ir^# «^ ^M :ibrarD of tl^ ffiusmm OF COMPARATIVE ZOOLOGY, AT HARVAP «IECE, CiHBRlDCE, MASS. The gift of tL. ^L/Jui. 'JhiAnjIl No.f(/Mp .■^^jip^ ,J{ji^_it,- 1^1(0 JOURNAL OF THE Elisha Mitchell Scientific Society VOLUME XII— PART SECOND J^ul^y==de:^oe:v[be:>p^ 18Q5 POST-OFFICE : CHAPEL HILL, N. C. ISSUED FROM THE UXIVERSITY PRESSES. CHAPEL HILL, X. C. 1895. TABLE OF CONTENTS. Notes on the Kaolin and Clay Deposits of N. C, J. A. Holmes 1 Description of Some Muscles of the Cat, H. V. Wilson and G. H. Kirby 10 Orig-in of the Peridotites of the Southern Appalachians, J. V. Lewis 24 Monazite, H. B. C. Nitze * 38 Table of Contents of First Twelve Volumes, 49 JOURNAL OF THE Elisha Mitchell Scientific Society NOTES ON THE KAOLIN- AND CLAY DEPOSITS OF NORTH CAROLINA. - BY J. A. HOLMES, CHAPEL HILL, N. C. As the Appalachian mountains reach their maxi- mum development in western North Carolina, we find also in that reg'ion indications of extensive dynamic dis- turbances and alterations uuderg-one by the rocks in connection with these mountain uplifts. Among- the minor results of these chang-es have been the formation of numerous dikes of "veins" of exceedingiy coarse granitic material>>, which in some places are mined for the mica which the\" contain, and in other places are quarried for kaolin. These dikes are filled with quartz, feldspar and mica, in varying- proportions, very coarsely crystalized. Sometimes the mica 'generally muscovitej, sometimes the feldspar fgenerall}^ albiteor orthoclasej, predominates. When the mica is present in considerable quantity and in large cr^'stals, the dyke is usually spoken of as a mica-vein, and is often worked for mica. Sometimes these crystals of mica occur on one side or the other, sometimes on both sides, * From the Transactions of the Am. Inst, of Mining- Eng-ineers, VoL XXIV, 1895. 2 JOURNAL OF THE and sometimes they are largely concentrated in the middle of the vein, or, it may be, distributed throug'h- out the dike with a considerable degree of uniformity. In many cases the crystals are too small and few to permit of the working of the vein as a mica mine; in other cases very little mica is present, and the feldspar constitutes the larger part of the material. This feld- spar of the dikes undergoes, throug-h the weathering action of the atmosphere, certain chemical chang-es re- sulting* in its alteration from feldspar into kaolinite — the kaolin of commerce. These dikes vary considerably in size, rang'ing from a few inches to several hundred feet in thickness, and up to many hundred yards in length. They are gene- rally parallel to the schistosity of the crystalline rocks, which, however, in some cases, they cross at various angles. The kaolin in those dikes which occur in the* Unaka or Smoky mountains is said to have been mined by the Indians, "packed" across the countr}^ to the seaboard, and shipped to England as early as the l7th century. From one of them, near Webster in Jackson Connty, kaolin is now mined (by the Harris Clay Co.) and shipped to Trenton, N. J. and other centers for the manufacture of fine pottery. This Webster dike con- tains very little mica and com.paratively little quartz. It has a maximum width of about 300 feet, and has been traced for a length of more than a half mile. It is mined to a depth of from 60 to 120 feet, below which the original feldspar has not been sufficiently altered, and is too hard for economic mining*. The kaolin is brought from the mine, crushed, and washed in a se- ries of settling-vats, for the purpose of separating it from the granular quartz. Its plasticity is increased ElvISHA MITCHELL SCIENTIFIC SOCIETY. 3 both by washing- and by the subsequent grinding- which it receives. The following analysis* of the washed and dried product ready for shipment shows the general character of such material. anaia'sis of kaolin, harris mixk, near webster, n. c. Per cent. Free silica, silicic acid and sand 2.28 Combined silica 41.62 Alumina .... 40.66 Oxide of iron 0.14 Alkalies ; 0.46 Lime none Mag-nesia trace Combined water 14.00 Moisture 0.84 Titanic acid none Total 100.00 Man}' simJdar but smaller feldspathic and kaolin dikes are found in the various other counties west of the Blue Ridge, and at a number of these the feldspar has been altered into kaolin for considerable depths be- low the surface, but none of them have been worked extensively for either the feldspar or the kaolin, except the Harris clay-mine just mentioned. Also at various points in the Piedmont Plateau, which extends east of the Blue Ridge for from 150 to 200 miles, there are to be found deposits of this kaolin which have doubt- less orig'inated in much the same way as those west of the Blue Ridge; but none of these are now worked to any considerable extent. The age of these crystal- line rocks in the Piedmont plateau and the mountain ^ Made for the Harris Clay Co. of Dillsboro, N. C. at the Pitts- burgh Pa. Testing Laboratory. 4 JOURNAL OF THE counties, and the exact time at which the disturbance took place which resulted in the formation of these massive g^ranitic dikes is, as jQt, a matter of doubt. So numerous are these dikes in certain places, and so long- have their feldspars been underg-oing- surface transformation into residual kaolin or clay, that one mig-lit expect to find in this reg'ion, as in some other countries, sedimentar}^ deposits of this material which had been transported for greater or less distances; but when we bear in mind the gfeneral elevation in the mountain-reg-ion and the consequent rapidity of its streams, we can readily understand that this product of denudation would scarcely be deposited until it had been carried so g-reat a distance from the original source as to be lost by comming-ling- in the lowlands with larg-er proportions of other and different materials. Along- the borders of the Piedmont plateau-reg-ion there are occasionally found deposits of this kaolin material which has evidently been carried but a short distance. Such occurrences are more extensively known on the western border of the Coastal Plain reg'ion, mainly in the Potomac formation, as in the neig-hbor- hood of i^iken, S. C, and Augusta, Ga., and in many other places, where considerable deposits of this kaolin- material occur, both in the form of arkose (where the kaolin is still mixed with the quartz and mica of the original granitic formation) and in the clay-beds, where it has been more completelv sorted, and the kaolin has been separated from the coarser materials, so as to form extensive beds of what is locally termed "china"- or potters-clay. In some cases, in the arkose material just referred to, the partially decayed crystals of feld- spar are frequently- found with the kaolinization in- complete; and mingled with these are fragments of ELISHA MITCHELL SCIENTIFIC SOCIETY. 5 other minerals, transported from the debris of the crys- talline rocks occurring along- the borders of the Pied- mont plateau, not man}^ miles away. The points above noted ma\^ explain, perhaps, the confusion which has arisen in the use of the term "ka- olin." The applicability of this name to the material described above as having- its orig-in directly in the larg-e g-ranitic dikes, I suppose no one will question. But if the residual material of dike-decomposition has been transported a short distance by the streams and deposited without further sorting- the materials, or if it has been transported to a much g-reater distance, so that the sorting- has become fairly complete, and the mineral kaolinite, while separated from the quartz and mica of the original mass remains unmixed with other foreig-u materials, so as to be itself fairly pure, — the question arises whether the term kaolin is still appli- cable in both cases; and if so, to what extent, in its transportation and sorting-, this material may become mixed w^ith other foreig-n materials resulting- from the decay of crystalline rocks in the reg-ion throug-h which it has been transported, before the term kaolin svould become inapplicable. In other words, where, in such a case, should we discontinue the use of the word "ka- olin" and apply the broader term "clay"? Further discussion of this question cannot be attempted in this paper; but it is mentioned here because the writer has recently heard a number of complaints from practical potters who use the clay-material on a commercial scale, that many people throug-hout the countr\^ were designating- all the samples of their material forward- ed as "kaolin," regardless of their color and other characteristics. Through many places, both in the mountain- and the b JOURNAL OF THE Piedmont