Ae iabteransh we bid? w\y ene worn was P yw i AKG “wy, tia: $5 ” | uM cae + absey yey" Ira TIM gai rrp tro epee in ll Pe LUSLIEDy Soy men ‘g oP rie ee A Hyg rgl cuprrnhttll naresereN tO aa Aare’ Se ea ae ny Wye | % SMe : iv MN att a | Lk i VEN ys ; Bet b ) ES @. Py Tiny slat hls al dale tefobebgletalaiey dhe lalate on “Ye the ‘wi Nyfneae at gt AA ee > rag 7) = @. 4 F TET GLEPPLEL Lig ° 4 é AN a ay fF bane Ze ) Seu. ¢ A Me 4 *gy>’ VeS~ nlite Hy Whit aes ete Tote \iMAA, ; s AP ee id { y . e's ¢ i | ; MN PT nee (Ae pare geet man uu} neweccn v ‘ Be, ay, ‘re 1 et ere, fm. A Are Aes. ww ai ta \ pre, ‘ ig Nple tris ™ \ Dee P ° © = , f vic a wer, 4 . ny Mac y) br na A Aaiaiibiay meni ‘ Ww ? $e. a ; I wAy “- “A, «AN | & on. . hia a a a 4 gin | 7 4! ST ld Pou, : ya Ott ; , : +! 1 » Q wo | | eae. . Ly Tamil cart 7 . PEUT ETE ee {| ALi 5 i Spaaezta’ ey Sn Ay eae Rea eas, VOSSSae UBReneeaee res ~ ee . 1 ne ‘vim, ~ ai Py \ or 4 “ss al bal i) “vy a oJ badd |, | Bahk] TLE e ne Ngee geeie Ma hg ce | NABRRE TINS ingens. wn { NighPe adn incatttrne ttl nt SDL eo Tae wot * be ‘A 44 | gure qT rt leh’ Pei Wire va dtd he Lb ee wf { ™ grag « .< Wy < , f Jone I Md Wo i. : pe aden SSF ieee et rer hy ha eben ratios MAA | Wve eee sada ukl git ae Pree PAG LL oles | nO earn, eS ( ‘gs. so oe ‘* es ev hh & 7) 2S tr ee tN Why , ‘ " ; rs a es “Q b W: RA: Yin adnisl mL bh I bie a nth AL) | | i ih Mousnent XA got ats, nal! : 3 TET Ni ry “hi x 6 4 : ae ae d mee ew tt , e yg ge qe Ie, = ort TNA a 19 pa Shahn yr 89 guna aa] Ki aot Va 5 Cee ea Ne a (oS oi wa ty! ma tt eer y wat = “VWOWie es i bed Ss a os es ee eee ti?) ie eh ee Jey ee Nhat A - ay, AMAA ee haan teegeayet te SRA Tit “ai ah Ay | x EEL es ce oe Th pein ag Pr cue al i} ai NTeA f re . Weer: soa 64,0 A ; AR ry vo Jods BoN ma Bas! WAS ™ Vig : Bins $¢ F “— ‘b. 2 $62». iN oy rtp , , ; ar b 7 C.<. 4 es 4 Jd a ° =e J as ‘in bcos teat wv at . vad ¢ no . 5 * OLereeae EE 4c - “fue «A a el» wae » f - La Py , > Ps v a - a Pt i e No . at g = ef ~~ ad, a i{{b ra oe acaniinrtrentin SPM PART) de ee pul | Caer ea PALA , é 4 a ca a « ¢_4 QM i>. a Reed ty why, nh. a r : g yx A] tet NAS Lian AAA GUPEP leit Daas eins Netti a «| Uy BS Stead \. Ln * 06 Ary . if ~ es. Se: \ » NAAAY nw an ¢s aE ny 8 PP Av: >: cere | pz alll RA eG ees ‘ag AW) ad ty! PAILS a a AD A "Nor Th) A = q wo‘ ™\ " “| ‘dao a' ¢ 4°» a-,\ 4 ; Alay a Seat Anan vosve® < 7% a = ae YAIR A ti Ae © te, ars ee Wee Ras ‘ fh att ToPeeen Seppe. Anttsananerest 1) NAAT r Ree Awe Sg Bae I LCE) aneethiadl mitatinin te oe ON eens ge ees | . Ue ys 4% “14dud “4 ants TH ~ ips Ue Pe adn a lalallala ZR a gorse TEETH 4 . : o~. So Or i a -\ ; Bet a ? we “\- 4a | a 4 i a { ( yy Sad — <¢ OY et nd . : guy —@e Me . a. «.» rveveee : : ‘eure eh a olalalaidialal TON RRA AA ‘peal TTT OlUy “4 oy Nit, Sh ; “4 1 | ae 7 os acted e { we SM pli Sie | USae OPTI r ed aaa sy, Vesey sei i Mca yily count OV Hind’ Aa sseeeg een . ea wt i 4 | 5 ONS tg au" 1° SUS "ul = rm Warcenyy +. Totpetcresnans wes 7) L era. 3 " ere AA | Lely 1M Gan SOP Wis }) yt | : eclttotne amdll lim secel trae agutaccints ty, . re sy Vy 4 belt oN ber © Ne abiehe ere Ths + shgtahags oe r ve & Hees ‘ \e Hh el f THY rhe Met me Paps ittg slay "8 bey Ana dh. Ser Qa, =) ars, SF SOS BUD Nd | “ney yes qd Ay, » 9 ay n rofin, eyes csc On. MALLE . \) ty ne tae, Wem ent m hy Ps ~S San . Paws fe! ie sal Dat ay W anh mh os ' ~~ Ay sal Neug ars, ue | : Sh PISA 32% 7 Hay ot ~ yt | My a ‘y Pian - ia No. 1.—JANUARY, 1896, WITH PLATE I. : NEW HAVEN, CONNECTICUT. 2 : 1. 19.6" TUTTLE, MOREHOUSE & TAYLOR, PRINTERS, 125 TEMPLE STREET. ublished monthly. Six dollars per year (postage prepaid). $6.40 to foreign sub- ers of countries in the Postal Union. Remittances should be made either by ley orders, registered letters, or bank checks. J museums. as ie adapted for educational work, are sold at very low prices. SCIENTIFIC BOOKS, PAMPHLETS, PAPERS, PROCEEDINGS — Individual Specimens. of rarest. and eheyues variety for collector Systematic Collections, comprising carefully tne and | : labelled type specimens of all common or important species, especially Special Collections, illustrating the various ores ; the uses of minerals ; $ a. their physical and chemical character, etc., etc. ee Minerals for Laboratory and nail purposes at lowest rates by the pound. Sof Detached Crystals in great Male stAs and pertaphen for students nd investigators. : . es ae Microscopical Mounts of exceeding beauty. D5 ED, WAGES UIEZ BS The quarries are opened in the southern quarter of a mile of the Quarry ridge, which extends thence northward for about a mile. The oldest quarry is at the end of the ridge; the others were opened in 1890. They are worked by Messrs. John §. Lane & Son, for railroad ballast and road metal; the rock being crushed alongside of a branch of the Consolidated =~ 7 Nee ) ii a \ \’ NS Davis—Quarries in the Lava Beds at Meriden. 9 railroad, by which it is hauled away. Some 800,000 tons have been sold in the last five years. In the rough sketch here given, the lower and inner part of the quarries is occupied by what I shall call the lower bed: the upper and outer part by the upper bed. The surface of, separation between the two beds may be traced without much difficulty; it is now very clearly shown at several places, but its distinctness varies with the condition of the quarry. Followed around the various faces of the quarry, this surface may be traced with more or less continuity for six or eight hundred feet ; and if all loose stuff were cleaned away, a length of a thousand feet might probably be measured. I have attempted to show the contact in fig. 3 by a line that breaks the shading at the back of the quarries. The deepest part of the lower flow now exposed is in the back of the southern one of the new quarries, where it is seen for about forty feet beneath its surface. Here it is of a bluish gray color, fine-grained and dense, rarely vesicular. Nearer its surface it becomes red or purplish, with an increasing propor- tion of vesicles. The red color does not seem to depend on the weathering of to-day, for it is fully developed deep in the quarries and in massive, unjointed rock: it should therefore be ascribed to weathering immediately subsequent to eruption, or to underground alteration. The color is often so strong near the upper surface that this part of the trap might easily be mistaken for baked sandstone. The largest regularly shaped vesicles in the upper part of the flow are nearly an inch in diameter; in the most vesicular rock the originally empty spaces must have occupied at least a third of the entire volume. Besides the spheroidal vesicles, now filled chiefly with calcite and chlorite, there are occasional irregular cavities, up to six inches in diameter, more or less completely filled with crystal- line minerals, chiefly calcite. Mr. Chapman, foreman of the quarry, assured me that these contain water when first opened in fresh broken rock. With all these secondary minerals, how- ever, this account has little to do. The lower bed nowhere exhibits any distinct columnar struc- ture, but has on the other hand a tendency to split into slabs four or five feet thick, along weathered joints parallel to its surface. The scoriaceous surface of the flow is occasionally stripped bare: thus at present there is, in the old quarry near its southern end, an area measuring about sixty by ten or fif- teen feet, where its rolling form is well exposed. It is a typ- ical example of the Hawaiian “ pahoehoe.” The adjacent swells of the surface measure three to six feet across, and their vertical relief may reach a foot or more. 10 Davis— Quarries in the Lava Beds at Meriden. The upper bed, as far as exposed in the quarries, is unlike the lower in many respects. The color of its deeper joint planes, where not affected by weathering, is dark steel-gray, greenish or bluish; the fresh broken rock is dark gray or greenish, without any of the bright reds or purples of the lower flow. The texture is dense throughout, vesicles and cavities being very unusual: but by following the quarry ridge to the northeast about a third of a mile, one may find the ordinary vesicular structure of the upper part of the bed there exposed on its back slope. ‘The density of the upper bed in the quarry is particularly noticeable where it rests directly upon the roll- ing surface of the strongly colored and highly vesicular lower bed. At the point in the old quarry where the surface of the lower bed is best exposed, the dense, dark lava of the upper bed fits closely into all the inequalities, the two beds being sometimes so closely welded together that weathering has not loosened them along the surface of contact, and hand speci- mens can be broken off showing the two kinds of lava: but as a general rule a weathered seam or parting follows the contact. The columnar jointing of the upper bed is fairly well devel- oped at many parts of the quarries, but it nowhere produces columns of notable regularity. Near the base of the upper bed, the rough joint columns are of small dimensions: half a foot to a foot in diameter. Some fifteen or twenty feet above the base the columns are often two, three or four feet in diam- eter. The weathered joints in the rocks near the surface of the ridge are yellowish in the upper bed, but brownish in the under bed. The former is hard enough for use as road metal to the very top of the quarry, but the latter is weak and rotten where it has been weathered. The stripping of the ridge in preparation for further quarrying exposes many well-glaciated rock surfaces. The various original features of the two masses thus described accord so perfectly with what occurs in modern vol- canic districts, and the secondary features result so naturally from the long lapse of time that the rocks have existed, that “a double-bedded lava-flow” seems to be a well warranted name for the whole structure. The most significant features of the structure appear over so large a quarry face, and main- tain throughout so constant a relation that no reasonable doubt can remain as to their meaning. In order to enforce the gen- eral recognition of a conclusion so well supported, I have replaced the ordinary term “trap sheets” with the more sug- gestive term “lava beds,” in the title of this article, and hope that the latter term may come into common use. The absence of stratified deposits between the two lava beds may be variously explained. The first bed may have filled the Davis— Quarries in the Lava Beds at Meriden. 11 shallow waters of the estuary into which it flowed, so that its surface was exposed to the air; then before the gradual depression that long prevailed in the region succeeded in sub- merging the lava surface, the second bed may have been poured out on the first. If the top of the first bed did not reach the surface of the estuary waters, it must be supposed that the second flow occurred before a perceptible deposit of sediment had time to form. The probabilities seem in favor of the first supposition: for the close study that has been given to many good contact specimens by Mr. Whittle failed to detect any trace of stratified deposits between the two lavas: but on the other hand, a few grains of quartz were found, not deriv- able from the lava and perhaps best explained as wind-born sand. The difficulty of recognizing the double structure of the “trap sheet” on the weathered, drift-covered, and wooded surface of the Quarry ridge prepares the observer not to expect manifest exhibition of double- bedded structure in the adjacent ridges of Cat-hole and Notch mountain blocks: but on the back of Notch mountain there are good indications that an upper flow occurs above the greater mass of the sheet.* The lava beds no longer lie horizontal as they must have at the time of their extrusion, but dip to the east-southeast at an angle of fifteen or twenty degrees. This is shown by the slope of the vesicular portion of the under bed; the slope is so manifest that it is well known to the quarrymen. The sandstones of the district have a similar dip, thus confirming the idea that the lava beds were essentially level when they were poured out, and that they were afterwards tilted along with the whole mass of the formation. Indeed on this point there can be no doubt with regard to the extrusive sheets; but whether the intrusive sheets of West and East rocks, near New Haven, were driven between the sandstone beds before or after tilting is not definitely settled. The opinion that they were intruded after the tilting of the stratified beds is generally accepted ; but the general accordance of their dislocations with those of the extrusive sheets in the Meriden district strongly suggests that both the intrusive and the extrusive sheets took their place in the bedded rocks before the tilting and faulting occurred. The evidence of faulting, both on a small and a large scale, is remarkably distinct in the Meriden quarries and their imme- diate surroundings. Within the quarries there are several fis- sures, of breadth varying up to five or more feet, filled with fractured rock, now more or less weathered. The greater part * Bulletin Museum Comp. Zodl., xvii, 1889, 81. 12 Davis— Quarries in the Lava Beds at Meriden. of the filling is of trap fragments; but the space between the angular blocks of trap is generally filled with a matrix of sand- stone or sandy shale; not fragments of bedded sandstone and shale, but a recomposed mass, presumably derived in the form of sandy or muddy powder from the grinding of overlying sandstones and shales at the time of faulting. All the ordinary features of fault structure may here be studied to good advan- tage.* The amount of movement on the fault lines reaches fifteen or twenty feet in the best examples now exposed ; this being on the little hill of rock left standing between the two northern quarries, where the vesicular surface of the lower lava bed is dropped by nearly twenty feet on the northwest side of the fault. This dislocation is very clearly exhibited at present. Several smaller examples of measurable throw occur in various parts of the quarries. The trend of the fractures varies from about north-northeast to northeast, the latter direc- tion being that of a little swampy water course that ob- liquely terminates the quarry ridge on the south.. The rock exposures nearest to this little valley are the weath- ered breccias on the margin of the southern quarry, and there is every reason to think that the valley itself is nothing but the topographic expression of a broad band of fractured rock, produced by a strong fault. The distribution of ledges and ridges in the neighborhood gives full support to this idea. Just across the little valley is the northern end of a ridge of conglomeratic sandstone, whose strike would carry its strata directly into the lava beds. Although the lavas were poured out previous to the deposition of the overlying fragmental beds and although the pebbles of the conglomeratic beds give evidence of rather active currents, no trap fragments are to be found among the pebbles; and this increases the presumption that the two masses—the lava beds and the conglomeratie sandstones—do not he in their original relative positions, but have been brought into their present relation by faulting. Adopting provisionally the general course of the little valley as the trend of the dislocation separating the two dissimilar ridges, and walking northeast or southwest, abundant confirmation of this idea is found. The sandstone ridges, coming northward towards the line of the supposed fault, end on reaching it in the submissive manner already described ; and at a distance of several miles to the northeast, the northern ends of the anterior, main and posterior sheets of the Lamentation block are simi- larly truncated. -The uplift of the fault is then seen to be on the eastern side, and its movement approaches two thousand * Details are given in the article on ‘‘The Faults in the Triassic Formation near Meriden,” already referred to. Marsh—WNote on Globular Lightning. 13 feet ; this being the great fault of the region, numbered 6 on fig. 2. The entire structural arrangement of the district is soon found to accord with the clue suggested by the quarries ; thus enhancing their value as a place for beginning the study of the region. The best approach to the quarries is by an old road leading north from the highway just east of the Fair Grounds. The observer then follows the ridge of conglomeratic sandstone above referred to for a little distance before the quarry ridge comes in sight. With the expectation that a strong topographic feature would suggest, he naturally looks forward to the con- tinuation of the ridge for some distance. Hence there is always an element of surprise when dark rocks in the old quarry first rise into view a few hundred feet to the north in line with the strike of the conglomerates. The chief problem of the region—the relation of the lava beds to the sedimentaries—is thus brought vividly to mind; and before returning from the quarries the problem may be advanced far toward solution. Harvard University, October, 1895. Art. I1.—Wote on Globular Lightning ; by O. C. Marsa. A GREAT deal has been written recently on the various forms of lightning, and the subject itself has so much scien- tific interest, that it may be worth while to place on record an observation of my own on globular lightning, made years ago, in which the main facts are different from any I have seen described. ; | On Tuesday, July 23, 1878, I was on board a large yacht at anchor in the harbor of Southampton, England. About two o’clock in the afternoon, when we were about to sail, a violent thunder storm came up from the west, and as it passed over Southampton, several bolts descended, one of which, as I afterwards learned, struck a church. As the first drops of rain came down on the yacht, I was standing in the after compan- ion way, looking forward, when my attention was attracted by a bright light apparently near the upper part of the foremast. When I first saw it distinctly, it was about half-mast high, and was falling slowly and directly toward the deck. This light was a ball of fire, a delicate rose-pink in color, pear-shaped in form, with the large end below, and appeared to be four or five inches in diameter and six or eight in length. 14 Marsh— Note on Globular Lightning. When it struck the deck, about forty feet from where I was standing, there was a loud explosion, and it was some minutes before it could be ascertained what damage had resulted. The mate, who was near the mainmast, about twenty-five feet from where I stood, was knocked down, but soon recovered. The same bolt, or part of it, also passed in front of the fore- mast, down a windsail ventilator, into the galley, where it knocked a large tin pan from the hands of a cook, and upset things generally in the culinary department, but injured no one seriously. Of the crew, some were on deck and some below, but none were really harmed, although a few were badly demoralized. A strong ozone-like odor was observed immediately after the explosion, and this remained perceptible for some time. The officer in command of the yacht, Captain Matthews, who was forward at the time, and escaped without injury, stated that just after the stroke, he saw “streaks of lightning running around on deck like snakes.” I was myself onl dazed for a moment by the explosion, and saw distinetly that the deck forward was illuminated with a bright confused light. The owner of the yacht, George Peabody Russell, and his other guests, had gone below when the storm began, and suffered no harm, except possibly from fright, as they were still further away from the stroke. As soon as the storm had passed, I made careful notes of the whole occurrence, with drawings and measurements, as I was much interested in the subject, and it was the first instance of the kind I had seen at close quarters. An inspection showed that the vessel itself had sustained no material damage, and not even permanent marks were left on the deck where the ball of fire exploded. A number of other yachts were at anchor quite near our vessel at the time, among them the white “Sunbeam” just home from her well-known voyage, but we saw no indications that any of these had been struck. I had no time to inquire, as immediately after the storm we sailed on a cruise to the eastward.* Yale University, New Haven, Conn., December 4, 1895. * Those interested in the rarer forms of lightning will find many observations recorded in ‘‘ Nature,” especially during the last fifteen years. Littlehales—Form of Isolated Submarine Peaks. 15 Art. Il—TZhe Form of Isolated Submarine Peaks; by G. W. Litrnenates, U. 8. Hydrographic Office. THEORETICALLY the form of an isolated submarine peak would be that of a solid of revolution in which the crushing strength of any section is equal to the combined weight of the portion of the formation above that section and of the superin- cumbent body of water. Let y denote the radius of any sec- tion, and # its distance from the top of the formation. Let K denote the coéfficient of crushing strength of the material composing the formation; 0, the weight of a unit of its vol- ume; and 0’, the weight of a unit of volume of sea-water. Assuming that the top of the formation just reaches to the surface of the ocean, of yds = =the weight of the formation above any section whose distance from the top is a, 270 fy. x. dy =the pressure of the water upon the formation above any section whose distance from the top is x, mKy* = the strength of any section to resist crushing. Then nif yrda+2ons'fy. a. dy = 1Ky?+C (1) in which C is a constant representing the excess of crushing strength in any section above what is necessary to withstand the pressure caused by the weight of the formation ome the weight of the superincumbent body of water. By differentiation, equation (1) becomes moy’. da+270'y. x. dy = 2xK. dy 0 dx _ dy oe 2 (K—6'x) y O da dy or 30! a) = y (2) = By integration, equation (2) becomes O 1 *) 48 sar los (oF == 10g: ¥ 20" K 7 lesy (3) or ¢° ao in which e is the base of the Naperian system of logarithms. Littlehales—Form of Isolated Submarine Peaks. 16 UOTNSOASUAL? POITLIA Suro? Abo 3 968 779+ Sb6L gp wis A PPUUL DY 02 S2 UOSA~LOLUOD VY “02s. 2SSROStp LIP UA Sas = 1000 fathoms. Each diiston y Littlehales—Form of Isolated Submarine Peaks. 1% This equation has been used in the generalized form, v= A+Be?4, to find from the observed bathymetric data in relation to Dacia Bank in latitude 31°-10’ N. and longitude 13°-40’ W., Seine Bank in latitude 33°-50’ N. and longitude 14°-20' W., The Salvages in latitude 80°-05’ N. and longitude 15°-55’ W., Enderbury Island in latitude 3°-10’ S. and longi- tude 171°-10’ W., and the shoal in the North Pacific ocean in latitude 32°-55’ N. and longitude 132°-30’ W., the equation to their average form. For this purpose the values of y, expressed in nautical miles, and «, in fathoms, were inserted in the above equation and a conditional equation was formed for each of the submarine formations. From these conditional equations normal equa- tions were found by the Method of Least Squares, which gave the values of the constants A and B. The resulting equa- tion is xe = +68°7985+641°8396 Eley , and the corresponding curve is shown in the accompanying diagram together with others which have been plotted from measured data for purposes of comparison. This investigation has an important bearing upon the inter- vals at which deep-sea soundings should be taken in searching for probable shoals in the open ocean and in developing the character of the bottom of the sea. It shows that isolated formations occupying comparatively limited areas at the bottom can and do occur in deep water, and we are able to assign at once the maximum interval that should obtain between deep- sea soundings taken in the above-mentioned operations. The minimum radius at the bottom which a dangerous shoal can have must vary directly with the depth, but on the average, in the deep sea, it may be stated as 10 miles. An interval of 10 miles coupled with an interval of 2 miles would be sufficient for general development, and would prove with certainty the existence or absence of any formation rising close to the sur- . face. Of all the possible ways in which a 10-mile interval could lie with reference to a submerged peak, that which would be most advantageous for a prompt discovery of its existence is the condition in which one end of the interval is at the bottom of the slope and the other near the apex, and that which would be least advantageous is the condition in which the interval is bisected by the position of the apex. In the latter case, there would be nearly equal soundings at both ends, but the soundings at the ends of the adjacent 2-mile intervals ~ would immediately disclose the slopes. Am. Jour. Sci1.—FourtsH Series, Vou. I, No. 1.—JANuaARy, 1896. 2 18 Wells and Foote—Double Fluorides of Art. [V.—On the Double Fluorides of Cesium and Zir- conium ; by H. L. Wetus and H. W. Foors. In connection with his comprehensive work on zirconoflu- orides, Marignac* has investigated the double fluorides of zirconium with ammonium, sodinm and potassium, and since the corresponding cesium salts have never been investigated, — we have undertaken a study of them. The following table gives Marignac’s ammonium and potas- sium salts, together with those which we have prepared with ceesium ; 3: 1 Type. 2:1 Type. 1: 1 Type. 2:3 Type. SNH F. ZnB, 2NH FE. ZF). 223. nscsee, ee sKF.ZrF, 2KE.ZrE, KE.ZrF,.H,0 (2) 2 ite e ee 2CsF.ZrF, CsF.ZrF,.H,O 2CsF.3ZrF,.2H,0 The analogy between two types of cesium and potassium salts is complete, while one type varies in each series. No evidence has been found that cesium, in this case, forms a greater variety of compounds than potassium. The symmetrical arrangement of the vacancies in the table, where no salts have been discovered, indicates that alkaline fluorides of lower molecular weight combine with a relatively smaller number of molecules of zirconium fluoride, while those of higher molecular weight combine with a greater number of such molecules. The 2:1 type is the only one occurring in all three series. This is the common and usually the only type of double halo- gen salts formed by tetravalent elements, hence its occurrence in all cases was to be expected. The single sodium salt described by Marignac, 5Nalk’.2ZrF,, does not correspond to any of the compounds in the above table, but it is to be noticed that the composition corresponding to this formula varies but little from that required for 2NaF’.ZrF’,. Although Marignac’s work on this salt was, as usual, very thorough and careful, it seems possible that his products may have been the 2:1 salt containing a small amount of some impurity, possibly a 3:1 compound. Marignac described the salts Mn,ZrF,. 6H,O, Cd,ZrF, .6H,O, Zn,ZrF,.12H,O and Cu,ZrF, .12H,O, all of which correspond toa4:1 type which has not been obtained with the alkali- metals. This type and those given in the preceding table make five varieties of zirconofluorides, one of which has been discovered in the present investigation. * Ann. Chim, Phys., III, lx, 257. Cosium and Zirconium. 19 The materials used for the preparation of the cesium salts under consideration were carefully purified by ourselves. Hydrofluoric acid was made from perfectly pure fluor-spar and sulphuric acid, using a platinum still and redistilling the product. Czesium carbonate, purified by the method described by one of us,* was used in preparing the fluoride. Zircon was used as the source of zirconium. The crude hydroxide was conveniently obtained by fusing the finely pulverized mineral with four parts of sodium carbonate, treating the resulting mass with hydrochloric acid, evaporating with an excess of sulphuric acid until the latter fumed, taking up with water, filtering and precipitating with ammonia. For purifying the zirconia, the method of Mitchell which has been advocated by Baskervillet was found convenient. This consists in dissolving the zirconium hydroxide in hydrochlorie acid, nearly neutraliz- ing with ammonia, adding a strong solution of sulphur dioxide and boiling. The precipitate, which, from the results of Venable and Baskerville,t appears to be a basic zirconium sulphite, cau readily be washed free from iron. The double salts were prepared by mixing solutions of the two fluorides in widely varying proportions, in the presence of more or less hydrofluoric acid, evaporating to the proper point and cooling. When cesium fluoride is in excess, even with very small amounts of zirconium fluoride, the salt 2CsI’. ZrF, is formed. It crystallizes in rather large, simple hexagonal plates, show- ing negative double refraction, and it can be recrystallized unchanged from water. When a larger proportion of zirconium fluoride is used, the salt CsI’. Zrl’,. H,O is obtained. This forms monoclinic erys- tals elongated in the direction of the 6 axis, and with faces which are usually too rough for accurate measurement. This salt also can be recrystallized unchanged from water. With a large excess of zirconium fluoride extremely small, difficultly soluble crystals of the salt 2CsF.3ZrF,.2H,O are produced. The small crystals havea slight action upon polar- ized light, but their form could not be made out. It does not recrystallize from water in a pure condition, the product being mixed with the 1:1 salt. To determine czsium and zirconium, the fluorides were con- verted into sulphates, then zirconium was separated from cesium by the use of ammonia, and zirconium oxide and ceesium sulphate were finally weighed. In order to determine fluorine a separate portion was dissolved in water, zirconium hydroxide was precipitated with ammonia, sodium carbonate * This Journal, xlvi, 188. + Jour. Am. Chem. Soc., xvi, 475. t Jour. Am, Chem. Soc., xvii, 448. 20 Wells and Foote—Double Fluorides, etc. was added in slight excess to the filtrate and all the ammonia was removed by evaporation. To the hot solution calcium nitrate was added, and the resulting precipitate, after ignition, was cautiously treated with dilute formic acid until, after evaporation on the water-bath, a further addition of the acid produced no effervescence. The calcium fluoride finally remaining after a final evaporation was washed, ignited and weighed. The results of the fluorine determinations were invariably somewhat low. The substitution of formic acid for the acetic acid usually used in removing calcium carbonate from the fluoride was sug- gested by the greater volatility of the first acid and the solu- bility of its calcium salt. We have found the modification to be an improvement as far as convenience is concerned, but we are not yet prepared to say that it is more accurate than the old method. Water was determined by heating the substance in a boat behind a layer of dry sodium carbonate in a combustion-tube, and collecting and weighing it in a calcium chloride tube. The following analyses of separate crops were made: 2CsF. ZrF, A. B. C. Calculated. Cesintng «2... -24- 56°41 at 55°51 56°60 ZITCONIMM 2 Fezeve.- 18°94 19°30 19°16 19°15 Minomne ssc. . DONS oT ee 24°25 Wetter -© 0 cee os", 160 0°98 0°97 The small amount of water found in the analyses was evi- dently mechanically included, for, under the microscope, bub- bles of mother-liquor could be occasionally seen within the erystals. CsF. 2rF, : H,O A. iB: Calculated. WS WIN Pe Rese he 38°44 39°58 ZAC OMG 22 ooo! 27°19 AES 26°79 Hliiornnes 2 ee 27°24 27°52 28°27 WV eatel ae eee ca ae 6°27 5°20 5°36 2CsF.. 3ZrF,. 2H.O A. B. Calculated. Cees se . ses 2 ae! 32°08 30°56 31°74 Z7irconiuiny sess. 22 32°45 33°48 30°99 Hhuorneree 2 eae. 31:09 30°43 S174 Waterco ee ae 4°40 3°96 4°30 Sheffield Scientific School, New Haven, Conn., July, 1895. Stanton and Vaughan—Section of the Cretaceous, etc. 21 Art. V.—Section of the Cretaceous at Hl Paso, Texas ;* by T. W. Sranron and T. WAYLAND VAUGHAN. THE detailed section of the Cretaceous of the Texas region here described is of interest because it is the most western of the outlying areas of the Texas Cretaceous as yet described, being 600 miles south of west from Denison, and 530 miles north of west from Austin.t Although the details of the El Paso section have not hitherto been published, many referencest have been made to it in the literature, and collections of fossils have been made there yew: (a and Dri EY A. Mearns, U.S. A.,~ Mr. Hill’s collection is at the Johns Hopkins University, and that of Dr. Mearns is deposited in the U. S. National Museum. The section§ here described was made in Mexico and New Mexico near the Initial Monument of the Mexican Boundary Survey, about three miles west of the city. The lowest part of it is exposed in the cutting of the Southern Pacific railroad on the west bank of the Rio Grande, where it cuts the pass through the mountains. The section extends from here to the top of the hill across the arroyo southeast of the Initial Monu- ment of the Boundary Survey. ‘The rocks are greatly faulted, folded and disturbed by igneous intrusions, so to obtain the sequence and thickness of the beds it was necessary to estab- lish horizons and measure between them where we could find them. Therefore, the highest beds of the section are not at the southeastern end of the exposures that we examined, but just north of a line joining the initial and second monuments of the Boundary Survey, where the rocks abut against a mass of hornblende porphyrite which is intruded through them. The downthrows of the faults are usually to the west. The following is a description of the section beginning with the highest bed : * Published by permission of the Director of the U. 8S. Geological Survey. + Gabb in vol. ii of the Paleontology, Geol. Survey of California, pp. 257-276, 1869, describes a collection of fossils made by Rémond from Sierra de las Conchas, near Arevichi, Sonora, Mexico. He recognized their resemblance to the Cre- taceous fossils of Texas. Mr. R. T. Hill says concerning these fossils, in this Journal, vol. xlv, p. 313, April, 1893, “The fossils are the characteristic fauna of the latest or Washita division of the Comanche Series, and resemble the varia- tions seen at Kl Paso and Juarez and throughout the northern littoral of the latest beds of the Comanche Series.” ¢ R. T. Hill: Bull. Geol. Soc. of America, vol. ii, pp. 517 and 518, May, 1891; this Journal, vol. xlv, p. 312, April, 1493; Bull. Geol. Soc. America, vol. v, p. 332, March, 1894; this Journal, vo}. 1, p. 233, September, 1895. T. W. ~ Stanton: In notes on a collection of fossils from near Belvidere, Kansas, pub- lished in R. T. Hill’s ‘‘ Outiying Areas of the Comanche Series in Kansas, Okla- homa, and New Mexico,” this Journal, vol. 1, p 218, Sept., 1895. § The description of the Denison Section published by R. T. Hill, Bull. Geol. Soe. of America, vol. v, pp. 303-304, and plate 13, March, 1891, should be con- sulted, while reading the description here presented. 22 Stanton and Vaughan—Section of the 10. Sandstone, white, yellow, or brownish, with shale beds... 52.2 aeee ae 200 ft.+ 9. Clay shales, with bands of limestone nodules, contain- ing great numbers of a large Hxopyra 722. eee 40 ft. 8. Hard limestone with the — same species of Hxogyra in the top... 2232 .eee 15 ft. . Sandstone, white or brown, 30 ft. + 6. Upper part of the bed, clay shales alternating with brown sandy flags. Brown sandy calcareous layers at base, containing Gryphea washitaensis Hill and Nodosaria texana in great abundance. The G. washitaensis is confined to base. 30 ft. above the base, Ostrea subovata. Ostrea quadri- plicata Shum. very abund- ant between 30 and 40 ft. 40 ft. above base zone of Kingena wacoensis _.--. 60 ft. 5. Flaggy argillaceous lime- stone, with shale partings. Layer of sandy flags at base. In the sandy flags, Cyprimeria. In the lime- stone bands Schleenbachia leonensis, Grypheu washi- taensis, Protocardia, Tri- gonia emoryt. ‘The zone of G. washitaensis is at the tOP. 21-25 5-423) -eee cee 4, Clay shales, with indurated calcareous bands. No fos- sils were found ._-.----- 125 ft. 3. Ledges of hard limestone. Fossils not numerous; at the base, there is a con- siderable number of an undetermined species of Gryphea, Gryphea tucum- carii, Neithea, and Schloen- bachia .2. 2. 2..° 2: 2 ee a. Kingena wacoensis, Ostrea quadriplicata, Ostrea subovata. 6. Grypheea washi- taensis and Nodosaria texana. c. Gryphcea washitaensis. d. Sandy layer with Cyprimeria. e. Sandy layer. ~T OF THE RIo GRANDE FROM KHL PASO, TEXAS. Seale =% in. = 100 ft. Washita Division. COLUMNAR SECTION OF CRETACEOUS NEAR INITIAL MONUMENT OF MEXICAN BOUNDARY SURVEY, ON WEST SIDE Fredericksburg Division. Cretaceous at El Paso, Texas. 2. Alternations of clay and soft argillaceous limestone ledges. Fossils: Hxogyra texana, Gryphea tucum- carti, Gryphea forniculata, Schlenbachia peruviana, 23 2 other species of Schlenbachia, and a large Neithea, 24 ft. 1. Argillaceous limestone, in thick ledges, weathering into nodular limestone, the nodules surrounded by clay. Fossils: Protocardia texana, Tylostoma peder- nalis, Trochus, Turritella, HExogyra texana, an Anchuroid genus, 2 sp. of Kchinoids, Pleuromya Mareen ONC Tee Usenid, CbC.. on oe 11 feet from the top Ostrea subovata (?) otal thieknesse 2350002 < 2. e698 ft. + The following is a list of species from the various beds: No. 9 of Section. Exogyra sp. A large species related to /. ponderosa Roemer. No. 6 of Section. Nodosaria texana Conrad. Terebratula (Kingena) wacoensis Roemer. Ostrea subovata Shumard. Typical examples of the species. Ostrea quadriplicata Shumard. Grypheza washitaensis Hill. Neithea texana (Roemer). Gervilliopsis invaginata White ? Plicatula incongrua Conrad. Modiola sp. Trigonia emoryi Conrad. Cardium (Protocardia) multistriatum Shumard ? Turritella sp. Related to 7. planilateris Conrad. No. 5 of Section. Pyrina parryi Hall. Enallaster texanus (Roemer). Epiaster elegans (Shumard). Gryphea washitaensis Hill. Neithea texana (Roemer). Lima n. sp. Lima wacoensis Roemer ? Modiola sp, Cardium (Protocardia) texanum Conrad. Turritella seriatim-granulata Roemer ? Aporrhais ? Distinct from those in No. 1. Natica ? : Schleenbachia leonensis (Conrad). 24. Stanton and Vaughan—Section of the No. 8 of Section. The fauna of No. 3 is essentially the same as that of No. 2, though the fossils are less numerous and not so well preserved. Gryphea forniculata White, Neithea tecana Roemer, Schlen- bachia belknapt Marcou, aud a few other forms were recognized. No fossils were seen in No. 4. No. 2 of Section. Enallaster texanus (Roemer). Holectypus planatus Roemer. Gry pbeea tucumcarii Marcou. Gryphea forniculata White. Exogyra texana Roemer.—Most of the specimens belong to a rather robust variety described by Conrad under the name Hxogyra fragosa. Plicatula incongrua Conrad. Neithea texana (Roemer). Pinna comancheana Cragin. Trigonia emoryi Conrad? Imperfect specimens that seem to be referable to this species. Cardium (Protocardia) texanum Conrad. Pholadomya sancti-sabe (Roemer) ? Pleuromya knowltoni Hill ?--Large specimens that prob- ably belong to this species. Tylostoma? sp. Turritella seriatim-granulata Roemer ? Schlenbachia peruviana von Buch. Schleenbachia belknapi Marcou. Schlenbachia sp. No. 1 of Section. Diplopodia texanum (Roemer). Enallaster texanus (Roemer). Ostrea subovata Shumard? A single specimen doubtfully referred to this species. Exogyra texana Roemer. Lima sp. Neithea texana (Roemer). Requienia texana (Roemer)? A large specimen probably of this species. Cardium (Protocardia) texanum Conrad. Pleuromya knowltoni Hill. Neritina n. sp. Tylostoma pedernalis (Roemer). Turritella sp. Aporrhais? Casts of two species. Cretaceous at Ll Paso, Texas. 25 Notes on the Fauna. Taking the fauna of the section as a whole, its essential iden- tity with that of the noted Tucumeari region in New Mexico and of other localities on the western and northern borders of the Lower Cretaceous area is at once apparent. The follow- ing list of species reported from Tucumcari is compiled by Mr RR. T. Hill im a recent, paper :* Turbinolia texana Conrad. Ostrea marshi Sowerby (=O. subovata Shumard). Ostrea quadriplicata Shumard. Gryphea dilatata, var. tucumcarit Marcou. Grypheea pitcheri Morton. Exogyra texana Roemer. Plicatula sp. Neithea occidentalis Conrad. Trigonia emoryi Conrad. Protocardia multistriata (Shumard). Protocardia texana Conrad. Cytherea leonensis Conrad. Pinna comancheana Cragin. Cardita belviderensis Cragin. Tapes belviderensis Cragin. Roudairia (?) quadrans Cragin. Cyprimeria sp. Turritella seriatim-granulata Roemer. Turritella marnochi White ? Ammonites leonensis Conrad. These species, according to Mr. Hill, come from sandy and shaly beds below the supposed Dakota sandstone with a total thickness of 115 feet. In the El Paso section, although there are one or two species that range from Nos. 1 to 6 inclusive, there are at least three well marked paleontological zones, each characterized by peculiar species that have been found useful as guide fossils in other parts of the Texan Lower Cretaceous area. Thus the fossils from No. indicate a horizon in the upper part of the Washita corresponding to the Denison beds, that portion of the Washita in the Denison section that overlies the restricted Fort Worth limestone. No. 5 contains Schlenbachia leonensis and Epiaster elegans, which are regarded as the most characteristic species of the Fort Worth limestone. * Outlying Areas of the Comanche Series in Kansas, Oklahoma and New Mexico, this Journal, III, vol. 1, p. 230, Sept., 1895. 26. Horbes—LEpidote from Huntington, Mass., and the Nos. 1, 2 and 3 may be grouped together as the lowest of the three zones, which shows a commingling of fossils that elsewhere occur in the lowest Washita and in the Fredericks- burg, with a few that range down into the Trinity division. In north Texas the same species of Ammonites occur in the Kiamitia and Duck Creek (the Preston beds) :at the base of the Washita and in the Goodland limestone, which belongs in the Fredericksburg. Gryphea forniculata is a Kiamitia species. Lxogyra texana is most abundant in the lower part of the Fredericksburg, though it ranges beyond both the lower and the upper limits of that division. The Echinoids also occur in both the Fredericksburg and the Washita. equienca tecana, Pleuromya knowltont and Tylostoma pedernalis are other forms that suggest a lower horizon than the Washita. The evidence of the fossils taken altogether is in favor of the view that a part of the Fredericksburg is represented here and that it grades into the basal Washita so imperceptibly that no paleontologic line can here be drawn between them. The Caprina limestone and its characteristic fauna, which in Central Texas, south of the Brazos river, usually affords such a distinct plane of separation, is absent here, as it is in north Texas and in all the outlying Lower Cretaceous areas of Kansas, Okla- homa and New Mexico.* ArT. VI.—On the Epidote from Huntington, Mass., and the Optical Properties of Epidote; by E. H. ForBEs. In the fall of 1892, Mr. W. L. Angell of Huntington, Mass., brought several crystals to the Mineralogical Laboratory of the Sheffield Scientific School for determination. They resembled zoisite in color and general appearance, but upon examination proved to be epidote. Their light color and the absence of the usual green of epidote indicated a low iron percentage and suggested a chemical and optical investigation. Abundant material was furnished by Mr. Angell, upon whose farm the mineral occurs, and some was also collected by Profs. S. L. Penfield and L. V. Pirsson, who visited the locality and found the epidote occurring in a seam in gneiss associated with quartz, biotite, albite and calcite. When the crystals project into cavities, they have a dark gray color and are clear and transparent, but when imbedded in the matrix they are much lighter in tone and less transparent, due undoubtedly to crush- ing, as the crystals are permeated by cracks and some are bent. * See this Journal, September, 1895, p. 234. Optical Properties of Epidote. 27 Carefully selected crystals were taken for the analysis, which was made in duplicate according to the usual methods. The ferrous iron was determined as described by Pratt.* The results are given below, together with the specific gravity determinations, which were made upon a chemical balance: Specific gravity.. 3°341 3°379 3°389 Average. .3°367 Average. Ratio. SiO, .... 38:20 37:78 37-99 633 6-00 Al,O, ... 29°00 30°06 2953 286 MO. 568) 571 5°67 ae ee MeOee O54, 052. 053. 007 MmOl 091 . 021 O21 003) 4386 413 CaO .... 23°92 23:37 23°85 426 MO... 205 208 2p4 1181-07 nr — Sd Total ___ 99°45 100°18 99°82 The above ratios approximate closely to 6, 8, 4, 1, which, regarding the water as coming from hydroxyl, gives the usually accepted formula H,Ca,A1,8i,O,, or Ca,(AlLOH)A1,Si,O,,, in which the calcium has been partly replaced by ferrous, and aluminium by ferric, iron. The analysis is interesting as show- ing a remarkably low iron percentage for epidote. The crystals, as is common with epidotes, show a prismatic development parallel to the 6 axis and some of the largest ones measured over 307" in length and 15™™ in diameter. They are usually simple, but some are twinned parallel to 100. Although the faces of the crystals were brilliant, yet the reflections of the signal on the goniometer were usually poor as the planes were commonly vicinal, which rendered the measurements somewhat uncertain. Some of the faces in the ortho-diagonal zone were deeply striated, so that there was a continuous series of reflections of the signal. Upon a suite of erystals which was measured, the following forms were identi- fied : fa, OIL vo, ML ie O12 Gq. 221 0, O11 0 HU. A'l of these were found upon a single crystal, as shown in the figure. On the simpler crystals the forms @, ¢, @é, 7 and nm were prominent and developed about like fig. 5, page 517, 6th edition of Dana’s Mineralogy. * This Journal, xlviii, p. 149, 1894. OT I > ——_— 28 Horbes—Hpidote from Huntington, Mass., and the The following is a list of some of the measured angles com- pared with the calculated ones derived from the axes of N. von Kokscharow Jr.* Q2b36= 75781421 21780362 =" 6 = Ge eb ioe } Measured. Calculated. Li Measured. Calculated. eng, 0012100 = G47 3IZUG622 So 5G. car, 001 « 10] = 63enr 62° 42’ anxm, 100), 110 = 54 34 54 59 54 cnk, 001,012) = sae ao) LO ene, 0014 101= 34 43 34 42 52 eam, ®0LAIIO=15 4 15 45 cao, OOLACIL = 58 32 58 27 45 ean, 001,111 > 1 Tt 15, Mey ance 100.101 = 29 48 29 54 caq, OOLA 221 = 89 32 89 42 Crt OOl A 102i — 34 00g as. 21 Each of the above is the mean of at least three separate measurements, and their agreement is very satisfactory. The optical investigations were made upon a single large crystal twinned about 100 and measured to make sure of the orientation. A section parallel to the clinopinacoid showed a slight zonal structure, and the central, or clearest portion, gave the following extinction: avec red, Li=1° 51’; yellow, Na, 2° 9’; green, Tl, 2° 12’, and for white light with the Bertrand ocular 2° 47’, all inclined to the ¢ axis in the obtuse angle P. The extinctions were measured on either side of the twinning plane and the above are half of the recorded angles. The value with the Bertrand ocular is considerably higher than the others, due undoubtedly to the influence of the blue and violet of white light. The Untersulzbach epidote with about 14 per cent of Fe,O,, gives, according to Klein,t aac for red 2° 56’ and for green, 2° 26’, but on the opposite side of the ¢ axis, that is, in the acute angle P. For the indices of refraction one prism was cut with a face parallel to 010 and its edge, as near as possible, parallel to the a axis, from which, by placing 010 perpendicular to the rays coming from the collimator, @ for yellow, Na =1°714 and y = 1:724 were determined. By minimum deviation @ was also found to be 1°714._ By means of a second prism parallel to the 6 axis, 8 was found to be 1°716. As the erystal under examination had a zonal structure, the results have only a rela- tive value, as they would be slightly different in another part of the crystal. The low double refraction y—a = 0:010 is below that of any recorded epidote. 7 From plates cut at right angles toa and c the axial angles were determined by measuring in a@-mono-bromnaphthalene (n, Na, = 16577) ona large Fuess axial-angle apparatus. From the plate at right angles to a, 2H,= 94° 33’ for yellow Na, and from the second plate 2H,= 93° 25’ were measured. From these values 2V, = 90° 32’, 2V,= 89° 28’ and 8 = 1°7144 were * Materialen zur Mineralogie Russlands, viii, p.44. + Jahrb. f. Min., 1874, p. 1. Optical Properties of Epidote. 29 obtained. In this crystal, a instead of being, as is usual, the acute becomes the obtuse bisectrix and the optical character is therefore positive. The dispersion was distinctly inclined. In the section at right angles toc, the color to the right of each hyperbola was violet and to the left red, but they were more diffused in one than in the other. The dispersion o, as in other epidotes. The pleochroism was quite striking, being in sections about 0:5™™ thick for rays vibrating parallel to 6 a lavender or pale plum-color, but parallel to a and c almost colorless, the former showing a slight tinge of lavender and the latter of green. By a comparison of the analysis of the Huntington with those of other epidotes, it was found that the light-colored one from Zillerthal in Tyrol, analyzed by Laspeyres,* was almost identical in chemical composition, as shown by the fol- lowing : SiO. . yas — 30 Lorbes—Epidote from Huntington, Mass, ete. For further comparison an iron determination was made, with the following results: Fe,O,= 6-97, FeO = 89 per cent, using as far as possible the green interior portion of the crys- tal, but owing to the scarcity of the material, some of the red part was also included. if only the green interior part had been used, these results would probably have been a trifle higher, but they are higher than those given by Laspeyres, whose analysis was made on carefully selected pale yellow crystals. The following table, containing the results already given, together with the optical determinations on epidote from Unter- sulzbach in Tyrol by Klein,* will show the changes in the opti- cal properties resulting from variations in the percentages of ferric iron : - 2Va, Na Locality. Fe,0; ‘Indices of refraction for yellow. measured 4 B ul V—2 over @. Untersulzbach, 14°0 1°73805 °1°7540 1°7677r-0°0372 "Ya a8 Zillerthal, 6°97 1°720 1°7245 1°7344y 0°0144 87 46 Huntington, 5°67 1°714 1716 1°724y 0-010) "OOMe2 With a decrease in the percentage of iron, the indices of refraction and the strength of the double refraction decrease, while the axial angle measured over a increases, so that in the Huntington epidote ¢ becomes the acute bisectrix and the erys- tal is positive with o MA MAvarrieas ee X.—Very Simple and Accurate Cathetometer; by F. L. O. Wapsworts. (With Plate °h) 222025 2 ere SCIENTIFIC INTELLIGENCE. Chemistry and Physics— Determination of Argon, SCHL@SING, JR., 49. “Graph: ea ite in Peematite. Moissan: Volumetric Determination of Cyanogen, DENIGHS, | 50 —Hmission of Light during Crystallization, BANDROWSKI, 51.—Properties of Titanium, Moissan: Proofs of ,Chemical Laws, CoRNISH, 52.—Molekularge-— wichtsbestimmung, Fucus: Manual of Organic Chemistry, CoHN: Organic | Chemistry, WHITELBY, 53.—Histoire de la Philosophie Atomistique, MABILLEAU: Copper Smelting, Prrers, Jr.: Light-emission power of bodies at high temper- | atures, Sr. Joan, 54.—Proof of the law of radiation, Winn and LuMMER: Plas- | § ticity of Ice, MU@ae: Magnetismus der Planeten, Leyst, 56.—Molecular Theory | of Matter, RisteEN: School Physics, AVERY, 57. Geology and Mineralogy—Gold Fields of Appalachians, BECKER, 57.—Geologi- cal Survey of New Jersey, 60.—Geological Survey of New York: Mollusca and | Crustacea of the Miocene of New Jersey, WHITFIELD, 61.— Climates of the | Geological past, DuBois, 62.—Geological Biology, WILtraMs: Mikroskopische Physiographie der massigen Gesteine, ROSENBUSCH, 63. —Rubies of Burma, | Brown and Jupp, 64.—New Minerals, 0. Botany—Pleiotaxy in the androscium of Epidendrum cochleatum, L.: Structure | and Development of the Mosses and Ferns, CAMPBELL, 72.—Phycotheca Boreali- | Americana, COLLINS, HOLDEN and SETOHELL, 73.—Introduction to the Study of | Seaweeds, Murray, 74. % Miscellaneous Scientific Intelligence—Comet of 1843 I, Krnurz: Harvard col. | lege Observatory Circulars: Magnetic Declination in ‘Alaska, Scuorr: Deter- minations of Latitude, etc., PRESTON, 75.—Relative Schwerebestimmungen : Physical Geography, Tarr: Herschels and Modern Astronomy, CLERKE: Justus von Liebig, SHENSTONE, 76.—Bibliotheca Zoologica II: Zoologisches _ Adressbuch: New York Academy of Sciences: Ostwald’s Klassiker der Exakten — Wissenschaften, 77. 4 Obituary—GrorGEe Lawson: M. S. BeBB: ANTONIO DEL CASTILLO, 78. Bara AMERICAN | JOURNAL OF SCIENCE. -Eprror: EDWARD S. DANA. f ASSOCIATE EDITORS 'Prorrssors GEO. L. GOODALE, JOHN TROWBRIDGE, -H. P. BOWDITCH awp W. G. FARLOW, or Camsrives, Prorrssors H. A. NEWTON, O. C. MARSH, A. E. VERRILL anpD H. S. WILLIAMS, or New Haven, _ Prorzssor GEORGE F. BARKER, or Pumapstenta, Prorrssor H. A. ROWLAND, oF Baxrimore, Mr. J. S. DILLER, oF Wasuinerton. FOURTH SERIES. VOL. I-[WHOLE NUMBER, CLI.] No. 2.—FEBRUARY, 1896. NEW HAVEN, CONNECTICUT. AS S.9 Gs f | : TUTTLE, MOREHOUSE & TAYLOR, PRINTERS, 125 TEMPLE STREET Published monthly. Six dollars per year (postage prepaid). $6.40 to foreign sub- ibers of countries in the Postal Union. Remittances should be made either by” mey orders, registered letters, or bank checks, MINERAL | : §WO NEW SPECIES. —-——-'Tn orthorhombic crystals of various habits and good lustre, from size suitable ae ee iol ee se __the microscope up to two inches in length, grouped i in Margarite schist. Resem-— ae na bles the pale sapphire and gray shades of Corundum or Cyanite. iv aes . Northupite, from Borax Lake (chlorocarbonate of sodium and magnesium)._ vA : new and perfectly stable compound, crystallizing i in beautiful octahedrons, which ae : show an interesting internal structure. us ; Opalized Wood, from Idaho. A recent discovery. It is accepted by all es have examined it, as the most perfect example of petrifaction ever seen. Sections ‘A of limbs three inches and upwards in diameter exhibit beautiful graining, with rings, | cells, and fibre of the wood replaced in minutest detail by a lustrous brown-yellow Se! opal. As with both Lawsonite and Northupite, we were fortunate enough to : secure the entire find. and are enabled to offer choice speeimens of all three at lh very low prices. : Also: Variscite, Crystallized Cryolite, Noselite Diadochite, Koninckite, Richellite, — Axinite, Beautiful quartzoid crystals of Endlichite, and many other rare species. {Any of the above will be sent on approval. : Illustrated Catalogue 128 pp., containing complete price lists with des- cription of collections, etc.; and all circulars free. Send for our last “ Sup- plement” descriptive of recent important arrivals. \ SCIENTIFIC BOOKS, PAMPHLETS, PAPERS, PROCEEDINGS AND TRANSACTIONS OF SOCIETIES, JOURNALS, ETC. ee An immense stock. Catalogues are published in all branches of Natural and Physical Sciences, such as Geology, Mineralogy, Mining, Botany, Eth- nology, Zoology, General Science, Microscopy, Physics, Electricity, eae: ‘ Agriculture, Education, and Medicine in all branches. Phiten All American books in print furnished on order, and foreign Res an obtained within a few weeks. ae ~ We make a specialty of Papers published in the Proceedings and Transac- __ ue tions of Scientific Societies, and many of the scientific men of the country, a tess : e ea es recognizing the advantage of such an exchange, have placed copies of their ao papers in our hands for sale. With our ever-changing and constantly be increasing stock, we can supply old and recent literature on all scientific sub- — a jects, and lists of books and papers referring to special subjects will be fur- = nished gratis on application. a Dr. A. E. FOOTE, ae | WaRREN M. Foote, Manager. | ae ec _~—i“‘<“=ésédcq BAA~-2G-2S North Forty-First Street,. oe PHILADELPHIA, PA, U.S.A. > Established 1876. ; ‘ i THE AMERICAN JOURNAL OF SCIENCE [FOURTH SERIES.,] oO Art. XI.— Researches in Acoustics ; by ALFRED M. Mayer. *# Paper No. 10 containing The Variation of the Modulus of Elasticity with Change of Temperature as determined by the Transverse Vibration of Bars at Various Temperatures. The Acoustical Properties of Aluminum. [Read before the British Association, at Oxford, Aug., 1894.] Summary of the Research. Potsson in his Zrazté de Méchanique (Paris, 1833, t. II, pp. 368-392)* discusses the laws of the transverse vibrations of a bar free at its ends and supported under its two nodes. He shows that the frequency of the vibrations of the bar is given by an equation, which, reduced to its simplest expression, fe = VX 1°027 9 in which N is the number of vibrations per second of the bar, ¢ its thickness, / its length and V the velocity of sound in the direction of the length of the bar. To ascertain how near the frequency of the transverse vibra- tions of a bar, computed by Poisson’s formula, agrees with the result obtained by experiment, the following method of experi- menting was used. Rods of steel, aluminum, brass, glass and of American white- pine (Pinus strobus)—substances differing greatly in their moduli of elasticity, densities and physical structures—were carefully wrought so as to have the length of 1°5-: meter; the thickness of 0°5°, the width of 2°’, and a uniform section throughout their lengths. The velocity of sound in these rods was determined by vibrating them longitudinally at a tempera- tute of 20°, while held between the thumb and forefinger, and ma * See, also, The Theory of Sound, by Lord Rayleigh, 1894, vol. i, ch. 8. Am. Jour. Sc1.—FourtH Series, Vou. I, No 2.—FEBRUARY, 1896. 6 82 A. M. Mayer— Researches in Acoustics. theingfrequencies of vibration ascertained by the standard forks of Dr. R. Keenig’s tonometer. Out of each of these long rods were cut three bars of the length of 20°™*, and these bars, also at 20°, were supported on threads at their nodes, vibrated transversely by striking them at their center with a rubber hammer, and their frequencies of vibration determined by the forks of the tonometer. The mean departure of the observed from the computed numbers of transverse vibrations (see Table I) is 3,; the computed frequency being always in excess of the observed, except in the case of glass, where the computed is 34, below the observed frequency. In Table I, 7 = length and ¢ = thickness of bar in centime- ters at 20°. V = velocity of sound in centimeters in bar at 20°. N = number of vibrations per second at 20°. The close agreement of the computed and observed values shows that, by vibrating a bar at various temperatures, the varia- tion of its modulus of elasticity with change in its temperature can be obtained. We observe, N, at various temperatures of the 2 bar; then V = 2 ee is computed, and the modulus M = ei f 1-0279- ; 7 As t, /, and d (the density of the bar), vary with the tempera- ture, the coefficient of expansion of each bar and its density at 4° were determined, so that the dimensions and density of the bar could be computed for each of the temperatures at which it was vibrated. Experiments were made on five bars of different steels, on two of aluminum, on one of St. Gobain glass, one of brass, one of bell-metal, one of zinc and one of silver. The results of these experiments may be summed up as follows: The modulus of elasticity of St. Gobain glass is 1:16 per ct. less at 100° than at 0°. te & the five steels ‘' 2°24 to3-09 s ie a brass a 3°73 tb ‘6 bb 3 : bell-metal tb 4:3 cc cc bb (a3 aluminum bb 5:5 « 6b 6c “cb silver a3 9°47 “cc 60° 73 c H zine “6°04 6208 ie The decrease of the modulus of elasticity of glass, aluminum and brass is proportional to the increase of temperature; straight lines referred to codrdinates giving the results of experiments on these substances. ‘The five steels, silver and zine give curves, convex upwards, showing that the modulus decreases more rapidly than the increment of temperature ; while bell-metal alone gives a curve which is concave upwards ; its modulus decreasing less than the increment of temperature. (See fig. 5, p. 95.) A. M. Mayer—fesearches in Acoustics. 83 The more carbon a steel contains the less is the fall of its modulus of elasticity on elevating the temperature of the steel. Thus, the modulus of the steel with 1:286 per cent of carbon is 2°24 per cent less at 100° than at 0°, while the steel con- taining 0°15 per cent of carbon has a modulus at 100° which is 3:09 per cent lower than its modulus at 0°. So far as experiments on a single steel containing nickel permits of any general deductions, it appears that the presence of nickel in a low carbon steel lowers its modulus of elasticity. Thus, steels Nos. 3 and 4 having respectively -47 and °51 per cent of carbon have a modulus of 2131 x 10°, while steel No. 5, containing -27 per cent of carbon and 3 per cent of nickel, has a modulus of 2080 x 10°, which is 2°12 per cent. lower than that of steels Nos. 3 and 4. The presence of nickel in a steel may, in a diminished degree, have the effect of carbon in lessening the lowering of the modulus when the temperature of the steel is increased. Thus, the percentage of the lowering of the modulus, by heating from 0° to 100°, of steel No. 5 containing 0°27 of carbon and 3 per cent of nickel, is the same as that of steel No. 3 with 0°47 per cent of carbon. If a bar of any one of the substances experimented on is struck with the same energy of blow, by letting fall on the centre of the bar a rather hard rubber ball from a fixed height, the sound emitted by the bar diminishes in intensity and in duration as the temperature of the bar is raised. Thus: Brass at 0° vibrates during 75 secs.; at 100° it vibrates during 45 secs. Bell-Metal Be Ke aT a “ T5ps Aluminum 74 te 40 44 tc ib 12 sc J.&C. Cast Steel * a ad x eg yee Bessemer Steel ‘ se Aas oe ee che we St. Gobain Glass ‘ a (Re Oe “ oe 3s Zine at 0° vibrated during 5 secs.; at 20° only during 1°5 secs. At 62° it vibrated for so short a time that it only gave three beats with forks of 1090 and 1082 v.s. At 80° it was not possible to determine the pitch of the bar, and at 100° the bar when struck gave the sound of a thud. The bar of silver acted in a similar manner to the bar of zinc—it was even less sonorous than zinc—thus flatly denying the “silvery tones” attributed to it. These phenomena do not depend on the fall of modulus but on changes in the structure of the metal on heating, which cause the blow to heat the bar and not to make it vibrate. Bell-metal was found to be an alloy peculiarly well suited for bells, as the intensity and duration of its vibrations were the same at 50° as at 0°; all other substances showing a / marked diminution of intensity and duration of sound at 50°. 84 A. M. Mayer— Researches in Acoustics. A bar of unannealed drawn brass, after it has been heated to 100°, has its modulus increased 35, per cent. See Table IIE and fig. 11, p. 98. In this research I had the good fortune to have had the assistance of Dr. Rudolph Keenig, of Paris. He not only placed at my service the resources of his laboratory and work- shop, but generously gave me constant assistance during the experiments; making the determinations of the numbers of vibration of the rods and bars with the standard forks of his tonometer. Without his aid this work could not have been done. For instance, in the cases of the bars of silver and zine the beats they give with a fork are so few that they cannot be compared with a chronometer; but Dr. Keenig, from his long experience in the estimation of beats, was enabled to form an accurate judgment of their number per second from the rhythm of the beats. The determination of pitches extending through such a range of vibrations as occur in this research can only be made with Dr. Kenig’s “grand tonométre,” a unique apparatus of precision, giving the frequency of vibra- tions from 32 to 43690 v.s., and really indispensable to the physicist who would engage in precise quantitative work in acoustics. We now proceed to give accounts of the several operations performed in the progress of this research. Determination of the Velocity of Sound in Rods. In the determinations of the velocity of sound in the rods of 1:5 m. in length I used the method of Chladni.* Kundt’s method of obtaining nodal lines of fine powders in a tube, by vibrating a rod whose end carries a cork which fits loosely the end of the tube, is not accurate. The weight and friction of the cork, the necessity of firmly clamping the rod at a node, and, above all, the want of knowledge of the velocity of sound in the air of glass tubes of different diameters, renders this method, so beautiful and ingenious, worthless for accurate measures of the velocity of sound in solids. The curves in fig. 1 show the very diverse determinations of the velocity of sound in the air in tubes of different diam- eters by the physicists Kundtt, Schneebelit, Seebeck§, and Keyser|. The velocity of sound in metres is given on the axis of Y ; the diameter of the tube in centimeters on the axis of X. Ku stands for Kundt, Sch for Schneebeli, Se for See- beck, and Ke for Keyser. The most precise measures of * Traité d’Acoustique, Paris, 1809, p. 318 e7¢ seq. + Bericht. der Akad. der Wiss. zu Berlin, 1867. +t Poge. Ann.. 1869, vol. cxxxvi. . § Pogg. Ann., 1870, vol. exxxix. || Wied. Ann., 1877, vol. ii, p. 218. A. M. Mayer—Researches in Acoustics. 85 velocities are those of Keyser, who closed the end of the tube with a cork attached to the end of a steel-bar, while the other end of the bar was securely clamped. The frequency of the transverse vibrations of the bar was registered by a style describing the sinusoids of the vibrating bar. Thus the I. 5 cues ER Ht +] ++ HE rH FE HHH -|--4 PE fit] HH ttt tt StH surissteciiort seiseiaet : seeeiiocit FH H ii tasee efeitos Beeeeed deeeeGeseeesert Ee dreeereeiteet nist Feber: feuae Heed Beseeee seas seriocttssis stitte Beeeerd eee sitertaret sarecaates 4 SegesdseesCeesacesessras! TT] Baaeen EHH REE EH | HoH seecniestiicdt See sastete fasteSEH SEGRE tte stict tie ee sissies Heap! sau SsESuaaneuaRucee’, On DBBSG8 BEDSDSEEERSoEEseoeSse HEH soecazoecees SRS SR CURE EREREES 40 CeeRs Jf fp SSGRS GEES PASPs SaSRaBeees Py CES RESREOE EbeSS2e/,. SR POR CRGESas Se2R2aaeeee DEGSS PRESS GBR0R GOSGR 00000 Se2e00R8 pp — So esenusase50aS0505000—SE05eRes5eeses5e505 BROOD GOSE GSN SOE. 22 O0Me0 SeOGo BREE oeeosS EEC EECH SP BEERE eaeS8E5 Por ae x SeueED dugegseees soeeen) Sou senneueees soeSceeUceese 0000 SOEs 6eVe seen eee OSRORDSRRS DBaaSF SERB eaEEs 8 GHSHR BORSE BBSRO SOB E8 Saab” GbE RE BRSEBEeoRs wee seeespueeeceued! Goce boccecessscsseunus rH sepgereycotsonssecttoe eecdtarrents ERE Tar cee ge ane tesuneul EEO) la tun RPTL bed ciesesenseers ee es ae : see Hefti sara! rH He 4nB Ho He rH CH eS A BeGESEaR8 HA Besas0ga ane i Pry BEE Ha EEE PEEEEEEH Het cI scare Eee et Eee ee eee eater ee aaa oH Su USEC CS peecepSH SUE EEEE CE CoeoE ceece Secu r Cec ce ccc ceeeceEeoaE aS caee H Sep ssceseuae oo) HERR sfectesient Sueceeeriieet seassueee Seaton) ats? featessteastin: ttasteestteates BpSresritesicae: g ee SE HE Serer nee esstncatt Saeunet gue seeuntee EEEEEEEEEEEEEEE fusee! (ee, 4 /Seuuseae megugeauees srfaes en) ftssteeat Sreiistisrtastenitictt Py setiate| etectecst sadtcsttsseeacttasttostt 3 Bia ce H Peon aH ier iaut scsteiieastiiitaas a iait iamragsnadial suetasariasiiasitt eae scce genus 34 Hysciteciscsttastosiastict 4 PAEEEEEEEEEEEEEEEE EEE E suiesstesttoctteciie: siittt 3 SRSEHESEHECeeaatoat cst Sosadtaceetfecrsttfessittfosuitifacuittte HE weight and friction of the cork introduced no error. In a similar manner I obtained the velocity, marked M in fig. 1, by vibrating a rod of aluminum. The frequency of the vibra- ~ tions of the rod were measured while the cork at the end of the rod was vibrating in the mouth of the tube. The result agrees closely with Keyser’s. It is needless to discuss the curves of fig. 1. ——— - eS a a = SS Ee “s arco 86 A. M. Mayer—fesearches in Acoustics. The method of Chladni, used exactly as that eminent man used it, remains the best we have. It is important, however, to note that the rod must be held between the thumb and forefinger when it is vibrated and not clamped when vibrated. When clamped it always gives a higher frequency, as shown by the following experiments : Steel rod clamped.._--.----- 3429°2 Aluminum rod clamped_-.-.____- 33770 Steel rod held between fingers 3428°4 Aluminum rod held between fingers 337674 The frequency of the vibrations of the rods of steel, brass, aluminum, glass and pine wood, when held at the middle of their lengths and vibrated so as to give their fundamental tones, gave exactly the octaves of these fundamental tones when held at one-quarter of their lengths and vibrated. Determination of the lengths of the long rods and of the lengths and thicknesses of the bars. The lengths of the rods of 15+ meters were ascertained by comparison with the rod of steel whose length was meas- ured at the Bureau International des Poids et Mesures. The lengths and thicknesses of the bars which were vibrated trans- versely were measured with micrometer calipers. The read- ings of these calipers were tested by comparison at 20° with a series of standards of various lengths of inches and fractions of inches, made for me with great care by Mr. George M. Bond, who has charge of the gauge department of the Pratt, Whitney Co. In reducing the comparisons to centimeters I adopted the value of the inch as equal to 25-4 millimeters. In obtaining the length of a bar, the mean of several measures in the axis of the bar and in directions parallel to the axis and at various distances from it was adopted. The thickness of a bar was taken as the mean of measures taken throughout the length of the bar at points designated by the intersections of lines drawn parallel and at right angles to the axis of the bar. The dimensions of the bars were measured at 20°, except those of steels Nos. 3, 4, 5, which were measured at 18°°25. Determinations of the Coefficients of Hxpansion of the Bars. To determine the coefficients of expansion of the bars I devised the apparatus shown in fig. 2. In a brass tube, T, the bar, B, rests in slots in the supports, 8,8’. The tube, T, is slightly shorter than the bar, B. Washers of rubber (shown in black in the figure), of the same diameter as the outside diameter of the tube, are placed in the screw-caps, C, C’. These washers are perforated with holes of diameters smaller than the thickness of the bar. When the caps are screwed up Researches in Acoustics. 87 A. M. Mayer the rubber washers press against the ends of the bar. This pressure is further increased by flat rings which surround the holes in the washers and are pressed against these washers by means of the springs, D, D’.. By this arrangement the sur- faces of the ends of the bars are exposed, while the contact of the washers on the bars makes a water and steam tight joint. bo Lal ANNAN 2 F R G Thus the bar may be surrounded with ice, or, with steam, or, with a current of water of different temperatures, and be cooled or heated up to its terminal planes, while the holes in the washers allow the micrometer screws, M, M’, to be brought in contact with the terminal planes of the bar. Two helical springs are attached to the column, A. The other ends of these springs are fastened to rods projecting from the tube, T. Thus the same pressure of contact is always made between the bar and the end of the micrometer screw, M. The tube, T, is supported in Vs, V, V’, and the greater part of the weight of the tube is taken off the Vs by helical springs fastened toa frame above the apparatus. The tension of these springs can be so regulated that the tube rests on the Vs with the same pressure when the tube has steam passing through it or when it is filled with ice. The column, A, and the Vs, V, V’, are _insulated from the base of the apparatus by thin plates of ebonite, e. Between the binding screws, E and E’, and con- nected by wires, are the voltaic cell, F, the galvanometer, G, 88 A. M. Mayer—Researches in Acoustics. and a box of resistance coils, R. The micrometer-screw, M’, with which the variations in length of a bar are measured, is mounted as follows: The screw passes through its nut ina massive brass plate which rotates around nicely fitted centers at H. These centers are supported by two side plates not shown in the figure. A spring, K, is fastened to the lower part of the swinging nut-plate and brings this plate against the plate, L, firmly fastened to the base of the apparatus. When the swinging plate is vertical and the axis of the screw hori- zontal the swinging plate fits accurately the surface of the fixed plate, L. By turning the rod, N, the swinging plate and its screw can be rotated away from the bar. This arrangement allows the screw to be swung out of the way while the tube, T, is being placed in the Vs. Also, it prevents any strain between the micrometer-screw, M’, and the column, A; which would take place if M’ were fixed and it should be brought in con- tact with a hot bar in the tube, T. With careful manipulation successive electric-contacts can be made on a bar in the tube, T, surrounded by ice, so that the variations in a series of measures will not exceed 4,7 mm., with a resistance of about 200 ohms placed in the circuit. It may be reasonably objected to this apparatus that when the micrometer-screw touches the bar at 0° it is cooled and shortened, and that when it touches the bar at 100°, or at temperatures higher than that of the screw, the latter is heated and elongated. This error, however, is quite small, and may be neglected in our work. If we assume that one centimeter of the screw is heated 10°, which is a large estimate, consid- ering the duration of contact of screw and bar during a measure, the shortening or elongation of 1 em. of the screw by gout or heating it 10° amounts to only 0012 mm., or TBDEtt us the length of the bar. This change in the length only -00000006. Determination of the Densities of the Bars at 4°. The bar, whose density was to be determined, was immersed in water at 4° for a couple of hours. The bar was then sus- pended by a platinum wire in water at 4° and weighed. The bar was then removed from the wire and a quantity of water equal in volume to the volume of the bar was added to the water in the vessel, and the platinum wire, now immersed ex- actly as it was when the bar was attached to it, was weighed. This weight, subtracted from the previous weighing, gave the weight of the bar in water. Every precaution was taken to prevent, by means of screens, the action on the balance of A. M. Mayer—fesearches in Acoustics. 89 the currents of cold air in the balance-case, which are _pro- duced by the constant descent of air from the sides of the cool vessel. The Apparatus in which the bars were heated and cooled. On the precautions used so that one is sure of having the real temperature of the bar when it is vibrated. The apparatus used to heat and cool the bar is shown in fig. 3. Ina brass box, C, is inclosed a box, C’, containing the bar, B, supported on its modes, N, N, by threads held by upright rods. From this central box two tubes, T, P, pass through the outer box C. The inner box is made water-tight and steam-tight by a rubber washer which is pressed between the top of the box and its cover by means of screws. Through the tube, T, the bar is vibrated by letting fall upon its center a rubber ball fastened to a light wooden rod. On the blow of _ the ball it rebounds and the rod is caught by the fingers in its upward motion. The cork is then at once replaced in the 90 A. M. Mayer—fesearches in Acoustics. tube, IT. The sound from the bar is conveyed to the ear, at EK, by means of a tube (fig. 4). One branch of this bifurcated tube leads through a rubber tube to the pipe, P, of the box, fig. 83. The other branch leads to the fork, F, the number of whose beats per second made with the vibrating bar is meas- ured by a chronometer. The pipe, 8, allows the steam to issue when water is boiled in the box, C’, by a gas lamp. The flow of gas through this lamp was neatly regulated by a stop- cock turned by a long lever. The box, C, is covered, except at the bottom, with thick felt. To determine the frequencies of vibration of a bar through a range of temperature from 0° to 100°, the following method was used: The box, OC, was filled with ice, surrounding the inner box, C’. It thus remained for an hour so that the boxes were cooled down to 0’, and the moisture in the inner box has been condensed so far as it can be at 0°. The bar, which has been in ice for two hours, is wiped dry and quickly intro- duced into the inner box. A thermometer, T (made by Baudin and corrected), entered the boxes through stuffing boxes, and whose bulb touched the under surface of the bar, was read till it became stationary. The bar was now vibrated and its frequency of vibration determined for the temperature given by the thermometer. 3 The lamp was now placed under the box and the water in it boiled till the thermometer reached its maximum reading and the reading remained stationary during a half-hour. The vibration frequency at this temperature was taken. The flame of the lamp was now lowered and the box allowed to cool very slowly, at the rate of 1° fall of temperature in about eight min- utes. When the thermometer read 80°, 60°, 40°, the flame of the lamp was carefully adjusted, so that these successive tem- A. M. Mayer—fesearches in Acoustics. vit peratures were maintained during 15 minutes. We then took the frequency of vibration of the bar. The numbers of vibrations of the forks used in the deter- -minations of the pitches of the bars were corrected for tem- perature by the coefficient -0001118, determined by Dr. Keenig in 1880 (Quelques Experiences d’Acoustique, Paris, 1882, p. 172, e¢ seq). The subsequent tables show the results of the experiments and give the computations of velocities and moduli founded on them. The curves express graphically the effect of change of temperature on the modulus of elasticity of all the bars experimented on. The circles, on or near the curves, give the data as determined by the experiments. In Table III, T=temperature of bars, =the length, =the thickness, and V=the velocity of sound through the bars, in centimeters. M=the modulus in grams per square centi- meter section of the bar. g, at Paris, equals 980-96. D=the density, and N=the number of vibrations of bar per second at temperature, T. All of the bars were annealed, except those of Jonas and Colver steel, of the French aluminum and of brass; these were experimented on just as they came from the draw- bench. For the analyses of the substances of the bars experimented on, [ am indebted to my colleagues, Professors Stillman and Leeds. TABLE I. | N computed by | IN obsertad 20°= Bar. 1. t. Vi at 20° = N=vV 1-02795 at 20°. Diff. Steel No. 1 __| 20°022 | -5025 |150-02x 3427-4 v.s.| 662-49 v.d. 660:8 |-i-69— 17 No. 2 __) 20:0246 |-5037 |=5141738ms G63;91ny OGD EP ONES “ No. 3 __| 20°0225 | 5022 GOZO 660°3 +1T=+st5 Aluminum. INO. U2 2. 4- 20°0253 | -4993 |150°05 x 3377 v.s. 64851 ‘* 646°6 +191= 5h, Ona eee 200296 | °4991 |—5s06719™s CAEO en 647°0 +0°9T= gthe INO: 8 2k... 20-0233 | -4998 | 649°80 “ 6480) ) i-llcS Brass. | INO: 1 Soe 20°02 °b0116 |150°05 x 23864 v.s.) 460°23 ‘ 4590 =| 4+1:23= ahy INO, 2) o2 5 20°02 50147 | =358079ems | 460°53 ‘* 458°95 |+158= 54) NiO. Bi... 20°02 50108 460°'16 ‘ 458°35 |+18l= sh St.Gobain | 23'516 [747 [152-2 x 3582 v. s. 147-03“ (ASE | OAs Glass ___.. Os oolocms | White Pine. [41°15 |-803 |171-18x 3072-75 v.s.| 25638 “« 2560 |+0°38= gt, Density =°365 —O2o9aus Mean departure of computed from observed value =z}, of observed value. 92 A. M. Mayer—Researches in Acoustics. TABLE II. Table of Analyses, of Densities at 4°, and of Coefficients of Hapansion of Bars. Iron. |Carbon|Silicon| Phos. sulph. Mang. Nickel eel | 98259] 1°286 | 0015) 0059] 0:031 0°350| ____ | Density @ 4°, 7°821, coef’t. expan., 98°738) 0°47 | 0°15 | 0°022| ___. ORG Za oe °848, ee a 98°628] 0-51 | 0°158) 0:024) __._ | 0°68 | .__- of CS dae xe af GoD OT ee OOO O34) sacs 0°69 | 3189 - on” Aceon, cia 99°03") 0:15 +) 0-02 +0709 | 0°06 0765) 2-22 me OSE, 74 Brass Bell Metal. Coppers. seh GaG me Sikes 5 cea g 64.34 Copper 2: .:c2. 2.) Eee eee 80°08 ZAMING aes o e es Se epee Se 8 34:97 Tin). i202 See 18°97 1 7c) Kp ma OB tea, Oe ME 58 Lead 2... .2.csggee eee 12 TROT es fen Bie eer ee eae iil Zi’. 2.2 eee ae 49 Density ieee nic ty saree mre 8°476 Density =. .2 23. ose eee 8°347 Coetts expan) sae seas 0000185 Coeft. ‘expan: 22 2.22 ees 0000187 Aluminum (Amer.). Aluminum (French). Aluminum Yh Ss {eeee ee eee 98:99 Aluminum, . <2 25ers meee 97°80 Free Carbon (graphite)__._..-- al Carbon with St. 2222 22e"— == 14 CombincdsCarbon.-se eee "16 “ frees... 2S eee 04 EDIE MOREY Saat es lea ARNE om stew S 21 «with, Copper 2225252222 09 Silicone feges Bin My i eee oe 32 Coppers : 2.2 222 eee 1°29 Tine is 5 acing et soe eee alts Silicon... - 2-5 cee eee "64 ID eVaNShliine re eer eb ae 2-702 Density —.... ee 5 230 Cockiviexpan: 2s. \ ee eee 0000232 Coeft. expan. . 20-2 aaa "000022 Silver, Pure. Zinc Density. oO Sask kee 10°512 Zine... 25.) ea 99°75 Coctisexpam. pas aoe ea 0000184 lron’.2e=. 2 26 ee 10 Lead 2232 04 Density .. 22-422 6°8107 Coeft. expan. --- <2 = 45 26 oe "0000296 St. Gobain Glass. Density scene ser Coeft. expan. _._.. LEB peste ot ph 2°545 afer h ae 00000777 (Fizeau.) 0000110 0000118 0000120 0000119 "0000122 —— Bar. Jonas & Colver Cast Steel. Steel No. 3. Steel No. 4. Steel No. 5. Bessemer Steel Brass when bar was cooled from 99°°6 tok. Brass before — bar was heat- ed to 99° 6. Bell Metal. —— 0°2 20 40 61 80 99°8 0 18°25 34 60 80 99°5. 0 18°25 40 80 100 te 21°8 40 60 80 99°6 0-4 20 A. M. Mayer—Researches in Acoustics. 20°0207 20°0246 20°0286 20°0328 29-0366 20°0406 20°3513 20°3558 20°3594 20°3657 20°3705 20°3752 20°3517 20°3562 20°3614 20 3643 20°3715 20°3755 20°3513 20°3559 20°3609 20°3657 20°3706 |20°3753 '20°451 (20°456 20°461 (20-466 (20-471 |20°476 |22°0128 22-0208 22-0274 (22-0348 22 0422 (22-0494 |20°0127 /20°0208 (22-2402 (22-2490 (22-2568 '22°2610 |22°2650 22-2780 '22:2818 5036 5037 5038 °5039 5040 5041 °64049 64063 64074 "64094 °64109 64124 "64295 64312 64328 | 64337 64360 "64373 64188 64202 “64218 64233! 64248 64264) 60144 60160) "601 76) 60190 "60205, 60220) 5009 | 011 *50126) "50144 "50162 50180) 5008 ‘D011 "82054 -82114| 82168! 