INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

Oversize materials (e-g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.

Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.

Bell & Howell Information and Learning 300 North Zeeb Road, Ann Abr, MI 48106-1346 USA 800-521-0600

LAURIE ALLYSON KETCH

A thesis submitted in conformity with the requirements for the degree of Master of Science Graduate Department of Botany University of Toronto

0 Copyright by Laurie Allyson Ketch 1998 National Library Bibliotheque nationale 1*1 of Canada du Canada Acquisitions and Acquisitions et Bibliographic Services services bibliographiques 395 Wellington Street 395. we Wellington Ottawa ON KIA ON4 OttawaON K1AON4 Canada Canada

The author has granted a non- L'auteur a accorde une Iicence non exclusive licence allowing the exclusive permettant a la National Library of Canada to Bibliotheque nationale du Canada de reproduce, loan, distribute or sell reproduire, prster, distribuer ou copies of this thesis in microform, vendre des copies de cette these sous paper or electronic formats. la forme de microfiche/^, de reproduction sur papier ou sur format electronique.

The author retains ownership of the L'auteur conserve la propriete du copyright in this thesis. Neither the droit d'auteur qui protege cette these. thesis nor substantial extracts fiom it Ni la these ni des extraits substantiels may be printed or otherwise de cekine doivent eke imprimes reproduced without the author's ou autrement reproduits sans son permission. autorisation.

ABSTRACT

Microbio~ogicalInvestigations of Geophagy in Chimpanzees Degree of Master of Science, 1998 Laurie Allyson Ketch Department of Botany, University of Toronto

Chimpanzees in East Africa are known to eat soil fiom Macrotermitinae mounds, but the reasons for and possible benefits of this behaviour are not well-understood. The purpose of this project was to use a microbiological approach to study geophagy. More specifically, it was to examine soils eaten by chimpanzees (Patroglodytes schweinpirrfhiz] in the Mahale Mountains and Gombe areas of Tanzania to determine if the soils had a characteristic or unusual microflora.

Results showed higher numbers of bacteria and lower numbers of hngi in most termite mound soils compared to control soils. These findings can be correlated to the activities of the termites and the clay mineralogy of the soils. The most commonly isolated fbngal genus was

Penicillium with over 500 isolates and 44 species. Penicillium cifrrfrrmrmwas isolated almost exclusively fiom termite mound soils and is considered a dominant member of the community.

Termite mound soils were found to exhibit greater similarity to each other than to control soils in hngal species composition. ACKNOWLEDGMENTS

I would like to thank Dr. David Malloch for the opportunity to work in his lab and for suggesting such an interesting project idea. His generous financial support as well as his help subculturing isolates and counting colonies is greatly appreciated.

Thank you to the other members of my supervisory committee; Dr. Frank DiCosmo for allowing use of his lab equipment and Dr. Tim Myles for identification of termite samples. Thank you to Dr. William Mahaney for use of particle size and clay mineralogical data in this thesis.

This project would not have been possible if it had not been for the collection of soil and termite samples by Dr. Michael Huffman and Mohamedi Kalunde and they are gratefully acknowledged.

I would especially like to thank my fellow graduate students: Cameron Currie, Michelle Hendrie and James Scott. Cameron Currie was a fiiendly and hn office mate and offered valuable help with statistical analyses. Thank you to Michelle Hendrie for her suppoG kindness and generosity. I am especially indebted to James Scott for patiently answering numerous questions, offering ideas and guidance, helping with identification of isolates and reviewing the thesis. I would also like to thank all three for their friendship over the past two years.

I would like to thank Brenda Koster for help subculturing isolates and in identifying Chaetmium species. Thank you to Wendy Untereiner for help in EupenicilZiunz measurements and also for her kind words of encouragement. Nazanin Alasti-Faridani, Swarnely Modi and Debbie Komlos are gratehlly acknowledged for plating PeniciIIium isolates.

Thanks goes to Alice Cheung and Bess Wong for always greeting me with a smile when I came knocking on their doors to borrow chemicals or equipment. - I would like to thank Carolyn Hutcheon and Jackie Wolfe for the many insightfd conversations pertaining to graduate work.

Thank you to my parents, Lome and Gladys Ketch, for their love and support. I am especially gratefbl to my father for continually helping with my numerous computer problems.

Last but not least I thank my partner, Brad Conrad, for his friendship, support and continued commitment and love. TABLE OF CONTENTS

TITLE PAGE ...... *...... i ... ABSmCT...... *...... *...... *...... *...... *...... *...... 111 ACKNOmEDGMENTS ...... iv TABLE OF CONTENTS ...... *...... v . LIST OF TABLES ...... VII ... LIST OF PLATES ...... VIII LIST OF FIGURES ...... *...... x

CHAPTER I: INTRODUCTION AND LITERATURE REVIEW ...... 1

WHAT ARE PICA AN11 GEOPHAGY? ...... I GEOPHAGY IN HUMANS ...... 4 1.2.1 Medicinal Purposes of Geophagy ...... 4 1.2.2 Geophagy During Pregnancy ...... :... 4 1. 2.3 Geophagy During Famine ...... 5 1.2.4 Soil As a Food Additive or Condiment ...... , ...... 5 1.2.5 Geophugy in Religion and Ceremony...... 7 ETIOLOGY OF GEOPHAGY ...... ~...... ~.~.....7 1.3.1 Humans ...... 7 I .3.2 Animals ...... 9 GEOPHAGY IN EAST AFRICA ...... 9 GEOPHAGY IN CHIMPANZEES ...... 10 CHIMPANZEE DIET ...... 10 MEDICINAL USE OF PLANTS BY CHIMPANZEES ...... -11 SOILS EATEN FROM TEMTE MOUNDS AND ANT HILLS ...... 12 TERMITES ...... ~...... 15 1.10 SCOPE OF PROJECT ...... 16 1.11 LITERATURE CITED ...... 17

CHAPTER 2: CHARACTERIZATION OF MICROBIAL FLORA OF TANZANIAN SOILS ...... 23

2.1 ABSTRACT ...... 23 2.2 INTRODUCTION ...... -23 2.3 METHODS ...... -24 2.3.1 Studysites ...... , ...... 24 2.3.2 . Sample collection...... -26 2.3.3 Laboratory AnaZyses ...... 26 2.3.3.1 Water Content ...... 27 . 2.3.3.2 Drlu~onPlating ...... 27 2.3.4 Statistical Analyses ...... -29 2.4 RESULTS ...... 29 2.5 DISCUSSION ...... 38 2.6 LITERATURE CITED ...... -42

CHAPTER 3: FUNGI FROM ~CROTERMITINAEMOUNDS IN TAN- ...... 45

3.1 ABSTRACT ...... 45 3 -2 NIRODUCTION ...... 45 3.3 METHODS ...... 46 3.3.1 Statistical Analyses ...... -46 3.4 RESULTS ...... 47 3.5 DISCUSSION ...... 54 3 -6 LITERATURE CITED ...... 59

CHAPTER 4: SOIL EXTRACT SUSCEPTIBUTY TESTING ...... 62

4.1 ABSTRACT ...... 62 4.2 INTRODUCTION ...... -62 4.3 METHODS ...... 63 4.3.1 SoilEktraction...... -63 4.3.2 InocuZa fion of Plates ...... 64 4.4 RESULTS ...... 64 4.5 DISCUSSION ...... 65 4.6 LITERATURE CITED ...... -67

CHAPTER 5: CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE WOm......

5.1 DISCUSSION ...... ,,. 5.2 RECOMMENDATIONS FOR FURTURE WORK ...... 5.3 LITERATURE CITED ......

APPENDIX 1: MEDIA RECIPES ...... 74 APPENDIX 2: PENICILLWM DESCRIPTIONS ...... 78

APPENDIX 3: LINE DRAWINGS AND PHOTOGRAPHS ...... 103 LIST OF TABLES

TABLE 1 Animals known to ingest soil ...... ,...... , ...... 3

TABLE 2 Diseases and ailments that soils have been used to treat or cure ...... 6

TABLE 3 Observations of soils fiom termite mounds and ant hills eaten by humans ...... -13

TABLE 4 Observations of soils fiom termite mounds and ant hills eaten by animals ...... 14

TABLE 5 Analysis of variance on the numbers of actinomycetes obtained from termite mound and control replicates ...... 31

TABLE 6 Analysis of variance on the numbers of non-filamentous bacteria obtained fkom termite mound and control replicates ...... -32

TABLE 7 Analysis of variance on the numbers of hngi obtained from termite mound and control replicates ...... 33

TABLE 8 Concentration in CFU/mg of filamentous bacteria (actinomycetes). non-filamentous bacteria and kngi with standard error of the mean in brackets ...... 34

TABLE 9 Concentration in CFU/mg of microorganisms from termite mound number one soils processed at ten and fifteen days post collection and standard error ...... 36

TABLE 10 Moisture content of soils ...... 36

TABLE 11 pH of soils ...... 36

TABLE 12 Percent sand. silt and clay of soils...... 37

TABLE 13 Mineralogy of the < 2 pm fraction of termite mound and control soils ...... 37 . TABLE 14 List of taxa recovered fio m termite mound and corkrol soils ...... 48

TABLE 15 Similarity matrix of soil samples ...... 52

TABLE 16 Shannon's index of diversity of soil samples ...... 52

TABLE 17 Number of taxa, isolates and percent coverage of hngi in soil samples ...... -53

TABLE 18 Zones of inhibition in mm of E-coli in the presence of chloramphenicol(30 pg) and erythromycin (15 pg) ...... 65

TABLE 19 Zones of inhibition in mm of Sauteus in the presence of chlorarnphenicol(30 pg) and erythromycin (1 5 pg) ...... 65

LIST OF PLATES (continued) .. PLATE 23 PeniciIIium oIsonzr ...... ,...... 125

PLATE 24 PeniciNium paxiIZi ...... 126

PLATE 25 PeniciUium piceum ...... 127

PLATE 26

PLATE 27

PLATE 28

PLATE 29

PLATE 30

PLATE 31

PLATE 32

PLATE 33

PLATE 34 LIST OF FIGURES

FIGURE 1 Location of Mahale Mountains National Park and Gombe National Park, Tanzania ...... 25

FIGURE 2 Location of termite mounds where chimpanzees have been observed to eat soil in the Mahale Mountains National Park, Tanzania ...... 28

FIGURE 3 Concentration in CFU/mg of actinomycetes in eaten and control soil replicates .... 31

FIGURE 4 Concentration in CWhgof non-filamentous bacteria in eaten and control soil replicates ...... 32

FIGURE 5 Concentration in CFU/mg of hngi in eaten and control soil replicates ...... 33

FIGURE 6 Concentration in CFUlrng of actinomycetes in soil samples ...... 34

FIGURE 7 Concentration in CFUlmg of non-filamentous bacteria in soil samples ...... 35

FIGURE 8 Concentration in CFtT/rng of fbngi in soil samples...... 35

FIGURE 9 Number of isolates of PeniciCIium cifrimm.PeniciIIium jmfhinellum and all other PeniciNium species obtained from each soil sample ...... 51 CHAPTER 1 INTRODUCTION AND LITERATURE REVIEW

Chimpanzees in East Africa are known to eat soil of Macrotermitinae mounds, but the reasons for and possible benefits of this behaviour are not well understood. While most work on the subject to date has involved using geochemical and clay mineralogical methods of investigation, the purpose of this project was to use a microbiological approach to study geophagy in chimpanzees. Geophagy is a complex phenomenon involving numerous factors, including behaviour, environment and sometimes culture. Organisms fiom at least three different kingdoms (chimpanzees, termites, fbngi and bacteria) are included in this study. Apart from the biological sphere of the project, there are also the geological and geochemical sides to consider. Literature relating to geophagy and to this study comes fiom a multitude of journals and sources, including anthropological, biochemical, entomological, microbiological, geographical, geological, medical, psychiatric, and zoo1ogical. The greatest body of literature relating to geophagy, however, is that occurring in humans. A review of some of these aspects is required for a better understanding of the system.

1.1 WHAT ARE PICA AND GEOPHAGY?

Pica is the ingestion of any unusual substance. Many definitions have been proposed for - this word, among them "the craving for oral ingestion of a given substance that is unusual in kind or quantity " (Lacey, 1990) and "manifestations of false or craving appetite and the deliberate ingestion of a bizarre selection of foods, non-nutritive substances and non-food items" (Parry- Jones and Pany-Jones, 1992). The Diagnostic and Statistical Mmual of Mental Disorders (DSM IV), lists the following criteria for pica: 1) frequent eating of non-nutritive substances for at least one month, 2) the behaviour is inappropriate for the developmental stage, 3) the behaviour is not part of a culturally sanctioned practice, and 4) the behaviour is not associated with another mental disorder, such as mental retardation, or, if the behaviour occurs in connection with another disorder it is severe enough to justify a separate clinical diagnosis (American Psychiatric Association, 1994). Commonly cited as items of pica are hair (trichophagia), burnt matches (cautopyreiophagia), feces (coprophagia), lead paint chips (plumbophagia), leaves, grass, and plant sterns (foliophagia), starch (arnylophagia), stones (lithophagia) and soil (geophagia) (Feldrnan, 1986; Lacey, 1990). The word pica may have come from the Latin name for magpie, which is Pica. It is unclear why the condition is named after the bird, but Cooper (1957) speculates it relates to the birds' range of vocalizations, and habit of eating a wide variety of bizarre substances. Geophagy, or earth-eating, is one form of pica known to occur worldwide, in a range of ethnic, religious and social groups. A wide range of animals have also been documented to eat earth, including buffalo, lions and warthogs. Many non-human primates, such as gorillas and chimpanzees eat soil as well (Table 1). 3 TABLE 1 Animals known to ingest soil

SPECIES TERMITE ANT OTHER MOUNDSOIL MOUNDSOIL SOIL Alcelaphus buselaphus (Bubd hartebeest) X Afces hericuna (Moose) Alouatta caraya (Howling monkey) Atefes (Spider monkey) Callicebus personatus melanochir (Uasked ti ti monkey) Cattle Cephalophus rufllatus (Red-flanked duiker) Centus canadensis (Elk) Chiropotes albinms (White-nosed bearded sakis) Crocuta crocuta (Hyena) Colob us guereza (Guerem monkey) Damaliscus lu~s(Topi) Diceros bicornis (Black rhinoceros) Girafla camalopwdalis (Giraffe) Gorilla gorilla (Mountain gorilIa) Hippotragus equinus (Roan antelope) Homo sapiens (Humans) Indri indri (Malagasy lemur) KOb us elfipsiprymnus(Defassa waterbuck) K kob (Buffon's kob) Lemur catta (Lemur) L- fihs (Lemur) Loxodonatu aficana africana (Bush elephant) L. africana cyclotis (Forest elephant) Lycanon pictus (Wild dog) Macaca mulatta (Rhesus macaque) Odocoileus hemionus (Mde deer) 0. virginiams (White-tailed deer) Oreamnos americanus (Mountain goat) Ourebia ourebi (Oribi) Ovis canadensis (Bighorn sheep) Pan troglodytes schweinfurthii (Chunpanzee) P-trogfodytes troglodjtes (Chimpanzee) Panthera leu (Lion) Papio anubis (Olive baboon) P. ursinus (Chacrna baboon) Parrots Phacochoerus aethiopicus (Warthog) Pongo pygmaeus (Orangutan) Presbytis senex (Purple-faced langur) P. mbicunda (Red leaf monkey) P. entellus (Gray langur) Rangifr arcticus (Caribou) Redunca redunca (Bohor reedbuck) Saguinus mystm: (Moustached tamarin) Syhricapra grimmia (Grimm's duiker) Syncem cafler (Atiican buffalo) Taurotragus derbimus (Giant eland) ~ra~eiu~husscriptus (Bushbuck)

* This was an assumption based on stomach contents and not actual obsecvation.

SOURCES: Cowan and Brink (1949); Davies and Ballie (1988); Diamond (1988); Ferrari (1995); Goodall (1986); Heymann and Hartmann (1991); Hladik (1977a); Mahaney (1993, 1987); MBhaney and Hancock (1990); Mahaney et al. (1995a, 1995b, 1996a); Miiller et al, (1997); Oates (1978); Po!lwk (1977); Ruggiero and Fay (1994). 1.2 GEOPEIAGY IN HUMANS

According to Parry-Jones and Parry-Jones (1992), the earliest English language description of pica involving geophagy is a translation made by John Trevisa in 1398 of the 13" century Latin work of Bartholomeus de Glanville, De Propriekztibus Rerzm. In the large medical part of this book bulimia ("bolismus") is considered, where it is stated that "sometyme the appetite chaungeth and desireth noel thinges, as coles, erthe, salt and other suche" (as cited by Parry-Jones and Parry-Jones, 1992). Soil has been eaten for a variety of reasons throughout histoly; these can be grouped accordingly as: 1) medicinal use, 2) use during pregnancy, 3) as a famine food, 4) as a food additive or condiment, and 5) for religious or ceremonial purposes.

1.2. I Medicinal Purposes of Geophagy Worldwide, the medicinal properties ascribed to clays are numerous. Soil has been used to treat a multitude of diseases and ailments. These have included a vast range of things; £?om a treatment of leprosy one thousand years ago to a perceived cure for acquired immune deficiency syndrome today (Table 2). Soil has been used both internally and externally. For example, it has been used externally to treat bubonic plague, eczema and herpes (Hooper and Mann, 1906; Laufer, 1930), and internally for treatment of such things as diarrhea and nausea (Anell and Lagercrantz, 1958; Bateson and Lebroy, 1978; Hooper and Mann, 1906; Hunter, 1973; Laufer, 1930; Verrneer and FerreH, 1985).

1.2.2 Geophagy During Pregnancy Geophagy is most commonly cited as occumng during pregnancy. From the 16&to the 20~century, it was widely believed to be dangerous to stop pregnant women from eating whatever they wanted. "Such a power hath this Pica and Malacia in women with child, that if they can not enioy the foode, or all other things as they desire, they or their young ones ar in danger of death " (Liebault, 1582, cited in Parry-Jones and Parry-Jones, 1992). Gelfand (1945) speculated that geophagy during pregnancy was related to the perceived fertility of the earth. This seems to be the case in Sudan, where indigenous peoples believe there is an association between geophagy and fertility (Anell and Lagercrantz, 1958). Another widely held belief is that women practicing geophagy will have children which are more handsome (Cooper, 1957), and that clay keeps the body from being marked at birth (O'Rourke et al., 1967). In some areas of Indonesia pregnant women eat earth so their children will be born with lighter skiq while in Egypt, expectant mothers engage in the practice in the belief their child will be born with a dark complexion (Anell and Lagercrantz, 1958). ORourke and colleagues (1 967) commented that it was not unusual to see visitors at their hospital in Georgia, U.S.A, bringing boxes of clay as gifts to obstetrics patients. The common practice of geophagy during pregnancy in some areas of the world may be put into perspective by a quote in Hunter (1993), when he cites an African female senior government doctor in Malawi as saying "It would be very surprising if pregnant women in Malawi did not eat clay. That's how you know when you are pregnant!" Prevalence of geophagy among pregnant women of Malawi, Zambia, Zimbabwe, Swaziland and South Afiica is estimated to be 90% (Hunter, 1993).

1.2.3 Geophagy During Famine Soil ingestion has often been related to famine. Laufer (1930) gives numerous examples of soil being used in this manner in China, New Zealand, New Caledonia, India, Mongoliq Germany, Finland, Africa and North and South America. Indigenous peoples of the Northern Territory of Australia are also known to use soil as a famine food (Bateson and Lebroy, 1978). Von Humboldt in his travel narrative of South America (1799-1804), wrote that clay was often eaten by the Otomac tribe along the Orinco, but when the river flooded the area for a two month period, the people seemed to survive solely on this earth (Cooper, 1957). Vermeer (1966) noted that among the Tiv of Nigeria it is common to qualify a bad famine by saying that "one only had dirt to feed their children,"

1-24 Soil as a Food Additive or Condiment Soil has also been used by various people as a food additive or condiment. It is sometimes added to breads and wild potatoes to decrease bitterness. Laufer (1930) wrote that the Oraibi of Arizona would mix clay with potatoes and that the Hopi would mix clay with certain berries and also tubers of wild bushes. The Porno of California were known to mix a red earth with acorn meal and cook it to make a kind of bread (Laufer, 1930). In some areas of Germany fine clay was placed like butter on bread (known as "stone - butter") (LauPr, 1930), while in the Cameroons it has been noted that some people use earth as a spice, and that it is often mixed with salt (Anell and Lagercrantz, 1958). TABLE 2 Diseases and ailments that soiIs have been used to treat or cure.

- - AILMENT REFERENCE GASTRO-I1VTESTI2VA.LDISORDERS cholera' hell and Lagercrantt, 1958; Hooper and Mann, 1906; Laufer, 1930 Constipa tion Laufer, 1930 Diarrhea Anell and Lagercrantz, 1958; Bateson and Lebroy, 1978; Hooper and Mann, 1906; Hunter, 1973; Laufer, 1930; Vemeer and Ferrelt, 1985 Dysentery Anell and Lagercrantz, 1958; Laufer, 1930 Dyspepsia Gelfand, 1945; Hooper and Mann, 1906 Nausea Bateson and Lebroy, 1978; Laufer, 1930 Relief of gastric irritation due to parasitic diseases Bateson and Lebroy, 1978; Hunter, 1973; Laufer, 1930 (especially hookworm) SEXUALLY TRANSMTTED DISEASES Acquired Immune Deficiency Syndrome Abrahams and Parsons, 1996 Herpes Hooper and Mann, 1906 Syphilis Anell and Lagercrantz, 1958; Christopherson and Cantab, 19 10; Cooper, 1957; Hunter, 1973 DEFICIENCY I1V VIITk'MINS OR MINERALS hernia2 Christopherson and Cantab, 19 10 Beriberi (wasting disease caused by lack of thiamine) hell and Lagercrantz, 1958 OTHER Absorption of tannins and plant material containing toxins Bateson and Lebroy, 1978; Johns and Duquette, 199 1; ZiegIer, 1997 Bubonic plague Laufer, 1930 Enema Hooper and Uann, 1906 Hemorrhoids Hopper and Mann, 1906 Leprosy Thorndike, 1923 (As described by Saint Hildegard of Bingen, 1098- 1 179) Liver problems Lairfer, 1930 Premensuual Syndrome Hooper and Mann, 1906 Relief of nervous tension O'Rourke et al., 1967 Treatment of sore2 or wounds Low. 1990: Vermeer. 1966

1 It was said that soil was taken to "allay burning sensations and cool the whole system." 2 There is still much debate over the relationship of geophagy and anemia. 3 Venneer (1966) noted that among the Tiv of Nigeria dried clays are ground, the powder applied to scabious sores for a short while, and then washed off. 1.2. 5 Geophagy in Religion and Ceremony Soil has been known to be eaten in religious ceremonies in Mexico, Barbados, China, Burma, and Malaysia (Laufer, 1930). Diatomaceous earth in China was often hailed as having a supernatural origin, and the finding and ingestion of this earth was viewed as a happy omen (Laufer, 1930). Geuphagy also occurs in various cultures in conjunction with the swearing of oaths.

1.3 ETIOLOGY OF GEOPaAGY

1.3.1 Hums: Cooper (1957) cites Aetius as considering pica to be the result of "suppression of the menstrual flow, due to pressure by the foetus. Whereupon, he explains, this bloody humor rises up and attacks the stomach, causing women so affected to crave various and odd foods, some salty, some acrid; some crave for sand, oyster shells and ashes" (Aetius-Aetios of Amida: Be Gynecology and Obstetrics of the nth Ceniury A.D.; translated fiom the Latin edition of Coronarius, 1542, by J.V. Ricci.) In fact, many early physicians attributed the phenomenon to changes or disorders in menstrual flow, and even more so to sexual frustration. Sexual hstration was thought to be such a common cause of pica during the eighteenth century that the most widely accepted treatment or cure was marriage (Parry-Jones and Parry-Jones, 1992). - Many references exist pertaining to the habit of earth eating by black slaves during the eighteenth century and the hypothesized reasons for its cause. Cragin (1836) gives reference to an 18 11 work in which the author states "it is the effect of relaxation, and its natural concomitant an impoverished state of the blood, arising commonly from mean diet, and may be produced by any other cause which induces laxity of the solids... and we find that negroes labouring under any great depression of mind, fiom the rigorous treatment of their masters, or from any other cause, addict themselves singularly to the eating of dirt." Thus, depression and inhumane treatment were often cited as cause for geophagy among slaves, as well as poor nourishment and even sorcery (Anell and Lagercrantz, 1958). Dirt-eating was considered to be so problematic that all possible means were used to stop it (Mustacchi, 1971). Often people have speculated that geophagy occurs as a result of a desire or need for salt. Examination of the literature, however, indicates this is unlikely in most cases. For example, in some areas salt is actually added to clay before ingestion. Vermeer (1984) noted that among certain tribes in Nigeria salt is purposely added to soil before it is eaten. Laufer (1930) commented that most individuals ingesting earth probably have-easier access to salt than to clay. Earth eating and the use of salt are apparently unrelated behaviours in both India and China. In the Lindi district of Tanzania geophagy is widespread but it is denied by those people who eat earth that it has any relation with "salt hunger" (hell and Lagercrantz, 1958). It is also written that in ancient China large amounts of salt were obtained fiom salty earth, but, in fact, this earth was not eaten, and soils free from salt were the ones consumed (cited in Laufer, 1930). In the twentieth century, hypotheses based more on scientific research than speculation were proposed to explain soil eating. For example, Vermeer and Ferrell (1985) presented evidence using X-ray diffraction that soils f7om the village of Uzalla, Nigeria, and widely sold throughout western Mica, are mineralogically similar to the commercially available antacid preparation Kaopectatem. That some ingested clays resemble this pharmaceutical was latter confirmed by Mahaney and colleagues (1995a) in their study of soils mined and eaten by mountain gorillas in Rwanda, and again by Mahaney and colleagues (1996b, 1997) for clays eaten by chimpanzees in Tanzania and Uganda. Some researchers have turned to analyzing the elemental composition of geophagical clays, attempting to explain the phenomenon as a need for various elements or minerals. Hunter and De Kleine (1984) analyzed clay tablets eaten in Central America. By comparing the amounts of elements present to recommended daily allowances of these minerals, they determined the tablets to be rich in iron, calcium and manganese, and may be helpful in supplying copper, potassium and zinc. However, Abrahams (1997), in studying soil eaten in Uganda, commented that it is difficult to determine the importance of elemental composition of soils to human nutrition. For example, elemental requirements are different for different ages and sex. As well, what one eats may influence the availability of the nutrients, there are possible competitive interactions among elements, and finally, the frequency and quantity of clay ingestion varies among individuals. At present it seems best not to view geophagy in terms of a conscious craving for a certain mineral element. Halsted (1968) noted that while geophagy to supplement mineral nutrition was a compelling hypothesis no objective and controlled data were available to support it. This continues to be true more than twenty years later as Johns and Duquette (1991) state "nothing supports a notion of a specific appetite for any nutrient as an acceptable explanation of geophagy .. . " Thus, although numerous hypotheses have been suggested to explain the occurrence of geophagy, or why the practice might be beneficial, few are supported by scientific evidence. Those based on scientific study have yielded some clues, but as of yet are not conclusive.

