58 THE LAMARCKIAN DOCTRINE 



course, if we choose, we may seek to pry deeper and speculate 

 about the composition, the ' architecture/ of the germ-plasm, and 

 the hidden springs of its behaviour. But, in the almost total lack 

 of verifiable evidence, we should be reduced to mere valueless 

 guessing valueless because incapable of being tested. We know, 

 however, that there must be a germ-plasm, a substance that is the 

 bearer of heredity, in the germ-cell, and that in the case of every 

 species of living beings it must possess definite hereditary 

 tendencies which differ from those of every other species, and we 

 have very strong reasons for supposing that it is derived by direct 

 and real descent from the germ-plasm of preceding germ-cells. 

 On these tolerably sure foundations we are able to build a very 

 great deal. 



89. It is more difficult to account for the differences in detail 

 between parent and child than for the resemblances in mass. 

 Some of these differences are readily explained by the fact that 

 the stimuli under which parent and child develop are never exactly 

 similar. The nutriment, use, and injury which awakened the 

 hereditary tendencies of the parent are sure to have differed some- 

 what from those which awakened the tendencies of the child. 

 These account for the acquired differences. The real difficulty 

 arises when we seek to account for * variations,' for those innate 

 differences which are founded, not on differences in stimuli, but in 

 germ-plasms. We have to answer the question, " How does it 

 happen that the child ' varies ' from the parent ? " Or more 

 precisely, " How does it happen that the germ-plasm of the child 

 differs from that of the parent ? " If the problem is extremely 

 complex and difficult, it is also very important. Indeed it is 

 perhaps the main problem of heredity. If we solve it correctly, 

 we shall have discovered not only the correct interpretation of 

 many of the phenomena of heredity, but, since all racial change is 

 founded on the variations of individuals, of evolution also. In this 

 matter, just because the problem is so complex and difficult, we 

 must, linking together a number of principles reached by induction, 

 consider heredity and evolution together. Variations, taken by 

 themselves, are capable of a number of interpretations, any one ot 

 which, as far as we know, may, or may not, be correct ; taken in 

 conjunction with the facts of adaptation it becomes evident that 

 only one interpretation can possibly be true. 



90. Before beginning our investigations we must make up our 

 minds to do some very careful thinking. I do not say this because 

 we are about to deal with exceptionally abstruse matters, nor 



