408 



JOHS. BOYK PKTKRSKN 



sU'dtii. Tilis vniicly is rt'foidt'd to iiavt- Ikcii IbuiKt l)y Ost r up in 13 

 samples from frosli watcr in Icclaiid. and is porhaps commonest in 

 watcr. bnl as alri-ady statcd, Laj^c rs t i'd t loiind il on i-aitli and mos- 

 ses« \1. c. , wliile Hu s ledt iound il on Uiipmuu mounds I. c- |). olil)). 



Pinnularia parva (Grej?. Cl. var. minuta Oslr. Østrup 1918. p. 3(), 

 Tab. IV, lig. 55. Pinnularia Ballouriana .\. Schm All. l'l. .313. lig. 29— 31. 

 (non Cli'vo' Bristol 192(1 p. (55. tcxt lig. 8, 3 5'.' Kig. nostra 29. 



L. 11— 23/^ br 1-6/^ str. 10-12 in Kl//. 

 Valvc linear or sliglitly lanccolatc witli rounded 

 or Iruncate apiccs. .Slriation sliorl. marginal, soino- 

 timcs uninterru|)tt'd all the way round, sometimes 

 interru|)tc(l in tlic middle eitlicr on one or both 

 sides. Slriæ nearly parallel, almost iinpercc|)libly 

 radiale in the middle and converging at the apices. 

 Ajjical area broad. aboul lialT as hroad as tlu' (.•oli. 



E. Icel. 25. 41, L. 61. L. 68, 69. L. 78. 92, 129 — 

 N. Icel. 133, 137. 218, 252. L. 253 - W. Ice 

 327 - S. Icel. 353. 



Fig. 29. Pinnularia 



parva (Greg.) Cl. 



v;ir minuta Ostr. 



X 1200 . 



297, 



Pinnularut Balfoiiriana \\a^ first published by (".leve fSyn. II. ji. 80, 

 Pi. I, lig. 18i alter a manuscri|)t descriplion by G r u n o \v accom|)anied 

 by a drawing by the same. The drawing shows a P. nearly elliptical or 

 oval in form wilh somewhat shortened slriæ, so thai the apical area is 

 cnmparatively broad. The slriæ radiale slightly, most near the a|)iccs. 

 The valve has the foUowing dimensions: L. 8—10 //, br. 4 //, sti\ 10 in 10 //, 

 and the species is classed under Disldulcs. Hu stedt gives a quite siniilar 

 figure (1922, p. 101, lig. 10, and 1924, Taf. 21, fig. 9, whereas the spcci- 

 mens ligured in A. Schm. All. PI. 313, ligs. 29— 31 belong to anothcr spe- 

 cies which can hardly be classed under Distantes but more probably 

 under Brcoislriuta: This form is almost Iwice as large as P. lUilfuiiridiui, 

 the striæ are parallel or slightly convergenl towards the apices, the 

 valve is linear or in small specimens slightly lanceolatc. The apices 

 are rounded or more fre(picntly almost truncate. This form which is 

 jirobably the same that is ligured by Miss Bristol fl. c.) is, as far as 

 1 tan judge, identical with N. jmrua var. minuta Øslr. I have tried in 

 vain to prove this (piite conclusively by examining Østrup' s pre|)aia- 

 tions. It is recorded to have occurred in one sample from llvibi col- 

 lected by Art h. Feddersen. The latter, however, collecled 3 samples 

 of algæ from Hvita. and in one sam|)le especially (puddles near Hvita) 

 there were sevcral specimens which I should refer to the present variely, 

 but none which had exactly the dimensions given by Østrup. Hence 

 I can hardly have seen the original specimen drawn and measured by 

 him, ncverthcless I do nol doubt thai il must bc rcferred to the same 

 variely as the specimens I savv in the |)reparalion . most of which had 

 no fascia and in which the outline was slightly lanceolatc' Since 

 Østrup cites .\. .Schm. Atlas le. under P. BalfourUuia it may be pre- 



') (Jn mcMsiiring a spfcinu-n I fouiul the followinf^ (limcnsions : L. 17/'. hr. \.f^ii. 

 str. 10 in 10/(, conscqiiciitly it was somewhat hroadcr tlian tlic specimen descrihed 

 by Øst ru p. 



