Page images



magnetic survey of the oceans under the direction of the Department of Research in Terrestrial Magnetism of the Carnegie Institution of Washington, left Brooklyn, New York, on her first cruise, August 21, and proceeded direct to Gardiner's Bay, Long Island. Here several complete swings of the vessel were made with both helms in order to test the instruments, train the observers, and, above all, to determine whether actual non-magnetic conditions had been secured at the various positions of the instruments. These tests resulted most satisfactorily, not only proving the absence of deviations in the three magnetic elements (declination, dip, and intensity), within the errors of observation, at all observation positions, but also showing that with the instruments installed and the methods of observation employed a high degree of accuracy can be obtained. Summary of Swings at Gardiner's Bay, Long Island,

New York, August 31 to September 2, 1909.

[ocr errors]
[blocks in formation]

Hamilton observations of September 12, 13, and 14, that perihelion should occur at 1910 April 18.63 G.M.T., and that the nearest approach to the earth should take place on 1910 May 19 at a distance of about 0.14, i.e. about 13,000,000 miles. Further, he points out that on May 18.14d. (G.M.T.) the earth and comet will be in heliocentric conjunction in longitude, the longitude being 230° 48'; the heliocentric latitude of the comet will then be -7', so that, according to the present eleinents, no actual transit of the comet over the sun's disc will occur, but a slight change in the elements might produce one. “ At any rate,

he says, “ it seems highly probable that we shall on May 18 be inside the tail."

In a communication to the Times (October 25), Prof. Newall announces that he observed the comet, visually, with the 25-inch refractor, power 214, on October 21. The magnitude was about 14.0 or 14.5, and the faint nebulous patch had neither stellar nucleus nor definite borders ; the diameter was estimated as 10 or 12 seconds of arc.

At the previous apparations, in 1759 and 1835, the comet was first seen 77 and 102 days, respectively, before perihelion ; Prof. Newall's observation was made about 180 days before the calculated perihelion passage of 1910, but he points out that this does not necessarily mean that the comet is so much brighter at the present apparition, for he would probably not have detected it had he not known its exact position as indicated by the previous photographic observations.

QUANTITATIVE MEASURES OF THE Oxyges BANDS IN THE SPECTRUM OF Mars.-In Bulletin No. 41 of the Lowell Observatory Prof. Very describes the methods by which he measured the relative strength of the B, oxygen, band in the spectrum of Mars, and discusses the results in their relation to the presence of, and quantity of, oxygen in the planet's atmosphere.

As Prof. Very points out, the B band is normally so intense, by the absorption in the earth's atinosphere, that only by a method capable of the minutest accuracy could it be expected that any slight extra intensification, due to the Martian atmosphere, would become measurable. He claims that, over a long series of measures, his improved spectral-band comparator is capable of measuring this added intensification. Briefly, although a visual examination shows no increase of intensity of B in passing from the spectrum of the moon to that of Mars, the comparator measures are surprisingly concordant in showing a positive value. in favour of Mars, several times greater than the probable error ; B, in the spectrum of the planet, is 15 per cent. stronger than in the lunar spectrum, and the probable error is 1.8 per cent. There is, as would be expected, a considerable variation among the individual measures, but no contradictory results.

STARS HAVING PECULIAR SPECTRA : New Variable Stars. -Harvard College Observatory Circular, No. 143, contains a list of seven stars exhibiting peculiar spectra, and twentyeight stars shown to be variable. For each star the position, for 1900:0, is given, and the class of spectrum indicated, whilst a series of notes summarises the observations. Some of the variables show a long range of magnitude, in one case amounting to 5.0.

Circular No. 151 is a similar publication announcing the discovery of twenty new variable stars in the Harvard map No. 49. It also describes a star in Taurus, at

R.A.= 5h. 43m. 125., dec. = + 19° 2.0', which varies more than five magnitudes, and exhibits a light-curve of the rare R Corona Borealis type. Long periods of normal brightness are followed by sudden diminutions over a wide range, the normal brightness being 10-2 and the minimum fainter than 15.5.

The Natal GOVERNMENT OBSERVATORY.--Mr. Nevill's report for the year 1908 deals chiefly with the meteorological observations, which, with the time service, form the chief work of the observatory, but it is noted that the large equatorial telescope was overhauled and repaired, and some observations were made with it by Mr. Hodgson. Among these was a new series of lunar photographs, for the determination of the real libration, and some sketches of the surface configurations of Mars, Jupiter, and Saturn.

