140 
The Pogoniris Section 
of affinity to the Regelia group, for the young shoots of /. Korolkowi and /. stolonifera are also deeply 
coloured. The colour disappears, however, when the leaves have grown to 2 or 3 inches in length 
and does not persist at their base as in the case of the Regelia species. 
/. MANDSHURICA 
Maxim, in Bull. Acad. P£t. XXVI. p. 530 (1880). 
M£l. Biol. X. p. 724 (1880). 
DISTRIBUTION. Southern Manchuria, where it was found by Giildenstadt near the river Suifun. 
South Ussuri district, 1897, Komarov (K). 
North Corea; Talu, 1897, Komarov (BM). 
North China, 1880, Webster (K). 
Diagnosis. 
/. mandshurica Pogoniris; inter /. flavissimam et /. Bloudowii ; ab ilia rhizomate magis compacto, 
foliis latioribus, ab hac spathis angustis viridibus differre videtur. 
Description. 
Rootstock, a slender, somewhat straggling rhizome with the bases of the growths wrapt in membranous 
sheaths. 
Leaves, ensiform, green, 6 — 8 in. by £ in. 
Stem, 3 — 6 in. long, sometimes bearing a lateral branch, set in a leaf-like bract (cf. Komarov’s 
specimen (K) and Observations). 
Spathc, 2 -flowered ; valves often unequal, lanceolate, i£ — 2 in. long, green with a membranous edge. 
Pedicel , short 
Ovary, narrow, tapering at either end. 
Tube, £ in. equal in length to the ovary. 
Flowers, bearded, of a greenish yellow. 
Observations. 
This very imperfectly known Iris is in some ways intermediate between I. flavissima and /. Bloudowii, 
that is to say that it has the spathes of the former and the leaves and rhizome of the latter. All our 
knowledge of it is derived from dried specimens, and these seem to show that the spathes are more 
distinctly green than in either of the other two specimens. 
It is probable that the branching stem of Komarov’s specimen (K) is abnormal. 
A better knowledge of this Iris may show that it is hardly sufficiently distinct from one or other 
of the allied species. 
II. The dwarf European Pogoniris. 
Great confusion prevails among this group of Irises, in which are included for the purpose of 
classification all those European bearded species which have unbranched stems. Complications have 
arisen from the fact that all the species are liable to colour-variation in their flowers. The secret of this 
variation has not yet been discovered but there seems little doubt that the yellow and purple colouring- 
matters, which are found in the flowers of this group, are very similar in their chemical constitution. 
It might have been thought that the various shades of colour were due to the various combinations of 
certain Mendelian characters, as has been established for Primulas, Antirrhinums, etc. But, in the case 
of Irises, there seems no doubt that other causes are also at work, for the colour of the flowers produced 
by the same plant varies from year to year. At present it is only possible to state the facts that have 
been already observed and to hope that at some future date the explanation of them will be forthcoming. 
The first point, then, is that plants, which in one year produce flowers of the purest yellow are apt, 
even if they remain untouched in the same position, to have the purity of their flowers spoilt in 
subsequent years by streaks and veins of dull purple. I incline to think that this only occurs when 
the soil is naturally deficient in lime or at any rate when a sufficient proportion of lime is no longer 
within reach of the roots. 
Another point is that an extract of the purple colouring-matter is immediately turned to a bright 
yellow by the addition of a solution of lime, although I have never seen any example of a purple flower 
streaked or mottled with yellow. 
Further, I have more than once been told by residents in the South of France that in some localities 
only one colour form of /. chamaeiris occurs, while in others several forms are found growing together. 
Herbarium material is unfortunately not always a trustworthy guide to colour, but I certainly incline to 
the belief that the same phenomenon occurs in the case of I. pumila, e.g. in Greece the purple form 
is comparatively rare and the yellow common. 
After these preliminary considerations, we can now proceed to the examination of those species which 
have been most commonly confused together, namely /. pumila Linn., /. chamaeiris Bert., I. pseudopumila 
Tineo, and a Balkan species, to which the name I. Reichenbachii Heuffel was probably the first given. 
