DIATOM MOTION. lOI 



In order to make what follows eudurable for the reader 

 who is not a specialist in diatoms, it will be necessary to 

 recall the main features of the diatom's structure. The living 

 substance, consisting of protoplasm with a nucleus, and 

 enclosing chromatophors and vacuoles, etc., is enclosed in a 

 peculiar, heavy, silicious box. This box is constructed on the 

 plan of a pill-box with its overlapping lid. The bottom of 

 the box and the top of the lid are alike known as valves. 

 The sides of the box are composed of two bands, one outside 

 the other, called girdles. The whole is known as a frustule. 

 These frustules vary endlessly in shape in the different species 

 and genera, though always of the described construction. 

 Also, they may grow in colonies, filaments, etc., or exist as 

 isolated individuals. It is the isolated species that mainly 

 concern us. Outside the silicious box is a layer of jelly, var- 

 iable in depth even in the same species, called the coleoderm. 

 It is ordinarily invisible in water, even when it has assumed 

 a very considerable bulk. It is not protoplasm, but an excre- 

 tion. Bismark brown has much affinity for it, though stain- 

 ing protoplasm with difficulty. Eosin, on the other hand, 

 prefers the protoplasm. By success in staining with these 

 two, fairly sharp delimitation may be had, at least when the 

 coleoderm is stiff enough to keep its shape under treatment. 

 This is not always the case. This coleoderm is of importance 

 in what follows. To return to the valves. A very large 

 number of species have valves more or less elliptical in shape, 

 giving the frustule somewhat the aspect of a boat. These 

 are the naviculoid diatoms. Their valves show conspicuous 

 axial lines running from each end toward the middle. These 

 lines are known as raphes. The raphe has been shown to be 

 a real cleft through the silicious wall, though so narrow that 

 the thinnest bacteria can scarcely pass it, but in the larger 

 species not too narrow for the same bacteria to insert their 

 vepy attenuated flagellae. Here, then, is one opening through 

 which the internal protoplasm may pass to operate upon the 

 environment directly. Now it is precisely the species pro- 



