NOVEMBER, 1903. MONTHLY WEATmR FtEVIEW. 509 ble to grain already in the ground, and that sown in September and Octo- ber germinated nicely, generally came up to good stands, and maintained a slow and steady growth. At the close of the month it was everywhere reported ta be in excellent condition.-A. 33. Wollaber. hmyZvania.-At t.he close of the month the soil was in good condi- tion in practically all sections and frozen sufficiently deep to cause a sus- pension of plowing; wheat and rye ranged from medium to very good, but were generally short on account of late seeding; considerable coru remained uncut; there was but little snow in any district for the protec- tion of grain or grass.-T. F. Toiuimencl. Porto Rico.-Heavy thunderstorms and high wiuds on the 23d and 44tli did some damage to fruits and small crops in some localities, but, in Fen- era], the weather was fa.vorable for growing crops and for fieldwork. The older canes commenced to arrow freely in the drier sect.ions of t,lie i early in the month, and near tlie close a few mills had begun grit The outlook for a good yielil was very promising. The yi)itng canes were in excellent condition. Coffee picking progressed rapiillq- tlurin;: the first half of tlie month uniler very favorable weather conditions, aut1 a t t,hi: end of the month t.his work was nearing conil)letion. Several small plot,s o f cotton were picked, and where the crop had proper care the ret,urn was very satisfautory. Some r k e ani1 corn were harvested, while ot,her plots were in tlie blossoming stage. Orange shiliping was active. Pastures aud stock continued in excellent. conclit.ion.-E. C. Thonipnoti. South CirroZinn.-The first six days itnd t,he niitldle of the seoond decatle were warm; the rest of the nionth was unusually oild, with killins frcist, on the 7th and 8th and the flrst ground freeze on the 18th and 19th. alter which the mouth rernainetl cold, with frequent frosts and grouuil freezes. The precipitation was light, but everilg distributed. Oats seed- ing was nearly finished, ])ut gerniinatil)u was slow and stands small. Wheat seeding made slow progress, and but little of t,liat sown came up to stand Many bolls of late cotton were destroyed iu the western countirs by frost. and IOW temperatures. The crop was picked (*loser t,hitn usual and pick- ing was practically finished. The cold weather injured fall awl winter truck, but was favorable for butchering hogs, making a saving of nearly one month of feeding.-J. W. Btriirr. South Dakota.-There was considerable cwld weat,lier aft,er the l l t h , and snows interfered locally to some extent with field farm work. but the cc3u- ditions were, on the whole, favorable. There was, however. cousideralile corn yet in the field in the Sioux River Valley at. the close of the mc)nt.h. Thrashing was ahout finished. Wintrr rye ancl t,he w r y limiteil amouut of winter wheat sowi~ were protected tiy snow during the cold weather and kept in satisfactory condition.-S. W. G h o i . T~~meaaea.-Collditious were generally favorable for gat,hering c!rops and seeding grain. Good raius fell itbout the first and niiddle of t.he month; otherwise i t was rather dry. The second half trf t,he nlontli was cold. Early sown wheat was generally looking well, ).but much of the crop was sown 1at.e and made slow progress; the acreage is much less than last year; there was some injury by freezes. ltye nud oats were doing fairly well. Corn and cotton were mostly gathered.-H. C. Bate. Decided falls in temperature occurred on the 18th and 19t.h and the 27th and 28th, Spring dover was injured ?.by the fall drought. Tesns.-The month was the driest Noveniber on record. giving freezing temperatures to the coast region. Conditions were ex- ceptionally favorable for the picking of cotton. About one-eight of the crop is still in the fields in the north portion, but elsewhere the crop is practically all picked. The freeze of the 18th and 19th killed the cotton plants. but, as there was little or no top crop, this caubed very slight damage. Wheat, rye, and oats that were up at the beginning of the month continued in fair condition, but nPeded rain. The dry condition o f the *oil greatly retardrd plowiug ana zowing. No damage was done to the sugar lane I T O ~ 11s the cold wea r. Cutting and grinding pro- grrssrd rapidly with very satisfactory ults. Fa11 gardens, paqtures, and the ranges were in need of rain-L. H. iMitrdoch. 