THE POSSIBLE NUTRITIONALEDICINAL VALUE OF SOME TERMITE MOUNDS USED BY ABORIGINAL COMMUNITIES OF NAUIYU NAMBIYU (DALY RIVER) ELLIO OF THE NORTHERN TERRITORY, WITH EMPHASIS ON MINERAL ELEMENTS. A thesis submitted for the deee of Master of Science the University of Queensland by FRANCOISE L. FOTI lngenieur Agronome (Nutrition and dietetique) Universite Catholique de Louvain-la-Neuve (BELGIUM) School of Chemisy d Eth Sciences Northe Ter ritory University Din, NT Mch 1994
323
Embed
THE POSSIBLE NUTRITIONAL/MEDICINAL VALUE OF SOME …espace.cdu.edu.au/eserv/cdu:34579/Thesis_CDU_34579_Foti_F.pdf · Nadia, Y asmin and Joshua and Hubby Tony, With Love •••
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
THE POSSIBLE NUTRITIONAL/MEDICINAL VALUE OF SOME
TERMITE MOUNDS USED BY ABORIGINAL COMMUNITIES OF
NAUIYU NAMBIYU (DALY RIVER) AND ELLIOTT OF THE
NORTHERN TERRITORY,
WITH EMPHASIS ON MINERAL ELEMENTS.
A thesis submitted for the degree of
Master of Science
in the
University of Queensland
by
FRANCOISE L. FOTI
lngenieur Agronome (Nutrition and dietetique)
Universite Catholique de Louvain-la-Neuve (BELGIUM)
School of Chemistry and Earth Sciences
Northern Territory University
Darwin, NT
March 1994
DECLARATION
The work presented in this thesis is, to the best of my knowledge and belief, original,
except as acknowledged in the text, and that the material has not been submitted,
either in whole or in part, for a degree at this or any other university
Eastwell (1979)" mentioned that one bush group of 40 people consumed 2 kg of termite
mound during April 1973. Eastwell (1984)44 reported that "people who live in the out
stations still eat some ant-hill with some of their meals, or have some clay"44 and that
"teenagers are seen carrying clay in plastic bags to eat at the outdoor cinema1143•
Sometimes, in Daly River, when there is a flood that renders the termite mounds
inaccessible, the people would dive down to collect clay from a favoured place. Clay
and termite mounds seem interchangeable with preference for termitaria in some places.
Levitt (1980)100 wrote that clay processed by animals, such as on the termite mound, was
considered to be safer than other kinds. Where clay is eaten, it is used in much the
same way as termite mounds: to allay hunget2.17, to cure stomach aches, diarrhoea, to settle the stomach, to treat worm infestation or to "line the stomach before eating yams,
or fish which may be poisonous"11.
4
.. 0
2 11-0
"" .. c ~ 0
~
PLATE 2 Mercia (Wawurr) of lhe Moil people at \lauiyu -.;ambiyu (eating some Nasutitermes tnodioetennlto.ria) accompanaed by Molly (Y JWalminy) and Mnlcolm Kurruwul
5
Bateson and Lebroy (1978)17 reported that children ate clay because they liked the taste.
On Groote Eylandt, some women ate clay "particularly if they had a craving for fish as
they said it tasted like fish"100•
1.1.1.1 Modes of Preparation
There are many different ways in which the Aborigines use termite mounds. It seems
almost as if every family has its own "recipes" . The simplest way is to break off small
pieces of the outer mound, crumble it in the hand to a powder, then drop it into the
mouth (Nauiyu Nambiyu1\ Groote Eylandt100), (Plate 2). In other communities, a large
piece of mound (hand size) is ground finely and mixed with water, milk or tea, then
drunk.14•82 Sometimes, only the extracted liquid is drunk (Wave Hill, Wattie Creek)17.
Honey-ants (Melophorus sp.) or plants can be added to the mixture of tennitaria and
water16. In Elliott14, pieces of mounds are first burnt in the fire, until black, then crushed
and mixed with water. One part of the "infusion" is then drunk and the other part used
to bathe the skin or as a poultice.16 When used as a mosquito repellent, the inner part
of the mound is slowly burnt in the frre4.
In general, the mounds are consumed on the spot. In some places, such as Numbulwar,
the dry earth eaten was probably tennitaria brought from another area for the
construction of the airstrip17• On some occasions, for example, just prior to the flood
caused by the rains of the wet season in Daly River, a big lump of the mound can be
taken horne for further use.
The mound is not the only part eaten, Dulcie Levitt (1980i00 reported that the tennite
soil tunnels in logs are also used.
Since tennite mounds have been so widely used in Australia and that most soft-bodied
insects that were available in quantity were eaten190, it is surprising that there is no
mention of termites themselves being used as food. Likewise, despite the abundance of
tennitaria in the Amazon region, few records exist of Brazilian Indians eating termites110,
compared with Africa where they are a favourite dish164•
" 0 u .. .. 0.. <) ;-;-,� '-' " 0 ..::::
:::.
'-' 0 0 E .. » .. 0 0 0 "" e:.
6
PLATE 3
PLATE 4
Typical Nasutitemus triodiae mounds in Daly River.
Mercia (Wawurr). Molly (Yawalminy) and Patricia Marrfura Me Taggart of the Daly Rjver showing the newly built material, that they favoured.
at the base of a Nasutllem1es triodiae mound.
1.1.2 In Two NT Aboriginal Communities
7
The selection of the two communities (Nauiyu Nambiyu (Daly River) and Elliott) was
suggested by Andy Barr (the Aboriginal pharmacopoeia project manager) and the late
Mrs Joan Chapman (the pharmacopoeia team pharmacist) who had worked closely with
the two communities and had received the approval by the two communities for further
study.
1.1.2.1 Nauiyu Nambiyu (Daly River)
The first contact with the Nauiyu Nambiyu community took place in June 1988. A
meeting was organised with Patricia Marrfurra Me Taggart of the Moil people at Nauiyu
Narubiyu (coordinator at the MajelJan Centre), her mother Moliy (Yawalminy), her
·grand-mother Mercia (Wawurr), her two children: Nathan (Culmungu) and Aaron
(Kingaruo), and Eileen Farrelly (adult educator of the Majellan Centre).
The information concerning the use of termitaria was gathered after a few subsequent
visits and field trips (August-October-November 1988, December 1988 and 1989, August
1990 and 1991), often also including members of the extended family.
1.1.2.1.1 Choice of Termitaria
Mercia and her family use three different types of mounds. They belong to:
'Nasutitermes triodiae, Tumulitermes pastinator and Amitermes vitiosus. Where
Nasutitermes triodiae is present, Amitermes vitiosus is not taken. The red mounds may
be thought to be more effective16• Coptotermes acinaciformis mounds have also been
reported by the pharmacopoeia team as being used in Daly river. Out of al� the mounds,
they prefer the freshly built parts of Nasutitermes triodiae mounds (Plates 3 & 4).
11Women like the taste of it" (Mercia). Just seeing the freshly built parts give them
watery mouth.
8
PLATE 5
PLATE G
A sample of Nasutitermes triodiae mound ready to be taken home (Daly River).
Quantity of termitaria Mercia (Wawurr) and Molly (Yawalminy) used to
take 2 or 3 times a day during thdr pn:gnancy (Daly River)
9
If the mound is attacked by red ants they would not eat it because it would be too old.
The 11old ones (are) not too sweet" said Mercia. Sometimes, they take "a big mob
home" for further consumption (Plate 5). They throw away the grass and termites by
shaking the portion taken and let it "dry" in the sun before putting it into their bag.
When tennitaria is not available (eg: during the flood of the wet season) they dig for a
clay (sub-soil) close to the mission.
1.1.2.1.2 Usage
Mercia said that they used to eat it 11more before the mission, now they forget about it".
She used to eat a handful (20 to 30 g) (Plate 6) at least 2-3 times a day when she was
pregnant or had diarrhoea and used to give termite mound pieces even to young children
with diarrhoea. She uses the clay in much the same way. Molly ate termitaria material
for "upset stomach, vomiting and diarrhoea" or when she was pregnant to stop the
craving. Molly used to eat "bits and pieces all day long11 during pregnancy. She
reported that they also used to eat it when they were hungry, when they had stayed in
the bush for a long time, or after eating yam or meat (turtle, wallaby, goanna and
porcupine). She mentioned that men eat termite mounds too. Patricia reported eating
termite mound when she was a child.
1.1.2.1.3 Mode of Preparation and Consumption
Small pieces of the outer casing are broken off, crushed between the hands and placed
on the middle of the tongue where they are left to melt before being swallowed.
10
PLATE 7 Lucy Hughes (Lababi) from Elliott, crushing someAmitermes vltiosus mounds during the preparation of the "infusion".
PLATE 8 In Elliott, Lucy Hughes demonstrates the poultice while Amy Lauder is getting ready to drink the "infusion.
11
1.1.2.2 Elliott
The first contact with the community took place in September 1988. As for the Nauiyu
Nambiyu community the information was gathered after a few meetings and field trips
with the Aboriginal informers (Amy Lauder, Molly Dixon and Lucy Hughes (Lababi)).
A) Choice of Termitaria
Amitermes vitiosus mounds are the dominant mounds in the landscape and are used by
Aboriginal community for medicinal purposes. Their local name is: "Bilaya".
B) Usage
It is used for diarrhoea, upset stomach, during pregnancy or to bring up milk after birth
and to "make the baby strong".
C) Mode of Preparation and Consumption
The mound is first pushed to the ground and broken into big pieces with an axe. Then,
the pieces are placed in the fire. They remain there for 15 a2Q minutes before being
placed in a billycan, half-full with water. The pieces are crushed to mud with a piece
of stick (Plate 7) and left there to infuse for I 0 minutes . The "infusion" is mixed well
and drunk hot. The remaining mixture of muddy composition is rubbed on the back and
the front of the person (baby and lactating mother), (Plate 8).
12
1.1.3 Other Countries
Native populations throughout the tropical and sub-tropical regions of the world have
been using termite mounds for many different purposes. As previously discussed in
chapter 1.1, only the physical aspects of their use have been considered.
Other than being used for pottery68•86•164 or to build roads86•143•68, tennis courts86•143,
housesw3•86•42•15 and ovens103•42•15, termite mounds have been widely used in agriculture
around the world (see chapter 1.4.2.5) and in geochemical explorations137• Termite
mounds are also an important source of nutrition for many species of animals. In
Northern Ghana, the indigenous population use Trinervitermes geminatus mounds as a
source of chicken food157• Pullan (1974)147 reported the use of termite mounds for
nutrient supply, for example, elephants may excavate tennitaria200 and eat the earth, and
antelopes may use natural excavations of termite mounds as salt licks. In Guinea
(Conakry), tennite mound excavations containing large white nodules (most probably
calcium carbonate) are given to the cattle as a natural mineral supplement by the
nomadic breeders around th� country (Personal communication 1981). Those "earths"
were also fed to mineral deficient sows at the University farm of Macenta (Guinea)
(personal observation 1981 ). Monkeys have also been reported eating termite
mounds75•74•40•210•73• Red leaf monkeys (Presbytis rubicunda), in Sabah (Northern
Borneo), were observed eating Macrotermes mounds and chimpanzees were reported,
in Gabon and Tanzania, eating 10 to 20g of earth (most of them from tennite mounds),
up to twice daily74•
This phenomenon (the habit of eating earth) is more commonly known as geophagy.
It is a fonn of pica which is the craving and eating of non-natural, and inedible objects70,
It has been reported in the human population for many centuries45•187•47, It is reported
all over the world70: Saudi Arabia70, TurkeY3•114, Iran146•188•1S4,Js9, Algeria12, Nigeria, Togo,
Liberia and Ghana186•185 Zambia147 South Africa189 USA181•180'47•46•56•53 South America1 10 • • • • •
Haiti63, India86 and Philippines41• "Earth eating cross-cuts ethnic, social, and economic
lines. In the United States, geophagy is found among whites and blacks, children and
adults, and in both rural and urban populations. "187 Its prevalence varies greatly from
13
culture to culture66• It seems to be more prevalent in the black Americans47 and among
"poverty·stricken" populations where the nutrition is less than optirnum65 and in
women. 6s·46•147•66•47•45•147 For example: it occurs among 57% of women and 16% of
children of both sexes in the black population of rural Holmes County, Mississipi; the
average daily consumption is 50g187• And more particularly in women during pregnancy:
out of 42 women studied in Alabama who ate between 6-130g of clay daily, 38 were
pregnant46• In Ghana, although clay (from sub·soil or termitaria) is mostly eaten by
adult women (34 to 68%, with the highest percentage in more remote areas), it is also
consumed by 14% of the adult male population (in Eweland)'"'. In South Africa, clay
from a termite mound is very popular with rural black women; 44% of rural black
women experience geophagy during pregnancy as opposed to 4.4% in mixed-coloured
women and 2.2% in Indian women; while pica did occur among white pregnant women,
it was rare1119• Cavdar et al (1983)33 report that geophagy was a common finding among
Turkish children and women in villages.
In some small southern towns in the USA, clay can be purchased at commercial
outlets46• In African countries it can be bought in the local market where a particular
clay called eko or Calabar could be imported from a distant region (more than 1500 Km away)186• In Western countries, capsules of dirt and clay are being marketed in health
food stores. The origin of the material used is not always specified.
1.1.3.1 Usages
Cesare Bressa (early 1 800), cited in Mustacchi (1971)121 described that the slaves in
Louisiana started eating soil during their illnesses (known later on as wet beriberi or
vitamin Bl deficiency); most often they wanted to eat earth and some seemed to prefer
hard earth while others liked clay121• In India, Joseph (1978)86 reported that the material
of the termite mounds was often eaten by local tribes, presumably for their mineral salts
content. In Nigeria, they used clay in traditional medicinal preparations for intestinal
problems (stomach and dysenteric ailments) or problems associated with pregnancy186•
The reasons given by the pregnant women (USA) for eating clay are numerous: to
14
relieve nausea, prevent vomiting, relieve dizziness, relieve headaches and many reported
eating more clay when they were upset46• In the Amazon, termite mounds are prepared
as remedies for bronchitis, constipation, goitre, sores, boils, ulcers and other ailments110
1982• Termite mounds are also eaten when food is scarce1 10• The Uaica Indians chew the
soil of pulverised termitaria, saying that it is good for them, it helps strengthening and
building up the body; they like the taste and eat handfuls at a time110,
1.1.4 Possible Therapeutic Activities
The external uses of clay are perhaps more familiar; poultices containing clay are a
known remedy for boils, while mud baths are recommended for the treatment of
rheumatism and arthritis. The popular literature claims that the benefits of termite
mound and clay consumption are quite numerous. According to Dextreit (I 976)41 clay
digestive disorders including ulcers, enteritis, dysentery, constipation and diarrhoea and
many other ailments. Dextre�t (1976t1 claimed that clay can cure iron deficiency, not
because of a high mineral content, but because it contains catalysts that work in
infinitesimal doses to stimulate failing organs. The more clay is exposed to sun, air, and
rain, the more active it becomes41 •
The most obvious reason given by many authors for geophagy is because o f the potential
as nutrient sources, mainly during pregnancy. For example, in Zambia, pregnant women
use te�ite mounds probably as a source of calcium and sodium147• Barbier et a/
(1986Y2 suggested that malnutrition causing growth failure and zinc deficiency could be
the cause of geophagy12• Edwards et a/ (1959t6 noted that possibly clay-eaters eat clay
to compensate a diet that was poor in calcium and iron. This will be discussed in more
detail in section 1.1.6
In the treatment of stomach complaints, including diarrhoea, clay (from soil or
termitaria) may act as an adsorbent antidiarrhoeal and might help to alleviate digestive
disorders40• The mineralogical analyses of the eko clay (Nigeria) are strikingly similar
15
to the clay in the commercial pharmaceutical Kaopectate (kaolinic composition)m.
Kaolin is a powdered hydrated aluminium silicate freed from gritty particles171• It has
long been used for the treatment of gastric disorders (diarrhoea, dysentery) in both
traditional (China) and modem pharmacologies178• The clay absorbent power is quite
extraordinary, according to Dextreit (1976t1, 5g of clay are enough to completely
discolour 10 cm3 of a water solution containing 0.1% of methyl blue41•
An interesting case, demonstrating the absorbent power of the clay, was reported by
Halsted ( 1968)", in which a prisoner, in Germany (158 1), who was given 6g of mercuric
chloride followed by one tablet of clay ("terra sigillata in olde wine") and did not die.
Halsted (1968)" reported that it is quite possible the clay had acted as an ion-exchange
resin and that elements in the clay could have exchanged with mercury, making it
unavailable for absorption.
. Research with rats has shown that geophagy occurs when rats are made acutely ill
(poison) or are stressed or arthritic32• The sicker the rat, the more kaolin it eats. Kaolin
intake is also a behavioural index of motion sickness in rats179• Geophagy could be a
response to generalised stress, since this state may involve gastro-intestinal malaise32·m.
A study done in South Africa (where geophagy is more common among black pregnant
women), on the frequency and severity of nausea and vomiting during pregnancy
reported that it affected only 3 .8 and 3 .19% of the black women respectively, while it
affected 19.8 and 17.8% of white women respectively189
Davies and Baillie (1988)'to observing monkeys in North Borneo suggested that geophagy
may serve different functions at different times: it might help to absorb toxins, to alleviate digestive disorders or to supplement the mineral intake. This latter point was
disputed by Hladik (1977ar3 who stated that the amounts of minerals in the termite
mound sample were very small but for iron, and that the iron content of the leaves and
fruits eaten by the primates was enough to cover all their physiological requirements73•
16
There may also be other determinants such as psychological and social. For one thing,
home treatment (eg: preparation and administration of the clay-tennitaria in illness) may
strengthen family ties because care and attention are exchanged within the family group.
The patient is surrounded and taken care of by his own people who feel responsibility
for the family1s health174 but also 11Eating clay is psychologically comforting.
Masticating it is gratifying, salivary flow is increased, and the stomach is comfortably
julf'43• Likewise, the most common reason for eating clay given by males (Ghana) is
that it provides "pleasure11185
1.1.5 Possible Complications
The medical literature on geophagy is meagre and research reports are often conflicting
in their findings45• In Australia, it is not viewed as a serious medical problem by the
nursing staff of different Aboriginal communities ( eg: Maningrida, Angurugu }, although
it can cause constipation when used to excess69 According to Hausheld (1975t9 it is a
traditional practice more likely to be beneficial to health than dangerous to it69•
However, Bateson and Lebroy (1978}17 (Northern Territory) warned that eating clay or
termite mounds can be dangerous as it can cause a partial obstruction, or even
perforation, of the colon. Thomson (1984}'x1 reported that geophagy, especially if kaolin
is involved, may lead to a zinc deficiency. Eastwell (1984)44 reported that it could also
interfere with iron absorption.
Dirt eating was listed as a cause of death in 1850121, probably because it was consumed
by dying slaves in their attempt to compensate their malnutrition. Geophagy is also
associated with severe iron deficiency anaemia,s6•33•188'146,n4,154 in addition to zinc
deficiency33•66• A syndrome characterised by geophagy, iron deficiency anaemia, zinc
deficiency, growth retardation and hypogonadism has been observed in both sexes in
Iran146•154•159, and in Turkey33·m for several decades; this syndrome is probably due to the
poor nutritional status of the population but possibly increased by prolonged geophagy33•
Clay ingestion has also been associated with myositis (muscle weakness) and
hypokalemia (potassium deficiency) or hyperkalemia (excess potassiumY3•81• Halsted
17
(1970t6 noted that a case of tetanus linked to the ingestion of clay containing spores of
the tetanus bacillus has been reported66•
1.1.6 Geophagy and its Relation to Mineral Deficiency
The traditional medicinal uses of termite mounds, such as for gastro-enteric disorders
and during pregnancy would suggest that a number of elements may be of importance,
in particular minerals such as: calcium, iron, magnesium, potassium and sodium. The
older literature suggests that iron-deficiency lead to geophagy. Halsted (1968)65 reported
that although it is an attractive theory, the literature evidence for anaemia resulting in
geophagy is meagre187 and the results are often conflicting. Barbier et al (1986)12
suggest that although geophagy was probably a spontaneous effort to compensate for
malnutrition (growth failure, zinc deficiency), it was also probably the cause of iron
deficiency anaernia12• This idea is supported by Cavdar er al (1983}33 who suggested
that the zinc deficiency could have been present before geophagy started. Indeed, the
nutritional status of Turkish villagers was poor: based on wheat bread and wheat product
with very limited animal protein. It had little zinc content and the zinc could have been
bound by phytate (inositol hexaphosphate) and fibres"'. But they also noted that the
geophagy may cause both iron and zinc deficiency: by reducing the appetite for normal
food, by decreasing the iron absorption and by causing malabsorption of zinc and iron
due to irreversible changes in intestinal epithelial cells associated with prolonged
geophagy. This same view is also supported by Walker et a/ (1985)189 who suggested
that "In the case of geophagy, while the practice may contribute significant amounts of
calcium and trace elements, under certain conditions it may exacerbate an existing iron
deficiency anaemia. The extent to which iron deficiency causes geophagy, or geophagy
promotes iron deficiency is not wholly clear ... ".
Different experiments have produced different results. For example, Patterson and
Staszak (1977)141 reported maternal anaemia and reduction in the birth weight of the
neonatal rats born to kaolin (20%) fed rats while the results of kaolin fed rats receiving
iron supplement were normal. Edwards et a/ (1983ts reported that small levels of clay
18
conswnption (20%) may be beneficial as it enhances physical development in rats. The
haemoglobin values of baby rats from female rats receiving 20% and 35 % of clay
through their diet were not different from control values45•
Eastwell (1979)43 studying Aboriginal clay eaters found no difference in the haemoglobin
levels of ingesters as compared with controls. Edwards et a/ (1964)47 reported that
although the total iron and calcium content of the clay studied was respectively:
20.8mg/100g and 21.4mgil 00g, only 0.03mgil00g ofiron and 0.2 mgilOOg of calcium
were "available" in the in vitro tests. Likewise, Venneer (1971)185 found very little
"available" minerals (0.1 N HCl extraction) in Ghanian clay. But he commented that
"It is possible, ... that even such small amounts may contribute to the overall nutrition
if that element is deficient in the body and if other dietary inputs is inadequate"m.
Although the ingestion of clay may slow down the mobility in the gastro-intestinal tract
and thus promoting absorption, its consumption may also impair the absorption of certain
other nutritive elements through chemical exchange185• Talkington et a/ (1970Y80
reported that while the ingestion of sizeable amounts of clays, just prior to iron intake,
did not appreciably reduce iron absorption; a red clay containing considerable iron
proved inefficient for correcting iron-deficiency anaemia while the admission of a
smaller quantity of ferrous sulfate did. They commented on the fact that iron absorption
in the same individual can vary appreciably, at different times, unrelated to the agent
being tested. They suggested that the degree of iron absorption reduction caused by clay
could vary widely, apparently depending upon the clay.
In their study on the effect of clays of different origins on 59FeS04 absorption, Minnie.
et a/ (1968)u4 reported that Turkish clay and soil reduced the amount of 59FeS04
absorbed from the intestinal tract while three other clays obtained in the United States
were less effective. The last clay tested was from New Mexico and it had no effect
upon iron absorption. Interestingly, a dose of MgO completely blocked radioiron
absorption in four of five subjects and reduced absorption significantly in the fifth.
