194 
DR. C. CHREE: DISCUSSION OF KEW MAGNETIC DATA 
Non-cyclic Ch ange. 
§ 4. In obtaining the diurnal inequalities the non-cyclic (n.c.) change C — i.e., the 
algebraic excess of the value at the second midnight (hour 24) over that at the first 
midnight (hour 0)— has been eliminated in the usual way, by applying at hour n the 
correction C (12 — ??.)/24. In the case of D the value of C for each month of the 
eleven years was given in a previous paper .* The value — 0'T2 assigned there to 
February, 1892, should have been — 0 /- 22. Allowing for this, the mean values for 
the year 1892, for the month of February, and for the whole 11 years, become 
respectively — (T'031, — (f'052 and — (fi'301. The change of force perpendicular to 
the magnetic meridian necessary to alter D by F varied gradually from 5’29y in 
Table III.—Non-cyclic Change (Unit O'Oly). 
Element. 
1890. 
1891. 
1892. 
1893. 
1894. 
1895. 
1896. 
1897. 
1898. 
1899. 
1900. 11 years. 
D 
- 7 
- 15 
- 16 
- 23 
-52 
+ 4 
- 16 
- 13 
-25 
-21 
+ 9 -16'0 
H 
+ 2 
+ 68 
+ 200 
+ 69 
+ 90 
+ 84 
+ 118 
+ 80 
+ 63 
+ 61 
+ 42 +79-7 
V 
-37 | 
-67 
- 55 
- 55 
-45 
| 
-40 
- 25 
-29 
-30 
-50 
-28 -42-0 
1890 to 5'36y in 1900, the mean for the eleven years being 5‘32y. Tables III. 
and IV. give the mean yearly and monthly values of the n.c. change for the ordinary 
days in the three elements, those for D being expressed in terms of the equivalent 
force. Ordinary days, it should be noticed, include the quiet, though, of course, not 
the disturbed. 
Table IV.—Non-cyclic Change (Unit O'Oly). 
Element. 
1 January. 
February. 
j 
1 March. 
April. 
>> 
a 
^4 
o5 
r* 
i—} 
JC 5 
August. 
September. 
October. 
November. 
December. 
D 
9 
-28 
- 41 
- 6 
+ 7 
- 26 
_ 2 
0 
-52 
- 24 
- 15 
+ 4 
H 
+ 41 
+ 97 
+ 202 
+ 75 
+ 49 
+ 6 
i + 54 
+ 132 
+ 64 
+ 87 
+ 84 
+ 67 
y 
1 - 115 
+ 11 
- 122 
-211 
• 219 
-245 
- 126 
- 2 
+ 98 
+ 214 
+ 141 
+ 73 
There is obviously a prevailing tendency for the n.c. change on ordinary days to be 
negative in D and V and positive in H. The inference that the D and V elements 
were falling, and the H element rising, would be correct, though unwarranted. The 
n.c. effect is influenced, of course, by the secular change but it is partly of instrumental 
“ ‘National Physical Laboratory Collected Researches,’ vol. Y. (Table IIa., p. 48). 
