MINERALOWCAL CHEMISTRY.Mineralogi c a1 Chemistry.21Minerals of the Cryolite Group recently found in Colorado.By W. CROSS and W. F. HILLEBRAND (Amer. J. Xci. [3], 26, 271-294) .-Cryolite and several allied fluorides have been recently dis-covered in the Pike’s Peak region, Colorado. The country rock ofthe whole district is a coarse reddish biotite granite of the ArchEanformation.The minerals described occur in two veins of massive white quartz,about one-third of a mile apart. In the vein designated A, cryolite,pachnolite, thomsenolite, gearksutite, prosopite, and, probably ral-stonite appear. In vein I3, prosopite, fluorite, and mixed fluoridesappear intimately associated with zircon, kaolinite, and mica.The cryolite from vein A is of a faint pink colour, and contains as avisible impurity the Fe,O, represented in the analysis-Fez03. Al.Ca. Na. H,O. F. Total. Sp. gr.0.40 12.90 0.28 32.40 0.30 53.55 99.83 2.972Previous to the analysis by J. Brand1 (Abstr., 1882, 1176) of pach-n d i t e , carefully selected by Groth, pachnolite and thomsenolite wereconsidered t o possess the same chemical composition. The resultsof all earlier analyses agree very well with the formulaNaF,CaF2,A1F3,Hz0,so that Brandl’s analysis agreeing with the formula NnF,CaF,,AlF,caused great surprise. Groth explained the supposed anhydrousnature of yachnolite *by assuming that the mineral analyeed waslargely contaminated with thomsenolite, but as the percentage of waternever falls. below 7 per cent. in any of the analyses published, thisassumption necessitates a most improbable admixlure of thomsenolite.All the analyses made by the authors correspond with the formulaNaF,CaF2,A1P3,H,0. The results of an analysis made upon crystalscarefully identified as pachnolite by the rhombic section, were asfollows :22 ABSTRACTS OF CHEMICAL PAPERS.81.Ca. Na. H20. F. Total.12-36 18.04 10.25 8.05 (51.30) 100*00Hence it, appears that the pachnolite from Pike’s Peak and thomseno-lite are identical in chemical composition, unless the fact of all analysesof thomsenolite showing slightly more water than that required forthe formula NaF,CaF,,AlF,,H,O, may indicate a partial replacement offluorine by hydroxjl.I n small cavities in the massive pachnolite, a mineral was found,occurring sparingly in a few specimens, which seemed to be differentfrom any known species.Al.Ca. Rlg. E. Na. F. Total.11.40 0.72 0.22 28.94 9-90 46.98 98-16The analysis gave-from which may be deduced a formula analogous to that of cryolite,in which about two-thirds of the Na is replaced by K. The authorsare of opinion that this mineral and ralstonite are not identical, butfurther investigations will be made as soon as better material can beobtained for examination.Itis made up of exceedingly minute colourless needles, and resembleskaolin in a remarkable degree. The analytical results obtained wereas follows :-Gearksutite is abundant among the minerals from this locality.LossAl. Ca. Na. K. H,O. F. as oxygen. Tot,al.15.31 22.30 0.10 0.04 15-46 42.07 4-78 100.00I n this Journal (Trans., 1883, 140) Flight described a mineralobtained from the cryolite bed of Greenland, which he considerednew, and named evigtokite; but from the agreement of his analyseswith those previously published, the similarity in occurrence aridphysical characteristics, there can hardly exist a doubt that Flightamalysed gearksutite.The name evigtokite should therefore bedropped.The rare mineral prosopite is more frequent in rein B than invein A. The analytical results obtained were as follows :-81. Ca. Mg. Nrt. H20. I?. as oxygen. Total.Loss22.1‘7 17-28 0.17 0.48 13.46 33.18 13.26 100.00B. H. B.Stibnite from Japan. By E. S . DANA (Amer. J. Sci. [ 3 ] , 26,214--221).-1n the size and beauty of the crystals, and in the greatcomplexity of their form, the Japanese stibnite far exceeds specimensfrom other localities.The finest specimens known to the author arethose recently acquired by the Yale Museum. They were obtainedfrom Mount Hosang, near Seijo, in South Japan. They are of re-markable size. A simple group consists of two prismatic crystals,with vertical axes nearly parallel ; the longer of the two measures22 inches, and the other 15 inches in length, the thickness variesfrom 1i to 2 inches. The lustre is very brilliant, and can be compareMINERALOGICAL CHEMISTRY. 