712 Miss Agnes M. Clerke 



bulbs, is theoretically about the same obtained by boiling out a space 

 with mercury. Practically, it turned out to be even higher. In 

 carefully prepared tubes, vacua so nearly perfect were produced that 

 heat had to be applied before a spark could be got to pass. Their 

 spectroscopic examination gave results of singular interest. Carbonic 

 oxide bands were generally present, but might be traced to emanations 

 from the glass ; they were associated with lines of hydrogen and 

 helium, and the distinctive yellow ray of neon. The path thus struck 

 out was pursued further in August 1900, when, by an improved 

 process, some tubes were filled at low pressure with the more volatile 

 gases of the atmosphere. Traces of nitrogen, argon, and carbon- 

 compounds, having been abolished by a bath of liquid hydrogen, 

 sparking brought out prominently the spectra of hydrogen, helium, 

 and neon, together with a number of less brilliant rays of unknown 

 origin. Excited by continuous electric discharges, tubes thus 

 prepared glow throughout with a strong orange light. The violet 

 and ultra-violet sections of the spectrum given by it seem, neverthe- 

 less, to rival the strength of its red and yellow rays, so far as could 

 be judged from spectrographs evidence. Sensitive plates received 

 impressions of great intensity up to a wave-length of 314, despite the 

 opacity of glass to such rapid vibrations. The photographs, it is true, 

 were taken with a quartz calcite train, but there was still the glass of 

 the tubes to be reckoned with. 



The wave-lengths of nearly three hundred rays in the spectrum 

 derived in this manner from residual atmospheric gases, uncondensed 

 at the temperature of liquid hydrogen, were measured by Professors 

 Liveing and Dewar, the spark-spectrum of iron serving for a standard 

 of reference. Of these, 69 were identified, certainly or probably, as 

 emanating from hydrogen, helium or neon, and it was noted, as a fact 

 of no small significance, that among them were four members of the 

 ultra-violet hydrogen series. For, under ordinary circumstances, 

 they are emitted only by gas carefully purified ; yet here they 

 emerged with comparative facility on plates exposed to light from a 

 heterogeneous mixture. An unexpected hint is thus afforded regarding 

 the conditions that may tend to modify the hydrogen-spectrum in 

 passing from star to star. Coincidences were diligently sought 

 among the unknown lines with nebular, coronal and auroral rays, 

 but with dubious or partial success. The possibility, however, was 

 not excluded that "nebulum" might actually lurk, almost infini- 

 tesimally, in the earth's atmosphere, since, from one tube which, 

 owing to its somewhat different treatment, preserved traces of 

 nitrogen and argon, a faint additional ray was derived, agreeing 

 approximately in position with the chief bright line of gaseous 

 nebulas at X 5007. Further observations were contemplated for the 

 verification of this curiously interesting suggestion. A good many 

 subordinate lines in the tube-spectra fell very near the places 

 assigned to radiations from the sun's corona ; yet here, again, con- 

 firmation was needed before the terrestrial presence of coronium 



