156 PALMER : LOCOMOTION IN SURIRELLA. 



alum-hgematoxylin show the keel-canal more or less full of a 

 substance continuous with the protoplasm of the rib-canals, 

 and staining like it, and showing the same granular consti- 

 tution. ^ 



I consider it demonstrated, therefore, that the protoplasm 

 of Surirella extends through the ribs into the keels, wherein it 

 runs from end to end of the diatom and occupies most of the 

 tube-like canals. Altogether it is quite evident that this 

 hyaline protoplasm, and not any hypothetical ''Gallerte" 

 flows back and forth in these keels, and operating upon the 

 surroundings through the minute clefts (where indeed there 

 may be a slight film of gelatinous substance) propels the 

 diatom, rolls it over, and stands it upon its end. This is a 

 kind of apparatus for diatom locomotion easily understood ; 

 and besides being in comfortable agreement with the nature 

 and peculiarities of the phenomenon to be explained, it has 

 the not inconsiderable advantage of being demonstrable. 



*Tlie marked contrast between this keel-canal protoplasm and typ- 

 ical "Gallerte," so far as concerns granular structure, was well shown 

 in the stained preparations, where the coleoderm of Navicvila Iridis 

 showed more like a jelly containing only the most minute grantiles, all 

 nearly of one size ; — in short, what might be expected of a structureless 

 secretion. 



