as co7i7iected with the Theory of Stibstitutions. 337 



comes slowly tithonized by the action of the daylight impinging 

 on it, union at last takes place. From this, therefore, we per- 

 ceive that chlorine which has been exposed to the sun will 

 unite promptly and energetically with hydrogen ; but chlorine 

 which has been made and kept in the dark shows no such pro- 

 perty. 



This form of experiment may be supposed imperfect, since 

 the chlorine is in a moist condition and confined by water. I 

 have therefore made the following variation. 



I took a tube A, fig. 4, six inches long and half an inch in 

 diameter, closed at one end and open at the other, and ce- 

 mented its open end on a piece of flat plate glass M N, one 

 inch wide and two long, ground on both sides, and having a 

 hole, p, one-sixth of an inch in diameter perforated through it. 

 This hole was not in the centre of the glass, but on one side, 

 as shown in the figure. The interior of this tube was per- 

 fectly clean and dry. 



A second tube B, consisting, as is shown in fig. 4, of two 

 portions, a wide portion B, and a narrower tube c, was ce- 

 mented on another piece of ground plate glass, similar to the 

 foregoing in all respects. The tube c was oj)en at its lower 

 extremity, and the entire capacity of B and c conjointly was 

 adjusted so as to be equal to the capacity of A. 



Next I filled A with dry chlorine and B c with dry hydro- 

 gen, and kept them from mixing until the proper time by 

 operating in the following way : — 



I placed the ground glasses face to face, as shown in fig. 5, 

 with a small quantity of soft tallow between them, arranging 

 them in such a way that the aperture which led to the inte- 

 rior of A was open. 



Through this aperture dry chlorine was conveyed. It was 

 generated by a mixture of peroxide of manganese and chloro- 

 hydric acid in the flask D, fig. 6, and passed along a tube E 

 filled with chloride of calcium. A slender glass tube/ con- 

 veyed it to the bottom of A, which was then filled by displa- 

 cing the atmospheric air. When A was supposed to be full 

 of chlorine, it was slowly lowered so as to bring the tube out 

 of the aperture, and as soon as it was disengaged the glass 

 plates were moved in such a manner by sliding them on one 

 another, that the aperture leading into A was shut, but that 

 leading to B was open. The vessel A was thus filled with 

 dry chlorine and securely closed. 



In the next place I filled B with dry hydrogen, which was 

 done as follows : — To a bottle G, fig. 7, containing dilute sul- 

 phuric acid and zinc, a drying tube, K, of chloride of calcium 

 was adjusted, and at its upper end a cork h arranged, so as to 



Phil, Mas:. S. 3. Vol. 27. No. 181 . Nov, 1 84)5. Z 



