November 4, 1909] 



NA TURE 



LETTERS TO THE EDITOR. 

 [The Editor does not hold himself responsible for opinions 

 expressed by his correspondents. Neither can he 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 Gallop of the Horse and the Dog. 



In a noto in Nature of October 28 (p. 526) it is stated 

 that Mr. Francis Ram, in a recent book, says I am in 

 error (in an article lately published by me) in regard to 

 the position of the legs and feet in a running dog. 



I have not seen Mr. Ram's book, but I should be glad 

 if you will print the enclosed outline figure of a running 

 dog taken from a series of instantaneous photographs of 

 a running dog by Mr. Edward Muybridge. 



The horizontal line AB gives the actual level of the 

 ground below the dog. The figure is one drawn for a 

 book which I have in preparation, and I think has con- 

 siderable value, since it serves to establish my suggestion 

 that the Mycensans (who were the originators of the pose 

 of the galloping hoise, which was never used by Greeks, 

 Egyptians, Assyrians, Romans, or Europeans, but 

 travelled, as Salomon Reinach has shown, across Tartary 

 to China and Japan, and came from Japan to England at 

 the end of the eighteenth century) did not invent the 

 well-known conventional pose, but observed it in the dog, 

 and very reasonably, but incorrectly, applied it to repre- 

 sentations of the horse and other animals which do not 

 really assume that pose. Tiie po^e in question satisfies the 



A. B. 



artist's judgment even when applied to the horse, because 

 the outstretched position of the hind legs, with upturned 

 hoofs and the forward-reaching position of the fore-le«s, 

 do succeed one another in the galloping horse so rapidly 

 as to cause, not a continuity of the retinal impressions, but 

 a continuity of the more slov/ly formed mental apprecia- 

 tions of the positions of the legs. 



It is an important fact that the late Prof. Marev, of 

 Paris, did not succeed in photographing the dog with all 

 the feet "off" the ground and the legs in the position 

 shown in Muybridge 's photographs, and consequently 

 archa;ologists have supposed tha't the Mycenjeans imagined 

 the pose as an artistic expression of rapid galloping. It 

 seems to me, on the contrary, certain that they constantly 

 saw and admired this pose in their hunting dogs. 



E. Ray Lankester. 



29 Thurloe Place, South Kensington, October 29. 



The Refractivity of Radium Emanation. 



We have read with special interest the communication 

 from Lord Rayleigh in Nature, October 28 (p. 519), on 

 the determination of the refractivity of gases available only 

 in minute quantity, because we ourselves have been work- 

 ing towards the same end at intervals during the last two 

 years. Our object in view was also the same, viz. the 

 determination of the refractive index of radium emana- 

 tion ; not only for the intrinsic interest of a knowledge 

 of the refractivity in question, but also because of the great 

 probability of the emanation being one of the series of 

 non-valent elements, and the determination would there- 

 fore enable us to extend the series of simple integers which 

 has been found by one of us to connect together the refrac- 

 tivities of the other elements in the series. 



The extremely minute quantity of emanation available — 

 not more, after undergoing the ordeal of purification, than 

 about one-tenth of a cubic millimetre measured at atmo- 

 spheric pressure — made it quite clear that the refracto- 

 meter to be employed must be on a minute scale, and the 

 NO. 2088, VOL. 82] 



form which it seemed to us would probably lead to the 

 most accurate results in the circumstances was one on the 

 principle of a Fabry and Perot etalon, partly on account 

 of the sharpness of the bands thus obtainable and partly 

 because it is the double thickness which constitutes the 

 path difference between successive interfering beams, and 

 consequently the gas contained is utilised twice. 



A capillary tube of glass (or fused silica) was sealed at 

 one end, and a transverse hole was drilled passing through 

 the extreme end of the bore. Two parallel faces perpen- 

 dicular to the axis of the hole were then ground on the 

 tube, and parallel plates of glass (or silica), silvered (or 

 platinised) on the inside, were then cemented on the faces 

 with Coate's cement. For this apparatus we had recourse, 

 as usual, to the excellent workmanship of Messrs. Hilger. 

 The result was a tiny interferometer vessel, 2-271 milli- 

 metres long and 071 mm. diameter, into which we could 

 compress the emanation through the capillary tube by 

 means of a mercury column in the usual way. When this 

 interferometer was set up in the path of the green beam 

 separated spectroscopically from the light given by a 

 Bastlan mercury lamp, and the light passing through was 

 examined through an astronomical eye-piece — the lens 

 system throughout being chosen so as to give best 

 illumination — the interference bands which 'were obtained 

 were all that could be desired, it being easily possible — 

 when the silvering was of the best thickness — to measure 

 micrometrically to the hundredth part of a band. 



The method of a determination, in general outline, con- 

 sisted in alternately increasing and decreasing the pressure 

 of the contained gas from and back to a practically zero 

 value and observing the number of interference bands which 

 passed over the cross-wire of the micrometer. In order to 

 determine the efficiency of the arrangement, observations 

 weie made for the refractivity of air, with the result that 

 we think we are justified in claiming that an accuracy 

 to within about 2 per cent, could be relied upon, so far 

 as the optical part of the experiments is concerned. 



The real difficulties begin, however, when we deal with 

 the emanation itself. The rapid generation of impurities, 

 originating in part in the action of the emanation upon 

 the resinous cement employed for fixing the parallel plates, 

 together with the lack of a knowledge of what these impuri- 

 ties are, m.ade it impossible to calculate the index of the 

 emanation from the experimental results, although it was 

 perfectly easy to measure the refractivity of the mixture 

 of gases existing at any time. The only datum known in 

 regard to the composition of the mixture was the approxi- 

 mate percentage of emanation present, this being found 

 by measuring the y radiation from it. The direction in 

 which the refractivity lies may, however, be inferred with 

 probability from the following observations. Starting with 

 emanation given off from a solution and containing a very 

 large amount of impurity, this was purified, first, by 

 explosion, drying, and absorption of CO., and afterwards 

 by freezing in liquid air and pumping off the volatile 

 impurities according to well-known methods. Testing the 

 refractivity from time to time, its value — at first of the 

 order of that of air — did not sensibly rise until the volume 

 was about one cubic millimetre. Continued purification 

 increased the refractivity, and the highest values obtained 

 in our experiments were 0-000840 when the volume was 

 0-205 mm.^ (at atmospheric pressure) and 0000916 when 

 the volume was 0-128 mm.' measured at atmospheric 

 pressure. The quantity of emanation was approximately 

 the same for both these measurements, and equalled 

 the quantity in equilibrium with 0-178 gram of radium. 

 Of course, if we could assume that the impurities were 

 the same in kind on the two occasions it would be possible 

 to estimate from these data the value for the pure sub- 

 stance, but the failure of this method on many occasions 

 to give consistent results took away all belief in its 

 applicability. For purposes of comparison we may state 

 that the higher of the above values is about twenty-six 

 times the value for helium, while the value for xenon — 

 the highest for any known gaseous element — is twenty 

 times, and for CO, thirteen times, the value of helium. 



One source of difficulty so long as the available amount 

 of emanation is so small is that the maximum pressure 

 to which it can be raised in the apparatus is onlv a fevir 

 (7 or 8) centimetres. The capillary correction thus becomes 



