130 
ME. H. N. MOSELEY ON THE 
Conclusions. 
The discovery of the hydroid nature of Millepora was made by Prof. Agassiz as long 
ago as 1859*. Agassiz determined that there were two forms of zooids present in the 
coral — the one form broad, with from four to six knobbed tentacles ; the other slender, 
with numerous similar tentacles disposed along their whole length. His conclusions 
with regard to Millepora have received only partial acceptation ; they have been uni- 
versally adopted in America and Germany. Claus and others have placed Millepora 
with Hydroids in their zoological systems. M. M.-Edwards, however f, did not accept, at 
all events in 1860, Prof. Agassiz’s results, and more recently Prof. Allman has 
expressed his uncertainty in the matter on the ground of our ignorance of the genera- 
tive system of the Milleporidee J, and awaits the result of further researches. Although, 
most unfortunately, no evidence as to the structure of the generative system of Mille- 
pora was obtained by the present investigations, the results obtained yield, nevertheless, 
I think, convincing proofs that this interesting form of coral is a true Hydroid. The 
peculiar structure of the hydrophyton, the forms of the zooids, the absence of all trace 
of mesenteries, the apparent septa present in the tentacles, the presence of the thread- 
cells of the form peculiar to Hydrozoa, and in fact every item of histological structure 
point irresistibly to the same conclusion. Professor Agassiz considered the Millepores 
to be allied to the Hydractinise, and Claus remarks on their resemblance in some points 
to the Corynidse. Both Hydractinia and Podocoryne resemble Millepora in having a 
hydrophyton which forms a continuous encrusting layer ; and in essential structure the 
hydrophyton of these two genera seems closely to resemble that of Millepora. The 
genus Podocoryne (Sars) has a “ hydrophyton consisting of a continuous adherent expan- 
sion formed by adnate inosculating canals, the deeper part, with its component canals, 
invested by a chitinous perisarc, while a layer of naked ccenosarc spreads over the free 
surface” §. In Millepora the canals are not adnate, being separated by the stout trabe- 
culae of calcareous matter which here take the place of the chitinous perisarc. The 
layer of naked ccenosarc on the surface is probably homologous with the layer in the 
hydrophyton of Millepora described in the present paper as the superficial layer of the 
ectoderm. The structure of the hydrophyton of Hydractinia is essentially similar to 
that of Podocoryne. Distinctive features in the hydrophyton of Millepora are the pre- 
sence in it of calicular excavations into which the zooids are retracted, the presence of 
large main branching canals, and the formation of successive superposed layers of 
hydrophyton, and consequent formation of lines of growth and tabulae in the calcareous 
skeleton. In having zooids of two kinds, mouthed and mouthless, the Millepores 
resemble Hydractinia echincuta , which bears likewise alimentary (mouthed) and spiral 
(mouthless) zooids. In the form of the zooids, however, and shape and arrange- 
* Bibl. Univ. de Geneve, Arch, des Sciences, Mai 1859, p. 80. 
t Hist. Nat. des CoraUiaires, t. iii. p. 224. 
$ Allman-, 1 Gymnoblastic and Tnbularian Hydroids,’ p. 3. 
§ Allman, l. c. vol. ii. p. 348. 
