The penetration of moisture into soils as affected by chemical … · 2020. 4. 2. · wate., but it is of less value in the matter of water move meat in the latter ease,. the continuá,tY

The penetration of moisture into soilsas affected by chemical composition andphysical properties of irrigation waters

Item Type text; Thesis-Reproduction (electronic)

Authors Ayers, Alvin Dearing, 1909-

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this materialis made possible by the University Libraries, University of Arizona.Further transmission, reproduction or presentation (such aspublic display or performance) of protected items is prohibitedexcept with permission of the author.

Download date 03/06/2021 04:42:44

Link to Item http://hdl.handle.net/10150/333156

http://hdl.handle.net/10150/333156

TEE PENETRATION OF MOISTURE INTO OILS AS. AFFECTED BY

CHEMICAL COMPOSITION AND PHYSICAL PEOPERTIES OF

IRRIGATION WATERS

Approved:

by

Alvin D. Ayers

Submitted in p rtial fu1fiient of the

requirements for te degree of

Master of Sdience

in the Co1leg3 of Azriculture, of the

University of irjzoua

yW,4014.

iaior adviser10 I

E777/

6dAL

ACKNOWLEDGYENTS

The author takes this opportunity to acznowledge the

help and assistance given by the members of the Department .

of Agricultural Chemistry, and especially that of Dr. T. F.

Buehrer, with whom the work was done., for his help and con-

structive criticisms in the supervision of this research.

The author also wishes to thank Mr. Karl Harris for his

assi3tance in obtaining the samples of soil used in this

study.

9496G

TABLE OF CONT-: S

PageI. Introduction 1-4

Importance of the problemPurpose of the inve stlgatiou

,,,eoretical aspects of the prob1e4Physical factors affecting depth and rate of'-penetrationAssumptions-involved in the application ofPoisouillets Law to water penetrationphenomena .

xII rvey of th lite.ature on water penetrationBriggs; Mechanics of soil moistureMing: Flow of water thrauel-soils14aaughli.n; Capillary movement of soil moistureecofieldt Uovement of water in irrigated soilsHardy: Percolation -la colloidal soilslicGeorge¡'Breazelle* vmd Burgess* Freezing up- of i13 - - -reGiorge: Effect of fertilizerson water,

penetratioaKarraker; Effect of soluble sa ts on soilmoisture

IT0 Experimental work 1Physical and chemical soil factors ' vestIgated

Solis studiedDescriptionMechanical analysisMoisture equivalent and saturation percentageChemical analysis of soli extractNature of base excbange coat, x

'Irrigation waters studiedDescription of sourceChemical tomPosition

Experiments on viscosityDescription of' vlsco-sime erPrecision of measurementsCalibration of vesitels

Calculation of relative viscosity from flow dataComparisons of results obtained with different

vesselsEalt solutions wed

ConcentrationsDensityViscosity measu ements

At various concentrationsAt various temperaturesAt various salt mixturesAt various pff values

Visotty determinations onextracts at various soilvw_r ratios

Irrigation watersSurface tension moasurclents

Pure solutionsSoil extracts at various soillwater rPt- sPure solutions at dirferent pli values

no of water througla soil columnsOutline of methodEffect or vary1-.ng flit i -_ content

V. Discussion of results 5

Vumaary 0C-.67

VII Bibliomtaphy S 70

LIST OF TABLS

PageI. MECHANICAL ANALYSIS OF SOIL SAMPLES 42II. , MOISTURE EQUIVALENT AND SATURATION PERCENTAGES

OF SOILS 23

III. SOLUBLE SALTS IN AIR-DRY SOIL 25IV. REPLACEABLE BASES IN SOILS STUDIED 28

V. ANALYSIS OF IRRIGATION WATERS 30

RELATIVE VISCOSITIES OF SODIUM CHLORIDE AT

DIFFERENT TEMPERATURES AS -DETERMINED IN FIVE

DIFFERENT VISCOSIMETERS

VII. RELATIVE VISCOSITIES OF SINGLE SALT SOLUTIONS

AT VARIOUS TEMPERATURES AND CONCENTRATIONS 45

VIII. RELATIVE VISCOSITY OF SOLUTIONS OF MIXED SALTS 49

IX. SURFACE TENSION DATA FOR SALT SOLUTIONS AND

SOIL EXTRACTS 57

L.COMPOSITION OF ZEOLITIC COMPLEX 65

LIST OF FIGURES

DESIGN OF VISCOSIMETERPage.53a

2. CALIBRATION CURVES FOR VISCOSIMETERS AT

DIFFERENT TEMPFAATORES 57a

a. RELATIVE VISCOSITY OF SODIUM CHLORIDE 45a

4. RELATIVE VISCOSITY OF SODIUM. SULPHATE 4Bb

5. RELATIVE VISCOSITY OF SODIU2 BICARBONATE 45c

6. RELATIVE VISCOSITY OF CALCIUM CHLORIDE 454

7". RELATIVE VISCOSITY OF CALCIUM SULPHATE 45e

8. AVERAGE RELATIVE VISCOSITIES OVER WERURE

RANGE OF 20° - 40° C. 45f

9. ABSOLUTE VISCOSITY AS A FOCTION OF TEMPERATURE 47a

10. RELATIVE VISCOSITY OF SOLUTIONS or SALT MIXTURES,

NaCI-Na2SO4 48a

11. RELATIVE VISCOSITY OF SOLUTIONS OF SALT MIXTURES

CaS0*-CaC12 s CaS044NaC1s NaCI-CaC12, NaC1 NaliCO3 48b

12. VISCOSITY OF SOIL EXTRACTS 54a

13. INITIAL MOISTURE CONTENT AS RELATED TO

PENETRATION 59a

THE PENETRATION OF MOISTURE INTO 80 S AS AFFECTED BY

CHEMICAL COMPOSITION AND PHYSICAL PROPERTIES OF

IRRIGATION WATERS

,orw1040Am0.0.ow...m.

In =alas=

Every agri6ultural region as well as every arable soil

type._may be regarded as subject to some limiting factor.* -

which's by its insufficiency* limits orHinhibits other desir

able:factors which are necessary for. optimum growth or yield

or crops, Usually such a limiting factor is attributed to

a single cause or condition in the soil* but it is more

likely to be the resultant of a number of different forces.

This . idea is embodied in a striking definition of a plant by

3, F Breazeale* Biochemist of the Arizona Agricultural Ex

pertinent Station* when he says :. oThe plant is the vector

sum _of all of the influences to Which it has been. subjected

during its era of adaptation°. In like manner* one might

define the soil as the vector sua of all of the forces*

chemical* physical biological climatic* or geological,.

which have at various times acted upon the parent rock to

bring the soil to Its preseat condition,

For this reason what appears to be a limiting factor

may in reality be the resultant of several forces* certain

of -which predominate over the rest. The effect of such a

limiting factor is complicated by the action of the forces

mentioned above, and, because of the simultaneous action of

these forces almost any problem in soil research faces its

principal difficulty in singling out certain variables for

experimental measurement while at the same time keeping the

,'other. -r.2nditions constant.

The limiting factors in soils are probably as varied

le 'soils themselves. In one soil it may be the acidity

or ,alkalinity, in -another, salinity, temperature, nutrient

deficiencies, or soil moisture. In arid regions soil moist--

ure is considered to be the limiting factor tor profitable

agriculture. Where irrigation is practised the problem of

water penetration is an all-important one if losses by

evaporation are to be reduced to a minimum and the plant is

to receive the amount of 'moisture needed for its best growth,

Irrigation practice has in recent years (1) concerned

itself primaril with the physical conditions of water appli-

cation, such as the head of water applied, the distribution

of the water as influenced by the slope of the land, fr

queny of irri ation, depth of application, and the downward

penetration of water and subsequent distribution. Such re

sults based upon sampling and moisture determinations upon

the soil at various 'intervals between p.Tiods of irri-ation

have yielded significant facts concerning the rate of water

movement in the soil.

should be noted, however, that in t,e highly saline

and gezeraUj alkaline soils of the desert, the accumulated

mi4eral .saltS rimy in time prove to be the limiting condition.

This 1s particularly true if unwise irrigation, practices:are

employed.. In many cases, to be sare,..the salinity or alka-

linity- are indigenous to the region and brave not necessarily

resulted from Vad irrigation practices. There are instances

*here even with a bountiful supply of good water, reclama

tion:.of such areas is .either impossible or impractical .

It is perhaps more ,proper, therefore, to refer tø the

limiting condition of arid soils as a combination of the tac -

1,tor:or soil moisture on the one hand and an excessive saline

or alkali. condition on the other. The high concentrations

.of. Oluble salts present in the water, as soon or shortly

after it is applied to the, soil, may very profoundly affect

the properties of water that are involved in water penetra...

tion From the standpoint of the plant, the accumulation of,alts in the feeding zone creates. a condition 'which in time

Tay avershadow, the soil moisture iactor. T11, nutritional.

'disorders. characteristic of. desert soilSi mentioned by

MCGedne and BreazeaIe. (2)0 are the resul,t tot only of

paired nutrition due to actual alkalinity tu t' also of

paireViaer intake. In ot4 r,Words, tho two factors of soil

moisture and salinity are int,r-de

the zazeexchange complex is an excellent eapIe of the effect

of alkali salts upon the penetration and distribution of

t. Th7a behavior of

moisture in the soi

oose A 4 es

The purpose of the present investigation was to study

certain paterrelationsblps of moisture penetration soil

alkalinity, and soil salinity. The factors studied re those

of sa lt composition, salt concentration temperature,

colloid content, and their effects upon those properties or

water wilich are ,closely linked up with water movement in

soil -,.. namely, viscosity and surfate teasion. These studies

' were sup lamented by similar measurements upm xtracts from

three typical soils of the Salt River Valley, each of which

exhibited st.r1kin1y different yroperties with respect to

their composition and water penetration. A correlation was

made between their mechanical compositions chemical consti-

tution and water penetration. Certain studies were also

made on the flow of water through these soils under controlled

conditions to observe, if possible, the e f et of varying the

initial moisture content. From these various studies it has

been, possible to draw some ton:Ausions wtich may prove helpful

in studying the probl _m of water penetration.

heo ca as " of te robler

The efficient us of water in agricultural practice may

depend as much upon the rate at 'which it penetrates into the

soil as upon the depth to which it may penetrate under a given

set of conditions. The time factor is it)portant since it is

desirable to reduce evaporation losses to-a.minimum and

not to waterlog the soil for any pariod of i,..e. However,

the growing plant is the main ran for irrigation and also

the ultimate judge of the efficacy of the irrigation method

euployed. The soil moistur6 should move. at least as fast as

the plant withdraws it and it muat not be present in a4ounts

whichwill interfere with the respiration of the roots. In

the Lecì:a21is1 of water penetratic;:a or-mOvement ',Ile depth to

which.it will penetrate is1 but slightly affected by gravity;

.it depends primarily upon th. sua2face tension of the water

and4ndirectly upon the physical and ch.claical factors'that

kTeOt the surface tension. The rate of penetration, however#

depends chiefly upon the viscoaity of the water (salt solu

.tion

Thase two factors, surface

quite closely interrelated, one

teasioa and viscosity, are

being approximately a linear

fUn6tion. of the other for a given solution -Of varylnE cancen

tratian but at constant teuperature. It does- not

110w-over, that these properties

with the temperature .

vlations appear as

affect the surface

and deviations are observe

will change at the .,:avie rate

In fact it is known that marked de-

the tepera

tendon arad

solutions changes.

