140 GORDONIA. 



Freziera, S. America. 

 Vismea, Canary islands. 

 Tremanthera, New Guinea. 



And from such a review of the genera, allied to Gordonia, we 

 turn with at least the suggestion that discontinuous distribution 

 is not so rare in the order as to afford any strong reason for 

 Korthals' and Pitard's endeavour to cut the Asiatic Gordonias from 

 the American. 



In fact what we have in the Theeae and Ternstroemieae is 

 such a noteworthy demand for damp air as to restrict them either 

 to damp seaboards or to mountain forests or to rain forests that 

 are not too hot. With a centre about the Pacific, local climates 

 have ruled where they should persist, and the discontinuous dis- 

 tribution is because all but a small part of North America is too 

 dry. 



instead of splitting gordonia, it apparently should be 



enlarged by the inclusion of at least the 



Asiatic species of Haemocharis. 



Xot only do I fail to follow the bias, but I find characters 

 contrary to Pitard's statement that Nabiasodendron has blunt cap- 

 sules and Gordonia pointed, and instead of cutting down Gordonia, 

 I believe that Haemocharis, at least in the East is not distinguish- 

 able from it; and I recommend that the two be examined with a 

 view to union. The only difference is in the presence of a style in 

 Gordonia and its absence in Haemocharis, a difference which puts 

 Gordonia singaporiana for instance on either side of the line accord- 



Fig 2. Ovary of 67. singaporiana at the time of the fall of the 

 corolla showing how undefined in the style. 



ing to the view of the moment, and prompted the remark of 

 Koorders and Valeton in regard to Haemocharis integerrima (Me- 

 dedeelingen 'S Lands Plantentuin, No. 16, 1896, p. 294) " stylis 

 fere dimidio leviter connatis apice obtuse diver gentibus, haec species 

 demarcationes inter genera Gordonia. et Laplacea (Haemocharis) 

 fere plane aufert," i.e. it comes within a trifle of breaking down the 

 boundary. 



There is one point to be set out before proceeding, namely that 

 the Asiatic species of Haemocharis owe their position in that genus 



Jour. Straits Branch 



