Robinson and Walkden.—A Critical Study of Crown Gall. 311 
bers and also the diversity of the organisms obtained from the broken galls 
are sufficiently accounted for by the former having lodged in crevices of 
the rough surface of the gall, or as the residue of the bacteria that originally 
penetrated some little distance below the surface of the cut shoot at the 
time of inoculation. 
The aerobic character of B. tumefaciens may account for the undoubted 
fact that the bacteria which are introduced to the interior either of vessels 
or of intercellular spaces of Chrysanthemum frutescens , which has relatively 
small intercellular spaces, do not multiply to any extent nor grow pro¬ 
gressively in the interior. 1 
Experiments, using plating methods similar to those described above, 
have demonstrated with certainty that, for several weeks after inoculation, 
there is a progressive increase in the numbers of B. tumefaciens present on 
galls left growing on the plants. 
The plating experiments thus show conclusively that, in the actively 
growing galls with rough surface on Chrysanthemum frutescens , there are 
enormous numbers of bacteria situated on the exterior of the gall, that 
a high percentage of these is B. tumefaciens , and that it is reasonable to 
assume that the presence of these progressively increasing numbers of 
B. tumefaciens on the exterior provides the progressive stimulus which 
leads to the continued growth of the gall. 
The facts regarding the distribution of the bacteria had obvious 
bearings upon the origin of the so-called secondary tumours and tumour- 
strands which have been so fully studied by Smith. These aspects of 
crown gall were therefore reinvestigated, and may now be considered. 
Secondary Tumours and Tumour-strands. 
As has been stated in the introduction to this paper, we have succeeded 
in producing on Chrysanthemum frutescens and on Nicotiana affinis galls 
similar in all respects and similarly distributed to the secondary galls 
figured by Smith. It has not, however, been found possible, as might have 
been assumed from some of Smith’s figures, to obtain secondary galls in 
Chrysanthemum frutescens at a distance from the primary inoculation by 
inoculating the shoot at some distance from the growing-point. With this 
plant we have obtained successful results only by the inoculation of the 
very young tissues in the vicinity of the meristematic apex. This led us to 
test a view suggested to us some years ago by Professor W. H. Lang, viz. 
that the appearance of secondary galls and of tumour-strands was due to 
the subsequent development, growth, and extension of the meristematic 
tissues under the influence of the bacteria after inoculation. 
1 Below it will be seen that in the shoot of Nicotiana the bacteria behave differently in 
undoubtedly multiplying and progressing in the interior. 
