Q L 5 6 l F T ■ At* 9 U trun*nf Koom d m 8) University of Ghana http://ugspace.ug.edu.gh iTHE BIOLOGY AW ECOLOGY OF MACRCHISCHOIDES ACULEATUS (HYMEKQPTSRA 8 FORMIC IDAS) IN COCOA FARMS IN (SIANA BY EBHIEZER iU£CRG ARXEETKT A thesis presented to the Department of Zoology of the University of Ghana, Leg on, for the Degree of Master of Science* AUGUST 1571 University of Ghana http://ugspace.ug.edu.gh ii DEPARTMENT CF Z0QL0G1 UNIVERSITY CF GHAKA LHKK , ACCRA GHAM This is to certify that tliis thesis faas not submitted for a degree to any other university# entirely my am work and eU help has been duly acknowledged. been It is m m E zm a k rc k g a h t e e t e t University of Ghana http://ugspace.ug.edu.gh TABLE QP COHTEJfTS ill Section 11 Section 2i Section 3* Section Ui Page General Introduction ••• ••• 1 The Study Area Introduction ... • •• 5 Description of the area ••• 5 Sampling and observation sites ••• 6 Distribution in space Introduction ... • •• 11 Materials and Methods Mapping ••• ••• 13 Radiotracing ... ... 1j| Results Mapping ... 1? Radioisotope experiments ... 23 Discussion ... ... 23 Seasonal population patterns introduction ... ... 33 Methods Sampling times ... ... 3U Sampling sites ... ... 3U Sampling method ... ... 35 Counting ... ... 36 Results ... 36 Discussion ... ... 61 University of Ghana http://ugspace.ug.edu.gh Section 5* Section 6t Section 7* iv Page Diurnal activity pattern, food and foraging habita. Introduction ... • •• 65 Materials and methods ... 66 Results and observations Activity pattern ... 67 Food and foraging habits 1. Observations on wild colonies 79 2. Observations on laboratory colonies 85 3. Food it«nut ... ... 66 4. Association with some 6occoidea, Apfcidoidea, Pantatanddae (Heoiptera) and Iycaenidae (Lepidoptera) 86 Discussion ... ... 92 latarrelatlonahlps of M. aculeatus with other ants. Introduction ... ... 99 Materials and aethod ... 99 Results and observations ... 101 Discussion ... ... 106 Interrelationships of M. aculeatus with other animals (excluding ants). Introduction ... ... 110 Materials and method ... 110 Results ... ... 111 Discussion ... ... 112 University of Ghana http://ugspace.ug.edu.gh VSection 8s Section 9t Nest a and nesting habits Introduction ••• ••• Materials and method Nest building ••• ••• Stimulus Tor nest building • • • Nest architecture ••• Descripticn of a cocoa tree ••• Results Nest building ••• Heat architecture • •• Type of nests ••• ••• Nesting habits ••• ••• Stimulus far nest building • • • Nest abandonment ••• Discussion ••• ••• Flight activity, colony founding and colony expansion. Introduction • •• ••• Materials and methods Mating flight • ... Incipient colonies ... Results and observationc Flight activity ... • Incipient colonies .»• 115 116 116 118 118 119 129 129 130 130 130 11*1 1ii6 11*7 1U7 11*7 11*9 Page University of Ghana http://ugspace.ug.edu.gh vi Behaviour of alate© ••• 1U9 Discussion *«• * * • 1 51 Section 10* Som© aspects of the life history Introduction #••'1 ««* 157 MaterlfiXe wnd aetbods Breeding cages . . . « u 157 Transfer of aaats into cage • • • **58 Obsearvaticm chmbers* . . . 158 Results •*• ... 160 Lervae • • • • • • 162 Pups© « •• 163 AduLts • • • • • • 16h Discussion ••• ... 165 Seetim 11* General Biscnasioa ••• ... 168 Section 12* SuOTS3?y and cc©aLiisiaBS ... ... 1?6 AcknoHledgemeots ... ... 179 Appendix ... .#. 180 References ... 161 Page University of Ghana http://ugspace.ug.edu.gh vll ABSTRACT The ecology and biology of ijacromigcholdes aculeatus have been studied# The ant vas found to be widely distributed in southern (Sham and its distribution overlapped that of the forest zone. In cocoa farms it occurred sore casaonly and more abundantly in areas vdth dense shade than in &reas of +Mn shade. Light, predators, haaoptera and food availability were found to influence its distribution. It was found to be largely crepusacular and nocturnal. M. aculeatus has bean found to be antagonistic to other dominant ants inT cocoa fanae but could occur tqgetLbr with Crematogaster clariventris cn sm e tre e s} this depends on a behavioural mechanism* As a predator it utilizes cocoa capsids and other insects, both living and dead, and feeds too an sugars fraa extraflarel nectaries at* fraa Homoptera. Hites nere found as external parasites and nezaatode vom s as internal parasites* Colonies m re polydaaous and largely aaaogynouu. Population changes are discussed. Alate production was periodic and flight activity had a peak between April and 2&y When queens founded colonies independently. Presence of brood uas found to stimulate nest building in workers. It is suggested that since the ant predates cocoa capsids it could be used as a biological control agent by, at least, encouraging its presence through good cultlvational practice rather than, as at present, killiiig it by the use of broad- spectrum insecticides. University of Ghana http://ugspace.ug.edu.gh 1 . GENERAL IKTRODUCTICK Hie ant genus ^ m»«iil«nhnldas was erected by Viieeler (1920) for *»><■ jyeciee, hitherto placed Id Macraaischa which is now restricted to the Heir World. The type locality of H. aculeatus is nwi/i Coast* that is Ghana, so we are dealing with topatypic material. ants are confined to the tropical forest zone of I jest and Central Africa (Wheeler, 1922). & Ghana though no country-wide survey has been conducted, it can be said from the habitat preference they exhibit that the ants are largely confined to the forest regions. They do not exhibit any nest-plant specificity since they occur cm large numbers of different plants both wild and cultivated. Ihe most prominent ones among the latter are cocoa, citrus and avocado. Nowadays these ants have became so closely associated with cocoa that one goes to cocoa fanas to look far than rather than to the bush. They are tree-dj*©H±Bg (arboreal) ants that make loose felt nests an the underside of or sandwiched between leaves (Uheeler, 1922). The nests are aggregated and it has been claimed that the queen is found onl^ in one of a large number (Anon, 19h7) • The importance of these ants in cocoa farms in Ghana was suggested as early as 1914* biy J. Nicol and his research colleagues at tiie then West African Cocoa Research Institute, Tafo (now Cocoa Research Institute of Ghana, Tafo). Ihe^ daisied that they had H&crcaaiBchoideB aculeatus on individual undamaged cocoa trees in badly attacked areas and suggested that the ants were controlling University of Ghana http://ugspace.ug.edu.gh 2the cocoa cspsids through direct predation. Following this discovery they carried out various experiments to employ the ants as biological control agents against the cocoa caps ids* They succeeded in a few trials as the following quotation Mdicates "....The effectiveness of the protection given by Macraaischoldes aculeatus has recently been illustrated by the severe damage suffer®! by some trees in the First Progeny Trial which once harboured but later lost the ant.” (ilnon, 19I;8). But the overall result ms discouraging. The failure is no surprise, since the ecology of the ants as well as of the capsids and of the general environment wsls then poorly known. Knowledge of these is always very essential in ell biological control measures (Smith, 1963). Subsequent to this attest, several other entomologists have looked at the problem fraa tine to time. Kotable among theia are Williams (1951;) who associated healthier cocoa with the presence of iMacromischoides or Qecophylla. His findings agree quite well with the previous ones of Hicol et al (i gj8J1)T - Direct evidence of predation was not however obtained until 1965-66, when Mar chart and Lest an (1969) did radiotracer experiments and made direct field observations to establish beyond all, reasonable doubt that the ants do prey on cocoa capsids (Marehart and Leston, 1968, 1969). More recently CoHingwood and ZKing (1970) using direct field count methods have also shown that the ants exert considerable predatory pressure on Dlatantiella tiieobroma. Moreover, in view of their University of Ghana http://ugspace.ug.edu.gh negative association with mealybug attending ants, especially, r,™Mn*tt.ap;aater spp*, it has boon suggested that they may holp check the spread of swollen shoot virus disease (Lestcn, personal caammications) • That Macaraaiacholdes aculeatus mv be used in direct control zaeasures against cocoa capsMs and indirectly against mealybugs is therefore strongly indicated# Becent progress made in the study of various aspects of the ecology of the cocoa faro, particularly css cocoa capsids and ants give© Infanaatian #iich could be develops! to give a successful introduction of biological control measures. This is very essential at this tiree sshen the chesjical approach is gradually failing as* The capsids hero developed resistance to scsae insecticides and XtOBtm (19?1 ) has advocated the use of the red tree ant, Qecoahylla l^gtooda in an integrated control scheme* Bat better results would perhaps be achieved if Ifecroaiachoidas aculeatus Here used in conjunction with _0« lopginoda. Although ranch is alreac^ Icnown about the ecology and Moicgy of the latter, virtually nothing is known about the f o e x c e p t for the wark noted above. This ccGaideration has prca&pted the present investigation, in sahich the ecology and biology of M. aculeatus are studied together* Topics investigated include distribution in apace and time, interrelationships with other ants, food and feeding habits, nests ®nd nesting habits, friends and enemies, colony founding and m pam lm , end certain aspects of life history* 3 University of Ghana http://ugspace.ug.edu.gh it Section 2. Hie study area University of Ghana http://ugspace.ug.edu.gh 2. TH£ STUDY AREA Preliminary surveys were carried out on the Akwapim Kange at Aburi, Itapong, and Adawso. Although the ants occurred in these areas their numbers were too small to constitute a good study population# Moreover the great heights of the trees and therefore of the nests made observations very difficult. These places were therefore abandoned In favour of Tafo, where conditions were found to be more favourable. General description of the areai All the observations were made in the plots of the Cocoa Research Institute of Ghana* Tafo. Tafo is in the Eastern Region of Ghana, and lies at latitude 6°17,N and longitude 0°22fW. It is 67 miles from Accra on the Accra-Kumasi road. Taylor 0952) classified it within the Celtis-Triplochiton subdivision of the moist semi-deciduous forests. But on a broader view it is in the Ghana- Guinean bloc of faoist forest at low and medium altitudes1 (Keay, 1959)# And according to Church (1957), Tafo has an equatorial type of climate with two equal maxima of peak of rainfall, though it rains every month. The two Wet seasons are March to July and September to October (Wills, 1962). The Dry season falls within December to February and then August. (Walker, 1962$ Wills, 1962). The severest period within the dry season is during Ilarmattan, which occurs around January. Very cool nights and extremely low relative humidity (dowi to 16# EH.) as well as hazy ekies are characteristic of harmattan. (Gibbs, Pickett L Lestan, 1968). University of Ghana http://ugspace.ug.edu.gh 6SMttnling and observation gitgftt Sampling and observations were carried out in five plots; Cl*, C5a, C5bf Dll*, and H10 (Fig.2,1 )• Plot Cli» This is a single big plot subdivided into two big blocks and planted with hybrid cocoa. The cocoa trees are of medium height end have very good canopy. The top shade is provided mostly by cultivated trees, for example Terminal ia spp. There is very little undergrowth and the trees are sprayed regularly. Currently it is being used for experiments on capsid population. Plot Cgai It is a relatively did plot. The trees are fairly tall in certain areas but those in the observation site are coppiced. The large number of branches per tree of the coppiced cocoa gives than a bushy appearance. Despite the seemingly continuous canopy, there are a few breaks hete and there in the canopy. Terminalia sp provide the top canopy. The undergrowth is very thick and certain parts of the plot are more or less overgrown with weeds. This plot is now being used for Herbicide experiments. Plot Cgb» A motorable road separates this plot from CSa. The cocoa trees are old and fairly tall. Several trees have been coppiced and these look very bushy. The cocoa canopy is very good in certain parts but e^ tranely poor in others. Top shade is provided by tall Terminal,ia trees. Weeding is not done to ground level. Currently the plot is being used for coppicing and caged experiments. flflLJBJl* It is a variety Trial plot with young cocoa of about six years old. The cocoa trees are of medium height and have very good cancfy University of Ghana http://ugspace.ug.edu.gh 7Fio 2.1 SCALE'. I F O O T : I M ILE FiE*2.1* A recent map of the Cocoa Research Institute, Ghana, Tafo, showing the study area (cross hatched). University of Ghana http://ugspace.ug.edu.gh 8n»ntr«> the other plots* top shade la provided by big, tall forest trees* There Is very little undergrowth except In areas receiving light due to breaks in the canopy. The observation site, near to a broken canopy area, is weedy. Plot HIQi This plot is the smallest among the five. It is divided Into four blocks (HLocke 1, 2, 3 and it). The cocoa trees are fairly old with good canopy, though breaks occur in certain places. Top shade is provided, In the main, by forest trees, though there are several cultivated trees i.e. SerminallB spp., SoXamai verbaadfoilnm and Hanlhot sp. tfadergrowfch occurs only In canopy-tardm areas. This plot is currently being used for systemic insecticide trials. Work started in raid-October (i.e. the rainy season) through December to February (dry season and haraattan) to the fallowing June (vet season). Thus two wet seasons sandwiching a dry were experienced In the course of t-M* work. Hie main climatic features of the area are smaarized in Tig.2.2. University of Ghana http://ugspace.ug.edu.gh Cp✓ Fig 2.2 IOO-i 80 o 40- 0900h c !300h l L J L M A M J J 1970 O N D J M A 1971 M Fd g *2 .2 1 feather data for Tafo during th© period of stucfcr. At r a i n f a l l , (histogram) and sunshine ( l i n e ) j B« tenroeraturei Cl relative humidity. University of Ghana http://ugspace.ug.edu.gh 10 3. Distribution in space. University of Ghana http://ugspace.ug.edu.gh 3# DISTRIBUTION Hi SPACE Introduction Macr^.3choides aouleatun le ccaafined to the tropical rarest ecsi© of Vest and Central Africa (Wheeler, 1?22). la Ghana, surveys so far ««»»riiircfca, CfJb and Dili, whilst the •acceptance-rej ectionf work was dene in plots C£b, &t ard H10# The area to be mapped was divided into 3 ^ 3 metre quadrats. The positions of all cocoa and shade trees occurring in each quadrat were accurately indicated cm the map and subsequently labelled. Boundaries of different shade regimes were marked to indicate areas with both top shade and good cocoa canopy, good cocoa canopy, top shade only, and poor canopy or open areas. Bach tree was then searched for ants of all species and a record made of then. Those Hhich could not be identified cn the spot were collected for later identification. All Macrcm&schoixios nests found on any one tree were labelled. Thus nests found on say tree h bore the labels 2*a, l4>, he etc. Each nest ims then removed from the tree with secateurs and enclosed in a small polythene bag, which was in turn put into a bigger polythene bag containing chloroform. They were later opened and a record made of the contents, especially of dealate females. Trees containing other dcaninant ants i.e. Qecophylla lenginoda arxi Cr^ afcogaater stadelmaml were also noted. Cdcny size was crudely determined by lining the 1 acceptance— rejection* method. This involved transferring workers fraa one nest 13 University of Ghana http://ugspace.ug.edu.gh uto anotner. Fighting easued if the wcaskers cam© from different colonies! if howwer, they belonged to the same oolaoy, there was no fighting. The ants were transferred hy means of a stick which had a band of a grease barrier (Ostico) about two-thirds of the vm& dam its length. Thus the ©:+jrina-n University of Ghana http://ugspace.ug.edu.gh Y ' l 1 O NEST I F2*>3.2* positions of neats cm cach cocoa ti £LaM£a, Tafo, 12th April 1711. University of Ghana http://ugspace.ug.edu.gh 16 SET-UP FOR RADIOISOTOPE EXPERIMENT — - ~ ,• Fig 3<3 University of Ghana http://ugspace.ug.edu.gh throu^i the slit. San? of the sugar solution was measured into the container. The branch was tied to the post with copper wire thus rendering the arrangement atom-proof. A broad band of grease was round the post. 0.0?ca3 of the radioactive phosphate with an activity of £0 microcurles per millilitre were then gently ntMwd to the sugar solution, stirred with a. glass rod and the cap of the container was screwed into position. A strip of filter paper was pushed through the slit into the bottom of the container to act as a wide. Lastly the leaf bearing the nest was tied to the poet with the almdnluB wire. 3h this way the nest was placed in contact with the container. The ants thus disturbed moved agitatedly, but soon discovered the sugar solution and started feeding an it. Saae entered the container while others remained on the filter paper. Each nest was monitored, using a Geiger counter, at 2h h intervals. Monitoring continued until constant counts were obtained in two or three consecutive days. Other Insects, particularly ante, found in the area were monitored dally. Results i. Bagging The results of mapping are shown in figures 3.1* - 3.6. They indicate that the ants show a preference for shaded areas. Tables 3.1 and 3.2 give the figures for the various surveys. A chi-squared teat of the presence or absence of the ants in the different regimes showed that the results were highly si&ilflcant. 17 University of Ghana http://ugspace.ug.edu.gh 13 University of Ghana http://ugspace.ug.edu.gh VT University of Ghana http://ugspace.ug.edu.gh 20 m Fig*3«6* Distribution of M» aculeatus and 0» lopginoda in relation to different shads i* plot in 01^ , Tafo, wot. ly?0 University of Ghana http://ugspace.ug.edu.gh 21 Distribution Mreraalaohcddes in different ahiide raises in Plots H10 and Cl*. Table 3*1 Plot Top shade mt i good cocoa cenapj Good cocoa canopy Poor cocoa cancpy or no shade | Totals H10BL Total Ho. of trees 20 10 8 36 Trees with Macro h 1 1 6 Total Ho. of nests 6 7 2 15 hiob: Total Ho# of trees 33 2 10 US Trees t*±th Macro 9 1 - 10 Total Ho* of oasts 2 h 1 25 ch Total Ho# of trees 12 5 3*2 seedling: 22 Trees id.th Macro 6 - - 6 Total Mo* of nesie U5 I mm 1*3" T 1 1 1 IB 1 University of Ghana http://ugspace.ug.edu.gh Occurrence of r&crcmlscfaoldes under different shade regimes 22 Table 3 .2 Hacranischoldes No. of tree under top ehwte plus good cocoa canopy No. of tree under- good cocoa canopy Iio. of tree under pocr canopy or no shade Total Present 2 1 22 Absent 16 15 22 83 Total 65 17 23 105 % of trees with Macro 29.2 11.7 1.5 20.9 ? of trees with H&cro as ccrapared to total trees 18.0 1.9 o.? 2 \ *» i£.02 Degress of freedcn » 1 P <. 