Pelee rarely raletetaloneletehgietatplpinies ewer ere ere ¥ eher ere’ pie Te ee ge Rigel re Whriy-atwle we nn ve — revo Rone § anew b euros ve ieehdadel ‘ee Bee where srirsine em ae alene Gonte Foe beeen er +4 hagethah eines Hehe r one terane en on Dn ee ee a ory ie POO Te wa we 1 Pale eee eee re ry De tele bet Wide wiv td mTeheitely! swe! elererrialees ede Pe a ee Ropsaninerctonmmteact PO NGO og AONE W Pee te elele nw ete ole rarer ont “ . Jr reirt hie Houbaphed a0 Ne 00! where etre re ee PE CUNT ee lade wlelatwrete stele le wnene ewe lewlete tems 4 hp ee were prahe'e ro%y wishes v'tiel® CRU h a ee oO ie i ke kk ee han) LahditeeSebd te ae "ele vee eeeetayie A didhabtbdlobabdataele eae cite ee en dicias eibhdivte Sediti-vedl ditt PVE TNND ROEM RY whebwre te teter sith tthe haa ics died etetatieerdetiehitndl aided athaac os be ee Pe ee oe eee Oe de lett tending Midi deat pate etd Nioetunnenebiors eelevrered Ses ee grew’ rrr nett a ee ee ae oe hen lene “ Vor wcyighetee wer w eum ig alah w 7 panemapreerpieterely te lebete ‘ghee wvolee viele PCO EO Tel enae ciwle letehete ev hehe ere etelghyns niet > er Oe Gan . pig vel ety veeeere many vie fete " meats url ehas wi iprete” ra by Be wie eee wreqare ee ererere ST re ete ree © wrere o oie aie! te tethers ip ere. brent pent tad AEE A OF Te 16 Oe VIE Om nia wher Se lalele ereiniele ere /nwivleher ae mirrenabans sora tehe tookalerny ew arey re one oberenencareTerss dbiedhdedtied aad oleate Spvsnes © wrwre shel ve elie were, Petetere'r wie oe v ’ a erelieaneeitescictctoeraes - re ves ed ee a heh elen et ee eterna whale ih dhdbdbah dba dicot Harte tipi bb mp go di vialot wt Sleeteateits wierwig wrote nr baronet ‘ne siahreiate epee ele wh wiprelens of . wel vier wears le ele ee oe tatetuteinielet iW ipiwietere rae eee rithteericute toe eae cheb weteteteteivialee heiytateratalgtytyieveleneie nee tecwt ein rein werety rere ee ane rane rirtare wale’ a BP ri eaetb dar tertndie ht Je ree verwlecnite ee ereleta ty Mebeewiprerviwre Cielereretejelerwgrerer ruiviat ates +e" ee ee , . el eee wa haters lw hal elwle viene eh eleven ronettngeie tere eae elated sl taeteearetitle ieee ween ee er ew ene ea nanutara rer aLdbabhedba aah M4 ottgteie® . te een Rha p aire! wie ew “ ares wravete rete vie vrrene severe "ete 20 reels coal eee le ETRE Then oe prehe wo eree etateretr rele Orr a ee ee . owoets pleat auiomtentehnonreer meee recgere"e: rai unuetna}anoeen nd 2 Mancha ieesorenersteeaatornnavonen eireontioeeemeee nebo ITTTNT Se ee as: yp tadtd- tind bvthaleg ee . PAN Maarten are Pe eee SO ee a Fe eee hvneiaa tages B70 we Tot ole wie wre ee ot a it ee a rye theres wow ee wR We whe wwe D 4 pe bh pn * ewe ree oe Ree Cem eee ee wretre hey view viewer? Fe eee 'e4 wren ntelert ey riety Cel han weet WiWeEl vinta velo wv a lethrsaethetielh Shibedareecorloa aber 7 . re wiviw Oe eit ta eet ° A fied!» " oreo reper Teel dhda dead dane here! Sarton ee ee TNT EWTN WHR eE ew ieee wre winleyalately lm Bi etete! werele otely we resiowyen ncn ay oreromeeny Poreneinl baie nes w! “Aner WP view, v0! ev ete wer “ tes wintering Sivle ore rre. ye hs anion a ° O tres ore wietele’s onions CE VED oH8 ely ere eet ee ee eee ore Parety! orate weer vee ley wiete le lula wt eligi Vee seoreieew treotiiben ‘Ave wi watetess wera eh eee eee eS ee ee 2 ths Ot 1 en 1 a kt ns the 1a on on oe ‘ 5 - ewe © - tes r : pm rir cmate ‘ : is a u r : r . epebendnannend terse ears we ste dal ie hdlalidind didi htt didlo ellen bel sth dy ice dona heeled e VW NeW Ow YY noctoroier ee errs tee Cra Let ce inn ~~ J eee PET Tee we ‘ . ver anh inargrt ntesnlFn oabtetainiri iene HS Dewesey cre sneeen eee misiarote wivig lynn bieieiete eseba nin te Veerwnee . ‘ . eee i 2 ee aint sermri gs emeneiewenereiwiey ° a ¥ ¥ v noTenerelenynrianr ne isre Hiottetindl a COOH WOVE TE WE TT ele e yee CPR Rg OME mee wie HED ee Bo erent vee ewe” soa wer vre tev iP ene ree pea dla di hied oN Cae ho 2a ‘i re yerewerrie'ele eee ere OF wie Fehe we ehaly teehee eae Phew erga te welvtele Aah idalitadl feted di Datta tesih--etopintish:t-Sesire Peleretae! Bele ere Tw et ee ee oda thid a ee Viviwle a whee epee ww ey vrwlere v - sm ‘ , win ha atwtn fenenerererel wala larg eer @errNra Os MEET H Tew le ele terE ely Met Lewes wividipieleie ete leme ‘ eM eeey Saruverelian tor ne tpaghetet 1ETE OMS e-wtprerpemetetetete (es wee vw eee whetet meter ere Wig eT er ee a ee a eee ee ee | ee Md bind PalelelNnd Wvlererbueree' ess ewe rsan of rovnginaae phe Pret wena wie ome oe PN ONE}! UP Vipeene tate ele ere we ow vere fc Seino thei oth asi pahareerwre wr airielsinie elvin'e evry Rina Hwy ee vee ve w wiesle Ole " nevoperovere’ ie ee eae ee ee ree Ra RETR FEE ® weeeueermenee lve roecyen tee gsah wie winde-te tots ipitetetet : Sabre cee ee cre tere ete CReIe OFwrE ETN RTE WWE RI BKYTS Here owinierqiylyiaG oben samira 107 casvanes ni baie meh he APneaD rath ts thane dnen ah sae Pereira eins airs. bnty soda Svat belies ee pomeeleaieaeargre geen ver suerauarenvartetveviotetetyna efer=-strse fs ovwteie ae Tete ete ele oe ree f Nerve ele we 4 eeeneenwt Lye wee gieleler® FEW wleheetel een er a eR ee a ee oe eee elem wea ee ee) eee ewer ew seer rat “ Z “ Ve iereel ere eerie eTEMO TITRE wieiale lyre PEW wer ofpteie iy ereneete eng vie Pe ee ead ete y Eee wie ene ht ele ie wheel teeter eles a held re eh eee oteteheapie tainty tate (ele rereietwrerereyele er rerr © wR ENP NLW Ely ly nt hergtenw Die wile sereemmy el elwiel# es Fe ‘ei orerutwipteere ’ ‘ wt wow ere vive tee erate ote owe © eer viwinrery oreo rele ‘whe elven win ge 6 a Ba ee weet etelte fore rel etietere te wily wraretvitene'y v ow ete etree Es toa-tgbncnds ont tot lr tere waporapmeet none ee ees Urn ete ee necmnpu peers ett rer cnet) wimere ee Ors or wre ewe ee Te de ee er he a ee eee Rs vv e-aretwertee-w a0 a wistela CENTRO we pipipnl gent . a Mer mcrridn tend aaa yy Sheeeeat sets : ee ee a 1k he we ’ . ~ Y wee lew IO wl 6 ee ae o'w Betecy ee! i ttf ian tetra shedete ent avcr slaesiribelicthdledath Rielbetin ‘ pleteletete: “rewe Oe ee ee ee . " ‘ tere eh * ‘ew ew leg rele Ye rpip ae wk Wletels tale ee 5 ‘ * ‘ ith Aidala didi de “e f died ' “ rere oes o$ Be lew eee weteieiy wiaeaain's ee eee eevee btbditiedh Aidit adhd abet bik Sak ak | 2 prep iwiecete wwe $09 rei enwurwnvraterve he ¢ i enn Cased tld ws apne? arerwrntetenatteleiel ~ ve Sietevere riarenpinrenenreteileterelWié sere ach srnteenan ret motrin ph Soe wens : * de eiptmewinlwlete er TR] We we eletT we ee tele atetwte pate abdpes ae) Mt 4 “ wot eles reer wrane ier dee eee” or ene urtoairirththetbd berthed 4 Titi didi itilit tietetdedir we grew ylelelerenrs ree ere es rt aah iededttadlb dite Z Cig Wererntelerer were Where Oy 'F wlele’ FW wero whe ie tolerelwiehele vewewes or ewe Tele eeln Oy Pieler Rye qierwivfereresereiviergte eo Terr erry 7 erwren wrrerrey ° Tee eae Dai de tae ee titi on tadbtiedn tadhpitteabnbnrnntbh irda) rieierere ghotalate Vee wow «ele Ve ewer GE Pe TNRINT INT We errtNNty vy whee Se ee de ee 1 Ta A he on ‘i deethestied chdedlindt 4 ssh tena ghib mbeen-mep a eee ee ee ee ee le ae hav ON ietetvioim el vee staerstererarr tare apytawiy nies Wide ere sewunnts Sve he ee oe oe ee Pee ee ee ed cmurerwrwnereretrretenie"t pv-erererytylereretarereie’ewwre owifute tailed IU"e rErarU Amn nU NON ION : etter eee oe rrelne Ce eeerreeee 09 1 Oe Were eee yp #0 fn cba whey eee rt ‘ewer rere every - rales . Vie wherrer ee tety utetarelenbie hetemieeenwete'v aha W" beteienihelel wvenwr® ca vo vvieietion~ ed eae ee ts. vreee a El a) ik Ait iid vO wow He taiere renee le eietetre Oe Seer ew Vw ewe oF a wreme we wravel elles tele eleinlelelere eee wINtel yw reese we ele lw le wlglene le a i winigiewwhereensiavoometertrn Poe ere eee eae Ce wet eie es WR eT wpe irks dirt erent eater eres we vo wrere vee ak teh tht halide ts vee vty ye Py vre'eorw way where se ye" rerew ns oF Pere pe pierre aero ewe ve PT ee OP Tt la tlh dt Bd ih . 4 he COTE Reel Oe O° Oe © For wl eels oe ele ewe tle errr Ce ee Vee ree ee eee e se Ke ewe CUR CE ee phew oerot re oh sel rn a sno Hive AS ewierery ere aera ta oe tednah dy Lac dado dedhaie indie teas betel tli Maint irl Anche rwierwewe eee t eer lorrw ble hind etal eee Ovieere Helse ere rupee srr ee al orey Se didi tee ee et em : oa ‘ , ta ae Sererere Se vee? ws dire inietird yer a ee oe eo ee oe ee ee ee we . wre ? J és wre © * ‘en y , > a _— r i . v ° . 7 aA NO natn teteate ets mann! Z poy eteretorn ¢ farvap ant es Qh ta _ } . yaa ‘ f oe ee ee ee ee a wirlee ote Vinee tee er : . oe CF ee Cd Fee Tiga PENT ae the dant et ae Verse wee ee | raw le r © te why teve re wwe ret oe ote wierey eere'e a eed owe vrw wr eee ee ee ee ee Bd (ER Oen a etel gnnmyat Fe whale wie lelele ily pele VV Pe ew els cowe mere Lidia Sik tie diedinde i. , NO Eee eH E € Fee WEE Pew we Hoe FH e RF > " ee ee ' ev 7% Pet oe a i ee Wee WIP RENIN eR ehew dale els oles eon eteiatentele es Vint: eed OT RWW OE Oe FD Bee F Wee € Ser ewIIwy wee a WROD Pw ehelele eels A ee wee Peewee CV Ure ee rere ee wee ws ewe , ‘é oe ‘ 7 ewe beuleee wes Vw ew wie eee ele ee we ad ; be Trey eee em, id piv ele opie we Hie ele a ete were ere De ee HC We Ge we OUP er oy trey Trwe WV re WE? Wrow FET Fe HER Oe Cer en esos ° rene i ow 2 ei rhretvriele owe er ee ioe ae FOF Fre o 8 pire) wre Wierelele eetE ig eRe Re SW We W'Elrrete By ries giaiae ee 8 Ee re phe PORE Opes wie ee eee T Cele D Pee wea Wee ere GH * tebe Ree ew? Pee ewer Eee OCP eer ee ree ree ee eee ere © eer eee FFF Ow ee pe . ie or tiadind nd tle vveere fs yes re sr oor Panay baeress Teena rer ewe vee . >» oer . ‘ ’ t vevree ‘ , e. : u . er were eee e Tere ee we single P'0'O OV GF 07018 O sw batiotee es aiprontotnrcn ene. V8 Pe OT ee PT OTe owervew ‘wanes ' i fi Pere eer] © ony - Vakelete vd ad dle Mata tiled teh i-th dig vee “ee e% : . ° ‘ le nA wt nly Sein ; aeprere rete bee wbee eretereiavew * * . a eT oe eee CTE eT © Wwe ol Rle se 6 e eater rs eve ty ree tr fe dul COEUR ERUT ET UR OownEe ,rrerer . . £ ee od ; Noes © » mes = * ‘ie ¥ : pew ee recta s UW eeve i dadad Aetadictths de thddntedi tte dade 2 ed - 9 baa tr ate SP A tw itbdlieli ted. eviews rrr ee ede peer Clee Oe ew rien Fr) © ‘ Ore P EE ERe PWR eal: O'Vawe 68 Oe Popeye a Shia dh noe th Lith doa dint gh aha . ower "reve everee yore eve@ueys rere POE ORT Oe Wr er eee e ® hot . y wei v'w'e werner vues + errr erry ee ee eee e's oe fe Ow © Tiere verReT viewer i . - —s n re 3 ‘ " . sll . -—- ~~ een A sin lat * ek hk OR te lng Te er St ee eS ot ee a atta, = PtP BAL 4 ee Ves 6 NA - 4 qr ~ tue > PL —_ oe =a Vw, wie oe ae j - iW del ~ ; \ ¢ & — a te * NA, eles ay 4 2 ide ; ry rte ; Ot wr, athe! Th tt y U ' h | " be v She ve : s* “aay nf The Ptr th eM a! | Li nt | M7 mes Sees ad re UCL apne Jobe pgest | || TTT w ‘ ame Tee ] f . j hae be + | rl” : 4 1 TAYE OPER - gé a Wg qurees eu. rp BLveyere por aan yd parr w'w ai ; vid reg! j Ne _[~ WL A ae yew 4 leh tele tT We ee. Pr wants TEL Athy or oy mannan y es ae 2) ong ht cage wAK ap q ro ke] og > 1 ira. HTT Ca akant® bales antec vl La aw iiNet caaauliet AS p warts WAY . Na had weasel wy eee ted of ol CSee : | ag THEE TL awa I . Aways : | wr ka oe \Aw wee “by habe IL alt eceg y UO: ee lye WN) | pr dight ee a WE een wrt ot; JUL) jd Wer oy bye “geo “We a el, Tim 6 Ss Ge Oa MD ws : Nh hated AA CTS SU yy dS aN CeCe) G ti “ow MAA ee wv ret AY, nhc . He . ae aMiyievrenrvever By they be } : CSS UTS eer al UL) Pigal eT ee Ut eee ecceeee terse Cos aca Sr Oy i Sy w! rab TS, b UA) bok Aa WN TT 1 partion | stbage' we “tan wee Coors Ty M he | oa: MATT TT : 2 | 7] rer rT 1) | Te gf eee) | Bierce oy VIEOTELY Ma gt as tat \ oi — 4 Ntvwy } % y oa ie | PC LTE oe Dayty Suet Oe ltee) Yee, & aS ra Vey twee Sree sue Wa gil Bec ext Prva Miva 8 Wee st soy a Sidi uy Dy eee rou t 3 ' he) bh i Ms rade? ' scye AT Peel We Bs Ss MAT riot ded sat meee ffiin HK ngs Maeve! tity PRC EDEN CS TTT ha, beanie ‘gl Lae Pn oe es ae ! ‘ ~G* -o™ wae er. gu? ee | | Neng? ne RA) Gren “ieee Mao oyyennntl rYUy ily WLLL |S we OL oa em Wye Op oct, Wigg Se aauecce aprceesareseseee erway, QIPONN TT! eeiVice vey as We Waly By* oy » 0 W Vocy bal | hi Ne Wath) OD ly A Sepa ty. wont *. LL 1] ity dd TES dn Wisoeuy. “ Ce r tt vivy wy Hey \ i. Rana Ny Wey \ 4 Ne ‘ £C : ae bt ‘ Ty ia meng shape’ a. , phe La ee , aah | ™: - vib" { | | seen ve Nis Wan by” vee "Wires SAB WN An ae a 1S Bes! On! are i} jd max. 4 “te & lee m ye | aren i] og “bak | Ng ‘"y a “Wiese ek aes ce thAs cee” bee Teo Shay ct we ie Se Ll Tal Oa wget Ogee Gra rm ve OW Ne itty 7 geo gn" i ‘ f a 6 Ty i q i ; ved be ty jaar Ye ae, WA De Vdv pha Wr Ta Mite hy meeitl yo pe ai ot. Bi ii aft gh) hha shhh E Al, f { wr) JS 1b Tata Gig hl” Shall aitial ne weTe ._ ee Pe Mi intact PE wyevivwabuut he Denon Ee a TRPASF ~~ (? oi COROT hg 5 OF 4 AMERICAN JOURNAL OF SCIENCE. Established by BENJAMIN SILLIMAN in 1818. EDITORS JAMES D. anp EDWARD 8. DANA. ASSOCIATE EDITORS Prorsssors JOSIAH P. COOKE, GEORGE L. GOODALE anp JOHN TROWBRIDGE, or Campriper. Prorussors H. A. NEWTON anv A. E. VERRILL, oF New Haven, } Prorrssorn GEORGE F. BARKER, or Paimapeeuta. = Bg od by THIRD SERIES. VOL. XLII.—[WHOLE NUMBER, CXLIL] Nos. 24%7—252. Ly TO DECEMBER, \1391. WITH XVI PLATES, NEW HAVEN, CONN.: J. D. & EB. 8. DANA. se ge A : A] nai ai PRESS OF TUTTLE, MOREHOUSE & TAYLOR, NEW HAVEN, CONN. CONTENTS OF VOLUME XLII. Number 247. Art. I.—The Solar Corona, an instance of the Newtonian eo Potential Function in the case of Repulsion; by F. H. (2 ELT fers eR Ss Sai ihe 7 9 lg ing iene te n= a 1 II.—Newtonite and Rectorite—two new minerals of the Kaolinite Group; by R. N. Brackerrand J. F.Witiiams 11 IlI.—Intensity of Sound—II. The Energy used by Organ Peer ey Cusnries K. Wiran 2.5 ..2.¢5...2-55 -+ PORE IV.—New Analyses of Astrophyllite and Tscheffkinite; by DRE ete Se reo RN BRIT I BO See 34 V.—Minerals in hollow Spherulites of Rhyolite from Glade Creek, Wyoming; by J. P. Ippines and 8. L. PENFIELD 39 Vi1.—Bernardinite: Isit a Mineral ora Fungus?; by Joszru emma Ey EW Bia) Fe Ey on oe 46 ViL—Development of Bilobites; by Cuartes E. BEEcHER. LE TELE St ingest pc 51 VIIL--Gmelinite from Nova Scotia; by Lovis V. Prrsson. 57 1X.—Analyses of Kamacite, Tzenite and Plessite from the Welland Meteoric Iron; by Jonn M. Davison --_----- 64 SCIENTIFIC INTELLIGENCE. Chemistry and Physics—Speed of the Explosive wave in Solid and Liquid Bodies, BERTHELOT: Relation between the Electrical Energy and the Chemical Energy | in Voltaic cells, LEvAy, 66.—Action of Heat on Carbon Monoxide, BERTHELOT: Electro-metallurgy of Aluminum, MINeEt, 67.—Detection of metallic Mercury in cases of Poisoning, Lecco: Tetrazotic acid and its Oxy- and Di-oxy derivatives, LOssEN, 68.—Polar light and Cosmic dust, Liveiné and DEwaAR: Phosphores- cence, WIEDEMANN, 69.—Reflection and Refraction of light by thin surface layers, DRUDE, 70. Geology and Mineralogy—Annual Report of the State Geologist of New Jersey, 70.—Two belts of fossiliferous black shale in the Triassic formation of Con- necticut, Davis and Loper, 72.— Illustrations of the Fauna of the St. John Group, No. V, G. F. Marruew, 73.—Etudes des gites minéraux de la France, R. ZEILLER, 75.—Genus Sphenophyllum, J. S. NewBrerry: Annuaire Géologique Universel, 76.—Tables for the Determination of Minerals, P. Frazer: Mate- rialien zur Mineralogie Russlands, N. von KoKSCHAROW, 77. Miscellaneous Scientific Intelligence—Voleano of Kilauea: American Geological Society, 77.—International Congress of Geologists: Physical Observatory at the Smithsonian Institution, Washington, 78. Obituary—CHARLES ARAD Joy, 78. 1V CONTENTS. Number 248. Page Art. X.—Some of the features of non-voleanic Igneous Ejec- tions, as illustrated in the four “ Rocks” of the New Haven Region, West Rock, Pine Rock, Mill Rock and East Rock; by James D. Dana. (With Plates II to | VIL). so 8. eee ts 22h 2 XI.—Notes on a Reconnaissance of the Ouachita Mountain System in Indian Territory; by Roserr T. Hii. -_--- 111 XII.—The Continuity of Solid and Liquid; by Cart Barus_ 125 XIII.—Note on the Asphaltum of Utah and Colorado; by Grorcr H. Stonm__ 022) 0 ee 148 XIV.—Photographic Investigation of Solar Prominences and their Spectra; by Grorce E. Hate. (With Plate VIII.) 160 XV.—A Gold-bearing Hot Spring Deposit; by WatrTer Harvey, WEED. 20° oo )0230. 36 2 ee 166 XVI.— Appenvix—Restoration of Stegosaurus; by O. C. Marsa. (With Plate 1X.).04 .5.u U2). ee SCIENTIFIC INTELLIGENCE. Chemistry—Chemistry of the Secondary Battery, Cantor, 169.—Dead Space in Chemical Reactions, LIEBREICH: A new Reaction of Carbon monoxide, BERTHE- LOT, 170. Geology—Relations of the Eastern Sandstones of Keweenaw Point to the Lower Silurian Limestone, M. EK. Wapswortu, 170.—Expedition to Mt. St. Elias in the summer of 1890 by ISRAEL C. RUSSELL, 171.—Glacier scratches south of the “terminal Moraine” in Western Pennsylvania, P. M. FosHay and R. R. HIcE: Losses of Cape Cod by sea-encroachments, H. L. MARINDIN, 172.—Der Pelo- ponnes Versuch einer Landeskunde auf geologischer Grundlage, A. PHILIPP- SON, 173. Botany—Botanic Gardens in the Equatorial Belt and in the South Seas, 173. Miscellaneous Scientific Inielligence—Die Denudation in der Wiuste und ihre geolo- gische Bedeutung; Untersuchungen tiber die Bildung der Sedimente in den Aegyptischen Wisten, J. WALTHER, 177.—History of Volcanic Action in the area of the British Isles, A. GzIkrE: Magnetic Declination in the United States for the Epoch of 1890, C. A. Scnorr: Telescopic Work for Starlight Evenings; W. F. Dennine: Ostwald’s Klassiker der Exacten Wissenschaften, 178. ERRATUM.—Page 108, bottom line, for one and a half, read three. — =. CONTENTS. ¥ Number 249. Page Art. XVII.—Capture of Comets by Planets, especially their Capture by Jupiter; by H. A. Newron.-_-.....--.-. 188 XVIII.—Pleistocene Fluvial Planes of Western Pennsy]l- faut ocby GRAN ny RMP. 222 oe eel 200 XIX.—A method for the Determination of Antimony and its condition of Oxidation; by F. A. Goocu and H. W. EEE Tats 2 Ane) Pi sc 2 oo ee aE3 XX.—A Method for the Estimation of Chlorates; by F. A. Pemmeen rare AG GG. AIPM, oe ee ee a 220 XXI.—Dampening of Electrical Oscillations on Iron Wires; MPO LEOWHEIDEH, 2283: 2. ee tes avi be te 223 XXII.—Genesis of Iron-ores by Isomorphous and Pseudo- morphous Replacement of Limestone, etc.; by Jamzs P. _. CREME 2 WOE SSSI) he ters he ae eee pe ee 231 XXIII.— Constitution of certain Micas, Vermiculites and Chlorites; by F. W. Crarke and E. A. SCHNEIDER..-- 242 XXIV.—A Further Note on the Age of the Orange Sands; Semis ONE srs) ee mer eee Le ede 252 XXV.—Note on the Causes of the Variations of the Mag- meme Needles by Prank HH. Biemrow ._-- 2. ------ 253 AppENDIx.—X XVI.—Notice of New Vertebrate Fossils ; ewe WE Nera es SS ee ee ed a 265 SCIENTIFIC INTELLIGENCE. Chemistry—Boron tri-iodide, MoIssaN, 256.—Hydrazine hydrate and the com- pounds of Diammonium with the Halogens, Curtius and Scuuuz: Synthesis of Indigo-carmine, HEYMANN, 257.—Lecons sur les Métaux, DiTtTs, 258. Geology and Natural History—Composition of the Till or Bowlder-Clay, W. G Crossy: Geology of the Rocky Mountain Region in Canada, Dr. G. M. Daw- SON: Greenstone Schist areas of Michigan, G. H. Winiiams, 259.—Some Bo- tanic Gardens in the Equatorial Belt and in the South Seas, 260. Issued August 17. vl CONTENTS. Number 2a: Arr. XXVII.—Some of the Possibilities of Economic Bot- any; by G. L. Goopama 22.) 22pm eee 271 XXVIII.—Vitality of some Annual Plants; by T. Horm. With Plate X...222..3/50241. ) 2302 304 XXIX.—Method for the Separation of Antimony from Arsenic by the Simultaneous Action of Hydrochloric and Hydriodic Acids; by F. A. Goocu and E. W. Danner 308 XX X.—Notes on Allotropic Silver; by M. C. Lea __.-._-- 312 XX XI.—Structural Geology of Steep Rock Lake, Ontario; by H.-L. Suyrn. With Plate X1 .-... 2) oa 317 XX XII.—So-called Amber of Cedar Lake, North Saskatch- ewan, Canada; by B. J. Harrineron, McGill College, Montréal. <2 soles oe ke ee ae XX XIII.—Geological Horizons as determined by Vertebrate © Fossils; by O. C. Marsu. With Plate XU _-232eae 336 SCIENTIFIC INTELLIGENCKH. Chemistry and Physics—Absorption Spectrum of Liquid Oxygen, OLSZEWSKI, 338. —Production of Ozone in Rapid Combustion, Inosvay, 339.—Sulphuryl Per- oxide, TRAUBE, 340.—Dictionary of Applied Chemistry, THoRPE: Measurement of time of Rotation, PRytz: Method of determining Specific Heat by means of the Electrical Current, PFAUNDLER, 341.—Optical relation of Organic Dyes, VoGcEeL: Maxim’s Flying Machine: Small Electrometers, Boys: Influence of brightness upon phenomena of interference of light, EBERT, 342.—Thought transference, LODGE, 343. Geology—Fifth Triennial Meeting of the International Congress of Geologists, 343.—Geological Society of America: United States Association of Govern- ment Geologists, 344.—Fauna of the Lower Cambrian or Olenellus Zone, C. D. Watcort, 345.—Relation of secular Rock-disintegration to certain transitional crystalline schists, R. PUMPELLY, 346.—Greylock Synclinorium, T. N. DALE: Report on the Arkansas Geological Survey for 1888, J. C. BRANNER: Tungsten minerals in Canada, W. F. FERRIER, 347. Botany—Some Museums and Botanical Gardens in the Equatorial Belt and in the South Seas, 347. Miscellaneous Scientific Intelligence—American Association for the Advancement of Science, 353.—British Association, 358. Obituary—WiILLIAM FERREL, 358. Ee — lO CONTENTS. Vil Marmber 951. Arr. XXXIV.—The Solution of Vulcanized India Rubber ; Seemerteay Psiouiel es te ye Le bee ie eek 359 XXXV.—Report of the Examination by means of the Microscope of Specimens of Infusorial Earths of the Pacific Coast of the United States; by A. M. Epwarps 369 XXXVI.—The Tonganoxie Meteorite; by E. H. 8. Barney. Page et ee kd ete | ee ae 2 Set Se 385 XXXVII.—Proposed Form of Mercurial Barometer; by Meri y) Seen. Foe Be Eek ele Oo 387 XXXVIITI.—Color Photography by Lippmann’s Process; by Peres eG eee Fes a Ee BU Chota 388 XXXIX.—New Analyses of Uraninite; by W. F. Hitie- EE AE) 2 SN ae Ee pee nee ne eee ee 390 XL.—The Tertiary Silicified Woods of Eastern Arkansas; Pee riswORtH CARE 24.0402 oe le bt kage 394 XLI.—Occeurrence of Sulphur, Orpiment and Realgar in the Yellowstone National Park; by W. H. Weep and L. SE IESE Inala i ie site ies sa Rates do ARON Open ee mee 401 XLIL.—Mineralogical Notes; by L. V. Pirsson .--- - -- 405 XLIII.—Peridotite Dikes in the Portage Sandstones near peices oh yah NRMP ee 410 XLIV.—New Locality for Meteoric Iron with a Preliminary Notice of the Discovery of Diamonds in the Iron; by meen. toon. With Plates XIV, XVe.u2.... -2..... 413 XLV.—The South Trap Range of the Keweenawan Series; Peele, tk Ao WwOIE okt ee oo ALT XLVI.—Geological Facts noted on Grand River, Labrador ; epee ee aeons BES oh ae oe ees esis 419 SCIENTIFIC INTELLIGENCE. Chemistry and Physics—-So-called ‘‘ Black Sulphur” of Magnus, Knapp, 422.— New form of Silicon, Warren: New Alkaioid from Conium maculatum, LADENBURG and ADAM, 423.—Iron-tetracarbonyl and Nickel-tetracarbonyl, MOND and QUINCKE, 424.—Sensitive Reaction for Tartaric acid, MOHLER, 425. —Photography of the Spectrum in natural color, VoGEL: Discharge of Elec- tricity through exhausted tubes without electrodes, THomSsoN, 426.—Ratio of Electromagnetic to Electrostatic units, THOMSON and SEARLE: Expansion of Water: Experiments in Aerodynamics, LANGLEY, 27.—Chemical Analysis of Tron, BLAIR, 428.—Die Fortentwickelung der elektrischen Eisenbahn-Hinrich- tungen, KOHLFURST, 429. Geology and Mineraiogy—Report of Exploration of the Glacial Lake Agassiz in Manitoba, W. UpHam, 429.—Geological Survey of Texas, 2d Annual Report, 1890, E. T. DumBLe: Preliminary Notice of a New Yttrium-Silicate, W. E. HIDDEN, 430.—Anatase from the Arvon Slate Quarries, Va., G. H. WILLIAMS, 431.—Ilvaite, G. Co. HorrmMann: Synthese du Rubis, E. Fremy, 432.—Brief notices of some recently described minerals, 433.—Catalogue of Minerals and Synonyms, T. Eeieston, 434. Botany—-Some Museums and Botanical Gardens in the Equatorial Belt and the South Seas, 434. Miscellaneous Scientific Intelligence—Leidy Memorial Museum: Bibliotheca Zoo- logica, O. TASCHENBERG: Catalogue of Minerals, 438. vill CONTENTS. Number 252. Page Art, XLVII.—Percival’s map of the Jura-Trias trap-belts of Central Connecticut, with observations on the up- turning, or mountain-making disturbance, of the Forma- tion; by J: D. Dana: “With a map, Plate XVieseems 4390 XLVIUI.—The Detection and Determination of Potassium Spectroscopically; by F. A. Goocn and T. 8. Harr... 448 XLIX.—The Ultra-Violet Spectrum of the Solar Promi- nences; ‘by G.. KE. HAs 2) 2 2 err L.—Phonics of Auditoriums; by E. Cutrmr ._-_-------.-- 468 LI.—The Secular Variation of Latitudes; by G. C. Comstock 470 LII,—Capture of Comets by Planets, especially their Capture by Jupiter; by H. A. Newron (i) 222233223 eee eee 482 LITI.—Distribution of Titanic Oxide upon the surface of the Earth; by KE. P. DUNNINGTION .:-- 22 = 2025 491 LIV.—Notes on a Missouri Barite; by C. LurpEKine and HH, A.’ WHEELER 5/20 ¢205 5... 225252 ee LV.—The Contraction of Molten Rock; by C. Barus -_--- 498 LVI.—Notes on Michigan Minerals; by A. C. Lanz, H. F. Kerirr and Ff. FE. Saarpumss 2.2) 02°) 23 499 SCIENTIFIC INTELLIGENCE. Chemistry and Physics—Two new Modifications of Sulphur, EneGeu: Chemistry of the Carbon compounds or Organic Chemistry, VON RICHTER, 509.—System of Inorganic Chemistry, Wm. Ramsay: An Introduction to the Mathematical The- ory of Electricity and Magnetism, W. T. H. EmraGe, 510.—Chapters on Hlec- tricity, S. SHELDON: Apparent change in electrochemical equivalent of copper, VANNI: Electrolytic generation of Gas in a closed space, CHABRY: Damping of electrical oscillations, BsERKNES: Velocity of Electrical waves in solid insulators, AVONS and RUBENS, 511. Geology—British Earthquakes of 1889, C. DAvISON, 512.—Formation of Graphite in Contact-metamorphism, BEoK and Luzi, 514.—Geological Survey of Alabama, EK. A. SmitH: Geological Survey of Missouri, Bulletin No. 5, A. WINSLOW: Geological Survey of Georgia, L. W. SPENCER, 515.—Geological facts on Grand River, Labrador, A. Cary: Index to the known Fossil Insects of the World, S. H. ScuppER: Stones for Building and Decoration, G. P. MERRILL: Manga- nese, its uses, ores and deposits, R. A. F. PENROSE, Jr., 516. Botany—Botanic Gardens in the Equatorial Belt and in the South Seas, 517. Miscellaneous Scientific Intelligence—Analysis of the water of the Salt Lake, Alia- paakai, on Oahu, Hawaiian Islands, Prof. Lyons, 522.—National Academy of Sciences: The Metal Worker, A. O. KITTREDGE, 523. Obituary—J. FRANCIS WILLIAMS. as EDITORS a JAMES D. ayo EDWARD 8. DANA. “S ASSOCIATE EDITORS ee ap | | Prowsssons JOSIAH P. COOKE, GEORGE L. GOODALE Anp JOHN- TROWBRIDGE, or CAMBRIDGE. & sions H. A. NEWTON ann A. E, VERRILL, oF canta New Haven, Prorusson GEORGE F. BARKER, or Purranetputa. a: mann ae THIRD SERIES. / VOL. XLIL—[WHOLE NUMBER, CXLIL] No. 247.JULY, 1891. WITH PLATE I. . NEW HAVEN, CONN.: J.D. & ES. DANA. os 1891. ae TUTTLE, MOREHOUSE & TAYLOR, PRINTERS, 371 STATE STREET. Ff : 2 F Published monthly. Six dollars per year (postage prepaid). $6.40 to foreign sub- Tibers of countries in the Postal Union. Remittances should be made either by ney orders, registered letters, or bank checks. the lot was unpacked, we still have a grand assortment at 25c. to $7.50. -ever-desirable mineral. oe etek WONDERFUL ENGLISH | MINE RAL iY SOAS POINT Sy we THE results of our MR. ATKINSON’ S visits to ithe fede mineral localities have far exceeded our most ‘sanguine « ' tions. veh, Egremont Calcite Twins in richest SrGaon, including the fm +e group of twins ever found, also the best ‘‘hexagonal right-angled” crystal, the best of the superb phantoms (showing an enclosed scaleno- _ hedron), a superb series of twins. groups and single crystals, the ship- ment as a whole comprising a far finer and larger collection of the Egre- mont Calcites than has ever before been brought to this country. — Expected in by July 1st. : Stank Mine Calcites of endless forms and colors, including two new types. Now onsale. The lot is so large and fine that notwithstanding we sold to one customer 65 specimens, aggregating $125, the day after A New Type of Bigrigg Mine Calcites, long, sharply pointed, eer ie liant white crystals, A splendid lot. Expected in by July 1st. ae 4 Fluorite. Over 1,000 selected specimens, of every imaginable colors My te large and small, low-priced and high-priced. The lot is expected in during July, and then will be the golden opporarnty to secure this i Other English Minerals secured by Mr. Atkinson include Specular — i ; Iron and Quartz of a new and beautiful variety; Specular Iron and Dolomite, very attractive; Aragonite crystals and groups; Barites in — oS great variety : ; Witherites, singly and doubly terminated ; Bromlite,a — ie ae: splendid lot; Tetrahedrite, a number beautifully iridescent ; Connetr ite ite ; ‘Brandtite ; Henwoodite; Lettsomite; Ludlamite, Tallingite ; Bis- Sor muthinite; Langite, etc., etc. at WHAT MR. NIVEN HAS SECURED ALREADY. rae ‘From Missouri: excellent specimens of Ruby Blende; Matra he new forms; Yellow Calcite; Galenite, bright crystals, etc. i From Colorado: Minium, the finest specimens we have ever had ; Cerargyrite, rich masses and good crystals; Cerussite, some choice specimens. a. a From Arizona: A wonderful strike of magnificent cave ees : specimens at Bisbee, consisting of ‘‘ stalactites tinted green; stal- actites of distorted, curious forms; Flos-Ferri; acicular crystals of Aragonite, similar to the English, only snow-white ; a few remarkable i turquois- -color stalactites ; rhombohedral crystals of Calcites on the . stalactites.” We secured, the exclusive right to all the specimens in this new cave and have laid in nearly a ton of the most exquisite speci- — mens imaginable, Mr. Niven having devoted no less than two weeks’ time to securing and packing the specimens. The find is worthy of ans elaborate description. Azurite crystals, some of the very finest we ever had (2-inch crystals), very bright and perfect. > Velvet Malachite, a fine lot! Cuprite Crystals, very brilliant group. = OTHER RECENT ADDITIONS. Hiddenite Crystals—the best we have ever had ; Cacoxenite from Pa. ; Polycrase from S. C.; Aguilarite, the new sulpho-selenide of silver : Fm Auerlite ; Siberian Topaz, extra fine crystals; Durangite crystals; Kronnkite ; Phillipite ; Libethenite, a fine lot; Monticellite from Ar- kansas ; Pollucite, Beryllonite and Childrenite from Maine ; Roscoelite ; Sperrylite, eic., etc. z 100 page Illustrated Catalogue, 15c. ; cloth bound, 25c. GEO. L. ENGLISH & CO., Mineralogists, 733 & 735 Broadway, New York. — ~e ePID! ‘ On WALCOTT. en a Fo BF thrall LJ a SEG THE AMERICAN JOURNAL OF SCIENCE [THIRD SERIES.] Oe ArT. L—TZhe Solar Corona, an instance of the Newtonian Potential Function in the case of Repulsion ; by Professor FRANK H. BIGELow. [Read before the National Academy of Sciences, Washington, April, 1891. Com- municated to the Academy by Professor Simon N ewcomb. | THE term Newtonian Potential Function, first employed by Neumann, is now generally accepted by writers on scientific subjects. It expresses the law of the attraction or the repul- sion, as the case may be, of the action of all the material sub- stance in the universe, the discrete parts acting mutually upon oneanother. The approximate value of the attraction between any two rigid bodies may be obtained by assuming that every particle of the one body attracts every particle of the other with a force directly proportional to the product of the masses of each pair of particles, and inversely proportional to the square of the distance between their centers. The true value is the limit approached as the bodies are subdivided indefi- nitely. From this case follows the whole subject treated as the attraction of gravitation. If we substitute in the defini- tion the word repulsion we derive the expression of the second case, and many of the formule in the mathematical discussion can be interchanged between the two cases by a proper use of the plus and minus signs. Electricity and magnetism depend upon this function for their analysis. The mystery underlying the physical condition of matter is as yet insoluble either by metaphysical speculation or by scien- tific investigation, but it is significant that this Newtonian Potential Function, using both algebraic signs, should reach to all the phenomena known to us up to this time. While I Am, Jour. Sct.—Turrp Series, VoL. XLII, No. 247.—Juty, 1891. ] 2 Ff. H. Bigelow—Solar Corona. cannot justify the idea, the suspicion keeps forcing itself upon my mind that matter some how, that is to say by change of conditions or environment, can be made to pass from the posi- tive to the negative form of the function and back again. We may yet discover that this is illustrated by the sun, when we get to the bottom of its mysterious nature. At present I am concerned simply with identifying the Solar Corona with the manifestation of the Newtonian Potential Function in the ease of repulsion. : Whenever the particles of a body, not-undergoing rotation, are free to move among themselves, the body assumes a spheri- eal figure about a center. This figure is modified by the rota- tion of the body. If in connection with such a spherical body there be present other material conforming to the case of repul- sion, the body is polarized along an axis, and the lines of force are parallel to the axis of polarization within the body, become discontinuous at the surface, and on leaving the surface form curves whose locus can be expressed by the formula n—&Z gus 3 i where N is the given line (7. @) the polar codrdinates of points measured from the axis of polarization and the center. We are not now discussing either the interior or the surface condi- tions, and in the case of the potential outside the sun we may believe that we have a nearly ideal distribution, on account of the prevailing conditions of freedom of motion. In the corona streamers, as displayed by the photographs, we must remember that the curves arising from the equation just given are modified by projection, and that therefore the measured curves must all be corrected for their distortion. At the outset we could not know the position of the axis of polarization or the center of reference, and as a first approxi- mation we supposed that this axis passed through the center of the sun, and coincided with the plane perpendicular to the line of sight from the observer, which also passes through the center of the sun. Fortunately in the eclipse photographs of July 29, 1878, Jan. 1, 1889, and Dec. 22, 1889, which were studied, these conditions were not very erroneous in their assumption. These three coronas are so similar in their appearance that they are spoken of as the ‘‘ American type,” the fact being that at the epoch of the eclipse, the pole of the corona in its rotation with the sun happened to be near the plane of reference just defined, so that the coronas were observed in their most symmetrical position relatively to the earth. The amount of this angular divergence between the axis of polarization and its trace on the plane of the disk was " F.. H. Bigelow—Solar Corona. 3 less than two degrees, and this had but little influence upon the traces of the curves themselves. The second point, namely, that the center of polarization coincided with the center of the sun was more inaccurate, and this was shown by the fact that in computing the angle through which the plane containing a given ray must be turned about the axis of polarization to produce the curve as it appeared on the photograph, it was found that this angle progressed in value for points of the curve, as we passed from the surface of the sun to its extrein- ity. This angle was however checked at the following step in the computation, by which each measured point on the ray gives the polar distance 0, at which the ray under discussion springs from the surface of the sun. We propose to rediscuss this question, in a second approximation, at some future time. The upshot of the whole matter is that of all the curves that theoretically exist in space, as surrounding a polarized sphere, only such occur in the corona as spring from a belt lying generally between the parallels of coronal polar distance 25° to 40° in each hemisphere of the sun. I subjoin a summary of my result for the three coronas. Hach value of 6@,, the polar distance of the base of the ray on the solar photosphere, is the mean of generally three, sometimes four or five points measured on the ray. The mutual agreement is substantial and convincing. ANGULAR DISTANCE FROM THE CORONAL POLE OF THE BASE OF THE RAYS. Corona of July 29, 1878. Ray. N. E. N. W. a S. E. 1 29° 427 Bd SS 32° 417 24 Rae + i 2 2& 32 an. AS 30 30 al 31 3 30 52 34 40 32 25 33 44 4 31 45 33 36 30-38 Bat 218 5 32 50 ao BT a4 21 33 55 ) 33 46 Al 34 36 616 34 «8 7 Sa 53 Seer 38 58 35-11 8 4) 41 Ber eA 40 54 oa. 1S Mean 2 eed | 33 39 d4 35 33 «56 33 49 Corona of Jan. 1, 1889. Ray. N. E. NSW: SeW. S. E. 1 34° 54’ 30° 337 1° 50! de Ba! 2 28 27 ZO awe 29 49 aie 23 3 24 39 By ae, ot 3a 30 00 “: ae ae UES 207 11 a2 Lb 5 pile 13 aol 2 34. 21 Bo * 21 6 37 49 32 45 29 20 33 48 7 42 51 2S fies 36 643 34 26 8 Liat ee Beis aus, eShathe 36 52 Mean al 8 31 44 31 41 32 30 31 46 4 I. H. Bigelow—Solar Corona. Corona of Dec. 22, 1889. Ray. N.E. N.W S. W. S. E. 1 30n 2a Ss sole SY ety Rae Bile a0 2 2 as ial Pape 1S) 31. 56 3 a0 29° 0 28 14 Zllite e bil 4 heats 26 ] 26 59 Hy eee 5 30 15 33 56 299 10 32) 6 SON 34 8 B47 2b 32) 20 7 Sp lye ale!) ae ee 30) ban Joo ee 8 2S ts aC Li 36 15 2 AST 9 3D 5D 39 ay ae ke 1 oe 10 38 1 hs Ne EOD Sees Ree | Ae —— Mean 30 49 Sy 6 Sil 24 a2 2 31°35 From this we proceed to the location of the coronal poles, or the points on the surface of the sun at which the axis of polarization pierces it. | The results are independent of each other as regards differ- ent coronas, and the two hemispheres are also independent for the same corona. North Pole. South Pole. Long. Lat. Long. Lat. ily 29, WBVEh SS? Blea peasy 12h 185°. 4) gaia Jan. 1, 1889, 43 <26 84 25 174 - 29) ) Beuige Dee. 22, 1889, BO. 1 Bee Oi 1384 52 86 2 Mean latitude, 85 32 85 24 Difference in Longitude. July 29, 1878, Ou" Ace Jan. 1, 1889, 100 53 Dee. 22, 1889, 98 33 Mean difference, 100 24 The axis of polarization is therefore at the surface of the sun about 43° degrees from the axis of rotation, and the southern end of it precedes by about 100 degrees in longi- tude. These codrdinates of latitude and longitude are al- ways referred to the plane of the sun’s equator, considered as celestial, and from the ascending node of the sun’s equator on the ecliptic, that is from the point whose longitude is 74° from the vernal equinox on the plane of the ecliptic. These coordinates are therefore celestial and, being independent of solar conditions, indicate the position of the axis of polarization without any complications. A computation of the distance apart in a great circle, from the center of the sun, of the north and the south coronal poles gives us for our three coronas: o>. F.. A, Bigelow—Solar Corona. 5 BF 9! 56 Lis: 28 Mean,.174° 6’. This will enable us to compute the position of the center of polarization of the sun, which is seen to be considerably eccen- tric, and from this our second approximation begins. It should be mentioned that although the existing photographs have served our purpose, and given results more satisfactory than was anticipated, yet no pains should be spared at approaching eclipses to produce pictures of much greater power than those we now possess. For this subject already opens up a vista of great interest in studying the physical nature of the sun.* From the results that have been quoted we may draw the eonclusion that the axis of polarization seems to be fixed in the body of the sun, the difference in longitude and the distance measured on a great circle being constant, within the errors arising from the measures, for epochs extending over nearly eleven years. Since the codrdinates of position of the poles are celestial, we have only to compute the periodic time in order to know the period of the rotation of the sun at a distance of 44 degrees from the axis of figure. It is de- sirable that this should be done, because the sun-spots, from which such a period is obtained for the equatorial regions, are confined to about 35° in latitude, and we shall thus be able to pass over the intervening 50° to the neighborhood of the solar poles. I have obtained the following results: For the period from July 29, 1878 to Jan. 1, 1889, 138 revolu- tions +194°°69; the mean daily motion is 13°°1353 in longitude. For the period from July 29, 1878 to Dee. 22, 1889, 151 revolu- tions +166°°68; the mean daily motion is 13°°1312 in longitude. For the pericd from Jan. 1, 1889-to Dec. 22, 1889, 12 revolu- tions +331°°99; the mean daily motion is 13°:0876 in longitude. AAs my conditions are of equal weight for each eclipse, a least ‘square solution gives me for mean daily motion in longitude 13°-138307=788’, at latitude 85°°5. This gives for the pidenrialyPeriod,. 27041171. = 27975 9h 59™ 595. Synodic Period, 29963580, = 299%" 15" 15™ 33°, in mean solar time. I propose the following formula for the rotation period of the solar surface at different latitudes, as * A paper containing the details of the work by which these results were ob- ‘tained, will be found in the Proceedings of the Astronomical Society of the Pacific, No. 17. 6 LF. H. Bigelow—Solar Corona. derived from the mean daily motion in Jongitude given by observations of the sun-spots and by computation at the coronal pole. X = 862'—76' sind, where X is the mean daily motion in minutes and / is the solar latitude. Other formula have been given : Faye, X = 862—186 sin’ J. Tisserand, =857'°6—157°3 sin’ J. Spoerer, =1011—203 sin (41°+/). Siderial Period Latitude. Faye. Tisserand. Spoerer. Bigelow. in days. 0° 863’ R58" 878 862-0 25°0577 10 857 853 853 848°8 25°4470 20 841 839 833 836°0 25°8370 30 816 818 819 824°0 262131 40 786 793 810 Suess 26°5613 50 754 765 808 S03°38 26°8729 60 2S 740 812 796°2 27°1288 70 699 719 821 790°6 27°3206 80 688 705 837 1312 27°4386 90 677 700 858 786°0 27°4806 If we compute back to the epoch 1878-0 we find the residuals in longitude for the three coronas, North Pole. South Pole. July 29, 1878, —0°°9 +0°°5 Sane le 1889, +7 °9 +8 ‘4 Dec. 22, 1889, —7 ‘0 —8 ‘9 I adopt as the longitude for 1878-0, North Pole, 2017-2: South Pole, 301°°6. days. Siderial Period, 27°41171. We can now readily locate the position of the coronal poles at any epoch, and I have done so for several past eclipses in order that comparison may be made between a model and the pictures obtained during the totalities of the eclipses. The model was constructed in the following manner. The body of the sun is represented by a five-inch globe. In the region of the coronal zones three parallels of coronal polar distance are taken, 29°, 34°, 39°, and on these somewhat at random, are in- serted wires having the proper form. Their curvature and their inclination to the normals of the sphere were calculated from the formule, a graphic representation of the locus of the curves made for a pattern, and the wires bent accordingly. t. A. Bigelow—Solar Corona. 7 For giving the proper position to the model for any eclipse this simple device is adopted. The circular stand is placed on a sheet of card board and two concentric circles are drawn upon it surrounding the base of the stand. On the inner one the figures represent the direction from the center towards the earth, equal to the sun’s longitude of date + 180°. A mark on the stand, drawn by regarding the inclination of the axis at 7° 15’, and representing the position of the node, is placed at the reading 74° on the circle just described.. For any eclipse turn the card board with the stand upon it about so that the reading (© +180°) is between the observer and the center. Furthermore adopting the data given above for the epoch 1878-0 as the elements of predicting the position of the poles of the corona, a table has been constructed for a series of cor- onas from 1857 to 1893. On the second circle the 0° reading begins at 74° of the first circle, and it is necessary to rotate the ball so that the North pole of the corona shall point to the reading that. was computed. The observer will then see the model in the position of the corona of the sun, if the eye is placed on the same level plane as that passing through the cen- ter of the ball. The following table gives the two readings necessary for setting the model. They are computed for the Greenwich mean time of conjunction of the Sun and Moon for the several eclipses as given by the Nautical Almanac. Observations made at any other time can be readily corrected. Table of Codrdinates for setting the Model of the Corona. Date of Eclipse. | Long. Long. Date of Eclipse. Long. Long. Earth. | Corona. Earth. | Corona. 1857-23254 G. M. fe) age" 294°4° |1876°71441 G. M. T.| 355° ae 1858°68616 - | 344 67°3 1878 57741 - | - 306 84°4 186054647 33 295 3511 1880-02999 Fi ip ig | 212°4 1862-00094 < ae 128°2 1882°37T597 Y 236 305°9 1864°34503 2 es) 248°6 (1883°34747 SORTA 226 286°1 1865°31645 . en 192°5 1885°68965 3 346 14 1867°66140 5 S36 281-0 1886°66132 - 336 342°4 1868°63032 « | 325 248°8 1887-59106 "3 311 122-47 1869 60209 3 |} 3165 23074. |1889-00517 3 101 65°T 1871-94778 os / 80 322-5 1889-97616 = si 43°5 1874:29174 4. | 206 46°3 1892 32003 7 216 1269 1875°26358 os 196 30°2 1893-29187 e 207 108 8 Three cases are presented for comparison. Figures 1, 3, 5, show the model placed in three positions, corresponding to the eclipses of Jan. 1, 1889, August 29, 1886, and July 29, 1878. Figures 2, 4, 6, represent these coronas, as drawn from the photographs. 8 I. H. Bigelow—Solar Corona. Soa dy Sali ffs ‘ Feet Uses rr AWS CA ie STIS SSM I. H. Bigelow—Solar Corona. 9 In making the comparison, it is proper to bear in mind a few obvious considerations. The wires that make the rays on the model properly represent only stream lines, or portions of the streamers of the corona. The curve is true for that part of the streamer which springs from the sun at the latitude corre- sponding to the axis of the wire. Inasmuch as the coronal streamer is large at its base the curvature of the ray must agree in all its parts with the lines springing from this region. The consequence is that each ray spreads out, as it recedes from the sun, to fill all the space occupied by the bounding curves, and we see as a result the curious forms of the curves of the cor- ona, which are definite and conform to this law. I would pro- pose this as a sufficient proof of the truth of the theory, even taking it by itself. The rays set into the model do not pretend to represent the lines measured for any particular eclipse, be- cause it is designed to illustrate the subject only in a general way. One ought properly to construct a model for each eclipse using the computed (a. @), the codrdinates of the base of the ray. Then photographing this, a comparison could be made between the individual lines. The model does not show the nebulous, structureless mass of material, which was proba- bly thrown up along these coronal lines, and is going through other transformations in its return to the sun. We miss also the radiant light which passes through this coronal matter and illuminates it, for the most part in radial lines up to the region of the streamers, where it is in a sense shut off, thus producing the effect of great equatorial extension. Coronal material may accumulate along the equatorial regions for immense distances, and then the radiant light streaming through it would produce the wings of the corona. It is evident that the quadrilateral forms are made by the perspective thickening of the coronal belt as it passes round the side of the sun. The polar rays are the individual streamers seen in projection. The reproduction of the eclipse photographs is necessarily such as to diminish very much their availability as objects of comparison. This should in fact be made with the glass nega- tives. Still it is easy to infer that there is an agreement in the following respects: (1) as to the general inclination of the corona as a whole to the piane of the ecliptic; (2) as to the general distribution of the larger and the smaller sides, sup- posing that the nebulous matter is supplied to the model by the imagination; (8) as to the trend of the stream lines wherever they are seen. We do not pretend to show all the individual lines, nor all the special solar outbursts in loco, nor do we pretend to account for all the imperfections of the pho- tographs or drawings. Those which are composite, or which are halated, or which are inadequate, must take their chances. 10 fH. Bigelow—Solar Corona. This comparison shows, however, that it is not the equatorial extensions which are interesting in this connection, but chiefly the individual stream lines which can be subjected to measures. Up to this point we have not been dealing in speculations, but in legitimate scientific data and their results. There are, however, two probable conclusions so apparent that I will not abstain from. mentioning them. If we regard these coronal streamers as the paths along which the sun is throwing off a portion of its energy, and consequently along which its material substances are being transported, whatever may be their physical conditions, we have only to suppose that near the extremity of these extremes these conditions change by loss of energy, cooling, condensa- tion, and so on, so that the repulsive power is lost and the gravitation of the sun sets in to take its place. What becomes of this material that has been ejected from high latitudes at the surface of the sun into high altitudes above the equatorial belts? Obviously it must descend again; the heavier or denser vertically, and as the model shows, this will fall directly over the sun-spot regions; the lighter or more finely subdivided in the ceaseless nebulous equatorial rain, which by its increase of angular velocity accelerates the mean daily motion of the sur- face of the sun itself, at the time of its impact with it. Itisa great solar whipping top. Much more might be said to illus- trate these statements, and yet but little can be added to the model itself in enforcing this conclusion. There is some evi- dence shown in the table of the angle @,, giving the polar dis- tances of the base of the streamers for the three eclipses, that the coronal belt has a motion in latitude on the surface of the sun, those of 1889 being more than a degree nearer the poles than that of 1878. This movement in latitude is illustrated by the motion of the maximum zones of the terrestrial aurora in latitude, and might be expected in view of the periodic nature of the activity of the sun, especially in the 11-year period. This fact would point to a more considerable motion in latitude of the ends of the streamers, by reason of the curvatures, and hence of the sun spots themselves, in case they are due to material coming from such a source. It is not unlikely that we shall sometime be able to pene- trate yet deeper into the mysterious nature that is implied in this most wonderful mechanism of the sun. We may well believe that it expresses the type of the common history through which all celestial bodies have to pass, in the process of construction and cooling. The aurora is an indication of this system on the earth, the residual being the permanent terrestrial magnetism. Now that we see more clearly the ele- ments of the problem, it will be easy to construct a rigorous \ : Brackett and Williams—Newtoniie and Rectorite. 1i solution, including the eccentricity of the center of polarization, the inclination of the axis of polarization to the plane of reference, besides such corrections as may arise from refraction, or diffraction or photography. The importance of the problem will certainly justify us in trying to take good photographs of the streamers at the future eclipses. Art. I].—Newtonite and Rectorite—two new minerals of the Kaolinite Group; by R. N. BRAcKETT and J. FRANcIS WILLIAMS. [Published by permission of the State Geologist of Arkansas. ] THE object of the following paper is to briefly describe two hydrous silicates of alumina, which we have every reason to believe have not before been observed, and to call attention to the relation of these new compounds to other members of the group. As is well known and generally admitted the commonest substance of this class, kaolin, or when crystallized called kaolinite, approaches the composition represented by the for- mula Al,O,.2Si0,.2H,O, and has the percentage composition : SiO, 46°50 Al,O, 39°57 H,O 13:93 = 100. Considering half of the water basic or as water of constitu- tion and dividing the formula by two, the constitution of kao- linite may be represented as follows :* OH Si—OW YO—Al+ 4(H,0). O- As there is reason to think that all the water represented in the original formula should be regarded as water of constitu- tion, the formula would become: EO HO si_oH HO—Si, See UE Giese Np SN ge A hap or writing this in the form suggested by F. W. Clarke in his paper on the Structure of the Natural Silicates,+ the following formula is obtained : * Kaolinite is thus regarded as a derivative of normal silicic acid Si(OH),, anal- ogous to a similar compound Ai,03. 2Si0. . 3H.O mentioned by Remsen. Inor- ganic Chemistry by Ira Remsen. American Science Series—Advanced Course. New York: Henry Holt & Company, 1889, p. 576. + Bulletin of the U. S. Geological Survey, No. 60, Washington, 1890, p. 16. 12 Brackett and Williams—Newtonite and Rectorite. _OH Ail—SiO,=H, “SiO =Al. Hither of these formulas suggests the possibility of the exist- ence of other hydrous silicates of alumina closely related to kaolinite, and indeed differing from it only in the presence of a larger or smaller proportion of water, while the relation of the silica to the alumina remains constant. It is readily seen that three other hydrous silicates of alumina may be derived by eliminating one molecule, or intro- ducing respectively one and two molecules of water into the formula, and that thus the following series would be formed : Formulas. Percentage composition. 8 Al,Os SiOe H,0 (1) Al,O, : 2810, A H,O 42°52 49°99 7°49 2) MALO 2810, ‘ 2H,O 39°57 46°50 13°93 3 LO) : 28510, . 83H,O 36°98 43°47 19°55 (4) AO: : 2810, : 411.0 34°72 40°82 24°46 Of this series of four theoretically possible hydrous silicates of alumina only one, No. 2 of the series, ordinary kaolin, has been described, so far as we have been able to find in the lit- erature at our command. From many of the published analyses of halloysite, this mineral might be supposed to correspond with No. 4 of the series, but, as will be shown below, this cor- respondenee is only apparent. This series will be designated as the Haolenite Serves,* and will include the Aaolinite Group, which was first established by J. D. Dana in 1858+ under the name of the Hadlloysite Group, but was afterwards called the Kaolinite Group by the same author.t The object of forming such a series is to classify if possible the already existing members of the kaolinite group, most, if not all of which will be found to fall under kaolinite ; and at the same time to have a definite place into which to put any new minerals of this class which, like rectorite and newton- ite, may from time to time be found, and which would at present hardly be classed under kaolinite itself if their water of constitution was properly determined. It is the hope of the authors to be able in a future paper to show the true chemical composition and microscopic structure of many minerals now existing as members of the kaolinite group ; and toassign them to their proper place in the above-mentioned series, by rede- * The word series is not used here in the sense in which it is generally applied in the natural sciences, but as it is emploved in mathematics to describe a se- quence of similar terms which bear some definite relation to each other. + This Journal, I!, vol. xxvi, p. 361, 1858. + System of Mineralogy, J. D. Dana, 5th edition, 1868. Brackett and Williams—Newtonite and Rectorite. 13 termining their water of constitution under the conditions mentioned below. Since kaolin approaches the composition represented by the formula ascribed to it only when it has been dried at about 110° C., and from the facts mentioned below regarding halloy- site, we propose to consider the whole series as based upon analyses of material dried at 110° C. or thereabouts. Considering the series in this way, at least one and probably two hydrous silicates of alumina lately analyzed in the labora- tory of the Geological Survey of Arkansas fall into this series. One of these corresponds to No. 4 and the other possibly to No. 1 of the series. Newtonite. The first compound which will be described, and that which suggested the series given above, is found on Sneed’s Creek in the northern part of Newton county (16 N., 23 W., section 1), in the State of Arkansas. At this place a mineral claim was laid and a shaft opened in 1889 by Mr. W. 8. Allen of Har- rison, Ark. The rocks of the region are for the most part sandstones and shales of the Barren Coal Measures, while the opening itself seems to penetrate some of the limestones of the Lower Carboniferous series. At a depth of eight feet this form of kaolin was found imbedded in a dark gray clay, through which it is scattered in lumps which vary from a few ounces to forty pounds in weight. Iron and a little manganese are also said to occur in the opening. Samples of the material were kindly furnished the Geological Survey of Arkansas by Mr. Allen, the proprietor of the claim. f On account of its occurrence in Newton county we propose the name WVewtonite for this, the fourth member of the Kaolinite Series. Newtonite is a.pure white, soft, compact, homogeneous sub- stance, and both chemical analysis and microscopic examination show it to be a remarkably pure substance. It is infusible before the blowpipe, and when in the form of a powder it has a specific gravity of 2°37. It is only slightly attacked by boil- ing concentrated hydrochloric acid, but boiling concentrated sulphuric acid decomposes it almost completely, with a separa- tion of silica. It is also decomposed by a boiling saturated solution of caustic potash with the formation of a compound insoluble in water but easily soluble in cold dilute hydrochloric acid. (See below.) Quantitative chemical analyses of newtonite gave the follow- ing results: 14 Brackett and Williams—Newtonite and Rectorite. Je INE. SiO, .o soa e a a RCO 40°22 Adore WS iia cece 35°20 35°27 Loss on ignition ..