92 PROCEEDINGS OP THE AMERICAN ACADEMY 



constant temperature. The water in the outer vessel never differed 

 by more than a small fraction of a degree from that in the inner one. 



To get the and 100° points the upper parts of the vessel above 

 the line a b were removed, and ice placed around the bulb of the air 

 thermometer, and left for several hours, until no further lowering took 

 place. For the 100° point the copper vessel shown in Fig. 3 was 

 used. The portion y of this vessel fitted directly over the bulb of the 

 air thermometer. On boiling water in x, the steam passed through 

 the tube to the air thermometer. It is with considerable difficulty 

 that the 100° point is accurately reached, and, unless care be taken, 

 the bulb will be at a slightly lower temperature. Not only must the 

 bulb be in the steam, but the walls of the cavity must also be at 100°. 

 To accomplish this in this case, a large mass of cloth was heaped over 

 the instrument, and then the water in x vigorously boiled for an hour 

 or so. After fifteen minutes there was generally no perceptible in- 

 crease of temperature, though an hour was allowed so as to make 

 certain. 



The external appearance of the apparatus is seen in Fig. 2. The 

 method of measuring the pressure was in some respects similar to that 

 used in the air thermometer of Jolly, except that the reading was 

 taken by a cathetometer rather than by a scale on a mirror. The 

 capillary stem of the air thermometer leaves the water vessel at a, and 

 passes to the tube b, which is joined to the three-way cock c. The 

 lower part of the cock is joined by a rubber tube to another glass tube 

 at d, which can be raised and lowered to any extent, and has also a fine 

 adjustment. These tubes were about 1.5 cm - diameter on the inside, 

 so that there should be little or no error from capillarity. Both tubes 

 were exactly of the same size, and for a similar reason. 



The three-way cock is used to fill the apparatus with dry air, and 

 also to determine the capacity of the tube above a given mark. In 

 filling the bulb, the air was pumped out about twenty times, and 

 allowed to enter through tubes containing chloride of calcium, sul- 

 phuric acid, and caustic soda, so as to absorb the water and the car- 

 bonic acid. 



The Cathetometer. 



The cathetometer was one made by Meyerstein, and was selected 

 because of the form of slide used. The support was round, and the 

 telescope was attached to a sleeve which exactly fitted the support. 

 The greatest error of cathetometers arises from the upright support 

 not being exactly true, so that the telescope will not remain in level 



