316 Robinson and Walkden.—A Critical Study of Crown Gall. 
At this stage there is a single protoxylem element extending down from the 
base of the leaf into the shoot, towards its rudimentary vascular cylinder. 
Secondary tumours arise when the needle-prick introduces the bacteria into 
this protoxylem element either in the leaf-base or in the stem. If the rudimen¬ 
tary leaves infected are somewhat farther advanced, the needle penetrates the 
region at the extreme base of the leaf where growth has almost ceased, the 
organisms are not carried forward in the stretching protoxylem, and, instead 
of an elongated series of galls, a single large gall arises at the base of each 
leaf wounded and inoculated. PL V, Fig. io, illustrates this effect. If, on the 
other hand, the leaf-rudiment is very young, the gall which arises involves 
the whole rudiment and thus obliterates the leaf. In other cases, where the 
protoxylem is entered by the bacteria, the young leaf may be so severely 
wounded that the development of secondary galls is accompanied by a par¬ 
tial arrest of the leaf as in PI. VI, Fig. 13. The facts thus briefly summarized 
for the inoculations of leaf-rudiments of Chrysanthemum frutescens were 
obtained by the examination of serial sections of a large number of apices 
at differing periods varying from three to fifteen days after inoculation by 
transverse or longitudinal needle-pricks. The observations were controlled 
in the case of each apex examined by serial sections of a corresponding 
wounded but uninoculated apex of the same age. The results have shown 
that the secondary galls on the leaves of Chrysanthemum frutescens are 
largely accounted for by the bacteria inoculated into the young protoxylem, 
being carried by the stretching growth of this to some distance from the point 
of inoculation ; our results on tobacco have shown that an actual migration 
of the bacteria is also possible in the protoxylem. The fact that the bacteria 
are restricted to the protoxylem in the secondary galls on the Chrysan¬ 
themum and that, though staining clearly shows them to be present, they 
are never so abundant as in the galls on the tobacco, fits with the undoubted 
difficulty which both Smith and ourselves have had in isolating the 
organism from these secondary galls on the Chrysanthemum. It would 
seem that there is, in these cases, little actual multiplication of the organisms 
after inoculation, while the undoubted very active migration and growth of 
zoogloeal strands in the shoot of tobacco clearly indicates the reverse. This 
may be correlated with the much greater ventilation of the tobacco shoot, 
its larger intercellular spaces affording more favourable conditions for the 
active multiplication of the highly aerobic Bacterium tumefaciens. 
It may be mentioned here that measurements of the stem of the 
tobacco after inoculating the cut surface show that there is very little 
elongation of the stem such as occurs subsequent to the wounding of an 
apex either of Chrysanthemum or of the tobacco. In the secondary galls 
which arise below the inoculated stem-surfaces in the tobacco there is, 
therefore, no question of growth elongation playing any considerable part in 
the form and arrangement of the galls. 
