August 12, 1909] 



NATURE 



189 



cus5cd here. Mr. Johnston will find this subject admirably 

 treated, from a theologian's point of view, in ]Mr. 

 Waggett's contribution (chapter xxiv.) to " Darwin and 

 Jlodern Science." 



The question of the antiquity of land-plants is of great 

 interest. Assuming, for the sake of argument, that the 

 highly organised lycopods and fern-like plants described by 

 Prof. Potonie as Upper Silurian (" Die Silur und die Culm- 

 flora des Harzes Geb.," Abhandl. d. k. Prcuss. Geolog. 

 Laiiilcsanstalt, Heft 3b, igoi) were really of that age, it 

 would appear certain that land-plants must then have 

 already passed through a very long course of evolution. 

 No one is likely, in these days, to suppose that Bothroden- 

 dron and Sphenopteridium were specially created. The 

 doctrine of mutation (of which Mr. Johnston appears to 

 be an adherent) does not materially help in hurrying up 

 the process of evolution, for, as Prof, de Vries himself 

 says, " Mutations do not necessarily produce greater 

 changes than fluctuations " (" Darwin and Jlodern 

 Science," p. 73). 



The problem confronts us, and its solution must simply 

 await further evidence. D. H. Scott. 



Oakley, Hants. 



The Perseids of 1909. 



The weather, fortunately, took a very favourable turn 

 on .August 3, and observations have been obtained nearly 

 every night since, though moonlight has rather seriously 

 interfered with the work and obliterated many small 

 meteors. Up to August 9, and including that date, the 

 display of Perseids was decidedly poor, and quite dis- 

 appointed expectation. On August 7 and 9, particularly, 

 there appeared to be very few meteors, but the moon was 

 shining rather strongly in the east. 



On .August S there were several brilliant Perseids 

 observed. At 10 p.m. a fine, long-pathed meteor shot from 

 Camelopardalus across ;.- Draconis towards ( Ursas 

 Majoris. At 10.8 p.m. another Perseid, equal in bright- 

 ness to Jupiter, passed from x to between /3 and -n Draconis. 

 .\t 10.34 P^m. a third shot exactly from a Cygni to close 

 to Cygni. These meteors left streaks, and, in com- 

 bination with a few other paths recorded on the same 

 night, indicated a radiant at 4i°-i-57°. 



The shower will probably arrive at its most abundant 

 phase on August 11 or 12 this year, so that the meteors 

 already reported are merely the vanguard of the approach- 

 ing main group. W. F. Denning. 



Bristol, August 10. 



The Ringing of House-bells without Apparent Cause. 



Kindly allow me space for a few remarks upon Sir 

 Oliver Lodge's theory, put forth in Nature of July 22 

 (p. oS), to the effect that " the bells get charged with 

 electricity (atmospheric), and are attracted to a neighbour- 

 ing wall or pipe, and then released suddenly by a spark." 

 Now, while it is conceivable that a bell might be rung 

 under certain conditions in this manner, during the pro- 

 gress of a thunderstorm or display of sheet-lightning, 

 and granting that ordinary non-electric bells have been 

 rung and wires fused when a house has been struck by 

 the electric current during such storms, still, this theory 

 is inadequate to explain those cases of mysterious bell- 

 ringing on record, and for one reason, among others, that 

 these ringings, often violent and prolonged, have been 

 extended over a term of several weeks or months, and 

 have constantly taken place when no storms or strong 

 electrical conditions were apparent, and when every effort 

 was being made to ascertain the cause. 



f speak from personal experience of a case which 

 occurred in my house when resident in the south. For a 

 period of two months there were constant ringings — often 

 violent, the bell lashing to and fro — of the indoor bells, 

 without apparent cause. In the case of one bell the wires 

 were cut, but still it rang. The utmost endeavours were 

 made to solve the mystery, but it defied all our efforts. 

