166 - T. Holm — Studies in the Cyperacece. 



From C. cryptocdrpd our species may be readily dis- 

 tinguished by the characters mentioned by Drejer, and as 

 may be seen from our figures (figs. 12-14) the squamae 

 and the perigynium, especially the former, exhibit a struc- 

 ture distinct from that of C. cryptocarpa (figs. 4-6). 



Carex cryptocfolcena nob. 8 



The most robust species of the Cryptocarpa, and very 

 characteristic by its heavy , cylindrical, pistillate spikes, 

 which are sessile or short-peduncled, erect or spreading. 

 So far as concerns the habit it resembles C. magnified 

 Dew., but the structure of the scales and perigynia are 

 more like that of C. cryptocdrpd; the large, pale green 

 perigynium is sparingly denticulate along the upper mar- 

 gins, a structure which has not been observed in C. crypto- 

 cdrpd, but in C. prionocdrpd, as stated above. 



With regard to the habitat Dr. Walter H. Evans col- 

 lected the species in Alaska: Kussiloff, on sands with 

 Elymus, while C. cryptocdrpd inhabits wet marshes, 

 borders of ponds, and meadows near the coast. 



As may be seen from the preceding pages the geo- 

 graphical distribution of the Cryptocdrpd is very limited : 

 C. cryptocdrpd inhabiting the shores of the northern 

 Pacific and Bering Sea, beside Iceland; C. Lyngbyei 

 known only from Faroe Islands ; C. cdpillipes only from 

 Iceland; C. prionocdrpd from Japan (Yeso) ; C. Jicema- 

 tolepis only from South Greenland, and finally C. crypto- 

 clil&nd from Aldskd. 



And while these species differ in the extreme forms, 

 they do unite in some characters, as if their descent might 

 be traced through modification from some common type. 

 We have seen the very important distinctions . derived 

 from the structure of the perigynium, the squamae, the in- 

 florescence to some extent, and to these may be added that 

 C. cryptocdrpd, C. prionocdrpd and C. cryptochlcend are 

 dphyllopodce, C. Lyngbyei, C. hmmatolepis and the Green- 

 land C. cryptocdrpd, on the other hand, are phyllopodce; 

 with respect to C. cdpillipes, the rhizome being not repre- 

 sented, we cannot decide whether it be aphyllopodic or 

 phyllopodic. 



The nearest allies of these species may be sought 

 among the Sdlince, and quite especially in the var. Kdtte- 

 gdtensis of C. salind, even though all the Sdlince are 

 phyllopodcB. The present geographical distribution in 

 connection with the association with certain types of the 



8 Holm, Theo. : New or little known Carices from Northwest America, this 

 Journal, vol. 20, p. 305, October, 1905. 



