Jan. 24, 1878] 



NATURE 



241 



tions for putting the models together, which consist of flat 

 cardboard marked in accordance with the printed descrip- 

 tions. The whole is contained in a neat quasi-envelope 

 (nine inches by seven). 



LETTERS TO THE EDITOR 



\The Editor does not hold himself responsible for opinions expressed 

 by his correspondents. Neither can he undertake to return, 

 or to correspond with the writers of, rejected manuscripts. 

 No notice is taken of anonymous communications. 



The Editor urgently requests correspondents to keep their letters as 

 short as possible. The pressure on his space is so great that it 

 is impossible otherwise to ensure the appearance even of com- 

 munications containing interesting and novel facts. ^ 



Glass for Reflectors 



Your last number (vol. xvii, p. 226) contains a very interest- 

 ing paper by Mr. Norman Lockyer, in which that gentleman 

 quotes the following passages from Mr. Grubb's paper : — 



" For the 4-foot disc of glass for the Paris reflector, in place 

 of that which has so recently resulted in failure, the St. Gobain 

 Glass Company require twelve months' time to perfect (although, 

 be it remembered, the quality of the glass is here of no conse- 

 quence whatever) ; and I have been myself in correspondence 

 with the principal glass manufacturers here and on the Conti- 

 nent, and not one of them is willing to undertake even a 6-foot 

 glass disc ; so that it would appear that, above that size, the 

 silver-on-glass mirrors are out of the question," . . . . " The 

 other great difficulty in the manu'acture of reflectors is the an- 

 nealing of the disc, and I believe it is this difficulty which limits 

 to so narrow an extent the production of glass discs for silver- on- 

 glass mirrors." 



It may be interesting to your readers to know that an attempt 

 is now being made to entirely overcome the apparently insur- 

 mountable difficulties so clearly pointed out by Mr. Grubb, and 

 to obtain at any time v/ithout delay, and at a very small cost, 

 discs of glass suitable for making silvered reflectors from 6 to 8, 

 or even 10 feet in diameter. 



It is almost impossible to over-rate the difficulty of producing 

 massive discs of glass such as the one employed for the 473-inch 

 reflector of the Paris Observatory, weighing, as it did, no less 

 less than 1,546 lbs, in the rough, for, however carefully annealed 

 such a mass of brittle, slow-conducting material may be, it will 

 always be liable to unequal expansion, deflection, and fracture. 



Fortunately, however, we have commercial plate glass to fall 

 back upon ; plates of i to 1 4 inch in thickness can be readily 

 made and perfectly annealed, and it is to the substitution of 

 these large and comparatively thin sheets of glass, in lieu of 

 thick cast masses, that my attention has been chiefly directed. 



It is perfectly well known that plates of i to i^ inches in 

 thickness, if of large area, are subject to a great amount of 

 deflection and consequent distortion of the image, which no 

 ordinary support or backing can prevent. Several modes of 

 converting such thin discs of commercial plate glass into efficient 

 reflectors are about to be put to the test of practical experiment 

 for the 5oJ-inch silver-on-glass reflecting telescope which I am 

 making and erecting at my residence on Denmark Hill. 



Any attempt to support a disc of this diameter of i J inch in 

 thickness against a cucheon of any kind, or loosely against a 

 plane, must end in failure ; nor can we hope to escape the diffi- 

 culty by cementing the glass to any foreign substance whose 

 power of conducting heat and rate of expansion diflers from that 

 of glass, as a giving way of the cement would be only a question 

 of time, while distortion would result from unequal expansion of 

 the two different materials. An intermediate course has there- 

 fore been adopted. 



A strongly ribbed hollow cellular casting is made of iron 52 J 

 inches in diameter, and 13 inches in thickness, weighing 1,400 

 lbs. ; after slowly cooling in its mould, it will be again heated to 

 about 900" F., and then be again slowly cooled ; the whole of the 

 external skin of the casting will be turned off' in the lathe, and 

 its face made into a true plane, less the final process of scraping ; 

