T. Holm — Studies in the Cyperacece. 305 



(P 2 and P 3 ) show the same structure as the larger ones, but 

 they are more open and only tubular at their very base. 



This structure of the clado-prophyllon observed in Cyjperus 

 is, also, to be found in that of other genera, where the function 

 is identical, but not in Duliehium or in certain Oarices where 

 these fore-leaves are enclosed by the sheaths of the subtending 

 bracts, and where the direction of the rhachis or peduncle 

 remain s unchanged . 



It is now interesting to notice that the involucral leayes and 

 the clado-prophylla, characteristic of Cyperus, etc., may be 

 traced in the inflorescence of certain Graminece. — Several years 

 ago, wheu we observed this peculiar structure and function of 

 the prophyllon in Cyperus, we naturally felt induced to look 

 for this same organ in the inflorescence of other families, 

 wherein the branches are able to perform the same move- 

 ment from vertical to horizontal in accordance with the devel- 

 opment of the flowers. Our attention became especially 

 directed to the Graminece, and to such genera in which the 

 inflorescence, the panicle, is ample and rich-flowered. We had 

 already studied the vegetative propagation of this order, the 

 Graminece, as represented in this country, and we had observed 

 several analogies to exist between these and the Gyjperacece, 

 for instance, the invariable presence of a fore-leaf at the base 

 of each lateral, vegetative branch. We had, also, learned that 

 these fore-leaves are "always" open in the Graminece, but 

 closed, more or less tubular in the Cyperaeece, a point which 

 explains their identity with leaf-sheaths. Among the GraminecB 

 these fore-leaves (clado-prophylla) are very large and quite 

 numerous in such genera of which the shoots are ramified, and 

 they are readily visible, for instance in Coix,Eleusine, Panicum, 

 Munroa, and many others. But if we wish now to demon- 

 strate their presence in the inflorescence also, we are obliged to 

 admit that we have, so far, not succeeded in finding them 

 developed as "free" leaves in any of the numerous species 

 which we have examined for this special purpose. The large 

 and open panicle of several species very often exhibits the pres- 

 ence of rudimentary bracts subtending the lateral branches, 

 but these branches themselves do not possess an independent 

 leaf at their base, which might correspond with the tubular 

 prophyllon (P 1 ) of Cyperus', all that is to be seen is a small 

 cushion-like body, which is located at the base of the secondary 

 branches, and on that side of the branch which turns away from 

 the mother-axis. It is, nevertheless, the peculiar structure of this 

 little body which enables the lateral branches to spread out at 

 a certain stage of the flowering period, and considering the fact 

 that its position answers exactly that of a prophyllon, besides 

 that the bract, which subtends the branch, is, also, present as 



