54 



NA rURE 



[November 21, 1901 



LETTERS TO THE EDITOR. 



[ The Editor does not hold himself responsible for opinions ex- 

 pressed by his correspondents. Neither can lie undertake 

 to return, or to co> respond with the writers of, rejccei 

 manuscripts intended for this or any other part of Natuke. 

 No notice is taken of anonvnious communications.^ 



Effect of a Magnetic Field on Spectra of Helium and 

 Mercury. 



The Department of Natural Philosophy in this Universiiy 

 has recently acquired, through a fund provided by the liberality 

 of the Bellahouston Trustees of Glasgow, a fine echelon grating 

 of twenty-six plates made by Hilger, London. The instrument 

 has excellent definition, and its great power enables it to show 

 Zeeman effects with moderate magnetic fields. In conjunction 

 with a new electromagnet which has been constructed for the 

 Department by Me.ssrs. Mavor and Coulson from the same 

 fund, we have been able to make some preliminary observations 

 which may be of interest to readers of N.^ture. The magnet 

 and spectroscope were shown at the recent British Association 

 meeting, when some account of the power of the set of instru- 

 ments was given. The magnet when excited by a current of 

 only five or six amperes gave a field of about 50,000 C.tl.S., 

 and was found capable of giving still higher intensities. 



We have examined the Zeeman cfi'cct for the yellow helium line 

 D3. The line, as is well known, is a doublet ; the wave-length of 

 the brighter component is 5875"8S3 Angstrom units, that of the 

 fainter 5876'2o6 units. The lines broaden with increasing field, 

 till at 6500 C.G.S. the fainter is a distinct doublet. As the 

 field is increased the comjionents of the doublet separate farther, 

 but we have not observed any further splitting. The brighter of 

 the Dj pair is not distinctly resolved till the field is 9100C.G.S., 

 when it appears as a triplet ; it remains so in higher fields. 



In a very high field, the strength of which was not observed, the 

 green (5460) line of mercury w.as resolved into nine components 



We have not so far come across any previous statement of 

 these results, though of course they may be well known to 

 observers of magneto-optic phenomena. We are now arrang- 

 ing for careful measurement by photography and otherwise of 

 displacements produced by fields of known strength, in order if 

 possible to answer some of the outstanding questions on the 

 subject. Andrew Gray. 



Walter Stewart. 



Physical Laboratory, The Universiiy, 

 Glasgow, November 15. 



Observations of Leonid Meteors. 

 Observations were made of the Leonid meteors with the 

 intention of determining the intensity and epoch of the shower. 

 Unfortunately, however, owing to cloudy weather on November 

 13 and 14, observations were restricted to the three following 

 nights. The annexed table gives the total number of meteors 

 observed each night : — 



Extreme limits of the 

 period during which 



Nov. 14 13 0-17 50 

 „ IS II 45-17 35 

 „ 16 12 15-14 15 



Total ... I if 202 .. 212 



During the whole of the time of observation the conditions 

 were practically perfect, save that on the l6th there was a haze 

 over the lower part of the sky. 



The Leonids were rather less numerous during the earlier 

 watches on the 15th than later on, but otherwise no well-defined 

 variation in their number was noticed during the period of 

 observation. The curious manner in which they came in groups 

 was, however, very marked. One rather striking example of it 

 occurred on the 15th, when three spr.ing out almost simul- 

 taneously, followed by another about four seconds later, and this 

 at a lime when the hourly rate was only about twenty. Their 

 brightness varied in most cases between the 15 and the 35 

 magnitude, which was considerably above the mean magnitude 

 of the other meteors. Only two meteors brighter than the first 

 magnitude were observed, both of which were Leonids. The 

 latter of these, which was seen on the i6that I3h. 5Sm., com- 

 menced its path at 1 1 Monocerotis, and passing directly over 

 NO. 1673, VOL. 65] 



II Leporis, disappeared three degrees beyond. It left a train of 

 irregular width, part of which remained visible for about six 

 seconds. 



No very systematic attempt was made to determine the 

 radiant, but as far as could be judged it was at 150° -f 23°, and 

 was sharply defined at any rate for much the larger proportion 

 of the meteors. It is, perhaps, worth mentioning that several 

 meteors very similar to Leonids in appearance were observed 

 to radiate from a point a little above the sickle, and it is pos- 

 sible that some of these, or some from other radiants, may have 

 been recorded as Leonids. Doubtful meteors were in all cases 

 counted with the class to which it seemed mos". likely they 

 belonged. 



Comparing these observations with those which I made last 

 year I should say that the shower was appreciably more intense 

 this year on the 14th and i6th, and much more intense on the 

 15th. I should remark, however, that last year I recorded the 

 maximum as occurring during the latter part of the night of the 

 13th, while on this occasion I have no observations for that 

 period. For the sake of comparison, however, I may say that 

 I consider the shower of the 14th this year about equal in 

 intensity to that of the I3lh last year. E. C. Willis. 



Southwell Lodge, Norwich, November 18. 



A Curious Flame. 



For some time past I have shown as a lecture experiment a 

 vibratory flame which illustrates in a striking manner phen- 

 omena similar to those to which Mr. Garbutt directs attention 

 in your issue of October 31. 



The flame (which is of the " washed-out" type) is produced 

 by means of a common form of spirit blast-lamp, the construction 

 of which will be sufficiently evident from the accompanying 

 figure. If after using the blast which issues from the nozzle -\, 



the small subsidiary lamp L be removed from the position 

 shown in the sketch and then placed from three to four inches 

 in front of the nozzle, a blue flame cone will in a second or two 

 dart back from the flame of L to N. The lamp L may now be 

 completely removed, and the flame cone will continue — adder 

 tongue like — to dart back and forth between N and a point three 

 to four inches distant for hours together. 



A baffle (in the form, say, of a glass tube or a knitting-needle) 

 held in the track of the vibrating flame at a distance from N 

 less than its normal traverse, will not permit the flame to pass 

 it. But clearly the baffle cannot in this case /*<rr//»(i««;///j' tether 

 the flame cone. It merely curtails the amplitude of vibration 

 without affecting m.aterially its frequency. 



If the safety-valve s be replaced by a cork carrying a U-tube 

 the bend of which contains water, there is a rise and fall of 

 water-level synchronising with the vibratory motion of the flame 

 cone. The apparatus thus becomes a heat-engine producing 

 reciprocating motion in an ideally simple manner. 



If a small compound strip of ferrotype plate and zinc foil be 

 used to baflle the flow of vapour from N, Ihe strip curls up 

 appreciably every time the tongue of flame licks it, uncurling 

 again in the intervals. By including this strip in an incomplete 

 electric bell circuit the bell may be caused to sound in .syn- 

 chronism with the vibrations of the flame. 



Blackheath, November 11. Douglas Carnegie. 



