Q-RUBB — Machinery for the Domes of Observatories. 485 



is to be attributed. For instance, in the celebrated case of the 

 dome constructed for Sir James South's telescope at Kensington, 

 nothing could exceed the care and pains that were taken to ensure 

 the most perfect accuracy and freedom of motion. Every part 

 was made with the greatest pains. The rail on which the rollers ran 

 was turned most carefully. The wheels were large in diameter, and 

 their axes revolved, not in bearings, but on large friction rollers. 

 The wood of which the dome was constructed was well seasoned, 

 and put together in the most careful manner. The result was that 

 the dome worked most beautifully when first laid on its rollers. 

 After a few days, however, it became somewhat stiff, and a little 

 later it got stiffer still, until finally it got so stiff as to be prac- 

 tically immovable. An attempt was made by the Rev. Dr. 

 Robinson, some years after, to get the dome into order. It was 

 again brought to its former state of eflQciency, only to return 

 as before to its inefficient state a few days after. The cause of 

 this was that every part was fitted so accurately that no allowance 

 was made for such alterations as are inevitable in such a structure 

 from temperature and hygrometric changes ; and the moment 

 the slightest change occurred in the form of the dome, all the 

 gearing became locked. 



In the construction of modern domes, any danger of such 

 locking is avoided by designing the revolving machinery in such 

 a way that no alteration of the form of dome within moderate 

 limits will affect the freedom of motion. 



I now proceed to describe the various types of revolving gear 

 used for observatory roofs. The simplest of all probably is that 

 of three or more grooved wheels working in bearings fixed to 

 domes, and rolling on a circular iron rail fixed to the walls (see 

 fig. 1). This form answers fairly well for domes up to a ton or so 

 in weight. The groove in the rollers should be considerably 

 wider than the rail on which it rolls to allow for alteration in 

 form of the dome. If instead of a simple piece of angle iron a 

 cast iron wall-plate be used for this, planed on top and sides, this 

 form answers perfectly for domes up to twelve or even fifteen feet 

 diameter, but when necessary to use more than three rollers, it 

 does not work as well, and there is considerable friction on side 

 flanges from any wind pressure. 



A better form is fig. 2, in which case there is no groove in 



2 Q2 



