114 



as the species. It belongs to the section of the genus which has 

 the male fiower-buds blunt, as the annexed figure shows, whereas 

 in //. brasilien.si^ tliev are acute. The female flowers of H. confusa 

 are further a little smaller and tlie male flowers considerably smaller 

 than in H. &ra.s«7i>M.s-t.s.* But a still more striking difference is 

 in the pose of the male flowers. 



On the left a female flower of H. confusa with bu'ls of two male flowers 

 (a third has been broken off). On the right a female flower of H. brasiliensis 

 and three male buds. Note the smaller flowers of the first and that the male 

 buds bend earthwards. 



On the left a male flower of H. confusa in section and on the right a 

 large one of H. brasiliensis which is very variable in regrad to size of its male 

 flowers. Note the blunt perianth lobes of the first. 



The panicle in H. hrasiliensis carries up to 300 flowers of 

 which, always if well developed, the terminal flower is female; and 

 file better developed the more female flowers are there, terminating 

 the stronger lower branches, up to about 7 in numljer. Thus a 

 panicle that is weak may be wholly male, and the stronger and larger 

 it is, the more in number are the female flowers on it. All these 

 female flowers take their position as regards the earth from the axis 

 tliat they terminate, and that position is generally in some measure 

 such that they are directed upwards or obliquely upwards : but the 

 divergence of this angle from the vertical is determined by the 

 angle at which the branch takes off from its parent axis and again 

 this by the angle at which the parent axis stands. The panicles 

 produced by tree Xo. 1844 — H. confusa — are narrower than those 

 of H. hro-nlirnsis, as much because the angle at which the side 

 axes take off is smaller, as because, at least in tree 1844, they are 

 of lesser size. The weakest panicles are Avholly male as in H. 

 hrasiliensis. and the stronger carry more and more female flowers 

 upon the lower side-branches up to 5 or 6. 



* J. Hnber, Novae contribuicoes para o generc Hevea in Boletim do Museu 

 Goehli, vii, 1910, pp. 200-216, has discussed at some length the size of the 

 flower, as a character b}- which species and groups of species may be distinguish- 

 ed f om each other. 



