306 ON THE EUROPEAN SPECIES OF FESTUCA. 



naturally be expected — are dispersed throughout an extended area 

 they are, F. ovinu, rubra, elatior, and varia. 



The followmg subspecies are monomorphic : — F. urina subsp. 

 Beckeri, Border ii, alpina, and hrevifolia ; F. rubra, subsp. heterophylla, 

 pyrenaica, nevadensis, and dumetorum ; F. varia, subsp. Eskia, 

 a/pestris, zantldna, fiavescens, and jmmila. 



Those species which have folded leaves are evidently later 

 developed forms, induced by dryness of both air and soil ; hence it 

 follows that these forms are more subject to variation. The con- 

 trary is the case as regards the flat-leaved species, which have now 

 almost ceased to vary, though we should except F. elatior. 



The folded leaf must, nevertheless, not be taken as necessarily a 

 sign of aftiuity. The flat leaf occurs in all three folded-leaf sections. 

 Neither is the form of the ligule of any use as betokening aftinity. 



The ovarium of the ancestral form appears to have been hairy. 

 In four sections it is universally so, and examples occur even in 

 the glabrous-ovary section, viz., in Ovina;. In those sections which 

 have a hairy ovary the species are more sharply defined and 

 separated, which is a token of antiquity. 



The adherence of the pale to the ovary is a late character. In 

 the older ;;pecies, viz., those with a hairy ovary, the latter is free 

 or only slightly adherent. Extravaginal growth of the bud is of 

 greater antiquity than intravaginal growth. We find that as the 

 tendency of the leaves to become folded increases the extravaginal 

 growth disappears. 



The oldest form Prof. Hackel believes will be found in the 

 group extravaginales of the section VaricE and in the section Montana,. 

 The first-named group [Amphigenes, Janka) inhabits widely separated 

 areas, viz., the Southern Alps, the mountains of the Balkan 

 Peninsula, the Carpathian Mountains, the Appennines, and Sierra 

 Nevada ; the areas are constricted, and some of the species are 

 exceedingly rare ; all which facts point to extreme antiquity. The 

 species of the Mo)ita7ia group are more widely distributed, but 

 their slight tendency to variation (ex. F. sijlvatica) favours great 

 antiquity. The section Scariosa, with its isolated representative, 

 F. granatensis, may be looked upon as offset from the Amphigenes 

 group. All three groups stand nearer to Poa than do the rest of 

 the Fescues (hence " Amphigenes,'' Janka, Poa sylvatica, Poll., the old 

 name for F. sylvatica; and Poa scariosa, Lag., the old name for F. 

 granatensis), and are therefore less specialised, less typical. The 

 section Subhulhosa: exhibits a strong African relation ; its members 

 show little aftinity with European Fescues ; jind possibly it may be 

 found to have more with African species. The members of the section 

 Bovina: undoubtedly exhibit great antiquity, for they have retained 

 many primitive characters. There are in this section, in Europe, 

 two strongly- marked species, one of which — F. gig(iiitea — has a 

 remarkably lu-oken area of distribution ; nevertheless the author 

 considers the species which constitute this section as later than 

 the Anipliigeiu's, for from it has evidently sprung the nearest of all 

 the Fescue forms, viz., the Orina group, which again have probably 

 sprung from /''. rubra. 



