Contributions to the Biochemistry of Growth. 237 



nervous tissue — which is free from ordinary non-phosphor ised fat — can be 

 accounted for is an indication of the accuracy of the method. Details of the 

 method will be published in a separate paper. Here it may be sufficient to 

 indicate the principle of the method, which consists in extracting the total 

 lipoids by chloroform, and estimating in aliquot portions of the extracts : 

 (1) the cholesterin and cholesterin esters by Windaus' method, using certain 

 precautions which are essential for the accuracy of the results ; (2) the 

 phosphatides by a phosphorus estimation ; (3) the cerebrosides by estimating 

 the galactose liberated after prolonged acid hydrolysis ; (4) the ordinary 

 non-phosphorised fats (glycerides of fatty acids) by difference. In calcu- 

 lating the phosphatides and cerebrosides from the data obtained by this 

 method, it is assumed that the average phosphorus percentage of the 

 phosphatides and the galactose content of the cerebrosides present in the 

 tumour cells do not differ essentially from the figures known for the 

 phosphatides and cerebrosides of nervous tissue. It is, of course, possible 

 that this assumption is not correct. In that case a recalculation of the 

 figures given in this paper for the phosphatides and cerebrosides would be 

 necessary. But since our arguments are based on a comparison between the 

 figures obtained for different strains of tumours, such a correction would 

 not materially affect our conclusions. In fact, in the case of the cere- 

 brosides, for which the differences between different tumour-strains are — 

 as will be seen — so great as to become qualitative, this possible source of 

 error can be disregarded. 



The observations refer to a rapidly growing rat sarcoma, and to two 

 different strains of a mammary carcinoma of the mouse, of which one always 

 grows rapidly (63/73b), while the other always grows slowly (72/21a and 

 72/22a). Since the rapidly growing mouse carcinoma showed marked 

 necrosis, the macroscopically necrotic parts were separated from the less 

 necrotic portion of the tumour and analysed separately. These two parts 

 are described in the following table as " necrotic part " and " healthy part " 

 respectively; but it must be understood that even in the macroscopically 

 healthy parts there was still a good deal of necrotic tissue. The rat sarcoma 

 showed little necrosis. 



The material subjected to analysis was obtained in the case of each strain 

 from several tumours belonging to the same subtransplantation. The 

 figures given in the following table represent therefore the composition of 

 the lipoids of a number of sister tumours at one particular period of their 

 growth, for the three tumour strains ; they do not merely indicate the 

 composition of one individual tumour. 



vol. lxxxvii. — B. 



s 



