38 F. E. FRITSCH. 



Diam. fil. ad bas. = 6 p (fil. inflat. 9-1 0/x); diam. trich. ad has. = 5 /A (trich. 

 iiiflat. 8-9 /A) ; diam. fil. in media parte = 3 5 /x ; diam. trich. in media parte = 3 /A ; 

 diam. heterocyst. = 5 5 /A ; diam. spor. = 5 /A ; long. spor. = 8-9 /A. 



Ilab. Gap pond, Winter Harbour, January 12th, 1904. 



The often much elongated and rather narrow filaments of this species present a 

 very graceful appearance under the microscope. In many cases there is quite a 

 gradual attenuation from base to apex (figs. 165, 172), but occasional filaments show 

 a prominent inflation of the base, as seen in fig. 166. False branching of the filaments 

 is rare (fig. 164). The heterocysts are slightly wider than the trichomes and are only 

 rarely found in an intercalary position (figs. 164, 170) ; their shape is remarkably 

 diverse (figs. 165, 166, 171). Only one filament with spores was seen (fig. 168), and 

 it is hardly likely that they were mature. 



The species comes near to C. sandricensis (Nordst.) Schmidle ( = Lophopodrum 

 sandvicense Nordst.), a form which has as yet only been found in the Southern 

 Hemisphere. C. gracilis differs from it in the different shape of the spores, in the 

 absence of a hair-like termination to the trichome, in the simple sheath, and the diverse 

 shape of the heterocysts. It also resembles C. cartilayinea G. S. West, superficially, 

 but this species has no heterocysts and its trichomes are attenuated into hairs. 



SCYTONEMATACE.E. 



Genus TOLYPOTHRIX Kiitz. 



57. TOLYPOTHRIX CONGLUTINATA. 



(PI. Ill, figs. 145, 146.) 



Tulypothrix eonglutinata Bovzi, Note alia Morf. o Biol. d. Alglie Ficocrom., Nuov. Giorn. bot. Ital., xi. 

 (1879), p. 371 ; Bornet et Flahault, Rev. d. Nostoc. hcterocyst., Aun. Sci. nat., Bot., 7 ser., v. (1887), 

 p. 125. 



Diam. fil. = 16-1 8 /A; diam. trich. = 8-9 yu, ; crass, parietis vagiuse = usque 

 ad 5 ju,. 



Halt. Freshwater pond, Granite Harbour, January 20th, 1902. 



Unfortunately only a single specimen of this form was found, but that was fairly 

 convincing. It formed an intricate mass of numerous branches, to which fig. 145 does 

 no degree of justice, but it was impossible to figure the confused tangle in its entirety. 

 The false branches were very numerous and followed closely upon one another. 

 They were superposed to such an extent as to make their course very difficult to 

 follow ; in many cases they were sharply incurved and more or less agglutinated with 

 the main filament and other branches. The filaments had very thick sheaths (up to 

 5 /A in thickness) with a rough edge, the sheaths being hyaline and not stratified 

 (fig. 146). The cells of the trichomes were very flat (often 3-4 times as broad as long) 

 with somewhat granular septa, the trichomes being slightly constricted between the 



