1886-87.] Movement of Water in Plants. 93 



view, on the ground that if you cut a plant close to the root 

 you can see, on rapidly applying blotting-paper to the section, 

 water oozing out of the section. All who have used this 

 method are against Sachs' view, Elfving (10) gives what is 

 tantamount to a direct proof of its incorrectness. Eosin 

 stains some substance in the cell-walls of coniferous wood, as 

 can easily be seen in a transverse section ; but if a branch of 

 Taxus is made to suck up coloured solutions, the walls are not 

 coloured, according to Elfving, while the cavities are full of 

 eosin. 



There are also some indirect ways of getting at the truth. 

 Sachs and Dufour (18), and latterly Darwin and Phillips (17), 

 tried the effect of making incisions in the stems. The latter 

 observers made two incisions, each as far as the pith on 

 opposite sides, and some distance apart. They found that 

 in Helianthus (where such treatment would cut almost all 

 the vessels) the transpiration was greatly diminished ; while 

 in Taxus, where continuity by the tracheids would not be 

 affected, there was practically no difference. 



The same observers (17) and also Kohl (18) and Vesque 

 (13), have found that when a stem is strongly squeezed 

 together in such a way that the cavities of the vessels may be 

 considered obliterated, no water can pass through; though, 

 on Sachs' view, it is quite possil^le. 



Another method, on which Sachs appears to rely, has been 

 adopted (at Wiirzburg) by Dufour (18). If a twig is bent 

 at a very sharp angle, Dufour found that transpiration is not 

 stopped. He considered that in this twig the vessel cavities 

 must be closed, as he could not inject twigs bent in this way 

 when separated from the tree. This conclusion is quite 

 fallacious, however. Russow (6) clearly shows that the cavity 

 of the vessels in this case is not closed, and Scheit (4) has 

 shown that if Dufour had cut the bent twigs under water 

 they could have been injected ; see also Vesque (13). 



A very ingenious proof is given by Weber (20), who found 

 that vessels whose walls were apparently changed by heat into 

 a gummy substance still continued to conduct water. 



Some remarkably conclusive experiments of Elfving (10) 

 appear to me to prove the point. His method consisted in 

 closing or rather filling up the cavities of the vessels by 

 melted coco-butter injected by the cut surface. The lumina 



