USEFUL HAND-BOOKS.

The Ornamental Penman's Pocketbook of Alphabets, for sign-writers, engravers, stone-cutters and draftsmen, 20 cts. A System of Easy Lettering, by Howard Cromwell, 50 cts. Practical Electrics: A Universal Handybook on Every-day Electrical Matters, 135 pp., fully illustrated, 12mo, cloth, 75 cts. Notes on Design of Small Dynamo, by G. Halliday, 79 pp., with a number of plates to scale, 12mo, cloth,

Introduction

A. Williams..

Periodical Cicada. C. V. Riley..

86

Notes and News..

86

W. Gillett, 87 pp., 12 folding plates, 12mo, cloth, $2 SPON & CHAMBERLAIN, Publishers, 12 Cortlandt St., New York. Illustrated and descriptive catalogues, 10 cts.

93

94

95

96

97

NEW METHOD OF PROTECTING BUILDINGS FROM LIGHTNING. SPARE THE ROD AND SPOIL THE HOUSE! Lightning Destroys. Shall it be Your House or a Pound of Copper?

PROTECTION FROM LIGHTNING.

What is the Problem?

IN seeking a means of protection from lightning-discharges, we have in view two objects, the one the prevention of damage to buildings, and the other the prevention of Injury to life. In order to destroy a building in whole or in part, it is necessary that work should be done; that is, as physicists express it, energy is required. Just before the lightning-discharge takes place, the energy capable of doing the damage which we seek to prevent exists in the column of air extending from the cloud to the earth in some form that makes it capable of appearing as what we call electricity. We will therefore call it electrical energy. What this electrical energy is, it is not necessary for us to consider in this place; but that it exists there can be no doubt, as it manifests itself in the destruction of buildings. The problem that we have to deal with, therefore, is the conversion of this energy into some other form, and the ac complishment of this in such a way as shall result in the least injury to property and life.

Why Have the Old Rods Failed?

When lightning-rods were first proposed, the science of energetics was entirely undeveloped; that is to say, in the middle of the last century scientific men had not come to recognize the fact that the different forms of energy heat, electricity, mechanical power, etc.- were convertible one into the other, and that each could produce just so much of each of the other forms, and no more. The doctrine of the conservation and correlation of energy was first clearly worked out in the early part of this century. There were, however, some facts known in regard to electricity a hundred and forty years ago; and among these were the attracting power of points for an electric spark, and the conducting power of metals. Lightning-rods were therefore introduced with the idea that the electricity existing in the lightning-discharge could be conveyed around the building which it was proposed to protect, and that the building would thus be saved.

The question as to dissipation of the energy involved was entirely ignored, naturally; and from that time to this, in spite of the best endeavors of those Interested, lightning-rods constructed in accordance with Franklin's principle have not furnished satisfactory protection. The reason for this is apparent when it is considered that the electrical energy existing in the atmosphere before the discharge, or, more exactly, in the column of dielectric from the cloud to the earth, above referred to, reaches its maximum value on the surface of the conductors that chance to be within the column of dielectric; so that the greatest display of energy will be on the surface of the very lightningrods that were meant to protect, and damage results, as so often proves to be the case.

It will be understood, of course, that this display of energy on the surface of the old lightning-rods is aided by their being more or less insulated from the earth, but in any event the very existence of such a mass of metal as an old lightning-rod can only tend to produce a disastrous dissipation of electrical energy upon its surface," to draw the lightning," as it is so commonly put.

Is there a Better Means of Protection?

Having cleared our minds, therefore, of any idea of conducting electricity, and keeping clearly in view the fact that in providing protection against lightning we must furnish some means by which the electrical energy may be harmlessly dissipated, the question arises, "Can an improved form be given to the rod so that it shall aid in this dissipation?"

New Store.
New Stock.

New Departments.

Send for our "Winter Bulletin," recently issued. Minerals, Gems, Microscopical Sections, Fine Lapidary Work.

GEO. L. ENGLISH & CO., Mineralogists, Removed to 64 East 12th Street, New York

GERMANIA A monthly magazine for the study

of the German language and literature, is highly recommended by college professors and the press as "the best effort yet made to assist the student of German, and to interest him in his pursuit." Its BEGINNERS' CORNER furnishes every year a complete and interesting course in German grammar. $2 a year. Single copies 20 cents. P. O. Box 151, Manchester, N. H.

