only have meaning if the effects of fibre and purity are incorporated in a cane payment system.

The reason why the effects of fibre and purity have not been incorporated into our cane payment system is probably the difficulty of assessing the fibre content of individual consignments and the uncertainty of the relationship between purity of crusher and mixed juice. The present method has the advantage of simplicity and an ideal method is difficult to visualise, but minor imperfections in such a new method should not stand in the way of its adoption if it will give a truer assessment of value and thereby lead to progress in the Industry.

184

Deterioration of Cut Cane The deterioration of cut cane or the sucrose loss

therein has often been the subject of investigation not only in this country but in other sugarcane producing countries and the general conclusion is that sugarcane starts deteriorating with accompany-ing sucrose losses almost immediately after it has been cut and the process accelerates so long as the cane is left unprocessed. And yet. there must be many in our Industry who do not realise the extent of the losses incurred and the economic importance of getting the cane milled as soon as possible. This may be due to the fact that as sucrose is being destroyed there is also generally a drying out of the cane which tends to raise the concentration of the remaining sucrose. Sucrose per cent cane, on which the deterioration may in this case be wrongly judged, may only show a slight fall or even occasionally a rise although the total amount of sucrose in the cane and its availability has fallen considerably.

A recent investigation carried out during Novem-ber and December 1959 at the Experiment Station on several commercial varieties, gave interesting results. The object here was to find out to what extent cane cut up in one foot lengths, as is likely to be done with a Massey Ferguson cane cutter, will deteriorate quicker than cane cut as whole sticks. Reference will here be made to the losses in whole cane only. The cane was divided into approximately 50-lb. bundles which were accurately weighed and the control bundles analysed. The remaining bundles were then kept under prevailing weather conditions in the open and re-weighed and analysed after 2, 4 and 6 days. The average results were as follows:

The table shows that although sucrose per cent does not drop in this case until the fourth day (and there is even an increase after two days), sucrose losses set in almost at once and after two days two per cent of the actual sucrose present, has been lost while a further quantity becomes unavailable: as a result of a marked drop in purity and an increase in fibre and the actual daily loss of recoverable sucrose increases progressively with time. The table indicates that a day gained on 2-4 days old cut cane, will lead to a saving of about two per cent or an annual saving of approximately £500,000 on a £25,000,000 crop.

It is conceivable that the losses during the cooler winter months may be lower than the figures here given but higher values may also sometimes occur.

The fact remains, however, that the necessity of getting cane: crushed as soon as possible after cutting can hardly be over-stressed. It is of vital economic importance to both the planter and the miller.

Fertilizer Application The importance and need for an adequate and

balanced fertilizer application to sugarcane have often been stressed and it is not intended to deal with the subject here in detail but rather to refer to some economic aspects affecting the issue.

Fertilizer responses almost invariably follow a law of diminishing returns over the range of appli-cations commonly made but the cost rises linearly with the amount of fertilizer applied. It therefore follows that there is an economic optimum amount above and below which the profit margin necessarily decreases and it is most important that this level be aimed at in practice, in advice given and be indicated in experimental results which are to be applied in practice.

As a general rule the response to the second 100 lbs. per acre N, P2O)5 or K20 is about one-third of the first 100 lbs. applied. Accepting this relation-ship for nitrogen and assuming the cost of 100 lbs. N to be £4 10s. 0d., the cost of cutting and transport as before 12.87s. per ton, the response to the first 100 lbs. N to be 8 tons cane per acre and the value of cane £1 18s. Od. per ton with no change in sucrose or cane value with nitrogen applications, what then is the optimum amount of nitrogen to be used?

The answer is about 120 lbs. N per acre with a yield increase due to N of 8.79 tons cane per acre and a profit of £5 13s. Od. per acre due to fertilizer application. An amount of fertilizer either greater or smaller will result in less profit but the change is small near the 120 lbs. N level and amounts varying from 100 to 140 will result in profits not very different from the optimum.

An increase in the price of cane and a reduction in fertilizer and cutting and transport cost will increase the amount of fertilizer which can profitably be applied. Thus if the value of cane is £3 per ton, the cost of 100 lbs. N £3, and cutting and transport cost 5s. per ton, then the optimum amount of nitrogen indicated is about 230 lbs. N per acre with a yield increase of 11. tons cane, per acre and a profit as a result of this nitrogen application of £23 8s. Od. per acre. The optimum amount of fertilizer to be applied is therefore apart from actual response very depen-dent on such economic factors as the value of cane and the cost of fertilizer and its application.

Let us now examine the results of an actual fertilizer experiment which has had an overall dressing of 800 lbs. superphosphate per acre and varying amounts of nitrogen and potash. The sucrose per cent cane' as determined at the Experi-

185

186

ment Station and without corrections for purity will be taken and cutting and transport cost will be taken as 10s. per ton. In this experiment both nitrogen and potash affected sucrose per cent cane and there was an interaction between nitrogen and potash. The experimental results and financial calculations are given in Table I I . The cost of 100 lbs. N is taken as £5.2, 100 lbs. K 2 0 as £1.7 and 800 lbs. Super as £4.

TABLE II

The results here show that the best yield in T.S.A. is obtained from 300 lbs. N and 400 lbs. K20 per acre, but the most profitable treatments were 100 lbs. N with either 100 or 200 lbs. K 20 per acre which yielded a profit in excess of £25 per acre due to fertilizer application. The results of this experiment show that while correct fertilizer application may be most profitable, incorrect, excessive or unbalanced applications may also result in severe losses.

All soils will not follow the same pattern of re-sponses nor will the results be identical to these here obtained in a plant cane crop on a Table Mountain sandstone soil, but fortunately soil and leaf analyses can indicate potash and phosphate responses and nitrogen responses may be generalised from field trials.

