422 Boodle and Hiley . — On the Vascular Structure of 
the protoxylems j 3 and 74 branches into two, thus giving yf, and 
y 4" (cf. Plate XXIX, Fig. 6). The protoxylems 73" and yf' take no part 
in the formation of the foliar bundle. 
The detachment of the leaf-trace from the stele of the rhizome takes 
place in the following manner. The radial band lengthens, and a gap is 
formed in the xylem between the protoxylems y 3' and yf, and through 
this gap the internal sclerenchyma, phloem, and endodermis become con- 
nected with the corresponding external tissues, the process being similar to 
the differentiation of the radial band, as described above. One end of the 
arched leaf-trace is thus separated (PL XXIX, Fig. 6). Soon afterwards 
a similar interruption is formed on the other side between the protoxylem- 
groups.yq' andyf', and the foliar bundle is then completely severed from 
the stele of the rhizome, leaving a leaf-gap, which, however, soon disappears. 
The protoxylems y%" and y 4" by division give rise to the usual four upper 
protoxylems of the stele, and the normal internodal structure is restored. 
There is no distinct thickening of the ‘ margins of the leaf-gap ’, such as 
G wynne- Vaughan (’ 03 , p. 700), inferred from the earlier description of the 
node (Boodle, ’01, p. 730). 
The xylem of the leaf-trace is of the horseshoe type with incurved 
ends, and is endarch, several groups of protoxylem being scattered on the 
inner face of the arch. The phloem is continuous on the outer side of the 
xylem-arch, and on the inner surface of the incurved ends, but is absent in 
the inner middle region of the arch. 
The description given above applies to nearly all the nodes examined, 
but in one node among the material sent by Mr. Jenman the structure was 
of a somewhat different type. In this case immediately the leaf-trace 
became detached from the stele of the rhizome one protoxylem-group of 
the leaf-trace (73' in PL XXIX, Fig. 6) became markedly mesarch, and 
a small group of fibres (accompanied by endodermis and phloem), branching 
off from the internal sclerenchyma, became immersed in the metaxylem in 
the neighbourhood of the protoxylem-group in question. In this way a 
group of tissue, resembling a small nodal island, came to be contained in 
the xylem of the leaf-trace near one of its extremities. Unfortunately the 
petiole had been cut off close to the base, so that it was not possible to trace 
the fate of the severed island. Another type of nodal structure was 
described by Boodle (’ 01 , p. 730) ; in this case similar groups of tissue, 
resembling small nodal islands, became embedded in the xylem of the stele 
in the neighbourhood of the protoxylem-groups y$" and j/4", and this took 
place just before the separation of the leaf-trace. These secondary islands, 
which differ from the typical nodal islands in being extended acropetally, 
were apparently not continued for any great distance. 
