Page:The Botany of the Antarctic Voyage.djvu/374

338 It appears to me that two very distinct species of this genus have heen confounded, partly together, and partly with the A. tenella, L., of Europe. The first is confined to the damp western portions of middle and southern Chili, Fuegia, and the Falkland Islands, and there are two or perhaps three varieties of it; I take it to be the A. alternifolia of Cavanilles, a variable plant, with the peduncles of the same length as, or not much exceeding, the leaves, and the capsule shorter than the calyx. The figure of that author is very inaccurate and at variance with his description; for the plant is represented erect, instead of creeping, and the leaves scattered, though said to be, approximate. Supposing Cavanilles' plant to form one variety of A. alternifolia, a second is larger and also creeping, with prostrate branches, 8-10 inches long, bearing broader, rounded and more acute leaves; it has been collected in Valparaiso by Mr. Cuming and Mr. Bridges. A third, intermediate between this and the Fuegian form, has the leaves more crowded, ovate-oblong, and smaller; it is possibly the state figured by Cavanilles, and has been gathered at Concepcion by Capt. King, at Valdivia by Mr. Bridges, and on the Andes of Mendoza by Dr. Gillies (Ruellia caspitosa, Gill. MSS.; and Anagallis herpestoides, Gill. MSS.). The fourth variety is what I have called densifolia; its leaves and stem are much smaller and crowded, and the whole plant is succulent.

Another extra-tropical South American Anagallis is the A. filiformis, Link, (A. tenella, &beta;. filiformis, St. Hil.), which approaches A. tenella so very closely, that M. St. Hilaire has united them specifically. It differs from A. alternifolia in the leaves being opposite, the stem slender, the peduncles longer, the calycine pieces narrower and twice as long as the capsule, and the whole plant not so succulent; from the European A. tenella in the leaves never being so broad, in the longer peduncles and rigid stems.

The variety densifolia has a large capsule, always equalling the calyx in length, thus differing from the plant figured by Cavanilles. The capsule, though described by D'Urville as having the dehiscence of a Lysimachia, evidently opens transversely in the specimens I have examined, though it is sometimes, from pressure, split at the top also. Its habit resembles the Abyssinian A. serpens, Hochst.

3. SAMOLUS, L.

1. littoralis, Brown, Prodr. p. 428. ''Duby in DC. Prodr.'' vol. x. p. 73. Scheffieldia repens, ''Forst. Nov. Gen.'' p. 18. t. 9. Chonos Archipelago and Cape Tres Montes, C. Darwin, Esq.

A plant common to New Holland, New Zealand, and South Chili, and very variable in the size of its parts in all these countries. I have not seen Chilian specimens from a lower latitude than Valdivia, between which and Cape Tres Montes it seems limited.

2. spathdatus, Duby, in ''DC. Prodr.'' vol. x. p. 74. Andros&aelig;a spathulata, Cavanilles Icones, vol. v. p. 56. t. 484., f. 1.

Strait of Magalhaens; Port Gregory, Capt. King. Elizabeth Island, C. Darwin, Esq.

The raceme, in most of Capt. King's specimens, is so much abbreviated that the flowers are almost capitate. The range of the species, between Port Desire and the Strait of Magalhaens, is remarkably limited.

1. PINGUICULA, Linn.

1. Antarctica, Vahl, Eunm. p. 192. ''Alph. DC. Prodr.'' vol. x. p. 31. P. obtusa, ''Banks et Sol. MSS. in Bibl. Banks''. (Tab. CXIX.)

Strait of Magalhaens; Port Famine, Capt. King; Good Success Bay, Banks and Solander; south part of Fuegia, C. Darwin, Esq.; Hermite Island, Cape Horn, J. D. H.