353 
section of the first root, however, the initials were not quite 
clear. Lang (22) has shown that the number of root imitials is 
not so much dependent on the age as it is on the bulk of the 
roots. Campbell (9), however, found always single initials. 
The cotyledon is not a scale-like rudiment as in B. lunaria, 
but a more or less developed structure. The narrow apical cell 
is already planly visible. À single vascular strand connects 
root and leaf. 
The position of the embryo in the prothallus is, as far as 
could be seen, quite the same as that in B. lunaria. The only im- 
portant difference between this form and B. simplex up to this 
stage is the large size of the first leaf in the latter. No trace of 
a vascular bundle appears in the cotyledon at this stage. The 
initials of the cotyledon could not be traced with any certainty. 
The Formation of the Stele. 
The second root makes its appearence at the same place as 
in B. lunaria. À series representing this stage 1s shown in Fig. 37. 
The second leaf 
has not yet deve- 
loped its leaf trace. 
The leaf trace in 
the cotyledon :1s 
planly visible. 
The second root 
appears very near 
to the place where 
the bundle of the 
second leaf will de- 
velop. The three Fig. 27 
traces Join in the 
foot to form a closed ring of xylem. Root initials could not 
be made out. 
The formation of the stele up to this stage proceeds therefore 
as follows : 
