The associations of the plant-ant Cladomyrma with plants in Southeast Asia Dissertation zur Erlangung des Doktorgrades der Naturwissenschaften vorgelegt beim Fachbereich Biowissenschaften der Johann Wolfgang Goethe-Universität Frankfurt am Main von Joachim Moog aus Frankfurt am Main Frankfurt am Main 2009 Vom Fachbereich ............................................................................................................ der J. W. Goethe-Universität, Frankfurt am Main, als Dissertation angenommen. Dekan: ............................................................................................................................. Gutachter: ....................................................................................................................... Datum der Disputation: ................................................................................................... Contents i Contents 1. Preface 1 2. Notes on terminology 4 2.1. Myrmecophyte 4 2.2. Domatium 6 3. Revision of the genus Cladomyrma 7 3.1. INTRODUCTION 7 3.2. MATERIAL AND METHODS 8 3.3. RESULTS AND DISCUSSION 9 3.3.1. Synopsis of species 13 3.3.2. Key to the queen caste 14 3.3.3. Species accounts 17 Cladomyrma andrei 17 Cladomyrma hobbyi 19 Cladomyrma maschwitzi 22 Cladomyrma nudidorsalis 23 Cladomyrma petalae 24 Cladomyrma yongi 27 Cladomyrma aurochaetae 29 Cladomyrma crypteroniae 31 Cladomyrma dianeae 33 Cladomyrma hewitti 35 Cladomyrma maryatiae 37 Cladomyrma sp. nov.? 39 3.3.4. Conclusion 40 3.3.5. Summary of the revision 44 4. Host plants of Cladomyrma 45 4.1. INTRODUCTION 45 4.2. MATERIAL AND METHODS 47 4.3. RESULTS 49 4.3.1. Fabaceae – Caesalpinioideae 49 Saraca thaipingensis 49 Saraca dives 58 4.3.2. Fabaceae – Papilionoideae 60 Callerya nieuwenhuisii 60 Spatholobus bracteolatus 65 Spatholobus oblongifolius 69 4.3.3. Loganiaceae 74 Strychnos vanprukii 74 4.3.4. Rutaceae 78 Luvunga sp. 78 4.3.5. Achariaceae 82 Ryparosa fasciculata 82 Ryparosa porcata 87 Ryparosa sp. nov. aff. calotricha 88 Ryparosa anterides 89 [Ryparosa amplifolia] 89 4.3.6. Crypteroniaceae 90 Crypteronia griffithii 90 Crypteronia macrophylla 97 4.3.7. Putranjivaceae 100 Drypetes longifolia 100 Drypetes fusiformis 105 ii Contents 4. Host plants of Cladomyrma (continued) 4.3.8. Rubiaceae 108 Neonauclea 108 Neonauclea artocarpoides 127 Neonauclea borneensis 127 Neonauclea calcarea 127 Neonauclea excelsioides 129 Neonauclea gigantea 129 Neonauclea longipedunculata 132 Neonauclea paracyrtopoda 132 Neonauclea pseudocalycina 135 Neonauclea sp. A 135 Neonauclea sp. B 138 Neonauclea sp. C 139 Neonauclea sp. D 139 Neonauclea sp. E 139 4.3.9. Summary of the results: host plants 143 4.4. DISCUSSION 144 4.4.1. Taxonomic diversity 146 4.4.2. Diversity in distribution 146 4.4.3. Diverse habitats 147 4.4.4. Diverse growth forms 150 4.4.5. Differences in ant-housing structures 150 4.4.6. Preadaptations for myrmecophytism 155 4.4.7. Evolutionary specialisation of ant-housing structures 158 4.4.8. The special case of Neonauclea and its sister taxon Myrmeconauclea 162 5. Host-choice 166 5.1. INTRODUCTION 166 5.2. MATERIAL AND METHODS 167 5.3. RESULTS 169 5.3.1. Summary of the results: host-choice 170 5.4. DISCUSSION 170 6. Acquisition of scale insects 173 6.1. INTRODUCTION 173 6.2. MATERIAL AND METHODS 176 6.3. RESULTS 181 6.3.1. Dispersal flight 181 6.3.2. Colony founding 181 6.3.3. Experimental exclusion of trophobionts 185 6.3.4. Active entry of immature mealybugs into founding chambers 187 6.3.5. Primary hole 190 6.3.6. Summary of the results: acquisition of scale insects 191 6.4. DISCUSSION 192 6.4.1. Co-dispersal 192 6.4.2. Colony founding 195 6.4.3. Secondary slits 196 6.4.4. Primary hole 197 6.4.5. Experimental exclusion of trophobionts 198 6.4.6. Active entry of immature mealybugs into founding chambers 202 6.4.7. Scale insect dispersal 204 6.4.8. Conclusion 205 Contents iii 7. Flood control 207 7.1. INTRODUCTION 207 7.2. MATERIAL AND METHODS 208 7.3. RESULTS 209 7.3.1. Summary of the results: flood control 212 7.4. DISCUSSION 212 8. Ant activity 215 8.1. INTRODUCTION 215 8.2. MATERIAL AND METHODS 216 8.3. RESULTS 218 8.3.1. Food rewards on the plant surface? 