MONTHLY WEATHER REVIEW. 2%
.. . .- .. ..
-" - 8 blazing fan.
Bancroft, I<ossutli coiiiitjy, Iowa : ail auroral tlisplay coni-
rnclimd a t 8.10 p. 111. of tlio 2ild ant1 continued, with \.;trying degrees of brightness, ulitil 11 1). in. I t colisisted of a bright
arch, with rapidly changing streamers darting from tlie uplw
snrfiice. Giirditior, Kouiiebec county, Miiinc : brilliant iiurOIY1I streltm-
ers were socii fi1slliiig fro111 above :I t l ~k C ~O U C ~ ilt 0 1). 111. of
&]le 33d. At 10 1). 111. the stre;iincrs hiid COiLleSced ;11ld tilt? (lis- pl:~y resembled inoonl iglit.
WililleilllicCi\, ?X~r:id:l: ;L faint R U ~O ~:L it^ visible for ;ibont
tlro Ilours (luring tlie evcniiiy of tlic 23d. I t \w s first socii
about 8 1). In. ;ind t1isni)penred at about 10 1). in. The liglit
fi~(le(1 iiwi~y so griidu:illy that tt was scnrcely possible to note
tile esact minute of its disappoaraiice. The display WLS
bi*iglitest at a few rnilnltes after 9 1). in.. though a t no time \vas it very brilliant. Its first y)pc:trance \vas in the forin of
8 rollow glow in the uortliern horizon, wliich bocamo inore dis- tiilct as the twilight faded :ind tlie sky became darker. I t tlioii
presented the ;q)l)c:irmce of ;in irregular arch of yellow light.
,At times a greonisli t i u t could be observed U e a r tile uppor edge of the ,ircli. The :izimutli of the ivosteru extremity w;w 1550 1'110 clogroe of bril-
liaiicy \wried a t tiinos, but tliere were 110 flashes of light, iior
ally inotioii oftlie arch, oscept that in t1isnppe:triug the size
of iirch diiniiiislied, seeniitig to sink down aut1 fiiiiillg (lis-
al)pear belom tlie liorizoii. Dulnth, Miiinesot:t: an aurorii was observed froin '3.30 1). ~1 .
of' the 23cl until tlaylight of the 24th. The a11ror;i 8s first soen
coilsistetl of' palc yellow light extending froin ;ixiiniith !)So to
260'. At 10.30 1). 111. i t had griLtlUiiIIy :IsSllIned the form of an arch, wliicli ostmentlcd froin :rzinintli 110' to 255O, with an :i]ti-
tlicle of 30O. Sti*cninors itppoaretl at 10.45 1). IU., and contiiiile(l to incruiiso i n size and iiuiiiber until 11.15, wiieii the t1isl)lay
was quite brillinlit. . Fort I3uf0rd, 1)akot;i : a fiiie tlllroriil display coiniuenced :Lt
10.25 p. in. of the 23tl. I t first appenrcd iLS two brilliant wliitc
arclies of light ostentliiig from iiortliwcst to east. Tlie Iicight
of tho u p p ~ iircll WIS :ibout 25'. Tho l o ~e r arch was bounclecl
beno:itli by it VO~J.' (lark sogiuctit tliroitgli wliicli tlic stws were
dimly seeu. At 11 1). in. :i third, and pirt of ;i fourth arch
were forined, the fourth extencliiig iie:Lrty to tlie zcnitli. At t h i s time the e:isterii extreinitios ofthc ;LrcIics begm to break into yollow strcniners wliicli shot u p like fl:imos to ;I lioight of
30'. At, 11.30 1). I U . the westcrii sides of the irrclies \wre per.
fect, while the otxsterti eiitlx axsunied the foriu of a large white
curtain xwayed by the wiiid. Tile t1iq)l:ty f'adod from viow shortly before iuiduiglit.
LS C~OSSC, Wisconsin : :I red auroriil light W;LS uoticed at
9.20 p. In. of tho 23d ill t k ? llortlierii sky. At 9.45 p. IU. a
pale wliitc arch h:id formed. Sliortly alter 11 1). in. n second
of the casterti S15O; altitude, 15O.
Dulutli , Minnesota.
14th .--E;wt,port, Maine ; Mackinaw City, MiC1iig;lu.
17tli.--Port Tottoil, Ihltot,a ; Mackinaw City, Michignu ;
2Otli.-Orono, Maine ; Borliii Mills, New Hanipshire; Ne\!--
2lst.-Lunenburg, Vermont.
