<5SO Cn i^ 4- ' V \ , LI B HARY OF THE U N IVERSITY Of ILLINOIS 550.5 FI RE^ C°P STORAG % 1 »,••• Explanation of Plate XLII Long Island meteorite as at present restored. X I- From a photograph. The position from left to right of the page is that probably assumed by the meteorite in falling. Field Columbian Museum Publication 64. Geological Series. Vol. I, No. 11 METEORITE STUDIES— I. BY Oliver Cummings Farrington, Ph. D. Curator, Department of Geology. Chicago, U. S. A. May, 1902. METEORITE STUDIES— I. By Oliver Cumaungs Farrington. LONG ISLAND, PHILLIPS COUNTY, KANSAS. MUSEUM NUMBER Me. 420. Nearly all of this great meteorite is possessed by the Museum and this has been the case since the opening of the institution in June, 1894, but it has never been fully described. A few lines were devoted to the meteorite and a cut of it shown in the catalogue of the meteorite collection published in August, 1895.* A petrographic description from fragments of the stone was also given by E. Weinschenk in i895-t No account of the finding of the stone seems ever to have been published however and there are many other features which are well worthy of description. For details regarding the occurrence of the stone I am indebted to Prof. Williston of the University of Kan- sas and Prof. Willard of the Kansas Agricultural College. Prof. Williston states that a fragment of the meteorite first reached him in the fall of 1892. Prof. Willard secured one at about the same time. On recognizing the meteoritic nature of the fragments sent them, Profs. Williston and Willard at once entered upon negotiations for the purchase of the mass and soon became its possessors. The work of collecting the pieces at the original locality was done by Prof. Willard, and to him I am indebted for information regarding the occurrence there. The meteorite lay, he states, on a slope of the ordinary soil of the upland prairie region. There is no outcrop of rock in the immediate vicinity and none within several miles, so far as he knows. Where there is an outcrop the rock is limestone. The distribution of the pieces of the meteorite as first seen by Prof. Willard was such as to indicate that the mass had struck upon the slope and its front portion being stopped, the rear portion had broken up and gone ahead. The four large pieces which are put together to make the mass shown in * Field Columbian Museum Publication 3, p. 59. t Tscb. Min. u. Petr. Mittlt., Vol. 14, p. 471. 283 284 Field Columkian Museum — Geology, Vol. I. Plate XLIII (Frontispiece) were together and in contact at the upper end of the fall. The top of these projected about four inches above the soil and the lowest point to which they reached was perhaps two feet below the surface. Beside these large pieces a quantity of smaller fragments more or less imbedded in the ground extended down the slope in a northwest direction for a distance of from 15 to 20 feet in a gourd-shaped area which was perhaps six feet wide at the widest point. The accompanying section and plan (Fig. 1) from a sketch by Prof. Willard will give an approximate idea of the manner in which the fragments lay. The location of the spot where the meteorite was found is about three miles west of the present town of Long Island, Fig. 1. Section and plan showing nature of occurrence in place of the Long Island meteorite. one-half mile east of the west line of Phillips County and three miles south of the Kansas-Nebraska State line. It is from the neighboring town of Long Island that the meteorite takes its name. With regard to the time of the fall no knowledge has yet been obtained. The stone was noticed by early comers to the region and was generally reputed to be a meteorite, so that visitors had in many cases taken away pieces as curiosities. That the mass had lain a number of years in place is proved by the coating of carbonate of lime, in some places two or three millimeters in thickness, which encrusts many of the pieces. Further evidence of the long exposure of the stone is given by the weathered character and rusty brown color of the surface of exposed fragments of the stone in contrast to the dark green color of their interior. The meteorite as collected by Prof. Willard was May, 1902. Meteorite Studies, I — Farrington. 285 shortly afterward purchased by Mr. George F. Kunz, of New York City, and after remaining in his possession for about a year was secured for this Museum. The entire weight of the meteorite as received at the Museum and made up of 4 large and 2,930 small fragments, was 1,184 pounds (537 kilos). This was supposed at the time to be the entire weight of the mass, but a year or two later Mr. Kunz obtained about 60 pounds (27 kilos) more, which is for the most part still in his possession. This additional material was chiefly fragments obtained from people in the region who had carried off portions of the stone for curiosities. A weight of at least 1,244 pounds (564 kilos) can therefore be posi- tively assigned the stone and there is little doubt that it originally weighed somewhat more than this since some pieces were probably carried off that will never be recovered. That the fragments all belonged to a single mass the manner of their occurrence in place leaves no doubt. Moreover their edges show no rounding or fusing as would have been the case had any of them made an independ- ent passage through any considerable part of the earth's atmos- phere. The stone is therefore much the largest single stone meteorite known to exist, its nearest competitors being the Bjurbole meteorite, which weighs 748 pounds (340 kilos) and one of the stones of the Knyahinya fall, which weighs 649 pounds (295 kilos). As soon as the installation of the stone was -undertaken at the Museum, it was at once seen that the four large pieces fitted together. When this was done the form shown in Plate XLIII (Frontispiece) was produced. Doubtless others of the fragments could be added to these, but as an effort to do this proved on trial to be likely to con- sume considerable time without giving any important results, the attempt was abandoned. There would be more hope of success if the Museum possessed the entire mass of the stone, but as it is, many of the fragments would be missing at best. The four large pieces weigh together 669 pounds (303 kilos), or more than half the weight of the stone. They hence probably give its essential form. Their weights are 269, 239, 89^ and 71^ pounds (122, 108, 46 and 32 kilos) respec- tively. The largest of the remaining fragments at the Museum weighs 22^ pounds (10 kilos), which is a weight much below that of the smallest of the four large ones. Mr. Kunz informs me that one of the fragments in his possession weighs about 35 pounds (15.8 kilos). The smaller fragments range from the weight above mentioned to those not over a gram in weight. Some have the true meteorite crust on one surface, showing that they are from the superficial portion of the stone, while the rough, irregular surfaces of the remaining frag- 286 Field Columbian Museum — Geology, Vol. I. ments show that they were wholly within the interior. The portion of the stone to which these fragments would be attached if a complete restoration of its form could be made would be that to the rear and to the left of the part shown in Plate XLIII (Frontispiece) or to the rear of the stone in the position in which it is shown in Fig. i, Plate XLIV. As will be seen by referring to Plate XLIII (Frontispiece) the portion of the stone to the right of a vertical line drawn through the middle of the plate has an almost wholly natural surface. Over this portion therefore the actual form of the stone is preserved. The form of the stone as at present restored is, as shown by the plate, roughly that of a low cone. The greatest diameter of the base of the cone is 34 inches (86 cm.) and the altitude from base to apex 20 inches (51 cm.) The conical form, as is well known, is the typical one to which meteorites are reduced in their passage through the atmosphere, from the fact that the portion of the mass in front receiving the 'brunt of the friction and heat is worn down rapidly to an apex from which the other portions slope away. That this is the position which the Long Island stone took in falling is further indicated by the smooth, unpitted character of the base of the cone (Ruckseite) as compared with the pitted surface of the conical portion, and further by the fact that the series of pittings (piezoglypten) on the surface extend in radial directions from the apex of the cone. It will be noted in the plate that the long axes of the pits run in directions nearly parallel to lines drawn from the apex to the base of the cone. These then were the directions of the air cur- rents. The planes along which the four large fragments were sepa- rated and along which they have now been joined together are not courses of ordinary irregular fracture, but are definite divisive planes. There are three of these planes, two being continuous each in its own direction while the third may be described as made up of two planes meeting at a very broad angle (1600). The planes run in three direc- tions nearly at right angles to each other. They meet, but only at one point do they pass through one another. If one will conceive of an apple cut in halves by a plane starting a little to one side of the bloom, one of these halves then cut through equatorially in a direction at right angles to the first plane by two planes starting a little above the equator, but meeting at it, then the quarter nearest the bloom cut through by a plane at right angles to the equatorial plane in a direc- tion running from the bloom to the stem, and passing into the other- wise uncut half for quite a distance, an idea will be gained pi the course of the division planes of this meteorite. Their course can also be seen by reference to Plate XLIV. May, 1902. Meteorite Studies, I — Farrington. 