380 BULLETIN OF THE 



less marked than the embryonic changes already known among many 

 animals. If we combine into a series all the changes thus far observed 

 among different families of corals, an unmistakable gradation appears 

 among them, akin to the series which may be traced among other animals 

 in their adult condition, when we take the complication of their structure 

 as a standard of their arrangement. Combining the evidence obtained 

 from adult coral stocks, and their young at various stages of growth, it 

 becomes evident that the representatives of the class of Polyps do not 

 stand upon the same structural level with one another ; but that there 

 are higher and lower types among them, recognizable without the 

 aid of embryological data, even though it was the study of the young 

 which led me to the recognition of their relative standing. This is 

 not the place for a discussion of the principles of classification of Polyps. 

 I will only state, what I trust I shall be able to prove hereafter, that 

 the Actinians proper stand lowest ; next to them the Madrepora- 

 rians, and highest the Halcyonarians. And as the Madreporarians form 

 the most prominent feature in the coral reefs, I may add that among 

 them the Turbinolians stand lowest, the Fungians next, then the Astras- 

 ans, and highest the Madreporians. Now it is a most interesting 

 fact that the successive changes which any representative of these 

 different groups exhibit during their growth recall the characteristic 

 features of the groups immediately below. For instance, young As- 

 traeans, before assuming their solid frame, are Actinia-like ; their 

 first coral frame is Turbinolia-like ; and from that stage they pass into 

 Fungia-like condition, before they assume their characteristic Astraan 

 features. 



I will only describe a few cases, in order to establish this correspond- 

 ence of growth and relative standing of adults upon a firm scientific 

 basis. Besides multiplying through eggs, Actinia? increase also by 

 budding, and this takes place by a spreading of their base of attach- 

 ment (abactinal area), from the margin of which new individuals arise 

 and finally detach themselves. Such a mode of enlargement or spread- 

 ing of a simple individual, by a widening of its base of attachment, I 

 have observed in many genera among Fungians, Astrasans, Oculines, 

 and Madrepores. If we take, for instance, a Siderastrcea, which, by the 

 way, is a Fungian, and not an Astraean, as is shown by the structure 

 of its tentacles, as well as of its coral stock, we find that the large 

 rounded masses formed by these corals are at first thin, spreading 



