FISHERY BULLETIN: VOL. 74, NO. 4 



llJ 

 < 

 > 



< 



u. 

 o 



UJ 



o 

 ir 



UJ 

 Q. 



70 

 60 

 50 

 40 

 30 

 20 

 iOl- 



" 1 



LABORATORY- REARED 

 LARVAE 



I 



0.005 0.025 0.045 0.065 0.085 0.105 

 SWIM BLADDER VOLUME (mm^) 



Figure 3. -Percent of northern anchovy larvae. 11.9 to 12.0 mm, 

 at night (solid bars) and in day (dashed bars) having swim 

 bladder volumes in various 0.01-mm' classes; numbers on 

 abscissa are midpoints of swim bladder volume classes. Upper 

 panel, larvae from CalCOFI ichthyoplankton collections 

 (preserved specimen length), A^ = 20 for night, and .V = 30 for 

 day. Lower panel, larvae reared in laboratory (live specimen 

 length), .V = 49 for night, and .V = 29 for day. Data from 2-h 

 after onset of dark and 2-h after onset of light were excluded in 

 laboratory-reared larvae. 



The mean swim bladder volume at night was 

 greater for wild than for laboratory-reared larvae 

 of the same length. The effect of preservation on 

 larval length for larvae of this size is not known 

 but a shrinkage of about lOSt in length in the 

 Formalin-preserved ichthyoplankton specimens 

 would account for this difference. The effect of 

 preservation on swim bladder volume is also 

 unknown. In some of the preserved specimens, we 

 noticed the bladder was filled with fluid but we did 

 not routinely make an examination of the bladder 

 contents. 



Swim Bladder Inflation and Larval Length 



The swim bladder was fully formed when larvae 

 reached 8 to 9 mm but it usually was not inflated. 

 To determine the larval size at which nightly 

 inflation commenced, night and day samples from 

 the laboratory were grouped into 1-mm length 

 classes (9.0 to 9.9 mm, 10.0 to 10.9 mm, etc.), and the 

 mean volume for day and night samples for each 



850 



class calculated, and compared using the t test. The 

 first 2 h after the onset of dark and the onset of 

 light were excluded from the classes. 



Some of the 9.0 to 9.9 mm larvae appeared to 

 have inflated swim bladders at night but the 

 night-day difference in swim bladder volume was 

 not significant (0.2>P>0.1). Mean volumes for day 

 and night samples were different in larvae 10.0 to 

 10.9 mm as were those for larvae in all succeeding 

 length classes (P<0.001). Thus, the threshold larval 

 length for nightly inflation of the swim bladder 

 occurred at about 10 mm, the point at which the 

 means for day and night volumes diverge (Figure 

 4).-^ From this point, mean volume of night sam- 

 ples increased exponentially with length whereas 

 that for day samples increased linearly. 



Relation Between Sinking Speed, 

 Swim Bladder Volume, and Larval Length 



We observed that larvae with inflated bladders 

 sank more slowly than those with uninflated 



•'Swim bladder inflation is reported to occur at 7 mm in E. 

 japoiiiciif! (Uotani 1973). Comparison of his illustrations to those 

 of Uchida et al. (1958) suggests Uotani's reported lengths are in 

 error and that E. Japan iciis also inflates the bladder at about 10 

 mm. 



020r 



to 



E 



E 



o 

 > 



Q 

 Q 

 < 



_l 



m 



CO 



18 

 016 

 14 

 12 

 010 

 008 

 006 

 004 

 002 



000 



O DAY 

 • NIGHT 



i 



\ 



^ * - ° 



I? ° 



i , I I I 



~\ — I — I — \ — I — r 



8 9 10 II 



"~T \ 1 1 1 1 1 1 1 — I \ 1 



12 13 14 15 16 17 



LENGTH (mm) 



Figure 4.- Mean swim bladder volume ± 2 SE for laboratory- 

 reared northern anchovy larvae for 1-mm classes of length 

 plotted at the midpoint of each class. Solid circles are night (first 

 2-h after onset of dark omitted) and open circles day (first 2-h 

 after onset of light omitted). 



