BEITINGER: THERMOREGULATORY BEHAVIOR OF BLUEGILL 



midpoint of the preferred range was 31.2°C 

 (Table 1). 



Pretreatment and posttreatment comparison of 

 preferred range midpoints for individual fish are 

 illustrated in Figure 1. Of the 20 control and cold- 

 treated fish ( lb, a), 18 had posttreatment midpoint 



33- 

 32- 



< 1 < 1- 



30 31 32 33 



PRETREATMENT MIDPOINT (°C) 



Figure 1. — Mean preshock and postshock midpoints of preferred 

 range for individual bluegills in each of the treatment groups. 

 Points falling on the 45° line indicate no change in midpoint 

 temperatures. 



temperatures within 0.3''C of their pretreatment 

 values. None of the individual control fish had 

 significant pretreatment and posttreatment 

 changes in mean avoidance temperatures U-test, 

 P>0.05). Among the cold-treated fish, two had 

 significant downward changes in lower avoidance 

 temperatures and one had a significant downward 

 change in its upper avoidance temperature. In the 

 control and cold-treated groups, five fish each had 

 lower posttreatment midpoint temperatures; 

 however, there were no significant trends (Wil- 

 coxon matched pairs, signed ranks; controls and 

 cold-treated P>0. 10). 



Eleven of the twelve heat-treated fish had 

 higher posttreatment midpoint temperatures 

 (Figure Ic). This trend was highly significant 

 (Wilcoxon matched pairs, signed ranks, P<0. 01). 

 The mean posttreatment midpoint temperatures 

 for heat-treated fish during each of the 3 days were 

 31.9°, 32.0°, and 31.9°C, indicating no return to- 

 wards the pretreatment preference level. 



Whereas all of the control and cold-treated fish 

 survived the treatment process, 7 of the 20 fish 

 (35%) exposed to 36.1°C died during treatment. 

 All fish that died lost equilibrium early in the 

 treatment and were dead within 5 min. The mean 

 pretreatment midpoint temperature of those that 

 died was significantly lower than that of the sur- 

 vivors (Mann- Whitney U testP<0.05); however, 

 temperatures experienced immediately prior to 

 exposure were the same for both groups. 



Although visual observations during the post- 

 treatment period of this study were limited to 

 avoid disturbing the fish, the typical immediate 

 posttreatment behavior of both the heat- and 

 cold-treated fish was submissive; often fish were 

 hiding behind objects in their experimental 

 tanks. However, at the feeding time, 4.5 h follow- 

 ing exposure, nearly all fish actively fed. 



A distinct diurnal pattern of activity was ob- 

 served for each of the treatment groups through- 

 out the 6-day experiment (Figure 2). Daytime 

 hourly activities were typically two to three times 

 higher than nighttime activities. The median 

 activity (tunnel passes) of the cold- and, partic- 

 ularly, heat-treated fish dropped appreciably the 

 hour following exposure (Figure 2). For general 

 activity comparisons (Figure 3a, b) diurnal and 

 nocturnal periods were separately analyzed 

 (dawn and dusk excluded). With the pretreatment 

 activity of each group serving as its own control, 

 a series of Mann-Whitney U tests, with ties 

 correction and z transformation were performed 



1089 



