22 
R. H. Compton 
this has its own “ protoxylem ” which is destroyed in the same way 
as the “ vaisseaux nlternes ” of the root, and corresponds to them 
functionally, but belongs to a much later stage of evolution. This 
is an example of what Chauveaud calls “ l’accdl6ration basifuge,’’ 
by which, as we ascend the axis from root to stem we have early 
stages of ontogeny successively omitted. 
If we study a seedling by means of transverse sections taken, 
not at successive times from the same place, but from different 
places at the same time, we behold the changes in the disposition 
of the tissues commonly known as the anatomical transition from 
root to stem. According to Chauveaud the different structures 
seen at different levels of the axis are all parts of his ideal scheme 
of alternate, intermediate and superposed: the differences being 
due to the fact that the initially formed elements belong to phases 
of vascular evolution successively later as we ascend the axis. In 
the root the first formed elements of the xylem are of the alternate 
phase: half-way up the hypocotyl they may belong to the intermediate 
phase, the alternate phase being unrepresented : at the summit of 
the hypocotyl ontogeny may start with superposed protoxylem 
elements, alternate and intermediate phases being omitted. Thus 
the transition from root to stem is the expression of basifugal 
acceleration. The protoxylem of the stem is not primitive in the 
same sense as that of the root. 
Chauveaud remarks that the sequence of phases is not strictly 
adhered to in all cases: “ the intermediate phase often seems to be 
suppressed, and the passage from the alternate to the superposed 
arrangement is abrupt. This is the case when the alternate vessels 
meet one another at the centre ; when the xylem groups are very nu¬ 
merous and closely approach the phloem groups ; or when the phloem 
bundles, separated at first by alternate vessels, fuse with one another 
after the suppression of these alternate vessels.” To these exceptions 
the present writer must add a case very pronounced in the large 
seedlings of Lupinus spp., and of such frequent occurrence in other 
plants as to become a normal and wide-spread type. In this type 
the alternate and the superposed vessels are alone present in certain 
regions of the hypocotyl, and are separated from one another by 
greater or smaller distances: in Lupinus hirsutus as many as 
fifteen parenchymatous cells may intervene on either side between 
the median polar protoxylem and the two great collateral bundles 
of common connection with it in the root. As we ascend from root 
to hypocotyl the polar xylem has the appearance of forking into 