218 8.3.2. Spatial and temporal distribution of ant activity 221 8.3.3. Activity on mature plant parts 228 8.3.4. Behaviour of the ants on young leaves 229 8.3.5. Response to disturbance of the host plant 229 8.3.6. Summary of the results: ant activity 230 8.4. DISCUSSION 231 8.4.1. Variation of ant activity and response to disturbance 231 8.4.2. Activity on mature plant parts 232 8.4.3. Preference for young plant parts 232 8.4.4. Does the host plant mediates the ant's behaviour by chemical signals? 233 8.4.5. What is the cue by which Cladomyrma ants can recognise young plant tissue? 235 8.4.6. Why do Cladomyrma species preferentially patrol young plant tissue? 236 9. Protection against herbivory 239 9.1. INTRODUCTION 239 9.2. MATERIAL AND METHODS 240 9.3. RESULTS 244 9.3.1. Ant response to insect herbivores 244 9.3.2. Field observations of unprovoked encounters with insect herbivores 250 9.3.3. Ant response to insect eggs (cleaning behaviour) 252 9.3.4. Ant protection of young leaves 253 9.3.5. Summary of the results: protection against herbivory 258 9.4. DISCUSSION 259 9.4.1. Protection against phytophagous insects 259 9.4.2. Cleaning behaviour 261 9.4.3. Does Cladomyrma provide protection against folivorous mammals? 263 9.4.4. Underestimation of the protective effect 265 9.4.5. Protection by a facultative ant, Crematogaster sp. 1 266 9.4.6. Is degree of folivory a good precursor of plant fitness? 268 9.4.7. Plant defence hypotheses 270 9.4.8. What is the role of biotic defences in relation to the plant's other defences? 273 9.4.9. Vulnerability of young plant tissue 274 9.4.10. The role of extrafloral nectaries 275 9.4.11. Biotic defence by Cladomyrma 278 10. Pruning of neighbouring plants 282 10.1. INTRODUCTION 282 10.2. MATERIAL AND METHODS 284 10.3. RESULTS 285 10.3.1. Summary of the results: pruning behaviour 290 10.4. DISCUSSION 291 10.4.1. Pruning intensity and climber toughness or age 291 10.4.2. The Cladomyrma maschwitzi case 293 10.4.3. Pruning in climber-inhabiting Cladomyrma species 294 10.4.4. Chemical 'warfare' 297 iv Contents 11. Concluding discussion 300 11.1. A plant’s perspective: benefits of being colonised by Cladomyrma 301 11.2. A plant’s perspective: costs of being colonised by Cladomyrma 304 11.3. Do host plants need Cladomyrma for their survival? 309 11.4. The ants’ point of view 315 11.5. Host specificity 317 11.6. Host range and specialisation in other plant-ant lineages 321 11.7. Evolution of host affiliation 323 12. Summary 329 13. Ausführliche Zusammenfassung 331 13.1. Einleitung 331 13.2. Revision der Gattung Cladomyrma 332 13.3. Das Wirtspflanzenspektrum 332 13.4. Unterschiede zwischen den Wirtpflanzen 333 13.5. Nutzen der Cladomyrma-Besiedlung für die Wirtspflanzen 336 13.6. Kosten der Cladomyrma-Besiedlung für die Wirtspflanzen 337 13.7. Brauchen Wirtspflanzen Cladomyrma für ihr Überleben? 338 13.8. Nutzen und Kosten aus Sicht der Ameisen 338 13.9. Wirtsspezifität 339 13.10. Evolution des Wirtsspektrums 339 14. References 341 15. Appendix 378 15.1. Table index 378 15.2. Figure index 379 15.3. Deposit of Cladomyrma specimens in Malaysia 383 15.4. AntWeb field guide to Cladomyrma (www.antweb.org) 385 16. Acknowledgements 388 17. Publications 391 18. Lebenslauf 394 19. Erklärung 395 Preface 1 1. Preface Ada gula ada semut Malayan proverb No plant can live in tropical regions without having personal contact with ants. What happens when these two worlds come together? What follows is a treatise on a rather extraordinary ant/plant complex. Specifically, this dissertation is about Cladomyrma plant-ants and their extremely extensive host plant range. In a broad sweep this study examines several aspects of the behavioural ecology of Cladomyrma, for example colony founding, host plant selection, acquisition of trophobiotic mealybugs, protection of their hosts against herbivorous insects, and effects of catastrophic events, such as flooding of nest chambers. Another major part of this study explores Cladomyrma taxonomy and host plant range