43 8 Farmer and Freeman. — On the Structure and 
The Foliar Bundles. 
The leaf-trace-strand as it leaves the stele is seen to be 
enfolded, though sometimes only on its outer and lateral 
faces, by an endodermis which is derived in the first instance 
from, and is in any case directly continuous with, that of the 
stele itself. As the trace bends out the stele is disconnected 
on one side first, and the foliar strand hinges on the still 
unbroken side in a somewhat characteristic fashion. It re- 
calls that of Gleiclmiia hecistophylla as figured by Poirault 
although the resemblance between the two cases does not 
extend to points of detail. 
After the trace has ceased to be connected with the parent 
stele, the endodermis is usually found to completely surround 
it. The strand soon betrays evidence that it consists of two 
main bundles. At first the protoxylem is in an almost 
central position conformably with the position which it 
occupied when the trace still formed a sector of the stele ; but 
the parenchyma soon becomes more markedly aggregated on 
the inner side of the central part of the trace, and the proto- 
xylem is then grouped in two main portions just outside this. 
Finally this parenchyma, which at first looks like a rather 
irregular pith, separates the xylem on the side directed 
towards the stele of the stem, and at the same time a similar 
division is effected on the opposite side ; and thus, by 
a gradual series of transitions, the trace comes to consist of 
two main collateral bundles with their protoxylem on the 
inner side of their chief mass of wood 2 . However, it may be 
observed that traces of the original mesarch structure can often 
still be detected (and this is true of bundles here and there in 
the petiole of the leaf itself), owing to the formation of a very 
few elements which arise on the inner side of the protoxylem. 
1 Poirault, loc. cit., p. 173. 
2 It is of interest to note in this connexion that the leaf-traces of Botrychium 
virginiamim belong, according to Jeffrey (Mem. of the Bost. Soc. of Nat. Hist., 
vol. 5, p. 160), to the concentric type. 