287 The area of each plane is approximately as follows: Plane A = 200 sq. in. (13 sq. dm.) Plane B = 196 sq. in. (12 sq. dm.) Plane C = 113 sq. in. (7.1 sq. dm.) The position of these planes makes it unlikely that they were developed by the blow of the meteorite in striking the earth, for one at least runs nearly at right angles to the probable direction of Fig. 2. Slickensided surface of Long Island meteorite. X 1. motion of the meteorite. Further, as stated more in detail below, the striae of the slickensided surfaces run in different directions. The plane marked (A) in Fig. 2, Plate XLIV, runs quite nearly in the direction of probable motion and it is interesting to note that near each end of the meteorite irregular cracks appear which are approximately parallel to this plane. Their position suggests that they may have been produced by the tendency of the base of the meteorite to continue its motion after the apex had been stopped by striking the earth. The plane marked (C) separating Pieces 2 and 4 can be noted continuing on in Piece 1 as a line which extends nearly 288 Field Columbian Museum — Geology, Vol. I. to the edge of that piece. This portion of the plane evidently was not sufficiently developed as a division plane to produce disruption of the piece when the meteorite struck the earth. That the three planes described represent a structure which existed in the meteorite before its entry into the earth's atmosphere there can be little doubt. They are too regular to make it possible to consider them planes arising from fracture by shock and there are several other lines of evidence pointing to their preterrestrial existence. The most important of these is that their surfaces are slickensided. The slickensided character of the surface resembles that seen in terrestrial rocks and is illustrated in Fig. 2. It is a smooth, shining, somewhat undulatory, like a roche moutonnte surface, and bears short striae which on the same surface run in one general direction, but take different directions on the three several planes. These several directions are indicated in Plate XLIV, Fig. 2, where one of the fragments is represented as removed. The color of the slickensided surfaces is somewhat darker than that of the crust of the meteorite, but there is no evidence of special heat having been developed by the force which produced the slickensides. This I have tested by cutting sections at right angles to the surfaces. The outlines of the individual grains were found to be sharp and unaltered up to the slickensided edge. Since slickensided surfaces on terrestrial rocks are so far as known produced by slow differential movement in the mass under considerable pressure and while in the solid state, they may in the absence of any evidence to the contrary be assigned to the same cause in this meteorite. The conclusion seems fair therefore that these planes and surfaces were formed during the preterrestrial existence of the mass and that the mass must, have been solid in its nature while in space. The three planes which I have described seem to me to resemble the joint planes of terrestrial rocks more than anything else I can think of and give us grounds for asserting the existence of joint structure in the rocks of space. I do not know that well marked joint structure has been observed in any other meteorites except that noted by Meunier in one of the stones of L'Aigle*. This stone he regarded as possessing a joint fissure, but it was not as well developed as the planes of the Long Island stone. If the occurrence of joint structure in the Long Island stone is deemed proved, it is significant as pointing to a considerable mass possessed by the body in space. Joint blocks of such size as this would not be likely to be developed in a small body. *Fremy's Encyc. Chimique, Tome II., Meteorites, p. 457. OJ ri^Ml!*. Ji.llj a Explanation of Plate XLIV. Fig. i. Drawing of restored portion of Long Island meteorite to show posi- tion and number of dividing (joint) planes. The position of the meteorite is as if it had been rotated upward from that shown in Plate XLII1, bringing the apex of the cone to the front. Fig. 2. Drawing of restored portion of Long Island meteorite in slightly different position and with Piece 2 removed, to show direction of striae on dividing (joint) planes. FIELD COLUMBIAN MUSEUM. GEOLOGY, PL. XLIV. A- Fig. 1. Fig 2. LIBRARY UNIVERSITY OF ILLINOIS URBANA May. 1902. Meteorite Studies, I — Farrington. 289 The natural surface of the more conical part (Brustseite) of the meteorite as it is at present joined together, is for the most part deeply pitted with characteristic meteoritic thumb-marks (piezoglypten). These pits vary considerably, as would be expected, in form and size, but still exhibit a certain uniformity. The majority have the form of an elongated ellipse whose major axis is about twice the length of its minor. The following dimensions may be considered as representing a fair average of the size of the pits: Major axis, 3.2 cm. (i^in.) Minor axis, 1.5 cm. (^8 in.) Depth, 3 to 10 mm. (}i to }i in.) The depression of each pit generally slopes uniformly toward the center of the ellipse, but often there are to be found pits, the deepest point of which is quite eccentrically placed and which have a more or less con- ical shape. Some pits have a nearly circular outline as contrasted with the more common ellipsoidal one. These circular pits are usually of small size, but one of large size and unusual depth is to be found at the point in the meteorite where the two planes A and C cut each other. This pit has for the most part the shape of a deep regular bowl, although the regularity of one portion is broken by two smaller conical pits. The depth of this pit is 3.2 cm. (i^in.) and its diameter 6.4 cm. (2^4 in.) The point of junction of the planes is almost exactly at the center of the pit. It is evident that this was a point of weak- ness in the stone at which the erosive action of heat and friction pro- duced during the passage of the mass through the atmosphere worked more rapidly than on other parts of the surface. Its occurrence at the point of junction of the planes is pretty good evidence that the latter existed in the stone previous to its entry into the atmosphere. This fact has also a bearing on the disputed question as to the origin of the pits in general. It shows that they owe their origin chiefly to an excavation by heat and pressure of the softer or more friable parts of the surface of the mass which is acted upon. Wherever there is a point of weakness there a pit will be formed. Vice versa, where a pit is formed, there was a point of weakness. The rear side (Ruckseite) of the stone is not pitted. It has a well developed crust, but the encrusted surface exhibits no marked depressions or elevations. The only portion of the meteorite as now restored which illustrates the Riickseite is that appearing in the upper right hand part of Plate XLIII (Frontispiece). Here the surface is slightly undulating, but there are no pits. The color of the crust of the meteorite is in general dark brown, but varies from almost black to light brown. At a little distance it appears perfectly smooth and in places shining, but on close examin- ation it is seen to be quite uniformly and coarsely stippled by the pro- 2qo Field Columbian Museum — Geology, Vol. I. trusion of the more resistant grains. In many places, especially in the vicinity of the pits, minute thread-like markings appear over the sur- face, sometimes in parallel and concentric series, but more commonly in arborescent forms which are often quite elaborate. These series or systems of markings do not appear to run in any common direc- tion, but are differently oriented wherever found. I have noted no system more than one inch (2.5 cm.) in length, but several of about this extent. They resemble closely the lines of flow such as have been noted on the crust of the Stannern and other meteorites, and doubtless are of this nature, being formed by a minute portion of the substance of the meteorite becoming momentarily fused and flowing in a diversified path until cooled. Their course in some cases seems to mark the swirling of the same air currents which formed the pits. More extensive and larger ridges are to be observed over some por- tions of the crust. Three nearly parallel appear on the portion of the Kiickseite just mentioned. Each is continuous for a length of from 3 to 5 inches. These do not appear to be of the nature of the lines of flow above mentioned, but more nearly resemble the veins which stand out on some meteorites and probably mark a line of more highly re- sistant constituents. Sections cut at right angles to the crust and examined with the microscope exhibit little if any alteration on the crust surface. The mineral outlines seem to be continued sharply up to the edge, and except for a certain smoothness of contour a crust surface could not be distinguished microscopically from the surface of an interior portion. Occasionally a metallic grain protrudes from the general outline, but so far as the contour as a whole is concerned it appears to be the result of erosion rather than of fusion. The weathering which the mass has undergone since its advent upon the earth has affected it considerably. Even the larger frag- ments when broken open will be found to be deeply invaded by rust which has penetrated along cracks in every direction. Doubtless the great number of small fragments into which the stone was found to be broken when first discovered was due to this process of sepa- ration through weathering rather than to shattering caused by the blow of the mass upon the earth. The weathering has affected chiefly the metallic constituents of the stone, causing their oxidation, and this rust has penetrated and stained the meteorite deeply. The color of the weathered, surfaces has thus been changed from the dark green of the unaltered rock to various shades of brown, a characteristic color being a light yellowish brown, almost white, spotted with dark or rust brown. The depth to which this discoloration has extended, except where May, 1902. Meteorite Studies, I — Farrington. 