516 Mil. G. BE5THAM ON CLASSIFICATION AND 



xanthum (PL IX. fig. 2) and Phalaris, or at the end of the spikelet,, 

 as in Melica. These have also been often termed neutral flowers, 

 although they hare no pretensions to be flowers at all. In some 

 genera, as in TIniola (Plate IX. fig. 4), from three to six of the 

 lower glumes are empty, and precisely similar to each other, only 

 differing from the flowering ones by being rather naiTower and 

 closer together ; and yet we are only allowed to call the two lowest 

 ones glumes, the others are termed flowers. We are not even 

 allowed to define glumes as the two lowest scales of the spikelet ; 

 for that of Leersia (PI. IX. fig. 1), which has two glumes, one 

 empty, the other flowering, is described as having no glumes but 

 two flowers. In KyUinga and Courtoisia, in Cyperacese, where 

 the fruit is similarly enclosed in two glumes, they are correctly 

 described as such, one empty, the other a flowering one. 



De Candolle and others have described the small outer glume 

 of Pcmicum as an additional bract ; and this may be so far correct, 

 inasmuch as all the glumes are bracts in the most extended sense 

 of the word ; but, as above mentioned^ it is most convenient iu 

 the Glumales to limit the special use of the term bract to those 

 which subtend the whole spikelet or the primary branches of the 

 inflorescence. These, though very general, are not universal in 

 Cyperacere, and very rare in Graminea). I am not aware of their 

 presence in any British genus except Sesleria. 



We now come to the contents of the flowering glume, and es- 

 pecially to the palea, commonly called the upper palea, which I 

 need scarcely repeat is neither homologous nor similar to the so- 

 called lower palea, or flowering glume. It is inserted on the axis 

 of the flower, and not on that of the spikelet, as may be readily 

 seen in any of our Grasses, and perhaps most distinctly in 

 the cultivated wheat, where it is perceptibly raised above the base 

 of the axis. It is differently shaped; and having instead of one 

 central rib or keel two prominent nerves, it is generally supposed 

 to be a double organ composed of the union of two scales. That 

 these two scales are the homologues of the two bracteoles of Hy- 

 polytrum and its allies is very probable ; indeed they most closely 

 resemble, in position as in structure, the united bracteoles of Hy- 

 polytrum pun gens and of Platylepis. At the same time this 

 homology may not be absolutely proved ; and as their presence 

 and general structure is almost as universal as peculiar in Gra- 

 ininea?, it is convenient to designate them by a special name, for 

 which the generally received term palea is not inappropriate, and 

 commits one to no special theory in regard to it. 



