the 



re IS 



th 



29 



almost constantly browse on these, and almost entirely neglect the Cynosums cristatus LoUi 

 perenne, and Poa trivkdis. The Welsh breed of sheep not being bred in the Pari i! * f 



ence to particular grasses must be the effect of something else than habit The b - 1 

 difference of opinion with respect to the merits of this grass: it certainly does not affo I ^ ■] r 

 a bite to cattle in the spring as many other grasses, and the culms are uniformly left I \ 



but this is more owing to the season in which they are produced, than to any particular defect* 



^ en a profusion of root leaves and herbage in general, which is always preferred by 

 cattle to the culms : when the grass is in flower, the cnlms are succulent, and contain much 

 nutritive matter; it is all, however, exhausted in perfecting the seed. If this grass is employed 

 only for the alternate husbandry, and its merits from thence estimated, it will be considered 

 an inferior grass, as it is by no means adapted for that purpose, either with respect to speedily 

 arriving at perfection, early growth, or quantity of produce; but its forming a close dense 

 turf of grateful nutritive herbage, and being little affected by the extremes of weather, where 

 other grasses, superior in the fore -mentioned points would be produced in tufts, and injured by 

 the extremes of weather. From these facts it is evident, a sward of the best quality, particu^ 

 larly under circumstances where sheep are a principal object, cannot be formed without an ad- 

 mixture or proportion of the crested dog's-tail grass. It flowers towards the end of June, ' and 

 ripens the seed towards the end of July. 



* * 



\ 



