26 MADEIRA METEOROLOGIC. PART in. 



Taking, however, first only the two months where 

 Scotland may compare with the two southern stations, 

 we find the rain-band, or the indicated amount of watery 

 vapour dissolved in the air, is double the Scottish amount 

 in Lisbon, and rather more than double in Madeira. At 

 the same time the " low-sun band " appears of nearly the 

 same strength at all three places, as indeed it should do ; 

 for this latter band is known to be an affair of dry gas, 

 and is only observed simultaneously with the rain-band 

 as a check on the ideal numbers in which any observer 

 may choose to record his observations, care being always 

 taken to avoid observing at times very near to either 

 sunrise or sunset ; and that was accomplished involun- 

 tarily for all three stations in June and July by the 

 observing hour being 9 A.M. 



But when we examine the whole year in Scotland, a 

 peculiar law of variation, though within narrow limits, 

 comes out for either band. For, first, the low-sun band 

 is at a maximum in winter and also in summer, but is at 

 a minimum at the two equinoxes. That summer maxi- 

 mum is probably due to the greater purity and strength 

 of the light, arising from the greater absence of the city's 

 usual coal smoke, at that season enabling all spectral 

 phenomena to be more clearly seen ; and the winter 

 maximum, to the sun being at the 9 A.M. observing hour, 

 on or near the horizon, and having, therefore, just then 

 the short-lived intensity after which it is named ; whence 

 it comes that the equinoctial minima of the journal are 



