VII. ESTIMATION 53 



result of their findings, described the first microchemical method for the 

 estimation of ehoHne. Roman obtained reproducible results within a range 

 of 5 7 to 5 mg. of choline with a maximum error of 5 %. 



Erickson and her collaborators^ improved the periodide micromethod of 

 Roman by using immersion filter sticks which facilitated washing the labile 

 choline iodide precipitate with minimum disturbance. The introduction of 

 this step in the procedure was desirable because the greatest source of 

 error in the periodide method was the volatile and unstable nature of the 

 choline precipitate. A further impro\'ement was the introduction of In'omine 

 oxidation to con^'ert the iodine to iodate preceding titration with standard 

 sodium thiosulfate. The latter modification offered the advantages of read}^ 

 solubility of the precipitate in the l^romine solution and a sixfold increase 

 in the final titration value. 



Potassium periodide yields precipitates with dimethylamine, trimethyl- 

 amine, betaines, certain cyclic bases, and many alkaloids.^" The presence 

 of these substances in plants makes it impossible to relate the results ob- 

 tained by the periodide method to a definite constituent in a plant extract. 

 Reifer^" claimed that the greatest part, if not all, of the periodide-precipi- 

 table substances in plant extracts, with the exception of choline, were 

 removed by lead acetate and Norit. 



At the present time the most widely used method for the quantitative 

 determination of choline appears to be by precipitation as reineckate. 

 Among the first to employ the reineckate method for the determination of 

 choline were Kapfhammer and Bischoff," who washed the choline reineck- 

 ate with ether and determined it gravimetrically. This method required 

 fairly large amounts of choline and, therefore, was not easily adaptable 

 for general use. A more sensitive method described by Beattie^- consisted 

 of the precipitation of choline with a freshly prepared solution of ammo- 

 nium reineckate, solution of the precipitate in acetone, and comparison of 

 the bright red color imparted to the solvent with that of a standard solu- 

 tion. Using a visual colorimeter, ciuantities of the order of 0.3 mg. of choline 

 in a concentration of 0.003 % were estimated bj^ this procedure with an 

 error of not more than 3 %. The use of photoelectric colorimetry greatly 

 increased the sensitivity of the method. '^'^'^ According to the procedure of 

 Jacobi et al.,^^ the ground sample was extracted wdth boiling 1:1 alcohol- 

 s' B. X. Erickson, I. Avrin, D. M. Teague, and H. H. Williams, J. Biol. Chem. 135, 

 671 (1940). 



1" I. Reifer, New Zealand J. Sci. Tech. 22B, iii (1941). 



'^ J. Kapfhammer and C. Bischoff, Hoppe-Seyler's Z. physiol. Chem. 191, 179 (1930). 



>2 F. J. R. Beattie, Biochem. J. 30, 15.54 (1936). 



'3 R. W. Engel, ./. Biol. Chem. 144, 701 (1942). 



'^ D. Click, ./. Biol. Chem. 156, (343 (1944). 



15 H. P. Jacobi, C. A. Baumann, and W. J. Meek, /. Biol. Chem. 138, 571 (1941). 



'« M. n. Thornton and F. K. Broome, hid. Eng. Chem. Anal. Ed. 14, 39 (1942). 



