12 ANNUAL RECORD OF SCIENCE AND INDUSTRY. 



to the eye by printing from his original negatives the juxta- 

 posed spectra of the sun and of the gas. In the photograph 

 can be Been each bright line of the oxygen spectrum matched 

 and prolonged by a corresponding bright line in the spectrum 

 of the sun. 



" While the metallic elements reveal themselves by dark 

 lines in the solar spectrum, oxygen shows bright, and this is 

 probably the reason why it has so long remained concealed; 

 for these bright lines or bands which indicate its presence 

 arc inconspicuous and easily mistaken for mere unoccupied 

 intervals between the multitudinous dark lines which abound 

 in the portion of the spectrum where they are found. The 

 attempt has always been to identify some of the dark lines 

 of the solar spectrum with those of the element in question, 

 and the bright bands escaped investigation until the photo- 

 graph brought out their significance. Why oxygen should 

 behave thus differently from the other substances before de- 

 tected it is not yet possible to say with certainty. A pos- 

 sible explanation is that its very abundance has hidden it. 

 A gaseous substance, sparingly present in the solar atmos- 

 phere, would declare itself by faint dark lines in the solar 

 spectrum ; but if the quantity of the gas should be gradually 

 and continuously increased, these dark lines, after growing 

 for a time more intense, would then by degrees fade away, 

 and when the quantity of the gas had become sufficiently 

 great, would be replaced by bright ones. But on this hy- 

 pothesis it is very difficult to understand why oxygen is not 

 conspicuous in the chromosphere; like the substance which 

 gives the so-called D 3 line, brilliant in the chromosphere 

 spectrum, but invisible (usually) in the spectrum of the 

 suifs surface. Probably the full explanation lies somewhat 

 deeper." 



Professor Langley, of Pittsburgh, publishes in the Monthly 

 Notices, K. A. S., a paper on the measurement of the direct 

 effect of sun-spots on terrestrial temperature. It is not in- 

 tended to show that the earth is, on the whole, cooler in 

 maximum sun-spot years, as the discussions on the paper 

 (a- reported in the Astronomical Register) indicate it to have 

 been misinterpreted to mean. The observations consisted in 

 measuring the relative amounts of umbral, penumbral, and 

 photospheric radiation. The relative umbral, penumbral, 



