HUNTER and SANCHEZ: CHANGES IN SWIM BLADDER INFLATION 



Table 3- Percent composition of swim bladder gas of laboratory-reared northern 

 anchovy larvae sampled at night, listed in order of time of sampling. 



Linnaeus, about 24 h (Wittenberg 1958). Con- 

 sidering the evidence presented here, and the 

 apparent lack of gas secretion in clupeoid fishes in 

 general (Brawn 1962), the most tenable hypothesis 

 is that swim bladder inflation is accomplished in 

 larval anchovy by taking in air at the water 

 surface. 



Vertical Migration 



If anchovy larvae fill their swim bladders each 

 night by swallowing air, they must either remain 

 near the surface throughout the day and night or 

 migrate to the surface at dusk. 



We reexamined the original data collected by 

 Ahlstrom (1959) to determine if any evidence 

 existed for vertical movements in northern an- 

 chovy larvae. Ahlstrom (1959) made extensive 

 horizontal tows for fish larvae with opening and 

 closing nets and presented the average number of 

 larvae of all lengths at various depths. We sepa- 

 rated his original length data into two length 

 classes: larvae <11.75 mm (preserved standard 

 length) and larvae 211.75 mm for night and day 

 collections; we omitted those collections occurring 

 near dawn and dusk. Unfortunately, only 14 larvae 

 sll.75 mm were taken in the day while 279 were 

 taken at night but the depth pattern in the day 

 collections was relatively consistent. Larvae 

 < 11.75 mm were more abundant: N = 6,456, night; 

 and A^ = 331, day. 



At night, over 50% of the larvae £11.75 mm were 

 taken in the upper 10 m whereas in the day the 

 upper 10 m contained less than 10% of the larger 

 larvae (Figure 7). About 50% of the larvae <11.75 

 mm occurred in the upper 10 m, but no obvious 

 difference between day and night samples existed. 

 These results are in general agreement with those 

 of Ida (1972) who studied the vertical distribution 

 of the Japanese anchovy, E. japonicus, a closely 



10 



20 



30 40 50 

 DEPTH (m) 



60 70 



80 



Figure 7.-The vertical distribution in the sea of northern larvae 

 at night and in the day for two length classes; length <11.5 mm, 

 upper panel, and length 211.5 mm, lower panel (lengths for 

 preserved specimens). Numbers of larvae taken at each depth 

 were cumulated starting at the shallowest tow (5 m) and 

 expressed as the cumulated percent of the total larvae taken. 

 Data are from Ahlstrom (1959). 



related species that also has a diel rhythm in swim 

 bladder inflation (Uotani 1973). Ida (1972) found a 

 striking diel change in the vertical distribution of 

 E. japonicus with the maximum numbers occur- 

 ring at the surface at night and at 20 to 30 m 

 during the day with the movement to the surface 

 occurring at twilight. Examination of the size 

 frequency histograms from some of the collections 



853 



