ECOLOGY: PORT JACKSON SHARKS 537 



POPULATION STRUCTURE 



Since H. portusjacksoni have such a wide distribution (from Western 

 Australia to beyond Newcastle in central N.S.W., with a doubtful record 

 in Moreton Bay in Queensland (Saville-Kent 1897)), and since the apparent 

 major oviposition and nursery areas are extensive but widely separated 

 (e.g., Port Stephens, Sydney area, and Jervis Bay area on the N.S.W. coast), 

 it would seem reasonable to expect that different populations of sharks 

 could use different breeding areas and hence be genetically different. Any 

 method by which genetic variation between individuals of a species can be 

 detected should yield information about the existence of separate popula- 

 tions within a species. McLaughlin and O'Gower were unable to distinguish 

 morphologically between individuals of H. portusjacksoni, but Gordon 

 (1947), using the frequencies of genes specifying morphological characters 

 in the fish Platypoecilus maculatis, was the first to demonstrate the exist- 

 ence of subpopulations in a marine vertebrate. Techniques and methodol- 

 ogy have advanced considerably since that time, and population studies 

 today usually involve genetic differences measurable at the molecular level. 

 Cushing (1964) reviewed the results of such studies involving marine verte- 

 brates, but his survey covered mainly serological techniques, high -resolution 

 electrophoresis of proteins being then in its infancy. In a later, very exten- 

 sive review, de Ligney (1969) surveyed the principles, methods, and results 

 of virtually all serological and biochemical studies of fish populations under- 

 taken to that time. This study reports the first attempt to define the popu- 

 lation structure of a species of elasmobranch by biochemical means, al- 

 though Sindermann and Mairs (1961) have described an erythrocyte antigen 

 system in Squalus acanthias and proposed that it be used in population 

 studies of this shark. 



A typical cellulose acetate gel, on which several shark haemoly sates have 

 been separated, is shown in Figure 5. The proteins running between the 

 origin and the haemoglobins are polymorphic (Nash and O'Gower, unpub- 

 lished data). The simplest genetic explanation for the protein variation as- 

 sumes a triple, codominant allelic system (a, b, and c) that segregates accord- 



Q 



ing to (p + q + r) , where p, q, and r represent the gene frequencies. 

 From Table 3 it can be seen that the populations from Sydney, Jervis 

 Bay, Victoria, and South Australia are in good agreement with the Hardy - 

 Weinberg law, the Western Australia population agrees at the 5% level, 

 and the Newcastle population differs significantly from the law. 



Comparisons of the gene frequencies show that superficially the popula- 

 tions are all different. However, the statistically more rigorous x 2 pair- 

 wise comparisons (Table 3) do not substantiate this belief. While there 

 are small, but significant, differences among the three eastern populations, 

 and among the Sydney, South Australia, and Western Australia popula- 

 tions, there is no significant difference (5% level) between the Jervis Bay 

 and the remaining southern and western populations. 



The results of the present study are probably as good as one could ex- 

 pect from a survey of natural populations, particularly when it has proved 



