i894. CHARACTERS IN BIOLOGY. 271 



it differs in all ; it is against all physiological and pathological 

 experience to think of, say, the sexual characters as independent 

 structures and functions superimposed on a common neutral basis. 



If characters are not regarded as varying independently of one 

 another, or as being present in some mystical way in the forms 

 which "transmit" them without "showing" them; if they are 

 not looked at as being any definite unities for the organism, but are 

 recognised as arbitrarily-defined groups of phenomena ; if, in short, 

 we make it clear to ourselves that they are only appearances of the 

 individual, which cannot with justice be divided into parts in the 

 consideration of heredity, and which if it varies in one respect varies 

 in its whole constitution, then the result for our theories of heredity is 

 obvious. The questions become at once more complex and more 

 simple. More complex because, being deprived of our false units, we 

 have to leave tabulating characters, and have to go to school to 

 physiology that we may trace the many aspects of the individual to 

 the functional peculiarity of the germ, the many differences to the one, 

 just as the above-mentioned authors have done for the complex 

 phenomena of sex ; and because a character ceases for us to be trans- 

 mitted or carried latent as such by the demon particles, and has 

 instead to be referred in its rise to the reactions of every part with 

 every other and with the world. They become more simple, because 

 we shall cease to be hampered by the difference in the habits of 

 authors as to the way in which they form characters, or by the 

 difficulty of reducing positive and negative relations of quantity and 

 quality, utterly incoordinate and incomparable as they are, to a common 

 formal expression ; and most of all because we shall cease to be 

 confronted with the question as to how the demon particles do their 

 strange work. 



Surely it is an unprecedented thing that men going about to 

 explain the shapes of animals and plants should leave on one side 

 what are clearly the causes of these shapes, namely, the processes of 

 development which lie to their hand in physiology and embryology, 

 and should rest content with an antique and fantastic scheme of 

 growth which makes all intermediate processes of no value ; so that 

 all the changes known as inheritance and variation can be studied in 

 relation to only two moments in the whole of life, namely, the adult, 

 as pure movphe, and the germ, fixed and stained. And since this has 

 been to so great an extent our method, it is not surprising that 

 Natural Selection is loaded with more than it can bear, or that the 

 expression which the great principle of use-inheritance receives is 

 artificial and exaggerated. 



George Sandeman. 