 it will then be thrice annealed in oil, each time slowly raising 

 the temperature from 60'' up to 600° F,, and each time slowly 

 cooling it again. When all undue tension has thus been got rid 

 of, its face will be finally scraped, to a true plane, and a small 

 spiral channel ^V of an inch deep, and the same in width, will 

 be formed on the flat face, the channels being about \ an 



inch apart from each other, and extending from the centre 

 nearly to the outer edge of the metal surface. One side 

 of the glass disc having been previously ground flat by 

 the plate-glass manufacturer, will have a second grinding 

 on the grooved plane, so as to insure perfect contact all 

 over its surface ; the emery having been all carefully removed, 

 the surface of the iron plane is to be slightly moistened with 

 olive oil, and the disc of glass replaced upon it. A flanged iron 

 ring will then be placed around the glass disc, and screwed firmly 

 to the iron surface, leaving a clear annular space of about | of 

 an inch wide between the periphery of the glass disc and the 

 ring ; a permanently tenacious viscous matter (of the character of 

 soft marine glue) will then be poured into this annular space, 

 forming an air-tight junction between the iron plate and the glass 

 surface, and at the same time admitting the glass to expand or 

 cortract freely. A partial vacuum will then be formed beneath 

 the glass by exhausting the air through a central hole communi- 

 cating with the spiral groove ; the glass disc will then be held 

 firmly in contact with the entire surface of the iron plane, which, 

 however, is free to slide under the glass when undergoing 

 expansion or contraction. I have found by repeated experi- 

 ments (many years ago) that plate glass (say of \ of an inch only 

 in thickness) so held on to an unyielding plane, may be 

 repeatedly struck by the rounded face of a heavy wooden mallet, 

 with the greatest violence, without producing a single fracture, 

 so complete is the support thus afforded. 



It is important to bear in mind that a glass disc so held does 

 not rest on its lower edge when placed in a vertical position, nor 

 are the upper portions of the plate allowed to press on, and be 

 supported by the lower ones, as is inevitably the case with a 

 mirror loosely suspended in a sling in the ordinary manner, but 

 on the contrary, every portion of the glass disc is sustained and 

 supported in position by atmospheric pressure, and held flatly 

 and firmly against a corresponding portion of the unyielding iron 

 plane, free from any accumulated downword pressure. 



The expansion by heat of plate-glass and cast-iron are in 

 round numbers as 19 is to 22, and the differential amount of this 

 expansion between the extreme range of summer and winter 

 temperatures, would cause the iron to exceed the diameter ot 

 the glass by about ^^^ of an inch ; — this minute sliding motion of 

 the two smooth planes upon each other would not in the slightest 

 degree alter the curved face of the mirror. 



The glass disc having been thus finally and permanently 

 attached to the iron plane, the latter would be supported in its 

 cell by bands passing round it as usual, and with a system of 

 triangular supports at the back. The weight of this strong-ribbed 

 hollow cellular plane, of 13 inches in thickness, is only 

 1,400 lbs., while a disc of equal diameter in speculum metal, if 

 only 4 J inches in thickness, would weigh about 2,700 lbs. ; 

 hence such a compound metal and glass reflector is lighter than 

 a solid cast glass one, and but little more than one-half the 

 weight of a reflector made of ordinary speculum metal, while its 

 thickness being three times as great as the latter, it would, when 

 in use, and also while undergoing the polishing operations, be 

 perfectly free from deflection. 



Hitherto I have spoken only of the mode of mounting" the 

 glass disc on its iron support ; it now remains to convert the 

 flat surface of the glass disc into a shallow concave reflector. 

 For this purpose I have made experiments in turning glass con- 

 caves with a diamond-cutting tool mounted on a slide-rest, and 

 I have found that in this way glass affords nearly the same 

 facilities for shaping in the lathe that iron or brass would do 

 under similar conditions ; it therefore follows that lenses of all 

 shapes and sizes may be brought approximately to a true figure 

 with very great ease and rapidity. 



Satisfied with this result, I am now erecting in my laboratory 

 a lathe of peculiar construction and specially adapted to this 

 purpose with a bed fifty feet in length, and having a fifty-four 

 inch diameter face plate at each end of the mandril. A massive 

 radius-bar or frame of double the intended focal length of the 

 reflector, moves on an adjustible pivot attached to the lathe-bed, 

 while the other end of the radius frame carries a slide- rest in 

 which a diamond-cutting tool is mounted, and by means of 

 which a spherical concavity is rapidly and truly turned over the 

 whole face of the glass disc, and of any desired radius, while a 

 second plate of glass or metal is turned into a convex surface on 

 the other face-plate of the lathe, thus furnishing a convex grinder 

 of the exact same radius as the concave reflector. Special 

 arrangements are made to neutralise any difference in the length 

 of the radius-frame by expansion or contraction during the 

 turning operation, and provision is also made for gauging to the 