Turkish clay and soil also removed iron from solution better than did the naturally acid
clays. The mechanism which most likely accounts for the observed effect is the cation
exchange capacity and the base saturation value (which correlate with the pH114). In the
19
study by Minnie et a/ (1968)1" the clay differed in effect depending upon their cation
exchange capacity (CEC). Turkish clay having a high CEC was more effective in
blocking iron absorption than were three other clays with lower CEC values. The iron
is exchanged for Ca, Mg, Mn, Na, K, and H ions with the formation of non-absorbable
iron compounds. They commented that the effect of clay and soil on iron absorption
may not be the only factor in the production of anaemia in geophagy, but it could be
contributory. Nutritional and parasitic (worms) factors may contribute to anaemia as
wel1114•
Halsted (1968)65 suggested that as geophagy leads to iron deficiency, it may also prevent
absorption of potassium and mercury and possibly zinc, as they have demonstrated a
high cation-exchange capacity for zinc by Iranian clay, in their in-vitro tests.
In the case of hypokalemia, some in-vitro tests have been conducted by Gonzalezet a! .
{1982)56 showing a moderate potassium-binding capacity of clay especially at pH 6.
They suggested that clay eaters could develop hypokalemia depending upon the type of
clay ingested, the daily dietary potassium intake, and the renal function status.
Severance et a/ (1988)162 reported that the potassium level returned to normal when the
clay ingestion was discontinued and potassium supplement was given162• Gelfandet a/
(1975)53, analysing hyperkalemia in five patients with chronic renal failure, suggested
that since river-bed clay contains as much as 100 meq of potassium in I 00 g of clay
(much of which is exchangeable at acid pH), hyperkalemia appears to be the result of
the absorption of potassium released from clay after ingestion. As in the hypokalemia
case, the hyperkalemia ceased to be a problem when the patients stopped eating clay53•
One may ask, after reading the medical literature, if the clay (and more specifically the
clay as in termitaria) is the cause of certain mineral deficiencies or is it the cure? Most
probably it will depend on the physiological state of the individual, but the composition
of the termite mounds (or clay) eaten is of prime importance. Most of the research to
date on human geophagy has concentrated on clays or earths. The mineral content of
the clays studied, as indicated by the literature review is generally very low in
comparison to the high mineral content of termite mounds.
20
1.2 Nutritional Aspects of Selected Elements and Recommended Dietary Intakes
(RDis).
The traditional medicinal uses of termite mounds, such as during pregnancy and for
gastro-enteric disorders, would suggest, as previous authors have already mentioned
(chapter 1 . 1 .4), that elements such as Ca, Cu, Fe, Mg, K, Na and Zn may be of some
importance. In order to determine the nutritional contribution of termite mounds towards
the human nutritional needs, a summary of the selected elements nutritional
characteristics is a prerequisite.
The selected elements studied in this thesis have been classified in human nutrition as:
- electrolytes (sodium, potassium),
- major minerals (calcium and magnesium),
- trace elements (cobalt, copper, iron, manganese and zinc),
- substances with no known essential nutrient function in man
(aluminium).
A summary of the selected element (calcium, cobalt, copper, iron, potassium,
magnesium, manganese, sodium and zinc) body contents, physiological functions,
Recommended Daily Intakes (RDis), sources and deficiency symptoms is given m
Table 1.2. In this study, mineral and element are used interchangeably.
The RDis have been defmed as "the levels of intake of essential nutrients considered,
in the judgement of the National Health and Medical Research Council, on the basis of
available scientific knowledge to be adequate to meet the known rmtritional needs of
practically all healthy people. "124 The human element requirement is difficult to
establish as it varies between individuals and changes according to age, environment and
physical condition. This has led to differences between various national and
international recommendations. However, these differences are, however narrowing as
more knowledge accumulates169• In Table 1.2, the Australian RDis have been chosen
21
whenever available; the American values were selected for cobalt, copper and
manganese.
The RDis should not be misinterpreted as a daily minimum or as a requirement for any
specific individual. The RDis exceed the actual nutrient requirements of practically all
healthy persons as the estimates of requirements for each age/sex category have been
increased by a generous factor to take into consideration the variations in absorption and
metabolism124• As discussed by Southgate et a! (1989Y69, the mineral nutritiona! value
of food and diets is not necessarily equal to their composition as determined by chemical
analyses. The intestinal absorption and subsequent metabolism of all the elements needs
to be considered. Only a proportion of the total ingested element is capable of being
used. The amount of element absorbed depends not only on the chemical form of the
mineral in the food but on the other ingredients in that food and of the rest of the diet
and also on physiological factors169• Certain food in the diet may enhance absorption
of a particular element and decrease another, for example: ascorbic acid and protein
from meat decrease copper absorption and increase iron absorption. The problems of
minerals are multiplied by their own interactions. For example, an excess of zinc, iron
or calcium decrease the absorption of copper (an increase of 5 to 20 mg of zinc results
in an increase of the copper need of 75%)19,
Iron deficiency is the most common nutrient deficiency disorder in the world125 and
pregnant women subgroup is one of the most at risk categories. The incidence of iron
deficiency anaemia among pregnant women varies from I 0% in adequately nourished
groups to 50% in poorly nourished groups with multiple, closely spaced pregnancy183.
The iron deficiency results from one or a combination of the following: inadequate diet,
impaired absorption, blood loss or repeated pregnancies20• The iron is absorbed in the
duodenum and upper jejunum through a complex but poorly understood process183• The
factors which determine the proportion of iron absorbed from food are complex. They
include the iron status of an individual, the iron content and the composition of a meal.
Only a small proportion of dietary iron is absorbed; normal subjects commonly absorb
5-10% of the iron of mixed diets, and iron-deficient individuals 15-20% or more of this
iron, but considerable divergence from these values can occurm. The iron absorption
22
is also markedly increased during pregnancy, being about 30% in the second trimester
and 40% in the third trimester183• The composition of a meal also determines what
proportion of its content can be absorbedm. The dietary factors enhancing the iron
absorption are : ascorbic acid, citric acid, meat, fish and alcohol while the inhibiting
factors are polyphenols (such as tannins), phosphates, bran, phytate (in cereals and
legumes) and cooked egg yolk169• All meat sources promote the absorption of iron from
other foods. Even relatively small quantities of meat and fish (50-75g) may markedly
improve the hie-availability of iron124• Concomitant doses of 200 mg of ascorbic acid
with iron salts may increase absorption by 25 to 50%119•
Adult males and post-menopausal women measured iron losses are about I mg/day
(Table 1.2). Additional iron losses associated with menstruation vary, and for 90% of
women it averaged at 1 .35 mg/day124• During the second and third trimesters of
pregnancy, 5-7 mg/day extra iron is necessary to provide for the large increase in the
blood volume of the mother and the foetal growth. In Australia, the composition of the
diet (particularly its content of vitamin C and meat protein) suggest an iron absorption
rate of 15-20%, and therefor� the minimum dietary iron intake necessary to meet the
physiological requirements of adult males and post-menopausal women is 7 mg. For
menstruating women, 12-16 mg/day are necessary.
All elements are potentially toxic in large dose. The RDis, even though known to be
excessive for at least 80% of the population, are also known to be safe for 100% of the
healthy population. Data which would permit the delineation of toxic levels of dietary
elements are meagre. The toxicity level of an element depends on the extent to which
other elements which affect their absorption and retention are present. This is
particularly true for copper. In animals, a particular level of copper intake can lead
either to copper deficiency or toxicity, depending on the relative intakes of molybdenum
and sulfur, or of zinc and iron184• Toxicity iron ingestion is the fourth most frequent
cause of poisoning of children in United States. The average toxic dose being 200 to
250 mg!Kg; an acute toxic dose may be as low as 150 mg!Kgm.
23
Aluminium has been selected not for its potential nutritive aspect but for its possible
toxicity. Aluminium has been considered to be a benign metal by many authors like
Davidson et a! (1973)39• They claimed that it is too insoluble in its natural form to be
absorb by the body and this very property gave them a use in human medicine. For
example, aluminium silicate (kaolin) is widely used as an absorbent in the treatment of
diarrhoea. In 1970, Berlyne et a/, cited in Kundu (1990)" reported that high levels of
aluminium in tap water used for renal dialysis equipment could be linked to a form of
dementia in patients who were undergoing treatment. Since then, aluminium as been
assodated with a variety of metabolic and neural dysfunctions. The extent of absorption
of aluminium from clay, however small, could be of importance. As it is believed that
a relatively small amount of aluminium enters through the mucosal lining of the mouth
and the gastro-intestinal wall. The great majority is excreted unabsorbed92. Brown
(1983), as reported by Kundu (!990)" has shown that while increased amounts of
aluminium associated with reduced pH is certainly toxic, small amount of calcium may
prevent the toxicity.
TABLE 1.2 Nutritional aspects ofselected elements: adult body content, physiological functions, deficiency symptoms, daily losses, o/o absorption from food, recommended dietary intakes (R.Dis) and sources.
Element
Cobalt
Copper
Calcium
lwn
Magnesium
Manganese
Potassium
Sodium
Zinc
Body content and primary concentration
1.1 mg Liver
80-120 mg Liver, brain, heart. kidneys 20.7 to 24.8 g per Kg of fat free body tissue1 99% of Ca deposited in bones
4-5 g1u 70% in haemoglobin, 25% in storage form (ferritin, hemosiderin) Liver, spleen, bone marrow
20-28 g165 55% in bone, 27% in musculature 12-20 mg Kidneys, pancreas, liver
J I .S to 131 g1SO Principal intracellular cation 83-97 giSO major cation in extracellular fluid
1-2.3 g in almost aU tissue, muscle 63%, bone 20o/o, blood 2%1n
Data from 601 otherwise specihed.
Physiological functions Deficiency Symptoms
Integral part of vitamin 812 Pernicious anaemia
Haemoglobin synthesis, bone Anaemia, growth retardation. mineralisation, enzyme function secondary iron deficiency
Bones and teeth, blood Osteoporosis, poor coagulation, storage and release developme'nt of teeth and of hormones, activation of bones, delayed coagulation enzymes systems Transport and utilisation of Anaemia, disturbance of bone oxygen, component of energy marrow function transfer oxidase
Cofactor in enzyme reactions Neuromuscular disturbances, behaviour disturbances, cardiac disturbance
Bones structure, reproduction, Impaired growth, skeletal activator of enzyme function abnormalities, ataxia,
convulsions, vomiting Osmotic pressure Lethargy and tetany
Osmose regulation, water balance Dehydration, nausea, anorexia, fatigue, muscular cramps
Metalloenzyme function, protein Poor growth and sexual metabolism, lipid metabolism development, impaired wound
healing
Daily losses in mglday flo absorption]
[30"/o]
12�100-150 [20%]
124Men, pmw®: I Menstruating women: I.35 pregnancy: 5·,. [15-20%]
@: pmw post�menopausal women #: only for second and third tnmesters of pregnancy
Sources
Liver, meal Varies on soil content on which food is grown Nuts, shellfish, dried legumes, liver Milk. cheese, nuts, green leafY vegetables, bones
Liver, meat, oysters, nu"
Grains, fiuits, vegetables, nuts
Green leafY vegetables, nuts, grain, tea Fruits and vegetables
Sodium Chloride, meat, fish, milk eggs
Meat, liver, eggs, seafood
t
25
1.3 Some Aspect of Traditional Aboriginal Health Concept and Background
1.3.1 Traditional Aboriginal Health Concept
The concept of health and illness in the traditional Aboriginal world is quite different
in philosophy and practice to western medicine. The maintenance of health is tied to
spiritual, religious and social welfare123 " ••• rather than through adequate nutrition,
exercise and the maintenance of a hygienic environment1195• Serious illness and death
may be attributed either to sorcerers or to the effect, direct or mediated, of the breach
of a religious law or social norm. The most commonly postulated cause of illness is
sorcery153• Treatment with medicine is often secondary to the spiritual healing processes.
"However, in the case of minor ailments, from which the patient would be expected to
recover in any case, such as colds, gastric troubles, wounds or skin diseases, the
treatment would probably be with medicines only"114• Usually women are the herbal
medicine authorities in the community82•174, although every adult possess a
comprehensive knowledge of bush medicines123• The two methods of healing (spiritual
and medicinal) may be used in conjunction 174•
1.3.2 Aboriginal Health Background (Mineral Nutrition)
Franklin and White (199It9 reported that it is now generally accepted that the average
pre·colonial Aborigines were generally in good nutritional health. They were vigorous
people with few but robust children (a quarter of their children had died by the end of
their fifth year). Their life span was 40 years, with injury (including warfare and
murder) being the most frequent cause of death before disease. In general they ate well
and "their environment provided ... food comprising both protein and vegetable foods
with adequate vitamins and minerals. "49 "There was no evidence of rickets or other
nutritionally·related disease in traditional Aboriginal groups"95• "Early data from
recently nomadic Aboriginals generally indicated high haemoglobin concentrations for
both men and women"95• But "the rapid shift to a sedentary life on cattle stations, . . . ,
26
had a dramatic effect on the nutritional and health status of Aboriginal people"9s. Their
transitional diet indicated that sufficient iron was still usually provided but many diets
were inadequate in calcium, vitamin A and vitamin C95• T o·day, malnutrition appears
to be a persi_stent problem for many Aborigines112• "Aborigines and Torres Strait
Islanders comprise the least healthy identifiable sub-population in Australia . . . death
rates are up to four times higher, and life expectancy is up to 21 years less"182•
"Intestinal infections and infestations remain a major cause of Aboriginal ill-health and
of hospitalisation. Dia"hoea/ diseases were responsible for more than 10% of the
deaths of Aboriginal children aged less than 5 years in the NT (1979-1983}"182• There
has been an increase in anaemia, in more recent studies, mainly due to iron deficiency95,
" ... nutrients such as zinc, vitamin C, vitamin D, iron, and vitamin A, may be major
contributing factors to poor health in Aboriginal children"95• But, in other areas, Lee
(1991)95 reported a surprisingly large ranges of both red blood cell and serum thiamine
concentrations, including some high values on remote aboriginal communities.
1.4 Termitaria Biological Background
1.4.1 Taxonomy and General Biology of Termites.
27
Termites are polymorphic eusocial insects comprising the order Isoptera. They belong
to the superclass Hexapoda and the infraclass Pterygota (Figure 1.2). They are closely
related to the order Blattodea (cockroaches)89'91• This small group of primitive insects
contains about 2300 species world-wide with some 350 species in Australia192 and about
60 species in the Top End of Australia113• The lsoptera order has five families present
in Australia (Figure 1.3) with the Termitidae (higher termites) being the largest and most
recent52• Termites are one of the predominant groups of tropical invertebrates86,85• They
are found mainly in the tropical and subtropical regions of the world54•97•9, approximately
between 45"N and 45CS96, In Australia, termites are absent from certain types of soil207
and there is a very small number of species in rainforest areas 51•
Termites are popularly known as White ants'. This name was given to them by the
English in the West Indies and later used by early naturalists (eg: Sir Joseph Banks in his journal (1768-1771) cited in Watson and Gay (1983)"1• This taxonomic
misconception is still used today and often people confuse termites {lsoptera) with ants
(Hymenoptera).
Their physical and social characteristics are among the factOrs that influence the features
of the colony. Termites are soft-bodied insects with cryptic habits. They are
hemimetabolous (have no pupa stage) and they live in family groups (colonies) with
polymorphic castes. Three major castes are recognised: reproductives, workers and
soldiers (shown in Figure 1 .4) The ratio of the castes may vary with time; it is kept in
control through pheromones, hormones and selective cannibalism176•
28
Phylum ARTHROPODA
I Sub-Phylum ANTENNATA
I Super-Class HEXAPODA
I Class INSECTA
I Sub-Class D!CONDILIA
I CERCOFILATA
I Infra-Class PTERYGOTA
I NEOPTERA
I DICTYOPTERA
Order I BLATTODEA I I ISOPTERA
FIGURE 1.2 Classification of Isoptera (redraw from Kristensen, 199 191).
29
I ORDER ISOPTERA I I
I I Primitive Recent
(no worker caste, (worker caste) pseudergates)
I I RlllNOTERMITIDAE TERMITIDAE Coptotermes Amitermes
Nasutitermes Tumu/itermes
IMASTOTERMITIDAEI I TERMOPSIDAE I KALOTERMITIDAE I -
FIGURE 1 .3 Tennite Families (redraw from Hadlington, 198764)
The primary reproductives , or kings and queens, are derived from the alates151• In all
but one species, Mastotermes darwiniensis, the fore and hind wings are similar, therefore
their name !so( same) ptera(wings). After the nuptial flight their wings are shed and only
some small scales remain (Figure 1.5). The alates are fully matured insects with
compound eyes, hard pigmented cuticle and gonads. Their function in the colony is
reproduction. In many species the queen' s abdomen becomes distended due, mainly, to
the enlargement of the ovaries. This phenomenon is known as physogastry (Figure 1 .6).
The queen can lay up to 2000-3000 eggs per day192 and can measure up to 12 cm106•
Both king and queen live for many years, often over twenty. In some species, the
colony may survive the death of their progenitors by producing supplementary
reproductives or neotenics192•
30
' I
\\\ ,\,
- .'\' :;··· . ·- _:· I ' . \ ;''' . ··/ l,i' .;,.;,_� � y
\\ - I ' !'
. .
B
2·5.m m
c
'�
FIGURE 1.4
Castes of Coptotermes acinaciformis, A, winged reproductive or alate; B, worker; C, soldier. (from Watson & Gay, 1991 192 )
TABLE 1.3 (conti.) Austalian termite mound and adjacent soil physical properties Location Termite species Type of material Grace I C. Sand F Sand Silt Clay
& & % Reference Position Texture (excluding gravel}, in %
*A: Lee & Wood (197lb)"' n= number of data set from literature values *B: Okello-Oloya et al (1985) #: n= 6 *C: Holt & Coventry (1981) + Coventry el al. (1988) 11#; � 18 •n: Birkm (1985) OM: Organic matter
TABLE 1.5 Termite mound and soil chemical data I (Lee and Wood, 1971 b)98• II samples containing high level of organic matter pretreated with peroxide
Location Termite species I Type of material
NT Berrimah
NT Howard Spring
NT S.Port Darwin
NT Daly River
NT Pine Creek
Soil
Mastolermes darwiniensis Carton in wooden build. #
Amitermes sp. Galleries und. bark #
Coptotermes acinaciformis M. outer casing
Carton from mound # Microcerotermes nervosus whole mound #
soil soil: A (0�25)
soil: 0(60-75)
Amitermes meridionalis
soil
Amilermes meridionalis
upper mound galleries
M. center
beneath mound
upper mound galleries
M. center
soil beneath mound
Microcerotermes nervosus Whole mound # soil A I I (0-10)
TABLE 1.5 (conti.) Lee and Wood (l97lb) termite mound and soil chemical data
II samples containing high level of organic matter pretreated with peroxide
Location Termite species/ Type of materiaJ
Soil
pH Org. C K,O
total
K
HCI
gi!OOg mgltOOg mgllOOg
NT Larrimalt
NT Tennant Creek
QLD Mareeba I
QLD Mareeba 2
QLD Mareeba 3
Tumulitermes hastilis
Amitermes vitiosus
soil
Drepanotermes rubriceps
soil
Mound galleries
Mound galleries
Mound base
AI (0·6)
B (25-30)
Mound external wall
Mound internal
A (0-to)
B (20-30)
Tumulitermes coma/us Mudgut in dead tree # Amilermes laurensis Mound
Tumulitermes pastinator Mound outer galleries
Nursery from mound # Nasutilermes triodiae Mound outer galleries
Drepanotermes rubriceps Mound outer galleries
soil A (0-8)
B (16-24)
Amitermes laurensis Mound outer galleries
core of mound
Nasutitermes triodiae Mound outer galleries
basal region of mound
soil A (0-20)
B (45-60)
Coptotermes acinaciformis M. outer casing
Carton from mound # G. within wood
Schedorhinotermes int.act G. over log
5.3
5.6
6.2
5.9
6.0
5.6
B 6.4
6.5
4.5
5.2
B 5.2
S.l
5.2
B 65
7.1
5.5
6.1
6.0
6.6
6.8
5.7
3.1
5.3
5.8
2.4
2.2
0.8
1.1
0.3
1.4
1.1
0.5
0.2
18.0
2.9
3.7
1 1.0,
4.4
1.6
0.8
0.4
2.1
1.3
2.0
4.0
0.5
0.3
2.7
44.0
2.8
7.4
370
350
350
490
370
1340
1310
1360
1430
4580
4340
4530
4840
5720
4790
1400
1420
1380
1440
3110
1510
220
180
260
74
64
53
61
64
170
170
130
160
260
180
250
270
310
260
100
380
120
98
140
150
38
160
100
54
94
97
c. HCI
mg!IOOg
60
80
50
60
30
130
90
70
40
170
60
90
260
70
30
<10
<10
100
70
50
160
30
20
90
300
40
210
� P Exc.Ca Exc.K Exc.Mg Exc.Na
HCI
mg!IOOg mgllOOg mgiiOOg mg/IOOg
13
14
13
14
13
I I
I I
10
8
1 8
I I
I I
22
10
8
8
8
16
14
12
12
8
20
10
20
9
I l l
64.1
62.1
38.1
36.1
26.1
114.2
92.2
60.1
42.1
50.1
84.2
68.1
24.0
2.0
2.0
1 14.2
64.1
88.2
142.3
30.1
38.1
164.3
104.2
240.5
15.2
12.5
9.4
6.6
6.6
19.6
23.5
14.5
11.7
5.9
11.3
7.4
5.1
2.0
2.7
16.4
10.2
23.9
46.9
5.5
11.3
39.1
34.4
32.1
19.5
26.8
14.6
12.2
12.2
21.9
10.9
10.9
9.7
34.0
64.4
64.4
55.9
8.5
79.0
32.8
12.2
31.6
47.4
4.9
18.2
42.6
36.5
34.0
mgiiOOg
0.7
1.1
2.5
1.4
0.7
1.1
1.8
0.7
0.5
4.6
25.3
17.9
9.0
1.8
10.6
3.7
1.6
1.4
3.4
0.7
0.9
3.4
2.8
3.9
QLD Atherton
QLD Townsville
QLD Warwick
NSW Canberra
NSW Tallangatta
Nas:utitermes triodiae
soil
Nasutitermes magrws
soil
Amitermes laurensis
Coptotermes acinaciformis
Nasulilermes longipennis
soil
Nasutitermes magnus
soil
Nasutitermes exitiosus
Coptotermes factus
soil
Coptotermes factus
Mound outer galleries A (0-10) B (30-50) Mound outer galleries Nursery of mound # A (0-15) n (25-35) Mound outer galleries core of mound M. outer casing Carton from mound # Mound A (0-8) B (16-24) Mound outer galleries Nursery from mound # AI (0-20) A2 (40-60) outer soil cap Carton from wall # Nursery from mound # outer soil cap Carton outer wall # Carton inner wall # Nursery from mound # AIA2 (0-20) Al (25-45) D (55-70) outer soil cap Carton from wall # Carton inner wall # Nursery from mound #
determine the particle size fractionation of termite mound material
to assess the selected elemental variations:
I) between different age material (new and old)
2) within mounds according to the sample position: top, middle and bottom
3) between mounds of the same species (and different species) in a same
location
4) between mounds of the same species (and different species) in different
locations
determine the bio-availability (in vitro) of the selected elements studied (with
emphasis on iron bio-availability)
CHAPTER TWO
MATERIAL AND METHODS
•
2 MATERIAL AND METHODS
2.1 Collection of Termitaria
71
The method of sampling termitaria was designed as a result of consultations with
Aboriginal communities (section: 1 . 1.2.1 and 1 .1 .2.2).
2.1.2 Method of Collection
In Daly River, the surface samples (O-J em) were taken at random on the outside of the
tennite mound (on both new and old parts of the mound) using a stainless-steel knife.
The deeper samples (0-lOcm) were taken, randomly, at different height on the outside
of the motmd with a I 0 em core fixed on a cordless drill. Core samples were also taken
from the middle section of two mounds (Nasutitermes triodiae and Tumulitermes
pastinator). A minimum of 200-250g of sample was collected when possible.
