Phenotypic plasticity in Australian cotton aphid (Homoptera: Aphididae): Host plant effects on morphological variation

Fourteen samples of cotton aphid, Aphis gossypii clover were obtained from different localities in Australia. Wingless parthenogenetic females from each sample were transferred to seedlings of various host plants in the laboratory and reproduced to form colonies. Winged aphids from 120 colonies were mounted on microscope slides. Twenty-five characters were measured on each aphid and analyzed for differences in origin and laboratory host plant. Aphids reared on cucumber and rockmelon were significantly larger than on cotton, broad bean, or eggplant. Large and highly significant differences among the means of different colonies on the same host plant species in many characters were probably caused by different quality of the individual seedlings. Significant morphological differences were detected among source-line means of some characters, but their magnitude was small compared with variation within the same source line. The inescapable conclusion from this study is that A. gossypii in Australia is phenotypically plastic and its morphology is affected by the host plant far more strongly than by genetic differences (if any) among means of local populations. Moreover, only 13% of A. gossypii host records in Australia are from native Australian plants, implying that it was introduced into Australia, perhaps by the Europeans in the 18th century Characters of 36 colonies on cucumber and 40 colonies on cotton were analyzed by analysis of covariance (ANCOVA) using wing length (a measure of aphid size) as a covariate. No significant differences in character means remained among source lines on cucumber, when the variation caused by aphid size was thus removed. On cotton, means of some characters still differed among source lines, after size adjustment by ANCOVA. Some evidence indicates that geographical samples of A. gossypii in Australia may differ (genetically?) in shape rather than size. When the 40 colonies reared on laboratory cotton were grouped by the host plant on which the source samples were collected, colonies originating in cotton fields were larger, in most characters, than colonies collected on other host plants. This pattern was not observed in the 36 colonies (of the same sources) reared on laboratory cucumber. If aphid size is a measure of fitness, then cotton-collected aphids had higher fitness than other lines when reared on cotton. The cotton aphids may then show traces of adaptation to their field host.