MINERALOGICAL CHEltIISTRY. Mineralogical Chemistry. 45 Behaviour of Minerals when submitted to the X-Rays, By CORNELIUS DOELTER ( Jcchrb. f. Mi%., 1896, ii, 87--106).-The several minerals examined show great differences of transparency for the Ront- gen rays, and no general relation between the density and transparency can be traced, but minerals with a density greater than 5 seem to be opaque. Also, no relation can be traced to the chemical compo- sition or molecular weight, but sulphur and arsenic compounds are the most opaque, boron and aluminium compounds, as a rule, the most transparent, and in silicates the opacity increases with the amount of iron. Dimorphous minerals do not show any important differences. Crystals show only slight differences in various directions. A scale of eight degrees of transparency is given :--I diamond, 2 corundum, 3 talc, 4 quartz, 5 rock-salt, 6 calcite, 7 cerussite, S realgar.Boric anhydride is even more transparent than diamond ; realgar is quite opaque. For distinguishing between precious stones and their imitations, the method has a practical application. L. J. S. By JAN W. RETUERS (Julwb. f. Mi%., 1896, i, 212--221).-Penfield (Abstr., 1896, ii, 216) has mentioned that as Retgers’ thallium-silver nitrate acts on mineral sulphides, it cannot be used for their separation ; this is due to the oxidising action of the nitrate. Blende is energetically attacked by molten silver nitrate with evolution of nitrous fumes, and separation Heavy Liquids for the Separation of Minerals.46 ABSTRACTS OF CHEMICAL PAPERS.of metallic silver and zinc sulphate. Thallium nitrate acts on pyrites, but, at a moderate temperature, not on blende. The acetates of thallium, silver, lead, and mercury, and various double and basic salts, as well as mixed acetates and nitrates of these heavy metals, have been examined, but the results are not very favourable, because, although the melting points are usually low, the specific gravity of the liquid is too low for use with mineral sulphides, being rarely above 3-4. Thallium-silver acetate (m. p. 75", sp. gr. 4.8) becomes opaque owing t o separation of metallic silver. Thallium-lead acetate (sp. gr. 3.6) is liquid a t the ordinary temperature, and has a high index of refraction. The only ones that would be of any use are thallium acetate (m.p. l l O o , sp.gr. 3.9) and thallium nitrate-acetate (m.p. 6 5 O , sp. gr. 4.5); the former, however, owing to its low sp. gr., would be only applicable in a few cases, and the latter decomposes a t tempera- tures above 100". L. J. s. Diamondiferous Sand fiom Brazil. By HENRI MOISSAN (Compt. 9*end., 1896, 123, 277--278).-Sand from Brazil, after removal of the siliceous constituents, was found to contain microscopic, transparent diamonds, black diamonds and graphite, together with particles of gold and platinum. It is noteworthy that in Brazil as well as at the Cape the diamonds are associated with graphite. Free Gold in Granite. By GEORGE P. MERRILL (Amey. J. Xci., 1896, [4], 1, S09-311).-A specimen labelled "gold ore, Sonora, Mexico " is an ordinary black-mica granite which, owing to weathering, is brown and friable ; throughout the mass are numerous specks, rarely over 1 mm.in diam., of native gold. Pyrites and other sulphides are not present. Microscopical examination of thin sections of the rock shows the gold enclosed in the quartz and felspar as beautiful arbofes- cent and platy forms; it therefore does not seem to be a secondary constituent which has been deposited along fissures, but to be a primary constituent which has crystallised from the original magma. Such a mode of occurrence of gold does not seem to have been before described. L. J. S. C. H. B. Nitrogen and Argon in Firedamp and in the Gas from the Rochebelle Coal Seam. By TH. SCHLOESING, Junr. (Compt. r e d . , 1896, 123, 302--305).-The firedamp, collected without admixture of air and containing no oxygen, was found to contain the following pro- portions of nitrogen and argon.Pressure of issuing firedamp in em. of Mine. water. Anzin ............... 400 BessBges ............ 4 to 5 Firininy ............ 16 Li6vin ............... 70 Plat-de-Gier ...... 75 Ronchamp ......... 8 Saint-fitienne ...... 600 Argon and nitro- gen in 100 vols. firedamp. 18 '1 3.8 0-74 8 -0 30.0 2.8 3.2 Argon in 100 vols. fire- dainp. 0'594 0.064 0'012 0'166 0'601 0'031 0'037 Argon in 100 vols. nitrogen and argon. 3 *28 1.63 1 -G7 2'32 1-83 1'09 1.17 I n the mines of Rochebelle there are sudden, and often violent, outbursts of gas which has the composition CO,, 98-13, nitrogenMINERALOGICAL CHEMISTRY. 