two procumbent teeth rest upon a pad or projecting palate which rises from the inner base of the upper incisors, and whose surface is nearly upon the level of the edges of the upper teeth themselves; the lower incisors, therefore, are only brought into contact with the upper incisors by protruding the jaw forward. I have, moreover, examined many specimens of the Macropus major, or kangaroo, and of varieties of the Halmaturus known as wallabies and pademelons, when they have been mortally wounded and under the influence of the spasmodic muscular contractions which occur at the point of death, and I have repeatedly found that they will alternately open the two incisors to their full extent, and unite them again with the energy which characterises all the muscular movements of an animal in its death-struggle.

If a small object, for instance the blade of a knife, is inserted between the teeth when fully extended, the animal will immediately grasp it with its incisors, which he will do without closing the jaw, showing that the movement is not absolutely dependent upon the action of closing the jaws, although, as I have said above, I believe it usually accompanies it.

The Phalangists or Australian opossums closely resemble the macropidæ in their dental formation, but they possess partially-developed canines in the upper jaw, whilst the latter have none in either jaw except in very early life; but although these opossums have their two procumbent incisors similarly situated, they probably do not possess the power of utilising them in the same manner; I have examined some specimens, but have failed so far to find more than the looseness of connection at the symphysis referred to by Prof. Owen.

In the genus which is represented by the Phascolarctos or native bear of Australia, which possesses the same lower incisors but distinct canines in the upper jaw, this arrangement is certainly wanting, as the rami of the lower jaw are firmly united.

This remarkable formation of the lower jaw of these kanga

Snow Flakes

WHILST walking home on March 26, about one in the morning, snow began to fall very gently; but instead of the usual powdery or feathery appearance, each flake consisted of a distinct plate, in some cases perfect six-pointed crystals. I measured some of them, and the largest were as much as five-eighths still more remarkable; instead of the white opaque body one of an inch across. On taking up a handful the appearance was usually sees, the mass was pearly and semi-transparent, and so strongly resembling boracic acid, that I should have had some difficulty in distinguishing a handful of each substance by sight alone.

Near the lamps the effect was very beautiful, more especially when the road became covered, luminous points appearing in all directions, which scintillated like stars as one walked along, nearing the ground. whilst many of the falling crystals reflected iridescent hues on

When out of the town I ignited a piece of magnesium wire, and the effect was most brilliant.

It was a cold, dull night, barometer falling.
Burton-on-Trent

Rats and Water Casks

FRANK E. LOTT

retention of a construction, and of a set of muscles in a class of animals which have constantly required their aid to sustain life, which in other families of the animal kingdom have become rigid by ossification and cartilaginous formations, and by atrophy of the muscles in consequence of disuse.

The great plains and deserts over which these marsupials wander in search of food afford an exceedingly precarious supply of pasture in consequence of droughts and bushfires, which not unfrequently follow a superabundance of herbage. These animals, by means of their procumbent teeth which they make use of as shears, are thus enabled to cut off any green shoots or half-buried remains spared by a scorching sun, and obtain nourishment where any grass-feeding placental would certainly

starve.

It is in consequence, I believe, of the power which is by this means given to these marsupials of eating scanty pasturage closer to the ground than any other animal, that in the great pastoral districts of New South Wales and Queensland it has been found that they are far more destructive of food than any stock that can be put upon the land, and in places where wallabies and pademelons are exceedingly numerous, it is noticeable that the native grasses in the particular localities which they frequent become completely destroyed, and that such places remain ungrassed until fresh seed is scattered over them by the winds. HENRY WELD BLUNDELL Gordon Downs, Queensland, December 5, 1878

Measuring the Velocity of Sound in Air THE following simple way of arriving at the velocity of sound in air occurred to me lately:-Standing on a straight staircase between two blank walls (brick, and papered), which I find to be 32 inches apart, I clap my hands. The effect from each clap is a brief musical sound, metallic in character, and of quite appreciable pitch. It arises, doubtless, from the disturbance travelling to and fro between the walls. The pitch I find to be, as nearly as possible, G sharp (in the fourth space). Now, the number of complete vibrations per second, corresponding to this note, seems to be about 205 (see Deschanel's "Natural Philosophy," p. 820). This implies that the disturbance, when I clapped my hands, made 410 excursions across the pace per second. Consequently, 410 x 32 = 13,325 inches = 1,100 feet. This is exactly the number Deschanel gives as the velocity of sound in air at 50° (approximately our mean annual temperature). M.

an ordinary vent-peg hole. The rest of the voyage a tub placed in the square of the main hatchway was kept constantly supplied with water, besides one or more square tins of water main deck.

the

UNIVERSITY

In the above voyage we stayed a week in Madras, and m loosing the foretop-gallant-sail on leaving, a rat and five or six young ones fell to the deck; and the sail was found to be so much eaten and full of holes, made to form and line the nest, that the sail had to be unbent and replaced. Gurnet Bay, March 31 E. J. A'COURT SMITH P.S.-The ship was the Cornwall, East Indiaman, Capt. Cow.

