48 



44 minutes, 28 seconds. This interval of time is called a synodi- 

 cal month, or lunar month. It commences from the moment 

 when the moon is directly between the sun and the earth, in 

 which position the moon is said to be in conjunction.'" 



After what I have said already, this passage will require little 

 comment. It is positively untrue, in so far as it asserts or im- 

 plies that the earth's advance in the ecliptic occasions such a 

 change in her relation to the sun, as to require an addition of 

 nearly two days and a half to bring the moon round to the 

 sun, after she has completed her revolution round the earth. 

 The earth's relation to the sun \s invariable^ (except, as already 

 so often mentioned, to the small extent occasioned by the ellip- 

 ticity of her orbit). She finds him in the same place positively 

 every day and every month, while, on the contrary, it is the star, 

 to which our author most erroneously ascribes this invariable- 

 ness, that experiences the change of relative position occasioned 

 by the earth's progressive movement in the ecliptic. It seems to 

 me the plainest dictate of common sense, as well as of right rea- 

 son, that if I were to place a dozen of posts round a circle on 

 the ground, to represent the stars, and one in the centre to re- 

 present the sun, and then to walk round the circle within the 

 line of posts, I should pass every one of them in succession, and 

 change my angular position with respect to each of them at every 

 step of my progress, while, with respect to the solar post in the 

 centre, the relation of the perpendicular axis of my own body to 

 that post would be invariable, always at precisely the same dis- 

 tance, and with precisely the same right angle subtended on each 

 side by my radius- vector with my orbit; and that, in conse- 

 quence, if I were, at each of the twelve posts, to turn once round 

 on my own axis, commencing with the solar post and the radius- 

 vector in line with my nose, my rotation surely would not be com- 

 plete till I found myself return to the same position, with my 

 face direct to the sun ; while, on the contrary, the post which I 

 had last left behind, and the corresponding post 180° distant, 

 would be seen to have changed their places to an angular ex- 

 tent corresponding to their positions in the circle, and pass my 

 nose and back proportionally sooner than the completion of my 

 rotation. And if, instead of making the rotation all at once at 

 each post, I were to turn myself in the same direction gradually 

 and proportionally, degree for degree, during my progress from 

 one post to another, the result would most certainly be precisely 

 the same. Is it not therefore more reasonable to consider the inva- 

 riable right angles of the radius-vector, rather than the continually 

 changing angles subtended by the stars, as the true indications of 

 the completion of the planet's or satellite's rotation ? Now look 

 to the diagram, fig. 6, which will answer equally well to illustrate 

 the subject of the earth's rotations and the moon's revolutions; the 