Instances have been seen, however, where potash and phosphates are applied even where soil and leaf analyses show abundant supplies and no likelihood of a response. The reason is apparently to maintain soil fertility and to put back what has been removed. This is a most laudable motive but very false economy, for in the case of potash excessive appli-cations will be lost in luxury consumption and in leaching without giving any return and may even cause a drop in yield and trouble in the factory.

Ratoons It is a common and generally correct observation

that the profit in cane farming lies in ratoons. This is oi course due to the fact that it is a costlv pro-

cedure to re-establish a cane field and that it invariably leads to a period of complete loss of cane growth. On the other hand old ratoons are inclined to be less productive than plant cane although their yields can often be improved very considerably by higher fertilizer application which of course again increases the cost. When then should a ratoon be ploughed out?

The question is so involved that it is often in practice side-tracked and an incorrect and over-simplified answer is given such as: "Plough out after second ratoon" or "Plough out as soon as the yield drops below 30 tons per acre". Obviously the answer will depend on the yield variation between plant cane and the various ratoons, the cost involved in each crop other than cutting and transport and the value of the cane. It will simplify calculations if we express the cost in terms of cane value, i.e. the value of the cane on the land, or the price of cane at the factory minus cutting, loading and transport costs. Thus if we assume for the purposes of calculation that the cost of a plant cane crop up to cutting time, is equal to the value of 20 tons of cane and the cost of a ratoon equals 7 tons cane, then these values must be subtracted from the crop yield to get the profitable cane yield.

Let us take a field which has given the following yields:

TABLE III 1st 2nd 3rd 4th 5th 6th

Plant Ratoon Ratoon Ratoon Ratoon Ratoon Ratoon

60.0 20.0

57.0 7.0

50.0

54.2 7.0

51.4 7.0

48.9 7.0

46.4 7.0

44 .1 7 .0

44.4 41.9

Yield T.C.A. Cost in T.C.A.

Profitable T.C.A.

If each crop was grown for two years and in addition the land before planting had one year's fallow, then the 40 profitable T.C.A. was the result of three years' endeavour or the annual production was 4 0 / 3 = 1 3 . 3 P.T.C.A., the first ratoon 5 0 / 2 = 25 P.T.C.A. and the second ratoon 4 7 . 2 / 2 = 2 3 . 6 P.T.C.A. The fact that there appears to be a fall in the second ratoon is, however, no reason for ploughing out the field, because the accumulated average annual production calculated from plant cane is still on the increase. Thus plant cane remains at 13.3 P.T.C.A. but at the end of the first ratoon we have had 40.0 + 50.0=90.0 P.T.C.A. in five years or an accumulated average of 18.0 P.T.C.A. and at the end of the second ratoon we have produced in all 137.2 P.T.C.A. in seven years averaging 19.6 P.T.C.A. per annum, which is an improvement on the first ratoon.

We may therefore arrange the results as follows: TABLE IV

1st 2nd 3rd 4th 5th (ith Plant Ratoon Ratoon Ratoon Ratoon Ratoon Ratoon

187

The correct stage to plough out this field is after the fourth ratoon because after that the accumulated annual return will diminish but this wall not be known until the fifth ratoon is cut and therefore in practice the field should be ploughed out after this ratoon and the loss in yield will be negligible. This stage is marked by a fall in the accumulated P.T.C.A. and of course also by the fact that the annual P.T.C.A. 19.7, becomes less than the current or preceding accumulated P.T.C.A. 20.2 or 20.3.

Should the original results have been the result of one year's growth with a year's fallow the calcu-lation would have been:

Here the critical stage where a plough out is indicated is not reached until the sixth ratoon.

Again to revert to a two year crop but with no fallow:

TABLE VI 1st 2nd 3rd 4th nth 6th

Plant Ratoon Ratoon Ratoon Ratoon Ratoon Ratoon

40.0 20.0

2 20.0

50.0 25.0

4 22.5

47.2 23.6

« 22.9

44.4 22 2

8 22.7

41.9 20.95 10 22.35

39.4 19.7 12 21.9

37.2 18.55 14 21.4

P.T.C.A. Annual I'.T.C.A. Total years. Ace. An. P.T.C.A.

Under these conditions the crop must be ploughed out after the third ratoon or even if the drop could have been anticipated after the second ratoon.

Let us now take a more realistic set of data where crops of different ages are cut and where it takes three months for ploughing out the old crop, pre-paring the land and planting the new crop. Two sets of data will also be given for the fourth ratoon: one for the normal procedure and another with an increased fertilizer application equal to the value of two tons of cane.

TABLE VII

The most profitable cycle for this farm would be to plough out after second ratoon or in practice after the third ratoon when the fall was noticeable. The fact that the fourth ratoon can be increased by six tons by applying extra fertilizer to the value of two tons cane must make it most tempting for the planter to continue ratooning, but while it is definitely more profitable to grow a fourth ratoon with the extra fertilizer than without it, the return is still too low to warrant a continuation of ratoon-

ing. For that to happen the yield obtained with the extra fertilizer should have been in excess of 42 tons cane per acre.

The calculation is of course further complicated by the fact that one season's growth is not always com-parable with that of another and the same applies to an even greater extent for the different months. Allowances can however be made. Thus if the average farm yield is 10 per cent down on that of the previous year, then this correction should be made.

The method here described should offer a means of determining the ratoon at which it becomes economic, to plough out by using the actual conditions and costs on the particular plantation or field. Fortun-ately in most cases there seems to be some latitude. Thus if the third ratoon is indicated as the most economic, then though ploughing out after the second or the fourth ratoon will certainly not be as profitable as after the third ratoon, the losses involved will not in general be very great.

