7048 - Laserfiche WebLink

Home Browse Search

Methods and Materials Microhabitat selection experiments were conducted from fall, 1986 through summer, 1987 with an artificial stream chamber measuring 1.8 m by 1.2 m by 0.9 m. Water was conveyed into the chamber via siphons, and baffles were placed at each end to reduce the flow through the chamber to zero. The habitat portion of the chamber was constructed so that it could be separated into two cells of equal size by raising a flexible nylon mesh attached to the bottom. In one cell clumps of vegetation were fixed and the other cell was left open with no cover. Chubs were introduced into the chamber and allowed to acclimate and distribute themselves, after which the partition was quickly raised from a distance. The number of chubs in each cell was recorded and the partition was lowered to allow free access once again. This experiment was done at dawn, noon, near sunset, and at midnight, and then replicated the following day. At the end of the second day the positions of the vegetation and open cells were reversed and the experiments were repeated for two more days. One experimental treatment consisted of 16 tests over a four day period and the treatments were repeated during each season. A total of 64 tests was made. Chi square analysis was used to test the null hypothesis that cell selection was random. Distribution of chubs in the CD headspring was examined by placing minnow traps (122 cm by 43 cm diameter with a 12.7 mm by 12.7 mm mesh) at locations in the headspring on several occasions during the course of the study. The locations represented different habitat types with different abundances of vegetation (Fig. 2). The traps were set after sunset, fished overnight, and inspected after sunrise the following day. Gut contents from 35 chubs taken from the CD headspring on four seasonal samplings were examined. Food categories were identified, counted, and weighed, and an index of relative importance (I.R.I) was derived by adding the percent number and percent weight, then multiplying the result by the frequency of occurrence (Pinkas et al. 1971). Results The chubs used in the 64 tests overwhelmingly selected for the vegetation cell (Fig. 3). Chi square analysis supported this observation: in 59 of the 64 cases, the null hypothesis was rejected (p = 0.05) and a significant majority of the chubs in each case selected for the vegetation cell. Of the remaining five cases, the chubs selected for the open cell in one case, and selection was random in four cases (p = 0.05). These five cases were all tests conducted during the midnight period where the chubs showed a higher incidence of open cell selection (Fig. 4). 16