| Dykes, The Genus Iris, 1913; Description. Rootstock , a compact rhizome, about an inch in width. Leaves , 4-8 in. by ½--1 in., ensiform, often narrowing suddenly to an incurved point. Stem , about 8- 12 inches in height, usually bearing small clasping leaves of varying length which almost entirely hide it, and usually only one flower. Spathe , 1-2 flowered; valves acute, often scarious in the upper half and sometimes tinged with purple, varying in length from 1½ to 3½ in., the outer slightly keeled. Pedicel , very short. Ovary , obscurely trigonal. Tube , 1½-2 in. long, with purplish stripes in the line of the standards. Falls . The obovate blade is not separated by any constriction from the broad wedge-shaped haft, which is veined with broad veins of brown purple on a whitish ground. The blade is of a deep blue purple, and the hairs of the beard are bluish in front, then white tipped with blue and finally with brownish yellow. Standards , obovate with a short, canaliculate haft, rather lighter in colour than the falls, with faint darker veins. The haft is veined with red brown on a pale ground. Styles , almost colourless with a central blue keel. Crests , triangular, acuminate. Stigma , entire. Filaments , pale violet. Anthers , bluish. Pollen , bluish. Capsule , not seen. Seeds , brown, wrinkled, pyriform.Brotero I.e. remarks that this Iris may have paler or yellowish white flowers. I have seen no specimens of this, but one has recently been obtained from the Moncorvo neighbourhood.Observations.This Iris was first observed by Clusius in the neighbourhood of Coimbra in Portugal about 1565 and described and figured by him as Iris lusitanica seu biflora on pp. 282-3 of his History of Rare Spanish Plants ( 1576). He there states that he found plants in flower in November and gave it the name of biflora on account of its habit of flowering in the autumn as well as in spring. Clusius' figure shows the small acute spathes and the clasping bract-like leaves on the stem and there is a similar figure in the Hortus Eystettensis with the name of portugalica. The biflora in that work is almost certainly the Central European aphylla, and it is probably by a misquotation of these two figures that Linnaeus first began the confusion between I. biflora and I. aphylla, which continued down to, and even after, the publication of Brotero's I. subbiflora which is expressly stated to be the biflora of Clusius but not of Linnaeus.On examining some specimens of this Iris in the Kew herbarium, I found one or two examples from Monsanto in the immediate neighbourhood of Lisbon of an Iris that appeared to differ from the typical I. subbiflora by the much longer spathes and tube. A description was published under the name of I. lisbonensis in the Gard. Chron. for March 5th, 1910, p. 146, and this led to some correspondence with Professor Henriques of Coimbra, who has most kindly sent me recently a series of plants, all collected in the neighbourhood of Lisbon and showing intermediate stages between the typical I. subbiflora and the plant that I described as I. lisbonensis. In the face of this evidence, we can only extend the limits of the dimensions of I. subbiflora and conclude that it is more variable in structure than most species of Iris fortunately appear to be.The cultivation of this Iris is not easy in England. The text that accompanies the figure in the Botanical Magazine (I.e.) states that the flowers were the first produced for fifteen years and my own experience has been similar. The plants need that rest in summer that they obtain on their native limestone rocks, and in cultivation in England this is perhaps best ensured by growing the plants in pots and leaving these to dry and bake in a frame for three or four months in summer. Growth begins again with the autumn rains and it was by this treatment that the flower was obtained from which Plate XXXIII was drawn. The soil in the pots should be made porous by a liberal addition of lime rubble but a fair amount of humus should be added in the shape of old leafsoil, for roots confined to a pot must have more concentrated food than when they can spread to greater distances in the open ground. The extraordinary summer of r 9 r I and the warm spring of r 9 r 2 caused some plants from Coimbra to flower well on an open rockery in my garden.In 1895 Foster received from Messrs Dammann an Iris which was collected in the mountains of Tunis. When it flowered it proved to be very similar in all respects to I. subbiflora. It is not now apparently in cultivation and no specimens from Tunis are to be found in herbarium collections. There exists, however, one from Tangiers and it is not impossible that this Iris will be found to be more widely spread in North Africa than is at present known to be the case.