188 A. EF. Verrilli—A ffinities of Paleozove 
—and this, I think, can easily be shown to be the case. Thus, 
or example, the genus Celastrea V., an undoubted Astrean 
forming true tabule; the genus Alveopora (figure 1, a), and 
others allied to Porites and Madrepora, have true tabule; also 
the genus Astreopsammia V., of the Eupsammide ; the species 
of Pocillipora, a genus closely allied in its animals, and other- 
wise, to Oculina and Stylophora, have very numerous and per 
a Fig. 1.* b 
fect transverse septa; even among the Alcyonaria, the genus 
Ti ubipora occasionally has transverse internal septa ; and the 
same is true of Millepora, belonging to the class of Acalephs. 
Notwithstanding the very slight basis upon which the group 
of “ Tabulata” was established, and disregarding the very great 
and important differences which exist among the corals thus 
unnaturally brought together, most writers upon corals, whether 
recent or fossil, during the past twenty years have adopted this 
classification without hesitation. 
And yet this is but another instance forcibly illustrating the 
general rule that classifications based on single characters are 
very likely to be artificial and erroneous. It also illustrates the 
manner in which such an error often leads to others of still 
greater importance, 
In 1857 Professor Agassiz made the very important discovery; 
that the animals of Millepora are not true polyps, but genuine 
hydroids, belonging to the class of Acalephs or Meduse. u 
since Millepora is a genus belonging to the “ Tabulata,” he 
immediately concluded that all the “ Tabulata” are, therefore, 
ydroid Acalephs! And not content with this sufficiently bold 
* Figure 1 a, a longitudinal section of Alveopora spongiosa Dana; 1 8, a VO ; 
cal view of some of the cells; both much enlarged, copied from Dana’s atlas 
ytes of th ixpl. Exp. For the use of this cut I am ic 
debted to vont tate & Mead, the publishers of Professor Dana’s new work on 
Siands. 
a Proceedings of the ‘Boston Society of Natural History, vol. vi, p- 373, | - 
oe fur Pourtales, in Illustrated Catal. of the Mus. of Comp. Zodlogy, No. !¥ P 
