400 Prof. Cornette on the Aurora of 1859. 
exception to this law between the equator and Mexico, and the 
exceptions are very rare at Springhill. Such an exception hap- 
— on the night of the Aurora Borealis (Aug. 28) when the 
arometer remained stationary from three to nine, and rose after 
nine when it should have fallen. 
2d. The temperature féll considerably but not until the next 
day (29th) under a northwest wind, which had not blown for a 
long time, and which is ordinarily cold. , 
3d. The tension of watery vapor in the air was slightly modi- 
fied. The mean degree of saturation on the 27th and 28th was 
21°-35, tension 18°‘7™m; 29th and 80th was 22°40, tension 19°9. 
The 28th at 9 in the evening during the phenomena 22°°26, ten- 
sion 19-7; 28th at 10 in the evening after the phenomena 23°47, 
q 
stricted. In high latitudes it is more disturbed and the ampli- 
tude larger. Three years observations in 4°, 14°, and 19° north 
latitude, leaving me no doubt upon the accuracy of these maxi 
ma minima. In these latitudes this daily march was dis- 
turbed only during earthquakes, a ed rmal 
order after the quaking. At Springhill (where I have followed 
