548 REPORT — 1886. 



of the i-ano;e, and so prevent the possibility of overwinding. It also carries an 

 index which moves along a graduated scale and counts the turns of the wire on the- 

 insulating cylinder. 



Id Joliu's rheostat, as already described, the two cylinders are geared together 

 directlv, and turn in contrary directions, the wire passing from the upper side of 

 one to" the under side of the other. In Sir William Thomson's instrument, as is 

 seen in the diagram, the toothed-wheels of the two cylinders are turned in the same 

 direction by the wheel on the screw shaft, and the wire passes horizontally from 

 the top of one cylinder to the top of the other. 



The conducting cylinder and the wire are both of platinoid, a metallic alloy 

 having properties which make it specially suitable for the purpose. It has very 

 hifh electric resistaDce, veiy small temperature variation of resistance, and it 

 remains with its surface almost or altogether untarnished in the air. On account 

 of the last-named property the contact between the wire and the conducting cylinder 

 is as perfect as can be desired ; and continuity of action, which was a gi'eat ditfi- 

 culty in the old form of the instrument, is absolutely complete. 



13. Description of Experiments for determining the Electric liesistance of 

 Metals at High Temperatures' Bij J. T. Bottomlet, M.A., F.B.8.E. 



This paper gives an account of apparatus and experiments for the purpose of 

 determining the electric resistance of certain platinum wires at temperatures varying 

 from 0° C. up to the temperature of dull redness, or even a little higher. The wires 

 experimented on were wires which have been used and are in use in an investiga- 

 tion on heat radiation at different temperatures in vacuum, and in air and other 

 gases. The necessity for knowing with accuracy the rate of variation of electric 

 resistance with temperature of the wires was explained in previous communica- 

 tions (1884 and 1885) to Section A of the British Association. The variation of 

 electric resistance of platinum with temperature differs so much in different speci- 

 mens that it is necessary to determine its value for each particular wire which is 

 used in the investigation just referred to. 



The principal apparatus described are an air-thermometer and a copper heating 

 iacket ; but a somewhat detailed account is also given of other heating apparatus 

 employed, and of the electrical arrangements. 



The air thermometer used is formed from a piece of thin glass tube from | in. 

 to i in. in internal diameter. This is allowed to fall in before the blowpipe and 

 drawn out to a capillary tube at each end ; and one of the ends (n) is conveniently 

 turned up along the side of the bulb. The length of the bulb is from 2 in. to 2^ in. 

 The thermometer is filled with perfectly dry pure air with the help of an aspirator, 

 and the capillary tube a is closed by drawing off a small portion by means of a 

 finely pointed blowpipe flame, care being taken that the blowpipe flame does not 

 play on the open end, which might contaminate the air within. When a number of 

 these bulbs are made and filled at one time, the end b is also closed, after the filling 

 with pure air, and the thermometers are then ready for use. 



When the thermometer is to be used it is placed in position, and very soon after 

 the heating commences the end b of the capillary tube is opened with a file. When 

 the bulb has taken the temperature which is to be measured, a hand blowpipe is 

 brouo-ht and b is closed again. The thermometer having been removed from the 

 hot space, it is allowed to cool, is carefully cleaned, if need be, and is then weighed. 

 The weighing so obtained (after a small correction for air contained which it is 

 usually unnecessary to apply) gives the weight of the glass. The thermometer is 

 then immersed under mercury or water which has been boiled and cooled (water 

 is preferred by the author for reasons stated), and the extremity of the capillary 

 tube a is cut off with a glass-knife. The water entering fills the space with the 

 exception of that occupied by the air which was left in the thermometer at the 

 hio-h temperature. A second weighing is then taken. Lastly the end b is opened 

 and the thermometer is filled with water, and a third weighing taken. The small 

 portions of glass cut off are of course weighed with the thermometer, and care is 

 taken as to drying the outside of the thermometer for weighings 2 and 3. The 



