364 KYTHER [CHAP. 17 



variable from day to day at a given place (see, for example, Fig. 7 ; also Currie, 

 1958). 



For the reasons given above, Steele's data are less variable and the seasonal 

 range is of smaller magnitude than appears to be the case with the studies 

 based upon 14 C measurements. Nevertheless, a well-defined seasonal cycle is 

 obvious, with spring and fall maxima and a summer minimum. Production 

 during the spring (April 1-June 15) was relatively constant for the three years 

 of the study, which is consistent with the hypothesis that the spring flowering 

 follows the establishment of the thermocline and is limited by the amount of 

 nutrients brought into the euphotic zone by winter mixing. During the summer 

 and fall periods, however, production was quite variable from year to year, a 

 fact which Steele has attributed largely to the variability of winds (Fig. 8) and 

 their effect upon the vertical transport of nutrients through the thermocline in 

 summer and the breakdown of the thermocline in autumn. 



The annual production at the Fladen Ground ranged from 57.4 to 82.3 g 

 carbon/m 2 . Steele's subsequent estimates for 27 locations in the northern North 

 Sea (Steele, 1958) ranged from 45 to 110. Thus production in this enclosed, rather 

 shallow, boreal environment appears to be somewhat lower than that estimated 

 for the open Atlantic 20° of latitude further south. It is noteworthy that the 

 difference is due primarily to the presumed absence of plant production during 

 North Sea winters. 



Other productivity studies in temperate and northern waters over shorter 

 periods of time deserve mention here because they illustrate local geographic 

 variations in production, and also because they emphasize the wide range and 

 extreme variability of the process in this general oceanic region. The measure- 

 ments by Saijo and Ichimura (1959) in the northwest Pacific south of Japan 

 during summer (May-September) show the contrast between the productivity 

 of the warm Kuroshio Current (0.05-0.10 g carbon/m 2 /day) and the cold, 

 coastal, more northerly Oyashio Current (0.25-0.50 g carbon/m 2 /day). Studies 

 of the same general area during the spring (Sorokin and Koblentz-Mishke, 1958) 

 gave equally low values for the Kuroshio, suggesting the absence of a seasonal 

 cycle of production typical of tropical water. Their measurements further north, 

 on the other hand, where the Oyashio is mixed with waters from the Sea of 

 Okhatsk and from the general sub -arctic water-mass, were extremely variable, 

 ranging from 0.006 to 5.1 g carbon/m 2 /day. The higher values were usually 

 associated with "cold spots", presumably indicative of local vertical turbulence. 



The boreal North Atlantic and the waters surrounding southern Greenland, 

 Iceland and the Faeroe Islands were investigated for several years during the 

 summer months by the Danish Institute for Fisheries and Marine Research 

 (Steemann Nielsen, 1958; Hansen, 1959). Productivity measurements of this 

 area (1954-1957) revealed a high degree of variability related to the complexity 

 of different water-masses and current systems. Summer rates of production 

 typically ranged from 0.1 to more than 2.0 g carbon/m 2 /day. 



In general, low production was observed in Davis Strait, the East Greenland 

 Polar Current, the Irminger Current and the North Atlantic Ocean west of the 



