540 



NATURE 



[April 4, 1901 



westward), and somewhat flatten the curved paths with the 

 wind ; but it would not probably affect materially the general 

 explanation above. 



" H. D. G." truly observes that, owing to the curvature of 

 the earth, his station was nearly 2000 feet below the horizontal 

 line of Spithead, but this also would not materially alter our 

 explanation. 



So far for the sound heard at distant stations not very far from 

 the east-west line from Spithead. When, however, we come 

 to consider the case of Oxford and other places nearly due north, 

 where the sound was heard, and in particular the duration of 

 continuous sound (20 seconds) and succession of sounds and 

 silences noted by Sir W. J. Herschel, our explanation entirely 

 fails. It is true that the differences of distance from Oxford to 

 the different ships were so small that the impulses from all the 

 forty-six ships. must have arrived within the interval of about 

 half a second, but they would have diminished in intensity 

 (according to the law of the inverse square of the distance, 

 which in the absence of wind in their direction must be 

 assumed) so as to be less strong than the impulse from a single 

 gun at the distance of ten miles, a distance at which, we have 

 seen, it would have been inaudible. 



Was there a southerly current in the upper atmosphere between 

 the Solent and Oxford ? Or was the state of the atmosphere 

 abnormal as to temperature, so that the upper regions were 

 warmer than the lower, as hardly seems probable? Or, lastly, 

 were the sounds heard by Sir W. J. Herschel diffraction effects 

 outside the upward curving sound rays, as the intervals of 

 sound and silence seem to suggest ? Or is any other explanation 

 possible from known laws ? 



Possibly some light may be thrown on these questions by 

 other correspondents, or such experts as Sir G. Stokes, Lord 

 Rayleigh or Prof Osborne Reynolds. 



The Cheibourg Peninsula at its northern end is about the 

 same distance to the south of Spithead as Oxford to the north. 

 It would be interesting to know if the sounds were heard by 

 ships in the channel between the Isle of Wight and Cherbourg. 



RoBT. B. Hayward. 



Ashcombe, Shanklin, March 12. 



The New Star in Perseus. 



In sending you a provisional light-curve of Dr. Anderson's 

 new star for publication in Nature, my principal purpose has 

 been to ascertain the nature of the curious fluctuations in the 

 latter part of its course. I have no doubt that they are real, as 

 even the slight irregularity in the descending curve, about 

 March 5, has been independently detected in Leyden, and pro- 

 bably elsewhere also ; but the periodicity that seems to establish 

 itself in the past six or seven days may be only apparent. In 

 this country the weather has been generally unfavourable for 

 some weeks, and it is possible that astronomers in other parts of 

 the world will be able to fill the gaps between the observed 

 parts of the descending curve (on February 25, March I, 3, 5, 

 6, 13, 17, 20, 21, 22, 23, 25, 27). 



This star is remarkable in still another respect. It is a well- 

 known fact that new stars have almost exclusively made their 

 appearance in the Milky Way ; moreover, it has been pointed 

 out recently by Sir Norman Lockyer in this journal that the 

 NovEe are not equally distributed along the galactic zone; like 

 the " Wolf- Rayet " stars, they seem to avoid the region com- 

 prised between Cassiopeia and Carina. Nova Persei 1901 is 

 no exception to the general rule, it being situated on a feeble 

 distance of the central plane of the Milky Way, but as in the 

 case with the other new star discovered by Dr. Anderson, in 

 Auriga, it lies in a relatively poor region of the galactic zone, in 

 which phenomena of this kind have but rarely occurred. Not- 

 withstanding this, I think that the tendency among the new stars 

 to group themselves in the opposite region of the Milky Way 

 holds good as a rule. C. Eas roN. 



Rotterdam, March 27. 



NOVA PERSEI. 

