A. Blood characteristics: The antigen - 

 antibody reaction has been employed in studying 

 differences (1) in blood sera and (2) in blood 

 cells. 



B. Muscle amino acids: Paper chroma- 

 tography as a tool for identifying muscle amino 

 acids has recently been used. 



While these different techniques are 

 rather exact and well known, they need extensive 

 testing in order to determine whether or not dif- 

 ferences observed through their use are a 

 reflection of genetic differences. 



VI . Genetic studies . 



While difficult, it is conceivable that the 

 direct approach of the geneticist in breeding ex- 

 periments could be applied to marine fishery 

 problems and, in the case of some fresh water 

 species, it has, in fact, been so applied. 



Some Recent Advances 



The post -World War II intensification of 

 interest in general fishery problems has, of 

 necessity, included an intensification of interest 

 in the subpopulation problem . As a consequence 

 new techniques have been developed, techniques 

 have been borrowed from other fields, previous 

 studies have been critically reviewed and new 

 studies have been carried out on fresh-water, 

 anadromous and marine fishes. Examples of all 

 these are included in the present collection and 

 are briefly mentioned below . 



Ordinarily comparisons of meristic or 

 morphometric data have been made by tests of 

 significance. These show, at some specified 

 level of probability, whether or not the samples 

 being examined could have been drawn from the 

 same universe. Royce points out that the stu- 

 dent of subpopulations (as opposed to, say, the 

 student of species) is not so much interested in 

 whether his samples were drawn from the same 

 or different subpopulations, but rather in how 

 much intermingling there is between the sub- 

 populations represented by his samples. This 

 is a very important distinction and one that has, 

 perhaps, not previously been fully appreciated. 



Because of the nature of their problems 

 and data, fishery biologists have been reason- 

 ably quick to adopt statistical procedures as 

 convenient tools for reducing data . Indeed, 

 many statistical procedures have been modified 

 or developed by fishery biologists seeking tools 

 appropriate to unique needs . Thus, Royce has 

 extended and modified previous methods so that 

 it is now possible to simultaneously compare 

 several characters and reach a maximum esti- 

 mate of the amount of mixing that could occur 

 between the populations sampled. It should be 

 pointed out that this method indicates only the 

 maximum mixing that could occur. It provides 

 no information as to whether or not such mixing 

 does in fact occur. Even so, this method will 

 be of great aid to fishery biologists and will, I 

 suspect, in some cases require extreme modifica- 

 tion or even rejection of previously held 

 hypotheses. 



With a certain maximum amount of mix- 

 ing possible, how can one determine how much 

 mixing, if any, actually does obtain? As Royce 

 points out, this information can be obtained 

 from a tagging experiment. Under ideal condi- 

 tions a tagging experiment could define the rate 

 or percentage mixing. Unfortunately, these 

 ideal conditions are seldom found and a tagging 

 experiment may only show that some mixing does 

 occur. Royce's method now enables us to set a 

 maximum amount to the mixing that does occur. 



A different and more restricted approach 

 to the problem of detecting mixing has been 

 taken by Widrig and Taft . They have considered 

 the problem of determining the amount of mixing 

 between two areas in some given time interval, 

 say one year. The method they develop is such 

 that if the mean value of some character is Known 

 for area A and area B at the beginning of the time 

 interval and the mean value of the character is 

 known for only area A at the end of the time in- 

 terval, the percentage of fish in area A at the 

 end of the time interval that came from area B 

 can be determined. 



As Widrig and Taft point out, there are 

 a number of limitations to this method. Since 

 the characters selected in practice may often be 

 phenotypic, it will be necessary to consider only 



