112 The Flora of Wiltshire. 



yields a larger crop than any other sorts. In general appear- 

 ance " T. pratense " is very like " T. medium," (not as yet observed 

 in Wilts.) Afzelius in his most elaborate paper in the " Linnaean 

 Transactions", has clearly distinguished them. The root is creeping 

 and more uniformly perennial in cultivation. The stems are zigzag 

 and more branched. Stipulas longer, linear, tapering to a point 

 and stand parallel to each other. Meads of flowers rather larger 

 and less dense. Cali/x slightly hairy except in the mouth, its two 

 upper teeth shortest, the rest gradually but not very strikingly 

 longer. Leaflets elliptical, various in width, a little glaucous 

 underneath, chiefly hairy at the margin. 



2. T. arvense (Linn.) Hares-foot Trefoil, called by old authors 

 lagopus or hares-foot, from {lagos) a hare, and {pons) a foot, from 

 its dense soft spikes somewhat resembling a hare's foot. Engl. Bot. 

 t. 944. St. 16, 3. 



Locality. Cultivated and waste ground on gravel, also in sandy 

 fields not unfrequent. A. Fl. July, September. Area, 1. * 3. 4. 5. 

 South Division. 

 1. South-east District, Sandy fields near Pewsey. 



3. South-west District, " Corsley," Miss Griffith. 



North Division. 



4. North-west District, "Chippenham," Dr. Alexander Prior. 

 " Tytherton," Mr. Coward. 



5. North-east District, Swindon. " Great Bed wyn," il/r. William 

 Bartlett. Probably not xmcommon in many parts of the county. 



Flowers small in pedunculate heads which are at first nearly 

 globular but soon become oblong or cylindrical, six to nine lines 

 long, appearing very soft and feathery, owing to the fine hairy 

 teeth of the calyx projecting beyond the very small corolla. 



[T. striatum (Linn.) and T. scabrum (Linn.) have both been 

 reported to grow in "Wilts, the former near Westbury, (District 2) 

 the latter at Market Lavington, {District 2.) There appears 

 nothing unlikely in the occurrence of these species in the county, 

 but it would be desirable to have well authenticated specimens 

 with their localities recorded for these plants, because the two are 

 not unfrequently mistaken for each other.] 