1.3.2 ANIMALS:

There has been a steady increase in research on geophagy in animals in the past thirty years. As is the case with human geophagy, various workers have hypothesized as to its stimulus in animals. Ruggiero and Fay (1994) proposed that elephants ingest termitarium soils due to sodium enrichment of the mound soil. Davies and Baillie (1988) concluded the main reason why red leaf monkeys eat soil is for mineral supplementation and also for the relief of digestive problems such as forestomach acidosis. Mineral supplementation was also accepted by Heymam and Hartmann (1991) in their study of geophagy in moustached tamarins. Oates (1978) suggested soil consumption in guereza monkeys aids in absorption of plant toxins and may help to regulate the pH of the stomach. As stated previously, clay mineralogical work on soils eaten by various primates has shown a composition similar to the pharmaceutical Kaopectatem, and thus it has been proposed the soils are eaten to alleviate or prevent gastrointestinal upset or diarrhea (Mahaney, 1993; Mahaney et a1.,1995a, 1996b, 1997). Again, while various ideas have been proposed, much work remains to be done to explain the phenomenon.

1.4 GEOPHAGY IN EAST AFIUCA

Geophagy is particularly widespread in Africa. Anell and Lagercrantz (1958) noted it is quite common in East aca,and give reference of earlier authors stating it was especially so on - Lake Tanganyika In fact, termite mound soil is sold throughout Tanzania and is often eaten by the Waha people in the Kigoma area (Goodall, 1986). In Dar es Salaam, Tanzania, close to fifty percent of the population is said to engage in geophagy, including the healthy and the sick young and old (Anell and Lagercrantz, 1958). Soils sold on the street in Kampala, Uganda, are often used as traditional medicine and are said to have a wide range of healing powers (Abrahams, 1997).

1.5 GEOPHAGY IN CHIMPANZEES

Chimpanzees (Pan trogZoa3,tes schweinfirthio in the Mahale Mountains National Park and Gombe National Park, Tanzania, have been observed to eat sail, as have Gabon and Ugandan chimpanzees (Goodall, pers. corn.; Huffman, pers. comrn.; Hladik, 197%; Hladik and Gueguen, 1974; Mahaney et al., 1996b, 1997). In Mahale and Gombe chimpanzees eat soil fiom termite mounds. According to Goodall (1986), chimpanzees in Gombe eat soil almost daily, usually fiom mounds of Pseu~canfhotemesmilitmis where they break off a walnut size piece to eat. In Mahale, chimpanzees have been observed to ingest soil from the mounds of termites, but they have never been observed to eat other soil. They break offa small piece fiom the top of the mound, usually approximately 2.5 cm3 in size and resembling a piece of fudge. Often they roll the piece of clay around in their mouth for a few minutes before swallowing. Chimpanzees do not ingest much soil at one time and chimpanzees of all ages engage in geophagy. However, controlled studies on the frequency of the behaviour and the health of the individual at the time of ingestion are lacking. Soil ingestion is a completely separate behaviour fiom termite fishing and eating. For example, to date, chimpanzees of the "Mugroup in Mahale have never been observed to eat termites, but they do eat soil from termite mounds. No other - mammals have been observed to eat soil from termitaria in Mahale, although, Huffman @em. comrn.) speculates that Colobus do as well.

Wrangham (1977) reported 140 species of plants used as food by chimpanzees in the Gombe area of Tanzania, thus adding to 61 other species previously recorded by other researchers in Gombe. Chimpanzees also eat insects, most notably ants and termites, but also those from the orders Lepidoptera, Coleoptera and Orthoptera (Goodall, 1986). They have also been observed to eat insect galls, honey and eggs (Goodall, 1986). Chimpanzees eat mammals, although mammalian prey capture seems to be correlated with its perceived vulnerability. Prey species include Colobzis baditrs (colobus monkeys), Potarnochoenis poms (bushpig), Trnge1aphzr.s scriptzis (bushbuck), Cercopithecus ascanius (redtail monkey), C. rnitus (blue monkey) and various bird species (Wrangharn, 1977). Chimpanzees will often inspect potential food sources before eating, using touch, sight and smell (Wrangham, 1977). Interestingly, some chimpanzees have been document'ed to eat things which would probably be considered items of pica in Homo sqiens. These include urine, feces, ashes (from cooking grates in fisherman's huts), soil contaminated with photographic chemicals and clay from termite mounds (Goodall, 1986).

1.7 MEDICINAL USE OF PLANTS BY CHIMPANZEES

Chimpanzees of the Mahale Mountains and Gombe regions of western Tanzania have been studied with respect to a variety of topics, including their influence on the environment around them, social structure and relationships, communication and culture (Goodall, 1986; Nishida, 1990). Rodriguez and Wrangharn (1993) proposed the term "zoopharmacognosy" to refer to the seiection and use of certain medicinal plants by wild animals for treatment and prevention of disease. Much work has been published on the use of medicinal plants-by chimpanzees (Huffman, 1997; Huffrnan and Seif5, 1989; Huffman and Wrangham, 1994; Wrangham and Nishida, 1983), and the chemical analyses of these plants (Jisaka et al., 1993; Koshimim et al., 1994; Ohigashi et al., 1994). For the past ten to fifteen years, evidence has been mounting that chimpanzees select certain plants to eat for their medicinal components. - Evidence comes from different circumstances, including the fact that ingestion does not provide

- ths chimpanzee with any great nutritional benefits (Wrangham and Nishida, 1983), observations of illness at the time of consumption (Huffman and Seifu, 1989), ingestion of plants during periods of high risk of parasite infection (Kawabata and Nishida, 1991; Huffman et al., 1997) and the low frequency of consumption of these plants (Huffman and Seifu, 1989; Wrangham and Nishida, 1983). One plant frequently used is Vernoniu amygdaIina, commonly known as "bitter leaf', and known as "kills goats" in West Afi-ica. It occurs from sea level to approximately 1700 m, in open woodland and ravine fringes throughout tropical and sub-Saharan Africa. Ethnomedicinal uses of I? mygdolina include intestinal upset, parasitosis, venereal disease, fever, cough and pneumonia (Huffman et al., 1996). Chimpanzees are the most cognitively advanced of the non-human primates. They are tool-users and thinkers and aspects of their culture shows similarities to our own (Tomasello, 1994). Chimpanzees have the ability to learn and to remember. If in fact, as the evidence suggests, chimpanzees are self medicating themselves through ingestion of certain plant species, then it becomes conceivable that soil eating may serve a similar purpose. The stimulus for primate geophagy is still in question. Potentially these animals are eating soil for the same purpose- self medication.

1.8 SOILS EATEN FROM TERMITE MOUNDS AND ANT HILLS

Termite and ant nests and mounds are the most common sources of geophagic soils cited in the literature. However, it is often somewhat unclear exactly which source is being referred to- ants or termites. For example, it is often unknown when authors refer to "ants" if in fact this is the local name for termites. Both humans and animals obtain geophagic soils from these sources (Table 3; Table 4). TABLE 3 Observations of soils from termite mounds and ant hills eaten by humans.

OBSERVATION REFERENCE

+ indigenous peoples in Australia are hiown to obtain ant-hiU earth Bateson and Lebroy, 1978 to use for stomach aches or diarrhea.

4 Clays from termite mounds are eaten in Nigeria. These clays are Vermeer, 1984; Vermeer and Ferrell, 1985 said to come fiom the interior of termitaria.

4 Clays from termite mounds are used among the Ewe tribe of Vermeer, 1971 Ghana

+ In western Kenya the source for geophagic clays by most school Geissler et al., 1997 children is "ant hill" soil (lop Iiel*) (this is actually termitaria soil). This source is favoured by 65% of the children while soil attached to tree trunks by termites (lop bie*) is preferred by 8%. (* Dholuo tenns) + Giant termite mounds are considered to be the most popular Hunter, 1993 source of clays for ingestion in Zambia and Zimbabwe, the major species being Macrotermes falciger, M. mossambicus and M natalensis. These mounds are also consumed by cattle. + Red ant-cIay and white antclay are eaten in Swaziland Hunter, 1993 + In northern Zimbabwe, clays eaten include "arboreal antclay, red Hunter, 1993 termite clay, white ant-clay, black (acidic) river clay, white river clay, and rnacrotenne mounds."

4 Soils eaten by the Shona people of Zimbabwe include soif fiom Aufielter et al., 1997 termite mounds on the ground ("Churu") and also those of termite mounds built on trees ("Muti").

+ Soils from termite mounds in the Philippines are mixed with water Anell and Lagercrantz, I958 and taken to cure a number of intestinai problems. + Nursing mothers in Morocco are said to eat soil fiom ant nests if Anell and Lagercrantz, 1958 they have difXiculty waking at night to breast feed; the rational being that "ants are supposed to sleep very Lightly."

+ Among some tnk along the south- slope of Mount Kenya Anell and Lagercrantz, 1958 "white ant earthn and a "black termite-earth" are eaten as they are believed to accelerate delivery. TABLE 4 Observations of soils fiom termite mounds and ant hills eaten by animals.

OBSERVATION REFE-NCE

+ In the Central Afiican Republic (CAR), African elephants and a range of other Ruggiero and Fay, 1994 animals eat soil f%omtermite mounds. Generally elephants seem to eat the core out of inactive mounds. Mounds eaten are probably those ofMacrotennes spp.

+ Red leaf monkeys (Presbytis rubicunda) in northern Borneo eat soil fiom Davies and Bailie, 1988 tennite mounds. Mounds eaten are often found near the bottoms of large trees and are probably those of Macrofermes, most likely M. gilvus, but possibly M. rnalaccensis.

+ Ia a study of geophagy by masked titi monkeys (Callicebus personatus Miiller et al., 1997 rnelanochir) in Brazil, geophagy was stated as taking piace "in I2 cases on the surface of leaf-cutting ant mound Vtta spec.)."

+ Of three observations of geophagy in moustached tamarins (Saguinus mystar) Heymann and EFartmanr~,1991 in northeastern Peru, one case was a tamarin taking a piece of soil from a "broken and abandoned mound of leaf-cutting ants Vttasp.)."

+ Orangutans in Borneo do not appear to look for clay, but instead seem to eat it Russon, pers. comm. - when they find it. Most of the soil eaten by these primates is found associated with termite nests or the roots of trees. According to locals, soils eaten are often fiom young tennite nests.

+ Chimpanzees in Tanmnia, Gabon and Uganda have been observed to eat soil Uebm, 1982 hrn~a&otennitinae mounds. including Pseudacanthotennes spiniger. Termites are polymorphic, social insects belonging to the order Isoptera Colonies of termites live in nests or mounds which they build. A colony is made up of many different castes which are morphologically dissimilar and which each have specific duties and roles to perform in the termite community. Larvae are able to develop into any group, depending on the needs of the colony (Lee and Wood, 1971). In many respects the higher termites resemble the leaf'-cutting ants in social structure and activities. Most termites exist in tropical and subtropical areas, but some species extend into the temperate zone to about 48% and approximately 45"s (Lee and Wood, 1971). The greatest species diversity exists in the Ethiopian region of Africa. Here 570 species have been recorded, representing 89 genera (Bouillon, 1970). The mounds constructed by termites range in size from small domed or conical structures only a few centimeters high and wide to the massive mounds built by certain species of African Macrotermitinae which can be more than 9 m in height and 20-30 m at the base (Lee and Wood, 197 1). The subfamily Macrotermitinae originated in the Tertiary period (Krishna, 1970). Twelve genera are included in this group. Ten of these occur in the Ethiopian region of Africa (Krishna, 1970). The Macrotermitinae are unique in that they have evolved a symbiosis with the fbngus Tennitomyces (a basidiornycete). These termites use their feces as substrate in construction of a garden upon which the fungus is cultivated (Thomas, 1987). Fecal pellets made up of poorly digested vegetative matter are produced by termite workers. These pellets are - placed on the top of the garden and mycelium quickly develops within it (Darlington, 1994). After approximately two weeks, the fungus starts to produce small, white, asexual fruiting structures. These are eaten by the workers, as are the older parts of the garden and the mycelium (Darlington, 1994; Thomas, 1987). Thus, there is a constant turnover with fresh fecal material being added to the top and old garden being removed from the bottom. The fbngal garden is an integral part of the termite colony and colonies will die if it is removed (Batra, 1975). Termites play an important role in soil ecology, transporting and mixing soil and organic material from different horizons. Termites have developed the ability to tunnel and to form . structures from soil and its components to a remarkable degree. Many termites have wide- ranging foraging areas over which they gather organic debris or living plant tissue which is then brought back to the nest (Lee and Wood, 1971). They aid in the physical breakdown of plant materials, such as cellulose, thereby creating a greater surface area available for microbial decomposition (Arshad et al., 1982). By assisting in the recycling of these materials, termites aid in soil fertility (Amund et al., 1988). Termites move soil particles, mix them with organic material and cement them together. As such they are an important part of the soil fauna in many areas of the world.

1.10 SCOPE OF PROJECT

Given the long history and widespread occurrence of geophagy in both animals and humans, it is conceivable the practice conveys benefits in some cases. To date, however, a conclusive reason for the practice in any organism is lacking. The aim of this study was to provide insight into the possible reason for geophagy in chimpanzees using a microbiologica1 approach. It has already been well documented that chimpanzees use medicinal plants when they are ill, and that some plants chimpanzees ingest at the time of illness have antimicrobial properties (Huffman, 1997). Given these facts, it is conceivable chimpanzees may also use soil for the same purpose- self medication. Many soil microorganisms produce secondary compounds with antimicrobial activities. The objective of this project therefore was to examine soils eaten by chimpanzees in the Mahale Mountains and Gombe areas of Tanzania to determine if they contained a characteristic or unusual microflora. More specifically, it was to determine if there were differences in the numbers of microorganisms present in ingested soils compared to non-eaten soils, or whether there were differences in the types and communities of microorganisms present. The hypotheses of the work were that termite mound soils would be found to differ quantitatively and/or qualitatively in microorganisms f?om non-eaten soils (control soils), and that termite mound soils would be more similar to each other than to control soils. 111 LITERATURE CITED

Abrahams, P.W. 1997. Geophagy (soil consumption) and iron supplementation in Uganda. Tropical Medicine and International Health 2 : 6 17-623.

Abrahams, P.W.and Parsons, J.A. 1996. Geophagy in the tropics: a literature review. me Geographical Jmrnal162: 63-72.

Amund, O.O., Onumonu, M.N., Onawale, T-A and Malaka, S.L.O. 1988. A study of microbial composition and lignocellulose degradation in the mound soil of Macroiemes bellicosus Smeathman. Microbios Letters 37: 69-74.

Anell, B. and Lagercrantz, S. 1958. Geophagical customs. SIudia Ethnographica Up~alie~a17: 1- 84.

Arshad, M.A, Mureria, N.K. and Keya, S.O. 1982. Effect of termite activities on the soil microflora. Pedobiologia 24: 161-167.

Aufieiter, S., Hancoclg R.G.V., Mahaney, W.C., Starnbolic-Robb, A. and Sanmugadas, K. 1997. Geochemistry and clay mineralogy of soils eaten by humans. International Journal of Food Sciences and Nutrition 48: 293 -3 05.

Bateson, E.M. and Lebroy, T. 1978. Clay eating by the aboriginals of the Northern Territory. The Medical Journal of Austraiia, Special Supplement 1: 1 -3.

Batra, S.W.T. 1975.. Termites (Isoptera) eat and manipulate symbiotic fungi. Journal of the Kansas EntomologrgrcalSociety 48: 89-92.

Bouillon, A. 1970. Termites of the Ethiopian region. In: Krishna, K. and Weesner, F.M. (eds.) Biology of Termites. Vo1.2. Academic Press, New York. pp. 154-273. - Christopherson, J.B. and Cantab, M.D. 1910. Earth-eating in the Egyptian Sudan. The Journal of Tropical Medicine and Hygiene 13: 3 -7. -

Cooper, M. 1957. Pica. Charles C. Thomas, U.S.A., 114 pp.

Cowan, I. McT. and Brink, V.C. 1949. Natural game licks in the Rocky Mountain national parks of Canada. Journal of Mammalogy 30: 379-387.

Cragin, F.W. 1836. Observations on Cachexia Africana or Dirt-Eating. American Journal of the Medical Sciences 17: 3 56-3 64.

Darlington, J.P.E.C. 1994. Nutrition and evolution in fungus-growing termites. In: Hunt, J-H-and Nalep a, C.A. (eds.) Nourishment and Evolution in Insect Societies. Westview Press, U.S.A pp. 105-130. Davies, kc. and Baillie, LC. 1988. Soil-eating by red leaf monkeys (Presbytis rubicunclir) in Sabah, Northern Borneo. Biotropica 20: 252-258.

Diagnostic and Statistical Manta2 of Mental Disorders IV. 1994. American Psychiatric Association, Washington, D.C.,886pp.

Diamond, J. 1998. Eat Dirt! Discover 19: 70-75.

Feldman, M.D. 1986. Pica: current perspectives. Psychosomatics 27: 5 19-523.

Ferrari, S.F. 1995. Observations on Chiropotes albinams fiom the Rio dos Marmelos, Amazonas, Brazil. Primates 36: 289-293.

Geissler, P.W., Mwaniki, D.L., Thiongto, F. and Friis, H. 1997. Geophagy among school children in west em Kenya. Tropical Medicine and International Heallh 2 : 624-63 0.

Gelfand, M. 1945. Geophagy and its relation to hookworm disease. Eart Afrian Medical Journal 22: 98-103.

Goodall, L 1986. llte Chimpanzees of Gombe: Patterns of Behavior. The Belknap Press of Harvard University Press, Massachusetts, 673pp.

Halsted, J.A. 1968. Geophagia in man: its nature and nutritional effects. The American Journal of Clinical Nutrition 21: 1384-1393.

Heymann, E.W. and Hartmann, G. 1991. Geophagy in moustached tarnarins, Saguinus mystax (Platyrrhini : Callitrichidae), at the Rio Blanco, Peruvian Amazonas. Primates 32: 533- 53 7.

Hladilg C.M. 1977a. A comparative study of the feeding strategies of two sympatric species of leaf monkeys: Presbyls senex and Presbytis enfellus. In: Clutton-Brock, T.H. (ed.) Primate Ecology: Studies of Feeding and Rangihg Behavimr in Lemurs, Monkeys and Apes. Academic Press, London. pp. 323-3 53.

Hladilg C.M. 1977b. Chimpanzees of Gabon and chimpanzees of Gombe: some comparative data on the diet. In: Clutton-Brock, T.H (ed.) Primate Ecology: Studies of Feeding and Ranging Behaviour in Lemurs, Monkeys and Apes. Academic Press, London. pp. 481- 50 1.

HIadilq C.M. and Gueguen, L. 1974. Ghphagie et nutrition minerale chez les Primates sauvages. Comptes Rendus de Shcesde L'Academie des Sciences, Serie D 279: 1393-1396-

Hooper, D. and Mann, H.H. 1906. Earth-eating and the earth-eating habit in India Memoirs of the Asiatic Society of Bengal 11: 249-270.

Huffinan, M.A 1997. Current evidence for self-medication in primates: a multidisciplinary perspective. Yeorbook of Physical Anthropology 40: 171-200. Huffman, M.A., Gotoh, S., Turner, L.A., Hamai, M. and Yoshida, K. 1997. Seasonal trends in intestinal nematode infection and medicinal plant use among chimpanzees in the Mahale Mountains, Tanzania. Primales 38: 11 1- 125.

Huffman, M.A., Koshimizu, K. and Ohigashi H. 1996. Ethnobotany and zoopharmacognosy of Vernonia amygdalina, a medicinal plant used by humass and chimpanzees. In: CaIigari, P.D.S and Hind, D. J.N. (eds.) Compositae: Biology and Utilization. Vo1.2. Royal Botanic Gardens, Kew. pp. 351-360.

Huffman, M.A. and Seifi, M. 1989. Observations on the illness and consumption of a possibly medicinal plant Vemoniu amyg&Iina pel.), by a wild chimpanzee in the Mahale Mountains National Park, Tanzania- Primates 30: 5 1-63.

Huffman, M.A. and Wrangham, R.W. 1994. Diversity of medicinal plant use by chimpanzees in the wild. In: Wrangham, R.W., McGrew, W.C., de Wall, F.B.M. and Heltne, P.G. (eds.) Chimpanzee Culfzires.Harvard University Press, Massachusetts. pp. 129- 148.

Hunter, J.M. 1993. Macrotenne geophagy and pregnancy clays in southern Africa. Journal of Cultural Geography 14: 69-92.

Hunter, J.M. 1973. Geophagy in Mica and the United States A culture-nutrition hypothesis. Geographical Review 63: 170- 195.

Hunter, J.M. and De Kleine, R. 1984. Geophagy in Central America. Geogrqhical Review -74: 157-169.

Jisaka, M., Ohigashi, H., Takegawa, K., Huffman, M.k and Koshimizu, K. 1993. Antitumoral and antimicrobial activities of bitter sesquiterpene lactones of Vemonia amyg&Iina, a possible medicinal plant used by wild chimpanzees. Bioscience, Biotechnology, Biochemistry 57: 833-834.

Johns, T. and Duquette, M. 1991. Detoxification and mineral supplementation as functions of geo phagy . American Journal of Clinical Nutrition 53: 448-456.

Kawabata, M. and Nishida, T. 1991. A preliminary note on the intestinal parasites of wild chimpanzees of the Mahale Mountains, Tanzania. Primates 32: 275-280.

Koshimizu, K., Ohigashi, H. and Huffman, M.A. 1994. Use of Vernonia amygdzlina by wild chimpanzee: possible roles of its bitter and related constituents. Physiology and Behavior 56: 1209- 1216.

Krishna, K. 1970. Taxonomy, phylogeny, and distribution of termites. In: Krishna, K. and Weesner, F.M. (eds.) Biology of Termites. Vo1.2. Academic Press, New York pp. 127- 152. Lacey, E.P. 1990. Broadening the perspective of pica: literature review. Public Health Repor& 105: 29-35.

Laufer, B. 1930. Geophagy. Field Museum of Narurul History, Anthropological Series XVIII: 99-198.

Lee, K.E. and Wood, T.G. 1971. Termites and Soils. Academic Press, New York, 25 lpp.

Low, T. 1990. Bush Medicine A Pharmacopoeia of Natural Remedies. Angus and Robertson, Australia, 23 8pp.

Mahaney, W.C. 1993. Scanning electron microscopy of earth mined and eaten by mountain gorillas in the Virunga Mountains, Rwanda. Primates 34: 3 11-3 19.

Mahaney, W.C. 1987. Behavior of the scan buffalo on Mount Kenya African Journal of ECOZO~Y52: 199-202.

Mahaney, W.C., Auf'reiter, S. and Hancock, RGY. 1995a. Mountain gorilla geophagy: A possible seasonal behavior for dealing with the effects of dietary changes. International .hodofPri~oZogy 16: 475-488.

Mahaney, W.C.,Bezada, M., Hancock, RGX, Aufieiter, S., Perez, F.L. 1996a. Geophagy of Holstein hybrid cattle in the northern Andes, Venezuela. Mountain Research and Development 16: 177- 1 80.

Mahaney, W.C. and Hancock, R.G.V. 1990. Geochemical analysis of &can buffalo geophagic sites and dung on Mount Kenya, East Africa Mmmalia 54: 25-32.

Mahaney, W.C., Hancock, R.G.V., Aufieiter, S. and Huffman, M.A. 1996b. Geochemistry and clay mineralogy of termite mound soil and the role of geophagy in chimpanzees of the Mahale Mountains, Tanzania. Primates 37: 121-134.

Mahaney, W.C., Milner, M.W., Sanmugadas, K., Hancoclg R.G.V., Aufieiter, S., Wrangham, R, and Pier, H.W. 1997. Analysis of geophagy soils in Kibale Foresc Uganda. Primates 38: 159-176.

Mahaney, W.C., Stambolic, A, Knezevich, M., Hancock, R.G.V., Aufieiter, S., Sanmugadas, K., Kessler, M.J. and Grynpas, M.D. 1995b. Geophagy amongst rhesus macaques on Cayo Santiago, Puerto Rico. Primates 36: 323-333.

Mtiller, K.H., Ahl, C. and Hartmann, G. 1997. Geophagy in masked titi monkeys (CaZZicebus personatus melanochir) in Brazil. Primates 38: 69-77.