-.---- 23°69 22°89 Be Onirn: iclad hela ee 0°21 0.21 OF pragerata nn asian mien Ge payere ba 0°3] 0°54 IVE) RL SE ee nt trace trace K.O pau 0:99 Na,O Be TLE VINEE BIT Pe 0°73 100°00 100°85 Waterat 110°-115 9 @..55. 5:53 5:44 If the impurities be disregarded and the silica, alumina and loss on ignition in analysis I ‘be recalculated to 100 per cent, and the same be done in analysis II, after first bringing the whole to 100 per cent, the following fiowres are obtained : Ta. Ila, Theory for Al,03.28102.4H,0. SiO (GL) gay 39°76 40:88 40°82 ANNO Yer ea) PO 36-01 35°85 34°79 Loss on ignition.. 24°23 2a 7 24°46 100°00 100°00 100°00 Although this compound closely resembles ordinary kaolin in its chemical properties, it shows thus a marked difference in composition, by containing for the same amount of silica and alumina double the quantity of water usually found in kaolin. That an apparent similarity exists between newtonite and halloysite when a comparison is instituted between the analysis of newtonite calculated on the material dried at 110° and the published analyses of halloysite where it is not stated whether the calculations are made on the air-dried material or that dried at the above-mentioned temperature, is shown in the following table : Newtonite. Halloysite (Indianaite.) Ta. IIa. BBE LNG: Op reat hE ae 39°76 40-882 39°35 38°90 Je et ae aaa 36°01 35°851: 36°35 37°40 Loss on ignition _ 24°23 23°267 22°90 23°60 100.00 100.00 98°60 99°90 "40 CaO 99°00 Analysis III is of a soft and IV of a hard, white variety of halloysite called indianaite.+ H. Pemberton, Jr., who made * Alkalies by difference. + Report of the Geological Survey of Indiana, 8th, ‘9th and 10th Annual Re- ports (1876-1878), p. 156. See also Sixth Annual Report (1874), p. 15. Brackett and Williams—Newtonite and Rectorite. 15 these analyses, kindly furnished the information that the c¢al- culations are made on the air-dried material, and that in analy- sis III, 8°68 per cent of the loss on ignition is given off at about 110°C. If analyses Ia and [la be calculated to the air-dried mate- rial the difference between them and the published analyses of halloysite is clearly shown, as is evident from a consideration of the following tabie : Tb. IIb. Halloysite. a Ea 36°83 37°96 39°35 2 eee eee 33°42 33°34 36°35 Loss on ignition._-.---.-.- 24°22 23°26 14°22 mewaterat 110°-115° CO. _.. 5°53 5°44 8°68 (at 100° C.) 10000 =1060°00 98°60 If it be assumed that the 8°68 per cent of water in halloysite is partly hygroscopic .and partly water of crystallization, this mineral would have the composition of kaolinite containing one molecule of water of crystallization. Judging from the newtonite analyses [b and IIb, this substance would, under like circumstances, have one molecule of water of crystalliza- tion, but would be represented by the formula A]l,O, . 2810, . 4H,O+aq, while the composition of halloysite would be ex- pressed by the formula Al,O, . 2SiO, . 2H,0+aq. Ordinary kaolin usually contains less than one per cent of loosely combined water. Hydrous silicates of alumina have, however, been analyzed in this laboratory, which have given off as much as five per cent of water at 110° C., but which differ from ordinary kaolin in no other respect, and it is probable that differences in origin and occurrence will account for these varying amounts of loosely combined water. A thin section of newtonite under the microscope when viewed only with low powers appears as a perfectly amorphous substance but when magnified to four or five hundred diam- eters 1t shows that it is entirely made up of minute rhombs or squares. The largest of these are not more than 0:005™ (sy45 of an inch) on an edge, while the smallest appear to be about half that size. Sometimes they seem to form pertect squares but in the majority of cases the acute angles have values rang- ing from 88° to 89°, as nearly as could be measured. There appear between these minute figures blank spaces where noth- ing can at first be seen, but by sinking the microscope tube somewhat, so as to focus a little lower down, an entirely new set of rhombs is discovered, while those above go out of focus. At first sight all the rhombs appear as squares and show small indistinet lines running from their corners toward the center, giving the appearance of the hopper-shaped crystals of salt. 16 Brackett and Williams—Newtonite and Rectorite. In addition to this there is a white rim about the edges which gives them the appearance of being higher than the rest of the surface. The cause of this is, however, not due to any mark- ing or relief on the surface but probably to internal reflec- tions whose origin it is hard to detect. In polarized light the rhombs extinguish sharply parallel to their diagonals, thus showing that they are faces of some anisotropic material and not, as might be supposed, sections of cubes which had been cut more or less obliquely. If these rhombs and squares are sections of rhombohedrons then one would expect to find also plane triangles correspond- ing to sections perpendicular to the principal axis. This, how- ever, is not the case and only in a very few instances have any triangular forms been found and even then they are very in- distinct and appear to be not in the upper surface of the plate but somewhat lower down. It is probable that in making sec- tions of this material the mdividual crystals are not eut, but are either rubbed away entirely, or are left undisturbed, so that what are seen under the microscope are not sections but crystal faces. By means of a selenite plate the positions of the axes of greatest and least elasticity were determined, and were found to lie respectively parallel to the shorter and longer diagonals of the rhomb. By powdering some of the material and allowing it to settle out from water, similar rhombohedral crystals were obtained. Rectorite. The second hydrous silicate of alumina, which is also to be regarded as new, is found in the Blue Mountain mining dis- trict in Marble Township, Garland county, 2 North, 19 West, section 27, about 24 miles nearly north of Hot Springs. It occurs in deposits which are very narrow near the surface but increase to the thickness of a foot or more in descending nine feet.. Several such deposits have been found. The wall rock is sandstone probably of Lower Silurian age. Specimens of this mineral have been furnished by Messrs. Ware and Arnold of Hot Springs, who are interested in developing the deposit. We propose the name /vectordte for this, the first member of the Kaolinite Series, in honor of Hon. E. W. Rector, of Hot Springs, Ark., who originated and has so unceasingly supported in the State Legislature the bills providing for the Geological Survey of Arkansas. | Rectorite, when pure, is a soft, white mineral occurring in large leaves or plates and resembling very closely in form that variety of asbestos known as “mountain leather,” and at the same time having somewhat the soapy appearance of steatite. Parts of it are often pure white, while other portions are Newtonite and Rectorite. 17 Brackett and Williams stained with hydrous oxide of iron and present a reddish-brown appearance. ‘The sheets tear apart easily and are very flexible and perfectly non-elastic. Some specimens of this mineral have been obtained through the kindness of Mr. Charles F. Brown, of Hot Springs, in which fine doubly terminated quartz erystals are imbedded. Some of the latter are at least one and a half inches in length and when surrounded by the rectorite form very beautiful and striking specimens. The hardness of rectorite is less than that of tale—say 0°5—although this is difficult to estimate exactly. When heated in the flame of a Bunsen burner it loses water and becomes brittle. It is infusi- ble before the blowpipe. Its behavior when treated with sul- phurie acid and caustic potash will be explained below. Two quantitative chemical analyses gave the following per- centage composition calculated on the material dried at 110° C.: iL VI. 0 Beads ai laa Ea ee 52°72 52°88 _ SED ee eee 36°60 35°51 a reer ree ( 0°25 0°25 BPMMetaie Shiro ns us | - 0-45 0°45 MgO + oncedetermined { 0°51 0°51 Ne Fai 5 oe ten 2 | 0°26 0:26 he ee | 2°83 2°83 Season ienition. ..<=.<.--=..-- 7°76 Li2 2S aes eae eee 101°38 100°41 ater an it0’ C........-.. , 878 8°33 If these analyses be brought to 100 per cent, then all save silica, alumina, and loss on. ignition be disregarded and the analyses again calculated to 100 per cent, the following figures result : Theoretical for Va. Via. Al1.,03 - 2Si0. 2 H.0. Si0, Jae DL ae eee 50°01] 49°99 A ee 37°69 36:96 42°52 Loss on ignition. ---- 7°99 8°03 7°49 100°00 100°00 100°00 If the calculations be made on the air-dried material the following figures are obtained : Vih. peiemrere site fli ai de Deal oe a ee 50*L8 Pea as eae, Sy Shee hs 33°72 Loss » PETES eee oe) Se ee eee 7-33 Water at “= SUSY 9 tab ag eee 8°78 100-00 Am. Jour. Sc1.—THIrD SERies, Vou. XLII, No. 247.—Jutty, 1891. 2 18 Brackett and Williams—Newtonite and Rectorite. If the water given off at 110°-115° C. be regarded mainly as water of crystallization it is evident that it corresponds to one molecule, and the compound would have the formula, Al,O, . 2810, . H,O-Faq-. Under the microscope a cleavage plate usually shows a few spots where it is evident that only one plate is included in the thickness, while the most of the section is made up of two or more plates lying one over another. In the single plate there is one comparatively distinct system of parallel lines in the direction of which a sharp extinction takes place. There is usually also a much less distinct system of lines which lie at nearly right angles to the first.* In the thicker portions of the plate two or more such pairs of line systems are often found superimposed one upon the other. In such cases the extinction parallel to either system is very indistinet. The index of refraction is low—lower than that of Canada balsam—and the peculiar structure of the plates gives to the thin section, especially when viewed without the microscope, a peculiar undulating and glistening appearance. In convergent polarized light, the simple plates show a strong double refraction, and give very beautiful biaxial interference figures. The acute bisectrix appears to stand perpendicular to the cleavage plane.t The angle between the hyperbolas varies much in size, in some cases being not more than 5°, and in others approaching nearer to 15° or 20°. The rings about the axes join each other forming ellipses so that the determina- tion of the dispersion of the axes and bisectrix is uncertain. It appears, however, as if the angle for red were greater than that for blue, p>v. Dispersion of the bisectrix appears to be wanting. The fact that in many cases two plates lie one over the other gives rise to apparent optical anomalies which are, however, only caused by this superimposition. Thus in some cases beautiful examples of what is known as the “ optical spectacles ” (Optische Brillen) may be observed. Among the inclusions of foreign material which appear in this substance may be mentioned the following: The hydrous — oxide of iron, which has already been noted, appears in small round masses or globules, which are for the most part deposited between the individual plates of which the mass is made up. Some member of the pyroxene or amphibole group has also been observed lying in the cleavage planes. These impurities occur in sufficiently large quantities to exert a decided influence over the results of the chemical * 86° and 88° have been measured. + A plate cut at right, angles to the cleavage plane seemed to show extinction parallel and perpendicular to that plane, but owing to the wavy form of the plate it was impossible to determine it accurately. Brackett and Wiliams—Newtonite and Rectorite. 19 analyses so that the discrepancy between them and the cal- enlated formula may well be ascribed to this cause. There was, however, no mineral detected which would account for the relatively large amount of alkali shown by the analysis, and it is possible that the soda should be considered as replac- ing some of the water and be brought into the formula. Fur- ther investigation will probably throw some light on this point. In view of the relatively large quantities of quartz of both macroscopic and microscopic dimensions, which have been observed intermixed with the rectorite, it may be allowable to consider the excess of silica found in the analyses as due prin- cipally to this cause. By recalculating the analysis after de- ducting just enough silica to bring that constituent down to the theoretical amount, the following percentages are obtained : Theoretical for Ve. VIe. Al,G, . 28i0.. H,0. <0 2 ee ZOE a Ot a 49°99 POpee econ 412G ; AOR ge 4969 eee 8°75 Si: ih yas gees 7:49 incall ovis 10000 = 100-00 100-00 In order to determine whether or not the soda found in the analyses really belonged to the rectorite, the following experi- ment was made. The mineral, in small flakes, was digested with concentrated hydrochloric acid for two hours on a sand bath. It was then washed and filtered, and the residue was boiled with sodic carbonate in order to remove any separated silica. The remaining substance was then washed with water, hydrochloric acid, and again with water, and was finally heated before the blast lamp. A portion of this dried and purified material was then analyzed with the following result : Var SIL 2S RR SRR he ie yee ee ee 57°10 ics NSSF RNa SO Bele ss HE a ae a a NR 40°53 ST gaan A ect data ipags a 97°63 Impurities (undetermined) -..------ .--- 2°37 Petals ire gs ies 8 S25 LOO"OO It appears from this that about half of the alkaline impuri- ties were removed, but that the silica and alumina had approx- imately the same relative values as before. If the theoretical amount of water be introduced into this analysis, and the silica be diminished as in the preceding case, the analysis then expresses very nearly the theoretical composition. 20 Brackett and Williams—Newtonite and Rectorite. Many points of similarity appear between rectorite and kao- linite, but in view of the peculiarity of the form which it assumes, and on account of its chemical composition, it is prob- able that it should be considered as a separate mineral. In confirmation of the above opinion the statements of two manufacturers of ceramics to whom specimens of rectorite were - sent for firing may be quoted. Homer Laughlin, Esq., of East Liverpool, Ohio, writes: “The sample of what you call kaolinite, sent me, was duly received, and carefully examined and tested under fire. The mineral is neither kaolin nor kaolinite, but just what it should be called I am unable to say, never in all my experience hav- ing seen any mineral of its kind. Unlike kaolin it will not dissolve* in water. It burns a white color and becomes very vitreous and strong. It cannot be finished with a smooth face or skin, but roughs up like a blotting pad. It is certainly a very interesting and curious mineral, but I can think of no use for it in ceramic manufacture unless it could, after careful experiments, be made into novel ornaments.” Messrs. Oliphant & Company of the Delaware Pottery, Trenton, New Jersey, write: ‘‘ Your sample of kaolinite came out of the kiln to-day, and would say that we are unable to make any report upon it. We do not know just what it is, therefore cannot say anything about its quality or market value.” It appears therefore from the above that its physical proper- ties when subjected to heat do not correspond to those of kaolin. Experiments were made in the laboratory on the relative solubility of newtonite and rectorite, and at the same time upon some specimens of true kaolin in the following manner : The fine powder of the various substances, was boiled with 10°¢ of concentrated sulphuric acid for five minutes, after hav- ing been digested with it for three hours on a sand bath. It was then diluted, decanted, treated with a strong solution of potassium carbonate, washed with water and hydrochloric acid, filtered and weighed. In all the cases, the results were very similar, so much so in fact that no characteristic differences could be detected. When treated with caustic potash the results were somewhat. different in the different cases. Powder from each specimen was boiled with 10° of a saturated solution of caustic potash for 20 minutes, diluted, filtered, washed and treated with dilute hydrochloric acid. The white floeculent residue which re- mained after the treatment of the powder with caustic potash * Mr. Laughlin does not mean dissolve in the chemical sense of the word, but disintegrate into a fine powder which remains partly in suspension. C. K. Wead—Intensity of Sound. 21 dissolved readily in cold dilute hydrochloric acid in all cases except that of rectorite. In order to dissolve the residue from the latter it was necessary to use much stronger acid and even then the solution was not complete. The composition of this residue has not yet been determined. From the foregoing facts and considerations, it is probable that three members out of the possible four, making up the above described series, are known, and the present status of the Kaolinite Series may therefore be concisely stated as follows: KAOLINITE SERIES Metwcctorite -) =. 2.2L. Al,0,2Si0,H,O +aq. Monoclinic (?). 2. Kaolinite and members Al,O,2Si0,2H,O Monoclinic or 0. of the Kaolinite Group Al,O,2Si0,2H,O +aq. v. Mepeeeee Es bss Al,O,2810,38H,0. 4, Newtonite__-.---__.- Al,O,28i0,4H,O-++-aq. Rhombohedral. In the ease of other hydrous silicates of alumina, as well as of magnesia and other bases, similar homologous series could be formed, which would tend toward a more systematic arrangement of the species than now exists. Chem. and Petrog. Laboratory of the Geol. Survey of Arkansas, Dec., 1890. Arr. UL—On the Intensity of Sound.—Il. The Energy used by Organ Pipes ; by CHARLES K. WEAD. | [Read in abstract at the Philadelphia meeting of the American Association, 1884. ] In a former paper* the case of a vibrating tuning fork has been considered as an important example of sounding bodies that gradually expend the store of energy originally imparted to them. We have now to consider one of the class that can store up little or no energy, viz: an organ pipe; and have therefore to determine, not the rate of loss as with the fork and piano-string, but the rate at which energy is supplied to the system from without. The experimental problem is very simple, and it seems strange that it has not been completely worked out. The literature of the subject is very slight. Lord Rayleigh,t in an oft-quoted experiment, measured the pressure and volume of air supplied to a whistle of 2740 d. v., and so found the rate of consumption of energy. Several years earlier Mr. Bo- sanquet in a very interesting and valuable papert discussed the relative amount of energy supplied to the several pipes of an * This Journal, xxvi, 177, Sept., 1883. } Phil. Mag., xliv, 1872. + Proc. Roy. Soc., xxvi, 248. 22 C. K. Wead—ILntensity of Sound. Open Diapason stop im an organ, but gave no absolute amount. He assumes as a matter of general knowledge that an organ builder furnishes a series of pipes of sensibly. equal loudness (and quality) throughout the scale; he quotes what he calls Topfer’s law, that the consumption of wind by pipes belong- ing to the same stop varies directly as the length of the pipe, and confirms it approximately by experiments ; and so he con- cludes that the amount of energy per second necessary to pro- duce sounds of equal loudness under similar conditions varies inversely as the vibration-frequency. On the other hand M. Allard* makes the assumption that the energy per second needed to maintain a sound just audible at a given distance varies directly as the vibration-frequency, and finds a satisfac- tory confirmation of his views in the experiments on the range of fog-horns made by various lighthouse boards. But the con- dition of the observer will be very different m the two cases ; so they are scarcely comparable. The experiments now to be detailed and discussed are suffi- ciently numerous and exact to disprove this alleged law of Topfer’s, so far at least as one organ is a fair sample of all. They were performed on a Hook and Hastings No. 11 Organ in the Congregational Church of Ann Arbor, Mich.; this instrument has two manuals of 58 keys each from © to a’, the great organ having 9 stops—the seven to be named in table ie a 22’ twelfth and a 3 rank mixture. The pressure of wind was very exactly 3 inches of water, and the total capacity of the bellows about 35 cubic feet; this quantity of wind would leak out in about 3 minutes. The only method of experiment available, unless one has a very large gas-meter at his disposal, is to fill the bellows and determine the time needed for the whole or any definite part of its contents to leak out; then determine similarly the time when one or more pipes are sounding. For example, 12 cu. ft. (=A) of air are used; if this leaks out in 60 seconds the leakage is A+60=‘20 cu. ft. per sec.; if when a pipe is sounding the time is 24 sec., the flow is tie A+24=°50 eu. ft. per sec., “and the pipe consumes the differ ence, that is 0°30 eu. ft. per second : if this is supplied under a pressure of 3 inches of water =15°6 Ibs. per sq. ft., the energy used by the pipe ='30xX156=47 ft. lbs. per sec. In this way the computa- tions have been made for the tables. Mr. Bosanquet limited his work to observing the times, and finding the difference of their reciprocals, thus getting the desired relative values. He used a string pendulum, finding the time needed for the bellows to empty itself, the ‘‘ feel” of the blowing lever indicating when the bellows is full or empty. * Comptes Rendus, xev, 1062. C. K. Wead—Intensity of Sound. 23 But the numbers he gives, especially for leakage, show such wide variations as to throw great doubt on the accuracy of the method. Therefore two modifications were made: first, a stop-watch indicating eighths of a second was used; and second, the movement of the wind-indicator above the key- board was observed through a space of 5U™™; to be sure only about one-third of the wind was used, but it is absolutely necessary to allow 10 to 20 seconds to elapse for the subsidence of the strong vibrations set up in the top of the bellows by the act of pumping. It was sometimes found that though no key was pressed the leakage was different according as the stop was drawn or closed, especially with one of, the pedal stops. One further modification of method was made: since the leakage is more than the amount of wind consumed by any ‘single pipe, except a few of the largest, the influence of errors of observation was diminished by combining several pipes so that they might all sound at once; two ways of doing this were tried : 1. A single stop was drawn, and several consecutive white keys, usually eight, were held down by a loaded block; thus we find the relative consumption of wind by different stops, or by pipes of the same stop in different parts of the scale. See table I. 2. Several stops were drawn as in ordinary playing, and a single key held down by a wedge. Im this case each pipe re- ceives less wind than when no unison pipe is near, a fact long known and further established by these experiments; but we may still find the relative wind-supply in different parts of the seales. See table II and part of III. Most of the results of the work can be given best in tabular form. In the tables the names of the stops need no explana- tion; where 9 stops were drawn they comprise all the stops drawn by the forte composition pedal, including the 7 named separately, a twelfth and a 3-rank mixture. The notation of the keys is used consistently always referring to the key, not the pitch, c’ corresponding to middle ¢(=268 d. v.) when an 8’ stop is drawn; it will therefore be seen that the absolute pitch of all the notes in the lower part of table I is the same; while in the next table pipes of five different lengths, besides the mixtures will respond to a single key. L of course means leakage. The time given is the mean of from 8 to 8 observa- tions: these agreed so well that the probable error of the mean is very rarely 1 per cent: take two examples at random ; Table WIC 224, 226, 926 205; mean 22°66 sec. Cc 62, 63 68, 62, 62; mean 6°375 sec. _ The following columns contain respectively 1--¢, and this quantity diminished by the leakage: this remainder represents 24 C. K. Wead—ILntensity of Sound. 813 =i49 Se 2480- 9Z0L- | G46 o-) Bina da sar ae yyuoeasLy ,Z LE1 on oa ae cae at 9V90- 00L0-| 82-71 | 2-2 eo aa “"~ @A8}00 /F 001 er roe eee pase fear 00F0- F990. | 90-81 7 i a hs jodmusy, 8 raoy | ventas 5 ae. “tg =a L070. | 19G90-| €8-L1 .) ose fee BuBlO[NG /8 FEL eC aa 2 an wad 3 ewe 9640. 0690: | 09-F1 ” gar ee BIPO[PW 8 881 tec eae ft es Fae pS 010. $060: | 90-11 | 2-2 | g ~~ uosedetq aed¢ ,g Gil ae sce oe) a. tine 0970. W090. (66-90 | 79 |) 8° RTS uopanog ,9T GTI ; ae ao tg oe ae? ie IL- 09PF0- | VI90- | 64-91 M4 ae ee ae ” 19T peak oe =: a Sy S790: 4640-| 9-81 | ,-,2 Coie eae uopinog /9] €6 ese LTO: GF G90- Z08- €L60- | 1890-/ 96-81 | 0-2 Bie mn oe " 911 ele €€40- fells GEG0- €&¢- C9P0- | 6190-| 91-91 | 2-0 eG ae . 818 Eo 9680- laser G180- ee 2180. | 9601: | 94-6 | 2-9 Ga ete YIMeayLy ,Z 19 ae a ole 4960: 099. 19Z0- IUGrO> GES | 7a Pia Be 15 a ” 66 om Pee ag) air LLE0- 6E8- G580 6PG0- | 12-81 | 2-0 fous, Some 8 i SIT ig 3 eree F1— 6790- GL¢- 1170. GZ90-| 00-91 | 2-0 ean Pera » 90% = Ty eae ee 9940- as 2680: 9460-/GZ-01 | of Sie a ae ey joduni t, /8 681 me Se lac) Sok ee go GG10- GOGO: OT SOc) eh. de 1» GGG = ioe 3-9 — are 6001- E9IT- | 09-8 io) Pe Jee 49 3oP es x oie? ay ae ee LO8T- T96I- | O1-¢ = eres 7 ice: 4 ; cee ee 2 ae as as ) Heneris O.40e) oe 0 |~“uosedeiqg uedg ,g bh as a tig os ae eee = Aco: GUROL | PO;0Gel g7ee7e 18 = 1 o 9 96 9.7+ 990. C= | 680: 08g. E8£0- 19G0- | €8-L1 | 19-9 | 8 : » g9oT oe ¢190.- [Gt = i 8690 | €89- 0990- 8680-|F6 IT | ,2-? ee ae ” GPG G.9— CgOT- 9-7— PIO, || 1699: 1960: GPII- | €4-8 p-o Lb r) clr I kee L7LI- Cero ler CEO Tyas wi S 0991. | SEsl-| FFs o-( g ~~ aosedeiq uedg ,g oo 3s eS 2 ae ‘Te | 8LT0-| 6-99 | Zz ee Meat a aovyeoT xX OL ‘4u00 ed | X <2. Neplomic.4-... x 5 Nealogic2=- 9 2-_= x Mplievolics-—— x Geratolocic? Figure 2.—Geuesis of Bilobites. a, nepionic stage (x4). Ordovician type like Platystrophia biforata. b, nealogic period ( x 4) at which divergence begins. c, Bilobites bilobus ( x 2). Epaemic form, Niagara horizon. d, Bilobites Vearneuilianus (x 2). Acmic form, Niagara horizon. e, Bulobites varicus (x2). Paraemic form, Lower Helderberg horizon. The genesis of the species is represented in the accompany- ing illustrations, in which it is shown, that all these species are alike in their development up to an early nealogic period, 56 C. Et. Beecher—Development of Bilobites. figure 6. B. Vernewilianus, figure d, diverges at this point, progressively increasing its variation from the normal direct growth, as exemplified in B. bilobus, figure c. B. varicus, figure ¢, passes through all the bilobus stages, and culminates in larger individuals, with less pronounced lobation of the shell. The direct line of development, or the epacme, is repre- sented by &. bdlobus, and it is significant that this form also has the greatest geological and geographical distribution. Next, the divergent and indirect line, or acmic group, typified by BL. Vernewlianus and B. acutilobus, is also widely dis- tributed, but less so than the first. Finally, the paraemie, or geratologous form, £. varicus, culminated and disappeared within very narrow time and regional limits. Yale Museum, New Haven, Conn. EXPLANATION OF PLATE I. Bilobites acutilobus, Ringueberg. FIGURE 1.—Outline of specimen from Niagara Group, Lockport, N. Y. x4. A ° o7- ° \ Bilobites Verneuilianus, Lindstrom. FIGURE 2.-—Common elongate form from Upper Silurian, Gotland, Sweden. x 4. Bilobites varicus, Conrad. FIGURE 3.--Dorsal view of youngest individual observed; showing inception of radiating strize and concealment of hinge areas. x18. FIGURE 4.—Profile of same; showing depth and extent of both valves. x18. FIGURE 5.—Hinge view of preceding. x 18. FIGURE 6.—Dorsal side of specimen; showing beginning of anterior marginal sinus. x 18. FIGURE 7.—Profile of same. x18. FIGURE 8.— Posterior view of same. x18. FIGURE 9.—Dorsal view of specimen, figure 15, showing concealment of ventral Blingtels <3) FiGURE 10.—Ventral view of same; showing dorsal area x 9. Compare this with dorsal view of larger specimen, figure 21, in series. FIGURES 11-26.—Series of specimens; seen from dorsal side; exhibiting ob- served stages of growth, variation and development of hinge, hinge area, and marginal sinus. x4. Figure 27.—Interior of ventral valve; showing teeth, muscular impressions, minute concave plate in apex of: fissure, and arrangement of puncte between nodes and ribs. x6. Lower Helderberg group. Albany County, N. Y. Bilobites bilobus, Linné. FIGURE 28.—Outline ; showing characteristic form of this species as occurring in © Upper Silurian of Gotland, Sweden. L. V. Pirsson—Gmelinite from Nova Scotia. 57 Art. VIIl.—Gmelinite from Nova Scotia; by Louis V. PIRSSON. THE zeolites of Nova Scotia have long been noted for the size and perfection of their crystals, and among them gmelinite has held a prominent place. Originally described by Jackson* under the name of ledererite, it was first proved to be identical with the gmelinite of European localities by Des Cloizeaux,t from crystallographic measurements. This was subsequently confirmed by analyses published by Marsh.{ Analyses have also been published by A. B. Howe, referred to later. Beyond these observations there seems to have been no investigation of the crystal form and physical properties of the mineral from American localities. This has been undertaken chiefly upon material collected during the past summer at Pinnacle Island, ene of the “ Five Islands” in the Basin of Minas, Nova Scotia. An analysis, which was made to control the results of the inves- tigation, having brought out some interesting facts, a discus- sion of the chemical composition has also been added. And since gmelinite has been referred by some authors, especially Tamnau§ and Streng,|| to chabazite, all points bearing on this question have been kept in mind and are here presented. The gmelinite from Five Islands occurs in seams implanted in a greatly decomposed trap. The crystals, often of large size, vary in color from a very pale flesh-red to a strong reddish- brown. Im thin section they are seen to be composed of a colorless outer shell or zone, inclosing a colored inner nucleus. In grinding the sections it was noticed that the outer shell was hard and tough, preserving the crystal boundaries, while the inner portion was spongy, cellular, somewhat friable and readily crumbled away. In large crystals the separation into parts of the colorless outer shell and the colored nucleus can be readily seen with the eye at a trihedral angle. There were no inclusions seen in thin section, only a slight discoloration along the cleavage cracks and occasionally ‘elsewhere. The erystals from Two Islands, Nova Scotia, and Bergen Hill, N. J., studied in connection with these are white, often with a pink tinge, translucent and apparently entirely homogeneous. Some in Professor Brush’s collection labeled Parsborough, Nova Scotia, are similar to those from Five Islands and may indeed have come from that locality. -* This Jour., xxv, pp. 78, 1834. + Man. de Min., pp. 398, 1862. ¢ This Jour., xliv, pp. 362, 1867. § Jahrb. f. Min., pp. 633, 1836 | Ber. d. Oberhess. Ges. f. Natur u. Heilkunde, xvi, pp. 74, 1877; also full abstract in Zeitschr. f. Kryst., pp. 519, vol. i, 1877. 58 L. V. Pirsson—Gmelinite from Nova Scotia. The following table shows the forms which have been observed on these crystals, several of which are new. In the first column the symbols are those of gmelinite as a distinet species, in the second the same are referred to the axes of chabazite. : AS gmelinite. As chabazite. As gmelinite. As chabazite. 6,0," 10001 O, 0001 1 Be OT 2 2023 m, I, 1010 By NOLO o, = OT -2, 0223 a, 4-2, 1120 i-2, 1120 q, ay 3032 RB, . Mbpiale l, i-2, 5270 i-L, 5270 6, 7 4 w4ane v7, 861421 Of these forms c, a, / and g are rare, the others occur on all crystals, almost without exception, from American localities. The basal plane ¢ occurs only on a few crystals from Two Islands and on a number of those from Bergen Hill. As noted by others, the face @ is generally characterized by the vicinal development of a pair of low scalenohedrons. The prism m is not generally striated in a horizontal direction, as observed on European forms. ‘The scalenohedron ¢ is invariat bly striated, oscillating with both the plus and minus rhombo- hedrons and in some cases, possibly, with a pyramid of the second order and a minus form of the same scalenohedron. Many crystals show on the goniometer, by revolving in the zone 7-9, a continuous band of light with the signals of these faces standing ont. The scalenohedron mentioned is, however, most prominent. The presence of this striated scalenohedron g is the most characteristic feature of the American forms, it is almost never lacking on any of the large number of specimens examined, A common appearance of one corner of the Pin- nacle Island crystals, where it oscillates with the rhombohedrons, is shown in fig. 3. While in general the crystal planes gave poor reflections of the signal, a number from Pinnacle Island were well suited, by the brilliancy and luster of the unit rhombohedron, for measur- ing the polar angle r~vr. This was done on a series of ten carefully selected crystals and the results are given in the fol- lowing table. Each measurement is the mean of five deter- minations and the greatest variation between the mean and any one determination is given in the second column. 685,107 .420 On nO Gras. 6Bo ose). 0° 07 ale 68 10 24 0 0 36 67 56 54 0 0 54 68! 16" 12 0 0° 18 68 TT 24 0-07-36 6B)) 3) 40 0 0 40 68 10 ks 0. Ay yika 68 4 18 07.0. as 68. 0.45 0 0 55 Average 68 3 Ol The table shows the degree of accuracy with which the angle could be measured. Of the above, that which gave 68° 08’ was selected as a fundamental. The reflections of the L. V. Pirsson—Gmelinite from Nova Scotia. 59 signal were extremely good and it is not far from the average of them all. From this we derive the axial ratio: a@:e:: 1: 0°734486 If we refer gmelinite to the axial ratio of chabazite, commonly accepted where 7A 7=85° 14’ and eae ss 4 =} -8G0 the prominent rhombohedron of gmelinite becomes 3, 2023, and this requires a length on the vertical axis of chabazite of ‘7240 and a polar angle of 67° 28’. These Pinnacle Island erystals would not permit of so great an error in the determi- nation, if they were referable to the axis of chabazite. The discussion of this point will be referred to later. The follow- ing table gives the calculated and measured angles which show the identification of the forms. The first column of figures gives the theoretical angles calculated for this species, the second gives the theoretical ones calculated by DesCloizeaux, and the third those calculated from the axes of chabazite : Forms. Cale. Dx. Chab. Meas. rar 1011, 1101 *68°08’ 67°34’ 67°28’ See above. r xm 10111010 49 42 *50 03 50 06% 49°407-50°02’ av. of 8=49°46’ rap 1011A0111 37 44} 37 27. 37 244 37 30-347 51 av. of T=37 39 Pee010 ~ S082 38.10. _2.. 938 34$. 38°48 @ ax 4377. 7347 29 214 _--. 29 04% 29 33 29 58 pepe Ott .4377 .16 044 ___. 15 56} 16 14 16 00 maa@ 101041120 30 Ee ape re at BOBS AG mae POLO S270- 16 06> 22. sl «16 30 For reasons stated before, none of these angles could be meas- ured with great accuracy, yet the averages agree better with the theory presented for these crystals than that given by Des Cloizeaux. The angles of the scalenohedron, as given above, was measured on a crystal from Two Islands where it was present almost without striations. This erystal is shown in fio. 1. In all of the figures the crystals are shown revolved 60° into the position of a minus rhombohedron, it having been found that this gave a better view of them. With the excep- tion noted, all the measurements given in the foregoing are upon crystals from Five Islands. Twinning.—The twinning of gmelinite has never to our knowledge been observed, beyond a brief note as to its possi- bility in an article by Howe, mentioned later. In examining a series of specimens I have discovered, however, numerous instances of a twinning on the basal plane. All that have been observed were penetration twins. They are often shown by the growth of the scalenohedron g and small g face, as pre- sented in fig. 3, directly out from the plane of the positive rhombohedron. This method of twinning is shown in fig. 2, an 60 L. V. Pirsson—Gmelinite from Nova Scotia. example which did not have the centers of the two individuals coincident. Further, the figure shows a common habit in the development of the forms. The second method of twinning is that in which the 3 rhombohedron becomes the twinning plane. This was first seen on a specimen from Parsborough, which presented a number of examples; it was afterwards observed on a number of other specimens. In all of these the twins were large sized crystals. An example of this method of twin- ning is shown in fig. 4, as well as another modification of the habit. The angle 7~7 was measured over the twinning plane in eight cases with the following results : 26° 04’, 26° 05’, 25° 58’, 25° 597, 25° 48", 25° 42" 25 Aan eee the average of ee) ale give an angle of 7 on the twinning plane of 77° 0’, and this shows the latter to be the 3 rhombo- hedron. If we use the elements already given (and "the angle rar was measured on one of the best of these crystals as 68° 09’ agreeing closely with that given as theoretical) the angles given above would be in theory 25° 04’. If, however, we use the elements of chabazite and geneMies the prominent rhombohedron on the gmelinite as %, our twinning plane becomes the unit rhombohedron of chabazite and the theoreti- cal re-entrant angle between the % rhombohedrons twinning on this plane would be 26° 182’. The last three measured angles, which were the best, are then about half-way between these two calculated angles. In hke manner pap was measured L. V. Pirsson—Gmelinite from Nova Scotua. 61 over the twinning plane in two cases and found to be 2° 21’ and 2° 55’, while theory would demand from our gmelinite ratios 4° 16’ and for the chabazite 2° 89’. It should be stated, however, that the two po faces on each pair.of the measured twins showed the low vicinal scalenohedron, characteristic of this face, and this of course tended to diminish the measured angles. Indices of refraction.—Three prisms were cut from differ- ent crystals, by using the plane m for one face of the prism and grinding another in the prismatic zone. In the first one only was any well defined double refraction detected by the eye; in the other two the image of the slit was measured by holding the analyzer in front with the shorter diagonal vertical and horizontal. The three gave: wNa 1°4760 1°4646 1°4770 éNa- 1°4674 1°4637 1°4765 There is therefore a very weak negative double refraction which. varies in different crystals, the averge was wNa—¢Na | for the above =-0033, while Negri* found wNa—eNa =:0018 in crystals from Montecchio Maggiore. Optical churacters.—In a section cut normal to the vertical axis it is seen under the microscope, between crossed nicols, that the section is not uniformly dark, but that slight optical anomalies present themselves, somewhat as in leucite. There does not seem to be any definite separation into parts, which would show the crystal composed of several individuals. In strongly convergent light the uniaxial interference figure is seen, and at some places, in revolving, this generally opens a trifle, with the arms of the cross assuming the position of hyperbolas. This is most marked in the hard outer shell, men- tioned before, where a small but distinct separation can be seen. These characters explain very clearly the variation in the indices of refraction in different crystals noted above. Cleavage.—The prismatic cleavage, first noted by Rose, is easily produced but is never very perfect. In a basal section, under the microscope, it is seen as a series of cracks parallel to the prism edges. An endeavor to determine whether a rhom- bohedral cleavage existed, or not, met with only partial success. A series of fragments, with the faces of the prism and unit rhombohedron upon them for orientation, were placed on the goniometer. Upon revolving, the prismatic cleavage always gave a reflection and in a number of cases there were reflected faint but distinct signals in the zone, from small faces, which gave measurements from the prismatic cleavage, as follows: Ae 4oesG Oe). 407 43’. 49° 55’ 50° 06’, 49° 027 * Zeitschr. f. Kryst., xiv, p. 584, 1888. 62 L. V. Pirsson—Gmelinite from Nova Scotia. Our theory demands for mar, 10101011, of gmelinite 49° 42’, In two cases the above were measured both as posi- tive and negative rhombohedron on the same fragment.’ This latter and the fact that the crystals are more or less cellular internally, renders it probable that these reflections came from minute interior faces. ) Chemical composition.—In order to obtain a control over the crystallographic work on the Five Islands gmelinite, two analyses A and B have been made. In A, the outer shell men- tioned before, was analyzed, and in B the inner nucleus. The material was easily obtained by taking fine crystals and split- ting off the shell by pressure. The fragments thus obtained were perfectly colorless, the inner portion had the usual flesh color. Both ground to a pure white powder. The analyses on the air-dried material were as follows : A B Sil ge te we Soh een Bete oe elles 50°35 50°67 ABE Ogee voles her coeds mys iy eee 18°33 18°50 BegOge he ee aes, SU, he 0°26 Onis CaO Aaene hee ins 101 1°05 K.,O SS a re Sten yey Pe 0°15 0:16 BS ek eS as pe RE S56 9°88 OO) Ba aN SAAS AE ke eo Ek 20°23 20°15 100°09 100°56 It will be observed that the analyses show no difference be- tween the two portions. Also the specific gravity carefully taken with the heavy solution was found to be 2:037, the same for both. The most marked result of the analyses is the very small amount of lime and large amount of soda indicated. Analyses of American gmelinite, on material from Two Islands, Five Islands and Bergen Hill have been made by A. b. Howe.* It will be observed that the one on Five Islands’ material is almost exactly like those just given. Two Bergen Five Islands 2. Theory. Hills}. Theory. Islands3. Theory. Ops seer 51°36 51:18 4867 48°79 50°45 49°74 WAU Ose, RPM aOn eA Sse 18°84 18°27 18°12 HesOa meee iat Bie a 0°10 races O17 a nian CAO es 22k 5°68 6°04 2°60 2°40 1:12 Tey, Oo etary 0:23 Brae trace pee 0:20 coe Nap One 2 3°92 3°89 9:14 8°69 9°79 9°75 el Gece ou 20°96 DAT 21°35 PAN OPAS' 20°71 PAPA 100°11 100:00 100°58 100:00 100°71 100°00 In the article previously quoted Streng has shown that chaba- zite may be considered a mixture of two isomorphous hydrated molecules, similar to the feldspars. If we consider gmelinite as a soda chabazite, we then have for these molecules: e=NaAlsi,O,.4H,0 y=Na,Al,Si,O, .4H,0 * This Jour., vol. xii, pp. 270, 1876. L. V. Pirsson—Gmelinite from Nova Scotia. 63 the first a hydrated albite molecule, the second a hydrated soda anorthite. If, according to this, we deduce the composition indicated by the analyses of the three gmelinites, replacing soda by lime to the extent observed, we have Two Islands =62+y in which Na: Ca::2 :3 Bergen Hill =3x+y = Mars Casares | Five Islands =4%+y i Nas CaS) 7:71 The theoretical composition for these formulas is given, for convenience, in the table after each analysis. The very close agreement of the theory, calculated for these simple relations, with the analyses themselves is very striking and a-strong proof of the correctness of Streng’s theory. In chemical com- position we may consider typical gmelinite as a soda chaba- zite, whose relation to the normal lime chabazite is the same as that for instance of lithiophilite to triphylite. Conclusion.—In considering the bearing of the foregoing facts upon the identity of this mineral with chabazite there is an apparent discordance. The result of the crystallographic work, points to a distinct difference in axial ratios and there is also a different habit and cleavage. On the other hand the twinning and the chemical constitution, both following that of chabazite present the strongest possible arguments for the identity of the species. To explain these apparent discrepan- eies the following hypothesis is offered. The analyses of chabazite and gmelinite, made by various chemists, show that soda and lime may replace each other to any extent, but that in gmelinite the soda is in excess, while in chabazite the reverse is true. If we consider then that the effect of the soda is to lengthen somewhat the vertical axis, the difference in angles and ratios would be accounted for and we might expect it to change also the habit and cleavage. While this cannot be considered otherwise than a hypothesis, the fact that in the Five Island material under examination these differences are greater than noted by any former observer, while at the same time the percentage of soda is also greater, points dis- tinctly towards it. According to this view gmelinite would bear much the same relation to chabazite that enstatite does to hypersthene, whether it should be considered a distinct species would be largely a matter of choice or convenience. ) In closing the author desires to express his thanks for the liberal use of valuable material to Professor G. J. Brush and to Professor S. L. Penfield, to the latter also for valuable advice during the progress of this examination. Mineralogical Laboratory, Sheffield Scientific School, New Haven, Feb., 1891. Am. Jour. Sci.—THixD SerRizs, Vou. XLII, No. 247.—Juny, 1891. 5) 64. J. M. Davison—Analyses of Kamacite, Tenite and Art. [X.—Analyses of Kamacite, Twnite and Plessite from the Welland Meteoric Iron ;* by Joun M. Davison. THE siderolite, which forms the subject of this paper, is described by Edwin E. Howell on pages 86-87 of the Pro- ceedings of the Rochester Academy of Science for 1890. Its analysis gave Fe 91°17 and Ni8-54. It is singularly free from troilite and schreibersite and thus offered an unusually good opportunity for the analysis of its separated nickel-iron alloys. On sawing the meteorite, the outside was found much decom- posed ; but between this and the compact center was a zone in which the oxidation was superficial and confined for the most part to planes of contact of the different nickel-iron alloys that form the Widmanstatten figures. It thus became possible to separate the kamacite and the teenite in quantities sufficient for analysis. The quantity of kamacite used for analysis was gm. 0-934, of tenite om. 04522. The physical characters of these alloys differ widely. The kamacite is brittle, breaking with a subconchoidal fracture, and is of the color of cast iron. It was coated with a thin film of black oxide which had often a resinous luster as if covered with lacquer, particularly where the teenite had been freshly stripped off. This oxide is attracted by the magnet, and is probably the magnetic oxide Fe,O,. Some pieces of kamacite of a millimeter or two in thickness were entirely altered to this oxide.. The kamacite shows, in places, a corrugated surface, in some specimens resembling bundles of rods, ike the columnar structure of hematite. Figures 1 and 2 show this columnar structure. In the latter the teenite which closely followed the form of the kamacite is laid back, but not detached. The teenite has a silvery luster with, when slightly oxidized, a tinge of bronze. It is flexible and elastic and fuses on the edges in the oxidizing flame of the blowpipe, turning dark. Its fusibility seems to be about 5. It resists oxidation better than the kamacite; the contrast between its comparatively fresh appearance and the dark film covering the other was marked, and facilitated their separation. Both kamacite and tenite were magnetic and exhibited a weak polarity which was more marked in the latter. Pieces of teenite floated directly on water, and of kamacite buoyed on a cork, arranged themselves in the magnetic meridian; the taenite promptly, the kamacite after being left for some time protected from air currents under a bell glass. The meteorite as a mass also showed polarity. The teenite is found separating the plates * Read before the Rochester Academy of Sciences and published in the Pro- ceedings for 1891, where it is accompanied by a plate, not reproduced here. we Plessite from the Welland Meteoric Iron. 65 of kamacite and enveloping the crystals of plessite. Figures 3 and 4 show plates of kamacite which were in close contact, and when separated were found to have been joined by a little triangular prism of the same substance. It was, at first, intended to analyze the plessite as a whole; but on examination its fine layers were so suggestive of kama- cite and tenite that the attempt was made to separate them, and to analyze each separately. It was found that one was brittle, the other flexible and elastic; one dark with superficial oxidation, the other showing the tzenite luster. Physically their correspondence, the one with kamacite, the other with teenite was exact, and in the kamacite-like part the columnar structure was ee Z on a diminutive scale, the diameter of the rods being from 4—$ mm. Their separation ‘then became simply a matter of patience, and with the aid of a watchmaker’s glass, and a magnetized needle to pick up the grains and flakes, most of which were too small for even delicate forceps to handle, there was obtained for analysis, of the part resembling kamacite gm. 0°5261, of that resembling tenite gm. 0°1314. The thickness ae phe kamacite was from 1-2 mm., that of the tenite from 4-34 mm. In the plessite the kamacite-like bands were from 1i-3, mm. thick ; ae Pele bands, as nearly as could be meas- ured, from. sho-sy7 MM. The method of analysis was the same in each case. The material was gone over repeatedly, piece by piece, with a watchmaker’s glass and very carefully assorted and cleansed, the pieces of kamacite being scraped bright. It was not possi- ble to do this to any extent ‘with the kamacite-like part of ples- site. It was dissolved in dilute hydrochloric acid by the aid of a weak galvanic current, at the positive pole of the battery. The carbon thus separated was collected on a Gooch filter and burned. The nickel and cobalt were separated from the iron by digestion in ammonium hydrate, the process being repeated four times. The iron was weighed, and the nickel and cobalt first determined together by electrolysis, then separated by potassium nitrite and each determined separately in the same manner. lor comparison, the analyses of kamacite and teenite are given each next to its corresponding part of the plessite. ‘ Kamacite. Plessite. Teenite. al a eS ee ee SSS Kamacite-like part. Teenite-like part. Fe 93:09 92°81 72°98 74°78 Ni 6°69 6°97 29781 24°32 Co "25 "19 °83 33 C "02 19 OL "50 — —____ —S 100°05 100°16 100°59 99°93 66 Scientific Intelligence. These physical and chemical correspondences justify, I think, the conclusion that in the Welland siderolite there are but two distinct nickel-iron alloys, viz: kamacite and tenite; and that the so-called plessite is mer oe) thin alternating lamelle of kamacite and teenite. It is unsafe to generalize on a single analysis, but an exami- nation of the markings of other meteoric irons suggests the thought that in them also there may be but two distinet alloys. Such are the Descubridora, the Glorietta Mt. and notably the Kiowa Co. and the Augusta Co., Va. meteorites. In sections. of the last two irons in Ward & Howell’s collection every piece of the so-called plessite in the Augusta Co. iron shows its thin lamellee, and in the Kiowa Co. pallasite the gradations of the markings are such, that in parts of the iron it would be difficult to say which should be called kamacite and which plessite. In etching meteoric iron, the kamacite is attacked by acid more readily than the teenite richer in nickel. The teenite and plessite stand in relief. Where lamelle do not show in plessite - may not closely crowded teenite bands have protected neighbor- ing kamacite layers from acid action, and might not more care- ful or prolonged etching develop lines in plessite that now appear homogeneous ? Reynolds Laboratory, University of Rochester, April, 1891. SCIENTIFIC INTELLIGENCE. J. CHEMISTRY AND PHYSICS. 1. On the Speed of the Explosive wave in Solid and Liquid Bodies.—BERTHELOT has studied the phenomena attending the production and transmission of the explosive wave in solid and liquid substances and finds that these phenomena do not have the regularity of progression observed in gases. In liquids the speed appears to be dependent upon the rigidity of the enclosing tubes, this speed being the greater the greater the resistance of the tubes to rupture. It is probably not possible, however, to prepare tubes which can bear without fracture the force of the explosion, since the volume of the bigh explosives is smaller as a rule than the volume of their decomposition products, even when these are compressed into the liquid condition. In methyl nitrate, the author finds that the explosion travels, when the liquid is contained in tubes of steel, with a speed of about 2100 meters per second.— C. &., cxii, 16; Ber. Berl. Chem. Ges., xxiv, (Ref.) 253, April, 1891. G. F. B. 2. On the Relation between the Electrical Energy and the Chemical Energy in Voltaic cells.—A series of experiments by Livay has been made to ascertain the amount of heat gener- aS ees Chemistry and Physics. | 67 ated by the current of certain voltaic cells, as compared with the amount generated by the chemical action going on in the cells; in order to determine the exact relation of these two quantities. The heat developed by the current was ascertained by means of a silver voltameter placed together with the cell, ina calorimeter. The heat evolved by the chemical action was deter- mined by direct calorimetric means. The cells examined were of | the Daniell and the De la Rue forms, three experiments being made with each. Asa result the author finds that with the Dan- iell cell, the heat equivalent of the current is greater than that generated chemically ; so that in the working of this cell, heat is absorbed. On the other hand the De la Rue cell shows a reverse effect, not all the heat proper to the chemical action going on appearing in the circuit. But in this case the author observed that the relative amount of electrical energy increases with the concentration of the solution in the cell. These results confirm substantially those of Jahn.—Ann. Phys. Chem., Ul, xlvii, 103; J. Chem. Soc., 1x, 513, May, 1891. G. F. B. 3. On the Action of Heat on Carbon Monoxide.—BERTHELOT has observed that when carbon monoxide is heated ina glass tube to 500° or 550°, a minute quantity—three or four thousandths —of carbon dioxide is produced ; and this without any simul- taneous separation of carbon. If, however, the carbon monoxide be passed through a porcelain tube, and the temperature of this be raised to a dull or even a bright red heat, while approximately the same quantity of carbon dioxide is observed to be produced as before, there is at the same time a distinct separation of carbon. Hence the author concludes that in this experiment carbon mon- oxide is not simply dissociated, but is at the same polymerized ; and that the product of this polymerization decomposes into earbon dioxide and carbon sub-oxide according to the equation C,0,=C,_,0, .+CO,; which sub-oxide at a higher temperature yields carbon monoxide and free carbon.—C. f#., exii, 594 Ber. Berl. Chem. Gies., xxiv, (Ref.) 348, May, 1891. Guan 4. On the Electro-metallurgy of Aluminum.—MInET has con- tributed further details concerning the reduction of aluminum by electrolytic methods. The steel crucible is now made smaller and is provided with an internal lining of carbon which serves as the negative electrode. The difference of potential between the two electrodes is 4°55 volts and the yield is 31°9 grams of alu- minum per horse power per hour, or 31°3 horse powers per hour for one kilogram of aluminum. ‘The author believes that it will be possible to reduce the difference of potential to 4-volts and under these conditions there will be no electrolysis of the sodium chloride and the yield will reach 70 per cent of the theoretical quantity. The loss of 30 per cent is due to the action of the fused fluorides on the aluminum and does not occur when aluminum alloys are made, since in this case the electrolytic cell is composed of the other metal and the liberated aluminum at once combines with it.—C. &., exii, 231; J. Chem. Soc., \x, 525, May, 1891. G. F. B. iil | AW! yk t mit et | i / f git | Tea 68 Screntific Intelligence. 5. On the Detection of metallic Mercury in cases of Poisoning. —It is generally assumed that metallic mercury when treated with hydrochloric acid and potassium chlorate goes readily into solution. But Lecco has observed that in destroying the organic matter in toxicological cases with these reagents, metallic mercury if present is only very slowly attacked. A human stomach in which minute globules of mercury could be distinctly seen was treated in this way until the organic matter was destroyed and then examined as usual, Scarcely a trace of. mercury could be detected in the solution, while in the residue minute globules of the metal were visible. Direct experiment showed that eI ey, itself is soluble with extreme difficulty under these conditions and hence the author believes that in examination for poisons this fact should be borne in mind. He recommends that the process of treating with hydrochloric acid and potassiam chlorate should be continued for some time after the organic matter disappears.