Yves Roisin

Biology of Termites: A Modern Synthesis

Biology of Termites, a Modern Synthesis brings together the major advances in termite biology, phylogenetics, social evolution and biogeography made in the decade since Abe et al Termites: Evolution, Sociality, Symbioses, Ecology became the standard modern reference work on termite science. Building on the success of the Kluwer book, David Bignell, Yves Roisin and Nathan Lo have brought together in the new volume most of the worlds leading experts on termite taxonomy, behaviour, genetics, caste differentiation, physiology, microbiology, mound architecture, distribution and control. Very strong evolutionary and developmental themes run through the individual chapters, fed by new data streams from molecular sequencing, and for the first time it is possible to compare the social organisation of termites with that of the social Hymenoptera, focusing on caste determination, population genetics, cooperative behaviour, nest hygiene and symbioses with microorganisms. New chapters have been added on termite pheromones, termites as pests of agriculture and on destructive invasive species, and new molecular and cladistic frameworks are presented for clarifying taxonomy, especially in the higher termites which dominate many tropical ecosystems. Applied entomologists, developmental and evolutionary biologists, microbial ecologists, sociobiologists and tropical agriculture specialists will all benefit from the new insights provided by this work.

Diversity and Evolution of Caste Patterns

Termite societies display a wide array of polymorphism patterns, rivalled only by the most complex ant societies. Caste differentiation in termites occurs when immatures of either sex deviate from the straight hemimetabolous pathway leading from the egg to the winged imago. Two kinds of irreversible deviations are found in primitive termites: one leads to the sterile soldier caste, the other to neotenic reproductives. Social tasks are primitively done by immatures which may postpone their maturation but retain a full range of developmental options. A major event in the evolution of termite caste patterns is the polyphyletic onset of a worker caste which is irreversibly excluded from the imaginal pathway. Termites with workers went through a remarkable behavioural and ecological diversification. After defining terms, I review here the diversity of caste patterns in termites in an evolutionary perspective. I provide examples of relationships between caste patterns and behavioural ecology, and examine the decision and regulatory mechanisms affecting caste determination. Trends in the evolution of termite caste patterns are then outlined, of which the most salient and enigmatic ones are the repeated occurrence of sexual dimorphism among workers and the frequent specialization of one sex in the formation of sterile castes.

Arthropod diversity in a tropical forest

Assessing Creepy Crawlies Arthropods are the most diverse group of terrestrial animal species, yet estimates of the total number of arthropod species have varied widely, especially for tropical forests. Basset et al. (p. 1481, see the cover) now provide more reliable estimates of total arthropod species richness in a tropical rainforest in Panama. Intensive sampling of a half hectare of forest yielded just over 6000 arthropod species. Scaling up this result to the whole forest suggests that the total species diversity lies between 17,000 and 40,000 species. Total arthropod species richness in a tropical rainforest can be best predicted by plant diversity. Most eukaryotic organisms are arthropods. Yet, their diversity in rich terrestrial ecosystems is still unknown. Here we produce tangible estimates of the total species richness of arthropods in a tropical rainforest. Using a comprehensive range of structured protocols, we sampled the phylogenetic breadth of arthropod taxa from the soil to the forest canopy in the San Lorenzo forest, Panama. We collected 6144 arthropod species from 0.48 hectare and extrapolated total species richness to larger areas on the basis of competing models. The whole 6000-hectare forest reserve most likely sustains 25,000 arthropod species. Notably, just 1 hectare of rainforest yields >60% of the arthropod biodiversity held in the wider landscape. Models based on plant diversity fitted the accumulated species richness of both herbivore and nonherbivore taxa exceptionally well. This lends credence to global estimates of arthropod biodiversity developed from plant models.