47 andargon 1.14, methane 0.73.100 c.c of the gas contains 0.021 C.C. of argon and 100 C.C. of the nitrogen and argon contain 1-87 C.C of the latter. Argon was found in all specimens of firedamp that were examined, but the proportion of nitrogen and argon in the gas varied widely. The ratio of argon to nitrogen also varied considerably, and often exceeded the ratio in atmospheric air. The most probable explanation is that the argon has been derived not directly from the atmosphere but from air dissolved in water. The possibility of some subterranean source OF argon is of course not excluded. By STEPHAN F. PECKHAM and LAURA A. LINTON (Amer.J. Xci., 1896, [4], 1, 193--207).-The authors have made analyses of the pitch found in and near the Pitch Lake of the Island of Trinidad. The pitch found within the Lake and also that outside have very much the same composition, and in all cases the pitch is fully saturated with moisture, usually containing some 25 to 30 per cent.About 38 per cent. of the residue is sand, the rest is bitumen and fragments of vegetation and disorgnnised cellular tissue. The pitch which rises in the middle of the Lake has a vesicular structure. When freshly dug, its colour is brown, but if left in the sun i t soon darkens, finally becoming bluish black. A large mass when placed in bright sunshine 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. C. H. B. Trinidad Pitch, Numerous analyses are quoted.J. J. S. Rutile, Cassiterite and Zircon. By HERMANN TRAUBE (Juhi-6. f. Min., 1896, BeiZuge, Bd. 10, 470-476).-By the action of potassium fluoride or of hydrogen potassium fluoride similar etch figures, indi- cating holohedral symmetry, were obtained on rutile, cassiterite and zircon ; this similarity cannot, however, be taken to support the view that zircon is isomorphous with the other minerals. Various analyses of rutile have shown small quantities of Fe,O, (up to 10 per cent.), Mn203.and Cr20,j and the author has artificially pre- pared rutile containing these oxides. By heating titanium dioxide with sodium tungstate and various oxides at a high temperature (1700") rutile crystals were obtained which contain up t o 5.4 per cent. Fe203, 3-01 Mn20, or 1.91 Cr,O,.Under the microscope, these crystals, as well as the natural crystals, do not shorn any inclosures, so that these oxides have been dissolved in the titanium dioxide. Cobalt and nickel oxides were not taken up in this way. A peach-coloured cassiterite containing chromium mas also prepared. The changes in colour which occur when crystals of rutile and cassi- terite me heated are described, RutiIe and cassiterite are acted on by potassium fluoride with the formation of K,TiO, and K,Sn06. L. J. S. Reniform Limestone from Villejuif. By LOUIS FRANCHET (BUZZ. Xoc. Philomuthique, Paris, 1896, [S], 8, 10--12).-1n the lime- stone at Villejuif are numerous reniform and globular masses very like the menilite of Menilmontant in appearance, and in fact they have48 ABSTRACTS OF CHEMICAL PAPERS, been described as such. The material is white or yellowish and very fine in grain.Sp. gr. 2.34-2.72; H. 4.5-5. Acid dissolves calcium carbonate and leaves a residue of silica. The following analyses show that tho material is a siliceous limestone and not menilite, which is a variety of opal. SiO,. A1,0,. Fe,O,. CaO. SrO. MgO. K20. Na,O. CO,. Total. 36.52 1-31 0.23 33-19 0.34 1.69 0.25 0.38 27-81 100 72 24-16 1.53 0.21 38-22 0.42 2.01 0.23 0.44 32.84 100.06 Also traces of manganese and sulphuric acid. L. J. S. Northupite, Pirssonite, Gaylussite and Hanksite from Borax Lake, California. By JULIUS H. PRATT (Anzer. J. Xci., 1896 [a], 2, 123--135).-The minerals here described are found associated with borax at Borax Lake, San Bernardino Co., California.Borax Lake is an alkali marsh which is usually dry, but some water collects in it during the wet seasons. Nortkupite (compare Abstr., 1896, ii, 184).--This was found in a single boring as isolated and unmod.ified regular octahedra. When pure it is colourless, but owing to impurities, probably of clay and organic matter, the colour usually varies from dirty white to dark brown. It is very brittle and shows no cleavage. Sp. gr. 2.380. It is isotropic, and for sodium light p = 1.5144. The mean of two analyses is Total (less CO,. CI. SO,. MgO. Na,O. K,O. H,O. Insol. 0 for Cl). 35-12 14.10 0.08 16.08 36.99 Nil 0.72 0.22 200.15 This agrees with the formula MgCO,,Na,CO,,NaCl. Cold water acts slowly on the mineral, but hot water decomposes it rapidly with separation of magnesium carbonate.It does not decompose on exposure to air. Fusible at 1 (Kobell's scale), with evolution of car- bonic anhydride, and leaving an alkaline mass. [For artificial northupite see Abstr., 1896, ii, 6101. Pirssonite.-This new mineral was found by C. €€. Northup as isolated crystals in the same boring as the northupite. The crystals are orthorhombic and hemimorphic; a : b : c = 0.56615 : 1 : 0.3019. The form e(131) is present a t one end only of the crystals, this being the pyroelectric analogous pole. The mineral is colourless to white, but is often darkened by impurities ; it is brittle, and does not show any cleavage. The optical characters are given in detail; with an increase in temperature, there is a small decrease in the value of the optic axial angle.The mean of two analyses is CO,. CaO. Na,O. K,O. H,O. A1,0,, &c. SiO,. Total. 