HEINRICH WILHELM DOVE

PROF. HEINRICH WILHELM DOVE was born at Liegnitz, Silesia, on October 6, 1803, and at the age of eighteen passed from the schools of that town to the Universities of Breslau and Berlin, where for the next three years mathematics and physics. In 1826 he took his degree of he devoted himself assiduously to the study of Doctor of Philosophy, his thesis on the occasion being an inquiry regarding barometric changes; and it is further significant of his future life-work that his first published memoir was a paper on certain meteorological inquiries relative to winds, these two subjects holding a first place in the great problem of weather-changes.

Dove began his public life as tutor and Professor at Königsberg, where he remained till 1829, being then invited to Berlin as supplementary Professor of Physics. His strikingly clear-sighted, bold, and original intellect turned instinctively to that intricate group of questions in the domain of physics which comprise the science of meteorology, and his success in these fields as an original explorer was so marked and rapid that he soon achieved for himself a seat in the Royal Academy of Sciences, and some time thereafter was raised to the distinguished position of the Chair of Physics in the University of

Berlin.

Among the scientific and fashionable circles of Berlin he took first rank as a lecturer, the combined qualities of accurate science, fine imagination, lucidity of style, com

manding presence, and the extent over which his utterances were heard, marking him out as the Arago and Brewster of Germany. Germany showered on him in profusion those honours and offices which it gracefully and gratefully bestows on learning and science; and perhaps there is no learned or scientific society of any note that has not the name of Dove enrolled among its honorary members. After a protracted and hopeless illness he died on Sunday last, April 6, in the seventysixth year of his age.

In the Royal Society's Catalogue of scientific papers, the lists under Dove specify 234 memoirs written between the years 1827-73. These show him to have been a successful worker and investigator in electricity, optics, crystallography, and in such practical matters as measures and the art of measuring, or the metric system of civilised nations. But it was to meteorological inquiries he devoted his full strength and all the powers of his mind, and, by his herculean but well-directed labours he has written his name in large imperishable characters on the records of science.

His fame rests on the successful inquiries he carried out with a view to the discovery of the laws regulating atmospheric phenomena which apparently are under no law whatever. The work he will be long best known by is his isothermals and isabnormals of temperature for the globe, in which work one cannot sufficiently admire the breadth of view which sustained and animated him as an explorer during the long toilsome years spent in its preparation. Equally characterised by breadth of view, and what really seemed a love for the drudgery of detail even to profuseness when such drudgery appeared necessary or desirable in attaining his object, are his various works on winds, the manner of their veering and their relations to atmospheric pressure, temperature, humidity, and rainfall, and the important bearings of the results on the climatologies of the globe; on storms and their connections with the general circulation of the atmosphere; the influence of the variations of temperature on the development of plants; and the cold weather of May-to which may be added the valuable system of meteorological observations he gradually organised for Germany, and the many full discussions of these which he published from year to year.

the British Navy, being the first vessel built wholly of steel; she is an unarmoured dispatch vessel, specially designed for high speed and great coal endurance. Her principal dimensions are: length between perpendiculars, 300 feet, breadth, extreme, 46 feet, mean load draught, 19 feet 9 inches, displacement, 3,735 tons.

Special attention has been paid to resistance to torpedo attack by constructing the hold in twenty-one separate compartments and the double bottom and bunkers in forty; with the additional weight thus introduced, it is still found that there is a saving of weight in the hull by the use throughout of steel amounting to 12 per cent., or 175 tons. The engines take 28 per cent of the displacement, and 20 per cent. is available for coal, which is estimated to be sufficient for steaming 7,000 knots at a speed of 10 knots per hour. The speed attained by the Iris on the measured mile was 18.6 knots, with an expenditure of 23 indicated horse-power per ton of displacement as compared with 14 indicated horse-power, required by a torpedo vessel.

