NOTE Forward et al.: Swimbladder deflation in Brevoortia tyrannus 



643 



data were arcsin-transformed. A Z-statistic was used 

 to test differences between two proportions, and a 

 Student's /-test was used to compare mean values of 

 swimbladder volume (Walpole, 1974). 



Results 



To determine whether deflation was evident at the 

 beginning of the light phase without the onset of light, 

 percent deflation (Fig. 1) and swimbladder volume 

 (Fig. 2) were measured in larvae kept in continuous 



12 13 14 15 



Total Length (mm) 



16 



Figure 2 



Swimbladder volume for different size larval Atlantic 

 menhaden, Brevoortia tyrannus, with (air) and without 

 (no air) access to the air-water interface. The times (e.g. 

 3 hours) indicate the time in darkness after the begin- 

 ning of the dark phase when larvae were sampled. Means 

 are plotted and the average sample sizes for calculating 

 the means in each experiment are as follows: air for 17- 

 23 hours=19; air for 14 hours=9; air for 3 hours=19; no 

 air for 14 hours=7; no air for 23 hours=6. The asterisks 

 indicate the mean volume was significantly different 

 (P<0.05; Student's /-test) from the mean volume 3 hours 

 after the beginning of the dark phase (air, 3 hours). 



darkness. When larvae had access to the air-water 

 interface, there was a significant (P<0.05; Z-test) 

 decrease in the percentage of fish with deflated 

 swimbladders between three hours after beginning 

 of the dark phase (23%) and two hours after the time 

 for beginning of the light phase (8%). This signifi- 

 cant decrease remained throughout the time for the 

 light phase. In contrast, when fish had no access to 

 the air-water interface, the percent deflation did not 

 change significantly (Z-test) over these time inter- 

 vals (Fig. 1). Thus, larvae do not sequentially inflate 

 their swimbladders at sunset and then deflate them 

 by gas diffusion or active processes by sunrise. Be- 

 cause the percentage of larvae with a deflated 

 swimbladder decreased through the night when lar- 

 vae had access to the air-water interface, larvae ap- 

 pear to continue to inflate their swimbladders when 

 in darkness. Measurements of swimbladder volume 

 (Fig. 2) support this suggestion. 



The swimbladder volumes over time in darkness 

 with and without access to the air-water interface 

 (Fig. 2) were compared to volumes after inflation 

 at the beginning of the night phase (air, 3 hours) to 

 indicate volume changes due to bubble ingestion 

 and removal, and gas diffusion. Swimbladder vol- 

 ume increased in darkness when fish had access to 

 the air-water interface and decreased when they 

 lacked access. These changes were apparent after 

 14 hours in darkness but only become statistically 

 significant (P<0.05; Student's /-test) after 17-23 

 hours (Fig. 2). 



Thus, larvae with access to the air-water inter- 

 face continued to actively inflate their swimbladder, 

 whereas swimbladder volume slowly decreased in 

 larvae that lacked this access. Nevertheless, shortly 

 after the beginning of the light phase (Fig. 2; no 

 air, 14 hours), the volumes in larvae without ac- 

 cess to air were not significantly lower (P>0.05; 

 Student's /-test) than levels after inflation at the 

 beginning of the dark phase (air, 3 hours). There- 

 fore, diffusion of gas from the swimbladder played 

 a very small role in normal deflation at sunrise. 



Relation of swimbladder deflation to light 



Larvae deflated their swimbladders upon exposure 

 to light. This response was demonstrated in the 

 initial experiment, in which, at the end of the night, 

 larvae were either maintained for three hours in 

 darkness or exposed to white light for this time. 

 The percentage of larvae with a deflated swim- 

 bladder in darkness was 12% (rc=75), whereas the 

 percentage in light was significantly (P<0.001; Z- 

 test) greater at 84% (n=25). Further studies indi- 

 cated percent deflation depended upon light inten- 



