282 RETIREMENT, TO 1897 : BOTANICAL WORK 

 May 24, 1893 : 



When you are again up in the hills, it would be well 

 to ascertain exactly, what are the Aconites whose roots are 

 eaten with impunity. . . . The subject of the alkaloids in 

 the roots of Aconites is now being studied with great care, 

 and there are anomalies amongst the Indian ones (as named) 

 which I can only explain by a confusion of nomenclature. 

 The Pharmaceutical Society is publishing separate researches 

 on the subject. 



... It is impossible in the Herbarium to limit the 

 Alpine varieties of several of the species, and to distinguish 

 those of one species from those of another. A careful 

 following of each species from a lower to a higher elevation, 

 would be a great thing to do. 



. . . The species of Iris of India are in a shocking 

 mess, I wasted days over them. Pray observe them well, 

 especially the * orris rooted ' wild and cultivated. 



January 7, 1896 : 



I can quite sympathise with you in your disappointment 

 at not being sent along with the Pauri Commission. ... It 

 may indeed be doubted whether the result would be worth 

 the expenditure ; what is now wanted is, not more collections, 

 but far more knowledge about what plants we have. The 

 number of genera, of which the species want careful examina- 

 tion, is very great ; in the Orchideae, for instance, all the 

 terrestrial species require revision ; Griffith set the example 

 of what could be done by the co-ordination of observing 

 with collecting. . . . The long and the short of it is, the N.W. 

 Flora wants revision throughout, by careful study of each 

 species, its range, variation, and obscure characters. 



February 22, 1895 : 



Have you ever yet taken up the Indian Irises system- 

 atically ? It can only be done by garden specimens and 

 drawings. You have also a fine field in the terrestrial 

 Orchids, which want analysis on living specimens. 



February 16, 1896 : 



Grasses are terribly difficult and full of snares ; but 



