Resumen

Studies of diversity and species distributions are crucial in light of a predicted 4- 6°C warming by the end of the 21st century, especially in the tropics where many species may be especially sensitive to climate change. Little is known about patterns of insect diversity and richness in Andean montane forest habitats. The main goal of the present research was to investigate the ecology of carabid beetles at different altitudes in southeastern Peru (Coleoptera: Carabidae). Changes in species richness, species assemblages, and wing and body attributes were estimated at different altitude zones along both disturbed and undisturbed gradients. My first objective was to estimate species richness patterns of carabid beetles on Andean slopes. Preliminary collections employed both hand collections and pitfall traps, but I found that hand collections were more efficient; subsequent data collection was carried out by hand collections. Preliminary data collected on an anthropogenically disturbed gradient from 1400 m to 3450 m revealed a decline in raw species number with altitude. When species number was adjusted for sample size (rarefaction) this trend produced a middle altitude peak in species richness. A truncated portion of the disturbed gradient (2000 m to 3450 m) was also compared to a parallel, old growth gradient, and no pattern was found with altitude for either gradient. However, overall species richness was lower for assemblages from the disturbed gradient. Species assemblages varied by altitude and by whether anthropogenic disturbance had occurred. The second objective was to characterize changes in body length and dispersal ability with altitude. For both morphological attributes, altitudinal patterns were highly variable among tribes of carabid beetles, exhibiting positive, negative and no correlation with altitude. When carabid beetle taxa were combined, however, body length and incidence of flightlessness were both significantly negatively correlated with altitude. Observation of morphological characters between the two gradients revealed that carabid beetles on an anthropogenically disturbed habitat were longer and had greater dispersal ability than those on an old growth gradient. The third objective was to compare assemblages of carabid beetles in the Andes of Peru to a temperate counterpart in the southern Appalachian Mountains, USA. Species richness of carabid beetle assemblages was found to be approximately twice as high in the Andean assemblage. Both assemblages showed a middle altitude peak in species richness. Furthermore, species composition of carabid assemblages had no overlap at the species and genus level, but species composition in both the Andes and the Appalachians varied with altitude. Body length was greater and dispersal ability was reduced for Appalachian as compared to Andean assemblages. Both regions exhibited similar patterns in body length in relation to altitude. However, while the proportion of flightless species was positively correlated with increasing altitude in the Andes, more than 90% of the species in the Appalachians were flightless at all elevations. I suggest the altitude-species richness trend may be more conclusive when a more complete altitudinal gradient is sampled, i.e., extending sampling along the entire length of the gradient. Furthermore, employing multiple study transects will enhance statistical validity of the results and greatly increase our knowledge of how carabid beetle assemblages vary along tropical altitudinal gradients.