VIII .
Astronomical Observations .
In two Letters from M. Francis de Zach , Professor of Mathematics , and Member of the Royal Academies of Sciences at Marseilles , Dijon , and Lyons , to Mr. Tiberius Cavallo , F. R. S. Read December 23 , 1784 .
SIR , Lyons , April 4 , 1783 .
I SEND you the account of the observations on the eclipse of the moon , which I have made together with the rev .
Father LE FEVRE , Astronomer at Lyons , in the Observatory called au grand College ; to which I shall add the observations of the vernal equinox ; some observations on Jupiter 's satellites , made at Marseilles by M. SAINT JAQUES DE SYLVABELLE ; and , lastly , a new solution of a problem that occurs in computing the orbits of comets .
If you think that these observations do in any way deserve the notice of the Royal Society , I shall be very glad you would communicate them .
In order to ascertain the going of the pendulum clock , I took several corresponding altitudes of the sun , which you will find in the following table .
On the day of the eclipse the sky was very serene , nothing could be finer , and it continued so during the observation .
I determined to use an achromatic telescope of 31/ 2 feet length , that shows objects in their natural position , because the diluted and uncertain termination of the true shadow of the earth appears more perfectly defined by small than by large telescopes , which magnify too much , and give too great a transit between the penumbra and the true dark shadow .
On that account some celebrated astronomers advise to use for the eclipses of the moon no greater telescopes than of four or five feet length .
It was remarked at Paris , that in an eclipse of the moon , observed through a telescope of DOLLOND , the focus of its object lens being 30 inches , and likewise through a telescope of five feet length ; the eclipse appeared to begin 4~ 7~ sooner , and to end 4~ 7~ later , through the small than through the long telescope ; the like has been remarked by several others , and it has been also observed by myself .
As to my observations I am tolerably satisfied with them , as they do not differ materially from those of Father LE FEVRE , though it is known that in eclipses of the moon no greater exactness than that of a minute can be obtained .
The moon 's spots were carefully observed ; for it is known , that the mean of the observations of the moon 's spots is sufficient to ascertain the longitude of a place to 4~ or 5~ nearly .
M. DE LA LAND comparing the observations of the moon 's spots in an eclipse , made the 22d of November , 1760 , in Vienna , by the Imperial Astronomer Abbe HELL , with those made at the same time in Paris by M. MESSIER , finds the difference of meridians to be 56~ 13~ , which agrees very exactly with that ascertained by other means .
Correspondent altitudes of the Sun taken with a quadrant of three-feet radius , in order to adjust the pendulum clock to apparent time .
Sun 's Ist obserSun 's IId obserSun 's IIId obserSun 's IVth obSun 's Vth obser18th March , 1783 .
altit .
vation .
altit .
vation .
altit .
vation .
altit .
servation .
altit .
vation .
The sun 's upper limb o ' h. ' ' ' o ' h. ' ' ' o ' h. ' ' ' o ' h. ' ' ' o h. ' ' ' at the horizontal 25 40 8 25 2 26 10 8 28 25 26 40 8 31 43 27 20 8 36 16 28 8 40 53 wire , eastern side Sun 's upper limb at the same altitude , 15 12 24 15 91 15 5 39 15 1 10 14 56 29 western side Dividing the sum by 2 23 37 26 23 37 26 23 37 22 23 37 26 23 37 22 Sun 's center on the meridian as marked 11 48 43 11 48 43 11 48 41 11 48 43 11 48 41 by the clock Equation of the day 12 8 15 12 8 15 12 8 15 12 8 15 12 8 15 Clock slower than 0 19 32 0 19 32 0 19 34 0 19 32 0 19 34 equated sloar time The mean of which I put 19~ 32~ Sun 's Ist observaSun 's IId obser19th March , 1783 .
Altit .
tion .
Alt .
vation .
