232 



NATURE 



\7nly i8, 1872 



within the principal focus. Before these holes short tubes were 

 fixed with screws ; in these tubes slide suitable holders for c;irry- 

 in^ electrodes or vacuum-tubes. The spark is thus brought at 

 once nearly into the axis of the telescope. Tlie final adjustment 

 is made in the following manner :^A bright star is brought into 

 the centre of the field of an ordinary eye-piece ; the eye-piece is 

 then pushed within the focus, when the wires or vacuum-tube can 

 be seen across the circle of light formed by the star out of focus. 

 The place of discharge between the electrodes or the middle of 

 the capillary part of the vacuum-tube is then brought into the 

 centre of the circle of light. The vacuum-tubes are covered 

 with black piper, with the exception of a space about a ] inch 

 long in the middle of the capillary part, through this small un- 

 covered space alone can the light escape to reach the slit. 



The accuracy of both methods of comparison, that by reflec- 

 tion and that by the spark within the tube, was tested by the 

 comparison of the three bright lines of magnesium and the 

 doulde line of sodium with the Fraunhofer lines /' and D in the 

 spectrum of the moon. I greatly prefer the latter method, be- 

 cause it is free from several delicate adjustments which are 

 necessary when the light is reflected, and which are liable to be 

 accidentally displaced. 



Spectroscope A is furnished with a sincile prism of dense glass 

 with a refracting angle 59° 42', giving 5° 6' from A to H. 



Spectroscope B has two compound prisms of Mr. Grubb's 

 construction, which move automatically to positions of minimum 

 deviation for the different parts of the spectrum. Each prism 

 gives about 9° 6' for minimum deviation from A to H. 



Spectroscope C is furnished with four similar prisms. 



The small telescopes of the three spectroscopes are of the same 

 size. Diameter of object-glass I J inch ; each is furnished with 

 three eye-pieces magnifying 5"5, 9'2, and i6-o diameters. 



Spectrum of the Nebula of Orion 



With spectroscopes A and B four* lines are seen. 



First line. — With spectroscope B and eye-piece i and 2, the 

 slit being made very narrow, this line was seen to be very 

 narrow, of a width corresponding to the slit, and defined at both 

 edges, and undoubtedly not durable. The line of nitrogen wlien 

 compared with it appeared double, and each component nebulous 

 and broader than the line of the nebula. This latter line was 

 seen on several nights to be apparently coincident with the mid- 

 dle of the less refrangible line of the double line of nitrogen. 

 This observation was on one night confirmed by observation with 

 the more powerful spectroscope C. 



The question suggests itself whether, under any condition^; of 

 pressure and temperature, the double fine of the nitrogen-spec- 

 trum becomes single ; and further, if this should be found to be 

 the case, whether the line becomes single by the fading out of its 

 more refrangible component, or in what other way the single line 

 comes to occupy the position in the spectrum, not of the middle 

 of the double line, but that of the less refrangible of the lines. 



I stated in my former p.rper that when for any reason the light 

 from the luminous nitrogen is greatly reduced in intensity, the 

 double line under consideration is the last to disappear, and con- 

 sequently a state of things may be found in which the light of 

 nitrogen is sensibly monochromatic when examined with a narrow 

 slit.f Under these circumstances the line of nitrogen appears 

 narrower, and the separate components can be detected with 

 difficulty, if at all. 



I stated also that the breadth of the component lines appears 

 to be connected with the conditions of density and temperature 

 of the gas. As was to be expected from theoretical considera- 

 tions, the lines become narrower and less nebulous as the pres- 

 sure is diminished. My observations of this change seemed to 

 show that the diminution of the breadth of the lines takes place 

 chiefly .at the outer sides of the lines, so that in the light from 

 very rarefied gas the double line is narrower, but the space of 

 separation between the components is not as much wider as 

 would be the case if the lines had equally decreased in width on 

 the sides towards each other. 



When the pressure of the gas is reduced to about 15 inches of 

 mercury, the line spectrum fades out to give place to Pliicker's 

 spectrum of the first order. During this process a state of 



in n-b;ili iS H. IV. (Phil. Tr: 



864, 



• The fourth lili was first 

 p. 441). 



t Phil. Trans. 1868, pp. 340-546. Observations on this point were subse- 

 quently made by Frankland and Lockyer (l^roc. Roy. Soc. vol. -xvii. p. 453). 

 It should be stntcd 'hat they mike no reference to my obs n-vations, though 

 they r;fer to a purely hypothetical suggestion contained in the same paper. 



things occurs when, for reasons already stated, the spectrum 

 becomes sensibly monochromatic when viewed with a narrow slit 

 and a spectroscope of several prisms. The line is narrower, and 

 remains double, and has the characters described in the preceding 

 paragraph. 



