502 Robinson and Walkden.—A Critical Study of Crown Gall. 
reaction of this plant to crown gall. These aspects of the disease lie outside 
the scope of this paper. 
The present work has been concerned with crown gall on Chrysan¬ 
themum frutescens and on Nicotiana affims. Bacterium tumefaciens was v 
independently isolated in this country in 1920 by Walkden ( 35 ) from 
naturally-occurring galls on Chrysanthemum frutescens. The cultural 
characters of the organism were fully studied by him, and it was proved by 
re-inoculation to be capable of causing the disease. Pure cultures so 
obtained by repeated re-isolations have been used for our experimental 
study of the disease (PI. V, Fig. 6). 
In the case of Chrysanthemum f rutescens the development of the galls 
on the cut surfaces of stems has been studied from the time of inoculation 
until the galls attain a large size, a strict comparison having been made at 
every stage with corresponding cut surfaces of uninoculated control stems. 
This study provided abundant material for an examination of the location 
of the bacteria causing the diseased growths, and we have obtained new and 
conclusive evidence regarding this very important matter. 
We have critically repeated Smith’s work on secondary tumours and 
tumour-strands both on Chrysanthemum frutescens and on Nicotiana 
ajfnis , and we have obtained galls similar in all respects and similarly 
distributed to the primary and secondary galls figured by Smith. Experi¬ 
mental evidence, supporting entirely different explanations from those given 
by Smith of the origin of these secondary galls and tumour-strands, is 
mven below. 
Although we have produced, by artificial inoculation both of Chrysan¬ 
themum and tobacco, structures precisely similar to the so-called teratomas 
figured and described by Smith, we do not propose to deal with these in 
this paper. The interpretation which we place on these so-called teratomas 
will, however, be briefly referred to in the discussion following the descrip¬ 
tion of our other results. 
It will be convenient first to describe the development of artificially 
produced primary galls on the Chrysanthemum from the earliest stages, 
then to deal with the distribution of the bacteria in the gall, and finally to 
describe experiments relating to the origin of secondary tumours and tumour- 
strands both in Chrysanthemum frutescens and in Nicotiana ajfnis. 
Development of Galls on Cut Surfaces of Shoots of 
Chrysanthemum frutescens. 
The material for the study of the development of galls after the 
inoculation of cut surfaces of stems was derived from healthy, well-grown 
plants of Chrysanthemum frutescens raised from cuttings and grown in 