The sun 's upper limb at the horizontal o ' h. ' ' ' o h. ' ' ' wire of the first telescope on the eastern 31 30 91 56 33 9 13 9 side Sun 's upper limb at the same altitude on 14 31 47 14 20 33 the western side of the meridian Dividing the sum by 2 23 33 43 23 33 42 Sun 's center on the meridian as marked by the pendulum clock 11 46 511/ 2 11 46 51 Equation of the day 12 7 57 12 7 57 Clock slower than equated sloar time 0 21 60 21 6 ' ' ' Clock slower than equated sloar time 19th March 21 6 18th March 19 32 Retarding of the clock upon 23 h. 58~ 8~ 1 34 I observed too the mid-day at the great gnomon of the observatory , and found at the same time the meridian line erroneous by 19~ , as you will find in the following tables .
h. ' ' ' h. ' ' ' h. ' ' ' When the center of the sun 's the the the image was on the meridian the 17th 11 50 50 18th 11 48 56 19th 11 47 3 time pointed by the clock was Mar. Mar. Mar. Equations of those days 12 8 33 12 8 15 12 7 57 Retarding on equated sloar time 17 43 19 19 the 20 54 Retarding the 17th 17 43 18th 19 19 Retarding of the clock during those 24 hours 1 36 1 35 I fixed therefore the retarding of the clock 1~ 35~ .
h. ' ' ' h. ' ' ' True mid-day concluded by the sun 's correspon the the 11 48 42 11 46 51 dent altitudes as the clock marked 18th 19th Equation of the mid-day Mar. 18 .
Mar. 18 Retarding of the clock at the rate of 1~ 35~ + 13 + 11 per 24 hours True mid-day the pendulum clock should 11 48 37 11 46 44 have marked Mid-day concluded at the gnomon of the 11 48 56 11 47 3 observatory Difference , the error of the meridian line 19 19 or gnomon From thence I concluded , h. ' ' ' h. ' ' ' Mid-day at true solar time 11 59 60 Mid-day at equated solar time 12 8 15 Mid-day the clock should Mid-day the clock should 11 48 37 11 48 37 have marked on the 18th have marked on the 18th Retarding upon true solar Retarding upon equated 11 23 19 38 time solar time Observations Observations of the moon 's eclipse the 18th March , 1783 .
My observations with an achromatic telescope of 31/ 2 Time marked True or apfeet length .
by the clock .
parent time .
IMMERSION S. The beginning of the eclipse very doubtful .
h. ' ' ' h. ' ' ' Shadow touches Grimaldi 7 41 45 7 53 39 Grimaldi all in the shadow 7 42 54 7 54 48 Shadow touches Mare Humorum 7 49 32 81 27 ~80 21 8 12 16 Copernicus all in the shadow 82 29 8 14 24 Tycho touches the shadow 86 18 8 18 13 Eudoxus 8 12 19 8 24 15 all in the shadow 8 13 43 8 25 39 Mare Serenitatis touches the shadow 8 16 78 28 3 the shadow in the middle 8 21 31 8 33 27 all in the shadow 8 25 54 8 37 50 Proclus touches the shadow 8 32 21 8 44 18 Mare Crisium touches the shadow 8 33 29 8 45 26 shadow in the middle 8 35 36 8 47 33 all in the shadow 8 36 56 8 48 53 Total immersion 8 38 57 8 50 55 EMERSION S. Beginning of the emersion 10 19 57 10 32 2 Grimaldi emerging 10 23 33 10 35 38 all out of the shadow 10 24 9 10 36 14 Mare Humorum emerging 10 29 34 10 41 39 Total emersion of Mare Humorum 10 35 37 10 47 43 Copernicus all out of the shadow 10 43 6 10 55 12 Mare Serenitatis all emerged 10 57 32 11 9 39 Mare Crisium all emerged 11 15 44 11 27 51 End of the eclipse 11 20 10 11 32 18 Total duration 3 39 0 Father LE FEVRE'S observations with a reflector 55 inches focal length , magnifying 300 times .
Time by the Apparent clock .
time .
