— 3— 



cal detail. Warnstorf's article is in spite of the oversight by no 

 means without interest and value, and the close relationship ot 5. por- 

 toricense and 5. imbricatum is indubitable, as had in fact been lecog- 

 nized before. 



Both species are best separable from others of the group by the 

 fact that their branch leaves in transverse section show the chloro- 

 phyll cells about equilaterally triangular with base ot triangle exposed 

 on inner surface of leaf and that the inner walls of their hyaline cells 

 where in contact with these are normally beset by so-called fringe- 

 fibrils, a series of close, nearly parallel, somewhat irregularly run- 

 ning ridges. 



S. portoricense is distinguished from its parent species (for this is 

 certainly the relation between the two) by its usually gi eater size, the 

 more distinct hyaline border of its stem and branch lea\es, by its 

 strong horizontally spreading branches whose leaves increase gradually 

 from the base, giving the branches a clavate effect, by the important 

 difference in cortical cells of branches, and finally by the fact that its 

 lower branch leaves are veiy shcrt. in some cases wider than long, 

 and all are more or less distincdy cordate at base, particularly, how- 

 ever, the lower ones. 



The leaf border in Inophloea, best represented in the stem and 

 lower branch leaves of this species, is something entirely different 

 from the border of Litophloea in that it is a border of resorption. 

 If examined closely in section and from both sides of the surface, it will 

 be found to consist of a single plate of hyaline membrane with a fine 

 mesh of darker lines, suggesting the remains of obsolete chlorophyll 

 cells, though the mesh is much finer and must play a part in main- 

 taining the rigidity of this rather broad but thin margin. This mar- 

 ginal membrane is a continuation of the inner leaf surface (the outer 

 being here completely resorbed away ), its outer edge is very irregular 

 in oudine, as is well illustrated in Warnstorf's figures in the article 

 alluded to. 



As to the cortical cells of branches, it may be noted generally 

 that all species of Sphagnum show at base of branches a cortex re- 

 sembling that of the stem, except that its cells do not constitute 

 so many layers ; usually the differentiated branch cortex shows its 

 most characteristic development at about the middle of the branch, 

 where the branch leaves are also best developed. In the case of 

 S. portoricense the strong branches show a cortex most distinct well 

 toward or quite at the apex, where the branch leaves are also largest 

 and most characteristic, i. e. most differentiated from the stem leaves. 

 The first noticeable peculiarly of these cells in 5. portoricense, apart 

 from their much greater size, is the fact that their outer walls do not 

 show the pores usually present in other species of Inophloea, a fact 

 noted already by Sullivant ; secondly, the dividing wall between each 