It has long be

6

ure rises. Salts do not all

viscozlity in the same manner,

the cr,ncentration of the

n recognized that textural and structural

characteristics have much to da with the cap nary movement

of water in soils. genorai, the finer the te ure., the.

slower the movement d the great retention os. : yiater

in the soil. . Various single-value .co Its have been pub-

lished in the soil literature to r present the physical state

aggregation -of the soil particles, ofar as such state

m ay affect . the moisture relations. - . For examp e the liporos.

it is a quantity often determ.ined in c-onnection with moist,ure studies*. it is to be . sure-, of the greatest importance

a a measure Of the a.hi-iit of the soil to retain or hold

wate., but it is of less value in the matter of water move

meat in the latter ease, . the continuá,tY of the capillary

pores as well as their size is the principal determining

factor. ., . .

Al ough there. is an extensive literature on the subject

of water p.enetratio.n, . it appears that a relatively limited

amount of work has been done to indicate and explain how th

chemical composition of the water applied to a soil affects

its penetrating properties. Of the various attempts to represent the flow of water through soils based upon Poisseuillel s

Law it is tacitly assuMed that the viscosity the penetrating soiution is the same as. that -ot pure wat or that it is

a constant irrespective of its content of di f erent salts

different concentrations. It will be hown m the results

of . this stu.dy that this assumption i ot tenable, but that

the '..effect of the dissolved salts and of d. .sersed colloida,.

material upon the viscosity may be so great as to render the

flow equations invalid, or at least in serious error. This

is especially true as found from studie at different &Mat

water ratios, that the viscosity rises rapidly as the soil:,

water ratio is increased.

aymtwn.e_it_tesza_tem...pintaio

. Before presenting the results of' the present investiga-

tion; it mill be of interest to review briefly the principal

research papers that have been published on this subject.

:The problem has been approached from various points of view,

dealing on the one hand with the pure mechanics of soil

moisture movement, the capillary forces involved, the capil-

lary potential, and the effect of texture, al.V on the other with

the chemical factors of character of the base exchange complex,

effect of fertilizers, and the effect of alkali salts naturally

present in soils which have been exten iv ly studied.

Briggs (3) has presented a general discussion of the fae .

tors which enter into the mechanics of ioil'moisture, Be In

cludes the effect of gravity sur4ace tension viscosity and

hygroscopic state of water upon the ret ntiah and movement of

water in the soil.' The theory of surface tension of liquids

in relation to the formation and stability of the moisture

film on the soil particle is also discussed. Briggs points

out that the effect of the viscosity of water is to retard

the establishment of Moisture equilibrium and that temperature

exerts a marked influence upan the flaw of water through soils.

As .the effect of a t.s upon the viscosity was not mentioned

by this author appears that he did not consider this

variable of importance.

King ( 17) in his- studies of the of water through

-soils found variations due partly to temperature and partly

-.to dissolved salts-. The farce of surface tension is shown

to retard the action of gravity upon the cap.ill.ary-held

mouture in the soil. However, the movement of water by way

of afilm from one particle to another" depends upon three

factors:. the viscosity of th e water which reta..rdds the rate

of such' adjustment a chau e In surface tension,. either allavi .

i.ng the film to stretch, or causing it. to be ruptured, and

curvature of the film as determined by the size of particle.

Briggs also (3) shows from considerations pressure of a

water film resulting fr©m ts radiys of curvature, that move

meut of moisture iu the film should occur from particles of

smaller size to those of larger size. Quoting Briggs (p.

nif equal volumes of two soils are placed in coftt.act and the

curvature of the surface is less in the first than in the : seo.,

ond then water will move from the first to the second. if

the ecoud sail contains a greater number of capillary spaces

than the first it will contain more water -when equilibrium

is establ.ished. During this adjustment,. water will actually

have moved from a soil contain ing a lower percentage of water

to one having, a higher percentage. n It i the form of the

surface which determines the movement of the water.

The effect of the finer sizes of capillary in the finer .

textured soils is also treated. In a coarse sandy son the

interstitial spaces are relatively largeoperinitting the water

to percolate and drain rapidly. As the texture becomes finer,. .

the interstitial spaces become smaller, and the number- of ac-

tual capillary spaces increases. Surface tension effects now

become pronounced, and the water tends to he held in the capil

larie- since the extent of capillary rise varies inversely

With the radius of the ca,illary. At the same time the viscos

ty effect increases to the point where movement is so slow as

to be negligible. Increased watr-holdiag capacity in soils

must therefore of necessity be attended by greatly decreased

movement, or percolation. These obfservations suggest that the

remedy for poor penetration is to found in a-condition which

will reduce the viscosity arid surface tension at the saxe time,

so that this film pressure effect as pointed out by Briggs

be minimized sufficiently to allow movement,to occur.

McLaughlin (4) has made an extensive study of the effect

of capillarity with the purpose of explaining losses of water

from irrigation systems, as well as the part it plays in the

distribution of irrigation water applied to the soil. xper

ments were carried out to determine the rate and extent of

capillary movement of moisture in columns of various types of

soil under conditions such that ca illa ity was ass1stec7 by

gravity, where it. acted. against gravit- and where gravity

a factor was e iminated. The columns in which gravity was to

ay

0

assist capillarity were inclined downwards at various angles

from the horizontal; those in which it acted against gravity

were inclined upwards at various angles from the horizontal;

and those columns in which gravity was entirely eliminated

were placed in an exactly horizontal position. Althou h the

factors of surface tension and viscosity, as affected by pre-

sence of soluble gaits, were not 'involved in this investigation,

it is of some interest to mention the general conclusions

reached- inasmuch as they do- indicate to what extent water

movement is influenced by the total,amount of moisture present

in _the $oil at any given time. The author points out that the

effect of soluble salts i not definitely known, and results

thus'far published are in many instanceS contradictory. He

also mentions the fact that while colloids influence the move-

ment of capillary moisture in one way, the organic matter exerts

an Influence in the opposite direction, and that it is not de-

finitely known to what extent these factqrs may compensate each

other, if at all.

When the results of capillary movement are plotted in terms

of inches rise against time in days, McLaughlin finds that for

a wide variety of soils the curves are approximately parabolic

indicating that the initial rise during he first few hours is

very rapid the rate then slowing down markedly until about the

fifth day when movement becomes quite Uniform. it was also

found that in the lighter soils the rate of movement is much

more rapid at the start and slows down much more quickly than

in the heavier soils. The effect of Inclination with the

horizontal was found to be quite pronounced being greatest

for downward inclination and varying directly with the angle

of inclination. These findings are of importance to the

irrigation engineer in that they show clearly what distribu-

- tionof,moisture may be expected when irrigation water is

applied to sloping land. It is significant, however, that

in those soils which are known to contain higher quantities

of soluble . salts, the slowing up effect appeared to be much

more pronounced than in, soils with a low salt content. This

might well have been the result of the gradually increasing

concentration of soluble salts in the water as the moisture

front progressed through the colum.

Scofield (5) 0 in an extended treatment entitled: "The

Movement of Water in. Irrigated Soilet has approached the

problem of water penetration from the point of view of the

effect exerted by the replaceable bases present in the ,Ase .

exchange or ziolitic complex. of the soil. He measured the

of water penetration into soil columns, by compacting the soil.

- (sifted through a -2-mm. sieve ) into glass tubes one-half inrh

in diameter and 5 feet long with perforated stoppers at, the

bottom. The tubes were filled to a 4-ft. depth, clamped In. a

, vertical position and distilled water added to a depth of 6

inches. From time to time additional portions of water were

added. The movement of the line of demarcation could be

readily followed, and readings were made in inches at the end

of certain intervals of time. One of the tubes showed an

excellent rate of penetration - about 33 inches in 3 hours

while the other required 14 months to penetrate 48 inches,

the entire depth of the column. Both soils were classed as

sandy loams and resembled each other so clo ely as to be al

most indistinguishable. The difference is attributed to the

condition of the zeolitic fraction in the soils as a result

of which on of the soils was much more disper ed than the

other.

Scofield also points out that the amounts of alkali ta

soils exerts an important effect u;on the rate of pen_ ration.

For example in some of the experiments, the soils were sub-

jected to the action of certain alkali salts. lt was found

that the rate of penetration was six to eight times as fast

in the samples that had not been so trea ed. It is further-

morepointed out that when saline soils are leached, and as

a result the soil solution is greatly diluted, the soils often

become in ermeable to water, which is the result of defloccu*

lation of the clay colloids. Scofield does not hoAever at-

te:apt to show what the limiting concentration for flocculation

of :ti clay colloid of the soil appears to be, and the effect

of pH in this process was not considered. A

Hardy (6) investigated typical siliceous and lateritic

oil- of the West Indies, a teraining a 1 rge number of physi

Cal constants including sticky point, hygroscopic coefficient

water absorption capacity, and volume expansion . The mechanical

-13-

composition,. humus contenti calcium carbonate cant nt,:p

value, and the moisture content of the air-dry Soil were also

determined. The so-called permeability constant, the volume-

of liquid that passes per second through a column of soil, of

unit cross-section and unit lenth was also determined for -

horizontal flow for the two soil classes, using kerosene and

rater respectively as the liquids. Hardy found that the per

meability constants did not differ markedly- when kerosene was

used-on the siliceous soils, but the lateritic soils gave

considerably lower values, probably due- to the naturally

finer subdivision of the soils. However, ,when water was

u ed, it was found that the lateritic soils are very re dily-

permeable to water: soil from Barbados for example which

showed a clay plus silt fraction equal to 75%, and therefore

largely colloidal, had a permeability cOnstant of the same,

order of -agnitu,.4e of at average sandy or silty non-colloidal

soil. The siliceous soils, on the other hand, were practically

Impervious to. water When liie was added, the permeability

constant for water greatly increased. For example, the sales.-

aus soil gave the following values for Pwx 104: Highly calcar-

eous, .20; Calcareous, 1.81; Non-calcareous 0.37. At the

same time, the lateritic soils give pe m a i ity constants to

ward water of 16.5 and 8.00 which are considerably higher.

