September 



1904] 



NA TURE 



515 



only 21,000 times the push. Halve the diameter again, and 

 the pull would be only 10,500 times the push. Reduce the 

 diameter to 1/42,000 of its original value, that is, to about 

 20 miles, and the pull would equal the push. 



In other words, a sun as hot as ours and 20 miles in 

 <lian)eter would repel bodies less than i cm. in diameter, 

 and could only hold in those which were larger. 



But it is, of course, absurd to think of such a small sun 

 as this having so high a temperature as 6000°. Let us 

 then reduce the temperature to 1/20, say 300° absolute, or 

 the temperature of the earth. Then the radiation would be 

 reduced to the fourth power of 1/20, or 1/160,000, and the 

 diameter would have to be reduced to 1/160,000 of 20 miles, 

 or about 20 cm., say 8 inches, when again radiation would 

 balance gravitation. 



It is not very difficult to show that if we had two equal 

 spheres each of the density and temperature of the earth 

 they would neither attract nor repel each other — their radi- 

 ation pressure would balance the gravitative pull — when 

 their diameters were about 6-8 cm., when, in fact, they were 

 about the size of cricket balls. 



It must be remembered that this is only true for spheres 

 out in space receiving no appreciable radiation from the 

 surrounding region. 



It would appear that we have arrived at a result of some 

 importance in considering the aggregation of small 

 meteorites. Imagine a thinly scattered stream of small 

 meteorites at the distance of the earth from the sun. 

 'I'hen, even if they be as large as cricket balls, they may 

 have no tendency to move together. If they are smaller 

 they may even tend to move apart and scatter. 



In conclusion, let me mention one more effect of this 

 radiation pressure. You will remember that radiation 

 presses back against any surface from which it issues. 

 If, then, a sphere at rest in space is radiating equally on all 

 sides it is pressed equally on all sides, and the net result 

 is a balance between the pressures. But suppose that it is 

 moving. It is following up the energy which it pours forth 

 in front, crowding it into a smaller space than if it were 

 at rest, making it more dense. Hence the pressure is 

 slightly greater, and it can be shown that it is greater 

 the greater the velocity and the higher the temperature. 

 On the other hand, it is drawing away from the energy 

 which it pours out behind, thinning it out, as it were, and 

 the pressure at the back is slightly less than if the sphere 

 were at rest. 



The net result is a force opposing the motion, a force 

 like viscous friction, always tending to reduce the speed. 



Thus calculation shows that there is a retarding force on 

 the earth as it moves along its orbit amounting in all to 

 about 20 kgm., say 50 lb. Not very serious, for in billions 

 of years it will only reduce the velocity by i in a million, 

 and it will only have serious effects if the life of the earth 

 is prolonged at its present temperature to hundreds of 

 billions of years. 



But here again size is everything. Reduce the diameter 

 of the moving body, and the retarding effect increases in 

 proportion to the reduction. If the earth were reduced to 

 the size of a marble, the effect would be appreciable in a 

 hundred thousand years. If it were reduced to a speck 

 of dust a thousandth of a centimetre in diameter, the effect 

 would be appreciable in a hundred years. 



Note what the effect would be. Imagine a dust 

 particle shot out from the earth and left behind to circulate 

 on its own account round the sun. It would be heated by 

 the sun and would be radiating out on all sides. As it 

 journeyed forward there would be a resisting force tending 

 to stop it. But instead of acting in this way the resistance 

 would enable the sun to pull the particle inwards, and the 

 fall inwards would actually increase the velocity. This 

 increase in the velocity would increase the resistance, and 

 at the same time the approach to the sun would raise its 

 temperature, increase the radiation, and so increase the 

 resistance still further. The particle would therefore move 

 in a more and more rapid spiral orbit, and ultimately it 

 would fall into the sun. Small marble-sized meteorites 

 Avould fall in from the distance of the earth probably in 

 a few million years. Small particles of dust w-ould be swept 

 in in a few thousand years. 



