Hydro- Electrical Currents. 91 



tion; it was attached near to the upper joints of the battery, on the 

 south face, and, like them, varied its temperature with the weather. The 

 poles weighed 1 grain each on 21st April. On 21st August, after four 

 months' exposure, the poles were reweighed, when the difference in 

 weight was found to be 0.12 of a grain ; the negative pole for the day- 

 current being the heavier, shewing that the silver has been deposited by 

 solar radiation. 



20. When the voltameter attached to the battery is maintained at a 

 uniform temperature of about 90° by means of a sand-bath, the current 

 generated chiefly by astral radiation, preponderated over the day cur- 

 rent in April, when the sun was more than twelve hours above the hori- 

 zon. To examine the influence of the fluctuations of the weather on 

 this arrangement : A battery, fig. 1, was placed with the upper half ex- 

 posed to sun and sky ; two copper wires were soldered to its poles, then 

 conducted inside the house, and permanently attached to a galvanometer 

 fixed at the distance of 15 feet from the battery. With this instrument 

 I was enabled to watch through the summer the changes in the force and 

 direction of the currents. The summary of these observations is, Ist^ 

 For a clear sky, the indications are constant as to direction ; from one 

 to one and a half hours after sunrise, to within the same period of sun- 

 set, the exposed joints of the battery are heated, and for the remainder 

 of the twenty-four hours, the same joints are cooled by radiation. The 

 time stated betwixt the rising or the setting of the sun, and the change 

 in the direction of the thermo-electrical currents, is given as the average 

 interval ; for there are occasions when the time intervening betwixt the 

 change of the currents and the rising or setting of the sun is much longer 

 or shorter. The most frequent cause of these changes in this interval 

 is, when the general aspect of the sky is clear, with clouds on the hori- 

 zon near where the sun rises or sets. 2dj In dry cloudy weather the 

 action of astral or solar radiation is much checked ; but this state of the 

 sky rarely alters the direction of the current due to the period of night 

 or day at which the observation is made. 3c?, In wet unsettled weather : 

 Rain falling for hours without intermission, nearly suspends the action 

 of the battery either by night or day. But by day the light or intermit- 

 ting summer showers produce by far the greatest variations in the force 

 and direction of the currents generated by the weather. The showery 

 weather, so common during our last spring and summer, by wetting the 

 exposed and previously warmed portion of the battery, produced a ther- 

 mo-electrical current corresponding to that derived from the radiation 

 to a clear midnight sky, and often exceeding it in intensity. 



21. The continuation through the summer and autumn of the experi- 

 ments where the battery, fig. 1, was employed to deposit silver in a vol- 

 tameter at a uniform temperature of 90°, enables me now to state the 

 observed changes through the four seasons. From January to the end 

 of April there was a constant deposit of silver by the current due to 

 astral radiation ; the rate of this deposit gradually decreasing as the 