[merged small][merged small][ocr errors][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small]

The above diagram will assist in the interpretation of the figures given in the above table, and at the same time make clear the general arrangements of the various instruments and the methods used. Consider the plane ABCD to be a vertical section through the four instrument positions and the fore and aft line of the vessel. Position A is on the bridge above the deck or chart-house, B is in the forward observatory, C in the after observatory, and D at the middle point of the chart-house, vertically below A. The sides of the parallelogram are about 133 feet.

At A is mounted the standard compass, one of a new type invented and constructed by the Department of Terrestrial Magnetism, and called the “ marine collimating compass”; it will be found described in the March (1909) issue of the journal Terrestrial Magnetism. The basis of

the instrument is a Ritchie 8-inch liquid compass with the stationed at D, who reads the ship's head by a spare card, however, removed, and an optical collimating system Thomson dry compass. The same kind of deflecting device with scale introduced, enabling the observer to note the by which the deflecting magnet is brought at right angles arc of motion of the magnet system while sighting on the to the axis of the detected card, thus admitting of the sun or star, hence knowing precisely to what part of the simple sine-computation formula, is likewise attached to arc the stellar azimuth applies. In all forms of compass the marine collimating compass at A. In the latter case azimuth circles hitherto used, the magnetic azimuth of the the angle between detlected magnetic system and the sun celestial body must be taken from whatever point the card (or true meridian) is read with a pocket sextant, thus in its oscillations to and fro has momentarily reached. In making one entirely independent of the yawing of the brief, practically the same method of observations can ship. now be used at sea as on land, where the magnetometer The declinations obtained at C are intended chiefly as .circle would be set to some convenient point on the magnet some control against any gross blunders which may be scale and then scale readings taken of the positions of the made at A; for a fairly smooth sea they compare favourmagnet during the interval of observations. The angle is ably with those got at A, but in rough seas the great next determined between the circle setting and some mark, superiority of the A results is very evident. or the true meridian, and the declination is finally deduced. in brief, then, the scheme of observations, whenever Similarly with the marine collimator compass. The angle fully carried out, will yield the following determinations in (say, iniddle of scale) between the magnet and some celestial about one hour's time by independent observers, with different body, as the sun, is read with a pocket sextant to the instruments, and at different positions on the vessel : -nearest minute of arc at a given time, and then scale At A.-Declinations (also horizontal intensity when a seadings of magnet and of watch are taken. With the aid celestial body is long enough visible to permit of full sets of the time readings, the motion of the sun during the of deflection observations). interval of observation is taken into account, and the true At B.-Two values of dip by the regular absolute method, azimuths determined, whereas the scale readings give the and two values of total intensity and of dip, using two varying positions of the magnet system.

deflecting distances. With this instrument, therefore, one is almost entirely The horizontal intensity is determined by computation independent of the yawing and rolling of the ship, making from the dip and the total intensity. it possible still to get satisfactory results when with all At C.-Two values of horizontal intensity independently other azimuth circle devices hitherto used at sea observa- with two magnets, and using each time two deflecting tions would be wholly impossible, or at least very uncertain. distances; also, whenever possible, magnetic declinations.

In five to ten minutes a value of the magnetic declina- D, as above described, is simply accessory to C, and does tion is now obtained possessing an accuracy attainable not furnish any direct result. with previous instruments only by most careful observa- It is thus seen that an effective control is obtained for tion and by laborious repetitions extending over a half-hour each magnetic element, and it is for this purpose a great or more to eliminate the motion of the card. Thus not gain indeed that it is now possible to compare at once the only has the accuracy of declinations at sea been in- values of horizontal intensity, for example, got at B and C .creased, but, what is equally important, the time has been without first waiting until the deviations are well deter. reduced and the possibility of getting useful results in all mined, as has hitherto been the case in all previous vessels kinds of seas greatly extended.

