S. ae eh a a al 
W. Gibbs on a normal Map of the Solar Spectrum. 9 
it each millimeter of the chart corresponds to the ere 
of a millimeter of wave length. The probable error of m 
ment in determining vs wave length does not pinaily xvod 
chart, while the seven principal io of Fraunhofer are denoted 
by the usual letters, Tn order to permit a comparison of the 
lines characteristic of different Saco closely related to each 
other in their chemical properties, or belonging to the same natu- 
ral groups, several impressions of the scale have also been printed 
which are produced by any one element. For aes same Treason 
it would probably be impossible at present to determine whether 
there is any law governing the wave lengths of the spectral lines 
belonging to the elements as regards their intervals or distribu- 
tion over the spectrum. The conviction that precise knowledge 
of these and similar points must depend upon a knowledge of 
wave lengths and not merely of indices of refraction, has led me 
to make this first attempt to form a normal map of the spectrum. 
s a first attempt merely it is a imperfect, but my 
object will be fully attained if I shall have succeeded in point- 
ing out the path to be followed in the fature discussion of the 
subject. 
a: grateful acknowledgments are due to Prof. J. KE. Hilgard 
er whose superintendence most of the curves which I have 
pone le for smaiepobathion have been drawn in the office of the 
Coast Survey. To Mr.S. P. Sharples I am also indebted for 
much assistance in the work of computation. . 
bridge, July, 1866. 
Postscript.—Since the above was written I have received a 
second paper by Ditscheiner* on the wave lengths of the spec- 
tral lines, in which the author gives the results of a determing 
tion of the absolute value of the interval between two suc 
sive lines of the grating or ruled glass surface employed by him. 
When this value is used the wave lengths of Dg and De become ~ 
* Sitzungsberichte der kaiserlichen Akad. der Wissenschaften, Bandi 280 : 
Am. Jour. Sct.—Seconp Sanies, Vou. XLITI, No. 127.—Jay., 1867. _ : 
2 
