Oct. U 



• KNOW^LEDGE 



327 



cloud-bands, there are two opposite points of the horizon, 

 whence the clouds seem to radiate as shown in Fig. 3. 



This, Ijy-the-way, may help to e.xplain a phenomenon 

 which is often seen in tropical and sub-tropical regions, 

 and which many observers seem to tind perplexing — ths 

 existence at the same time of two .sets of radiations from 

 the setting or rising sun : for it is clear that if there are 

 a series of straight and parallel rays from the sun near 

 the horizon, these, like the parallel cloud-bands, would 

 appear as a series of radiating beams, not only around the 

 sun, but around the point of the horizon opposite to him. 

 (To be continued.) 



THE MENACING COMET. 

 Bv R. A. Proctor. 



IN my last I promised to show that this comet was seen 

 where the comet of 18-i3 and 1880 could not have 

 been seen — within two degrees. I am unable to do 

 this, simply because there is every reason to believe 



to point out ray own mistake here. I tell him that the 

 mistake arose from an exuberant anxiety to find out the 

 truth, and tell it to readers of KxowLKDOh:. To this he 

 replies, " That's all right, and now's your chance." 



Seriously, I doubt whether many readers of Knowledge 

 could have found out my mistake if I had chosen to leave it 

 alone, but it is a clear duty to correct it I wrote so soon 

 as I had detected it (which was within an hour of my 

 receiving first copy of Knowledge) to the six leading daily 

 papers, to correct it, and, with one exception, they kindly 

 gave me the opportunity. 



Now to give a more correct account of matters : — 

 In Fig. 1, c, p, (I, represents the perihelion portion of the 

 orbit of the comet of 18-13 around the sun S, g^ >)eing the 

 rising node, ?S the descending node, i! Di is the orbit of the 

 earth near the time of the vi^rnal equinox, tlie position of 

 which is indicated by the line S 'Y'- The places of the earth 

 on Sept. 12, 17, 18, IS), 24, and 28, are shown at Q, A, B, 

 C, D, and E respectively, and the direction lines from the 

 earth on those days to the new comet are shown. (The 

 last was added after the block had been already cut) The 

 plane of the paper is supposed, for convenience of reference, 



that the comet which circled close around the sun on 

 Sept 17 is no other than our friend the Menacing Comet, 

 come back not in fifteen years, as the Speclnlor said I said 

 it might, but in less than two years and eight months. 



I had corrected the proof of Prof. Smyth's letter on the 

 comet, when it occurred to me (for my sins, no doubt) to 

 make a rough calculation to see whether the orbit of the 

 oomet of 1813, seen from the earth's place on Sept 18, 

 crossed the position occupied by the comet on tliat day. I 

 turned to my comet portfolio for my diagram of the orbit 

 of comet 1843 — or rather of the part of the orbit near the 

 sun. Unfortunately, the arrow put in to represent the 

 direction of motion had been carelessly drawn the wrong 

 way, making the comet's course direct instead of retrograde, 

 and thus altering the rising node into th(? descending node, 

 and vice vernd. Having occasion for dealing rather quickly 

 with the matter, I did not notice the error, and naturally 

 came to an erroneous conclusion. The Editor of Know- 

 LEDOB remarks on this, that as I have no hesitation in 

 pointing out the mistakes of others, I ought not to liesitato 



to correspond with the plane of the comet's orbit (the 

 comet of 1843 is here referred to), and as the inclination 

 of this plane to the ecliptic is about 3.5J degrees, we must 

 suppose i D t' to be inclined at this angle to the plane of 

 the paper, t being above that plane by tlie distance k /, and 

 r below it by the equal distance k' I'. So much under- 

 stood, the real direction of the lines Qq, Aa, Bh, Cc, Drf, 

 E»?, can be at once recognised. It will l>e seen that they 

 i agree with the observations recorded elsewhere for these 

 respective dates. 



These direction lines are so situated that they cut the 

 plane of the orbit of the comet of 1813, in the points 

 a, b, c, and J, lying on or very near the orbit of tliat 

 comet About the line Qi/ there is room for doubt, the 

 telegram from M. Cruls at Rio Janeiro, having probably 

 been incorrectly rendered. 



The other observations (which are given in detail at 



p. 319) agree so well with the th(>ory that the comet is 



, moving in the orbit of the comets of 1813 and 1880 (at 



, least in the part of tlie orbit near the sun, for at aphelion 



