1800 ] 
geve the lower degree of elevation ; and 
the higher, when winds from the oppo- 
fite quarter. prevailed. 
I refrain here from the numerous con- 
clufions which from thefe obfervations 
might be drawn, with refpect to places dif- 
ferently fituated. It would be aéting too 
precipitately to do this from fo few.exam- 
ples, the explication of which is only ha- 
zarded, perhaps hardly true with refpect 
to the places where the obfervations were 
made, and confequently requires further 
proof and confirmation. I fhall only re- 
mark, that another fcale for correction on 
account of the temperature of the air does 
not remove the above variations, nor thofe 
which in rainy weather 1 have obtained 
for places of this diftri¢ét which lie near 
to one another. It was therefore natural, 
to fuppofe that thefe differences were ow- 
ing to the effett of the vapours, which 
have fo great an influence on the preffure 
of the air, and which feem to recommend 
. the ufe of an hygrometer in obfervations 
_ relative to places far diftant from one an- 
other; the more fo, as thus we might 
perhaps be enabled to find a rule for cor- 
recting the errors arifing from the varia- 
tions produced by vapours, by which means, 
with a few obfervations, a re/ult nearly 
approaching to truth might be obtained. 
“But till then, fuch obfervations require 
the oftener to be repeated, the more the 
places are diftant from one another: and 
the refults of thefe obfervations fhould be 
claffed in a particular manner for calcu- 
lating the final mean elevation, which 
muft not be deduced from all thefe refults 
promifcuoufly thrown together. [It is 
probable that the true elevation will be 
_ nearly the medium of thofe average num- 
bers which are obtained from the refults 
for oppofite winds, and for moift and dry 
air. Hence, for determining the eleva- 
tion of very remote places, it may be ne- 
ceflary to make obfervations at different 
times during direétly contrary ftates of the 
weather; and little confidence can be 
placed in fuch as are obtained from only a 
few, or even from many, obfervations with- 
out any regard to the variations of wind 
and weather. The elevation of Bern 
above Mulheim would, from cbfervations 
» during the prevalence of winds from the 
third quarter, which he continue a long 
time, have been found 70 feet lefs than 
when the winds blew from the firf ; and 
it is merely accidental, that the medium 
deduced from all the obfervations agrees 
with that obtained from a comparifon of 
the winds, 
_ The conne&ting of the hygrometer with 
MONTHLY MAG. NQ, 58. 
Spanifp and Portuguefe Packets. 
337 
barometrical obfervations is ao doubt fub- 
ject to many difficulties. It fhould be 
portative, not expenfive, and harmonife 
with the barometer. It will indicate the 
drynefs or moiftnefs of the air which im- 
mediately furrounds it, the ftate of which 
may be different from that of the great 
ma{s of air: and it may be doubted if it 
can ever accurately indicate the moilture 
which produces an effeét on the barome~ 
ter. However, notwithftanding all the 
difficulties attending them, De Luc alrea- 
dy exhorts naturalifts to make trials there- 
with; and what he fays on.this fubject in 
the fecond volume of his Modif. fur 1 Atm. 
is well worthy of an attentive perufal. 
I hardly need remind the readers, that 
the vanes muft be advantageoufly placed, 
be fenfible to the leaft impreffion of the 
wind, and fhould indicate both the direc- 
tion and the force of the wind, and confe- 
quently fhew likewife when there is a 
calm. From the direction of our common 
vanes, we often conclude that the wind 
blows which is indicated by them, though 
fuch indication is only the confequence of 
a preceding wind, to which fucceeded a 
calm in the atmofphere. 
ee 
Zo the Editor of the Monthly Magazines 
SIR, 
AM a-canftant reader of your excel- 
Ji. lent Magazine, and would confider my- 
felf much obliged-if any of your contribu- 
tors would have the goodnefs to flate a 
method by which the neutral falts may’ be 
feparated from barrilla and kelp, leaving 
‘the alcalt.. The. method ‘wanted is fuch 
as may be ufed in large by a manufac 
rer. Your reader truits you will oblige 
him in this requeft, if not, will you have 
the kindnefs to acknowledge in your next 
Magazine the receipt of this, and point out 
where the withed-for information may be 
met with by your-friend and férvant, 
Gatfbead, 3 April, 1800. M.-P. 
EEE 
For the Monihly Magazine. 
COMMUNICATION Jbeteveen spatn and 
PORTUGAL, and their COLONIES. 
By C. A. Fisuer. 
WERHAPS fome account of the cor- 
refpondencia ultramarina, or of the 
packet-boats feat to the Spanifh and Por- 
tuguefe colonies, may not be unaccepta~ 
ble: as even Bourgoing makes no mention 
of the former, although they have been ¢:~ 
tablithed ever fince the year 1744. 
There are, in Corunna, feven fri- 
gates and’ fix brigantines: the former of 
from 160-350 tons and 12-20 guns, the 
latter of 120-150 tons and 16-20 guns- 
Xx oO aGmae 
4 
ie 
