I56 EVANS 



vations are developed show but a single pair of pericheetial 

 bracts. In other words there is an abrupt transition between 

 the bracts with their explanate lobules and the normal leaves 

 just below them with their well-developed water-sacs. H. 

 intermedia offers a certain exception to this rule, the leaf below 

 the innovation being distinctly intermediate between a normal 

 leaf and a bract (PI. VIII, figs. 1, 2). In this leaf the lobe is 

 larger than on ordinary leaves and also less convex, while the 

 lobule is acutely pointed and almost plane. The underleaves 

 also show a gradual transition toward the bracteole, but this is 

 a much more usual condition. 



The new species is also of interest because in some respects 

 it is intermediate between the genus to which it has been referred 

 and Drcpanolcjeunea. The most important differences between 

 these 2 genera are found in the underleaves. In typical 

 species of Harpalejeunea these are divided by a shallow sinus 

 into 2 broad and divergent divisions, rounded at the apex and 

 usually 3 or 4 cells wide. The radicelliferous region is 

 commonly indistinct. In Drepanolejeunea the divisions of the 

 underleaves are setaceous and usually widely spreading ; in 

 most cases they consist of from 2 to 5 elongated cells in a 

 single row, but they may be 2 cells wide at the base. These 

 divisions arise from a basal portion in which the radicelliferous 

 region is bounded by a distinct margin of larger cells. In H. 

 intermedia the underleaves show a basal portion with a fairly 

 distinct border (PL VIII, fig. 7), and the divisions vary at the 

 apex from rounded and 2 cells wide, to pointed and tipped with 

 2 superimposed cells (PI. VIII, figs. 8, 9). They therefore 

 combine the underleaf-characters of the 2 genera. In some 

 respects these underleaves bear a resemblance to those of H. 

 pseudoneura Evans, 1 of the Hawaiian Islands, which is also a 

 somewhat aberrant member of the genus. 



H. intermedia is apparently the first species of Harpalejeunea 

 which has been recorded from Asia, and it has no very close 

 allies among species known from other parts of the earth. H. 

 pseudoncura, with which its underleaves have just been com- 

 pared, is at once distinguished by the continuous row of ocelli 



■Trans. Conn. Acad. 10 : 427.^/. 50, f. i-g. 1900. 



