102 STEVENS: NortH DAKOTA PLANTS 
+ MaApIA GLOMERATA Hook. Spring Brook, Aug. 17, 1915, 
a quantity in a prairie slough near the town. 
+ ARTEMISIA PABULARIS (A. Nels.) Rydb. Mandan, Sept. I1, 
1020. Determined by Rydberg. There seems to be no other 
specimen among our material which resembles this. Mr. 
Thysell of the Northern Plains Station showed me the plants, 
only two places where they were growing being known. 
} SENECIO MANITOBENSIS Greenman. Bottineau, July 8, 
1917. Determined by J. Lunell. In the meadow of a dried up 
pond in the Turtle Mts.; Towner, July 13, 1911, Lunell (speci- 
men in the Gray Herbarium). 
7 CENTAUREA PIcRIS Pall. Spring Brook, Aug. 11, 1918. 
Determined by Standley. Received from Jacob Widman who 
referred to it as abundant in a field. Especially a few years 
ago the akenes were often found in Turkestan alfalfa seed, but 
this is apparently the first record of its having become established. 
SONCHUS ARVENSIS L. Ellendale, July 24, 1919. Several 
spots around buildings in the town. 
{SONCHUS ULIGINOSUs Bieb. S. arvensis of Bergman’s Flora, 
Fargo, Aug. 1, 1916; Oakes, July 18, 1919; Crosby, June 7, 1919, 
J. H. Phelps; Williston, Aug. 1915. Specimens of this and of 
the preceding were examined by Standley, who stated that no 
North American specimens were in the U: S. National Museum 
collections. A specimen sent to the New York Botanical - 
Garden was reported by Rydberg as apparently new to this 
country*. This seems rather strange in view of the fact that it 
is the common form through the Red River Valley, and westward 
in the northern part of North Dakota to somewhat beyond Devil's 
Lake, southern Manitoba, and western Minnesota. It is said 
to have first appeared at-Portage Plains, Manitoba, about 1900. 
This form differs from S. arvensis in the absence of glandular 
hairs on the upper stems and involucres. While I have not 
been able to compare carefully fresh material, I believe the heads 
are smaller, paler, and the rays more inclined to become recurved. 
Vegetative reproduction is by long horizontal. roots and not 
rootstocks as commonly stated. The development of the 
terminal head is often stopped before flowering and several flower- 
ing branches arise just below it. 
* 
mall has recently listed the species as an addition to the American 
flora, from Pennsylvania, where it was collected in 1921 (see Torreya 21: 100. 
1922). 