Thus the sun is ever at work keeping the space round 

 liim free from dust. If the particles are very minute he 



NO. 182 I, VOL. 70] 



drives them forth into outer space. If they are larger he 

 draws them in. It is just possible that we have evidence 

 of this drawing in in the zodiacal light, that vast dust-like 

 ring which stretches from the sun outwards far beyond the 

 orbit of the earth, and is at once the largest and the most 

 mysterious member of the solar system. 



T 



PHYSICS AT THE BRITISH ASSOCIATION. 



HE number of communications made to Section A this 

 year was again so large as to necessitate duplicate 

 sittings on several days, an arrangement which appears to 

 bring home to members in a forcible manner the impossi- 

 bility of being in two places at once. For some undis- 

 covered reason the subcommittee which arranges the order 

 of the papers is generally held responsible for this limit- 

 ation, and gets a considerable amount of abuse. The dis- 

 advantage of the division was particularly evident at the 

 discussion on the units used in meteorological measure- 

 ments opened by Dr. W. N. Shaw. A subcommittee of the 

 council of the association appointed to consider the question, 

 recommends the use of the absolute zero of temperature 

 with either the centigrade or Fahrenheit degree as the unit, 

 but preferably the former, and the introduction of a new 

 " degree of pressure " which is equal to 2000 C.G.S. units, 

 and involves a graduation of the barometer in nearly i/i6th 

 of an inch (006 in.), and the use of a vernier down to 

 1/160 inch. The meeting before which the matter was dis- 

 cussed was disposed to dw^ell mainly on the cost of effecting 

 the changes proposed, and owing to the scant attendance 

 of physicists, rather lost sight of the advantages of adopt- 

 ing what is practically equivalent to the C.G.S. system. 



.\ttwood's machine as an aid to the teaching of dynamics 

 was much discredited during the discussion of a paper by 

 Mr. Eggar on an apparatus for verifying Newton's second 

 law. Mr. Eggar finds that the movement of a truck down 

 an inclined plane the angle of tilt of which can be altered, 

 is much more convenient and effective than the fall of a 

 weight. 



The coefficient of expansion of hydrogen at various 

 pressures down to low temperatures was the subject of a 

 communication from Prof. Witkowski. He finds that the 

 coefficient increases with decrease of temperature, and de- 

 creases with increase of pressure, a result which must have 

 an important bearing on our standards of temperature. 



Dr. Glazebrook's account of the recent work of the 

 National Physical Laboratory made one hope that the efforts 

 to cope with the demands made on it by our manufacturers 

 for tests of materials and for scientific help of other kinds, 

 will not be hampered by the insufficiency of the financial 

 support the institution receives from the Government. In 

 order to establish a scale of temperature. Dr. Harker has 

 compared up to 1000° C. the constant volume nitrogen 

 thermometer with a thermojunction previously standardised 

 at the Reichsanstalt, and a platinum thermometer. Mr. 

 Smith has constructed and compared a number of mercury 

 standards of resistance. Dr. Stanton has been engaged in 

 determining the amount and distribution of the pressure 

 on structures due to wind, Dr. Carpenter has investigated 

 the solidification of iron-carbon alloys, and a number of 

 other important investigations have been carried out for 

 manufacturers and for the Government. 



Problems connected with radiation played a prominerit 

 part in the proceedings of the section. Prof. Poynting's 

 interesting afternoon address, which appears in another part 

 of the present issue, dealt with the applications of the 

 laws of radiation to the solar system. Taking Stefan's 

 law as a basis, the temperature of the sun works out as 

 6250° C, and that of a black body at the distance of the 

 earth from the sun at 27° C, which agrees well with 

 the average temperature of the earth. A description 

 of an apparatus by means of which he had measured the 

 tangential stress on a surface due to the oblique impact of 

 light, was also given to the ordinary sectional meeting by 

 Prof. Poynting. If E is the stream of momentum per 

 sq. cm. per second due to the light incident at an angle 9, 

 and ^ is the fraction of the incident light reflected, the 

 tangential pressure on the surface is (i— /i)/2.E sin iS, and 

 although in general it is smaller than the normal pressure, 



