50 THE BIOCHEMISTRY OF B VITAMINS 



extraneous effects. 31 Even so, the growth period in the test is 10 days, 

 and in view of later developments the use of this organism has little to 

 recommend it. 



Niven and Smiley 32 have suggested the use of Streptococcus salivarius 

 as a test organism. It is extremely sensitive to minute amounts of thia- 

 mine (0.1 to 2 /i,g per tube) and yields satisfactory assay values. Thiamine 

 fragments are inactive; cocarboxylase is 40 per cent more active than 

 thiamine. The method has the disadvantage that the medium has a pH 

 value of 7.4 and thiamine must be added to it aseptically after auto- 

 claving. 



Sarett and Cheldelin 33 have developed an assay method using Lacto- 

 bacillus fermenti, which responds in their test to from 0.005 to 0.05 /xg 

 of thiamine. The recommended growth period is 16 to 18 hours and under 

 these conditions, cocarboxylase is 30 per cent more active than thiamine. 

 The method has been applied by the authors to a considerable number 

 of foods, to animal tissues, and to urine with excellent results. 34 



Biological Assays. Probably the most widely used biological assay 

 methods for thiamine involve curative tests on rats. In the U.S.P. 

 method, 35 rats are kept upon a thiamine free diet until they show signs 

 of acute polyneuritis, whereupon at least eight such rats are given a 

 standard dose of pure thiamine which "cures" the polyneuritic condition 

 for from 5 to 15 days, depending upon the dose given and the condition 

 of the individual rats. As soon as the rats have regressed and reached 

 the same stage of polyneuritis as before dosing, a single dose of the 

 unknown material to be tested is administered. If the curative effect of 

 this preparation lasts as long or longer than that of the standard (based 

 upon the sum of the cured days for each animal), it is determined to con- 

 tain as much or more thiamine than the standard. 



It is obvious that the assay of unknown mixtures by this method 

 involves the use of a large number of animals and a large amount of 

 time, since the depletion period may last as long as 8 or 10 weeks and 

 the assay period for a single test is from 10 to 30 days longer. 



A less time-consuming curative test is one devised by Smith 36 and 

 modified by Birch and Harris. 37 In this modified test animals are placed 

 upon a thiamine-deficient diet containing, however, a small amount of 

 thiamine in the form of 0.4—0.5 per cent brewers' yeast. On this diet 

 they develop polyneuritic symptoms more regularly and do not die of 

 extreme thiamine deficiency before they become useful test animals. 

 After depletion, which may require 50 days or more, they are injected 

 intravenously with the material to be tested and the minimal amount 

 required to produce a three-day cure is determined. If a larger than 

 minimal dose is given, the excess can be judged by the longer duration 