17ah.-Teniperatiirrs during the month were genwally above normal, rxrepting during the latter part o f the second decade, when abnormally cold M rather 111 evailed. Prw4pitation was above normal over the niiitlreru half of tlie section, ptaving the soil in good condition and fa- aud srowth of fall grain, which was coming up southern half. whale, howe\ er, but little fall any piecipitati4)n fell aud the grouncl was dry ge5 were i n giiod condition.-R. J. Hyntt. l*irg;tiin.-Crop ~~rojirebs during the nionth wah niuch retarded by weather conditions that wcre Iboth colder and drier than normally. Early sown wiiiter grain wtl5 not materially injurrtl on account of its more ad- cl(i\ er ma5 dainageil, especially on M 5. -E&oartl A. Eilaiia. vy rainfall in the weqtern fie+ Wwhitigtott.-The nicluth wah onr tiou aut1 an uuurual amount of rain and snow fell in eahtern section. Tlir fir-t rlwaile wah vaiiii, the sewncl druaclr ccild with heavy frost and while t h r thirll clrl.alle was modrrately warm. On orniy weather, the month was unfavorable for farm urfivial to the groa th of fall w)\vn wheat. Late crops wrrr mostly gatherecl in all districts.-(:. N. Salisbury. Wmt li'rgii&i.-The dry wrather, fc)llowrd by the fieezing tempora- tures with no suow prt)tection cloring the latter half of the month, was very UlifaVoralJle for the growth of winter wheat, rye, and oats, and at the close of the niontli they were in poor condition. The acreage of wheat sown \vas not ab large as usual. Stock wit* generally in gi~ud condition ancl fewling began earlier than usual. Some corn wah still in shock, and the pruspwth were for a better c ~~p than had been expected. I t was too dry for turiiips.-E. C. Vow. Wiscmiti.-The uiunth was generally fair aut1 pleasant during the first ten days with tt'mperaturrs ~~J O W normal, Lilt from the 12th to the end of the month, decideclly cold weather for the season prex ailed. Moderately Ilea\ y rains occurrrd on thr llth, turning to snow. Snow occurred again on tlie 17th. 43d, and 2Ktl1, alii1 ranged in delith at the end of the month froni two to ten inches. Wintw grain5 and grashes were amply protrcted Ly the snow, and were reported in good conditiou.- W. L%f. Wikon. IVyottPing.--UuniisuallS Ideahaut weather with iuilcl temperatures pre- vailed over tlie State during thr first ancl last two weekz of the month A coli1 wave on the 17th aud 18th wab general , but war not severe on. stock. Practically all of the precipitation of the mouth fell during the storniy period froin the 7th to the 17th of the nionth.- W. 8. Palmer. \ancet1 stage of gro\vtll, but the 1 ing of aheat. oat#, rye. and SPECIAL CONTRIBUTIONS. STUDIES ON THE CIRCULATION OF THE ATMOSPHERES OF THE SUN AND OF THE EARTH. By Prof. F R l N R H. BIOELOW. 11.-SYNCHRONISM OF THE VARIATIONS OF THE S@LAR PRODI- INENCES WITH THE TERRESTRIAL BAROMETRIC: PRES- SURES AND THE TEMPERATURES. SEVERAL OPINIONS ON THE SUBJECT OF YYNCHRONISRI. The numerous studies during the past fifty years into the ap- parent synchronism between the solar variations of energy and the terrestrial effects, as shown in the magnetic field and the meteorological elements, have been on the whole unsatisfactory, if not disappointing. Just enough simultaneous variation hm been detected in the atmospheres of the sun and the earth to fascinate the attentive student, i f not to justify a large espendi- ture of labor, in view of the great practical advantages to be obtained in the future as the result of a complete understand- ing of this cosmical pulsation. The attack upon the problem has really consisted in rather blindly groping for the most sensitive pulse in the entire cosmical circulation, and in disen- tangling the ssveral interacting types of impulses. It is eri- dent that the partial failures hitherto attending this work hate been due to two principal causes: (1) The comparison was made between t,he changes in the spotted areas of the sun and the terrestrial variations, but these solar changes were not sensitive enough to regirrter a complete account of the action of the solar output. Discussions of the spots are being replaced by others upon the