291 it has followed cracks and fissures, is usually scarcely a millimeter, the color changing beyond this through reddish to black before the dark green of the unstained stone is seen. Over a large part of the surface of the stone as found appeared a white amorphous coating which adhered very firmly. It could be removed by treatment with weak acid and most of it has been taken off in this way since the arrival of the stone at the Museum. When its substance is examined chemically it is found to be carbonate of lime containing a small percentage of clay. There can be little doubt that this coating is derived from the calcareous soil in which the stone lay for an unknown period, the carbonate of lime from the soil doubt- less spreading over the meteorite surfaces through capillary attraction and cementing upon the stone some of the surrounding clay. In some cavities of the stone a much greater proportion of soil is held, and at many points the cementing agent is iron oxide, derived doubt- less from the oxidation of the metallic grains of the meteorite. The unaltered stone when exposed by fresh fracture is of a dark green color, varying to black, although the latter shade may be due to staining from terrestrial oxidation. The stone is fine-grained, tough and compact. Occasional portions exhibit a slight porosity, giving a slag-like appearance. Such areas are however small and the pores of small size. The proportion of metallic ingredients is not large but they are quite uniformly distributed. The metallic grains show most plainly on a polished surface, the distribution and quantity being illustrated in Fig. 3. Occasionally well-marked aggregations of these may be seen. None of the surfaces that I have examined show arrangement of the grains in lines or systems of lines such as have been noted in a number of stone meteorites by Reichenbach* and Newtonf. The largest metallic grain I have seen in the Long Island meteorite has a diameter of 1.5 mm. From this size all gradations may be found down to the minut- est grains, examination with a lens bringing out many not visible to the naked eye. The bronze-yellow color and comparative softness of many of the grains as exhibited on a polished surface mark them as troilite, in con- trast to the silver-white color and greater hardness of those composed of nickel-iron. Further identification of the grains can be obtained by isolating them or by treating a polished surface of the meteorite with copper sulphate. On the polished surfaces examined the number of troilite grains is evidently much in excess of those of nickel-iron. ♦Ueber das Gefflge der Steinmeteoriten. Pogjjendorff' s Annalen 1859, vol. 108, pp. 291-311. + .\mer. Jour, of Science, 1893, 3rd ser., vol. 45, pp. 152-3. 292 Field Columbian Museum — Geology, Vol. I. As individual grains they are, however, smaller in size. Often the nickel-iron and troilite can be seen to be intergrown in a single grain. Before the blowpipe a fragment of the rock fuses even in the oxidiz- ing flame with a fusibility of about 4.5, the entire fragment blackening from the formation doubtless of FeO. In the reducing flame the fusibility is, as would be expected, greater on account of a more rapid formation of FeO. Evidently the mixture of minerals forms an Fig. 3. Polished surface of Long Island meteorite showing size and distribution of metallic grains. Fully half of these are troilite. The light irregular lines mark the position of fractures of no significance. X y. aggregate fusible at a lower temperature than any of its components, for the component minerals are practically infusible. The specific gravity of the stone, determined as an average of three separate portions weighing 50, 18 and 7 grams respectively, is 3-45- As previously stated, a description of the petrographic characters of the stone has been made by Weinschenk*, and as it seems desir- able to collect here all the important literature bearing on the mete- orite, a translation of Weinschenk's article follows: "From the Long Island, Phillips County, Kansas, occurrence I *Loc. cit. May, 1902. Meteorite Studies, I — Farrixgtox. 293 have at my disposal four pieces, amounting in weight to 20-30 grams. They possess a rusty weathered surface. Many hundred similar pieces were found (in part with crust), having a total weight of 1,184 pounds. The meteorite of Long Island is a compact, dark stone, which appears dark green on fresh fracture and shows numerous metallic specks. The crystalline structure is megascopically visible; there are numerous shining cleavage surfaces and the meteorite resembles the fine-grained harzburgite from Riddles, Oregon. Chondri are only now and then to be seen. Under the microscope it is clearly seen that chrysolite and bronzite are the characteristic ingredients. The structure as well as the relations in quantity of the two constit- uents are very variable, the chrysolite now being in excess and now again the pyroxene, and the general porphyritic structure passes com- monly enough over to a purely granular one. Chondrus-like forms are found throughout, but they are seldom developed in an especially characteristic way. Ragged particles of metallic iron, numerous grains of iron sulphide (troilite?) and chromite complete its compo- sition. The chrysolite occurs generally in porphyritic, more or less idiomorphic crystals, and in fragments. In the fresh condition it is colorless, but on slight heating it becomes reddish brown and pleochroic, and at red heat completely opaque, indicating a high con- tent of iron. The cleavage of the mineral is always clearly developed, and this shows in many cases undulatory extinction. It is very rich in inclusions, generally appearing as dark-brown rounded forms which often show regular arrangement. In the weathered portions there occurs beside iron hydroxide, a serpentine-like substance as an alteration product of the chrysolite. The orthorhombic pyroxene is likewise colorless and transparent and may be classed as bronzite. It tends to form groups of larger individuals where the stone has granular structure; in smaller crystals it occurs also as a constituent of the ground mass in the porphyritic forms. Its distribution in the stone can best be seen if a section is treated with hydrochloric acid. This dissolves out the chrysolite but leaves the pyroxene unattacked. In sections so treated it can especially well be seen that the bronzite where it occurs as a constituent of the ground mass often exhibits skeleton growths which lie imbedded in a colorless substance and are not attacked by hydrochloric acid. This has weak refraction and between crossed nicols shows irregular illumination so that it is not improbable that it is a glassy substance possessing optical anomalies through strains. Rarely, besides the orthorhombic pyroxene there is to be seen a monoclinic augite in single grains, with the properties of diallage. The solid iron occurs in angular particles and often in 294 Field Columbian Museum — Geology, Vol. I. zonal growths with chromite which also occurs alone, widely distributed in the stone. The little grains of the latter mineral appear brown, translucent. Also pyrrhotite (inagnctkies) is present in considerable quantity and generally in large individuals. The structure of the whole stone indicates a cooling from a fused liquid, a view also sup- ported by the porphyritic crystals of chrysolite and the skeletons of bronzite in the colorless base. There is no trace of breccia structure and the occurrence of few well-defined chondri gives no further proof. As has been often observed in meteorites, the whole stone has much more the character of a suddenly cooled mass, a character which is also indicated by the undulatory extinction of the chrysolite, the skeleton growths of pyroxene and the sudden variations in composi- tion. The Long Island meteorite in mineralogical characters belongs to the harzburgites. If among terrestrial rocks we look for masses which in a structural and mineralogical way can be compared to the Long Island meteorite, it will be found that the number is a. very limited one for the reason that rocks of similar composition have suf- fered in most cases much decomposition, by which their structure becomes indeterminable. But at all events it seems probable from the few observations on, for example, the terrestrial basalts of Green- land, that similar structures as they are here observed and in many other meteorites are formations characteristic of cooled rocks in which silicate of magnesia plays an important part, and that no grounds are given for the belief that formations of this kind in any of the terres- trial rocks have originated in any different way." To the observations of Weinschenk there is little of importance to be added. The crystalline structure is perhaps hardly as promi- nent megascopically as one would judge from Weinschenk's account, while the chondritic structure is easily recognized in all the sections I have examined. There are numerous polysomatic porphyritic chry- solite chondri and typical fibrous ones of enstatite. One of the latter observed was 2.5 mm. in diameter and it is evidently not cut through its center. A black, seemingly carbonaceous matter, borders its outer edge. The fibers are minute and lie in parallel groups extending in various directions. A porphyritic chrysolite chondrus seen had a diameter of 1.25 mm., a single grain reaching the size of .025 mm. Another monosomatic chrysolite chondrus seen was made up of chrys- olite porphyritically developed in glass and with a distinct circular border of chrysolite all extinguishing simultaneously. This chondrus also contained a large grain of troilite. The crystal outlines of the chrysolite individuals whether developed in the chondri or out are often well defined, the predominant habit being short stout crystals May, 1902. Meteorite Studies, I — Farrington. 