76 Dr. Meyer Wilder man on the Velocity of 



part of the tubes, and the four joints of platinum and 

 manganin are immersed in the liquid and kept at the same 

 temperature. Though the thin copper case assumes the 

 temperature of the liquid very readily, having a very large 

 surface of contact, and the air-space is only 4 mm., direct 

 investigations have shown that the velocity with which it 

 assumes the temperature of the liquid is only sufficient for 

 the investigation of freezing-points and boiling-points, but 

 not for the velocity of reaction. Better results are obtained 

 when the platinum wire is covered with a thin layer of 

 shellac, and the copper case is filled with liquid paraffin, 

 space being allowed for the expansion of the coil. 



Fig. Pt t shows the most sensitive pyrometer ; the platinum 

 spiral /3 is drawn on two pieces of mica which are fixed 

 perpendicularly to each other, and the platinum wire directly 

 immersed in the liquid. The compensating-leads and the 

 ■current-lead were made each of four thin wires of manganin, 

 soldered to four large flat pieces of copper, and their resis- 

 tance at 23°* 2 was 0*6607 and 0'6601 ohm. The resistance of 

 the pyrometer was, directly after the annealing, 19*3002 ohms 

 (at 23°' 2 ?) . It was standardized for the range of temperature 

 for which it had to be subsequently used ( — 5° to about + 60°) , 

 readings on my bridge of zero deflexion having been taken, 

 while the thermometer was immersed in 4| litres of water, 

 which were allowed gradually to cool, and the temperature was 

 simultaneously read on a standard mercury thermometer. 

 The rest of the construction is seen from the diagram. The 

 thermometer is protected from the stirrer by a glass jacket C, 

 consisting of four U-shaped rods fixed to the piece of ebonite 

 on the top of the thermometer, as given in PI. I. To avoid 

 leakage of electricity the platinum coil was coated with a 

 solution of indiarubber in chloroform until tests with a very 

 good conducting liquid showed that the coil was sufficiently 

 insulated. Thus one experiment gave the following results: — 

 The obtained deflexion was, on my galvanometer, when the 

 Pt-Qoil was immersed in a concentrated NaCl solution con- 

 nected with one lead of two dry cells, and brought near the 

 other lead of cell — a copper wire — 0"5 cm. for each of the 

 two leads of the Ptf-spiral, and 0*2 cm. for each of the two 

 leads forming the compensating-leads ; the deflexion is here 

 the greatest ; lm 00 q ampere gives on my galvanometer a 

 deflexion of 9'5 mm. The velocity with which this thermo- 

 meter assumes the temperature was measured on a photo- 

 graphic curve, and found to be not smaller than the velocity 

 with which the galvanometer-mirror approaches its point of 

 rest when once deflected. 



