S31 
071 the Countries tf Congo and Loango. 
in the same manner. This, notwithstanding the height of the 
tree, is easily accomplished. The climber provides himself with 
a tough woodbine hoop, the circumference of which embraces 
the tree and his body, but with so much space intervening, 
as permits him to lean back at ai'ms-length from the tree, 
thus enabling him to hx his feet firmly against the knobs. In 
this way, by jerking the hoop upwards, he ascends very quick- 
ly- 
The wine is always extracted from the male tree ; the female, 
which bears the nuts, being too valuable to use in that way. 
The nut is nearly of the size and figure of the walnut. Each 
tree produces three or four bunches, which are sometimes so 
large that a single cluster ,has been known to weigh above 100 
pounds. 
( To he continued,) 
Art. XIV . — On the Method of finding the Dip or Depressicni 
of the Horizmi. By Adam Anderson, Esq. A. M. 
F. B. S. E. 
rp ^ ... . 
HE altitude of a heavenly object, as it is obtained by ob- 
servation, being an arch of a vertical circle intercepted between 
the object and the apparent horizon, it is evident that this arch 
must be greater as the eye of the observer is raised above the 
plane of the horizon. Let A, Plate VI. Fig. 4. be any place 
on the surface of the earth, b AB the sensible horizon of that 
place. A' a point directly vertical to A, and 5'A'B' the corre- 
sponding horizontal plane ; then C being the centre of the earth, 
if O be an object at a great distance, the angle B'A'O, will not 
differ sensibly from the angle BAO, the apparent altitude at 
A ; but at A' the apparent altitude will not be the angle B'A'O. 
but the angle OA'E, which differs from the former by the angle 
B'AE. This difference, which varies with the altitude of the 
observer, is called the Dip of the horizon, and is equal to the 
angle at the centre ACE. We shall compute the magnitude of 
it for a particular altitude in feet, and then give a general ex- 
pression, by which it may be determined with sufficient accuracy 
in all other cases. 