36.07 23.38 255'0 0.15 14.73 0.13 0.29 100.45 Formula, CaC0,,Na2C0,,2H,0 ; this is like gaplussite, which has, however, 5H,O; the two minerals are also somewhat similar in appearance. Practically all the water is expelled below 150." Before the blowpipe, the mineral decrepitates and fuses at 2-2.5 (Kobell's scale) to an akaline mass. The name is given after L. V. Pirsson. Sp .gr. 2.352 ; H. 5-3-'5.MINERALOGICAL CHEMISTRY. 49 Gciylussite. - Crystallographic and complete optical determinations have been made on very pure crystals. Sp. gr. 1.392. HcmLksite.-The refractive indices for sodium light are given as o = 1.4807, E = 1.4614. The usually accepted formula for this mineral is 4Na,SO,,Na,CO, ; and the potassium and chlorine shown in the two previous analyses have been considered as impurities (salt and sylvite).The author, however, finds that chlorine is present in all crystals, and that microscopical examination shows the presence of only slight impurities. The following analyses have been made on pure material ; I, on tabular crystals, and 11, on prismatic crystals. SO,. CO,. Na,O. C1. K. Insol. Sp. gr. IT. 45.78 5.63 43.61 2.28 2.39 0.12 2.545 I. 45.93 5.65 43.35 2.21 2.48 0.19 2.567-2.553 The close agreement of both these with previous analyses indicates that chlorine and potassium are not accidental. The formula is now given as 9Na2S0,, 2Na2C0,, KC.1. Potassium and sodium do not seem to be isomorphous here, for there is just enough potassium to form potassium chloride.Few minerals contain three acid radicles, as tJhis does. L. J. S. Mysorine (Anhydrous Copper Carbonate). By Lours FRANCHET (Bull. SOC. Philomccthiqwe, Pccvis, 1894, [S], 8, 61-65).-Thomson’s mysorine from Mysore, although no water is shown in his analysis, has usually been considered to be an impure malachite. I n the present paper, material which probably came from the Urals is described. It is amorphous and compact, of a greenish-grey colour, and contains specks of malachite and iron oxide ; fracture conchoidal ; lustre slightly resinous ; sp. gr. 4.398, H. 5.5. Excepting the sp. gr., these characters agree completely with those given by Thomson. The material dissolves in hydrochloric acid with effervescence and separation of sulphur and iron oxide.Analysis gave CUO CUO (combined (combined Loss on CO,. with CO,). with S). 8. A1,0,. Fe,O,. SiO,. ignition. Total. 15.73 56-50 15.67 3.95 2.28 4.90 0.09 0.56 99.68. This corresponds with 72.23 per cent. of copper carbonate and 19.62 per cent, of copper sulphide. The loss on ignition probably represents the water contained in the small amount of malachite present. Beudant has stated that when malachite is gently heated it loses water and leaves a product having the characters of mysorine; the author has, however, been only able to obtain friable, black copper oxide, as the water and carbonic anhydride are expelled together. Mysorine appears to be a definite mineral species quite distinct from malachite.L. J. S. A Green Mineral from Brisbane, Queensland. By HENRY G. STOKES (Proc. Roy. Soc. Queensland, 1894, 10, 11-13} -Lining joints and cavities in the schists of Adelaide Street, Brisbane, is an amorphous, compact mineral of an apple-green to dark sea-green VOL. LXXII. ii. 450 ABSTRACTS OF CHEMICAL PAPERS. colour. It sometimes shows a mammillated surface, and is often streaked with white. Brittle ; smooth sub-conchoidal fracture. Thin splinters are sub-translucent. Hardness, 4-5; sp. gr. 2.6. On exposure to air, it decomposes to a white powder. It is infusible, and is slightly attacked by acids. Analysis by E. Hall gave P2OP A1,0,. V. H20 (loss on ignition). Total. 48.25 29-07 trace 23.6 1 100.93 This corresponds approximately with 4A1,0,,5P,05,18H,0.The colour appears t o be due to vanadium. The mineral somewhat resembles turquoise, L. J. S. Wardite, a new hydrous basic Aluminium Phosphate. By JOHN M. DAVISON (Amev. J. Xci., 1S96, [a], 2, 154-155).-0ccasionally, the decomposition of the variscite from Utah (Abstr., 1894, ii, 321) has given rise to cavities in the nodules ; wardite incrusts these cavities. It is light-green or bluish-green, with a vitreous lustre, and is con- cretionary, sometimes approaching to oolitic, in structure. Sp. gr. 2.77; H. about 5, being a little harder than variscite. Analysis gave P,O,. FeO. CuO. MgO. Na,O. K,O. H20. A1,0, [diff.]. 