64 
POPULAR SCIENCE REVIEW. 
group by M. Milne-Edwards, belonged to the genera Mittepora , Heliopora 
(representing the family Milleporidae), Pocillopora (referred to the Favositidse), 
and Seriatopora (the type of the Seriatoporidae) ; and the fossil forms were 
more or less closely associated with these in accordance with the resemblances 
apparently presented by the respective polyparies. The great majority of 
the fossils were from palaeozoic deposits, and the well-known forms of 
Favosites and Halysites may serve as typical examples of the group. 
Although MM. Milne-Edwards and Haime established the section 
Madreporaria Tabulata for the convenience of having some group to which 
to refer a considerable series of fossils with which they were obliged to deal, 
it is very clear that they felt by no means convinced that it was a sound 
zoological group; and as early as the year 1860, we find M. Milne- 
Edwards himself, in the third volume of his Histoire Naturelle des Coralliaires 
(p.223), remarking that * according to recent observations made by M. Agassiz 
upon the Millepores, it is probable that it will have to undergo great altera- 
tions, or perhaps be placed entirely in another class of the animal kingdom.’ 
This passage alludes to the discovery by Agassiz of the Hydrozoal nature 
of Millepora , which has since been amply confirmed by Mr. Moseley during 
the voyage of the Challenger , and by Professor P. Martin Duncan from the 
researches made many years ago by General Nelson at Bermuda. Further 
investigations have shown that notwithstanding the correctness of Professor 
Agassiz’s observations, the removal of all the Tabulate Corals to the Hy- 
drozoa, as suggested by him and supported by some other naturalists, would 
have been a mistake, the recent genus Pocillopora, especially, proving to be 
a true Zoantharian allied to the well-known Oculinidae. The morphological 
importance of the ‘ tabulae,’ or transverse floors, the presence of which con- 
stituted the chief bond of union between the members of the group, was also 
shown to have been overrated, as similar structures occur in a good many 
genera of undoubted Zoantharian corals. The combined researches of 
palaeontologists in fact showed pretty clearly that the group would never be 
removed ‘ en entier,’ as suggested by M. Milne-Edwards, to any class of the 
animal kingdom. 
Professor Nicholson enters in considerable detail into the history of the 
section Tabulata, which he shows clearly must be 1 broken up and undergo 
redistribution.’ The performance of this task is, however, a matter of no 
small difficulty, the great majority of the species referred to in this group 
being old Palaeozoic fossils, with no very distinct relationships to any existing 
forms. Our author is inclined to recognize about twelve distinct groups of 
these corak, the affinities of which he discusses (pp. 12-29). The Mille- 
poridae, here greatly reduced in compass, must, as we have seen, be referred 
to the Hydrozoa in the vicinity of Hydractinia ; and the Pocilloporidae have 
been shown by Professor Verrill’s researches upon the animal of Pocillopora 
to be Zoantharians allied to the Oculinidae. Seriatopora, a recent coral, and 
with Milne-Edwards the type of a distinct family, is regarded as so closely 
related to Pocillopora, that it must also be transferred to the same neighbour- 
hood. The Favositidae (Honey-comb Corals) which constitute by far the 
greater part of the section as proposed by Milne-Edwards, are pretty gene- 
rally regarded as true Zoantharia, to which the Columnariadae and Syrin- 
goporidae also probably belong, although their precise position is much more 
