| Dykes in Genus Iris p.149, 1913.Description. Rootstock , a compact rhizome with crowded shoots. Leaves , ensiform, slightly glaucous, 3-6 in. by 1/3-1/2 in. at flowering time. Stem , 1-10 in., bare in the upper part, with one or two reduced leaves attached below the centre; one headed. (In very rare cases a lateral I-flowered branch may develop.) Spathes , 1-2-flowered; valves green or scarious in the upper third, slightly inflated; one valve may be very slightly keeled, 1 1/2--2 in. long. Pedicel , very short. Ovary , obscurely trigonal. Tube, about 1 in. long. Falls . The obovate blade tapers gradually into the wedge-shaped haft, which bears diffuse veins on a light ground. The colour is either blue or red purple, yellow or white, with a contrasting beard, either yellow, or white tipped with yellow or even bluish. Standards , rounded oblong with a canaliculate haft, slightly shorter than the falls, but usually a little wider. Styles, paler than the segments except along the deeper-coloured keel. Crests, small, trigonal. Stigma , entire. Filaments, white or tinged with purple. Anthers, cream or tinged with purple. Pollen, cream. Capsule , 1 1/2-2 1/2 in. long, oblong or oval in outline, the valves remaining attached together at the extreme tip after they have split open lower down. The section is rounded trigonal with a raised line running down each face. Seeds , red brown, pyriform, with wrinkled skins.Observations.This is the common dwarf Iris of the South of France and North Italy and is also the ancestor of most of the garden forms that are so commonly known as I pumila. It is distinguished from the true pumila by the obvious stem, by the shorter tube and by the more inflated and less membranous spathes'. In cultivation it has the advantage that the new leaves attain some length before winter, but owing to this early habit of growth, imported plants are often less able to resist our winter than I. pumila.Like the latter, I. chamaeiris is very variable in colour and in dimensions and has received many names. Thus Lamarck's I. lutescens is clearly a yellow flowered example such as are common in the south of France. Henon's I. olbiensis, with its various colour forms, differs in no essential from the more westerly specimens and I. virescens D.C. from Sion in the Rhone Valley is probably only an introduction from the south, for it grows on a hill, crowned with ruins, on which there also grows an Opuntia, that is likewise probably an escape from a formerly cultivated area. ln cultivation the Sion plant is indistinguishable from examples of Henon's I. olbiensis from the neighbourhood of Hyeres. In this connection it must be remembered that there was in ancient times a town of Olbia, near the actual site of Hyeres, and that this must not be confused with another Olbia on the coast of the Black Sea, a little to the East of Odessa. The confusion of these two names was probably the origin of the term "Crimean," applied by trade catalogues to dwarf-bearded Irises. Any specimens from the Crimea are always found to be I. pumila and not I. lutescens. It does not seem possible to separate a variety italica, based on Parlatore's I. ita!ica (v. supra). Baker's statement (Hdk. p. 27) that it differs by having dark violet flowers does not agree with Parlatore's original description of a variety C, "flore flavo." The height of Italian specimens is as variable as those from France already mentioned on p. ·I4r. Cf. Groves' (E) specimens: that of 1861 has a 3 in. stem while that of r 878 is no less than 10 in. high. The cultivation is the same as that of I. pumila except that a position which gives some protection from frost and cold winds is desirable, especially for imported plants.