Joseph Baxendell, F.R.S. 51 



west, and in the second period (1862-68) under winds from 

 about south south-east. Comparing the mean temperature 

 under south-west winds with that under south and south- 

 east winds, from the differences he finds a maximum in 

 i860 and a minimum in 1867 ; it is therefore evident that 

 the distribution of temperature under different winds like 

 that of barometric pressure, is very sensibly influenced by 

 changes which take place in solar activity. He makes a 

 comparison of rainfall during the two periods, 1858-62 

 and 1862-68, under different winds; he finds that the 

 amounts of rainfall under south-west and west winds 

 are invariably greater than those under south-east and 

 south winds during the years when the number of solar 

 spots was above the average, and invariably less in the years 

 when the number of solar spots was below the average, and 

 further, that the greatest difference in the first series of years 

 occurred in i860, at the time of a solar spot maximum, and 

 that in the second series, in 1866, at or very near to the 

 time of a solar spot minimum. — " On Distribution of Rainfall 

 under different winds at St. Petersburg during a solar spot 

 period." He states that the hypothesis which led to this 

 investigation requires that great diversity should exist in 

 the relative amounts of rainfall under different winds at 

 different stations, but whatever may be the nature of the 

 distribution at any station, the changes to which it will be 

 subject will take place in a period identical with the solar 

 spot period. After examining the tables of rainfall at St. 

 Petersburg during the eleven years, 1854-64, he finds that 

 the close agreement which exists at St. Petersburg between 

 the maximum and minimum period of solar spots, and 

 those of variation in the distribution of rainfall under 

 different winds, gives increased value to the results derived 

 from the Oxford observations, and affords additional sup- 

 port to the hypothesis which he had advanced in a former 

 paper, that changes in solar activity, and consequently in 



