8 
THE VOYAGE OF H.M.S. CHALLENGER. 
tlie flagellated chambers in shape and size, and frequently occupy the place that would 
otherwise be taken by these chambers, but sometimes they project beyond the general 
surface of the sponge-lamella into the sinus of a canal. 
In their earliest observed stage they exist as single, finely granular, deeply stained oval 
cells situated in the. mesoderm, and about 0'0197 by 0 ’01 58 mm. in size (PL XLI. fig. 13). 
This primitive oval cell by repeated segmentation gives rise to cells continually in- 
creasing in number and diminishing in size, till the mature spermatozoa are produced. 
The sperm-clusters lie within definite cavities of the mesoderm, lined by so-called endo- 
thelium ; they are not provided with a “ cover cell.” The cavities are mostly larger 
than the clusters, owing to shrinkage consequent on preservation in spirits. The largest 
cluster of mature spermatozoa observed measured 0‘045 mm. in diameter, the con- 
taining vesicle 0‘064 mm. 
Some of the commoner stages represented furnish the following characters. In a 
cluster 0'032 mm. in diameter, the cells are spherical granular bodies 0‘0079 mm. in 
diameter, with a clear vesicular nucleus and small deeply stained spherical 
nucleolus. In more advanced clusters 0‘04 mm. in diameter, the cells have 
diminished to 0*004 mm. in diameter, the general protoplasm is much diminished in 
quantity, the nucleus with its darkly stained nucleolus constituting the greater part of 
the cell. In still more advanced stages scarcely anything but nucleus and nucleolus 
remain, the little spherical cells now measure 0*002 mm. in diameter. Finally we 
meet with the mature spermatozoa, the heads of a round or oval outline, 0*001 mm. 
in diameter, furnished with tails, the length of which it was not found possible to 
measure (PL XLI. fig. 20). The head has every appearance of being the transformed 
nucleolus. 
Problematical Body (PL XLI. fig. 21). — Occupying a position similar to that in which 
the sperm clusters occur, and deceptively similar to them in structure, is a curious little 
body of which a series of sections were obtained, one of which is shown in the figure. 
It is a spherical cluster of cells, each of which measures about 0*008 to 0*01 mm. 
in diameter ; their outline is circular or oval, and they consist of very faintly stained 
protoplasm, embedding a large oval deeply stained nucleus, 0*003 mm. in diameter, and 
one or more small, deeply stained granules, which are surrounded by a clear spherical 
space. That this is not a cluster of developing sperm-cells, is shown by the presence of 
a structureless, colourless, transparent membrane, within which the cell-aggregate is 
enclosed. By this feature also the possibility of its being a segmenting ovum of the 
sponge is excluded ; and it would appear to be a stage in the segmentation of the ovum 
of some other organism foreign to the sponge. 