plateau-reg'ions, there are deposits of clay resulting' from the decay of g-ranites, o"neisses and crystalline schists. Many of these have a structure which would indicate that the materials have been transported for g-reater or less distances. But in, per- haps, many other cases, the materials have evidently decayed in place, since the g-radations can be traced from the cla}^ down into the partly altered rock below. These clays, of course, var}^ in composition with the character of the rocks from which they have been formed. They have frequently a reddish or yellowish color, due to the oxides of iron present, thoug-h in many places the colors are much lig-hter, the iron having* been re- moved througfh the action of org-anic matter. As will be seen from the above statement, these may be classed as partl}^ residual clays and parLly transported clays. They have been worked on a small scale in many plac- es for brick; and in a few places, as at Biltmore (Bun- combe County) and at Pomona (Guilford County) they have been used in the manufactvire of tile-, drain-, and sewer-pipes; also at Pomona for lire-brick; and near Grover (Gaston County) for fire-brick and vitrified or paving-brick. The ag*e of these transported clays of the mountain- and Piedmont plateau-counties is unknown. Some of them, upon careful investig-ation, may be shown to be- long* to certain definite recent g-eologic periods; but most of them, probably, cannot be ascribed to any def- inite g'eolog'ic time, but must be attributed simply to local conditions; and their ag*e is probably recent. The clay and brick-loam deposits alon^- the river terraces of the mountain and Piedmont counties which, in many places, are well adapted to the manufacture of brick, may be Columbian or older in ag-e, ELISHA MITCHELL SCIENTIFIC SOCIETY. ( Those residual clays of these regions which have been formed /fi situ are the result of the processes of decay, the operation of which cannot be limited to any definite epoch, but may be ascribed, in g-eneral, to re- cent g-eolog-ic time. The most extensive beds of cla\' known in North Carolina are those found in the Coastal Plain reg-ion. In the Potomac (lower Cretaceous; formation, there are extensive beds of laminated, dark-colored cla3's, exposed along- the banks of rivers crossing- the coastal- plain reg-ion, notabh' on the Cape Fear river, for fifty miles below Fayetteville. These clays are usually dark in color, owing- to the vegetable matter which they contain; and, in some cases, the}" are hig-hly lig-nitic. The thin laminae are frequentlv separated by still thin- ner parting's of sand; and frequently within a short distance (from a few feet to a few hundred feet) the clay-laminae become thin and disappear, while the sand-parting's g-radually thicken, so that the whole as- sumes the character of a sand-bed instead of a clay- bed. This feature, which indicates plainly the shift- ing- conditions under which these deposits were laid down in certain localities, illustrated in the accompa- nying- sketch of the river bluff at Prospect Hall on the Cape Fear river, 21J miles below Fa3"etteville. In some portions of these clay-beds, pyrite occurs in such quantities as would probabl}' interfere with their industrial use; but the larger portion of the deposits appears to be free from pyrite, and will probabl}' prove to possess considerable economic value. Thus far no efforts have been made to utilize them; but both ana- lytic and practical tests of them are being- made at the present time. Along the western border of the Coastal Plain re- "7 M rt (U ■'■ rt "J. ."^ :l 1 rt ^ J -" - -zf. A * a; '^ 7) >^ CJ »" ^ o -J :=H GO p: « be O ^1 00 ^ O Ji •/i' o •n <3-> X ^ ^ ^ rt o 5 1^ ^ ^ 5^ > I 13 1 c' 0) a> rt P:! ElvISHA MiTCHELIv SCIENTIFIC SOCIETY. 9 gioti, especially in Moore and Harnett Counties, there are limited exposures of silicious Kocene deposits (over- lying- the Potomac series, and capping- some of the sand-hillsj which have recently been tested for fire- brick with very satisfactory results. These deposits are from 5 to 15 or more feet in thickness, and are over- laid b}^ but a few feet of loose sand. The following analysis of this material, collected two miles N. E. of Spout Si)rings," shows its general composition. ANALYSIS OF "fire-clay" ( EOCENE) TWO MILES N. E. OF SPOUT SPRINGS, N. C. Silica 87.70 Alumina 3,29 Ferric oxide 2,81 Lime 0.48 Magnesia 0,40 Alkaline chlorides 1.48 Loss on ignition 3.15 Total 99.31 Among the Miocene deposits, there are, in places along- the river-bluffs of the Coastal Plain region, es- pecialh^ on the Roanoke and the Tar, somewhat exten- sive exposures of "blue marl," a calcareous clay ^vhich may prove to be of some value, but of which no prac- tical tests have yet been made- The Lafayette (Pliocene; materials, which are spread over so large a portion of the Coastal Plain region, are g-enerally gravell}^ or sandy in composition, with a large admixture of loam in many places. No extensive deposits of clay have been observed among- the mate- rials of this formation, though doubtless limited depos- its of cla}^ will be discovered as more extensive explo- rations are made. ^ Made in the laboratory of the N. C. Geological Survey. 10 JOURNAL OF THE The later deposits bordering- the river-courses and covering- the river-terraces, at elevations from 20 to 100 feet above sea-level, which may be desig-nated as the river phase of the Columbian formation, contain extensive beds of sandy clays and brick-loams; and, throug-hout the entire Coastal Plain region of the South Atlantic States, it is these deposits which are most largely used m the manufacture of brick. A DESCRIPTION OF SOME OF THE MUSCLES OF THE CAT. BY H. V. WIIvSON AND G. H. KIRBY. The most complete account extant of the cat's mus- cular system is that found in Strauss-Durckheim's An- atomic du Chat^ The expense of the work, however, renders it inaccessible to many. And in spite of the g-reat beauty and general accuracy of the plates, it is not well adapted to the purposes of the dissector, ow- ing chiefly to the length and detailed character of the descriptions (e. g. psoas-niuscles and quadratics lum- horum) and in a less degree to what we regard as the misleading subdivision of the muscles of certain groups (e. g. fcctoral gxo\y^). With due deference to such a magnificent work, the results of dissection lead us to differ from Strauss-Durckheim in certain points of de- tail, to some of which attention is called. Perhaps the most generally useful work on the cat is that of Mivart^ . The descriptions and accompanying 1 Strauss-Durckheim. Anatomie Descriptive et Comparative du Chat. 2 vols, and Atlas. Paris, 1845. 2 St. George Mivart. The Cat. New York, 1895 (latest issue). ELISAH MITCHELL SCIENTIFIC SOCIETY. 11 figures enable the dissector to identify without trouble most of the muscles. But this part of the work, at least, contains too many inaccuracies to make it a safe guide. In like manner the value of a recently pub- lished hand-book^ is much impaired by inaccurate state- ments regarding- the origin and insertion of many of the muscles described. Finally, that most original and valuable work, the "Anatomical Technology," only includes a description of the muscles of the chest, shoulder and fore-leg. The accurate nature of all the descriptions in the "Technology" has long been recognized, and the few points in which our description of the pectoral muscles differs from that of Wilder and Gage*, are doubtless varying points. In spite, however, of its ac- curacy, the account given in the "Technology" of the cat pectorals seems to us an unnecessarily difficult one to follow. This is in part due to the manner in which the group is subdivided. ABDOMINAL MUSCLES. External oblique. This muscle arises along its ex- ternal border from the nine last ribs and from the lum- bar fascia. The attachments to the ribs interdigitate with the slips of the serratus niag-nus, and lie beneath the latissirmcs dorsi. The lumbar fascia from which the muscle arises is the superficial fascia which may be peeled from the underlying transversalis abdom- inis, and mav be traced to the spinous processes of the vertebrae. Along its inner border the muscle passes into a broad, 3 Gorhain and Tower. A Laboratory Guide for the Dissection of the Cat. New York, 1895. 4 Wilder and Gag-e. Anatomical Technology as applied to the Domestic Cat: New York and Chicago, 1882. 