Mustacchi, P. 1971. Cesare Bressa (1785-1836) on dirt eating in Louisiana, a critical analysis of his unpublished manuscript De la Dissolution Scorbutique. Journal of the American Medical Association 2 18 : 229-23 2. Nishida, T. 1990. A quarter century of research in the Mahale Mountains. In: Nishida, T. (ed.) l3e Chimpanzees of the Mahale Mountains.. Sedand Life History Strategies. University of Tokyo Press, Tokyo. pp. 3-3 5.

Oates, J.F. 1978. Water-plant and soil consumption by guereza monkeys (Colobus guereur): a relationship with mineral and toxins in the diet? Biotropicu 10: 241-253.

Ohigashi, H., Huffman, M.A, Izutsu, D.,Koshimizu, K., Kawanaka, M., Sugiyama, H., Kirby, G.C., Warhurst, D.C., Allen, D., Wright, C.W., Phillipson, ID., Timon-David, P., Delmas, F., Elias, R., and Balansard, G. 1994. Toward the chemical ecology of medicinal plant use in chimpanzees: the case of Vernonia amygdalina, a plant used by wild chimpanzees possibly for parasite-related diseases. Journal of Chemical Ecology 20: 54 1- 553.

O'Rourke, D.E., Quinn, J.G., Nicholson, J.O. and Gibson, M.H. 1967. Geophagia during pregnancy. Obstetrics and Gynecology 29: 58 1-584.

Parry-Jones, B. and Parry-Jones, W.L.L. 1992. Pica: symptom or eating disorder? A historical assessment. British Journal of Psychiav 160: 3 4 1-3 54.

Pollock, 1.1. 1977. The ecology and sociology of feeding in I'd-i indri. In: Clutton-Brock, T.H. (ed.) Primate Ecology: Studies in Feeding and Ranging Behaviour in Lemurs, Monkeys andApes. Academic Press, London. pp. 37-69.

Rodriguez, E. and Wrangham, R. 1993. Zoopharmacognosy: the use of medicinal plants by animals. In: Downum, K.R, Romeo, J.T.and Stafford, H.A (eds.) Recent Advances in Phylochemistry 27: 89-105. Plenum Press, New York.

Ruggiero, R.G. and Fay, J.M. 1994. Utilization of termitarium soils by elephants and its ecological implications. African Journal of Ecology 32: 222-232.

Thomas, RJ. 1987. Factors affecting the distribution and activity of fbngi in the nests of Macrotermit i nae (Isoptera). Soil Biology and Biochemistry 19 : 343 -3 49.

Thorndike, L. 1923. A History of Magic and Merimentat Science During the First nirteen Centuries of our era. VoI 11. The Macmillan Company, New Yorlq 1036 pp.

Tomasello, M. 1994. The question of chimpanzee culture. In: Wrangham, R.W., McGrew, W.C., de Wall, F.B.M. and Heltne, P.G. (eds.) Chimpanzee Cultures. Harvard University Press, Massachusetts. pp. 30 1-3 17.

Uehara, S. 1982. Seasonal changes in the techniques employed by wild chimpanzees in the Mahale Mountains, Tanzania, to feed on termites (Pseudacunthotermes spiniger). Folia Primatologia 37 : 44-76. Vermeer, D.E. 1966. Geophagy among the Tiv of Nigeria. Annuls of the Association of American Geographers 56: 197-204.

Vermeer, D.E. 197 1. Geophagy among the Ewe of Ghana Ethnology 9: 56-72.

Vermeer, D.E. 1984. Geophagical clays from the Benin area of Nigeria: a major source for the markets of West Africa. Ndional Geographic society ~&earchReports 16: 705-71 I.

Vermeer, D.E. and Ferrell, R.E. 1985. N~geriangeophagical clay: a traditional antidiarrheal pharmaceutical. Science 227: 63 4-63 6.

Wrangham, R.W. 1977. Feeding behaviour of chimpanzees in Gombe NationaI Park, Tanzania. In: Clutton-Broclq T.H. (ed.) Primate Ecolom: Shrdies of Feeding and Ranging Behavimr it1 Lemtrrs, Monkeys and Apes. Academic Press, London. pp. 503-53 8.

Wrangham, R.W. and Nishida, T. 1983. AspiIia spp. leaves: a puzzle in the feeding behavior of wild chimpanzees. Primates 24: 276-282.

Ziegler, J. L. 1997. Geophagy: a vestige of palaeonutrition? Tropical Medicine and lntemationd Health 2: 609-6 1 1. CHAPTER 2 CHARACTERIZATION OF MICROBIAL FLORA OF TANZANIAN SOILS

2.1 ABSTRACT

The dilution plate method was used to determine abundance of microorganisms in termite mound soils eaten by chimpanzees (Pan trogIodytes schweinfrfhii) and corresponding control soils fhm Tanzania. Results show higher numbers of bacteria (both filamentous and non- filamentous) in most termite mound soils as compared to control soils. However, the opposite trend was true of the fingi in most cases. These findings may be related to the activities of the termites and the clay mineralogy of the soils.

Geophagy in primates has received increasing attention in the past two decades. Numerous workers have documented geophagy in non-human primates, including gorillas, chimpanzees and macaques (Davies and Baillie, 1988; Heymam and Hartmann, 1991; Mahaney et al., 1997; Miiller et al., 1997; Pollock, 1977). Investigations have been ongoing by some researchers, most notably Mahaney and workers, in an attempt to determine the reason for this behaviour. Studies to date have mainly focused on the geochemical, heavy mineral and clay mineralogical properties of the ingested soils.- Thus, Mahaney and colleagues have investigated the elemental concentrations and clay mineralogy of soils eaten by chimpanzees in Tanzania and Uganda (Mahaney et al., 1996, 1997), gorillas in Rwanda (Mahaney et al., 199Sa) and rhesus macaques in Puerto Rico (Mahaney et al., 1995b). Probably the most notable point of this work is the fisding that it is often clays with properties similar to the commercially available pharmaceutical Kaopectatem that are chosen for ingestion. Termite mounds are a common source of geophagical soils. Termites belonging to the subfami I y Macrotermit inae are widespread throughout Africa and their mounds fonn a conspicuous part of the Mcan landscape. These termites have evolved a symbiosis with a basidiornycetous fingus which they cultivate and eat. The insects play an important role in the soil ecosystem by moving and rotating large quantities of soil. Few studies have investigated the microorganisms present in these mounds. The majority of existing literature has focused on microorganisms present in the ftmgal garden, or those occumng in the gut or on the exoskeleton of the insect, with few dealing with microorganisms present in the outer areas of the mound- such as those ingested by chimpanzees. Preliminary work by Malloch (pers. comm.) in 1996, suggested that soils from termite mounds eaten by chimpanzees (Pan trogIodytes schweinfrthii) in the Mahale Mountains National Park, Tanzania, contained high numbers of bacteria. Interestingly, particularly high numbers of actinomycetes were isolated. The biological properties of soils eaten by primates seems to have been largely ignored. This fact, together with preliminary findings of high bacterial counts suggested it would be interesting to investigate the microbial communities of these soils. Therefore, the purpose of this study was to use the dilution plate method to determine quantities of microorganisms present in termite mound soils eaten by chimpanzees and to compare these numbers to non-eaten, control soils.

2.3 METHODS

2.3.1 Study Sires:

The Mahale Mountains National Park in western Tanzania (6'73, 2g055'E), is one of only two field stations in Africa where long-term study of chimpanzees has been conducted, the other station being Gombe National Park (Figure 1) (Nkhida et al., 1983). Both parks contain groups of the same chimpanzee subspecies, Pun iroghfytes schweinzrfhii. Mahale boarders Lake Tanganyika, one of Africa's three largest lakes. The area is characterized by abundant tall peaks. Nkungwe, the tallest, is 2462 m above sea level and composed of granitic gneisses and schists (Collins and McGrew, 1988). The mountains are divided by various valleys, some with permanent streams feeding the lake. Mist and cloud development ofken occur in Mahale as a result of damp air blowing f?om the lake (Nishida, 1990). This dampness sustains montane forests above 1880 m as well as the concentration of gallery forests which can be found at the northwestern foot of the mountains (Nishida, 1990). These gallery forests, present from 780- 1300 m, provide food and shelter necessary for the survival of the chimpanzees. The area has a single rainy season and the dry season is from May to October (Collins and McGrew, 1988). The mean daily maximum temperature ranges from 28.4OC to 30.S°C and the minimum from iT Bujumbura

- - IflNatl. Park \

Lake Rukwa

FIGURE 1 Location of Mahale Mountains National Park and Gombe National Park, Tanzania. 18.4OC to 19.S°C, while the average annual rainfall is 1836 rnm (Collins and McGrew, 1988; Huffman, 1997). Soils in the area tend to be relatively young, stony and porous. In culour they are brown or grey-brown or in some areas yellow to red (Collins and McGrew, 1988). In areas containing an organic horizon it is usually 1-2 cm deep (Collins and McGrew, 1988). Gombe National Park (4O4OYS,29'38'E), is located north of Mahale, and also borders Lake Tanganyika (Figure 1). In Gombe the highest land points are about 1500 m, and composed of hard quartzite and some gneiss and sandstone (Collins and McGrew, 1988). Like Mahale, Gombe has a single rainy season and the dry season occurs fiom May to October. Average maximum temperatures range in the wet season fiom about 2S°C to 26S°C and in the dry season from about 27°C to 30°C. Mean minimum temperatures do not vary much with season and remain between 18S°C and 21°C throughout the year. Humidity is very high, approximately 60% to 100% during the wet season and 30% to 70% during the dry season, with average annual rainfall of approximately 1600 mm (Goodall, 1986; Huffman, 1997). Chimpanzees at Gombe range throughout the park from the bottoms of the valley floors to the tops of the hills (Ransom, 198 1). Like Mahale, the Gombe soils are relatively young and stony. In colour they are brown or grey-brown or sometimes grey (Collins and McGrew, 1988). When present, the organic horizon is usually 1-2 cm deep (Collins and McGrew, 1988).

2.3.2 Saple Collection:

Soil samples were collected from Mahale Mountains National Park and Gombe National Park in October 1996 by Dr. M.A. Huffman (Primate Research Institute, Kyoto University). Samples were collected fkom the outer parts of termitaria using an alcohol flamed spade. A sample of from 200 g to 600 g of soil was collected at each site and placed in a Zploc@bag. Additionally, soil samples were taken from approximately 5 m from each termite mound to serve as control samples. Control soils had never been observed to be eaten by chimpanzees. Samples were placed in a second ~i~loc@bag and then into nylon bags and placed in a foam cooler for shipment. Samples were received in Canada 10 days following collection.

2.3.3 Laboratory Analyses:

Five termitarium soils and their corresponding control soils were analyzed in this study. Four of the samples were fiom Mahale Mountains National Park (1, 2, 6 & 10) while the fifth was from Gombe National Park ("G") (Figure 2). Samples were coded with an E or a C to indicate whether they were from a tennite mound (E = "eaten soil") or were a control soil (C).

2-3- 3.1 Water Content Water content of the soil samples was determined gravimetrically (Gardner, 1986). Samples of fiom 1.5 to 2.0 g of each soil were weighed separately into 30 ml glass French square bottles. Bottles were placed in an oven at 50°C and dried to a constant weight. All soil samples were run in duplicate.

2-3.3.2 Dilution PIan'ng The dilution plate method (Waksrnan, 1927; Johnson et al., 1959) was used to isolate microorganisms fiom the soils. For each soil sample dilution series were prepared for five subsamples. 1.00 g of soil was weighed into a test tube containing 9 ml of sterile peptone water (0.1% W/V)using a heat sterilized spatula. This was vortex mixed for four minutes. A 1 ml aliquot of this mixture was then transferred using a sterile pipette to a second test tube containing 9 ml of peptone water. This procedure was continued until a final dilution of lo-' was achieved, with all subsequent dilutions vortexed for 90 s. For isolation of fungi, 1 ml of the dilutions lo-' to 10" were pipetted individually into sterile petri plates. Molten (= 40°C) dextrose-peptone yeast-extract agar @PYA) (Papavizas and Davey, 1959) supplemented with streptomycin, chlorotetracycline and penicillin G (50 pg/ml each) was poured into each plate and the plate rotated prior to solidification to ensure even distribution of spores. Similarly, plates were prepared for recovery of actinomycetes and non-filamentous- bacteria. Starch-casein agar (Kiister and Williams, 1964) was used to isolate actinomycetes while tryptic soy agar (TSA) (Martin, 1975) was used to isolate non-filamentous bacteria. These media were supplemented with the antifungal antibiotics cycloheximide and nystatin (50 pg/d each). Dilutions of lo5 to lod were used for recovery of actinomycetes while dilutions of lo4 to lo-' were used for non-filamentous bacteria. One soil sample and its control were processed each day for five days. On the sixth day, the sample and control which had been processed on the first day were repeated. Plates were incubated at room temperature and colonies were counted four days after inoculation. From the dilution series, plates were also prepared for the recovery of themotolerant and thermophilic fingal species. For each subsample, 1 ml of the dilution 10" was pipetted into a FIGURE 2 Location of termite mounds where chimpanzees have been observed to eat soil in the Mahale Mountains National Park, Tanzania (only samples from sites 1, 2, 6 and 10 were analyzed in this study). sterile petri plate and DPYA added. All plates were incubated at 45°C in moist chambers.

2.3.4 Statistical ,-inalyses:

All statistical analyses were performed on loglo transformed data to correct for heteroscedasticity (necessary for the proper use of parametric analysis of variance and related tests). A single factor analysis of variance was used to determine if mean counts of eaten and control replicates were comparable. The Students t-test (Zar, 1984) was used to determine if mean counts of microorganisms fiom the termite mound and control soil at each site were equal. All analyses were performed using Microsoft Excel version 8.

2.4 RESULTS

Colony forming units per milligram soil (CFU/mg) of actinomycetes, non-filamentous bacteria and hngi were calculated for each of five subsamples of five termite mounds and their five corresponding control soils. Mean CFU/mg of termite mound soil replicates were compared in the three cases with mean CFU/mg of control soils using a single factor analysis of variance. Although both actinomycetes and non-filamentous bacteria appear to occur in greater numbers in termite mound soils than control soils, this difference was not found to be statistically significant (Figures 3, 4; Tables 5, 6). Fungal counts appear higher in control soils than in termite mound soils but statistically this was not significant (Figure 5; Table 7). The soil from termite mound 1 has far higher numbers of non-filamentous bacteria and actinomycetes than any other soil sample (Table 8). If each termite mound is compared to its corresponding control soil (bearing in mind that subsamples within each soil sample do not represent a statistically valid replication) then significantly higher numbers of actinomycetes were found in termite mound soils from sites 1, 2, 10 and Gombe (G), and significantly higher non-filamentous bacterial counts were found in termite mound soils from sites 1 and 10 (P = 0.05) (Figure 6, 7). Fungi were found in significantly higher numbers in control soils f?om sites 1, 6, 10 and G (Figure 8). To determine if time from collection to processing affected the number of microorganisms obtained, soils ficsm site 1 were processed twice; once at ten days following collection and then at fifteen days. Colony forming units of both filamentous and non- filamentous bacteria increased slightly in this five day span in both the termite mound and control soils, while hgal counts decreased a small amount in the termite mound soil and increased slightly in the control soil (Table 9). However, this difference was significant only for non-filamentous bacteria in termite mound soil. Low numbers of fungi were recovered when plates were incubated at 4S°C. At a dilution of no plate for any soil or subsample yielded more than' 10 colonies. Soils from termite mound 10 and the Gombe mound did not yield any colonies at 45OC. Control soils yielded more thermophilic hngi than termite mound soils. Moisture content was determined for the samples. In all cases the control soils had a slightly higher moisture content than the termite mound soils but this difference was not significant. All soils were less than 1% water by weight (Table 10). Termite samples were sent by M.A.H. for sites 2 and 10 in Mahale. These were identified by Dr. T. Myles (Faculty of Forestry, University of Toronto). Termites from the mound at site 2 were identified as Odontotermes sp. (Isoptera: Termitidae: Macrotermitinae) while termites from the mound at site 10 were identified as Pseuducmthotemes sp. (Isoptera: Tennitidae: Macrotermitinae). Unfortunately termite samples were not collected at the same time as soil samples and the three other study sites had broken down at the time of termite collection. Soil pH, particle size and clay mineralogical data were obtained f?om Dr. W.C.~ahane~ (Department of Geography, York University). Control soils were more acidic (pH 4.4-5 -4) than termite mound soils (pH 4.7-6.7) (Table 11). Particle size data indicated that eaten soils had higher clay contents (30-68%) than control soils (19-48%) (Table 12). The mineralogy of the less than 2 pm fraction of the soils was dominated by metahalloysite, illite and illite-smectite, but there did not seem to be any real distinction between the ingested and control soils (Table 13). TERMlTE MOUND (EATEN) SOIL SPMRES

FIGURE 3 Concentration in CFU/mg of actinomycetes in eaten and control soil replicates (bars represent standard error of the mean).

TABLE 5 Analysis ofvariance on the numbers of actinomycetes obtained from termite mound and control replicates (df = degrees of freedom, MS = mean of squares, F = Fisher's F value, P = probability that the null hypothesis is true).

SOURCE df MS F P Between Termite Mound and Controf Soils 1 0.75 3 -34 0. I 1 Within Termite Mound and Control Soils 8 0.22

Total 9 TERMITE M3UM (EATEN) son SAMPLES

FIGURE 4 Concentration in CFUhg of non-filamentous bacteria in eaten and control soil replicates (bars represent standard error of the mean).

TABLE 6 Analysis of variiince on the numbers of non-filamentous bacteria obtained fkom termite mound and control replicates (df = degrees of fieedom, MS = mean of squares, F = Fisher's F value, P = probability that the null hypothesis is true).

SOURCE df MS F P Between Termite Mound and Control Soils 1 0.16 0.72 0.42 Within Termite Mound and Control Soils 8 0.22

Total 9

TABLE 8 Concentration in CFU/mg of filamentous bacteria (actinomycetes), non-filamentous bacteria and hgiwith standard error of the mean in brackets (TM= termite mound soil, C= control soil).

SAMPLE ACTINOMYCETES BACTERIA FUNGI TM C TM C TM C

10 TERMflE MOUND (EATEN) mCONlROL I

-

2 6 SOL SAMPLES

FIGURE 6 Concentration in CFUImg of actinomycetes in soil samples. [QTERMITE MOUND (EAfEN) 0 CONTROL 1

SOlL SAMPLES

FIGURE 7 Concentration in CFU/mg of non-filamentous bacteria in soil samples.

[U TERMITE MOUND (EATEN) mOONTROL ]

2 6 SOlL SAMPLES

FIGURE 8 Concentration in CFU/mg of hngi in soil samples. TABLE 9 Concentration in CRl/mg of microorganisms in soils from site 1 processed at ten and fifteen days following collection with standard error of the mean (SE).

ACmOMYCETES BACTERIA FUNGI TM C TM C TM C Day 10 15 10 15 10 15 10 15 10 15 10 15 CFU 36000 45000 3000 4000 45000 68000 7810 8400 169 159 420 500 SE 2300 2000 260 290 11000 7800 330 1100 20.8 12.9 36 50

TABLE 10 Moisture content of soils.

SAMPLE TEIXIMI'E MOUND CONTROL

TABLE 11 pH of soils (data courtesy of W.C. Mahaney).

SAMPLE TE-MOUND CONTROL 1 6.4 5 -4 2 6.7 -* 6 6.2 5.0 10 6.1 5.3 G 4.7 4.4 * pH was not determined for this sample. TABLE 12 Percent sand, silt and clay of soils (data courtesy of W.C. Mahaney).

1E 1C 2E 6E 6C 10E 10C GE GC Sand (>20 pm) 41 59 36 28 45 28 34 40 68 Silt (2 pm-20 pm) 29 15 12 10 21 4 18 15 13 Clay (< 2 bm) 30 26 52 62 34 68 48 45 19

TABLE 13 Mineralogy* of the < 2 prn £taction of termite mound and control soils (data* courtesy of W.C. Mahaney).

SAMPLE 1E 1C 2E 6E 6C 10E 10C GE GC

* The minerals identified include kaolinite (K), metahalloysite @&I),iIlite (I),illite-smectite (I-S), smectite (S), quartz (Q),orthoclase (0), mica (M), and plagioclase feldspar (P). Semi-quantitative amounts of each mineral are given as trace (tr), small (-I-), moderate (*), abundant (+) or not detected (-). 2.5 DISCUSSION

Clay minerals can be defined as mica-like particles less than 2 pm in size with a large surface area compared to volume (Velde, 1995). Because clays are sheet-like in structure, they can be stacked closely together in parallel layers. Approximately half of the ions present in clays are oxygen, with silicon and aluminum dominating as cations. These ions will bind covalently to form networks (Velde, 1995). Networks are usually made of interlinked polyhedra composed of oxygen and silicon or aluminum. Four oxygen ions can bind to a silicon ion and because of this the polyhedron formed is a tetrahedron. Aluminum, magnesium or ferrous ions, csn bind six oxygen ions and thus form octahedral units. The binding of various tetrahedral and octahedral units allows for-the formation of different types of clays. Combinations seen in natural minerals are: one tetrahedral plus one octahedral layer (a 1:1 structure), two tetrahedral layers plus one octahedral layer (a 2: 1 structure) or two tetrahedral layers plus two octahedral layers (a 2:l + 1 structure) (Velde, 1995). Clays with a 2:l structure are often known as expanding clays. This relates to the fact that these clays generally have a very low layer charge. As such they are not able to bind interlayer cations with enough strength to cause neighboring layers to contract, and exchangeable cations, water, and organic molecules can exist between layers (Weaver, 1989).. Numbers of microorganisms obtained from termite mound soils and control soils in this study can be correlated to the clay mineralogy. Soils corn termite mounds contained a higher percentage of clay than control soils. The clay fraction of the soils is dominated by metahalloysite (a 1: 1 mineral) and illite and illite-smectite. Illite often has expanding properties (Weaver, 1989). It has been shown that the amount and type(s) of clay minerals present in a soil, can affect the kinds of microorganisms occurring in that environment and their survival (Stotzky, 1980). Clay minerals of the 2: 1 type often have a marked influence on the microbial community (Gray and Williams, 1971). In certain cases, montmorillonite (smectite) can decrease the rate of fungal respiration while increasing that of bacteria. As stated above, this may relate to the clay's lattice structure which can expand in the presence of water, and which has a significant cation exchange capacity. These properties therefore allow it to exert an influence on the pH of systems. Bacteria often favour a neutral or slightly alkaline pH for growth (Williams et al., 1983). In effect, expanding clays may help to maintain a soil pH more favourable to bacterial growth by replacing hydrogen ions created by metabolism with basic cations £?om the exchange complex (Stotzky, 1966; Stotzky and Rem, 1966; Stotzky and Rem, 1967). According to some workers the number of actinomycetes recovered from soil using dilution plating can be increased if bentonite (a clay mineral dominated by smectite) is added to the soil (cited in Martin et al., 1976). These clay properties may help to explain why higher numbers of bacteria were found in the clay-rich termite mound soils in this study. Stotzky and Rem (1967) tested a range of kngi (zygomycetes, ascomycetes, basidiomycetes and hyphomycetes) for their growth response in the presence of kaolinite and montmorillonite. They found that the respiration of most species tested decreased in the presence of montmorillonite at 2% or greater, whereas kaolinite was usually required at concentrations greater than 40% to inhibit respiration (Stotzky and Rem, 1967). This result may be due to two reasons. First, it is possible that this effect is due to poor oxygen diffusion rates in the presence of an expanding clay like montmorillonite. Secondly, since fbngi are typically more tolerant of low pH than bacteria, the buffering capability of a clay might not be a deciding factor in the growth of fingi in that environment. Again, these observations could help to explain the lower numbers of fungi found in most termite mound soils as compared to control soils and also correspond to pH data which showed termite mound soils to have a higher pH than control soils. Thus, microbial numbers appear to be heavily influenced by soil properties, which &e in turn influenced by the termites and their activities in mound building. In studying mounds formed by Macrotemes species in the Ibadan area of Nigeria, Nye (1955) suggested there was an upper limit to the size of particles used to construct the mound. He found these termites to prefer particles less than 2 mm in construction, and suggested they may not be able to carry particles greater than 4 mm. Based on work on Macrotemes mounds in Uganda, Kenya and Tanganyika (Tanzania), Hesse (1955) concluded that mound soil is directly related to the surrounding subsoil. Different mounds did not have the same amounts of clay unless they occurred on the same type of soil. Hesse proposed that the insects do not specifically select certain clay size particles for construction, but that the reason mounds are often reported to have higher clay contents than the adjacent topsoil is because the mounds are constructed from subsoil (which generally is higher in clay content than topsoil). Similarly, Stoops (1964) determined that the soil of Macrotemes mounds in Leopoldville (Zaire) varied little from the surrounding subsoil. However, Stoops also studied mounds constructed by Cubiiennes (Termitidae: Termitinae: Cubifermes)and found an increase in clay, organic matter and cations compared to subsoil. Watson (1975), found increased clay content in Macrotermes mounds in Rhodesia (Zimbabwe) but only when the mounds occurred in areas with minimal rainfall (c 700 mm per year). Arshad et al. (1982) investigated Macrotennes mounds in Kenya and found the outer casing of a M. subhyalims mound to be 36% clay whereas the outer casing of a M michaelseni mound was 43% clay. Adjacent soil ranged fiom 22%-49% clay, depending on depth (Arshad et al., 1982). Whatever the reason for the concentration of clay in Macrotermitinae mounds, it is probable that it influences the numbers and kinds of microorganisms found there. It is diff~cultto determine how numbers of microorganisms obtained in this study relate to what one might expect from such soils. Few workers have investigated microorganisms from the outer casing of termitaria. Also, no single technique will isolate all bacterial or fingal genera present. Colony counts of actinomycetes are second only to the bacteria and range from lo4 to 10' in most soils, but again with no technique to isolate ail taxa, this is an underestimate (Kutzner, 1981; Williams et al., 1983). In the "average" soil colony counts for non-filamentous bacteria and kngi may range from 106-1 o8 and 104- lo6, respectively (Alexander, 1977; Burges, 1958). Boyer (1955) studied Macroterrnitinae mounds in the Ivory Coast and found high numbers of certain types of bacteria, notably aerobic and anaerobic nitrogen fixers and aerobic cellulose decomposers. Meiklejo hn (1 965) investigated various bacterial groups occurring in Macrotennes mounds and associated control soils in Rhodesia (Zimbabwe). She also found that cellulose decomposers along with ammonifiers, ammonia oxidizers, nitrate oxidizers and Pseudomonas and D~nitrobacillusoccurred in higher numbers in termite mound soils, while Beijerinckia and Clostridm species were higher in control samples. Amund and colleagues - (198 8) studied microorganisms associated with soil samples of a Macrotennes bellicoms mound in Nigeria and corresponding control soils. Amund's findings were again similar to those of Boyer (1955) and Meiklejohn (1965) in that he found higher numbers of aerobic and anaerobic bacteria, nitrogen fixers and cellulose decomposers in termite mound soil. This worker, however, also isolated actinomycetes and hngi and determined the numbers of these organisms to be higher in termite mound soil as well (Amund et al., 1988). However, Amund reported 3000 CF'U/g actinomycetes in termite mound soil and 220 CFU/g in control soil, which is much lower than anything reported in this study. Probably the study comparing the best to this one is that of Zoberi (1979) in a study of a mound of Macrotennes natalensis in Nigeria. Zorberi (1979) determined bacterial numbers ranged from 3 160 CFU/mg to 10,080 CFU/mg in the upper layer of the termitarium, depending on the month sampled. Fungi ranged from 45 CFU/mg to 203 CFU/rng. However, filamentous and nonfilamentous bacteria were grouped together in this study and the surrounding soil was not analyzed (Zoberi, 1979). Ideally, using the dilution plate method one would want to collect and process samples in a short amount of time. Given the logistical problems of this study, samples were not processed until ten days after collection. For this reason the soils processed on the first day (site 1) were repeated again (15 days following collection) to determine if there were any significant differences in the number of microorganisms obtained. Although there were some differences, they were only significant in one case, and possibly may be attributed in part to the problems inherent in the dilution plate method itself. Although the statistical methods used in this study revealed no significant difference in the numbers of microorganisms occurring in termite mound soils as compared to proximal control soil replicates, this should not rule out the possibility of biological significance. The soils from termite mounds 1 and 10, for example, have far higher numbers of non-filamentous bacteria and actinomycetes than the rest of the soil samples. Soil from site one has twelve times more actinomycetes and almost six times more non-filamentous bacteria in the termite mound soil as compared to the control soil. Soil from site ten has just under twelve times more actinomycetes and just under four times more non-filamentous bacteria in the termite mound soil as compared to the control soil. At the very least it seems that the termite mounds at these two sites are unique in terms of the numbers of microorganisms present. The differences in the soils from these sites seem so great that they probably should not be deemed insignificant on the basis of statistics alone. It is conceivablethat the positive feedback threshold required to support a behaviour such as geophagy may be less than statistically significant. That such high numbers of bacteria were found in termite mound soh 1 and 10 relative to control soils makes it conceivable that this could serve as a stimulus for geophagy. 2.6 LITERATURE CITED

Alexander, M. 1977. Itrtrodz~ctionto SoilMicrobiology. 2nded. John Wiley and Sons, New York, 467pp.