 There were no rats, the house having been made rat-proof, 

 nor did we see one rodent during our stay. The wires 

 were carefully traced and examined. Pendulums were 

 affixed to all the bells to detect slight motion, and they 

 were strongly illuminated hv a powerful light and a watch 



NO. 2076, VOL. 81] 



kept, sometimes all through the night. The chief offender 

 among the bells was one communicating with a private 

 room. The wire from this ran, high up near the ceiling, 

 upon the varnished paper, except where it passed through 

 a wall, which it did through a half-inch pipe. It was 

 impossible for a rat or mouse to touch it all along its 

 course. This bell rang repeatedly from early morn to 

 late at night. The room was thoroughly searched and 

 secured — the shutters put up and barred and the door 

 locked. Still the bell rang, and defied all our efforts to 

 elucidate the mystery. On one occasion, when the whole 

 household was together in another room, some little 

 distance away, one of them said, " I wonder if it will 

 ring to-night? " The words were scarcely spoken before 

 the bell rang out, first faintly, then so violently that the 

 bell lashed from side to side. All ran out and saw it 

 swaying. I can state that during the whole period we 

 had no thunderstorm, it being winter, and the ringings 

 were so frequent that it would have needed scores of 

 storms and abnormal electrical conditions to produce them, 

 even if these had been the cause. This theory is ingenious, 

 but one doubts whether Sir Oliver advances it seriously. 

 Whatever is the cause of these mysterious ringings, it is 

 patent to anyone having had experience of them, or know- 

 ing the cases on record, that it is not electricity, atmo- 

 spheric or other. C. L. Tweedale. 

 Weston Vicarage, Otley, Yorks, .August 6. 



Variation in Relative Intensity of Helium Lines 



It has long been an open question whether the spectra 

 of gases were subject to any general law of intensity 

 variation such as the Wien-Paschen displacement law for 

 incandescent solids. Reasoning from Kirchhoff's law, it 

 is frequently assumed that the Wien-Paschen function 

 must be an envelope to the radiation spectrum of any 

 gas, at least for certain selected lines. Pfliiger's work 

 with the mercury lamp [Ann. Ph., July, 1908) indicated 

 some such variation in relative intensity, but the energy 

 (temperature) effect was obscured by the effect of vary- 

 ing vapour density. The latter is known to be large, so 

 that the existence of the former was not proven. 



We have recently taken up this question, using a pair 

 of helium lamps containing very pure gas at 5 mm. pres- 

 sure, and found no variation whatever when the intensity 

 was varied by a factor of eighteen. The method was to 

 hold one of this pair of lamps constant at a moderate 

 current (5 milliamperes per mm.^), while the intensity of 

 the other was varied from 015 to 270 times this value. 

 Spectrophotometric settings were made on the brightest 

 red line (X 668) and the bright blue line X 447. The un- 

 certainty in the ratios of high/low intensities was less 

 than 2 per cent, in both red and blue ; the observed differ- 

 ence in these ratios between red and blue was less than 

 I per cent. The red and blue lines then increase in the 

 same proportion over a wide range of intensity. 



.According to the Wien-Paschen function, an incandescent 

 body should, for an increase in total radiation of 18 fold, 

 increase in temperature 2-06 fold. If this increase is 

 from 1500° to 3090°, then \ 668 would increase in that 

 range 1721 fold, and A. 447 would increase 66,850 fold in 

 intensity, so that the blue increases 38-8 times as much 

 as the red, whereas for helium we found no difference 

 so great as i per cent. P. G. Nutting. 



Orin Tugman. 



Department of Commerce and Labour, Bureau 

 of Standards, Washington. 



Kohlrausch's "Physical Measurements" 



I SHOULD like, through the medium of Nature, to 

 direct attention to an erratum in Kohlrausch's " Physical 

 Measurements " (third English edition, Churchill, 1894), 

 p. 434, table 8. The value for k{s = 2o) is given as 0-458, 

 its true value being 0-457(1). Other editions, English or 

 German, not being at hand, I do not know whether the 

 mistake is repeated or not. As this occurs in a very 

 useful table in a universally used reference work, it 

 should be of interest to a number of your readers. 



E. W. Nelson. 



The Laboratory, Citadel Hill, Plymouth, 

 August 2. 



