Notes on the Design of Telescope Tubes. 21 



In the Melbourne Telescope the large speculum is a very 

 ponderous affair indeed, containing with its surroundings 

 some tons of metal ; while the small mirror situated at the 

 opposite end of the tube is by comparison a mere feather- 

 weight. Hence the point of attachment of the tube to the 

 declination axis (upon which alone it is supported) is placed 

 very near to the end where the large speculum is fixed. 

 The lower portion of the tube from the main speculum to 

 a point a short distance on the other side of the declination 

 axis is a hollow cylinder of riveted plates of metal very 

 similar to the outside shell of a steam boiler. From that 

 point to the extreme further end it consists of an open 

 latticed arrangement of metal bars. In the Great Paris 

 Reflector — a somewhat similar instrument in other respects 

 — the whole tube consists of a continuous cylinder of boiler- 

 plate. This latter arrangement, while admirable in point of 

 stiffness, is objected to as giving rise to a certain circulation 

 of currents of air of unequal refractive power, and thus 

 impairing the optical performance of the instrument. The 

 former system — that adopted in the Melbourne Telescope — 

 is free from this somewhat serious objection. 



We have thus arrived at these conclusions — 1. That the 

 greater part of the length of the tube of a large reflector 

 must consist of an open framework of thin bars. 2. That 

 this framework will be supported at one end only, where it 

 is united to the cylinder tube, and will be loaded by its own 

 weight and that of the small speculum. 3. That the frame- 

 work must be so arranged as not to intercept any of the 

 rays of light in their course through the instrument. 4. 

 That the framework must be so designed as to secure a 

 maximum of stiffness with a given amount of material ; and 

 5. That it must be equally stiff in every direction. 



In order to comply with condition 3, the bars must be 

 placed altogether exterior to the solid cylinder of rays pro- 

 ceeding to the main speculum, and may be appropriately 

 arranged in the surface of a cylinder or a prism of polygonal 

 section. And in order to comply with condition 4 it will be 

 necessary to revert to the fundamental principles of design 

 of framed structures, and to adopt a method of investigation 

 similar to that employed in designing girders, roofs, and 

 bridges. In fact, the design of our telescope tube is but a 

 particular case, or extension of the old familiar problem of 

 designing an open framed bridge girder; the main difference 



