1882.] 



relative Thermal Intensity of the Sun. 



339 



A, B are the two bulbs, 20 millims. in diameter, one of which, A, is 

 blackened in the usual way, and then sealed into the larger clear glass 

 globe, C, which has a small neck at c for attachment to a Sprengel 

 pump. As soon as a good vacuum has been obtained, the neck is 

 sealed off before the blowpipe, as shown in the figure. The other bulb, 

 B, is shaded by a small zinc arch, /, 3 inches long, painted on both sides 

 with " flatted " zinc- white. These bulbs are connected by a tube bent 

 twice at right angles and furnished at d with a branch and stopcock. 



The tube from the bulbs to e e is of the diameter of that of a self- 

 registering spirit thermometer, but the remaining part of it is much 

 wider, in order to diminish the friction of the column of mercury 

 moving in it. The upright tube attached to B is provided with a scale, 

 and the usual steel index is enclosed in the capillary part of it. The 

 length of the scale is quite independent of the capacity of the bulbs, 

 provided the volume of that part of the capillary tube in which the 

 mercury oscillates is nearly a vanishing quantity in relation to the 

 volume of the bulbs. It is well to have the two bulbs of nearly the 

 same size, but a difference of capacity does not interfere with the 

 accuracy of the instrument. The difference of the level of the 

 mercury in the two limbs will be exactly proportional (neglecting the 

 volume of the capillary tube) to the difference of temperature in the 

 two bulbs, and the degrees of the scale, g, will therefore be equal 

 throughout. The length of these degrees, however, though constant 

 for any one instrument, will vary with the temperature and pressure 

 at which the instrument has been filled, being greater the lower the 

 temperature and the higher the pressure. At 0° 0. and 760 millims. 

 mercurial pressure, the difference of mercury level corresponding to 

 1° 0. difference of temperature would be 2*784 millims., and the full 

 length of these degrees may be practically obtained by making the 

 wide portion of the tube longer in the limb A e than in B e, so that a 

 minute depression of the mercury in A e will cause a great rise of the 

 column in B e. The relative capacities of the capillary and wide tubes 

 can be readily determined, and the necessary correction made in the 

 readings for the depression of the mercury in A e. 



The instrument may, however, be so constructed as to make the 

 mercury rise and fall equally in both capillary tubes ; in which case the 

 rise in one limb would be accompanied by a corresponding fall in the 

 other, and consequently the indicated degrees on the limb B e would be 

 only half as long, 1'39 millims. corresponding to 1° 0. — a graduation 

 which is sufficiently open for all practical purposes. As a greater 

 difference than 60° C. between the shade and sun thermometers has 

 never been yet observed, a scale 163 millims. long in the one case, or 

 half that length in the other, is sufficient, except perhaps for obser- 

 vations at very great altitudes. 



As shown in the figure, the bulbs are supported upon a firm table, 



2 b 2 



