Journ. of Gen. 6(2):75-80. December, 1916
(With Plate I.)

SINCE the work of Baur and Winkler the extraordinary interest of plant-"chimaeras" has been universally recognized. In Winkler's examples various combinations of Solanum lycopersicum (Tomato) and S. nigrum were made by grafting, and "graft-hybrids" were thus produced, having one or more cell-layers of the one species superposed over a core of the other species. From his studies of variegated plants Baur successfully interpreted these cases. There are many sorts of variegation, but in the commonest a "skin," one or more cell-layers thick, devoid of green chloroplasts and consequently white, overlies a core of green. Such plants have exclusively albino offspring, since the germ-cells are derived from the sub-epidermal layer which is albino. In the converse, but much rarer form of variegation (Pelargonium, Holly, Coprosma) a green skin overlies a white core, and from such plants (though experimental proof has only been obtained in Pelargonium) the seedlings will be all green.

1 It must be borne in mind that the distribution of fertility in C. adami has not yet been dealt with. Usually, at least, the purpureus flowers have bad pollen and the adami flowers good pollen.

Baur extended his interpretation to Cytisus adami1, which has a skin of C. purpureus over a core of C. laburnum, and to the various Bronvaux Medlars, which have one or more cell-layers of Crataegus mespilus over a core of C. monogyna. On the last subject a paper giving many details has lately appeared by J. Meyer (Zts. f. Abstammungs- u. Vererbungsl. 1915, xiii. p. 193).

The object of this note is to point out the fact that collateral evidence shows some unsuspected plants to be in reality of this nature, namely periclinal chimaeras, having an outer layer or cortex distinct in genetic composition from the inner core. My attention was called to the subject by reading a report of an address of Mr. C. E. Pearson to the Horticultural Club (June 30, 1914), in which he stated that some Bouvardias, and the class of Pelargoniums known as "Regals" did not come true from root-cuttings. Now the buds formed on true roots arise by endogenous growth from the central tissues which, pushing through the outer cortex, grow into plants exhibiting the characters proper to the "core."

1 Plants entirely of this heterozygous yellow also exist.

Whenever therefore plants grown from root-cuttings differ from those grown from stem-cuttings, we may infer that the plant is a periclinal chimaera. As cases accumulate it will be interesting to learn the classes of distinctions by which the cortex may differ from the central tissues. Presumably they will include the various kinds of distinctions for which genetic factors are responsible. Apart from the graft-hybrids, in the only examples hitherto known the distinctions have been in the presence and nature of the chloroplasts. The cortex may be white and the core green or vice versa, and in Pelargonium we also know a form having the core green and the cortex in the heterozygous yellow condition1 which Baur first identified in Antirrhinum. From such a plant both green and yellow seedlings have been raised here, and doubtless total albinos are also produced.

2 A writer in The Garden, 1916, p. 122,
makes the same statement as to this variety.

In Bouvardia some doubles are said to give singles from their root-cuttings, and in several cases differences in colour are reported. In answer to my questions Mr. Pearson kindly wrote that the doubles said to have given singles are Alfred Neuner2, President Garfield and Hogarth. In our own experiments with these three varieties however this result did not occur, for the products of both root- and stem-cuttings were identical. Nevertheless I see no reason for doubting the correctness of Mr. Pearson's record, which is moreover supported by a statement made in general term's by J. C. C. in The Garden, 1889, 1. p. 347 that many, if not all of the plants raised from root-cuttings of double Bouvardias will produce single flowers. The nature of the discrepancy between the two results will be considered later.

1 Cp. The Garden, 1889, i. p. 380.

As regards distinctions in colour between root-cuttings and stem-cuttings J. C. C. (loc. cit.) says that he has "never known the ordinary cuttings to sport," but that he has had "many instances where, when plants have been raised from root-cuttings, all those sorts with red or pinkish-coloured blooms have reverted to the white-flowered forms1." In our own experiments one case only has given a positive result. Bridesmaid, a double, having the outer surface of corolla pink and the inner faces of the petals pinkish white (see coloured plate), has given uniformly from its root-cuttings plants agreeing in every respect with the well-known Hogarth, a double carnation-scarlet. Over sixty such plants have been raised from the roots of Bridesmaid. Negative results occurred, as already stated, with President Garfield (double pink), Alfred Neuner (double white) and Hogarth itself. The evidence is so far clear that Bridesmaid is a periclinal chimaera having a skin or cortex of pinkish-white over a core of Hogarth.

The "Regal" Pelargoniums are a class of plants having flowers partially double, with a curious crumpling of the petals. We attempted to raise root-cuttings of several but failed, doubtless through a mistake in treatment. That such plants should be periclinal chimaeras is exactly what one might expect. The crumpling or buckling of the petals is presumably a consequence of an unconformity of growth between the outer and inner constituents. Similar buckling of the leaves is a striking feature of Winkler's Solanum graft-hybrid which he called Gaertnerianum, and I have little doubt that from root-cuttings of these Regals the ordinary flat-petalled forms will be produced, as Mr. Pearson stated.