Amund, O.O., Onumonu, M.N., OnawaIe, T.A. and Malaka, S.L.O. 1988. A study of microbial composition and lignocellulose degradation in the mound soil of Macrotermes beNicosus Srneathman. Mlicrobios Letters 37: 69-74.

Arshad, M.A., Mureria, N.K., and Keya, S-0. 1982. Effect of termite activities on the soil microflora. Pedobiologia 24: 16 1 - 167.

Boyer, P. 1955. Premieres etudes pedologiques et bacteriologiques des termitieres. Comptes Rerrdrcs Hebdomudaires des S.ances de L 'Acudkinie des Sciences 240: 569-57 1.

Burges, A 1958. Micro-organisms in the Soil. Hutchinson and Company, London, 188pp.

Collins, D.A and McGrew, W.C. 1988. Habitats of three groups of chimpanzees (Pan hog~odytes)in western Tanzania compared. Journal of Human Evolution 17: 5 53-574.

Davies, A.G. and Baillie, LC. 1988. Soil-eating by red leaf monkeys (Presbylis rtrbicunda) in Sabah, Northern Borneo. Biotropica 20: 252-258.

Gardner, W.H. 1986. Water Content. In: Mute, A. (ed.) Method of Soil AmzZysis Part I: Physical and Mhwalogicd Metho&. 2" ed. American Society of ~~ronom~-~oil Science Society of America, U.S.A. pp. 493-544.

Goodall, J. 1986. 7he Chimpunzees of Gombe: Patterns of Behavior. The Belknap Press of Harvard University Press, Massachusetts, 673 pp.

Gray, T.R.G. and Williams, S .T. 1971. Soil Micro-organisms. Oliver and Boyd, Edinburgh, 240pp.

Hesse, P.R. 1955. A chemical and physical study of the soil of termite mounds in East Africa. Journal of Ecology 43: 449-46 1.

Heymann, E.W. and Hartmann, G. 199 1. Geophagy in moustached tamarins, Saguimrs mysfm (Platyrrhini: Callitrichidae), at the Rio Blanco, Peruvian Amazonas. Primates 32: 533- 537.

Huffman, M. A 1997. Current evidence for self-medication in primates: a multidisciplinary perspective. Yearbook of Physical Anthropology 40: 17 1-200.

Johnson, L.F., Curl, E.A., Bond, J.H. and Fribourg, HA. 1959. Methods for Studjing Soil Microflora Pht Disease Relationships. Burgess Publishing Company, Minneapolis, 178pp. Kutzner, H.J. 1981. The family Streptomycetaceae. In: Starr, M.P., Stolp, H-, Trirper, H.G., Balows, A. and Schlegel, H.G. (eds.) The Prokaryotes A Hnndbook on Habitats, Isdation andldentj?cafionof Bacteria. Vol. II. Springer-Verlag, New York pp- 2028-2090.

Kuster, E. and Williams, S.T. 1964. Selection of media for isolation of streptomycetes. Nature 202: 928-929.

Mahaney, W.C., AuEreiter, S. and Hancock, R.G.V. 1995a. Mountain gorilla geophagy: a possible seasonal behavior for dealing with the effects of dietary change. Internafirnafiond hU~alofPrmafology 16: 475-48 8.

Mahaney, W.C. and Hancock, R.G.V.,Aufieiter, S., and Huffman, M.A. 1996. Geochemistry and clay mineralogy of termite mound soil and the role of geophagy in chimpanzees of the Mahale Mountains, Tanzania. Primates 37: 12 1- 134.

Mahaney, W.C., Milner, M.W., Sanmugadas, K., Hancock, RGX, Aufieiter, S., Wrangham, R, and Pier, H.W. 1997. Analysis of geophagy soils in Kibale Forest, Uganda. Primates 38: 159-176.

Mahaney, W.C., Stambolic, A-, Knezevich, M., Hancock, R.G.V., Aufieiter, S., Sanrnugadas, K., Kessler, M.J. and Grynpas, M.D. 1995b. Geophagy amongst rhesus macaques on Cayo Santiago, Puerto Rico. Primates 36: 323-333.

Martin, J.K. 1975. Comparison of agar media for counts of viable soil bacteria. Soil Biology.ynd Biochemisfry 7:40 1-402.

Martin, J.P., Filip, 2. and Haider, K. 1976. Effect of montmorillonite and humate on growth and metabolic activity of some actinomycetes. Soil Biology and Biochemistry 8: 409-413.

Meikiejohn, 1. 1965. Microbiological studies on large termite mounds. Rhodesia, ZmZambia und Malawi Journal of Agricul~luraZResearch 3: 67-79. -

Mtiller, K.H., Ahl, C. and Hartmann, G. 1997. Geophagy in masked titi monkeys (Callicebus personutus melonochir) in Brazil. Primates 38 : 69-77.

Nishida, T. 1990. A quarter century of research in the Mahale Mountains. In: Nishida, T. (ed.) The Chimpanzees of the Mahale Mountaim Sexual and Lije History Strategies. University of Tokyo Press, Tokyo. pp. 3-35.

Nishida, T., Wrangham, RW., Goodall, J. and Uehara, S. 1983. Local differences in plant- feeding habits of chimpanzees between the Mahale Mountains and Gombe National Park, Tanzania. Journal of Human Evolution 12: 467.480.

Nye, P.H. 1955. Some soil forming processes in the humid tropics. Journal of Soil Science 6: 73- 83. Papavizas, GC. and Davey, C.B. 1959. Evaluation of various media and antimicrobial agents for isolation of soil fungi. Soil Science 88: 112-1 17.

Ransom, T.W. 1981. Beach Troop of the Gombe. Associated University Press, U.S.A., 3 19pp.

Stoops, G. 1964. Application of some pedological methods to the analysis of termite mounds. In: Bouillon, A. (ed.) Etudes sur les Termites Aficuns. Universite Lovanium, L6opoldville. pp. 379-398.

Stotzky, G. 1980. Surface interactions between clay minerals and microbes, viruses and soluble organics, and the probable importance of these interactions to the ecology of microbes in soil. In: Berkeley, R.C.W., Lynch, J.M., Melling, J., Rutter, P.R., and Vincent, B. (eds.) Microbial Adhesion to Surfaces, Ellis Korwood, England. pp. 23 1-247.

Stotzlq, G. 1966. Influence of clay minerals on micro-organisms. III. Effect of particle size, cation exchange capacity and surface area on bacteria. Canadan Journal of Microbiology 12: 1235-1246.

Stot*, G. and Rem, L.T. 1966. Influence of clay minerals on micro-organisms. I. Montmorillonite and kaolinite on bacteria. Canadian Jozirnal of Microbiology 12: 547- - 564.

Stotzky, G. and Rem, L.T. 1967. Influence of clay minerals on micro-organisms. N. Montmorillonite and kaolinite on fungi. Canadan Journal of Microbiology 13: 1535- 1550.

Velde, B. (ed.) 1995. Origin and Mineralogy of Clays, Clays in the Environment. Springer- Verlag, Berlin, 334pp.

Waksman, S.A. 1927. Principles of Soil Microbiology. Bailliere, Tindall and Cox, London, 897pp. -

Watson, J.P. 1975. The composition of termite (Macrofermesspp.) mounds on soil derived from basic rock in three rainfall zones of Rhodesia. Geodema 14: 147-158.

Weaver, C.E. 1989. Clays, Muds, andSMes. Elsvier Science Publishers, Amsterdam, 819pp.

Williams, S.T., Lanning, S. and Wellington, E.M.H. 1983. Ecology of Actinomycetes. In: GoodfeIlow, M., Mardarski, M., and Williams, S.T. (eds.) The Biology of Actinomycetes. Academic Press, New York. pp. 481-528.

Zar, J.H. 1 984. Biostutisticul AnaZysis- 2nded.P rent ice Hall, New Jersey, 7 1 8pp.

Zoberi, M.H. 1979. The ecology of some fungi in a termite hill. Mycologia 71: 537-545. CHAPTER3 FUNGI FROM MACROTERMITINAE MOUNDS IN TANZANIA

3.1 ABSTRACT

The dilution plate method was used to isolate fungi fiom Macrotermitinae mound soil ingested by chimpanzees (Pan iroglodyles schweinfurrhii) in Tanzania, and non-ingested control soils. The most commonly isolated genus was PenicilIium with over 500 isolates representing 42 species. The five most frequently isolated species were PeniciIIium citrinum. Penicillium junthinelium, Paecilomyces cf. marquandii, Gongronella bzrtleri and Eupenicilh shearii. Penicillium cilrinum was isolated almost exclusively fiom termite mound soil and is considered to be a dominant member of this community. Termite mound soils were found to exhibit greater similarity to each other than to control soils in species composition.

3.2 INTRODUCTION

Few studies have investigated the microorganisms existing in the outer parts of mounds created by members of the Macrotermitinae. Perhaps the group that has been ignored to the greatest degree is the fungi. Although a number of papers have dealt with fungi occurring in the fbngal gardens of these termites (see Batra and Batra, 1966; Das et al., 1962; Thomas, 1987% 1987b), hardly any have studied the fungi that might occur in other areas of the mound. Zoberi (1979) was one of the few workers to identify hngi from the outer casing of a termitarium, but he obtained only seven species of fungi from a Macrotennes natalemis mound in Nigeria. In the first part of this investigation, numbers of microorganisms occurring in termite mound soils eaten by chimpanzees and non-eaten control soils were determined (see chapter 2). Higher numbers of fungi were found in four out of five control soils. However, a second hypothesis was that the fungal species present in termite mound soils might be unique and different fiom those in non-ingested soils. Therefore, the purpose of this study was to identify the fungi occurring in the soils, and to determine if termite mound soils have a characteristic . my coflora. The dilution plate method as previously described in Chapter 2 was used to isolate fingi from the soil. Colonies were subcultured £?om plates between four and 20 days post-inoculation. Plates at a dilution yielding 15-35 colonies were selected for subculturing. All colonies fhm these plates were then transferred onto separate plates of an appropriate medium. If a dilution series contained no plates giving the desired number of colonies (i-e. plates had less than 15 colonies or more than 35 colonies) the best plate was chosen and all colonies were taken from a randomly selected half or quarter of the plate.

3.3.1 STA TISTICAL ANAL YSES

Shannon's index of diversity was calculated using the BASIC program supplied in Ludwig and Reynolds (1988) and the formula:

where H' is the average uncertainty per species in an infi~nitecommunity made up of S* species with known proportional abundancesp,, p2, p3, .- ., ps* (Ludwig and Reynolds, 1988)- A modified Sorenson's similarity index was used to determine whether termite mound soils were more similar in terms of species composition to other termite mound soils than to control soils. This was calculated using ths formula:

where aN is the total number of individuals in site 4 bN is the total number of individuals in site B, and jN is the sum of the lower of the two abundances for species found in both sites (Magurran, 1988). The index is designed so a value of one indicates samples are completely similar while zero means they are completely dissimilar. To determine the proportion of the fhngal population represented by species isolated in . the individual soil samples ("percent coverage"), Good's hypothesis 8 (1953), as modified by Moore and Holdeman (1974) was used. According to this formula the percent coverage is equal to 1- (number of species observed once/ total number of species) multiplied by 100. 3.4 RESULTS

More than 3400 microbiological isolates (2122 bacteria and 1371 fungi) were subcultured f?om soil dilution plates. Thirty-one genera of fbngi were identified, representing at least 126 species (Table 14). Ninety taxa were identified fiom termite mound soils, 68 from control soils while 29 species were found in both termite mound and control soils. The most commonly isolated genus was Peniciliium with over 500 isolates and 42 species. The five most frequently isolated species were Penicillium cifrinum (172 isolates), Penicillium janthinelltrm (1 3 3 i so t ates), Paecilomyces cf. marqzrandii (1 18 isolates), Gongronella butleri (88 isolates) and Eupenicillium shearii (66 isolates). Fifty-five species were onIy isolated on a single occasion. Interestingly, Penicillium cifrintrm was found almost exclusively in termite mound soils (169 of 172 isolates), while GongronelIa butleri was found only in control soils and 124 of 127 isolates of Penicillium janihinellum were found in control soils. Paecilomyces cf marpandii was found in all soils (Figure 11; Table 14). Isolates of all species except Gliomasiix novae-zelondae, LiberteNa sp., PenicilIium chrysogenum, P enicillizrm paxilli, Penicilliurn species #9 to #29, Penicifium sp. #3 1, "mycelium sterile" and the "coelomycete" and "yeast" species were deposited with the American Type Culture Collection. Similarity indices showed termite mound soils to be more similar to each other in species composition than to control soils. Similarity values for termite mound soils compared to other termite mound soils ranged from 0.25-0.42, while values for termite mound soils compared with wntrol soils ranged fiom 0.04-0.3 3 (Table 15). In all cases termite mound soils were found to have a higher species diversity than control soils. Diversity values for termite mound soils ranged from 2.25-2.91, while those for control soils ranged from 1.28 - 2.61 (Table 16). Using Good's modified formula it was determined that the taxa isolated accounted for 40.7% to 57.1% of the total taxa viable in the soil. Thus, the probability of obtaining a new isolate with the next sample is about 50% (Table 17). TABLE 14 List of taxa recovered fkom termite mound and control soils and the number of times each taxon was isolated. Sample 1 was analyzed twice and both data sets are presented.

TAXON 1E 1E2 1C 1C2 2E 2C 6E 6C 10E 10C GE GC TOTAL Absidia cylindrospra Hagcm 003200010 0013 19 Acremonium stricturn w. Garns 000000207 0 00 9 Acremonium cJ fusidioides picot) W-~ams 00 0 0 00 1 0 0 0 0 0 1 Acremonium sp. #I 000000000 001 1 Acremonium sp. #2 000000000 100 1 Aspergillus amleatus f izuka 000000000 100 1 Aspergillus auricornus (Gucgcn)Saito 310001000 120 8 Aspergillus candidus # 1 Link ex Link 000000600 0 00 6 Aspergillus candidus #2 c ink e?r Link 210000000 0 00 3 Aspergillusflavipes (&in. & S~R)~hom & 530020001 3 11 16 Church Aspergillus giganteus wchma 000000000 001 1 Aspergillus jmus Raper & Thorn 000000000 0 30 3 Aspergillus malodoratus Kwon-Chung & 000000100 000 1 Fcnnell Aspergillus melleus Yukawa 001000000 0 00 1 Aspergillus niger van Ticghem 110000000 0 00 2 Aspergillus niger #2 van ~hieghcm 210000000 000 3 Aspergillus parasiticus s peare 000000000 005 5 Aspergillus sclerotiorum HUM 110000000 0 00 2 Aspergillus sulphureus @resen) Weher 968000000 0 00 23 Aspergillus ustus ah) mom gr church 000000002 0 00 Aspergiflus sp. # 1 100000000 000 Aspergillus sp. #2 420000000 0 00 Aspergillus sp. #3 000001000 000 A ureobasidium sp. 200000000 000 Blastobotrys cf: nivea KIO~O~CIC 000000010 0 00 Chaetomium cochliodes pall. 000001000 000 Chaetomium longicolleum Krzem & Badura 00 0 0 00 0 0 1 0 0 0 Chaetosphaeronema sp. 200000000 000 CIudosporium sp. #1 000000000 0 10 Cladosporium sp. #2 000000000 0 10 Cunninghamella sp. 0020200140 2 00 Eupenicillium shearii SIOI~gt scou 3 2144 02302 3 3 2 1 Eurotium sp. 000000010 000 Fusariurn sp, #1 20 114000 130 0 Fusariurn sp. #2 000000000 0 13 Fusarium sp. #3 000010100 00 Gliocladium roseum ~ain 000030000 000 Gliocladium sp. #I 000100000 000 Gliocladium sp. #2 001000000 0 00 Gliocladium sp. #3 000000001 000 Gliomastixnovae-ze Iandiae HU* gc 000000001 000 Dickinson Gliomastix sp, 000000200 000 Gongronella butleri (Lmdncr) Peyronclgt D~I 00 3 1 13 00 0 11 0 28 0 5 vcsco TABLE 14 continued

TAXON 1E 1E2 1C 1C2 2E 2C 6E 6C IOE 10C GE GC TOTAL Liberteffa sp. 000010100 000 2 Metarhitiurn anisopliae (M~~SCIUL)SO~O~ Microsphaeropsispseudasppora surton Mucor cf: racernosus ~tes. Mucor sp. Myrothecium sp. Nectria sp, Oidiodendron griseum Robak Paecilomyces cameus Hcim) AKS. Brown & G. Sm Paecilomyces cf, marquanc?ii (M-+ HU~~CS Paecilomyces marquandii massee) ~ughcs Paecilom~essp. Penicillium charfesii G. smith Penicillium chrysogenum Thorn Peniciflium citrinum norn Peniciffiumcf coryfophilum Dicdx Penicillium griseolurn G.smith Peniciffiumjanaewskii Zaleski Penicilfiumjanthinellum Biourge Penicilliunr jensenii meski Penicillium olsonii Bain & Sartory Penicilfiumpdi MIL Peniciffiumpiceum Raper & Fcnncll Peniciffiumrestricturn Gilman & ~bbott Penicillium rugulosum Thorn Penicilfium sclerotiorum van Beyma PeniCifhm simpficissintum (Oudem.) Thorn Penicillium sp. # 1 (Subgen Aspergiffoides) Peniciffiumsp. #3 (Subgen Penicillium) Penicillium sp. #4 (Subgen Penicilliunt) Penicillium sp. #5 (Subgen Peniciffium) Penicillium sp. #6 (Subgen Penicilfium) Penicillium sp. #7 (Subgen Penicillium) Penicillium sp. #8 (Subgen Furcatum) Peniciflium sp. #9 (Subgen Furcatum) Penicillium sp. #10 (Subgen Furcatum) Penicillium sp. # 1 1 Penicillium sp. #12 Penicillium sp. # 13 (Subgen Furcatum) Penicillium sp. #14 Penicillium sp. #16 (Subgen Furcatum) Penicillium sp. #17 (Subgen Furcatum) Penicillium sp. #18 Penicilliurn sp. #19 (Subgen Furcatum) Penicillium sp. #2 1 (Subgen Furcaturn) 200000000 000 2 TABLE 14 contimed - TAXON 1E 1E2 1C lC2 2E 2C 6E 6C 1OE LOC GE GC TOTAL Peniciilium sp. #23 100000000 0 00 1 Peniciilium sp. #24 (Subgen Furcatum) Penicilh sp. #25 (Subgen Furcatum) Peniciiiium sp. #26 (Subgen Furcatum) PenicilIium sp. #27 (Subgen Biverticillium) Peniciilium sp. #28 (Subgen Furcaturn) PenicilZium sp. #29 Peniciliium sp. #30 (Subgen Biverticillium) PenicilZium sp. #3 1 Pestalotia sp, Phoma sp. Phomopsis sp. Scedosporium state of Pseudallescheria sp. Scopulariopsis sp. SesquiciIhm candelabrum W.~ams Talarornyces wortmannii ~;lackcr)C-R Benjamin Trichoderma crassuni Bissett Trichoderma hanianum Rifai Trichoderma koningii Oudem. Trichoderma Wens(Miller, Giddens & Foster) von Am Verticiilium chlamydosporium Goddard Verticiihm tenerunt (Nw ex Pm)Link Verticilliurn sp. Volutella ciliak2 Alb.& Schw. : Fr. Coelomycete #1 Coelomycete #2 Coelomycete #3 Coelornycete #4 Coelomycete #5 CoeIomycete #6 - Coelomycete #7 Coelomycete #8 MyceIiwn sterile Nectria anamorph Yeast*. #1 Yeast sp. #2 Yeast sp. #3 Yeast sp. #4 Yeast sp. #5 Yeast sp. #6 Unidentified

1E 1E2 1C 1C2 2E 2C 6E 6C 10E 10C GE GC TOTAL Total Number of Isolates Identified 101 56 133 60 96 68 142119 112 75 71 108 1141 TotaI Number of Taxa 26 14 22 15 26 10 27 15 24 22 29 27 90 -

IP.abinum E P. janthinellum OAIl Other Species of Penicillium

FIGURE 9 Number of isolates of Penicilliurn cihimrm, Penicillium janthinellum and all other Penicilh species obtained fiom each soil sample. TABLE 15 Similarity matrix of soil samples. Values in dark grey cells were obtained by comparing termite mound soils. Values in light grey cells were obtained by comparing control soils, while values in non-shaded cells were obtained by comparing termite mound soils with control soils.

1C 0.22 0.17 0.19 0.09 0.21 ...... 2c 0-1 I 0.07 0.08 0.07 0.12 ...... ::~.~i~:;;;;. 6C 0-15 0. 16 0 10 0 04 0 20 :~~~0~3:0~!~~~!~:@~3~2~~:~~~~...... '...... 1 ...... 1oc 0. 19 0.33 0.15 0 09 0-19 ;;;,;0;!47;3;:0;;$;7;;;;;;;;0;3g;;,;;;...... t::.: ...... I:::. ..t' ...... GC 0.09 0-12 0.1 1 0 07 0.12 ~;,~~~;f~,;;,;.;;;.@;gg;,;:;;;;;;j~;;$~;;;,;;j;;;~;;~~;;;;;;......

TABLE 16 Shannon's index of diversity of soil samples.

SAMPLE H' TABLE 17 Number of taxa, isolates and percent coverage of fungi in soil samples.

