384 Cannon : Anatomy of Phoradendron villosum Nutt. 



walled cells. The latter are densely filled with finely granular 

 matter the nature of which I did not learn. I did not succeed in 

 demonstrating the presence of sieve tubes in the sinkers ; their 

 existence may be questioned. The long, thin-walled cells of the 

 central cylinder are gradually shortened, and their walls thickened, 

 as they approach the end of the sinker, and they are there indis- 

 tinguishable from the other cells that make up the tip. The ducts 

 also, at the end of the sinker, merge into the cells that compose 

 it, so that there is a gradual transition from the cells of the central 

 cylinder to those of the end of the sinker. The cells at the end and 

 on the margin of the sinker are in contact with the host ducts and 

 the union is such a close one that there is apparently only one cell 

 wall between them. The pits of the sinker are in apposition with 

 those of the ducts (Fig. n) and the status of the parasite as a 

 partial, or water, parasite, is thus established. 



In the region of the cambium of the host, as has already been 

 mentioned, the sinker cells are smaller and with thinner walls than 

 those either toward or away from its tip. These cells have large 

 nuclei and dense protoplasmic contents. They are typical meris- 

 tematic cells. The thickness of this part of the sinker is some- 

 what more than that of the cambium of the host adjacent to it. 

 By the means of active cell division in the meristematic region of 

 the sinker it is able to accommodate its length to the increasing 

 diameter of the oak branch. And it is also possible that the ele- 

 vation of the aerial haustoria from the host is in part accomplished 

 by the same agency. 



Longitudinal and cross sections of the sinker in the region of 

 its meristem show that the thin-walled conductive system does not 

 unite with the phloem of the host, sieve-tubes do not unite with 

 sieve -tubes as in complete parasites. There is, however, a transfer 

 of foods from the phloem of the host to the sinker as in the corti- 

 cal haustoria, but it takes place by osmosis only, because there do 

 not appear to be any cells of the sinker with epithelial characters. 

 The sinker, because of the lack of cells for digestive purposes, does 

 encroach upon the phloem of the host. This is shown by longi- 

 tudinal sections (Fig. 12). 



The aerial haustoria are those cortical haustoria which have 

 become more or less freed from the host cortex, are partly or 



