Thanks to the progress in genetic methods in population ecology, many critical discoveries were recently made in the field of reproductive strategies of social insect. Among them is the description of mixed reproductive strategies, combining advantages of sexual reproduction with thelytokous parthenogenesis. The queens of such species produce sterile castes through classical sexual process from fertilized eggs, while future queens develop asexually from unfertilized eggs. This original breeding system was first described in several genera of ants, but it was found very recently, that it is not restricted to social Hymenoptera, since it has been identified also in the phylogenetically remote eusocial clade of termites. Switching between the sexual process and thelytokous parthenogenesis, now known as Asexual Queen Succession (AQS), which enables a continuity of genetically almost identical queen generations after the death of the founding primary queen, was first reported only in one genus of lower termites, i.e. Reticulitermes. Recently, our research group participated at the identification of AQS in four other species from two subfamilies in higher termites. One of these species is Silvestritermes minutus. This species is locally abundant in French Guiana and lives in small, well shaped nests on...

The traditional view of a lifelong monogamy between a king and a queen has recently been challenged in termites. In several species, multiple parthenogenetically-produced secondary queens replace the primary queen and mate with the primary king; this strategy is referred to as “Asexual Queen Succession” (AQS). The aim of my thesis was to investigate the modalities of reproduction and the ecology of neotropical soil-feeding termites from the Termitinae, with a focus on the inquiline termite Cavitermes tuberosus in the Termes group.In the first axis, we investigated the modalities of reproduction of C. tuberosus. (i) AQS is the main reproductive strategy of this species. (ii) The evolution of AQS requires the propensity of parthenogens to develop into neotenic queens. In C. tuberosus, secondary queens develop from a developmental stage of “aspirants” which participate to the social tasks usually undertaken by workers, as long as the primary queen is alive. (iii) In AQS species, a female-biased sex ratio is expected in the dispersing reproductives. In C. tuberosus, sex ratio varies among years and according to the type of reproductives, and the population sex ratio is balanced. These results raise hints on queen-king conflict over the sex ratio.In the second axis, we described the ecology and symbioses of C. tuberosus. (iv) Wolbachia, an endosymbiotic bacterium mainly known for manipulating the reproduction of arthropods in order to enhance its own transmission, infects all individuals in societies. This bacterium, particularly abundant in a gut-associated bacteriome, may play a role in the nutrition of C. tuberosus; both partners would have evolved a mutualistic symbiosis. (v) Inquiline termites live in a nest built by other termite species and do not forage outside. Physico-chemical measures and microbiota sequencing revealed that C. tuberosus is a generalist nest-feeder.Finally, we expanded our study of the breeding systems in the phylogenetic proximity of C. tuberosus. (vi) We described Palmitermes impostor, a new genus and species as a sister-group to the genus Cavitermes. (vii) AQS is the main reproductive strategy in P. impostor, and queens of Spinitermes trispinosus and Inquilinitermes inquilinus are able to reproduce parthenogenetically. Therefore, it appears likely that the conditional use of sexual and asexual reproductions is a preadaptation common to the whole Termes group, and that it evolved into a stable element of their breeding system at least in some species.Overall, our results open new perspectives in the understanding of reproductive strategies in termites and their relationships with their bacterial symbionts. / Doctorat en Sciences / info:eu-repo/semantics/nonPublished

Altruism is defined as a trait in an individual that increases some other individual’s fitness at the expense of her own. Therefore, existence of such traits in a population is an evolutionary paradox, as natural selection should eliminate such a trait. Extreme altruism in the form of eusociality where individuals relinquish their own reproduction to help raise other’s offspring has been an enigma in evolutionary biology since Darwin. Primitively eusocial organisms provide one with a unique system to study the evolution and maintenance of altruism as in these kind of species most of the individuals are capable of developing their reproductive organs, although at a certain point in time, only one or a few individuals actually reproduce. Ropalidia marginata is a primitively eusocial wasp belonging to the insect order Hymenoptera, Family Vespidae. R. marginata colonies are monogynous, although serial polygyny is observed in a colony’s lifetime. Colony initiation happens either by single founding or multiple founding. Newly founded colonies may accept individuals from other colonies, but mature colonies seldom do. Production of males is irregular, and once eclosed, they generally leave their natal nest within a week. The haplodiploidy of Hymenopteran species, i.e. the males being haploid and the females diploid, make them uniquely genetically predisposed for eusociality to evolve as was shown by William Donald Hamilton in his kin selection theory. Primitvely eusocial Hymenopteran species, being susceptible to experimental manipulation, allows one to test the predictions of this theory. In this thesis I have addressed three aspects of the biology of R. marginata using microsatellite markers, which are the following: 1) Distribution of nestmate genetic relatedness in early founding (pre‐emergence) and mature ( post‐emergence colonies) and their comparison (Chapter 3) 2) Role of relatedness and fertility in predicting the queen’s successor (Chapter 4) 3) Genetic structure of populations (Chapter 5) CHAPTER 1. INTRODUCTION: This chapter gives a brief outline of the field of molecular ecology putting its techniques to the context of insect sociobiology. CHAPTER 2. METHODS: This chapter gives a general outline of the molecular genetic methods involved. In addition, the issue of the mutation process in R. marginata microsatellites has also been addressed. There are two main models of mutation for microsatellite evolution i.e. infinite alleles model (IAM) and the step‐wise mutation model (SMM). To understand the actual process of mutation in R. marginata, sets of alleles with continuous sizes were sequenced and aligned. This was repeated for several of the loci. Seven out of the nine loci genotyped revealed clear step‐like mutation pattern and was binned accordingly. Two loci were dropped as the actual nature of step‐sizes in these two loci was unclear. Therefore, the final dataset consisted of genotype for 7 loci. This chapter also discusses the initial steps in data formatting and analysis. CHAPTER 3. GENETIC RELATEDNESS IN DIFFERENT STAGES OF COLONY DEVELOPMENT: In this chapter I have estimated nestmate genetic relatedness using seven polymorphic microsatellite loci in two different stages of colony development of the primitively eusocial wasp Ropalidiamarginata and compared them. In both kinds of nests the average colony relatedness was observed to be less than 0.75, i.e., what is expected for full sib females in Hymenoptera. Moreover, it was observed that the nestmates at the initial colony founding stage are on average less related to each other than in mature colonies. From this, one may postulate that the indirect component of inclusive fitness plays a relatively minor role than its direct component as individuals chose to leave a higher relatedness background in favour of a lower relatedness background. As newly founded colonies are relatively smaller in size than mature colonies, the probability of an individual wasp becoming the queen in this kind of colony is higher than in mature colonies. CHAPTER 4. TESTING THE ROLE OF RELATEDNESS AND FERTILITY IN PREDICTING THE QUEEN’S SUCCESSORS: R. marginata colonies are headed by docile queens. When this queen dies or is removed, one of the workers becomes extremely aggressive. She is known as the potential queen because within a few days she becomes the new queen of the colony and her aggression comes down. Predicting the successor in the presence of the queen has eluded most of the approaches attempted so far. The probability of an individual becoming the queen has been found to be uncorrelated with her body size, aggression, ovarian status or mating status. The only trend that has been observed till date, is a positive correlation with age, but the pattern is not perfect. However, the workers themselves seem to be perfectly aware of who their immediate successor going to be. In this chapter, I have tested several models of queen succession constructed in an inclusive fitness framework. These models have been tested both using relatedness alone as well as using fertility along with relatedness. Predictions of none of the models actually matched the observed sequence of successors. The wasps do not seem to be choosing their successor to maximize their inclusive fitness. CHAPTER 5. GENETIC STRUCTURE OF NATURAL POPULATIONS: I have also looked at the genetic structure of R. marginata populations in a large part of its natural distribution. I have used both F and R statistics to estimate the level of structuring and compared them. Both Fat as well Rst were found to be significantly larger than 0. Also Fis and Ris both were small and not significant suggesting lack of inbreeding. Rst was observed to be higher than Fst. Permutation test revealed a higher contribution of mutation in this structuring than migration, suggesting Rst to be a better measure of genetic structuring in this case. Similar pattern was observed with Anlysis of MOlecular VAriance. Pairwise Fst/(1‐Fst) values were found to be uncorrelated with distance, whereas barely significant trend was observed with Rst/(1‐Rst). The scatter across the trend line in both the cases suggested lack of migration drift equilibrium, with drift being more relative to migration. Higher level of structuring was observed at the level of the colony. However, colonies were rather outbred as was suggested by high and negative values of Fia and Ria values. This is not at all surprising as nestmates are related to each other. The pattern of isolation by distance at the colony level was similar to that observed in case of the populations. However, there was even higher degree of scattering of the individual points in this case. CHAPTER 6. CONCLUSIONS: Hamilton’s inclusive fitness theory has received a wide attention from and acceptance by sociobiologists, and relatedness have been measured in a wide variety of social insects. In this thesis relatedness in the context of colony founding was measured and compared with mature colonies. Also, several models constructed in an inclusive theory framework were experimentally tested. In both, support for indirect fitness was found wanting. The population genetic structure of R. marginata revealed that the sub populations are small in size and migration among them low. It also suggested significant contribution of colony level structuring on the population genetic structuring. Using more modern molecular genetic and statistical techniques, these and similar other questions can be addressed with higher precision and rigour, and such studies are expected to greatly advance our understanding of the basic premise of this thesis, i.e., how can eusociality evolve and be maintained? We hope that the current work will encourage others to ask such questions in other species.