As the electrical energy involved manifests itself on the surface of conductors, the improved rod should be metallic; but, instead of making a large rod, suppose that we make it comparatively small in size, so that the total amount of metal running from the top of the house to some poiut a little below the foundations shall not exceed one pound. Suppose, again, that we introduce numerous insulating Joints in this rod. We shall then have a rod that experience shows will be readily destroyed-will be readily dissipated when a discharge takes place; an i it will be evident, that, so far as the electrical energy is consumed in doing this, there will be the less to do other damage.

The only point that remains to be proved as to the utility of such a rod is to show that the dissipation of such a conductor does not tend to injure other bodies in its immediate vicinity. On this point I can only say that I have found no case where such a conductor (for instance, a bell wire) has been dissipated, even if resting against a plastered wall, where there has been any material damage done to surrounding objects.

Of course, it is readily understood that such an explosion cannot take place in a confined space without the rupture of the walls (the wire cannot be boarded over); but in every case that I have found recorded this dissipation takes place just as gunpowder burns when spread on a board. The objects against which the conductor rests may be stained, but they are not shattered, I would therefore make clear this distinction between the action of electrlcal energy when dissipated on the surface of a large conductor and when dissipated on the surface of a comparatively small or easily di-sipated conductor. When dissipated on the surface of a large conductor, a conductor so strong as to resist the explosive effect, - damage results to objects around. When dissipated on the surface of a small conductor, the conductor goes, but the other objects around are saved

A Typical Case of the Action of a Small Conductor. Franklin, in a letter to Collinson read before the London Royal Society, Dec. 18, 1755, describing the partial destruction by lightning of a church-tower at Newbury, Mass., wrote, "Near the bell was fixed an iron hammer to strike the hours; and from the tall of the hammer a wire went down through a small gimlet-hole in the floor that the bell stood upon, and through a second floor in like manner; then horizontally under and near the plastered ceiling of that second floor, till it came near a plastered wall; then down by the side of that wall to a clock, which stood about twenty feet below the bell. The wire was not bigger than a common knitting needle. The spire was split all to pieces by the lightning, and the parts flung in all directions over the square in which the church stood, so that nothing remained above the bell. The lightrire passed between the hammer and the clock in the above-mentioned wire, without hurting either of the floors, or having any effect upon them (except making the gimlet-holes, through which the wire passed, a little bigger), and without hurting the plastered wall, or any part of the building, so far as the aforesaid wire and the pendulum-wire of the clock extended; which latter wire was about the thickness of a goose-quill. From the end of the pendulum, down quite to the ground, the building was exceedingly rent and damaged.... No part of the aforementioned long, small wire, between the clock and the hammer, could be found, except about two inches that hung to the tall of the hammer, and about as much that was fastened to the clock; the rest being exploded, and its particles dissipated in smoke and air, as gunpowder is by common fire, and had only left a black smutty track on the plas tering, three or four inches broad, darkest in the middle, and fainter towards the edges, all along the celling, under which it passed, and down the wall."' One hundred feet of the Hodges Patent Lightning Dispeller (made under patents of N. D. C. Hodges, Editor of Science) will be malled, postpaid, to any address, on receipt of five dollars ($5).

Correspondence solicited. Agents wanted. AMERICAN LIGHTNING PROTECTION CO.. 874 Broadway, New York Citv.

Littell's Living Age.

THE ONLY WEEKLY ECLECTIC. 1893.

1844. "The Oldest and the Best." It selects from the whole wide field of EUROPEAN PERIODICAL LITERATURE the best articles by THE ABLEST LIVING WRITERS In every department of Literature, Science, Politics and Art. OPINIONS.

"Only the best has ever filled its pages; the best thought rendered in the purest English. Nothing poor or unworthy has ever appeared in the columns of THE LIVING AGE."-The Presbyterian, Phila.

"Considering its size, it is the cheapest of literary periodicals, and no collection of magazine literature is complete without this foremost of eclectics."-Educational Courant, Louisville, Ky.

"It is one of the few periodicals which seem indispensable. It contains nearly all the good literature of the time."-The Churchman, New York.

"The fields of fiction, biography, travel, science, poetry, criticism, and social and religious discussion all come within its domain."— Boston Journal.

"To read it is itself an education in the course of modern thought and literature."— - Buffalo Commercial Advertiser.