0.1 University of Ghana http://ugspace.ug.edu.gh ii. Radioisotope experiments! The results of the radiotracer eiipertaent are shewn in Figs. 3.7 - 3.9 and Tables 3.3 - 3.5. They show colony boundaries and iixJieate that there are "no-man's~lands" between colonies. They provide further information an colony sizes. Table 3.5 shows the rate pf spread of the radioactive sugar. This trill be discussed under 'Daily activity patterns# foraging and food' in Section 5. Discussion Hie results of th© mapping prove beyond all reasonable doubt that the ants prefer areas with dense shade, i.e. those with both top shade and good cocoa canopy. 06.3% of the total number of trees with HacraaischoideB occurred in this area. Ihe next preferred areas are those having good cocoa canopy (9.1 %}, The least preferred areas are those with either poor cocoa canopy or no shade at all (5.6£). Incidentally this least preferred area is where Qecophrlla longinoda abounds. Results of several laboratory experiments seem to substantiate the field observations. In the laboratory the ants always retreated into dark comers of stools and tables to nest. Only on four occasions did they nonage to construct half-finished nests on cocoa seedlings in the open in the laboratory. However, t&en placed In a dark roco they constructed a nest on top of a octal bar. Thus light intensity is a major factor that influences choice of nesting site, .toother factor worth considering is the greater leaf area of good canopy cocoa which offers better opportunity for 23 University of Ghana http://ugspace.ug.edu.gh Metres o 1 .2 3 4 5 10Metres 2U O O o o o o o GOoQ_ Q. O o a «5> o o o o o o o o o o o o n/m o 1 o \ I o _ o- 9 . £ 2 .o o c OO , 2 <~0 , 33 <•*= GO *~*z> zr> 3o -v-> _c— a CL Ozn a)JZr> O O D OO O O aQ O O O O O O O O O O o o o o o o O O O o o o □ O O O O O O O O O in \S) ■ cvl o D n 3 ^ S' §1 S plot C5*> Tafo, University of Ghana http://ugspace.ug.edu.gh Table 3.5 Saaaary of radioactive test results 27 Plot C5a C5b Dllj BIO Total number of nests 22h 77 56 m Number of nest radioactive 7h 38 52* 1li0 Bate at spread in days 1 3 3 5 University of Ghana http://ugspace.ug.edu.gh Hate of spread of radioactive sugar# Trial 2 20 Table 3 . it Date No. of nest S. active in C5a 17.iv.71 7h 18.iv.71 7h 19.1v.71 7h 22.iv.71 7h 25.iv.71 7h University of Ghana http://ugspace.ug.edu.gh Rate of spread of the radioactive sugar. Trial 1 Table 3.3 Date Ho. of oests R. active in C0b Ho. of nests R. active in H10 Ho. of nests R. active In Oil* 19.111.71 37 92 mm 20.111.7* 36 107 1*7 21.111.71 36 127 53 22.1ii.7l 38 135 53 23.111.71 38 11*0 51* University of Ghana http://ugspace.ug.edu.gh ■Hghtiwg f(n> segting sites to nfitfid females (Billosj It la «i«« likely ♦>»»«•- shaded cocoa trees experience lass violent stores tiMB mmhwdad ones. Thus ant neats in this area may be saved Aron stonn-daosge, a factor thathaa been recently found to cause i«mw deaths to both Qecophylla and Macranischoldes (Leatcn, personal corammicaticn). Stona-dsmage will be particularly high in Hacromlsohoidee which constructs nests sandwiched between leaves, since the atom can easily force the leaves apart. Ctace the ants fall to tha ground, they become disorientated and their behaviour profoundly upeet. They therefore perish by either predation or deprivation. The maps cculd not define accurately colors size and colony boundaries. Tha radiotracer experiment achieved this In a very clear cut way. Die limit of the radioactivity is the limit c£ a colony, since only ants from the same colony have an access to the radioactive sugar. Colony boundaries could thus be dram with ease and the number of radioactive nests found within the demarcated area constituted the size of the colony. Colony sizes were found to be very variable indeed. By the ' acceptance-re J action' method three colonies ware found to contain 2, 36 and 1|7 nests respectively while four colonies containing 38, 53, 7h end 11*0 nests respectively were Identified with the radiotracer. The number of nests may a*?**** cn the age of a colony, since a colony is founded by a single queen. Obviously mare mature colonies will have more nests than Incipient ones. Other factors however, such as shade regimes, food availability 30 University of Ghana http://ugspace.ug.edu.gh and predators way greatly influence tho siz e of the colony* Accanling to Cooler (1922) several hundreds of nests may be found on a single plant in the Gongo forest?. If his observation Is correct, then colonies raay contain several thousands of nests® The rnasber of nests per tree in our cocoa ferns is, however, very small* the highest number observed so far being twenty-three . Although actual colon:/ size may be sn a il the ante cover a wide area due to the mall number of nests per tree. Fraa a practical viewpoint this is an advantage, in that the ants am by this means protect many cocoa trees free capsid attack, and moreover vill not hacper fam work, such as harvesting of cocoa. Bius this warie has shown th&t the ants nest nainly in uell- shaded areas, that they are polydaaous, end colonies are separated lor free zones* Otter findings such as number of queens per colony, rate of spread of and relationship with other ants will be discussed subsequently. 31 University of Ghana http://ugspace.ug.edu.gh 32 Section U. Seasonal population patterns. University of Ghana http://ugspace.ug.edu.gh 1*. gsASCKAI. POPULATION PATTEBHS Introduction Periodicity of brood and adult workers and of sexual caste production of ants In temperate regions is a cccsaon phenomenon has been reported in many cases* for example Leptothorax sp. by Headley (I9li3) and Mmaica sp. by Brian (1957). Brian (1957) ovplning that periodicity is Inevitable if the life cycle of any ant is long compared with an annual climatic cycle. In the tropics, seasonal changes in ant populations have also been reported. Ledoux (I9k9, quoted by Sudd, 1967) reports that though Oecophylla lenginoda produces aeocnals throughout the year In Infest Africa, they are more ccranon Just at the end of the rains in Koveraber. Gibbs and Leston 0970) walking at Tafo, Ghana, however found the peak of 0. longinoda alate female liberation between March and May. They also demonstrated periodicity in alate faaale production in Odontcoachus haaaatodus. Periodicity has not however been explicitly demonstrated in M. aculeatus. It can, nevertheless, be inferred from Lang's (quoted by leiheeler, 1922) statement that nests of H. aculeatus in the Congo, "sometimes" contained brood and winged indiiiduals but •often" only a few workers or were enpty. Billes et al t^ ^ ^ working at Tafo noted the period of emergence of foundress queens between April and Hay. Apart from these c-oww] observations no 33 University of Ghana http://ugspace.ug.edu.gh effort hae been node to deejonstrat © tho changes In th© population in relation with th© seasons. Ih© present stuly Alma at investigating th© changes In th© population of K. aculeatus during the period of saapling frco October 1970 to June 1971 • This Includes period of ovipositian, changes In adult worker and worker brood peculations, sexual production and population changes, to attfjnpt will also be made to relate the population changes to the seasons as classified by Gibbs and Leaton (1970). Methods Sawrxling times i Initially, £rom 1l*th October to 23rd Kovaaber 1970, sampling was done at weekly intervals. This was, however, charged to fartmghtly for th© rest of the sampling period. Saapllag sites a Preliminary survey 3 were conducted to select ffcdrly large nest aggregates since before the radioisotope eaqjeriioent it was not easy to delimit colony boundaries. Thus it was not certain whether two or three nest-aggregates found in the saae area belonged to one colony or different colonies. Sa&jpling was therefore done in different nest Aggregates partly due to this initial uncertainty about colony boundaries and also th© smallness of the nest—aggregates. Two or three saraplings in one area resulted in complete collapse of th© nest aggregate. It was then necessary to move on to the next nest-aggregate and so on. Sailing was made in PH10, Blocks 1-1* and frcci two nest-aggregates in Plot C5a. 3h University of Ghana http://ugspace.ug.edu.gh Sampling; methods & collecting the nests* a polythene bag, measuring 30cm. x 38cm wee pushed half-way up a nest and held there with the mouth open. Then, with the pair of secateurs* the petiole was cut so that the leaf bearing the nest fell into the bag. The mouth of the bag was quickly held together and tied with the rubber band. % this means few or none of the agile workers was lost. When however the nest was at such a place that the bag could not be held under it, the leaf petiole was simply cut with the secateurs and dropped into the alresdy-open bag. A few workers were sometimes lost by this method. Prior to cutting the nest a label written in pencil cn a piece of paper was dropped Into the bag. 3Ms bare the tree end nest numbers. Any first tree sarpled cm any particular date was labelled 'A'. Similarly the first nest m tree A was designated A1. Thus three nests on tree A were labelled Al* A2 and A3. Tree A of a particular date might or night not be tree A the following or subsequent sampling periods. The nests collected were packed into a bigger polythene bag* tied with a rubber band and conveyed to the laboratory. 3h the laboratory the ants were killed with ethyl acetate* chloroform, Campy1 s fluid or occasionally carbon dioxide. F-woh bag was then opened* nest taken out and contents emptied on a large sheet of white paper. The nest was then examined under a dissecting microscope for any food remains, inquilines* or predators. The different castes and developmental stages were sorted into different petri-dishes with a forceps. A small quantity of 70% alcohol in 3S University of Ghana http://ugspace.ug.edu.gh 36 the petrl dish facilitated larval and egg sorting. Dieso together with the nests were oxacdnod under the microscope together with the debris left after separation of the ants# Any parasites, inquilines, or food remains collected wore put into 70% alcohol and labelled. Countingt Since the ants were sciall and numbers obtained very large, at times running into thousands, counting was done in calibrated tubes# One hundred workers, or larvae or pupae were counted into Hat-bottoaed perspex tubes and marked. The rest of the tube was calibrated upon this first mark. An egg-aeasuriqg tube was also made, but it was designed to count in thousands# Queens, alate moles and females, reproductive larvae and pupae were actually counted. Vtarker pupae and larvae were counted if their nraber was email. Record was then aade of the various numbers obtained# Results The data for the population counts are given in Table li.1 # Figure &.1 shows the graph of the mean number of workers and worker brood per nest per week or fortnight# It also contains inf amotion on aviposition and times when queens were obtained in the different sampling sites# This figure Indicates that worker brood was present almost throughout the sampling period, and, with but a few fluctuations, xaeanJSorker population remained more or less steady throughout the period# There was however a marked drop in October, University of Ghana http://ugspace.ug.edu.gh Ta bl e lj .1 Ta fo , MO b^ i 1U .X .7 0 University of Ghana http://ugspace.ug.edu.gh Ta bl e 4* 1 (c on ti nu ed ) 38 ! t * IOS £ 8 2 ........... 3 i . ^ * H CVS rH r-3 ^ i nlA d s i &0 8*3 t t I l i i I i I • i « i 8 i ( a i • I « n 3 -a i i • i a i i 8 i i i i i i i i i i i i « 3 f | ^ ^ | | | | | • | I • • I • • I I I -4* cl o « r <4« ! I I | I • I I S I I I I I I I I I • I • ! s § § § l l § g § § § ?5§ § & § K' § 3 | ; * 2 • • o % S5 H W ^ 4 I A H W h W H W K > _ CM *«» C\3 W\< C < < < < I E 5 P 5 0 0 K < 3 A * r i s ri CM CQ ONH H i ~* H I H r 4 H I H H H University of Ghana http://ugspace.ug.edu.gh Ta bl e 4* 1 (c on ti nu ed ) Pa fo Hi Ob . 26 * £. 70 39 «o * r5€* 83 8* & 53 4* © 5 *< ae to a & 03 w- fc i D© S'S J S< &< «O85 £ £ M 4 I • I O • d I I I • I I I K > £Csi » i i i i • « i i i « i • • t I ft ft I I I I t t • 8 • C \VO g 8 w 8 3 § ^ 1 ?) 3 S ?, 1 er** r-» cH 8 • • I » « I • I I I | « I $ I § 8 8 3 3 S 8 84 H CO a } ^ H H -4 “ g g o o o liS H r i fH CM i*\s « ja ^ o « fca K> 4 I f t VO N C3 C N O rH CMH H H H University of Ghana http://ugspace.ug.edu.gh im r- r- r— • i t n 1 • 8 8 1 1 • 1 1 1 9 8 8 8 ® ^ r- C- ^ 0 © 1 § 1 3 l 1 1 1 1 1 1 I I 1 1 1 1 1 1 1 1 • 1 I 8 1 1 8 R 8 8 8 4 S g ^ S § 3 C3 CVi o SS3 K M i • • i • • • 1 2 | S 8 I I I 9 I * I CO C M O O Q O O C ) . CO O O »A Q OM fH u\ «h *> »n I I I I I I I I I I i i « • i s i • o e m K § 9 * , S r § 8 ? R I I I • 0 • I • I • 9 I • I I • I I I • I I I • • I I I I • I § i a s § § g § R ! i $ *H W 4 SO N CO C\ o H I X P> wf > University of Ghana http://ugspace.ug.edu.gh fc ? *I Cvi 3 13 8 * ^ /T H 2 -I g o • l «€-* e-4 «xJ g £' O (& 03 Wk 0)* isO I I 2£ I « I I I • • • • » I • a © Q ^ |JS£> *a» C3© o tS*3 3 1 a0 Gj r? E raCD 5*3 * so «* r !■!-« in o m in*> co q m CM H PS ^ VD s i s ! ! ? a a s » 3 r f e » a s f s ? 5 j » University of Ghana http://ugspace.ug.edu.gh xe fo , Hl Gt^ & b, 22 .X li .7 0 n* Sw w e s*P5 I& © • a © o •as s © o $ $ S 23 © ©Q Cm © O £ £ CO •© o25 25 «w O •u • ?o ©r5 l"5 hh 8 4 I | I I H I lA CO VO ^ \o tA ir\ | I g « I CM ON IfN I • N I i<\ N W » n » rH 4 N N * 1 & 8 § 8 8 8 r2 h h w W ' h »<> rH ^ 2 O H W CO H • O 8 I I I I I • • I I • I I I I *"N *"N 8 I N I i f \ CM CM 8 § S ^ 8 8 8 ^ ° ON CM lf \ - t iT\ Q KN r-4 3 § a P I S ^ 5 i 3 S S S ? e ? { ? l S & ® 5 § 8X fH c y j ^ 4 i o s o s (00N O HH t-4 CM a * University of Ghana http://ugspace.ug.edu.gh Fi 3* ” i 1 i f 8 * - &H O i £ © * ? « nJ e 8 &cm s •» E © * o o © •a s CQ © Hi &■ ' ( • a4-> © «J »H H « ^3 S 3i/i fti H ^ © © —i w . W © © H 4* «a g l i 16 | « I I I • I I • • • CMi s I t I I I I eg t N i o 8 • § fl B ° ^ H 0 * - • £ K \ GO CO • s o i l ll> I I I t • I I I I I O Q O O Oir\ cm in coCNi | S 8 § S § | ^ +»03 •a> o O °3 ■t* • «r> © H H W H W 4 tA < « « PQ O fi P « 3 cm ^ 4 u> vo co o\ IX University of Ghana http://ugspace.ug.edu.gh OB 5*0 W n j 8 S £ t a 03 ^ « I O 02 © H 63 J5S « M '■“> •| 9 a « § « - ^H O 1 d*>sr **? J3 «*» H © •a E O o © n•p © cu H n J © © ©•P r-»cu « ft £- © © r-4•P aj 5JB 4*»K •© o2S S O w • ?© O£5 25 . • • g » h6 ( • « • • • ‘ • • • 8 50* » 5 I CJ O r-l Q I i • I I I t I I H « S ® £ • CO •o 4^ - O o o CVI » n « 1 I I H «tf\ Olf\ I 6 ON c \ 1 • 1 l^A LTV CM e s a > cl 5 S 4 S 6 I 9 0 CM | | | | | • 0 8 I fl C\J r-t I 8 fl 8 C\j J CO CM CJ S 0> X> t Oeg 0 0 0 0 0 0 0 0 0 0 0 0 0 0 v.,nl i i s s R a 3 a u 3 n 3 5 a s f e g d s g s d s f & r t s s o tA 00 r“l w ^ 4 i f \ V O N c o O v o fH CsJ University of Ghana http://ugspace.ug.edu.gh mIs-* R 8 * HI O 4 e iH O (d cy fc« *h ©Mr-,iW © © *3 53Q "3 JJ* ■ £0> ■fio @ 63-P 5) I'iS i QrH a© Q)(A £ © ** d i i 5 O O*a• *- O© ts3« • I I I • • i 0 i Q § t e i i 9 i 8 i 8 & * & ^ « g 8 8 af*’! • r | I I I CM H K \ ^ i * i i i • i i i i ^ r- • i • i • i i i i * <2 !3- 7 •C\i 8 C| 8 I i*\tS\V3^C£j 8 • I 9 8 • I I • I CD r*l \S\ 8 O Q O O O t A O O l I 8 I 8 I I I * I £> © Q O O r-f O if> . ^ •^ p-8 OJ r-» I ^ rQ 1 1 1 6 9 8 1 6 8 6 8 I CM 8 8 I I r 4 | I 8 8 1 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 1 8 8 » I I < i 8 I I 8 6 8 I 8 tA I I I I o H VS ^ O § l S i l S g | i 3 « 3 8 S s | § S 8 S 8 § *H CO -4- r-« CMrr%^*u^v0 r^co ^ 3 H H H tH r) H H H H H OJ I « University of Ghana http://ugspace.ug.edu.gh xa OL W £|. #x \c on 'd : p i w I* A * § ♦ %UQ Cu p C5K JZJ 8 3 U3 CM H *ri a? »a 8 I • i • I « rH O CM 8 rH CO o -4 O • I I • I 9 Ia> h o ia I cm 4 I t i l l • r^ • • • • i I i i I i • § § § I § ° § ® S 8 Ift 00 on VOIs ^ H 4 h W h<; la H W 4 v O r ^ c Q O N O H H •n vo ® 3 ........................................... 1 G j ^ R ^ f=t S ' 1 • • * ‘ ‘ vj J I I I I I • I £• 00to • tn 3 C\o *J CM * diin *n *n ON I K University of Ghana http://ugspace.ug.edu.gh It? Ft ■g w I • t* * - sJ E? i 8 ^ £ g * { O 8 •a23 o Q *3 $ £ 2& « © oa4? 0 1*3<* 23 • © rH S a s 5 ^ -51*•*« Cm & *O u ° 4® o# § 'r\ r 4 < 2 Q » ' * O ^ C 0 C ) » «A ON u ) OS C\J IT\ »A CV VD CO VO H W ^ 4 l f \ V 0 I N ( O C \ CM CMc* •^inwinONON^ift'X|N •V O n i ^ O N C O H ^ 4 H Q lO » s | § 3 8 $ R 3 » S S S * 3^ fO K\ ?» . * 5 KN -T • I • I I I I I CM C\i •00 IK University of Ghana http://ugspace.ug.edu.gh Ta bl e 4. 1 ( I S3 s I * \ © i I I i i i i 3 2 4*m • © o £5 S5 %.-s ®,S 50 « 3 £ $ $ £ s ! S ! 3 « ; * j C>3 CQ ^ r *jr*t r-l COtrfcf,»“ {3DV©O*f*CHir*>uF\G\C0 •B S C M ^ r t W ^ H ^ i A H g PJ 3 « 3 S t S 8 S a t » S 8H H H «~4 ON 0% I I I I • • • I • I 8 3 l o n r » 2 ® § ! 3 § » I S S § I § I « H W * A 4 U \ V O N ( S ) < y \ 2 .H g ?? University of Ghana http://ugspace.ug.edu.gh Ta bl e 4. 1 (c on ti nu ed ) si fi B £ CMJO 38 • I I I I I t I • • I CO © as 3 3 so o S s 1 3 CO• fc$F! wV V , Vr- s* ir\ 1 1 1 1 • 1 1 1 1 • co cvjtrN-4-rw0 kn vo cm r*- -4 • H H r=S r-l 3 CO 8 § S 8 S 8 q 8 8 S S UNi A K > U \ K > 4 u > 4 4 ^ t ' l O N ONMNK% *=> IK University of Ghana http://ugspace.ug.edu.gh Ta bl # Um l (c on ti nu ed ) Ta fb f C5 &, 13 * IV * 71 s ®aoV5 © •333 <9 © , 1 $2 • • i i I i i I -f c?CM ^ H O ' VO CO I « N H W M)t£J K\ | C M I H I M 8 0 4 ^H Cv ^ | s I 5 5 3 S 3*3 5 8l«4 © i i : 9 « O ON0 ro*> 03 VO J .CM CM § 3 « » 3 2 3 g s s g g £ • • • I I I I I | | H CM 35 § § § § ! § § § ! & 3 H *> «■}> r-4 C* 4 U > V O f ^ C O G\ .* *? 14 University of Ghana http://ugspace.ug.edu.gh 53 I Fi I fc* 8 \£W 04 3 | I fM ^ J10 $ 8 a **3 cJ * s *OPi «QQ ■a 3S © 173 •o pr*. fcs o O fl) £5 525 • I I • VO I I I I C * a i ^ i i n - ^ CO • r-4 rj O co o r l ^ 'O 4 h cm h w H W K \ 4 i A ' O N (Q(n University of Ghana http://ugspace.ug.edu.gh Ta bl e 4. 1 ( c o n t i n u e d ) Fi >• 8 «* •a © f i ofSi t f I •tt 2a 5 s Iotat ■a^3 6) © *311a. § I & • a49 © 3 3 « S | 5 3cj c3 3 -pCQ •© o u ° 4* O ©*5? 5R a • I I 9 I 8 I I I 8 OJ9 9 9 CM 9 9 9 I 9 9 CJ • O **> ro 4 C 0 H S N N N C D g ^ N H * 9 9 9 9 9 1 9 9 0 9 CM O *H l l \ (M SO CO ^ l A | A ^ r-i H U> ^ 8 S 8 ^ § 8 8 3 15 ri H -4 u\ cm cm iH rH 4 4 H N l A ^ ^ V O N c o O S IN University of Ghana http://ugspace.ug.edu.gh H J A 5» •a I&• i £ - 2Io nty *3 a a£?