engaged in magnetic work-even on the Galilee, which, Furthermore, sufficient attention has not always been before the Carnegie, had the smallest deviations of any paid in previous ocean magnetic work to a proper control ship. of constants. Thus, e.g., with the compass azimuth Besides the great improvement resulting from having a circles, as usually constructed, there are movable parts non-magnetic vessel, and from the perfection of the instrusubject to wear, such as the axes of mirrors or of prisms ments themselves, the conditions and opportunities for and of the azimuth circle on the bowl. The wearing of observing have also been materially bettered the these parts may easily bring about the same effect as Carnegic. Thus the instruments at B and C are under tliough the compass were not mounted in the fore and aft shelter, being mounted in observatories with revolvir.g line, i.e. introduce a quantity A not due necessarily to the domes and movable slides, permitting of both magnetic ship's magnetism, but to instrumental error, which is likely and astronomical observations, with full protection to the to vary with extent of use of azimuth device.' To control observer and instrument from wind and weather. such errors, all instruments were invariably dismounted in Outstanding Difficulties.—These are chiely due the case of the vessel (the Galilee) employed in the Pacific meteorological conditions and the state of the sea. Thus Ocean work, whenever a port was reached, and correspond- in the absence of sun or star no magnetic declinations can ing observations made between land and ship instruments. be obtained, though dip and intensities may be got even With the present instrument, there being no such wearing in a pouring rain, because of the introduction of the paits, there will not be the source of error described. sheltering observatories. It is hoped that some instrument

It may also be pointed out that the effect of drag of based on the gyroscope compass may be soon perfected magnet system moving in the liquid during changes of the having the desired accuracy, with the aid of which the ship's head is overcome in the present instrument, as well direction of the magnetic needle may be referred to an as in the one mounted at C, for the method of observation invariable plane to be controlled whenever a celestial body involves turning the compass bowl opposite to the ship's becomes visible. niction.

Were one to wait for a calm or a fairly smooth sea it At B is the gimbal stand for mounting an L.C. dip would occur quite frequently that no magnetic results circle, as modified for the Galilee work, with which the would be forthcoming. In fact, on the entire recent dip is determined absolutely (i.e. in all positions of circle October passage of the Carnegie from St. John's, Newand needle, inclusive of reversal of polarity by an electric foundland, to Falmouth, England, there was not a single coil) with two regular dip needles and again by the method day which would fall in the usual category of favourable of deflections, and total intensity is obtained using two days of observation, but, instead, on nearly every day there deflecting distances.

was a gale, the sea was rough, the vessel yawed through At C is a Ritchie liquid compass provided with an angle of 10° or more, and rolled through an arc of 20° improved azimuth circle and a deflecting attachment, both to 30° and more, and vet observations were secured on designed and made by the Department of Terrestrial every day except one. That utilisable magnetic results have Magnetism. With this instrument declinations are obtained, still resulted under such very adverse conditions is due to and also the horizontal intensity of the earth's magnetic the perfection of the instruments, the cutting down of 'force, by the method of deflections independently with two time required for observations to get a desired degree of deflecting magnets, the magnetic moments of which are accuracy, thus reducing to a minimum the condition of controlled from time to time by shore observations when- stcadiness of ship, and, of course, to the skill of the ever the vessel is in port, and using two deflecting observers. Still further improvements are being striven distances. In these deflection observations the yawing of for with regard to independence of steadiness of ship. the ship, or the changes in the lubber-line from which the It is thus seen that while the endeavour is steadfastly deflection angles are counted, is controlled by the recorder held in mind to measure the magnetic elements at





with every possible precision, the improvements made, and ' vessel. The installation is in the charge of the chief yet to be made, are along the line of reduction in time engineer, Mr. D. F. Smith, who has had both theoretical required to achieve the desired result-in other words, and practical experience with gas engines. Mr. Carl towards simplicity. With proper instrumental means and Smith, an acknowledged gas-engine expert of the technomethods it need not require any more time to make logical branch of the United States Geological Survey, is accurate observations than to get indifferent results with consulting expert in connection with this work.

Should instruments not adapted to the conditions to be met. this type of engine be made a practical success for marine

As a proof of this is given below the table of results 'purposes, even though it be but for auxiliary uses, it will of the work on board the Carnegie from September 1 to be a valuable achievement, being the most economical form October 18, or in about six weeks' time. After the tests of motor now employed. in Gardiner's Bay referred to above, the vessel proceeded, The Magnetic Results.—In submitting now the magnetic under the command of Wm. J. Peters, to New London, results obtained to date, it should be stated that the comConnecticut, to have some slight alterations made. She putation and revision of results goes on apace on board with left this port on September 11 bound for St. John's, the observational work, not only that possible errors may arriving there September 25. Here the director rejoined quickly be detected, but also that the results may be made her, having left the vessel in Long Island Sound to attend known promptly. It is the intention, hereafter, to publish the meeting of the British Association at Winnipeg, the results at intervals of about three months. Canada.