295 bounded chiefly by pinacoids. The chromite more often has a red tone than the brown described by Weinschenk, its deep red grains being frequently seen in the sections. Both nickel-iron and troilite grains sometimes enclose small siliceous particles of what is probably chrysolite, indicating the latter to be the earlier formation. As regards classification, the Long Island meteorite is classed by Wulfing as a crystalline spherical chondrite, Cck.* Beaver Creek, Bethlehem, Lumpkin, Menow, Prairie Dog Creek, Richmond and Savtschenskoje are other meteorites included in the same class. Brezina classifies Long Island as a crystalline chondrite Ck.,| in which group are included Erxleben, Klein-Wenden, Kernouve and many others. By Meunier, Long Island is put in Class 34, Erxlebenite, which includes monogenic meteorites of fine grain made up chiefly of chrysolite and bronzite and containing visible grains of nickel-iron. Bluff, Erxleben, Kernouve, Klein-Wenden, Menow and Pipe Creek are among the other meteorites brought by Meunier into this class. Thus the place of Long Island in classification seems to be quite generally agreed upon. Differences can, of course, be noted from other meteorites with which it is classed, it being, for instance, more compact and of finer grain than Beaver Creek and containing much less nickel-iron than Pipe Creek. Of its well-marked crystalline character, however, there can be no doubt, nor, to my mind, of its monogenic origin. Absorption by a siliceous magma, of iron in preference to nickel, seems to me to afford a reasonable explanation of the high percentage of nickel in the metallic portion of the stone shown in the following analysis. Such a high percentage of nickel in the nickel-iron of stone as compared with iron meteorites is common and must be of some significance. If the meteorite is simply tuffaceous in origin, one would expect the nickel-iron to have the composition of that of the iron meteorites uninfluenced by the accompanying silicates, but such is not the case. Again, the outlines of the crystal individuals in the Long Island meteorite are sharply and fully developed and are in stable and mag- matic position with reference to each other. Some of them are larger than the individual chondri and yet exhibit no sign of wear or fracture. Accordingly the believers in the tuffaceous character of all stone meteorites would find, I think, little to support their views in an examination of this stone. I can see no indications in its structure of any other origin than one of cooling in place from a fused magma, *Die Meteoriten in Sammlungen, Tubingen, 1897, p. 453. fDie Meteoriten Sammlung des K.K. Naturhistorische Hofmuseums, Wien, 1895, p. 3^3. 296 Field Columbian Museum — Geology, Vol. I. and this applies to the chondri as well as to every other part of the stone. In this view I agree fully with Weinschenk as quoted above. An analysis of the meteorite was made by Mr. H. W. Nichols, Assistant Curator of the Department, in the laboratory of the Department. For this analysis a fragment of the meteorite free from visible oxidation was pulverized and dried at ioo° C. A portion of 3.3863 grams was weighed out for the major part of the analysis, experiment having shown that better results could be obtained from portions of this size than from a larger portion of about 16 grams. The nickel-iron was separated by Eggertz's method of solution in iodine, the stone being found to be of too compact a nature to admit of magnetic separation even if the iodine method is not to be preferred in any case. The siliceous portion remaining was separated into two parts in the usual manner by treatment with dilute hydrochloric acid and potash. The separated portions were not weighed as it was found that sufficient oxidation occurred while burning off the filter to vitiate the results, but were analyzed separately and their weights calculated from the analyses. The insoluble portion was fused with sodium carbonate and a small amount of nitre. Silica was determined after the common method. Nickel and cobalt were separated from the iron by three precipita- tions and long digestion with ammonia and a large excess of ammonium chloride. Cobalt was separated from nickel by potassium nitrite. Nickel was titrated with KCn and cobalt weighed as sul- phate. Magnesium was weighed as pyrophosphate and calcium as oxide after an oxalate precipitation in the usual manner. Chromium was weighed as lead chromate after oxidation by bromine in acetic acid solution. Phosphorus was separated by the acetate process and weighed as molybdate. Water above ioo°C. was determined in a separate portion by Penfield's method of direct weight. Sulphur was determined in a separate portion as barium sulphate. Iron and insolu- ble alumina were determined in a separate portion, the iron being tit- rated by permanganate. The alumina was weighed directly, that of the soluble portion as alumina, that of the insoluble portion as phosphate. The alkalies were determined in a separate portion after separation by platinic chloride as usual. They were found to occur wholly in the insoluble portion. Ti02 was present in distinct although unweighable quantities. A precipitate of ammonium manganese phosphate also proved to be not quite large enough to weigh. A search for copper gave only a very faint brownish coloration with hydrogen sulphide which was not sufficient to verify. May, 1902. Meteorite Studies, I Farrington. The analysis gave the following results: 297 Metallic. Sol. in MCI. Insoluble. Total. SiO oil 26.54 35-65 AL(>3 . . ,.64 1.44 3 08 FeO . 17.19 5.66 22.85 WgO . 8.99 13 75 22.74 CaO 0.02 1 38 I.40 Na,0 . O.OO 0.25 O.25 K,0 * 0 OO 0.03 O.03 11.' I above 100 ° 1.52 .... I 52 TH >, . Tr. Tr. P . O 036 0.024 0 060 s . I .QOO 1 oii grams. This does not represent the entire amount, since of some stones Mr. Ward was unable to obtain exact record, but at least this amount has been found. The majority of these, so far as their place of find has been recorded, have come from the neighborhood of Franklinville, a village about five miles south of Ness City. The first one described by Mr. Ward, however, came from a place nearly twenty miles to the east of Franklinville, the exact locality being *Amer. Jour, of Science, 4th ser., vol. 7, p. 233. tAmer. Jour, of Science, 4th ser., vol. 9, p. 112. May, 1902. Meteorite Studies, I — Farrington. 303 given by Mr. Ward as Section 2, Township 20 S. , Range 21 W. The village of Wellmanville is not far from this locality and this aerolite may therefore be called the Wellmanville stone. For investigating Preston's hypothesis two lines of inquiry may be followed; (1) The probable course of the meteor and (2) the con- stitution and structure of the stones. (1) The probable course of the meteor: Starting from Franklin- ville and going in a general northwest direction the Kansada and Jerome stones will be found nearly in the same line at distances of seventeen miles to Kansada and thirteen miles further on to the locality of the- Jerome meteorite.* Thirty miles further on in the same line appears the Oakley find. (See Plate XLV.) That a meteor may have moved along this line dropping fragments as it passed is conceivable, although no observed shower has a greater length of distribution than sixteen miles. If these stones are from a single meteor the direction of movement was undoubtedly from southeast to northwest rather than the reverse, since the smaller fragments would undoubtedly fall first. f The Wellmanville stone is somewhat eccentric to this general course but it is quite conceivable that it may have come from the same meteor so far as the latter's course is concerned. This would give a length of distribution of forty-six miles, or, if the Oakley stone is included, of eighty-six miles. The Long Island and Prairie Dog Creek meteorites evidently lie far outside of this course, although the exact location of the Prairie Dog Creek find seems not to be recorded. Brezina gives it| as 39°3o' N., 99°o' W., on Sappa Creek, Decatur County, Kansas, which is an utterly impossible location. However, Prairie Dog Creek, Decatur County, may be assumed to be its approx- imate place of find. The Long Island and Prairie Dog Creek locations form then, as shown by Preston, in connection with the Jerome and Ness County stones, a parallelogram 117 miles long by 35 miles broad. This parallelogram extends in a north-northeast direction, a course at about right angles to that which I have just traced. It is * Dana, as quoted by Washington, states that the Jerome meteorite was found on the Smoky Hill River fifteen miles east of Jerome.— Amer. Jour, of Science, 4th ser., vol. 5, p. 447. tl do not know that attention has been called before to this method of deducing a meteor's course, but it seems evident as a matter of reasoning that the smaller fragments would reach the ground first since the greater momentum possessed by the larger fragments would carry them farther. Meunier is of the opposite opinion (Meteorites, p. 424), but as a matter of record in all showers of which I have been able to obtain statistics the larger stones are found at the farther end of the meteor's path. This was the case at New Concord, Orgueil, Lanc6 and Butsura. For the purpose of gaining further evidence on this point it is quite desirable that observers should in the future note .the weight of the stones in connection with their location when picked up after a meteoric fall. J Die Meteoriten Sammlung, etc., Wien, May 1895, p. 359. 