58 
by the acid. The outline of the valve is extremely varied ; it 
may be circular, triangular, quadrangular, pentangular, semicircular, 
wedge-shaped, fusiform, or wandlike. Some genera secrete a semi- 
gelatinous thread, and become parasitic upon larger plants. The 
cell contents consist of a golden brown endochrome, and sometimes 
two or more oily globules may be observed; cyclosis has been 
detected in several species when stimulated by light and warmth. 
That the colouring matter or endochrome is identical with that in 
the desmids, and the chlorophyl in the larger plants, is, I think, 
satisfactorily proved by spectrum analysis, the absorption band 
occupying the same position as the chlorophyl band. 
The most remarkable jDhenonienon connected with these organ- 
isms IS the power of locomotion, and the means by which it is 
accomplished have not as yet been made out. This movement is 
most conspicuous in the bacillar or wand-shaped forms, and con- 
sists of a series of jerks ; it has been used as an argument in 
favour of their animality, but if attentively studied, it will be 
found to bear little resemblance to that of even the lowest animal 
organisms. If its course is impeded by any obstacle, it does not 
move on one side, but remains motionless for a short time, and 
eventually goes back by a similar series of jerking motions. This 
movement has been supposed, by some authors, to be produced by 
the expulsion of minute currents, but unfortunately for this theory, 
the minutest particles of matter are not disturbed by the diatom, 
unless actually touched by it. Professor Smith has timed the 
rate of progression in several species, and he found that the most 
rapid moved over of an inch in a second, or rather less than 
an inch in three minutes, and the slowest of an inch in one 
second, or nearly an hour to accomplish one inch. 
At the risk of being tedious, I will, with your permission, say a 
few words on the reproduction of these organisms. You will, 
peihaps, remember I stated that the cingulum is the only part 
that grows ; within this, two other valves are found, and when 
this has taken place, the cingulum separates, and we have now two 
new frustules. Under favourable circumstances, this goes on with 
great rapidity, but as every new frustule is formed within the old, 
a gradual decrease of size takes place, and at last self-division 
ceases,^ and the species would die out, did not a reproductive pro- 
cess, similar to that in desmidia, place two frustules approximate. 
