October, 1913. 



KNOWLEDGE. 



379 



such aggregates of moving individuals, e.g., the spiral 

 nebulae, will appear perhaps like a stationary or 

 wavy streak of cloud maintaining a fictitious appear- 

 ance of rest during long periods ; and yet they are in 

 all probability the seat of turbulent centrifugal 

 action the ends of which it will take aeons to 

 discern. 



Professor See, of Mare Island, California, in his 

 recent monumental work on the nebular theory, 

 maintains that he has adduced evidence that the 

 older forms of these mysterious germs of worlds 

 approximate to the Archimedean form of spiral, the 

 later tending to the logarithmic. These curves can 

 express respectively arithmetical and geometrical pro- 

 gressions such as Perrin found ; and it is noteworthy 

 that upon any such theory of gravitation, as a 

 residual mode of energy, as that of Lorentz a resist- 

 ance varying in the inverse square must be added. 

 Professor See in a recent letter to the writer 

 suggests that sufficient attention has not yet been 

 given to such attempted correlation ; such was also 

 the opinion of Poincare. It appears not improbable 

 that free electrons passing through atoms, or some 

 stars, such as 1830 e Groombridge, through globular 

 clusters, would do so, not upon long ellipses, but 

 rather on elliptiform helices, their orbital motion 

 being gradually damped while passing through the 

 centre of the cluster. Yet in no science must we be 

 more vigilant against spurious appearances of 

 irregularity or regularity than in optics. In that 

 science a speculative element is positively a saving 

 grace ; as Faraday expressed it : " That man only 

 is condemned who cherishes fixity of opinion." Do 

 not the canals of Mars teach us a like lesson, that these 

 questions of law or ultimate expressions of fact are 

 still open to faith in a beneficent and volitional 

 purpose ? 



In pursuance of such ideas Mr. Rankin and the 

 writer have recently (Nature, June 19th and August 

 21st, 1913) been able to show that X-rays are 

 " diffracted," or at least characterised by contact or 

 close proximity to metals as well as "crystals," and 

 even by substances such as plate glass (see Figure 

 411), which must be regarded as either structureless 

 or at least in a transitional stage, seeing that it is a 

 super-cooled liquid. Some of these effects have also 

 been obtained through cast, as well as wrought, iron 

 plates (up to one centimetre in thickness), used as 

 obstacles without apertures, showing a geometrical 

 pattern on special rapid plates which can scarcely 

 in all cases be attributed to the mechanical rolling 

 of the metals. It seems not improbable (see Figure 

 410) that the primary beam of X-rays has under- 

 gone some displacement or shift, as though it were 

 subject to some repulsive agency which begins to 

 act perceptibly at short distances, effects which are 

 visible only to the photographic eyes of science. In 

 this experiment a thin lamina of mica was used to 

 cover the aperture in a lead screen, and was placed 

 at an angle of 85° to the rays. The curious bands 

 upon the reflected portion of the beam are still, we 



believe, unexplained. A series of upwards of two 

 hundred experiments has been made with the 

 object of throwing light upon some of the obscure 

 questions raised by the recent reflection of X-rays. 

 A short resume only of the results can be given in 

 the present article. 



Halos, similar to those of Figures 406 and 407, 

 were obtained both with and without mica, or other 

 crystal used to cover the apertures, which were 

 generally about a quarter of an inch in diameter, in 

 metal screens, lead, iron, brass, and others. One 

 hypothesis of these halos attributes them to atmo- 

 spheric secondary radiation. They are probably not 

 halation effects, though if this were the case it would 

 be tantamount to a reflection of X-rays from a glass, 

 i.e., a non-crystalline surface, the back of the 

 " special rapid " plate. The term " halo " is care- 

 lessly used by photographers, and should we 

 think be confined to appearances such as Figure 

 406, where the feature is independent of the image 

 of the direct rays ; in Figure 407 the white band 

 alone secures the effect from being possibly due to 

 secondary radiations, or some cause not implicating 

 the primary rays. In experiments (see Figure 409) 

 to study the disposition of the reflected spots from 

 mica placed normally to the aperture in an iron 

 screen it was found in several cases that these 

 markings occur at increasing distances from the 

 centre, indicating a spiral sequence ; the exterior 

 spots not well seen in the reproduction, appear 

 elongated or drawn out into bands which are not 

 concentric. Figure 408 shows one arm of a bright 

 cross which had appeared on one negative where no 

 mica or crystal had been interposed in the aperture 

 of an iron screen. This cross covered the whole 

 plate, i.e., extended beyond, and was apparently 

 independent of, the image of the aperture, and 

 suggested a possibility that the primary rays might 

 possess some structure or be polarised or quenched 

 in special directions by the use of certain crystals. 

 But in further experiments black bands parallel and 

 normal to the arms of a cross were obtained even 

 with the direct or uninterrupted rays received upon 

 a plate within a wooden box containing an intensify- 

 ing screen. This raised the question whether these 

 bands might not be purely photographic effects due 

 to some development errors or the mechanical pro- 

 cess of plate manufacture, though plates of various 

 makers had been used and all the non-essentials of 

 the experiment repeatedly changed. A crucial trial 

 was therefore devised to determine how far photo- 

 graphic errors could effect the general results, and 

 this showed that the possibility of vibrations at right 

 angles to the plate in process of manufacture causing 

 inequalities of disposition of the sensitive material 

 in process of drying, must be taken into consideration 

 This particular effect could not be due to a primary 

 structure of the X-ray beam, i.e., a structure ante- 

 cedent to all obstacles, such as metallic crystals, 

 whether in the wires, anti-cathodes, or glass of bulb 

 or plates. The experiment was arranged as follows. 

 Four recording plates, two made by the well-known 



