We compared the efficacy of 400-m line transects and sets of three point counts at detecting avian richness and abundance in bottomland hardwood forests and intensively managed cottonwood (Populus deltoides) plantations within the Mississippi Alluvial Valley. We detected more species and more individuals on line transects than on three point counts during 218 paired surveys conducted between 24 March and 3 June, 1996 and 1997. Line transects also yielded more birds per unit of time, even though point counts yielded higher estimates of relative bird density. In structurally more-complex bottomland hardwood forests, we detected more species and individuals on line transects, but in more-open cottonwood plantations, transects surpassed point counts only at detecting species within 50 m of the observer. Species richness and total abundance of Nearctic-Neotropical migrants and temperate migrants were greater on line transects within bottomland hardwood forests. Within cottonwood plantations, however, only species richness of Nearctic-Neotropical migrants and total abundance of temperate migrants were greater on line transects. Because we compared survey techniques using the same observer, within the same forest stand on a given day, we assumed that the technique yielding greater estimates of avian species richness and total abundance per unit of effort is superior. Thus, for monitoring migration within hardwood forests of the Mississippi Alluvial Valley, we recommend using line transects instead of point counts.