48 PROCEEDINGS OF THE 



a most beautiful spectrum of the fivst class. After interposing the jar, a splendid 

 spectrum of the second class will be seen. But here the case is more complicated 

 yet. The above meniioned spectrum of the first class is not a simple one, but it is 

 produced by the superposition of two spectra of the same class. Ignited nitrogen 

 at the lowest temperature has a most beautiful colour of gold. Wheu its temper- 

 ature rises, its colour suddenly changes into blue. In the first case, the corres- 

 ponding spectrum is formed by the less refracted bands extended towards the vio- 

 let part ; in the second case, it is formed by the more refracted band of the paint- 

 ing extending towards the red. Nitrogen, therefore, has two spectra of the first 

 class and one spectrum of the second class. The final conclusion, therefore, is that 

 sulphur has two, nitrogen three, different allotropic states. It may appear very 

 strange that a gaseous body may have different allotropic states— z. e., different 

 states of molecular equilibrium. It may not appear, perhaps, more strange that a 

 substance, hitherto supposed to be an elementary one, may really be decomposed 

 at an extremely high temperature. From spectral analysis there cannot be taken 

 any objection that sulphur and nitrogen may be decomposed. Chloride of zinc 

 (or cadmium), for instance, exhibits two different spectra. If heated like sulphur 

 and then ignited by the discharge of Ruhmkorff's coil, you will get a beautiful 

 spectrum either of chlorine or of the metal, if either the Leyden jar be not inter- 

 posed or be interposed. There is, in this case, a dissociation of the elements of 

 the composed body in the highest temperature, and re-composition again at the 

 lower temperature. You may consider the dissociation as an allotropic state, and, 

 therefore, I may make use of this term as long as the decomposition be not proved 

 by the separated elements. 



'On the Star Chromatoscope,' by Mr. A. Claudet. — The scintillation and change 

 of colours observed in looking at the stars are so rapid that it is very difiicult to 

 judge of the separate lengths of their duration. If we could increase on the retina 

 the length of the sensations they produce we should have the better means of ex- 

 amining them. This can be done by taking advantage of the power by which the 

 retina can retain the sensation of light during a fraction of time which has been 

 found to be one-third of a second — a phenomenon which is exemplified by the 

 curious experiment of a piece of incandescent charcoal revolving round a centre, 

 and forming a continual circle of light. It is obvious that if the incandescent 

 charcoal during its revolution was evolving successively various rays, we could 

 measure the length and duration of every ray by the angle each would subtend 

 during its course. This is precisely what can be done with the light of the star. 

 It can indeed be made to revolve like the incandescent charcoal, and form a com- 

 plete circle on the retina. When we look at a star with a telescope we see it on a 

 definite part of the field of the glass ; but if with one hand we slightly move the 

 telescope the image of the star changes its position, and during that motion, on 

 account of the persistence of sensation on the retina, instead of appearing like a 

 spot, it assumes the shape of a continued line. Now if, instead of moving the 

 telescope in a straight line, we endeavour to move it in a circular direction, the 

 star appears hke a circle, but very irregular, on account of the unsteadiness of 

 the movement communicated by the hand. Such is the principle of the instru- 

 ment employed by the author to communicate the perfect circular motion which 

 it is impossible to impart by the hand. The instrument consists of a couicai tube 



