DR. M. T. MASTERS ON SOUXn-AFRICAN JiESTIACE^, 211 



plants, it does not follow that they really belong to the same 



species. 



In addition to the positive characters I have mentioned, the 

 student of these plants gets a kind of intuitive perception that 

 such and such are sexual forms of the same species, though it is 

 not always possible to make the fact obvious to others by de- 

 scription. But little stress, however, can be laid on this matter, 

 as the intuitive perceptions of various observers may happen to 

 differ widely one from the other ! 



It will be seen that I have considerably reduced the number of 

 genera that have been proposed, and that I have in many cases 

 revised and reconstructed the generic characters. Such a course 

 was absolutely essential from the very great confusion in which 

 the matter was involved, a confusion arising from imperfect 

 material and opportunity, in great degree, but still more from 

 the fact that observers have necessarily worked independently 

 of each other. By some authors, as by Endlicher, genera have 

 been combined or separated, as it would seem, solely from a 

 comparison of the descriptions (often very imperfect) given by 

 other botanists, and without reference to the actual specimens. 

 It can easily be understood how mischievous such a practice 

 may be. The numerous type specimens in Dr. Bonder's herbarium 

 have been of the greatest service to me in this particular ; and if I 

 have succeeded in more perfectly circumscribing the various genera, 

 it must be attributed mainly to the examination of the speci- 

 mens so liberally pli^ccd at my disposal by Dr. Sonder. The 

 principal points on which I have relied in the circumscription of 

 the genera have been the inflorescence of the male and female 

 plants, its nature, whether alike or diflerent in the two sexes, the 

 number of spikelets, &c. The number of the flowers in the female 

 spikes is also a matter of great importance, as in those cases where 

 there is only a single flower the whole structure of the flower is 

 modified in accordance with the power that the flower now has 

 of growing equally on all sides, and of not being compelled to ac- 

 commodate itself to pressure from the bracts or other florets. 

 The nature of the fruit, whether dehiscent or indehiscent, and the 

 number of loculi also furnish valuable generic characters. The 

 persistence or otherwise of the sheaths or vaginae is also a good 

 mark of distinction. The stamens are so uniform in their charac- 

 ters throughout the whole of the African species of the order, 

 that it is but rarely I have thought it necessary to describe them 

 fully even in the description of species, 



v2 



