[From the Journal of the Bombay Natural Hist. Soc, August I, 1927-j 



REVISION OF 

 THE FLORA OF THE BOMBAY PRESIDENCY 



BY 



E. Blatter, s.j., Ph.D., f.l.s. 



Part III 

 ( Continued from page 917 of Vol. xxxi) 



GRAMINE^S (Cke. ii, 907) 



BY 



E. Blatter and C. McCann 



We have decided to take up the Graminece for several reasons. Since 

 Cooke's publication of this family several new species have been described and 

 McCann has added over 50 species which are new to the Presidency. More 

 important than the numerical additions are the changes made during the last 

 fifteen years with regard to the general arrangement of the grasses and the 

 definition of certain genera. A great amount of work has been done in Europe 

 as well as in America. In Europe it was chiefly O. Stapf who, with his vast 

 experience of the grass-flora of many countries and the rich material of the 

 Kew Herbarium at his disposal, has advanced our knowledge of the grasses 

 and their systematic co-ordination more than anybody else. A comparison of 

 his monograph of the grasses in the Flora of Tropical Africa which is still 

 in progress with that of the Flora Capensis shows at once in which direction 

 and to what extent progress has been made. As regards the second sub- 

 family Pooidece the two works reveal only minor changes ; but considerable 

 changes were found necessary in the first subfamily Panicoidecz. Here 

 again it is chiefly the genera Andropogon and Panicum and their allies which 

 have been affected. In this respect, especially where the tribe of Panicec? 

 is concerned, we owe a good deal to American botanists. It was chiefly 

 the fact that there were no definite dividing lines for the genera of Andpro- 

 gonecs and Panicece, that induced Stapf not 'to unite the groups wherever 

 intermediate links can be detected,' which would bring about endless 

 confusion, but ' to be satisfied with approximately definable groups, which can 

 on the whole be easily grasped and remembered.' In other words, it is 

 preferable from a practical point of view to adopt smaller genera than 

 unmanageable large genera for merely theoretical reasons. As the Kew 

 Herbarium, as far as the grasses are concerned, owes its systematic arrange- 

 ment entirely to Stapf, and as colonial workers will always appeal to Kew in 

 their difficulties, we thought it advisable to follow Stapf in the definition and! 

 sequence of the tribes as laid down in the Flora of Tropical Africa. As of 

 late the grass-problems have received renewed attention by the Agricultural 

 Departments in India, we do not consider it superfluous to bring the systematic 

 account of the grasses of Bombay up-to-date. 



An asterisk in front of a name means that the particular genus or species has 

 been introduced. 



Two complete keys, one natural and one artificial will be given at the end. 



SUBFAMILY I. PANICOIDE.^E 



The mature spikelets fall entire from their pedicels or with them, all are 

 alike or differ in sex and structure. Perfect spikelets with 2 heteromorphous 

 florets, the upper hermaphrodite, the lower male or barren. Rhachilla not 

 continued beyond the upper floret. 



TRIBE I. Mayidecs. — Sexes borne on different inflorescences on the same 

 plant or the female spikelets at the base of the inflorescence, and the male 

 above them. The male spikelets in pairs, one sessile, the other pedicelled, or 

 both pedicelled, in spike-like solitary or panicled racemes, 2-flowered. Glumes 



[i] 