These experiments give a hint as to the probable mechanism

of water movement in highly colloidal soils. The volume expaa

sion (swelling) for the siliceous soils ranged from 60.8A to

66.9%. That of the lateritic soils was 112 and 23.3%. cilific

14-

acid is a hydrophilic colloid, a gel, and it affects water

penetration in three possible ways: (1) the force of imbibi-

tion developed by the colloidal component when the soil is

moistened with water; (2) the conseauent swelling of the col

bid; (3) the extent to which water passes thrbugh the colloid

phase. The process may be explained as follows: As soon as

the colloidal particle is wetted with water, imbibition of

water occurs and a tension or forae of imbibition is developed

which is added to that due to capillarity ad which draws the

water from the wet to the dry portions of the soil. The ef

feet of this force may be lessened by the closing of the pores

by swelling and to some extent by the c-,,alescence of the par

ticles. We conceive, then, that the water moves _in part as a

film over the surfaces of the soil particles, but also to

appreciable extent through the colloidal coating of the parti-

iiei. It has not yet been definitely established whether the

soil Colloids behave as semi-permeable membranes* or whether

they act esseiltially like porous materials having re1attve1y

large capillaries.

The above deductions wre made for siliceous and lateritic

sails here the soluble salt content was very low. The situa-

lOn. is quit different in ,ilkaline calcareous soils. The

flocculating action of the 'alkali salts upon the soil colloid

s to some extent offset by the high p0 'which tends to disperse

tLeni The soils investigated by -Hardy were neutral or acids

the siliceous soils having a pa of 7.5 to 7.1, and the later-

itic soils 6.7 to 5.9. The soils foried tlie basis of

15-

the present stuty were actually alkaline, ranging tu pH from

8.3 to 9.7. The imbibition of water by theclay colloids in

alkaline soils is very closely coirelated with the concentra'

tion of electrolyte and the pH, and the isoelectric point is

doubtless shifted as the concentration of electrolyte decreases.

Then the factors of pit and soluble salt content are taken, into

account one finds a fairly close correlation between the sili

ceous soils of Barbados and the colloidal soils of Arizona.

Still another factor, however, enters into the pr.obie m of

permeability of alkaline soi12 as shown by McGeorge, Breazeale

and Burgess (7). They established the fact that, on account.

of the free alkalinity in alkaline carcareous soils appreciable

azo-unts of soluble aluminum exists in solution as sodium alum-

inate. The presence of aluminates was established by electro-

metric titration. Capillary experiT2ants were conducted on

columns of silica sand with which had been mix d certain.

amounts of calcium and sodium carbonates... The tubes contain-

ing these mixtures were than placed in aluminum chloride soIu

tions and the-rise of capillary -water noted in each case. It

was shown that when the pH ranges betireen 7.6 and 11.0, aluminaM

hydroxide precipitates out.and the movement of water through

the column stops. This same effect is shown to occur when alum

or other similar soluble salts are added to soils as amendments

to correct the alkaline condition.

It was also shown by Burgess and McGeorge (8) that the

process of zeolite formation occurs in alkaline soils by means

6

of th'. reaction, between sodium aluminate and sodium siUcate .

If the pH for any reason drops below 8.6 the aluminate present

may precipitate ap gelatinous aluminum hydroxide in the soil

and 'freeze a it up, so that water penetration is extremely

slow if it is not entirely stopped. These authors ree=mend

as a reclamation procedure leaching of the sdil durinf., the

Sii141MST when the t-mperature of the soil is high, for under

these conditions the aluminum hydroxide is.recipitated out

in the lesscolloidal form. Also alternate leaching and dry

in', tends to destroy the colloidal condition and render the

soil more permeable. The effect of alkali salts in this case

is largely indirect; however, the effect of alumirlates-on the

.viscosity of water would be such as to greatly retard water

penetration,' even without the possibility of -aluminum hYdrox

ide precipitation.

McGeorge (9) in a paper entitled: The Effect of Fertil-

izers upon the Physical Properties of Hawaiian Soilsa studied

the effect of a large number of fertilizers upon such proper-

ties as capillarity, percolation, flocculation, cohesion, ap-

parent specific gravity, vapor pressure, and hygroscopic

moisture. The soils were typical Hawaiian soils: yellow

silty soil, manganiferous soil, red-clay soli, titaaiferous

so and 'dust a soil, He obtained some rather significant

results. For example, fertilizers considerably increase the

resistance to percolation, which by reason of the large appl

cations of fertilizer salts used, must have very materially

17

increased the viscosity of the water. He finds also that

flocculation Is increased by the addition of the salts. The

lowering of vapor pressure, as indicated by McGeorge, cannot

be explained from a consideration of the surface tension of

the added salts. A lowering of vapor pressure might, of

course, arise from the presence of the dissolved salts in the

'i.ater film, but it might ,also arise from the fact that the

smaller particles which have a smaller radius of curvature,

would have a less stable film, and therefore, a higher yapor

pressure. It would follow from this reasoning that the par

ticles must have increased in size, which is also ,consistent

with the observation that flocculation and cohesion often

increase the particle size.

Karraker (9) studied_the effect on soil moisture of changes

in the surface tension of the soil solution brought about by

the addition of soluble salts. The ob4eat was primarily to

ascertain a correlation between the moisture condition of the

soil and the surface tension and viscosity of the solutions

added ae used sodium nitrate, ammonium sulf,;,te manure extract

sodium chloride, potassium chloride, mono-calcium p-Losphate

sodium carbonate, and other. The solutions were made up by

dissolving 10 grams of each in i liter of water. The single

salts increased the surface tension but the effect was not very

marked. The viscosity was also increased in all cases, but no

cognizance was taken of the fact that the pH of some of these

solutions were very markedly different. He then added these

solutions to different soils and determined the moisture con

tents after the lapse of a certain number of days- He con

eludes that the surface tension effect is too spril to ac..

count for the observed effects, attributing the effect of

the salts primarily to their effect upon th'structure of the

soil. There is some weakness in Karrakerts argument Jaow -

ever a the respect that he correlates the surface t-ns1on0

of the pure Solutioas with those of the resulting soil solu

tioas. This may-lead to decidedly erroneous results. By

reason of base exchange, éqssolution of calcium carbonate,

and other reactions occurring in the soil_ the concentration

of the soil solution might easily have been decidedly differ

.ent from that of the solution added. if he had mzasured' the

sUrface tensions of the soil extracts t.emselves, he mould

have .secured a closer correlatiaa between the rate of moist

ure movement and the surface tension of the solution. The

viscosity would also probably have been in close correspon

dence.

19-

lalljauLsl_jamjImmkapntal Work on. Wate, Penetratl 11

In order to present the experidiental work of this !laves-

tigation in a systematic manner, we may first present a general

outline of the determinations made and the points it was de

sired to investigate and establish. The factors affecting

penetration which were studied will be brie tir outlined as

followst

Physical and chemical soil factors inveStigatedSoils studied

Description_Mechanical analysisMoisture equivalent and sturation percentageChemical analysis of soil extractNature of base exchange complex

Irrigation waters studiedDescription of sourceChemical comnositiam

Experiments on viscosityDescription of viscosimeter

Precision of measurementsCalibration of vessels

Calculation of .relative viscosity from flow data. Comparison of .results obtained with different

vesselsSalt solutions used

ConcentrationDensityViscosity measurements

At various concentrationsAt various temperaturAt various salt mixture,At various pH values

Viscosity determinations onSoil extracts at various soil:water ratiosIrrigation waters

surface tension measurementsPure solutionsSoil extracts at various soil :water ratios.Pure solutions at different pH Nalues

Flow of water through soil columnsOutline of method.Effect of varying initial moisture content

Since there are still various aspects of the penetration

process which are not definitely understood the emphasis in

this investigation was directed principally to gaining a clearer

understanding of the effect of the-composition of the soil

moisture upon the various forces which play a part in the

movement of water, chief of wnich are the viscosity and surface.

tension. In addition, it was planned to obtain, if possible,

some definite relationship between the nature of the soil

solution and the colloidal complex.

P______ ohe S2111-at110-ed

Three typical desert soils froT, the Salt River Valley

were selected for these experiments. The reason for their

eiectiôn was a pronounced difference in their water penetra-

tion characteristics.

No 1. Glassford Ranch is located west of Phoenix ia

T. 2 N.A. 2 E., N.W. 1/4 Sec... 32. This soil is classified

as a silty clay ,loam. It had been planted to 'whe-t and alfalfa

and was being used for pature Samples were taken at depths

of 11 2, 31 and 4 feet,by means of .a soil tube. This Soil

gave a reaction with pheaolOhthalein showing that it wa s. al-

kaline, yet it was known to give good water penetration.

No. 2 Befsnes Ranch, located south of Tempe in T. 1

È. 4_ E. N.W. 1/4 of LW. i Sec, 27. This souls cla si-

fled as a sandy clay loam, high in alkali and had been

leached some time previously to reclaim it. A test crop of

-21

Hegari had been grown on it. The saipies were taken where

the Hegari failed to m ke a stand, apparently a typical

slick spots'. It had a hard top crust, was very alkaline,

and was notably poor from the standpoint of penetration..

The first foot was extremely hard, the second fairly.soft;

at three and one half feet the soil tube stru. k. a hard pan

about _4 inches in thickness, thence to 4 feet it was again

quite soft.

.No. 3. State Land. This plot is located south of

Mesa and across the road from the Groehler ranch in T. I N.,

R. 5 E., S.E. 1/4. Sec. 34. This soil classif4ed as a typical

clay, and was exceso vely hard throughout the four feet sam-

pled, it was alkaline in .reaction calc Galls, and shOwed ex-

cessively poor penetration. There had. been no evident attempt

to reclaim this land by leaching= The test crop of "agari

was apparently quite uniform, but on the whole rather poor.

.We fina in these zoiis, then, three markedly different .

conditions insofar as water penetration is concerned. In the

first we have a soil definitely alkaline and containing a.

high salt content. (Table III) and with. excellent water penetra-

tion characteristics. In the second, we have another soil

with unusually high soluble alkali salt 'concentration but much

more alkaline in react on and very poor in respect to penetra-

'tion. In the last we have an almost pure clay, low in alkali

salts, containing some free alkalinity, and excessively poor

in penetration.

ch n- alN o'

The textural compo .,tion of these soils was determined

by the hydrometer method of Bouyoucos (10) in which 50 grams

of soil are first dispersed with sodium bydroxide dlutidn

and ,sodium oxalate in a high speed ixing-lalachine for a period

of .fifteen minutes. The dispersions are then placed in a

speciai cylinder' inverted several times, and hydrometer

readings taken at 40 seconds,'I5 m_nutes, and.one hdur. The

sands are calculated as having settled out _tn. 40 seconds.

The 1.5 minute rending represents the amount of colloidal

matter still in suspension, and also Includes the finer par-

tIdes of silt. The hour reading indicates the amdunt of

matter still remaining in suspension Dud is calculated as .

clay. By this method the textural composition was obtained

as shown in Table I.

TABLE I.

