i 74 DIFFUSION, OSMOSIS, AND FIITRATION. 
funnel was proportional to the difference of concentration of the solu- 
tions on either side of the membrane. 
Jolly 1 specially studied the ratio between the amount of water 
passing into the solution and the amount of dissolved substance passing 
out, using salts with pig's bladder as membrane. This ratio he termed 
the endosmotic equivalent of the salt, and maintained that it is constant 
for the same membrane, concentration of the salt solution, and tem- 
perature. For some years after this the whole attention of those interested 
in the matter of osmosis was directed to a fuller study of this ratio in 
the case of different substances. 2 
As a result of these researches, it was seen that even with the same 
membrane it was only within slight changes of concentration of the 
solution that constancy of the endosmotic equivalent was obtainable, a 
result in accordance with expectation, seeing that the physical nature of 
an animal membrane must necessarily undergo change with the amount 
of water imbibed, a quantity variable with the concentration of the solu- 
tions in which it is in contact. With a strictly semipermeable membrane, 
the endosmotic equivalent is evidently infinite, while the more permeable 
the membrane to dissolved substance the lower will be the equivalent. 
Thus, according to Harzer, 3 the endosmotic equivalent for sodium chloride 
is with fish-swim-bladder, 2'9 ; ox -pericardium, 4 - ; ox-bladder, 6*4. 
It must therefore be admitted that, in spite of the great labour that 
has been expended on the determination of endosmotic equivalents of 
different substances with different membranes, the results obtained are 
of little value to the practical physiologist, who deals with membranes 
in the living body, whose physical characters are by no means necessarily 
those of the structures used in such experiments. The only value that can 
be attached to these determinations is an orienting one, as to the diffusi- 
bility of the substances into water, through dead animal membranes, 
under the conditions of the experiments. 
Before we can attempt to answer the question, How is the process of 
diffusion modified when in an osmose experiment an animal membrane 
is placed between solution and solvent ? it is obviously necessary to 
know the physical structure of the membrane. Of this we must admit 
great ignorance. To Brticke 4 we owe a theory of "pore diffusion." 
Assuming capillary pores in the membrane, it maintains that, by 
attraction, a layer of pure water lines these, while an axis of salt solu- 
tion, whose concentration falls from axis to mantle of the cylindrical 
pore, lies centrally. The highest concentration in the axis must be that 
of the salt solution in the experiment, and along the axis ordinary 
hydrodiffusion takes place, water entering the salt solution and salt 
entering the water. Along the mantle, however, only water can pass 
into the salt solution, so that the stream of water exceeds that of salt. 
If the pores are very narrow, it is conceivable that there is no central 
core of salt solution, in fact the membrane becomes semipermeable. 
1 Ztschr. f. rat. Med., 1849, Bd. vii. p. S3 ; Ann. d. Phys. u. Chen., Leipzig, 1849, 
Bd. lxxviii. S. 261. 
2 Fick, Untersuch, z. Naturl. d. Mensch. u. d. Thiere, 1857, Bd. iii. S. 294 ; W. Schmidt, 
Ann. d. Phys. u. Chem., Leipzig, 1857, Bd. cii. S. 122; Beitr. z. Anat. v. Physiol. 
(Eckhard), Giessen, 1855, Bd. i. S. 97 ; 1S60, Bd. ii. S. 1, 31, 147 ; Hoffmann, ibid.. 
1860, Bd. ii. S. 59. 
3 Arch.f.physiol. Heilk., Stuttgart, 1856, Bd. xv. S. 194. 
4 " De diffusione lmmorum per septa mortua et viva," Berlin, 1S42 ; Ann. d. Phys. u. 
Chem., Leipzig, 1S43, Bd. lviii. S. 77. 
