NOVEMBER 19, 1897.] 
representation of the surface features of the 
planets. Although the moon has been pho- 
tographed with splendid success, and the 
planets with results that are encouraging 
and suggestive, we still rely (in the case of 
the planets) on the old method of hand 
drawing used by Galileo. The fallibility of 
the draftsman is well known. It has been 
illustrated again and again. Yet there 
seems to be a curious habit among some 
observers of regarding a drawing, when 
once made, as invested with high authority 
—as that of a standard established by legis- 
lative act. A photograph, if it could be 
made, would be free from the errors of the 
draftsman, and from a personality which is 
recognizable in all hand drawings, and 
which, though it is scarcely to be classed 
as an error, it would be desirable to avoid. 
Here, then, is another opportunity for the 
new methods. There is no reason to sup- 
pose that it is impossible to obtain photo- 
eraphs of the planets which will show all 
that the eye can see, although there are 
many reasons to know that it will be very 
difficult to do so. The instruments for this 
purpose would have to be quite different 
from those in general use, and there would 
be few occasions, in even the most favored 
regions of the earth, when they could be 
employed. Difficulties would also arise 
from the rapid rotation of some of the 
planets. But this is not the place to dis- 
cuss the necessary conditions. It is only 
fair to say that Professor Schaeberle, of the 
Lick Observatory, has already been experi- 
menting in this direction—with what suc- 
cess is not yet generally known. 
Passing to stellar spectroscopy, a field 
broader even than that of solar physics is 
opened before us; for the sun, although 
paramount in his ownsystem, is only one of 
the stars. In a general way, the spectra of 
the stars have been observed, and classified 
according to their character, and objects of 
unusual interest have been noted for fu- 
SCLENCE. 
703 
ture investigation—many a rare specimen 
has been meshed in Harvard’s widely ex- 
tended net; but the detailed study of indi- 
vidual spectra has just begun. For this pur- 
pose large telescopes are desirable, if not 
absolutely necessary. Many observations 
of precision required in the older astronomy 
are best made with small telescopes. But 
in stellar spectroscopy light is all-important; 
and while much can doubtless be accom- 
plished with small telescopes, there is prob- 
ably nothing that cannot be done better with 
large ones. Hven in solar spectroscopy, 
where the supply of light is abundant, a 
large image is required for the study of in- 
dividual parts of the sun’s surface. 
No department of astrophysics has prof- 
ited more by the introduction of photo- 
graphic processes than stellar spectroscopy. 
To the advantages of photography already 
mentioned there is here to be added another 
not less important. Owing to atmospheric 
disturbances the image of a star dances 
about on the slit-plate of a spectroscope 
placed in the focus of a telescope. The 
spectrum is not only faint, but tremulous, 
and to measure the lines in it by visual ob- 
servation is like trying to read a printed 
page irregularly illuminated by flashes of 
light. These irregularities do not appear 
on the photograph. They disappear in the 
process of integration. Negatives obtained 
with the spectrograph can be directly 
measured under a microscope, or enlarge- 
ments can be made from them in the usual 
manner. In this way photographs of star 
spectra are now made which are compara- 
ble, with respect to accuracy and wealth of 
detail, to Kirchhoff’s famous map of the 
solar spectrum. ‘It is simply amazing,”’ 
says Professor Young, with reference to the 
Draper memorial photographs, ‘‘ that the 
feeble, twinkling light of astar can be made 
to produce such an autographic record of 
substance and condition of the inconceiy- 
able distant luminary.” 