solar prominences and fttculie, which respond much more esactly to the working of the sun's internal circu- lation. (2) The magnetic and the meteorological observations have not been handled with suecient precision to do justice to the terrestrial side of the comparison. It is evident that all these physical data a t the sun and a t the earth must be com- puted with an esactnem comparable to that of astronomical ob- servations of position, i f meteorology is to be raised to the rank of a cosmical science. When one considers the crudeness of the meteorological data, taken the world over, clue to the character of the instruments employed, the different local hours of observation, and the divergent methods of reduction, i t is no wonder that the small solar variations have been swal- lowed up in the bad workmanship of meteorologists. The prevailing methods hate been suEcient for forecasting and for climatological purposes, but they are entirely inadequate for the cosmical problems whose solution will form the basis of scientific long-range forecasts over large areas of the earth, that iR, for forecasting the seasonal changes of the weather from year to year. It is perfectly evident that if secular varia- tions of any kind, such as the annual changes in terrestrial pres- sure, temperature, or magnetic field, are to be attributed to solar action, the original observations must be finally reduced to a homogeneous Bystem. The local peculiarities of each station 510 MONTEILY WEBTHER REVIEW. NOVEMBER, 1903 must be carefully eliminated, and the data of numerous sta- tions must be concentrated before anything like quantitative cosmical residuals can be obtained. When we consider that there have been numerous changes in the elevations of barom- eters, various methods of reducing the readings, and many groups of selected hours of observations entering into the series at the same station, how could i t be expected that any thing better than negative results in solar problems would be obtained? The skeptical attitucle of conservative students, who declare that the many indecisive results already obtained mean that there is no true and causal solar-terrestrial spn- chronism, is, of course, quite fallacious until it has been demon- strated by the use of first-class homogeneous data that the suspected physical connection is imaginary. There is but little question that the existing uncertainty is in fact baRecl upon the use of the very imperfect niethoils of observation and reduction which have prevailed in meteorological ofices, rather than upon the unreality of the phenomena in nature. At present the di6culties of the research are diminishing by reason of two improvements; ( 1) a better knowledge of where to make the comparison, and (2 ) the gradual acqnisition of reliable secular data. Thus, the prominence data are super- seding the sun-spot numbers, and it has now become compara- tively easy to traverse the magnetic and the meteorological fields with our improved stanclard curve of comparison, and to bring out the fundamental typical synchronism in nearly erery series of observations, so far as the annual means are concerned. The importance of emancipating this subject from the pre- vailing skepticism is evidently in the interests of advancing cosmical science. If we can prove that other forces than the Newtonian gravitation and radiation are interacting between the sun and the earth, it becomes a conclusion of vital interest to astronomers. As an example of the present state of opinion, me note Prof. Simon Newcomb's address" before the Astro- nomical and Astrophysical Society of America on December 39, 1902, in which he sags: The conclusion is that spots on the 311n and inagiletic storins are due to the same cause. This cause can not be any change in the ordinary radiation of the bun, because the bert r~cords oP the teiuperature show that, to whatever variations the sun's radiation may be sul)jectecl, they do not change in the period of the sun spots. , We shall, on the other hand, show in this paper that ter- restrial temperatures do, as a whole, change with the varia- tions of the solar prominences, and this will tend to modify Professor Newcomb's inference. The question whether the connection is direct or indirect, by a magnetic field