October 9, 19 19] 



NATURE 



113 



ItETTERS TO THE EDITOR. 



XThe Editor does not hold hhnsclf responsible for opinions 

 expressed by his corrcspondotits. Neither can he undertake to 

 return, or to correspond ivith the ivriters of, rejected manu- 

 scripts intended for this or any other part of NATURE. No 

 notice is taken of anonymous conimunications.\ 



Temperature in the Sun. 



Whenever a spell of hot weather occurs it is 

 'Common to see published accounts of the "tempera- 

 ture in the sun." These sun-tempeiatures have little 

 meaning unless the other surrounding conditions are 

 also stated. 



Comparatively few people realise that a thermo- 

 meter indicates nothing except the temperature of the 

 fluid in its bulb, and that to draw any useful inference 

 from that temperature it is necessary to know how 

 the heat which produced it was supplied. 



Heat may enter a thermometer from the air bv 

 iconduction, aided by convection currents and wind, 

 .and also by radiation from distant objects. In 

 ■general, both these sources contribute to the total. 



The true temperature of the air is indicated only 

 when the thermometer is screened from the radiation 

 of any body which is not at that temperature, and 

 the ventilated shelters in which meteorologists place 

 their instruments are intended to secure this condi- 

 tion. In ordinary cloudy and windy weather thev 

 answer the purpose, but in sunshine and calms the 

 whole shelter becomes heated, and the thermometer 

 readings are too high. 



When a thermometer is fully exposed to the sun 

 a large part of the heat received is supplied bv radia- 

 tion, and the apparent temperature will varv with the 

 character of the surroundings, including the nature 

 of the glass of which the bulb is made. 



Of the total radiant energy falling on the bulb part 

 IS regularly reflected and the remainder scattered or 

 absorbed, but it is only the energy absorbed during 

 its passage through the glass of the bulb which raises 

 the temperature of the contents — at any rate, in 

 mercury thermometers. The limiting temperature is 

 reached when the surface of the bulb loses, bv con- 

 duction and dark radiation, as much heat as will 

 balance the supplv. 



If the bulb is smoked there is scarcelv anv reflec- 

 tion, and thus a bulb coated with lampblack will 

 nach a higher temperature than a black glass bulb, 

 and this, in turn, will be higher than if the glass is 

 , transparent, and if the exterior of the bulb is silvered 

 I there will be an even greater difference. 



Thus, in the same place and in the same sunlight, 

 four different temperatures might be indicated bv 

 accurate thermometers, each reading differing from 

 the others by several degrees, the differences depending 

 on the different absorptive and emissive qualities of 

 the glass and its surface. 



Tlie actual difference between the apparent " tem- 

 perature in the sun " and the air temperature may in 

 this country be as great as 50° F. In the tropics 

 I believe it may be considerablv more. 



Darwin, when in the Galapagos Islands, wrote : — 

 " On two days the thermometers in the tent stood for 

 some hours at 95°, but in the open air in the wmd 

 and 'sun at only 85°. The sand was extremely hot; 

 the thermometer placed in some of a brown colour 

 immediately rose to 137°, and how much above that 

 it would have risen I do not know, for it was not 

 graduated any higher. The black sand felt much 

 hotter. ..." The true air temperature was probably 

 about So°, so that the sun's radiation heated the 

 ground 60° or 70° more than air. 



I remember seeing in a sunny window in January 

 the thermometer standing at 108° when the room 

 temperature was about 60° ; and in the recent warm 



NO. 2606, VOL. 104] 



weather, when the air temperature was about 80°, 

 a thermometer shielded from draught by a thin 

 smoked glass tube indicated 128°. 



These facts show how little meaning can be 

 attached to "temperatures in the sun" unless all 

 the conditions are stated. 



If a blackened thermometer is enclosed in a good 

 vacuum chamber of transparent glass, and is care- 

 fully screened from all ground radiation, its readings 

 in the sun will give a good comparative measure of 

 the transparency of the air to radiant heat; but if the 

 true temperature of the air is required, the thermo- 

 meter should be surrounded by two or more con- 

 centric silvered glass tubes ' through which a rapid 

 draught is maintained. In this way the effects of 

 radiation are almost eliminated, and all the heat 

 received is supplied by conduction. 



\. Mallock. 



6 Cresswell Gardens, South Kensington. 



Percussion Figures in Isotropic Solids. 



The accompanying photographs are of interest as 

 illustrating the manner in which an isotropic solid 

 breaks down unde- the stresses set up by impact 

 when these exceed the limits of perfect recovery, and 

 have a bearing on the theory of the collision of 

 elastic solids developed mathematically by Hertz. 



Figs. I, 2, and 3 are pictures of the percussion 

 figure, taken from three different points of view, 

 produced on the surface of a thick glass plate by the 



Fig. 



Fic. 3. 



impact of a polished hard steel ball. Near the centre 

 of th-e region of contact between the sphere and the 

 plate the stresses are mainly in the nature of a 

 volume-compression, and fracture accordingly does 

 not originate there, but occurs at or near the margin 

 of the compressed area in the form of a fine circular 

 crack which spreads inwards into the plate obliquely 

 in the form of a surface of revolution. This is clearly 

 shown in Fig. i, which is a front view of the percus- 

 sion figure by reflected light, the dark circle in the 

 middle being the uninjured area of contact between 

 ball and plate. The circular interference-rings seen 

 in the picture are a measure of the separation of 



