SECT. XV CLASSIFICATION OF CRUSTACEA 263 



the group in the light of what is known as to the 

 orisfin of the whole class from a bent Annelid. We 

 find, then, no group of early Crustacea from which we 

 can actually deduce them. They are distinctly lower 

 in the scale of development than any of the early 

 groups which we have already described, and proved 

 to be the most primitive. We are thus driven to the 

 conclusion that they must have originated from some 

 larval form. There is no difficulty in this sup- 

 position. Among the enormous number of free- 

 swimming and independently feeding Nauplii, it 

 would almost certainly be an advantage to some to 

 remain but little advanced beyond the Nauplius, the 

 more pronounced character of the adult bringing them 

 at once into danger. If we assume that they are 

 modified larvae of early Apodid^e, the conditions, as 

 far as we know them, would be fairly well satisfied. 

 The Apodidae were driven from the open sea by some 

 foe or foes, and would have been exterminated had 

 they not, in the manner described in the early part of 

 this book, taken refuge in shallows and lagoons, and 

 finally in freshwater puddles. We may well suppose, 

 therefore, that while one division of the Apodidae 

 thus retreated inland and were able there to develop 

 into adults, another probably found safety in remain- 

 ing at the larval stage, their smallness, their trans- 

 parency, and the rapidity of their motion rendering 

 them comparatively safe. Whether the organisation 

 of the Copepoda can be explained on this hypothesis 

 we are not able to decide. The view that they are 

 really equivalent to larvae finds some support in the 