• Is* Ua|o o a•£> 3-3 3,3rig w ■S £ft 5*4 © © 1*3 S?S 6) 15O 4*03 •0) O J2J ZS 04 °!o ©'*-£* 525 55 , i i i i i i • • • i i i i « ■ a i • i i i i i i i • < t i i i i i i i < i i i i i i i i t i i i CQ | I I I I <» I I I • © • , , « o r t | - H CM NO - J j £ I I I I I I I I • • , , . S CM . rt in 5J £ £ a S § 8 § § R I I ' 8 | < < pn « o o o o a I X ~ c r-i N K> 4 tf\ VO S © ON O 51 4. 0 7. 7 37 .4 University of Ghana http://ugspace.ug.edu.gh 56 * t W r4 £ £ ^-I o i s« f-» I s <0 I s* 5 O « *i* •■s 5* O fa « «) -»-* © ,3*3*s a ctf 3^iH e $ r® 0 o «7l ^J8 s . ££ «u • W 2£ i i • • • • • • • i i t i • i • i i * i i i a « i i i i 6 i • i « i • i i i i i i i t § « i « i i Nf> S q% tn 9 « Q o • * *O ^ S% H H LA ■r. 1 I * I I l 9 9 t*> H C>1 iA CM r r \ | f * . CM » s s a a * * * i srH K> St a £3 & g £ 13 a #H* cm ^ 4 tf\ VO i^* qj . * 8 3 University of Ghana http://ugspace.ug.edu.gh S I 62 O ^ •s £ .3 o 4 » 8 M* ^ M r-f 53 J ^ 1 1 03 0) 0) § §• 8a fc* c *> 55 O *»1 f*4 rH R3 Cl13 ^ fa> o w a» 5 ^«3 R i rH /P | c» ->:**1 c -VsKi •© ofe sa 8 W rM C' iA »S 2! 4> Si vc> 3A $ cv M3 803 u CM o o r* CO K \ o cl K \ a IM Ci1.'? f3 University of Ghana http://ugspace.ug.edu.gh Fig 4.1 58 University of Ghana http://ugspace.ug.edu.gh fallowed by a steady build-up to a peak in December, then a slight drop In Jaeraaiy after which the population was maintained at a nearly steady level, until it started fluctuating widely in Bay with a sharp drop in Jane. Pupae alternate, more or leas, with adults. Pupal production started at a high peak between October and December but declined gradually till it readied nearly zero in Hay* With a few discrepancies, larvae alternated with pupae* Fig. Iw1 shows also that large miabers of eggs ware obtained in October, very few in November and fairly large mrnbers in January and Hoy. In all nine queens were collected. Six of these were obtained in plot RlObU, one in BlObl and b2, and two in C5a. With the exception of Hi Obi and h3, eggs were obtained each tine a quean was found. Fig. li.2 shows the seasonal changes in sexual caste population. It demonstrates a strong periodicity. Sexual caste production started in December, readied a maximum between March and May, and declined nearly to zero in June. Build-up of both adult immature sexual individuals was usually followed by a sharp decline in their nwbers. Initially slate females were mare abundant than males. Female population growth was gradual and drawn-out as compared with the more or less exponential build-up of males. Thus very soon the outnumbered the females by varying ratios, but at the peak it is about Utl. Male and female pupae alternated with male and female adults respectively . larvae also alternated with pupae. 59 University of Ghana http://ugspace.ug.edu.gh Fig 4- 2 University of Ghana http://ugspace.ug.edu.gh Ttm greatest production of sexuals was in the first Wet sunny season? but thoir developnent started earlier in the first Dry sunry season* DiflCUgBBlCP The reei&ts demonstrate a periodicity in sexual caste production beyond all reasonable doubt. The present finding that sexuals have a peak between March and May agrees with Bille’s (19h6) earlier observation. This period falls within the first wet, sxmny season of Gibbs and Leston's classification. The large flush that occurs early in this season nay produce large leaves i&ieh mey be retained far eight or nine months. Since according to NicdL (19hh) the sis© end the retention period of leaves are zaajor factors that determine colonization of M. acisleatus this seascn would be an ideal one. Moreover the large range of insects (Gibbs and Leston, 1?70) available during this season xdUL enmre an abundant food supply for the deve&qpSng and developed sequels as well as their attendant workers. Further, considering the fact that the ants are largely markedly photophobia, the first %mt9 dull season uhich immediately fcUotTE sexual production, development and liberation, w m afford them more hours of activity, since cloud cover is mayimnl darting this se&scn* The "wide fluctuations In adult slate populations may denote flight periods. Worker caste population was maintained at a relatively high level throughout the sampling period. This is understandable, for tho xjorkor must be there to prepare for the advent of the sexual a 61 University of Ghana http://ugspace.ug.edu.gh and idien they arrive the workers are still needed to nurse and feed them* Any large decline in their numbers will therefore affect the entire colony adversely. It is interesting that unlike the adult worker population, worker brood numbers dropped nitfa the production of sexuals, but wore raaintained at low levels nearly throughout the period. The fact that eggs were not obtained throughout the ftmqiUiJiQ period does not necessarily mean that oviposition is periodic. Sine© Macror&scholdes is largely moetogynoua removal of the queen from a colony early during the sampling period would mean that no nore eggs wuld be obtained in subsequent collections. If it had been possible to identify the quemfa nest aM leave it uncut or had the sampling method not been destructive a different result raigjit have been obtained. It is interesting that the eggs laid early in the sampling period developed oaLy into workers, except at the beginning of the first dry sunny season in December ttim serials were produced. Schneirla (1957) found a sinilasr situation In the genus Bciton. a Hew World Darylinae. Gibbs and Leston (1970) have suggested that the high tc$if>eratures in the dry sunny season night •accelerate* the development of sexual larvae of Qecogh^J^ lonninoda. Hiis may, pexfcaps, be the explanation for the production of sexual brood of Ilacrc^ tLscholdes in the dry season. In view of the destructive nature of sampling and the fact that I moved from one colony to another as necessary, no valid 62 University of Ghana http://ugspace.ug.edu.gh 63 or meaningful inferences can bo node on brood population fluctuations free the graphs. So far it has not bean possible to determine the duration of all the different larval instars In the laboratory. Uar, unfortunately, can mich be learnt fi’cc Figs.it.1 and h»2 about this subject. University of Ghana http://ugspace.ug.edu.gh 6li Section 5. Diurnal activity pattern, food and foraging habits. University of Ghana http://ugspace.ug.edu.gh 5* DItRHAL ACTIVITY PATT5KE, FOOD AID FCRAGIHG HABITS ,;orka on ant diurnal activity, food andforaglng habits are now very numerous. The few to be mentioned include the works of f i i 0953), Ayre (1958) and Leston (1960) cn dally activities of different ant species, notable among the workers an ent foraging or hunting are Brian (1955), VewflL&s (1955) and (1961i). Lange ((i960 cited by Sudd, 1967), Chauvin et al (I96I) and i*nn«» (196S) have also worked an food traffic among ants. Marks cn scent trails and trail laying by Carthy (1951 )» kllson (1959) and Sudd (1967) are also well-known. There are, however, no data cn this matter specifically referring to Hacro- mischoldea aculeatus. The present study attempts to provide infonnaticn an this subject. Ants are well known to have fluctuating peaks of activity. Scoe species nay have one peak of activity in 2h h, which nay be either in the day or during the nightj others nay have two or several peaks a day. Several attempts have been made to determine the environmental factors that directly or Indirectly influence this activity pattern. Seme authors have gone even further to suggest that there is an Innate rhythm of activity. Qoudsley-Ihcnpscn (1953) and Pittendrigh and Bruce (1957) claim that the activity rhythms in PeriglffPgta and Oroaonhila respectively are tanperature dependent. 65 University of Ghana http://ugspace.ug.edu.gh 66 Jh Fornix Ayre (1958) found that light and humidity were the significant factors but he could not differentiate their effect from temperature and atmospheric pressor® sine© they were all interrelated under field conditions, toother worker, such as Sehnelrla (19hh) has linked activity patterns with the presence or Absence of tarood-eggs, larvae and pupae. Bse presence of brood, he elates, always causes high activity in a colony, since the brood acts as a drain on food reserves of the colony, aheeler (1?10) Just called the responsible stimulus •hunger*, and claimed that it was one of the nain stirauli to Increased activity. Ih this study the possible meteorological factors, such as air temperature, relative hoaldlty, and light intensity as well as other environmental factors, i.e. presence or absence of predators that may influence daily activity pattern have been eranined. Materials and methods Ten nests were selected randomly Trm a colony and labelled A, B, C .....J. Census points were established about one foot a*®y from the nest on the branch bearing the nest. £21 connections with the nest other than the selected census point were severed. These preparations ware made a day previous to the day of counting. Hourly counts were made of all ants passing the census point during a period of five ndnutes using a tally counter. Atnoepherlc temperature and relative humidity wore recorded with a'Moisture content meter* and light intensity with •'Weston Master V” exposure meter. University of Ghana http://ugspace.ug.edu.gh Continuous 2h h counts ware node. Three replicates wore made on different dates and In different plots or blocks; one during tho dry season and two In the wet season. To investigate the Influence of other dominant ants on the activity pattern, one 2li-h count toes made in plot Dili# #iere ?^ icra rischoidcs aeuleatus end Crcs^ogaster clarlventria occurred together on tho sane trees# The results of the activity counts and observations on food and foraging habits are given belo*;. Kesults and observations Activity patten.: The results of the activity counts are given in Tables 5*1 - 5*5# snd Figs. 5*1 - 5*5# They indicate that the ants are iiiainl^ crepuscular and nocturnal in their activity. 3h the absence of Crsraatogaster clariventrls activity starts at about 15*00 hours rising to a peak at 17*00 hours and nay continue either steadily throughout the night (Fig#5*1) , or show a steady decline (Fig.5#2)# Peak activity *3&s found to be very variable# In one case it -aac between 17*00 and 18.00 hours. (Fig#5*1) whilst in another it fell between 10.00 and ?3*00 hours (Fl£i5#2| and 5#3). The period of low- activity ims betuoan 09.00 hours to 15*00 hours# This coincides uith periods of hi£h tQrtyer&tvros and light intensities, and law relative humidities. Activity build-up usually fcUoivod a rise in relative humidity and fall in atao spheric temperature and light intensity (Fics. 5.1 - 5.3). There is a change in tliic activity pattern in presence of C^caat^aster o^rlventris, Unlilx the clariventrls free areas 67 University of Ghana http://ugspace.ug.edu.gh Hun* % 100 100 100 100 82 70 70 78 80 86 100 loo 100 100 100 100 100 100 100 100 100 100 loo luO 68 Table 5*1 SjftcroaifiOiioldGs count i4»xii07l B C » — r K e P s t & s Ji X J Total 0 84 £8 152 0 6 0 0 37 245 0 74 13 38 0 0 0 1 13 l£5 0 29 16 20 0 1 0 1 3 78 0 10 1 3 0 0 0 0 7 26 0 80 0 34 0 8 1 1 7 1?4 0 17 7 9 0 0 4 34 8 65 0 £0 5 10 0 1 0 0 34 59 0 6 2 36 0 0 2 0 5 55 0 8 8 1 0 0 0 0 3 30 0 12 29 134 0 2 6 0 47 36 0 249 7L 123 0 6 9 0 51 594 8 155 50 219 2 5 41 1 36 556 0 45 28 155 3 3 20 1 10 316 0 54 35 119 0 18 3 1 6 347 1 81 29 93 3 42 0 2 14 352 7 105 34 90 2 30 6 1 22 408 1 60 73 96 4 44 4 5 5 386 2 85 40 87 4 90 0 0 0 348 10 72 36 80 4 54 2 i 23. 361 2 86 21 124 4 90 0 20 16 453 16 41 28 143 4 70 2 0 41 417 0 15 25 73 6 137 3 42 6 356 6 57 32 185 15 16 5 6 13 426 15 71 8 150 o 37 3 0 1C m 68 1516 599 2154 47 660 111 97 405 6759 University of Ghana http://ugspace.ug.edu.gh 69 Table 5.2 Uacromi oohoides countf 14* xii# 71 Tine A B C D N © E s P t s G It I J Total enther Tenp. Hun. of 9.00 0 3 17 14 12 2 3 0 5 12 58 79 100 10. (30 0 3 17 3 14- 2 2 0 7 9 57 83 100 11.00 4 11 7 4 1 2 30 0 1 26 88 88 82 12.00 1 1 1 0 0 0 5 0 2 10 20 90 70 13.00 4 2 21 1 0 0 10 0 2 8 48 93 70 14.00 2 0 0 0 0 0 7 0 1 31 41 91 78 15.00 2 0 0 0 0 0 3 0 16 101 122 90 80 16.00 25 29 32 22 31 1ft 12 0 20 105 286 82 86 17.00 110 104 84 32 100 69 42 18 54 162 785 79 100 18,00 114 99 115 89 85 89 41 70 46 158 875 75 1j0 19.00 89 119 50 67 58 62 50 87 25 116 732 73 100 20.00 113 97 87 50 47 56 53 76 32 99 710 72 100 21.00 9'j 126 84 94 42 80 47 71 21 92 756 70 100 22.00 74 75 68 83 36 58 51 62 6 57 570 70 100 23.00 104 76 52 47 22 61 34 85 10 60 551 70 100 24.00 89 50 70 81 59 80 73 72 12 63 649 70 100 1*00 109 79 34 53 56 74 10 76 8 77 566 70 100 2.00 73 117 64 38 61 39 16 89 6 62 565 72 100 3.00 81 65 69 94 53 56 43 74 10 54 599 71 100 4.00 34 62 J1 40 10 47 32 58 7 65 386 71 100 5.00 62 69 115 88 26 86 24 46 19 57 592 70 100 6.00 57 126 75 91 7 64 13 42 11 41 526 71 100 7.00 43 104 40 32 14 19 6 45 1? 43 363 71 100 8.00 14 114 28 17 9 20 9 38 10 53 3i2 73 100 To tel 1308 1533 1151 1040 745 976 616 1009 343 1366 10287 University of Ghana http://ugspace.ug.edu.gh 70 Table 5*3 ttiaoromi goiioidss count© 20« i.v« 71 Time A B C P W S © 0 F t & 3 H I J Total Weather Venn. Hum. of ; 8.00 12 6 52 47 1 2 4 7 0 0 111 84.2 100 7.6 9.00 0 0 9 7 4 1 1 0 0 0 22 84.0 99.6 8.5 10.00 8 0 9 1 0 0 1 0 0 0 19 83.0 86.0 9.2 11.00 1 0 2 1 0 0 0 1 0 0 5 91.8 81.0 9.4 1S.00 0 0 26 0 0 0 0 0 0 0 26 93.4 77.0 9.8 13.00 0 0 1 0 0 1 1 0 0 0 3 93.3 77.1 9.2 14.00 1 0 0 0 0 0 0 0 0 0 1 92*8 77.4 9.0 15.00 0 0 0 0 0 0 0 49 9 0 58 82.9 38.7 7.6 16.00 35 9 7 21 3 7 0 5 2 0 89 82.6 100 7 .2 17.00 1 5 3 38 3 7 1 96 13 6 155 78.1 1 0 2.0 18.00 50 36 74 30 ,11 134 111 135 26 79 556 74.0 100 0 19.00 93 30 56 43 21 14 0 88 44 29 418 74.0 100 0 20.00 62 43 91 20 26 115 11 95 & 40 553 74.0 100 0 21.00 95 51 106 24 58 88 E 96 20 42 580 74.0 1.0 0 22.00 82 36 163 36 52 40 3 101 20 46 576 74.0 luo 0 23.00 82 34 116 27 45 124 £ 108 19 41 596 74.0 100 0 24.00 55 46 76 17 21 135 H 94 10 24 478 72.0 ico 0 1.00 63 45 98 11 22 85 X 79 13 25 441 72.0 100 0 2.00 60 64 71 53 26 106 £ 102 20 37 539 72.0 100 0 3.00 82 30 66 17 20 90 £ 85 17 25 432 72.0 100 0 4.00 86 28 29 19 7 148 93 22 31 463 72.0 100 0 5.00 55 30 87 20 5 66 Q6 17 13 379 72.0 lvO 0 6.00 31 24 81 2 4 39 50 6 10 247 73,0 100 0 7.00 19 18 76 0 0 15 15 20 0 163 74.0 100 0 Total 973 535 1253 414 329 1387 19 1387 339 44fi 6886 University of Ghana http://ugspace.ug.edu.gh 71 Teble 5.4 Macroalscl oldeg counts 23. Iv. 71 Time N e a t 6 w ee.ther A B C V E F & H I J "Total m__ Hup. A Li^t 8.00 6 1 1 0 0 4 1 4 0 0 17 80.5 94.4 7.7 9.00 0 0 0 0 0 2 1 5 1 0 9 85.6 83*2 8.4 10.00 0 0 0 0 0 0 0 2 0 0 2 90.5 75.6 8.9 11.00 0 0 0 0 0 0 0 0 0 0 0 90.5 71.5 9.6 12.00 0 0 0 0 0 0 0 0 0 0 0 90.0 70.0 9.1 13.00 0 0 0 0 0 0 0 0 0 0 0 90.0 68.2 8.6 U.00 0 0 0 0 0 0 0 0 c 0 0 91.3 67.3 8.5 15.00 0 0 0 0 0 0 0 0 1 1 2 90.5 69.7 8.0 16.00 0 0 0 0 0 0 0 0 0 0 0 89.3 74.9 7.9 17.00 1 0 0 0 1 0 0 2 2 0 6 85.7 82.4 2.5 16.00 6 4 2 0 a 8 0 23 10 61 135 82.8 89.3 0 19.00 29 6 33 1 59 61 16 61 54 42 362 80.8 99.0 0 20.00 13 7 40 0 40 40 25 38 96 47 348 79.^ 100 0 21.00 4 44- 30 0 31 26 43 43 77 57 355 76.0 100 0 22.00 40 68 49 3 39 37 20 40 61 69 426 76.2 100 0 23.00 19 68 84 0 91 34. 12 21 122 65 516 75.0 100 0 24.00 24. 84- 46 21 114 20 7 18 134 64- 512 74.0 100 0 1.00 21* 61 75 6 71 32 16 14- 115 51 465 74.0 100 0 2.00 9 84 62 5 % 44 27 46 13£ a 567 74,0 100 0 3.00 16 45 58 4 68 25 14 81 82 61 454 74.0 100 0 4.00 29 66 42 12 73 38 14 20 90 55 439 72.2 100 0 5.00 36 172 38 16 136 30 4 23 52 22 529 72.0 100 0 6.00 8 62 4 0 39 24- 2 14 C 3 162 72.8 100 6.5 7.00 4 8 4 0 3 7 0 6 1 0 33 75.8 100 8.1 Total 268 780 566 68 880 432 202 461 1018 662 5339 University of Ghana http://ugspace.ug.edu.gh 72 Table 5.5 Macranischoldes count 23.iv.7l Time 4 B C D E F 0 H I J Total Tcnp. Hueu Sab Light r- % _________ 8.00 1 0 0 0 15 3 0 1 0 1 21 80.5 9l*.lj 7.7 9.00 0 0 0 0 2 2 0 0 0 0 h 85.6 83.2 8.1* 10.00 0 0 0 0 1 0 0 0 0 0 1 90.5 75.6 8.9 11.00 0 0 0 0 0 0 0 0 0 0 0 90.5 71.5 9.6 12.00 0 0 0 0 0 0 0 0 0 0 0 90.0 70.0 9.1 13.00 0 0 0 0 0 0 0 0 0 0 0 90.0 68.2 8.6 lli.00 0 0 0 0 0 0 0 0 0 0 0 91.3 67.3 8.5 15.00 0 0 0 0 0 0 0 0 0 0 0 90.5 69.7 8.0 16.00 0 0 0 0 0 0 0 0 0 0 0 89.3 71*. 9 7.9 17.00 0 0 0 0 0 0 0 3 0 0 3 85.7 82.1* 2.5 18.00 0 6 0 2 3 2U i* 3 21* 9 75 82.8 89.3 0 19.00 18 8 9 59 25 29 I* 8 28 29 217 GO.G 99.0 0 20.00 8 65 122 cc 55 CO 10 13 90 67 596 79.1* 100 0 21 .(X) 8 10 59 39 39 71 1? 8 23 91 367 78.0 100 0 22.00 5 33 119 57 hz 1*3 15 10 72 33 1*29 76.2 100 0 23.00 11 121* 110 62 29 58 25 13 83 21 536 75.0 100 0 22i.OO 2 213 131 93 30 72 27 19 72 62 721 71*.0 100 0 1.00 7 285 21 I* 66 28 62 21* 5 71* 52 £17 71.0 100 0 2.00 8 85 159 ii8 31 38 IjO 11* 81 30 531* ?l*.o 100 0 3.00 3 99 168 38 51* W 11 13 112 72 598 71*.0 100 0 4.00 3 11* 1li3 28 36 62 23 11 86 78 616 72.2 100 0 5.00 8 11*3 155 1*3 12 1*7 7 5 1*0 53 513 72.0 100 0 6.00 0 1* 20 1* 7 6 3 3 10 11* 71 72.8 100 6.5 7.00 Total 1 1 3 1* I* 10 831222 1152 629 1400 6)jS 6 1 1 0 218130 776 612 26 611*7 75.8 100 8.1 University of Ghana http://ugspace.ug.edu.gh University of Ghana http://ugspace.ug.edu.gh University of Ghana http://ugspace.ug.edu.gh 006 75 University of Ghana http://ugspace.ug.edu.gh Ccl fl. ci yg .nt ri? University of Ghana http://ugspace.ug.edu.gh C cl ar iY en tr is 77 M» . ec ul oa tt fg an d C. cl ar tv et tt rl fl ac ti vi ty co un ts in pl ot DU ** Ta £o f 23 rd Ap ri l 1^ 71 I*i £t ha nd sc al e 0, 3 • 9) t li gh t in un it s (W eo to n Ha ir to r V ex po su re me te r) Ri gh t ha nd se tt s (0 *1 0 • 1Q 0) | te mp er at ur e in rj R. H. pe rc en t® University of Ghana http://ugspace.ug.edu.gh there was virtually no activity during the day and start tot activity was delayed till about 18*00 hours* Macrcroischoldes activity Increased as that of clariventrie declined (Figs*5*i & 5*5) • Peak activity was found between 2l*.00 hours and 02.00 hours* The presence of a predator such as Qecoph.vlla longinoda i6 another factor which effects a change In the pattern* Ctoce Qecophylla lays siege to the nest entrance* Kacrcanischoides Hill not caae out until the former retreats* Thus the day activity pattern is modified* No quantitative work was done on this* Rather surprisingly, there was a sharp increase in activity at 05.00 hours on one occasion* Following seme ants it was found that large numbers of the© were drinking water (condensed dew)* This has since been found on several occasions as shown in Figs* 5*1 - 5*3 and 5*5* The drop in tenperature and light intensity that precedes rainfall does not seem to affect activity* During a rain storm however the ants were seen to stop still, brought the legs close to the body and gripped the substratum finaly* l»8 Association of ra^ aaiachaLdefl with lycaenid caterpillars not Tree No* of larva© Mo* of ants No. of larva* without ants HlObl 1 ! 