12 JOURNAL OF THE thin aponeurosis, which is closely bound to the subja- cent rectus abdo?ninis, and unites with its fellow of the opposite side along- the li7?ea alba. This aponeurosis extends from the sy?nphysis pubis, to which it is at- tached, forwards to the level of the ninth costal carti- lag"e. The anterior part of the aponeurosis underlies (is dorsal to) the most posterior pectoral muscle. The posterior and external edg-e of the aponeurosis, extend- ing- from the symphysis pubis obliquely dorsally and posteriorly, is known as Pouparfs lig-ament. Near the symphysis the aponeurosis is perforated by an ap- erture, the external abdominal ring"^ which is the ex- ternal opening" of the iiig-uinal canal. This aperture lies between Pouparf s ligament and the rest of the aponeurosis. Synonymy. Oblique-interne abdominal, S.-D., vol. II, p. 314; external oblique, M., p. 141; external oblique, G. & T., p. 28. Mivart is wrong" in stating* that the muscle arises from the eig-Jit last ribs; and that Poupart's lig-ament extends from the symphysis pubis to the Ilium. The latter statement is true of human anatomy; but in the cat the lig-ament (or free edg-e of the aponeurosis) passes in front of the ilium, having- no connection with it. Gorham and Tower commit the same mistakes. The account g-iven of the muscle both by Mivart and Gorham and Tower would indicate that Poupart's lig-ament is something- distinct from the tendon or ap- oneurosis of the external oblique. Mivart in this mat- ter, is not intellig-ible. He states that the aponeurosis divides into external and internal tendons, between which lies the "external abdominal ring-," bounded in front by Poupart's lig-ament. Possibly the "in front" is a misprint, for the lig-ament lies behind the abdomi- nal ring-. ElvISHA MITCHELL SCIENTIFIC SOCIETY. 13 hiternal oblique. The muscle arises along- its ex- ternal border from the superficial lumbar fascia, di- rectly beneath the origin of the external ohlique\ and from the ventral margin of the ilium. At its poste- rior end the muscle arises from the pubis. The fibres run anteriorly and ventrally. At some distance external to the rectus abdominis, they pass into a thin aponeurosis, which throughout its anterior extent is firmly attached to the underlying (/. e. dorsal) trajisversalis abdonmiis, and is thus dorsal to the rec- tus abdominis . Throughout its posterior extent the aponeurosis extends ventrally over the rectus abdom- inis, being- indisting-uishably united v^ith the aponeu- rosis of the external oblique. At its anterior end, near the dorsal border, the mus- cle is attached directly b\^ its fibres to the last rib. The muscle may here be seen to be a continuation of the layer formed by the interyial intercostals. The in- ter7ial intercostal between the 12th and 13th ribs, is in fact directly continuous with the internal obiique. Synonymy. Oblique-interne abdominal, S.-D., vol. II, p. 315; internal oblique, M., p. 142; internal oblique. G. & T., p. 29. Mivart is wrong- in stating the muscle to be "insert- ed inside the cartilag-es of the last ribs." Gorham and Tower inexactly state "the more anterior fibres to be inserted on the cartilages of the ribsS' According- to Strauss-Durckheim the relations of the aponeurosis of the internal oblique to the rectus are quite the same as in human anatomy, in that anteriorl}^ the aponeurosis splits into two layers which ensheathe the rectus dor- sally and ventrall}^ This ventral limb of the anterior part of rectus sheath, if it ever is present, must con- 14 JOURNAL OF THE sist of a very fe'v and delicate fibres. It was not ob- served in any of the several cats dissected by us. Transversalis abdominis. The muscle arises along- its dorsal border, just beneath the erectores spinae, from the fascia covering- the ventral surface of the lat- ter (middle la3xr of the lumbar fascia in human anato- my). It also arises from the cartilaofes of the false ribs and from the ventral margin of the ilium. Along" its inner border it is anteriorly, and for the g^reater^ part of its course, inserted into a broad, thin aponeurosis which lies dorsal to the rectus abdominis , and is continuous with its fellow of the opposite side as'^an independent aponeurosis. Posteriorly, however, the fibres ^of the muscle pass into a fascia, lying ven- tral to the rectus abdo7mnis, and inseparably united with the combined aponeuroses of the external and in- ternal oblique. Synonymy. Latitudinal, S.-D., vol. II, p. 317; trans- versalis, M., p. 142; tTansver sails abdoyyiinis, G. & T., p. 29. According to Mivart the muscle ends in an aponeu- rosisHying dorsal to the rectus. Gorham and Tower give the muscle as ending "in an aponeurosis beneath the rectus." Both authors thus overlook the import- ant difference in position between the anterior and pos- terior'portions of the aponeurosis. Strauss-Durckheim evidently recog-nizes this difference, thoug-h he does not explicitly state it. He designates the aponeurosis as the third layer of the abdominal aponeurosis, and goes on to say: "Ce feuillet — s'unit a celui de I'oblique in- terne, et se comporte du reste comme lui." Rectus abdominis. The muscle arises from the sym- physis-pubis. It is completely separated by the con- nective tissue of the linea alba from its fellow of the ELISHA MITCHELL SCIENTIFIC SOCIETY. 15 Opposite side; and is inserted on the first, second and third ribs, the chief insertion being- on the first. At about the level of the ensiform cartilage, the mus- cle g-ives off a small slip, which is inserted along- with its fellow of the opposite side, into the aponeuroses of the external obliques, in the median line, beneath the posterior (fifth) -pectorals. The rectus abdominis is ensheathed in the aponeu- roses of the two obliques and the transversalis. Pos- teriorly all these aponeuroses lie ventral to it. Ante- riorly the aponeurosis of the external oblique is ven- tral, but the aponeuroses of the internal oblique and transversalis are dorsal to it. Synonom3\ Droit-abdominal, S.-D., vol. II, p. 307; rectus abdominis, M., p. 142; rectus abdominis, G. & T., p. 29. The slip g"iven off in region of ensiform cartilag-e, is not mentioned by the writers cited above. It probably corresponds to the fibres said to be attached to the en- siform cartila^Jfe in human anatomy."^ * Quain's Anatomy, 10 ed., vol. II, pt. 2, p, 334. Our account of the rectus sheath differs from that of Strauss-Durckheim, in that we do not find the apon- eurosis of the internal oblique dividing- into ventral and dorsal limbs which embrace the rectus anteriorly, as already mentioned in connection with the internal oblique. Mivart, and Gorham and Tower, overlook the insertion of the posterior part of the transversalis, and thus g-ive an imperfect account of the sheath. PECTORAL MUSCLES. The pectoral muscles form a triang-ular mass on the front of the chest, running- from the median line of the body to the arm. The mass is divisible into ^\^ dis- 16 JOURNAL OF THE tinct muscles. These may be arranged in two groups* One group includes the first, second, and third ^^6^/0- rals, all of which muscles arise from the anterior por- tion of the sternum and run outwards. This group corresponds in a general way with the more superficial {P. major) of the two pectorals of man. The other group includes the fourth and fifth pectorals, which muscles in general pursue a course from the sternum outwards and anteriorly to the proximal end of the hu- merus. This group corresponds with the inner human pectoral {P. minor). In a general way it may be said that the muscles of one group cross those of the other. (On this crossing of the pectoral muscles, see Wilder and Gage, p. 235). The first pectoral \^ the most superficial of the group. It is rather narrow and band-shaped, and arises from the presternum and the raphe in front of the latter. The raphe in question is the median connective tissue septum between the posterior portions of the sterno- mastoids, and between the anterior portions of the sec- ond pectorals. The greater part of the muscle arises from the mid-ventral line of the presternum, along the anterior three fourths of that bone. Only the extreme anterior part arises from the raphe. Part of the muscle is inserted along with a shoulder muscle, the cephalo-humeral. Part is inserted into the fascia of the fore-arm, along with a slip from the third pectoral. Synonymy. Pecto-antehrachial, S.-D., vol. II, p. 352 (the slip from the third pectoral is regarded as the "second chef" of the pectoa7itel)rachial) 'pecto-a?ite- brachial, W. & G., p. 236 (above-mentioned slip is counted as the "caudal division of the pecto-antebra- ElylSHA MITCHELL SCIENTIFIC SOCIETY. 