186 The Nepalensis Section 
there is not really a beard although the much crinkled crest is surmounted here and there by a few 
thread-like processes. 
The chief peculiarity of /. nepalensis is to be found in its rootstock, which in the resting state 
consists of a bundle of fleshy roots, often swollen out towards their extremities and bearing a consider- 
able resemblance to the rootstock of a Hemerocallis (cf. Fig. 25, p. 185). These roots are attached 
to a small hard disc, surrounded by the fibrous remains of old leaves, from the centre of which the 
new growths emerge. 
The exact relationship of /. Collettii to the typical plant is not yet established, although the 
evidence that has accumulated tends to weaken its claim to specific rank. Unfortunately I have never 
yet seen /. Collettii in flower, though I believe that I now possess specimens from Yunnan, which I owe 
to the kindness of the Abbd Ducloux. The typical /. nepalensis seems to differ from /. Collettii in its 
taller growth, its glaucous narrow leaves and in the structure of the latter. These in the type 
bear on one side a single prominent central rib, and on the reverse two similar ribs, one on either side 
of the centre. In /. Collettii the leaves of the non-flowering tufts are broader and have two ribs on 
one side and three on the other of the leaves. 
Moreover, I. Collettii appears to be far more floriferous. Two or more stems, which are sometimes 
very short and sometimes 2 — 3 in. long, rise close together, and the spathes usually produce two and 
sometimes three or even four flowers. 
In cultivation it was Foster’s experience that /. Collettii was easier to manage than /. nepalensis , 
which stands entirely apart from all other Irises in its needs. True to its habits in its native home, 
I. nepalensis lies dormant for six months during the time when the dry N.E. winds would be blowing 
and only grows during the period of the heavy rains of the S.W. Monsoon. Thus even in England 
the leaves die down late in October and do not appear again until April or May. When growth once 
begins it is rapid, and the plants are usually in flower during the latter part of June or the first half of 
July. The individual flowers are very shortlived and do not last more than twelve hours unless the 
weather is very dull and cool. Since this is so, there is less likelihood of pollination being effected by 
insects, and if it is desired to obtain seeds it is best to pollinate the flowers artificially. Capsules then 
form readily and seeds are produced in abundance. 
The seeds should be sown in the autumn rather thinly in large pots, which may be treated in the 
same way as pots of other Iris seeds (see p. 235). Germination, however, will not take place until 
April or May, and care must be taken that the soil is not allowed to become too dry at that time. The 
young plants should be kept well supplied with water throughout the summer and should be allowed to 
remain in the pots. When the leaves turn yellow water should be withheld, and the young plants can 
then be shaken out of the soil and stored in dry sand for the winter. It will be found that each has 
formed from one to three of the tuberous fleshy roots. 
The old plants should also be lifted and stored in dry sand when the leaves wither in autumn, and, 
together with the seedling plants, they should be planted out in March or April in rich, moist soil. 
I incline to think that they do better in a position where they are shaded during part of the day than 
in an entirely open situation. 
/. Collettii 
•Hooker, fil. in Bot Mag. t. 7889 (1903)*. 
Synonyms. 
I. nepalensis forma depauperata, Collett and Hemsley in J. L. S. XXVIII. p. 136 (1890). 
I. nepalensis var. LetJia, Foster in Gard. Chron. 1892, II. p. 458. 
/. Duclouxit, L^veill^ in Fedde, Repert. Nov. Sp. VI. p. 113 (1908). 
Distribution. North Burma, Siam and Yunnan. 
Burmah ; Shan Hills, 1888, Collett, no. 765 (K). 
Upper Burmah; Fort White, 1891, Stone (Foster MS.) (K). 
[" Fort White is situated on a spur of the Northern Chin Hills called the Letha Range.”] 
Siam; Chcngmai, Doi Sootep, 1909, Kerr (1600 — 2500ft.) (K) (B). 
Yunnan; no locality, 1887, Delavaye, no. 2684 (P). 
Mongtse, 1893, Tanaut (P). 
Sen-Tchong-chan, 1905, Ducloux, nos. 349 and 820 (P) (L). 
Eastern Flank of Lichiang Range, 1906, Forrest, no. 2205 (K) (E). 
Tali Range, 1906, Forrest, no. 4840 (E). 
Yunnan-sen, , Maire, 1532 (E). 
Yunnan-fu, 1909, Ducloux (HortD). 
Eastern Flank of Lichiang Range, 1910, Forrest, nos. 5654 and 5929 (E). 
Diagnosis. 
I. Collettii I. nepalensi valde similis sed folia saturate viridia nec glauca, scapus multo brevior, 
nonnunquam omnino obsoletus. 
1 A mistake was made in the Bot Mag. description, which gave the tube as short, for the specimens (K) from which the 
plate was prepared have a tube of more than 1 in. long and they agree with Collett’s original specimens. 
