1879.] 



Report to the Committee on Sola? 1 Physics. 



115 



The vertical portion between the brackets is here all that is ex- 

 posed. 



12. Before concluding this description of onr method we ought to 

 remark that it can only be considered as correct for periods not far 

 removed on either side from that for which the series or book (as we 

 call it) was originally framed. Thus, in the present instance, while 

 the results of Table I will readily indicate inequalities that lie six 

 divisions (of 0'067 day each) either to the right or left of the normal 

 series, it is apparent that +12 divisions from the normal must neces- 

 sarily exhibit the same result as — 12 ; it will not therefore do to push 

 the method nearly so far. 



13. When this method is applied to Table I we obtain the following 

 results : — 



Table II. — Exhibiting the results of the above method applied to 

 the numbers of Table I. 



Divisions from Exact period Magnitude of 



normal. 



in days. 



inequality 



-8-0 



23 7133 



1438 " 



-7-5 



23 7469 



1757 



-7-0 



23 7804 



1546 



-6-5 



23 "8140 



1047 



-6-0 



23 -8475 



1707 



-5-5 



23 -8810 



1141 



— 5 -0 



23 -9146 



1002 



-4-5 



23 -9481 



2009 



-4-0 



23 9817 



2954 



-3 5 



24 -0152 



3109 



-3-0 



24 -0487 



2504 



-2-5 



24 -0823 



1289 



-2-0 



24 1158 



1068 



-1-5 



24 -1494 



1473 



-1 -o 



24 -1829 



1700 



-0-5 



24 -2165 



1513 





24 -2500 



1162 



+0 '5 



24 -2835 



. . 1115 



+ 1-0 



24 -3171 



1398 



+ 1 -5 



24 -3506 



867 



+ 2-0 



24 -3842 



1062 



14. The results of Table II are exhibited graphically by means of a 

 «urve in Diagram I, fig. I, which accompanies this paper. 



In this curve the abscissas denote periods, while the ordinates 

 represent corresponding inequalities. It will be noticed that we have 



i 2 



