IRRADIATION OF LIVING PROTOPLASM 629 



amount of salt that penetrates. The permeabihty coefficient n is defined 

 as 1 — C/C, where C is the plasmolytic concentration of salt and C the 

 concentration physically isotonic with the plasmolytic nonelectrolyte. 



Trondle studied the palisade and parenchyme cells of the leaves of 

 Tilia cordata and the parenchyme cells of Buxus semper vir ens. Lepesch- 

 kin used the pulvinus cells of Phaseolus, Spirogyra filaments, and Rheo 

 leaf cells. In both studies daylight of undetermined intensity was the 

 source of radiation. Both authors found an increase in permeability 

 to KNO3 with increased illumination. Thus, in Tilia, 28 experiments in 

 cloudy weather gave an average /x of 0.24; in direct sunlight the average 

 of 22 determinations was 0.32. With an artificial source of light, a 

 32 candle-power bulb, the value of n could become even higher. Trondle 

 does not report exposure time intervals, though from later experiments 

 (362) he plotted a time curve showing that illuminated cells ultimately 

 became less permeable than darkened cells. 



Lepeschkin's results are similar. In cells of Phaseolus kept in sunlight 

 the permeability coefficient was 1.5 to 2 times as great as that of the 

 darkened controls. 



This method of isotonic coefficients has been criticized by many 

 reviewers (46, 101, 349). Zycha (404) has sharply attacked the results 

 of Lepeschkin and Trondle, and with considerable justice. In his 

 experiments Zycha used the same material as his predecessors and 

 employed the same method of isotonic coefficients with much regard to 

 exactness of technique. Examining carefully the sources of error in the 

 method, he calculates that it can be as high as 30 per cent. His experi- 

 mental results are diametrically opposed to those of Lepeschkin and 

 Trondle, but he discards them as w^ell as the latter because of the intrinsic 

 uncertainty of the method. Lepeschkin (221) criticizes Zycha's use of 

 parallel slices of leaf for the determinations of the plasmolytic concentra- 

 tion of salt and of sugar, respectively; the variation among slices, he 

 claims, does not permit this. Zycha had insisted upon this procedure 

 because plasmolysis, occupying an hour or more, must injure the cell 

 to such extent that the second determination is made with an abnormal 

 cell. Both of these claims may be just and together invalidate the 

 method entirely. Nadson and Rochline-Gleichgewicht (269) found that 

 plasmolysis by sucrose causes a release of Ca in cells of Elodea and Pterygo- 

 phyllum as evidenced by the formation of calcium oxalate crystals. 

 One of us (Mazia) was able to verify this observation in Elodea cells and 

 found, further, that plasmolysis by salt solutions could cause this Ca 

 release. These facts would support Zycha's view. 



Brauner (43) used a plasmolytic method, with controlled illumina- 

 tion. An exposure of cells from the coleoptile of Avena to a source of 

 50,000 meter-candle sec. for 30 min. oaused an increase of 25 per cent in 

 the amount of NaNOs penetrating in a fixed period of time. After 



