THE MODES OF REPRODUCTION 23 



must consist of the physiological units postulated by 

 Spencer. 



Much space has been spent on this subject : firstly, 

 because it is satisfactory to correlate theory with 

 actual structure, so far as is possible ; secondly, 

 because it is plainly necessary that we should pursue 

 our study of the nucleus to the very last limits ; 

 and finally, because this great conception of physio- 

 logical units has lately acquired a new importance — 

 clearly anticipated by its originator — in relation to 

 the remarkable theory of variation which traces its 

 origin to the work of the Abbe Mendel forty years 

 ago, but which has only within the last year or two 

 taken its place as one of the most important of the 

 constructive efforts of modern biology.^ 



It wdll now be necessary for us to study the repro- 

 ductive cell, not in the abstract, but, so to speak, in 

 situ, as an actual structure found, in given conditions, 

 in all but the lowest animals and plants. So far as 

 is possible, we must make our study of heredity a 

 study of observed facts. The less we talk al)out 

 " forces " and the more about positive entities and 

 sequences that can be actually observed, the more 

 likely are we to reach conclusions that will reward 

 our labours. Fortunately, we live to-day not in 

 the dawn of embryology, which is now, despite its 

 difficulties, an astonishingly luminous and con- 

 spicuous science. 



One further note may be made. At first sight it 

 may seem almost incredible that a heritable organi- 

 sation of great complexity can iind its physical basis 

 in a microscopic ovum and in a spermatozoon which 



1 See Chapter VII. 