23only with that of highly polished steel. The complexity of formobserved among Japanese stibnites is their most remarkable character.Previous t o 1864 b u t 16 planes had been identified. Krenner (Ber.Wien.Akud., 1864, 436) added 28 new planes, and Seligmann (Juhrb.f. Mia., 1880, 1, 135) added one more. Of these 45 planes, 30 halebeen observed on the Japanese crystals, and, in addition t o these,40 new planes have been determined, thus increasing the list to 85.This list could be considerably enlarged if the planes admitting cfonly doubtful determination were to be added.*By W. C. EUSTIS (Chem. News, 48, 98).-The specimen examined had a dark-coloured incrustation containingiron, a little manganese, and traces of nickel and cobalt. The mineralwas greyish-white to nearly c~iourless, the surface was marnmillary,and the structure radiate o r fibrous. Lustre, sub-vitreous or slightlysilky in some parts; hardness, 3 ; sp.gr., 2.4. Before the blowpipeit whitens, and loses water, but does not fuse. Analysis yielded thefollowing results :-B. H. B.Gibbsite from Brazil.A1,03. H,O. Fe20,. CaO. MgO. SiOa. Total.63.81 25.85 0.49 0.30 0.03 0.20 = 100.68No aluminium phosphate could be detected. D. A. L.Corundum Gems in India. By C. U. SHEPARD (Amer. J. Xci.[3], 26, 339-340).-A remarkable deposit of sapphire and ruby hasbeen discovered at Sungchang in the Himalaya Mountains. Thematrix is a schistose or slaty rock, and the vein consists of quartz,amethyst, and several varieties of corundum, all beautifully crystal-lised. In addition t o these, massive corundum, chlorite, and a littlemagnetite also occur. There is a great analogy in the mode of occur-rence and other particulars between the crystals found in India andthose produced at several of the American localities, notably a t theLaurens district, S.Carolina; Pelham, Mass., and Burke Co.,N. Carolina. The resemblances are, in fact, sufficiently important tolead t o the belief that valuable corundum gems may yet be found inthe United States. B. H. B.Cassiterite, Spodumene, and Beryl from Dakota. ByW. P. BLAKE (Amer. J . Sci. [3], 26, 235).--Caseiterite has been dis-covered in the central region of the Black Hills, Dakota, in a mass ofcoarsely crystalline granite. The felspar, quartz, and mica are inunusually large crystals ; and, in addition to these minerals, there isan abundance of spodumene in gigantic crystals ranging from 2 to 6feet in length, and 8 to 20 inches in diameter.I n the midst of thespodumene and felspar crjstals, the first masses of tin-ore werefound. The chief associate of the tin-ore is a dense aggregation ofsmall mica, crystals. At another locality, where muscovite mica has* Since the above memoir was published, a much finer group of Japanese stibnitecrystals has been purchased from Mr. Henson for the British Musenm. The groupconsists of 120 crjstals from 0.5 to 2’5 inches wide and 13 inches long; weight,150 lbs.-B. H. B24 ABSTRACTS OF CHEMICAL PAPERS.been mined, the author found large crystals of beryl embedded ing u art z .Descloizite from Mexico. By S. L. PENFIELD (Amer. J. Sci. [3],26, 361--365).-The mineral analysed was said to be from Zacatecas,Mexico.In all physical properties it seems to be identical with trito-chlorite recently described by A. Frenzel (Abstr., 1882, 473). Theanalytical results obtained are given under I. Frenzel's analysis oftritochlorite is also given (11) for comparison :-B. H. B.V,O,. A S ~ ~ , . P,O,. PbO. CuO. ZnO.I.. ... 18.95 3.82 0.18 54-93 6.74 12.2411. .... 24.41 3.76 - 53.90 7-04 11.06FeO. H20. Si02. Total. Sp. gr.I. .... 