Published WEEKLY at $8.00 a year, free of postage.

Club Rates. -For $10.15 THE LIVING AGE and SCIENCE will be sent for a year, postpaid.

Rates for clubbing THE LIVING AGE with other periodicals will be sent on application.

Sample copies of THE LIVING AGE, 15 cents each. Address,

Architectural publication in the country. Littell & Co., 31 Bedford St., Boston, Mass. Interesting articles on architecture, Sanitation, Archæology, Decoration, etc., by the ablest writers. Richly illustrated Issued weekly. Send stamp for specimen copy to the publishers,

BRENTANO'S,

Publishers, Importers, Booksellers.

Ticknor & Co., 211 Tremont St., Boston. branches of science, and in all languages.

1869.

THE

1893.

We make a specialty of technical works in all Subscriptions taken for all American and foreign scientific periodicals.

Our Paris and London branches enable us to im

RACES AND PEOPLES.

By DANIEL G, BRINTON, M.D.

"The book is good, thoroughly good, and will long remain the best accessible elementary ethnography in our language."-The Christian Union.

"We strongly recommend Dr. Brinton's 'Races and Peoples' to both beginners and scholars. We are not aware of any other recent work on the science of which it treats in the English language." !-Asiatic Quarterly.

"His book is an excellent one, and we can heartily recommend it as an introductory manual of ethnology."-The Monist.

"A useful and really interesting work, which deserves to be widely read and studied both in Europe and America."-Brighton (Eng.) Herald.

"This volume is most stimulating. It is written with great clearness, so that anybody can understand, and while in some ways, perforce, superficial, grasps very well the complete field of humanity."The New York Times.

"Dr. Brinton invests his scientific illustrations and measurements with an indescribable charm of narration, so that 'Races and Peoples,' avowedly a record of discovered facts, is in reality a strong stimulant to the imagination."-Philadelphia Public Ledger.

"The work is indispensable to the student who requires an intelligent guide to a course of ethnographic reading."-Philadelphia Times.

Price, postpaid, $1.75.

THE AMERICAN RACE.

By DANIEL G. BRINTON, M.D. "The book is one of unusual interest and value."Inter Ocean.

"Dr. Daniel G. Brinton writes as the acknowledged authority of the subject."-Philadelphia Press.

"The work will be of genuine value to all who wish to know the substance of what has been found out about the indigenous Americans."-Nature. "A masterly discussion, and an example of the -Philadelphia Ledger..

Subscription, $3.00 per year.

RESTORE YOUR EYESIGHT

Cataracts, scars or films can be absorbed and paralyzed nerves restored, without the knife or risk. Diseased eyes or lids can be cured by our home treatment. "We prove it." Hundreds convinced. Our illustrated pamphlet, 'Home Treatment for Eyes," free. Don't miss it. Everybody wants it. "THE EYE," Glens Falls, N.Y.

The Boston Medical and

BOSTON,

Surgical Journal.

Price, postpaid, $2.

N. D. C. HODGES,

874 BROADWAY, NEW YORK.

Pennsylvania Bedford Springs Mineral Water

For Liver, Kidney and B dder Troubles.
For Gravel, Gall Stones, Jaundice.
For Dyspepsia, Rheumatism and Gout.
For Dropsy, Bright's Disease, Diabetes.
For Hemorrhoids, Etc.

It has been used medicinally and prescribed by physicians for nearly one hundred years.

DIRECTIONS:-Take one or two glasses about a half-hour before each meal.

Case One Dozen Half-Gallon Bottles, $4.50.
Case Fifty Quarts (Aerated), $7.50.

Bedford Mineral Springs Co., Bedford, Pa.
Philadelphia Office, 1004 Walnut St.

BUILDING BOOKS.

DRAWING

1893 Catalogue

of Books on Building, Painting, and Decorating, also Catalogue of Drawing Instruments and Ma, terials, sent free on appli cation to

Wm. T. Comstock, 23 Warren St., New York.

NEW YORK, AUGUST 18, 1893.

BOTANY IN JAMAICA.

BY JAMES ELLIS HUMPHREYS.

