RECENTLY DISCOVERED GASES. 271 



very low pressure the spectrum of argon appeared. There was here, 

 therefore, no trace of a new gas to be found. 



It was not deemed worth while to investigate the ammonia, since I 

 had already prepared nitrogen out of this and Lord Rayleigh had 

 determined its density; he found this to be exactly the same as the 

 density of nitrogen from different chemical sources. It remained, how- 

 ever, possible that the sought-for gas could combine with hydrogen, 

 and that such a compound might possess an acid character; in this 

 case it might have entered into combination with the magnesium. On 

 account of the possibility that such a compound might be soluble, the 

 magnesia was extracted with water, the solution evaporated and treated 

 with sulphuric acid in a vacuum. A gas was evolved, but it proved to 

 be exclusively carbon dioxid. We should have carried the treatment 

 of the magnesium further had not the argon at last become sufficiently 

 pure to subject it to the refrigerating action of liquid air, and it seemed 

 to me there was more hope of finding the new substance in the argon 

 from the atmosphere than in this residue of magnesia, which it would 

 require much time and labor to work up. 



Dr. Hampson, the inventor of a very simple and practical machine 

 for the preparation of liquid air, which is based upon the same princi- 

 ple as that of Herr Linde, was so kind as to place large quantities of 

 liquid air at my disposal. In order to become acquainted with the art 

 of working with so unusual a material, I asked Dr. Hampson for a 

 liter; with this Dr. Travers and I practiced and made different little 

 experiments to prepare ourselves for the great experiment of liquefy- 

 ing argon. 



It seemed to me a pity to boil away all the air without collecting the 

 last residue; for, though it seemed improbable that the looked-for ele- 

 ment could be here, yet it was, indeed, possible that a heavier gas might 

 accompany the argon. This suspicion was confirmed. The residue 

 from the liquid air consisted chiefly of oxygen and argon, and, after 

 removing the oxygen and nitrogen, beside the spectrum of argon were 

 two brilliant lines, one in the yellow, which was not identical with D ;i 

 of helium, and one in the green. This gas was decidedly heavier than 

 argon ; its density was 22.5 instead of the 20 of argon. We had, there- 

 fore, discovered a new body, which was an element, for the ratio 

 between the specific heats was 1.66. To this element we gave the name 

 "krypton." Up to this time we have not followed further the study 

 of this element; we have, however, collected and preserved many resi- 

 dues which are rich in krypton. It was, however, our first intention 

 to examine the lightest part of the argon. In many, however, we 

 remarked, in passing, that the wave-length of the green line of krypton 

 is exceedingly close to that of the northern lights, being 5,570, while 

 the latter is 5,571. 



Our whole supply of argon was now liquefied in the following manner: 

 The gasometer containing the argon was connected with a series of 



