Acta Societatis pro Fauna et Flora Fennica, 49, N:o 7. 55 



The general proportions well agree with those of 

 A. glaucesceus although its size is but one third of the 

 measures given by Wittrock. It seems to be a very 

 doubtful question whether this species is a desmid at all 

 or perhaps a Tetraedron. It is by no means impossible, that 

 there are two species, very like each other, as maintained 

 by Dr Borge: one Arthrodesmus and one Tetraedron. 

 The cells under my observation were most probably an 

 Arthrodesmus. 



Hab. — (SW?), leg. Elfv. — PI. 3 fig. 54-55. 



A. convergens Ehr. var. incrassatus Gutw. Forma 

 spinis longis rectis divergentibus, membrana in cellulis a 

 latere visis conspicue incrassata. Long. 42, lat. cum spin. 74, 

 lat. sine sp. 42, isthm. 9,5, crass. 22, long, spinae 16 \i. 

 Compare also Borge, Siissw. Chlor. Archang. pi. 3 f. 35. 

 [N:o 843]. 



Hab. Gov. of Archangel. — PI. 3 fig. 52—53. 



A. octocornis Ehr. Forma spinis inferioribus bre- 

 vioribus, superioribus plus] horizontaliter dispositis, apice 

 minus concavo. Cum spin. long. 34, lat. 34 ,«. [N:o 1066]. 



Hab. Karkku (W), leg. Hj. Hjelt. PI. 3 fig. 56—57. 



A. impar (Jacobs.) nob. (Synon. = Xanth. octocorne 

 f. impar Jacobs., Desm. Danem. pi. 8 f. 28) Cellulae parvae 

 medio incisura introrsum non acuta profunde constrictae. 

 Semicellulae trapeziformes lateribus et apice valde excavatis, 

 angulis quasi productis inflatisque et spinis singulis (vel 

 interdum geminatis) longis firmis oblique sursum directis 

 instructis; a vertice visae depresse ellipticae medio necprotube- 

 rantia nee incrassatione ulla. Cum. spin. long. 76 — 80, lat. 

 76—80, sine spin. long. 42—49, lat. 34—42 isthm. 10—13, 

 crass. 19, long. spin. max. 23 ,a. [N:o 761]. 



West has suggested the relationship between Jacob- 

 sen's form and X. smithii (Vide Mngr. IV page 113). 

 Again, the Karelian form, which seemed to agree very well 

 with Jacob sen's description and figure, is doubtless an 

 Arthrodesmus: the central area in the semicells is always 

 undifferentiated and the spines are normally unpaired. It 