MECHANICAL ANALYSIS OF SOIL SAMPLES

Description1. Glassford Banch

1st foot2nd foot3rd foot4th foot

2. Refanes Ranch1st footknd foot3rd foot4th foot

7 4ate Land'1st foot2nd footZrd foot4th foot

% Sand % Silt % Clay

46.0 26.0 28.040.26 28.14 31.6039.64 t,O.W. 29.4049.53 25.11 25.36

' 49.8652.77

24.7027.43

A.5.3419.eo

50.42 24.1151.78 19.51 28.71

42.00 23.82 34.1843.14 20.69 36.1741.22 19.21 39.5740.07 11.99 47.94

The moisture relations were next determined. The moist-

ure equivalents were determined by the standard method of

Briggs and McLane (11). The saturation percentage was deter

mined by the method of Scofield WI as follows: 1,11, sample

or the air-dried soil, equivalent to 100 grams of moisture-

free soil i weighed into a soil can having a tight cover.

Distilled water,is added and the soil stirred with a spAtula

until a condition of saturation is reached. In its saturated

condition th soil mass should be plastic enough to flow

slightly when the container is tipped. The surface should

glisten it reflects light and the air should all be dis-

placed. Allow the sample to stand for several minutes, then

test again to make sure that the soil is still saturated.

-The can with its saturated soil is then weighed, and the weight

of the contents in excess of 100 grams is recorded as the

saturation percentage.

- -Table II shows the results of the moisture determinations.

TABLE II.

, MOISTURE EQUIVALENTS AND SATURATION PßRCNTAGES OF THE &QILE3

Soil Moisture Equivalent% ir-dry basis

Saturation Percentage% water free basis

Glasford 21.12 35.3

Refsnes 19.28 35.2

State Land 25.44 42.9

The data in this table indicate that both moisture equi-

valents and saturation percentages are consistent with the

clay content of these soils, both of which are proportional

to the amount of moistuIP held by this fraction under differ

cat conditions: the ,oisture equ,.valent, the amount held

under 1000 times gravit- and the saturation percentage 1-, ing

the total water holding capacity under ordinary conditions.

an e tratioI of the D_ .v!

T4._3 pg of these soils was dot2r2ined by mean* of the

hydrogen electrodt, using 10 gram: of soil shaken with 50 cc.

of.dtstiUed (CO2 free) water. The measurement was made in

each case with the zoii in contact with the 1:5 soil water

mixture.

The soluble salts present in these soils were determined

from the 1:5 COz free water extract. The ions 1, ermined were

Ca, Mg SO4, Cl, CO3 and LIC03.- These were determined by

methods used in the Agricultural Chaaistry Laboratory, Univer-

city of Arizona. The calcium and magnesium were determined

by soap titration, sulphates turbidmetricall, chlorides, car-

bonates and bicarbonates by standard volumetric procedure and

the sodium calculated by difference from the reaction values

of the ions. The analytical data are shown in Table III.

-25-

TABLE III.

SOLUBLE SALTS IN P.P.M. IN AIE DRY SOILS FO 1 SALT

RIVER VALLEY AS DETERIIINED FROM 1:5 WATEF EXTRACT

Description it ft. 2nd ft.Glassford

Total sol. salts 312 1240Calcium o oMagnesium 8 8Sodium 81 388Chlorides 40 240Sulphates trace 165Carbonates trace traceBicarbonates 183 439

3rd ft. 4th ft.

2648 225860 12023 15

834 633850 920600 350trace trace281 220

pH* 8.33 8.85 8.42 8.31

e4snesTotal soluble salts 2121 3608 4738 3594Calcium 0 - - 0 0 0Magnesium 0 0 , 8 8Sodium 741 1314 1637 1302Chlorides 310 950 1370 1350Sulphates 75 350 700 375Carbonates 324 384 84 108Bicarbonates 671 610 939 451pli* . - 9.69 9.68 9.69 93$

State LandTotal soluble salts 796 793 770 898Calcium 0' 8 0 eMagnesium 8 8 0 8Sodium 240 234 237 456Chlorides 170 170 130 110Sulphates trace trace trace traceCarbonates trace trace trace traceBicarbonates 378 366 403 52410* 8.39 8.39 8.26 8.64

*The pi' was determined by hydrogen electrode method on a

solution containing 10 grams of soil in equilibrium with 50

cc. of water.

-23-

The data show selmral remarkable relationship. In

the first place the soils are all very decidedly alkaline.

The Glassford soil had a relatively hizn amount of soluble

salts but was found to exhibit good water penetration

ilhereas the State Land, a clay soil with practically the

same pH and with a much lower total soluble salt content,

scarcely takes water at all. On the other hand, the Refsnes

soil which has an unusually high pH (0.69) and a high alkali

salt content is evidently in a dispersed condition in spite

of its high salt content as it takes water with great diffi-

culty

. It was also noticed that while the Glassford and Refsnes

soils increased in soluble salt content with the depth,

State Land soil showed practically the same percentage of solli

ble salt throughout the four feet. In the foregoing two soils,

there is evidence that irrigation is gradually carrytng the

salt to lower levels. In the Refsnes soil, it was noted that

there was a layer of hardpan between .3 and 3-1/2 feet which -

was very difficult to penetrate. The table shows that at

three feet, the salt concentration is 4700 p.p.m. of the air-

dry soil, whereas below and immediately above the level, it

is about 3G00 p.p.m. This disparity would indicate that the

soil moisture percolates down under gravity carrying the alkali

salts with.it until it strikes the hardpan and can go no fur-

ther. This results in an accumulation at that level. The high

-27

concentration of ublack alkali' 1r the form of sodium carbon

ate and the almost total absence or calcium salts in the ex-

tract account for the high degree of di p rsion and resulting

poor penetration.

Na .re of baEte-exchange coRr.nlex

The method of determining replaceable bases in calcareous

soils is somewhat different from that uzed in non- alcareous

soils because of presence of calcium carbonate in the solid

phase. Burgess and Breazeale (13) recognized this and made

corrections for this factor. The method used here is a modi-

fication of the method used by Magistad and Burgess 14) us-

ing alcoholic BaC12 as the replacing solution. The soils were

first shaken with distilled water and filtered to remove all

the soluble salts. Fifty grams of the air-dry sample and 250

cubic centimeters of N/10 BaC12 in C8% ethyl alcohol was added

and the mixture shaken on a shakIng machine for one hour. The

solution was then filtered off and 200 cc more of the alco-

holic BaCl2 added and shaken for one hour. This solution was

filtered off. The soil was then all washed onto the filter

and several hundred cubic centimeters of the alcoholic barium

chloride filtered 'through until the filtrate skiowed no test

for calcium. The filtrates were then all added together and

evaporated to dryness on the sieam bath to drive off the alco-

hol. The residue was taken up with water and diluted to 250 cc.

The replaceable calcium and magnesium were determined by

the soap method after first precipitatin out the barium

,r m the solution with odium chromat

Sodium was determined by the Uranyl-Zine-Ac tata Method

as used by the Division of Western irrigation. Agriculture (15)

which was adapted from the method of Barber and Kolthoff (16).

The results for the determination of exchangeable bases are

tabulated in Table IV.

TABLE IV

REPLACEABLE BASES IN THLEE TYPICAL SALT RIVER VALLEY SOILS

Soil M.E.*odium M.E. Calcium ILE. Magn, ium Na/Ca

Glasford 0.584 i336 0.28 0.06

Refsnes 4.26 .74 0.28 1.14

State Land 10.00 11.02 0.58 0.91

*M.E. Milligram equivalent or milliequivalents per 100 gramssoil.

The che4ical nature of the base exchange complex corre-

lates closely with the penetration characteristics of the'

'soils. The Glassford soil shows a predominance of cal(!ium in

the complex and also shows good penetration. In the Refsnes

soil, the sodium exceeds the calcium .s14ht1y, but they are

both of about the same order of magnitude. In th. case of

the State Land, the calcium actually predominates slightly

11 M.E. of Ca per 100 gms. soll with 10 !I.E. of Na. Still,

the soil is so heavy in clay, that it is nearly impenetrable.

The ratios of Na/Ca are very nearly equal for the Refsnes and

State Land soil and both take water only with difficulty.

;it1ot? waters used

Four irrigation waters were used in this investigation.

The sampleS were taken March '"4 1934, at the following

pointSt

I. Arizona Canal at end of. Seventh Street north of the

City of Phoenix.

2. anal near Tempe along D. S. Righway No 80.

3. Head of Canal north side,of Sacaton dam, on the

Gila River.

.4. Canal south of Randolph carrying water fro a Picacho

Lake.

These waters were selected partly because of the differ-

ence in their sources and their resulting chemical composi-

tion. The first two waters spring from the Salt and Verde

Rivers, and represent the most widely used irrigation water

in the Salt River Valley. The Gila River water, sampled at

%,acaton is to be used on the new Indian irrigation project.

The water sampled south of Randolph comes from the Coolidge

Darn on the Oila River, being stored in Picacho Lake as an

intermediate reservoir, and is used on an extensive area now

under irrigation cultivation near Randolph, Coolidge, and.

Casa Grande.

The analyses of these waters are, shown in Table V.

TABLE V

ANALYIS OF IRRIGATION WATERS

Total soluble saltsCalcium (Ca)Magnesiu= Mg

' Sodium NaChlorides 'CIvulphates kS0Carbonates COBicarbonates (HC0)

Arizona Canalp.p.m4

Tempe Sacatonp.p.mh p.p.m.

Picachop.p.m.

730 860 1143 117345 CO 83 12019 g,=z,,,, 30 23

173 205 274 253240 294 450 34870 80 150 1800 0 0 0

185 128 156 249

Experients on IrSCO3itl- of Solutions

As Stated in the Introduction, the vizposity factor_is

-important in water penetration in that it controls the rate

at which water flows throUgh the Soil capliTlarlea and thus

governs the rate at which moisture equilibrium_ is established.

Numerous data on viscosity, covering a large number of com

pounds dissolved in different solvents, may be found in the

literature. Ratschek (18) in his book on rViscosity of

Liquidstf devotes considerable attention to the viscosities of

solutions in relation to such physico-chemical properties as

conductivity of electrolytes.

The colloids of the soil classify in ceneral into inor-

ganic and organic. The inorganic colloids comprise chiefly

the silicates, alumino-silicates, certain oxides like those

of iron and aluminum, and certain inorganic salts which readily

go into colloidal suspension. The organic colloids comprise a

large class of compounds found in the intermediate decomposi-

tion of organic matter in the soil. These different products

-31

have various isoelectric points and as a result one may find

some organic matter in a state of dispersion in the soil at

almost any pH. This condition contributes to the viscosity

of water as it penetrates into the soil, It is found, for

example, that while the suspensoids, that is the hydrophobic

colloids, do not'markedly affett the vistosity, the emulsoids

of .hydrophilic colloids do greatly affect the viscosity. In

the case of the hydrophobic Colloids the viscosity varies in

some Measure with the concentration, but does not become ap-

preciable until higher concentrations are reached. The vis-

cosity is not greatly differeat from that of the solvent in

dilute solutions. However, in the case of the hydrophiles

such as the organic colloids, silica gel, and the alumino-

silicates, the effect of their dispersion upon t he viscosity .

is far more pronounced4 -Small concentrations of agar, such as

0.370, increase the viscosity of water up to 2.5 times that of

pure water. This effect is partly due, to the swelling or the

colloid as a result of which a certain amount of water In the

bound form must be carried along by the particle as it moves

through the solution. There is still another aspect of the

behavior of hydrophilic colloids which is of intprest in this

connection, that is the fact that the viscosity changes with

the time, increasing as the degree of dispersion and the con-

centration of the dispersed phase builds up. This factor may

be important in explaining the characteristic slowing up of

the process after the first intervals of time. A certain

amount of gel-formation may also occur. If so, part of the

mechanism of penetration would then involve an endosmotic pro

cess which is exceedingly slow. This process very probably

accounts for the extreme sIownes of penetration which is

sometimee observed on nslick spotsn of alkali soils.