The Evolutionary History of Termites as Inferred from 66 Mitochondrial Genomes

Termites have colonized many habitats and are among the most abundant animals in tropical ecosystems, which they modify considerably through their actions. The timing of their rise in abundance and of the dispersal events that gave rise to modern termite lineages is not well understood. To shed light on termite origins and diversification, we sequenced the mitochondrial genome of 48 termite species and combined them with 18 previously sequenced termite mitochondrial genomes for phylogenetic and molecular clock analyses using multiple fossil calibrations. The 66 genomes represent most major clades of termites. Unlike previous phylogenetic studies based on fewer molecular data, our phylogenetic tree is fully resolved for the lower termites. The phylogenetic positions of Macrotermitinae and Apicotermitinae are also resolved as the basal groups in the higher termites, but in the crown termitid groups, including Termitinae + Syntermitinae + Nasutitermitinae + Cubitermitinae, the position of some nodes remains uncertain. Our molecular clock tree indicates that the lineages leading to termites and Cryptocercus roaches diverged 170 Ma (153-196 Ma 95% confidence interval [CI]), that modern Termitidae arose 54 Ma (46-66 Ma 95% CI), and that the crown termitid group arose 40 Ma (35-49 Ma 95% CI). This indicates that the distribution of basal termite clades was influenced by the final stages of the breakup of Pangaea. Our inference of ancestral geographic ranges shows that the Termitidae, which includes more than 75% of extant termite species, most likely originated in Africa or Asia, and acquired their pantropical distribution after a series of dispersal and subsequent diversification events.

Morphology, development and evolutionary significance of the working stages in the caste system of Prorhinotermes (Insecta, Isoptera)*

SummaryThe existence of a worker caste in Prorhinotermes inopinatus was determined with the help of biometrical analyses and morphological observations. No dichotomy between alate and worker lines was detected among the young larval instars. Mandible sclerotization and wear show that the larvae become active by the fourth instar in mature colonies, and by the third in incipient ones. Prothoracic finger-shaped expansions characterize the most dependent larval instars. The antennal structure is the best feature for distinguishing late instars. The developmental schema is very flexible: the late larvae of stages 5 to 8 (at least) can give rise to wing-padded nymphs, to the next larval instar, to presoldiers, or most likely to neotenic reproductives as well; the wing-padded individuals (nymphs) normally moult direct to alates, but they can also undergo a regression of their wing pads and revert to worker-like individuals or differentiate into soldiers or neotenics. Slight variations in alate size and antennal morphology result from their origin in different larval instars. Whole-colony censuses revealed that the major part of the colonys work force is composed of pseudergates, i.e. late larvae that do not follow the most direct pathway to the alate. The pseudergates of Prohinotermes do not constitute a distinct worker caste irreversibly diverted from the alate developmental pathway. For this reason the caste system of Prorhinotermes is analogous to that of the Kalotermitidae and Termopsidae and much more primitive than those of the other Rhinotermitidae so far studied, Reticulitermes, Schedorhinotermes and Coptotermes.

Philopatric reproduction, a prime mover in the evolution of termite sociality?

Summary: Philopatric reproduction commonly occurs even in the most basic isopteran lineages and is almost certainly a primitive potentiality of these insects. Several authors hypothesized that opportunities for philopatric reproduction may have been a prime mover for the three most remarkable events in the evolution of termite sociality: the origin of helping behaviour, the origin of soldiers, and the origin of a permanently apterous worker caste. A critical assessment of these hypotheses and a review of relevant evidence are presented here. I first discuss the hypothesis that termite helpers derive from individuals choosing to postpone dispersal because of nest inheritance prospects. Such individuals would have developed helping as a strategy yielding indirect reproductive benefits while they wait for reproductive opportunities. However, it appears that prospects of philopatric reproduction need not be invoked to explain the origin of helping behaviour, which can be justified more parsimoniously by a favourable intrinsic benefit/cost ratio. Second, the hypothesis that soldiers originated through selection for fighting abilities among neotenic reproductives is found to face important difficulties, the major one being how to explain the absence of fighting devices in almost all present-day neotenics. Finally, the hypothesized link between apterism and chances of philopatric reproduction, which might have favoured the onset of the worker caste, is poorly supported by empirical evidence. It appears thus that philopatric reproduction, notwithstanding its importance in the biology of many extant termite species, is unlikely to have been a prime mover in the evolution of helping and of altruistic castes in termites.