1 11 - B 2 1 1 mm Hl0b2 3 1 o✓ - f? n 1 6 - u if 1 2 - 0 n 1 8 - II « 1 5 - n 14 <» 1 - i» 5 1 1 « • Total 5 9 hh am University of Ghana http://ugspace.ug.edu.gh Perhaps by being nocturnal the ants ©scape high predation, since most of their predators* such as, QeccgferHfe lon&inoda, aam Eeduviinae (e.g. Acanfchagpis billneolata), raaay skinfce (e.g. *>•) are day feeders, but most geckos ere nocturnal and arboreal, perhaps their nocturne! activity pattern ie a device to synchronise their foreging period ■Kith availability of their prey. But a look at the list of prey (Table 0.6) does not sees to suggest this. Perhaps the isost important fidvpntagf: is an escape fraa desiccation. Macrctaischoiries constantly nerve brood fraa me nest into another during their active period. If this were done during the day when the evaporation, rat© is M fft das to high temperatures am1 lou relative hmidities, the larvae ia5#it die of desiccation os? at least be adversely affected. A relative humidity of about 60 percent and a taEperature of 23«6°C to 26.6°C seen to be suitable for activity huHd-’2p. Their inactivity during the day may also be determined by a negative response to light, since the ants always moved fraa any bright light shone cn them during the night. But light may be ploying a more icqaortant raLe than this* It Is very likelj that light rather than tes^ erature and hiaEidity is the main trigger that sets normal activity in aoticn and barings it to a stop. A careful scrutiny of figs.5.3 -5.5 seems to support this, for on all the three occasions when light intensity was measured, increase activity started just after a sharp drop in light intensity at about 17.30 hours, and continued throughout the night only to decline at 06.00 hours the following mowing after a sharp 9h University of Ghana http://ugspace.ug.edu.gh rise in litfvt intensity. Suitable Uwpemtwce and huntaity perhaps help to ttMyfnt.fl-j'n -the activity after it has been triggered by light* But in the day tiei©* under naraal fiaM conditions, 3Jght intensity 13 positively correlated with tec$>erature and hence with humidity, thus making it difficult to determine ^ ahich is responsible for the low activity level although all three m«y be jointly responsible {A&re, 1953) * Qhlike the situation in the clarivmtries free area, t&igfq thare vas same slight activity daring the day, the ants regained totally inactive daring the day in areas %ivere clarlventris is present# Thus the presence of doriventrls is responsible for the complete absence of activity# Mien the activity of c&ariventgis declines that of aculeatu3 increases# The two ants are therefore able to coexist because they are segregated in tixae. However the same meteorological factors that influence activity in the dariventris free areas are In operation sihore it occurs as well. Perhaps the habit of Macmaiseholdes of searching all crevices end other dark areas such as forks, angles between pod and stem, and branch apices is idiat makes it effective predator of capsids# Xoudeteei (1968) has shown that capsids usually aggregate in the dark places between the pod and the stem, inhere the females oviposit# Thus by frequenting these spots ovipositdon will be disturbed and th© tender first instar nyzsphs predated# i02 the trail interruption experiments outlined really show cnly that th© ants Hill not cross clean ground. They do not show *ahat University of Ghana http://ugspace.ug.edu.gh 96 ms on the ground or how the ants use it (Sudd* 1?67). 3h® only evidence of trail laying was the ant that moved with the body low mentioned earlier. Perhaps it uses its sizing with secretion fraa Dufour’s to lay the trail, as this ie a ccanmcn feature of the anhfanrfly ffcrmieiriac. The anta do flollow trails since the recruited workers moved an the marked path of the scout. Trail laying and following are big assets to nocturnal insects like Haeroaischcddes. since they can go out to forage and return to the nest without difficulty. Their food is highly varied as the list shows, though there sere few capsid renains, this doss not man the ants are ineffective against the cocoa capedds. Thus Colltngwood (17fi ) has categorically stated that I&eraRlscholdes is "consistently negatively associated with Distantialla thecfarqaa1’. This may iqply that the capsid nyisphs being so soft-bodied are preferred to other hard-bodied insects like the ants. The scarcity of capsid remains found Bay also be due to their being soft-bodied. Maorosaiacholdes obtains its protein supply iby feeding on other animals. But carbohydrates are very essential in ant diet. These it obtains by attending coccids, aphids and lycaenid caterpillars which produce honoy-dew and other saccharine liquids. Another source of carbohydrates may be fpcsa the soft juicy coating of the seeds of Solanua varbasclfoilium which are ccEEWily found incorporated in the nest in areas where the trees occur. Lestcn (personal coEnmunication) flnda that MaCT«aiBc^ ildea utilises the sugars of e&trafloral nectaries of at least three cannon University of Ghana http://ugspace.ug.edu.gh cocoa Tana shrubs* Minerals aud essential sal is tfcc^ obtain perhaps £ro& bird or reptile droppings they feed on# Qfer (1 >'?Q) has reported a similar finding in Polyrachis siaple&,ln Israel* The scalo-iasects sad xaoalybugs ■sdaich these ants attend are not vectors of the s&dlcn shoot virus disease (Strickland^ 1951 )# The aphids are act serious pests in the cocoa fanas* iiiile the l^ caenide are relatively lasissportant defoliators* Lodos (1968) claims that tfe© ants attend the sgga of B» thalassina and drive a&sy parasites but do not carry the eggs Contrary to tftis observation soroe eggs «er© found in the nests aad those found being attended i*ere all parasitised* These observations are not in conflict* If the ants found the eggs only after they nor© parasiti&ed, they sight not kaa® the difference* VbaA does a©ea likely is that if to^ ^attend the eggs, they labor predate the first instar nyraphs* The rate of sproad of the radioactive sugar through the various colonies within such short periods (Section 3) denotes a high rate of inter-nest laovement* It is very likely that the rate was aided W foed^ sharing, a com m behaviour of ants* Chauvin et si* (1961) found that radioisotope give© to Formica puJLycteaa was transmitted to g* ruffe by oral exchange* Iloi^ evor, M* aculoatus has never been seen indulixing in food sharing outside the nest* Perhaps this takes place inside the nest* >7 University of Ghana http://ugspace.ug.edu.gh 98 Section 6. Iaterpelatlonahips of EL aculeatos with other ants University of Ghana http://ugspace.ug.edu.gh 99 6. INTEREIATICKSHIPS OF M. ACULEATUS WITH OTHER ANTS Introduction One of the few available c aments on the interrelationships of M. aculeatus other ants such as Qscoohylla longinoda and Creaatog aster species In its environment was by Coll ingwood and King 0970). These authors, using direct field scoring methods, claimed that M« aculeatus was significantly negatively associated with Qecophylla longinoda and Crematogaster spp. Thqy also stated that this negative association was not absolute for 0. longinoda and M. aculeatus occasionally occurred together whilst M. aculeatus and Creraatogaster extremely rarely coincided* The species of Crematogaster worked on were not however mentioned. Leston (personal oomication) however suggested a positive relationship between M. aculeatus and C. clariventris free the results of the analysis fcf his field data. Roaa (personal cdonunicatlon) has demonstrated this relationship with fuller data and a better sampling technique. The present work was undertaken to attempt to establish in more visual, and thus more concrete fora, the interrelationships between M_._ aculeatus end its neighbour ants such as 0. longinoda. C. » C. clariventris and other minor ants. Materials and methods tohen mapping the distribution of M. aculeatus with respect to the different shade regimes, dominant ants, such as Q« longjnnriA atm^ Gj. 3tadelmanpi occurring In the area were also napped. Two more areas University of Ghana http://ugspace.ug.edu.gh were later mapped to show the distribution of aculeatus and C* clarlventria. The degree of coexistence was detemined by calculating the Coexistence Index (Ed) and the coefficient of coexistence (Sc) for each mapped area uglier Hay ash Ida *s (1960) method* A number of other indices* using the same parameters, are available* (Southwood, 1966). Ed measures the degree of coexistence of one species (A) to another (B) in a whole area* It is simply calculated by the formulas Ed - lOO.h a h and a are respectively the number of samples in which both A and B or A only was discovered* For our purpose Macraaischoides was taken as (A) and any other ant occurring with it as (B)* Coexistence in each habitat, Ec, which for our purpose are the different shade regimes, was obtained by the formulas Be - hn ab a, b, h are respectively the number of samples in vhich either species A or B occurred or in which both species were discovered together and n is the number of samples in each habitat* The coexistence positively or negatively deviates from the chance score, of which statistical significance is given by X2 test with Yatefs correction* However a change was made in the foxnxula as given by Hsyashida to suit the present purpose* In the first place Ec was not calculated for each habitat, such 100 University of Ghana http://ugspace.ug.edu.gh as each shade regime j rather each ^ mapped area was taken as being a habitat# Thus in fact both Ed and Be measure the same thing, such as the degree of coexistence. Secondly chi-square was not used to check the significance because some cell values were too small, scroe even had zero. Ant predators i Field observations were made to find out the predators, and Oscophylla longlnoda and Polyrachis laboriosa in the radioisotope experimental area were monitored for radioactivity. Results and observations Figs.3.ii - 3.6 and 6.1 - 6.2 show the distribution of M. aculeatus in relation to 0. longinoda. C. stadelmarmi and C. darlventris respectively. Table 6.1 shows the interaction of the ant species. Macromischoides, Oecophylla interactions Qocophylla longinoda is a day fieeder that nests on shade and cocoa trees usually in more open areas. It generally forages on Ilacrocsdschoides nest-trees but occasionally the two nest an the same tree. The two are antagonistic and fierce fighting ensues if they meet. Oecophylla even besieges, attacks and predates Hacropdschoides. They were found to be radioactive when monitored, and a worker was found carrying a radioactive Macronischoides larva. Moreover three out of five Oecophylla nests in the area were radioactive. Sane Oecophylla remains were however found among food collections in Macromigcholdes nests. The test for the degree of coexistence gave 20 per cent as the greatest value in the three mapped areas (Table 6.1 ). HacromischoMes» C^_ stadelmanni interactions Ifalike Oecophylla Iccginoda Cm gtadabaami nests only on shade trees with its carton nests high up 101 University of Ghana http://ugspace.ug.edu.gh M ET RE S £0 102 “Oc C v> Q LU CO o t/> CO c cu I/} 3 o _CU _ o~EnJZ Q _ co cx> £ w c > Cr> c k— •+-J C ” 5 - ) _ c 1 2 C D O _g u O 2 O <+- U o s u a 2 o Ck) O O d Dk_ ^ Io cO o 8 « O o _ g 3t>JZ CL 8 ■O 00 (/)Cx) c ^ ccu= _ g J ZC L O C c o o o o ^ o -8 O © 0 IS □ i *13 c a> 3 -+-> O 0^ jQ Wcr> m cn C ZJ O c/) 0 -- o 1 I I I o © © o © o © © o o o o o o o o o o © © ©^ © © o o o o o o o o o o o o o o o o o o o o o o o o \ w/o" o o o o ©Do © o o o 0 © © o o o © © o © © © i ° n ° o © o © o" o o ©^ © o o © o © o o o © © © o o o o o o o o o o © o o o © © © 9 i o o o © o o o © © © © o o © © o cy oDo o o © © © © © © o o o o o o o" o o o o o o °i° © © o o o © o o o o 0 □0 o © © o o © © © © o 0 cmdD o o’ o o o o o o o o © o (m • •x\© o o o o o 6 o o o o o|o ©^ 9 • ♦ vO o o’ o o o o o o o o °($ 4ft ♦ /O)♦ »)l o o o o o o o o o o o o'V ® / C * f o 1 5ff Q Q O g o o o1 o o1 o o_ ,_L . o o o o1o o o o O u t/ ) o to UJcc UJ z ! I »S g- I CVi v0 University of Ghana http://ugspace.ug.edu.gh f 5 I 5 8 I o I I 8 <*4 O IAVT\r - o n n r - « - CM O O fl O wcvj o CM CM M • 1 u T | l l a P* «S E-» 6-« © & JS3 sO CM CO c- VO•-o fl a * c0 a 1 !9 ~=t 5 -3 S a E-i u\ c^ oo ° s s I I £ C& XA £ 3r - CM O COVO-^ 1e-« CQ«o 50 I VO *“ fl % o 25 N S 6-t co o r - VTWO I) £ ® *4£o o35 co 04 \OCOCM *M *A T- O o 8 a * » 1 1 * 5 1 3 £L* *4! fc-« vO VO CM t - CM a >o cm co .O r - « -HjC ^ ' lto o^r-^ 1 1 * 3© W 4? o 0 1 a % » o a s miAQD -g CM ^ \ A O CM CO O o O p-CM t * - O © CM CM -Z? I © r- c— <3 O I s c-» *H *-x /*\ 2 ia§ £ Vi ?■§S S 3 4. §w o University of Ghana http://ugspace.ug.edu.gh 105 the tree* It forages however co cocoa trees and erven an the ground* The two ants usually fight when in contact. Fig.3.6 shows that there is no overlapping whatsoever between them. 3am deserted Etecromischoides nests were however found on trees now occupied by stadelmnni. 3d value for the two is zero per cent. Macramischoides. C. clariventris interaction* C. clariventris nests on tall shade trees and like C. stadelmanai forages and nests on cocoa md other trees and shrubs. It seems that there is little hostility between the two species as they seldom fought when in contact. On few occasions moreover, clariventris workers were even found resting in deserted aculeatus nests. The two ants coexist fairly well as Fig.6.2 indicates. There seems however to be a superficial negative association between them in plot Cli as Fig.6.1 shows. The association tests were therefore positive in plot Dili, but negative in Cli. MacrcBd-schoides versus other antss PolOTachis laboriasa coexists with M. aculeatus as both very often nest on same trees. Moreover the former moves fl*eely axaong the latter. They became radioactive during the radioisotope tagging experiment. F. riLvoill does not construct nefefes of its own. It nests rather in prefomed cavities in dry cocoa pods and trees. Thqy were found on several occasions occupying deserted M. aculeatus nests. P. rivQiij were vanquished by M. aculeatus when a straight encounter was organised between them. University of Ghana http://ugspace.ug.edu.gh Carapoaotua acvuMviny^ flj g does not seem to be hostile to M. acplftatrns* C. acvapifflensis moves freely but rapidly among Haaromisaholdes. Mother ant, found occupying same trees with aculeatus is Cataulachus sp. Pheidole magacephala was found occupying trees in Macraaisclioldes areas. In C5a Pheidole sp. attacked and destroyed an incipient colony. 2c the laboratory they always attacked and annihilated cultures. Discussion The results obtained show that 0. longinoda and M« aculeatus are mutually exclusive. This observation agrees with CoUingwood and King's (1970) finding. There is however an occasional overlapping, which is more transitory than permanent, since Qecophylla nests were found on trees previously occupied by Macromischoides. It is very likely therefore that 0. langinoda pushes M. aculeatus out if they both occur an same tree, especially so since both nest on leases. This suggestion is further substantiated by the earlier observations that Qecophylla actively besieges and predates Macrcraischoides. The fact that Qecophylla nests became radioactive and a worker was found with radioactive Kacroalschoides larva proves beyond doubt that Qecophylla exerts predatory pressure on Macraalschoides. Gibbs (1969) asserts however, that Qecophylla succeeds in reducing the Hacrcciischoides population only in areas where it is already low. JEhfcre seesns to be mutual predation between the two species as some Qecophylla rmains ware found among Hacroroischoides food store. M. aculeatus not attack Qecophylla but aay kill Qecophylla in self- or nest-defence, or they way still obtain Oacpptella corpses. 106 University of Ghana http://ugspace.ug.edu.gh Gibbs (1969) claims further that Craiatogaater is largely excluded by Qecophylla. This was ccnfimed by Colllngwood and King (1970) and the present results indicate a very strong negative association between the two ants. Hie straight fight between Oecopfrylla longinoda and Creraatogaster dariventria in plot D1J* is evidence of the hostility between them. The same degree of hostility exists between Hacromischoides and some Creiaatogaater species, especially C« stadelmannl* The two ants are antagonistic and fight x±ien In contact* The same cannot however be said of C# clariventris. This ant coexists fairly well with Macrcmischoides« In fact the two are isutually tolerant and rarely fight when in contact under normal circumstances • This does not exclude the possibility of fighting if the two ants are suddenly brought together# Hew clariventris cooes to occupy aculeatus nest Is not certain# But it may not be an active process, such as fighting# It is ve*y likely that the occupied nests had been previously deserted by aculeatus # The interdigitatlon of the two ants in Plot Cl±, (Fig.6#1) laay be an Initial stage of coexistence# Perhaps they are able to coexist since, though they occur esi same tree, they occupy different parts of it# Hacrceiischoides nests on leaves, and being largely nocturnal does not usually make use of the branches and stems where C, clariventris rests and forages in the day time# Mhen clariventris activity declines in the evening, the branches and stems become free for Macroraischoides to use# Urns differences in nesting and resting sites as well as seggregation in time play an i m p o r t a n t pert in making their coexistence possible* Collingwood and 10? University of Ghana http://ugspace.ug.edu.gh King *8 0970) claim of the rarity of coincidence of frlacroraischoideg and all Crematogaster ante is thus not £rue. Among the minor ants P. laboriosa is closely associated with H. aculeatus* Cataulachus sp. and Gacgxaiotua acvaprimensis also scan to be neutral but to what extent is not known. P. rlvoill may occupy deserted M. aculeatus nests. Pheidole ants are very hostile and responsible for the destruction of many Incipient colonies. 106 University of Ghana http://ugspace.ug.edu.gh 109 Section 7* Interrelationships of M. aculeatus with other animals (excluding ants). University of Ghana http://ugspace.ug.edu.gh 7. 