34.46 0.76 0.04 2.40 5.98 0.24 17-87 [38*25] Formula : P,05, 2A1,0,, 4H,O = Al,(OH),PO, + $-H20 ; or, perhaps, AlNaPO,, Al(OH), + &H,O. The following series is made out.Peganite .................... AI,(OH),PO, + l$H,O Wardite ..................... Al,(OH),YO, + &H,O Turquoise ..................... Al,(OH),PO, + 1 H,O Before the blowpipe, the mineral swells up and turns white. It is only partially decomposed by acids; hot aqua regia leaves 20 per cent. undissolved, this is, however, soluble after being ignited. When first ignited, about 11 per cent. is insoluble in acids. The name is given after H. A. Ward. Analyses of the variscite agree with that of Packard (Abstr., 1894, ii, 321), but, contrary to his statement, this mineral is completely soluble in aqua regia and in sulphuric acid. At 100-130" it gives off 22.22 per cent. of water, and on ignition 0.50 per cent. more. A pure white variscite from this locality is mentioned, the usual colour being green.L. J. S. By CHARLES PALACHE (Amw. J . &i., 1896, [4], 1, 389-390).-A crystallographic description is given of crocoite from the silver-bearing lead deposits of Adelaide mine on Mount Dundas, west coast of Tasmania. The light hyacinth-red crystals rest on a matrix of lamellar limonite ; they are prismatic in habit, and are translucent with adamantine ' lustre (compare Abstr., 1896, ii, 65'7). By FRANK W. CLARKE (Bull. U.8. Geol. Survey, 1895, No. 125,l--109. Compare Abstr., 1888,659 ; 1891, 529).--The author here summarises his views on the constitution of the Crocoite from Tasmania. L. J. S. Constitution of the Silicates.MINERALOGICAL CHEMISTRY. 51 naturally occurring silicates and their relations to one another, and in a complete and connected account of the whole series discusses each species in detail.Even when the empirical formulz have been definitely determined, which, owing to impure material, isomorphous replacements, and defective analyses, has not yet been done in very many cases, there is no method of determining molecular weights, so that attempts t o arrive a t the constitution of the silicates must be more or less speculative. Simplicity of structure is, however, t o be inferred from the occurrence of only a limited number of definite minerals, which are usually exceedingly stable salts and have been formed under special conditions. Of importance in showing the rela- tion in structure of one mineral to others is the study of its alteration products, as shown by pseudomorphs and by such experiments its those of Lemberg, Friedel and others.All the silicates are considered as salts of one or other of the follow- ing silicic acids :-Ortho-, H4Si04 ; meta-, H,SiO, ; diortho-, H,Si,O,, ; dimeta-, H,Si,O, ; tri-, H,Si,O,. Besides normal salts, there are also double salts, and basic and acid salts, and it is often impossible to dis- tinguish between these; for example, Al,SiO, may be expressed as a basic metasilicate by three different structural formulE, or as an orthosilicate by two. A large number of minerals may be most eagily interpreted as orthosilicates, and, most of these being salts of aluminium, they can be considered as substitution derivatives of the normal salt, A14(Si0& which is taken to be a fundamental molecule in this theory of the silicates.This salt, which is possibly represented by the mineral xenolite, can be written structurally in several different ways ; the one here adopted is Ali[(SiO,)i All,. The following groups are then made out, and, under the various heads, the constitution and relations of all known silicates which can be referred to mineral species are minutely discussed. Orthosilicates of aluminium.-I. The nephelite type.-Here one atom of aluminium is replaced by R‘, ; for example, nephelite, Al,(SiO,),Na, ; muscovite, Al,(SiO,),KH, ; topaz, A13(Si04)3(AlF2)3 ; andalusite, A1,(Si0,),(A10)3. This brings out the relation between topaz and andalusite, and their alteration to muscovite. 11. The garnet-biotite type.-Here the general formula is Al,( Si04)3R’6.Of the three subtypes, the first includes natrolite, Al,(SiO,),Na,H, ; the second biotite, AI,(SiO,),Mg,HK ; and the third the garnets,R’,(SiO,),R,, and theepidotegroup, R’”2(Si04)3R’2(R’’’OH). The relations between garnet, epidote and idocrase are discussed. Sodalite, Al,(Sio,),Na,(AlCl), and cancrinite, Al,(SiO,),Na,H(AlCO,), are placed in this group. 111. The felspars and scapo1ites.-There is a parallelism between these two series, and both yield muscovite and kaolin as altera- tion products. The end members of the series are expressed as : albite, A1(Si,08),Na,Al, ; anorthite, [A1(Si0,),A12Ca],Ca ; marialite, Al(Si,O,),Na,Al(AlCl) ; meionite, [A1(Si0,),A1,Ca]2Ca,0. Kaolin is given as OH*Al(SiO,),AlH,. Leucite is related to the feslpars and garnets, and is written as a pseudometasilicate, Al(Si308),( SiO,),K,AI,. IV.The normal zeolites.-These are related to nephelite and the felspars, and may be called felspathic zeolites (Abstr., 1894, ii, 459). 4-252 ABSTRACTS OF CHEMICAL PAPERS. V. The micas and ch1orites.-(Abstr., 1890, 460 ; 1892, 125, 794 ; 1893, ii, 78; 1896, ii, 37.) VI. The tourmaline group.