In Elliott, the samples (approximately 10cm3) were cut using an axe, as used by the
Aboriginal people. All samples were stored in paper soil bags.
Ten centimetre cores of soil were collected, at each site, using a l Ocm auger. Triplicate
cores (50cm apart) were obtained for each sample site and bulked.
2.1.2 Site Locations
The samples were collected at 4 geographically different localities: Berrimah. Howard
Springs, Daly River and Elliott.
Geographical references concerning the different sites are given in Table 2.1.
;::! TABLE 2.1 Site locations
Site Termite species Location Longitude Latitude Grid reference Notes collected • (map)
Daly River (site I) Tp,l11 5070 Daly River 130" 43' 13" 44' 52LFK865803 5.2 Km on the right when coming from the Daly River mission
Daly River (site 2) Av 5070 Daly River 130" 42' 13' 45' 52LFK847785 3.4 Km on the left when coming from the Daly River mission
Daly River (site 3) Nt,Tp 5070 Daly River 130" 44' 13' 39' 52LFK883897 20.2 Km from the Police Station on the Daly River Road to Darwin
Daly River (site 4) Nt,Av,Ca 5070 Daly River 130' 49' 13" 3 1 ' 52LFL961 040 2.9 Km after Lichfield road when coming from the mission; 64.6 Km from Tipperary
Elliott (site 5) Av SE 53-5 N.waters 133' 30' 17" 35' 53KLA4055 8.4 Km on the left side of the road leading to Lake Woods and the Longreacb Waterhole
Howard Springs (site 6) Nt,Tp 5073-2 Darwin 130" 59' 12" 28' 52LGMI58207 0.5 Km after the Yarrawonga Zoo, on the left when coming from Darwin
Berrimah (site 7) Nl 5073-2 Danvin 130" 55' 12" 28' 52LGM093213 I .8Km before East Ann Settlement, on the left side of the road coming from Darwin
D : Indicates the geographic location (D: Daly River, E: Elliott, H: Howard Spring)
e : Site number in a specific location
An underlined sample is a collected on the outer surface of the mound (depth: 0-lcm).
An asterisk (*) at the end of a sample number indicates that the sample has been
collected on a newly built part of the mound by opposition to the "old'' material. Old
does not have a connotation of time (specific age)� it is a part of the mound material that
has been weathered. It is distinguished from the new built material by the different
texture and color.
Sample numbering example: Nt31D4* = Nasutitermes triodiae sample number 3 1 from
Daly River, site 4, collected on a new part on the outer surface of the mound (depth of
0-lcm).
2.1.3.1 Detailed Mound Study
One mound of each of the three major species used by the Aboriginal communities
(Nasutitermes triodiae, Tumulitermes pastinator andAmitermes vitiosus) was studied in
detail. The physical characteristics of the mounds are given in Table 2.2. A total of 20
samples were taken from Nasutitermes triodiae mound (Site 3), 17 from the
74
Tumulitermes pastinator mound (Site 3) and 9 from Amitermes vitiosus mound (site 5).
The different number of samples was related to the size of the mounds and the way the
Aboriginal people sample them. A detail of the 3 mound sampling is given in Table 2.3
and associated soil samples are given in Table 2.4.
TABLE 2.2 Physical characteristics of 3 termitaria selected for more detail sampling: Nasutitermes triodiae (Nt) from Daly River (site 3), Tumulitermes pastinator (Tp) from Daly River (site 3) and Amitermes vitiosus (Av) from Elliott (site 5).
Mound characteristics Termite species -
Nt Tp Av
Site number 3 3 5
Mound number I I I I
Basal circumference (em) 460 320 127
Middle height circum (em) 310 250 98
Top less lOcm circum (em) 72 130 53
Height (em) 310 70 64
Tota1 mound samples 20 17 9
Total soil samples 4 3 4
75
TABLE 2.3 Detail mound sample summary for 3 termitaria: Nasutitermes triodiae (Nt) from Daly River (site 3), Tumulitermes pastinator (Tp) from Daly River (site 3) and Amitermes vitiosus from Elliott (site 5).
Mound number/
0-lcm depth from outer casing
0-IOcm depth from outer casing
0-1 Ocm depth from middle section
site --------------------------'--_;_-
l / site 3
I / site 3
1 1 I site 5
Top Middle Bottom Top Middle Bottom Top Middle Bottom
NtOl "' Nt02
Nt04* Nt08* NtlO Ntl2 Nt05 Nt09 Ntll Ntl3
Nt03* Nt06*
Tp27',Tp28 Tp29,Tp30*
Tp31
Nt07
Tp32 Tp33 Tp34 Tp35
Tp36 Tp37
Ntl4 Ntl6 Ntl8 Ntl9 NtiS Nt17 Nt201
T38 Tp39
Tp40 Tp41 Tp42 Tp431
ns ns ns Av21 Av23 Av26 Av27 Av28 A29 Av22 Av24
Av25
•: newly built material 1: material from nursery ns: not sampled
TABLE 2.4 Soil samples collected at 0-!0cm depth in Daly River (site 3) and in Elliott (site 5).
Location
Daly River
Elliott
1-m away from any mounds
15D3, 16D3, 17D3, 18D3, 19D3
OlE, 02E
�1m away from any mounds
20D3, 21D3, 22D:l'
03E, 04E
76
2.1.3.2 Termitaria Sampling
In general three samples were taken per mound (Top, middle, bottom). If the mound
height was below !.2m, only 2 samples were taken (top and bottom). In the sample
summary tables (2.5, 2.6, 2.7, 2.8), the diameter + height gives an indication of the
relative volume of the mound. The circumference was taken at lm height for mounds
over l . m otherwise at half-height of the mound. All samples were 0-10 em depth,
unless indicated. The underlined samples were superficial samples (depth: 0-lcm). The
* indicates that the sample was taken on a newly built part.
A minimum of 3-4 soil samples were taken at each major site: Elliott (site 5), Howard
Springs (site 6), Berrimah (site 7) and Daly River (site I, site 2, site 3 and site 4). Each
sample was taken at a minimum distance of 1-m from any termite mound on the site and
at a depth of 0-lOcm. Another soil sample was also collected near site site 2 (50m on
the right to the Daly River mission) as it is favoured by Mercia' family during the
annual wet season flood, when the mounds are not available.
2.2 Sample Preparation
All the tennitaria and soil samples were initially crushed in plastic bags by gentle
rubbing with a piece of wood (the tennitaria often being in solid lumps). This allowed
the removal of the grass and tennites by sieving and shaking. The samples were then
81
dried, in soil paper bags, at 60° in an oven. The coarse sample was crushed using a soil
crusher (Conservation Commission Berrimah Soil laboratory) and sieved through a 2mm
sieve. A I OOg sub-sample of material was removed for physical analyses. The
remaining sample was pulverised using a ring grinder (<75 microns). A sub-sample of
the homogenous pulverised sample was taken for chemical analyses.
2.3 Particle-Size Analysis
The aim of the particle-size analysis was to subdivide the soil minerals into different
categories according to the particle diameter
clay <0.002 mm (<2 �m)
silt 0.002-0.02 rom (2-20 �m)
ftne sand 0.02-0.2 rom (20-200 �m)
coarse sand 0.2-2.0 mm (200-2000 �m)
The particle-size analysis method used was based on the Pipette and sieve method
described by Coventry and Fett (1979)37 in which sodium tripolyphosphate was replaced
by Calgon (hexametaphosphate) as dispersing agent. It was conducted at the NT Conservation Commission Berrimah soil laboratory. A chemical pretreatment of termite
mound samples rich in organic matter (>0.5%) with hydrogen peroxide was conducted
according to the method of Mcintyre and Loveday (1974)10s.
2.4 Acid Extraction of Termitaria and Soils
A number of digestion methods, using combinations of the acids HN03, HCl04 and
H2S04 were investigated. The aim was to select a method that would provide
reproducible results with high reproducible recovery rates. It was not necessary to
obtain a total extraction for this study. The elements analysed were aluminium, calcium,
cobalt, copper, iron, magnesium, manganese, potassium, sodium and zinc. The
82
determination of the element concentrations was performed using a Varian SpectrAA 40
atomic absorption spectrophotometer (AAS) (see 2.4.3).
2.4.1 Extraction Trials
The trials were conducted in order to select the most appropriate:
All glass-wear was washed with high purity water and detergent (decon) before being
placed in a detergent bath (2% decon) for a several hours. They were then washed three
times with high purity water, soaked overnight in 10% nitric acid, rinsed thoroughly
with high purity water (Pennutit), dried in an oven.
An homogenous pulverised sample ( 1 g) was weighed into clean 200 mL test tubes (in
triplicate). One mL nitric acid (AR) was added to the test tube. The tubes were covered
with plastic film and the mixture was allowed to stand overnight in a block digester at
room temperature. The following morning, 4 mL perchloric acid (Aristar grade) was
added to the mixture. Gradually, over a period of I hour, the temperature was increased
83
to a maximum of 180' C and maintained at 180' C for 3 hours. Triplicate blank and
2 reference samples (in duplicate) were carried out with each batch of 50 samples;
The digest was allowed to cool before bringing to volume (20mL) with high purity
water (permutit). The digest was mixed thoroughly using a vortex mixer before being
transferred into SOmL polypropylene centrifuge tubes and centrifuged for 10 minutes at
12,000 rpm in a Beckman model N' J2-21MIE centrifuge.
The centrifuged samples were filtered through Whatrnan 541 filter paper into 50 mL polyethylene bottles and stored in a fridge at 4'C prior to analysis.
Cu 324.754 0.100 - 1 .000 Na 589.592 1 .000 - 10.00
Fe 238.204 0.100 - 0.500 Zn 213.856 0.100 - 1.000
0.500 - 50.00
90
2.6.3.2 Analysis of Fe(II)
Prior to being analysed for the ionisable iron (Fe(II)), the sample solution was first
acidified with {6M) HCI, then the Fe(III) was complexed as (FeF6l by addition of
potassium fluoride (2M) to the sample solution, l 0 minutes before adding a-a'
bipyridine (0.25 % w/v) and ammonium acetate · acetic acid buffer (pH 4.5). The Fe{II)
was measured colorimetrically in pH 1.35 and pH 7.5 extracts at the absorbance of 523
run on a Perkin-Elmer 552 UV-Visible Spectrophotometer, using a lcm plastic cell, as
HF (etching acid) is present in the filtrate. The calibration curve was established using
a range of standards from 0.05 ppm iron(ll) to I 0 ppm.
2.6.4 Quality Assurance and Quality Control
The precision of the technique was established in the same way as for the
nitric/perchloric analyses. A termitaria reference sample Nl60 (< 2 nun fraction) was
chosen and subjected to the analysis 3 times in triplicates. Sample N160 was also run
with every batch of samples.
CHAPTER lliREE
RESULTS
3 RESULTS
3.1 Site and Mound Characterisations
91
Four species of termite mounds (Plates 9-16), including the three species that are used
by the Aboriginal communities of Daly River or Elliott, have been studied from 4
geographically different localities as described in section 2.1.2.
3.1.1 Site Characterisations
A typical site with Nasutitermes triodiae and Amitermes vitiosus mounds in Daly River
is shown in Plate 9. The type of vegetation is open woodland (site 4) to woodland
(Eucalyptus) in Daly River, Howard Springs and Bertimah and open grassland with
scattered trees in Elliott.
3.1.2 Termite Species and Mound Characterisations
Four termite species were collected and identified. The identification was based on the
size and physical structure of the soldier caste, the termite distribution and the type of
mound. The identifications were later confirmed by Leigh Miller from CSIRO Division
of Entomology, Canberra.
All the termites studied belong to the recent family: Termitidae (termites with worker
caste) and are grass-eating termites. Three belong to the sub-family Nasutitennitinae:
Nasutitermes triodiae (Froggatt) (Plates 10 and I I), Tumulitermes pastinator (Hill)
(Plates 12 and 13) and Tumulitermes hastilis (Froggatt) (Plate 14). They have nasute
head (Figure 3.1). Nasutitermes sp are distinctive with the soldier's head is not
constricted (Figure 3.1-A), while Tumulitermes sp have the soldier's head constricted
near its centre (Figure 3.1-B). Amitermes vitiosus Hill (Plates I S and 16) belongs to the
~
,....
>
--'
-rl
-c 3 0 :::
::l c.. .,.
~ ~ ~ ::- " " "'0 et " ~
l6
93
Amitermitinae subfamily. Amitermes sp have mandibulate heads with mandibles toothed
and sabre-shaped (Figure 3 . I -C).
The termitaria have characteristic features, although there is a degree of variation in
mound shape within species. A general representation of the size (height +
circumference) of the termitaria sampled at each site, is given in Figure 3.2. As seen
in Figure 3.2, Amitermes vitiosus and Tumulitermes hastilis have mounds that are small
and usually narrow, those of Tumulitermes pastinator are low but wide and those of
Nasutitermes triodiae are tall and large.
A. Nasutitermes sp
Nasute head
B. Tumulitermes sp
Mandibulate head
C. Amitermes sp
FIGURE 3.1 Dorsal view ofnasute head: A: Nasutitermes sp and B: Tumu/itermes sp and mandibulate head: C : Amitermes sp. (from Hadlington, 198764)
PLATE 10 Nasutitermes triodiae mound (3m height), Daly River, site 3 .
PLATE I I Vertical section of Nasutitermes triodiae mound at site 3, showing compact basal portion-galleries and nursery (middle part of the mound, ground level).
\C .a:..
1000
BOO
600 8
400
200
0
FIGURE 3.2
3.1.2.1
95
Av Nt
8 CIRCUM [l HEIGHT
z 4 5 . , 3 • 3 • • 7
SITE
Size comparison (height + circumference in em) of Amitermes vitiosus (Av), Tumulitermespastinator (Tp1 Nasutitermes triodiae (Nt) and Tumulitermes hasti/is (Th) mounds at different sites.
Nasutitermes triodiae (Froggatt)
These nasute soldiers (4.5 ± 0.25 mm) have a dark brown head extended into a thin
nasus (Figure 3.1-A) through which they can expel a sticky repellent secretion associated
with the defence against ants and other enemies. Nasutitermes triodiae occur widely in
Northern Australia. They are also known as "spinifex termites". They construct various
types of mounds that are the largest of any of the Australian species, reaching a height
of 6 metres. The mounds collected in Daly River are the "cathedral" type, (Plates 9, 10
and I I), they are only found in the Top End, north of Pine Creek. Elsewhere,
Nasutitermes triodiae builds large mushroom-shaped mounds or columnar mounds
without the Top End elaborations151•72,
In these mounds, the inner region (beside the nursery N) is solid and extremely hard.
Around this solid central core, there is a zone of open galleries and cells in which
chaffed grass (8.61 ± 0.51 mm) is stored in considerable quantities during the dry
season97• As the mound grows, the outer storage chambers are abandoned and re-packed
River, site 3. Vertical section of Tumulitermes pastinator mound
at site 3 (Daly River), showing alveolar type of
structure and the nursery (N).
\C �
97
3.1.2.2 Tumulitermes pastinator {Hill)
Tumulitermes pastinator is a small species of variable size and colour; the soldiers
measure 3.5 to 3.75 mm long, they have a nasute head usually light to dark brown72•
The species extends across Northern Australia from Queensland to Western Australia52•
The mounds are generally low dome-shaped structures (Plate 12) about 60-80 em high and 60-90 em in diameter, but occasionally they reach a height of 1.2 m and a basal
diameter of 1 .5 m. The outer wall is made of repacked soil material and is very thin and
dense. It covers an open alveolar interior made of softer repacked soil material (Plate
13). The vast number of chambers and galleries are packed with fragments of chaffed
grass stems. The central nursery [N] (Plate 13 ground level) is made of a soil/carton
mixture97
3.1.2.3 Amitermes vitiosus Hill
Amitermes vitiosus is a variable species with a dark orange head. The soldiers measure
4-5 mm and are of the mandibulate type72 (Figure 3.1-C). It is a fairly common species
in the northern parts of Queensland and the Northern TerritorY2• It is often found in
association with other grass�feeding mound�building species, Nasutitermes triodiae
(site 4) and Tumulitermes hastilis. The mounds of this termite are remarkable for their
abundance in certain localities and for their diversity in form. Their structures are
intensely hard (concrete� hard material), with a thin, undifferentiated outer wall which
is often deeply sculptured (Plates 14 and 15). They range in colour from light gray
(site 2) to dark gray (site 4) and to deep mahogany red (site 5, Plates 14 and 15)
according to the colour of the surrounding soi1'2• The commonest form consists of a
colunmar mound up to 1.2 m high with a basal diameter up to 60 em (site 2). On sandy
soil, it builds small conical mounds (site 5). On areas subject to seasonal flooding (such
as site 4), the mounds closely resemble those of Amitermes meridionalis in being
laterally flattened and oriented more or less on a north·south axis (Plate 9).
PLATE 14 Amitermes vitiosus mound (50cm height), Elliott. PLATE IS Vertical section of Amitermes vitiosus mound in
Elliott showing the concrete hard structure.
\C 00
99
The mounds show little obvious differentiation in gross structure. The internal structure
consists of repacked soil with interconnected galleries. The gaJleries are usually lined
by very fine, black organic residues (Sites 2 and 4). In the outer part of the mound,
they are often filled with fragments of grass, leaves and organic materials, including the
bodies of dead termites 38.
3.1.2.4 Tumuliterme.� hastilis (Frogatt)
The soldiers measure 3.5-4.5 mm. The species has a wide distribution in the inland low
rainfall areas of Queensland, Northern Territory, Western Australia and South Australia.
It builds tall narrow mounds up to 1 . 5 m height and 50 em wide at ground level52 In
Daly River site I, the mounds are relatively small and narrow (see Plate 16), they do not
exceed 90 em height. The construction material is mostly soil97. The interior consists
of large numbers of small chambers stored with grass ·and pieces of unidentifiable
vegetable debris.
Plate 16 Tumulitermes hastilis mound (55cm height), Daly River. site I .
� <:> TABLE 3.1 Quality control of selected elemental composition of reference material (BCSS-1, MESS-I <:>
and 1AEA SOILS), following perchloric/nitric acid (4:1) extraction (mgl100g).
nd: not detennined • nc: not certified, OSS (Jabiru) value
101
3.2 Acid Extractable (PerchloridNitric Acids) Selected Elements from Termite
Mounds and Soil Together with Particle Sizes.
3.2.1 Quality Assurance and Quality Control
The results of the quality control and quality assurance are shown in Tables 3.1, 3.2 and
3.3. Table 3 .1 shows the results of selected elemental composition of reference material
(BCSS-1, MESS-I and IAEA SOILS) following perchloric/nitric acids (4:1) extraction.
This extraction does not result in a total concentration in all elements. However, the
efficiency for Fe is greater than 95 % for all the reference materials and the extraction
efficiencies for Co, Cu, Mg, Mn and Zn are generally greater than 80 %. The Ca
extraction efficiency varies between reference materials (54 % in MESS-I to .76 % in
BCSS-1). Na and K are poorly extracted from soils and sediments. The efficiency
being of the order of 42 to 60 % for K and 5 to 48 % for Na.
. The extraction efficiency depends on the mineralogical nature of the sample. It was
therefore important to check with reference material more closely matched to the
samples of this study. As there is no termitaria reference material available, internal
reference materials were established. Four termite mound samples: Amitermes vitiosus
(Daly River, site 4), Amitermes vitiosus (Elliott, site 5), Tumu/itermes pastinator (Daly
River, site 1), Nasutitermes triodiae (Daly River, site 4) and three soil samples: Daly
River (site 4), Elliott (site 5) and Howard Springs (site 6) were sent to a private
laboratory for comparative analyses. The termitaria samples were analysed by X-ray
fluorescence for AI, Ca. Fe, K, Mg and Zn and following a mixed acid digestion (HCI,
HCI04 and HF) by Inductively Coupled Plasma Optical Emission Spectrometry (ICP
OES) for all the selected elements but Co and by Inductively Coupled Plasma Mass
Spectrometer (ICP-MS) for Co. Soils were analysed by ICP-OES following mixed acid
digestion (HCI, HCI04 and HF)
Tables 3.2 and 3.3 show the results of the selected elemental composition of the internal
reference materials (mounds and soils) following perchloric/nitric acids ( 4: 1) extraction.
TABLE 3.2 Quality control of selected elemental composition of internal reference termitaria material (Av44D4, Av22E, Tp23Dl and Nt24D4), following perchloric/nitric acid (4:1) extraction (mg/lOOg).
The method shows good extraction efficiencies compared to the XRF external laboratory
results, except for Mn in the Tumulitermes pastinator and Nasutitermes triodiae mound
samples; and good recoveries compared to ICP-OES and JCP-MS, except for Na (where
percentage recovery varies from 53 to 68 % in the mounds and 43 to 63 % in the soils).
The K percentage recoveries were highest in the Elliott samples: 100 % in the mound
and 86 % in the soil, while at the other sites it was arotu1d 60 to 70 %.
The general precision of the method is indicated by the selected element standard
deviations given in Tables 3.1, 3.2 and 3.3. These values are the total of the values
obtained for all the runs, during testings and analyses. They were usually very low
(below 5 %) but K and Na were higher (± 10 %).
3.2.2 Overview, General Correlation
A global overview of termite mound selected elements and particle sizes data from
different areas and different species is presented in Table 3.4. It shows strong negative
and positive correlations between most of the variables. Out of the 91 variable
cOrrelations, 72 were significantly correlated. For example, the iron was positively
correlated to aluminium, cobalt, copper, manganese, zinc, clay and fine sand. It was
negatively correlated to potassium, magnesium, sodium, silt and coarse sand. No
significant correlation was observed between iron, calcium and silt. The calcium was
positively correlated to copper, manganese and zinc and negatively correlated to
potassium and sodium. The clay was significantly correlated to most of the variables:
positively correlated to aluminium, cobalt, copper and iron and negatively correlated to
potassium, magnesium, silt and coarse sand. No significant correlation has been found
between clay, calcium and fine sand.
While a complete study of the correlations would be interesting, it is not the focus of
the project which has concentrated on Aboriginal community use ofterrn.itaria. For this
purpose, a finer detailed investigation between: age of sample collection, sample
position, size of mound, mounds of different species and different sites has been
necessary.
� = ...
TABLE 3.3 Quality control of selected elemental composition of internal reference soil material (OlE, 25D4 and 29H), following perchloric/nitric acid (4:1) extraction (mg/IOOg).
@: for C1(planation of soil reference material number n:fi:r to chapter 2.3.1 t: mixed acid digestion - HCI, HCIO,, HF; #: ICPMS
TABLE 3.4 Pearson correlation (PC) matrix and probabilities (P t) of selected elements and particle size of 87 termite mounds (n=189) of all the species and sites studied (depth=l).
Element/ Aluminium Calcium Cobalt Copper Iron Potassium Magnesium Particle size PC p PC p PC p PC p PC p PC p PC p Aluminium 1.000 ...
Calcium 0.051 N 1.000 ...
Cobalt 0.749 ••• 0.056 N 1.000 •••
Copper 0.801 ... 0.160 • 0.818 ••• 1.000 ••• Iron 0.842 ... 0.037 N 0.764 ••• 0.817 ••• 1.000 •••
One mound of each termite species selected by the Aboriginal communities of Elliott
and Daly River, was analysed in detail. The detailed study included comparisons
between:
a) the age of the outside material of the moWld (old and new) in Nasutitermes triodiae
and Tumulitermes pastinator mounds;
b) the depth of sample collect where: depth�O, represents the Q.J em fraction of the
outside mound; depth= 1 , represents the 0� 10 em fraction of the outside mound; and
depth=2, represents the 0-10 em taken from the inside central axis of the mound.
