140 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. 
The low acaulescent habit, simply pinnate leaves, and other characters of L. east- 
woodae and L. macounii, so unlike our other species of Ligusticum, has suggested segre- 
gation as a genus. Species of similar habit, however, such as L. aromaticum Banks 
& Soland, and L. enysii Kirk, occur in New Zealand, and such segregation would 
demand an extended study of the whole genus. 
In addition to the species described above, numerous specimens of Ligusticum have 
been collected which do not seem to belong to them, but the material is either too 
scanty or too incomplete for satisfactory determination. We cite some of this mate- 
rial in the hope that collectors may be able to supplement much of it. 
A plant of the mountains of southeastern California, collected by Purpus (no. 5661) 
at ‘ Farewell Gap,” altitude 1,090 meters, in 1897, is not associated geographically 
with any known Ligusticum. It suggests L. porteri somewhat, which grows in north- 
ern Arizona, but it is acaulescent, with small fruit (4 to 5 mm.) and prominent coni- 
cal stylopodium. 
We can not determine Kellogg & Harford 304 and 309, June 24, 1868, from Oakland 
Hills, California. The plants are apparently different from either L. apiodorum or 
L. californicum of the same region. 
Undetermined forms from the upper Rogue River region of southwestern Oregon _ 
are Applegate 2638 (Jackson County), July, 1898; and Coville & Applegate 1184 (Jef- 
ferson Mountain), August, 1898. , 
A form collected in flower near Montesano, Chehalis County, Wash., by A. A. & 
E. Gertrude Heller (no. 3973), June 28, 1898, has foliage quite different from any of 
the known species of Ligusticum. 
In the Cascade Mountains of Washington, and especially in the region of Mount 
Rainier, there is a group of forms which have been confused with L. aptifolium, and 
which possibly represent several species. Some of them are as follows: Suksdorf 58 
(Mount Adams), September, 1882; Tweedy 288, in 1882; Piper 629 (Mount Rainier), 
August, 1889; Smith 1050 (Mount Rainier), August, 1890; Henderson 378 (Mount 
Adams), August-October, 1892; Gorman 669 (head of Poison Creek), September, 
1897; Gorman 816 (Horseshoe Basin), October, 1897; Horner 216, July 17, 1897; 
Elmer 1222 (Mount Stuart, Kittitas County), August, 1898. 
Undetermined forms from northeastern Washington are the Wilkes Exped. plant 
from Fort Colville and Spokane River, and Sandberg & Leiberg 731, in 1893. 
In Canby’s Alaskan journey of 1897, plants were collected in the Selkirk Mount 
tains at Glacier, Canada (no. 96), and in southern Alaska, which seem quite dis- 
tinct from any known species of Ligusticum. 
A Ligusticum from the Coast Mountains near Waldo, Oreg., collected by Howell 
(no. 121), June 13, 1884, seems to differ from any known species. It may possibly 
represent Cynapium nudicaulis Nutt. in Torr, & Gray Fl. 1: 641. 1840, the deserip- 
tion of which it seems to answer, but it is somewhat out of its range. The specimen 
has no basal leaves, so that no complete characterization is possible. 
43. COELOPLEURUM Ledeb. FI. Ross. 2: 361. 1844. 
Calyx teeth obsolete. Fruit oblong, slightly flattened laterally if 
at all, glabrous. Carpel with very thick and prominent corky ribs, 
becoming hollow, the laterals broadest or all equal, each with a large 
group of strengthening cells. Oil tubes small, 2 to 4 on the commis- 
sural side and 1 or 2 under each rib (in addition to the one in the 
interval), all adhering to the seed, which is loose in the pericarp, and 
with plane or somewhat concave face. 
