AFFINITIES OF ASTACID^. 301 



important task to determine the share which can with any cer- 

 tainty he ascribed to the effects of atmospheric agency in the 

 transportation of fresh- water animals. At present we cannot 

 do this even approximately. The investigation would he un- 

 commonly difficult, for, in order to get a clear idea of it, it 

 would be necessary to contemplate at the same time the ques- 

 tions, first : Whether many nearly-allied forms, or forms which 

 to our eye appear as identical, might not have originated in two 

 or more distinct localities by what is known as polyphyletio 

 descent, and secondly : How old. the different forms may be his- 

 torically in the development of the animal world. In every 

 case we should thus be led to an exact inquiry into the genea- 

 logical affinities of the animal. One example will suffice. True 

 Astacidse or river cray-fish occur in Europe, in America, and 

 Australia, while they are absent from the intervening countries 

 and islands. Now, it would certainly be more than bold to 

 derive either of these groups directly from one of the others by 

 any theory of transportation through the air on the feet of 

 water-birds ; consequently the qiiestion at once arises, whether 

 we here have an instance of polyphyletic descent or not. Now, 

 so far as is known, there is no Crustacean living on land or in 

 fresh water, in either of the three continents, which can be 

 regarded as the parent stock of the Astacidce living there. But 

 in the different Oceans we do indeed find crustaceans — as, for 

 example, the species of Paranephrops — which have been con- 

 sidered as the nearest allies of the river crustaceans ; we will 

 not here discuss whether with justice or no. Now, if these 

 different marine Astacidse had gone through the same migra- 

 tions into rivers or on land, independently of each other in the 

 three continents, and had passed through analogous modifications 

 corresponding to those migrations, the extraordinary resem- 

 blance of the river Astacidte at such wide distances from each 

 other would be satisfactorily explained. But this method of 

 explanation obviously presupposes that our views as to the 

 essential affinity of the animals in question must be in fact per- 

 fectly accurate."^ 



Though, in the instances we have thus far been considering, 

 it has been difficult, or even almost impossible, to point out the 



