1058 



abundantly in some barren fields near Ragland. I acknowledge that 

 the plant exhibits in other particulars the closest resemblance to F. 

 dentata, especially to the variety F. eriocarpa, but I cannot persuade 

 myself that so marked a difference in the fruit is of no value. The 

 Valeriana of the valleys is V. sambucifolia. The leafits are four or five 

 pairs, with a trifid terminal leafit. Near Monmouth I sometimes found 

 five or six pairs with a quinato-pinnatifid terminal leafit. The cup from 

 which the processes of the calyx arise, is sometimes fully as wide as the 

 widest part of the conico-ovoid seed, at others it is divided quite down to 

 the base. The rhizoma is sometimes almost perpendicular, exhibiting 

 one set of fibres above the other ; but I did not notice this in any 

 flowering specimens, and in such cases the lower root, not the upper, 

 sends forth slender runners. The upper is probably the production 

 of the year, instead of a flowering stem, and it would flower, and send 

 out runners the next year. Independent of this descending rhizoma, 

 there are, both for the flowering and barren plants, superficial run- 

 nel's, often very numerous, producing in their course pairs of scales, 

 and sometimes a single leaf; but I could not see any pairs of leaves, 

 except at the end, were the fibres of a new root were produced. 



Triticura caninum, which, I think, is not mentioned in the Bota- 

 nists' Guides, grows near Pont-nedd-vechan, both on the Purthin and 

 on the little Nedd, and there are several tufts of Gnaphalium margari- 

 taceum below the Dinas Rock, a large limestone rock rising abruptly 

 at the head of the open valley. Lepidium Smithii grows at Pont- 

 nedd-vechan, and at Pont Walby. At the latter place, and, I believe, 

 at Cil-Hepste, there is a curious form of Equisetum, perhaps of E. ar- 

 vense, in which the lower branches are subdivided, sometimes even 

 producing secondary whorls. It has the long terminating division of 

 the frond which has been noticed in E. umbrosum, with which in 

 other respects it does not well agree. Its situation on wet banks, or 

 in the spray of a waterfall, is very different from that in which we usu- 

 ally find E. arvense, and it may perhaps prove a distinct species. 

 Orobanche Hederae grows at Penrice and at Sully. Genista Anglica, 

 which I do not see mentioned in the ' Botanist's Guide' for Glamorgan, 

 grows at Rhydgroes, near the station of Rubus suberectus. Mr. Wat- 

 son in the Bot. Gaz. i. 59, has given a list of plants not very rare, but 

 not hitherto recorded in the botany of South Wales. Of these Ra- 

 diola millegrana occurs on the ascent of Cefn Brin, in Gower ; and 

 Silaus pratensis is common on the bog in the eastern part of the 

 county. 



Near Pyle, among the sand hills on the shore, is a large pond, 



