r 149 ] 



XVII.— NOTE ON THE EFFECT OF FLEXUEE ON THE PER- 

 FORMANCE OF TELESCOPIC OBJECTIVES. By 

 HOWARD GRUBB, F.R.S. 



[Read, June 18, 1883.] 



During my work on the 27-iiicli objective for the Yienna Refrac- 

 tor, I was led to study more closely than heretofore the effect of 

 flexure on one or both lenses" of a large objective, and as the results 

 which I arrived at, both in actual practice and on consideration of 

 the subject theoretically, were very different to what might at first 

 sight be considered natural, I thought it might be interesting to 

 place the matter on record. 



Let me first, however, explain that this flexure is of most 

 serious consequence in the preparation of object-glasses of large 

 size. It may be thought by those not conversant with the sub- 

 ject that any flexure that could occur in a disc of glass, say of 

 twenty-four inches diameter and from one to two inches thick, 

 from, or due to, its own weight, must be so infinitesimal as might 

 be fairly neglected in the general results ; but this is not so. I 

 can easily demonstrate to yc^u that such flexure is a mechanically 

 measurable quantity, and one-tenth of the smallest quantity thus 

 measurable would be quite sufiicient to make all the difference 

 between a good objective and a bad one. 



This last assertion I am not in a position to demonstrate, but 

 anyone who has attempted to make an objective or a speculum for 

 a reflecting telescope will bear me out in what I say. 



The flexure that I propose to consider in this communication is 

 that due to the weight of the disc of glass itself, which alters the 

 figure or shape of the disc as it lies in its cell, supported round the 

 edge only on three fixed points, or on three fixed and any number 

 of adjustable points, but in all cases round the edge only. So long 

 as the glass is on its polishing machine undergoing the various 

 processes of grinding, polishing, figuring, &c., it can be supported 

 on a perfectly equable support of some kind, such as reflectors are 

 laid on both for manipulation and for actual working ; but it is 



SCIEN. PROC. E.D.S. VOL. IV. PT. III. 



