Milton and Blaber: Sexual maturity and spawning of tuna baitfish in the Solomon Islands 



229 



< 



(a) Munda 



eggs 



I t , , , T , fOvT-f-r-r . 



0.3 



0.2 



0.1 



Larvae 



AMJ JASONDJFMAM 



(b) Vona Vona 



AMJJASONDJFMAM 



7 



Pt 'l 'l' l ff^ i ^ T -r-T^ 



AMJJASONDJFMAM 



AMJJASONDJFMAM 



0.0-'-' — r 



AMJJASONDJFMAM 

 Time 



AMJJASONDJFMAM 

 Time 



Figure 8 



Variation in monthly abundance of Encrasicholina eggs and larvae ( ± range) at three sites in the Solomon 

 Islands, April 1988-May 1989. 



with wind and time since rain. At Vona Vona there was 

 a negative correlation with wind strength and positive 

 correlation with zooplankton biomass and tide range. 

 Spawning at Tulagi correlated most strongly with time 

 since heavy rainfall (>25mm), temperature, and zoo- 

 plankton biomass (Table 4). When all data were in- 

 cluded, time since heavy rain was the only significant 

 correlate. 



Stepwise regression analysis of the six independent 

 proximate stimuli (Table 5) showed a much poorer fit. 

 The only site that showed a significant relationship was 

 Vona Vona, where full moon, greater tidal range, and 



low wind accounted for 40% of the variation in spawn- 

 ing oiE. devisi. Although not significant, days of rain 

 had a similar coefficient in the equations of best fit at 

 both Munda and Tulagi (Table 5), which suggests that 

 rainfall had a similar effect on spawning at these sites. 

 The significant proximate stimuli for E. heterolobus 

 differed between sites (Table 4). The best fit was ob- 

 tained at Vona Vona (r 2 0.44, P<0.01) where greater 

 spawning occurred when cloud cover was low and 

 moon phase approached full. Spawning was nega- 

 tively related to total rainfall and days since rain at 

 Munda (r 2 0.3, P<0.05), and there was a negative 