82196 82224 82266 82338 1 93 TABLE III. N | a | 1 : 102795, g | 0012564 7-828 |662:0 | 512602 Dee er san 5108 ‘0012562 7-823 |661°0 | 511908 |2089805486 2090 x 10 0012561, 7°818 |659°6 | 510112 |2080863837 2081 x 108 0012560 %7°814 16586 | 509389 |2072936852) 2073 x 108 ‘0012554, 7-809 65771 | 508470 |2064153151) 2064 x 108 0012551) 7°804 |655°14) 507065 |2050537870| 2050 x 10° 0015464 7:849 |820°38) 516124 |2131430318) 2131 x 108 0015460 7-844 |818°91| 515329 |2123515662) 2123 x 108 ‘0015457 7-8395|817°50| 514539 |2115797406| 2116 x 10° 0015453) 7-832 |814-°70) 512907 |2100383667 2100 x 108 0015449 7:8265:812°40/ 511586 !2088031273) 2088 x 108 0015446 7-821 |810°00| 510175 (2075047288) 2075 x 108 0015523, 7-846 |824°71/ 516238 /|2131556968) 2135 x 108 0015520 7°841 |822-71/ 515708 |2125826869| 2126 x 108 0015517) 7°8345|821:25) 514893 |2117356581) 2117 x 108 0015514) 7°831 |819°67) 514002 (2109091886) 2109 x 108 "0015508) 7-823 |816°30| 512088 2091259452) 2091 x 108 ‘0015505 7-819 |814:00) 510745 /|2079256951) 2079 x 108 0015498) 7-852 |813:29| 510527 |2086250350) 2086 x 108 0015494) 7-847 |811°84| 509757 2078637463 2078 x 108 ‘0015490 7-841 |809-68) 508529 |2067052961| 2067 x 108 ‘0015486 7°835 /807°31) 507168 2054430194) 2054 x 106 0015482) 7°829 |805°02) 505856 2042249532) 2042 x 108 0015479) 7824 |802°71) 504506 |2030066264) 2030 x 108 0014380) 7°8421|/761:37| 515093 Peete ve aaa ne ‘0u14377| 7°8364)759-80| 514032 |2110783287| 2111 x 108 0014374) 7-8306|757°70| 512960 |2100440462! 2100 x 108 0014370, 7-8248)755°41| 511409 /|2086216940 2086 x 108 -0014367| 7°8192/752°90| 509710 |2070893085| 2071 x 108 0014363) 7°8134/749°90| 508026 (2055702283) 2056 x 108 | 0012506) 8-4774/460-64) 358337 |1109672142| 1109 x 108 0012501) 84677|458:95| 357307 |1102010322| 1102 x 108 0012497) 8:4592/457-40| 356103 1093479554) 1093 x 108 0012492) 8-4498/455-70| 355052 1085864045) 1086 x 108 0012488] 8-4404/454°00| 353727 |1076578495) 1076 x 108 0012483) 8 4312|/452°30) 352678 |1069044150| 1069 x 108 0012504| 8-4778|/460-°04, 357928 |1107191685/ 1107 x 106 ‘0012501| 8-4685/458°35| 356698 |1098388907| 1098 x 108 0016589) 8-3490/572°05| 335300 956864118/956°8 x 108 0016588) 8°3390/569°50| 333806 947214927/947-2 x 108 - 0016587) 8°3302/567°12) 332411 938324839 938-3 x 108 -0016587| 8°3256/566°02, 331766 934172231/934°2 x 108 (0016586) 8:3208/564:94) 331133 930074354|930°1 x 108 0016586] 8°3144/563°61| 330354 924989117/925°0 x 108 0016584) 8°3025|561'21) 328946 915914824/915-9 x 106 94 A. M. Mayer— Researches in Acoustics. Ban vr | 2 cae De) ayo Ya ar M G 102795 g Aluminum 0-4 |/21°612 | -55176| 0011813) 2°7027/621:00 511423 720621232/720°6 x 106 (American). |20 |21°622 | -55200/ 0011807} 2°6990/618:10) 509032 712929083/712°9 x 108 45 /|21°6346! *55232| 0011800); 2°6943)613:90, 505993 703215278)/703°2 x 108 © 60 (21°6422! -55250| -0011796| 2°6915|/611°60| 504516 698390199|698°4 x 106 82 |21°654 | 55278] -0011789| 2°6874/607°88) 501447 688868686)688°8 x 10® 99°5 |21°662 | -55200/ 0011785) 2°6840/604°71| 499247 681974176 681-9 x 106 Aluminum 0°5 |20°0170) -49914/ 0012457} 2°7306|650°00) 507632 717306505/713-°0 x 10° — (French). 20 = |20°0253) -49930| 0012451) 2°7270|646°60, 505375 710002959|710°0 x 108 40 |20:0340) -49950| 0012445) 2°7232|642°78) 502388 700672761/700°6 x 108 60°5 |20°0428 -49965} 0012438) 2°7194/639°19) 499977 692975831/692°9 x 106 81 200518) -49980) 0012430) 2°7156)635°50 497477 685124447 685°1 x 106 100 (200600 °50000/ °0012425| 2°7120)632°00, 494734 67T6TV0317|GT6°7 x 108 | | | Silver. 0°3 |17°2176| -4614 | °0015564|10°5142/437°93) 273736 803135399|803°1 x 106 20 = |17°2250) -46158) -0015557|10°5022/437°35| 273489 800762642 800-7 x 106 30 =|17°2284) 46168) -0015554/10°4962/436°80; 273201 798729900/798°7 x 106 40 {17-2316 °46176| 0015551/104900/435°80| 272626 794797589|794°8 x 106 60 17-2380 46194) -0015545|10-4778/433°00) 270979 783307979/783°3 x 108 Zine. 0°3 |18°2094, *44517| -0013426| 6°8130/559°84) 405663 1142925404) 1143 x 106 20 =|18°2200 °44534) -0013415| 6°8010|557°84, 404501 1134426237) 1134 x 108 40 18°2308 -44552| 0013405) 6°7890/553°'76| 401844 1117556599| 1117 x 106 50°5 |18°2364 -44560| °0013399| 6°7826/551°22; 400294 1107903320) 1108 x 108 62 |18:2426) 44570} 0013390; 6°7758/543°61} 394291 1073843790) 1074 x 108 St. Gobain 0:3 |23°496'| -74898] 0013566] 2°5452/750°65| 538313 751865836 751°8 x 106 Glass. 24°5 |23-501 | °74902| 0013562) 2°5436/749°67| 537769 749885431/749°8 x 108 | 40 |23°503 | °74910| °0013561} 2°5424/749'12| 537403 788511034'748'5 x 106 160 |23°507 | -74922| 0013558) 2°5411|\748°35| 536909 746742620 746-7 x 106 (80 |23°510 | -74934) -0013557| 2:°5397/747°62| 536497 745196510/745°2 x 108 99°5 |23°514 | °74945 eae Re 2°5384,746°70) 535955 743311215/743°3 x 106 TABLE IV. Variation of Modulus of Elasticity with change of Temperature. Tn this table the modulus of each substance is taken as 100 at 0°. In comput- ing this table the moduli taken were those obtained from the curves passing through the mean positions of the points determined by the experiments. The results contained in this table are expressed graphically in fig. 5. | | | J.& C. Steel Steel | | Bessemer Bell te Steel. No. 3. No. 4. | SEU 5 Steel. Be Metal. 0°; 100:00 | 100-00 100:00 | 100-00 100°00 100°00 | 100:00 20° 99°61 99:51 is 299 -T0e tae 90 BS 99°53 99°26 | 99°03 40° 99°25 | 99:10 | 99°31 99°10 |» «98:99 98°51 98°09 60° 98°87 98°58 | 98:8 | 99°54 | 98:35 97°76 97-21 80° 98°42 98°00 5) 98722 io oS | 97°63 97:03 96°38 100° OTEG ONsS2 = EOI b aa siete pot ohorge 96°27 99°70 Aluminum Aluminum 4 he | St. Gob. (Amer.). | (French).| °YY°™ | me. | Glass. 0° | 100-00 10000 | 10000 | 100-00 | 100-00 20% 98°92 98°86 99°73 99°26 99°88 40° 97°83 - Sey Ip Shep 97°78 99°76 60° | 96°75 96°58 97°53 | 93-96(62°)| 99°53 80° 95°67 Te! Sat May oa sh ieee 99°30 100° 94°59 94°31 ees =e 98°84 A. M. Mayer—Researches in Acoustics. 95 naa EnnEE i] Gaceee Seetsstitiatteasats inesaguess si seersieiiet Hi + S ti peers : 96 P= TA Tk bE O° tH HT ; biliees - iaaespede cH ita es phesahs + : peasy + At r oe] itt ; = H i 2 tf SS Ht SE Bs essa { tH= +H THF # i ri : Fetes] perl parr = Ss i. ; ++ * 31 + res: + ti : t 4 =i = bss bees! a in Soaps baat CSopy staat poe] ey fears tt peat eatsesesi perstaserasuiel aaetese H hb: — #4 es : petesks tes n = Ts A. M. Mayer—Researches in Acoustics. i} 6. ne TAT 7 aa a) Nee pi pase onped ++ t + + ; f tht + # +r $5 pers cs t tt + t esyeg ba 4 fH : pase soa { ve i cs : tre : } = =: ooee Tt +t H + + i HH LQ” 20, 3.9 Go ow qo ‘80 ~ 90 199 97 A. M. Mayer— Researches in Acoustics. lanl ae ee Sd 3 i 9 pat ape HH s 4 00 re Sy ° a 3 2 : i) = b [en] aa i & tH Ho | Ay s earem| “I ae) a $ See Ht) i i fon i a ° 3 Oo 7) : fo el wd | & 1 5) fa - is a H & Heeeefeaeta 4 +h : H ila +H : 4) f re yo tcth se 3 AAT eB : at i) iti! since S iE a | Teittt: 4 ° +H cy °) Lal ih & Wee “ oA ; Lf i aragetevedatt setae i HEE 6 Hatha ts ia HHH (om) EERE E 2 2 te 3 Hit rf = j t 1H i ihe Fp S if + Het si = lta Re tg EA fe a zs . a \ ps | i e ! - a sre z - a 9 =e ey TN a Yo ov Cie Ih Je A. M. Mayer—Researches in Acoustics. 10. i f205 sti : ; a Sst 2 19 = : : = - 20, 30 Jo 5o- Ge ; ; 4 7e $a. i get” | Peer EEte i {yt peqecEUEOs coded subst Thats [ey + - = EET tH ae a TEL Ese rspadta a esaal spans eases Suuuy paauy ny [EEE saya : \o =H fs Shara ee ais o5r] pups feces shecs juke seat egal heed eee i [SSR tl F399 23 RErET 4 ! B Ee ae Ee HERTHA EE : Sees esieeaeaiae ssdzs Pesca segue nats Etaese: par 99 49a” [ooa- A. M. Mayer—fResearches in Acoustics. ‘ clag EAReRS { ho ° x 96 950 ‘94 ; Ft 93 5 ci 4 Boe aee: qiistase ff 2. - Ht He SSEEsee : 5 SHHSREIPREESSEEEEEEEREEEEEE ne Bs ; fe Uspeasescstes psp i t HE seit Hf HH t + 3 Ba i j ci gas 2 EEE 525 iss 1 Breageegeests Hee Fa ephetesseesy areenpae: Tate HH bree batsEret et Hate ber pees oO 10 20 30 40 50 13), Fava Fteilteeee igsaq egal] Eada) (PRN yal reeaedans Geb Vg eqacdedpadeat Be eer 5 beset Es nti £ = t is re : esese! at J x Ro + if H ses $ Saee : + t + Ht tT Eocyevsbessaugsaueclenal Tl or fit aa TTT H ‘ =e i. ; = +t aneg f +1 pe H == HH Srstneieaistte : its E 25 rc v1 + = : t t Ha : ot : F : ; 33 (egg aa a - - : Sissi i + rrr {68 cabewes ss Be : S panels Ht 344 t + O x 5 ; poh vows 10 20 30 qo 50 40 7° Soe 9° 1900 100 A. M. Mayer—fesearches in Acoustics. 14. a PERE SS ess pr 40 29 3o 40 50 Go 70 Bo \go loo i 3 a 5 H+ Bs so EEE s PtH -O 4 4 ap miss . x Bes see 2 = : ErEEeeeeteeeeeee i a4 ER 3 Sent + ao cE Bon 4 : -| 4 Bo = go 3 ot i | t T +++} Bs guaEs ot r 796 tt : Ht Ht 794 geeteeeziae : Hose : etsobas t t t + Bag cuseccacn == b9 t Tat saserer = 4 at Sh t seau Hs ct 4 Tr H at o t+ + Boece z t rebgyeeiez i iaguecugecenes t t 4 ; ; 79 SEETESS sujssseeatesta pssccassese sieneieserges peace a Hat H ibeed sbecebedee Ht = Fesuape! 2 Se ee eee + . i Ht peep Bs pousesessee gos SopeepRese [esesaes sebeee + ee . eepgeus anaes. oF swans ee sespased seessueet sare: HH ees ; ise! Peas Se eed proses as 4 bsasaefeprgescessers beticeae: HH HH plgecstesdseeaiy eeepaetessapesa ot + Ht ; + chen se panepes. + enlopes beer Ha a aitees iftt chun Gedceseguguas phopasausuape Bs easspuaes ; iaga Bt +h : oe easba t ceo Tete SS puane pues siege neues couea: + + + PEE : Hae 3} eeapadaas +] Bpsusssayee! auseea| tH gaahetiten duaageeedasseqreeass it SH a biti gains Soc URESPESEEE : geseas : - oa neees : f - eet = peepee esesnezos cat zi sheeeat ct ; si jens = --- trate 2 - aaa it t + . - . HH lea ppoeececwncse ppucseee + basi cpecscecbubetestce Bs + aes + s eee penesg He 782 : ; ss Fe HH Susuye : { Gascume 4 tt pases: cneteecae F ; r 1 HH =H 3 aes = 3 4 7 3 HH : eg t baat oA Hea I t + a roo 1G) "20 39 40 50 Go 79 89 99 LOL 6 7 1 1 AUTRE TE TEE pe HHH oF Le HPSUHE TE PES TTLs LAR eae RAE ETRE 9 E it oo REE iF a a Har RATES EE He f a HEE E oe TTT aT Hy ti HEHE HE i Be 5 Hak HE RH HEE ce e a HEH HHH HH Hl | ae ao ae ee il ae . HH PSUS EH Hae deta aT He a HEHE a HITE Het ea ane a ee lh ; H at i! f ae tt a * i ! HIER HEE ee HEA it We HHT i 2 i Ht f HE HEMT A, M. Mayer—fesearches in Acoustics. | eee L = _ a . _ 5 a ane ae 7 ee a ae a ' HEE a HH a ce Leebeel Vo 102 A. M. Mayer— Researches in Acoustics. fesults obtained by other Hxperimenters on the change of the Modulus of Elasticity with change of Temperature. I have found five researches on this subject. I here give abstracts of results from these papers. Wertheim, 1844. Ann. de Chim. et de Phys. IRON. Modulus 5:2 per cent greater at 100° than at 18°. Modulus 19:1 per cent less at 200° than at 100°. IRon WIRE. Modulus 4°9 per cent greater at + 10° than at —11°°6. Modulus 7:42 per cent greater at 10u° than at 18°. WIRE oF ENGLISH CAST-STEEL. Modulus 23°23 per cent greater at 100° than at 18°. Modulus 9°46 per cent less at 200° than at 100°. Modulus at 200° is 11°57 per cent higher than modulus at 18°. STEEL WIRE TEMPERED TO BLUE. Modulus 1°97 per cent higher at + 1¢° than at —10°. Modulus 5'1_ per cent higher at 100° than at 18°. CAST-STEEL. Modulus 2°8 per cent less at 100° than at 18°. Modulus 5°73 per cent less at 200° than at 100°. SILVER. Modulus 5 per cent less at + 10° than at —13°°8. Modulus 1°87 per cent greater at 100° than at 18°. Modulus 12°87 per cent less at 200° than at 100°. CorprPpER. Modulus 6°53 per cent less at + 10° than at —15°. Modulus 6°58 per cent less at 100° than at 18°. Modulus 20° per cent less at 200° than at 100°. WirE oF BERLIN BRass (Cu.=67'55 . Zn. =32°35). Modulus 7°95 per cent less at + 11° than at —10°. Kupffer, 1856. Mem. de l’Acad. de St. Petersb. Modulus of iron wire 5:5 per cent less at 100° than at 0°. Modulus of copper wire 82 per cent less at 100° than at 0°. Modulus of brass wire 3°9 per cent less at 100° than at 0°. A. M. Mayer—fesearches in Acoustics. 103 Kohlrausch and Loomis, 1870. Pogg. Ann. Modulus of iron wire 5 per cent less at 100° than at 0°. Modulus of copper 6 per cent less at 100° than at 0°. Brass 6°2 per cent less at 100° than at 0°. H. Tomlinson, 1887. Phil. Mag. Says, “my own experiments show that both the torsional and longitudinal elasticities are decreased about 24 per cent when the temperature of steel is raised from 0° to 100°.” M. C. Noyes, 1895. The Physical Review. Modulus of a piano wire of 54,™™ diam. 5 per cent less at 100° than at 0°. The results of Wertheim’s experiments giving an increase to the modulus, as the temperature rises, of iron, iron wire, wire of English cast-steel, steel wire drawn to blue and silver, have not been confirmed in any instance by subsequent experiments; only for cast-steel vod and copper did he obtain a diminution of modulus for a rise of temperature from 18° to 100°. Yet, he found that a wire of English cast-steel had a modulus 23 per cent higher at 100° than at 18°. On the Acoustical Properties of Aluminum. The low density (2°7) of aluminum combined with a modulus of elasticity of only 712x10° renders this metal easy to set in vibration; a transverse blow given to a bar of this metal causes it to vibrate with an amplitude of vibration greater than that which the same energy of blow would have given to a similar bar of steel or of brass. This fact has given rise to the popular opinion that aluminum has sonorous properties greatly exceeding those of any other metal. This opinion is erroneous. If a bar of aluminum and a bar of brass having the same length and breadth and giving the same note, are struck transversely so that the bars have the same amplitude of vibration, the bars give equal intensity of sounds; but the bar of aluminum from its low density and because of its internal friction will vibrate less than one-third as long as the bar of brass. Thus, a bar of aluminum and a bar of brass of the same length and width and of such thickness that they gave the same note, SOL, of 768 v. d., were vibrated so that the sounds at the moment of the blows were, as near as could be judged, of the same intensity. The duration of the sound of the brass bar lasted 100 seconds; the sound of the aluminum bar lasted 30 seconds. The readiness with which a bar of aluminum vibrates when acted on by aerial vibrations of the same frequency as those 104 A. M. Mayer— Researches in Acoustics. given by the bar, gives one the means of making many charm- ing experiments in which ‘sympathetic vibrations” come into play. I here describe an experiment which I devised to show the interference of sound in a manner similar to analogous experi- ments in the case of light. The resonant box on which Koenig mounts his UT, (1024 v. d.) fork is open at both ends and has a length of nearly a half wave of the sound of the fork. If this resonant box is held with its axis vertical, above an aluminum bar in tune with the vibrating fork, the bar does not enter into sympathetic vibration with the fork, because the sonorous pulses, on reaching the aluminum bar from the two openings of the resonant box, differ in phase by one-half wave- length. But if the axis of the box is held parallel to the axis of the bar, then the sonorous waves reaching the bar have travelled over equal lengths from the openings at the ends of the box, and these waves conspire in their action and the aluminum bar enters into sympathetic vibration. | As this experiment is an interesting one, I here give details as to the manner of making it. The bar of aluminum has a large surface, having a length of 17™* and a width of 5ems) =6The two nodal lines, which are distant from the ends of the bar equal to 2ths of its length, are drawn on the bar. The bar is supported under these nodal lines on threads stretched on a frame. This frame is of such a height that the under surface of the aluminum bar is 84%, or one-quarter wave length, above the surface of the table, so that the vibra- tions of the bar and those of the waves reflected from the table will act together. The upper surface of the bar is covered with a piece of thick cardboard, in which is cut a rectangular aperture, having for length the distance between the nodal lines and a width equal to that of the bar. As this piece of cardboard rest on supports which lift it a slight distance above the sur- face of the bar, the latter, when it vibrates, does not send to the ear the vibrations of the surfaces of the bar included between its nodal lines and its ends, and which vibrations are opposed in phase to those given by the central area of the bar. Thus the sound emitted by the bar is much increased and the experiment rendered more delicate and im- proved in every way. I have found that the experiment succeeds best when the center of the resonant box is held ADOUty Sos ee Orne T above the surface of the aluminum bar. This experiment works best in the open air, away from the action of sound-waves reflected from the walls and ceiling of a room. A AL. Mayer— Researches in Acoustics. 105 The fact that aluminum gives, from a comparatively slight blow, a great initial vibration and that its vibrations last for a short time, render this metal peculiarly well suited for the construction of those musical instruments formed of bars which are sounded by percussion and the duration of whose sounds is not desirable. I had hopes that the aluminum would prove to be a good substance out of which to make plates on which to form the acoustic figures of Chladni. Experiments have shown that aluminum is not suited to this purpose. I had plates of aluminum carefully cast, with 2°™* of thickness. These plates were turned down on the face-plate of a lathe to thick- nesses of 2™™ and 3°8™™. Three of these plates were quite hom- ogeneous in elasticity, for the Chladni figures when obtained on them were symmetrical. Yet the Chladni figures were difi- cult to produce, because it is difficult to obtain a pure tone from an aluminum plate. The sound is generally more or less composite ; therefore, the plate in its vibration tends to form two or more figures at the same time, and the consequence is that either no figure is formed or one is given that is not sharply defined. One square plate of 30°8°"* on the side and ‘38 thick, gave quite clearly the three following tones: UT, (1), SOL, (2), and SOL, (8). Corresponding respectively to the Chladni figures of (1) two lines drawn from opposite points of the center of sides of plate; (2) figure formed of the two diagonals drawn from the corners of plate; (3) figure sim- ilar to (1) but with corners of plate cut off by curved lines. Fig. 4 corresponded so nearly to the sound of SOL, that a vibrating SOL, fork when held near the plate set the latter into vigorous vibration. Another difficulty met with in using plates of aluminum for Chladni’s figures is that sand, even when entirely free of salt and of the globular grains of wind-blown sand, does not move freely over a vibrating surface of aluminum, whether this surface has been polished or has been slightly tarnished and roughened by the action of alkali. There is one serious objection to the use of aluminum in the construction of musical and acoustical instruments, and that is the great effect that the change of temperature has upon its elasticity. Ifa bar of aluminum and a bar of east-steel be tuned at a certain temperature to exact unison, a change from that temperature will affect the frequency of vibration of the aluminum bar 2¢ times as much as the same change of tem- perature will affect the bar of cast-steel. 106 G. W. Littlehales—Isolated Shoals in the Open Sea. Art. XII.—On the Improbability of Finding Isolated Shoals in the Open Sea by Sailing Over the Geographical Positions in which they are Charted ; by G. W. LitTLEHALEs, U. S. Hydrographic Office. Many of the isolated shoals that are represented on nautical charts of the oceans have been located from the reports of mariners who have discovered them incidentally in making voyages of commerce. Previous to the year 1860, when there was no exact knowledge of the depths of the oceans, the vague reports of navigators, often doubtless based upon the observation of floating objects and of misleading appearances of the surface of the sea, caused the charting of many dangers for the existence of which there is no substantial foundation. But, as our knowledge of bathymetry increased, the existence of many of them was disproved, and they were removed from the charts. As a result of these experiences, there arose a traditional distrust among mariners and hydrographers of the existence of many of these dangers that still appear on the charts with well founded evidence, and there is perhaps a disposition on the part of many to claim that they should be removed upon scant evidence of their non-existence. It is not uncommon for a mariner to report that, being in the vicinity of a eharted rock or shoal, he laid his course so as to pass over the geographical position assigned to it with one hundred fathoms of line out or with lookouts posted aloft, but was unable to detect any evidence of its existence, and that he does not believe, there- fore, that the rock or shoal exists. It seems necessary, there- fore, to inquire into the degree of confidence that can be placed in such a piece of evidence of the non-existence of a danger, and to establish what probability there would be of finding it under these conditions. Suppose that A discovers, in the open ocean, a shoal 7 miles in radius, and determines the geographical position of its center subject to extreme errors of m miles in longitude and m miles in latitude; and that B, who is able to establish his geographical position within the same limits of extreme error as A, attempts to find the shoal again by proceeding to the geographical position assigned to it by A. What is the proba- bility that he will find it ? If A, after making the discovery, had revisited the shoal a great number of times and had deduced the latitude and longi- tude of the same spot, under the same circumstances, at each visit, the latitudes would all differ from the true latitude and, G. W. Littlehales—TIsolated Shoals in the Open Sea. 107 likewise, the longitude from the true longitude. If we call the differences between the true latitude and each deduced latitude errors of latitude and lay them off, according to their signs, to the right and left of an assumed origin, and then, correspond- ing to each error as an abscissa, erect an ordinate of a length proportional to the probability of that error, these ordinates and abscissas will be the coordinates of the probability curve. And, likewise, if the errors in longitude were found and plotted in conjunction with their probabilities, a similar curve would be developed. ' In this investigation the probability curve, ordinarily repre- sented by Laplace’s formula, y=ce—%* will be replaced by two equally inclined straight lines AB and AB’ as shown in figure 1. B’ oO Cc BD a This substitution, which has been employed by Helie in his ‘Traite de Balistique Expérimentale and referred to by Wright in his work on the Adjustment of Observations, causes an appreciable but extremely small error which has no practical significance when we consider that, from the nature of the calculations about to be made, absolute precision is not to be sought. The probability of having an error between Oc = « (figure 1) and «+ 4x, to the right of the axis OS, is equal to s.dx. As, in this case, OB and OB’ measure the extreme errors, all pos- sible errors are comprised between zero and OB, and zero and OB’; and the sum of all the elements which are singly rep- resented by s. da, or the area of the triangle ABB’, should be equal to unity, which is the measure of certainty. The equa- tion to the straight line AB will be, calling m the extreme error OB and ¢ the intercept on the axis of S, Seas bon ; 1 . but, since the area ABB’ =bxm=1orbd= - this equation becomes 108 G. W. Littlehales—TIsolated Shoals in the Open Sea. x sm + —=1 m Mm —2X or $s = m And since x can only vary between zero and m, the proba- bility of having an error between w and «+ 4z Mt — & eh i a (1) The causes which produce the grouping of a number of deduced geographical positions around the true one are of two kinds ; one tending to place the deduced latitude to the north or south of the true latitude, and the other tending to place the deduced longitude to the east or west of the true longi- tude. So that a particular deduced geographical position P will be the result of having an error OA in latitude and an error OB in longitude. The probability that the geographical position deduced by A, upon his discovery of the shoal, occupies a certain position with reference to the true geographical position of the shoal is, therefore, easily deduced. Through the true geographical position of the shoal let two rectangular axes, OX and OY, be passed as shown in figure 2. Upon the former conceive errors in longitude to be measured, and upon the latter, errors in latitude. ‘The position P, of which the codrdinates are # and y, results from the concurrence of two conditions, the error of a miles in longitude and the error of y miles in latitude. The probability p, of the error x is, as shown by equation (1), Mi — X = . ax P ne and, in the same manner, the probability , of the error y will be eae, P, = a dy te) In these formulas, m and m represent respectively the extreme errors in longitude and latitude in miles. G. W. Littlehales—Isolated Shoals in the Open Sea. 109 The probability p of having, at the same time, the error « and the error y, or of deducing the geographical position P as the position of the shoal, will be the product p,, p,, or pp (m—x) (n—y) Be (3) mn? an equation in which x can vary from zero to m, and y from zero ton. It is, therefore, applicable to the first right angle of the axes OX and OY, but, in order to make it applicable to other quadrants, it is only necessary to change the signs of « and y. Bie nation 3 then expresses the probability that A’s determi- nation of the geographical position of the shoal is in error by x miles in longitude and y miles in latitude. If the center of the shoal were really located in the geo- graphical position assigned to it by A, and B should succeed in coming within 7 miles of it, he would find the shoal since its radius is 7 miles. We have, therefore, as the second step in the solution of the problem, to determine what is the probability that B will come within a circular area, 7 miles in radius, having its center any- where within the rectangle described about the true position of the shoal with sides equal to the extreme errors to which the determinations of latitude and longitude by A and B are subject. To find the probability, P, of coming within any portion of the rectangle of extreme errors inclosed by a curve whose equa- tion is y= f(a), it is sufficient to integrate the expression (8) between limits depending only upon y= f(x), and we shall have, in the first right angle, P= —— ye doc Jim) (n—y) dy (4) For a circular area of radius 7, we shall have for the first quadrant, V7? — 4p? PS : = saina fl (m—2) ae . (n—y) dy and for the whole circle, 4 ‘a V7? — 52 P= af m—2% anf n—y) dr mn ed , ( ) A ( y) ay or? 1 27 OW 9" or Ps ees — _ (5) Inn 2 3m 38n. Amn The probability that B will find the shoal depends upon the concurrence of two independent conditions whose separate 110 T. L. Walker—WNotes on Sperrylite. probabilities are represented by equations (8) and (5) respec- tively, and is, therefore, equal to P. p, or 27 12 Qr am r (m—z2x) et 23 (6) mn mn? 2 om oan 4mn Integrating the two expressions which make up equation (3) between the limits « and x+ 4x and y and y+4y, respectively, the above expression becomes : or “a Dp) Qi ie “(1 et a ( ran) ES oe Me rise. le 2 3m 37m 6 4mn m 2m n 2n which, for 7 = 1 mile, c = 2 miles, y = 2 miles, m = 10 miles, nm =10 miles, and dn and Jy each equal to 1 mile, becomes eqyq- That is, under the conditions stated, B would stand one chance in 6173 of finding the shoal. Art. XIII.—Wotes on Sperrylite ; by T. L. WALKER. THE interesting mineral sperrylite was first described by Professors Penfield and Wells in 1889.* Being a diarsenide of platinum and crystallizing in pyritohedral forms, it serves to link the platinum group of metals with the iron group, since in the latter group diarsenides and disulphides commonly erys- tallize in pyritohedral forms. In June, 1893, the writer obtained permission from Mr. H. P. McIntosh, secretary of the Canadian Copper Company, to visit the Vermillion mine in Algoma, Ontario, and to obtain specimens of sperrylite and associated minerals. The notes here submitted are derived from an examination of the material collected. Professor Penfield described four crystal forms on sperrylite, (111), (O01), (110) and 7(210), but he also adds: “Some erys- tals appear to be somewhat rounded by other faces, but none of the latter could be determined.” + Some of the larger and more promising crystals were exam- ined under the microscope. The common forms observed were (111) and (001), generally in combinations in which the former usually predominates. The form (110) was not recognized on any of the crystals examined. (210) could be seen on most of the larger crystals, though the faces were generally small. Another form was observed which replaces the angles formed by (111), (001), and (210). The edges formed by the intersection of these new faces with z(210) are parallel, and hence the new form belongs to the same zone as 7(210). The * This Journal, xxxvii, 67, 71, 1889; Zeitschr. f. Kryst., xv, 285 and 290. | Zeitschr. f. Kryst., xv, 291: T. L. Walker—WNotes on Sperrylite. EE edges formed by the intersection of the cubic faces with the octahedral faces and with the new faces, form on the cubic faces plane angles which gave the following measurements: 155° 35/7 156° 15’ 155° 55’ 156- 20! 156° 10’ 156° 50’ Average measurement, 156° 9’. Calculated for 7(10°5'2) 156° 49’, The calculated symbol for these new faces agrees very closely with the di-dodecahedron z(10°5-2), which has not been previ- ously determined for sperrylite; 2(11°5-2) occurs on pyrite and is very close to the faces here described. There are still other faces present on some of the crystals, but they are too small to be determined on the present material. Small erystals are frequently observed, half-imbedded in the cubic faces of the larger ones. These guest crystals appear to possess the same orientation as the host, and generally show combinations of (001) and (111). The hemi- hedral faces may occur on the smaller individuals, but could not be determined. In intergrowths of pyrite, the pyritohe- dral faces are so disposed that the whole is regarded as inter- penetration twins with the twinning axis normal to (110). This is well seen in the so-called “iron cross.”” That the crystals of sperrylite follow the same law of twinning is highly probable, but we cannot regard it as proved from the present observa- tions. The detection of pyritohedra or di-dodecahedra on the guest crystals and the determination of their position with regard to that of the corresponding faces on the larger crystal, would settle the question beyond dispute. _ Sperrylite occurs in loose decomposition products of chaleo- pyrite, pyrrhotite and other iron-nickel sulphides. The con- centrates obtained by “panning” consisted principally of grains of chalcopyrite and pyrrhotite along with crystals of magnetite, sperrylite, and cassiterite. The chalcopyrite grains showed on closer examination that they frequently contain erystals of sperrylite. The fragments of pyrrhotite were care- fully examined, but in no case could erystals of sperrylite be observed on them. ‘The chalcopyrite is therefore the original host of the sperrylite. In this connection it may be men- tioned that in all the copper-nickel mines of the Sudbury dis- trict, traces of the metals of the platinum group are found, and also that nickel matte from mines low in copper contains very little nickel, while mines richer in copper afford a matte proportionally richer in platinum. In short, the platinum con- tents of nickel matte in the Sudbury district is directly pro- portional to the copper contents, viz: proportional to the amount of copper pyrites in the original ore. This fact, taken 112 T. L. Walker—Notes on Sperrylite. in connection with the detection of sperrylite in fragments of chalcopyrite, while a careful search did not reveal it in the pyrrhotite fragments, renders it very probable that the plati- num found in the Sudbury district, in the copper-nickel mines, in general occurs as sperrylite and that this mineral occurs generally if not exclusively in chalcopyrite. It should be mentioned here, however, that according to some analyses,* polydymite, which occurs very sparingly in some of the Sud- bury mines, contains from 0-006 per cent to 0:024 per cent platinum. Sperrylite dissolves slowly in strong hot hydro- chloric acid, more readily in aqua regia, while in nitric, sul- phuric and hydrofluorie acids it is practically insoluble. An analysis of Manhée matte from Murray mines gave the following results : Nickel, (with tracescobalt).- 2202 see 43°82% Copper... aes Soe oe 25°92 OW se ate ye eae a re ee 2°94 Sul plur oe oe oe 22°50 Gold. * Sec. Pee ce 000075 pilver *- io oo ee oe Platinum 2.23"! 2 Pee ee ee Indi S ee BOSSE eee 000056 Osmium 2G les Bs eee "000057 uhogiin 37 Ske ee eee small quantity Palladium sd!e bert onto eee small quantity Potal ws a2 ssese les ete In this as in similar analyses made by others the presence of iridium and osmium is noted. In Professor Wells’s analysest of sperrylite these metals are not detected, though he was at special pains in searching for iridium. Baron H. B. von Foullont concludes, from the presence of these metals in the Sudbury ores, that there is another mineral present which con- tains the iridium. Thisis possible, but more probably in some cases part of the platinum in sperrylite is replaced by the elements iridium and osmium. This would not appear to be the case, however, in the Vermillion mine sperrylite as shown by Professor Wells’s analyses. The above observations were carried on in the laboratories of the School of Mines at Kingston, Ontario, and at the Mineral Institute at Leipzig. For advice and assistance I am especially indebted to Dr. W. L. Goodwin and Professor Nicol of Kingston and also to Herrn Geheimrath Professor Zirkel of Leipzig. Leipzig, Saxony, November, 1895. * Bull. U. S. Geol. Survey, No. 64, p. 21. + This Journal, 1889. + Jahrb. d k. k. geol. Reichsanstalt, 1892, p. 301. Kiimmel—Glaciation of Mountains in Pennsylvania. 1138 Art. XIV.—Wote on the Glaciation of Pocono Knob and Mounts Ararat and Sugar Loaf, Pennsylvania; by Henry B. KtmMmet, Pu.D. [Published by permission of the State Geologist of New Jersey. ] DuRInG the past field-season an opportunity was given me, in connection with my work on the glacial deposits of Northern New Jersey, to visit Pocono Knob, Monroe County, and Mounts Ararat and Sugar Loaf, Wayne County, Pennsylvania. The Pennsylvania State geologists* have held that during the gla- cial period these peaks were nunataks. Pocono Knob is an outlier of Pocono Plateau, situated about nine miles northwest of Stroudsburg. The terminal moraine is well developed on its north and south flanks, about two-thirds of the way up the slope. Since this knob was examined by Lewis and Wright, a wagon road has been constructed up the northwest side and along the top of the hill for more than half amile. This road affords almost continuous exposures to a depth of from one to three feet, along the top of the knob. Here was found a considerable variety of material, chiefly shales and sandstones of different color, texture, and hthological con- stitution, with some coarse conglomerate. Not a few of these had been worn to subangular form with more or less well marked planation surfaces. Fragments bearing strie, of whose origin there can be no doubt, are not abundant, but ten or twelve cobbles were found, which bore unmistakable glacial scratches. Some of these occurred not more than thirty feet below the summit. The evidence is conclusive that the ice covered the western part of the hill to within at least thirty feet of the highest point. That it also covered the crest is very probable. When one leaves the road and examines the weathered material on the surface, hidden, as it is, by vegetation, it is next to impossible to convince oneself, that there is any glacial debris on the knob, and during the earlier part of my investi- gation, before I had examined the exposures along the road, I held the same opinion as those who had earlier studied the ground. On the highest point of the hill, the surface is strewn with large bowlders of disintegration, and the ledges show no signs of glacial action. No evidence of glaciation could be found on that part of the hill which lies east of the summit, where there are no exposures, but in view of what was found along the road, where the opportunity for observation is good, there can * Lewis, Second Geological Survey of Pennsylvania, Terminal Moraine, Z, pp. 75, 271. White, idem., Susquehanna and Wayne, G5, pp. 25, 159. Am. Jour. Sct.—FourtH Series, Vou. I, No. 2.—FEBRuaRY, 1896. 8 114. Atimmel— Glaciation of Mountains in Pennsylvania. hardly be any doubt but that the whole knob was covered by the ice. The amount of material left by it was, however, small. North of Dry Gap, as the col connecting the knob with the plateau is called, the characteristic topography of the terminal moraine is strongly developed. Fresh cuts along the recently constructed line of the Wilkesbarre and Eastern railroad give fine exposures of the glacial deposits. Since the data now at hand prove that the ice covered the knob, the connection between these two parts of the moraine is probably through Dry Gap,* but the typical morainic topography is not developed at that point. Sugar Loaf and Ararat, in the northwestern part of the State, are 2475 and 2650 feet high,t respectively, and rise about 500 and 700 feet above the general level of the surrounding plateau. Both are thickly covered with underbrush and_ timber, and exposures are almost entirely wanting. On the north face of Sugar Loaf, glacial material was found along an . old wood road, two-thirds of the way up (as far as the road extended). Above that height nothing definite could be made out for lack of exposures. The surface, however, was not © radically unlike that of the lower part of the hill. The few rock ledges which occur do not bear strize, nor have they roche moutonnée surfaces. However, ledges of sandstone so exposed to the weather could hardly be expected long to retain glacial markings. On Ararat the facts are much the same. From want of opportunity for critical examination I was unable to prove that the ice covered the highest points, nor was I able to satisfy myself that it did not. Although in the present state of things, it is impossible to obtain conclusive data, there are certain considerations which indicate that the excepted view is probably not the correct one. At the Delaware Water Gap, the ice filled the gorge and overrode, with very little deflection, the crest of Kittatinny mountain, which rises 1800 feet above the river and 600 to 700 feet above the general level of Kittatinny valley. If the ice was of sufficient thickness to accomplish this at points less than seven miles from its margin, it seems improbable that the ice-sheet was not thick enough to override hills such as Ararat and Sugar Loaf, which rise less than 700 feet above the sur- rounding plateau, and which are seventy miles north of the margin of the ice. There can be no doubt but that, were the erests of these hills cleared and excavations made, traces of glacial drift would be found here as at Pocono Knob. Geological Survey of New Jersey, Trenton, N. J. * This was suggested by Lewis in a supplementary note to his report (1. c., p. 271), although in the report he had mapped the moraine as extending around the eastern end of the knob. twine; 1. ¢:, p. 17. C. Barus—Counter-twisted Curl Aneroid. 115 Art. XV.—The Counter-twisted Curl Aneroid ,* | by CARL Barus, Hazard Professor of Physics at Brown University. INTRODUCTORY. 1.—IT seems plausible to argue that much goes on in the atmosphere immediately related to pressure which the ordinary pneumatic barometer merely integrates, and of which it can give no detailed account. I refer both to the changes referable to the gusty character of the wind + and to pressure variations of a more subtle nature,t the origin of which may be considered in relation to the earth’s magnetic and electrical field. 2.—The problems, therefore, are beyond the immediate scope of instruments of large mass like the ordinary mercury or water barometers. The conditions to be fulfilled are (1) great sensitiveness and (2) instantaneous indications ; (3) regis- try subject to corrections of a purely scientific kind. It is when these three conditions are simultaneously demanded that the problem becomes formidably difficult. No matter what form of mechanism is selected, one is brought face to face at once with viscosity and with the thermal variations of both vis- cosity and of elasticity. I desire in this paper to find out how far one can go with suitable modifications of the Bourdon tube. In certain meas- urements§ of high pressure made with such a tube coiled heli- cally I received much encouragement, inasmuch as the instru- ment could be read off closely enough, without the aid of subsidiary mechanism. The difficulties which [ then encoun- tered were purely technical. In flattening and coiling the necessarily heavy tube, I had to remove the temper at a sacrifice of strength and resilience, and the gauge actually burst at 1000 atmospheres. I abandoned it simply because of the difficulty in fashioning this cumbersome apparatus in the laboratory. In relation to low pressures, however, all of these difficulties fall away at once, and it becomes merely a question of patience to find the limit of constancy and precision to which a gauge of this kind can be pushed. The conditions of sensitiveness were discussed not long ago * The present research was encouraged by a fund kindly placed at my disposal oy the Secretary of the Smithsonian Institution. + Cf. 8. P. Langley: ‘The Internal Work of the Wind,” Smithsonian contribu- tions, No. 884, 1893, Washington, D. C. ¢ Considerations of this kind originated, I believe, with the late Prof. Balfour Stewart. Recently the subject has been attacked more seriously, notably by Prof. F. H. Bigelow. § Barus: Bulletin U. S. Geological Survey, No. 96, p. 29; 1892. Cf. Proc. American Acad,, xxv, p. 106, 1890. 116 C. Barus—Counter-twisted Curl Aneroid. by Prof. Worthington,* and in a more elaborate article by Prof. Greenhill.t There is not much succor to be gained from theory. The most lucid expression of the case is due to Lord Rayleigh.t His explanation meets the conditions of the present paper very fully, since the gauges are to be constructed with sharp edges and a spindle-shaped section, in order that the changes of shape occurring may be pure bending. § 8. Stretch- ing would necessarily introduce resistances large as compared with the feeble forces to be measured. Since in the flexure of an inextensible surface, the product of the principle radii of curvature at a point of the surface remains constant (Gauss), any uniform sectional flattening of the walls of the tube due to increased external pressure must be compensated by an increased curvature of the axis ; and vice versa. If the edges are sharp, there seems to be no advantage in increasing the height of the section, for the elastic resistances increase at the same time as the external pressure. There is, on the other hand, a disadvantage in wide tubes, for the length of the helix is thereby necessarily increased and the curl made more cumbersome, while the shape of the section cannot long remain regularly and uniformly arched. SIMPLE CURLS. 3. Apparatus.—There is, therefore, a demand for extremely thin metallic tubes, flattened nearly to the point of actual con- tiguity of the walls and left with almost sharp edges. ‘Tubes made of metal as thin-walled as 0-01 are manufactured abroad;§ but American makers usually fall short of this remarkable accomplishment. I found, however, that it was quite possible to derive full advantage from the American tubes by dissolving off the outside layers in a suitable acid bath. The tube at my disposal was made of copper, being like the material I formerly used in my ealorimetric work.| This metal is elastically unpromising, apart from the technical advantages of smooth solution in diluted nitrie acid, of easy flattening and bending, and of a small modulus of elasticity. Results, however, which can be obtained with copper tubes, can be much improved by the use of other more suitable metals, and for this reason the data of sensitiveness and con- stancy obtained would be an index of the possibilities of the gauge erring markedly on the side of safety. Two methods suggest themselves for flattening: the tube * Worthington: Nature, xli, p. 296, 1890. + Greenhill: Nature, xli, p. 517, 1890. { Rayleigh: Proc. Roy. Soc., Dec., 1888. § I owe most of my information in relation to thin tubes to Prof. Langley. || Barus: Proc. Am. Acad., xxvi, pp. 316-317, 1891. The liquid was examined in a very thin tubular helix of copper, to insure rapid cooling in the calorimeter. OC. Barus—Counter-twisted Curl Aneroid. 117 may either be compressed between steel rollers, then neatly coiled on a steel mandrel at the lathe with the turns one or two millimeters apart, so that the finished tube has the general form of a curl, and finally dissolved to the required thinness of wall in the nitric acid bath; or the tube may first be dissolved to the required lightness, then flattened and coiled as stated. The second of these methods is the best, though it requires a long tubular bath with a cistern at the top to avoid spilling the acid during the more or less violent effervescence in solution. The ends of the tube are stopped with corks, covered over with a cement of resin and beeswax, and all imperfect parts of the tube are similarly protected. The progress of solution is tested by taking weights from time to time, and when the tube is so light as to begin to float, it is suitably held down. A spring balance arrangement attached to the tube and out of the reach of the acid fumes, obviates the need of repeatedly removing the tube from the bath. Holes are sometimes eaten through the walls at defective places, probably from galvanic action. Should this occur, the tube must be cleaned at once, and cut apart at the defective place. The first of the methods sketched above is more convenient, as the curl requires a much smaller bath. Unfortunately solution is most active at the edges; the tube be- comes very fragile and is in danger of fis- suring on continued use. Tubes rolled down and coiled under a tense pull are apt to be quite closed. The way in which this can be remedied by coun- ter-twisting will be described below (§ 8, et seq.), where closed tubes may even be an advantage. But in this place it is best to stretch two fine wires of brass or copper (say 0-02*" in diameter) from end to end of the tube, so that complete closure is avoided. An interesting structural result was ob- served in this work. Open helices, with the turns about 0°3™ apart, after solution became closed helices with the turns all but touching at the edges, showing that the layers of the metal were subjected to unequal strains with a maximum of deformation at the out- side.* As a rule tubes must first be cleansed internally (to remove pulverulent or fatty matter) with ammonia, hydrochloric acid and water, and then thoroughly dried by a current of warm air. The finished curl is closed at one end, with the other Simple curled Bourdon tube. * Compare this with the gradual loss of explosive tendency of a Prince Rupert drop, when the external layers are successively etched off (Barus and Strouhal, 118 C. Barus—Counter-twisted Curl Aneroid. soldered to a suitable terminal tube to connect it with the air pump. The accompanying figure 1 shows the curl ready for experiment; @ is the terminal tube leading to the air pump, 6b the curl aneroid, ¢ a mirror or other form of index, to regis- ter the rotation of the lower end of the curl relatively to the upper, when the air within is being exhausted. For conven- lence in drawing only a few turns of tube are shown in the figure. 4. fvesults.—In the first experiments the object sought was some guidance as to the effect of varying the diameter of the curls. Copper tube, with walls 0°025™ thick, was flattened till the spindle- shaped section was about 0°5°" high and 0:08" wide, then wound so as to make an open helical spring. The diameter of the curl was successively reduced by winding on smaller mandrels, and the corresponding sensitiveness was found by comparing the curl with a mercury gauge when both were joined to the receiver of a Sprengel pump. The index at the bottom end of the centered curl moved over a graduated circle about a foot in diameter and divided in quarter degrees. Table 1 contains an example of the results. The curl lengths are approximate, due to unavoidable irregularities of winding. In addition to the direct readings I will give the pressure difference in centimeters of mercury per deeree of deflection (i.e. per degree of deviation between the terminal tangents of the helix), and the same datum taken per turn of the curl and per centimeter of the length measured along the turns of the helix. In general pressure denotes the difference of pressure between the outside and the inside of the tube and is always given in centimeters of mercury. It appears therefore, that within the range of accuracy of these experiments, the untwisting of the curl in degrees of are is very nearly proportional to the pressure difference in em. of mercury, remembering that the pressure on the outside of the tube acts on the cistern of the barometer. In the second place, the sensitiveness of the coil for a given length of tube is some- ’ ee 5, ee what greater for small radii than for large radii; thus for the diameter 3°1°", a degree of are corresponds to 4:07" of mer- cury, whereas for the diameter 2:0, a degree corresponds to 35°" of mercury. The changes, however, are small and irregu- lar and possibly largely influenced by temperature, which was not taken. At all events, the data for cm. of mercury per degree of deflection, per turns of curl, increases so rapidly, that in view of the objectionably great length of curls of small diameter, they offer no advantage. As a rule the diameter 25° will be adhered to in the present paper. this Journal, xxxii, p. 182, 1886); and with another result (this Journal, xxxiv, p. 183, 1887), in which the limit of torsional resilience of soft iron was reached whenever the obliquity of the external fiber of the wire exceeded about 0°003 radians. C. Barus—Counter-twisted Curl Aneroid. 119 TaB_E 1.— Behavior of Curls of different diameters, for the same length of tube coiled to a helix. Cm. of Hg Mercury Inflection Cm.ofHg Cm.of Hg per degree Mano- of per degree per dexyree per cm. Description of curl.* meter. curl, ofare. perturn. of diameter. Diam. of eurl Q-]cm cm Care Turns of curl 21 76°7 18°7 4°07 85 les Length of curl 18™ 67:9 16°6 4°07 61°5 15°0 4°08 84°3 13°2 41°5 10°1 a2 7°8 Digm., 2°3°° 76°3 7 AE | 3°55 87 1°5 Turns, 24 70°5 19°5 3°55 Length of curl 19°5™ = 65°3 18°0 Sail 59°8 16°4 3°63 52°5 14°4 47°4 13°0 41°4 13°38 29°35 8:0 iam, 20°" 60°3 17-4 3°45 Lov 17 Purns, 35 52:0 14°9 3°49 Length of curl 22°" 31-0 8 9 0:0 0°0 ~ Practically, in estimating the effect produced on thinning the walls by etching, I may therefore either express the num- ber of em. of mercury which correspond to a degree of are per turn of the tube, in which case the same diameter should occur throughout or be specified; or I may make similar reference to the diameter; or finally (and probably best), I may state the centimeters of mercury per degree of deflection, of the coil per unit (em.) of length of the tube used in winding the curl. Lumping the results of Table 1, where the length of tube used was about 210° throughout, it appears that in these initial experiments a pressure of abont 780° of mercury would produce 1° of deflection between the ends (tangents) of a tube 1™ long, and having the size and thickness specitied. The endeavor was now made to dissolve down this curl, but the action was carried too far and it was eventually lost in the process. After about 4 of the weight had been etched off however, 61°" of mercury corresponded to a degree of twist per turn for the diameter of curl 2°9°7, or 550™ of mereury per linear cm. per degree, showing a decided gain over the general datat of Table 1. 5. Another curl, No. V, was now wound, etched off and tested with the following results, after having been exhausted for some time. The original weight was 39% and after etching * The length of the eurl and the length of the tube used in winding the curl must be carefully distinguished. + A slight digression was made in coiling the next tube like a watchspring ; but the experiment showed no special advantages. 120 C. Barus—Counter-twisted Curl Aneroid. 208, so that the walls must have been reduced in thickness from 0:025™ to about one-half this value, Diam. of curl, 2°8°™ Pressure, 71°9°", Hg Deflection, 111° Turns of curl, 22 59-5om 97° Length of curl 14°7™ 0:0 20° (later) 0:0 0° (next day} 0:0 —2° Pressure per degree of are per turn, 14°6™, He. Pressure per degree per linear cm., 130°, Hg. There has therefore been a considerable gain in sensitive- ness,* inasmuch as the deflection of one degree between the ends of a linear centimeter of the coiled tube is now equivalent to 180° of mercury. A new feature has been introduced in the apparent occurrence of marked viscosity, as shown by the gradual displacement of the “zero” reading for pressure dif- ference. The cause of this will be further studied, but it is probably due to the tendency of external pressure to quite close the tube,t so that it takes the air some time to reénter. This curl was now cut in two parts, V, and V,, of 9°5 and 11 turns respectively, in order to test whether the air had quite reached the extreme parts of the tube. An examination showed the following results : Pressure : Deflection: ( Diameter of curl, 2:88°™ 72°00" sete: 49°6° | Turns of curl, 9°5 64°3 43°7° ; Length of curl, 55°8 38°3° V d 33°3 93°4° reg 0:0 te (later) 0:0 0:0° | Pressure per degree per turn, 14°0™. L Pressure per degree per linear cm., 122°, ( Diameter of curl, 2°8°™ 62°72 56°3° Turns, YI 53°9 49°6° | Length, - 46:2 43°8° Vv i 35°6 35°3° -) 0-0 79° (later) 0:0 00° | Pressure per degree per turn, 14:3. [ Pressure per degree per linear em., 127°". These results agree substantially with the data for the uncut curl, the discrepancies being due to the difficulties of estimat- ing the number and diameter of the turns and to the occur- rence of viscosity. It is seen that viscosity has in no way dis- appeared, though it is much more marked in V, than inV,. * Data of the table taken in accord with other similar experiments on the same eurl. + There may also be a displacement of the zero due to friction of the walls of the tube on each other. Such displacement, however, would probably be per- manent. C. Barus—Counter-twisted Curl Aneroid. 194 From this it follows that the viscosity in question must be in large measure apparent but not real. While V, was preserved for other purposes, V, was now further etched off. In so doing some of the turns had to be sacrificed, being eaten through. The results in these cases were more irregular, and apparent viscosity more marked. [ will not therefore give them in full. After the first solution, 9 turns of V, showed 13™ pressure per degree per turn, or 107™ Hg per degree per linear cm. After the second solution V, showed 8°7°™ pressure per degree per turn, or 77°" pressure per degree per linearem. It was imposible to carry the solution further, because the edges were too thin to withstand flexure without fissuring ; but the walls were 0°01™ in thickness, and might easily have been reduced. cf. $3. Summary.—Remembering, therefore, that if viscosity were excluded, the curl would have been more sensitive, I may state that it takes from 50 to 75° of flat copper tubing, the walls being 0:01 thick, to make a curl aneroid such that a degree of are shall correspond to 1™ of the barometer. If, therefore, the reading be made with mirror and telescope, with the scale at a distance of 286° from the mirror, 10 linear em. of thescale cor- respond to 1°” of the barometer. If the tube be wound ona mandrel a little over 28°" in diameter, 7 or 8 turns would suf- fice, and the length of the curl need therefore not exceed 8. Hence, 10 such curls joined in series end to end and suitably supported, would show 100™ at the telescopic scale per cm. of the barometer, and the curl would be less than a meter long. § 12. COUNTER-TWISTED SYSTEMS. 7.—In the results thus far, the object has been merely to exhibit the possibilities of the curl aneroid. If the apparatus is to attain precision, the apparent viscosity must be brought quite under control, and the effect of the temperature of the medium evaluated and reduced to the smallest margin. I have in a measure fulfilled both these conditions by using counter- twisted systems in the way presently to be specified. As a first step in this direction, I will cite some data obtained with curl VIII, in which two fine brass wires (diameter 0:02™) were drawn through the tube before flattening and coiling. The walls were 0°013°™ thick, and all but 5-2 turns were lost during solution. Diam. of curl, 2°8°" Pressure, 73°7°7, Hg. Deflection, 38°3° 63°2 33°4— Turns, a2 Length of curl, 3:°3°™ 485 26°0° 0°0 0°6° 0:0 0:0° Pressure per degree, per turn, 10°4°™. Pressure per degree, per linear cm., 91°, 122 C. Barus—Counter-twisted Curl Aneroid. The small residue of viscosity left here is very probably not even yet a true phenomenon, i. e. it is due to strictures which prevent the free passage of air, and to friction between the con- tiguous walls of the tube and wire; but the improvement over the preceding cases is obvious. 8.—To enter into this question somewhat more fully and from a different point of view, a wider tube was selected (the walls of which were but 0-12°™ thick), flattened and coiled without closing the section. Indeed a blunt edge was left and the section was about 0°06 broad and 1°6™ high. The data for this coil (LV) were as follows: Diam. of curl, 3°2°° Pressure, 75°9°°, He. Deflechionsaieas Turns of curl, 4°5 69°5 10°37 Length of curl, 14:0°™ 57°8 8°8° 41:3 6°6°° 36°8 5°92 0-0 zero. 0°0° Pressure per degree, per turn, 32°5°. Pressure per degree, per linear cm., 330°". In this curl there 1s no evidence of viscosity, but the relation of pressure-difference and deflection is not quite linear, as was the case with sharp-edged coils. The data are mean ratios. The sensitiveness (830, Hg, per degree per linear em.) is of low order, in spite of the thin walls (0:012™) and broad tube. § 2. This curl was now replaced on the metallic mandrel, and the edges hammered quite sharp from end to end. On removing the curl from the mandrel I found that no azr could be sucked through it. The walls, therefore, overlapped each other, 7m- perviously to air. When, however, the curl was somewhat uncoiled in the hands, the air came through quite freely. This suggested a novel method of making the curl aneroid, requiring no inclosed wires, and partaking of other advantages, since the uncoiling can be done with a suitable spring. In a counter- twisted system of this kind: (1.) The sharp-edged coils can be opened by an amount com- patible with the free access of air. Therefore this system 1s adapted for extreme sensitiveness. (2.) The system is differential ; or, in other words, the dzffer- ences of viscosity and of elasticity of curl and spring, and the differences of the thermal variations of these quantities come into play. Thusif a spring and curl could be made having the same effective viscosity* and the same thermal coefficients of viscosity and elasticity, respectively, the system would be perfectly elastic and independent of temperature ; or, * Depending therefore both on the material and on the lengths of the two helices. C. Barus—Counter-twisted Curl Aneroid. 123 (3.) If the adjusted curl aneroid, exhausted and therefore twisted by external pressure, be untwisted to the same degree by the spring, the curl is now nearly without strain, excepting such of higher orders; and the viscosity and thermal relations of viscosity and rigidity of the counter-twisting spring alone come into play. Hence a metal of low rigidity may be chosen for making the curl, while the counter-twisting spring is selected for high viscosity. One would use, for instance, hard steel annealed at 400° or even a sufficiently thick quartz fiber. It is clear that the counter-spring must be weak, so that it may make a number of rotations for each rotation of the curl; for the resistance of the counter-spring is incre- mented at the expense of the deflection of the curl. Hence a long fiber, or spiral spring, or a watch spring is adapted. Finally, to prevent knotting of the spring, a weight is (temporarily) added to the sys- tem, preferably suspended in a basin of - water to deaden vibrations. The curl ane- roid has now taken the form of figure 2, where ce is the curl, communicating at one end, #, with the air pump. ‘The other end is closed and earries the horizontal, radial stylus 7, to which the stiff wire dd of the spring bb, the mirror e and the damping plate g are attached symmetrically to the eurl. The plate g is of lead, thus serving as a weight (as well as a damper) to pull! out the spring 6b. A dish / (shown with g in cross section) containing a liquid, surrounds g. The spring 6 is attached above to a rigid clamp at a, by means of a stiff wire J, the clamp (not shown) being so constructed that the spring can be both raised and twisted. The weight g and the damping arrange- ment are only permissible in calibration work like the present. If the registry of the aneroid is to be instantaneous, the mass must be kept down to the lowest limit pos- sible, and no ballast is to be attached. Hence a delicate watch-spring, or a double Counter-twisted system helical spring is preferably employed, and of curled Bourdon tube, a suitable method of damping vibrations SPs and appurte- must be investigated. An elastic fiber fas- , tened at both ends, with the end of the helix attached near the middle, is also contemplated. 124 C. Barus—Counter-twisted Curl Aneroid. 9.—To obtain preliminary evidence, I took curl IV, just dis- cussed (§ 8), though it had become somewhat leaky from hard usage. The counter-twisting was done by a long steel wire with the following results: Turns uncoiled from { Pressure per degree, per turn, 13°4°, Hg. 4°5 to 3°7. Pressure per degree, per linear cm., 167. Turns uncoiled from { Pressure per degree, per turn, 5-1, Hg. 4°5 to 3:0. Pressure per degree, per linear cm., 79°™. In all these cases the deflections were instantaneous, and there was no apparent viscosity. The curl itself has been improved from one of low sensitiveness (830°) to nearly the same range of high sensitiveness (77°), actually obtained in § 6 for extremely thin copper tubes. This result deserves special study; but it already appears that so long as the coil is sharp-edged in section—so long as the strain is a case of nearly pure bending—the coil increases in seusitiveness as its spindle-shaped section is more highly arched. ‘The smaller the medial radius of curvature of a right section of the tube, the greater proportionately is its variation for the same pressure-difference ; and the greater proportion- ately must the corresponding variations of the longitudinal radius of curvature (coil radius) also be, since the product of two radii is to remain constant. Hence a rotundly arched, spindle-shaped section, maintained in an excessively thin-walled tube, 1s compatible with the greatest amount of rotation at the registering inden. 10.—A copper tube having walls 0:01 thick was flattened and rolled down as usual. The curl (No. VII) contained 9 turns and it was 2°8™ in diameter and about 12°5™ long, each turn in section being 0°8™ high, with blunt edges about 0:05°" wide. So constructed, the sensitiveness (conformably with the data in § 8) was low, for the pressure in em. Hg per degree per turn was 50, and the pressure per degree per linear em. of the curl 440™.—in spite of the thin copper walls stated. This tube was now hammered flat and sharp-edged on the steel mandrel. A part, VII,, free from flaws was then cut from this tube and used in the following tests, Table 2. For want of better material a helical spring of brass spring wire was used in the counter-twisting and the system was weighted to prevent knotting (cf. § 8). These results are. given graphically in the chart figure 3, where the ordinates are pressures in cm. of mercury, corre- sponding to a deflection of 1° for a length of 1° measured on the turns of the curl. The abscissas show the amount of counter-twisting at the lower end of the system in degrees. The sensitiveness therefore increases in marked degree with C. Barus—Counter-twisted Curl Aneroid. 125 the amount of counter-twisting, and apparently reaches a limit at about 150°. From here on, however, the evidences of per- manent set were met with. They are in the direction of the uncoiling, showing the spring to have acted on the curl. At the end of the experiment, when the spring was released, about 90° of permanent set had been imparted to the curl. I believe therefore, that the curl could have been made even more sen- i] } EERE BOEOr ooo ooneoeesee I BES G0RonOoooooonr, oT Neoeneenem a Coe Degdoo SSS0SS000 00080807 207 40/50 5720 Sees Sur a a EEEEEEEEEEEe GRE SBaS BESESoeo8 Bags SoGES Cees SSEEE SEES SEESSSEEEETSEEESE i eS HEEEEE+HH BEEEC EEE bed SEER PSRs Bee < | : : F : [4 Lin : : : Hl F |_| 4 | ; A lal A | | [meh] / =H jauegugee06H i iL | F |_| : H of a as : iH : eit aa co cI CeCe Ho ESSEEEEEEEEEEEEEEEH BES ECeEe EEE Leet saa08 BF d Ei iapaa| CoC CCL oe we a tara Varaval PO LU Am 90 PET a ta t Chart showing the inverse sensitiveness (ordinate) of a counter-twisted system varying with the amount of counter-twist (abscissa). sitive than shown in figure 3, if its material had been resilient and of high viscosity. At all events, the datum reached (50™ Hg corresponding to a degree of deflection between the ends of a linear cm. of the turns of the curl) is markedly in excess of the results given at the end of §7, while the apparent vis- cosity of the metal has now no serious significance. Indeed since the viscous set above was actually negative (uncoiled curl), whereas in the simple curl ($5 et seq.) it is positive, a virtual elimination of viscosity is clearly possible. TABLE 2.—Showing the increased sensitiveness resulting from mereased counter-twisting. Counter-twist Pressure Pressure per Turns of in degrees per degree, degree, per VII, of are. per turn. linear cm. Remarks. No. .o) Yous delay Komp males 2°5 0 13°6 120 No viscosity. 2°3 60 9°9 95 <° 2°1 100 6°6 69 ss 1°9 140 4°3 50 ee 2 7 180 nes 56 Curl shows negative viscosity. . ° 2°3 (released) 0 10°0 98 Carl shows 90° set after releasing spring. 126 C. Barus—Counter-twisted Curl Aneroid. 11.—I want finally to put the remarks made in the above paragraphs to an actual test in continuous series of measure- ments with mirror and scale. A figure of the apparatus is given in $8. A curl, No. II, was wound for the purpose, of copper tube. A very smail leak was purposely left, so that the exhausted curl might gradually (several days) fill with air, while comparison with the attached mercury manometer were in progress. Preliminary tests made on the graduated circle showed the following results : Pressure Pressure per de- Pressure per degree, gree, per Description. Pressure. Defiection. perdegree. perturn. linear em. Diameter comys2:3°" em, ie. aire Turns, lara 76°80 84°75 0°926 14°] ” 124 Length, Len 57°92 =64°80 = 0894 40°70 45°75 0°0 0:0 The relation is not linear, though the discrepancy may be thermal. The curl is only moderately sensitive (for in §10 this sensitiveness has been doubled), and not quite free from viscosity. It is therefore interesting to see in how far preci- sion can be obtained with this short and not very favorable coil, when suitably counter-twisted. The spring used in counter-twisting was of brass spring wire, 0°05 diameter. It consisted of 31 turns about 1:5 in diameter and when stretched was 1:7™ long. The counter- poise weighed something over 258. The system was adjusted by first exhausting the curl aneroid to a nearly perfect vacuum, after which the mirror was brought back to its original position (that is before exhausting) by the spring. There was thus but little strain left in the curl. The method adopted was therefore the third in § 8. A thermometer hung beside the mercury manometer and the curl, giving the temperature of both, though not closely enough (as observation showed). A fine bulb thermometer placed within the coils is essential. The scale was at a dis- tance 3°5™°' from the mirror. Of the two continuous series of observations made I shall, for brevity, only give an example of part of the second. The first is defective in temperature. Table 3 contains the results, showing in the first three columns the height (0) of the mer- cury in the manometer (reduced to 0° C.), the temperature (7) of the air near the curl, and the reading (s) in the telescope of the deflections of the curl. Observations were made in groups of four or more, during different parts of the day for different days. Temperature was very variable, thus adding to the severity of the test. j C. Barus—Counter-twisted Curl Aneroid. 12a The column headed B is the virtual barometer, to be ob- tained in a way presently to be given. Taste 8.—Showing the behavior of the curl aneroid as compared with the mercury manometer. b Ss t B ) b 8 t B Tem- Manometer Tem- Manometer Manometer. Curl. perature. at 23°. Manometer. Curl. perature. at25°. cm., Hg. cm. Oe cem., Hg. cm., Hg. cm. ORC: cm., Hg. 75°49 54°75 25°6 75°53 73°69 58°80 29:0 75°99 74°89 49°40 25°8 74°95 Towle 54°30 28°8 75°44 74°58 46°70 26:0 74°65 74°36 46°30 28°4 746] iargo 40°30 26°3 74°05 (ay 32°00 280 Wo 19 75°58 56°50 27-3 CONS 76°06* 57°40 Zak 15:92 75°38 54°60 27:3 (Gea UO 50°70 Mail 75°15 74°95 50°60 27°5 75.14 74°72 45:00 23 2 74°59 74°66 47:90 28°0 74°88 74:02 38°40 23.6 13°92 75°59 57-60 29°2 75.90 76°01 58:00 23°5 75°90 75°15 54-00 29 6 75°50 75°69 54°90 23°5 75°58 44:22 44-70 30°0 74:59 ome 50°00 23°5 15°07 arog 41°60 29°8 74°24 74°76 45°80 24°2 74-70 15°57 57°90 30°2 75°96 7646+ 60°70 Des 76°22 15:35 56°50 30°2 75°74 75:70 54°55 22°3 75°50 74°95 52°30 30-0 Waro2 75°22 50°50 22°8 75°05 71°92 21°10 29°6 C225 Tare 37 85 23°4 73 82 oD: ayy al) 29°4 15°87 76°36 63°00 26°0 76°44. 74°95 52°40 29°2 75°26 75°86 57-30 26°1 15°92 14:45 46°60 29°2 74-76 15°42 54°50 26.3 (hensy? 74:02 42°70 29°2 74°33 7478 47°60 26°2 74°87 15°64 58°50 29°4. TOEOt 16°34f 57°60 20°8 76°02 (3:16 54°30 29°4 75°49 5°93 54°65 21-0 75°63 74°33 45°80 ole 74°63 75°49 51°10 21°2 75°20 73°38 36°10 29°0 73°67 74°86 45°20 21°6 74°61 * One day after. + Three days after. t Four days after. If the deflections (s) are platted as a function of the pressure (b,) a series of detached lines are obtained, which usually lie at some distance apart and are not quite straight even for a single group. Inspection shows these discrepancies to be principally due to temperature. It is not at once obvious how the tem- perature coefficient is to be computed, unless it be assumed, conformably with § 4, that the corrected loci are really straight. The following is a method of calculating the temperature corrections without entering into unduly complex computa- tions: Let s be the deflection (scale reading) at the pressure } and the temperature 7, let s’, 0’, ¢’ have a similar relation, and let s, c and a be constants. For any two groups, put (s+s))(1+ at) =be, a (s’ +5)(1+at')=0'e. ) If by aid of the chart of detached lines just referred to, the values of 6 be taken for s=s’, then b’—b 128 C. Barus—Counter-twisted Curl Aneroid. Proceeding thus, I found #=0:001; and with this value I reduced the observations by placing (1) under the form ar a ee (3) Hence s is a linear function of 4; and this postulate can be tested graphically without computing ¢ and s. J is given in the last column of Table 3. The conditions under which these data were obtained were very unfavorable. Asa result of the delicate suspension, the jar of wagons and cars passing the laboratory often made it impossible to read the scale millimeters in the telescope. Moreover I was unprepared for so large a temperature coeffi- cient, a, as has just been adduced; and I did not therefore take such special precautions against currents of air which should have been taken, seeing that the thin metallic helix adjusted itself at once to temperature, whereas the thermome- ter follows sluggishly after. Nor was the latter placed near enough to the coil. Hence since a single degree of tempera- ture corresponds to nearly 0°1°" of the barometer, discrepancies* of half a millimeter of pressure must have been incident to the work, and the table and chart bear this assertion out. Points which lie out of position are isolated in each group and the discrepancy is the result of a current of air. In spite of the unfavorable choice of metals (copper and brass) the behavior of the copper curl is thus seen to be satis- factory so far as the present purposes go: for a development of principles of construction has only been aimed at. The tel- escope reading for this small curl is 9°8°" per em. of the mer- curial barometer. 12. Concluston.—Having reached this stage of progress, it seemed expedient to break off the work, for it would not be worth while to proceed with fine measurement without making the system of more highly viscous material at the outset. In brief: tubing preferably of resilient brass 3-4 millimeters in diameter or less, with walls 0°01™ thick or less (by solution) is desirable. The tube should be heated in steam to remove excesses of drawn strain and thereafter flattened and coiled until the walls all but' touch. The counter-twisting spring is to be either helical or watchspring-shaped, and of steel annealed at 400° C., or a suitable long quartz fiber. A long curl would not be self-sustaining; but it could be made so without interfering with its free action, by a series of equidis- * In the face of the large thermal effect discussed in the text it would be use-. less to endeavor to evaluate the effect of viscosity. Nor would a metal of low viscosity, like a spring of drawn brass, be chosen in tests of a final character. Penfield and Forbes—Chrysolite-Fayalite Group, etc. 129 tant radial spokes attached at the inner face of the curl with their central ends fastened to a fiber of silk. If, therefore, the conditions investigated at the end of §§ 6, 10, be called to mind, a non-differential curl aneroid (§ 8, ease 3) less than a meter long, provided with a mirror for regis- try, will give account of variations of atmospheric pressure of a thousandth of a millimeter of the barometer, provided the mounting is sufficiently free from tremor, and temperature be kept constant to a few thousandths of a degree during the inter- val of observation. These conditions will be much less severe if the parts of the counter-twisted system are especially chosen (as stated in § 8, case 2) and twisted with reference to viscosity, rigidity and temperature. Indeed the chief result of the present paper is the exhibition of the properties of the counter-twisted system. A continuous mechanism has been brought forward which not only minimizes the hurtful effects of viscosity and of the ther- mal changes both of viscosity and of rigidity, but which accomplishes these desirable results in such a way as to remark- ably increase the sensitiveness of the instrument. Wilson Physical Laboratory, Brown University, Providence, R. I. Art. XVI.—Fayalite from Rockport, Mass., and on the Optical Properties of the Chrysolite-Fayalite Group and of Monticellite ; by S. L. PENFIELD and E. H. Forsss. In the fall of 1890, Mr. J. H. Sears of the Peabody Academy of Science, Salem, Mass., while visiting the Rockport Granite Quarry, found a crystalline mass of a dark colored mineral, which proved on examination to be fayalite. The occurrence of this mineral is so unusual that it was considered worthy of a special investigation and was sent for that purpose to the Mineralogical Laboratory of the Sheffield Scientific School. The material offered an excellent opportunity for an investi- gation of the optical properties of fayalite, which had never been determined, and the results were of such a nature that it seemed best to extend the study to the different members of the chrysolite group, in order to determine the effect upon the optical properties due to the mutual replacement of iron and magnesium. Fayalite. The material from Rockport was a crystalline mass, found at a depth of 60 feet, near the base of a large boss, or vein of Am. Jour. Sci1.—FourtH Srrtes, Vou. I, No. 2.—FEBRUARY, 1896. 9 130 Penfield and Forbes—Chrysolite-Fayalite Group, ete. pegmatite. It occurred at one side of the pegmatite vein in the massive hornblende-biotite granite, as a lenticular shell of varying thickness, about 12 to 16 inches in diameter, filled on the inside with loose earthy material and enveloped by a layer of magnetite about one inch thick. The material showed no crystal faces, but two distinct cleav- ages at right angles to each other. The color of the mineral on a fresh fracture is a dark resinous greea, but thin edges transmit a yellowish light. On examining fragments with a microscope it was found that they were fresh and transparent, but permeated with grains of magnetite. The material for analysis was therefore prepared by pulverizing and sifting to auniform grain and freeing from magnetite with a magnet. The material thus prepared was of exceptional purity. The specific gravity taken with the pycnometer and the results of the analysis are as follows: Specific gravity 4°323, 4°316, 4°317. Average 4°318 T. IL. Average. Ratio. S10, eee Oo 30°05 30°08 “501 1:00 FeO) 2: 68°04 68°19 68°12 "946 Mi@)22 FHT 65 “(2 "010 + 1°004 2°00 H,O aya "88 87 *80 ‘048 Total... 99°80 99°76 99°80 The ratio indicates almost pure ferrous ortho-silicate Fe,SiO, with only a trace of manganese and no magnesia, although a careful test was made for the latter. The small amount of water is considered as basic. A high temperature is needed to expel it, and limonite, the most likely mineral to be formed by decomposition, was wanting. If the water is disregarded, the analysis shows a slight excess of silica. The cleavages served for the orientation of the optical proper- ties and plates cut parallel to these, as described by Penfield and Pratt,* proved to be parallel to the pinacoids 001 and 010. The first of these gave no interference figure in polarized light, while the second showed the emergence of the acute bisectrix. The divergence of the optical axes was measured on a large Fuess axial angle apparatus, and although it could be measured in air, it was found best to measure it in a-mono- bromnaphthalene. The plates also served for determining the pleochroism. The indices of refraction were obtained from prisms carefully oriented by the cleavages. The results are as follows: for yel- * This Journal, vol. 1, p. 387, 1895. Penfield and Forbes—Chrysolite-Fayalite Group, ete. 131 low Na, a = 1°8236, 8 = 1°8642, 7 = 18736, 7 — a= :050, the orientation being @a=c, b=a, ¢= b as iner -ysolite. The plane of “the optic axis is the base and a is the acute bisectrix. The double refraction is, therefore, negative. The dispersion is p>», 2H, Li=57° 27’: 2H Na 56", 32) OF. Tl = 55°’. The index of refraction for amono -bromnaph- Gitene being 1:6577 at 23° C. for yellow, V, = 24° 55’ and 2 = 103° 25’, 9E, was also measured and found to be 103° From the values of a, Band 7, according to the usual aia V, was found to be 64° 42’, hence V, = 25° 18’, agree- ing closely with 24° 55’ as given above. The pleochroism is distinct, in sections about 0°5™™ thick for rays vibrating parallel to b orange yellow, parallel to a and c greenish yellow. To make sure of the orientation of the optical properties as given above, where only cleavages were available, a comparison was made with the excellent er ystals from the obsidian of the Yellowstone Park described by Iddings and Penfield.* For this purpose, material was supplied by Mr. Arnold Hague of the U.S. Geological Survey, to whom the authors desire to express their thanks. On these transparent, but minute crys- tals, the basal cleavage and the emergence of the obtuse bisec- trix at right angles to the pinacoid 100 were distinctly seen, the plane of the optic axis being 001. No indication of a cleay- age parallel to the pinacoid 100 was observed on the fayalite from Rockport, and the statement in many mineralogies of a cleavage in that direction is probably erroneous. Hortonolite. Under this name Professor Brusht has described a mem- ber of the chrysolite-fayalite group, found in an iron mine at Monroe, Orange Co., N. Y., and characterized by its dark color and high iron percentage. In appearance it resembled the fayalite from Rockport. In order to ascertain the chemical composition of the mineral upon which the opti- cal properties were to be determined, a new analysis was made upon material from a large mass, showing cleavage in two directions, which served for the optical orientation, this cleav- able material being better adapted for optical work than the erystals at our disposal. Grains of magnetite were disseminated through the mineral and the material for analysis was therefore purified as described under fayalite. The results of the analysis are given below, together with those obtained by Mixterand quoted by Brush: * This Journal, vol. xxx, p. 58, 1885. + Ibid., vol. xlviii, p. 17, 1869. 132 Penfield and Forbes—Chrysolite-Hayalite Group, ete. Specific gravity 4:047, 4.030, Average 4-038. Mixter. 1B INE Average. Ratio. Sp. gr. 3:91 SiO) 2221 33°60. (33°94 Bord © "62 ‘962, “E00 33°59 BeQe) 2. 47°19 47°32 (47:96 658) 44°37 MnQOrc:. 4°76 4°32 4°54 064 4 1:093 1:95 4°35 MeO 2) 14:02) 18°745 13°88) “Bt7 a 16°68 ERO E act 0°48 0°48 OrdB 9) “O26, K,O 0:39 H,O 0:26 10:05 QU 78059 99"93 — 99°64 The ratio of the SiO,: RO = 1: 1°95 ornearly 1:2, giving the formula (FeMgMn),SiO,, and, therefore, the mineral holds an intermediate position between fayalite and the iron- rich chrysolites. The cleavages are poorer than in fayalite, but parallel to the same faces 001 and 010. Plates and prisms were prepared as in fayalite, the optical orientation and the character of the double refraction being the same. The following results were obtained : For yellow hght Na, a= 1-684, S—1°7915, y= Pagar y—a 084i, 2H, li= 76° 59; OH Na= We. eee Tl = 75° 45’; hence with 2 Na for a-mono-bromnaphthalene = 1°6567 at 25° C., V,, = 34° 42”. From the valuesof a, 8 andy, V, was found to be 54° 55’; hence V,= 35° 5’, which agrees closely with V, as given above. The pleochroism was similar to that of fayalite but weaker, for rays parallel to a and c very pale yellowish green, and parallel to 6 pale yellow. Chrysolite. For the optical constants of chrysolite we are indebted to Des Cloizeaux* who gives a = 1°661, 8 = 1:678, y = 1°697, for yellow Na, 2V, =87° 46’, the double refraction being posi- tive and the dispersion p v. The last three are the determinations of Des Cloizeaux and Zimanyi already referred to. 134 Penfield and Forbes— Chrysolite-Layalite Group, ete. Per Cent Axial Angle 2V Material. Locality. of FeO. measured over 4. B: Fayalite, Rockport, 68°1 49° 50' 1°86 Hortonolite, Monroe, 47°83 69 24 1°791 Chrysolite, Auvergne, 13°0 89 36 1692 - Vesuvius, 12°6 89 42 . a Hawaii, 10°3 91° 2 oi Egypt, 9°2 91 19 1°678 es New Mexico, 8°6 91 24 sf Unknown, ? 