TRANSLOCATION OF CARBOHYDRATES 471 



in dififerent ways according to the plant chosen. In the case 

 of Impatiens parviflom, it first left the bundle sheaths, next 

 the adjacent cells and finally the more distant cells of the 

 mesophyll. In Hydrocharis morsus-rmm^ on the other hand, 

 it left the more distant mesophyll cells first and apparently 

 accumulated in the bundle sheaths. Later these became starch- 

 free, the small sheaths first, then the larger ones, until finally 

 no starch remained even in the largest sheaths. 



Leaves of Impatiens parviflora left several days in darkness 

 became sugar-free ; in this case it was found that the " glucose " 

 left the mesophyll first, next the finest sheaths and later still 

 it could only be detected at the bases of the stronger veins. 

 Ultimately the sugar left the lamina altogether and could 

 only be demonstrated in the petiole. A number of experiments 

 of this kind led Schimper to conclude that sugars travel almost 

 -exclusively in the bundle sheaths. But he only followed their 

 passage as far as the bundle sheaths ; his method was not 

 well adapted for observing the contents of delicate cells like 

 sieve-tubes, situated in the middle of the tissues of the veins, 

 indeed he appears to have neglected these elements almost 

 entirely. His observations were therefore incomplete, though 

 accurate as far as they went. 



It will be seen that similar changes in the distribution of 

 sugars would occur if they passed from the mesophyll into 

 the small sheaths and were then withdrawn by the sieve- 

 tubes which the sheaths enclose. Continuous diffusion from 

 the sheaths into the sieve-tubes would eventually remove 

 the whole of the sugar from the leaf; naturally the smaller 

 sheaths would first become empty. 



I have frequently observed stages in the process of emptying 

 in which the sheaths of the small veins contained no sugar, 

 while their sieve-tubes contained a considerable amount. In 

 the light of the anatomical facts previously considered it seems 

 not unfair to infer from such a distribution of the sugar that 

 it had passed from the sheath into the sieve-tubes and was 

 in course of translocation in these elements. This is made all 

 the more probable by the fact that in still later stages sugar 

 could only be detected in the sieve-tubes nearer the bases 

 of the veins. (Fig. 19.) 



As previously pointed out a certain amount of sugar passes 

 directly from the mesophyll into the nerve parenchyma of 



