258 



Fishery Bulletin 91(2). 1993 



30 50 70 90 



Time (min)in Darkness 



110 



Figure 6 



Percentage of larvae 11-16 mmTL filling their swim- 

 bladders after different times in darkness. Number under 

 each point is the sample size, and asterisk is the first 

 time that a proportion was significantly greater (p<0.05l 

 than the proportion of larvae with inflated swimbladders 

 initially in light, which is plotted at time zero. 



enous rhythm. After this time in constant light, 

 variation in inflation after introduction to dark was 

 not evident in the present study. There are two 

 possible explanations for this result. First, the 

 rhythm could fail to continue be- 

 cause larvae were kept in constant 

 light, a condition that frequently 

 suppresses an endogenous rhythm 

 (Hastings et al. 1991). Second, the 

 rhythm could consist of one cycle 

 in which inflation is suppressed 

 during the light-phase and larvae 

 become "ready" to inflate at the be- 

 ginning of the dark-phase. Readi- 

 ness then continues until a dark cue 

 is received, which resets the endog- 

 enous clock. 



Clearly, Atlantic menhaden lar- 

 vae are adapted for swimbladder 

 inflation at sunset. Their rhythm in- 

 dicates they are most responsive to 

 a light-intensity decrease at this 

 time and most inflation occurs 

 within 20 min. Such a dramatic re- 

 sponse suggests swimbladder infla- 

 tion has an important functional 

 advantage. 



Menhaden larvae are negatively 

 buoyant even with a fully inflated 

 swimbladder. Nevertheless, infla- 

 tion reduces their sinking rate (Hoss 

 et al. 1989). Past investigators have 



suggested that swimbladder inflation acts as an energy- 

 saving mechanism, allowing larvae to expend less energy 

 for maintaining their position in the water column at 

 night when they are not feeding ( Hunter & Sanchez 1976). 

 During the day, a fully inflated swimbladder may reduce 

 the speed of movement and, thereby, the effectiveness of 

 prey capture and predator avoidance. 



In addition, Uotani ( 1973) proposed that inflation allows 

 larvae to decrease their movement at night, which serves 

 to reduce detection by predators that hunt by vibrations, 

 such as chaetognaths. Field studies show some indication 

 that menhaden larvae undergo reverse diel vertical migra- 

 tion (DVM) in which they descend in the water column 

 near sunset and ascend near sunrise (Hoss et al. 1989). 

 Chaetognaths exhibit the opposite pattern of nocturnal 

 DVM (Pearre 1973, Sweatt & Forward 1985). Reverse DVM 

 is proposed as a mechanism for avoiding zooplankton preda- 

 tors that undergo nocturnal DVM (Ohman et al. 1981, 

 Neill 1990). A slower descent rate at sunset by menhaden 

 larvae due to inflated swimbladders may reduce detection 

 by chaetognaths that are ascending toward the surface. 

 Since the percentage of menhaden larvae with inflated 

 swimbladders increases with size, the importance of re- 

 duced sinking rate for predator avoidance may increase 

 with size. The threat of predation to menhaden larvae is 

 probably reduced during their ascent at sunrise because 

 the descending chaetognaths have been feeding all night. 



0800 1000 1200 1400 1600 1800 2000 2200 1000 

 Time (hrs) 



Figure 7 



Percentage (A) of larvae ll-16mmTL that filled their swimbladders before ( ) 



and after ( ) exposure to darkness for 2h over the solar day. The swimbladder 



volume of 11 mmTL larvae (B) after exposure to darkness is also plotted against 

 time in the solar day. Means and standard errors are plotted. Number near each plot 

 is the sample size. Arrow indicates the times of the beginning of the dark phase of 

 the rearing LD cycle. 