"On

In striking contrast with the Iris we have the monster proposed by Rear-Admiral J. H. Selwyn in his paper the most Powerful Ironclad." The author of the paper has long advocated some modification of the circular ironclad first proposed by Mr. Elder some years ago, and carried out with some alterations by Admiral Popoff. The vessel here proposed is 370 feet in length 220 feet in breadth, with a draught forward of 18 feet and aft 13 feet. Her armament is to consist of twenty 80ton guns, or eight 100-ton and eight 80-ton; these are to be mounted in two gun-pits on the Moncrieft hydro-pneumatic principle. The guns are carried on a turn-table of the full size of each gun-pit, the floors of which are composed of steel bars set on edge to provide for ventilation, but to keep out shell fragments; the breast-work round each will consist of 30 inches of armour-plating. The guns would be raised by hydraulic power to fire over the breast-work, recoiling automatically under cover for re-loading. The vessel would be protected with a belt of 30-inch armour round the water-line, and a thickness of 25 feet of coals stowed inside it. There would be two Perkins hydraulic engines of 21,000 horsepower for propulsion and steering, and these would be at once available for keeping the vessel afloat in case of a leak. The author estimates that if a hole 10 feet square were made by a torpedo, the engines would be able to keep the water under, while danger of sinking by such damage is much lessened, if the engines are partially disabled, by the large number of water-tight_compart

It cannot be denied that the Russian Popoffkas have been far from successful, especially in facility of steerage, which was one of the main advantages claimed for them, but it can only be determined by an actual experiment whether our naval authorities can overcome the difficulties in speed and steering which have baffled the Russian Admiralty. Even if a vessel as here proposed could not be made sea-going, or to attain a 16-knots speed as claimed, she would at least be more valuable as a harbour defence than a Spithead or Plymouth breakwater fort, and could be adapted to some sites at a less cost in proportion to the weight of armament.

"Armour for ships" by Mr. Barnaby, C.B, Director of Naval Construction, consists of a general review of the progress of armour-plating from its introduction in 1854 down to the present time. The description of the steel turret-plates manufactured by Messrs. Schneider at Creuzot, 32 inches thick, and weighing 65 tons, is not without significance in the present state of depression in the iron trade of this country, but some consolation is to be obtained from the account given of the steel-faced plates of Messrs. Brown and Cammell which shows that some progress is still being made nearer home. The paper by Admiral Sir R. Spencer Robinson, K.C.B.,

F.R.S., gives an exhaustive statement of the experiments that have been made on various targets at Shoeburyness, representing the armour of different ships. A table is given showing the displacement, thickness of armour, and proportion of the former to the latter in ships of different types; this ratio varies from 6.38 in the Warrior, 4'00 in the Alexandra, to 295 in the Dreadnought, and 2.50 in the Glatton; thus the last may be considered the most heavily-armoured vessel in proportion to size in the navy. The penetration of shot of different diameters and weights with various velocities is given, and the experiments show that it is proportional to the energy of the shot on impact whether due more to velocity or weight, and inversely proportional to diameter of shot; also that the resistance of solid plates is proportional to the square of their thickness. The resistance of composite targets is treated at some length, and a comparison drawn between the various forms adopted in existing ships and the Millwall shield designed by Mr. Hughes, in which the latter is shown to be preferable; but the questions of steel and steel-faced armour which are now attracting the attention of artillerists are not gone into, and are only referred to with the evident feeling that the end of the battle between guns and armour has not yet come.

The paper (6 On the Resistance given to Screw-Ships by the Action of the Screw-Propeller, and how to Remedy it," by Robert Griffiths, points out an important difficulty in screw-propulsion which has only recently been recognised. A screw-propeller obtains the resistance to drive the ship forward by accelerating the velocity of the currents of water flowing past the stern of the vessel; as in different parts of the screw's disk these currents are encountered at different velocities, the resistance to a blade is not uniform throughout a revolution. In experiments made at Devonport by towing a screw-pinnace, it was found that the water flowed through the lower half of the screw disk nearly at the speed at which the boat was towed, but in the upper half it was so dragged by the boat as to flow past the screw at only half that speed. In dynamometer diagrams, taken with H.M.S. Rattler, it was shown that the thrust of the screw varied from 2'9 to 4'1 tons in each revolution. The increase in the resistance of the ship, due to the working of the screw above that due to the ship herself when towed at the same speed, and which Mr. Froude has shown to be 40 or 50 per cent., is considerably greater when the upper currents are more accelerated than it would be if the acceleration were uniformly given to the whole column of water passed through by the screw disk. The author proposes a screw-propeller so constructed that the blades always meet with equal resistance. The blades are so made that more than half their surface is aft of the centre line, so that the pressure on their surface tends to lessen the pitch; they are also made movable in the boss, but so connected that by decreasing the pitch of one, that of the other is increased; when, therefore, one blade meets with more resistance than the other, the increased pressure causes it to turn and throw some of the work on the other.