92 14 1678 ec East Indies, y 92 45 1°670 Fosterite, Vesuvius, Die 93 50 1°657 The effect of a decrease in iron is to constantly increase the value of 2V, which, at the iron end of the series, changes much more rapidly than at the magnesia end, as may be seen by the curve which has been plotted, and where the percentages of FeO have been taken as ordinates and tae values of 2V as abscissas. The decrease in FeO is accompanied by a decrease in the value of # and also in the strength of the double refrac- tion. With the FeO about 12 per cent 2V for yellow equals nearly 90°. Chrysolites con- taining less than 12 per cent FeO have c, or the erystallo- graphic axis @ for the acute bisectrix and are _ optically positive with dispersion p v. Monticellite. In order to make the investigation of the chrysolite group more complete, we have included monticellite CaMgSi0,, which in its crystallographic relations is very close to chrysolite, the crystals from Magnet Cove, Arkansas, investigated by Genth and Pirsson* furnishing excellent material for the pur- pose. An abundant supply of this rare mineral was generously furnished to us by Messrs. Geo. L. English & Co. of New York, to whom we take pleasure in expressing our thanks. As * This Journal, vol. xli, Dp soo oole Penfield and Forbes—Chrysolite-Fayalite Group, ete. 135 Genth’s analysis showed a loss by ignition of 2°29 per cent and 2°08 per cent of P,O,, it evidently was made on impure mate- rial and consequently a new analysis seemed necessary. The material which we were able to select was of exceptional purity and yielded the following results: Te ae Mean. Genth. Specific gravity 3:022 3°047 3°035 Analysis I. IDE Average. Ratio. iO 36°86) 36:70) 36°78) 613 Gs) b200 33°47 FeO 4°61 4°89 4°75 °066 ) 5°01 MnO 11°58 1°67 1 G2i O23 G29) earl O2 ea NEGO) 21-44)5 21°75 21°60. +540 J 20°61 2On) 34:23) 9934-39) 34:31) “613 613 1:00 35°25 H,O 97 "93 "95 fen 2°29 FiO, 72:03 Total 99°69 100°33 100°01 NEON a Soy 99°95 The ration of Si0,:(MgFeMn)O:COa0 = 1:00: 1-02: 1-00 is very satisfactory, and gives the usually accepted formula CaMgSiO, in which a little Mg is replaced by Fe and Mn. A careful test was made for P,O, but none was found, and un- doubtedly Genth’s material, as assumed by him, was contami- nated with apatite. rom the crystal from which the material for analysis was taken, a few colorless and transparent grains were selected which when heated im the closed tube gave only the merest trace of water, the pure mineral is therefore practi- eally anhydrous, but in order to obtain sufficient material for the analysis it was necessary to include some slightly brownish grains, and these were permeated with cracks along which decomposition had commenced, which accounts for the water. For the determination of the optical properties a single crys- tal like the one figured by Pirsson was used. The indices of refraction, determined by means of prisms, are as follows: for yellow light, Na, @=1-6505, 3=1-6616, = 1-6679,;, —a= 0:0174 ;also 8, Li = 16594 and f, Tl = 1:6653. The optical orien- tation is @=c,b=a, and c=b asin chrysolite. The plane of the optical axes is 001 and ais the acute bisectrix, consequently the double refraction is negative. The dispersion is o > v as fol- forme bln (aco , 2 Na= (5° 21 sand 2HeTl = 74> 52. The measurements being made in a-mono-bromnaphthalene, Wen —ool Vee Na 37° Sl and V2, 0F—936 282 while from the values of a, 8 and + as given above V,,Na by ecaleula- tion, was found to be 37° 9’. Laboratory of Mineralogy and Petrography, Sheffield Scientific School, New Haven, October, 1895. 136 Scientific Intelligence. SCIENTIFIC INTELLIGENCE. I. CHEMISTRY AND Puysics. 1. Recent investigations on hydrogen peroxide.—Our know]l- edge of the chemical nature of this interesting substance was much advanced by the work of TRaupg, chiefly recorded in the “ Berichte ” of the German Chemical Society from 1882 to 1893. Traube came to the conclusion that hydrogen peroxide cannot possess the same constitution as the peroxides of manganese, lead and silver, because it is not produced, like these, at the anode in electrolysis, but is produced along with hydrogen at the kathode, especially when oxygen is blown against this electrode. He con- sidered it, therefore, not an oxidation-product of water, but a reduction-product of molecular oxygen. This view is also sup- ported by the fact that hydrogen peroxide is never formed by the oxidation of hydrogen by nascent oxygen or other oxidizing agents, while, moreover, it reduces all the powerful oxidizing agents or is destroyed by them. Traube therefore rejected the generally-accepted di-hydroxyl formula, HO.OH, and advanced the view that the substance is a hydride of molecular oxygen, as. expressed by the formla H[O:O]H. The interest taken in hydrogen peroxide has been considerably increased since WOLFFENSTEIN showed that the substance could be purified and concentrated, with but small loss through decompo- sition, by distillation under diminished pressure, (Berichte, 1894, . 3307). To perform this operation it is necessary that the solu- tion should be free from compounds having an alkaline reaction as well as from every trace of salts of the heavy metals and from solid substances of all kinds. Wolffenstein recommends a simple distillation of the commercial 3 per cent solution under a pres- sure of about 68™™ for the production of small quantities of the pure solution. For larger quantities he advises evaporation at ordinary pressure upon the steam-bath until a concentration of about 20 per cent is reached, then concentration under dimin- ished pressure to 50-55 per cent, after that, extraction with ether (Briihl considers this step dangerous, as will be seen beyond), and finally distillation under diminished pressure. B repeated fractionation he obtained a product which boiled at 84— 85° at 65™™, and which contained 99:1 per cent of the sub- stance. Spring (Zeitschr. anorg. Chem., viii, 424) has prepared large quantities of hydrogen peroxide by Wolffenstein’s method and has determined some of its physical properties. He found it to have a blue color about half-again as intense as that of water, and concludes that this fact supports Traube’s theory concerning the constitution of the compound, because Olszewski having shown that liquid oxygen has a blue color about fifty times as strong as that of water, he concludes that oxygen retains more of Chemistry and Physics. 137 its original properties in hydrogen peroxide than in water. Spring concludes also, from determinations of the specific heat of hydrogen peroxide, that the potential energy of the elements is not completely used in forming this compound. Brtut (Berichte, 1895, 2847) has recently described some important work upon the subject under consideration. Using W olffenstein’s process, usually omitting, however, the extraction with ether, he has prepared hydrogen peroxide in a water-free condition. The boiling-point of his purest product was 69°2° at 26™™. He observed that the distilled substance is much more stable than that which is less pure, and that the nearly water- free liquid suffered scarcely any decomposition when kept cool and ina dark place for five or six weeks. Agitation promotes decomposition, as does also an increase in the surface with which the substance is in contact. Sharp points and rough surfaces have a marked action, and when the compound is brought into contact with a ground-glass surface the decomposition is violent. Paper, cloth, wool and asbestus cannot be used for filtering the concentrated liquid ; in fact, wool is ignited by contact with a small amount of it. Gun-cotton does not decompose it and can be used for its filtration. Brtih] considers the specific gravity of hydrogen peroxide as the best criterion of its purity. The highest specific gravity which he obtained was, d2=1-4584, while a liquid containing 99°48 per cent of the pure substance, accord- ing to analysis, had a density of 1°4094. He experienced no trouble from explosions when solutions were concentrated and purified simply by evaporation and distillation, but some material which had been repeatedly treated with ether gave off much gas with an odor of ozone upon distillation, and an oily, colorless resi- due, which did not volatilize at 100° was left. This fluid, amount- ing to only 1 or 2° showed but slight explosive properties when a drop of it was conveyed into a Bunsen flame by means of a platinum spatula; but when a glass rod with a sharply cut end was brought into contact with the remainder, a fearfully violent explosion took place which caused much damage in the labora- tory. hU6vluaw | oe +) (meters | dynes | dynes | dynes | dynes | dynes Austin. Texas (Capitol)_._|30 16 30 97 44 16} 170 (979°274 979-307 979°333/979°369| —-062 Austin, Texas (University) 30 17 11 97 44 14| 189 979°269) 979°306 |979°332|979°370| —-064 Miaredo, Texasvi 1.22 wal 27 30 29 |99 31 12} 129 |9%9-068| 979-095 |979-109|979-160) —-065 Galveston, Texas___.___- 29 18 12 94 47 29 3 |979°258| 979°259 |979°259/979-294| —-035 INew ‘Orleans ia,. 222.2 29 56 58 90 @4 14 2. (979°310) 979-311 |9T9°311/979' 344)" —=-033 Calais, Maite 26226 oJ" 45 1111/6716 54| 38 980°618, 980°626 |980°632'980°647, —-021 Residuals, g re- duced minus g com- puted. Faye’s reduc- tion dynes — 036 — ‘038 —051 — 035 — ‘033 —'015 The continuation of the investigation of the question of reduction to sea level and of the anomalies of gravity is essen- tial to a satisfactory application of pendulum observations in geodesy. The four stations in the south are of interest in this connection as well as from a geological point of view because of their comparative location. Two, Austin and Laredo, are situated toward the interior about 150 and 135 miles (241 and 217 kilometers) respectively from the Gulf coast, and in a region of erosion. Galveston is on a sandy island close to the main land, from which it is separated by a bay into which empty some small rivers. New Orleans is on the banks of the Mississippi in the midst of the vast alluvial region built up by that river, but is about 85 miles (137 km.) from the present mouth of the river. Galveston and New Orleans are about equally distant (nearly 100 miles or 161 km.) from the 100 fathom (183 meter) curve in the Gulf of Mexico, beyond which line the water depths rapidly increase. It has been estimated that an area of about 1,800,000 square miles is drained by the Mississippi, Rio Grande and neighboring rivers emptying into the Gulf of Mexico from the north, and that the enormous amount of sediment carried by these rivers is depos- ited within an area of 300,000 square miles along the northern part of the Gulf.* If this added load accumulating for ages and displacing sea water of only about one-third its density, were sustained by a rigid earth, it would seem that there must be an effect on the force of gravity measured at the surface which would be quite within the range of observation, and gravity in the loaded region would be found greater than normal. If, on the other hand, there existed a perfect condition of equilibrium, and an area so loaded continually adjusted itself and sank in proportion to the load, in accordance with the theory of isostasy, we would expect to find the force ot gravity quite normal. Should there be a lag between cause * These drainage and deposition areas are shown in the map, here reproduced, with the author’s permission, from Mr. McGee’s paper, later referred to. The positions of the gravity stations are also shown. Deposition area of Gulf of Mexico WW Putnam—fesults of Pendulum Observations. 189 and effect, that is, should the load accumulate to a certain extent before the sinking occurred because of the partial rigidity of the crust or some other retarding cause, we would again expect to find gravity in excess, but only to a small amount. It is of interest, therefore, in this connection to examine the gravity residuals at the four southern stations (differing as they do but little in latitude) as given in the last two columns of the table. The minus sign indicates a defect of gravity as compared with that given by the theoretical Degradation area tributary to = Gulf of Mexico Stations: N. O., New Orleans; G., Galveston; A., Austin; L., Laredo. formula used, but it is only the comparative values between the coast and interior stations that need be considered here. There is practically no difference in the results for the two coast stations, New Orleans and Galveston. With Bouguer’s ‘reduction, gravity at the coast stations, however, is apparently about -030 dyne greater than at the interior stations, but this is probably due to the fact that the interior stations are more elevated, as negative residuals almost invariably appear in ele- vated regions with this method of reduction. With Faye’s reduction, gravity at the coast stations is (017 dyne greater than 190 Putram—fesults of Pendulum Observations. at Laredo and only ‘003 greater than at Austin, or an average excess of -010.* As far as it is safe to draw an inference from so small a number of observations (and considering the unfav- orable location of the station at New Orleans), gravity is appar- ently slightly in excess near the Gulf coast as compared with interior stations. A determination nearer the present deltat of the Mississippi would be of interest in this connection, as well as observations in the same latitude on the other coasts of the continent. The smallness of the differences found indicate a close approach to the condition of hydrostatic equilibrium in - this region. The fact that the slight excess is at the coast sta- tions points possibly to some retardation in reaching this condition. It is interesting to compare the conclusion reached by Mr. McGee in discussing the condition of the same region as deduced from geological and other considerations, in these words: “So the data relating to the condition of the earth’s crust derived from the modern Gulf of Mexico indicate that throughout the vast geologic province of southeastern North America, isostasy is probably perfect, i. e., that land and sea bottom are here in a state of hydrostatic equilibrium so delicately adjusted that any transfer of load produces a quanti- tatively equivalent deformation.” While coast stations have in general shown less apparent irreg- ularity in the force of gravity than is often found in the interior, doubtless because they are situated so near the sea leveland are therefore more free from the uncertainties asso- ciated with the sea level reduction, still greater discrepancies appear than can be attributed to errors of observation. In the following table are collected the residuals observed minus computed gravity, for the stations on or near the coasts, thus * The attraction of an extended horizontal plate of rock of average density 31 feet (9'4 meters) thick corresponds to ‘001 dyne force of gravity. + The interesting fact has been recently brought out by an Engineer officer, that a rise of one foot, since 1877, in the level of mean high water in the Gulf of Mexico is indicated by the tidal observations at Port Hads at the delta of the Mississippi. This change would correspond to a subsidence of the land of like amount, but may possibly be accounted for by a local settling near the bench mark. t‘‘The Gulf of Mexico as a Measure of Isostasy,” W. J. McGee, this Journal, Sept., 1892, vol. xliv, p 189. The following conclusions are also quoted as bear- ing on the same subject: ‘‘Tt appears that the crust, in the form in which it exists, must be in a condi- tion of approximate hydrostatic equilibrium, such that a considerable addition of load will cause any region to sink, or any considerable amount denuded off an area will cause it to rise.” ‘Physics of the Earth’s Crust” (2d edition), by Rey. O. Fisher, p. 355. ‘“‘Tt may be laid down as a general rule that where great bodies of sediment have been deposited over extensive areas, their deposition has been accompanied by a subsidence of the whole mass.” ‘‘On some of the greater Problems of Physical Geology,” by Major C. E. Dutton, Phil. Soc. of Washington, vol. xi, p. 55. Putnam—fesults of Pendulum Observations. LOE far determined in the United States (exclusive of Alaska) and for the two methods of reduction to sea level already men- tioned. (— indicates gravity smaller than normal, + greater than normal.) Summary of gravity residuals for Coast Stations. Residuals, observed minus computed g. Coast. Station. Latitude | Bouguer’s| Faye’s* north. reduction. | reduction. Pay 78 dynes dynes North Atlantic _|Calais, Maine ._.-| 45 11 —'021 | —-015 Cambridge, Mass. | 42 23 | —‘009 | —:003 Boston, Mass. ---- 42 22 —'007 | —:001 Hoboken, N. J. --| 40 44 +:008 | +:019 Princeton, N. J..-| 40 21 —'038 = a2 Philadelphia, Pa._| 39 57 +:004 +011 Baltimore, Md.._._| 39 18 —'034 | —-023 Washington, D.C.| 38 53 +:014 + °026 Gulf of Mexico .|New Orleans, La. _| 29 57 —'033 ="033 Galveston, Texas_| 29 18 —'0385 | —:035 North Pacific .. .|\Seattle, Wash. __.| 47 36 —°135 — ‘090 San Francisco, Cal.| 37 47 —°016 — 049 The variations in these residuals, while not considerable except in‘a single case (Seattle), are nevertheless of much inter- est. It will be noted that with either method of reduction the largest excess of gravity appears at Washington.t A computation of the amount of flattening of the earth has been made from the 33 results for gravity obtained in the United States in 1894 and 1895. These observations are of course not well suited for this purpose except in combination with results in other parts of the world, as the work of 1894 was intended to develop the effect of elevation and continental position, and not the variation of gravity with latitude, and for this reason those stations differed little in latitude, and even with the few southern and northern stations added during the ast season the extreme range in latitude is less than 18° (from 27° 80’ to 45° 11’). Nevertheless it is of interest to see how * Some slight differences in the last column of this table from values previously given, are due to revised estimates of average elevation. The results from Hoboken, Baltimore, Seattle and San Francisco are from earlher determinations. + Commandant Defforges, comparing results obtaized on the shores of various oceans came to the conclusion that each ocean possesses a characteristic anomaly in the force of gravity along its shores. ‘Memorial du Dépot Général de la Guerre, Observations du Pendule,” vol. xv, p. 194. 192 Putnam—fesults of Pendulum Observations. reliable a value can be obtained from such meager data, derived entirely within the borders of the United States. The values reduced to sea level by Faye’s reduction were used, and to simplify the computation were averaged together for each degree of latitude. Conditional equations were formed of the form gy = «+ y sin’ ¢g, representing the variation of gravity with latitude, where x is gravity at the equator, y is the differ- ence between gravity at the pole and the equator, and g, is observed gravity at latitude g; « and y were then computed by the method of least squares, and substituted in Clairant’s theorem. Using Helmert’s expansion of Clairant’s theorem,* the value 3;'; was obtained for the amount of flattening, or the difference between the earth’s polar and equatorial axes divided by the equatorial axis.t Helmert in 1884 obtained the value s5/5.3 from a discussion of pendulum observations in various parts of the world, and Clarke from a similar discussion in 1880 obtained 5,45. Two of the most important values obtained from are measurements are those of Bessel (1841) soy.y and Clarke (1880) 544.5. A variation of one unit in the denominator of these ratios corresponds to a change in the dif- ference between the earth’s major and minor axes, of about 481 feet (147 meters), and from the discrepancies in the above and other results one may judge that the actual uncertainty may be nearly a mile.t The value derived from the pendulum observations in the United States is not very different from Helmert’s and Bessel’s, but it is of significance only in showing that a fairly accordant result may be obtained from so few observations very narrowly distributed in latitude, and in fur- ther confirming the validity of the reduction to the sea level tentatively employed, and the theory of the condition of the earth’s crust on which that reduction is based, the equilibrium or isostatic theory.§ This result has been obtained by combin- ing observations made at altitudes above sea level ranging from 6 ft. to 14085 ft. (2 m. to 4293 m.), and in a great variety of continental locations. * Geodasie,” by F. R. Helmert, vol. ii, p. 83. + A result for the flattening of 54, was obtained by comparing the four southern stations, Austin, Laredo, Galveston and New Orleans, with four stations in nearly the same longitude near the 39th parallel in the central plains, St. Louis, Kansas City, Ellsworth and Wallace. + Prof. Harkness says: ‘“ Indeed the facts thus far advanced scarcely warrant any conclusion more definite than that the flattening probably lies between 34, and z1,, but we shall see presently that there is some further evidence which tends in the direction of the smaller limit” [,45].—“ The Solar Parallax and its. Related Constants,” p. 103. § For discussions of the relation of the Coast and Geodetic Survey pendu- lum observations to these theories, see papers by Mr. G. K. Gilbert, Bulletin Phil. Society of Washington, vol. xiii, p. 61: by Rev. O. Fisher, Nature, vol. lii, p. 433, Sept. 5, 1895; and by M. Faye, Comptes Rendus de l’Académie des. Sciences, 20 May, 1895. Peckham and Linton—Trinidad Pitch. 193 Art. XX.—On Trinidad Pitch; by 8S. F. PEckHAm and Lavra A LINTON. THE bitumen found on the Island of Trinidad in the so- called Pitch Lake and in its neighborhood, has entered com- merce under the name of Trinidad Pitch. That which is found within the lake is called “ Lake Pitech;” that found outside the lake is called “‘ Land Pitch.” As it occurs it is a unique substance found nowhere else in nature. It consists of a mixture of bitumen, water, sand, decayed vegetation and gas in such definite proportions that within certain limits the composition of the entire mass is uniform. The bitumen has never yet been investigated in such manner as to determine its relations to other bitumens, but it appears to be of vegetable origin and convertible into solid asphaltum by processes of nature. In its natural condi- tion about one-third of it is water. Deprived of water it is about one-third sand. When the bitumen is dissolved away from the sand under the microscope, the silica avec re to be in exceedingly minute amorphous particles from zyiqq tO a9'o0 of an inch in thickness. When freed from organic matter by burning, the silica appears in small sharply “angular grains, stained by iron and a small quantity of bluish clay. The organic matter not bitumen consists of fragments of vegetation and disorganized cellular tissue, with products of the decompo- sition of wood. As the bitumen rises in the center of the so-called lake it is inflated with gas. When the masses are broken into the structure resembles vesicular lava. The gas cavities are of all sizes, some of them very large and in the aggregate occupy at a rough estimate from one-third to one-half the volume of’ the pitch. At any point in the deposit removed from the center of the lake, the gas, in part, has escaped from the asphaltum and the mass become more compact. Both within and with- out the lake the pitch is saturated with water. It is in this condition without viscosity and can be trodden upon or squeezed in the hands without adhesion to either hands or feet. In this condition it cuts like cheese, hence the name, “ cheese pitch.” When freshly dug the color is brown, but if left in the sun it soon darkens, finally becoming a bluish-black. If a mass of any considerable size is laid in the sun, it will melt to a thin pellicle upon the exposed surface, and retain the larger part of the water at a temperature sufficient to remove every trace of water if it were dried in the shade. A mass exposed to the air out of the sun, immediately begins to dry out and Am. Jour. Sci.—Fourte Series, Vou. I, No. 3.—Marcu, 1896. 194 Peckham and Linton— Trinidad Pitch. in a short time loses nearly all of the water, which is in part readily re-absorbed if again exposed to dampness. The evaporation of the water precipitates within the pitch a small percentage of saline matter, chiefly sulphates of the alkalies and alkaline earths, that the natural water holds in solution. The hygroscopic property of the pitch is no doubt largely due to the presence of these salts. In selecting specimens that would fairly represent the char- acter of the mass of pitch both within and without the lake, we were largely governed by the appearance of the pitch and the relation of the several localities to one another and the cen- ter of the lake. No. 1 was picked up at random from the piteh taken from an excavation from which the cargo of the bark “Ella” was dug, during February and March, 1895. The excavation was upon a village lot about three-quarters of a mile from the lake towards Point La Brea. No. 2 is from a village lot which we have named the “ Pho- tograph Lot.” It was here that a pit was dug and photographs taken of the pit at intervals of ten days to determine whether any movement in the pitch was in progress by which a cavity dug in the pitch would refill. No. 2 was taken from the pitch removed from the pit. This lot had been excavated about six months previous and had nearly refilled, and was then being uncovered preparatory to the removal of a fresh supply of several thousand tons. It was about twenty rods nearer Point La Brea than No. 1.. No. 3 is so-called “Iron pitch” from the Photograph lot. This is pitch that has been melted and deprived of its water and gas. It is solid, ot a bluish-black color, with a dull earthy fracture and is slightly sonorous when struck. No. 4 was taken from a lot on the right hand side of the road approaching the lake, that was being excavated by Mr. Ghent. It came from a point 10 or 15 feet below the surface on the western border of the mass filling the ravine down which the overflow of pitch from the lake has taken place, and nearly on the opposite side of the road from the point from which No. 10 was taken. Fi No. 5 is No. 4, boiled to form Epureé, in Mr. Ghent’s boil- ing works near Point La Brea. | Nos. 6 and 7 were from opposite corners of a mass about 12 inches square and four inches in thickness. This mass was taken from a point on the northeast side of the lake on the outside of and near to the tramway, and was selected of con- venient size from among a quantity that had been broken with a pick preparatory to removal in the tram cars or carts by the Trinidad Asphalt Co. Peckham and Linton—Trinidad Pitch. 195 No. 8 is from an average from the same piece made up by breaking fragments from many points upon its surface. No. 9 is from the center of the lake or near it. The mass was soft enough to flatten in the shade, but did not stick to the paper in which it was wrapped. After drying it became ridged and brittle. No. 10 is an average from a large piece taken from an exca- vation being made by the Trinidad Asphalt Co. on the Belle- vue estate near the road leading to the lake. The excavation extended along the road for perhaps 1500 feet and was narrow. The pitch was clean and pure but was covered by rank vege- tation that grew upon and in the pitch itself, and not upon soil that covered it. This fact accounts for the high percentage of organic matter not bitumen, although the piece was taken several feet below the surface. No. 11 is a decomposition product of the pitch from the photograph lot. No. 12 is another decomposition product from the same lot. It resembled coke and may have been heated. It is the only material resembling coke that we saw in or around the lake and the amount was only a few pounds. No. 13 is also a decomposition product resembling No. 11, from the south side of the lake. It was enclosed by a pelli- cle of sun-dried, melted pitch, within which it was of a light brown color with a columnar structure, like starch, and was very easily powdered. It had the external appearance of asphaltene that had been precipitated from solution. No. 14 is from a pile of land pitch melting on the beach at Point La Brea, said to have come from the same lot as No. 1. No. 15 was brought from the lake about 1865, by the late William Attwood of Portland, Me. No. 16 is from the southeast side of the lake inside the road and was cut from the surface at a spot free from vegetation. The point was about half way from the tramway to the border of the lake. No. 17 is from the west side about mid-way of the tramway loop, where men were loading tram ears. It was picked up from under the feet of the men. No. 18 is iron pitch from the northeast side of the lake near where the left limb of the tramway, looking south, enters upon the lake. No.‘19 is refined land pitch, from the refinery of the Trini- dad Bituminous Asphalt Co. at Jersey City, N. J. It came from the same lot as No. 2. No. 20 is refined lake pitch, purchased in New York of the Warren, Scharf Co. 196 Peckham and Linton— Trinidad Pitch. No. 21 is from the northeast side of the lake near the left limb of the tramway looking south. No. 22 is from the northeast side of the lake near the right of the left limb of the tramway loop looking south, about one hundred feet from No. 21. No. 23 is from the northwest side of the lake on the west side of the right limb of the tramway loop looking south. No. 24 is from the south side of the lake near where the road ieaves the lake. No. 25 is from the northwest side of the lake on the west side of right loop of tramway looking south near a “ blow-hole.” No. 26 is from the south side of the lake near where the road leaves it, about one hundred feet from No. 24. No. 27 is Epureé from the boiling works of the Trinidad Asphalt Co., at Point La Brea. It was made by boiling a mixture of No. 10, No. 8 and No. 9. INOE:.0; 01,05, 9 and 17 represent commercial lake pitch. Nos. 16, 21, 22, 23, 24, 25 and 26 represent the contents of the lake occupying the annular space outside the tramway and embracing hundreds of thousands of tons. The area is about 60 per cent. of the surface of the lake. Nos. 1, 2, 4, 10 and 14 represent an average of commercial land pitch. Nos. 5 and 19 represent refined land pitch. Nos. 20 and 27 represent refined lake pitch. Nos. 3, 11, 12, 18 and 18 are rubbish so far as commerce is concerned, and are introduced here to show that there is rub- bish in the lake as well as outside of it, and also the relation of alteration products to the commercial pitch. The locations of the several specimens are shown on the accompanying map. No specimens were taken from near the border of the west side of the lake, because the pools of water were so wide as to make it quite difficult to get around among them. These specimens are believed to furnish a fair representation of the pitch as it occurs both within and withont the so-called lake, and also the refined pitch and Epureé made from the same. As before stated, the condition of the pitch in the entire deposit is that of complete saturation with water. Water is reached everywhere within a few feet of the surface and often stands in the areas from which pitch has been excavated. Both outside and upon the borders of the lake it appears to render the re-filling of the areas less rapid. In and near the center of the lake, the enormous volume of gas constantly rising, forces the pitch into any excavation more rapidly. The pitch is removed from near the tramway soon after it is dug, and before it has time to melt is weighed full of water Peckham and Linton—Trinidad Pitch. 197 and immediately thrown into the hold of the ship. The pitch from other parts of the deposit is dug up in large pieces and removed in carts to the beach, where it is immediately put into lighters and transferred as soon as possible to the hold of the THE PITCH LAKE —AND ITS — ENVIRONS = —— FROM — OFFICIAL SURVEYS. A=fR0AD TO OROPOUCHE. B-Roap To LAKE AND GUAPO, CG AnD D= Roaps To PoinTE Boyer. F=ArrroxiMATE LOCATION OF Power HousE AND TRAMWAY, Sl BRIGHTON ESTATE LAN BELLE VUE ESTATE. ship anchored off shore. In either case the pitch reaches the hold of the ship containing from 25 per cent to 30 per cent of water and considerable gas, especially that removed from the lake. Before being discharged, heat, the motion of the ship 198 Peckham and Linton— Trinidad Pitch. and the weight of the mass upon itself have rendered the mass within the hold of the ship nearly solid and the material is no longer the natural crude pitch, but something more or less removed from it by loss of water and gas. _In the case of samples like those taken by ourselves and packed in a trunk, labelled and carefully wrapped in paper, the loss of water was nearly complete before they reached New York. In fact it required only a week or ten days in Port of Spain to completely transform the cheese pitch from a moist, porous substance, cutting with a knife like cheese, to a hard brittle solid, readily broken into fragments that could only be cut with considerable difficulty, provided it was kept out of the sun. It is therefore manifest that commercial samples of crude pitch are not samples of natural crude pitch ; nor is it possible to bring away from Trinidad samples of “cheese” pitch in the natural condition. We therefore determined to analyze the specimens selected free from water and gas, and thus render the results comparable. The samples were severally coarsely powdered and air dried by placing them upon the laboratory table in the sun. In dry weather they soon dried to a constant weight. In damp weather they lost and gained within narrow limits indefinitely. Heated inan air bath to 50° C., they were soon brought toa con- stant weight. Heated at 100° C., to a constant weight, a vary- ing loss of volatile oils invariably followed, which showed that determinations of water at 100° C. as a constituent of the pitch leads to vitiated results from two sources: first, the percentage of water is not constant in the same specimen but varies with the condition of the atmosphere; second, pitch that is appar- ently very dry gives off an appreciable amount of volatile oils below 100° C. The samples were therefore dried to a constant weight, at a temperature below 50° C. Of course, if, for any reason, the amount of water in a given specimen of pitch is desired, it is easily ascertained, but it should not be reported as a constituent of the pitch, as the varying per- centage of water causes all of the other percentages to vary in the same specimen at different times. The dried specimens were then exhausted with petroleum ether. In the present instance the petroleum ether used for all the specimens came from the same barrel and was of specific gravity 74° B. The exhausted residues were dried at 100° C. and the difference in weight was computed as petrolene. The dried residues were then exhausted with boiling spirits of turpentine, washed with ethyl alcohol and dried at 100° C. to a constant weight. The loss was noted. The dried residues were then exhausted with chloroform and dried and the loss noted. The loss by turpentine plus the loss by chloroform is Peckham and Linton—Trinidad Pitch. 199 estimated as asphaltene. The distinction made by the use of these two solvents will be noted farther on. The dried residue from the chloroform exhaustion was then put into a platinum erucible and the organic matter burned off. The residue was inorganic or mineral matter, sand, and the small percentage of soluble and non-volatile salts present. The pitch was thus divided into that portion soluble only in petroleum ether, or “ netrolene,” that portion soluble only in boiling spirits of tur- pentine, and chloroform which together form “ asphaltene,” and with the petrolene constitutes the “total bitumen” ; also the “organic matter not bitumen ” and the “ inorganic matter.” Table No. 1 represents the results of the analysis of the 27 specimens described above. Any attempt to classify them as to the locality from which they were obtained by means of these analyses will inevitably fail. Our analyses prove that an average of specimens will show an increase in the proportion of “mineral matter” and of “organic matter not bitumen,” as the point from which the specimen is taken is removed from the center of the lake; yet, the exceptions to this rule are so numerous and marked, that no certainty can attach to the use of these criteria. Great value is attached by some experts on asphaltum to- the determination of the specific gravity and temperature at which the different specimens soften and flow. These tests might have some value if applied to a pure bitumen which had been proved to change in specific gravity, etc., as it underwent chemical and other changes; but no such investigation and proof in relation to any asphaltum has been attempted. Such a relation has been assumed but not proved. The fact that Trinidad pitch is a mixture so indefinite that it is almost, if not quite impossible, to select two pieces that have the same pro- portions (as is proved by Nos. 6, 7 and 8) is a sufficient reason why no such distinctions can be based upon such determinations. The average proportions of mineral matter to bitumen in the 5 samples of commercial lake pitch taken from near the center of the lake is 100:151. No two of them are exactly alike, and the extremes are, lowest 100 :148, highest 100:155. The seven specimens from outside the tramway show greater uni- formity, yet no two are alike and all are below the lowest of the five mentioned above. The average ratio of the land pitch is still a little lower, while the extremes of variation are 134 and 146. It is manifest that between these extremes of pro- portion of 100: 134 and 100: 155 a marked variation in specific gravity and flowing test must occur, as the sand is about twice as heavy as the bitumen. These extremes of variation include one-sixth of the average amount of bitumen present. 200 Peckham and Linton—Trinidad Pitch. These observations apply with equal force to the ingredient of the mixture denominated “ organic matter not bitumen.” In the five specimens of commercial lake pitch the average amount of this material is 10°651 per cent. The extremes of variation include 1:482 per cent, which is 14°8 per cent or nearly one- sixth of the average amount present. The smallest amount is found in the average pitch from near the center of the lake,. yet No. 17, which was picked up from under the feet of the men who were loading the tram cars on the west side of the tramway near the middle, contains nearly 12 per cent. This average lake pitch is found where the mass is in constant motion from escape of gas. Here there is no organic matter added by vegetation to the amount originally found in the pitch. ‘The growth of vegetation upon and in the pitch itself is the source from which the excess of organic matter found in the lake pitch taken from points outside the center and from the so-called land deposits is derived. As this excess consists. mainly of coarse roots it is removed by refining, so that when the pitch is ready for use the difference in the organic matter has largely or entirely disappeared. See numbers 2 and 19, 8 and 20. Table No. 2 shows the results of a comparative examination of the bitumen contained in the different samples without regard to the amount present. The first division of this table shows the percentage of the crude pitch dissolved only by petroleum ether, boiling spirits of turpentine and chloroform respectively. The middle column gives the percentage of the total bitumen in the ernde pitch. The next three columns give the percentages of the total bitumen present dissolved only by petroleum ether, boiling spirits of turpentine and chloroform respectively. The last column shows the percent- age of the total bitumen dissolved by boiling spirits of turpen- tine. This item is represented by adding together the items of the first and second columns, as all of the material that is dissolved by petroleum ether is soluble in boiling spirits of tur- _pentine. A comparison of these numbers along each hori- zontal line shows that there is no necessary connection between the amount of crude pitch dissolved by petroleum ether and the guality of the total bitumen. As an example, in No. 1, which is a land pitch, 100 parts of bitumen are mixed with very nearly 100 parts of sand and organic matter, not bitumen, while in No. 9 from the center of the lake 100 parts of bitu- men are mixed with about 92 parts of foreign matter. Now while the percentage of crude pitch dissolved by petroleum ether from No. 1 is 2°333 per cent less than the percentage of No. 9 dissolved by the same menstruum, the proportions of the total bitumen dissolved in the two eases are almost identical, Peckham and Linton—Trinidad Pitch. 201 viz: 66°5 per cent and 66°544 per cent. Again, No. 17 was broken off a piece of pitch as a negro raised it and threw it into atram car. It fell under his feet and was secured asa piece of convenient size for aspecimen. On analysis it gave 34:2 per cent soluble in petroleum ether, while the total bitumen was 52°997 per cent. These results give 64°531 per cent of the total bitumen soluble in petroleum ether and 89:372 per cent soluble in boiling spirits of turpentine. Of the five samples of commercial land pitch, Nos. 2, 4 and 10 contain 33°62, 33°736 and 33°730 per cent of matter soluble in petroleum ether. The average is 33°705, yet the average per cent of the total bitumen present soluble in petroleum ether is 64:283 per cent, almost identically the same as that yielded by No.17. In fact No. 2 and No. 17 represent the extremes of location from the west side of the tramway to the village lot furthest from the lake and the difference in the percentage of the total amount of bitumen soluble in petroleum ether is only (‘255 per cent) two hundred and fifty-five thousandths of one per cent—a difference wholly without significance. No. 9 was from the center of the lake and No. 1 from a village lot about 20 rods nearer the lake than No. 2, yet the difference in the amount of total bitumen present soluble only in petroleum ether, is only forty-four thou- sandths of one per cent. Arranged in a table these samples of commercial pitch appear as follows: INL cand: se so Ee enn? 66°500 No. 9. Lake (center).-.....--.- 66°544 Pee alndy 4.4) es No. 10. BNO OTA CC Capea seis vie ia 2 64:283 Nose Lipliner. 3 ea ete es 64°531 Of the specimens representing the pitch filling the annular space outside the tramway and beyond, to the boundaries of the lake, Nos. 21 and 22 were taken from the bottom of an excavation on the right of the left limb of the tramway loop as it descends upon the lake and within 140 feet of each other. Nos. 23 and 25 were taken from points very near each other on the right or west side, of the right limb of the tramway loop. These four points are on the north side of the lake and near the border. No. 16 was taken from a point on the south- east side of the lake about half way from the tramway to the border of the lake. The spot was free from grass, yet it was within the area covered with vegetation. Nos. 24 and 26 were from points on the south side of the lake directly opposite Nos. 21, 22, 23 and 25 and about 100 feet apart. 202 Peckham and Linton—Trinidad Pitch. The percentage of total bitumen soluble in petroleum ether is shown in the following table : ING) D0 eee ee ae 51°555 sn Ua icy os daoleb i 29 65°809 pales 1 Mina eff le 64°960 Bee. ee Se ae en 62°974 CONDO. 2a ee Soars 5G. SHERI | es eS 70°691 “Fa SG. Lee Ete Hy ee 66'933 Nos. 25 and 26 from opposite sides of the lake and very near the border, are the highest in the list and higher than No. 9 from the center of the lake, and higher than No. 6 which is the highest commercial lake pitch. The differences and identi- ties of these different groups, as well as between the individual members of the groups, can be readily traced by reference to table No. 2. The portion soluble in petroleum ether is called “ petrolene.” It is a constituent and essential part of the pitch, and embraces all that is most volatile in the pitch, including those etherial or oily fluids that are given off at a temperature below the boiling point of water, and which are found in all specimens from all parts of the deposit (land and lake) that have not been previously heated or decomposed. It is contended that petrolene is the cementitious portion of the pitch, because the remaining portion of the bitumen is solid and insoluble in residuum oil. It might just as well be contended that water is the cementitious principle of glue and that glue has no cement- ing properties because it is not soluble in alcohol. The fact is, that the bitumen of Trinidad pitch consists of asphaltene dis- solved in petrolene and that its cementitiousness is just as much due to one as the other. Sand cannot be cemented with either petrolene or asphaltene alone, neither can wood be cemented with either water or glue alone. The cementitious- ness of the pitch depends upon the amount and quality of the bitumen present. What meaning is intended to be given the word “dry” in connection with pitch is not very clear. It cannot be freedom from moisture, for no specimen of crude pitch is entirely free from water. As found in the deposit, the pitch both outside and inside the lake is saturated with water, and its condition after removal from the deposit depends entirely upon what is done with it. The use of the word “dry” appears to imply that the pitch from outside the lake has lost the whole or a large part of the most volatile oils originally contained in it. No proof whatever has been offered to sustain such assertions. It has been contended that the sun heats the land pitch to Peckham and Iinton—Trinidad Pitch. 203 140°-150° F. Outside the lake the deposit is covered with from 2 to 15 feet of earth, rubbish and vegetation. The large area in the center of the lake from which, for convenience, the commercial lake pitch is removed, is bare and black, exposed to the full rays of the tropical sun. Shallow pools of water on the surface of the lake appear to have a temperature of about 100° F. The pitch is probably hotter. It is therefore reasonable to suppose that if evaporation of light oils were taking place that the pitch in the lake would be “driest.” Our examination has shown us that both land and lake pitch con- tain oils volatile under the boiling point of water in about the same proportion; small in both cases. The use of both turpentine and chloroform as solvents for asphaltene is based upon observations made a year ago upon the methods employed for the technical analysis of asphaltum. Is was found that in the United States carbon disulphide has been almost exclusively used as a solvent for asphaltene, while ‘in Europe spirits of turpentine has been used for the same pur- pose. Careful experiment showed that neither of these sol- vents would dissolve all of the bitumen from the specimens in our possession, among which were those from the valley of the Rhone. It was observed that turpentine left a large and vary- ing residuum when applied to nearly all of the American specimens, including Trinidad, and that only a very small per- centage was left from the Rhone specimens and those from the Indian Territory. It was aiso found that in either case chloro- form alone effected a complete extraction of the bitumen. Later we received a specimen of Neufchatel asphaltic rock from which turpentine completely dissolved the bitumen. This led to an examination and classification of the various bitumens with reference to the action of turpentine. It was found that a large percentage of the asphaltene of Grahamite and a varying percentage of the asphaltene of Trinidad pitch and the asphaltums of California is insoluble in turpentine. It was also found that the asphaltene of the bitumens of Texas and the valley of the Rhone is almost wholly soluble in tur- pentine, and further that when the bitumen is removed from solution from these asphaltic rocks it is not a solid asphaltum but a semi-solid viscous fiuid, that does not become solid by exposure, but has remarkable stability in the atmosphere. These facts lead to the belief that the proportion of asphaltum soluble only in chloroform furnishes an indication of the extent to which a bitumen has been affected by “aging.” It has been asserted that land pitch had matured through geological time and reached a condition approaching “ glance pitch.” The word glance as applied to pitch has nothing to do with its age, or with any other property except its appear- 204 Peckham and Linton—Trinidad Pitch. ance. The word glance is from the German word “ glanz,” which means glistening. Pure asphaltum that has been melted has asmooth glistening fracture like rosin or anthracite coal. It would be impossible to produce glance pitch by melting a material containing so much mineral matter as Trin- idad pitch. Iron pitch is the nearest approach to it that can be found in the neighborhood of the Pitch lake, and that was. found both within and without the lake. If it has been intended to convey the impression that asphaltum becomes glance pitch by aging, and that the land pitch is farther on the way through geological time towards glance pitch than that in the lake, it must also be admitted that so far as anyone knows to the contrary the whole phenomenon of the pitch lake may have been produced within five hundred years. Our analyses have not furnished the slightest evidence that the bulk of the pitch outside the lake has aged any more than that within it. Our analyses also show that the bulk of the pitch is in good condition throughout the deposit; and that the effects of aging are about equally distributed. These analyses do not sustain the allegation that land pitch is any less uniform in composition than lake pitch. The fol- lowing figures represent the extremes in the percentage com- position of total bitumen in the five samples of commercial lake pitch and the five samples of commercial land pitch soluble in Land. Lake. Petroleum:ether2va2s 0s 52228 3°445% 3°601 Boiling spirits turpentine --.-- 9°7464 7°110 Chioraformeswscd ses eu BQIF 9°716 Total soluble in turpentine_-- 9°6214% 9°696 The correspondence between them is remarkable. It makes no difference whether these results of analysis are taken as a whole, or compared severally, or in the different elements that make up each analysis, the same conclusion is inevitable, viz: that the entire deposit both within and with- out the boundaries of the lake is one and the same substance and in substantially the same condition. There are five specimens in the collection that represent the rubbish of the deposit. Nos. 3 and 18 are respectively land and lake “‘iron-pitch,” Nos. 11 and 13 are decomposition prod- ucts from the land and lake respectively, that may be what has been called “chocolate pitch,” No. 12 may be what has been called “grey pitch.” None of these are commercial articles, yet they are shown by these analyses to have scientific rela- tions to the commercial pitches full of interest. By compar- ing these five specimens with any other five in the tables, it will be found that they are low in material soluble in petro- Peckham and Linton— Trinidad Pitch. 205 leum ether; high in material soluble in boiling spirits of turpentine and chloroform, and at the same time they are low in the percentage of total bitumen soluble in boiling spirits of turpentine. This apparent contradiction is easily accounted for when the high percentage of material soluble only in chloroform is observed. The proper interpretation of these results in reference to the aging of asphalt awaits the comple- tion of investigations now in progress. TABLE No. 1.—Analyses of Trinidad Pitch. Total Organic No. Petrolene. | Asphaltene. BitaimewilaNot Bitaciow Inorganic. a. \33°600 17-150 11-625 [37-625 b. |33°635 17-200 11°850 37-300 1. Mean 33-617 17-175 50°791 11-737 37-462 a. |33°640 18-888 11°347 1367123 b. |33-600 18-468 11:652 36°280 2. Mean 33°620 18-678 62:297 | 11-499 | 36-201 a. |33°637 23 621 7-840 34-902 b. 33511 23-034 | 8-988 34-467 3. Mean | 33-574 23327 56°901 8-414 | 34-684 | | a. Bas 18°764 | 10-564 36:903 b. |33°703 18-498 11:138 36°555 4, Mean | 33°36 18631 52:367 10°851 36°729 a. 33-600 '18-000 | 10-500 '37-900 b. |33-650 18-500 | 9800 38-050 6. Mean 33-625 18-250 51:875 | 10°150 | 37-975 a. |36°650 17-200 _ 10-800 35-350 b. 36°650 117-250 | 10-775 35°325 6. Mean | 36-650 17°225 53-875 10-787 | 35°337 a. |36°392 17-151 10-491 35-961 b. 36°392 17-245 | 10-410 35-990 7. Mean | 36°372 17198 53°570 | 10-450 35-975 a. \36°300 17-975 | | 9900 35°725 -b. |36 650 17-800 | 9°850 35°700 8. Mean 36°475) 17°887| 54:362 | 9°875 35°712 a. |35°950 18-050 | | 10°85 35°125 b. 135-950 18-100 | 10:690 35-260 9. Mean 35-950 18-075 54-025 10-782 35°192 a. |33°730 18-948 | 11390 35-930 b. |33°730 18-750 | 11°667 35-842 10. Mean 33°730) 18849 52°579 11°528 | 35°886 a. \21°200 30-375 | 9-600 '38°500 b. |21°525 30-250 10-100 38-450 11. Mean | 21-362 30°312) 51-674 9°850 38-475 a. | 0-000 3-000 | 43-525 53-475 b. | 0-000 3-000 | | 43°550 53°450 ee SS SS EE EE ES ee ae eee eee Peckham and Linton—Trinidad Pitch. Petrolene. 12. Mean 13. Mean 14. Mean 19°200 19°300 31°800 (31°750 32°176 '32°450 15, Mean | 16. Mean Wie Mean 18. Mean 19. Mean 20, Mean | 2c Mean 29. Mean | Dias Mean 24. Mean! 25, Mean 35°375 35°425 34°200 34'200 22°275 22°225 39°350 39°300 '38°150 38125 | '26°925 26-925 34-750 34-700 34-400 34-425 33°275 33°100 34:950 34°850 35-600 35-125 » |(B4°725 34°775 0°000 19°250 a1 TTS 32°313 35°400 34°200 22°250 39°325 38°137 26°925 34°725 34°412 33°187 34°900 35'°362 34°750 Asphaltene. Total Bitumen. 33°050 33°575 18-725 18°525 43°649 43°300 17-475 17-600 18-775 18 820 22°225 22°475 17-175 17-400 18°750 18°921 25-275 18°035 18°050 18°550 18°575 19°325 19700 16°100 16°100 14650 14°675 20°175 19°625 3°000 33°312 18°625 43-474 22°350) 18°562 19°572 16°100 14°662 19:900 3°000 52°562 50°400 75°787 52°987 52°997 44°600 56612 56°973 52°225 52°767 52°974 52°699 51°000 50°024 54°650 Organic 9°650 9°475 11°925 12°355 5°570 5°960 11°000 10°925 11-275 ~ 11:440 8°925 8°950 9225 9'100 8°100 7°923 11°375 11100 10°970 10°875 11°500 11°325 11°100 11°150 10°600 10°475 11°100 11°325 8°425 8'925 Not Bitumen. 43°537 9°562 127140 5°765 10°962 11°357 8:937 9162 8011 10°922 11°412 11125 10°537 11°212 8675 Inorganic. 38:000 37°750 37°550 37°370 18°605 18°290 36°150 36°050 35°750 35°540 46°575 46°350 34°250 34°200 35°000 35°031 36°375 36°700 36°245 36°375 35°550 35°675 36°300 36°050 38°350 38°575 38°650 38°875 36°675 36675 53°462 37875. 37460 18-447 36'100 35°645. 46°462 34:225 35°015 36°537 36°310 35612 36°175 38°462 38°762 36°675 Peckham and Linton—Trinidad Pitch. 207 Percentage of Crude Pitch TABLE No, 2.—Analyses of Trinidad Pitch. only soluble in No. |Petroleum Boiling Chloro-| Spts. Tur- form. Ether. 1é 33°617 2. 33°620 3. 33°574 4. 33°736 5. 33°625 PEG: 36°650 (A B18) B3 (74 8. 36°475 9. 35°950 10. 33°730 He 21°362 iV Sy 19°250 14, S105 15. 32°313 16. | 35°400 ie 34°200 18. 22°250 19. 39°325 20. 38°137 ile 26°925 22. 34°725 ae 34°412 24, SBS i) 25. 34:900 26,94) SoeaG2 oi: 34°750 pentine. 10°887 | 10°494 | 13-802 10511 | 15°575 12°800 11°683 | 15°750 12-310 15°670 15°200 1°700 20°487 11°875 21°249 12°300 11575 SIS 35) 11°285 12°538 18°612 P35 13°100 14:237 9°200 9°862 10°712 6°287 8183 9°819 8°120 2°675 4-425 5514 2°137 5°762 S179 15°112 1300 | 12°825 6°750 | 22-295 5237 7:222 | 12°615 | 6002 6:297 6°687 4-867 5°462 5275 6°900 4°800 9:187 ] | | Percentage of Total Bitumen only soluble in Total |Petroleum Boiling Chloro- Total ‘Bitumen.| Ether. Spts.Tur- form. in pentine. ©: Pur: 50-791 | 66-500 | 21-250 | 19-250 |87-750 52°297 | 64:°276 | 20°066 | 15°658 (84:342 [eot-195-9|-58002, \o 24-131 9) 17-16% |82°833 52°367 64°422 | 20°071 | 15°506 84°493 51875 | 64°819 | 307022 | 5:159 |94°841 | 53°875 68°132 | 23-700 8'168 (91°832 | 53°570 67°896 | 21°862 | 10°242 |89-758 54°362 67:096 | 28°972 3°932 96°068 54:022 | 66°544 | 22-788 10°668 (89-332 | 52579 | G4:151 | 29°812 6°036 93°963 51°674 | 41°340 | 29°415 | 29°245 (70°755 3°000 Alter ation Pro- duct 52°562 | 36°621 | 38°976 | 24-403 |75°597 50-400 | 63°055 | 23°547 13°398 (86°602 T5187 | 42°636 | 28°038 | 29°326 |70°674 52-937 | 66935 | 23°162 | 9-903 |90°097 | 52°997 | 64531 | 21°841 | 13648 86372 | 44-600 | 49°921 | 21°837 | 28252 |71-758 | 96°612 69-465 | 19°933 | 10°602 (89°398 | 56°973 | 66959 | 22°016 | 11°025 |88°975 no2- 224) |) 5'-b55 3526405) 12°805° (87-195 | 52-767 | 65°809 | 24°968 9223 190-77 527974 | 64°960 | 24-720 | 10°320 89:°680 52°699 | 62°974 | 27:016 | 10°010. 89°990 51-000 | 68-431 | 18°039 | 13°530 /|86:470 50°024 | 70°691 19-714 | 9°595 (90°405 54°650 | 63°526 4 University of Michigan, Ann Arbor, Michigan, November 30, 1895. 19°592 | 16°882 (83°118 208 EL. A. de Schweinite—Meteorite from North Carolona. Art. XXI.—A Meteorite from Forsyth Vo., North Carolina ; by E. A. DE ScuweiniTz, M.D., Pu.D. AxBouT three years ago there was ploughed up on a farm in the southwestern portion of Forsyth Co., a mass about 50 Ibs. in weight which, upon cursory examination, appeared to be pure iron, probably of meteoric origin. The mass had an irregular wedge-like shape and was covered with a thin scale of oxide of iron. The metal beneath was exceedingly tough, could be cut with the greatest difficulty, and fragments ob- tained by means of chipping with a cold chisel showed a erys- talline structure. A polished surface etched with nitric acid did not show the Widmanstitten figures distinctly, but a mot- tled crystalline. structure. The appearance of the meteorite when found can be seen from the accompanying figures, 1 to 4 (one-fourth natural size). Subsequently, two slices were cut from the meteorite and etched respectively with nitric and hydrochloric acids, without producing any very characteristic markings. A preliminary analysis made from a piece chipped off the end of the meteorite by means of a cold chisel gave the follow- iug results : EAS pe A a Papa pees 94°90 per cent. S Le te Feat. eae oe aes oe 99 6¢ ONS hee ee 2 taeda Ar1g «3 COtet Sie eee eee oie a Pe een ee hoe Be ee trace From this it would seem to be closely allied to the Guilford County meteorite, possibly a chip of the same find. This iron is now in the possession of Mr. G. F. Kunz of New York. Biochemic Laboratory, Columbian University, Washington, D. C., Novémber 30, 1895. 209 na. te from North Carol (MZ tz— Meteo ene A. de Schwet 5 e E et 14 Am. Jour. Sc1.—FourtH Series, Vou. I, No 3 —Marcu, 1896. 210 Adams and Harrington—Alkali Hornblende and Art. XXII.—On a new Alkali Hornblende and a titaniferous Andradite from the Nepheline-Syenite of Dungannon, Hastings County, Ontario; by FRANK D. Apams and B. J. Harrineron, McGill College, Montreal. IN a paper which appeared in the number of this Journal for July, 1894, the discovery of a large area of nepheline- syenite in the township of Dungannon, in the Province of Ontario, was announced and the geological relations and min- eralogical characters of the mass briefly described. One of the many peculiarities of this rock is the absence from it of the mineral pyroxene, which is usually the chief iron-magnesia constituent in rocks of this class, its place being taken by hornblende and mica, but even these minerals are present in comparatively small amount. Of the hornblende two varieties, occurring in different parts of the mass, were distinguished. The first, from near the York river, has a large axial angle with strong pleochroism in tints varying from pale yellow to deep green, and although containing a considerable amount of soda, probably approaches common green hornblende in composition. The second variety, which occurs in a series of exposures about two miles to the east of the village of Ban- croft, is quite different in character, having a small axial angle with high extinction and a much stronger pleochroism in the bluish tints suggestive of arfvedsonite. A number of additional thin sections have been prepared and in the present paper the results of a further investigation of the optical properties and chemical composition of this second variety of hornblende are presented. Hornblende—The mineral occurs in hypidiomorphic grains, which show the usual hornblende cleavages; it is optically negative, a being the acute bisectrix, but the double refraction is weak. It possesses, as has been mentioned, a strong pleochroism as follows: a = yellowish green. 6 and c= deep bluish green. The absorption isc =6>a. bandc,if not quite equal in absorption, are nearly so, hence sections cut at right angles to the acute bisectrix show but little pleochroism and are nearly isotropic. c lies nearest the vertical axis, but whether toward the acute angle # or on the opposite side cannot be determined as the mineral does not possess a good crystalline form; it makes with the vertical axis a large angle the extinction amounting to 30°. The plane of the optic axes is the clinopina- Titaniferous Andradite from Ontario. 211 eoid, and there is a strong dispersion—red greater than violet. What drew especial attention to this hornblende in the first instance was the fact that it appeared to be nearly uniaxial. When a section, cut at right angles to the acute bisectrix, is examined between crossed nicols in convergent light, a black cross is seen somewhat thickened toward the intersection of the arms. This cross, on revolving the stage, divides into two hyperbolas, but these separate from one another but very little, and appear to separate less than they really do, on account of the fact that the low double refraction and deep color of these sections causes the hyperbolas to be ill-defined, while the whole field is very dark. The dispersion, however, makes itself evident in the varying colors on the sides of the hyper- bolas. When, however, a gypsum plate giving a red of the first order is inserted above the objective the hyperbolas become a little better defined, although still not sufficiently definite to allow the axial angle to be accurately measured. The axial angle is found to be over 30°, possibly as much as 45°, which, however, is still very sraall for hornblende, being about one-half the usual value. Our thanks are due to Pro- fessor Rosenbusch for his assistance in working out these optical relations. On examining a large series of thin sections of nepheline- syenites representing most of the important occurrences hith- erto discovered, only two rocks were found which contain a hornblende at all similar to that above described. The first of these is the nepheline-syenite from the Corporation Quarry at Montreal, in which hornblende with the same small axial angle, low double refraction, intense color and pleochroism, large extinction angle and high specific gravity, occurs intergrown with the augite. The second is the hornblende described by Hackman under the name of arfvedsonite and which occurs intergrown with aegerine in the nepheline-syenite from Umptek in the Kola peninsula.* This mineral, however, differs from typical arfvedsonite in having an extinction of about 40° as well as in several other important respects. It possesses more- over a very small axial angle, although this fact is not noted by Hackman, while in true arfvedsonite the axial angle is very large. This Kola hornblende is much lighter in color than the hornblende from either of the above mentioned Canadian localities. In order to determine the chemical composition of this somewhat remarkable variety of hornblende from the Dungan- non rock, it was decided to separate a portion for analysis. A considerable quantity of the rock was accordingly reduced to * “ Petrographische Beschreibung des Nephelinsyenites vom Umptek,” von Victor Hackman. Kuopio, 1894, p. 14. 212 Adams and Harrington— Alkali Hornblende and powder and passed through a sieve of 43 meshes to the inch— the rock being rather coarse in grain—and after having been freed from dust was treated with Thoulet’s solution, having a specific gravity of 3°13, in a large separating funnel. In this way an almost complete separation of the colored constituents was effected. These latter, which sank in Thoulet’s solution, were subjected to the action of a bar magnet and then treated with dilute hydrochloric acid, and various impurities thus removed. The purified powder was then treated first with Klein’s solution, having a specific gravity of 3°22, and then with methylene iodide, having a specific gravity of 3°328. In both fluids practically everything sank, only a few composite grains floating. A microscopic examination showed the pow- der now to consist of grains of hornblende and of garnet with some composite grains consisting partly of nepheline. Fur- ther separation became difficult since, as was subsequently ascertained, the hornblende had a specific gravity of 3°433, and the specific gravity of the garnet was 3°739, while many composite grains consisting of garnet and nepheline had a specific gravity practically identical with that of the horn- blende. As the electro-magnet was found to be useless, both minerals being readily attracted by it, Retger’s silver nitrate method was employed.* The silver nitrate was fused in a properly arranged test tube, and after the introduction of the powder, potassium nitrate in powder was gradually added to the fused mass until the garnet fell, the whole being frequently stirred and maintained at a temperature of from 200° to 240° C. On allowing the mass to solidify, a portion of the powder was found to have collected at the top of the mass, while the rest was at the bottom, the intervening part being quite free from mineral grains. The solid mass was then cut in two and the salts dissolved by treatment with water. After three suc- cessive separations the hornblende was obtained quite free from grains of garnet—the only impurities present being some composite grains consisting of garnet and nepheline. This powder was then placed under a lens and all the composite grains picked out by means of a fine needle. In this way a quantity of pure hornblende sufficient for purposes of analysis was obtained, while the garnet was obtained directly in a state of purity without the necessity of a final separation by hand. Both minerals were found to be quite fresh and bright and quite unacted upon by the fused salts. The hornblendet was then analyzed by Dr. Harrington with the following results: *“Teber Schwere Flussigkeiten zur Trennung von Mineralien.” Neues Jahrbuch fiir Mineralogie, etc., 1889, ii, p. 190. + We would suggest Hastingsite as a varietal name for this hornblende, con- necting it with the region where it occurs. Titaniferous Andradite from Ontario. 218 Silicate Can sepa tees aac et i BAIL BS itaniimMraioxide nie a2. ie aoe ty OM, SENT TO ee eA a ee Sct oe ORE CCO KIC Cpe see ee cee 12°621 IMERFOUS] OXIGCGs eo a pay ose 21°979 Manganous oxide.-.------ pers ba7 8 629 ret. ey ee Syst ea ts ie ea bre cg 9°867 Maonesiake: ee. Sesrsee ewe hos TZOtAC pusm meee taal Palas hers he oe 2-286 Sere eee a WS Pepe a is SE he ey es ane ae 3°290 WN ontieic se eee eats cnet ees 348 99°601 DPeCile CraNvlby 2. 4.22 ee ones 3°433 The atomic and quantivalent ratios deducible from the above analysis are as follows: Atomic. Quantivalent. 1 — 929 aes Ov apes | 2386 2356 i Se eee 226X3= 678 at Be isa eae yr ae Hee a 305K 2= oe : Win oe hh 9x2= 3 | one COA es ay Gein : 2 NiGS=..-=- Behe 4°10 aia sper 410 Stenzelberg ---- - Sedaats 4:02 3. 9 4°38 Dungannon .----- Srhies 3°84 Sees ead Scharizer adopts the foregoing ratios (8: 1:38 and 3:4) as those of syntagmatite in calculating the composition of horn- Ie ek Ill blendes intermediate between (R,R), R,Si,O,, and actinolite. He assumes in the first place that all the alumina and ferric oxide belong to the syntagmatite molecule (=). The sum of the Al,O, and Fe,O, molecules (from the molecular ratio) mul- tipled by three, gives (SiO,)= on the one hand and (R,O+ RO)s on the other. The sum of (R,O+RO)=z divided in the proportion of 3:4 gives (R,O+CaO)s and MgO+FeO)s. Subtracting (MgO+ FeO): from the sum of the correspondin molecules deduced from the analysis gives (MgO+ FeO),s—that is the number of molecules of magnesia and ferrous oxide belonging to the actinolite molecule (A)—and (MgO+FeO)a divided by three (see actinolite formula) gives the lime mole- cules of the actinolite (CaO)s. This value subtracted from the total number of lime molecules gives (CaO)s, and (CaQ)z subtracted from (R,O+CaO}s gives the alkali molecules (in some cases including H,O). Finally (MgO+CaQ), gives (SiO,)a. These statements will be made clearer by the follow- ing example, one of those selected by Scharizer. HORNBLENDE FROM EDENVILLE, ANALYZED BY RAMMELSBERG. | Molee. BR. | Original Original deduced Syntag- | Calculated analysis analysis. (i trom | - matite. Actinolite. | composition. | calc. to analysis. | 100. Si@eeee = 5)c6 (1) S61) )) 1222 609 51°97 52°66 Oe 6 5°75 56 56 4 we | 5°99 5°86 Bet 2"! 2°86 18 18 see | 300 2°91 MeO™ 2. 23°37 584 127} 457 24°35 | 23°82 Water. 19°49) 999. | 76 | 1B a ENS 12°66 Na Or 2? 0°75 12 1D 8 See O07 8 wal Oras FO Oee 292)" 0784 | 9 9 | ane | 0°88 9°86 PLO 327 '0°46 25 4 | oe 0°07 0°47 98°12, | | 10000 100-00 216 Adams and Harrington— Alkali Hornblende and Here (SiO,)= == 3(56 + 18) se (R,O + RO)s = 3(56+18) = 222 (R,O +CaO) = = ee — = =='95 (MgO): = ee ise — = 127 (MgO) « = 584—(MgO) > = 584—127 = 457 45 (CaO) See = fe = 152 (Ca0j}z = 222—(CaO)a = 999-152 = 70 (Na,0 +K,0+H,0): =(R,0+0a0)s —(Ca0)s=95 —70=25 But (Na,O+K,O)s = 1249 =21 “(ELO) Ss 74 Finally (SiO,), = (MgO +CaO) a= 4574152 = 609 Having thus deduced the molecular ratios of the syntagma- tite and actinolite, the numbers for each constituent are multi- plied by the corresponding molecular weights, in order to obtain the theoretical relative weights of the constituents of the mixed hornblende. Syntagmatite. Actinolite. 22/2 C60) = 113320 609 x60 = 36540 56 10256] Va745 . ae Le byes LESUGO: == 2880 apes 8 ea eo! 12 aA ee D080 45740 = 18280 1036 56. =F 3920 152 X56 = Sate 1x26 205 744 peas 2 Naa 9x 94 = 846 eters eee eae Pal Be 2 sapips eee 32607 63332 Then, (32607 + 63332): (13320 4- 36540): 31008 x and # = 51:97 = p. ¢. of SiO, in the mixed hornblende. And in like manner the percentages of the other constituents are calculated. | But 32607 ; 63332 practically as 1:2, and therefore the formula of the Edenville hornblende might be regarded as Lee R.B,Si,0,, + 2(Mg,CaSi,0,,) or as Scharizer gives it IIL 10(R,R,8i,0,,) + 20(Mg,CaSi,0,,) Titaniferous Andradite from Ontario. 217 The analyses selected by Scharizer agree remarkably well with his theory, but there are aluminous hornblendes whose constitution cannot be readily explained in this way and which at the same time cannot be referred to the pargasite orthosili- cate.* Garnet.—In the hand specimens the garnet is seen to possess a deep reddish-brown color. In the thin sections it is a paler brown although still deeply colored. Itis not found in all parts of the mass and where it does occur is usually present only in small amount. It possesses the usual high index of refraction and is quite isotropic, occurring usually in irregular shaped grains but in some few cases showing distinct crystalline form. It frequently holds a few large inclusions which usually con- sist of calcite in single individuals, although the garnet is per- fectly fresh and the calcite shows no distinct evidence of a secondary origin. It moreover sometimes holds inclusions of the hornblende above described, of pyrite, iron ore and even of nepheline. A garnet resembling this occurs in small amount associated with a similar hornblende, as above mentioned, in the nepheline-syenite of the Corporation Quarry at Montreal, and it also contains as inclusions most of the other constituents of the rock. The same is also true of the melanite in the nepheline-syenite of Alné.t Before analysis the garnet was purified by several separations with fused silver nitrate and on careful examination with the microscope the grains appeared to be entirely free from foreign matter. With the pycnometer their specific gravity at 16° C. was found to be 3°739. Chemical analysis gave the following results : SUL Oo Wir apart cash a ns ph Dale apr en eS 36'604 itaninmudioxtdess ec" 23 Fe 1078 ANI RET ING Wee pe wig Rr oe a ce 9°771 MerricrOxides eset ine irs, Se 15°996 METNOUCROXIMe, ae a2 er wees fat oP a S289 Manganous oxidé<2-925../--—- 2425 1°301 Weim eoe. Beet ei as i ee 29°306 e INIMOMCSIA se feet oe oe ee el BOF oss ome nition - 21 etek Se "285 99°577 The atomic and quantivalent ratios deduced from the above analysis are as follows: * See Scharizer’s paper, loc. cit., p. 156. + ‘‘ Ueber das Nephelinsyenitgebiet auf der Insel Alnd,” von A. G. Hégbom. Geol. Foren. i. Stockholm Forh., 1895, p. 144. 218 Adams and Harrington—Alkalit Hornblende, ete. SUES Ms Die Bet -<2e re ee ae oe er eer ee = Atomic. 610x« 4 Lo x4 2440 52 Quantivalent. 2492 2492 The ratio for RO: R,O,: (SiTi)O, is 629 : 196 ; 628, or, calculating the titanium as Ti,0,, 629 : 203; 610=3:13 3. The analysis therefore accords well with the ordinary garnet Thy Ler formula 3RO, R,O,, 38810, or R,R,Si,O,,, and the mineral may be regarded as a titaniferous andradite, with a considerable proportion of the ferric oxide replaced by alumina. position it resembles somewhat the brown garnet from the Island of Stok6, analyzed by Lindstrém.* By way of comparison the analysis of the Stok6 garnet and also one of a garnet from the nepheline-syenite of the Island of Alnot are included in the following table. In com- Stok6. Molec. R. Alno. Molec. R. Dungannon. Molec. R. Si0,_....1 36:63" 610" 610 |\/ 31-15) 519 36-604 610 Oral eae i Se 6°73 a 603 Sianie a 2 iO, Pena t9o1 oe 314 31 | onl 96 Fe,0, --. 13°45 af N82 | 93-83 pat 180 | 15-996 eer a BOE a 228 32) ce ap ela 3-852 53 MnO 2. ea 0 58 8 | 1301 18 Gi0_. |... 35:90 641 | 33-44 5975616 | 29:306 [523 | MEO Sp CDRTe MgC Pe Cea 1384 Be pe Na,O--- aa 7 | | 68 Ls) Socc Soc | Teme = 16 93) | 285 16 J | 99°30 | 99°55 99-577 * Zeitschr. fur Kryst. u. Min., xvi, 160, 1890. +Sahlbom, in the paper by Hégbom already cited. R. Bell—fising of the Land around Hudson Bay. 219 ART. XXIIL.— Proofs of the Rising of the Land around Hudson Buy; by Roperr BE tt, of the Geological Survey of Canada. [Read before the Geological Society of America, Philadelphia, 27th December, 1895. Abstract. | In the provinces of Ontario and Quebec, it has been found from actual levellings by Gilbert, Spencer and Upham, that the old shore-lines are not perfectly horizontal, but that they slope upward in a northeasterly direction at rates varying in different regions from a few inches to a foot and even two feet per mile. If this upward slope were continued in the same direction to the northeastern extremity of Labrador, 1300 miles from Lake Huron, the increase in the elevation might there amount to 1000 or 2000 feet. It is scarcely probable that the differential elevation is constant and regular for such a great distance. Still, it is a fact that well preserved shore- lines are to be seen at great heights in the northern parts of Labrador. In my Geological Survey Report for 1884, I have mentioned ancient beaches at Nachvak, 140 miles south of Hudson Strait, which have an estimated altitude of 1500 feet above the sea. The two sides of Hudson Bay present very different phys- ical characters. The eastern is formed mostly of crystalline rocks and, as a rule, is more or less elevated, with a broken sur- face sloping somewhat rapidly westward or towards the bay ; while the western side is mostly very low and much of it is underlaid by nearly horizontal Silurian and Devonian strata. These low shores are accompanied by shallow water extending far to seaward. The head of James Bay, which forms the southern prolongation of Hudson Bay, is extremely shallow, but the various rivers which flow into it have cut channels through the soft shallows, and by means of these the land may be approached with sea-going vessels. The whole of Hudson Bay may be said to be shallow in proportion to its great area, as the soundings show that it does not average more than 70 fathoms in depth. The shores of the bay everywhere afford abundant evidence that there has been a comparatively rapid rise in the land and that the elevation is still going on. I have mentioned numer- ous proofs of this in my various official reports on the geology of these regions from 1875 to 1886, and I shall now recall a few of those and give fresh ones in addition, some of which came to my knowledge on a journey to the bay during the past summer. It is well known to those who have paid any atten- tion to the subject that since the establishment of the posts of 220 LR. Bell— Rising of the Land around Hudson Bay. the Hudson’s Bay Company in the mouths of the rivers around the bay, 200 years ago, there has been an ever-increasing diffi- culty in reaching these establishments from the sea. On the eastern side the most striking evidence of the rising of the land is afforded by the numerous well-preserved and conspicuous terraces cut in the till and other deposits. Near the sea these may be seen at various heights, up to about 300 feet, but above this elevation the scarcity of soft material out of which terraces might be excavated, renders this kind of evi- dence less apparent than it might otherwise be, at higher levels. On this side of the bay, one of the best evidences that the elevation of the land is still going on is furnished by the long lines of driftwood which one sees in many places far above the reach of the highest tides. The old beaches, on which this wood is plainly seen, occur at various levels up to about thirty feet above high tide, but the remains of rotten wood may be detected in some localities up to nearly fifty feet, above which it has disappeared from the ancient shores by long exposure to the weather. This driftwood consists principally of spruce, but a little white cedar and other kinds, which have been brought down by the rivers, are also mixed with it. The bark having been worn off by the action of the waves while the trunks were still fresh, has tended to their preservation. Owing principally to the salt water and the cold climate, wood endures for an incredibly long time in exposed situations in this region wherever it has an opportunity of drying quickly after rain. Some of the wood which may still be seen upon the higher levels may be upwards of 600 years old. It has been suggested that all this driftwood along hundreds of miles of coast may have been thrown up by some extraordi- narily high tide. But there are many reasons why this is quite wnlikely. It seems impossible that any modern tide could rise to such a great height and deposit so much wood at different levels all at once and in such even lines, following all the sinu- osities of more than one of the raised beaches. The supposi- titious extraordinary tide would necessarily be of brief duration and would be accompanied by a tremendous gale blowing upon the coast. This would have the effect of throwing the wood in confused heaps and only into situations favorable for catch- ing it, such as angles of the shore. But instead of this, we find it at different levels laid longitudinally all along, as if accumulated by slow degrees with moderate winds from every quarter. The fact that the wood is freshest along the lower lines and becomes progressively more and more decayed as we ascend, and that finally only traces remain on the higher levels, shows that it must have been stranded from time to time as the land was rising above the sea, and we are forced to adopt this obvious view of the case. PR. Bell— Rising of the Land around Hudson Bay. 221 In support of the paroxysmal tide theory, it is related that once during a northern gale the tide was forced as high as the front gate in the palisaded enclosure at Rupert House near the head of James Bay, and it is added that this would be equiva- lent toa height of about thirty feet. When at Rupert House last summer, I could hear no authentic account of such an extraordinary rise in the water and besides the gate referred to did not appear to be more than fifteen feet above the sea-level. But even if such a great rise in the water had once occurred at this place, it would prove nothing in regard to the raised beaches on the long straight shore out on the open sea. Hud- son Bay is about 1000 miles long and its outline is funnel- shaped, with James Bay representing the contracted extremity. Rupert House is situated near the end of this narrow continu- ation, so that just here we should expect very high water with a spring tide and northern gales driving the sea in from the broad expanse outside and heaping it up at the extremity of the constantly narrowing termination. The gravel terraces seen at various elevations around the coves and upon the thousands of small islands along the east coast of James Bay are remarkably sharp and well-preserved and almost as fresh-looking as if they had been formed but yes- terday. They are generally bare of trees or bushes and the yet smooth surface-pebbles are only partially covered by lichens. Similar terraces may be seen farther north on this coast and in Hudson Strait, wherever material exists out of which they may be formed. On Marble Island the raised beaches are very plainly visible on account of the whiteness of their smooth, quartzite shingle. On the west side of Hudson Bay the land is generally too low to admit of the relatively higher sea-levels of former times having been recorded in the shape of terraces near the present shore line, but if we go back into the woods we shall find unmistakable evidence of the existence of such higher levels at comparatively recent periods. These consist of long, low ridges of drifted materials, such as we see in a fresher state at the present high tide mark. They are made up of driftwood and other vegetable debris in a completely decayed condition, covered by moss and having trees and shrubs growing upon them. In some places we may still trace the forms of the larger trunks which had been cast ashore by the waves at high tide. Between these ridges and the present shore there is a thick growth of the coniferous forest and the ground is carpeted with moss, over which the tide has never passed. Examples of these low ridges may be seen near the head of tide-water at the mouth of Nelson River, at Attawapishkat River and in places between the latter and Albany River. 