NOVEMBER 19, 1897. ] 
ordinary properties are well understood. 
But under the conditions met with in the 
sun and stars it behaves in a mysterious 
manner. Notwithstanding its considerable 
atomic weight, it floats quietly high above 
the surface of the sun, where other heavy 
metals are only occasionally present in 
consequence of violent eruptions. It is 
true that the apparently abnormal spectrum 
of calcium under these conditions has been 
shown by Sir William and Lady Huggins 
to be merely the result of extreme tenuity 
of the luminous vapor; but the existence of 
ealcium at such great heights, under any 
conditions whatsoever, seems to point to 
some remarkable property of the element 
which is unrecognizable by the methods of 
ordinary chemistry. 
The spectrum of a substance is not the 
same under all circumstances. In some 
cases a change occurs suddenly when cer- 
tain critical conditions are reached; in 
others the change is gradual and progress- 
ive. By studying these changes in labora- 
tory experiments, and comparing them 
with what we see in the observatory, we 
are able to arrive at some definite con- 
clusions respecting the conditions which 
prevail in the stars, while the same com- 
parison often throws light on the phe- 
nomena observed in the laboratory. It 
has been shown, for instance, that the 
spectrum of magnesium gives a means of 
estimating the temperatures of the stars; 
and the same criterion enables us to recog- 
nize in the stars temperatures vastly ex- 
eeeding the highest that have been pro- 
duced on the earth. Thus the science of 
astrophysics allows us to extend our inves- 
tigations to temperatures which the re- 
sources of the laboratory cannot furnish. 
It may be well to mention an example of 
the difficulties, to which I have referred, 
arising from our imperfect knowledge of the 
laws which underlie phenomena constantly 
observed. Recent comparisons of thespectra 
SCIENCE. 
Tol 
of the sun and metals, made at the Johns 
Hopkins University with the concave grat- 
ing spectroscope of Professor Rowland, have 
proved that spectral lines may not merely 
be widened by increased pressure of the radi- 
ating vapor, but that they may be shifted 
bodily ; while the still more recent investiga- 
tions of Zeeman show that a line may be wid- 
ened (and at the same time doubled) under 
the influence of a strong magnetic field. It 
is true that in both cases the effect produced 
is very small. It could not lead to mis- 
takes in identifying stellar lines, or to ap- 
preciable errors in measuring celestial mo- 
tions. But the fact that the spectrum of 
a substance varies according to circum- 
stances which are as yet only imperfectly 
understood, shows us the necessity of exer- 
cising caution in interpreting the spectral 
phenomena presented to us by the heavenly 
bodies. At present these spectral variations 
increase the difficulties that the astrophys- 
icist has to contend with. Hventually they 
will become additional and most valuable 
sources of information. 
The discovery, by Kayser and Runge, of 
line series in the spectra of the common ele- 
ments has a most important bearing on the 
work of the astrophysicist. It provides 
him with the means, long greatly needed, of 
deciding with certainty whether or not lines 
in celestial spectra are identical with lines 
in the spectra of terrestrial substances. On 
the other hand, as we have already seen, 
he is sometimes able to supply the physicist 
with missing ‘data. 
From the point of view of the old astron- 
omy the most important result of the in- 
troduction of the new methods has been the 
determination of motions in the line of sight 
by means of the spectroscope. The method 
has been tested so often, and with such uni- 
form success, that there is no longer any 
doubt as to the correctness of the principle 
on which it is based, or to the accuracy of 
the results which it is capable of yielding 
