108 



tionably nearly allied, some few remarks upon their differential cha- 

 racters are necessary. Adopting only those derived from the flower, a 

 plan which deference to long habit has rendered all but peremptory 

 in botanical science, the structure of the spikelets in the entire series 

 is so strikingly similar, that, apart from proportion, pubescence, and 

 other accidental circumstances, no plausible pretext could be found for 

 their separation. Such separation then can only be established by 

 reference to features usually considered beyond the pale of floral 

 orthodoxy ; the value, however, of some of these which are intimately 

 associated with plant life, and distribution, cannot be disputed, even by 

 the most rigid disciples of the species-condensing school, of which, as 

 in p. 73 of the present work, I have already confessed myself a 

 member. It is upon some such features that my belief is founded of 

 an actual specific distinction existing between two at least of the grasses 

 now before us. Cultivated for many years, and raised occasionally 

 from seed, they all maintain their original characters without the 

 slightest deviation being produced by difference of soil and situation ; 

 and admitting F. vimpara to be an understood variety of one of the 

 others, and the possibility that F. coesia may bear the same position with 

 regard to F. ovinq,, which I doubt, F. duriuscula and F. rubra still 

 remain with a difference of habit so marked as to render their sup- 

 posed identity with the typical species of this section incompatible. 

 The stouter rounded stems and broader channelled, not bristle-like, 

 leaves of the Tall Fescue Grass, the pyramidal form of its larger and 

 looser panicle, are constant features of distinction between them. In 

 regard to F. rvibra, its stoloniferous character affords very positive 

 ground for retaining it as a species, though misled by an occasional 

 tendency in F. duriuscula to produce, in light soils, weak root shoots, 

 some botanists have confounded them. Had these latter witnessed 

 the pertinacity with which the scions of the Purple Fescue, properly so 

 called, are extended in the hardest ground and beneath their garden 

 walks, they would possibly be of a different opinion. The general 

 aspect of the two plants is otherwise about as dissimilar as can well be 

 conceived. 



AU the specimens I have met with of F. vivipara appear to belong 

 to F. duriuscida rather than to F. ovina, and are very unlike the long- 

 leaved variety of that species found in the Highlands of Scotland, 

 F. tenuifoUa, of Sibthorpe. This last grass is figured by Dr. ParneU, 

 plate 57 of his ' Grasses of Scotland,' under the name of angustifoKa; 

 the absence of awn at the extremity of the palea formerly led to the 

 supposition that it might be a distinct species ; the opportunity of 

 making it such is, however, allowed to pass by modern writers, and, as 

 apart from the non-development of a terminal hair, length of leaf is 

 the only characteristic, the bare notice of its existence is sufficient. 



* * * Radical leaves flat, broader than those of the cuhn. Awn absent, 

 or arising from below the summit of the outer palea, never ter- 

 minal. Ligule of the uppermost leaf prominent, obtuse. Outer 

 palea three-veined. 



Schedonorous — Palisot de Beauvois. 