222 Lf. Bell—Lising of the Land around Hudson Bay. To the west and southwest of James Bay the till, covering the nearly flat Silurian and Devonian rocks, is generally over- spread by stratitied clays. Marine shells are found in these up to an elevation of 400 to 500 feet, but on the eastern side of the bay no fossils have yet been detected at such high levels, owing perhaps to the scarcity there of marine deposits and to the fact that but little search has yet been made for them. In the sandy deposits among the hills about twenty miles south of Cape Wolstenholme, I saw abundance of Saxicava rugosa and Tellina Grenlandica with smaller numbers of a few other species, at heights varying from the sea level up to about 200 feet; and last summer I found brackish water varieties of a number of the commoner species of our northern marine shells up to 70 feet above the sea in the clay banks along the lower portion of the Noddawai River. Around the head of James Bay and up its western side the encroachment of the outer lines of the forest upon the wide alluvial flats which extend all along these shores and are con- stantly broadening towards the sea is good evidence that a ris- ing of the land is now going on. The existing condition in this part of the bay is well described by Mr. A. P. Low in speaking of Agoomski Island. On page 24 J. Geol. Survey Report for 1887, he says: ‘“The island closely resembles the adjoining mainland in phys- ical character, being very low and swampy. The shore-line above high-water mark is made up of muddy flats covered in part with grasses and sedges, followed further inland by thick growths of small willows, these in turn giving place to small black spruce and tamarac as slightly higher ground is reached. The line of these trees is often over two miles inland from high-water mark, itself a long distance from the sea at low water.” No living mollusks are to be found in James Bay except perhaps in the northern part, owing probably to the muddy and brackish nature of the water, but abundance of the dead shells of a considerable number of kinds are washed out of the clays forming the present shores. Some of these belong to moderately deep-water species and are well-preserved, retain- ing the epidermis. This, of course, shows a recent elevation of the sea bottom. 7 Richmond Gulf on the eastern side is separated from the main bay by a high bar of stratified rocks, which strike with its length and dip westward or towards the open sea. This bar is cut through by several gaps, all resembling one another, except in their heights above the sea, and all bearing evidence of their having been well worn channels of communication at more or less remote times according to the greater or less eleva- R. Bell— Rising of the Land around Hudson Bay. 228 tion of their beds above the sea. Only one narrow passage now remains open or low enough to admit the water, but two others are as yet only slightly raised above the tides. Some of the aboriginal geographical names around the head of James Bay are significant of considerable changes in the topography since these shores became inhabited by the natives who still occupy them. The large peninsula between Hannah and Rupert bays is called Ministik-oo-watum, which means wooded island with a cove or hole in it, ministik being the Cree for a wooded island and watum for a cove or hole. The heads of the channels, which now run in behind the present peninsula from the opposite sides, are separated by a strip of low ground some ten miles long covered by bushes. Midwa across this strip, the elevation is estimated to be about fifteen feet above high tide. The most prominent point on the coast between Moose Factory and Fort Albany is now called “ Cock- ispenny” by the whites, but the Cree name is Ka-ka-ki-sip- pin-a-wayo Minis, or Island where the Crow-duck (Cormorant) lays eggs. Since this island became connected with the main- land, bushes have taken the place of the grasses and sedges which first grew upon the low ground between them, and the former are constantly acquiring a stronger growth. Man years ago the winter trail of the coast passed over the neck of this peninsula, but now it has become necessary to go outside of it, because the bushes have grown so large that they catch the snow which, in such situations, remains too soft for dog teams and snow-shoers. The salt marshes along the west coast of James Bay and also in the vicinity of York Factory, which used to attract vast numbers of wild geese and ducks, have been gradually drying up, much to the inconvenience of the Hudson’s Bay Company’s people, who depended largely upon them for food. The character of the lower portions of such rivers as the Moose, Albany and Attawapishkat shows a recession of the sea. This is particularly observable in the lower thirty miles of the Moose, where very long and narrow or ribbon-like islands run parallel to one another for many miles. The process of their formation appears to have been a constant drawing out of their lower extremities as the sea receded from them, just as the lowest islands of the present day are growing. On the east-main coast, where the land is comparatively high, the grade of the rivers is rapid as they approach the bay, and in some of them, as the Nastapoka and the Langlands, there are perpendicular falls of about 100 feet almost directly into the sea. This condition indicates recent elevation. One of the best evidences of the modern rising of the land is to be found in the beach-dwellings of the Eskimos, which 224 hk. Bell—fising of the Land around Hudson Bay. may be seen at all elevations up to about 70 feet. In summer these people generally camp on the shore, and their favorite locations are at the mouths of small streams into which the sea trout run at high tide. Here they construct weirs of stones, which impound the fish when the tide retires. On Outer Digges Island, I have found these fish traps and the rings of stones and other structures marking their old camp- ing places, up to a height estimated at 70 feet. Among the historical evidences bearing upon this question since the advent of the white man, may be mentioned the fact that in 1610, Henry Hudson, the navigator, wintered in a bay full of islands on the east coast south of latitude 53°. None of the bays in this region would now be possible for this purpose, showing that a considerable change in the level of the sea has taken place in less than 300 years. In 1674, Charles Bayley, then local governor for the Hud- son’s Bay Company, sailed through in a sloop between Agoomski Island and the main west shore of James Bay. It would now be impossible to pass here in a sea-going vessel of any kind. In 1886 I found it difficult to get through in bark canoes, drawing only a few inches of water. The shoaling is not due to a silting up, since the almost dry bottom consists of a level surface of till with bowlders scattered thickly over it. From 1675 to 1685 the Hudson’s Bay Company’s establish- ment in the mouth of Moose River was upon Hayes’ Island, which, it is to be presumed, was selected for convenience of landing goods from their vessels and shipping out their returns. This island is now unapproachable except by canoes and small boats. For more than 200 years the factory* has stood upon Moose Island, the next below Hayes’ Island. ‘The annual ship from England anchors in the channel cut through the sands off the mouth of Moose River. On account of the risk of rough water, it is necessary to discharge the cargo by schooners. Witbin the memory of living men, these schooners could ascend to a wharf built opposite the large storehouse of the factory. But for many years, the same schooners have been unable to ascend all the way, and the cargo requires to be trans- ferred into scows, which complete the trip to the wharf; and the distance to which the schooners can ascend is constantly diminishing. In the beginning of the present century Prin- cess Island, a narrow bushy strip immediately in front of the factory, was separated by a channel with a good depth of water at the lowest tides. Last antumn I saw it quite dry on several occasions during ebb tide. It is well known to every one who has lived at this post in the present generation that every now and then a new “lump” will appear in the bed of the river * Factory, a residence of a factor or agent. R. Bell—fiising of the Land around Hudson Bay. 225 and become permanent, growing higher and higher, eventually escaping submergence at most tides and at length becoming covered with grass and then with bushes. Some islands which were covered only with bushes forty or fifty years ago, now ‘support a growth of young trees. The small one on the west side of Middleboro’, below Moose Island, is an example of this and the appearance of the trees upon it is within the memory of Mr. Broughton, the gentleman now in charge of Moose Factory. Middleton Island, between the mouths of Rupert and Noddawai rivers, lies close to the east shore of Rupert Bay. Up to a few years ago, canoes and boats could pass at high tide through the long narrow grassy channel behind this island, but last autumn I found it impossible to do so with my canoes and we were obliged, at great inconvenience, to go round outside. Two hundred years ago, the ships of the Hudson’s Bay Company appear to have had no difficulty in entering the mouths of various rivers on the Eastmain coast which cannot now be used as harbors. In old times the principal post of the company on that coast wasinthe mouth of Eastmain River, which had no doubt been chosen because it afforded a good harbor. It is only a few years since the mouth of Little Whale River, several hundred miles farther north, had to be aban- doned as a harbor on account of the increasing shallowness of the water. At York Factory there is a “ship hole” in the channel of Hayes’ River, directly in front of the storehouse. The sea- going vessels of light draft employed in the Hudson’s Bay Company’s trade have been accustomed to anchor in this hole and formerly they remained afloat at all stages of the tide, but of late years, vessels drawing even less than those of former times have begun to “take the ground” at low water. In objection to the belief that the land is rising it may be said this may be due to a silting up of the hole, but on examining the material brought up on the flukes of the anchors, I found it to consist of light colored stiff bowlder clay or till. In 1782, after the French Admiral Lepeyrouse had destroyed Fort Prince of Wales at the mouth of Churchill River, he landed with cannons on the southeast side of Nelson River and hauling them across the point between it and Hayes’ River, captured York Factory. Two ships belonging to the Hudson’s Bay Company which were then lying in Hayes’ River, laden with valuable cargoes, escaped under cover of the darkness of the following*night and got safely to England. At the present time, it is only possible for a sea-going vessel to get out from this river at the top of high water with favorable wind and careful piloting in daylight. To say nothing of the difficulty Am. Jour. Scl.—Fourts Series, Vou. I, No. 3.—Marou, 1896. 15 226 L. Bell—fising of the Land around Hudson Bay. caused by the darkness, it is unlikely that all the other condi- tions now necessary to enable a vessel to leave the river, con- spired to aid the escape of these ships. It is much more reasonable to believe that the water was deeper then than it is now. The landing of Lepeyrouse with his guns on the shore of Nelson River abreast of York Factory was a feat the like of which could not be accomplished at the present day, owing to the extreme shallowness of the water. The present Fort Churchill or “ New-Fort,” as it is still called, was built in 1782 on the west side of the river about four miles and a half above Fort Prince of Wales as soon as the French had retired after destroying the latter establish- ment. The residents now suffer much inconvenience on account of the continued shoaling of the water and they have been obliged to lengthen out their “launch” or long landing tressel from time to time in order to be able to reach the outer end of it with their coast boats. Off the western side of the lagoon within the mouth of Churchill River is Sloop’s Cove, a small elliptical pond con- necting with the lagoon by a very narrow entrance, through which the water barely passes at high tide. On the arkose rocks beside this little cove many inscriptions have been cut and some ring-bolts have been fastened, for mooring vessels, all of which indicate that the cove was used for wintering ships in old times. Indeed it is known that the “ Furnace” and the “Discovery,” two small ships commanded by Captain Middle- ton, passed the winter of 1741-42 in this cove. I have exam- ined the place on various occasions and have copied most of the sketches and inscriptions on the rocks, and it always appeared to me that the conditions which we observe indicate a rise in the land since the last ship wintered there. At the present time, the tide does not rise high enough to allow of the passage into it of crafts larger than ordinary row-boats. No sea-going vessel could now enter it, which would indicate an elevation nearly equal to the draft of the ships formerly fre- quenting it. It would be a boon to the agents of the Hudson’s Bay Company at Churchill if they could now winter their small schooner in this cove instead of being obliged to send her every autumn to winter at York Factory. The captain who commands her happens to be the person now in charge of the company’s post at Churchill, and both he and his crew are obliged to walk back 150 miles through the mud from York Factory after leaving their vessel there in the autumn and to walk the same distance again to bring her back in the spring. Mr. J. B. Tyrrell visited Sloop’s Cove in the autumn of 1898, and in a paper published in the Geological Magazine for August, 1894, says he thinks the land is here in a state of equi- R. Bell—Rising of the Land around Hudson Bay. 227 librium. Two inscriptions which he saw on the rocks, namely, “May 25thand May 27th, 1753,” were about seven feet above the present high tide and he thinks these were cut by men standing on the ice. This, however, does not prove much, for the men were quite as likely to have sat as stood while engray- ing these inscriptions. As the tide still enters the cove and keeps it full of water, the average relative level of its ice to the rocks surrounding it may not have differed much from what it is now. When I visited Fort Prince of Wales in 1879, oak planks brought from England while the fort was still occupied as well as timbers of native wood, all charred by Lepeyrouse’s fire, were found stranded far out of reach of the present tides and still in perfect preservation. On the occasion referred to, I met at the “ New Fort” children of some of the people who were living at the “Old Fort” when it was cap- tured by the French, and from them some information could be obtained as to the conditions at that time. We have, besides, the description and illustrations in the book by Samuel Hearne, who was then in charge of the place. Any light which these accounts may throw on the state of matters then as compared with the present time, points in the direction of some elevation having taken place. Among the photographs which I took around Fort Prince of Wales in 1879 is one which shows strips of dry land grasses alternating with little parallel ridges of gravel thrown up by the waves and now above the highest tide-mark, but below the level of the spot which was pointed out to me as the landing place of Lepeyrouse. The ground on which the fort stands was an island during high tide at the time the place was occupied and a bridge was thrown across the narrowest part of the little separating channel to connect the island with the main land. This channel is now entirely dry. If anything further were wanting to show that an elevation of the land is now going on in this region we have some direct personal evidence in the lifetime of the witness himself in sup- port of the facts already cited. About twenty years ago, a very aged Indian, who was said to have ‘‘ seen more than a hun- dred winters,” and who was quietly passing the last years of his extraordinarily long life at Norway House, told me in presence of the factor, Mr. Roderick Ross, and the other gentlemen of that establishment that he had, when a boy, witnessed the land- ing of Lepeyrouse and the destruction of Fort Prince of Wales. He gave graphic details of every circumstance, which agreed perfectly with Lepeyrouse’s own account, and he answered all my questions on other points entirely satisfactorily and with- out a moment’s hesitation. Among other things, he mentioned that the spot where the Frenchmen’s boats landed was quite 228 fh. Bell—Rising of the Land around Hudson Bay. close to that portion of the western wall which they under- mined and blew up with gunpowder. He said that when all was ready, they laid “a rope” (train) of gunpowder across the beach and setting fire to the end of it, ran off to a safe dis- tance to witness the effect. It is now a considerable distance from this spot to the nearest point of water at high tide. The proofs of the rising of the land around Hudson Bay in post-glacial times would be admitted by any geologist, and the question of the continuance of the movement at the present time is, I think, answered in the aftirmative by the actual general shoaling of the water which is going on and the encroachment of the land on all sides, some proofs of which have been given in the foregoing pages. All the facts which have been mentioned (and many more might be added) point in the same direction, while there appears to be no evidence of a contrary character. The officers of the Hudson’s Bay Com- pany are an intelligent set of men, and their universal opinion, based upon lifetimes of observation, is that the land all around the bay is rismg. The following is part of a letter recently received from Mr. Joseph Fortescue, lately a chief factor in the Hudson’s Bay Company, in answer to my request for his opinion on this subject: “ Regarding the rising of the shores of Hudson Bay, I have no doubt whatever. When I was at York Factory, I heard several Indians say that the sea or tide had retired two miles from places they remembered when they were young, and my own observa- tions during twenty years there would lead me to entertain the same opinion. When I revisited Moose Factory, after nearly forty years absence, I found a great change in the appearance of the coast and river. Channels which were navigable at all times of the tide formerly, could now only be used at high water.” Penfield and Pratt— Occurrence of Thaumasite, etc. 229 Art. XXIV.—On the Occurrence of Thaumasite at West Paterson, New Jersey; by 8. L. PENFIELD and J. H. PRATT. In 1878 Baron von Nordenskidld* described a mineral from the copper mines of Areskuta, Jemtland, Sweden, which, according to the analyses of Lindstrém ails had the composition CaSiO,, CaCO,, CaSO,, 14H,O and to which the name thau- masite was given, from Savpdcerr, to be surprised. The min- eral was not found in distinct crystals but was crystalline and on a fracture showed a fine fibrous structure. Its homogeneous character and its right to be considered a distinct mineral species rested upon the following: The material seemed to be homogeneous when examined with the microscope, and the three analyses of Lindstrém, made upon material collected in the early part of this century by Polheimer, in 1859 by Nor- denskidld, and in 1878 by Engberg, agreed not only very closely with one another but also with the theory demanded by the formula. That a mineral with such a remarkable composition was capable of existence was not accepted by all mineralogists, and Bertrand,{ on examining thin sections of it with the microscope was led to believe that it was a mixture, composed of a uniaxial mineral with negative double refraction supposed to be calcite, of a biaxial mineral gypsum, and of a third mineral, the optical properties of which could not be made out, probably calcium silicate or wollastonite. The idea of Bertrand’s that thaumasite was a mixture was not accepted by Nordenskidld, and the latter to sustain his position presented the following arguments,$ which were very convincing: First, if it were possibly a mixture it certainly would be very remarkable that three independent samples, collected at such widely separated periods, should agree so closely in percentage composition. Second, there is no known hydrated calcium silicate which, when mixed with calcite and gypsum, could yield a product containing over 42 per cent of water. Third, it would not be possible for a mixture of cal- cite, gypsum and wollastonite, with specific gravities of 2°72, 231 and 2:90 respectively, to yield a product with such a low specific gravity as thaumasite, 1°877. Specimens were moreover sent to Lacroix for renewed * Comptes Rendus, vol. lxxxvii, p 313, 1878. + Ofv. Ak. Stockholm, vol. xxxv, No. 9, p 43, 1878. t Bull. Soc. Min. de France, VO 11) ps 159: 1880, and vol. iv, p. 8, 1881. § Geol. For. Férhandl., Stockholm, vol. v, p. 270, 1880. 230 Penfield and Pratt— Occurrence of optical examination, and in a letter to Nordenskidld he states* that the material was found to be practically homogeneous, uniaxial and with negative double refraction, but whether hexagonal or tetragonal could not be determined. The uni- axial material which Bertrand had taken for ealcite was in reality thaumasite, and Bertrand in a letter to Nordenskidld+ withdrew his objection. He gives also the approximate indices of refraction w=1°503, e=1:467, which differ from those of calcite. In 1890 Widmant described specimens of thaumasite belonging to the mineral collection of the University of Upsala, which are reported to have been found at Kjélland, about thirteen miles from the original locality Areskuta, and two analyses by Hedstrém quoted by him agree very closely with the ones made by Lindstrém. From Hedstrém’s analy- ses the formula CaSiO,, CaCO,, CaSO,, 15H,O was derived, and as pointed out by Widman this slight change in the formula agrees satisfactorily with the analytical results of Lind- strom, who really had found over fourteen and one-half mole- cules of water. It is with pleasure that the authors are able to announce the discovery of this unusually interesting mineral at Burger’s quarry, West Paterson, New Jersey, the material having been first brought to our notice by Mr. Geo. L. English, of New York, who sent a specimen of it to the mineralogical labora- tory of the Sheffield Scientific School for identification. The mineral occurs as an aggregate of prismatic crystals, sometimes so loosely held together that the individuals can be separated by crushing between the fingers, while more often the masses are firm and have somewhat the appearance of white alabaster. Occasionally distinct prismatic crystals were observed, aver- aging 0°5™" in diameter and 2 to 4"™ in length, but they were poorly formed and without distinct terminations. Some of the masses showing fine prismatic crystals have a decidedly silky luster. There is a distinct prismatic cleavage. Measure- ments were only possible in the prismatic zone and approximated to 60°, which determine the crystallization as hexagonal. On examining fragments imbedded in Canada balsam ones can readily be found which show a uniaxial interference figure with negative double refraction. Using a polished plate, the index of refraction for the ordinary ray was determined by means of total reflection in a-mono-bromnaphthalene and found to be 15125 for yellow Na. By means of a prism of 32° 58’ the following values were also obtained for yellow, o=1°519 * Geol. For. Férhandl., Stockholm, vol. ix, p. 35, 1887. filloid., vol. ix; pelsly nssi- t Ibid., vol. xii, p. 20, 1890. Thaumasite at West Paterson, WV. J. 231 and «=1°476. It must be stated, however, that a prism cut from a crystalline aggregate cannot yield wholly satisfactory results, as the light does not traverse a single individual, and that for example which yielded the extraordinary value above was vibrating in crystals whose vertical axes were approxi- mately and not perfectly parallel to the edge of the prism. Levy and Lacroix* give w=1°507 and «=1°468. In order to be absolutely sure of the uniform character of the material for analysis, selected pieces of the mineral were crushed and sifted toa uniform grain and separated by means of methyl iodide CH,I, which was diluted with ether. That every particle of the mineral in the separator floated at a specific gravity of 1-887 and sank at 1°875, a difference of only 0-012, is sufficient proof of the homogeneous character and great purity of the material. Lindstrém gives as the specific gravity of the Swedish mineral 1:°877 and Widman gives 1°83. The results of the analysis are as follows: i JUL, Til. Average. Ratio. SO) serene 9°23 9:33 9°23 9°26 "155 “97 CO) se eae Ste 27 6°82 "155 7) SOM e. 13°56 13°32 13°44 168 =: 105 CAOr 2... MOS 20 NOP 23 "484 3°04 EEO se 42°81 42°72 42°77 2°377 15°00 INGO ae “39 "39 REO 1 “18 18 99°99 The ratio of SiO,:CO,:SO,:CaOQ:H,O is very nearly 1:1:1:38:15, demanded by the formula CaSiO,, CaCO,, CaSO,, 15H,O. The analytical results are, moreover, very close to those obtained upon the Swedish mineral by Lindstrom and Hedstrom. A slight amount of alkali sulphate is prob- ably present as impurity, therefore the alkalies have been neglected in making the above calculation. That Na,O and K,O are not isomorphous with CaO is shown by the following experiment: 1:1765 gram of the powdered mineral were treated in a platinum dish for over two days with cold water, the insoluble mineral was then filtered off and the soluble por- tion analyzed, with the following results: SiO,, 0°39 per cent ; SO,, 0°56; CaO, 0°56; Na,O+K,O, 0°25. These indicate that tbaumasite is slightly soluble and that the alkalies have an independent existence, for a quantity of Na,O+K,O equal to about one-half of that found in the original analysis was * Les Minéraux des Roches, p. 286, 1888. 232 Penfield and Pratt—Occurrence of extracted, while relatively only a very small proportion of the calcium was dissolved, a result which would not have taken place if the alkalies had belonged with the thaumasite. > Utewn E ie ee: ~~ @yrad Fie. 1.—Sketch map of the Bearpaw Mts. Outside of the military reservation the region is settled wherever there are level tracts suitable for cultivation or for the cutting of hay. The heavy growth of grass and the gentle, open nature of the country have been found especially suitable. for stock raising. 286 Weed and Pirsson— Bearpaw Mountains, Montana. The highest western point of the mountains is known as Centennial Butte, and the isolated, conical elevation standing at the extreme east end is called Eagle Butte. A few other names 1n common use among the settlers are given upon the accompanying sketch map. The mountains have never been surveyed. They appear on most maps of the State, but only the merest approximation to the drainage is given, and the intervening space is filled with meaningless hachures. The accompanying sketch map, com- piled from various sources and our own notes, is the best obtainable. General Geology. The Bearpaw Mountains are the dissected remains of a group of voleanoes of Tertiary age. Denudation has laid bare the cen- tral cores or necks of the old vents, surrounded by altered sedi- mentary rocks through which the conduits were made. Radial dikes traverse the sedimentary foundation, and accumulations of scoria and various fragmentary volcanic rocks and lava flows form the outer part of the mountain region and also a portion of the central area. The accompanying figure represents a dia- grammatic cross-section through the range, showing the stock of granular rock exposed on Beaver Creek and the relations of the effusive rocks to the basal platform of Cretaceous shales. The geology is interesting, not only from these fea- tures, which have their counterpart in the neighboring High- wood Mountajns, but also on account of the character of the eruptive rocks and the bearing of their occurrence upon the broad problems of petrology. The mountains have not been glaciated, although the confluent terminal moraines of the two great continental ice sheets skirt the flanks on all but the southern sides. So _ Cretaceous Massive Tufts Strata. Igneous Rock. and flows. Fig. 2.—North and South section of Bearpaw Mountains through Bearpaw Peak and Beaver Creek Core. The sedimentary rocks are Cretaceous, and in part at least of Laramie age. They consist of a thickness of 2,000 to 3,000 feet of dark-colored shales; with interbedded sandstones and Weed and Pirsson— Bearpaw Mountains, Montana. 287 occasional lenticular intercalations of impure limestones. So far as observed, the rocks have proved nearly barren of fossils, but asa coal-bearing series occurring along Milk River is also exposed beneath the drift along the stream courses north of the mountains, fossils will probably be found upon a close examination of the beds.* Older Cretaceous strata also occur to the north, near Havre, where they have been identi- fied by Mr. T. W. Stanton, and in the bad lands of the Missouri River, where they have been studied by one of the authors (W. H. W.). About the cores of granular igneous rocks the sedimentary beds are generally highly metamor- phosed, more especially where the eruptives are of basic types. In such cases the original character is obscured, and the shales and sandstones are converted into dense, hard horn- stones and quartzites of those light-colored, compact and flinty rocks of various shades of lavender, gray, green, and of adinole-like character, which so often characterize the con- tact zones of igneous intrusions. These baked rocks possess a marked cubical jointing, which causes them to break into small angular fragments, so that imposing exposures are seldom seen. As, however, they often resist erosion better than the granular rocks of the volcanic cores, the contact ring generally stands in bold relief, forming ridges about the igneous centers. Where dikes occur thickly clustered, as is.the case on the ridge at the head of Snake Creek, the sedimentary rocks are also much indurated and altered. From what has already been stated, it is evident that the chief interest in the region lies in the igneous rocks. The unusual character of the types represented make their study of importance, for which reason the field occurrence and the petrographic character are described somewhat in detail. Extrusive Rocks. The extrusive rocks are most abundant, forming the highest peaks and many of the lesser summits of the region; they are the usual rocks of the foot-hills, where their richly-colored, rough outcrops form crags and oddly shaped castellated masses in strong contrast to the smooth and grassy slopes about them. They consist of dark-colored basaltic tuffs, breccias and lava flows, which are parts of the former volcanic cones. The finer tufaceous varieties are often washed down and form dark-colored, sandy soils, generally well grassed, while the vesicular slaggy rocks and coarser breccias form * We have been informed by Prof. O. C. Marsh that the remains of a Dinosaur Hadrosaurus breviceps Marsh) described by him (this Jour., vol. xxxvii, p. 335, 1889, also ibid, vol. xxxix, p. 423, 1890), are stated by the collector to have been _ obtained in the Bearpaw Mountains. 288 Weed and Pirsson—Bearpaw Mountains, Montana. rough masses standing above these slopes. The rugged hills forming the eastern end of the range are largely covered by these crags, whose creamy tints of purple, red, brown, and gray are a striking feature of the scenery. Within the mountain area the effusive fragmental rocks are quite irregularly distributed. In general they rest upon a formerly eroded surface that was hilly in nature; so that the exposures occur both as cappings to hills and in masses at lower altitudes; no definite bedding is recognizable. The irregular disposition of these masses is well shown in the accidented region between Clear and Beaver Creeks, where the effusive rocks and sedimentary beds are exposed at various altitudes and without apparent order. The rocks of this part of the mountains are, in part at least, more acidic than those generally prevailing, the breccias including fragments of trachytic rocks. The hills in this interior part are smoothly contoured, well grassed, and show the black patches of gravelly debris so characteristic of these basaltic accumula- tions and which accentuate and bring into strong relief the grassy ridges and intervening hollows. The rocks here also frequently form rough, craggy exposures, almost invariably found upon the southern side of the hills. Tn the effusive rocks the breccias predominate, with associ- ated tuff beds and lava flows. The rocks vary from compact, dense basalts showing no porphyritical crystals, to open porous forms which pass into scorias. They have a wide range of colors, red, brown, gray, in warm rich tones, more rarely green and buff. Though varying in appearance they consist almost wholly of leucite basalts; many varieties have no megascopic phenocrysts, but most of them show a dark ground-mass peppered with small white specks of altered leucite, while occasionally erystals of augite are seen, and more rarely olivine. Sometimes the breccias contain blocks of a dense, black rock with large phenocrysts of altered mica. The tufis vary from pale-buff or dull-green to dark red, brown, or gray in color, and are largely altered. A specimen selected for examination representing a common variety seen throughout the mountains but coming from Bull Hook Butte, is typical of the rock forming the larger blocks and fragments in the breccias. The rock is of a very dark- gray color, rather compact on the whole, though on jointing surfaces it seems rather porous, and this is probably due to the weathering out of an iron-bearing component; on a completely weathered surface it becomes of a leather-brown color. On fresh surfaces one readily detects yellowish grains of olivine, passing at times into a chestnut-red color and 1 or 2™™ Weed and Pirsson— Bearpaw Mountains, Montana. 289 long, black prisms of augite of about the same size or larger, and round white or very pale greenish grains of leucite, which thickly pepper the dark ground-mass and which vary in size from mere white dots up to 0°%5™™ across. The rock strongly recalls certain of the south Italan leucite rocks in which the leucite phenocrysts are very small. In thin section the rock is composed of the above minerals thickly crowded, so that there is little base. The augite is of the usual character in rocks of this class; in section it becomes very pale in color and of a brownish rather than greenish tone and has a strong zonal structure, is very idiomorphic, frequently twinned on a@(100), has a wide angle of extinction up to 40° or more, shows excellent cleav- age and is very fresh. The olivine is generally idiomorphic though occurring at times in polysomatic groups; the large orystals are fresh and clear ; the smaller ones frequently changed into a deep-brown or reddish-brown colored substance, which appears to be simi- lar to alterations of olivine seen in other occurrences,* in this change the smaller crystals are completely altered, larger ones only partially, and in the largest the alteration appears only as red-brown bands following cleavage planes. We believe this to be due to a change in the iron oxide in the mineral, and that it is especially iron-rich olivines which are lable to it. The Jewcites under the microscope are seen to be very thickly crowded, composing the bulk of the rock. They are of all sizes, from very minute individuals up to those previ- ously mentioned. They are of rounded forms though not generally bounded by an absolutely sharp, definite line, but fade out into the ground-mass in a rather ragged indefinite way. They do not contain any of the inclusions zonally arranged which are so frequently seen in this mineral, but are, like the larger leucite phenocrysts in the leucitic rocks, entirely free from them; like these larger phenocrysts they are fre- quently cracked or appear in grouped polysomatic forms. They are not colorless and limpid like the leucites of the fresh Italian rocks, but are of a light-brown color and appear with a low power very like a kaolinized orthoclase in character; they are perfectly isotropic between crossed nicols. Studied with high powers they appear filled with excessively fine granules, shreds and leaves of some substance so fine that they can barely be discriminated ; this material is unevenly distributed through them and does not act on polarized light. It is possi- * Rosenbusch, N. Jahrb., 1872, p. 59, Phys. d. Min., 1892, p. 472; Michel Lévy, Bull. Geol. Soe. France, 3d Ser. xviii, 1890, p. 831; Iddings, Geol. Eureka Dist., Mon. xx., U. S., Geol. Surv., Appendix B, pp. 388-390; Pirsson, this Journal, vol. xlv, 1893, p. 381; Lawson, Geol. Carmelo Bay, Bull. Univ. Cal., vol i, 1893, p. 31. 290 Weed and Pirsson— Bearpaw Mountains, Montana. ble that it may be original and represents extremely fine inclusions, which, had they been gathered into larger grains, might have arranged themselves zonally. We believe, however, that they really represent a stage of alteration and that the leucites, since they appear so exactly like kaolinized orthoelase, are in fact partially kaolinized. That some alteration has taken place seems possible from the fact that the rock contains over 3°5 per cent. of water, though this in part may well come from the including base. Qualitative tests showed the absence of SO, and Cl, and the mineral is therefore not hauyn, nosean, nor sodalite. | The rock appears closely related to the analcite basalts of the Highwood Mountains which have been described by Lindgren,* who showed that in the material studied by him the mineral resembling leucite was really analcite, and from the freshness of the other components he was inclined to regard it as of primary origin. In the course of our own work in the Highwood Mountains we have gathered a large and varied collection of these leucite-like rocks and have found them to consist of several types, and we hope that the investigation which we are now making of them will throw some light on the character and origin of the leucite-analcite minerals. The discussion of these similar Bearpaw types is deferred in conse- quence until our publication upon this material, which will ap- pear at a later date. The ground-mass in which the other minerals lie is very small in amount and consists mainly of minute black grains of magnetite, with tiny microlites of augite imbedded in what appears to be 4 colorless glass. The composition of the rock from the foregoing is thus shown to be a very simple one—magnetite, augite, olivine, and probably altered leucite. There are no signs of any feldspar present, and the rock is thus a leucite basalt. The occurrence of leucite in this locality we shall have occasion to mention later, in connection with its presence in some remarkably fresh, unaltered, and interesting lavas from Bearpaw Peak. Intrusive Rocks. kagle Butte.-—The eastern end of the mountains, near Cleveland, is formed of the sedimentary rocks tilted near the postoftice at angles of 15° to 20° to the southeast, while the prominent buttes seen near by are formed of igneous intrusions of basaltic rocks. The extreme easterly point of the moun- tains is an isolated conical hill of considerable elevation, called Eagle Butte on the map. It shows a cap of black debris * 10th Census U.S., vol. xv, p. 719. Proc. Cal. Acad. Sci., Ser. 2, vol. ili., p. 51. Weed and Pirsson— Bearpaw Mountains, Montana. 291 extending downward in tongues cutting the grassy middle slopes. The base of the butte is formed of rough outcrops of purplish, reddish, and steel-colored rocks, with fragments of a light-colored variety which seemed to form a dike or intrusive mass, as seen from the summit of Gray Butte. The rock is a mica-trachyte, having a medium gray color, is of fine grain and filled with many very small glittering black tablets of biotite and with small whitish spots and specks. Under the microscope it appears very badly altered, much more so than the megascopic appearance would suggest. The mica is fresh and recalls that of minettes; there is, moreover, a great deal of it, and it is strongly pleochroic and very idio- morphic. The rock is filled with masses of calcite, which appears due to the decomposition of augite; some may possibly have been introduced. Secondary quartz also appears filling cavities. There are a number of phenocrysts of feldspar generally show- ing both albite and carlsbad twinning; one of these, oriented in the zone 001 on 100 perpendicular to 010, gave extinction angles with the Bertrand ocular for the albite twins a = 19° a’ = 19, the carlsbad twin gave 9°, and hence the plagioclase is andesine. The ground-mass is made up of lath-shaped feld- spars, which are so much altered that it cannot be safely told whether a plagioclase which is present or alkali feldspar pre- dominates. The alkali feldspar appears, however, to predom- inate. It is certainly present, as a little interstitial quartz permits the recognition by Becke’s method that some less altered, unstriated granules, have in all positions less refraction than the quartz. The rock appears, therefore, to be a transi- tion form between a mica-andesite and a mica-trachyte so far as can be told. Gray Butte is the most western of the isolated group of three elevations, which forms such a prominent feature of the east end of the mountains. Unlike its neighbors it is rugged in appearance, showing bold exposures of massive rock with talus of large blocks covering the slopes below. The lower part for 200 to 300 feet above the base is a steep slope cut in soft black shales that show no fossils and are apparently horizontal. The shaly beds are cut at the southwestern base by a dike of decomposed, hornblendic trachyte. The butte above is formed of an intrusive mass of massive, alkali guartze-syenite-porphyry, breaking up through the shales in a boss a half mile wide. The gray color of the outcrops and debris piles which gives the name to the butte is due to lichens, as the rock is a pale-green on fresh fracture, becoming pink on weathered surfaces. The shales are slightly altered for a few yards distant from the coutact. The rock occurs in place on the summit of the butte, 292 Weed and Pirsson— Bearpaw Mountains, Montana. where it forms a massively jointed wall extending north and south along its crest, while below large talus blocks cover a steep slope extending down to the top of bold cliffs of the same rock. The cliffs, seen from the surrounding country, form the most noticeable feature of the butte. On three sides they form an almost insurmountable wall, over a hundred feet in height. In these cliffs the syenite-porphyry has a massive, platy structure, the lamination being parallel to the outer sur- face of the cliffs, dipping on all sides away from the butte at 30° to 45°. Though the rock sometimes breaks along these planes and forms a smooth rock slope, the exposed face of the cliffs is generally much steeper and corresponds to a jointing of the rock. The debris consists of blocks, often 8 feet in diameter, with very little material less than a foot across. The weathered rock has a botryoidal surface that is rather characteristic. No difference is recognizable between the rock forming these cliffs and that of the summit of the butte, either in the nature or relative amounts of the component minerals, or in the granularity or physical structure. It appears prob- able, however, that erosion has as yet only exposed the outer part of the boss, and that both summit and side show only the outer portion of the intruded mass. On a surface of fresh fracture the rock is a very pale-gray color mottled with very pale-pink; on a somewhat weathered surface these contrasts of coloring are accentuated. Upon closer examination the rock is seen to be chiefly composed of large, thickly crowded crystals of feldspar, held in a dense pale-green ground-mass that is speckled with small black augite crystals. The pale-pink tone of the rock is due to these thickly crowded phenocrysts of feldspar. The black augites which occur plentifully scattered through the ground-mass rarely attain a length of 2™™.. The feldspar phenoerysts are stout in habit and vary in size from one to one-half centimeter in length. They usually possess the orthoclase habit, are very rarely clear glassy and sanidine-like, and are developed in short, stout, columnar forms on the clino axis; they are apt to be more or less rounded and anhedral.* When idiomorphic they show the common planes m(110), 6(010), and c(O0L). They are frequently carlsbad, less often baveno twins. A few sporadic grains of quartz have also been observed, though they do not appear to be at all common. * A term suggested for rounded formless masses which do not show outward crystal form. See report of the winter meeting of the Geol. Soc. Am. 1895, in its proceedings, also abstracts in this Journal, p. 150, vol. i, 1896; Jour. Geol. vol. iv, 1896, p. 128. It has been suggested that the terms anhedron and anhedral conflict with allotriomorphic; this is by no means the case, since allotriomorphic ig a term of structure denoting a formal relation of parts, while anhedron is indi- vidual and used in a crystallographic sense. Weed and Pirsson—Bearpaw Mountains, Montana. 293 Under the microscope the rock is found to be composed of apatite, egrrite-augite, alkala feldspars and quartz. The egirite-augite has the usual characters of this mineral ; a clear light grass-green color, pleochroism, though not strong, into tones of yellow, an excellent cleavage, and is usually quite idiomorphic. It is zonally built, increasing in the egirite molecule towards the periphery, and has in these parts a corresponding decrease in the angle of extinction. It has in many cases a pronounced dispersion of the optie axes, which is probably owing to the presence of titanium. Since the mineral composition of the rock is, as indicated above, so simple, we may reckon out of the bulk rock analysis the chemical composition of this augite, which then has approximately the following composition : SiO. Fe203 FeO MgO CaO Na.O 45°3 22°1 1°8 8°6 15°6 8°6=100°0 corresponding in round numbers to two molecules of diopside to one of egirite. The apatite has been seen only in a very few, rather large, scattered grains; the small amount of it is indicated in the minute quantity of phosphoric anhydride shown in the analysis. The large feldspar phenocrysts between enoseed nicols appear homogeneous and in general untwinned, except an occasional carlsbad. In some places areas of kaolinization occur, otherwise they are quite fresh. The cleavages parallel to 6(010) and ¢(001) are both excellent and furnish good plates for optical examination: such plates parallel to ¢ (001) extinguish at about one degree from the trace of 6 (010) while plates parallel to 6 (010) extinguish positively at 10° 30° to 11° 30’ from the trace of ¢(001) and show in convergent light a positive bisectrix nearly centered in the field. The feldspar is, therefore, not orthoclase but anorthoclase, as must indeed be expected ‘from the rock analysis. It seems to be closely rela- ted to the cryptoperthite described by Brogger.* 2-2 as - XXXII.—Notes on Glacial Gravels, in the Lower Susque: hanna Valley; by H. B. BasHore.....2-. eee 281 XXXIV.—The Bearpaw Mountains, Montana. First Paper; by W. H. Weep and L. V. Pirsson Peete a ples Be 2) XXXV.—Pleistocene Marine Shore-Lines on the South Side Page ~ of the St. Lawrence Valley; by R. CuHatmers.-_-_------ 302 XXXVI.—Occurrence of Frée Gold in Granite; by G. P. , MMB ROLE ee aoe, woe OI es at nee XXXVI. eens of the “ X- “Rays; 9 ‘by A. A. MicuEtson 312 SCIENTIFIC INTELLIGENCE. Chemistry and Physics—Some physical properties of argon and helium, Lorp RAYLEIGH: Mixed double halides of platinum and potassium, HERTY, 315.— Relative atomic weights of oxygen and hydrogen, J. THOMSEN: Synthesis of caffeine, EK. Fischer and L. AcH: Free hydrazine, L. de BRUY¥YN,316 —Praetical Tnorganic Chemistry, G. S. Turpin: Chemical Experiments, General and Ana- lytical, R. P. Witutams: Text Book of Gas Manufacture for Students, J. HornsBy: Methods of determination of Dielectric Constants, W. NeERNST, 317.— Dielectric Resistance, P. DRupE: Index of refraction and reflective power of water and alcohol for Electric Waves, A. D. Cote: Rontgen Rays, J. J. THom- SON, 318.—Experiments with the Rontgen Rays, DoELTER: Elements of Phys- ies: A College Text-book, E. L. NicHors and W. S. FRANKLIN, 319. Geology and Natural History—Contributions to the Cretaceous Paleontology of the Pacific Coast, T. W. STANTON, 320.—Contribution a l’étude des lapiés, M. EMILE CHAIX: Geographical distribution of marine animals, 321.—Charles Lyell and modern Geology, T. G. BonNnEY: Beitrage zur Geophysik, G. GER- LAND, 322.—Structure-Planes of Corundum and some Massive Minerals (simple rocks) from India and Australia, J. W. Jupp: Remarkable phosphoresence in Wollastonite, W. F. HILLEBRAND, 323.—Propagation of the sugar-cane, J. H, ~WaAKKER: Systematic arrangement of Australian Fungi, together with Host- Index and List of Works on the subject, D. McALPINE, 324.—Der Reis-Brand und der Setaria-Brand, O. BREFELD, 325.—-Entwickelung des Peritheciums bei Sphaerotheca Castagnei, R. A. HARPER, 326. Obituary—DR. JOHANN MULLER, 326. Miscellaneous Scientific Intelligence —Catalogue of Scientific Papers (1874-1883), compiled .by the Royal Society of London: Yellowstone National Park, H. M. CHITTENDEN: Mechanics and Hydrostatics: R. T. GLAZEBROOK, 327.—Consti- tution and Functions of Gases, the Nature of Radiance, and the Law of Radia- tion, 8S. J. Corrigan: Algebra for Beginners: Introduction to the Algebra of Quantics, E. B. Ettiotr: Problems in Differential Calculus, W. HK. BYERLY: Plane and Solid Geometry, W. W. BeMAn and D. E. SmitH: Computation Rules and Logarithms with Tables of other useful Functions, S. W. HOLMAN, 328. D. Walcott, a po Pema RATE At fe I wh eae : re oartiee : a Say a . S. Geol. fey, i a 7. ae 1 eae ie Rg AMERICAN JOURNAL OF SCIENCE. Kpitorn: EDWARD 8S. DANA Pe! ASSOCIATE EDITORS Prorrssors GEO. L. GOODALE, JOHN TROWBRIDGE, H. P. BOWDITCH awnp W. G. FARLOW, or Camprines, Prorerssors H. A. NEWTON, O. C. MARSH, A. E. VERRILL AND H. 8S. WILLIAMS, or New Haven, , a EEE RE TR ey S — a r —— Po - ° a : Ae ae oc AOR TER ee ee ee: ae Pe 0 a a + ' a ’ ‘ . . : =i eS: ’ Soria, - >. taal Ne fs eee ey Seah wy ‘ * ms z a et See I AEs = — ~ “9 >» * ee ahie we kasisee ae. a i i ; Paes Prorrssor GEORGE F. BARKER, or Puiavetpuia, Prorsessor H. A. ROWLAND, or Batrimorgz, | Mr. J. 8. DILLER, or Wasuineron. FOURTH SERIES. VOL. I—[WHOLE NUMBER, CLI.] No. 5.—MAY, 1896. PLATE IX. NEW HAVEN, CONNECTICUT. LS. TUTTLE, MOREHOUSE & TAYLOR, PRINTERS, 12 TEMPLE STREET ; Published monthly. Six dollars per year (postage prepaid). $6.40 to foreign sub- - scribers of countries in the Postal Union. Remittances should be made either by _ money orders, registered letters, or bank checks. e ~ ty a ay We" The mineral collection of W. M. Foote, including the following rarities Rea i American and Kur opean mineral localities, ‘has just been placed on sale with us. Most of the specimens are of moderate price, having originally been retained more because of their neat appearance and perfect crystallization than for their — commercial value, and in these regards they may be considered “ first pick” from numerous large lots. In the present breaking up of the collection an opportunity is offered which has already been recognized ‘by several of our patrons, and collectors will do well to “secure at once such specimens as they may wish. They are not offered as bargains, but as first class selected specimens at pares ~ usually obtained for such. (@s" Specimens sent on approval and sold singly. Laurionite, Laurium, Greece, a Ee 9 EE ae a ees ee ners $3.00 | Chalcocite, var. Redruthite, Corn- wall, England. 2”x2" $3.00; 1” 1.25 Cinnabar, Spain. Brilliant crystals on gangue. 24°x24" ___-__.---- 5.00 | _Nadorite, Algeria. Cryst.. 3"x2$" 2.00 | Libethenite, Cornwall, England. Olivenite, , Cornwall, DAU Med ae Pa hae Sey pea 1.50 Phosgenite, Cromford, Derbyshire, England. Bright sharp crystals. aie | aak Vy wars) ee AES 6.00 | Barite, Cumberland, England. The neatest and most perfect example of this lovely mineral we have Chalcosiderite, Cornwall, England. Daas PUN tah goes re ae 3.00 Crocidolite, S. Africa. Blue, pol- ISHOC Ges A LS pe Ne Pa ie 3.50 | Quartz pseudo, after Crocidolite, “Tiger’s Hye,” 8. Africa, pol- ished. 33x33" (Fibres 32” long) $5.00; 44°x12"_._----2- 2.50 Mazapilite, crystals, Mexico. Octahedrite, Switzerland. 34"x3" 2.50 Psilomelane, Cornwall, England. Sialacubic 2s das See nL Fluorite, coated with Quartz, Corn- wall, England. Very handsome specimens) +6: x4) (EL. oo 3.50 Chalcedony, blue, Queretaro, Mex- The UV ANGSS Rae Sel ate cea Lea oF 2.50 Hye Agate, Brazil,» 2°x2"_ 22255 .- 1.00 Bowrnonite, Cornwall, England. Seas SCA NOR ae aa Sotay 3.50 Melanophlogite, Sicily. 23"x2"____ 2.00 Ouprite, Cornwall, England, 34"x3" 3.50 Sylvanite, Hungary. 3’x2$"_____ 5.00 Sulphur, Sicily. Fine crystals. Exceptionally choice. 32"x3"__ 7.00 Zoisite var. Thulite, Norway. Pol- ished. Rich color. 34’x3"__.. 1.00 Dr. Garnet, var. Melanite, Franklin, ; NJ... 8"x37i3 7a $2.50 | Dioptase, Siberia. Brilliant erystals of rich color. ()12!x1t ve ee 10.00 Chilorastrolite, Lake Superior. Very large polished pebble. 14”x1}" 17.50 | ORs Ld" fe ae ae 2.00 Wulfenite and Pyromorphite, Phoe- — nixville, Pa, oxi 3" SoS 1.50 | Liroconite, Cornwall, England. 1}? xh!" i eee 2.50 | Anglesite, Phoenixville, Pa. 34’x24" 2.50 ever eed. Dy. xars te ee, STB Leadhillite and Sie es Scot- land.!.: 4". ee ee 5.00 Cerussite, Phoenixville, Pa. 24”x2" 2.50 Azurite, Arizona. Grouping of balls. - 24x14" eee 1.50 Anglesite, Sardinia. Brilliant erys- — tals; ) 2"x2" 222 ee ee 2.00° Native Lead, Sweden. On rock. Dx ee a ee ee 1.00 | Serpierite, Laurium, Greece. 2”x2" 1.50 — Molybdenite, Canada. Very large surface. 8'x3g" en See 15.00 Ner y fine crystals in rock. 3”x3"_30.00 re 2”x3"_ 5.00 Rhodochrosite, Hungary. Delicate pink color. ‘“‘ Himberspath ” : BAK". ne i aes 3.50 Flos Ferri. Santa Eulalia Mines, Mexico. Beautiful form. 4"x3" 1.75 Gold, Hungary. 2”x14"$3.00____ 5.00 Mimetite, Cumberland, England. Very fine. 34:33 020 cau wea 7.50 Siderite, single crystal, very rare type, on Quartz, Cornwall, Eng- land. 13" 0. YS es ae 15.00 Bismuthenite, Cornwall, England. AIR! A ees 3.50 Domeykite, var. Condurrite, Corn- wall, England. 2 0z, .----...- 2.50 A. E. FOOTE, WARREN. M. Foote, Manager. 1224-26-28 North Worty-First Street, PHILADELPHIA, PA., U.S A. THE AMERICAN JOURNAL OF SCIENCE [FOURTH SERIES.] Oe Art. XXXVIII.—Carbon and Oxygen in the Sun; by JOHN TROWBRIDGE. - In 1887 Professor Hutchins of Bowdoin College and myself brought forth evidence to show that the peculiar bands of the voltaic-are spectrum of carbon can be detected in the sun’s spectrum. They are, however, almost obliterated by the over- lying absorption lines of other metals, especially by the lines due toiron. In order to form an idea of the amount of iron in the atmosphere of the sun which would be necessary to obliterate the banded spectra of carbon, I have compared the spectrum of carbon with that of carbon dust and a definite proportion of iron distributed uniformly through it. The car- bon dust and iron reduced by hydrogen was formed into pencils suitable for forming the voltaic are.* Chemical analysis showed that the iron was uniformly mixed with the carbon dust. Speci- mens, taken from different sections of the carbons which I burned in the are, gave twenty-eight per cent of iron and seventy-two per cent of carbon. The method of experimenting was as follows: That portion of the spectrum of the sun which contains traces of the pecu- liar carbon band lying at wave length 3883-7 and which has been almost obliterated by the accompanying lines of absorp- tion of other metals, among them those of iron, was photo- graphed. The pure carbon banded spectrum was photographed on the same plate immediately below the solar spectrum, and the spectrum of the mixture of iron and carbon immediately below this. The sun’s spectrum can be regarded as a com- *T am indebted to Mr. John Lee.of the American Bell Telephone Co. for his skill in making the carbons and for analyses of the composite carbons. Am. Jour. So1.—FourtH Srriges, Vou. I, No. 5.—May, 1896. 330 3S. Trowbridge—Carbon and Oxygen in the Sun. posite photograph, and the iron and carbon ean also be regarded as a composite photograph. It was speedily seen that from twenty-eight to thirty per cent of iron in combination with seventy-two to seventy per cent of carbon almost completely obliterated the peculiar banded spectrum of carbon. This proportion therefore of iron in the atmosphere of the sun, were there no other vapors of metals present, would be suffi- cient to prevent our seeing the full spectrum of carbon. The iron in the carbon terminals which I employed greatly increased the conductivity, as will be seen from Table I, which | was obtained in the following manner. The carbons were separated by means of a micrometer screw and the current and difference of potential were meas- ured with different lengths of arc. Table I gives the results for pure carbon. Table II for twenty-eight per cent of iron and seventy-two per cent of carbon. TABLE I, Length of arc in mm. Amperes. Volts. 1 Jae 25 2 23 24 3 22°5 20 4 20 18 5 16°5 15 TABLE II. Length of arc in mm. Amperes. Volts. 1 * 30'°5 30 2 29 - 180 3 27°5 28 4 24 25 5 22. 20 6 20 20 ff 18 19 8 16 18 The length of the are could be nearly doubled with the same current and the same voltage by the admixture of 28 per cent of iron. The light was apparently greatly increased, but the difference in color between the pure carbon light and the iron carbon light made measurements unreliable. } Moissan* has shown that the carbon in an electric oven through which powerful electric currents have flowed is free from foreign admixtures. Deslandres has confirmed this and finds only a trace of calcium present. The purification comes from a species of distillation of the volatile impurities. The purest carbon is found at the negative pole. The light of the * Comptes Rendus, CxX, pp. 1259-1260, 1895. J. Trowbridge—Carbon and Oxygen in the Sun. 381 electric furnace is due to the combustion of carbon. Can we conclude that the sun is also a vast electric furnace ? If the voltaic are is formed under water its brilliancy dimin- ishes greatly. On the other hand, an atmosphere of oxygen greatly augments its vividness. The question, therefore, whether oxygen exists in the sun is closely related to questions in regard to the presence of carbon, when we consider the temperature and light of the sun. , If suppositions also are made in regard to the magnetic con- dition of the atmosphere of the sun, it is of great interest to determine whether oxygen exists there, for oxygen has been shown by Faraday, and later by Professor Dewar, to be strongly maenetic. Professor Henry Draper brought forward evidence to prove the existence of bright oxygen lines in the solar spectrum. Professor Hutchins of Bowdoin College and myself examined this evidence and, after a long study of the oxygen spectrum in comparison with the solar spectrum, came to the conclusion that the bright lines of oxygen could not be distinguished in the solar spectrum. We published our paper in 1885. I have lately studied the subject from another standpoint, having carefully examined the regions in the solar spectrum where the bright lines of oxygen should occur if they manifest them- selves, in order to see if any of the fine absorption lines of iron in the spectrum of iron were absent, for it is reasonable to suppose that the bright nebulous lines of oxygen would oblit- erate the faintest lines of iron. The method adopted by Draper for obtaining the spectrum of oxygen consisted in the employment of a powerful spark in ordinary air. ‘To obtain this spark the current from adynamo running through the primary of a Ruhmkorf coil was suitably interrupted. By the use of an alternating machine and a step- up transformer, powerful sparks can be more readily obtained. Since the time of exposure with a grating of large dispersion is long, considerable heat is developed in the transformer from the powerful currents which are necessary to produce a spark of sufficient brilliancy. I have therefore modified the method in the following manner. The spark gap is enclosed in a suit- able chamber, which can be exhausted. When the exhaustion is pushed to a certain point, the length of the spark can be increased ten or twenty times over its length in air, and a suitable spark for photographic purposes can therefore be obtained by the employment of far less energy in the trans- former. A pressure of eight to ten inches of mercury in the exhausted vessel is sufficient. A quartz lens inserted in the wells of the exhausted chamber serves to focus the light of the spark on the slit of the spectroscope. 3382 J. Trowbridge—Carbon and Oxygen in the Sun. The following table gives the oxygen lines and iron lines im the same region of the spectrum. 0 Fe in Sun. Intensity. 4631 4629°44 4730°22 4630°91 . 4361°61 4656 4654°7 4657°71 4683 4683:°04 1 4683°93 2 4601°5 4600°09 1n 4601°8 1 4602°11 4 4007 4604°84 1n 4605°52 1 4606'°34 4607°79 6 4618 4611°38 8 4613°35 4 1 6 1 aS eS ee 4614°29 4693°5 4691°52 4696°97 The faintest iron lines are therefore not obliterated in the spaces where the oxygen lines should occur. If we examine the great absorption region about the K line, we find that between wave lengths 3930-29 and 3938°55 Rowland, it gives 8 lines which coincide with iron lines. From the table of wave lengths of iron lines in the are spec- trum given in the report of the British Association for 1891, I find the following lines given between these limits. 3930°37* 3931:22* 3932°-71* 3933-01" 3933°75 3934°47 3934 81* 3935°40* 3935°92* 3937427 2938°16 3938°59 The starred lines are probably the iron lines given by Row- land in his list of standard solar lines. The iron lines that are not starred apparently are obliterated in the great absorption band near the calcium line K. H. Jacoby—Determination of the Division Errors, etc. 333 Lord Salisbury, in his address before the British Association at Oxford, 1894, remarks: ‘Oxygen constitutes the largest portion of the solid and liquid substances of our planet, so far as we know it; and nitrogen is very far the predominant con- stituent of our atmosphere. If the earth is a detached bit, whirled off the mass, leaving the sun we cleaned him out so completely of his nitrogen and oxygen that not a trace of these gases remain behind to be discovered even by the sensitive vision of the spectroscope ?” Although we have not succeeded in detecting oxygen in the sun, it seems to me that the character of its light, the fact of the combustion of carbon in its mass, the conditions for the incandescence of the oxides of the rare earths which .exist, would prevent the detection of oxygen in its uncombined state. Notwithstanding the negative evidence which I have brought forward, I cannot help feeling strongly that oxygen is present in the sun and that the sun’s light is due to carbon vapor in an atmosphere of oxygen. Jefferson Physical Laboratory. ArT. XXX1IX.—On the Determination of the Dwision Errors of a Straight Scale; by HAROLD JACOBY. 1. LyIne as it does at the base of all exact metrology, the above problem has received the attention of many investigators. Numerous methods have been devised and employed for its solution, the object generally sought being a result of sufii- client accuracy accompanied with the minimum amount of labor. Probably the first investigation in which the highest precision was attained or even aimed at, was carried out by Hansen.* His method amounts to comparing the spaces of the scale under investigation, which we will call scale A, with a series of spaces marked upon an auxiliary scale B.. This was done in such a way that every one-space of scale A was com- pared with an auxiliary one-space ; every two-space of scale A, with an auxiliary two-space, etc. The method may of course be varied by using, instead of spaces on an auxiliary scale, a fixed interval between the two microscopes of a comparateur. In fact, Hansen’s process really consisted in comparing inter se all the smmgle spaces, two-spaces, three-spaces, ete., of scale A.. The series of observations made in this way, Hansen treats by . the method of least squares, giving a solution whose numeri- cal application requires a very considerable amount of compu- tation. * Abh. d. math. phys. Classe der Kgl. Sachs. Gesell. d. Wissenschaften, vol. xv. 384 HT. Jacoby— Determination of the Division This method of Hansen appears to have received no mate- rial improvement until 1888. In that year, Gill, acting upon a suggestion made by Marth, began to determine the errors of the straight scales of the Cape heliometer. Gill employs an auxiliary scale B which is an exact duplicate of scale A. In fact, in the case of the heliometer, each of the two scales is used as an auxiliary for investigating the other. Instead of then comparing, as Hansen did, every one-space, two-space, etc., of scale A with one auxiliary one-space, two-space, ete., on scale B, Gill compares every one-space of scale A with every one-space of scale B; every two-space with every two-space ; etc. This method must be regarded as an important extension of Hansen’s. With comparatively little additional labor in the observations, it causes a considerable increase of precision in the results. Precisely the same method of observation employed by Gill has been discussed very recently by Lorentzen,* and used in determining the errors of the Bamberg heliometer. He begins by applying the method of least squares rigidly to the reduc- tion of the observations; but the numerical work required being very great, he concludes by recommending a slight depar- ture from the rigid least square method of reduction. This brings him to a method which is identical with Gill’s, both in the observations, and in the calculation of the division errors. They differ, however, in the calculation of the weights of the division errors determined for the several lines of the scale. Gill comes to the conclusion that all these division errors are determined with the same weight. This is not rigorously correct. Lorentzen does not come to the same conclusion; but as we shall see further on, he deduces a weight formula which is quite correct. 2. The method of Hansen, both in its original form, and as extended by Gill, in common with all other methods of. deter- mining division errors, has the defect just mentioned. The various lines are not determined with rigorously equal weight. Since, in general, there is no reason @ priort why we should determine some of the lines more accurately than the others, a method which will produce true equality of weights for all the lines still remains a desideratum. Now the method of Gill would appear to be observationally exhaustive, since all possi- ble combinations of spaces on the two scales have been com- pared. Nevertheless, the method may be observationally . varied, the general weight greatly increased, and even the desired equality of weights may be produced, if we vary the precision with which the several comparisons between the two scales are effected. This need not increase the labor very * Astronomische Nachrichten, 3134, 3236. Errors of a Straight Seale. 335 much, as we shall see further on. First, however, I will briefly describe Gill’s method of observation and reduction. The latter admits of demonstration without resorting to extensive mathematical formule. The scale under investigation, A, and the auxiliary scale, B, are supposed to be graduated as nearly alike as possible, and to be mounted so that one can be compared with the other. The observations are now effected in a series of separate operations, the diagrams given below showing how they would be arranged for a scale of four spaces. Operation 1. . A OP ys 2) tials 74 Coir Ce De O° B Operation 2. A One Tee Oi a # AN SO DG B Operation 3. A OF HGR Dewan A! BI) BCWMONG B Operation 4 A Ore Deiat As Odi NC ND a B Operation 5. A OD esa emaine” 6a B Operation 6. A OL 203i v4. Oh AE MP AD B Operation 7. A Or 84 OOO WO B. 336 H. Jacoby-—Determination of the Division Scale A is numbered 0, 1, 2, 3,4; and scale B is numbered é, d, c, b, a, to avoid confusion. Operation 1 consists in plac- ing scale B opposite scale A in such a way that division ¢ comes opposite 38, and d opposite 4. The microscope of the comparateur is then brought over the divisions 3 and e, and readings are taken on each of these divisions. The microscope is then brought over the divisions 4 and d, and these are also read. The operation is then repeated in the reverse order, the scales remaining unmoved all the time. If we suppose the microscope readings to decrease as the readings of scale A increase, and if we put: d, = excess of reading on division 4 over that on division d, é, = excess of reading on division 3 over that on division e, it is obvious that operation 1 furnishes the quantity : é,—d.,, or the excess of the space 3, 4 of scale A over the space e, d of scale B. In exactly the same way operation 2 furnishes the quantities : d,—c, and e,—d,, or the excesses of the spaces 2, 3 and 3, 4 overe, d oe d, ¢ respectively. Proceeding in this way, we may arrange the results of all the oper anions as in the following table: TABLE I, 0,1 1,2 2,3 3,4 (ca) Operation 1. (¢,—d.) | (d,—e.) E 5 (e,—d,) | (d,—e,) Oy “ 3. (e,—d,) (d, = Cy} (Cc, == 0p) (6,—a,) 6c 4, (d,—c¢,) (c, — Up) Ole ) ape - 5. (c,—6,) (0, —4d,) ae Le ce 6 Ce Come) Was “ 7. K, K, K, Ky It will be seen that each of the quantities in parenthesis is the excess of one of the spaces of scale A over one of the spaces of scale B. Thus the first column contains the excesses of the space 0,1 of scale A over each of the four spaces of scale B. Similarly the other columns contain the excesses of the spaces 1,2; 2,3; and 3,4. Moreover the results of each separate operation will be found in the same horizontal line of the table. The quantities K,, K,, K,, K,, are the means of the quantities above them in the columns. Thus: Errors of a Straight Scale. 337 K = 2[ (€,—d,) ar (d,—¢,) 19 (c,—6,) a (6,—a,) | It is evident that K, is the excess of the space 0, 1 over the true fourth part of scale B, since it is the mean of the excesses of 0,1 over the four constituent spaces of scale B. Consequently, if we assume scale B as standard, the space 0, 1 is too large by the quantity K,, which we may express as fol lows : Division error of line 1, scale A = K. Similarly the space 1,2 is too large by K.,, and therefore the space 0, 2 is too large by the quantity K,+K,; or Division error of line 2, scale A, = K,+K, Similarly : Division error of line 3, scale A, = K,+K,+K, Division error of line 4, scale A, = K, 4 2 +K, Since we have assumed scale B as the standard, the quantity : K +K,+K,+K, or the division error of line 4 of scale A, is really the differ- ence in length of the two scales; so if we put: K, er Bg Mi eR), K, will be the excess of the true average space of scale A over the true average space of scale B. Consequently, if we wish the division errors of scale A to be expressed in terms of that scale itself as a standard, we must write: Division error of line 1 = K,—K, Division error of line 2 = K,+K,-2K, Division error of line 3 = K,+K,+K,—3K, Division error of line 4 = K,+K,+K,+K,—4K, = 0 Table I also furnishes the corresponding division errors of scale B. If we take diagonal means as follows : ea | (e,—d,) + (¢,—d,) +(€,—d,) +.(4, ey Q. = H(d—<,) +(d,=6,) + (4,6) + (44-6) etc. ete. we have: Division error of line d= —(Q,-—K,) Division error of line ce = —(Q,+Q,—2K,) Division error of line 6 = —(Q,+Q.+Q,—3K,) Division error of line a = —(Q,+Q,+Q,+Q,—4K,) = 0 since, necessarily, ie ke KK, ——@,- 0 +079) Generalizing these formule we have for the division error of the line numbered m on scale A, supposed to have m+1 lines, numbered from 0 to n, the following: 338 H. Jacoby— Determination of the Division K,= 1 > K ] | es ( (1) Division error of line m= 2 K—mK, j 1 And for the other scale, Division error of line m = — (= Q—mK,) d 3. The above is Gill’s method, substantially in the form described in the Monthly Notices (vol. xlix, p. 110); though the way in which it is presented has been somewhat modified. The formule (1) are the same as Lorentzen’s final formule, given in Astr. Nach., 3336. The next step is to calculate the weights of the division errors determined. ‘This can be done very easily, if we notice in Table I that the sum of any number of consecutive paren- theses in the same horizontal line is known with the same pre- cision as any one parenthesis alone. Thus the quantity: (e,—d,) + (d,—e¢,) + (¢,—8,) is known with the same precision as the equivalent quantity : é,—9, This circumstance will of course affect the sums of the K’s, the mean error of CARL for instance, being by no means as great as : 4/2 (mean error of K,), as it would be if K, and K, were independent. Now let us put: | €= mean error of any of the observed quantities, as ¢,, 6,, etc., which are supposed to be all observed with equal precision. A,,= division error of line m on scale A. Sm = OmtCmtdmt ss ey so that s,, is the sum of all the letters except the first and last, after each has received the subscript m. Then we get at once from equations (1) AVEO Se ee ie nN 1 7 m+i1 Also we see from Table I that n3K = 8,—S,+ (e,+¢, + wee $lp—1)—(4,+4,+ .. . +ap) 1 Das = Sm—Sn-+ (Ent mir oss + + Opa) —( Ansa +E Onze = 2 + Gn) m+1 Substituting these values in the previous equation gives: Errors of a Straight Scale. 339 m+1 A,= "S| », +3e— 3a |+% s| se a This equation is not intended for caleulating A,,, since equa- tion (1) is more convenient for that purpose. But in the present equation, all the quantities are independent, so that it ean be used at once for getting EH, the mean error of A,,. Since e is the mean error of any one of the observed quantities, we shall have: (mean error of s as (n—1)ée (mean error of Seyramee (mean error of Se)’ = (n—m)éé (mean error of s,,)* = (n—1)éé€ etc. etc. - Consequently,. BE | [aa | | x1 ~ mm | | ee [ n—1 -- =| [a1 nm + nam | + | ec n n = a n*(a—1) +m(n—m) |~ If, therefore, we assume our unit of weight such that: coat we shall have for computing the weight P,,, of line m on scale A, the equation : ere GD) 2m oe pale: 1) (2) 1 j Pp has a maximum value fon : VL) Nh) = oy We therefore have: if $ on— se ieee ee 2 (3) The minimum value of subject of course to the condition = y] that m may only vary from 1 to n— 1, will occur when: m=lorm=n—I1 In either case : 2(7%—1)(n’?+1) n* (4) 1 Sun P (minimum) = 340 H. Jacoby—Determination of the Division The difference, therefore, is: 1 dS Sechoeee _ $(n—2)* | P (maximum) — P (minimum) = ; ai The formula obtained by Lorentzen for computing P,,, is, allow- ing for the difference of notation (Astr. Nach., 31384, Eq. 15): 2 2m = 7 —1) apes (n—12) P,, which agrees with equation (2) just obtained. But Lorentzen’s value for the weight of a quantity he calls 0, which equals the excess of the true average space of scale A over the true aver- age space of scale B, is not quite exact. In our notation this excess is —K,, and as we have seen, it is equal to: nr “ (the sum of all the parentheses in Table I). We have also seen that the sum of all the parentheses in any horizontal line of Table I, has for the square of its mean error the quantity : . 2EE And, as there are 27—1 horizontal lines, the sum of the whole table has for the square of its mean error: 2(2n—l1)ee Therefore the square of the mean error of 0 is: : 2(2n— Gn), nv and the weight of 0, ee being taken as unity, is: n* 2(2n—1) which may be written : Lorentzen gets for the same quantity the slightly incorrect — value : n° 4 In Table II, which will be given further on, are contained the maximum and minimum weights obtained in Gill’s method for various values of n. The table is computed by means of equations (8) and (4). When m is an even number, the maxi- mum weight applies to the middle line of the scale; and when Errors of a Straight Scale. 341 m is an uneven number, it applies to either of the two lines nearest the middle of the scale. 4, Having now described Gill’s method, I shall proceed to show how it may be improved very materially. For this pur- pose, let us resume the equation : A, = = S,+ “Se— 3a |+ 7 5,- yee la Now let us suppose that the various observed quantities con- tained in Table I, have been determined with different weights as follows: Weight of a, = = Weight of ¢, = a ete. etc. In other words, let us indicate the number of times each letter is observed by the reciprocal of that letter pramed. Then if we put: Sie 08 Gh eis at following the analogy of s,,, already employed, we shall have: (mean error of s,)’ = s',e€ (mean error of s,)° = $’,,€€ m—1 m=—1 (mean error of 2 e)’=«eZe’ 0 0 ete. ete. Consequently we shall have for computing the square of the mean error of line m on scale A, assuming ee = 1: (EE). = = iS (w—m)*| s+" Se + al | mm 0 1 +m'| s, +3e43a' | (5) m+1 This equation may be used in a very simple way for strength- ening Gill’s method. Let us put the weights of all the exterior quantities in Table I equal to each other, i. e., let us put: es | 1 1 1 Toei 1 1 1 1 Pei Ms CHEE, aCe Te. frre. re | CAE BIS ik Qty) Ga Nig ae and suppose any one of these quantities to be represented by 1 J —. In other words, let us observe every comparison between the two scales in which either of the four end lines appears — 342 LH. Jacoby—Determination of the Division times instead of once, and use the mean in the further compu- tations. We then have, since ce =1: Sm =n—1, 8’, = (n—1)Q, 8’, = (n—1)¢ m—1 m n—1 n = e=mg, 2a’ =mg, Ze = (n—m)q, Za' = (n—m)q, 1 m m-+1 So that if we put all the other weights equal to unity, equation (5) becomes : 1 1 1 2m Pee, (n—1) +| = (m1) +5 (n—m |g (6) This equation agrees with (2) if we put unity for g. I shall call Gill’s method, when modified in the manner just explained, method 1. In Table II given below, will be found the maximum and minimum weights belonging to method 1, placed for comparison next to those belonging to Gill’s method. In order to bring out more clearly the advantage of method 1, I have also put in the table the total number of observations required for carrying out Gill’s method, and the additional number required for method 1. This number is: For Gill’s method, (n+1)’?—2 Additional for method 1, (4n—2) S —] ) In computing the column “ Time used” I have taken as the unit the tirhe required to make one observation in Gill’s method. I have then assumed that the additional observations of method 1 each require only one-fourth as much time, since they simply consist in additional settings after the microscope, etc., have been brought into position and adjusted. In com- puting the table, I have put: which is equivalent to observing all the comparisons involving the end lines four times. It does not seem desirable to carry the number of observations beyond four for any one compari- son, since it is doubtful if the mean of a large number of observations of this kind really possesses a weight correspond- ing to that number. For the sake of comparison, I set down here the theoretical weights which would result from a rigid least square discussion of the observations in Gill’s method, according to Lorentzen. Errors of a Straight Seale. 343 . Weight. n Max. Min. 3 2°15 2°15 4 PTET 2°68 3) 3°35 3°19 6 3°99 3°69 10 6°49 5°74 It will be seen that these weights are nearly the same as those obtained in Gill’s own method of reduction. TABLE II. Z Time used | Gill Method. | Method 1. | Max. Weight.|No. of Obs. (arbitrary) Min. Wt. x 10? ” Weight. Weight. | Min. Weight. aT addit.| unit). | Time used. Pes Mai Man |) Ga ee OE ey ean acl Gabe) 1 3/ 2:03 | 2:02 | 3:45| 3-44) 1°00 | 1:00 | 14 | 30) 14 | 21:5) 14°3 | 16:0 A| 2:49 | 2-44 | 4:13] 4:10| 1°02 | 1°01 | 23 | 42 | 23 | 33°5| 10°5 | 12-2 5| 3°00 | 2°95 | 4-93] 4-88] 1:02 | 101 ; 34 | 54) 34 | 47°5| 85 | 10-2 6| 3:47 | 3:40 | 5°65| 5-62! 1:02 | 1:01 | 47 | 661 47 | 63:5] TI 8:8 i| 4:08 | 400 | 6°62) 658| 1:02 | 1°01 | 62 | 8 | 62 | 81-5] 6-4 81 8| 439 | 431 | 7-09] 7-04) 2-02 | 1:01 | 79 | 90 | 79 |101°5| 5:6 6:9 9| 5-15 | 5-05 | 8:33| 8-26| 1:02 | 1:01 | 98 | 102 | 98 |123°5| 5:1 6:7 10| 5:49 | 5-41 | 8-93| 895) 102 | 1-01 |119 | 114 {119 147-5} 4:3 6:0 15| 8:06 | 7-94 |12°90/12°82) 1:02 | 1°01 |254 | 174 |254 |297°5| 3:2 4-3 20:10:53 |10°42 |16- 86|16°78) 1:01 | 1°01 |439 | 234 [439 |497-5| 9-4 3:4 The last two columns of Table II bring out very clearly the advantage of method 1 over Gill’s. It may be objected that these columns depend upon the somewhat arbitrary assumption that the additional observations require only one-fourth the time needed for the first ones. But it should be noted that even without this assumption, these columns would show an increasing advantage for method 1, as soon as 2 becomes larger than 10 or 12. The same would be true even for the smaller - 1 : values of m if we were to put — equal to 2 or 8 instead of 4. But the assumption seems fully justified by the experience gained by Dr. Gill and others in investigating the scales of the Cape heliometer. 5. The above method 1 leaves very little to be desired on the score of variation in the weight with which the several division errors are determined. Consequently the investiga- tion of a method which will make these weights all equal is of interest chiefly from the theoretical point of view. Still an occasion might possibly arise when it would be worth while to take the extra trouble necessary to bring about this equality of weights. This can be done by the following method, which we will call method 2. Let us resume equation (5): 344 H. Jacoby —Determination of the Division 1 ; m— m n' — = n’s',,+(n—m)’ EE: pa 3a | : 0 1 ae n—1 n +m’ | sn Be +2 a! | (5) m m+1 and put m+1 for m init. This gives: 1 ; m ; m+1 n* —— = 1078 mi, + (m—m—1)’ E pte + Sa’ | | ee 0 1 n—1 n +(m+iy [s, + Se + Sal | m+1 m-+2 In order to bring about equality of weights, we have only to make : Baie ie = 1 NS Le : If we do this, and let — have the same signification as before, we get, after some simplification : 3 So = “4 [2-14 2m(n—2)] (7) This equation shows that the sum of the reciprocals of all the weights (except the top and bottom weights) in any column of Table I exceeds that in the following column by the quantity : 29 a [2—1+2m(n—2) | And if this equation be satisfied for all values of m from m=1 to m=n—2, the division errors will be determined with equal weight. If we substitute successively 1, 2, 3,.,2—2 for m in equation (7), we get a series of quantities in arithmetical pro- gression. Summing this we derive: 9 As 2¢ Sn = —f (n—1) aires [m(n—2)+1]m (8) This equation shows that s’, can never exceed 8’, by a quantity greater than 2gn; and since an inspection of Table I shows that s’,—s’,, cannot possibly be made greater than n—1, it fol- lows that 2g cannot be taken greater than unity, or g greater - than 4. We have already explained that it is not desirable to make g less than 4. It will therefore be best to put as before: dite It is obviously desirable to make: s,=n—-l or in other words, to observe once all the quantities in Table I that have the suffix 1, except the top and bottom ones ¢, Errors of a Straight Seale. 