Conclusion Some factors which influence the economy of

sugar production have been discussed. Some of them have a direct bearing on the planter and others affect the availability of the sugar in the factory. All of them, however, must influence the economy of the Industry as a whole.

Profitable tons sucrose per acre defined as (sucrose per cent cane — x) (tons cane per acre)) with x a cost factor associated with cane and expressed in terms of sucrose per cent cane is a better yardstick than tons sucrose per acre for judging varieties, etc.

The availability of sucrose should be taken into account in judging experimental results and in determining the value of cane.

The importance of cane deterioration after cutting can hardly be over-stressed.

Fertilizer application and the growing of ratoons are important economic factors in cane production but there is an economic limit to both.

S.A.S.A. Experimental Station, MOUNT EDGECOMBE. 13th February, I960.

Mr. Wilson (in the Chair) said that practical guidance from the Experiment Station to growers was usually expected to start in the field and finish in the field. Profit per acre, not gross yields of cane or sucrose per acre, was however the main criterion in deciding the real value of any recommendation and in consequence it was impossible to divorce agricultural research either in field or laboratory from economics.

Mr. Bentlcy criticised the suggestion that the time to plough out was at the peak of a grower's return. He though he should carry on until his return from a ratoon crop was the same as that from plant cane. He asked for further information on this aspect.

Mr. du Toit replied that it was accumulated annual profitable return which was important.

Mr. Pollock said if it cost him as much as 12s. per ton for transporting and cutting cant; he would not be able to continue cane growing.

Mr. du Toit replied that some planters would criticise the figure he had given as being too low. He considered his figure to be something like the average estimated by the Cane Growers' Association. He realised that for The Tongaat Sugar Co. the figure would be much lower. His paper merely gave the outline of the method which he considered necessary to give a true assessment of value and the figures he had given were subject to alteration to suit individual cases.

Mr. W. J. G. Barnes agreed that 12s. per ton was not very far from the average.

Dr. McMartin said he welcomed a paper which emphasised the importance of economics in fertilizer application. It was known tha t increase in yield was not directly proportional to the increase in fertilizer used but that each increase of the latter produced a smaller increase of the former. There was thus a point at which yield was at a maximum, but this was not necessarily the point at which maximum profit was obtained, and the object should be to determine the latter. Another point which could be taken into account was the hazards involved in obtaining a profitable return; for example, under irrigation the hazard of drought was removed, but under dry land conditions it might be argued that a smaller expenditure on fertilizer should be decided upon to counterbalance the effect of a bad year: he wondered if the author had any views on this aspect of the subject?

Mr. du Toit said that one should work on the average of good and bad years.

Mr. W J. G. Barnes asked how could the grower tell when he had cut his second ratoon, that his third ratoon would not be better.

Mr. du Toit said it might be necessary to go to one extra ratoon to find out when profit began to decrease. This would not be a great mistake.

Dr. Dodds thought it was advisable to arrange an Economic Advisory Department at the Experi-ment Station as well as a Fertilizer Advisory Service. Another point was that he thought the Java ratio was no credit to the Industry. Fibre was a very

important factor. He said that planters who sent in cane with a low fibre were penalised, while a grower supplying a high fibre in cane was given an unduly high payment.

Mr. du Toit said even the use of the common Java ratio had certain advantages because it was so simple. He mentioned work done by Mr. Hugo on the same subject of assessing value as far as profitable sucrose was concerned, and when a ratoon should be ploughed out. Mr. Hugo came to the same con-clusion as himself although they had worked inde-pendently. Mr. Hugo had evolved a formula but he himself had tried to make his approach more simple, and he had completed his paper before; his attention was drawn to the work of Mr. Hugo.

Mr. de Robillard stated that from the figures given in Tables IV and VI wo saw that with a long fallow the field should be ploughed after the 4th ratoon and after the 3rd ratoon with a short fallow. It would then appear that long fallowing was more economical. But taking the time factor into account in 11 years the long fallow had given a total of 105 tons of profitable canes, and in 8 years the short fallow has given a total of 90.8 tons of profitable canes; reducing to profitable tons canes per two year crop, the long fallow had given an average of 19.1 profitable ton canes and the short fallow in the same two year crop had given 25.2 profitable tons canes. The short fallow then was more economical as it gave 6.1 tons more profitable tons canes per crop.

Mr. du Toit said that if the same yield could be assumed from a short as from a long fallow, then obviously the short fallow was more advantageous because with a long fallow you had to continue cropping for a longer period to make up for the time wasted. His opinion was that land should not be out of production unnecessarily. If one was going to plant in March, green manuring would be advan-tageous but not when spring planting was done.

Mr. de Robillard asked if anyone had any views comparing the yields of short and long fallows, because apparently it needed an extra yield of 6 tons per acre with the long fallow to make it more economical than the short fallow.

Mr. du Toit said this was really outside the scope of the present paper.

Dr. Dick said that, if sucrose content were to be given its true value as a criterion in selection, more care would have to be exercised in sampling variety trials for sucrose.

Mr. du Toit agreed that sampling should be as exact as possible but the figures shown were an average of a number of tests and thus more reliable than individual variety trials.

189

THE DEVELOPMENT OF THE MECHANISATION COMMITTEE OF THE SOUTH AFRICAN SUGAR ASSOCIATION

By GEORGE S. BARTLETT

The attention of the Sugar Industry was first drawn to mechanisation in 1947 when Mr. O. W. M. Pearce was commissioned to make a. survey of developments in mechanisation. As a result of this, and also the enthusiastic efforts of a small group of men, a Mechanisation Sub-committee was formed in 1948. Among these early pioneers of mechani-sation were Messrs. W. F. C. Jex, A. A. Lloyd, O. W. M. Pearce, U. W. M. Campbell and J. Garner, and as their activities grew, other names, such as A. C. Barnes and B. H. Abrahamson and then later van der Walt, Woods and many others were heard when talk revolved around mechanisation establish-ments, cane harvesters and self-loading trailers.

