Body mass and geographical range are two main drivers of diet in animals, yet how these factors influence diet in the morphologically and ecologically diverse avian group of Psittaciformes is little known. We reviewed current knowledge of the diet of Neotropical parrots and assessed the relation between diet (breadth and composition), phylogeny, body mass and geographical range. Diet has been documented for 98 of 165 species, but information is available only for 34 of 59 threatened species, and countries with high species diversity (> 20 species) had few studies (one to seven). Neotropical parrot species consumed 1293 plant species of 125 families. When assessing the relative frequency of different food items in the diet (seed, fruits, flowers, leaves, nectar, bark and stems), we found that parrots mostly exploited seeds (41.9%) and fruits (38.3%) of native species. Diet overlap was very low among genera (0.006-0.321). At the species level, geographical range and body size explained the variation in diet composition. In particular, small parrots of restricted distribution had a distinct diet composition relative to either large or widely distributed species. Although body size and geographical range showed phylogenetic inertia, diet was independent of phylogenetic history. Our review not only reveals ecological factors explaining diet in a generalist group but also exposes information gaps across the Neotropical region.

Human-induced degradation of Chaco forests has led to a mosaic of habitats with different forest conditions, offering different habitat suitability characteristics to the native fauna. Abundance and microhabitat use of Teius teyou, Liolaemus chacoensis, Cnemidophorus ocellifer, and Tropidurus etheridgei were analyzed in the western Chaco forest of Argentina. A mature forest that has remained undisturbed for the last 30 years (Los Colorados Biological Station, LC) and a highly disturbed forest (Campo Grande, CG) were compared through diurnal visual encounter surveys along 74 1-km transects. Lizards were assigned to the microhabitat category according to the site where they were first observed (bare ground, litter/grass/herbs, shrubs or trees). T. teyou preferred bare ground in LC, avoiding litter/grass/herbs and using shrubs according to availability, whereas in CG the species preferred shrubs, avoiding bare ground. C. ocellifer used microhabitats according to their availability in LC, whereas in CG, this species preferred shrubs and avoided the remaining microhabitats. L. chacoensis preferred bare ground in LC, using litter/grass/herbs and shrubs according to availability. In CG, the species used bare ground according to availability, preferred shrubs, and avoided litter/grass/herbs. T. etheridgei preferred bare ground in LC, avoiding litter/grass/herbs, and using shrubs according to availability, whereas in CG both bare ground and shrubs were preferred, and litter/grass/herbs was avoided. Our results suggest that microhabitat selection by lizards in the Chaco is a plastic process influenced by vegetation structure, and probably regulated by lizards' thermal requirements.

No abstract

Las bandadas mixtas (BM) son asociaciones de diferentes especies que se forman como estrategia para aumentar la eficiencia de forrajeo y/o reducir el riesgo de predación. El objetivo del presente trabajo fue determinar la variación en la composición y estructura de las BM, a lo largo del gradiente latitudinal de la selva montana de las Yungas argentinas. Ocho sitios fueron estudiados a lo largo de los 700 km de distribución de las Yungas en el país. En cada sitio se establecieron 10 transectas visitadas de junio a septiembre entre 2006 y 2009. Se determinó riqueza y abundancia de las BM. Se observaron 325 BM, con un total de 72 especies participantes y una abundancia de 2480 individuos. Diecisiete familias estuvieron representadas, 15 del Orden Passeriformes y 2 no-Passeriformes (Trochilidae y Picidae). Las familias más importantes fueron Tyrannidae (12 especies), Furnariidae (10 especies) y Thraupidae (nueve especies). Doce especies fueron omnipresentes en los ocho sitios de estudio y las especies que tuvieron los valores de ocurrencia más altos fueron: Syndactyla rufosuperciliata, Phylloscartes ventralis, Chlorospingus flavopectus y Myioborus brunniceps. Aunque los análisis estadísticos multivariados indicaron que el gradiente latitudinal no tendría influencia directa sobre las características estructurales (número de bandadas, número de especies e individuos involucrados) de las BM, la composición de especies varió en asociación con la latitud. Algunas causas asociadas al gradiente, como el clima local, características de la vegetación y disponibilidad de alimento, podrían influir en los resultados obtenidos.

No abstract