606 



Fishery Bulletin 97(3), 1999 



Eggs/larvae 



Size class 



Reserve 



Fishing area 



Catch 



Catch 



Eggs/larvae 



Catch 



Size class 



Figure 1 



Flow diagram for the model. Lines represent contributions from one size class to 

 itself and others. Populations in the fishing area and reserve interacted through 

 larval dispersal between them, whereas adults remained in the area where they 

 settled. Populations in the fishing area experienced fishing mortality ("catch"), 

 whereas reserve populations did not. Otherwise, populations in reserve and fishing 

 areas experienced the same population phenomena. This figure is an example of the 

 model construction for white grunt, Haemulon plumieri — the number of size classes, 

 size class at reproductive maturity, and size class at recruitment to the fishery dif- 

 fered for other species. 



Appendix for relevant equations). To convert these 

 continuous measures into probabilities, we used the 

 standard von Bertalanffy parameters to estimate the 

 lengths of the smallest and largest individual in each 

 size category for the following year. We calculated 

 the proportion of this size range that fell into the 

 next size class and used this value to represent the 

 probability that a fish of this size class grew to the 

 next size class. The remaining individuals stayed the 



same size over the next year, with a probability de- 

 termined by subtracting the probability of growing 

 from 1. We chose the size-class interval for each spe- 

 cies such that newly settled fish had exactly 100% 

 chance of growing to the next size class during the 

 first year (see Appendix for formula). Consequently, 

 fish were never able to grow more than one size class 

 in a year. As with all von Bertalanffy growth rela- 

 tionships, growth slowed with age — in our case from 



