434 JOHS. bOYi: pktkhsi.n 



hc callt'd Sihizoj/Diiiiim crispiim und .S. mumle. Tlic former was statod 

 to bc rccofinisabk' by ils soIIit. as a rulc sorncwbal tbiniicr lilamonls, 

 which sbowfd a jirealcr toiidoiicy to form a Ihallus ol" a band-likc or 

 Hat oxpaiulcd lorin. and hy ils oflcn bein}" funiisbcd wilb rhizoids. 

 S.iviinih'. on tbc olber band. was supposed to bc roc-oj^nisable by its 

 slider lilainenls. as a lule soinewbat tliicker and destitulc ot" rliizoids. 

 but oc'casionaliy fiirnislKHl wilb |)hirici'lhilar side-brancbes. and scddom 

 attainini^ a fiirliier dcvelopnienl tlian thc lormation ol' llie band-like form. 

 Brand, liowever. bas sbown 1. c.) tbal thcsc cbaraclcrs do not ahvays 

 occiir tofjc'lbcr. Tbiis he linds tiial tiiose Horniiitinni-forms tliat sbow 

 tbe grcatost Icndoncy to form a Prasiola Iballiis arc nol Ibe Ibinncsl 

 ones, bul on tbc contrary as a lulc Ibc Ibickcst. 1 Imvi' arrivcd al tbe 

 same result on the basis of tbc present material. In eacli of tbc samples 

 al band I bave mcnsiircd tbe tliickncss of tbe Ibickcst and liie liiinnest 

 //n/7;i/(/H//;i-lilamenl I could lind and lakcn tbe averaj>e. All Ibe averaj^es 

 for tbc samples in wliieb tbere occurred longitudinal divisions, i. e. for- 

 mation o[ Schizononiiim- or Prasiolu tballus. wcre added and divided by 

 tbe nnmber-. and tbe averaj^es from tbe samples in wbicb longitudinal 

 divisions did not occur were trealed in a similar way. Il Iben jjroved 

 tbat the ^Hormidiiim lilaments in samples wilhout longitudinal divisions 

 of tbe lilaments showed an average tbickness of 12.6//. wbile in the 

 samples wilb longitudinal divisions tbev bad a tbickness of 14.9 /r Nor 

 do 1 Ibink tbat tbe ])rcsence or absence of rbizoids can be used as a 

 distinguisbing cbaracter. In this respect tbe difference between tbc 

 various growlhs is more probably due to cxlernal conditions. Hence I 

 folldw Brand and unite tbe two species under one. 



Brand ,1914, p. 305; states tbat be bas not in any case been ablc 

 to sbow thc presence of starch wilb iodinc in lliis s|)ecics. I iiave 

 treated most of niy samples wilb cblor-zinc-iodine. and afler this treat- 

 ment I found a distinct laycr of small slarcb grains surrounding tbe 

 pyrcnoid in almost all spccimens Cf. also Wille 1901. j). IS. 



Pnisiola ciis])a is presumably in Ibe main a terreslrial alga but 

 wbile it can grow at considcrable beigbls above the ground in Denmark 

 when tbc light conditions etc. are favourable. thus e. g. on trec-trunks. 

 it never reaches sucb beigbls in Iceland. Tbere it is predominantly 

 found on tbc ground, but Ibence it spreads up tbe lower jkhI of tiie 

 turf walls of tbe houses, especially in piaces where tbe water runs 

 down in rainy weather. Cf. Hclgi .huisson 1895, p.54. 1900, p. 58, 

 59j. Thc species is also vcry common on bird-clilfs. a laet which is 

 well known from otber |)arls loo. I myself have only bad tbe oppor- 

 tunity of examining the bird-cliffs of thc Vestmannaeyjar. 



It rarely occurs on woodwork in Iccland; I bave only found il 

 sparingly developed on a beam in IsafjortVir. and in a Plcnrococciis 

 form on a fencc |)ok' at Moiiruvellir in Kjos. In a couplc of placcs I 

 found it on <lamp walls, and on large stones in thc ravine 'St(Sra (Ija« 

 near Myvaln. Twice I found il on loosc lava gravel, on the edge of an 

 <dd craler at SkutuslaiMr, and on tbc beath between the Isafjon^ur and 

 Flateyri in N. W. Iccland. Apart from tbcsc laller localities I have only 

 found it ncar human habilations or the haunts of animals (b'''<I-fl''1^- 

 caves in which shccp sheller), and 15 rand is no doubl right when be 



