Shrinkage and growth compensation of artificially damaged common sunflowers (Helianthus annuus) were studied in central North Dakota during 1981-82 in an effort to increase accuracy of estimates of blackbird damage to sunflowers. In both years, as plants matured damaged areas on seedheads shrank at a greater rate than the sunflower heads themselves. This differential shrinkage resulted in an underestimation of the area damaged. Sunflower head and damaged-area shrinkage varied widely by time and degree of damage and by size of the seedhead damaged. Because variation in shrinkage by time of damage was so large, predicting when blackbird damage occurs may be the most important factor in estimating loss. Yield/occupied area was greater (P < 0.05) for damaged than undamaged heads and tended to increase as degree of damage inflicted increased, indicating growth compensation was occurring in response to lost seeds. Yields of undamaged seeds in seedheads damaged during early development were higher than those of heads damaged later. This suggested that there was a period of maximal response to damage when plants were best able to redirect growth to seeds remaining in the head. Sunflowers appear to be able to compensate for damage of -<15% of the total head area. Estimates of damage can be improved by applying empirical results of differential shrinkage and growth compensation. J. WILDL. MANAGE. 50(3):513-520 Sunflowers, grown primarily for the oil seed market, have become an increasingly important cash crop in the United States in recent years. In North Dakota, South Dakota, and Minnesota about 21,450 km2 of sunflowers were harvested in 1979. Production losses occur when birds, especially red-winged blackbirds (Agelaius phoeniceus), yellow-headed blackbirds (Xanthocephalus xanthocephalus), and common grackles (Quiscalus quiscula) damage ripening sunflower heads. In 1980, sunflower losses to blackbirds in North Dakota, South Dakota, and Minnesota were estimated to be 32,100 metric tons (R. L. Hothem, pers. commun.) worth $7.9 million. Estimates of damage losses generally are conducted near harvest by measuring the damaged (vs. undamaged) areas on a sample of sunflower heads. Sunflower heads shrink as they mature, and recently, investigators have discovered that d maged areas on heads shrink more than undamaged areas, resulting in underestimates of damage (J. L. Guarino and D. F. Mott, pers. commun.). To more accurately assess sunflower eed loss to blackbirds, refinements of damage estimation techniques are necessary. Yet even when differential shrinkage between damaged and undamaged areas of sunflower heads is taken into account, damage estimates may not reflect true yield loss by weight. Because blackbird damage peaks before seeds are fully mature (Cummings and Marsh 1980), damaged plants may be able to direct energy to the remaining undamaged seeds, compenThis content downloaded from 157.55.39.223 on Wed, 24 Aug 2016 05:11:15 UTC All use subject to http://about.jstor.org/terms 514 SUNFLOWER DAMAGE ESTIMATES * Sedgwick et al. J. Wildl. Manage. 50(3):1986 sating for part or all of the loss in yield caused by blackbirds. This apparently occurs to some extent in corn (Woronecki et al. 1976, 1980), where low levels of damage do not result in a decrease in yield. If sunflowers can compensate for blackbird damage, it may be necessary to adjust estimates of losses downward. The objectives of this study were to provide information to refine assessments of sunflower loss to blackbirds. To do so we evaluated the effects of differential shrinkage between damaged and undamaged sunflower head areas on damage estimates and determined how shrinkage varies with time of damage, head size, and degree of damage. We also investigated the extent of growth compensation in sunflowers and determined the effects of sunflower time of damage, head size, and amount of damage on growth compensation. We would like to thank P. Y. Sweanor, L. J. Barrie, and S. A. Boyle for their assistance in the field and laboratory and N. P. Groninger for computer programming advice. We thank R. T. Sterner and J. L. Cummings (Denver Wildl. Res. Cent.) for their field and technical assistance. We also gratefully acknowledge Denver Wildl. Res. Cent. biologists J. L. Guarino, C. E. Knittle, T. L. Burst, J. F. Besser, and G. M. Linz for technical assistance and manu-