230 SCIENCE PROGRESS. 



vinced that conjugation in unicellular and sexual reproduc- 

 tion in multicellular organisms are the means of producing 

 variation (4). But the ultimate cause of hereditary varia- 

 tions he considers to be due to the constant recurrence of 

 slight inequalities of nutrition in the germ plasm, or repro- 

 ductive substance of the organism. Weismann also considers 

 that variation depends upon the action of environment, but 

 as he does not believe that variations so produced can be 

 transmitted to the next generation, he is compelled to adopt 

 the hypothesis mentioned. His views are best illustrated 

 by a concrete example, the case of twins in the human race. 

 As a rule, twins differ from each other in mental and physi- 

 cal characteristics just as much as do children of the same 

 parents born consecutively. Occasionally, however, the 

 twins show so close a resemblance that they cannot be dis- 

 tinguished from each other by strangers. These may be 

 called " identical " twins, and the former " dissimilar " twins. 

 In all probability dissimilar twins are derived from two ova, 

 which have of course been fertilised by two different sper- 

 matozoa, whilst "identical " twins are derived from a single 

 ovum and spermatozoon. On careful physical measure- 

 ment, it is found that " identical " twins do show slight 

 differences from each other in respect of the size and 

 relation of parts. These differences are probably due to 

 variations in the external conditions of environment during 

 growth. The large and obvious differences of " dissimilar" 

 twins are, on the contrary, due to differences of the germ 

 plasm, and are determined at the time of fertilisation of the 

 ovum. 



To mention still another theory as to the cause of 

 variation, Montgomery considers that it is due to the dis- 

 turbance of correlation of the organs, resulting from the 

 change of environment, and is thus only indirectly due to 

 change of environment (2). 



All the theories which have been advanced as to the 

 causes of variation may thus be practically divided into two 

 classes — (i.) theories which regard them as due to some 

 inherent tendency of the organism to vary, or to variations 

 of the germ plasm ; that is to say, theories which are 