-Four types of tourmaline are given, and are written as two A1(SiO,),R”’,(A1*BO2) groups connected by Al(BO,\NaH, where R”’ is Al, R’, or MgH; this formula explains the alteration of tourmaline to mica, Axinite and some other borosilicates are considered here. VIT. Miscellaneous species.-For staurolite is suggested [ Al(SiO,),( A lO),Fe],Fe. O~thosilicc6tes of the Dpcl Bccses.--It is here necessary in most cases to assume polymerisation; for example, the members of the humite group, as recently shown by Penfield and Howe, are derivatives of the salts Mg,(SiO,),, Mg,(SiO,),, and Mg,(Si0,)2.,4nd owing to the exist- ence of trimerite, Be,Mn,Ca( Si0,)31 phenakite is written with the triple formula Be,(SiO,),. The constitution of serpentine is discussed, and is expressed as H,(MgOH)Mg,(SiO,),. The structural formulae of these are written as rings or series of rings, others are written as open chains, for example, dioptase, Cu,(SiO,),H,Cu. 04LosiZicates of Tetmcl Bc6ses.-To zircon is given the polymeric expression Zr(Si04),Zr3, and from this other zirconium silicates are derived. Auerbachite is Zr,(Si,Os)( SiO,), j and eudialyte and eucolite are mixtures of tri- and orthosilicates forming a group like the felspars.Thorium and titanium silicates find a place here. Biop.thosiZiccttes.-Although ethereal salts of H,Si,Oi are known, the metallic salts are uncertain. The typical member of the group is barysilite, Pb,,Si,07, and other lead silicates belong here. Apophyllite is given as Ca,(Si,07)3H,,(CaOH)2, and to it okenite and gyrolite are related. Cordierite seems to be the only aluminium salt. &!eta- and BinzetccsiZiccctes.-These are not easy to interpret, and in the case of the pyroxenes and amphiboles the analyses show variations from the metasilicate ratio. Wollastonite and pectolite are given as Ca,( SiO,),Ca and Ctt,(SiO,),NaH respectively, as being chemically dis- similar from the noymal pyroxenes.The fact that spodumene splits up on alteration into eucryptite and albite suggests that it is a mixed ortho- and tri-salt, to be written as A1,(Si,0s),(Si0,),Li6 ; by analogy the other pyroxenes are also written as pseudometasilicates, R”,(Si,O,)~( SiO,), ; and Tschermak’e aluminous constituent of augite is taken six times as A1,(Si0,),(Si0,),(A10~Mg)6, As the amphiboles have a less specific gravity than the pyroxenes, they may be supposed to have smaller molecules (this is opposed to the usually accepted formulae CaMgSi,Os and CaMg3Si,0,, of diopside and tremolite respec- tively), and, considering them as pseudometasilicates with the bases replaceable by fourths, the formula may be written as R”,(SiO4)(Si3O8) ; glaucophane is Mg,Na,( A10),(Si30,)2. The pyroxenes are then bipoly- nierides of the amphiboles, and the character of the structure is the same for both groups. Petalite, AlLi(Si,O,),, and milarite, HKCa,Al,(Si,O,),, are dimeta- silicates, and so are mordenite and ptilolite (Abstr., 1893, ii, 77).3;. J. S,MTN ERA1;OGICAL CHEMISTRY. 53 Microcline from the Spessart. By E. PHILIPPI (Ber. Senckenb. Ges. Fmmkfurt, 1896, 125--133).-1n the gneiss of the Spessart Moun- t aim near Dltmm and Aschaffenburg are pegmat,ite veins containing pale flesh-red felspar. The cleavage angle, 001 : 010, of this felspar varies from 90" 1' t o 90" 11' ; macroscopically, no albite lamelh are seen. As shown by micro-chemical reactions, the small amount of soda varies in quantity, and, for the following analysis, a specimen containing ap- parently the average amount was selected.Loss on SiO,. A1,03. Fe,O,. CaO MgO. K,O. Na,O. ignition. Total. 63.84 19.74 0.03 0.21 0.06 13.42 1.82 0.39 99.51. Sp. gr. 2.562. The amount of calcium corresponds with 1.10 per cent. of anorthite, and the sodium with 15.75 per cent, of albite. After de- ducting this plagioclase material, and calculating the remaining silica, alumina, and potash to 100, the composition of the potash felspar is SiO, 63.84, A1,0, 19.85, K,O 16.31 ; here there is an excess of 1-42 per cent. of alumina over the theoretical amount, which indicates commencing alteration, a fact also shown by the microscopical examination. The angle of optical extinction on b(O1O) is 5-7", as in orthoclase ; on c(OOl), i t varies froml2" to 16". The mineral is a microcline-perthite, with albite lame112 usually in one direction only ; the microscopical characters of these lamellae are described in detail.L. J. S. Rocks and Asbestos from Corsica. By DI. OELS (Juhb. f. illin., 1896, i, Ref., 46 ; from 1naug.-Diss. Edangen, 1894).--Actinolite- asbestos from Liiri and Morosaglia, Corsica, gave the following results on analysis. SiO,. Al,O,. Fe,Oe FeO. MgO. CeO. H,O. Total. Sp. gr. 55.65 1.73 2.45 0.13 23.56 14.64 0.98 99-14 2.99 56-84 2.64 1.05 0.36 24.05 14.18 1.89 101.01 3.09 Several analyses of serpentine, as well as of some other rocks, are given. L. J. S. Pyrophyllite from the Urals. By FRANZ LOEWINSON-LE.SSING (Verh. ~uss.