WHAT IS A "TENDENCY" ? 145 



characters breaks down altogether. All are the outcome of 

 the hereditary transmission of a tendency or predisposition 

 possessed by the original germ. This hereditable tendency 

 to vary lies at the root of all evolution, and of the acquisition 

 of all new characters and functions. 



Whether this tendency to variation in organic germs 

 is equally strong in all directions, or whether, indeed, there 

 is such an unlimited tendency, is a question the discussion of 

 which would lead us beyond the limits of our present in- 

 quiry. I have already attempted to point out, in a paper 

 read before the British Association at its meeting in Liver- 

 pool in 1870, 1 how difficult it is, on mathematical grounds, 

 to explain the phenomena of Mimetism by the application 

 of the law of Natural Selection acting on an unlimited 

 tendency to variation in all directions. Twenty-five years 

 ago I wrote: 2 "The discovery of this law (Natural Selec- 

 tion) marked an era in the history of natural science, and 

 gave a wonderful impulse to original research. The danger 

 now is that the law will be pressed into services which 

 have no claim upon it ; and that, in the hands of injudicious 

 partisans, it will become a hindrance rather than an aid to 

 science, by closing the door against further investigations 

 into other laws which lie behind it. To claim for Natural 

 Selection the main agency in the creation of the countless 

 forms of organic life with which we are surrounded, is 

 straining it beyond its strength. An era of equal im- 

 portance will be marked by the discovery of the law which 

 regulates the tendency to variation that must necessarily 

 underlie Natural Selection." Are we much nearer to a 

 solution of this problem now than we were a quarter of a 

 century ago ? 



My chief object in the foregoing remarks has been to 

 show that, accepting Weismann's own definition of "ac- 

 quired " characters, there is still in these characters a 

 hereditable factor, which renders it impossible to draw any 

 satisfactory distinction between them and "non-acquired" 



1 See Nature, vol. iii., p. 30, 10th Nov., 1870. 



2 Tom. tit., p. 33. 



