i8o SCIENCE PROGRESS 



The evidence seems to point very definitely to a value 

 of about + io"'5 for the lunar acceleration, and to a value 

 for the solar acceleration lying between + i'' and +2", but 

 which is probably closer to the latter figure. 



Interference Methods in Astronomy. — The application of 

 interference methods to observations with the loo-inch Hooker 

 telescope of the Mount Wilson Observatory has recently 

 directed attention to the advantages of this method, which, 

 although not new, has not received much attention from 

 astronomers. The method, however, has for some purposes 

 such marked advantages that it seems desirable to summarise 

 here the principles which underlie it, and the results which have 

 already been obtained at Mount Wilson. It may be recalled 

 that the image of a distant star (point-source) formed in the 

 focal plane of a telescope is not a point of light, as would be 

 inferred from the geometrical theory, but consists, in the 

 normal case, of a circular aperture of a bright central disc sur- 

 rounded by a series of circular diffraction rings whose angular 

 diameters can be calculated. The angular diameter of the 

 first ring is, for instance, given by 1-22 X/D, D being the dia- 

 meter of the aperture and X the wave-length of the light. If 

 two close distant stars are observed, their diffraction patterns 

 will be superposed. If the central image of one falls on the 

 first diffraction ring of the other, a diminution in the intensity 

 between the two nuclei will be observed, and the star will 

 be recognised as double : if the nuclei are closer still together, 

 then a variation in intensity cannot be definitely asserted, 

 though the star-image may be perceived to be elongated. 

 Lord Rayleigh proposed that this limiting case should be 

 considered as the limit of resolution of the telescope, i.e. stars 

 at an angular distance of less than 1-22 X/D cannot be con- 

 sidered as separated by the telescope. This gives a theoretical 

 resolving power of a telescope of aperture D (expressed in 

 inches) of about 57^. Thus, properly to resolve a double 

 star of separation o"-2 5 would require a 20-inch telescope. 



In the case of the image of a planet or planetary satellite, 

 geometrical theory would indicate an image with a perfectly 

 sharp edge. Diffraction, however, produces a more or less 

 general falling off of intensity at the theoretical circumference 

 instead of the abrupt transition. For a given magnification, 

 the greater the aperture of the telescope the more rapid is the 

 falling off in intensity. As a result, the determination of the 

 angular diameters of small satellites, etc., is a very delicate 

 matter, and the values obtained in general depend upon the 

 size of the aperture, the magnification employed, and also on 

 the brightness of the object ; the smaller the telescope, the 

 larger will the measured diameter be. 