-rE. n ~ i 7 ~ . Ges., 1895, [a], 33, 283-287).-The pyrophyllite occurring in radial aggregates at Yychminsk, near Beresovsk, is shown t o consist of a mechanical mixture of two minerals which may be separated by means of Thoulet's solution.The heavier portion gave analysis I, this corresponding with 3H20,3A1,0,,1 IXiO,, and so being near to the formula usually given for pyrophyllite, namely, H20,A1,0,,4Si0, ; I the lighter portion gave analysis II., this corresponding with 3R0,8H,O,4A1,0,,9SiO,, and is therefore distinct from pyrophyllite,54 ABSTRACTS OF CHEMICAL PAPERS. which it closely resembles in appearance ; the name psezcdopy.ophyllite is given to this lighter mineral. Loss on SiO,. Al,03. Fe,O,. FeO. MgO. CaO. ignition. Total. Sp. gr. I. 64.94 29.22 - - 0.16 0-67 5.90 100.89 2-782 11. 43-68 32.60 3.13 0.30 9.08 0.59 11.52 100.90 2.687 Hermann’s analysis of the Pychminsk mineral agrees fairly closely with a mixture of three parts of pyrophyllite and one of pseudo- pyrophyllite. The pseudopyrophyllite is harder, and greener in colour than the pyrophyllite; both are orthorhombic, with a very perfect basal cleavage ; the acute optical bisectrix is perpendicular to the cleavage, and the dispersion of the axes is feeble with p > 21.Pseudo- pyrophyllite differs from pyrophyllite in having a larger optic axial angle, and in being optically positive. The microscopical examination points to a variability in the composition of pseudopyrophyllite. L. J. S. A Crystal of Labradorite from Gabbro. By N. H. WINCHELL (Bull. Nus. Bist. ATctt. (Payis), 1896, 160--161).--Blocks of gabbro from Bearev Bay, Minnesota (N.E.shore of Lake Superior), contain large crystals of labradorite ; one of these crystals is described. Optical extinction angle on 6 (010) 25-27’, on c (001) 7-11’; sp. gr. 2.72. Analysis gave SiO,. Al,03. Fe,O,. FeO. CaO. MgO. K20. Na20. Total. 50.75 32.80 0.22 - 13.69 0.04 0.12 2.60 100.22. This is between Ab, An2 and Ah, An3, and corresponds with labradorite- bytownite. L. J. s. Alteration of Glauconite. By K. D. GLINKA (Annzcnis*e GZoZ. Min. de Zcc Russie, 1896, 1, Mem. 1--3).-GUmbel has stated that glauconite alters to limonite, and Koudriavtseff gives silica as an alteration product of this mineral. It is pointed out that mineral waters from glauconite rocks contain iron and potassium, and the hydrated iron oxides which cement these rocks have probably been derived from the glauconite.Analyses of the altered mineral shorn that with a decrease in the iron and potassium there is an increase in the aluminium, the end product then being a ferruginous clay. Reasons are given for doubting Koudriavtseff’s conclusion that silica is an end product, Artificial Production of Ro’cks by Fusion in the Presence of Various Agents, By CORNELIUS DOELTER (Jahrb. f. Min., 1896, i, 211-212). K. Schmutz, under the direction of the author, has fused natural rocks in the presence of various agents; the fused products consist of a glassy base containing the minerals mentioned below. When eclogite was fused with calcium and sodium fluorides, the re- sulting glass contained meionite and plagioclase. Leucitite with cal- cium chloride gave orthoclase ; with sodium fluoride and potassium silicofluoride, it gave scapolite, mica and magnetite ; and with sodium chloride the same rock gave augite, scapolite and magnetite.Granite L. J. 5.MINERALOGICAL CHEMISTRY. 55 with sodium chloride and potassium tungstate gave plagioclase, augite, and tridymite; in the presence of other agents, olivine, augite or scapolite accompanied the plagioclase. Nephelite-basalt with cal- cium fluoride, sodium chloride and boric acid gave magnetite, anorthite and augite. A chlorite-schist containing pyrites when fused with sodium and aluminium chlorides gave oligoclase, mica, pyrites, and haiiyne. L. J. S. Analcite Diabase from California. By HAROLD W. FAIRBANKS. (Ja1wb.f. Min., 1896, ii, Ref., 89-90 ; from Bull.Bept. Geol. Univ. Ccdi- fmh, 1895, 1, 273-300).-1n San Luis, Obispo Co., California, are three isolated occurrences of analcite-diabase in large dykes; these rocks are interesting on account of their relations to teschenite and theralite. The rock of the most important of these, the Cuynmas dyke, is much decomposed, and contains, ifi order of formation, magnetite, olivine, a plagioclase near labradorite, augite, and analcite ; the structure is granular. Intersecting the main dyke are numerous smaller dykes, in which the rock is fresh and contains the same minerals as before, with the exception of olivine; here the structure is panidiomorphic. The analcite occurs under four different conditions : (1) lining cavities, (2) filling angular spaces between the felspar crystals, (3) replacing felspars, (4) in one of the dykes in the form of hexagonal or rounded grains partly enclosed within the felspars ; it is considered to be an alteration product of nephelite.Secondary felspar and prehnite re- place analcite. Analysis by V. Lenher of the rock from one of the secondary dykes gave- SiO,. A1,0,. Fe,O,. FeO. CaO. MgO. K,O. Na,O. 50.55 20.48 2.66 4.02 7.30 4.24 2-27 8.37 H,O. C1. Total. 