DISCOVERY 



13 



this case too, as in the Bizzaria Orange, the graft did 

 not succeed, but later a branch grew out from near the 

 insertion of the graft, which in its foliage and flowers 

 was intermediate between the stock and the graft. 

 By many it was considered that it might perhaps have 

 been a seed-hybrid which the nurserj-man had en- 

 deavoured to graft upon the Laburnum, but no such 

 hybrid is known, nor has it been possible to obtain this 

 seed-hybrid of these two plants by the ordinarv' methods 

 of crossing. Moreover, Cytisus Adami exhibits what is 

 one of the most striking phenomena of graft -hybrids, 

 namely the tendency to revert back to the constituent 

 plants, which may be regarded as its parents. Thus three 

 different tj'pes of foliage and flowers may after a time 

 be observed on most of the specimens of Cytisus Adami. 

 The flowers are represented by long grape-like inflores- 

 cences with flowers of yellow colour, shorter clusters of 

 somewhat purplish flowers, while on the branches 

 exhibiting the foliage of the Purple Broom small purple 

 flowers are bom in pairs. Such a segregation of char- 

 acters on one plant is very rare in the case of seed- 

 hybrids, but characteristic of graft-hybrids. A de- 

 tailed microscopic investigation of the plant by J. M. 

 Macfarlane also revealed the fact that the hybrid 

 portion seemed to be "wrapped round, so to speak, by 

 a skin " of the Purple Broom, which feature we may 

 fittingly compare with the " orange shell and lemon 

 pulp " of some of the fruits of the Bizzaria Orange 

 mentioned above. Indeed, this pecuUar form of 

 segregation of characters in definite layers, as well as 

 the tendency" to reversion, seem to be characteristic 

 of many graft-hybrids. 



Earl}' in the present centuiy Dr. Hans Winkler set 

 out definitely to produce other graft-hybrids, selecting 

 for his experiments the genus Solanum, which seemed 

 to him for certain reasons a very suitable subject. 

 Grafting yoimg shoots of the Tomato on the stem of 

 the Nightshade and vice versa, he fomid that he readily 

 got these two related species to imite. He then cut 

 transversely through the region of the graft, thus 

 exposing the living tissues of both plants. The wound 

 so inflicted was rapidly covered by a healing tissue 

 known as callus, upon which numerous new buds were 

 formed. Some of these grew out into branches of 

 Tomato or of Nightshade according to their point of 

 origin, while in some cases where the two tissues were 

 close together, a branch might be formed half of which 

 was Tomato while the other half bore leaves and 

 flowers of the Nightshade. Such growths he caUed 

 ChimcBras, after the fabulous monster of antiquity, 

 part lion and part dragon. Very occasionally, how- 

 ever, shoots were produced in which the characters of 

 the two constituent plants were blended, and these 

 he regarded as graft-hybrids. They also showed a 

 tendency to revert to their constituent parents, and an 



important feature was the fact that, though the shape 



of their leaves was intermediate, their outer covering 



! was purely that of the Tomato or of the Nightshade 



according as to whether the latter or the lonner was 



FLOWERING BK.\>rcKES OF THE GR.\FT-HYBRID [CYTISUS 



ADAMI) A-XD OF ITS PARENTS. 

 A. The Purple Broom [Cytisus purpureas) with smailleaves and short 

 purple flowers. B. The common Yellow laburnum. C. The graft- 

 hybrid, with leaves of intermediate size, and flowers in which the outer 

 layer of the petals is purple, while the inner tissues are yellow. This gives 

 the flower an intermediate colour. 



the stock. It was, therefore, argued that in these 

 experimental hybrids there had been no true vegetative 

 fusion of ceUs comparable to the fertilisation of a seed, 

 but that they represented the core of one plant sur- 

 rounded by the skin of the other, very much as a finger 

 may be covered by a glove. Further exaoiination 

 proved this to be correct. The internal cells partook 

 of the nature of the stock, while the skin, or epidermis, 

 was that of the inserted graft. Buds of this kind 

 might readily arise from the woxmd tissue in which the 

 cells of the two plants must be variously arranged, and 

 in the production of a bud the inner cells of the stock 

 might easily be covered by a cap of cells belonging to 

 the scion. It has been suggested that, as these two 

 types of cells largely preserve their own characteristics, 

 these so-caUed graft-hybrids are reaUy more of the 

 nature of the chimaera mentioned above, but that, 

 instead of the tissues of the two plants being side by 