0.06 2-70 0-12 99.74 6.2011. .... - - - 100.17 6.25The minerals have the same density, and are alike in many payti-culilrs, but i t seems hardly probable that Frenzel could have overlooked2.70 per cent. of water. The difference in the percentage of vanadicacid is also considerable. B. H. B.Manganese Ores.By M. LILL and L. SCHNEKDER (Dingl. poZyt.J., 248, 471).-The authors have analysed the following samples ofmanganese ores from Bukowina :-I. From Upper Arschitza ; 11.Lower Arschitza, and III and IV, from the mines Theresia andSchara. The following results were obtained :-MnO.. ....&ln,O, ....FeO ......Fe?03 ....AI,O, ....CaO.. ....Bn,O ......CUO.. ....co .. t...Alkalis.. ..SiO ?...... so, ......P20, ......( 3 0 2 ......H,O ......MnO, ....MgO ....I.0.828.2054.2 70.7716.710.461.080-47tracestracestraces0.4810-95traces0.8425.25-11.1-947-6039.580.6527.341-731.030.280.004traces0.3813.000.0941.1115.55I-111.0.67€Pi952.520.3016.272.021.800.30tracestracestraces0.3010.900.08G.53traces5.25IV.1-497.0147- 140-5 112.780.853-500.570.0060-4118.10traces0.630.1 86.50--100302 100-289 99.73 99.679D.B.Native Ferrous and Aluminium Sulphate from Mexico.By T. P. LIPPITT (Chem. News, 48, 98).-The specimen was a cornMINERALOGICAL CHENISTR Y. 25pact mass of flexible fibres resembling asbestos. It was of a palegreenish-white colour, and had a silky shining lustre. Hardness, 2 ;sp. gr. 1.89. When heated in a glass tube, it gave off acid water.A small quantity of gypsum was intermixed with the mineral, whichwas otherwise quite homogeneous. The mineral is soluble in water,and yielded on analysis the following numbers :-Fe (ferrous). Al. Ca.so,. H?O. Total.7.81 4.92 0.52 41.59 43.60 = 98.44The iron is all feryous. Subtracting the constituents of gypsumfrom the results the formula Fe,”Al,( S04),(H,0), is obtained, whichwould be equal to a mixture of 2 mols. of halotrichite + 1 mol.melantrite.2 { Fe”A12( S O&. (H,O) 22 ] + Fe”S 0, ( H20) ,.D. A. L.Occurrence, Association, and Probable Mode of Formationof Barytes, Celestine, and Anhydrite. By DIEULAFAIT (Compt.rend., 97, 51-43).Barytes is generally found associated with metallic ores containingsulphur, the metals present in these ores being antimony, arsenic,lead, silver, mercury, copper, cadmium, bismuth, zinc, and manganese.With the exception of the last two these metals have a strong attrac-tion for sulphur; they are precipitated by hydrogen sulphide fromacid solutions, and they are generally found in nature in the form ofsulphides, which appear to be their most stable compounds.Zincis frequently found as sulphide, b u t its most stable form is theoxide, although the tendency of the natural sulphide to pass into thestate of oxide does not appear. to be very great. The most stable form ofmanganese is the dioxide. Some of the sulphides with which barytesis associated are very volatile, others are decomposed when heated inpresence of air (which generally has access to metallic veins), and allare decomposed when fused with alkaline chlorides. These and otherfacts show that the barytes and the minerals with which it is associatedhave never been subjected to the action of even a moderately hightemperature, and have certainly not been formed in contact with thefused alkaline chlorides.CeZestine, unlike barytes, is generally found in saliferous deposits,in which it is associated with gypsum and rock-salt, and it exists iiismall quantities in the gypsum of such deposits. If the gypsums arerich in organic matter, as is the case in all those in which free sulphuris found, polysulphides of calcium and strontium are formed in theinterior of the rock.These polysulphides are dissolved and washedout by percolating water, and when brought ill contact with the airthey are decomposed, with formation of calcium sulphate, calciumcarbonate, strontium sulphate, and free sulphur, and the frequentoccurrence of crystals of celestine on the surface of crystals of sulphurshows that the former must have been deposited a t a temperaturebelow the melting point of the latter,Bqzhyldrife frequently alternates with gypsum, which agrees withGorgeu’s hypothesis, but, on the other hand, it is quite as commo26 ABSTRACTS OF CHEMICAL PAPERS.in