Glum 



<;i; win \< I .;. 





4 



...t 



the joints, <xn ered with a coat of silex, sometii 

 alternate, with a split sheath, and a membranous expansion (li| 

 ind blade. Flo were green in littli died locus tie, arret 



or panicled manner. Flowers usually J, sometimes monoecious or ] 

 ing oi imbricated bracts, "t' which the most exl called glumes, th< 



diately enclosing the Btamena palese, and the innermost at tliebase of n 

 < iliuiKs usual!) •_', alternate ; somi times single, most commonly unequal. . 

 nate ; the low« r or exterior simple, the upper or interior compi 



_ i ■< .11-, margins, and usually with 2 keels, together forming a kind 

 sometimes wanting ; if 2, collateral, alternate with tl 

 the lower of them ; either distinct or united. Stamens hypi 



. 1 of which alternates with the 2 hypogynous scales, and is thei -,t the 



lower palese ; anthers versatile. Ovary simple; styles 2 or 3, vet 

 into one ; Btigmas feather] or hairy; ovule ascending by a broad ba 

 Pericarp usually [indistinguishable from the Beed, membranous. Albumen foi 

 embryo lying on one side of the albumen at the base, lenticular, with a I road cot 

 and a developed plumula ; and occasionally, but very rarely, with a Becond cot^ 



<m the outside of the plumula, and alter- 

 nate with the usual cotyledon. 



This iiinst important Order off! 



great singularities in it~ organisation, 



although it is one in which, formerly, 



botanists the least suspected anomalies 



to exist. They found calyx and corolla 



and nectarii - here with the same facility 



as they found them in a Ranunculus ; ! 



and yet Buch organs exist in no one 



genus of Grasses. Their so-called 



flowers consist of green Bcales, not 



placed in whorls, but arranged one 



above the other, and are undoubtedly 



constructed of bracts alone. Not a trace 



is discoverable among them of calyx ur 



corolla, properly bo called, unless certain scales usually pn 



the ovary, are to be so considered. Brown's account of then 

 Btruction i- Mill the best that has been published. He says, — 



•• The natural or most common structure ol Gramineae i* to have thi ir - \e...l i 

 Burrounded by the floral env< lop« a, each of which usually consist of two distil 

 but both "t these envelopes are, in many genera of the order, subject tovarii 

 of imperfection or even suppression of their parts. The outer envi 

 Jussieu, in most cases containing several flowers with distinct and often disl 

 tions on a common receptacle, can only be considered as ai he bract 



in volucrum of other plants. The tendency to suppression in thi* envel 

 ).. great r in the i xterior "i low< r valve ; so thai a .nun consisting 

 in all cases, be considi red as deprived of its outer or inferior valve, bn 

 with a simple spike, as Lolium and Lepturus, this is clearly proved by th< stru 

 the terminal flower or spicula, which retains the natural number of ; 

 genera not admitting of this direct proof, the fact i> established bj 

 showing it< gradual obliteration, as in those 8] Panicum which c 



genus with Paspalum. On the other hand, in the inner envelo] 

 obliteration first takes place in the inner or upper valve ; but ; 

 of one central nerve, two nerves equidistant from ii- axis, I consider it 



two fluent valves, analogous to what takes place in th 



irregular flowers of other class* a ; and this conflneno 

 towards its obliteration, which is complete in many sj 

 Pappophorum, Alopecurus, Trichodium, ral other " 



nature of this inner or proper envelope of Grasses, il i 

 its structure now given, in reducing its parts to the usual I 

 ledons, affords an additional argument for considering it a~ tl 

 argument, however, is not conclusive, for a similar conflui • 

 two inner lateral bracfe BB ol the gr< at- t p .• I of hides 



tion of its valves, the proper envelo] 

 accord than with a genuine perianthium. It", th< n fore, th 



i lxxiv. 



I.WV. 



LXXIV.— 1. S 

 Pig. LXXV.-S 



