SCIISNTIFIC KEW8. V^» 



be »ttrpnie(ito find, ^hat foroiulft have occurred to Mr* 

 Poisson, the absolute summing up of which appeared to 

 him impracticable. His object was to examine the influence 

 of the term of the stcond oider in the expression of tlie 

 velocity of the Earth's rotation. These terms arise from 

 .expanding? into a series the function expressing the sum of 

 the products of the pjass of each body attracting by thats^ 

 of the body attracted, divided by the mutual distance of 

 these bodies. As it is impoi^^ible to calculate all these 

 jterqis, the object is to bring forward only those that merit 

 attention. Mr. P. accordingly examines in the first place* 

 whether even those that depend en the Sup might not be 

 neglected; and he finds, that they are always in fact v«ry 

 small. 



As to i]ye figure of the Earth, Mr. P. supposes, tbat»^ 

 without the action of the Sun and Moon, the Earth would 

 turn precisely round one of its principal axes. This i» 

 justified by the physical state of things, since we do Dot- 

 perceive in the altitudes of the pole, observed at different 

 places, any of the oscillations, that would result from a dif- 

 ferent hypothesis, and the duration of which would be 

 about one year. By similar considerations he expunges 

 the terms relative to the otlier two principal axes, which 

 can never become sensible but on hypotheses of little pro- 

 bability, which would give to the rotary motion of the Earth 

 periods of less than two y<^ars, which have never been oh*" 

 served. He afterwards shows, that the equations to be 

 summed up in the successive approximations preserve the 

 same form ; whence be concludes, that the axis of rotation 

 will always coincide nearly with the shortest of the Earth's 

 principal axes, and that the poles will always answer to the 

 same points on the surface. 



But, though the latitudes may not vary so as to deserye Mar th«re 

 any attention, or to be perceptible to astronomers, is the J^^J^jg^^*^^,^' 

 rotary motion so uniform, fi$ has been supposed ? If its ine- 

 qualities be of a very short period, and not very perceptible, 

 they may escape our notice, and yet in a certain degree 

 affect all our observations, and the consequences deduced 

 ft-om them. Suppose, for example, that the pole, instead of 

 the 3^^' of its circle, passes only through 350* ^ and that the 



latitudes 