Samples were taken from depths= 1 and 2 for Amitermes vitiosus and from depths=O,
I and 2 for Tumulitermes pastinator and Nasutitermes triodiae mounds;
c) the vertical position of the sample in the mound: top, middle and bottom
The physical characteristics of the termitaria selected are given in Table 2.2 and the
detailed mound sample summary for the 3 termitaria is given in Table 2.3. The detailed
results of the analyses are given in Appendices lc, le, lg, lllc, llle, lllg.
3.2.3.1 Amitermes vitiosus (Elliott, Site 5)
The effects of depth and position on the selected elements and particle sizes of the
mound together with ANOV A probability of differences are presented in Table 3.5 and
Table 3.6 respectively.
107
TABLE 3.5 Amitennes vitiosus mound detailed study (Elliott, site 5): depth effects on selected elements (mg/lOOg) and particle sizes (o/o) (mean ± standard deviation) together with ANOV A probability of differences (P) between depths.
Element/ Depth-I Depth-2 p Particle size n""6 n=3 � Aluminium 3428 ± 128 3548 ± 191 N
Calcium 136 ± 17 127 ± 22 N
Cobalt 0.32 ± 0,03 0.35 ± O.Ql N
Copper 0.91 ± 0.03 0.92 ± 0.03 N
Iron 1644 ± 43 1666 ± 24 N
Potassium 154 ± 2.8 160 ± 3.0 N
Magnesium 92 ± 2.1 92 ± 5.5 N
Manganese 7.93 ± 0.33 7.76 ± 0.23 N
Sodium 6.19 ± 0.32 5.98 ± 0.43 N
Zinc 1.09 ± 0.10 1.10 ± 0.11 N
Clay 16.5 ± 6.5 20.8 ± 2.3 N
Silt 10.4 ± 5.0 9.7 ± 5.3 N
Fine sand 36.3 ± 6.0 33.8 ± 1.1 N
Coarse sand 38.0 ± 1.3 37.5 ± 0.9 N l: N: P>O.OS
In the mound studied, there were no statistically significant differences associated with
depth of the sample in the mound.
The ANOV A shows that there are highly significant differences in calcium, magnesium
and zinc with position (Table 3.6). The pairwise comparisons (Tukey test) indicate that
the differences are associated with increases in concentrations in the top and middle
sections of the mound (Figure 3.3).
108
160 l 1.2 -
� Bottom 0 Middle
120 1 Fj l Lc Too_l 09 0, 0 0 -' 0 80 � El � El � 06 5 \1 0 0 40 � El � El � 03
0 -'--- 0.0 I 1=1 ,
FIGURE 3.3
c. MINERAL
Mg zo MINERAL
Position effects (mean ± SE) on calcium, magnesium and zinc (mg/lOOg) in Amitermes vitiosus mound sampled in Elliott (site 5) at depths I and 2
109
TABLE 3.6 Amitermes vitiosus mound detailed study (Elliott, site 5): position effects on selected elements (mg/lOOg) and particle sizes (%) (mean ± standard deviation) together with ANOV A probability of differences (P) between positions.
Element/ Top Middle Bottom Particle size n=3 n=4 n=2
Aluminium 3529 ± 135 3521 ± 1 1 7 3271 ± 82
Calcium 133 ± 3.4 147 ± 5.5 105 ± 1 .00
Cobalt 0.34 ± 0.03 0.32 ± 0.02 0.33 ± 0.04
Copper 0.90 ± 0.02 0.93 ± 0.03 0.92 ± 0.05
Iron 1655 ± 70 1653 ± 21 1641 ± 1.0
Potassium 158 ± 3.5 153 ± 3.4 160 ± 3.9
Magnesium 94 ± 2.3 93 ± 0.8 87 ± 1.3
Manganese 7.88 ± 0.40 7.93 ± 0.32 7.74 ± 0.15
Sodium 6.36 ± 0.21 6.16 ± 0.31 5.67 ± 0.01
Zinc 1 . 15 ± 0.63 1.14 ± 0.02 0.94 ± 0.06
Clay 12.1 ± 6.1 20.6 ± 3.1 21.3 ± 0.4
Silt 15.5 ± 2.4 8.4 ± 3.2 5.6 ± 0.7
Fine sand 39.9 ± 7.0 32.5 ± 0.9 34.9 ± 0.3
Coarse sand 37.3 ± 1 . 1 38.3 ± 1.2 37.6 ± 1.4
l: N: P>O.OS; . . : O.OOJ<P<O.OI
3.2.3.2 Tumulitermes pastinator (Daly River, Site 3)
p t N ..
N
N
N
N . .
N
N ••
N
N
N
N
The effects of age on the selected elements and particle sizes of the mound together with
the ANOVA probability of differences are presented in Table 3.7. The results of the
ANOV A show that the effects of age on the proportion of the selected elements and
particle sizes were not significant for the Tumulitermes pastinator mound at site 3
(P>0.05).
110
80
60 � ... a. 0 0 � ' § 40 � 1111! Sl 0 0
20 � .I.Ellli!
1000
750 � ' IIIII
500 � 1 rn
250 � 1 rn
1 2
g
6
3
l8l Depth-2 g Depth•1 D Depth .. O
0 -'-----'- 0 -'---__u o L--.t...r;:
FIGURE 3.4
FIGURE 3.5
c.
40 ,
I 30
� z
K MINERAL
Mo
Depth effects (mean ± SE) on calcium, potassium and manganese (mg!IOOg) in Tumulitermes pastinator mound sampled in Daly river (site 3)
l ll!l Deoth•2 El Depth-1 0 Depth-0
T
� 20 a: w a_
10
0 I I '7 Sill Coarse sand PARTICLE SIZE
Depth effects (mean ± SE) on silt and coarse sand (%) in Tumulitermes pastinator mound sampled in Daly river (site 3)
111
TABLE 3. 7 Tumulitermes pastinator mound detailed study: age effects on selected elements (mg/lOOg) and particle sizes (o/o) (mean ± standard deviation) together with ANOVA probability of differences (P) between ages. (Deptb=O).
Element/ Old material New material p Particle size n=5 n=2 + Aluminium 3983 ± 498 3986 ± 161 N
Calcium 16.5 ± 1.3 24.1 ± 7.6 N
Cobalt 0.44 ± 0.04 0.45 ± 0.01 N
Copper 0.86 ± 0.08 0.85 ± 0.01 N
Iron 2942 ± 250 2912 ± 108 N
Potassium 701 ± 83 755 ± 67 N
Magnesium !55 ± 10 164 ± 24 N
Manganese 5.43 ± 0.41 5.96 ± 0.94 N
Sodium 14.9 ± 3.0 15.3 ± 1.5 N
Zinc 1 .24 ± 0.10 1 .26 ± 0.14 N
Clay 18.3 ± 1.3 19.4 ± 0.3 N
Silt 16.4 ± 1.0 16.9 ± 0.1 N
Fine sand 37.0 ± 0.3 38.4 ± 4.1 N
Coarse sand 31.5 ± 1.7 30.2 ± 5.0 N
I' N: P>O.OS
The effects of depth and position on the selected elements and particle sizes of the
Tumulitermes pastinator mound together with the ANOV A probability of differences are
presented in Table 3.8.
The ANOVA shows a highly significant difference (P<O.Ol) in calcium, potassium,
manganese, silt and coarse sand and a significant difference in clay (P<0.05) associated
with depth. The pair-wise comparison between depths indicates that the inner section
of the mound ( depth=2) contains a higher level of calcium, potassium, manganese
� � TABLE 3.8 Tumulitermes pastinator mound (Daly River, site 3) detailed study: depth and position effects on selected
...
elements (mg/lOOg) and particle size (%) (mean ± standard deviation) together with ANOVA probability of differences (P) between depths and positions (at deptb=l).
Element/ Depth=O Depth= 1 Depth=2 P (Depths) Top Middle Bottom P (Positions)
Particle size n=7
Aluminium 3984 ± 412
Calcium 18.7 ± 4.9
Cobalt 0.44 ± o.m Copper 0.85 ± 0.08
Iron 2934 ± 210
Potassium 717 ± 78
Magnesium 158 ± 13
Manganese 5.58 ± 0.57
Sodium 15.0 ± 2.5
Zinc 1.24 ± 0.10
Clay 18.6 ± 1.2
Silt 16.5 ± 0.9
Fine sand 37.4 ± 1.8
Coarse sand 3 1 . 1 ± 2.6
n=6
3826 ± 446
23.3 ± 3.9
0.42 ± 0.04
1.50 ± 0.55
2917 ± 198
651 ± 79
152 ± 19
5.97 ± 0.83
13.8 ± 2.1
1.54 ± 0.28
2 1 .2 ± 2.4
15.6 ± 2.2
36.9 ± 2 . 1
28.8 ± 2.9
n=4
4039 ± 269
60.2 ± q.o 0.48 ± 0.04
1.34 ± 0.46
2939 ± 186
783 ± 61
175 ± 16
10.0 ± 2.1
16.8 ± 1.4
1.66 ± 0.31
20.2 ± 1.3
18.7 ± 0.8
38.1 ± 2.4
23.7 ± 2.7
I ' N: P>O . O S ; * : O . O l<P< O . O S ; * * : 0 . 00l<P<0.01
t n=2 n=2 n=2 t N 3795 ± 233 4280 ± 382 3403 ± 145 N
•• 23.8 ± 1.2 19.8 ± 1.71 26.2 ± 5.34 N
N 0.42 ± 0.04 0.44 ± 0.06 0.39 ± 0.01 N
N 1.22 ± 0.64 1.81 ± 0.25 1.48 ± 0.82 N
N 2821 ± 204 3117 ± 175 2814 ± 55 N
• • 629 ± 51 732 ± 87 593 ± 20 N
N 165 ± 3.2 159 ± 20.8 131 ± 5.6 N
• • 6.08 ± 0.77 5.32 ± 0.40 6.50 ± 1.10 N
N 1 3 . 1 ± 3.5 15.2 ± 1.80 13.0 ± 0.57 N
N 1.46 ± 0.28 1.75 ± O.Q2 1.43 ± 0.42 N
• 21.2 ± 1.9 23.1 ± 2.7 19.4 ± 2.3 N
• • 16.1 ± 1.2 17.4 ± 1.9 13.2 ± 0.3 N
N 35.9 ± 1.0 37.5 ± 0.1 37.4 ± 4.1 N
• • 28.8 ± 1.7 26.0 ± 0.8 31.7 ± 2.5 N
113
There are no significant differences associated with position of elements or particle sizes
content at depth= ! (Table 3.8). The other depths (0 and 2) were not investigated
because of the size of the mound.
3.2.3.3 Nasutitermes triodiae (Daly River, Site 3)
The effects of age on the selected elements and particle sizes of the mound together with
the ANOVA probability of differences are presented in Table 3.9. The results of the
ANOV A show that the effects of age on the proportion of the selected elements and particles size were not significant for the Nasutitennes triodiae mound at site 3 (P > 0.05).
Table 3.10 shows that at Daly River site 3, there were significant differences for cobalt,
iron, sodium and coarse sand with depth. The pair-wise comparison between depths
indicates that there was an increase in cobalt in the inner p;rrt of the mound (depth =2),
and an increase of iron, sodium and coarse sand in the outer part (depth =0) (Fignre 3. 6).
The calcium concentration was higher at depth=2, however the increase was not
significant due to a very high standard deviation.
At depth= I , the AN OVA shows significant differences in calcium and coarse sand with
positions (Table 3 . 1 0). The calcium concentration was higher at the bottom position of
the mound and the coarse sand was lower at the top (Figure 3.7).
114
3600
� 3500 a 8 -
' ..§' 3400
0 z 0 ()
3300
3200
,, MINERAl
FIGURE 3.6
30
" 8 25
r � 20
19 23 l!':l Qgpth-2
El Deoth- 1
0 Depth•O
18 22
17 � 21 w 0 �
16 � 20
15 19
14 18 No Coarse sand
MINERAL PARTICLE StZE
Depth effects (mean ± SE) on iron and sodium (mg/lOOg) together with coarse sand (%) in Nasutitermes triodiae mound sampled in Daly river (site 3)
22
2 1
� iii � 20
1 9
� Bottom
0 Middle
El Top
1 5 .L.._----=:1..,.- 18 .1._---=L,--
FIGURE 3.7
Co
MINERAL
Coarse sand
PARTICLE SIZE
Position effects (mean ± SE) on calcium (mg/JOOg) and coarse sand (%) in Nasutitermes triodiae mound sampled in Daly River (site 3)
115
TABLE 3.9 Nasutitermes triodiae mound detailed study: age effects on selected elements (mgllOOg) and particle sizes (%) (mean ± standard deviation) together with ANOV A probability of differences (P) between ages. (Depth=O).
Element/ Old material New material p Particle size n=l n=S t Aluminium 4772 ± 158 4722 ± 303 0.776 N
Calcium 23.1 ± 3 . 1 25.1 ± 4.7 0.501 N
Cobalt 0.52 ± 0.03 0.53 ± 0.02 0.718 N
Copper 0.82 ± 0.14 0.90 ± 0.19 0.490 N
Iron 3493 ± 169 3596 ± 232 0.485 N
Potassium 940 ± 81 987 ± 72 0.389 N
Magnesium 265 ± 37 252 ± 16 0.550 N
Manganese 9.61 ± 0.89 9.79 ± 1.79 0.859 N
Sodium 18.1 ± 1.9 18.4 ± 2.3 0.836 N
Zinc 1.63 ± O.o7 1 .62 ± 0.09 0.903 N
Clay 19.8 ± 1.6 23.6 ± 7.4 0.355 N
Silt 25.5 ± 5.0 26.0 ± 7.5 0.908 N
Fine sand 33.1 ± 1.8 33.3 ± 1.9 0.900 N
Coarse sand 23.5 ± 4.2 21.5 ± 1.7 0.350 N
t: N: P>O.OS
� TABLE 3.10 Nasutitennes triodiae mound (Daly River site 3) detailed study: depth and position effects on selected
� "' minerals (mg!IOOg) and particle size (%) (mean ± standard deviation) together with AN OVA probability of differences (P) between depths and positions (at depth=!).
Element/ Depth effects Position effects
Particle size Depth=O Depth=l Depth=2 P (Depths) Top Middle Bottom P (Positions) n=9 n=6 n=5 t n=2 n=2 n=2 t
A number of hypotheses have arisen from the way the Aboriginal communities selected
and sampled tennite mounds. The depth considered in this chapter is depth=! (unless
indicated), which represents the 0-10 em fraction of the outside of the mound. Depth= I
is the depth normally used by Aboriginals. The detailed analysis results of all the
samples are given in Appendices I to V.
3.2.4.1 Hypothesis 1: The New Material of Nasutitermes triodiae Mounds Contains a Higher Element Content, in Particular Iron and Calcium, and Has a Higher Clay and Silt Content than the Older Part of the Mounds.
Samples from new and old material were collected from 5 Nasutitermes triodiae mounds
(n=30), at Daly River (site 4). The results (mean ± standard deviation), per age group
of the mineral analyses (mg/1 OOg) and the particle size analyses (percent of the fine
fraction: <2mm) together with the probability of differences between old and new parts
of the mounds are given in Table 3.11 . The sample depth studied is depth=O, it
represents the 0-1 em fraction collected from the outside of the mound.
The results of the ANOV A show that the only significant differences between the ages
of samples were for aluminium, copper and iron, where the increase in the old material
was highly significant (P<O.OI) and for potassium significantly different (P<O.OS)
(Figure 3.8).
118
5000 ..,
4ooo I a, 0 ;2 3000 i ' � 5 � 2000 -i ()
10] FIGURE 3.8
5000 "")
4000 i a, 0 � 3000 1 ' � 5 S? 2000 -l 0 ()
100: 1 FIGURE 3.9
1.0 12 New marerial
n 1 0 Old rna !erial 1
DB
I
I
I � 0.6
� 04
I I B 02
[)! 0.0 AI ,, K c"
MINERAL MINERAL
Age effects (mean ± SE) on aluminium, iron, potassium and copper (mg/lOOg) in Nasutitermes triodiae mounds sampled in Daly River (site 4)
e.! New malenal Aluminium -
2000 1 Iron 0 Old material
� 1 m
l r,a
I I Top
� _L 1600
l r,a I I,!, 1200
l r,a I P1 BOO
I I I I 400
0 Middle Bottom Top Middle Bottom
POSITION POSITION
Age effects (mean ± SE) on aluminium and iron (mgllOOg), at three positions (top, middle, bottom), in Nasutitermes triodiae mounds sampled in Daly River (site 4)
119
TABLE 3.11 Age effects on selected elements (mgllOOg) and particle sizes (%) (mean ± standard deviation) in 5 Nasutitermes triodiae mounds (Daly River, site 4). ANOVA probability of differences (P) between ages. (Depth=O).
Element I Nasutitermes triodiae site 4 Particle size Old material New material p
The aluminium, copper, iron, and to a lesser extent potassium, mean content is
consistently higher in the old part of the mound, at each mound position (top, middle
and bottom), as shown in Figure 3.9 for aluminium and iron. The age effect
probabilities per position in the moWld (Table 3.12) were not significant (P>0.05) in all
cases and the age/position pairwise comparisons (new-old material I top-middle-bottom
section) indicated no significant differences for any of the elements or particle sizes, in
the Nasutitermes triodiae mounds (site 4), except aluminium (Table 3.13). For
aluminium, the pairs top-old versus bottom-new and middle-old versus bottom-new were
significantly different (P<0.05).
TABLE 3.12
Element/ Particle size
Aluminium
Calcium
Cobalt
Copper
Iron
Potassium
Magnesium
Manganese
Sodium
Zinc
Clay
Silt
Fine sand
Coarse sand
*= N: P>0.05
Age (new, old) and position (top, middle, bottom) effects on selected elements (mg/lOOg) and particle size (%) (mean ± standard deviation) in 5 Nasutitermes triodiae mounds sampled at site 4. AN OVA probability of differences (P) between ages. (Depth=O).
Top Middle
Old material New material p Old material New material n=5 n=5 t n=5 n=S
Matrix of Pairwise Comparison Probabilities (P:) (Tukey test) between different positions of material (T=top, M=middle, B=bottom) and age (o=old, n=new) for aluminium content in Nasutitermes triodiae mounds sampled at site 4.
T·o
N
N
N
N
N
•
T·n
N
N
N
N
N
M·o
N
N
N
•
M·n
N
N
N
B·o
N
N
B·n
N :: N: P>0.05; *: O.Ol<P<0.05
3.2.4.2 Hypothesis 2: There is No Difference Between Samples Taken
from Different Positions of Termitaria for
Selected Elements and Particle Size Content.
Three species of termitaria at different sites were analysed separately for position effects:
a) Amitermes vitiosus rnotmds: top and bottom positions;
b) Tumulitermes pastinator mounds: top and bottom positions;
c) Nasutilermes triodiae mounds: top, middle and bottom positions.
A) Amitermes vitiosus
The results of selected elements, particle sizes and ANOV A analyses on the position
effects in Amitermes vitiosus mounds from Daly River (sites 2 and 4) and Elliott (site
5) are given in Table 3.14.
For the Daly River site 2, there were no significant differences (P>0.05) for the ten
selected elements and for particle size composition between the top and the bottom
sections of the mounds. This contrasts with the mounds at Daly River site 4, where the
top section showed a highly significant increase in calcium, copper, magnesium,
TABLE 3.14
Element/
Particle size
Aluminium
Calcium
· Cobalt
Copper
Iron
Potassium
Magnesium
Manganese
Sodium
Zinc
Clay
Silt
Fine sand
Coarse sand
Position effects on selected elements (mgllOOg) and particle size (%)(mean ± standard deviation) in Amitermes vitiosus mounds sampled at sites 2, 4 and 5. ANOV A probability of differences (P) between positions.
Da1y River (Site 2} DaJy River (Site 4) Elliott (Site 5)
Top Bottom Probability Top Bottom Probability Top Bottom Probability n=4 n= 4 t n = lO n= 10 t n=ll n= 1 1 t
3042 ± 143 3209 ± 300 0.353 N 3903 ± 353 3750 ± !59 0.228 N 3771 ± 394 3526 ± 359 0.136 N
:1:: N: P>0.05; *: 0.01 <P<0.05; .. : 0.001 <P<O.Ol; ***: P<O.OOI n: number or mounds
� ... ...
123
manganese, zinc and silt; a significant increase in cobalt and a highly significant
decrease in clay and coarse sand, as seen in Figures 3.10 and 3 . 1 1 .
In Elliott, there were significant (P<O.Ol) differences between position for calcium,
magnesium and zinc (Table 3.14). Here, as at Daly River site 4, the increase was in the
top position, with the calcium concentration: 1 1 7 ± 17 mg/1 OOg in the top section and
95 ± 20 mg/IOOg in the bottom section.
In general, there were no highly significant differences between the top and bottom
positions in Amitermes vitiosus mounds for aluminium, cobalt, iron, potassium, sodium
and fine sand content for the three sites studied. For the other elements, the differences
were highly significant but only at site 4.
B) Tumulitermes pastinator
The results of the analyses comparing top with bottom sections of Tumulitermes
pastinator mounds from sites 1 and 6 are given in Table 3.15. The results of the
ANOV A show that the effects of mound position (top/bottom) on the selected elements
and particle size content, were not highly significant (P<0.01) for all the termitaria
studied at the 2 locations. The only significant differences (0.01 <P<O.OS) were for
aluminium, calcium and manganese content at site 1 (Table 3.15).
124
150 ] 10
120 � 8
" 0 0 90 j I F==llil 6 � ' 0 .§ 0 60 j ""' = 4 z 0 0 3: 1 I I 2
0 c. Mg Mo
MINERAL MINERAL
FIGURE 3.10 Position effects (mean ± SE) on calcium, magnesium and manganese (mg/IOOg) in Amitennes vitiosus mounds sampled in Daly River (site 4)
50
40
� 30 w 0 a: 1i: 20
10
O I E Clay Slit Fine sand
PARTICLE SIZE
� Bottom � Top
Coarse sand
FIGURE 3.11 Position effects (mean ± SE) on particle size ( (%) in Amitermes vitiosus mounds sampled in Daly River (site 4)
125
TABLE 3.15 Position effects on selected elements (mgllOOg) and particle sizes (%) (mean ± standard deviation) in Tumulitermes pastinator mounds sampled at sites 1 and 6. ANOV A probability of differences (P) between positions.
Element/ Daly River (Site I)
Particle size Top Bottom n=l5 n=Il
Aluminium 3 1 19 ± 176 2942 ± 196
Calcium 21.4 ± 5.2 26.7 ± 7.8
Cobalt 0.29 ± 0.01 0.29 ± 0.01
Copper 0.57 ± 0.03 0.57 ± O.o3
Iron 1359 ± 101 1321 ± 77
Potassium 514 ± 23 503 ± 19
Magnesium 64.3 ± 7.4 65.7 ± 6.6
Manganese 3.16 ± 0.32 3.57 ± 0.46
Sodium 27.8 ± 2.5 26.4 ± 1.6
Zinc 0.48 ± 0.03 0.48 ± 0.03
Clay 14.8 ± 1.3 14.8 ± 1.4
Silt 8.7 ± 1.0 8.9 ± 1.3
Fine sand 43.3 ± 3.1 43.8 ± 2.0
Coarse sand 32.7 ± 3.7 32.8 ± 2.6
l: N: P>0.05; •: O.Ol<P<O.OS n: number of mounds
C) Nasutitermes triodiae
Howard Springs (Site 6)
p Top Bottom p I n=I3 n=9 I
0.024 • 6300 ± 703 6125 ± 524 0.532 N
0.048 . 52 ± 12 47 ± 10 0.373 N
0.376 N 0.92 ± 0.23 0.84 ± 0.26 0.463 N
0.691 N 1.67 ± 0.33 1.65 ± 0.39 0.881 N
0.302 N 4307 ± 658 4813 ± 971 0.160 N
0.185 N 38.1 ± 7.7 35.6 ± 6.4 0.449 N
0.628 N 48.9 ± 5.3 48.9 ± 4.8 0.975 N
0.014 "' 6.54 ± 1.89 6.61 ± 2.04 0.938 N
0.116 N 6.11 ± 0.66 6.45 ± 0.67 0.258 N
0.997 N 1.14 ± 0.16 1.16 ± 0.25 0.876 N
0.999 N 27.3 .± 3.4 25.8 ± 3.5 0.322 N
0.560 N 6.41 ± 1.0 5.9 ± 1.1 0.358 N
0.605 N 45.9 ± 4.3 46.7 ± 2.2 0.647 N
0.967 N 18.8 ± 2.3 19.8 ± 3.3 0.405 N
The results of the analyses comparing the top and bottom positions of Nasutitermes
triodiae mounds, at sites 4 and 6 are given in Table 3.16. As for Tumulitermes
pastinator, the results of the AN OVA show no highly significant differences between
mound positions for any of the elements or particle sizes. The only significant
differences (O.Ol<P<0.05) were in the site 4 samples, with a significant increase in
aluminium and copper in the top section of the mounds. No position effects were
observed in the Howard Springs mounds.
