MICHIGAN ACADEMY OF SCIENCE. £9 



longer tenable, either us an explanation of osmotic pressure, or of osmosis 

 in general. It is not only inadequate,, but is opposed to experimental facts 

 of a purely physical nature. 



Physiological experiments on plasmolysis and operations related to it, go 

 to show that we must look to other sources for explanatory ideas, than to 

 the famous van't Hoff's law. The activities of an osmotic character in the 

 plant cell cannot be explained on that ground, consecjuently we must look 

 to other processes. 



There have been several other theories offered to explain osmotic action, 

 but none of them has received the general prominence or acceptance accorded 

 to van't Hoff's law. The more important of these theories may be designated 

 as follows: Surface tension, capillarity, molecular attraction. 



Without going into detail explanatory of these, it may be said that there is 

 less or more to be said in favor of each. Of the last we have this to say 

 that if apjjlied generally, enough it proves satisfactory. But in writing this 

 ])aper, we have in mind an explanation defining and explaining this attraction 

 to suit the needs of the case. 



The explanation here offered postulates two things: (1) That moleculeo 

 of substances have by nature an attraction for one another and that the 

 amount of such attraction will dej^end upon the character of the substance. 

 (2) That a meml^rane may be obtained which is more porous to the molecules 

 of one sulistance than to those of another. By this is meant that a membrane 

 (e. g. copper ferro-cyanide) may offer more resistance to the passage of a 

 molecule of one l^ody than to that of another whether it be due to the size 

 of the molecule, to the relative shape of the molecule and pore, or to an 

 attraction, i:>ositive or negative, l)et'\\een the substance of the meml^rane 

 itself and that of the molecule passing. These two postulates are so self 

 evident that no explanation is necessary. When liquids are arranged with 

 a memljrane l^etween, as in any ordinary osmometer, a molecular attraction 

 comes into play, Ijut it operates only at very minute distances, hence to 

 develop high osmotic pressures it is necessary to stir the liciuids. The 

 amount of pressure which a given substance may exert will depend on the 

 specific character of the molecule, as well as upon the number of molecules 

 present. 



As this is a mere abstract, details of explanation cannot be gone into as 

 fully as might be wished. It is, however, intended that the attraction theory 

 abo\'e referred to affords a basis for a reasonaljle explanation for osmotic 

 phenomena of every character. It is, as already indicated, contended also 

 that van't Hoff's law is not a valid hypothesis when relating to osmotic 

 pressure. 



East Lansing, April 1, 1909. 



