52 Transactions. 



its position in the mature leaf of two species of Podocarpus ; 

 while in the third species (totara), as has already been pointed 

 out, he was unable to find any at all. I therefore feel more at 

 liberty to express an opinion with regard to this group. It 

 seems rather a premature proceeding to confine the origin of 

 transfusion tissue in all gymnosperms to centripetal wood when 

 the evidence is conclusive only in the lowest groups. 



Now, in the Podocarpece — of which, for the development of 

 transfusion tissue, P. totara may for the present be taken as a 

 type, the development being similar in the following species — 

 in no section either of the cotyledon or of the mature leaf was 

 there any great development of centripetal xylem, the elements, 

 if any, being very occasional even in the cotyledons, where 

 we should most expect to find them. From the cotyledons 

 upwards the transfusion tracheids were always at the side of the 

 centrifugal wood, and in many cases, as will be seen from the 

 drawings of the bundle, there were direct transitions to them 

 from the px through the centrifugal tracheids which extended 

 out towards the sides. In every species there was a marked 

 increase in the number of transfusion tracheids from the earliest 

 to the later stages, where there is no evidence of any centripetal 

 xylem ever having been formed. These transitions, which in 

 many cases make it hard to distinguish which is to be regarded 

 as centrifugal wood and which as transfusion tracheids, to- 

 gether with this gradual increase in number from the earliest 

 to the later stages, seems to give almost conclusive evidence 

 in these species of their origin not from the centripetal but 

 from the centrifugal xylem. Near the apex of the young cotyle- 

 don we actually see the wood of the bundle passing out to the 

 sides, and serving as transfusion tracheids. When one or two 

 elements of centripetal wood have been formed, in many cases 

 they have been preserved and used on the ventral surface as 

 transfusion tracheids, but I see no reason because of this why 

 we should regard all transfusion tracheids as having been 

 formed on this side of the px, and then as passing out and 

 attaching themselves in direct communication with the centri- 

 fugal tracheids at the sides. 



The character of these elements does not in any way alter 

 this opinion : there are transitions here out through tracheids 

 at the sides from the px. In the case of P. totani it will be seen 

 from the longitudinal section of the shrub-leaf how greatly 

 modified are these elements on the outer edge, appearing almost 

 like parenchyma cells, and very hard, in many eases, to dis- 

 tinguish from these. I have found undoubted cases where the 

 walls are only very slightly lignified, the reaction of the wall 

 being more that of cellulose, but which have undoubted bordered 