WE are apt to think, when speaking of American botany and botanists, only of those of the United States and Canada, assuming that our southern neighbors, both continental and insular, have not yet reached that stage of civilization that encourages the cultivation of the sciences. And so far as those regions are concerned which have felt the influence chiefly of Latin civilization, this is measurably true. But some of the neighboring islands have been under Anglo-Saxon rule for two centuries or more, and have felt different influences. Not, indeed, that their people, as a class, have been much affected by contact with their rulers, but in the British islands the mother country has especially fostered botanical study from an early time, and British residents have carried with them the scientific impulse.

Jamaica has been a British colony for fully two hundred years, and it is now more than one hundred since its first botanic garden was established at Bath. At first privately supported, it afterward received spasmodic government support. But eventually the site was abandoned and a new location was chosen beside the Wag water and among the beautiful hills of the interior nineteen miles north of Kingston. From this time the support of the government was constant and effective, and the Castleton garden grew steadily in consequence, under competent directors sent out from England. It has now an especially notable collection of palms and orchids, besides its economic collection.

Meantime the Hope Gardens, near Gordon Town, and six miles from Kingston, begun for private pleasure when the island was in the full tide of its prosperity from the profits of sugar and rum, have been taken up by the government and are destined to be the chief botanical centre of the island. This collection is newer than that at Castleton and therefore does not possess as many fine specimens and, in some other respects, does not equal it. But most of the propagating and active work of the department is now done at the Hope Gardens. As must inevitably be the case with most government establishments, the chief work of the Botanical department of Jamaica, as of other British colonies, is economic, the study of the useful plants of the colony, their propagation and products. Its work is at present ably directed by Mr. William Fawcett, F. L. S., formerly of the British Museum.

A third establishment in charge of the department is the experimental Cinchona plantation far up the Blue Mountains. Here, also, is the official residence of the Director, in an almost ideal location and climate. Indeed, it is said, to quite justify the enthusiasm of an admirer, who called it "the loveliest spot in the British empire."

This place, called Cinchona, can be reached only by a narrow bridle-path that runs twelve miles upward into the heart of the mountains from Gordon Town.

The department issues a periodical bulletin of the results of its work.

Ever since the time of Patrick Bowne and Sir Hans Sloane, the higher plants of the island have found devoted

students. And among them must be specially mentioned Grisebach, whose "Flora of the British West Indies," London, 1863, remains the only hand-book of the subject. But the Thallophytes of the region have received little attention and offer a very attractive field.

The wife of the present energetic governor of the island, Sir Henry Blake, some time since proposed the raising of a fund to establish a permanent marine biological laboratory as a memorial to Columbus, who landed on the island on his second voyage. The idea is an admirable one, but the project remains, so far as can be learned, in statu quo. A small and well-equipped laboratory at a suitable point on the island, open to the zoologists and botanists of the world, might be of the greatest service in affording means for the collection and preservation of the numberless tropical forms of life in which Jamaica and the surrounding waters abound. A party of zoologists from the Johns Hopkins University has this year, for the second time, established a temporary laboratory at Port Henderson on Kingston harbor; but I understand that this choice of a location has been largely governed by the presence of suitable accommodations. It will be agreed that, in determining the site for a permanent laboratory, the abundance of available vegetable, as well as animal, life should be consulted. After a somewhat careful examination of the marine flora of the easterly part of the island, as far west as St. Ann's Bay, the writer can say that several of the ports on the north side are far more favorable, botanically, than Kingston harbor. And perhaps no region is, on the whole, more favorably situated or richer in its vegetation than the neighborhood of Port Antonio. This port has more frequent communication with the United States than even Kingston, from its extensive fruit trade. And the journey from Europe to Jamaica is less monotonous and less expensive, as well as quite as quick, via the United States, as by the Royal Mail from England.

Another factor of considerable importance lies in the much cooler and more healthful climate of the north side of the island, as compared with the south side.

In Jamaica, then, the botanist finds evidences of past and present activity in certain lines, and the sympathy and aid of fellow workers. It is much to be hoped that he may soon be able to find, also, the laboratory facilities, which will enable him to study to the best advantage the unsolved problems of tropical vegetation.

INTRODUCTION OF WEEDS IN GRASS SEED.

BY THOMAS A. WILLIAMS, STATE AGR'L COLLEGE, BROOKINGS, S. D.

