well be the case if the fig. 868, in the illustrated edition of the 'Handbook of the British Flora' be accepted as characteristic of P. Persicaria. It has, however, all the habit of P. maculatum, and the stifles are represented unconnected and deflexed as in P. lapathifolium.
About London P. Persicaria, L., presents itself in the typical form, well figured in a new plate by Syme in E. B. MCCXXXVII., and by Reichenbach (Iconog. Bot. ic. 684). This has a stem rarely spotted, with rather loosely arranged divaricating branches and stout, rather short (compared with P. maculatum), blunt, oblong or cylindrical, remote racemes. The plant figured by Curtis (Flor. Lond. f. 1) is probably a luxuriant state of the type. The old figure of P. Persicaria, E. B. 756, now used by Syme to illustrate the variety elatum, is one from which all definite characters seem to have evaporated, and is almost valueless. The authentic specimens of P. Persicaria in Sowerby's herbarium are really P. maculatum, and it is by no means impossible that this plant was in part copied by the artist, though the details are from P. Persicaria.
A much rarer form of this species occurs occasionally in moist cultivated ground. It may be distinguished from the type by its more succulent and swollen stem, ascending and not divaricate branches, and racemes not cylindrical, but longer and attenuated upwards, contracted when young into a panicle. The foliage also is somewhat weaker in texture, and of a livelier green. This is the P. biforme of Wahlenberg (Suec. n. 437) and of Fries (Mant. ii. p. 28), who quotes Curtis's figure, in which he is perhaps right,[1] and also Reichenbach's, which belongs to the more typical form. Authentic specimens of P. biforme, from Fries, in the Kew Herbarium, it is noticeable have ochrese without cilia, though this does not accord with the description. P. Persicaria, β. elatum (Gren. et Godr. Fl. de Fr. vol. iii. p. 48), is another synonym, and it has already been shown to be probably the true P. nodosum, of Persoon. P. laxum, Reich., which seems to have nothing really to do with P. maculatum., may be also cap- able of inclusion here. Some writers (Boreau, Fl. du Cent, de la Fr. ed. 3, t. ii. p. 557, Gren. et Godr. 1. c. p. 49) regard P. laxum, Reich., as a hybrid between P. Hydropiper and P. maculatum. If so, it is remarkable that a glandless plant should be the offspring of two glandular parents. Fries says of this plant "transitus ad Mite," (Mant. vol. ii. p. 26).[2] A curious variety referable to the typical form of P. Persicaria, but which may be really a hybrid, has the peduncles slightly glandular hispid. It has been distributed by Mr. Watson from Esher, and has been noticed at Kilburn and Haverstock Hill. There are also specimens from Essex in E. Forster's herbarium.
It will be convenient to give the characters and synonymy in a col- lected form.[3]
- ↑ In Fl. Midd. p. 244, we have also quoted it, but our notion of this form was at that time more comprehensive than now.
- ↑ Reichenbach, in his description in the 'Iconographia Botanica,' was not disposed to accept the suggestion of hybridity, remarking "nimis enimprodigiosa mihi videtur hybriditas hodierna." However, in the 'Flora Excursoria,' 3696, he gives Hydropiperl-nodosum as a synonym, with the explanation "Inflorosccntia densa gracilis floresque parvi rosei P. nodosi, folia contra et ochreoe P. Hydropiperis."
- ↑ Linnæus seems to have founded his P. lapathifolium on a species of Tournefort's, and the only locality he gives is "Gallia"; as he describes the plant as "floribus . . . semidigyuis," it is somewhat doubtful what he really meant. The Linnean Herbarium only confuses matters, the specimen named "lapathifolium" being a widely