845 and a,. By means of equation (8) we can then find out how many times to observe the quantities having the suffix m. Of course we cannot apply equation (8) rigorously, as that would make it necessary to observe some of the quantities a fractional number of times. In fact, if we observe even one of the quantities in column m twice instead of once, we shall have made: $ —S', =F and we cannot possibly make it less than 4, unless we leave it zero. It will therefore be best to begin to make s'.—-S'n =F as soon as its theoretical value exceeds 4, and to continue it 4, until its theoretical value reaches 2. We can then make it 1, until its theoretical value exceeds 3, ete. ‘To do this con- veniently, we must solve equation (8) for m, after putting g=+4. This gives, finally: _V4(n—1) (n—2) + 8n?(n—2) (s',—s,) +1 —1 (9) fy 2(n—2) We shall now apply this to an example. Suppose 7 = 4, as in Table I. Equation (9) then becomes : 4/25 +256(s', —s',) —1 m m 4 from which we get: for 8 —8'm = £ = 25 ¥ SS pS m = 3, 4 We should, therefore, in Table I begin to observe one of the interior quantities twice for m= 3. For this purpose we might select either 6, or ¢,. The above method 2 renders all the weights almost exactly equal. What the common weight is can be computed from equation (5). Since the weight is always the same, we may put in equation (5): Mig NS in = We, therefore, also have: So = @—l¢ s, = (n—1)¢ m—1 n—l Brie == 1g = e'= (n—1)q 0 m Sa'= q ya'= (n—1)q 1 m—1 Consequently : NONE eel (n—1) (n*—1) | Powe [ n' Lies Z Am. Jour. Sc1.—Fourtse Series, Vou. I, No. 5.—May, 1896. 23 346 H. Jacoby—Determination of the Division which does not differ very much from the weight obtained in method 1. 6. As an example of the use of method 1, I have caused two of my students, Messrs. Schlesinger and Ling, to make a deter- mination of the non-periodic errors of one of the screws of the double microscope of the Repsold machine for measuring photographs at Columbia College Observatory. The method is admirably adapted for this purpose, and I am not aware that it has been used elsewhere for examining screws. It is merely necessary to take readings on the successive divisions of the millimeter scale of the machine, arranging them in accord with the scheme explained above. Such a set of readings is set’ down in Table A below, the number attached to each screw- reading being the weight of that reading, or the number of times the line of the scale was bisected at that point. The exterior quantities have been observed three times, instead of four, as recommended in this paper. This was done to save time. Table B is obtained from Table A by taking the dif- ferences of the successive numbers, and corresponds exactly to Table I given on page 336. In Table C the K’s are set down, and their summation gives the corrections of the screw at the points 5, 7, 9, 11, 138, 15, revolutions. In column four of the table the necessary corrections are applied to give results in terms of the screw as a standard, instead of the scale. This is done by means of equation (1). Column five contains the corrected division errors, as determined by the observer S. Columns six and seven contain the results of the observations by L. and of a second set of observations by 8. The last column gives the mean. In all these tables, unity in the fourth decimal place corresponds to one-twentieth of a micron on the scale. It will be noticed that in one case only does the differ- ence from the mean amount to half a micron, or zy4y, of an inch on the scale. This implies that in measuring any object with this microscope, we need not fear errors of greater amount than sy4y, inch, on account of the non-periodic errors of the screw. It is to be noted also that this method gives the errors of the screw without the assumption of any law of error, while in many investigations by other methods it has been customary to assume that the non-periodic screw errors could be represented by a parabolic curve. | Errors of a Straight Scale. 347 TABLE A. Horizontal argument : Scale readings. i} 7gmm | 77mm 7Gomm | 75mm | 74mm 73mm | 3 | | Operation 1 13 0027 (3) 14:9983 (3) | 2 11-0023 (3) 1370005 (1)/14-9960 (8) 13-0075 (1) 15-0038 eo) 11-0055 (1) 13°0075 (1) 15-0013 (3 8-9940 (1) 10°9985 (1) 12°9950 (1 14°9917 (3) 39-0070 (3) 11-0088 (1) 47-0062 (3) 9°0075 (1) 5 49945 (3) 69935 (1) ( ee NGI | 5:0045 (3)| 69995 (1) 9:0035 (1)/11-0035 (1) 13°0022 (3) RPS | 4-9970 (3) 69980 (1)| Socio nea (3) 638! | 50013 (3), 70015 (1), 89992 (3) we. 9) | _ 50003 (3) 7-0002 (3) TABLE B. Excess of Two Revolutions of Screw over Space on Scale. Serew. | Screw | Screw | Screw | Screw Foy ale at 9-11 | 11-13 | 13+15 Seale 78-77 —-0010 | +-0013| +0018 | —-0018 | —-0044 “ 77-16" | —-0025 +-0005 | —-0020 | —-0013 | —:0045 tt Y6=75 +0010 | +:0050 | +-0045 | +-0020 | —-0037 574 +0002 | +°0020 | —-0005 | —-0035 | —-0062 “14-73 —0001 | —0023 | —-0018 | —-0013 | — 0033 | | | | | Means —-0005 | +°0013 | +°0004 | —-0012 | —-0044 K, Ke K; | Ky, K; I ! } TABLE C. | } Obs’r. Ss la xOlotaeaeS). Obsir i: 2 Screw KN ZK —MKo Div. error | Div. error Ph set Mean | Div. error | | | 5 | ~:0000 | -0000) -0000 -0000 0000 =| 0000 “0000 7 | —-0005 |—-0005) +:0008 | +°0003 +0016 | —-0002 | +:0006 9 | +:0013 |+:0008 +-0017 | +-0025 +0018 | +:0016 | +:0020 11 | +°0004 |+-0012) +:0025 +°0037 | +°0031 ~ | +:0027 | +0032 13 | —-0012 | -0000| +0036 | +-0036 +°0036 +0033 | +:0035 15 | —0044 |—-0044) +-0044 | 0000 “0000 “0000 “0000 These division errors are to be subtracted from observed readings of the screw. Columbia University Observatory, New York. 348 T. Holm—Studies upon the Cyperacew. Art. XL.—Studies upon the Cyperacee ; by Toro. Houm. (With Plate IX.) I. On the monopodial ramification in certain North-American species of Carex. Two kinds of ramification are known to be represented in the genus Carex; the sympodial and the monopodial. The first of these is the most common, and species which exhibit this ramification show only the development of one single axis, which in its first year of growth merely develops leaves, until it finally, two or three years later, passes over into a flower- bearing stem. In this case the leaves and the flower-bearing stem are both developed from the same bud. But in the other kind of ramification, the monopodial, two special forms of buds develop, the main one of which continues to develop leaves, while the floral buds are constantly lateral. In this case there are two different axes, and species which pos- sess such kind of ramification are, therefore, biaxial. These two forms of ramification are easily recognized in our species of Carex. The sympodial shows us a central flower- bearing stem, the base of which is surrounded by more or less faded leaves from the previous year, while the monopodial shows a central leafy shoot with a number of laterally devel- oped flower-bearing stems. The terminal shoot, when purely vegetative, can continue to grow for several years, and when it finally dies off, one or more vegetative buds develop from the axils of its leaves, which grow out and repeat the same kind of ramification. It is now interesting to note, that the mono- podial ramification is only known as very rare in the genus Carex. The German botanist Wydler deserves the credit for being - the first one to observe it in Carex, and he described it as characteristic of two European species: C. digitata L. and C. ornithopoda Willd.* Some years later Alexander Braun attributed the same form of ramification to Carex pilosa Scop., C. pendula Huds. and @. strigosa Huds.t It is difficult to know, however, whether this statement of Braun is really cor- rect. Doell has, at least, contradicted him so far as concerns C. strigosa and C. pilosa,t and he states that by examining a large number of specimens of C. strzgosa, the flower-bearing stems were constantly found to be central, and that he was unable to detect any central vegetative shoot in C. pilosa. * Ueber die Achsenzahl der Gewachse, Bot. Ztg., 1844. + Das Individuum der Pflanze in seinem Verhaltniss zur Species, Abhdlg. d. k. Acad. d. Wissensch, Berlin, 1853, p. 90. t Flora von Baden, Carlsruhe, 1857. T. Holm—Studies upon the Cyperacee. 349 Celakovsky, however, has supported Braun’s opinion, and declared Doell’s statement to be incorrect in regard to @. pulosa.* The flowering stems are actually lateral in this species, but the central vegetative axis dies off very soon. Only one more species has been added to these, namely C. globularis L., by Callmé.t If we now consider our North-American species, it is sur- prising to see that the systematic authors keep perfectly silent in this matter, although Boott has illustrated it so exceedingly correct in several of his figures.¢ The monopodial ramification is very well represented in this country, and exists in several of our commonest species. lt is characteristic of all the species, which Bailey§ has enume- rated under the group “ Lawiflore” of Kunth, viz: C. lawi- flora Lam. and its varieties, C. Hlendersona Bailey, C. laxiculmis Schw., C. ptychocarpa Steud., C. digutalis Willd., C. Careyana Torr., we plantaginea Lam. and U. platyphylla Carey ; besides that it, also, exists in the remarkable species C. /raserz Andrews of the “ Physocephale.’ We have examined several speci- mens of these species and the character appears to be constant. The accompanying plate (IX, fig. 1) illustrates this ramification as shown in (. platyphylla, the specimen collected in the early spring. We see here a perennial vegetative shoot, the base of which is surrounded by a number of buds, which will develop into flower-bearing stems during the same spring. These floral buds are all developed in the axils of old leaves, which are situated on the same axis as those of the upper part of the shoot, which are now in their prime of growth. By examin- ing the axils of the younger leaves, we find, also, buds, which will develop into floral shoots one year later. It cannot, how- ever, be decided with any certainty how long the vegetative shoot can continue its growth, but it seems, at least in the species enumerated above, that it can persist for more than three years. (. digitalis ‘forms perhaps an exception, since the vegetative shoot in most of the specimens examined did not seem to have persisted for a longer time than two years. The growth of the individual is, however, as mentioned above, secured by the development of a few lateral buds next to the floral ones. There are some other species of Carex, which probably are monopodial like the above mentioned. These are, e. g., C. pubescens Muhl., C. oligocarpa Schk., C. grisea Wahl., C. gracillima Schw. ‘and C! arctata Boott. These * Pflanzenmorphologische Mittheilungen, Lotos, vol. xiv, 1864, p. 20. + Ueber zweigliedrige Sprossfolge bei den Arten der Gattung Carex, Berichte d, deutsch. botan. Gesellsch , vol. v, 1887. } Illustrations of the genus Carex, London, 1858-67. § A preliminary synopsis of North-American Carices, Proceed. Am. Acad. of Arts and Sciences, April 14, 1886. 350 T. Holm—Studies upon the Cyperacee. species do not show, however, any persistent central vegetative shoot, but the flower-bearing stems appear, nevertheless, as if they were lateral. We might suppose in these cases that the age of the vegetative shoot does not exceed one year. C. gracillima and C. arctata are perhaps to be compared with the above described (C. pilosa. A marked characteristic of these monopodial species is that the floral shoots are surrounded at their base by scale-like leaves. In the sympodial the flower-bearing stems are com- monly, if not always, surrounded by proper leaves with closed sheaths and long, green blades. The monopodial ramification is, therefore, represented in this country by a number of species, most of which are common and abundantly occurring. These species are not exclusively southern forms, since several of them extend as far north as Canada, according to Macoun.* The character of being monopodial does, therefore, not seem to be favored by any special climate ; we can only state that it seems to be especially well represented on this side the Atlan- tic, and we might perhaps feel justitied to add, that it is preva- lent among our sylvan forms. Purely vegetative shoots are, however, not exclusively connected with the monopodial rami- fication. There are among the sympodial species some, which yearly develop assimilatory shoots, which die off in the first year of their growth. Such species show, then, besides the flowering stems, a number: of very densely leaved shoots, for instance C. tribuloides Wahl. with its varieties, C. JLuskin- gumensis Schw., C. straminea Willd. and C. Sa rtwelli Dewey. These sterile shoots develop sometimes, as floured on our Plate IX (fig. 2, by 6) small lateral buds in the axils of their leaves, which, however, die off at the same time as the entire shoot itself. It is, at the same time, interesting to observe the base of the shoots, both the floral and vegetative, of C. tribuloides and C. Muskingumensis (Plate IX, figs. 2 and 3). We find here distinct internodes which are not hidden within the sheaths of leaves, but which are free to quite a considerable extent. These basal internodes are even sometimes geniculate and show the nodes almost as distinct as in the Graminee. But otherwise, as it is stated in the Manuals, etce., the basal internodes in Carex are very short and perfectly invisible on account of the densely crowded leaves. Washington, D. C., February, 1896. EXPLANATION OF PLATE IX. FIGURE 1.—Specimen of Carex platyphylla at its vernal stage, showing a central vegetative shoot and a number of lateral floral ones. Natural size. FIGURE 2.—Specimen of Carex tribulotdes, showing one fioral and one vegetative shoot. The base shows distinct internodes. About one-third of the natural size. FIGURE 3.—Base of a floral shoot of Caren Muskingumensis with long internodes and distinct nodi. Natural size. * Catalogue of Canadian plants, Part IV, Endogens, Montreal, 1888. Weed and Pirsson—Bearpaw Mountains, Montana. 351 Art. XLI.— The Bearpaw Mountains of Montana. First Paper; by WALTER HARVEY WEED and Louis V. Pirsson. [Continued from p. 301.] Beaver Creek Core. A FEW miles north of Bearpaw Peak the deeply trenched valley of Beaver Oreek cuts through an igneous center or intrusive stock of granular rock which presents a highly inter- esting example of an igneous mass which, intruded into sedi- mentary strata, has there experienced a differentiation in place, producing a gradational series of rock types. An excellent illustration of this process, already described by the authors,* occurs at Yogo Peak in the Little Belt Range, the front of the Rocky Mountain Cordillera, 120 miles to the southward, and it is interesting to find similar rock types repeated here. The brief visit made to the Beaver stock did not permit of a thorough study of the mutual relations of the different types involved, but the most basic rocks were found near the periph- ery of the intrusion, which is in accord with the usual occur- rence of such rocks. The locality is accessible by wagon road from Fort Assini- boine to the prospect claims located on several small metal- liferous veins occurring in the stock. The tract is one of gently contoured, grassy summits trenched by abrupt and deep drainage ways. The topographic relief affords no hint of the presence of the massive rock, and the slopes are not scenically attractive. The accompanying diagram (fig. 4) represents an Fig. 4. Section through intruded stock at the head of Beaver Creek, Bearpaw Mountains. ideal east-and-west section through the'stock. The intrusion is laccolithic in character, the sedimentary rocks dipping away from it in every direction, as shown in the figure. The exposed surface of the intrusion is about a mile across. The sedimen- tary rocks are highly altered and metamorphosed in the con- tact zone, and as usual these hardened strata resist erosive agencies better than the granular rock forming the higher * Weed and Pirsson, Igneous Rocks of Yogo Peak, this Journal, 1, 467, 1895. 352 Weed and Pirsson—Bearpaw Mountains, Montana. ridges and crests about the stock. The rocks of the contact zone are well exposed on the south side of the core, where they form the cliffs on either side of Beaver Creek, dipping up stream and away from the stock at 20°, which becomes but 15° farther away from the core. Detached masses of the altered sedimentary rocks are in one place found resting upon the granular rock, as shown in fig. 4. On the crest of the spur showing these detached masses the line of contact is seen to be somewhat irregular, and the igneous rock breaks through the baked sedimentaries in chimney-like masses. The western border of the stock shows the best exposures of the contact zone. Above the uniform crest of a ridge of massive rock, the slopes of altered sedimentary strata rise abruptly to a high ridge forming the divide between the waters of Beaver Creek and the stream to the west. The ridge is formed of highly altered, hardened, metamorphosed shales forming adinoles of white and light creamy tones of brown, green, lavender and pink, together with sandstones altered to dense compact quartz- ites, the beds dipping at 20° away from -the stock. The metamorphic influence of the igneous rock is noticeable in a zone about a half mile in width. The massive rock shows considerable variation in character. A rather finely granular, even-grained, syenitic rock showing equal proportions of feldspathic and of ferromagnesian min- erals, forms the main body of the mass, but this passes rapidly in some places into a dark, highly basic type, and on the north- ern boundary a,highly feldspathic variety is seen, a syenite, which as it is the most acidic type occurring at this core will be considered first. Beaver core syenite.—This variety occurs on the north side of a branch creek, from the east, near a miner’s cabin. It forms a debris pile of rudely platy blocks, is hard and tough, ringing under the hammer, and breaking with difficulty. The blocks are lichen-covered, but the rock is quite fresh though stained with iron leachings near the surface. On a freshly fractured surface the rock is seen to be fine grained, evenly granular, and compact. The color is gray, slightly stained with iron rust. The rock is essentially feldspathic, and of somewhat aplitic aspect. Close examination shows certain feldspars developed with a tabular habit which give long cleavage planes with parallel arrangement. This and the “Schlieren” seen in the field show a rude flow-structure. Examined with the lens, it is seen to be composed mainly of light-colored feldspar, with a few inconspicuous small spots of green pyroxene. Weed and Pirsson— Bearpaw Mountains, Montuna. 353 In thin section under the microscope the following minerals are found: alkali feldspar, augite, iron ore, quartz, a very little bzotite, hornblende, and titanite. The auwgite is a clear green diopside. It occurs in small anhedral grains and there is very little of it. The 27on ore is about equal in amount; of the dzotite only a few shreds have been seen. Ztanzte is rare. The feldspar forming the main constituent is entirely of the alkaline series; no plagioclase feldspars have been found. In the main it has a rude, thick, tabular habit on 0(010), giving square or rectangular cross sections whose interstices are filled by smaller grains and quartz. It shows throughout a pro- nounced and beautiful microperthite structure, more developed on the boundaries and thus fringing the sections with palisade- like edges. The remaining feldspar consists also of varying mixtures of the albite and orthoclase molecules, giving spotted, cloudy effects in polarized light. At times the interlaminated lamellee are so fine as to be seen only with high powers. The feldspars in fact present most perfectly the varying effects so well described by Brogger and by Rosenbusch in their articles on the alkali feldspars. The numerous contacts with quartz show that they have always, in all positions, a lesser index of refraction than the quartz. In a section parallel to 0(010) an obtuse positive bisectrix emerges, and in some areas measured from the trace of the good cleavage c(001) the extinction-angle is 7° plus; these are of orthoclase. In other areas it rises to 19° which are albite; these are marked by a higher birefrac- tion. Others give varying angles between them and are mix- tures; on this face the albite interlaminations follow the direc- tion of a very steep dome. In sections perpendicular to 6(010) the albite patches show often the albite twinning whose maxi- mum extinction-angle was measured at 15° on either side of the twinning line. We believe that these feldspars were once homogeneous anorthoclase and have split up into these variable masses through secondary processes. ‘The average composition is given after the analysis. The quartz fills small spaces left between the other com- ponents. It frequently contains vast quantities of slender hair-like interpositions which are indeterminable, but are believed to be of rutile like those found in certain granites. The analysis of the rock by Dr. H. N. Stokes yielded the results shown on the following page. 354 Weed and Pirsson— Bearpaw Mountains, Montana. 16 IDE S80 Ta, SiO Gil ee lags tn EB deo. GeBR 65°54 1139 DVO Oe Jas 5330, a tlie 17°81 149 WesOuajanis tia c+? 1:90 1:98 74 012 HeQ 2 apes ek "84 ZU) 1°15 "012 1026 ie eet aap Renan 7) 17 98 ‘014 CAO Rees ee "92 1°32 1°92 "016 NWO So ee 5°45 6-49 5°55 089 J5G 0 betel iia 5°62 5°76 5°58 060 UO TIO tee 15 08 HO 110) es 30 mee aA TOs CSOs ee 21 "22 11 MnO ere lis Saco eOr trace trace BaOr ork (Sees reais ‘29 not det. Sr@ Baie ene "04 "06 not det. THO ele eee none trace trace PaO b2iieek. Saree ‘13 10 trace Cy, iain en: eeRES sae 04 04 epee a ge ao i Sabi none trace ae SO aie sicls ees eed ek he ge ea "02 aes 99°95 99°97 99°92 Oat Ol 99°94 99°96 I, Quartz syenite, Beaver Creek stock, Bearpaw Mountains, H. N. Stokes anal. II. Quartz syenite porphyry, Gray Butte stock, Bearpaw Moun- tains, H. N. Stokes anal. III. Quartz syenite, Highwood Peak, Highwood Mountains, L. V. Pirsson and W. L. Mitchell anal. Ta. Molecular proportions of No. I. The low lime, iron, and magnesia show that the rock belongs in the alkaline series. A couple of analyses are quoted for comparison from the first part of this paper. In Ia@ are given the molecular proportions of the essential oxides. ‘These show some interesting features in their mutual relations and may be computed into the minerals shown in the section as follows: Ke Ons 06011 } Na,O =-:089 = 1 Ori 10339° Al,O; = -060.= 1. Orthoclase Al,0; = 089 = 1 tat Ab; 2) =41°3 SiQa) =""360==16 SiQa), = "034 = 6 Ota. 7 = "13"2 MgO =‘014=1 FeO =‘012=1 Pyr. = 311 CaO =-‘014=1}Pyroxene Fe.0; = 012 =1 t Magnetite .Magn. = 2°5. SiOp i) 028: —2 -_— Si0g = -°217% = free Quartz 100.0 The only deviation here is some 0°14 per cent of lime, which belongs in the minute amount of apatite and titanite present and of which no account has been taken. The rock is thus Weed and Pirsson—Bearpaw Mountains, Montana. 355 shown to consist of over 80 per cent of alkaline feldspar of varying mixtures, but whose average composition would be nearly Or,Ab,. It therefore stands on the boundary in the alkali series between the granites and quartz syenites; it might be termed an augite aplite or granite or a quartz syenite. It is a notice- able fact how high a silica content purely feldspathic rocks may earry (albite with SiO, = 68°7%) and in this case high silica may still yield little free quartz. _ Basie syenite or monzonite ( Yogoite).—This variety consti- tutes the main mass of the stock. It seldom occurs in prom- inent exposures, but forms talus slopes whose luxuriant cover- ing of grass conceals the angular debris blocks into which the rock breaks. This type occurs in different parts of the mass, showing identical characters. The specimen described comes from the base of the slopes east of Beaver Creek. The rock is of a medium gray color, isevenly and moderately fine granular, the average size of grain being about 1™™” in diameter. It is easily seen to be composed of a white felds- pathic component mixed with about equal quantities of a dark augite and biotite, the angite dominating the biotite in amount. - They do not appear porphyritic in development but in grains like the feldspar; more rarely the biotite is seen in somewhat larger, ragged, poikilitic plates. In thin section the rock is found to be composed of the fol- lowing minerals: Lron ore, apatite, diopside, biotite, soda-lime feldspar varying from labradorite to oligoclase, and orthoclase. The diopside is a pale green, and is apt to occur in short, stout prisms which have rough, rounded exteriors giving the mineral an anhedral habit. The dzotzte is also allotriomorphic with respect to the other minerals; it has a strong pleochroism between very pale yellow and deep olive brown. Its period of formation overlaps the pyroxene but commenced later; it is very common to find the pyroxenes with an interior zone filled with biotite shreds; within this it does not occur, and this marks the commence- ment of the biotite crystallization. The biotite frequently surrounds grains of iron ore. The plagioclase feldspar occurs in short, thick laths which are very small in size compared with the other components, and are generally quite idiomorphic. In amount it is much less than any of the other components. The small laths lie scattered around among the other components, and are enclosed in the orthoclase without orientation. It shows carlsbad, albite, and rarely pericline twinning. In composition it is somewhat vari- able, the interior parts being as usual more basic ; the interior part is labradorite, and the outer portion varies through 356 Weed and Pirsson— Bearpaw Mountains, Montana. andesine to oligoclase. Sometimes the crystals are not zonal and these may be of labradorite. Thus a section in the zone 001 on 100 perpendicular to 010 gives for the albite twins in one carlsbad half 35°, 35°, in the other half 17° and 19°, then d=17°, and the feldspar is a labradorite between Ab,An, and Ab,An,. The index of refraction is also higher than the orthoclase, in which the laths lie. Its period of formation laps that of the biotite. The orthoclase, which was the last miveral to crystallize, fills the interspaces and encloses the other components in a poiki- litic manner over considerable areas. Its determination as orthoclase rests on the fact that in numerous sections perpendi- cular to an obtuse positive bisectrix the angle of extinction varied from 5 to 7 degrees from the trace of the cleavage 001 in the obtuse angle f, the section being oriented by this cleav- age and by inclusions which give the direction of the vertical axis as well as a parting which appears to be parallel to the prismatic faces. Thus in such a section the angle was measured 63 and 8 = 63° 54’. The extinction measured with the Ber- trand ocular is 7° 30’ from the trace of the good cleavage 001 in the obtuse angle. The albite molecule is surely present in it to a certain extent, and some examples contain patches of interlaminated albite. All of the minerals are very fresh, and the structure is hyp- idiomorphie, typical of granular abyssal rocks. The differences in the mineral composition of this rock and the syenite just described are of the same kind as those noted between the two types described from Yogo Peak, but the contrast is even greater, as will be apparent from the chemical analysis. An analysis of this type, made in the laboratory of the U. 8. Geological Survey by Dr. H. N. Stokes, is given in the first column of the table on the next page. Not knowing the exact proportions in which the elements enter into the various minerals present, it is not possible to calculate their relative percentages, but it is clear that the albite molecule must be largely present in the alkali feldspars. It will be seen that the agreement between this type and the one from Yogo Peak is very satisfactory, as shown by the small differences of only about 1 per cent in SiO,, Al,O,, and MgO. The mineral composition and structure are also the same. Since our paper on Y ogo Peak was published the first half of the new edition of Rosenbusch’s ‘ Mikroskopische Physiographie ” has come into our hands, and almost simultaneously a notable work by Brégger chiefly on the rocks of Predazzo in Tyrol. The latter shows that the granular rocks there exposed form a transition group between the alkali feldspar and plagioclase groups with-all grades from those rich in feldspar to those free Weed and Pirsson— Bearpaw Mountains, Montana. 357 from it; further, the author proposes to erect a new series co- equal with the granite-syenite and diorite-gabbro series; the middle members of this series are the “ monzonites” and in minerals, structure, and chemical composition they are the equivalents of our Yogoites. My Jule II. Ta. SU) Se a ee eta 54°42 54°20 8801 AO) qh aa eietae 15°66 14:28 15°73 1520 Fe Oy aie aia 3°06 3°32 3°67 ‘0191 MeO. 2.2 see 4°76 4°13 5°40 ‘0661 Me@e eit) 21 Os 4g9 6°12 3°40 1947 (Ve Oe ll eda Mia ane! T57 UU: 8°50 "1352 INanO Pu soo. 3°60 3°44 3°07 ‘0580 HO Pi ade 4:29 4°42 0513 ROE THO 8 S21. 93 38 ae EO MO 16 2) FRO fla Viet): Tih +80 ‘40 LIC Cae gene eee trace eer as ae Cll eS eee ee ‘07 ea eave EOYs ee ee “75 “59 50 SO) et a eee trace pete sporhes IN a) «SE eee aa trace 10 "70 1652) ieee eee 24 "32 fe roi Gael an es 09 13 ? MpOeies 28 2. trace. = trace ra 100°24 100°19 100°49 OE CP ee aut 02 100°22 I. Monzonite, ‘‘ Yogoite,” or basic syenite, Beaver Creek, Bear- paw Mountains. H. N. Stokes anal. II. Monzonite, “‘ Yogoite,” or basic syenite, Yogo Peak, Little Belt Mountains, Montana. W. F. Hillebrand anal. Ill. Monzonite (Brogger, Erupt. Gest. Predazzo, 1895, p. 24). V. Schmelck anal. Ia. Molecular proportions in No. I. This is also clearly shown by analysis No. ILI, which is that of a typical monzonite. Rosenbusch,* however, regards these as being representatives of the basic syenites, or that member of the syenite family rich in the dark-colored components as its distinguishing characteristic. He calls it the monzoni type. me this type of granular rock corresponds so closely in all essential features and in chemical composition with our yogoite and the name “monzonite” has priority in date, we desire to * Mikr. Phys. Mass. Gest., 3d ed., p. 123, 1895. . 358 Weed and Pirsson—Bearpaw Mountains, Montana. withdraw the term “ yogoite” and substitute “ monzonite” in its place and thus avoid confusion in the nomenclature. In connection with the associated occurrence of this monzo- nite and the alkali syenite previously described, it is of interest to recall the keen remark of Rosenbusch* on the probable pres- ence of an alkaline syenite at Monzoni. Leaving aside then for the present Brogger’s radical proposal for the new group, the augite syenite family, taking into con- sideration the relative quantities of light and dark minerals according to our method of classificationt would be as follows: sanidinite, augite syenite (akerite and laurvikite types), monzo- nite, shonkinite, pyroxenite. This leaves out of consideration the plagioclase in the monzonite; if it is considered and Brég- ger’s monzonite group be adopted, a gap is left to be filled by some type of hitherto undescribed alkaline syenite, rich in augite and free from plagioclase. Gradational types between the monzonite and the more basic shonkinite were collected and their occurrence is quite what we should expect. Shonkinite type—The shonkinite type, the complementary rock of the acidic syenite, occurs in the outer or peripheral part of the stock. This rock crumbles readily in weathering, and good exposures donot occur. Pyritic impregnations have, however, led mining prospectors to drift into the rock, and the tunnel face and dump heaps afford quite fresh unaltered mate- rial. At the Zortman claim the rock is mottled and streaked with feldspathic material often in stringers and slender extended portions (Schlieren), and large irregular feldspar crystals occur. The normal rock is, however, free from these streakings, the specimen analyzed being entirely devoid of such material. It-is a very dark, basic-looking rock of a moderately coarse grain and strongly micaceous appearance from the light reflected from innumerable cleavages of biotite. In the description ywhich follows, the type from the Bearpaw mine referred to in the first part of this paper is included, as it differs in no essen- - tial from the present occurrence. Under the microscope it is found to consist essentially of the same minerals as the monzonite previously described, though in different proportions; they are apatite, iron ore, diopside, biotite, alkali feldspar, and in addition a very little olivine and probably a little nephelite. The apatite is present in short, stout crystals which are clear and colorless. The diopside is present in rather long, slender prisms and in short, round anhedral grains; it is of a very clear * Mikr. Phys. Mass. Gest., 3d ed., p. 124, 1895. +See this Jour., Rocks of Yogo Peak, p. 478, vol. 1, 1895. Weed and Pirsson— Bearpaw Mountains, Montana. 359 pale green, of a wide extinction angle, good cleavage, and con- tains a few inclusions of glass, biotite, and iron ore. The dzotite is of a brown, strongly pleochroic variety, which passes at times into a green one. It has the character of a granitic biotite and is not veryidiomorphic. Although so pro- | minent megascopically, it is less in amount than the pyroxene. It frequently encloses iron ore and patches of iron ore and biotite are found which are evidently recrystallizations after a former olivine. In a few cases a little olivine remainder has been foundin them. These are not like the opacite resorptions found in effusive rocks, but composed of rather coarse group- ings of the minerals. The feldspar is found in broad masses which yield wide plates in the section. Only in one specimen was the feldspar developed in thin, flat tables on 6 (010) with irregular boun- daries. The broad fields of it enclose all the other minerals in a poikilitic manner. In its composition it is wholly anortho- clase or soda-orthoclase. Thisisshown by the fact that sections perpendicular to an obtuse (2H>115°) positive bisectrix give extinction anglesin a positive sense of 9° for the Bearpaw mine type and 12° 30’ for the Beaver Creek occurrence from the trace of the cleavage c (001). The orientation of the section is given by this cleavage, and the direction of the vertical axis c by lines of altered inclusions and by a parting which appears. to be parallel to a prism face; between these the angle § was measured 68° in each case. The feldspar, moreover, presents all the characteristics of the soda-orthoclase group, which have been so frequently men- tioned in these papers and shown to be dominant in the Bear- paw rocks. It has the homogeneous appearance with low powers, and the patchy varying appearance with high ones between crossed nicols. In one or two instances plates have shown an extremely fine, delicate, scarcely perceptible twinning following the albite law. The feldspar is indeed similar to the eryptoperthite of the Norwegian alkali rocks. No trace of any lime-soda feldspar has been found in these shonkinites. It did not occur in the original type at Square Butte and was very sparingly present in the Yogo Peak variety, a noteworthy fact when the high per cent of lime is considered. The presence of a little nephelite is suspected as the rock powder yields a little gelatinous silica (not due to the olivine) on treat- ment with very dilute acid. We have not found it in the see- tions and only a mere trace can be present. _ The structure is the typical hypidiomorphic one characteris- tic of even-grained abyssal rocks. The chemical composition of the Beaver Creek type is oe in the following analysis made for us by Dr. H. N. tokes : 360 Weed and Pirsson— Bearpaw Mountains, Montana. I. II. II. Wie, \vpewe SiO, 2a 7 es oe 50°00 48°98 46°73 50°43 °8333 AYO ste pee a 9°87 “12°99. 10:05 \10@i. -0nge BeQeeewe. 13:46 2°88 3°53 ) oo Gamee hc Oene BeOe eek 501 5°77 8°20 0696 MgO by NSpes 5 eae 12:92 9°19 9°68 5°58 "2980 CR) erin See 8°31 9°65 13:22 14°82 "1484 Na,O ia SCS = 2:4 222, 1°81 1°48 ‘0388 K,O See ee 5°02 4°96 3°76 3°70 "0532 Or Si 73 1:44 ‘78 ? RO 102 Db Pong 26 me ae FEO 110° Ls Se S16 56 CHO Sates | trace ae LARS Mn@Oie ss. niin trace 08 28 a Bate INTO ss au 2) Sa 07 Eb hd 4349 BAO psa 32 ‘43 undet ? SPO ai ee es 07 08 undet ? Bile keyg Pht oe! 5 "16 ‘22 Lees gy wr CaCG ox. t a et eee nee 52 (i) pag a ow sare ‘08 rss | 18 Soe | hepa ah ‘81 ‘98 151 70 SO re ee "02 peas trace ee Oe se eee 24 La Sais f tee Thi a5 ee a trace trace trace oe ae 100:01 99.99 100°97 99.88 O=Cl & Fl... 08 08 04 99°93 99°91 100°98 I. Shonkinite, Beaver Creek, Bearpaw Mountains, H. N. Stokes anal. IJ. Shonkinite, Yogo Peak, Little Belt Mountains, Montana. W. F. Hillebrand anal. III. Shonkinite, Square Butte, Highwood Mountains, Montana, L. V. Pirsson dnal. IV. Shonkinite, Monzoni (Lemberg, Zeitschr. d. deutsch. G. G., 1872, p. 201), Lemberg anal. Ia. Molecular proportions in No. I. It will be seen in the above that the rock has all the char- acters of the type, high lime, iron, and magnesia and also high alkalies with potash dominating thesoda. We have introduced the analyses of the previous types described by us and also one of a pyroxene-orthoclase rock from Monzoni, being indebted to Brégger* for the reference. It should evidently fall under this type. Bréggert remarks that shonkinite is “a peculiar pyrox- enite rich in plagioclase” ; this is certainly true, but the same * Kruptionsfolge eruptivgesteine Predazzo, 1895, p. 66. + Loe. cit. Weed and Pirsson—Bearpaw Mountains, Montana. 361 definition would apply as well to augite-syenite, or gabbro could be called a peridotite rich in plagioclase. A pyrox- enite, according to our conception, is essentially a non-feld- spathic rock, and we desire to emphasize at- this point that in the shonkinites described by us the alkali feldspars are a constant and essential component, true pyroxenites, the practi- cally non-feldspathic member of the series, never having been observed at any of the localities. As a matter of fact, if the syenites be estimated as a series practically free from plagioclase and we adopt Brogger’s mon- zonite group as an alkali-feldspar- plagioclase series as he defines it, then the shonkinites are the syenite equivalents of the olivine monzonites, while the pyroxenites—the end terms—are alike in each, the two series approaching as the feldspars diminish until they unite in the last term, the pyroxenites. This makes the fourth occurrence of this rock type described by us from widely separated and distinct loealities in central Montana. We have material, also, from others which we hope to describe at a future date. It is of interest to note that theralite (a plagioclase-nephelite, granular rock) which shonkin- ite strongly resembles in the character and abundance of the ferro-magnesian minerals, also occurs in this region, in the Crazy Mts., but appears to be a very rare type, the shonkinite being a more common and dominating one.* Differentiation at the Beaver Creek core.—To enter into detail concerning the facts of the differentiation of igneous magmas at the Beaver Creek core and their bearing on theo- retic petrology, would be merely a repetition of the discussion given in our former papers. We have thought it best, there- fore, to merely present the analyses (p. 8362) in comparison with those at Yogo Peak and let the figures tell their own story. The lower figure i in the case of each oxide refers to the Beaver Creek, the upper one to the Yogo Peak series. At Yogo Peak the series began with a quartz-syenite por- phyry,f and next to it came the syenite. The contact, if any, between them is covered, and we cannot teli whether the quartz-syenite porphyry is a differentiation in place, as seems . *While this article was passing through the press we received a paper on “‘ Malignite, a family of basic plutonic orthoclase rocks rich in alkalies and lime,” by Prof. A. C. Lawson (Bull. Univ. California, vol. i, pp. 337-362, March, 1896). These rocks appear closely related chemically and in part mineralogically to the theralites and shonkinites, and although the author does not appear to recognize it they clearly belong in Rosenbusch’s theralitic magma series (Mass. Gesteine, 3d ed., p. 385, 1895). + The analysis of the quartz-syenite porphyry given is not that of the rock occurring at Yogo Peak but of a precisely similar type from Big Baldy Mountain, a few miles distant. It represents perfectly, however, the Yogo Peak differenti- ation. Am. Jour. Sco1.—Fourts Series, Vou. I, No. 5.—May, 1896, 24 362 Weed and Pirsson—Bearpaw Mountains, Montana. most probable, or a later intrusion. In any case it is a differ- entiated phase of the same magma. As seen in the table below, its correspondent exists at Beaver Creek, but the one corresponding to the syenite is wanting. It probably exists, but was not observed in the hasty examination we have been able to make. Quartz Monzonite : Syenite. Syenite. ‘“'Y ogoite.”’ Shonkinite. $0 § 67:04 61°15 54-49 48°49 | 68-23 ska 52°81 50:00 15°25 15°07 14:28 12°29 Al,0, 15°12 ars 15°66 9°87 1°69 2°03 3°32 2°88 Fe,0, 1:90 uae 3°06 3°46 1°13 2°25 4°13 5°77 ae 84 iW 4°76 5°01 1°75 3°67 6-12 9°91 i 54 aT 4°99 11°92 2°17 4°61 7°72 9°65 Cae 92 AX 757 8°31 4:09 4°35 3-44 2°22 Na,O 5°30 posi 3°60 2°41 : 5°10 4°50 4°22 4°96 K,0 5°57 ator 4°84 5°02 The other types are clearly shown. In general a compari- son of the two localities shows the Beaver core as the more alkaline of the two; and the most important difference is that at Beaver Creek; the Al,O, and MgO show a much greater degree of concentration and inverse variation than at Yogo. This inverse variation is especially marked in the Beaver Creek monzonite and shonkinite, which otherwise do not greatly differ from one another. They thus furnish a most instructive example of the fact that a nearly similar silica percentage may exist in two types of rocks which differ greatly in appearance, one with high AJ,O, being clearly feldspathic and recalling many medium diorites in appearance, while the other, with high MgO, is dark, basic-looking, and with evidently prepon- derant ferromagnesian minerals. : There are a number of interesting dike rocks connected with the Beaver stock; these have been studied, and an account of them, together with some other occurrences of rare types of both intrusive and extrusive rocks in the Bearpaw Mts., will be shortly published in a second paper in this Journal. Washington and New Haven, March, 1896. UC. Lea—Réntgen Rays not present in Sunlight. 368 Art. XLII.—Réntgen Rays not Present in Sunlight; by M. Carey LEA. Ir Prof. Rontgen’s views as to the nature of the X-rays are correct, it would seem that they ought to be found amongst the many forms of radiant energy received from the sun, and various observers have thought that they sofound them. Some experiments, the most important of which will be here briefly stated, do not seem to support this opinion. 1. A very sensitive dry plate (S. 27) was placed between the leaves of a book so that 100 leaves and the red paper cover should be between the sensitive film and the sunlight. The book was then packed in a box frame to exclude all light from the sides. A large and thick lead star was then fastened on the outside of the book and the arrangement was exposed to exceptionally bright sunshine from 11 A. M. to sunset, March 7. The plate when put into a developing bath behaved as if unexposed. A prolonged development did not bring out a trace of an image of the lead star. It will be remembered that Prof. Rontgen found that the X-rays penetrated easily through a book of 1000 printed pages. Indeed G. Moreau has recently stated that in his hands the X-rays had penetrated through “ several meters” of cardboard.* So that the above experiment seems to be very significant. 2. A piece of sheet aluminium 1:2"™ thick was accurately fitted into a frame. A very sensitive plate was placed behind it and a lead star in front. With three hours’ exposure not a trace of an image conld be obtained. This experiment was varied by substituting thin aluminium foil for the plate, also by using bromide paper as the sensitive surface. No images in any case were obtained. 3. The sun’s rays or some portion of its radiation passes readily through wood, if the latter is not too thick. Thus through a piece of white pine 8, of an inch thick, images that could readily be developed were obtained by three minutes exposure toafternoon sunlight. With half an hour’s exposure the images were brilliant. | A panel about 12 inches square was removed from an inside shutter and replaced with a piece of white pine + inch thick. When the room was thoroughly darkened, reddish light could be seen to pass through the board. So that wood of this thick- ness is plainly translucent to the sight. The sun’s light may be examined for X-rays also by fluores- cence. *C. R., cxxii, p. 238: quoted Chem. News, Feb. 21, 1896, p. 85, (No. 1891.) 364 M. 0. Lea—Réntgen Rays not present in Sunlight. 4. The panel just described was replaced by one of stout book board. With the sun shining on this book board directly and not through glass, paper marked with a saturated solution of barium platinocyanide exhibited no indications of fluores- cence when placed behind the board. 5. Three thicknesses of Bristol board were pasted together, a circle was cut out, to one side of which barium platinocyanide was applied. The circle was then placed in a pasteboard tube, (an arrangement, I believe, proposed by Prof. Magie.) When the sun was looked at through this tube the barium salt exhib- ited fluorescence. But the interposition between the card and the sun of very thin aluminium foil sufficed to cut off the fluo- rescence. These concurrent results seem to indicate the absence of X-rays from sunlight. Charles Henry* quotes an opinion of H. Poinearé that all bodies’ whose phosphorescence is sufficiently intense, emit in addition to luminous rays the X-rays of Rontgen, whatever may be the cause of their fluorescence. Henry quotes confirma- tory experiments of his own made with zine sulphide. It seemed worth while to ascertain if this principle is of general application. A dilute solution of uranin was exposed to sunlight, using a large surface of solution so as to get the best effect. A short distance over the surface was placed a sensitive film protected by aluminium foil 4, of a millimeter in thickness and with a lead star interposed. Two hours exposure gave no result. The experiment was repeated with acid solution of quinine, with which five hours exposure gave no result. I have also examined the Welsbach light for X-rays. This hight is usually burned under a chimney which increases the brightness but interposes glass between the source of light and the sensitive film. Even without a chimney the light is bright. The experiment was therefore made both ways. No X-rays could be detected. Nothing capable of passing through alumi- nium foil =, of a millimeter in thickness by five hours exposure to the uncovered flame. * (©. R. cxxii, 312; Chem. News, Feb. 28, 1896, p. 98. W. B. Clark—Potomac River Section, ete. 365 Art. XLIil.—TZhe Potomac Liver Section of the Middle Atlantic Coast Eocene; by WM. BULLOCK CLARK. Our knowledge of the Tertiary geology and paleontology of the middle Atlantic slope has been largely increased since the days of Conrad and Rogers, yet few fields afford better oppor- tunities for continued observation, and in none is there greater need of a careful revision of results. Very divergent opinions have prevailed and to-day find expression in the different interpretations of the data. General Features of the Formation. The Eocene strata of the middle Atlantic slope, described by Darton* under the name of the Pamunkey Formation, forma belt of varying width extending from northeast to southwest somewhat to the west of the center of the coastal plain. This belt has been traced almost continuously from the southern portion of Neweastle County, Delaware, to the valley of the Nottaway river in southern Virginia, and although at times buried beneath later deposits it presents fine exposures along all the leading stream channels, while not infrequently broad expanses of the formation outcrop at the surface in the inter- vening country. Lithologie Characters.—The deposits are typically glauco- nitic and are found in their unweathered state either as dark gray or green sands or clays. The glauconite varies in amount all the way from very nearly pure beds of that sub- stance to deposits in which the arenaceous and argillaceous elements predominate, although the strata are generally very homogeneous for considerable thicknesses. At some horizons the shells of organisms are found commingled with the glauco- nitic materials in such numbers as largely to make up the beds, producing what is known as a green-sand marl. These beds are at times indurated, forming true limestone bands. This latter phase is seen typically developed both at Fort Wash- ington, Maryland, and Agquia Creek, Virginia, interstratified with the unconsolidated green-sand layers. When the glauconite is weathered the deposits lose their characteristic gray or green color, and generally become lighter gray with reddish or reddish-brown streaks or bands, or may become entirely of the latter color. In this condition they are often cemented into a ferruginous sandstone. This change * Bull. Geol. Soc. Amer., vol. ii, p. 411, 1891. 366 W. B. Clark— Potomac River Section particularly characterizes the Eocene deposits of Delaware and the eastern shore of Maryland as well as of Anne Arundel County on the western side of the Chesapeake. .A very coarse phase of the consolidated sand-rock is seen at Mt. Misery on the Severn river. In the less completely weathered portions of the formation the change is shown in the mottled yellow and brown character of the more superficial beds, the glauconitice grains still show- ing their green color when crushed. Thin iron crusts at times appear under these conditions. When the glauconite is largely or more rarely entirely absent, the deposits consist of black or grey sands or clays, the latter at times micaceous and in a few instances carbonaceous. A pebble bed has been found at some localities at the base of the formation. Sirike, Dip and Thickness.—The strike of the Eocene deposits in Delaware and Maryland is approximately N.E. to S.W. while in Virginia the prevailing trend is more nearly N.to 8. This change in the direction of strike is attained in the area between the Patuxent and Rappahannock rivers, chiefly in the Potomac basin. The dip of the strata differs in the various portions of the area, aS shown by section-lines and well-borings, from 10 to 20 feet to the mile, but along the Potomac river section, where detailed measurements were made by the writer, it is on an average about 123 feet to the mile. The results obtained froma study of the various section-lines and well-borings show that the average thickness of the deposits is somewhere in the vicinity of 200 feet, although estimates based upon the Potomac river section as well as upon well borings in the area to the east of Fredericksburg show that it reaches quite 300 feet in that portion of the middle Atlantic slope. Potomac River Section. The most extensive section of the Eocene in the middle Atlantic slope is found in the valley of the Potomac river, a nearly complete sequence of the several members of the forma- tion being exhibited in the bluffs between Aquia Creek, Stafford County, Virginia and Pope’s Creek, Charles County, Maryland. Detailed stratigraphy.—The accompanying columnar section shows the relative thickness and character of the deposits in thisarea. The several members of the formation are numbered in ascending order. The full thickness is about 300 feet. of the Middle Atlantic Coast Hocene. No. 1. The thickness of the Eocene beneath the basal strata of the Aquia Creek bluff, cannot fall far short of 60 feet. Some exposures are seen in the ravines to the west of the bluff, but. no complete sequence of the beds has been found. The almost entire absence of fossils renders it impossible to say anything regarding the faunal relations of the strata. The deposits are typical green-sands, at times somewhat argillaceous and with a basal pebble-bed over- lying the Cretaceous at several points. ~ No. 2. This bed is composed of dark uncon- solidated green-sand packed chiefly with the shells of Crassatella aleformis, Dosiniopsis lenticularis and Cytherea ovata. The bed is about 12 feet in thickness at the upper end of the Aquia Creek bluff, but gradually declines in elevation until it passes below water level about half way to Marlborough Pt. The same bed appears at water level on the opposite side of the Potomac river at Clifton Beach. No. 3. The green-sand marl composing this bed is generally indurated so as to form a firm band, 2 to 3 feet in thickness. The limestone is highly glaucdnitic and of dark color, and is filled with the shells, or more commonly the easts of the same species as the previous bed, together with Ostrea compressirostra and here and there a specimen of Turritella mortoni. No. 4. This bed is a typical unconsolidated green-sand containing large numbers of the same forms as No. 3. It reaches about eight feet in thickness. ‘ No. 5. This limestone bed is very persistent, and forms a conspicuous ledge along the face of the Aquia Oreek bluff until it passes below tide level near its eastern extremity. It is commonly about two feet in thickness"and is packed with fossils among which the forms mentioned below are conspicuous, in addition to the species already mentioned as characteristic for zones 2 to 5 which still remain the most com- mon types, viz: Pholadomya marylandica, Panopea elongata, Tellina virginiana, Pholas (¢) petrosa, Fusus sp., Caricella sp., ete. No. 6. The dark characteristic green-sand overlying the limestone ledge is packed with the 17 16 367 Woodstock Fauna Aquia Cr. Fauna. Potomac River Section of the Middle Atlantic Coast Eocene. 368 W. B. Clark—Potomac River Section common species of the previous zones. In the main portion of the Aquia Creek bluff this bed is only one foot in thickness, but thickens to the eastward, and just above Marlborough Pt. contains among other forms several species of corals, including Hupsam- mia elaborata, Turbinolia acuticostata, and Paracyathus (2) clarkeanus. _ No. 7. Overlying the preceding bed, and really a continua- tion of it, is a zone in which the fossils are few in number and much broken. This bed is about seven feet in thickness, No. 8. The highly characteristic green-sands and green-sand marls of the previous zones are succeeded by a bed some thirty feet in thickness, in which the glauconitic grains have been exten- sively weathered, giving the strata a ereenish -grey appearance which changes to a reddish-brown in the upper layers. Several irregular bands packed with Zurritella mortoné are present both in the Agquia Creek and Potomac Creek sections, while associated with that species at both localities are Z’ urritella humerosa, Cucullea gigantea, Crassatella aleformis, Ostrea compresst- rostra and other forms. The upper portions of this bed have afforded most of the species from the Potomac Creek bluff. This zone forms the base of that bluff, while it is more than thirty feet above water level in the Aquia Creek section three miles above. No. 9. The thick-bedded limestone layers which compose this zone are almost exclusively made up of the shells of Zwrra- tella mortoni, forming a true Turritella rock. Between the in- durated bands are layers of unconsolidated and much-weathered green-sand which contain very few fossils of any description. Great masses of this Turritella rock strew the shore at the base of both the Aquia Creek and Potomac Creek bluffs. The bed is about ten feet in thickness in the Aquia Creek bluff, but reaches fourteen feet in places in the Potomac Creek section. No. 10. The greenish-grey sand overlying the Turritella bed is more argillaceous than the underlying beds of the Eocene. The glauconite grains have been much weathered and nearly all trace of the shell substance removed from the few forms recognized. The casts found at the Potomac Creek bluff are chiefly those of a Cytherea. No fossils were found at the Aquia Creek bluff. The bed is about twenty-five feet in thickness. No. 11. A thin highly indurated layer of argillaceous green- sand overlies No. 10 in the Potomac Creek bluff, and among several indeterminable casts a few specimens of Venericardia planicosta were found. No. 12. This bed of greenish-grey argillaceous sand still shows some unweathered grains of glauconite, but is devoid of fossils so far as observed. It reaches eight feet in thickness. of the Middle Atlantic Coast Hocene. 369 No. 13. This bed consists of a light grey glauconitic sand somewhat weathered and filled with shells of Venericardia planicosta. It reaches three feet in thickness. This zone is very persistent and has been found outcropping in several of the ravines to the east and south of Potomac Creek, as well as in the bluff upon the river front. No. 14. Overlying the Venericardia layer is a bed of green- ish-grey argillaceous sand some four feet in thickness, which contains a great number of bands filled with gypsum crystals No fossils were observed. No. 15. This bed consists of greenish-grey argillaceous sand in which the glauconite grains have been extensively weathered. No fossils were found. The bed has a thickness of twenty- five feet. No. 16. In this zone have been placed the green-sand strata intervening between the upper layers of the Eocene in the Potomac Creek section and the base of the Woodstock section. - Comparatively little is known regarding this portion of the series, as no satisfactory outcrops appear on the river bluffs, although the strata are found in an unfossiliferous condition in some of the ravines to the west of the Woodstock area. The estimated thickness of these beds is fifty feet. ; No. 17. The highly glauconitic beds at Woodstock, Virginia and Pope’s Creek, Maryland are grouped together in one zone, as no satisfactory separation could be made. The deposits are very homogeneous, although an inconstant indurated layer appears about six feet above the base of the Woodstock sec- tion with a band of Venerzcardia planicosta below it, while a thin bed of Ostrea selleformis also occurs in the lower part of the zone, althongh evidently not always at the same horizon. Otherwise the fossils are the same throughout, so far as observed. The most common forms are Protocardia virginiana, Cytherea - subumpressa, Corbula nusuta, Corbula oniscus, Ostrea selle- Jormis, Pectunculus idoneus, Leda improcera, Leda parva, Nucula magnifica, Lucina dartoni, Lucina uhleri, Lucina whiter, Ringicula dalli and Cy lichna venusta. Paleontological character oe —The paleontological charac- teristics of the several zones indicate two very distinct faunal stages in the middle Atlantic slope Eocene, the first typically developed in zones 2 to 9, and the second in zone 17. The characteristics of zone 1 and of zones 10 to 16 cannot be readily made out in the Potomac River area on account of the extensive weathering of the beds, although in some of the adjoining districts there is an intermingling of some of the forms of the two stages in the beds intervening between 9 and 17. To the two faunal divisions, the names of Aguwia Creek Stage and Woodstock Stage have been already given by the writer.* * Johns Hopkins Univ. Circulars, vol. xv, p. 3, 1&95. 370 W. B. Clark— Potomac River Section Aquia Creek Stage.-—The most common species of the Aquia Creek stage are Zurritella mortoni confined chiefly to zones 8 and 9, Cytherea ovata and Crassatella alwformis confined largely to zones 2 to 7, Cucullea gigantea mainly found in zone 8, Ostrea compressirostra* most common in zones 6 and 7, and Dosiniopsis lenticularis, for the most part limited to zones 2 to 5. Other species found in this stage are ZLhecachampsa marylandica, Trionyx virginiana, Ischyrhiza (¢) radiata, Myliobatis copeanus, Carcharodon polygurus, Lamna (?) obliqua, Oxyrhina hastalis, Odontaspis elegans, Galeocerdo contortus, Nautilus sp., Tornatella bella, Pleurotoma harrisi, Volutilithes (Athleta) tuwomeyt, Volutila- thes sp., Caricella sp. Mitra marylandica, Pyropsis sp., Frusus (Levifusus) trabeatus, Fusus (Strepsidura) perlatus, Fusus sp., Fulgur argutus, Lunatia marylandica, Natica cliftonensis, Turritella humerosa,. Calyptrea trochiformis, ermetus sp., Scala virginiana, Gibbula glandula, Solarium sp., Gastrochena sp., Pholas(?) petrosa, Coralliophaga bryant, Tellina williams, Panopea elongata, Pholadomya mary- landica, Lucina aquiana, Venericardia planicosta, Crassa- tella aquiana, Leda protexta, Modiola potomacensis, Pecten johnsoni, Pecten sp., Ostrea sp., Serpula sp., Paracyathus (?) clarkeanus, Turbinolia acuticostata, Hupsammia elaborata, besides many species of Foraminifera. Woodstock stage-—The Woodstock stage is characterized by the following common species, viz: Protocardia virginiana, Cytherea subsmpresst, Corbula nasuta, Corbula oniscus, Ostrea sellefornis, Pectunculus idoneus, Leda improcera, Leda parva, Nucula magnifica, Lucina dartoni, Lucina uhleri, Lucina whiter, Venericardia planicosta, Ringicula dalla and Cylichna venusta. Among other species found at this horizon may be mentioned Carcharodon polygurus, Lamna (¢) obliqua, Oxyrhina hastalis, Odontaspis elegans, Galeocerdo contortus, Cythere sp. Mangilia (Pleurotomella) bellistriata, Fusus (Levifusus) trabeatus, Lunatia marylandica, Cadulus bellulus, Teredo virginiana, Solemya petricoloides, Corbula aldrichi, Tellina virginiana, Cytherea ovata, Diplodonta hopkinsiensis, Yoldia cultelliformis, Modiola potomacensis, Pecten rogersi, together with numerous species of Foraminifera. | Correlation of the Deposits.— By common consent the diversified and extensive Eocene deposits of the Gulf area have come to be regarded as the type for the eastern border region and the various Eocene deposits of the Atlantic coast states have been assigned to a position in this series, although very different limits have been given by the different authori- ties. The Eocene deposits of the middle Atlantic slope have * Several immature oysters found at this horizon bear a strong resemblance to the young of Ostrea selleeformis. of the Middle Atlantic Coast Eocene. 371 been regarded by some to represent a single minor division of the Gulf section, while others have regarded them as an equiva- lent of a larger portion of that series. The latter conclusion seems to the writer, after a consideration of both the geological and paleontological data, to be the only tenable position. In the past too little attention has been given to the geological phenomena, while, at the same time, the knowledge of the fossils has been wholly insufficient for a proper interpretation of the faunal characteristics of the formation. The Geological Criteria.—The lithological and stratigraph- ical characteristics of the Eocene in the middle Atlantic slope afford some important criteria for the correlation of the strata. To begin with, the homogeneous nature of the deposits is a characteristic feature, indicating conditions throughout the period of Eocene deposition, undisturbed by important phys- ical changes. Again, the fact that the strata are so largely made up of secondary materials shows that the position of accumulation was in the vicinity of a coast reached by no large rivers bearing sediment, while at the same time sufficiently removed from the coast line to be unaffected by shore condi- tions. It is further evident that these deposits, which are so largely glauconitic, were accumulated with exceeding slowness, as has been shown in the case of the formation of greensands upon the beds of existing seas. » Now when we compare these conditions of accumulation in the middle Atlantic slope with the conditions that prevailed in the Gulf region, marked differences appear. In the latter area numerous rivers, draining the interior of the continent, dis- charged great quantities of material throughout much of Eocene time, making the deposits highly diversified. Instead of the green-sands and greenish and black clays of the middle Atlantic slope, which no longer to any large extent characterize the strata, are found coarser beds of sand and clay, often partly calcareous, which give every indication of more rapid accumu- lation. To compare, therefore, the 200 feet and more of green sands and clays of the middle Atlantic slope with one or two subdivisions of hardly equal thickness in the Gulf region would, even upon stratigraphical grounds without the aid of fossils, hardly be attempted. The strata of the Middle Atlantic slope must be represented in the Gulf by deposits many times their thickness. The general relations of the strata, occurring as they do between the Cretaceous and Neocene along both the Atlantic and Gulf coasts, give some indications of the continental move- ments to which each province was subjected. Although the movements may not have been absolutely contemporaneous they afford nevertheless satisfactory criteria for the broad correlation of the deposits, their more exact parallelism being determined upon other grounds. 372 W. B. Clark—Potomac River Section Paleontological Criteria.— Although life-zones are frequently of great extent and may be accepted as the most trustworthy evidence of geological contemporaniety, yet the subdivisions of a fauna recognized in one area under one set of physical con- ditions may not be found in another area, distant from the first, where the conditions are wholly different. It is scarcely to be expected that the vertical range of the species will be the same in the two regions, while the time occupied in migration is a factor that cannot be ignored in most classes of organisms. Forms, likewise, which, from their persistence under one set of physical conditions, may be regarded as typical, are often entirely wanting in an adjacent province. ‘The presence also of a large number of new species is of itself evidence of change in physi- cal surroundings, and renders it necessary to proceed with great caution when detailed correlations of the strata are attempted. Especially is this true when the areas are widely separated in latitude so that temperature differences occur. When we come now to compare the faunal charactisties of the Eocene of the middle Atlantic slope with those of the Gulf we find first of all that the assemblage of forms is very different in the two areas. The great majority of the species is unlike, while the identical forms are mainly of wide vertical range. Most of those regarded as the same also show certain dif- ferences, as the result of the dissimilar conditions under which they lived, so that the determination of the middle Atlantie coast forms often involves certain doubts as to their identity with Gulf species. The sequence of fornis is likewise different, a differentiation into the great number of subdivisions recog- nized in the Gulf, not occurring in Maryland and Virginia. The Aquia Creek fauna which is typically developed in zones 2 to 9 in the Potomac area occupies, so far as can be with certainty determined, only about 70 feet of strata some 60 feet from the base of the formation and contains among other Gulf species Zurritella mortoni, T. humerosa, Tornatella bella Volutilithes (Athleta) tuomeyi, Fusus (Strepsidura) perlatus, Dosiniopsis lenticularis, Venericardia planicosta, Cucullea gigantea, and Ostrea compressirostra. The general aspect of this assemblage is Lzgnitic, some of the forms being found in the Gulf area in the middle, or in the middle and lower mem- bers of that division, while others range into its upper portions, and are also found at higher horizons. At the same time quite 60 feet of strata are found beneath the Aquia Creek fossilif- erous beds in which as yet only a few indistinct casts of Zurri- tella sp. have been observed. If the fossiliferous zones represent approximately the middle, or the middle and upper Lignitie, this lower zone (1) may he regarded in a general way as the equiva- lent of the lower Lignitic. — The Woodstock fauna, which is typically represented in of the Middle Atlantic Coast Eocene. 373 zone 17, embraces so far as can be determined about 45 feet at thetopof theseries. It containsamong other Gulf forms, Husus (Levifusus) trabeatus, Solemya petricoloides, Corbula nasuta, Corbula oniscus, Venericardia planicosta, Nucula magnifica and Ostrea sellwformis, the latter species increasing in number toward southern Virginia and affording thick beds on the Pamunkey and James rivers. At the same time the common forms of the Aquia Creek stage are wanting. Although not pos- sessing the number of distinctive species found in the precedin divisions, the Woodstock stage is nevertheless in all probability the representative of the Clazborne of the Gulf, showing a closer parallelism, perhaps, with the beds beneath the fossiliferous sands than with the upper horizon of that division. Between the fossiliferous beds carrying the faunas of the two stages are very nearly 125 feet of strata in which few fossils have been found outside of Venericardia planicosta. Many of the beds seem to be wholly barren of organic remains, while in others only a few indeterminable casts appear. No satisfactory paleontological data for correlation are therefore afforded by these deposits. If now the Aquia Creek fauna should be held to be suffi- ciently similar to the Bells Landing fauna of the Gulf to warrant its restriction to that sub-stage (middle Lignitic); and the Ostrea selleformis bed of the Woodstock stage, the exact equivalent of the Ostrea selleformis zone of the Claiborne (middle Claiborne), then we find the 600 feet between those horizons in central Alabama represented by only 125 feet in the middle Atlantic slope and perhaps by considerably less. The representatives of the Woods Bluff and Hatchetigbee stages of the Lignitic together with the Buhrstone and lower portion of the Claiborne would thus be here included. The upper beds of the Woodstock stage might then be regarded per- haps as the equivalent of the upper Claiborne, while the 60 feet below the Aquia Creek fossiliferous beds would approximately represent the earlier portions of the Lignitic. As these lower beds are much more glauconitic than the beds above the Aquia Creek stage, they doubtless accumulated more slowly. It is apparent, however, that the sequence of organic remains in the middle Atlantic coast Eocene does not afford the neces- sary data for a detailed parallelism of the subdivisions of that area with the Gulf stages. It seems altogether probable that the Pamunkey formation is the equivalent in a broad way only of the lower and middle divisions of the Eocene of the Gulf, and may even represent portions of the upper division as well. Regarding the latter reference there is little paleontol- ogical evidence, but undoubtedly less change in faunal devel- opment would be produced under the stable conditions that prevailed in Eocene time in the middle Atlantic slope than in the Gulf, so that the more highly developed fauna of the 374 W. B. Clark— Potomac River Section, ete. upper portion of the series in the latter area may have existed contemporaneously with older forms outside the region. With- out a much fuller knowledge of the characteristics of the Eocene fauna in the intermediate district, this cannot be defi- nitely determined, although it seems highly probable. — Considering all the facts, the writer is strongly of the opin- ion that the Eocene deposits of the middle Atlantic slope represent the greater portion of the Eocene series of the Gulf, its upper members alone excepted. Compared with the sec- tion recognized by Prof. E. A. Smith, in the Alabama area, it undoubtedly comprises all or the major part of the Lignitic, Buhrstone and Claiborne and, possibly, also portions of higher horizons. This reference does not, however, necessarily involve the assumption that the basal beds of the Potomac section are the exact equivalent of the basal beds of the Lignitic, since deposi- tion may have commenced in the one area somewhat earlier than in the other, although the difference was probably not reat. Conclusions.—1. The Eocene deposits of the middle Atlantic slope constitute a single geological unit already described under the name of the Pamunkey formation. 2. The deposits are remarkably homogeneous, consisting typically of glauconitic sands and clays which reach a thickness of nearly 300 feet. 3. Two clearly defined faunal zones are found, viz: the Aquia Creek Stage and the Woodstock: Stage. | 4. The Aquia Creek fossiliferous zone is approximately middle, or middle and upper Lignitic, the Woodstock zone middle, or middle and upper Claiborne. If restricted respec- tively to the Bells Landing sub-stage of the Lignitic and the Ostrea selleeformis bed of the Claiborne, as seems hardly prob- able for the reasons above cited, the 600 feet between those zones in the central Alabama area would then be represented by only 125 feet or perhaps considerably less in the Potomac region. The upper beds of the Woodstock stage might then perhaps represent the upper portion of the Claiborne while the beds below the Agquia Creek fossiliferous zone would stand as the approximate equivalent of the lower Lignitic, with- out however necessarily assuming that the basal beds of the Potomac section are the exact equivalents of the basal beds of the Lignitie. 5. The middle Atlantic slope Eocene, therefore, répresents in a broad way all or the major part of the Lignitic, Buhr- stone and Claiborne of Smith, and, when the physical conditions affecting range and migration of species are considered, perhaps even more. Both the geological and paleontological criteria are wholly inadequate for establishing the great number of local subdivisions recognized in the Gulf area, and in fact the sequence of forms indicates that no such differentiation of the fauna took place. HLS. Washington—Ischian Trachytes. 375 Art. XLIV.—On some Ischian Trachytes; by HEnry S&S. W ASHINGTON. In the fall of 1894 I had occasion to visit Ischia, in the Bay of Naples, and collect the representative trachytes of the island. Most of these are too well known to need descrip- tion, but some specimens from Mt. Rotaro showed a certain rather rare structure in such perfection that an account of them seems not unworthy of publication. Mt. Rotaro is a small volcanic cone in the northeastern part of the island, some 315 meters high,* and with a well-preserved crater 107 meters deep. As far as can be seen it is composed of fragmentary material, scorie and blocks of trachyte and obsidian, which show stratification in places. The whole rests on marls and clays containing late Tertiary fossils:+ The ridge of Mt. Tabor,t composed of the well known sodalite trachyte, is a lava-stream of this small voleano which has flowed to the north. According to Fuchs,§ Mt. Rotaro is the site of the great eruption of the first half of the fourth eentury B. C., which drove the inhabitants to the mainland. The trachytes of Mt. Rotaro are of two kinds. Oneisa not very compact, light grayish-brown rock with phenocrysts of orthoclase and augite. It occurs in relatively small quan- tity, and under the microscope presents no features of special interest here; large phenocrysts lying in a typically trachytie holoerystalline groundmass, which shows a marked flow-struc- ture among its component feldspar laths. Corresponding to this is a hght brown obsidian, extremely brittle and somewhat vesicular, showing some orthoclase phenocrysts. Under the microscope it presents a clear, light brown glass, with rare green biotite and augite, and glassy sanidine phenocrysts, with a few feldspar microlites. Many air cavities are present. The black trachytes and obsidians are of greater interest. For reasons which will be seen we shall begin their descrip- tion with the obsidians. No. 524| is coal black, with a vitre- ous luster, very brittle, and shows numerous stout phenocrysts of glassy sanidine. An analysis by Fuchs] of a similar black obsidian from Mt. Rotaro is here inserted : * This is the height given on the Government Topographical Maps; Fuchs gives 301°4, and Fonseca 277-4. + Fonseca, Geologia dell’Isola d’Ischia, Florence 1870, 25. t+ vom Rath, Zeitschr. d.d geol. Gesellsch. xviii, 628, 1866. S Fuchs, Isola d’Ischia. Florence 1873, 47; also, Tsch. Min. Mitth., 1872, 237, || The numbers are those of specimens in my collection. “| Fuchs, op. cit., 40. 376 H. §. Washington—TIschian Trachytes. SIO Beet ahs oi ie 60°77 A ONES set 06) uy on cr 19°83 Me, Oi) eeu ses... dels so KeQ, - p22 Su ee pags} IMO Ge Ue ot oe le ee ie Trace CaO 1 vs 6 a etapa Be OS re MOO cc oe ee 0°34 INBEO) op cee ee 4:90 TOO eee ose eee 6:27 Pon 3 Be 0-002 Ponit... 2202 22a ee 0°24 100°55 Under the microscope the largely predominating, clear glass basis shows a rather light coffee color. The phenocrysts of sanidine are clear and not generally twinned. Inclusions are uncommon and consist of glass, magnetite and an occasional bio- tite, zircon or apatite. They are quite constantly surrounded by a narrow zone of deep brown glass, in which lie small prismatic microlites of orthoclase, generally at right angles to the domal and basal planes of the erystal. Besides these large feldspars are seen a few smaller phenocrysts of magnetite, green augite and biotite. Scattered through the basis, often sporadically, but generally in clusters or irregular streaks, are small prismatic orthoclase crystals, which seldom exceed 0:05™" in length, and are gen- erally only 0°92 "™ long. These are usually simple, twinning not having been seen, and only rare cases of forked forms or those brushy at the end. They also are surrounded by a border of darker substance, or he in large irregular patches or long streaks of this, where they occur together. This substance has no appreciable action on polarized light and shows but few signs of definite strue- ture, though a tendency toward the formation of small spheroidal masses is noticeable under high powers, and is rendered more evident by the presence of a little globulitic dust. When we examine the next less glassy specimen (No. 553) we find that, while the phenocrysts remain similar in general characters, the narrow dark border has disappeared and the feldspar needles have increased in number, giving rise to a fringe of fine orthoclase needles, separated from each other by extremely minute trichites and globulites. This fine narrow fringe is especially prominent on the ends of crystal sections, cut perpendicular to the plane of symmetry, the sides, as a rule, showing no such borders, but ending cleanly against the groundmass. In one case a rather small and narrow sani- H. §. Washington—Ischian Trachytes. 3877 dine crystal has been cracked across and slightly bent, and from the broken edges have grown minute needles, prolonging the edges and crossing each other (fig. 1, a). Fie. 1. The groundmass crystals (almost solely of orthoclase), which show a decided flow-structure, have increased greatly, not only in number but in size (attaining lengths of 0°3"™) and in complexity. ‘They are seen on examination to be elongated par- allel to the @ axis. Notwinned forms were seen, but ‘almost all the small prisms show more or less forked forms, some of which are shown in fig. 1,b—#. The forms are irregular, and are rather tree or twig-like, than sheaf-like, in habit, as they do not show much tendency toward equal development of the two ends. A number of stubby brush forms are to be observed among the smallest of them (fig. 1, ¢, 7); and among these especially are noticed forms which are more highly devel- oped and which resemble closely those to be described presently. The basis proper is a clear glass, of a pale café au lait color, showing here and there an air vesicle. Surrounding the small forked crystals is a darker brown, slightly oranular substance, which exerts some action on polarized light, as is evidenced by a very feeble aggregate double refraction. This substance also occurs in isolated spots and irregular anastomosing streaks Am. Jour. Sci1.—FourtTH SERIES, VOL. I, No 5.—May, 1896. 25 378 Hf. S. Washington—Ischian Trachytes. not inclosing erystals. A fine globulitic dust occurs in it to some extent and there is a marked tendency toward spherulitice or axiolitic forms, with a faintly developed radiated structure. This gives rise to rather mammillary or botryoidal borders around the feldspars, of which it has been attempted to give an idea in fig. 1, ¢, d, e. The last Mt. Rotaro specimen to be described (No. 528) is properly a trachyte, not an obsidian, and it may be mentioned here that the majority of the blocks of Mt. Rotaro are of this rock or the black obsidian. It is dark-brownish black in color, of a duil luster and quite compact. There are very many white glassy sanidine phenocrysts, varying in diameter from: » town. Under the microscope the phenocrysts are seen to be the same as in the preceding types, though augite, biotite and -magnetite are rather more rare. The sanidine phenocrysts show a much greater development of the fringed borders than in the last case,and the fringe not infrequently reaches a depth of 0°5™™, being usually the deeper the smaller the crystal. In some a horn is seen on each side, being a narrow continuation of the feldspar substance, with the fringe of bright needles and interstitial trichites and dust between them. These horns are sometimes slightly bent, diverging outward, and give the crystal the appearance of a shark’s ege- sac (“sea-purse’”’). The sides also show occasionally a narrow border of fine needles and trichites, the needles lying parallel or nearly so to the crystal edge. The groundmass consists of a colorless glass basis quite thickly sprinkled with small black grains and globulites. Its most striking feature is the great abundance of small sheaves of orthoclase needles, a development of the forms just described. These sheaves show a well marked flow-structure, in a given area, the long axes lying approximately parallel. In length they vary from 0:2 to 0°5™™, comparatively few being either longer or shorter, and their width at the widest part is about half their length. While varying considerably in details, yet the general structure is much the same. Some typical forms are shown in fig. 2, a—A, though the great deli- cacy and complexity in most cases is only roughly given. | These sheaves are composed in general of a single, straight, untwinned crystal; which at its center is narrow, but which at the ends is split up, the split portions diverging but preserv- ing their continuity with the main body. The fission and diver- gence have gone on, as a rule, approximately equally at both ends, and in all directions around the axis, producing quite symmetrical forms. Besides the diverging needles which are obviously connected with the waist or main crystal, are others H. S. Washington—Ischian Trachytes. 