The Ants of Africa Chapter 1 - Introduction

The past forty years have seen greatly increased interest in tropical forests and in the crops grown in once-forested areas. Among the most abundant animals in the tropics are the ants (Hymenoptera: Formicidae). For instance, Entwistle (1972) related findings from Ghana (by Leston & Gibbs) and from Nigeria (by Booker) which showed that as many as 87% of the total number of individual insects found on cocoa are ants. This, coupled with the colonial lifestyle of many species and their activity as predators, surely makes them major components of both the terrestrial and arboreal ecosystems.

Tree crops, especially cocoa, are of great economic significance in the forest zone of Nigeria and the ants which can be found on all the trees profoundly affect both pests and diseases of the crops. As members of a team studying Black Pod disease of cocoa, caused by one of two fungi, Phytophthora palmivora (Butler) Butler, or the closely related Phytophthora megakarya Brasier & Griffin, my staff and I made extensive collections and surveys of ants at the Cocoa Research Institute of Nigeria (CRIN) and elsewhere in the cocoa growing areas of western Nigeria. Our findings on the distribution and inter-specific relations of the ants and their role in the annual Black Pod epidemic appeared in the Final Report of the Project (Gregory & Maddison, 1981) and in specific research papers (Taylor, 1977; Taylor & Adedoyin, 1978). Although the number of ant species which were both abundant and involved in Black Pod disease was found to be quite small, we encountered some 100 species foraging or nesting on cocoa trees and another 50 species in cocoa plantations or in the surrounding forest. This contrasted greatly to the report by Booker (1968), who collected only 19 ant species in repeated surveys of the tree canopy of two blocks of cocoa at CRIN, despite using regular pyrethrum knockdown (pkd) for nearly two years, and the list of 27 species given by Eguagie (1971).

The problem of identifying tropical insects is widely recognised and ants are no exception. Strickland (1948), whose bionomic studies in Ghana were among the earliest on the ants of West Africa, described how -

"Nearly 50 species of ant have so far been taken attending P. njalensis colonies in the field" but "Only nine of the species have so far been identified".

In Nigeria, during 1974-76, the output from our daily surveys and monitoring work included a bewildering array of small brown-black moving objects, ants, some or all of which might be playing a role in the transmission of cocoa black pod disease. In order to satisfy the crucial need to rapidly separate the numerous ants, I started to make small annotated illustrations of the worker and soldier castes of each species. Two slices of good fortune combined, however, to ease the task of naming the species. First, CRIN at that time had a moderate collection of ants which had been curated by Barry Bolton, now one of the world's great ant taxonomists. Secondly, Bolton, by then at the Natural History Museum, also had just published keys to the genera of West African ants (Bolton 1973a). The combination enabled me to separate out all the specimens to genus level. The labels on CRIN species and the first of Bolton's revisionary studies (Bolton, 1969, 1970-71, 1971a, 1971b, 1972, 1973, 1974a, 1974b, 1975a, 1975b) enabled me to assign specific names to a good proportion of the taxa we collected. In many instances, separating them to species level still meant either using the code letters or numbers of specimens in the CRIN collection, some of which we did not encounter, or devising further code designations. The major problem at that time, however, was the poor state of taxonomy, especially at the species level, and over 50 of the species could be labelled only as "species 1, species 2, species A, species B, etc.", my additional species were labelled "species T¹, species T², and so-on". The literature on research in Ghana, which is referred to extensively later in the text, shows similar code numbers, etc., for many species, probably reflecting the fact that CRIN originally was an outstation of the main institute in Ghana (first known as the West African Cocoa Research Institute; now The Cocoa Research Institute of Ghana, CRIG). The source of some of the codes, moreover, may have lain elsewhere, especially the British Museum (Natural History). For instance, Kemp (1951), reporting work on Pheidole in (now) Tanzania, listed two species as Pheidole species A and Q, with the identifications being attributed to H. St. J. Donisthorpe of that Museum. The latter also had written two short papers which dealt with ants collected by workers based in Ghana (see Chapter 2). The abundance of such coded "forms" can be seen by a quick glance at the two Taxonomic Name Indices which deal with Forms A to N and Forms O to Z.

The initiative, in 1975, to begin publication of a CRIN Research Bulletin encouraged me to turn my illustrations into a proper field guide which could be used by anyone needing to identify ants of the area. The original monograph, Ants of the Nigerian Forest Zone, was written in early 1976, and appeared in five parts (Taylor 1976, 1978, 1979, 1980a, 1980b). Advances in ant taxonomy since that time have been considerable and an update has become both desirable and possible. Extensive revisionary work, mostly by Barry Bolton, has given proper names to many of the species and synonymized a number of others.

