338 On the Structure of Heteropora neozelanica, Busk. 



in tins part of their course (and it is one that I have never 

 failed to recognize) is that their cavities are here crossed 

 by transverse calcareous plates or "tabulae" (the "septa" 

 of Prof. Busk and Mr. Waters), which, though few in 

 number, are "complete" and in every way well developed* 

 (fig.2,C). 



On the other hand, in the peripheral portion of their course 

 (where the appearances are precisely the same as in the cor-. 

 responding region of a transverse section) the tubes have very 

 much thickened walls, and the walls are crossed at right 

 angles by numerous canaliculi, which open at both ends into 

 the cavities of the tubes by trumpet-shaped apertures. In 

 all parts of the section, also, where the inner surfaces of 

 the tubes are brought into view, these exhibit numerous 

 rounded apertures or pores, which represent the mouths of the 

 said canaliculi, and which have been well described and 

 figured by Prof. Busk and Mr. Waters {he. cit.). It is 

 very difficult in the outer part of these longitudinal sections 

 to distinguish between the proper zocecia and the interstitial 

 tubes or cancelli, their size being very much the same, and 

 their internal structure being exactly alike; and this leads me 

 to make a few remarks upon another point. When, namely, 

 such a section as I now speak of is examined with the 5-inch 

 objective, it is seen that the wall separating contiguous tubes 

 exhibits a central light space, limited on both sides by dark 

 and definite boundaries, and crossed by the transverse canali- 

 culi which have been already described (fig. 2, D). There is 

 thus created an appearance of a central tube in the interior of 

 the wall ; or, rather, what I have here described as the wall 

 might possibly be taken to be really one of the smaller inter- 

 stitial tubes divided longitudinally. Apart, however, from 

 the difficulty of conceiving how the canaliculi could be con- 

 tinued across and through the cavity of an interstitial tube, 

 we have in tangential sections, as previously remarked, the 

 conclusive proof that this is not the case, but that we really 

 have to deal with the wall of the tubes. These sections, in fact 



* Transverse partitions or " tabulae " are well developed in various 

 other Polyzoa (e. g. JEidalophora, Ceramopora, and Heteroclictya), but, of 

 course, cannot be homologous with the " tabulae " of the Coelenterates. 

 In a recent number of the 'Annals' (ser. 5, vol. vi. p. 244) Mr. Carter 

 announces the discovery of transverse partitions or "tabulae" in the well- 

 known "stellate canals" of a Stromatopora, and adds that it is at once 

 thus "proved that the Stromatoporce could not have been sponges and 

 that they were Tabulate Corals." This conclusion could only have been 

 penned by Mr. Carter by inadvertence, since "tabulae" occur not only in 

 many corals, but also in various Hydroids, and, as just remarked, in 

 several unquestionable Polyzoa, to which last group some good observers 

 have referred the Stromatoporoids. 



