GENERA OF FAVOSITID^E. 157 



and which more or less extensively anastomose with one an- 

 other, precisely as in the genus Michelinia, De Kon. The 

 tabulae are continued into the hollow horizontal expansions of 

 the tubes, where they continue to present the same characters 

 of irregular anastomosis as in the cavities of the corallites 

 themselves. Transverse sections (PI. VIII., fig. 2 b) exhibit 

 the corallites themselves, cut across at right angles to their 

 lonof axis, tos^ether with irreo-ular moieties of the hollow con- 

 necting-floors, which come into view in consequence of their 

 not lying in a single horizontal plane. The only special fea- 

 tures exhibited by sections of this nature are that the corallites 

 are provided with distinct and separate walls, and that the 

 septa are represented by radially-disposed spinules. 



From the preceding description it will be clear that we have 

 to deal in Chonostegites Clappi, E. and H., with a coral in many 

 respects closely allied to Michelinia. If we take typical ex- 

 amples of this species, in which the corallites stand wide apart, 

 each enclosed in its own epitheca, and united with its fellows 

 by hollow horizontal expansions derived from the visceral cham- 

 ber, then, it Is true, we should feel no hesitation In separating 

 Chonostegites from Michelinia as a distinct genus. Even In such 

 examples, however, the hollow periodic expansions of the tubes 

 may be properly regarded as homologous with the "mural pores" 

 oi Michelinia. If, on the other hand, we examine a portion of 

 the colony of Chonostegites Clappi, In which the corallites are In 

 actual contact and have their walls perforated by numerous 

 mural pores, then we feel at a loss to produce a single charac- 

 ter by which the species could be generically separated from 

 Michelinia. Moreover, the curious AI. {Eimnonsia) cylindrica 

 of Edwards and Haime Is a distinct connecting-link between 

 the present species and the typical forms of Micheliniay since It 

 exhibits slight periodic constrictions of Its tubes, and the mural 

 pores are generally confined to the dilated portions of the 

 corallites. 



Upon the whole, however, I prefer keeping Chonostegites dis- 

 tinct from Michelinia, In the meanwhile at any rate. It may be 