23tl.-New Hnreii, New London, Hartford, North Colebrook, :~nd Southiiigtou, Coilneeticut ; Wobstor :uid Fort Totten, Dakota ; lLiley, Illiuois ; Fort Xadison, Iudepeiideiice, N m - cliestor, :uid Monticollo, Iowa; Allison, l<:ius:hs ; E;lstport,
I Portlalid, l h r Hai*bor, ;in.d 0 ~0 1 1 0 , Maine ; Wootlstock, Mary-
I laritl ; I3luo Hill Observatory, Cambridge, Boston, Fall Riror,
I Miltoil, ~rincetoii, Soiiiersot, aiicl Westborougli, M:WW
' / cliusetts; Griuitl Hnveir , Maclriiinm City, and Traverse City,
I Micliig;in ; Moorlie:itl, Miiiuesotn; 1'opl:tr River, Montaun ;
' Viilontiiie. Nebriiska ; Kiisliiia, New I3ampsliirc ; Ihrerly,
Xiytoii, I)over, Mooivstown, mtl Upper Rlouiitclair, New
Jersey ; AIb:itij, Cooporstowii, I+ictoryville, Mouiitniiiville,
ind Setiwlret, New York ; Bctlilelietn, Cilt;iwiss;t, Dyberiy,
iiirl Wcllsborougli, Poiinsy1v:ini;t; V;iriety Mills, Virginia ;
Fort Spok;uie, Washington Territorj ; Manitomoc ant1 Prairie
In Cliico, Wisconsin.
24tli.--Nem Londou, Connecticut ; Lunenburg mid Post Mills, Vertnoiit.
25th .-Pokiii, Illiiiois.
2Gtli.-l'ekin, Illinois; Atlantic City, New Jersey; New-
2'7tli.-Windsor, Illinois.
l(itll.-Westborougl1, hlnssacllnsotts.
New port, Verniori t.
port, Vermont.
32tl.-Block ISl~t1id, ltllode Isl;ind.
,art, Vermont.
THUNDER-STORMS O F AUGUST, 188G.
[Hy Jr. Prof. 11. A. H\z~.:s.]
Duriiig August there wore received from roluotary observers 531 reports of distinct storins ; from Sigu:d Service observers
337 ; ;tiid from special thunder-storm obsorvers 1,884, m;iking
R totill of' 2,853, or 470 I U O ~O t h l tluring June, alld 144 mort)
tli;ui cluri tig July. The distribution by states and districts will
be seeii ill the :tcconipaiiyiiig table. This tablo docs not giro
MI itlea of the relative fiwpency in the different states, 2s
4onic 1i;ive mtny tnore observers tltau others, but it mill serve
for coinp:wisoii with ainiilar tnl)les i n previous months. The
days of greatest number weru lst, 210; lltli, 203; Utli,
238; 13tli, 273; ;ind lGtli, 3'38; :ind of least naniber, were ad,
19; 31, 12; Ytli, 14; l!)th, 17; 2Otli, 18; 24th, 6; 25tli, 17;
26th. 13; mid 31st, 18. I t will be SOCII that tliere wero two
well-market1 poriotls of thniider-storm activity, froin the 11th
to 17tl1, :ind from the 37th to 30th. Oii the lGtli the conditions
were specially interesting, it being tho hottest day of the
iuonth, tho teniperitturo rising to 104' i n the afternooil. On this t l i t t e there wcro tlic most storms of any t h y , and soine idea
of tho conditiolis 011 this date inity be glenuotl froin c1i:irt i s .
Most a11 the storm on this thte occurred beforo noou, lieuce
the map of isobars, isotherms, mtl mind-directions has beell
choson for 7h.00. It will be seeu that, as i n tlic l)rcvioiis
iuonth, nearly every storni is in the southeast quadrant of tho
low area. This clinrt is spociiilly interesting as showing tho
conditions usliering in tile tortindoes of the aftcruoon of this
date, elsewlicro described in detail.
2 30 MONTHLY WEATHER REVIEW. AUGUST, 1886.
-. __ __ -. - . .-- -- - _- _. ._
Thunder-storms hy districts; August, 188G.
.........
31. I Total.
I / 6
I
-- -!-
I , 4
S a g
a I 79
4
, 19
....... 1 0
...... .....
.......................... Went \'irginia.. ................
CHART OF ELECTROMETER READINGS.
[liy Prof. 1'. e. ~IF:SIlESIIAI.I., i ~S S i S h l l t .]
0bserv;Ltions Iiave been in;ulc during tlic nioutli of Augnst,
:LS USu:tl, at all stations.
At Boston, Massachusetts, negative indications occurr~d 011
the followiiig dates: l s t , 3d, 4th, titli, 9tli, 12tli, 14th, lGtll,
35th, mtl 30th. Raiu occurred during night of 1st; atternooii
Of 3tl ; t~lirilig night of 5th ; during night of 7th ; duriiig after-
110011 Of 12th ; morning of 14th ; ant1 ;Lfternoon of 30th. The
moruing of tlio 25th mas cloudy afid threatening. Tlie oiily
(late during tlie moritli on wIiicIi rainfall w a s not accomp:rnied
or preceded by negative intlicatioris was August 7th. This was
a foggy and damp day with fresh east winds, a i d was cllarac- terizetl by unut;nally high positive potentials.