Accuracy of Magnetic Results.-In general, the declinaAfter the necessary shore observations, the Carnegie set tions given may be taken as correct within 0.1°, and only sail from St. John's on October 2 direct for Falmouth, in but a few instances, when the conditions of sea were England, arriving there on October 14, having had favour- unusually bad, will the error be 0.2°; under favourable able winds. Owing to the lateness of the season the conditions the observation error in declination with the Hudson Bay cruise for this year had to be abandoned. marine collimating compass for the mean result will be


Deck view of the Carnegie, showing the chart hưuse with Marine Collimator (A in diagram) in the middle and the two observato ies.

The forward one is uncovered, and inside it position B of the diagram, where the L.C. Dip Circle is mounted ; inside the after
observatory (the covered one) is mounted the Deflector and Compass (Position C).

After the completion of the harbour and shore observations about +0.05° (see the Gardiner's Bay results). About the at Falmouth, the Carnegie, continuing under the command ' same statement as made for the declination applies to the of Mr. Peters, will proceed to Madeira, and will return dip. The horizontal intensities, as at present given, may early next year to New York via Bermuda, the director be assumed correct within 1 unit in the third decimal for returning to Washington.

the severe conditions encountered on the greater part of Personnel.-Besides the director of the department and the cruise thus far ; for fairly good conditions of the sea the commander of the vessel, the scientific and navigation the error need not be more than 5 in the fourth decimal, staff is composed as follows :-C. E. Littlefield, sailing and may be made less, as shown by the Gardiner's Bay master; J. P. Ault, E. Kidson, R. R. Tafel, observers; work. Dr. C. C. Craft, surgeon and observer ; and D. F. Smith, In making these preliminary statements, it should be chief engineer. The ship's personnel includes, furthermore, remarked that every possible source of error is considered two watch officers, two cooks, eight seamen, and one in other words, absolute accuracy, not relative accuracy, mechanic.

is meant. For example, when it is declared that, given Auxiliary Propulsion.--Of considerable interest in marine : fairly good conditions, it is possible to get the dip on circles, aside from the magnetic work on board this vessel, board the Carnegie within 0.05°, i.e, within three minutes are the experiments being made in the perfecting of a absolutely, this means more than may at first appear. producer-gas engine for marine propulsion. Such a plant- Thus dip circles-especially ship dip circles-have instruand an almost entirely non-magnetic one-is aboard the mental corrections exceeding frequently the relative error Carnegie, of 150 horse-power, sufficient to drive her at six of observation. Accordingly, the dip circle on the Carnegie knots' speed in calm weather. This engine has already has been compared, not only with various observatory dip proved a useful adjunct to the vessel's equipment, facili- circles, but also with an earth inductor at Washington, for tating the entering and leaving of ports, and such tests as the range of dip from +88° to -60°. Further control will were made at Gardiner's Bay requiring swinging of the be had at the various ports of call during the progress of

the work. This serves as an illustration of the care re- Corrections of the Present Magnetic Charts.-The table quired and being taken, not only with this form of magnetic requires no explanation other than is already given in the instrument, but with every instrument used aboard.

headings, with the exception of the last three columns, We have preferred to underestimate our absolute accuracy which exhibit the quantities to be added to or subtracted from rather than to overestimate it. In any case, it may be the declinations scaled from the lines of equal magnetic said that the magnetic elements are now being obtained on declination (“ variation of the compass ''), first, as given the Carnegie with sufficient accuracy, not only for prac

by the British Admiralty chart, 1907; nest, by the United tical demands, but also for purely scientific ones.