or by some special action of radiation, is to be decided finally by an appeal to the observations. Dr. J. Hann writes in his Lehrbuch der Meteorologie, pages 636, 627: These can lead to the discovery of the period, IJUt it is very difficult to find the true length of the Iieriod, sinre the auiplitucle of the veiiation of the meteorological elements within the period ih not x ery great, becau.;e so many other influenves are preqent. w1iit.h stand in the way of deriving more accurate mean values out of long intervals of time. As yet no one has succeeded in surely deducing for any one meteorological element a cyclic variation of consitlerable aniplitude. These efforts have been applied to variations of tempera- ture, clouds, rainfall, thunderstorms, hail, barometric pres- sures, cyclones, and winds, especially with the view of finding an 11-year period synchronous with that of the sun spots. It should be noted that a shorter periocl, of about three years, is probably the better period of synchronism to he studied. Also, synchronous movements need not be truly periodic. Indeed, there may be true correspondence with very irregular and aperiodic changes. It is easier to coiinect loosely constructed variations in the prominences of about three or four years duration with terrestrial variations than to establish synchro- nism in the 11-year sun-spot period. Dr. A. Sprung, in his Lehrbuch der Meteorologie, pages 366, 367, writes: __-- l7 Science, January 23, 1003. Therefore, a connection between the sun-spot frequency and the changes in our atmosphere can not well be denied. It is probable that the pe- riodic changes in the atmosphere are not caused directly through the sun spots, but that both phenomena are brought about through one common or by several interacting causes, whereby a displacement of the periods relative to one another becomes possible. Prof. Cleveland Abbe has frequently expressed in the MONTHLY WEATHER REVIEW a very doubtful view regarding the advisability of such researches, with the object of discouraging further efforts to unravel the solar-terrestrial net. Thus, in the MONTHLY WEATHER REVIEW for June, 1901, page 264, he writes: d s the periodicities in sun spots, t,he width of the spectrum lines, the magnetic and auroral ihenomena are sufficiently well marked to he satis- fal:t.cirilg ilemonst,rable, while c~)rrespouding variations in pressure, tem- perature, wind. and rainfall are small, elusive, and debatable, we must caution our readers aexinst being carried away by optimistic promises. It is certainly iinl'ressive to the t,lioughtful mind t,o realize that there is even a slight conneetion bet,weeii solar and terrestrial phenomena, but the delicacy of t,liis connection is such that it still remains true that t.he study of meteorology is essentially the study of the earth's atmosphere as actetl upon 11s a cotiHtmt source of heat froin the sun. None of these astrophysical studies should tempt the meteorologist to wander far from t,he study of the dynamics of the earth's at,niosphere and the effectti of the oceans and contiuent,s that diveysify the earth's surface. I I I I FIc. 5.-Solar and terrestrial synchronism. We have, nevertheless, merely to recall the works of many scientists in order to realize how strong a hold this problem has upon the astrophysical meteorologist: Herschel, 1800; Chitier, 1844; Fritsch, 1854; Arago, 1855; Zimmermann, 1856; Wolf, 1859; Meldrum, 1870; Koeppen, 1873; Hill, 1880; NOVEMBER, 1903. MONTHLY WEATmR REVIEW. 511 van Bebber, 1882; Blanford, 1889; Brucliner, 1890; Locliyer, 1898; Carrington, Spoerer, Wolfer, and many others. The number of students who are taking up the problems of COS- mica1 meteorology is rapidly increasing, and this shows that there is encouragement for such work. The present paper continues the discussion of an inresti- gation first published in 1894," which brought out the fact that there is a synchronous variation in short cycles of about three years duration superposed upon the 11-year sun-spot periocl. In Bulletin No. 21, Solar am1 Terrestrial Magnetism. page 127, i t was said: A comparison of the mean American meteorological ciirve 1% itli the European magnetic curve certainly shows confnrmity to snch an extent as to exclude merely atacidental physic-a1 relations. Should sucli a reqult be obtained also in the future, i t will Ibe a demonstration o f the syn'~1iru- nism of the two sghtems of forces untler c~on~iileration. t20 0 - 20 -10 t 10 0 -ro tzo 0 -a tzo 0 -10 tm 0 -a cm 0 -20 I FIG. 6.