Edith Chick on Torreya Myrisiica. 
89 
type of thickening (a combination of scalariform thickening and bor¬ 
dered pits), then scalariform metaxylem elements and next proto- 
xylem elements which are loosely spiral and annular. To the 
inside of these again we get long narrow tracheides, of greater 
diameter than the protoxylem. Some of these have loose thicken¬ 
ings almost of the character of protoxylem thickenings, but the 
majority are scalariform, or sometimes pitted. In many places 
there can be seen a complete transition from these, through 
tracheides which are shorter but of greater diameter, to the 
scalariform and pitted elements of greater diameter and almost 
cubical form which make up the transfusion-tissue here. These 
transfusion-elements are found ventrally, and in longitudinal 
section are seen to occur for the most part in vertical rows, the 
row being interrupted occasionally by a parenchyma cell exactly 
similar in form, but with no thickenings. It certainly looks here 
as if the transfusion tissue in this position were being formed from 
the parenchyma outside the bundle. 
According to Mr. Worsdell, however, the presence of complete 
transition in size and other characters which is found between the 
elements of the centripetal xylem and the transfusion-tissue is an 
argument against this, and supports the origin of the transfusion- 
tissue as a direct extension of the centripetal xylem. 1 
In one pair of cotyledons (PL vii., fig. 2) there was an expanded 
region near the tip of each which might fairly be described as a 
lamina. In this region the bundle was very greatly elongated and 
there was a great development of transfusion-tissue on it’s flanks, 
shewing a tendency to extend on either side in a dorsal direction. 
It was almost absent in the central ventral position. There were 
also a good many centripetal xylem elements present. At the tip 
the transfusion-tissue had increased enormously in amount, and 
was now found completely enclosing the flanks of the bundle and 
extending to take up a central dorsal position. In two of the 
seedlings there was a lobing (in one case incomplete) at the tip of 
the cotyledon; in these the bundles divided just before the lobing 
took place, and the transfusion-tissue of each branch behaved 
exactly as above. In one case one of the lobes showed a cambium 
present on the dorsal side of the bundle, and a few secondary 
elements were formed from it, so that finally one had a concentric 
bundle with the original protophloem in a central position. 
1 W. C. Worsdell 011 Transfusion-Tissue, Trans, Linn. Soc., 
Dec., 1897. 
