78 Proceedings of Royal Society of Edinburgh, [jan. 17 , 
its readings. It measures, however, chiefly the intensity of the solar 
radiation that is received at the surface of the earth, as modified by 
the surroundings of the bulb. The flat black surface, on the other 
hand, gives the intensity of the solar radiation as modified by air 
currents ; the stronger the wind the lower the temperature recorded 
for a given intensity of radiation ; whereas the black bulb in vacuo 
is not much affected by wind, though it is affected by other and 
rather obscure influences * the flat black surface therefore gives a 
better indication of the climate than the black bulb. In fine 
summer weather it was found that the flat black surface generally 
read about 12 per cent, above the black bulb; but in winter this 
difference has been found to be very much greater, on account of the 
more perfect calms at that season permitting the exposed surface to 
be more highly heated. 
But further, it is found that there is a close relation between the 
difference in the readings of the two radiation thermometers and the 
errors of the Stevenson screen. An examination of the readings 
taken in December show that as the ratio of the temperature of the 
black surface to that of the black bulb increased, the error in the 
screen increased along with it. When the ratio of the black surface 
to the vacuum black bulb was under 1 - 5, the error of the Stevenson 
screen was about 1 degree, and as the ratio increased, the error in- 
creased along with it. This error attained its maximum, as far as was 
observed, about mid-day on the 13th of the month; at that hour 
the temperature of the air was 34°, the vacuum black bulb 64°, and 
the flat black surface 93°. The black bulb was thus 30° above the 
temperature of the air, while the black surface was 59°, or in the 
ratio of nearly two. At that hour the Stevenson screen reached its 
maximum error for the day, being 2° '9 above the standard. It may 
be noted that when the error of the screen was at its maximum the 
vacuum thermometer did not register the maximum temperature for 
the month ; indeed it was one of the lowest on the bright days, but 
owing to the entire absence of wind at the time, the difference in the 
heating effect of the sun’s rays on the two forms of radiation thermo- 
meters, and the error of the Stevenson screen attained their maximum. 
While on this subject, it may be as well to consider an effect of 
these two tendencies of air when heated and when cooled, — the one 
to instability and currents, the other to stability and calmness. Given 
