Systematics, Ecology and Behaviour of Orthopteran Insects
Dr. Rohini Balakrishnan

The Order Orthoptera (crickets and grasshoppers) is one of the evolutionarily ancient orders of insects and is distributed world-wide, with several representative groups in both temperate and tropical regions of the Old and New World. The order may be broadly sub-divided into three sub-groups: the Ensifera (or crickets), the Caelifera (or grasshoppers) and the Gryllacridoidea (whose phylogenetic relationship with the other two groups is currently unclear). In contrast to the Caelifera and the Gryllacridoidea, in which many of the species do not produce sound, members of the sub-order Ensifera often exhibit acoustic communication. The acoustic signals are typically produced by adult males in the context of mate attraction. The acoustic signals or calling songs produced by the males are highly species-specific, with each cricket species producing a unique song.

The Ensifera may be further sub-divided into two families: the Gryllidae (or true crickets) and the Tettigoniidae (or bushcrickets). Both gryllids and tettigoniids produce sound by rubbing the forewings together: a specialised structure called the plectrum (on the inner margin of one forewing) scrapes against a row of pegs on the stridulatory file on the ventral surface of the other fore wing to produce sound. The sound is then amplified by special resonating areas of the forewing, such as the harp and mirror. Most gryllid and tettigoniid species also possess tympanal ears, which are located, one on each foreleg, at the proximal end of the tibia.   

Cricket species of the sub-order Ensifera provide excellent systems to study the perception, function, diversity and evolution of animal communication signals because they use a repertoire of acoustic and non-acoustic signals for both long- and close- range communication in the context of mate-finding. Since they are very widely distributed over a range of habitats from suburban gardens, scrub and grasslands to tropical rainforests, and occupy a variety of micro-habitats ranging from the ground level to the canopy, they also offer opportunities for studies in community ecology. Both behavioural and ecological studies require, however, to be placed in an evolutionary context and the reconstruction of phylogenetic relationships between the focal species or genera therefore acquires significance.


Studies of ecological communities of orthopteran insects, including the diversity and distribution of species, can only be rigorously carried out if species can be identified with a high degree of certainty and if the sampling methods for estimation of species richness and relative abundance are efficient, reliable and unbiased. In contrast with insect groups such as butterflies and ants, however, orthopteran insects, particularly crickets (sub-order Ensifera) are poorly studied and unambiguous taxonomic identification to the species level is difficult, if not impossible, for several ensiferan groups from the Indian subcontinent. The reasons for this include imprecise definition of characters, incomplete species descriptions, ambiguities in the available taxonomic keys and inaccessibility of the holotype specimens for species verification (since the holotypes are typically in museums in Europe and North America). Under the circumstances, it is obvious that detailed taxonomic studies will have to precede any efforts to estimate species richness and relative abundance.

We (my research group and I) have therefore begun a research program in orthopteran systematics currently focused on two gryllid sub-families: the Oecanthinae (or tree crickets) and the Gryllinae (or field crickets). We have used the Genus Oecanthus, with only four described species from India (all of which occur sympatrically in and around Bangalore), as a model system to investigate the utility of a numerical taxonomic approach (using multivariate statistical techniques) to the delimitation of species boundaries and the identification of species. We have used these techniques on both song and morphological characters independently in order to contrast their effectiveness in correctly delimiting and diagnosing species. Our results suggest that, using multivariate statistical techniques, both song and morphological characters can be used to correctly delimit and identify the tree cricket species.

We aim in the next few years to test this approach on the much larger sub-family Gryllinae (field crickets), using the subset of species represented over Southern India. For this purpose, we plan to record the calling songs of as many field cricket species as possible so that we may again be able to contrast the efficacy of acoustic versus morphological characters in species identification. Our aim in the long run is to develop Web-based taxonomic keys for species identification of at least some groups of Indian gryllids. We also propose to reconstruct the phylogenetic relationships between species of some gryllid sub-families that may be of special interest for behavioural and ecological studies.


Sampling methods

Ecological sampling methods that are commonly used to estimate insect species diversity and relative abundance include sweep netting, pitfall trapping and light trapping. Whereas sweep netting is a relatively efficient and reliable method to obtain estimates of grasshopper species diversity and relative abundance, it is not very effective at trapping ensiferans (crickets). In our experiments using sweep netting, only about 10-12% of the total catch consisted of ensiferans, and these were restricted to very few species.

Field crickets (sub-family Gryllinae) can only be caught in pitfall traps, where they constitute less than 10% of the total insects captured. In a series of experiments using pitfall trapping, we found the efficiency of capture of field crickets to be extremely low. Further, several field cricket species that could be located by acoustic or ad lib sampling were simply not represented in the pitfall trap catches, in spite of the traps being placed in a number of micro-habitats and sampling over several weeks. Pitfall trapping is thus neither an efficient nor a reliable method to obtain accurate estimates of field cricket species richness and relative abundance. We are therefore trying to develop acoustic sampling as a reliable, efficient and non-invasive method of sampling field crickets. Acoustic sampling requires knowledge of the species-specific calls, and we are currently engaged in building a library of calls of different cricket species that could be used for acoustic sampling in the future.

None of the trapping methods described above are effective in obtaining estimates of tettigoniid species richness and relative abundance: very few tettigoniid species are actually trapped in sweep nets, pitfall traps or light traps and, in this case as well, acoustic sampling may provide a better alternative.

Cricket communities in evergreen forests

Natural communities of acoustically communicating orthopterans have rarely been studied in any depth in tropical forests world wide and even less so in India. Most studies to date have been short-term and carried out in a qualitative manner. Whereas communities of birds and frogs have been studied in considerable detail, we know little about the spatio-temporal patterning of orthopteran communities. We have recently initiated studies in evergreen forests of the Southern Western Ghats aimed at characterising the structure of communities of acoustically communicating ensiferan insects. We are currently engaged in obtaining the calls and identifying the species that constitute the acoustic insect community and aim in the long run to understand the patterns of habitat selection and temporal dynamics of these communities, as well as their response to habitat fragmentation and degradation. These studies should provide information of relevance to bio-diversity conservation.


The species-specific calling songs of crickets are used by females to identify and approach conspecific males. Females are thus tuned to the calling songs of males of their own species and tend to ignore heterospecific males that may be calling at the same time. The neural basis of recognition of conspecific song has been an important area of research in the field of neuroethology for several decades. Whereas the basis of tuning to spectral features of the song is fairly well understood, the neural basis of temporal pattern recognition continues to pose a challenge. In collaborative research with the University of Erlangen, Germany, a neural model for song syllable recognition in the duetting grasshopper Chorthippus biguttulus has been developed. We have also examined additive and synergistic interactions in the processing of different features of the signal within the female receiver.

The role of calling song as a pre-zygotic mechanism of species isolation is currently being investigated in two sympatric species of tree crickets of the genus Oecanthus, which are also synchronous in their seasonal breeding cycles. These two species are to be found co-existing in areas of natural vegetation (dry grasses and shrubs) on the campus of the Indian Institute of Science and are being studied in the field. The calling songs of the two species, Oecanthus henryi and Oecanthus indicus, are similar in several of their features, raising the question of the effectiveness of song in maintaining species identities in this case. Ethograms of the courtship behaviour of these species have also been developed in the same context. We aim to use these species in order to investigate the role of calling and courtship signals in species isolation and mate choice.

Contact Address
Centre for Ecological Sciences,
Indian Institute of Science,
Bangalore 560012, India
Ph: 3600985 Extn: 103