IMMERSION S. h. ' ' ' h. ' ' ' Grimaldi touches the shadow 7 41 43 7 53 37 Kepler touches the shadow 7 52 283 57 all in the shadow 7 53 24 85 19 Copernicus touches the shadow 80 22 8 12 17 all in the shadow 82 26 8 14 21 Mare Serenitatis touches the shadow 8 16 78 28 3 all in the shadow 8 26 28 37 58 Mare Crisium touches the shadow 8 33 28 8 45 25 all in the shadow 8 36 56 8 48 53 Total immersion 8 38 54 8 50 52 EMERSION S. Beginning of the emersion 10 19 42 10 31 47 Grimaldi emerged 10 23 24 10 35 29 Kepler all out of the shadow 10 35 43 10 47 49 Copernicus all out 10 43 4 10 55 10 Mare Serenitatis all out 10 57 19 11 9 26 Crisium all out 11 15 50 11 27 57 End of the eclipse 11 20 22 11 32 30 Total duration 3 39 20 The observation of the vernal equinox was made at the gnomon .
The height of this gnomon , taken from the center of the hole by which the beams of the sun come in , is 1878 lines of a French inch ; the distance from the bottom of the gnomon to the equinoctial point is 1928 ; the distance from the upper limb of the sun 's image to the equinoctial point was found 16,7 ; the distance from the under limb 23,4 ; the diameter of the hole = 6 ; therefore the distance from the bottom to the upper limb 1928 16,7 +3= 1914,3 , to the under limb 1928 + 23,4 3= 1948,4 ; which gives the time the equinox happened the 20th of March , 5 h. 56~ 52~ P.M. Observations of Jupiter 's satellites at Marseilles .
Apparent Observa1782 time .
tion .
h. ' ' ' April 3 Immersion of the Ist satellite at 2 22 56 good May 19 Imm. Ist sat. 2 48 12 good June 7 Imm. IVth sat. was not total , but its light diminished sensibly without ever disappearing ; the sky was serene , and Jupiter had six belts very distinctly .
20 Emersion of the Ist satellite 1 29 46 good July 5 Em .
Ist sat. 11 43 59 good 13 Em .
IId sat. 9 17 28 good 20 Em .
IIId sat. 9 27 13 good 20 Em .
IId sat. 11 51 59 good 27 Em .
Ist sat. 10 1 11 good 27 Imm. IIId sat. 10 40 33 good Aug. 6 Em .
Ist sat. 8 21 20 good 13 Em .
Ist sat. 10 18 49 good 14 Em .
IId sat. 8 55 39 good Sept. 1 Em .
IIId sat. 9 40 44 doubtful 14 Em .
Ist sat. 86 48 good IT is known , that the indirect method to calculate the orbits of comets in a conic section , by means of three observations given , is rendered more easy and expeditious if there is a possibility of drawing a graphical figure that represents nearly the orbit under consideration , by means of which the calculation is directed , and the required elements of the comet 's path may be rigorously determined .
To draw the orbit of a comet that moves in a parabola or ellipsis , the problem is reduced to find the position of the axis and the perihelial distance ; this position of the axis will be determined as soon as the angle is known , that the axis forms with another line , whose position is given ; this line may be an ordinate to a given point of the curve , or a tangent , or a radius vector , &amp;c. The latter is to be employed in preference , because the perihelial distance being a constant quantity , the angle of position then becomes the true anomaly of the comet ; but as the data of this problem are only geocentric longitudes and latitudes of the comet , deduced from the immediate observations of right ascension and declination , the heliocentric longitudes and latitudes must first be calculated ; but as those data are not sufficient , what is not given must be arbitrarily supposed , viz. the shortened distances ( distantias curtatas ) .
This supposition is changed and altered until the calculation will agree with the three observations , then the difference between two longitudes is the angle comprehended between the two shortened distances in the plane of the ecliptic ; the whole reduced to the plane of the comet 's orbit by means of the heliocentric latitude , gives the difference between the anomalies comprehended by two radius vectors , the problem then is reduced : two radius vectors being given , with the angle comprehended , to find the two true anomalies , the perihelial distance , and the time the comet puts in running its anomalies .
