Influence of venom availability on a spider's prey‐choice behaviour

Abstract

Summary A neotropical hunting spider, Cupiennius salei , was studied to test whether spiders optimize venom use. A prey‐choice experiment was performed with several cockroach species. These easily accepted prey species are especially suitable for such an experiment because the LD 50 varies over a range of 40. Cupiennius salei spiders had either full venom glands (control) or their venom glands were experimentally emptied by electrical milking or by allowing them to bite into a series of crickets. Two pairs of cockroaches were used: Blatta orientalis (LD 50 : 0·4 nl mg −1 ) vs Nauphoeta cinerea (LD 50 : 17·5 nl mg −1 ) and Periplaneta australasiae (LD 50 : 0·75 nl mg −1 ) vs Periplaneta americana (LD 50 : 10 nl mg −1 ). These pairs provide a difference of 40‐fold and 10‐fold, respectively, in venom sensitivity. Cupiennius salei preferred B. orientalis significantly over N. cinerea when the venom glands were empty, but not when the full amount of venom was available. No effect was evident for the other pair, indicating that a factor of approximately 10 is too low. We conclude that the spiders know how much venom is available in their venom glands and that they select prey items that are appropriate for the amount of venom available.