;64 



NATURE 



[StPTEMBEP. 2 2, I9IO 



from near Wellington College. 1 have seen brighter 

 meteors, but never one that remained so long in sight, and 

 its course was marked for a long way by a streak of light, 

 showing very clearly the route it had taken. When 1 first 

 caught sight of it, it appeared to be rising in the sky, 

 through the Camelopard, and it passed almost exactly over 

 ^ and 7 of the Little Bear, over ri Draconis, between tt and 

 p Herculis, and over a Ophiuchi, vanishing perhaps 

 15 degrees further south in the Serpent. Rising and fall- 

 ing in its flight like a thrown cricket-ball, it seemed to be 

 quite close at hand. 



Edmund J. Wkbb. 

 Burghclere, Newbury, September g. 



Mr. Edmund J. Webb's highly interesting account of the 

 fireball of September 2, in conjunction with other descrip- 

 tions which have now come to hand, enable the real path 

 to be well determined. 



The radiant point of the meteor was near /3 .\urig;E, or 

 at about 87° + 41°, and the height of the object from about 

 98 to 44 miles from over the North Sea to S.S.W. coast 

 of England. The meteor had an unusually long flight 

 right across the country from N.N.E. to S.S.W., and its 

 visible course of 352 miles was probably traversed at a 

 velocity of 40 miles per second. It is only rarely that a 

 fireball is seen in this country with such an extended 

 trajectory. Most of the observers only saw a part of the 

 path. The radiant was near the horizon in N.N.E. 



The fireball was seen bv the Rev. F. C. Lees, Sutton, 

 Surrey ; Rev. C. L. Tweedale, Otley, Yorks ; Col. E. K. 

 Markwick, Boscombe, Hants ; and many other observers. 



W. F. DliNNING. 



The Law of Definite Proportions. 



Perhaps a reader of Nature will be good enough to 

 solve the following question. 



If an amount of heat is supplied to a volume of ice, 

 water, and vapour at the triple point, and yenxaining al 

 the triple point, and the same volume, while the heat is 

 being supplied, are water and vapour formed in definiti- 

 relative proportions from the ice? That is, is the ratio of 

 vapour to water independent of the amount of heat sup- 

 plied, or of the original proportions of the three phases? 



C. E. 



King Edward VIL School, Sheffield. 



FIRE TESTS WITH TEXTILES.' 

 'T'HE frequent accidents caused by tlie ignition of 

 ■*■ liighly inflammable wearing apparel have 

 directed wide attention to the possibility, by suitable 

 treatment, of rendering materials like flannelette non- 

 inflammable. The interest aroused by the subject is 

 further increased by the fact that most of the fatal 

 accidents occur to very young children, and apparentlv 

 the number of such accidents is not diminishing. 



Thanks must be given to the British Fire Prevention 

 Committee for the efforts being made to investigate 

 the subject in a thoroughly scientific manner, and for 

 the report before us, which contains the results of 

 experiments on 456 samples of cloth. These were 

 divided into the five following- groups : — (a) Flannel- 

 ette ("non-flam," commercial); (b) flannelette ("non- 

 flam," special); (i) flannelette (ordinary); (d) "union" 

 (a mixture of cotton and wool); (c) flannel; (/) flan- 

 nelette (fine finish). The method of testing employed 

 was briefly as follows : — A yard of the cloth was sus- 

 pended from three hooks fixed in a beam, the lower 

 edge was kindled by the flame of a wax taper or 

 spirit lamp. .At the end of sixty seconds any flame 

 was extinguished, and the portion of material burned 

 carefully measured. In many cases photographs were 



' Fire Tests with Textiles. Flannelette known as " Non-Flam " FLin. 

 relette, Ordinary Flannelette, '"Union" Fl.inneletle suhmitted for test by 

 Messrs. Whipp Bros, and Tod, Ltd., Manchester. The Committee's Report, 

 pp. 48 (" Red Books " nf the British Fire Prevention Committee, No. t48.) 

 (London: The British Fire Prevention Committee, igio.) Price 51. 



taken before and after the ignition, and these supply 

 more vivid illustrations of the results than the pages 

 of statistics which follow. In some cases made-up. 

 garments were suspended on wire frames and tested 

 as before. The different samples were also tested be- 

 fore and after repeated washings. Manifestly this is 

 a point of great import.'ince, and it was proved that 

 in the case of " non-flam " materials there was practic- 

 ally no difference as regards fire resistance between 

 samples washed once and- those washed twenty times. 



The general nature of the results may be briefly 

 stated. Unquestionably the flannelette known as 

 "non-flam" justifies its name. Samples of this 

 material are only charred where they have been in 

 contact with the flame ; they are non-inflammable. 



Ordinary flannelette as received from the manu- 

 facturer burned up through the centre of the sample. 



NO. 2134, VOL. 84] 



Demonstration Tests with Garments: in) Flannelette (" Non Flam 



Commercial) at 120 seconds. {,/•) Flannelette (Ordinary) at 60 



seconds. — From "Fire Tests with Textiles." 



and from 25 to 40 per cent, of the material was con- 

 sumed, while after one washing from gz to 100 per 

 cent. v\-as destroyed. The material known as 

 "union," a mixture of cotton and wool, as might be 

 expected, is less inflammable than flannelette; from 

 57 to 66 per cent, of the material, after one washing, 

 was burnt. 



In the case of flannel the charring only reached as 

 far as the power of the flame extended. ' Lastly, the 

 flannelette (fine surface) is shown to be very like the 

 ordinary, and in many cases the sample was com- 

 pletely consumed. This investigation appears to have 

 been carefully conducted, and the report should be 

 widely circulated. The illustrations "explain them- 

 selves : (a) a "non-flam," made-up garment, after 

 120 seconds; (b) a made-up garment, ordinary flan- 

 nelette, after 60 seconds. 



