1879.] 



Report to the Committee on Solar Physics. 



307 



for Kew, fig. (3) that for Utrecht, and fig. 4 that for Toronto. 

 This means that the temperatnre-ranges at these three places when 

 treated in a precisely similar manner indicate the existence in each of 

 inequalities of the same periods. The results are not, however, pre- 

 cisely the same for each, which may perhaps be accounted for by 

 certain local influences which have not been entirely obliterated, but 

 which might be overcome by a longer series of observations. 



The sums of the three vertical columns too are not very different 

 from each other. We have, Kew= 98247, Utrecht= 101284, and 

 Toronto = 841 73, a result which might seem to indicate that the 

 magnitude of these inequalities is nearly the same at the three 

 stations. All this is in favour of their cosmical origin. 



5. The next point is to obtain the average results for the three 

 observatories. To obtain this it will not answer to add the three 

 columns together horizontally, and then divide by 3. For although 

 these periods may have a cosmical origin, and be due to the sun, yet 

 the corresponding phases may not be simultaneous at those places. 

 Indeed, this is rendered probable by the fact observed by Toynbee and 

 others that certain weather changes which are in a sense caused by 

 the sun appear to travel from west to east. Now, if we should find 

 that these inequalities take the same course it is not impossible to 

 imagine that they may be allied to, or even in a sense identical with 

 those other well-known changes of weather. We may be as it were 

 on the track of those changes which may perhaps be due to solar 

 inequalities. 



6. How then are we to decide whether the various phases of these 

 inequalities occur at Toronto before they occur at Kew ? or occur at 

 Kew before they take place at Utrecht ? in fine, how are we to ascer- 

 tain that these appearances travel from west to east ? 



This may be ascertained in the following manner. Any one of the 

 numbers in Table III is the sum of the 24 terms of an inequality which 

 have been added together without respect of sign. Thus the number 

 for O'O is for Kew, 3912 ; for Utrecht, 2586 ; for Toronto, 3428. It 

 does not, however, follow that if we add algebraically together the 

 three inequalities corresponding to these three numbers, we shal 

 obtain an inequality the sum of whose terms shall be equal to the 

 sum of these three numbers. For the three inequalities, even if pre- 

 cisely the same in type, may not have their corresponding phases 

 occurring together. The signs of the numbers we add together may 

 therefore sometimes be different, and we shall then have to subtract 

 the one from the other. Thus again if there be a want of simultaneity 

 of phase in two such inequalities, the algebraic addition together 

 of the two will give a result less than the sum (without reference to 

 signs) of the 48 terms, and this falling off will be greater the greater 

 the want of correspondence in phase. Now let us regard Kew as 