� TABLE 3.16 Position effects on selected elements (mg/lOOg) and particle size (o/o) (mean ± standard deviation) in Nasutitermes
... a-triodiae mounds sampled at sites 4 and 6. ANOV A probability of differences (P) between positions.
Element/ Daly River (Site 4) Howard Springs (site 6)
Particle size Top Middle Bottom Probability Top Middle Bottom Probability n=l5 n=IO n= l5 * n=S n=S n=S *
To test the influence of the mound size on selected elements and particle sizes, 10 to 15
Amitermes vitiosus, Tumulitermes pastinator and Nasutitermes triodiae mounds were
sampled at two locations per species. The correlation between the size of the mounds
and the selected elements and particle sizes was calculated by the Pearson Correlation
and Probability Test. Equality of variance and normality were checked by Bartlett's
Test. The results of the analyses, given in Table 3.17, indicate that 85 % of the variable
correlations are not significantly correlated. The significant 15 % of correlations are not
consistent across species and sites. For example, calcium is positively correlated to size
only in Amitermes vitiosus mounds at site 5; iron is not significantly correlated for the
three species at all sites and clay is significantly positively correlated in Nasutitermes
triodiae mounds at site 4. There is no consistency in. the way the variables are
correlated. For example in Amitermes vitiosus mounds the iron is positively correlated
(but not significantly) to size at site 4, while it is significantly negatively correlated to
size at site 5. Overall, there are no significant correlations between mound size and
selected element and particle sizes.
3.2.4.4 Hypothesis 4: There are Differences Between Mounds of the
Same Species at the Same Site.
The differences between mounds of the same species at the same site were tested by
ANOV A for Amitermes vitiosus at sites 2, 4 and 5, Tumulitermes pastinator at sites 1
and 6, Nasutitermes triodiae at sites 4, 6 and 7 and for Tumulitermes hastilis at site I .
The probability differences (P) are given in Table 3.18. The concentration data are
given in Appendices I to V. Interestingly, at two sites (Amitermes vitiosus site 4 and
Tumulitermes pastinator site 1), there were no significant differences between mounds
of the same species, except for iron at site 4 (0.01 <P<0.05), but significant and highly
significant differences for the other species on the same site (Nasutitermes triodiae at site
4 and Tumulitermes hastilis at site 1). An example of the variation between mounds is
128
120 2500
100
� 80 � ' 0 .§ 60
........ � 2000 0 0 0 � 1500 .§
0 15 0 40
� 1000 0
• 0 20
l � § s § § .1: 500
0 o ' ''''''''''''''' 1 � 3 • e o r a a w •• n ffl u � • 2 3 • a e r e e � ,, 12 13 u �
Nt MOUND SITE 4 Nt MOUND SITE 4
FIGURE 3.12 Mound effects (mean ± SE) on calcium and iron (mg/lOOg) in Nasutitermes triodiae
mounds sampled in Daly River (site 4)
129
given in Figure 3.12, for calcium and iron content in Nasutilermes triodiae mounds at
site 4. At the other sites, the differences between mounds are often highly significant
for all the selected elements.
For the particle sizes, the differences are less marked. There was only one site with
highly significant differences for silt.
TABLE 3.17 Pearson correlation (PC) and probability (P�) matrix of mound size (height + circumference) with selected elements and particle sizes of three species® at different sites.
site 4: 10 mounds (n=20) site 5: I I mounds (n=26) site I: IS mounds (n=26) site 6: 13 mounds (11'"'25)
site 4: IS mounds (n=41) site 6: 5 mounds (n=J5)
130
TABLE 3.18 Probability differences (ANOVA) PW between mounds of each species (Av, Tp, Nt and Th) per site for selected elements and particle sizes.
Element/ Av Tp Nt Th -
Particle size Site 2 Site 4 Site 5 Site I Site 6 Site 4 Site 6 Site 7 Site I n=12 n=20 n=26 n=26 n=25 n=41 n=15 n::JO n:IO
Aluminium • N .. N .. .. .. . .. ...
Calcium N N .. N ... ... ... ... ...
Cobalt ... N .. N ... .. ... .. ..
Copper ... N .. N ... .. .. • ...
Iron .. • ... N • ... .. ... N
Potassium .. N .. N .. N .. N N
Magnesium • N N N ... .. ... .. ...
Manganese • N ... N ... ... ... ... ...
Sodium • N .. N ... • N • N
Zinc N N N N .. ... .. N ...
Clay . . N N N • .. ...
Silt .. N N N N N N
Fine sand N N N N .. .. N
Coarse sand N N ... N .. • N £ N: P>0.05; 1: O.OI<P<O.OS; ": O.OOI<P<O.Ol; n•: P<O.OOI
3.2.4.5 Hypothesis 5: There are Differences in Selected Elements and Particle Sizes Between Termitaria of Different Species at the Same Site.
The data were analysed statistically by one-way analysis of variance (ANOV A), with
selected elements and particle size distribution as the source of variance between species
mounds. An overview of the termitaria selected elements and particle sizes (mean ±
standard error), per species and per site, is presented in Figures 3.13 (A to D).
7000
6000
- 5000 8 2 4000
g 3000 ;;: 2000
1000
0
150
"' 100 8 ' 0 .§ • 50 0
Av
2 4 '
Av
2 4 '
131
Nt Th
3 6 3 4 6 7 SITE
!'!! Th
7
TABLE 3.13 A Selected elements (mg!IOOg) and particle size (%) (mean ± SE) of Amitennes vitiosus
(A v), Tumulitermes pastinator (Tp ), Nasutitermes triodiae (Nt) and Tumulitermes hastilis
(Th) at sites 1-7.
Aluminium and calcium concentrations (mgllOOg)
1.00
OEO 0 8 060 � ' rn ..s 0.40 0 0
0.20
000
2.00 l 150
0 0 0 � � 1.00 s , 0
0.50
000 I
AY In I'!! Th 5000 J Av I In ., , Nt 4000
0 8 3000 � ' � s 2000 m "-
1000
0
2 4 5 I 3 6 3 4 6 7 I 2 4 5 I 3 6 3 4 6
1000 Av I In � I Nt I Th l Av I In ' ""' Nt
1 ..+, BOO
0 0 600
- � 0 �
� � - � I "' s 400
rlil "'
200
vp 9 r�(J '''(' ¥(4 r\Q II"(I !Cf' Kie "I" ' I 0 2 4 5 I 3 6 3 4 6 7 I I 3 6 3 4 6 2 4 5
SITE SITE
TABLE 3.13 B Selected elements (mg/IOOg) and particle size(%) (mean ± SE) of Amilermes vitiosus
(A v), Tumulitermes pastinaror(Tp), Nasutitermes triodiae (Nt) and Tumulitermes hastilis
(Th) at sites 1�7. Cobalt, copper, iron and potassium concentrations (mg/IOOg)
hh I � "' N
7
I Th
7
300 .., I I� I Th I 40 Av In Nt
30 roo ] I -a 8
I -
E3 r 20 � 100 El El • = z 10 J i I iii i �li i � ;li I 0
2 4 5 I 3 6 3 4 6 . 7 I 2 4 5 I 3 6 3 4 6 7
10 , 2.00 Av In Rl Nt Th
8 1.50 -a -a 0 6 I
0 0
I 0 -
� - � -' '(» 1.00 0 s 4 5
c
� c " N 2 0.50
0 ')" "I'' "�' I "j � ·y· ,,, ,,�. "j¥ , ... .,,,. . "i' 0.00 2 4 5 I 3 6 3 4 6 7 I 2 4 5 I 3 6 3 4 6 7
SITE SITE
TABLE 3.13 C Selected elements (mg/IOOg) and particle size(%) (mean ± SE) of Amitermes vitiosus (A v), Tumulitermes pastinator(Tp), Nasutitermes triodiae (Nt) and Tumulitermes hastilis
(Th) at sites 1-7. � "' "' Magnesium, manganese, sodium and zinc concentrations (mg/100g)
TABLE 3.13 D Selected elements (mg/IOOg) and particle size(%) (mean ± SE) of Amitennes vitiosus
Nt � � I !" I
4 6 7
Nt
4 ' 7
(A v), Tumulitermes pastinator(Tp), Nasurilermes triodiae (Nt) and Tumulitermes hastilis
(Th) at sites 1-7. Clay, silt, fine sand and coarse sand content (%)
-... ...
135
A) Tumulitermes pastinator versus Tumulitermes hastilis mounds at site 1 (Daly
River).
The mean (± standard deviation) of the selected elements, the particle size distribution
and the pairwise comparison probabilities between Tumulitermes pastinator and
Tumulitermes hasti/is mounds, sampled at site 1 , are given in Table 3.19.
As seen in Table 3.19, the aluminium, potassium, sodium, zinc, clay and to a lesser
extent fine sand content are highly significantly increased (P<O.Ol) in Tumulitermes
pastinator mounds, while the calcium, magnesium, manganese and coarse sand content
are higher in Tumulitermes hastilis mounds. For example, in Tumulitermes pastinator
motu1ds: aluminium is 38 % higher; calcium 4.5 fold lower and coarse sand 22 % lower.
For the other selected elements and particle sizes (cobalt, copper, iron and fine sand)
there were no significant differences detected at site 1 .
136
TABLE 3.19 Selected elements (mg/100g) and particle sizes (%) (mean ± standard deviation) of Tumulitermes pastinator and Tumulitermes hastilis mounds sampled at site 1. Probability of differences (P) between the two species mounds@.
Element/ Tumulitermes Tumulitermes p Particle size pas tina tor hastilis �
n=26 n=S
Aluminium 3044 ± 201 2209 ± 247 • •••
Calcium 23.7 ± 6.8 107 ± 66 ...
Cobalt 0.29 ± 0.01 0.30 ± 0.04 N
Copper 0.57 ± 0.03 0.57 ± 0.07 N
Iron 1343 ± 92 1249 ± 1 13 • N
Potassium 510 ± 22 403 ± 18 • ...
Magnesium 64.9 ± 6.9 76 ± 14.2 ..
Manganese 3.33 ± 0.43 5.91 ± 1 .65 ...
Sodium 27.2 ± 2.2 18 .4 ± 1 .02 • ...
Zinc 0.48 ± 0.03 0.42 ± 0.07 • ..
Clay 14.8 ± 1.3 12.3 ± 2.6 • ..
Silt 8.8 ± 1.1 9.6 ± 0.6 N
Fine sand 43.5 ± 2.7 38.9 ± 10.1 • •
Coarse sand 32.8 ± 3.2 42.0 ± 1 1 . 1 ...
@: 15 Tumulitermes pastinator mourids and Tumulilermes hastilis mounds
I ' N: P>O.OS; $; O.Ol<P<O.OS; U; O.OOI<P<O.Ol; •••: P<O.OOI .o. : mean variable content higher in Tunrulitermu pastinator mounds
B) Tumu/itermes pastinator versus Nasutitermes triodiae mounds at site 3 (Daly
River).
The mean (± standard deviation) of the selected elements, the particle size distribution
and the pairwise comparison probabilities between Tumulitermes pastinator and
Nasutitermes triodiae mounds, sampled at site 1 , are given in Table 3.20.
The probability of differences (ANOV A) between the two species' mounds indicates a
significant increase of fine sand (P<O.OS) and a highly significant increase of coarse sand
(P<O.OOl) in Tumulitermes pastinator mounds. However, although the copper content
is 1.50 ± 0.55 mg/IOOg in Tumulitermes pastinator mounds and 0.97 ± 0.30 mgl100g
137
m Nasutitermes triodiae mounds, no significant difference was indicated. In
Nasutitermes triodiae, there is a highly significant increase in aluminium, cobalt, iron,
potassium, magnesium, manganese and silt content (P<O.Ol) and a significant increase
of sodium (P<O.OS). For example, the mean magnesium and manganese content are
respectively 69 and 57 % higher in Nasutitermes triodiae.
There are no significant statistical differences detected at site 3 for calcium, copper, zinc
and clay content between the two species.
TABLE 3.20 Selected elements (mg/IOOg) and particle sizes (%) (mean ± standard deviation) of Tumulitermes pastinator and Nasutitermes triodiae mounds sampled at site 3. Probability of differences (P) between the two species mounds®.
Element/ Tumulitermes Nasutitermes p Particle size pastinator triodiae t
n=6 n=6
Aluminium 3826 ± 446 4488 ± 167 ••
Calcium 23.3 ± 3.9 23.1 ± 4.72 • N
Cobalt 0.42 ± 0.04 0.52 ± 0.04 ••
Copper 1.50 ± 0.55 0.97 ± 0.30 • N
Iron 2917 ± 198 3434 ± 124 ...
Potassium 651 ± 79 897 ± 65 •••
Magnesium 1 5 1 ± 19 256 ± 24.9 •••
Manganese 5.97 ± 0.83 9.38 ± 0.99 •••
Sodium 13.9 ± 2.1 16.5 ± 2.03 •
Zinc 1.54 ± 0.28 1.68 ± 0.18 N
Clay 21.2 ± 2.4 19.4 ± 1.0 • N
Silt 15.6 ± 2.2 29.4 ± 1.3 ...
Fine sand 36.9 ± 2.1 33.6 ± 2.1 • •
Coarse sand 28.8 ± 2.9 20.4 ± 1.5 • ...
@ : I Tumuliterm£s pastinator mound and I Nasutitermes triodiae mound
I' N: P>O.OS; *: O.Ol<P<O.OS; .. : O.OOI<P<O.Oi; •••: P<O.OOI " mean variable content higher in Tumulitermes pastinator mounds
138
TABLE 3.21 Selected elements (mg/100g) and particle sizes (%) (mean ± standard
deviation) of Amitermes vitiosus and Nasutitermes triodiae mounds
sampled at site 4. Probability of differences (P) between species
mounds®.
Element/ Amitermes Nasutitermes p Particle size vitiosus triodiae �
@: 10 A.mitermes vitioswr mounds and IS Nasulitermes triodiae mounds
!: N: P>O.OS; •: O.Oi<P<O.OS; �•: O.OOI<P<O.Oi; •n: P<O.OOI & : mean variable content higher in Amilermes vitiosus
C) Amitermes vitiosus versus Nasutitermes triodiae mounds at site 4.
139
The mean (± standard deviation) of the selected elements, the particle size distribution
and the probabilities (P) between Amitermes vitiosus and Nasutitermes triadiae mounds,
sampled at site 4, are presented in Table 3.21.
The results of the ANOV A betweenAmitermes vitiosus and Nasutitermes triodiae mound
contents, indicate highly significant differences (P<O.Ol ) for cobalt, potassium,
manganese, sodium, clay, silt, fine sand and coarse sand and a significant di�erence
(?<0.05) in calcium content. Nasutitermes triodiae mounds have a larger proportion of
clay (36 % higher) and coarse sand (52 % higher) than the Amitermes vitiosus mounds.
Amitermes vitiosus mounds have a higher calcium, cobalt, potassium, manganese (51 %),
sodium, silt (1 1 1 %) and fine sand (%) content than Nasutitermes triodiae mounds.
No significant differences were observed for aluminium, copper, iron, magnesium and
zinc content between the Amitermes vitiosus and Nasutitermes triodiae mounds studied
at site 4.
D) Tumulitermes pastinator versus Nasutitermes triodiae mounds at site 6 (Howard Springs).
The mean (± standard deviation) of the selected elements, the particle size distribution
and the pairwise comparison probabilities between Tumulitermes pastinator and
Nasutitermes triodiae mounds sampled at site 6, are given in Table 3.22.
Significant differences are observed between the two species for: aluminium (P<0.05),
calcium (P<O.OOl), iron (P<O.Ol), manganese (P<0.05) and silt (P<O.OOI). No
significant differences are observed for the other variables.
140
TABLE 3.22 Selected elements (mgllOOg) and particle sizes(%) (mean ± standard
deviation) of Tumulitermes pastinator and Nasutitermes triodiae
mounds� sampled at site 6. Probability of differences (P) between the
3.2.4.6
two species mounds.
Element/ Tumu/itermes Nasutitermes Particle size pas tina/or triodiae
n�25 0""15
Aluminium 6300 ± 658 5738 ± 733
Calcium 50.0 ± I I . 7 83.7 ± 23.2
Cobalt 0.86 ± 0.24 0.79 ± 0.21
Copper 1 .63 ± 0.35 !.56 ± 0.17
Iron 4527 ± 780 3649 ± 759
Potassium 37.3 ± 7.4 38.5 ± 4.6
Magnesium 49.3 ± 5.6 50.3 ± 7.62
Manganese 6.40 ± 1.88 5.03 ± 1.68
Sodium 6.22 ± 0.73 5.89 ± 0.79
Zinc 1 . 15 ± 0.22 1.04 ± 0.16
Clay 26.9 ± 3.5 27.6 ± 3.0
Silt 629 ± 1 .1 7.9 ± 1 2
Fine sand 45.6 ± 3.8 44.7 ± 3.0
Coarse sand 19.5 ± 2.9 19.2 ± 2.5
fl: 13 Tumulitermes pastinator mounds and S Nasuli/ermes triodiae mounds
N: P>0.05; •: O.Ol<P<0.05; n: O.OOI<P<O.OI; u•: P<O.OOI .o : mean variable content higher in Tumulitermes pastinator mounds
•
•
•
•
p t
•
...
N
N ..
N
N
• •
• N
• N
N ...
• N
• N
Hypothesis 6: There are Differences in Element and Particle
Size Content for Same Species Mounds at Different Sites.
A) Amitermes vitiosus
The mean (± standard deviation) of selected elements (mg/IOOg) and particle sizes (%)
of Amitermes vitiosus mounds at sites 2, 4, 5 and at all sites (2+4+5) are presented in
Table 3.23 together with the probability of differences (P) between the Amitermes
vitiosus termitaria samples collected at the three sites.
141
TABLE 3.23 Selected elements (mg/100g) and particle sizes (%) (mean ± standard deviation) of Amitermes vitiosus mounds per site (2, 4 and 5) and for all sites (2+4+5) together with the probability of differences (P) between sites.
Element/ Daly River Daly River Elliott Total p Particle size (Site 2) (Site 4) (Site 5) (Sites 2+4+5) j
The ANOVA shows that there are highly significant differences (P<O.OOl) in nearly all
the variables studied: aluminium, calcium, cobalt, iron, potassium, magnesium,
manganese, sodium, zinc, fine sand and coarse sand; significant differences (P<0.05) in
copper and silt and no significant difference (P>O.OS) in clay content (the mean clay
content remained at approximately 16 %). The aluminium, copper, iron, magnesium,
manganese and fine sand mean content, although highly significantly different, varied
within a narrow range. This contrasts with the mean contents of calcium, potassium,
sodium, zinc and coarse sand, which vary greatly between sites (Figure 3.13, A to N ).
142
B) Tumulitermes pastinator
The mean (± standard deviation) of selected elements (mg/1 OOg) and particle sizes
analyses (%) of Tumulitermes pastinator mounds at sites 1 , 3, 6 and at all sites (1+3+6)
are presented in Table 3.24, together with the probability of differences (P) between all
Tumulitermes pastinator samples collected.
The AN OVA shows that there are highly significant differences P(<O.OOl) between sites
for all the selected elements and particle sizes. Figure 3.13 (A to N) shows the wide
variation between sites. For example, the mean content varied from 3044 ± 201 to 6300
± 658 mgllOOg for aluminium, 1343 ± 92 to 4527 ± 780 mg/lOOg for iron and 27.2 ±
2.2 to 6.22 ± 0.73 mgllOOg for sodium.
TABLE 3.24 Selected elements (mg/IOOg) and particle sizes (%) (mean ± standard deviation) of Tumulitermes pastinator mounds per site (1, 3 and 6) and for all sites (1+3+6) together with the probability of differences (P) between sites.
Element/ Daly River Daly River Howard Springs Total Probability Particle (Site I) (Site 3) (Site 6) (Sites 1+3+6) P t SIZe n�26 n=6 n�5 n�57
Most of the lowest mean values are found in the Daly River site I samples (except for
potassium, sodium and coarse sand); while the highest mean values are found in the
Howard Springs samples. Site 3 has a particularly high copper, potassium, magnesium,
manganese, zinc and silt mean content.
C) Nasutitermes triodiae
The mean (± standard deviation) of selected elements (mg/l OOg) and particle sizes (%)
of Nasutitermes triodiae mounds per site (3, 4, 6 and 7) and for all sites (3+4+6+ 7) are
presented in Table 3.25 together with the probability of differences (P) between all the
Nasutitermes triodiae samples collected.
The ANOV A shows that like the Tumulitermes pastinator mean contents, there are
highly significant differences between sample content for all the selected elements and
particle sizes. Figure 3.13 (A to N) shows the wide variation between sites. For
example, the mean content varied from: 23.1 ± 4.7 to 83.7 ± 23.2 mg/100g for calcium,
890 ± 134 to 3649 ± 759 mg/1 OOg for iron and 38.5 ± 4.6 to 897 ± 65 mgllOOg for
potassium. The variation between clay content remained small: 19.2 ± 5.9 % to 27.6
± 3.0 %
144
TABLE 3.25 Selected elements (mg/IOOg) and particle sizes (%) (mean ± standard deviation) of Nasutitermes triodiae mound samples per site (3, 4, 6 and 7) and for all sites (3+4+6+7) together with the probability of differences (P) between sites.
Element/ Daly River Daly River Howard Berrimah Total (Sites P l Particle size (Site 3) (Site 4) Springs (Site 7) 3+4+6+7)
The Howard Springs site had the highest aluminium, calcium, cobalt, copper, iron, clay
and fine sand content; while the higher potassium, magnesium, manganese, sodium, zinc,
silt and coarse sand mean values were found principally at the Daly River, site 3. The
iron content is exceptionally low at the Berrimah site while the sodium content is the
highest.
3.2.4. 7 Hypothesis 7:
145
There are Differences in Elements and Particle
Sizes Between Different Species at Different
Sites.
The minimum, maximum and mean (± standard deviation) of selected elements
(mg/l OOg) and particle sizes (%) of the species of mounds (Amitermes vitiosus,
Tumulitermes pastinator and Nasutitennes triodiae) chosen by the Aboriginal
communities, sampled at sites (1 to 7), together with the probability of differences
between the three species, are given in Table 3.26.
The ANOV A between the moWldS of the three species shows that there are a highly
significant (P<O.OOI) differences for all the selected elements and particle sizes, except
for zinc, which was significantly difference (P<0.05). The Tukey pair-wise comparison,
performed after the Bartlett test for homogeneity of group variances, indicates that the
differences between Amitermes vitiosus and Tumulitermes pastinator mounds are more
important amongst all the elements and particle sizes than-between Amitermes vitiosus
and Nasutitermes triodiae mounds and between Tumulitermes pastinator and
Nasutitermes triodiae mounds (Table 3.26).
