596 
observations, covering a comparatively small but very 
interesting part of the ocean, is found in the account 
of the Gulf of Guinea by Lieut. Brito-Capello of the 
Portuguese navy.! The method of record here em- 
ployed is easily understood, but does not give a meas- 
ure of the number of observations used in averaging. 
Each fan is composed of arrows, showing the direc- 
Fie. 4. 
tion and velocity (estimated from distance made by 
sailing-vessels) of the local winds fora month. The 
angle of the fan measures the variability of the 
winds; and their length gives velocity, the distance 
between the cross-bars corresponding to the rather 
indefinite sailing-rate of ‘one knot an hour close to 
the wind.’ The apex of the fan is toward the source 
of the wind. The current-arrows show average di- 
rection, when the original observations accorded; but 
. near the coast, where observations do not agree close- 
ly, averaging was not attempted. The length of the 
arrow is on a scale of one millimetre to two miles 
and four-tenths a day. Fig. 4 is from the chart for 
March and April; the fine-ribbed fans being for one 
month, the coarse-ribbed for the other. 
Three admirable monographs of a certain number 
of the ten-degree squares (ten degrees of longitude 
by ten degrees of latitude) into which the ocean is 
divided for the convenient partition of observations, 
have been prepared under the direction of Capt. 
Henry Toynbee of the British meteorological office,? 
from logs of British vessels collected mostly between 
1855 and 1870. The equatorial squares from north 
latitude 20° to south latitude 10°, between Africa 
1 Of this, I have only the French translation: Guide pour 
Pusage des cartes des vents et des courants du Golfe de Guinée 
par M, de Brito-Capello. Paris, 1862. 
? Monthly charts of meteorological data for square 3; with 
remarks. London, 1874. 
Monthly charts of meteorological data for the nine ten-degree 
squares of the Atlantic, which lie between 20° north and 10° 
south latitude, and extend from 10° to 40° west longitude; with 
remarks. London, 1876. (The results for square 3 are repeated 
in this second volume.) 
Meteorological charts for the ocean district adjacent to the 
Cape of Good Hope; with a volume on the gales of that region. 
London, 1880. 
SCIENCE. 
[Vou. III, No. 67. 
and South America, were first studied, as the dol- 
drums of this region often cause vexatious delays 
to vessels crossing the line; and it is a problem of 
much importance to determine where the passage 
should be made. The group of squares about the 
Cape of Good Hope was next chosen as of difficult 
navigation, from their numerous gales and conflict- 
ing currents. ‘These monographs are of the 
° greatest value, not only to the seaman, but 
to the student as well; for they contain a 
large amount of information most carefully 
sorted out in time and place. Winds and 
isobars, temperature of air and water, cur- 
rents, weather, clouds, etc., all have special 
record for ten subdivisions of every ten- 
degree square, and forevery month. They 
are shown, first, in diagrams that summa- ~ 
rize all available observations; and, second, 
most of these elements are given in charts 
of average monthly results. It is greatly to 
be hoped, both for practical and theoretical 
ends, that this excellent series of publica- 
tions may be continued. Fig. 5 illustrates 
the concise method of detailed record of cer- 
tain data for each of the ten subdivisions 
of the equatorial ten-degree squares. In 
the original, every thing relating to currents 
is printed in red. The number of observas 
tions and mean force (Beaufort scale) of winds 
are given numerically, for every point of the com- 
pass from which they blow, on the two inner circles, 
and are further marked by arrows, whose length 
shows relative frequency, crossed by a curved line, 
NW. Winps NE 
o 1.98 48) “Gite 
LL tr Mtrate 10 
mr No Carr® 
[HX OVO. 
Hea 
Temp IFF. 
13S rf 
Ere. 5. 
whose distance from the circumference shows force. 
Currents are plotted for the points of the compass 
toward which they flow in the two outer circles, 
the figures representing the number of observations 
and mean rate in miles per day. ‘They are aver- 
aged for quadrants on either side of the circles; and 
an average of all currents is given in the upper right- 
hand corner of the square. Winds are similarly aver- — 