--Variatious of the aiinual prehsiire in the direct t >p . Since that time advances have been made as follows: The magnetic curve has been extended from 1811 to 1900; the barometric pressures of the United States have been re- duced to a homogeneous systeni; the curves of prominence frequency on the sun have been computed by Lockyer and in- dependently by myself; the variations of the prominences have been closely associated with the changes in the angular veloc- ity of the solar surface rotations in iliflerent zones, especially in the polar latitudes; the type of internal circulation neces- sary to produce this polar retardation, and to transform the solar mass into a polarized magnetic qhere, has been indicated. I I I t I I I -20 0 t 30 -zo 0 +20 -20 0 t 20 -20 0 t 20 -20 0 tM Frc:. 'i.-Variations o f the annual pressure in the inverse type. 512 MONTHLY WEATHER REVIEW. NOVEMBER, 1903 of the earth. These have a variation in direct synchronism with the prominences, in certain parts of the earth, but under special conditions of orography the synchronism is of the i ~- wrse type. This chain of evidence is strong enough to incliice conticlence in regard to the fact that this solar-terrestrial phys- ical synchronism really exists. THE UNSATISFACTORY STATE OF THE OBSERVATIONAL DATA. The two prevailing clificulties in extracting suit,aLle &ita from the published reports of meteorological observatories, and reclucing them to a homogeneous system, are the ~iiinier- 011s changes in the elevation of the lmrometers, and in the very different hours of making the observatioas. Without the espencliture of labor entirely beyond the capacity of a siiigle office to bestow upon the task, when the research for synchro- nism is extended to the entire earth, it has been necessnry to t 20 0 - &% FIG. x.-lTati:ttions of tlir annual prehsure in the inilifferent t j lie. use some simple devices for the sake of arriving at approxi- mately hon;ogeneons residuals. The work for the United States is complete for the pressures. and is in progrebs for the tein- peratures. By inspecting my Barometry Reportly it i< easy to see the reason for the necessity of the reduction. I n order to give some idea of the state of the data in other countries, we note the following with respect to the barometric prehsiires : For Russia-Siberia, several stations changed elei ation more than once. India, there are ~i~imertms changes of e1e.r ation. south Africa, numerous changes of eleration, ant1 alho of New South Wales, the monthly means of observations alone the hours of observation. 'SReport of the Chief of the Weather Bureau, l ' J U U -l D c l l , Vol. 11. v - mere published. These hac1 to be collected before the annual means could be computed. Argentina, the monthly means of observations alone were pub- lished, and these also had to be collected before tlie annual means could be coiaputetl. Iceland and Greenland, very few changes in elevation, but not long records. I n general all the annual pressure curves were plotted, and a metin pressure ant1 iiormal graclient, were determined, from which the aiuplitucle variations were taken of€ as residuals. Since our purpohe was simply to secure the most probable aiiiiual residuals this gr:tphic method was substituted for the exact coinpiitations which ought to be made. Frequently the secular gradient slope m as so prominent throughout the series for a single station as to suggest x grac1u:d change in the cor- rection of the Laroiueter relative to a norinal standard. With respect to the teiiiperat,ures, the annual means were extractetl froin tlie repnrts, anel the iiieaii values for the several series were COlUlJUted, so far :is they mere apparently homo- geneous, mil from these the residluals mere foriueil. As the cosuiical aiinual \ariation of temperature is only 1" to 3" F., i t \\:is often 1)ossible to brenk 111) a long series a t the same station iuto hoinogeneous sections: hut tliis was done cau- tiously, ant1 only after clear erideiice of a discontinuity in the local conditions. The great clificulty with the temperature dnta conbibts in the niimeroiis hours of observntion that have lwei1 adoptecl. or in the nuiiierous selected groups of hours froiii which the iiieans were derived. Many of these differ- ences arose froni artificial attempts to obtain an approximately correct 24-hour mean, to which in fact all meteorological data sliould lie T ery c*arefully reduced. Some of the combinations of hours nsed are as follows: United States. W :d h &m iiiean time, 7:35, 1:33, 11 :85; 73.5, 435, 1l:OO; 7, 3, 11. Sexenty-fifth meridian time. 7, 3, 11: 7, 3, 10; 8. 