304 Field Columbian Museum — Geology, Vol. I. over this area that Preston suggests a meteoric shower might have extended. But an extension of a shower over an area so large, espe- cially in width, would be quite unprecedented so far as present obser- vations go. It seems to me, therefore, from a consideration of the probable paths, that Long Island and Prairie Dog Creek must be regarded as of separate origin from Ness County, Kansada, Jerome and Oakley, but that the four latter may, if only the paths are taken into consideration, belong to the same fall. (2) The constitution and structure of the stones: On this point, unfortunately, little evidence is as yet at hand. As Preston remarks, the six finds are megascopically very similar. They arev all about equally oxidized and coated with carbonate of lime, indicating that they have lain about the same length of time in the soil, and are all of compact texture and possess about the same quantity of metallic grains except Oakley, which contains much more metal than the others. The crust of the large stones is, however, thin and dark- brown in color while that of the small stones (Ness County) is, except Wellmanville, so far as I am able to learn, thick and black. This would indicate some difference in the character of the stones, but per- haps not sufficient to warrant considering them different falls. From the point of view of structure, Weinschenk states that Prairie Dog Creek is sharply distinguished from Long Island*, for in Prairie Dog Creek the chondri are very numerous and make up the greater part of the stone, while in Long Island the chondri are obscure and the structure has a marked crystalline character. Taking into consideration, therefore, their distance from the other finds and their differences from each other, there seems to be good reason for regard- ing Prairie Dog Creek and Long Island as separate single falls. Among the remaining four, Oakley, Jerome, Kansada and Franklin- ville, to which should perhaps be added Wellmanville as distinguished from Franklinville, Oakley seems on the whole to possess a distin- guishing character in its larger quantity of metal. According to Preston's determination it contains 14.44% °f rnetalf, presumably nickel-iron, while the percentage of metal (nickel-iron) in Jerome is, according to Washingtonf, only 4.25%. Oakley is also of coarser grain and possesses more bronzite than Ness County, as I have been able to learn by comparing microscopic sections. Through the kindness of Dr. Washington I have also been enabled to compare a section of the Jerome meteorite with sections of Oakley and Ness County. Considerable differences are thus brought to light which make it very improbable that Jerome belongs to either of the ♦Tschermak's Min. u. Petr. Mitth. vol. 14, p. 474. \Loc. cit. r uH 10 MO:T*i- Explanation of Plate XLV. Map showing location of meteorite finds in counties of northwestern Kansas. X marks location of finds. That of Prairie Dog Creek is approximate only. m r Q o a o o m (0 z o r\ r- z n r r r n ■< c m » a m 5 < o 3 2 03 m > z z > z c (/) | m c o B X H J> r O 0 5 * -i z o X C CO o z \ 1 > 2 W 0 fc H O > i H - .. 4 . \ \ \ #1 \ \ ( \\ 5( «• A s 5; ( w \* -a r Z R 0 < z PI X 5 > z \Q > }% c V " J* 3 B 3 z m © (A H X ni D X J> I z 0 X H O > z S = z (0 0 > 2. 3 3 = 1 0 1 o 3 ** I •< l y 0 X 1 o m X X » O m 0 r 3 i- o o < X c r r 0 r ■o (0 M 0) ■ z r r < LIBRARY UNIVERSITY OF ILLINOIS URSA May, 1902. Meteorite Studies, I — Farrington. 305 other falls. Its structure is much more highly chondritic than that of either of the above, and the peculiarities of the chondri, which have been so fully described by Washington that they need not here be again enumerated, render them unique. There remain then only Kansada and Wellmanville to be compared as to structure with the other finds. Regarding these two, however, no further data can at present be obtained. No details as to their intimate structure have been published and I do not know the present whereabouts of the stones. It would not be surprising, when an opportunity for com- parison presents itself, to find that Kansada could with good reason be connected with either Jerome or Franklinville and Wellmanville with Franklinville. Still, each might prove to be a separate fall, for as may be noted, there is no inherent improbability in supposing falls to take place within a short distance of each other at different times. The falls of Homestead and Hartford, Iowa, were, for instance, separated only about thirty miles and the character of the stones was not very different. The interval of time was twenty-eight years. Castine and Searsmont were separated by only about twenty miles in distance and twenty-three years in time. These are quite similar stones. The Estherville and Forest City falls were distant not over sixty miles from each other and took place within an interval of eleven years. Here, however, the character of the stones was quite different. Doubtless many other instances of falls approaching near each other in space and time could be found by searching. The citing of even the above is, however, sufficient to lead one to the conviction that classing together into one fall meteorites found in different localities is a work that should be performed with caution. TOLUCA (LOS REYES), MEXICO, D. F. MUSEUM No. Me. 454. This meteorite was obtained for the Museum in the spring of 1897 from Mr. E. O. Matthews of the City of Mexico. It was brought him by some native Mexicans or peons who reported that they had found it some months before, at Los Reyes, while ploughing. This is all the evidence obtainable regarding the manner of its discovery. The meteorite is of the metallic variety (aerosiderite) and is a complete individual. Its weight entire is 43 pounds (19.5 kilos). Its form (illustrated by the accompanying cuts, Plate XLVI), is roughly that of a steep triangular pyramid whose greatest length is 24 cm. (9^ inches), and greatest width 15 cm. (6 inches). The sides of the pyramid are 306 Field Columbian Museum — Geology, Vol. I. deeply hollowed and rounded so that the contours of the mass are curved, and at one of the edges it extends out in the form of a thin wing. On one side near the base are two especially deep and well-marked pits side by side, one somewhat conical in shape, the other broadly concave. The diameter of the conical pit is about 45 mm. (1^ inches) and its depth 20 mm. (^ of an inch). The concave pit is about 63 mm. (2^ inches) in diameter and 12 mm. (}4 of an inch) deep. These pits (fully shown in Plate XLVI) probably mark areas of schreiber- site which were fused out during the passage of the meteorite to the earth. The surface of the meteorite is of a uniform dark brown color from oxidation, but the depth to which oxidation has penetrated is very slight, as the merest scratch with a file reveals the nickel-white color of the interior. The meteorite is not of the " sweating" variety and exhibits no tendency to further alteration. Fig. 4. Etched surface of Toluca (Los Reyes) meteorite, showing character of Widmanstfttten figures. X £. Its substance is tough and malleable to a high degree. It is medium hard, cutting with some difficulty with a hack-saw. It takes a good polish, a polished surface being of silver-white to nickel-white color. Relative to copper sulphate the meteorite is active. The iron has not been sliced, but a triangular area 63 mm. x 25 mm. (2j4 inches x 1 inch) was made smooth and etched with nitric acid. The surface etched easily and exhibited well-marked Widmanstatten figures which are shown in Fig. 4. Two other smaller surfaces were also etched on other portions of the meteorite. The figures of the meteorite show that it is to be classed with Brezina's group 46 (Octahedrite with lamellae of medium width) or Meunier's group 7 (Arvaite). The bands of the etching figures are not of uniform .IVJX 3T*J^ 10 HOTA»UJ9x3 A Front and side views of Toluca (Los Reyes) meteorite. X|£. FIELD COLUMBIAN MUSEUM. GEOLOGY, PL. XLVI. Fig. 1. Fig. 2. LIBRARY UNIVERSITY OF ILLINOIS URUAKA May, 1902. Meteorite Studies, I — Farrington. 307 width nor do they extend continuously for any great distance. They are of the type described by German writers as ** wulstige" (swollen). The longest one on the etched surface figured accompanying is n mm. (54) of an inch in length and its contour is very irregular. Only the two alloys kamacite and tsenite seem to be present. The former is iron gray in color and occasionally has a well-marked granular structure. The latter, filling the areas between the kamacite bands, is now more or less ribbon-like and now occurs in curvilinear areas. Much of it appears connected through the section, giving the impres- sion of a network in which the kamacite is imbedded. It shades to a bronze color as contrasted with the iron gray of the kamacite and is left standing in relief by the etching. Under the lens its surface appears very rough, the etching of the acid acting upon it more irreg- ularly than upon the kamacite. The only other mineral appearing in abundance in the meteorite is schreibersite, which occurs in long narrow bands or in irregular star-like forms. These areas are bounded by kamacite (swathing kamacite). Decomposition has taken place usually along the schreibersite bands, and these decomposed areas appear as dark marks on the etched surface. Troilite seems to be almost entirely absent from the meteorite. Only two minute nodules are to be seen on the surfaces which have been etched and the percentage of sulphur obtained by analysis cor- responds to a content of only 0.07%. The presence of cohenite is indicated by the carbon found by analysis, but it was not observed on the etched surfaces. An analysis of the meteorite was made by Mr. H. W. Nichols, the methods employed being briefly as [follows: Material for the analysis was secured by a boring made with a ^-inch drill. The amount of substance used was 2.4353 grams. In order to prevent loss of sulphur and phosphorus the borings were placed in a flask and first treated with fuming nitric acid, to which they remained pas- sive, and then hydrochloric acid was gradually added cold until solu- tion was complete. Sulphur was weighed as barium sulphate. Phos- phorus was determined by Eggertz's method as phosphomolybdate, the quantity being too small to allow of a magnesium pyrophosphate determination. Iron was separated by one ammonia and three basic acetate and one final ammonia precipitation. Manganese was sepa- rated by the sodium acetate method. Copper, cobalt and nickel were" precipitated as sulphides in acetic acid solution, cobalt and nickel separated by potassium nitrite and all weighed from electrolytic depo- sition. Carbon was determined in an independent sample by oxida- tion in chromic acid after the method described by Blair.