Dec. 14, 1871] 



NATURE 



125 



No. I .... Casella . Specially made with an extra- 

 thick cylinder bulb to defy 

 compression. 



3 . . . . „ . Spherical bulb ; extra-thick 



glass. This thermometer 

 was made, at the special rc- 

 questofoncof IVtr. Casella's 

 workmen, in order to resist 

 effect by pressure. 



4 . . . . „ . Short cylinder bulb : extra- 



thick glass. 

 6 . . . . „ .A glass cup fitting over bulb, 



designed by Mr. Siemens. 



All the above were Six's thermometers with the bulbs 

 unprotected. 



No. 2 . . . . Casella . Glass-encased bulb, as pro- 

 posed by Dr. Miller, but 

 with the case nearly filled 

 with spirit. 



5 . . . . „ . Long cylinder bulb at the 



back, encased in glass, and 

 nearly filled with spirit. 



These instruments were first compared in air and then 

 immersed in a tub of water. No. 57 being placed in an 

 iron bottle. Set the indices and placed the thermometers 

 in the cylinder of the press, and pumped on a pressure 

 equal to 250 fathoms, and kept it on two hours. 



It is useless to record the result of this first experi- 

 ment ; or it may rather be stated that the results were 

 ;///, except ascertaining the weak points of the process 

 adopted. 



The Miller-pattern thermometer subsequently proved 

 so near perfection it was decided to use that as a standard 

 for the Hydrographic Office pattern.* 



It was found necessary to reduce the number of ther- 

 mometers, and also of the readings, to a minimum. 



With the view of testing the efficiency of Dr. Miller's 

 pattern (No. 2) it was placed in the cylinder with No. 57, 

 and subjected to a pressure of 4,032 lbs. (about 1,480 

 fathoms) for a quarter of an hour, with the following 

 result. 



Experiment No. i (pressure = 1,480 fathoms). 

 Dr. Miller reading. 



This experiment at once proved the efficacy of the 

 encased bulb ; and the experiment was repeated with 

 more thermometers, with the same pressure and for the 

 same period of time. 



It was found by this experiment that while the mean 

 difference of the encased bulbs was only o°'95, that of the 

 two made to defy compression was 7 "'25, that with the 

 cover 10^5, the Hydrographic Office pattern the same as 

 in No. I, 7''5, and a Phillip's Alpine thermometer 7o"-3. 



The "Phillip's" was an ordinary make, with a very 

 small bulb ; and the great difference shown by it proved 

 that the amount of compression is in proportion to the 

 thickness of the glass ; but in immediate connection with 

 the subject the experiment clearly demonstrated two facts, 

 viz. : 



1. That very nearly all the difference, or error, is due to 

 pressure on the full bulb ; and 



2. That by encasing the bulb we have nearly a perfect 

 instrument. 



Notwithstanding the satisfactory result obtained in 

 enabling us to decide on a thermometer for future use, it 

 was necessary, if possible, to establish a scale whereby 

 temperatures already taken with instruments of the Hydro- 

 graphic Office pattern might be corrected for pressure, 

 and also to ascertain if all, or wt\at part, of the difference 

 shown under pressure in the Miller pattern was due to 

 calorific effect produced by sudden compression of the 

 water in the cylinder or by compression of the unprotected 

 parts : preparation was accordingly made to continue the 

 experiments. 



It being necessary, as before stated, to reduce the 

 number of the thermometers, and also the readings, to a 

 minimum, the following were selected, viz. :— 

 Nos. 2 and J Casella . Encased bulbs. 



56 and 57 „ . Hydrographic Office pattern. 

 73 .... Pastorelli „ „ „ 



67 .... Elliott . _ „ „ „ 



9641 .... CaseUa . Alpine. 

 These were attached to a float (to avoid immersing the 

 hand in the water) and placed in the cylinder filled with 

 water, to remain all night ; the cistern, from which the 

 water is pumped into the cylinder, was filled, and also a 

 tub of water for replenishing placed by the side in order 

 that the water in each might be, as nearly as possible, of 

 the same temperature in the morning. 



The thermometers were read in the order in which they 

 are placed ; w'ncn all were read, the indices were set as 

 quickly as possible, and the instruments at once lowered 

 into the cylinder and the pressure applied. 



May 5. The;first series of experiments were made, Mr. 

 Casella reading. 



First Series of Experiments. Errors at different pressures. {Abridged from original) 



* I was not aware at that time of the existence of the enclosed Phillip's thermometer as designed by Sir Wliham Thomson. 



+ The instrument was taken out safely, but while reading off the full bulb cracked right across. 



X Broke at a pressure equal to 1,848 fathoms, 



§ This insulated thermometer is a Phillip's encased in a glass cylinder containing a little spirits designed by Sir William Thomsoi 