146
TABLE 3.26 Selected elements (mg!lOOg) and particle sizes (%) (minimum,
ElementJ
maximum and mean ± standard deviation) of Amitermes vitiosus, Tumulitermes pastinator and Nasutitermes triodiae mounds sampled at sites 1 to 7, together with the probability of differences (P: Av· Tp-Nt) between the three species and the pairwise comparison probabilities between species (P: Av-Tp, P: Av-Nt and P: Tp-Nt).
Av, Tp and Nt p p p p Particle size
Mean ± SD Av·Tp·Nt Av·Tp Av-Nt Tp-Nt
Min Max l l l l Aluminium 2676 7745 4192 ± 1 1 78 ••• ... ... N
Table 3.26 also illustrates the vast differences between mound elements and particle sizes
between species and sites. For example. the minimum iron content in the mounds studied
was 782 mg/IOOg and the maximum 6519 mg/IOOg (more than 8 times higher). The
results of the probability of differences (ANOVA) between soils of different sites
indicated highly significant differences (P<O.Ol) for all the variables tested (Tables 3.27-
3.28).
The probability of differences between soil and termite mounds by location and species
is given in Table 3.29 together with the percentage differences between the soil and
mound content
TABLE 3.27 Selected elements (mg/100g) and particle size (%) (mean ± standard deviation) of soil samples (0-!0cm) collected at all the site studied: I to 7.
The elements mean content of the soil compared to the mounds and the finer particle
sizes (clay and silt) were generally lower (but not necessary statistically) in most of the
soils studied. An example is given in Figure 3.14 for calcium, potassium and sodium.
The finer soil particle fractions were generally higher in the mound while the larger
particle size fractions (fme sand and coarse sand) were usually higher in the soil at any
given site (Table 3.29) (see Figure 3.15 for clay and coarse sand).
TABLE 3.28 Selected elements (mg/lOOg) and particle sizes (%) (minimum, maximum and mean ± standard deviation) of soil (0-lOcm) sampled at sites 1 to 7, together with the probability (P) of differences between soils.
Element/ Sites 1 to 7 p Particle size Min Max Mean ± SD l
Aluminium 1385 5508 2771 ± 1033 •••
Calcium 4.06 69.1 18.1 ± 19.2 •••#
Cobalt 0.19 0.97 0.38 ± 0.17 ...
Copper 0.37 1.51 0.69 ± 0.28 ...
Iron 505 7371 2219 ± 1777 ...
Potassium 24.0 943 385 ± 260 ...
Magnesium 20.5 201 75.7 ± 52 ...
Manganese 1.66 12.9 5.52 ± 3.29 ..
Sodium 3.75 23.6 1 1 .54 ± 5.73 ...
Zinc 0.20 7.73 1.50 ± 0.76 ..
Clay 6.21 22.05 12.9 ± 4.4 ...
Silt 4 . 1 1 27.17 1 1 .8 ± 6.5 ...
Fine sand 26.34 58.37 41.7 ± 7.6 •••
Coarse sand 20.18 54.09 35.7 ± 9.8 ...
t *: O.Ol<P<0.05; U; O.OOI<P<O.Ol ; •••: P<O.OOI
TABLE 3.29 Probability of differences (P) between soil and termite mound by location and species.
Element/ P t P t P t P t P t P t P t P t P t P t P t Particle size Site I Site 1 Site 2 Site 3 Site 3 soil Site 4 Site 4 Site 5 Site 6 Site 6 Site 7
soil - Tp soil - Th soil - Av soil - Tp • Nt soil - Av soil - Nt soil - Av soil - Tp soil - Nt soil - Nt
Clay - ... - ... . N - ... . .. . N - ... . N - ... - ... • •
Silt . N . N • • . N . .. • • N .. N . N N
Fine sand N N N .. ... N ... . N • • . N
Coarse sand • . N N N N .. . N ... ... .. . •
t: N: P>0.05; *: O.Oi<P<0.05; **: O.OOI<P<O.OI; ***: P<O.OOl -: indicates that the mean of the soil content is lower than the mean of the mound content (but not necessarily statistically). � Abbreviations: Tp = Tumulitermes pastinator; Th = Tumulitermes hastilis; A v = Amitermes vitiosus; Nt = Nasutitermes triodiae .... "'
150
1 50 l
8 1 oo J I 0 � ' "' E � "' 50
0
0
1 000
-Ol 0 0 � ' Ol E �
"
BOO
600
400
200
0
40
0, 30 8 �
ci, 20 s � 1 0
0
Site 1 -
Site 5
§ I I I� Site 7 Site6
Tp Th S Av S Tp Nt S Av Nt S Av S Tp Nt S Nt S -
Site 3
Site 4
Site 1 Site 2 Site 7
Tp Tb 5 Av S Tp Nt S Av Nt S Av S Tp Nt S Nt S
Site 7 Site 4 .--
Site 1
Tp Th Av S Tp Nt S Av Nt S Av S Tp Nt S Nt S
SPECIES I SOIL
FIGURE 3.14 Soil�mound effects (mean ± SE) on calcium, potassium and sodium (mg/lOOg) in mounds of different species and soils (0-lOcm) sampled at different locations.
S = Soil; Species abbreviations as indicated previously.
30
20 .. ->"'
0 10
0
60 50 � 0>
0 0 40 -' 0> 5 30 "0 c 20 "' Cf.! 0 10
0
- Site 6
Site 3 Site 4 Sit• ' I� Site 7
T
Site 1 Sitel
T al T T
J.
Ill <=! Ell I �II§
Tp Th S Av S Tp Nt S Av Nt S Av S Tp Nt S Nt S
SPECIES I SOIL
Site 5 Site 1 Site 2
Site 7
Site 6
Tp Th s A• s Tp Nt s A• Nt s A• s Tp No s No s
SPECIES I SOIL
FIGURE 3.15 Soil-mound effects (mean ± SE) on clay and coarse sand (%) in mounds of different species and soils (0-lOcm) sampled at different locations.
S = Soil; Species abbreviations as indicated previously.
151
152
3.3 Hot Water ("Infusion") Extractable Selected Elements from Amitermes
viliosus Mounds (Elliott, Site 5).
Eleven mounds (top and bottom sections), from Elliott (Site 5), were analysed for
selected elements following the method described in chapter 2 (2.5.1). The "infusion"
analyses were performed in duplicate, using two separate samples for each position. The
infusion's results for selected elements were compared to those obtained following acid
(perchloriclnitric acids) extraction (chapter 2, 2.2.4.2) performed on the same mounds
(I I mounds, top and bottom sections).
3.3.1 Comparison of Hot Water ("Infusion") Element Extracts and
Perchloric/Nitric Acid Extracts.
The concentration (mean ± standard) deviation of selected elements of Amitermes
vitiosus mounds and soils (Elliott, Site 5), extracted with hot water ("infusion") and acid
extraction (perchloric/nitric acids), together with the percentage recovery between
extractions, are presented in Table 3.30.
The results show that three of the selected elements were not detected in the "infusion"
extracts: cobalt, copper and zinc; and three other elements: aluminium, iron and
manganese had a very low percentage of recoveries: <0.3 %. The remaining four
elements: calcium, potassium, magnesium and sodium had percentage recoveries of 6.3 7
%, 6.33. %, 2.56 % and 7.31 % respectively. In the soil extracts, the recoveries were
lower for calcium, potassium and magnesium (1 .95 %, 1.84 % and 0.94 % respectively)
but higher for sodium (9 .3 % ).
153
TABLE 3.30 Selected element concentrations (mean ± standard deviation in mgllOOg) following hot water ("infusion'') and acid (perchloric/nitric acids) extractions of eleven Amitermes vitiosus mounds (Elliott, Site 5) and soils, together with the percentage recovery between extractions.
Element Mounds
"Infusion" Acid % n=44 n=22 recovery
Aluminium 0.56 ± 0.59 3664 ± 396 0.02
Calcium 6.69 ± 3.21 105 ± 21 6.37
Cobalt nd 0.30 ± 0.05 nd
Copper nd 0.91 ± 0.08 nd
Iron 0.10 ± 0. 1 1 1738 ± 156 0.01
Potassium 10.3 ± 7.83 162 ± 12 6.33
Magnesium 2.37 ± 1.55 93 ± 7.1 2.56
Manganese 0.02 ± 0.02 7.45 ± 1.96 0.23
Sodium 0.47 ± 0.54 6.49 ± 0.73 7.31
Zinc nd 1.08 ± 0.09 nd nd: not detected; detection limit (mg/JOOg): Co and Zn • 0.02;
3.3.2 Position Effects on Selected Elements
"Infusion" n=2
1.01 ± 0.24
0.92 ± 0.07
nd
nd
0.13 ± 0.01
1.78 ± 0.1 1
0.45 ± 0.03
o.oz ± 0.00
0.38 ± 0.02
nd Cu .. 0.01
Soil
Acid n=2
1847 ± 330
47 ± 1.9
0.27 ± 0.01
0.68 ± 0.09
1 154 ± 128
97 ± 12
48 ± 5.9
6.37 ± 0.66
4.03 ± 0.01
0.59 ± 0. 1 1
% recovery
0.05
1.95
nd
nd
O.oJ
1 .84
0.94
0.27
9.34
nd
The comparison between the two types of extractions with respect to position (top and
bottom), using ANOVA, is presented in Table 3.31. A significant increase in calcium
in the top section of the mounds is observed following the two types of extractions. A
highly significant increase in iron in the bottom of mounds after hot water extraction
may be attributed to the very low iron concentrations.
154
TABLE 3.31 Position effects on selected elements (mean ± standard deviation in mg/lOOg) in eleven Amitermes vitiosus mounds (Elliott, Site 5) following two types of extraction: hot water ("infusion") extraction and perchloric/nitric acids extraction. ANOV A probability of differences (P) between positions.
Mineral
Top n""22
"Infusion"
Bottom n=22
p �
Aluminium 0.39 ± 0.47 0.73 ± 0.64 N
Calcium 8.06 ± 3.42 5.33 ± 2.35 ••
Cobalt nd nd nd
Copper nd nd nd
Iron 0.05 ± O.o7 0.16 ± 0.1 1 •••
Potassium 12.2 ± 9.09 8.34 ± 5.93 N
Magnesium 2.74 ± 1.27 2.01 ± 1.75 N
Manganese 0.02 ± 0.03 0.02 ± 0.02 N
Sodium 0.62 ± 0.70 0.33 ± 0.24 N
Zinc nd nd nd
t: N: P>O.OS; *: O.Oi<P<0.05; ••: O.OOI<P<O.Ol ; •••: P<O.OOI nd: not detecll:d; detection limit (mg/IOOg): Co and Zn:0.02; Cu: 0.01
Perchloric/nitric extraction
Top n=11
Bottom n=l l
3829 ± 392 3526 ± 359
1 16 ± 17 95 ± 20
0.31 ± 0.05 0.30 ± 0.05
0.94 ± 0.08 0.89 ± O.o7
1 775 ± 160 1701 ± 141
167 ± 12 159 ± 12
97 ± 6.1 89 ± 6.1
7.77 ± 2.01 7.21 ± 1.89
6.67 ± 0.80 6.35 ± 0.63
1 . 12 ± 0.08 1.04 ± 0.08
p � N
•
N
N
N
N •
N
N •
3.4 Soluble Iron, Ionisable Iron and Selected Element Concentrations of
Termitaria and Soils Following Pepsin-Hydrochloric Acid Incubation.
The soluble iron, ionisable iron together with the elements AI, Ca, Co, Cu, K, Mg, Mn,
Na and Zn analyses were performed in triplicate on selected mound and soil samples
(Table 2.1 1) at each site (I to 6) according to the method described in chapter 2.6.2.
3.4.1 Pepsin Concentration Effects on Soluble Iron, Ionisable Iron and Selected
Element content following Pepsin-HCI Acid Incubation.
The results of variation of pepsin concentration (0-0.5 %) in the pepsin-HCl incubation,
on soluble iron, ionisable iron and selected element concentrations (mg/l OOg), together
with the probability of differences between pepsin concentration, are given in Table
3.32.
•
TABLE 3.32 Selected element composition (mg/IOOg) of termitaria reference material® (Nt26D4), following O.IN HCI acid (pH 1.35) extraction with different pepsin percentage v/w (0%, 0.1 o/o and O.So/o), together with the probability P(t) of differences between pepsin concentrations.
Element Pepsin-HCI pH 1.35 extract
0% 0.1% 0.5% P(t)
Aluminium 20.3 ± 0.9 19.8 ± 0.9 19.7 ± 0.9 N
Calcium 33.6 ± 1.0 33.7 ± 0.7 33.7 ± 0.4 N
Cobalt 0.01 ± 0.01 0.01 ± 0.00 O.ot ± 0.00 N
Copper 0.08 ± 0.01 0.09 ± 0.02 0.08 ± 0.01 N
Iron 1 1 .5 ± 0.2 1 1.3 ± 0.3 1 1.2 ± 0.3 N
Potassium 65.8 ± 3.1 65.5 ± 2.9 64.4 ± 1.9 N
Magnesium 44.4 ± 1.4 44.0 ± 0.7 43.6 ± 1.3 N
Manganese 2.39 ± 0.10 2.35 ± 0.09 2.32 ± 0.08 N
Sodium 7.65 ± 0.41 7.43 ± 0.28 7.34 ± 0.24 N
Zinc 0.08 ± O.ot 0.09 ± 0.02 0.08 ± 0.02 N
Iron(II) 8.12 ± 0.27 8.00 ± 0.32 8.04 ± 0.33 N
@: for tennitana material number explanations see chapter 2.1.3. t: N: P>0.05; •: 0.01 <P<0.05 nd: not detected; cobalt and zinc detection limit: 0.02 mgfl OOg.
0%
2.05 ± 0.53
28.7 ± 1.2
nd
0.03 ± 0.00
126 ± 0.27
63.3 ± 1.8
38.2 ± 1.2
1 .89 ± 0.07
0.02 ± 0.02
0.34 ± 0.15
pH 7.5 filtrate
0.1% 0.5%
2.49 ± 0.58 2.90 ± 0.61
28.1 ± 1.6 27.8 ± 1.3
nd nd
0.03 ± O.ot 0.03 ± 0.00
1.49 ± 0.37 1.69 ± 0.37
62.7 ± 2.3 62.4 ± 2.0
37.2 ± 2.2 35.8 ± 2.1
1.82 ± 0.10 1.77 ± 0.12
0.02 ± 0.01 nd
0.38 ± 0.12 0.67 ± 0.92
P())
•
N
N
•
N
•
N
N
� "' "'
156
25.0
� 200 � ' _§ 15.0
f--ctl 1 0 0
2 w _j w 5.0
Pepsjn-HCI Extract fpH 1.35)
0.0 _c_ __ _ 0.0 0. 1
% PEPSIN
§ Aluminium
ISl Soluble iron
� lonisable iron
0.5
FIGURE 3.16 · Pepsin concentration effects on aluminium, soluble iron and ionisable iron (mean ± SE in mg/lOOg) in Nasutitermes triodiae mound (Nt26D4, Daly River, Site 4) following Pepsin-HCI pH 1.35 extraction (n=9).
3.0 pH 7.5 Filtrate
� 2 0 � ' L � Aluminium I � ISl Soluble iron
!:11 lonisable iron 5 f--z w 1.0 2 w _j w
0.0 0.0 0.1 0.5
% PEPSIN
FIGURE 3.17 Pepsin concentration effects on aluminium, soluble iron and ionisable iron (mean ± SE in mg/IOOg) in Nasutitermes triodiae mound (Nt2604) in pH 7.5 filtrates (n=9).
I
I ,
157
The pepsin concentration does not significantly (P>0.05) affect the amount of soluble
iron, ionisable iron and selected elements present in pepsin-HCl acid (pH 1.35) extracts
(see Table 3.32 and Figure 3.!6). In pH 7.5 filtrates. there is a significant
(0.01 <P<0.05) increase in alwninium and soluble iron for the 0.5 % pepsin concentration
(see Figure 3.17) and a significant decrease in magnesium for the same pepsin
concentration. Although the ionisable iron concentration mean is higher in the pH 7.5
filtrate from the 0.5 % pepsin-HCl extract, there is no significant difference. Likewise,
no significant difference has been observed in the other selected elements. In view of
the results obtained (significantly higher soluble iron concentration and higher ionisable
iron mean) a 0.5 % w/v solution of pepsin was used for all incubations as in the in vitro
test for predicting the bioavailability of iron in foods122•
3.4.2 Quality Assurance
The results of the quality assurance are shown in Table 3.33. The internal reference
sample (Nt26D4) was run in duplicate with every batch of samples. The precision of
the method is indicated by the selected element standard deviations (n=40) for the internal reference sample mean contents. From the means of the pepsin-HCl extracts,
the standard deviations are below 5 % for AI, Ca, Fe, K, Mg, Mn, Na, Fe(II). The mean
concentrations of Co, Cu and Zn are very low, 0.02 to 0.07 mg/IOOg and their standard
deviations were 50 %, 14 % and 40 % respectively. The standard deviations from the
mean concentration of selected elements (including soluble iron and ionisable iron) of
pH 7.5 filtrates are much higher.
158
TABLE 3.33 Internal quality control of selected elemental composition (mgllOOg) of pepsin-HCI acid (pH 1.35) extracts and pH 7.5 filtrates of termitaria sample (Nt26D4).
Element Internal reference material: Nt26D4® n=40
Pepsin-HCI
Aluminium 20.6 ± 0.8
Calcium 34.3 ± 1.6
Cobalt 0.02 ± 0.01
Copper O.D7 ± 0.01
Iron 1 1.3 ± 0.5
Potassium 64.1 ± 1.4
Magnesium 46.2 ± 1.3
Manganese 2.39 ± 0.06
Sodiom 7.42 ± 0.31
Zinc 0.05 ± 0.02
Fe(ll) 9.1 7 ± 0.41
pH 7.5 filtrate
2.36 ± 1.12
31 .0 ± 1.8
<0.02
0.03 ± 0.01
1.01 ± 0.35
61 . 1 ± 4.4
42.4 ± 2.6
1.77 ± 0.28
<0.02
0.30 ± 0.1 1 @:lor explanatiOn of temutana reference matenill number reler to chapter :t.J.I
159
3.4.3 Soluble Iron, lonisable Iron and Selected Element Composition of Pepsin
HCI Acid (pH 1.35) Extracts and pH 7.5 Filtrates.
The concentration mean (± standard deviation) of selected elements of Amitermes
viliosus, Tumulitermes pastinator, Tumulitermes hastilis and Nasutitermes triodiae
mounds and soils at different sites (I to 6), in 0.5 % (w/v) pepsin - O.IN HCI acid
(pH 1.35) extracts and pH 7.50 filtrates are given in Tables 3.34 to 3.40, together with
the concentration mean (± standard deviation) of selected elements following
perchloric/nitric acid (4:1) extraction and the percentage recoveries between treatments.
Due to the inherent large degree of variability associated with the exchangeable method
used (O.lN HCl) and the small number of replicates, comparative data between different
species at the same site and same species at different sites have not been statistically
presented, as nearly all comparisons showed no significant differences. Instead,
graphical comparisons are given in Appendices Vl and VII.
3.4.3.1 Soluble Iron, Ionisable Iron and Selected Element Comparisons of
Pepsin-HCI Acid pH 1.35 Extracts, pH 7.5 Filtrates and
Perchloric/Nitric Acid Extracts.
A) Soluble Iron and Ionisable Iron
The concentration mean (± standard deviation) ofperchloric/nitric extractable iron, total
soluble and ionisable iron following pepsin-HCI acid incubation, together with the
percentage of ionisable iron in soluble iron are given in Table 3.34.
As seen in Table 3.34, very little of the iron present in the perchloric/nitric extracts is
released following pepsin-HCl (pH 1.35) incubation. The percentage of soluble iron in
the pepsin-HCl (pH 1 .35) extracts compared to the perchloric/nitric iron varied in
mounds from 0.17 % in Tumu/itermes pastinator at site 6 to 1 1 % in Amitermes vitiosus
at site 4 and in soils from 0.10 % at site 6 to 3.2 % at site 4.
TABLE 3.34 Soluble iron and ionisablc iron content (mean± standard deviation in mg/lOOg) oftermitaria and soils, in perchloric/nitric extracts, pepsin-hydrochloric (pH 1.35) extracts and pH 7.5 filtrates together with the percentage recovery between
Site
Site 1
Site 2
Site 3
Site 4
Site 5
Site 6
ionisablc iron and soluble iron in pepsin-HCI extracts and pH 7.5 filtrates. Depth= I, n=3 unless indicated
Abbreviations: nd: not detected; detection limit in mg/IOOg: ' 0.06; iron(ll) 0.20 ; Nt:Nasutitermes triodiae I fOil
%
26
39
I I I I
59 21
32
24
� "' Q
161
The highest ionisable iron contents in pepsin-HCI extracts were found in mounds where
the concentrations ranged from 6.22 mg/1 DOg in Tumulitermes pastinator mounds at site
6 to 74.7 mg/lOOg in Amitermes vitiosus mounds at site 4; while in the soil samples the
ionisable iron ranged from 0.84 mg/JOOg at site 5 to 6.48 mgiiOOg at site 4. In either
case (soluble and ionisable iron) the highest content was found in Amitermes vitiosus
mounds at sites 2 and 4.
As the pH increased from 1.35 in the pepsin-HCl extracts to 7.5 in the filtrates, both the
ionisable and soluble iron decreased. However, the decrease in the ionisable iron was
of a greater magnitude. The ionisable iron contents were very low, ranging in the
mounds, from 0.08 to 0.58 mg/lOOg. Even in Amitermes vitiosus mounds at site 4,
where previously both soluble and ionisable iron were the highest, in pH 7.5 filtrates,
the level was comparable to the other species' mound levels. In soils, at all sites, no
soluble iron nor ionisable iron was detected in pH 7.5 filtrates (for example, soil at site
4: see Table 3.34 and Figure 3.18).
162
1 00
80
f-z 60 w 0 8i 40 Q_
20
perchlortc/njtdc
I f'SW Rr,. 0 Alt"'e �• "' "' • A
1 00
Extraction Procedures: Pepsjn-HCJ
pH 1.35 pH 7 5 filtrate
PerchloricJnjtric Pepsin-HC! pH 1 35
pH 7.5 filtrate 80
f-z 60 w 0 8i 40 Q_
B
20
Q AI CaFe KMg
1 00 .
AI CaFe KMg /\I CaFe KMg
� Aluminium
• Calcium
0 Iron
bl Potassium
113 Magnesium i
Perchlortctnjtric Pepsjn-HCI pH 1.35
pH 7.5 filtrate 80
f-z 60 w 0 8i 40 Q_
20 � Q AJ Ca Fe KMg u � 1.
AI CaFe KMg AI CaFe KMg
C ELEMENT
FIGURE 3.18 Selected element percentage variations between perchloric/nitric acid extracts, pepsinHCI acid pH 1.35 extracts and pH 7.5 filtrates oftennitaria (A: Amitermes vitiosus; B:
Nasutitermes triodiae and soil (C), from Daly River site 4.
TABLE 3o35 Comparison of selected element concentration (mean ± standard deviation in mg/100g) of termite mounds (Tumulitermes pastinator and Tumulitermes hastilis) and soils (0-10cm), sampled from Daly River (site 1) in: A- pepsin-HCI acid (pH 1.35) extracts; 8- pH 7.5 filtrates and C- perchloric/nitric acid (4:1) extracts, together with the % recovery between treatments.
