108 MR BROUN ON THE RELATION OF THE VARIATIONS OF THE 
to be a day or two after the new moon, and of the greatest heat a day or two 
after full moon, in the same way as our periods of greatest cold and heat are 
after the winter and summer solstices. This seems to agree with the periods of 
maximum and minimum horizontal intensity. If M. Cournor be right, or if Sir 
Joun HerscHEL’s supposition be insufficient, then we must look to the solar ema- 
nations reflected or radiated from the moon for the causes of the variations of the 
earth’s magnetism, and to our atmosphere for a cause of the supposed retardation 
of epoch.* 
21. The connexion of lunar phase and horizontal intensity was first noticed 
by me in July 1845. Iam not aware of any investigations on the relation of the 
horizontal intensity to the lunar month, excepting a paper by M. Hanstesn, of 
which I have lately merely seen the title, which refers to the connexion of the 
horizontal intensity with the moon’s ascending node. 
22. Having mentioned some time ago to Professor Forszs, that I was engaged 
in examining the relation of the lunar periods to the variations of the earth’s 
magnetism, I learned from him that M. Kreiiu of Prague had stated, in his 
volume of observations for 1842, that the horizontal intensity was greater at the 
moon’s passage of the inferior meridian, than at its passage of the superior meri- 
dian. I know not whether M. Krerzy has verified his statement, or to what extent 
his observations prove it.t I have now discussed the observations for 1844, with 
reference to this period, and have verified my results by a similar discussion of 
the observations for 1845. I shall, at present, merely state the leading facts, and 
leave the details to another communication. 
23. The observation at the hours on which the moon was on the meridian were 
termed 0 hours, the observation the hour after one hour, and so up to 24; as the 
moon takes about 25 hours to return to the meridian again. On some occasions 
there were only 24 observations between the two passages; in these cases (few 
in number) the hour of passage was reckoned as 24 hours, and also as 0 hours of 
the next day. The summations for the hours were made for each month; I shall 
only speak of the means for the whole year in this communication ; these means 
have been projected. The large disturbances have been eliminated from the sum- 
mations for 1844 and 1845. (See Curves, No. 9, Plate IV.) Any observation in 
1844 which shewed a difference from the monthly mean, for the hour at which 
* Tt is evident that the variations of horizontal intensity may be due either to changes of the 
total intensity, or of its direction ; any reasoning, therefore, on these facts must be necessarily incom- 
plete, until we are certain of the actual effect. 
+ I have, since this was written, been favoured by Professor ForBEs with a copy of M. Krerz1’s 
table for the horizontal force during the moon’s hour angle. It indicates a minimum of intensity 
about two hours before the meridian passage, and maximum peaks at 12" and 15", giving the interpo- 
lated period of maximum about 13 hours after the inferior meridian passage; the latter period agrees 
completely with my own conclusion, the former differs about three hours from my result. 