In the course of some experiments on forage plants, which were begun last season on the Station grounds, quite a large quantity of grass and clover seed was purchased from various seedsmen, principally from Hendersons, of New York. At the time of sowing some of the packages were found to contain more or less seed of various weedy plants. The plots were watched closely, and the following plants were found to have been introduced:

Cruciferae. Nasturtium palustre, (L.) D. C.; Sisymbrium officinale, (L.) Scop; Camelina sativa, (L.) Crantz; Brassica arvensis (L.) B. S. P.; Brassica alba, (L.) Gray; Brassica nigra, (L.) Koch; Brassica campestris, L.; Erysimum cheiranthoides, L.; Erysimum orientale (?) L.; Diplotaxis tenui

folia (?) (L.) D. C.; Raphanus raphanistrum, L.; Raphanus sativus, L.

Capparideae. Cleome integrifolia, Torr. and Gray.
Violarieae.-Viola tricolor, L.

Caryophylleae.-Saponaria officinalis, L.; Saponaria vaccaria, L.; Silene antirrhina, L.; Silene noctiflora, L.; Lychnis alba, Mill; Agrostemma githago, L.; Cerastium. arvense, L.; Stellaria media, (L.) Smith; Spergula arvensis, L.

Geraniaceae.-Geranium pusillum, L.; Erodium cicutarium, (L.) L'Her.

Leguminosae.-Vicia sativa, (L.) Koch.

Umbilliferae.-Carum carui, L.; Coriandriem sativum, L.; Daucus carota, L.

Rubiaceae. Galium sp. 2; Galium tricorne, With.; Galium verum, L.

Compositae.-Anthernis cotula, L.; Achillea millefolium, L.; Carduus nutans, L.; Centaurea cyanus, L,; Taraxacum officinale, Web.; Sonchus arvensis, L.; Sonchus asper, Vill; Sonchus olenaceus, L.

- Borragineae.-Lithospermum arvense, L. Plantagineae.-Plantago lanceolata, L.

Polygonaceae.-Rumex crispus, L.; Rumex acetosella, L.; Rumex acetosa, L.

Gramineae. Panicum crus galli, L.; Panicum glabrum, (Schrad,) Gaud; Panicum sanguinale, L.; Avena fatua, L.; Eragrostis, major, Host; Eragrostis pilosa, (L.) Beauv; Bromus mollis, L.

It is interesting to note the spread of weeds in a new State. Saponaria vaccaria is found along the railroads, together with Anthemis cotula, over the whole of eastern South Dakota. The former has even followed up the freighting trails over the range between Pierre and the Black Hills, where it is quite common, particularly at watering and camping places. Man is evidently the one who is responsible for the distribution of this weed.

PERIODICAL CICADA.

BY C. V. RILEY, UNITED STATES DEPARTMENT OF AGRICULTURE, DIVISION OF ENTOMOLOGY, WASHINGTON, D. C.

DURING the present year two broods of the Periodical Cicada, or so-called "Seventeen-year Locust" (Cicada septendecim L.), one of the seventeen-year (septendecim) race, and one of the thirteen-year (tredecim) race, will make their appearance in different parts of the country.

The following list of localities has been prepared from previous records. Any evidence giving the extent of territory over which they appear in any county or state, or any well-attested dates of their appearance in previous years, will be thankfully received and appreciated.

BROOD XVI. Tredecim-(1880, 1893.) AlabamaLowndes County. Georgia-Cobb and Cherokee Counties. Tennessee-Lincoln County. North CarolinaLincoln and Moore Counties. This brood is but little known, and all localities require further confirmation this year.

BROOD XI. Septendecim (1876, 1893). North Carolina From Raleigh, Wake County, to the northern line of the State; also in the counties of Rowan, Davie, Cabarrus and Iredell. Virginia From Petersburg, Dinwiddie County, to the northern line of the State; Bedford and Rockbridge Counties; Valley of Virginia, from the Potomac River to the Tennessee and North Carolina

lines. District of Columbia-Woods north of Washing

Maryland Southern half of St. Mary's County. Kentucky-Trimble County. Indiana-Knox, Sullivan and Posey Counties. Illinois-Madison County. KansasDickinson and Leavenworth Counties. Colorado-Cheyenne Canyon. This is a well-established brood, most of

the localities in the Eastern States, as well as those in Indiana and Illinois, having been verified in past years; but the localities in Kentucky and Kansas require confirmation, and that in Colorado is extremely doubtful.

