24.0 REPORT—1 896. 
unrestricted monthly mean is, in each month of the year, for each 
element, as follows (average of five years 1890 to 1894) :— 
Greenwich | Jan. Feb, | Mar. Apr. May | June | July } Aug. | Sept. | Oct. | Nov. |Dec. 
le | 
PORE In declination 
aiuinte | —0"04 | +012 | +0"24 | —0"10 | +012 | +010 | 000 | +010 | —0’06 | —0"08 | +0/-24)0"00 
Fs esa In horizontal force 
value | | +35 | +42 | +24 | +37 | +381 | +20 |4+22|] +5 | +27[ +55] +51 [+57 
all days In vertical force 
value . 
| +31 | —26 | —10 | —13 | —i1 | +10 | -li | a5 %, i pie ii 
Mean yearly excess of ‘quiet’ day value in declination = + 0/05 
es -4 a in horizontal force = + 34 
“8 oe is in vertical force = — 8 
The corresponding separate yearly differences are :— 
| Greenwich 1900 | sor | te02 | 1893 | 1894 
H In declination 
+0"01 | —O"12 | +0'07 | +0719 | +012 
Excess of absolute In horizontal force 
‘ ] ? 
quiet’ day value - : 7 
above all days value | +L eee) + BL. | Bel ieee 
In vertical foree 
ee eg 8 |. oo 
I have not sufficient opportunity at the present moment to add much 
by way of discussion of these numbers, but taking the element in which 
the difference of absolute value is most marked, that of horizontal force, 
some few remarks may be offered. We see that the uniformly positive 
non-cyclic change on ‘quiet’ days is accompanied by an increased absolute 
value of horizontal force on such days, as compared with the value from 
all days, as we should perhaps expect. At Greenwich magnetic disturb- 
ance commonly causes diminution of horizontal force, after which the 
value works back to amore normal one. But there are years in which 
the magnetic registers are unusually quiet, with few disturbances of even 
moderate amount, as in the years 1878 and 1879. The inference would 
be that in such years the difference between the absolute value for the 
especially ‘ quiet’ days and that for all days should be small, varying to a 
certain extent in different years with the more or less prevalence of 
magnetic disturbance. I cannot, however, for the moment refer to the 
Pawlowsk differences of which Dr. Chree speaks, to ascertain whether 
they exhibit variations of this character. Further, whether the rise of 
value on ‘quiet’ days represents a recovery from abnormal loss during dis- 
turbance, or whether the abnormal fall during disturbance is in any way 
a consequence of preceding abnormal rise, may be a question. But the 
view that the recovery on ‘ quiet’ days is rather a consequence of abnormal 
fall during disturbance, that is, that the disturbance is really the primary 
dominating factor, appears to receive support from the following considera- 
tion. When disturbance suddenly arises it seems to break out over the 
