276 



THE JOURNAL OF BOTANY 



distant branches only. Leaves rather closely set, erect- spreadimj, 

 only very slightly falcate-secund at the tips of the branches, narrow- 

 lanceolate, gradually tapering to a lon(i,jine, sharply -toothed siibiUa. 

 Nerve narrow, scarcely reaching half-way up the leaf. Cells long, 

 narrow, thick-waJled, the basal somewhat incrassate, slightly porose. 



Hab. In a bog near Craig-lyn-Dyfi, Llan-y-mawddwy, Meri- 

 oneth, North Wales, alt. 2000 ft., Sept. 1898, ^liss M. Roberts. 



Allied to var. setiforme Eeu., differing in the shorter leaves, the 

 colour of the tufts, mixed with a purplish red, and especially by the 

 cells with the walls incrassate, the basal ones slightly porose. This 

 character brings it near to the j'alcatuin group, and especially to var. 

 anylicuDi f. Holtii Sanio {Auihlysteiiiuin jiiiitans var. *j Holtii Sauio, 

 Braithw. Brit. Moss Fl. iii. 51), from which it differs in the 

 colouring, glossy leaves, less-branched stems, etc. The line of 

 demarcation between the groups amphibiiun and j'alcatum is at 

 times not very clearly defined, and there are certain forms for 

 which it is not easy to determine the most satisfactory position, 

 the present variety being one of these. It is a very pretty plant, 

 the glossy, variegated colouring being unusual. 



H. FLUiTANS L. (am/iJiibinm) var. squalidum Ken. & Dixon, 

 var. nov. In dense intricate masses, pale dull t/reen above, uf a dirty 

 reddish brown below. Stems little-branched, rather robust, the leaves 

 somewhat complanate-spreading, falcate in upper part of stem and 

 branches, rather large, lanceolate, somewhat abruptly terminatiny in 

 a fine, faintly -toothed, almost pili/onn subula. Other characters as 

 in var. Jeanbernati Ren. 



Hab. In stagnant water, Dawley, Shropshire, May 11, 1896; 

 Rev. W. H. Painter. 



l^is variety is nearly allied to var. Jeanbernati Ren., from which 

 it differs in the fine subula of the leaves, as well as in the habit. 

 The stem-leaves are widely complanate-spreading, giving the ap- 

 pearance of some aquatic forms of H. riparium L. — it was indeed 

 at first sent me as H. riparium var. splendens De Not. 



A third variety has perhaps a still greater value than the two 

 described above, seeing that its ascertained distribution is much 

 wider. A plant collected somewhat extensively by Mr. J. A. 

 Wheldon and Mr. A. Wilson in Lancashire, on elevated moorlands, 

 and again by Mr. W. Ingham in Yorkshire, has for some time 

 given rise to discussion. Its short nerve, the scarcely secund, wide 

 leaves, shortly and broadly pointed, the loose areolation and very 

 indistinctly defined auricles, gave it a very different character from 

 that usually obtaining in this species. Under his somewhat hetero- 

 geneous group ^' a obsoletum,'' Sanio has described a var. Holleri 

 having very nearly the same characters, and M. Renauld at first 

 referred our plants to this, ranking them under var. Jeanbernati as 

 f. Holleri Sanio. In a letter recently received from him, however, 

 he writes: " Cette determination est, a la rigueur, acceptable; 

 cependant comaie la var. Holleri Sanio est mal con9ue et mal 

 decrite, et que d'ailleurs le groupe obsoletum Sanio est tres confus je 

 prefere aujourd'hui fonder une variete nouvelle qui doit etre placee 

 a cote de la var. Jeanbernati Hen.'' For this variety M. Renauld 



