472 TRANS. OF THE ACAD. OF SCIENCE. 
superante; capitulis magnis densissime multi-(30-80)floris ; 
floribus majoribus virescentibus demum stramineis; sepalis 
lanceolato-subulatis exterioribus longioribus ; antheris linea- 
ribus filamento paulo brevioribus; ovario lanceolato in sty- 
lum brevem sensim abeunte; capsula pyramidato-subulata vix 
exserta.—J. megacephalus, Wood, Bot. 724, non Curtis. 
This species takes a much wider geographical range than 
the last, including the whole of North America north of Mex- 
ico, with the exclusion of the south-eastern States; but the 
different forms occupy different geographical regions. Var. a. 
is found throughout British North America from Canada and 
the Hudson Bay regions to the Rocky Mountains and the 
North-west coast, and extends southward to Pennsylvania, 
Porter, Hb. n. 70, Ohio, Lapham, Michigan, Bigelow, Hb. n. 71, 
and Wisconsin, Lapham, Hale; 1 have seen no specimens 
from further south, though the older authors credit it to Vir- 
ginia and Carolina, quoting, among others, Bose as their au- 
thority. Var. 6 has been solely found in Western Texas, Lind- 
" heimer, 545, Wright, Buckley. Var. y meets a on Lake Ontario, 
where also J. alpinus and articulatus join, and extends from 
thence westward to Michigan, Bigelow, Hb. n. 74, and south- 
westward to Illinois, Missouri, the northern Red River, Hub- 
bard, the Saskatchawan, Bourgeau, the Yellowstone, Hayden, 
Colorado, Parry, Hb. n. 75 (a dwarfed form), New Mexico, 
Fendler, 849 Wright, 696 & 1926, Texas, Lindheimer, 546, and 
others; and to Arizona, Coues & Palmer, and California, Coul- 
ter, 809. It flowers from July to August. 
Our plant is very closely allied to the last one, and is often 
confounded with it; but the number of stamens and the mark- 
ings of the seeds will readily distinguish any of the forms which 
may be mistaken for one another, e. g. J. scirpoides, p echina- 
tus, and J. nodosus, y megacephalus ; besides, the slender sto- 
lons which terminate in a chain of small bulbs, probably the 
only part that sustains the life of the plant during winter, are 
quite characteristic of all the forms of this species. Another 
peculiarity of var. 6 and y is the direction of the leaves, 
especially the upper ones, which are patulous, making a very 
distinct angle with their sheathing base, while in var. a the 
leaves are erect, forming an almost straight continuation of 
the sheath. Var.a and more rarely var. y exhibit sometimes 
that degeneration of the heads into bunches of sheaths or 
leaves which has been spoken of in another place. The 
seeds are 0.22-0.27 line long, the length being nearly equal 
to 2, rarely to 24 diameters; commonly 8 ribs are visible on 
one side. 
The northern form, var. a, is the genuine J. nedosus of 
Linnzus, who described it from specimens sent by Kalm 
(most probably from Canada), as Prof. Gray ascertained in 
the Linnean herbarium itself; he informs me that “ Linneus’ 
