98 BOTANICAL GAZETTE [AUGUST 
or constant in position, although they are most numerous in con- 
tact with the medullary rays (fig. 12). That their origin is the 
same as that of the ordinary pitted tracheid is seen at once in a 
transverse section of the wood (fig. 17). 
Still another form of tracheid is found in the large medullary 
rays (fig. 18). These tracheids are scalariform, are irregular in 
outline, and are nearly erect at their upper end, but become nearly 
Yu 
WS 
al 
UU 
CS) 
aN 
we 
LOE 
y) 
Ny) 
@ 
) 
OK 
OVS 
i 
yo Ww 
WY 
IN 
\( 
- 
Ss 
IN 
© 
+ 
IG. 15.—Dioon spin 
ro > 4 J 3 . 
horizontal deeper down in the ray, so that a tangential section of 
the ray shows them in both longitudinal and in transverse section. 
Every large medullary ray would show these peculiar tracheids 
at some point or other, and they are particularly numerous near 
the pith. They connect the leaf trace bundle, which is found in 
every large ray, with the secondary xylem, a connection hitherto 
unknown in cycads. Such a secondary connection of the leaf 
trace is a prominent feature in angiosperms. Professor R. B. 
THOMSON examined preparations of the large medullary rays, and 