As the pressure is diminished, the double line fades out 

 entirely, and the spectrum cf the second order gives place to the 

 spectrum of the first order. When, however, the pressure 

 becomes exceedingly small, from o'l inch to 005 inch, or less, 

 of mercury, there is a condition of the discharge in which the 

 line again appears, while the other lines remain very faint. 

 Under these conditions I have always been able, though with 

 some difficulty on account of the faint light when the necessary 

 dispersive power (spectroscope B with second or third eye-piece) 

 and a narrow slit are used, to see the line to be double, but it is 

 narrower than when the gas is more dense, and may be easily 

 mistaken for a single line. I have not yet been able to find a 

 condition of luminous nitrogen in which the line has the same 

 characters as those presented by the line in the nebula, where 

 it is single and of the width of the slit. 



Upon the whole I am still inclined to regard the line in the 

 nebula as probably due to nitrogen. 



If this should be found to be the case, and that the nebular 

 line has originally the refrangibility of the middle of the double 

 line of nitrogen, then we should have evidence that the nebula 

 is moving from the earth. The amojnt of displacement of the 

 nebular line from the middle of the nitrogen double line corre- 

 sponds to a velocity of 55 miles per second from the earth. At 

 the time of observation the part of the earth's orbital motion, 

 which was from the nebula, was I4'9 miles per second. From 

 the remaining 40 miles per second would have to be deducted 

 the probable motion from the nebula due to the motion of the 

 solar system in space. This estimation of the possible motion 

 of the nebula can be regarded as only approximate. 



If the want of accordance of the line in the nebula with the 

 middle of the double line of nitrogen be due to a recession of 

 the nebula in the line of sight, there should be a corresponding 

 displacement of the third line as compared with that of hydrogen. 

 For reasons which will be found in a subsequent paragraph, I 

 have not been able to make this comparison with the necessary 

 accuracy. 



In my former paper* I g.ave reasons against supposing so Large 

 a motion in the nebula ; these were based on the circumstance 

 that the nebular line falls upon the double nitrogen line, which 

 the present observations confirm. I was not then able to use a 

 slit sufficiently narrow to show that the nebular line is single 

 and not coincident with the middle of the double line of 

 nitrogen. 



I am still pursuing the investigation of the parts of this in- 

 quiry which remain unsettled. 



Second line. — This line was found by my former comparison to 

 be a little less refrangible than a strong line in the spectrum of 

 barium. Three sets of measures give for this line a wave-length 

 of 4,957 on Angstrom's scale ; this would show thit the line 

 agrees nearly in position with a strong line of iron. At present 

 I am not able to suggest to what substance this line belongs. 



This line is also narrow and defined. I suspect that the 

 brightness of this line relatively to the first line varies in different 

 nebulce. 



Third and fourth line. — My former observations show that 

 these lines agree in position with two lines of the spectrum of 

 hydrogen, that at F and the line near G. 



These lines are very narrow, and are defined ; the hydrogen, 

 therefore, must be at a low tension. 



The brightness of these lines relatively to the first and second 

 lines varies considerably in different nebulne ; and I suspect they 

 may also vary in the same nebula." at different times, and even in 

 different parts of the same nebula, but at present I have not 

 sufficient evidence on these points, t I regret that, in conse- 

 quence of a continuance of bad weather, I have not yet been 

 able to obtain decisive observations as to the possible motion of 

 the nebula in the line of sight. With spectroscope B and eye- 



• Phil. Trans. 1868, pp 342. 543- 



+ .Since writing this sentence I have s en a note by Prof. D'Arrese in the 

 *' Astronomische Nachrichten," No. 1,885. Speaking of the nebula H. IV. 

 37. he says : — " Sein Spectrum istjausser von Hnggins bisher nur noch von 

 Dr. H. Vogel untersucht worden. In No. 1,864, Ast, Nachr. theilt Letzterer 

 mit, trotz'er im Jahre 1S71, im Widerspruch mit Huggins' Angabe, die 

 Linie Neb. {^)=^{2), bisweilen sogar (2) <(3) gefunden haben. Auch Hug- 

 gins war dagegen im Jahre 1864 positiv (2) > (3). 1st Vogel's Beobachtung, 

 wie ich nicht bezweift':, znverliissig, so wird seine Vermuthung einer Veran- 

 derung hier in der That begriindet sein, denn diesen Winter, namentlich ira 



