650 
Fishery Bulletin 95(4), 1997 
Table 5 
Results of analysis of variance performed on the percent- 
age of time spent inactive by fish A and fish B during the 
four phases of the tidal cycle, df = degrees of freedom; SS = 
sum of squares; MS = mean of squares. 
Probability 
Source 
df 
SS 
MS 
F-value 
value P 
Date 
10 
5,492.5 
549.3 
1.74 
0.089 
Fish 
1 
33.1 
33.1 
0.10 
0.747 
Tide 
3 
1,320.8 
440.3 
1.39 
0.252 
Fish x tide 3 
771.5 
257.2 
0.81 
0.490 
Error 
70 
22,117.3 
316.0 
Total 
87 
29,735.3 
ences between the tidal phases for the length of in- 
activity bouts of fish Aand fish B (F 3 70 =1.57, P=0.204; 
F 1 ? =0.4, P= 0.756), percent of time spent inactive by 
fish C and fish B (P 356 =0.50, P= 0.685; F 13 =0.79, 
P=0.505) and length of inactivity bouts of fish C and 
fish B (P 356 =0.82, P=0.4986; F x 3 =1.94, P=0.134). 
Time of 1 year, i.e. seasonal factors, accounted for 
58.7% of the variation in the percentage of the diur- 
nal activity period spent inactive, i.e. the amount of 
time spent in shelter between the commencement and 
cessation of daily activity (Table 4). Cunners spent an 
increasing proportion of the diurnal period inactive 
from June through November. This factor accounted 
for 12.7% of the variation in length of inactivity bouts 
(which increased over time as well) and for 6.6% and 
22.3% of the variance in onset and cessation of activ- 
ity, respectively (Table 4). The trend was for activity 
to begin later in the morning and to end earlier in 
the evening (in relation to sunrise and sunset) as the 
season progressed. Time of year accounted for 92.3% of 
the variation in duration of diurnal activity. As the sea- 
sons progressed and the photoperiod became progres- 
sively shorter, there was a corresponding decrease in 
the duration of diurnal activity (Fig. 1). 
Thus the duration of cunner diurnal activity, per- 
centage of time inactive, and length of inactivity 
bouts were all closely related to day length. By com- 
bining data from all subjects, the average elapsed 
time between onset and cessation of activity (i.e. 
duration of diurnal activity) was 16.5 h (n= 38, 
SE=0.10) during June-July, 13.5 h (n= 8, SE=0.19) 
during August-September, and 11.0 h (n=27, 
SE=0.25) during October-November. Cunner spent 
22.9% (n=35, SE=15.5) of the diurnal period inactive 
between June and July, 47.0% (n=13, SE 24.3) of the 
diurnal period inactive in August-September, and 
71.8% (n= 30, SE=13.2 ) of the diurnal period inactive 
during October-November; length of inactivity bouts, 
on the other hand, increased from an average of 6.4 
min (n=64, SE=0.75) in June— July to 9.3 min (n= 15, 
SE 1.56) in August-September, to 12.7 min (tz= 41, 
SE=1.14) in October-November. 
These data were used to calculate the number of 
hours a female cunner spent out of 
its shelter per day. On average, dur- 
ing June-July, female cunners were 
active for 12.5 h/day, during August- 
September 7 h/day, and during Octo- 
ber-November only 3 h/day. The re- 
mainder of the year was spent in win- 
ter torpor. 
There were no significant differ- 
ences (P>0.05, paired comparison t- 
test) in any of the activity parameters 
for the single female cunner tracked 
during both the prespawning and 
spawning periods. 
Discussion 
20 
15 
io » 
Figure I 
Relation between length of daily photoperiod and duration of diurnal activity 
(US) for eight female cunners tracked in Conception Bay, Newfoundland, be- 
tween June 17 and November 24. Maximum daily seawater temperature at the 
study site is also shown. 
Pottle (1979) reported that in New- 
foundland territorial male cunners 
undergo periods of daylight quies- 
cence under cover. Whoriskey (1983) 
also observed cunners in Massachu- 
setts underneath boulders at those 
times during the day when they were 
