16 BRITISH 



IRELAND. Errigal, Donegal ! Slieve Commedagh, 

 Down ! Frequent in Mayo and Gralwav ! 



V t- 



Geogr. Distribution. France. Germany. Italy. 

 Eonmania. Norway. Sweden. Denmark. Faeroes. 

 Greenland. Central China (a form). Celebes. United 

 States. Patagonia. 



Brebisson's original figure of Cosmarium latum does 

 justice neither to the angularity of this Desmid nor to the 

 disposition of its granules. The figure given by Messrs. Roy 

 and Bissett in 1894 was the first one to portray its salient 

 features. We find it impossible, however, to separate C. 

 latum as a species from C. cons per sum. The differences are 

 too slight, and many intermediate conditions exist. The var. 

 latum is much more generally distributed than typical C. 

 conspersum, and differs chiefly in its slightly greater breadth 

 and more convex apices. The granulation of the two forms 

 is very similar, and in both cases the cell-wall between the 

 granules is smooth. The ratio of the breadth to the length 

 in C. conspersum is 1 : 1'35, whereas in the var. latum the 

 average ratio is 1 : 1*18. 



NOTE : COSMARIUM LATUM Breb. var. MINOR Roy & Biss. 

 (Scott. Desm. 1894,, p. 105, t. 2, f. 11 [= C. similatum Roy 

 & Biss. MS.] ; Gutw. Wykaz. Glonow Wadow.-Makow. 1897, 

 p. 148 ; W. & G. S. West, Alga-fl. Yorks. 1900, p. 88 ; Wittr., 

 Nordst, et Lagerh. Alg. Exsic. 1903, no." 1481 ; faac. 35, p. 13). 

 This Desmid was obtained by Messrs. Roy & Bissett from 

 above Loch Etchachan on Ben Macdhui in Aberdeenshire, 

 and from Canlochan in Forfar, and we have obtained it from 

 Boston Spa and Oo-clen Clousfh in West Yorkshire. It was 



L C? O 



described as "very like the type [C. latum] in form, but 

 smaller and more closely granulated. Length 60-67 /i ; 

 breadth 48-55 jit ; isthmus 20-22 /u." We give a copy of their 

 figure on PL XCIX, fig. 7, but we consider that this Desmid 

 is not correctly placed with the forms of Cosmarium conspersum 

 Ralfs. The semicells are too rounded, and the granulation is 

 both finer and denser. Its dimensions and its granulation 

 are those of Cosmarium Logiense Bissett [vide Vol. Ill, PL 

 LXXX, figs. 1 and 2], with which species it should more 

 rightly be placed. It might be called C. LOGIEKSE Biss. forma 

 EXPANSA nob. 



Var. rotundatum Wittr. (PI. XCIX, fig. 3.) 



C. consperstimKalfs var. rotundatum "Wittr. Skandinav. Desm. 1869, p. 13, 

 f . 4 ; Wille, Ferskv. Alg. Nov. Semlj. 1879, p. 35 ; De Toni, Syll. Alg. 



