Notes on the Anatomy of Stangeria paradoxa. 23 
It should be mentioned that mucilage canals are very abundant 
all through the plant. Some are large and lie free in the cortex, 
while other smaller ones run next the phloem in connection with 
the bundles, following them right into the laminae, where the great 
amount of stain taken up by the mucilage often makes the structure 
of the small bundles difficult to ascertain. 
The Leaf. 
The main features of the leaf are given by Kraus and Nestler 
and with their results my own observations generally agreed. The 
number of strands in various parts of the petiole and midrib is not 
such a definite thing as Nestler gives one to understand: in the 
leaves examined the numbers were seen to be less. My investiga¬ 
tions were, however, mostly along a different line: they were 
directed chiefly to a study of the “ mesarch ” bundles of the leaf. 
The petiole shows in transverse section a number of bundles 
distributed in relation to the two rows of pinnae and forming the 
usual inverted omega, Pavolini records the occurence of “ medullary 
strands ” in the petiole, but there was none present in the material 
which I examined. 
It is convenient to deal first with the connection between the 
endarch cortical leaf-trace and the “ mesarch ” petiolar bundle. It 
was found that the transition from one type to the other took place 
in the basal centimetre of the petiole and a series of seven sections 
across this basal centimetre was cut. They were cut at 
approximately equal intervals and called a— 1 7 in acropetal order. 
I have drawn typical bundles from each section by means of the 
camera lucida and these form the diagrams of Fig. 5. In a the 
structure was entirely endarch but in (3 there was a small group of 
centripetal elements. The centrifugal xylem here was much more 
loosely arranged than in a or in the cortical leaf-trace bundles. 
There is a fairly definite differentiation of the centrifugal xylem 
into an outer secondary portion in which the elements are arranged 
in rows and an inner primary portion whose tracheids are in less 
regular arrangement, and are connected with the centripetal 
elements. A cambium is present here as in all bundles of the 
petiole and midrib, at whatever level, cutting off, on the outside, 
phloem, and on the inside, centrifugal xylem, both tracheids and 
parenchyma, y—?; show a progressive increase in the size of the 
centripetal portion of the xylem and a decrease in the amount of 
the centrifugal wood. The centrifugal becomes more completely 
of the secondary type, i.e,, it appears in connection with the rows 
