Gardeners’ Chronicle, pp. 753-754 (June 20, 1891)

J. M. Macfarlane

THOUGH I have not yet been able to peruse the extended account given by Professor George Henslow (Gardeners’ Chronicle, p. 618) of his interesting observations on the colour of hybrid Rhododendrons, I may be permitted to make a few remarks on the same subject. Reference will also be made to the flowering period and constitutional vigour of hybrids, as of equal or even greater importance.

The results of Professor Henslow's examination might lead one to consider that colour-production is in some cases fickle and variable, and while, in the present state of our knowledge, this may have to be granted, I feel that it will eventually be possible, in the great majority of cases, to predict the exact colour which the hybrid will show, especially if the colour in each parent be due to the presence of one pigment only. The examples chosen by Professor Henslow, however, are rather complicated by the frequent presence of two pigments—a dissolved red, and a granular yellow, in at least one of the parents. If we compare parents which each develop one pigment, or one of which only is white, i.e., devoid of colour, it may be laid down as a broad general rule, that the hybrid will be intermediate between the two, having regard to the size of the floral parts of each. If a deeply-coloured, small-flowered form, be crossed with a pale or colourless form, bearing large floral parts, the hybrid will appear to lean to the light-flowered parent, since half the amount of colour of the former has been distributed over the larger surface of the hybrid.

To cite examples of well-known Rhododendrons, R. praecox is intermediate between the purple-crimson-flowered parent, R. atrovirens, and the pink-white parent, R. ciliatum; the rich cerise-coloured R. Nobleanum, between the scarlet-flowered R. arboreum and the white-flowered R. caucasicum; R. Greivei, with pale whitish-pink blossoms, between R. ciliatum and the dull pink R. glaucum. The allied hybrid, Bryanthus erectus, is neatly intermediate between the pale pink R. Chamaecistus and the rose-pink Menziesia empetriformis var. Drummondii (not M. coerulea, as is generally stated); so also is Erica Watsoni, between E. tetralix and E. ciliaris.

From every order in which hybrids have appeared, illustrations might be given, as anyone can readily gather from the perusal of your pages, or of Focke's Pflanzen Mitchlinge, Seldom is it better exemplified than when tints of yellow, or yellow and white, are crossed, if the colour be due wholly to yellow chromoplasts in the cell-protoplasm. During the last week or two I have gathered a large number of hybrid Oxlips alongside their parents, the Primrose and Cowslip. One has merely to pluck a blossom of each, and place these in rows to see how closely the hybrid is the mean between the parents. A set of hybrid Hedychiums, raised at the Royal Botanic Garden, gives additional confirmation. Thus, the orange-coloured H. Gardnerianum, crossed by the white H. coronarium, gave H. Sadlerianum with a tint exactly intermediate; this, recrossed by the latter parent, gave H. Lindsayi, the blossoms of which are of a pale maize-white in bud, becoming white in blossom. Crosses between H. Gardnerianum and H. angustifolium also verified what one might have predicted.

*The late Professor Henslow's paper in the Cambridge Philosophical Transactions for 1833, leaves nothing to be desired further in the present connection.

Hybrids of Dianthus, Geum, Saxifraga, Gesnera, Gloxinia, Digitalis,* Orchids, Montbretia, &c., might all be noted for their floral parts, as might pitchers of Sarracenia and Nepenthes for their vegetative leaf-parts.

But knotty points often arise in the study of hybrids, whose parents—one or both—have two or more pigments in their cells. A comparatively simple case of such is Masdevallia Chelsoni, in which, as in both parents, we have a dense background of yellow cells, while from the surface arise layer sacs filled with a purple pigment. But when yellow, red, or blue occur in the same or neighbouring cells of a tissue, the hybrid product may take after one or other of the parents in an apparently arbitrary way. Even in such cases, however, I think that the appearances can be explained in a manner that clears away many of the apparent difficulties which beset one when a minute study of hybrids is first entered upon. The full explanation—which cannot be given in a short paper like this—I hope to publish shortly; suffice it to say, that I regard many of the unequal blendings in hybrid colour and structure to be due to incompatibility in chemical or molecular union, and the resulting predominance of that colour which is the more stable or readily evolved of the two.

Study of the period of flowering of plants is still in its infancy, but I venture to think that a comparison of the flowering period of hybrids with that of their parents will yield in the end the most valuable phenological harvest, for we deal then with a connected series of three. During the last twenty years a record of the flowering periods of a few has been kept at the Edinburgh Botanic Garden, and since 1880 a similar record, for fully 800 plants in the rock garden (including several hybrids and their parents) has been kept; while in the Transactions of the Edinburgh Botanical Society, vol. xvii., part 2, Mr. Lindsay has recorded the time of flowering of 1408 species during 1887. These, supplemented by limited observations of my own, all point distinctly to a flowering period in hybrids closely intermediate between the parents. To give only one or two examples verified during the present season, Rhododendron atrovirens first opened on January 21, and was in full bloom by the 27th; R. praecox opened on March 1, but was immediately after destroyed by frost; judging, however, from previous years, it would have blossomed well by March 10; while R. ciliatum opened on April 25, and was in full blossom by May 4.

I was rather puzzled for a time to account for the early flowering of R. Nobleanum, which from an average of twenty years opens on March 1, while R. caucasicum opens from April 25 to 30. I could scarcely imagine that the other parent, R. arboreum, would start, even in a milder climate than our own, in the early part of January; but on going over references in Hooker's Himalayan Journals, it appeared that in 1848 he saw it flowering on December 7, for he says (vol. i., p. 274):—"The descent from the Talloong ridge was very steep, and in some places almost precipitous, first through dense woods of Silver Fir, with Rhododendron Falconeri and Hodgsoni, then through Abies Brunoniana with Yew (now covered with red berries), to the region of Magnolias and Rhododendron arboreum and barbatam. One bush of the former was in flower, making a gorgeous show." And he mentions it again and again, up till the end of May. It must be difficult to compare a Himalayan winter climate at 8000 to 10,000 feet with that of our country, but correspondents of this paper might aid towards a satisfactory solution. Great accuracy and copious data have still to be reached, and the inquiry can afford the aid of many workers, while for gardeners there is the possibility of obtaining many crosses which might adorn our cool and hot-houses during winter.

On the constitutional vigour of hybrids I should greatly desire information, rather than attempt to give it, but the behaviour of Montbretia Pottsii, Tritonia aurea, and M. crocosmaeflora, in the Edinburgh Garden during the past winter seems suggestive. The corms of the first appear scarcely to have been injured. Those of the hybrid have been largely killed off—at least, to the extent of 60 per cent.; while Tritonia—never hardy in exposed ground—has survived only where it is planted against, and can creep along, the outer side of a hot-house wall.