SECT. 2] ORGANIC REGULATION OF PHVTOPLANKTON FERTILITY 173 



of water-masses, and already poses some new puzzles. It is interesting for 

 instance to compare the C/N ratios found in fresh water and sea-water. Birge 

 and Juday (1934), in grouping the data from several hundred lakes according 

 to total organic carbon, found a C/N ratio of 12.2 for lakes containing 1.0- 

 1.9 mg C/l. Duursma found in North Atlantic deep waters of similar C content 

 a C/N ratio varying from 2.5-6.5 with a mean of 2.7. The surface waters 

 (0-200 m) of the same transect (a meridional section running southward from 

 Cape Farewell in Greenland, 60°N, to about 43°N) show a tendency toward 

 large variations in C/N ratios (up to 20-30) at higher temperatures ( > 7°C) 

 whereas at lower temperatures (<7°C), and especially in September, the C/N 

 ratios are closer to the values of deep water. Living phytoplankton and zoo- 

 plankton have an average C/N ratio of 5.7-6.7 with limits between 4 and 14 

 (Fleming, 1940), proteins of 2-3, and skeletons of invertebrates 2.9-3.3 (Vino- 

 gradov, 1953). Why the striking difference in C/N ratios between fresh waters 

 and sea-water? Are the high values in fresh water only due to the influence of 

 the land on lakes, or to physical conditions such as depth and size of the basins? 

 Hutchinson (1957) has calculated from the data of Birge and Juday the amounts 

 of the two types of dissolved organic matter found in lakes : one (allochthonous) 

 derived from bogs, peat and soil has a brown color and an approximate C/N 

 ratio of 45-50 ; the other (autochthonous) derived from the decomposition of 

 plankton has an approximate ratio of 12. Skopintsev (1959) found that the end 

 product of complete decomposition of marine organisms is a "water humus", 

 a carbon-protein complex (possibly a pectin or uron) of high biochemical 

 stability. This product has a C/N ratio of 10, remarkably close to the one 

 calculated by Hutchinson. In the open sea we cannot expect to find significant 

 amounts of allochthonous dissolved organic matter but we could expect to find 

 the autochthonous, yet we find lower ratios (2-6) in some surface waters and 

 in deep waters. The warm surface waters of the North Atlantic have variable 

 and high ratios, while the cooler surface waters have ratios similar to the deep 

 waters. This could be due to a slower decomposition in cool waters : the dead 

 organisms will be decomposed only partially before they sink ; in deep water 

 the scarcity of micro-organisms and the cool temperature may account for 

 the stability of C/N ratios of the dissolved organic portion. If so, the dissolved 

 organic matter in deep water should still be a good nutrient and not a product 

 resistant to bacterial attack. Many experiments with oxygen bottles in fact 

 show that the deep waters can support a good population of micro-organisms 

 (provided solid surfaces are available). This, however, does not explain the 

 difference, presumably in the early phases of decomposition, between fresh 

 water and sea-water. From the starting point of 6 (ratio in living plankton) 

 decomposition produces in fresh waters higher ratios (i.e. N is preferentially 

 consumed) and in sea-water lower ratios (C is preferentially consumed or high-N 

 compounds are produced). If we assume that, during decomposition in sea- 

 water, high-N products are released in large amounts, then the low C/N ratios 

 found do not exclude the presence, though in low quantities, of stable products 

 like "water humus" which have high C/N ratios. Quite likely decomposition 



