358 THE JOURNAL OF BOTANY 



Eoxburgh collected in the Eajmahal Hills, which are in this 

 country, and there is no reason to suppose that his contributions 

 to the Calcutta Garden consisted of seeds only. G. helicterifolia 

 Wall, practically replaces G. hirsuta Vahl to the north of the 

 Vindhya chain of mountains, and extends over the Indo-Gangetic 

 tract as far as the Panjab Salt Eange (Indus basin). Wight and 

 Arnott had no doubt observed that the only authentic types of 

 " G. inlosa" at Kew (there are none apparently in any other 

 collection) had been grown at Sibpur ; but it does not follow that 

 this was the only material to which Eoxburgh had access. It 

 would be premature at this stage to discuss the validity of species 

 in this difficult group; but assuming that " helicterifolia" Wallich, 

 as understood by Eoyle, which is easily recognizable in the field 

 and the herbarium alike, is valid, then there can be little question 

 that Eoxburgh's No. 12, as regards the locality and parts of the 

 description, corresponds to G. helicterifolia. 



No. 13, G. polygama, remains ; and it must be admitted that 

 in this case too much weight was placed on the polygamous in- 

 florescence. All the species of the G. hirsuta group, which has 

 meantime been enriched from the Indo-Chinese region, tend 

 probably to be polygamo-dioecious. On the other hand, the 

 characters which mark off G. 2)olygama from its allies, though not 

 very easy to describe, are manifest and constant ; they will be 

 indicated in a key to certain of the Indian species with which 

 these notes will be concluded. The type of polygama is in Smith's 

 herbarium (a male shoot only) and there are duplicates at Kew. 



To sum up as regards this group, the Eoxburghian species 

 emerge thus : — 



No. 9. HIRSUTA Vahl, Teling. Juvellikee. 

 Hab. Southern and Central India. 

 = G. hirsuta Vahl. 



11. CARPiNiPOLiA Juss. Teling. Nullee. 



= G. flavescens Juss. 

 Hab. Arid subtropical belt of Asia, Africa, South and Central 

 India to Senegambia. 



12. piLOSA E. (pro parte). 



Hab. Northern India from the eastern spurs of the Vindhya 

 chain to the Sewaliks northwards, and west to the Indus basin. 

 = G. helicterifolia Wall. 



13. POLYGAMA E. 



;= G. polygama Eoxb. 

 Hab. Eastern Bengal ; South India ('?). 



The following recapitulation will assist the reader in forming 

 conclusions on the intricate questions handled above. 



The numbers cited from the Flora Inclica are those actually 

 printed in Carey's edition (1832), and the specific name is that 

 assigned l)y Eoxburgh in the same work, but the arrangement of 

 the species follows that adopted in Vrsiins Bengal Plants, i. 281-4. 



(1) No. 11, Fl. Ind. ii. 587, " carinnifolia Juss." is properly 



