Mr. J. Ralfs on the species of Desmidium. 373 



straight, and are scattered in loose bundles in the water, or form 

 a thin gelatinous fleece at the bottom of the pool. They are ge- 

 nerally twisted in a regular manner ; this is shown by the waved 

 darker line in D. Swartzii, and in D. cylindricum and D. Bor- 

 reri by the apparent variableness in the diameter of the filament, 

 and by the different position of the angles, which causes the cre- 

 natures either to become less distinct or entirely to disappear at 

 certain intervals. 



Whether D. mucosum is also twisted, I am unable to determine, 

 the crenated appearance of the joints being caused by a shallow 

 groove running all round the joint : all the joints seem to be si- 

 milar, and if twisted would not present any perceptible difference. 



As the angles of the joints in all the species are, in a greater or 

 less degree, really or apparently bicrenate, this character is rather 

 a generic than a specific one. 



Most commonly each joint in the species of this genus has its 

 endochrome divided into two portions separated by a paler line, 

 which runs across between the crenatures, and in D. mucosum 

 corresponds with the groove passing round the joint. A trans- 

 verse view shows the endochrome disposed in a stellate manner. 



The species are found during great part of the year in clear 

 shallow pools or in old peat-pits ; those at present known are few 

 in number, and possess well-marked characters. 



* Filaments with a mucous sheath. 



1. D. cylindricum, Grev. Filaments slightly compressed, with two 

 angles ; crenatures strongly marked ; joints rather broader than 

 long*. Plate VIII. fig. 1 . Grev. Scot. Crypt. Fl. t. 293 ; Desmaz. 

 n. 1110. 



In old peat-pits about Dcjgelley, N. Wales ; sparingly near Pen- 

 zance. 



D. cylindricum, Swartzii, Borreri and mucosum, growing together 

 in a watercourse on Chiltington common near Pulborough, Sussex : 

 Mr. Jenner ; Cheshunt, Mr. Hassall. 



The filaments are as thick as those of D. Swartzii, which spe- 

 cies it much resembles in the water. The joints, inclusive of the 

 angles, which are colourless, are rather broader than long, and of 

 an oval form, with a small sharp notch in each angle. Those 

 joints in which the angles are not visible are about as long as 

 broad. The joints seem to be connected by a thickened border. 

 The mucous sheath is not noticed by Dr. Greville or Agardh, 

 whose description in the ' Conspectus criticus Diatomacearum' is 

 taken from Greville. The description in the ' Scot. Crypt. Flora' 

 was probably drawn up from dried specimens, in which the sheath 

 is less evident. 



* Length and breadth have here their usual meaning when applied to the 

 joints of the filament of a Conferva. 



