THE CAUSATION OF DISEASE. 1 45 



those with long beaks, large feet. In all these cases, the parts 

 sympathetically affected must, in some way, be anatomically and 

 physiologically united, so that an E affecting one of them must 

 necessarily affect the other, or others, at the same time. It may 

 not be easy to trace this connection in all cases, but in some it is 

 possible. The teeth, for instance, are dermal appendages, and 

 anything interfering with the development of the skin will tend 

 to affect the formation of the teeth. Wherefore the effects of 

 '■'correlation" are wrought through modifications of internal 

 cell-E, and the latter is ultimately dependent upon the ex- 

 ternal-E. 



(e) Inheritance. — We have seen that the offspring cannot be 

 like either parent, and hence one cause of variation. We also 

 saw, however, good reasons for assuming that upon an ultimate 

 analysis the variation is primarily due to E. " But," it is argued, 

 "the principle of heredity is fickle." "No one can say . . . why 

 the child often reverts in certain characters to its grandfather, 

 or grandmother, or more remote ancestors ; why a peculiarity is 

 often transmitted from one sex to both sexes, or one sex alone." 

 Now, reasons have already been given which favour the view 

 that reversions are often brought about by peculiarities of E. 

 Thus, a peculiarity of E may hinder development at a cer- 

 tain stage, and the individual may be born with some remote 

 ancestral characteristics ; and it is probable also that special 

 likeness to one or other parent is chiefly connected with the E 

 of germ and sperm during their birth and development. We saw 

 that, if the parents are of unlike species, there is a tendency to 

 reversion. But this is no real exception to the law of heredity, 

 for the offspring is a mean structural product of the parents, 

 leaving out of account all those characters which have been 

 acquired since the period of ancestral divergence, and which 

 refuse to blend in the offspring. 



It is necessary now to observe that individuals differ vastly in 

 the readiness with which they respond to different kinds of E. 

 In other words, one organism will more readily take on 

 new characters than another, or, in the usual language of the 

 biologist, one organism will more readily "vary" in a particular 

 •direction than another. This diversity in respect of the 