Let therefore ~~~~ represent the ecliptic at an infinite distance ; QPR the apparent elliptical or parabolical path of a comet ; S the sun 's center ; P the comet 's perihelion ; T the place of the earth when the comet was first observed in C ; I the earth 's place when the comet was observed in K ; ST = d , SI = ~ , the distances from the earth to the sun at the first and second observation known by astronomical tables ; let Cm and Kn be two perpendiculars to the plane of the ecliptic , it will be Sm = u , Sn =v the two shortened distances .
The observed geocentric longitude of the comet in T=a=arc ~~ G ; the observed geocentric longitude of the comet in I=~=arc ~~ H ; the geocentric longitude of the sun by tables in T=b=arc ~~ A ; the geocentric longitude of the sun by tables in I=~=arc ~~~ B. Now for the first observation the angle of elongation is b a ; for the angle ATG = arc AG =~~~A~~G= long .
~ long .
comet .
=b a ; the angle of the annual parallax SmT = sin. ( b a)d/ u = e ; 0~ the angle of commutation mST = 180~ e+ ( b a ) = f ; from whence the heliocentric longitude of the comet = b 180~ +f= g. The same at the second observation in ~ .
Angle of elongation =~ ~ ; Angle of annual parallax ~= sin. ( ~ ~)~/ ~ ; 0~ Angle of commutation ~= 180~ ~+ ( ~ ~ ) ; heliocentric longitude of the comet in I=~ 180~ +~= ~ ; putting now the heliocentric latitude seen from S= k ; the geocentric latitude seen from T= l ; the heliocentric latitude will be sin. f. tang .
l/ sin. ( b a ) = tang .
k ; the same with Kn it will be sin. ~ .
tang .
~/ sin. ( ~ ~ ) = tang .
x heliocentric latitude in K. Having thus determined the heliocentric latitudes of two observations , the radius vectors will easily be found in the supposition made for the shortened distances , for they are in the same ratio to the radius vectors as the cosine of the heliocentric latitudes are to the radius = 1 ; therefore the radius vector m of the first observation will be = u/ cos. k and the radius vector of the second observation ~= v/ cos. x~ Taking now the difference between the found heliocentric longitudes , we get the heliocentric motion of the comet upon the ecliptic between two shortened distances , which is to be reduced upon the comet 's orbit , this heliocentric motion is therefore ~g= m. Now to reduce this motion we have , first , sinus totus =1 is to cosine m : : as cotangent k is to the tangent of an angle which I put = n , and 90~ k~n will give an angle which I put = q. Lastly , the analogy cos. n : cos. q : : sin. k : will give the cosine of an angle ~ , which is the required motion upon the orbit , or the angle comprehended between the two radius vectors m and ~ .
Let therefore ECPMND be the apparent parabolic path of a comet ; S the sun 's center ; M and N two places of the comet , the angle MSN equal to its motion in longitude , or the comprehended angle ~ ; P the perihelion ; it is required to find the two anomalies PM , PN , that is , PSM and PSN , the perihelial distance SP , and the time the comet employed to come from its perihelion P to M and N. Resolution .