3OTESRELATICHSHIPS CF M. ACHLEiT® WITS OTIES AHIML3 (EXCLUDING AHTS) Iptroductian A vast, number of arthropods or© ©ssoclated with ants in one is&y or another and far various reasons* These insects or other arthropods that inhabit ant—nests either throughout lif© cs* during one or rsore of their developmental stages are collectively known, as myn&eccphiles or ant-guests in a broad sense. Wheeler (1910) subdivided the E©rjaeccfihiles Into four major groups* Hfeose which live as scavengers or predators end which the ants treat with hostility! those which attract less attention due to their smallness cr sluggislmess and are thus indifferently treated} the irue-guests which are thus amicably treated and even attended; and the parasites both external and internal oases. Other arthropods or invertebrates in general and aotie vertebrates predate ants and are called raynuecophags. A vast literature exists on all aspects csf the above subject. Notable and relevant ones for our present, purpose are the works of iiiasmann, Janet, Wheelers -tea and Leixfy. Not such attention was given to this subject as a whola. However the little information obtained on sane aspects of the subject is given in the following paragraphs. Materials and methods Nests brought to the laboratory for population counts were examined for inquJlinoo, parasites sad predators. Those found were 110 University of Ghana http://ugspace.ug.edu.gh placed In 10% alcohol and labelled. In the field several ant® infested with ectqparasitic mites were collected on leaves far sway from their nests. The number of mites and their positions on the body were recorded. Ant workers with distended gasters were dissected and the nematode warns were rmoved and placed in 10% alcohol. Gut analysis was made on a aklrfe suspected of predating the ants. Results The main Inquilines found were the psocids which live in the nest wall. These are very small end fast moving. Their number in the nests was very variable, but as many as UQ were counted in one nest. Since they quickly vanished out of sight t&en a nest "was opened they were never observed feeding. Several CdHembola were found living with the ants in a laboratory culture. These were scaahraagers living on the food resaains especially the decomposed or dried ones. The Collerahola, though not living inside the ant nests, moved freely among them. Louis (personal cQncminication) lias found out that the nymphal instars of some Reduviinae, e.g. Acanthaspis bHineolata predate the ant and deposit the corpses on their backs. Several Macromischoides ants were found in the gut contents of the skink, Habuva sp in the laboratory. The ectoparasites found were socie red and brown mites. The red ®ites attach to the lags, especially the femora, and there night be as many as four on one ant. The infested ants looked very sluggish and usually remained motionless on leaves as if resting. The brown mites are much bigger and attach to the body rather than the legs. HI University of Ghana http://ugspace.ug.edu.gh Long whit© nematode warms vrere c o l l e c t G d fran. th © gaeter of many workers* The warns* usually coiled several "times over, occupy the visceral cavity and may measure up to 30mm. Discussion Th© psocids (Psocoptera) are synachtrans that feed on the organic waste of th© colony* Qfer (1970) reported that they had been seen to attack and harm the cocoons of pupae which had been separated frcm a colony of Polyrachis simplex* 2h his list of British synoeketes, Ingas (1925) Included the Collembdla of the genus Crahodmua* There is however no report in the available literature of their being harmful to their hosts* Possibly since they multiply fast, much faster than their hosts, they might interfere in one way or another with the day to day activity* But in the field M. aculeatus preys on Colleznbola* Skinks are cccsaan in the cocoa farms* They are day feeders that move mainly on the ground but occasionally climb cocoa trees* Though they were never seen predating the ants in the field, it is not unlikely that they do so since the ants forage both an trees and on the ground* The laboratory finding seems to confirm this* It is very interesting that several ants infested with the parasitic gemasid mites were collected outside the nests and few were found in the nest populations. The reason for this is not clear* Such infested ants will not survive since, as has also been shown by Qfer (1970) working with Polyrachis simplex. they are unable to fetid themselves. The nematode worms may also add to the mortality rate. 112 University of Ghana http://ugspace.ug.edu.gh Leldy (1851) described a similar worn in a grasshopper and claimed that the worm cofcld extend from the abdomen into the head* This may not be possible in the ant which ha3 a constriction between thorax and gaster and may explain the extensive coiling of the worn in the gaster# Perhaps the worst enexay of M» aculeatus is rain star®u Storm. damage has recently been found by Leston (personal coeimmic&tian) to cause mass deaths in H* aculeatus and 0* longinoda. The storm forces sandwiched nests apart and dislodge both exposed and sandwiched ones* 113 University of Ghana http://ugspace.ug.edu.gh lilt Section 8* B ests and nesting habits. University of Ghana http://ugspace.ug.edu.gh us 8. NESTS AND NESTING HABITS Introduction There is very little infaraation in the vast ant literature on the nest and nesting habits of Macrocdsdhoides aculeatus* The earliest mention of it was by Santschi in 1309 (quoted by Wheeler, 1922) who described the structure of the nest as consisting of the !lea£ of a tree or shrub rolled v%> and lined with a felt-work of very fine vegetable debris and of a nyceliun bearing fructifications0 • In 1916 For el (quoted by wheeler, 1922) commenting cn the variety rubroflava remarked that it was found in nests woven of silk and fixed to leaves. Wheeler (1922) tod: both authors to task for their r>iaiscanprehensi<3Qn in regard to the structure of the nest* He correctly described the nests as consisting of particles of the iaost diverse vegetable substances, bits of bark, dead leaves, trichases etc* loosely felted together sad invaded by a fungus r^ yeelium bearing no fructificaticais* Hied and others working at the then W*A*C*R*I* (now C.R.I.G.) described the nest as "carton” end built on the underside of leaves of trees and shrubs (reported by Voelcker, 12h7) * Ledoux (1958) describing the nest erred by including soil particles in the building materials* Ledoux’s illustrations are inaccurate* The present work aims at correcting saae of the errors of the earlier authors and at providing more information on the description, materials, construction, architecture, and type of nests, as well as the light and height preferences of the nesting sites, not leaving out information on the causes of nest abandonment and the stimulus for nest building* University of Ghana http://ugspace.ug.edu.gh 116 Materials and zaethodg Neat hirLldinfr M Several workers from three nests together with brood and one queen were released into a glass cage containing three cocoa seedlings as veil as leaves, pods and twigs of cocoa, fresh and dry leaves and also fruits of Solemuxa verbascifalium, a shrubby weed ccssmm in cocoa farns, in addition to litter frcm the cocoa farm, and ©oil in petri dishes* The aim was to simulate the cocoa farm environment as much as possible* The lege of the table on which the c&ge sat were stood in cans containing gas oil to keep out other insects* Water was provided in petri dishes and the ants were fed on insects obtained by sweeping. The contents of the sweeping net were squashed to weaken the insects and then the entire contents were emptied on the board or cage for the ants to select their prey. Several replicates were set up* but big open boards were substituted for glass cages* One such arrange­ ment is shown in Plate 6.1 * The effect of light was determined by perfuming a similar experiment in a dark room* Field observations were also made cn th© method and materials of nest building* Many ents carrying building materials were Intercepted end robbed of their load* Stimulus of nest building i Experimental arrangements similar to the above were used to investigate the stimulus for nest building* Workers from the s ssoe or different nests but belonging to the same col any were diviaed into two groups* One group was given brood only or brood and reproductive male end female alates* The second group was either left without brood whatsoever or in seme cases with reproductive a or brood University of Ghana http://ugspace.ug.edu.gh 117 Plata 8,11 Laboratory arrangement for rearing M. aculeatus* University of Ghana http://ugspace.ug.edu.gh introduced at a later period. In the £ield the workers that came out to attack when a nest was disturbed were made to move on to a twig and carried far away fraa their nest tree to be treated as 1-without brood* • Those remaining with the brood inside the nest were forced out together Kith the brood an to the ground. The two groups were then observed for nest building. Nest architectures To investigate the internal structure of and arrangement of brood and castes in nests several nests were removed froa the trees with a pair of secateurs and innediately dropped into a large polythene bag containing a large amount of either chlarofom or ethyl acetate. The ants became unconscious and movements ceased almost instantly, thus much of Hie internal arrangement of brood and castes was preserved. Back in the laboratory each nest was sectioned with a pair of scissors into strips of about 5cm wide frcn petiolar to apical rads. The contents of each strip were examined and recorded. Description of a cocoa trees A typical cocoa tree has a short main stem (usually straight) which forks into 3-5 almost horizontal branches (the so-called •fans1 or jourquette1) about 1-2 metres above ground to form the first storey or tier. Suckers grow frcci the fans into straight vertical stoas (chupons) which give rise to more fans thus constituting a second storey. Trees may be found with three to five storeys depending on plantation practice. The maximum height is however about 10 metres (Urquhart, 1955) although at Tafo the averagcjfoeight is about 5 metres. The trees usually have a bushy appearance as the branches touch the ground. This is especially so in coppiced cocoa which have several chupons growing from the cut stem. 118 University of Ghana http://ugspace.ug.edu.gh Heating habitat Height preference j A long stick marked in metres was held vertically against each tree containing M. aculeatus nests and number of nests found within the various height bands were scored. Seat abandonment! To find out the rate and causes cf nest abandonment several nests were labelled* These were observed 28 days later and a record male of the abandoned nests and visible causes of their abandonment, such as ageing leaves* mechanical i.e. nests forced apart by wind, and presence of enemies. Record was also made of new nests built during the period. Besults The caged ants first took shelter temporarily on the moist soil surface. The fallowing morning about ten workers were noted beneath a silk matrix span an the under surface of a leaf, stretching froa cne edge of the leaf blade to the midrib. A depression roofed over by the silk matrix was therefore created. Several workers started carrying building materials composed of pieces of fine rootlets, chips of dry cocoa leaves, twigs^ pods, vegetable debris, sod pieces of deserted spider webs to the building site. These were disposed readonly on the matrix. Each worker normally worked its own piece of building material into the matrix by the help of the mandibles and the first pair of legs. Several workers would however co-operate in the incorporation of large pieces. Work was at first concentrated on the firmer edges, i.e. areas bordering the midrib and the leaf edge, of the matrix. After these areas were nearly covered, attention was turned to 11 'J University of Ghana http://ugspace.ug.edu.gh th© opening near the petiole* Meanwhile sam worker® had started carpeting th© floor (loaf surface) with very fin© vegetable debris glued to to© leaf surface* At this stage evacuation of brood froa the soil surface into the nest started* Ttiae building and brood transporta­ tion went m side by side* The large opening at the pet&olar end ms partitioned* This was dm© by building a broad-based pillar on the leaf surface and extended to touch the roof. Thus the single opening ms divided into two* Later one opening was sealed leaving only one small opening which served as the main entrance* The opening at the apical end was similarly divided into two by a pillar, but both were later sealed* Another pillar was seen under construction Inside the nest, but it became difficult to find out what was happening Inside the nest since it ms nearly covered over. % the third day extension work had began on the other half of the leaf blade* Materials brought in were glued to the main lateral veins as shown in Fig.8.1 * These served as foundations upon which walls were built* The ants successfully roofed over the space between two adjacent veins* It was noticed at this stage that the number of wockers on the Job had started decreasing and warfc was almost coming to a stop* Further investigation revealed a Jarge nest that the ants had built underneath the table* This became their persaanent nest. Several nests were constructed in the laboratory in corners of stools| indeed the ants nearly always Ignored the seedlings provided them* 3h all, four nests were successfully built on leaves of cocoa seedlings but a silk matrix was used in only one* Scsae of the nests built in the comers of 120 University of Ghana http://ugspace.ug.edu.gh NES T CO NS TR UC TIO N 121 2: O w <, > K _J LU < b tr- ! § 3 5 CD 5 cr <_j O hUJ Q_ Fi q W University of Ghana http://ugspace.ug.edu.gh 122 stools and other obscure places are shown In Plates 8«2 * 8.6. D* ants kept In the dark built a nest on top of a metal plate that fastened the tap In the rocm to the wall. Under normal conditions In the field the ants usually chewed the pieces of building material into a pulp, thus making it very difficult to recognise them. Ch almost all occasions building started right on the leaf surface along the main veins. Chi one occasion in plot C$a that sons ♦displaced* ants were found depositing building materials an a deserted spider web, while others took shelter in a crevice in the stoo* This nest was however never finished. Building can go on during the day but more usually It coincided with foraging In the night. Thus both food and building materials were transported simultaneously during the activity period. Nests could be built either of homogenous or heterogenous materials, but always of vegetable origin. Plate 8.5 shows a long tubular neat built in a comer between a board and a stool. Hiis nest was built with cotton vodl fibres only. However, acme paper material scrapped from a empty sugar packet shorn In Plate 6.1 was incorporated later. Paper used in the field to label the nests was chipped off, reduced to pulp and used in nest building. Ihe white part of the nest shown In Plate 8.7 la an example of this. Several nests found in the field were built solely with chewed leaves and fruits of Solanum verfcaacifollnn. During the haxmattan whan kapok or silk cotton, the fibres surroundii® the seeds of Ceiba pentandra (Boofoacaceae), was very comaon extensions were rasde to msqy existing nests using only this material. Mere ccraonly, University of Ghana http://ugspace.ug.edu.gh 123 University of Ghana http://ugspace.ug.edu.gh 12U Plate 8.3» Nest constructed, cn a top corner of a glass cage University of Ghana http://ugspace.ug.edu.gh 125 Plat© 8«i*i Two oat of four nests constructed under a laboratory stool. University of Ghana http://ugspace.ug.edu.gh 126 KUte8*5* ^th cotton wo<*or®s Isa. a comer fo@troe& a board sad a ©tocOL. University of Ghana http://ugspace.ug.edu.gh 127 Plate 8.6* ka unfinished nest an the side of a stool. University of Ghana http://ugspace.ug.edu.gh 120 Flat© 8.7 a An exposed nest with the white part constructed from paper end the black portion from the usual building materials. University of Ghana http://ugspace.ug.edu.gh however, nasta are constructed of heterogenous vegetable materials as already described above# On the 4-loor and walla of laany nest© were ramifications of fungal mycelium# Si© mils of sarae few nests seemed to have a moss covering# Usually the floor of the nest has a wet shiny appearance* as if polished with a wax-like substance. Nest architgctujrgt The various sections cut showed that a nest has two jaain architectural plane* It iaay either be single tier or several tiers# Single tiers were very canon in sandwiched nests where growth in depth was hampered by the leaves that fom the floor and the roof. These nets were longer and broader than they were deep# Exposed nests usually tended to be more compact and contained two or even several tiers of intercocmttiicating caapartaents# It was also evident frasi the sections that no special cells vere given to a particular caste car instar Thus eggs, larvae, pupae, and workers could be found in all the compart­ ments though there might be a predominance of a particular developmental stage, say, larvae, in a particular chamber# Ho special queen cell was also found# ffypes of nests i As has already been outlined above there are two main types of nests s exposed and sandwiched nests# The former are usually single cctopaci nests constructed on the ad&xial parts of a leaf or leaves In this type the leaf surface forms the floor whilst a definite roof is constructed# These nests are .usually taore complicated, contain more cells or ccrapartcjants than the sandwiched ones. The sandwiched ones iaay be built between leaves either on the same branch or different branches cm the same tree or f!rc*a different trees# Here the ab&sdal side University of Ghana http://ugspace.ug.edu.gh of one leaf form the floor, whilst the roof is farmed ty the adajdal surface of another leaf* Type® of nests are shown in ELate*8.8£ & b* Nesting habitsi The results of the investigation to deterzsine the preferred height for siting nests ere shown in Table 8*1 and Figs.8.2 end 8.3. It \ms found thet the preferred height lies between one and two metres fraa the ground as shorn in Fig.8.2. Few nests are found above three metres from the ground end ame above four, fetian the ants were displaced they took shelter in cavities* crevices and forks of trees In the field. Sam workers from a laboratozy culture broke