FROM ORDINARY DAYS OF THE ELEVEN YEARS 1890 TO 1900. 
199 
* 
from the whole 11 years, and diurnal inequalities for the year from the 11 years and 
the groups of years of sunspot maximum and minimum. 
To save decimals O'ly is employed as the unit in the tables relating to elements of 
force. The extreme hourly values, he., the algebraic maximum and minimum, are in 
heavy type. In addition to the hourly values, the tables give the range or algebraic 
difference of the extreme hourly values, and the quantity described as A.D. (or 
average departure from the daily mean). The latter quantity represents the result 
obtained by dividing by 24 the numerical sum of the differences of the 24 hourly 
values from their arithmetic mean. 
If readings were taken at every minute, instead of every hour of the day, larger 
values would in most cases be obtained for the range of the inequality, because it 
must be exceptional for the extreme values to fall at exact hours G.M.T. The under¬ 
estimate, however, thus arising is usually very small, as is easily recognised from the 
shape of the curves representing the inequalities. The value of the A.D. is naturally 
less affected by the accident of time, and in most cases it probably gives a better idea 
than the range of the activity of the forces to which the diurnal inequality is due. 
This is more especially the case when the inequality shows a double daily variation with 
two maxima and two minima. 
§ 7. The inequality data in Table V. are shown graphically in fig. 1. As in other 
similar cases, the general features are most readily recognised in the curve, while for 
details recourse is desirable to the numerical data. Fig. 2 shows the H inequalities 
for the three seasons and the year, contrasting the data for the whole eleven years in 
Table Y. with those for the sunspot maximum and minimum groups of years in 
Tables YI. and YII. 
The composite character of the diurnal inequality in H derived from the whole 
year is most clearly seen by comparing some of the seasonal data in the morning 
hours in Table Y. At 6h., for instance, while the winter value has its maximum for 
the day, the summer value falls below the daily mean. The transition from plus to 
minus in winter does not occur until nearly 9h., while it occurs in equinox shortly 
after 7 a.m., and in summer shortly after 5 a.m. 
The minimum or principal minimum for the day occurs at lOh. from May to 
September, and at llh. in the remaining seven months. It is the most constant and 
dominant feature in the inequality. In the four summer months the afternoon 
maximum—then the only maximum—is almost equally prominent, but in the other 
months the afternoon maximum resembles a plateau rather than a peak, and in the 
winter months the principal maximum occurs in the forenoon. 
The forenoon maximum exists also as a secondary maximum in March and October, 
and even at midsummer the appearance of the curves suggests some influence 
delaying the plunge to the minimum at lOh. In most months the rise to the 
afternoon maximum seems to lag somewhat near 4 p.m., and a distinct secondary 
minimum is then recognisable in the case of January. 