6 : iuaaimum, ininiiiiuiu. The stations have quite short records. Nen South Tt'ales, 9 a. in.; 9, 3, 9 ; Iuaximum, miiiinium. South Australia, 9, 3, 9; 9, 12, 3. 6, 9; masiiuum, minimum. West Australia. 9, 3; 9, 12, 3; 9 a. in.; G , 6; inasimum, mini- Ocean Idanils. hourly; 9, 3, 9, minilnuin; 6, 9, 1, 3 , 3:.56. C'hina, hourly: 10, 4, 10. India, S. 10, 4 ; 10, 4; &hourly, or 10, 4, 10, 4; 9:30, 3:M; Russia-Siberia, 7, 1, 9; 7 , 3 , 9: 9, 13. 9; S, 1, 9; hourly. Europe, 7. 3, !). !); 7:45, H: 6. 2, lo; :I -h (~~r l y ; masimum, niinimuin; 7, 10, 1: 4, 7, 11; 7, 1, 7 ; 6, 9, 12; 8. 6, 9; 6, 12, 9; IllUlll . Jtzl)a11, 9:30, 3:30, 9:30; 4 -h 0 ~~l y , 01' 2, G , 10, 3, (;, 10. 9, 4; 1O:YO. 3 :3 0 : masiiniiin, iiiiiiiinum. 11ourly. Azores-RIacleira, 9, 3, 9. North ilfrica, 7 , 3, 9; 7, 11. 2, 5 : 7, 1, G ; 9, 3, 9. South Africa. 6. 12, 6; f i , 3 ; 9, 9; S, S; S a. 111. South America, 7. 2, 9; hourly. Icelan~l-C:reenland, H, 2. 9. Froin tiucli an exhibit it is no wontler that meteorology has not yet coutributecl its proper share to :tccurate cosniical phys- ics. I t ih iieedless to recount the reason for this state of afiairs, lait only to urge as speedy a remedy :is is possible. It might be argnecl that no rebnlts (*an be derived from such data; but tliih is not true. as a stncly of the resitluals sumniarizecl in this paper aiiipl? co~~firiiis. It is, perhap, surprising that valuable results can be extmctecl from the data, ~n c l this only proves how iiiiportant such work might be made if suificient care were exercised in selecting the hours of oherration, and estab- lishing rigorous niethods of reduction. It frequently happens that a t a xiveil station the sniiie hours continue to be used for many yearh. se) that in effect its own rehiduals are nearly homo- geneous. The iiieans of the iTarious combinations of selected hours generally approximate a true 24-hour mean, SO that on tlie whole there is something like honiogeneitg in the cliffer- NOVEMBER, 1903. MONTHLY WEATHER REVIEW. 513 I t I I I I I 4 .0 0 -z.o +z.o 0 -Z.0 I I I I I I I I I I I I I I CZAl 0 -2.0 f I.0 0 --LO +LO 0 -LO +LO 0 -Lo LZ. 0 0 -2.0 L1.0 0 -z.0 I I I I LZ.0 0 -2.0 -r.o 0 -1.0 -2.0 0 ent changes. The fact that residuals synchronous with solar variations actually siirviw, is a satisfactory e\ iclence that the causes producing them are wliir and not local terrestrial. ihle to print iii the DIOHTHLI. TS’E:imER REI IEM the table of residuals for encli station, ant1 we iiiiist coiifiiie ourselves to the curves representing the inem resiclunls for a group of stations, the 1iunil)er 1)eing entered in counechtioii with the name of the country. Thu.;, for New Snnth MT:des the pressure curve, fig. G, was cleterniinecl from si1 stations, Albany, Brzthurst, Deniliqniii, C+oulborn, Newcastle, Sytlney. RE\ITLT\ UF THE OBroadly, the region around the Indian Ocean gives direct synchronism, South America and North America give inverse synchronism, while Europe and Siberia give an indif- ferent type. Greenlancl and Iceland seem to have direct type NOVI~XBEB, 1903. MONTHLY WEATHER REVIEW. 616 FIG. 12.