366 
MR. E. G. BILHAM ON THE SPARK SPECTRUM OF 
Table I (continued). 
Wave-length. 
Intensity. 
Class. 
+10 atmospheres. 
+ 5 atmospheres. 
AA 0 
111 rcnro A - u - 
No. 
of plates. 
AA 0 
111 ToVo A - U - 
No. 
of plates. 
3769-62 E. 
5 
IV. 
77 
3 
55 
2 
3775-74 
5 
II. 
22 
3 
12 
2 
3783-67 
5 
I. 
16 
3 
12 
2 
3807-29 
7 
II. 
18 
3 
9 
2 
3831-87 
2 
I. 
36 
2 
— 
— 
3849-70 E. 
5 
IY. 
149 
1 
— 
— 
3858-50 
8 
I. 
21 
3 
12 
2 
3889-80 
3 
IV. 
— 
— 
— 
3972-32 
2 
III. 
28 
1 
28 
i 
3973-75 
3 
Y. 
58 
1 
82 
i 
4015-65 
3 
IV. 
— 
— 
91 
i 
4121-48 1 Co. 
2 
III. 
26 
1 
48 
i 
4142-47 
i 
III. 
32 
1 
— 
— 
4288-20 
5 
V. 
121 
1 
— 
— 
4331-83 
4 
III. 
44 
2 
— 
— 
4359-76 
2 
III. 
106 
2 
— 
— 
4401-77 
6 
V. 
102 
2 
— 
— 
4459-25 
6 
V. 
93 
2 
— 
— 
4462-65 
3 
V. 
99 
2 
— 
— 
4470-70 
6 
V. 
89 
2 
— 
— 
4592-76 
5 
V. 
118 
1 
— 
— 
7. Comparison with Previous Measurements. 
Humphreys'* has measured the displacements often nickel lines in the arc spectrum 
at pressures of 9f, 12|-, and 14|- atmospheres. The mean shifts per atmosphere of those 
lines which have also been measured at South Kensington have been calculated. The 
results are given in Table II. 
Table II. 
Wave-length. 
e> 
Mean shift per atmosphere in ydo-o A -U. 
Humphreys 
(Arc). 
Bilham 
(Spark). 
3458-62 
2-08 
1-80 
3461-84 
1-74 
2-50 
3501-01 
2-17 
1-50 
3515-21 
2-67 
2-75 
3524-69 
2-40 
2-90 
Means . . 
2 ‘21 
2-29 
* ‘ Astrophysical Journal,’ VI., p. 204. 
