340 Prof. Spencer U. Pickering on Delicate 



comparison with a standard occupies the greater part of an 

 hour. It is possible to do three determinations at different 

 temperatures with it in a day ; but this means a very hard 

 day's work, and could only be effected after much practice. 



Although the results obtained with this thermometer may 

 appear satisfactory as measured by the average error, they 

 still leave much to be desired. It is impossible to feel 

 absolute confidence in an instrument which possesses such 

 serious imperfections ; and, apart from this, I was desirous of 

 investigating the matter further for its own sake, and pro- 

 ducing an instrument in which delicacy did not involve the 

 sacrifice even of convenience. 



It occurred to me that the substitution of a double bulb for 

 the single bulb of .No. 62839 would probably result in a 

 considerable increase in strength under pressure. Such an 

 instrument, therefore, I had constructed. Considerable diffi- 

 culty was met with in boiling the mercury in a double bulb 

 thoroughly, but this was overcome in a masterful manner, 

 while at the same time the other parts of the instrument were 

 so perfect that I believe it is the finest thermometer of its 

 kind ever constructed, in this country at any rate. The stem 

 was of the required stoutness, and the bore, while being flat, 

 was well rounded at the edges, thus exposing no crevices into 

 which the mercury would find it difficult to enter, as is usually 

 the case with flat bores, and as was so with No. 62839. The 

 length of each bulb of this instrument, No. 63616, was 55 mm., 

 the total weight of mercury in them being 30*5 grams, con- 

 siderably less therefore than in the other thermometer and 

 less than double that in the bulbs of some of the less delicate 

 thermometers. The scale was rather less open than in 62839, 

 the total range of 60 cm. representing 3°*85 C, or 154*3 mm. 

 to each degree, one estimation-figure representing therefore 

 _J ° 



3000 * m m ... 



The results of two series of experiments with this instrument 

 similar to those with 62839, last described, gave the following 

 results : — 



I. 



Registered. 



o 



Immersed after being cooled 1° 45*69 



heated 1 45"765 



cooled 1 45-685 



heated 1 45*815 



cooled 10 45-720 



heated 10 45*850 



cooled 10 45-815 



heated 10 45*947 



Theory. 



Difference. 



45-688 

 45-790 

 45-703 

 45733 

 45-768 

 45-899 



+•077 

 -•105 

 + •112 

 -•113 



+ •082 

 -•084 



Mean 



-096 



