Dec. i, 1923 
Morphology of Crowngall 
433 
region inoculated. Consequently, a series of sections were cut both 
longitudinally and transversely through the growing tips of the sweet 
Dea, sunflower, and Paris daisy. 
A median longitudinal section through the sunflower may be seen in 
Plate 6, A. Intercellular spaces have been located in the pith as far above 
the usual position of the puncture (base of arrow) as the distance indi¬ 
cated by the arrow. The great condensation of the intemodes is shown 
by the leaves and leaf primordia on the sides. Those shown are, of 
course, by no means the total number. It is estimated that only about 
one out of four of the total number of young leaves appears on one side 
of the section. Thus it may be seen that the area covered by the arrow 
extends over sixteen intemodes. Of course, it must be borne in mind 
that the buds of individual plants vary somewhat. 
The cross section of the tip of a sunflower stem shown in Plate 6, B, 
was taken 0.13 mm. below the apical cell of the growing tip at the 
place indicated by the transverse line in A. Here it may be noted that 
the intercellular spaces are already formed in a few places in the pith 
(pi. 6, D) and are comparatively well developed in the leaves that are 
differentiated. Well formed though small intercellular spaces have been 
observed even in a portion of the smallest leaf section (pi. 6, C). Since 
three of the older leaves which have their origin above the point of punc¬ 
ture do not show in B, it is quite evident that the intercellular spaces 
occur in more than seventeen intemodes above the position of the 
puncture. 
So it appears that the liquid released by a puncture in the rapidly 
elongating region of the sunflower bud could find a channel in the inter¬ 
cellular spaces of both cortex and pith to spread over the distance occu¬ 
pied by more than fifteen intemodes. At the same time, the linear dis¬ 
tance of migration of the organism at the time of inoculation to cover 
this number of expanded intemodes, as may be seen in Plate 6, need 
not be more than seven mm. It seems unquestionable that the bacteria, 
once inside such a wound, would migrate to the limits of the avenue 
provided by the flooded intercellular spaces. Thus “secondary galls" 
separated by a number of intemodes from the puncture might be pro¬ 
duced at any position or series of positions where conditions were favor¬ 
able. The actual distance of separation would depend upon the amount 
of expansion as the water-soaked region elongated. 
It is not surprising then that “secondary galls" and “tumor strands" 
occur as far as 60 cm. away from the point where the needle en¬ 
tered. Rather it is more surprising that “secondary galls" are not pro¬ 
duced in a higher percentage of trials. In actual practice it appears that 
the water soaking does not progress even as far as the thirteenth node 
except in 1 or 2 per cent of the inoculations in the region of elongation. 
At the same time, it is evident that “secondary galls" and “tumor 
strands" are not really secondary at all from the standpoint of invasion. 
They are provided with bacteria from the same inoculation that produced 
the “primary gall." Their appearance of being secondary is the result 
of their separation from the point of entry by the processes of elongation, 
and also of the presence of a much smaller number of bacteria in a region 
that is less completely occupied by liquid. This accounts for their ap¬ 
pearance later than the “primary gall" and their being smaller in size. 
A similar situation has been found as the result of examinations made 
of longitudinal and cross sections of Paris daisy and sweet pea. Photo¬ 
micrographs of median longitudinal sections through the growing tips of 
