238 TOILERS IN THE SEA. 



laps slightly the analogue of the oral opening (fig. 5 i). 

 Out of this opening comes a second cylinder, to termi- 

 nate in four other tentacles in the same way ; and in 

 some polyps there is a further growth, so that there 

 are two or more rows of tentacles separated by the 

 tubular cylindrical tissue. It is evident that Agassiz 

 saw young, ill-developed, and probably injured po- 

 lyps, which had not attained their second row of ten- 

 tacles. The number of tentacles may be, therefore, 

 four, eight, twelve, &c. In looking at this description 

 there is a probability that the Millepore is an Alcyo- 

 narian, and there is no proof that it is a Hydroid." l 

 It is also stated that "every variety of tentacular 

 and disk apparatus may exist in either ; but the 

 external development of the gemmules, ova, and 

 embryonic forms, must be recognised before any 

 Coelenterate animal can be associated with the Hy- 

 drozoa. Here is the point where Agassiz fails. His 

 researches are only suggestive, until the generative 

 organs are recognised on the protruded polyps." 



In his enumeration of the different groups of 

 animals which contribute to the formation of coral 

 reefs, Dana includes the Sea- fans, and their allies, 

 and the Sea-mat animals (Bryozoa}. The former of 

 these he admits contribute but little to the material 



1 "On the Actinozoan Nature of Millepora," by R. G. 

 Nelson, and P. M. Duncan, in "Annals of Natural History," 

 vol. xvii. (1876), p. 354. 