Much interest was aroused throughout the Sugar Industry by the activities of these gentlemen and as they grew, it was found that a more permanent establishment was required. Thus, in August of 1951 Mr. Abrahamson occupied the Mechanisation Committee's Central Headquarters at Umhlali where work commenced in earnest on the Abrahamson Cane Harvester, the Abrex Planter and a variety of other mechanical devices. In addition, it organised a number of demonstrations and carried out investi-gations into the usefulness of particular machines in sugarcane agriculture. A series of five bulletins were issued for the general information of all growers and a close liaison was kept with manu-facturers.

The Committee was re-constituted in 1952 as a Mechanisation Working Committee under the Experi-ment Station Committee which was responsible for laying down the broad policy under which the Mechanisation Committee operated. The Com-mittee's efforts eventually came to a standstill for a variety of reasons and the establishment at Umhlali was terminated on August 31st, 1953. The Committee, however, continued its liaison work with the growers and manufacturers. In 1954 Mr. A. C. Barnes, at the request of the Committee submitted a report entitled "Research on Field Mechanisation for the Sugarcane Industry". This report called for the establishment of a Research Organisation which was to be staffed by a Director, Assistant Agricultural Engineer, a workshop mech-anic, clerk and necessary skilled and unskilled labour.

Following this, Mr. Walter Boa was seconded from the National Institute of Agricultural Engineering in the United Kingdom for a period of approximately one year and arrived in South Africa in December, 1954. During the following year, Mr. Boa examined

the conditions of the Sugar Industry and prepared an extensive report entitled "A Field Mechanisation Establishment for the South African Sugar Industry" which was submitted in August, 1.955. This report was similar in content to that of Mr. A. C. Barnes but went much further into the details of the purpose, organisation and control of the establishment.

In the meantime, a limited programme of field testing was carried out and the beginning of a workshop was erected at the South African Sugar Association's Experimental Farm at Chaka's Kraal. It was intended that this should be the start of an "Agricultural Machinery Testing Service" and the headquarters for experimental work with farm machinery. Numerous manufacturers and agents were informed of these developments and many replied expressing their interest in this scheme.

In September 1955, the recommendations con-tained in Mr. Boa's report were put to Council but no decision was taken, and from that date no further progress was made until August 1957 when the matter was again raised in Council.

At this time, however, an energetic group of Zululand growers had formed their own "Zululand Cane Mechanisation Committee" under the leader-ship of Mr. Bruce Morris. The efforts of the original Committee had stimulated new ideas from these growers and they were making a solid contribution to remedy the labour problems in their area. Their functions were as follows:

(a) Collecting and disseminating to growers infor-mation relevant to mechanisation.

(b) Disseminating information to growers regarding proper and most efficient use of machinery.

(c) Training operators in the proper use of machinery.

(d) Testing machines supplied by manufacturers for adaptability in the Industry.

(e) Suggesting alterations to the manufacturers of machines.

(/) Assisting in and promoting the development of prototype machines.

(g) Recommending the adoption of suitable machinery by growers.

In view of this, Council agreed to refer the matter to the Growers' Association for consideration and recommendation. In due course a sub-committee was formed of representatives of both the Growers' Association and the Millers' Association to draft

190

the constitution of the present Mechanisation Com-mittee. This provided for three regional committees consisting of representatives from both the Millers' and Growers' groups and an Industrial Field Mechanisation and Labour Saving Committee which was the policy-making and controlling body of the organisation. This organisation is laid down in the first bulletin issued by the Committee entitled "Organisation and Administration". In due course, I was appointed to the post of Mechanisation Field Officer and commenced my duties in May, 1959. Since then a secretary has also been appointed.

During the past ten months the Committees have endeavoured to perform their duties as laid down by the. Constitution. The Industrial Committee has been concerned primarily in establishing an overall policy of the Sugar Industry toward sound field mechanisation and co-ordinating this policy through-out the Industry. It is also responsible for adminis-tering the funds granted to the Committee by the S.A.S.A. which includes the granting of financial assistance to worthwhile projects.

This policy is put into practice through the three Regional Committees whose members endeavour to maintain a close liaison between the Industrial Committee and the individual growers. It is at this level that many of the problems or ideas from growers are first examined, and if found to merit further attention are then passed on to the Industrial Committee.

A close liaison has been kept with manufacturers and their representatives so that all new implements appearing on the sugar belt are brought to the attention of the Committee while, in return, the Committee can pass on the Mechanisation needs of the Industry to the manufacturers. As a result of this, much work has been done by the machinery firms in developing new implements or modifying existing ones to suit the conditions existing on the cane belt. Demonstrations of new production or prototype machines have been organised for the various committee members where constructive criticism and discussion has occurred. A close contact has been kept with some of the smaller engineering firms where advice and in some cases technical assistance, has been granted by the Com-mittee members and myself. This liaison has also been extended overseas and the Committee has established contacts which keep it informed of developments in other sugar-producing countries. Engineers and executives from overseas implement firms have called on the Committee for advice regarding the mechanisation needs and field con-ditions of our Industry so that our needs will be better catered for. In return information and assistance has been provided to overseas enquiries regarding specialised machinery developed in South

Africa, such as the self-loading trailers and the Inca cane cutter.

