88 



KNOWLEDGE. 



March, 1911. 



the concave mirror M^. It is then acted on by the 

 mirror-prism combination, falls on the other 

 concave mirrOr Mo. and is brought to a focus on 

 the photographic plate at P. The concave mirrors 

 are placed at their calculated distances and an\- 

 necessar\- adjustment is then done bv sliding in or 

 out the slit until the image of the latter, when 

 illuminated with Na. light, is perfectly sliarp nn the 

 ground glass plate. Then, since the focal length of 

 a mirror is independent of the colour of the light, all 

 the spectrum is in focus away into the ultra-violet. 

 Suppose that a quartz prism is being used and 

 that the photographic plate is replaced by a fixed 

 evepiece with crosswires. Then, if the mirror M., is 

 adjusted so that the Na. lines coincide with the 

 crosswires when the\' arc at niinimuni dex'iation. and 

 the prism table be rotatecl. e\L-rv line comes into 

 minimum deviation as it reaches the crosswires. 

 That is, we obtain maximum definition automaticallv. 

 The same holds if we are examining the infra-red 

 with a linear thermopile. The thermopile remains 

 fixed and we move the spectrum across it and ever\' 

 line as it reaches it mo\es into perfect focus, a 

 pleasant contrast to the quartz spectroscope where, 

 for ever\- wave-length, we have to adjust both for 

 minimum deviation and the correct distance of the 

 theruKipile from the lens. 



Other advantages of this mirror spectroscope are 

 its compact form and the absence of diffuse light. 

 When light falls on a lens, eight per cent, is reflected 

 back : here the light not used is absorbed by the 

 mirror. Also, there are no tubes to reflect light at 

 grazing incidence. The instrument may also he 

 used as a monochromatic illuminator. 



For the mirrors I have used plate glass and cheap 

 concave lenses silvered. As it is the outside surface 

 of the silver that is used, it is better to send the 

 glass to an optical firm to be silvered. In calculat- 

 ing the position of Mj and M., allo\\ance must be 

 made for the obli<]uit\- of the incidence. The 

 correct distance between S and Mj or P and M., is 



, , r cos '/' , . , , , ,, 



not r/i but 7^ where </> is tiie angle between tlie 



incident beam of light and tile normal to the 

 mn ror. 



The disadvantage oi silver mirrors is that the)' 

 reflect light at 310 /u/n. ver\" poorly. Consequently, 

 that part of the si)ectrum is usually wanting, 

 although the region beyond comes out well enough. 

 Spiegel magnaliuni mirrors, howexer. reflect well to 

 the very end of the si)ectrum ; I ha\e no experience 

 of them. but. according to the tables, if they are used, 

 the spectrum should be ex'erywhere as bright as w ith 

 ipiartz lenses. 



A NATIVE iX .MULRXI.XG I-UR A DUG. 



By the courtesy of Mr. Fretlerick lionneN' we are 

 enabled to imblish the accompan\ing photograph 

 of an old native woman of Australia 

 who is shown in "half-mourning" 

 (num-muyn-ka). Mourning con- 

 sists of cox'ering the head with 

 plaster, and in half - mourning a 

 broad band of the latter is put 

 over the head from front to back. 

 The remains of this are seen in 

 the picture, the special interest 

 of which is that the signs of 

 sorrow were put on, not owing 

 to the loss of a human relative, 

 but on account of the death of a 

 favourite dog. 



After the period of mourning is 

 over, the plaster must not be 

 removed, but allowed to crumble 

 away. Incidentalh' it ma\- be said 

 that the bag which the wDinan is 

 netting is one that is usualh' 

 carried suspended between the 

 shoulders, the ends being tied 

 under the chin. In such a bag, 

 the women carry about small 

 possessions or an infant if they 

 are moving about with one. 



The photograph was taken in Ajiril. 1S,S 

 Momba. l\i\'er Darliii!/. New South Wales. 



1. at 



An Abiiri.^inal Wnin.in of Australia nettint,' a bag. 



The plaster on her licid is the remains of half-mourning after the death of her dog. 



