386 



Year. 



I. 



II. 



III. 



IV. 



1840 



+ 0'-06 



-0'-30 



+ 0'-43 



-0'-19 



1841 



+ 0-04 



-0-01 



+ 0-33 



-0-35 



1842 



+ 0-71 



-0-27 



+ 0-25 



-0-70 



1843 



+ 0-65 



-0-48 



+ 0-29 



-0-45 



1844 



+ 0-47 



-0-12 



+ 0-43 



-0-78 



1845 



+ 0-32 



-0-48 



+ 0-68 



-0-54 



1846 



+ 0-38 



-0-14 



+ 0-60 



-0-84 



1847 



+ 1-12 



-0-29 



-0-62 



-022 



1848 



+ 1-14 



-0-49 



+ 0-24 



-0 88 



1849 



+ 0-59 



-0-44 



+ 0-66 



-0-81 



1850 



+ 0-59 



-0-91 



+ 0-29 



+ 0-04 



Means 



+ 0'-55 



- 0'-36 



+ 0'-33 



- 0'-52 



" These results leave no doubt of the existence of a lunar 

 period. In fact, there are but two instances, out of forty-four, 

 in which the rule announced does not hold. The mean differ- 

 ence of the first and second quarters is 0'-91, and that of the 

 third and fourth 0'-85. 



" It has been stated, that the law of the variation is not ex- 

 hibited distinctly, from day to day, in the separate years. 

 But when Ave combine the results of the eleven years, we ob- 

 tain a series of numbers, in which the law of the change from 

 day to day is evident. The following are the results : 



Day. 



Eange. 



Day. 



Range. 



Day. 



Range. 



Day. 



Range. 







9'-78 



7 



9'-66 



15 



9 A 59 



22 



8'-79 



1 



10-27 



8 



8-91 



16 



10-07 



23 



902 



2 



1015 



9 



8-94 



17 



9-94 



24 



916 



3 



10-77 



10 



8-87 



18 



10-23 



25 



9-14 



4 



10-27 



11 



9-12 



19 



10-10 



26 



8-83 



5 



1024 



12 



8-92 



20 



1000 



27 



938 



6 



9-55 



13 



9-33 



21 



952 



28 



9-60 







14 



10-14 







29 



8-71 



We see here, as before, that the range is greatest in the first 

 and third quarters, and least in the second and fourth. The 