SAMPLE NUMBER OF TOTAL NUMBER OF TAXA COVERAGE ISOLATES TAXA SEEN ONCE (%) 1E 101 26 10 61.5 1C 133 22 6 72.7 2E 96 26 14 46.2 2C 68 10 5 50.0 6E 142 27 14 51.9 6C 119 15 8 46.7 10E 112 24 13 45.8 10C 75 22 1 I 50.0 GE 72 29 16 44.8 GC 108 27 16 40.7 3.5 DISCUSSION

Various techniques have been used to study microbial populations in soil. However, no single method will isolate all genera present, or, in the case of hngi, indicate accurately the amount of vegetative hyphae in the soil. The total number of fbngal species in any soil community (species richness) is unknown, and probably indeterminable with current methodologies. This is contrary to richness of vascular plants in temperate environments where species-area curves usually level off at some point (Christensen, 1981). With soil hngal communities species richness is directly proportional to the number of samples examined and the intensity of the study. There are two fundamental means to enumerate fungal communities; colonies can be viewed without being identified, or they can be identitied without being seen in situ. The later technique was chosen for this study by using the dilution plate method. Dilution plating is biased towards those species producing numerous spores, such as Penicillizim and often neglects slower growing species. Particular groups such as the Basidiomycetes often are very limited or do not develop on dilution plates (Waksman, 1944). However, spore production is indicative of biological activity. Spores fi-om different substrates are likely brought to the soil surface by such things as wind, insects and animals, but it is unlikely that spores of unadapted species would be recorded as important members of the soil community since they are likely destroyed quite quickly upon incorporation into the soil (Christensen, 1969). The dilution plate method will often give rise to fungal species different fiom those obtained from soil plates, hyphal isolation, isolation from rhizomorphs, sclerotZ or fruitifications (Warcup, 1965). However, the method will generally isolate the greatest number of species and is probably the best choice for a general survey such as this. In comparison with other methods of isolation, dilution plating allows fungi to be isolated from a greater volume of soil and thus species occurring on small pieces of organic or mineral matter are not as likely to be excluded or greatly overestimated (Christensen, 1969). The fact that all soils in this study were processed identically means the same sorts of fkngi should be isolated in all soils and therefore one soil can be compared to another. Thus, the dilution plate method is an attempt to reconstruct the fhgal community with viable spores. The problems inherent in this are somewhat analogous to using a seed bank to try to determine what a forest looks like. Similar to determination of soil fbngal communities, seed bank densities can be determined either by direct counting or seedling emergence (Simpson et al., 1989). Direct counting determines total number of seeds in the soil sample, but gives no information on viability. Seedling emergence methods give an estimate of viable seeds, however germination is dependant on numerous factors, such as temperature, light, soil type and oxygen availability. Thus, just as the dilution plate method underestimates the numbers of viable fbngal propagules in a sample, so too do seedling emergence technhues in estimating viable buried seed abundances (Simpson et al., 1989). Only fungi were identified in this study. Time considerations did not allow the bacteria to be identified. However, it is probable that the fbngi can be identified with less difficulty and greater accuracy. The number of described bacterial species of medical origin far outnumber those described fiom soil. Furthermore, there has been some suggestion in recent years that the number of bacterial genotypes found in nature is far greater than once believed. By isolating total DNA and determining the reassociation rate, Torsvik and colleagues (1990) estimated heterogeneity of bacterial DNA from Nonvegian soil. They determined there existed at least 4000 unique genomes of bacteria, which, according to Tiedje (1995) may suggest 20,000 to 40,000 bacterial species per gram of soil. Little has changed with respect to the difficulties in identification of soil bacteria since Chester (1900) wrote "The animal pathologist deals with a comparatively few forms which he can readily identify. The agricultural bacteriologist, on the other hand, can scarcely take up a piece of work before he meets with scores of bacterial forms of which he knows nothing, and which he is unable to identify" (Chester, 1900, cited in Lochhead and Taylor, 1938). Two hundred and thirty hngi were left unidentified in this study. In most cases this was because they were not sporulating. Ail isolates were subcultured to Weitzman and Silva- Hutner's medium (Weitzman and Silva-Hutner, 1967) and/or V8 agar (Wickerham et al., 1946) in an attempt to induce sporulation. Due to time constraints other methods were not employed to try to induce fruiting. All unidentitied hngi were rare species not isolated more than once. The types and numbers of fhgi isolated in this study showed that the fbngal community in the soils follows a model similar to the Raunkiaer pattern found in some plant communities. From Raunkiaer's Law of Distribution of Frequencies (Raunkiaer, 1934), it can be concluded that the majority of species present in a sample occur in small numbers or are unevenly distributed (Daubenmire, 1968). This pattern is similar to that found in studies of soil fungi by Bettucci and Rodriguez (1989) and Gochenaur (1984) in that there are few taxa common to all soils that are also very abundant. Most taxa are found only in one or two soils and are rare. Termite mound soils were determined to have greater similarity to each other than to control soils in species composition. This supports part of the original hypothesis of the study which was that termite mound soils would be found to have more in common with each other than with control soils. Termite mound soils were also found to have a higher species diversity than control soils. This is interesting given that fingal colony forming units were lower in termite mound soils and may indicate an uniqueness of the soils. Using Good's hypothesis, it was determined the taxa isolated accounted for 40.7%- 57.1% of the total taxa viable in the soil. This means the probability of obtaining a new isolate with the next sample is about 50%. However, as the majority of unidentified isolates were likely unique, the actual percentages may in fact be lower than this. Gochenaur (1984) and Bettucci and Rodriguez (1989) obtained much higher percent coverage values in studies of soil penicillia. The percent coverage for soils in this study is increased if the penicillia are analyzed separately. This is probably related to the biases of the dilution plate method to heavily sporulating species, such as PeniciIIium. However, the values obtained when penicillia are analyzed separately are not as high as those claimed by Gochenaur (1984) who reported 94.7-99.0% coverage for Penicillium species isolated fiom a Long Island oak-birch forest. Gochenaur's results seem questionably high and it is possible that they may have been due to overlooking rare species by repeatedly regrouping visually similar cultures. It seems Gochenaur only examined 187 of 35,560 PeniciIIium isolates microscopically. In this study almost every culture was examined microscopically. Regardless of the low values obtained for percent coverage, it is likely that the prominent species were recovered. If more soils had been processed a greater number of species would have been isolated. However, that these species were missed is probably not significant in terms of the ecology of the area and what the important members of the community are. It is doubtful that analyzing more samples would have caused a rare species to become a numerically important one. Penicilhm citrinum and P. janthinellum were the most fkequently isolated species in this study. P. citrirturn occurred almost exclusively in termite mound soils while P. janthinellum was restricted primarily to control soils. Diagnostic characteristics for P. citrimm isolates included such things as vesiculate metulae and greyish-turquoise conidial colour, yellowish reverse and pale yellow exudate on Czapek's yeast autolysate agar (CYA) amended with trace metals (Pitt, 1979) (see Appendix 2). Penicillium cifrirmm produces antibiotic compounds, most notably citrinin. Citrinin has been shown to inhibit the growth of SfaphyIococci, in particular S. aureus and S. albus (Reddy and Berndt, 1991). Synthesized analogues of citrinin and dihydrocitirnin also inhibit Bacillus slrbtilis and B. mycoides (Warren et al., 1962). Anti parasitic effects against Trypanosoma gumbiense, Trichomonas vagirzaIi.s, T. foetus and Entamoeba histolytica have also been noted (cited in Betina, 1989). Citrinin is absorbed quickly no matter how it is administered, but is nephrotoxic and due to this never became a therapeutic agent, even though it was once hailed as the antibiotic of the fiture (Betina, 1989; Reddy and Berndt, 1991). Kojic acid, another compound with antibiotic properties is also produced by Penicillium citrit~zim(Cole and Coq 1981; Domsch et al., 1980). Peniciilium janthinellum was found predominately in control soils. Diagnostic characteristics for P. janthinellum included ampulliform phialides with very long slender collula, and subglobose to pyriform roughened conidia (see Appendix 2). Interestingly, the reverse of most isolates on CYA was yellowish in colour at seven days, but would turn to reddish colours in 11-30 days. During this time sclerotia, reddish exudate and diffusing pigments were often produced as well. There are few studies with which to compare the species of hngi found in this investigation. Fungi have been identified from the inner parts of Macrotermitinae nests, but rarely fiom the outer parts. By using the dilution plate method and a general soil fungus medium, Thomas (1987a) recovered twelve species and sterile mycelium from various parts of the fungal garden and food store of a Macrotermes bel2icos1i.s mound in Nigeria. The species recorded were Absidia corymbifra, AspergiNus flaws, A. niger, A. sulphureus, A. terricola, Acremonium sp. (= Cephulo.porium), Cladosporiuum sp., Cunninghamella sp ., Ftisarium sp ., Mucor sp ., Paecilomyces variotti., Penicillium sp p. and Trzchodenna sp p .. Penicillium, Paecilomyces and Aspergillus were among the most frequently isolated genera, which is comparable to results obtained in this study. Using the same technique, Thomas (1987b) isolated hngi from the gardens of Mucrotermes szibhyalims and two species of Microtennes. The M. subhyalinus nest structure and garden yielded similar fungi as reported by Thomas (1987a) for M. bellicosrrs. MMicrotermes hngal gardens yielded fewer fungi. Temifomyces, Aspergillus fumigutus, A. nidulms, A. niger, Acremonium sp. (= CephuZo~poriuum),and Czrmdaria sp. were isolated from a Microtermes grassei fingal garden, while a Microtermes lepidtcs garden yielded Temitomyces, Acrernonium (= Cephalosporium) sp., CZadosporium sp., and Paecilomyces vmiotii (Thomas, 19876). Using the dilution plate method and 2% malt and yeast-extract- cellulose agar Zorberi (1979) isolated seven fingal species from the outer part of a Macroiemes natalenisi mound in Nigeria. These included Aspergillus niger, Peniciliium sp ., S'cephalesfrum racernosum, Trichodema sp., Rhilopus stolonifer, Neurospora sp ., and Botryodiplodia sp.. Comparison with these studies indicates that the outer casing of Macrotermitinae mounds harbor a more diverse group of kngi than previously reported, and a more diverse group than exists in the inner parts and fingal garden of the mounds. PeniciZIizim griseolum is reported here but this name had previously been synonymized with Penicilliz~mrestricfzrm by Pitt (1979). Penicillium griseolzrm was used for a number of monoverticillate isolates showing the slow growth rate typical of PenicilZiurn restricturn, but with distinct characteristics not fitting this species. Nineteen isolates were obtained and all required more than seven days incubation for sporulation to occur. Conidial colour was very grey and macroscopically the isolates did not look like Penicillium. It was difficult to make a good mount of the fbngus for microscopic examination as it tended to stick together making it hard to tdase the conidiophores apart. Conidiophores were even smaller than those of PeniciZZiurn restricturn and often would have only one or two phialides. Conidia were very rough and quite large relative to the size of the conidiophores. The isolates fit well with Smith's (1957) original description of the hngus and with line drawings in Ramirez (1982). The name PenicilZium charlesii was used for a group of isolates which superficially resembled A~pergiZZus. Pitt (1979) synonymized P. chmlesii with P. fellutamrm. The thirty-one isolates obtained from soils in this study produced both monoverticillate and biverticillate conidiophores, and had stipes or metulae with highly inflated apices (see Appendix 3). The original line drawings of P. fellutamm (Biourge, 1923), show metulae that are not inflated. The isolates fit the description and illustrations in the original description of P. charlesii (Smith, 1933), and also with the description of the type in Raper and Thorn (1949). Table 14 lists Aspergi2Zus cmtdidus #1 and A. candidus #2. Both groups of isolates keyed to this species, but were different from one another. A. crmdidus #1 had mycelium which was mostly submerged, while A. cmdidus #2 had abundant aerial mycelium, black sclerotia and yellow diffising pigments. Likewise in the Aspergilli are listed two different forms of A. niger. Aspergillus niger #2 was distinguished fiom Aspergillus niger #1 by abundant aerial mycelium, clear exudate and cream to dulI yellow sclerotia. Cultures were kept for a number of months but did not produced teleomorphs. From this investigation it is not possible to identify a £&gal "fingerprint" characteristic of termite mound soils eaten by chimpanzees. It is, however, possible to conclude that certain hngal species are a common part of the mycoflora of these sites. Most notably, Peniciiiium cifrimrn was found almost exciusively in termite mound soils and was also very prevalent, indicating it is likely an active member of the habitat. Thus, although the soils fiom each termite mound varied Eom one to the other in hngal species, all contained P. cifrimm. Also, these soils were more simiIar to each other than to control soils. Therefore, it is conceivabIe that the mycoflora present in these termite mound soils is sufficiently different £?om other available soils to be selected for by chimpanzees.

3.6 LITERATURE CITED

Batra, L. R. and Batra, S.W.1966. Fungus-growing termites of tropical India and associated fungi. Journal of Kamm Entomological Society 39: 725-738.

Betina, V. 1989.Mycooxilrr Chemical, BioIogr'caI and Environmental Aspects. Elsevier Science Publishing Company, New York, 438pp. - Bettucci, L. and Rodriguez, C. 1989. Composition and organization of the Peniciliium and its teleomorphs taxocene of two grazing land soils in Uruguay. Cryptogamie: Mycdogie 10: 107-1 16.

Biourge, P. 1923. Les moisissures du groupe Penicillium Link etude monograph. La CeIZzde 33: 1-33 1.

Christensen, M. 198 1. Species diversity and dominance in fungal communities. In: Wicklow, D.T. and Carroll, G.C. (eds.) The Fungal Community Its Organization andRole in the Ecosystem. Marcel Dekker, New York pp. 201-232.

Christensen, M. 1969. Soil microfbngi of dry to mesic conifer-hardwood forests in northern Wisconsin. Ecology 50: 9-27. Cole, R.J. and COT R.R 198 1. Handbook of Toxic Fungal Metabolites. Academic Press, New York, 937pp.

Das, S.R., Maheshwari, K.L., Nigam, S.S., Shukla, R.K. and Tandon, R.N. 1962. Micro- organisms fiom the fungus garden of the termite Odontotermes obesus (Rambur). In: Termites it1 the Hzcrnid Tropics: Proceedhgs of the New Delhi Symposizm. United Nations Educational, Scientific and Cultural ~r~anizatioh,Paris. pp. 163-166.

Daubenmire, R. 1968. Plant Commzcnities A Textbook of Plant Syneco[ogy. Harper and Row Publishers, New York, 300pp-

Domsch, K.H., Gams, W., and Anderson, T. 1980. Compendizmm of Soil Fungi. Vol. 1. Academic Press, London, 859pp.

Good, LJ. 1953. The population frequencies of species and the estimation of population parameters. Biometrika 40: 23 7-264.

Gochenaur, S.E. 1984. Fungi of a Long Island oak-birch forest II. Population dynamics and hydrolase patterns for the soil penicillia. Mycologia 76: 21 8-23 1.

Lochhead, A.G. and Taylor, C.B. 1938. Qualitative studies of soil micro-organisms I. General introduction. Conadan Journal of Research 16: 152-161.

Ludwig, J.A. and Reynolds, J.F. 1988. Statistical Ecology A Primer on Methoak and Comput!ng. John Wiley and Sons, New York, 337pp.

Magurran, A.E. 198 8. Ecological Diversity and its Measurement. Croom He1 m Limited, London, 179pp.

Moore, W.E.C. and Holdeman, L.V. 1974. Human fecal flora: the normal flora of 20 Japanese- HawaiiansJIpplied Microbiology 27: 96 1-979.

Pitt, J.I. 1979. The Gem Peniciliium and its Teleomorphic States Eupenicillzm and Talarm~yces.Academic Press, London, 634pp.

Ramirez, C. 1982. Manual and Ath of the Pe~~icillia.Elsevier Biomedical Press, Amsterdam, 874pp.

Raper, K.B. and Thorn, C. 1949. A Mamal of the Penicillia. Williams and Wilkins, Baltimore, 875pp.

Raunkiaer, C. 1934. Be Life Forms of Plants and Statistical Plant Geoqqhy. Oxford University Press, Oxford, 63 2pp.

Reddy, R.V. and Bemdt, W.O. 1991. Citrinin. In: Sharma, R P. and Salunkhe, D.K. (eds.) Mycotoxins and Phytoalexiins. CRC Press, U. S.A pp. 23 7-250. Simpson, R.L., Leck, M.A., and Parker, V.T. 1989. Seed banks: general concepts and methodological issues. In: Leck, MA, Parker, V.T. and Simpson, R-L. (eds.) Ecology of Soil Seed Ba~zh.Academic Press, California. pp. 3-8.

Smith, G. 193 3. Some new species of Penicillium. Transactions of the British Mycological society 18: 88-91.

Smith, G. 1957. Some new and interesting species of micro-fungi. Transacfionsof the British Mycological Society 40 : 48 1-488.

Tiedje, J.M. 1995. Approaches to the comprehensive evaluation of prokaryote diversity of a habitat. In: Allsopp, D., Coiwell, RR. and Hawksworth, D.L. (eds.) Microbial Diversity and Ecosystem Function. CAB International, UK. pp. 73-87.

Thomas, R.J. 1987a Distribution of Tennitomyces Heim and other hngi in the nests and major workers of Macrotennes bellicosus (Smeathman) in Nigeria. Soil Biology and Biochemistry 19: 329-333.

Thomas, RJ. 1987b. Distribution of Termitomyces and other hngi in the nests and major workers of several Nigerian Macrotermitinae. SoilBiology andBiochemistry 19: 335 -341.

Torsvik, V., Goksqr, J. and Daae, F.L. 1990. High diversity of soil bacteria. Appliedond Environmental Microbiology 56: 782-787.

Waksman, S.k 1944. Three decades with soil fungi. Soil Science 58: 89-1 15.

Warcup, J.H. 1965. Growth and reproduction of soil microorganisms in relation to substrate. In: Baker, K.F. and Snyder, W.C. (eds.) Ecology of Soil-Borne Plant Pathogens Prelude to Biological Control. University of California Press, Los Angeles. pp. 52-68.

Warren, H.H., Finkelstein, M., and Scola, D.A. 1962. The synthesis and antibacterial activity of analogs of citrinin and dihydrocitrinin. The Journul of the American Chemical Society 84: 1926-1928,

Weitzman, I. and Silva-Hutner, M. 1967. Non-keratinous agar media as substrates for the ascigerous state in certain members of the Gymnoascaceae pathogenic for man and animals. Sabouraudia 5:33 5-340.

Wickerham, L.J., Flickinger, M.H. and Burton, K.k 1946. A modification of Henrici's vegetable juice sporulation medium for yeasts. Journal of Bacteriology 52: 61 1612.

Zorberi, M.H. 1979. The ecology of some fungi in a termite hill. Mycologia 71: 537-545. CHAPTER 4 SOIL EXTRACT SUSCEPTIBILITY TESTING

4.1 ABSTRACT

In a previous investigation, termite mound soils eaten'by chimpanzees were found to have significantly higher numbers of actinomycetes if each termite mound soil was compared to its corresponding control soil. To test the hypothesis that these soils might also contain compounds with antibiotic properties, the Kirby-Bauer disk diffusion method was used to screen soil extracts against Escheria coli and Staphylococczrs aureus. Using this method, no evidence of antagonism against E. culi or S. aureus was observed

4.2 INTRODUCTION

Large numbers of soil microorganisms produce compounds with antimicrobial and/or antiparasitic activity. This likely relates to the fact that soil is an environment with high numbers of antagonistic organisms. Thus, natural selection has probably favoured those species which could defend themselves, such as by producing and excreting compounds toxic to the growth of other organisms (Vining , 1990). Fungal metabolites have been demonstrated to be responsible for plant and animal diseases and poisonings, resistance of fbngi to microbial infection, and antagonism towards other species (Gloer, 1995). Many soil fingi have also been shown to produce bacteriolytic enzymes (Thorn, 1997). The production of antimicrobial compounds may help to keep substrate resources from being depleted by other organisms so that the hngus can survive and reproduce (Wicklow, 198 1). Among the pro karyotes, the actinomycetes are great antibiotic producers, producing 75% cbf all known compounds and the majority used in chemotherapy (Kutmer, 1981). Most interest in actinomycetes over the years has been in obtaining antibacterial compounds. Thus, the search for novel species, and the mass screening of species for secondary metabolites has been a major focus of researchers and pharmaceutical companies worldwide. The capacity for secondary metabolite production by actinomycetes continues to be unequaled, however, why this is so is not clear (Goodfellow et al., 1983). In the first part of this investigation (see Chapter 2) termite mound soils eaten by chimpanzees were found to have higher numbers of actinomycetes than non-eaten, control soils. While the bacteria were not identified, there did appear to be many isolates of Streptomyces. In general it seems that areas with high numbers and diversity of Streptomycetes also have higher numbers of antibiotic producers (Kutzner, 198 1). ~urthermore,Penicilhrn citrinum, known to produce the antimicrobial compound citrinin, was found in high numbers in termite mound soils. Therefore, the purpose of this study was to do a bulk extraction of the soil samples to determine if the extractions were antagonistic against bacteria in vilro.

4.3 METEODS

Soil samples from termite mounds number one and ten were chosen to be used for extraction in this study because of the high bacterial counts which had been recorded from them previously (see Chapter 2). The Kirby-Bauer disk diffision method as outlined in Barry and Thornsberry (1991) was followed. Two bacterial species which are common agents of chimpanzee infections (see Hubbard et al., 1991) were used to screen extracts: the gram positive Staphylococnrs azrrelrs (ATCC 25923) and the gram-negat ive Escheria coli (ATCC 25 922). The antibiotics erythromycin (15 pg) and chloroamphenicol (30 pg) on commercially available filter paper discs were used in this experiment as a control.

- 4.3.1 Soil Eitractiotr : For each soil sample, an extraction using 10 g of soil and 50 ml of 1: 1 dichloromethane / methanol was performed. Soils were crushed with a mortar and pestle prior to being weighed. The soil/dichlorornethane:methanol mixtures were shaken at 230 rpm for 2 hours. They were then vacuum filtered and the filtrate evaporated to dryness using a rotary evaporator. Samples were then resuspended in 3 ml of ethanol. Filter paper discs 7 mm in diameter were made using a hole punch and were sterilized by autoclaving. Twenty discs were prepared for each extraction. To each disc, 10 p1 of the extract was applied and the solvent evaporated. This was then repeated giving a total of 20 pl of extract applied to each disc. Ethanol was also applied to 20 discs as a control. 4.3.2 InocuZation of Plates: Four or five well-isolated bacterial colonies were inoculated into tryptic soy broth and incubated at 35OC until the culture began to become turbid (= 3 hrs). The turbidity of the culture was then checked against a McFarland 0.5 standard (Barry and Thornsberry, 1991) prepared by adding 0.5 ml of 48 rnM BaClz to 99.5 ml of 0.36 N ~2~04(a McFarland standard of 1 corresponds to approximately 3 x 10' cells/ml; Acar and Goldstein, 1996). Petri plates of Mueller-Hinton agar were inoculated using a sterile cotton swab to streak the bacteria in three directions to yield a full bacterial lawn. Five plates of Mueller-Hinton agar inoculated with S. mrezis were used to test for susceptibility to chloramphenicol and another five to test for susceptibility to erythromycin. Similarly, five plates inoculated with S. arreus were used to test for susceptibility to the extract from sample one, five for sensitivity to the extract from sample ten and an additional five plates were used for ethanol control discs. In all cases four discs were applied to each plate. This procedure was repeated using E. coli as the test organism. Plates were incubated for 18 h and then examined and zones of inhibition measured. Zones of inhibition were compared to tables listing microorganisms and their expected inhibition zones with various antimicrobial agents to determine resistance or susceptibility (see Acar and Goldstein, 1996; Bany and Thornsberry, 199 1; Brown and Blowers, 1978).

4.4 RESULTS

No zones of inhibition were seen when discs inoculated with soil extracts were applied to plates of E. coli and S. meus. Commercially available discs of erythromycin and chloroarnphenicol yielded expected zones of inhibition when screened against the bacteria (Table 17; Table 18). Thus, E. coli was found to be susceptible to chloramphenicol but not erythromycin and S. aureus was found to be susceptible to both antibiotics. This experiment was replicated and the loading of discs doubled. However, again no inhibition zones were observed. TABLE 18 Zones of inhibition in nun of E coli in the presence of chlorampheniwl (30 pg) and erythromycin (15 pg). Numbers are the mean diameter of inhibition zones of four discs applied to each pIate.

GNTIBIOTXC PLATE NUMBER

TABLE 19 Zones of inhibition in mm of aureus in the presence of chlorampheniwl (30 pg) and erythromycin (IS pg). Numbers are the mean diameter of inhibition zones of four discs applied to each plate.

