30
GORDON A. RILEY
C : N ratio by weight was 5-7 :1 in the surface layer and 3.6 :1 in waters
of 300-3 000 m. Menzel and Ryther (1964) obtained essentially similar
results in surface waters of the temperate North Atlantic. I n some of
their deep water analyses the ratios were as little as 2.7 : l . Values
reported by Duursma (1960) in the North Atlantic and Dal Pont and
Newel1 (1963) in the South Pacific fell within this same general
range.
Duursma (1960) and others had found that C : N ratios in dissolved
organic matter in sea water also were of a similar order. Provasoli
(1963) called attention to the fact that these ratios were much lower
than in fresh wat,er or soils, and the results suggested that humic
materials, which ordinarily have a C : N ratio of 8 : 1 or more, may be
scanty in the sea.
However, recent work indicates that the situation is more complicated and variable than it appeared to be in the early 1960's. HolmHansen et al. (1966) obtained a ratio of 5.5 :1 in the surface layer of
their station off the California coast, but the ratio increased below the
surface, and the average for the whole water column was 1 2 . 5 : l .
Subsequently Gordon (1970b) analyzed three sets of cruise data obtained in temperate and subtropical waters of the North Atlantic.
Altogether about 120 samples were analyzed, and the means were 7-8 :1
and 8.3 :1 in the first two cruises and 15-5 :1 in the third.
The possibility that the apparent variability was due to differences
in analytical technique cannot be entirely ruled out, although this
seems unlikely. All except the last author cited used wet oxidation
methods, which are possibly suspect on the grounds that oxidation of
carbon may not have been complete. Gordon used dry combustion,
and his higher ratios could have been due to differences in method.
However, Holm-Hansen et al. also obtained markedly high values by
wet oxidation methods. Here one might argue further that conditions
might have been atypical in the isolated basin that these investigators
were studying. However, the argument is not very convincing, because
high ratios were obtained well above the level of the sill depth of the
basin.
Handa (1968) also obtained moderately high C : N ratios at a station
in subtropical waters off Japan (35"N, 139"36'E), and he found that
the ratio increased with depth. He regarded this as indicating that
protein is utilized more readily than carbohydrate as the material sinks
into deep water. His biochemical analyses, which will be summarized
later, indicated that a large proportion of the carbohydrate existed as
polysaccharides which one would expect to be refractory.
There is no longer much doubt that observed variations in C : N
GORDON A. RILEY
C : N ratio by weight was 5-7 :1 in the surface layer and 3.6 :1 in waters
of 300-3 000 m. Menzel and Ryther (1964) obtained essentially similar
results in surface waters of the temperate North Atlantic. I n some of
their deep water analyses the ratios were as little as 2.7 : l . Values
reported by Duursma (1960) in the North Atlantic and Dal Pont and
Newel1 (1963) in the South Pacific fell within this same general
range.
Duursma (1960) and others had found that C : N ratios in dissolved
organic matter in sea water also were of a similar order. Provasoli
(1963) called attention to the fact that these ratios were much lower
than in fresh wat,er or soils, and the results suggested that humic
materials, which ordinarily have a C : N ratio of 8 : 1 or more, may be
scanty in the sea.
However, recent work indicates that the situation is more complicated and variable than it appeared to be in the early 1960's. HolmHansen et al. (1966) obtained a ratio of 5.5 :1 in the surface layer of
their station off the California coast, but the ratio increased below the
surface, and the average for the whole water column was 1 2 . 5 : l .
Subsequently Gordon (1970b) analyzed three sets of cruise data obtained in temperate and subtropical waters of the North Atlantic.
Altogether about 120 samples were analyzed, and the means were 7-8 :1
and 8.3 :1 in the first two cruises and 15-5 :1 in the third.
The possibility that the apparent variability was due to differences
in analytical technique cannot be entirely ruled out, although this
seems unlikely. All except the last author cited used wet oxidation
methods, which are possibly suspect on the grounds that oxidation of
carbon may not have been complete. Gordon used dry combustion,
and his higher ratios could have been due to differences in method.
However, Holm-Hansen et al. also obtained markedly high values by
wet oxidation methods. Here one might argue further that conditions
might have been atypical in the isolated basin that these investigators
were studying. However, the argument is not very convincing, because
high ratios were obtained well above the level of the sill depth of the
basin.
Handa (1968) also obtained moderately high C : N ratios at a station
in subtropical waters off Japan (35"N, 139"36'E), and he found that
the ratio increased with depth. He regarded this as indicating that
protein is utilized more readily than carbohydrate as the material sinks
into deep water. His biochemical analyses, which will be summarized
later, indicated that a large proportion of the carbohydrate existed as
polysaccharides which one would expect to be refractory.
There is no longer much doubt that observed variations in C : N