States Hydrographic Office for 1910; and, lastly, by the Were we able to choose the time to observe and wait Deutsche Seewarte for 1905.' in all cases the values for fairly smooth sea, the magnetic elements could be scaled from the charts have been referred to the present determined at sea with an accuracy practically the same time with the aid of the secular variation corrections as as in the determinations for land magnetic surveys.

given on the respective charts. It will be seen that, in Diurnal Variation Corrections.-No corrections


the case of each chart, the corrections are usually less than diurnal variation need, in general, be applied. The attempt 1°, and, considering the miscellaneous data at the disposal is being made to get the magnetic elements at such times of the makers of the charts and the uncertainty of the of the day when these corrections are small or are of the secular variation reductions, the general correctness of the order of the error of observations. Thus, for example, the charts is most gratifying. most favourable condition for the declination work is However, one fact, revealed by the prevalence in sign when the sun is low, i.e. early in the morning or late in of the quantities appears to be of sufficient practical the afternoon, and at these times the diurnal variation importance to require attention. The data given in the corrections are small and frequently of opposite sign. table apply pretty closely to the tracks followed by the Should there be evidence of magnetic disturbance during Transatlantic steamers between New York and England. the observations, or as may appear later from observa- Along the portion from New York to a point somewhat tory records, the observations, if necessary, will be beyond Sable Island the corrections for each chart are rejected."

positive, amounting in the maximum to 1°; they then Geographic Positions.-Equal care is bestowed upon the change sign, with eastwardly progression, reaching a determination of the geographic positions of the points negative value of nearly 1°. In other words, for the track where the magnetic observations are made. The astro- pursued by the Atlantic liners from England to a point nomical observations and computations are made in dupli- off Sable Island, the present magnetic charts, in general, cate, and at times in triplicate, by the observers, and thus show too large westerly declination, whereas on the rethe positions are effectively checked. Six well-tested mainder of the route to New York the charts give too small chronometers are carried aboard. With the methods westerly declination. followed, it would appear that the errors in the final posi- It can readily be seen what the effect of these systematic tions assigned will, in general, be less than three minutes errors of the charts would be on the course of a vessel of arc in latitude and in longitude.

sailing from England to New York if, during the entire Magnetic Results obtained on the Carnegie," September i

passage, no sun or stars were visible, as sometimes occurs, to October 18, 1906, in the North Atlantic Ocean.

so that the course of the vessel would have to be shaped

entirely by the compass and the log. At the end of the Corrections bf 2000 miles of the great circle route the vessel, off Sable

Variation Charts Island, would be about thirty miles too far north of her No.

regular track, and if the set of the current were in the

same direction the vessel would easily be exposed to shipBritish U.S.


wreck. From Sable Island to New York the effect of the

chart errors, being reversed in sign to what they were 41'1 Sept. 1

before, would be in the opposite direction, i.e. the vessel 72'1 0133


+0.5 + 2



+06 would be put out of her true course in a southerly direc40'9 70'4

-F08 to'g +07

tion. 694

0*185 +0'22 +0'4 to'4 For a vessel going eastward the effects are just reversed; 4009

13'9 71'9 0182 +0.8 +04 to'z 40'9

hence on the course from New York to Sable Island the to'8

+03 413 162 7199

to'3 to'o

tendency of the chart error would be to set the vessel to o'177

to'y to's the northward, hence again towards the source of danger : 612

0176 +13

+0'6 +06 608

thereafter the vessel would be set to the southward of her 72'5


too to'4 589

outlined course. 237 72'7 o'171 +15 +07 +13

So that, if proper allowance were not 55'7 0*169

made, a captain would have his vessel turned off towards 47'3 73'5 0'158

Sable Island or Cape Race, whether he came from the 29 75 0'159

east or from the west. 51'4 Oct.


to'g 16

735 o'157

It appears to be known to some captains, at least, that 434 48'0 4

to 4 +03

there is some such systematic change which, judging from 4895 477 73'0 O 161

+o't +1'i conversations had with them by one of the present writers 487 4695

72'5 o'161


(Bauer), during various Atlantic voyages, is ascribed by them

to a systematic change in their ship deviations. Two cap503 32'1

707 o'171

-04 tains have told him, independently of each other, that after 50-6 288


-07 24 24'o

their compasses had been carefully adjusted they have noticed -06

-09 o'174

bv repeated observation a systematic change in the deviations, 26 505


which reversed in sign when Sable Island was passed. 50'3 17'2


The effects would be opposite for the eastward and the 49'9

09185 -07


westward cruises. The deviations referred to by these

to'y 49'5 75


captains are the differences between chart values and those 50 175 +01

observed with their adjusted standard compasses; hence 50'I 66'5 0'187 to'2 +0'2

they are not pure deviations, but are the sums of ship No. 1 in Gardiner's Bay; No. 14 at St. John's, Newfoundland ; No. 32 in


deviation and chart error. The work of the Carnegie has Falmouth Bay.