ANTIBIOTIC PLATE NUMBER 1 2 3 4 5 Chloramphenicol 19.8 19.5 19.8 19.5 19.5 Erythromycin 21.8 21.3 21.8 22.0 22.3

4.5 DISCUSSION

Given the inconsistencies of definitions of primary and secondary metabolites, perhaps the best definition of secondary metabolites is that they are not primary metabolites. Primary metabolites have a known fbnction in the life of the organism, while the finction- of secondary metabolites is often unknown (Griffin, 1994). Primary metabolism is usually associated with growth and is ongoing throughout the life of the organism. Relative to secondary metabolism, there are generally few intermediates produced during primary metabolism and for the most part these do not accumulate (Cooke and Whipps, 1993). Production of secondary metabolites is often associated with growth, but with the lag phase of the growth curve. Whereas primary metabolites are usually produced by all species in a genus, secondary metabolites are generally only manufactured by some members (Demain, 1996). Many secondary metabolites show antibiotic activity and accomplish this through interfering with primary metabolic processes (Vining, 1990). Secondary metabolites frequently have complicated chemical structures and although their fimction is often unknown it is widely accepted that many of these compounds play a role at the ecosystem level in microorganismal antagonism and defense (Wicklow, 198 1). Soil biologists and chemists have long had problems trying to demonstrate the in vivo production and role of antimicrobial compounds. The generally accepted idea if antibiotics are not detected in soil is that it does not necessarily mean they are not present. More probably difficulties in detection relate to adsorption by clay minerals, the sensitivity of many secondary products to chemical and biological reactions, and the relatively small quantities produced at a limited number of microsites (Soulides, 1964; 1969). Kriiger (1961) investigated the adsorption of antibiotics in soils and found the amount of antibiotic adsorbed usually increased with increasing clay and organic content. However, different antibiotics varied largely in adsorption. Penicillin and novobiocin were adsorbed only slightly while others, such as erythromycin were adsorbed more strongly. Pinck and colleagues (1961a) also studied adsorption of antibiotics by soils. These workers concluded antibiotics could be divided into three categories depending on their interaction with clay minerals and charge. Strongly basic antibiotics such as streptomycin, and amphoteric antibiotics such as bacitracin were found to form complexes in variable amounts with kaolinite, montmorillonite, vermiculite and illite. However, acid (e-g. penicillin) or neutral antibiotics (e.g. chloramphenicol) were only adsorbed by small amounts and only by montmorillonite. In a second study, Pinck and colleagues (1961b) created antibiotic clay complexes and used the cylinder-plate technique to examine bacterial zones of inhibition. This method involves adding an antibi-otidclay complex to glass cylinders 10 rnm in diameter and then placing them on an agar medium inoculated with the test organism. They found the basic antibiotics, streptomycin and dihydrostreptomycin, were released fiom kaolinite, and dihydrostreptomycin was released fiom illite. However, none of the basic antibiotics tested were released fiom moritmorillonite or vermiculite. The amphoteric antibiotics tested were released f?om all of the clays (Pinck et d., 1961b). Some workers have also found that in order to recover small amounts of antibiotics from soil, much larger amounts must be added. In a Cecil sandy clay loam soil, 900 pg of streptomycin per 100 g of soil was needed to recover the minimum amount for detection (I)In Fargo clay loam soil (montmorillonitic) the amount of streptomycin needed to recover the minimum amount was 10,000 pg per 100 g of soil (Soulides, 1965). Although the soil extracts obtained in this study did not inhibit the growth of E. coli or S. aureus, this may not mean the soils contain no compounds with antibiotic properties. As stated in chapter two, termite mound soils contained large quantities of clays. It is possible the extraction procedure did not allow compounds to be released fkom these complexes. Many variables could be changed in an attempt to deteci antibiotic activity in these soils. In this study extracts were only screened against two bacteria. It is possible the extracts were not effective against these organisms, but may in fact be antagonistic to other bacteria, hngi or parasites. The solvents used in extraction were chosen based on a paper by Weiss and colleagues (1957) and in discussion with DiCosmo (pers. corn.). However, any number or combination of solvents could have been used. Ratio of soil to solvent, time given for extraction and amount of extract applied to discs could also have been varied.

4.6 LITERATURE CITED

Acar, J.F. and Goldstein, F.W. 1996. Disk susceptibility test. In: Lorian, V. (ed.) Antibiotics in Laboratory Medicine, 4& ed. Williams and Wilkins, USA pp. 1-51.

Barry, A.L. and Thornsbeny, C. 1991. Susceptibility tests: diffusion test procedures. In: Bdows, A-, Haulser, W.J. Jr., Hemnann, K.L., Isenberg, H.D. and Shadomy, H.J. (eds.) MmaI of ClinicalMicrobioIogy- 5&ed. American Society for Microbiology, Washington. pp. 1117-1 125.

Brown, D. and Blowers, R. 1978. Disc methods of sensitivity testing and other semiquantitative methods. In: Reeves, D.S., Phillips, I., Williams, J.D. and Wise, R. (eds.) Laboratory Methods in Antimicrobial Chemotherapy. Churchill Livingston, Edinburgh. pp. 8-30.

Cooke, RC. and Whipps, J.M. 1993. Ecophysidogy oflungi. Blackwell Scientific Publications, Oxford, 3 37pp.

Demain, A.L. 1996. Fungal secondary metabolism: regulation and functions. In: Sutton, B. (ed.) A Century ofMycology. Cambridge University Press, New York pp. 233-254.

Gloer, J.B. 1995. The chemistry of fungal antagonism and defense. Canadian J'rnul of Botany 73 (SUPPI.1): S 1265-S 1274.

Goodfellow, M., Williams, S.T. and Mordarski, M. 1983. Introduction to and importance of actinomycetes. In: Goodfellow, M., Mordarski, M. and Williams, S.T. (eds.) Ihe Biology of the Actinontycetes. Academic Press, London. pp. 1-6. Griffin, D.H. 1994. Fzingal Physiology. 2nded.Wi ley-Liss, U. S. A., 458pp.

Hubbard, G.B., Lee, D.R and Eichberg, J.W. 1991. Diseases and pathology of chimpanzees at the southwest foundation for biomedical research. American Journal of Primalology 24: 273-282.

Kriiger, W. 1961. The activity of antibiotics in soil. I. ~dsorptionof antibiotics by soils. South African Journal of AgricuIfziral Science 4: 1 7 1- 183.

Kutzner, H.J. 1981. The family Streptomycetaceae. In: Starr, M.P., Stolp, H., Triiper, H.G., Balows, A. and Schlegel, H.G. (eds.) Tibe Prokaryotes A Handbook on Habitats, Isolation. mtd ldentifcation of Bacteria. Vol 11. Springer-Verlag, New York. pp. 2028 -2090.

Pinck, LA,Holton, W.F., and Allison, F.E. 1961a. Antibiotics in soils: I. Physico-chemical studies of antibiotic-clay complexes. Soil Science 91: 22-28.

Pinck, L.A., Soulides, D.A., and Allison, F.E. 1961b. Antibiotics in soils: ][I, Extent and mechanism of release. Soil Science 91: 94-99.

Soulides, D.A. 1964. Antibiotics in soils: VI. Determination of micro-quantities of antibiotics in soil. Soil Science 97: 286-289.

Soulides, D.A. 1965. Antibiotics in soil: W. Production of streptomycin and tetracyclines in soils. Soil Science 100: 200-206,

Soulides, D.A 1969. Antibiotic tolerance of the soil microflora in relation to type of clay minerals. Soil Science 107: 105-107.

Thorn, G. 1997. The fungi in soil. In: van Elsas, J.D., Trevors, J.T. and Wellington, E.M.H. (eds.) Modern Soil Microbiology- Marcel Dekker, New York pp. 63 - 127.

Vining, L.C. 1990. Functions of secondary metabolites. Annual Review of Microbiology 44: 395-427.

Weiss, P.J., Andrew, M.L. and Wright, W.W. 1957. Solubility of antibiotics in twenty-four solvents use in analysis. Antibiotics and Chemotherapy 7:3 74-3 77.

Wicklow, D.T. 198 1. Interference competition and the organization of fungal communities. In: Wicklow, D.T. and Carroll, G.C. (eds.) The Fungal Community Its Organization and Role in the Ecosystem. Marcel Dekker, New York. pp. 35 1-375. CHAPTER 5 CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE WORK

5.1 DISCUSSION

The relationship between geophagy and health has long been debated. Mason (1833) was one of the first to suggest the controversial idea that soil eating might be a cure, not a cause of disease, due to the presence of useful elements in the soil such as iron and alkalis. Geophagy in relation to anemia has been wideiy argued; some workers claiming geophagy causes anemia (Halsted, 1968; Pany-Jones and Parry-Jones, 1992), others claiming anemia causes geophagy (Crosby, 1976, 1982). Many hypotheses have been proposed to explain why geophagy occurs and what its possible health benefits are. However, overwhelming support does not exist for any one hypothesis. In all probability, there is not a single, all encompassing answer to why geophagy occurs. Given geophagy is so very widespread and practiced under a variety of circumstances by many different animals, it seems unlikely the reason would necessarily be the same in all cases. However, it is also conceivable that if the practice was detrimental or of no consequence that it might be expected to be less prevalent. For instance, it appears probable the benefits of geophagy often outweigh associated costs. Geissler and colleagues (1988) investigated soils eaten by Kenyan school children. They found termite mounds to be the most common source of geophagic material. However, this was also where eggs of the parasite Ascaris lumbricoides were most eequently found. Due to poor sanitary conditions in this area of the world, people often defecate behind termite mounds since they provide some privacy. This likely increases the parasite load in the area. Human defecation around termite mounds may not be a factor contributing to parasite load in soils eaten by chimpanzees in the jungle. However, certain monkeys are known to use the mounds as a look-off and it has been hypothesized that they also defecate there (Geissler et al., 1998). If an animal is eating soil likely contaminated with parasites, it would seem the benefits of the behaviour must outweigh the potential risk of parasitic infection. Organisms with high cognitive abilities, such as chimpanzees and humans, would likely become conditioned to avoid geophagy if it caused illness. Food generalists, such as humans and chimpanzees are at a greater risk for food poisoning than food specialists (i-e. koalas and pandas). Generalists avoid poisoning by a variety of mechanisms, including rejecting bitter- tasting substances, physiological responses such as vomiting, and through learning by experience (Nachman et al., 1977). In numerous mammals, association between food ingestion and sickness can be made over delays of many hours (Kalat, 1977). Given these facts it seems likely that soil ingestion would be avoided by chimpanzees if it caused illness. Perhaps Johns and Duquette's (1991) statements on geophagy and humans apply equally well to chimpanzees. They wrote "We believe geophagy plays a usefbl role in its proper context and should be appreciated as a normal human behavior. In traditional societies it is usually a nonpathologic activity that in appropriate circumstances can contribute to human health. Like any behavior, it can be inappropriate when engaged in out of context or to excess" (Johns and Duquette, 1991). Termite mound soils have potential for providing antibiotic compounds. High numbers of actinomycetes were found in the termite mound soils. These bacteria are prolific producers of secondary compounds and produce the majority of known antibiotics (Kutmer, 1981). Among the fungi, PeniciIIium citrimm stood out as the dominant species in termite mound soils. This species produces at least one compound, citrinin, with antibacterial and anti-parasitic properties (Betina, 1989; Reddy and Berndt, 1991; Warren et al., 1962). However, alone this evidence is insuff~cientto state chimpanzees are obtaining antimicrobial benefits from soil ingestion. If the soils had been shown to have antimicrobial properties, the significance of this would still be difficult to assess. There are three factors to consider in determining health impacts; the amount of soil eaten, whether soil particles are enriched in a certain microbe, microbial product or chemical, and if so, how much of these bound materials are available in the gastrointestinal tract (are there sufficient amounts to yield active serum concentrations?) (Sheppard, 1988). Had all other factors been determined, it would still be unknown whether the conditions of the stomach would allow these compounds to be released from the soil. This probably relates much to the type of clay mineral predominating in the soil as well as to the chemical properties of the compound. If a mineral is used to carry a pharmaceutical compound, such as an antibiotic, the compound must be able to be gradually, but readily released (Theng, 1974). Certain clays might serve as excellent vectors of secondary products, with the acidic conditions of the stomach effectively releasing the bound compounds. It is interesting to note, for example, that citrinin can be obtained fiom Penicillizim citrimm growing in liquid medium by precipitation with hydrochloric acid (Arnbrose and Deeds, 1946; Hetherington and Raistrick, 1931). However, the conditions of the stomach might not allow antibiotics to be released fiom all types of clays. If antibiotics were to remain bound they would likely pass through the system without exerting an effect upon it, somewhat analogous to giving activated charcoal after poisoning to adsorb toxins. The general conclusions which can be drawn fiom this study are as follows: most termite mound soils eaten by chimpanzees have higher numbers of viable prokaryotes than non-eaten soils and lower numbers of fungi. However, hngal species diversity is higher in ingested soils and some species seem to favour these sites. Termite mound soils show a greater similarity to each other than to control soils in fungal species composition. Penicillium citrimrm was the most fiequently isolated fingal species and was concentrated in termite mound soils. All of these results support the hypothesis that termite mound soils would differ quantitatively and/or qualitatively in microorganisms fiom control soils, and that they would exhibit greater similarity. It is possible these factors might be influencing the types of soils chosen by chimpanzees for ingestion. Further studies are needed to supplement this one to provide additional information.

5.2 RECOMMENDATIONS FOR FUTURE WORK

The findings of this investigation warrant firther studies. More work is needed to provide additional information supporting or discrediting the notion of geophagy as a means of obtaining antimicrobial compounds. A number of things are needed to increase the value of a study like this. First routine observations of chimpanzees must be made to determine how fiequently they eat soil, whether it is eaten with or after anything else and what the observable health of the individual is at the time of ingestion. The only published information to date on the health of Mahale chimpanzees at the time of soil ingestion occurs in Mahaney and colleagues (1996). Here it is stated that of five cases of geophagy by three individuals, four of those cases could be linked at the time with severe diarrhea and later to parasitic infection at the time of ingestion. Secondly an improved sampling protocol is needed. Concentration should be placed on two or three mounds formed by the same species of termite and known to be frequented by chimpanzees. Mounds should be roughly of the same age. From each mound five to ten individual soil samples should be taken. Likewise, five to ten wrrespoiiding control soils should be taken. To rule out any possibility of obtaining erosion products of termite mounds in control soils, controls should be taken at a distance greater than five meters from the mound. These changes would decrease the number of variables and should make data easier to interpret. An important area to investigate is the prokaryotic community of the ingested soils, more specifically, the actinomycetes. Usefbl information would be gained by the identification of the actino mycetes, most importantly the Streptomyces species occumng in the soils. If an antibiotic hypothesis as explanation of geophagy is to be pursued, biochemical work must be done. The extraction and screening of soils for antibiotic potential in this study was very preliminary. As was stated in chapter four, there are numerous obstacles to overcome to demonstrate the presence of antibiotics in soil. Better success would be likely with this approach if it was part of a longer study where various types of extraction procedures could be carried out and multiple organisms could be screened for susceptibility.

5.3 LITERATURE CITED

Arnbrose, A-M. and Deeds, F. 1946. Some toxicological and pharmacological properties of citrinin. Jmnral of Pharmacology and Erperimentd IXerapue tics 88: 173 - 186.

Betina, V. 1989. Mycotoxins Chemical, Biological and Environmental Aspects. Elsevier Science Publishing Company, New Yorlg 438pp. - Crosby, W.H. 1976. Pica: a compulsion caused by iron deficiency. British Journal of HaematologV 34 : 34 1-342.

Crosby, W.H. 1982. Clay ingestion and iron deficiency anemia. Annals of Internal Medicine 97: 456.

Geissler, P.W., Mwaniki, D., Thiong'o, F. and Friis, H. 1988. Geophagy as a risk factor for geohelminth infections: a longitudinal study of Kenyan primary school children. Transactions of the Royal Society of Tropical Medicine and Hygiene 92: 7-1 1.

Halsted, LA. 1968. Geophagia in man: its nature and nutritional effects. The American Journal of Clinical Nutrition 21: 1384-1393.

Hetherington, A.C. and Raistrick, H. 193 1. Studies in the biochemistry of micro-organisms part W.-On the production and chemical constitution of a new yellow colouring matter, citrinin, produced from glucose by PeniciIIiurn cifrimm Thorn. Phihsophical Truns~~ctionsof the Royal Society of London. Series B. Biological Sciences 220: 269-295.

Johns, T. and Duquette, M. 1991. Detoxification and mineral supplementation as fbnctions of geophagy. American Journal of Clinical Nutrition 53: 448-456.

Kalat, J. W. 1977. Biological significance of food aversion learning. In: Milgram, N. W., Krames, L., and Alloway, T.M. (eds.) FuodAversion Learning. Plenum Press, New York. pp. 73-103

Kutner, HJ. 198 1. The family Streptomycetaceae. In: Starr, M.P., Stolp, H., Triiper, H.G., Balows, k and Schlegel, H.G. (eds.) The Pro~otesA Handbook on Habitats, Isolaion, and Identzfzcaiion of Bacleria- Vol 11. S pringer-Verlag, New York. pp. 2028 -2090.

Mahaney, W.C., Hancock, R.G.V., Aufieiter, S., and Huffman, M.A. 1996. Geochemistry and clay mineralogy of termite mound soil and the role of geophagy in chimpanzees of the Mahale Mountains, Tanzania. Primates 37: 121- 134.

Mason, D. 1833. On atrophia a ventricula (ma1 d'estomac) or dirt-eating. Edinburgh Medical md Surgical JmmaI 3 9 : 28 9-296.

Nachman, M., Rauschenberger, J., and Ashe, J.K 1977. Stimulus characteristics in food aversion learning. In: Milgram, N. W.,Krarnes, L., and Alloway, T.M. (eds.) Food Aversian Leanling. Plenum Press, New York. pp. 105-13 1.

Parry-Jones, B. and Parry-Jones, W.L.L. 1992. Pica: symptom or eating disorder? A historical assessment. British Journal of Psychiatry 160: 341-354.

-Reddy, RV. and Bemdt, W.O. 1991. Citrinin. In: Sharma, R. P. and Salunkhe, D.K. (eds.) Mycotoxins and Phytoalexim CRC Press, U.S .A pp. 237-250. - Sheppard, S.C. 1998. Geophagy: who eats soil and where do possible contaminants go? Environmenlal Geology 33: 109- 114.

Theng, B.K.G.1974. 23e Chemistry of CZq-Organic Reactions. John Wiley and Sons, New York, 343pp.

Warren, H-H., Finkelstein, M., and Scola, D.A 1962. The synthesis and antibacterial activity of analogs of citrinin and dihydrocitrinin. The Journal of the American Chemical Society 84: 1926-1928. APPENDIX 1 MEDIA RECIPES

CZAPEK YEAST AUTOLYSATE AGAR (CYA)

Sucrose 30 g Yeast Autolysate or Extract 5g K2mo4 lg Czapek Concentrate 10 mI Trace Metal Solution 1 ml Agar 15 g Distilled Water 1000 ml

NaN03 30 g ZnS04 7&0 Ig KC1 5 g CuC12 m0 0.34 g MgS04 7&0 5 g Distilled Water 100 ml FeS04 - 7H20 0.1 g Distilled Water 100 ml

Pin, J.I. 1979. The Genus PeniciZZiurn and its Teleomorphic States EupeniciZZiurn and Talaromyces. Academic Press, London, 634pp.

DEXTROSE-PEPTONE-YEASTEXTRACT AGAR @PYA)

Dextrose Peptone Yeast Extract -03 K2mo4 MgS04 7I&O FeCI3 6H20 Oxgal 1 Sodium Propionate *gar Distilled Water

Papavizas, G.C. and Davey, C.B. 1959. Evaluation of various media and antimicrobial agents for isolation of soil fbngi. Soil Science 88: 112-1 17. MALT EXTRACT AGAR (MEA)

Malt Extract Peptone Glucose AEwr Distilled Water

Raper, K.B. and Thorn, C. 1949. A Manual of the Peiticillia. Williams and Wilkins, Baltimore, 875pp.

MODIFIED* LEONIAN'S AGAR

Maltose Malt Extract MgS04 7H20 m2po4 Peptone Yeast Extract Agar Distilled Water

* Original Leonian's formula was modified by Dr. R.F. Cain, University of Toronto, by addition of yeast extract.

Leonian, L.H. 1924. A study of factors promoting pycnidium formation in some Sphaeropsidales. American Joumui of Botany 11: 19-50.

Cain, R.F. and Farrow, W.M. 1956. Studies of coprophilous ascomycetes III. The genus Triangularia Crmadm Journal of Botuny 34 : 689-69?.

MUELLER-BUNTON AGAR

Beef Infirsion Acid Hydrolysate of Casein Starch Agar Distilled Water

* This medium is available as a premixed powder from Difco Laboratories and Oxoid Unipath.

Atlas, R-M. 1993. Handbook of Microbiological Media. CRC Press, US.A., 1079 pp. PENICILLIUM REFERENCE MEDIUM (PRM) Recipe for one liter

Glucose 9-1g NH&I (1M = 53.5M) 5 mI Major Salts Stock Solution I00 ml Micronutrients and Vitamin Stock Solution 100 mi Buffer Stock Solution 10 mi Agar 1Sg Make to volume with distilled water

MJOR SALTS STOCK SOLUTION

KC1 4.85 g MgSO4 7H20 4.93 g CaC12 2H20 - 0.441 g NaCl 0.88 g Distilled Water 1000 ml

A47CROIVUTRENTS AND P7TAMIN STOCK SOLUTTON One liter stock prepared by mixing 5 ml of each of the following:

NZL&Po4 - HzO FeCb 6H20 chelated with 17 gll of disodium salt of EDTA &BO3 MnCI - 6H20 ZnSO4 m0 Na2Mo04 2H20 coc12 6H20 Gus04 SH20 Thiamine Chloride Biotin Vitamin B12 -

BUFFER STOCK SOLUZ7ON

Tris Buffer adjusted to pH 8.0 with HCl 60.6 gA STARCH-CASEIN AGAR

Starch Casein (vitamin free) KN03 NaCl K2mo4 MgS04 El20 CaC03 FeS04 - 7&0 Agar Distilled Water

Kiister, E. and Williams, S.T. 1964. Selection of media for isolation of streptomycetes. Nature 202: 928-929.

V-STMAGAR

V-8'TMVegetable Juice 200 ml CaC03 3 g Agar 20 g Distilled Water 1000 ml

Wickerham, L.I., Flickinger, M.H. and Burton, K.A 1946. A modification of Henrici's vegetable juice sporulation medium for yeasts. Journal of Bacteriology 52: 6 11-6 12.

WEXTZMAN AND SILV-4-KUTNER'S MEDIUM

Alphacel 20 g MgS04 7H20 h3 -Po4 1.5 g NaN03 lg Tomato Paste log Pablum Baby Oatmeal log Agar 20 g Distilled Water 1000 ml

Weitzman, I. and Silva-Hutner, M. 1967. Nun-keratinous agar media as substrates for the ascigerous state in certain members of the Gymnoascaceae pathogenic for man and animals. Sabouraudia 5: 335-340. APPENDIX 2 PENICILLIUM DESCRIPTIONS

Penicillia descriptions were made uz seven days on PeniciIIium reference medium (PM)imd Czapekyeast azitolysate agar (CYA). The colozrr code usedfor descriptions was that of Komertip and Wamcher (1978). irz memirernerrts of conid& D/d refers to the mean of lengfhwidth ratios.

Komerup, A. and Wanscher, J.H. 1978. Methrren Hmtdbookof Colour. 3rd ed. Eyre Methuen, London, 252 pp.

PRM: Colonies 7-20 rnm, plane, velvety, sometimes with a tuft of mycelium in center, without exudate or with small amounts of clear exudate, with an even outline, without a well-defined margin. Sporulation moderate to heavy, dark green (25,26,27F6-7 "myrtle green"; 26F4 "bottle green"), dull green (25D3, 26E3), greyish turquoise (24E4) or dark turquoise (24F3). Without diffusing pigments. Reverse varying from uncoloured with pink, brownish orange (5C5 "topaz") or reddish brown (8D4) areas, to yellowish grey (3B2), dark yellow (4C8 "curry yellow") or brown (7E8 "henna").

CYA: Colonies 13-30 mm, felty, sometimes with a light wooly overgrowth in center, radially folded with a few or numerous pleats, without exudate or with numerous drops of clear or light yellow exudate, with an even to slightly floral or amoeboid outline and a poorly-defined to well- defined white margin. Sporulation moderate to heavy, dull to greyish green (25E4-s), dark green (25F4), greyish turquoise (24DE3-4), or bluish grey (23B3 "Persian blue"). Without diffusing pigments. Reverse ye1 lowish grey (3 B2), dull yellow (3 B3), greyish yellow (4BC4-5 "champagne", "corn", "blond", "bamboo"), yellowish brown (5F4 "beaver"), brownish orange (5C5 "topaz") or orange (5B8).

Conidiophores biverticillate with smooth to finely roughened stipes. Many monoverticillate conidiophores. Conidiophores resembling Aspergills. Metulae generally in verticils of two to four, extremely vesiculate, vesicle width 5.3-8.7 prn. Ampulliforin to acerose phialides 8.6-13-1 pm long. Conidia subglobose to ellipsoidal, smooth to finely roughened, 2.2-(2.5)-2.9 X 1.8- (2.2)-2.4 pm, Dld = 1.0-(1.2)- 1.4 (Plate 16) PeniciZfium chrysogenum Thorn

PRM: Colony 17 mm, plane, velvety, without exudate, with an uneven outline, without a well- defined margin. Diffusing pigment light yellow (IAS "sulpher yellow"). Sporulation heavy, dark blue (ZFS). Reverse greyish yellow (1B6). ,-

CYA: Colony 28 mm, felty, lightly radially folded with approximately five pleats, with yellow exudate in colony center, with an even outline and a well-defined white margin. Sporulation heavy, dark blue (23F4). Greenish yellow (1A7) diffusing pigment extending 10 mm from colony margin. Reverse greyish yellow (3337 "Naples yellow").

Conidiophores terverticillate with smooth stipes. Many biverticillate conidiophores. Metulae in verticils of two to four, somewhat divergent and inflated. Ampulliform phialides. Conidia globose to subglobose, smooth to finely roughened, 1.7-(2.2)-2.6 X 1.8-(2.2)-2.7 pm, D/d = 0.9- (1 -0)-1.2.

Peniciuium cil'rinum Thorn

PRM: Colonies 12-27 mm, plane, velvety, without exudate or with a few drops of clear exudate, with an even to uneven or amoeboid-like outline, without a well-defined margin. Sporulation poor to heavy, dull green (26,27,28DE3-4), greyish green (ZSES), greenish grey (27CD2), or dark green (25,26F3-6 "jungle green", "fir green", "bottle green"). Without diffising pigments. Reverse generally uncoloured but sometimes greyish yellow to greyish green (1BC4-5), greenish grey (1C2), yellowish grey (3B2) or olive brown (4DEF6 "honey yellow"). -

CYA: Colonies 18-35 mm, felty, radially folded with a few or many pleats, some isolates with concentric folds giving a somewhat wrinkled appearance, without exudate or with a few or numerous drops of clear, light yellow or yellow-orange exudate, with an even, scalloped or somewhat angular outline and a well-defined white margin. Sporulation poor to heavy, sometimes zonate, generally greyish turquoise (24CDE3-4; 25E4) or turquoise grey (24B2), but also dull blue (23DE4-5 "China blue"), dull green to greyish green (25DE3-5)or dark green (25F4-5 "jungle green"). Without diffising pigments or with yellow diffusing pigments. Reverse yellowish grey (2,3BC2-3 "silver white", "wax white", "oyster grey" ; 4AB2 "putty"), greyish yellow (4BC3-5 "ivory", "champagne", "corn", "beige" ; 385 "wax yellow"), yellowish white (4A2), brownish orange (5C3-6 "golden blonde", "topaz", "Pornpeian yellow"), olive yellow (3C8) or yellowish brown (5D6 "oak brown").

