PROFESSOR THOMAS GRAHAM's SCIENTIFIC WORK. 199 



lar movements in each of the two cases. A glass cj^liuder, .57 meter 

 liigb, had the lowest tenth of its height filled with carbonic gas. Then, 

 after ditTerent intervals of time, the uppermost tenth of air in the 

 cylinder was drawn off and examined. In five minntes the carbonic 

 gas in this npper tenth of air amounted to .04, and in seven minutes 

 to 1.03 per cent. ; or 1 per cent, of carbonic gas had diffused to the 

 distance of half a meter in seven minutes, being at the rate of 73 mil- 

 limeters per minute. Now, the conditions of this movement always 

 prevail in the air of the atmosphere, and, using the words of INIr. Graham, 

 " it is certainly most remarkable that in perfectly still air its molecules 

 should spontaneously alter their position, and move to a distance of 

 half a meter in any direction in the course of five or six minutes." 

 By similar experiments made with an inverted cylinder, 1 per cent, of 

 hydrogen was found to diffuse downward at the rate of 350 millimeters 

 l^er minute, or about five times as rapidly as the carbonic gas diffused 

 upward. 



With regard to Mr. Graham's experiments on the difiusion of gases 

 through porous septa, his earliest results on this subject were communi- 

 cated to the Royal Society of Edinburgh, in a paper " On the law of 

 the diffusion of gases," already referred to as the first-born of what may 

 be considered his great papers.* Prior even to Dalton's above-mentioned 

 expermients on free diffusion, Dr. Priestly, when transmitting different 

 gases through stoneware tubes surrounded by burning fuel, perceived 

 that the tubes were porous ; and that not only was there an escape of 

 the gas, under pressure, from within the tube outward to the fire, but 

 that there was also a penetration of the exterior gases of the fire into 

 the tube, notwithstanding the superior jiressure of the current of gas 

 passing through the tube. 



Mr. Graham, however, appears to have had his attention originally 

 directed to the study of the transmission of gases through porous 

 diaphragms by the curious observations and experiments of Dobereiner, 

 who, having occasion to collect and store some quantities of hydrogen 

 over water, accidentally made use of a fissured jar, and was surprised 

 to find that the water of the pneumatic trough rose in this jar to the 

 height of an inch and a half in twelve hours, and to not far short of 

 three inches in twenty-four hours. Having assured himself of the 

 constancy of the ])henomenon, Dobereiner attributed it to capillary 

 action, conceiving hydrogen to be alone attractable by, and, on account 

 of the assumed minuteness of its atoms, admissible through the fissure. 

 In repeating Dobereiner's experiments, however, Mr. Graham soon 

 observed that the escape of hydrogen outward was always accompanied 

 by a penetration of air inward, the volume of air finding an entrance 

 through the fissure amounting to about one-fourth of the volume of 

 hydrogen making its escape ; or the fissure proved permeable to the 

 grosser air as well as to the finer hydrogen. Having arrived at this 



Ediuburgli Royal Soci.'ty Trausactious, xii, 1834, p. 2'^2. 