17 chiaV)', ^ectoralis, -part /., M., p. 145; pectoralis, -part a, G. & T., p. 30. There is no actual union, in cases at least, between the muscular part of the distal end of the slip from the third pectoral and the cpitrochlear, as W. & G. state (p. 238). The slip is in close juxtaposition to the epi- trochlear along- it's posterior border. The oriofin of the muscle, as g-iven by Mivart, is in- accurate: "from beneath the manubrium and attach- ment of the first two costal cartilages." Gorham and Tower g*ive the same orig^in as Mivart. The second pectoral is a wnder muscle than the first. It is superficial just in front and just behind the first, but the greater part of it is covered by the latter. It arises from the presternum and the median raphe, above referred to. The muscle arises from the mid-ventral line of the presternum along the entire length of the latter. The anterior portion arising from the raphe is noticeably thicker than the rest of the muscle. The chief insertion is on the outer surface of the hu- merus, along the middle third of that bone, external to the attachment of the third pectoral. The fibres ly- ing along the posterior border of the muscle are in- serted into the bicipital arch. Synonymy. Premier chef dii large pectoral, S.-D., vol. II, p. 343; lamina ectalis of the ectopectoralis, W. & G., p. 'z^2)S\ pectoralis, part j (partially), M., p. 147; pectoralis, part e (partially), G. & T., p. 31. Mivart's part j apparently only corresponds to the anterior portion of our second pectoral. M. is inaccu- rate regarding the origin, which he says is from the ma- nubrium: ''The ffth ^ " is the most anterior. It arises from the manubrium." The origin from the median 18 JOURNAIv OF THK raphe is thus overlooked. Gorham & Tower agree with Mivart. The third pectoral is a wide muscle. Anteriorly it is covered by the /^r^/ and second pectorals, but the pos- terior half or more is superficial. It arises from the above-mentioned raphe, from the presternum, and from the next two sternebrae, its orig-in thus extending- as far back as the attachment of the fourth costal cartilage to the sternum. Onh^ the extreme anterior portion of the muscle arises from the raphe. Behind these fibres, the muscle arises from the ventro-lateral surface of the presternum, along the whole length of the latter. The portion of the muscle arising from the next two divis- ions (sternebrae) of the sternum is superficial. The chief insertion is on the humerus, and is a very long one. The proximal part of this insertion is by a thin aponeurosis to the extreme outer surface of the humerus. The middle part is directly by muscle fibres to the ventral surface of the humerus. The distal part is again by aponeurosis to the ventral surface of the humerus. The most anterior fibres of the muscle are inserted into a small tendon which quickly divides. One divis- ion passes externally to the shoulder joint, and is at- tached to the outer end of the clavicle. The other di- vision passes internally to the shoulder-joint, and ter- minates in the fascia on the mesal surface of the scap- ula in the interval between the supra-spinatus and the sub-scapularis. The muscle is also inserted into the bicipital arch. As has already been mentioned, a slip lying along the posterior border of this muscle separates from it, and is inserted along with t\i^ first pectoral. Synonomy. Second chef du larg-e pectoral, S.-D., BLISHA MITCHELL SCIENTIFIC SOCIETY. 19 vol. II, p. 343; lamina oitalis of the ectopect oralis, W. & G., pp. 239-240; ^6^r/o;Y^/^, /^/r/ 2 (partially), M., p. 145; pectoralis, pari b (partially), G. & T., p. 30. Wilder and Ga^e divide the muscle into two subdi- visions, cephalic and caudal. This division seems to us an arbitrary one. We do not iind the muscle natu- rally so divided. Mivart's part 2 apparently includes our third pectoral and the posterior part of our second pectoral. The fourth pectoral is a w^ide, thick mass, being much the largest muscle of the group. ±\X its origin it is a single, undivided muscle, though distally it splits into two subdivisions, anterior and posterior, having different insertions. The muscle arises from nearly the whole sternum. The origin beg-ins just posterior to the presternum, and includes all the mesosternebrae and the anterior two-thirds of the xiphi-sternum. Of the two subdivisions into which the muscle splits distally, the anterior is inserted on the great tuberosity of the humerus, coming here into close connection with the supra-spinatits. The insertion of the posterior sub- division is complex. The chief part is inserted on the proximal half of the ventral surface of the shaft of the humerus, anteriorly by muscle fibres, posteriorly by aponeurosis. The fibres forming the posterior portion of this subdivision are inserted, along with the latis- sinius dorsi, into the bicipital arch. Into the posterior border of this subdivision run fibres belonging to the great cutaneous muscle, th.Q paii7iiciihcs carnosus. Synonomy. Sterno-trochiterien plus premier chef du grand pectoral, S.-D., p. 337, p. 341; ento-pecto- rali^, W. & G., p. 2\\, pectoralis, part j, M. p. 147; pectoralis, part c, G. & T., p. 30. 20 JOURNAI. OF THK Mivart ag'ain g-ives an inaccurate orig-in: "from the sternum between the second and sixth costal carti- lag-es." Gorham and Tower agree with Mivart. The fifth j^cct oral, the most posterior member of the g*roup, is long" and comparatively narrow. The mus- cle arises from the aponeurosis of the external oblique, which here forms a ventral covering- for the reetus ab- dominis. In the anterior portion of the orig-in, the muscle fibres arise directly from the median line, where there is raphe common to the several muscles of the ab- dominal wall. Posteriorly, however, the muscle fibres arise from the aponeurosis, along- a line which extends obliquely, in a dorsal and posterior direction, from the mid-ventral line. The muscle is inserted by aponeurosis both into the bicipital arch, and the proximal end of the humerus. The latter insertion includes both tuberosities. Synonomy. Seeo)id chef du grand pectoral, S.-D., vol. II, p. 341; xiphi-hiimeralis, W. & G., p. 2^^^ t^c- toralis, part 4, M., p. \M\ pectorals, part d, G. & T., p. 31. The details of the orig-in of this muscle are doubtless variable. Wilder & Gag-e state that the muscle fibres are connected to the median raphe by a thin, wide ten- don. Mivart is, however, entirely mistaken as to the orig-in. He g-ives the muscle as arising- "from the ster- num between the fifth costal cartilag-e and the root of the xiphoid." Gorham and Tower g-ive the same ori- g-in as Mivart: "from the sternum between the carti- lag-es of the fifth and eig-hth ribs." SOME MUSCLES OF THE HIND-EEG AND BACK. Tensor vag-inae femoris. The muscle arises from the anterior end of the ilium, and from the fascia ex- BlvISHA MITCHELI^ SCIENTIFIC SOCIETY. 21 tending- between this muscle and the anterior division of the g-hiteus maximns. This fascia is closely bound to the underlying- g-liiteits niedms, and is continuous with the dense fascia covering- the sacrum dorsally. The muscle is inserted by a fascia {fascia lata), which covers the external and anterior surface of the thigh, and dipping in between the vastus cxterniis and the adductor, is attached to the outer surface of the fe- mur, along- nearly its whole length. The tensor vag'iiiae near its origin, is inseparably connected (in cases at least) with the g-liitens viediiis. It is also directly continuous posteriorly with the an- terior division of th.Q gluteus maximus, the two actually forming a sing-le muscle. S3'^onym3^ Couturier et droit interne plus t\v^fas- cialis, S,-D., vol. II, p. 402, p. \^Z\ tensor vaginae femoris, M., p. 154; ditto, G. & T., p. 39. Gluteus maximus. The muscle is divided into two portions. The anterior arises directl}^ by muscle fi- bres from the transverse process (lateral mass) of the third sacral vertebra, and the transverse process of the first caudal vertebra, and from the fascia covering- the gluteus medius and the sacral region. It is in- serted into the femur, just below the great trochanter, by means of the extreme upper end of the fascia lata, and by a few independent fibrous strands, noticeable in old subjects. The posterior part arises directly by muscle fibres from the transverse processes of the first two caudal vertebrae. It is inserted into the fascia lata. It is only in young subjects that the two parts of the muscle are distinct at the orig-in. In older subjects, the muscle has a continuous origin from the sacral fas- cia, sacrum and caudal vertebrae. 22 JOURNAL OF THE S\^nonymy. Fesser plus the paranieral, S.-D., vol. II, p. 395; o-luteus niaximus, M., p. 154; g-luteus max- imus, G. & T., p. 39. The fesser of S.