NOTES AND NEWS.

more

In this age of rapid advancement in all lines of knowledge, especially in science, people have learned that combined organized labor accomplishes far more exact results than individual effort. Every department of science has its organization for the promotion of that science. Such an organization is the Wilson Ornithological chapter of the Agassiz Association, for the promotion of American ornithology. It is composed of active, associate and honorary members. It is in no respect a rival of the American Ornithologists' Union, but has its work conducted on a cooperative plan, and therefore necessarily largely systematic. While furnishing the advanced with ample material for work, it also offers such opportunities to the younger and less experienced as are best suited to their needs. It seeks to educate those just beginning and those pursuing a dilatory course into the highest usefulness as working ornithologists. Active members pay an initiation fee and a yearly assessment of $1.00, and are limited to 100 in number. This number is now nearly reached. Associate members pay a yearly assessment of 50 cents and are unlimited in number. All working ornithologists are invited to join and aid in the work. Applications for membership should reach the President or Secretary before Sept. 20, to insure insertion in the list of candidates for the October election. Address either Willard N. Clute, Sec., Binghampton, N. Y., or Lynds Jones, Oberlin, Ohio.

-William Beverley Harison published on the 15th "The Foreigner's Manual of English." This is prepared for use in mixed classes of foreigners, and can be used without any knowledge of the several languages, as English only is used throughout. It has been carefully corrected to embody all of the suggestions of Gouin, whose book appeared after completion of first MS., and during revision the MS. has been successfully used in teaching Chinese, Polish Jews and others absolutely ignorant of both written or spoken English. The lessons are arranged to give in each a concrete subject, and a useful vocabulary is given to enable the student to talk from the beginning.

-The Chain Hardy & Company, Denver, Colo., have just ready the revised and enlarged edition of the "Geology of Colorado and Western Ore Deposits." This work of Professor Lakes, of the State School of Mines, has already run through one edition as applied to Colorado. Now that the Western States have been included the sale is expected to be quite extended. The plates illustrating the geological formations are very elaborate, and illustrate the peculiarities of veins and ore deposits. The book is designed for a text-book, and is also adapted for general reading by those interested in mining.

-Rand, McNally & Co. have in preparation the Proceedings of the Bankers' and Financiers' Congress held in Chicago from June 19 to 24.

-The Scientific Publishing Co. have just ready a work on "Universal Bimetallism and an international monetary clearing house, together with a record of the world's money statistics of gold and silver," etc., by Richard P. Rothwell, editor of the Engineering and Mining Journal.

-Macmillan & Co. have just ready "A Treatise on the Theory of Functions," by Prof. James Harkness, of Bryn Mawr College.

SCIENCE:

PUBLISHED BY N. D. C. HODGES, 874 BROADWAY, NEW YORK.

SUBSCRIPTIONS TO ANY PART OF THE WORLD, $3.50 A YEAR.

To any contributor, on request in advance, one hundred copies of the issue containing his article will be sent without charge. More copies will be supplied at about cost, also if ordered in advance. Reprints are not supplied, as for obvious reasons we desire to circulate as many copies of SCIENCE as possible. Authors are, however, at perfect liberty to have their articles reprinted elsewere. For illustrations, drawings in black and white suitable for photo-engraving should be supplied by the contributor. Rejected manuscripts will be returned to the authors only when the requisite amount of postage accompanies the manuscript. Whatever is intended for insertion must be authenticated by the name and address of the writer; not necessarily for publication, but as a guaranty of good faith. We do not hold ourselves responsible for any view or opinions expressed in the communications of our correspondents.

Attention is called to the "Wants" column. It is invaluable to those who use it in soliciting information or seeking new positions. The name and address of applicants should be given in full, so that answers will go direct to them. The "Exchange" column is likewise open.

THE ATMOSPHERE OF STELLAR SPACE.

BY G. D. LIVEING, CAMBRIDGE, ENGLAND.

