212 Proceedings of the Royal Society 
The following Communications were read : — 
2. Why the Barometer does not always indicate the Real 
Weight of the Mass of Atmosphere aloft. A continuation 
of the Paper on this subject laid before the Society in 
Session 1876 and 1877. By Robert Tennent. 
Meteorological phenomena are in all cases to be regarded as being 
both a cause and an effect ; owing to this, and also to the imperfect 
state of our knowledge on this subject, it maybe safely asserted that 
exception proves or tests the rule. Taking up the subject in this 
point of view, much assistance is in this way to be acquired when 
attempts are made to explain the phenomena which are under con- 
sideration, and hence these negative conclusions can to a large 
extent be taken advantage of. As an illustration of this, can it be 
assumed that a barometer placed on the surface of the earth, and 
which always correctly indicates the amount of pressure on its 
cistern, will also always correctly indicate the weight of the mass of 
air aloft, both when it is at rest and when it is in rapid motion, 
accompanied by the important element of friction. Equal identity 
of pressure in both such cases is assumed, although no observations 
to justify the correctness of this assertion have been made. With 
the atmosphere in a state of perfect rest, setting aside just now the 
effects of heat and vapour, there will be an upward normal diminu- 
tion of pressure, but with the same superincumbent mass of air no 
longer at rest, but attended with the dynamical element of motion in 
the rapid upper strata, will there not then be an upward abnormal 
diminution of pressure, due to the u lifting” of the air on the surface 
and its accumulation aloft ? It is this point which was shown in 
the former paper, and is now here to be more fully explained. In 
both such cases, and under the same superincumbent mass of air, 
can the barometer, placed on the surface , possibly show the same 
amount of pressure ? It is only by a series of barometers placed 
vertically above each other to a great height, and not very far apart, 
that the amount of pressure in both such cases will be found to be 
the same. 
Mathematical meteorological investigations cannot be here intro- 
duced with perfect accuracy, and only to a small extent, where so 
much complexity, irregularity, and want of uniformity is to be 