-D. corresponds to our anterior por- tion of the muscle, thoug-h S.-D. omits the origin from the third sacral vertebra. The parameral of S.-D. corresponds to our posterior portion. Here ag^ain we differ slig-htly from S.-D. regarding' the orig-in, which according' to him is from the second and third caudal vertebrae. The difference may easil}^ be due to var- iation. Quadratus lumborum. The muscle arises directly by its fibres, and also b\^ tendinous orig-ins, from the dorsal part of the anterior border of the ilium; and from the transverse processes of all the lumbar verte- brae. The fibres of the muscle run forwards and in- wards (mesially), and are inserted both directly and b}^ means of narrow tendons into the centra of all the lumbar and the posterior three dorsal vertebrae. This muscle exhibits a remnant of the metamerism, characteristic of the trunk muscles of the embryo and lower vertebrates, in that the narrow tendons are in- serted successive!}^ into the bodies of the vertebrae, thus imperfectly dividing- the muscle into myotomes. Synonymy. Ivongfs-sous-intertransversaires de lom- bes, S.-D., vol. II, p. 282; quadratus lumborum, M., p. 156. Psoas mag-nus. The muscle arises directly by its fibres from the dorsal part of the anterior border of the ilium; by a few fibres from the transverse process of the last lumbar vertebra; at the extreme anterior end, externall}^ from the aponeurosis covering- the quadratus lumborum; along- its inner border from the centra of the lumbar vertebrae. Over its dorsal sur- ELISHA MITCHELL SCIENTIFIC SOCIETY. 23 face the fibres of the muscle pass into the quadra- tus lumborum, with which muscle the psoas magnus is blended anteriorly, excepting* that part which arises from the aponeurosis covering- the former. The muscle is inserted on the lesser trochanter of the femur. Synonomy. Psoas, S.-D., vol. II, p. 406; psoas mag-nus plus iliacus, M., p. 156;, psoas magnus, G. & T., p. 42. The fibres arising from the anterior border of the ilium correspond to Mivart's iliacus. M. is wrong in stating that the muscle arises "from the transverse processes of all the lumbar vertebrae." Gorham and Tower make the same statement. Psoas parvus. This muscle may be regarded as a part of the psoas magnus, which has a separate inser- tion. It is indistinguishably united with the p. mag- nus throughout its anterior portion, and thus has with the latter a common origin. It is inserted by a tendon on the rim of the pelvis, at the ileo-pectineal eminence. Synonymy. Mipsoas, S.-D., vol. 2. p. 287; psoas parvus, M., p. 156. University of North Carolina, February, 1896, ORIGIN OF THK PERIDOTITES OF THE SOUTHERN APPALACHIANS." BY J. VOI^NEY IvEWIS. INTRODUCTORY NOTE. It was my privileg'e to spend the summer months of 1893, 1894 and a part of 1895 in mapping and studying in the field that portion of the Appalachian belt of per- idotites which extends across North Carolina, almost parallel to its western boundary, from Virginia to Georgia and western South Carolina. A number of excursions were also made to other portions of the belt in Georgia and Pennsylvania. The work was done in the preparation of a report on "Corundum and the Ba- sic Magnesian Rocks of Western North Carolina," which has just appeared as Bulletin No. 11 of the State Geological Survey. The Bulletin was confined to a presentation of the field results; the various rock types are described and the distribution and modes of occur- rence of the peridotites and corundum are given in some detail. Discussion of doubtful points, and especially of those already involved in controversy, was withheld for a future publication, in which the results of micro- scopic study and other laboratory work on these rocks would be presented, with such discussion of origin and other relations as these results might justify. A por- tion of this work has already been done, and some in- teresting results have been attained, but they will *Published by pennission of the State Geolog-ist of North Caro- lina, who also allows the use of some illustrations from plates pre- pared for Bulletin 11 of the Survey publications, ElylSHA MITCHELL SCIENTIFIC SOCIETY. 25 hardly be ready to put into systematic form for publi- cation before the close of the year. A number of points have been broug-ht out in the field-work, however, that are considered to have sufficient bearing* on the origin of the peridotites to warrant their presentation. THE ROCKS AND THEIR RELATIONS. The accompanying- map of the Appalachian cr3^stal- line belt (Plate I) shows the distribution of the perid- otites throug-hout this reg^ion. In Pennsylvania and Maryland the basic mag^nesian rocks, whether orig-i- nally peridotites or pyroxenites, have been entirely al- tered into serpentine; and hence have lost most of their orig-inal characteristics. The Virg'inia occurrences are not so well known. Rog-ers' Reports show that many of them have altered into serpentine and talcose rocks, and this is doubtless true of the g-reat majorit}^ of out- crops. Fresh olivine rocks, however, are found pass- ing* over into that state from Alleg^hany county, North Carolina, and they are doubtless to be found at a num- ber of places on the belt, south of Lynchburg*. With few exceptions, the peridotites south of Vir- g*inia are remarkably fresh, the alterations consisting*, in the majority of cases, of scarcely more than surface discoloration or an occasional local chang-e into massive serpentine. The country rocks of the reg*ion are g-neisses and mi- ca schists, sometimes bearing* considerable bodies of sheared and massive g*ranites. The lamination has a prevailing* strike of north 30° to 45*^ east and dips at hig-h ang*les either northwest or southeast, often pass- ing* throug*h the vertical from one to the other within the space of a few feet. Constituting* a small propor- tion of the area of this reg*ion are the rocks of the per- 26 JOURNAL OP THE idotite belt, which occur t3^pically in blunt lenticular form with a long'er axis of from a few hundred to a thousand feet or more and oriented with the lamina- tion in the surrounding- gneiss. Sometimes they take the form of narrow strips two or three miles in length.* Three types of ma.irnesian rocks are found in this belt: namely, jyeridotitcs, pyroxenites, and amphibo- lites, characterized by the predominance of olivine, py- roxene and hornblende respectively. Pyroxenites and amphibolite frequently occur in small masses in close association with peridotites, and sometimes form im- portant independent masses, but probabl}^ more than nine-tenths of the outcrops are peridotites. The Peridotites. The accepted classification of these rocks is used here merely as a matter of conveni- ence. No such division into distinct classes is possible in the field; and the names here g-iven represent sim- ply mineralog-ic varieties of the same petrog'raphic unit. Occasionally these varieties form separate mass- es, but g-enerally they g-rade insensibly into each other, and sometimes within the limits of a sing-le small outcrop. Dtinite, essentially the pure olivine rock, with acces- sory chromite or picotite, is by far the most important, since it constitutes almost the entire body of a g-reat majority of outcrops. Harzburgite (Saxonite), the olivine-enstatite rock, constitutes larg-e masses in some of the northwestern counties of North Carolina, but it is usually found as a local variety of dunite. "Perid- osteatite" and "g'linkite" seem to be only partially altered forms of this rock, consisting- of large olivine crystals (or serpentine) and talc. Amphibole-ficrite, * See map (Plate Ij published in Bull. 11, N. C. Geological Survey, 1895. EIvISHA MITCHELL SCIENTIFIC SOCIETY. 27 the olivine-actinolite variety, has been observed only in connection with dunite, thoug-h some chloritic olivine rocks are doubtless altered forms of this variety. Troc- tolite (forellenstein) consists essentially of olivine and anorthite, with zones of intermediate fibrous minerals always separating- them, and sometimes also bears larg-e dark hornblende. By gradual transition it pass- es, on the one hand, into dunite, and, on the other, in- to the feldspar type aiiorthosite. The Pyroxenites, — This class is represented by two types, which seem to be quite distinct, as no transitional forms have yet been found. Enstatite-rock, as the name indicates, is composed of orthorhombic pyrox- ene, and seldom contains any other mineral in prom- inent proportions. It often occurs in small masses with dunite and harzburg-ite, thouo-h considerable outcrops in some localities are composed of it entirely. In some of these instances there is apparently evidence of its origin from peridotites. Websterite, the enstatite-di- opside type, is found in the midst of the peridotites about Webster, Jackson count^s North Carolina. This type was first recognized and described by the late Dr. G. H. Williams," and similar rocks were studied from Baltimore county, Maryland, where transitional forms were found connecting it with peridotites, bron- zitite, gabbros, and norite. At the Webster locality, however, no such intermediate types have been ob- served. Aynfhibolite. — Only one type of this class has been found in the peridotite belt. It is composed of a bril- liant, grass-green, aluminous amphibole, which often bears considerable anorthite and is only occasionally * Am. Geologist, VI, 1890, 40-49. 