XXXVlll 



entirely original ; for there was no available information except the memoirs 

 of Mudge and Edwards, which were only applicable to specula of four or 

 five inches diameter. Lord Rosse's starting-point was the necessity of using 

 speculum metal of the highest standard. For this he combined four equi- 

 valents of copper and one of tin : this compound is very brilliant and 

 resists tarnish far better than if there be a slight excess of either ingre- 

 dient. A striking example of this is a compound speculum poUshed in 

 1830, which lay neglected in his laboratory, yet a few years ago was still 

 quite bright. Arsenic and other metals which have been recommended in 

 small quantity, he tried, but preferred the simple alloy. This, however, is 

 nearly the m^ost intractable of all materials — harder than steel, more brittle 

 tha«n glass, friable, crystalline, and, worst of all, flying in pieces with any 

 sudden change of temperature. In consequence, the common process of 

 the founder does not avail here, except on a small scale ; and all large 

 specula which had been previously made contained a larger proportion of 

 copper than the above, that they might have the necessary toughness. 

 Yet more, he shared the general belief that the polishing could only be 

 done by hand, that it was essential to feel the nature of the contact, 

 that therefore only small surfaces could be accurately figured. For both 

 reasons he was led to build up a speculum of small pieces. His first 

 attempt was very ingenious. He combined a central disk with annular 

 zones tvfo or three inches broad, and ground and polished them spherical , 

 In such a surface, each zone is of shorter focus than those within it, and 

 the resulting image would be indistinct. But by a fit adjustment, each, 

 beginning at the centre, was drawn back till their foci coincided ; and the 

 action of the compound was good even with an aperture of eighteen inches. 

 Yet the complexity of the arrangement and its liability to change with 

 temperature were weighty objections, and he was led by a new fact to 

 devise another plan. In polishing these rings, he found the outer ones too 

 large for hand- work, and made in 1828 a machine which gave a rectilineal' 

 motion to the polisher, while the speculum revolved slowly. This was 

 only expected to produce a spherical figure, but it caused rings and other 

 irregularities, which he saw vt^ould be corrected by adding a second excen- 

 tric, which, by giving a lateral motion to the polisher, changed its course 

 into a curve which might be varied from a right line to an ellipse almost 

 circular. This had the desired effect ; but on watching its action he saw 

 that it would also give whatv^^as the great desideratum, a change of curva- 

 ture from the centre to the circumference which could be varied at pleasure, 

 and therefore could be made to give a true aplanatic figure. The machine so 

 altered was employed through his whole life with only one important change, 

 which was a means of setting the six-feet vertical while on the polishing ma- 

 chine, and resting it by a terrestrial mark. This was added in 1862, and 

 very much facilitated the obtaining a perfect figure. He met the difficulty 

 of obtaining a large speculum by making a strong frame of a peculiar brass 

 which has the same expansion as fine speculum-metal, and soldering on its 