In his paper on naval guns, Mr. C. W. Merrifield vigorously attacks the Woolwich type of gun, pointing out the disadvantages and absolute futility of the increasing twist in rifling at present adopted. It is now four or five years since this was first done by Prof. Osborne Reynolds, and, aided by the Thunderer explosion, it is to be hoped that the time is drawing near when the subject will receive the consideration of the War Department. The author also lays great stress on the advantages of breech-loaders over muzzle-loaders, regarding the latter now, with its complication of gear and fittings, as inferior to the former, even in the simplicity always claimed for it

Amongst other papers read at the meeting are the following:-" On Sir William Thomson's Navigational Sounding Machine," by P. M. Swan, in which the accu

racy of this now well-known apparatus is amply testified by a large number of observations; and a paper by Mr. J. Scott Russell, F.R.S., "On the true Nature of the Wave of Translation, and the Part it plays in Removing the Water out of the Way of a Ship with least Resistance.”

OUR ASTRONOMICAL COLUMN

NOTE ON 72 OPHIUCHI (0. 2. 342). The publication of the entire series of observations of this suspected double star, made at Pulkowa to 1876, does not lessen the difficulty of arriving at a definite conclusion as to its duplicity or otherwise. On November 1, 1841, it was noted double magnitudes 4 and 7 on Struve's scale, and, no doubt attached to the observation; on May 14, 1842, it appeared single, but at the epoch 1842'72 it was again double, the measures giving for position, 156°6, and distance, 1"3. Subsequent observations gave the following results :

1844'85

1845'62 1846 49 1847'50

1847'70

1848*79 1850*50 1851 51 1851'67

1852.63

Single, or with only a suspicion of elongation at 63°; images excellent.

With very good images; no companion seen. Single, or perhaps slightly wedged at 87°. Pos. 162° 4, dist. 1" 61, but there was a doubt if the object observed was not an optical illusion. Pos. 1681, dist. 1"6. M. Struve says: "I feel sure of the duplicity, but the images are not very good."

Single.

Pos. 166° 3, dist. 1"49.

After the observation a

note was added "This is only an optical deception."

Single; under excellent atmospheric conditions.

This last observation appearing decisive, M. Struve considered that 72 Ophiuchi should be omitted from the list of double-stars, and in the following years only examined it once (185966), when it was again single under very favourable conditions. But in 1876 he found reason to modify his view: at 187667 the satellite was seen very distinctly, with position 1560, distance 1"60; a fortnight later there were only very slight impressions of a satellite, and M. Struve remarked that the principal star of 70 Ophiuchi presented an analogous phenomenon, though less distinctly. Hence arose the suspicion that the said impressions were caused by accidental conditions of the air and the instrument. Nevertheless, on considering the preceding observations and the fact of their being made without the least recollection of anterior ones, M. Struve thinks their approximate agreement cannot be attributed to chance, and that we are necessarily led to infer that the star is 'really double, but the companion undergoes considerable and rapid variation of brightness. It is worthy of note that only three weeks before the Pulkowa observation of 1859, when the star was pronounced single, Secchi had recorded of it: "Certainly double, and well separated," his measures giving the position 3°75, distance o′′61.

THE VARIABLE STAR X CYGNI.-According to the later observations of Dr. Julius Schmidt at Athens, it is probable that the next maximum may occur on or about April 25, and the next minimum about December 14. At the last observed maximum on March 14, 1878, the star was hardly a fifth magnitude, which is about the mean brightness in that phase, the extreme limits of variation being two magnitudes or 4m.-6m. according to Prof. Schönfeld; at minimum it descends to 13m. No formula has yet been deduced which will represent satisfactorily the totality of the observations, commencing with those of Kirch the discoverer in 1686; considerable

irregularities following no law so far discovered occasionally presenting themselves. This is particularly evident if we compare Argelander's last formula in vol. vii. of the Bonn observations with the observed times of maxima during the last fifteen years. The place of the true x Cygni of Bayer, which is the variable, is, for 1880'o, in RA. 19h. 45m. 57'3s., N.P.D. 57° 23′ 18′′; it therefore follows the star to which Flamsteed attached this letter, 4m. 4s., and is south of it 50'6; Flamsteed's star ought to be called by his number, 17 Cygni. At the times when he was looking for Bayer's x, as Argelander has remarked, the variable would be near a minimum; hence his observing the nearest star of similar brightness.

