and the Line-spectra of Hydrogen and Oxygen, 365 



nizable as such, provided that the elementary O-spectrum is 

 known sufficiently exactly, which is unfortunately not the 

 case with very faint rays. 



We are thus in possession of mathematico-spectroscopic 

 criteria for the hydrogen-rays, which possess an unmistakable 

 resemblance to the chemical reactions for the detection and 

 recognition of substances. 



III. 



The Dissociation of Hydrogen in the Sun's Atmosphere. 



A careful comparison of the lines of the above-described 

 spectrum of the primary element b with the corresponding 

 Fraunhofer and chromospheric lines, as given above, shows 

 that the element b must occur in the free state in the sun's 

 atmosphere, and must be identical with helium, of which 

 hitherto only one ray, D 3 5874*9, is known. Hence, as well as 

 by the close agreement of the lines of the a-spectrum, also 

 given above, with corresponding lines of the solar spectrum, 

 it is proved that hydrogen occurs in the sun's atmosphere in a 

 dissociated condition. 



But if this is the case the constituent a must not only be 

 present wherever the dissociation takes place, together with the 

 helium b, but as the lightest of all gases it must exist above 

 those regions of the photosphere which can only be reached 

 by the much heavier helium and the other elements excep- 

 tionally in powerful eruptions. 



But if this occurs, and if in these very high and relatively 

 cool regions the helium b and the primary substance a meet, 

 then, under the extremely small pressure existing there, all 

 the conditions are present for the formation of the modified 

 ammonium-like hydrogen, which radiates the second com- 

 pound hydrogen-spectrum. Such regions, which correspond 

 to the extreme portion of the corona, will therefore specially 

 exhibit the lines of the second hydrogen-spectrum together 

 with those of ordinary hydrogen. 



The hitherto unknown corona-substance appears to be a gas 

 with the properties indicated above, which emits the so-far 

 solitary known ray at 1474 of Kirchhoff's scale, or A = 5315'9 



(about) of Angstrom's scale, and which I may perhaps be 

 allowed to call " Coronmm." It is easy, then, to assume that 

 the primary constituent a of hydrogen is identical with this 

 coronium, which, in any case, must be a different substance 

 from helium, since the corona-line 53 15' 9 often remains 

 unaltered, whilst at the same time the helium-line D 3 is 

 either broadened, or distorted and displaced, i. e, exhibits 