Another aspect of the liaison with t h e manu-facturers is the field of testing of prototype machinery. So far this work has been rather limited due to the broad field of endeavour which has to be covered. It is hoped, however, that this service will be exten-ded as the Committee organises itself to meet this need.

Another activitity of the Committee is t he assess-ment of a new idea or invention brought forward by individuals seeking financial assistance for its development. Each case has been thoroughly examined, and it is the policy of the Industrial Committee that the idea must not only be basically sound engineering-wise, but that there mus t also be a definite need in the Industry for the finished product.

It is the policy of the Industrial Committee to keep in close touch with field experiments and research in agronomic practices. Any new develop-ments or methods of cane agriculture which will improve the ultimate cane crop are passed on to the machinery firms so that the machines will keep abreast with these new developments. To this end the Director of the Experiment Station has been co-opted to the Industrial Committee to keep it informed of any pertinent developments at the Station. Agronomists employed by the various Miller - cum - planter companies have also been approached from time to time to give their expert opinion on the way a particular operation should be carried out.

An effort was also made to get growers a n d tech-nicians together to discuss the problems of infield cane handling. This took the form of a symposium which was held in the Library of the Experiment Station. All members of the four committees were invited along with company agronomists a n d mem-bers of the Experiment Station staff. Although no definite conclusions were reached, I feel it can be safely said that the symposium did stimulate thinking among all those who attended, especially regarding the problem of soil compaction. As a result of this, I have been informed that the Experi-ment Station is to speed up its investigation in to the physical characteristics of the various soil types throughout the Sugar Industry. Not only that, but I have also noticed an increased interest in soil compaction on the part of the manufacturers of self-loading trailers and other types of infield machinery, since a report on the symposium was published in the Sugar Journal.

The third field which the Committee endeavours to cover is its contact with the grower. Th i s is a tremendous task due to the number and geographical

191

location of the growers throughout the Industry. The Regional Committee members report their activities and any new developments to their particular planter group.

In addition, a number of field demonstrations were organised in the earlier part of 1959 but it was found that these became too numerous to handle. Con-sequently, it was decided that a single annual demonstration and show be held at which will appear all the cane, machinery and implements available in the cane belt. This year's demonstration is planned for September and will be held in the Chaka's Kraal area for a period of two or three days. So far over 30 companies have, expressed their desire to take part. In addition to this, the Committee has used the facilities of the Sugar Journal and has presented slide and film shows to disseminate among the growers as much news as possible.

This has been a brief outline, of the history and present activities of the Mechanisation Committees. I have, made no attempt to discuss any of the many individual problems which have been tackled.

During the past few months there have been many arguments and discussions not only between members of the various Committees but also between growers, engineers, salesmen and a variety of other interested parties. Some of these have led to dead ends, while others have yielded results which, whether large or small, have contributed to the store of knowledge and experience which is being built up through the existence of the Mechanisation Committees.

Now if you will bear with me I would like to present to this Congress some of my own ideas and opinions for your consideration.

During the past ten months I have had considerable opportunity to discuss mechanisation problems with many members of the Sugar Industry. As a result of this, I feel that the activities of the Mechanisation Committee can be divided into four main groups, viz. Research, Machinery development, Field testing and Education. Of these, I would like to discuss Research and Education in detail while just mention-ing the others in passing.

Research Considerable discussion has occurred in the past

regarding the research activities of the Mechanisation Committee. Much of this has revolved around research aimed at developing new machinery for use within the cane belt. It has been proposed that a workshop be established and that a variety of individuals be engaged to develop new machinery for the Industry.

While I must acknowledge that these proposals were made at a time when the Industry was ex-periencing considerable difficulty in obtaining suitable

native labour, T would like to ask whether it would be advisable to follow this policy today. The efforts of the previous Mechanisation Committee in attempt-ing to develop machines for the Industry may not: have met with complete success, but it can be said that the existence of the Mechanisation Committee at the time stimulated other individuals and organi-sations to improve on each others ideas which ultimately resulted for example, in the great variety of production models of self-loading trailers which are available today. I believe that the word "stimu-la te" is the key to this whole problem. There are hundreds of people in Natal who are extremely stimulated by any idea which, when developed, will return to them a certain margin of profit, and might I add at this stage, that it is only the ideas which will make a person a profit which will ever reach the production stage and thereby be of benefit to the Sugar Industry as a whole.

The problem which presents itself, therefore, is how the Mechanisation Committee can "stimulate." the Machinery Industry in Natal to provide the machinery and implements we require. It can try to do this by endeavouring to pioneer some research itself in the hope that it might turn up something which will be taken over by private enterprise. There is a great risk in this policy, however, of entering into a field which will undoubtedly be a heavy drain on the resources of the Committee with the probable chance of it not meeting with any success.

The question must also be asked as to how far the Sugar Industry is prepared to commit itself. Many at tempts have been made to develop a cane harvester, for instance, and the costs of a local effort have been estimated to run as high as £10,000 so far, and another £10,000 to take it to its next stage of development. It has been reported that up till the beginning of the 1959 season the Massey-Ferguson Company had spent over £60,000 in developing its cane-harvester with the chance of spending a total of £150,000 before it reaches the production stage. And this, I might add, is a world-wide organisation with a vast resource of engineering personnel and technical know-how, which is in the business to make money. Of course, I realise that I am now talking about a machine of major propor-tions and that these vast sums of money do not necessarily apply to the development of the less expensive forms of mechanisation. It is my opinion, however, that our local Engineering Industry is prepared to undertake the development of these smaller items with very little financial assistance from the Sugar Industry, if any at all. Is it advisable, therefore, to establish a research station to develop equipment which could quite easily be developed by private enterprise? I do not think so, I feel that the Sugar Industry should exploit the vast resources of

192

knowledge, equipment and finance of our local industry since it is they who are geared to this type of work, and it is they who will reap the ultimate profit once a machine or implement reaches the market. If any money is to be spent at all, I feel it should be spent to stimulate our local industry into doing this developing. This will be discussed further on.

