whole. We have corrected those so far as practicable. And the student will be further aided by our notice of the author and place of first description of each specimen. Their early notice of the meteorite gives a certain probability to their truest knowledge of the date.

We have given the weights of our specimens in two columns. The first gives weight of our largest piece, the second the total weight which we possess of the kind. We follow usual custom in measuring this weight in grammes; we differ from the majority of catalogues in ignoring any fraction of a gramme.*

As a rule our specimens are of many grammes. Indeed, the average of the individual weights of our 603 falls, after eliminating the great masses from the estimate, is, as given on page 105, about 4 pounds-nearly 2 kilogrammes each. A collection with so large a number necessarily includes many falls which were of small weight at the outset, and of which only the large museums have specimens, and these perforce very small-of a few grammes each. There is here no criticism to be made of the specimen being small, but congratulation on the fall being represented at all. In this feature of the size of the individual specimens it is evident that the smaller collections have opportunity for higher average. Entire boloidsmasses which have not been broken since they reached our earth, and are covered on all sides with the crust-are interesting as showing the treatment of the piece by aerial friction and heat action. And the larger they are the greater the surface on which such phenomena are registered. We have a few such entire boloids-notably Baratta, weighing 175 pounds and nearly two feet in length, with several much larger iron masses. In other instances we have specimens showing how small are some entire boloids when they reach our earth after the tribulations of the "middle passage." We have such meteorite integers of the Pultusk, Forest City and Estherville falls, which are but little more than a centimeter in diameter, and weigh but 2 or 3 grammes.t

Of some of these abundant showers we have several score of specimens of very different sizes. These are of highest interest as showing the breaking up of large masses in an early part of their passage through the air-belt of our planet. A single sample of a few grammes— which we possess of meteoric dust brought by Baron Nordenskiold from the snow-fields of Northern Finland is of high interest as probably showing the ultimate trituration of meteoric matter. In our large meteorite series are specimens which illustrate the phenomena of pitting, striation and furrowing of their external surfaces both among Aerolites (Baratta, Knyahinya, Tabory, etc.) and among Siderites, as Cañon Diablo, Glorieta, Youndegin and others. The inner features of the mass, Chondri (Allegan and Bjurbole), Veins (Farmington, Schönberg and Zavid), Breccias resulting from the reunion of distinct mineral or rock fragments (Parnalee, Mezo-Madaras, Fukotomi), and metamorphism analogous to that of our marbles (Tadjera) are shown in a diversity of specimens in this collection. Also the different iron structures are brought out in the Widmanstäten figures-octahedral, hexagonal, etc., alloys and inclusions, together with instances of curved lamellae (Glorieta, Toluca),

*Life is hardly long enough in our estimation to watch the scales in deciding whether one of our meteorites weighs 9170 grammes or 9170.01 grammes! An old catalogue of the British Museum notes its specimen of Rancho de la Pila as weighing 46,512.4 grammes. Can they weigh it a second time and get the same fraction?

+The smallest meteorite known, or strongly supposed, to have been a distinct entire fall (not one in a meteorite shower) is the Mühlau Aerolite, which was found at the village of that name near Innsbruck in the Tyrol in 1877. It weighs 5 grammes, and is sacredly preserved in the Royal Vienna Museum.

The deposits found at the bottom of the ocean by the Government exploring ship Challenger and described by Mr. John Murray are thought by him and by the astronomer Proctor to be the submarine equivalent of this meteoric dust, and alike of cosmic origin.

faults (Puquios), slickensides (Tennassilm), etc. We have made no enumeration of the score or more of Pseudo-meteorites-fragments of stone or iron purely of terrestrial origin which are from time to time brought forward as true cosmic bodies. These are not unfrequently enumerated in catalogues-even those of the great museums. We consider it a true misfortune that prominence should thus be allowed to the unreal, and that ancient blunders should be given a continued lease of life.

