44 Contributions to Western Botany. [ZOE 
and as it is often the case that the full development is arrested by 
the dry weather, se the majority of the carpels are by 
necessity indehiscen he fruit is broadly ovate to rhombic 
and when the ah occur above the middle of each carpel then 
itis oval, but never ‘‘ orbicular” in any specimen from the Great 
Basin that I ever saw. ‘The peduncles are always short and like 
the type. The plant is three to six feet high, grows among 
other brush, is widely branched and closely resembles J. 
opulifolia in general appearance, but is a little stiffer. It ranges 
from 7000 to gooo feet altitude, and prefers the north side of 
steep mountain sides as there only can it get enough moisture. It 
is common in the mountains, and I have it from many localities 
in all stages of development. 
Comparing my notes with those of Mr. Greene I find no 
character left to separate it from /V. monogyna and only the flattened 
pod to separate it from JV. ofulifolia, while he gives WV. monogyna 
as having a somewhat inflated pod which destroys the last valid 
distinction. : 
Watson reports the type monogyna as from the East Humboldt 
Mountains, Nevada, and from Stansbury Island, in Great Salt 
Lake. I have not seen his specimens, but presume they are the 
var. alternans, 
Since writing the above I have found a fine fruiting specimen 
in my collection from Albuquerque, New Mexico, which belongs 
to the type of NV. monogyna. ‘The calyx is much inflated or little 
so, lobes often emarginate; carpels two or three in each calyx, 
tips needle-like and widely divergent, carpels separate to below 
the middle, fully as inflated for their size as the less inflated 
form of iV. opulifolia from Colorado described above, or perfectly 
flat and silique like, scarcely over half the length of the calyx or 
one-fourth longer, one to three-seeded, seeds very broadly 
obovate, scarcely yellow, and angular by being crowded in the 
carpel, carpels dehiscent and bivalvularly so to the middle. 
ese variances all occur on the one specimen. The only way to 
uphold /V. monogyna seems to be that adopted by Gray to keep up 
nti namely by an aggregation of characters no one of which 
$s permanent, but some of which are always present when the 
a fail. 
