REPORT ON THE TETRACTINELLIDA. 
173 
more or fewer of the spherical bodies are usually torn out of the cell, and those which 
remain are separated by clear intervals from each other ; it then becomes evident that as 
a rule no network of protoplasm is associated with these bodies, they alone constitute the 
contents of the cell. 
It is evident that some kind of relation exists between these pigment-cells and the 
cells previously described, and this is rendered certain by the occurrence in the latter of 
occasional spherical bodies of the same character as those of the pigment-cells (PI. XIX. 
fig. 21). The question then arises as to whether the pigment-cell is a metamorphosed 
protoplasmic network cell, or whether the latter is an exhausted pigment-cell. The 
latter would appear to be the more probable view ; the pale circular bodies of the proto- 
plasmic cells have much the appearance of pigment spherules which have lost their 
pigment. They frequently contain a small body about 0‘002 mm. in diameter, which 
shines brilliantly with reflected light, and which is evidently a contained air-bubble ; 
now air-bubbles when unconfined always work their way out of balsam, and their presence 
in these bodies is therefore probably due to confinement within some limiting membrane ; 
and the question suggests itself whether they may not occupy cavities left vacant by the 
exhaustion of the pigment, from pigment granules now represented by the pale circular 
bodies, in which the air-bubbles occur. The apparent absence of protoplasm from the 
pigment-cells may be more apparent than real, but till this is shown to be the case it 
leaves the question of the relations between the two kinds of cells in a very unsatisfactory 
position. 
In addition to the preceding cells a third element (PI. XIX. figs. 14, 15) enters 
into the composition of the ectosome, and this is of an altogether problematical character. 
It presents itself as minute oval or elongate finger-biscuit-shaped bodies, exceedingly 
numerous, evenly and darkly stained, evenly and very finely granular (fig. 15), and 
sometimes marked with faint, longitudinal, undulating striations (wrinkles ?) (fig. 14). At 
one end towards the side a small, more darkly stained oval body sometimes occurs, and 
may represent a nucleus. Each of these bodies, on an average O’Ol by O'OOG mm. in size, 
lies within an oval vesicle in the surrounding collenchyma, about O'OIG by 0‘012 mm. in 
size. 
Choanosome . — The mesoderm is a well-marked sarcenchyma, in which, however, oval 
ceUs containing a protoplasmic reticulum, like those of the ectosome, and pigment-cells as 
well, are abundantly present ; in addition are met with, lying each in its own cavity, 
large (0‘024 mm. in diameter) rounded or polygonal, deeply stained, granular, naked, 
protoplasmic cells, with a large oval nucleus O’Ol 2 mm. in diameter, which is surrounded 
by a darkly stained margin, and contains a spherical nucleolus 0’004 mm. in diameter. 
These cells are numerous in places, as many as seven sometimes being seen together in 
one portion of the field of the microscope. The figure (PI. XIX. fig. 13) represents one 
of these cells ; the wash of colour in the cavity it occupies should have been omitted, and 
