* 



286 BOTANICAL GAZETTE [October 



tangential walls when in contact with similar cells are perforated 



with openings which vary from simple pits to relatively large sieve 



plates with numerous small apertures. The transversely elongated 



elements derived from the pericycle by tangential divisions are, 



as pointed out by Seward and Ford (io), sieve elements which 



are of later origin than the sieve tubes of the protophloem. The V 



sieve plates of these elements, as asserted by Faull (4), show 



callus plugs, and all the essential features of true sieve cells are 



present. There was no evidence of the appearance of phloem in 



the internal parenchyma in any of the material investigated, nor 



were the sieve cells ever observed to appear in the parenchyma 



tissue of the foliar gaps. 



As already stated, there is a single row of nucleated paren- 

 chyma cells which surround the xylem in the first stages of the 

 stele. At higher levels these elements increase in amount by 

 tangential divisions. In longitudinal view they are observed to be 

 narrow, elongated, with acutely oblique terminal walls. These 



resemble very closely the earliest xylem parenchyma. Are these 

 elements to be considered as of cortical origin, as asserted by 

 Jeffrey (7) and Faull (4), or are they to be considered as dis- 

 tinctly stelar and representing undeveloped potential xylem 

 elements differentiated by the meristematic stem apex? In the 

 young stele these cells do not resemble the cortical tissues either in 

 topography or cytology. The central ones at first have distinctly 

 acute terminal walls, and in general topography are altogether 

 like the tracheae, except, of course, for the secondary thickenings 

 of their walls. At higher levels these internal unthickened elements 

 frequently have transverse terminal walls or are but slightly 

 obliqued from the horizontal. Further evidence for the stelar 

 origin of these elements is found in the occurrence of tracheids in 

 the same linear series with parenchyma cells above and below 

 (fig. 29). It was further noticed in such instances that the tracheids 

 were terminated by walls which were transverse or almost so, 

 instead of having the strongly oblique parenchymatous type so 

 characteristic of the normal tracheae of the xylem axis. Cells of 

 this type were found on the inner edge of the xylem in contact with 

 the parenchyma, and in a few instances on the outer border where 



f 



* 