Perhaps one of the most important factors affecting the

viscosity of these hydrophilic colloie is that of pH. It is

found, for example, that the hydrophiles have definite iso-

electric points at which tne viscosities are a minimum, and

above or below the, point the viscosity curve rises quite

, rapidly. If, Vierefore lthappenz that the soil has a pH

which corresponds with the isoelectric point of the predomin-

- ating c011oid or group of colloids, one may expect the optimum

penetration. FUrthermore if the soil should happen ta be

either more acidic or more alkaline than corresponds to the

isoelectric point, the. viscosity of the suspension should be

expected to rise, and pcnetratiam slow up. This is exactly

the case In the siliceous soils of BarbadoS as found by Hardy.

In the case of Hardyts results it was fdund that when clcium

carbonate was added to the acid ziliceoUs toils, the penetra-

ion was increased about 5-fold, which is a simple corrobora-

tion of the above theory.

The isoelectric points of organic colloids are as a rule

found to be on the acid side of neutrality, usually between

pa of 35 or 6.8. How'ever, there is some evidence to show that

the more complex zeolitic colloids have isoelectric points at

neutrality or slightly on the alkaline side.

astiremel t on visccities of oiution s. and e: acts

of soils

In the experimental work on viscosity, measurements were

made on solutions of pure salts, ext acts of iis, irriga-

tion -waters, and colloidal suspeasions It s eered to be of

some interest to study solutions of pure salts, although it

is.to.be expected that the effect of the colloidal'fraction

upon the viscosity might overshadow that of the pure salts.

The principal reason for making these measurements, however,

wa that the data in the literature on the viscosities of

pure salt solutions are confined to a large extent to more

concentrated solutions, whereas in the soil we meet with re-

latively dilute sOlutions. It is evident that as the solu-

tions become more and more dilute,- the change in viscosity

becomes rather slight. and a more refined aad accurate tech-

nic of measurement must be developed. The viscosimeter used

will be first described.

The'viscosimeter designed and constructed in,this work

was a modification of the original Ostwald type, with dimen-

sions as shown in Figure 1.

-33a-

Fá urc /.

V/JcOs//:14e2zJrseo % . - =1em

The vessel was constructed of Pyrex Glass and incorpor-

ates several desirable features which add considerably to

the precision of measurement. In the first place, the reser

voir bulb was constructed in the form Of a large ellipsoid,

as suggested by ashburn so that the down..flow of solution

from the bulb abOve would not cause an appreciable back

pressure and thus reduce the rate of flow during the course

Of.a measurement.. The tubes immediately above and below the

deliVerybulb were made of tubing of sufficient diameter to

make accurate readings possible and at the same time to ob-

viate-the effect of capillarity. A small spherical bulb was

pladed between the stopcock and delivery bulb to serve as a

trap and also as a reservoir from which normal flow could be

set up before the .liquid passes the upper mark of the deli-

very bulb. It is more desirable to have such. normal flow set

up-before observations are b gun than to start the flow from

rest .as it was found by experiment that such procedure gives

more accurate and reproducible results. A cap was placed

over the open end of the funnel. tube to keep out foreign matter.

The temperature of the thermostat was varied successively

over a range of 20 to 400 C. in five-degree intervals and could

be easily regulated to 0.01° C. The thermometer employed had

been calibrated by the U. S. Bureau of Standards, and the cor-

rections were applied at each temperature. The viscosimeter

and solution was kept at each temperature a sufficient length

of time to allow thorough temperature equilibrium. Tests

were made as to the length of time required for the tempera-

ture of the solution in the vessel to come to the tempera-

ture of the thermostat. It was found that 25-30 minutes

were sufficient. Apparently the glass walls of the vessel

were thin enough to come to temperature equilitrium quickly.

The average dimensions of the various parts of the

vessel were as follows;

Vessel No. Vol. deliverybulb cc.

Vol. reservoirbulb, cc.

17 19917.5 178

3 17 20745

16.516

180201

The average bore of the capillary was about 0.75 ram. and

the average diameter of the reservoi.. bulb, 80 mm. It is

diff:l.cult to construa't the latter so that they will be

absolutely uniform; a tet was made, however, by using a

cathetometer and the increase in pressure head in the reser-

voirs of each of the above vessels was determined. The

readings, beginning with vessel No. 1, were as follows:

3.4 mm., 3.5, 2.9, 3.30 and 3.3 mm. These data show that

the magnitude of back pressure produced was practically the

same in all cases and could be considered as equal.

It was found also that the volume of solution placed

in the reservoir solution affected the rate of flow. An

experiment using different volumes of water was performed

with the following results:

Volume (cc.) Time (min.)

75 2.5603.00 2.630125 2.7$7150 2.870

A constant volume of 100 cc. wa chosen for all subsequent

measurements.

The time measurements were made with stop-watches which

had been checked against each other and found to be in sub-

stantial agreement. In the course of a given et of measure .

ments, observations were made on each visco ¡meter with each

watch. When these measurements were found to .heck , it was

concluded that the value so determined was very probably the

correct one. If there was lack of agreement, additional ob-

servations were made with the varticular watch which gave

the discordant result, and in most cases agreement with

other watches was obtained. Variations were attributed to

possible error in observing the in,,:tant when the meniscus of

the solution crossed the mark on the viscosimeter, or to the

lack of temperature equilibrium. In some cases the first

observations were at variance with those subsequently ob

tained. This was particularly true in one vessel where the'

elliptical 'bulb had been made of somewhat thicker walled

glass. In cases where the observed tire S of flow never came

to a constant value, it was concluded that there must have

been a minute (invisible) obstruction in the capillary. When

the measuring bulb 'was filled by suction the particle may

l'Arf"'1,1"'"'

have loosened up and later, in the course of draining it

may have lodged in the capillary atad again affected the

rate of flow. When this difficulty diD manifest itself, the .

vessel was removed and thoroughly cleaned. the visco imeters

were mounted on a frame made of strap brass Nelicn could eas-

ily be clamped in place in the thermostat or removed when

desired, Care was taken to see that the vessel was mounted

vertically*

Calibration of the Ijimuir_rialt.52.

Since relative vi cosities were to be measured, it was

necessary to calibrate the vessels-under the same conditions

as described above,' with water as the refreence liquid. A

graph.how1ng the ty¡Acal calibration carves' is given in

Figure t.

. The vessels were calibrated at each of the tempraturesof weaurement. Since the determinations were relative in

character, and there was no simple experimental means of de

termining the absolute accuracy of the determinations, an

analysis was made of the calibration data somewhat as fol

lows; it was assumed that the initial values of tifile of flow

at 20° C. were most probably. correct. Using these figures as

a basis, the percentage decrements were calculated for each

successive temperatùre interval. If the vessels all func-

tioned in a similar manner, and being similar in their struc-

tural features, they should, the percentage decrement (or

21,A.L17X,7.:1IJ.7,.1orsiz,t

'All "003i11

.. ,.- ..-._ .. ... . . .. . - C....... .:.. . . ;, 1: u , ._.- ::.::..:. . OF : a .fi.E;; S7 ...i..:...:Pa9 .. .s : .... :.. . ... .a. ...... . ...... CE Ers .. . . . . ,. . . : . : :: . . i , . . . 4 . Q. . . . r........e.....l.$.!!i@f@! !!!!s{{.es.e...P L Pe 7 IPLl=P9ms69.9 ... .Pfii.E ?..a fir@mefie ...n..iP:sfiaL.m s.ss.2...fiPrLa9rL ..:CC7ie:slPPEa p ! a g { agç.fi .Pagg..; 'sE .9Er9rP @ErEEr .i.rfiiE ..rfifi' !:'P1... . r'r:P:LC ' CrrE : ss 7 C:.

. 9 .. .,::rE::E ' fi !' C ' fififififi i

e e:::LSelllli 9.EE: 9 PE. Y..., . . . . . . . . .

E . !!. ...a.a. .:.er EEP:::LP Ee PE9Pr:PeEE:...eEa:fifi.EL....e..n{E PE:i.sii.P9. :sPElEP .,.rE1ifiE::e .L. ... ...YYY.ú... . . . . . . . . :. .. . .. .. ..... . . . .. . . . . . . ..... ....... . 1: Q:e ::: ::::a :... ::::::::: ::: :: :a: :.::s a a m :;E : L se:sQ::. 9 aE . . sa1:s e1 --Q;::;:::ae:::fi::sa:fii:man i :9m . pp!. P...a .fiEss.{ . E. .c 7 EE. :.. .! 9 fi 9s.9 @. E E9 fi:.a ...:e.:fi. s:.E..H ...... ,. . 7 gg 1 ú . ..:, E: C@C ::y a aLLa7aaLSeera Pr a d:E sE .:fi r E: ..::s...{ ..:[email protected]:ss.E:r.sm..e...Yi..... .. . . . .. ... . . . . Yn,.. p . .. p... .... [email protected]= ï r{ ..e 1fi me 7

.. r'r9r r9 :firn :'r!a!9' e: ! : 7::....irfifi:s .... fisr.: e::P fir.i.. ..i:::::{:::es e:.E1Qï :: ::EE9QlPE ..: e:::.... L sa 9 i: P ! 7 :L K :. . ...rs ee: çç: ! s . . : C: . ::: e7 : :: :: : 7:: e:EE::i::::e:..E ::e.a :E E @ mE99 ! : P zri e! a 9E :E EssilP i: : se : a :s9: P : eE a: :: ú 7Ps:mPs: a ru H.HU u u u Huu Yr:° r :9r 99gç ! !' E ë 9EEP'Efi:. . .aE::99E:E°fi: . .. . @ ! r r!! 1 EPr9P9P á E:ae:...i.i..i. :: r : P. =L'r:fi1.'EE9?. .. . P. ua@::P....r 7IB s9P ! 9 7P::PE$P99r.:iE. .9EE . . . . .

e ; ; . . .i....r::ErE. . .: .m . 7. . r! i iP : : P é s ..:e... .is :.. .iL7r...ai....E..fi.E.9 i..e.E.1..iLa E:.: :.

s:E. ...Y.. . .9..s.... ..a.L.. { s E....a.. .....e. .HEse. .... Y ... .. .... . ...fi E... i.LL:QE PPPlrP::m QE : {...:ae 'EEE . .pp.aa :Qr:: : ::::::: :n i. i .. .. . ::: :::. :. :: : : :: ... .. :::: : .. i;. .e::Es,9EEu::L::::P:E:::eP:::lEaE99éEE!?EEEErEEPEá:rá3á@l99EEEE E'EPEEEEEÑE=1.:aE _! i7é'@á=:ú: 1'iE'Pé=:ß P"'.:.E:r

=P :.ifiE.i: r..... s.. . . .. . .. .. ....fii: : :::r:s::::: ::: {.. ... i i..i .ii....i ..n... ;; . ; .. . .::: .. .. Y.. . ....a.E : E EC {Q .P..PLP x977g @ P. ...0 e ..:..rEE7r: C,:::::9::: a:.PC:::fi:: rEi. 7Lr , ,:Pi:,.: úE:99Er nEsPPlIPPPssrearE:mePPrr9PPr.: e ::7:ri ::9: rE:úaEr: úP..EiQ :EE.fieEfirEnPfi:l .iQ{ ° ePfifi:ifi: Paï r:a:P: P;:sE@E:=: =i.; " i 7.9 ...::::9L.: 1Q " i ie ..i.i.E;.a :...Q....L..:r. ... .p .. ... . .. .. . ... ... .. H . .. e. . .. g eg ...:..... L...:9111::s.p.úE.EE19E9@9E 9EEeE::E91:r PseP=rerr7?ém9E:99::::1: : ú:Y....:::::E:Pr:E7e{E : :7::::.