now proved that the cause of the systematic change and 1. Magnetic Storms and Northern Lights.-It so happened on board the

reversal of ship deviations encountered by these painstaking Carnegie that po magnetic observations were in progress during the severe

captains was due, in part at least, to systematic errors in portions of the nagnetic storms of September 25 and October 18-19. No the variation charts. Greenwich time signals could be obtained at St. John's from Heart's Content The corrections for the chart of the lines of equal mason the morning of September 25 on account of heavy earth currents. Northern lights were seen becween 8 and 10.30, local apparent time, of the

netic dip (British Admiralty Chart of 1907 and the Deutsche evening of September 21 in latitude 43' N. and longitude 60 W. of Green.

Seewarte of 1905) are generally less than }', being somewich. A glow of white light with occasional streamers extended upward timos plus and sometimes negative. about 10° above the horizon from a low bank of cumulo stratus clouds. The display was in the magnetic north.

The lines of equal horizontal magnetic intensity as laid The evenings of September 22, 23, 24, and 25 were cloudy or foggy.

I The latest chart of the Deutsche Seewarte is not at present to hand. NO. 2087, VOL. 81]

Hor. Int.

[ocr errors]
[ocr errors]
[ocr errors]







63'9 68.4 664 611



[ocr errors]

42'0 42'5

[ocr errors]

9 IO

14 120


2 20'2
21 2008
3 304




428 438 45'5


72 7


13 14 25


[blocks in formation]

476 478 482

o'o -O'2


[merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][ocr errors][subsumed][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][merged small][merged small]


down on the British Admiralty Chart for 1907 are in error obtained by Mr. Duddell; some of these are published in by amounts reaching 16 units in the third decimal C.G.S., the Journal of the Institution of Electrical Engineers (vol. and those of the Deutsche Seewarte for 1902 (the most xxxix., No. 186, 1907, pp. 545-6). They were obtained recent chart not being available to us just now) require by speaking into a solid-back telephone transmitter, corrections running up to 12 units in the third decimal and transmitting the periodic current obtained thereby on the values obtained by the Carnegie. A part of this through the high-frequency electrical oscillograph invented error is, of course, due to secular variation, but the major by Mr. Duddell. The regularity of these curves is very portion appears to be due to defective data. For both striking; a perfectly definite curve corresponds to each charts there are five negative corrections, amounting in the vowel sound, and with some practice it would be possible maximum to 4 units in the third decimal, four zero values to decipher a telephone message by inspecting the curve and thirteen positive ones for the “ B.A." chart, and two whichi corresponds to the telephone current. It may be zero and fifteen positive corrections for the Seewarte mentioned, in passing, that the analysis of the sounds chart. On the average for the Atlantic, from Long Island which constitute speech is now acquiring commercial to Falmouth, the chart values are too low by about 1/45th importance in connection with the problem of telephone part of the value of the force, even at times amount- transmission. ing to 10 per cent. to 20 per cent. In the Pacific Ocean Mr. Bowron has constructed an acoustic oscillograph the intensity charts gave, in general, too high values by intended to show the vibrations comprised in ordinary about 1'25th part.

speech; but although this gives some interesting informa

tion, the curves obtained are not nearly so good as those Since the above was written the Carnegie has been obtained by Mr. Duddell. The arrangement used is someswung in Falmouth Bay, observations of the three what like that due to Dr. Erskine-Murray, but the slack magnetic elements (declination, dip, and horizontal in- membrane is replaced by a ferrotype telephone diaphragm. tensity) being made on eight equidistant headings. The In practice it is found that Mr. Bowron's oscillograph does mean results of the entire swing are in excellent agreement not give good curves for sounds of low pitch ; this is due with what would be deduced from Rücker and Thorpe's to the fairly high frequency of the natural vibrations of magnetic charts of the British Isles applying secular varia- the diaphragm. It has been pointed out by Mr. A. tion corrections deduced from the records of the Campbell (Jour. Inst. Elec. Engineers, vol. xxxix., No. Falmouth Magnetic Observatory. The declinations and 186, p. 533) that even a microphone transmitter has certain dips agree within two or three minutes, and the horizontal definite free periods, and that sounds of the corresponding intensity within 1/2000th part. This again proves that pitches are greatly reinforced. Mr. Campbell's experithere are no deviation corrections of whatever nature to ments in proof of this are worthy of mention here. A be applied to the Carnegie results; also that the instru- solid-back microphone is put in circuit with a battery of mental constants have been well determined. We have 6 volts or 8 volts and the primary circuit of a fair-sized here also a satisfactory proof that if the distribution of spark coil. The secondary of this coil is connected to a magnetism is uniform, as appears to be the case here, the reflecting electrostatic voltmeter reading up to 10 volts. sea values, upon careful measurement, will be found in The deflections of the volumeter afford a sensitive indicaagreement with the shore values.