28 JOURNAIv OF THE massive, g'enerally presenting- a g*neissic lamination. The amphibole constituent has long* been erroneously called "smarag-dite," solel}^ on account of its color. Analyses show over 17 per cent, of alumina, and Pro- fessor Dana calls it edenite in his new "Systemof Min- eralogy." This rock forms dikes in the peridotites at Buck Creek, Clay county. North Carolina, and is asso- ciated with peridotites in many places in this and the neig-hboring- counties of Georg-ia. No absolute contacts have been observed between the peridotites or pyroxenites and the enclosing- g-neiss. There is always a border of schistose talc, from two to three feet in thickness, developed between them. Of- ten there is also a variable zone of chlorite or vermicu- lite, or of both tog-ether, intervening- between the talc and the peridotite mass; and in many cases this bears corundum in considerable quantit3\"'^ EVIDENCE BEARING ON ORIGIN. We have seen that the peridotites occur, as a rule, in lenticular masses and sheets, having- their long'er axes oriented with the lamination of the g'neisses and schists of the country. Some of these lenses are short and blunt, and occasionally the outcrops present a ver}" ir- reg-ular outline and cover areas of several hundred acres, as at Buck Creek, in Clay county, North Caro- lina (Plate II). The smaller masses are in just the form one would expect to find in small intrusions of molten mag-ma into thoroug'hly laminated crystalline 4- For more detailed description of the rocks and modes of occur- rence aud distribution of both the peridotites and corundum, the reader is referred to ''Corundum and the Basic Mag-nesian Rocks of Western North Carolina," Bulletin ii, N. C. Ccol. Survey, and to "Corundum of the Appalachian Crystalline Belt," Traus. Am. lust. Mining Etigineers, XXV, Atlanta Meeting-, 1895, ElvISHA MlTCHElvI. SClENTIi^IC SOCIETY. 29 rocks; and the more irreg-ular outlines would naturally have been produced by larger intrusions accompanied or preceded by intense folding- and contorsion of the gneisses. Furthermore, both the irregular and the more typical lenticular outcrops frequently present a forked outline or send off small apophyses into the gneiss in such a way as would be wholly inexplicable in a rock of sedimentary origin. Prominent examples of this character are presented on the maps of Buck Creek, Corundum Hill and Webster areas (Plates II, III, and IV). A forked mass is found on the mountain slopes at the head of Elijay creek, Macon county, N. C, and a larger one on the spurs of Klk Ridge, in Ashe county. Several others of a similar nature might be added from North Carolina alone. The planes of least resistance in gneiss lie along the lamination, and hence, as stated above, we should ex- pect the axes of intrusive masses to coincide with these planes. This is found to be true in the great major- ity of cases, and the apparent exceptions are doubtful. The gneisses and schists bend closely around the en- closed mass, being only temporarily diverted from their normal course. The axes of the bifurcate masses are also found to follow the same general direction. (See Plate III). The last requirement in geotecnic evidence would seem to be fulh^ met by the gneiss area, more than a quarter of a mile long, entirely surrounded by perido- tites, on the northern border of the Webster area. (PI. IV). Indeed, this Webster region alone, though once made the basis of a theory of sedimentary^ origin for the peridotites," affords some of the most conclusive * "The Dunite Beds of North Carolina," by A. A. Julian, Proc. Nat. Hist. Soc, Boston, XXII, 1882, 141-149. 30 JOURNAL OF THE evidence of their intrusive character. The form pre- sented by the lamination planes of the g-neiss is that of an eroded dome-shaped anticline, somewhat elon- gated in almost a north and south direction. Even after considerable observation in the field, it was my impression that the peridotites of this area represented a continuous sheet conformable with the lamination of the g-neiss. But close, detailed mapping brought out not only the enclosed body of gneiss mentioned above, but also the irregular, protuberant apophysis at Ad- die, and other projecting- arms at the crossing- of Scott's creek, three miles west of Addie, and two in the vicin- ity of Webster. Furthermore, there are ^ve distinct breaks in the continuity of the eastern side on Cane creek, and the isolated portions have the typical lenticular form found in the majority of the outcrops. In the vicinity of these breaks there is a narrow strip passing- off the border of the map in a northeast direction, which is also completely disconnected from the principal masses of this reg-ion. This Webster rock shows the higfhest development of lamination found in the whole belt, and the micro- scope shows that the laminated rock is composed of thin layers of finely granulated olivine alternating' with coarser typical dunite. This condition, taken in con- nection with the g-reat development of schistose talc in the narrower portions, g-ives evidence of considerable shearing-, which is most naturally ascribed to the move- ments that gave rise to the anticlinal structure. The other characters pointed out are explicable only as the results of igneous intrusion. Much larg-er areas of g-neiss than those enclosed at Webster are almost surrounded in the Buck creek area by the peridotites and the later dikes of amphibolite. ELISHA MiTCHElvIv SCIENTIFIC SOCIETY. 31 The manner in which these dikes cut the peridotites or pass up throuofh the same opening- beside them, both in this reg-ion and in a number of places in the adjoin- ing- portions of Georg'ia, clearly indicates that the per- idotites occupy areas of weakness in the g^neiss, through which the dikes have found their easiest exit. It may be stated here that no direct influence of the amphibo- lite dikes upon the dunite was noted except an inter- ming-ling- of the constituents of the two rocks along- bands of three or four inches in thickness at the con- tacts. Another character which is perhaps as important as that of form, is the entire absence of lamination, or its development to only a slig-ht extent, in the g-reat ma- jority of peridotite outcrops. Dr. Julian, as noted above, studied the outcrop at Webster, where the lam- ination is so marked that the dunite bears a striking- resemblance to a thin-bedded sandstone; and this is, in fact, what Dr. Julian considered it to be. As pointed out by Dr. Wadsworth, however, "the chief defect in Dr. Julian's reasoning- is that all the evidence which he g-ives in support of this view^ could exist equally well if the rock had some entirely different orig'in. "^ If Dr. Julian had only studied some of the typical massive du- nite to be found within a day's journey of Webster, it is safe to say that his conclusion on this point would have been entireh^ different. Aside from the lamination in the body of the perido- tite, there is often abundant evidence that it has been subjected to considerable shearing in the schistose char- acter of the talc along' the borders and in the frequent slickensides encountered in the corundum-bearing- chlo- * Science. Ill, 1884, 486, 487. 