20 SCIENCE PROGRESS 



other growths. There is thus a degree of protection which is 

 common, and a certain degree which is specific. The spontaneous 

 disappearance of tumours occurs, entirely beyond our control as 

 yet. But the protection it confers can be imitated artificially. 

 The same degree of common protection can be conferred by the 

 preceding inoculation of tumour material if followed by no 

 growth, i.e. if absorbed, and this is not so remarkable as the fact 

 that it can be induced also by the injection of the normal tissues 

 of the mouse and most readily by normal blood ; but not by the 

 cancers or tissues of alien species. The far-reaching signifi- 

 cance of the protection which can be induced by carcinomatous, 

 by sarcomatous, or by normal tissues, causes me to emphasise 

 that our observations and Ehrlich's are in essential agreement. 

 We require to learn whether the protection thus conferred is 

 actively induced or only passively conferred. In the case of 

 blood this question was most easily settled. The protection is 

 not passively conferred by the serum, but is actively induced by 

 the blood cells. To sum up, we are able to so modify the soil 

 provided by mice that cancer cannot grow in it. For indications 

 of the change we are dependent on the behaviour of the living 

 cancer cell in the living mouse; we are dependent on experiment 

 absolutely. 



Jensen recorded the disappearance of tumours in 1902-3. 

 The first occasion on which we noticed a large transplanted 

 tumour disappear was while studying a batch of mice with his 

 tumour in August 1904. This tumour was exhaustively studied, 

 and the important part phagocytosis played in the process 

 discovered, not only in the case of this entire tumour, but in 

 more limited areas of a very large number of tumours. We 

 have since recorded the same process for other carcinomata, and 

 its importance has been emphasised by others, especially by 

 Clowes. We have observed the same process in localised areas 

 of spontaneous tumours. There can therefore be no doubt that 

 the animal primarily attacked attempts to protect itself by the 

 same means as a normal animal after successful inoculation. 



The investigation of the nature of protection is bound up 

 with very great difficulties, which are not diminished by the 

 absence of any analogy with what we know of immunity to 

 infective diseases. Up to the present I have spoken of the 

 propagation of the tumours as if it meant a mere culture in vivo 

 of the cancer cells. This, however, is only part of the truth. 