tion of sounds received by the microphone. If the nearly L. A. Bac'ER. pure note of 1 stopped organ-pipe be sounded, a deflection W. J. PETERS. of the voltmeter is produced, and if the pitch of the note

be gradually raised the spot of light rushes off the scale

when the note attains certain definite frequencies. This THE ANALYSIS OF SOUNDS USED IN

arrangement might be used with advantage in cases where SPEECH.

measurements of the intensities of sounds are required. THE characteristics of the simpler sounds, which form

There can be little doubt that much remains to be done the elements of speech have been studied by many

in connection with the analysis of the sounds used in physicists. At first, attempts were made to reproduce the

speech. Thus the curve obtained by Mr. Duddell for the vowel sounds synthetically; Helmholtz achieved noticeable

sound in “coo" scarcely differs from a sine-wave successes in this endeavour, but by themselves synthetical

a slight alteration of the mouth introduced the methods can never be quite satisfactory, since it is neces

octave of the fundamental vibration. According to the to prove that the ear does not possess properties

accepted theory of vowel sounds, each vowel is distinguished similar to those possessed by the eye; it is well known by the reinforcement of those partials which lie near to that two mixtures of light may produce identical effects

certain definite pitches. The question could be answered on the eye, although the component waves may be quite definitely if we possessed a diaphragn which would vibrate different in the two cases.

indifferently for all pitches; and since the drum of the ear Many attempts have been made to analyse complex does this, it may be hoped that we may be able to realise sounds, but the results obtained up to the present have not

the conditions necessary to the solution of the problem. been quite conclusive. Dr. Erskine-Murray used a thin

Edwin EDSER. membrane stretched slackly over the small end of a conical tube; a light mirror was adjusted so that it was rocked to and fro by the motion of the membrane, and a beam

HEAT TRANSMISSION. of light was reflected from this on to a revolving mirror and thence on to a screen. Such an arrangement is very A PAPER on

heat transmission was read bv Prof.' sensitive for sounds of low pitch, but it is mechanically W. E. Dalby before the Institution of Mechanical unsuitable for analysing sounds of high frequency. Engineers on Friday, October 15. The object of the reHermann, McKendrick, and Bevier have attacked the search was to place before the members a general view same problems by analysing phonograph records, and have

of the work which has been done relating to the transobtained much useful information ; but there is small mission of heat across boiler-heating surfaces, and in reason to suppose that a phonograph record, however good carrying this out more than 500 papers have been read it may be, affords an exact equivalent to human speech. and abstracts of the more important prepared.

For some time past Mr. G. Bowron, of 57 Edgware In a furnace heat is transmitted by three methods Road, has devoted attention to the construction of an simultaneously, viz. radiation, convection, and conduction. oscillograph which projects on a screen a curve represent

It is extremely difficult to analyse the results of experiing the sounds produced by a gramophone; a brief notice ments, so that the heat transmitted by each of these of this oscillograph was given in NATURE of May 21, 1908 methods may be stated separately. Formulæ of simple (vol. lxxviii., p. 69). The curves obtained are

very in

type have been devised to express the results of definite structive, and they appear to possess as much detail as sets of experiments; the application of such formulæ should those obtained

the elaborate method of Prof. be strictly limited to cases in which similar conditions McKendrick ; but the problem of the analysis of the sounds prevail and fall within the range of the original experiused in speech can scarcely be solved in this way.

mients. The best curves representing vowel sounds have been In discussing radiation, the author gives Stefan's law





« PreviousContinue »