330 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1&16. 



The last 20 years constitute indisputably a new period in the his- 

 tory of transformism where the field of discussion has been reneAved, 

 and scientists haA^e sought to give it a much more positive and ex- 

 perimental character. Two kinds of investigation have been devel- 

 oped in this direction : On one hand the methodical study of varia- 

 tions, and on the other that of heredity and especially of hybridiza- 

 tion. These two categories overlap. 



Note that this new point of view is not, properly speaking, a study 

 of evolution. According to it, variation and heredity in themselves, 

 under present conditions, are analyzed independently of all hypo- 

 thetical previous states of the organism. Afterwards the results 

 obtained with the Lamarckian, Darwinian, and other succeeding 

 theories will be confronted. 



The sum of these researches, which are now in high favor, is a 

 new and important branch of biology, which has received the name 

 of genetics: It defines for us in particular the hitherto very vague 

 notion of heredity and seems certain to lead us to an analysis of the 

 properties of living substance somewhat comparable to that which 

 the atomic theory has afforded concerning organic chemistry. We 

 can not maintain too strongly its great importance. As far as the 

 theory of evolution is concerned, the results obtained up to this time 

 have been rather disappointing. Taken together the newly dis- 

 covered facts have had a more or less destructive trend. In truth 

 the results obtained do not agree with any of the general con- 

 ceptions previousl}^ advanced and do not show us how evolution may 

 have come about. They have a much greater tendency, if we look 

 only to them, to suggest the idea of the absolute steadfastness of the 

 species. We must evidently accept these facts such as they are. 

 But what is their significance? On the one hand they are still 

 limited, on the other hand, as I have already stated above, and as I 

 shall try to show in the following lectures, the advances made by 

 the study of heredity in organisms at the present time and under 

 the conditions in which we are placed, does not permit us to accept 

 ipso facto the doctrines of heredity for all past time and under all 

 circumstances. 



To use a comparison which has only the force of a metaphor but 

 which will make my thought clear, the biologist who studies 

 heredity is very much like a mathematician who is studying a very 

 complex function with the aid of partial differential equations and 

 who tries to analyze the properties and the function about a point 

 without being able as in the case of an elementary function to study 

 it in itself, directly, in all its aspects. The properties ascertained 

 about one point are not necessarily applicable to all space. 



As far as the organisms are concerned, the conditions of their 

 variability have not certainly been the same in all periods. The 



