TO BIRD’s-EYE FORMATION IN THE WOOD OF TREES. 31 
slight extent, but, being held firmly in position by their long 
Spindle-form and their pointed ends dovetailing into each other, 
they gradually cut into the soft protruding tissue of the root. 
Such a patch may be seen to the right in Plate XI., Fig. 43, lying 
at the bottom of the deep incision which it has made in the 
cortex of the root. 
A comparatively early stage of development is shown in Plate 
XI., Fig. 42, which is a section transverse to the stem and longi- 
tudinal to the root-rudiment. That this structure has nothing ~ 
whatever in common with the vascular supply to the leaves or 
buds was easily seen on comparing it with a transverse section of 
a normal stem. We have here a medullary ray which has 
become abnormally broad and developed a large swollen apex 
projecting into the cortex and causing a corresponding papilla to 
appear on the outside. Over the apex of the protruding 
medullary ray may be seen a dark semi-circular cap, which is 
formed by the crushed and flattened remains of cortical cells, 
against which the apex of the outgrowing protuberance is 
pressing. The continuation of the cambium round the swollen 
end of the ray can be traced. 
The dense coral-like masses at the base of the shoot (Plate X., 
Figs. 36 and 37), are formed by closely-packed adventitious 
roots which branch copiously. 
Elm. 
There are various elm-trees (U/mus campestris) in the Garden 
which have likewise thrown out numerous peg-like protuberances 
which, on microscopical examination, were also found to be 
medullary ray-roots. Plate XI., Figs. 44 and 45, give a general 
impression of the appearance and arrangement of those adventi- 
tious structures. In a radial longitudinal section of both stem 
and adventitious root (Plate XI., Fig. 49), the continuation of 
cambium and xylem of the stem into the papilla can be easily 
traced. In transverse section those excrescences showed distinct 
root-like characters. 
The aerial roots in this tree had not pierced the periderm or 
cork-layers of the stem, and it would seem that, for a time at 
least, the phellogem keeps pace with the outgrowing root. On 
this tree I have not yet observed any cases where the cork-layer 
has been broken through. 
