REPORT OF WORK IN 1915 IN KANSU AND TIBET. 335 
/. ensata (F 496), on the contrary, sheets all the loess region of these 
parts in one ocean of green, which in the end of May becomes a pale 
sea of blue, washing up even into comparatively alpine bays at ten 
thousand feet. For its form is far better than elsewhere — generous 
in shape and of the loveliest, softest blend of milky blue and lavender 
and cream and white, yet further enhanced by its delicious fragrance 
of hyacinth. In front of the Tien Tang Abbey itself, the huge level lawn 
was one solid sheet of the Iris, and to walk through it is to be trans- 
ferred to Holland in hyacinth time. I think hard, hot, poor soil in a 
hot place will give us our best chance of getting the best out of this 
/. ensata, which, in richer, damper places, tends to run to leaf and 
rarity of bloom. Albinos occurred to me several times, and once a 
form with flowers of darkest sapphire velvet ; otherwise, but for minute 
differences in shape, the species is wonderfully stable. (Painting.) 
I. Potaninii (F 500) I only saw once, on the ledges of one hot 
dolomite cliff, facing the full sun, at the foot of Wolvesden Pass. It is 
like a little Iris of the Chamaeiris group, forming neat tuffets from 
which spring almost scapeless the small Flags, with sharply tucked- 
under falls, in tones of straw-yellow or musty purple. In the same 
cushion, though not probably from the same roots, you will get 
the two colours ; the purple-flowered is slightly larger, the straw- 
yellow distinctly the prettier. (Painting.) 
/. tenuifolia (F 499) is the most beautiful of the year. Its wide old 
grassy hassocks do not love the loess, but delight in the open loamy 
green lawns of the cool alpine foothills, especially about the folds of 
the fell in which Chebson Abbey lurks. Here, in May, it was lavish 
with its large and lovely flowers of rich blue, each on what seemed a 
stem of six inches or so, but all of which was nothing but the pre- 
posterous tube of the flower, for actually in the ground developed 
the fat rosy pods, just protruding their pink bulges sometimes, but 
often, it is evident, lurking undiscovered for years in the heart of the 
clump, nursing what still seems good seed. /. tenuifolia seems slow of 
growth even at home, where the clumps show masses of dead leafage- 
stumps, and even hassocks of dead matter from which spring scant 
sprouts. Nor does it seem certain either in flower or seed. The 
species has an enormous range, and the specimens already in cultiva- 
tion hail from Quetta, not a land of promise so far as our gardens are 
concerned. So that their unalterable sulkiness need not give any 
rule for the behaviour of a fresh stock from so far away, and from 
conditions so absolutely different as the cool high downs of Chebson, 
far up on the northernmost limits of Tibet. (Painting.) 
I. Tigridia (F 498) ranks second to /. tenuifolia. It haunts only 
the torrid amphitheatre of cliffs and loess-banks behind Tien Tang 
Ssu, loving to grow on the rim of the steep breaks, sprouting from 
under some slight covert of scrub in a tuffet of short greyish foliage 
from which stand up the beautifully-balanced little flower-de-luces 
in blended tones of amethyst, claret, and blue-violet, on stems of three 
or four inches, in May-June. The seed had mostly fallen when next 