THE MINOR PLANETS IN 1879.-Advanced sheets of the Berliner astronomisches Jahrbuch for 1881, containing places of the small planets during the present year have been circulated amongst observers, the ephemerides for the planets coming into opposition early in the year, some time since. There are positions of the first 187 members of this group, with the exception of Nos. 99 and 155, for which sufficient data are not available. Only two out of the number approach the earth at opposition, within her mean distance from the sun: Isis, on June 20, is distant o'995, with a south declination of 25°, and Hertha, on September 12, o'988, just upon the equator. No. 154 travels as far south as 50° about July 14.

BRORSEN'S COMET.-The observations of this comet made at Arcetri and Kremsmunster from March 10 to 19 with Dr. Schulze's other elements, fix the time of perihelion passage to about March 30'5716 G.M.T., which is nearly twelve hours later than that assigned by calculation. The following ephemeris is founded upon this corrected epoch for arrival at perihelion :

A COMMUNICATION in yesterday's Daily News, from a New York correspondent of that paper, gives a glowing, and to all appearance justifiably so, account of Mr. Edison's success in attaining a form of electric lighting that seems to be in all respects much superior to anything hitherto produced. The first impression made on the correspondent was the mild effect of the light on the eyes, its steadiness, and the absence of that ghastly hue which seems to be an invariable accompaniment of the carbon. This new form of light has only been attained after many disappointments on the part of Mr. Edison, who, however, has all along been confident of success.

During the past two months the progress towards its present perfection has been very rapid. Chiefly contri

buting to this result has been the discovery of a new alloy, the fusing-point of which is much higher than either platinum or iridium, in fact, than any known metal. This discovery is spoken of by some of Mr. Edison's chief employés as the greatest achievement of his life. This alloy also reduces the cost of the valuable metals used in each lamp to such a point as to do away entirely with Prof. Tyndall's criticism. It is said to possess properties heretofore unknown, or at least undefined by scientific men. Not only has it cheapened the cost, but the union of the metals has increased the illuminating power to such a degree that six lights are now obtained per horse power where only four were possible with the pure platinum coil. Six lights per horse-power is the number authoritatively stated, but Mr. Edison's chief assistant does not hesitate to predict that eleven lights will eventually be obtained for each horse-power. This is not expected from the Gramme machine, however, which is now used; but is hoped for after the completion of the new generator, which a dozen of the most skilled workmen at Menlo Park are now engaged in constructing.

The lamp itself takes many forms. In some instances it is attached to the wall, like a gas bracket, and in many others it hangs from the ceiling and takes the external form of a glass globe, capped by brass or nickel attachments. There is none of the hissing, sputtering, and flickering observable in the carbon lamps. The lamp which attracts most attention is, in appearance, a St. Germain student lamp, without the reservoir for the oil. covered copper wires alone connect it with the main conIt stands in the middle of a small table, and two fine In this the ducting cables from the Gramme machine. Daily News correspondent tells us, are embodied all the latest improvements. He also tells us that there cannot possibly be any mistake, as Mr. Edison has taken crucial precautions in all directions. There is nothing in the lamp itself that gives any idea of its construction. The cunning device for rendering the flame steady is in reality the idea of the quadruplex telegraph applied to heat instead of electricity. Now that the new alloy has been discovered, its twofold purpose of preventing fusion and steadying the light is no longer served. The expansion of the tiny key, or switch, breaks the current for the fraction of a second, and permits the actual, though imperceptible, cooling of the incandescent coil. This connection is made and broken many times during each second, so that to human eyes the light is constant as the sun. The movement of a finger and thumb converts the glowing meteor before us into a night lamp for a sickroom. Again, it is seen at one-candle power, then at two, and so on. It is as manageable as a tallow dip, and much more satisfactory. It will not go out of itself, and needs no care. The little coil of wire is hermetically It is not in a vacuum, but sealed in the glass chamber. the chamber is filled with air. There is a sensitive spot on the metal cap in which the glass tube sits, and the expansion of the air manipulates the switch. The heat of the metal itself, therefore, is no longer relied on. inventor explains that after all manner of severe tests this has been found the easiest and the least easily deranged manner of controlling the light. The difficulty of making thin plates of metal of equal density and weight rendered the previous method impracticable for small lights, although it will probably be the best form in which to secure the desired result where the lamps are to show lights of great intensity.

