FROM SELECTED DAYS DURING THE SEVEN YEARS 1898 TO 1904. 
321) 
previous paper* of 1896, which dealt however only with eye observations at one or 
two hours during some months of 1895 and 1896. The results obtained are generally 
similar, except that in 1896 I found an apparent association of low potential with 
long previous sunshine. Even here the contradiction may be more apparent than 
real, as the influence of sunshine may depend on the hour of the day. An intensifica¬ 
tion for instance of the afternoon minimum may he accompanied by a compensating 
increase of potential during the night. Considerations of this kind must in fact be 
borne in mind in other cases besides that of sunshine. 
§ 25. In 1896 I paid special attention to aqueous vapour, as Exner’s theory, which 
associated potential gradient and aqueous vapour m a definite formula, was then 
rather in the ascendant. I advanced grounds—whose cogency will now, I tllink, be 
hardly questioned—for believing the theory in its general form to be untenable. The 
1895—6 data were from a Kelvin portable electrometer, whose insulation was never 
in doubt. I he present data are from an electrograph, whose insulation may depend, 
especially at night—sometimes it may be feared not inappreciably—on the greater or 
less dryness of the air. Also, for one and the same amount of aqueous vapour in the 
air, it seems all important so far as potential is concerned whether fog forms or not, 
and in the long nights of winter no satisfactory data exist as to the occurrence of fog, 
or its thickness when present. I have thus omitted consideration of the effects of 
aqueous vapour as too large a subject to be included in the present paper. 
A similar remark applies to wind direction. There are here two main difficulties : 
first the fact that the direction is perpetually shifting, and the extent of the variation 
in a single day is often large ; second the fact that when the wind is light—and there 
are few fine weather days when this does not happen sometime during the 24 hours— 
the record of direction is often uncertain, the strength of the wind being insufficient to 
turn the vane of the Robinson cup anemograph. In particular is this true of foggy 
days in winter, the precise days in which the potential tends to be highest. 
Appendix. 
On the Relation between the Regular Diurnal Changes of Barometric Pressure 
and Potential Gradient. 
§ 26. I he only meteorological element whose diurnal inequality presents two 
prominent maxima and minima is the barometric pressure. Everett seems to have 
been the first to call attention to the similarity between the regular diurnal inequalities 
of potential and barometric pressure, and the potential data which he considered more 
especially in this connection were those from Kew. Having no Kew barometric data, 
Everett employed instead some from Halle. Confining himself to the mean 
VOL. CCVI.—A. 
* Loc. cit. (for summary of results, see pp. 130, 131). 
2 u 