— Ber. Berl. Chem. Ges., xxiv, 928, April, 1891. Gs ae 6. On Tetrazotic acid and its Oxy- and Di-oxy derivatives.— In consequence of the observation that by the action of acids up- on benzenyl-amidine nitrite, an acid is produced having the for- mula C,H,N,O, and therefore of the composition of a di-nitroso- benzenyl-amidine, W. Lossen undertook further researches in this direction and has obtained some noteworthy results. He finds (1) that similar compound acids are yielded by other ami- dines provided that in them the hydrogen in the group C7 ONAL, is not replaced by alkyl radicals; (2) that the acids thus obtained X . CN,O,H, called dioxy-tetrazotic acids and of which the above benzenyl-dioxytetrazotic acid- C,H, . CN,O,H is an example, are reduced by sodium amalgam to oxy-tetrazotic acids, X . CN,OH, such as benzenyl-oxytetrazotic acid C,H, . CN,OH, ‘and to tetra- zotic acids X . CN,H, as for example C, ie ON, H benzenyl- tetra- zotic acid ; (3) the ‘dioxytetrazotic acids decompose spontaneously when set free from their salts, and their metallic salts when dry are extraordinarily explosive; (4) the oxytetrazotic acids, in regard to their permanence are intermediate between the unstable dioxyte- trazotic acids and the quiet permanent tetrazotic acids, although the latter and also its salts are explosive; (5) by Raoult’s method the molecular formulas of benzenyl-tetrazotic and benzenyl- oxytetrazotic acids were found to be C,H,N, and ©,H,N,O respect- ively ; (6) with reference to the constitution of these acids, the author states (a) that the hypothetical free benzenyl- -dioxytetra- 2N.NO zotic acid bas apparently the formula O,H,. Crt oN | NOW since it gives Liebermann’s nitroso-reaction and decomposes into benzoni- trile, nitrogen and nitrogen dioxide; (b) that benzenyl-tetrazotic acid, by the action of conecentrated hydrochloric acid, decomposes according to the equation C.H.N,+(H,0),=C,H, .NH,+CO,+N,+NH, Chemistry and Physics. 69 though “aa an intermediate stage occurs as follows: a ZANOH .- ©E.O- AN H+ (H,0),=C,0,CC¢ oy +N,4+NH, the benzhydroxamic acid splitting into aniline and carbon dioxide ; benzenyl-tetrazotic acid may be considered cither as a phenyl- yg tetrazo : , analogous to tne isomeric compoun 1 C,H, Ex l I | d NH— rar | discovered by Bladin CHK || ,oras an imido compound N(C,H,).N ey corresponding to the benzoyl-azoimide of CurtiusC,H,. CO. NG Hee mis a NN in the latter case having the formula C,H,C(NH). Ne Il; (¢) since N benzenyl-oxytetrazotic acid does not give Liebermann’s reaction, it is not a nitro-compound. According to C. Lossen, benzenyl- oxytetrazotic acid crystallizes from: boiling water in rhombic needles which fuse with decomposition at 175°. With one mole- cule of crystal water the acid is permanent, but when deprived of this water at 105°, it readily decomposes evolving nitrous vapors. Its salts with potassium, barium and silver are described. Benzenyl-tetrazotic acid crystallizes from hot water, better from alcohol in rhombic hemimorphie colorless needles, fusing ait 2 120 to 213° with decomposition. By slowly heating it, a beautiful red mass is obtained; while on rapid heating a violent decomposi- tion results, often with ignition, a dark green tenacious residue being left in the test-tube, whose vapor is red or violet. The dioxy- tetrazotic acid affords a meta-nitro-derivative m-nitrobenzeny]l- dioxytetrazotic acid.—Liebig’s Annalen, celxili, 73; Ber. Berl. Chem. Ges., xxiv, 332, May, 1891. G. F. B, 7. Polar light and Cosmic dust.—Livrine and Dewar ob- tained metallic dust by means of electrical discharges between terminals of different metals inserted in a glass receptacle— from this the dust was conveyed by means of a stream of hydro- gen into an end-on-tube, through which electrical discharges were passed. The spectrum of these discharges showed no trace of the lines of the finely divided metals although the finely divided dust was present in great abundance. They therefore conclude that if the northern lights are due to great electric dis- charges through rarified air filled with cosmic dust, conditions must exist which are different from those in the experiment devised by them.—- Proc. Roy. Soc., xlviii, p. 487-440, 1891. nena 8. Phosphorescence.—K. WiEDERMANN has investigated the character of the light given out by Balmain’s paint under different conditions of exposure. He expresses his belief that a source of light which sends forth proportionally more light waves than heat waves, as Langley maintains is the case with the fire fly, is not = es = a a ee oes == SS == ————— = eS = " il Rey ntl Phat 70 Scientific Intelligence. necessarily the cheapest source of light. In order to estimate the ‘cheapness of a light account must be taken of the entire trans- formation of the energy of the light in the process of vision.”— Beiblatier zu den Annalen der Physik, No. 4, 1891, p. 281. 3.7. 9. Reflection and Refraction of light by thin surface layers.— P. Drupr examines mathematically the conditions which must hold for the reflection and refraction of light by thin layers of metals such as Professor Kundt has experimented with in obtain- ing indices of refraction of metals. The paper is long and ex- haustive; but is not supported by experimental results. The author hopes to obtain suitable surfaces to verify his theoretical conclusions. ‘These are as follows: (1.) In the expression for the absolute amplitude, ratios and dif- ference of phase of the reflected and the transmitted light—three constants depending upon the nature of the layer enter. In Cauchy’s formula but one constant depending upon the boundary enters. (2.) For refraction and ordinary reflection the formulas are identical with those of Cauchy. (3.) A lower limit for the thickness of the layer transmitting light is given by elliptical polarization. (4.) No ellipticity is shown if the layer is contained between the media ot the same index of refraction. If the plate is wedge- shaped of snrall angle; in reflected light the bright bands have the normal polarization angle. The dark bands deviate from this, and a conclusion can thus be drawn in regard to the index of refraction of the layer in case the layer is homogeneous. (5.) In the dark band the reflected light is linear polarized— the transmitted light elliptically polarized. In the bright bands the reflected as well as the transmitted light is elliptically polar- ized. (6.) From observation upon the light transmitted by thin metallic layers and on light reflected the true optical constants of the metals can be computed.—Ann. der Physik und Chemie, No. 5, 1891, pp. 126-157. 3, Il. Gronocy AND MINERALOGY. l. Annual Report of the State Geologist of New Jersey for the year 1890. 305 pp. 8vo. 1891.—Since the death of Prof. Cook, Prof. G. C. Smock has been appointed the State Geologist of New Jersey with F. L. Nason and C.’W. Coman as assistant geologists. ‘This report contains an article on the age of the Sussex Co. crystalline limestones by Mr. Nason; an account of geological work in the southern part of the State by C. W. Coman, treating especially of the strata overlying the upper marl bed, and a report on the water-power and water-supply of the State by C. C. Vermeule. Mr. Nason’s paper contains the important announcement that the bluish, semi-crystalline limestone of Sussex Co. and the asso- Geology and Mineralogy. 71 ciated sandstone, have afforded Dr. C. E. Beecher Lower Cam- brian fossils ; and that in one case the sandstone contained, near by, the mineral graphite. The main purpose of the article is to give the evidence obtained by the’author in favor of the conclu- sion that the white crystalline limestone of the county, containing chondrodite and other minerals, which has been supposed to be Archean, is really of the age of the blue limestone. The evidence given is, briefly, the occurrence of graphite in both the white and blue limestones; the passage of one into the other at some localities ; and the inference that the white limestone owes its crystallization to contact with eruptive rocks, (granite, etc.), and exhibits various contact phenomena. ‘The Franklinite iron ore- bed of the county is associated with the white limestone, and is made therefore of the same age. The conclusion is a wide- reaching one, and the facts should have full investigaticn before it is adopted. ‘The evidence drawn from the graphite is of un- certain value as the mineral occurs in rocks of much later time. Prof. Cook regarded the iron ore beds and the limestones as part of the gneissic formation of the region, the gneiss being not in his view foliated granite ; and the writer’s examinations of the rocks associated with these ores have led him to the same con- clusion. Moreover, it is an impossibility that the crystallization of the white limestone formation should have been produced by contact with the dikes of igneous rocks, or even with protruded granite ; for the rock of a dike cools outside too rapidly for such a result. The trap dikes of New Jersey illustrate this point abundantly. Melted granite injected through a cold rock would not be true crystalline granite against the walls or make the limestone adjoining coarsely crystalline, like the white limestone, even for a hundred feet. Again dikes of a hornblendic scapolite rock are described. But it is impossible that melted scapolite injected into cold rocks in fissures four to six feet wide or wider than this, should become on cooling crystallized scapolite, even of a granular form, alike from wall to wall, with “ perfect folia- tion” parallel to the walls, so that it has been mistaken for gneiss. For such crystallization the enclosing limestone should be hot enough for its own crystallization—the condition attend- ing metamorphism. The actual passage of the blue limestone into the white has weight, if the observation is beyond question. The writer doubts the conclusion as to actual passage because he has ob- served in Hast Lee, Massachusetts, an apparent passage of the kind between the Stockbridge limestone and another which is chondroditic, and saved himself from inferring their identity by finding the latter associated in a part of the area with a very different class of crystalline schists. In other cases over eastern Berkshire chondroditic limestone was met with; and in each it was associated with rocks that were in part so unlike the schists of the Stockbridge limestone or Taconic belt, viewing them through its whole course, from Vermont, Massachusetts, and _ Sara ae Se eee =e eS eee —— aa = oS Sew aren ait ih Wall) jai | ii! inal i berrey ty Ih ii | nt | | Cn jenveni ' 72 Scientific Intelligence. Connecticut to New York island, that it was accepted as evi- dence of Archean age. Superposition of the later limestone on the earlier and subsequent changes may account for the cases of apparent passage. Limestone belts have determined the positions of the chief valleys of Berkshire ; and in some cases Archean limestone was first in the work. One of the most comprehensive facts in the geology of Eastern America is the general identity of strike and dip, in associated metamorphic or crystalline rocks of Archean and later time. In eastern Berkshire the writer failed to detect the limit between the Taconic schists and the Archean, after several trials ; and the same was true for the ridge southwest of Cornwall, Conn., where chondroditic limestone occurs; and also in Putnam County, N. Y., where there are Archean iron ores. In each case the quartzyte of the Taconic series was followed by gneiss of like dip and this by other gneisses, and the Archzean limit was not dis- ‘covered. The question was left for a later and more thorough investigation, which has not been made. It is now in other hands, with a promise of success. Taking the evidence which strike and dip afford as of itself conclusive, it is probable that nearly all the so-called Archean rocks of the Appalachian Pro- taxis could be proved to be Paleozoic. The problem which Mr. Nason has investigated in Northern New Jersey is one of great importance and difficulty. It is a part of a wider problem—that embracing all the Archzean schists and ore-beds of New Jersey. Jo DAR 2. Two belts of fossiliferous black shale in the Triassic forma- tion of Connecticut, by W. M. Davis and 8. Warp Loper. 16 pp. 8vo. (Bull. Geol. Soc. America, vol. 1, April, 1891.)— Professor Davis commences his paper with a summary of his conclusions respecting the Triassic formation in the vicinity of Meriden, Conn., and its associated trap. His list of papers mentions five subsequent to the one published in this Journal in 1886, with the title “Triassic formation of the Connecticut Valley.” Under the same title, he published a fuller paper in the Report of the U. S. Geological Survey for 1888. Since then the following have appeared: ‘“ The ash-bed at Meriden and its structural relations,” in the Proceedings of the Meriden Scientific Association for 1889; “On the Topographic development of the Triassic formation of the Connecticut Valley,” in vol. xxxvii of this Journal, 1889; “On the faults near Meriden, and on the intrusive and extrusive trap sheets of the Connecticut Valley,” in the Bulletin of the Museum of Comparative Zoology for 1889. In the present paper the following general conclusions are stated. Three overflow trap-sheets in the vicinity of Meriden are now well made out; the first, thin and amygdaloidal, the second, thick and massive and sometimes a double flow, the third, thin like the first. Beside these overflows one great intrusive sheet, exists, and apparently several smaller ones. The great sheet, as implied in a note, is that of West Rock, of the New Haven re- Geology and Mineralogy. — 73 gion. The east-and-west ridge called Mt. Carmel, situated about half way between New Haven and the Meriden trap ridges is a ‘“‘oreat mass of dikes,” which ‘“‘may be regarded as the locus of the volcanic pipes up through which rose the lavas now seen in the extrusive and intrusive sheets.” The existence of these vol- - canoes is spoken of as without direct evidence, but probable. In the two figures of the paper these volcanoes are represented as buried in the sandstone formation and are entitled “the Group of buried volcanoes,” “The lost volcanoes.” The tilting of the sandstone with the intercalated sheet of trap, giving the forma- tion its eastward dip throughout the region, probably followed the time of deposition and eruption. Even the intrusive dike, West Rock, is probably ‘‘of earlier date than the tilting and faulting of the formation, and hence of roughly synchronous date with the overflows.” The faulting of the sandstone accompany- ing the uplifts was probably guided in direction by the planes of foliation in the underlying schists. The two belts of black shale contain fossil fishes and plants. One of them is that of the well-known Durham locality and others of the same belt. The second occurs in a small brook north of the village of Westfield, Conn., and has been opened also at four other places along a line of about fifty miles. The latter, Mr. Loper states, has afforded one species of fish, Jschypterus gigas, not found in the Durham Jine, and two species of plants also absent from it, Lgwisetum Rogersi Sch., and Ctenophyllum Braunianum Sch. 3. Lllustrations of the Fauna of the St. John Group, No. V.; by G. F. Matrrnew.—Mr. G. F. Matthew’s paper under the above title, though read before the Royal Society of Canada in May, 1890, has only recently been distributed. The author has made a study of the fauna of the lower rocks of New Brunswick, especially near the city of St. John, and has given the results of his labors in numerous papers of interest. In the present one after discussing the structure of the St. John Basin, and various sec- tions of the strata, he describes several new species of fossils and presents remarks upon some old ones, especially upon trilobites. This section is followed by a third treating mainly of tracks and markings, and upon this we offer some remarks and criticisms. He gives, in the first place, a short sketch of Nathorst’s obser- vations upon Meduse, quoting the descriptions of Medusites princeps Torell (sp.) (= MW. jfavosa of Nathorst), MW. radiata Linrs., (sp.) and J. ee Torell (sp.) (=. Lindstromi of Nathorst). All these forms were described from beds of Cam- _ brian age in Sweden, and Mr. Matthew says that in the St. John group indications are found of some of these ‘“ medusa-like forms as Nathorst considers them.” He then proceeds to describe a new genus Medusichnites, founded for the reception of certain trails or tracks “which appear to have been produced by such creatures.” Not that there is any indication they were positively made by JMedusce, ‘but rather that they are probably due to 1 i i | 74 Scientific Intellagence. those Radiate animals which Nathorst has referred to Wedusites.” The name Zaonichnites had been previously suggested by him for some similar form, but he now advocates discarding the old name and substituting a new one. Why the new forms should not have been included with the old one we are unable to under- stand. Fortunately no specific names are given to the series of markings referred to this genus. The author prefers, instead, to i designate them as “forms,” and of these he describes and illus- a iit i | trates five. In these descriptions we no longer find any doubt expressed as to the manner in which the tracks were made or the kind of animals which made them. One is “the imprint of tentacles resting on the bottom.” Another, from the Animikie group of the Lake Superior region, and the original of Taonichnites, he says “‘is a good illustration of certain impressions which have been mistaken for rill. markings, but which are really of organic origin. It has been made by a Medusite swept along by a cur- rent above the surface of a bed of very fine sandy mud.” It is interesting to know that the same genus ranged from the Animi- kie into the Middle Cambrian. Another new genus proposed is Hoichnites, the name being a substitute for Hophyton. Mr. Matthew gives an account of _ Kophyton and of its supposed nature by the original discoverer, Torell, and figures what he calls Hoichnites Linneeanus Torell (sp.), from the St. John group. These figures, while they bear little resemblance to the typical Hophyton Linneanum, are very much like some of the figures of Medusichnites, and should be | placed there if that genus be a good one. Ce A third new genus established is Ctenichnites, adopted for : i! | markings which Mr. Matthew says Torell and Linnarsson con- ne | founded with Hophyton.: We are told the markings resemble Loichnites so far “that they might easily be supposed to have been produced by larger individuals of the kind which made the Hoichnites.” This genus has one species, C. zngens, and it is described with considerable detail. A table is given of the chief at | varieties. Then the sort of animal supposed to have made them ite is discussed, the exposures yielding the following inferences to the author: “1. That the animal lived in schools. 2. That it had a rapid, on a direct, darting motion. 3. That it had three or four flexible, A | fleshy arms. 4. That these arms were furnished with sharp nl | (horny?) spines. 5. That it had an easy motion through the || water so that sometimes the arms of one side touched the bot- tom, sometimes the other.’ ) | Then having found. reason to believe Ctenichnites to be of | animal origin,” Mr. Matthews concludes that it might have been a | | il i t Ht naked cephalopod. Then the habits of squids, and the nature of coprolites, and the armature of trilobites are examined in turn, and the final conclusion is that all the facts point to squids of some sort being the probable source of the Ctentchnites markings. ih | | ‘i mt I | ' Geology and Mineralogy. 75 A comparison of the figures of Ctenichnites ingens with those of Hoichnites and some forms of Medusichnites induces the Opinion that they are not to be separated even specifically. Some of the first have the lines wider apart than those of the second, and are less curved than the third, but the variability in all is so great that to draw a line anywhere between them is a task that few would dare undertake. Under Psammichnites he refers to certain specimens found in St. John which seem to him probably ‘‘ may have been made by by a Psammichnites.’ From this it would appear that Mr. Matthew considers Psammichnites to be an animal form of some sort. The general opinion is (Hancock, Haughton, Torell and Nathorst) that. the forms described under that name are only tracks. Under Frena a new species, / ramosa, is described, and under Arenicolites also, a new form, A. brevis, is given. Still a fourth genus, Goniadnichnites, with one species, G. trichiformis, is created, and on very slender grounds. Small, slender and thread- like, the name is given because of their resemblance to tracks made by recent Goniada. The figure bears some resemblance to certain branching forms of graptolites like Dendrograptus tenui- ramosus, from the Utiva Slate of New York. Last of all the new species is an addition to Torrell’s Monocra- terton, under the name of M+ magnificum. From the plate it is well named, for from a central cavity two inches in its longer diameter, and one and a quarter inches in its shorter, spread out filaments, called “tentacles,” three inches in length: and this figure is reduced one-third. If this burrow were made by a worm, it must have been a gigantic creature. In studying this paper of Mr. Matthew’s we cannot but regret that he has made his many new genera and species upon such scanty material. As objects illustrating some phase of sedimen- tation, or the possibility of some sort of life having existed, these markings are of interest. But it is a useless burden upon science to give to them generic and specific names. JosEPH F. James, Washington, D. C., June 13, 1891. 4. Etudes des gites minéraux de la France. Bassin Houiller et Permien d’ Autun et d’ Epinac. Fasc. II, Flore fossile, Pre- miere partie, par R. Zei~ruER. Pp..1-304. Atlas, xxvii plates, 4°.—This fascicle begins the third of a series of valuable recent works on the flora of the Carboniferous epoch in France. Of these three, the first, on the flora of the Valenciennes basin, by M. R. Zeiller, dated 1888, is the most important work in French on the Paleozoic flora since the “ Histoire” of Brongniart, with which it will take a place as a classic in paleobotanical literature. The first part of the second work, on the Commentry flora, in which the ferns are monographed by Zeiller, bears the same date; but the second part, under the joint authorship of MM. B. | Renault and Zeiller was not finished until 1890. The present | i ie iM a] if i i ‘i | (ana! (a we ‘al il ii ni ! ie at Mn i! itt quit! el i | Wt} it nh Ha, i i ‘il, ah | / svn ‘halt wet il! ae a i | jolt | k Mell i feoahentel| qelinit 1} i A ‘ ii nt ' i | | aH) itt (eg | Sly wo 00) Wal fl it") i ‘ . i ! \" jugdinins* Te". Scerentifie Intelligence. work includes the flora of the Epinac and Molloy stages of the Upper Carboniferous and the [gornay-Lally, Cornaille-Chambois, and Millery stages of the Lower Permian. The Millery horizon, in the Autun basin, is celebrated as the source of the wonderfully preserved silicified plants that formed the basis of the many important works on the organization and fructification of the plants of the Paleozoic by Brongniart, Grand ’Eury, Renault, Bertrand, and Zeiller. In this fascicle Zeiller treats the ferns, prefacing their description with an illustrated résumé of the classification of the types represented in this flora according to their discovered fruiting forms. Considerable new and interest- ing material is here brought to hight. About forty species, many of them new, are described from foliar and fruiting characters. The last 120 pages contain descriptions and illustrations of the trunks and petioles of the ferns, belonging to Ptychopteris, in- cluding Caulopteris gigantea, F. & W., to Psaronius, represent- ing trunks of Pecopteris and Scolecopteris, and comprising an extinct tribe of the Warattiacew, and to Myeloxylon, including Medullosa (pars), Myelopteris and Stenzelia, which he regards as petioles and rachises of Alethopteris. Odontopteris and Neur- opteris, representing a group, with pithed petioles and a cen- trifugally developed:secondary woody zone, perhaps intermediate between the Ophioglossacew and the Marattiacew. The flora is interesting as showing many transition forms between the Upper Carboniferous and the Permian types. ‘The second part of the work, dealing with the remaining groups, is in preparation by M. B. Renault. D. W. 5. The Genus Sphenophyllum, by J. S. NEwserry. Journ. Cincinnati Soc. Nat. Hist., vol. xiii, 1891, pp. 212-217, pl. xix. —In this short paper Dr. Newberry reiterates the view pro- posed first by himself in 1853, and afterwards independently by Coemans and Kickx, that in certain species of Sphenophyllum in which the leaves are normally wedge-shaped and dentate or serrate, the deeply dissected, fimbriate, or capillary forms, simu- lating ‘Asterophylllites, belonging to the same species, represent only ‘portions of the same plant that were submerged. Several figures illustrate different parts of S. erosum, including the forms known as S. saxifragefolium. The author also gives a few of the characters of six species of this genus with which he is familiar in this country. Dr. Newberry regards Sphenophyllum, whose affinities have for over fifty years been the subject of controversial discussions, as representing a peculiar and extinct family whose nearest living relative is Aguisetum. D. W. 6. Annuaire Géologique Universel, Année 1889, Tome VI. Paris 1890.—This geological Annual, founded by Dr. Dagincourt, is now under the direction of Dr. L. Carez for Geology, and M. H. Douvillé for Paleontology; and besides, it has many able co-work- ers from among the geologists of France and other countries. The Annual for 1889 is a closely printed large-octavo volume of 1200 pages. The first 120 pages are occupied with lists of the Miscellaneous Intelligence. a geological and paleontological papers, memoirs, maps, etc., of the year, arranged according to subjects and countries : and after a catalogue of the authors in the lists, the following 1000 pages of the volume contain quite full abstracts of very many of these publications. Not only the names of new species are given in the Paleontological part, but, to a large extent, descriptions of genera, and among the Vertebrata of many of the species, besides a review of new deductions and opinions. The Annual is ee to the geologist who would know about the yearly progress of the science over the world, and keep himself informed of dis- coveries bearing on his own ‘work. 7. Tables for the Determination of Minerals by physical properties ascertainable with the aid of a few field instruments, based on the system of Professor Dr. Albin Weisbach by Persiror Frazer. Third edition, entirely re-written, 113 pp. Philadelphia, 1891 (J. B. Lippincott Company).—Professor Frazer’s tables have already been found of much practical value by many workers, and in their present revised and improved form, their sphere of usefulness should be widely extended. 8. Materialien zur Mineralogie Russlands, von N. vy. Koxs- cHarow. Vol. x, pp. 225-351. St. Petersburg, 1891.—The part now issued forms the conclusion of volume x. It includes descriptions of jeremejewite, eichwaldite, columbite, also sup- plementary notes on euclase, zircon, topaz and other species. Ill. MiIscELLANEOUS SCIENTIFIC INTELLIGENCE. 1. Volcano of Kilauea.—In a letter of May 9th, Rev. E. P. Baker reports that after the eruption of March 6 the lava first appeared in the bottom of the empty basin of Halemaumau on the 10th of April. He visited the crater on the 29th of April and found the lake of liquid lava within it 100 or 200 feet across, and a blowing cone by the side of it which was throwing up globules of lava. The lake had a thin scum-like crust over it. While look- ing at the lake from the edge of the basin, 300 feet perhaps above the lake, the lava began to run off through an orifice beside the cone until the basin was nearly empty. The next day the lava had wholly disappeared. Again on the 6th of May Mr. Baker was down in the crater and found no liquid lava in the basin; but from the cooled lava on its sides it appeared that the lava had in the interval risen to a higher level than on April 30th. It thus seemed that the lake was rising and falling—rising through the accession of new lavas from below, and falling through discharges. The cone continued to throw up occasionally ¢ globules of lava. 2. American Geological Society.—The summer meeting of the society is to be held Monday and Tuesday, August 24 and 25, in the Columbian University, Washington, D. C., and will doubtless be one of unusual interest. The meeting will be preceded August 19-22, by the meeting of the American Association for the Ad- vancement of Science, and will be followed by the International mm hi | iW \r i a, \ Ke : , i) I, id rh janice ition 78 Miscellaneous Intelligence. Geological C Jongress, which meets August 26, and remains in session one week. The three societies will meet in the same building. The foreign members of the International Geological Congress are to be invited to read papers before the Geological Society, and their papers will be given precedence on the pro- gram. A number of excursions will probably be arranged. The local arrangements are in the hands of a committee, Mr. G. K. Gilbert, chairman. 3. International Congress of Geologists—5th Session, Wash- ington, 1891.-—Circular of information, No. 11, has been recently issued by the Secretaries, H. S. Williams and S§. F. Emmons, giving full information in regard to time and place of meet- ing (see above), program, transportation, excursions and hotel accommodations. Correspondence should be addressed to 8. F. Emmons, 1330 F street, Washington. 4, Physical Observatory y at the Smithsonian Institution, Washington.-—-Prof. 8. P. Langley announces (in a letter to the Editors, dated June 1, 1891) that there has been established at Washington, as a department of the Smithsonian Institution, a Physical Observatory, which has been furnished. with specially designed apparatus for the prosecution of investigations in radiant energy and other departments of telluric and astro- physics. The communication of new memoirs bearing in any way on such researches is requested, and for them it is hoped that proper return can be made in due time. Prof. Langley also states that he has resigned the titular directorship of Allegheny Observatory. OBITUARY. JHARLES ARAD Joy, for many years Professor of Chemistry Columbia College, died May 29 at Stockbridge, Mass. He was born in Ludlowville, Tompkins County, N. Y., Oct. 8, 1823. He was graduated from Union College in 1844 and from the Har- vard Law School in 1847. The same year he was appointed on the Geological Survey of the Lake Superior region under Josiah D. Whituey and Charles T. Jackson. Subsequently he went abroad and studied chemistry in Berlin, at Gottingen, and at the Sor- bonne in Paris. On his return he was called’ to the Chair of Chemistry at Union College. He held this position until 1857, when he was made Professor of Chemistry at Columbia College, which position he beld until 1877. Professor Joy’s labors were devoted to chemistry and allied branches, and he was the author of many papers especially of a popular character upon scientific subjects. When a student in Gottingen he carried on a series of researches on the combination of alcohol radicals with selenium and later he investigated the compounds of glucinum, the results of which were published in this Journal (1863). He also made contributions to the sub- ject of mineral chemistry. Professor Joy was one of the jurors at the International World’s Fairs of London, Paris, Vienna, and Philadelphia, and was a member of many scientific societies. ah ae No. 6 Miia Street, New York, revs s of Balances and Weights of Picsision for Ghent stayers, Jewelers, Druggists, and in general for every use z: ‘accuracy j 1s Ra ay ca, d to Chemisty, jrhaes sass cups Geography, eee = 08 Ass ci ite ‘Bilitors Ji P2 COOKE, JR, GEORGE L. GOODALE, and JoHN TROoWw- G of Cambridge, H. A. Newron and A. E. VERRILL, of Yale, and G. F. , of the University of Pennsylvania, Philadelphia. umes of 480 pages each, published annually in MONTHLY NUMBERS, — ournal per its jirst series of 50 SE asa pi ae he in 1845, and its The monthly seric« eom- ey ‘Twenty copies of each original communication are, if a iemer in advance eee author without charge ; and more at the aaa s expense, prov should be sent in two months before the time of issuing the number ap ‘tae aie are intended. Notice is : always Ae be given when communications ul peeacts price $6; 50 cents a number. A few sets on sale of the first nd second series. © aS : Ten: volume index paesiahen on Phand for the second and third series. The index g volume XXxXI to XL (3d series) was issued in January, 1891; ; price 75 cents. | Address the PROPRIETORS, a D. and E. S8. DANA, New Haven Conn. ‘. a OE Ts 5 ee ee 6% ibis iq s,s anaes * s i “%': i 4 x a Sa "Established ~ BENJAMIN SILLIMAN in 1818. - war AT LS 1) TH Fal BRS" OD 5 Te ot LU J # ant > 4 THE AMERICAN cs “ EDITORS JAMES D. ann EDWARD 8. DANA. 5 ASSOCIATE EDITORS a oressors JOSIAH P. COOKE, GEORGE L. GOODALE : anp JOHN TROWBRIDGE, or Camsripce. “1 Be. | Provessons H. A. NEWTON anp A. E. VERRILL, oF 2, . New Haven, Prorrssor GEORGE F. BARKER, or PxHiILavELpPHia ; & THIRD SERIES. a VOL. XLIL—[WHOLE NUMBER, CXLIL] - | ee No. 248.—AUGUST, 1891. l WITH PLATES II-IX. | “ | | NEW HAVEN; CONN.: J. D. & HE. §. DANA. | eae * 1891. | . ‘TUTTLE, MOREHOUSE & TAYLOR, ee 371 STATE STREET. | il lished monthly. Six dollars per year (postage prepaid). $6.40 to foreign ae ; of countries in the Postal Union. Remittances should be made either by orders, registered letters, or bank checks. Tei ges NOTABLE ARRIVALS OF MINERALS. is not surprising, therefore, that we can announce this month as haying witnessed the receipt not only of the largest accessions ever mae to our stock, but probably the finest also. Collected by our Mr. Atkinson. — ENGLISH MINERALS.—Hgremont Calcites, twins, groups, single crys-— tals, phantoms—the finest lot ever sent over. Bigrigg Calcites of anew form, extra choice and very cheap. Stank Calcites, good. Fluorites, over “ag 000, personally selected, every color and size—all cheaper than ever. ‘Specular Iron and Quartz, the best we have had. Dolomite, iri- descent; Aragonite groups; Barites, a great variety at rock-bottom — prices ; Witherites, singly and doubly terminated ; Bromlite, a splendid — lot; Tetrahedrite, iridescent ; Connellite, Henwoodite, Lettsomite, Tal- lingite, Bismuthinite, Ludlamite, etc. CONTINENTAL MINERALS.—Proustite, choice little specimens, also fine — little specimens of Pyrargyrite, Stephanite, Polybusite, Argentite and Acanthite, all well crystallized; Dioptase, magnificent specimens; — Topaz from Siberia, extra fine ‘erystals ; Aquamarine, some doubly terminated and highly modified ; Epidote from Tyrol, the finest brought to this country for many years: ; Aainite, fine lot; Sylvanite, xld.; Native Lead; Lehrbuchite ; Aikintte ; Byelkite : Zinkenite « Amalgam - Orangite ; Scheelite; Polianite; Herrengrundite; Szaibelyite ; Roselite ; Phosgenite ; Hvansite ; Leucochalcite;: Babingtonite: Pseudomalachite ; Pyrosmalite ; Polyarsenite; Pleonectite ; Cataplevite; Friedelite; Sar- kinite; Bismutosphaerite; Alexandrite; Retzbanyite; Messelite ; Calo- mel; Dewalquite ; Sarcolite ; Pucherite ; Phillipsite ; ‘Argyrodite ; Lieb-— igite ; Leuchtenbergite ; Pseudobrookite ; Sternbergite ; Strengite; Ines- ite; Allaktite; Hessite; Herschelite ; and hundreds of other varieties. ELBA MINERALS.—Rubellites, choice large lot ; Hematites, fine. Collected by our Mr. Niven. Minium, splendid specimens from Colorado. Azurite, a number of extraordinary crystals and groups. Velvet and Tufted Malachite, choice. tare Cuprite, some of the best we have had. Stalactites, incomparably beautiful, some stained green with Mala- chite, others of the exquisite Flos Ferri type, others covered with dis- tinct "crystals of Aragonite—an endless variety. Aguilarite, a few more small but choice specimens have been secured. This is the new sulpho-selenide of silver recently described. Pyrargyrite, Argentite, Embolite, Chaleanthite, Aurichalctte, &c., &c. Bastndsite and Tysonite are now en route from our Colorado collec- — tor, and other good Colorado minerals are expected. Yellow Sphenes from Tilly Foster Mine. One of the most interesting finds ever made in this country has recently come into our possession. The crystals (nearly all twins) rival the best European specimens both in quality and size. They are worthy of elaborate description and are meeting with a warm recep- tion from our best customers. The foregoing includes but a fraction of the important additions to our stock during the month. Blowpipe Minerals are being strongly added for the fall trade. 100 page Illustrated Catalogue, 15c. ; cloth bound, 25c. GEO. L. ENGLISH & CO., Mineralogists, 733 & 735 Broadway, New York. . ; = oe r “~~ Three Experienced Collectors have devoted their time during July to. securing new supplies of minerals for us.—Mr. Atkinson of our firm has been in Europe, Mr. Niven of our firm in Mexico, and our regular collector has been at work in the most promising Colorado localities. It \ jade 4 age ne Sait Reis tat Bt ae ~ THE AMERICAN JOURNAL OF SCIENCE \ [THIRD SERIES] } Oe i} Art. X.— Some of the | Jeatures of non-volcanic Igneous H Lyections, as ulustrated im the four “ Rocks” of the New Ny Haven Region, West Rock, Pine Rock, Mill Rock and 1 East Rock; by James D. Dana. With Plates [I to VIL. | iI THE observations on the igneous ejections of the New Haven | region here recorded and discussed were mostly completed il during the years 1879 and 1880, shortly after the publication i (in 1877) of a detailed topographical map of the region by the i U.S. Coast and Geodetic Survey, made under the special i direction of R. M. Bache. As this map is on the large scale a || of zo¢y7> OF about 64 inches to the mile, and has 20-foot contour lines, it afforded a very convenient basis for the record of geological facts. i A reduction of a portion of this map to a scale of two miles Hi to the inch, is presented on Plate II.* Excepting the hills in | the southwestern corner of the map, its whole area, even that Al of the New Haven plain, is underlaid by the Jura-Trias Red- i sandstone formation. (The excepted hills are part of the bor- | der of metamorphic schists that bounds the Jura-Trias region i} * This map is a portion of Plate II in the writer’s paper on the ‘‘ Phenomena }) of the Glacial and Champlain Periods about the mouth of the Connecticut Valley, i or the New Haven Region” (This Journal, xxvii, 113, Feb. 1884). The limit of i the New Haven plain is marked by a dotted line at the base of the hills, and the | contour-lines over it are omitted, the heights instead being given after a special i survey. The small nearly circular depressions marked on the map represent Kettle-holes.” The New Haven plain was of river-flood origin and it is pre- sented on the map with the outlines and height unaltered by the gradings for road-making, and by the making of mill-dams; and hence the map is a map of { the region of New Haven before 1640, as stated in its title. i {| Am. Jour. Sci.—THIRD Series, VoL. XLII, No. 248.—Aveust, 1891. 6 80 J.D. Dana—Features of non-volcanic Igneous Ejections. on the west.) The map shows the positions of the four trap ridges—more strictly trap-and-sandstone ridges—West Rock, Pine Rock, Mill Rock and East Rock, and gives their heights above mean tide. . These rampart-like elevations are now two to three miles from New Haven Bay; but they bear evidence of having been for a time the headlands of a much larger bay. The ridges are part of the Jura-Trias Mountain-range of the Connecticut Valley. (1) East Rock and West Rock are like the other north-and-south ridges of the range in their form, structure and direction, and. West Rock ridge after a course of seventeen miles, dies out just where the higher trap ridges of the Mt. Tom line commence, showing an interlocking with the rest of the system. (2) They consist of Jura-Trias sandstone with an intercalated sheet of trap (as the igneous rock is pop- ularly called). (8) The sheet of trap in the ridges has a rising inclination westward, or a dip eastward, like the associated beds of sandstone, the liquid rock having been extruded from a fissure or fissures situated somewhere to the eastward. (4) Asa consequence of these common features, denudation by water and ice has given to the New Haven ridges the features typical of the range,* namely, a steep western front, consisting of sandstone below and the harder trap above, a top of bare trap, and eastern slopes of sandstone, that is of the overlying sandstone. . From such common features the inference as to a common method of origin is natural. Still, as Professor Davis claims, it needs also other support for acceptance. We note also (4) that these Rocks are situated at the south- ern extremity of the Jura-Trias Mountain-range ; for the Con- necticut Valley and its Jura-Trias beds do not extend over Long Island. Instead of this, Long Island pertains to an east- and-west system of mountain-structure. Whether nearness in position to this east-and-west range has occasioned any of the features of the Rocks is an interesting question for con- sideration. 1. SUMMARY OF THE PRINCIPAL FACTS AND CONCLUSIONS. The facts.—The facts relate to the sandstone of the New Haven region as well as the trap; for the sandstone was broken through to give exit to the liquid trap, and it broke as such a sandstone would break. (1) The sandstone, as the rock is comprehensively called, varies from fine-grained to coarse, and beyond this, to a fine *In the writer’s paper on the Geology of the New Haven region of 1869, (Trans. Conn. Acad. Sci., ii, 4. 1870), he observes that “the sandstone mass with its intersecting dikes of trap constituted the block out of which the future New Haven region was to be carved by various denuding agencies.” J. D. Dana—Features of non-voleanice Igneous Ejections. 81 and coarse conglomerate, even cobble-stone-gravel conglomer- ate. When fine-grained and shaly it is not a firm laminated rock, but divides or crumbles readily to thin chips. The more massive kinds are usually traversed with fractures; and none has much firmness except where consolidated by heat from the trap-ejections, or the hot vapors produced thereby. Conse- quently, fissures made though the formation should have great irregularities, from irregular fracturing and the tumbling into them of masses of sandstone and large sections of their walls. _ (2) The thickness of the sandstone intersected by the fissures over the center of the New Haven region was at least 3000 feet, as proved by borings at a point half way between the bay and the west end of Mill Rock. Along the West Rock line the depth was probably less, as this ridge is within a mile and a half of the western metamorphic limit of the Connecti- eut Valley of Triassic time. Beneath the sandstone the fissures came up through underlying crystalline rocks, in which they would probably have great regularity in course, width and con- tinuity. (3) When the heat from the trap, or the hot vapors gener- ated by it, consolidated the sandstone, it generally made hard, durable rock of the coarser kind, but left the finer beds, alternating with the coarse, fragile and chip-making ; and this Was so, apparently, because hot vapor penetrates most easily the coarser beds for the cementing work. The heat, through the penetrating vapors, generally discharged more or less com- pletely the color of the beds it consolidated, producing an ash- ‘ray and brownish shade ; made in them steam tubes with blanched walls; produced blotches of impure chlorite, or epidote, and erystallizations of hematite and epidote, and less commonly garnet. But the finer beds that alternate with the coarse commonly retain, except perhaps for a few inches, their red color, and even have it deepened to a dark purplish red— as if by the reduction of some of the red coloring matter (oxide of iron) to magnetite. Moreover, the sandstone often loses all the old bedding. These varying effects from the heat have added much to the original irregularities of the beds. (1) Of the four Rocks, East and West belong to the prevail- ing north-and-south system, as already stated; the other two, Pine Rock and Mill Rock, to a transverse system. (2) In East Rock and West Rock the sheet of trap made by outflow from the opened fissure or fissures has a length westward of 100 to 500 yards. (3) The supply fissure, or its filling, the dike, descends be- neath the eastern slope with a large eastward pitch: the angle of pitch in the case of East Rock being about 50°. 82. J.D. Dana—Ffeuatures of non-volcanie Igneous Ejections. (4) In Pine Rock and Mill Rock, the trap is in dikes, there being no evidence of any outflow. Yet these dikes have in some of the outlets the great breadth of 150 to 300 or more feet. | (5) The pitch of these dikes is to the northward ; and its angle 18° to 40°—both characters of unusual interest. (6) Although neither East Rock, Mill Rock nor Pine Rock has a length exceeding a mile and a half, each has three or four distinct outlets of trap, separated by intervening sandstone ; moreover, there is wide diversity between the Rocks in the form and arrangement of these areas of extruded trap, as the map illustrates. (7) The trap of the several ridges, according to examinations by E. 8. Dana, is true doleryte, free, or nearly so, from chlorite and other evidences of interior alteration, and not at all vesicular. (8). Columnar fractures give the rock a rudely columnar structure, in which the halt-defined columns are four to eight feet in diameter. In the west fronts of the north and south ridges the rude columns have usually an inclination nearly at right angles to the mean dip of the associated sandstone—accord- ing thus with the usual rule: perpendicular to the cooling sur- faces. But among the columnar fractures, whatever the incli- nation of the columns, that plane of fracture or joint which is transverse to the sides of the dike or trap-mass and nearly ver- tical is the most strongly developed, and consequently the trap often cleaves into nearly vertical plates or laminz of great extent, much like a laminated rock. There usually is also a second easy cleavage-direction, nearly at right angles to the former so that rectangular columns sometimes come out with great prominence. (9). The outflows of trap have a floor either of an inclined layer of the sandstone or of edges of the upturned layers. The principal conclusions.—(1). The igneous eruptions of the New Haven region took place after the sandstone had been upturned ; that is, after the evolution of the Connecticut-valley mountain-range in this part of the valley had made great progress. (2). None of them were volcanic eruptions, for there was no center of action, no pericentric discharge of volcanic materials. (8). In the outflows from the fissures (those of East and West Rock) the liquid trap did not escape into the open air and spread over the surface, but entered between layers of the sandstone. (4). Moreover the flow was not by gravity into spaces that had been previously made, but a forced flow that opened J. D. Dana—Features of non-volcanic Igneous Ejections. 83 spaces or chambers for its occupation, the liquid rock thus lifting the overlying sandstone as long as the discharge was continued. By such means the sheets of liquid trap attained, in some cases, a thickness of 300 or more feet. This forcible opening and filling of a chamber in the sandstone by the up- thrust lavas, is a laccolithie process, it according with that of the typical laccoliths ably studied out and described by Gilbert.* (5). The intrusion of the flowing rock between the sandstone layers took place at comparatively shallow depths, where the pressure of the rock was not too great to prevent it. (6). It was favored, in each case, by the fact that the oblique fissure supplying the lava was inclined in the same direction with the layers of the uplifted sandstone—both in- clining westward, the dip being eastward. (7). The termination of a fissure in several outlets, exempli- fied in three of the Rocks, was largely due to the great ineli- nation and depth of the fissures opened through the weak upturned and faulted sandstone, and thence to great downfalls of the hanging wall. The same cause led to irregularities in the width and forms of dikes, and influenced the outlines and surface-features of outflows. (8). The course and dip of supply-fissures was not deter- mined by the foliation or bedding of the schists underneath the sandstone. 2. SPECIAL FACTS FROM THE SEVERAL ROCKS ILLUSTRATING THE ABOVE CONCLUSIONS. The ridges, Pine Rock and Mill Rock, containing simple dikes are first considered, and then East Rock and West Rock, which include dikes and outilows from them.t 1. PINE ROCK. The general form of Pine Rock is shown on Plate II, and still better on the following larger map.{ It is only three- fourths of a mile long and trends N. 67° E., or east-northeast. This small ridge has three, perhaps four, independent outlets of trap, A, BB’, CC’ and D. The first, at the west end, is a small dike 15 to 20 feet wide, trending north 20° west, and traceable for 220 feet. It dips eastward 25°, and thus proves * Geology of the Henry Mountains by G. K. Gilbert, 4to, 1877. + In justice to Percival, the author of the Report on the Geology of Connecti- cut of 1842, it should be here stated that there is scarcely an outlet or area of trap mentioned beyond which is not recorded on his map or described in his Report. ¢ The contour lines on this map, and also those on that of Mill Rock on page 87, are copied from the Bache Coast Survey map. 84. J.D. Dana—Features of non-volcanic Igneous Hjections. that it is not an outlier of West Rock, but part of the Pine Rock group. The other three are, more evidently, outlets from one great fissure. The width of the larger mass, CO’, is about 300 feet; and it is therefore one of the widest of dikes. The dip of the dike is 50° to 55° northwestward. This inclined i. 500 1600 ft. t Map of Pine Rock. Heights reckoned from high-tide level. Areas of trap with dotted outline. position (85° to 40° from a vertical) is given the dike in fig. 2, in which D I K E represents a section of it between its sandstone walls before denudation, and d7 K E, the same through the De 3. sore highest point of the Rock as it now is—or was before recent quarrying. The cross-lining gives the direction of the columnar fractures. The other figure, fig. 3, is a section through » on the map, where the removal of the sandstone of the southern wall (v, in the section) has left a depression ealled the Cave. (The sandstone of these sections is now concealed by the debris, and outside of this by the Terrace formation.) J. D. Dana—Features of non-volcanic Igneous Hjections. 85 The southern wall of the dike is the roof of the cave; the rock has the fine texture and fissured surface usual where it cooled in contact with the sandstone. Just above the cave, Inclined columus of Pine Rock, above the ‘ Cave.” where the exterior is removed, the surface is made up of the ends of rude columns. A _ profile view of these inclined columns from a point just south is shown in fig. 4.* At w, (see the preceding map) the north wall of the inclined dike is uncovered for a height of 50 feet, the sandstone having been carried off by the glacier. -At the eastern extremity of Pine Pock (near ©’), the trap of the north wall may be seen in contact with hard-baked’ sandstone. In the large quarry just south, the rock exhibits finely the transverse Jamination crossing the dike—referred to on page 82. The lamine incline 10° to 15° to the eastward, the dip being 80 to 85° to the westward. The surfaces of the plates are usually yellowish-brown with limonite for scores of feet from the summit, owing to the waters that penetrate from the surface downward and oxydize the iron of the rock ; but in the transverse joints or cracks, which are less accessible to the waters, there is usually a coating of stilbite and sometimes * From a photograph by G. N. Lawson, of the class at Yale of 1890; taken in December, 1890. + The shaping of the northern slopes of the Pine Rock ridge is a part of the same work of the ice; and the trend of the mass, like that of Sachem’s Ridge, (Plate II), indicates the direction of movement of the glacier. The same is true for the northern slopes of Whitney Peak and Indian Head. 86 J.D. Dana—Ffeatures of non-volcanic Igneous Hjections. of other zeolites, as chabazite, analcite, heulandite.* The dike has a few transverse courses of fracture containing prehnite and occasionally apophyllite, but no longitudinal have been observed. A sandstone ridge connects A and BB’, in which the rock is hard, and has the strike N. 40°-45° E., and the dip 45°S., becoming N. 30° E. and 30° to 35° in dip more to the west. It is mostly a coarse sandstone; but some layers contain stones 4 to 5 inches in diameter. Origin of the Features of the Rock. The existence of so many outlets of trap in the small space, and the irregular forms of the areas are unusual facts. BB’ is short, broad and blunt, shield-shaped; and CO’, is duck-like in shape, the irregular bosses at the northwest end (EE’) making the neck and head. These bosses are not in the line of the dike, and must be due to a local catastrophe. In view of the great inclination of the fissure, and its depth of 2000 to 3000 feet in the weak sandstone, a caving in of some part of its northern or hanging wall would be of extreme probability. Such a catastrophe would account for the stoppage of the out- © flow and the separation thus of BB’ and CO’; and such a stoppage of the up-thrust lavas would explain their escape by one or more extemporized outlets, and for the actual position of the apertures on the north side of the fissure ; and thereby for the making of the bosses. The obstructed lavas of the fissure may also have found exit in the western dike, A. The trap-mass D is possibly a resnlt of a second smaller catastrophe of like character; but its separation from CQ’, may be a result of erosion. | Another consequence of the great inclination of the fissure is the exposure of the dike of heavy trap to degradation through the removal of the supporting sandstone on the south side. Such undermining has produced the steepness of the southern front. And sea-shore waves or breakers were probably the chief agent—the shores being those of the broad center, or a central arm, of the New Haven Bay. 2. MILL ROCK AND THE WHITNEY RIDGE. Mill Rock is one mile distant from the east end of Pine Rock. Its length to Whitneyville or Mill River, is four-fifths of a mile. This small area, as is seen on Plate II, and better in the following larger map, has four independent outlets of trap— * The surface of the crust of zeolites is frequently tinged with the red iron oxide—which is a probable indication of heat as high at least as 200° F. dur- ing the formation of the minerals. J. D. Dana—Features of non-voleanie Igneous Hjections. 87 the western, AA’, the eastern, Bb’; north of the gap between these, a short narrow dike C, and farther north, the isolated area, D. The width of the first, AA’, (as measured at its west end) is 200 feet; of the second, 140 or 150 feet; of the third, 1 to 10 feet; of the fourth, 50 feet, the length being 150. The mass BB’ continues to Mill River where the surface of the country declines to tide level. But the trap does not stop here ; it crosses the river and extends on eastward, with an in- creased width, 180 feet, to the summit of Whitney Peak. The Whitney Peak dike belongs therefore to the Mill Rock region, although topographically part of the East Rock area. The trend of the Whitney Peak portion is 8. 68° E.; of AA’, S. 78° E. The mean course for the whole series to the summit of Whitney Peak is about S. 72° E. Map of Mill Rock, excepting its eastern extremity. Trap areas with dotted outlines. The dip or pitch of the main dike is about 72° to the north- ward, or 18° from the vertical. This inclination and the course of the ‘columnar fractures are well exhibited at the west end of the dike, A, and are repre- sented in figure 0. Besides the columnar frac- — tures at right angles to the walls, there are also longitudi- nal fractures in interrupted lines, parallel to the walls. Two are seen at the west end of the Rock and are indicated in the above figure. They are now mineral veins. The more south- Section of Mill Rock, west end. 88 J.D. Dana—Ffeatures of non-volcanic Igneous Ejections. ern one, a, contains chiefly prehnite, with traces of copper ore, and the trap along its course is solid or little altered. The other is situated about half way between the sides. It con- tains abundantiy the very hydrous mineral laumontite and the trap along it is decomposed; it contains also impure chlorite, and is fragile for a breadth of six to ten inches. A ean lanmontite vein, but nearer the north wall of the dike, is seen at Whitneyville, and also in the trap of Whitney Peak. The junction of the Whitney Peak part of the dike with BB’ takes place in the bed of the stream at Whitneyville, and is not now exposed to view owing to the dam and the build- ings below it.