TO DR. R. WIGHT. 



xli 



popular, it is certainly a handsome and valuable timber. We have 

 supplied several first rate ones to the Arsenal, and this shews that 

 in Bengal it will succeed for practical purposes, always supposing the 

 north west to be out of the way. I have ordered you a set of tree 

 and shrub seeds, which I hope will please you. Our handsomest 

 trees are the Acacia serissa, A. elata, Conocarpus acuminatus, Ter- 

 minaliae several, Nageia Putranjiva, Dalbergiae two or three Pterocar- 

 pus dalbergioides, Cassia sumatrana : this last is admirable for groups, 

 Schleichera trijiiga, Parkia Brunonis, elegant etc. Now I am ac- 

 quainted with most, I know how to group them in plantations to 

 produce great effect. There will be plenty of seed of Poinciana 

 regia, it is a beautiful tree. Your name is down for that also. I will 

 send you for experiment before long, cuttings of some very pret- 

 ty things. Quoad Bentinckia. I think I told you I had a pigmy 

 one from Malacca, this I found out from Martius. 



I have no opinion of long generic descriptions, what useless re- 

 petition ! what absurd sectional characters you often meet with 

 in books, it would require a logician to see the differences. Now 

 I say all differential characters should be evident, if drawn from 

 the roots no matter so that they carry conviction. Science is not 

 wanted, but practical characters ; this is one reason why I loathe 

 the subdivisions going on. Who did not know a Convolvulus 

 formerly ? ; it requires now a Botanist to know one. Besides 

 consistency requires that all groups of the same value should be as 

 equally separated as possible, subdivisions do not allow of this, but 

 on the contrary, give us some genera so far as we know, quite dis- 

 tinct, others quite indistinct. Therefore I hope you will set your 

 face against this practice when you come to Convolvulaceee. I will 

 remember your hint about the dates of genera, and if Wallichia can 

 be restored, I will do it on principle. DCs. system I always thought 

 poorly of as a system. He did a great deal doubtless by it, and his 

 Vegetable Organographie, Theorie Elementaire de la Botanique, and 

 Essai sur les Prop. Medicales ; the fact is. Botanists are far behind. 

 Depend upon it, that our three sub-kingdoms, the three divisions of 

 Compositse, the three of Leguminosae are not indicated to us by nature 

 for nothing ; yet all divide and sub-divide, make tribes, and sub- 

 tribes, without suspecting that nature may have a numerical system. 

 Nothing great will be done until number, the study of deve- 



/ 