-Distribution o€ the pressure types. like the Indian Ocean. Fig. 13, ‘( Distribution of the tempera- ture types,” shows that there is synchronism of the direct type for the Indian Ocean, Africa, South America, the West Inclies, and the Pacific islands generally-that is to say, throughout the Tropical Zone. The inverse or the indifferent types pre- vail in Asia, Europe, and North America generally-that is, throughout the North Temperate Zone. Taking the earth as a whole, the temperatures spchronize di- rectly with the solar energy in the Tropical Zone, and inrersely in the temperate zones. The indifferent type prevails in the plateau districts of the continental areas, probably because t,he solar type is there so broken up by the local climatic conditions as to practically obscure the synchronism. I n the pressures the Eastern Hemiqihere tends to direct synchronism, except in Europe and Russia, where the indiEerent type prevails, and the Western Hemisphere to the inverse type. It may not be practicable to explain all that t,his means, but apparently we are dealing with the complicatioa caused by superposing an atmosphere in circulation upon the unequally heated surface of the earth. The surging of the atmosphere as a whole from one hemisphere to the other, or from the continents to the oceans, is concerned in producing these eflects. The trend of the great mountain systems strongly differentiates the circu- lat,ion of the lower strata. Thus, the Himalaya Mountains, running east and west,, check the flow of air from the Tropics to the Asiatic Continent, while the Rocky Mountains and the 516 MONTHLY WEATHER REVIEW. NOVEMBER, 1903 Andes system favor the flow along the meridians, especially in the United States. As a result, the number of cyclones crossing the United States is many times the number crossing Siberia, which is in fact singularly deficient in cyclones. South America shows a similar defect in circulation, because it lies too near the Tropical Zone. The United States is covered by an active circulation be- tween the Tropics and the north Polar regions, Siberia by a stag- iiaiit atmosphere, and Europe generally by a mixed and in- different circulation, since the American cyclones tend to break up upon the territory of Europe after crossing the Atlantic Ocean. Hence, the region about the Indian Ocean is favor- able for detecting direct synchronisms of pressure and tem- perature with the solar prominences by reason of its quiescent atmosphere, and the United States is well placed to respond to an iuverse synchronism, by reason o f its actmire circulation with a pronounced component from the north Polar regions. Europe does not possess an atmosphere which registers the so- lar and terrestial synchronism i a a very efficient manner. This may account for the fact that the E~iropeaii at,tenipts to estalJlish a definite synchronism have issued generally with negative results. As has already been suggested, too nmch empliasis lias been put upon the failures to make out the connection between the solar and the terrestrial synchronisins. It shoulcl be noted that C'. Norcl~nann?~ and A. Angot" cle- duced for certain tropical stations sm:tll residuals of tenipera- ture which are iri rrrse to the su1i-spot curve, but apparently synchronous. These authors have snioot,litcl their cmves by grouping successive years, ant1 have reached siiiall residuals. Since the synchroiiisin should display the annual variations intact, as given above, i t iuay be questioned nhether any pro- cess for eliminating the ininor deflections from year to J ear is desirable. We also note the important fact that tlie wide amplitudes which are cliaracteristic of the 11-year sun-spot curve, and which it has been chiefly souglit to discover in the ineteoro- logical elements, does not, according to this research, appear at a11 proniinent,ly in the residuals. It is only t,he short period of about three years that displays the solar terrestrial s p - chronism. I am not, a t present. able to indicate what this re- sult. implies in solar physics, hut i t certainly carries with it a change in our method of approaching the eutire problem. THE PROBLEM OF THE CYCLONE. Ry F. J. R. CORI~FI€:II, d.iti 11 h c ~l ~o i f , 1: 1 , Septeiiilicr 5, 1902.1 It was Lord Iielviii who showed that a mass of fluicl in vor- tes inotioii acquires all the properties of a solitl, the chief of which are rigidity and elasticit?. It was on this cleinonstra- tion that he fouiidecl his astonishing vortex tlieory of nintter. He showed perfectly that an atom of matter might possilJly he nothing else than tlie frictionless fluid ether in a vortes st:tte. A ~o r t e s in the ether woulc~ thus possess rigidity, elasticity, inertia, and all other properties of matter. I n the mine way *o The periuilivity of siin spots auil the mriatious o f the iiieitii aiiuunl temper;ttnres of tlie atmoqphwe. M. Charles Noralmann. C'olu1)trs Reiiclus. Paris, *Julie, 191)3. Translation in RIontlily Wwtlirr Ervirkv, August. 