184 SOLAR CHANGES OF TEMPERATURE. 



represent the growth or diminution of the percentage of the whole 

 number of lines due to iron in the spots, are at a minimum when the 

 others are at a maximum, and are at a maximum when the others are 

 at a minimum, and that again they have fallen to a mininuun at the 

 time of the maximum in 1893. 



These lines which are at a minimum when the sun spots are at a 

 maximum, and vice versa, are the known iron lines, but there is also a 

 dotted curve whose maximum and minimum agree with those of the 

 sun spots, and those are the lines whose provenance we do not know. 



These are the solar phenomena. We shall see the same thing restated 

 in its apparent connection with terrestrial phenomena b}' looking at 

 Plate II, which is drawn on the same scale of years. Here we see 

 that the close of the famine year in each case agrees with the period 

 when the dotted line crosses the solid line; that is, where certain lines 

 due to temperature appear or disappear. Hence, at certain periods of 

 about eleven years we have spectroscopic evidence that there are 

 changes of temperature in the sun, and from what follows, that these 

 apparently coincide with the close of the famine years in India, as 

 well as with other terrestrial phenomena. This is the important coinci- 

 dence pointed out b}- Sir Norman Lockyer. It does not seem pro])able 

 that it can be due to chance, but we must await further observations 

 before feeling that this most interesting observation has acquired the 

 character of entire demonstration. 