off frcm the mother colony and occupied a hole in the soil in sshich a seedling ims gram * This is shown in Plate 8.9# Stimulus for nest building i Workers with either brood only or brood* plus reproduetives always built nests and very fast too. They initially found temporary shelter for the brood under stones* leaves and other sh®d©d and moist places before nest building started. In all the four laboratory end three field trials, nests t#ere built in each case. In one out of the seven trials, workers with alate reproductdires built cnly an unfinished nest in the laboratory, lidrkors without brood, however, almaps failed to build nest, but if the brood was added later thay 'fchen constructed a nest. The results are shorn in Tteble 8.2. Host, abandonments Analysis of the data obtained shou that 62? of nests found deserted mbr® on nomal green leaves, 2Q% were on either dead or dying leaves and 10$ on both normal and ageing leaves (Table 8.1*). Also, in a period of 28 days, 90 neats or 18# of the total number of tjo University of Ghana http://ugspace.ug.edu.gh 131 Table 8*1 Height of neats above ground Tree Ira 2m 3m liia Total Tree Jlm 2m 3m ijd tlOh, 1 7 2 0 0 9 36 2 0 1 0 3 2 0 6 2 0 8 37 0 1 0 0 1 3 0 r*> 2 0 7 38 0 5 0 0 5 h 0 1 0 1 2 39 0 7 1 0 8 $ 0 2 0 0 2 hP 0 1 •1j 0 h 6 0 3 0 0 3 Ifl 2 3 0 0 5 7 0 h 2 0 6 U2 0 1 0 0 1 8 8 9 3 0 20 U3 0 9 0 0 9 9 3 1 2 0 6 hh 0 2 3 0 5 0 0 13 U 0 17 1*5 0 U 1 0 5 1 0 9 8 1 18 I46 1 2 0 0 3 2 0 h 1 0 5 hi 0 3 0 0 3 3 0 2 0 0 2 I18 0 9 0 0 9 it 0 2 0 0 2 W 2 ii 0 0 6 5 0 1 0 0 1 $ 0 9 1 0 10 siob2 51 2 6 0 0 6 52 2 3 1 0 6 6 0 5 0 0 5 53 0 1 2 0 3 7 8 9 !0 0 0 0 0 2 2 12 0 0 1 1 1 0 0 0 0 2 3 13 1 C5a 51* 55 56 3 3 0 1 1 1 0 1 0 0 0 0 h 5 1 flOb„ 57 1 h 2 0 73 .1 0 6 0 0 6 5859 0 0 3 2 1 0 0 0 ii 2>2 0 2 0 0 2 60 0 1 0 0 1 !3 9k 0 0 1 0 0 1 0 0 1 1 6162 63 0 0 6 6 1 7 0 0 0 0 1 0 6 2 130 H >5 1 0 1 0 2 6b 0 1 0 0 1 >6 1 ii 0 0 5 65 1 1 1 0 3 >7 0 2 0 0 2 66 0 1 0 0 1 28 0 1 0 0 1 67 0 2 0 0 2 29 0 6 0 0 6 68 0 1 0 0 1 30 3 U 2 0 9 69 0 1 0 0 1 31 1 1 0 0 2 — — — mm 32 33 1 0 2 1 1 0 0 0 h 1 Total 53 221 5* 3 328 31 0 0 1 0 1 35 3 2 0 0 B \ — -- - — University of Ghana http://ugspace.ug.edu.gh J 50 0 0 1 5 0 2 0 0 No. of nest 132 Height above ground in metres — k> \ u PREFERRED HEIGHT FOR SITING NESTS University of Ghana http://ugspace.ug.edu.gh Table 8.2 Stimulus for nest building 133 Treatment Results Workers only No nests Workers plus alates Ho nest Workers plus brood (fraa start) Nest built Workers plus brood (later) Lest built Workers plus eggs plus queen Nest built University of Ghana http://ugspace.ug.edu.gh 7 0 0 0 0 0 0 8 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 3 1 1 0 0 3 3 1 5 8 h 5 1 9 5 5 3 3 9 6 10 6 6 3 h 5 1 7 h 2 1 6 2 3 1 3 1 2 1 1 iji Table 8. Height of nests al Total 2 0 0 9 6 2 0 8 $ 2 0 7 1 0 1 2 2 0 0 2 3 0 0 3 h 2 0 6 9 3 0 20 1 2 0 6 13 k 0 17 9 8 1 18 h 1 0 5 2 0 0 2 2 0 0 2 1 0 0 1 5 0 0 5 2 0 0 2 2 1 0 3 12 1 0 13 0 1 0 1 6 0 0 6 2 0 0 2 1 0 0 1 0 1 0 1 0 1 0 2 h 0 0 5 2 0 0 2 1 0 0 1 6 0 0 6 a 2 0 9 1 0 0 2 2 1 0 h 1 0 0 1 0 1 0 1 2 0 0 £ ground Tree Vb 2m 3m 36 2 0 1 37 0 1 0 38 0 c: 0 39 0 7 1 IP 0 1 3 W 2 .3 0 >42 0 1 0 Ii3 0 9 0 Wi 0 2 3 1*5 0 it 1 hf> 1 2 0 hi 0 3 0 he 0 9 0 h9 2 h 0 0 9 1 51 2 6 0 52 2 3 1 53 0 1 2 C5a 51* 3 1 0 55 3 1 1 56 0 1 0 57 1 h 2 58 0 3 1 59 0 2 0 60 0 1 0 61 0 6 0 62 0 1 0 63 6 7 0 6k 0 1 0 65 1 1 1 66 0 1 0 67 0 O4 0 68 0 1 0 69 0 1 0 — — — Total 53 221 51 i ■ _ _ University of Ghana http://ugspace.ug.edu.gh 50 100150200 132 PREFERRED HEIGHT FOR SITING NESTS University of Ghana http://ugspace.ug.edu.gh Table 8.2 Stimulus for nest building 133 Treatment Results fcfcrkers only No nests tekars plus aletes So nest Workers plus brood (£rcga start) Nest built Markers plus brood (leter) Nest built Workers plus eggs plus queen Lest built University of Ghana http://ugspace.ug.edu.gh 13U Table 8.3 Causes of nest abandonment Plot INd of nests feest Ho. H10 1 55 - + 2 1 36 ♦ - 3 I 37 ♦ - h I 38 + ♦ 5 30 ♦ - 6 I 10 ♦ - 7 27 | 8 2h - 9 I 6 - 10 | 28 : + mm 11 ! 5 ♦ - 12 71 + + 13 73 - + 1li 82 15 81* ♦ - 16 86 ♦ - 17 <89 • 18 91 - 4* 19 95 - 4* 20 120 - + 21 121 - ♦ 22 115 ♦ «» 23 112 ♦ Condition of leaf Fresh Ageing Remarks A new nest nearby «■> New nest nearby Increase population In 39 Increase population in 29 New nest nearby •m Leaves forced apart New nest being built Only nest area brown Nest washed off by rain Nest washed off by rain New best nearby Oaly nest area brown Mly nest area brom Nest dropped. Hew nest nearby Qecqphylla around area University of Ghana http://ugspace.ug.edu.gh 135 Table 0.3 (continued) Plot No. of nests Neat No. Condition of leaf Fresh Ageing Remarks C5b 21* 31 *#■ - 25 1 - + Nest dropped 26 35 ♦ - Few workers present 27 11 ♦ - New nest nearby* 26 36 ♦ - Dropped a? 1*1* ♦ - New nest nearby 30 8 ♦ - - 31 1*7 + - •m 32 7 - - Missing 33 68 - ♦ Dr capped 31* 6 ♦ - - 35 19 ♦ - • 36 12 ♦ - • 37 11* + - - 38 15 - ■f New nest nearby 39 16 + - JUO 17 •f - 1(1 ?1 - Jlli 1*2 32 ♦ «» fc3 35 - ♦ Nest area only brown increase population in 36 1*1* 31 - * Nest area only brown 1*5 16 ♦ loaves forced apart University of Ghana http://ugspace.ug.edu.gh Table 8 .3 (continued) Plot Mo. of nests Hest No. Condition of leaf Fresh Ageing Remarks h6 28 L7 33 ♦ - New nest nasrby hB U8 * Increase population in 19 2*9 13 + - - 50 6 llest area only brown University of Ghana http://ugspace.ug.edu.gh Breakdown of deserted nesttf Table 8#i* Fresh leaves Ageiag leaves Both A and B (A) (B) No. of nests 31 11* 5 Percentage 62 28 10 Table 8.5 Further analysis of Table 8.3. Nests labelled Deserted nosis Neu nests Total Ho. 277 50 9 Percentage 18.0 3.2 University of Ghana http://ugspace.ug.edu.gh Plate 2.8a. Exposed, nest of II. aculeatua found in the field. University of Ghana http://ugspace.ug.edu.gh 1 39 Plate 8.8b. Typeo of Baodiriched neats of M. acuLeatus found in the field. ' University of Ghana http://ugspace.ug.edu.gh lltO University of Ghana http://ugspace.ug.edu.gh nests (277) were deserted, end within the same period 9 new nests faming 3.2$ of total number of neats were constructed as shewn in Table 8.5. Table 0.3 gives the raw data. Discussion The occurrence of silk in H* aculeatus nest does not sem to be normal. In fact silk **as found in only two nests out of the thousands of nests esmined both in the field and the laboratory© Cfct each of the two occasions the ants vjere disturbed. They were therefore farced by circumstances to seek shelter. Since Hacrqaischoides will* if disturbed, build nests anywhere using any available vegetable naterialo provided brood is present it is not surprising that in them two cases they incorporated a deserted spider web into the nest* Similar occurrences sight have misled Santschi (1909) and Forel 0916) to include silk in their description of the nest, ignorant of the fact that adult ants do not produce silk and Ifyriaecine ant larvae have no silk glands (^ heeler, 1922)* Usually building materials are deposited directly on the leaf surface and glued to it, as described tey Ledoux (1952). From observations sade it appears the glue caaes from the mouth, as the caster was never involved in the building process. Ifaschwits ct &L (1970) clataod that Lasius fuliginosus workers use hcney-dew and sugar solution as glue and not secretions frcui the loandibular glands. But since II. aculcatus is able to build nests in the laboratory where there uas no hcney-dew or sugar solution suggests that sane natural secretion *?as being used. Moreover the®*© no disceraiblo difference in the consistency of the nests built in thefield and laboratory . The secretion is possibly University of Ghana http://ugspace.ug.edu.gh derived tcan the glands. Ho soil particles or for that «u»»ti?r any earthen materials war© ewer found among the building materials* contrary to the findings of Ledoux (195G). Moreover the fact that no solid particles remained after burning the nest proves the absence of any particles. ledoux (1950) also that elongated pieces of leaf and moss are used in building nests. Leaves are certainly used bat they could seldom be recognised as tiho ants usually chav them and reduce them to pulp. Hose occurred very infrequently in nests; ahen it did occur it was growing mainly in the outer vail and not incorporated in the nest as Ledoux has claimed. The part $Layed ty the fungus mycelium found in the nests is not wall understood. The claim by some authors* for exseiple Elliott (1JH5) dealing with Lssius. that the fungal hyphae give stability to the nest rsall by binding the building particles together is not applicable to the case of HaercBsischoides since not all nests collected fraa the field and none constructed in the laboratory had fungal ramifications} yet their walls were as firm as those v&ich contain raycellaa. Wheeler’s (1922) suggestion that the high moisture content of the environment encourages fungal growth soetis reasonable. Moreover the ccjpeting substance* as well as honey-dew from the coccids found in aeet floor, are very likely to fam a very good substretun fcr fungel £rawth. In the field, nests were nearly always found constructed cr* leaves although an two occasions nests were found built on a stem over a group of young coccids (Plates 5.1 - 5«3). Thus the habit of construct­ ing neats under stools in the laboratory rather than on leaves may be d u e to a negative response to light. The building of a nest on top of a 1JU2 University of Ghana http://ugspace.ug.edu.gh metal plate in a dark room lends support to tide suggestion. Thus light may play an important part in the selection of nesting sites on a tree# Perhaps the preference of the ?-? metre-level is due, among other things to the low level of light intensity. Similarly the habit of taking shelter in crevices and holes can be explained in toms * of light avoidance. Although there is no qtaantitative evidence, field observations show that sandwiched nests are found more caaaoaly In open areas than in itesvily sliaded areas mthin a colony* It is also interesting that three out of the four neste built in the laboratory caa cocoa seedlings in the glass cage were sandwiched. Sms light seems to influence both the siting of nests and their type. Brood, it has been claimed, encourages foraging activity. The present work has established that brood not only encourages foraging activity but nest building as well. Thus, perhaps, the faster the queer* lays eggs end more larvae and pupae are produced, the more the workers are stimulated to build nests to enlarge the colony. It is very difficult to detera&ne the causes of nest abandonment. It may possibly be due to absence of brood, workers vacating their own nest© to join others tahieh have brood. This is supported 1 the fact that after Hie mating flight, a •regrouping1 of workers takes plaeej a few nests become highly populated rahil© large numbers are deserted. The populated nests almost invariably' contain brood. Thus it appears that after the nuptjttal flight, the few brood left in various nests are concentrated into a few nests followed by their nurses. Similar movments 12*3 University of Ghana http://ugspace.ug.edu.gh during the pre-flight period roa^/ account fear tlie abandonment of nests. The observation that 18# of nests were deserted while only 3•3% new ones wore constructed over the same period, probably, indicates that several of the inmates of the deserted nests noved into other already existir^ nests. University of Ghana http://ugspace.ug.edu.gh 1 us Section 9* Flight activity, coloaosr founding ffid colca^ mpsn&ixxu University of Ghana http://ugspace.ug.edu.gh 9. FLIGHT ACTIVITY, COLCNY FOUNDING AND CGLCHT EXPAHSICK Introduction Per a better understanding of the ecological and geographical distribution of ant species, a knowledge of mating flights and of the behaviour of fertilized females in founding colonies is essential* This is so because the mating flights constitute the only dispersal phase in the life of ant colonies and the behaviour of the mated females in selecting nesting sites will determine the distribution of col€Kd.es« Much work has been done on both aspects. Investigators on flight activities include Talbot (191x3, 1&8, 1956, 1957) and Talbot and Kennedy (19140)# Wheeler (1933) has also laid the foundation far the study of colony founding behaviour of fertilized females# Hich raork has been dene on colony founding behaviour of the tenperate species of the prtaltive subfamilies, I^ raecinae and Ponerinae. Hotable among these are works of liheeler (1933) and Haskins and Haskins (1950) an IferaLcia and Qiontmachus* These primitive ants have an Independent method of colony founding, by which the female forms a brood chamber which has an external opening penaitting her to forage for food while her eggs develop. Ihey therefore fora a link between the true social hymancptera and the solitary members of the Order. Colony founding behaviour of the Ifermicinae lias horaver been relatively neglected. It is intended to study the nuptial flight, colony founding ®^ P®nsioa in M. aculeatua a tropical member of the subfamily in this work. H14j6 University of Ghana http://ugspace.ug.edu.gh Materials and methods Mayngj^ ygiyfc* Ho light trap was ran personally. However data obtained by J. Major £raa one he ran at Kade frcm February 1971 to !fey I97I, thus coinciding with the period of this work, have been used with his perraisBion. Inci’ofeent colonises (i) Field? Several trees were gsssdned for incipient colonies (single m U nests). Es& incipient colony found was carefully lifted 13? with a pair of forceps and queen gently driven oat. The contents of the nest were examined and record aade of nosfoer of @g&sf larvae, ptipae end adults (callous)* A few incipient colonies were examined from time to time to investigate the frequency of egg lading. (ii) laboratory* Six mated fenisles (dcalates) collected from around li$it in a house at M o were put into match bases and petri dishes and provided with food (insects and sugar solution) end water. One half of the petri dish m s covered. They were supplied with cotton wool. Seme alate males ant females were also placed together in two petri dishes to observe mating behaviour. 3h the laboratory ten alate females, seven alate males end eight workers were placed in a glass breeding cage (Fig. 10.1) and provided with food and water. Results and observations Flight activity* 3he results shown in Fig. 9.1 are based on light trap catches and indicate that flight activity started in February and with but few fluctuations remained low until April, when there was a sudden rloo during the third week, reaching a peek the following week. This then followed by a sharp drop in the first week of !%-. 1 hi University of Ghana http://ugspace.ug.edu.gh To ta l np of al at es 7 .0 0 0 6.000 5 .0 0 0 4 .0 0 0 3 .0 0 0 2.000 1,000 o n g *9*11 K U ®ht activity Of E. aealefttus alatea. Sad©. U.7. trap. 118 4 5 Weeks 6 7 f rom 20t'h 9 10 February University of Ghana http://ugspace.ug.edu.gh Incipient colaLtar.t Table 9.1 svaaaarisea the results of the analysxs of the nest contents of the 2h incipient colonies examined. The results indicate that more than one mated feraale may alight and found a coloiy on one tree. Four were found on one tree, three on another and three again on a thirdr It is evident that the initial batch of eggs is vei*y mall, varying from two to six* This ia inferred from the number of larva© (both advanced and young) obtained £raa each colors * Even the next batch of eggs Boom to be quite snail and not xaore than 2fa* Field observations tend to show that the queen starts laying as soon as the foundation of the nest has been laid; and that egg lay in g and nest eonstructico proceed simultaneously* The evidence for this is the mated female in nest 18, Table 9*1 * This feanale vas found when she vaa putting building materials on a leaf to start a nest and on examination one egg ms found in thenest. On one occasion an incccplete nest was found with four eggs in it* At the end of a week tne nuaber of eggs had increased to thirteen, and the nest \am nearly ecsapleted* Hone of the mated females put into match boxes and petri dishes lived long enough to lay eggs* They all died within four days without attempting to construct nests* Behaviour of slates s The alate pairs in petri dishes did not mate* The females were found to be snare active than the males* Usually a female nay prod a male with her antennae, and at times climb on top of the stale# At times several females sight croud over a roale* No mating took place, as ncei© of the females shed hear wings* Similar observations were made in the laboratory breeding cages* Six out of the seven males t*ere found 111? University of Ghana http://ugspace.ug.edu.gh Table 9.1 Colony founding Incipient colonies (C5a - 25th April 1971) Tree Nest No* of JaggB No. of Larvaa Jo. of Pwpae No. of Callous Remarks 1 1 1 21 • • - 1 2 - m t - a* - 2 3 1 9 - - - 2 1* - mm mm - - 3 5 6 mm 1 9 6 h big ° 2 smell “ - 5 7 1 6 L 3 bigU 1 ami»n mm - 6 8 1 12 31 l i t 22 Q a same tree with fairly old Incipient colony (1 cost)* 7 9 1 15 2 - - 8 10 1 20 2 mm 9 9 11 12 1 1 2 17 mm a h big ° 2 snail mm mm - Nest quite old. 9 13 1 17 2 mm mm 9 11* Not Pound 10 3 mm - 10 15 1 20 1* mm - 11 16 1 16 1* • mm 11 17 1 6 6 k 2 small li h 11 18 1 1 • - - Nest under construction. 12 19 1 22 l i - • 12 20 • - - «• 12 21 1 3 • - OB 13 22 1 13 - • «■» 11* 23 1 8 - tm 15 21* 1 21* 3 - - 15 21* 20 251 77 18 26 University of Ghana http://ugspace.ug.edu.gh dead with pert of their gaator eatenj a similar occurrence has been found in nests. She ten feraaLes were always found surrounding the rwaJning melee Biey were kept for sixty-five days* but th^ y did not sate, the females retaining their wtags throughout. Cfae, probably rnted female s from another laboratory cuLture in a glass cage* was however found on a cocoa seedling trying to drop her wings. Thie she did by vigorous flexing of the thorax and constantly brushing the wings with the hind legs. Unfortunately she did not succeed in dropping her wings before she fell into a dish of water. She was rescued fraa drowning but did not continue the operation. Discussion Flight activity starts In February at the end o f the dry, massy season and continues to rise steadily to reach a peak in April-Bay Just before the heavy rains begin in June. This finding is in ccnplste agreement with that of Billes (1 9h6) and also with ay earlier results on seasonal patterns in section Ij. The period of swarming and thus of cdLaay founding coincides mainly with the first wet, sunny, and partly with the first wet-dull seasons of Gibbs and Lesion's classification. These period® are characterised ter low teng>eratures, moderately low light intensities and high humidities. These weather conditims say perhaps be Meal for swarming. This sort of weather is possibly essential for the first batch of eggs, if they are going to escape desiccations since they are deposited in unfinished nests and thus exposed to the atmosphere fop m long aa th© nests raaain unfinished. 