233 
Iris Hybrids 
In its corm and in its single-valved spathes it is more like a Moraea than an Iris but since its 
segments coalesce to form a tube for a short distance above the ovary, it is considered as an Iris. 
Each stem usually produces several heads of flowers and each spathe contains a number of buds. 
These open in succession and though the individual flowers only last from about u or 12 in the 
morning until 5 or 6 p.m., yet a constant succession of them prolongs the flowering period over a 
fortnight or more. 
The plants succeed best in a heavy limestone soil in a hot and dry position. It is best to lift 
them yearly when the foliage withers and store the corms in sand until September. If they are left in 
the ground, the corms sometimes fail to ripen, or growth may begin too early, in which case the foliage 
suffers during the winter. 
IRIS HYBRIDS 
It is impossible to give an account of all the hybrid Irises that have been raised, and these 
notes will therefore only endeavour to indicate the main lines upon which hybridisation has proceeded, 
and to make suggestions as to the origin of those plants which have long been in cultivation in our 
gardens but whose origin is unknown. Several have been described as species, and have been long 
regarded as such, but the absence in the chief herbarium collections of any indication of their existence 
in uncultivated areas points to their hybrid origin. 
The Iris genus is subdivided into sections by certain morphological characters (see p. 13), and the 
fact that hybrids between the various sections are both extremely rare and have been hitherto invariably 
sterile, seems to show that these characters indicate very real differences in the nature of the plants. 
For instance, no hybrid between a bulbous and a non-bulbous Iris has yet been recorded, notwith- 
standing the fact that it was once declared that /. Kaempferi and I. xiphioides had been combined to 
produce Deleuil’s I. kyerensis. I have seen the so-called I. hyerensis growing in Deleuil’s garden at 
Hydres, and have no doubt that it is merely a form of /. spuria. When it is remembered that the 
latter as well as I. xiphioides and /. Kaempferi produces abundance of seed and is almost certainly 
self-fertile, it is easy to see how a supposed cross may have been produced in a garden where all 
three Irises were growing. Moreover, not only has no bulbous Iris been combined with a non-bulbous 
species, but also it has not been found possible to raise hybrids between the members of the various 
groups that compose the bulbous section. A xiphium will not readily cross with a reticulata , nor a Juno 
with either, and it is even possible to go further and say that the members of the various subdivisions 
of the Juno group are sterile to each other’s pollen (see p. 188). 
On the other hand, with the exception of /. xiphioides all the members of the xiphium group 
seem capable of combination, and the same is probably true of the reticulata species, even though no 
hybrid of /. Danfordiae has yet been made. 
In dealing with the non-bulbous Irises, it will be found that not only do the Apogon Irises form 
a section which is far more widely separated from the other sections, such as the Evansias, the 
Oncocyclus species, and the Pogoniris, than are the latter from one another, but also that the various 
species of Apogon Irises are in many cases as widely separated from one another as are the various 
sections which form the rest of the genus. This, at any rate, is the conclusion to which we are led 
by a review of the facts, which show that, although hybrids between Oncocyclus and Regelia and 
Pogoniris species are relatively common, and those between Pogoniris and Evansia not unknown (see 
p. 98 and Plate XXIV), yet interspecific hybrids of the Apogon section are extremely rare. 
A few instances are recorded, as for example the Japanese hybrids of /. Kaempferi , in the 
production of which it is possible that /. setosa may originally have been used, although the only trace 
it would appear to have left lies in the development of the falls at the expense of the standards. 
Other instances were the combination of I. fulva and /. foliosa, which are however both members of 
the closely allied hexagona group, and of I. Clarkei and I. Douglasiana, which may perhaps be taken 
as an indication that such hybrids are after all far from impossible. 
On the other hand, repeated attempts to fertilise such species as /. psetidacorus and /. foetidissima 
have always remained fruitless, and Foster’s answer to the question whether he had ever succeeded in 
combining an Apogon and a Pogoniris was to show me a large clump of what looked like /. germanica , 
and to tell me that it arose from a single seed obtained by pollinating that species with pollen of 
/. spuria. He added, however, that the only fact that made him think that “anything had happened" 
was that the plant had never flowered, though it had grown vigorously, and although no other Pogoniris 
had remained flowerless for so long when in such good health in the Shelford garden. 
Foster’s notebooks show that he found it possible to combine many of the Oncocyclus species, 
though from what he showed me and told me in his garden towards the end of his life, it was evident 
that these hybrids were no more easy to maintain in good health than the Oncocyclus species 
themselves. 
The idea of combining the shape and large size of the flowers of the Oncocyclus Irises with the 
greater amenability to cultivation of the Regelia group seems to have originated with Foster. He 
