208 PROVASOLt [CHAP. 8 



cells in different waters. Consequently, the bioassay of vitamins in direct assays 

 of sea-water should be based on (Droop, 1955) or include (Daisley, 1958) 

 "internal standards". The sample of sea-water is split in two portions equally 

 enriched with N, P, Si and chelated trace metals. Two parallel growth curves 

 are obtained either by diluting one portion 3 x with artificial sea-water and 

 adding to both portions graded amounts of B12 (Droop, 1955) or by adding B12 

 to one portion and diluting both portions stepwise with H 2 (Daisley, 1958). 

 One can calculate, by the interval separating the two regression curves, the 

 amount of Bi 2 present in sea-water without referring to an "external standard" 

 (i.e. the growth curve obtained with increasing amounts of vitamins in artificial 

 media). Assays based on B12 extracted from sea-water or separated by dialysis, 

 do not need internal standards. 



Since we cannot transfer laboratory sensitivity data directly to ecological 

 situations, other means must be devised to judge whether vitamins are limiting. 

 One method was found by Johnston (in press). He employed bacteria-free 

 cultures of Skeletonema : no, or poor, growth occurred in over 200 samples of 

 sea-water from the northern seas enriched with N, P, Si, with and without 

 trace metals. Since the same types of waters enriched with S36 medium — which 

 has vitamins — give much better growth, the content of vitamin Bi2-like 

 cobalamins was limiting in these waters. Over sixty of these samples were 

 collected in winter. According to the data of Co way (1956) one would expect 

 that 1-2 mfxg/1. of vitamin Bi2-like cobalamins were present. This quantity, 

 according to laboratory sensitivity data (Provasoli, 1958a), is above the 

 minimal quantity required by several marine algae. Unfortunately we have no 

 laboratory data on the sensitivity of Skeletonema ; however, a good density was 

 obtained by Droop (1955a) in the second serial transfer in "no B12" from a 

 culture grown in 100 m[i.g/l. (the dilution factor for each transfer was 100). 

 Similar assays with bacteria-free cultures of the important ecological species of 

 algae of a given environment can tell directly whether the waters are deficient 

 in vitamins. The assay with bacteria-free strains of the local ecological species 

 permits a reliable judgment. The only inconvenience may be the duration of 

 the assay, but this disadvantage is offset by an avoidance of the measurement 

 of the quantity of vitamin (which has little meaning except for a comparison 

 between different waters) ; this in turn requires differential assays and the 

 determination of the vitamin specificity patterns of each organism of the 

 environment. Bacterized unialgal cultures obviously cannot be used for this 

 purpose — bacteria produce vitamins. 



Trace metals are the other new parameter. The bioassays of Kylin, Thomas, 

 Ryther and Guillard, and Johnston, extending the pioneering work of Harvey, 

 prove that waters from the northern seas, as well as the Sargasso and tropical 

 Pacific, benefit greatly from the addition of trace metals. As noted earlier, 

 Johnston demonstrated that, in the northern waters, the deficiency is due not 

 to lack of trace metals but to their unavailability ; the addition of a metal 

 chelator, because of its solubilizing power, is as effective as the addition of 

 trace metals. Conceivably other waters, especially the tropical ones, may be 



