A REVIEW OF RECENT STUDIES OF SUBPOPULATIONS 

 OF PACIFIC FISHES 



By 



Elbert H. Ahlstrom- 



1/ 



INTRODUCTION 



The chief objective of "population ' studies 

 has been to determine the extent of heterogeneity 

 within the range of widely distributed species. 

 Is a population composed of two or more groups 

 that maintain a high degree of separateness, or 

 is there mixing throughout the range of a species? 

 If a population is made up of several groups that 

 can be shown to possess structural differences, 

 are the differences due to environment or to 

 heredity? 



According to the definition of a subpopula- 

 tion given by Marr in the first article in this 

 series, a subpopulation is a fraction of a popula- 

 tion that is itself genetically self-sustaining. 

 Hence, even to establish that a subpopulation 

 exists, it is imperative to distinguish between 

 the effects of environment and differences in 

 heredity. 



Actually, this has not been done for any 

 species that will be considered in this paper. 

 Hence, in a strict sense we shall not be dealing 

 with studies of subpopulations, but rather with 

 studies of population heterogeneity . In fact, as 

 was frankly expressed by Rounsefell and Dahl- 

 gren (1935) in the paper on "races" of herring 

 in southeastern Alaska "whether the differences 

 in the characters chosen are due to heredity or 

 to environment has not been considered as being 

 of great importance, as long as the characters 

 are fairly stable within each population so that 

 significant differences indicate very slight inter- 

 mingling, if any, between adjacent stocks of 

 herring, (ibid: 120). For convenience, 1 refer 

 to the studies under review as "subpopulations" 

 studies Distinguishing "subpopulations" was 



1/ Assistant Chief, South Pacific Fishery 

 Investigations, U. S. Fish and Wildlife Service, 

 P. O. Box 271, Lajolla, California. 



clearly the intent if not the accomplishment of 

 most workers . 



There are two common approaches to the 

 problem of recognizing subpopulations. One is 

 an indirect approach and employs average mor- 

 phometric or meristic differences between 

 groups of fish to determine their probable sep- 

 arateness. The other is a direct approach- - 

 and uses marked (tagged) members to assess the 

 extent of mixing between groups of fish. The 

 morphometric approach has been widely used 

 because it can be done simply and inexpensively . 

 Often the specimens to be studied can be obtained 

 from the commercial catch . An effective tagging 

 program requires a considerable outlay in money 

 and personnel . 



The hypothesis underlying morphometric 

 and meristic studies of subpopulations of fish is 

 simply this: under conditions of partial or com- 

 plete isolation of groups of fish, slight differences 

 in body proportions or meristic characters will 

 be preserved in each group. These small dif- 

 ferences will not necessarily be apparent in 

 individual specimens but often only in an average 

 of a large number of specimens . The significance 

 of the differences is appraised by means of 

 statistical procedures based on the theory of 

 probability. The differences might be due to 

 either environmental or hereditary factors. It 

 is usually extremely difficult to determine whether 

 differences are phenotypic or genotypic, yet 

 knowledge of the cause(s) of the differences is 

 essential to an understanding of their significance. 



One of the critical points in this approach 

 is one of interpretation . It has often been assumed 

 by workers that if two groups of fish can be shown 

 to differ significantly in one or several morpho- 

 metric characters, the groups therefore are 

 distinct with only slight intermixture. This con- 

 clusion has been reached in many of the papers 



44 



