LEVINE: STUDIES ON PLANT CANCERS—IV 237 
lated into the stem and leaf axils of decapitated plants by the 
trocar method. Here in the thirty-nine plants studied, the 
largest crown galls were produced with the weakest suspension, 
6e. Yet the variations were slight and if a sufficient number 
of cases had been available, it is believed the average size of the 
crown galls would probably have been more nearly alike. The 
fact still remains, however, that the smallest crown galls were 
produced with suspensions bearing the largest number of bacteria. 
The inoculation of the upper axillary buds of decapitated 
stems showed that 6c and 6e gave results more than double 
those of 6b and 6d, respectively. The dilution in each case 
was half the strength of those that gave the smaller growths. 
Inoculating the midveins of leaves with the dilutions of 
Emulsion III listed above, gave more consistent and uniform 
results. In fifteen plants so studied the average diameter of 
the crown galls was 0.3 cm. The inoculations of the stem gave 
quite a range in size of the galls produced, but here again as 
shown for the other two series the number of bacteria did not 
obviously influence the size of the growths. The sizes of the 
crown galls seemed to be correlated with the individual vari- 
ations in the sizes of the plants, but here again emulsion dilution 
3d (1 : 15) gave crown galls almost twice as large as those ob- 
tained with 3a, which is the mother emulsion. 
Inoculating these plants by pricking them with a needle 
capable of holding a uniform quantity of bacteria gave approxi- 
mately equal results, which shows that the number of bacteria 
or strength of the emulsion within the range tested is not 
directly concerned with the size of the crown gall. Emulsion II 
was used in inoculating thirty-two plants by dipping the point 
or spoon-like end of a filled hypodermic needle into emulsion 
Ila, b ,and c of Bacterium tumefaciens, and then pricking the 
plant with the point of the needle. The inoculated plants 
showed at the end of the growing season that the more con- 
centrated suspension of Bacterium tumefaciens did not produce 
the largest gall. As it happened, it produced the smallest. 
Comparing the trocar and the needle method of inoculation 
as used, cultures Ila and I[c—IIa’ and IIc’ (the latter two cul- 
tures indicate similar dilutions but used with the trocar) show 
somewhat larger galls for the trocar method. The differences, 
however, do not appear to be significant. The weak dilution 
