COMPARATIVE ANATOMY OF THE CYCADACEX. 451 
tracheides, none of which are centrifugal; but there is a large 
amount of phloem. 
The young stem shows four bundles in transverse section, of 
ordinary collateral structure and with centrifugal, but no centri- 
petal, xylem. Besides these central bundles there are six others 
running obliquely inward. These are the cotyledonary bundles, 
three belonging to each cotyledon. Before uniting with the four 
central bundles they would doubtless fuse into a smaller number, 
but I did not succeed in obtaining sections at a level low enough 
tosee this. Higher up in the stem the four central buudles are 
seen to gradually split into a greater number, which respectively 
and mutually alter their positions. The larger products pass 
outwards to provide the first one or two leaves. The smaller 
products remain in the central region and form an are which is 
the nucleus of the youngest leaf ; immediately on the side of this 
is the apex of the stem, consisting of meristematic tissue. 
The transition from stem- to root-structure was, unfortunately, 
missed. 
The uppermost part of the primary root shows a triarch stele. 
Between the three protoxylem groups are equally far-protruding 
xylem-masses, each with well-developed phloem on its outer side. 
These are partly the result of the secondary thickening which is 
present. Lower down, at two or three different levels, a lateral 
strand is given off, which, in one or two cases, appears to arise 
opposite one of the groups of xylem and phloem with secondary 
thickening situated between the protoxylem groups, as the con- 
spicuous phloem-mass seems to be carried onward. The xylem 
becomes entirely surrounded by phloem as the strand passes 
outwards and downwards in an oblique direction, appearing all 
the time in transverse section. Its tracheides are angular and 
isodiametric. Lower down it is seen to pass outwards into an 
endogenously-arising organ. It may be that the first-formed 
lateral roots in the upper part of the primary root arise endo- 
genously from the outermost part of the cortex, and that all the 
later-formed roots arise in the normal way from the pericycle. 
But I cannot be at all certain as to the exact nature of this lateral 
appendage. It is probably homologous with the structures 
observed in Cycas revoluta and Macrozamia spiralis (see above). 
The youngest part of the root appears to be diarch in 
structure, 
LINN. JOURN.—BOTANY, VOL. XXXIII. 2L 