1:fi: u::7rsfiae ....1: mg s. ....fiPfi. .i...i...r:rea:::rfi....ii...ii..i :./.Y. :C::C.. ... . . 7:Ea::::r:::e : ::::::::. , :/:r::::EEPPfiP 'lrEE'r:r: E. i.i ú ' .i .e.fi:.E....: E e....: P i. : P:9P E ?: :: E:s fi :.....L....H.:::77C:3ElL.mL.... .. . m . : ..r ::: . : aC a :::éi::!=E?E=9ré@çé?ilééá@l::á@@'erE á::7::P:a7:s=:e i.C9°:EEiEE9E?ë ° :g::1:9:::7:sez:ál:E:9==P9é9EEe i : rue =P'cBE"fiPfiP3E'Q::: : .:::: 1:m: maa ::e. a: .::r::r ::: .... :: : ,. :C s: ::: ee:eE .e. :Q. ..7 : . ...:e P.a ! .. r ' Esa 1 a :: a :: r 1i: : a9fi a.P E.. r::9ls9EEPPP9!Eme9s::: i 7 : i 7 aeem7e: reE LIE.: i..a..p.e.. .L.: . "wir:: =1r:1EérC"r:P9CP @1é1111éC7C:PÉ=1i1PPEE@CEl=EC

EEnfiL :aPE: elm :

i':QiE=:7G:117:L C:r ='e :

::C:S e:a:7Er.ErEu.i.1 PE'r : E'E2ï EE=:pplP=:a7:sE:r::::=Pr'::P:EEfiEfi:::::.i 7 E°:...i e:Ee i"'7 W Amu.Pe::7aC:efi¢a9iiKPesL:e// Pir......ép..psa1::: 7 7:C:L" : : :;::e: s... aP9'rlEPPP .E' r7Q:79::9 E.p :::7pne E r::L iea:Er9E:E9Eee .wr: Wills....... ..p:E fi...e.Cl:::aa::9'P::::r:7 ú7eeel:ea.r..Er:.r :.E P rú... ... .. . :. .1.. ..:H..:Cu..Y..:me..:s:rr Y...ú,:::::'7::;:7:aa::' :::ppP:: 1 : ::::a : mH:::1::::r7 "" "

e e@E' çaEmCiP6=e°.;'r, ... s. :,i 'On Ill 7 :rial

71ee

eQ.1.C::EEr ...EE:r7e:ru::9. .... .:.:...:....:. .... .

:1E'

i99'°:E::fiIEEEPP@iPe6 =--w_:Hú7:::r r :::::r:r:::7 rr::r:::Ew...W...P....r:e:7: :99PP. .. 7:: : ... 9Pae r:E9:C: "° : :i r : : e::E::PEP: `r 7::E:CC i.E.a:.nsLe:C r T -- L:.. e : e1u7 7 L E:e:::i'rum: iG"'e : : r: _. '1 .,1/...Qr :fi:::C::CE:QL:: lull nnin: ' ia7 r -=i:::Ç11111111111""

1

"°"ë .... :ássm r:r 7r w .....;:Ce=::7.7oH.... p EEBEEP.=.EEE.7.::..::E99QE fiPr..: r :.. .... H...:..7 E_p:,nW. rH:s::Gr =' C"nr.'pí'Y e,E L u i

su

...e..

7

.sumi.:

C..eTe:ee::=in?7::.. =

P9'Pp.. P::Q: :':: :ra"

MSS-.w _ '

uúwm ::

ai`

ar777

:

.íru a nuC a7

:: .E aps P. s L i::ai:r7:C 7::ruse....n.p....,...:.C`

H

E 7 e. a m '..

7 7 r L :n....... Hr. L' 1L:.Hnpu ,m'. . . 7 Qmien .:..

._1 n ..CX -np.r

-- a:

.... .. .

_ , ::`?_E : :: sr a'H=::-: 7 7 ... .úie .. ...

....

:7....E.:.E..Y7.:.C. ' . .

....CC

.

:C;uuuu u...-

.. .. c .p'.... Q ...

... Cr

rHnQPG.-

.. . ..

i

.. .... .'

:....:: L . ..i cn :.

Q.... ....:..we::' D ::::K; 7L .... ..e:aee. .... C C.....2L.::HH....i::r: ...Q a: .. .. ._ . °- _:..: : . _ ....p.r ...er r L. . ú...:.... L .:. ... ..:n ..... .i...n.p.....ú..e.. .E,. a. :E:fi::9?PE.u.ww r ...m..H.: e::P::á 9::LE9:::. ... . .¡¡¡ ... i.... - .., ..Yn?...ww.......n ..°L z.... ....H... :o.:., .,_., ,a r:PErE°6r: rE°°:: p ....Y.rPEPe:e. fiaLS...9ú ... .... ... . '..:ú. ... : a9 H C K... .. .. H..::7 a: :... .. ::...W::..: .i : L... ..y.aT

\ ; ', s°':::r:::::::::.::::LC LE:rLL : :r L:L: r::r.::: :: CC :.'_-

--0---..---!'- "iE.::::E.r.0E'P9T.H.H'i C L r ::°.aripÑL. . !:

C:::.K. e. . . 9 9 Si.' T.CC1': : a9:: 9 :...., ....;

i` \ nEE:.W.nW...... .EE:: e..Y..... _ .e ...\p.p. : i.....n........ \ , fi7 :.H..wCCC:u:9 ::LE.Ñ:1

: : Y' :ne ' H1Er ::Ke:::L:2111

. ,HHn :: :HH C .. _: : . n. :a:ar?.'aP eun7: : : :He ' ' ,_ \ \ H.: w e 9 . ..9:.: .i. ee :

::11

::::r:rti:rr:::::: r::::.rHn e.n.sLWp..m::ra9

CC :.....p...Lfi7:1:eC::nm....:.....

::::Y....i.. CpQ ..::Es:7...

C::C.Y...Y .:.:Li . H.ú:::r:u CC::e4eQ r9p:u1 : i

ni'..n::e: 7r::eu

enp..Y..... .u

ú

OW:

s

H'11:11 re::.Y..C....i

e

e

rQrem.HU ::::Q...: ::::s

..:WY.:.../p Emu. w Qi . rii

iH..H_

:1. Q

::W ,7,!::::

::::_ :an

eQ

eel: brit ....PrEEEáriEE9r1E:..p.7

h " t

7L.

::::::C:. C:::='1 _Q:::::::E;::...i....iQ

n:r:nnnr r:uuYu

s

e

p..nn.Y1

7turi:

rHrr.e:

1 1 111.1.i r::::.m:m:LNe:Ni . HQ...! : w;^ 1pH i.1.1.:L 77:

P W7.7..Q:r:QriQQr'E....:::::

a

e

e

Q

1 Q7 C:.. ........ Q ...........u...7

Q : iQ L:,n C1 w. e ....u::L .Y.....E.H........H.nYEr::7:r:::a ......WQ : C:... .. ... a7::ú:::e:e.......1 e w :

Cn....n._.seesppiú

: ...r Y.Y. ...../:Y..Y Efi111...:..1.

! :

.r:::a ........ :E::::CH.;;;i.u... :i7EEfi:=:EP:C'r_r:r"

7 ::y III :::::: :r. :E::r::annEr. 9E:E 9PE: um, : uEEEfie... :r:77Y...°..':. °: =Q:ne:: ..ee ::iü:úr : e:a:m..r....1 K:::S:::: :::.z::G::z .pnuQ úru9 su, : ç3ru a mnwp: H: uunWUúsuussii3. Hpnw .. -wiEC ::

: :Q r r.pr :L:ECiEQ:aEEr:ww.:. e , rEp wnwr.\. .,Y.. .qIV/YYWnw ..r pH..../....1uiuHnu 4iiylii:Qa. imríú::: úvfiC,i::{1Er:ii%'i iiQ:'1' iiiilii:ir i::

r.::: .... ..r.... .eE:E: _::::: 111119111664111::::11.31..::s:.={{{{{{{{fi{@{{{{{{e{{{{{{ip7u:i,HL1n u HU:::iwnnw::::=an1 1:::Eá:a::E::m i::E.1p.::r:uE:sáEEäC=t.ws Q":..::ww:ï% r7::: :

:E::::C171

: EQ.n..1.Y'uHur--.. c .acsa-" -r nE.en.Epr n ,....n.i.:r:m. :. °s:C: í`':iï..._ L:r

1 1

G1

IS' 1fj73sS

t'oj77S'rj W

r:::r:::npi ' nn........W . N . _ _ _ - _ _ .

........:

r. .

. 7ï

Q :{

-

w.E

Q

r r'

n

i

.EYuuuu HCCE

ii:r:::rá.m..i H :...Leare' HrE

iree: {mhos

Mil EE4u0rsal7bJw n

i.... H:::E:p ....

SUM n1a: :í.... iuu.s:\:Q pni:::rnr :

-

-38

increment) ti the rate of flow should be the same for all

vessels through any given temperature interval. As an ex-

ample, the percentage decrements for the interval

were found for five vessels to be as follows:

O. 530,0.2760 0.273, 0.271 0.273.

The agreement is found to be satisfactory in all

o 350

ept the

firat vessel. The same lack of agreement for Vessel No. 1-

was also observed_ at the other teperatue intervals studied

and ttwas therefore concluded that some conditions possibly

lodgment of foreign matter In the capillary and not necessar-

ily lac% of temperature equilibration, was reSponsible for

the discrepancy. When such disazreement was foUnd, the ves-.

sel as always recalibrated. mall .other ca.ses it was con-

sidered safe to conclude that the vessels had 'attained true

temperature equilibrium and were functioning normally in

every respect.

