NO. 2 MOSSES OF AFRICAN EXPEDITION — DIXON 9 



minor point regarding the pores. These are remarkable among the 

 section Cuspidata, being very numerous on both surfaces, especially 

 the dorsal, where they are arranged in chain form along the com- 

 missures as in many of the Subsecunda, but are much larger; on the 

 ventral surface they are also numerous, but somewhat less so, and 

 are smaller, while here and there occur very minute pores in the 

 angle of the cell or on its face. Warnstorf describes the pores of 

 the ventral surface as " beringt," those of the dorsal surface as 

 " schwach beringten," while in the present plant the dorsal ones are 

 decidedly more strongly ringed than those on the ventral surface. 

 In every other particular the plant agrees exactly with the var. graci- 

 lescens as described. 



ANDREAEACEAE 



ANDREAEA KILIMANDSCHARICA Par. 

 Andreaea striata C. M., not A. striata Mitt. 



Loc. 3,630 meters, Nos. 1584, 1588. No. 1588 is in fruit, and the 

 few capsules show a rather remarkable peculiarity. It will be recol- 

 lected that Hooker f. and Wilson divided Andreaea into two sub- 

 genera, separating A. Wilson! (as Acroschisma) from all the re- 

 maining species (Eu- Andreaea) , on the ground of the capsule, which, 

 instead of splitting to the base or nearly so into four to six valves, 

 splits only about one-fourth the length, the lower part of the capsule 

 remaining entire, shortly cylindric. Wilson placed also in Acro- 

 schisma another species, A. dcnsi folia Mitt., from the Himalayas, 

 but Brotherus includes it in Eu- Andreaea. According to Roth, 1 Mit- 

 ten's specimens show a capsule narrowly elliptic to almost cylindric 

 and split into valves from the middle or from two-thirds its length 

 upwards, so that it remains a question as to which subgenus should 

 claim it. The Mt. Kenia plant (No. 1588) shows a few capsules 

 quite normal, one or two entire to about the middle, and one or two 

 entire to about two-thirds the length, only the upper one-fourth or 

 one-third split into valves (cf. pi. 1, fig. 5, a, a'). The species, it can 

 hardly be doubted, is a Eu-Andreaea, and the peculiarity of the 

 capsule form an abnormality ; it is possible that this may be the case 

 with A. densi folia Mitt. The three species, it may be remarked, are 

 not in other respects nearly related to one another. 



1 Die aussereuropaischen Laubmoose, Band 1, p. 21. 