At this stage, I would like to mention another aspect of Research for your consideration, viz. agronomic research in relation to mechanisation.

If we are to ask our local manufacturers to provide us with the machinery we need, it is going to be necessary for us to be able to provide them with details of the type of operation we wish to perform. At the present time there appears to be considerable diversity of opinion regarding certain basic agricul-tural practices and I would like to suggest that steps be taken to obtain reliable information based on experimental data. I am now referring to such operations, for example, as seed-bed preparation, planting and harvesting. I would personally like the Mechanisation Committee's present state of co-operation with the Experiment Station extended to include a series of experiments designed primarily to investigate the mechanical problems involved in performing the operation. I would like to see official recommendation based on experimental data pro-vided to answer some of the many questions which at this stage cause controversy, among these are: how little seed-bed preparation can I get away with; how deep must I plant or how shallow can I plant; how close can I plant; where must I place the ferti-lizer; how much filter press should I apply and where? I believe that there is still much to be accomplished in this field if we are going to progress towards sound and economic mechanisation.

Machinery Development As I have stated earlier, it is my opinion that the

major portion of machinery development should be financed and carried out by private concerns. This can be greatly stimulated by the Mechanisation Committee being able to keep the manufacturers fully aware of the needs of the Industry, and well supplied with all pertinent experimental data which will result in the production of a more efficient and practical machine. A close contact should be kept with developments in overseas sugar-producing countries so that any new ideas developed there can be exploited by local manufacturers wherever possible.

Competition between manufacturers will always lead to a greater effort to meet the needs of the Sugar Industry. This can be achieved by growers becoming more aware of the merits of various makes through large scale demonstrations, field test reports

and extension work. Detailed investigations into the costs of performing various operations with various implements should be carried out and these made available to the growers. This will encourage the manufacturers to provide more economical and efficient machines especially at the present time when there is the threat of restricting cane produc-tion and growers are concentrating on raising the efficiency of their operation.

Although I have stated that the local companies should develop new machinery, I must agree that many worthwhile ideas originate from the growers themselves. These should be encouraged and where advisable financial assistance should be grunted to either the grower or a local firm to proceed with the development as is clone by the Industrial. Committor at the present time.

Field Testing It has been found in the past that many imple-

ments and machines have been marketed in the sugar belt without having undergone an adequate programme of field testing. This has resulted in some machines having to be scrapped at considerable financial loss to the grower, and a loss in reputation for the manufacturer.

It is because of this, that provision has been made in the constitution of the Committee to organise a field testing unit. The aim of this unit will be to carry out an impartial field test programme on any machine submitted by manufacturers to the Mechani-sation Committee. It is felt that such an arrangement will promote the development of better machinery for the Industry and will result in more satisfied growers and machinery suppliers.

In order for such a programme to be a success, it is essential that the unit be correctly constituted and staffed with men of ability and integrity. It is because of this, and also because of the pressure of other work, that the development of this programme has been rather limited during the past year. The Committee recognises this need, however, and it is intended that more progress will be made during the coming months.

Education Finally I would like to discuss the problem of

education; a problem which I feel should receive serious consideration by all members of the Sugar Industry.

It is my personal opinion that there is a great lack of training and educational facilities throughout all levels of the mechanical side of the Sugar Indust ry . This includes training facilities not only for nat ive labour but also for the youth of the Sugar Indust ry at the high school level as well as the College and. University level. This especially surprises me when

193

I stop to consider the magnitude of our Industry and the vast amounts of capital which must certainly be invested in machinery.

In most forms of business, the staff and personnel have usually received some form of specialised train-ing which allows them to perform their duties more efficiently. It is generally accepted that the firm which takes advantage of the new knowledge and skills resulting from research by various organi-sations, is more likely to progress further than the firm which retains the philosophy of "if it was good enough for my father then it's good enough for me". I say this with all due respect to the efforts of those who have pioneered all forms of development in the past. However, we are living in an age where the search for knowledge is advancing at an unprece-dented rate. It is those persons who recognise this fact and gear themselves to this development who are going to come out on top in the long run.

Recognition of these facts has been accepted in many progressive countries such, as the United Kingdom, Canada, the United States and Australia. It has been said in Canada that farming was a "way of life" and not an industry. In the days prior to the First World War it was the farmer's smart sons who obtained the higher education and went to the cities to work while his slower thinking brothers remained at home to walk behind the plough. With the surge of industrial development which resulted in competition for labour and ultimately mechanisation, however, it soon became apparent that the farmer and his workers would have to become highly skilled persons or otherwise be forced out of business. Consequently the Government and educational institutions of these countries have geared themselves to raise the standard of education of their agricultural workers. University degrees in agriculture have been extended to all branches of the profession and short and diploma courses have been increased to correctly train field workers for their jobs. This has been supplemented by large-scale extension services being organised to keep farmers informed of new developments.

One might ask where all this is leading to. Well, one hears many stories of men like Warren North of Indiana, who single-handed operates an extensive feeding station handling 400 Herefords, and 500 Hampshire hogs from a single instrument panel in his barn. I personally know of a farmer outside Toronto who by himself handles over 20,000 broilers, and there are many other such cases. The day of automation in agriculture is fast approaching in these more progressive countries due to the increasing shortage of labour and the application of new skills developed through education and research. There is even talk of operating a fleet of tractors in the field by remote control from a central control tower. This might be looking well into the future and the

question might be asked, how does the South African Sugar Industry fit into this development?

