176 PROCEEDINGS OF THE AMERICAN ACADEMY. 



pended from the periphery by four stout phitinum wires symmetrically 

 placed, which in turn were made fast to another hard rul)ber tu])e (f). 

 A hard rubber pulley and suitable bearing (e) were fastened to the tube 

 (f ) and permitted the mechanical rotation in the horizontal plane of this 

 part of the system. Small platinum wings attached to the outside 

 of the receptacle below caused the rotation to stir thoroughly the 

 surrounding acid. The alkali itself within the very small can likewise 

 needed agitation, which was effected by a tiny platinum stirrer (s) 

 suspended by a hard rubber rod within the tube (f). Upon the effec- 

 tiveness of the stirring within these receptacles depended the speed 

 with which thermal homogeneity could be established, and the arrange- 

 ments were found to be very effective for this purpose. 



The platinum bottle was immersed, as has already been stated, 

 within the liquid to be studied in a large platinum calorimeter capable 

 of holding about 0.7 liter. 



The jacket arrangement for obviating the cooling correction should 

 also be described in detail, as it differs from that used in the last inves- 

 tigation of this kind. As in other recent calorimetric work done in 

 this laboratory, the whole calorimeter and its appurtenances were 

 enclosed in a water-tight vessel, with tubes above for the necessary 

 connections. This vessel was immersed in a pail of dilute alkali. 

 As the reaction within the calorimeter progressed, acid was added to 

 this alkali, so that the change of temperature in the surrounding pail 

 should parallel that in the calorimeter itself. The pail (A) was made 

 of copper and furnished with a powerful propeller-stirrer (J). The 

 water-tight "submarine" vessel containing the calorimeter was 

 furnished with a cover clamped in place and made water-tight by a 

 smooth rubber washer. The submarine vessel was nickelplated and 

 polished inside, and was large enough so as to allow a narrow air 

 space between it and the calorimeter, from which it was kept apart 

 by small wedges of cork. Through suitable tubes at the top of the 

 "submarine" were introduced not only the thermometer and the 

 hard rubber tubes already mentioned, but also the rods bearing the 

 two-stage platinum stirrer which stirred the calorimeter itself, thus 

 equalizing the temperature in the liquid immediately surrounding the 

 wide mouthed platinum bottle. The hard rubber stirrer rods were 

 as small in diameter as possible ; for large rods cause a pumping effect 

 and thus introduce outside air into the calorimeter. Great pains 

 must be taken to prevent the circulation of air through these tubes, 

 for the adjustment is so delicate that if the temperature of the room 

 varies considerablv from that within the submarine, a distinct effect 