(E,E)-α-Farnesene, an Alarm Pheromone of the Termite Prorhinotermes canalifrons

The behavioral and electroantennographic responses of Prorhinotermes canalifrons to its soldier frontal gland secretion, and two separated major components of the secretion, (E)-1-nitropentadec-1-ene and (E,E)-α-farnesene, were studied in laboratory experiments. Behavioral experiments showed that both the frontal gland secretion and (E,E)-α-farnesene triggered alarm reactions in P. canalifrons, whereas (E)-1-nitropentadec-1-ene did not affect the behavior of termite groups. The alarm reactions were characterized by rapid walking of activated termites and efforts to alert and activate other members of the group. Behavioral responses to alarm pheromone differed between homogeneous and mixed groups, suggesting complex interactions. Antennae of both soldiers and pseudergates were sensitive to the frontal gland secretion and to (E,E)-α-farnesene, but soldiers showed stronger responses. The dose responses to (E,E)-α-farnesene were identical for both soldiers and pseudergates, suggesting that both castes use similar receptors to perceive (E,E)-α-farnesene. Our data confirm (E,E)-α-farnesene as an alarm pheromone of P. canalifrons.

Intragroup conflicts and the evolution of sterile castes in termites

A hypothesis is proposed to account for the onset of a helper phenotype in the ancestors of termites, for the regular expression of this phenotype by a fraction of the individuals composing the colony, and for the subsequent development and diversification of sterile castes (soldiers and workers) This hypothesis rests on the gradual enhancement of behavioral and physiological traits preexisting in noneusocial ancestors Ancestors of termites probably lived in parent-offspring groups, in which parents assumed reproduction while older offspring helped to raise younger siblings. I demonstrate that conflicts (in particular, competition among helpers to reach the alate stage) are expected within such groups. The ancestral character of such conflicts in termites is substantiated by empirical evidence for their existence in all major groups of extant lower termites, in which they can still play an important role in caste determination I suggest that such conflicts produced individuals with lessened abilities for alate development, which would invest preferentially in siblings rather than in their own regeneration. Selection could then favor any increase in the efficiency of this helper phenotype at carrying out social tasks and a fine-tuning of the switch mechanism between the two phenotypes-the helper and the sexual nymph-as to fit Hamiltons rule as closely as possible. This process could have led to the regular expression of nondispersive phenotypes (i.e., the onset of permanent helper castes), followed by a wide diversification of the helpers, as observed nowadays in the higher termites.

Feeding ecology and phylogenetic structure of a complex neotropical termite assemblage, revealed by nitrogen stable isotope ratios

1. In the current ecological classification of termites, four feeding groups (I–IV) are recognised, corresponding to a gradient of decomposition from sound wood to highly mineralised organic matter in the soil.

Replacement of reproductives in Nasutitermes princeps (Desneux) (Isoptera: Termitidae)

SummaryThe production of replacement reproductives in experimentally orphaned societies was investigated in Nasutitermes princeps (Desneux). From 1 to 180 replacement queens and multiple kings, all adultoids, were found in the 13 nests that were re-collected. Normal imagos, microimagos, or a mixture of the two forms were present, depending on colony composition at the time of orphaning. In colonies with alates, the corresponding forms were found as replacement reproductives. Where only young nymphs were present, microimagos differentiated. Neither the number of queens nor their level of physogastry was correlated with the time elapsed since orphaning, but the total egg-laying rates were. We suggest that competition among queens rather than time determines individual physogastric development. Observations show that the most likely cause of accidental queen death in nature is probably nest fall from the supporting tree. Even in this case, the queen may be able to migrate to a newly rebuilt nest. An experimental simulation of this situation showed that colony migration can occur. The data indicate that the replacement of the primary queen after her accidental death cannot by itself account for the high rate of polygyny (60%) encountered in N. princeps. Two other possible causes are the replacement of the ageing foundress as a normal event of colony life, and colony reproduction by budding off new nests with adultoid reproductives.