Within the alphabetical arrangement of the meteorites of this catalogue we have chosen the three main divisions first announced by Story-Maskelyne, and still continued in the catalogue of the meteorites of the British Museum of Siderites, Siderolites, and Aerolites; the former division including all these meteorites whose composition is almost wholly iron, more or less alloyed with nickel. Those in which silicates-notably Olivine, Enstatite and Bronzite-abound, with little or no iron as aerolites; while the siderolites stand as an intermediate group in which there is a mingling of metallic nickel-iron with stony matter. The former of these groups is the most constant in its composition as well as its structure; the latter is the least constantly and sharply defined. We have given to each meteorite fall a letter-symbol indicating its position in a taxonomic classification. The detail of this classification will be found on pages 97-103. It is the latest expression of Dr. Brezina of Vienna on this subject. The system is essentially that published in his catalogue of the Vienna Museum meteorites in 1896, with its groups based on structural peculiarities augmented by some groups newly found or newly determined. Of the former is (12) Leucituranolite, based on the Schafstädt aerolite (fell June, 1891) and lately described by Professor Klein of Berlin; (43) Crystalline Enstatite Chondrite, based on Hvittis, fell 1901; (62 and 65) on the alike new falls of Kodaikanal (India) and N'Gourema in the Soudan. Among groups based on new determinations are (27) veined black chondrite-Farmington-separated from black chondrite; (44) Mezosiderites and (45) Grahamite have been separated from each other. The Hexahedrites and the Ataxites have been rearranged according to numerous researches of Cohen and Brezina, and new definitions have been given for them. A number of meteorites have changed their places in the system according to fuller researches on better material—a thing which is likely to continue in the future. It probably can be claimed by no system of meteorite classification that it has further value than a measure of adaptability to bring together falls of generally similar structure and appearances. Analysts and petrographers have still imporant work to do here. It is to be hoped that they may employ some more natural and less empirical bases for classificatory purposes. We have shown on page 104 how the present collection represents all of Brezina's 74 meteorite groups, with 95% of all the falls.

NOTEWORTHY SPECIMENS

Turning over the pages of our catalogue, we find not a few score of meteorites which present points of especial interest. First among the siderites, Arispe-the Sonora Iron of late (1888) discovery--besides its important size, has special interest in its tripartite structure. A section of the mass shows three areas with differently orientated series of kamacite bands showing distinct centers of structural growth. Our main slice is the type specimen of a description of this iron. Another iron from West Africa presents a feature superficially similar which has been the subject of two memoirs by Professors Berwerth and Brezina of Vienna and Professor Cohen of Greifswald. The former describes four distinct areas of

this iron as due to the twinning of a gigantic crystal. Our series of specimens of Cañon Diablo is very large, from small, thin, sharp-edged nuggets to masses of several hundredweight each. The largest mass, weighing 383 kilogrammes, has two holes several inches in diameter passing directly through the mass. Several of the other masses have these holes, which were doubtless once filled with cylindrical nodules of Troilite. Indeed, one most interesting specimen contains the Troilite filling still remaining at the bottom of a half-emptied hole. Sections of the Bella Roca iron, as also the Toluca, show alike large Troilite inclusions, while the Australian Youndegin has the deep concavities and bores quite the counterpart of Cañon Diablo. In like manner are inclusions of Schreibersite profusely present in our slices of Chupaderos and Tombigbee River irons. In the latter, the sulphid shows itself through the mass in zigzag lines strongly suggesting Hebrew characters.

