[BARNES] TEMPERATURE RECORD OF NOCTURNAL RADIATION 133 



ture of the trace from the line of equal temperatures is shown. The 

 return of the trace to the zero line and over to the warm side shows 

 when the sun's rays came on to the instrument. 



Before commencing the traces on the flat thermometers at the 

 rhvsics Building, a number of zero traces were obtained which shows 

 the small inequalities of air currents on the Hat thermometers. Fig. 5 

 illustrates this very well. Figs. 6 and 7 show characteristic traces 

 with glass covers over the boxes. The wooden cover was open over 



..^r-^^ 



y^ ^6 yT yS 



Fig. 5. Zero Trace. 



thermometer P, and the trace travelled to the left. The maximum 

 cooling effect was 2° Cent. Figs. 8 and 9 show direct radiation 

 traces with the glass plates removed, the maximum cooling is seen to 

 be slightly less than 4° Cent., 7-2° Fahr. The next three Figs., 10, 

 11 and 12, are similar traces, illustrating the different conditions of 

 clearness of the sky. In every case the time of opening the wooden 

 cover is clearly shown and was generally between 5 and 6 o'clock in 

 the afternoon, when the sun was shadowed by the building. Usually 

 the cover was closed again in the early morning to avoid the effect of 

 the sun's rays. In Fig. 11 the sun had affected the readings before 

 the cover was closed. 



/Vûcn/rna/ Rad/ahon. 



Fig. 6. Radiation Through Glass Plate, 1 mm. Thick, Air Temperature 

 O'' Fahr. Sky very Clear. 



Fig. 13 illustrates the screening effect of glass and a saturated 

 solution of salt. The glass covers were placed over the thermometers 

 and the wooden cover of P opened at 7.45 p.m. The departure of 

 the trace from the zero line previous to this was caused by some adjust- 

 ments at the recorder, and does not indicate a displacement of the 

 zero. At the time indicated on the diagram by the line B, a wooden 

 box, provided with a glass bottom of similar thickness to the glass 



