288 SCIENTIFIC RECORD FOR 1884. 



size of the thermometer bulb and of the inclosure is also important. 

 [The complete theory of the Arago-Davy conjugate thermometers or 

 actinometer has been worked out by Professor Ferrel and is fully 

 given in his recent memoir on "Temperature of the Earth's Surface."] 

 [Nature, xxix, p. 208.) 



118. J. T. Bottomly communicated to the Montreal meeting of the 

 British Association for the Advancement of Science, the preliminary 

 results of a course of experiments on the loss of heat by radiation and 

 convection for bodies of different dimensions. His experiments are 

 made on wires of different sizes both covered and bare, both in ordi- 

 nary air and very rarefied air, the wire being heated by an electric cur- 

 rent; he finds that, other things being the same, the smaller the wire 

 the greater the emissivity ; for a diminution of air pressure the emis- 

 sivity decreases slowly down to one-half or one-third of the ordinary 

 air i)re6sure, after which it becomes very great as the vacuum becomes 

 more complete. {Nature, xxx, p. 523.) 



119. Dr. Maurer, of Zurich, contributes a study on the application of 

 air thermometers to meteorological observations. He employed a thin 

 polished cylindrical copjier vessel 5""' in diameter and 44<"^ long, and 

 also a thin brass cylinder 10*"" radius ; in each receptacle was hermet- 

 ically sealed a normal mercurial thermometer and the copper vessel 

 then filled with i^ure dry air. This was then established in a double- 

 louvre thermometer shelter of sheet zinc in which were also two mer- 

 curial thermometers, the whole being constructed entirely in accordance 

 with the results of Professor Wild's investigations into the proper 

 method of thermometer exposure for thedetermination of the tempera- 

 ture of the air. 



The two free thermometers at a distance of 25<"" from each other 

 agreed always within one or two tenths of a degree, but with regard 

 to the thermometers within the metallic receptacle the following may 

 be noted: If the temperature variations are slight and always slowly, 

 steadily, rising or falling, then the inclosed thermometers follow these 

 changes quite rapidly ; if ttie temperature changes are rapid and large, 

 followed by a constant temperature, the receptacles also follow closely ; 

 but if the changes are rapid in opposite directions, as happens in thunder- 

 storms, partly cloudy weather, or alternating calms and winds, then 

 there is no agreement whatever between the free and the inclosed ther- 

 mometers, differences of one or more degrees being quite common. He 

 therefore concludes that all records with such self- registering air ther- 

 mometers must be accepted with great caution, as they frequently give 

 results that have no scientific value. From a theoretical point of 

 view it can be easily shown, as he has done, that the conduction of heat 

 alone will bring the thermometers to equilibrium in about four minutes 

 if inclosed within a brass sphere as above and exposed to a sudden 

 change of air temperature of one degree. {Z. 0. G. M., xviii, p. 334.) 



120. Dr. 0. Lang, of Munich, describes the simple method of deter- 



