April 1 8, 1872] 



NATURE 



On the Colour of a Hydrogen Flame 



Accepting, for the time being, the experiments of Mr. 

 Barrett as sufficient proof that a pure hydrogen flame does not 

 exhibit a blue colour, my "elaborate theory" must, I suppose, 

 seek refuge under the actinic power of the electric light. 



Mr. Murphy refers this actinism to the fact "that the electric 

 light is /'A/iv-ihan solar light," the blue rays of the sun's liyht 

 having been abstracted by absorption. This is a bare fact, and 

 deals solely with the relative prO|-iortions of the different coloured 

 rays which reach us from the two sources — it conveys no clue to 

 the reas' n u<Iiy the blue rays have an entity in the first instance. 



I would not have it understood that I consider all the liigh 

 refrangible rays to be due to secondary waves ; but I think it 

 possible that some, at least, of those emitted from sources of a 

 very higli temperature may owe tlieir existence to this cause. 

 Considering for the moment the eleciric light, we have a centre 

 of the most intense commotion sending off waves in all direc- 

 tions — a condition necessary, and at the same time eminently 

 favourable, for tlie production of secondary waves. 



With respect to Mr. Barrett's experiments, I intend to repeat 

 them as soon as I can command the time. The absence of the 

 higher refrang ble rays in a hydrogen flame does not, however, 

 affect the mechanical possibility of the fxistence of secondary 

 waves ; although it would be reasonable to expect their presence 

 in a pure oxy-hydrogen flame, the amplitude of the disturbed 

 particles being necessarily very great. A. G. Meeze 



Hartley Institution, Southampton, April 15. 



Another Aurora 



A MAGNIFICENT aurora, scarcely inferior to that of February 

 4, was observed here on the evening of the loth inst., between 

 8'' 30'" and gli 30". 



The display was at its greatest beauty about gi" o™, when the 

 crfamy-white streamers attained an altitude of at least 60° above 

 the N. horizon, and formed a fine contrast with a pale rose-pink 

 background. The streamers appeared to proceed from behind a 

 dense mass of stra'us cloud which, although a moderate breeze 

 was blowing from the S.W., remained almost stationary and un- 

 altered during the display. The N. horizon was hghted up with 

 a glow as intense as the early twilight on an evening In June. 



With a small direct-vision spectroscope by Browning, I could 

 see the line in the green near F, but no others. It was remark- 

 ably bright and sharply defined. 



Bedford, April 12 Thos. Gwyn E. Elger 



Brilliant Meteor 



Yesterday afternoon, whilst standing on the lawn of the 

 Observatory with my hack to the sun, which was brightly 

 shining, I saw a splendid meteor fall in the south-east. The 

 sky at the time was of an intense blue and cloudless, with the 

 exception of a few cirri in the north and north-west, and the 

 meteor as seen against it presented the appearance of polished 

 silver. The flight of the meteor was almost vertical at an altitude 

 of about 30°, its extent was about 10°, and the trail which seemed 

 to hang in the air and fade away like the trail of a rocket, was 

 at the instant of explosion probably 3° in length. There was no 

 report accompanying its disruption, or it would certainly have 

 been heard, the neighbourhood being very still at the time. 



Immediately on its disappearance I looked at my watch, it 

 was 4I' 36s r.M. G.M.T. 



Had the fall occurred after dark I have no doubt but that the 

 meteor would have exhibited a magnificent spectacle, for its 

 brilliancy far exceeded that of the moon as seen by daylight. 



During the aurora on the evening of the loth I observed at 

 9. 16 r.M. a peculiar well-defined patch or short band of bright 

 red light, the position of which, as seen from here, was N.N.E. 

 altitude 40° lo 45°. Perhaps other observers may have noticed 

 it, and their observations will give data which may serve to 

 assist in determining the true height of the auroral discharge. 



The magnetic disturbance on the ich commenced abruptly at 

 2 r.M., and was greatest during the hours of daylight, so it is 

 extremely probable, the sky being but partially clouded, that if 

 the aurora was visible before night, some observers may have 

 seen it. I cannot say I have ever seen it myself in the daytime, 

 although I have repeatedly seen cirrus clouds assuming a form 

 very similar to auroral streamers. However, on looking at the 



magnets and finding them undisturbed at the time, I have con- 

 cluded that no aurora was taking place. 



Kew Observatory, April 13 G. Mathu.s WhipI'LE 



Tide Gauges 



The subject of the tides is now one in which much interest is 

 taken by the committee of the British Association, and it would 

 be a great boon to many who are in a position to give attention 

 to it, if some of your readers would su 'ply a description of a 

 self registering gauge for recording the heights, which should do 

 its work effectively and not very expensively. Many plans are 

 suggested ; the difficulty is to know which is the best. 



Vicarage, Fleetwood, April ii James Pearson 



NOTES ON THE RAINFALL OF iZji 



THE following are a few particulars of the rainfall of 

 the past year, deduced from daily observations with 

 Glaisher's (Hall's improved) rain gauge* at Fulwell,t 

 near Twickenham, Middlesex, the place of observation 

 being in lat. 51° 26' o" N. long, o'' 20' 53" W. 



The orifice, or receiving surface of the gauge, which is 

 placed horizontally, is 800 inches in diameter (;o'26 in 

 area), the height of the same above the ground being one 

 foot, and, as determined by spirit leveUing from Ordnance 

 E.M., 47 feet above mean sea-level. 



The results of the observations have been calculated in 

 the imperial system, and metric equivalents are placed in 

 brackets, the use of which (biackets), for the sake of 

 distinction, has been avoided in all other formulas ; they 

 have, in each instance, been calculated to two or three 

 places of decimals, but are here given, so far as is practi- 

 cable, in whole numbers ; the nearest integer, in each 

 instance, having been taken ; t/iey have further been 

 cahulaied upon tlie hypothesis tJtat the rainfall was equally 

 distributed. 



In the following table : — 



a = depth of rainfall in inches \ Total fall 

 (3 = depth in centimetres j per month. 



y = number of gallons i Equivalents 



S = number of hectolitres \ per acre. 



The total depth during the year was 22-42 in., or 56-947 

 centimetres. 



The rainf all on a square mile during the year was 

 22,500 X 640 X 22-42 = 322,848,000 gallons (-=-22-024 = 

 14,658,918 hectolitres), or 640 X 4S40 X 9 X 22-42-^-12 

 = 52,oS6,i44 cubic feet (-=- 35-31658 = 1,474,835 cubic 

 metres). 



A cubic inch of distilled water at a temperature of 

 62° Fahr. (i6-66 C.)isa standard of weight ; this quantity 

 has been determined to -weigh 252-458 grains, of which 

 437-5 make one ounce Av. ;% therefore, a cubic foot weighs 



* Vide Scientific Opinion, -Vol. iii., pp 



t Although the observations reftr e: 



probably be scarcely the less interesting. 



J Practical Meteorology, hy }o\mt>i^Vi, Ph.D., sec. 



420, 44Q (May 18, 1870). 

 pecially to this locality, they will 



