516 DESMIDS AND DIATOMS. 
glassy bar takes upon itself an individuality, and though 
still connected with the rest, moves as with an impulse of 
its own. The middle bar alone is motionless. Those on 
either side slide gently on the last, the one to the right, 
the other to the left, with clock-like regularity. The 
movement of the first is the signal for all the rest. Each 
in turn slips quickly along his neighbor’s side, until from 
a long and ribbon-like band, we now have a séries of 
glassy steps, each crystal bar resting slightly upon that 
below it. But the change is transient. When the whole 
series has thus unfolded, as it were, it begins to slowly 
recoil again. Each plant or bar resumes its former place, 
and the ribbon-like band again hangs motionless from 
leaf to leaf. 
The cause of these motions has been severally assigned 
to the action of minute vibratory cilia, to an undulatory 
motion of the outer membrane, and to the mechanical 
effects resulting from thé absorption or discharge of 
water. The subject, however, is one which yet remains 
in great obscurity. — To be concluded. 
EXPLANATION OF PLATE 13. 
igs. 1-5. Dersmrps,—1 and 3, Euastrum; 2 and 6, rs eT 
— fossils found in flint; 4, Micrasterias; 5, Closterium. 
-21. DIATOMS, — Fresh-water. 7, Navicula; Al Nitschia; 
9, baoa: 10, Pinnularia; 11, Eunotia triodon; 12, Meridion ver- 
; 18, Ta bellaria peo a, —a, front view, b, side view; 14, CY- 
clotella Zap aaa side view; 15, Acnanthes; 16, Diatoma floc- 
culosum ; 17, peonpi 18, Bacillaria paradoxa; 19, Mastogloia; 
20, Licmopho ra; 21, Odontidium 
Figs. 22-36. DIATOMS n,- Marine Forms. 22, Amphiprora; 23, Am 
ee , forming a zigzag chain, —a, a frustule about to divide into 
two, b, two Da newly formed but not yet separated, the “com” 
_ necting membrane” having fallen off; 24, Asteromphalus, a peautiful 
deep-sea form, taken from below 2,000 fathoms in the sea of Kams- 
chatka; 25, Podosira ; 26, Navicula didyma; 27, Triceratium; 28, Nits- 
