43 8 PHYSICAL GEOGRAPHY. 



in most they are undulate, so that on a front view they look somewhat 

 like written characters, which has led to the name Grammatophora being 

 given to the genus. 



Nearly related to the genus just mentioned is one known as Rhabdo- 

 nema. It likewise is found growing in chains attached to algae, and 

 similar submerged objects in the ocean. In this the septa are not con- 

 tinuous, but look as if they were punctured with several holes. The two 

 last mentioned genera are attached to their supports by a small gelati- 

 nous cushion, but there is a genus named AcJinantJics, in which this 

 cushion is lengthened out into a long stalk, and as it is attached to one 

 of the corners of the frustule, the whole individual looks very much like 

 a flag floating out straight from a staff. This stalk is called the "stipes," 

 and is remarkably developed in other genera. Thus, Gomphonema con- 

 sists of a number of wedge-shaped frustules attached by their pointed 

 ends to long and forked stalks, while in Synedra the stipes has again 

 shrunk down to a cushion. The frustules of Synedra are like little sticks 

 attached to the cushion by one end, and sticking out on all sides like the 

 spines of a porcupine. 



There are many other forms which the various genera of diatoms 

 present, but we have had space only to describe and figure a few of them. 

 The possessor of a microscope will soon, if he searches, come across 

 others ; and if he is encouraged to ask for more information concerning 

 the beautiful atomies he sees, as it is to be hoped will be the case, he will 

 be able, in the works of W. Smith, Rabenhorst, Kiitzing, Ralfs, and others, 

 to find them more fully described, and better and more thoroughly illus- 

 trated. So to those works we must refer the reader who desires to learn 

 more than we have been able to tell him in this little sketch concerning 

 the diatomaceae. 



PART SECOND. 



Movements of the Diatomaceae. 



It has been said that it is extremely probable that all species of dia- 

 tomaceae are at some period of their lives free, while for a short period 

 perhaps for some, but always for some time, they are fixed or attached to 

 some submerged object, as rocks, plants, woodwork, or similar substances. 

 This opinion is not held by all observers ; so much so that many, and 



