! 
BOTANY. ened 157 
of California. It begins to flower when scarcely an inch high, but at length, in good soil, attains 
the height ofa foot or more. The flowers vary greatly in size, from less than an inch to an inch 
in diameter. They are of a bright rose color with a yellow centre. We refer here E. Muhlen- 
bergii, Benth. Pl. Hartw., apparently a mere dwarf, small-flowered state of the plant, which we 
have from several collectors. This species is known in California by the name of Canchalagua, 
a bitter tonic of considerable reputation; but E. tricantha is often confounded with it, and 
possesses similar virtues. The same name has long been applied to E. Chilensis. Gyrandra 
speciosa, Benth. Bot. Sulph. р. 127, t. 45, differs from this species chiefly in the shorter filaments, 
and in being more spreading. 工 have noticed at least in one species that the filaments increase 
in length after flowering, as do the tube of the corolla and the style. In authentic specimens of 
Berlandier, (whose plant is certainly annual, not as stated by Grisebach, perennial,) the stem is 
4-angled, as in all the species of Erythrwa. The genus Gyrandra seems to have no characters 
by which it can be distinguished from Erythrea. The segments of the calyx are somewhat cari- 
nate, but not winged as in some genuine Erythrwa, The corolla is of the same form in both 
genera. The anthers of E. Centaurium, as well as of other species that I have examined, are 
as destitute of a connective as are those of Gyrandra. 
Евүтнвжа TEXENSIS, Griseb. Gent. р. 139, din DC. Prodr. l. c., p. 58. Rocky places on 
the Pecos; Bigelow ; and along the Rio Grande ; Schott. Aprii—September. 
Евутнвжа Вкүвїснц, Torr. & Gray, Fl. 2, p. —, (ined.;) Torr. in Воі. Marcy’s Rep. р. 291, 
2. 13. E. tricantha, 8 angustifolia, Griseb. in DC. l. с. 60. Rocky banks of the Pecos and 
San Pedro; Bigelow, etc. Sabina creek, Texas; Thurber. (No. 1662, Wright.) 
ERYTHRZA TRICANTHA, Griseb. Gent. р. 146, and in РО. Prodr. 9, p. 60; Benth. Pl. Hartw. р. 
322. In various parts of California, especially near the coast. We have from Sir William 
Hooker specimens of this plant, collected by Douglas, which agree in all respects with ours. 
The lobes of the corolla are not “linear and very acute," but lanceolate and rather obtuse. 
Grisebach probably drew his description from specimens which were dried without pressure, so 
that the lobes of the corolla had become involute and apparently very narrow as well as acute. 
A fter a careful comparison of original specimens of E. tricantha and E. floribunda, (the California 
plant of Benth. Pl. Hartweg,) I have little doubt that they are forms of one species. E. flori- 
bunda differs only in the broader lobes of the corolla. 
GENTIANA DETONSA, Fries. Griseb. in DO. Prodr. 9, p. 101 ; Torr. Fl. N. York, 2, p. 108, t. 82. 
Babacomori to Santa Cruz, in wet places, September; Thurber. (No. 1658, Wright.) 
GENTIANA AFFINIS, Griseb. in Hook. Fl. Bor.-Amer. 2, p. 57. Hills and rocky places near 
the Copper Mines, August—October; Bigelow. (No. 1657, Wright.) 
GENTIANA QUINQUEFLORA, Гат. Dict. 2, p. 643; Griseb. 1. с. Western Texas. No. 1659, 
Wright. ` a 
Eustoma ROUSSELIANUM, G. Don, Gen. Syst. Gard. 4, р. 175; Griseb. in DC. Prodr. 9, p. 51. 
Valley of the Rio Grande, etec., in sandy soils; common, May—October ; Thurber. On the 
Gila, Sonora; Thurber. (No. 1660, Wright.) Leaves varying from oblong and obtuse to 
narrowly lanceolate and acute. The flowers also vary greatly in size and in the proportionate 
length of the segments and tube of the corolla Perhaps not distinct from E. exaltatum. 
SABBATIA campestris, Nutt. in Trans. Amer. Phil. Soc. (n. ser.) 5, p. 197. Prairies on the 
Guadalupe river, Texas, May—June; Wright. (No. 2568, Berlandier.) 
