278 REPORT — 1844. 



which the line 29*500 is crossed. They appear in fact to be converging see* 

 tions of a wave having its posterior trough near and to the north of Longstone. 



(4.) The western section from Scilly to Longstone presents a nearly un- 

 broken descent, and it is probable that the wave which is so apparent on the 

 central and eastern sections did not extend to this ; there is a slight rise a little 

 to the north-east of Bardsey, which may probably result from the extremity of 

 the wave. The section from Brussels to Scilly is also of an unbroken character. 



(5.) The central and southern sections are crossed by two others, which ex- 

 hibit the distribution of pressure over the area included by Dublin, Munich and 

 Geneva ; thesesections ( Plate XL VI.) are extremely interesting, they s upply defi- 

 ciencies in the sections already given, and are both characterized by two bulges 

 or superposed waves. The lowest of these waves (A 1) stretches from the coast 

 of Wales to the coast of Holland on the Munich line ; and from Bardsey to the 

 coast of France on the Geneva line. From a careful consideration of the sections 

 drawn in other directions, there can be no question that the bulge on the eastern 

 and central of the English sections and this wave is the same. It covered nearly 

 the whole of England at this time (Nov. 9.3.0), its apex or crest stretched from 

 Cornwall to the shores of Suffolk, taking a somewhat circular direction about 

 midway between London and Birmingham, and it is probable that it extended 

 considerably towards the east over the German Ocean, and on the continent. 



(6.) Of the wave preceding this we obtain but a small portion ; the two 

 sections from Dublin to Munich and Geneva indicate that the direction of 

 the crest was nearly parallel to that of the wave we have just examined. A 

 section exhibiting the distribution of pressure from Longstone to Geneva, 

 constructed from interpolated ordinates, also gives this wave. 



(7.) The curves exhibiting the distribution of pressure are constructed 

 from observed and interpolated ordinates. The latter are measured on those 

 points of the curves where they intersect others, and where the pressure must 

 necessarily be the same in both. The sections are consequently mere approxi- 

 mations to those that would be given were the stations of observation suffi- 

 ciently near to each other to pfoject from them the true curves of pressure 

 on a line, as from Dublin to Munich ; in fact we require a much longer line 

 than this, which probably does not give the half breadth of the normal M'ave. 

 The same principle of short intervals, which Sir John Herschel has so effi- 

 ciently applied to time in the solstitial and equinoctial observations, applied to 

 space in arranging series of observations on certain lines, would make us ac- 

 quainted with the true distribution of pressure over a tract of country, and 

 the two combined would give us the march as well as the distribution. 



(8.) The waves deduced from the observations are affected by two circum- 

 stances, namely, the diurnal oscillation, which must to a certain extent inter- 

 fere with the form of the wave, and the influence of the aqueous vapour dif- 

 fused in the atmosphere. The pressure of the aqueous vapour varies in a 

 different manner to that of the gaseous atmosphere, and will materially modify 

 the forms of the waves, &c. On the occasion we have been examining the 

 pressure of the vapour increases, while that of the 7«/<o/e atmosphere decreases. 

 It appears quite as important to examine the distribution of the vapour over 

 a tract of country as that of the M'hole pressure ; and in pursuing the investi- 

 gation, it is necessary either to get rid of the effects of the vapour, or to exa- 

 mine how far it influences the forms, directions, amplitudes and progress of 

 the waves deduced from the whole pressure. In order to render barometric 

 observations as efficient as possible, it will be absolutely necessary to observe 

 at the same time the wet and dry bulb thermometer, and it would be still better 

 to accompany them with readings of Daniell's hygrometer. The determina- 

 tion of the co-efticients of the diurnal oscillation will also become of para- 

 mount importance, that the forms of the waves may be corrected. 



J 



