PROCEEDINGS OF THE BRITISH ASSOCIATION, 
87 
cular orifice freely with the interseptal spaces, and through them with the interior 
of the tentacula. The inner surface of the stomach exhibits numerous vertical 
plaits, corresponding with the attachments of the septa, two of which, more 
developed than the others, divide the animal into halves, giving it a bi-lateral 
symmetry, and internally at least destroying its radiated character. The gemmi¬ 
ferous bodies , about 200 in number, form elongated masses on the inner edges of 
the delicate mesenteric expansion of the septa. Under the microscope, each gem¬ 
miferous body is seen to consist of a compact layer of gemmides , round at first, 
but changing their form from mutual pressure as their development increases, and 
indicating by a well-marked central depression, the situation of the oral aperture, 
but without tentacula. When the gemmules are large, they raise the investing 
membrane, and project considerably from the surface of the gemmiferous mass, in 
which state they may be easily detached by slight pressure with the point of 
a needle. There is no regular gradation of size in the gemmules from the top to 
the bottom of the gemmiferous mass, as if they successively arrived at maturity, 
as they are represented to do by Spix, but their size in each mass is uniform, 
except that there are a few small ones scattered indiscriminately among the 
rest. The development of the gemmules, however, varies in different gemmife¬ 
rous masses of the same Actinia ; and in different Actinia at the same season of 
the year, from March to November inclusive. The colour of the gemmules 
varies in the same species. Attached to the internal border of the gemmiferous 
body, from top to bottom, there is a soft minutely-convoluted vermiform filament 
of a milk-white colour, about the thickness of a Horse-hair, but so minute and 
delicate superiorly that the author has not been able to disentangle its convolu¬ 
tions, and detect its commencement; but inferiorly the filament increases in size, 
its convolutions are less intricate, and finally it passes in a simple wavy line to 
the stomach, where it terminates. During the life of the animal, the filament 
exhibits a distinct vermicular motion, and detached portions of it placed in sea¬ 
water slowly coiled themselves, and exhibited considerable locomotive power for 
a few hours, when it gradually ceased. The functions of the vermiform filaments 
are very obscure, some naturalists considering them to be oviducts, but in this 
view the author cannot coincide, for reasons which he assigns, and in particular 
because he has never been able to detect the patent extremity of the tubular 
filaments which this office implies, nor any of the gemmules in their progress 
through the tube, notwithstanding the frequency of his observations, the immense 
number of the gemmules, and the great length of the filaments exceeding the 
length of the animal some thousands of times. In the absence of direct evidence, 
the author suspects that the filaments are elongated follicular glands, analogous 
to the salivan r , pancreatic, and hepatic follicles of animals a little higher in the 
scale of organization, supplying secretions subservient to the digestive process. 