0.44 trace 100.33 L. J. S. Comendite, a new Rhyolite. By S. BEETOLIO (Jc~hrb. f. B i ~ z , , 1896, ii, Ref., 76-77 ; from Rend. Accccd. Lincei, 1895 [5], 4, Sem. 2, 48--50).-The island of S. Pietro, off the west coast of Sardinia, con- sists of very acid volcanic rocks; these are lipazites rich in alkalis, and containing much quartz and chalcedony.Basic minerals are subordinate; biotite and hornblende occur in some of the rocks, bnt more usual is a pyroxene resembling sgyrite, which shows the pleo- chroism c bright yellow, b green-yellow, u leaf-green. Liparites with an alkali pyroxene form a new group, to which the name comendite, from the locality Comende, is given. The rock also contains an am- phibole resembling arfvedsonite, with pleochroism c brown, a greenish- yellow. In a previous paper (Boll. Conz. Geol. Itccl., 1894, 25, 407), the author gives the following analysis of this rock, SiO,. Al,O,,Fe,O,. CaO. 1790. E20. Na,O. Total. 80.3 9.2 trace 0.6 3-9 5.5 99.5. L. J. S. Green Slate from Llanberis. By JOHN H. COSTE (Bev., 1896, 29, 2450-245 I).-Green slate from the Dinornwig quarry near Llanberis gave the following results on analysis : the cornpasition is56 ABSTRACTS OF CHEMICAL PAPERS.similar to that of a clayey soil. hydrochloric acid gave I, and the insoluble portion 11. The portion soluble in strong SiO,. Al,O,. Fe,O,. FeO. CaO. Illgo. Kot det. I. - 4.79 1.52 3-96 0.22 2.14 - II. '77.37 18.68 1.60 - trace - 3.35 Na,O, K20. P,O,. Tnsol. ignition. Total. I. 0.13 0.41 85.06 1.34 99.57 The presence of phosphoric acid is noteworthy. Seminormal hydro- chloric acid acts appreciably on the rock, and extracts 0.089 per cent, of phosphoric acid. Sp. gr. of the rock = 2.S18. By JOHN E. WOLFF (Arne?.. J . S'ci., 1896, [4], 1, 271-272).-This rock forms a large intrusion in Tertiary limestone on the Atlantic slope of the Costa Rica volcanic plateau. It is a dark grey spotted rock, and shows distinct crystals of augite and biotite, and rounded areas of radiated zeolites. The microscope shows : augite, plagioclase (labradorite), sanidine, nephelite, a mineral of the sodalite group, olivine, biotite, magnetite, apatite, and analcite and other zeolites.The main differences from the Montana type are the absence of sgyrite, the distinctly basic plagioclase, and the small amount of sanidine ; it is also probably poorer in alkalis. This is the nearest approach yet described to the theralite of Rosenbusch, namely, a plutonic plagioclase-nephelite rock. L. J. 8. Altered Vesuvian Lava. By FRANZ LOEWINSON-LESSING (Annuai./.e G601. Min. de In Russie, 1896, 1, Section 11, 10-11 ; from C. R. SOC. Naturalistes St. Petersbouyg, 1895, No, I, 15-16).--A lava stream of 1631 forms a sea-cliff a t La Scala in the harbour of Resina; t.he long action of a warm saline solution in the form of spray has decomposed the rock t o a soft, friable mass containing porphyritic crystals of augite and leucite.Analysis of this decomposed rock gave the results under I ; this is compared with the average composition of Vesuvian lava 11. Loss on L. J. S. Theralite from Costa Rica. SiO,. AI,O,. Fe20,. FeO. CaO. MgO. Na,O. 11. 47.82 18.85 5.24 5-12 9.51 4.40 2.65 I. 48.31 L--- 24*23---- 9.54 8.24 4.83 Loss on K,O. ignition. Total. 2-77 1.73 99.65 6.41 -- 100~00 This shows that ferrous iron and potash have been replaced by magnesia and soda respectively. The material contains 1.32-1.55 per cent. of chlorine, which can be extracted by water as chlorides of sodium and magnesium.L. J. S. By E.~IIL W. COHEN (Ja1wb.f. Mim., 1896, ii, Ref., 42-43 ; from Anla. k.k. Naturhist. Hofmzcseums, Wiem, 1894, 9, 97--118).-1n No. 111. of his " Meteoric Iron Studies " the author gives analyses of several irons, and of the isolated rhabdite needles. It is shown that the acicular rhabdite and Meteoric Irons-[Rhabdite and Schreibersite].MINERALOGICAL CHEMISTRY. 57 I I 1 49'06 35'48 I1 51.60 30.89 I11 55.01 28-62, I V I 55'54 26'73 VII 1 47'22 30.16 I the platy schreiberaite have the same chemical composition, namely, (Fe,Ni,Co),P ; the two occur together and only differ in habit. Goniometric measurements of rhabdite needles showed the existence of a tetragonal prism zone. The irons examined mere :-I See-L'iisgen, Prussia ; I1 Bolson de Mapimi C = Coahuila], Mexico ; 111 Sanchez Estate, Mexico; IV Hex River Mounts, S. Africa; V Schwetz, Yrussia ; VI Rasgata, New Granada ; VII Lime Creek [ = Claiborne], Alabama.The mineralogical composition of these is as follows. 0'45 13.35 0.48 0.55 0.63 ' 100 03 0.70 1 14-63 1 0.78 - 0.96 1 99-56 0.38 I 14'16 , 0.95 1'16 I 1 I 95.57 0.60 15'24 0'43 - -- ' 100'19 0.46 15'10 I 0.25 0.30 98'91 __ _____ I Nickel iron ........................ I 98 *76 98 *344 Ni-Fe phosphides ................ 1 *21 1'615 Carbon ........................... 0.01 0.011 Chromite and silicates . . . . . . . . , 0.01 0.003 Daubreelite .................... 0.01 ' 0 087 Tznite.. ............................ i z / - - Angular fragments .............__ Insoluble.. I - - .......................... - V 98.55 0.17 0.01 o*ox 0 '89 0 '36 __ _I - 97.10 1'73 0.07 0.03 0'15 0.56 0'36 - - - VII _ - 96.96 2.91 0.01 0.09 0.03 - - - - . - The action of acid on the different irons of similar chemical composition is very variable, dissolution taking place in a few days, or only after an interval of months. Analysis of the portions soluble in dilute hydrochloric acid gave I 92.23 1 7'24 1 0'47 0 04 94.22 5 -17 0.54 I 0'06 92.47 VI 92'29 0.15 VII , 93-86 I 0'02 0'01 0.0148 0'02 0'01 Analysis of the isolated rhabdite needles gave P. 1 Cr. IS (calcd.)l Residue.1 Total. I I ! Fe. 1 Ni. I With 111, there is 0.28 per cent. of carbon ; in IV and VII are small quantities of chrornite and silicates. Analysis of schreibersite from IV gave Fe.Ni . co. P. Cr. Chromitc. 8. Total. 61.46 21.31 0.34 15.20 0.32 0.25 9.39 99.27 L. J. S.58 ABSTRACTS OF CHEMICAL PAPERS. The Meteorite of Hamblen Go., Tennessee. By GEORGE P. NERRILL (Anze~. J. Xci., 1896, [4], 2, 149--153).-Eakins’s analyses of this meteorite (Abstr., 1894, ii, 56) showed that olivine could be present in only small quantities, for in the 37.63 per cent. of the stony portion which was soluble in hydrochloric acid only 1.34 per cent. of magnesia was present ; and at the time no satisfactory con- clusion was come to as to the mineral compositio’n, beyond that the insoluble portion consisted mainly of pyroxene. Microscopical examination by the present author shows a holo- crystalline, granular ground-mass of enstatite, diallage, and anorthite, with porphyritic pyroxenes and some indeterminable material between the grains.Olivine is inconspicuous, and cannot be separated. The anorthite gave the following results on analysis. SiO,. A1,0,. FeO. CaO. MgO. Na,O. Total. 42.02 37.77 trace 16.41 0.96 not det. 97.16 The enstatite was found to contain magnesium, but no calcium or aluminium. These minerals do not, however, altogether satisfy the requirements of the analysis of the soluble portion. After separating the anorthite as far as possible, some of the material is still gelatinised by acid, and, as magnesium and calcium go into solution, the presence of monticellite is suggested. When the stony portion of the meteorite is boiled with water, chlorine, sulphuric acid, calcium and iron are extracted; this suggests the presence of gypsum (derived from oldhamite) and lawrencite. The minerals probably present are then : nickel-iron, en stati t e, diallage, anorthite, olivine (or monticelli te), oldhamite (or secondary gypsum), lawrencite, troilite, and schreibersite.The meteorite is to be classed as a mesosiderite; but the composition and structure are very variable, €or the stony and metallic portions each in turn predominate in different parts. In the coarser portions, especially when near the metallic iron, there is a strongly marked cataclastic structure, which is well shown in the anorthite. The average sp. gr. of the fragments is 4-32. [Eakins in a later note (Amer. J. Xci., 1893, 46, 482) had corrected some errors in his formula, and suggested that the soluble portion might consist of olivine and’anorthite.] By EUGEN W. HILGARD (Amer. J. Xci,, 1896, [4], 2, 100-107. Compare Abstr., 1893, ii, 165).-The presence of alkali carbonates in mineral waters is usually explained by supposing that carbonic anhydride has extracted the alkalis from silicates ; but it is better explained by the fact that alkali carbonates are formed when a solution of an alkali sulphate or chloride is treated with calcium or magnesium carbonate in the presence of carbonic anhydride. The action of alkali carbonates in soils and in hardening muds is discussed, as is their efficacy in the alteration and metamorphism of rocks, and in the formation of mineral veins. L. J. S. L. J. S. The Geologic Efficacy of Alkali Carbonate Solutions. Artesian Waters of N. S. Wales. By JOHN C. H. MINGAYE (Austrcdiccn Assoc. Adv. Sci., 1895, 8, 265--277).--Nineteen analyses are given of the water from artesian bores in the western district ofPHYSIOLOGICAL CHEN ISTRY. 59 New South Wales; the total solids, except in five cases, are small, and consist mainly of sodium and potassium carbonates, and sodium chloride, with less calcium and magnesium carbonates, alumina, silica, and organic matter, The value of these waters for irrigation and other purposes is discussed. Analyses of the soluble salts in two samples of soils are given, The deteriorating action on plants of alkali carbonates introduced by irrigation may be counteracted by gypsum. L. J. S.