gypsum, especially in the triassic deposits in the south-east ofFrance, where the anhydrite often passes insensibly into gypsum.Itcannot be supposed that the anhydrite alternating with gypsum hasheen deposited from fused sodium chloride, since the gypsum beginsto lose its water even below 100". I t would appear, therefore, thatGorgeu's hypothesis cannot be accepted as an explanation of themode of formation of barytes, celestine, and anhydrite existing inveins and in saliferous deposits, although it may be of value in con-sidering the chemical changes which accompany volcanic action.C.H. B.New Locality of Chalchuite. By W. P. BLAKE (Amer. J. Sci.[3], 25, 197-200).-The occurrence in New Mexico of a green tur-quoise, chalchuite, has already been described by the author (Amer.J. Sci., 1858, 25, 227). Chalchuite also occurs in Cochise County,Arizona, in an outlying ridge of the Dragoon Mountains, now knownas " Turquoise Mountain." The mineral is identical in appearance withthe New Mexican variety. It is of a light green colour, with a sp. gr.of 2.71-2-82. It is peculiarly interesting archEologically. It was ingeneral use among the Aztecs before the arrival of the Spaniards, andthe author proves its identity with the caZZais or callnina of Pliny.B.H. B.Scovillite, a new Phosphate of Didymium, Yttrium, andother rare Earths from Salisbury, Conn. By G. J. BRUSH andS. L. PENFIELD (Amer. J. Sci. [3], 25, 459--463).--This mineral wasdiscovered by J. S. Adam, occurring sparingly as an incrustation onsome of the iron and manganese ores from the Scoville ore bed. Theincrustation is one-sixteenth of an inch in thickness, and is frequentlybotyroiclal or stalactitic. It is of a pink to yellowish-white colour,and presents a radiated fibrous structure. H. = 3.5. Sp. gr. 3.94-4.01. The results obtained on analysis were as follows :-P,O,. Y203 + Er203. La203 + Di,O,. Fe203. Combined H,O.24.94 8.51 55.17 0.25 5-88HaO lost at 100".CO,. Total.1.49 3.59 99.83The presence of the carbonate is regarded as due to an admixtureof lanthanite, (La,L)i),(C02),,9H,0. The mineral as analysed is calcu-lated to be a mixture of 17-04 per cent. of lanthanite, and 82-79 percent. of the new phosphate, R2( POa),,H,O. Calculating the 82.79 percent. of Scovillite up to 100 per cent., we obtain-Y205. (YEr),O,. (LrtDi)203. Fe203. H,O. Total.30.12 10.28 55.73 0.30 3.57 100.00B. H. B.Analyses of Lithiophillite. By S. L. PENFIET~D (Amel-. J. Sci. [3],25, 176) .-The author has already published analyses of triphylliteand lithiophillito from the various known localities (Abstr., 1879,695). He now adds to these two analyses, one of lithiophillite froMINERALOGlCAL CHEMISTRY.27Norway, Me (I), the other (11) of a variety from Branchville, Conn.The results obtained were :-Sp. gr. P,O,. FeO. MnO. CaO. Li,O.I. ... 3.398 44.40 8.60 35.98 0.78 8.5011. ... 3.504 44-93 16.36 28.58 0.05 8.59Na-0. H20. Gangue. Total.I. .... 0.14 1-19 0-12 99.7111. .... 0.21 0.54 0.13 99.39These analyses fully substantiate the formula Li(MnFe)POj,already made out for the species. B. H. B.Determination of Alkalis in Lepidote from India. By M.PAGE (Chew,. News, 48, 109--IlO).-The mineral examined, of a leadgrey colour, was found in granite associated with a violet-red lepidote,quartz, and occasionally tinstone. The total alkalis and sodium andpotassium were determined in the usual way ; lithium was determinedas phosphate ; rubidium by fractional precipitation with platinic chlo-ride and numerous washings with boiling water.The results percent. are:-K20. Li20. Na,O. Rb,O.8.595 1.754 0.609 0.070D. A. L.Topaz from Maine, U.S. By C . M. BRADBURY (Chem. News, 48,IO9).