379 which seem to be detached. Closer examination shows that most of these (if not all) are in reality of one substance with the main group, the connection being either masked by tri- chitic matter, or having been destroyed by the position of the plane of the section. : BIG 2 All these needles diverge at generally not large angles (up to about 30°) from the nucleus or from each other; though here and there some are seen which form angles up to 90° with the main prism. Though usually straight, or nearly so, many show marked curvature, the concavity being always outward. In some cases needles along the side form curves which are quite hyperbolic in character (fig, 2, @, 2, 7). Between the needles lies a colorless, or very faintly brown, isotropic substance, which is very thickly sprinkled with minute globulites and curved black trichites. These bring out the structure very plainly, the colorless needles standing out bright against the dark background. Examination with the mica plate shows that in both the nuclear prisms and the divergent needles the axis of greatest elasticity a, lies parallel to the length, and that hence the crys- tals are elongated in the direction of the @ axis—the usual. habit. 380 H. S. Washington—Ischian Trachytes. For such divergent erystal forms, which, as will presently be seen, are due to the ramification and growth of a single indi- vidual, and which correspond to the spherokrystalle of Lehmann and Rosenbusch, I would propose the name keraunoid (Gr. xepavves, a thunderbolt). This word, which may seem some- what fanciful, is chosen on account of the narrow waist and divergent, equally developed ends, with symmetry about an axis which give them a striking resemblance to ancient Greek representations of the thunderbolt of Olympian Zeus. It must be understood that the mode of formation is connoted with the term. Besides these well-defined forms are seen in smaller number oblong spots of about the same size, of a dark gray eolor and dusty appearance, showing a very fine indistinct structure of parallel or slightly divergent lines. Under high powers these are resolved into bundles of colorless orthoclase needles, less clearly cut than the others, and with such a large proportion of interstitial trichitic basis that their true character is masked. It is probable that these are less well developed forms, as they show a great resemblance to the fringes at the ends of the sani- dine phenocrysts. A few small clusters of needles radiating from a point are also seen—evidently sections cut through the ends of keraunoids at large angles with the axis. The extinction of the needles is parallel or only slightly inclined, so that as the stage is revolved a rather broad dark band swings across the keraunoids, with occasionally, in the thicker ones, a short bar at right angles. In consequence of the approximate parallelism of the keraunoids due to flow- structure, the field under crossed nicois has an appearance simu- lating that of many spherulitic groundmasses, being sprinkled with short, thick, black bands lying parallel to one another, with here and there a thick elongated cross. An examination of all the other Ischian slides in my posses- sion (about seventy-five) revealed the presence of the kerau- noids in only two of the blocks of black trachyte from the tufas of the Scanella Cliff, on the southwest coast, where fine sections of interstratified tufa and lava beds are exposed by wave action. One of these shows scarcely any phenocrysts, and the colorless glass basis is rather dusty. The keraunoids here are made up of much finer needles, and with more interstitial trichitic basis than in the Mt. Rotaro specimens, but the structure is quite the same. The other is much more like a normal trachyte in its groundmass, which is almost holoerystal- line. True keraunoids are wanting, but many of the ground- mass orthoclase crystals show divergent forms; and this may be held to be a type representative of the holocrystalline development of the glassy trachytes described above. Some H. 8. Washington—Ischian Trachytes. 381 small forked orthoclase crystals were also seen in the ground- mass of a black glassy obsidian from the tufa of Monte di Procida, near Cuma; and Rosenbusch* speaks of similar forms in obsidian from Ponza, though I could find none in my specimens. In regard to the origin of these peculiar forms, it will have been evident that they are most certainly not due to twinning, and also that they cannot be referred to skeleton forms con- sequent upon growth along certain crystallographic axes or inter- axes. They must be either an aggregation of individuals about a common central point or axis, or else due to the continuous splitting up and growth of what would be under other circum- stances a compact individual. Many cases of such ramified crystals are described and figured by Lehmann+ and all the evidence points to the Ischian keraunoids being of this character, and not of the nature of aggregate growths of separate individuals. ‘This is clearly seen on examination of the series; passing from the most glassy obsidians with small and seldom forked crystals, through the brown glassy trachyte, with individuals which are evidently split and whose diverging needles can clearly be seen to be continuous in substance with the central crystal, to the trachytes with colorless basis, where the forms have grown quite complex but in which the continuity can be made out by careful study. According to Lehmann, the cause of this ramification is “the existence of internal tensions which cause the crystals to split here and there at the surface, producing a discontinuity which cannot be overcome by further growth. ...... The broken parts grow independently and so form ramifications from the erystal which are no longer oriented exactly parallel to the main mass.” These secondary needles may also split and ramify and thus complex forms result. That the existence of such internal tension is a sufficient proximate cause is not to be denied, and the facts as shown by Lehmann and elsewhere apparently prove that it is the true one also. It is, however, more difficult to give the cause of the existence of such tensions; or to determine whether they are inherent in the crystal during growth under certain conditions, or whether they are due to the physical action of the magma on the crystal substance. That the conditions were extremely local in their character is to be inferred from the few instances of ramification found among the Ischian trachytes, as well as in the Hawaiian basalts to be mentioned presently. It is scarcely possible that the * Rosenbusch. Mikr. Phys. II, 565, 1887. + Lehmann, Molecularphysik, Leipzig; 1888, I, 378 ff. 382 Hi. §. Washington—Ischian Trachytes. formation of skeleton crystals is the primary cause, as Lehmann suggests (p. 390); at least in the trachytes in question no traces of skeletal growth are to be seen, and such growth is well known to be rare in the feldspars. That fission and ramification took place after the mass had come to rest, is to be inferred from Lehmann’s conclusion (p. 880) that a certain degree of viscosity and acceleration of the crystallization is necessary to the process. It is also indi- cated by the extreme delicacy of the forms themselves, which would hardly be able to withstand the action of amoving magma. The existence of the easy cleavages parallel to the base and clinopinacoid probably assisted the fission materially; and, as the prisms are elongated parallel to the axis @, may explain the symmetrical arrangement about an axis and not on each side of a plane. The fringes at the ends of the sanidine phenocrysts seem to be due to the fission of feldspar substance which erystallized out of the surrounding zone and was oriented like its erystal nucleus, rather than to the fission of the phenocrysts themselves, which latter idea the sharp straight edges of the crystals de- cidedly disprove. The horns so often seen on either side may be due to skeletal growth, and their divergence is explained by the expansion due to fission of the enclosed feldspathic matter. The forms just described are closely analogous to the “ feather forms” assumed by augite in certain Hawaiian basalts described by E. S. Dana,* to whose kindness I am indebted for an exami- nation of the original slides. These augitic groups are much larger than the Ischian ortho- clase keraunoids, are generally coarser, and differ as well in being much more curved and complex. Though the nuclear crystal is not, as a rule, as prominent or as constantly present in the Hawaiian forms, there cannot be much doubt but that they are due to the same cause as the Ischian—an internal tension splitting the crystal and subsequent growth enlarging it on these lines. A tendency among the augite keraunoids towards greater development at one end could be observed in many ceases, which might be connected with the hemihedrism of pyroxene.t It was also interesting to observe among these basalts, especi- ally the finer grained ones, a tendency of the feldspars to fork and form rudimentary keraunoids somewhat resembling, though much coarser than, those seen in No. 553 of Ischia. The presence of feldspar microlites with “ curved processes” apparently due to fission, and surrounded by a dark spherulitic zone, in a Hawaiian basalt glass,t is of great interest in this *K.S. Dana, this Journal, xxxvil, 443, 1889. +G. H. Williams, this Journal, xxviii, 115, 1889. t Dana, op. cit., 451, fig. 5. H. S. Washington—Ischian Trachytes. 383 connection, as being very closely analogous to the Ischian forms and an evidence of the existence of internal tension in the erystals of these rocks. Rosenbusch* mentions several cases of similar forms, and they have been described by Iddingst in a rhyolite of the Eureka District in Nevada. Analogous forms have also been noted by Herzt in a diabase from Guagua Pichincha, by Vogt§ in a slag, and also by H. Vogelsang.| It is of especial interest, however, to compare with the Ischian forms the branching feldspars observed by Iddings{ in the Obsidian Cliff in the Yellowstone Park, and by Cross** in rhyo- lites of Custer Co., Colorado. The ‘branching feldspars in these cases are orthoclase and are found as constituents of cer- tain spherulites. Some of them are not due to fission, but are branched erystallographically and the prisms are elongated parallel to cither the ¢ or @ axis in different parts of the group. Others seem from the descriptions and figures to be quite identical in nature with those of Ischia, the prismsand needles being also elongated parallel to a. The conelusions to which these two observers come regard- ing the structure and origin of spherulites in general are especi- ally noteworthy. Iddings points out that we must ‘“ base the general definition of spherulitic structure on some other charac- ter than outward form,” and that this fundamental character- istic is “their mode of crystallization”; spherulitic growth consisting in “ the formation of radiating or diverging groups of crystals,’ whatever may be their outward shape, or whether _ the divergence takes place from one or more points or a plane. While this definition may perhaps be thought to be in some ways too broad, (since strictly according to it we could call the radiating groups of tourmalines in granites spherulitic), yet that it applies to the larger number of true spherulites, if not all, can hardly be denied. According to this definition the Ischian (as well as the Hawaiian) keraunoids are true spheru- lites, though they are not groups of radiating, separate, rami- fied crystals, but one individual. They are in fact, as has been said, the “ sphewrokrystalle” of Lehmanntt and Rosenbusch.t+ This term is not appropriate, * Rosenbusch, Mikr. Phys., i, 36, 628; Taf. iii, fig. i; ii, 494, 548. + Hague, Geol. of Eureka District, Mon. xx, U. 8. G. S8., 1892, 378, Pl. III, fig. 1 | Reiss and Stubel, Reisen in 8. Amer. Hochgeb. Rep. Heuador, I, Berlin, 1892. 88, 8 Vogt. Mineralbild. in Schmelzmassen, Kristiania. 1892, 179, Pl. II, 15. | Die Krystalliten, Bonn, 1875, Taf. xiv, figs. 4 and 6. 4] Iddings, Bull. Phil. Soc., Washington, xi, 1891, 445, cf. Obsidian Cliff 7, Rep, U.S. G.S, 276 ff 1888. ** Cross, Bull. Phil. Soc. Washington, xi, 1891, 411. ++ Lehmann, op. cit., i, 379. tt Rosenbusch, op. cit. i, 36. 384 H. 8. Washington—Ischian Trachytes. as Cohen* and Crosst have pointed out; since a true, or even approximately, spherical shape is an extreme form, and one very rarely attained by the process. For this reason I sug- gested the use of the term keraunozd, which better fits the majority of the forms, and since the need of some term for this type of radiate growth seems to exist. Cross (Joe. cit.) denies to such forms the right to be called spherulites, but if Iddings’ definition be accepted they properly come under this head; as it is the radiate manner of growth, and not the number of indi- viduals or the outer form which is the determining classifica- tory characteristic. One can infer from the observations of these two writers, as well as from those of others and from general considerations, that true spherulites in this sense may be produced by three distinct processes. These are: by the divergence of many .separate, simple, prismatic crystals from a common point or points—the most common type, especially in the smaller spheru- lites; by the crystallographic branching of radiating prisms, as seen in Iddings’s “ porous” and Cross’s “ hollow” spherulites ; or by ramification, which may take place in only one crystal, as in the examples of Cross and Iddings. We might distinguish the two forms of this third type by the use of Tschermak’st terms monosomatic and polysomatic applied to meteoric chon- drules. The same terms may be used for the “branched” spherulites ; while those formed by the first named process are, in the nature of the case, always polysomatic. Any two, or all three, of these processes may combine, simultaneously or successively, to form complex spherulites. In regard to the mode of growth of spherulites, Cross comes — to the conclusion that, antecedent to the crystallization of the feldspar, there was a development of colloidal substance within the area of the spherulite; and that the fission of the crystals is due to the tension assumed to exist in this solidifying col- loidal mass. That such a tension external to the crystal would induce strains within it, or at least aid the process of fission, seems highly probable. But apart from this an internal tension must exist in the crystal, as shown by Lehmann, and as several facts noted above (such as the curvature of the needles) go to rove. 5S In connection with this idea of the formation of a colloidal substance the presence of a zone of dark brown, feebly polariz- ing substance around individual crystals, and in patches con- taining many forked crystals, in the Ischian trachytes, (and in * Cohen, Gotting. gel. Anzeigen, 1886, 915. + Cross, op. cit., 432. t Cf. Cohen, Meteoritenkunde, Stuttgart, 1894, i, 260. H. §. Washington—Ischian Trachytes. 385 the Hawaiian basalt glass), is especially noteworthy.* This brown substance seems to be identical with the “ supplemental spherulitic growth” of Cross (p. 424), and may be supposed to represent this colloidal substance, which in the more quickly cooled obsidian has not had time to erystalize, while in the trachyte proper (No. 523) it has entirely disappeared. In the Colorado rhyolites Cross supposes this substance to be composed of hydrous silica and feldspar, in accordance with the chemical composition of the rock and the intimate association of quartz or tridymite with the orthoclase in the spherulites. This supposition cannot be made in the case of the Ischian trachytes, where the silica does not reach 61 per cent, and the HO content is less than one per cent. Here we must suppose it to have been almost entirely of feldspathic matter, with some iron which went to form the trichites. It may be of interest to note that the ramified orthoclase crys- tals described in the preceding pages, as well as the Hawaiian aug- ites, show a certain similarity to the diverging pyroxene needles forming chondrules in certain meteoric stones, as those of Montrejeau (1558), Tadjera (1867), and Tieschitz (1878), as described and shown in photographs by Meunier.t The resemblance is even greater with the artificial crystals of enstat- ite obtained by the same scientist and figured on page 339 of the work cited. As I have not been able to examine sections of these falls it will not do to push the analogy far, and I can at present only call attention to their apparent resemblance, and suggest that chondrules be examined from the point of view of spherulitic growth. * A similar brown zone may be seen about many of the spherulites in the rhyo- lite of the Alter Schloss, near Schemnitz, in Hungary. +S. Meunier, Les Meteorites, Paris, 1884, 240, 242, 523. 386 M. C. Lea— Atomic Weights of the Elements. ART. XLV.—On numerical Relations existing between the Atomic Weights of the Elements; by M. Carty La. In the first part of a paper on the ions it was shown that the elements were divisible into three great classes: those whose ions were always colorless, those whose ions were always colored, and a smaller class whose ions were colored at some valencies and colorless at others.* It was also shown that the first class, those whose ions were always colorless, could be arranged in vertical lines so that the horizontal lines contained each a natural group. Also that the elements having both colored and colorless ions were much more closely allied to these than to the group having always colored ions. This last named class does not divide into groups at all, but forms series with the atomic weights immediately following one another. Therefore as long as an element has any colorless ions it really seems to belong to the class with the ions all colorless. So much so that when the first class is tabulated the members of this transitional class find vacant spaces into which they naturally fall. To make this clear and to elucidate what fol- lows, I here reproduce Table II from the first part in a con- densed form. (Transitionals in italics.) I iL 1 | Siaae WG) Cl 355 Br 80 Jie Ane peg ian Na 238 Ke 39: Rb 85 Cs 132 Be Of iM Be asa Ca 40 Sr. 88. Ba isn mbt ju & 2 ea Sel Se 44 x 90 La 139 Sana LV: a Sap Ti 48 Pie Baath paw V. cee Soke | eae | Nb 94 Ta 183 Ay ee Wil: une rie EE Mo 96 W 184 isp tbh iE eee Sas Cu 63 Ag 108 Au 196 Bape i Be 9, Me 24 Yneas Cd 119° He sont eee. TPES Bello) WAT 2754 Ganeo In’ 11 a7 een apes Ve Ort2 Si 28 Ge 72 Sn 118 Pb 206 Th 2384 Wie TON S14 ee loi) Aer Te Sb 120° - Be 20s Aas Vi. O 16 S 382 Se 79 Le 2125 sgt! Sees In the case of some few elements it has not been easy to find published data sufficient to determine with absolute cer- tainty the color of the ions. Further study of this subject has led me to make a slight change in reprinting the above table. I had previously classed the metal cerium as transitional. The ceric ion is undoubtedly colored, but as to the cerous ion there is some uncertainty. Its compounds are nearly colorless, but as they exhibit a slight red tinge the ions may perhaps be colored. It, therefore, seems better in this uncertainty to place cerium * This Journal, May, 1895; the second part will appear next month. M. C. Lea—Atomic Weights of the Llements. 387 alongside of the other members of its group. It is, therefore, omitted from the above table. Also in the paper just referred to gold was placed amongst the elements having colored ions only. The auric ion is certainly colored, but about the aurous ion there is some doubt; the oxids and the haloids are colored but as they are insoluble they give no positive information. It appears, however, that when aurous chloride is dissolved in sodium chloride it yields a colorless solution from which color. less crystals are obtained. Various other double salts form both colorless solutions and colorless crystals. The following are examples: Ammonium aurammonium sulphite, sodium aurothiosulphate, potassium aurocyanide, etc.* As it seems characteristic of the soluble salts to be colorless, I conelude that gold must be considered as having both colored and colorless ions. It is, therefore, a transitional element and finds its place in the above table where elements of that class appear in italics. It is, however, interesting to observe that just as these two metals cerium and gold are at the very limit- ing point between two classes, so there are spaces open for them in each of these classes, a circumstance that can hardly be fortuitous. Table of Differences. If the respective numbers of the first column in the preced- ing table be each subtracted from the corresponding number of the second column, the second column from the third and so on, we obtain a series of differences which are given in the table below. 18 16°5 44:5 47 16 16 46 47 opin Laae 48 49 sate SRI 45 49 aS Hite 44 88 Be is aa Le 88 nae Late 45 88 15 413 46°7 88 16 49 45 90 16 44 46 88 17 44 45 88 NGI ipa cas) AG 46 ae It has been remarked, I think as far back as the time of Dumas that differences of 16 between the atomic weights fre- quently presented themselves and occasionally differences of forty-five or thereabouts. But these were scattered cases. In the above table of differences ald the elements are repre- sented with the exception of the comparatively small group having ions always colored. As has been already said, this * Roscoe and Schorlemmer, Ist ed., ii, 2, pp. 380-1. 388 Series of: all the Elements in Numerical Order. 2"° COLORED GROUP M. C. Lea—Atomic Weights of the Elements. * Ath %, 4T*COLOREO BROUP 3 COLORED GRoup 2. {ST COLORED GROUP HoLi, Bald (Gath MOM RE NaiMaAlSS PAO GIEKCGaSe. @ fill ase sues (7 9 I 12 4 16 19 23 24 27 28 3) 32 35539 40 44 “Th Sa CHiIAMSATS Dalai CatBallae mee ae ea SE ee 65 69 72 75 79 80 BS 88 90 2 4 18 120 125 127 132 137 139 See Avternate Series 9 CoLoRLESS 9 CoLtorLess ’ 9 CoLoRLEss 9 CoLoRLEss last group of elements having nothing in common with other elements cannot be classed with them and will probably be found, when we know much more than we do now, to have a wholly different constitution. There are several things worthy of remark in the above table. It will be seen that the differences at first approximate to 16. Then comes a long set of differences, twenty in number, beginning with 41°3 and gradually but not regularly increasing to 49. The remaining differences are all exactly 88 with one exception of 90, but as this is the dif- ference between two metals of which com- paratively little is known and whose atomic weights cannot be considered as being aceu- rately fixed (indium and thallium), it is quite probable that this exception may hereafter disappear. It should be remarked that when the set of differences approximating 16 is once left behind, this difference does not reappear in a single case. The same is true of the differ- ences averaging 45: this class of differences does not once reappear when 88 is reached. For some time past it has been believed that the oxygen group does not end with tellurium but contains still another member with a higher atomic weight. It will be ~ observed that in both the above tables there remains a space for such an element. Its atomic weight should exceed that of tellu- rium by the difference 88 and should there- fore be 213. Finally it may be said that the blanks in the table of differences are due to corresponding blanks in the table of atomic weights. The change just mentioned in the classifi- cation of cerium and of gold makes a slight - change in the table representing the whole range of elements in series. I therefore reproduce this table as corrected. It will be observed that all the elements having only colorless ions appear on the base line. The elements having ions always colored appear on the upper line parallel to the base line. The transition elements appear on the inclined lines. C. Palache—Crocoite from Tasmania. | 389 Art. XLVI.—Crocoite from Tasmania ; by CHARLES PALACHE. THROUGH the kindness of Mr. Stephen A. Douglas of San Francisco, the writer came into possession some time since of specimens of crocoite from Tasmania representing, so far as his information extends, a new or undescribed locality for this min- eral. Of the two specimens available for study one is now in the mineral cabinet of the University of California, the second in the writer’s possession. They consist of masses of crocoite erystals clustered upon bases of lamellar limonite, each mass measuring several inches in diameter. The mineral occurs in a silver-bearing lead deposit known as the Adelaide mine on Mt. Dundas, west coast of Tasmania; but of the nearer geo- - logical relations unfortunately no information is at hand. It is said to occur in considerable abundance, a statement borne out by the appearance of these specimens and the many others in Mr. Douglas’s possession. Besides the limonite there is no trace of gangue or wall rock nor are there any other lead min- erals such as might be expected to accompany the crocoite. The crystals are of a light hyacinth red color, quite translu- cent and with adamantine Iuster. They vary in size from minutest needles to prisms of 2° length and 3™™ diameter. The habit is prismatic and the crystals are never doubly termi- nated, being attached at one end to the limonite. The larger erystals are often cavernous, giving rise to hollow prismatic forms. As is usually the case with crocoite, the crystal planes are even and brilliant, giving good reflections on the goniome- ter. The faces of the prism zone are, however, strongly stri-. ated parallel to the prism edges and this renders the identifica- tion of some forms doubtful. Four crystals were subjected to measurement and showed the following forms, most of which were present on each crys- tal. The letters used are those of Dana. m (110) I & (101) 14 e720) (-2 g (O11) 1— d (210) i-2 w (012) 42 *S (10:30) 7-72 y (021) 2-2 *T (530) i-3 Grell) ae b (010) i oe (Taal a c (001) 0 The fcllowing table shows some of the measurements and the angles calculated from Dauber’s elements. 390 C. Palache—Crocoite from Tasmania. Average Number of Calculated. measurement. times measured. Limits. mam*> 1104110 86° 19” 86° 14’ 14 85° 59'-86° 317 CPi Aggie Wit bee$ 011 83.31 83 38 al waw 0124012 48 11 48 12 y nf OON ROL T2135 121 40 1 kav 1014111 SOrs0 35 41 1 A 0 5530010 60 38 59 55 ae | Sa b 10°30. 010 Tie IETS Casale 3 73°-75° 157 The two prisms (10°3°0) and (5°30) were represented by exceedingly indistinct faces, reflections from which were only dimly visible with the 0 ocular of the Fuess instrument. On this account they are considered doubtful and are not intro- duced into the figure. The first form S (10°30), is unrecorded; the second, T (5°3°0), is enumerated among doubtful forms by Dauber. The remaining forms are shown in the figure in about an average development ; but their proportions vary widely in various crystals with either ¢ (111), w (111) or forms of the climodome zone predominating. This combination of forms is exceedingly like that shown by Dauber* on a erystal of crocoite from Berezov in the Ural, which is somewhat surprising considering the widely different paragenesis of the mineral in the two localities. Mineralogical Laboratory, Harvard University, January, 1896. * Berichte Akad. Wien., xlii, fig. 93, Pl. II, 1860. we Chemistry and Physics. 391 SCIEN TIETC INTELEIGEN CE. I. CHEMISTRY AND PHYSICS. 1, Investigations with sulphide of nitrogen.—In a preliminary notice, CLEVER and Mutumann have described a few products obtained by the action of several reagents upon this well-known explosive compound. The substances obtained are so remarkable in composition and behavior that it is evident that the authors have opened up an important field of investigation. By deter- mining the elevation in boiling-point of the carbon disulphide solution, the authors have confirmed the recent results of Schenek, who used the depression of the freezing-point in naphthalene solu- tion, in arriving at a molecular weight 184 for sulphide of nitro- gen, corresponding to the formula N,S,. By the action of an excess of bromine upon a solution of nitrogen sulphide in carbon disulphide, a well crystallized, bronze-colored compound, N,S,Br,, was obtained. By exposure to the air, the substance just men- tioned loses sulphur bromide by evaporation and is changed into — a yellow amorphous compound, probably having the composition N,S,Br,. If bromine vapor is allowed to act upon dry nitrogen sulphide, it is greedily absorbed, liquefaction takes place and after some time large garnet-red crystals of a very unstable body with the composition N,S,Br, are formed. By exposure to the air this substance yields the previously-mentioned yellow amorph- ous body. ‘The most interesting compounds, however, are those which will be now mentioned. By the action of nitrogen dioxide (NO,) upon a solution of nitrogen sulphide in carbon disulphide, a very deliquescent, white crystalline compound, probably NSO,, is produced. It reacts with water with the evolution of NO and the formation of sulphuric acid, probably according to the equa- tion, NSO, +H,0 = NO+HS0,. Upon acting upon the previously-mentioned substance, N,S,Br,, while suspended in carbon disulphide, with nitrogen dioxide, a canary-yellow, substance is deposited in microscopic crystals. This is free from bromine and probably has the composition rep- resented either by the formula NSO or N;S,0O,. This compound decomposes suddenly, giving forth a brilliant light, even when a tube containing it is exposed to the heat of the hand. It dis- solves in water, giving a yellow, neutral solution, which, upon warming, deposits a black substance. The latter quickly decom- poses into sulphur and a gas of peculiar odor, which has not yet been investigated, but which the authors surmise to be a lower oxide of sulphur. The canary-yellow compound behaves in a different manner with alcohol, giving a dark red solution which gives off sulphur dioxide upon boiling and deposits a crystalline substance which had not yet been analyzed. Another curious 392 Scientific Intelligence. substance, probably N,S,O,, was obtained by the action of nitro- gen dioxide upon N,S,br,. The authors intend to complete the study of the substances that have been mentioned, and to extend the investigation to the action of chlorine, iodine and the chlor- ides of phosphorus upon nitrogen sulphide.— Berichte, xxix, 340. Heh. W. 2. Manganese carbide——Motissan has prepared, by means of his electric furnace, a well-characterized compound CMn,. This result confirms that of Troost and Hautefeuille, who obtained the same substance by the use of a wind-furnace. Moissan used 200 parts of manganese protosesquioxide and 50 parts of sugar char- coal and heated the mixture in a carbon tube, closed at one end. The compound is readily attacked by various chemical agents, and the action of water upon it is especially interesting from the fact that equal volumes of marsh-gas and hydrogen are evolved. The following equation represents the reaction: CMn,+6H,O = 3Mn(OH),+CH,+H, No acetylene or ethylene is produced.— Compt. Rend., exxii, 421. : H. L. W. 3. The preparation of pure strontium compounds.—SORENSEN has made a critical study of various methods of obtaining stron- tium. salts free from barium and calcium. 'The process which he recommends, after an elaborate series of experiments, is briefly as follows: The greater part of the barium is removed by adding concentrated hydrochloric acid to a solution of the chlorides until strontium chloride begins to crystallize out upon cooling. A precipitation by means of sulphuric acid is then made in the hydrochloric agid solution, whereby most of the calcium is left in solution. The sulphates are decomposed by warming with strong ammonium carbonate solution, the washed carbonates are dis- solved in nitric acid, the solution is evaporated to dryness, the residue is dissolved in water and filtered, and the barium is com- pletely removed by repeated fractional precipitations with small amounts of dilute sulphuric acid in a solution containing ;', of its volume of 66 per cent nitric acid. This separation is not con- sidered complete until the last precipitate of sulphate is free from barium. The nitrate solution is evaporated to a semi-solid condi- tion, the mass is extracted with alcohol and the residue is washed with the same liquid. This residue is dissolved in water and the operation is repeated until the calcium has been completely removed, ‘The author obtained a yield of 76-77 per cent by this method. The author calls attention to a rale which has an important bearing upon the separation of barium, strontium and calcium compounds from each other. He considers the rule to be quite self-evident, but he believes that it has not always been taken into consideration in the preparation of chemical products. It is expressed by the statement that ‘‘ Corresponding, isomorphous salts of closely-related elements are more difficult to separate than Chemistry and Physics. 393 corresponding salts which are not isomorphous.” As examples, the author notices that he was unable to completely separate strontium chloride from calcium chloride, although the two salts, separately, have an entirely distinct behavior when concentrated hydrochloric acid is added to their solutions. On the other hand, barium and strontium chlorides, although their behavior with concentrated hydrochloric acid is similar, can be readily separated so as to obtain pure barium chloride, because these salts are not isomorphous, Calcium nitrate, having a crystalline form which is distinct from that of barium and strontium nitrates, can be separated from the latter. The sulphates of calcium, strontium and barium are isomorphous, but although they differ considera- bly in their solubility, for instance in dilute hydrochloric acid, and in-their stability towards the alkali-carbonates, it is never- theless impossible to separate them when they are precipitated together, and not each by itself—Zedttschr. Anorg. Chem., xi, 305-378. H. L. W. 4, A new class of compounds of metallic salts with ammonia. —Wiepe and Hormann have described the salts C,S,Co,(N H,),, CS,Ni(NH,), and C,8,Fe,(NH,),2H,O, which are evidently deriva- tives of thiocarbonic acid, and members of a new class of the interesting metallic-ammonia salts. The simplest method of pre- paring the compounds is by the action of aqueous ammonia and carbon disulphide upon the metallic hydroxides. The cobalt and iron salts are black while the nickel compound has a ruby red color, and all of them were obtained in a well crystallized condi- tion. It is remarkable that iron forms such a compound.—Zeiéschr. Anorg. Chem., xi, 379. H. L. W. 5. Influence of light on the form of discharge of a Holtz machine.—E.stER and GrITEL have shown that the brush and spark discharge of a Holtz machine, between a spherical anode and a disc cathode, can be made to disappear under the influence of ultra violet light. A concave disc of amalgamated sheet zinc constitute the negative poles. When sparks are passing between the sphere and the disc, they can be made to disappear by light- ing a piece of magnesium ribbon in the neighborhood of the cathode. When the magnesium light no longer illuminates the cathode the sparks reappear. In this case the illumination restricts apparently the discharge instead of increasing it. Elster and Geitel return to the consideration of this phenomenon, and conclude that the brush discharge and spark of a Holtz machine is supplanted by a glow discharge when the cathode is illuminated, and that this last form of discharge carries over a less quantity of electricity than the brush and spark discharge in the dark.— Ann. der Physik und Chemie, No. 3, pp. 401-407, 1896. Jes 6. Hluorescence of Vapors.—K. W1EDEMANN and G. C. Scumipt State that sodium and potassium vapor fluoresce brightly, the first green and the last red. The fluorescence spectrum of sodium vapor gives continuous and also channelled bands, together with the sodium line. Sodium and potassium vapor show also Am. Jour. Sci.—FourTH SERIES, VoL. I, No. 5.—May, 1896. 26 394 Scientific Intelligence. under electrical excitation of fluorescence continuous bands in the green and also the red. Stokes’ law apparently applies to the fluorescence of metallic vapor, and the fluorescence of the latter affords a means of explanation of various astrophysical phenomena. —Ann. der Physik und Chemie, No. 3, pp. 447-453, 1896. 3. 7. 7. Interference of Electric Waves. — Viktor von Lane employs Quincke’s well known double U-tube form of apparatus for showing the interference of sound waves to exhibit also the interference of electrical waves. Righi’s apparatus is used to generate short electric waves. These waves are sent into the Quincke tube and by suitably changing the length of the arms of the tube, interference is produced which is detected by a species of coherer such as was used by Branly, and developed by Lodge. One arm of the Quincke tube was also partially filled with certain dielectrics and their index of refraction measured. The value of the latter for parafline was 7=1°701. Righi obtained u = 1°43. For sulphur «= 2°'333, while Righi obtained p= 1°87.—Ann. der Physik und Chemie, No. 3, pp. 480-442, 1896. a: D 8. Réntgen’s Discovery.—Since the last issue of this journal considerable progress has been made in the art of cathode photog- raphy. Greater detail has been obtained by what are called focus tubes, which consist in the main of a modification of that form of Crookes tube which was employed to show the convergence of the cathode rays proceeding from a concave metallic mirror, upon a thin sheet of platinum placed at its focus. When this sheet of platinum is inclined atan angle of 45° to the line connecting the cathode and the ancde, it apparently serves as a center from which sharp shadows are obtained. It seems probable also that the anode reflects the cathode rays in a similar manner. ‘Tesla, and O. N. Rood bring forward evidence to show that the cathode rays can be reflected. Tesla early stated that it was best to use one terminal of a high-frequency coil. I have found this method of great advantage in diminishing the liability to breakage of the tubes. A wire isled from one terminal of a Thom- son or Tesla coil to one terminal of a Crookes tube, the other terminal of which is connected toa large metallic plate. The system should be brought into resonance. The method, however, presents this difficulty. The high electromotive force and the electrical oscillations drive out the occluded air and the tubes require to be re-exhausted. It is desirable, therefore, to have the tubes always connected with a mercury pump, or to exhaust the tubes originally by the aid of a high-frequency coil. Tubes which have been electrically excited by the ordinary Ruhmkorf coil while being exhausted soon depreciate under the action of the high-frequency coil. Se 9. The Temperature of the Carbons of the Hlectric Arce.— Witson and Gray in a paper before the Royal Society give the results of measurements of the temperature of the positive pole of the electric arc. The temperature obtained was 3600° abs. or 3300° C., Geology and Natural History. 395 This agrees very closely with the approximate estimate given by Violle (1893), viz.: 3500° C. The method here employed is that used by the same authors in an earlier investigation on the effective temperature of the sun (Phil. Trans., A., vol. elxxxv, 361, 1894). This consisted essentially in balancing the radiation from a platinum strip against that of the carbon of the are. Further, taking 3300° as the temperature of the crater of the positive car- bon, that of the negative carbon is found to be about 2400°. No estimate is attempted of the temperature of the arc itself — Proc. Roy. Soc., lviii, 24. eras 10. Melting points of some of the metals.—S. W. Hoiman, R.R. Lawrence and L. Barr have recently given the following deter- minations of the melting points of several of the metals, which they believe to be more reliable than previous data. The values are based upon 1072° C. as the melting point of gold, as given by Holborn and Wien. They are as follows: PAM UTA TUNEL: es ete Share Te he SN 660° SWIG) EE canes CUA eo Bis Meat eae ee 970° SOG eee erp AEE 2/8, tn) ae i [1072°] assumed Wop per eee see ed Seu es 1095° Aspire ee eee lee GO All of the samples experimented upon were of a high degree of fineness except the platinum, which may have contained 0°5 p. c. of impurity. The paper, from which the above data are quoted, gives in full the method of experiment followed. Other papers by S. W. Holman, also recently published in the Proceedings of the American Academi y, have the subjects: Calorimetry, methods of cooling correction; also Pyrometry, calibration of the Le Cha- telier thermo-electric pyr ometer. Il. GroLtocgy AND NatTuRAL HISTORY. 1. Economic Geology of the Mercur Mining District, Utah ; by J. Epwarp Spvurr, with introduction by 8. F. Emmons. Six teenth Annual Report of the United States Geological Survey, Part I, pp. 343-454. (Author’s abstract.)—The Oquirrh moun- tains are one of the parallel ranges of the Great Basin, and the first lying west of the Wasatch range and the Great Salt lake. Close to the southern end of this range the Mercur mining dis- trict is situated, in a well-marked topographical basin which has been called the Mercur basin. ‘The rocks exposed in the Mercur basin consist of about 12,000 feet of strata, chiefly massive lime- stones with intercalated calcareous sandstones and occasional shale beds ; fossils from various points in this series show it to be all of Carboniferous age. In the lower part there are intruded sheets of quartz porphyry of two distinct varieties. With one of these varieties, which has a distinct granophyric structure, all of the ore deposits of the district are associated. In the produc- tive region this porphyry is reduced to three sheets, averaging 396 Scientific Intelligence. | ten or fifteen feet in thickness, and within one hundred feet of each other. The lowest sheet is characterized by the presence of silver ores, to the exclusion of gold; the middle sheet by gold ores, with no silver; while the uppermost sheet, which is espec- ially thin and intermittent, is not mineralized to any extent. . The gold and silver horizon are known as the Gold and Silver Ledges respectively. The Silver Ledge is marked by complete silification of the limestone, and by barite in irregular masses, with some stibnite and a little copper and silver. It is probable that the metals were originally deposited as sulphides, and that they were intro- duced into their present position, together with quartz and barite, by ascending waters; and the phenomena accord best with the idea that the mineralization was accomplished by waters excluded from the porphyry during its consolidation, and that thus the ore- deposit is a special case of contact-metamorphism. The Gold Ledge, which is situated about one hundred feet ver- ticaily above the Silver Ledge, is characterized by a softened con- dition of the ores, whether in the normal condition of sulphide, or in the zone which has been bleached and altered by surface oxidation. Its most characteristic feature is realgar, which occurs in large amounts in the unoxidized ores, with frequent cinnabar and gold in small, but in certain zones nearly uniform, quantities. The gold is in extremely finely divided condition; but it is prob- able that it exists in the unaltered form as telluride, and that on oxidation it has become free gold. Evidence shows that the mineralization of the gold-horizon took place at a distinctly later date than that of the silver-horizon; and that the mineralizing agents were probably in a vaporous rather than in a liquid form. 2. Catalogue of the Fossil Fishes in the British Museum, Part IIT, containing the Actinopterygian Teleostomi of the orders Chondrostéi (voncluded), Protospondyli, Aetheospondyli, and Isospondyli (in part) ; by A. 8. Woopwarp ; pp. 1-544, pls. I-xvill. London, 1895.—Mr. Woodward’s introduction, of about twenty pages, is full of important conclusions, derived from his exhaustive study of the Actinopterygian fishes, regarding the phylogenetic relations of the several families. In the arrange- ment of the material in the catalogue, the author has attempted to record, in as nearly a natural order as possible, the variation of each type at the time of its dominance. The origin of the Chondrostéi is obscure, but that they are later than the Crossop- terygians is evident. When they first became dominant in the lower Carboniferous they exhibited a remarkable sense of modifica- tion and thereafter suffered very little essential change. The genus Acentrophorus of the upper Permian is the first of the sub-order Protospondyli, and it is observed to be the most generalized mem- ber of the family to which it belongs (Semionolidz) which is also the most generalized family of its series. The same fact is noted regarding Ophiopsis, the most general- ized genus of the family Macrosemeide, and the earliest to ap- pear. Geology and Natural History. — 397 The Pycnodontide, ueealle on account of the character of the axial skeleton and the mandible, are placed among the Protospon- dyli. The author is led to place little value upon the characters of the scales for purposes of classification. The combination of thick, rhombic, firmly-articulated scales of the abdominal region with delicate, cycloidal, deeply-overlapping scales of the caudal pedicle in the interesting genus Aetheolepis of New South Wales, and the second case of TZetragonolepis haying both thick and thin scales, furnish a good reason for departing from the long established usage in this respect. Attention is also called to the interesting fact that the higher fishes, like the highest of terrestrial vertebrates, are characterized by a comparatively simple mandible. And the author remarks in this connection that, “on acquiring this simplication of the jaw, the Teleostomes seem to be infused with new vigor, vertebral centra invariably occur, at first as simple rings, then as robnst amphiccelous bodies ; and a still more varied series of families arises, including analogies of all the principal modifications observed among the lower races, these being superinduced upon the new and advanced type of skeletal frame.” H. 8. W. 3. Catalogue of the Mesozoic Plants in the Department of Geology of the British Museum. The Wealden Flora. Part IT.— Gymnosperme,; by A. C. S—ewarD; pp. 259, plates xx. London, 1895. eee. following quotations from Mr. Seward’s conclusion drawn from the study of this material will be of inter- est to the geologist : ‘‘The general characters of the vegetation would certainly seem to point to a tropical climate,” p. 239. “The evidence of palcobvtany. certainly favors the inclusion of the Wealden rocks in the Jurassic series,” p. 240. Regarding the evolution of angiospermous plants, he says: “The true Wealden v egetation would seem to have been without any examples of the highest class of plants, and may be looked upon as the last of the Mesozoic floras in which gymnosperms represented the limit of plant development. One genus, how- ever, carries us a few steps towards the next stage in botanical evolution; the inflorescence of Bennettites marks a distinct advance in the differentation ot reproductive structures beyond the char- acteristic cycadean type,” p. 241. 4. Catalogue of the Perciform Fishes in the British Museum, 2d edition, Vol. 1, containing the Centrarchide, Percide, and Serranide (part) ; by Grorce A. BouLENGER: pp. 394, pls. xv. London, 1895.—The materials forming the basis for this second edition are the collections received by the museum since the year 1870. Atthe time of the publication of the first edition 29,275 specimens had been registered. The acquisitions since that time amount to 29,375 specimens. These have been obtained from all parts of the world; among the most important contributions aie those brought home by the “Challenger,” comprising littoral, pelagic, and bathybial forms from almost every part of the ocean 398 Scientific Intelligence. traversed by the ship. Also fresh-water specimens from out of the way places in Asia, Africa, South and North America and the isles of the seas. The author has given special attention in his revision to the study of the osteological characters, and the chief types of cranial structure are illustrated by figures in the text. Special acknowledgments are made of the assistance derived from the revisions of North American ichthyology by our country- men Jordan, Gilbert and EKigenmann. For the families studied the report gives a thorough revision of present knowledge. 5. Descriptive Catalogue of the Spiders of Burma, based upon the Collection made by Eugene W. Oates and preserved in the British Museum ; by T. Tooreti; pp. 406. London, 1895.— This is an exhaustive descriptive catalogue, in Latin, of this unique collection containing 310 species, of which 206 are new to Burma and 153 new to science. 6. Lhe duration of Niagara Falls and the History of the Great Lakes ; by J. W. SpeNcER; pp. 1-126, figs. 1-27, pls. i-v. (2d ed.)—This is an excellent series of papers explaining the geological features and history of Niagara Falls and environs and republished in book form under the direction of the Commis- sioners of the N. Y. State Reservation at Niagara and accom- panying their eleventh report. The chapters, nine of them, were originally published by Dr. Spencer in this and other journals; to them are added a few full-page reproductions of photographs of the falls and river. 7. Illinois State Museum. Bull. No. 7.—New and interesting species of Paleozoic fossils. pp. 1-89, pls. i-v, Dee. 5; 1895. Bull. No. 8.—Descriptions of new and remarkable fossils from the Paleozoic rocks of the Mississippi valley; pp. 1-65, pls. i-v, Feb. 1896.—In these two bulletins the authors, S. A. Mirter and Wm. F, E. Gurweny, have described and figured a large number of specimens of fossils, chiefly crinoids, and from Niagara, Hamilton and various Carboniferous formations, in large majority from the latter. 8. Oblique Bedding in Limestones.—A remarkable structural condition is described by Professor Calvin in the Le Claire lime- stones of Iowa.* This limestone is the second stage of the Niagara formations of that state. The author states: “‘In the first place it varies locally in thickness, so much so that its upper surface is exceedingly undulating, the curves in some places being very sharp and abrupt. In the second place it differs from every other limestone of Iowa in frequently exhibiting the pecul- iarity of being obliquely bedded on a large scale, the oblique bedding often affecting a thickness of fifteen or twenty feet. The phenomena suggest that during the deposition of the Le Claire limestone the sea covered only the southwestern part of the Niagara area, that at times the waters were comparatively shal- low, and that strong currents, acting sometimes in one direction * The Le Claire Limestone by Samuel Calvin, Bull. Lab. Nat. Sci. State Univ. Iowa, vol. ili. pp. 183-189, pls. i-ii, Mch. 16, ’96. Geology and Natural History. . 399 and sometimes in another, swept the calcareous mud back and forth, piling it up in the eddies in lenticular heaps or building it up in obliquely-bedded masses over areas of considerable extent. The oblique beds observe no regularity with respect to either the angle or direction of dip. Within comparatively short distances they may be {found inclining to all poiuts of the compass.” 9. Geological History of the Chautauqua (N. Y.) Grape Belt ; by R. 8. Tarr.—Under this title, Bulletin 109, of the Cornell Uni- versity Agricultural Experiment Station, publishes some topo- graphical and structural facts regarding the land bordering the eastern end of Lake Erie, which will be of interest to students of surface geology. 10. Geological Literature.—The assistant librarian of the Lon- don Geological Society has prepared a second pamphlet under the above title, containing a list of the geological literature added to the library during the year ending December 31, 1895. It con- tains a full subject index as well as list of titles of papers arranged alphabetically by authors. Although not exhaustive (157 pages), it is a convenient reference catalogue. 11. Norges Geologiske Underségelse.—The following are recent publications from the Geological Survey of Norway: No. 10. Tagskifere, heller og vekstene; af Amund Helland. No. 11. Lagfélgen pa Hardangervidda ogden sakaldte ‘‘ béjfjeldskvarts ; ” af W. C. Brogger. No. 12. Norges granitindustri af Carl C. Riiber. No. 13. Gausdal. Fjeldbygningen inden rektangelkartet Gausdals omraade ; af K. Q. Bjorlykke. No. 14. Aarbog for 1892 og ’93; udgivet af Dr. Hans Reusch, underségelsens bestyrer. No. 15. Dunderlandsdalens jernmalmfelt (i Ranen, Nordlands amt, lidt sénden- for polarkredsen) ; af J. H. L. Vogt. No. 16. Jordbunden i Jarlsberg og Larviks amt; af Amund Helland. No. 17. Nissedalens jernmalmforekomst (i Thelemarken); af J. H. L. Vogt. 12. Hruptionsfolge der triadischen Eruptivgesteine bei Predazzo in Sidtyrol; von W. C. Broeerer. (Vidensk. Skrift. 1 Math. natur. Klasse, Kristiania, 1895, 8° 183 pp.)—In this publication, which is part ‘evo of the series (part I, Grorudit-Tinguait Serie) by Prof. Brogger on the eruptive rocks of south Norway, the author mentions that having undertaken an expedition into South- ern Tyrol—a region classic in geological history—for purposes of study and comparison, he has been led to publish his observations and to draw some general conclusions from them. He first shows that “the typical rocks of Monzoni and Predazzo are not syenites, nor are they diorites (or diabases or gabbros) but nuonzonites. They form a well-characterized and particular group of rocks, which are distinguished by the fact that they occupy an intermediate position between the alkali-orthoclase rocks on the one hand and the plagioclase rocks, rich in lime and poor in alkali, on the other. The monzonites are sharply defined orthoclase- plagioclase rocks.” The relations are shown in the following table, where the monzonite group is inserted between the alkali series and the lime series. 400 Scientific Intelligence. Orthoclase-plagioclase. Plagioclase rocks. Monzonite group. Orthoclase rocks. % SiOz % SiO. {Acid quartz mon- |% SiOz | Acid quartz diorite. 67-82 Granite. 67-73 zonite. 66-72 (Tonalite.) | (adamellite) Quartz-Syenite. | Medium acid Medium acid quartz 63-66 | (Nordmarkite, 63-66 |quartz monzonite. 63-65 diorite. etc.). | (Banatite.) Syenite. 49-62 | Monzonite. 48-62 | Diorite. 50~62 |(Plauenite, Laur- \46 52 | Olivine-Monzonite, 44-53 Gabbros, vikite, ete.). | etc. | ete. Peripheral portions of the masses at Predazzo, composed of pyrox- enites, are regarded as differentiation products in place, while dikes of camptonite and complementary feldspathic rocks (lieb- nerite poryhyry) are mentioned. The order of succession of the various magmas at Monzoni and Predazzo is discussed very fully and in connection with that of the rocks of South Norway. in carrying out the discussion a notable contribution to the origin of granite and the mechanics, etc., of lac- colith formation is added with illustrations drawn chiefly from the south Norwegian localities. The author believes that in endeavoring to discover the law of succession of eruptive magmas it is wrong to confound the abyssal and effusive rocks together, since the two do not of necessity cor- respond. ‘The order of succession, basic, less basic, acid, appears in fact with abyssal rocks to occur so commonly that we must regard this succession as the normal one; the sudden recurrence to basic is known in many localities, but appears just as often to be lacking. 7" A very thorough digest of the literature is given and it can be truly said that in the light of the new petrology the author has cleared up and put in order a vast mass of confused and often con- tradictory information about this well known locality in the Tyrol. Field geologists as well as petrographers will find the work replete with fruitful and suggestive ideas. LeeVicgies 13. Mica-Peridotites in Bengal.—In two short papers by Mr. T. H. Hottanp (Records Geol. Surv. India, vol. xxvii, 1894, Pt. 4, p. 1389 seq.), some interesting occurrences of mica-peridotites. are given; one of them, which contains over eleven per cent of apatite, appears to us of especial interest in connection with Vogt’s work on apatite deposits (Zeitschr. ftir prakt. Geol., Nov., Dec., 1895). Several varieties of these rocks are described, some of which break up through coal-bearing strata. 1s Vie 14. Zusammenstellung petrographischer Untersuchungsmethoden nebst Angabe der Literatur, by E. CowEn. 3d edition, 53 pp. Stuttgart, 1896.—This little pamphlet, which contains a very full bibliography of the literature pertaining to the various methods of petrographic investigation and discrimination of rock-forming minerals, will be found a very useful addition to the working library of every mineralogist and petrographer. 1, Vas Geology and Natural History. 401 15. Jadeite from Thibet, by Max Bauer, Jahrb. ftir Min., 1896, i, p. 85.—In studying some specimens ‘of jadeite from an unknown locality in Thibet, Prof. Bauer has made the very inter- esting discovery that the mineral is a component of a rock com- posed of jadeite, plagioclase and nephelite; at times the jadeite predominates to the extent of practically supplanting the other components. The author remarks that if the jadeite at this locality is, as elsewhere, a member of the family of crystalline schists, then we have here an eutirely new method for the occur- rence of nephelite, until now restricted to eruptive rocks. It appears to us, however, that this interesting occurrence of nephelite may prove to be of great importance in explaining the origin of jadeite, whose significance i in the crystalline schists has never been understood, and it may also furnish one explanation why, among all the varieties of metamorphosed igneous rocks occurring among the crystalline schists, those containing nephe- lite have not been found. Ei ViaiP. 16. A Dictionary of the Names of Minerals including their History and Ktymology ; by AuBert H. Cuester. 320 pp., 8vo. New York, 1896 (Wiley & Sons).—The author gives in this work, to which it is evident that he has devoted a vast amount of careful, patient labor, the most complete history of the names of mineral species that has ever been attempted. How fruitful his efforts have been will be appreciated from the fact that of nearly five thousand naimes included, full information with reference to the original authority, also derivation and so on, is given in the case of all but about one hundred and fifty. This work will be highly valued by all mineralogists interested in the history of their science, and will be useful to mary others, includ- ing those concerned with the etymology of English words in general. 17. Minéralogie de la France et de_ses Colonies: Description physique et chimique des Minéraux, Etude des conditions géolo- giques de leurs gisements; par A. Lacroix. Tome Premier (2 Partie), 723 pages. Paris, 1895 (Baudry et Cie.).—The first part of Professor Lacroix’s important work was noticed nearly three years since (vol. xlvi, p. 76), and its originality in scope and method remarked upon. The part now issued concludes the first volume and carries the subject on from the micas through the pyroxenes and amphiboles. All interested will look forward with interest to the final completion of the entire work. 18. Mineralogy ; by Frank Ruttey. Eighth edition, revised and corrected, 240 pp. London, 1887 (Thomas Murby).—This little book must have been an aid to many in acquiring a knowl- edge of elementary mineralogy, for since its first appearance it has gone through numerous revisions and the eighth editidn is now given to the public. The various subjects embraced under general mineralogy are treated concisely in the first seventy pages and the remainder of the book is given to the description of species. 402 Screntific Intellagence. 19. Determination des Feldspaths dans les plaques minces ; par A. Micuri-Livy. 2d fase., pp. 71-108, Pls. ix—xxi, (Baudry et Cie.) Paris, 1896.—The first portion of this work which has been noticed in this Journal (vol. xlvilil, p. 173, 1894) has already proved of great service to mineralogists and petrographers, and this addition will also be found extremely useful. A new addi- tional method of determination for the plagioclase series is intro- duced, depending on the equal illumination (éclatrement commun) of zonal bands. ‘To it is added a résumé of the optical proper- ties of microcline accompanied by a diagram. Ls We P. 20. Allgemeine Krystallbeschreibung auf Grund einer verein- Jachten Methode des Krystallzeichnens bearbeitet und mit einer Anleitung zur Anfertigung der Krystallnetze und Krystallmo- nelle herausgegeben von Dr. Aue. Nixes. Mit 182 Originalzeich- dungen im Texte. 8vo. 176 pp. Stuttgart, 1895. (EK. Schweizer- bart’sche Verlagsbuchhandlung—-E. Koch.)—The aim of the author in this volume is to present the elements of descriptive crys- tallography in simple form, and as developed by means of a new method of drawing crystalline forms which, as he states, he has used in his instruction with much success for a number of years. This method is based upon the determination of the angular points of the form, projected by means of the methods of analytical geom- etry, these being plotted upon sheets of paper ruled in squares of 4™™ on the side. The simple mathematical relations needed are explained and instructions given for the application of the method described. The symbols of Naumann are employed throughout, only modified to allow of the designation of each individual jace. Numerous figures drawn by the author’s method show its practical application. III. MiscELLANEOUS SCIENTIFIC INTELLIGENCE. 1. On the Diurnal Periodicity of Earthquakes ; by CHARLES Davison, M.A., F.G.S. (Abstract received from the author.)— Reference is made to the previous work of De Montessus and Omori, the former endeavoring to show that the diurnal perio- dicity of earthquakes is apparent rather than real, and the latter pointing out that a marked diurnal periodicity characterizes the after-shocks of great earthquakes in Japan. The results of twenty-six registers obtained by means of continuously recording instruments in Japan, the Philippine Islands, and Italy are sub- jected to harmonic analysis with the following conclusions :-— (1) The reality of the diurnal variation of ear thquake-frequency seems to be proved by the approximate agreement in epoch (mean local time) of the first four components (24, 12, 8, and 6 hours) for the whole year at Tokio and Manila, and for the winter and summer halves of the year at Tokio. (2) In ordinary earthquakes, there is in nearly every case a marked diurnal period, the maximum generally occurring between 10 4. M. and noon. The semi-diurnal period, though less promi- Miscellaneous Intelligence. 403 nent, is also clearly marked, the maximum occurring as a rule between 9 A.M. and noon and between 9 p. um. and midnight. Other minor harmonic components are also occasionally important, the first maximum of the eight-hour component probably occur- ring about 6.30 a. m., and that of the six-hour component about 3 or 4 A. M.; but for these two epochs the results are not always concordant. (3) Though the materials are insufficient for any general con- clusion, the weaker shocks seem to be subject to a more marked diurnal periodicity. (4) In the case of after-shocks of great earthquakes, the diurnal periodicity is as a rule strongly pronounced. The maximum of the diurnal period occurs within a few hours after midnight, but the epochs of the other components are subject to wide variation, possibly on account of the short intervals over which the records extend. A special feature of after-shocks is the prominence of the eight-hour and four-hour components. The epochs of the first four components representing the diurnal variation of seismic frequency are compared in several cases with those for barometric pressure and wind velocity. While the variation of the former cannot be attributed exclusively to either of the latter phenomena, it seems not improbable that the diurnal periodicity of ordinary earthquakes may be due chiefly to that of wind velocity, and the diurnal periodicity of after-shocks chiefly to that of barometric pressure.—Proc. Roy. Soc., London, vol. Ix. 2. Transactions of the American Microscopical Society ; edited by the Secretary. Highteenth Annual Meeting held at Cornell University, Ithaca, N. Y., August 21, 22, and 23, 1895. Volume xvil, 376 pp., Buffalo, 1896 (The Wenborne-Sumner Co.). —This volume gives a full account of the Eighteenth Annual Meeting of the American Microscopical Society with the papers then read. Among these may be noted the following, all of which are illustrated by plates: Some modifications of stems and roots for purposes of respiration, by H. Schrenk; The lateral line system of sense organs in some American Amphibia and comparison with Dipnoans, and On the Spermatheca and methods of fertilization in some American newts and salamanders, by Dr. B. F. Kingsbury ; Comparative Morphology of the brain of the soft-shelled Turtle and the English Sparrow, by Susanna P. Gage. The next meeting of the Society will be held at Pittsburg, Pa., August 18 and 20, 1896. 3. Handbook of Arctic Discoveries, by A. W. GreEzty, 257 pp., 12mo. Boston, 1895 (Roberts Brothers, Columbian Knowl- edge Series, No. 3).—The interest and value of this little vol- ume are quite out of proportion to its size. It deals with a subject which has not only the highest interest from the scientific side, but which perhaps more than any other tends to excite the imagination of the general public. Notwithstanding its brevity it gives a well-digested and very readable account of Arctic ex- plorations from the earliest times. Its interest is increased in no 404 Scientific Intelligence. small degree from the fact that it has been prepared by an author who has himself a thorough acquaintance with the Arctic and whose own exploits in this field will never cease to excite admira- tion. 4. James Clerk Maxwell and Modern Physics; by R. T. GLAZEBROOK, F.R.S., 224 pp. 12mo. New York, 1896 (Macmillan & Co.—The Century Science: Series).-Modern physics owes to Maxwell perhaps more than to any other of recent physicists, for his keen insight into the scientific problems in which he was interested gave a rare value and originality to all that he wrote. His Treatise on Electricity and Magnetism has been before the public for many years, but no one would venture to say that this mine has been exhausted. The present volume is written by a ~ physicist well fitted to deal with a subject of more than usual difficulty and no one interested in the recent progress in physics, or in the personality of those to whom this progress is due, can fail to be profited by its perusal. 5. North American Birds, by H. Neuruine, Part XIII, pp. 193-240, Milwaukee, Wis. (Geo. Brumder).—The thirteenth part of this interesting and handsomely illustrated work has appeared recently. 6. Ostwald’s Klassiker der exacten Wissenschaften, (Wm. Engelmann, Leipzig).—The latest additions to Ostwald’s valuable series of classic scientific memoirs are the following : No. 67. Entwurf einer Theorie der Abel’schen Transcendenten HErster Ordnung von Dr. A. GOpel (1847). No. 68. Das natiirliche System der chemischen Elemente. Abhandlungen von Lothar Meyer (1864-1869) und D. Mendelejeff (1869-1871). No. 69. Ueber #araday’s Kraftlinien, von James Clerk Maxwell (1855- 1856). No. 70. Magnetische Polarization der Metalle und Erze durch Temperatur- differenz, von “Th. J. Seebeck (1822-1823). m .(m—1) a? sat WOT 1.2 a No. 71. Untersuchungen tiber die Reihe 1492 - N. H. Abel (18286). Nr. 72. Chemische Analyse durch Spectralbeobachtungen, von G. KIRCHHOFF und R. BUNSEN. (1860.) 174 pages. Nr. 73. Zwei Abhandlungen tuber Spharische Trigonometrie, von LEONHARD Huber. (1753 und 1779.) 65 pages. Nr. 74. Untersuchungen tiber die Gesetz der Verwandschaft, von CLAUDE LouIs BERTHOLLET. (1801.) 113 pages. Nr. 75. Abhandlung itiber die Herleitung aller Krystallographischer Systeme mit ihren Unterabtheilungen aus einem einzigen Prinzipe, von AXEL GADOLIN, (1867.) 92 pages. The Origin and History of Contract in Roman Law down to the end of the Republican Period. (Being the Yorke Prize Essay for 1393.) By W. H. BUCKLER. London: 1895, 228 pages (C. J. Clay & Sons). Elements of Botany ; by J. Y. BERGEN. 57 pp. 12mo. Boston and London, 1896 (Ginn & Company). Las Rocas Eruptivas del Suroeste de la cuenca de Mexico, par EZEQUIEL ORDONEZ. (Boletin d. Inst. Geol. d. México, Num. 2), pp. 1-46. Mexico, 1895. _OUR SPRING BULLETIN “Issued during April, will be sent free on application. A glance at it will convince you that our stock is overflowing with fine minerals, indeed probably never before have so many new and really choice speci- mens been offered for sale in any of the markets of the world. Send for the Bulletin and then for a ship- ment on approval. NEW MULTI-COLORED GEM TOURMA- LINES. Work is being pushed at the new Connecticut Tourmaline locality, and many fine crystals have been secured. Our former prices have been materially cut and never before have such choice gem crystals been sold so cheaply. The crystals are brilliantly terminated and often transparent throughout (!), and prices range from 25c. to $20.00. Thecrystals found recently are the most beautiful yet discovered, the blending of delicate pink, pale to deep green and rich lemon-yellow being most attractive. - fae NEW PECTOLITE IN ISOLATED CRYSTALS, from Bergen Hill is unquestionably the finest ever found, even the splendid finds of last summer at West Paterson being left deep in the shade. Wenow have about 150 very fine specimens at 50c. to $5.00, besides a number of good little pieces at 10c. to 25e. Thespecimens are made up of slender crystals 1¢ to 2 inches in length with crystals occasionally branching out singly into cavities, but more commonly loosely aggregated in bright, snowy-white radiating masses associated with richest green prehnite. WEST PATERSON. ZEOUCITES at half to one-quarter the prices charged by other dealers, and of superior quality. Beautiful pearly Heulandites, 10c. to 50c. (you would pay 25c. to $2. ‘1. elsewhere) ; large Apophyllites, 10c. to T5e.; ; splendid Pectolites, 10c. to $1.00. THAUMASITE in unrivaled specimens, 50c. to $3.50; small pieces as low as de. each. NEW FINDS OF PA. AMETHYSTS! Two visits have been made recently to the Pa. Amethyst localities and every available specimen has been purchased, about $200’s worth of very fine quality crystals being secured. They have a marvelously se and rich purple color and are frequently doubly terminated. Sizes 14 to 216 inches. Price 10c. to $7.50. ARKANSAS BROOKITES. A lot of the best specimens found for many years. Brilliant, good-sized erystals on the matrix, 25c. to $2.50. A NEW FIND OF AUSTRALIAN OPALS. 28 ny. beautiful and fiery specimens, remarkably cheap, $1.00 to $10.00 SPLENDID LAKE SUPERIOR MINERALS including fine crystallized Copper, beautiful, clear Calcite crystals enclos- ing Copper, bright native Silver, massive Datolite, etc. OTHER RECENT ADDITIONS. Lorandite, the new Thallium mineral, in fine crystals with beautiful crys- tallized Realgar, at half old prices, $1.00 to $5.00. Minium, one splendid specimen, $15.00. Kylindrite, the new Bolivian tin mineral, d0c. to $1.50. Lithiophorite ; beautiful Blue Anhydrite ; a lot of 100 fine Witherites at 25c. to $7.50. GEO. L. ENGLISH & CO., Mineralogists, 64 East 12th St., New York City. CONTENTS. Art, XXXVIII.—Carbon and Oxygen in the Sun; by J. : TROWBRIDGE ©5225. 505.0 2S yee (829 |F XX XITX.—Determination of the Division Errors of a Straight. | Seale;"by HH.) Jacopy .- 2/225) 1322 a ee 333 XL.—Studies upon the Cyperacee; by T. Horm. (With PlateTX) (8. oe 348 XLL—Bearpaw Mountains of Montana; by W. H. WeEp- anh: tu: V..\PIRSSON 2. oe TCE B51. XLII.—Réntgen Rays not Present in Sunlight; by M. C. Lza 363 — XLIII.—Potomac River Section of the Middle Atlantic Coast — Eocene; by W. B.\Ciark’ 2.2... 52. 23S ee 365, _ XLIV.—Ischian Trachytes; by H. S. Wasnineron._------ B15 XLV.--Numerical Relations existing between the Atomic | Weights of the Elements; by M. C. Lza____.._.__._1_ 886 XLVI.—Crocoite from Tasmania; by C. PauacHE ._....-. 389 SCIENTIFIC INTELLIGENCE. Chemistry and Physics—Investigations with sulphide of nitrogen, CLEVER and MUTHMANN, 391.—Manganese carbide, MOISSAN: Preparation of pure stron- tium compounds. SORENSEN, 392.—New class of compounds of metallic salts with ammonia, WIEDE and HorMaNnNn: Influence of light on the form of dis- charge of a Holtz machine, ELSTER and GEITEL: Fluorescence of Vapors, EK. WIEDEMANN and G. C. Scumipt, 393.—Interference of Electric Waves, VIKTOR VON LANG: Rontgen’s Discovery: Temperature of the Carbons of the — Electric Arc, WILSON and GRAY, 394.—Melting points of some of the metals, S. W. Homan, R.*R. LAWRENCE and L; Barr, 395. Geology and Natural History—Economic Geology of the Mercur Mining District, Utah, J. E. Spurr, 395.—Catalogue of the Fossil Fishes in the British Museum, Part III, A. S. Woopwarp, 396.—Catalogue of the Mesozoic Plants in the De- partment of Geology of the British-Museum, Part II, A. C. Sewarp: Catalogue of the Perciform Fishes in the British Museum, G. A. BoULENGER, 397.—De- scriptive Catalogue of the Spiders of Burma, preserved in the British Museum, T. THORELL: Duration of Niagara Falls and the History of the Great Lakes, J. W. SPENCER: Illinois State Museum: Oblique Bedding in Limestones, 398.— Geological History of the Chautauqua (N. Y.) Grape Belt, R. S. Tarr: Geolog- ical Literature: Norges Geologiske Undersogelse: Hruptionsfolge der triadi- schen Eruptivgesteine bei Predazzo in Siidtyrol, W. C. BROG@GER, 399.—Mica- Peridotites in Bengal, T. H. HOLLAND: -Zusammenstellung petrographischer Untersuchungsmethoden nebst Angabe der Literatur, HK. Conzn, 400.—Jadeite from Thibet, M. Bauvgsr: Dictionary of the Names of Minerals including their History and Etymology, A. H. CHESTER: Minéralogie de la France et de ses Colonies, A. LAcRorx: Mineralogy, F. Ruttey, 401.—Determination des Felds- paths dans les plaques minces, A. MicHEL-Livy: Allgemeine Krystallbeschreib- ung auf Grund einer vereinfachten Methode des Krystallzeichnens, A. Nius, 402. Miscellaneous Scientific Intelligence—Diurnal Periodicity of Earthquakes, C. Davison, 402.—Transactions of the American Microscopical Society: Hand- book of Arctic Discoveries, A. W. GREELY, 403.—James Clerk Maxwell and Modern Physics, R. T. GLAZeEBROoK: North American Birds, H. NEHRLING: Ostwald’s Klassiker der exacten Wissenschaften, 404. ) Page — — Sl a hae ch gn Pe ae, ee a g- INS iyee Car Reh Resi sc Ss lh oy hs s. D. Walcott, PR ee AO UTEEIG, J. S. Geol. Survey. ae ie. WHtA ‘¥ oe, JUNE: 1896. SR ee a : Acs “eS BAY * _ Established by BENJAMIN SILLIMAN in 1818. THE AMERICAN | JOURNAL OF SCIENCE. Epitorn: EDWARD S. DANA. ASSOCIATE EDITORS Prorsssors GEO. L. GOODALE, JOHN TROWBRIDGE, H. P. BOWDITCH ann W. G. FARLOW, or CamMBripGE, \ 1 Prorzssors H. A. NEWTON, O. C. MARSH, A. E. VERRILL 4 anp H. S. WILLIAMS, or New Haven, Prorzssor GEORGE F. BARKER, or Pumaperpuna, Prorrssor H. A. ROWLAND, or Batrimors, Mer. J. 8. DILLER, oF WasuHineron. FOURTH SERIES. VOL. I—[WHOLE NUMBER, CLI.] No. 6.—JUNH, 1896. PLATES X—XIII. NEW HAVEN, CONNECTICUT. 1896. TUTTLE, MOREHOUSE & TAYLOR, PRINTERS, 12 TEMPLE STREET 4 Published monthly. Six dollars per year (postage prepaid). $6.40 to foreign sub- scribers of countries in the Postal Union. Remittances should be made either by i proney orders, registered letters, or bank checks. ROR MINERA Eee | SYSTEMATICALLY ARRANGED IN EDUCATIONAL COL LECTIONS AND FOR GENERAL EXCELLENCE OF INDIVIDUAL SPECIMENS EXHIBITED, we were awarded FIRST PRIZES at the followtan WORLD’S EXPOSITIONS: PHILADELPHIA, 1876. NEw ORLEANS, 1884-85. Lonpon, 1887. Paris, 1889. ESTABLISHED, 1876. ne The Largest and Most Varied Stock of Minerals in the World. We supply every kind and quality of mineralogical material required for scientific or educational purposes. —~- = In no other establishment in the world is displayed such a number and variety of beautiful and rare minerals. Every attention is given to scientific classification and the correct labelling of all specimens is ouaranteed. INDIVIDUAL SPECIMENS, including many rare species of every grade and price, for museums, ~ collectors, or students, COLLEGE AND SCHOOL COLLECTIONS, comprising carefully selected and accurately labelled. specimens, ar- ranged to evenly and completely illustrate the science of mineralogy. TEACHERS’ AND STUDENTS’ COLLECTIONS to accompany text-books in class-room and study. PROSPECTOR’S COLLECTIONS for reference in the prospecting or collecting field. SPECIAL COLLECTIONS illustrating crystallization, hardness, color and other physical and chemical properties. i SERIES illustrating the occurrence of different ores; uses of minerals in the arts, agriculture, etc. ; LABORATORY MATERIAL for blowpipe or experimental purposes. SINGLE CRYSTALS in great variety and perfection, for teachers and Co MOUNTS FOR THE MICROSCOPE. Send for ‘‘ILLUSTRATED MINERAL CATALOGUE” containing com- plete price lists of collections and single specimens, laboratory material, etc., etc., and circulars mailed free. Dr. A. E. FOOTE, MINERALS. SCIENTIFIC BOOKS. 1224-26-28 North Forty-First Street PHILADELPHIA, PA, U.S A. ti mbar a ies 3. cag ping oe SAL Te eal i Or Or Evidently this method will not distinguish between selenious and selenic acids, but it will be found of much value in point of time saved in the determination of either separately, or of the total selenium in case both occur together. The kindly advice and suggestions of Prof. F. A. Gooch are gratefully acknowledged. G. I. Adams—Extinct Felide of North America. 419 Art. XLIX.—The Haxtinet Felide of North America; by Geo. I. Apams, Fellow of Princeton College. (With Plates mo, X IT.) THe following paper is a result of studies by the author in the Department of Paleontology of Princeton. It is an attempt to summarize the literature on the extinct Felide, to add such new points of knowledge as it has been possible to discover and to propose a classification for the family. 1 wish here to express my thanks for privileges of study kindly extended by Prof. H. F. Osborn and Dr. J. L. Wortman of the American Museum, to Prof. Cope for suggestions and use of material, and to Mr. Dixon and Dr. Nolan of the Philadelphia Academy for assistance in examining specimens and literature in that institution. JI wish also to acknowledge my special indebt- edness to Prof. W. B. Scott, whose valuable criticism and kindly interest have been an inspiration to me in my work, and to Mr. J..B. Hatcher, from whom I have received much information and assistance. The illustrations are by Mr. Rudolph Weber and their excellence is due to his care and skill. Osteology of Hoplophoneus Primevus. Hoplophoneus primevus is at present known from a short description of the type skull by Leidy (Geol. Sur. Wis., Iowa, Minn. and Neb. 1852), and later from a description of a speci- men nearly agreeing with the type along with a second larger skull which belongs to a different species (Extinct Fauna of Dak. and Neb.) A restoration and brief description has been published by Scott (Bull. Mus. Comp. Zoology, Harvard, 1887). The material here described consists of a nearly complete skeleton, which is well preserved (No. 10741, Princeton Col- lection) and a skull somewhat crushed (No. 11013). In addi- tion there is in the same collection a skeleton (No. 10934) not very complete but having associated with it the anterior and posterior portions of the skull which agree very closely with the above mentioned specimen and is supplemented by another lacking the occiput (No. 10540.) This latter specimen is slightly smaller but is a young skull just losing the decidu- ous canines. The principal material is that which was referred by the writer to 7. primevus in the American Naturalist for January, ’96, and corresponds with the original type. This species of Hoplophoneus is of special interest inas- much as it agrees very closely in size with Dinzctis felona, the osteology of which is known from the description by Dr. ree 420 G.I. Adams—Extinet Felide of North America. Seott (Proc. Phila. Acad., 1889). In describing it I shall make comparisons with the lynx (Lynx canadensis) and occa- sionally with the lion, since the bones of the lynx, although they are nearly the same size, do not express the feline charac- ters so well as do those of the lion. I shall not attempt to quote from previous descriptions or give minute descriptions where the characters agree with those of recent felines. The osteology is illustrated by Plate X. In general Hoplophoneus primevus is comparable in size with the lynx, although having a longer head, the distal por- tions of the limbs relatively much shorter and the tail long. The skull.