Even with all the taxonomic work, however, several important, perhaps the most important, genera remain inadequately classified. The most urgent needs are for revisions of Camponotus, Crematogaster and Pheidole. Wilson (1976) had speculated that the three genera had 641, 274 and 391 members respectively. The three genera form what Hölldobler & Wilson (1990) termed the crux myrmecologorum, but they noted that the extreme example, Pheidole, is thought to have more than 1,000 members, two and a half times as many as was thought only 14 years earlier. Thus, even very recent reports of ant-related research in West Africa, such as Campbell (1994), have had to resort to using the name Pheidole megacephala to cover several, perhaps as many as eighteen, morphologically and behaviourally similar taxa, and Belshaw & Bolton (1994b) listed eleven undetermined species. The magnitude of the taxonomists' task was illustrated by Bolton (1981b), when he described the then state of knowledge of the subfamily Myrmicinae. There were some 43 genera recognised from sub-Saharan Africa; of which 29 genera had 1-10 species, 9 had 10-40 species and 5 had more than 40 species. Of the last, Tetramorium, Pheidole and Crematogaster were thought to have well over 100 species each in the region. His own revisionary studies had enabled him to recognise 176 species of Tetramorium (Bolton, 1980) and 33 species of Triglyphothrix (Bolton 1976, the genus was synonymized subsequently with Tetramorium, Bolton, 1985). Recently, Wilson, together with Bill Brown of Cornell University, announced that a start is being made on the task of preparing a monograph on Pheidole (Hölldobler & Wilson, 1994).

Taking the foregoing, I decided to adopt a two-pronged approach. This was to compile the fullest possible taxonomic catalogue, illustrated by my numerous drawings of species from Nigeria, and to pull together all useful knowledge of the biology and strategic importance of ants in the Forest Zone and the adjoining Guinea Savannah Zone of West Africa. Although my illustrations cover only some 40% of the known, described species, they do show almost all the relatively common species and at least one representative of most of the genera. Perusing the rather battered copy of Wheeler (1922) held by the Royal Entomological Society led me to decide to incorporate relevant drawings (originally by Mrs Helen von Ziska, and denoted by W '22) and illustrations both to complement this text and to further enhance their value for posterity. To aid those who wish to look further into the literature and to identify the many other species; wherever possible, I have drawn on modern publications to provide Keys to genus members. Then I have added brief notes on the particular distinguishing features of each of the listed species, including indicating illustrations in the reference literature. If available, information on habitats, nest sites, etc. is provided. Unless indicated, the information comes from the collections by my staff and myself. The emphasis remains on ants in cocoa farms simply because of the paucity of other research.

A final justification, if any is needed, for this book is to bridge the gap between anglophone and francophone entomologists. While in Nigeria, I met entomologists from Ivory Coast and Cameroun who were working on cocoa problems but I have to admit to being ignorant of the work under way at Lamto in the Ivory Coast, although that was in its earlier days. My contemporary compatriots, such as Mike Bigger and Colin Campbell, who were studying aspects of cocoa entomology in Ghana and published a little after my main papers, make no reference to work published in French, excusable perhaps as they were not dealing specifically with ant mosaics or general biology. In converse, moreover, Lévieux (1983) wrote how little or nothing had been written on the biology of Myrmicaria, whereas Myrmicaria striata featured in my papers (notably Taylor & Griffin, 1981).

Receipt (in mid-1999) of a copy of Francis Bernard's treatise on the ants of Mt. Nimba, Guinea, over 100 pages of fascinating text, showed again how much the English-speaking world has failed to appreciate texts written in other languages. I have endeavoured to do justice to Bernard's work and, thus, have included (my) translations of his descriptions of previously new species and of many of his commentaries. He examined several of the major museum collections of ants and I have taken his naming of species as valid, especially as, obviously, that is better than the code designations I have given for many unassigned forms. Given my own difficulty in obtaining a copy of his treatise, I have also incorporated Bernard's illustrations (distinguishing them by "B'52").