-The sample examined consisted of large colourless translucenttopaz crystals; sp. gr. 3.54. An analysis yielded the followingfigures :-A1 27.14 ; Si 14.64; F 29.21 ; 0 28.56 ; total, 99.55. ThisIS very anomalous, as three-quarters and not half the oxygen, corre-sponding with silicon, is replaced by fluorine in ordinary topaz.Berzelius’s analytical method was employed. D. A. L.Analyses of Franklinite Ores from New Jersey. By P.RICKETTS (Dingl. polyt. J., 248, 523).Si02 ................ZnO ................FeO, Pe,03, and IE’e,04Al,03 ..............MnO ................CaO ................MgO ................c02 . .. . . . . . . . . . . .cu ..................I.11-8534.1 328-480.5814.1 35-510.134-960.0711.11.5940-8329.94traces8.354-160.794.12111.8.6434-7028.34traces15.505-701 *446.26tracesIV.1 O . i O33.0931.05traces15.514.590.274.38traces - -99.84 99.78 100.48 99.59The following composition corresponds with the latter analysis :28 ABSTRACTS OF CHEMICAL PAPERS.Franklinite. ............... 51.51Red zinc oxide ............ 6.40Rhodonite ................ 11.13Willemite ................ 20.23Manganese carbonate ...... 1.24Limestone ................ 8-7699.27-Analysis of the most important constituents gave :-Zinki te.Franklinite. Willemite.ZnO.. .. 95.20 Zn0.. .. 20.72 ZnO.. .. 69.97MnO. .. 3.19 MnO ... 12.72 MnO ... 1.14FeBOa.. . 63.90 Fe30a.. . tracesSiOz ... 16.81D. B.Chrysocolla from Arizona. By W. C. EUSTIS (Chem. News, 48,109).-The specimen for analysis was associated with a bluish-greenvariety of chryaocolla, and with much carbonate. The sample analysedwas emerald-green, transparent, and pseudomorphous, with vitreouslustre. Sp. gr. 2.3; hardness, 3.5. Analysis yielded CuO, 32.28;Si02, 34.08 ; H20, 31.65 ; total, 98.95. These figures approximate to theformula 3Cu0,4Si02,13H,0, which corresponds with some specimensof chrysocolla of the constitution CuSiOs,3H20, in which one quarterof the copper is replaced'by hydrogen ; the mineral analysed may belooked upon as a copper hydrogen silicate.D. A. L.Rocks of the Yellowstone Park. By W. BEAM (Amer. J. Xci.1. Porphyritic Obsidian.-Colour greenish-black, semi-transparent.[3], 25, 106).H. = 6. Sp. gr. = 2.4. The analysis gave-Al,O, andSiO,. Fe,03. CaO. MgO. Na20. K20. HzO. Total.77.00 13.40 1.25 1.19 3-43 3.62 0.70 100.592. Pebble of quartz trachyte covered with a deposit from E'chinusGeyser.-The pebble analysed was about 1.5 inches in diameter, of alight fawn colour, and contained small masses of colourless silica. Theanalytical results were as follows :-SiO,. A1,O,+Fe,Os. CaO. MgO. K20. Na,O. H,O. Total.77.90 14.55 0.40 trace 4.63 2.10 1.00 100.58B. H.B.Volcanoes of Northern California, Oregon, and Washington.By A. HAGUE and J. P. IDDINGS (Amer. J. Xci. [3], 26, 222-235).-During the year 1870 the geologists attached to the Geological Explo-ration of the Fortieth Parallel, made a preliminary survey of theextinct volcanic cones of North California, Oregon, and Washington,but a further study was never undertaken. As the rock specimensthen collected may be considered as representing the principal typeMINERALOGICAL CHEMISTRY. 29of the ejected lavas, a large number of thin sections have recentlybeen made, and their microscopic examination has been followed upby chemical investigaiion.The four great cones, Lassen’s Peak, Mt. Shasta, &It. Hood, andMt. Rimier, which may be taken as typical of the chain, are all ande-site volcanoes, with extrusions of basalt.Similar variations inmineral composition and minute details of structure are found a teach of the volcanoes. All the socks from these volcanoes niay beclassified under the heads of basalt, hypersthene-andesite, hornblende-andesite, and dacite.Hypersthene-andesike.-These rocks are generally very porous,varying in colour from blue-black t o steel-grey. They occur in allstages, from crystalline, dense forms to glassy pumice. They aregenerally crowded with very small porphyritic crystals, of which thefelspars are the most noticeable, the iron niagriesium silicates being moseapparent in the light-coloured varieties. A complete chemical analysisis given (I) of a pumice from which the hypersthene was extracted.The ground-mass is an almost pure glass, and the microscope sliows thatthe hypersthene was the first of the essential minerals to crystalliseout from the original magma.The analysis of this pumice does notvery materially differ from the analysis of the