222 The Experimental Analysis of the Growth of Cancer. 



The behaviour of some strains of Jensen's tumour present a parallel to the 

 other tumours now under consideration. We have already alluded to the 

 negative results sometimes obtained on transplanting Jensen's tumour. In 

 the graphic record (fig. 12) the steps are shown by which a strain of Jensen's 

 tumour, at first giving a high percentage of success, progressively exhibits 

 weaker and weaker powers of proliferation, till finally the tumours obtained 

 gave negative results on transplantation. Such strains are not uncommon ; 

 they have frequently been followed to a finish during our experiments. Thus 

 one chapter, as it were, in the life history of Jensen's tumour reproduces the 

 entire life history of other tumours under artificial propagation. These results 

 are difficult to harmonise with the assumption that the apparently con- 

 tinuous proliferation of Jensen's tumour is purely vegetative. Together with 

 the facts of spontaneous absorption they strengthen the conclusion derived 

 from a study of the details of that proliferation, that a cyclical process is 

 involved. 



The importance of the preceding analysis of the growth of propagated 

 cancer is obvious in appraising the results of attempts to modify growth 

 experimentally. The experimental conditions whose variations cause 

 irregularities in the success of artificial propagation must be taken account 

 of. In particular, the age of the animals would seem to call for especial 

 attention, because the short duration of the life of a mouse magnifies the 

 effect of the lapse of time involved in procedures for inducing immunity. 

 Specially adapted control experiments must be performed in order to obviate 

 the fallacy which the ageing of the animals introduces. Those fluctuations 

 which cannot be referred to the experimental conditions but are natural 

 features of the proliferation of the tumour cells are an even more urgent 

 reason for caution in interpreting the results of therapeutical experiments. 

 The difficulty or even imposssibility of predicting the time at which 

 spontaneous absorption will affect the propagated tumours indicates the 

 necessity for accurate records of their previous history. 



In another paper with Dr. Cramer we shall discuss the results we have 

 obtained on re-inoculating mice in which the absorption of well-established 

 tumours had occurred spontaneously and under the action of radium. 



