THE BIOLOGICAL STANDARDISATION 

 OF THE VITAMINS 



SIR HENRY DALE has Said that "the object of every committee on biological 

 standardisation is self-extermination." Chemical and physical determi- 

 nations are more quickly carried out and, in general, are more accurate than 

 biological determinations, and one of the chief functions of workers on 

 biological assay is to correlate their results with those obtained by chemical 

 and physical methods in order to find out how far these latter methods really 

 measure the activity in question. It therefore behoves workers in the biologi- 

 cal field not only to make their determinations as accurate as possible under 

 the very difficult conditions inherent in the nature of their task, but also to 

 apply to their results accepted methods of estimation of the accuracy of their 

 determinations. Otherwise it is only too easy to blame the inaccuracy of the 

 biological method for differences between chemical and biological results and 

 for differences between two or more biological results. 



The same statistical examination of results is needed when two determina- 

 tions appear to be equal, for there is nearly always good internal evidence from 

 the experiment that there is a fair chance that the results are not in fact equal. 

 The accuracy of a biological determination is generally stated as, for example, 

 a one-in-two chance (50:50) that the result obtained lies outside, plus or minus 

 a certain percentage, of the true result ("the probable error"), a one-in-three 

 chance that it lies outside somewhat larger limits ("the standard error"), a 

 one-in-twenty-two chance that it lies outside larger limits, a one-in-a-hundred 

 chance that it lies outside larger limits still. These chances are often stated 

 as probabilities that the result lies within the corresponding limits when they 

 are designated as the limits of error for P = 0.50, 0.67, 0.95, 0.99 respectively. 

 Every biological determination should be stated as the result obtained to- 

 gether with the limits of error corresponding to the probability most likely 

 to be of use in evaluating the result. The Sub-Committee on the Accuracy of 

 Biological Determinations, of the British Pharmacopoeia Commission, chose 

 to express the limits of error of the tests it described for P = 0.95 as being the 

 figure in rather general use, and also for P = 0.99 as being a figure which a court 

 of law would probably consider large enough in possible legal proceedings.^ 



Essentials of the Biological Method 



Workers on the biological assay of vitamins now generally recognise certain 

 essentials for their task. Some of these were apparent almost at the beginning 

 of the attempts to measure these factors; others have become evident only 

 through arduous and most painstaking labours. 



1. Standard of Reference— Every worker is familiar with the variation in 