—The skull is one-fourth longer from the condyles to the premaxillary border than that of the lynx. This is due to a greater proportionate length of the face and palate. In LT. primevus the distance from the condyles to the line of the upper molars is nearly the same as from that line to the premaxillary border. While in the lynx the first measurement is about the same as that of A. primevus, the latter is only about one-half as great. The brain case is not quite so large as that of the lynx and the post-orbital constriction is much more marked. The face is not only proportionately longer but wider, the width at the canines being as great as the width of the brain case at the parieto-temporal suture. The zygomata are expanded about as much as in the lynx in proportion to the length of the skull, but the space enclosed is proportionately longer and much wider because of the constriction of the brain case and the smaller size of the orbit. The orbit is consider- ably smaller than that of the lynx and is horizontally oval. The post orbital processes of the frontals and malar are short and rather stout. The face is strongly arched transversely ; seen from the side the angle it makes with the posterior portion of the skull is about the same as in the panther or lion, but it is straighter and the frontal region not so full. The bones of the skull are much thicker than those of the lynx and the processes and borders much more massive. This is seen in the inion and post-tympanic process. The latter is. trihedral and truncated distally. It is directed slightly towards the post-glenoid as if tending to approach it. The zygomatic process is strong and drooping with its glenoid portion look- ing slightly inward. The lambdoidal and sagittal crests are rather high and thick. The fronto-temporal ridges diverge at the fronto-parietal suture. The sutures of the skull are very similar to those of recent felines. The nasal processes of the frontals do not extend so far anteriorly, being separated by a considerable space from the ascending rami of the premaxillaries. The nasals are long G. I. Adams—Extinet Felide of North America. 421 and inserted well into the anterior border of the frontals, their posterior borders being broadly rounded. The bulle are seldom preserved, but judging from a cast of one they are moderately expanded. -The foramina of the skull present peculiarly primitive char- acters, as has already been recognized. The condylar and carotid are distinct from the foramen lacerum posterius. There is a post glenoid foramen. The foramen ovale enters the outer end of a deep transverse groove situated in the base of the zygomatic process. There is an ali-sphenoid canal, the posterior opening of which is in the inner portion of the above mentioned groove and its anterior opening is just back of the anterior lacerated foramen. The opening of the foramen rotundum is concealed within the alisphenoid canal. The optic foramen is in the same relation to the anterior lacerated as in the cats. There is also an ethmoidal foramen present. The palatal foramina need no special description. There is a post-parietal foramen and sometimes two. The infra-orbital is large and vertically oval. The mandible-—The mandible is quite characteristic of the Macherodonts. Its anterior portion consists of two vertical nearly plane surfaces meeting at a wide angle at the symphysis. The lateral face of the mandible is separated from the anterior by nearly a right angle marked by a distinct ridge, posterior to which is a shallow fossa for the superior canine. At this place the lower border of the mandible is produced into a moderate flange on the surface of which the fossa is continued. The symphysis is lower than the ramus and abuts against the flange on its inner side. The ramus is long and quite straight, being heaviest at the sectorial. The condyle is on a line with the alveolar border and is semi-conical. The coronoid is small and evenly rounded. The masseteric fossa is deep, the angle being well out from the plane of the coronoid. The dental and mental foramina are as in the modern cats. On the anterior or symphysial surface there are two foramina on either ramus. Dentition.—There are three stout subconical incisors which are slightly recurved and are placed in nearly a semicircle. Their posterior surface, which is a little over one-third the circumference of the tooth, is slightly flattened and separated from the anterior by a sharp line which is slightly denticulate in an unworn tooth. They increase in size outward, the external being largest. The superior canines are long, com- pressed and slightly recurved. They are implanted by a strong fang which reaches well up to the frontal bone. Their anterior and posterior cutting borders are denticulate. When the mouth is closed the canines rest in the fosse of the mandible, extending nearly as low as the flange. There is a small space Am. Jour. So1.—FourTH SERIES, VOL. I, No. 6.—JUNE, 1896. ——— 422 G.I. Adams—EKxtinet Felide of North America. anterior to the canine and one about twice as great posterior to it. There are three superior premolars; the anterior one (p’) is considerably the smallest and may lack distinct anterior and posterior cusps. The second (p’) is well developed but may not have a distinct anterior cusp. The third (p*) is the sec- torial. This tooth is different from that of the lynx or lion in not possessing an internal cusp ; the inner root, however, sup- ports a convex buttress which descends from the principal cusp. There is also an anterior basal cusp which is rather incipient and situated high on the principal cusp. The pos- terior cusp is a long cutting heel. The upper molar is rather better developed than in the modern cats and is inserted as in the lion by two roots in a transverse line. The incisors of the lower jaw along with the canines form a regularly curved series, the canines being not much larger than the external incisors. ‘The internal incisors are much compressed and in some specimens are hardly more than rudiments. They are slightly divergent and have the same general structure as the upper ones. The canine is curved slightly backward and has a rather stronger posterior border. Its greatest diameter is nearly jin an antero-posterior line. It is also denticulate when unworn, but the border soon becomes smooth. Back of the canine there is a diastema about twice as great as in the upper jaw. There are two well developed premolars, the anterior being, however, considerably the smaller. The inferior molar is a sectorial and differs from that of the modern cats in having a low heel and a somewhat variable postero-internal cusp. The teeth of the molar series differ from those of the modern eats in being more compressed and in having sharper borders which when unworn are feebly crenulated. The vertebre.—There are sixteen presacral vertebre of one specimen, all of which are considerably mutilated so that it is possible to judge only of the relative size and length of the centra. They indicate a much stronger vertebral column than that of the lynx and one which is proportionately shorter in the lumbar region. Five lumbar vertebree measure considera- bly less than any five of same region of the lynx. On the other hand, the cervicals are longer than those of the lynx and the axis particularly so. The thoracic vertebree are, as near as can be judged from comparison with those at hand, a little longer and, proportionately to the other vertebree, also more massive. From what I know of other representatives of the genus I think it safe to say that the processes were stout and well developed. The sacrum, as would be expected, is heavier and wider than that of the lynx and its centrum is not so much depressed. The caudals are not preserved in the speci- mens of this species, but from the other species it is evident that the tail was long like that of the recent cats. Gok. Adams—Extinet Felide of North America. 423 The pelvis.—I have only fragments of the pelvis, but they give an idea of its relative size and strength. ‘The acetabulum is one-fifth greater in diameter than that of the lynx. Asa whole the pelvis is more massive and considerably longer. At the sciatic notch, however, the ischinm has about the same diameter as that of the lynx. The scapula.—The distal portion of the scapula is con- siderably larger than in the lynx, the glenoid cavity being about one-fifth larger. There is a short stout coracoid much the same as in the modern cats. No other features are pre- served in the specimens which are at hand. The fore-limb.—The humerus of H. primevus is the same length as that of the lynx, but fully one-half more massive. The head presents a large articular surface which is very sim- ilar in shape to that of the cats. The great tuberosity is par- ticularly prominent and rises considerably above the head and is well set off from it. The smaller tuberosity is low but rugose and the bicipital groove quite broad. The prominent character of the bone is the bold deltoid ridge which has a straight sharp border on the internal side extending to the great tuberosity. On the outer side, the lower portion has a similar border which runs slightly divergent from the inner border, but farther up curves toward the smaller tuberosity and becomes a mere line on the convex surface. ‘The prominence of the del- toid ridge makes the antero-posterior diameter of the humerus at its middle portion twice as great as the lateral diameter, a feature which is not met with in the Felide even in the lion. Below the deltoid ridge the anterior surface retreats rapidly as it descends, becoming an even convex surface. The supinator ridge is also very bold and is in fact a thin prominent border as far up as the lower portion of the deltoid ridge. Above it, extending in the same line but not connected with it, is a line for muscular attachment, extending to the base of the head. The ent-epi-condylar foramen is large and formed by a free arch, being but slightly depressed into the body of the bone. The trochlea is very similar to that of the lynx but slightly more oblique. The anconeal fossa is deep and large, but not perforated. The ulna is practically of the same length as the humerus. Its olecranon is proportionately much longer than in the lynx and lion, being one-fifth the entire length, while in the lynx it is not quite one-eighth, and in the lion a little over one-seventh. The sigmoid cavity is long and defined much as in the eats. Except as to proportions, the ulna presents no especial pecu- harities. 424 G.I. Adams—EKatinct Felide of North America. The radius is three-fourths the length of the humerus and is a short bone compared with that of the modern eats. Its head presents a distinct notch on the dorsal margin. The bone has sharp lines and the lower portion is quite rugose. The styloid portion is heavy and the process short. The scapho-lunar articular surface is rather smaller than would be expected. The manus.—The manus is particularly short and the digits divergent. The scapho-lunar, although not much different from that of the lynx in general structure, is wider and has a better developed tubercle. The line of union of its two ele- ments is visible on its distal surface in the facet for the mag- num. ‘The pyramidal is very seldom preserved with a speci- men, but is fortunately retained in this case. It presents a coneave surface for articulation with the unciform. On its external surface are two facets, the proximal for the pisiform and the other for the styloid process of the ulna. It also articulates slightly with the fifth metacarpal. The pisiform is well developed, articulating as in the Felidee with the pyra- midal and ulna. The unciform and magnum need no special description. In the specimen which I have the trapezium and trapezoid are absent. They are present in specimens of Hoplophoneus insolens. The trapezoid in them nearly excludes the second metacarpal from lateral contact with the magnum. In the specimen which I am describing the relation of the meta- carpals is such as to indicate the same position. In this respect the carpus differs from that of the modern eats, in the lion the articulation being very large. The metacarpals are sur- prisingly short, being only about two-thirds as long as those of the lynx. The first metacarpal is about as much reduced as in the lynx or lion. The phalanges are large and the unguals have well developed hoods which are usually preserved. The hind-limb.—The femur is the same in length as that of the lynx, but of course much heavier, and its extremities are rather larger in proportion to the strength of the shaft. The head is presented slightly more inward and forward. It pre- sents about the same relative articular surface as the lynx and has a deep pit for the round ligament. The great trochanter is separated from the head by a more distinct notch, which is deepest at the neck. The digital fossa is deep and the posterior border of the great trochanter is reflected over it fully as much as in the lynx. The second trochanter is very prominent and its position is quite different from what it is in the modern eats. It is relatively farther below the head and not as much removed from the inner border of the femur. When the bone is viewed directly from in front a considerable portion of it is seen, while in the recent cats it is concealed by the shaft. The femur also differs from that of the recent cats in having a G. I. Adams—Extinct Felide of North America. 425 distinct third trochanter which is connected above by a ridge with the great trochanter and is continuous below with the linea aspera externa. The shaft is more rugose than in the lynx and is not quite as straight. The patellar surface is broad and shallow and the condyles large. The tibia is absolutely and relatively shorter than in the lynx, being not quite four-fifths as long as the femur, while in the lynx they are of the same length. In the lion the tibia is five-sixths the length of the femur. The condylar surfaces are large and are separated only by a low spine. The shaft of the tibia is compressed laterally and the cnemial crest is high. The anterior tuberosity is rather far from the condyles. The distal portion is quite different from that of the lynx or lion, inasmuch as the articular surface is very oblique and the astra- galar groove only slightly indicated. The malleolus is heavy and straight. A distinct ridge rises on its anterior surface and extends a short distance up the shaft. . The fibula needs no special description. It is quite stout and has sharp lines for muscular attachment and is free the whole length of the shaft. Its distal end presents a large articular surface for the calcaneum. The pes.—The calcaneum is slightly shorter than that of the lynx owing to its tubercle being not quite so long. The sus- tentacular portion is situated rather farther distally and the facet is presented more nearly upward. There is no facet to support the head of the astragalus asin the lynx. This last mentioned bone extends much farther distally than in the lynx although the neck is not proportionately longer. The astragalus of Hoplophoneus is larger, particularly the body por- tion. Its trochlear surface is only slightly grooved and its outer border very oblique. It is in these two respects very primitive. The cuboid is somewhat larger than that of the lynx, while the navicular is considerably wider and the tubercle not so much reflected upon the head of the astragalus. Its distal surface presents three distinct facets for the cuneiforms. In these bones the relationship as regards the articulation with the metatarsals is similar to that of the trapezoid and magnum in the manus, the second metatarsal being nearly excluded from any lateral articulation with the ectocuneiform. In the modern cats this wedging in of the second metatarsal is a very great element of strength. The ento-cuneiform is not so much reduced as in the lynx since the first metatarsal is not so rudimental. The length of the metatarsals is proportionately even shorter than the metacarpals, being only four-sevenths as long as those of the lynx. The first digit is not much reduced and carries an ungual. 426 G.I. Adams—Lxtinet Felide of North America. Summary. Hf, primevus differs from the recent Felide and the lynx in particular, in the following points: The skull is Machzerodont, is large in proportion to the body and long anteriorly. The brain case is relatively much less expanded and the post orbital constriction is very marked. : The mandible has a distinct vertical anterior face and a moderate flange. The post-tympanic is large, sub-cylindrical and shows a ten- dency to approach the post-glenoid process. The zygomatic processes are drooping. There are distinct carotid, condyloid, post-glenoid and post- parietal foramina and an alisphenoid canal. The dentition is I8C1Pm3M1. The superior canine is very long, recurved, compressed and when the mouth is closed extends nearly as low as the flange of the mandible, resting in a slight fossa. The superior sectorial has an anterior basal cusp but no internal, the internal root supporting instead a convex buttress which descends from the principal cusp. ‘The inferior sectorial has a small postero-internal cusp and a_ low heel. The incisors are subconical and slightly divergent. The inferior canine is not much larger than the external incisor. All the teeth when unworn are denticulated or feebly crenulated and the superior canines are permanently so. The skeleton is about the size of the lynx but more massive as in the lion. The cervical region is rather long and the lum- bar short as compared with the lynx. The tail is long as in the lion. The humerus has a strong massive deltoid ridge, the femur a third trochanter. The ulna and radius and the tibia are proportionately much shorter than in the recent Felide. The pes and manus are very short and broad. The astra- galus is only slightly grooved and its tibial surface 1s oblique. The scapho-lunar shows the line of union of its two elements. The second metacarpal is nearly excluded from lateral articula- tion with the magnum and the second metatarsal from lateral articulation with the ecto-cuneiform. | The unguals have heavy hoods and were retractile. The position of the feet was as in the modern eats. A comparison with Dinictis felina. The osteology of this species of Dznictzs, as has already been stated, has been described by Scott. In addition to the material G. I. Adams—Extinet Felide of North America. 427 which was known at that time, the Princeton Museum contains besides duplicate portions of the skeleton, a humerus and a few bones of the carpus. These need no special description for my purpose here. The genus Dinictis stands ancestral to the genus Hoplophoneus and the species felina agreeing so closely in size with H. primwvus makes a comparison very interest- ing. I shall state here the points in which PD. felina differs, althongh many minor points which would be apparent to the eye are necessarily omitted. The skull is slightly larger and is higher in the frontal region. The mandible has a smaller flange. The glenoid process is not so low and the post-tympanic does not show a tendency to approach the post glenoid. The dentition is 18C1Pm3M4. The incisors are small, spatulate and in an even transverse row. ‘The outer one is con- . siderably larger than the others. The canine is larger at the base and not quite so long. The superior sectorial has an inner cusp but no anterior basal. The inferior sectorial has a better developed heel. The second lower molar is much reduced but _ constant. The foramina are the same. The skeleton is of nearly the same size but not so massive. The limb bones have more slender shafts and not quite so large extremities. The ulna and radius and the tibia are not so short in propor- tion to the lengths of the humerus and femur, which are almost identical in length with those of HZ. primevus. The manus and pes are narrower and longer. The unguals have weak hoods but were retractile, although perhaps not perfectly so. The generic distinctions are to be found in the structure of the sectorials, the character of the incisors and the unguals, although in distinguishing the two the other points above men- tioned can be relied upon. Measurements. H. primeevus. D. felina. Length of skull condyles to premaxillary ! mondemp ee0s 22 2 Docu. Oa he Stele 148 154™™ Length of cranium to anterior rim of orbit... 99 108 enethvol bony palate 22-2342. 22S) A 79 72 Breadth of bony palate at sectorials -__..-- 61 69 Breadth of skull at canines ___-.-- i Mea Es 4A” 50 Breadth of skull at post-orbital constriction, 35 33 Length of mandible from condyle to lateral MIEISORMcE me pee Ak i. Oa See 112 eo y 428 G.I. Adams—Extinet Felide of North America. H. primevus. _—OD. felina. Length, of ,homeras 2.25 %)233-2 42 ha aoe See 12. ONO O eh iL) ee SUN Toney POMS Ly MEN 163 toe TAC Ame Tee his bs OL oy cas 122 AT PSTN ap eh a, ey ae 185 190 Plata ee | CN gee eee 168 CAlCAMEUIN ei pt ae et eee ane 43 43 MLeCAtATeal LV. ores © ate ee aii 53 mletacanpal LV" sect ser ee 38 a The Hoplophoneus Series. Hoplophoneus primcevus Leidy and Owen. This species was the first Machzrodont found in North America and was described as Machcerodus primcvus and later as Dre- panodon primevus. The establishment of the genus Hoplo- phoneus Cope, removed it to that group. In Leidy’s description in the Ancient Fauna of Dakota and Nebraska a skull is referred to this species which is considerably larger than the original type and quite different from it. The species as here used is as limited by the writer in the American Naturalist, January, 1896. ‘The description of the osteology preceding this makes further men- tion of it here unnecessary. Dental formula I3C1Pm3M1. Hoplophoneus robustus Adams. This species was proposed as representative of the larger skull, which was referred by Leidy to H. primevus. As compared with that species, it shows an increase in size and the skeleton is more massive. The skull is relatively large and the first superior premolar (pm’) more reduced and in old specimens may be absent. To this species should be referred the specimen determined by Osborn and Wortman as H. primevus (Bull. Amer. Mus., 1894, p- 228). The figure of the skull and the measurements here given are from that specimen. I3C1Pm2;2M1. Length of skull, condyles to premaxillary border, 180™™ ae humeras fic 2 see 2s ee ee 170 ee Wna pce et ho oe eee s¢ MON UNG Sasa nce al cya dee ee 195 2 tibiae ee re 160 Hoplophoneus occidentalis Leidy. This is the largest species of the series. It was first proposed by Leidy on a fragment of a mandible (Extinct Fauna of Dakota and Nebraska). It is best known from a fine skull and nearly complete skeleton described by Williston (Kansas University Quarterly, June, °95), as Dinotomius atrox, which name, as has been shown by the writer, 1s a Synonym (Amer. Nat. , January, 96). The dentition is [3 301Pm3 Mi. The inferior sectorial is very strong and thick at the base, the postero-internal cusp is wanting and the heel is reduced. G. I. Adams—Extinect Felide of North America. 429 Length of the skull, inion to premaxillary border, 260™™ “ INUNITE RUS eee eke Ss 240 cs Ci 1a eee ee ee 237 Hoplophoneus insolens Adams. The determination of the skeletal characters of 4. occidentalis made it apparent that the specimen determined as such by Osborn and Wortman (Bull. Amer. Mus., 1894) is a quite distinct species, and it together with material in “the Princeton collection was the basis of this species intermediate between 1. robustus and H. occidentalis. As compared with A. robustus the skeleton is much larger, the limb bones being longer but not much heavier, the extremities being relatively smaller. The dental formula is 13C0iPm3M1. Length of skull, condyles to premaxillary border, 190™™ Pana SI a sav Ate bit fe ae OO cs ieee RAC R Ue ae ee ef Oe ad PEI C IVD 8 eee OS ba ea os PUN 8 ee Ee 2: a Eee ee 250 ee EN Oy 1G pat SS SURES 5 Se Ne Weare Saeh pear eee 188 Hoplophoneus oreodontis Cope. This species is the type of the genus. It was first described by Cope under the genus Macherodus, but better material enabled him to separate from that genus. The specimen figured in the series is interesting as supplementing the material already known. This skull (number 10515 in the Princeton collection) differs from that described by Cope in showing the roots of a much et tea! pm’, thus changing the dental formula of the species I3C1Pm253 M. ie It is the smallest species known from the White River. sath of skull, condyles to premaxillary border, 135"™ Hoplophoneus cerebralis Cope. This is the smallest species and the most peculiar. It is the only representative of the genus thus far found in the John Day. It is peculiar in showing a short temporal space, a very convex profile, and a nearly vertical and abrupt occiput. The superior sectorial has a better developed anterior basal cusp than the other Species; in this respect it approaches Macherodus, as has been noted by Cope. Dentition I°/C’/Pm’M’/. The species is known only from a skull. Length of skull, condyles to premaxillary border, 120™™ There are thus six species of Hoplophoneus disregarding, 4. strigidens Cope, which being known only from a fragment of a canine which presents an unusual form, is not characterized by any features which refer it to that genus rather than any other. The series of skulls figured in Plate XI show a great variation in size and a study of the skeletons shows a like variation. 480 G.I. Adams—Fatinct Felide of North America. The Dinictis Series. Dinictis felina Leidy. | This is the type species of the genus and is well known from the original description by Leidy, and the osteology by Scott. A summary of its essential differences from -Hoplophoneus is given in this paper, consequently no further discussion will be given here. The skull figured is a well preserved specimen in the Princeton Museum (number 10972). Dental formule 1301 Pm3M3. Length of skull, condyles to premaxillary border, 163™™ oC humerusicci sverse) wos) oe 172 a femur. 2328 bs re 190 “2 tibia c 2 oS AAS ee 168 Dinictis squalidens Cope. This species was first described from a portion of a deciduous superior canine and a fragment of a mandible supporting the deciduous dentition. Later a mandible containing the permanent dentition was referred to it. The skull figured in the series (number 11379 in the Princeton Museum) contains the permanent dentition except the superior canines, which are just on the point of being replaced. Its reference to D. squalidens from the char- acter of the mandible is not to be doubted. The specimen although immature is considerably smaller than D. felina and differs from it sufficiently to warrant its reference to a distinct species. Length of skull, condyles to premaxillary border, 140™™ Dinictis fortis Adams. This species was described principally from a skeleton. D. bombifrons, which, was described from a skull, has since been deter- mined to be a synonym of the former (Amer. Nat., January, ’96, p- 50, foot-note). Further material has shown D. fortis to be quite distinct from D. felina, and now that the skull is known the species may be considered as established. The skull here figured is the one originally described as D. bombifrons. Length of skull, condyles to premaxillary border, 185™™ RS NUNICTUS!. se cceee ee Stee 192 Tela 2 bes As he 19] ec TOMUL 204 etek See 205 ef tibia: 22 RGR Se a ee 186 D. cyclops Cope. With this species we take up the John Day forms. The out- line drawing here given is from the type skull figured by Cope. The striking features of it are its convex profile, the round orbit. and the short temporal space. In a general way it-is comparable with Hoplophoneus cerebralis, also from the John Day. Lengtk of the skull condyles to premaxillary border, 150™™ G. 1. Adams—Eztinet Felidew of North America. 431 Dinictis brachyops (Pogonodon brachyops Cope). As will be further pointed out in this paper, Pogonodon must be considered as a synonym of Dinictis. The tooth structure is identical and the dentition differs only in the absence of the second inferior molar, the reduction of which is indicated in other species of Dinictis. The figure given in the series is from the fragments of the skull described by Cope. The posterior portion of the skull is reversed and the restoration of the outline is conjectural. The skeleton is in part known and is very Dinictis- like. This species is from the John Day. Dinictis platycopis (Pogonodon platycopis Cope). This is the largest of the series and with it is probably reached the culmination of the Dinictis type as regards size. The profile of the skull was probably more convex than is indicated in the figure, which is taken from Cope’s Tertiary Vertebrates, since the specimen is slightly crushed. It is comparable with HZ. occiden- talis although exceeding it in size. ‘The skeleton is not known. The type specimen is from the John Day. Length of skull, condyles to premaxillary border, 935mm The foregoing series of Dénictis species, of which the skulls are figured in Plate XII, although less perfectly known than the Hoplophoneus series, iscomparable with it. The genus Dinictes stands ancestral to the genus Hoplophoneus and, from what we know, seems to have been more conservative. In both there is a striking gradation in size. Hoplophoneus is greatly diversi- fied in the White River. Ddnzctis remains are not so abundant -and three of the six species are known only from the John Day, while only one species of Hoplophoneus is known from that formation. The range of the species in so far as I can report them is expressed in the following table. WHITE RIVER. JOHN DAY. Titanotherium Oreodon |Protoceras Beds. Beds. Beds. Winretiqtortis — 222 fs x x Pee MMe aes Se ls x x ae ascualidens 3: os. 5. 52 x Bier ORAChYVOPS 925. 2. = 5 x PeMeplaLyeOpIs: 2) _ = Fees x PEC VICLONS eas Bey x Hoplophoneus primeevus- ---- x a oreodontis ___-_ x Be robustus ...._- x a msolens 2. - x x ‘f occidentalis __- x ss cerebralis -___- x 432 GL. Adams—Extinct Fede of North America. Nomenclature and Synonyms. The following short account of the history of the genus Macherodus may not be out of place here, since this genus gives its name to a subfamily and its priority seems to be some- what questioned. Isolated teeth were first noticed by Cuvier in 1824, to whom specimens discovered in the Vald’ Arno were exhibited by Nesti. From evidence relative to their association with re- mains of Ursus, Cuvier was induced to refer them to that genus under the name Ursus cultridens (Supplement to Ossa- mens Fossiles, vol. v, Pt. ii, p. 517). The first description is due to Nesti, according to M. de Blainville, who cites his ‘“‘Lettera terza dei alcune ossa fossile non peranco descritte al Sogn. Prof. Pali Savi, Pisa, 1826,” in which the name Ursus trepanodon is used, whence later the genus Drepanodon, from the species name ¢repanodon, was evidently made. Later M. Bravard described Felis meganteron (now known as Machero- dus meganteron) and conjecturally restored a portion of a skull by adding to it a canine of the character previously described as Ursus cultridens, referring it to the genus Felzs under the specific name cultrzdens proposed by Cuvier (Mono- graph de deux Felis, p. 148, 1828). Kaup in Description d’Ossemens fossiies der Muséum de Darmstadt, 1833, laid stress on the differences which the falciform canines present as compared with known bears and felines, pointing out their dif- ferences from the teeth of other carnivores in the curved form and denticulate margins and proposed a distinct genus, dMa- cherodus. Am associated incisor he referred to a new genus Agnotherium, not recognizing that it probably belonged to the same individual. The real affinities had been recognized by Bravard, but Kaup was the first to propose a generic name for this type of dentition. In De Blainville’s Osteographie, under a description of Felzs meganteron (Macherodus meganteron), a very full discussion is given in which he presses Bravard’s claims. Pomel has also sought to substitute the species name meganteron for Kaup’s Macherodus. Finally Bronn in Lethea Geonostica has at- tempted to combine Smilodon Lund, and Macherodus Kaup under the generic name. Drepanodon, using Nestis’ species name. The first Machzrodont fossil found in North America was described by Leidy and Owen as JMacherodus primevus. Later Leidy used for this species the generic name Drepanodon. Likewise Cope first described species under this genus Macher- odus. Later, however, he removed them all to new genera. In referring to European species he used the genus Drepanodon G.I. Adams—Eatinect Felide of North America. 483 with Macherodus as a synonym, but retained Smilodon as a distinct genus (Extinct Cats of North America, Amer. Nat., 1880). It 8 not been definitely shown as yet that there are any representatives of the genus J/achewrodus in North America. In the American Naturalist, 1887, Cope proposed JZ. catacopis on the anterior portion of a mandible having a moderate flange and containing the roots of the incisors and a canine with a posterior denticulate border. It is not distinguishable from Hoplophoneus however. Cragin in Science, 1892, proposed Jf. crassidens on feline remains from the same beds and Williston has since (Kansas University Quarterly, 1895) referred other bones from the same locality to this species. None of the species can be said to be established as belonging to the genus Macherodus since the generic characters of dental formula, structure of the molar series and the basi-cranial foramina are not determinable in the specimens thus far found in the Loup Fork. : Smlodon is distinctively an American genus. It was estab- lished by Lund, 1842, on a specimen from Brazil (S. neogaeus K. Danske Vid. Selsk. Copenhagen). Leidy described a frag- ment of a maxilla containing a sectorial under the name Trucifelis fatalzs. In structure the tooth agrees with Smzlo- don and a second specimen induced Cope to refer it to that genus. Smilodon jfloridanus Leidy is known from a well pre- served skull from which the teeth have been lost. The genus Husmilus was established by Gervais (4. perar- matus, Journal -de Zoologie, 1875). Previously Filhol had described Jf. bidentatus, which is synonymous, consequently Gervais genus and Filhols species are accepted. Dinobastis Cope, although founded on very limited material, will probably prove to represent the latest development of the Macheerodont type in North America. The genus is based upon the absence of the internal root of the superior sectorial. The genus Denectis has remained well defined since first proposed by Leidy. To it, however, should be referred the two species of Pogonodon described by Cope. The genus Pogonodon as proposed by Cope does not differ from Dinictis as regards tooth structure, and the absence of the second inferior molar, which in Dinictzs is much reduced, is not suffi- cient grounds for retaining it as a distinct genus, since in several specimens of /znzctis it is variable in size and in one is absent from one side. The genus Dinictis as thus constituted shows the reduction of this tooth, just as Hoplophoneus shows the reduction of the superior premolar. The genus Hoplophoneus proposed by Cope embraces, besides new species, others previously described under the generic 434 G.I. Adams—Extinet Felide of North America. names Macherodus and Drepanodon. Dinotomius atrox Williston is a synonym of ZH. occidentalis, as has been shown by the writer (Amer. Nat., January, 1896). The genera Archelurus and Vimravus have remained as first defined. Our knowledge of North American fossil forms of the true Felingz is very limited. lLeidy has described Pseudelurus intrepidus, referring it to the European genus established by Gervais. There are several imperfectly known species of Felis. Leidy has described /’ atrox and /. augustus. Felis impervalis Leidy is not sufficiently characterized to be retained. Scott has described ?./. maxima from a humerus found in the - Loup Fork of Kansas. It will be remembered that remains from that deposit have been described under the genus Macherodus, and it is not probable that this humerus belongs to a different genus although further material must be found before the affinities of these forms can be determined. The genus Uncia Gray cannot be distinguished with certainty in fossil forms and is here included in felzs. The Dentition of the Felide. The modifications of the dentition of the Felidee consist, on the whole, of the reduction of the number of teeth until almost the entire function of the molar series is performed by a single large sectorial in each jaw and, accompanying the reduction in number, an increase in size and complexity of structure whereby the individual teeth become more perfect- cutting instruments.* The specialization of one tooth as a sectorial to the ex- clusion of the others appears to be due to the following causes. It is to be observed that when a cat devours a carcass it cuts off masses by a shearing action of the jaws. In so doing the part to be divided is brought to the canthus or angle of the soft wall of the mouth, which is just at the front of the masseter muscle, at which point the greatest amount of force is gained, since the weight is brought immediately to the power. The first inferior molar situated at this point can be most effectively used and has developed into a sectorial. In- asmuch as the inferior and superior teeth do not directly oppose each other but form an alternate series, the first inferior molar more nearly opposes the last upper premolar and this tooth has become the superior sectorial. The shortening of the jaw which accompanied the reduction of the molar series has brought the canines nearer to the powerful muscles of mastication and they are as a result more effective weapons in the Felide than in other carnivores. * Origin of Specialized Teeth of the Carnivora, Cope, Amer. Nat., p. 171, 1879; Mechanieal Genesis of Tooth-forms, Ryder, Proc. Acad. Nat. Sci., 1878. G. I. Adams— Extinct Felide of North America. 485 -The cause of the reduction of the molar series seems to lie in the following facts. If the jaws of one of the primitive Felidee are examined it will be found that when they are closed the teeth which first come in contact are the sectorials, because of their elevated crowns. The premolars being further from the condyles are the last to come in contact. The function of seizing and holding is performed by the greatly developed canines and the function of cutting by the sectorials, hence no important function devolves upon the anterior pre-molars and posterior molars. Inasmuch as the lower series bite in front of the upper, it will be seen that the last inferior molar and the first superior premolar are only partially opposed by the teeth of the opposite series, conse- quently they are less functional and are the first to disappear. The most primitive form of superior sectorial among the Felide consists of a principal, posterior, and antero-internal cusps, as is seen in Dinictis. To this form there is added in Hoplophoneus an incipient anterior basal cusp which in Macherodus is well developed. In Smilodon there is a second anterior basal cusp. The internal cusp present in Dinictis disappears in later genera, where it is represented only by a con- vex buttress, which descends from the principal cusp, and, as a more perfect shearing action is developed, this becomes less prominent until in Dinobastis it is absent and the internal root which supported it is lost. The inferior sectorial in the most primitive genera consists of a principal, anterior, posterior and postero-internal cusps. The fate of the internal cusp is similar to that of the internal of the superior sectorial. Since it meets with no opposing cusp it is soon lost. The posterior cusp or heel is reduced in genera in which the superior molar is rudimental and does not oppose it. In Lusmilus dakotensis, it becomes a mere sharp line on the ' posterior border of the principal cusp. The function of the inferior sectorial devolves chiefly upon the principal and anterior cusps, which are well developed. As the number of teeth was reduced the individual teeth become larger and the premolars developed posterior basal cusps by the elevation of the cingulum, as is seen in felis. The incisors also became more robust in the Machzrodont genera. In Dinictis they are small and form an even series. In Hoplophoneus they are more robust, and in Dinobastis we see their most specialized form. In this genus they have minute ~ basal cusps and their margins are crenulated. The development of saber-like canines characteristic of the Machzerodont type is easily traced in Archelurus, in which the inferior and superior canines are sub-equal; in “urogale, where the posterior border is denticulate, and then in Dénctis, where both the anterior and posterior borders are denticulate. ee ee ee 436 G.I. Adams—LKxtinet Felide of North America. With the higher genera they attained an enormous develop- ment, such as would seem to have been a positive hindrance in biting and seizing. In the true cats the canines remained sub-equal although developing to a great size. The development of a flange on the anterior inferior portion of the mandible is to be correlated with the lengthening of the superior canines. Its function seems to have been to pro-. tect these teeth, but in the latter genera the length of the flange evidently was not as great and the canines extending below the mandible were effective weapons when the mouth was closed. The most primitive forms probably possessed no flange since the canines were short and sub-equal. In Arche- lurus wesee the first indication of a developing flange in the obtuse angle of the mandible and the shallow fossa in which the superior canine rests. The evolution of the Felide is best indicated in the characters of the dentition. It is upon the dental formule and dental structure that generic distinctions rest, and a careful study of these points will reveal the genetic and phylogenetic relation- ships of the family. The Succession of Genera. There are two distinct types of development among the Felide. Of these the Machzrodont type seems to have had its origin in America, since with the exception of lurogale and possibly Macherodus all the genera have been found here. The Old World was probably the home of the true cats, as it seems to be to-day. The only representative of them in America previdus to Pliocene times was Pseudelurus, of which genus only one specimen has thus far been found. This genus may have been acquired through Oligocene intermigra- tion, and the genera #lurogale and Machairodus are proba- bly European descendants of American forms. | Any attempt at a phylogenetic arrangement of the Felide must coincide with the occurrence in time, the order of reduction of the dental series, and the development of the individual teeth, particularly the sectorials. It is proposed herein to show that there is a succession of genera of the two types above mentioned which meets these requirements. Of the Macherodont type the somewhat problematical form from the Bridger described by Wortman as ? Patriofelis leidyanus is the most primitive. From it Dinictis* is deriv- able, through the reduction of the size of the postero-internal * D. fortis has been described as having the second lower molar very rudi- mental. In a specimen in the American Museum this tooth is absent from one side. In D. paucidens described by Mr. Riggs in the Kansas University Quar- terly, April, 1896, it is absent from both sides. The dental formula of D. fortis is thus M ai D. paucidens probably being a synonym. G. I. Adams—Extinet Felide of North America. 487 cusp of the inferior sectorial and a modification of the heel. Thus is reached the most primitive species of the genus. Within the limits we can trace the reduction of the second inferior molar from a small two-rooted tooth to a mere tubercle and finally find it entirely absent in D. platycopis. The genus Hoplophoneus is well separated from Dinictis through the development of an anterior basal cusp on the superior sectorial and the loss of the internal cusp. The internal cusp had already shown some signs of reduction in certain specimens of Dznzctis in that it had become lower and less distinct. Moreover, /oplophoneus has lost the second inferior premolar and within the genus we can trace the reduction and final disappearance of the second superior premolar. The postero-internal cusp of the inferior sectorial, which had shown signs of variation in Menzctis, is absent in the higher species of Hoplophoneus. ‘The rapidly diminishing dental series reaches its maximum reduction in Husmilus. Of this only the inferior dentition is known. It differs from Hoplo- phoneus occidentalis in the loss of an incisor, the reduction of the heel of the sectorial in £. bedentatus and its absence in ff. dakotensis. This is the only genus in which the incisors have suffered a reduction in number, although in some speci- mens of Hoplophoneus the internal ones are much reduced. With the formula 13C1Pm2M?1 the most stable form of Macheerodont dentition seems to have been reached, since it has given rise through the addition of basal cutting lobes to genera which persisted until Pleistocene times. The genus J/achero- dus differs from the genus Smzlodon in having a second basal lobe developed on the molar series. They are further distin- guished by the fact that in Smzlodon the post-glenoid and post- tympanic are codssified. The absence of the entepicondylar foramen in the specimen of S. necator figured by Cope* can- not be considered of generic importance since Burmeister has shown that it is present in three specimens which he has described.t Its absence in the case above mentioned is prob- ably due to individual variation. The most specialized genus as regards the structure of the teeth is Dinobastis. In Hoplo- phoneus the internal cusp of the superior sectorial has disap- peared and the root supports a convex buttress. This is also the case in Machwrodus and Smilodon, but in Dinobastis even the internal root has disappeared. The incisors also present a peculiarity, which is seen in some specimens of J/uchwrodus and Smzlodon, in the possession of minute basal cusps and erenulate margins. With this genus the Macherodonts cul- minated in specialization of tooth structure as they did in * Amer. Nat., 1880, p. 857. + Description physique de la Republique Argentine, Text in Atlas, p. 106. Am. Jour. Sc1.—Fourtu Series, Vou. I, No. 6.—JuNnsE, 1896. 4388 G.I. Adams—Extinct Felide of North America. Eusmilus in the reduction of the dental series, and with this form they become extinct although it is probable, judging from fossil remains, that the dominant genus in Pleistocene times was Smilodon. The genera Archelurus, dvlurogale and Vimravus seem to represent specialized forms of more primitive genera, forming a side line of development. Of these Archelurus is most primitive. It is not directly derivable from any known form nor does it stand ancestral to the genera of Machzerodonts previously discussed. The superior sectorial consists of a prin- cipal and a posterior cusp. There is a slight convexity in the position occupied by the anterior basal cusp in Hoplophoneus. The same thing is seen in Vimravus, but it does not develop into a true cusp. The principal cusp has a convex internal buttress as in Hoplophoneus. The inferior sectorial consists of an anterior, principal and posterior cusp or heel. From this genus 4lurogale could be derived by the loss of the first superior premolar. There is also a different development of the canine, the tooth becoming longer, recurved and acquiring a posterior denticulate margin. WVimravus differs in dental formula from 4 lurogale in the loss of the second inferior premolar. The canine, however, is the most striking feature of this genus: it is nearly straight and spike-shaped, a form which is seemingly not derivable from “urogale but rather from Archelurus. Alurogale is a European genus and although indicative of the order of reduction of the teeth in this small and seemingly aberrant group, it cannot be said to occupy an intermediate position between the other two genera. These three genera differ from other Macherodonts in the order of reduction of the teeth, all having retained the second inferior molar. This tooth is, however, on the point of disap- pearing in Vimravus, as is indicated by its absence on one side in a second specimen of LV. gomphodus. Archelurus exhibits a peculiarity which should not be overlooked, in the strange deflection or exostosis of the border of the mandible posterior to the molar series. Of this a trace is seen in Vemravus. Of the true cats there is but one genus which is known only as an extinct form, namely Pseudelurus. Felis, Lynx and Cynelurus, however, are known as fossils. Pseudelurus presents the most primitive dental formula and one which can be compared with Dinictis as regards tooth structure, but it differs in the absence of the second inferior molar. Pseudelurus probably gave origin to /élis through the loss of the second inferior premolar. Cynelurus differs from Felis in the absence of the internal cusp of the superior sec- torial and the somewhat imperfect retractility of the claws. It forms aside development from the main line of descent. In G. I. Adams—Extinct Felidee of North America. 439 the genus Felis there is a tendency to the loss of the second superior premolar, it frequently being absent in old individuals. and thus is indicated the probable derivation of the genus Lynx. This genus occasionally loses the superior molar and so in exceptional cases we arrive at a dentition comparable with that of Husmilus, the extreme reduction, however, taking place in the other jaw. In several species of /élzs posterior cutting lobes are added to the premolars and an anterior basal cusp is indicated on the superior sectorial. Comparing the dentition with that of Machwroedus or Smilodon, it differs in the reten- tion of the internal cusp of the superior sectorial and the presence of an additional superior premolar. PHYLOGENY OF THE FELID&. Present systems of classification. Cope, in his publication on the “Extinct Cats of North America,’* has based his classification upon the basi-cranial foramina, making two families, the Felide and Nimravide, according to the following definitions. | Felidz.—No distinct carotid foramen nor alisphenoid canal, eondylar foramen entering the foramen lacerum posterius. No post-parietal and generally no post-glenoid foramina. Nimravidze.—Carotid and condylar foramina entirely dis- tinct from the foramen lacerum posterius; an ali-sphenoid canal and post-glenoid and post-parietal foramina. This classification includes Macherodont members in each family. Proclurusis considered as the genus from which the true Felidgee may have been derived through Pseude@lurus, if indeed these two genera are not the primitive members of that family. The Machzrodont members of the Nimravide are suggested as the ancestors of the Felide. It should be re- marked here that Cope’s classification, based upon structural characters, does not permit.of a phyletic interpretation, since the families as constituted by him are of polyphyletic origin as I interpret his meaning. Zittelt establishes three subfamilies, the Prozlurinz, Ma- cherodine, and Feline, defining them a follows : eect. —Dental formula 3, 4, 4, 3, lower carnassial with strong cutting talon skull elongated. Lower jaw slender, small in front with curved lower border ; ; long-legged, fore and hind feet semi-digitigrade and pentadacty]. Macheerodinee.—Dental formula 2; + 4=2 +. Upper canine large and powerful, saber-shaped, compressed with anterior and posterior cutting edges. Front pms more or less reduced. * Cope, Amer. Nat, 1880, p. 834. + Handbuch der Paleontologie, 1893. 440 G.I. Adams—Katinet Felide of North America. Upper carnassial with or without anterior basal cusp. Lower carnassial with talon. Lower jaw flattened at symphysis and separated from the lateral face by an angle. Lower border straight, more or less flared down in front. Femur frequently with third trochanter. ae 1 Felinz.—Dental formula $ + 3=3 +. Canine conical, upper and lower of about the same size; the two anterior pms lost. Upper carnassial with strong anterior basal cusp. Lower carnassial without talon. Upper molar reduced, lower one lost. Lateral surface of mandible not separated from the symphysis by an angle. | a The Prozlurine include Proewlurus Fil. and Pseudelurus Gerv., but some doubt is expressed concerning the relations of Procelurus. From the creodont Palewonictis he derives the Macheerodinze, using Wortman’s* suggestion in this respect, while from an independent source in the Creodontia he derives the Cryptoprocta and Feline through Prowlurus. Schlossert in Affen Lemuren, u. s., in criticising Filhol, expresses the opinion that Pseudelurus edwardsi is a true eat, while Prowlurus, Pseudelurus intermedius and the prob- lematical Cryptoprocta belong to another line. He points out clearly the Viverine relationships of Proclurus and rejects it as a probable cat ancestor; otherwise he accepts Cope’s classi- fication of the Felide. Proposed phylogeny. As will be seen, these authors mutually criticise each other. The determination of the relationships of Proglurus removes the possibility of its being the ancestor of Pseudeelurus and the true cats, as has been suggested by Cope, and also destroys the subfamily Proailurine as established by Zittel. The unusual method of classification employed by Cope and accepted by Schlosser in this family, whereby Machzerodont genera are grouped with the true cats, is avoided by Zittel’s subfamilies Machzerodine and Feline. These two subfamilies are the equivalents of the Macherodontine and Feline pro- posed by Gillt and his classification is here accepted, since it has the priority, however referring Cynelurus to the Feline, inasmuch as the separation into a distinct subfamily Guepar- dinee cannot be maintained. The points upon which the Gue- pardinee was established are the non-retractility of the claws and the absence of the inner cusp of the superior sectorial. The claws are retractile although less perfectly so than in other * Wortman, Ancestry of the Felidze, Bull. Ann. Mus., vol. iv, p. 94. + Schlosser, Affen Lemuren, u. s., 1887, p. 420. +Theodore Gill, Smithsonian Miscellaneous Collection, 230, Arrangement ‘of the Families of Mammals, 1872. G. I, Adams—Fixtinet Felide of North America. 441 Felidz, and the absence of the cusp is not of sufficient import- ance, as would be judged by the absence of the same character in certain of the Machzrodontine. The subfamilies may be defined as follows, slightly modify- ing Gill’s definitions : Macherodontine; superior canines large and powerful, usually saber-shaped with posterior and anterior denticulate borders, the inferior canine not greatly exceeding the outer incisor. Symphysial portion of the mandible separated from the lateral surface by an angle, the anterior inferior border of the ramus produced into a flange or making an obtuse angle with the symphysial portion. Felinge.— Canines sub-conical and sub-equal, inferior and lateral faces of ramus continuous with the symphysial. The Macheerodontine as thus constituted include a group which is in some respects intermediate between the two sub- . families but which do not stand ancestral to any genera of either of them and are not derivable from them. They are Archelurus, the European genus dlurogale and the peculiar Nimravus. Archelurus and Nimravus have been called by ‘Cope the “ false-sabre-tooths.” Arvchelurus in general appear- ance is very suggestive as to what was the probable ancestor of the Macherodontine, but its late appearance and specialized tooth structure show that it could not have given rise to any known form outside of the genera above mentioned. On the removal of the matrix from one of the specimens I found that the internal cusp of the superior sectorial is wanting in Arche- lurus. The same also proved to be true in Vimravus, although reported as present in each by Cope. Moreover the postero- internal cusp of the inferior sectorial is wanting in Vzmravus and probably in Archelurus. Thus there is found to be the same tooth-structure in these two genera that is seen in 4lu- rogale. It is therefore demonstrated that although the denti- tions of these genera are more primitive as regards numbers, they could not have given rise to other Machzrodonts or to the Feline. These three genera are related in the order of the reduction of the dental series, the character of the man- ‘dible and the structure of the canines. The canines of Arche- lurus have a convex anterior border and a posterior cutting edge, but in its present state of preservation shows no signs of -denticulations. The canine of #lurogale has the posterior border denticulate. ‘The peculiar spike-shaped canine of Vzm- ravus is very suggestive that this genus marks the end of an aberrant phylum. The subfamilies stand entirely distinct from each other and well separated from the Creodonts. Inthe subfamily Macheero- dontine there is, as has just been shown, a small group which stands. in a peculiarly isolated position. There is, how- VINOGOTYD SMILODON SMILODON MaAcH2RODUS [| Macamonus | HopLoPHONEuS | HoplLopHONEUS AELUROTHERIUM ZELUROGALE ee ARCHZLURUS \ NIMRAVUS PSqobsecoRUS Lynx Lynx 442. G.I. Adams—Faxtinet Felide of North America. ever, a specimen which may be considered as a probable ances- tral form of all the Macheerodontine and brings them a step nearer the Creodonts. It is the jaw fragment from the Bridger Eocene (number 11875 in the Princeton Collections) which Wortman has described as ? Patriofelis leidyanus (Am. Mus. Nat. Hist. Bull., 1892) as a probable connecting link between the Creodont Paleonictis and the Felide. A further knowl- edge of Patriofelis induced him to remove it from that genus (ibid., 1894). or the sake of reference I propose the generic name 4lurotherium. -dtlurotherium leidyanum, following the description given by Wortman, consists of a jaw fragment containing the third and fourth premolars and the first molar or sectorial. It also bears a distinct trace of the alveolus for the canine, the position of which is such as to preclude the pos- sibility of there being more than three premolars. The sec- torial, as will be seen from a comparison with Dinéctis, presents the same elements as that genus but has a much better devel- oped postero-internal cusp and heel, thus fulfilling the require- ment of an ancestral form. From what can be judged from the wear of the cusps and the relation of the teeth in the Felidze it seems probable that it possessed the dental formula P4Mé as was inferred by Wortman, thus exceeding the dental formula of Dznectes by a premolar. It will be noted that the dental formula of 4'lurotheriwm was probably more primitive than that of Archeelurus, thus not precluding the possibility of its being ancestral to that genus also, although probably quite removed from it. The following arrangement according to subfamilies expresses the probable line of descent of each as well as the occurrence in time of the various genera. EocENE OLIGOCENE MIOCENE PLIOCENE PLEISTOCENE RECEN?D BRIDGER WHITE RIVER Joun Day Loup ForRE DINOBASTIS EvusMILus PY NILNOGOLP HIV Jy - FELIS FELIS FELIS FELIS \ CyN2LURUS CyNALuRUS | CYNALURUS PHOSPHORITES St.. GERAND SANSAN VAL D’ARNO ; LE Puy ’ : WNIT] G. I. Adams—EKatinct Helide of North America, 448 There is a point of difference among the two sub-families which should be discussed here. It is the character of the basi-cranial foramina which was made by Cope a basis of classi- fication. ‘The arrangement of the foramina and the presence of the alisphenoid canal as found in those genera which he made to constitute the Nimravide is such as is found in primi- tive forms. It is not unnatural that the higher forms should have paralleled each other in the loss of the alisphenoid canal and the disposition of the foramina, so that in Smzlodon and Felis we find the same conditions although the two genera represent the most specialized forms of the phyla in which the basi-cranial characters are known. The two subfamilies ex- hibit a most remarkable parallelism in development. The point of divergence from a common ancestor is quite removed from the position by the most primitive of the well known genera. Indeed it is not impossible that the Felidae may have had a separate origin from the Creodonts, but our knowledge of their early relationships is too meagre to justify any definite conclusions on this point. As we know them the phyla are distinct. The following key for determining the genera of the Felidee is submitted. #lurotheriwm not being well known is con- trasted with Dinzctis, which it most resembles in the characters known. Husmilus, being known only from the lower denti- tion, is contrasted with Hoplophoneus. I. Superior canines large and powerful, usually saber-shaped with posterior and anterior denticulate borders. Inferior canine not greatly exceeding the outer incisor. Mandible with the symphysial portion separated from the lateral by an angle, the anterior inferior border of the ramus with a flange or mak- ing an obtuse angle with the symphysial portion. MacuH &RODONTIN A. A, Anterior inferior border of mandible with an obtuse angle. a. Superior sectorial without internal cusp. a.a. Superior canine recurved posterior border, not den- , ticulate. 1. Dentition 13C1Pm4M3 Archelurus. 6.6. Superior canine recurved posterior border, denticulate. 2. Dentition I8C1Pm3M3 _Zlurogale. ¢.c. Superior canine spike shaped. 3. Dentition I8C1Pm3M4 Mimravus. #. Anterior inferior border of mandible with a flange. 6. Superior sectorial with internal cusp. d.d. Inferior sectorial with strong postero-internal cusp and talon. 4. Dentition ?13C1Pm4M3 Hlurotherium. e.€ Inferior sectorial with small postero-internal cusp, talon reduced. 444 G. I. Adams—Lixtinct Felide of North America. 5. Dentition 1$C}Pm3M *— Dinictis. e. Superior sectorial without inteeal cusp. Jf. Anterior basal cusp of superior sectorial, incipient premolars without basal cusps. 6. As Dentition I8C1Pm3252M1 Hoplophoneus. Dentition 120/1Pm/1 M/1t Lusmilus. d, Anterior basal cusp of superior sectorial well-developed, no internal cusp, premolars with basal lobes. gg. Superior sectorial with single anterior basal cusp, post- glenoid and post-tympanic processes distinct. 8. Dentition 18C1 3M1 Macherodus. hh. Superior sectorial with a second anterior basal cusp, post-glenoid and post-tympanis processes codssi- 9. fied. Dentition I8C1 Pins ma ; Mt 1 Smilodon. at. Superior sequal se there ruse re) root. LO; Dentition probably like Smilodon, Dinobastis. II. Canines sub-conical and sub-equal, inferior and lateral faces of ramus continuous with symphysial. FEIN 2. #&. Anterior and inferior borders of mandible continuous. é. Superior sectorial with internal cusp. ile Dentition 18C01Pm3M+t Pseudelurus. . Dentition 13C1Pm432M?1 Felis. Dentition 8C1Pm3M17° Lyne. si Superior sectorial without internal cusp. FIGURE FIGURE FIGURE FIGURE FIGUKE FIGURE FIGURE FIGURE FIGURE FIGURE FIGURE FIGURE FIGURE FIGURE FIGURE FIGURE FIGURE FIGURE FIGURE 14, Dentition I13C1Pm3M1 Cynelurus. = EXPLANATION OF PLATES. PLATE X. Hoplophoneus primevus, all x 4. 1.—Skull (number 11013 Princeton Museum restored from number 10540.) 2.—Manus (number 10741 Princeton Museum.) 3.—Pes 4.—Femur 66 ab 5.—Ulna andradius “ 6.—Tibia and fibula is 7,.—Humerus ae PLATE XI, Hoplophoneus series, all x 4. 1 -—Hoplophoneus cerebralis (after Cope. ) y— 3.— 4,.— 5.— 6.— 6b 6c te oe oreodontis (10515 Princeton Museum.) primevus (11013 Princeton Museum.) robustus (650 American Museum.) insolens (11022 Princeton Museum.) occidentalis (after Williston.) PuaTe XII. Dinictis series, all x 4. 1. —Dinictis squalidens (11379 Princeton Museum.) Oa SS 4,— 5.— 6.— 6 be é¢ 6 cyclops (after Cope.) felina (10972) Princeton Museum.) fortis (10502 Princeton Museum.) [reversed.) brachyops (Cope’s type, Amer. Mus., posterior portion platycopis (after Cope.) A. Hague—Iqneous Rocks of the Yellowstone Park. 445 Arr. L.—TZhe Age of the Igneous Rocks of the Yellow- stone National Park ; by ARNOLD HacukE, of the U. S$. Geological Survey. THE region embraced by the Yellowstone National Park and the adjoining country to the north and east, has been a center of great volcanic activity. By far the greater part of the surrounding mountains, and the entire park plateau, have been built up by the pouring ont of vast accumulations of vol- canic material. The region, throughout Tertiary time, was undoubtedly the most active center of eruptive energy to be found in the northern Rocky Mountains. Under the authorization of the U. 8. Geological Survey, aided by an able corps of assistants, 1 have carefully studied the geological history of volcanic eruptions in the Park coun- try, and the order of succession of the different masses of lavas which make up this enormous body of igneous rocks. To a proper understanding of the geological relations of these vol- canic lavas to the earlier crystalline and sedimentary rocks, as well as their relations to the pre-existing mountain ranges, a brief sketch of the geology of the country seems necessary. In all the mountain ranges which surround and shut in the park, at least a nucleus of highly crystalline rocks of Archean age is exposed. The Tetons, to the south, consist mainly of an Archean mass, which towers high above all later rock for- mations. In the Absaroka range, stretching along the entire east side of the park and formed mainly of igneous rocks, granites and schists are exposed at the northern end, which soon pass between the later lavas. The Snowy range, which shuts in the park on the north, is largely made up of Archean schists, gneisses and granites, associated with the more recent outbursts of lava. In the Gallatin range, on the west, a body of crumpled gneisses and schists form the nucleus around which have been built up a complex mountain structure, in which the volcanic rocks play an important part in the struc- tural features of the mountains. Resting unconformably upon these Archean bodies, which formed either a broad continental mass, or a group of closely related islands, occurs a great thick- ness of sedimentary beds, made up in greater part of material derived by the slow processes of denudation of the earlier con- tinental land surfaces. These sediments, thus slowly deposited, built up during Paleozoic and Mesozoic times a series of sand- stones, limestones, and shales conformable from base to summit. ‘The basal beds, resting directly upon the Archean rocks, con- sist largely of siliceous material, passing gradually over into a 446 A. Haque-—Age of the Igneous Rocks shaly limestone, carrying a fauna characteristic of the Middle Cambrian period. These beds have been designated the Flat- head formation. Above these beds, in their regular order of succession, come the Upper Cambrian, Silurian, Devonian, Car- boniferous, Trias, Jura, and every grand division of the Cre- taceous recognized in central Montana and northern Wyoming, including the Dakota, Colorado, Montana, and the Laramie sandstone at the top. ; The Montana formation exhibits throughout its entire devel- opment a singularly uniform sandstone epoch, the argillaceous beds of the Pierre shales being poorly represented other than by a great thickness of arenaceous deposits. The overlying Laramie, on the other hand, although essentially a sandstone- making epoch, indicates frequent and abrupt changes in its sedimentation. Constantly changing beds of shales, clays, and impure standstones, afford abundant evidence in their mode of occurrence, that the material forming the beds was deposited in a shallow sea. With the close of the Laramie sandstones, the conformable Paleozoic and Mesozoic strata came to an end. The entire region was again elevated above the sea. The sedi- mentary beds were everywhere uplifted, and very large areas tilted up at high angles. A profound orographic movement took place and the region became one of mountain building on a grand scale, accompanied by plication, folding, and faulting. It was this movement that blocked out the mountains which surround the Yellowstone Park plateau, and indeed all ranges of the northern Rocky mountains. It has been designated the post- Laramie movement. This orographic disturbance was doubt- less coincident with the similar movements described by Mr.. S. F. Emmons as clearly defined in Colorado, though in the northern country no unconformities such as have been noted to the southward, have as yet been definitely recognized between the Middle Cambrian and the Laramie, as described by Mr. Emmons.* This limitation of the Laramie formation to the beds at the top of the upturned conformable series of sediments, is the line of demarcation first proposed by Mr. Clarence King.t It is in accord with the great physical break which played so important a part in the building up of the northern Rocky Mountains, and brought to a close a period of a distinctly marine or brackish fauna. In this region there are no evidences that igneous rocks played any part during Paleozoic and Mesozoic times, prior to the post-Laramie movement. There are no intrusive masses nor interbedded flows contemporaneous with the deposition of * Orographic Movement in the Rocky Mountains, 8. F. Emmons, Bull. Geol.- Soc. Amer., vol. i, 1890. + U.S. Exploration of the 40th Parallel, Systematic Geology, vol. i, p. 357. of the Yellowstone Park. 447 sediments, nor are there any beds derived wholly, or in part, from volcanic material of later age other than the eruptive masses of the Archean. On the other hand, as will be shown later, closely following and possibly coincident with the Lara- nie uplift, came the first outpouring of igneous rocks, which, beginning in late Cretaceous time, continued throughout a long period, lasting till near the close of Tertiary time. With the blocking out of the mountains, which accompanied the post- Laramie movement, a largely increased continental land area was formed. Coincident with this elevation of the land, denu- dation of the continental mass began, and deposition of fresh sediments uncomformable to the uplifted Laramie took place. Within the park region such sediments, if deposited, were either carried away by erosion or buried beneath outflows of later igneous rocks. Along the northern slopes of the Snowy range, about 45 miles north of the park, these post-Laramie sediments were deposited several thousand feet in thickness under conditions singularly favorable for their preservation. The lowest rocks resting directly upon these upturned sedi- mentary beds, are largely composed of friable sandstones derived from the degradation of the Laramie strata. They are soon followed by sands, conglomerates, clays, and grits, varying greatly in texture in different localities, and consisting mainly of material derived from the disintegration of igneous rocks. In places, they consist almost wholly of coarse, acid, andesitic agglomerates, which attain a development of nearly 2,000 feet in thickness. The voleanic nature of the great body of this ae eee was recognized by Mr. Walter H. Weed,* who first carefully studied it, and this view has been generally accepted by all geologists ‘who have since visited the region. According to Mr. Weed, the beds reach a maximum development of nearly 7,000 feet. He has proposed to designate them as the Liv- ingston formation, after the town on the Yellowstone river where they are well exposed. Since their deposition the Livingston beds have been uplifted and now lie inclined, dipping away from the mountains, hav- ing taken part in subsequent orographic movements which elevated the Rocky Mountains. The limited invertebrate fauna obtained from these beds is as yet too restricted to be of any special value in determining the geological position of the terrane. As regards the flora, however, the collections which have been made from time to time, are more satisfactory and definite as regards specific characters, and Prof. F. H. Knowl- *The Laramie and Overlying Livingston Formation in Montana, U. S. Geol. Survey Buil., 105. r 448 A. Hague—Age of the Igneous Rocks ton has been able to determine a distinct Livingston flora higher than the Laramie, but far more closely allied to a Cretaceous than a Tertiary flora. The importance of the Livingston for- mation in the geological history of the region can hardly be over-estimated, from the fact already pointed out that it is the earliest occurrence of an accumulation, either wholly or in part, of volcanic material following the uplift of the Paleozoic and Mesozoic beds. To the east and north of the Livingston formation, but farther removed from the mountains, occurs a still later series of beds made up of the disintegrated sediments of the earlier rovks, including those of the Livingston formation. Apparently they are identical in age with similar deposits found near the junction of the Missouri and Yellowstone rivers, where they in all probability he upon the Laramie sandstones without the intervention of any beds referable to the Livingston formation ; at least, the latter have not as yet been recognized in this locality. These beds along the Missouri and Yellowstone rivers have long been designated as the Fort Union formation, and have also been of special interest on account of their varied fossil flora, first carefully described by the late Prof. J. S. Newberry. This Fort Union flora is now generally regarded as of Hocene age. Returning to the immediate region of the Yellowstone Park, the Laramie sandstones are found at a number of locali- ties, but neither the Livingston nor Fort Union beds are ex- posed, and if ever deposited, have been removed by erosion or buried beneath later ignecus outflows. Subsequent to the post-Laramie movement and after a very considerable erosion of the uplifted sandstones, volcanic action broke out with great energy in the Park country. From numerous centers of eruption lavas were thrown out, submerging broad areas of country. Volcanoes of great size surrounded the Park on the east, north, and northwest. The Absaroka Range was built up by the pouring forth of vast accumulations of volcanic ejectamenta, burying the greater part of the earlier range be- neath many thousands of feet of accumulated lavas. These volcanoes followed the lines of orographic movement which took place at the close of the Laramie, along what are now the Absaroka, Snowy, and Gallatin ranges. To-day, this ancient center of eruption lies in the interior of the continent just eastward of the continental watershed which separates the waters of the Atlantic from those of the Pacific, but at the time this vast body of lavas was ejected, they built up a re- markable chain of coast voleanoes along the western shore-line of a gradually retreating ocean. of the Yellowstone Park. 449 This voleanic energy continued throughout a long period of time with intervals of comparative rest, as shown by the evi- dences of erosion of the ejected material, and followed by re- newed activity in the outpouring of fresh lavas. In many localities in the worid the age of volcanic flows has been more or less definitely determined by the fortunate preservation of a fauna or flora in some lake beds formed in a depression or basin in the volcanic rocks, and these again preserved by beds of later compact lava or mud flows. These in a number of in- stances determine the age of the underlying and overlying flows, but are of slight value in determining the relative age of other eruptions in the adjoining regions. It is doubtful if any region in the world furnishes a more complete geological record of volcanic eruption from beginning to end throughout so long a period of time as the Yellowstone Park and the ad- joining mountains. As regards the problem connected with the relative age of the different masses of igneous rocks, the prosecution of the work has been greatly advanced by the field study of the so-called fossil forests and the many localities where a varied flora has been well preserved in the tuffs and waterlain deposits. The mode of occurrence of the beds carrying a fossil flora has been carefully examined and large collections made from time to time. Prof. Lester F. Ward and Prof. F. H. Knowlton passed the season of 1887 in the Park. In the fol- lowing year, Professor Knowlton accompanied me to the field for the purpose of collecting from all the known localities, | adding largely to the store of material previously obtained. | All these collections were submitted to him for study, and ‘his | results when published will form an important chapter in the | monograph upon the Yellowstone Park, soon to be published by the U. 8. Geological Survey. All references in this paper to the genera and species found here are based npon the identifications of Professor Knowlton. The oldest extrusive flows recognized in the Park area series of light-colored coarse breccias and fine tuffs composed of hornblende-andesite, and hornblende-mica-andesite, varying greatly in texture and physical habit, but grouped together under the general description of early acid breccias and flows. It is evident that they were thrown out from different centers : of eruption, and in every instance, so far as can be told, they were the earliest eruptions from those recognized centers. In general, these early acid breccias occur as relatively small isolated bodies exposed by erosion of the later rocks. In this way they occupy the bottom of the deep valley of Cache Creek in the Absaroka Range, they are found in the deeper gorges of Tower Creek basin, and are found as far westward | he 450 A. Hague—Age of the Igneous Rocks as Sepulchre Mountain. They are, however, best exposed in the northwest corner of the Park on both sides of the Yellow- stone River, where the overlying lavas have undergone the greatest amount of denudation. Here they rest directly upon the Archean rocks of the Snowy Range, and in sharp contrast to the later lavas carry a large amount of crystalline gneisses and schists imbedded in the breccias. On Crescent Hill, just south of the Yellowstone River, these early acid breccias are well shown over 1,000 feet in height, in a steep slope facing northward. The finer tuffs, ashes, and mud flows have afforded exceptionally favorable conditions for the growth and preservation of a rich and varied flora. At Crescent Hill, Elk Creek, and on both banks of the Yellowstone River, be- low the mouth of Elk Creek, large collections of plant re- mains were obtained, which have since been studied by Professor Knowlton. From this abundant flora he has identi- fied 79 species, of which 42 species are either new to science or now specifically determined. Of the remaining 37 species that have been found elsewhere, 17, or nearly one-half of them, are already known in beds belonging to the Fort Union for- mation. Of the remaining species, 5 come up from the Laramie, 5 are found in the Livingston and Denver beds, and 11 species are also common to the Auriferous Gravels of Cali- fornia. This flora from the acid breccias is otherwise so closely allied to early Tertiary forms that Professor Knowlton has no hesitation in referring the entire group to the Eocene period, and correlating it with the Fort Union horizon. The following characteristic species have been determined from the acid breccias: Sapindus offinis, Cornus acuminata, Populus speciosa, Sequoia couttsie, Asplenium magnum, Taxites ulriki, Populus daphnogenoides, Populus xantholi- theusis, Betula iddingsu, Fagus undulata, Aralia notata. This reference to the Eocene is in accord with the geological evidence which makes these andesitic lavas carrying the flora younger than the Livingston formation, and among the earliest of the extrusive rocks of the Park. Overlying this group of early acid breccias comes a great accumulation, in places several thousand feet in thickness, of volcanic ejectamenta, and also like the earlier rocks, made up of coarse and fine material presenting great differences in texture and physical habit, whose character is determined in great part by their mode of occurrence and the distance from their sources of eruption. Like the early acid breccias, they have been grouped together into one series representing a dis- tinct phase of the volcanic phenomena of the region. Unlike the earlier breccias, however, they are distinctly basie in com- of the Yellowstone Park. 451 position and consist mainly of hornblende-pyroxene-andesites, pyroxene-andesites, and basalts. By far the greater part of this erupted material is formed of coarse agelomerates, somber in color, with interbedded sheets of compact basalt varying in extent, and of greater or less thickness. These basic breccias cover large areas of country, extending from the Gallatin Range westward across the Snowy Range, along the nothern border of the Park. They consti- tute nearly all the northern portion of the Absaroka Range. Indeed, they may be said to cover the greater part of the northwest corner of the Park. Over this extensive area the basic breccias are only seen in a few localities resting directly upon the earlier acid rocks, but where exposed, they are for the most part easily recognized by their sharp contrast in color, texture, and mineral composition. In places along their contact the acid breccias show evidence of considerable erosion before the pouring out of basic flows, but in other localities they exhibit a transition from one series of rocks to the other and occasionally a mingling of both groups. On the ridge about one mile south of Yanceys Station and within a short distance of Elk Creek, and again near Lost Creek, these basic rocks lie directly upon the acid rocks in a series of fine agglomerates and mud flows. The acid rocks present a very hilly and uneven surface, with the basic rocks frequently vine at a lower level and occupying depressions in the older ava. Possibly in some instances erosion washed down the softer material from both series of rocks, causing a mingling of both acid and basic lavas, and rendering it impossible to discrimi- nate between beds. Apparently similar conditions prevailed on the opposite side of the Yellowstone River along the north- ern escarpment of Specimen Ridge, extending eastward as far as Crystal Creek, where both series of rocks are again seen together. It is evident that the belt of country including the localities mentioned was* at one time favorable to a vigorous and varied vegetable growth. Atanumber of localities the same flora occurs, carrying well-preserved leaf impressions, together with silicified trunks of trees still standing firmly planted in the muds and breccias. The flora from this limited area contains 30 species, of which 18 are new to science. Only 2 of the new species are found in the acid breccias, and only 3 have been collected from the basic breccias. Nine of the species identified are common to both acid and basic breccias. Of the species previously described as occurring elsewhere than in the Park, most of them have a wide geological and geographical distribution, ranging from the Laramie well up through the Miocene. The 452 A. Hague—Age of the Igneous Rocks following species may be mentioned as characteristic of this horizon: Platanus montana, Quercus yanceyt, Laurinoxylon amethystenrum, Populoxylon wardir. For the present this group of plant remains has been pro- visionally designated the intermediate flora, as according to Professor Knowlton, it possesses many distinct features of its own, yet appears as a transition flora between that of the acid rocks and the flora of the basic breccia which is of later age. This flora has been referred to the Lower Miocene. Directly to the south of the region where both types of rock occur together, the acid breccias suddenly disappear, and later flows bury everything beneath an enormous mass of basic lava. In the deep ravines cut in the abrupt escarpments alone the south and west side of Lamar Valley, lies the famous Fos- sil Forest of the Yellowstone Park. The ravines or gorges expose to view vertical sections nearly %,000 feet in thickness across the bedded agglomerates, mud flows, and basalt sheets of these early basic breecias. From the base of the breccias as exposed along the valley, nearly to their summit on the top of Specimen Ridge the beds carry abundant evidences of an extinct fossil flora. They have so far yielded 70 species, more than one-half being new to science and described by Professor Knowlton for the first time. Those species that are identified as belonging to both the acid and basic breccias are mainly such species as have a wide geological distribution. This flora Professor Knowlton regards as distinctly different from that of the acid breccias, and of much later age. In the identification of.species and in its affinities it bears a close resemblance to the flora of the Auriferous Gravels of Califor- nia. It has been referred to the Upper Miocene period, and named the Lamar flora. The following species are selected as characteristic of the Lamar horizon: Platanus guillelme, Laurus californica, Magnolia spectabilis, Planera longifo- lia, Magnolia culvert, Aralia whitney?. The flora presents many of the same general characteristics throughout the 2,000 feet of beds, which were derived wholly from eruptive material. It has been found impossible as yet to discriminate between the flora of the lowest beds and those at the top. This would indicate much the same climatic and physical conditions during the entire time required for the accumulation of such a mass of volcanic material with its accompanying flora buried beneath successive layers of tufts and muds. That there were long periods of rest during the piling up of this material is evident from the great size of the trees, whose trunks are still standing at different elevations in the lavas. of the Yellowstone Park. 453 On both sides of the Yellowstone, stretching from Crescent Hill to Junction Butte, occur several isolated exposures of an acid lava, remnants of a much more extensive body, the greater part of which has been carried away by erosion. Along the Yellowstone Valley denudation of the overlying material has been so excessive that the relations of these acid rocks to other igneous rocks are somewhat obscure. In many instances it would be impossible to make out their geological position, as they lie completely isolated from other lavas. In some instances they have been preserved from erosion by eappings of basalt, but the age of these basalt flows is by no means iden- tical. They have been grouped together and designated trachy- tic-rhyolite, from certain resemblances in mineral composition to both these types of igneous rocks, although it can not be stated definitely that they all belong to the same period. For the present they have been correlated as being contemporaneous in age, and are regarded by the writer as of later age than the series of early acid and basic breccias. As regards their geological relations to other eruptive masses and the general sequence of lavas, they play an unimportant part, and tor the purposes of the present paper, which deals more especially with the age and duration of volcanic energy in this region, they are of slight interest, being limited in amount, covering small areas, and carrying no evidences of a fossil flora. Resting directly upon the irregular surfaces of the early basic breccias comes a succession of distinctly bedded basalt flows, which in places have attained an aggregate thickness of nearly 1,000 feet. They cap several of the higher ridges of the Absaroka Range, and on the west side of the Lamar Valley west of the Range, they present a grand escarpment forming the abrupt wall of Mirror plateau. From this plateau the basalts stretch southward and eastward, sharply defining the series of early acid and basic breccias from later accumula- tions of igneous rocks. These basalts have been designated early basalt flows, as they mark a distinct period in the geolog- ical history of the volcanic eruptions, although flows of basalt frequently occur in the older breccias. Following these early basalts comes a second series of acid and basic breccias similar in mode of occurrence, mineral com- position, and in the varying manitestations of the different phases of eruptive energy. Apparently the centers of volcanic action moved southward and the later series of acid and basic breccias built up the western portion of the southern half of the Absaroka Range. Unlike the early acid breccias, the late acid breccias are not so deeply buried beneath the more recent lavas, but form the tops of several high peaks, and cover large Am. Jour. Scl.