226 PHYSIOLOGY OF NUTRITION. 



withers. In Rhus typhina the splint is so thin that the ringing 

 injures it, and the conduction of water is necessarily interrupted. 1 

 In my experiments with Rhus typhina the part above the ring 

 was withered even after a few hours. Rhus glabra does not be- 

 have in the same manner. 



We cut off a shoot of Impatiens noli-me-tangere, or I. par vi flora, 

 and place it with its cut surface in an aqueous solution of methyl 

 green. The stalks of these plants are very transparent, so that 

 we can observe in a remarkably beautiful manner the phenomena 

 here under consideration. In an experiment with Impatiens 

 parviflora it was found that the colouring matter had already 

 risen in the stem of a fairly actively transpiring shoot to a height 

 of a few centimetres at the end of a quarter of an hour, and 

 microscopic examination of transverse sections of the stem showed 

 that only the wood of the vascular bundles, which are arranged in 

 a circle, was stained. 



The results of this and similar experiments have recently ac- 

 quired great significance from the researches of Wieler 2 and Stras- 

 burger. 3 They may indeed be utilised to establish the position 

 that the movement of water in the plant takes place in the wood. 

 In discussing, however, the results of these experiments with 

 pigment solutions, it is always necessary, especially as regards 

 details, to be very cautious. 



A leaf-bearing branch of Robinia, about 20 mm. in diameter, 

 which has been standing for some time in water, is placed in an 

 aqueous solution of eosin, without its cut surface being exposed 

 to the air. The eosin solution must be so dilute that a thickness 

 of 10 cm. is still transparent. Examination at the end of a few 

 hours teaches that only the outer parts of the wood, not the 

 central parts, are stained. The latter no longer conduct water. 

 If the experiment does not continue too long, it is also easy to 

 prove, from the distribution of the eosin, that only the tracheal 

 paths, and not wood fibres, take part in the conduction of the 

 water. The latter remain unstained. 



If Tilia branches are treated in the same way, and examined 

 after a few hours, the eosin can only be detected in the vessels 

 and tracheides. Wood fibres and bast remain unstained. In ex- 

 periments with Aristolochia the wood stains ; the ring of scleren- 

 chyma fibres remains unstained. 



Having thus determined that the movement of the water takes 

 place in the tracheal channels, we will now show that the conduc- 



