ox A TABLE OF STANDARD WAVE LENGTUS OF THE SPECTRAL I.TVr«. lllU 
wave lengtlis of the metal lines arc very nearly the .^amo a"^ ubtaliunl from any 
portion of the visible or ultra violet spectrum proves llie accurary of the lallcr, 
as well as that of the metallic wave lengths. 
7. Measurement of plates having metallic >^pectra of dilTorcni orders. 
Advantages of the Pkocess. 
The only other process of obtaining relative wave Iciiglbs is by means of angular 
measures. Supposing the angle to be about 45', an error of 1" will make an crr^r 
of about 1 in 200,000 in the sine of the angle. When one considers tlic r-li.mges 
of temperature and barometer measuring on one line and then nnollier, togeilM-r 
with the errors of graduation, it would be a difficult matter to nie;iv„ro thif» angle 
ma 
error of 1 in 100,000, or about ^ division <A Aug 
Looking over the observations of principal standards made under the dlrecti(»u of 
Professor Vogel in Potsdam, with very poor gratings but an excellent ^^pectroineler, 
find the averaire nrobable error to be about ±T^yjo of the wave IcMigtli, wlii.'li 
we , „,-. , 
is not far from the other estimate. This does not include constant errors and 1 
believe the probable error to be really greater than this. 
The method of coincidences by the concave grating gives far superior rcMih*. 
The distance to be measured is very small, and the equivalent focal length of a 
telescope to correspond w^ould be very great (21i feet), t 
of b 
the dispersion of the air, which, when known, can be corrected for. ^ It^ is not to 
be wondered at that this method is far superior to the fo 
ndeed, reduced to ±i^r^., or even less for the best lines A^ 
d thermometer are eliminated at once, except the small effect 
1 
can be made on photo-rapl.., this probable cn-or is scarry incrca^r,! at all; but 
talcing it at twice^ tbe above esti.nate, the n,olho.l even then re„,am< from 
- . .. . r „.,_,!„.. r..nn<nvomont Indeed, the un- 
even ^ 
three to five times as accurate as that of 
o 
made on mv mind in looking over Vogel's Pot^lam ob 
j^ncr).^iuil lllUUt; un lll^ lliinu m I'^wivi..^ ^.^^ - ^ T 1' 1 'fill 
my tables and process ar6 ten times as accurate as theirs; and t in' an> cnie 
student of both processes will come to a similar conclusion. 
Tbe wonderful result that can be obtai 
the new micrometer, which 
m 
of photog 
mea^^nre plates over twenty iiK'be. long. !> i-arllv 
»-»" Lue new micrometer, wineii uiui mcasuic ^jk-il^ ^„au «r a 
' ^ ^ r .X- niplios tbe wave lengtri oi a 
.seen in the table. Where the distance is only a f.vv inciios, tne ^ ^^ v^.;,j,, 
3ries of lines can be measured with a probable error of less (ban o . ■ 
iti-^rt cA fhnt the r)r()i>:*'»l* ^ttoi 
f An,..t,-om. Indeed, a series would detormnio any hne 5n that r 
VOL. XII. 
15 