I am not for one moment suggesting that our conditions are such that we should consider such drastic steps. But what I am suggesting is that farming is no longer a "way of life" but is an industry which should be geared for development and progress just like any other industry. This will be rejected by many traditionalists who would like to maintain the pleasures of the old way of life. However, I think I would be quite safe in saying that the changes which will be seen in South African agricul-ture during the next thirty years will be beyond our imagination. The point that troubles me is, are we preparing ourselves for these changes? Might I ask whether we are training enough South Africans to meet the demands for technicians and research workers in the future. At this stage I am probably sounding dramatic, but as an engineer engaged in promoting mechanisation in our Sugar Industry, I cannot help but feel that we are lacking so much basic know-how. In many cases we are taking raw natives, giving them very little training, and then expecting them to operate and maintain a highly engineered machine worth in many cases over £1,000. In the past, the abundance of native labour has tended to supress the need or the desire on the part of the owner to increase his knowledge of mechanisation. So often I have heard on my jour-neys through the Sugar Belt, tha t equipment must be made "Pondo-proof". This is a statement which I feel is contrary to any mechanised principles, because it basically means that either the operator or the owner of the equipment is not sufficiently educated to make the proper or efficient use of the highly capitalised equipment.

You are probably wondering what all this is leading up to. Well I would like to propose a programme of education in Agricultural Engineering and Farm Mechanisation to this Congress. I am fully aware tha t the success of such a programme depends entirely upon the att i tude of our individual growers and for this reason I hope tha t this address will stimulate more concern and thought on this matter.

The programme which I am proposing covers five basic fields, viz. native operator training; grower extension service; short courses; diploma courses and university degree courses.

Starting with the worker, I believe that there is a great need in the Industry for some sort of lead towards providing facilities for operator training. Individual schemes have been tried in the past but these have failed through lack of interest on the part of the growers. I am confident that the Mechani-sation Committee has the co-operation of the larger tractor distributors in any training programme we might envisage. Such a programme must have the

194

support of the Industry as a whole if it is to be a success. I appreciate that there are many factors affecting such a development and I feel that it would be necessary for this matter to be discussed in full so that a suitable industrial-wide policy can be adopted. This is an urgent need, however, and it is for this reason that I would like this Congress to promote more thought and discussion on the matter.

I emphasise this point because at the moment mechanical equipment is basically in the hands of the non-European who, although adaptable and reasonably proficient, is far from being a trained or skilled operator capable of carrying out small adjustments as and when necessary. It stands to reason, that with properly trained operators, the equipment in the field will last longer, be better maintained and downtime will be reduced con-siderably which should naturally result in a lower cost of operation.

The second field of education I envisage is that of a grower extension service. At the present time I am the only trained technical person actively engaged in advising sugar growers on mechanisation, but owing to the pressure of all the other duties which I have to perform I am not achieving the result I should like to achieve. We must accept that this service is a necessity in order that growers be kept constantly aware of new developments and ways and means of making their operations more efficient. In the past growers have had to rely on their own resources to overcome their problems, however, I feel it is too much to expect one man to be an expert on mechanisation, irrigation, agronomy, entomology, plant pathology and finance. Once growers realise what such a service holds open to them I am confident they will demand that it be provided.

Not only will a mechanisation representative act in an advisory capacity but he will also organise discussion groups and field days in various areas where growers can get together to compare notes. This will have a great stimulating effect on the search for new knowledge. "Tractor Clubs" should also be organised for the younger members of the com-munity wherever possible so that they will grow up with a sound practical knowledge of the operation and maintenance of their power units. The scope for this sort of service is extremely broad once the enthusiasm of the grower has been aroused. Of course such a service requires adequate staff and it is up to the Industry to decide whether the need warrants it.

Short courses held at suitable locations should be available to growers and their sons. These should be organised through the Natal University and the Cedara Agricultural College and should be

designed to provide those attending with the basic principles of mechanics and machinery as well as the more practical aspects of mechanisation. These courses could be held while the full-time students are taking their annual holidays and should run for a period of from two to six weeks. I might add that I have been informed that the S.A. Sugar Tech-nologists' Association has already investigated this possibility and I was gratified to hear that plans are being drawn up for a course to be held next year. I feel that it is essential for ultimate financial gain that the owner of mechanised equipment be com-pletely conversant with his tools.

Diploma courses and University Courses should be designed to provide training to equip students for two broad fields of the machinery business, viz. agricultural mechanics and agricultural engineering. These can be differentiated by saying that the agricultural mechanic is a more practical person who would probably be employed as an extension officer or a farm machinery salesman or service expert. The agricultural engineer on the other hand should be a professional man actively engaged in research and development work or executive matters. It is these people who will be the leaders of the future and since it takes years to build up an adequate supply of trained personnel the Sugar Industry should encourage its young men to enter this pro-fession. This might even take the form of scholar-ships or bursaries which would be granted to deserving persons.

Summing up I would like to place four points of interest before you. Firstly, I maintain that any active research in which the Mechanisation Com-mittee should engage itself, should be limited to agronomic and economic research done in con-junction with the Experiment Station.

Secondly, the development of new or improved machines should rest primarily with private enter-prise which should look to the Mechanisation Com-mittee for technical data and reliable information on the needs of the Industry. In extreme cases financial assistance might be granted.

Thirdly, the Mechanisation Committee should eventually establish a highly respected field testing unit to which manufacturers will send their machines for testing.

Finally, the Mechanisation Committee should actively encourage a programme of education in Agricultural Engineering and Farm Mechanisation aimed at ensuring that the Sugar Industry and the farm implement manufacturing industry, will be adequately supplied with trained engineers and technicians to meet the demands which will be placed on us in the years to come.