* ♦The Chemical Analysis of Iron, 3rd edition, p. 136. 3o8 Field Columbian Museum — Geology, Vol. I. The analysis gave the following results: Fe Ni Co Cu Mn P C S Si Insol. 00.56 7.71 1.07 0.14 Trace 0.24 0.0 1 0.025 0.006 0.09 99.85 Omitting silicon and insoluble matter the analysis indicates that the meteorite has the following mineralogical composition: Nickel-iron (Fe, Ni, Co, Cu, Mn.) 97.98 Schreibersite 1.55 Cohenite 0.15 Troilite 0.07 9975 As the locality where the meteorite was found may be said in a certain sense to be in the vicinity of Toluca, it becomes an important question from the standpoint of the collector to determine whether the specimen is to be regarded a portion of the Toluca fall. Los Reyes is about forty miles (sixty-two kilometers) in a direct line east of Toluca. It is the little station at the southern end of Lake Tex- coco where the Morelos division of the Interoceanic Railway joins the main line, about twelve miles southeast of the City of Mexico. On the same line of railroad twenty-five miles from the City of Mex- ico is the town of Ameca-Ameca, where the find of another iron mete- orite has been reported by Castillo.* Castillo classes this iron with the Toluca meteorites,! and describes the "zone" of Toluca mete- orites as extending from Ameca-Ameca on the east to Xiquipilco in the valley of Toluca [on the west]. If Castillo is right in this conclu- sion the Los Reyes meteorite comes within this zone, as Los Reyes is some fifteen miles (twenty-three kilometers) nearer Toluca than Ameca-Ameca. Castillo unfortunately gives no description of the 'Ameca-Ameca meteorite by which its resemblance or otherwise to the known specimens from Toluca can be determined. He simply describes it as a " small nodule of meteoric iron found in the village [of ♦Catalogue Descriptif des Meteorites du Mexique, Paris, 1889, p. 3. t Op. cit., p. 11. May, 1902. Meteorite Studies, I — Farrington. 309 that name] and now preserved in the National Museum of Mexico. " If it is correct thus to group the Ameca-Ameca meteorite (and hence Los Reyes) with Toluca, a distribution of fifty or sixty miles at least must be conceded to this fall, Ixtlahuaca and Xiquipilco, the two local- ities in the Valley of Toluca where many of the Toluca meteorites are found, being ten miles farther from Ameca-Ameca than Toluca itself. It will be remembered that Fletcher, after a careful study of Mexican meteorites with especial regard to the supposed occurrence of wide- spread meteoritic showers,* reached a negative conclusion as regards the wide extent of such showers, this opinion being similar to one in d to such showers in general which he had expressed in an earlier paper. f According to Fletcher the distribution of the Toluca meteor- ites as they have been reported from J localities distant from Toluca was probably due to human agency. With reference to the Ameca- Ameca meteorite he states that "Ameca-Ameca is a town where there arc now iron foundries, and where ploughs, castings, smoothing irons, mill wheels and other articles are manufactured," to show that Toluca iron might have been carried there for manufacturing purposes. With regard to this report of the state of manufacturing enterprises in Ameca-Ameca I fear that the distinguished authority of the British Museum has been misinformed, for I have spent weary days in the town without having learned of the existence of such industry. The fact brought out by Fletcher to the effect that no known meteorite shower has a greater distribution than sixteen miles is a more important one in the study of this case, and the evidence at hand in this instance is hardly sufficient to enable us to assert that the Toluca shower had a wider extent. The meteorite may of course have reached Los Reyes by the agency of man, but on the whole the indications are that it fell where it was found. The statements of the finders were plain and simple, the meteorite bears no marks showing any attempt to use it for eco- nomic purposes, and the price at which it was purchased was lower than any one who had brought it from Toluca would probably have sold it for. If the iron fell where it was found it is important to determine whether it was an independent fall or whether its resemblance to known Toluca irons is sufficient to make it probable that it fell at the time of the Toluca shower. Here again no positive evidence is at hand, but the chances are, in my opinion, in favor of the latter con- clusion. The meteorite certainly does not differ sufficiently from known Toluca irons so that its independent origin can be asserted, *Mineralogical Magazine, vol. IX, No. 42, pp. 91-179. TMineralogical Magazine, vol. VIII, p. 225. 3io Field Columbian Museum — Geology, Vol. I. and on the whole it resembles them considerably. Published analy- ses of Toluca irons give percentages varying somewhat widely, within which limits the Los Reyes values may certainly be included. For purposes of comparison of analyses, several that have been made of Toluca irons by different authorities are given below: i. Taylor, American Jour. Sci., 3d ser. XXII. 374. 1856. 2 and 3. Pugh, Annal. der Chem. and Pharm. XCVII. 385. 1856. 4. Nason, Jour. Prakt. Chemie. LXXI. 123. 1857. Mn S C P X 0.18 0.63 = 100.46 Insol. residue. 0.03 O.24 O.34 = 99.88 X 0 . 20 o . 62 0.22= 09 . 06 Tr 0.376 2.225= 99-975 Insol residue. Tr. 0.025 °-01 °-24 O.096 99.85 The resemblance in chemical composition to the average of To- luca irons is thus seen to be close. Further, the etching figures come within the limits found in Toluca irons, since these vary considerably in detail as is well known. The meteorite will be designated, there- fore, as Toluca (Los Reyes). Fe Ni Co Cu I. 90.72 8.49 0.44 2. 90.74 7.78 0.72 0.03 3- 87.89 4- 90-I33 9.06 1.07 N v 1 7.241 .... Los Reyes 90.56 7.71 1.07 0. 14 HOPEWELL MOUNDS, ROSS COUNTY, OHIO. MUSEUM No. Me. 480. Among the objects obtained from the Hopewell Mounds of Ohio, and now in the Anthropological collections of this Museum, are a number made of iron. These include a part of a head and ear ornament, some celts, a number of beads, and lastly a small unwrought mass weighing about 130 grams (5 ounces). Dr. G. A. Dorsey, to whom I am indebted for calling my attention to them, informs me that they were all found associated with a single human skeleton near an altar of one of the mounds. They were considerably oxidized, so that the original metal is in most cases obliterated, but the unwrought mass above mentioned was found to be oxidized only on the surface. A quali- May, 1902. Meteorite Studies, I — Farrington. 3ii tative analysis of some filings from this mass showed the presence of nickel and indicated, as might be expected since no other source of iron probably lay open to the Mound Builders, that the objects were made of meteoric iron. Upon removing the rust from one surface and submitting the area so exposed to the etching action of nitric acid, the meteoric nature of the iron was proved beyond question by Fig. 5. Outline of Hopewell Mounds meteorite, with etched portions showing curving of the Widmanst&tten figures, probably due to heating and hammering that the mass has received. X %. the appearance of Widmanstatten figures. The nature of these fig- ures is shown in Figs. 5 and 6, where the structure of bands of kam- acite separated by thin ribbons of taenite can. be plainly discerned. The width and continuity of the kamacite bands varies consid- erably. Some are at least a millimeter in width and from these they grade down to not over twice the width of the corresponding taenite ribbon. While many are continuous in a general way for a length of from 10 to 20 millimeters, the taenite runs through them all in a series of anastomosing branches and in places gives the impres- 312 Field Columbian Museum — Geology, Vol. I. sion of a network in which grains of kamacite are imbedded. The contour of the figures is for the most part curved and wavy, especially near the borders of the section. The most reasonable explanation for this seems to be the treatment probably given the mass by the ancient workmen. If heated until it became somewhat plastic and then hammered, just such curving of the plates might be pro- duced. Owing to the distortion of the figures it is impossible to positively classify the iron. Apparently it is an octahedral iron hav- ing lamellae of medium width. While two alloys, kamacite and taen- ite, are plainly discernible, no troilite or schreibersite can be seen, Fig. 6. Etched surface of Hopewell Mounds meteorite showing Widmanstatten figures curved and interwoven, probably on account of heating and hammering that the mass has received. X 8. although the presence of the two latter is indicated by the percent- ages of sulphur and phosphorus found on analysis. At one end of the mass are three large irregular pores such as might have been pro- duced by the falling out of crystals of chrysolite or other stony mat- ter. There is no other evidence, however, that such stony matter was at one time present and the cavities may have been produced in a purely mechanical way. This seems rather the more probable from the fact that the rest of the mass is quite compact. The iron is rather soft, cutting easily with a hack-saw, and malleable. It is active to copper sulphate. For purposes of quantitative analysis a small piece was sawed from one end of the mass and cleaned from rust by filing and scraping. May. 1902. Meteorite Studies, I — Farrington. 