78 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 
among themselves than between different species of these genera. In other 
words, they would form one genus if the perianth alone were the basis of 
classification. But this is not the feature relied upon, but the nwmber 
of pollen masses—i.c. a single feature, and therefore, so far, an artificial 
character—just as Linneeus would unite the Ash tree, Veronica, a Grass, 
and the Duckweed because they have two stamens. Turning to the 
Genera Plantarum” we find that Cattleya has four pollinia. Then 
follow three genera with eight in two series; those of the upper series 
very often much smaller than the lower. Then comes Lelia, also with 
eight pollinia in slightly unequal series.* 
Here, then, is obviously a closely graduated series based on a single 
character, and a purely artificial one. The classification is therefore 
not strictly natural, though the series of so-called ‘“‘genera’’ may be. 
Consequently, though we may call the cross-products ‘“‘ bi-geners,”’ it is 
only so from Bentham and Hooker’s classificatory point of view. 
These observations apply to Hpidendrum and Sophronitis as well. 
Similarly with the so-called bi-gener between Lapageria and Philesia. 
Those genera stand together in the “ Gen. Pl.,”” being No. 10 and No. 11 
in Liliacee: both are mono-specific, and both live in Chili. The 
distinguishing features are recorded as being, in Lapageria, ‘‘ Leaves 3-5- 
nerved,” and ‘“‘ the segments of the perianth sub-equal.’’ In Philesia, 
“ Leaves 1-nerved,” and “ the exterior segments of the perianth are much 
shorter than the interior.” But much ereater differences in nervature 
occur in species of Plantago ; and also between the outer and inner 
whorls of the perianth of species of Iris. 
Consequently, to be true to principles of natural classification, it 
would seem that the above two genera should be regarded simply as two 
species of the same genus. 
Once more, in the “Gen. Pl.” Gloxinia (gen. 6) and Achimenes 
(gen. 7) belong to the sub-tribe Gloxinee ; while the genera Gesnera (gen. 
18) and Sinningia (gen. 19) are in the sub-tribe Hugesneree. There is, 
however, no special feature to separate them—a fact which Dean Herbert 
perceived and discussed at length some seventy years ago. 
He mentions also that Simningia and Gloxinia produced fertile 
hybrids. 
Gloxinera—i.e. Gloxima x Gesnera—was raised in 1894.+ 
Selenipedium and Cypripedium are genera which Bentham and 
Hooker admit to be scarcely distinguishable except by the ovary being 
one-celled in the former, from a want of cohesion of the placentas, 
and its habitat, viz. South America, the nearest home of the latter 
being Mexico. 
Though it has been found difficult to cross these, yet Mr. Swinburne, 
of Winchcombe, near Cheltenham,$ raised small plants from S. Schlin 
x C. Spicerianum, male parent; also between S. Donvinianum x C. Cham- 
berlaint. The plants were raised in 1896, but have not yet flowered. 
* See observations by Mr. C. C. Hurst relative to this matter, Jowrn. R.H.S. 
1898, p. 475. 
+ On Hybridisation amongst Vegetables. 
{ For successful Bi-geners and Hybrids in Gesneriacee, see Burbidge’s Prop. and 
Improv. of Cult. Plants, p. 331 seqq. 
§ Gard. Chron. Oct. 10, 1896, p. 435. 
