Fighting over defense chemicals disrupts mating behavior

Abstract

Studies on intraspecific contest behavior predominantly focus on contests between individuals of the same sex, however contest behavior is also expected to occur between individuals of the opposite sex including possible mates. Here we investigate potential trade-offs between mating and fighting behavior in the turnip sawfly (Athalia rosae). Adults of this species collect chemical defense compounds (clerodanoids) directly from plants but also indirectly by nibbling on conspecifics that have already obtained clerodanoids, a highly aggressive behavioral interaction. An A. rosae individual without clerodanoids may therefore be the potential mate or attacker of an individual of the opposite sex that has gained clerodanoids. To test the effect of clerodanoids on agonistic and mating behavior we paired females and males with or without clerodanoid access in a two-way factorial design. We show that asymmetrical clerodanoid acquisition between female-male pairs causes an increase in agonistic nibbling behavior, irrespective of sex, and moreover that conflict between individuals delays mating behavior. Our study highlights the importance of investigating agonistic intersex interactions, which can occur when adults are able to acquire valuable non-reproductive resources from a potential partner.

Keywords: chemical defense; conflict; dyadic contests; fighting behavior; mating behavior.

Figure 1

(A) Outline of two-way factorial experimental design showing the four different treatment levels and number (N) of Athalia rosae pairs used in the behavioral assays. Green boxes denote those individuals who had access to clerodanoids via nibbling on Ajuga reptans leaves (C+), those without green boxes had no access to A. reptans leaves and therefore do not have clerodanoids (C−). (B) Illustration of the general hypotheses for each treatment with the saturation and direction of each arrow representing the likelihood of individuals being motivated to display a certain behavior. (C) A full list of the behaviors measured and parameters tested alongside specific predictions for each set of parameters (see Supplementary Material S2 and methods for further information on each behavior and the parameters respectively).

Figure 2

Effect of clerodanoid access (C+ = nibbled on Ajuga reptans leaf, C− = not nibbled on A. reptans leaf) on (A) the occurrence of successful copulation, (B) log time until copulation commenced, (C) the occurrence of successful nibbling, (D) whether fighting occurred before copulation, (E) the number of individuals that successfully nibbled their opponent split within treatment by whether or not they had clerodanoid access, and (F) time taken until copulation commenced depending on whether pre-copulatory agonistic behavior also occurred in adult A. rosae. Treatment levels are listed with females on the left and males on the right. Boxes in box plot show the median, the first and third quartiles (the 25th and 75th percentiles) at the hinge, and the whiskers extend to the largest or smallest value no further than 1.5 × IQR from the hinge for the upper and lower whiskers, respectively. Points with error bars in other plots are model predictions with 95% confidence intervals.