290 sun's distance. 



the south pole, the movement of rotation shortens both durations in 

 no very unequal degrees. If we take a station between the south 

 pole and the southern limit of illumination, the motion of rotation 

 tends to lengthen the duration, which by virtue of southernly position 

 is shorter, and thereby the inequality is diminished. Thus it appears 

 that the transit of 1874 cannot be used with the same advantage as 

 that of 1882 for determining the Sun's horizontal parallax. 



An examination of the characteristics of these transits, and also of 

 those which occur in the month of June, (as 1761 and 1769,) when 

 the north pole of the Earth is turned towards the Sun, suggests the 

 following remarks: The transits favorable for the determination of the 

 Sun's horizontal parallax are those in which the part of the Sun's disk 

 crossed by Venus has the same name (north or south) as the pole of 

 the Earth which is turned towards the Sun. Now, in general (but not 

 always) the transits of Venus will occur in pairs, (as 1761-1760, 1874- 

 1882,) with an interval in each case of eight years. This interval 

 depends on the circumstances that the transits can only be visible 

 when the conjunction of the Earth and Venus takes place very near to 

 one of the nodes of the orbit of Venus on that of the Earth; and that 

 in eight years Venus has revolved almost exactly thirteen times, so 

 that a conjunction at any one degree of heliocentric longitude is fol- 

 lowed by a conjunction very near to the same degree after an interval 

 of eight years. But in consequence of the proportion of 8 : 13 being 

 not quite exact, and because in eight years Venus revolves a little 

 more than thirteen times, the successive conjunctions take place in 

 2^ days less than 8 Julian years. Therefore, at the second conjunc- 

 tion Venus is less advanced in respect of the node than at the first. 

 At the December conjunctions Venus is near the ascending node; at 

 the June conjunctions she is near the descending node. In the former, 

 therefore, she will be at the second transit more southerly, and in the 

 latter more northerly, than at the first transit. These indications 

 correspond with those of favorable transits. Therefore, in all cases 

 the second transit of each pair is the more favorable for determining 

 the Sun's horizontal parallax. The exceptional case is when Venus 

 crosses the middle of the Sun's disk, as then the latitude of Venus is 

 too great at both the next preceding and the next following 8 -year 

 interval to give a visible transit. 



In the explanation, up to this point, we have gone on the supposi- 

 tion that the observations of transit to be employed are those of 

 duration of transit. And this method possesses the very important 

 advantage that it is entirely independent of the assumed longitude of 

 the place of observation. But there is another method, namely, that 

 of observing the absolute time (as referred to Greenwich time) of 

 ingress only, or of egress only, at different stations on 'the Earth. 

 The best way of considering this is to conceive that figure 3 is not in 

 the plane of a meridian, but in the plane passing through the observ- 

 ing station and through the Earth's centre. Then it is plain that the 

 apparent disturbance of the point S, from the point at which Venus 

 would be seen from the Earth's centre, is in the plane which passes 

 through the observing station and through the Earth's centre. Now, 



