356 THE CAUSATION OF DISEASE. 



The above remarks are borne out by what we know of 

 nutgalls, the structure of which depends, as is well known, 

 upon the structure of the plant-tissue affected, and upon the 

 nature of the parasite. Darwin speaks of the important share 

 taken by the latter in determining the structure of the gall. 

 Many species, he observes, may grow on the same tree, and 

 their difference is due to the subtle differences in the chemical 

 composition of the poison emitted by the several species of 

 gnats. A subtle shade of difference in chemical poison is, 

 then, capable of producing a palpable difference of structure. 

 Surely, a most noteworthy fact. 



I have said that the transported malignant cells share in the 

 production of secondary growths. Consider the carcinomata ; 

 their complex structure is faithfully reproduced in the secondary 

 growths, even though the tissue secondarily affected contains 

 no epithelia. Assuming that the transported cells do not 

 multiply, the secondary growths may be accounted for in two 

 ways, (i) We may suppose that the specific bacterium is 

 capable of producing the same tissue-transformation in what- 

 soever tissue it may happen to lodge, for, be it observed, the 

 essential part of my hypothesis consists in this : that under 

 the bacterial irritation the tissues primarily affected undergo 

 dissolution, manifesting some or other of their many poten- 

 tialities — undergoing, in fact, what may be called vitiated 

 reversion. But it would be ridiculous to assume such an origin 

 for all cases of secondary growths — to suppose for instance, that 

 a secondary epithelioma, whether squamous or cylindrical, of — 

 let us say — a lymphatic gland, could originate by reversion ; 

 there is no ground whatever for supposing that adenoid tissue, 

 belonging as it does to the connective group, contains within 

 itself the potentiality of such a complex gland-like structure. 

 (2) The other hypothesis is sufficiently designated by the word 

 " spermatic," and, judging from the attention it has received 

 in works on pathology, it has been widely accepted. Yet it is 

 utterly unscientific. It assumes that the cells of one tissue 

 can, by mere contact with the cells of another tissue, compel 

 the latter to grow like themselves. What warrant have we for 

 such an assumption ? Certainly not one biological fact ; in- 



