98 
The Evansia Section 
Pollen, deep yellow. 
Capsule, oblong, i — in. long, trigonal with deeply channelled sides. 
Seeds, small, dark brown, with curious, pale wings at either end (see Plate XLVIII, Fig. ii). 
Observations. 
This curious Iris was first discovered by Pallas and since that time it has been introduced and 
reintroduced into cultivation at intervals. It has often been confused with Pardanthus ( Belamcanda ) 
chinensis and the growth of the two plants is certainly very similar. The flowers, however, are very 
different, for the segments of those of the Pardanthus are all approximately equal while the standards 
of /. dichotoma are noticeably smaller than the falls. The seeds too are very different, those of the 
Pardanthus being relatively large and spherical, with glossy black iridescent coats, and thus easily dis- 
tinguished from those of I. dichotoma (see Plate XLVIII, Fig. n). 
This Iris probably produces more flowers on each stem than any other Iris. The stem is much 
branched and even the branches often issue in pairs at the same point. Moreover from each spathe as 
many as five or even more flowers are produced in succession. Each flower, unfortunately, lasts only a 
few hours and often only opens in the afternoon — a character which gained for the plant the synonym 
of /. vespertina. However, such is the profusion of flowers that there are usually four or six to be 
found open at once on each plant. 
Another peculiarity of this Iris lies in the fact that it does not begin to bloom until about the 
middle of August and then continues in flower for about three weeks or a month. Each flower as it 
dies twists up in a curious spiral and often falls off together with the ovary between which and the 
pedicel there is an articulation. In all the other known Irises, if the withered flower falls very easily, 
it always snaps off above the ovary, except perhaps in the case of I. japonica. 
Iris dichotoma appears to be variable in colour (cf. Franchet, PI. David, i. p. 298), although at 
present only a dull greenish white form mottled with brown purple, seems to be in cultivation. The 
Botanical Magazine figure represents a purple flowered form, and Foster had in flower in August 1905 
a similar plant that he raised from seeds sent from Mukden, which he described as having falls of a 
reddish lurid purple with a dead white signal patch and standards of a somewhat faint red purple. 
It is not a difficult Iris to cultivate especially if it is raised from seed. The young growths suffer 
occasionally from late frosts in spring but in a warm sheltered corner it does well and seedling plants 
will flower in little more than a year from the time of germination. Seed is unfortunately not very 
readily obtained in England even with artificial pollination. In Southern Europe, however, seed is set 
readily. 
I. dichotoma does not appear to be exacting as to soil but the individual plant is probably short- 
lived, and so much of its energy seems to go into the flowering stem that no lateral shoots are formed, 
with the result that the whole plant not infrequently dies after flowering. 
With all its defects, however, /. dichotoma is not to be despised, for it provides us with Iris 
flowers at a season when few other species can be depended upon to bloom. 
THE EVANSIA SECTION 
This section comprises the comparatively small number of rhizomatous Irises in which the central 
line along the haft and part of the blade of the falls is developed into a raised linear crest. 
The first member of this section to be described, namely /. japonica, was so distinct from all the 
known Irises that Salisbury (Trans. Hort. Soc. 1. p. 303, 1812) made a new genus with the name of 
Evansia for it, after one Thomas Evans of the India House and Stepney “who had introduced, besides 
it, so many other plants into this country.” 
With the exception of the crest on the blade of the falls which is the distinguishing mark of the 
section there is probably no other character in which all the plants here grouped together agree. 
Indeed, it almost seems that it is not really a natural group, so great is the divergence found among 
its members. /. tectorum is obviously nearly related to I. Milesii both in its foliage and in its capsule 
and seeds but it is difficult to see much relationship among the others. 
An interesting question arises when we attempt to discover the origin of the crest. Is a beard a 
development from a crest or have both bearded and crested Irises been evolved along parallel lines 
from a beardless Iris ? No definite answer to these questions has yet been given but the attempt to 
find some reply has produced some interesting results. On Plate XXIV there is a drawing of an Iris, 
which was produced by fertilising the Loppio variety of /. Cengialti with pollen of I. tectorum. Many 
previous attempts to cross I. tectorum had apparently been made but without success and this particular 
cross has been, with me, the only one that has produced sound seeds. 
A reference to the plate will show that the hybrid has lost the erect standards of the seed parent, 
Loppio, and assumed the flattened outline of the pollen parent, /. tectorum. The colour is a compromise 