For those interested in ants there is no better starting point than the magnificent book The Ants by Bert Hölldobler and Edward O. Wilson (Hölldobler & Wilson, 1990). An easier read but no less fascinating is its companion book Journey to the Ants (Hölldobler & Wilson, 1994). Contemporaneously, the narrower but fascinating aspect of ant morphology and taxonomy has been well served by Barry Bolton (1994), whose great book Key to the Ant Genera of the World has 522 superb photographs, taken with a Scanning Electron Microscope by Laraine Ficken, and showing almost all recognised genera. Complementing that work now is his comprehensive A New General Catalogue of the Ants of the World, in which one finds a full list of what, at 31 December 1993, were 16 subfamilies, 296 genera, and 9536 species (Bolton, 1995). More than that, however, he gives extensive coverage of subspecies, junior synonyms, junior homonyms and unavailable names. The whole is completed by references to all the source papers for the names and many ancillary papers. All three have proved very useful to me in writing this text.

Coming across the website of the Japanese Ant Color Image Database filled me with admiration for the outstanding quality of the colour images. Importantly, the "Database" includes several "tramp species" found in West Africa, plus some genera not pictured on my site. Thus, I have added links to my site which, writing this on 1 January 2000, show how the internet really can open up the guarded world of the museum specialists.

The Third edition - 2001 showed how readily the medium of electronic publishing enables the addition, enhancement and improvement of a "publication". Contributions and support from fresh contacts, such as Professors Doyle McKey and Alain Dejean are enabling me to progress the site in ways I had not envisaged. For instance, the receipt of numerous specimens from the Doyle McKey team has enabled me to add a considerable amount of information and scanned images of the species. The specimens include several species proviously recorded only from the Congo Basin and, so, I have decided to expand the overall list by adding records of species known from that area.

What I have decided now (late-2001) is to endeavour to add original descriptions from publications from as much as a century or more ago. This can be seen in my translations and use of the key to Myrmicaria written by F. Santschi in 1925. I have also added descriptions from Wheeler (1922) - prompted by my identifying several species in the Doyle McKey team collection which previously had been reported from Zaïre or elsewhere in the Congo Basin.

Fourth edition - 2002

The catalogue now includes species known from the Congo Basin (most of Zaïre, Congo and Gabon), including information from the writings of F. Santschi and A. Forel. Other fresh material is from a paper by Cedric Collingwood and a very interesting set of specimens of Driver Ants sent to me by Tatyana Humle, of the University of Stirling, who is studying chimpanzee behaviour, especially the use of tools, such as for "ant-dipping". I have added a spreadsheet summary catalogue listing 902 definitive species and 268 forms denoted by taxonomists from modern surveys, etc. Another update is the inclusion of an "album" of "scanned" images of ants from the Cameroun and Guinea.

Fifth edition - 2003

I have continued the process of revision and updating. Notably, I have included a link to the Hymenoptera Name Server from almost all of the definitive species (essentially those listed in Bolton, 1995). Note that some of my own suggested revisions may not be incorporated on the server. The link appears as and clicking the image will take you to the appropriate page of the server (as an example, the link here leads to the species Tetramorium aculeatum).

Finally, much has been written in recent years about ants but I would like to convey something of my own fascination with these widespread and, in the tropics, ubiquitous insects.

Although most ants appear to be brown or black, one of the most notable variations is bicolouration. Typical of this is Monomorium bicolor, with an orange head and thorax but a dark gaster; a reverse pattern is shown by Crematogaster clariventris, with a distinctive orange gaster. More vivid is the gold and silver pubescence and long hairs of some Polyrhachis, which is made more attractive by their remarkable sculpturation and large spines.

Other remarkable forms are - the slender bodied Tetraponera with large eyes, obvious high visual acuity and rapid evasion; the sculptured, arboreal Cataulacus, again with large eyes well able to see attackers but able to roll into a ball with nothing exposed, and, if still attacked, dropping off the tree; and, the furry little "teddy bears", of Meranoplus inermis, Tetramorium brevispinosum (formerly Triglyphothrix, with bizarre divided hairs), and the species which I labelled Acantholepis T²´³ now Lepisiota cacozela .

No-one can fail to be impressed by the overwhelming power of the rivers of Army Ants, genus Dorylus subgenus Anomma, which for me was most vividly illustrated by a column perhaps 20-30 ants wide emerging from a teak plantation and flowing along the path beside my Onipe 1/1 study area for some 200 m before re-entering the secondary forest. Much of the trail was covered with soil and vegetable debris, but open spaces were bridged by the larger ants especially the soldiers. The column was moving and continuous when I arrived to check the cocoa plots and was no different some two hours or more later when I came back to the path. When they are not migrating, one learns of their activities by sudden sharp pains to the nether regions. For me, working in cocoa meant looking upwards, so blundering into a Doryline foraging area not infrequently resulted in a pressing desire to dash some distance, down trousers (I gave up shorts in tropical field work sometime before) and rapidly remove all the ants which were biting with their needle-like mandibles.

