View larger version (49K):
[in a new window]
|
Fig. 1. Experimental design used to assess the importance of natural predators on
cricket behaviour and survival. For each treatment, cages were built so as to
allow for a single type of predation. (A) Cages for the `air' treatment had an
open top to allow flying predators to access the encaged crickets, and a
plastic border at the top of the cages on which fluon was applied to prevent
escapes or unwanted intrusions. (B,C) Cages for the `spider' (B) and the
`control' (C) treatments were closed at the top with mosquito screen. All
cages had a wooden frame of 0.5 mx0.5 mx0.25 m covered with clear
plastic on the sides and a fabric bottom. In total, we used 225 juvenile
crickets of stage 2-3, 225 juvenile crickets of stage 4-5 and 90 adults (45
males and 45 females), all caught from the local population. Behavioural and
survival data were collected once a week from each cage. The experimenter
first watched the inside of the cages from above for 1 min and counted the
number of crickets from each age category that were visible at the surface of
leaves, and then made an exhaustive count of the number of crickets still
alive. The computed proportion of individuals visible at the surface of leaves
was compared among treatments using ANOVA. To ensure that the effect seen on
the proportions was not a byproduct of the total number of individuals still
alive, proportions were weighted by numbers still alive before being compared
among treatments. Given that total numbers still alive were not found to
differ significantly among cages, the weight used was the mean number still
alive for each cage over weeks. Survival data were directly obtained through
the exhaustive count of crickets. Survival data were compared among treatments
independently for juveniles of stage 2-3, juveniles of stage 4-5, and adults,
using Tarone-Ware survival tests.
|