SM =m In the right-angled triangleSMRand SNV we have SN =~ MR = OS =m sin. ( ~ ~ x ) NV = QS =~ sin. x ; MSN =~ therefore OP = 1/ 4 pm ( sin. ~~ x ) and PQ = NSB =x 1/ 4 p~~ sin. x ; but by the nature of the parabola MSB = ( ~ ~ x ) we have SM = AP + POandSN = AP + PQ ; that is Parameter =pm= 1/ 2 pm ( sin. ~~ x ) ~= 1/ 2 p~ sin. x m+m ( sin. ~~ x ) = 1/ 2 p~~~ sin. x= 1/ 2 p m ( 1 + sin. ~~ x ) = 1/ 2 p~ ( 1 ~ sin. x ) = 1/ 2 p and 1+ sin. ( ~ ~ x ) = p/ 2m 1~ sin. x= p/ 2~ ; by putting into a sum 1+ sin. ( ~ ~ x ) +1~ sin. x= p/ 2m + p/ 2~ ; reduction made 2~ sin. x+ sin. ( ~ ~ x ) = ( m + ~/ 2m~ ) p ; but by trigonometrical formulae we have sin. ( ~ ~ x ) = sin. ~ cos. x~ sin. x cos. ~ .
Substituting this expression in its place we obtain , 2~ sin. x+ sin. ~ cos.x~sin .
x cos.~ = ( m + ~/ 2m~ ) p. By the same formulae we have cos.2 x=1 sin.2 x and cos. x=~1 sin.2 x. Substituting it comes out , 2~ sin. x+ sin. ~ ~1 sin.2 x~~sin.x cos. ~= ( m+~/ 2m~ ) p and sin. ~ ~1 sin.2 x~ = ( m+~/ 2m~ ) p 2~sin .
x~sin .
x cos. ~ .
By casting away the sign of the square root , and reducing to 0 , ( m+~/ 4m2~2)2 p2 +4+ sin.2x + sin.2x cos.2~ 2 ( m+~/ m~ ) p~ ( m+~/ m~ ) p. sin. x~ ( m+~/ m~ ) p. ( sin.x cos.~ ) ~4 sin. x cos. ~+2 sin.2 x cos. ~ sin.2~ + sin.2x sin.2~ = 0 ; disentangling the equation , m2p2 + 2m~p2 + ~2p2/ 4m2~2 +4+ sin.2 x+ sin.2 x cos.2~ 2mp-2~p/ m~ ~ mp sin.x~~p sin. x/ m~ ~ mp sin. x cos. ~~~p sin. x cos.~/ m~ ~4 sin. x~4 sin. x cos. ~+2 sin.~ x cos. ~ sin.2 ~+ sin.2 x sin.2 ~= 0 .
But p being equal to 2~~2~ sin. x , substituting this value , we have 4m2~2~8m2~2 sin. x+ 4m2~2 sin.2x + 8m~3 + 8m~3 sin.2x~16m~3 sin.x + 4~4 + 4~4 sin.2x~8~4 sin.x/ 4m2~2 +4+ sin.2x + + sin.2 x cos.2 ~ 4m~ ~ 4m~ sin. x 4~2 ~ 4~2 sin. x~ 2m~ sin.x 2m~ sin.2 x~ 2~2 sin.x-2~2 sin.2x~/ m~ ~ 2m~ sin.xcos.~-2m~ sin.2xcos.~~2~2 sin.xcos.~-2~2 sin.2x cos.~/ m~ ~ 4sin.x~4sin.xcos.~ + 2sin.2xcos.~ sin.2~ + + sin.2x sin.2~ = 0 .
Dividing the numerator and denominator of the first fraction by 4 ~2 , and of the second by ~ , and putting the whole to the common denominator m2 it comes m2~2m2 sin.x+m2 sin.2x+2m~+2m~ sin.2x~4m~ sin.x+~2+~2 sin.2x~2~2 sin.x+4m2+m2 sin.2xcos.2~+m2 sin.2x-/ m2 -4m2 ~ 4m2 sin. x 4m~ 4m~ sin. x~ 2m2 sin.x 2m2 sin. 2x ~ 2m~ sin. x 2m~ sin. 2x ~ 2m2 sin. x cos. ~ 2m2 sin. 2x cos. ~ ~/ m2 ~2m~ sin. x cos. ~ 2m~ sin. 2x cos. ~~ 4m2 sin. x~4m2 sin. x cos. ~+ 2m2 sin. 2x cos. ~ m2 sin. 2~+ m2 sin. 2x sin. 2~/ m2 = 0 .