3i3 The analysis, made by Mr. H. W. Nichols, and using the methods noted above for the Los Reyes meteorite, gave the following results: Amount of substance taken, 2,166.3 grams. Fe Ni Co Cu Mn Sn S P 95.20 4.64 0.404 0.035 trace trace 0.13 0.07 100.48 The other meteorites known to have been found in Indian mounds of this country are those of Octibbeha County, Mississippi, and the Turner Mounds, Ohio. In the Octibbeha County iron the quantity of nickel reaches 59.7%, and this sufficiently distinguishes it from any other known meteorite. The Turner Mound meteorites include masses from two different mounds, wThich were analyzed by Kinnicutt* with the following results: Fe Ni Co From Mound No. 3. From Mound No. 4 I. 2. 86.66 88.37 89.00 12.67 10.00 10.65 o-33 0.44 0-45 It will be remembered that Kunz concluded from a comparison of the Turner Mounds meteorites with those of Kiowa County, Kansas, that on account of the marked similarity in constitution and structure they belonged to the same fall.* The Hopewell Mounds are only about seventy-five miles distant from'the Turner Mounds in an east- erly direction, and it might be expected ^that the same meteoric iron would have been used for the construction of the objects found in these mounds. The results of the analysis above given do not, how- ever, permit this conclusion, the differences in the percentages being greater than are known to occur among the individuals of a single fall. Comparison of etching figures is out of the question on account of the distortion of those of the Hopewell Mounds specimen, but the lack of any-content of chrysolite such as characterizes the Turner Mounds ♦Reports Peabody Museum of Archaeology, vol. 3. p. 382. et seq. ♦American Journal of Science, 3rd series, vol. 40, pp. 316-318. 314 Field Columbian Museum — Geology, Vol. I. masses is a further point of difference. It seems impossible at pres- ent therefore to connect the Hopewell Mounds mass with any known meteorite and the specimen will therefore be designated as the Hope- well Mounds meteorite. TAENITE FROM THE KENTON COUNTY METEORITE. One of the sections of the Kenton County, Kentucky, meteorite in the Museum collection (Museum No. Me. 134), tends to decompose along the planes of structure marked by the Widmanstatten figures. The result of this decomposition is a separation of the mass into frag- ments bounded by octahedral planes, of a homogeneous alloy of iron- gray color between which lie thin, elastic plates of a tin-white color. The first alloy is undoubtedly kamacite, the second taenite. In order to compare this taenite with that known from other meteorites, some study of it was made. The fragments which it was possible to sepa- rate rarely exceeded four square millimeters in surface. As plates, they were thin, elastic and magnetic. The only feature noted regarding their surface was that it is often marked by parallel rows of minute ridges extending across the plate. Corresponding striations usually appear on the adjacent kamacite. The plates are soluble in copper ammonium chloride, and fusible with difficulty B.B. In separating plates for analysis care was taken to use only those which could be completely isolated and showed'no rust. This proved a laborious operation, and after considerable toil the amount that could be secured for analysis was only 0.022 grams. The analysis was made by Mr. H. W. Nichols. Iron was determined by titration with an n/100 potassium bichromate solution, and cobalt and nickel isolated by means of two ammonia and three basic acetate separations and then precipitated electrolytically. While the extremely small amount used for analysis makes the chances of error larger than is desirable, it is believed that fairly accurate results were attained. The analysis gave: Fe 80.3 Ni and Co 19.6 99.9 May, 1902. Meteorite Studies, I — Farrington. 315 This corresponds nearly to the formula Fe13 Ni., the percent- ages for such a compound being: Fe 80.5 Ni and Co 19.5 100.00 Other analysts have obtained for the percentages of iron in taen- ite from other meteorites, values ranging from 86.44% to 57-x^%» and corresponding to formulas varying from Fe6 Ni to Fe4 Ni„. Judg- ing from our present knowledge of alloys it is hardly to be expected that taenite should have a uniform composition. Kamacite, being the substance with the lowest freezing point, is to be considered the eutectic of the series and as such has a fairly uniform composition, but this would not be expected of the other alloy or alloys. INDEX. Page Acoustic phenomena of falling meteorites 9, 1 1, 14 Aerolites 18 Analyses 25 Classes 24 Definition 18 Description 23 Aerosiderites 18 Analyses 19 Definition 18 Description 18, 22 Aerosiderolites 18 Analyses 23 Definition 22 Description 18 Agates, Theory of formation of.. . 197 Age of the "Pillar of the Consti- tution" 254 Allen, J. A., quoted 200 Almy, John I)., Communications with 22 1 Altitude of Ixtaccihuatl, various determinations of 107 snow line on Popocatepetl. . . 92 Popocatepetl, Various deter- minations of ' 87 Altitudes near the Ranch of Tla- macas 00 Amaga\ Iron mining District of 131, 143 Town of 143 Amalfi, Mining District of 131, 137 City of 137 Ameca, Town of 78, 308 Ameca-Ameca, same as Ameca. Amphoterite 24 Amyzon shales 199 Anaime, Mercury ores of 169 Analysis of — dolomite 230 Hopewell Mounds meteorite. 313 inesite 223 Long Island meteorite 297 Page Analysis of — taenite 314 Toluca meteorite 308 Analyses of meteorites, quoted 19,23,25,310, 313 Anchitherium 185 Andesite, Amphibole, of Ixtacci- huatl 119 Hypersthene.of Popocatepetl 100 Anori, History of 138 Mining District of 131, 138 Antioquia, Department of 130 District of, West 141 Mining districts and ores of.. 130 Anza, Mining District of. 131, 141, 142 Aragonite 249, 254 Arsenopy rite, in needle form 147 Ascent of- Ixtaccihuatl 104, 105 Popocatepetl 82 Asiderites 18, 23 Astroccenia conica 215 Awaruite 19 Atlantosaurus 281 Baker, Frank C, quoted. .86, 105, in Ball.S. H 181 Bats, Distribution of in caves 148 Becker, G. F., quoted 265 Bibliography of Geology of Co- lombia 1 74 of Popocatepetl and Ixtacci- huatl 73 Biela's Comet 29, 30 Bird Remains 199 Anatidae 198 Fringillida? 200 Galla; 200 Blatchley, W. S., quoted. 247, 249, 253, 264, 265 Bradbury, Dr. S. M., Correspon- dence with 267 Breccia from weathering of pyr- rhotite 142 317 3*8 Index. Page Brongniardite, Occurrence in Co- lombia of 156, 157 Brontosaurus 276 Bronzite — of Long Island meteo- rite 293, 299 Ness County meteorite 301 Bryan, Wm. A., quoted 198 Bustite 24 Calcite, Crystal forms of 232 crystals, Coan's Cave 265 " Joplin 232 " " Differentiation of surface planes of 236 crystals, Joplin,etching figures 237 " " Forms of large crystals 233 crystals, Joplin, Forms of smaller crystals 237 crystals, Joplin, Forms of twinned crystals 238 crystals, Joplin, Localities of large crystals 232 crystals, Joplin, Type I.. .232, 233 " II.. 233, 234, 235 " " "III 237 " IV . . . . 238 deposited in quiet and mov- ing waters 266 Golden 229 twins, Joplin 238 " Guanajuato 240 in stalactites 249, 250 Camarosaurus 280 Catalogue of the Meteorite Col- lection 1 Caledonite 224 Blowpipe reactions of 225 Crystal forms of 225 Occurrence of 224 Capillary attraction, Influence of on forms of stalactites 251 Caramanta\ Mining District of. 132, 151 Carbonaceous meteorites 17, 27 Castillo, Antonio, quoted ' 308 Cauca, Mines of 127, 153, 155, 157 Cave Hill Cemetery, Marengo Cave 257 Caves, Observations on Indiana. . 247 Page Cenotes 248 Cephalopoda, Development of . . . 209 Chalcedony 195 Chassignite 24 Cluna River, Mining District of. . 165 Chladnite 24 Chondri 26 in Long Island meteorite 294 in Ness County " .... 301 Chondritic Structure 26 Chromite— in Long Island mete- orite . . 293, 294, 295, 297, 299, 300 in Ness County meteorite 302 Chrysolite, Absence of 313 of Long Island meteorite. 293, 298 of Ness County " 301 Circular Halls in Wyandotte Cave. 247 Form of 247 Origin of 247, 248 Structure of 248 Classes of Meteorites 18 Cleavage in Meteorites 22 Coal in Colombia 129, 143, 148 Coan's Cave 264 Bats in 249 Entrance to 264 Pool in 265 Collection of Ores from Colombia 125 Collections of Ores, Character of ideal 1 26 Collet, — ., quoted 253, 254 Colorado, Grand River Valley of. 267 Colorado ore, defined 132 Colombia, General Geology 128 Gold production 125 Mines of 121 Ores of 121 Physical features of ^128 state of geological science in. 125 Como Beds 267, 271, 276, 279 Compounds in meteorites 17 Contributions to the Palaeontology of the Upper Cretaceous Series 205 Cope, E. D., quoted. .181, 183, 185, 200, 281 Copper Ores in Colombia 141, 169 Crater of Popocatepetl . . 94, 95, 96, 97 Dimensions of 96 Ki'i \. 3*9 Page Crystal Cave, Joplin. Mo 234 Crystal forms of calcite 229, 232 caledonite 225 epsomite 228 gay-lussite 227 inesite 221 Crystalline structure of meteorites 19,21, 293, 295, 301 Crystals, experiments on growth of 265 Crystals from quiet and moving water, Characters of 266 Crust of Meteorites 1 Long Island Meteorite 289 N ess County Meteorite 300 Dakota group of the Grand River Valley 270 1 )a\ -ison, J. M., quoted 20 '•Diamond Dome," Marengo Cave 259 Dinosaur Beds of the Grand River Valley 267 Diogenite 24 Distribution of Bats in caves 248, 249 Meteorites 12 Sciuromorph rodents 187 Dolomite, California 231 used as Indian money 230 Dome-shaped Halls in Caves. 247, 253 Form of 247 Origin of 247 Structure of 248 Dorsey, G. A., On Mines in Co- lombia . 1 27 cpioted 230, 310 Dunite 24 Dust spouts in Mexico 87 Kchandia, Mining District of 153 Effects of heat on meteorites 16 Egg, Fossil 193 Measurement of modern 199 Elements in meteorites 16 Elliott, D. G., quoted 260 E psomite 228 Eroded stalactites in Shiloh Cave 262 Erosion in Marengo Cave 256 Erosive and solvent action of water in caves 248 Page Etching figures on Calcite .... 