* The level of the trap beneath the dam is but a few feet above and below tide level. The height of the Whitney Peak dike increases eastward; first by a sudden rise of 100 feet, and then more oradually in the last 500 yards to 280 feet. Whitney Peak has a bold front to the eastward with sandstone at its base showing a sudden stoppage of the fissure in that direction; and at the same place it widens southward— not by overflow, as the precipitous eastward front and the depth of the trap shows, but through the opening of a trans- verse fissure. The Rock has a steep wall 70 to 80 feet high, on the north side of the summit for nearly 100 yards; but this is due to the removal of the sandstone by glacier action, expos- ing the north wall of the trap dike. The narrow dike C is about 110 feet long. It is situated in the face of a bluff of sandstone; and from the evidences of heat in the hardness of the rock, its mottled and light gray color in places, its steam tubes, and epidote, it is plain that the ejection determined the resisting power of. the sandstone against denuding agencies. The following figures represent two cross sections from the western half, and a map of the last 40 feet of the eastern half. At 65 feet the outflow is divided, * To the fact of this continuation I have recent testimony from Mr. Eli Whit- ney, who has superintended the constructions made there during the past forty years. Besides mentioning that the dam was built along the junction of the trap and sandstone, he says that below the dam for some distance, there is trap rock only, no sandstone outcropping there to his knowledge. The gun factory at Whitneyville was established there by his father, the in- ventor of the cotton-gin, in 1798, for the manufacture of muskets for the United States Army. J. D. Dana—Features of non-volcanie Igneous Hjections. 89 a narrow stream of trap (fig. 9), coming out above a layer of the sandstone 5 to 6 feet thick, the main part of the dike appear- ing below. This envelope of sandstone by trap continues for 30 feet, when the two parts come together again. The depth _at which the side stream goes off from the main dike is not known. ‘The inclination of the dike is mostly 25° to 28° (fig. 7) from a vertical, but at 45 feet from the west end it becomes 40° (fi. 8), and 10 feet beyond this, 30°. The sandstone of the Mill Rock region is of all degrees of coarseness up to cobble-stone conglomerate ; and no distinction is observable between that of the west and east ends. Origin of the Mill Rock features. The subdivision of the trap into its four masses may be ex- plained in the same way as that in the Pine Rock area. A downfall of the northern sandstone wall of the fissure, the hanging wall, would account for the separation of AA’ and BB’. Further, the obstruction thus occasioned to the great as- cending stream—its width 150 to 200 feet—would have forced upen passages to the surface for the discharge of the liquid trap, and thus may have been produced the small dike OC, situated near the fissure wall, and the remoter mass D. The irregularities of the little dike C, and the situation of both C and D to the north of the line of the dike, accord with this idea of a downfall of a part of the northern wall. The liability te such a catastrophe in a wall made of the rude sand- stone 3000 feet or more high, and having a large inclination, was augmented in both Pine Rock and Mill Rock by the tilted position and faulted state of the sandstone. The beds had already received their eastward dip of 15° to 25°, and breaks and faults innumerable that had been made in the adjustment to the new tilted position; it was therefore a tottlish structure overhanging a profound abyss. The fact here introduced that the eastward pitch of the sandstone was given it before the ejection of the trap is sustained by facts reported beyond. But an argument for it is afforded here: for if this eastward pitch were of subsequent origin, then the Whitney Peak end of the system should be the lowest. In- ae of this it is greatly the highest; the ridge slopes west- ward. It is possible that the fissures of AA’ and BB’ were, from the first, independent fissures to a considerable depth; for they are not in precisely the same line. If this were so, the above explanation, while in the chief points right, would require some modification. As in Pine Rock, so with Mill Rock but to a less degree, the northward pitch of the dike made it easy of degradation 90 SF. D. Dana— Features of non-volcanic Igneous Ejections. by sea-shore action. Through such means, beyond doubt, the part of it extending from Mill River westward for 300 yards, was reduced to a width above ground of 40 to 50 feet. This narrowing commences just west of the Pumping House of the City Water Works (p, fig. 5), and continues without inter- ruption to the river. It is part of the evidence of a greater New Haven Bay at some former time. Why the range falls gradually to so low a level at Whitney- ville, appears to be explained only on the view that less trap here came to the surface. I have elsewhere shown that it cannot be due to glacial removal. Neither is it probable that fluvial or marine waters have produced it. We have to attri- bute it to some condition existing or produced in the supply- fissures of eastern Mill Rock and Whitney Peak, at the time they were opened. Besides the dikes of Pine Rock and Mill Rock, there is another transverse dike of special interest which intersects the West Rock ridge just below the margin of Wintergreen Lake, or about one and a quarter miles north of the southern termination of the ridge and four miles from New Haven Bay. It descends the eastern slope of West Rock in an interrupted ridge, forms part of the southern bank of Wintergreen Lake, sinks to the level of the West Rock surlace at the summit, but stands out lke a buttress along the steep west front of the Rock. From the last feature I have called it for the past twenty years, the “ Buttress dike.” It extends south-westward through the metamorphic region of the towns of Woodbridge and Orange to the mouth of the Housatonic—as long since mapped and described by Percival. This dike has a pitch northward, amounting to 25° from a vertical in the part of it intersecting West Rock, but in that through the metamorphic rocks it is nearly vertical.* The strike of the inclined columns in the buttress portion is S. 30-32° KE. It is an example of a dike made subsequently to the cooling of another dike, that of West Rock. It has great importance in this connection, since it brings into the Jura-Trias system of mountain-movements a dike intersecting the metamorphic rocks outside of the Connecticut Valley, and one that branches off from the southern or New Haven part of the system. 3. THE EAST ROCK SERIES. The form of the East Rock area and its position between Mill River and the Quinnipiac, are shown on Plate II. Through * The rock of the dike is sparsely porphyritic; and the feldspar distributed through it in crystals a fourth to a third of an inch Jong is anorthite, as shown by G. W. Hawes (this Journal, IIJ, ix, 188, 1875). This character makes it easy to identify the several parts of the dike; it is the only case in which this mineral has thus far been found in the Connecticut Valley trap. Percival’s account of the Buttress dike and its extension southwestward is on page 399 of his Report. J. D. Dana—Features of non-volcanic Igneous Ejections. 91 denudation by the sea, rivers and ice, it has lost all of the sandstone formation that may have covered the summit, and for the most part that over its slopes above the 200-foot con- tour-line. The form of its upper portion is therefore largely that of the trap in its constitution—the hard rock that was most successful in resisting wear. This fact gives special interest to the larger and more detailed topographical map making Plate III, as will appear beyond.* To the north is Whitney Peak, which has already been described as the eastern extremity of the Mill Rock series. South of this and of a large area of sandstone, are East Rock and Indian Head, one in trap surface, but in fact the result of two independent outflows. To the south of Indian Head is Snake Rock, which also has its large trap mass, but is peculiar in having ridges of hard-baked sandstone that are higher than those of trap. The East Rock areas of trap here referred to are lettered on the map BB’, CC’C”, DD’. Besides these there is a more northern one, lettered AA’, which lies near the eastern foot of Whitney Peak. The trap-mass A A’—This northernmost mass, is about one hundred yards long. At its northern end it is only forty feet distant from the trap of Whitney Peak, and it is a question, therefore, whether it is not a part of the latter dike. But it is separated from it by outcropping sandstone, except where the interval is narrowest, and at this point there was until recently drained, a standing pool of water, a pretty good indi- cation that sandstone exists beneath, since trap is commonly too much fissured to hold water or afford springs. Moreover, the mass AA’ has the trend of the East Rock series; and, * The map of East Rock Park which is the basis of Plate III, was obtained from the Engineer department in New Haven, through the City Engineer, Mr. A. B. Hill. The roads of the Park from the termination of Orange St., around by the north to the summit of East Rock are lettered F, and the others E. These letters refer to two citizens of New Haven, Henry Farnam and James E. English, who liberally bore the expense of their construction. The topography is in part from the Bache Coast Survey map: but the accuracy of its contour lines was not sufficient for their transfer to the Park map. The heights are reckoned from high tide. The map is indebted to Prof. S. E. Barney. for the determination by leveling of the height of the highest point of East Rock, just south of the monu- ment (3584 feet) and also of other points on its south and east sides, and for that of the junction of the trap and sandstone on the west front near Orange St. bridge (155 feet). The height of the bolt at the Coast Survey Station he found to be 343 feet, and the height of the top of the first step leading to the terrace about the monument, 355 feet. (Prof. Barney’s figures are underscored on the map). The circuit road about the summit has a height of 320 to 350 feet; and the nearly parallel road on the east rises from about 216 feet near the quarries south of the sum- mit, to 270 near the junction of the ‘‘ Farnam drive” and “‘ English drive,” and thence declines northward to about 250 where it bends westward. The lettersS on the map indicate an outcrop of sandstone in the vicinity of junctions with the trap. In giving the topography of the Rock, the quarry excavations on the south side above a level of 216 feet are not introduced, it seeming best to represent the Rock in its original form. They are separately mapped on the plate. 92 J.D. Dana—Features of non-volcanie Igneous Ejections. besides, ledges of trap along the east side appear to indicate that the supply of liquid rock was from the eastward, like that of East Rock: On this view it is the northern mass of the East Rock series. Hast Rock proper.—The trap mass BB’, or East Rock, curves around from N. 25° E., on the north to N. 75° W. at the southwest extremity. Adding to it the Indian Head mass, it ends in an east-and-west dike, and is a complete crescent in outline. It has a bold columnar front, in which the columns incline about 22° from a vertical—the position, being, as-is usual, at right angles to the mean dip of the tilted sandstone. A view of the southwest front of the Rock is presented on Plate IV. Plate V_ illustrates the character and inclined position of the columns, and shows the contrast in the latter respect with Pine Rock. The upper 200 feet are of trap. The junction of the col- umnar trap with the sandstone is exposed to view at several points along the front. One such exposure may be seen when crossing the Orange Street bridge. The view in Plate LV, in which the bridge appears in the foreground, has the exposure half way up the front to the right. The height of the junction plane above mean tide at this place is 155 feet. Another is faintly indicated on the same plate directly below the Refresh- ment House; the height of the junction is there 150 feet. In other exposures of the junction-plane to the north, the height is less and becomes only 85 feet near the Rock Lane bridge ; and it is also less to the south being but 1324 feet at B', the southwest angle of the trap mass. Since the strike of the sandstone of the region is about N. 30° W., the sandstone (or the junction plane) has its greatest height, 155 feet, where the front has this direction; and the bedding of the sandstone in the section for this reason appears to be horizontal. The diminished height to the northward is owing mainly to the exposures. being at a lower level on the junction-plane because of the changed direction of the front, it becoming N. 10° E. near Rock Lane bridge. Through this interval the trap retains its thickness of about 200 feet. North of Rock Lane bridge the underlying sandstone is wholly covered by debris, so that the position of the junction-plane is doubtful. The supply of the trap forming East Rock came up, as the slope of its surface shows, from the eastward ; and it continues rising westward to the western and southwestern margin of the summit. The slope from the summit eastward and north- ward is gradual for about 300 yards, and then it pitches off at an angle of 45° to 50’ along the course of one of its dikes. The position of the dike, and thereby of the supply-fissure, is well exhibited at dc. A bare wall of trap, 50 to 55 feet in J. D. Dana—Features of non-volcanic Igneous Hjections. 93 height, descends at the angle mentioned. Since the surface there exposed became solidified against the northern sandstone _ wall of the fissure, the rock is of fine-grained texture and has an irregularly rifted aspect. The foot of the wall is about 200 feet above high tide. and from it the land, underlaid by sandstone, slopes off gently to the eastward. Since the direc- tion of this wall of trap is 8. 15° W., or that of the movement of the ice over this region in the Glacial era, the wall escaped the tearing action of the glacier, and so retains its original surface. Farther south, along a line from d to e, there is a similarly steep slope, but it is made of displaced blocks of trap. At its base there is a flat, terrace-like surface, which is near 200 feet above tide level. This steep slope appears hence to have been the course of the wall of another part of the supply-fissure. The flat terrace, although nearly 100 feet wide, is without stones over its surface of either trap or sandstone except in its southern portion, and there occur sandstone in fragments along with trap, and an outcrop of sandstone over trap at S. This fact and the occurrence of a perennial spring in this southern part (at the point toward which the two paths on the map, Plate III, descend) make it probable that the terrace rests on sandstone, and that this sandstone was that bounding on the east, the supply-fissure above referred to. But there is trap again to the east of this terrace, showing that the lower eastern slopes were supplied from a more eastern fissure. Along from c¢ to d, the trap of the outer fissure appears to have flowed over and coalesced with that ot the inner. Again south of e, the distinction of the two fissures cannot be made out. But the fact that the supply-fissures, one or both had a large inclination—not far from 45°—is evident from the very steep slope of the surface. Sections of the dikes of trap are nowhere exposed, and hence we are ignorant of the width of the supply-fissures. Judging from those of Mill Rock and Pine Rock, it may have been 150, 200 or 300 feet ; but it was possibly much less. The Outflows.—In East Rock, the trap which overlies the sandstone along the front, was that of outflows from the fissures westward between layers of the tilted sandstone. The fact that the columns of trap have a position at right angles nearly to the inclined layers of sandstone is believed to be good evidence of this intrusion of the melted trap. Fig. 10 represents the view that has ordinarily been held with regard to the relative positions of the trap and sandstone. According to it the trap left the dike to flow westward between sandstone layers having a dip of 20° to 25°. . A space was opened between the layers of sandstone which the liquid t 94 JS. D. Dana—Features of non-volcanic Igneous Ejections. trap filled. It is plain that this chamber could not have been so opened in advance of the inflow; for the hanging wall of the weak sandstone inclined 65° would have had no support. It is hence evident that the ascending stream of trap, forced Ideal Section of Kast Rock before the removal of the sandstone from the summit. along its course, opened a way between the layers; that a tongue of trap first entered, which would have been partly cooled against the cold rock; but the flow was kept up below this first intruding portion until the trap had all entered, the lifting of the overlying sandstone going on as it needed more space. This lifting would have brought a strain on the sand- stone that would have broken the connection between the lifted portion and that either side, to the northward, westward and southwestward. To the question, therefore, how far did the trap flow westward, the conditions reply: to the wall of such a fracture; and it may not have extended many rods beyond the present limit. The sandstone of the western wall has disappeared in the general denudation over the New Haven region, excepting a small part at the southwest angle, where a zigzag path (Z, Plate III) ascends to its top; the height of this sandstone is 185 feet, which is twenty-five feet above the base of the trap where highest to the northward, and fifty feet above that just south at A’. The locality of this sandstone and the zigzag path is seen on the right margin of Plate IV. The sandstone of the northern wall remains to a height of 196 feet at m-: the sandstone between Whitney Peak and East Rock is what is left of it. The dip of this sandstone at m, near the junction, is 30°, in the direction N. 73° E.; and the inclination of the columns of the trap just above is also 30°. The theoretical section of East Rock in fig. 10 represents correctly the fact of the intrusion of the melted trap between sandstone layers. But since the bottom over which the flow took place is concealed from view, it is not quite certain that the sandstone layer on which the flow began continued to he the floor to its western limit. Moreover, there is a large discrepancy between the pitch of the trap over the summit and that in the section. An actual section of the rock from J. D. Dana—Features of non-voleanic Igneous Hjections. 95 east to west (or more exactly E.S.E. to W.S.W. since this is approximately the direction of a transverse diameter) drawn to a scale, fig. 11, throws some light on these points. Section of East Rock, showing the correct profile. This section is essentially right in its profile, but more or less doubtful in its interior lines. The height of the upper surface of the outflow where it left the dike at d’ is 265 to 270 feet. It was not less than this; for we have this height for the top of the bare, unabraded wall of trap (adding the part of it under the Summer House west of the road). The length of the overflow to the present western front, is, as already stated, about 800 yards. The height of the western brow of trap in the section is 355 feet ;* and that of the bottom of the trap in the western front, 155 feet. These are facts; and the diver- gence here from figure 10 is very great. » Further, the mean angle of the trap surface over the summit is 10° instead of 22°, the mean dip of the sandstone. The latter dip is shown in the lines dv ; and if the floor had originally this pitch through- out, the thickness of the trap would have been about 450 feet, this being the distance on the scale of thé section between dn and d’n', while actually it is only 200 to 210 feet. The question arises: How was the lower:slope of 10° at- tained, and how the lessened thickness. Are they a result of wear by glacial or other methods; or was the present slope approximately the original slope of the outflow? A large amount of observation over trap ridges leads me to believe that the loss over East and West Rocks by abrasion has been small, probably not over 50 feet. The glacier, as it was shoved along, might easily have torn off columns from the front, but it would have made little impression on the exposed surfaces. Moreover glacial abrasion would hardly have left the highest points of the summit so near the western edge. If the outline of the summit approaches that of the original outHow, then—d being the lower limit of the trap on the front —a line drawn from d nearly parallel to the summit plane, * This is the height 80 feet north of the Summit Refreshment House, just west of the road, this being the highest point over this northern half of the summit area. AM. JouR. Sci.—THIrD SERIES, Vou. XLII, No. 248.—Aveust, 1891. q 96 J. D. Dana—Features of non-volcanic Igneous Ejections. - would probably represent the position of the bottom of the outflow. The line dl” l’/ has been drawn on this view. It supposes that the trap, on leaving the dike, passed between two layers of sandstone from / to /’ and that afterward it broke away the layer beneath it and flowed on, either over the edges or surfaces of layers as the conditions favored. The only spot where a section of the floor or plane of junc- tion of trap and sandstone, is seen, is at A’, the south-south- west corner of the trap-mass, by the road-side. There, for afew yards, the trap rests on upturned ledges of sandstone, and not on one continuous layer. The section is too short for any reli- able conclusion were it not sustained by facts from West Rock. The section, fig. 11, also represents the wnner and outer dikes described above, with the intervening (?) sandstone. The doubts with regard to the widths of these dikes and the area of sandstone have already been the subject of remark. Columns stand out boldly on the steep western front of East Rock. But they have none of the normal forms, for the angle between the most prominent faces frequently approaches a right angle, resulting from a combination of the plane of frac- ture at right angles to the trap-mass and another transverse. The direction of these planes varies along the course of the Rock on account of the curve in its outlines. At the quarry, on the south side of the summit, at the termination of the zigzag path Z, there is a fine display of broad surfaces in the two directions meeting nearly at a right angle. The courses here are about N. 35° E. and N. 55° W. The surface of one of them for many square yards is covered with rosettes of garnets and scattered minute crystals of magnetite, their faces brilliant in the sunshine. Along the whole western front of the Rock there is a remarkable predominance of planes conforming to its plane through all its changes of direction. This is apparent on Plate IV. and some of the right angles are seen on Plate V. The upper half of the columnar front (see Plate IV), down to a level of about 220 feet above tide-level, has columns four to eight feet in diameter ; below this the size is in general half less; and for the lower twenty feet above the sandstone, they are quite small. Indian FHead.—Indian Head is much lke a small edition of East Rock. The length of the outflow is 100 yards; the height 310 feet (813 above mean tide). A section made on the same principle with fig. 11 of East Rock is given in fig, 12. Indian Head stands quite apart from East Rock. The gap now separating them, where highest, is about 200 feet above high tide, and therefore nearly 160 feet below the top ——- J. D. Dana—Features of non-voleanie Igneous Ejections. 97 of East Rock and 110 below that of Indian Head, and proba- bly sandstone intervened for the greater part of this depth ; for the two Rocks face one another with steep slopes, as well brought out on the map, Plate III. These continue to be 12, Section of Indian Head. steep to the very foot of each, where they approach one an- other down the eastern slopes. Their bases are here in inde- pendent valleys, designated on the map by the letters E and I, separated by a low trap ridge, R, so that East Rock and Indian Head, although the trap extends over the surface of the gap from one to the other, are nowhere united at base. The eastward sloping valley, I, lying at the northeast foot, of Indian Head is continued in a westward sloping valley I’, at its north- western foot, and the two together define its outline. The low trap ridge R, between E and I, although consisting at surface mostly of blocks of trap, has a solid ledge in its lower part. It probably crosses the gap westward ; and the Summer House, near 201 on Plate III, may be on its western part. The valley E, at the southeast foot of East Rock, is perhaps, a result of glacial action; but why there should be two valleys side-by-side if erosion made either, is not explained. The trap of Indian Head rises from the bottom of the small valley just mentioned apparentiy in two half-separated streams instead of one even stream; but this feature may be a result of erosion. The eastern outline of the trap (see Plate ITI) is in a line with the eastern of the East Rock trap, indicating that the supply-fissure corresponded in direction with the outer and not the inner of the East Rock courses of fissures. The two Rocks, although alike in features, are to a large degree inde- pendent. Abrasion helped to deepen the gap between them, but more by the removal of sandstone than of trap. Indian Head is peculiar in having a long eastward projection from the southern end. It is described on a following page. The mode of origin ot the trap-masses of East Rock and Indian Head—by a forced flow of lava, opening through its uplifting action, a chamber in the sandstone for its accommo- dation—entitles the two to be called daccoliths. Through degradation, stripping them of the covering of sandstone, they stand side-by-side—a pair of laccoliths. ; 98 J. D. Dana—Features of non-volcanic Igneous Ejections. Snake Rock.—In Snake Rock, a broad mass of trap measur- ing about 900 by 450 feet in its two diameters lies encased in sandstone. The greatest height of the trap is but 160 feet, and that of the sandstone west of it over 200 feet. The trap covers the eastern slope of the Rock nearly to its foot, thus showing that the supply-fissure was on that side, as in other parts of the East Rock series, and also indicating by its steep- ness that the fissure was much inclined. At the south end of the Rock, in the yard behind the north corner of the Basser- mann house, at a junction of the trap and sandstone, the dip is about 45°; and this is direct evidence as to the inclination. The area of trap of Snake Rock has on the north the width of that of Indian Head; and the mass may hence owe its in- creased width northward to an outflow. If so, Snake Rock contains a half-emerged laccolith, its summit exposed, but the western wall of sandstone still standing and overtopping the trap. The sandstone shows everywhere the effects of hot vapors in all their varied forms, and before encroachments were made by a brewery there was a fine display of columnar sandstone in the southwestern bluff. Origin of the breaks in the Hast Rock series. The prominent breaks in the East Rock series are that be- tween Indian Head and Snake Rock, and that separating the small northern area, AA’, from the main East Rock mass, BB’. The Indian Head and Snake Rock masses, CC’ and DD’, approach one another bluntly within a hundred yards, and the area of sandstone between has parallel sides, as the map, Plate III, shows. In view of the steep pitch of the supply- fissure, a catastrophe to the western or overhanging wall is a most probable explanation of the break between them. The checking of so great a stream for a length of 100 yards might be expected to open escape-ways in some direction. The long eastern tail-like projection from Indian Head, C’O”, is the result of outflow along an east-and-west fissure. The pitch of the fissure, as the position of the trap shows, was about 25° to the northward. Its southern front is steep and rocky, the northern, gentle and grass-covered. It may be that this supply fissure was the escape-way then made, and the trap the part of the stream that would have occupied the interval had no such catastrophe occurred. The relations of the northern trap-mass of the series, AA’, to BB’ are doubtful. Yet it is probable that the trap of AA’ was ejected from the north end of one of the two East Rock fissures, or lines of fissures. The ledge of very hard sandstone which extends southward from near the south end of AA’, passes by the east side of the dike-wall dc; and it probably derived its position and its excessive consolidation and lost bedding to J. D. Dana—Features of non-voleanie Igneous Ejections. 99 a catastrophe that closed the fissure for the interval between them, which is only 200 feet wide, yet left it giving out heat, and generating volumes of hot vapors for the consolidating work. The East Rock masses of trap may therefore be traced to two ranges of fissures. The western was the probable source of the most northern area, AA, and of the summit portion of that of BB’ on East Rock. The eastern, contributed to the lower slopes of East Rock; and also through its continuation southward gave origin to the trap of Indian Head and Snake Rock. But for the accident to the hanging wall of the great fissure, the trap of Indian Head and Snake Rock would have made one continuous mass, and the columnar front of the former might have been continued over part of the present Snake Rock area. The areas of trap in the East Rock series narrow both to the south and the north. 4, WEST ROCK. The facts and conclusions relating to the West Rock region derive prominent interest from their pertaining to one of the long trap-ranges of the Connecticut Valley region. The area is represented on the accompanying map, Plate VI, from a survey made by the author with chain and hand-level in 1879 and 1880. The 20-foot contour-lines of the steep western and. southern fronts of the Rock and the geographical positions are from Bache’s Coast Survey map; but the other contour- lines exhibiting the surface features, which required for map- ping detailed measurements, are those of the author.* Featwres.—(1.) While the general course of the West Rock Range is north-and-south, the western foot of the blunt south- ern extremity bends round to an eastward course, and ends with north 30° east. The summit of the ridge also curves, in its last 500 yards, around to 8. 70° E. or nearly to east-by-west. Its height in this part is 899 to 405 feet above high tide, the geodetic station at the extremity being 399 feet. The eastern foot of the ridge has no corresponding bend. (2.) The trap of the Rock is a continued mass instead of being divided into several masses through a multiplication of outlets. But it has a large bay of sandstone, of triangular out- line, in its southeastern portion, which from its form is called the Triangle. (3.) South of the Triangle there is a prolonged hook-like point making the southeast termination of the trap. (4.) North of the Triangle commences the tiap of the west slope of the mountain. For a distance of 500 feet near the foot, increasing to 800 feet above, the surface of the trap is here elevated sixty to eighty feet or more above the level * The dotted line on Plate II is the north limit of the map, Plate VI. Heights C to Oa are plane-table results of Prof. H. A. Newton, from Bache’s 399 as base. 100 J. D. Dana—Features of non-voleanic Igneous Ejections. farther north. Moreover it is raised into rounded ridges, and some of these ridges have a high inclined wall on the south side. ‘The first of these walls adjoins the Triangle and has a height of seventy-five feet, a slope of about 45° and an even flat surface free from marks of abrasion. Another similar wall farther north is thirty feet high. The smaller troughs are mostly one to three yards deep. The angle of slope in the embossed surface between the 300-foot and 100-foot contour- lines is less than 17°; and in the surface north of it less than 14°. (5.) The long, hook-like point, above referred to, is not a simple ridge of trap, like that from an ordinary fissure, but consists, as seen along its northern side (Plate V1), of a series of rounded ridges which increase in height to the westward, like those of the elevated surface of trap on the other side of the Triangle. Moreover, all these wrinkle-like ridges, concave troughs and oblique walls, have a general parallelism. (6.) The embossed surface north of the Triangle has lost, through glacial abrasion, as a consequence of its elevation above the general level, all of the sandstone once covering it, even to the foot of the mountain, excepting small portions in two of the troughs. Farther north the sandstone remains in some places nearly to the 800-foot contour-line. (7.) The trap of the embossed area that was thus uncovered suffered little from the abrasion ; for the rock of the surface has the fineness of grain and other characteristics of the contact rock. This is true also of the trap of the southeast point. Moreover, in many places on this point below 300 feet, the trap contains imbedded fr ag- ments of the sandstone which fell into it while it was still liquid. The trap of other parts of West Rock ridge rarely shows evidence of abrasion below a level of 300 feet. On the contrary, above this level it has lost by abrasion the fine-grained, brittle crust-portion, and presents at surface the coarseness of crystalline texture that belongs to the interior of the mass. (8.) Another very important feature of West Rock is its affording a long east-and-west section through the breadth of a great trap range, exhibiting the contact-plane for several hundred feet of the outflowing trap and the underlying sand- stone, as described and figured beyond. The map, Plate VI, has the walls, troughs, and ridges of the surface shaded, to bring out better these features of the original surface of the trap. The southern front of the Rock has been made by degradation and hence has no shading. The southeastern point owes its straight outline on the south side to the quarrymen and the joints in the trap. The map shows what remained of the point in 1880. There is much less now. The Supply-fissure.—The inclination and width of the fis- sures supplying the liquid trap for the West Rock range are J. D. Dana—Features of non-volcanic Igneous Ejections. 101 undetermined. Exposures that will afford the facts are most likely to be found along the eastern base of the ridge. At one place where the surface of trap had been uncovered but not abraded, which was seemingly favorable for a safe conclusion, the slope was 25° to 30°, and suggested the angle of 30° for the inclination. But the trap at the place may have been part of the outflow, and not that of a dike. Observations along the eastern slope of the range farther north may obtain decisive facts. The Outflow.—The slopes of the higher parts of the West Rock ridge, the pitch of the columns of the western front, and the resemblance in features of West Rock to East Rock, lead to a like conclusion for the two, that the outflow was lacco- lithic; in other words, that the liquid rock forced its way between layers of sandstone, and made the chamber it occu- pied. The present thickness of the mass is nearly 250 feet. The overlying sandstone is to a large extent the weak, chip- making rock of dark red and purplish color already described. It is remarkable that a rock of so feeble coherence could have been lifted in the way mentioned. The questions suggested by East Rock here come up again: Whether the feeble slope of the surface from the west edge of the summit eastward to the 300-foot contour-line, and the small thickness of the trap, are due to abrasion, or whether the present conditions are nearly those of the original outflow. As the length of the outflow is nearly 500 yards, the mass, if forced up between layers dipping. 25° eastward, would have had a much larger amount to lose by abrasion than in the case of East Rock.* Speculation is here set aside by the actual east-and-west section of the Rock which is presented along its southern front, and is shown in part on Plate VII, from a photograph.t It exhibits the trap resting, to the eastward, on * The thickness does not admit of calculation, because the only datum besides the dip of the sandstone, is the height of the bottom of the trap over the sand- stone on the west front (about 200 feet); the height of the outflow where it left the fissure is not ascertainable. + The fine photograph was taken by M. W. Filley, of the firm of Bundy & Filley, of New Haven. The sandstone has here been exposed to view by the removal of the debris for macadamizing. The irregular line in the plate a third of an inch above the sandstone was the limit of the talus or debris slope; and the line below the sandstone is the profile of the quarry wagon road. Along the part of the section represented, the height of this road is ninety to one hundred feet. If the debris were wholly removed to the bottom of the slope, the height of the sandstone exposed to view would be, where greatest, over 150 feet. The photograph does great injustice to the view in the diminution of the vertical as compared with the horizontal scale, and also in flattening the angle of dip in the sandstone. 200 feet measured on the quarry road reaches from the eastern point of the sandstone section westward to within twenty-five feet of the line of the deep notch in the columnar front of the Rock (the place where the first section of sandstone ends); but this length applied vertically to the front above the road would make it only 180 feet in height, when in fact this height where greatest is over 300 feet. This error arises partly from the fact that the view was taken from the terrace opposite, which is only sixty feet high, but more from the error in an ordinary lens. 102 J. D. Dana—Features of non-voleanic Igneous Ejections. a tilted layer of the sandstone, the dip of which eastward is 25°. We are left to conjecture as regards the eastward and downward continuation of this layer to the supply-fissure (which the further removal of debris might perhaps uncover). But we know that the trap continues up this sloping layer for seventy-five yards from the commencement of the outerop. It conforms to the theoretical view of an outflow as presented in fig. 10, on page 94. But on reaching the end of the seventy-five yards, there is a change. The trap beyond rests on the edges of the layers in a. series of ledges of the sandstone. Moreover there is but little rise westward along the floor; for a line drawn along the top of the ledges would be almost horizontal, and have therefore near parallelism to the surface of the trap at the summit west of the geodetic station. The following figure represents the eastern extremity of the sandstone for a height of fourteen feet, together with the 13. base of the overlying trap. The rock is partly a hard-baked granitic sandstone, and partly the feeble shaly chip-making purplish-red sand-rock. The trap columns above the sandstone have in the lower part an inclination of 20°, approaching thus verticality to the surface of the sandstone; but, higher up the bluff front, there is a gradual change to 5°, which is the prevailing inclination.* The upper layer of the sandstone where uncovered shows a surface without breaks or much unevenness. A section of the sandstone, with the trap above, for the next seventy-five yards is represented in the following figure. The fact that the trap when melted flowed over the upturned edges is manifest. The chip-making rock constitutes much of the mass, and at its contact with the trap it is searcely changed in color or texture. The trap is far more finely columnar than that to the east over the single sandstone layer, and probably because moisture reached the trap freely from between the upturned layers. Other sections farther west are of similar * The angles of inclination here recorded are those presented to an observer in the front view of the rock here described. J. D. Dana—Features of non-volcanie Igneous Ejections. 103 character, excepting that the apparent dip is less) They may be followed westward along the quarryman’s road for 400 yards, when they begin to pass into the normal sections of the western front, that is, sections in which the lines of bedding are horizontal because they are in the line of strike of the sandstone. = ; fi thy, a reg a TT A ee ae | tN Me ‘4 1 £ mers / iy df y) 4 f: /, y ty! Yj f : Ue 4 / hil @ // | Jaa gh },/ 19 NG as pfig , yy (nt J ‘Wt er ALENT cial The question here arises: Did the flowing trap, owing to its movement and weight, wear off the layers of sandstone and so make the succession of ledges on which it rests; or did it escape from its confining cover of sandstone into the open air and cover in its flow the exposed ledges of the region. The former is probably the correct view. Had the flow become subaerial there would have been at once a decline westward in the level of its upper surface; for the level would have fallen as soon as the resistance from confinement ceased. There is no evidence of such a decline. From points on the summit close to the western precipice the surface for the first 300 yards has generally a slope eastward of 1 to 4, or 1 to 5, correspond- ing toapitch of 14° to 11°. The decline is eastward; not westward. Such a rise westward, even if only 5°, would be an impossibility except in a covered passage-way, that is, in the present case, one having a cover of the sandstone. Other evidence bearing in the same direction is afforded by the position of the columns along the western front, which pitch westward 15° to 20°. The summit slope eastward ot 14° to 11° is less than the dip of the sandstone, and favors the conclusion that the underlying sandstone was in many places torn up by th: heavily moving liquid trap, while left in place elsewhere. The floor so made consisted of alternations of wide strips that had the regular dip of the sandstone, with others abraded down to nearly flat and ledgy surfaces; and the former prevailed sufficiently to determine the direction of the contractional 104 J. D. Dana—Ffeatures of non-voleanie Igneous Lyections. fracture-planes or the columnar structure. A reduction so nearly to horizontality as that shown in the south front of West Rock along with parallelism in the profile of the sum- mit may not be common. West Rock teaches that the section of East Rock in fig. 11, p- 95, may be no exaggeration. Yet it is more probable that the original condition was intermediate between this position and that indicated in this diagram. Sections similar to that in the south face of West Rock may be looked for, with some probability of success, among many of the trap-ranges of the Connecticut Valley wherever they terminate in transverse sections. All that is necessary to ascertain the truth is to remove the talus of trap debris. Three miles east of New Haven (in East Haven) a section was opened in cutting for a carriage-road through the second trap ridge west of Saltonstall Lake; it is but a few rods west of the railroad station. The facts are in all respects similar to those of West Rock, as shown in the annexed figure. The trap covers a series of ledges of upturned sandstone, and shows no traces of displacement subsequent to its cooling. The sandstone is intersected by extensive nearly vertical frac- tures, whose surfaces, owing to friction, are scratched and polished; and the larger planes extend up through the sand- stone without any appearance of corresponding displacement in the trap. Moreover these polished slickensided surfaces have the white porcellanous coating common in the region; probably made by the grinding of the feldspar of the sandstone in the mutual friction of the walls.* * Atall the East Haven quarries, and in the ledges elsewhere exposed to view, these evidences of displacement and of much friction attending it abound. Frag- ments as large as the hand, slickensided on both surfaces and over planes of cross-fracture, are common; and so are walls of various inclinations hundreds of square yards in area. The sloping upper surfaces of the sandstone layers laid bare in the quarrying are sometimes polished and scratched in the direction of the dip for many square rods. There is abundant evidence of a vast amount of movement, though movement in a small way, during the progress of the upturn- ing in which the sandstone received its universal eastward dip. The section represented in fig. 15 has lost much of its original distinctness by the sliding down of debris from above. ~ J.D. Dana—Features of non-voleanie Igneous Ejections. 105 The trap of this ridge, at a higher level above the sandstone, is more or less chloritic and in many places amygdaloidal. Part of the amygdules are slender cylinders, two to three inches long and like pipe-stems in size, occurring often in groups—the result probably of the sudden vaporization of particles of liquid carbonic acid. In the railroad gap through the Saltonstall Ridge, the first west of Saltonstall Lake (‘*Pond Ridge” of Percival), the sandstone appears to lie in a similar manner unconformably beneath the western extension of the trap. but the section is now too much covered by debris for a satisfactory observation. Two miles east of the Saltonstall ridge in Branford, as de- scribed by Mr. E. O. Hovey,* the trap of a short range, the easternmost in this part of the sandstone region and near the gneiss boundary, overlies the upturned edges of the sandstone, and there is between the two rocks a layer of sandstone con- glomerate containing nodules of trap, which he attributed to the rubbing action of the flowing trap on the sandstone. These facts, ranging in this part of the Connecticut Valley over the whole breadth of the Jura-Trias formation, from the west side of the New Haven region where the trap is of the compact non-vesicular kind to the dikes of vesicular trap toward and near the eastern gneissic border, have great impor- tance in their bearing on the subject of the other Jura-Trias ridges. The more eastern are placed by Professor Davis among the ridges made of horizontal subaerial flows, ejected before the upturning of the sandstone; and the more western he has regarded as horizontally ejected and subsequently up- turned, although admitted to be interstitial intrusions. Neither of these conclusions are sustained by the facts which have been presented. The facts prove further that the era of disturbance or of the upturning of the sandstone was not due in any way to the ejection or heat of the igneous rock. The latter event, although so extensive, was simply incident to the disturb- ance; the upturning preceded the eruptions. Liffects of Obstructions to the outflow.—Although the trap of West Rock—that is of the southern part of the West Rock ridge—is not divided into several areas, other effects of obstruc- tions may be looked for, since the hanging wall of a large inclined fissure is sure to have its downfalls. The gaps or notches in the ridge indicate incipient division, and may be among the effects from such a canse. They may have been produced also by local narrowings of the fissure through horizontal or oblique movement of its walls, or in other ways; and it is a question whether the results of these two modes of origin can be dis- * This Journal, vol. xxxviii, p. 361, 1889. 106 J. D. Dana— Features of non-volcanie Igneous Ejections. tinguished. The deeper and more abrupt notches we should be disposed to refer to the former cause. As the Bache map of West Rock ridge indicates by its con- tour lines, within a mile and a quarter of the south end of West Rock, there are three gaps. Two are included on Plate II. At the first, the height of the ridge falls off sixty feet in the course of 500 yards. The second, situated 300 yards farther north, and called the ‘“ Judges’ Notch” because near the “Judges’ Cave,” is similar to the first in depth, but narrows more.down the western front. Half a mile farther north is the third, called the “ Wintergreen Notch.” It is one of the larger gaps in the ridge. Along the sum- mit, both from the north and the south, there is a descent of 100 feet, from a height of 440 feet to 340. Figure 16, from the Bache map, exhibits the facts.* The decline is gradual on the south side, but very rapid northward; in the latter direction the level of 460 feet is reached at the same distance from the cen- ter of the gap as 440 on the south. This third gap is probably one of those caused by obstructions to the outflow, whatever the fact with the others. The stream, in con- sequence of the obstruction, reached a height at the gap of but 3840 feet; but just beyond, the lavas that had been held back, made the abrupt rise in the ridge to 440 and 460 feet. The correctness of this explanation appears to be sustained also by the abruptness of the rise in the slopes east of the gap, as the contour lines in the figure show, and the great breadth of the nearly horizontal area farther east. It will be observed also that the summit of the ridge north of the gap is farther to the west than that on the south. (Arrows are inserted to make this dis- tinct.) It is so because any given amount of trap depends for its height on the distance it flowed westward up the inclined sandstone layers. It may be observed that not only the height, Wintergreen Notch. * The west side of the ridge in this part, as elsewhere, is the precipitous side, bold columnar above. Its upper 200 to 225 feet usually consist of trap, and the part below of sandstone; but the junction-plane at the Notch is concealed by trap debris, so that its actual height is not determinable. J. D. Dana—Features of non-volcanic Igneous Ejections. 107 460, but also 440 on the north side is to the west of 440 of the south side ; but the height of 440 to the north is probably pro- duced with a less thickness of trap. This notch is 300 yards south of the Buttress dike described on a former page; the position of this dike is shown on the above figure at 0. This example will suffice for illustration. Other gaps in the ridge occur farther north, but they are outside of the region here under consideration. Obstructions to the outflow of lava while it was making its way between the layers of sandstone are also possible through any cause that would prevent the lifting of any portion of the overlying rock. The area of the Triangle has been described as an area of sandstone within the proper limits of the trap range. ‘This sandstone was not lifted like the rest of the over- lying stratum. Instead of this, it remained in place for the most part, and hence, forced the liquid rock to pass to one side of it. The lava, mainly took the north side; and so the trap of that side had its surface raised in level above the rock north and became the elevated embossed area already described. The great sloping trap wall making the north side of the Triangle is the wall of an oblique fissure in the sandstone for- mation. Along this fissure—45° in inclination,—the sandstone of the south side, or that of the Triangle, lay unmoved or nearly so, while that of the north side was shoved up as the lavas came in below. Other walls, and the small ridges both north and south of the Triangle, are evidences of similar frac- tures, in parallel directions, with analogous results. The unlifted sandstone was in some way put under a strain that produced the parallel fracturing and movements. The origin of the southern or western walls of West Rock is sufficiently explained in the remarks on this asa respect- ing East Rock (page 94). The southern front of West Rock has a paieaieisess aspect. But in reality no columns stand out with the boldness they have in East Rock. The surface is mostly made up of the cleavage surface or joints that are in its plane; and where there has been quarrying, these joints have great width as well as height. 3. RELATION OF THE EAST-AND-WEST AND NORTH-AND-SOUTH FISSURES, AND THE ORIGIN OF THESE COURSES. _ These two courses of fissures are so locked together in the New Haven region that they evidently are results of one sys- tem of movements. They occur together in Pine Rock; and West Rock has the general trend of the Pine Rock ridge represented in the embossed area and the southeast point. 108 J. D. Dana— Features of non-volcanie Igneous Ejections. Mill Rock ends to the eastward in a south-southwest fissure, transverse to its main course which is apparently parallel to the adjoining part of the East Rock trap. East Rock com- mences with a nearly north-and-south course, but bends around to east-southeast. Mill Rock and Pine Rock are not neces- sarily synchronous in eruption with East Rock or West Rock, but they belong to one epoch of disturbance. The origin of these courses is not fully ascertained. I have long explained the north-by-east trend of West Rock, and of the other ridges of like direction to the north, on the general principle that the mountain-making forces of Eastern America operated over any part of the area, as a general thing in the same direction from Archean time onward, examples occurring in the Taconic and Jura-Trias elevations of the western half of New England. In accordance with this view the strike of the Jura-Trias should be that of the underlying erystalline rocks. It does not follow that a like dip prevails in the schists be- neath. It is true however that the predominant dip in them, and in the Jura-Trias fissures and bedding, is eastward. This last fact seems to favor the suggestion of Professor Davis that the foliation of the underlying schists has determined the courses of fissures in the Jura-Trias area. This sug- gestion would have support in the fact, were it not that in New Jersey, where the same is true as to the dip of the underlying schists, the Jura-Trias fissures and bedding dip westward. In the New Haven region, the idea of an accordance between direction of foliation in the schists and of fissures in the Jura- Trias finds no support. The West Rock ridge crosses the line of strike of the metamorphic schists two miles west of it at an angle of 20°. East Rock has an east-of-north course only in its northern extremity, and curves around through nearly half a circle. Pine Rock and Mill Rock cut across any probable course of foliation in underlying schists and do it on lines that differ 50° in trend. The origin of the east-and-west courses, which commence in the extremity of West Rock and continue to Whitney Peak, four miles, may have its explanation suggested by the remark on page 80. Or, it may be a consequence of the movement attending the production of the north-and-south fissures, and local to the New Haven region. The subject at present is one of conjectures. On account of the interest of the dynamical question here brought into view, I introduce another illustration of the facts from a transverse ridge only six miles north of Whitney Peak and Mill Rock. It is called Mt. Carmel. The ridge is only one and a half miles long. It is higher than those already J. D. Dana—Features of non-volcanie [gneous Kjections. 109 considered, the most elevated point being 736 feet above high tide.* But height means here, not larger accumulation of igneous rock or trap, but, simply, greater emergence above the sea-level; for this increase northward of height runs parallel with a like increase in the height of the metamorphic ridges just west; and it is continued, at a diminished rate, into Massachusetts. Mt. Carmel has resemblances to Pine Rock. Its mean course is E. N. E.; and a north-and-south trend exists in its western part. but the north-and-south portion in Mt. Carmel is a large feature in the ridge and has direct continuity with the east-northeast portion. The ridge is divided by a very deep and open gorge, into an eastern and a western section. The gorge is often called the “Neck,” and the high summit adjoining it on the west, the “Head” of the “Sleeping Giant ”—a name suggested by the form of the ridge as it appears lying on the northern horizon. Both have northern and southern slopes of sandstone, the southern going about half way to the top above its base, and the northern reaching a greater height. The western section, while high and massive at its eastern extremity, falls off rapidly to the westward, and in half a mile is reduced to a narrow trap ridge not exceeding 100 feet in height above the adjoining country. Through this part within 300 yards, pass Mill River, a north-and-south carriage road (N. 20° W.) without change of grade, and, a few rods farther west a railroad. Along the railroad, and between the carriage road and the river, the course of the trap changes from about north-and-south to N. 10° E.; and as it crosses the river to N. 20° E. Thence it continues on to the summit, widening and increasing rapidly in height and curving still farther eastward. At the section in the railroad cut, the trap is seen resting on its south wall of sandstone, the wall dipping about 45°—appar- ently indicating that the dike has this pitch. Between the carriage-road and Mill River, the north side of the trap has in many places a westward dip of the same angle, confirming the conclusion from the railroad section as to the large dip of the fissure. It is thus proved that the western section is a con- tinuous mass of trap of gradually changing course and mag- nitude ; and that it is strictly “transverse” in direction only along itseastern end. It isa dike to the westward and probably so throughout. The eastern section is made one continuous mass of trap by Percival, and one also with the western portion. It is divided * According to the leveling of two parties under Mr. Bache. 