Conidiophores biverticillate with smooth to finely roughened stipes. Sometimes with monoverticillate and/or terverticillate conidiophores. Metulae in verticils of three to five, appressed to slightly divergent, generally of equal length, in some isolates occasionally vesiculate while in others always vesiculate. AmpulIiform phialides. Conidia subglobose to ellipsoidal, generally smooth but sometimes finely roughened, 2.0-(2.5)-3.1 X 1.7-(2.1)-2.5 pm, Dfd = 0.9-(l.L)-l.4 (Plate 17).

PRM: Colonies 23 mm, plane, velvety, without exudate, with an even outline, without a well- defined margin. Sporulation poor, difficult to characterize. Without diffusing pigments. Reverse white with dark green center approximately 12 mm in diameter.

CYA: Colonies 3 1 rnrn, radially folded with six full pleats and several partial pleats, with an even outline and a well-defined white margin. Sporulation moderate, dull green (25DE3). Without diffising pigments. Reverse yellowish grey (3B2), except for the center 10-15 mm which is dark green (25F8).

- Conidiophores bivertici Hate with smooth to finely roughened stipes. Some terverticillate-like conidiophores present. Metulae in verticils of two to five, generally unequal in length, somewhat divergent, often slightly vesiculate. Ampulliform phialides with long slender collula. Conidia subglobose to slightly ellipsoidal, smooth, 2.2-(2.4)-2.5 X 1.7-(1.9)-2.2 pn, D/d = 1.1-(1.3)-1.4 ('Plate 18).

Penicihm griseofum G. Smith

PRM: Colonies at seven days 7-10 mm and at twenty-one days 23-38 mm. Varying somewhat in appearance, plane or very slightly raised in center, felty to slightly wooly, without exudate, with an even or uneven outline, without a well-defined margin. Sporulation generally absent at seven days and sometimes taking more than sixteen days to occur, poor and confined to colony center, or moderate and evenly distributed across colony. Very grey, greenish grey (29,30F2), brownish grey (6E2) or greyish brown (5E3 "mouse grey"). Without diffusing pigments. Reverse uncoloured or yellowish white to pale yellow (3A2-3).

CYA: Colonies at seven days 8-12 mm, and at twenty-one days 34-44 mm. Felty, thick, sometimes raised in center, radially folded with five fit11 pleats and several partial pleats or with fractal-like pleats occurring only at margin. With a few or numerous drops of clear, light yellow or pinkish white (7A2) exudate, with an even or slightly uneven outline, without a well-defined margin or sometimes with a poorlydefined white margin. Sporulation generally absent at seven days, and at twenty-one days poor and confined to colony center or moderate, light grey (Dl), greenish grey (30F2) or brownish grey (SE2 "elephant skin"). Without diffusing pigments or rarely with small amounts of reddish brown diffising pigment. Reverse light yellow (4A4), greyish yellow (4B4 "champagne"), olive brown (4D4-5 "khaki") or pale to greyish orange (5AB3).

Conidiophores extremely tiny, monoverticillate, nonvesiculate with smooth stipes. Many conidiophores with only aisingle phialide and the greatest number of phialides per stipe appears to be three. Conidia varying in shape; globose, subglobose, slightly pyriform, ellipsoidal or fusiform, rough to echinulate, 2.4-(2.8)-3.3 X 2.2-(2.6)-2.8 irm, Dld = 0.9-(1.1)- 1.3 (Plate 19).

PeniciIliurnjanczewskii ~;lleski

PRM: Colonies 11-21 mm, plane, velvety, without exudate or with small amounts of clear exudate, with an even to slightly uneven outline, without a well-defined margin. Sporulation heavy, dull blue (22E4), greyish turquoise (24E34) or dark green (25F8). Without diffusing pigments. Reverse varying kern uncoloured to orange (5B7 "golden yellow").

CYA: Colonies 21-29 mm, felty, radially folded with three or four fill pleats, with a small amount of clear exudate, with an even outline and a well-defined white margin. Sporulation absent to moderate, ranging f?om bluish grey (23BC3 "baby blue", "fog blue") to dull blue (23D4). Without diffising pigments. Reverse greyish yellow (4B4 "champagne"). Conidiophores biverticillate with smooth stipes. Many terverticillate conidiophores present. Metulae in verticils of two to five, of equal length, sometimes vesiculate. Sometimes with septate metulae or phialides. Conidia green, globose to subglobose, conspicuously roughened and very Aqergillus-like, 2.1-(2.4)-2.8 X 1.9-(2.2)-2.7 pm, D/d = 1.0-(1.1)-1.3 (Plate 20).

PRM: Colonies 18-3 1 mm, plane, velvety or slightly wooly, without exudate or with a few drops of clear exudate, with an even to somewhat uneven outline, without a well-defined margin. Some isolates with sclerotia. Sporulation poor to heavy, dull green (25,26,27E3-4; 29DW-4),greenish grey (25,26,27DE2), turquoise grey (24DE2), greyish turquoise (24DE3), bluish grey (22,23 CD3 "eye blue", "fog blue") and rarely dark green (26F4 "bottle greenN). Without diffising pigments. Reverse pink (12A5) or bluish red to greyish ruby (12BC7), but most isolates with reverse in yellow colours; greyish yellow (1B3-4; 2B3-7 "wax whiten, "canary yellow"; 3B5-7 "wax yellow", "mustard yellow", "Naples yellow"; 2.3CS "linden green", "absinthe yellow"), yellowish white (2,3A2), olive yellow (2C6-7), pale yellow (3A3), pastel to light yellow (3A4-5) or dark yellow (4C8 "cuqyellow"). I

CYA: Colonies 25-42 mm, felty, sometimes slightly thick, radially folded with a few or several pleats, without exudate or with a few drops of clear or pinkish exudate, with an even to somewhat uneven outline and a well-defined white margin. Some isolates with sclerotia. Sporulation poor to heavy, dull green (25DE3-4),greenish grey (2SD2), turquoise grey (24D2), greyish turquoise (24CDE3-4) or bluish grey (22,23CD3 "eye blue", "fog blue"). Without diffising pigments. Reverse at seven days generally in yellow colours; pale yellow (2,3,4A3 "cream"), yellowish white (3A2), yellowish grey (4B2 "putty"), greyish yellow (3B4-5 "straw yellow", "wax yellow"; 4B3 "ivory"; 4C6-7 "goldenw, "brass"), olive brown (4E5; 4DE7), yellowish brown (5D5;"clay"; 5E6 "mustard brown") or brownish orange (7C4).

Colonies at 14 days52-73 mm, reverse of many isolates in reddish shades; brownish orange (5C3-4 "golden blonde"; 7C3), dull red (8C3), reddish brown (9DE5-7 "rosewood", "oxblood") or violet brown (IOEFB) Many isolates beginning to produce reddish-brown diffising pigments and increased numbers of isolates with sclerotia and pinkish exudate. Conidiophores biverticillate with smooth to finely roughened or roughened stipes and metulae. OccasionalIy with monoverticilIate and te~erticillateconidiophores present. Metulae in verticils of two to four, appressed to divergent, of equal and unequal lengths. Ampullifom phialides with long slender collula. Conidia globose to subglobose or pyrifonn, finely roughened to roughened, 2.3-(2.8)-3.4 X 2.0-(2.5)-3.0 prn, Dld = 0.9-(1.1)-1.4 plate 21).

PRM: Colonies 16-17 mm, plane, without exudate, with an even to slightly uneven outline, without a well-defined margin. Sporulation poor, generally concentrated in colony center, difficult to characterize, but probably close to greyish turquoise (24E3). Without diffising pigments. Reverse pale to greyish orange (SAB3).

Colonies at 20 days 25-28 mm, sporulation heavy, bluish grey to dark blue (23F3-4).

CYA: Colonies 21-22 mm, plane, without exudate, with an even outline, without a well-defined margin. Sporulation absent, but colony dour is orange white (6A2). Without diffusing pigments. Reverse 0range:white (6A2). I

Colonies at 20 days 67-69 mm, sporulation poor, unable to characterize.

Conidiophores biverticillate with smooth stipes. Also with many nonvesiculate monoverticillate conidiophores and numerous abnormal-looking conidiophores. Ampulliform phialides with long slender collula. Conidia varying in size and shape, globose to subglobose to ellipsoidal, rough, 1.9-(2.3)-3.1 X 1.8-(2.1)-2.7 pm, D/d = 1.0-(1.1)-1.4 (Plate 22).

PRIM: Colonies 1 1-22 mm, plane, velvety, sometimes slightly raised in places, without exudate or with small amounts of clear exudate, with an even to uneven or amoeboid-like outline, without a well-defined margin. Sporulation heavy, dark green (27,28F4-5) or dull green (25DE3;27F4). Without diffising pigments. Reverse uncoloured, yellowish white (3,4A2) or greyish orange (SB4). CYA: Colony 20-40 mm, felty, lightly radially folded or with several full pleats and many partial pleats, without exudate or with a few drops of clear or light yellow exudate, with an even or slightly floral-like outline and a well-defined white margin. Sporulation heavy, dull green (25DE3-4),greenish grey (26DE2), or greyish turquoise (24CD3; 24E3-4). Without diffusing pigments. Reverse yellowish white (3A2), yellowish grey (3B2), greyish yellow (4B3-4 "ivory", 'champagne", "sand"), dull yellow (3B3) or pale yellow (4A3 "cream").

Conidiophores extremely large, terverticillate with wide, smooth stipes and appressed rami and metulae. Often quaterverticillate or even greater branching. Metulae sometimes inflated, short. Ampulliform phialides or sometimes with rectangular-shaped phialides. Conidia varying somewhat in size and shape, subglobose to ellipsoidal or sometimes slightly pyriform or cylindrical, smooth to finely roughened or roughened, 2.2-(3.0)-4.O X 2.0-(2.5)-3 -7 pm, D/d = 1.0-(1.2)- 1.5 (Plate 23).

PRM: Colonies 20-26 mm, plane, velvety, without exudate, with an even to somewhat uneven outline, without a well-defined margin. Sporulation poor to moderate, dull green (27E3) to dark green (27F4). Without diffusing pigments. Reverse uncoloured.

CYA: Colonies 21-31 mm, felty, radially folded with five or six fill pleats and several partial pleats or sometimes with wavy or fiactal-like partial pleats. Without exudate or with a few or numerous drops of clear exudate, with an even to somewhat uneven outline and a well-defined white margin. Sporulation absent to heavy, dull green (25DE3) or sometimes greyish turquoise (24E3). Without diffusing pigments. Reverse yellowish grey (3B2), greyish yellow (4BC3 "ivory", "beige"), yellowish white (4A2) or pale yellow (4A3 "cream"), sometimes with olive brown (4DE3)areas in center.

Conidiophores biverticillate with smooth to roughened stipes. Occasionally with terverticillate- like conidiophores. Metulae in verticils of two to five, appressed, of equal length. Ampullifonn to almost acerose phialides, sometimes with long distinct collula. Conidia subglobose to slightly ellipsoidal, smooth, 1.9-(2.4)-3.1 X 1.6-(2.0)-2.1 prn, Dld = 1.0-(1.2)-1.5 (Plate 24). Peniciffiumpiceum Raper & Fennel1

PRM: Colony 19 mm, floccose, without exudate, with an uneven outline without a well-defined margin. Sporulation moderate, dark green (26F3 "fu green"). Colony uneven in colour and texture; center dark green overlaid by white flocculent mycelium, remainder of colony with light yellow (1A4) mycelium. Without diffusing pigments. Reverse varying from pale yellow to greyish yellow (2AB3 "wax white").

CYA: Colony 18 mm, slightly floccose and raised in center, without exudate, with an even outline, without a well-defined margin Colony colourfbl, pinkish mycelium in center interspersed by dark green conidia and with a yellow margin Sporulation moderate, dark green (26F4 "bottle green"). Without diffusing pigments. Reverse reddish brown (8E8 "Persian red").

Cooidiophores biverticillate with smooth stipes slightly inflated at apex. Metulae in verticils of six or more, appressed. Sometimes metulae septate or producing another conidiophore above the frst. Acerose phialides. Conidia varying somewhat in size and shape, subglobose to ellipsoidal,

smooth, sometimes with a "bleb" at one or both ends, 2.2-(2.4)-2.8 X 1.8-(2.0)-2.2 pm, D/d = 1. L(l.2)-1.4 plate 25). : I

Penicilljum restricfunt GiIman & A b bott

PRM: Colonies 8-15 mm, velvety but often growing very poorly and thinly, with numerous drops of clear to golden exudate, with a somewhat uneven outline and a diffise margin. Sporulation poor to moderate, dark green (26F4-5 "bottle green") or dull blue (23DE3). Without diesing pigments. Reverse uncoloured, white or pale yellow (3A3).

CYA: Colonies 12-17 mm, growing poorly and thinly, sometimes slightly raised in the center or radially folded with a couple of partial pleats, without exudate or with a few drops of clear exudate, with an uneven or even outline, without a welldefined margin. Sporulation absent to moderate, difficult to characterize, but probably close to dark green (26F3 "fir green"). Without diffusing pigments. Reverse uncoloured, greyish yellow (4B3 "ivory", "sand") or pale yellow to pale orange (4-943 "cream"). Conidiophores monoverticillate, generally nonvesiculate but sometimes slightly vesiculate. Smooth short stipes usually less than 10 pm long but ranging from 5-16 pm. Some conidiophores with only one phialide. Conidia globose to subglobose to pyriforrn, rough to echinulate (roughness may be occurring in bands), sometimes adhering in chains, 2.1-(2.3)-3.1 X 1.9-(2.2)-2.9 pm, D/d = 1.0-(1.1)-1.2 (Plate 26).

Penicilhm rugu~osumThorn

PRM: Colony 10 mm at ten days, sporulation absent. Colonies at eight weeks with a light dusting of sporulation.

CYA: Colony 10 mrn at ten days, sporulation absent. Colonies at eight weeks with a light dusting of sporulation

Conidiophores biverticillate with smooth stipes. Terverticillate and quaterverticillate conidiophores also present. Terverticillate conidiophores with three rami. Metulae in verticils of two to five, fairly appressed. Acerose phialides sometimes flared at apices. Conidia globose to subglobose or slightly pyridorm, finely roughened, 2.4-(2-8)-3.2 X 2.1-(2.5)-3.0 pm (Pate 27).

Penicilfium sckrotiorurn van Beyma

PRM: Colonies 18-23 mrn, plane, felty, sometimes with a light wooly overgrowth, without exudate or with limited amounts of clear exudate, with an even to somewhat uneven outline and a pale yellow (3A3) to yellow (3A6) margin. Often with orange mycelium (5A7) and sclerotia. Sporulation light to moderate, difficult to characterize, but probably greyish turquoise (24DE3), turquoise grey (24E2) or dull blue (23E4-5). Without diffusing pigments. Reverse greyish yellow (3B6 "mustard yellow"), orange yellow (4B8 "yolk yellow") or orafige (6Al38 "deep orange", "mandarin orange", "orange peel").

CYA: Colonies 25-35 mm, felty, radially folded with four or five full pleats and several partial pleats, with a few or numerous drops of clear, light orange or orange exudate, with an even or angular outline and a well-defined white margin. Sporulation poor to moderate, difficult to characterize due to orange mycelium, but probably greyish turquoise (24D3). Colonies often zonate orange (SA6 "melon yellow") and pale orange (5A3). Without diffising pigments. Reverse greyish yellow (4B5-6 "corn", "amber yellow") or brownish yellow (5C7 t'yellow ochre").

Conidiophores monoverticillate with slightly vesiculate to vesiculate stipes. Stipes smooth to finely roughened. Ampulliform phialides often with long slender collula. Conidia subglobose to ellipsoidal, smooth to finely roughened, 2.0-(2.5)-3.1 X 1.7-(2.1>2.4 pm, D/d = 1.0-(1.2)-1.4 (Plate 28).

PRM: Colonies 25-31 mm, plane, velvety, without exudate, with an even outline, without a well-defined margin. Sporulation poor to heavy, sometimes confined to center 12-14 mm of colony, dull green (25E3, 27E2-3) or olive (3F3-4 "goose turd"). Without diffising pigments. Reverse uncoloured.

CYA: Colonies 3 1-44 md felty, radially folded with a few or numerous pleats, without exudate or with a few drops of clear or pale yellow exudate, with an even outline and a narrow well- defined white margin. Sporulation absent to moderate sometimes zonate or occurring in sectors, greyish turquoise (24DE3) or bluish grey (23CDE3 "fog blue"). Without diffising pigments. Reverse yellowish white to pale yellow (4A2-3 "cream") or greyish yellow (4B3-4 "ivory", "sand", "champagne").

Conidiophores biverticillate with finely roughened to almost worty stipes and metulae. Metulae in verticils of two to five, appressed to somewhat divergent, generally equal in length but sometimes unequal. Ampulliform phialides, sometimes with long distinct collula. Conidia globose to pyrifonn, rough to echinulate (roughness may be occumng in bands), varying in size, shape and texture 2.7-(3.2)-3.8 X 2.3-(2.7)-3.2 pm, Dld = 1.0.-(1.2)-1.4 (Plate 29). PRM: Colonies 22-24 mm, velvety, plane, without exudate, with an even outline, without a well-defined margin. Sporulation heavy, dark green (25F5-8 '?jungle green", "myrtle green"). Without diffusing pigments. Reverse uncoloured.

CYA: Colonies 25-34 mm, felty, sometimes with a light wooly overgrowth in center, radially folded with five or more fill pleats, with a few or numerous drops of clear exudate, with an even outline and a well-defined white margin. Sporulation heavy, greyish-turquoise (24E4) or dull blue (23E4). Without diffising pigments. Reverse yellowish grey (3B2) or greyish yellow (4BC3 "ivory", "sand", "beige", "flaxenf').

Conidiophores monoverticillate, vesiculate with smooth stipes. Ampulliform phialides, 6.7- (8.0)-9.8 X 2.1-(24-2.9 pm. Conidia subglobose, finely roughened to roughened, varying in

size, 22.-(2.4)-3.0 X 1.9-(2.3)-2.8 pm, D/d = 0.9-(1.1)-1.2 (Plate 30).

Penicillium s p. #3

PRM: Colonies 11-18 mm', plane, velvety, with numerous drops of clear exudate, with an even outline, without a well-defined margin. Sporulation moderate to heavy, dark green (25,26F3-4 "fir green", "bottle green"). Without diffusing pigments. Reverse uncoloured but sometimes with a slight hint of pink in places.

CYA: Colonies 23-25 mm, felty, somewhat thick and deep, radially folded with a couple of fill pleats and a few partial pleats, without exudate or with a few drops of clear exudate, with an even outline and a well-defined white margin. Sporulation poor to moderate, turquoise grey (24D2) to greyish turquoise (24DE3 4). Witho~tdiffising pigments. Reverse greyish yellow (4BC3-4 "ivory", "champagne", "beige", "bamboo"), greyish orange (5B3) or greyish yellow (4B4 "champagne") in center with brown (7E6) margin, or with entire reverse brown.

Conid io p hores terverticillate with smooth to finely roughened stipes. Many biverticillate conidiophores and irregularly branched conidiophores. Metulae in vert icils of five or more, appressed or slightly divergent, of equal and unequal lengths, sometimes slightly inflated. Ampullifom phialides with long slender collula. Conidia subglobose to slightly pyriform, smooth to finely roughened or roughened, varying in size, 2.1-(2.5)-3.0 X 1.8-(2.3)-2.7 pm, Did

= O.g-(l. l>lA plate 3 1).

Penicillium sp. #4

PRM: Colonies 17-19 mm, plane,. felty, without exudate, with an even outline and a well- defined white margin. Sporulation heavy, dark green (2SFS "jungle green"). Without diffising pigments. Reverse yellowish grey to greyish yellow (3 C2-3 "smoke grey", "ash blonde").

CYA: Colonies 25-29 mm, felty to granular, sometimes slightly raised or sunken in center, radially folded with seven or eight fill pleats, with a few drops of clear to light yellow exudate, with an even to somewhat angular outline. Sporulation light, sometimes zonate, turquoise white, pale turquoise, turquoise grey or greyish turquoise (24AB2-3). Overall colonies look turquoise with light yellow (4A4) centers. Without diffising pigments. Reverse light yellow to orange yellow (4A4-7 "butter yellow", "maize", "buttercup yellow", "saffron").

Colonies at three weeks 43-64 mm, with small amounts of pastel yellow (3A3) or reddish brown diffusing pigments.

Conidiophores terverticillate with two or three rami often unequal in length. Stipes and rami finely roughened. Large numbers of biverticillate conidiophores. Metulae in verticils of three to four, appressed, 7.3-(11.2)- 13.4 X 3 -0-(3.7)-4.2pm. Ampulliform phialides. Conidia globose to

subglobose, rough, 1.6-(2.2>2.7 X 1.7-(2.0t2.6 ~III,D/d = 0.9-(1.1)- 1.4.

PRM: Colonies 18-21 mm, plane, velvety, sometimes with a tuft of sterile mycelium in center, without exudate or with a few drops of clear exudate, with an even outline, without a well- defined margin. Sporuiation heavy, dull green (25E3) to dark green (26F4 "bottle green"). Without diffising pigments. Reverse uncoloured. CYA: Colonies 24-25 mm, felty, sometimes slightly raised in center, radiaily folded with four or five fbll pleats and a couple of partial pleats or with partial pleats only, without exudate or with a few drops of clear exudate, with an even outline and a poorly-defined to well-defined white margin. Sporulation poor to moderate, in most isolates occurring randomly and in sectors, greyish turquoise (24E3-4). Without diffusing pigments. Reverse reddish brown to dark brown (8EF4; 8F8).

Conidiop hores terverticillate with smooth stipes. Bivertici Hate wnidiophores and occasionally quaterverticillate conidiophores present. Rarni unequal in length, divergent and very long (e.g. one conidiophore with rarni 52 pm and 46 prn long). Metulae in verticils of two to four, divergent, fairly long, of equal and unequal lengths. Ampullifom phialides with long slender collula Conidia globose to pyriform, smooth, 1.9-(2.5)-3.0 X l.8-(2.1)-2.4 pm, D/d = 1.0-(1.2)- 2.4 (Plate 32).

PRM: Colony 16 mm, plane, felty, without exudate, with an uneven outline, without a well- defined margin. Sporulation heavy, dark green (27F8). Without diffusing pigments. Reverse uncoloured.

CYA: Colony 23 mm, felty, radially folded with six full pleats and a few partial pleats, with several drops of clear to pale yellow exudate, with an uneven outline and a thin well-defined white margin. Sporulation heavy, dark green (26F4 "bottle green"). Without difising pigments. Reverse greyish yellow (2B3 "wax whitef').

Conidiophores terverticillate with smooth stipes and fairly wide and robust looking conidiophores. Often with three rami, appressed, of equal and unequal lengths. Metulae appressed. Conidia subglobose, finely roughened to roughened, 2.2-(2.8)-3.0 X 2.1-(2.5)-2.9

D/d = 1.0-(1.1>1.3. PRM: Colony 11 mm, colony appearance dominated by copious amounts of clear exudate and sclerotia, with a somewhat uneven outline, without a well-defined margin. Sporulation poor, difficult to characterize. Without diffusing pigments. Reverse uncoloured.

CYA: Colony 27 mm, felty, radially folded with numerous fill pleats, with copious amounts of clear exudate and sclerotia in colony center. Sporulation absent. Without difising pigments. Reverse greyish yellow (4B3 "ivory", "sand").

Conidiophores terverticillate with smooth to roughened stipes. Biverticillate wnidiophores present. Rami of unequal length and divergent. Metulae in verticils of five to siq of equal and unequal lengths, sometimes slightly inflated. Conidia subglobose to slightly ellipsoidal finely roughened, 2.0-(2.4)-2.9 X 1.9-(2.1)-2.3 ptn, D/d = 1.0-(1.2)- 1.4.

PRM: Colony 18 mm, pl'pne, without exudate, with an even outline, without a well-defined margin. Sporulation poor, only occurring approximately 6 mrn in colony center, difficult to characterize but probably close to greyish turquoise (24D3). Without diffusing pigments. Reverse brownish orange (SC5 "topaz") in center approximately 6 rnm in width, orange white (5A2) at margin.

CYA: Colony 24 mm, felty with a light wooly overgrowth in center, radially folded with four full pleats and several partial pleats, without exudate, with an even outline and a well-defined white margin. Sporulation moderate, dull green (26DE4). Without difising pigments. Reverse light yellow (4A4).

Conidiophores biverticillate with smooth stipes. Metulae generally in verticils of four, not closely appressed or widely divergent, equal in length. Ampulliform phialides, many with long slender collula Conidia subglobose to slightly pyriform, finely roughened, 1.9-(2.4)-2.9 X 1.8- (2.1)-2.6 pm, Dld = 0.9-(1.2)-1.4. PRM: Colonies 23-25 mm, plane, velvety, without exudate, sometimes with a tuft of sterile mycelium in center, with an even outline, without a well-defined margin. Sporulation moderate, dark green (26F4 "bottle green"). Without diffusing pigments. Reverse uncoloured.

