of the Fishery Board for Scotland. 



3 



The information which we have also obtained as to the limited 

 number of times very many salmon spawn, sheds an important light 

 upon the direct result which may be produced upon the stock of fish 

 in different localities. It is clear that far more grilse and salmon 

 pass prolonged periods, including spawning seasons, in the sea, than 

 was formerly supposed. The grilse and salmon which enter our 

 rivers represent the fish which, in any given year, are alone 

 responsible for the reproduction of their species that season. 



Grilse are virgin fish, which still show to a great extent the habit 

 of moving in shoals, or at any rate of making simultaneous migrations, 

 in the manner seen in the earlier smolt stage of development. The 

 smolts descend to the sea in April, May, and June. The first 

 reappearance of these fish in fresh water, or on the coast near river 

 mouths, is in the following summer, when, as grilse, they have 

 attained the age of three to three and a half years. 



Grilse are, however, taken in large numbers in bag-nets set at 

 considerable distances from the mouths of rivers. Their wanderings 

 are evidently pretty wide, but the noticeable fact is that they do not 

 so occur till May, and do not as a rule become plentiful till June. 

 Great captures are made, as it were, quite suddenly when the fish 

 "strike the coast" in numbers. Certain favourable conditions are 

 known to coast fishers when the takes of both salmon and grilse will 

 probably be large, but salmon do not occur in the same numbers as 

 grilse. Salmon in most localities are netted from the opening of the 

 fishing season. Grilse do not come within the netted zone till 

 summer. Yet the grilse captured outnumber the salmon captured 

 in, I believe, all netted localities in Scotland, and enormously out- 

 number the salmon in many districts. 



I may, I think, regard it as axiomatic that when large numbers of 

 grilse are taken by net, large numbers are entering our rivers to 

 spawn. In such seasons the fact that rod fishers find a plentiful 

 supply of grilse in fresh waters is, I consider, sufficiently good proof. 

 To demonstrate the fact is unnecessary, although reference to the 

 point seems desirable for the sake of logical sequence. That a con- 

 siderable proportion of grilse do not enter fresh water at this stage is, 

 however, equally true. Investigations by marking and by the study 

 of scale growth show that the small spring fish, which are so 

 numerous in many early rivers, e.g., Dee, Brora, Helmsdale, etc., are 

 the grilse of the previous summer which did not enter fresh water to 

 spawn, but continued feeding and growing in the sea till after the 

 spawning season. I may mention in passing that in certain large 

 rivers of Norway, as apparently also in the Rhine, grilse are or have 

 become comparatively scarce. In the returns respecting the Dutch 

 salmon fisheries I find that in 1873 the grilse showed, in the total 

 catch, a percentage of 57 ; for a considerable period of years after 

 that date the grilse formed 23 to 24 per cent. In 1903 the grilse 

 formed 6 per cent. 



In Scotland certain rivers which used to hold abundance of grilse 

 do not appear now to do so. In the Tweed quite recently, as I 

 am informed, the absence of grilse has been evident to all. My 

 Tweed figures do not cover recent years, so that I am unable to show 

 how the netting records stand. In other cases we have rivers which 

 apparently never attracted grilse in any numbers, and rivers which 