110 J.D. Dana—Features of non-voleanie Igneous Hjections. from east to west, as he states, by a valley, and in the valley there is a spring giving out a streamlet which flows northward. There are gaps in both the southern and northern sides, divid- ing them into a series of elevations. These elevations are indicated on Percival’s map, so as to look as if he regarded them as separate dikes; but this is contrary to the description in his Report. I have looked for sandstoné in two of the gaps of the south side, east of the “neck,” and have found evidence in each that the trap is continuous, and descends in these gaps nearly half way to the base of the mountain. In the east-and- west valley the spring and streamlet are probable evidence that there is sandstone beneath; and on this ground, it may be that there are, in this eastern part of Mt. Carmel, two parallel east-and- west dikes, Mt. Carmel appears to be a combination of dikes, without the “buried voleanoes” supposed to exist there by Professor Davis. In the view from the west side of Mill River there are in sight nearly 600 feet in height of massive trap, having no subdivision into sheets or layers, and nothing to suggest the idea of lava-streams in the depths below. The union in this small ridge of approximately north-and- south and east-and-west courses is further proof of their mutual dependence in the system of movements attending the Jura-Trias mountain-making of the Connecticut Valley. But its origin remains unexplained. Concluding Remarks.—A review of the principal conclu- sions in this paper is given in its introductory remarks (page 82), and a recapitulation here is therefore unnecessary. The reader may have been led to the idea that the author would make the West Rock Ridge typical for other ridges of like features in the Connecticut Valley region, in disagreement with the conclusion of Professor Davis who holds that in the case of most of these ridges, if not of all, the trap was poured out in one, two or more horizontal sheets, separated, and over- laid horizontally, by beds of sandstone, and that the whole was afterward faulted and tilted so as to make the ridges. The author acknowledges that he is inclined to make the conclusions he has reached general. He, however, admits that he has not made the structure of the other ridges of the valley a special study. He believes his observations sufficient, however, to authorize the statement that a more intimate knowledge of the facts is required before any adverse views can be regarded as established. LR. T. Aili—Ouachita Mountain System, ete. PEt Art. XI.—Wotes on a Reconnarssance of the Ouachita Moun- tain System in Indian Territory ; by Rop’r T. Hruu.* Synopsis.—General topographic features of Indian Territory including Oklahoma. The northern, middle and southern belts. The middle or mountainous belt. 1. The Eastern or Arkansas-Choctaw Division. 2. The Central or Chickasaw Division. 2a. The Wapenucka Sub-division. 2b. The Tishomingo Granite. 2c. The Arbuckle Mountains and Washita Water Gap. 3. The Wichita Divis- ion. Partial record of history recorded in the Ouachita System. LITTLE has been written concerning the geography and geol- ogy of Indian Territory, and the writer presents this prelim- inary paper in hope that it will direct to that interesting region more careful and detailed study. Topographically Indian Territory, especially its southern half, presents a great diversity of mountain, plain, forest and stream. Within this area is found the extension of nearly every topographic unit from the Missouri-Kansas region on the north to the Texas on the south, from the Great Plains of the west to the forests of Arkansas on the east; there are also many unique characteristic features of the region itself. The territory may be provisionally divided into three par- allel east and west belts, each containing a marked diversity of geologic structure and corresponding topographic expression. The northern or Cherokee-Oklahoma belt includes the coun- try north of the Canadian; the greater part is prairie with spots of timber decreasing in density toward the west. This belt may be sub-divided into three districts; the eastern or Cherokee, the middle or Oklahoma, the western or Arrapahoe. The Cherokee division, with the exception of a small area of Ozark hills in the northeast corner, is mostly composed of Car- boniferous rocks with an undulating topography similar to that of southeast Kansas. The Oklahoma section is a typical red bed region in its western half, with undulating prairies and soft disintegrating structure. The Arrapahoe division is the ragged eastern border of the great plains country, with its characteristic fresh water deposits of sands and grits occupying the flat divides, as originally described in the adjacent west Kansas region by Dr. J. 8. Newberry and more recently by Pro- fessor Robt. Hay.t These plains are the newest or culminating formation in western Texas, Kansas and Indian Territory ; they are now slowly receding westward because of the head water erosion of the streams that indent this eastern border, * To Mr. James S. Stone, of Newton, Massachusetts, the writer is greatly in- debted for his faithful assistance in conducting this investigation. Also to Mr. W.L. Davidson, a student of the University of Texas. + See Bulletin 57, U. 8. Geological Survey. Am. Jour. Sci.—THIRD SERIES, Von. XLII, No. 248.—Aveust, 1891. 8 LR. T. Hill—Reconnaissance of the 112 ry mi => AW iy, ) SS Ds | RX Ginny) (a None bp a os Py a TEA ‘s FURS ee a : shaq EY a2unigohhnunysy, == 20) Bag ‘SIUPPSIULNT UMIANITO SP. ENS $7? tay ae. p)S7097793-4). aay (H) 442722127, Lay Yso W) 22APLY IDA J )ALDULZZON) Bry FOC BSAC GV PFLYIDNG Kay, sw FH Wy, WG yuu (I) S72041asMU0QTD (S) $27 B9 “ ALAMO'T nll %y, Ouachita Mountain System in Indian Territory. 118 and in this manner the underlying structure and topography are revealed. The northern belt of Indian Territory distinctly belongs to the Kansas division of the United States and the writer leaves its further description to St. John, Cragin, Hay and Jenney, investigators who possess more facts con- cerning its geology. “The middle or mountainous belt lies south of the Canadian- Arkansas River. A mountain system traverses it from east to west and marks the great barrier between the upper Missis- sippi Valley and the Texas-Arkansas regions of the United States.* Toa description of these mountains this paper is mostly devoted. The third and southern belt, the description of which must be left to a future paper, includes the region between the mountainous belt and Red River. It is the northern termina- tion of the Texas region of the United States. It includes many topographic and geologic features which are the result of neozoic sedimentation against the southern border of the mountains. The Mountain Region of Central Indian Territory.—With the exception of the Ozark hills in the extreme northeastern corner, the mountains of Indian Territory are the direct west- ward continuation of the Ouachita system of mountains which has been described+ as the mountainous area between Hot Springs Arkansas and the Staked Plains of Texas, including the various points known as the Poteau, Seven Devils, San Bois, Shawnees, Jack’s Fork, Black Fork, Winding Stair, Sugar Loaf,{ Cavenal, Stringtown Hills, Limestone Ridge, Potato Hills, Arbuckles, Wichitas, Navajoes and other moun- tains. These mountains are south of the Arkansas-Canadian drainage and must not be confused with the Ozarks of south- western Missouri. Dr. J. C. Branner’s coming reports will doubtless give us needed light on this relation. The mountain belt has three distinct sub-divisions: (1) an eastern or Arkansas, (2) a central or Chickasaw, (8a) western or Wichita. Its areal extent may be compared to an arch whose apex is southward, as marked by the course of the Canadian, Arkansas and Red River drainage ; its eastern mem- ber in Arkansas and the Choctaw nation is a forest area of vertically folded Carboniferous shales and sandstones resem- bling the Appalachian country ; the western member in the Chickasaw and Comanche nations, is a mostly treeless region and consists of low folds of hard white and blue Silurian lime- * See this Journal, April, 1889. + Arkansas Geological Survey, 1888, vol. ii. The geology of Southwestern Arkansas, by Robt. T. Hill. ¢ Near Fort Smith, not the Cretaceous butte of the same name east of Caddo. 114 ft. T. Hill— Reconnaissance of the stones. and eruptives; the keystone or central Chickasaw region, consists of an area of granite and Silurian limestones. 1. The Lastern or Arkansas-Choctaw Division.—The north- ern two-thirds of the Choctaw nation and the northeastern Chickasaw country are a direct continuation of the mountains and geologic features of west-central Arkansas. This region consists of numerous timber-covered ridges varying in altitude from 2700 feet along the Arkansas line to 1200 along the Missouri, Kansas and Texas railroad. The ridges are usually elongated, timbered, devoid of sharp peaks and owe their pres- ent form to the unequal erosion of the exaggerated structural folds. The general trend of these mountains, corresponding with the strike of the folds, is south of westward, but often, as seen near Stringtown and along the Kiamitia River, it is nearer north and south. The ridges consist of sandstones, clays and shales apparently of the Carboniferous period, but further investigation may reveal older rocks. The rocks occur in numerous parallel, overlapping folds, which are nearly vertical in the southern and central portion of their extent, but become horizontal along their northern outline. a Scale of rrtles 2 Section north and south across Red Bird Mts.. showing relation of Mountain folds to Cretaceous Prairies. The Saint Louis and San Francisco railroad, from Fort Smith, Arkansas, to Paris, Texas, passes through the heart of the region, and the type structure, as seen along this route, consists of vertical eastward folds dislocated by another and later movement, as seen south of Tushka Homa, the Choctaw capital. This road follows for miles the water gap of the Kiamitia River, which apparently flows in an anticlinal valley: A hundred miles west of this railway, the Missouri, Kansas and ‘l'exas road affords another parallel north and south section of the mountain system, but owing to the gradual cessation of timber and decreasing altitude entirely different scenic effects are revealed. The latter road follows the valley prairies between the mountain ridges, which here have the contour and altitude which, in Kentucky, would be ealled knobs. The railroad follows the strike of the structure from Atoka to Limestone Gap. The differences in elevation are the result of unequal weathering of the crumbling shales and the more resisting sandstones and limestones, the former being treeless valleys while the latter persist as mountainous ridges. (Fig. 3.) Timber grows upon the sandstone outcrops while the Ouachita Mountain System in Indian Territory. 115 ' prairies occupy the more compact clays of the valley. Even where the vertical outcrops have been eroded to a level plain, the alternations of sandstones and clays can often be traced for miles by the timber which follows the sandstone outcrops in narrow ribbon-like parallel belts. (See fig. 2.) The northern half of this area contains coal strata whose extent and known occurrence are indicated on the map. An admirable paper upon the structure of these coal beds has been published by Mr. H. M. Chance.* Mr. Arthur Winslowf has equally well defined them in Arkansas. Mr. J. T. Munson of Denison, Texas, has much unpublished information con- cerning the formation of this region, and to him the writer is indebted for his invaluable assistance and data. The coal fields, for which the name Fort Smith-McAllister area 1s most appropriate, are of great commercial importance, for they are the chief source of fuel supply for the Arkansas- Texas region. These extend along the northern border of the mountains and are terminated on the southwest by the Silurian and granite field of the Tishomingo district which are an apparent barrier between this and the Texas-Ardmore coal 2. S 3 n 7 PIII GF, ‘rr Carborulerous ee) Se? L Fielderberg.? ? ? Seale of miles F Section north and south through Woodford, showing structure of Prairie and Mountaio. Continuation of fig. 1. field, the fuel of which is of an entirely different character and should not be confused with itt geographically, structurally, or economically. Mr. Chance has published a section of the rocks of the eastern division. He estimates at least 8500 feet of coal- bearing strata, but the total thickness of the Carboniferous and Permo-Carboniferous, as seen in the folds near Ardmore, is greater by the addition of the uppermost or Permo-Carbonif- erous which here has a thickness of several thousand feet. The most marked feature of these mountains is the ex- cessive, compressed and vertical folding which the whole region has undergone, and the displacement of these folds by a lateral dislocation which has squeezed them into S-shaped flexures. So excessive is this folding that every stratum in * Geology of the Choctaw Coal Fields by H. M. Chance. Transactions Ameri- can Institute of Mining Engineers, Feb., 1890. + Arkansas Geological Survey, Report for 1888, vol. iii. { The writer is inclined to believe that the greater excess of ash in the coals of the More horizontal Texas region is due to the calcium carbonate and other im- purities deposited in the joints during their long submergence beneath the Creta- ceous seas, while the McAllister coals have remained above water. 116 Rk. T. Hill— Reconnaissance of the the mountain region south of the coal fields can be, said liter- + ally to be standing vertically as shown in the figures. This system of folding is complicated and the writer has not had time for the minute study necessary to interpret it. In gen- eral, two great trends or strikes are conspicuous, the first and oldest is about 25° south of west; this is frequently dislocated by an apparently later movement resulting in northeast and southwest trends, all of which are accompanied by overlapping and lack of continuity.* The direction of the folds has a marked effect on the political features of the region, all lines of transportation and public highways practically following the valleys of erosion in the trend of the folds. The proof of two great disloca- tions of the Carboniferous strata is found in the mountains north of Atoka and in Limestone ridge where the vertical folds of the first epoch are defiected by S-shaped dislocations into the southeast course. Of the many illustrations of this folding one of the finest is found in the peculiar limestone ridge which extends from near Lehigh to Lime- stone Gap and eastward. This is the principal limestone stratum of the Carboniferous system ; it occurs at the base of Mr. Chance’s section. : It consists of about 200 feet of {mam gine, © assive blue limestone and dolomite Limestone Ridge, showing deflected Somuang pee fe eth tote ealdes Wapenucka via. Lehigh to Lime- stone Gap, thence eastward to the St. Louis and San Francisco railroad, it forms a sharp ridge rising 100 feet above the adjacent valleys, a plan and cross section of which are given in the accompanying figure (8). The Missouri, Kansas and Texas railroad, between String- town and Limestone Gap, follows the valley east of this ridge ; at the latter place a tributary of the Red river has cut through the ridge which, from this point, trends eastward as shown in SSS LS * Dr. John C. Branner, on page 30, vol. i, of his report, has previously expressed an opinion that in Arkansas these folds are of overlapping rather than of con- tinuous strike, as stated by Comstock in the same volume. Ouachita Mountain System in Indian Territory. 117 Mr. Chance’s map. Several sigmoid or S-shaped flexures occur along this section, and, also, in the sandstones of the Coal Measures of eastern Indian Territory and across the Territory to the Arkansas line. The southern border of this old system has been degraded* by the shore lines of the ancient Cretaceous and Tertiary seas which overlapped it and planed it northward for many miles. The vertical edges of the planed off strata are buried beneath the Cretaceous sediments as shown in my former section along the Arkansas-Texas line, resulting in the complete interment of the Carboniferous system southward, throughout the great central denuded region of Texas where the only exposures of Carboniferous rocks are through erosion of the overlying Cre- taceous layers. The structure of these mountains is of the Appalachian type, and Mr. Chance says that ‘“ topographically and structurally the Choctaw coal fields represent in miniature the features of the anthracite regions of Pennsylvania.” 2. The Central or Chickasaw Division.—In the northeastern part of the Chickasaw nation the continuity of the Carbonif- erous rocks is terminated by an extensive area of Silurian lime- stones, which, in turn, are succeeded southward by underlying granites whose exact relation to the complicated Coal Measures is not determined, but which are exposed by the erosion of the latter and are.unconformable beneath them. 2a. The Kastern or Wapenucka portion of this area is inter- esting, but little explored. It lies west of Boggy station along Delaware Creek at Bill Jackson’s ranch, and near the quaint old Chickasaw academy of Wapenucka. There is a series of low limestone hills—apparently remnants of anticlinal folds— along whose strike flows the Delaware creek. In places these limestones resemble the blue Silurian limestone to be described in our discussion of the Arbuckle Mountains, but they are more horizontal in outerop. In the collection of Mr. J. T. Munson, of Denison, who first called my attention to this in- teresting region, are fossils apparently Silurian in age Ortho- ceras as Brachiopoda, from Bill Jackson’s ranch on the Dela- ware. Crinoidal limestones of Carboniferous age are the prevalent rocks and were collected near the academy at the southern border of the district, and the sandstones of apparent Carbon- iferous age and shales of that age begin there again. A single specimen of Favosites, of Silurian age, was collected from one * Principal Events in North American Cretaceous History as revealed in the Arkansas-Texas Region, by Robt. T. Hill. This Journal, April, 1889. + Professor Alpheus Hyatt, to whom I sent this specimen, says that he thinks there is little doubt that it is a fragment from the Hudson River group. The Or- thoceras being closely related to one found at Cincinnati, and the brachiopod being probably Orthis testudinaria. 143 3a LR. T. Hill—Reconnaissance of the of the Delaware Mountains near the Hudson River limestone. The Delaware Mountains proper are a few long limestone ridges and detached buttes in the beautiful valley of Delaware Creek. Seven miles west of the academy, near Bill Jackson’s, they are composed of limestone underlaid by the above men- tioned Favosites sandstone—a porous gray quartzite with an occasional patch of limestone. The buttes are peculiarly dis- torted, their strata being disturbed at a very slight angle in many directions, which may be compared to the uneven curva- ture of a saddle. The Delaware mountains were mentioned by Mr. Jules Marcou, who followed the old Fort Smith and Fort Washita trail which passed by them. He referred them to the Sub- Carboniferous or Mountain Limestone.* The scenery in the Valley of the Delaware is exquisite, the contrast between the - low rounded hills and the extensive valleys with their peculiar buttes present a restful and varied landscape. The region promises rich scientific treasures to some future student who has time and facilities to work out its structure and history. It was impossible to trace the relation of the Wapenucka dis- trict to the Arbuckle Mountains to the westward, owing to dangers of exploration in a country where geologists are not welcome, but there is evidently a close connection if not con- tinuity between them. , 2b. The Tishomingo Granite.—In the heart of the Chick- asaw nation south of and underlying the Wapenucka limestone district is an extensive granite area. This is the central divis- ion of our mountain region. It isa triangular area of sandy prairie land with low rounded granite hills and undulations, lying between the Santa Fe and M. K. and T. railroad and running east and west from Boggy depot to six miles west of Tishomingo, and northeast to Mill Creek and beyond. The granite is well displayed two miles southwest of Boggy station; in Pennington Creek; at Tishomingo and other places. At its eastern outcrop it is composed of red feldspar, white and black mica, quartz and hornblende with numerous pegmatitic veins. Its composition and occurrence is nearly identical with the Burnet Texas granite, and it is unlike the igneous rocks of the Wichitas to be described later. In the western part of this area the feldspar is albite. There are numerous dikes of black rock intersecting this granite specimens of which from Pennington Creek have been sent to Professor J. F. Kemp for study. Concerning these he says: ‘‘They are a typical dia- base. They are mostly idiomorphic plagioclase crystals, doubt- less labradorite from the extinction angles, irregular greenish augite and a little magnetite. They show the so-called ophitic * Geology of North America. Ouachita Mountain System in Indian Territory. 119 structure of diabase in a very marked degree.” The dikes run west 20° S. and are seen at the crossing of Mill Creek road and Pennington: Creek. The northern margin of the granite area is overlaid by hard metamorphosed, sub-horizontal Silurian limestone of the same cherty and flaggy lithologic aspect as the Upper Potsdam rocks of Burnet County, Texas, but I could find no fossils. Carboniferous rocks cover its eastern point at Boggy station. Its southern border was the sea-shore of the ancient Trinity and other Mesozoic and Cenozoic seas and is buried beneath the Trinity sands. The western border is covered by Silurian and Carboniferous rocks.* I saw no evidence that this granite was of later age than the oldest of the Paleozoic rocks which rest upon it. 2c. The Arbuckle Folds —West of the Washita River the mountains again present a new and entirely different aspect. An elongated mass of low rounded barren limestone folds stands about 500 feet above the plain and extends east and west, between Wild Horse Creek and the Washita River for about forty miles, forming an almost impassable barrier for wagon travel. They are composed of folds of hard Silurian limestones. The trend of the mountains—north of west— corresponds with strike of the folds, but is opposite in direc- tion to the prevalent trend of the Choctaw-Arkansas division. These folds are the’hard persistent core of the structure, the softer and exterior Carboniferous layers having been eroded to the level of the Ardmore prairies. (See figures.) West of Duncan the limestone hills are buried beneath the red beds for twenty miles, but again appear in the neighbor- hood of Fort Sill forming a low ridge north of ‘and parallel to the Wichita Mountains, as is explained later. The Arbuckle Mountains constitute a great and wonderful development of the Silurian system, although this has not been hitherto appreciated, and afford a superb example of folded structure. ‘This folding is beautifully shown in the valley of the Washita which has cut a deep and tortuous water gap through these mountains where, unobscured by forest growth, fold after fold of the stratified limestones and shales appear in startling boldness. Several journeys through this gap only increased the appreciation of the greatness of the task of thoroughly delineating the section, the complexity of which may be inferred from the accompanying figures. Twenty miles south of the Arbuckle ridges proper, and separated from it by a valley based upon Carboniferous shales and sands, near the crossing of Hickory Creek and the Santa * The only previous mention of this important granite area of which I am aware was made by Dr. R. H. Loughridge in the 10th Census Report on Cotton Production. ; H ii 120 hk. T. Hill—Reconnarssance of the Fe road is a smaller but similar and parallel ridge of folded Silurian rocks extending westward to Healdton (see map). For these mountains there is no local name, and I have ealled them Red Bird from an adjacent post-office. They serve to prove the great width of the folded belt. Us o ZAM) luau Section across Indian Territory from south to north along Atchison, Topeka and Santa Fe Railway. The accompanying north and south section and profile from Gainsville, Texas, to Guthrie, Oklahoma, gives at least an idea of the sequence and foldings of the Arbuckle region. Pro- ceeding southward along the line of the Atchison, Topeka and Santa Fe, the typical gypsiferous red beds of Texas, Kansas, Indian Territory and New Mexico—the alleged Triassic* —are seen from Guthrie to Oklahoma City, lying in a disturbed, but comparatively sub-horizontal position, showing greater dips than the Cretaceous, but none of the complicated folding of the Paleozoic strata. South of the Canadian, the Carboniferous clays and sandstones appear with the excessive dips of the Ouachita folds. At Buckhorn Creek, east of Dougherty, the coal-bearing beds of the Carboniferous are seen dipping north at an angle of 65°, and involved in the folds of the adjacent limestone hills. In this vicinity there are terranes at the base of the Carboniferous, the age of which I could not determine, especially a great thickness of soft sandstone, but the succeed- ing limestones are undoubtedly a part of the Silurian system as determined for me from fossils by Professor Henry 8. Williams. Proceeding southward from Dougherty to Berwin the lime- stones, shales and sandstones of the pre-Carboniferous succeed each other, but so complicated is the vertical folding, that the writer must confess his utter inability to determine their pro- per succession, even after considerable study. These rocks occupy in cross-section, almost invariably a sub-perpendicular * The basal portion of these Red Beds is of Permian age as shown in their Texas continuation by Boll, Cope and White. See American Naturalist, June 1879, September 1880. Ouachita Mountain System in Indian Territory. 121 position for a distance of twelve miles. From north to south, however, the following distinct sub-divisions are apparent. Their relation however is indefinite, owing to folds and faults: 1. Massive, hard blue limestones. Strata of 20 feet in thick- ness alternating with thin flaggy layers. Thickness feet. manertupted by a great-fault)__22....1_2+-2-.-22-- +280 2. Massive limestones, but in thinner and more flaggy PemerGene te Netty. 2 cjs nos ig. doled aig Sule 2 se +100 3. Thin shaly argillaceous beds, fossiliferous, excessively folded and crumpled. Aggregate thickness includ- Sn POUT TA RS Se aye genes ets Pea ee nA Ton eae +360 4, A massive bed of pure white loosely cemented sand- stone, similar to that seen above the Lower Helder- Perea VV OOCIOEC 2 oro ee eee aie 95 5. Thin flags and shales, mostly concealed but seen in con- | mcomch Gat south side of fiver ~....2--.....--.-- ? 6. A massive, yellow-blue limestone; finer grained than No. 1; rich in fossils (Trilobites, ete.) South bank of Washita at railroad bridge (Trenton) --_- .-.----- +140 . Concealed interval. . Dark blue shales of great, but undetermined thickness. . Carboniferous shales and sandstone, Berwin to Overbrook. Cc CO -T Concerning the age of the pre-Carboniferous rocks only a little can be said, but sufficient to confirm the impression that they include Trenton (No. 6), Niagara? (No 1), Lower Helder- berg, (No. 2). Could accurate collections be made, many other terranes would no doubt be shown to exist. The basis for these determinations are as follows. Near Woodford post office, ten miles west of the railroad, I collected from strata which are continuous with and apparently the same as No. 2, the following fossils, kindly determined by Professor H. S. Williams: Spirifera lamellosa, Strophomena rugosa (=rhom- boidalis), Rhynchonella nucleolata, Lingula ? rectilatra. Con- cerning these he says: “It is safe to say the horizon is Upper Silurian and probably equivalent to the Lower Helderberg of New York. It is above the Niagara, and this is an interesting feature.” Concerning the fossils from No. 6, he says: ‘ They are not very satisfactory but a Zrinucleus concentricus shows No. 1092a to be of Lower Silurian, probably Trenton age.” A fine specimen of Lituites beckmani Whittield, in my pos- session, I have cause to believe came from this same locality, although I had previously been greatly deceived by its collec. tor as to its locality and horizon. It is not my desire to attempt any classification of these pre- Carboniferous rocks, but I believe from stratigraphic evidence that the shales at the south end of the gap may prove Devo- nian. Beneath the Trenton rocks there are exposed still older eer oN R. T. Hill—Reconnaissance of the terranes, especially in the Red Bird Mountains, which may be Cambrian. Continuing southward along our section the mountains cease coincident with the limestones, and after a mile of black shales (No. 7) the well-defined Carboniferous sands and shales begin near Berwyn and continue for twenty-nine miles along the railroad to the vicinity of Overbrook. These all occur in vertical folds, apparently coincident with or at least a part of the same system to which the Silurian limestones belong, but which, owing to their disintegrating character, have been leveled down to a low undulating plain. Ten miles south of Ardmore, the Trinity sands, the base of the Comanche series, rest unconformably against the Carboniferous (the Red beds being absent), and upon these in turn to the southward the sub-horizontal beds of the Lower Cretaceous, which I shall make the subject of another paper.* A parallel north and south section twenty miles west of the Santa Fe road shows the presence of the Red beds and the absence of the Cretaceous, the lattér having deflected south- ward through Texas. It is not alone in the mountains of the Paleozoic areas, how- ever, that this remarkable vertical structure is seen, but much of the Carboniferous prairie regions east of the Red beds are based upon it. For twenty miles north from the Red Bird to the Arbuckle Mountains the undulating prairies, void of any high relief whatever, except slight rises where the sand- stones prevail, are based upon the almost vertical Carboniferous shales and sands, as shown in our diagrams. The wonderful degradation these folds must have undergone exceeds all possi- bility of description. Yet, as] have shown in my Arkansas re- port, there are many miles of planed-off folds buried beneath the Cretaceous sediments. This is the only instance in the southwest of a level upland plain underlaid by vertical structure. The great unconformity of sedimentation between the Silurian rocks and the supposed base of the Carboniferous is seen both at Buck- horn on the northern margin of the Silurian and at Hickory Creek near Red Bird on the southern side, as shown by differ- ence of dip, and the presence of conglomerates in contact with the Silurian rocks, especially at the last-named place. | 3. The Wichita Division—The Arbuckle folds west of Dunean are buried beneath the Red beds for some thirty miles, but outcrop again some eight miles north of Fort Sill, marking the northern margin of the Wichita Mountains, forming a low foothill which is comparatively inconspicuous, owing to the overshadowing height and sharpness of the adjacent eruptives of the Wichita Mountains proper. * See vol. ii, pp. 503-528, Bulletin Geological Society of North America. Ouachita Mountain System in Indian Territory. 128 These mountains rise abruptly above the level of the Red bed prairies, which surround them on every side, and their sharp jagged outlines present striking and exquisite scenery. The ragged peaks of igneous rock present a strong contrast to the stratified ridges of the eastern and central divisions of the system. Although in Arkansas the latter have a similar eleva- tion above the surrounding plain, they have not the rugged peaks and points of the Wichitas, and are covered by forests. Their aspect is Appalachian—the arid Wichitas remind us of the Rockies. The eastern Ouachitas are the eroded remnants of stratified rocks with their characteristic topography, the Wichitas consist of igneous rocks—hard, firm, ragged and barren. These mountains extend westward from Fort Sill 120 miles to the 100th meridian and were partially mapped out by Marcy and McCleland years ago,* and T’. B. Comstock has recently made an interesting reconnoisance of them.t The most prominent of the many peaks are Mt. Scott and Mt. Sheridan ; the former is 2400 feet above sea level, 1200 feet above Fort Sill on the plain below, and 1700 feet above Red River fifty miles distant. Though neither high nor extensive, the Wichitas are models of topography and mountain structure. Mt. Scott is a solid mass of red feldspathic granite with little quartz, while neighboring mountains are composed of green- stones, basalts, etc., indicating two widely different types of igneous rocks. The westward continuation of these mountains is buried beneath the Tertiary sediments of the Staked Plains and with it the history of the relation of the Ouachita system to the Rocky Mountains. At one or two places in No Man’s Land and north of Clarendon, Texas, I am told that erosion has cut down to the rocks of this mountain system but I have not been able to find the localities, The composition of the Wichitas is unlike that of any mountain area of the southwest, and, so far as I could see, presents no structural resemblance either to the basin-sur- rounded mountains of the Trans-Pecos, or the early Paleozoic buttes and denuded folds of the central Texas region. Their age is not determined. They are certainly Post- ‘Silurian and the Red beds have in part participated in the movements but the eruptives may be Post-Cretaceous or even later. The apparent absence of the Lower and Upper Cretaceous in the composition of the Wichitas is especially noticeable. Their trend and composition plainly places them in the Ouachita system. * See Exploration of Red River of Louisiana, Marcy. + See First (Second) Annual Report of the Texas State Geological Survey. Austin, 1889. 124 R. T. Hill— Ouachita Mountain System, ete. Résumé of History recorded in the Ouachita System. 1. There are evidences of a Post-Silurian movement in the Buckhorn and Red Bird unconformities. 2. The great folding and elevation of the system were after the close of the Carboniferous period, probably during the Per- mian, as shown by the participation of the rocks of the former period in the movement, and Pre-Triassic, if the upper Red Beds are of that age. 3. A second or lateral movement must have taken place after * this folding by which the folds were bent into S-shaped flexures. This movement preceded the Red Bed epoch. 4. The marked but not excessive disturbance of the Red Beds indicates movement and displacement after their deposition and previous to the Trinity epoch. 5. The Lower Cretaceous Comanche series—which may be partly Jurassic—was deposited against and not over these moun- tains, and show in themselves no folding or other disturbance except such faulting as may be attributed to the Post Upper Cretaceous continental movement. 6. The Upper Cretaceous, the Marine Eocene and the Quater- nary along the southeastern and eastern border of the system in Arkansas were also deposited against and not entirely over the system, and, like the Comanche series, reveal no participation in adjacent mountain folding, but merely alternations of subsidence and elevation. | 7. This system has undergone extensive erosions throughout Post-Carboniferous time, and its sediments have contributed to all later deposits. 8. The western portion of the mountain system was in parts submerged during the Red Bed epoch [Triassic ?] and completely degraded or buried beneath the sediments of the great Tertiary lake which constitutes the formation of the Llano Estacado. 9. The relation of this system to the Rocky Mountain move- ment is to be determined. The mountains of the Ouachita system, including the eastern or Arkansas-Choctaw division, the central or Wapenucka Lime- stone district, the Arbuckle division and the Wichitas, should no longer be omitted from our maps, for together they consti- tute the foundation of all later geologic structure in the Texas region, differentiating it from the Kansas-Missouri region in both present and past geologic times back to the earlier Meso- zoic epochs, and influencing all the main river courses of Indian Territory whose great southward bends are an adaptation to the strike of this mountain system, the Washita alone having cut through it. : The mountains are also interesting from their exceedingly diverse structure and composition, and from the fact that, with the exception of the Uintas, they are the only east and west system on our continent. . a C. Barus— Continuity of Solid and Liquid. 125 Art. XII.—The Continuity of Solid and Liquid ;* by ; CARL BaARUS. Introductory. 1. My earlier paperst entered somewhat minutely into the volume thermodynamics of fluid matter. The behavior of matter passing from liquid to solid and back again was only incidentallyt considered. This feature, however, is the very one which gives character, or at least a more easily interpret- able character, to the whole of the volume phenomena of the substance; and it was therefore reserved for special research. The problem may be looked at from another point of view: Let it be required to find the relation of melting point to pres- sure. My results have long since shown§ that in a compre- hensive study of this question the crude optical and other methods hitherto used as criteria of fusion (criteria which have no inherent relation to the phenomenon to be observed) must be discarded. In their stead the striking volume changes which nearly always accompany change of physical state, in a definitely constituted simple substance, are to be employed. The literature of the subject I will omit here, since the _ more important work has entered the text-books and since I shall probably have occasion to refer to it elsewhere. The present experiments were made with naphthalene only. They are no means even near the degree of precision of which the applied plan of research admits. Thus far my chief object has been to carry the method quite through to an issue, pre- liminarily, and to test it at every point. The data are suffi- cient, however, to show that the procedure adopted, though I approached it with diffidence, can be brought under control throughout; and that the attainable accuracy need only be limited by the patience, skill and discernment of the observer. _I was in some degree surprised, therefore, to find that my method led to new results at the outset. 2. Harlier allied experwments.—In applying the principle of $1, I first made direct volume measurements with substances enclosed in capillary tubes of glass. In the case of naphtha- * Geological interpretations are in the hands of Mr. Clarence King, by whom the work, as a whole, was suggested. + This Journal, III, xxxvili, p. 407, 1889; xxxix, p. 478, 1890; xl, p. 219, 1890; xli, p. 110, 1891. Phil. Mag., V, xxx, p. 338, 1890. ¢ This Journal. xxxviii, p. 408, 1889; xxxix, pp. 490, 491, 494, 1890. § This Journal, l.c. More pointedly with an indication of methods in Phil. Mag., V, xxxi, p. 14, 1891. || 1 will merely mention Sir William Thomson (1850), Bunsen (1850), Hopkins (1854), Mousson (1858), Poynting (1881), Peddie (1884), Amagat (1887), Battelli (1887) and some others. Cf. §§ 29, 30. 126 C. Barus— Continuity of Solid and Liquid. lene and some others, I thus obtained satisfactory results.* Such work is, however, limited to relatively low pressures (600 to 800 atm.); it does not adm of sufficient correction for the volume changes of the glass, and from the small quantity of substance examined, and the relatively frequent occurrence of nuclear condensation, volume lags are often obscured. Hence the definition which I was inclined to adopt after making these experiments, viz: that a pressure which when acting isothermally for an infinite time will just solidify the liquid and will just liquify the solid, stands to the given temperature in the relation of melting point and pressure, is not in accord- ance with facts.t In a second methodt I endeavored to ascertain the positions of the characteristic specific volumes by passing current out of the mercury index through the hot walls of the thin glass tube which contained contiguous columns of both the substance and the mercury. Supposing the tube surrounded by a liquid conductor transmitting pressure, the changes of resistance of the arrangement indicate the motion of the index and hence the degree of compression produced. Here, however, a new and unexpected annoyance was encountered, inasmuch as both the medium of oil contained in the piezometer and the glass possess seriously large pressure coefficients.§ Moreover it is only with great difficulty that the perfect insulation of an apparatus, in which water jackets form an essential part, can be maintained. I therefore abandoned the work. In a third method similar to the preceding, I expressed the motion of the mercury thread or index in terms of the resist- ance of a very fine platinum wire, passing through the axis of the tube. Successive intercepts thus indicated the changes of volume to be observed. This method gave good indications of the pressure position of the melting points of the sample. It failed, however, to give serviceable values for the fiuid volume changes. I found on trial that the contacts in such a case are essentially loose, and that thermocurrents can only with diffi- culty be eliminated or allowed for, seeing that the successive isothermal temperatures are to be considerably above the at- mospheric temperature. | Finally all the methods here described must necessarily fail after the substance has been solidified; for in this case the thread or index is split up and forced into the interstices of the solid material. Thus it is manifestly impossible. to retain the * Cf, this Journal, xxxviii, p. 408, 1889. + A considerable number of experiments made with naphthalene in this way showed the melting points 83°4°, 92°3°, 100°, to correspond to the pressures 100 atm. 350 atm., 565 atm. respectively. Thus the factor is +°036° C./ atm. § 27 Phil. Mag., xxxi, p. 14, 1891. é Ibid, pp. 18 to 24, et. seq. CO. Barus—Continuity of Solid and Liquid. 127 original meniscus, and therefore impracticable both to arrive at the volume behavior of the solid and to rigorously codrdinate successive series of experiments. 3. Advantages of the method of this paper.—Hence I en- deavored to modify Kopp’s* specific volume flask, in a way to make it available under any temperature or pressure. Here the readings are independent of the unbroken character of the meniscus immediately in contact with the solidifying substance, whereas on the other hand (as I shall presently show), the volume measurements can be made electrically, with almost every desirable degree of accuracy. Furthermore by charging the flask with suitably apportioned quantities of substance and of mercury, the error due to the compressibility of the glass may be eliminated in any degree whatever, and an apparatus be obtained which is practically rigid in relation to pressure. The data show that from each single series of experiments I thus obtain the isothermals and isopiestics and therefore also the isometrics, both for the liquid and for the solid state, admitting the latter to be less accurate; further, the relation of solidifica- tion and.of fusion to pressure, and finally, the pressure changes _of the isothermal specific volumes of solid and liquid, at solidi- fying and melting points. From such results the character of fusion, and the probable positions of critical, $26, and of tran- sitional points, §28, can already be pretty well predicted. It is then only necessary to examine a number of substances, normally existing under widely different conditions of thermal state,t in order to broaden the evidence and possibly to reach results of a uniform -bearing on matter in general. Thus I endeavor to avail myself of the enormous internal pressure through incremental pressures applied externally. Apparatus. 4. Temperature.—Inasmuch as pressure varies at a mean rate of over 30 atm. per degree of melting point, so that tem- perature is as it were the coarse adjustment and pressure, the fine adjustment for the conditions of fusion, the method of experiment should be such that temperature may be kept rig- orously constant while pressure is varied at pleasure. To obtain constant temperature, I constructed a series of brazed * Kopp: Ann. Chem. u. Pharm., xciii, p. 129, 1855. The results of this fine memoir are too rarely quoted. + The absolute expansion and compressibility of mercury being now known. Si. . t ‘‘Instead of tracing the isothermals of a single substance throughout enormous ranges of pressure, similarly comparable results may possibly be obtained by ex- . amining different substances conceived to exist in widely different thermal states.” This Journal, 1. c., xxxix, p. 510. Am. Jour. Sc1.—THiRD Seriss, Vou. XLII, No. 248.—Auvueust. 1891. 9 128 C. Barus—Continuity of Solid and Liquid. vapor baths of thin sheet iron, thickly jacketed with asbestus. They were cylindrical in form, 20™ high and 10™ in diameter. Axial tubulures, the upper of which projected outward, the lower both inward and outward, allowed the vertical tubular piezometer to pass axially through the vapor baths, and suita- ble stuffing boxes obviated leakage. Again the upward pro- jection of the lower tubulure (both of which fit the piezometer snugly), formed an annular trough with the walls of the vapor bath, in which a sufficient quantity of the ebullition liquid could be placed, and boiled, by aid of the flat spiral burner below. The top of the vapor bath was provided with two other (lateral) tubulures, one of which served for the perma- nent attachment of a vertical condenser, and the other for the introduction of a suitable thermometer or thermocouple. Here also the quantity of ebullition liquid present, could at any time be tested, its amount increased or diminished, and its quality directly purified by fractional distillation or otherwise (an operation necessary, for instance, when amyl alcohol is used). §20. With a good condenser, the boiling may be kept up indefinitely, for the condensed vapor falls back into the trough below. At temperatures below 100°, it is expedi- ent to avail oneself of the high latent heat of water* and to boil this liquid under diminished pressure. Temporarily attach- ing Professor R. H. Richards’ jet pump to the end of the con- denser, pressure may be reduced at pleasure, and any boiling point between 50° and 100° reached and maintained indefi- nitely. For higher temperatures toluol, amyl alcohol, turpen- tine, naphthalene, benzoic acid, diphenylamine, phenanthren, ‘sulphur, etc., subserve similar purposes more or less thoroughly. Temperature was measured by a Baudin thermometer of known errors, and also computed from the vapor tension of steam under known conditions. 5. Pressure.—To obtain pressures as high as 2000 atm., I employed the screw compressor described elsewhere.t I made use, however, of a vertical piezometer, identical with the hori- zontal form described, except in so far as it could be removed from the barrel as a whole. As before, the piezometer is-insu- lated from the barrel. When in adjustment the former was surrounded by the following parts, enumerated from below: an insulated guard preventing spilled water, ete., to reach the insulation; the lower cold water jacket, the flat burner, the vapor bath, and finally (wherever necessary) an upper cold water jacket. Internally the piezometer was filled with thick mineral oil.t * T shall in future experiments also boil water under pressure. + Proceed. Am, Acad., xxv, p. 93, 1890. + Phil. Mag., (V), xxxi, p. 10, 1891. OC. Barus—Continuity of Solid and Liquid. 129 For pressure measurement 1 am now able to avail myself of superb Amagat ‘‘manométre a pistons libres,” which can be attached to my compressor without further mechanism and with advantage.* The instrument is adapted to measure 3000 atm. 6. The volume tube.—This is shown in the annexed figure (diagram). It consists of an external cylindrical & envelope AB of glass, closed below, open above, about 26™ long and -4™ or ‘5° in diameter. Throughout the greater part of its length, the tube is divided into two coaxial cylindric compart- ments, by a central glass partition tube CkC, open at both ends, and fused te the tube AB along the ring CC, about 7™ from the top. CkC Ck is about 17™ long and ‘13° in internal diameter, drawn as thin-walled and even in calibre as possi- ble, so that the greater part of its lower length may be available for measurement. The substance to be examined is introduced into the annular space “4, care being taken that when fused under the highest temperature and lowest pressure to be applied, its lower boundary may be 4 or more above the end &. Immediately in contact with H# and extending upward into the central tube is a plug of mercury /’/, with its \\E WSS RSS S NS WG, Yj Z % \F ©) Zia SN ‘ Yj VL 13 Milde. 14 ZIONS SS SSS SSSSSSSS SSS WSU Mpc : : NAN free meniscus at g. When ZZ is solid, g must be , KNNN (say) 2 above the end #4, and when EE is liquid B \.\ b g must even in the extreme case be at about an equal distance below the end CC of the tube CkC. The remainder of this tube, above g, is quite filled with a con- centrated solution of zine sulphates G/g, into which an amal- gamated zinc terminal 1, has been submerged and fixed in position by the platinum wire a, fused to the sides of the tube AB as shown. The other terminal ) passing through the sealed bottom of A.B, is in metallic connection with the mer- eury /’F therein contained. The tube thus adjusted is completely submerged in the oil within the insulated tubular piezometer, with which the termi- nal a connects. The terminal 6 completely insulated from the piezometer by a coating of glass tube, is in metallic connection with the barrel. Thus the tube AB is held in position by tensely stretching the fixed wires a and 4, and so adjusting their lengths that the parts #2 and hg with reference to which the measurements are made, may lie wholly within the vapor * Results thus obtained in comparing various high pressure gauges and methods of manipulation will be given in a current number of the Phil. Mag., xxxi, p. 400, 1891. 130 C. Barus— Continuity of Solid and Liquid. reservoir of the cylindrical vapor bath surrounding the piezom- eter. Many of these operations are delicate, but descriptions must be omitted. An inspection of the figure shows at once that if a current enter the outside of the barrel, it will pass through 0, k, h, D and a, into the outside of the piezometer, and thence back to the battery. The only relatively significant resistance en-- countered in such a course, can be confined to the path between g and A, through the thread of the zine sulphate solution ; but this resistance, cat. par., varies directly with the length of gh and therefore proportionally to the volume contraction of the substance HH! If Kohlrausch’s method* of intermittent cur- rents, bridge and telephone be used for the resistance measure- ment of the electrolyte, solidification or fusion of “’# breaks upon the ear with a loud roar, whereas the ordinary volume. changes {solid or liquid) are indicated by intensifications of the sound in the telephone, sufficiently pronounced however to subserve the purposes of measurement. | It is seen that any breakage of the surface of separation . between H#’ and /’/’ is entirely without influence on these 2 results, and that even in case of solidification of #4, when the mercury is forced into the interstices left after contraction, the compressibility of ’#’ will still be measurable. | The charging of the tube free from air, is an operation : which I have not yet accomplished satisfactorily. If a volatile : | | substance like naphthalene be filled into H# and fused in vacuo, vapors objectionably condense in the tube Ak. If HE be not fused, I doubt whether the air can be eliminated in vacuo. Hence in the present work, the substance was not air- - free, a condition to which I gave less attention because I do not believe the melting points are appreciably influenced by dissolved air, nor that the other measurements made are seri- ously distorted by this error. In further measurements, how- ever, I will endeavor to meet the difficulty by fusing the end A of the inverted tube AB, to the top of a barometer tube, provided with a lateral tubulure leading to a Sprengel pump. If then, after exhaustion the lower meniscus is adjustible, so that the whole barometric column can be raiséd quite into the tube AZ or withdrawn from it at pleasure, a thorough vacuum filling may be effected. Rubber connections must be serupu- lously avoided. Method of Measurement. 7. Constants of the tube.—In order that the present meas- urements may be carried out absolutely, it is necessary to * Kohlrausch: Verh. med. phys. Ges. Wurzburg, xv, p. 1, 1880; Wied. Ann., xi, p. 653, 1880. Long: Wied. Ann., xi, p. 37, 1880. . C. Barus—Continuity of Solid and Liquid. — 1381 know: (1) The volume of the charge at a fiducial temperature and pressure ; (2) The volume of the plug of mercury under the same conditions; (3) The volume of the central tube kA (figure 1) per centimeter of length; (4) The resistance of the thread of zine sulphate solution, per centimeter of length, under all the stated conditions of temperature and pressure. From (3) and (4) there follows at once (5) the resistance of the thread of zinc sulphate per unit of volume, under any stated conditions of temperature and pressure. Thus it is necessary to investigate preliminarily (6) the isopiestic relation of resist- ance and temperature of the given concentrated solution of zine sulphate, and (7) the isothermal relation of resistance and pressure of the same solution. In other words one must know what may be called the isoelectrics of the measuring electrolyte. Furthermore it is necessary to find (8) the compressibility of the glass in its relation to pressure and temperature and (9) the compressibility of mercury under the same conditions ; finally (10) the thermal expansion of the glass and (11) the thermal expansion of mercury under given conditions of pressure. The measurements (8) to (11) I have not thus far made directly. They are here of small importance, seeing that the substances on which I operate are all characterized by rela- tively large volume changes. Such measurements, however, are easily feasible, since both the expansion constants and the compression constants of pure mercury (thanks to the recent labors of Tait, Amagat* and Guillaume) are now thoroughly known, and it is also known that the thermal changes of the elastics of glass are of no relative consequence,t even as far as 200°. If therefore the tube AB, figure 1, be filled with mer- cury, replacing the substance 4, the expansion and com- pression constants may be found by the method above stated, $3, once for all. In the present paper I assumed the compres- sibility of my glasst to be -0000022, that of mercury,$ being 0000039 ; moreover the coefficient of thermal expansion of the glass| to be .000025, that of mercury4| between 60° and 130° being 000182. 8. Volume of the charge.—Clearly the fiducial conditions to which the volumes are to be referred, are given by the (normal) melting point, under atmospheric pressure. By weighing the tube before and after charging, I found for the mass of naph- thalene enclosed, ‘763g. In a special and duplicate set of pycnometer measurements, I furthermore found for the density of fused naphthalene at 82°, ‘724. Hence the volume of the * Cf. E. H. Amagat: Ann. ch. et phys, VI, xxii, p. 95, 1891. + Ibid., p. 136. f Ubid.j ip: 125: § Ibid., p. 137. || Landolt u. Boernstein’s tables, 1883, p. 69. Sielbitd: p..3't. ll 132. Barus—Continuity of Solid and Liquid. charge at 82° is 552 em’, which I took for the volume at the normal melting point (80°). 9. Hxpansion and compressibility of envelopes.—The plug of mercury weighed 774g. Its volume was therefore 571 cm’, at 20°, and its mean volume between 60° and 130° (being be- tween °575 and °582) sufficiently near ‘58 cm*. Thus the volume of the glass tube containing both the charge of naphthalene and of mercury, was 113 cm’. Its expansion per degree centigrade -000028 cm*, while the ex- pansion of the mercury in place was ‘000105 cm‘, per de- gree, whence the apparent expansion ‘00007 cm®* per degree. Therefore if in place of the fiducial volume 532 cm®* (§ 8), the following volumes be substituted, viz: 60° *5565 cm.? 100° 5535 em.? 80° °5550 120° 0519 SO D542 130° “5311 the tube may be treated as free from thermal expansion. Here at 80°, °555 appears instead of °552, to allow for the fiducial volume of the stem A (fig. 1), as will be shown in § 17. Again the compression of the 1:13 em.* of glass, and the *58 cm.*, of mercury will be: 100 atm.; glass, 00025 em.*; mercury, 00023 cm.?; difference, :00002 cm.* ? 1 500 124 113 1000 249 226 23 1500 373 339 34 2000 A497 452 45 Thus the corrections which would individually be appreci- able (affecting the increments say 3 per cent) are differentially negligible (.8 per cent) where they fall below the electrical pressure coefficient of the zine sulphate solution. §14, ef. §3. 10. Lesestance measurement.—Using the interrupter and telephone ($6), I facilitated audition by connecting the dia- phragm cup with a graphophone tube, and listening with both ears. ‘The resistances, however, were rather higher than con- templated in Kohlrausch’s method, when an ordinary Bell telephone is used. Hence the measurements particularly near and in the solid state are far below the limit of attainable accuracy. I shall in future measurements wind a telephone specially adapted for my purposes, and endeavor to use both ends of the magnet to actuate diaphragms. When zine sul- phate is enclosed between terminals of zinc, continuous cur- rents and the galvanometer are available. In this way, I made most of the calibration measurements. Supposing the mercury index to be slightly deadened in its electronegative qualities by zinc, it may also be used in ease of the tube. Should the measuring thread of mercury gh, figure 1, break into parts alternating with threads of zine sulphate (a possi- C. Barus—Continuity of Solid and Liquid. 