It was an interesting speculation that Sir R. Ball opened up in this journal, a short time since, with regard to the lunar atmosphere. His argument might easily be carried further, and would take us, as I shall try to show, into the realms of stellar space. It has been objected to his theory that the velocity of the particles of air at ordinary temperatures, though on the average about five hundred yards per second, is not enough to carry a particle so quickly away from the moon that it would not be drawn back again by its gravitation. This objection vanishes if we consider, not the average velocity, but the velocities of individual particles, and the changes those velocities rapidly undergo in consequence of frequent collisions among the particles. It is not easy to grasp the numbers involved in my argument, but I will state them on the authority of Lord Kelvin's popular lecture on the size of atoms. He gives the number of particles in one cubic centimetre, or one-sixteenth of a cubic inch, of atmospheric air at ordinary barometric pressure and at ordinary temperature, as not less than a million million of millions, or 1018. Maxwell, in his article on "Atoms," in the Encyclopædia Britannica, makes the number greater. These particles cannot move far, not more on the average than about one hundredth of a thousandth of a centimetre, without encountering one another, so that each particle collides with one or another of its neighbors no less than five thousand million times in every second. If we suppose the density of the moon's atmosphere to be only a millionth of that of our atmosphere at the earth's surface, there will still be at least a million millions of particles in one cubic centimetre of it, and the frequency of their encounters with each other will still be some thousands per second for each of them. These encounters will cause them to be perpetually changing their velocities, and while some will have, at any given instant, velocities many times greater than the average, others will move at correspondingly slower rates. The directions, also, of their movements will be constantly changing from the same

cause.

If we suppose two particles, moving with equal velocities in directions at right angles to one another, to come into direct collision, one of them will have its veloc-. ity increased in the ratio of the square root of two to one, or rather more than seven to five, while the other will be reduced to momentary rest. If, now, the former come into a similar collision with a third particle, one of these two

will acquire a still greater velocity. And considering the prodigious number of the particles and the short distance they can move without encountering others, it is evident that there must be an immense variety of rates of motion amongst them, and many of them must have velocities far exceeding that necessary to carry them clear away from the moon, or the earth, or even from the sun. In fact, amongst so many millions of millions the chance that some one will go on increasing its velocity at every one of a large number of successive encounters is very great indeed, practically a certainty. If this be granted, some, if it be but a small fraction of the whole, will be always escaping from the outer surface of the lunar atmosphere into the planetary space; and the like must go on from the atmospheres of other planets, only the fraction of the whole which get clear away from the bigger planets will be so much less because of the greater attraction of the bigger masses.

One interesting consequence of this escape of only the quicker moving particles, is that the temperature of interplanetary space must be thereby raised above that of the outer regions of a planet's atmosphere. For the temperature is directly proportional to the average square of the velocities of the particles, and as only the quickest fly off for good, the average velocity of the remainder must be less than that of those that break away. The process of dissipating an atmosphere into space might be stopped by its own cooling effect. But it is obvious that there is another cause which prevents anything like this. The planets are continually sweeping through the interplanetary space where the escaped particles are moving about, and even if the density of this interplanetary atmosphere be only a millionth of a millionth of the density of that at the earth's surface, still there will be at least a million particles in each cubic centimetre, and some of them will get swept up by the planets in their course and will not get away again. Hence the process of dissipation will cease when a planet picks up in its course through space just as many as it loses by diffusion in the same time. It follows from this that there must exist in planetary space an atmosphere, greatly reduced in density, it is true, but of the same chemical constitution as the earth's atmos

phere. That is to say, the chemical constituents will be the same, though not quite in the same proportions. For the average velocity of the particles of nitrogen is a trifle greater than that of the particles of oxygen, and so the former will escape into space rather more frequently in proportion to their numbers than the latter. Besides, the effect of gravity is to increase very slightly the proportion of oxygen to nitrogen in the lower strata of the atmosphere. Hence, for both reasons, the atmosphere of planetary space will be a trifle richer in nitrogen than the air we breathe. There is so very little free hydrogen in our atmosphere that we cannot detect it, but for all that, it is most probable that there is a very little. And as oxygen particles are sixteen times as heavy as those of hydrogen, the proportion of free hydrogen to the other gases will be proportionally greater in the upper regions of the air than in the lower; and since hydrogen particles move four times as quickly as oxygen particles, it follows that the former will escape from the earth's attraction about four times as fast, and so the proportion of hydrogen in planetary space may be sensibly greater than in air we are able to test. A similar argument will apply to particles of water vapor, which are little more than half as massive as particles of oxygen. If all the planets are thus losing continually some of their atmospheres and picking up an equal amount from the space they move in, it follows that all the planets must have atmospheres of similar constitution to our own. For each planet has for ages been losing some of its own and acquiring some of the air