no. 3 



RADIATION OF THE ATMOSPHERE ANGSTROM 



71 



Instead of being side by side, the strips are here placed one above the 

 other beneath a thin horizontal plate of brass. When the instrument 

 was in use, a blackened screen was placed beneath it, so that the 

 lower strip was exchanging radiation only with this screen, which 

 subtended a hemisphere. The upper strip was exchanging radiation 

 with the whole sky. The radiation was calculated from the current 

 necessary to heat the upper strip to the same temperature as the 

 lower one. 



Even in the use of this instrument in its original form, it is difficult 

 to avoid some systematic errors. One is due to the difficulty of pro- 

 tecting the screen with which the lower strip exchanges radiation, 

 from absorbing a small fraction of the incoming radiation and in this 

 way giving rise to a heating of the lower strip. And secondly the 

 convection is apt to be different, the effect of rising air currents being 

 greater for the upper strip than for the lower one. The error in- 



Table XI — Radiation of the Sky 



Before sunrise 



Noon 



After sunset 



Total sky radiation.-. . 



Sept. 5 



— 0.169 

 +0.062 

 —0 . 208 

 +0.250 



Sept. 6 



—0.20S 



+0 . 092 



—0.225 

 +O.307 



Sept. 7 



—0 . 208 



+0.047 

 — 0.220 

 +O.261 



Mean 



—O.I94 

 +O.067 

 —O.218 

 +O.273 



troduced by these causes may possibly amount to 10 or 15 per cent. 

 In this instrument as well as in the original Angstrom instrument, 

 the error, when we attempt to measure the sky radiation during the 

 day, tends to make this radiation appear weaker than it really is. 



Table XI gives some results of observations with the last named 

 instrument, taken by Dr. Abbot and the author. My measurements 

 of the nocturnal radiation during the preceding and following nights 

 are given in the same place. The total diffuse sky radiation is calcu- 

 lated on the assumption that the effective temperature radiation dur- 

 ing the daytime is a mean of the morning and evening values deter- 

 mined by the nocturnal apparatus. The sky was perfectly uniform 

 during the observations but was overcast by a faint yellow-tinted 

 haze, ascribed by Abbot to the eruption of Mount Katmai in Alaska. 

 The energy of the direct solar beam at noon was, for all three days, 

 1.24 to 1.25 cal. The sun's zenith angle at noon was 32 . From the 

 table it may be seen that there was always an access of radiation from 

 the sky, indicating that the diffuse radiation from the sky was always 