133 bility when the thread is worked up and down many hundred times, particularly in view of the suddenness of solidification), the constants of reduction are not thereby necessarily vitiated, always supposing the number of such breaks to be small. The shifting of codrdinates thus produced can be corrected by check-work at a given temperature.* Long continued passage of intermittent currents, charges the mereury with zine, but solution of mercury can not become serious, since the column is being continually washed by the terminal JY. Some advan- tage would be gained by using zine sulphate in the strength (1:286, Kohlrausch) which corresponds to maximum conduct- ivity. 11. Calibration.—The tube hk, figure 1, being of insuffi- ciently uniform caliber, volume must be expressed as a func- tion of length. This I did by weighing threads of mercury, whose length had been measnred in successive parts of the tube, obtaining the results of the first two columns of table 1. The fiducial zero is here arbitraily placed 2°" below the ring CC. Similarly the resistance of the filament of zinc sulphate hg must be expressed as a function of length, referred to the same fiducial zero, at some convenient atmospheric temperature. To do this, I drew a zine wire down to a diameter slightly below the caliber of the tube. Opening the bottom of AB, and closing the top so as to hold the terminal D firmly in posi- tion, | inverted the tube and quite filled it with the solution. AB was then placed in a cold water bath, with the terminal a insulated, and the terminal 6 replaced by the zine wire referred TABLE 1.— Volumes per unit of length. Llectrical resistance per unit of length 6=17°8°. Volume per unit of resistance 92=17:8°. Length. Volume. Length. Resistance. || Resistance. | Volume. cm. cm.3 cm. ohms. ohms. cm.? 3°00 0491 —'06 2720 2800 "0000 11°15 "1609 2°13 5780 5530 °0350 Sr a ee 4°15 10190 8850 ‘0640 9-79 1430 7:96 16200 12530 0920 15-49 2145 12-06 24340 16270 "1190 i 3-30 0538 1:38 4630 20250 *1450 7-20 1076 3°43 7920 24250 "1705 15°60 ‘2152 9°00 18220 | 215 0470 11°49 23140 6:10 "0940 | 9°65 ‘1410 13°40 *1880 * This I should have done after obtaining Table 9, $21; but the full details of manipulation could not all be foreseen at the outset. 134 C. Barus—Continuity of Solid and Liquad. to, and so adjusted that I could slide it up or down and fix it in any position at pleasure. Measuring the distance between the ring CC and the free end of the wire, with Grunow’s cathetometer, and measuring at the same time the resistance corresponding to this length, I obtained the data necessary for constructing resistance as a function of length, for the tempera- ture of the bath. In this way the second and third columns of table 1 were found. Combining the results of these four columns by graphic interpolation, I obtained the fifth and sixth columns in which volume is expressed in terms of resistance, at the temperature §—17°8° with regard to the fiducial mark in. question. 12. Electrolytic resistance and temperature.—The investiga- tion of this relation is a general problem, quite apart from the TABLE 2.—The relation of electrical resistance to temperature and pressure, in case of a concentrated solution of zinc sulphate. || eat eeu eng Pressure. Resistance. _|| L°™PETA- pressure,| Resistance, ture. | | ture. ws aim: | ohms.) \ “i / Finn || 2“ C. atm. ohms. | R/Ryioo *6°3 100 | 26960 6°060 99°6 191 1237 "990 6°5 100 | 26960 6-060 | SES 186 1242 "994 66°8 140 | 6178 1°388 99°6 | 471 1222 ‘978 67°3 140 6120 1:376 | 99°6 448 1210 ‘968 68:0 140 6050 1360 | 09°62") ) LOrd 1198 "958 678 140 6060 1362s) 99°6 4 981 1203 "962 67'8 140 6075 1°365 996 503 1232 "986 67°7 140 6075 1°365 99°6 507 1232 ‘986 100°0 137 4550 10235 7 99°6 126 1262 1°010 100°0 137 4540 1:020 126°0 157 1062 "850 100°0 479 4470 1°005 126°0 Ly 1062 "850 100°0 469 4480 1°007 127°8 14S 1058 "846 LO0s0™ |) TOTS 4370 "982 12 ine 149 1058 *846 100°0 | 996 4380 «985 160°5 154 980 “784 100°0 | 150% 4310 ‘969 160°5 154 984 "787 100°0 1443 4320 00 160°5 154 984 “T8T 100°0 129 4360 "980 $64 138 7440 5°952 100°0 139 4360 ‘980 6:4 138 7500 6°000 127°6 158 3740 "840 6°4 138 7500 6-000 127°8 158 3730 °838 16°4 147 4980 3°984 127°8 158 3730 838 16°4 147 5000 4-000 6°5 116 27140 6°100 16°4 147 5020 4°016 6°3 116 27000 6°067 61:3 170 1900 1°520 +6°7 72 7280 5°824 61:3 170 1900 1°520 6:7 465 7090 5 672 61°3 170 1913 1°530 6°7 458 7120 5700 85°8 180 1359 1:088 67 906 7020 5°616 85°8 180 1359 1:088 6°7 830 7020 5°616 85°8 180 1364 1/091 67 498 7130 5704 99°6 189 1247 1:000 6°7 | 492 7050 5°640 99°6 189 1253 1°002 6°7 139 7190 5°752 99°6 189 1253 1:002 Gt ulay 7160 5728 + Second Series. + Third Series. Diameter of tube, ‘30°™. Diameter of tube, ‘30°. C. Barus—Continuity of Solid and Liquid. 135 special apparatus used. Nevertheless I made two sets of measurements, in the first of which I determined the resistance of the thread Ak, fig. 1, between fixed terminals of zinc, when the whole apparatus was kept at successive constant tempera- ture, and under pressures sufficient to insure the condensation of all polarization gases and the presence of a continuous liquid thread of zine sulphate solution. The four columned table 2 contains these results, where A/Z,,, is the relative resistance at any stated temperature in terms of the corre- sponding datum for 100° C. At 100° moreover pressures are varied for the measurement of the pressure coefficients dis- cussed in the next paragraph. In the second and third parts of the table, the above tube AB was replaced by a plain straight tube. Resistances are much smaller here, but the column //£#,,, makes all the data comparable. If the values 2/f,,, be compared graphically, as a function of temperature for nearly the same pressures, the results of all the series in table 2 are in good accord. Moreover the results for the large interval 6° to 160°, lie on a curve whose form closely resembles an hyperbola. From this point of view _ the data are remarkably interesting: for if it be true, then a suitable inversion of the locus indicates that the electric con- ductivity of the electrolyte varies linearly with temperature. Such a result would not only possess theoretic interest, but would make measurements of the kind necessary in the present paper feasible with a high degree of certainty. The interpola- tions of this paper were made empirically however, and I must withhold further opinion until I can trace the locus as far as 300°. I may add that inasmuch as a solution of maximum conductivity is accompanied by a smaller temperature coefii- cient, advantages of such a solution are suggested, $10. 13. Volume in terms of resistance.—With the data of §$11 and 12 in hand, it is now possible to express the volume of the capillary tube Ax, figure 1,in terms of the resistance of the thread of electrolyte, observed at any temperature. With this object in view, I computed tables for each of the tempera- tures of the isothermals below, §$15 to 21, facilitating the further reduction by graphic methods. Being merely of pass- Ing interest the tables are omitted here. 14. Pressure coefficient of the electrolyte.—The results in table 2 for variable pressure and constant temperature are summarized in the small table 4, below. Here @ denotes the temperature, 7 the resistance of the thread, and k=d R/£, dp where p symbolizes pressure, the pressure coefficient sought Ff, holds at 0°C. 100 136 C. Barus—Continuity of Solid and Liquid. TABLE 4.—Pressure coefficients of concentrated zinc sulphate solution. @ Pressure. kx 108 6 (Pressure.! kx 106 | 6 (|Pressure.! kx 106 T° 119) I 200° | 15% |} 100° | | 134 ARON ole Gir al 482 tel 1007 kc 119 eho | 474 ae L135 slag: | —35 | | 996 § | 1475 The mean value is k= —45/10°, being negative, inasmuch as the resistance is here decreased by pressure. The pressure coefficient is nearly independent of temperature, and decreases somewhat with pressure. The results, however, are not quite consistent, and a detailed construction of the data in table 2 shows a difference of march in the pressure on and the pres- sure off movements. I have yet to learn whether this be due to insufficiently fixed terminals, or to. polarization, as well as to find the conditions (change of concentration or of the solu- tion) under which the pressure coefficient may be a minimum. As the results stand the mean value is probably within 20 per cent of the truth, and hence in the extreme case of 2000 atm., the uncertainty of the pressure coefficient will not affect the volume increments more than 2 per cent. In an earlier paper,* I found k=—50/10° between ( and 150 atm., agreeing substantially with table 4. I then called attention to the strikingly close proximity of this datum to the corresponding coefficient for mercury k= —30/10°. The pres- sure coeflicient is of considerable interest, inasmuch as it indi- cates a certain relation between elasticity and the chemical equilibrium of the solid or liquid operated ont specially for zine sulphate, it may be noted that whereas the conductivity of a nearly concentrated solution (density > 1°29) decreases on further concentration, compression (which might be regarded as having a concentrative effect on the solution between the terminals) increases the conductivity. Results of the measurements. 15. Arrangement of the tables.—The following tables 5 to 10, in which the isothermals of naphthalene are fully given, are constructed as follows: The first column contains the time. in minutes at which the observation was made, the initial date being arbitrary. The (uncorrected) resistance as actually found at the pressure given, is shown under in ohms. To this the correction for pressure coefficient, Ap per unit of &, is to be * This Journal, xl, p. 219, 1890. The work of this paper was done some two years prior to the publication. + Phil. Mag., V, xxxi, p. 24 et seq., 1891. OC. Barus—Continuity of Solid and Liquid. 137 added, after which 7 can be expressed as a volume increment, referred as yet to an arbitrary fiducial zero, $$ 13, 14. The corresponding volume (last column of the tables) is deduced from this by inserting the initial volume values of §9. Cf. $17. Two data are usually given for each step of pressure, the second of which, obtained after long waiting (5™ or more) is more nearly isothermal than the first. In most cases, a small additional volume decrement takes place after solidification, either viscously or as the results of gradual decrease of tem- perature. Parentheses occur to show that for the data enclosed the measurement was made along an (upper) part of the tube Ad, figure 1, whose calibre was not sufficiently uniform. Without knowing the full expansions at the higher temperatures and lower pressures, it is a priori impossible so to fill the tube that all measurements fall within calibrated parts, and all other calibration conditions are complied with, §9. These approxima- tions however refer to the liquid state, and are thus of less consequence in this paper, $1. If the isothermals of the liquid only were sought, it would be advisable to make the tube hk very much more nearly capillary from the outset. The experiments were made on different days, and together extended over more than a week. This is too long a time to employ the tube without special readjustment, and some shift- ing of codrdinates may therefore have occurred. §$10, 25, 26. I may add finally that the melting point of naphthalene in air is about 80°, its solidification point below this, under proper conditions. The density of the solid is 1:14, and that of the liquid at 82°, is 724, §8. Hence naphthalene melted in water sinks or swims, according as its temperature is sufficiently below or above 80°. 16. Solid wsothermal, 63°.—Clearly the data obtained in operating on the solid will be less accurate than the liquid TABLE 5.—Jsothermals of (solid) naphthalene, at 63°5°, referred to 55 cm.?, at the normal melting potnt. Time. Pressure. PR. | kag Volume. Time. Pressure. R. _“? , Volume. | | | x 10 x 10 | | | | | | m. | atm. |ohms.| ome. |) Mm). atm... ohms. em. 16 1 petGO0) 3 ALD \)° 38 859 8240 39 3965 oye) GS 7430; 3 | 4156 || 40 850 =| 8120 38 *3988 20 289 1760) 13. | 4085 || 42 | 972 | 8310) 44 *3945 Ao, 28h 2) (630) 13. |. 4109) |) 44 966 | 8310 44 *3945 25 | 491 8160 22 ‘4003 | 46 565 | T600 25 “£097 29. | 481 8010 22 4027 | 49 565. © |. 1605) 25 "4097 34 | 675 | 8050) 30 | -4010 || 51 100 7270, 4 "4183 37 | 662 | 1880) 30 | 4039 || 67 65 7340 3 ‘4173 138 C. Barus— Continuity of Solid and Liquid. data, §6. For in addition to relatively greater importance of the corrections for the compressibility of the envelopes, the fissured or honeycombed structure and the high resistances, §10, interfere with sharp measurement. Nevertheless by com- paring the data with similar solid isothermals obtained at much higher temperatures, their validity may be inferred. With reference to the series it is interesting to note that re- coil of volume (pressure decreasing) is more rapid than compres- sion. ‘The reverse of this would have been anticipated, sup- posing that mercury lodged in the interstices. The liquid in the above case was allowed to solidify under pressure. | TABLE 6.—Isothermals of naphthalene, at 83°, referred to ‘55 cm.* at the normal melting potnt. Time. Pressure. R. | _*P 3| Volume. || Time. |Pressure.| A. | ‘p 3 Volume. / x 10 | x 10 m. atm. ohms. | cm.2 | mm. | atm. ohms. | eom.3 36 39... 6923), 2-1 5498-5 1) F984 129 |. aaa “4241 3900) 206 Vo PSs) Oe 5S Uk Se We gg ANAT) 4. 4358 44 221 P1581 TOe PMGasS2. hi 8d 93 4348 4 4453 46 250 W231 V1 ‘35D |) 6 94 93 4348 4 4453 51 244 1198) 11 ‘5365 || 96 70 3651) 3 4630 FD, vole s0b 1232). 12 5348 of 70 3256] 3 ‘4732 59 | 265 P22e V2 *‘DSO1 fr TOA, area 2031) 3 5062 60 | 303 1288 13 "SO22re | MLE al POG: 1695} 3 5172 63 | 322 1326, 14 "5306 POO VERS ale 1480, 3 5250 GEST uo Wk 5667 14 4138 122 .| 49 920), 2 | 5508 (Sacer BU rs. a alas 127 52 905; 2° | ° *Bb08 75 127 | 5250! 6 “4241 TABLE 7.—Jsothermals of naphthalene, at 90°, referred to 55 cm.® at the normal melting point. Time. |Pressure.| 2. _/P, Volume. Time. |Pressure.| &. es Volume. m. atm. ohms. | om? mM. atm. ohms. | ems 16 517 Tilia.) pA 60 316 4618 14 4311 18 231 972/ 10 | 5432 65 322 4555 14 4326 24 222 95/71) 210) SP 5439 67 280 3855| 13 | “4504 26 422 | DES! We | 65329 he 283 2333) 13 | °4929 aL 406 t 1Tob) 18> in *ba3sé | %6 283 1985| 13. | 5028 33 490 |. 1247| 22 ‘D208 81 278 1618) 18° >) *biea AQ.) (AT4. 1 ABBR DIE. |. ses02 83 175 897, 8 |. “B42 40)... DOT 1258| 23 | °5292 89 179 902) 8 | 5468 44 | 532 1288) 24 | ‘5276 91 81 193) = 5530 46 | 555 5250| 25 "4147 95 83 802 4 6524 57 | 544 5290, 24 “4139 17. Liquid-solid isothermal, S3°.—Three independent. series are in hand. The example given in table 6, is the second in number. It is from these isothermals that I obtained the C. Barus—Oontinuity of Solid and Liquid. 139 fiducial zero of the stem AX, figure 1. For at pressure zero the increment is ‘0024 em.*. Hence the value -5524 of §8, increased by ‘0024 is the fiducial volume (°555 cm.* nearly) here to be used, and from this the others are derived. § 9. 18. Liquid-solid rsothermals at 90°—Two independent series of results are in hand. The example given in table 7 is the second in number. 19. Liguid-solid isothermals at 100°.—Four independent series are in hand. The example given in table 8 is the second in number. TABLE 8.—Jsothermals of naphthalene, at 100°, referred to °55 cm.? at the normal melting point. | | | | Time! Pressure. FR. | P| Volume. Time. Pressure. R. kp Volume. | |x 10 | x 10 | aaa | m. aim. | ohms. OnE A" tet! atm Womans. | ems 20 ie ube GO). Ao |.Cb5itS) 148 | 906 4882 41 | ‘4124 ay) 116 | 667| 5 | (5575) || 150 | 769 | ayia) 35 | -4175 29 Peeeh esha 1S). )> 4500 1) ee Ue AT14) 35 | -4175 36 B74 | +835) 12 | "6464 |) 159: | 644 4556, 29 | *4224 45 460 . | 1020| 21-| -5360 165 652 4556 29 | °4224 47 Bude) ba23)° 30" |) 52557 || 167 |" 560 3808) 25 | °4425 Ba) '653 | 1198) 29 | 5266 || 168 | 560 | 3202) 25 | +4599 56 WO), 4) VS3t), 35.) |,-°5204,,.)), 169 562. | 2846) 25 | -4707 63 766 TanO 34 83°. Solidification gradual, the other solidifications take place at once. + Temperature 129°6°. sie M. P. and pressure, 80° to 130°, 28°5 atm./°C., or 0351 °C./atm. c Hi SE aa bah C. Barus—Continuity of Solid and Liquid. 145 Table 11 gives the values of the pressures corresponding to solidification and to fusion at the different temperatures, together with other relevant information, as sharply as these statements can be made. M. P. denotes melting point, S. P., solidifying point. | Let the solidifying points and melting points be constructed as functions of pressure. Figure 4 shows the points to le on a spindle-shaped figure, running diagonally across the chart. They are again taken out of all the four sets of results and numbered accordingly. §25. The parts of the curves actu- ally observed are given in full lines, the inferential prolonga- tions in dotted lines. From the nature of the case the solidification points are not sharply determinable. $23. Even leaving the nuclear solidifi- cation induced by inconstancy in the linear distribution of temperature out of sight (a marked example of which occurs in the series for 117°, §20, where solidification and fusion are practically coincideat), all percussion and jarring, too rapid increase of pressure, a vibratile wire running through the column as in some of my earlier experiments, will cause the whole labile structure to topple into solidification. Hence the solidifying points must be fairly crept upon and surprised, and hence my present results in which these precautious were taken show high solidifying points as compared with my other work. §2, note. These conditions do not hold with like importance in case of fusion; for the melting points as a rule show much greater coincidence. In figure 4 I have therefore placed chief reliance on the data of series II obtained as they were with the experience of series I to guide me. 28. Transitional point.—lf the two curves be prolonged in the direction of zncreasing temperature, their eventual coales- cence is presumptive by §$ 25, 26. Clearly the occurrence of volume lag must cease when the paroxysmal volume changes vanish. If the two curves be prolonged in the direction of decreasing temperature, then the data themselves indicate the probability of an intersection in the region of negative pressure. Beyond this, therefore, there would be an inversion of the conditions of fusion: in other words, the substance would solidify at a lower pressure than that at which it fuses, and fuse at a higher pres- sure than corresponds to solidification. I believe this remark- able suggestion to be interpretable as follows: The normal type of fusion changes continuously into the ice type of fusion, through a transitional type, characterized by the zero of volume lag. The position of the latter for naphthalene, so far as can 146 C. Barus—Continuity of Solid and Lnquid. now be discerned, may be placed at (say) 50° and (say)—1000 atm. It is noteworthy, that with the understanding here laid down, the normal type of fusion is reached from the ice type, in the direction of increasing temperature.* 29. Solubility and pressure.—In view of the detailed analogy which holds between many characters of fusion, and of solu- tion, much that can be investigated for the simpler of these phenomena (fusion apparently) will be applicable to the other. A substance may be transferred from the solid into the liquid state either (1) by heating it, or (2) by dissolving it. In gen- eral, excess of temperature, or of solvent favor the diminution of viscosity here in question. A liquid on the verge of solidi- fication or a concentrated solution is solidified or deposits solid on cooling; and in both cases the nice adjustment of labile molecular equilibrium is accompanied by volume hyster- esis,—under-cooling, etc., in the one case, supersaturation, etc., in the other. Hence I conclude that if under proper thermal conditions pressure alone can solidify a liquid, it can also under proper solutional conditions induce crystallization or the deposit of solid from solution—thereby trenching upon, or (from a new point of departure) approaching the modern chemical doctrines which originated, I believe, with van’t Hoff. I am the more justified in drawing these inferences as in my last articlet on the solvent action of hot water on glass, I have already adduced the necessary evidence. Since from one point of view, the isothermal compressibility of silicated water is increased proportionally to the time during which the solvent action has been going on; and from another, with the amount of basic silicate dissolved,—the deduction is closely at hand, that what pressure actually did in this instance, was a mere precipitation of a proportionate amount of the dissolved sili- cate. The volume changes thence resulting were blindly put into computation as increments of compressibility, because the precipitated silicate is again dissolved when pressure is with- drawn. t * [Added to proof—Throughout the present paper, I have avoided the discus- sion of the isopiestics, since I shall consider them in detail in connection with special experiments. It is well to state, however, that the transitional temperature is related to the prospective intersection of the prolonged liquid and solid isopies- tics, of a given substance, at the same pressure in both cases. Thus a reason why hysteresis may vanish is again suggested. A given substance on one side of the transitional temperature would differ molecularly from the same substance on the other side. ] + This Journal, xli, p. 110, 1891. t [Added to proof—tIn justice to myself let me say that the manuscript left my hands on Feb. 23d, some five months ago, and before the kindred deductions of Grme Masson (Nat. xliii, p. 345, 1891), or of Ramsay (Nat. xliii, p. 589, 1891) had reached me. I have not in any way altered § 29. In fact, what these gentlemen have deduced from the solution behavior liquid-liquid, I had legitimately derived from the solution behavior solid-liquid, as set forth in my own work. My preced- C. Barus—Continuity of Solid and Liquid. 147 Thus the work has a bearing on the nature of solution; for to my thinking, what I have ventured to call cohesive affinities* cannot differ except in degree from the affinities determining valency. At least proceeding on this assumption, I am led naturally to a theory regarding changes of the physical state of ageregation in general, which I will indicate elsewhere. 30. Conclusion.—In the above pages I have merely sought to describe the results directly given by experiment in so far as I understand them, and to draw conclusions which in the light of known facts seemed to be admissible or even obvious. In how far these conclusions are to stand or fall, will depend on similar investigations, to be made with a variety of other substances specially selected with reference to their position in a scale of thermal state. $38. How such selection is to be made, I am now unable to intimate. Substances for instance which fuse continually, like glass or sealing wax, might at first sight be referred to positions near their critical temperatures : but I believe these cases are mere solution phenomena of rela- tively small interest. At all events at the outset, the experi- ments must deal with bodies of definite, simple and preferably erystalline character, to the exclusion of mixtures. I feel con- fident that in an examination of many types, some will be found lying relatively nearer the critical poimt, while others lie nearer or even beyond the-transitional point; and that if the above method be applied with greater rigor than was done in the present paper, light will be thrown on the long neglected department of fusion and solution thermodynamics as related to pressure. T'rom this stage of progress it will then be possi- ble to approach nearer the next of the kindred phenomena, which I conceive to be nothing less than the kind of hysteresis or higher order of volume lag known as chemical affinity. [Added to proof.—To obviate the occurrence of a bald statement like the last, I will indicate my views on the distribution, or successive orders of volume lags. These are to be sought—I, during the passage of a given atom into the next con- secutive in a scale of decreasing atomic weights; IJ, during the occurrence of dissociation of the molecule, including solutions gas-fluid. They are demonstrable, III, in the region of Andrews’s vapor tensions, including the Alexéef-Masson solutions liquid-liquid; IV, in the region of the solid-liquid phenomena of the present paper, including solutions solid-liquid; V, in the region of solid-solid phenomena categorically distinguishable as “permanent set” (Osmond, Carus- Wilson, Barus). They are to be sought for finaliy, VI, during the passage of a given atom into the next consecutive in a scale of increasing atomic weights. The enumeration is systematic, and inasmuch as VI is virtually identical with I, the inherent nature of these changes is periodic. Hence under suitable ther- mal conditions, and continually increasing pressure, the evolution of atoms, of molecules, of changes of physical state, are successive stages of periodically recur- ring hysteresis. ] ing paper is at fault only in postulating an unnecessary change of hydration of the silicated water (l. c., p. 116). i It is gratifying to note that evidence of the similar solution behavior solid-solid is forthcoming, and to be found in the work of Osmond, of C. A. Carus-Wilson (Phil. Mag., xxix, p. 200), and of myself, as I have already pointed out (Phil. Mag., xxxi, pp. 26-28).] Pid: ps 115. 148 G. H. Stone—Asphaltum of Utah and Colorado. Art. XIIL.—Wote on the Asphaltum of Utah and Colorado ; by GrorGE H. STONE. _Durine the past year the writer has visited all the known asphalt fields of western Colorado and northeastern Utah, save those situated within the reservations of the Ute Indians, and two other exceptions: noted below. It is intended at some future time to complete a map of the asphalt exposures and to publish a more detailed account of them than is possible in this preliminary paper. Petrography of the Deposits.—The following named classes of deposit are represented : 1. Asphaltic sand-rock, known also as sand-asphalt and bitu- minous rock. This is the most abundant of all the asphaltic deposits. It consists of a sandstone the grains of which are in contact with each other and the spaces between the grains are wholly or partly filled with asphaltum. The proportion of as- es varies up to about 15 per cent by weight and 27 per cent y volume Of course sandstones will contain different pro-- portions of asphalt in their inter-granular spaces since those spaces depend on the sizes and shapes of the constituent grains and often on the presence of other cementing sub- stances. When a bituminous sandstone contains more than about 15 per cent of asphalt, it may be assumed that it has not been under pressure of superincumbent rock suflicient to cause the grains to come in contact with each other.* The thickest stratum of fully charged rock that I have seen was near 40 feet in thickness. Usually the strata of high grade rock are not more than 4 to 10 feet thick and they alternate with lower grade or barren strata of sand-rock, and sometimes with marls, shales or limestones. Hence the amount of rich rock,—‘‘ pay rock ’’—has often been enormously over-estimated, no account having been made of the poor strata. This is par- ticularly the case with some of the published accounts of the asphalt beds of the valley of Ashley Creek, Utah. 2. Bituminous Shales or Marls.—Black or blackish marls or shales cover large areas both in Colorado and Utah. The richer layers have the smell of asphalt, though like Wurtzilite, they are difficultly soluble. The specimens examined by me * The analyses of the California bituminous rock given in Seventh Annual Report, Wm. Irelan, Jr., State Mineralogist, Cal., 1887, pp. 51-53, show from 1:10 to 8 per cent of fixed carbon, and of volatile carbonaceous matter from 9°40 to 46°20 per cent, with small proportions of lime, ete. An asphaltic sand con- taining so large a percentage of asphalt as afforded by some of these analyses would probably have been produced by a relatively small quantity of sand being washed or drifted on to an outflow of soft asphalt. G. H. Stone—Asphaltum of Utah and Colorado. 149 yield no paraffin, or at most a mere trace to solvents and boil- ing water, and a considerably larger proportion after destruc- tive distillation. They approach cannel coal in composition, but contain a very large proportion of ash, so that none of them contain more than 10, or perhaps 20 per cent of carbon- aceous matter. The richer layers are commonly known in western Colorado as “ oil rock,” and burn readily with a bright, furious flame, leaving pieces of shale having the same size and shape as they had before being burned. These facts indicate that in their natural state these bituminous shales (they all contain so much lime as to be more nearly marls than shales) are asphaltic rather than paraffinic rocks. The richer layers are seldom more than 4 feet thick and are found in the midst of low grade rocks (shales, marls, and limestones). At one place I noted ten of the rich layers each two to four feet thick, distributed at intervals through about 400 feet of rock. 8. Bituminous Limestones.—Limestones and marls consti- tute a large part of the Tertiary rocks of the region under- discussion, i. e. of the asphalt-bearing formations. Almost all the limestones are somewhat bituminous, and some strata will burn like the shales.. They are colored from gray to yellowish- brown, light color, rather than blackish like the bituminous shales. Usually they do not contain distinct fossils, but are often oolitic, pisolitic or -coarser coneretionary, i. e. they are semi-crystalline. Fetid layers are not rare, and some of them are particularly offensive. Cavities in the bituminous lime- stones are often filled with hard asphalt, in some places taking the form of Wurtzilite, in other places Uintaite. The color of the Uintaite varies from the deepest black to brown and even gray-brown. The lighter colors are found in the centers of | the lumps or in the cavities less open to the air. I have found asphalt in five classes of cavities in these limestones. 1. In small irregular or somewhat amygdaloidal cavities in fine granular limestone and having no visible outlets. 2. In fractures that cross the strata for only a short distance (gash veins of the miners). 3. In deep fissures (true fissure veins). 4. In eaves or channels of subterranean streams, in which the asphalt was brought in after the stalagmitic growths were _ completed or nearly so. 5. In the interior of shells, or in the cavities found in the centers of concretions and nodules contained in the limestone. The limestone yields on destructive distillation several per cent of volatile and combustible carbonaceous matter. In all cases un- less in the fissure veins and stream caves it is evident that the as- phalt must have been derived from the country rock, 1. e. a bitu- minous liquid oozed out of the limestone into the cavities. Since 150 G. H. Stone—Asphaltum of Utah and Colorado. the limestone is of a light color, this liquid must have acquired its dark color during the process of being changed into hard asphalt. This conclusion is confirmed by the lighter color of the least exposed asphalt. Evidently the bituminous matter that is now in the rock is not in the condition of ordinary black asphalt, but the liquid which oozed out of the rock was capable of being changed into such asphalt, hence the bitumin- ous limestones may well be classed with the asphalt-producing rocks. The rather light color of some of these masses of hard asphalt, which have all the properties of Uintaite except the deep black color, suggests the question whether the color of asphalt be not due to disseminated fixed carbon, in a state ap- proaching charcoal, the product of partial oxidation, more than to the natural color of the hydro-carbons proper. I began some experiments and analyses to determine this point, but the work is incomplete.* - 4. Outflow or Overflow Asphalt.—Under this class are here included all forms of asphalt that have oozed out of the rock that originally contained them. Some of these had the black color before, others have acquired it since the outflow. I leave it as an open question whether these oils were true asphalts before acquiring the black color.t+ Mineralogically the outflow asphalts present the same difli- culties of classification as do the petroleums. There are per- haps a dozen different grades in Utah and Oolorado that might be described as distinct minerals by those on the alert for new species. The more important generic terms (they are all generic rather than specific) are the followmg: 1. Maltha, asphaltic tar, brea, mineral tar or pittasphalt. Here are in- cluded the viscous liquids. In Utah they all have an aromatic odor and black color. By degrees they harden to a solid, sometimes tough, waxy or horny, sometimes brittle. The * Mr. 8. H. Gilson, of Salt Lake, informs me that he has obtained by distilla- tion of the limestone out of which a mass of Wurtzilite had oozed, a dark yellow- ish tarry material that closely resembles and appears to be identical with the distillate from the Wurtzilite. + That the lighter constituents of petroleum can be changed to more viscous oils by protracted exposure to oxygen, appears to have been proved by experi- ments made some years ago by W. P. Jenney. The same conclusion is enforced by the hardening of the brea of California, also by the finding of asphalt in eavi- ties in the Devonian and Silurian petroliferous rocks (see Report of Professor Edward Orton on the Trenton Limestone as a source of Natural Gas and Petro- leum in Ohio and Indiana. Eighth Ann. Report Director U. 8. G. S., 86-87). Such asphalt cavities have been observed by Shaler, Newberry, Linney, Orton and others. In the present state of the argument it is permissible to assume as a working hypothesis that the harder asphalts were derived from the softer or pittasphalts, and they in turn from more liquid bitumens. under exposure to the air or perhaps to aerated waters. How much this is quantitatively due to oxygen or other chemical agencies, and how much to evaporation of the lighter com- pounds, remains to be determined G. H. Stone—Asphaltum of Utah and Colorado. 151 hardened outflow is known as outflow or overflow asphalt. The maltha is found in small pools, or spread over the ground and often penetrates the spaces between the broken rock of the talus or sub-soil in a complex network of stringers, small veins and sheets. 2. Uintaite or Gilsonite. hee Fe 3 Bini. atl ie? we oo * A he Pe: a re ie Lee thaatel a oat vs Rarern yee: a eens et tien re er pan fs vil Fa aie te st _ AMERICAN JOURNAL OF SCIENCE. FOUNDED BY PROFESSOR SILLIMAN IN 1818. EDITORS: JAMES D. DANA and EDWARD Ss. Dana. RY Associate Editors: J. P.CooKe, Tr., GEORGE L. Gosonik: and JoHN TROW- DGE, of Cambridge, H. A. Newron and A. E. VERRILL, of Yale, and G. F. BARKER, of the University of Pennsylvania, Philadelphia. _ ‘Two volumes of 480 pages each, published annually in MONTHLY N UMBERS. wid his Journal ape Jirst series of 50 volumes as a quarterly in 1845, and its d series of 50 volumes as a two-monthly in 1870. The monthly series com- enced in 1871. Twenty copies of each et communication are, if requested, in advance struck off f for the author without charge; and more at the author’s expense, provided the iD ae of copies desired is stated on the manuscript or communicated to the printers of the J ournal. = ‘The title of communications and the names of authors must be fully given. At ticles should be sent in two months before the time of issuing the number for which they are intended. Notice is always to be given when communications “offered, have been, or are to be, published also in other Journals. * Subscription price $6; 50 cents a number. A few sets on sale of the first oF and ‘second series. - Ten-volume index numbers on hand for the second and third series. The index * volume XXXI to XL (3d series) was issued in January, 1891; price 75 cents. 4 - Address the PROPRIETORS, _ UNIVERSAL ENGLISH-GERMAN AND GERMAN- ENGLISH DICTIONARY. ow by, Dr, FRLrx FLUGEL. j A y Ath edition, edn: revised, of Dr. J. G. Fligel’s Complete Dic- tionary of the English and German Languages. Complete in 12 numbers of 14 signatures each; a number appears _ monthly. Subscription price for each number, 3 Marks. GEORGE WESTERMANN, Braunschweig, Germany. B. WESTERMANN & Co., New York City. CHE Y ate otes on a Reconnaissance of the Ouachita Mountai ART. x. —Some of the features of non-voleanic Igneous. Eje tions, as illustrated in the four “Rocks” of the ‘New ens: Haven Region, West Rock, Pine Rock, Mill Rock a a’ East Rock; by Jamus D. Dana. (With Plates II to — System in Indian Territory; by Roperr T. Hit ere XII.—The Continuity of Solid and Liquid; by ‘Cart Barvus_ XIII-—Note on the Asphaltum of Utah ut Colorado; by oY IG BORGER FEL 1 @ Nin ie eS Oe Ors ena Fee ae XIV.—Photographic Investigation of salen Prominences and their Spectra; by Grorcs E. Hatz. (With Plate VUL) 16¢ XV.—A Gold-bearing Hot Spring Depaits by Watrer gs Hapvey. WERD 2 £0 ee ee eee at XVI.—AprEenvix—Restoration of Siewasanree : by, oe ©. es Marsu. oe Plated ete er eee ae Pe z SCIENTIFIC INTELLIGENCE. Chemical Reactions, LIEBREICH: A new Reaction of Carbon monoxide, Burrus LOT, 170. | US ea Geology—Relations of the Eastern Sandstones of esr denee Point to the Lowe Silurian Limestone, M. E. Wapsworts, 170.—Expedition to Mt. St, Elias in» summer of 1890 by ISRAEL OC, RUSSELL, 171.—Glacier scratches south of 1 ) ‘terminal Moraine” in Western Pennsylvania, P. M. Fosuay and R. R. Hice: Losses of Cape Cod by sea-encroachments, H. L. MARINDIN, 172. —Der Pel Pelo- eenee Versuch einer Landeskunde auf geologischer Grundlage, A. Pun LU SON, 173. Botany—Botanic Gardens in the Equatorial Belt sae in the South Seas, juste Misceilaneous Scientific Intelligence—Die Denudation in der Wiste und ihre geolo- | gische Bedeutung; Untersuchungen tiber die Bildung der Sedimente in den Aegyptischen Wiisten, J. WALTHER, 177.—History of Volcanic Action in ‘the area of the British Isles, A. GEIKIE: Magnetic Declination in the United | State for the Epoch of 1890, ©. A. ScHorr: Telescopic Work for Starlight Evenin; W. F. Denning: Ostwald’s Klassiker der Exacten Wissenschaften, 178). ve ERRATUM.—Page 108, bottom line, for one and a half, read three. Chas. D. Walcott, 7 oe ii tS ae U. S. Geological Survey. 3 hay, oa 5 ath nes rege oe ee is TEMBER, 1891 : Established by BENJAMIN SILLIMAN in 1818 . on, : ct T H E \ j i 4 EDITORS. | i JAMES D. anv EDWARD 8S. DANA. | Bee ASSOCIATE EDITORS ae. " Prowessons JOSIAH P. COOKE, GEORGE L. GOODALE ann JOHN TROWBRIDGE, or Camsriver. " 43 4 | Prorzssons H. A. NEWTON anv A. E. VERRILL, or ; 3 New Haven, | re. GEORGE F. BARKER, or Pamaperputa. | THIRD SERIES. | VOL. XLIL—[WHOLE NUMBER, CXLIL] ee. No. 249.—SEPTEMBER, 1891. a NEW HAVEN, CONN.: J. D. & E. S. DANA. oY ay a AS PO a ae | TUTTLE, MOREHOUSE & TAYLOR, PRINTERS, 371 STATE STREET. . op - Published monthly. Six dollars per year (postage prepaid). $6.40 to foreign sub- é id eethers of countries in the Postal Union. Remittances should be made either by Z i money ea registered letters, or bank checks. snl Sieg cx = a S =e : - Pies i ie = arn. = * “Ye Os wi ey ne x soe apenas Pe OE Files EN IR Pe ss * a one Fie? ae . es wa) any OE EY AD 2 - Diablo, Arizona. The largest mass weighs 201 lbs., and anaes 154 tbs. by him Page the Academy of Natural Sciences. gave 25% ' per cent. nickel, and in the only piece which has yet been cut eo 1S. : exposed containing small diamonds. As this is the first instance of » the oc ar-- though of no commercial value. Professor Foote was assisted in his sear pa _ five men, and secured every specimen that had been found before te a4 during the past seven months: ‘: purchased by me at the locality was only about two inches square, Rue oe Pyramids. be tks From Arizona.—The new species described by Professor Koenig, Pipe’ aoe conite, an oxide of copper, resembling Anatase. in fine large crystals. Footeite, an it oxy-chloride of copper in blue crystals. Chalcotrichite, Azurite, Cuprite Crystals, Ait Chrysocolla, Descloizite, Vanadinite, Yellow and Red Wulfenite, Free Gold, Embolite, ete, — +s e vate _ METEORITES. Professor Foote personally collected some remarkable. met orl rence of this gem in meteoric iron, the find will be of great interest to scien ite . except parts of a 40-lb. one that had been broken up. : ee. Hence, in spite of the great interest attached to them, they are offered a following low prices: Small complete masses, 25c. to $5.00; larger masses, p10 0 to $50.00; polished surfaces about 257 in advance. MINERALS—N EW ARRIVALS. Bena for circular ‘giving full description. The following minerals have been collected by Professor Foote : at the ocali From Mexico.— Cuprodescloizite, now Hse cia as a divi ‘sueeeane and botryoidal at one-tenth the price ever sold before, 25c. to $5. 00. Aguilarite, the new sulpbo- “selenide of silver in crystals, some quit $1.00 to $10.00. *% I am promised ‘a supply ‘of larger specimens of. this species, The larges tals were as fine as any seen. ya _ Argentite crystals and groups, g500, to $LO. 00. Pyrargyrite crystals and groups and massive pieces, 50c. to $50, 00. One very fine crystal, about 1} inches long and 4 inch in diameter, shows fine red translucency. This is one of the finest epecrnepe ever found in Mexico, and is worth $50.00. pa ine Acanthite crystals, $1.00 to $5.00. Lr eee Calcite in greater variety of twinning forms, and more beautiful than have. ‘tre been seen before. Polybasite, Stephanite, Embolite, Cerargyrite, Amethyst, Quartz, with moving bubbles. Obsidian implements and antique pottery er eos hi on a “~s’ a en mes From California. — Hanksite, Cinnabar and Sulphur crystals. Thenardite, Te markably well crystallized Gold, etc., ete. ait From Colorado.—Beautiful Blue Barite in crystals and groups. Patri, Hessite, Gold, brilliant Pyrite, etc.. etc. ig PRE Garnets from Salida, perfect dodecahedrons, from 10c. upward. — eis weigh: * ing 54 Ibs. $5.00 to $7.50. ae From Utah.— Olivenite, Conichalcite, Clinoclasite, and all Tintic minerals. Uintah: — Re ite, Salt in fine cubes, Brochantite, etc., ete. SP aes best at From Nevada — Thinolite, Gay- Lussite, Boras, Glauberite, ete., etc: 7 ae, From New Mexico.— Yellow Wulfenite, Cerussite, Anglesite, F. los Ferri, Trurquots, SS ores Satin Spar, Chrysoprase, Descloizite, etc., etc. Send for our circular giving descriptions of Laurionite, Nadorite, \MalamieNlogtie® and other rare and beautiful minerals, which were collected by Professor Foote in — Africa, Spain, Italy, Greece, and other parts of Europe in 1889 and 1890. Our 100-page Illustrated Catalogue of Minerals free to all purchasers. To others: Light paper, 5c.; heavy paper, 10c. pe fo Send for our 32-page list of Books on Geology. fii pare i . PH E AMERICAN JOURNAL OF SCIENCE [THIRD SERIES.] ———__ + 9+ ——__—_—__- Art. XVII.—On the Capture of Comets by Planets, especially their Capture by Jupiter; by H. A. NewrTon. 1. Some years ago I obtained and published* a formula ex- pressing in simple terms the total result of the action of a planet in increasing or diminishing the velocity of a comet or small body that passes near the planet. This formula is practically a modification of the integral of energy, the smaller terms in the perturbing function being omitted. A very brief and partial treatment of it was presented to the British Asso- ciation for the Advancement of Science in 1879 at its Sheffield meeting.t Within the last two or three years several astrono- mers have made special study of the manner of Jupiter’s action in changing the orbits of comets that pass very near him. M. Tisserand has given us an expression connecting the major axis, inclination and parameter of the orbit described before coming near to Jupiter with the corresponding elements of the orbit after leaving the neighborhood of the planet.{ M. Schulhof has applied the formula of M. Tisserand as a criterion for de- termining the possible identity of various comets whose orbits pass near to Jupiter’s orbit.§ Messrs. Seeliger, Callandreau and others have continued these investigations. The interest thus shown in the problem has led me to resume the study of the subject, and to work out the results of the formula obtained by me in 1878 more fully than they have been hitherto devel- oped. * This Journal, III, vol. xvi, p. 175, 1878. + Report, 1879, p. 274. + Sur la théorie de la capture des cométes périodiques, Bull. Astron., Tome vi, juin and juillet, 1889. § Notes sur quelques Cométes a courte période, Astron. Nachrichten, No. 2964, Am. Jour. Sci.—THIRD Series, Vou. XLII, No. 249.—SEpreMBER, 1891. So ee 4 Oe nig 3 / Ueki 184 ZH. A. Newton—Capture of Comets by Planets. 2. One of the remarkable distinctions between the comets of long (or infinite) periods, and those of short periods, is that the orbits of the latter have almost without exception direct motions and small inclinations to the plane of the ecliptic, while the orbits of the former have all possible inclinations between 0° and 180°. At first sight this seems to imply that the two groups of comets are radically distinct in origin or nature one from the other. The most natural line of investi- gation therefore is the effect of perturbations in bringing or not bringing the comets to move with the planet after the perturbation. 3. The algebraic processes by which was obtained the form- ula for the change of energy which a small body experiences from passing near a planet were given in the article cited, and they need not be here reproduced. The following was the resulting equation, viz: 4mfu'v, COS @ sin a i Pr, si and it was obtained from the general differential equations of motion by making assumptions not greatly differing from those used in obtaining Laplace’s well known theorem, that a sphere of suitable magnitude may be described about the planet as a center and that for a tolerable first approximation the comet may be regarded as moving when without this sphere in a conic section of which the sun is the focus, and as moving when within the sphere in a conic section (an hyperbola) of which the planet is the focus. In other words, only perturba- tions of the first order of magnitude are taken account of. A comet is treated throughout this paper as a small indivisible body whose mass may be neglected. 4, Notation. The symbols used in (1) and also other sym- bols which I shall have occasion to use may be thus defined. i Let €, be the orbit of the comet about the sun before the comet comes under the appreciable action of the planet ; € the orbit of the’comet about the sun after perturbation by the planet ; the hyperbolic orbit of the comet relative to Jupiter when near the planet ; the elliptic orbit of Jupiter about the sun ; the point on @, which is nearest to J; the point on J which is nearest to € ; the length of the straight line EA being the perpendicu- lar distance between the orbits at their nearest ap- proach ; the angle between the tangent of €, at A and the tangent to J atE; Qmprw © 8 H. A. Newton—Capture of Comets by Planets. 185 Let ibe the distance which the planet has yet to pass over to reach E when the comet is at A (A may be negative) ; m the mass of the planet, sun’s mass=unity ; a the unit of distance, in general the mean distance of the earth from the sun; the sun’s attractive force at the unit of distance ; the planet’s velocity in its orbit at E; ) the comet’s velocity in its orbit C when the comet en- ters the sphere of Jupiter’s perceptible influence ; ees fe) ~ v the comet’s velocity at A relative to the sun ; 0) 8: @, the semi-axis major of (7, (negativeif (is an pay @ the semi-axis major of € (negative if €@ is an hyperbola) ; p the perpendicular from the planet upon asymptote to C; a the acute angle between the transverse axis of C and the asymptote to C. p the angle between the tangent to J at O (drawn in the direction of the planet’s motion) and the line from the planet to the vertices and center of C; A the semi-transverse axis of C; B_ the semi-conjugate axis of C (hence equal to p) ; the distance of the planet from the sun; _ the distance of the comet from the sun; r, the distance of the comet from the planet ; p,and p distances of the comet from the sun at selected epochs before and after perturbation ; u,and u the velocities of the comet at the selected epochs ; 2a, Lingfar r SIS A the increase to which v*—~ receives by - the ° / ie} ° e e planet’s action during the whole period in which the comet is passing near to Jupiter. 5. If we assume two epochs, one before and one after the perturbation, at which the comet is equally distant from the planet, the term 2mfa’/r, is the same at both instants, and it disappears from the value of A. Therefore 25, 2fa* ae ee But by the well-known formulas from the law of gravitation, 1 I Utes fa = Sy ; fe (| = va 1 and =) TO (= _ ca} 1 1 hence A = fae al 3) 1 1 4m cos p sin a that is, from (1) Grr ar: i 186 H. A, Newton—Capture of Comets by Planets. | This equation is valid whatever be @, the major axis of the orbit €,, and may be used to determine the major axis of either orbit from the elements of the other. My present pur- pose is, however, to study the action of Jupiter in changing orbits that are originally parabolas, and hence in general @, will be taken infinite. In that case abs we Se 4m cos @ sin a (2) It will be found that the second number of ( (2) depends on wo, d and A, and these are known quantities when the elements of @, and "3 are given. The use of the equation is moreover oreatly simplified and enhanced by the fact that the plane of the planet’s orbit is involved only in so far as that it must contain the tangent to J at E. 6. In the second member of (2) all the factors are positive except cos g, hence, if ¢<$7, @ is positive and the orbit & is an ellipse; but, if gy >$7, @ is negative and € is an hyperbola. This result may be thus expressed; 2f the comet passes in Front of Jupiter the kinetic energy of the comet is diminished, of it passes behind the planet the kinetic energy of the comet is encreased. 'The reason for this may also be given in general language. If the comet passes in front of the planet the comet’s attraction increases the velocity, and hence increases the kinetic energy of the planet and vice versa. But the total energy of the two bodies is constant ; so that when that of the planet is increased, that of the comet is diminished and vice versa. 7. It is desirable now to transform the value of @ given in equation (2) so as to be able to determine the major axis of the new orbit of the comet directly from the circumstances of its initial approach to the planet before perturbation; in other words, to find @ in terms of , d and A. For this we must find in terms of w, d and A, values for s ,p, a and So find s.—In fig. 1 let A and E repre- sent the two points A and E as defined above (Art. 4), and the line AE represent d. Let AY be the tangent to @, at A, and HO the tangent to J at E. It is an admissible sup- position that the planet is describing the straight line OK, and that the comet in its unperturbed orbit is describing the stracght line YA. At some certain moment the line joining the planet and the unperturbed comet must evidently be perpendicular toOE. Let OY be the line joining the bodies at that moment, so that the planet is at O when the comet is at Y, and EOY is a right angle. Instead, however, of supposing the i. HH. A. Newton— Capture of Comets by Planets. 187 planet to move from O towards E we may apply an equal, opposite motion to the comet, and consider the planet to remain at rest at O. Draw AC parallel to EO and make AB equal to the distance described by the planet during the time that the comet is moving from Y to A. Join YB. Then since YA and BA represent in direction and magnitude the motions of the two bodies in a given interval, the third side YB of the triangle represents in magnitude and direction the motion of the comet relative to the planet. The angle YAB is the angle », and the three sides of the triangle YA, YB and BA are proportional to v, v, andv, Let the angle YBCO be @; then from the triangle Y AB we have VU, =U, —2¥,v cos W+’, and , U:V0,1¥,::8in 9: sin (@Q—o@): sin @w. (3) Since v and v, can be computed from the given elements of the orbits of the planet and comet, we may readily compute from @ the value of s, orv,/v, But if the planet is at its mean distance from the sun, and the comet’s orbit is parabolic, vy = 2v’, and we have s* = 3 — 2,/2 cos ow. (4) Also from the triangle 2u, =v," + 2v,v,cos 0+ ,’, or 2s cos 9@=1—s", (5) 9. To find p.—The planet being regarded at rest at O and the relative unperturbed motion of the comet being along YB, this line may within admissible limits of error be treated as one asymptote of the relative orbit C. The perpendicular from O upon YB will then be by definition (Art. 4) the line p. Draw OX from O perpendicular to OY and OK, and let these three lines be codrdinate axes. Let the line AB meet the plane XOY in C. Join OC, let fall OD perpendicular to YB, and join CD. Since EA is perpendicular to AY and also to EO, and so to its parallel line AC, therefore it is perpen- dicular to the plane YAO. Hence OC, parallel to EA is per- pendicular to the plane, and so perpendicular to CD. Again CDY is a right angle; for OD’+DY’=OY’=O00°+CY’, and ~OD'=0C*+DC*. Hence DC’?+DY’=CY’, and consequently CDY isa right angle. The quantity A is the line BO; for hf is the distance which the planet, when the comet is at A, has yet to pass over before reaching K. But the comet was at Y when the planet was at O, and the planet describes BA, while the comet describes YA, leaving BC as the distance yet to be described or 4. But the angle CBD is @, so that we have p =O = OC" = CD? =a" +h? sin’ 6. (6) 188 H. A. Newton—Capture of Comets by Planets. 10. Zo jind a—The angle a is the acute angle between the asymptote and the transverse axis of the hyperbola, and hence from the nature of the hyperbola tan a=B/A. By known formulas we have, if the ska is at its mean distance vou 3) 1 1 Dee = ae + oa) Therefore 6. = eee or A= ess ] Os A Ss (7) Bp s(@4hisin’®) ( Hence from (6) tana= oe aoe ae sol 11. Zo find yo.—The orbit of the comet relative to Jupiter lies in the plane YOB. Let 2 be the inclination of the plane YOB to YOX, measured positive from x positive to 2 positive ; let 2 be the longitude of the direction YC, measured in the plane YOX from OY, that is, the angle made by YC with OY produced; let % be the longitude of the direction YB mea- sured in the plane YOB from OY, that is, the angle made by YB with OY produced. Imagine now a sphere deseribed about Y as a center that shall cut the three planes XOY, BOY and BCY in three sides of a right angled spherical triangle. The hypotentise of this triangle is A, the base /, the perpen- dicular 47—0, and the angle opposite to the perpendicular is 23 hence we have cos A = cos / sin 8, (8) cos 6= sin z sin A, (9) cot 7 = sin é tan @. (10) Also from the triangles OCY and BCY | OC d tan /= tan OYO = — Vie Ree (11) The angle g is by definition the angle between the direction OE, and a line in the plane YOB that makes with YB an angle a. Hence we have readily cos p= sinz sin (A+ a). (12) These equations enable us to compute ¢ in terms of d, A and @; for in succession @ may be computed by (3), l by (11\, rX by (8), az by (10), and @ by (12). 12. These values of s, p, a and @ give by equation (2) the value of @. The suppositions that the planet is at its mean distance, and that €, is a parabola, are involved in that equa- tion, but they are not necessary to the determination of @ when no such hypotheses are made, and changes in the equation H. A. Newton— Capture of Comets by Planets. 189 that are not serious would make it applicable without these limitations. The quantities in the several equations may be regarded as having values :— d positive, h positive or negative, @ positive and less than $7, a, 9, p and @ positive and less than z, Zand A positive and less than 27. 13. We may, however, also find directly the value of @ in terms of d, h, and the known functions of o. From (12) cos Psin a=sin?2 sinA cosasina+sinZ cosA sin’ a. From (7) cos asina= - and sin* a = a a a= 73, By ek Fe Fr From (10) and (8) : in 6 in 6 ee cos / sin ( a Ravi _coté sin 6 z (1+sin*/ tan’ 6)? (sec’ 6+ cot? l)? hence from (6) and (11) an h sin® 6 h sin’ 6 SIG? LOS An ==)4- = p= +e (d° +h’ sin’ 6)? From these and (9) cos @ sin a (A* + B*) = AB cos6+ AB sin’ 8, and hence from (2) ae ea _ 8 Att¢d*thisin'd |, en “A cosO+hsin®? 6 4m °> AcosOtAsin’ (13) Since m is the known mass of the planet, and @, s and A are known functions of », equation (13) gives directly the value of @, the semi-axis major of the new orbit € in terms of d, A and o. 14. For a particular case of approach, equation (13) is con- venient for computation. We may, however, now treat d, A and @ as independent variables whose varying values may express all the different possible cases of approach of the comet to the planet so far as change of periodic time of the comet is con- cerned. The dependence of @ upon the three variables cannot be very easily represented graphically in a single plane dia- gram. But by giving to successive values in multiples of MO viz: o— MW) 20" 302) ete, to 170°, 1 have prepared a series of diagrams to exhibit in each case in succession the relation of @ to the other two variables. The values of 0, s and A for the several values of were needed in making the diagrams and they are given in Table J. Equations (4), (5) 190 HA. A. Newton—Capture of Comets by Planets. and (7) are used in making the table. The disturbing planet is assumed to be Jupiter, so that m was taken equal to 1/1050 and 7=5'2. TABLE I. @ 6p Le Vig are Vie igs 6°. 