32 JOURNAL OF THE rite zones. On the other hand, we find many outcrops that present none of these evidences of shearing-, but are perfectly massive and structureless; and we must conclude that these represent the least altered condi- tion of the rock. The massive character of the perid- otites, with often scarcely a trace of shearing*, is whol- ly incompatible with the theory that they are of con- temporaneous orig-in with the g-neiss and have passed throug-h the same cycles of disturbance. The line of separation between the peridotites and the g-neiss is always sharp; there are no transitional forms, either in chemical or mineralog^ical composition. The peridotites are extremely basic mag^nesian rocks (40-45 per cent, silica); the constituents of the gfneiss are all aluminous minerals, and the rock is hig-hly acid (60-70 per cent silica). That such dissimilar rocks would have been deposited as contemporaneous sedi- ments or precipitates over wide areas without some- where producing* an intermediate type is, at least, hig-hly improbable. Even a mere casual examination of these rocks with the microscope reveals a thoroug*hly crystalline g*ranu- lar structure, g*enerally a coarse texture, and in all re- spects the characteristics of a deep-seated, ig-neous rock. In fresh specimens the olivine g*rains always pre- sent ang-ular outlines, sometimes having* crystal form, and always fitting* perfectly tog*ether, without intersti- tial spaces or cementing* material of any kind. (PI. V., fig*. 1). In the first stag*e of alteration to serpentine the grains are separated by thin films of this mineral; but the irreg*ularities of adjacent g*raius still remain perfect counterparts. (PI. V, fig*. 2.). Only in the more advanced stag*es do the corners become rounded and the intervening* serpentine assume the appearance PLATE I. PLATE n. CONTOUR INTERVAL FIFTY FEET : FIGURES ON CONTOUR LINKS GIVE ELEVATIONS ABOVE SEA LEVEL. PLATE III. ("Showirig directions of strike and dip.) Mica Schist Corundum Workings MAP OF CORUNDUIVl HILL Macon County, N.C. By J. Volneu Leicis, 1S95. Topography by Chas. E Cooke. Contour Interval 10 feet. SCALE OF feet: 100 200 300 6NG'D BV AV-RICAN BANK NOTE CO.,N.V. FIGUREb ON CONTOUR LINES GIVE ELEVATIONS ABOVE AN ARBITRARY BASE— THE FLA" ROCK BED OF THE BRANCH NEAR THE SOUTHWESTERN CORNER OF ThEMAP^ PLATE IV LEGEND: strike and dip.) MAP OF THE lv'v'Vv'v<. Dunite. WEBSTER PERIDOTITE AREA JACKSON COUNTY, N.C. |Hfi|^| Websterlte. j5y J. Volneij Lewis, 1895. Topography by Chas. E. Cooke. Contour interval 100 feet. 1 J^' 1 Talc-Schist. SCAIE OF MILES: 0 3^ 1 , J \// . 1 Gneiss. 1 <^-^ (Showing- directions of ES3RAVED 8Y AMERICAN BANK NOTE CO.,f FIGURES ON CONTOUR LIKES GIVE ELEVATIONS ABOVE SBA LEVEL. PLATE V. EJIvISHA MITCHELL SCIENTIFIC SOCIETY. 33 of a cementing* material in a clastic rock; and this is true only when thin sections are viewed in ordinary lig-ht. (PI. V, fig-. 4.) For, so long as numerous oli- ivine remnants remain embedded in the serpentine, they are frequently found to exting-uish together over con- siderable areas, showing that the fragments belong- to the same crystalline individual, representing the larger g-rains of the orig-inal rock. Such fragments may be as widely separated by the alteration product as are the Z ^rtions of entirely distinct individuals. E^ven after all li aces of unaltered olivine have disappeared, the orig- inal granular character, and occasionally the outline of a crystal form, are still shown in some of the ser- pentines by the narrow, reticulating bands (* 'mesh- structure") of parallel-polarizing, fibrous serpentine, representing- the first stage of alteration along the bor- ders and fissures of the olivine grains. Plate V. is reproduced from photomicrographs of thin sections of typical dunite. A perfectly fresh spec- imen is shown in figure 1. Figure 2 shows a slight beginning- of serpentinization about the borders of the olivine. Figure 3 is taken from a fresh specimen in which the original g-rains, like those in figure 1, have been broken into great numbers of smaller g-rains, with- out perceptibly disturbing their orientation; this sec- tion is seen between crossed nicols. A more advanced stage of alteration than that of figure 2 is shown in fig-ure 4; the olivine grains, the light portions of the field, are reduced to mere remnants. Figure 5 repre- sents the final result of serpentinization, in which no fragment of unaltered olivine remains. With the ex- ception of spots caused by segregation of iron oxides, the rock appears perfectly homogent'ous in ordinary light. When viewed between crossed nicols, however, 34 JOURNAL OF THE the 'mesh-structure' is distinctly brouo-ht out, and the orig-inal <2franular nature of the rock clearly seen, as shown in iig-ure 6. The variation between the different mineralog-ic va- rieties within the same rock mass, and the essential unity of the whole peridotite .15 FesOo .36 1.01 4.63 trace 1.30 1.07 FeO 1.10 MnO .08 .24 CaO .84 .91 4.83 1.05 1.50 .90 McfO .03 trace .04 ZrOs .66 1.54 SnOo 22 .08 PbO .58 .26 TagOg i 6.27 F TiOs HoO .27 .35 1.61 .38 1.36 .78 ' .20 CeO ErgOs i 149.35 4.76 i 1 40 JOURNAIy OF THE (1) to (4) inclusive are from peg-matite veins of southern Norway, by C. W. Blonistrand. (5) From I^ake Ilmen, Russia, by R. Her- mann. (6 From Arendal, Sweden, by C. F. Rammelsberg-. (7) From Ottawa County, Quebec, by F. A. Genth. (8) From Burke County, N. C, by S. Iv. Penfield. Penfield" deduces the molecular formula: (Ce,La,Di),03 : P.O^ =1:1; THO2 ' SiOs =1:1. The former corresponds to the normal phosphate of the cerium metals (RgPsOj; the latter corresponds to the normal thorium silicate, which, in combination with a small percentag-e of water, makes the mineral thorite or orang-ite (ThSi04.H2 0). He concludes, therefore, that monazite is essentially a normal phos- phate of the cerium metals, in which thorium silicate is present in varying- proportions as an impurity in the form of the mineral thorite or orangite. Dunningtont had soniewhat previously come to the same conclusion. Rammelsberg'si formula of thorium free monazite from Arendal, Norway, was RgPgOg = (Ce, La, Di)2P2 08, thus ag-reeing- with Penfield. Blomstrand,§ from his analysis of Norwegian and Siberian monazite concludes that the mineral is a nor- mal tri-basic phosphate, an excess of bases being* com- bined withSiOs. Thus: ^^/(SRCPgOs) + 2R6,Si02 -f ^HjO, where 7?i = 5 to 20, and p = less than 1 usually. He does not believe, as Penfield does, that the thoria is originally combined with silica as thorite, but that * Am. Jour. Sci. (3) vol. XXIV, 1882, p. 250; vol. XXXVI, 1888, p. 322. Zeitschr. fur Kryst., vol, VII, 1883, p. 366; vol. XVII, 1890, p. 407. t Am. Chem. Jour., vol. IV, 1882, p. 138. t Zeitschr. Deutch. Geol. Gesell. Berlin vol. XXIX, 1877, p. 79.; Zeitschr. fiir Kryst., vol. Ill, 1879, p. 101. § Zeitschr. fur Kryst., vol. IX, 1887, p. 160; vol. XX, 1892, p.367. ELISHA MiTCHElvIv SCIENTIFIC SOCIETY. 41 it is present as a phosphate, either in combination with the cerium or as an isomorphous mixture, thus: IV III 1 I IV Ce.Ce.COsPO). and R.ThCOsPO)^; and that it is altered to the silicate by siliceous waters. PHYSICAL PROPERTIES. The crystalloo-raphic form of monazite is monoclinic, and the commonly occurring- planes are ortho- and cli- no-pinacoids and domes, the unit prism and the unit pyramid. The basal pinacoid is rare, having* been observed only on crystals from the Urals and from Al- exander County, N. C. The usual crystal habit is tabular, parallel to the ortho-pinacoid; also short columnar, and sometimes elong-ated parallel to the prism. Twins are not com- mon, the twinning* plane being* usually parallel to the ortho-pinacoid. These crystals vary in size from the microscopic nee- dles of cryptolite, which have a thickness of 0.00015 to 0.00062 inch, to the abnormally large crystals of Ame- lia County, Va., 5 inches in leng-th. The more g-eneral variation lies between ^^ and 1 inch. The cleavag-e of monazite is most perfectly devel- oped parallel to the basal pinacoid, it is also distinct, as a rule, parallel to the ortho-pinacoid; and some- times visible parallel to the clino-pinacoid. The mine- ral is brittle, w^th a conchoidal to uneven fracture; the hardness is 5 to S.S\ specific g-ravity 4.64 to 5.3; lustre resinous to waxy; the crystal faces are splendent in fresh, pure specimens, dull in weathered, impure spec- imens; the color is honey yellow, yellowish brown, am- ber brown, reddish brown, brown or greenish yellow; the purest specimens are transparent, becoming- trans- lucent, and even opaque in the impure varieties. 