—Fourts Series, Vou. I, No. 6.—JuUNE, 1896. 454 A. Hague—Age of the Igneous Rocks areas along the summit of the range. The late basic breccias buried in great part the late acid breccias, and in turn built up the extreme southern end of the Absaroka Range, or at least that portion lying within the limits of the Yellowstone Park. The basic breccias stretch southward far beyond the bounda- ries of the park, and constitute the top of the Wind River plateau, which serves to unite the Wind River and Absaroka Ranges. In places the late basic breccias are shown in precip- itous cliffs, exposing a thickness of over 2,000 feet of volcanic material. Unlike the early acid and basic breccias, neither the late acid nor basic breccias have as yet yielded any valuable contribution to our knowledge of an extinct flora. Fragments of silicified wood are common enough, but as yet no bed of tuffs or fragmental material has been found to carry a well- preserved flora. Occasionally fragmental impressions of stems, twigs, and leaves have been collected, but they are too poor for specific identification and of no special significance for com- parative purposes. It may be said that the Absaroka Range has nowhere as yet afforded any important plant-bearing beds, the localities furnishing the material lying to the westward of the range. Penetrating both the late acid and basic breccias, and conse- quently of later age, occurred extensive outflows of hornblende- mica-andesite. As these dense rocks withstand atmospheric agencies better than the underlying breccias, they often form the summit of isolated peaks in the more elevated portions of the range. These hornblende-mica-andesite flows denote a distinct period in the phases of volcanic phenomena in the Absaroka Range: After the cessation of volcanic energy which built up the vast pile of andesitic and basaltic breccias and flows, a long period of erosion followed. An enormous amount of material was swept away. Broad valleys and deep canyons were carved out and the present mountains outlined somewhat as they appear to-day. Then with renewed activity volcanic eruption again broke out and immense masses of rhyolite were ejected. These rhyolite flows changed the depressed basin lying between the mountains which surround the Park into an elevated plateau, the lavas accumulating to a thickness of nearly 2,000 feet. The appellation of Park Pla- teau has been given to this striking physical feature which em- braces an area not less than 50 miles in length and 40 miles in width. On all sides the lower slopes of the pre-existing ranges were buried beneath rhyolite, and in some instances the deeply eroded valleys in the andesitic breccias of the Absaroka range were partially filled by the later rock. Underlying the rhyo- lite at one or two localities occur limited flows of basalt and of the XY ellowstone Park. 455 again thin interbedded sheets of basalt are found in the upper portion of the rhyolite, but in general the great body of rock presents a singularly uniform mass over the entire plateau. Nowhere have any aqueous deposits or accumulations of detrital material between successive flows been recognized. Neither have any marked evidences of erosion or long intervals of comparative rest been observed during the pouring out of these rhyolites. Although no plant remains or invertebrate fossils have as yet been obtained associated with the rhyolites of the park, they have been referred on other evidence to the Pliocene period. Along the abrupt wall of the grand canyon of the Yellow- stone, just north of Tower Oreek, occurs a bed of coarse con- glomerate, made up of Archean and andesitic material. On the west side of the canyon the conglomerates rest directly upon andesitic breccias. On both sides of the canyon the con- glomerate is overlain by basaltic flows in’ which columnar structure is well developed. In the conglomerate on the west wall, and directly beneath the overlying basalt, a few vertebrate remains have been collected, which, though fragmentary, were sufficient to enable Prof. O. C. Marsh to determine them as belonging to tne skeleton of a fossil horse of Pliocene time. A careful search of the canyon conglomerate failed to find any pebbles of rhyolite mingled with the other material, showing conclusively that none was present in the immediate region when the conglomerate was deposited, as the configuration of the drainage basin is such that had any rhyolite been present, evidences of it would have been detected in the waterlain material. The basalt on both sides of the canyon is covered by rhyolite, a portion of the great rhyolite field in all proba- bility belonging to the earlier flows of this period. After the cessation of the rhyolite came the final phase of voleanic activity, the pouring out of the recent basalts. They oceur along the northern and western boundaries of the Park, skirting the outer edges of the rhyolite mass, where in broad, but comparatively thin, sheets they overlie the former rock. They are admirably shown in the region of Bunsen Peak and Gardiner River but are best developed in Falls River basin, stretching far westward into Idaho in somber, monotonous beds. Occasionally dikes of basalt penetrate the rhyolite, as shown in the escarpment of Madison canyon, but in the broad central portion of the Park plateau outflows of basalt are entirely wanting, even near such a powerful center of eruption as the Sheriden voleano. Upon the cessation of these recent basaltic outflows, after a long period of activity, eruptions of igneous rock in this region came to anend. That this activity lasted throughout the greater part of Pliocene time seems well 456 A. Hague—Age of the Igneous Rocks, ete. established, but there is no evidence of its continuance into the Pleistocene. Slight outbursts of lava may have reached the surface at different times after the coming in of the Pleisto- cene, but there is no direct evidence that such was the ease. If they did break out they were limited in extent and were only evidences of continued internal heat, such as is shown by the hydrothermal phenomena for which this region is still so- celebrated. The facts brought together here clearly demonstrate that the pouring out of igneous rocks began with the post-Laramie uplift or closely followed it, and from the time of the first appearance of these rocks, volcanic eruptions continued with greater or less energy throughout Tertiary time. It is evident that from the time of the post-Laramie uplift there was, as shown in the geological history of the region, a succession of events of great importance in the development of the Rocky Mountains, and that each period of this history was character- ized by distinct phases of volcanic phenontena. The great value of paleobotany as an aid in determining the age of the geological formations is singularly well illustrated in the re- gion under discussion. At least five distinct and important geological pericds are defined by their fossil flora, and four of them are exposed in the Park within a few miles of each other. Along the abrupt west wall of Mt. Everts, the Laramie sandstones contain several species of plants, sufficiently characteristic to determine the Laramie age of the beds. It has been shown that the Livingston formation resting uncon- formably upon the Laramie carries a distinct flora, well defined from the Laramie below and the Fort Union above. About ten miles east of Mt. Everts the early acid andesitic breccias contain a varied Eocene flora, referred to the Fort Union formation. At a number of localities within a few miles of Crescent Hill, the breccias afford a special grouping of fossil plants designated as the Intermediate flora, and re- garded as more closely related to the Lamar than to the Fort Union flora. They have been referred to the Lower Miocene. Above these latter beds in a great development of basic breccias occurs a flora referred to the Upper Miocene period. Tt has been named the Lamar flora, and correlated with the Auriferous Gravels of California, with which it is closely allied. The following table may serve to bring out more clearly the relationships between the different geological formations and the floras which characterize them : 3 H. W. Foote—Oceurrence of Pollucite, ete. 457 Formation. | Flora. Age. Basic breccia. Lamar. Upper Miocene. Intermediate breccias. Intermediate flora. Lower Miocene. Acid breccia. Fort Union. Kocene. Agglomerates, Waterlain Livingston. Cretaceous. igneous material. | Sandstone. Laramie. Cretaceous. In a thin bed of conglomerate exposed by the cutting of the grand canyon of the Yellowstone, vertebrate remains of a fossil horse were discovered, sufficient to determine the Pliocene age of the deposit. Upon this conglomerate rests the great body of rhyolite and the still later basalt. Soon after the dying out of these recent basalts climatic conditions changed and the Yellowstone Park was covered by glacial ice which distinctly marked the coming in of Pleistocene time. ART. LI.—QOn the Occurrence of Pollucite, Mangano-Colum- bite and Microlite at Reumford, Maine ; by H. W. Foor. DuRInG the spring of 1885, Mr. E. M. Bailey of Andover, Maine, sent to Prof. S. L. Penfield for identification some ‘specimens from black Mountain, Rumford, Me. Among these, one which had the appearance of ordinary white quartz or beryl proved to be the very rare and interesting mineral pollucite. The following summer, the locality was visited by Professor Penfield accompanied by Mr. Bailey and a supply of pollucite and its associated minerals was obtained. The minerals occur in a ledge of coarse pegmatite which has been worked without success for gem tourmalines, but the locality is of unusual mineralogical interest and the work has been continued to some extent for specimens. In addition to quartz, feldspar and muscovite, which are crystallized on a large seale, there were found pink, green and white tourmalines in imbedded crystals, pink lepidolite, both granular and in erys- tals, with hexagonal outline, spodumene occasionally in ecrys- tals but usually only in cleavage masses, amblygonite rather abundant and in cleavage masses only, beryl not very common, eassiterite rather abundant as irregular masses and rarely in small but distinct crystals, black columbite, pollucite and, very rarely, mangano-columbite and microlite. Pollucite. Pollucite has previously been found in the United States at Hebron, Me., where it was discovered ky Mr. Loren B. Mer- 458 H. W. Foote—Occurrence of Pollucite, rill while mining for gem tourmalines. A full description of this occurrence together with an analysis and a discussion of the chemical composition of the mineral has been given by Prof. H. L. Wells.* The material at Hebron was ali found as loose pieces in two cavities, while at Rumford, which is about thirty miles north of Hebron, it is found intimately associated with quartz, albite, muscovite, tourmaline, lepidolite and spodumene. The specimens are not very attractive in appearance as neither the pollucite nor the associated minerals, with the exception of tourmaline, occur well crystallized. The irregular masses of pollucite are sometimes quite large, so that, for example, for a distance of ten centimeters there will be continuous pollucite. Small particles of the mineral are colorless and perfectly transparent, but the masses look as if they had been crushed and the appearance is therefore white. The quartz at the locality is mostly smoky so that it is readily told from the pollucite .at a glance, but some of it is quite white, and then it is almost impossible without physical or chemical tests to distinguish the two minerals apart. A great deal of credit is due to Mr. Bailey for having ob- served that the pollucite was something different from quartz, which needed investigation. Although occurring in masses of considerable size, the pollucite is not very abundant, but it is hoped, now that the mineral has been identified, that it will be more earefully looked for and saved. It is possible and even quite probable that pollucite is not at all a rare mineral at the tourmaline and lepidolite localities in Maine, but has been overlooked as it resembles quartz so closely and does not occur in character- istic crystals. Material for a chemical analysis was separated in a very pure condition by means of the potassium mercuric iodide solution. That which was used for the analysis ranged in specific gravity between 3-029 and 2°938, a difference of 0-091. Wells gives 2°986 and 2-976 for the mineral from Hebron. The method of analysis was as follows: The substance was digested with strong hydrochloric acid for several hours until completely decomposed. The silica and alumnia were then determined in the usual manner. The filtrate from the alumnia precipitation was evaporated to dryness and ammonium salts driven off with extreme caution. The chlorides of the alkalies were dissolved in very little hydrochloric acid and enough lead chloride added to combine with all the cesium to form the salt Cs,PbCl,.+ Chlorine was then passed into the hot solution which was * This Jour, (il, xl. 213,189 16 + Ibid., xlvi, p. 186, 1893. Mangano-Columbite and Microlite in Maine. 459 gradually cooled, finally with ice. This operation removed nearly all the cesium, which was filtered on a Gooch crucible, washed with hydrochloric acid saturated with chlorine, dried at 100° and weighed as Cs,PbCl,. The filtrate was evapo- rated to dryness, the residue taken up in water, the lead re- moved with hydrogen sulphide and the potassium and re- maining cesium precipitated and weighed as_ platinum chlorides. The platinum was then determined, from which the potassium and cesium were calculated. Lithium and sodium were freed from platinum and separated with amyl aleohol. Only a trace of rubidium could be detected by the spectroscope. Water was determined by strong ignition of the substance over a powerful blast lamp. Following are the results of the analyses, together with the analysis of the mineral from Hebron, by Wells: Ie Il. Average. Ratio. Hebron. SiO, 43°75 43°54 43°64 oN eatk 9°00 43°51 IOS 167-7 16°90 16°84 °165 2°03 16°30 Cs,O 36°25 36°03 36°14 128 ) 36°10 K,O 0°33 0°42 037 004 | 168 OG 48 Na,O 2°06 2°11 2°09 035 r 1°68 Li,O 0°03 0°18 0°08 003 | 05 H,O 1°57 1°59 1°58 "082 1°01 1°50 CaO—0:22 100°76 100°72 100°74 99°84 The results of the analyses are remarkably close to those ob- tained by Professor Wells and the ratio which is nearly 9:2:2:1 gives the formula H,Os,Al,(SiO,), as deduced by him. In a recent article, ‘On the Constitution of the Silicates,”’* Clarke has discussed the constitution of pollucite and con- sidered two formule, one, H,Cs,A1,Si,O,,, deduced from the analysis of the mineral from Elba by Rammelsberg, the other that of Wells, The latter had shown, however, very con- celusively that his formula was correct and that it agreed better with the results of the earlier analysis than did that of. Rammelsberg, and this new analysis fully substantiates his view. Many thanks are due Professor Wells for his many helpful suggestions as to methods of analysis. * Bull U. S. Geolog. Survey, No. 125, p. 31, 1895. 460 HH. W. Foote—Occurrence of Pollucite, Mangano-Columbite. This mineral is found in imbedded crystals which are sel- dom over 10™™ long. The color is a dark reddish brown, very closely resembling that of rutile. The available material was not sufficient for a quantitative analysis, but a qualitative examination showed the the presence of manganese, tantalum and niobium. A specific gravity determination on 0°32498 of material taken very carefully on a chemical balance gave 6°44, which would indicate a chemical composition about mid- way between a niobate and tantalate and near that of the man- gano-columbite from Branchville, analyzed by Comstock.* iB 2. The crystals show a considerable variation in habit, being usually prismatic in the direction of the ¢ axis, fig. 1, some- times lengthened parallel to a, fig. 2, while occasionally they are about equally« developed in all directions. They show a very good cleavage parallel to 6, 010. The forms which were observed are similar to those found on columbite and are as follows: a, 100 m, 110 J5 02 Uu, 133 b, 010 g, 130 e, 021 m, 121 d, 730 k, 108 o, 111 n, 168 They gave afew excellent reflections which served to es- tablish the lengths of the crystallographic axes. The values are given below, together with the axes for columbite as de- termined by E. 8. Dana.t+ Mangano-Columbite. Columbite. m A177, 110.110 = 79 0G Too ae Pry?) 02102 5540 56 28 G2 Oc = 078350 a0; Sens 0°8285 :1: 0°8898 The variation in the ratios is undoubtedly due to a change of both acid and base, the measurement by Dana having been * This Journal, xix p. 131, 1880. - + Zeitschr. Kryst. xii, p. 266, 1886. Pp , Mangano-Columbite and Microlite in Maine. 461 made on columbite from Standish, Me., having a specific grav- ity of 5°65, and which according to the analysis of Allen* con- tains only a little tantalum and manganese. Several forms gave rather unsatisfactory reflections. Some of the measured and calculated angles are as follows : Calculated for Calculated. Measured. Columbite. ew 5 202, Ay O02 bd 4.07 omen KO DEV lee re NOS) Gl Odes 38 47 38 52 39 23 CE SO OO = 39 25: 39 56 39 6 me WO A LO =) 7947 TON AS AG) Wey Gg 180 A130! = 136, 31 136 51 136 10 GA. SOA T30'=) 43) 29 Lv) 43 50 Oune. OZ Oho Ons a LB BO MAO, LLOATIT = 3605 B35 5 25. gS Ceo IW ==) eu 44 77 29 ONO LIU ATI = 162304 62 274 TATE 5 WON TE = Bs © SG 8 55 30 pag 121) AO 1 3) LO 82 100 59 pei 63.168 = 19 42 19 54 nei NGOS N163 = 118 10 118 20 Mee 183 A 133i) 79) 32 One variation from columbite shown by the habit of these erystals is that the form 7, which is the predominating one at the termination of all of the crystals, is of rare occurrence and has previously been observed only with slight development on columbite. Microlite. Very beautiful crystals of microlite averaging 2™™ in diam- eter of a honey-yellow color and high luster are found spar- ingly at the locality. The prevailing form is the octahedron, modified by the dodecahedron and sometimes by the icositetra- hedron, 118. The habit is then very much like that of the pyrochlore figured on page 762 of the sixth edition of Dana’s Mineralogy. ‘The specific gravity, taken by Professor Penfield on 0°2642 gr. of material was found to be 5°17, which is some- what low for microlite and due probably to the presence of a little more niobium than usual. A qualitative analysis indi- eated the presence of calcium and of tantalum in considerable excess over niobium. In closing, I wish to express my thanks to Professor Penfield for his kind assistance during the entire investigation and to Mr. Bailey for a generous supply of the rare and interesting minerals from this locality. Laboratory of Mineralogy and Petrography, Sheffield Scientific School, New Haven, November, 1895. * Zeitschr. Kryst., xii, p. 272, 1886. 462 A. J. Moses—Drawing of Crystal Forms. Art. LIL—A Device for Simplifying the Drawing of Crystal Forms; by A. J. Mosss. In a former article* I described a graphic method for obtain- ing any axial cross from any projection of the isometric axes by use of a quadrant and scale; any axial length, sine or cosine being measured on the scale and quadrant and laid off on the vertical axis and the proportionate length obtained on any other line through the center by connecting the ends and drawing a line parallel to the connecting line from the point on the vertical axis. Necessarily the vertical axis was either the length of the radius of the quadrant or proportionate dividers were used. The method is made still simpler by laying off all measure- ments upon a “scale line” drawn at will from the centre of the axial cross and of a length equal to the radius of the quadrant and by using a metal quadrant shown in fig. 1, the center of Na, Se which is at B. The edge AB and the are AC are tapered to a thin edge for greater exactness in marking. With AB ten centimeters long the results will be correct within the limits of a drawing. Axial lengths are transferred directly from AB to the scale line approximately to the third decimal. Sines and cosines are transferred as follows. If the edge BC and the scale line are made coincident and then, by use of a triangle, the quadrant is slid along in a direction at right angles thereto (BC remain- ing parallel to the scale line) until the scale line cuts the are at the given degree and approximate minute, the intercept on the scale line will be the sine. Similarly with the edge AB coin- cident with the scale line and a motion at right angles thereto, the intercept on the scale line will be the cosine. * School of Mines Quarterly, xv, 214-218. Hutchins and Robinson— Concerning Crookes lubes. 468 All measurements are transferred to other lines, as above described, by lines parallel to the direction between the ends of the scale line and the unit line. The device can also be used in the measurements needed in obtaining edge directions. Fig. 2 represents the construction of the axial cross of a monoclinic crystal in which a:b: c¢=1:092;1:0°589 and B=74° 10’. XX YY ZZ is the isometric axial cross; OS isthe scale line. Of is sine 74° 10’ when radius is OS and trans- ferred to OY is On; Od is cosine 74° 10’ when radius is OS and transferred to OZ is Or; by completing the parallelogram ¢ and O¢result. For the axial lengths, Oz is 1:092 times OS and Oe is 0°589 times OS and transferred are respectively Oa=1-092 x Oz and Oc=0°589 x OZ. Munich, March 16, 1896. Arr. LIII.—Concerning Crookes Tubes ; by C. C. HutcHins and F. C. ROBINSON. WE would offer the following contribution to the rapidly increasing literature on the X-rays of Rontgen. It has to do with a part of the subject upon which very little has been written, and for that reason may be helpful to other experi- menters. One of the chief difficulties in the way of experimenting has been the cost of the bulbs or tubes. We have proved to our own satisfaction that the making of them need not be beyond the resources of the ordinary laboratory ; for within a few weeks time we have made and tested more than one hun- dred tubes, and have frequently made one and exhausted it and used it, all within an hour’s time. All that is required is some little skill in. glass-blowing and in the manipulation of the pump. The glass—A hard German glass, or its equivalent, free from lead, has proved the best. It gives a strong green fluo- rescence under the action of the current, and what is of great importance, resists without softening the heat generated by the cathode ray at its point of impact. Unfortunately it is not to be had free from bubbles, and these are the cause of the destruc- tion of many tubes; the glass being chipped away into the bubble by the action of the current and the tube ruined. It is also rather difficult to put in the electrodes so that they will stay, and it may be necessary to use three kinds of glass,—first 464 Hutchins and Robinson—Concerning Crookes Tubes. the tube itself, then a bit of softer glass, and upon that very soft lead glass for the seal. Shape of the tube.—A good tube should throw shadows as sharp as possible and develop the rays as powerfully as possible. It should easily appear that the ordinary spherical form meets neither of these conditions. To produce a sharp shadow the radiant must be small. It was found that a picture could be taken upon any side of a spherical bulb; making it probable at least that the entire sur- face is a source of radiation. In the matter of strong action also, the spherical form is inferior. This is for two reasons :—first, glass more or less extinguishes the rays according to its thickness, therefore, the larger the bulb the more opaque it must be, for it must be thick enough to stand the atmospheric pressure. Secondly: there is a comparatively large amount of radiant or conducting matter within the spherical bulb which diffuses the energy of the discharge. Proof of the second point was obtained as follows. A moder- ately thick bulb about three inches in diameter was blown, and upon this a spot one inch across was blown out very thin, forming a smaller hemispherical bulb upon the first. Opposite this thin window was the concave cathode. This bulb proved better than the ordinary sort, but far inferior to tubes about to be described. A second experiment was made with a tube blown thin along one side for a space of three inches, and opposite this was the cathode in the form of a quarter cylinder. The performange of this was also inferior. Without going into the details of many similar experiments, it will be sufficient to say that we have found that a simple straight tube from one-half to one inch in diameter, having a small and very thin bulb for a cathode window, has given the most satisfactory results. In length it may be from four to eight inches. The bulb may be blown at the bottom of the tube, the cathode placed at the top, and the anode across the tube just above the bulb. Better results are, however, produced by using a bit of platinum foil for an anode, inclining it about forty-five degrees to the cathode ray. In this case the small bulb may simply be blown out upon the side of the tube and the electrodes put in at the two ends, so that the cathode ray will be reflected into the bulb. Shape and disposition of the electrodes.—We have made the cathode in the form of a wire, a flat plate, a convex plate and a concave plate. The concave form proves the best in every case. We have made it of varying size, up to an inch or more in diameter, and have not come to any conclusion as to which Hutchins and Robinson—Concerning Crookes Tubes. 465 is best. It is very difficult to have other conditions sufficiently uniform to enable one to judge, where differences are small. We have made the anode in the form ofa wire of aluminum, a flattened strip of it, and, as stated above, in the form of a platinum reflector. As yet we have got our best results from the platinum. One rather interesting result obtained was, that when the anode was in the form of an aluminum disc parallel to the cathode and nearly large enough to close the ~ tube, it gave little or no interference with the X-ray. We made one on a hinge so that it could be swung out of the path of the ray or in at pleasure, and the effect on the photographic plate was the same in either position. Source of the rays.—Being able to construct tubes of any form, we have made many experiments as to the source of the rays, whether from the cathode or anode. One was in this way :—Two tubes were joined together parallel so that they were exhausted together. The cathode rays could be made to pass down one tube and the anode rays (if such existed) down the other, and either screened off at will. We found that the anode rays affected the plate but slightly, and that practically all the effect came from the cathode. Intensity of effect—We do not intend to convey the impres- sion that these home-made tubes we have described are simply good enough for experiment and valuable from their cheap- ness. We believe also that they are more effective than others. We have made good negatives of bones of the hand, arm, including the elbow, foot, ankle, ete., all with remarkably short exposures; have taken impressions perfectly distinct through nine inches of wood in less than five minutes; have taken perfectly the bones of the hand through thin sheet zine in two minutes and through the slide of the plate holder in five seconds. The ordinary coin and key impression require not over one or two seconds with our best tubes. fiemarks upon pumping.—The interest in the subject at present may make some remarks upon pumping here in place, most of all, since many have found great difficulty in this respect. It is here supposed that the pump has a three-way cock above its bulb, opening in its two positions between the bulb and fork and the bulb and outer air; and that above this three- way cock are one or two cocks of the ordinary kind. Let the three-way cock be called A, the others B and C in order. Let the position in which A puts the bulb in communication with the fork be position 1; and that in which it puts the bulb in connection with B, G, and the outer air, position 2. The ordinary process of pumping with the use of A alone is supposed to be understood. After a greater or less number of 466 Hutchins and Robinson—Concerning Crookes Tubes. strokes it is observed that no more air is obtained. The pump contains air, however, condensed upon the glass walls. To remove this A is put in position 2, and the mercury raised until a drop passes B. JB is then shut and the mercury dropped until only a drop remains above A.’ A is then shut and the movable mercury tank dropped to its lowest point, when A is put in position 1. Pumping now goes on as before only with B shut, and the tank is raised only a third as high as before. After four or five strokes it is well to pass the mercury again above B. : If the highest possible degree of exhaustion is desired this process can be repeated between B and C, but this is not neces- sary in exhausting a Crookes tube. As soon as the stage of pumping with B shut is reached, the tube which is being exhausted must be strongly heated, mov- ing the lamp flame over every part of it, and after two or three strokes more the current from the coil is turned into the tube. By the combined action of the heat and current the occluded air is driven from the glass and exhaustion proceeds rapidly. It should not occupy over twenty or thirty minutes for a mod- erate sized tube. Allowing the tube to cool, if short sparks can be drawn from the bulb and there is little or nothing to be seen in it except green light, the exhaustion is complete. There is danger of carrying it too far, for the vacuum very much increases during the first hour that the tube is used; but of these matters a little experience is the best teacher. Searles Science Laboratory, Bowdoin College, May 6, 1896. A. M. Mayer—fesearches on the Rontgen Rays. 467 Art. LIV.—Rescarches on the Rontgen Rays; by ALFRED M. Mayer. ContTENTS: 7. The Réntgen rays cannot be polarized by passing through herapathite. 2. The density of herapathite. 3. The formule of transmis- ston of the Réntgen rays through crown glass, aluminum, platinum, green tourmaline and herapathite. 4. The actinic action of the Réntgen rays varies imversely as the square of the distance of the sensitive plate from the source of the rays. 1. The Réntgen rays cannot be polarized by passing through herapathite. __HERAPATHITE is an iodo-sulphate of quinine, discovered by W. B. Herapath in 1852,* and named herapathite by Haidinger. Herapath gives several formula for its production. The one which succeeded the best with me in giving crystal plates of large area is contained in Phil. Mag., Nov. 1853. It is as follows : Bisulphate of quinine..-...........-..-. 3°24 grams. evrolomeous acid 222. 2.2 42-24-44 -- 56° PMcOMOl (95 per cent). ee e2 te = 56" Solution of iodine (1 grm. in 11° alcohol) 50 drops. The bisulphate of quinine is added to the mixture of pyro- ligneous acid and alcohol and heated to 55° C., and then the iodine is added drop by drop while constantly stirring the solu- tion. The vessel containing the solution is then placed on several layers of thick felt resting on a firm support to prevent vibration, and it thus remains about 18 hours at a temperature of 8° C. At the expiration of that time large crystals will generally be seen floating on the surface of the liquid, the majority being at the bottom of the vessel. The solution gives, of herapathite, only 7 of the weight of bisulphate of quinine. As suggested by Herapath, a microscope cover glass is cemented to the end of a glass rod, with its plane at right angles to the rod, and is carefully brought under the floating erystal, very slowly brought up to it and the crystal is thus se- cured on the surface of the glass. The mother liquid is then absorbed from the glass by blotting paper. If several crystals are desired on a cover glass, a glass tube closed by the finger is brought over the crystals at the bottom of the vessel, the finger removed, then replaced and the tube taken out of the liquid. The crystals are allowed to sink into the drop at the end of the tube by holding the latter for some time in a vertical position ; * Phil. Mag., Mar. 1852 468 A. M. Mayer— Researches on the Réntgen Rays. the drop is then brought in contact with a cover glass. This is the manner in which I placed on the glass dises and blotting paper the plates of herapathite used in my experiments. Roéntgen has shown that generally the lower the density and always the thinner the substance the less it screens from a pho- tographic plate the action of the X-rays. ~ Herapathite is, therefore, eminently fitted as the substance on which to make experiments which are to decide whether the X-rays can or cannot be polarized by having traversed crystals which, like herapathite and tourmaline, transmit only one polarized beam, the: other being absorbed by the crystal, for the density of herapathite is 1-557 and plates of only -012™™ thick are sufficient to answer the question. When crystals of herapathite -012™™ thick have their optic axes crossed at 90° these crossed portions viewed against incident light appear black, so powerful is the polarizing property of this substance. If the X-rays be polar- izable these black portions should act like thick lead and com- pletely screen the sensitive film from the action of the X-rays. The fact is that the crossed herapathites do not screen the X- rays any more than the herapathites do when superposed plates have their axes parallel. In the latter case the crystals freely transmit light with a faint olive-green tint. The thickness of crystal plates used in the experiments varied from -01™™ to -025™™, as found by focussing with a micrometer-screw a powerful objective on the top surface of the crystals and on the glass on which they rested. Six discs of glass, -15™" thick and -25™™ in diameter, were covered with herapathites in the manner described. They crossed one another at various angles; where they crossed at 90° the crossed portions were black. Ona piece of yellow blotting paper, #™™ thick, were also placed several layers of herapathites, so deep that they reflected a green metallic luster like the elytra of cantharides. These dises and the blotting paper were fastened to the slide covering the photographic plate. This slide was impervious to two hours’ exposure to the actinic action of the sun’s light. On the slide were also three discs of thin glass, so overlapping that the X-rays had to pass through 1, 2, 3 thicknesses of the glass before reaching the sen- sitive plate. These served as standards with which to compare the screening effects of the herapathites. The slide so prepared and covering a sensitive plate was exposed to the radiations of a Crookes tube in three experi- ments, for 4, 1, and 24 hours. On developing the plates, not the slightest trace of the presence of the herapathites was visi- ble. The photographs of the glass discs had not the slightest mottling on their surfaces; appearing to the unaided eye and when examined through a magnifying glass with uniform illu- A. M. Mayer— Researches on the Réntgen Rays. 469 mination and grain throughout, and exactly like the photo- graph obtained by the X-rays passing through a similar glass dise with nothing on its surface. The herapathites used in the experiments were so thin that they did not appreciably screen the X-rays, whether the axes of the superposed crystals were arallel or crossed. But the action of the rays on the square of blotting paper proved even more conclusively that the X-rays cannot be polarized in this manner, for where this paper covered the photographic plate nothing was visible, except by careful scrutiny and with a favorable illumination, and then a mere ghost of the paper was detected, but with no traces what- ever of the herapathites. These experiments confirm in a convincing manner what Rontgen found by his experiments, viz: that the X-rays can- not be polarized. At least, they cannot be polarized by pass- ing through herapathite, which is by far the most powerful polarizing substance known. It is unlikely that polarization will be detected by using any doubly refracting substance which transmits two beams polarized in planes at right angles to each other; for if polarization exist, the thickness of the substance required to get a measurable departure of the two rays so screens the X-rays that a very small fraction of them (at least by calcite) will be transmitted ; also, Rontgen is of the opinion that if the X-rays be refrangi- ble, the index of refraction is nearly unity even in such a highly refracting substance as ebonite, which has an index of about 1°6.* It is therefore reasonable to suppose that the difference in the refraction of the ordinary and extraordi- nary beams will be too small to be measurable in the faint shadowgraphs obtained by traversing doubly refracting sub- stances. It remains to be decided whether the X-rays can or cannot be polarized by reflection, Professor Rood having recently proved conclusively that they are reflected from platinum. 2. The density of herapathite. Herapath gives 1:89, at 60° F., as the density of the very remarkable substance he discovered.t As its density is interesting to have in connection with the experiments described above, I made two determinations of it by weigh- ing about °3 ers. of the substance in water, and also by the displacement it gave of the water in a specific gravity flask. The mean of these two measures was 1°6. As this : rs the Physical Properties of Vulcanite, by A. M. Mayer, this Journal, Jan. 1891. + Phil. Mag., May, 1855, p. 369. Am. Jour. So1.—Fourts Srrizs, Vou. I, No. 6.—Junu, 1896. 3! 470 A. M. Mayer—fesearches on the Réntgen Rays. number differs greatly from that of Herapath, I thought that the small mass I had used was the cause of the discrepancy. I then made about two grams of herapathite, freed it of its mother liquid by washing with water at 0°, and dried it at 25° in vacuo. When it had ceased to lose weight I weighed it in a specific gravity flask holding about 10 ¢.c., then just covered it with water, placed it in vacuo and agitated it, so that it would be freed of any air that might be contained in its mass. The flask was then nearly filled with water and again placed in vacuo, then entirely filled and weighed. I found from this carefully made experiment that the density of herapathite is 1557 at 20° C. 3. The formule of transmission of the Réntgen rays through glass, aluminum, platinum, green tourmaline and herapathite. The Rontgen rays, after their transmission through various substances, produce actinism on a photographic plate, and by the degree of this actinism we have attempted to obtain the formula of transmission peculiar to each substance. This action of the X-rays is cumulative and evidently is entirely different from the transmission of light and radiant heat where the maximum of transmission is instantly reached and under proper and controllable conditions remains constant and there- fore can be accurately measured. But with the X-rays the amount of their action on the plate varies directly as the time of their action (see 4), depends on the distance of the sensitive plate from source of radiation, and on the energy of the source. Therefore this« method of determining the constants of the formule of transmission of the X-rays is somewhat arbitrary ; but the exact conditions in the determination of the constants having been given, these conditions can readily be obtained by other experimenters. Thus, to have the same conditions as existed in our experiments, one has only to place a pile of crown giass plates 5°5™™ thick, at such a distance from the radiant source that on two hours exposure the X-rays will have just not visibly acted on the photographic plate. Some assump- tion has to be made as to the nature of the X-rays, otherwise no progress can be made in determining the constants of their formule of transmission. We have assumed that they are homogeneous. The formule, as determined, hold good till they have conclusively been shown not to be homogeneous, which is very likely to happen in the progress of research on their nature. The method used owes whatever merit it may have to the use of the wire netting devised by Professor Rood to give accurate indications of the relative permeability of various A. M. Mayer—feesearches on the Réntgen Rays. 471 substances to the X-rays. This wire netting he places on the slide covering the sensitive plate, and on the netting he places substances of various thickness. The netting completely screens the X-rays, and its image on the negative is the brightest possible to obtain in the given conditions. If a plate of a sub- stance should also entirely screen the X-rays, then the image of the netting is invisible in the photograph of this substance, and as plates of substances allow more and more of the X-rays to traverse them the images of the netting in the photographs of these substances are more and more bright. If the image of the netting in the photographs of two substances should be equally bright, then these two plates transmit equal actinic action to the sensitive plate. Thus this ingenious device serves as a very delicate photometer in determining the fact just men- tioned. It occurred to me that the wire netting could also give me data with which to determine the constants of the formule of transmission of the X-rays through various substances. The method devised is as follows: On a wire netting with 8 meshes to the inch placed on the slide of the plate-holder, are cemented piles of glass discs, (each glass about ;;"™ thick); these piles gradually increase in thickness. These piles of glass were exposed to the action of the X-rays, so that the photographic plate was 25°" distant from the radiant source. In the appara- tus used this distance could be accurately measured. After an exposure of two hours it was found, on developing the plate, that the netting was just visible in the photograph of the pile 5°35™™5 thick, and that it was not visible in the photograph of the pile 5°5™™ thick. In this last case, though the X-rays had pene- trated to the sensitive film, yet the difference in the screening effect of the netting and of the glass was not visible, because the eye cannot. distinguish between the illumination of jux- taposed surfaces which differ in illumination by about ;4,. Henee, through the last pile of 5°5™™ about ;1, of the actinic intensity of the incident beam had passed. Now the formula of transmission of rays through a substance which does not reflect these rays is I’=Ia*. Where I’ = the intensity of the transmitted beam; I= the intensity of the incident beam; @= a constant depending on the substance, and ¢ is an exponent of a, and 7 is the thickness of the sub- stance. We have taken ~,™™ as the unit of ¢. From the conditions of the experiment I’ = Ia*=,1,, and as t is known, @ is readily computed. The accuracy of the determination of @ depends on the value of the least percep- tible difference in illumination that the eye can distinguish in two juxtaposed surfaces. We have adopted =A, as the most probable value.* But suppose that the fraction is not =3, but * Photometric Experiments, O. N. Rood, this Journal, July 1870. 472 A. M. Mayer— Researches on the Réntgen Rays. zis, OY, sz, then the departure of these fractions from 2; will affect the constant by 4 units in the third decimal place. These experiments, made in the conditions we have indi- cated, give tor the formula of transmission through the glass used, a crown glass made by Chance & Co.: T=I~x :92¢ Having this formula as a basis, it is comparatively easy to determine the constants of another substance, by placing the substance on the netting with piles of glass of graded thickness and exposing them to the X-rays. We then see what thick- ness of glass gives the same illumination to the image of the netting as does the known thickness of the substance. Thus, the netting in the photograph of a disc of herapathite -9™™ thick had the same brightness as in the photograph of a pile of glass 69" thick. By the formula glass -69™™ thick transmits ‘5636 of the incident beam and herapathite of -9™™ transmits the same. From this we compute that I’ = 1 xX -9382* is the formula for the transmission of the X-rays through herapathite, which for ¢ = 69 gives ‘5636 of transmission. In the same manner it was found that 2°05™" of aluminum transmits the same as 2°44 of glass 02 =“ platinum “ ce Seg 2°0 «¢ green tourmaline “ S012 S00 tae ‘ggnm (44 herapathite c¢ ce °9 CE 36E From these determinations we have computed the formule of transmission of these substances : Glas€io oo oe ee eee PAS ria aia eee oe es l = besos: Platina S25) ek eee eee Il’ = 1> :00063' Green tourmalines 9242.32 l= f97; EHerapathites: 2s) aia eee T a1 338. Taking the amount of transmission through aluminum of zjmm and of 1™™ as unity, we have, in the following table, the . relative transmission through the substances experimented on. ae {= 1", mamas eee ee ile ice (Hla scy ae nene see ee 1'016 1°180 _Green tourmaline .. 1:016 1°180 Herapathite - _--_-- 1:036 1°435 - Platimnmess: ae _ °000696 Platinum -06™ thick is practically impervious to the rays from the Crookes tube used, transmitting only ‘005 of incident beam. This method of determining the transmission of the X-rays through various substances may be criticised, because in the experiments I obtain not alone the transmission through the Aa MM. Mayer— Researches on the Leéntgen Rays. 473 substances placed on the slide of the plate-holder but at the same time the transmission through the slide itself. If the slide has a measurable transmission, or, screening effect, then its equivalent in thickness of the substances placed on it should be added to them. This criticism, however, is nought, because experiments have shown that if the thickness of a portion of the slide is doubled by placing on it a piece of the material of which it is made, no screening of the X-rays by this piece can be detected. 4. The actinic effect of the Réntgen rays varies inversely as the square of the distance of the sensitive plate from the radiant source. The slide covering a photographic plate had on it the wire netting, and the plate was exposed for 30 minutes to the X- rays at a distance of 10 inches from the radiant source. Another similar plate in the same plate-holder was exposed for two hours at a distance of 20 inches from the radiant source. These distances could be accurately measured in the apparatus used. These plates were taken from the same box and developed side by side in the same tray. Indeed all the conditions were care- fully made the same in the two experiments except the dis- tance of plates from the radiant source. On developing the images on the plates they were exactly alike; the image of the netting had the same illumination on each plate, and the density of the films was the same. This experiment shows that the X-rays act on a sensitive plate according to the law of the inverse squares. From this law it follows that the actinic power of the X-rays is not sensibly absorbed in traversing the air; also, that these rays are not sensibly diffused by radiation from the molecules of the air they traverse, the air being at ordinary barometric pressures. These deductions, which necessarily follow from the law of the inverse square, are, however, at variance with the facts observed by Professor Pupin, who states in Science of April 10, 1896: “There was evidently a diffuse scattering of the perayceintheir passace) through the air’)... °. 2. “These experiments prove beyond all reasonable doubt that the Rontgen radiance is diffusely scattered through bodies, gases not excepted.” These opinions of Professor Pupin are founded on experiments on the action of the X-rays on a fluo- rescent screen, or, rather on a “fluoroscope,”’ not on experi- ments on their actinic effects. In the latter case I am confi- dent that the X-rays act according to the law of the inverse squares, and therefore are not sensibly diffused. In the former case Professor Pupin finds that they are diffused in traversing the air, and very sensibly diffused if I understand aright his 474. A. M. Mayer—Researches on the Réntgen Rays. paper; but as he does not give any photometric measures, or estimates, of the intensity of the fluorescence in the geometric image of the slit, and outside of this image, one cannot form an opinion of the amount of the diffusion he describes. The facts. of diffusion described by Professor Pupin are opposed to those discovered by so distinguished and experienced a physicist as. Rontgen, who, in section 10 of his first paper, writes: “I find, using a Weber’s photometer, that the intensity of the fluorescent light varies nearly as the inverse squares of the distance between screen and discharge tube. This result is obtained from three very consistent sets of observations at distances of 100 and- 200". Hence air absorbs the X-rays much less than the cathode rays. This result isin complete agreement with the previously described result, that the fluorescence of the screen can be observed at 2 metres distance from the vacuum tube.’” Rontgen does not mention any diffusion observed by him in these experiments, and it is certain that he would have men- tioned the existence of diffusion in experiments made to deter- mine a law of radiation, and which diffusion necessarily would have invalidated the law of inverse squares. The radiation from the Crookes tube, used in the experi- ments described in this paper, came from a calcined shell sup- ported by the anode wire and placed opposite the cathode. The tube was not sensibly heated during the experiments and the actinic power of its radiation remained constant. It was proved by taking a pin-hole photograph of the naked tube, with an exposure of two hours, that by far the larger propor- tion of the X-rays emanated from a 4™™ square surface of the shell which was acted on by the cathode rays. The walls of the glass tube, as was proved by other independent photo- graphic experiments, furnished a small percentage of X-rays but not enough to make their presence known in the pin-hole photograph, which had furnished quite a dense image of a por- tion of the shell. Furthermore the X-rays from the glass wall itself were cut off and prevented from reaching the photo- graphic plate by a diaphragm of lead with a circular opening of one-half inch, and at one inch distant from the radiant source. It is also to be remarked that this second source of the X-rays was nearer to the photographic plate only by 74 inch minus ‘06 inch, the thickness of the tube, hence any effects due to them may be neglected, as indeed the results in the experi- ments on the Inverse squares show. The experiments described in this paper were made in the private laboratory of Professor Rood in Columbia University. Professor Rood not only gave me the use of his apparatus, which he had made the subject of a special investigation before investigating with it, but he also gave me the advantage of the experience obtained during his researches on the X-rays. Marsh—On the Pithecanthropus erectus. 475 ArT. LV.—On the PITHECANTHROPUS ERECTUS, from the Tertiary of Java; by O. C. Marsu.* (With Plate XIII.) NEAR the beginning of last year, a discovery was announced that excited great interest throughout the scientific world, espe- cially among those interested in the origin and antiquity of man. The announcement first made was that remains of a veritable missing link between man and the higher apes had been found in Java, in strata of Pleistocene age. The dis- covery was made by Dr. Eugene Dubois, a surgeon in the Dutch army, who had been stationed in Java for several years, and had devoted much time to the vertebrate fossils of that island. The first definite information received in this country was in December, 1894, when Dubois’s memoir on Prthecanthropus arrived.t One of the first copies reached the late Professor Dana just as he was printing the last pages of his great work on geology. Heat once wrote to me in Washington, asking me to look up the memoir, and telegraph my opinion of the dis- covery, so that he could refer to it in his book. On inquiry, I ascertained that this memoir had not then been received at any of the scientific centers in Washington, and that the discovery itself was not known. On returning to New Haven, I founda copy of the memoir awaiting me (received December 29, 1894), and at Professor Dana’s request, I wrote a review of it, which appeared, with illustrations, in this Journal for February, 1895. The memoir of Dr. Dubois was an admirable one, and, although written in Java, with only limited facilities for con- sulting the literature on the subject and for comparing the remains described with living and extinct forms to which they were related, the author showed himself to be an anatomist of more than usual attainments, and fully qualified to record the important discovery he had made. In my review, therefore, of this important memoir, I endeavored to state fairly the essential facts of the discovery, as well as the main results reached by Dr. Dubois after a careful study of the remains. * Abstract of communication made to the National Academy of Sciences at Washington, April 24, 1896. + Pithecanthropus erectus. Kine menschenaehnliche Uebergangsform aus Java. Von Kug. Dubois, Militairarzt der niederlagndisch-indischen Armee. Mit zwei Tafeln und drei in den Text gedruckten Figuren. 4to, Batavia, 1894. ¢ The figures then given in Plate II are repeated in the plate accompanying the present article. 476 Marsh—On the Pithecanthropus erectus, My own conclusions in regard to this discovery, briefly stated in my review, were as follows :— Pt as only justice to Dr. Dubois and his admirable memoir to say here, that he has proved to science the existence of a new prehistoric anthropoid form, not human indeed, but in size, brain power, and erect posture, much nearer man than any animal hitherto discovered, living or extinct. Whatever light future resear rches may throw upon the affinities of this new form that left its remains in the volcanic deposits of Java during later Tertiary time, there can be no doubt that the discovery itself is an event equal in interest to that of the Neanderthal skull. ‘The man of the Neander valley remained without honor, even in his own country, for more than a quarter of a century, and was still doubted and reviled when his kinsmen, the men of Spy, came to his defense, and a new chapter was added to the early history of the human race. The ape-man of Java comes to light at a more fortunate time, when zeal for explora- tion is so great that the discovery of additional remains may be expected at no distant day. That still other intermediate forms will eventually be brought to light no one familiar with the subject can doubt.” In most scientific quarters, however, both in this .country and in Europe, Dr. Dubois’s discovery was not received with great favor, and the facts and conclusions stated in his memoir were much criticised. Among a score or more of notices of this elaborate memoir which appeared subsequent to my review, I do not recall a single one that, in attempting to weigh the evidence presented, admitted the full importance of the discovery made by Dr. ‘Dubois. The early conclusions seemed to be that’ the various remains discovered were human, and of no great age; that they did not belong to the same individual ; that the skull apparently pertained to an idiot; and that both the skull and femur showed pathological features. -In fact, the old story of the distrust aroused by the discovery of the Neanderthal skull, nearly forty years before, was repeated, although in a milder form. Dr. Dubois has stated in a late memoir that, with the exception of Professor Manouvrier of Paris and myself, no one else, until very recently, regarded the remains as evidence of a transi- tional form between man and the apes. It was a fortunate thing for science that the Dutch govern- ment appreciated the importance of the discovery made in its Javanese province by Dr. Dubois, and last summer allowed him to return to Holland and bring with him the precious remains he had found, and so well described. Not only this, from the Tertiary of Java. ATT but he was also permitted to bring the extensive collections of other vertebrate fossils which he had secured from the same horizon and in the same locality where the Pithecanthropus was discovered. All these were shown at the International Congress of Zoologists, held at Leyden, in September last, and on the 21st of that month, Dr. Dubois read an elaborate paper on his original discovery and on his later explorations in the same region. This communication was in many respects the most important one of the session, and its presentation with the specimens themselves was a rare treat to the large audience present, especially to those fitted to appreciate the evidence laid before them.* Professor Virchow of Berlin was president of the meeting on that day, and had brought many specimens to illustrate the remarks he was to make in the discussion. The famous Leyden museum was also drawn upon for an extensive series of speci- mens of man and the higher apes, so that, if possible, the true position of Prthecanthropus might then be determined once for all. _ Dr. Dubois, moreover, kindly invited Professor Virchow, Sir William Flower, and myself, to come an hour before the meeting, and personally examine the remains he was to discuss, and this invitation was most gladly accepted. The first sight of the fossils was a surprise, as they were evidently much older than appeared from the descriptions. All were dark in color, thoroughly petrified, and the matrix was solid rock, difficult to remove. Theskull-cap of Prthecanthro- pus was filled with the hard matrix, firmly cemented to it. The roughness of the superior surface, especially in the frontal region, was apparently due to corrosion after entombment, and not to disease, as had been suggested by some anatomists. The femur was free from matrix, but very heavy in consequence of the infiltration of mineral matter. The exostosis on its upper portion is a conspicuous feature, but of course is pathological. This feature is of little consequence, as very similar outgrowths occur on fossil bones of even Eocene age. The two teeth showed no characters that indicated their interment under cir- cumstances different from that of the skull or femur. All the physical characters impressed me strongly with the idea that these various remains were of Pliocene age, and not Post- Tertiary, as had been supposed. The description of the local- ity and the account of the series of strata there exposed, as given by Dr. Dubois in his communication, confirmed this opinion, and a later examination of accompanying vertebrate fossils placed the Pliocene age of all beyond reasonable doubt. * Compte-Rendu des Séances du Troisiéme Congrés International de Zoologie, Leyden, September, 1895, pp. 251-271, 1896. See also Transactions Royal Dublin Society, vol. vi, pp. 1-18, February, 1896; and Anatomischer Anzeiger, Bd. xii, pp. 1-22, 1896. 478 Marsh—On the Pithecanthropus erectus, The facts relating to the discovery itself, and the position in which the remains were found, as stated by Dr. Dubois in his paper, together with some additional details given to me per- sonally, convinced me that, im all probability, the various remains attributed to Prthecanthropus pertained to one indi- vidual. Under the circumstances, no paleontologist who has had experience in collecting vertebrate fossils would hesitate to place them together. In figure 1, below, a geological section is given, showing the series of strata exposed in the bank of the river Bengawan, near Trinil, in central Java, where all the remains of P7the- canthropus were found. The exact positions of these various specimens when discovered are also indicated. ; a 7 ee ee 0 > = EB SAGE vig SPS SRO RG 77 G ies eae Ta ‘O° oa “oo ooo °9 a + LEE 93 . —) D> cz . <- Figure 1.—Section of the bone strata at Trinil. (After Dubois.) A, vegetable soil; B, sand-rock ; C, bed of lapilli-rock ; D, level in which the four remains were found ; E, conglomerate ; iy clay-rock ; G, marine breccia; H, rainy season level of river s TL; dry: season level of river. The above section, taken from Dr. Dubois’s Dublin paper, makes clear many points as to the locality where the discoveries were made, which were left doubtful in the original memoir. from the Tertiary of Java. 479 The three specimens originally described, the tooth, the skull, and the femur, were found at different times in the same horizon, all imbedded in the same volcanic tufa, as indicated in figure 1, D. The tooth was found first, in September, 1891, in the left bank of the river, about a meter below the water level during the dry season, and twelve or fifteen meters below the plain in which the river had cut its bed. A month later, the skull was discovered, only a meter distant from the place where the tooth lay. In August, 1892, the femur also was found, about fifteen meters distant from the locality where the other specimens were imbedded. Later, in October of the same year, a second molar was obtained at a distance of not more than three meters from where the skull-cap was found, and in the direction of the place where the femur was dug out. The fossils thus secured were all carefully investigated by Dr. Dubois, who regards them as representing a distinct species and genus, and also a new family, which he has named the Pithecanthropide, and distinguished mainly by the following characters : Brain cavity absolutely larger, and, in proportion to the size of the body, much more eapacious than in the Semudea, yet less so than in the Homznide: Capacity of the skull about two-thirds the average of that of man. Inclination of the nuchal surface of the occiput considerably greater than in the Simide. Dentition, although retrogressive, still of the simian type. Femur equal in its dimensions to that of man, and like that adapted for walking in an upright position. Of this skull, the upper portion alone is preserved, the line of fracture extending from the glabella backward irregularly to the occiput, which it divides somewhat below the upper nuchal line. The cranium seen from above is an elongated oval in outline, dolichocephalic ; and is distinguished from that of other anthropoid apes by its large size and its higher arching in the coronal region, as shown below in figure 3. The greatest length from the glabella to the posterior projection of the Gcciput is) 1s5""_) The greatest breadth is 1307", and the smallest, behind the orbit, is 90"™. The cranium in its original condition must have been of somewhat larger dimensions. The upper surface of the skull is without ridges, and the sutures all appear to be obliterated. This dolicocephalic skull, with an index of 70°, is readily distinguished from that of the Orang-utan, which is decidedly brachycephalic. The absence of the characteristic cranial crests will separate it from the skull of the adult Gorilla. In its smooth upper surface and general form, it shows a resemblance to the skull of the Chimpanzee, and still closer to that of the Gibbons (fylobates). 480 Marsh—On the Pithecanthropus erectus, A figure of the present specimen and the skull of a Gibbon for comparison are shown in figure 2, Plate XIII. These figures and those that follow are reproduced from illustrations in Dr. Dubois’s memoirs. In comparing the cranium of Pthecanthropus with skulls most nearly allied, both human and simian, the outlines given in figure 3, below, will prove especially instructive. The basis of this cut is the figure given by Dr. Dubois in his Leyden paper. This I have modified by omitting the outline of the microcephalic idiot, and substituting that of the well-known Neanderthal skull.* FicurEe 3.—Profile outline of the skull of Pithecanthropus (Pe), compared with those of a Papuan man, the man of Spy No. 1, Neanderthal man (Nt), man of Spy No. 2, and Hylobates leuciscus (Hl), Semnopithecus maurus (Sm), and Anthropopithecus troglodytes (At). (Modified from a figure by Dubois.) Gl, glabella; Op, opisthion; Jn, linea nuchz superior; Lmni, linea nuche inferior. Dr. Dubois’s conception of the skull of Pzthecanthropus, when entire, is indicated by his attempted restoration shown in figure 6, on page 481. Future discoveries must determine the accuracy of this restoration. * In presenting the present paper before the National Academy of Sciences at Washington, I was fortunately able to exhibit a cast of the Pithecanthropus skull, recently sent to me by Dr. Dubois, and also to compare this with a cast of the Neanderthal skull. The latter was not available during the discussion at Leyden. Srom the Tertiary of Java. 481 The tooth, the first specimen found, is represented in figure 4, below. It is the last upper molar of the right side, and is in good preservation. It indicates a fully adult, but not very old, animal. The crown is subtriangular in form, with the corners rounded, and the narrowest portion behind. The antero- posterior diameter of the crown is 11°3™", and the transverse diameter 15°3"™. The grinding surface of the crown is con- cave, and less rugose than in existing anthropoid apes. The diverging roots are a simian feature. 4, Figure 4.—Third right upper molar of Pithecanthropus erectus, Two-thirds natural size. (After Dubois.) a, back view ; b, top view. ~ FIGURE 6.—Restoration of the skull of Pithecanthropus erectus. Two-fifths natural size. (Reduced from a figure by Dubois.) c, sutura coronalis ; L, sutura lamboidea; 0, foramen occipitale. The femur, which is from the left side, is in fair preserva- tion, although it was somewhat injured in removing it from the surrounding rock. It belonged to a fully adult individual. In form and dimensions, it resembles so strongly a human femur that only a careful comparison would distinguish one 482 Marsh—On the Pithecanthropus erectus. from the other. The bone is very. long, its greatest length being 455™". The shaft is slender and nearly straight. The general form and proportions of this femur are shown in figure 5, Plate XIII, with a human femur for comparison. These various remains of Pthecanthropus were again described in detail and compared with allied forms by Dr. Dubois in his paper at Leyden, and in the discussion that fol- lowed, the whole subject was once more gone over by anthro- pologists, zoologists, and geologists, in a most thorough and judicial manner. To attempt to weigh impartially the evidence as to the nature of P7thecanthropus, presented by Dr. Dubois in his paper and by those who took part in the critical discus- sion that followed its reading, would lead far beyond the limits of the present communication. I can only say that this evidence was strongly in favor of the view that the skull. of Pithecanthropus is not human, as the orbital and nuchal regions show, while at the same time it indicates an animal much above any anthropoid ape now known, living or extinct. Opinions differed as to whether the various remains pertained to the same individual, but no one doubted their importance. The varied opinions expressed in regard to the anatomical characters of each of the specimens have already been pub- lished, and need not be repeated here. Dr. Dubois, in his papers above cited, has met all the principal objections made to his views since he announced his discovery. He has also given full references to the literature, which promises to -be voluminous as the importance of the subject becomes better known. Among the authorities thus cited may be mentioned Cunningham, Kéith, Lydekker, Turner; Manouvrier, Pettit, Topinard, Verneau; Haeckel, Krause, Martin, Ten Kate, and Virchow, who have all taken part in the discussion. ‘ After a careful study of all the Prthecanthropus remains and of the evidence presented as to the original discovery, the position in which the remains were found, and the associated fossils, my own conclusions may be briefly stated, as follows: (1) The remains of P2thecanthropus at present known are of Pliocene age, and the associated vertebrate fauna resembles that of the Siwalik Hills of India. (2) The various specimens of Pithecanthropus apparently belonged to one individual. or (3) This individual was not human, but represented a form intermediate between man and the higher apes. If it be true, as some have contended, that the different remains had no connection with each other, this simply proves that Dr. Dubois has made several important discoveries instead of one. All the remains are certainly anthropoid, and if any of them are human, the antiquity of man extends back into the Tertiary, and his affinities with the higher apes become much nearer than has hitherto been supposed. ~ One thing is certain: the discovery of Pithecanthropus is an event of the first importance to the scientific world. Chemistry and Physics. 483 SCIENTIETOC INTELLIGENCE. I. CHEMISTRY AND PHYSICS. 1. The molecular weight of sulphur.—Previous investigations have shown that the sulphur molecule probably corresponds to the formula §,, or to §,, at a temperature slightly above its boiling point, and that as the temperature rises the molecular weight becomes smaller, until between 860 and 1700° it is constant and corresponds to a normal molecule, 8,. Previous work upon the molecular weight of sulphur in solution, as determined by the boiling-point and the freezing-point methods, has in most cases pointed to the formula 8,. Ornporrr and TERRASSE have now described an elaborate series of experiments in which the boiling- point method has been used with various solvents, and they con- clude from their results that at temperatures below its melting- point the molecule of sulphur is 8,, while with solvents whose boiling-points are above the melting-point of sulphur, the mole- cular formula is $,. The interesting fact was observed that when dissolved in sulphur chloride (S,Cl,) the molecule of sulphur corresponds to the normal formula, 8,. The authors have also made the observation that in carbon disulphide, from which only orthorhombic sulphur crystallizes out, and in benzene and toluene, from which the monoclinic form alone separates, the same mole- cule, 8, exists— Am. Chem. Jour., xviii, 173. H. L. W. 2. The absorption of the Rénigen rays by chemical compounds.— Vv Nov4k and Sutc have examined nearly 300 substances in this respect. Their method of investigation consisted in attaching rings of glass to a sheet of paper and placing uniform layers of the finely pulverized materials in the different rings so that the thickness of the layer was 0°4°" ineach case. The paper with the rings was then placed over a photographic plate which was envel- oped in black paper, and exposed to the Réntgen rays for a period of 20 to 25 minutes. By comparing the photographic effect of the rays where the substances were interposed, the relative absorptions were“determined. The authors found that a great number of organic compounds containing only carbon, hydrogen, oxygen and nitrogen, are equally penetrable, and hence they con- clude that the absorption has no relation to molecular weight or the arrangement of the atoms. Organic halogen derivatives were found to possess much greater absorption, which increased with the number of halogen atoms present. ‘This effect increased with the atomic weights of the halogens, two atoms of bromine having a greater effect than six chlorine atoms, while iodine derivatives were entirely impenetrable under the conditions used in the experiments. This indication of the influence of elements of varying atomic weight led the authors to examine a series of ele- mentary substances, all of rather low atomic weights. The absorptive power was as follows: 484 Scientific Intelligence. S=Pe>Al>Mg>B—Co(=Ain) on wo le teow mn Ban 214 This series agrees in a striking manner with the order of the weights of the atoms, and the metallic or non-metallic character of the substance seems to be without influence. Salts of different metals with the same acid showed a variation, in like manner, with the atomic weights : NH eS < a7 (Nh H Seu cl i My Ay Oo] a S\\\ MS WOO UII \ ee N INNA » 7s = Y a / S M4 RWeber, del. Am. Jour. Sci., Vol. |, 1896. Plate III. ea AMS” = WH}, Ay A~< Le PD (6 Y DW S aw). Y, j yj ee Zr = VEE FO s\ mi HNO TS — I / Clin RWeber. del. - = =~ = —- = = = — Ee _— — = a Sw s on =u se = — = eae “York. Plate IV. \S YS Wit Ml Si\ \) =< Yy e\ STN - |, 1896. Am. Jour. Sci., Vol. Am. Jour. Sci., Vol. I? 1896. Plate V. Am. Jour. Sci., Vol. |, 1896. Plate VI Seale, two-thirds. — = = = Sons DENIES ELE OT II Plate VII. ae two-thirds. 1. Seale Peooe: SGip eV Ol: oe Am. Jour. Am. Jour. Sci., Vol. |, 1896. Plate VIII. ZZ SS, Q Q Q Q 2 Coe k SU = CSE — ae 0 M AN VY NNSA \ROESSS D \ I\\ \ AN SSS * ® Ih ay : ‘= VS A \ & \S S x x TRIARTHRUS BECKI Green. ——=~ Sree — ——- \\ A = \ = S| A ite << = aN MASS BAN ON SS Sa \ SS NSD \\ ‘ ‘mA \ ON im Z ON . yl Up y ooo YG TR, Wy SQ V4 Sone VAN \ B, SSS55) ‘ ®W, 2 Wy, [: ge \\ 5 ory 2 sorta ( : Fe es Ty Se Se | [| 1) S Y vi a Wk YAK? S 2 : ie NNN Z ATI A = Asie Ain rr! N ay SS FAZED sie 2 | y: CD aa oe eB Bee FTO Ee A é Ce AE: \A gece oe SS SS ss \ \ ANN hws : Ds = \ 7 ea \ Tw TMI SY Fay Ah TW dutirW\ = aS MD — Wy pee) UY, elees ell /; Js MU LA ue a Wife Y) J, Ly ie, Z . CONTENTS. Art. XLVII.—Color Relations of Atoms, Tons | an cules; by M. C,:Lea. Part, II: .. 2.2.2 ee k Ph, XLVIII. Se uimetite Determination ry: Selenium ;_ | baer: W. Prigcn oo... 2-2. 2222.22.22 Ibs XLIX.—Extinct Felide of North Amennae i bas . ApAms. (With Plates X, XI, XIL) 72333geee Sous Sea L.—Age of the Igneous Rocks of the Yellowstone National — [ Park; by A. HaGun?_2- 2...) ieee ede as Bhee 2 SPAS ie: of Pollucite, angen ee and Mi ae crolite at Rumford, Maine; by H. W. Foors <-ee S .- 457 LII.—Device for Simplifying the Drawing of Crystal Forms ; by A. J. Mosus __.. 22. ._-. 1: _252 2 462 LIII.—Concerning Crookes Tubes; by ©. C. Hureniys and | F. C. Rowpinson _-....-.---.'-22 320 er 463 Paar LIV.—Researches ov the Réntgen Rays; by A. M. Maven 467 ‘ie LV.—PITHECANTHROPUS ERECTUS, from the Tertiary Of. © ~ at _ Java; by O. C. Marsu. (With Plate XH) 2a ATS Chemistry and Physics—Molecular weight of sulphur, ORNDORFF and een Absorption of the Réntgen rays by chemical compounds, NovAK and SuLo, i 483.—Dictionary of Chemical Solubilities, Inorganic, A. M. Compy: Fermenta- — tions, P. ScHUTZENBERGER, 484.—Répertoire des. Réactives Spéciaux, F. Jean and G. MERCIER: Existence of two Orthophthalic Acids, W. T. H. Hows: Na- | ture of the X-Rays, D. A. GOLDHAMMER, 485.—Recent work with Roéntgen Rays, — 486.—Diminution of the intensity of Sound with the distance, K. L. ScHA#FER, 487. ; ae , SCIENTIFIC INTELLIGENCE. j Geology and Natural History—Fifteenth Annual Report of the United States Geological Survey, 487.—Topographical maps of the United States Geological Survey: Summary Description of the Geology of Pennsylvania, 488.—Geology of the Road-Building Stones of Massachusetts, with some Consideration of Similar Material from other parts of the United States, N. 8. SHALmR: Univer- — sity Geological Survey of Kansas, conducted under authority of the Board of | Regents of the University of Kansas, HE. Haworts, 489.—Geological Survey of | Canada, 490.—Tierreich, eine Zusammenstellung und Kennzeichnung der | rezenten Tierformen. Comparative Morphology of the Galeodide, H. M. Bernard, 491.—Characese of America, T. F. ALLEN: Kryptogamen-Flora von Deutschland, Oesterreich und der Schweiz, Hysteriacez, Discomycetes (Peziz- acese), H. Rrum, 492.—Phycotheca Boreali- Americana, F. 8. CoLnins, I. HOLDEN and W. A. SETCHELL. Miscellaneous Scientific Intelligence—National Academy of Sciences, 493.—New _ F England Meteorological Society, 495. INDEX, 495. (iy "AUUUE NYA AeA ASL) AMA er naa ebaaa nn, PO ag | HERR EEEEEE TE AASSLOS GRP boii ee seve | - lalalateh-eiplamie When, pains nd \ jayne | a ail ] ene Lsanehia’ AR | toa rit Wii AN Woy (1H gepaa Me awe aL wry Rae-~ y A 4 | ““Ne Vw Les ~ , | don, \ aca ee ra ly Mb, 4 j | wh bh, ~ se, te : ee lée ‘7 ps phen ry ae HUR RES. 0 ag f 4 Y 2 4 “thy ™ Se ek aT ladle TTY Li aye \ » } . | PD ni itp Ne | By oe errs fhe 8. AHA \ R : , Asse p 8 oy Bay @ we wl LN - Mim y TY. 5, | a tg coon FAMED ia tS ae ts ae ry ~e, e ie \ ay SN oe da : PNT te aaa 4 e ‘e ow Se ie ~~» yy pr abahs La gs Le oye | bande alba ry is y. hed Sealine! || seoseamg tl aii bai acetate ie eo Naan -=e: aN sre ERT ® P 3 we ra AN VW pes ay | Bt delete il? Mu PUPA yi, Triad 20 OR i atnd - Da,- ‘a ON ane aia Pye + PPD wasasadaia, les 4 .. pepe An Ue. ) = ” se VT gabe Waa!) HO 1 DID aS a y gan a RE gor My bg WP Bey er Bs "t oe kG “A aay A AAW? , 7 MULE eG Ay, REP ye x; {* ~ ya Aye | 4 al an 7 Sal 4s »! ma MT Way AL) vA DANY bse 4 a _ - smANSD qT Waa Se Kee mo be ‘wane ETT ty btndne |yet Sy aaaen! > b By qc! ik = Ae Maa seseits aa YN etteint | alo an atc Db bg MR pur Vay 3 Pe | TT ast if? aye n° \ uudtan ance ABs ee May Boe i~ : yes, ao - ere). | v \@ VOR Bee ~ pera | TX ing TARA cheese ma umuneree iimeans Aa fw? 2N\h SRS er a ub -~ Vf a Ny WA - ee SUMMS ste ease i, poaequeune Ns jee ola i, * wananuatale aR iy Pyat Po ah. peur NS WY t vidi, Y) wenn ert a | Tee, | | Pits oat i Zee pat SSAC yeuery, Aa Ped dd whl By ME j we Se ea ae a We VA, : ganneee Be Py Pe ide if a “vib in Kham, nm ooo Maa, i aaah, yi os lal ol ater nowy ec UNA vy yyy vYrre a iN SLL LT Muy PMA Syl ne Maan Tl m1) vil "AAV eengu, vase . WIP IN/RADaiaPhya ab? ‘i NAMA, uaeereaie 7 Osea. dt Ved Nupsiyiiis,_. wae tI MY Ved 4 ae pe epi ~~! > Ns 4\ by al Von | al Wd Fesaceeoubte retepaiieeten ny aaa mad EQ ST tabi i. \ Viueu al? 1 “ Pa ari Ay I api ‘ wake i i ani hl i Ka, ‘ft P| Oren US ge aA ee et AALS od ld wae a atime ca con aad a wii Bs s aE : ws twa sy WI 3 9088 01298 5529