The well-developed stings of ponerines and myrmicines are rarely experienced but once, when climbing a step ladder in the canopy, I brushed my face against a leaf with a Tetramorium aculeatum nest, resulting in a combination of pain and numbness in my right cheek for several minutes. The attack pose of Crematogaster species is to turn their tails over their head. In contrast, their cousins the Myrmicaria always keep their tails held tucked under the hind legs - one wonders, being soil-dwelling, do they expect an attack from below? Other posers are the Odontomachus with their mandibles outspread like the moustaches of RAF pilots from World War II, but able to snap the moustache shut against the substrate and propel themselves backwards.

Well-known are the Oecophylla, or Weaver Ants, which use larval silk to bind live leaves into a ball to form their nest. They are very sharp-eyed watching anything that approaches their territory and confronting the intruder with their tail above their head ready to spray the defensive formic acid.

Polymorphism is at its most extreme in Dorylus from the small (3-4 mm) relatively inoffensive workers through several castes to the large and very vicious soldiers (10-12 mm), with needle-like mandibles. Among the myrmicines examples include Atopomyrmex species, with the largest morph being bulkier and much more coarsely sculptured, and Crematogaster (Atopogyne) species, for instance Crematogaster africana, with a range of sizes but not of shape. Then there are the dramatic soldiers of many Pheidole species, with almost absurd big-heads; the overall size being up to twice that of the workers but the heads perhaps three times bigger and bulkier. Formicines display no more than dimorphism, such as Camponotus species with large headed majors, perhaps one and a half times bigger than the minors. Also dimorphic but with the smaller caste rarely seen away from the nests, is Oecophylla longinoda.

The ability of ants to build nests and other structures also reveals remarkable variations in form and sites. The latter range from inside little dead twig ends, like the apparently highly successful tiny (1.3 mm) Plagiolepis brunni which despite, or perhaps because of, its small size, could be found on nearly all the cocoa farms in Nigeria and on some 8% of the trees; to the massive structures of carton, the millions of years old forerunner of our chipboard, built high on forest trees by Crematogaster africana and Crematogaster depressa. Silk is used to bind leaves by Oecophylla longinoda and to bind various vegetable fragments by Polyrhachis species. Those ants which live in the soil reveal little of their nest structure but the tents built over Homopteran "cattle" by Myrmicaria striata show strong cementing of the soil particles, this is revealed also in their semi-subterranean trails from the nest in open ground to the trees. Ponerines clearly move soil in large crumbs, seen in the pyramidal nest entrances of Pachycondyla tarsata and Pachycondyla analis, and the crude tents built by Odontomachus troglodytes. Also revealed by the nest entrances of Pachycondyla tarsata is their diet of insects and other arthropods, seen as skeletal remains. Food remains appear also in the debris nests in tree crevices of some semi-arboreal species of Pheidole.

Diet related modifications abound. The tent-builders clearly use Homopterans (aphids, mealybugs, hard scales, psyllids, etc.) as sources of sugar. General carnivores such as many ponerines have simple but powerful mandibles. Many species, however, have elongated mouthparts used for seizing and manipulating smaller soft-bodied prey, such as the Collembolans (springtails) known to be the food of the Strumigenys species. Others with long mandibles are Leptogenys, Plectroctena and Psalidomyrmex.

Ants are well known for their well-developed systems of communication. Most important among those systems is the use of chemicals and a useful summary of the state of knowledge in the early 1980s was provided by Bradshaw & Howse (1984). West African ants included in their review are Camponotus sericeus, Myrmicaria natalensis (as Myrmicaria eumenoides) and Myrmicaria striata, Odontomachus troglodytes, Oecophylla longinoda, Pachycondyla analis (as Megaponera foetens), Pachycondyla soror (as Bothroponera soror), and Pachycondyla tarsata (as Paltothyreus tarsatus).

©1998-2000, 2001, 2002, 2003, 2006 - Brian Taylor CBiol FIBiol FRES 11, Grazingfield, Wilford, Nottingham, NG11 7FN, U.K.

href="intro.htm"