History of Agropyrum JDonianum. 263 



matures the awns grow larger, till they become (as Professor 

 Hackel remarks) " very ob\ious even for the naked eye." 

 I am compelled, therefore, to think that this character is a 

 constant one, and if constant, then A. Donianum is a good 

 species. 



Of the British species of Agropyrum, A. Donianum is 

 most like A. repens. The leaves of A. Donianum are more 

 rigid, and more scabrid on both surfaces ; the spike (erect in 

 both species) has coniparatiA-ely shorter and broader glumes ; 

 the rachis is more strongly spinulose, not only on the edges, 

 but on ribs within the edges ; and the axes of the spikelets 

 have longer internodes, which are hairy. The inner pale in 

 A. repens is shortly bifid or emarginate at the apex. 



From A. violaceum the chief distinction lies, as already 

 mentioned, in the lateral awns of the inner pale, but there 

 are other differences. The apex of the inner pale is more 

 acute, and the spinulose pectination of the ribs is more 

 dense, and not so long; the glumes and outer pale are 

 proportionately longer and narrower ; and the axis of the 

 spikelet appears to be somewhat shorter and more longly 

 spinulose. In colour the spike of A. Donianum is usually 

 of a glaucous green, rarely slightly tinged here and there 

 with violet, whilst A. violaceum derives its name from the 

 violet colour of the spike. This colour character is probably, 

 however, of comparatively little importance, since I have 

 found both A. repens and A. caninum with A-iolet-coloured 

 spikes, and Swedish specimens of A. caninum with violet 

 spikes have been distributed as A. violaceum. The spikelets 

 of A. DoniaMum are three to four flowered. 



Though A. violaceum is described as being altogether 

 without stolons, this is, it seems, not quite the case, since 

 Professor Hackel tells me that he has a specimen of the true 

 A. violaceum, which shows short runners, but diffei'ent from 

 the long and scaly runners of A. repens. The stolon pro- 

 duced by my plant of A. Donianum is also short and 

 scaleless. 



As regards the specimen seen by Mr Mitten, and called 

 by him T. hijloru^n, there is, judging from the description, 

 some doubt, since our plant has not " leaves smooth on both 

 sides," nor " usually two-flowered spikelets." A. caninum 

 differs from A. Donianum by its often-nodding spike, softer 



