Gardeners' Chronicle 11(263): 46 (January 9, 1892)

Subterranean Grafts
Lucien Daniel

The Journal de la Société Nationale d'Horticulture de France for September contains a review of M. Daniel's paper "Sur la Greffe des Parties Souterraines des Plantes." Grafting, says the author, is one of the most important horticultural operations, and, except upon the tubers of Dahlias, is not generally performed upon underground stems or roots. To ensure success in grafting, it is usually considered necessary that the two portions should be of plants of the same genus, or at least of the same family. M. Daniel, wishing to prove if this applied to subterranean shoots, as well as to stems and branches, set grafts of various plants on the roots or tubers of species belonging to different families. In most cases this plan was not successful; however, a graft of Saponaria, a Caryophyllaceous plant, with a free or superior ovary struck on the root of Oenothera biennis, which belongs to a family of Onagrads in which the ovary is adherent or inferior. Hitherto there had been no ground for supposing that this could be done. In the experiments which failed, the grafts were affected in different ways. Under some circumstances, says M. Daniel, the graft at first throve at the expense of the stock until it developed adventitious roots, which enabled the plant to obtain nourishment for itself; it then grew as does a cutting, and if it lived, the stock was of no further use. This was the case when Cabbage and Burdock were grafted on Ranunculus tubers, Parsley on Valerian, &c. In other cases the graft at first drew its nourishment from the stock, but no adventitious roots were developed, and as the stock died after a few months, the graft perished from want of nourishment. This was the cause of the failure of the trials of Lily of the Valley and Primroses on Crocus bulbs. One fact worth notice is, that grafts with a growing bud set on herbaceous stocks, such as Spruce and Cedar on Potato tubers, thrived at the expense of those stocks until they became woody. Grafts made between plants of the same family were successful, but often the graft sent out adventitious roots, and became independent. This was the case when Lettuce and Chicory were put in on Dandelion, Cabbage on Wallflowers, &c. Though the union might be perfect, M. Daniel found that the graft was not sufficiently nourished by the stock, since the suppression of adventitious roots caused shrivelling, and finally death. On the contrary, the graft grew well when Salsafy was put on Scorzonera, Barkhansia on Dandelion and Hypochaeris, although with few or no adventitious roots. This may be explained by the fact that in the first case the membranes and the graft could not be traversed by the nutritive matters with which the stock could furnish it, while in the second case they were quite permeable by these substances. M. Daniel believes the following general conclusion to be correct:—that the failure of many grafts is easily to be explained by insufficient nourishment, without its being necessary to take into consideration the more or less problematical relations between genera or between species of the same genus.

There was another unexpected result of M. Daniel's trials. The chief thing in grafting is to connect the graft and stock so as to bring the growing layer or cambium of each in connection with that of the other; a layer which in dicotyledons is situated between the wood and the liber, or fibrous bark. It is, in fact, the young and active tissue which alone is capable of effecting fusion between the two. Now the experimenter finds that this is not absolutely necessary. "In a certain number of grafts," he writes, " I have cut only the bark and liber without going down to the cambium. The union then took place, nevertheless, by the aid of the cellular tissue, which was existent in a state of meristem (that is to say, cellular tissue essentially young and living), as well as the medullary parenchyma of the stock. It is, therefore, sufficient if living tissues are brought into contact—either in a state of meristem, as in the growing layers, or in a tissue which may pass into that condition as the liberian or medullary parenchyma.

Daniel bibliography