MORPHOLOGY OF THE FLORAL ORGANS 123 



We may also conclude that the 2 organs inclosing the flower of 

 Homalocenchrus are lemma and palea and that the 2 glumes are 

 absent, since in Oryza, a closely aUied genus, the glumes are present, 

 though small. 



The glumes are sometimes awned, but less frequently 

 so than are the lemmas. They are variously modified and 

 distorted in a few genera, the first glumes of a group of 

 spikelets together forming a sort of involucre around the 

 group (Anthephora), thickened like a bird's head (Lopho- 

 lepis), globose and pitted (Rytilix). The large second 

 glume is covered with hooks in Nazia so that the group of 

 spikelets becomes a bur. In Alopecurus the glumes are 

 connate, that is, grown together along the edges to form 

 a cup (Fig. 39). 



153. Anomalous glumes. — The glumes of some genera 

 of Hordeae show certain anomalies. In Lolium and in a 

 few allied genera the spikelets are sessile on a flattened 

 rachis but stand edgewise to this instead of crosswise as 

 is usual in other genera (Fig. 57). But one glume (the 

 second) is present and this on the outer side of the spike- 

 let. It is longer and larger than the lemmas, sometimes 

 longer than the spikelet, and looks like a subtending bract. 

 In the terminal spikelet of the spike, however, both 

 glumes are developed. In Sitanion and some species of 

 allied genera the glumes are reduced to subulate awns, 

 these forming a sort of involucre to the groups of spike- 

 lets. In certain species of El3nnus (e.g., E. virginicus L.) 

 the glumes of the lateral spikelets stand in pairs in front 

 of the spikelets. 



Hochstetter states that the glumes of Hordeum and many 

 species of Elymus are single but cleft into 2 parts. Schenck thinks 

 that they are sterile spikelets. (For full discussion, see Bot. Jahrb. 

 Engler40:97. 1907.) 