195

Mr. Wilson (Chairman) pointed out that since the Exper iment Station had only one professional officer on its establishment for agronomic research and only four extension officers to cover the whole indust ry it was not surprising that there were serious gaps in our knowledge in the agronomic field nor that Mr. Bartlett found himself called upon to fill so many roles during the course of his duties. There seemed every possibility that the staff position would be considerably improved in the near future. Meanwhile steps had already been taken to initiate experiments designed to fill some of the more obvious gaps .

While agreeing entirely with Mr. Bartlett 's analy-sis of requirements, he was disappointed that Mr. Bar t le t t had stopped short of what appeared to him the obvious conclusion, particularly with regard to research and extension, that mechanization should be pa r t of the Experiment Station.

Apar t from the loading and transport of cane after harvest which involved problems purely of a mechanical or engineering nature, all other operations involving mechanization, from land preparation to harvesting, were inseparable from agronomy and co-ordination of the work would be simplified if all came under unified direction.

He questioned whether any special implements other than those already available to the industry need in fact be required for any operation except possibly planting and cutting and even with these operations, a number of machines available else-where still remained to be tested here before the need to develop our own machines was proved essential. The problem seemed more one of knowing which machine or implement was most suitable for a part icular operation under a given set of conditions, t h a n of developing new machines specifically for this industry.

Mr. Frost, speaking as a machinery inspector, said it would be realised how much trouble was given to engineers in the factories by the inspections carried out and the requirements demanded. He noted with disappointment tha t the functions of the Committee did not include safe operation of the various machines used in the cane fields. Under the Machinery Act, as it was at present, a vehicle was excluded from the definition of machinery, so t ha t machinery inspectors were powerless. The present Act was now being amended and under the definition of machinery it was proposed to include any device that the Minister might declare as machinery. He said he had been approached by medical superintendents and asked what could be done to stop people being mangled by agricultural machinery. He would recommend for a start, that the Bell trailer would have to be classified as

machinery and, as such, would have to be properly protected. The training of operators should be compulsory. He said that safety and production went together.

The Chairman said that the Mechanisation Commit-tee was more than conscious of its responsibilities in respect of the safety of machinery. A good safety margin was always insisted upon by the Committee in the specifications of any implement or machine which it was called upon to consider. He agreed with Mr. Frost that more cognizance of the safety devices should be made.

Mr. Bartlett said that the need for a P.T.O. shield on the Bell Trailer was considered by the Committee some time ago. This resulted in the field testing of a shield with the co-operation of the Hulett Sugar Company at Darnall. He said he had publicized the need for a protective shield for this trailer and had informed growers where these were obtainable, but so far the demand had been very limited. The particular P.T.O. shield tested is called the 'Atko' shield, however, there are many other makes avail-able in England and the United States which could be used in the Sugar Industry.

Dr. Van der Pol wished to associate himself with all the remarks made by the author under the heading of education. As Mr. Bartlett had pointed out the Technical Education Committee of this Association was engaged in trying to introduce a course of Agricultural studies. As Convenor of this Commit-tee, he wanted to stress that this Association did not have the final say and that this matter was at present under discussion by the Technical Training Committee of the S.A. Sugar Association. He ex-pressed the hope that Mr. Bartlett 's paper would bring about a more sympathetic att i tude towards Technical Education on a whole, which would assist greatly in establishing a Course which would satisfy the demands of the author and his Committee, as suggested in the paper. He asked whether the period of two to six weeks was intended to be devoted to Mechanisation only.

Mr. Bartlett said that six weeks could easily be devoted to only a section of mechanisation if the time and facilities were available. Under present circumstances, however, any period made available to mechanisation could be used to advantage. This, he considered, would lead to further development and encouragement of students interested in this field.

Mr. Pearson said in Australia, where the farmer was his own operator, tools attached to the tractor were more important than the tractor. He noticed that in this country the reverse seemed to be the case.

196

INSTRUCTIONS TO AUTHORS

1. All papers for the Congress must be in the hands of the Technical

Secretary fourteen days before the meeting. It is requested that authors

will endeavour to send their papers earlier than this date so as to facilitate

the work of printing.

2. Where possible the manuscript should be typewritten; when not possible

the paper should be submitted in a form easily read.

3. References at the end of the paper should be arranged as follows: Name

and initial(s) of author; year of publication in brackets; exact t i t le of

paper; contracted tit le of periodical; volume number; beginning page

number of article. Thus:—

Brown, S. J. (1933): Study of Streak Disease. Proc. S.A. Sugar Tech. Assoc, 7, 101.

4. All the authors referred to should be arranged in alphabetical order.

5. Illustrations and diagrams accompanying the papers must be carefully drawn

on smooth, white drawing paper in Indian ink so as to be suitable

for reproduction. Any lettering on margins should be in pencil so as to

allow of easy replacement by printer's type. The size of the largest dia-

gram, curve, etc., should not be greater than 10 by 15 inches. Curves

should be drawn on black graph paper, which can be obtained from the

Technical Secretary.

6. Careful correction of typographical and other errors should be carried out

by authors; this would save time on the parts of the Editor and Printers.

7. A summary of the main features described in the paper should be given

at the end of the paper, immediately before the REFERENCES.

8. Authors may have 25 copies of their papers on application to the Technical Secretary. Any additional copies required by the Author will be charged for.

9. Papers read at the Congress will not necessarily be published in the Proceedings.

PROCEEDINGS OF THE THIRTY'FOURTH ANNUAL CONGKESS...proceedings of the thirty'fourth annual congkess i960 of the south african sugar technologists' association . the south african ...

Documents