CONFERENCE ON SALMON PROBLEMS 



87 



with a view to elucidating certain appar- 

 ently anomalous sea trout fisheries I was 

 much struck by the fact that all the waters 

 which had suitable spawning ground were 

 inhabited by sea trout while all those which 

 were locally reputed to be devoid of sea 

 trout had no suitable spawning ground for 

 the fish. 



Apart from these observational data we 

 have also evidence from the fish themselves. 

 In the course of body lengths calculated 

 from the scales we have found that each 

 individual river has a different average 

 length of the smolts at migration and that 

 the type of parr scale growth as well as 

 the smolt age differs considerably from 

 river to river. Differences in length, aver- 

 age age and type of growth are not neces- 

 sarily connected with different parts of the 

 country but may occur in adjacent rivers, 

 e.g., the Aberdeen Dee and Don are only a 

 few miles apart and yet the parr differ so 

 greatly in type that occasionally among fish 

 caught in the Dee one may occur with a 

 scale growth so different from the average 

 and so like that of the Don as to suggest 

 that it is a chance wanderer. A compari- 

 son of the weight to length ratio of salmon 

 from different rivers, when all are caught 

 just fresh from the sea, also shows that 

 each river has a certain class of either rela- 

 tively fat or comparatively long and thin 

 fish. This particular feature can be traced 

 not merely for the whole stock but for each 

 sex and each year class individually just 

 as the average smolt length and the type 

 of parr scale growth is found to vary not 

 merely for the whole catch but also when 

 each age class from different rivers is com- 

 pared by itself. 



One final essential difference between dif- 

 ferent rivers has to be referred to before 

 the situation is summed up. It is that at 

 the same times and with the same age 

 groups of salmon those of one river may 

 have an average length and an average 

 weight which are both greater than the 

 corresponding details for the other river. 



It is true that even if specific local stocks 

 for individual rivers do exist we cannot 

 expect that the distribution will be abso- 



lutely perfect and that a few fish will not 

 occasionally enter, possibly temporarily, a 

 wrong river. But if we can quote as be- 

 tween different rivers and in strictly com- 

 parable circumstance differences of weight, 

 length, condition (ratio of weight to 

 length), distribution of age groups, smolt 

 lengths and ages, and types of parr growth, 

 then I do submit that the only explanation 

 can be the existence of separate biological 

 units in each river which through course 

 of time and in response to the local environ- 

 ment have developed these characteristics. 

 When such features are reinforced by more 

 equivocal circumstances such as the varia- 

 tions in the times of run and in the propor- 

 tions of different age classes as well as the 

 total number of salmon in each river any 

 other conclusion seems to be impossible. 



The development of the special scale 

 characteristics can also be used to check the 

 migrations of salmon marked in the sea and 

 to ascertain if those which make long mi- 

 grations from the place of marking to a 

 river have done so by chance or with a defi- 

 nite destination in view. In all cases which 

 I have tried, the test has shown that the fish 

 are of a type which belongs to the particular 

 area or the particular river where they were 

 finally found. For instance in the scale of 

 a salmon which was marked on the Scottish 

 northwest coast and which was recaptured 

 on the Norwegian coast. Professor Dahl 

 agrees with me that the parr scale is typical 

 of that of a Norwegian west coast river; 

 it is certainly quite unlike that of any Scot- 

 tish river. On the other hand the Forth 

 salmon have a particularly typical parr 

 growth which shows on the scale a small 

 first year and a good second year. Let us 

 consider the scale of a salmon marked in 

 Norway and recaptured in the Forth and 

 that of another marked at Loch Inchard 

 and again taken in the Forth. The first 

 scale practically duplicates the growth of 

 the type scale and although the second is 

 not so clearly marked the calculated lengths 

 obtained from it are of the same order. 

 Finally parr growth and consequently scale 

 growth in the Tweed is exceptionally good 

 and there are concurrently a considerably 