C 1 ula on vi

flow a., d de

a ; ttne _a es

' the solution,

The relative viscosity data as obtained by the foregoing

experimental procedure were computed from times of flow and

densities of the solutions by means of an equation derived

from Poinaeuillets Law:

4V =,

I, rt

where V = volume of liquid delivered in tie T.

P = difference in pressure under which the liquid flows

through the capillary.

'70

R = radius of the capillary.

L 22 actual vertical distance in capillary traversed

by the liquid during flaw, and maintaining pressure differ-

ence P.

= viscosity of the fluid.

The pressure difference under which flow of the 1iquid

through the capillary occurs may be expressed in terms of

the density of the solution and .the dtffereice In levels

between the marks above and below the delivery bulb, for a

given liquid of density d1;

PI b, di g

and for any other liquid who e viscosity relative-to it is

to be determined:

Pa h da g

The common form of Poiseuillets Equation is as follavs:

PR8V1

The relative viscosity becomes:

Pl R4 Tl

T

'1 =e V

p2 R4 Tx

8VlSubstituting for P1 and P2 in the foregoing equation, and

cancelling out constants and like terms, we obtain the final

form:

dl T1.

; d TA2 2

40

This equation was employed in calculating the relative vis-

cosities of the solutions used in this investigation. The

values of d2 and T refer to density and time of delivery of

pure water, and the terms with subscript 1, to the solution

being sltudied. As the solutions become more and more dilute,

the density ratio obviously approaches unity, and can be

neglected as soon as the deviation from unity is found to lie

within limits of experimental errore

obtained f

iven tion when measu ed simulta_ ou lv

a

different

vessels

It was considered of interest to measure the relative

viscosity of a given solution in five different vessels under

identical experimental conditions to determine how closely

the results would agree. Such a test would also afford evi-

d.ence of the extent to which the precision of the values ob-

tained was dependent upon the dimensional peculiarities of

the vessel. The solution chosen for this test was N/10 Noel

at two different temperatures, 30° and 35° C. The results

are shown in Table Vi.

-41

TABU VI.

Viscosit of N. 10 NaCl at Different Temperatures as

Determined in Five Different V , cos1rneters.

Vessel Temp.No.C. Rel.Viscosity1 30 1.01444 0 1.011D3 It 1.01154 a isolai5 v 1.0130

1 ,z-tap 1.0134,4,,

ft 1.01103 it - 1.01234 It 1.01335 II 1.0119

Ave.Viscosity

Deviationfrom Avg.

1.0129 0.00151.0129 0.00101.0129 0.00141.0129 0.00081.0129 0.0001

1.0124 0.0010 -1.0124 0.00141.0124 0.00011.0124 0.00091.0124 0.0005

The data show that none of the determinations varied

more. than 0.15A from the average, and the majority of them

showed much better agreement. We may assume, therefore,'

that If the experimental conditions, especially that of

attainment of true temperature equilibrium, are maintained

constant during the measurements, the viscosity measurements

will be reproducible to within 0.10 per cent. Table VI also

shows that the values of relative viscosity are substantially

independent of the vessel S used in their measurement. How-

ever, in nearly every ease, the final values as given in sub-

sequent tables, are averages of two or more determinations

made 6n the respective solutions with different vessels and

at different times.

-42-

Viscos t' Determinations on Salt Solutions

As stated in a previous section, it was considered of

Importance to investigate somewhat extensively the viscosity

relations of salt solution, particularly those which make

up in large part the soil solutim or closely approximate the

solution represented by the irrigation water after its

cation to the soil It is remarkable that the data on vis

cosities of salt solutions at high dilutions as recorded in

the International Critical Tables 19) and similar reference

works are rather limited in extent. They do not extend over

any appreciable range of concentrations below tenth-normal.

The choice of salts for this study was determined in a

large measure by the average composition of irrigation waters

and soil extracts. The salts most commonly found in irriga-

tion waters are the sulfates, chlorides anc-vbicarbonates of

sodium, calcium and magnesium. The sodium salts usually

predominate, and the concentration of calciva salts is usu

ally about twice that of magnesium. Small amounts of carbon-

ates, nitrates, fluoridesi potassium and boron may also be

present in the water, but the concentrations are so small

that their physical or chemical effect upon the water may be

neglected. Of these latter salts, carbonates, if present in

any considerable quantity, give water an alkaline reaction by

hydrolysis and affect water penetration through their dis-

parsing action on the soil colloids. The salts accordingly

chosen for this study were those mosteammonly present, namely

sodium chloride, sodium sulfate, sodium bicarbonate, calcium

chloride and calcium sulfate.

The concentrations to be studied were also given careful

consideration' Irrigation waters may vary from 300 to MOO

p.p.m. of total dissolved salts. The latter figure is

tremely-tigh but cases are on record Where auch waters have

been succssfully ilsed in the production of alfalfa and other

alkaliresistant crops. If such a water contained only one

saite the concentration would be less than 0.1 normal. How-

ever, the soil solution itself reaches a much higher concen-

tration. Waters are usually composed of several salts whoie

individual and total concentrations are less than 0.01 normal.

To cover all possible èasesi the concentration range chosen

was N/I, N/100 N/1000 N/1060, and N/104000, and the tempera-

ture range chosen was from 200 to 40° C. in five degree in-

tervals,

.The solutions were made up from the-crystalline salts

Of C.P. grade, which showed upon qualitative test that all

elements which might be present in appreciable quantity and

interfere with the determinations-were absent. The solutions

were made up with distilled water which WIIM tested showed

absence of chlorides, sulphates and ralcium. In. the cRse.of

sodium sulfate, a solution somewhat stronger than N/10 was

made up, analyzed, and the dilutions made accordingly. The

saturated solution of calcium sulfate was analysed and the

dilutions made up from this solutian.

Densities of the solutions were determined by means of

Sprengel density pipettes and with ordinary pyknometers

having capillary stoppers. These vessels were kept in the

thermostat at the desired temperature and filled with the

proper solution (also kept at the same temperature) before

weighing. The pyknometers had previously been calibrated

against wr,..ter at the several temperatures.

The viscosity measurements were made as previously

described 3.00 cc. of solution being placed in the viscosi-

meter and time being allowed for the establishment of tem-

perature equilibrium. To eliminate a possibility of error

in the time observation, sets of observations on the time

of Slow of each vessel were taken with four or five differ-.

ent stop-watches until substantial agreement was obtained.

From the density and time data, the relative viscosities o;

the solutions were calculated by the formula previously

described. The times of flow were found to be reproducible

to about 0.005 minute in a total time period of 2.5 to 6

minutes. This agreement is regarded as unusually satisfac-

tory in view of the fact that different watches were used,

. and different observers made part of the measurements. The

temperature in each case was held rigorously constant.

A summary of all relative viscosity data for the single

salt solutions is given in Table VIZ. These results have in

many cases been rechecked to confirm their accuracy.

TABLE VII.

Relative Viz,co.Aties of Lì 1e Salt Solutions a

Salt

Na01

Na2SO4

Normality p.p.m.

0.1 548500.01 - 5850.001 580.0001 5.8

0.1 701000..01 710

- 0.001 71

Temperature20$ 251 301 a5t 40,

1.0136 1.0140 1.0130 1.0125. 1.01681.0037 1.0036 1.0028 1.0028 1.00471.0022 1.0022 1.0022 1.0034 1.00221.0027 1.0017 1.0011 1.0022 1.0017

1.0267 1.0279 1.0266 1.0244 1,02451.0029 1.0029 1.0034 1.0030 1.00311.0027 1.0022 1.0009 1.0005 1.0010

0.0001 7.1 1.0010 1.0008 1.0010 1.0012 1.0012

NaHCO3 0.1 80400 1.0328 1.0318 1.0316 1.0319 1.02850.01 840 1.0020 1.0018 1.0014 1.0012 1.0013-0.001 84 1.0016 130012 1.0005 1.0017 1.00050.0001 8.4 1.0010 1.0000 1.0015 1.0017 1.0000

CaC12 0.1 « 5,550 1.0291 1.0322 1.0318 1.0320 1.02690.01 555 1.0019 1.0016 1.0025 1.0029 1.00360.001 55 1.0008 1.0016 1.0015 1.0002 1.00000.0001 . 5.5 1.0005 1.0016 1.0010 0.9937-1.0014

CaSO4 Saturated 2,080 '1,0114 1.0105 1.0114 1.0101 1.0090- 0.01 680 1.0026 1.0021 1.0020 1.0016 1.0029

,04001, 68 1.0011 1.0013 1.0016 1.0017 1.0011..0.0001 6.8 1.0011 0.9992-049997 1.0006 .1.0011

The data included in this table are plotted as shown

in Figures 3, 40 50 60 and 70 in which relative viscosity

is the Ordinate and the abscissae are represented as logar-

ithms of the reciprocal concentrations. This latter unit

of concentration was chosen so as-to_cive the widely varying

concentrations equal importance or significance in plotting.

Thus N/10 concentration corresponds to the value 10 N/100

represented by 2* N/1000 by 30 and so forth.

g .9J/7o &troyod/ad.,ero. yo 907

zO91 Z If(

'A 'N "OD 3 1, NiaEEiaaaiiiiil IiIN EE .:... .iálsiaaa .. .. . e .. w ma 1 mm . ... . . . Nc. . . . . .. . . . . . ...gE: ... ... Ns1.....c1a E E1 1 E 1=1d1 : IE E E E3 = 1. : .. . .. .. .. .. .....: .. . Ii1aE :a. IEEi i i.` ae aa Eaiai i 1e .e=a1E ' a:m1Ee7. a. a m7 7 7a1a a7s' :: °: : c'L9 L7a: aG G : L C :!'cL' 7ú'7 .. . r7777"'7"G:E i 1 E E .. ... . '1.. LEaiiliiilllll iilR IIE1E:1°EIE3a.111=ma1i IIIIi Eililiaell=a1a=Ei1aE1a1elliEilaiaaiaaEga;I111iEIlIi11==ie;alla'1111=aa11 ú1gL1171Cú 7g::. . : . . ! :. '. . . _: .r .. . . .' . . . : '. . . .... .. .. 'E'I.EII IE111i1lllelllIE==ei$ei9iéillil°iaEá=ai1=IIIIIIIiI==EilE=li==.a=r;I::E;. . .. . r . .N ... . .. . . . . .. .. . . .. .. ...... .. .N .1191111 n . .1E.L . E 11 7 S :IC E e° 1 a a E_= 11= L am77 L .ligi 77 ;;G ..a. ..a7=17=:E

gg= 7a ' EEIIIIE iiii Ili i illlili III==aii EIa1G=iEImE '11 I 3a1 iliE I ''1EIEIaú Iai1Ea1 "' EIe i 1i . .. ." `3.S..13..SaaS.. aa a.. a i. .g N .. 1 7 a. ELEC . E.. 7:aLe eE ;° 1a===..a.7..a=:77:==:1':iE`I

179;a =y'i e ` 7 E1.n 1: r u . rr .7 7 L.a 77. .m .m L 1 1 a L.EL a6 1 ep!1_ °1=1m:: :1:: N .:..7.i=..:gN..=..alEialllil IlEi1 iI1I111Ii11°==7E°=E I=..r.=171EICIIIi Ia Iiia7ila1E1=°E=úE1°1=a1il1i1111111ai1IE731:=.1::1:CiLa:aC1L=a= = ú a11i13E.iE:7. ....N . . . .. . . .. . .. ...

