SAND-GRAINS AGGLUTINATED INTO FLOCCULENT MASSES. 185
dissociated from one another, but joined in small agglomera-
tions. Moreover, in many cases it could now be seen quite
plainly (figs. 1, 2 and 3) that the only connections between the
grains were the cells of the Scenedesmus. Certainly it was not
possible to prove that this was true in all cases, but the difficulty
in doing so appeared usually to be due to the overlapping of the
sand-grains, thus making it impossible always to get clear views
of the spaces between them. In a very few cases sand-grains
were seen adhering to one another without the aid of the algae, or
Sand-Grains Agglutinated by Scenedesmus Antennatus.
indeed, of any visible binding material, but, even so, the possi-
bility exists that the cementing substance may have been derived
from specimens of the Scenedesmus previously adhering to the
grains, but subsequently detached.
The adhesion of the cells of the alga to the sand-grains was
always by the slightly expanded or knob-like ends of the hyaline
cusps of the cell. These knobs, which are one of the character-
istic features of the species, evidently secreted or, at any rate,
were covered with some sticky material, for not only were the
cells attached to the sand-grains by their means, but in the