Published online by Cambridge University Press: 20 January 2017
Giant ragweed seeds have high nutritional value, consisting of 47% crude protein and 38% crude fat, and may be an important food source for rodent and invertebrate populations in agricultural and early successional ecosystems. We investigated temporal patterns of postdispersal giant ragweed seed predation on the soil surface of a no-tillage cornfield as affected by involucre (seed dispersal unit) size and presence or absence of crop residue. Cage exclusion experiments indicated that rodents and invertebrates were the principal predators of giant ragweed seed, and total predation of involucres over a 12-mo period beginning in November was 88%. Rodents were the greatest predators of giant ragweed involucres during fall and winter, and cumulative predation by February 1 in treatments with rodent access ranged from 39 to 43%. In contrast, giant ragweed involucre predation by invertebrates occurred mainly from May 1 to November 1. When rodent access to involucres was prevented, total involucre predation by invertebrates over a 12-mo period ranged from 57 to 78%. Rodents showed an initial preference for large involucres (> 4.8-mm diameter), and invertebrates preferred small involucres (< 4.8-mm diameter). Involucres covered with corn plant residue underwent less predation by rodents from November to February than uncovered involucres, but residue cover had no effect on seed predation by invertebrates. In a laboratory feeding trial, the carabid Harpalus pensylvanicus preferred seed of smooth pigweed and yellow foxtail to giant ragweed seed, suggesting that giant ragweed seed is an incidental rather than a preferred food source for some carabids. Because giant ragweed exhibits relatively low fecundity and short seed bank persistence, results of this study suggest that postdispersal predation may directly reduce giant ragweed recruitment the next year by reducing new seed bank inputs. However, seed losses from predation alone may be insufficient to maintain giant ragweed populations below economic threshold levels in no-tillage cornfields.