500 ME. H. M. BERNARD ON THE AFFINITIES OF 



now and again a case is found whicli shows this homology to 

 be the correct one. 



I have somewhat elaborated this, to my mind, almost obvious 

 conclusion, because it is essential to a right understanding of the 

 morphology of the genus. Just as the corallum of the Madre- 

 poridse, as I have elsewhere shown, is built up above the epi- 

 theca out of the radial structures — septa with their synapticular 

 supports — so also is the corallum of AJveopora built up within 

 the epithecal cup out of fusing septa. But there is a fundamental 

 difference between the two : in Alveopora there are no continuous 

 septal laminae, no synapticulse, and consequently no inner thecal 

 wall supplementing the epitheca ; hence no costse and no ccenen- 

 chyma. These peculiarities of the genus, which separate it from 

 all other recent Madreporaria, will be repeatedly insisted upon in 

 the following pages, but are mentioned here in this sketch of the 

 genus, inasmuch as v. Koch *, who was not so fortunate as I have 

 been in having young stages to esamine, thought that the skele- 

 ton of the bud was walled round by thin synapticulse which united 

 the septa, and that hence the budding might be called " Zwischen- 

 knospung." Whereas it seems to me that if the calicle-walls of 

 Alveopora are really primarily due to the fusion in one plane 

 of septal spines, not only is the word " theca" not applicable, but 

 the word " synapticulse " is equally inapplicable ; and, lastly, that 

 the budding cannot, at least very felicitously, be called "Zwischen- 

 knospung." The method of reproduction should in fact be called, 

 in the ordinary terminology, "intracalicular gemmation," although 

 it is increasingly clear to me that much of this terminology 

 will require revision. Por, as we shall see later on, this is in 

 all probability in reality a case of lateral budding. 



One result of this method of increase is that the original 

 parent polyp of the colony ceases very early to be recognizable. 

 Only in vertical sections could the identification be made, the 

 parent being the one which runs down into the apex of the 

 original epithecal cup. 



The growth forms assumed by Alveopora are not numerous. 

 If the budding takes place chiefly close to the epitheca, expanded 

 layers are formed with a prominent epithecal rim. If the bud- 

 ding takes place chiefly in the centre of the young colony, the 

 stock forms nodules or thick columns which may branch. In 



* Gegeiibaur's Festschrift, ii., 1896; see also Morph. Jahrb. xxiv., 1896. 



