140 BOTANY 



are being illuminated by chemical investigations into the nature of the proximate sub- 

 stances which are becoming recognised as the agents responsible for some at least of the 

 differences which separate nearly related forms. Mendelian analysis is proving a more 

 complicated affair than many people at first anticipated, but the recognition that a 

 character or a group pf characters may require the interaction of two or more " factors " 

 is a considerable step in advance. The researches on enzymes (especially the oxydases 

 and peroxydases, specially investigated by Chodat and his school) in the determination 

 of colour development and colour patterns are especially suggestive in this connection. 

 Whiteness, for example, is a character which may be due to a variety of causes, in other 

 words colour-absence is a term which includes more than one category. The develop- 

 ment of a particular hue in the flowers of a considerable number of plants is recognised 

 as depending on the simultaneous co-operation of several " factors " all of which must 

 be present in order that the colour in question may be forthcoming, and in the absence 

 of one or more of them the flower may be an albino or white. Oxydases and peroxydases 

 above referred to, are known to play an important part in many of these colour produc- 

 tions. They clearly represent one of the factorial members of the complex, the others 

 being made up of chromogens, substances ultimately derived, in all probability, from 

 glucosides. But the matter is further complicated by the circumstance that whiteness 

 may also result from the operation of a chemical " inhibitory " substance which arrests 

 the colour production that ought to ensue from the interaction of enzyme and chromo- 

 gen. The recognition of the influence of this factor, and especially its experimental 

 investigation by Keeble and Armstrong, has served to clear away much that previously 

 obscured the understanding of the occurrence and distribution of whiteness in flowers. 

 See M. Wheldale, "Plant oxydases and the chemical inter-relationships of colour varie- 

 ties," Progressus rei botanicae, Bd. iii, 1910. Also Keeble & Armstrong, papers in the Proc. 

 Roy. Soc., vol. 85, 1912.) 



Graft-hybrids. Much light has, within the last few years, been thrown on the nature 

 of the so-called graft-hybrids, especially by the work of E. Baur and H. Winkler. The 

 general result of their investigations indicates that graft-hybrids consist of growths in 

 which the cellular tissues of both stock and scion take a share, but the tissues, genetical- 

 ly derived from each of these two sources, however much they may be intermingled, 

 remain organically distinct as far as the retention of their respective individual or 

 specific characters is concerned. Sometimes the entire half of the " hybrid " shoot 

 resembles one parent, while the other half obviously belongs to the other or one (stock 

 or scion) may be arranged so as to occupy a smaller sector of the shoot regarded in trans- 

 verse section. These " sectorial chimaeras " are more easy to recognise than those 

 belonging to the " periclinal " category. Periclinal chimaeras were first correctly 

 explained by Baur, and arise when the tissues of stock and scion are so arranged that the 

 one more or less covers the other in a concentric fashion. Although the apparent inter- 

 mingling of characters especially in the periclinal chimaeras is sometimes intimate, it is 

 to be understood that there is no fusion of the protoplasm of the composite structure of 

 which the chimaera is made up. What has happened is the production of a sort of mo- 

 saic or interweaving of the tissues belonging to the respective constituents, each of which 

 preserves its own characters intact, or only exhibits such modification as the correlative 

 disturbance of nutritional conditions can effect. Thus it may happen, as in the well- 

 known example of Cytisus Ada-mi, that the tissues of one of the two constituents may lag 

 in development. The resulting shoot then resumes the habit and colour of the original 

 constituent of which it is then composed, and if the separation has really been complete 

 no reversion either to the hybrid form or to the character of the other constituent can 

 or does take place. 



Winkler, who has successfully produced a number of " graft-hybrids," clearly states 

 that the two co-operating tissue systems of the respective constituents continue to re- 

 tain their own specific characters. The composite organism is a mosaic of the cells and 

 tissues of the grafted individuals only. In no instance (save in one doubtful one) does 

 the union extend to a commingling of those finer elements which are ultimately responsi- 



