A MULTISTAGE RECRUITMENT PROCESS IN 



LABORATORY FISH POPULATIONS: IMPLICATIONS FOR 



MODELS OF FISH POPULATION DYNAMICS' 



David G. Hankin^ 



ABSTRACT 



Laboratory studies have been previously used to examine fundamental aspects of fish population 

 dynamics and may be explicitly structured to examine the stock-recruitment relation. Previous studies 

 have shown that cycling of population numbers occurs in refuge-free environments, but provision of 

 refuge areas allows maintenance of stable population numbers. Results of these studies may be 

 adequately explained by simple stock-recruitment theory. 



Laboratory experiments described here show that manipulation of refuge habitat quality can 

 profoundly influence interactions among population components. Complex interactions among fry, 

 juveniles, and adults created erratic pulses in numerical population growth. Numerical population 

 dynamics could not be adequately explained by simple stock-recruitment theory. 



Based on experimental observations, a multistage adult-juvenile stock-recruitment relation was 

 developed and was found, through statistical analyses, to adequately describe observed numerical 

 dynamics. The biological plausibility of complex multistage recruitment processes argues that 

 expectations for empirical support of simple stock-recruitment theory may be unreasonable and 

 inappropriate. The simple theory may often not be biologically appropriate and more complex models of 

 numerical population dynamics may be required for biological realism and for meaningful data 

 analysis. Whether collection of data necessary to allow use of such complex recruitment models is 

 economically feasible and, if so, whether more complex models may prove of practical use for 

 management of fish populations is at present unclear. 



One poorly understood population process is the 

 so-called stock-recruitment relation (Ricker 1954) 

 describing the dependency of input of new indi- 

 viduals, Rf, on the density of adult parents some 

 time previous, St-T- Although the theoretical 

 basis of the stock-recruitment relation is well 

 established ( Ricker 1954; Beverton and Holt 1957) 

 and recent study in theoretical ecology (May 1975; 

 Oster 1975) has emphasized the impressive variety 

 of population behaviors suggested by simple dis- 

 crete-time models of the form /?; = St-r  GiSt-r), 

 remarkably little empirical support for the theory 

 exists. In part this reflects severe restrictions in 

 data collection. In temperate populations, for 

 example, only one observation of recruitment may 

 be obtained annually and this observation is 

 normally related not only to parent stock but also 

 to fluctuating environmental conditions and mor- 

 tality (from birth to recruitment) which may 



Based on a dissertation submitted in partial fulfillment of 

 the requirements for the degree of Doctor of Philosophy, Depart- 

 ment of Natural Resources, Cornell University. 



Department of Natural Resources, Cornell University, 

 Ithaca, N.Y.; present address: Department of Fisheries, Hum- 

 boldt State University, Areata, CA 95521. 



strongly influence the ultimate size of a recruited 

 year class or cohort. The data collection process 

 is exceedingly slow, and exogenous factors may 

 confound the dependency of recruitment on parent 

 stock. 



Further, the theory itself is simplistic. Chief 

 limitations are the requirements that feedback be 

 exerted at only one point in time and that the 

 responsible population component consists solely 

 of adults. Alternative feedback control mechan- 

 isms could involve either juveniles or the adult 

 stock at more than one point in time. For example, 

 in largemouth bass, Micropterus salmoides, adults 

 are in contact with developing larvae for only a 

 short period of time during which adult-related 

 density-dependent mortality might occur. Adults 

 leave inshore nesting and nursery areas shortly 

 after spawning, but yearling bass, produced by the 

 adult stock a year previous, remain in inshore 

 areas where they may prey extensively on younger 

 juveniles (Ricker 1954). In Dungeness crab, Can- 

 cer magister, and other cannibalistic species, re- 

 cruitment may depend not only on parent stock 

 but also on adult densities when juveniles first 

 enter the adult population and are extremely 



Manuscript accepted: February 1980. 

 FISHERY BULLETIN: VOL. 78, NO. 3, 1980. 



555 



