Development of T€tradm»i Cellidosiuii Hall. — Riiedemajui. 23 
like sections (fig. 14). The rounded ends may separate in toto 
(fig. 13) or divide again lengthwise and give off a branch ffig. 
14 a). The latter process, which is the more common of the 
two, together with the fast multiplication of the corallites at 
both ends of the bands, produces very broad branches, which, 
however, often contain only two and in some places even only 
one row of corallites. It is the peculiar appearance of these 
bands, which, no doubt, has induced some investigators to 
compare Tetradium with Halysites. But, as the develop- 
ment of this species shows, the origin of the chain-like coralla 
of Tetradmni is very different from that of Halysites the for- 
mer originating from a lengthwise splitting of round coralla, 
the latter from a continued stolonal gemmation in one direc- 
tion. 
Figure 15 shows that with the development of larger cor- 
alla a general and remarkable decrease in the size of the cor- 
allites, and a becoming more irregular of the shape of the lat- 
ter are connected. Sections of younger and older coralla 
present, on this account, a differing aspect and could easily 
be mistaken as representing different varieties or even species 
(compare figs. 15 a and b). In sections of older, massive cor- 
alla all corallites. are of nearly equal size and their plications of 
equal length; multiplication by fission, therefore, has appar- 
ently been greatly retarded. As only such sections previously 
came under observation, it can easily be understood why the 
four fissional plications continued to be regarded as septa or 
pseudosepta. With the septa of the Hexacoralla and Tetra- 
€oralla, they cannot be directly compared, because they have 
a different function, serving primarily the multiplication by 
fission. Neither can the term "pseudosepta," as used by Mose- 
ley* for slight plications of the wall of Heliopora , which in 
number and arrangement are independent from the mesenteries^ 
be applied here, for the pseudosepta of Heliopora do not de- 
velop into fissional partitions and the plications of Tetradium 
are too long to have been independent from the mesenteries. 
The nature of the plications, therefore, cannot be used 
to distinguish Tetradium from CheeteteSy the projections of 
♦Report on certain hydroid, alcyonarian and madreporarian corals 
procured during the voyage of H. M. S. Challenger. Voyage of H. M. 
is. Challenger, Zoology, vol. II, 1881. 