41 A | te be Kae at 0° O° 307) Oral 02886 || 100° | 231° 487) i868 "00142 LOS: 32 1 0°463 "02309 | 110) 4 ASS. 1:992 “00125 20 55 «4 0°585 "01448 || IDO. | Ae | D1 27101 00112 30 ey) 0°742 -00900 | 130°>7\— 250 267)" Bales ‘00103 40 84 46 0913 00594 | 140 +} 156-26 | 2°23 00096 50 | 94 47 1:087 "00419 | 150 9} 162,, 22°) Bee "00091 60 LOS 220 1°259 "00312 || 160 168 16 2-379 | -00088 70 Se 1°426 00244 LUO en es See: 8 2°405 "00086 80 lash Ar 1°584 00197 || 180. | 180. 0.) 3220 “00085 90 125 > 6 wee ‘00165 | Fe 15. Using these values of 6, s and A we may now represent graphically the dependence of @ upon the other two variables d and / for each specified value of w.. Let d and A be Carte- sian coordinates, then for each point of the codrdinate plane there is a value of @. The ambiguous sign will be fully satis- fied by giving positive and negative values to A. For an assumed value of @ we shall have a curve whose equation is (13), and each point of this curve represents values of d and A for which the total action of the planet upon the comet will be to reduce the energy of the comet a constant amount. This locus will be called an zsergonal curve. 16. Laisceau of isergonal ellipses.—The equation (13) of the isergonal curve may be written 4m@ (A cos 6+/ sin’ 6) = s(A’*+d*+h’ sin’ 6), and this is the equation of an ellipse. As @ changes its value we may treat it as a parameter and we have a faisceau of simi- lar isergonal ellipses, each ellipse symmetrical with the axis of h. The radical axis of the faisceau A cos 0+A/ sin* 0=0, and the imaginary ellipse A°+d@°+/’ sin* @=0, are theoretically two members of the faisceau. For points on the radical axis @= « and therefore for this locus there is no change in the energy of the comet. 17. Center and area of the isergonal ellipse.—The center of the isergonal ellipse is upon the axis of 4; making d=0, and solving for 4 we have 1 2n@ 2m@ As Of —— K — — ~- 14 0 Ss de $ sit AC (cos oe ; ( ) The first term of the second member of (14) is the ordinate of the center, and the second term is the semi-axis major of the H. A. Newton—Oapture of Comets by Planets. 191 ellipse. The ratio of the axes being 1:sin 0, and As’ being =mr, the area of the ellipse will be equal to eee (: — (cos a) ) s* sin 0 2as 18. Maximum action of the planet.—For two particular values of @ the isergonal ellipses become points. These values of @ result if the maximum effect of the planet in increasing and in decreasing the energy of the comet takes place, and they are obtained by making the two values of / equal to each As : na +1. Since at other in (14), that is, by making cos 6— the same time h=2m@/s, we obtain A As = cos 6-+- 1’ eo 2m(cos 6+ 1) ae) Let fh’ and 4”, and @’ and @” be the positive and negative values of 4 and @ in (15) and we may construct the following table of their values. As in Table I Jupiter is assumed to be the perturbing planet. TABLE IT. i | | | | oO / h’ h”’ @’ ‘ @” / a h’ | h”’ @’ @” | | | | 0°; -01443 a SoA 0 oC | 100° :00426 —-:00085 4:17 —0°83 10 "01250 |— 15174 3°04 —36°90 | 110 | 00489 —-00072 5°12) —0°75 20 | -00927/—-03307 2°85 —10°15 120 -00598 —-00062 660 —0-68 30 00690 |—-01290 2°69 | — 5-03 || 130 | -00789| —-00055) 9-09 —0°63 4Q | °00544|!—-00654; 2-6] — 3°13 || 140: | -01149; —-00050| 13°71) —0-60 50 00457 |—-00387 261 — 2°21), 150 -01934 —:00047 23°70 —0°57 60. 00407 |—°00253 | 2°69 | — 1°68]; 160 | -04192 —00044 52°36 —0°55 70 | -00382 |—"00179,, 2°86 | — 1:34) 170 | °16336 —-00043) 206-30) —0°54 S05) 00377 | — "00134 | 3°14 | — I-11 || 180 aK — 00043, a — 0°54 90 | 00390 |—:‘00105 3°55 — 0°95 | 19. Explanation of Table 1[.—The meaning of the numbers in this table may be explained by an example. If a comet moving in a parabola passes near to Jupiter, and the directions of the two original motions at nearest points of the orbits make an angle of 10°, then the greatest action of Jupiter (during the whole period of transit) in diminishing the velocity of the comet in its orbit about the sun will take place if the two orbits actually intersect (d=0), and if the comet in its unper- turbed orbit arrives first at the point of intersection at the instant when Jupiter is distant therefrom :01250 (the earth’s mean distance from the sun being unity). The resulting semi- axis major of the comet’s orbit about the sun will be 3:04. On the other hand, the greatest effect in increasing the velocity of the comet will take place when the two orbits 192 H. A. Newton—Capture of Comets by Planets. actually intersect, and the comet in its unperturbed orbit reaches the point of intersection later than the planet and when the planet is distant therefrom 0:15174. The semi-trans- verse axis of the resulting hyperbolic orbit about the sun will be 36:90. 20. Lesulting orbits of maximum perturbation.—The posi- tion of the relative orbit about Jupiter in these cases of maxi- mum perturbation for given values of @ is easily determined. From: the equations (7), (6) and (15) tan a=B/A=/A sin@/A=sin @/(cos 0-1). The positive sign gives 2a=0, and the negative sign gives 2a=7+0. But the angle 2a in the first case is the angle of the asymptotes enclosing the branch of the hyperbola described about Jupiter by the comet. Since the two original orbits intersect, the plane of the relative orbit contains the planet’s path, so that the comet passes directly in front of the planet and being turned backward leaves Jupiter exactly in the direc- tion of Jupiter’s quit.* The place of encounter with Jupiter will be near an apse of the comet’s resulting orbit about the sun. The comet leaves the planet with the relative velocity v,, 80 that if s<1 the motion about the sun in the new orbit will be direct; if s>>1 the motion in the new orbit will be retrograde. That is, by (4) when < 47 the resulting motion is direct; when > {7 the resulting motion is retrograde. In the second case the angle 2a, being greater than 180°, stands for the angle between the asymptotes exterior to the orbit. Hence the comet passing behind the planet will be turned forward and will leave the planet in the direction of Jupiter’s goal, and have a velocity that will send it perma- nently out of the solar system. 21. The results of Art. 20 assume that » is given. To find for what value of the period of the resulting orbit is the shortest possible we may put As’=mr and 1—s’=2s cos 0 in (15) so that r Gy eal ie Se U To find the minimum for @ place — =0 in this equation. This gives s=+1, in which result since s is inherently positive only the positive sign is used. But when s=1, @=37, h=mr and w=4a. Hence the greatest effect of perturbation of a planet moving in a circular orbit in shortening the periodic tume of a comet originally moving in a parabola is obtained of the comets original orbit actually intersects the planet's orbit at an angle of 45°, and if the comet is due first at the * The goal and the quit of a moving body are those two points on the celestial sphere towards which and from which the body is moving. H. A. Newton—Capture of Comets by Planets. 1938 point of intersection at the stant when the planet’s distance therefrom ws equal to the planet's distance from the sun mult- plied by the ratio of the mass of the planet to the mass of the SUN. The relative velocity of the comet on leaving the planet’s sphere of action would be equal to and directly opposite to the planet’s velocity (s=1), and the comet would be left entirely at rest to fall to the sun. This case could not happen for planets like the earth where mr is less than the semi-diameter of the planet. In the case of the earth mr is less than 3800 miles, and actual collision would result. But for Jupiter mr is greater than the distance of the second satellite from the planet. The nearest approach of the comet to the planet would be mr (/2—1) which is more than four times the radius of Jupiter. Hence this case of maximum diminution of major axis could occur near Jupiter. | Bie! 2 == 10°: Big) 3+ o= 170". 99. Isergonal ellipse for o=10°.—lf we make #—10° the vanishing points of the isergonal ellipses will be (Table IJ) at d=0, h=-01250, and d=0, h=— "15174. In fig. 2 let OE and OH be the axes of d and f/ respectively. The vanishing points will be on the axis OH at distances /’ and 4” above and below O. Upon this diagram are shown the halves of four isergonal ellipses. The scales used for d and h are not equal to each other, since the use of the same scale for both codrdinates would make the figures of inconvenient shape. In this, and in all the figures 2-18, the unit in d is to the unit in A, as 1 to sin. But to indicate more clearly this scale, and at the same time to give a kind of shading to a part of the area, there are drawn above the radical axis ae lines parallel to OE, and parallel to OH, at intervals of -01; that is, the sides of each of the small rectangles in the quadrant HOE are ‘01, or about | | | 194 H. A. Newton—Capture of Comets by Planets. 925,000 miles. Only the positive values of d are represented in the figures. The positive vanishing point being 1:250 of these divisions above O, and the negative vanishing point 15,174 below O, we lay off Oa=4(/’ + h’’)=—6-962 divisions, and draw ae for the radical axis. The smallest positive value of @ is (Table II) 3:04. As @ increases from 3:04 the ellipse increases in size, and the innermost curve represents what it becomes when @=5. The second curve (separating the blank and shaded areas) corresponds to @=20. Any parabolic comet passing Jupiter with an original angle of o=10°, and having d and / such as to be represented by a point within the blank area of fig. 2 will leave the vicinity of the planet in an elliptic orbit whose semi-axis major is less than 20, and whose period therefore is less than 90 years. Fig. 4; o=20°. Pio. 5; = Tops 0 SS ee ee ee ee Fe ope me cae ey es ed ars eg ee | Gere Ee Ge ear ee es Ee ee RS ee eee = 1 | —} | — ee SR RET SS SES ee oe SN See ee Se eo oe ae ae te ea eee peor ope ee i eed SS f SS a ee eed ae ee ee ae oes ee a GS] (9 (EN () Pe ed EE ee ES RE RT se }| ak \ NON \ « NX \ . SOQ The larger curve that lies above ae in the shaded area is the isergonal ellipse for @=50. As @ increases the lower part of the curve tends to approach the radical axis ae, with which it coincides when @=a. For points in the area below ae (dis- tinguished by the oblique-line shading), the planet increases the velocity of the comets, and the comet would be thrown per- manently out of the solar system. The smallest semi-trans- verse axis, the one corresponding to the vanishing ellipse is (Table II) 36°90, and the isergonal curve for @=—50 is drawn in the figure. H, A. Newton— Capture of Comets by Planets, 195 23. Isergonal ellipses for #=170°.—In figure 3 are drawn the three ellipses corresponding to the values of @, —5, —20, and —50. The ellipses above ae do not appear, inasmuch as the smallest possible elliptic orbit has a semi-axis major of 206°3 (Table II), and a period of about 3000 years. The radi- eal axis ae is ‘08146 (or over 8 divisions) above OE. 24, Figures 4 and 5 are like diagrams for o=20° and o=160°. With altered numbers the explanations of arts. 22 and 23 apply with slight change to these figures. The line ae’ in figs. 4 and 5 is nearer to OK than is the same line in figs. 2 and 3. In Hig. 6 +030". Fig. 7; o=150. H fig. 4 the line for @= —20 appears below ae, while above ae are the three curves for +5, +20, and +50, respectively. In fig. 5 the ellipse for @=50 is wanting since the minimum ellipse has a semi-axis major 52°36 (Table IL), while below ae the three curves are present. In figures 6 and 7 are contrasted in like manner the isergonal curves for the angles o=30, and w=the supplement of 30°. In fig. 6 the curve @=—5 is wanting, and in fig. 7 the two curves @=5, and @=20 are both wanting. In like manner are to be explained the figs. 8-18. The numbers needed for drawing the figures are furnished by equa- 196 H. A. Newton—Capture of Comets by Planets. tion (18). The curves that in each figure separate the shaded area from the non-shaded area are the ellipses for @=20, and @=-—20 The shading is introduced in order to compare more readily the corresponding curves in the figures. 25. The dotted curve in the several figures represents those values of d and / for which the total change of direction in the relative orbit is 10°; that is, a=85°. It is that curve whose equation is A tan 85°=B, or d’-+ A’ sin® 0=A’* tan’ 85°. It is therefore an ellipse whose center is the origin of codrdi- nates, and it is similar in each figure to the isergonal ellipses. Fig. 8; o=40°. Hie, 9s o=140} H ttt CO ee Ss ee (a 26. Hypotheses about the parabolic cometary orbits.—It will be convenient to make two assumptions about the distribution of the parabolic comets, and the distribution of the goals of their motions. There seems to be no very well marked rela- tion between the ecliptic, or to speak more strictly the invari- able plane of the solar system, and the known parabolie comet- ary orbits. The following two assumptions do not seem likely therefore to introduce any very serious error into our reason- ings. HT A. Newton— Capture of Comets by Planets. 197 If about the sun as a center a sphere % be described with an arbitrary radius 7, it will be asswmed that near the surface of &, space is filled equably with comets. We may express this by supposing that in each cubic unit of space near 9, there are at each and every instant n comets. As the orbits are all assumed to be parabolic, the n comets have a common velocity v. Fig. 12; o=120°, Fig. 10; o=50°. Fig. 12; wo=60. Wigs 1h o=130. fee) Cee J Se Ee ode as Be Ca oS a 8 ea i It will be furthermore assumed that the directions of the comets in each cubic unit of space near % are at random, that is, that the quits and goals of the comet’s motions relative to the sun are distributed equably over the surface of the celes- tial sphere. 27. Number of comets entering G.—If about a normal to & as an axis there be described two cones cutting the celestial sphere in two small circles distant from the point where the normal meets the celestial sphere ~ and w+dvw, then of the n comets there will be $n sin Wd comets whose quits are between the two circles. Each of these comets will move per- pendicularly to the spherical surface § with the velocity 198 HH. A. Newton—Capture of Comets by Planets. v cos w. Hence in a unit of times $nv cos W sin dw comets will cross a unit of the surface G going towards the sun. The total entering the sphere in the unit of time will be this number multiplied by the number of units in the surface of G, or T 2 ae nv cos wp sin ~ dip = mnvr"’. 0 : = 0) Vie. 15: o=H10- 28. Distribution of parabolic comets as to perihelion dis- tance.—This supposition of equable distribution of the goals of comets as they cross the spherical surface § involves also a law of distribution of comets as to perihelion distanee. The number of comets that enter the sphere in a given time whose motions make with the normal angles between y and w+ dw is proportional to sin cos dy. If N be the number of comets that enter § in a given period of time with an angle with the normal less than w, we may write dN=£ sin > cos way, where & is some constant. But if g is the perihelion distance of a comet which at the distance 7 from the sun moves at an angle with the radius equal to y, then g=7 sin* w, and dg=2r sin ~ cos Wdy. But comets that enter § with angles to the H. A. Newton—Capture of Comets by Planets. 199 normal between Ww and wW+dW, have perihelion distances between g and g+dqg. Hence N may also represent the num- ber of comets that in the given period of time pass their perihe- lia, and whose perihelion distances are less than g. Therefore NV — is a constant, and we conclude that if comets be grouped f according to their perihelion distances the number of comets whose perihelion distances are less than g is proportional to q. Fig. 16; w=80°. Fig. 17; w=90°. Fig. 18; w=100. 29. It follows as a corollary to art. 28 that if the two assumptions of art. 26 be made for. the spherical surface G, the like distributions are true for every smaller concentric spherical surface. It would be but a reasonable extension of the assumptions to make them apply to larger spheres, if finite. [To be continued. ] AM. Jour. ScI.—THIRD SERIES, Vou. XLII, No. 249.—SzepPremBer, 1891. 14 200 FE. Levereti— Pleistocene Fluvial Planes of Art. X VIII.— Pleistocene Fluvial Planes of Western Penn- sylvania ; by FRANK LEVERETT. In the November number of this Journal there appeared a paper by Mr. P. Max Foshay, entitled “ Preglacial Drainage and Recent Geological History of Western Pennsylvania,” in which certain views are expressed which do not seem consistent with facts Im my possession; and in which, although valuable suggestions are made regar ding the possibilities of changes in drainage which the region may have undergone since the be- ginning of the glacial period, adequate data are not presented to sustain the broad and positive conclusions drawn. It there- fore seems important that the subject be further considered and that attention be called to facts which render the problem more complex and its solution less certain than the paper leaves the readers to suppose. A few general statements respecting the fasted planes of the region (following essentially President Chamberlin*) will aid in showing the bearing of the facts I wish to present. In the district immediately southeast of the drift-covered portion of western Pennsylvania there are three well-developed fluvial planes distinct from the present flood-planes of the streams, representing as many distinct episodes in recent geological history. The lowest fluvial plane is the rock floor of the val- leys, which in much of the region is at a lower level than the beds of the present streams. The middle fluvial plane is rep- resented by the moraine-headed terraces which take their rise in the bulky outer moraine of the “Grand River lobe.” These are at a somewhat higher level than the present flood-planes, for the altitude of the region now is such that the streams are deepening their channels. The highest fluvial plane is repre- sented by elevated terraces 250 feet above the present streams. This is much broader than the middle and lowest planes. Its remnant is a rocky shelf capped with distinetly fluvial material varying in depth up to 40 feet or more. Abandoned loops or “oxbows” occasionally occur, giving a complete cross-profile. The time sequence of these planes is as follows: the high terraces are the oldest, the moraine-headed terraces are the youngest, while the rock floors of the buried channels are of intermediate age, for they represent the limit of an erosion and deepening that took place after the high terraces were formed and before the moraine-headed . terraces were built up. High-level terraces.—President Chamberlin has set forth in the bulletin referred to the fact that the high terraces were * Bulletin No. 58, U. S. Geological Survey, pp. 24-37. Western Pennsylvania. 201 fluvial planes as late as the early glacial period. His observa- . tions together with the earlier ones of Dr. H. M. Chance of the Pennsylvania Geological Survey, are to the effect that Explanation of Map.—The shaded portions represent moraines. Their map- ping is complete only between the Cuyahoga river and Lake Chautauqua. Striae are represented by arrows and indicate the general divergeuce from the axes of the lobes. The numbers indicate villages and cities as follows: (1) Lottsville, Penn.; (2) Titusville, Penn.: (3) Meadville, Penn.; (4) Erie, Penn.; (5) Ashta- bula, O.; (6) Painesville, O.; (7) Akron, O ; (8) Canton. O.; (9) Braceville, O.; (10) Leavittsburg. O.; (11) Niles, O.; (12) Youngstown, O.; (13) Lowellville, O. ; (14) Edenburg, Penn.; (15) Newcastle, Pa.; (16) Greenville, Pa., (17) Raymilton, Penn.; (18) Oil City, Penn.; (19) Beaver, Penn. fluvial material containing crystalline erratics of Canadian derivation occurs along the Allegheny river on this terrace. This determination is of great importance since it brings all 202 Lf Leverett— Pleistocene Fluvial Planes of . the erosion of the lower 300 feet of the Allegheny valley within the earlier glacial and the interglacial epoch, and throws doubt upon the preglacial age of the buried channels, although they are plainly older than the moraine-headed ter- races of this region. The fact that the Allegheny and Monongahela river-beds had become coated to considerable depth with fluvial débris (40 ft.) by the close of the earlier glacial period is evidence that the streams had reached a base level at a still earlier date, and makes it probable, though not demonstrative, that the ex- cavation to the level of the upper rock plane of the several lines of drainage which subsequently united to form the Allegheny was preglacial. If not preglacial, it must have been accomplished during the earlier part of the first glacial epoch. 3 Me Mr. Foshay’s paper the high level terrace’ along the Bea- ver river is discussed as “‘an old base-level plane,” and Pro- fessor I. C. White, in his report on Beaver County, Pennsy]- vania, calls it the ‘fourth terrace.”* It has been identified by these writers as far north as the mouth of the Connoquenessing where the terminal moraine of the later drift hes upon it. Mr. Foshay ealls attention to the important fact that this terrace has a northward descent from the mouth of Beaver river to the terminal moraine, and President Chamberlin has shown that the high terrace of the lower Allegheny and the Monongahela descends with the present streams to their junction not far above the mouth of the Beaver river; all of which evidence favors the hypothesis that the lower factors of the Allegheny river and the Monon- gahela discharged toward the Lake Erie basin along the course of the Beaver river before the first glacial epoch. This de- cline of the high level terrace from the mouth of the Beaver north to the point where it is lost under the moraine seems in itself to be good evidence that the old river took this north- ward course instead of that now followed by the Ohio, and this determination by Mr. Foshay is a valuable addition to our knowledge. It is somewhat short of a conclusive demonstra- tion of the northward course of the stream in the fact that the decline is only 25 feet, that the distance is only ten miles, that the observations are few, (apparently only two), that the two remnants may not belong to strictly identical planes, that the decline is not greater than the possible differential northward depression of the region, and that the non-continuance of the high-level plane down the present course of the Ohio has not been demonstrated. If it shall be shown that no such high ter- races follow down the Ohio, the presumption in favor of the * 2nd Geol. Survey of Penn. Q, pp. 11, 12. Western Pennsylvania, 2038 Beaver river route will be strong. If high-level terraces occur, as they doubtless do, on the Ohio between the mouth of the Beaver and Wellsburg, West Virginia, the supposed old divide, and these terraces decline toward the mouth of the Beaver, i. e. contrary to the present stream, then the demonstration that the old course was to the north through the Beaver valley will be essentially complete. | It is to be hoped that Mr. Foshay, who is practically on the ground, will pursue to a demonstration the hypothesis he has already rendered so highly probable. As to the course which this old river pursued north of the moraine on the Beaver (assuming that it took this course)— whether along the present route of the Mahoning or that of the Shenango—there seems to be no demonstrative evidence. None of Mr. Foshay’s data bear definitely on this point. The Shenango valley all the way from its mouth, 7-8 miles north of where the high-terrace is lost under the moraine, to the Pymatuning swamp on the Erie divide where it connects with the valley of Ashtabula creek flowing into Lake Erie, is broader than that of the Mahoning from its mouth to the Erie divide near Warren, Ohio; its bluffs are less abrupt and its general aspect that of a valley older as well as larger than that of the Mahoning. Moreover the lower Mahoning valley be- comes very narrow in the‘vicinity of Lowellville, Ohio, having abrupt bluffs with a breadth at base of but about one-fourth mile, which is too narrow to make it probable that it is a con- tinuation of the old river under consideration, whose breadth above is much greater and whose slopes are more worn and receding. The narrowing at this point fits well the hypothesis that here was the preglacial divide between a stream running _ northwest into the Grand river basin and one running south- east to join the old river under discussion at the present mouth of the Mahoning. Furthermore, the main preglacial valley of the Grand river basin seems to have entered, not from the southeast along the lower Mahoning but from the south along the upper Mahoning from the direction of Alliance, Ohio, there being a comparatively low belt several miles wide along the upper (north flowing) part of the Mahoning, with low bluffs and a gradual rise both to the east and the west of the river. The relative elevations of the present divides on the Mahoning and Shenango routes respectively, do not help us much in this question, since, in the first place, we cannot trace, or at least have not traced, the terraces which mark the old river bed,—the present surface divide and the present rock divide being matters of more recent formation—and, in the second place, an eastward differential uplift is known to have taken place. The uplift referred to is well shown by the 204 Lf, Leverett—Pleistocene Fluvial Planes of highest of the beaches in the eastern part of the expanded Lake Erie. This beach is fully 80 feet higher immediately north of Chautauqua lake than it is at the Grand river basin. Since this amount of differential uplift has occurred during the short time since the lake occupied this beach, it becomes necessary to allow for even greater changes either of depres- sion or of uplift in the much longer period that has elapsed since the high-level terraces along the streams of western Pennsylvania were formed. On the whole, therefore, the balance of evidence favors the Shenango as against the Mahoning route, but the question is still an open one. The definite conclusions of Mr. Foshay supported by map and proposed name do not seem to be war- ranted by the present state of evidence, or even to represent the probabilities of the case. Interglacial valleys (Buried channels).—As the high-level base-plane has been demonstrably connected with the earlier glacial epoch by Dr. Chance and President Chamberlin, the channels cut in it are obviously of later age; and it is impor- tant that the existing broad distinction between the interglacial and preglacial channels of this region be kept in mind; the preglacial channels have, so far as yet identified, a fluvial plane far above that of the present streams, while the interglacial channels have a rock floor in large part below the present streams. The study of the profiles of the valley floors within this drift-covered region, when combined with an attempt to restore former systems of drainage, is calculated to impress one with the fragmentary nature of available evidence. It is true that portions of Big Sandy, Oil and French creeks and of the Allegheny and Conewango rivers are sutticiently well explored by the numerous oil-well borings to givea satisfactory knowledge of the slope of the valley floors, but outside of the oil district the valleys have been explored only so far as is necessary to obtain water or to prove that oil and gas are not to be found. Throughout much of northwestern Pennsy]l- vania and northeastern Ohio the depth of drift in valleys is known only at intervals of several miles, and very seldom has a series of borings been made that test the entire breadth of a valley. | Mr. Foshay calls attention to the very low altitude of the rock floor near the junction of the Mahoning and Shenango rivers, where it is said to be 50-75 feet below the level of the floor of the Ohio near Pittsburg, and perhaps lower than at the mouth of the Beaver, and bases his ‘“ Spencer River” channel largely upon this deep portion of the valley, no bor- ings having yet been made farther up-stream along either the Western Pennsylvania. 205 Mahoning or the Shenango that reveal a rock surface so low as that near the junction of these streams, as may be seen from the following table which represents the deepest borings of which I have knowledge. Table showing principal borings along the Mahoning-Grand River route. : | | + | | Location. he ee. | Altitude. Drift. |Rock floor. Authority. | | = Lawrence Junct., Pa. ._----- 0 miles. 760 feet. 150 feet. 610 feet. White. Hdeneure Pa. _.__..---..--- eG ESOe SP UAGO 2 8 THB Og ee Stare Line, O. and Pa, ------ rts) ISRO ts 180 Se O30.) Newberry: Mawellyalie sO...» -- NOs ee OAOe Piet ay am ere. Haselton, near Youngstown__15°6 “ (831 “ 90 ‘ (741 ‘ |Foshay * ies, 0. Jie 26 «(1854 1900 gy4 Ee Near Southington, O. _____-- 40 “ (870 “ 222+ 648—“ Leverett. Mesopotamia, 0. .._---.--- ais ee ee SOO mt TI 2OS Mee | GAD Her |e: at! Ger iC ne ice IS 20h Fo Oa Gb0— | OS * Given in letter to the writer. The thickness of drift at Edenburg is a disputed question, some citizens maintaining that the greatest amount penetrated was 140 feet where the level of the well-mouth was 12-20 feet above the river, while others hold the opinion that the drift extends about 200 feet below the river. The borings were made nearly thirty years ago, and no records are known to have been preserved, consequently much allowance should be made for inaccuracies. It may be necessary to add 75 feet to denote the true altitude of the rock floor at this point. However, as there is a possibility that the rock floor is as low as indicated I leave it as given by Prof. White and Mr. Foshay. The borings at Niles, Ohio (No. 11, on map), are cited by Mr. Foshay as fixing the position of the old channel at that point, if northward differential uplift be taken into account. The amount of differential uplift required if the rock floor is but 580 feet A. T. at Edenburg, is 72 feet per mile, allowing the stream no fall between Edenburg (No. 14), and Niles; and 93 feet allowing the stream a fall of one foot per mile, Niles being 12 miles farther north than Edenburg and about 22 miles distant by the stream, while the valley floor there is 94 feet higher than at Edenburg. If we are allowed to assume an uplift of 93 feet per mile, or’even of 74 feet, nearly every large stream tributary to the Ohio from the State of Ohio, as well as the lower Allegheny and the Monongahela, could be carried into the Lake Erie basin, and we could if we saw fit attach to the Lake Erie basin all the southern tributaries of the Ohio from West Virginia and eastern Kentucky. 206 Lf, Leverett— Pleistocene Fluvial Planes of In the above calculation a continuous deep channel from Edenburg to Niles is assumed but there is evidence against the validity of this assumption. At Lowellville, Ohio (No. 13), the Mahoning, as has been noted by Dr. Newberry,* has a rocky _ bed, an examination of the valley at this village for the pur- pose of finding, if possible, evidence of a deep channel leading through it convinced me that not only is there no evidence of its existence but on the contrary the rock is exposed at fre- quent intervals throughout the whole width of the valley, the village as well as the river bed being upon rock. The out- crops are so frequent that there appears to be no room for a gorge so much as 100 yards in width, much less for one suffi- cient to be the outlet of such a stream as must have been dis- charged by the Monongahela and lower Allegheny rivers. It therefore appears that the hypothesis of a discharge northward along the Mahoning route involves a hypothetical uplift of an improbable amount, wholly unsustained by evidence, and further, that the constriction of the valley at Lowellville makes the route an inherently improbable one. I have also examined the Shenango valley for the purpose of discovering a northward outlet for the deep channel at Edenburg. The rock floor of this valley is struck at several points about 125 feet below the present stream and seems to slope with the present stream, southward, instead of toward the Erie basin. At Greenville several borings have been made which test quite well the valley throughout its entire width, and no channel of greater depth exists unless it be a narrow gorge inadequate for the passage of a large stream. The level of the lowest part of the valley floor through much of the busi- ness portion of the city is about 815 feet A. T. This is 155 feet above the valley floor at Newcastle and at least 160 feet and possibly 235 feet above the rock floor at Edenburg. A calculation of the amount of northward differential uplift that must be assumed and subtracted from the altitude of the valley floor at Greenville to bring it to the lower of the levels at Edenburg gives 844 feet per mile, and if enough uplift be assumed to give the ancient stream a northward descent of one foot per mile this amount will be increased to about 10 feet per mile,—a greater uplift than it is legitimate to assume. Additional evidence against the northward discharge of the waters from the great drainage basin of the Monongahela and lower Allegheny some 13,000 square miles in area, may be found in the narrow gorge of the Beaver above Beaver Falls; but the character of the evidence from the Ohio valley itself appears to render unnecessary further consideration of the probabilities of northward drainage. This valley receives * Geology of Ohio, vol. iii, p. 804. Western Pennsylwania. 207 glacial terraces from its northern tributaries above Wellsburg, West Virginia, and all these terraces continue down the Ohio and have a fall as great as the present stream, showing that an open valley existed previous to the later glacial period and that its stream has since this glacial period been reéxcavating a channel partialiy filled by glacial gravels. Furthermore, gas well borings at Wellsburg, West Virginia, where Mr. Foshay has placed the old watershed, show the rock floor there to be but 590 feet A. T. or 10 feet lower than it is known to be at any point along the Ohio in Pennsylvania. A carefully pre- pared report of a well has been sent me by Millard E. Boyd, Esq., city engineer of Wellsburg, in which the character and thickness of drift are given and the altitude of the rock floor is referred to low water in the Ohio, from which it appears that the drift below the level of low water mark is gravel, showing vigorous drainage, and the rock floor is 40°58 feet below low water. Mr. Boyd states that within a radius of two miles about thirty wells have been made and that those on the same bottom with the one reported all show the rock floor to have about the same altitude (590 feet A. T.) This evidence from the Ohio valley seems conclusive that the Monongahela and Allegheny rivers had their present course down the Ohio in the interglacial period and have held it con- tinuously from that time to the present. In view of the results arrived at by the study of this portion of the Upper Ohio district, we are naturally led to examine the nature of the evidence put forth by Mr. J. F. Carll some years ago,* as a demonstration that the buried channels of the upper Allegheny, Conewango, Oil, and French creeks have an outlet into the Lake Erie basin. In discussing this evidence the buried channels (interglacial fluvial planes) only are con- sidered and no account is taken of the high terraces (pregla- cial fluvial planes) since these high planes have not been sufficiently investigated north of the glacial boundary to enable one to form an opinion concerning them. The numerous oil well borings show that the valley floors of several northern tributaries of the Allegheny have higher altitudes near the mouths of these tributaries than they have a few miles upstream. For example, in the Conewango valley the rock floor is 129 feet lower at Fentonville, near the State line of New York and Pennsylvania, than it is at the mouth of the stream, 13 miles south, and the rock floor of Little Broken- straw valley is 148 feet lower at Lottsville, Penn. (No. 1), than where its waters join the Allegheny 15 miles below. In other tributaries of the Allegheny the descent of the rock floor begins a few miles above the mouth; thus in Oil creek the * Penn. 2d Geol. Survey III, 1880, pp. 330-366. 208 F. Leverett—Pleistocene Fluvial Planes of valley floor has its highest point near Titusville, Penn. (No. 2), and there is a descent of 66 feet in eight miles upstream. The valley floor of French creek rises for five or six miles upstream, but near Meadville (No. 8), 25 miles above its mouth, the rock floor is about 150 feet lower than at the mouth of the © creek. The floor of Big Sandy creek rises from its mouth to. the vicinity of Raymilton (No. 17), but descends above that village, beg fully 40 feet lower at Sandy Lake than at Ray- milton. Borings are sufficiently numerous to show a strong probability that these valley floors have no channels deep enough to drain them southward, but unfortunately they are not sufficient to demonstrate whether or not there is a continu- ous descent to the Lake Erie basin from any of the points noted. Mr. Carll has shown that there appears to be no obstacle to the northward continuation of the Conewango val- ley past Cassadaga lake into Lake Erie, though it is necessary to assume about 500 feet of drift filing at the watershed. Similarly, to give French creek a northern outlet by way of Conneaut lake and Conneaut creek a drift fillmg of over 300 feet at the watershed north of Conneaut lake must be assumed. Inasmuch as the northward drainage of these buried chan- nels remains an open question, a brief consideration of other hypotheses to account for the phenomena seems called for. A certain amount of northward descent may prove to be due to erust-deformation. The beaches about Lakes Erie and Ontario, and those of the Glacial Lake Agassiz, as is well known, indicate clearly a northward differential uplift aceom- panying the retreat of the ice, bunt they indicate nothing as to the depression that preceded this uplift. In the opinion of those who have given most attention to these beaches, the uplift was due mainly to the withdrawal of the load of ice. This hypothesis involves a previous depression occasioned by its accumulation and an imperfect restoration, owing to the removal of material from portions of the glaciated district and the presence of a load of drift and large bodies of water after the ice withdrew in parts of the glaciated district not thus encum- bered in preglacial times. So far as we may reason from theo- retical grounds, there should be expected a residuum of north- ward depression in the region under discussion, since a large amount of drift was deposited here. And this may prove to have been an important factor in giving these valley floors a northward slope, though it is hardly probable that it was the chief one. A more important factor in the production of the peculiar valley phenomena of this region may prove to have been erosion effected beneath the ice either by the ice itself or by subglacial waters. Whether the ice greatly deepened valleys Western Pennsylvania. 209 through which it flowed is an open question, but that sub- glacial waters exerted a peculiar eroding power in certain places near the ice margin is conclusively shown in various parts of the glaciated district by the presence of large channels made by them. Some of these are remote from present streams and have been little affected by post-glacial erosion. Their trend is in line with the striation and approximately at right angles with the moraine. They are often occupied by osars and hence are called “osar troughs.” These troughs or channels sometimes rise toward the moraine at the rate of several feet per mile, and yet the material in the osars lying in them shows conclusively that the flow was in that direction. The, water seems to have been forced upward toward the ice margin by the weight of the ice sheet and by hydrostatic pressure. These osar troughs were formed just before the ice made its final retreat, but the eroding power, of which they are the product, was probably in operation in earlier stages of the ice invasion. The outer moraine in the district under discussion is a complex one, the equivalent of several moraines farther west that indi- cate a succession of advances and retreats of the ice front. In the early stages the rock floors of these valleys may have been deepened in places by the subglacial streams in the same man- ner as the osar troughs were produced. By reference to the accompanying map it will be seen that every valley in which avery low rock floor has been reported has a trend approxi- mately at right angles with the moraine and in line with the ice movement; that is, such a trend as to invite the flow of ice and of subglacial waters. Furthermore it appears that in every case the lowest known point of the rock channel in these several valleys is near. the inner border of the moraine. In case it is found that no northward outlets exist the most plausi- ble explanation for the low altitude at these points would seem to be a deepening of the channels here below their main out- lets by subglacial waters assisted, perhaps, by the ice itself. Summing up all the available evidence, it appears that no northward outlets have been found for the low channels just. within the moraine on these several streams which are not embarrassed, either by a rise in the rock floor or an extraordi- nary amount of drift. In the streams under special considera- tion, the Shenango, Mahoning and Beaver, it appears that the rock floor rises in all directions from Edenburg, unless there be a descent down the Beaver. The obstacles to a northward discharge of these streams seem, on the whole, greater than those in the way of a southward discharge. In the Mononga- hela, lower Allegheny and the Ohio valleys, the available evidence all indicates southward discharge along the present course of the Ohio from the interglacial period to the present time. 210 FI’, Leverett— Plestocene Fluvial Planes of Taking into consideration all the known facts, it certainly seems premature to urge, without distinct qualification, the acceptance of a hypothesis of northward drainage for any of these streams during the interglacial epoch, and much more so to impose a name for the unproven river. Moraine-headed terraces.—President Chamberlin’s descrip- tion of the moraine-headed terraces and general remarks upon the history of the several fluvial planes embody so well the essential facts that further remarks are unnecessary. From his paper the following extracts are taken verbatim.* “The third group of terraces are sharply distinguishable from those which have just been considered; first, in the fact that, instead of being rock platforms covered by fluvial mate- rial, they are made up bodily of coarse alluvium, mainly gravel. They have their chief development in the rivers entering the Ohio from the north, and when traced up they are found to head on one of the moraines of the later glacial epoch, or at least of a later glacial epoch following at a considerable inter- val an earlier one. The uppermost of these terraces has for its surface plane the ancient flood deposits of the glacier-fed streams. The lower terraces have been cut out of it by subsequent erosion. Near the moraine this upper glacial flood surface. may be traced continuously, rising somewhat rapidly as the moraine is approached, and passing gradually into a series of undulations which merge into the gravelly knobs and _ basins, and thence into the unassorted hills of the moraine. This relationship was satisfactorily observed by Mr. Gilbert and myself, separately or jointly, on Conewango creek, near Rus- sellburg; on the Little Brokenstraw, near Freehold; on the Big Brokenstraw, near Horn’s Siding; and on Oil creek near Hydetown. On Sugar creek, French creek, and Sandy creek phenomena of similar significance appear, but they are less clear in their import. On Beaver river and Little Beaver creek analogous features are more satisfactorily displayed. The streams of gravel starting in these morainic heads run down through the rock channels cut below the old river bottom as above described. The surfaces of these later glacial gravel streams are generally much below that of the earlier terrace deposits, but as they slope more rapidly there is no constant difference. An interval of from 100 to 200 feet may be taken as representative. The bottom of these later glacial gravels extends below the present river-beds, reaching depths varying from 40 feet to 250 feet or more, showing a considerable depth of channel before this late filling. These terraces reach their greatest height above the present stream, so far as observed, at the junction of the Beaver river with the Ohio. There the * Bulletin No. 58, U. S. Geol. Sur., 1890, pp. 32-36. Western Pennsylvania. : 211 terrace rises 127 feet above the Ohio, according to a lock-level measurement by Mr. Gilbert. Similar moraine-headed terraces occur in Ohio on the Mus- kingum, Scioto, and Mad rivers and their tributaries, and seem to have their equivalent in terraces on the lower stretches of these rivers and on the Ohio. In other words, there is a general system of deep valley gravels, starting from the mo- raines indicated and sweeping down the valleys, growing pro- gressively finer in material. Out of these glacial flood deposits a system of terraces has been cut by subsequent erosion. The still later glacial episodes seem to have introduced modifying elements, but these are unimportant in this connection. The time and manner of origin of the moraine-headed terrace planes are placed beyond question by their morainic connec- tions. They are clearly the products of the streams that issued from the glacier during the moraine-forming epoch. The carving of the terraces out of these planes was chiefly a subsequent work, of relatively minor importance in the present discussion. The coarseness of the gravels of this series indi- cates vigorous drainage, which in turn implies an open valley and at least a fair gradient below. It is equally evident that terraces of a much higher level and different gradient could not have been formed at the same time. Minor side-valley terraces night have been formed at flood stages, but only to the height of the maximum floods, and these must have had the same slope as the broad flood planes. It is clear that the upper gravel-bearing terraces were not formed at the same stage as these moraine-appended ones, for they are not only of a different type, being alluvinm-covered rock platforms, but they stand high above most of the morainic heads of the later deposits and show much greater antiquity in the erosion of their surfaces. For example, at Warren the old gravels have an altitude of 1415 feet above sea-level, with a terrace at 1395 feet, while the moraine-headed flood deposits of the later epoch at Russellburg, eight miles upstream, occur at about 1275 feet. On the Beaver river the moraine-headed gravel stream has an elevation.of about 830 feet, while along the valley below pebbles referred to the earlier epoch range from 900 to 950 feet, and ten miles below there is a wide rock- based terrace at about 885 feet. But these higher gravels con- tain pebbles of granite and other crystalline rock, whose pres- ence is only to be accounted for through glacial agencies, and the explanation of their origin must embrace that element.” “The higher glacial gravels antedated those of the moraine- forming epoch by the measure of the erosion of the channel through the old drift and the rock, whose mean depth here is about 300 feet, of which, perhaps, 250 may be said to be rock. = — ee ee 212 FE. Leverett— Pleistocene Fluvial Planes, ete. The excavation that intervened between the two epochs in other portions of the Allegheny, Monongahela, and Upper Ohio valleys is closely comparable with this. In view of these facts it seems scarcely less than proven that it was the earlier invasion of ice that reversed the drainage and partially filled the valleys with debris, forming the capping of glacial gravel that rests upon the upper terrace.” ~ “From the fact that the fluvial material in these abandoned channels and on the corresponding terraces in the Monongahela valley is wholly local, or southern, while among the analogous material of the Allegheny there mingle crystalline erratics of Canadian derivation, and from the evidence given above, we draw the inference that the partial filling was coincident with some stage of the earlier glaciation, presumably a late stage. This view gathers some support from the now well sustained belief that a general depression and slackening of drainage accompanied the earlier glaciation. Following this episode of valley-filling and earlier glacia- tion there was a prolonged epoch ot rapid erosion of the valley bottom, which was apparently coincident with an interglacial epoch, and was, perhaps the result of the resilience of the land after the glacial depression. During this epoch the rock gorges were cut down to the rock bottoms that now lie forty feet or more below the present river bottoms. Then came the later invasion that halted at the outer terminal moraine, whose overloaded floods, like those of the preceding glacial incursion, filled the valley bottoms with glacial alluvium ; only, in this instance, in harmony with the more vigorous character of the later glaciation, the filling reached, at some points, 300 feet. Since that time there has been another stage of reéxcavation, giving origin to the lower gravel terraces. This is doubtless far from being the whole history of events, and may be divergent from the truth in minor phases, but I believe with some confidence that it represents the general truth respecting the history of the abandoned channels and chief terrace deposits of the system of benches under consid- eration.” Madison, Wis., March 14, 1891. Gooch and Gruener— Determination of Antimony, etc. 213 Art. XIX.—A method for the Determination of Antimony and its condition of Oxidation; by F. A. GoocH and H. W. GRUENER. [Contributions from the Kent Chemical Laboratory of Yale College.—VIII.] BUNSEN’S method of determining qualitatively the condi- tion of oxidation of salts of antimony, by boiling these sub- stances in solution with potassium iodide and hydrochloric acid and noting whether the liquid takes the color of free iodine, has been applied successfully to the quantitative deter- mination of antimony in its highest condition of oxidation by Weller,* who distils the iodine from the solution, collects it in the distillate and, determining it volumetrically, calculates from the amount of it found the antimonic salt which sets it free according to the equation SbCl, + 2H-I=SbCl’ + 2HC1I+L-L The advantage of treating the residue, rather than the distil- late, in analytical processes in general which involve distillation is so obvious as to constrain us to seek conditions under which Bunsen’s reaction may be applied in such manner that the antimony shall be held and estimated directly in the residue. The general plan of work was laid down in a similar process elaborated in this laboratory for the reduction of arsenic acid.t+ According to this process the arsenic to be reduced is taken in a solution of appropriate dilution, and treated with sulphuric acid in adjusted amount and an excess of potassium iodide. The liquid thus prepared is boiled to a definite degree of con- centration, the iodine then remaining unexpelled, if any, is bleached by the very careful addition of dilute (centinormal) sulphurous acid, and the liquid is immediately diluted and neutralized. After cooling, the reduced arsenic is titrated by standard iodine in presence of starch. We found in preliminary experimentation that the same general plan of treatment is available in the handling of anti- monic compounds, but it is necessary to take precautions to prevent the deposition of the antimony from solution upon the addition of the sulphuric acid. Tartaric acid accomplishes this effect satisfactorily and does not, as the result proved, intro- duce undesirable complications. It transpired also that the dilution of the solution at which the crystalline iodide or oxy- iodide separates out during the boiling is greater than is the case when similar amounts of arsenic are dealt with. It appeared, for example, that concentration to 45cm.° was sufh- cient to cause crystallization and slight sublimation when the * Ann. d. Chem. u. Pharm., cexiii, 246. +Gooch and Browning, this Journal, xl, p. 66. 214. Gooch and Gruener—WMethod for the Determination amount of antimonious oxide present (with excess of potas- sium iodide and 10cm.* of sulphuric acid, 1:1) was approxi- mately 0°2 grm. Otherwise the process as employed in the reduction of arsenic appeared to be applicable to the similar treatment of antimony. The following quantitative experiments were undertaken to discover the condition of concentration best suited to the reduction of antimonic salts under circumstances otherwise like those adapted to the reduction of arsenic, and to test the perfection of the process. Definite amounts of tartar emetic, purified by recrystallization, were used to make the antimonic salt to be afterward reduced, the antimony being raised to the highest degree of oxidation by titration with standard iodine after the addition of sodium tartrate (to prevent the precipita- tion of the antimony during the process of oxidation) and hydrogen sodium carbonate in the usual excess, In this process starch was sometimes employed to give the end reac- tion, and sometimes reliance was placed upon the appearance of the color of free iodine, experience having indicated that the use of the starch is not essential when the solutions are sufficiently small in volume, though as a matter of course, the correction demanded for the excess of iodine necessary to give color to the body of liquid is greater when starch is not used. This treatment of the tartar emetic served the double pur- pose of providing a perfectly definite antimonic salt and re- standardizing the solution of standard iodine, which was to be used subsequently in reoxidizing the antimony after its reduc- tion, against the tartar emetic; and thus the imperfection of the process, whatever it may be, whether in the reduction or elsewhere, becomes apparent and is measured immediately by the difference between the amounts of iodine employed in the two oxidations. This mode of standardizing the iodine appears to be peculiarly advantageous in view of Fresenius’s demonstra- tion* that the iodometric estimation of antimony yields too high results, at least in the case of tartar emetic, when the standard iodine is standardized in the usual manner and, as is undoubtedly best, the characteristic starch-blue is taken for the end reaction rather than the premonitory and somewhat indefinite reddish tint. The larger amounts of tartar emetic were weighed out dry ; the smaller quantities were secured by measuring out definite portions of a solution of fixed strength. To every portion was added, in an Erlenmeyer beaker of 300 em.* capacity, one gram of tartaric acid previously treated with an excess of hydrogen sodium carbonate, and the oxidation was effected, as described, by iodine dissolved in potassium iodide to a solution * Quant. Anal. 6% Aufl., 817. of Antimony and its condition of Oxidation. 215 approximately decinormal. Four grams of tartaric acid were added, and dilute sulphuric acid, if the solution still remained alkaline, to faint acidity. Im addition 10 em.’ of a mixture of sulphuric acid and water in equal parts were intro- duced, and the liquid was boilied after introducing a platinum spiral to prevent bumping, and a trap made of a two-bulb dry- ing tube cut short and hung, large end downward, in the mouth of the flask, to prevent mechanical loss. At the chosen degree of concentration, determined by marks upon the flask, the boiling was stopped, the color bleached by the cautious addition of sulphurous acid (approximately centinormal), and the solution, nearly neutralized with sodium hydrate, made alkaline by hydrogen sodium carbonate added in an excess amounting to about 20 cm.* of the saturated solution, was titrated with the standard (decinormal) iodine after the addi- tion of a fresh portion of starch. Table I, contains the account of experiments in which the larger amounts of antimony were employed. TABLE J. Tartar Iodine used Final emetic S$b,.03 in final ‘Sk.03 Error. yolume. taken. ~ taken. oxidation. found. em.? grm. erm. orm. erm. grm. 100 075021 0:2178 023522) "| 0°2004 0:0174— 80 | 0°5030 02181 | 0°3784 | 02153 | 0:0028— 60 0°5008 | O22 0°3768 | 02144 | 0:0028— 60 0°5010 Ont. | 03780 | O;2151 0:0022 — 60 | 075010 0°2173 0°3809 0°2168 0:0005— (55 075023 0:2178 | 03827 | 0°2178 | 0:0000 | 55 05015 0°2175 | 0°3806 | 0°2166 | 0:-0009— 4 50 05007 | 0°2172 | 0O-3814 | O-2171 | e-0001— | 50 0°5039 | 072185 | 0°3839 | 0°2185 | 0:0000 | 45 0-5001 | 0°2169 | 0°3818 0°2173 | 0°0004 + | 45 | 05004 | O-2 ETO," | 0:3825 | O-2176 | 0:0006+ The results of these experiments indicate unmistakably that complete reduction may be brought about under the conditions, but that concentration to a volume of from 45 em.’ to 55 em.’ during the boiling is not only advantageous but necessary. The mean error of the determination in which the final volume fell within these limits was zero between limits of -0:0009 grm. — or 0°0006 grm.+. In both determinations in which a final volume of 45 cm.