CANADIAN FISHERIES EXPEDITION, 191.'rl5 



109 



1915-16 also seems to suggest that the process of mixing is gradual and may require 

 some time, in other words, the group comprising the mature fish can at certain 

 periods be more or less heterogeneously composed, as was the case to a very high degree 

 in 1916. 



Inv'estigations as to age among the Norwegian herring have been carried out 

 under favourable conditions, inasmuch as it has been possible to obtain samples not 

 biassed owing to tho methods of capture employed. Obviously, when gill-nets are 

 used, some doubt may easily arise as to whether the age-distribution, apparent in a 

 sample, may be more or less a result of the selective effect of the net itself, and a 

 sample of such netted herring cannot be credited a priori with the same representative 

 value as one taken with the fine-meshed seine. It was therefore fortunate for the 

 Norwegian investigations that the herring fishery of Norway happens to a great ex- 

 tent to be carried on by seine (shore and purse-seines), and these more reliable seine 

 samples furnished a starting point from which to investigate those taken in other nets. 



The best means of carrying out such investigations would probably be to set nets 

 of different mesh out in a standing seine. This would furnish excellent material for 

 the purpose. Unfortunately no such experiments have hitherto been carried out, and 

 for the present, all we can do is to utilize the material available, and compare seine 

 samples with net samples taken at approximately the same place and time. In so 

 doing, however, we have to reckon with a disturbing factor, to wit, a certain relation 

 existing between time and place of capture, on the one hand, and implement on the 

 other. Thus drift-nets are employed for the capture of herring out at sea, and espe- 

 cially early in the season ; the whole of the " large herring " fishery is carried on with 

 drift nets. The shore seine, on the other hand, can only be used when the fish come 

 close in to land. Stake nets, again, are fixed on the bottom, while the purse-seine is 

 worked near the surface. The Norwegian material includes a number of paired sam- 

 ples more or less satisfying the above requirements. These are set out in pairs in 

 table 4 so as to permit of comparison between the age-composition in the samples. 



Table 4. — Comparison between samples from seine and net-hauls. Each net sample 

 compared with the seine sample nearest adjacent in point of time and place. 



Localitj'. 



Svinfihavet. 

 Gursko. . . . 

 Karnisund. . 

 Rovccr 



Karrnsund 



Skudf^nes. ....... 



Inside Skudeiies. 



Date. 



Mar. 5, 1914 

 „ 26, 1914 



Feb. 19, 1914 

 M 13, 1914 

 ., 19, 1914 

 M 19, 1914 

 •, 18, 1915 

 ,. 22, 1915 



The general impression given by the table is that the columns of net-samijle fig- 

 ures resemble very closely those for the seine-caught fish, such discrepancies as occur 

 being small and without apparent regularity. Particularly interesting are the 

 columns for the samples for Karmsvuid and KuvaT, where tiie two seine samples are 

 from hauls made on the same day and quite close together. The impression, furnished 

 by such paired comparisons between seine and net samples, leads us to the conclusion 

 that the age-composition of mature Norwegian herring has been very much the same 

 in the netted samples as in those taken with the seine. The same result is arrived at 

 if we consider the material as a whole, the whole of the netted material tending in the 

 same dir<^ction as that furnished by the seines. And bearing in mind the fact that 



