ESTIMATION 



This produces a blue pigment, presumably the indophenol 



CHoOH 



o=< 



HOCH2 



CI 



\===/ N 



CI 



>'CH, 



Pyridoxine did not give the reaction in presence of a borate buffer/^ 

 whereas 4-ethoxymethyl -3-hydroxy- 5 -hydro xymethyl- 2 -methyl -pyri- 

 dine and 4: 5-epoxydimethyl-3-hydroxy-2-methyl-pyridine gave colours 

 with the reagent both in the presence and in the absence of borate, so 

 that they could readily be distinguished from pyridoxine.^^ Unfor- 

 tunately, however, it is not possible to correct for the presence of these 

 substances in pyridoxine merely by subtracting the value found in 

 presence of borate from that found in its absence, as the different 

 compounds react with borate at different rates. It is preferable to 

 remove interfering substances before colour development. 



2 : 6-Dichloroquinone-chloroimide was also used by Bird et al.,^"^ 

 who carried out the coupling reaction in a veronal buffer solution, 

 pYL 7'8 to 8-0, and by Hochberg et al.,^^ who coupled in aqueous iso- 

 propanol solution. The method was said to give results in close 

 agreement with the biological method when applied to foodstuffs ^^ 

 and to rice bran,^^ but not to liver or yeast. ^o 



Pyridoxine exhibits a characteristic half-wave potential at the 

 dropping mercury electrode, ^^ but the polarographic method does not 

 seem to have been used for the estimation of pyridoxine. 



IVIicrobiological Assay Methods 



In recent years, there has been a marked increase in the use of 

 microbiological methods of assay for the estimation of vitamin Bg in 

 foodstuffs. These led incidentally to the discovery^ of compounds 

 with vitamin Bg-activity other than pyridoxine. 



The first microbiological method to be used was a yeast growth 

 method due to Atkin et al.^^ The organism was a strain of Saccharo- 

 myces cerevisiae (No. 4228), which requires pyridoxine for growth. 

 Extracts were prepared for assay by acid digestion, and the growth 

 of the organism was measured turbidimetrically. A similar method 

 was used by R. J. Williams et al.^^ and by Siegel et al^^ The latter 

 group of workers autoclaved the test material in acid suspension to 

 liberate bound pyridoxine. 



An attempt was made by Emery et al.^^ to develop a pyridoxine 

 assay method using another yeast, Kloeckera brevis, which requires 

 pyridoxine for growth, but the results were frequently at variance 



311 



