SALPA IN RELATION TO EVOLUTION OF LIFE. 203 



Still more impressive and significant is the fact that all craspe- 

 dota have a hydra-stage in their life history. Conclusive evidence 

 shoAvs that they are all descended from a hydra-like ancestor. 



This undoubted truth has been assumed to involve the belief that 

 this common ancestor was a fixed hydroid-cormus, although this 

 implication is by no means inevitable. 



Every one now believes that the hydro-medusa is an expanded 

 and perfected expression of the hydroid type, and that it has been 

 evolved from a simple, hydra-like, starting point, but the writers 

 who have most clearly seen this homology, and who have demon- 

 strated it most conclusively have gone a step further, and have 

 assumed that the locomotor medusa-stage has been added on to the 

 life of hydroids for the purpose of distributing the species, and that 

 it has been evolved, according to the law of the division of labor, 

 by the gradual specialization of certain ones among the members of 

 a polymorphic hydroid community. 



This view, which seems to have commended itself to all students, 

 and which seems to derive support from the well-known fact that 

 the blastostyles are, actually, nutritive hydranths which have been 

 secondarily differentiated and specialized, by division of labor, in 

 adaptation to the physiological needs of the cormus as a whole, this 

 view has become so firmly established that it is now regarded as a 

 settled and closed question. 



I have no hope of effecting any change in views which are so 

 firmly rooted, but I shall now try to show that this established 

 opinion will not stand the test of searching examination. 



From the earliest times the very ancient ciliated planula stage 

 has provided for the distribution of coelenterates, and there is no 

 need of other means of dispersal, nor is there any reason to think 

 that the distribution of the species is any less important now than 

 it has been in the past ; yet one of the most remarkable and note- 

 worthy peculiarities of hydroids, is the pronounced tendency of the 

 medusa to degenerate and to lose its locomotor habit, and to become 

 a sessil and degraded gonophore. The view that the sessil medusa- 

 buds of hydroids, like hydractinia, are nascent medusae has, very 

 properly been abandoned, and all morphologists admit that they are 

 degenerated and that their sessil condition is secondary. 



