Mr. H. J. Carter on the Genus Rossella. 113 
alimentary apparatus will afterwards be developed. Near the 
boundary between the middle and the girdle-like parts this 
layer entirely ceases, and in the girdle-like part (region of the 
formation of the arms) and further to the pole of the nutritive 
vitellus and round the latter we only meet with the cells of 
the dermo-muscular layer, as has already been stated. Con- 
sequently the lower or intestino-fibrous layer of the second 
germ-lamella, as may easily be seen, originates by transverse 
division of the originally one-layered second germ-lamella, and 
therefore in the same way as the latter lamella itself from the 
cells of the one-layered blastoderm or the upper germ-lamella. 
The cells of both layers of the middle germ-lamella are always 
rather smaller, but are more numerous than those of the upper 
lamella. In form they are generally oval, not unfrequently 
extended (in the wall of the yelk-sac); their protoplasm is 
dark, fatty ; and the nucleus (or often two) enclosed in each 
cell can scarcely be detected without reagents. None of the 
cells of either the second or the upper germ-lamella contain 
any trace of membranes. 
It is not without a purpose that I have dwelt so long on 
the mode of formation, the individuality, and the distribution 
of the first two germ-lamelle, seeing that the only extant 
memoir treating of this subject (namely that of EK. Metschni- 
koff*) is not quite sactisfactory. In the first place, this 
naturalist has not recognized the second or inferior lamella 
(“ parenchymatise ’’) as the middle one; and secondly, he has 
not referred to its cleavage into the two layers above described, 
which play so important a part in the formation of the em- 
bryonal organism. I regard it as almost unnecessary to add 
that my wearisome investigations of the development of four 
different species of Cephalopoda completely contradict the 
opinion put forward by Kélliker +, according to which both 
the germ-lamelle are denied to the Cephalopoda. 
[To be continued. | 
XIV.—On the Genus Rossella (a Hewactinellid Sponge), 
with the Descriptions of three Species. By H. J. CARTER, 
HR. ce, 
[Plate X.] 
In 1872 I published some figures of two forms of sponge- 
spicule which were found abundantly adhering to fragments 
of a Tethya (T. antarctica, C.) that had been dredged up from 
* Toc. ct. p. 19. + Loe. cit. p. 167. 
Ann. & Mag. N. Hist. Ser. 4. Vol. xv. 8 