1# University of Ghana http://ugspace.ug.edu.gh 1$2 *hai ©mironmentsl factor or factors serve as 8 rele&ser for nest building is not Imam, but it is likely that the presence of the first batch of eggs nay serve as a sttouXua for its completion, since the presence of eggs (queen was present) has been shown to stimulate workers to build a nest (Section 6). Tii© lif© of a cocoa leaf has been shorn tqt McMLvie (1962) to be about eight or nine laonths* Thus nests constructed csi the new leaves formed after the leaf fall in the dry seascea are likely to be retained for that Icng* This then make the colony more stable than if the leaves had a short retention, period* The large area of leaves during the rains and after, any also help the alighting of f©sales and establishment of nests (Billes, 191j6)* Incipient colonies* It is unfortunate that the ants were not seen Bating* Marikovaky (1961) lias reported that females of Formica rufa bit off the abdenen of males nating with the©* thus perhaps the sales fourd dead with their gaster bitten off ussy have attempted to mte with females* The finding that several mated females can establish incipient colonies on one tree, and several incipient colonies can be found on several contiguous trees is interesting* It has been established earlier that a colony of several nests noraally has only one queen* Thus some elimination of Incipient colonies must take place* Inestimably one colony adgfet take over tho workers of another, while killing one of the queens, or kill both workers mid queen* Perhaps the fecundity of one queen, as well as the numerical strength coupled with the ferocity of the University of Ghana http://ugspace.ug.edu.gh ®arly workers might toe contributory to success# iifcil© doubtless many young nests fall victims to predation by other insects or are last as a result of the dropping of leaves. It will I® useful at ♦*>*« stage to construct a continuous story of colony founding from the infomation available. There is a mass flight of males and females frcta the nests, mostly at night. It is not yet known tðsr swsraing occurs from all nests of a colony at the same time or whether the slates in different nests of a colony may swam at different times or even whether all alatea in a nest leave cn the sme night. Females are fertilized, after which they shed their wings by brushing them with the feet and flexing the thorax. Each dealate female constructs a small, done-shaped, open nest on the underside of a leaf as shown in Plate 9.1 • Several females may nest cn one tree or on contiguous trees. Egg laying starts as soon as the female starts putting down material far a nest and raey continue until the nest is finished. Initially a small number of eggs is laid, usual!; four. These are tended and hatched aid the young fed ty the fenale„ The female is very active during this period. By the time the first batch of workers emerge, more eggs may have hatched and they, together with the queen, nurse the young larvae. Fran field observations it seems as if the queen remains active for a long tine, foraging and nursing brood, but eventually with an increasing worker population the queen as mines the role of egg-laying only. As the nunber of workers increases and more brood are produced, the original incipient colony is expanded and new nests are built away from 153 University of Ghana http://ugspace.ug.edu.gh 15U Plate 9*1 * A young nest built bjy a quean* University of Ghana http://ugspace.ug.edu.gh 155 the mother nost. Eggs, lervac and pupae are continually removed frcm the queen's mat into the satellite ones. The s ta r / aa outlined here agrees in essential details with Wheeler1® (cited by Ilaskias & Haskins, 1950) hypothesis cxaxcerning the 1 normal1 method of colony found In the higher ants, as well as in Kfroecia and the ponerines including (Mon tcraachus» University of Ghana http://ugspace.ug.edu.gh 156 Section 10. Setae aspects of tha life history University of Ghana http://ugspace.ug.edu.gh 10. SOME ASPECTS OF THE LIFE HISTORY Introduction It appears from the general accounts of the life history of ants given by Uheeler (1910), For el (1928) and Inns (1925) that ants exhibit the usual hdoraetabaLous development of egg-larva-pupa-adult. Bat usually interposed between the larval and pupal stages is a saoi- ar prepupa. a© ainplost description of this will be "pupa in larval skin”. Few accounts, however, exist cn the life history of specific ants.” The most recent one by Ctfer (1970) on Poivrachla simplex shows that the egg, larval, and pupal stages lasted 15* 2h and 21 da^ s respectively. The life history of Haarcnlscholdes aculeatus has not been worked out. Uheelar (1922) has described the adult worker, male and female as have the larvae by Vlieelar and fchesler (1955). Hie eggs and pipae have not been described. An account of the few observations made on life history as well as adult and brood behaviour are given below. Materials and methods Breeding canest The Fiolde model was used with slight modifications. Walls of about 5m height and with a basal width of about 15cia but tapering upwards were raised round the four edges of a glass slab, using either plasticine or Plaster of Boris. The former material was later rejected due to its strong aaell. A partition wall was built to divide the cage into two unequal, interccBmunicating chambers* The smaller chamber was to be usod as the *nest% while the bigger one 157 University of Ghana http://ugspace.ug.edu.gh represented the 'outside world* where the ants could forage or dusap refuse, a piece of cotton wool soaked with water xjas placed inside the nest to raise the humidity. Loose pieces of cotton wool were also added to serve as building materiel for the ante. Food in the farm of insects end sugar solution, as well as water were provided In the bigger chamber In wm**3.i plastic cups and petri dishes. A second glass slab was then placed on top, making sure that all escape routes were blocked with cotton wool# A thick bock placed on top of the nest-side served to darken the mailer chamber (Fig. 10.1 )• Transfer of ants into cages Initially the ants were immobilized fey placing thm briefly in a chloroform or ethyl acetate chamber. This method was abandoned later as it was not producing the desired result due to timing difficulties connected with the use of the anaesthesia. Carbon dioxide was also used for sccietiiae, The method met frequently used was siraply to open the nest without Immobilizing the ants in any way. The contents were quickly shaken into the nest and covered. The remaining ones were trapped In cotton wool fibres and placed into the cage. This was done fey dabbing a piece of cotton on the ants. Carbon dioxide was however used whenever the nest contents needed sorting. Observation chambers t Either eggs, larvae or pupae were put Into a email petri dish together with workers to attend them. Wet or dry blotting paper or cotton wool was placed on the floor of the dish depending on the type of brood. Eggs and larvae received the wet treatment, whilst pupae were given dry tftotting paper or cotton wool. The petri dish was 1$8 University of Ghana http://ugspace.ug.edu.gh 159 o Ll) CD < o z Q Ll)Ll) QC CO cr II University of Ghana http://ugspace.ug.edu.gh 160 placed under the dissecting microscope and observation could be carried out for long periods of tine, especially when the ante 1fere behaving normally after the initial period of agitation. Results and observations VJhen emptied into the breeding cages the workers quickly collected the brood into the nest# Sggs, larvae and pupae were all mixed up. Eggs and larvae were however more numerous in the wettest part of the nest near the cotton wool, with pyp&e more in the periphery* The queen was also crowded over by workers *&o gradually led her into the nest# The queens The queen was never seen casing out to forage. She sits in one place with numerous workers round her, constantly groaning her especially the gaster. Presumably she was also fed but this was not observed. QT all the queens kept only cue laid some few eggs. The eggs are elongate, elliptical and transparent t&ito measuring about 0.2m in length (Fig.10.2a). They are very sticky and usually several egg® adhere together forming big egg-maases# The eggs laid in the laboratory did not hatch before they disappeared froa the nest# It was therefore impossible to determine how long it took theeggs to hatch. Very small larvae were found among theegg-oasses collected fraa the field. These were taken to be first instar larvae. The eggs were also licked by the workers. University of Ghana http://ugspace.ug.edu.gh Fig 10- 2 161 Fl g# l0 #2 t St ag es of M» fl ci de at us t University of Ghana http://ugspace.ug.edu.gh Larvae ifonaholotrvt The first instars are extremely small only slightly longer than theeggs, having an average length of about 0.1; millimeters. They also are very sticky and attach to any object with which they come into contact. Since they are not immediately separated out on hatching, they are usually found intermingled with the eggs. They can hoirevar be distinguished frcm the eggs by their slightly more elongated and segmented body as well as the dark-brown or brown midgut (meconium) which is conspicuous in the abdominal region and discernible through the transparent cuticle* liheeler and Wheeler (1955) have adequately described the larvae and provided diagrams. Fig.10.2b is a drawing of a larva. Caret The larvae are constantly licked and moved from one place to another. In transporting a Larva under normal circumstances, a worker stands astride it with the head of the larva pointing posteriorly* The worker then seizes the mld-scctlon of the larva with her mandibles and carries it away. The head-end of the larva 3a thus protected under the worker’s body. In an emergency the larva is however, Just grabbed and carried assy without any special orientation . Feeding a Wien feeding, a larva is usually placed on its back with the mouth pointing up. A worker then coses and places its mouth against that of the larva. What fallows is a vigorous 'swallowing* raovsaent which is Indicated by the bonding of the larval head ventrslly. The worker then moves away and another one canes to repeat the process. Feeding is period ic, as workers retire now and then especially outside foraging 162 University of Ghana http://ugspace.ug.edu.gh 163 hours to carry out self and mutual groaning# Whenever a larva is disturbed, such as fcy touching it, it gives out a big drop of dear liquid from tho anus. The larva passes through thedifferent stages of development, which are not easily distinguishable, into a pupa# The prepupal stage can however be distinguished. The borders of the sccdtes, becane Inconspicuous due to the distension of the body. The posterior end becomes transluscent, perhaps narking off the gaster. As development proceeds, the legs, and head appendages can be easily seen, though folded and closely applied to the body# The body shrinks at the end of tlis stage# Moulting takes place# Two prepupae were observed moulting into pupae# Ihe workers consumed the exuviae# Pupae Morphology! The nandy emerged pupa which already looks exactly like the adult ant in fam is completely unpjgnapted# Even the mouth parts are not tamed# Scnetiraes the gaster appears yellowish# The legs and antennae are folded and applied closely to the body as shown in Fig #10# 2c. Pupal changes* The eyes intially colourless start developing colour 2li-hours later. Snail reddish spots appear dotted cn the caapound eyes; these spots spread out and intensify to deep red, broim and finally black. The claws of tho legs are unexpanded at the initial stages, and the tarsus, though demarcated is not well defined. The antennae lode very much like legs. The spines of the thorax are present right frcm eclosicn, though not tanned* llesnwhile colour lias started fording esi University of Ghana http://ugspace.ug.edu.gh the re st of the body, which passes through shades o f brown to black. The pupal in star takes about 23 days a t 27°C, fo r two larvae which pupated between 29th and 30th November emerged cm the 21st and 22nd December respectively . Caret Pupae are frequently groomed. They are a lso sh ifted about lik e the larva©. Adults The newly ejaarged edulta are usually p aler than the older ones and are referred to as callou s. The workers o f incip ient colonies are sm aller than those o f matured colonies and are thus called a&nies. In Hacromischoides aculeatus nanics have an average body length of 2*?m and h@adid.dtli o f O.&ia whilst adult workers from matured co lo sies generally measure 3 .tan long and have headiddth o f O .trn. Morphology s The adult has been described by Vfoeeler (1922). He described workers as im ll as ©ales and fem ales. F ig .10.2d ahosrs a draining of the adult worker. Behaviours The w rkers stop work, nmi and then, and engage in m tual grooming. One stands s t i l l while the other on© lic k s i t , anteriorly tn m the abdomen to the mouth p a rts . At times the one being licked leans sidewy© to ensure proper groorolog. Che marker may lic k or nib&Le one or two of i t s nest mate®, xihile i t s e l f being sd jailarly treated tzy another. Self-grooming: takes place more ft^quently than mstual grooming. In self-grooming a worker mainly dean s the leg s and antennae, and a t times bends the gaater forward through the lege to reach th e mouth. 161* University of Ghana http://ugspace.ug.edu.gh 165 The workers also carry their nestmates. Discussion It is most probable that the concentration of eggs and larvae In the moist area is to prevent desiccation and ease larval ecdysis. Pupa© can the other hand are more resistant to desiccation, and as Sudd (1^ 67) states* they are usually separated out into a warm dry part of the neat. This type of domestic arrangement vaa# however, not observed in the nests collected from the field* Perhaps it existed but was destroyed through mishandling of the nest or maybe the humidity in the natural nest is nore unifoasa than in the artificial nests. Queens are usually nursed, fed and in certain cases even •delivered1 (Sudd, 1967)* It was not surprising therefore that the workers were seen crowding the queen and tending her, and since the queen waa never seen coming out to forage, it is most probable that she is fed. Her failure to lay eggs in the laboratory may be due to mechanical disturbance in handling or to the chemicals used to immobilize the ants. As reported the only eggs laid in the laboratory disappeared! presumably they were eaten by the workers and or the queen since no larvae were found® This is not uncommon in ants. Thus according to Preeland (1958) Kynnecia forceps queens never eat anything but eggs! Brian 0953) reasonably points out that queens would be unable to survive an their ora eggs only, and that in this circumstance the egg pile must be angmented by large quantities of sterile workers1 eggs® Ko workers1 eggs (usually smaller than queen’s eggs) were noticed among M. a.cu^ atus ©gg masses. This does not however rule out the University of Ghana http://ugspace.ug.edu.gh possibility of workers laying eggs* Workers seen with distended gasters had nematode worms in thorn* not eggs* As the workers lick the eggs continuously, they leave a salivary coating on them which makes them sticlgr and possibly reduces water loss as veil* Neglect of the eggs by the workers results in the loss of the coatingj when this dries out, the eggs no longer adhere loss of coherence and consequently the egg-piles collapse The licking is also said to protect both eggs and larvae against fungal attack (Sudd, 1967) • wheeler and Tdheelerfs description of the larvae agrees well with my own observations. The first instar larvae are not segregated frees the egg pile. They feed an fe eggs, a s reported by Brian (1953) for Hramica ruEtea end Weir (1959) for I-tmnica rubra microgyna. They are only removed after their first moult and are then fed by the workers. The subsequent larval instars are fed by the workers. It was not possible to determine on what the larvae were fed, whether malaxated insect meat or regurgitated food. Perhaps the sclerotised mouth-parts of larvae enable them to feed on solid or hard food. The hairs on the body of the larvae serve various functions. Qfer (1970) described five types in Polyradiis stoplex=*but could not assign functions to each. Wheeler and Mieeler (1955) found three types of hairs in I^ cromischoides aculeatusa which they described as (1) few, simple, minute, (0.001 -0.027»sa), (2) with denticulate tip, 0*027-0.072m long, the most abundant typef (3) anchor tipped, about 0.22pm long* The anchor-tipped type may help the attachment of one larva to another, thus 166 University of Ghana http://ugspace.ug.edu.gh 167 keeping them In larval piles. This aggregation of larvae into piles is possibly advantageous because when danger threatens a large number of larvae would be carried all at once to safety. The denticulate- tipped hairs, being the most numerous may perhaps fora a protective covering on the body of the larva* The few, simple minute type ° could be sensory in function, since ants are known to possess hair sensillae. (Sudd, 1y6?). The constant removal of brood from one part of the nest to another may serve to place each larva in the most suitable conditions obtaining in the nest; this also nay be the reason why brood is moved from one nest into mother in the wild. The 23 days duration obtained for the pupa is not vastly different from that reported by Gfer (1970) for P. slrolex. The absence of pigmentation in the pupa is a caramon occurrence in ants, as reported by Wheeler (1910). Fore! (1928) as well as other xaodera authors such as Sudd (1967). Full colour is developed some day's after emerging into adult. Perhaps callows perform intranidal duties such as nursing the larvae and licking the eggs. They start foraging only «fe©n they are fully pigmented. This suggestion is lent weight by Fr©eland1 s (195(5) report that nanitlc workers of ffiyrmecia forceps were normally seen near the brood and refrained from foraging while there