2 Pomologie Française, ii. 1846, s.v. Epine Vinette Ordinaire.
The seedless fruit is figured by Nicholson, The Garden, 1889, i. p. 264,

I was disposed to apply the same interpretation to a curious case quoted by Darwin (An. and Plts. i. 384). There is a form of Berberis vulgaris with seedless fruit, which can be propagated by cuttings or layers, but its "suckers always revert to the common form which produces fruit containing seeds." Darwin adds that his father repeatedly tried this experiment, and always with the same result. But Duhamel (Traite des Arbres Fruitiers, 1768, I. p. 151) states that the seedless condition is acquired with age. The seedless variety is found wild in Northern France and on being transplanted into gardens develops seeds, though subsequently the same plants return to the seedless condition. Poiteau has a passage to the same effect. I have only lately obtained the plant. Formerly it was well known under the name "Maiden Barberry," and was grown for preserves, but now it has almost disappeared from cultivation. I was fortunate in finding it with Messrs Smith of Newry, who keep it under the name Berberis vulgaris, var. asperma. It will now be further investigated. Seeing that the seedlessness is retained when the plant is propagated by "cuttings or layers" the suggestion that return of fertility is merely a matter of rejuvenescence seems scarcely probable, though pending examination one cannot assume that the "suckers" arise from roots.

Apart however from this doubtful case the distinction between cortex and core-to use non-committal terms-may relate to chlorophyll-production, doubleness or singleness of flowers, colour of flowers, size of parts (as indicated in the Regal Pelargonium). With further observation of the produce of root-cuttings the list will no doubt soon be greatly extended. There seems a priori, to be no reason why the distinction may not affect any factorial character. For example, both Bridesmaid and Hogarth Bouvardias are short-styled, but there would be nothing surprising if one or other had been long-styled.

Having in view the fact, which can scarcely be doubtful, that these various chimaeras arose as seedlings, their peculiar constitution must be recognized as having been produced by somatic segregation. How the characters are distributed among the embryonic layers of the plants is however as yet uncertain. All that can be declared is that, in those plants which give a distinct form from their root-cuttings, the plerome is factorially distinct from the outer layers; but the constitution of the periblem as distinct from that of the dermatogen cannot be yet decided, though perhaps on this point further genetic analysis might throw light. In Bouvardia I have found seed-production very difficult to obtain, and hope for some more amenable subject for such investigations. It would besides be interesting to know whether somatic segregation occurs with any special frequency between the three embryonic layers-whether, in other words, these layers have any prerogative individuality in that respect. The fact that the plerome has the power, when it forms adventitious buds, of providing out of itself all the parts normally arising from periblem and from dermatogen, is manifestly beside the question here suggested for consideration.

1 Gard. Chron. 1867, p. 1000, and 1867, p. 74; see also Rev. Hort. 1866, p. 429.

Proof that a plant is a periclinal chimaera may of course be obtained from adventitious buds arising in internodes of the stem, as well as from those formed on roots, for such adventitious buds are also extensions from the plerome. At the time when plants with variegated leaves were much cultivated it was often observed that they (e.g. Zonal Pelargoniums) gave only green plants from their roots. The observation presumably applies generally if not solely to plants in which the variegation affected the skin periclinally. Moreover Hally1 records that when the variety Mrs Pollock was propagated in such a way as to cause adventitious buds to arise from the internodes the resulting shoots were green.

2 Gard. Chron. 1867, p. 952. Cp. ibid. 1872, p. 1321, Pyrus japonica said to produce flowers borne directly on "roots " which had been cut for propagation by cuttings. Probably the " roots" were underground stems in both cases.

Meehan (signing T. M.), contributing to the same discussion, notes that though several variegated plants produced green shoots from root-cuttings, those raised from Pyrus japonica, variegated, came true. Thus far I have failed to obtain this plant and I have not met with any one who remembers it. Probably its variegation was not of the periclinal kind, or perhaps the pieces used for propagation were not true roots.

Pending further evidence it is natural to interpret these cases as examples of heterozygous plants in which there has been somatic segregation of a factor at an early stage. On a general survey of the phenomena of chimaera-production and of bud-sports, the indications suggest that such a segregation may occur at many and perhaps at any cell-divisions by which the parts of the embryo are constituted, or the organs of the plant differentiated. It may well be that segregation is most commonly relegated to the divisions in the germ-cell cycle, but I am unwilling to regard segregation postponed to the reduction division as a process distinct in kind from those somatic segregations of which bud-sports are the visible manifestation.

At the beginning of this paper it was remarked that certain double varieties of Bouvardia which had been reported as giving singles from their root-cuttings did not in our experiments behave in that way. But taking the periclinal variegated plants as an analogy, just as we are familiar with the formation of solid shoots composed entirely of tissue having the characters which in the chimaera are confined to the cortex, so may the double Bouvardias have formed shoots entirely double, without any core of singleness. Wholly albino shoots for example are common on variegated plants with white cortices, being doubtless formed by periclinal divisions in the subepidermal layer. Still more frequently patches may be found in the leaves, in which the white layer is more than one cell thick. In some variegated plants, e.g. Euonymus japonicus, var. latifolia, irregularities of this kind occur in almost every stem, and totally white shoots are often produced.

Similar proliferation in the layer lacking the factor for singleness might easily lead to the formation of a shoot altogether devoid of singleness, and in all probability such a shoot would be then quite indistinguishable from the form previously recognized as that of the variety, though no longer a periclinal chimaera.

[Note. Nov. 1916. Since this paper was written a fresh crop of Bouvardia root-cuttings has been raised. They are just beginning to flower. Those from Bridesmaid are Hogarth as before. Whether the point is significant or not I cannot say, but it may be worth mentioning that whereas in Bridesmaid the inner surfaces of the petals are nearly white, the outer surfaces being pink, in Hogarth the outer surfaces are whitish and the inner scarlet.

The Barberry mentioned in the text proved to be the ordinary form with seeds. I should be grateful to any one who can supply the genuine "asperma" variety. W. B.]