288 RETIREMENT, TO 1897 : BOTANICAL WORK 



for all new species in the place of my own whenever we agree. 

 The latter [agreement] will I do hope be the rule, but consider- 

 ing the very different materials, in amount of specimens and 

 number of collections upon which we work, there must be 

 discrepancies in details. 



To Mr. Duthie 



Jan. 7, 1896. 



I am not sorry that Hackel has cut me out of Duthiea ; 

 he could and has discoursed on it in a way I could not have. 

 My impression is that the plant is as good an Avenacea as 

 Festucacea — but I am not going to interfere with Hackel's 

 determination. After all Hackel's paper shows that the 

 tribes and sub-tribes of Poaceae are a sad lot. 



He was doubtful, however, as to the natural position near 

 Bromus assigned to Duthiea by Hackel. A new Duthiea turned 

 up in a fresh consignment of Indian grasses, 



which satisfies me (and Stapf) that I was right in placing the 

 genus next Danthonia. When Hackel's paper appeared I, 

 of course, followed him, and in the clavis of genera (printed 

 off months ago) I put it next to Bromus. I shall ' hale it 

 up ' in the body of the work to its right place. It corrobo- 

 rates my view of the close affinity between Bromus and 

 Avenaceae. (Feb. 16, 1896.) 



Amid such complexities of this, ' the hardest work I ever 

 did,' the solution of which depended upon the critical observa- 

 tions on the spot that he so much desiderated, it was hardly 

 to be wondered at, however annoying, that errors crept in. 

 For these he was constantly on the look out, and on February 6, 

 1898 (the Flora was published in 1897 and he was then at work 

 on the Ceylon Grasses for the completion of Trimen's Flora of 

 Ceylon), he writes to Mr. Duthie : 



You may be shocked to hear that my new genus Ney- 

 raudia, over which Stapf and I spent no little time (to make 

 sure !) turns out to be a known not uncommon Trir aphis ! 

 and is not even Arundineous. The fact is that Gramineae 

 are still in a shocking state. Stapf is making discoveries 

 constantly of mistaken affinities amongst well-known genera, 