in order to evaluate the degree of host specificity. All these aspects touch some fundamental biological questions concerning mechanism of maintenance of diversity, e.g. habitat and/or host selection, intra- and interspecific competition and stochastic versus deterministic colonisation. Other major subjects in biology that are touched by this study are plant anti-herbivore defence systems and the evolution of interspecific mutualisms. Such mutualisms are widespread, but how they evolve is not clear. What factors promote the intimate interaction between unrelated organisms? Do these ant/plant associations show the potential for reciprocal evolutionary change, i.e., for coevolution? Naturally, these subjects, being of major importance to biological science, cannot be treated here in depth. On the contrary, it is my intent to maintain a broad but rather superficial course by highlighting various different aspects, thus providing a first glance from different angles on this newly discovered ant/plant complex. The result is a puzzle with widely separated pieces and many gaps. Still, I hope that a picture emerges that serves as the framework for promising future research. 2 Preface Let me say a few words regarding the so-called 'big science'. We presently live in a time when molecular biology and other types of big science receive far more attention by grant giving institutions than 'simple' field research. This trend will continue to reduce interest in whole-organism biology and natural history in both universities and the students they produce. But, as DEVRIES, CHACON & MURRAY (1992, p.123) have stressed earlier, "without data from the real world, no matter how sophisticated laboratory techniques or models become, in the absence of natural history they are unlikely to broaden our understanding of the myriad interactions among organisms." In my opinion, future insights into tropical biodiversity in general, and ant-plant mutualism specifically will demand a great deal more field work. The study on Cladomyrma/plant associations presented here illustrates this need. The tropical rain forests display a biodiversity unparalleled by that of any other vegetation type. However, over the past few years I experienced a dramatic loss of tropical rainforest in some of my study areas. On the island of Borneo, for example, the natural forest cover is disappearing at a fast rate. Sabah, a federal state of Malaysia on Borneo, has almost depleted its primary forests outside conservation areas (UEBELHÖR & VON DER HEYDE 1993, RAMAKERS 1999, CHUNG et al. 2000, FITZHERBERT et al. 2008). Today, more than two thirds of the commercial forest reserves have been turned into second growth forests. Secondary growth and selective logging, in turn, greatly enhance fire danger and, indeed, in the last two decades, fire has become one of the greatest threats to Bornean tropical rainforests (SIEGERT & HOFFMANN 2000). For the Bornean province East Kalimantan (Indonesia) alone it is estimated that during a severe drought caused by the 1997/98 El Niño phenomena, some 3 million hectare of forest (700,000 ha of protected forests and 2.3 million ha of natural forest concession areas), about 15% of the entire province, were affected by fire (HOFFMANN, HINRICHS & SIEGERT 1999). As yet there is no clear trend of a return to pre-disturbance conditions, which indicates that the burned forest may remain in a severely degraded state for a prolonged period of time (CLEARY & PRIDJATI 2005). Certainly, several of the studied ant/plant associations will be strongly affected by these intense logging activities and increased fire danger. Although a few ant associations with pioneer host plant species seem to benefit - at least on a short-term basis - by these activities, I fear that in the long run they will be diminished as well. My observations suggest that the association with Cladomyrma is lost - even in pioneer hosts - when the distance to rainforest sources increases. A recent study by TANAKA et al. (2007) corroborates my assumption that deforestation can drastically decrease the