236, 237 of Meteorites 20, 21, 306, 311 Eucastor 184 Eutectic of meteorites 315 Farrington, O. C, author. Crystal forms of Calcite 232 Fossil Egg from South Dak. . 193 Handbook and Catalogue of the Meteorite Collection ... 1 Meteorite Studies, I 283 New Mineral Occurrences. . . 221 Observations on Indiana Caves 247 Observations on Popocatepetl and Ixtaccihuatl. Fasciolaria 214 Felix and Lenk, on Ixtaccihuatl.. 104 on Popocatepetl 83 Fissure systems, Wyandotte Cave 248 Fletcher, L., quoted 3, 309 Flos Ferri, Origin of 249 Foote, H. W., quoted 251 Fore Leg and Pectoral Girdle of Morosaurus 275 "Fortress Monroe," Marengo Cave 257 Fossil Birds 199 Anatidae 198 Fringillidae 200 Gallae 200 Fossil Egg 193 Origin 197 Resemblance to those of Ana- tine Birds 198 Fossils of the Grand River Valley 272 Upper Cretaceous Series 205 Fossil wood, Organic Matter in . . . 195 Franklinville meteorite. .302, 303, 304 Frias Mine of Guayabal 163 Mining District of 163 Frontino, Mining District of. . 131, 141 Fusus 215 Gamba/f. Pereira.on Colombian Ores 125 Gay-Lussite 226 Crystal Forms of 227 General conclusions upon the mines and geology of Colombia . 1 74 Gunnison River 268 Heddle, H. Forster, quoted 197 320 [ndex. Page "Helen's Dome," Wyandotte Cave 247 Heilprin, Angelo, quoted. .80, 106, 111 Account of ascent by 105 Heilprin Peak 112 Hess, Wm. H., quoted 249 Holmes, W. H., quoted 248 Hovey, H. C, quoted 254 Hyatt, Alpheus, quoted 209 Hystrix refossa 183 lbague, City of 167 Mining District of 167 Iddings, J. P., loan of crystals by. 232 Index to meteorites mentioned in Museum Handbook and Cata- logue 62 Indiana Caves, Observations on. . 247 1 ndian Money 230 Inesite 221 Analysis of 222 Blowpipe reactions of 221 Forms of 221 Formula of 224 Localities of 22 1 Iron Ores, in Colombia 143 Ixtaccihuatl, Observations on 67 Glacier on 111 Jaguas, defined 147 Jamesonite 148, 150 Jerome meteorite 302, 303, 304 Joint planes in Long Island me- teorite 286 Joplin, Mo., Calcite from 232 Jurassic Strata of the Grand River Valley 26c) Kamacite, Composition of 17 in Hopewell Mounds meteor- ite 311 " Kenton County meteorite. 314 " Toluca meteorite. 306 Nature of , 20 Kansada meteorite 302, 303, 304 Kenton County meteorite, Analy- sis of taenite from 314 Kermesite 1 52 Knight, Prof. W. C, Communica- tions from 227, 228 Kunz, Geo. F., Meteorites pur- chased from 8, 34 quoted 285 Pasre La Plata del Libano, Mines of,.. 165 Leaf Stalactites 262 Libano, Mining Districts of 165 Limit of Vegetation on Popoca- tepetl (;2 Loaiza Hill, Silver Mines of 153 Logan, Wm. N., author, Contribu- tions to the Palaeontology of the Upper Cretaceous Series 205 Long Island meteorite, Analysis of 297 Catalogue of eg Description of 285 Occurrence of 283 Los Reyes, Find of meteorite in . . 305 Location of 308 Ludwig and Soret, quoted 265 Magdalena River, Geology of 159 Mammoth Cave, Rotunda in 247 Manizales, City of 1 56 Mining District of 156 Map of Antioquia and Tolima. . 177 meteorite falls in Northwest- ern Kansas ^04 Popocatepetl and Ixtaccihuatl 120 Marengo Cave 256 Abundance of stalagmites in 257 Floor terrace in 256 Origin of peculiar stalag- mites in 257 Rate of growth of stalagmites in 261 Stalagmo-Stalagmites in 259 Stream deposit in 257 Mariquita, Ancient City of 159 Marmato, Mining District 153 Marsh, O. C, quoted 275 Matthews, E. 0., Meteorite ob- tained from 305 Measurements of eggs 199 Morosaurus bones 278, 279 Medellin, City of 140 Meek, F. W., quoted 211 Meniscomys hippodus 183 Menke, H. W., Photographs by 191, 267 Mercury ores in Colombia 169 "Mermaid," Marengo Cave 259 Merrill, G. P., quoted 249,251, 254 Index. 3^ Page Mesogaulus 184, 185, 186 ballensis 181 monodop 181 sesquipedalis . 181 Meteorites, Handbook and Cata- logue of I Studies of 283 Meunier, S., quoted 288, 295, 303 Milne, Sir Alexander, quoted 255 "Miln.y's Temple," Wyandotte Cue 217 Molecular arrangement of stalag- mites and stalactites 260 Molino ore 154 Money, Indian 230 "Monument Mountain." Wyan- dotte Cave 248 Moraine of Porfirio Diaz Glacier. 11; Morosaurus agilis 27; Fore Leg and Pectoral Gir- dle of 275 £rand"s 2-z,, 276. 278, 279 «mpar 2J- lentus 275 robustIM 275. 280 Morton, Xye F., on the ores of Antioquia , 1^0 "Mount Vesuvius." Marengo Cave 259 Mylagaulida* 181 Pliylogeny of 184 Mylaugaulua 181 Ness County Meteorites ^oo New Mineral Occurrences 221 Nichols, H.W.. Analyses by.. 296. 307 313. 3U Author. Ores of Colombia 121 Collections by 232 Fxperiments by 265 Nitrates in cave earths 249 < )akley meteorite ^o^, 304 Observations on Indiana Caves.. 247 Popocatepetl and Ixtaccihuatl 67 "Odd Fellow's Hall," Wyandotte Cave 2^1 Ores of Colombia 121 Organic matter in fossils 195 Origin of stalagmites in Marengo Cave 2;- ores in tuff 149 Page t feborn, IL F., quoted 276, 281 Ostrea 213 ! HstributJon of 213 beloiti 214 Packard, A. S., quoted 77, 86, 103 Palache, Chas., Communication from 239 Pamplona, City of 170 Mines of 170 Peale, A. S., on the Jurassic of the Gunnison Valley 267, 270 Pectoral girdle of Morosaurus. .. 275 Petrified wood. Organic, matter in 195 Petrography of Ixtaccihuatl 119 Long Island meteorite 293 Ness County meteorite 301 Popocatepetl • . . 100 Pliylogeny of the Mylagaulida?. . 184 Physical features of Colombia.. . . 128 Pico del Fraile 94, 99 "Pillar of the Constitution" 252 Age of. 254 Pirsson, L. V., quoted 240 Pits on meteorites 13, 289, 300, 306 Po (Indian money) 231 Popocatepetl. Ascents of 79, 82 83, 84, 85, 86 Description of 80 Observations on 67 Porfirio Diaz Glacier in Collecting ground of 112 Fvidences of former extent of 115 Middle ground of 113 Terminal portion of 114 " moraine of 115 Prairie Dog Creek meteorite.. 303, 304 Preston, H. L., quoted 302 Preslwich, J., quoted 250 Prionotropis woolgari 211 "Prison Cell," Marengo Cave 2;; Protogaulus hippodus 183 Pseudoperna wilsoni 215 Pyrargyrite 156, 158, 160, 164, 166 Quiuna, Gold mine of 142 Randolph, John C, on Colombia i?9. 164 Kcclus, Flisee, quoted. . . 143, 160, 173 32 2 Index. Page Remedies, City of 132 Mining District of 131, 132 Restrepo, Vicente, quoted 144, 162 Riggs, E. S., author, Dinosaur Beds of the Grand River Valley of Colorado 267 Fore Leg and Pectoral Girdle of Morosaurus 275 The Mylagaulida? 181 Robertson's Cave 258 " Rock of Gibraltar," Marengo Cave •. 257 Rodents, Distribution of Sciuro- morph 187 Rothroclc, H. A 253, 254 "Rothrock's- Cathedral," Wyan- dotte Cave 247 " Rotunda," Mammoth Cave ..... 247 " Sand Pit," Marengo Cave ...... 257 San Pedro Mining District. . . 131, 140 San Sebastian de la Plata 1 59 Santa Ana Mine, History of 162 Santana, Mining District of 163 Santa Rosa de Osos, City of 139 Mining District of 131, 139 Scaphites 207 Distribution of 207 nodosus 209 cequalis 207 Ontogeny 208 Paleontogeny 208 Phylogeny 208 ventricosus 211 warreni 209 " revised and- enlarged description of. 210 Scheiderz, defined 196 Schlemm, W..H 221 Schreibersitc 17, 19, 22, 307 Sciuromorph Rodents, Distribu- tion of 187 Scott, W. B., quoted 181 " Senate Chamber," Marengo Cave 247, 248, 252 Senft, F., quoted 249, 250, 254, 257 Shiloh Cave 262 Eroded stalactites in 262 Leaf stalactites in 262 Stalagmo-stalagmite from-. . . 259 Page Sierra de Ahualco 73, 75 Route to the Mountains of . . . 77 Former fauna of 76 " glaciation of 77 Time of origin of 76 Lavas of 76, 77 Sigmogomphius 184 Slickensides, in Long Island meteorite 288 in meteorites 27 Origin of 288 Snow Line, altitude of on Popo- catepetl k 92 Soledad, Mining District of 161 Specific gravity, of fossil egg v. 1*96 of inesite 223 of Long Island meteorite.... 292 of meteorites in general.. 19, 23, 26, 39 of Ness County meteorite 301 Stalactites, Eroded 262 Leaf ; 262 Molecular arrangement of. . . 260 Modified by capillary action. 251 Vermiform 249 Stagmalites, Definition of 261 Rate of growth of 261 Stalagmites, Abundance of in Marengo Cave 257 Bermuda 255 Origin of 258 Molecular arrangement of. . . 260 from Robertson's Cave 258 Variations in the form of 258 Stalagmo-stalactitcs 259 Stephanite in Colombia 1 56 Stibnite in Colombia 151, 152, 158, 161 Stratigraphy of the Grand River Valley 268 Stream deposit in Marengo Cave 257 Subcarboniferous strata in Tolima 129 Subcarboniferous limestone from Santander 129 Sucre, Mining District of 161 Sulphur Mines on Popocatepetl 89, 90 Supia, Mining District of 155 Syenite, Use of the term in Colom- bia 133 Index. 3^3 Page Taenite, Analysis of from Kenton County meteorite 314 Composition of 17, 315 Description of 20, 307, 312 Tellurides of gold in Colombia . . 141 Terminal Moraine, Porfirio Diaz Glacier 115 Terrace on cave floor 256 Thames, Rate of erosion by ..."*. 256 Thumbmarks of meteorites . . 14 Titiribi, City of 144 Mining District of 132, 144 Origin of ores of 149 Tolinia, Department of 159 Mining Districts and ores of. 159 Volcano of 160 Toluca meteorites, Analyses of, quoted 310 Distribution of 308 Triassic strata of the Grand River Valley 268 Troilite 17, 22, 291, 302, 307 Turner Mounds meteorite, Analy- ses of, quoted 313 Uncompahgre Plateau 267 Upper Cretaceous Series, Palae- ontology of 205 Venadillo, Mining District of 167 Vermiform stalactites 130 Vetas de cajon 130, 139 Yiviparus 272 Volcano of Popocatepetl 80 Page Volcano of Tolima 160 Ward's Natural Science Estab- lishment, Meteorites obtained from 8, 34 Ward, Henry L., quoted 302 Washington, H. L., Loan of sec- tion by 304 quoted 303 "Washington's Monument," Ma- rengo Cave 258 Weinschenk, E., quoted 292, 304 Weller, Stuart 209 Wellmanville meteorite. .303, 304, 305 Western Antioquia Mining Dis- tricts 131, 141 Widmanstatten figures, Descrip- tion of 20, 21 of Hopewell Mounds mete- orite 311 of Toluca meteorite 306 Weyman, Henry, on Crystal Cave. 234 White, Robert, on petrography of Colombia 129 Willard, J. T., quoted 283 Williston, S. W., quoted 267, 283 Wyandotte Cave, Circular halls in 247 Distribution of bats in 248 Fissure systems of 248 " Pillar of the Constitution" in 252 Vermiform stalactites in 249 Zancudo Establishment 144, 145 Mfcik Stiti Labowifn if litwl HUifj