were larger workers which could do so. Ifte mutual raid self grooming noted in the adults prevent fungal attack and also help dislodge mite parasites. University of Ghana http://ugspace.ug.edu.gh S@cy.OOi 11 • 163 Gesieral Discussion University of Ghana http://ugspace.ug.edu.gh 11 # gejjeraIi biscussicij H a c r a t s l s c h o l d e a aculeatus is widely distributed in southern Ghana aa Fig.3 .1 indicates. The ant has became closely associated with cocoa farms. It also nests on a wide variety of wild trees including Colcaaia acandens (Araceae), Ficus exasperatus. Ficus agperlfclia (Maraceae) Milletia sp. (Papilionaceae) Riialodlscua unlingatus. Rauwolfia vcmiteria and Sdamaa verbascifoliura (Solanaceae). It is attracted to cocoa, perhaps due to its broad leaves which are retained for a fairly long time. Moreover the typical structure of a cocoa farra with top canopy or upper storey and cocoa canopy make the place dark and humid and this way offer special attraction to the ant which is largely photophobic. These conditions are not obtainable in i/iany other crop farms and this may account for the absence of the ant in those places* Their sparse occmrence in cocoa farms in the Akwapim areas may be due to the dearth of shade trees in these tam s. Hie ant is largely crepuscular and nocturnal. Since normally temperatures are low and relative humidities high in the night the ant, especially the eggs and larvae, will escape desiccation as brood circulation taker place in the night. It may be safe, to sense extent, fraa predation by its diurnal predators, Qecopfaylla longinoda and riabuva spp. Furtheraore, its habit of nesting on the under side of or between leaves may increase its chances of escaping predatAcn by a predator like Sfebura that hunts by sight, since the nests are difficult to detect. The nests are also thus sheltered frcra rain and perhaps to some extent from light. Sandwiched nests are, however, not so safe since either strong wind or a 169 University of Ghana http://ugspace.ug.edu.gh n o rainstorm can easily tear th«a open* Since however, the habit of constructing such nests persists, they must have seem other advantage over the simpler pattern. Fungal raercoLia found in M. aeulmtom nests do not seem to play any part in holding the nest materials together, as nests with or without the Rtyrceli& have the same consistency and stability. It is more likely that a secretion from the mandibular glands serves as glue for holding idle building materials together. Hie claim by early authors of silk in the nest is not completely wrong, as the ants are capable of incorporating a deserted spiderfs web into nests. Since however, the larvae are incapable of producing silk, any silk that appears in the nest is of foreign source. The nests ere unequally and sparsely distributed on the trees; thus a colony of seventy-four nests was found to occupy as many as seventeen trees. It will be interesting to investigate what factors detenaine the number of nests to be constructed on a tree. This colony pattern is an advantage to cocoa cultivation since several trees may be protected by a single colony against capsid attack and perhaps indirectly against swollen shoot, M» aculeatus being antagonistic to the SSV - vector attendants, except Creraatogaster clariventris* An overall average of 363 workers per nest was obtained over the period of sampling . Thus a colony of 11+0 nests may contain as many as 50,000 workers, and even a colony of two nests may comprise more than seven hundred workers. If really the ants are active predators, a tree carrying only two or three nests will be University of Ghana http://ugspace.ug.edu.gh kept free £rm oapaM attack. The dispersal of cocoa capsids takes place during the maximum population peek between Hovaaber and December* this coincides with the r^ r-Umm population peak of Kiacroiai3clioides workers. It is voiry liirafly *tbat those c^ jsids which on treos contfliniDg Hacrcsaischoides will not survive either to cause damage or to reproduce. Moreover, since during this period large sexual larvae are present end the presence of brood stimulates Mgh activity in the workers, the predation rate will b© high. Efy this means M. aculeatus inay conserve the health of good cocoa. Unfortunately not much can be learnt frcxi the results of the seasonal population counts, since sailings had to be made in different colonies, and these 3aay or may not be in phase. The results however, substantiate the earlier observation of Billea (t 9h6) on the anergeuee of sexuals between March end Hay, and also agree with the results obtained with the Xi^ ht trap. Since Kade where the light trap was run is about IS miles tram. Taf o the production of sexuals is probably in phase over a wide area. Tt& method of colony founding after the flight period individu&l queens is worth considering. The earlier account on this subject In section nine showed that several females could start colonies on one tree. Eventually however, only can& colony containing usually caie queen rsaaaizaa In the area. What happens to the other incipient colonies is not known. Ihe results of the radioisotope experiment showed however, that a small colony of a few nests night be almost surrounded by a different 171 University of Ghana http://ugspace.ug.edu.gh 172 ®Bd usually large colony. In C5b and H10, a few nests never became radioactive. The inmates of these neats were attacked by those from the big colony whm the twtv^ought together. This showed that those nests were alien to the larger colony. Fro® these observations it can be concluded that a certain fora of elimination was going cn. Knowledge that each mated female is capable of founding a colony independently can be advantageously utilised to establish the ants over a wide area in a biological control scheaae. If several incipient colonies are found an one tree, or on several contiguous trees, all but one or two, can be removed and re-established on other trees situated wide apart. % this means many more trees might be made to aaarboiir* the ants, thus conserving their health. This can, however, be done only after removing other species of dominant ants from the area either mechanically or chemically. Macromisoholaes aculeatus colonies are not ccopletely mcnogynous, since in a few nests more than one queen has been found, although generally not exceeding two or three. This is in accord with Wilson's (1959) observation that monogyny was the normal pattern in the ant species he studied in the tropical rain forests in New Guinea, but that, in exceptional cases where there were multiple queens, their number did not exceed two or three. The early batches of workers, or nanies as they are often on account of their smallness, take over the d u t y of nursing, foraging, building and fieience frcm the foundress queen, who concentrates her energies University of Ghana http://ugspace.ug.edu.gh on egg production* Tho seasonal population studies have sfcowa that production of sexuals is seasonal, starting tram December to about June* There are two main hypotheses to explain sexual production* The oogenic hypothesis claims that sexuals are produced from special eggsj whilst the trophogeeic hypothesis holds that different types of food fed to the sane kind of larvae cause them to develop into workers or sescuals. As the number of workers increased the quern is stimulated to 1ay more eggs (Bahian & Brian, 1951 )• The vrorkers are in turn stimulated by the increasing nunber of eggs, larvae, and pupae- to build new nests, and expand existing ones* *016 incipient colony nest, the one built by a queen usually measures 1 .5cms across and 1 .Ocas high, whilst some nests built by workers are very big measuring about 17*5 x 11.0 x 1 .Sea in sandwiched nests, and 19*2* x 7*6 x 2.5cm in exposed nests* The fact that a cocoa loaf has a life of about eight to nine aonths indicates that a nest could last for that long* It does not, however, mem that the colony collapses with the fall of the leaves* Hew nests are continually built while old ones are deserted# The reason for nest abandonment is not yet knetm. (be possibility arises fraa the fact that a cork layer foms at the junction of the leaf petiole and the stem, and this, long before visible signs of leaf fall appear, cut out TOter supply to the leaf* As a result transpiration is reduced and the humidity in or around nest may fall. Another possible result is that the lecanilds inside the nest nay not be able to obtain enough water to produce hon^ r-dew. Such factors could be signals for 173 University of Ghana http://ugspace.ug.edu.gh neat desertion and may explain why large number of nests on apparently fresh leaves are deserted. Depletion of food resources is not likely to be responsible since the new nest is usually built very near the abaxadonned one. Macromischoides aculeatus has been found to be both an active predator and a carrion feeder. It has been explained elsewhere that the fact that few capsid remains were found in their nests did not imply that they were poor predators of capsids. They «ay feed on the early instars of the cocoa capsids without leaving any residue, since these are soft- bodied. The observations of Nicol et al (191*7), Marchart and Lestonfs (1966) radioisotope tagging experiment, Gollingwood^ (19 71) statement, and iqy own recent observations, all establish the fact that M. aculeatus is a predator of cocoa capsids. The degree of predation has, however, not been assessed. Finnegan's (1969) mechanical devise for assessing predation by ants on insects might be used to determine the effectiveness of M. aculeatus as a predator of cocoa capsids. To supplement their protein diet with carbohydrates the ants attend coccids, aphids and lycaenid caterpillars for honey dew, and visit extra floral nectaries. Fortunately the coccids attended are not vectors of swollen shoot viruses and it is generally believed that aphids are not harmful to cocoas There is a form of mutualism between the ants and the lycaenid caterpillars. The latter are protected while the fcorner get seme sugar substance in return for the protection. Hie caterpillars are classified as a minor pest of cocoa in Ghana. 171^ University of Ghana http://ugspace.ug.edu.gh It seems strange that the ants should obtain their mineral salts £r I9hh~h6* 26-27• Anon, 19U8. Biological control, Rep. II. Afr. Res. Inst. Tafo» I9k6~u7* 39-bP. Ayre, G.L. 1?S6. Some meteorological factors affecting the foraging of Fonalca aubnltens Creighton (Hymenop .Fomicidae) 1C1 Ihsectes soc. 5*li.i7-157 • Billes, D. J. 191|6. Macrqaischoides investigations, U. Afr. Cocoa Res. Inst. TaTo, fcth Quart.Rep. 7-6 (xn^ ublished) Brian, M.V. 1953* Brood-rearing in relation to worker number in the ant Itoraica rubra L.« Physiol. Zeal. 261 355-36fi. Brian, H.V. 1955* Food collection by a Scottish ant cccMunity, J. iiniau ScaL. 2l*s 336-351. Brian, M.V. 1957# The growth and development of colonies of the ant ifernioa. Insects soc. lit 177-1 90. Brian, M.V. and Brian, AJD. 1951. Insolation and ant population in the west of Scotland, Trans. R. ent. Soc. Legid. 102* 303-330< Bruce, V.G. and Pittendsigh, C.S. 1957* Endogenous rhythms in Insects and roica*oorganifflas, Aiaer. Nat. 91 s 170-195« Carth^ r, J.D. 1951* The orientation of two allied species of British^ nt II. Odour trail laying and following in Acanthoesyops (Lasius) fuliidnogua. Behaviour. 3s 30k-3*l 8. Chsuvin, R., Court ois, G. and Leccaate, J. 1961. Sur la transmission. d9 isotopes radio-actifs entre deux fourrnilieres d'especes differentes (Fonaica rufa et F. polyctena). Insects soo. 8* 99-107. Church, R.J.H. 1957. Uest Africa. Longmans, London. Cloudesley-lkGdpson, J.L. 1953* Studies in diurnal rhyfchas. HI. ffcoto- periodim In the cockroach PerJUxLaneta arasricana (L.). jyon. Hag. nat. Hist.. London b9t 233-23!?. Collingwood, C.A. 1971. Predators, Interim Rep. International Qap^d Research Team. 196^9?Y,7 (unpublished). Collingwood, C.A. and King, A.B.S. 1970. Ant-Capsid-Canopy relations, Rgp_. Cocoa Res. Inst. Tafo. 1966-69* 56-62. University of Ghana http://ugspace.ug.edu.gh 182 ^can-Weatherle^, A.H. 1953. Some aspects of the biology of the mound ant, Irdbdkaaymex detectus (Smith), Auat. J. Zool. 18 178-192. ELllott, J.S.B. 1915. Fungus in nests of ants# Trans# Br» rayeol. Soc#. 5t 138-1*2# Finnegan, ft#J# 1969# Assessing predation by ants on insects, Insectes Soc# 26# 61*66# For el, A# 1923# The social world of the ants# Put nan* s, London and New Z W t ~ Freeland, J# 1958# Biological and social patterns in the Australian Builds ants of the genus Hynaeda species, Augfc# J# 2oqIa 61 1-18# Gibbs, D# 1969# Qecophylla interaction with other ants, Hep# Cocoa Bos# 1967-68, 68-69. Gibbs, D#G#, 4 Leston, D# 1970# Insect phenology in a forest cocoa-fam locality in Kest Africa, J# appl. 2cal# h 519-5^ 8# Gibbs, D#G#, Pickett, A.D# end Leston, B. 1968# Seasonal population changes in cocoa capsids (Bemiptera, Mtridae) in Ghana, Bull, ent. Res# 56* 279-293. Haskins, C.P. and Haskins, E.F. 1950. Kates on the biology and social behaviour of the archaiae ponerine ants of the genera Hymecia and SSSBBB2BSiS» Ann, ent. Soc. Mter. h3t it61-li9l. Hayashida, K. 1960# Studies on the ecological distribution of ants in Sapporo and its vicinity, Insectes soc# 7# 126-162# Headley, A#E# 191*3. Population studies of two species of ants, Leptothcxrax Icngigroinoaus (Eoger) and Lcpto thorax curvigpinosus Utor). Am#"ent. Soc# &aer. 361 7U-7^3# te, A.D# 1925. A general textbook of entcndLogy. Methuen 1 London 720-726. Keay, R.W.J. (Ed) 1959. Vegetation saap of Africa South of the Tropic of Cancer, O.U.P#Loa*ion. Lange, E# 1960# Food-traffic between nests, Z# Tiarpgychol. 17s 389-1*01. University of Ghana http://ugspace.ug.edu.gh Ledoux A* 1958. La construction du Mid chez quelques Fourmis arboriccxLes de France et dfAfrique tropicale, Proc. 10th int# Congr> ^tcnd> Montreal 2i 521-8# Leioly, J. 1851# Flora and fauna witliin living animals* Smithson. Contrlb. Knowledge 5s 6# Leston, D. 1968. Diurnal activity of Decophylla looglnoda. Rep. Cocoa Res. lost* Tafo. 1965>-66s 62-72. Leston, D. (Personal connmmication) Leston D. 1971* Ants, capsids and swollen shoot in Ghana« interactions and the implications for pest control, Proc. 3rd Int. Cocoa Res. Conf.. Accra (in press). Lodos, H. 1968. Potential major pests Bathycoelia thalaasina (H.S.), Rep. Cococa Res. Inst. Tafo. 19c6-67s 1*1-k2. Louis, D. (Personal coesmmlcation). Majer, J.D. (Personal ccraaunication). Marchart, H. & Leston, D. 1969# Predators of Dtatantlella theobroma. Rep. Cocoa Res. Inst. Tafo 1967-6oj o3~oU. Marchart, H., & Leston, D. 1968. Radioisotope tagging experiments, Rep. Cocoa Res. Inst. Tafo. 1965-66| 52-51*. Marikovsky, P. I. 1961. Material on sexual biology of the ant Formica rufa L., Insectes age. 8t 23-30. Maschwits, U and K&Lldobler, B. 1970. Der Kartonnestbau bei Lasius f^ Li^ lnosus Latr. (Hym.Fariaicidae), Z. Vergl. Physiol. McKelvie, A.D. 1962. Cocoas B. Physiology, in Agriculture and Land Use in Ghana (Ed. by J.B. Wills), 256-&). O.U.P. London. Nicol, J. I9I4I4. Bionomics of the cacao capsidss Predators, W.Afr. Cacao Res, last. Tafo. 1st Quart. Rep. 6-7 (unpublished). Qfer, «*• 1970. Pol^ rachis sigftlex the waver ant of Israel, Ihsectes soc. Rocra, P. Personal communication University of Ghana http://ugspace.ug.edu.gh 18U Schnelrla T.C. 19hh» Tiia reproductive functions of the army ant queen * as pace-makers of the group behaviour pattern* J.N.Y. ent. Soc. 52t 153-192. Schneirla, T.C. 1957. Comparison of species and genera In the ant sub­ family Darylinae with respect to functional pattern, Ihaectes Soc. lit 259-298. Smith, R.F. 1966. Recent developments In Integrated control, Proc. Lth Brit. Insecticide. Fungicide Goaf. 2* • Strickland, A.H. 1951 • Oie entomology of swollen shoot of cacao II, Bull, cot. Res. 1*21 65-103. Sudd* J.H. 1967* An introduction to the behaviour of ants. Bdward Arnold (Publishera) Ltd., London. Talbot, M. 191*3- Population studies of the ant Prenolettls lroparls Say, Ecology 2ii» 31-UU. Talbot* H. 19li0. A comparison of two ants of the genua Formica. Ecology 29* 316-325. Talbot, M. 1956. Flight activities of the ant Dollchodarus (Hypochlnea) aarlae (Forel), Psyche 63» 13U-I39. Talbot, N. 1957. Population studies of the slave-siaking ant leptothoraz duloticua and its slave, Leptotharax curvianlnoaus. Ecology jSt U*9-Ai56. Talbot, K. and Remedy, C.H. I9li0. The slave-making ant* Formica aaneuinea auhlntegra Haeny, its raids* raytial flights and nest structure, Ann, ant. Soc. Am. 33« 560-77. Taylor, C.J. 1952. The vegetation:, zones of the Gold Coast. Dull .Forest. Dept, bold Coast lit 1-12. Urquhart, 0. 1955. Cocoa. Longmans, Green, London 8-10. Voalclcer, O.J. 19li7. Biological control, Rep. VI. Afr. Cacao Res. 3nat. Tafo, 19l5-li6| 37. Vowles, D.M. 1955. Hie foraging of ants, J. Anlra. Bahav.. 3* 1-13. Walker, H.O. 1962. Weather and climate, In Agriculture and La"** Use in Gh«e& by J.B. Kills), O.O.P. London, 7-50. Wallis, D.I. 1962. Ihe relation between hunger and worker function in ant colony, Proc. zool. soc. Land. 139* 589-605. University of Ghana http://ugspace.ug.edu.gh Wallis. D#I# 196k# The foraging behaviour of the ant Faraica fusoa, ftahavioar 23* 1 1*9-1 75 Weir, J.s. 19^ 9. Ejgg m asee and early larval growth in Hmaiea# TraMMStBB SOC# 6? 180—201* Wheeler, W#H# 1910# fl« Colmbia tfc&v# Pressj Hew Xosrk 378-1*22* Wheeler, W.M# 1920# The subfamilies of Foraieldae, and other ta^ OiOoie aotee, Psyche 27* 1*6-55 • Wheeler, W#M. 1922. Ants of the Belgium Congo, M U Im# 1fes. Nat#Hist# 15* 187-190. Iflheeler, W#M# 1933* Colony-foigKiing gsaoag ants with an account of sosae primitive AustrsJLian species* Harvard Uhlv# Press, Cambridge, Massachusetts • Wheeler* G.C. and VJheder, J# 1955* Tho ant larvae of the Megseex&rra i'ribe Leptothoracini, Ann, ent. Soc# Am# 1*8* 17-20. M I U m b G. 1951i« Held observations on the cacao Hirids, Sahlbereella slngularls Hagl. and Dlstantiella theofaroroa (Hist.) in the Gold Coast# Part IU. Peculation fluctuations, Bull, ent. Ees# 1|5* 72>7Wi. Walls, J#B. 1962# The general pattern of land use, in Agriculture ad Land Use in Ghana (Ed# by J#B# Wills), 0#U#P# London 2$-2&0. Wilson, E.0# 1963# The origin of polOTorrfiisa in ants. C# Sev# Biol# 20* 136-156. Wilson, E#0# 1959* Some ecological characteristics of ants in He*? Guinea rein forests, ZctiLotst I4O* 1*37-1^ 7. Youdeiwei, A# 1968# Notes on the feeding and habits of the cocoa xairid Sahlbeggella singular^ Hagl# in U&stern Nigeria, Nigerian oat. Mfag#, 1 * ?lt-98# University of Ghana http://ugspace.ug.edu.gh