Species Method/ Element ± standard deviation mgllOOg
% recovery Aluminium Calcium Cobalt Copper Iron Potassium Magnesium Manganese Sodium Zinc Iron II
Abbreviations: Tp = Tumulitermes pastinator; Th = Tumulitermes hastilis; nd = not detected Detection limit (mg!IOOg): Co and Zn = 0.02; Cu = 0.01; Fe = 0.06; Fe(II) = 0.20
nd 0.67 ± 0.16 0.16 ± 0.05 -- l30 42
348 ± 26 31.2 ± 0.5 3.36 ± 0.24 15.9 ± 2.0
0.66 1.7 I I 3.5
0 2.2 4.6
nd
-0.44 ± 0.08
23
0.54 ± 0.02
3.3
0.06 ± 0.06 2.11 ± 0.18
nd nd
0.40 ± 0.01
1 5
� "' ...
TABLE 3.37 Comparison of selected element concentration (mean ± standard deviation in mg/100g)) of termite mounds (Tumulitermes pastinator and Nasutitermes triodine) and soils (Q-10cm), sampled from Daly River (site 3) in: A- pepsin-HCI add (pH 1.35) extracts; 8- pH 7.50 filtrates and C- perchloric/nitric acid (4:1) extracts together with the % recovery between treatments.
Species Method/ Element ± standard deviation mgllOOg
%recovery Aluminium Calcium Cobalt Copper Iron Potassium Magnesium Manganese Sodium Zinc Iron II
Abbreviations: Tp = Tumulitermes pastinator; Nt = Nasutitermes triodiae; nd = not detected ~
'"' Detection limit (mgllOOg): Co and Zn = 0.02; Cu = 0.01; Fe= 0.06; Fe(ll) = 0.20 "'
-"' TABLE 3.38 Comparison of selected element concentration (mean ± standard deviation in mg/100g) of termite mounds (Nasutitermes triodiae) Q\
and soils (D-10cm), samples from Daly River (site 4), depth"" 0 (new/old material), in: A- pepsin-hydrochloric acid (pH 1.35) extracts; 8- pH 7.50 filtrates and C- perchloric/nitric acid (4:1) extracts together with the% recovery between treatments. Probabilities (P) of differences between ages (new/old).
Species/ Method/ Elementl ± standard deviation mg/IOOg
P()) % recovery Aluminium Calcium Cobalt Copper lro" Potassium Magnesium Manganese Sodium Zinc Iron II
Abbreviations: Nt = Nasutitermes triodiae; nd = not detected; P = probability of differences between material ages (new/old) Detection limit (mg/lOOg): Co and Zn = 0.02; Cu = 0.01; Fe(ll) = 0.20 P(t): N: P>0.05; *: 0.01< P<O.OS; u: O.OOI<P<O.Ol; ***: P<O.OOI
21.5 ± 5.4
20
N
N
0.02 ± 0.02 0.48 ± 0.11
15 25
0.43 ± 0.04
24
3.7
N
N
N
N (0.083)
N (0.202)
TABLE 3.39 Comparison of selected element concentration (mean ± standard deviation in mg/100g) of termite mounds {Amitennes vitiosus and Nasutitermes triodiae) and soils ({).10cm), samples from Daly River (site 4), in: A· pepsin-HCI acid (pH 1.35) extracts; 8- pH 7.50 filtrates and C- perchloric/nitric acid (4:1) extracts, together with the% recovery between treatments.
Species Method/ Element ± standard deviation mg/1 OOg
%recovery Aluminium Calcium Cobalt Copper !roo Potassium Magnesium Manganese Sodium Zinc Iron II
Abbrev~atlons: A' Amitermes vmosus; Nt Nasu/ltermes tridwe; od not detected Detection limit (mg/100g): Co and Zn = 0.02; Cu = 0.01; Fe= 0.06; Fe(ll) = 0.20 ~
"' ....
TABlE 3.40 Comparison of selected element concentration (mean ±standard deviation in mg/100g) of termite mounds (Tumulitermes pastinator ~ and Nasutitermes triodiae) and soils (D-10cm), samples from Howard Springs (site 6), in: A- pepsin-HCI (pH 1.35) extracts;: 8- pH 7.50 QO
filtrate and C- perchloric/nitric acid (4:1) extracts, together with the% recovery between treatments.
Species Method/ Element ± standard deviation mg/IOOg
%recovery Aluminium Calcium Cobalt Copper Iron Potassium Magnesium Manganese Sodium Zinc Iron II
Abbreviations: Tp = Tumulitermer partinator; Nt = Narutitermer triodiae; nd = not detected Detection limit (mg/IOOg): Co and Zn = 0.02; Cu = 0.01; Fe= 0.06; Fe(II) = 0.20
169
B) Selected Elements
As Tables 3.35 to 3.40 show, very little of the predominant elements (aluminiwn and
iron) present in the mounds and soil, at all sites, were released with pepsin-HCI acid at
pH 1.35. The percentage of recoveries of aluminium in the pepsin-HCI acid extracts
compared to the perchloric/nitric acid extracts were very low, less than 2 % for all
species, at all sites. The highest recoveries in the pepsin-HCl extracts of mounds,
compared to the perchloric/nitric acid extracts, were for calcium (greater than 78 % for
all species, at all sites).
The recoveries varied with respect to the origin of the sample. Generally, the percentage
recoveries with pepsin-HCl extracts for calcium, potassium, magnesium, manganese and
sodium were lower in the soil. This trend was not observed between mounds and soils
sampled in Elliott, where the percentage recoveries were similar for calcium, potassium,
magnesiwn and sodiwn (Table 3.36).
When the pH was increased from 1.35 to 7.5, the amount of aluminium, cobalt, copper,
soluble iron, ionisable iron and zinc decreased dramatically for all species, at all sites.
The cobalt, copper and zinc contents in the pH 7.5 filtrates were close to the detection
limit or not detected at all. Very little of the calcium, potassium and magnesium were
lost after neutralisation to pH 7.5, the percentage recoveries of these elements remained
between 80-100% for all species and at all sites (Table 3.35 to 3.40). The manganese
recoveries in the pH 7.5 filtrates compared to pepsin-HCl extracts were more variable
with an average of 65 % for all species termitaria and 49 % for all soils. Here again
as in the pepsin-HCl extracts, the percentage recoveries for calcium, potassium,
magnesium and manganese were generally lower in the soil than in the mounds, as
shown in Figure 3.19 for Nasutitermes triodiae, Tumulitermes pastinator and soil from
site 3.
170
100 D Nt
80 j ~ I bl Tp >- Ill! Soil a: w 60 > 0 () w 40 a:
"' 20
0 Calcium PotaSSIUm Magnesium Manganese
ELEMENT
FIGURE 3.19 Selected elements (calcium, potassium, magnesium and manganese) percentage recovery in pH 7.5 filtrates of Nasutitermes triodiae (Nt) mounds, Tumulitermes pastinaror (Tp) mounds and soils·at site 3.
171
Figure 3.18 shows an example of the change in the percentage distribution of aluminium,
calcium, iron, potassium and magnesium in perchloric/nitric extracts, pepsin-HCl extracts
and pH 7.5 filtrates. At all sites and for all species, the aluminium and iron were the
dominant elements in the perchloric/nitric extracts. In the pH 7.5 filtrates, the 3
dominant elements were calcium, potassium and magnesium. Their relative percentages
varied according to the type of samples. For example, in the pH 7.5 filtrates, at site 4
calcium is the dominant element in Amitermes vitiosus mounds while potassium is the
dominant element in Nasutitermes triodiae mounds and soils.
3.4.3.2 Age Effects on Selected Elements (Depth=O).
As shown in Table 3.38, the results of the AN OVA between ages of mound material
(old and new), indicate a significant increase (0.01 <P<0.05) in the new material for
soluble iron in the pepsin-HCl (pH 1.35) extract and a significant decrease for iron
following the perchloric/nitric acid extraction. A smali ionisable iron increase (P=0.083)
in the new material was detected in the pepsin-HCI extract. Although highly significant
decreases were observed in new material for aluminium and copper in the
perchloric/nitric extracts, no other differences were observed in the pepsin-HCl extracts
and pH 7.5 filtrates.
3.4.3.3 General Oveniew of Different Species Studied at Different Sites with
Relation to the Adjacent Soil (0-lOcm).
The minimum, maximum and mean ± standard deviation of selected element contents
(mg/1 OOg) of the species of mounds. chosen by the Aboriginal communities, and the
adjacent soils (0-lOcm) sampled at sites I to 6, in pepsin-HCl extracts and pH 7.5
filtrates are given in Tables 3.41 and 3.42 respectively.
172
TABLE 3.41 Selected elements (mg/100g) (minimum, maximum and mean ± standard deviation) of Amitermes vitiosus, Tumulitermes pastinator and Nasutitermes triodiae mounds and soils (O~lOcm) sampled at sites 1 to 6, following pepsin-HCl incubation (pH 1.35).
Abbreviations: Av= Amitermes vitiosus; Tp= Tumulitermes pastinator; Nt= Nasutitermes triodiae nd= not detected; detection limit (mgflOOg): Co and Zn c 0.02
The selected element content differences (%) between the soil and the termite mounds
at a given site in pepsin-HCI extracts and pH 7.5 filtrates are given in Table 3.43. In
pepsin-HCI extracts, the element mean contents of the soil compared to the mounds were
generally lower in most of the soils studied, with the exception of aluminium where they
were higher in some cases. In pH 7.5 filtrates, the element mean contents of the soil
compared to the mounds, when detected, were always lower.
173
TABLE 3.42 Selected element content (mgllOOg) (minimum, maximum and mean ± standard deviation) ofAmitermes vitiosus, Tumulitermes pastinator and Nasutitermes triodiae mounds and soils (0-lOcm) sampled at sites 1 to 6, in pH 7.5 filtrates.
3.25) and Tumulitermes pastinator mounds 37.3 ± 7.4 mgllOOg (Table 3.24). The same
relationship was found for the maximum mean potassium concentrations, amongst all the
sites studied, the potassium content was the highest in the soil, in Nasutitermes triodiae
mounds and Tumulitermes pastinator mounds at site 3 (Daly River): 694 ± 158, 897 ±
67 and 651 ± 79 mg/lOOg respectively. The same relationship between soil and mounds
was found for the other selected elements (Tables 3.26 and 3.28).
The selected mean element cgncentrations and the fmer soil particle size (clay and silt)
were generally higher (but not always statistically) in the termite mounds studied than
in their adjacent (0-10 em) soils (Table 3.29 and Figures 3.14 and 3.15). At all the sites
studied, the termite mounds had a higher clay content than their surrounding soils
(Figure 3.15 and Table 3.29); the increases were always sigoificantly or highly
significantly different in Tumulitermes pastinator and Nasutitermes triodiae mounds but
no significant differences were found between Amitermes vitiosus mounds and their
adjacent (0-10 em) soil. The majority of studies have shown an increase in clay content
in mounds in comparison with unmodified soils (section 1.2.3.1-B). In Australia, with
the exception of one site where the clay content of the (0-1 0 em) soil horizon was
exceptionally high (53 %), the clay content was always higher in the mounds of the
three species (Amitermes vitiosus, Tumulitermes pastinator and Nasutitermes triodiae)
than in the adjacent (0-10 em) soils (Table 1.3). Lee & Wood (197lb)98 and Holt eta/
(1980)71 reported that the level of clay content in the mounds resembles the content of
deeper horizons which could have levels even higher than in the mounds (Table 1.3).
This finding is important in relation to the medicinal use of termite mounds as the
193
Aborigines prefer termite mounds instead of surface soil. It is easier to take a piece of
termite mound than to have to dig to the subsoil to obtain the same amount of clay,
particularly during the dry season when the soil becomes rock hard. However, during
the wet season, when the termite mound sites are flooded and inaccessible, the Daly
River Aborigines dig the flooded soil close to the mission to collect some subsoil which
has a clay content higher than the monnds (30.2 ± 0.7 %clay, hydrometer method).
According to Nye (1955)134 no general agreement about the differences between the
chemical composition of the mounds and the surrounding soil has been found~ Most
studies throughout the world and in Australia showed an increase in the chemical
composition (organic carbon, nitrogen and exchangeable bases (calcium, magnesium))
of the termite mounds compared to the adjacent soils77•98
•102
• The many factors
responsible for the increase in element concentrations in the mounds have been discussed
in section 1.2.4. However according to a munber of authors (section I .2.4) not all
mounds have a higher element content than their surrounding soils.
The results of this study together with literature show that although the termite mound
composition reflects the composition of the adjacent soil (0~ lOcm), the selected element
content in the monnds was generally higher than in the surronnding top soil (Table
3.29). Nine soil~species pairs out of eleven were significantly higher in aluminium, iron,
magnesium and sodium, eight soil~species pairs were significantly higher in calcium and
seven pairs were significantly higher in potassium and zinc. The increase of element
concentrations in mounds has been attributed to a number of factors (section 1.2.4) such
as the use of richer sub-soil by termites and the incorporation of vegetation, saliva and
excreta (in parts of mounds). The increase could further support the fact that the
Aborigines prefer taking soil from termitaria in preference to the adjacent soil (0~ lOcm).
There are of course other possible reasons for selecting tennitaria, such as, the belief that
soil processed by animals is considered to be safer than that which is not processed100,
194
4.3 Hot Water (''Infusion") Extractable Selected Element Concentrations from
Amitermes vitiosus Mounds (Elliot~ Site 5)
The purpose of the hot water "infusion" extraction was to obtain a measure of the
concentration of selected element present in the "termite mound tea" drunk by the
Aboriginals of Elliott. Potassium, calcium and magnesium were the three principal
elements extracted following hot water "infusion" from mounds with: 10.3 ± 7.83, 6.69
± 3.21 and 2.37 ± 1.55 mg/lOOg respectively (Table 3.30). These concentrations
represent a very small fraction of the nitric/perchloric acid extract: 6.33, 6.37 and 2.56
%. The extraction of these elements was even lower from adjacent (0-10 em) soil
material: 1.84, 1.95 and 0.94 % for potassium, calcium and magnesium respectively.
This higher concentration could indicate that the selected elemental increase in termitaria
"Infusion" extract (Table 3.29) may be of a more available nature as it may come, at
least partially, from tennite by-products (for example, saliva and excreta) which are
more bioavailable. Potassium was the dominant element in the "infusion", and as
mentioned in section 4.2.1, it could be in a different form (more soluble) in Elliott
mounds and soils than at any other sites. The influence of the sample position in the
mound on the selected element concentrations was comparable to the one observed in
the nitric/perchloric extract analyses in regards to the calcium (section 4.2.5 and Table
3.31). A higher calcium concentration was observed in the top section of mounds. A
highly significant increase in iron in the bottom of mounds, after hot water extraction,
may not be very relevant as the iron concentrations were very low.
In comparison to the human recommended dietary intakes (see Table 1.2), the
concentration of selected elements extracted are minimal (Table 3.30) but, nevertheless,
could contribute to the global intakes. For example, if all the calcium present in the
"infusion" was available to the human body, 1.5 to 2.2 litres of "infusion" would be
necessary to cover the daily calcium losses. But in nonnal food, the percentage
absorption of calcium is around 20 %125; therefore, at least five times more 11infusion"
may be needed.
195
Amongst other usages, the "infusion" is given for gastro-intestinal disorders, such as,
diarrhoea. It could help to restore body fluid and a fraction of body salts lost with
diarrhoea fluids and thus prevent dehydration. On the other hand, as the finer fraction
of the soil (clay and silt) is preferentially selected in the infusion, with the heavier
fractions, fine sand and coarse sand, sinking more easily to the bottom and therefore not
drunk, it may possibly be a more selective and pleasant way of eating clay. The heavier
fraction is recycled as poultice and the combination of both ("infusion" and poultice) are
used to bring up milk after birth. In a hot and harsh environment, smearing the poultice
on the chest and back provides a cooling effect due to evaporation and may benefit the
nursing mother, but also, drinking and relaxing while being surrounded by the attention
of family members, could help in releasing oxytocin, (hormone from the pituitary gland
which causes contraction of the muscle fiber of the milk glands, forcing the milk into
larger ducts), therefore promoting the "let-down" reflex. This reflex may be inhibited
if the nursing mother is worried or physically uncomfortable183• The poultice could have
other beneficial aspects as mentioned in the popular literature, to cite a few: clay
poultices placed on the lower abdomen for several days before menstruation prevent
pain; clay increases circulation and flow of oxygen to the skin all over the body; and
miraculous cures have occurred in patients, in a Swiss phthisiology centre, who had their
entire thorax coated with clay41• Unfortunately, no scientific explanations have been
offered for these observations.
4.4 Soluble Iron, lonisable Iron and Selected Element Concentrations of
Termitaria and Soils Following Pepsin-Hydrochloric Acid Incubation
As discussed in section 1.2, the total amount of element present in the diet is not
necessarily available to the hwnan body. For example, only 5-20% of dietary iron is
absorbed from the diet; and in general, cereal based food products have low iron
availability with absorption of 1-7 %for rice, com and whole wheat flour125• Most
study in Australia has focused on the effects resulting from termite activities in soils
and on the exchangeable or 'available' elements in relation to biological, ecological and
pedological significance of termite modified material. As the primary objective of the
196
analyses of this study was focusing on human nutrition, and no standard bioavailable
methods were established for all the selected elements, a method which would predict
the bio-availability of iron from food was selected. This method, suggested by
Narasinga Rao & Prabhavathi (1978), simulates the digestive conditions in the stomach
(pepsin-HCl, pH !.35) and the intestine (pH 7.5)122• As most elements are mainly
absorbed in the intestine 183, the method was extended to the other elements selected in
this study; this would reflect the potential nutritional value of termitaria more closely as
it would give a more realistic idea of the selected element content present at the
absorption site.
4.4.1 Quality Assurance
The Fe(II}-bipyridine spectrophotometric analysis had a severe interference due to the
high concentration of ferric iron, Fe(III), present in termitaria, it was not possible to
obtain a reliable measure of bioavailable iron, Fe(JI), as the absorbance of the Fe(JI)
a,a.'-bipyridine complex was observed increasing with time. This type of reaction has
previously been described by Lee eta/ (1948) with Fe(II)-1,10-Phenanthroline complex.
They found that the absorbance of the Fe(ll) complex increased with time because the
Fe(Ill) complex is slowly reduced to the Fe(II) complex. They attributed this reduction
to the higher stability constant of the Fe(!!) complex than the Fe(lll) complex". A
similar type of reaction will occur for the Fe(III) and the Fe(II)-u,u' -bipyridine
complexes with time. By complexing the Fe(lll) as [FeF6]3" with potassium fluoride
prior to the addition of the a,a'-bipyridine reagent, it was possible to obtain a reliable
and constant measure of Fe(II), as no increase of the Fe(II) complex was observed.
The effect of pepsin concentration on the selected element concentration was
investigated. A significant increase of aluminium and soluble iron in the pH 7.5 filtrates
for a 0.5 % w/v solution of pepsin, was observed but it was not followed by a
significant increase of the ionisable iron (pH 7.5) (Table 3.32), nor other differences at
all concentrations tested. Therefore, as suggested by Narasinga Rao & Prabhavathi
197
(1978) a 0.5 % w/v solution of pepsin was used in addition to HCl for all the
extractions.
An internal termite mound reference material was used to monitor possible variations of
the selected element composition during analyses. The results showed that the precision
of the method was very high for the selected elements of the pepsin-HCI extracts (Table
3.33) considering the low concentration of certain elements: cobalt, copper and zinc.
The standard deviation from the mean concentration of selected elements (including
soluble iron and ionisable iron) of pH 7.5 filtrates were much higher for a number of
elements: aluminium, copper, iron, manganese and iron (II). It could be due to the
relatively low concentration of those particular elements and to co-precipitation during
neutralisation to pH 7.5.
4.4.2 Selected Element Comparisons ofPepsin-HCl (pH 1.35) Extracts, pH
7.5 Filtrates and Perchloric/Nitric ExtraCts.
A summary of the selected element concentrations of different termitaria and soils in
pepsin-HCl (pH 1.35) extracts, pH 7.5 filtrates and perchloric/nitric acid extracts is given
in Tables 3.35 to 3.40, together with the percentage recovery between treatments. The
graphical comparisons between mounds of different species at the same site and between
sites are given in Appendices Fl-F2.
High element concentrations in perchloric/nitric acid extracts does not necessarily reflect
their bio-availability. A method that simulates the human digestion (pepsin-HCl pH 1.35
followed by neutralisation to pH 7.5) is more likely to represent the quantity really being
extracted in the human stomach and the quantity available for absorption in the intestine.
Pepsin-HCl extractions are far less rigorous than perchloric/nitric extractions and could
result in variations in the ratios of concentrations of element extracted by the two
different methods between different species of termite mounds -at different sites. Some
elements may be in a "more extractable" form in one species than in another or in one
site than in another (see 4.2.1).
198
4.4.2.1 Influence of Age of Mound Material on Selected Element
Concentrations and Particle Size (Depth=O)
As indicated in section 3.4.2.2, the only significant difference between ages of
Nasutitermes triodiae mound materials at site 4, was in soluble iron in pepsin-HCl
extracts (Table 3.38). Although highly siguificant differences had been observed in
aluminium and copper in perchloric/nitric extracts, no differences were observed in
pepsin-HCl extracts and pH 7.5 filtrates. The constant mound modifications by termites
(rebuilding and extending part of the mound, bringing material from inside to the
outside), could explain the lack of differences in concentration of elements between new
and old parts of termitaria.
While in the perchloric/nitric extracts there was a significant decrease in iron in the new
materials, the opposite was observed for soluble iron in the pepsin-HCI extracts where
a significant increase of soluble iron was found in the new material (Table 3.38). No
significant differences in soluble iron and ionisable iron were observed in pH 7.5
filtrates, although their mean~ were higher in the new materials. In the old material the
soluble iron mean was 0.44 ± 0.42 mg/lOOg while in the new material it was 2.24 ± 1.96 mg/IOOg. The lack of significant differences in soluble iron and ionisable iron in
pH 7.5 filtrates was mainly due to the high standard deviations from the means. In
respect to the preference of Aboriginals for the new parts of the mounds over the old
parts, the soluble iron increase in pepsin-HCl extracts in the new parts of termitaria
could be of importance, as it could be related to the Aboriginal usage during pregnancy
where the iron needs are markedly increased (see section 4.4.3). No other increases in
major elements were detected in pepsin-HCI extracts and pH 7.5 filtrates.
4.4.2.2
199
Comparison Between Different Species Mound Composition at the
Same Site
A) Comparison Between Tumulitermes pastinator and Tumulitermes hasti/is
Mounds Composition in Daly River Site 1.
With the exception of aluminium and sodium mean contents in pepsin-HCI extracts,
which were of the same order in both Tumulitermes pastinator and Tumulitermes hastilis
mounds, the mean concentrations of elements were higher in Tumulitermes hastilis than
in Tumu/itermes pastinator mounds (Table 3.35 and Appendix FI). In particular, the
calcium mean was three times higher in pepsin-HCI extracts and pH 7.5 filtrates; the
soluble iron was 16 times higher in Tumulitermes hastilis mounds (0.13 versus 2.07
mg/1 OOg) and the ionisable iron was not detected in the pH 7.5 filtrates of Tumulitermes
pastinator mounds but had a relatively high concentration of 0.54 mg/1 OOg in
Tumulitermes hastilis mounds. The potassiwn which was highly significantly higher in
Tumulitermes pastinator mounds in perchloric/nitric extracts, was 1.5 times higher in
Tumulitermes hastilis mounds in pepsin-HCI extracts and pH 7.5 filtrates.
The concentrations of selected elements in pepsin-HCl extracts and pH 7.5 filtrates
would not appear to be the reasons why the Aboriginals prefer Tumulitermes pastinator
mounds over Tumulitermes hastilis mounds at site I, as the concentrations of elements
were generally higher in the extracts from Tumulitermes hastilis mounds.
B) Comparison Between Nasutitermes triodiae and Tumulitermes pastinator
Mounds Composition in Daly River Site 3 and Howard Springs Site 6.
