34 THE VOYAGE OF H.M.S. CHALLENGER. 



is brought about mainly by the super-heating of the surface and thence of the lowermost 

 strata of the air. But as regards the open sea, this mode of disturbing the atmospheric 

 equilibrium cannot take place, inasmuch as the influence of solar radiation is only to 

 raise the temperature of the surface of the sea not more than a degree. Hence it is 

 probable that the disturbance of the equilibrium of the atmosphere in the case of 

 thunderstorms over the open sea, is brought about by the cooling of the higher strata of 

 the atmosphere by terrestrial radiation. 



An inspection of the curves of thunderstorms for Oxford, or of thunder with 

 lightning, and of lightning without thunder (PI. II. figs. 26 and 27), shows that they 

 are quite different from each other, — the difference, and it is a vital one, being that 

 while the curve for thunderstorms is coincident with the afternoon minimum, the curve 

 for lightning only is coincident with the evening maximum of atmospheric pressure, or 

 from five to six hours later. Part of this, but no more than an insignificant part, is 

 due to those instances of heat-lightning which are but the reflection of distant flashes 

 of lightning, the thunder accompanying which is not heard. By far the majority of the 

 cases of heat-lightning are not connected with thunder, as is conclusively shown by 

 the curve for August at Oxford, where the very pronounced maximum occurs during 

 the two hours from 9 to 11 p.m., long after darkness has set in, and when the curve for 

 thunderstorms has fallen from the daily maximum to near the minimum. I have 

 calculated or otherwise collected the averages for the curves of these phenomena for 

 nearly two hundred places in all climates of the world, and the result is to show that 

 the two curves are essentially distinct and different from each other, showing conclu- 

 sively that many electric discharges are not accompanied with thunder. 



As explained, the diurnal maximum of heat lightning is coincident with the 

 evening maximum of atmospheric pressure, that is, during those hours when the upper 

 strata over the place are having poured over them a warmer and moister stratum of air 

 which has its origin in the ascending current of the longitudes immediately to west- 

 ward, where the afternoon minimum of pressure is then taking place. In this con- 

 nection it is highly significant that while in May the number of cases of lightning 

 was 17, in August, when the ascending current has much greater relative and 

 absolute humidity, the number of cases was 82, or about five times greater than in 

 May. 



Over the open sea, the diurnal curve of lightning is closely coincident with the 

 evening maximum of pressure, the maximum occurring about midnight (see Table, 

 p. 3 1 ). The relations of the maximum of lightning to thunderstorms over the open sea is 

 essentially different from what obtains over land. Thus, while over land the maximum 

 of lightning occurs from five to six hours later than that of thunderstorms, over the 

 ocean it occurs about four hours earlier. The order of occurrence of these phenomena 

 in the summer months is this — thunderstorms over land, from 2 to 6 p.m. ; lightning 