At Ithacn, rJew York (Cornell Univt:rsity), negative vnlues
occurred on the 1st during thunder-storm j on the 3t1, followiug
rain; 011 the 2d; 011 the Gth, precetling rain serer:il Iioursj 011
the 7th, followiog rain; on lGtli, preceding rain ; aiicl on 17tl1,
unaccompanied b1 rain. A very Nevcre thunder-storm occurred on the eveniiig of tile llt11, not preceded by any unusual indi-
cation. A Revere tliunder-storm occurred 011 tlie lGth i n wliicli
the indications were tliroughout positive, ranging from 152 to 2,625 volts.
The first diagram of chart vi represents a series of mean values for that part of the day between 9 a . m. and 3 1). in.,
obtained i n the followiIig way : All dates of thunder-storms
and abnormal variations are omitted i n this discussion of ineau values, as it is evitleiit that the great values obt:iiaed at these
times would eitlier masque or distort tile diurnal variation due
to more regular c:mses. Groups of five dajs are taken, and the mean of each of the four dnilp observations ilotermined.
This, i n the case of Washington City, is represented by the
line numbered 1 ; the menil observations made on J u n o 29t11,
30th, July 3d, 4th, and 6th ; interroning dates being omitted
for the re:isom above stated. The inean values for the 9 antl 11 a. In., 1 1). m., 3 1). m. observations were, 138, 126, 138, and
114 volts, respectively. Similarly for tlie G t l i , 7tl1, 8tl1, loth,
and lGtli, tho curve iiumbcred 2 represents tlie incan values
92, 95, 100, mtl 91 volts. For the 17th, 18th, 20tli, 22tl, alld
24th, curve numbered 3 represents 75, 91, 102, and 109 volts;
and for tile 26t11, 27th, 28th, and August 4th a~id Btti, ciirve
num1)ered 4 represents 78, 90, 996, a~icl 107 volts. Tlic mean. of these values, i n turn, is represented by the dark-liiied curve
ilurnbered 5. Tho values are 7G.G, 80.4, 85.0, and 84.0; or, for this time and locality, tlic curve sccuis to sliow a steady rise
during tlie foreuoon, reaching a maximiirii about tlie early part
of tlic afternoon.
The second di;igram nu chart vi. represents ;i like tletermin-
tioii for Uoston, M:issachusetts. Observations are Inade here
nt tlie Massachusetts Iustitute of Technology, antl for miny
reasons we ouglit not anticipate any striking corrospontlence.
Curve number 1 represents the mean of tlie observations inado
011 June 2930th, JuIJ- Ist, 2d, 3d, giving as v:ilues-minus 8.9,
minus 11.9, minns 7.G, ;inti plus 18.3. From the 4th to tlie 8th,
inclusive, tlie values aro throughout on tho 1ieg;ttive Bide, 13.7,
12.8, lG.!I, ;ind 15.4. From tlie 9th to 13tli, inclusive, the ralues
are positlive, 9.6, 7.8, 6.3, :uid G.2, representetl by curve num-
bered 3. Curve nuinberd 4 gives tlio values from the 14th to
the l8tl1, inclusive, viz., G,10, 2G.2, 14.2; curve numbered 5, the wilucs from the 18th to tlie 23d, 65.8, 90.7, 128.2, mid 125.1,
and ciirve numbered G , tlic v;iliies of tlio 24th, 25tl1, 2Gtl1, 27th,
and 29t11, minus 1.3; plus 43.8, 44.5, 54.3. Tlie nitmi for tlie
month is represented by tlio curve numbered 7, re1)resentiiig
the following mean values for tlie four observations at 9 and
11 a. m. :uid 1 alii1 3 p. m., 9.4, 21.3, 2 9 4 33.8, respect-
ively. The tliird portion of tlio first t1iagr:im of cliart vi gives
the values, obtained i n tho mine way, for New Ilnvcii, Con-
necticut. From June 2&29tIi, the mean v:iliies are 9.4,8.2,0.0,
and 8.G; from J u n o 30th to July 6tli, 11.6, 15.8, 8.6, 8.8; from
July Gtli to lOtIi, represented by ciirve numbered 3, !).4, 16.4
8.6, 9.0, froin Jn1y 12th to 17tl1, represented by ciirve nnmbered
4, 12.0, 7.0, 6.8, 12.0, antl from 19tli to 24tl1, onlittirig tliu 21st,