42 REPORT— 1900. 



It appears therefore preferable to use the more general term ' radio- 

 calorimeter ' for this particular instrument, as it was not intended, like 

 the ' pyrheliometers ' of Pouillet or Angstrom, for the direct measurement 

 of solar radiation. The tube form of radio-calorimeter consists of a pair 

 of concentric tubes about nine inches long separated by an annular space 

 of about a twentieth of an inch, through which water is caused to circulate 

 in a spiral fashion by a helix of copper wire nearly fitting the space 

 between the tubes. The inner tube has a diameter of about one inch, and 

 is furnished with a series of sliding copper diaphragms, which can be set 

 at suitable points to screen off any lateral radiation, and prevent internal 

 reflection from the walls of the tube. The blackened copper disc for 

 receiving and measuring the radiation is supported near the centre of tlie 

 tube by means of the fine wires of the thermo-couple. The diameter of 

 the disc is 1-30 cm., and it is set close behind a diaphragm of 14 mm. 

 diameter, so that the whole of its surface is exposed to the radiation. In 

 this arrangement the quantity of radiation measured is determined solely 

 by the diameter of the disc and not by that of the apertures. The disc 

 can be accurately centred and directed on the source of radiation by 

 looking through a small hole at the back of the tube. The cold junctions 

 of the thermo-couples are contained in fine copper tubes soldered to a 

 sliding tube which carries the disc, and is a good fit for the inner tube of 

 the water-jacket. Water at the temperature of the laboratory is con- 

 tinuously pumped by a small motor from one large copper tank to another 

 at a higher level, and flows back continuously and uniformly through the 

 water-jacket of the radio -calorimeter. By this means the temperature of 

 the jacket is maintained very constant Avithout the necessity of making 

 the instrument itself massive or unwieldy. 



Observations with, different Coatings on the Disc. — With this apparatus 

 it was possible to obtain much more consistent results owing to the 

 greater steadiness of the experimental conditions and the greater ease of 

 adjustment and manipulation. Among other tests, some comparative 

 measurements were made of the relative efficiency of different coatings 

 of black for the disc, of which the following may be taken as samples : — 



1. Copper disc clean but not polished. Final excess 1-223° C, 

 <7=-00512, r= -00626, 1= -00379. 



2. Copper colour just visible through a thin film of smoke-black. 

 Final excess 2-528° C. (7=-00595, r=-01505, I=-00910. 



3. Copper disc covered with thick opaque film of smoke-black. Final 

 excess 2-373° C. (^=-006354, ^=-01508, 1= -0091 2. 



4. Copper disc covered with dead-black varnish of shellac and smoke- 

 black. Final excess 2-159° C. ^=-00703, 7-=-001517, I=-00918. 



5. Same disc, but with new thermo-couple, thick smoke-film. Final 

 excess 2-328° C. »7=-00642, r=-0]494, I=-00904. 



6. Same disc and couple, but thin black varnish ; back also covered. 

 Final excess 1-831° C. g=-00812, r=-01487, I=-00900. 



It will be observed in the above results that the final excess tempera- 

 ture {rjq) and the coefficient of cooling, q, vary considerably under differ- 

 ent conditions, but that the results for the rate of heating, r, and the 

 intensity of radiation, q, agree fairly well for the different coatings of 

 black. The same focus-lamp was used as a source in each case, at the 

 same distance, and it is probable that the actual variations in the inten- 

 sity of the radiatiori did not exceed one part in 500. The voltage on the 