Ballinoo, of which we brought the main mass from West Australia, is the only iron which presents two zones of alteration—the outer one shining, the other dull. This and Tazewell, of which latter we have a handsome slab, have the added and most exceptional feature of showing dodecahedral lamellae besides the octahedral ones. There are several pieces of Glorieta, one of them a slice with curved lamellae, a feature which shows better here than in any other meteoric iron. The other is a lengthened mass of flattened cylindrical shape and weighing about 2 kilogrammes, which has upon its lower surface distinct shallow cavities about 1 centimeter in diameter, filled with a pale yellow Olivine. The Puquios iron (first brought by us from Chili) shows a clear faulting in some of the kamecite bands. One large slice of Casas Grandes-the great mass of which is in the National Museum at Washingtonis a prehistoric iron found in a cave with mummied objects in the State of Chihuahua, Mexico. Other irons in the collection are Charcas, State of Luis Potosi, Mexico, and Victoria on the Saskatchewan River in British America, both of which have been objects of worship by the indigenous people within historical times. The oldest iron, and indeed the oldest well authenticated meteorite, is Elbogen, which was known from early in the fifteenth century. Of this we have a piece, as also of Brannau, which was seen to fall in 1847, and through the study of which Widmanstädt first called attention to the structural figures which have since borne his name. Among siderolites we may notice several unusually large slices of the Brenham Pallasite, with the olivine-filled cells about equaling in volume the iron net-work. Of the Siberian Pallasite Pavlodar (Jamyschewka) we have the largest known piece, with a still larger piece of Marjalahti, a Finland congener which fell two years ago on the west shore of Lake Ladoga. One of the rarest pieces of the collection is a piece weighing one kilogramme of Veramin, a celebrated meteorite in the possession of the Shah of Persia. Finally, we have a series of nearly fifty pieces varying in size from 5 grammes to 10 pounds of the Estherville, Iowa, meteorite.

AEROLITES.

Of the aerolites we have among our 333 localities many which are of especial rarity or notable from structural or mineralogical interest. Noticing them alphabetically, Baratta, obtained two years since from the place of its fall in Australia, is the largest piece of its fall and one of the largest of aerolites, being nearly two feet long, and is crusted and pitted over its entire surface. It is also noteworthy from the very different sizes of its abundant chondri. Bjurböle, from Finland, is noteworthy from the great size of its chondri, which are of marked

fibro-crystalline structure and are loose in the matrix. Ensisheim is the oldest of recorded aerolite falls-1492. Ergheo is a brecciated chondrite from the northeastern corner of Africa Somali Land. Farmington, the second greatest Kansas meteorite, is represented by a large slab in which are well seen the fissures which, as has been suggested by Preston, have been filled at a later period with veins of black molten metallic matter. Hvittis, a Finland meteorite of recent fall, is interesting from its unusual per cent of the mineral Oldhamite. Indarch is the largest and heaviest known piece of this or any other of the limited group of carbonaceous meteorites a noble crusted mass, weighing over 18 kilogrammes. It is accompanied by all the other members of the group, five in number, including among them a magnificent mass of Mighei, also unique in size. Kesen, a well crusted and deeply pitted meteorite, is interesting as a stone which was given sacred honors for many years in a Buddhist temple. MacKinney, a black chondrite, is a piece of nearly a hundredweight. Of Ness County, Kansas, we have many pieces, all handsomely covered with a thick crust. Of Nobleborough the rarest American aerolite-we have a large piece, with shining black crust. The Russian diamond-bearing meteorite Novo urei is represented by a handsome specimen. Of Pipe Creek we have the largest mass, weighing nearly 4 kilogrammes. Of the interesting meteorite Saline, we have a noble slice, as well as an outside crust. Professor Farmington, describing this meteorite in Science, notices its structure, a veined spherulitic chondrite, as allied to Werchne Tschirskaya (Russia) and Trenzano (Italy), both of which, like Saline, fell in midNovember on the date of the Leonid star showers. We note further that Bath Furnace, Kentucky, of which we obtained the main mass, is also a veined chondrite and fell on the same date (15th of November) in 1902. Also, of the Russian meteorite Tabory (Ochansk; see cut on title page) we have two masses of several kilogrammes each, one well crusted.

Finally the Lujan, from Buenos Ayres, which is the only recorded instance of an undoubted geological meteorite.

In closing we enumerate thirty meteorite falls—about equally divided between Irons and Stones of which the largest single piece or part in any museum is now in the Ward-Coonley collection.