Effect of sperm depletion and starvation on female mating behavior in the water strider,Aquarius remigis

The resolution of intersexual conflict over mating should be dependent on the current state of each individual. In this study, I used a factorial design to examine the influence of two physiological factors, sperm depletion and food deprivation, on resistance to mating by females of the water strider, Aquarius remigis. Females employ several different mate-resisting tactics during an encounter with a male. Five measures of female resistance to mating were identified: jumping, rolling, dunking, time spent dunking, and struggle duration. Jump, roll, and dunk rates were highly correlated with each other and combined into one metric of resistance to mating (PC1) using principal components analysis. Time per dunk (T/D) and struggle duration were also analyzed. Discrete male behaviors during the struggle could not be identified. Two measures of female resistance, PC1 and T/D, were significantly lower in sperm-depleted females than in sperm-replenished females. Struggle duration did not differ between the two treatments. Starvation had no effect on any of the measures of resistance. Sperm depletion significantly enhanced the probability of mating (54% vs. 24% for replenished females), while starvation had no effect on the probability of mating. I pooled all the females and compared females that mated with those that did not mate. Nonmating females resisted significantly more than mating females in all three measures of resistance. Path analysis indicated that PC1 was the only measure of resistance that was significantly negatively related to the probability of mating. Almost half (46%) of sperm-depleted females showed no resistance to males, while only 3% of sperm-replenished females were nonresistant. When nonresisters were removed from the analysis, sperm depletion had no effect on any of the measures of female resistance to mating and no effect on the probability of mating. In A. remigis, female resistance appears to be a yes/no phenomenon with respect to sperm depletion and not affected directly by starvation. Received: 2 September 1994/Accepted after revision: 9 September 1995