PRESSURE SENSITIVITY OF 

 ATLANTIC HERRING, CLUPEA HARENGUS L., LARVAE 1 



David R. Colby, 2 Donald E. Hoss, 2 and J. H. S. Blaxter 3 



ABSTRACT 



Larval Atlantic herring, Clupea harengus harengus L., are known to change their vertical distribu- 

 tion by day and night, but it is not clear if they can sense their depth by perception of hydrostatic 

 pressure. Two experiments were designed to test whether herring larvae would respond to imposed 

 pressure changes by making appropriate compensatory vertical movements and whether such 

 ability could be related to the development of the bulla system (stage I, bulla absent; stage II, bulla 

 liquid-filled; stage III, bulla gas-filled). In the first experiment, pairs of larvae were exposed to a 

 fixed sequence of pressure changes (&P) from ±13 cm H 2 to ±66 cm H 2 0. Members of simul- 

 taneously tested pairs tended to be influenced by one another when responding to pressure change. 

 The response of stage-I larvae tended to first increase and then decrease over a20-min test period for 

 a given AP. Stage-II and stage-Ill larvae showed better performances in compensating for imposed 

 pressure changes than did stage I. Larvae exposed to a sudden pressure increase of 5 atm (atmos- 

 pheres) (5,000 cm H 2 0) before the experiment did not perform as well as those not so exposed, but the 

 differences were not statistically significant. A second experiment tested the response of individual 

 larvae to randomized sequences of pressure changes. Stage-Ill larvae moved most frequently to 

 compensate for the pressure changes, but stage-I and stage-II larvae also responded to changes in 

 pressure. Both experiments show that herring larvae of all three stages compensate for applied 

 pressure changes by moving up when pressure is increased and down when it is decreased, but that 

 they rarely move sufficiently far in the vertical plane to fully compensate. 



Larval fish are known to change their vertical 

 distribution diurnally. Although this behavior is 

 probably controlled by changes in light inten- 

 sity, it is not clear whether hydrostatic pressure 

 perception is important in limiting or controlling 

 the depths reached at different stages of the ver- 

 tical migration cycle. A few workers (e.g., Qasim 

 et al. 1963) have shown that fish larvae can re- 

 spond to pressure changes; in particular, Bishai 

 (1961) and Blaxter and Denton (1976) have 

 shown that Atlantic herring, Clupea harengus 

 harengus L., larvae are pressure sensitive. 



The most likely site for a pressure receptor is a 

 gas-filled structure, such as a swim bladder, 

 which, if compliant, undergoes large changes in 

 volume during vertical movements (10 m change 

 of depth being equivalent to 1 atmospheric pres- 

 sure). However, clupeoids, together with some 

 other groups such as mormyrids, have gas-filled 

 bulla. In herring the bulla appears to be sensitive 

 to pressure changes (Allen et al. 1976; Denton 



'Contribution No. 82-11B, from the Southeast Fisheries Cen- 

 ter Beaufort Laboratory, National Marine Fisheries Service, 

 NOAA, Beaufort, N.C. 



Southeast Fisheries Center Beaufort Laboratory, National 

 Marine Fisheries Service, NOAA, Beaufort, NC 28516. 



3 Dunstaffnage Marine Research Laboratory, Oban, Scot- 

 land. 



Manuscript accepted January 1982. 

 FISHERY BULLETIN: VOL. 80, NO. 3. 1982. 



and Blaxter 1976). In herring the prootic bulla 

 has two parts: one filled with gas, the other with 

 perilymph. The two parts are separated by an 

 elastic membrane. This membrane responds to 

 pressure changes, driving the perilymph in or 

 out of a fenestra, which is situated close to the 

 utriculus of the inner ear. The gas-filled part of 

 the bulla is also connected to the swim bladder by 

 a very narrow gas duct. This connection allows 

 the prootic membrane to adapt to slow changes of 

 pressure. If the pressure increases, the mem- 

 brane bows in and being elastic tends to return to 

 its resting position. The swim bladder wall is 

 compliant and the pressure differential created 

 along the gas duct causes gas to flow into the 

 bulla from the swim bladder. If the pressure de- 

 creases, the membrane bows outward (into the 

 perilymph space) and gas flows from the bulla 

 back to the swim bladder. 



In the fully functional system described above 

 the bulla may respond to hydrostatic pressure 

 changes, but because the system adapts in 15-30 

 s, there will be no perception of absolute pres- 

 sure. In the very early larval stages of herring 

 (from hatching to 18 mm TL) no bulla is present; 

 the bulla appears at about 18 mm and usually is 

 filled with gas by 26 mm. The swim bladder is 

 not fully formed until 35 mm or more (Blaxter 



567 