In the perchloric/nitric extracts (Table 3.20) significant and highly significant increases
were found in Nasutitermes triodiae mounds at site 3 in all but three of the selected
elements (calciwn, copper and zinc). However, only two elements (magnesiwn and
sodiwn) remained higher in pepsin-HCl extracts (magnesiwn and sodiwn) and three in
pH 7.5 filtrates (magnesiwn, soluble iron and ionisable iron) than in Tumulitermes
200
pastinator mounds (Tables 3.37 and Appendix F2). No soluble iron or ionisable iron
was detected in pH 7.5 filtrates of Tumulitermes pastinator mounds and only a small
amount was detected in Nasutitermes triodiae mounds (0.36 ± 0.38 and 0.14 ± 0.17
mg/lOOg respectively). In pepsin-HCl extracts, higher concentrations of copper and zinc
were present in Tumulitermes pastinator mounds and in pH 7.5 filtrates aluminium and
copper contents were higher than in Nasutitermes triodiae mounds. The other selected
element concentrations were of the same order in both species.
The small amount of soluble and ionisable iron found in pH 7.5 filtrates from
Nasutitermes triodiae mounds and the increase in magnesium and sodium could possibly
be related to the Aboriginals preference for Nasutitermes triodiae mounds in favour of
Tumulitermes pastinator mounds for nutritional! medicinal purpose.
In Howard Springs site 6, when comparing Nasutitermes triodiae and Tumulitermes
pastinator selected element composition in perchloric/nitric extracts, it was found that
only calcium was highly significantly higher in Nasutitermes triodiae mounds and
aluminium, iron and manganese were significantly higher in Tumulitermes pastinator
mounds (Table 3.22, Appendix F6). This contrasts with the concentrations of calcium,
soluble iron, ionisable iron, potassium, magnesium and sodium in pepsin-HCI extracts
and calcium. soluble iron, ionisable iron, potassium and magnesium in pH 7.5 filtrates
being higher in Nasutitermes triodiae than in Tumulitermes pastinator mounds (Table
3.40). At site 6, low concentrations of soluble iron (0.03 ± 0.05 mg/lOOg) and no
detectable ionisable iron in pH 7.5 filtrates was found in. Tumulitermes pastinator
mounds.
While no general pattern was found for the concentrations of selected elements between
the two species at the two sites in perchloric/nitric extracts, in the bioavailable extracts,
when differences occurred in element concentrations between the two species, the
majority of increases were found in Nasutitermes triodiae mounds.
201
C) Comparison Between Amitermes vitiosus and Nasutitermes triodiae Mound
Composition in Daly River Site 4.
In the perchloric!nitric extracts the Amitermes vitiosus mounds results indicated higher
element contents (calcium, cobalt, potassium, manganese and sodium: see Table 3.21)
than in Nasutitermes triodiae mounds. In the pepsin-HCI extracts, Amitermes vitiosus
mounds had higher concentrations of aluminium, calcium, cobalt, copper, soluble iron
(1 0 times), ionisable iron (6 times) than the Nasutitermes triodiae mounds while
potassium and sodium were higher in Nasutitermes triodiae mounds (Table 3.39 and
Appendix F4). This contrasts with the concentration of elements between the two
species in pH 7.5 filtrates where out of ten selected elements, seven were found in
similar concentration in both species and three (aluminium, potassium and magnesium)
were higher in Nasutitermes triodiae mounds (Table 3.39).
The Aboriginal preference for Nasutitermes triodiae mounds over Amitermes vitiosus
mounds may possibly be related to the higher potassium and magnesium content in a
potentially bioavailable fonn in Nasutitermes triodiae mounds, to the higher clay content
(section 4.2.8.3) and to the fact that it is physically easier to sample Nasutitermes
triodiae mounds than Amitermes vitiosus mounds.
4.4.2.3 Comparison Between Mound Composition of the Same Species at
Different Sites
A) Comparison Between Mound Composition ofAmitermes vitiosus at Different
Sites
In perchloric/nitric extracts all the selected element concentrations were significantly
different between sites with narrow ranges of variation in aluminium, copper, iron,
magnesium and manganese and larger variations for calcium, potassium, sodium and zinc
(Table 3.29 and section 4.2.9.1). In pepsin-HCI extracts (Tables 3.36 and 3.39;
Appendix F7), a narrow range of variation was observed for aluminium, potassium,
sodium and wide variations were observed for soluble iron (at site 5: 18.2 ± 2.5
202
mg/IOOg and at site 2: !57± 77 mg/IOOg), ionisable iron, calcium (at site 5: 119 ± 15
mg/IOOg and at site 2: 24.2 ± 8.7 mg/IOOg), and magnesium in pH 7.5 filtrates.
In relation to the Aboriginal use (selection of Amitermes vitiosus mounds at Daly River
site 2 and rejection of Amitermes vitiosus mounds at site 4), the differences in selected
element composition do not indicate reasons for selection at one site over another, as the
rejected site (site 4) had higher soluble iron, ionisable iron, calcium and magnesium
concentrations than mounds at site 2.
B) Comparison Between Mound Composition of Tumulitermes pastinator and
Nasutitermes triodiae at Different Sites
In Tumulitermes pastinator (Tables 3.35, 3.37 and 3.40; Appendix F8) and Nasutitermes
triodiae mounds (Tables 3.37, 3.39 and 3.40; Appendix F9), important variations in
selected elements between sites were observed. For example in Tumulitermes pastinator
the soluble iron varied in p~psin-HCI extracts from 8.68 ± 1.43 mg/lOOg in site 6 to
18.0 ± 10.0 mg/IOOg in site 3 (Tables 3.37 and 3.40). In both species, a high content
in perchloric/ nitric extracts was not necessarily followed by a high content in
bioavailable extracts. For example, in the perchloric/nitric extract iron content was
higher in Nasutitermes triodiae at site 6 than at site 4 by a factor of 2.6 and the soluble
iron content was similar in pepsin-HCl extracts and markedly different in pH 7.5 filtrates
where it was 15 times higher in site 4.
4.4.2.4 General Overview: Comparison Between Mound Composition of
Different Species Studied at Different Sites and their Relation to the
Adjacent Soil (0-lOcm)
As in perchloric/nitric extracts (section 4.2.10.2), the majority of selected element
content in the mounds compared to the surrounding soil was higher in pepsin-HCl
extracts and pH 7.5 filtrates (Table 3.43; Appendices Fl-F6). While the differences
between the soil and the termite mound rarely exceeded 150 % in the perchloric/nitric
203
extracts (Table 3.29), it frequently exceeded 200 % in the pepsin-HCl extracts and pH
7.5 filtrates (Table 3.43). In pH 7.5 filtrates, soluble iron was never detected in the
soils but was present in most mounds (Table 3.34). The selected element increases
observed in the mounds (section 4.2.1 0.2, Figure 3.19) have been attributed to different
factors (section 1.2.4) and amongst them the termite use of richer sub-soil material97, and
the incorporation of saliva, excreta and vegetation in the mound by termite activities.
The increased differences between mounds and soils in pepsin-HCI extracts and pH 7.5
filtrates could reflect the higher bio-availability of elements derived from plants and
termite by-products.
Although it is difficult to compare the results of this study with those in the literature
as the methods used are so different (section 1.2.4.3), similar trends have been observed.
For example, Lee and Wood (197lb)98 reported, at Daly River, a higher total potassium
concentration (XRF) in the soil with 2950 mg/lOOg compared to 1730 mgi!OOg in a
Nasutitermes triodiae mound; and the opposite was found following exchangeable
extraction: 5.9 and 39.1 mg/lOOg in soil and mound respectively (Table 1.5). In tltis
study at Daly River site 4 the concentration of potassium in perchloric/nitric extracts was
551 ± 144 mg/IOOg in the soil and 687 ± 44 mg/1 OOg in Nasutitermes triodiae mounds,
which represented a 20 % increase. In the pepsin-HCI extracts, the potassium
concentration was 5.85 mg/lOOg remained in soil and 54.9 mg/lOOg in mounds (or 9
times more in the mounds). In Okello-Oloya et al (1985)136, the same trend was
observed between soil and mounds for sodium but not for potassium where the soil
mound ratio for different analyses (HF and BaCliNH4Cl) remained constant (Table 1.7).
In Davies and Baillie (1988)40 study in Sabah, Northern Borneo, they found that
although aluminium was higher in Macrotermes sp. mounds than soil (515 and 447 ppm
respectively) following digestion in hydrochloric acid after pre-ignition at 800°C. \Vhile
with potassium chloride extraction the levels were much higher in soil than mounds:
33.0 and 3.9 me/lOOg respectively. Similar observations have not been made in this
study, the aluminium in perchloric/nitric extracts was generally higher in mounds than
in soils (Table 3.29); in pepsin-HCl extracts it was higher in the mounds at certain sites
204
and lower in other sites and in pH 7.5 filtrates, when detected, it was higher in mounds
(Table 3.43).
A general increase in 'bioavailable' element concentration in tennitaria compared to the
soils and no soluble iron or ionisable iron being detected in pH 7.5 filtrates from soil
(Tables 3.41 and 3.42) could be contributing factors in explaining the Aboriginal
preference for mounds over soils.
4.4.2.5 'Bioavailable' Composition of Different Mounds at Different Sites in
Relation to Human Needs and Foods.
Vermeer ( 1971) 185 reported very low concentrations of "available" minerals in 0.1 N HCl
extract of a popular Ghanian clay, with calcium, potassium and magnesium values of 12,
16.5 and 3.1 rog/lOOg respectively. Their calcium and magnesium values are close to
those found in this study in soils (Table 3.41) and the potassium value is closer to the
mound value found in this work. In Alabama, Edwards et al (1964)47, analysing some
clay eaten by pregnant women of the rural communities found that 0. 03 mg/1 OOg of iron
and 0.20 mgi!OOg of calcium were available (0.1 N HCl). The quantity of clay
consumed varied from 6 to 130 g per day in Alabama. This is comparable to the
quantity eaten by the Aboriginals in the Northern Territory (section 1.1.2.1.2).
The daily average quantity of termite mound consumption can only provide a small
portion of the RD!s for adults (Table 1.2). For example, if 50 g of termitaria is eaten
(Table 3.41), 9 % of calcium, 7 % of copper and 6 % of magnesium RDis could be
covered. As mentioned in section 1.2 the RDis exceed the actual daily nutrient
requirements to take into consideration the variations in absorption and metabolism. The
daily losses are much smaller, for example, to cover the daily calcium loss in an adult
male, 100-150 mg of calcium are necessary (section 1.2). If all the calcium present in
the pepsin·HCl extracts was available (Table 3.41), one would need to eat 250-350 g of
termitaria daily. To satisfy potassium, sodium and zinc needs, much greater quantities
would be needed.
205
Table 4.1 shows the relationship between termitaria, Aboriginal bushfoods and other
common food is given in Table 4.1. However, one must be very cautious in trying to
compare termite mounds with other Aboriginal bushfoods, not only are the methods of
extraction very different (dry ashing or nitric/sulfuric acid extracts138} but also the
organic composition of food compared to mineral/organic composition of termitaria
render the bushfood more available to humans.
TABLE 4.1 Composition of selected Australian Aboriginal bushfoods30 and Western foods18 in mg per lOOg edible portion.
@ : Sample number explanation: Tp Tumu.litemaes pa.stinaror
01-26 : Sample number
Dl : Daly River, site 1
• : Newly built mound material
APPENDIX le: Particle size of Daly River, site 3, termite mound samples (O.lOcm on the outside of the mound unless indicated) from one mound of Tumulitermes pastinator.
@ : Sample number explanation: Nt Nasutitermes trioditu:
92-111 Sample number
H Howard Springs (site 6)
B Berrimah (site 7)
• : Newly built mound material
248
APPENDIX Ij: Particle size of termite mound samples (0-lOcm on the outside at the middle height of the mound) at Daly River, site 1:
Tumulitermes hastilis
Sample Mound Particle size ~%of dry weis_ht~
Number@ Number Oay Silt Fine sand Coarse sand
ThOIDI I 8.7 10.2 51.0 29.6
Th02Dl 2 15.2 10.0 46.9 32.3
Th03Dl 3 14.5 9.2 38.3 42.7
Th04Dl 4 11.1 9.2 32.2 49.8
Th05Dl 5 12.3 8.7 26.5 55.6
@ : Sample number explanation: Th ' TJ~mulitermes hastilis
01-05 : Sample number
Dl Daly River, site 1
249
APPENDIX Ik: Particle size of soil samples (0-lOcm depth) collected at different sites: Elliott (site 5), Daly River (1-4), Howard Springs (site 6) and Berrlmah (site 7).
Soil Termite Particle size (%of d~ weiB,ht~ Number sEecies@ Clal Silt Fine sand Coarse sand
OlE Av 10.28 5.10 34.61 51.17
02E - 12.03 4.55 29.36 54.09
03E - 12.01 6.85 38.61 43.55
04E - 17.65 4.11 26.34 53.26
05D1 Tp,Th 6.66 9.25 42.18 43.36
06DI - 6.21 9.49 47.32 36.21
07D1 • 6.24 7.11 46.18 40.93
08DI • 6.67 8.03 39.39 43.53
09D2 Av 13.33 13.44 35.02 39.61
10D2 • 15.59 11.44 35.41 40.40
1102 • 13.64 9.24 32.65 43.98
15D3 Tp 10.93 22.06 40.14 30.53
16D3 • 8.19 10.58 39.05 44.45
17D3 Nt 17.44 27.17 40.33 21.58
18D3 • 15.91 26.24 41.29 20.18
19D3 • 17.39 25.75 40.15 20.33
20D3 Tp 7.23 12.08 41.67 40.23
2!D3 Nt 14.1 19.30 44.87 25.02
22D3 Nt,Tp 14.76 18.23 40.09 29.30
23D4 Nt,Av,Ca 14.24 10.34 47.21 32.15
24D4 • 16.35 14.41 47.33 26.21
25D4 • 1241 8.13 39.15 44.51
26D4 • 11.42 8.64 51.65 33.75
27H Nt,Tp 20.59 7.09 46.34 30.73
28H • 22.05 10.52 47.31 24.47
29H • 14.3 6.12 56.12 26.35
30H • 18.93 6.70 54.08 25.72
31B Nt 8.77 11.56 35.83 39.08
32B • 8.3 11.19 58.37 26.83
33B • 13.26 8.90 32.49 39.87
@: Termite species studied at these soil sites
See list of abbreviations p.(xxxii).
Appendix II: Selected elemental composition of termite mounds sampled at Elliott (site 5) following hot water Minfusion~ extraction.
Appendix Ilia: Selected elemental composition of termite mounds sampled at Daly River, site 2, following perchloric/nitric acid (4:1) extraction. Species: Amilermes vitiosus.
For selected termite species studied at each site, refer to APPENDIX Ik.
... ~
APPENDIX lVa: Selected elemental composition of termite mouods sampled at Elliott (site 6) and Daly River (2 & 4) following 0.1 N pepsin- IICI (pll 1.35) extraction
@;for explanation of ~.ample number refer to APPENDIX Ia, Ib and Ic.
nd: not detected. Zn detection limit: 0.02 mg/lOOg.
... 8i
... a, a,
APPENDIX IVb: Selected elemental composition of termite mounds sampled at Daly River (sites: 1 & 3) and Howard Springs (site 6), following 0.1 N pepsin- IICI (pH 1.35)
• extraction on 2mm fraction size samples. Species: Tumulitermes pastinator.
@; for e11:planation or sample number refer to APPENDIX Id, le and If.
nd: not detected. Co detection limit = 0.02 mgftOOg
APPENDIX lYe: Selected elemental composition of tennite mounds sampled at Daly River (3 & 4) and Howard Springs (site 6), following 0.1N pepsin- IICI (pH 1.35) extraction on 2mm fraction size samples. Spedes: Nasutitermes triodiae.
@; for explanation of sample refer to Appendix Id, Ie and Ir.
nd: not detected. Detection limit (mg/lOOg); Co and Zn = 0.02; Fe(II) .. 0.2
Fe(ll) nd nd nd nd nd nd nd nd nd
"' .... ~
APPENDIX Vc: Selected elemental composition of termite mounds sampled at Daly River (sites: 3 & 4) and Howard Springs (site 6), following 0.1N pepsfn-IICI ... extraction and neutralisation (pll 7.50) on 2mm fraction size samples. Species: Nasutitermes triodiae. tj
For selected tennite spedes studied at each site, refer to APPENDIX lk..
nd: not detected. Detection limit (mgllOOg): AI and K = 0.05; Co and Zn = 0.02
Cu = 0.01; Fe = 0.06; Fe(II) = 0.2
"----... .----~ ----.
-"' 0 0 ~
' "' E ~
<(
-"' 0 0 ~
' "' E ~
"' 0
-"' 0 0 ~
' "' E ~
0 0
Pepsin-HCI (pH 1.35)
75 Site 1 Site2 Site3
60
45
30
15
0 Tp Th s Av s Tp Nt s Av Nt s Av s Tp Nt s
150 Site 1 Site2 Site3 Site4 SiteS Site6
100
50
0 Tp Th s M s Tp Nt s Av Nt s Av s Tp Nt s
0.15 Site 1 Site2 Site3 Site4 SiteS Site6
0.10
0.05
0.00 -'--,-~ Th S ~ S ~ ~ S ~ Nt S ~ S ~ ~ S
SPECIES I SOIL
275
APPENDIX VI-A Graphic representations of the soil - mound effects (mean ± SE) on element concentrations (mgllOOg) following the in vitro test on termitaria of Amitermes vitiosus (Av), Tumub1ermes pastinator (Tp), Tumulitermes hastilis (Th), Nasutitermes triodiae (Nt), and soil (S) (S) (0-IOcm), sampled at sites 1-6, following pepsin-HCI (pH 1.35) extractions: aluminium, calcium and cobalt.
276
-"' 0
Pepsin-HCI (pH 1.35)
1.30 -,-----,----,---,---,.----,------, Site 1 Site2 Site3 Slte4 Site 5 SiteS
1.04
0 0.78 ~
d, 5 0.52 ::> 0
-"' 0
0.26
0.00 I "i" C(l ..,.. I ~ Th S ~ S ~ M S ~ M S ~ S ~ M S
225-,-----,---,---,----,---,---~ Site 1 Site2 Site3
180
0 135 ~
E "' lL
g 0 ~
' "' E ~
=
90
45
0 I B!lll ~ Th S ~ S ~ M S ~ M S ~ S ~ M S
100 Sb6 Site 5 Site 3 Site 4 Site 1 Site 2
80
60
40
;f 20
0 I 171 171 "? I 171 "7' I ~ Th S ~ S ~ M S ~ Nl S ~ S ~ M S
APPENDIX VI-B
SPECIES I SOIL
Graphic representations of the soil • mound effects (mean ± SE) on element concentrations (mg/IOOg) following the in vitro test on tennitaria of Amitermes vitiosus (Av), Tumulitermes pastinator (Tp), Tumulitermes hastilis (Th}, Nasutitermes triodiae (Nt), and soil (S) (0-IOcm), sampled at sites 1-6, following pepsin-HCI (pH 1.35) extractions: copper, iron and iron (ii).
APPENDIX VI-C
SPECIES I SOIL
Graphic representations of the soil • mound effects (mean ± SE) on element concentrations (mgllOOg) following the in vitro test on tennitaria of Amitermes vitiosus (Av), Tumulitermes pastinator (Tp), Tumu/itermes hastilis (Th). Nasutitermes triodiae (Nt), and soil (S) (0-lOcm), sampled at sites 1-6, following pepsin-HCI (pH 1.35) extractions: potassium, magnesium and manganese.
278
12 Site 1
10
-CD 8 0 0 ~
' 6 CD E -"' 4 z
2
0
1.75 Site 1
I 1.40 -CD
0 0 ~
1.05
' CD E 0.70 -c N
0.35
0.00
APPENDIX VI- D
Pepsin-HCI (pH 1 .35)
I Site 2,/ Site3
I Site2j Site3 I Site4 j SiteS j SiteS
SPECIES I SOIL
Graphic representations of the soil - mound effects (mean ± SE) on element concentrations (mg!IOOg) following the in vitro test on tennitaria of Amitermes vitiosus (Av), Tumulitermes pastinator (Tp), Tumulitermes hastilis {Th), Nasutitermes triodiae (Nt), and soil (S) (0-lOcm), sampled at sites 1-6, following pepsin-HCI (pH 1.35) extractions: sodium and zinc
0
APPENDIX VII-A
Nt S Av S Tp Nt S
Site 5 Site 6
Graphic representations of the soil ~ mound effects (mean ± SE) on element concentrations (mg/1 OOg) following the in · vitro test on termitaria of Amitermes vitiosus (Av),Tumu/itermes pastinator {Tp), Tumu/itermes hastilis (Th), Nasutitermes triodiae (Nt} and soil (S) (0-!0m),sampled at sites l-6, following pepsin-HCI (phl.35) extractions and neutralisation (pH 7 .5): aluminium, calcium and cobalt.
280 pH 7.5 Filtrate
0.20 -,----,.---,------.----c--r----,
c;, 0.15 0 0 ~
' ~ 010 -::> 0 0.05
Q.QQ I ep
Site 4 SiteS Site 6
Av Nt S Av S Tp Nt S
7.50 -,----,-----,----r--,---,-------,-------, Site 1 Site 2 Site 3 Site 6
c;, 0
6.00
0 4.50 ~
do s 300
" u_
1.50
000 I '1" Tp Th S Av S Tp Nt S s
0.7 B I Sitol I Site2 I Sit• 3 I h•I• I <a. • I oa. < I - 060 OJ 0 0 :::: 0.45
]' 030 -
" u_ 0.15
1;j;1 I I 1;j;1 I tj'l tj'l I 0.00 I 1 I I I I I I
APPENDIX Vll-B
~ S ~ M S ~ M S ~ S ~ M S
SPECIES I SOIL
Graphic representations of the soil - mound effects (mean ± SE) on element concentrations (mg/1 OOg) following the in vitro test on termitaria of Amitermes vitiosus (Av),Tumulitermes pastinator (Tp), Tumulitermes hastilis (Th), Nasutitermes triodiae (Nt), and soil (S) (0-l Om),sampled at sites I-6, following pepsin-HC (ph 1.35) extractions and neutralisation (pH 7 .5): copper, iron and iron (Ii).
60 Site 1
50 ~ 40 a> 0 0 30 ~
' a> 20 E -
" 10
0
-10 Tp Th
0
APPENDIX Vli-C
pH 7.5 Filtrate 281
Sitel Site 3 Site 5 Site 6
Av s Tp Nt s Av Nt
s
SPECIES I SOIL
Graphic representations of the soil - mound effects (mean ± SE) on element concentrations (mg/1 OOg) following the in vitro test on termitaria of Amitermes vitiosus (Av),Tumu/itermes pastinator (Tp), Tumulitermes hastilis (Th), Nasutitermes triodiae (Nt), and soil (S) (0-lOm),sampled at sites 1-6; following pepsin-HCI (phl.35) extractions and neutralisation (pH 7.5): potassium, magnesium and manganese.
282
pH 7.5 Filtrate
O.D1 Site 1 Site 2 Site 3 Site 5 Site6
0.01 -Ol 0 0 001 ~ . a, 5 000 c N 000
000 ~ Th S ~ S ~ M S ~ M S ~ S ~ M S
SPECIES I SOIL
APPENDIX VII-D Graphic representations of the soil - mound effects (mean ± SE) on element concentrations (mg/IOOg) following the in vitro test on termitaria of Amitermes vitiosus (Av),Tumulitermes pastinator (Tp), Tumu/itermes hasti/is (Th), Nasutitermes triodiae (Nt), and soil (S) (0-IOm), -sampled at sites 1-6, following pepsin-HCI (phl.35) extractions and neutralisation (pH 7.5): zinc