PAPER BY- PROF. HELMHOLTZ. 81 



strata of the atmosphere also diminishes the effect of the viscosity here 

 under consideration. 



This computation also shows how extremely unimportant for the 

 upper strata of the air are those effects of viscosity that can arise on 

 the earth's surface in the course of a year. 



Only at the fixed boundaries of the space that the atmosphere fills, or 

 at the interior surfaces of discontinuity where currents of different ve- 

 locity border on each other, do the surface forces remain the same when 

 the scale of dimensions is increased and the coefficient of friction is 

 not simultaneously increased, and this allows us to recognize that the 

 annulment of living force by viscosity can take place principally only at 

 the surface of the ground and at the discontinuous surfaces that occur 

 in vortex motions. 



A similar relation obtains with regard to those temperature changes 

 that can be effected by the true conduction of heat in the narrower 

 sense, namely, the diffusion of moving molecules of gas between the 

 warmer and colder strata. The coefficient h of conduction for heat, 

 when we choose as the unit of heat that which warms a unit volume of 

 the substance by one degree in temperature (or the thermometric co. 

 efficient of conduction), is, according to Maxwell (Theory of Heat, page 

 302): 



5 / A-- 



37/ V 



'?) 



where y is the ratio between the two specific heats of gases. 



In order to solve the corresponding problem for the conduction of 



k 2 

 heat this u is to be substituted in equation (2) instead of -, and if we 



p lU y —\,\i it is seen that in the above-assumed atmosphere of uniform 

 density under a pressure of 76 centimetres of mercury and at a tern- 

 perature of 0° an interval of 3(3164 years would be necessary in order 

 by conduction to reduce by one-half the final difference in temperature 

 of the upper and lower surfaces. Therefore also in the interchange of 

 heat only its radiation and its convection by the motion of the air need 

 be taken into consideration, except at the boundary between it and the 

 earth's surface and at the interior surfaces of discontinuity. 



On the other hand, simple computations have frequently shown that 

 an unrestricted circulation of the air in the trade zones can not exist 

 even up to 30° latitude. 



If we imagine a rotating ring of air whose axis coincides with that of 

 the earth and which, by the pressure of neighboring similar rings, is 

 pushed now northward and now southward, aud in which we can neg- 

 lect the friction, then, according to the well-known general mechanical 

 principle, tbo, moment of rotation of this ring must remain constant. 

 We will iudicate this moment as computed for the unit of mass by 12, 

 SO A 6 



