614 SCIENCE PROGRESS 



The Lofoten Cod-fishery 



So far this appears to be the best example of a periodic 

 sea-fishery, the fluctuations of which depend on the fluctuations 

 of the Gulf Stream drift in northern latitudes. The great 

 Lofoten cod-fishery usually begins in January and ends in 

 April, but the variations from these dates are very notable. 

 Generally the maximum catches are made in March, but in 

 some years the fishery is only beginning in that month while 

 in other years it is then practically at an end. Fortunately 

 the Norwegian fishery authorities have collected statistics 

 which give not only the quantity of cod taken from week to 

 week during the progress of the fishery, but also the quantity 

 of cod-roe, and cod-liver obtained, these latter being commercial 

 products of great importance. The fishery records also include 

 statements of the air-temperature at the Lofoten Islands, and 

 during the last few years there are very precise data with 

 regard to the intensity and variations in the periodicity of the 

 Gulf Stream flow in those latitudes. It will be seen from fig. 2 1 

 that these islands lie in the direct core of the Gulf Stream drift. 



I reproduce here the figures prepared by Helland-Hansen 

 and Nansen in order to show the variations in the productivity 

 of this fishery and the associated intensity in the volume of 

 the Gulf Stream drift. An arbitrary date — March 15 — has been 

 selected, and the numbers of cod taken before that day have 

 been calculated and expressed as percentages of the total catch 

 of the whole fishery season. The greater the percentage offish 

 taken before March 15, the earlier is the appearance of the shoals, 

 and vice versa. This percentage is shown in fig. 7 as a thin 

 broken line (III.). It was maximal in 1902, minimal in 1903-4, 

 and has decreased since the latter year. The same figure shows 

 the air-temperature at Lofoten as a thin line (II.), and this was 

 minimal in 1902, maximal in 1903-4, and then decreased. The 

 thick continuous line (I.) shows the average temperature of the 

 sea beneath the surface at the Sognefjord line. This value 

 varies similarly to that of the air-temperature at Lofoten, but 

 the phase is different ; that is, the sea-temperature at Sognefjord 

 is a year earlier than the air-temperature — and, therefore, the 

 surface sea-temperature — at Lofoten. 



There is, therefore, an inverse relationship between the con- 



1 See Pt. I. of this article, Science Progress, January 1910, p. 479. 