Casting away m2 , and reducing , we have m2 + ~2 + ~2 ~2 sin. x+ m2 sin. 2x cos. 2~ 2m~~ 2m~~2m~ sin. x cos. ~ 2m ~ 2m~ sin.2x cos. ~ m2 sin.2 ~+ m2 sin. 2~+ m2 sin. 2x sin. 2~~ ~2m~ sin. x~ 2m2 sin. x cos. ~= 0 .
Putting together sin. 2z and sin. x ( ~2 + m2 cos. 2~ 2m~ cos. ~+ m2 sin.2 ~ ) sin. 2x ~ 2~ ~ ( 2~2 ~ 2m~ cos. ~~ 2m~ ~ 2m2 cos. ~ ) sin. x+ m2 2m~ + ~2 m2 sin. 2~ = 0 .
Substituting cos2 ~=1 sin. 2~ we recover a quadratic equation , ( ~2 + m2 2m~ cos.~ ) sin.2 x~ ( 2~2 ~ 2m~cos .
~~ 2m2 cos. ~~ 2m~ ) sin.x + m2 2m~ + ~2 m2sin.2~ = 0 , and sin.2 x~ ( 2~2 ~ 2m~ cos. ~~ 2m2 cos. ~~ 2m~)/ ( ~2 + m2 2m~ cos. ~ ) sin. x= 2m~ m2 ~2 + m2 sin2 ~/ ~2 + m2 2m~ cos. ~ .
This equation resolved gives sin. x~ ~2 ~ m~ cos. ~~ m2 cos. ~~ m~/ ~3 + m2 2m~ cos. ~=~ 2m~ m2 ~2 + m2 sin.2~/ ~2 + m2 2m~ cos. ~~ ( ~2 ~ m~ cos. ~~ m2 cos. ~~ m~)2/ ( ~2 + m2 2m~ cos. ~)2 which gives farther , sin.x = ~(2m~ m2 ~2 + m2 sin.2 ~ ) .
( ~2 + m2 2m~cos .
~ ) ~ ( ~2 ~ m~ cos. ~~ m~)2/ ~2 + m2 2m~ cos. ~~ ~2 ~ m~ cos. ~~ m2 cos.~~m~/ ~2 + m2 2m~ cos. ~ .
Putting 1 sin.2 ~ instead of cos.2~ , after all reduction made , we recover sin. x=~ 2m3 ~ sin.2 ~ 2m3 ~ cos. ~ sin.2 ~/ m2 + ~2 2m~ cos. ~~ ( ~ ~2 ~ ~m cos. ~~ m2 cos. ~~ m~)/ ~2 + m2 2m~ cos. ~ ; that is , sin. x= ( ~ ~~ m ) .
( m cos. ~ ~ ) ~~ 2m3 ~ sin.2 ~ ( 1 cos. ~ , which gives at last the simple expression sin. x= ( ~ ~~ m ) .
( m cos. ~ ~ ) ~m sin. ~~ 2m~ ( 1 cos. ~)/ m2 + ~2 2m~ cos. ~ .
The angle x defines therefore the position of the axis and the two anomalies required , the perihelial distance being p= 2~ ~ 2~ sin. x , it will be known also by the angle x. In order to find the time the comet employs in running its anomalies , let the perihelial distance just now investigated p be equal to the radius of the earth 's orbit , the parabolic area swept by the radius vector will be by the nature of the parabola 2/ 3 PO x OM + 1/ 2 SO x OM = 4PO x OM + 3SO x OM/ 6 .
Now the periphery of the earth 's orbit is 7 : 22 : : 2p : 44/ 7p ; therefore the whole area 44/ 7p 1/ 2p = 22/ 7 p2 .
It is known that the velocity of a heavenly body moved in a circular path , is to the velocity in a parabolic path in the ratio ~2 : 1 .
