VARIATION AND HEREDITY 189 



it was formerly supposed that all variation 

 must be, a progress from a lower degree of 

 complexity to a higher, but the converse. 

 When from a single wild type man succeeds 

 in producing a multitude of new varieties, 

 we may speak of the result as a progress in 

 differentiation : but we must recognize that 

 the term is only applicable loosely, and that 

 the obvious appearance of increased com- 

 plexity may in reality be the outcome of a 

 process of simplification." Similarly, " rever- 

 sion occurs when the sum total of the factors 

 returns to that which it has been in some 

 original type." The return may be brought 

 about by the omission of an element or 

 by the addition of a missing element. If 

 certain kinds of variation may be called 

 " unpacking," reversion is re-packing. 



In discussing the bearing of Mendelism on 

 the theory of evolution, Bateson makes three 

 important suggestions. (1) " One has only 

 to glance over trays of birds' skins, the port- 

 folios of a herbarium, or drawers of butterflies 

 and moths, to discover abundant ' species ' 

 which are analytical varieties of others," i. e. 

 differing in the presence or absence of definite 

 factors. " The principles of heredity we 

 trace in our experimental breeding are operat- 

 ing throughout the natural world of species." 



