June 23, 1921] 



NATURE 



527 



higher velocity. It will also be observed that 

 ammonium nitrate itself under a powerful initial 

 impulse gives rise to a notable pressure, so that 

 that ingredient is not to be looked on as a diluent 

 of the T.N.T. , but as an explosive substance, as 

 well as a purveyor of the oxygen in which T.N.T. 

 is deficient. 



Smoke. — For the purpose of correct ranging 

 and locating the position of burst, an explosive 

 developing smoke is desirable. Amatol 80/20, 

 when used alone, had the disadvantage that it 

 gave no smoke, as the products of the detonation 

 are colourless gases, thus : — 



2C7H3N306 + 2iNH4N03 = 24N2 + 47H20 + i4C02 ; 



whereas, when picric acid or trinitrotoluene de- 

 tonates, a large quantity of unconsumed carbon is 

 set free, affording a black cloud useful for the 

 purpose of observation. 



Mixtures capable of producing a white smoke, 

 useful for aerial observation, were then added, and 

 as a result of investigations as to the best method 

 of securing its dissociation, ammonium chloride 

 in conjunction with the ingredients of amatol was 

 localised at the base of the filling. 



Needless to say, there were many other develop- 

 ments in explosives practice during the war, but 

 the example of the train of detonation leading up 

 to the complete detonation of a high explosive 



shell was chosen to exemplify the subject of this 

 discourse, since it included many features and 

 new problems which had an intimate connection 

 with the technical development of the subject. 



To secure the high percentage of detonations that 

 our artillerists obtained with the freedom from 

 prematures which they always demanded, it was 

 necessary to have each part of the somewhat com- 

 plicated train as nearly perfect as possible 

 not only in design, in order to withstand 

 the effects of rough usage and of set-back 

 in the gun, but also in workmanship, both 

 mechanical and chemical as to purity of 

 materials. This was achieved by the co-ordination 

 of a large number of industries organised on a 

 scientific basis, and these were becoming every 

 day more and more efficient. War is now so 

 highly organised that for its successful prosecu- 

 tion all the technical industry of the country is 

 brought under requisition, and to succeed requires 

 a higher development in research, applied 

 methods, and industrial progress than belongs to 

 the enemy. 



The effort made by this country in the time of 

 stress to overcome deficiencies in these respects 

 was successful as a great technical achievement, 

 and should be an encouragement to us to look for- 

 ward to an equal development of our scientific 

 industries under the stress of a competitive peace. 



Stellar Parallax. ^ 

 By Sir Frank Dyson, F.R.S. 



T N the past ten years a number of the large 

 •*■ telescopes of the world have been applied to 

 the determination of stellar parallax. The prin- 

 ciple of the method is well known and is ex- 

 tremely simple, merely consisting in the detection 

 of the small annual movement of a near star with 

 reference to more distant stars caused by the 

 different position occupied by the observer in con- 

 sequence of the earth's annual revolution round 

 the sun. The whole difficulty consists in the ex- 

 treme minuteness of the angle to be measured. 

 If two railway lines, starting at King's Cross, 

 instead of remaining parallel, met at Newcastle 

 the angle between them would be of the order of 

 the angle to be measured in finding the distances 

 of the nearest stars. To form an idea of what is 

 now being done by large telescopes using photo- 

 graphic methods, imagine two plumb-lines 5 ft. 

 apart. They are sensibly parallel, but actually 

 meet at the centre of the earth, and the angle 

 between them is 005". An angle of this size is 

 measured with an accuracy of ±001'^. Results of 

 this high value were first obtained by Prof. 

 Schlesinger at the Yerkes Observatory. At the 

 present time the observatories of Allegheny, 

 ■Greenwich, McCormick, Mount Wilson, Yerkes, 

 and a number of others are engaged on a 



4 From a discourse delivered at the Royal Institution on Friday, April ^g, 



NO. 2695, VOL. TO7] 



comprehensive programme. At Greenwich we 

 determine the parallaxes of fifty stars a 

 year ; at some of the American observatories 

 many more. 



Necessarily, a good deal of care is required 

 both in taking the photographs and in measuring 

 them. The image of a star may have a diameter 

 of 2" or 3", and the position of its centre should 

 be measurable to between i/50th and i/iooth of 

 this amount. The methods of measurement 

 present some points of interest which need not be 

 described now, but a word or two about the pre- 

 cautions to be observed in taking the photo- 

 graphs may be of interest. The images must be 

 as circular and uniform as possible. (i) The 

 guiding of the telescope must be as perfect as 

 possible. (2) The lenses of large object-glasses 

 must be adjusted with great care so that there 

 may be neither tilt nor eccentricity between them. 

 (3) Photographs should all be taken with the tele- 

 scope pointing in the samfe direction. One cannot 

 be taken when the field is east and another when 

 it is west. Atmospheric dispersion and possibly 

 minute flexure of the lenses cause slight deforma- 

 tion of the images which may be scarcely visible 

 to the eye, but appear in measures. (4) The star 

 the parallax of which is being determined and 

 the comparison stars should have approximately 



