FROM ORDINARY DAYS OF THE ELEVEN YEARS 1890 TO 1900. 
269 
1890 by 259y, or about 
alter D by 1/ was 
1'4 per cent. If we go to 3 figures, the force required to 
5'29y in 1890 and 1891 
5’30y „ 1892 
5'31y „ 1893 and 1894 
5\32y „ 1895 
5'33y „ 1896 
5-34 y „ 1897 and 1898 
5’35 y „ 1899 
5’36y „ 1900. 
This variation, though quite appreciable when we are dealing with large ranges, is 
hardly worth considering in the case of the minimum ranges in Table XLVI. 
Examining that table we see that the winter months, especially December, are 
conspicuous for small ranges. The fact that the mean values of II' for winter months 
fall conspicuously short of those for other seasons is due, not so much to the absence 
of large values of R/, as to the presence of a considerable number of very small values. 
In summer, really small ranges are scarce. In fact, in the half year from April to 
September no single day of the 11 years had a range under 5'. In 1893 no April 
day had a range under 12 /- 5, though the mean value for the month was only 17'T, 
and in June the lowest range was 12 /- 3 though the mean for the month was only 
16/'4. The uniformity of the range in June 1893 was very remarkable. In 21 of the 
30 days the range lay between 14 /- 0 and 1 8 r ‘5 and the highest value of the month 
was only 21 /- 6. The absolutely smallest range of the 11 years was 2 ,- 2, or 12y in 
force. 
Turning to Table XLVII., we see that the lowest range recorded was 7y. The 
winter months, as with D, supply all the outstandingly small ranges, December being 
especially conspicuous. 
During the eleven years the sensitiveness of the Ivew H nragnetograpli was always 
near 1 mm. = 5y. Thus a range of 7y implies a variation of only 1'4 mm. in the 
length of the ordinate throughout a portion of curve whose length of abscissa is 
360 mm. This means a slope so gentle everywhere that the recognition of the 
positions of the maximum and minimum is no easy matter. Again, a change of 1°F. 
in temperature would alter the ordinate fully 0’3 mm. Thus not improbably 7y may 
have been somewhat an under-estimate of the true minimum range. 
At the other end of the scale the largest observed ranges were 
in D, 1° 25'T ( = 457y) on March 15, 1898 ; 
in H, 720y on February 13, 1892 ; 
in V, 639y on July 20, 1894. 
It is possible that the D range on March 15, 1898, may have been exceeded on 
2 o 2 
