HORTICULTURE. 951 



that hardness, density, and elasticit)' of wood play a secondary role 

 in the success and duration of grafts. l)ut it is not the same with con- 

 duction. When the ditferences of sap conductions are too great the 

 grafts will not succeed. As an example may be mentioned the grafts 

 between the lilac and ash, cherr}" and almond, cotoneaster and chest- 

 nut, etc., which grow the first year, then die without fructifying. The 

 duration of the graft is then very variable and depends for its value 

 on the differences in conduction between the scion and the stock. Thus 

 the pear grafted on the quince endures for a shorter period than the 

 pear grafted on the pear seedling. When the differences of conduc- 

 tion are too great between two plants, the mixed graft is sometimes 

 used successfully where the ordinary graft fails. By the use of the 

 mixed graft the author was able to unite Vernonia prmalta and Xan- 

 tJmim maGTocarimm., which failed by the ordinary process of grafting. 



The author has succeeded in grafting plants whose cell contents 

 presented very marked differences. Thus the grafts of Chicoriacese 

 and of divers Euphorbiacete, etc., show that plants with different latex 

 contents succeed, although it has been previously held that plants with 

 a milky juice could not be grafted. 



A number of grafts were made to determine what influence reserve 

 material in plants might have on grafting. The easy grafts on roots of 

 the carrot and parsnip show that the presence of reserve material is no 

 obstacle to success. The graft of the tomato on the potato, annual 

 sunflower on the Jerusalem artichoke, etc., show that the formation of 

 tubers on the stock takes place even when the scion is incapa])le of 

 producing tubers itself. In grafting in September a young cabbage on 

 the purple-topped turnip, which would have commenced to form its 

 tuber in Octo])er. the turnip tuber formed only the month of April 

 following, when the scion became plethoric. It is then the scion which 

 by its mode of nutrition commands the function of reserve material in 

 the stock. 



The inverse graft of plants suscepti))le of forming tubers on a plant 

 which does not yield tu])ers may be realized. Thus the author suc- 

 ceeded in grafting TTdhinfliiiK ht'tijJoni.^, species with enlarged rhizome, 

 on IT. a/unn/,s, an annual species not forming tubers. The scion grown 

 entirely above the soil was unable to form tubers. The reserves which 

 were formed passed into another form in the stock, which took a devel- 

 opment altogether a])normal and l)ecame yevy ligneous. The potato 

 grafted on eggplant and tomatoes has been observed to form aerial 

 tubers and thus store up its reserve material. 



Analogy in habitat seems to be a more or less important factor. 

 Thus Pltlox decusHata^ which grows in humid soils, has not been success- 

 fully grafted by the author with P. suhidata, which grows on dry soils; 

 though parsley, which prefers a dry soil, succeeds when grafted with 

 Sison ammoiimm, which prefers a humid soil. In the case of trees, 



