98 



SCIENCE 



[N. S. Vol. XLV. No. 1153 



after which it remains fixed. In CaCL the 

 permeability at first decreases until a cer- 

 tain minimum is reached: after this it be- 

 gins to increase and finally reaches a con- 

 stant value (as in NaCl), which signifies 

 death. 



Further experiments showed that all sub- 

 stances which affect permeability may be di- 

 vided into two groups, (1) those which act 

 like NaCl; (2) those which act like CaCL. 

 This led to the following hypothesis : Sub- 

 stances of the first group antagonize sub- 

 stances of the second group and vice versa. 



Experiments were then made to test this 

 lypothesis. It was found that substances 

 which behave like NaCl with respect to 

 antagonism (in experiments on gro^vth) 

 behave like NaCl in their effect on perme- 

 ability. Substances which behave like 

 CaCla with respect to antagonism also be- 

 have like CaCU in their effect on perme- 

 ability. Moreover, substances like LaClg, 

 which antagonize NaCl more powerfully 

 than does CaCl,, are found to affect perme- 

 ability more powerfully than CaCU. 

 There is therefore a striking parallel be- 

 tween effects on permeability and the 

 antagonistic effects observed in experi- 

 ments in which growth and length of life 

 are employed as criteria of antagonism. 



Equally remarkable is the outcome when 

 permeability is used as the criterion of an- 

 tagonism. It is found that all solutions 

 which permit normal growth are likewise 

 solutions which preserve normal permeabil- 

 ity. 



These experiments which were originally 

 made on Laminaria were afterward ex- 

 tended to other algse, to flowering plants 

 and to animals. 



Using permeability as a criterion of an- 

 tagonism, the speaker has made investiga- 

 tions on a great variety of substances. 

 Time is lacking to describe these, but it may 

 be said that the outcome in every case has 



supported the hypothesis. This was strik- 

 ingly shown in investigations on organic 

 substances (non-electrolytes), a number of 

 which were found to belong to the second 

 group. It turned out that all of these sub- 

 stances were able to antagonize NaCI, as 

 is required by the hypothesis. 



This result greatly strengthened the 

 speaker's confidence in the hypothesis 

 which seems to serve a useful purpose by 

 enabling us to predict what substances will 

 antagonize each other. 



As the result of these investigations we 

 seem to be justified in concluding that there 

 is a close connection between antagonism 

 and permeability. Conclusions concerning 

 such fundamental relations should be 

 tested, whenever possible, by a variety of 

 methods. This task was undertaken by 

 Dr. Brooks, who confined himself chiefly 

 to the following methods: (1) diffusion 

 through living tissue, (2) exosmosis, (3) 

 change of curvature of strips of tissue.^ 



In the first of these methods different so- 

 lutions were placed on opposite sides of a 

 piece of tissue. The diffusion of salts 

 through the tissue was then measured. 



In the second method the tissue was 

 placed for a short time in a salt solution 

 and the rate at which substances subse- 

 quently diffused out of the cell was meas- 

 ured. 



In the third method strips of the ped- 

 uncle of the dandelion were placed in hy- 

 pertonic salt solutions and the rate of pene- 

 tration of the salt into the protoplasm was 

 calculated from the rate at which the strips 

 recovered their normal shape after being 

 curved by the action of the hypertonic so- 

 lution (the strips remaining in the solu- 

 tion during recovery) . This gives the same 

 kind of information as plasmolysis but 

 avoids the most serious errors of that 

 method. 



2 Brooks, S. C, Proc. Nat. Acad., 2: 569, 1916. 



