427 
DR. W. M. HICKS: A CRITICAL STUDY OF SPECTRAL SERIES. 
calculated frequencies for m > 2. These for m = 3, 4, 6 are 22277'28, 25148'60, 
27569‘40. The last gives d\ = '40. No lines correspond to the others, hut 22211 
observed by C. and R. with Cu electrodes and (15) 25107T4 (R.)) (9)---G‘G4 (W.) are 
respectively 66'28 and 41’56, (41'96) less. Now a dj displacement on the limit gives 
13‘96, so that these correspond within error limits to a (5(5i)F(qo), d\ = —’5, and 
(3^i)F (oo), d\ = —’06, and stand in some analogy to what has been seen to happen 
in X. The intensity of the second line however would seem very great for m = 4. In 
support of the C.R. line with its large possible error is (2)22196'92 (Royd) at 13‘08 
behind it or an extra It would make 22196 = {^^\) F, and consequently F = 22280'68, 
i.e., dX = —'7. If the 5640 lines exist, they all lie outside the region of observation 
except in the case of the first line 15846 where the corresponding triplet set is found 
(see above). However, anticipating the value of the links obtained later and using 
them as sounders, viz., e = 23678'3, u = 11191‘8, v = 13680‘6 with some uncertainties 
we can test for their existence. Taking the observed lines 22196‘92, 22211 we expect 
lines about 27838, 27858, only to be observed if strong. There are none, but there 
are (3) 14166 (W.) 2816 16982'0 (C.R.) which with the v link give 27846’6 +1'9 + du, 
30662’6±3 + du, separation = 2816 + 5. These suggest the triplet set [Sy] F, viz., 
22196-9 5649*7 
22210-9 5635 + l*9p + du 
27846-6 +l-9p + du 2816 + 3p'-l*9p 30662-6 + 3^ + du 
The results of sounding are indicated in the following, where, since e, u, v are only 
approximately known, their values are supposed corrected by de, du, dv. The letters 
after wave-lengths refer to observers :— 
m = 3. (5(li) F. 
OiUside (3) 14166 (W.).t; 16982 (C.R.).?; 
22196-9 5649-7 
22210-9 5635 + 1 -9p + dv 27846-6 + 1-9/; +(fe 2816 + 3p'- l-9p 30662-6 + 3/; + f/y 
With the calculated value of Fi the first separation is 3-5 larger. The first line is 
displacement on 22210. If the former does not belong to the system, we may take 
C.R.’s direct line which has possible error of 4 or 5. 
Outside 
25107-14 5640-6 + 4p + du 
?;.(0) 15136 (W.).e 
m = 4. (3^i) F. 
(28i) 19583 (C.R.).m ' (28,) 22398 (C.R.).?4 
{3074:7 - 8 ) +ip+ du 2814-2 + 5p'-4p (33562) + 5/; + (/?i 
f;.(0) 20784-31 (W.).e 
?;.(-8i)(l) 20799 (C.R.).e 
(30782-6) 
?;.(0) 23598-2 (W.).e 
2814-9 
(25134-3) 5648-31 + 46-47 
[25148-60] 5648 
?;.(!) 20799.e 
2816-0 
(33596-5) 
?«.(!) 21120-0.e 