 'X'HE observations of the new star in Perseus have not 

 *• decreased in interest since they were last referred to 

 in Nature. Strictly according to precedent a nebular 

 spectrum, somewhat similar to that observed by Gothard 

 in Nova Aurigfe, followed the disappearance of the dark 

 lines in the spectrum ; but about the same time a new 



NO. 1640, VOL. 63] 



phenomenon in relation to Nova was observed ; the star 

 behaved like a " collision-variable." 



Sudden changes of magnitude have been one of the 

 most interesting features of this new star. Since the 

 titne (February 23) that the Nova attained its greatest 

 brilliancy, the star gradually diminished in brightness, 

 decreasing rather rapidly till the 13th March, and some- 

 what more slowly up to the 17th. Since this date 

 periodical variations have occurred, the star decreasing 

 to a 5 '5 magnitude star and rising to about 4*2 in a 

 period of three days (about). Thus minima were ob- 

 served on March 19, 22, 25 and 28. On the evenings of 

 the 30th, 31st and April i the star was of mag. 4-2, 4*3, 

 4'4, so that either another minimum had occurred 

 between the times at which these observations were 

 made or the periodicity is undergoing a change. It is 

 important, therefore, that the light of the Nova should be 

 observed as often as possible, so that such changes may 

 be accurately determined. 



Whether this result is due to the complete capture of 

 the denser swarm or to other changes brought about in 

 the sparser one, it is as yet impossible to say. 



We append some extracts from a paper communicated 

 lo the Royal Society by Sir Norman Lockyer last 

 Thursday. 



Colour. — The colour has undergone some distinct changes 

 since the observation on March 5 last, when it was shining 

 with a clarety-red hue. On the 9th and loth it was observed to 

 be much redder, due probably to the great development of the 

 red C line of hydrogen. 



On the 23rd and 24th the star was noted as yellowish- red, 

 while on the 25th (after the sudden drop in magnitude) it was 

 very red with perhaps a yellow tinge. 



Since that date the star has again become much less red than 

 formerly, and on April i was distinctly yellow with a reddish 

 tinge. 



The Visual Spectrum. — Since March 5 the spectrum has 

 become much fainter, the bright lines of hydrogen being 

 relatively more prominent than they were before ; indeed, C 

 and F throughout this period have been the most conspicuous 

 lines, especially the former, while the bright lines XA5169, 

 5018 and 4924, and the line in the yellow at or near D, were 

 the most prominent of the others. 



All these lines have been gradually becoming weaker, but 

 there is an indication that A 5018 has been brightening rela- 

 tively to \ 5169. 



Accompanying the great diminution in the light of the Nova 

 observed on the evening of the 25th, the spectrum was found to 

 have undergone a great change : the continuous spectrum had 

 practically disappeared, and a line near D (probably helium D*) 

 became more distinct. The other lines were hardly visible. 



The Photographic Spectrum.— The spectral changes recorded 

 in the photograph in one part of the spectrum follow suit with 

 tho'?e observed visually in the other. 



On March 6 the photograph was very similar to those obtained 

 in the earlier stages, the only apparent difference being in the 

 relative intensity of the bright hydrogen lines as opposed to 

 those having other origins, most of which have been shown to 

 be probably due to iron and calcium. The hydrogen lines 

 have sensibly brightened, while the others have become much 

 feebler. 



The photograph of March 10 .'shows a further dimming of 

 the bright lines other than those of hydrogen. 



On Match 25, when the next good photograph was taken, the 

 spectrum had undergone great modifications. The hydrogen 

 lines are still very bright, though they do not show the structure 

 which they did in the photographs taken between February 25 and 

 March 10. The bright lines other than those of hydrogen which 

 are seen in the earlier ^photographs have now disappeared and 

 other lines become visible. The continuous spectrum has also 

 greatly diminished. 



Rough determinations of the wave-length of these new lines 

 have been made by Mr. Bixandall by interpolation between the 

 hydrogen lines. They are as follows : — 



387. Brood and merging into 11^(3889). 



436. Faint. 



447. Not very strong. Probably Helium (A4471 '6). 



