Extremely long mouthparts in flower-visiting insects

Extremely long mouthparts that serve for the uptake of nectar in flower visiting insects provide ample opportunity to examine constraints on organ evolution. The majority of the flower visiting insects are regarded as short-tongued since their mouthparts are shorter than the head, while extremely long proboscides, i.e., exceeding body length, are rare. Advantages of long proboscides have been previously formulated and tested in nectar feeding from long spurred flowers. The project represents the first attempt to evaluate the costs of long mouthparts. The research will compare flower-visiting insects which have an average-sized tubular proboscis with related species having extremely long mouthparts. The study will include species of hawk moths (Sphingidae) with proboscides longer than 250 mm, neotropical butterflies (Riodinidae) and orchid bees (Euglossini) with proboscis lengths greater than 35 mm. The South African tangle-veined flies (Nemestrinidae) with a variable proboscis length of 20-50 mm within the species is chosen for one of several case studies. The research hypotheses postulate that (1) anatomical costs are optimized in those regions of the proboscis which are disproportionately elongated to the extent that they contain fewer muscles and sensillae compared to corresponding regions in average-sized proboscides of closely related species. (2) Anatomical costs of variously long proboscides within a single species will provide an estimate of the extra expenses which accompany advances of proboscis elongation. (3) Biometry and biomechanics of the suction pumps correlate with the proboscis length. (4) Flower handling time should greatly increase in butterflies and nemstrinid flies with extremely long proboscides. The comparative studies will use microanatomical methods (electronmicroscopy and micro CT), biometric analyses and field studies in the Tropical Biological Station La Gamba (Costa Rica) and South Africa. The biomechanics of suction pumps will be studied in living butterflies using synchotrone x-ray imaging in cooperation with the Argonne National Laboratory (U.S.A.). The estimation of anatomical and functional costs of extremely long proboscid mouthparts should contribute significantly to our understanding of the evolution of form and function in context with insect feeding behaviour on flowers.

Extremely long mouthparts that serve for the uptake of nectar in flower visiting insects provide ample opportunity to examine constraints in organ evolution. The majority of the flower visiting insects are regarded as short-tongued since their mouthparts are shorter than the head, while extremely long proboscides, i.e., exceeding body length, are rare. Benefits of long proboscides have been previously formulated and tested in nectar feeding insects on long spurred flowers however material costs were studied for the first time in this project. The research project was the first attempt to evaluate costs of proboscis length by comparing insects which have average-sized proboscides with related species having extremely long mouthparts. The study investigated Neotropical butterflies (Hesperiidae, Riodinidae), sphingid moths (Sphinigdae), orchid bees (Euglossini), and South African flies (Nemestrinidae, Tabanidae). All groups included representatives that possessed a proboscis at least twice as long as the body while related species showed average-sized proboscis lengths. Convergent evolution shaped the morphology of the mouthparts in insects that are adapted to take up nectar from long spurred flowers. In species which exhibited mouthparts exceeding the body length, the mouthparts were characterized by a very thin proboscis, a short apical nectar uptake region and muscular nectar suction pumps in the head. We found allometric scaling relationships between proboscis length, suction pumps and body size. These mouthpart structures which composed the food tube were disproportionately elongated. The musculature that was responsible for proboscis movements and the muscles of suction pumps showed allometric correlations to body size. In this way these organs were adapted to the proboscis lengths however additional expenses came from enlarged muscle mass of nectar suction pumps and organs of proboscis movements related to proboscis elongation. Anatomical costs were optimized in those parts of the proboscis which were disproportionately elongated to the extent that they contained fewer muscles and lower numbers of sensilla than corresponding regions in normal sized proboscides of related species. Field observations and experiments indicated greater nectar uptake rates in longproboscid species however flower handling times increased in species exhibiting extremely long proboscides compared to species with medium-sized feeding apparatus.The examination of anatomical and functional costs of extremely long proboscides of flower visiting insects contributed to a better understanding of evolutionary constraints of form and function in context with insect feeding behaviour on flowers and helped to explain why only few nectar feeding insects evolved particularly long proboscides.