more compact lavas.After separating the heavier minerals from the felspar and glass, t’helatter were subjected to Thoulet’s solution, in order to isolate theglass from the felspar. Two separations of felspar were obtained,one with sp. gr. from 2.66-2.68, and the other from 2.64-2.66.Analyses of these are given (I11 and IV). These probably representthe same felspar rendered impure by foreign ingredients. An analysisof the glass which forms the base of the pumice is given under V.Sp.gr. = 2.29.SiO? ............FeO ............CaO ............MgO ............R,O ............NaQO ..........Mn 0. ...........TiO, ............P,Oj ............Ignition ........A1203 ..........I.62-0017.844.405.372-641.474-29trace0.1 70.291 -6611.50.330.9 722-001-8823.29 --0.64111.56.412 7.390.699-870.090.365.43IT.56.9527.47trace9.100.020.485-78Total .... 100.13 99.11 100.24 99.80Dacite.-This rock occurs near the highest point of Lassen’s Pea,k.It is probably the most recent extrusion from the peak, and is quiteunlike the rocks obtained from the other volcanoes. The specimensvary from a moderately compact rock to others of a pumice-like cha-racter.The rock is composed of quartz, mica, hornblende, andplagioclase, and is from 6-9 per cent. richer in silica than the pre30 ABSTRACTS OF CHEMICAL PAPERS.vailing rocks of the four main cones (Analysis VI).felspar (VII) and of the glass (VIII) are given.bably a mixture of andesine and oligoclase.Analyses of theThe felspar is pro-V.SiO,.. .......... 6994Al,O, .......... 15.63Fe,O, -FeO ............ 1.89CaO ............ 2.49MgO.. .......... 0.28K,O ............ 2-8.5Na,LO.. .......... 3.83Jgnition ........ 3-25Total .... 100.16..........VI .69-3616.230.881.533.1 71-343.024-060.45VII.65.7721.51trace5-720.835.920.34-VIII.76.7512.321-361.183983.550.54--200.04 100.09 99.68The four types of rocks, basalt, hypersthene-andesite, hornblende-andesite, and dacite, exhibit four well charscterised groups, but theyare not sharply defined and distinct forms, as between any two in theseries all possible intermediate varieties exist.Meteoric Iron from Georgia.By C. U. SHEPARD (Amer. J.[3], 26, 336--338).-The mass here described was found in 2879about 14 miles north-east of Dalton, Whitfield Co., Georgia. Itweighed 117 lbs. The surfaceis black and very little oxidised. The sp. gr. is 7.986, which is some-what higher than is usual in meteoric iron. The analysis gave thefollowing results :-B. H. R.Its shape is somewhat that of a pear.Fe. Ni . co. Total.9 4 66 480 0.34 99.80with traces of phosphorus, chromium and manganese.B.H. B.Geyser Waters and Deposits. By H. LEFFMANN (Amer. J. SC;.[3], 25, 104--105).-The specimens from which the following ana-lyses were made were collected from the Pellowstone Park in 1878.&st of the geysers and hot springs are siliceous, and in most of thewaters examined the silica is in the free condition. All the resultsare given in grains to the imperial gallon:-1. Pearl Geyser.CaSOd. Na2S 04. NaC1. SiO,. Total.1.40 1-89 61-39 7.84 72.52At the bottom of the bottle containing this water was a quantity ofgelatinous matter, which was found on analysis to be composed of79-1 per cent. SiOz and 4.9 HzO, together with traces of A1203,Fe303, and CaO.2. Jug Xpwhg.CaCO,. Na,CO,. N~I$~OJ. NaCl.SiOz. Total.0-79 49.14 2-12 31.57 14.56 98.1MINERALOGICAL CHEMISTRY. 313. Opal Spring.NaCl. CaSO,. CaCl,. SiO,. Total.72-18 3-22 4.06 53.76 143.22This is not a geyser but a spring having the temperature of 90' F.The water is opalescent.4. Deposit from Bronze Spring.-This occurs in convoluted layerswith bronze-coloured surfaces and fawn-coloured streak, H. = 5.5.The results obtained on analysis were as follows :-Fe203 Organic matterSiO,. and A1203. and water.83.1 1.2 13.6B. H. B.Solid and Gaseous Constituents of Sea Water and OceanicDeposits. By H. TONROE and L. SCHMELCK (Bied. Centr., 1883, 217-231) .--These investigations were made by the Norwegian NorthAtlantic Expedition in 18i6-78, when a tract of ocean lying between60" and 80" N.and 12"--37" E. was examined.Solid Constituents.