42 JOURNAL OF THE The optical properties of mouazite are: — thin sec- tions, b}^ transmitted ligfht, arc colorless to yellowish; pleochroism is g"enerally scarcely noticeable; absorption b is g-reater than c = a; the plane of the optic axis is perpendicular to the plane of of S3'nimetry, that is the clino-pinacoid, the positive a.cute bisectrix lies iu the oblique ang-le /?, hence sections parallel to the basal pinacoid show the full interference figure; the extinc- tion angle varies from l'^04^ to 5" 54^ ; the optical angle is small, 2E (red) = 25»22\ 2E (yellow) ^ 24°56\ 2V (yellow) = 12°44S (from Schiittenhofen, Bohemia); the dispersion is weak and horizontal; the single refrac- tion is high, and the double refraction considerable, y — a= 0.0454, y — ^ 0.0446, fi — a =0.0008 (from Aren- dal, Norway). DISTRIBUTION AND MODES OF OCCURRENCE- Until comparatively recently the localities in which monazite was found were few and far between. The original specimen of turnerite came from the Dau- phine in France; in 1826 Menge discovered some crys- tals in the Ilmen Mountains of Russia; it was then found in the United States at Norwich and Watertown, Conn. Up to the present time, it has been found in over 75 localities in the United States, Canada, South America, England, Sweden, Norway, Finnish Lap- mark, Russia, Belgium, France, Switzerland, Ger- many, Austria and Australia. And the probabilities are that these localities will be rapidly added to in the future. Monazite is an accessory constituent of the granite eruptives and their derived gneisses. It has been found in these rocks over widely separated areas of the Earth's surface, and further search and study is liable to reveal its more general presence in similar rocks, ELISHA MITCHEIvL SICE^NTIFIC SOCIETY. 43 than was formerly supposed. Thus Derby, "'^ by exam- ining- the heavy residues of a number of hand speci- mens, selected at random from the collection in the National Museum at Washing-ton, D. C, described the occurrence of monazite in certain granites and gneisses of Maine, New Hampshire, Rhode Island and Massa- chusetts. In Norway, Silesia and Bohemia, and in some of the mica mines of Canada, Virginia and North Carolina, monazite has been found in pegmatite dikes. Derby has found the mineral in a red s^^enite at Serra do Stauba, in the province of Bahia, Brazil. The turner- ite of the Saacher Lee (which is an extinct volcanic crater) near Coblenz, in Prussia, was found in a druse in a sanadine bomb, the only known occurrence of mo- nazite in an undoubted volcanic rock. The turnerite of Olivone, Switzerland, occurs in a quartz vein 20 to 30 cm. wide, traversing- crystalline schists. The cryptolite of Norway occurs as inclusions of very fine, needle-shaped crystals in apatite. In Cleveland County, N. C, monazite has been found intergrown in cyanite. The percentage of monazite in these rocks is exceed- ingly small, often infinitesimal; thus Derby"" states that the g-ranite dikes of Serra de Tingua, near Rio, are rich in the yellow mineral, carrying 0.02 to 0.03 per cent, and a fine-grained granite dike on the outskirts of Rio de Janeiro, showed 0.07 per cent monazite. Monazite has not been found in the sedimentary rocks, although it may be i)resent in some of these as a sec- ondary mineral of transportation. * Proc. Rochester Acad. Sci., vol. I, 1891, p. 294. * Proo. Rochester Acad. Sci. vol, I, 1891, p. 294. 44 JOURNAIv OF THE The monazite is contained in the main constituents of the granitic rocks, in the quartz, feldspar and mica, thoug-h it appears to be more g-enerally confined to the feldspar. Zircon may be regfarded as a constant associate; among" the other usually associated minerals, of coeval origin with the monazite, are xenotine, ferg'usonite, sphene, rutile, brookite, ilmenite, cassiterite, magnetite, and apatite; sometimes beryl, tourmaline, cyanite, co- rundum, columbite, samarskite, uraninite, gummite, autunite, gadolinite, hielmite and orthite. Among the principal secondary minerals found in as- sociation with monazite, are rutile, brookite, anatase, epidote, orthite, g^arnet, sillimanite, and staurolite. The economically valuable deposits of monazite are found in the placer sands of streams and rivers, in the irregular sedimentary sand deposits of the adjoining* bottom lands and in the beach sands along* the seashore. The decomposition and disinteg^ration of the crystal- line rocks, the original source of the mineral, has pro- ceeded to considerable depths in certain localities, par- ticularly in the southern unglaciated countries. By erosion and secular movement the material is depos- ited in the stream beds and there undergoes a natural process of sorting* and concentration, the heavy mine- rals being* deposited first and tog-ether. The richer portions of these stream deposits are thus found near the headwaters. The accompanying* plate shows one of these small valleys (Lattimore's), three miles north- east of Shelby, N. C, where all of the underlying* g*ravel is being* dug* and washed for monazite (see also plate facing p. ), and where the sand in the bed of the small stream is also being' washed for the same pur- pose. The geographical areas over which such work- ElvISHA MiTCHELIv SCIENTIFIC SOCIETY. 45 able deposits have been found up to the present time, are quite limited in number and extent. In the United States the placer deposits of North and South Carolina stand alone. Similar deposits exist in the provinces of Bahia, Minas Geraes, Sao Paulo, and Rio de Janeiro, Brazil; in the river sands of Buenos Ay res, Argentine Republic; in the gold placers of Rio Chico, at Antioquia, Colombia; and in the Bakakui placers of the Sanarka River, Russia. In Brazil there are also important deposits in the beach sands in the southern part of the province of Bahia, near the island of Alcobaca. But little reliable information is at hand concerning- these foreign deposits, and the remainder of the pres- ent paper will be taken up with a description of the Carolinian deposits, and the methods of mining and cleaning the sand, emplo3^ed there. The Carolinian area includes between 1600 and 2000 square miles, situated in Burke, McDowell, Ruther- ford, Cleveland, and Polk Counties, N. C, and the northern part of Spartanburg County, S. C. The principal deposits of this region are found along the waters of Silver, South Muddy, and North Muddy creeks, and Henry's and Jacob's forks of the Catawba River in McDowell and Burke counties; the Second Broad River in McDowell and Rutherford counties; and the First Broad in Rutherford and Cleveland coun- ties, N. C, and Spartanburg County, S. C. These streams have their source in the South Mountains, an eastern outlier of the Blue Ridge. The general out- lines of this area are indicated on the accompanying map. The country rock is granitic biotite gneiss and dio- ritic hornblende gneiss. The existence of monazite 46 JOURNAL OF THE here in commercial quantities was £rst established in 1887. The thickness of these stream g; ravel deposits is from one to two feet. The percentage of monazite in the original sand is very variable, from an infinites- imal quantity to one or two per cent. WASHING AND CLEANING MONA^ITE SAND. The monazite is won by washing- the material in sluice boxes, about 8 feet long- by 20 inches wide by 20 inches deep, exactly after the manner that placer gold is worked. Magnetite, if present, is eliminated from the dried, concentrated sand by treatment with a large hand magnet. Many of the heavy minerals such as zircon, menaccanite, rutile, brookite, corundum, g"ar- net, etc., cannot at present be completely separated. The commercially prepared sand, therefore, is not pure monazite. A cleaned sand, containing from 65 to 70 per cent, monazite is considered of good quality. The most systematic washing method employed is by the use of two sluice boxes, the mouth of one dis- charging into the head of the other, placed below. The gravel is charged on a perforated plate at the head of the upper box, and the clean up is so thoroughly washed as to g-ive a high grade sand, often up to 85 per cent, pure. The tailings discharge directl}^ into the lower box, where they are rewashed, producing a second grade sand. At times the material is siibjected to as many as five similar consecutive washing- treatments in the sluice boxes. A further concentration of the dried washed sand is sometimes made b_v pouring from a cup in a line, steady stream from a height of about 4 feet, on to a level platform; the lighter quartz and black sand, with the line grains of monazite (tailings) fall on the perijohery of the conical pile and are constantly brushed aside with hand brushes; these tailings are ELISHA MITCHELL SCILXTIFIC SOCIETY. 47 afterwards revvaslied. Or, instead of pouring" and brushing-, the material is treated in a winnowing* ma- chine, similar to that used for separating chaif from wheat. Although the best grade of sand, produced by the above complicated treatment, may be as high as 85 per cent, pure, its quantitative proportion is small as com- pared with the second and other inferior grades, and there is alwa^^s considerable loss of monazite in the final tailings. It is impossible to conduct this ^vashing" pro- cess without loss in monazite, and equally impossible to make a /6?;y^