3i6 



NATURE 



\Aug. lo, 1876 



Bombay, were reversed in direction and taken together 

 with the latter, showed a tendency to a cyclonic circula- 

 tion of the atmosphere around the Peninsula during 

 falling pressure, and an anticyclonic circulation with 

 rising pressure. Moreover, the east and west components 

 agreed almost exactly in epoch with the north and south 

 components, the result being a movement of air from the 

 north-west, with falhng pressure, and from the south-east 

 with rising pressure. These facts, taken in conjunction 

 with the positions of Bombay and Calcutta, on opposite 

 sides of the Peninsula, seemed to point to the differential 

 conditions of land and water being probably concerned 

 in the phenomenon. Another and not less important fact 

 connecting the winds with the diurnal oscillation of the 

 barometer appeared at the same time. When the wind 

 variation was analysed by Bessel's method, there appeared 

 an east and west oscillation of considerable magnitude, 

 corresponding in epoch with the barometric inequality 

 expressed by the first periodical term of the barometric 

 formula. This was easily distinguished from the oscilla- 

 tion of the sea and land winds, since the latter is nearly 

 north and south at Calcutta. At Bombay where the sun 

 and land-breezes are nearly east and west, such an oscil- 

 lation would be undistinguishable, even if it really exists. 



The east and west oscillation of diurnal period indicates 

 an outflow of air to the eastward during the daytime, an 

 inflow from the east during the night, and the former 

 phase of it evidently corresponds to the hot winds of the 

 Gangetic plain and northern India, and indeed to the 

 day- winds of the dry months of the greater part of India. 

 They blow towards the sea from the eastward, only in the 

 western portion of the Dakhan, Mysore, &c. This 

 system of day- winds consists of an outflow «f air from 

 the Peninsula towards the sea on both coasts, the 

 westerly direction greatly predominating. 



The next step in the inquiry was to ascertain what 

 general cause would operate to produce this efflux and 

 influx of air ; and the obvious suggestion was that it must 

 consist in the differential action of the sun's heat on dry 

 air and water. 



Let V be any volume of dry air at pressure /*, and 

 absolute temperature 7", and let t units of heat be com- 

 municated to it, raising its temperature from Z" to Z -f /, 

 while the volume remains constant. The pressure will be 

 thereby increased from /* to /• -+- /, wherein 



(0 



Also 



t^p{T+>-,)^P 



-V'^^f'c, 



(2) 



^X 



wherein r is the density of air at the standard pressure P 

 and temperature T^, and c its specific heat at constant 

 volume, compared with water as unity. 



If now the same quantity of heat t be employed in 

 evaporating water at temperature T (the whole being con- 

 sumed as latent heat), and filling the volume of air V with 

 vapour at pressure /', the total pressure will become 

 P +/', and 



P T 



where s is the hypothetical density of water vapour at 

 P and Toy and X its latent heat at temperature T. Sub- 

 stituting for f its approximate equivalent f s 



r=r|.|'rx (3) 



and equating (2) and (3) and eliminating common factors, 



Ptc = p%\ 



i^ = P% (4) 



From (i) and (4) 



p.p' =P 



^ Tc 



(S) 



which gives the ratio of the increase of pressure pro- 

 duced by the same quantity of heat, employed in the one 

 case simply in heating dry air, and in the other in charg- 

 ing it with vapour. At a temperature of 80° Fahr. = 

 2" =541, 



P = 7*36/; 

 that is to say, when a given quantity of heat is employed 

 in heating dry air at the temperature of 80° it raises its 

 pressure more than seven times as much as when it 

 simply charges it with vapour without altering the tempe- 

 rature. With lower values of T the difference will be 

 still greater. 



This great difference is no doubt much reduced in 

 nature by the effects of radiation ; and while some evapo- 

 ration is effected on the land surface, there is some 

 increase of temperature over the sea, but it may be 

 expected that some part of this difference will manifest 

 itself in the greater intensity of the forenoon pressure in 

 the lower strata of the atmosphere on the land as com- 

 pared with the sea, and in fine clear weather as compared 

 with cloudy weather, when banks of clouds present an 

 evaporating surface. With regard to this latter point, it 

 has been shown by Lamont and Kreii's investigations, 

 that between clear and cloudy days, there is a difference 

 of this kind, and that it is manifested not only in the 

 greater magnitude of the diurnal co-efficient W, but also, 

 although to a much less degree, in that of the semi- 

 diumai co-efficient U" of the barometric formula. Fur- 

 ther evidence of the same kind is afforded by the values 

 of these co-efficients for the several months at Calcutta. 



The driest months in Northern India being March and 

 April, while July is the wettest and most cloudy. 



On Espy and Kreii's hypothesis of the cause of the 

 double oscillation, there is no apparent reason why the 

 evening maximum, arising from contraction and dynamic 

 pressure, should be equal to the morning maximum, 

 which seems unquestionably due to the increased tension 

 of the lower atmosphere in consequence of heating and 

 the introduction of vapour ; and any inequality will, of 

 course, appear in the value of U', or of the co-efficients 

 of other terms of odd periodicity. But the fact esta- 

 blished by the anemometer that an outflow of air from a 

 heated land area takes place during the day-time, at 

 once assigns a cause for the greater part of the equality, 

 viz., an alteration of the static pressure. This is not an 

 overflow in the upper regions of the atmosphere, but an 

 outflow of the lower strata, or a tendency in that direc- 

 tion. It does not, of course, follow that to produce a 

 reduction in the mass of air over a continent, there should 

 be an actual motion of the air outwards in all directions. 

 The very small forces in action will be manifested even 

 more in retarding in-flowing currents than in accelerating 

 efflux ; and it is only in very dry and highly-heated 

 regions, such as India, that they produce well-marked 

 diurnal surface winds, blowing outwards towards the sea ; 



