Forward et al.: Swimbladder inflation of Brevoortia tyrannus 



255 



& Sanchez ( 1976): V = 4/3 4ii ab 2 , where b = half the 

 bubble width and a = half the bubble length. The total 

 volume of all bubbles was used as the swimbladder 

 volume. 



There were three sets of experiments. The first was 

 designed to determine the relationship between light 

 intensity and swimbladder inflation. Larvae were re- 

 moved from the rearing tanks just before the beginning 

 of the dark phase and maintained under approximately 

 the same light intensity (cool white fluorescent 

 lamps: intensity = 1.7xl0 15 photons crrr 2 s ' ). A con- 

 trol group of the larvae was measured for total length 

 (TL), presence of gas bubbles in the swimbladder and 

 alimentary canal, and size of the bubbles. Similar stan- 

 dard measurements were made on the remaining lar- 

 vae after being exposed to the test light intensity for 

 3h. The light-stimulus source was a 300 W incandes- 

 cent lamp filtered to the blue region with a Corning 4- 

 96 filter. The transmitted wavelengths encompassed the 

 major spectral-sensitivity maxima of most fish (e.g., 

 Munz 1958, McFarland & Munz 1975). All intensities 

 below the maximum level were controlled by neutral 

 density filters. Each larva was only measured once. 



Since the total-length range of larvae was 9-20 mm, 

 an important question is whether all sizes were equally 

 responsive to lighting changes that induce swimbladder 

 inflation. An answer to this question was provided by 

 considering ( 1 ) the percentage of different-sized larvae 

 that inflate their swimbladder before and after expo- 

 sure to new lighting conditions, and (2) the percentage 

 of the total number of larvae sampled in each size- 

 category 



The second set of experiments was designed to deter- 

 mine the time-course for swimbladder inflation. Larvae 

 were again removed from the rearing tank near the end 

 of the light-phase and illuminated at about the intensity 

 used for rearing. Standard measurements were made on 

 a control sample of larvae. The remaining larvae were 

 put into darkness at the beginning of the dark-phase, 

 and standard measurements made on subsamples at vari- 

 ous time-intervals (5, 15, 30, 60, 90, 120 min). 



The third experiment tested for the presence of an 

 endogenous rhythm in swimbladder inflation. A large 

 group of larvae was removed from the rearing tank 

 at the beginning of the light phase and placed in 

 constant temperature (22°C) and light (cool-white 

 fluorescent lamps: intensity = 1.7X10 15 photons cm -2 

 s _1 ) both of which were similar to the light intensity 

 and water temperature of the rearing tanks. Standard 

 measurements were made at 3 h intervals on a 

 subsample of larvae. At the beginning of each of these 

 measurements, a similar subsample was placed in dark- 

 ness and standard measurements made 2h later. A 

 change in the percent inflating over the day would 

 suggest the presence of an endogenous rhythm. 



The percentage of larvae with inflated swimbladders 

 and the total volume of bubbles in each larva were 

 calculated for each observation within each experiment. 

 Means, standard deviations, and standard errors of 

 percent data were calculated after the data were arcsin- 

 transformed. A Z statistic was used to test differences 

 between two proportions, while a Student's t-test was 

 used to compare mean values (Walpole 1974). 



Results 



Relation of swimbladder inflation to light 

 intensity 



The percentage of larvae inflating their swimbladder 

 under all experimental conditions changed with size 

 (Fig. 1). Prior to exposure to new light conditions, the 

 percent of larvae with inflated swimbladders remained 

 relatively low for all sizes of larvae. In contrast, when 

 subjected to a decreased light intensity, the percent 

 response changed dramatically with size. None of the 

 9mmTL larvae inflated their swimbladder, and only a 

 low percentage of lOmmTL larvae showed inflation 

 (Fig 1). Percent inflation increased as size increased, 

 reaching 100% for larvae >17mmTL. The following 

 study will focus on larvae 11-16 mm in length, be- 

 cause the percent inflation can vary with lighting 

 condition. 



Equal numbers of larvae were not sampled in each 

 size-category ( Fig. 2), as most larvae were 11-12 mmTL 

 and there were few >15mmTL. This size distribution 



10 12 14 16 18 



Total Length (mm) 



20 



Figure 1 



Percentage of larvae inflating their swimbladder versus size 

 for all larvae in the swimbladder inflation versus light inten- 

 sity experiment (Fig. 3). The dashed line shows the percent- 

 ages before exposure to different light conditions, and the 

 solid line is the experimental percentages. 



