1884.] India during the Sunspot Period. 203 



is less during the summer, and vice versa. Again, in those years which 

 in the diagram are marked Hh , and which immediately precede the 

 years of sunspot maximum and minimum, the order above pointed out 

 obtains only in a slight degree ; in other words, these are years of transi- 

 tion. 



That these facts are purely the result of accident seems very un- 

 likely, for the}^ are supported by three other series of concurrent facts ; 

 which are as follow : — 



1st. The plus years begin immediately after the sunspot maxima, and 

 the minus years begin immediately at the sunspot minimum. 



2nd. There are more transition years during the slow descent of the 

 sunspot curve than during its rapid ascent. 



3rd. The oscillations of both the winter and the summer rainfall are 

 of greater amplitude during the negative years than during the positive. 



"With reference to the first of the above series of facts, it might be 

 supposed that, as the minus years begin immediately at the sunspot mini- 

 mum, so for perfect analogy the plus years should begin immediately at 

 the sunspot maximum. But this is by no means necessary, for the slight 

 delay in the coming in of the positive years agrees very well with the 

 slow descent of the sunspot curve as compared with its rapid ascent. 



The main fact which I have pointed out in this paper, — namely, 

 that the smaller variations of the winter rainfall are the same in character 

 as those of the summer rainfall during years of maximum sunspot ; and 

 opposite in character during years of minimum sunspot, — if it can 

 be established as a general rule, will be an important one ; for it will 

 indicate that, whatever be the cause which produces the general opposi- 

 tion in character between the eleven yearly variations of the winter and 

 of the summer rainfalls, that cause operates chiefly during the years of 

 minimum sunspot, and during three years of maximum sunspot it operates 

 only in a very minor degree, and in two of those years (namely, the 1st 

 and 2nd) it probably does not operate at all. By thus limiting the period 

 during which the cause operates, a valuable point is gained, and a clue to 

 a knowledge of the cause possibly afforded. 



It is also interesting to notice that not only do the rules above in- 

 dicated obtain qualitatively, but that there is also a near approach to a 

 quantitative relation between the short period oscillations of the summer 

 and the winter rainfall respectively. The nature of this relation in the 

 years which I have denoted as positive, namely, in the 1st, 2nd, and 3rd 

 years of the sunspot cycle, will be seen at once on inspecting the " minor 

 oscillations " curve of the diagram. It will be observed that the oscil- 

 lation which takes place in the two curves during those three years is 

 not only the same in phase, but is nearly the same in amplitude. The 



