, 
APRIL 4, 1884.] 
eye-memory the observations plotted on several sepa- 
But this combination can be made easily 
rate charts. 
SCIENCE. 
403 
the circle of three hundred mile radius, where the 
longest arrows represent hurricane violence, or twelve 
of the Beaufort scale; and, second, that 
these winds have, with notably few ex- 
Pi . ~ ceptions, a distinct departure from a cir- 
Pa| ~.600 MILES cular path to an incurving spiral. Very 
a \ ye “kROM CenTRe evidently, therefore, the centre of this 
~ ve Z S storm would not bear ‘ about eight points’ 
a / a aan to the left of the wind’s course, as the 
Sega poy Lee 3 \\ older mariner’s guides put it, but gener- 
/ LY as ally about six, or in some cases even as 
‘eS ae \3o00 \ ¥ little as four, points to the left, as has 
j \ \ | / ro, cs \ been shown in many other examples. 
/ ; “ Me i“ sen \ Fig. 2 shows the result of similar treat- 
H \ i. we \ i ment of several days’ records of storm- 
fF ess / 1 winds in our northern states, as mapped 
i { \ in the Signal-service daily bulletins for 
} ae \\ | i October, 1877. Observations north of 
} WA \ co me \A | } the centre are unfortunately rare, aS we 
\ | eae ! J have not as yet sufficient stations in the 
as a 7 | British possessions. Although the three 
. \\ ae ! hundred arrows show numerous discord- 
\ \} eM 7 ! ant directions, the general motion indi- 
\ as S Sr } ] Je cated by them is again clearly an incurv- 
\ ta 1 U ing spiral. 
\ = ? A This method of concentrating obser- 
\ ‘\ y vations on a single diagram may prove 
‘\ / = 7 of service in several directions of storm- 
zee \ \. f study, being applicable to the determi- 
ee ~~“ oo nation of the general form of isobars, 
ae ag and areas of cloud and rain, as well as 
mS is to the investigation of the inclination 
TSS and' velocity of the wind in different 
‘quadrants of the storm, or at stations of 
| different situation as to distance from 
and accurately by tak- 
ing off the records of ye =U, 
successive dates on a Z t pre Goo 
single sheet of tracing- Wwe ° z 
paper. A cross on the = Bang ec Se ‘XN 
paper marks the storm- | y PANE \ f 
centre, always to be */ Se \ S x 
placed on the middle of | / A 4 ee 4 
the area of low pressure; / - 2 \ ae: 1 
and a north line, laid sre aa NI ae) . WA 
parallel to the meridi- Ly : ° : NG \ 
ans, serves aS a means recy! in| \ iN © Nos 
of orientation. Alarge {| _. /, v SS N Vatce es 
number of observations \ SF Sse Mut if. eee ~ — 
may thus be transferred \ Kl lam oe a S 
to a single figure, and 
every one of themfalls ~ 
in its proper position (4 
with respect to the cen- 
tre of the storm. 
The synchronous 
charts of the North At- 
lantic for August, 1873, 
prepared by Capt. Toyn- 
bee of the British me- iy * 
teorological office, yield Ho) iN ° 
a hundred and fifty-sev- re x 
en wind-records within 
six hundred miles of the 
centre of a cyclone that 
passed from the West . 
Indies along our coast 
in the ten days from the 
18th to the 27th of that 
month. These are all 
brought into the accom- 
panying diagram (fig. 1), which shows very clearly, 
first, that the stronger winds are nearly all inside 
| | 
ule 
x |r] ia. iN 
Te / Se a WN 
A eh aah 
[sy me 
Ne Sly, 
ha ) SN eA 
4 \ 14 
Se ae ~~ 
> NS nl ee 
— 
TGs) 2% 
the coast-line, or elevation above sea-level. 
Cambridge, Mass. W. M. DAVIS. 
