WELWITSCh's AFRICAN FRESHWATER ALG^E. 271 



2. PoRPYROSiPHON Kiitz. (1850-52). 



1. p. NoTARisii Kiitz. Tab. Phyc. ii. (1850-2), 7, t. 27, f. i. 

 Crass, fil. 21-29 /x ; crass, trich. 12-5-17 /x. This interesting and 

 characteristic alga appears to have been abundant in the district 

 traversed by Welwitsch. 



Pungo Andongo. Alga terrestris, latissimas palustrinm ter- 

 rarum plagas colore sanguineo-ferruginea tingens, prope Coudo ; 

 March 1857. No. 9. Alga ca)spitosa, glauco-viridis. inter muscornm 

 cfespites in cryptis editiorum mentis de Cazella crescens; Dec. 1856. 

 No. 107. (These specimens appear to be old ones ; many of the 

 sheaths were empty, and the red colour has mostly disappeared.) 



Huilla. In udis sylvaticis et apricis inter Monino et lac. Ivantala 

 latas plagas colore fusco-sanguineo obducens ; April 1860. No. 14. 

 Frequens in pascuis pluv. autumnalibus frequentius inundatis, 

 demum exsiccantibus ad lacum magnum de Ivantala ; March 1860. 

 Nos. 17 and 18. "Alga terrestris, atro- vel fusco-sanguinea, late 

 ca3spitosa, crespitibus lubricis planis e filamentis teuacibus dense 

 intertextis et contextis versus marginem radiantibus conflatis, 

 terrfie arete adhserentibus ; loca ubi frequens obvia hroc Alga, 

 tempore spongiosa facta et multis aliis stirpibus uliginosis solum 

 prrebent." 



This alga occasionally branches ; these branches are few and of 

 the same thickness as the primary filament, around which they are 

 closely coiled. The sheath of the branch becomes solid at the base, 

 and is gradually merged into that of the primary filament, its 

 trichome not being in any way connected with that of the latter. 

 The branching really consists of a close application of the base of 

 one filament to the sheath of another, the fusion being for such a 

 short distance that the two trichomes are never present within the 

 same sheath. It is of the same type as the branching present in 

 Schizothrix, but the division into branches is deeper, causing the 

 branches and primary filaments to be of equal thickness. The 

 above remarks prove that Purphyrosiphon is very closely related to 

 SchizothrLv, and emphasizes its position in the Vaginariecc. This 

 affinity was inferred by Gomont (in Ann. Sci. Nat. 7e serie, Bot. 

 XV. 332) from a consideration of an entirely different character, viz. 

 the structure of the sheath. Gomont's statement on the same page 

 that Porphyrosiphon has no branches is, however, erroneous. 



3. Polychlamydum, gen. n. 



PlantfB aquatic.T ; fila subbrevia (1-6 mm.), flexuosissima et 

 sine ramis ; vaginns amplissimre, e serie tubarum concentricj\rum 

 constituta', tubis interioribus firmissimis et coloratis, iis extcriorilnis 

 hyalinis et plus minusve difiluentibus, apice aperto et levitcr expanso ; 

 trichomata solitaria, nonnunquam 2 vel rarius 3 intra vaginam ; 

 apice trichomatis recto ; cellula} brcvissima\ 



1. P. insigne, sp. unica (tab. 365, figs. 1, 2). P. tubis firmis 

 interioribus vaginarum flavo-brunneis 5-7, subarctis, rcgularibus et 

 in ambitu glabris ; tubis extcrioribus 2-4, hyalinis, achrois, doli- 

 catissimis, iis extimis difiluentibus cum marginibus multo irregu- 



