Analysis of Magnesium and Carbon. 347 



and in carbon) in the same chemical state in each of them, but 

 in combination with different kinds of matter, the intensities 

 of the rays due to c will be variously affected by the action of 

 these different kinds of matter. This is the reason why, for 

 instance, many rays of c are seen in the oxygen-spectrum which 

 disappear from the spectrum of magnesium or carbon and vice 

 versa ; so that without mathematical analysis it is almost im- 

 possible to recognize this element and trace it spectroscopically 

 in different substances, even when it occurs in them in the 

 same state. Lately the author has succeeded in detecting e 

 in nitrogen also, and in the same state as it occurs in oxygen, 

 carbon, and magnesium. 



The third group of magnesium-lines consists of all the lines 

 and sharp edges of bands, thirty in number, with wave-lengths 

 between 3896 and 3329*1 observed by Liveing and Dewar 

 and by Hartley and Adeney, and, according to our author, 

 forms part of the large group of lines in the hydrogen- 

 spectrum produced by the primary element b in the chemical 

 state in which it occurs in hydrogen under the influence of a 

 or "coronium." The wave-lengths of this group, when mul- 

 tiplied by 4/5, give the wave-lengths of corresponding lines 

 of the water-spectrum due to b as it occurs in hydrogen in 

 water ; and when multiplied by 46/41 and by 70/59, give the 

 wave-lengths of lines of oxygen due to b as it occurs in 0' in 

 oxygen and as it occurs in 0" in 0' in oxygen, respectively ; 

 and when multiplied by 23/32x46/41 and 21/32x70/59, 

 give the wave-lengths of corresponding groups of lines in the 

 water-spectrum. None of the group have as yet been ob- 

 served in the spectrum of hydrogen, but several of them have 

 been observed in that of the oxyhydrogen-flame. As regards 

 the water-spectrum, the agreement of the observed with the 

 calculated lines is close and very complete throughout the long 

 list of some ninety lines. As regards the oxygen-spectrum, 

 only a few of the calculated wave-lengths correspond to lines 

 of which observations are recorded. The existence of these, 

 hitherto unnoticed, oxygen rays is therefore an inference 

 from the existence of the rays of water which are harmoni- 

 cally related to them. 



The fourth group of magnesium -lines includes 3336*2, and 

 all the lines more refrangible than it observed by Liveing 

 and Dewar and by Hartley and Adeney. This group forms 

 part of the large group of lines of the water-spectrum due to 

 b as it occurs in hydrogen in water- vapour ; and for nearly 

 every one of the fifty lines of the group, a line with the 

 same wave-length within one unit of Angstrom's scale has 

 been observed by Liveing and Dewar. What differences of 



