402 



JOHS. BOYK i»i:ti:hsi:n 



Fig. 23. 



Pinniilari'i borealis "Ehrb. 



a. f. siibcapilata n. f. 

 h. - oiutlis — 



c. - liinceolata — 



these forms. :imon'> wliifh transilional forms mav no doubt hi' Ibiind. 

 anii tlie nalurc of wliich is on tlu- whok' doublfiil cf. Introiiiiclion . In 

 tlie Ici'hindic iiialerial I myscif liavc lomid llic forms j^iven bclow. If 

 in addition llie dimensions werc considered, a j^reat many more forms 



mij^hl l)c set up. I have. howcver. conlined 

 mysclf to thi' foUowinj^ whieh I name pro- 

 visionally. 



1. f. tfipica (A. Sihm. .\tl. Taf. 45. lig. 19, 

 20). The valve Hncar. with rounded or less 

 frcquenlly somewhal llattened apices. Siiows 

 considcrahle variation in size. 



2. f. Srhmidlii .\. Schm. Atl. Taf. 45, fii^. 1 (> ). 

 Tiic valve hncar. wilh more or less rostrate- 

 truncate apices. The valve often slightly con- 

 strii-ted in thc middle. This form approachcs 

 P. htla V. lidhcnhorstii. 



3. f. sul'ntfjilnld n. I. The valve narrowly 

 lanceolale, with roslrate-subcaj)itale a|)ices. 

 ri<^. 23 a. 



4. f. ovalis. The valve oval in outlinc. 

 aboul Iwicc as long as broad. Fig. 23 b. 



5. f. lanceolaki. The valve rhombic-lan- 

 ccolate, 3 linies as Ion« as broad. Fig. 23 c. 



F. Ii/pica is much the commonesl of these forms. It occurred in 

 ncarly all Ihe samples ciled. 



F. SvhmuUii occurred in samples 8, 182, 218, 254. 



F. siibcnpilala is the commonesl form in Iceland after f. li/pira. and 

 forms intermediatc between them are often found. It occurred in samples 

 14, 41, 09, 121. 129. 135. 1()0. 161. 182. 19(). 217. 307. 



F. ovalis is pcrhaps merely abnormal. A few specimens of it were 

 found in samples 268 and 121. 



V. liincvolala is likewise infreipient. It occurred in samples 138. 

 185, 295. 



P. borealis has from olden times been recognised as an aerial spe- 

 cies and is probably spread practically all over the globe. Being so 

 common, it will always be |)resent in small quantilx in fresh water 

 samples Østrup found it in 195 samples from Iceland, but never in 

 greal number . It has been found on red snow in Anlarclica Fritsch 

 1912, p. 321 and on tiie inland ice of Greenland (Boye Petersen 

 1924 II, p. 315 , further in numerous localities in arctic, antarctic. tempc- 

 rale, and tropical countries. It will be easily understood tliat ils valves 

 are common in tiie atmospheric dust. Ostrup 1893. p. 140. This 

 species mav Iherefore correctly be termcd ubi(|uist. 



As regards ils occurrence in Iceland I may stale that it was found 

 in the grealest (|uantity in samples that did not contain much mnnure, 

 1 have, howcver, found it in almost evcry kind of locality. That it 

 withslands desiccation well will appear partly from Bristol's (1920, 

 p. 42,, partly from my own investigations i^oye Petersen 1915, p. 24), 

 and partly from my observations in Nature whiih wouM secm to show 

 that il may live in Iceland cvcn on the driest ground. 



