12 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. 
ary line between the States of Jalisco and Zacatecas. It was first observed by the 
writer south of Huajuquilla, and later in suitable localities southward nearly to 
Guadalajara. 
Specimens examined: 
Jalisco: On the road between Huajuquilla and Mesquitec, J. N. Rose, August 25, 
1897 (no. 2571, type); bulbs from near Monte Escabedo, August, 1897, and 
between Bolafios and Guadalajara, September 21, 1897 (no. 3088); also 
about Guadalajara, Dr. Edward Palmer, 1896 (nos. 345 and 346); same 
locality, C. G, Pringle, 1889, 
A goodly quantity of bulbs was sent to Washington, but they have not been prop- 
erly cared for, 
12. Polianthes geminiflora (Lex.) Rose. FiGure 4. 
Bravoa geminiflora Lex, in Llave & Lex. Nov. Veg. Dese. 1: 
6. 1824. 
Bravoa coetocapnia Roem, Syn. Monogr. 4: 245. 1847. 
Coetocapnia geminiflora Link & Otto, Ie. Pl. Rar. Hort. Berol. 
35. 1828. 
Zetocapnia gemmiflora Link & Otto, loc. cit. pl. 78, 1828. 
Bravoa graminiflora Wemsl. Biol. Centr. Am. 3: 338, 1884. 
(The species name evidently an error.) 
Sulbs rather shallow, giving off several thickened roots; 
stem 4 to 7 dm. tall, perfectly glabrous throughout; basal 
leaves several, 3 to 5 dm. long, 10 to 13 mm. broad; stem leaves 
Fic. 4.—Flowers of Poe 3 or 4, becoming successively shorter; racemes 2 to 4 dm. 
lianthes — geminiflora, 
. long; flowers in a few (6 or more) distant pairs; pedicels slen- 
natural size. 
der, 6 to 10 mm. long; perianth rather pale red or orange- 
colored, about 2 cm. long, strongly bent downward near the base; lobes short and 
broad, rounded at tip and shightly puberulent just below the tip. 
Specimens examined: 
England: Cultivated specimens taken from garden in London. 
State of Mexico: Meadow swells, valley of Toluca, C. G. Pringle, August 19, 1892 
(no. 4215), and also from low meadows at Flor Maria, July 28, 1890 (no. 
3633); near Tultenango, J. N. Rose, July 13, 1901 (no. 5420). 
Type locality: “In montibus Micciacanis, et prope Vallisoletum,” or, as Hooker 
translates,@ ‘‘ Mountains of Valladolid and Miciciacan.’? Valladolid (now Morelia) 
is a town in the State of Michoacan. Although discovered more than seventy-five 
years ago, and in cultivation since L838, it has very rarely been collected. It is native 
of the high mountain valleys of cenral Mexico, the type coming from the eastern part 
of the State of Michoacan. The material collected by the writer in 1901, though 
from the State of Mexico, was from its extreme northern part and very near the 
eastern line of Michoacan. It agrees also closely with the cultivated material which 
may have been derived from the early supplies of this species. This being true, the 
yay is cleared for separating (as graminifolia) the material from the State of Jalisco 
which has been distributed as Bravoa geminiflora. 
My cultivated material was kindly sent me by Mr. E.G. Baker of the British 
Museum. The bulbs have flowered and I have had a good colored drawing made. 
Illustrations: 
Herbert, Amaryl. p/. 72, figs. 5 and 6. 
Bot. Mag. 79: pl. 4741. 
Fl. Ser. 5: pl. 520. 
Link & Otto, Ie. Pl. Rar. Berol. pl. 78, under Coetocapnia (not seen). 
Hooker has figured in Botanical Magazine@ as Bravoa geminiflora a plant which 
seems to differ from my material as follows: The basal leaves are more numerous and 
« Vol. 79, plate 4741. 
