408 



Mr. F. Chambers on the Variations [Dec. 7, 



Now if the whole atmosphere participates in the movements which 

 are observed in the low^er strata, an increase or decrease of the pressure 

 of the air, very nearly in direct proportion to the quantity of air which 

 enters or leaves any blockade, must result. Assuming that the whole 

 atmosphere does participate in these movements, and taking 30 inches as 

 the mean barometric pressure for each of tKe blockades, the increment of 

 pressure i for each hour of the day is obtained by the proportion 



i\/r QA • • M X 30 

 a 



where a is the area of a blockade in square miles (2,287,336 square miles 

 in the present case), and M is the number of square miles covered by the 

 air which enters or leaves the blockade per hour. Adding successively 

 these hourly increments, we obtain the barometric variation which results 

 from the variation of the wind from which the values of M were 

 calculated. The variations corresponding to the successive values of 



N E 



, , of Table II. are given in the following Table (p. 409), and 



graphically represented by figures 10 to 12. 



Figure 10 (p. 404) shows the diurnal variation of the barometer that 

 would result from the diurnal variation of the north component of the wind 

 only. It is remarkably regular in form, and its turning points corre- 

 spond almost exactly with those of the observed barometric variation in 

 low latitudes. The diurnal movement is greater than the nocturnal one, 

 as with the observed variation; but the morning maximum, in accordance 

 with what might be expected from the notion of a gradually decreasing 

 oscillatory movement of the atmosphere, is perhaps not as proportion ally 

 high as in the curve showing actual barometric variation. It is, however, 

 higher than the evening maximum, a fact which could not be explained 

 by the supposition that the movements of the N component of the wind 

 are simply of the nature of oscillatory movements resulting from the 

 midday disturbance, but which requires that some other cause should 

 also be in action at the same time. This is probably to be found in the 

 gradual cooling of the air during the night hours resulting in a gradual 

 return towards the equator of the air which had moved outwards during 

 the day hours, causing the outward oscillatory movement of the night to 

 be smaller than the subsequent inward movement. And this is probably 

 also the reason why at places near the equator, like Batavia, the diurnal 

 barometric inequalities for the night hours have greater positive or 

 smaller negative values than at places further removed from the equator, 

 like Bombay. Fig. 10 has, as might be expected, a smaller range than 

 the actual variation of the barometer. 



Fig. 11 is of the same general form as fig. 10, with the exception that 

 the morning minimum occurs at two hours instead of between three and 

 four hours as in the latter figure. This is probably an irregularity which 

 would disappear with a longer continuation of the observations. The 



