12 AIR MASS ANALYSIS 
quently observed with winds having a 
southerly component, and tropical air 
masses not infrequently have sections 
in which the wind may blow from the 
northwest. This is true particularly 
in upper levels. In the placement of 
fronts, however, winds are very im- 
portant. Some of the fundamental 
concepts of this phase of the problem 
will be taken up in a later article in 
this series. 
Of the six elements named above, 
the thermal and hygrometric quanti- 
ties are the best indices to be used in 
following and identifying air masses. 
Of these, the equivalent-potential tem- 
perature is the most conservative, 
combining the conservative qualities 
of both the potential temperature and 
the specific humidity. The above dis- 
cussion, necessarily brief and sketchy, 
will enable us to take up the Rossby- 
diagram in the next article of this 
series. 
Ill. THE ROSSBY DIAGRAM—PLOTTING ROUTINE 
It is of primary importance in 
synoptic meteorology that air masses 
be followed as they move from area 
to area over the earth’s surface; the 
weather at any given locality is, of 
course, largely dependent upon the 
type of air mass present and the 
modification which the body of air 
has undergone during its history. In 
the preceding article it was pointed 
out that the use of representative 
observations is necessary for the 
identification of air masses from 
different source regions. The upper 
layers of the atmosphere being com- 
paratively free from surface effects, 
it is best to use data from upper 
air soundings. It was also shown that 
the most conservative quantities that 
can be used for purposes of identifi- 
cation are not the ones directly meas- 
urable—temperature and relative hu- 
midity—but those indirectly obtained 
—potential temperature and specific 
humidity. These two quantities, as 
will be seen later (Article X), are 
also used in isentropic analysis. 
Realizing the importance of a quan- 
titative method for identifying air 
masses, Professor C.-G. Rossby, of the 
Massachusetts Institute of Technol- 
ogy, developed the diagram which 
bears his name. As might be sup- 
posed, the diagram makes use of the 
most conservative quantities—poten- 
tial temperature and specific humid- 
ity. Potential temperature is the 
ordinate of the diagram, specific hu- 
midity the abscissa. Since the equiva- 
lent-potential temperature is a func- 
tion of potential temperature and 
specific humidity, another set of 
lines representing constant equivalent- 
potential temperature may be con- 
structed on this diagram. The sig- 
nificance of these lines as well as the 
interpretation of various curves on 
the Rossby diagram will be discussed 
later. At present we shall concern 
ourselves with the mechanical pro- 
cedure of constructing, on these dia- 
grams, curves representing aerolo- 
gical soundings. 
In the first article of this series it 
was pointed out that, in an adiabatic 
process, the relation between tempera- 
ture and pressure is given by the 
formula: 
T, D1 0.288 
T2 ( pez ) 
where 7: is the temperature at the 
pressure pi, and 7. is the tempera- 
ture the particle assumes at the pres- 
sure p2 If we now specify that the 
particle, originally at temperature 7, 
and pressure py: be compressed to 
1000 mb pressure we have 
T., D1 0.288 
Te ( 1000 
But if the particle is reduced dry 
