266 



NATURE 



\Au^: I, 1872 



indication. This will be readily comprehended. Fig. 2 

 (p. 267) represents the bulb of a thermometer exposed 

 to the rays when the sun's zenith distance is 65 ; 

 Fi^. ,1 representing the bulb when the zenith distance 

 is iS' 23', the latter being the minimum at the 

 Observatory of the Roman College, where the thermo- 

 heliometer has been long employed for the purpose of 

 ascertaining the intensity of solar radiation. Referring 

 to Fig. 2, it will seen that the blank crescent c, whose 

 varying thickness indicates very nearly the amount of 

 heat imparted at each point of the spherical surface pre- 

 sented towards the sun, occupies a nearly vertical posi- 

 tion Tlie mercury contained within the space indicated 

 by the said crescent, having its specific gravity reduced 

 by the radiant heat, will ascend ; while the mercury on 

 the opposite side, which retains its specific gravity, will 

 descend ; thus a circulation will be established hv menns 



of which the heat received from the sun will be gradually 

 communicated to the entire mass of mercury in the bulb. 

 But, when the latter is exposed to the sun's rays under a 

 zenith distance of about 18^, as shown in Fig. 3. the heated 

 mass of mercury contained within the crescent a has so 

 slight an inclination that scarcely any circulation t.akes 

 place. Consequently, if it can be shown practically that 

 mercury is incapable of transmitting heat from particle to 

 particle with sufficient velocity, it will be evident that 

 thermometers and thermohcliometers, with spherical 

 bulbs are worthless as means of measuring the intensity 

 of solar radiation. It will be perceived that if the bulb 

 in Fig. 3 be surrounded by an enclosure, as in the thermo- 

 heliometer, the mercury contained within the space indi- 

 cated by the crescent b will radiate far less heat towards 

 such enclosure than the mercury within the opposite 

 heated crescent i7. It will also be perceived that by in- 



creasing the size uf the bulk the transmission of heat from 

 a to /' will be retarded unless the conductivity of mercury 

 be perfect. Hence the si-c of the bulb is an element 

 affecting the accuracy of the indication — a circumstance 

 fatal to the employment of a spherical bulb in the thermo- 

 heliometer. 



The nature of the illustrated apparatus constructed for 

 the determination of the conductivity of mercury will be 

 readily understood from the following description : — Fig. 

 I represents a longitudinal section through the vertical 

 plane, a is a boiler, with a flat bottom and semicircular 

 ends, supported on tivo columns, / and i,--, resting on the 

 bottom of the cisterns c and d. The column / is com- 

 posed of wrought copper plated with silver, highly 

 polished. The column /^ consists of a cylindrical vessel 

 of glass open at the top, filled with mercury, and sur- 

 rounded with a socket, /;, composed of polished silver. The 

 cisterns c and d, supported on nonconducting substances, 

 are plated with polished silver, and provided with funnels 



j shaped openings at the top, through which thermometers 

 are inserted. These cisterns, as well as the columns /and 

 •;, are surrounded with nonconducting coverings, /,/5, and 

 0,0. A lamp, b, is applied behind the cisterns for heating 

 the water in the boiler. It is scarcely necessary to ob- 

 serve that the pohshed silver plating of the copper column, 

 and the polished silver socket round the mercurial column, 

 are intended to prevent loss of heat by radiation, while the 

 coverings before mentioned are intended to prevent loss 

 of heat by convection attending atmospheric currents. 

 The inside diameter of the cylindrical vessel t,-", it should 

 be noticed, is o'5 in., corresponding exactly with the dia- 

 meter of the copper column/, thetop of which is on a level 

 with that of the mercurial column. The lines k I and 

 m n are in the same horizontal plane, their distance below 

 the upper ends of the columns/ and .;'• being precisely two 

 inches. 



The object of the apparatus being that of comparing 

 the conductivity of m.ercury to that of some other metal 