The

As there seems no reason to distrust the evidence of the Daily News correspondent, it may be accepted that Mr. Edison has succeeded in going a long way to solve some of the difficulties connected with the practical adoption of electric lighting. It is stated that in a few months the Edison Company will be prepared to supply the light to such private consumers as may desire it at at least one-third or one-fourth the cost of gas.

EXPERIMENTAL RESEARCHES ON THE REPULSION RESULTING FROM RADIATION1

HA

II.

AVING completed the experimental investigation of the amount of repulsion produced by radiation on disks of various kinds, and coated with different substances, I turned my attention to the amount of repulsion produced when polarised light is allowed to fall on a plate of tourmaline suspended in vacuo in a torsion balance. It was originally thought that a slice of tourmaline, being black to a ray of light polarised in one plane, and white to a ray polarised in the other plane, would be repelled when the incident light was quenched by it, and not affected when the incident light passed through it. Experiments, however, prove that this action does not exist in any

The positive repul

repelled, causing negative rotation.1 sion was, however, rather stronger than the negative repulsion, so that, when both sides were illuminated, the force was only that due to the difference of these repulsions.

A hot glass shade is a convenient means of heating the bulb, by immersing it in a hot-air bath, without the liability of introducing action of rays other than those emitted by hot glass. On inverting a hot glass shade over the bulb in the above experiment, negative rotation was produced which changed to positive on cooling. Both these rotations were stronger than that given by the candle. The experiment was varied (1) by 6 millims. of the sides being turned up instead of 4; (2) by folding the plates across the vertical diagonal and then across their horizontal diagonal; (3) by attaching flat

appreciable degree, the repulsion resulting from radiation being almost entirely a surface action, whilst the action of a tourmaline on a ray of polarised light is one in which thickness is necessary.

I next examined the effect of shape in influencing the amount and direction of repulsion. These experiments were for the most part tried with the apparatus shown in Fig. 3 (p. 513, part I.). Through the open top access can readily be obtained, and disks, plates, &c., can be quickly tested by being fixed to the extremities of a pair of aluminium arms, with a glass cap in the centre, rotating on the needle-point. Plates, 12 millims. square, cut from thin aluminium foil, were mounted diamondwise on arms, and supported on the needle-point inside the bulb. The plates were lampblacked on sides facing

plates to the arms at an angle of 45°, blacking them on the insides away from the bulb, and repeating the experiment with plates blacked on the outsides. The results obtained show that when flat plates are taken blacked on alternate sides, the rotation is normal or positive, i.e., the black side is repelled. When the outer corners of each plate is turned up so as to keep the blacked surface on the concave side, the positive rotation is either diminished, stopped, or converted into negative rotation, according to the amount of surface of the plate which has been turned up. The favourable presentation of the surface of the vanes to the inside of the bulb has more influence on the movement than has the colour of the surface. Radiometers constructed with silver flake vanes set at an angle

Elevation. Scale

FIG. 5.

opposite ways, and the apparatus was well exhausted. The vanes behaved like an ordinary metal radiometer in respect to light and radiant heat. Fig. 4 shows the elevation and plan of the fly, the dotted side representing the one which was lampblacked. The arrows show the direction of positive rotation when exposed to the light of a standard candle 3'5 inches off. The outer corners of the aluminium plates were now turned up at an angle of 45°, 4 millims. of the two sides being turned up, leaving 8 millims. flat, as shown in Fig. 5. They were lampblacked on the inside, as shown in the figure by dots. A lighted candle 3.5 inches off caused very slow and feeble positive rotation. On shading the light from the black side, the bright side was repelled, causing positive rotation; and on shading the light from the bright side the black was

• Continued from p. 514.

FIG. 7.

of 45° and blacked on the outside prove the most sensitive for light hitherto constructed.

I now endeavoured to clear up many anomalous results which had attended the application of heat either by hot shades or by hot water to radiometers. There was an antagonistic action between the effect of shape and that of colour of surface, the two actions sometimes acting together and sometimes in opposition.

bulb, shape of vanes, and degree of exhaustion, only Five radiometers were made exactly alike in size of differing in the material of which the vanes were composed. No. I was made of mica, 0003 inch in thickness; No. 2, of mica, 0'0005 inch in thickness; No. 3, of pith,

I call the rotation positive when the black or driving side is repelled, and negative when the side which under ordinary circumstances would be the driving side, moves towards the light.