56 SCIENCE PROGRESS. 



mal. More especially should we expect to find this state of 

 things in those instances in which the two local races are 

 what are called climatic varieties, varieties that is to say 

 associated with conditions which we can recognise as 

 distinct (without postulating any direct relation of cause 

 and effect). It is not rare, for instance, to find a species re- 

 presented in the North by a northern race, and in the South 

 by a southern race. In such a case, if the two forms inhabit 

 a continental area, divided by no natural barrier, we should 

 expect, on the hypothesis indicated, that in the intermediate 

 area there should be an intermediate normal. Or, more 

 precisely, we should expect in travelling from North to 

 South to pass through country inhabited by a whole series 

 of normals, passing in unbroken continuity from the dis- 

 tinctly northern form to the distinctly southern. 



Again, if a lowland species is represented in the moun- 

 tains by an alpine race we should expect as we travel up the 

 valleys to pass through a series of normals, each appropriate 

 to its own level. We may concede that a full chain of inter- 

 mediates as normals need not be expected in every case, 

 but that the state of things ought, on the whole, to bear out 

 this expectation is surely clear, and if the hypothesis of an 

 essentially gradual evolution is true, the geographical transi- 

 tion from the one race to the other should be essentially a 

 gradual transition. 



It is very likely that cases could be produced where 

 such essentially gradual transitions occur, but, as will be 

 shown, there are others which cannot readily be so de- 

 scribed. Whether there are any general features distinguish- 

 ing the two classes of cases we cannot yet say. It is the 

 object of these remarks to call attention to the paramount 

 importance of such phenomena as subjects for investigation. 



Few indeed are the instances which we can yet with 

 confidence refer to either class. Collectors and systematists 

 have hitherto been content as a rule with the bare know- 

 ledge that there is sometimes intergradation between local 

 races. Statistical evidence of the modes of intergradation 

 are almost entirely wanting. Commonly the information is 

 of the most meagre description ; and even in the case of 



