568 BACTERIAL DISEASES OF PLANTS 



time destitute of any killing excretions, so that it shall be able 

 to live on good terms with the host-cell, and you have my con- 

 ception of a tumor-parasite, and in the chain of phenomena set 

 up b}^ its presence (H-ionization, disturbed respiration, etc.) 

 the origin of all tumors which have the power of self-centered 

 continuous growth. 



This physico-chemical hypothesis, which does not always 

 require the presence of a parasite in animals any more than in 

 plants, since we may suppose that there are in. animals slow- 

 growing benign tumors quite like the non-parasitic growths 

 under stomata and lenticels, serves also to explain the develop- 

 ment of fetal fragments in tumors and the formation of teratomas 

 in the absence of tumors, thus uniting and correlating all types of 

 abnormal growth. I have shown for crown, gall that when the 

 tumor develops under or in a dormant bud, as for example in 

 the axil of a leaf, it always contains embryo-fragments and may 

 be full of perishable leafy shoots (Figs. 333, 334, 338, 342). 

 I have shown, furthermore, that preformed dormant buds are 

 not necessary for the production of these embryomas, but that 

 they occur whenever the crown gall organism is inoculated into 

 a cambium, that is into a tissue containing totipotent or pluri- 

 potent cells (Figs. 335, 336, SS9A, 340), that a cambium which 

 normally produces only bark-cells may under the crown-gall 

 stimulus produce also totipotent cells, and that the nature of 

 these embryonic inclusions, i.e., the kind of organs included in 

 the tumor, such as roots, leaves, stems, flower-buds, depends on 

 the location of the tumor or, in other words, on the type of 

 mother cells reached and stimulated. 



Recently, I have discovered how to cause dormant totipotent 

 cells to develop in the absence of tumor cells. This also I 

 believe I have accomplished by increased acidity due to the 

 abstraction of water (and along with it oxygen) from very young 

 tissues. For this purpose I used buds of the proliferous hothouse 

 plant, Begonia phyllomaniaca. On this plant, which is probably 

 a hybrid, and which is very sensitively balanced to loss of water, 

 it is possible to produce a veritable forest of shoots on leaves and 

 on internodes by wounding the roots and also directly in the 

 lips of wounds if these wounds are made in very young tissues, 