If the parabolic area of the comet is divided by ~2 it comes out 4PO x OM + 3SO x MO/ 6~2 equal to an area that the earth describes in the very same time ; put therefore A equal to the time of a sidereal year , we shall recover the analogy ; the whole area of the earth 's orbit is to the time in which it is described as the parabolic area is to the time consumed in sweeping it ; therefore 22/ 7 p2 : A : : ( 4PO + 3SO ) MO/ 6~2 : 7A ( 4PO + 3SO ) MO/ 72p2~2 ; but OM = SM .
sin. anom .
PSM and OS = SM .
cos. anom .
PSM ; let the anomaly be = ~ , we have OM =m sin. ~ , and OS =m cos. ~ ; therefore PO =pm cos. ~ .
Substituting we obtain 7A ( 4p 4m cos. ~+ 3m cos. ~ ) m sin. ~/ 72p2~2 which is 7A ( 4p m cos. ~ ) m sin. ~/ 72p2~2 , whereby the time is found in parts of a sidereal year .
I am , &amp;c. SIR , Lyons , May 4 , 1783 .
LATELY I received from the Observatory at Marseilles the observation of the transit of Mercury , which happened the 12th Nov. 1782 .
The sky not being very favourable , only the two internal contacts were observed ; the first internal contact was observed by M. ST .
JACQUES DE SYLVABELLE , at 3 h. 18~ 30~ apparent time ; the last internal contact by M. ST .
JACQUES , at 4 h. 30~ 16~ ; by M. BERNARD , his Adjunctus , at 4 h. 29~ 13~ .
The nearest distances of Mercury 's limb to that of the sun in the northern part of its disk were at h. ~~ 3 33 14 31 3 42 57 34 parts of the micrometer .
4 22 17 19 The apparent diameter of the sun was 2174 parts of this micrometer : I suppose the before-mentioned 2174 parts = 32~ 26~ , 9 .
I conclude farther , by the observations , the middle of the transit at 3 h. 54~ 7~ , 25 , whereas I fix , by interpolation , the distances of the limbs at 3 h. 54~ 7~ , 25 = 35~ , 6 ; I have therefore semi-diameter of the sun = 16~ 13~ , 4 35~ , 6= 15~ 37~ , 8+ semi-diameter of Mercury = 6~ = 15~ 43~ , 8= to the least distance of centers of the sun and Mercury .
By M. DE LA LANDE'S tables it is 15~ 42~ , only a difference of 1~ , 8 .
M. WALLOT at Paris has observed this transit at the Royal Observatory , h. ~~ First external contact 2 56 28 First internal contact 323 Second 4 17 18 Second external 4 22 53 I only add an important remark upon the diameter of Mercury , which the astronomers supposed in this transit = 12~ .
Let ABC represent the sun 's disk ; in P an external in Q an internal contact ; ANC the apparent path of Mercury over the sun .
The semi-diameter of the sun = 972~ , this of Mercury in our supposition = 6~ , MN = 942~ the least distances of the centers .
In the right-angled triangle MNP it is MP = 972~ + 6~ = 978~ , MQ = 972~ 6~ = 966~ ; therefore NP will be found = 260~ and NQ = 210~ : now NP NQ = PQ = 50~ , which converted into time gives 8~ 14~ for the time the diameter of Mercury employed to run over the limb of the sun ; but by the observations of M. WALLOTI find this time constantly in both contacts 5~ 35~ ; therefore 8~ 14~ : 12~ : : 5~ 35~ : 8~ , 137 , which should be the diameter of Mercury ; and indeed M. WALLO T , by an immediate measure , taken with an excellent wire-micrometer , finds this apparent diameter not greater than 9~ , which sufficiently shows that this diameter supposed 7~ in the mean distance is also too great .
Philos. ~ .
Vol. LXXV .
Tab. VI .
p. 152 .
~ page 145 .
B AP ~S~ T m I C n Q K R G H~ AAA PPP OMOMOM QN BVCVB CBSRSR SRV ENQ DEDQN ED page 147 .
C AN PQ

