Pursuant to the arrangements made with the Carnegie In- 

 stitution at Washington, in September 1906, Mr. John Daggett 

 Hooker placed an order with the French Plate Glass Companies 

 at St. Gobain, France, for the casting of a glass disk one hun- 

 dred inches in diameter to be used for a reflecting telescope on 

 Mt. Wilson, near Los Angeles, California. 



In December, 1908, the glass disk, 102 inches in diameter, 

 I31/0 inches thick and weighing 4% tons, reached Pasadena. 

 Prof. Ritchey immediately discovered that in many respects the 

 disk was imperfect. It had been cast or poured from three pots 

 in succession, giving an effect of three layers, and it contained 

 millions of air bubbles and striae which gave it a milky ap- 

 pearance ; therefore it showed an entire lack of homogeneity so 

 necessary for equable expansion and contraction of a perfect 

 mirror disk. 



It was then thought that this disk could not be made avail- 

 able and Mr. Ritchey departed almost immediately for France, 

 with the request from Mr. Hooker that he stay long enough 

 to see the institution of proceedings for a perfect casting. Mr. 

 Ritchey remained at St. Gobain several months and gave his 

 time and experience to this problem, but although many cast- 

 ings have been made, down to the present time, no perfect disk 

 has been achieved. 



The use of metal for reflectors has practically been aban- 

 doned. Metal is fibrous and granular in the finest castings, and 

 irregularities in its surface will be apparent after the most 

 careful polishing. No combination of iron, tin, zinc or alumi- 

 nium can give a surface equal to what may be obtained from a 

 vitreous substance; hence the insuperable superiority of glass, 

 quartz or other vitreous matter, which can be polished to such 

 a degree of perfection that a microscopic inspection will reveal 

 no irregularities in its polishing or in its silvering. It will ap- 

 pear as limpid and pure as a transparent liquid surface. An- 

 other factor in the superiority of glass over metal is its greater 

 •rigidity, so that it is less subject to flexure. 



It has been a subject of inquiry why a glass disk, even 

 with air bubbles and other imperfections in the interior, will 

 not be as good as a perfect casting, inasmuch as its surface will 

 be silvered, and no part of the glass itself will be visible or used 

 as a reflector. If these imperfections are so far within the disk, 

 below the surface, that in grinding and polishing they are not 

 reached, it will be possible to use the castings, but by reason 

 of its not being homogeneous, it will be subject to unequal or 

 irregular expansion and contraction throughout the entire body. 

 The manufacturers at St. Gobain overlooked this fact, and al- 

 though they knew of these interior blemishes, in sending the 

 casting they evidently thought that it would answer, as it could 

 be given a perfect surface. 



Upon a more careful examination of this disk, Mr. Ritchey 

 found that the nearest imperfection would be about 14 of an 

 inch below the surface when polished and figured, and after 



44 