. aala..1' E' EaE ' ;=1131e E EEEEI E11a1 aEEy ;eaGLa'111a1 1el 1g:g19=; e: E1Ee ...........=:I:Ca:a ' 1EaEa ail allii lai N$ II .. ' 71úY : 3e EE I 3 aa "e 1 ep p p

. . . . a Lm 1l:... n .. . .... . ... ..N... ..........::1`E''Ea° E 1 EE

1 mE9gggúú7aa1 G711a y 7 úCanY aúaL Y1Ia 1 /EEYag Ega:7E: aúg.y;úúCLaE ;C 7 g ga e y aelE E.L eE gg E ÿ G.L..Ca7C.m .1= ` a. Ililllli ie1 IIaI1a1' Ii111 lúa. 1 1E m3CIli1leeeaal 1 úa7 1aC g L Ea ` ié `E 1: E a E='lgGGe''7 7. . . .. úaú. .. .. ..7.7 ... .77.CN l 7 ... . ... ..... . . . .. ... ...N . ....n ... . .. .... ...... ... . N...... ... ...g.7 a I L 9Eaaú1n77: ggg LE21 1 e1E`. y':3a1 .

e... . . . N. ... .. .. .. . . .... ..IipgiaC1 illpem. .

aag N 7777a.;. a ia 1 gay7.7..pg .7 71.. 1 . 71Ee 7=1.. ;:.7 ;L.GG;;.. ..

a ' E E 'EIiElElil IEIaEaaamalle:. ' ' . : .. a: E :a = ° e EEa = E I'E1°77E1e1G7NE:e g Çg'77 7. ... .. .. . !0 :G 1úC 7 Y: eaaa C a.a ! 1 G 7 a G : G.:s 1:ú1a s 1 =e1 ° 3 E1L:a Ls::la: . . ... . E1: E '1 "a' c 7 1 Ea a E" '1= . . "'1= ' "11 "°=L' 77a:Y.e C. Lla e.. .:. : .e:. .. ú..1:.9 . .. 7ú.. 1 e.... . ...

i1EE1EaEE1=g ag

ç; ' a;g ; 77 . =.n a ' E 7l Y ;; . . E 7 . C:7'G',E .. 'C '77.7... =CC77:77:'.IlEii IIIIIaGa1:11Iial1EE=m ICm11E maEm1..:Y..Cm....aG.a:BI.Le...e CC7aNE:.Ia.a.aB.LCaI...Ga;.:.am..7C L..LL 7 LeS

.I= E'sll.i:° Ee=EealE7galaa°IlillaailElll=EeEEai aé'gC.7a'=g ludo:.

°1i'31 . 7EE=l:E'a : Ea'g i:Ygg GL EE°Ea1aCL77Gg=.;..;..;..; L CL9 G. aL l 1 ! p p' '!m : l.Y.:..::a:m: m2: LG 71 7a. : ... . ' :. E .a:wlaaw 1.7r7 n: 7 : :g:ú:77I `' _ 1=3E`ë `E= E` "17 aC 1 EüE aöiaEEE _ "' :7 :=CG1 E 1 " E :7`EE e . ..: ,rp p :.:7 : .. .C.N Y .:C 1 . L . 7 7.:7. . 7lC . .a;.; I;.. T: ?:."-1_ _ . . ' pE°I .; ;' ; I. .. .. ; .;. 7 ..L.'7 ' GIIc.. . .I. . .. ;7 L 7g;. ..ilEililla117.:al.aaalelaaEEsEIaE 7EiLl.e.mlL...,Nacml:7aE.1E.Clalaa11E1.71S.mC.L.1 10Ea....7..0 7.CIa1Ya7Sl1:GG:E.ú71La:a a'E i°'.gE 11==' E_ 'EEö Y'' paE°gEEEea e °Ca=1 eEE:.=E7 c 113E ággEEGa:G17aE=G7C71gmc77L=laClaaeggggegYggEagggeaEY 7GE gEe=°=7=7g=7...e

71i:ga=E111éiCeaalLaaag°7iiili5:eagG7:7:7aE77T:7 a EaaEaaE...i.a.... . . . .. . . .g ... ...;..7.7E..7LL7l9Emei : : n¡¡

7 1 a L.a.aL.l..7l.m.mú7. a...71.N. ..... ..,S e 7S1' d 'i''aii

r. . ..1 aú:.N....;; yN a7 p ' 7 .y1. .r ... ú: r.. 7C7 :'-: '

1 1! :. a E gal : 1 ... .. ... 11 a 13e a==1E =C1.7. ...aaa..m.17..7a1 ú1LLL71377...G3 10E777 ú7 77 .-_ ....7.1= ..7.111141111111111111111111111111111111111111111lE1gEi. g.: IIa111I11C1EELailllEllaa1eL71ee CE171em1i=;.C;.y.G.; 71:11"7'77"lCYa'úarlal.=3.,°GC.;; úa 7lLC7L!'1'=""1E;: ;71I I= '=m Ia ' 11 g ` T'ael=11E17CC E``GE.....17:L 7eL '''77 °g11 C 1'7:. - 7 úYL7G7CG 'y p p a Ç.:-:.xx . .- e'7

EeSS1E::.

grEG3 I:i ; iE 11 ; úrCä 17. ....N.. ...y.E...aGEC.. y:áT.=Ga7:e!9:G...7.E1 Ii E Ii alp

I 1 C G. .. E. u....:`T N..7 :hillkliPMEIOPPIIIIPAginal7g1 =a Ea' gCELLE1EIEg:Eg:::7a:a7:=; :77:I1gLIiú73= e...7111:7::..c>L'°:á:::C::1::1::'e1i`Y1C1: 11=7L'7G==E 1..=7777=: °iea=1 E1g a7 . r;.: °C a==e1C1a a°_i:='" 5_.. : z"Y'=°'"Nre __ ° : °E1m. N . uN Nnnn u .rr.nLa -iaäo` 7 17 ú

g1:: Isgi:==E111;.;3;.11; .. .. L 71I7Y : =.7C1 Iaae°=-=;..:_...,1E .. 7 ICaplGig7 Ai CL7:7 g 17a'yi . r . . . p ._ ...:.... .. ...=:7:7=l: :'cúú 3 .. E17a E a 7L 7 = -- .... .1..7 . .1 . .>>--.. .- . . ...:CG C . . ..ws5

p Y1.a1:.Ia7SaEa....n.1G111aE.Eai.Ep.I.ElE1aa.E 7....5.r..:.a..Ç. ...a..7.a.z 1.SL...1 57.7.C.L7. ... LCL7LN. 17L: °e'E'l:Sa.

Ca9LúG:a7=l':e 11EE1=3 1a'I YE' 311,' =::?.'::=GG' CC :a: .e::m...p.7: ageli: =gC1eeE111i7ú9711E11mE ; 1. ..

..7 : 11.1...a13.a e.71.1 ....y.GaES"G.L 7L.. .7...r.N.a 31 .1.77.N. 7 ' 17 7C7 1 ..1..

11111111111

a7 7....41:1:111:1111:11T: Y'La 777 1 ü1!Da'ra7lúr. . . N. ... .n.C 3 LaL . u .N.n.. D- .... ... a. 1=1Y1G' L ...L3='="a1lY17 7 N a.77..11. ú. .. !.. 7. ..a. ....C....1:.... 7Z: .. 1... ..pp.C.7 e.7 7 11..13=1aú=1'CCCL1 1 gú1Y 1 ;;1!l....0 ! !:l / úaúL mL.1.a. 71.L ..N11...5.7 1N.1C/ .

eüi::==ii7ú`eLC.NpE19l1' 7777ú:::a771i7ggi

;1;;;y Q m..... / a 1 1 C:... a.L "n... N . .. . 'L .: . 1e..aa71 ú7C.7 . L ú7G7 7L''. CClCL': r1 _ ...NEn.a.1 7.19 e77..1ä7 e .... gNal.. . a.1 e.ä77. n 7 1e7LILE IaC7ú71` ...... .. a LC. eE7

/ LúY 7' a..11:.L.e.'7C;NÖ Ne.i1e377171171E CLL' y1EL'9111 :1G71a1=17771°.e7=L'LLL:C ;Y.177a:L 7E . .. .I ..l ...a.lGe..L..G7..;1. . . 1...7. .ú...gs .L .l..aa...1L.N.;!m 'aCm7'G.;..7;g...; S.m...;.. ..,N.. .. .. .... ... ... .......... .r

=Ea1 ú 711117:g7ClIIE1 i r;;1:7:"laaaL77Tae1. o:11iT1eee;;.N7777g'e°'e°,Eñ`17a3Y'aa ! 1111EpG=E3`=Ea1l. _a;.a.C..NE1 LT=Eg1m11 um.,.. =177C1ú7ú :ú'a1....NLaGCGúúa777G771'E.C..Cee;;ENe...;;aLEN1a...yC GL7......n....Ngn N uunu m m unN uu u. .. .u'i':Gl. ..n9 Y ;11111-;

.1. C. :l.L..1 ..7..C... 1 aC.L..Na...1. .CL.C. 77ä. L. NN N 1 791u . Lg;. G177Lgg1.: ;;.... 7ú . 7 n.N.m u'a l....7:C..:....1.. .7 C..L...;;...;..; .IC..CNnN... Em:LG' y;Y.. ....331;;...G.;e77...:..:..:....L;..L7 g.L..L.: :a :.7.N.n_I ' L 7.N..rr ... .. N .1N.NNn.GN .... 1nNNN . ... .. .. N.. . - .... N .. , .. . N . .. .... . . n ... ... ... . .... ...N... r .. ;97G..=::...;_ ?=7177I7Ge7e77'C ;g7;7 : g 77.. .ú Y... ..7... ...7. a ..1"" 7 C LC C"C 1:,;..L...; y...C.;..;.:;....7N...7 :...N77L...n...g7...g ;uunmul 7un;. ymu .7. ú77!lE1 .. .. ;. ;uar n.1u

311=a=1:aEE1Ñ1::::E11=T1 ; ......E9:91é " g11e1EE1EE1.1=E1eae17 91,.P`717úú:E11NN gY.gmmaaa'ag1a1E11g331/11111.;.. .Ç