1898] THE MOVEMENT OF DIATOMS 411 
ment in Pinnularia is next described, and will be found illustrated 
in figs. 6 and 7. It can be seen to greatest advantage in a lateral 
view of the diatom in Indian ink emulsion (fig. 6). Small granules 
of Indian ink, set in motion in the region of the anterior! terminal 
nodes, move outwards, and then flow, at some distance from the 
surface of the cell, as far as the central node of the raphe. The 
granules borne along by these anterior streams (gs.) are not closely 
packed, and exhibit a more or less active independent move- 
ment amongst themselves. Arrived at the middle of the cell the 
streams of granules turn sharply towards the openings of the central 
nodes, forming there larger or smaller accumulations, and from these 
points arise granular threads (gt.) which always run obliquely back- 
wards, forming an acute angle with the surface of the frustule. In 
watching the formation of these granular filaments, Lauterborn ob- 
serves that the process strongly suggests the forcible expulsion, by 
fits and starts, from the central node, of a gelatinous thread, to.which 
a single row of granules adheres. This view is strengthened by the 
circumstance that the granules at the moment of union with the 
thread at once lose their active molecular movement perceptible 
whilst within range of the anterior streams of granules, also that 
the granules are all moved backwards intermittently with the same 
velocity, as the thread lengthens, suggesting the presence of a 
common substratum binding all of them together. It often happens 
that the series of granules is interrupted at some point, nevertheless 
the direction and intensity of motion in the granules behind such a 
gap correspond with those in front of it, and this circumstance 
appears to offer additional evidence of the presence of a common 
bond in the shape of a gelatinous thread, absolutely invisible in 
clear water until brought into view by the adherence to it of foreign 
bodies. These threads, which lengthen as the diatom moves onward, 
may be five or six times as long as the frustule. 
The phenomena described above are best observed when the 
diatom is lying on its side, as in fig. 6 ; but in a surface view (fig. 7) 
the streams of granules along the raphe on both sides of the frustule 
can be demonstrated by altering the focus. Occasionally the granule 
stream and thread on one side of a diatom move in the opposite 
direction to those on the other side, so that the two streams 
neutralise one another to a certain extent, and sometimes only one 
granular thread is developed. 
In a paper published in 1891 Biitschli considered it highly 
probable that a causal connection existed between the backward 
prolongation of the granule threads and the local movements of 
the diatom, stating that both he and Lauterborn were disposed to 
1 The terms ‘anterior’ and ‘ posterior’ are here used with reference to the direction 
of movement of the diatom. 