-There is but very little variation of the sp. gr.found ; a few samples, however, were much lighter, but this was dueto the nejghbourhood of ice, or river mouth, as the solids still bore aconstant ratio to one another. As regards the percentages of chlo-rine, lime, magnesia, and sulphuric acid present in water taken fromvarious depths, they remain practically constant, neither are theyaffected by the latitude ; in those cases in which potassium has beenestimated (as chloride) no variation has been foiind. The authorsfound only 0.0025 gram per 100 C.C. of or ganicmatter, and no hydrogensulphide. When sea water is boiled, the evaporated water being con-stantly replaced, all the carbonic anhydride is expelled, and a precipi-tate of magnesium salt free from lime is formed; but if on the con-trary, the water be evaporated to say one-half, only part of the gas isdriven off, and the precipitate consists of calcium carbonate andgypsum ; in both cases the water becomes alkaline ; further concen-tlratiori causes the precipitation of gypsum and sodium chloride, themother-liquor containing all the magnesium and potassium salts.Air itt 8ea Wader.-Jakobsen in 1871-72 found less oxygen at lowerdepths than a t the surface.Buchanan found the percentage of oxygen in surface water to varywith the latitude: so do the authors, obtaining 33.93 per cent.(mean) in the German Ocean, 34.94 per cent.in latitude 50-70" N.,and 35.54 per cent. in latitude 70-80".As these results do not tally with those obtained by Bunsen, whostates that the composition of the absorbed air is constant for all tern-peratnres : experiments to elucidate the question were made, and theconclusion arrived a t was, that seeing that the percentage of nitrogenwas normal, there must be some unknown factor which causes thewater to be supersaturated with oxygen.When comparing the volumes of oxygen in the water collected fromdifferent depths, it was noticed that the reduction was at first veryrapid, but soon became less SO, and a t 1000 m.the minimum wa32 ABSTRACTS OF CHEMICAL PAPERS.reached, increasing slowly afterwards, being 0.4 per cent. a t 3000 m.The curves representing percentages of salt and nitrogen at differentdepths, show that the percentiige of nitrogen is inversely proportionalto that of salt.Deposits.-Schmelck collected 300 samples of the sea bottom, andall may be classed under the name of clay, but of this clay there arefive kinds-grey, transition, biloculine, rhabdamine, and volcanic.The grey clay covers the whole surface of the bed of the sea, but issuperposed a t depths of more than 1000 fathoms by biloculine ; inshallow seas, as along the coast of Norwa,y and Spitzbergen, muchsand, flint, and mussel shells are found, the percentage of calciumcarbonate being 9 per cent.At 500 fathoms, brown clay overlies thegrey; this brown clay is distinguished from tbe brown hiloculinewhich only appears a t 1000 fathoms by the coarse sand present, andthe paucity of Foraminifera ; it is this clay which has been termed“ transition.” Biloculine, which is distinguished from grey clay byits colour, and from trarisitional by its s’tructureless fineness and uni-formity of external appearance, is rich in Foraminifera, and conse-quently in lime : the principal Foraminifera present being Globigerinabiloculina, lituola, and nonionina ; the minerals are microscopic grainsof quartz and flakes of mica ; one difference is remarkable, that whereasthese particles are sharp-edged in this clay, these are all rounded inthe transition clay. The green or rhabdaminic clay is found in theshallow waters which lie between Norway, Beren’s Island, and NovaZembla, and cont’ains a large number of Foraminifera, but littlecalcium carbonate, much silica. Volcanic sands and dark green sandyclay is found about Jan Mayen’s Land, and contains lava, tufa, felspath,augite, hornblende, magnetit and olivin. All of these deposits varyconsiderably in their percentage of ferrous oxide, the lighter colourecikind containing least. E. w. P