0. N. Rood on a Secondary Spectrum. 175 
primary spectra ; in the secondary spectra, also, sometimes all 
were present, though usually the potassium line was too faint 
to admit of accurate measurement, and sometimes this was the 
case with the violet hydrogen line. 
Mode of experimenting.—Light from the slit was rendered 
parallel by the collimator and fell upon the prism, which was 
placed in the position of minimum deviation for the Na line; 
emerging from this, it fell upon the grating, which was also 
arranged at its position of minimum of deviation for the same 
ine. The vertical axis of the observing telescope was placed 
directly under the grating, prism and grating being arranged in 
the sense of balancing each other's action. With the aid of the 
chemical lines above enumerated and the micrometer eye-piece, 
a map of the secondary spectrum was then noted down, an 
afterward by two distinct processes, maps of the two primary 
spectra were obtained, the prism and the grating, as_before, 
being in the position of minimum deviation for the Na line. 
furnishes correctly, not only the order and direction in whic 
the lines are disposed, but also their actual distances apart. If 
: and perfect achromatism attain 
“ A sa If, sag the other hand, we reduce one of 
“ yee original spectra to a point, and as 
Na Y before complete the construction, a real 
Hg resultant spectrum will be found on the 
* on line AB, correct in 1 g an of 
HB half the size of the actual physical resul- 
Na tant spectrum (which in this case is the 
a He remaining original spectrum), and so on 
B for all pe cases; or in general, if S= 
ay x, the length of the physical resultant or 
secondary spectrum, or the jen h of 
any portion of it, and g: = the corresponding length of the 
Same spectrum as constructed, then S 35 : 
Seale, the resultant or secondary spectrum in the construction 
