SPHAGNUM. 7 



Var. /3. confertum Lindb. More compact, with shorter and 

 denser branches. Branch leaves rounded, obtuse, very concave. 



Var. y. stenophyllum Lindb. Branch leaves narrower, ovate- 

 oblong, less concave and almost entire above. 



Hab. Peat bogs and ditches. Rare. The var. $. in drier places. 



I have no hesitation in sinking S. papillosum to a sub-species of S. cymbifolium, 

 with Cardot and other authors. The papillae on the cell walls are often reduced 

 to a minute size, and moreover are entirely absent in plants which otherwise exactly 

 resemble typical 5 1 . papillosum (vars. sublave Warnst. and lave Warnst.). If we 

 consider the latter form as belonging to .£ cymbifolium, we must admit that the 

 points of distinction derived from characters other than the papillae are valueless, and 

 the distinction rests on these alone. And since there are forms (var. sublceve Warnst.) 

 in which the papillae are extremely minute and are even found only in the basal cells 

 of the leaves, while wanting in the middle and upper part of the leaf, it seems 

 clear that this character alone cannot be held to give full specific rank. I have speci- 

 mens of both the above vars. from N. America, which have the distinctive facies of S. 

 papillosum, but with the papilke almost absent in the one and entirely wanting in 

 the other. In the latter case it becomes, of course, a question to which of the two 

 plants it should be referred. The fact that minute papillae have also been recently 

 found in some forms of S. squarrosum and other species, and that .S. medium Limpr. 

 has the walls sometimes smooth and sometimes papillose, considerably minimises the 

 value of these structures as specific characters. 



I do not find the outer layers of the cuticular cells of the stem free from fibres, 

 as Braithwaite describes them, though usually not so strongly fibrose as in S. cymbi- 

 folium. The inner ones are strongly fibrose. 



The chlorophyllose cells are usually more or less emergent on the ventral surface, 

 but sometimes enclosed on both faces. 



2. Sphagnum Austini Sull. (Tab. VI. C.) 



Habit and colour of S. papillosum, but less rigid. Walls of 

 the hyaline cells, adjoining the chlorophyllose, thickly studded 

 with linear papillae, much longer than those of the former plant. 

 Chlorophyllose cells on the ventral surface, more entirely covered 

 by the hyaline on the dorsal face, broadly triangular in section. 



Var. (3. imbricatum Lindb. In dense tufts of a darker brown ; 

 branches crowded, with closely imbricated leaves. 



Hab. Wet heaths and bogs, especailly near the coast ; very rare ; the type, 

 Lyth Moss, Westmorland ; Lewis ; Kirkcudbright ; the var. j8 in almost the same 

 localities. 



There seems somewhat more reason for according this plant the full specific rank, 

 partly on account of the greater distinctness of the papillae, which however tend to 

 be greatly reduced in some forms, and partly because of the difference in the chloro- 

 phyllose cells. The division of the Sphagna into species must of necessity be based 

 on more or less arbitrary grounds. It appears most satisfactory to consider it a 

 principle in their classification, that when two types, in themselves not very strongly 

 distinct, are also united by a fairly well graduated series of intermediate forms, 

 they should be re-united as species and sub-species. This is certainly the case with 

 5. cymbifolium and S. papillosum ; but it is less clearly established in the case of the 

 former species and S. Austini, and I have therefore kept them apart. I should add, 



