[ *85 ] 
tion of the Annulus, was confiderably Below the Weft 
Point of the Sun's Disk $ and the fecond Contaft, at 
the Diflolution of the Annulus, feemed to be about 
io Degrees Eaftwards from the North Point or Zenith 
of the Disk : But I did not find that the Pofttion of 
thofe Points of Contad could be eftimated with E# 
actnefs on feveral Accounts. The Breadth of the An- 
nulus towards the South-eaft Part of the Sun's Disk, 
was at leaft double of its Breadth towards the oppofitc 
Part, about the Middle of this Appearance. An Ap- 
paratus, by which I was in Hopes of being able to de- 
termine thofe things more accurately, was not ready. 
I propofed to have made fome Eftimation of the Ratio 
of the Continuance of the annular Appearance, where 
it was central to its Continuance at Edinburgh , from 
that of the Arithmetical Mean betwixt the Numbers 
that fhould exprefs the Proportion of the greateft and 
lead: Breadth of the Annulus to the Geometrical Mean 
betwixt the fame Numbers 5 or from the Ratio of the 
Radius to the Sine of half the Arch intercepted be- 
tween the two Points of internal Contaft > but I did 
not obtain thefe Ratio's with fufficient Exa&nefs. 
At 3 Ho. 31 Min. 43 Sec. the Annulus was dif 
folved, after having continued y Min. 48 Sec. And 
here again our Signals were heard immediately after 
one another: The Middle of the Eclipfe was there- 
fore at 3 Ho. 28 Min. 49 Sec. In this the Time by 
Obfervation did not agree fo well with the Time by 
Computation as in the Beginning of the Eclipfe, the 
Difference being here about four Minutes. The Irre- 
gularities of the Moon's Surface occafioned the fame 
Appearances, in fome meafure, as at the Formation of 
the Annulus . When I returned to the Heliofcope, 
A a 2 there 
