June 3, 1920] 



NATURE 



42 



Letters to the Editor. 



[The Editor does not hold himself responsible for 

 opinions expressed by his correspondents. Neither 

 can be undertake to return, or to correspond with 

 the writers of, rejected manuscripts intended for 

 this or any other part of Nature. No notice is 

 taken of anonymous communications.] 



The Flight of Flying-fish. 



It is disputed whether the flight of flying-fish is a 

 genuine flight or simply a leap and a glide. The 

 question is referred to in the section devoted to flying- 

 fish in the Natural History Museum, South Kensing- 

 ton. Recently I have had ample opportunity to study 

 these fish in the tropical waters of the Atlantic and 

 Pacific Oceans. 



The observations which I have made and the con- 

 clusions at which 1 have arrived are corroborated bv 

 the officers of the R.M.S. Victoria. Many of these 

 gentlemen were surprised to hear that there was any 

 doubt on the matter. That the flight is a genuine 

 one is proved by the following facts : — 



(i) During flight these fish are able to turn at right 

 angles, and even at a very acute angle. More than 

 once I have seen a fish turn with great rapidity at 

 an acute angle and come back in a direction opposite 

 to the direction in which it set out. A mere glide will 

 not enable any animal to do this. 



(2) Standing at the bow of the ship directly above 

 some flying-fish which were in a hurry to get out of 

 the way, I saw the wings flap as distinctly as the 

 wings of any frightened bird. 



(3) Some of these fish fly for a distance of from 

 150 to 200 yards without rising more than a 

 couple of feet above the surface. They rise over the 

 crests of the waves and sink into the hollows. They 

 could not do this by a mere leap and a glide. 



(4) Besides flying low over the surface of the 

 waters, they are also able to rise to a considerable 

 height, and not infrequently come on board large 

 steamers. When they fall on deck their wings can be 

 seen, as well as heard, flapping. It is true that they 

 are unable to rise from the deck, but the same is true 

 of many sea-birds. 



(1;) When in full flight the outlines of the wings are 

 indistinct and blurred in contrast with the clear out- 

 line of the bodv. This can only be due to the very 

 rapid movement, as in the case of hovering flies and 

 humming-birds. J. McNamara. 



3 Holland Road, Kensington, W. 



An Experiment on the Spectrum. 



In school and college courses little experimental 

 work is done on the infra-red and ultra-violet 

 parts of the spectrum. The student is, of course, 

 told about these regions, and how they can be de- 

 tected respectively by the heating and actinic pro- 

 perties of their rays. But he is not allowed to in- 

 vestigate these rays himself, nor are their properties 

 demonstrated before him. This is on account of the 

 elaborate nature of the apparatus necessary ; the infra- 

 red region requires a thermopile or bolometer, to- 

 gether with an expensive galvanometer, and the ultra- 

 violet requires photographic methods and a spectro- 

 graph. So much apparatus cannot be afforded for one 

 experiment, and besides is apt to distract the student's 

 attention from the simple nature of the facts involved. 



If, however, a very intense spectrum is used, the 

 infra-red can be mapped roughly with an ordinary 

 thermometer, and the ultra-violet with a photographic 

 exposure-meter. Neither galvanometer nor spectro- 

 graph is necessary. The thermometer I have used is 

 NO. 2640, VOL. 105] 



a Fahrenheit one, range 0° to 220°, the bulb of which 

 is blackened by dipping it into lamp-black shaken up 

 with methylated spirits; the bulb is 5 mm. in dia- 

 meter. The exposure-meter is the Imperial exposure- 

 meter for dull light and interiors, which costs is. 6d. 

 together with a refill. In this instrument a piece of 

 sensitive paper is exposed to the light, and the time 

 noted that it takes to darken to a standard tint. 

 J The sensitive paper supplied darkens two or three 

 times as fast as ordinary P.O.P. As source of light 

 I have used a little 5-ampere arc lamp, which is run 

 off the lighting circuit through a rheostat. T^he anode 

 is horizontal and the cathode vertical ; they are both 

 enclosed in a glass cylinder which restricts the supply 

 of oxygen, arj^d so lengthens the life of the carbons. 

 Lamps of this pattern burn very steadily, and have 

 come into wide use during the past ten years. It is 

 because so many laboratories have these lamps that 

 I describe this experiment here. .An arc of this pat- 

 tern is absolutely necessary; a pointolite or half-watt 

 lamp is of no use for the purpose. 



As lens I have used a spectacle lens of 25 cm. focal 

 length, and as prism a single equilateral dense flint 

 ij in. high. The spectrum and arc are equidistant 

 from the lens. As slit I have used the crater of the 

 arc, which measures about 3 mm. in diameter, since the 

 carbons in this t\pe of lamp are onlv 5 mm. thick. If 

 the rays of light from the anode fell squarely on the 

 lens we should have a point image of a point source, 

 and • the spectrum would be only 3 mm. high, but 

 by setting the lens obliquely, rotating it through 

 30°, an astigmatic line image is formed, and we get 

 a reasonablv sharp spectrum 12 mm. high. Strav 

 light is excluded by enclosing the arc in a box. 



The following table gives a set of results taken in 

 somewhat less than an hour : 



Scale 

 cm. 

 70 



II 



8-5 

 90 



9-5 

 lo-o 

 105 

 no 



"•5 



I2-0 

 125 

 13-0 



The ends of the visible spectrum were at 9-2 and 

 12 cm. The first column gives the readings on a 

 centimetre scale placed along the spectrum, the second 

 column the name of the colour, the third the rise of 

 temf>erature experienced by the thermometer in three 

 minutes, and the fourth the reciprocal of the time 

 in minutes taken by the paper to darken to the 

 standard tint. In the case of the last two readings 

 the exposure-meter was illuminated by stray light. 

 It is possible to go further into the ultra-violet if a 

 crown glass prism is used. The infra-red measured 

 goes to 2-2 /u or thereabouts. If a piece of P.O.P. 

 is exposed to the spectrum for a couple of minutes, 

 it shows bands — one from 10-11 cm., another from 

 II-3-II-7 cm., and a third from ii-8^i2-7 cm., the 

 positions dll being measured on the centimetre scale 

 above referred to. 



It is interesting to remember that when Sir William 

 Herschel discovered the heat spectrum in 1800 he 

 iised thermometers. The source of light was the 

 sun, and the arrangement was similar to N«wton|s 

 original one — the prism was placed close up to a slit 

 at a window, no lenses^ were used, and -the spectrum 



