XEW OBSERVATIONS OF THE PLAXET MERCURY. A'^^) 
The discordance between the calculated and observed times of the phenomenon 
was therefore : 
5 days 20 h. 
2 days 30 h. 
2 days 23 h. 
3 days 2 h. 
and the mean : — 3 days 4 h. 
To translate these times of retardation into phase angles of lu,^,-, uc may li.if^ 
recourse (1) to the calculated phase; (2) to the phase loss of the drawings nt tlir liih^. 
Thus obtained, the phase loss at these times respectively w.is: 
(1) (2) 
11.0 .... 1 11 .... 1 
18.7 .... 1 13 .... 2 
13.6 .... 4 li . . . . 4 
7.6 .... 3 7 . . • • 3 
and the means, (1) 11°.9 (2) ll-°4 
From the measures of the drawings we find further that the observed i-hnso W^ 
at the times of theoretic dichotomy at these four elongations wa^^ respectively: — 
September 16, 9.6 ... . 1 
October 22, 13.7 .... 1 
January 8, 8.1 . . . • 1 
February 10, 8.0 ...» 1 
Giving a mean of 9°.9 
36. Explanation of it. — The cause of the discordance between the apparen 
theoretic times of dichotomv has not hitherto been satisfactorily explained. D 
of dichotomj 
f Table IX. will, I think, betray 
If the centre of the terminator and 
phn«e th 
inability in the eye to detect the oblique illumination out to its theoretic l.nnt, no d 
cordance between the observed and calculated times of dichotomy conid, ni the ca>- 
a sphere, ensue ; in the case of an ellipsoid, there would be a difference between t 
observed and the calculated times of dichotomy, but as observed dichotomy won 
of 
be just as likely 
crepancy w^ould 
1 lue long ran, tu ul.^ui v^.. —- ---- ^ 
the mean be zero, ^vherea, the discor-lunce oh.ervc^ „ alw.j 
Vol. XII. No. 4. —4. 
