511 F. W. Very — Transmission of Terrestrial Radiation. 



mometer, also of limited aperture ; and the curve might be sum- 

 med zonally, were it not that such a summation could have no 

 useful significance. The really important fact is that given by 

 the lowest reading, or that in the direction of the zenith ; and 

 since there appears to be an impression among some leading 

 astrophysicists that the day sky ought to give positive deflec- 

 tions, I will call attention to- the fact that a clear sky never 

 gives any other than negative readings, whether b} r day or by 

 night ; and that on the date of the above-mentioned curve — a 

 hot, sunshiny day in July — the apparent temperature of the 

 sky was zero Centigrade. 



It is not yet sufficiently recognized that the radiation which 

 proceeds from the earth's surface, or at least that portion of it 

 which is absorbed by the atmosphere, does not continue very 

 long as rectilinear propagation in the original directions, but 

 becomes almost immediately involved in the meshes of molec- 

 ular kinetics, and thenceforth is handed on in a vertical 

 direction in connection with those processes by which the 

 normal upward change of air pressure is maintained. 



At my present station, which is not far from Boston, skies 

 good enough for measurements of air transmission are rather 

 rare in winter. This may possibly be due to the neighborhood 

 of the ocean, since, when the water is considerably warmer 

 than the land, the diffusion of water vapor from the ocean into 

 colder air over the land may very likely produce a thin mist 

 in certain layers. It is extraordinary how a scarcely perceptible 

 mist will take down the sky transmission, and, indeed, my ex- 

 periments on the transmission of radiation from a boiling 

 radiator at a distance of about 100 meters suggest that it is not 

 necessary for the atmospheric water to be actually precipi- 

 tated in liquid form in order that it shall do this. It is 

 enough that saturation be imminent, because this condition 

 conduces to the formation of highly absorbent hydrols, even 

 before actual precipitation has begun. This will be made 

 evident when my long-continued observations on these radia- 

 tions under every possible condition of atmospheric moisture 

 shall have been published. 



In summer the land is warmer than the ocean on sunny days, 

 and any vapor coming from the ocean to air over the land is in 

 a fair way to recede from the saturation point. Hence there 

 are plenty of days in summer when transmission measures can 

 be made. From long experience at the Allegheny Observa- 

 tory, I can say that, notwithstanding Pittsburgh smoke which 

 sadly dims the visible rays, nevertheless, transmission of ter- 

 restrial radiation is but little affected by the smoke, unless 

 saturation is so near that the carbon particles serve as con- 

 densation nuclei ; while the absence of the ocean is very likely 



