564 REPORT — 1881. 



The Nicol prism lias several defects. Its field is narrow, being limited on one 

 side by the ordinary ray, which ceases at a certain angle to be totally reflected ; and 

 at the other by vanishing of the extraordinary ray by total reflexion ; the limit of 

 the field in this direction being further reduced by tlie occurrence of a blue iris, 

 within which faint interference-fringes may be discerned. The limitation of the 

 extraordinary ray ari,ses from the arrangement of the prism, in which tlie 

 inclination of the incident rays to the optic axis difters widely, producing the 

 effective residt that at a certain angle the extraordinary index is at a value not 

 only greater than its minimum, but greater than that of the balsam. 



The author's new prism obviates this, by employing spar so cut that the optic 

 axis is exactly at right angles to the axis of the prism, and that the balsam film 

 is parallel to the optic axis. The extraordinary rays, which lie in a plane containing 

 .the axis of the prism and at right angles to the optic axis, are aU therefore 

 transmitted, and the limitation on this side of the field Is therefore removed. Of 

 other rays It will be noted that, as the angle at which they cut the optic axis 

 diminishes, their angle of incidence on the balsam film increases, instead of 

 diminishing. This is, however, of no moment, as tlie aperture of the field In a 

 principal plane of section containing the optic axis of the crystal is still over 90°. 

 The available aperture of the new prism Is about 10° greater than that of the Nicol 

 Prism, and Its ' lateral ' aperture Is as great as that of the Nicol. 



The end faces are also more ueai-ly normal to the axis of the prism, thereby 

 reducing loss of light by reflexion. 



12. On an Overlapping Spectroscope. Bij James Love, F.B.A.8., F.G.8. 



In the in.strument described by the author, one spectroscope was fixed, another 

 being capable of rotation about Its own axis, and of a small angular movement In 

 their common plane, about a line passing through the point of intersection of their 

 axes. The light which had passed through the movable spectro.scope was refiected 

 at the surface of the dispersion prism of the fixed spectroscope, and entered the eye 

 looking along its axis. The results obtained by the author, in working with 

 this instrument, dlfier in some respects from those already described by previous 

 observers. 



13. On Change of Density at the Melting Point. 

 By James Love, F.B.A.S., F.G.8. 



It is commonly supposed that bodies, as a general rule, contract their 

 dimensions, and rapidly become more dense, as they pass from the liquid to the 

 solid condition ; and that those bodies which expand and become less dense on 

 solidifying, such as water and cast iron, are exceptions to the general rule. 



For the purpose of testing this so-called rule I have experimented on a few 

 bodies at their melting points, and find that, as far as the experiments go, the 

 ' exceptions ' are sufficiently numerous to form a class by themselves. The method 

 used was to heat the body to its melting temperature and observe whether the 

 solid woulijloat or sink in its own liquid. The bodies experimented on may be 

 arranged into two classes, iiamely: Class A. Bodies whose solids sink in their 

 own liquid, and Class B. Bodies whose solids ^oat in their own liquid ; or, for 

 shortness, sinking solids and floating solids. They are as follows: — 



Class A. — Sinking Solids. 

 Lead ; tin ; solder = 1 sn. + 1 pb. ; 

 Resin ; tallow ; butter ; lard ; beef fat ; mutton fat. 

 Class B. — Floating Solids. 

 Antimony ; copper ; zinc ; cast Iron ; 

 Wrought iron upon cast ; phosphor bronze ; 

 Gunmetal = 16 cu. + 2 sn. + 1 zn. ; 

 Brass = 16 cu. + 8 zn. + 1 pb. ; 

 White metal = 1 cu. + 6 sn. + 12 zn. ; 

 Another wliite metal = 2 cu. + 3 sb. -t 24 sn. ; 



With most of these bodies the observations were easily made, as the sinking 



