December i, 1892] 



NATURE 



\Q>\ 



acteristic of " Bishop's Ring," and distinguishable at Honolulu 

 for two years. Apparently there has recently been a great re- 

 inforcement added to the material in the-upper atmosphere, w hich 

 produces the afterglows. 



Is this owing to the August eruption in Alaska, which is said 

 to have distributed ashes at a distance of 250 miles ? 



Prof. C. J. Lyons, in charge of tidal observations in Honolulu, 

 reports the period of highest mean tide to have extended itself 

 this year into November, or fourteen months later than the last 

 similar period. The mean sea level is now over ten inches higher 

 than it was last April. It is also somewhat higher than has been 

 shown by any previous tide registers in Honolulu. Mr. Lyons 

 regards this as of special importance, taken in connection with 

 the oscillation of the earth's axis, now established by the com- 

 bined observations at Berlin and Honolulu. 



Honolulu, November 8. Sereno E. Bishop. 



O' 



OSMOTIC PRESSURE. 



)F the various properties which have found a common 

 explanation in the new theory of solutions, there 

 are none perhaps to which more interest attaches than to 

 osmotic pressure ; and although, on account of the experi- 

 mental difificulties, the observations as yet accumulated 

 on this subject are but scanty, they have so largely con- 

 tributed to the novel ideas involved in the new theory, 

 that they merit special attention. 



Since accounts of osmotic pressure are finding their 

 way into few English text-books, it may be worth while 

 glancing at the main features which have led up to the 

 present state of the question. 



It has long been known that if an aqueous solution — 

 say, of sugar — be separated from pure water by a piece 

 of animal membrane, that movements of the water and of 

 the sugar take place through the membrane. If the 

 solution be contained in an open vessel, the base of which 

 is composed of membrane, on partially immersing the 

 vessel in water it is easy to see that more water enters 

 the vessel than solution leaves it. The level of liquid 

 within rises above that without the vessel, different pres- 

 sures being thus set up on opposite sides of the mem- 

 brane. 



To this process wherein currents pass through a mem- 

 branous septum, the terms " osmosis," " osmose," and 

 " diosmose " have been applied. The last of these is per- 

 haps to be preferred, as it serves to indicate that two 

 currents are involved in the phenomena. Investigations 

 carried out as indicated above were concerned with the 

 measurement of what was termed the " endosmotic 

 equivalent." That is the ratio of the amount of water 

 passing into the solution to the amount of dissolved sub- 

 stance passing in the opposite direction. Consistent 

 measurements of this quantity could not be obtained, 

 however, for it was found that the nature of the mem- 

 brane exercised a marked influence upon its magnitude. 

 The kind of membrane employed, or, with the same 

 membrane, its thickness or freshness, or even the direc- 

 tion in which water passed through it, was of importance. 

 Thus in illustration of the last point, water passes more 

 readily outwards through eel's-skin, more readily inwards 

 through frog's-skin. 



To obtain quantitative relations in this field it thus 

 became essential to eliminate the influence of the mem- 

 brane, and more recently this end seems to have been 

 attained by the use of membranes artificially prepared. 



These artificial membranes differ from those of animal 

 origin in the remarkable particular that although they 

 allow water to pass through, they present a barrier to 

 the passage of certain dissolved substances. On this 

 account they have been termed semi-permeable mem- 

 branes, and by their use measurements of osmotic 

 pressure have been made possible. 



To carry out such measurements the first point to be 

 solved was to obtain a membrane of sufficient strength. 



NO. 1205, VOL. 47] 



The substance which has been found to be most satis- 

 factory as a membrane-former is copper ferrocyanide. 

 When aqueous solutions of potassium ferrocyanide and 

 copper sulphate are carefully brought into contact a 

 pellicle of copper ferrocyanide is formed where the two 

 solutions meet. In this condition the pellicle is much 

 too fragile to sustain even slight differences of pressure ; 

 but by the following simple device, emplo>ed first of all 

 by W. Pfeffer, satisfactory results have been obtained. 



If a cell similar to the ordinary porous pot of a voltaic 

 battery be lowered into a solution of copper sulphate 

 while at the same time a solution of potassium ferrocya- 

 nide be poured into its interior, the two solutions meet 

 somewhere within the walls of the cell and deposit a film 

 of copper ferrocyanide. Little diaphragms of membrane 

 are thus produced stretching across the pores of the 

 cell-wall, which furnishes the necessary support, and by 

 taking suitable precautions a membrane may thus be 

 obtained capable of withstanding a pressure of several 

 atmospheres. 



The behaviour of a solution when separated from pure 

 solvent by such a semi-permeable membrane differs 

 markedly from what takes place when an animal mem- 

 brane is employed. In the latter case, at the outset 

 water adds itself to the solution ; the level of liquid and 

 the pressure on the solution-side of the membrane thus 

 rise until a maximum pressure-head is attamed, which, 

 roughly speaking, is greater the stronger the solution 

 u?ed. Seeing, however, that dissolved substance is con- 

 tinually escaping from the solution through the mem- 

 brane, as soon as the maximum is reached the pressure- 

 head begins to fall until eventually it vanishes, the 

 levels of liquid on either side of the membrane being the 

 same. 



If, on the other hand, a semi-permeable membrane be 

 employed, as before, a maximum pressure is attained ; 

 but since dissolved substance cannot leave the solution, 

 this maximum pressure as well as the concentration of the 

 solution remain constant. 



When this constant state of things is established the 

 excess of pressuie on the solution-side of the membrane 

 over that on the solvent-side, whatever it may mean, is 

 termed the "osmotic pressure" of the solution. It is 

 therefore customary to reserve the term osmose to pheno- 

 mena relating to semi-permeable membranes ; diosmose 

 being used in cases where, as with animal membranes, 

 dissolved substance as well as solvent can traverse the 

 membrane. It is obvious that when the pressure is 

 established as indicated above, the original concentration 

 of the solution has been altered by the entrance uf sol- 

 vent, and the observed osmotic pressure refers of course 

 to the solution having the final concentration. If, how- 

 ever, we imagine the vessel containing the solution to be 

 closed at the top, a quantity of air being imprisoned over 

 the solution, pressure may be set up by compressing this 

 air, only a small quantity of solvent being allowed to 

 enter. If, further, the air enclosure be tapped by a ma- 

 nometer, measurements of the pressure may be taken, 

 and by making the air enclosure and the volume of the 

 manometer small enough the quantity of solvent enter- 

 ing while pressure is being established may be neg- 

 lected, the original concentration of the solution remaining 

 practically unaltered. This is the principle of the method 

 employed in measuring osmotic pressure in absolute 

 units. 



The question now arises, " Are these measurements 

 really independent of the nature of the membrane ? Has 

 the difificulty which beset the older experiments been 

 overcome?" To this question an immediate answer is 

 for thcoming, for, as pointed out by Prof Ostwald, it 

 follows from theoretical considerations that if the mem- 

 brane employed is really semi-permeable, the observed 

 osmotic pressure of a given solution must be the same, 

 no matter of what material the membrane is com- 



