514 MR. G. BENTHAM ON CLASSIFICATION AND 



mtion of the term glume could be given, unless it were applied, as 

 in Cyperaceae, to the whole of the primary imbricate scales attached 

 to the main axis of the spikelet, whether enclosing a hermaphro- 

 dite unisexual or semi-abortive flower or quite empty — and that 

 other names must be given to the secondary scales inserted within 

 the glume under the flower on the secondary or floral axis. I 

 drew up my generic and specific characters accordingly ; and it was 

 only after they were in print that I ascertained that similar views 

 had been independently propounded by Hugo Mohl, Doll and 

 others, in Germany, and more recently by Germain de St. Pierre, 

 in France. The consequent terminology has since been adopted 

 by Dr. Hooker and by Prof. Oliver, but not, I believe, by syste- 

 matic botanists generally. Those who have long devoted them- 

 selves specially to grasses have been so used to consider the 

 flowering glume and its enclosed palea as two paleae, and the lower 

 empty ones alone as glumes, that they find it inconvenient to 

 adopt a new language, and have adduced arguments in support of 

 their practice, which may appear specious, and may require 

 some further observations in reply ; in illustration of which I shall 

 select examples from some of the most familiar genera of Grasses, 

 of which a few diagrams are given in Plate IX. 



Whilst insisting on two great laws in terminology — that homo- 

 logous and generally similar organs should be designated by the 

 same name, and that where the want of homology of two organs has 

 been demonstrated they should be called by different names, I would 

 nevertheless admit, as an exception to the first rule, that icherein 

 a series of homologous organs a certain number of them are regularly 

 modified in consequence of a difference in the functions they are 

 called upon to perform, it may be convenient to describe them under 

 a different name. If, therefore, those glumes which in the spike- 

 let of Gramineae enclose a flower, and are thus called upon 

 partially to replace the perianth, were constantly different from 

 the empty ones, it might have been convenient to give them a 

 distinct name ; but it is easy to show that no such difference is at 

 all regular, and that even where the flowering glume is in a slight 

 degree modified the assimilating it to the palea is most inappro- 

 priate and misleading. 



In several of our large genera of Grasses, the only difference 

 .Between the one or two outer empty glumes and the flowering 

 ones is that they are rather smaller or rather larger, as is so fre- 

 quently the case in Cyperace® and Restiace©, where no difference 