CYA: Colonies 30-3 1 mm, felty, sometimes raised or with a light wooly overgrowth in center, radially folded with numerous pleats, with a few drops of clear exudate, with an even outline and a well-defined white margin. Sporulation poor, dark green (25FS "jungle green"). Without diffusing pigments. Reverse yellowish white to yellowish grey.

Conidiophores biverticillate with smooth stipes. Metulae generally in verticils of three to five, appressed, occasionally slightly inflated. Conidia subglobose, smooth, 2.1-(2.5)-2.9 X 2.0-(2.3)- 2.7 pm, D/d = 1.0-(1.1)-1.3.

PRM: Colonies 7-8 mm, sometimes slightly raised in center, with copious clear exudate and a diffuse, poorly-defined margin. Sporulation moderate but restricted to 4 mm in colony center, difficult to characterize but probably close to greenish grey or dull green (25E2-3). Without diffusing pigments. Reverse uncoloured.

CYA: Colonies 14- 15 mm, velvety, slightly raised in center, with copious clear exudate, with an even outline and a well-defined white margin. Sporulation poor, dull green (2SE4). Without diffusing pigments. Reverse yellowish white (4A2) or greyish yellow (4B3 "ivory", "sand").

Conidiophores biverticillate with smooth stipes. Some poorly developed terverticillate conidiophores. Short metulae and phialides. Ampullifom phialides, sometimes with a cellarette. Conidia subglobose to pyriform, very rough, 1.9-(2.4)-2.7 X 1.6-(2.0)-2.4 pm, Did = 1.0i1.2)-1.5. Penicilliunr sp. #I1

PRM: Colony at seven days 23 mm and at thirty-two days 60 mm. Uneven in height, with an even outline and a submerged margin. Sporulation absent at seven days and moderate at thirty- two days, greenish grey (27EF2) and occurring in secton. Without diffusing pigments. Reverse yellowish white (3A2).

CYA: Colony at seven days 29 mm and at thirty-two days 75 nun. Low and wet in appearance, with a few partial pleats, with numerous drops of clear exudate, with an even outline and a submerged margin Sporulation poor, occumng in only two small sectors, grey. Without diffusing pigments. Reverse pale yellow to greyish yellow (3AB3 "wine yellow").

Conidiophores ranging from monoverticillate to terverticillate but without any "normal"- looking conidiophores. Bizarre branching patterns. Smooth stipes. Monoverticitlate conidiophores varying from nonvesiculate to vesiculate. Biverticillate conidiophores with metulae equal or unequal in length, appressed or divergent, sometimes with only one phialide per metula. Conidia globose to subglobose, finely roughened to roughened, 2.2-(2.7)-3.1 X 2.1-

(2.7)-3.0 PITI,D/d = 0.9-(1.0)-1.1. I

PRM: Colony 16 mm, velvety, without exudate, with a floral-like outline, without a well- defined margin. Sporulation moderate, dull green (26E3). Without diffusing pigments. Reverse yellowish grey (3B2).

CYA: Colony 31 mm, felty, radially folded with five pleats, with numerous drops of clear exudate, with an even outline and a well-defined white margin. Spomlation moderate, greyish turquoise (24DE4). Reverse yellowish brown (5D4 "dark blonde").

Conidiop hores biverticillate with smooth stipes. Metulae generally in verticils of four, usually appressed but occasionally slightly divergent, of equal lengths, slightly vesiculate. Ampullifonn to almost acerose phialides. Conidia subglobose to slightly ellipsoidal, smooth, 2.0-(2.3)-2.7 X 1.7-Q.9)-2.4 pm, Dld = 1.Oil 2)- 1.4. PRM: Colony 30 mm, plane, velvety, without exudate, with a somewhat uneven outline, without a well-defined margin. Sporulation moderate, dull green (25E3). Without diffising pigments. Reverse light brown (7D5) in center, margins greyish yellow (4B3 "ivory", "sand").

CYA: Colony 35 mrn, felty, radially folded with ten full pleats and a couple of partial pleats, with a few drops of clear exudate with an angular outline and a thin well-defined margin. Sporulation heavy, greyish turquoise (24E3). Without diffusing pigments. Reverse greyish yellow (3B3).

Conidiophores biverticillate with smooth to finely roughened stipes. Monoverticillate conidiophores present. Metulae in verticils of two to three, closely appressed to slightly divergent, equal to somewhat unequal in length. Ampulliforrn phialides, some with very long and slender collula. Conidia subglobose to slightly pyriform or ellipsoidal, smooth, sometimes remaining attached to each other, 2.6-(2.9)-3.3 X 1.9-(2.2)-2.7 pn, D/d l.l-(ll.3)-1.5

Penicilfium sp. #14 I

PRM: Colonies 13 mm, plane, velvety, sometimes with a light-wooly overgrowth, without exudate, with an uneven floral-like outline and a fairly well-defined white margin. Sporulation heavy, dark green (26F4 "bottle green"). Without diffusing pigments. Reverse yellowish grey (2B2 "silver white").

CYA: Colonies 26-29 rnm, felty, radially folded with six or seven full pleats and no partial pleats or with four full pleats and several partial pleats, with a few or numerous drops of clear or light yellow exudate, with an even outline and a well-defined white margin. Sporulation moderate, dull green (25DE3). Without diffising pigments. Reverse greyish yellow (4C3-4 "beige", "flaxen", blonde", bamboo").

Conidiop hores biverticillate with smooth stipes. Terverticillate-like co nidiop hores present. Metulae generally in verticils of four, appressed, of equal length. Ampullifom to acerose phialides. Conidia subglobose to slightly ellipsoidal, smooth, 2.2-(2.5)-2.8 X 1.7-(2.0)-2.3 pm, Did = 1.1-(1.2)-1.4.

PRM: Colony 31 mm, plane, with a tuft of sterile mycelium in center, with an even outline, without a well-defined margin. Sporulation poor, confined to 15 mm in colony center, difficult to characterize but probably close to dark green (27F4). Without difising pigments. Reverse uncoloured.

CYA: Colony 34 mm, felty, radially folded with seven full pleats and several partial pleats, with numerous drops of clear exudate restricted to the outer 9-10 mrn of colony, with a slightly angular outline and a well-defined white margin. Sporulation moderate but confined to outer 4-5 mm of colony, very grey, greenish grey (28E2). Without difising pigments. Reverse greyish yellow (4B3 "ivory", "sand").

Conidiophores biverticiltate with smooth stipes. Monoverticillate conidiophores present. t Metulae generally in verticils of two to three, sometimes appressed but usually slightly divergent, of equal and unequal lengths. Ampullifom phialides, sometimes with long slender

collula. Conidia subglobose to ellipsoidal, smooth, 2.2-(2.6)-2.9 X 1.9-(2.2)-2.7 ~lm,D/d = 1.0- (1 -2t1.3.

PeniciIhm sp. # 17

PRM: Colony 15 mm, with a tuft of mycelium in center, without exudate, with an amoeboid outline, without a well-defined margin. Sporulation poor, difficult to characterize. Wthout difising pigments. Reverse uncoloured.

CYA: Colony 33 mm, radially folded with a few partial pleats and a few concentric pleats causing buckling, without exudate, with an even outline, without a well-defined margin. Sporulation absent. Without difising pigments. Reverse greyish yellow (4B3 "ivory", "sand"). Conidiophores biverticillate with smooth stipes. Metulae generally in verticils of two to four, usually slightly divergent but sometimes widely divergent or appressed, of equal to slightly unequal lengths. Ampulliform phialides. Conidia subglobose, smooth, 1.9-(2.2)-2.6 X 1.7-(2.0)-

2-5 PIII., D/d = 0.9-(1.1)-1.3.

Penicillium sp. #18

PKM: Colony 19 mm, plane, velvety, without exudate, with an even outline and a somewhat well-defined white margin. Sporulation heavy, greyish turquoise (24E4). Without diffising pigments. Reverse brownish orange (5C3).

CYA: Colony 28 mrn, felty, without exudate, radially folded with a few full pleats and numerous partial pleats, with an angular outline, without a well-defined margin. Yellowish mycelium in some areas. Sporulation poor and zonate, difficult to characterize. Without diffusing pigments. Reverse close to orange yellow (4A6 "maize").

Conidiophores biverticillate, fairly wide and with smooth stipes. A few terverticillate conidiophores present. Metu~aein verticils of four to five, appressed, slightly vesiculate. Conidia subglobose to slightly ellipsoidal, 1.9-(2.2b2.4 X 1.8-(2.1)-2.2 pm, D/d = 0.9-(1.1)-1.2.

Penicillium sp. # 19

PRM: Colonies 10-13 mm, plane, velvety, without exudate with an even to slightly uneven outline, without a well-defined margin. Sporulation moderate to heavy, dull green (29E4) or greyish green (30C3 "water green"). Without diffising pigments. Reverse uncoloured.

CYA: Colonies 19-21 mm, felty, sometimes raised in center, radially folded with numerous pleats, with several drops of clear to bright yellow-orange exudate, with an even outline and a thin, well-defined white margin. Sporulation moderate, greenish grey (25C2). Without diffusing pigments. Reverse pale yellow (4A3 "cream") to greyish yellow (4C4 "blonde", "bamboo").

Conidiophores biverticillate with smooth stipes. Mehllae generally in verticils of three to five, appressed, of equal and unequal lengths, sometimes slightly vesiculate. Phialides ampulliform to acerose, long. Conidia varying in shape; subglobose, pyriform or lemon-shaped, rough, 2.1-

(2.7)-3.2 X 1.8-(2.2)-2.7 ~IYI,D/d = 1.0-(1.2)-1.5.

PRM: Colonies 9-15 mm, plane, velvety, with copious amounts of clear exudate, with an uneven outline, without a well-defined margin. Sclerotia orange at seven days, turning darker with age. Sporulation heavy, dark green (25F6, 26F4 "bottle green"). Without diffising pigments. Reverse uncoloured.

CYA: Colonies 21-30 mm, felty, sometimes raised in center, radially folded with a few pleats and several partial pleats, with numerous drops of clear to pale yellow exudate, with an even outline and a well-defined white margin. Without diffising pigments. Reverse yellowish grey (3B2) or brownish orange (5C4).

Conidiophores biverticillate, robust-looking with smooth stipes. Stipes sometimes vesiculate at apices. Some terverticillate conidiophores with divergent rami also present. Metulae in verticils of two to five, appressed divergent, of equal and unequal lengths, slightly vesiculate to very vesiculate. Arnpulliform phialides, sometimes with long distinct collula. Conidia globose to subglobose, smooth, 2.2-(2.5)-2.8 X 1.8-(2.3)-2.6 pm, D/d = 1.0-(1.1)- 1.4.

Penicilium sp. #23

PRM: Colony 8 rnm, plane, velvety, with several drops of clear exudate, with an even outline, without a well-defined margin. Sclerotia abundant, orange-brown and immersed under mycelium. Sporulation moderate and occurring in two sectors, dull green (25E3). Without diffusing pigments at seven days, but with yellowish brown (5DS "clay") diffising pigments at 3 months. Reverse yellowish grey (4B2).

CYA: Colony 18 rnm, velvety, raised and then sunken in center, radially folded at margins, without exudate, with an even outline and a 5 mm wide white margin. Sporulation moderate but restricted to colony center, dull green (2%3 -4). Without diffising pigments. Reverse greyish yellow (4C7) and almost florescent in appearance. Conidiop hores bivertici Hate, robust and Asper'ills-li ke. Metulae sometimes very vesiculate. Ampulliform to acerose phialides. Conidia globose, subglobose or pyrifomn, rough, 2.1-(2.5)-2.8

X 1.9-(2.2)-2.5 ~III,D/d = 0.9-(1.1)- 1.4.

Penicilfium sp. #24

PRM: Colony 20 mm, plane, velvety, without exudate, with an even outline, without a well- defined margin. Sporulation heavy, dark green (26F3 "fir green"). Without diffising pigments. Reverse uncoloured.

CYA: Colony 27 mm, felty, slightly raised in center, radially folded with five fill pleats and numerous partial pleats, with an even outline, without a well-defined margin. Sporulation absent. Without diffising pigments. Reverse yellowish white (4A2).

Conidiophores biverticillate with smooth to finely roughened stipes. Monoverticillate conidiophores present. Metulae appressed or divergent, of equal and unequal lengths, sometimes vesiculate. Ampulliform phialides with long distinct collula. Conidia globose to subglobose, smooth, sometimes adhering in chains, 2.2-(2-7)-3.0 X 2.0-(2.5)-2.8 pm, Dld = 0.9-(1.1)-1.2.

Peniciffiumsp. #25

PRM: Colonies 11-13 mrn, plane, velvety, with a few drops of clear exudate in center, with an even outline and a well-defined white margin. Sporulation heavy, dark green (26F2-3 "fir green"). Without diffising pigments. Reverse greyish brown (5D3 "nougat"), or uncoloured with greyish brown center.

CYA: Colonies 15-18 mm, felty, lightly radially folded, without exudate or with a few drops of clear exudate, with an even outline, without a welldefined margin. Sporulation poor to moderate, greyish turquoise (24D3). Without diffusing pigments. Reverse yellowish white (4A2) or brownish orange (5C4).

Conidiophores biverticillate with smooth stipes. Some te~erticillate-like conidiophores present. Metulae in verticils of five or more, appressed. Ampulliform phialides with long distinct co1lula. Conidia globose, subglobose or pyriform, smooth to finely mughned, 2.6-(2.9)-

3.2 X 2.3-(2.6> 3.0 pm, D/d = 1.0<1.1)-1.3.

Penicillium sp. #26

PRM: Colony 35 mm, plane, velvety, with yellow mycelium, without exudate, with a slightly uneven outline, without a well-defined margin. Sporulation moderate, greyish turquoise (24E3). Without diffusing pigments. Reverse dark yellow to olive brown (4CD8 "curry yellow").

Coaidiophores biverticillate with smooth stipes. Many irregular-looking conidiophores present. Metulae often very long. Ampulliform phialides. Conidia large and smooth, varying in shape; generally broadly pyrifonn or lemon-shaped but sometimes subglobose, 3 -0-(3-7)-4.4 X 2.4-

(2.8)-3.2 ~III,D/d = 1.1-(1.3)-1.6.

Penicillium sp. #27

PRM: Colonies 14-15 mm, velvety, without exudate, with an uneven outline, without a well- defined margin. Sporulation poor and occurring in only two small areas at colony margins, difficult to characterize at seven days, but at sixteen days dull green (27E3). Reverse uncoloured.

CYA: Colonies 19 mm, felty, slightly buckled in center, radially folded with two to four full pleats and a few partial pleats, with an even outline, without a well-defined margin. Sporulation absent. Reverse greyish yellow (4B3).

Conidiophores biverticillate with smooth to finely roughened stipes. Metulae in verticils of six or more, short and appressed. Phialides crowded; difficult to determine shape. Conidia subglobose, smooth, 2.q2.3)-2.7 X 1.8-(2.1b2.4 pm, D/d = 1.0-(1.1>1.3. Penicillium sp. #28

PRM: Colony 12 mm, velvety, with copious amounts of clear exudate, with an even outline and a well-defined white margin. Sporulation heavy, dark green (25F4). Without diffising pigments. Reverse uncoloured.

CYA: Colony 17 mm, felty, radially folded with seven full pleats and also with two concentric pleats, without exudate, with an even outline, without a well-defined margin. Sporulation poor, difficult to characterize. Without diffusing pigments. Reverse brown (7.6= "eye brown").

Conidiophores biverticillate with smooth stipes. Monoverticillate and terverticillate wnidiophores also present. Metulae varying from closely appressed to widely divergent. Ampulliform phialides, sometimes with long slender collula Conidia globose, subglobose or pyriform, finely roughened, 2.4-(2.9)-3 -4 X 2.3 -(2.6)-3.2 pm, D/d = 0.9-(1.1)- 1.3.

Penicillium sp. #29

PRM: Colony 15 mm, plane, with a few drops of clear exudate, with an even outline, without a well-defined margin. Sclerotia present. Sporulation absent at seven days but at sixteen days dull green (27E3). Without difising pigments. Reverse unwloured but with a yellowish white (4A2) to greyish yellow (4B3 "ivory", "sand") center.

CYA: Colony 21 mm, felty, slightly raised and then sunken in center, without exduate, with an even outline, without a well-defined margin. Sporulation absent. Reverse pale yellow (4A3 "cream ").

Conidiophores biverticillate with finely roughened stipes. Metulae in verticils of five, generally appressed and of equal lengths. Ampulliform to acerose phialides. Conidia subglobose, smooth,

2.242.3)-2.5 X 1.8-(2.0)-2.3 ~IYI,D/d = 1.0-(1.2)- 1.3. Penicifliumsp, #30

PRM: Colonies 27-31 mm, plane, velvety, without exudate, with an even outline, without a well-defined margin. Sporulation heavy; dull green (27E4; 29E4) or greyish green (29E5). Without difising pigments. Reverse dark green (27-29F5-6) or olive (2EF5).

CYA: Colonies 22-34 mm, felty, often with a light wooly overgrowth, radially folded with four fill pleats and sometimes with a few partial pleats, with a few or several small drops of clear to light yellow exudate, with an even outline and a well-defined white margin. Sporulation moderate to heavy, greenish grey (26-27DE2-3) or dull green (25DE3). Without diffusing pigments or with light yellow (1A4) difising pigments. Reverse olive (3EF5) with a pale yellow margin (3A3), dull green (29DE4), or greyish green (29E5) with a yellowish-white margin (3 A2).

Conidiophores biverticillate with finely roughened to roughened stipes. Metulae in verticils of four or more (at least six can be distinguished), appressed. Acerose phialides. Conidia

subglobose, smooth to finely roughened, 2.2-(2.5)-2.8 X 2.0-(2.2)-2.5 JL~,D/d = 1.0-(1.1)-1.3 (Plate 33). i

~enici~liurnsp. #3 1

PRM: Colony 15 mm, with a few large drops of clear exudate and large flesh coloured sclerotia, with an even outline, without a well-defined margin. Sporulation poor, difficult to characterize. Without difising pigments. Reverse uncoloured.

CYA: Colony 32 mm, slightly raised in center and radially folded at margin, with copious amounts of clear exudate and abundant flesh to grey coloured sclerotia. Sporulation and difising pigments absent. Reverse close to brownish orange or yellowish brown (5CD4 "golden blonde", "dark blonde").

Conidiophores biverticillate with finely roughened stipes. Some terverticillate-like conidiophores present. Metulae generally in verticils of three to four, somewhat appressed. Ampullifom phialides. Conidia subglobose to ellipsoidal, smooth, 2.0-(2.4)-2.9 X 1.6-(2.1)-2.5 pm, D/d = 0.9-(1.1)-1.5.

Eupenicillium shehi S tolk & Scott

PRM. Colonies 9-19 mm, colony appearance dominated in most isolates by abundant cleistothecia and copious clear exudate. With an uneven outline, without a well-defined margin. Sporulation poor to moderate, greyish turquoise (24BCD3 "aquamarine"). In most isolates colony colour dominated by cleistothecial colour; orange grey (6B2 "birch bark), reddish grey (7,8B2) greyish red (7B3) or medium grey (3E1). Without diffising pigments. Reverse ye1 bwish white to yellowish grey (2,3,4A2).

CYA: Colonies 15-25 mm, thick, colony appearance dominated in most isolates by abundant cleistothecia and copious clear exudate. Radially folded, some isolates with concentric folds giving a wrinkled appearance, with a somewhat uneven to scalloped outline and a well-defined white margin. Colony colour dominated by cleistothecial production in the center with spomlation at margin. colony centers reddish grey to brown grey (7BC2). Sporulation poor to moderate, greyish turquoise (24BCD3 "aquamarine") or bluish grey (23BC3 "baby blue", "fog blue"). Without diffising pigments. Reverse dull yellow (3%2), yellowish white (4A2), pale yellow (4A3 "cream"), yellowish grey (3,4B2 "putty") or greyish yellow (4BC3 "ivory", "beigelt).

Asci round with eight ascospores, not produced in chains. Ascospores ellipsoidal, smooth, with two longitudinal ridges, 2.1-2.8 X 1.5-2.0 pm.

Coaidiophores biverticillate with smooth to finely roughened stipes. Some isolates with an occasional or a few terverticillate conidiop hores with widely divergent rami. Metulae in verticils of two to five, appressed to slightly divergent, generally equal in length, sometimes slightly vesiculate. Phialides varying from arnpullifonn to acerose. Conidia subglobose to ellipsoidal, smooth to finely roughened, 1.9-(2.3)-2.8 X 1.5-(1.8)-2.3 pm, D/d = 1.1-(1.3)-1.6 (Plate 34). APPENDIX 3 LINE DRAWINGS AND PHOTOGRAPHS ( I Absiaia cylindrospora Hagem

Fig. 1 Sporangiophores with intact and deliquesced sporangia Scale bar = 20 prn Fig. 2 Sporangiospores Scale bar = lOpn Fig. 3 Zygospore Scale bar = SOW PLATE 2 Cunninghamelk sp.

Fig. 1 Sporangiophores with sporangioles ScaIe bar = 20 p.rn Fig. 2 Sporangioles Scale bar = 10pm PLATE 3 Glwccadium sp. #I

Fig. 1 Conidiophores Scale bar = 20 pm Fig. 2 Conidia Scale bar = 10p PLATE 4 Glwcladurn sp. #2

Fig. 1 Conidiophors Scale bar = 20 pn Fig. 2 Conidia Scale bar = 10pm PLATE 5 Gongrunella butleri (Lendher) Peyronel & Dal Vesco

Fig. 1 Sporangiophores with intact and emptied sporangia showing apophysate columellae. Scale bar = 10 pm Fig. 2 Sporangiospores Scale bar = lop PLATE 6 Mucor cf. racernosus Fres.

Fig. 1 Sporangiophores Scale bar = 20 pm Fig. 2 Sporangiospores Scale bar = 10 pm Fig. 3 Chlarnydospores Scale bar = 10 prn PLATE 7 Mucor sp.

Fig. 1 Sporangiophores Scale bar = 20 pm Fig. 2 Sporangiophores Scale bar = 50 pm Fig. 3 Sporangiospores Scale bar = 10 p PLATE 8 Neclria anamorph

Fig. 1 Conidiophores S4ebar = 20 pn Fig. 2 Pigmented hyphae Scale bar = 20 pin Fig. 3 Conidia Scale bar = 10 pm PLATE 9 Oidiodendron griseum Robak

PLATE 10 Paecilomyces carneus (Duch6 & Heim) A.H.S. Brown & GSm.

Fig. 1 Conidiophores Scale bar = 10 jmt Fig. 2 Conidia Scale bar = 10 prn PLATE 11 scopulariopsis sp.

Fig. 1 Conidiophores Scale bar = 10 p Fig. 2 Conidia ScaIe bar = 10 prn Trichoderma crassurn Bissett

Fig. 1 Conidiophores Scale bar = 20 pm Fig. 2 Conidia Scale bar = 10 pm PLATE 13 Trichodenna harzianum Rifai

Fig. 1 Conidiophores Scale bar = 20 pm Fig. 2 Conidia Scale bar = I0 pm - Fig. 3 Chlamydospores Scale bar = 10 pm Fig. 1 Conidiophores Scale bar = 20 pm Fig. 2 Conidia Scale bar = 10 pm Fig. 3 Chlamydospores Scale bar = 10 pm

PLATE 16 Peniciliium charlesii Smith Fig. 1 Monoverticillate conidiophores, x 1000 Fig. 2 Biverticillate conidiophores, x 1000 Fig. 3 Conidia, x 1000 PLATE 17 PeniciIIium cifrinum Thorn Fig. 1 Conidiophores, x 1000 Fig. 2 Conidia, x 1000 PLATE 18 Peniciliwn cJ corylophilum Dierchx Conidiophores, x 1000 PLATE 19 Pewiciilium griseolum G. Smith Fig. 1 Conidiophores, x 1000 Fig. 2 Conidia, x 1000 PLATE 20 Penicillium janczewskii Zaleski Conidiophores, x 1000 PLATE 21 Peniciilium juts thinellum Bio u rge Fig. 1 Conidiophores, x 1000 Fig. 2 Conidia, x 1000 PLATE 22 Penicifliumjensenii Zaleski Fig. 1 Conidiophores, x 1000 Fig. 2 Conidia, x 1000 PLATE 23 Penicilfium olsonii Bain. & Sartory Fig. 1 Conidiophores, x 1000 Fig. 2 Conidia, x 1000 PLATE 24 Penicillum paxilli Bain. Conidiophores, x 1000 PLATE 25 Penicifiium piceum Raper & Fennell Fig. 1 Conidiophores, x 1000 Fig. 2 Conidia, x 1000 PLATE 26 Penicr'llium restricturn Gilman & Ab bott Fig. 1 Conidiophores, x 1000 Fig. 2 Conidia showing various degrees of roughness, x 1000 PLATE 27 Penicilliurn rugdosum Thorn Conidiophores, x 1000 PLATE 28 Peniciilium sclerotiorum van Beyma Fig 1. Conidiophores, x 1000 Fig 2. Conidia, x 1000 PLATE 29 Peniciilium sinzplcisssinrunr (Oudem.) Thorn Fig. 1 Conidiophores, x 1000 Fig. 2 Conidia showing variation in shape and roughness, x 1000 PLATE 30 Penicillirrn sp. #1 Conidiophores, x 1000 PLATE 31 Penicillium sp. #3 Conidiophores, x 1000 PLATE 32 Peniciiiium sp. #5 Fig. 1 Conidiophores, x 1000 Fig. 2 Conidia, x 1000 PLATE 33 Peniciliium sp. #30 Conidiophores, x 1000 PLATE 34 Eupenicillium shearii Stolk & Scott Fig. 1 Conidiophores Fig. 2 Conidia Fig. 3 Asci