10 
ME. E. CHAMBERS ON THE DIURNAL VARIATIONS 
a manner that the ratio of one to the other would be constant. The numbers in the 
third line of Table V. show the mean diurnal ranges of the land- and sea-breeze between 
and 15ij hours in each month of the year, the values for these hours having been 
obtained in the same manner as those given in Table II. They appear to show that the 
two variations are not the effects of the same persistent local influences. 
10. Our attention has thus far been chiefly confined to that part of the diurnal varia- 
tion of the wind which manifests itself in the direction of meridian lines ; but the same 
considerations would lead us to expect an east and west diurnal variation also, of a some- 
what similar character, showing that the wind blows away from, not towards, the sun, 
and therefore from the east during the morning hours, and from the west during the 
afternoon hours, and having similar oscillatory movements during the night. Reasons 
for believing that such a variation does really exist have been given in paragraph 5. 
Its times of maxima and minima will probably be about the same time at which the 
meridional variation is at its mean positions, and vice versa ; and its directions will, unlike 
the meridional variation, be the same at the same times at places in both hemispheres 
that are equidistant from the equator. 
Unfortunately the observations for Bombay alone will not suffice for the exact deter- 
mination of the nature of the mean longitudinal diurnal variation of the wind for the 
whole year, until the precise character of the land- and sea-breeze has been found by a 
comparison of the observations with similar ones made elsewhere ; but, judging from what 
has preceded, we may assume as a first approximation to the truth, and for the sake of 
carrying the investigation a step further, that it is similar to the meridional variation, 
when the north components of the latter are taken to represent the west components 
of the former, but earlier in phase by about three hours. 
Taking, then, the meridional variation as a type of the longitudinal variation, and com- 
bining the two, we have the epitrochoidal curve represented by fig. 9, which may be 
regarded as a typical curve representing the diurnal variation of the wind in low northern 
latitudes ; and again combining the two after reversing the meridional variation, we get 
the curve represented by fig. 10, which may be regarded as a typical curve representing 
the diurnal variation of the wind in low southern latitudes. 
The curves show a double diurnal right-handed rotation, in the same direction as the 
hands of a watch, for the northern hemisphere, and a double diurnal left-handed rotation, 
in the opposite direction to watch-hands, for the southern hemisphere. 
11. In the curve (fig. 1) showing the mean diurnal variation of the wind at Bombay, 
the doable diurnal rotation is converted into a single rotation by the influence of the 
land- and sea-breeze ; but the tendency towards a double rotation is well marked, and 
the mean diurnal variation for the months July and August, when the effect of the 
land- and sea-breeze is very slight, clearly exhibits the double rotation. This variation 
is given in Table III., and represented graphically by fig. 11. 
12. The following extracts from the discussion of the diurnal variations of meteorological 
phenomena for St. Helena, 1841 to 1843, by General Sir Edward Sabiee, are almost an 
exact description of what would follow from the superposition of a diurnal variation like 