19113. P. 371. ?1 The hiultaneouh variations of siin G 1 )o t s null o f terrrstlinl atnlos- pherit. ternperaturra. Prof. Alfred A n p t . Annuairr rlr la SuciCtS. RI6- tt!orologique de France. June. 1903. Tranhlation iu Monthly IVrathrr Review, August, 1903. P. 371. I The Editor l i i i s retainrtl this palier fl)r a year in Iiolws that till. auth,,r x+onltl rlnl~orate the ~iiatlrrmntit~d Iletlul.tiou of thc formultr that l i e , l i u t the Iattrr ha* tlitiuglit lirst t r i bii~iply all11 a few rrfrrr~i~ -~-_________ Major Bnrnanl. The renilrr \r i l l trcate~l in inany nlo(lc~ru works c riro rehts hi* theairy t-ntirrlg i u i nlny deal with tlie cyc4oiir as if it nrrr it rot pitper vf any special intrrrst t l i the stutlrnt (i f results will. it ih hopell, 1r:ttl others tu R niorr ri subjec.t.-ED. a vortes in the atmosphere acquires shape and preserves it like any solid, as well as rigidity and elasticity. Professor Tait's smoke rings, which suggested to Lord Kelvin his ethereal vortes atoms, have all the properties of solid bodies. So. when I treat a revolving mass of air as being dynamically the same as a solid I do what Lord Kelvin has shown is perfectly ad- missible.' Poisson's general equations for rotary motion of a solid having one fisetl point O are gireii in most wc)rks on mechanics alld read as follo\vs: In these ecliiatioiis (I . v , ani1 U J are the angular velocities of rot:ttioii of a solid lmly with referelwe to tlie three coordinate a w h , X, 1 , :a i d Z. fixed in space and intersecting at the fixed point 0. A, I;, a i d (' are the in(ments uf inertia of the solid ninsib with referelice to its own tliree principal ases, the latter being in niotioii relative to the three fixed axes. L, ,VI, and LV are the moiueiits of tlle accelerating forces that n u t upon the body froin wit81iont taken n itli referellce to the tllree principal If we ;apply these ecluation< to a spiiiietricd solid of revo- lution. sucli as a ring, or an ellilmid (J f relolution llaviiig its fixed 1)oint in its axis of figure, then we oljtain the equations for the moTeiueiit of ii gyroscope or rutascope, or a top, and l i e are able to explain a11 the motions of those bodies with reference tu the support on n1iic.h they stand. If, however, instead of supposing the revolving bocly to have a fixed poiiit, ~e give the latter also a clefinite motion, BS. for instance, when the gyroscope, with its siiplmrt, is carried with the eartli around the earth's axis in its diurnal rotation, we can tlien deduce the moveiiient of the gyrowope with reference to the mericlian of the locality. If the disk of the gyroscope be snpposed to be horizontal, or nearly bo. and revolving rapidly :about an asis th:tt is rer- tical, or iiearly so, nncl if its axis is not constrained, but free to mole on the earth's surface, we have a case apparently analogous to tlir movement of a cyclone or hurricane, at least in so far as the latter consists of a mass of air rotating in a horizontal plane. Practically the air within a cyclone is known to be either ascending or descending and climigiiig coatinu- ally, so that energy is brought into i t from without and car- ried outward from it. If tlir energies thus nclded and lost couiiterhnlnuce eavh nther, me iiiay perhaps hope t o deduce from the laws of the gyroscopic inotioii of a solid swue insight into tlie laws of the iuotion of tlie hurricane :along tlie earth's surfwe. The aboi-e general equations o f rotation were in lS5X put into a convenient form fur the study of the gyroscope by Major, afterwarcls (+eneral, J. (3. Barnard, of the Army Engi- neers, :m c l his p p e r is reprinted :as No. 90 of Tali Nostrancl's Science Series.' In RIajor Baraarcl's little volume the reader will find decluced from fuiiclmnentnl principles the law of gyroscopic niotion. which is this: If a spinning gyroscope or a spinning wheel be tnriietl about :~ii asis perpendicular to its onn axis of rotation, a cletiectile force will be developed per- axes. ~_____ __ -~ 1 Ic'e11 n1,alldI lllrll.-EIl. J. C;. Bamaic!. d u n l j s i s of E4 Italy IIotioii its Alq)lirtl t l ) thr Ggrowolw. By Major D. Van Nobtrand, pul~li~her, New Tork, 1887.