Genetics and Ecology of Drosophila Melanogaster Larval Foraging and Pupation Behaviour

In this paper we show that, (1) Drosophila melanogaster larvae utilize a variety of pupal microhabitats in an orchard, (2) variation in larval foraging path length, pupation distance from the food and pupal microhabitat preference (on or off the fruit) is genetically based and, (3) variation in these behaviours can be maintained in a spatially heterogenous environment since there is a reversal in pupation site suitability in wet and dry pupal microhabitats. Differences in path length in both laboratory and natural populations can be attributed to genes on the second pair of chromosomes and is under simple genetic control, whereas differences in pupal height are polygenically inherited (the second pair of chromosomes influences pupal height three times more than the third pair). Pupae collected from on-fruit sites had shorter foraging path lengths and lower pupal heights than off-fruit populations. Populations from the orchard maintained their field pupal microhabitat preferences even after 1 year of rearing them in the laboratory. Larvae with the sitter larval phenotype (short path lengths and low pupal heights tended to pupate more on-fruit than those with the rover phenotype (long path lengths and high pupal heights). To determined if these genetically based differences in microhabitat preference contributed to fitness, larval pupation behaviour was studied in a “field assay” (dish with fruit on soil) with soil water content varied. At low soil water contents, pupal survivorship was significantly better on the fruit whereas, at high soil water contents, survivorship was better in the soil. There was a reversal in which microhabitat (dry or wet) was a better site for pupation. In the field environment where soil water content fluctuates in space and time, such a reversal would explain the maintenance of genetic variation for these larval behaviours. Another selective agent acting on D. melanogaster larvae in our orchard is parasitization by Asobaru tabidu. This parasitoid parasitizes larvae with high locomotory scores (e.g. rovers) significantly more than those with low scores (sitters). This study relates laboratory phenotypes to field phenotypes thereby linking the ecological, behavioural and genetic components of larval habitat selection in D. melanogusrer. Key Word Index: Microhabitat morphism, rover/sitter, parasitoid selection, genetics, ecological, larval behaviour, Drosophila, polyHabitat choice occurs through differences in behavioural preferences as well as selection (Powell and Taylor, 1979). Natural selection is thought to favour individuals with the most fit of alternative behaviour patterns in a particular environment. It is common to speculate about the adaptive significance of behavioural traits (Brown, 1975). Inherent to this speculation is the assumption that there is a genetic basis to differences in behaviour. The study of the genetics of behaviour is a relatively new field of investigation (Ehrman and Parsons, 1981). Behaviour, like morphology, is a phenotype. A behavioural phenotype is influenced by both genotype and environment. A genetic basis for differences in behavioural preferences must be demonstrated before the role of natural selection in habitat choice can be implicated. A genetic basis for intraspecific variation in habitat selection has not been conclusively demonstrated for any species (Parsons, 1983). Examples of intraspecific variation in behaviours potentially important to habitat selection are numerous. Alary polymorphisms, the occurrence of several morphs differing in their ability to fly, are found in aphids (Lees, 1966; Hardie, 1980). In migratory locusts, gregarious and solitary morphs differ behaviourally and morphologically (Kennedy, 1975). In the gypsy moth, Lymantria dispar, the size of the hatching larva which influences dispersal tendency (Barbosa er al., 1981) is proportional to the egg size (Barbosa and Carinera, 1978). Wellington (1957, 1960, 1964, 1965) has reported a foraging polymorphism in the western tent caterpillar Mulucosoma pluviule with morphs differing in locmotory, phototactic and food location behaviour. Denno et al. (1980) demonstrated differences in habitat selection between the wing forms of the dimorphic planthopper, Prokefisiu marginata. Nevertheless, studies which distinguish between genetic and environmental contributions to the components of variation between any of these morphs are lacking. Rausher (1978) has identified a polymorphism in the searching behaviour of adult female pipevine swallowtail butterflies, Buttusphilenor. Females search for host plants in one of two modes, the broad leaf or the narrow leaf mode. Papaj and Rausher (1983) suggest that these differences in search mode can be genotypically based, environmentally induced and/or experiential. They provide preliminary evidence that