26 Lord Oxmantown on a new Reflecting Telescope. 



Two modes have been hitherto adopted for diminishing the 

 spherical aberration in reflectors, the one by rubbing down the 

 outer surface of the speculum from the edge to the centre, so 

 as to make its figure approach to that of a paraboloid, the 

 other by increasing the focal length in proportion to the aper- 

 ture. It is certainly extremely probable that a very skilful 

 workman, who has devoted the greater part of his life to the 

 construction of reflectors, may succeed in some instances, par- 

 ticularly when the instrument is what is technically termed a 

 dumpy, in forming a surface approaching to the paraboloid, 

 which will perform better than one which is truly spherical ; 

 but when we consider the extreme accuracy necessary, and that 

 a true surface can only be obtained by the process of polish- 

 ing, when two motions are combined, the one in some degree 

 at right angles to the other, and that a spherical surface is the 

 only surface which can be formed by these two motions,* it 

 will be evident that when we attempt the parabolic form we 

 abandon an essential requisite to the formation of an accurate 

 surface. It is .scarcely worth while remarking, that in every 

 attempt to improve a speculum of an accurate spherical figure 

 I have invariably rendered it worse ; — these attempts were not 

 on very dumpy instruments. The other method of diminish- 

 ing the spherical aberration by increasing the focal length in 

 proportion to the aperture is certainly unexceptionable ; but it 

 will be immediately evident that it has its limits, and that in- 

 struments become unwieldy after they exceed a certain length. 

 I will now immediately proceed to describe one of the instru- 

 ments I have constructed, with a view of diminishing the 

 spherical aberration without introducing either of these defects. 

 In Fig. 1, Plate I. AB is a brass plate turned true on both 

 sides by means of a slide apparatus, which at the same time 

 renders its sides parallel. The dimensions are seven inches by 

 five-eighths thi^k. CD is another brass plate, made true by 

 the same means, one-half an inch thick. The two plates were 

 then screwed together in a temporary manner, their centres 

 coinciding; three holes were then bored through them one- 



* A plane is a spherical surface with an infinite radius, or in practice 

 with a very great radius, and is extremely difficult to execute. 



