1 2 Mr. J. Herapath on (lie irregular 



meriting room, I formed about three hours before we began 

 all round the glass, but not touching it, a close thick wall of 

 books, rising from two to three inches above it. At 8 h 22', 

 the mercury was 46° 23', and at 9 h 10' the air 46°-13. At 

 this time we commenced our experiments, going through the 

 usual process of heating and cooling the bulb (to 46°'2) before 

 taking each temperature after the first. 



Wishing to know whether the air and mercury for a short 



time afterwards varied their temperatures much, I observed 



with the same instrument, without any heating or cooling, at 



9h 32' the mere, was 46°'33 the air 46°'13 



9 37 ....'. 46-40 • ' 46-80 going out and 



returning at 9 52 46 '53 . . 46 -80 



Exp. 15 and 16. Another experiment made with a closer 

 and higher wall of books gave a similar result *13. Varying 

 the experiment so as to measure the temperature of the mer- 

 cury with the thermometer in the same state a second time, 

 before each repetition of heating, had the advantage of show- 

 ing the real alteration of the mercurial temperature at equal 

 intervals, during the course of repetitions. No difference 

 however having been found in the general value of anomalous 

 indication, it is unnecessary to enter into details. 



In this last experiment I used water for heating the ther- 

 mometer, which was only 1 06° ; in which I put the bulb, and 

 let it remain for about 5 minutes to see whether any difference 

 in the result would arise from the water being of a less tem- 

 perature than before. The difference of indication came out, 

 however, nearly the same, '17 or rather more. 



The mean difference of indication in all these experiments, 

 after allowing in the 12th for the advance of temperature, is 

 about -174; the air and thermometrical tube being about 

 50 o- 7, or in round numbers 50°. 



During the making of the preceding experiments I made 

 one or two efforts likewise to ascertain whether any difference 

 existed in the indications, by first immersing the bulb in 

 water of 105° with the stem oblique, and then raising it to a 

 perpendicular ; but, for want of a steady temperature in the 

 water, I could not succeed. 



On inquiring into the cause of the phscmomena I have de- 

 tailed, 