* was reached, and in one of the experiments in which the final volume was 50 em.’, the formation of the crystalline antimonious iodide or oxyiodide in the liquid was noted, and the deposition of a very slight sublimate of the same salt in the trap. It is evident, therefore, that it would be hazardous to attempt to push the concentration further. Am. JouR. So1.—THIRD SERIES, VoL. XLII, No. 249.—SEPTEMBER, 1891, 15 216 Gooch and Gruener—Method for the Determination In all these experiments hydriodic acid was present in amount equivalent to 1:1 grm. of potassium iodide—0°5 grm. intro- duced as iodine and 0°6 grm. introduced as such in the stan- dard iodine where it plays the part of solvent. In the experiments recorded in Table II, smaller amounts of antimony and correspondingly smaller quantities of the oxidiz- ing solution were employed ; otherwise, the same general mode of proceeding was fellowed. The limits of concentration fixed upon were, however, varied somewhat. The previous experi- ments showed plainly that anything like a complete reduction of the antimony could not be anticipated when the final volume was greater than 60 cm.*, and the experience with the smaller amounts of antimony treated in the second series pointed to the fact, as the work progressed, that for them the erystallization and sublimation did not occur until the concen- tration had brought about a decrease in volume to 35 cm’. The limits of final volume were placed therefore, for these experiments, at 60 cm.* and 35 cm*. Centinormal iodine was used for the oxidations and bleaching with sulphurous acid was found to be unnecessary, the amount of icdine liberated in these experiments being so small as to vanish in the concentra- tion so completely that no color was visible (nor was it brought out by starch) after washing down the trap and cooling. There did remain a trace of color before the addition of the water but this seemed to us to be due in all probability to the incip- ient formation of the antimonious iodide or oxyiodide which is decomposed by the action of more water. At all events it disappeared on the addition of water and no reoxidation of the antimony was found subsequently. TABLE II. Tartar | Sb203 | pees used Sb,03 Final Emetic | in final Error. Volume. taken. | taken. | oxidation. | found. 4 ee - pees emt) orm. | erm. | orm. erm. orm. 60 0: 0500 | 0:°0217 0:0239 00136 | 00081— 60 0°0500 | 00217 | 0°0258 00147 | 0-0070— 60 0:0500 | O02) | 0:0261 | 00148 | 0:0069— 50 | 0:0500 | 00217 | 0:0316 070180 | 0-0037— ( 40 | 0:0500 | 00217 | 0°0385 070219 | 0:0002+4 35 |" *0-0500 00217 | 0°0380 | 00216 | 0:0001— 4 35 0°0500 O'O21T a 0:0381 00217 | 0°0000 | 35 | 0:0500 | 0°0217 | 0:0382 | 0°0218 0:0001 + | 35 | 0:0500 | 00217 | 0:0382 070218 | 0°0001+ These results show that for the smaller amounts of antimony the reduction was completed only by pushing the degree of concentration somewhat lower than was found to be necessary of Antimony and its condition of Oxidation. 217 in treating the larger amounts. The only point in which these experiments differ essentially from those of the previous series is in the quantity of the iodine solution employed to effect this oxidation. So far as concerns the free iodine itself the conditions are similar in both series; for the iodine is con- verted in both cases to hydriodic acid exactly equivalent in amount to the antimony acted upon. The potassium iodide which is added in the iodine solution produces by action upon the sulphuric acid present an excess of hydriodic acid, which is, of course, dependent upon the absolute amount of the iodine solution employed. The hydriodic acid is the active agent in the reduction of the antimony, and to the greater mass-action in the former series of experiments might be attributed the more complete reduction for equal degrees of concentration. Accordingly the determinations of Table II were made to put this point to the test. In these experiments the conditions were identical with those of the determinations of Table I, excepting that in every case 1 grm. of potassium iodide was added to the liquid before boiling, thus bringing the total amount of hydriodic acid present to an equality with that present in the experiments of Table I, in which the larger amounts of antimony were treated. The results of these experiments bear out completely the hypothesis concerning the mass-action of the hydriodic acid—the smaller amounts of antimony being completely reduced in the presence of the large excess of hydriodic acid even at a final volume of 60 cm.* with a maximum error of 0:0002 grm.— TABLE III, Tartar Sb203 Todine used Sb203 Final Emetic in fina Error. Volume. taken. taken. - Oxidation. found. em.? grm. erm. grm. - grm. grm. 60 0°0500 0°0217 0°0378 0°0215 0°0002— 60 0°0500 0°0217 0°0379 0°0216 0:0001— 60 0°0500 | 0°0217 0°0379 0°0216 0-0001— It is plain therefore that we have in the phenomena de- scribed the basis of a good method for the iodometric deter- mination of the condition of oxidation of antimony; for, the amount of antimonious salt present In a mixture of antimoni- ous and antimonic salts may be determined by direct titration in alkaline solution, and the total amount of antimony present is given similarly after the treatment by boiling, as described, with potassium iodide and sulphuric acid, the amount of antimonic salt being immediately calculable from the difference between the quantities of the standard iodine used as the ‘ if 7 —E 218 Gooch and Gruener—WMethod for the Determination oxidizer before and after reduction. The best method of proceeding appears to be that in which the concentration was restricted so that the point of sublimation and crystallization was not reached and in which the presence of an excess of potassium iodide was assured. It seemed desirable, in this connection, to test the applica- bility of the method, as outlined, to the reduction and estima- tion of antimony and arsenic associated together, as so often happens in practice. The preceding experiments establish the fact that it is undesirable to attempt, in treating antimony, to force the concentration of the solution below 50 em.*, under the conditions laid down and when the amount of antimony present is equivalent to the maximum with which we have experimented, avout 0°2 grm. of antimonious oxide. In the parallel process for the determination of arsenic concentration to 40 em* was recommended in all cases (the maximum amount treated being equivalent to about 0°33 erm. of arsenious oxide), but it was not shown in the elaboration of that process that reduction would not take place at a concentration not quite so extreme. In the results recorded in Table V, which relate to experiments which duplicate the conditions found most favorable to the reduction of varying amounts of anti- mony,—the presence of the equivalent of 1-1 grm. of potassium iodide, and concentration to 50 em*——and differ from these only in the fact that arsenic was associated with antimony in every case, it appears that the reduction of arsenic may be effected simultaneously with that of the antimony. TABLE IV. i | | | | Difference | Hirsor inte 5 | | 2 | - a Ss of aq 'Tartar | Sb,03 | As.O, Iodine used Iodine ngeq| Pehween one Ss 5 |Emetic in first | infinal | iodine = oa / ES | taken. taken. | taken. oxidation. oxidation. | j, He ba Sb.03 | AseOz > | | | oxidations. cm.3) erm. | grm. |. grm. emis) | en | em.* grm. | grm. 50! 071530 0:0870) 0:0500 19°37 19°43 006+ | 0°0004+ 0°0003— 50 | 0°1503 0°0855) 0°0495 19°05 19°02 | 0°63— 0:0002— 0:0001— 50 0°1503 0°0855 0°0544 20°05 199754) 0-08 — 0:0006— 0°0004— 50' 071503 0°0855' 0°0495 19°05 19°007 4 0°05— | 0:0004—' 0:0003 + It is plain that the error in these results, whether reckoned as falling upon the antimonious oxide or upon the arsenious oxide, is quite within the limits allowable in volumetric deter- minations by means of decinormal solutions. One point, however, in the determination of the combined amounts of antimony and arsenic by the method here proposed deserves special consideration. It has been shown in the work to which reference has been made that arsenic is reducible by the pro- of Antimony and its condition of Oxidation. 219 cess outlined and determinable with accuracy by titration with iodine standardized against arsenious oxide. In this later work we show that antimony may be reduced similarly and -estimated satisfactorily by titration against iodine standardized against tartar emetic. These two methods of standardizing do not yield identical results, and so we are confronted with an inherent error in the process for estimating antimony and- arsenic at once, which cannot be overcome unless the individ- ual amount of one or other constituent may be otherwise determined. If the determination of either the arsenic or antimony is possible it is, of course, easy to calculate with the use of the appropriate standard the amount of the solution of iodine which is really engaged in the oxidation of this particu- lar constituent, and the remainder of the iodine actually em- ployed, gauged by the second standard, will give the corrected amount of the second constituent. In ease no such correction is feasible it becomes a matter of interest to note the magnitude of possible error. Our experi- ence, based upon many determinations throughout the course of the work detailed above, pointed to a difference in the value of the two standards amounting to about one-half of one per cent. If, therefore, the weight of reduced oxide amounts to the maximum which we have experimented with—about 0-2 grm.—the greatest possible error will be 0-0010 grm.+ or 0-0010 grm. —, according as the entire 0-2 grm. is antimonious oxide estimated by the arsenic standard, or arsenious oxide estimated by the tartar emetic standard. The essential features of the process which we propose for the reduction of antimony and the determination of its degree of oxidation are recapitu- lated briefly in the following statement. The salt of antimony, not exceeding the equivalent of about 0-2 grm. of antimonious oxide, is titrated, in presence of 1 grm. of sodium tartrate and the usual excess of sodium hydrogen carbonate, by means of iodine standardized against tartar emetic. The result of this titration gives the amount of antimonious salt present. To the solution are then added 4 grms. of tartaric, dilute sulphuric acid, if necessary, to neutralization, an excess of 10 em.° of half and half sulphuric acid, and enough potassium iodide so that there shall be present of hydriodic acid the equivalent of a little more than 1 grm. of the iodide. The liquid is diluted to 100 em.¥*, boiled in an Erlenmeyer beaker until the volume is decreased to 50 em.*, the precaution being taken to introduce a platinum spiral to prevent bumping and a trap, as described, to obviate mechanical loss. The color remaining after concentration, if there be any, is bleached by dilute sulphurous acid (approxi- mately centinormal). The solution is nearly neutralized with 220 Gooch and Smith—Method for the sodium hydrate, treated with an excess of sodium hydrogen carbonate amounting to 20 cm.’ of the saturated solution, cooled, and titrated in presence of starch by the standard iodine. This final titration gives, of course, the entire amount of antimony present. The difference between the indications of the two titrations is the measure of the antimony in the higher condition of oxidation. The method as outlined is accurate and rapid, and so simple as regards manipulation that a number of determinations can be carried through simultane- ously with the use of ordinary apparatus. Art. XX.—A Method for the Hstimation of Chlorates ; by F. A. Goocu and C. G. SMITH. [Contributions from the Kent Chemical Laboratory of Yale College.—VII.] It has been shown in recent work in this laboratory* that under conditions properly controlled, arsenic acid in excess is capable of expelling the iodine from hydriodic acid at the boil- ing temperature of the solution, being itself reduced corres- pondingly according to the equation H,AsO,+2H-I = H,AsO,+H,O+TL-1. On cooling the liquid remaining after such treatment, and neutralizing, the arsenious oxide produced in the reaction may be reoxidized iodometrically in the usual manner, the iodine added to accomplish this purpose being the exact measure of the iodine originally present as hydriodic acid and expelled from the acid solution during the process of boiling. If other sufficiently energetic and easily decomposable ox- idizing agents are present at the same time with the arsenic acid, it would be natural to suppose that these substances will act similarly upon the hydriodic acid, and, furthermore, that the oxidizing power of the arsenic acid will not be called into play until that of the more unstable oxidizers has been ex- hausted. Chloric acid, for example, acts with great ease upon hydriodie acid, and it would be natural to suppose that in a mixture of chloric, hydriodic and arsenie acids the mutual action of the chloric and hydriodie acids will be manifest first and will go on steadily to completion, and that when this effect is accomplished, and then only, the action of the arsenic acid in liberating iodine from the residual hydriodic acid and in registering by its own reduction the amount of iodine thus set free will appear. It should be possible, therefore, if this: theory of the reaction between these substances is correct, to found upon the method referred to for the estimation of iodine * Gooch and Browning; this Journal, xxxix, p. 188. Estimation of Chlorates. 221 a method for the estimation of chlorates—this to consist In heating the chlorate, in acid solution and under conditions otherwise appropriate, with a known amount of potassium iodide, somewhat in excess of that theoretically equivalent to the chlorate, and in presence of an excess of arsenic acid, the arsenious oxide produced in the process being determined lodometrically and serving to measure the amount of iodide left undecomposed by the chlorate. Of course, the difference between the amount of iodide left undecomposed and _ that originally introduced should be the measure of the chlorate entering into the reaction. That a better form of iodometric method than those we have had heretofore for the estimation of chlorates is desirable is obvious when it is recognized that Bunsen’s original process—consisting in heating the chlorate _ with hydrochloric acid and potassium iodide, distilling and estimating the iodine collected in the distillate—tfails (owing to the formation of the comparatively non-volatile iodine chloride in the simultaneous action of the oxidizer upon hydrochloric and hydriodic acids) to show the entire amount of iodine cor- responding to the chlorate; and that Finkener’s substitute for this process—which prescribes the heating of the chlorate, under pressure in a closed bottle and in an atmosphere of car- bon dioxide, with a mixture of hydrochloric acid and potas- sium iodide previously prepared by treatment with sulphurous acid, boiling and subsequent cooling in an atmosphere of car- bon dioxide—though excellent when properly carried out, demands careful preparation of materials and skillful handling in the execution. We have studied the applicability of the process outlined above and record our experience in the following account. A solution of potassium iodide, approximately decinormal, was standardized according to the method to which reference has been made and which may be summarized in brief, as fol- lows: Portions of this solution were measured from a burette into Erlenmeyer beakers capable of holding 300 cm.’*, 2 grins., approximately, of pure dihydrogen potassium arseniate were added in solution, 20em.* of a mixture of sulphuric acid and water in equal volumes were introduced with enough water beside to increase the entire volume to a little more than 100 em’. Mg Fe_o> Mg \ Si0-=H. \ SiO =H, \Si0 =Al each plus three molecules of water, in the ratio 14:6:9. As in the case of jefferisite and kerrite, the three molecules of loosely combined water are unlike; two being given off over sulphuric acid, and the third retained rather more tenaciously. Reducing the original analyses to 100 per cent, uniting all similar oxides to similar type, reckoning FeO as MgO, Na,O as K,O, TiO, as SiO,, etc., we get the following comparison between observation and theory: 244 FP. W. Clarke and E. A. Schneider— Constitution of Henderson Co. Protovermiculite. Found. . Cale. Found. Cale. SiO, 39°70 39°90 34°10 34°18 ALG. TA sag, 14°25 14°52 14°78 Fe,O, Load 1 Sa BS) VIZ 120 MgO 16°32 17°08 22°41 22° K,O br iy 6°17 US ps H,O, essential 4°83 4°87 5°43 5°40 hydration 5°75 4°58 15°82 15°65 100°00 100°00 100°00 100°00 These results, taken in connection with our work on jefferi- site and kerrite, and with the mica theory upon which all our formule are based, are exceedingly suggestive. Kerrite is essentially a trihydrated hydro-phlogopite. Protovermiculite is the same substance, commingled with a tri-hydrated hydro- clintonite, in the ratio 1:1 very nearly. Jefferisite is a similar mixture of hydro-biotite and hydro clintonite, also trihydrated, and in the ratio 1:1. The Henderson County mica is essen- tially a biotite, about half way transformed into a vermiculite, and is interesting as a transition product. The hydration of its several admixed molecules is naturally uncertain. At an early date we hope to imitate experimentally the process by which a mica becomes transformed into its corresponding vermiculite. But although the above-named minerals appear to be very simple in their structure and relationships, a like simplicity does not characterize all of the vermiculites. In some mem- bers of the group there seem to be a small admixture of chlo- ritic molecules, and it is even probable that many intermediate stages between mica and chlorite may exist. As bearing upon this question we have a series of vermiculitic minerals from Chester and Delaware Counties, Pennsylvania, some of which have already been studied by Cooke, Gooch, Leeds and others, while some have escaped examination hitherto. To begin with we may consider the hallite, from Nottingham, Chester County, and the vermiculites from Lenni, (not Lerni), Delaware County. The hallite, received through the kindness of Mr. W. W. Jefferis, was dark bluish green, and agreed perfectly with the published descriptions. The Lenni mineral, partly from the collection of the late Isaac Lea, and partly gathered in the field by one of us, is represented by several varieties, which in a large series of specimens, are seen to shade into each other. Three varieties were examined: one, silver white, resembling outwardly an ordinary mica; a second, bronzy brown, like jefferisite; and the third, dark green, similar to clinochlore. All four substances were examined microscopi- certain Micas, Vermiculites and Chlorites. 245 cally by Mr. Lindgren, who found in the hallite some spear- shaped, rhombic, or more rarely hexagonal inclusions of a dark brown mineral, not further identified. His optical notes will be published in an official bulletin, later. Analyses as fol- lows, with itemized water ere naions é A. B. €. D. Hallite. Lenni l. Lenni 2. Lenni 3. J. 2 aa 35°54 36°72 35°09 34°90 i... ae undet. 0°18 0°58 0°10 Al,O, .. 2 9°74 10°06 L205 10°60 Fe,O, __ ee 9°07 Sot 6°67 8°57 Cr,O, ... 2 eee BM ei 0°26 0°46 0°23 3. 2 0°28 0°12 0-11 Or22 Ll. |e 0°25 0°31 0-27 O-17 =e 0°16 0-20 0°20 0°19° MgO. Meroe | 2 OOD 29°40 27°62 28-21 So. . Se siaie sft trace rae ie EO; _ S\ e 2°64 6°40 5°70 4°99 ns) 300°... 2 126 2°68 1°98 1°60 Poered neat __._._ 10°91 8°69 9°22 9°88 99°87 100°39 99°95 99°66 Loss over H,SO, .--.-undet. 6°92 5°84 5-21 In these analyses we at once see that the combined water is mostly in excess of the crystalline water, and that the formule deduced must be correspondingly modified. The molecular ratios are as follows: pe B. C. D. oo. 2 ae "592 614 ‘591 583 0 "152 "134 163 159 Ree 2k 760 “744 698 “712 cll Si? ——e 606 483 512 "549 [. 71 as rr "215 504 ‘427 "366 In order to learn something as to the distribution of the hydroxy] indicated by these ratios, resort was had to the process of heating in dry, gaseous, hydrochloric acid, as described in our former paper. From this test, however, the brown Lenni vermiculite was omitted, as being intermediate in its character between the white and the green. Each experiment was made at the temperature 383°- 419°, A. B. D. figure mentedy 2 SSL 164 16 17 iO) removed... 24. 3°42 1:08 1°56 MeO « Ene pane te-09 6°30 6°57 Molec. ratio MgOH eyo aes Oe "202 158 164 Here it is assumed, on the grounds of our former work, that the magnesia rendered soluble by gaseous HC] is present as 246 F. W. Clarke and E. A. Schneider—Constitution of MgOH. Representing this by the symbol R’, the three ver- miculites give the following anh, formule: Hallite( 44245 4-46 soe RR ahh cca l® obo Linia(1O,) sO) are White Lenni ... 2.22. Re me he oH (SiO) aac 304 aa: ¢ Greeny iR yh cat sae es as iM H,,, (S10,,) 2, Oscss 0 Oe These reduce at once, subject to small uncertainties as to hygroscopic water, to mixtures of molecules of the hydro- elintonite and hydro- phlogopite types, with small amounts of chloritic compounds Meg(SiO,),(MgOH), and Mg,(SiO,),H,. Upon this basis the three minerals become : VAMGe x28 oe ee Al(Si0,) ,Mg.H, 3 aq. 7 molecules. Al1O MESO, H, 18 re Mg(Si0,), (MgOH), White Lenni __---- AIOM OH 3 aq. Green Lenni 22._-- ARSIO, ) Mg(SiO.). Niet, The actual ratios observed were slightly more complex, but the foregoing expressions accord well with the analyses. Here, as previously, we may reduce the analyses to typical form and 100 per cent, reckoning Fe,O, as Al,O,, ete. The comparison is as follows: mm CO eS OI SO dw WH cS Analyses reduced. A (Hallite). B (White). D (Green.) SiO: oshiE nichn alg Oe Mea 36°93 37°56 36°33 ATMO Stale as, beaten cine 16°13 14:92 16°83 IMO JE. sat Negi ee 31°58 30°41 29°44 PL Ope yee cee NE res! 11:34 8°86 10°26 Ag, tee HORNS hy BN Eee 4°02 9°25 6°84 100°00 100°00. 100:00 MoO tin MeO 22. 2u5as6- 8:09 6°30 6°57 Calculated aN B. D. SiO ce 102) as apenas BOAT 38°11 36°61 ANODE.) Meee eee 15°82 14°25 16°97 MoQ: see vane Sa 31°75 30°49 29°95 HOURS os ean eere 11:27 8°92 10°47 Aq 5 te eet en ea ee 4°69 8°23 6°00 100°00 100°00 100°00 MeO.in Mo Qiiierre. {aie 8°93 6°10 6°66 certain Micas, Vermiculites and Chlorites. 247 When we consider the nature of the vermiculites, as products of alteration, the agreement here shown is fully as close as could be expected. Many well crystallized minerals, fresh and unaltered, are less simply interpreted. On the 19th of February, 1891, Professor Tschermak read _ before the Vienna Academy a paper on the chlorite group, in which he referred certain vermiculites to that class of minerals. He also put forth some views concerning the constitution of the chlorites, which, however, we cannot discuss until they have been published in full.* One fundamental molecule, regarded by T'schermak as a constituent of most chlorites, we may adopt for present purposes, under slightly different struc- tural form from his. This is the “amesite substance, SiA], Mg.,H,0,, written by Tschermak SiA!,H,O,MgOH),. In de- fault of experimental evidence this may be transformed into OMg,Si,(Al1H,O,),, when it becomes part of a natural chloritic series parallel with the micas—thus : Normal orthosilicate --- -- ELifsiO )., Mg,(SiO,), Whence we derive, Micas. Chlorites. Al,(Si0,),R’, Mg,(Si0,),R’, Al,(Si0,) i, Mg,(Si0,),R’, Al (Si0,),R’, Mg (Si0,),R’, LON pn Meg : CN ee OC Mae sO = B’. \Si0=R’, “Me In other words, the “amesite substance” in our chlorite series is the basic equivalent of the clintonite molecule among the micas, and is applicable to the solution of certain obscure problems. Some of the vermiculites, as Tschermak suggests, are probably chlorites, and two examples have come under our notice in which this view is partly sustained. Both were originally received from Mr. Jefferis; one from the corundum mine at Newlin, Chester County, Pa., and the other from Middletown, Delaware County, in the same State. The New- lin mineral was dull green, and much resembled culsageeite both outwardly and optically. The Middletown vermiculite was bright golden yellow; strongly exfoliating before the blowpipe and fusible on the edges. It was found upon the farm of Mr. James Painter, whence Mr. Jefferis named it provisionally “ Painterite,’ a name which seems also to have been applied to a peculiar brownish, waxy, feldspathic matrix in which the broad golden laminz were imbedded. A second sample of it was later collected by one of us. According to * This paper has appeared in extenso since this was written. We cannot, how- ever, discuss it thoroughly at present. Am. Jour. Sci.—THIkD SERIES, VoL. XLII, No. 249.—SeEPTEMBER, 1891. 248 FL. W. Clarke and E. A. Schnecder—Constitution of an optical examination by Mr. Lindgren the matrix of the ‘‘painterite” is a mixture of plagioclase, probably labradorite, with serpentine. The “painterite” itself shows hexagonal markings on the surface, and contains, Mr. Lindgren says, in- clusions of ferric oxide. Optically he found it to show slight double refraction between crossed nicols, the angle of the optical axes being small but distinct. In ‘the Newlin mineral the axial angle was usually large, being at least 25°. Analysis as follows: A, Newlin. B, Painterite from Jefferis. C, Pain- terite collected by Schneider. D, matrix of painterite. A. B. E. D. SiO, 31°28 34°86 33°95 52°47 TiO, gc ye trace trace ‘none Al,O, 17°52 11°64 12°52 21°72 Cr On" 0-14 i ae oo Fe,0, 4°70 3°78 4°40 1:23 FeO 1°20 0°20 0°20 OF MnO 0°20 ree _ ee Babs - NiO 0°33 0°14 “os ia eee oa oe MgO 31°36 31°32 30°56 9°26 CaO oes 0:07 none 3°25 K,O Bee Se a ae 0°63 Na,O re mgr aoe 5:09 EHO, 105" 1°08 1°64 1°56 1°14 «¢ 250°-300° 0°40 1°03 0°59 cee «< ignition 12-15 Leo, 16°46 4°74 100°31 100°43 100°47 99°70 Upon treating the three vermiculites with gaseous hydro- chloric acid at 383°-412°, the following results were obtained : A. e C. Hours heated 8 124 19 R,O, removed 1°09 “80 ‘78 MgO ss 5°86 8°26 9°56 Molec. ratio MeOH "146 207 239 The molecular ratios are: rey B. C. SiO, 520 ‘581 566 R,O, 202 _ 188 150 RO ‘808 ‘789 ‘770 H,O ‘675 ‘875 ‘914 Aq ‘080 "148 119 In these examples the water (Aq) expelled below 300° is so small in amount that it may be left out of consideration. Part of it undoubtedly represents hydrated molecules, which, how- ever, are relatively so few in number that they may be for present purposes disregarded, certain Micas, Vermiculites and Chlorites. QA9 From the remaining ratios, writing MgOH as R’, the sub- joined empirical formule are directly derived : Newlin Bee ats nee R's 1am (Si0,).,., O 2g Painterite “¢ 1e Fuh. RY oe0 Roo: 13 a (ered lee O ” C. ob det) jaa H (810,) Reduced to structural form these give less satisfactory results than the previously considered vermiculites. The Newlin mineral may be regarded as nearly a hydroclintonite, AlO,MgSiO,H, with an admixture of an amesite-like com- pound .Mg,OSi0O,(MgOH), in the ratio 4:1. In reality the mixture is more complicated, and must contain other mole- cules. The ‘ painterite”’ C, is wholly chloritic, containing the amesite molecule Mg,OSi0,(AlH,O,),, with the molecules Mg(SiO,),(MgOH), and Mg(SiO,),H,, in the ratio 16:4:18. These compare with the actual analyses, reduced to typical form and 100 per cent, thus: 908° 709° 1589 566 763° Newlin. ‘‘Painterite.”’ — ee a “~~ OF Found. Cale. Found. Calc. Si0, 32°42 alah 35°03 oo og Al,O, 21°39 D1AT 16°22 16°13 MgO eT 33°69 3177.2 \I80"84 H,O 12-62 j1S-27 16°98 17°44 100°00 100°00 100°00 100°00 MgO in MgOH .. 6:09 8°42 8°56 9°49 The ‘“ Painterite” B reduces less easily, but satisfies all the required conditions. It is like C, but contains other chloritic molecules in somewhat complex ratios. It must be remembered that all these minerals are mzxtwres, and the fact that they are reducible at all to simple expressions is a strong point in favor of the theory adopted for the chlorites and micas in general. A very interesting example of the way in which the chloritic vermiculites approach the serpentines in composition and char- acter has been furnished us by Mr. G. P. Merrill of the U.S. National Museum. It was found by him at Old Wolf Quarry, Chestnut Hill, Easton, Pa., and is described by him as follows: “Tt occurs in the form of bright yellowish green inelastic scales of all sizes up to an inch in diameter, associated with a compact tremolite rock which is here quarried and pulverized for use as a filler in paper manufacture. The character of the rock is greatly varied, but at the quarry opening the prevailing material is tremolite more or less altered into serpentine, the vermiculite, and other secondary products, including calcite in both fibrous and granular forms. The vermiculite, although occurring in plates of considerable thickness readily separable into thin folie, never, so far as observed, shows good crystal outlines. Optically it is biaxial 250 F. W. Clarke and E. A. Schneider—Constitution of and negative, though the axial angle is small, basal plates in the thin section showing a black cross which searcely opens at all during the revolution of the stage. Cleavage plates a milli- meter or more in thickness show plainly the biaxial character, though the figure is somewhat distorted. Dispersion p < ». The surface of the plates is at times plainly marked by sharp lines crossing at angles of 60° and 120° and along which the mineral frequently separates readily. Before the blowpipe the mineral exfoliates and fuses readily on the edges to a thin lass.” / ‘ According to Mr. Merrill this mineral is sometimes seen in cabinets labelled ‘‘ tale ;’ and indeed in its appearance it resem- bles both tale and serpentine. Upon analysis the following results were obtained, the percentage of K,O representing two identical determinations. Analysis. Molec. ratios. SIU 2 a. Caper ol 8 ee aed 728 Peles iat Moret S09 es 3°59 "035 Be, O.. saeetes 23. ee a090 006 MeO 2 ange te 38°58 ‘964 iO abi ee Gos) 023 Na OW tere. lem 0°13 002 BO; 10sp gue 38 0:46 mea! CE D5 Or SOO), Denies 0°09 errs ‘Sp dommpiony See 10°70 "594 100°38 Treated with dry, gaseous HCl at 383°—412° for 164 hours, 4°36 per cent of magnesia became soluble, corresponding in molecular ratio to 109 mol. MgOH. Hence the mineral, although resembling serpentine in general composition, differs from the latter in its proportion of this molecular group. Upon treatment with aqueous HCl of sp. gr. 1°12, a small portion remained undecomposed. Ten grams of the mineral were therefore digested with the acid for three days on the water-bath, and the residue was afterwards boiled out with a solution of sodium carbonate to remove liberated silica. The remaining residue, amounting to 3°10 per cent of the original material, was then analyzed separately, and found to contain : SLO ae ee ek es 64°53 ALCL HG: on tees aed 2°03 Mion. es eee rae 33°04 99°60 All the potash went into solution; whence it seems probable that no muscovite was present. The ratios of the insoluble residue agree very closely with those of tale, and we may there- certain Micas, Vermiculites and Chlorites. 251 fore assume that mineral to be present as an impurity. Deduct- ing from the molecular ratios given above the quantity of tale indicated by experiment we get for the empirical formula of the mineral the expression | male We Met Si0,) ae eas which becomes, if the excess of oxygen is regarded as hydroxyl, with (MgOH),,, as observed, Mg,..K,,H (MgOH) (AlH,0,),.(OH),. The small excess of hydroxy] is probably due to undistributed errors of analysis, and may be added to the MgOH, bringing the latter to 118, and reducing the Mg to 820. Then, general- izing, by uniting all the univalent groups and atoms we get as an ultimate formula 1170 889 109 Mgo.(S10,) -a5F” _ which equals, almost exactly, 125 Mg,(SiO,),R’, +223 Mg,(SiO,),R’,, a result in accordance with our serpentine-chlorite theory. The distribution of the several components of R’ is, however, not clear, and remains to be ascertained. No other discussion of the analysis appears to give as satisfactory results as this, and we have tried several methods of reduction, representing various hypotheses. : One other mineral examined during this investigation remains to be noticed ; a pale yellowish-green mica collected by Mr. G. P. Merrill at a granite quarry in Auburn, Me., near where the Maine Central railroad crosses the Androscoggin river. It occurs in direct contact with ordinary, broadly foliated mus- covite, sometimes forming marginal growths about the plates of the latter mineral, like lepidolite. Analysis gave: 11399 Oe a ea CRU CIN: th 46°54 Pe iA! ee asa” MY Bagg aay epmmenny | BST. EME eA "32 HOA Erte ue Hi BE NS HE AMOSS EN AR Oe 030 0 RU eed. a ae 0°41 Bese wl Tee Boe aes: (none H,O, 105° WSIS CPC iE al Seay IS BIOMj nls? See Wet) 4°72 99°63 This is the composition of muscovite, which the mica undoubt- edly is. The case is interesting, however, as showing a secondary growth of muscovite on muscovite, with a marked difference in outward appearance between the two formations. Laboratory U.S. Geological Survey, Washington, D. C., April 27, 1891. 952 R. D. Salishury—Age of the Orange Sands. ArT. XXIV.—A Further Note on the Age of the Orange Sands ; by R. D. SALISBURY. In arecent number of this Journal, President Chamberlin and the writer set forth what seemed to us sufficient reasons for believing the whole of the Orange Sand series of sands and gravels to be of Pre-pleistocene age. The arguments there adduced we still believe to be sufficient to warrant the conclusions drawn from them. Since that article appeared, some new facts have come to our knowledge which afford new and more direct proof of the correctness of the position then taken. Until this season’s work in the field began, it was not known to us that the Orange Sand gravels reached so far north as the southern bor- der of the glacial drift. They had been searched for along the southern border of the drift north of the area where they are best known, in the hope that they might be found beneath the glacial deposits, but this search had been fruitless, so far as the particular question at issue is concerned. During the early part of this season’s field-work, the writer spent some time in the region between the Mississippi and Illinois rivers above the point of their junction. In this region, in the counties of Calhoun, Pike, Adams and Hancock, the Orange Sand gravels were found to obtain a considerable development. These counties are well north of the southern limit of the glacial drift, and the gravel is uniformly found to occupy a position beneath it. Among other places, this rela- tionship is well shown near Bloomfield, Adams county, where till may be seen resting directly on the brown flint gravels. Here, as at several other localities, the gravel is cemented by iron oxide into a firm conglomerate, though at other points but a few rods away, the gravel is but partially or not at all cemented. It will be remembered that this habit of being firmly cemented at one point, and nearly or altogether free from cement at another, is one of the prominent characteristics of the gravel farther south. But the Pre-pleistocene (presumably Tertiary) series of the counties named is not limited to the brown flint or “ Orange” gravels. Accompanying these, there are very considerable beds of sand, essentially like those accompanying the corre- sponding gravels to the south. These are best exposed, so far as the writer’s knowledge goes, a short distance south of Lib- erty, Adams county, but they have a considerable development in various parts of this county. At the above locality, till may be seen resting on the sand. F. H. Bigelow— Variations of the Magnetic Needle. 2538 Apart from the obvious proof of the preglacial age of these gravels and sands afforded by the superposition of the drift upon them, the character of the till affords a further proof of the same thing. If the sands and gravels occupied the region before the ice invasion, they should have made their contribu- tion to the drift. This they have done, and so generously that at many points and over considerable areas the character of the drift has been in very large measure determined by this contri- bution. To the arguments adduced in the article referred to above for the Pre-pleistocene reference of the Orange Sand gravels and sands, must now be added the further arguments of (1) superposition of the earlier glacial drift upon them, and (2) the contribution of these sands and gravels to this drift. Subsequent to the writer’s determination of the existence of Pre-pleistocene material in the region indicated, reference to the reports of the Illinois Geological Survey revealed the fact that in the reports on Pike and Hancock counties, the Illinois geologists had made note of the fact that ferruginous flint gravels occur beneath the drift in these counties, and that they had further correlated them with the gravels in the south- ern part of the state. To them, therefore, belongs the credit of the first recognition of these gravels, as wholly distinct from the drift.* ArT. XX V.—Wote on the Causes of the Variations of the Magnetic Needle; by Professor FRANK H. BIGELOW. In May, 1890, I published in Bulletin No. 18 of the U.S. Scientific Expedition to West Africa, a preliminary statement of a new theory of terrestrial magnetism which had been con- ceived in order to account for the observed variations of the free magnetic needle. Since that time my efforts have been directed towards obtaining a clear conception of the mode of action of the forces whose relations were indicated in the Bulletin, and [am now prepared to add a note asa further preliminary statement of the progress made in this study. On re-examination of the mode of analysis already published, I find that the main conception is not to be modified and that the successive steps are correct. When making an attempt to reduce the observations by means of this treatment, namely, the combination of current functions by the use of harmonics, it was evident that a very complex system of computation *T am unable at this writing to refer to the Illinois reports, and therefore can- not cite the exact references to the statements therein made. R. D. 8. 254 F H. Bigelow— Variations of the Magnetic Needle. would be required. My endeavor was, therefore, to simplify the fundamental treatment so as to secure not only a sound theory, but also a working process for handling the observations. To do this two distinct sets of trial computations were made, first by the theory of moments about the rectangular axes. whose origin was in the spherical surface passing through the north end of the needle, concentric with the surface of the earth, and forming equations whose solution would give the required constants of the phenomenon. This also failed to be sufficiently simple and direct to show the action in its general relations. The second attempt was an empirical one, for the time abandoning theory, and building up from the simultane- ous observations in various parts of the earth such an exhibi- tion of facts as would display the real nature of the laws behind them. This trial has been successful far beyond antici- pation, and that too in a simple and practical form. The theory is at best complicated, as it depends upon the laws of magnetic induction in their most complex conditions, but it unifies and classifies harmoniously all the visible motions of the needle. My method and result are, briefly, as follows: The month of June, 1883, was selected because of the material collected. in the publications of the International Polar Commission applicable simultaneously over a large area of the earth, also because the north polar stations were at that time exposed to sunlight throughout the twenty-four hours. The stations. used were: Point Barrow, Fort Rae, Kiugua Fjord, Jan Mayen, Bossekop, Sodankyli, Pawlowsk, Wilhelmshaven, Vienna, Tiflis, Za-Ki-Wei, Cape Horn, South Georgien. The monthly means for each hour local time of the horizontal and vertical forces and the declination were reduced to the coérdi-. nates, x positive to the north in the mean magnetic meridian of the month, y positive to the west, 2 positive inwards along the normal, the plane zy being the horizon at the surface of the earth or through the north end of the needle. The differ- ences between the mean and the hourly values, namely JH, AD, 4Z, were plotted on paper, smoothed out, the resulting values dx, dy, dz, combined to show the total deflecting force at the station with its magnetic azimuth and altitude, this. form of azimuth being finally transformed into north geo- graphical azimuth. My idea was that the needle floating freely in a magnetic line indicated simply its direction, and that the deflections were produced by a component coming to it from space, the motive being to discover the condition of such components over the earth at the same time. Next, a large model was constructed on which these component forces were represented in direction and magnitude. By assigning F.. H. Bigelow— Variations of the Magnetic Needle. 255 certain meridians for the hours, and supposing the permanent pole to take up its position from one meridian to another, there was finally collected upon these meridians representing a series. of local hour angles, now referred to the sun as if the earth had ceased to rotate on its axis, an exhibition of what exists. over the globe at the same instant of time. The result is most interesting and gratifying, but I can only indicate now what could be elaborated by a mass of computa- tions. It is difficult to convey any view of the complicated system of lines of force produced by inserting a magnetized or polarized sphere in a field of force, supposing the sphere at rest; if it rotates it is much more troublesome. These refer- ences, however, may be cited: Sir W. Thomson in § xxxii, p. 486, Papers on Electrostatics and Magnetism, illustrates some of the forms produced in the case of symmetry, that is, the axis of polarization being parallel to the field; in article 434, Vol. Il, Maxwell’s Electricity and Magnetism a similar illus- tration is found; in article 436 of the same is given an ex- ample of the sphere being placed at an angle to the field. The mathematical treatment of these cases, when once the constants involved are known, leads to certain typical lines of force entering the sphere at definite angles corresponding to the latitude of the point.. Furthermore when the sphere is rotated the whole system recedes through an angle depending on the constants, as indicated in the Bulletin. My model gives the angles and directions corresponding to such a system, if we take the radiant sunlight as the uniform field of force, direc- ted positive towards the sun. The entering and emerging forces are on the respective sides of the earth, and the whole system is receding by about twenty-three degrees. The peculiar form of the polar station lines and the inclination of their planes of action to the meridians is well displayed. The sta- tions all over the world bear the same testimony. The action of the coronal field is entirely similar but not strong enough to appear on the model. The separation of the two fields is merely a question of close computation, There remains one more important point. The positive direction of the earth’s permanent magnetism is from the north towards the south side of the ecliptic y, the uniform field is positive towards the sun 2, the motion of the earth in its orbit a, is perpendicular to the field. If these are taken as the usual rectangular x, y, 2, they form a consistent positive or right-handed cyclic system. In a word, the permanent mag- netic condition of the earth may be principally due to the orbital motion of the earth through the radiant field of sun- light. The rotation of the earth on its axis causes a modifica- tion of the direction of the axis of polarization, by diminishing 256 Scientific Intelligence. the angle between the two axes, and as the result of the annual motion may cause it to rotate in a secular period about the axis of figure, or if the magnetization has already become set in the body of the earth, may cause a succession of secular waves to sweep over it from east to west, as is shown to be the case in the history of the agonic lines and the long period deflections of the needle. This surprising identification of magnetic and light action of the radiations of the sun in direction will be recognized as harmonizing with the conclusions arrived at by Maxwell and Hertz in their investigations. If light is to be studied as a magnetic phenomenon it adds a large field to the work allotted to meteorology. Furthermore, several important physical con- stants relating to the cosmical action of the sun and the earth, and also the implied nature of the sun and the earth as physical bodies are becoming accessible. Attention is directed to the | fact that such a force acting towards the center of the sun, being cosmical or universal, is of the kind required to account for the outstanding motion in the perihelion of Mercury, not included in the development or the law of gravitation or its positive side. My next step is to form the necessary equations of condition and solve them for the constants involved in the magnetic observations. Washington, D. C., July 31, 1891. SCIENTIFIC INTELLIGENCE I. CHEMISTRY. 1. On Boron tri-iodide.—According to Moissan, boron tri- iodide can be obtained in three ways : either (1) by passing boron chloride and hydrogen iodide through a red hot porcelain tube, (2) by acting with iodine upon boron directly at 700°—800°, or (3) and most conveniently by acting upon amorphous boron, pre- viously dried in a current of hydrogen at 200°, with dry hydrogen iodide gas, the boron being heated in a combustion tube to a temperature near that of the softening of the glass. In this way purple colored scales are obtained containing some free iodine ; from which they may be freed by solution in carbon disulphide agitation with mercury, and evaporation of the solvent. The boron tri-iodide thus obtained is colorless but becomes colored on exposure to light. It is very hygroscopic, fuses at 43°, boils at 210°, burns in the air at a red heat, has at 50° the approximate density of 3°3 and is easily soluble in carbon disulphide, carbon tetrachloride and benzene. By water it is decomposed into boric and hydriodie acids, and it reacts with phosphorus, silver fluoride and magnesium (at 500°) with combustion; though not with Chemistry. 257 aluminum, sodium or silver. With alcohol and ether it reacts, yielding ethyl iodide and ortho-boric acid in the former case and ethyl iodide and ethyl ortho-borate in the latter.—C. &., exil, 717; Ber. Berl. Chem. Ges., xxiv, (Ref.) 387, May, 1891. G. F. B. 2. On Hydrazine hydrate and the compounds of Diammonium with the Halogens. —The researches of Curtius and Scuunz have shown that hydrazine hydrate N,H,.H,O, prepared by distilling the sulphate with potassium hydrate, is a liquid boiling at 118°5° under 739°5 mm. pressure and having at 21° the specific gravity 10305. Its molecular mass at 100° in vacuo is 50, corresponding to the formula N,H,.H,O. At 170° under the ordinary pres- sure, the hydrate is completely dissociated into diamide and water. At higher temperatures the molecular mass diminishes markedly not reaching 50 again at ordinary pressures even at 300° to 400°. Ina lead bath, however, numbers approaching 100 were obtained. Hydrazine hydrate in aqueous solution gave ap- proximately the molecular mass 68, corresponding to the composi- tion N,H,. (H,O),. Comparing this hydrate with ammonia, as to its action on indicators, this action was shown to be as sharp in all cases except that of phenol-phthalein. When aqueous solutions of hydrazine are neutralized with a haloid acid, and evaporated first on the water-bath, then over potassium hydrate, halogen-diam- monium salts are formed, by preference with two equivalents of acid. The bromide and iodide with one equivalent of acid are formed when the free halogen is made to act on an alcoholic solu- tion of hydrazine, a portion of the hydrazine being decomposed. The bi-acid salts crystallize in the regular system, are soluble in water, almost insoluble in alcohol. The mon-acid salts are easily soluble in water and warm alcohol. Tri-hydrazine di-iodhydrate NH, . (HI), is formed when iodine is added to an alcoholic solu- tion of hydrazine so long as crystals appear. As to the molecu- lar mass of the halogen diammonium compounds in aqueous solution, it is found to be with the mono-halogenides, the difluoride and the sulphate equal to one-half, with the di-halogen- ides generally equal to one-fourth and with the tri-hydrazine di-iodhydrate equal to one-fifth the simplest formula.—J. pr. Ch., xlu, 521; Ber. Berl. Chem. Ges., xxiv, (Ref.) 256, Apr. 1891. G. F. B. 3. On the Synthesis of Indigo-carmine.-—HEYMANN has suc- ceeded in effecting the synthesis of indigo-carmine, the disulpho- acid of indigo, by acting upon phenyl-glycocoll with fuming sulphuric acid. If, for example, phenyl-glycocoll be mixed in a test tube with ten to twenty times its mass of fuming sulphuric acid containing 20 to 25 per cent of sulphuric oxide, and gently warmed, it dissolves with a yellow color, evolving sulphurous oxide gas. On pouring the solution upon ice, it rapidly assumes the greenish blue color of indigo-carmine. For its production, the following method gives the best results : One part of phenyl- glycocoll is mixed with 10 to 20 parts of sand and then intro- 250 Scientific Intelligence. duced into 20 times its mass of fuming sulphuric acid, warmed to 20° or 25°, containing 80 per cent sulphuric oxide; the temperature not being allowed to rise above 30°. The glycocoll goes easily into solution with a yellow color which at once with evolution of sulphurous oxide passes into the deep blue color of the indigo-solution. To remove the concentrated acid, the mass is diluted with sulpburic acid of 66° B. The coloring matter is isolated by farther dilution with ice and the addition of salt. As so prepared the product is completely pure indigo-carmine. The colors obtained in dyeing with it far exceed in brillianecy those obtained from the best varieties of commercial indigo. Its identity with the natural product was established by means of its chemical reactions, by dyeing tests and by spectroscopic exami- nation. The yield is about 60 per cent of the glycocoll taken.— Ber. Berl. Chem. Ges., xxiv, 1476, May, 1891. G. F. B. 4. Lecons sur les Métaux, professées 4 la Faculté des Sciences de Paris. Par AtFrrep Dirt, Professeur de Chimie 4 la Faculté. Premier Fascicule. 4to, pp. 44, lvili, 621. Paris, 1891. (Vve Ch. Dunod.)—To judge from the part of Professor Ditte’s book now before us, the complete work will be a valuable addition to chemical literature. It is written largely from the standpoint of energy. In his preface the author says: ‘“ The principles of Thermo-chemistry and the consequences which flow from them, teach us not only to explain reactions, but also frequently to fore- see them and to discover in advance what phenomena will be produced when two or more substances are put together under determined conditions. . On the other hand when two reactions are simultaneously possible the laws of dissociation enable us to define rigorously the conditions of eqilibrium which must be established between them. In general a rational application of these principles and these laws enables us to say, often even before making the experiment, why one given reaction is certain to result, while another reaction is impossible; why an action which begins without difficulty, ceases after a time; and finally why a particular phenomenon occurring under certain circum- stances, does not take place under other circumstances entirely similar apparently in appearance.” The introductory portions of the baok are therefore devoted to calorimetry and the general principles of Thermo-chemistry as laid down by Berthelot. In the First part, a general study of the metals is given, covering about four hundred pages. It includes the principles of metal- lurgy, the physical properties of the metals and their alloys, their compounds with the non-metals, the action of water, acids, etc., on the metals, and lastly metallic salts. The Second part is. devoted to the study of the metals specially. Throughout the book all the reactions are given as energy-reactions, and repre- sent the heat-changes concerned, thus : SnCl, + 2HO =SnO, + 2HO},,.. + [67°9 + 2.39°3 — 64°6 — 2,.34°5] [+ 12:9]. Geology and Natural History. 259 ‘The notation used in the book is the old equivalent notation ; which seems unfortunate since it is not in accord with that based on the atomic theory now generally employed. The great advantage of considering the heat-changes in all reactions, and the evident care with which the descriptive part has been written, will make Pro- fessor Ditte’s book acceptable to the chemist. We shall look with interest for the remaining parts. Goji Be Il. GEoLocy AND NatTuRAL HiIstTory. 1. Composition of the Till or Bovwlder-Clay; by W. G. Crossy (Proc. Bost. Soc. Nat. Hist., xxv, 1890).—In this paper Professor Crosby gives the results of an investigation of the glacial deposits in the vicinity of Boston. His analyses show that the proportion of true clay in the till is small and that of rock-flour, or very finely pulverized rock, is large. He concludes that the proportion of stones over two inches in diameter is not over 5 to 10 per cent. His results give for the gravel, 24:90 per cent ; the sand, 19°51; the rock-tlour, 43°86 ; the clay, 11°67 = 99°94. In his table, each of these divisions of the material, is farther subdivided into coarse, medium and fine. Moreover, he gives his results for each of the different localities studied. In the redistribution of the material by the glacial flood, the rock- flour goes with the clay, adding to its volume, so that the clay- beds embrace fully half of .the original material of the till, The rock-flour was found to be essentially quartz-flour—this being the final result of disintegration and the consequent decomposition— according with Daubrée’s observation that the milky turbidity of the Rhine, even for hundreds of miles from the Alpine glaciers, is due chiefly to impalpable quartz. It is further con- cluded that of the material of the till, one-third is probably of preglacial erosion, and two-thirds of glacial erosion. The amount of rock-flour is evidence in favor of this. But the fact does not, Professor Crosby observes, lend support to the view that the _ glacier “profoundly modified the topography of the glaciated area.”’ ‘These are a few of the important facts and conclusions in Professor Crosby’s excellent paper. 2. Geology of the Rocky Mountain Region in Canada with special reference to changes in Elevation and to the History of the Glacial Period; by Dr. G. M. Dawson.—The eighth vol- ume of the Transactions of the Royal Society of Canada, con- tains, among its papers, the very valuable Presidential Address of Dr. G. M. Dawson on the above subject. The Mesozoic and Tertiary history occupies 22 pages, and the Glacial history the following 50 pages. 3. The Greenstone Schist areas of the Menominee and Mar- quette regions of Michigan ; by Professor G. H. Witiiams. 218 pp- 8vo, with plates and cuts. Bulletin U.S. Geol. Survey, No. 62.—The important subject here discussed ably and with great fulness, by the author is—the Methods in which a massive crys- 260 Scientific Intelligence. talline rock may be modified by the action of orographic forces. The three methods mentioned—the Macro-structural, Micro- structural and Mineralogical, are severally considered, and the results under each, as recognized by the author, are described in detail, and illustrated by his microscopic study of the Green- stone schist and the associated rocks. 4, Some Botanic Gardens in the Equatorial Belt and in the South Seas. (Second paper.)—The voyage from Colombo, Ceylon, to Adelaide, South Australia, is not far from 4,400 miles, After leaving the harbor, land covered with tropical vegetation and shores fringed with mangrove are kept in sight, until the once prominent port at Point de Galle sinks from view, and then a fairly straight run is made for Cape Leeuwin. Rounding this, the distant shores of West Australia are skirted as far as the Head near King George’s Sound, at Albany, after which no land is seen until Kangaroo Island is reached about the fourteenth day out. The landing is made in a steam launch which runs in all weathers, sometimes in pretty rough water, through an open roadstead, up to a jetty at the Semaphore, the terminus of a suburban railway leading to the City of Adelaide. Passengers by the Peninsular and Oriental line land at Glenelg, a little farther south. The clouds of grasshoppers which met us at the landing did not presage a very happy condition of things in the fields and gardens. But the mischief thus far wrought by them had been local and hardly so severe as had been dreaded. It was now the middle of December (the southern summer) and the ground seemed dry, but the crops around the city were in good color and strong. OES comparatively new city, Adelaide is fairly well shaded with trees. ‘The suburbs are attractive. Northeast of the city proper, and within a few minutes walk from the principal streets, the University and the Botanic Garden are found near together. Adelaide.—The Botanic Garden occupies an area of about forty acres, and adjoins parklands which are used as an arboretum. From his entrance at the main gate, throughout his whole tour of the garden, the visitor is struck by the more or less successful attempts at decorative management of shrubbery and marble statuary, indicating that there has been a desire to make a place which is easily accessible very attractive to the public. The result is generally pleasing; in fact, it is all good, except in the case of the water, which leaves much to be desired. Australian plants are represented by pretty good specimens, but the conditions for culture are not favorable. The soil ap- peared thirsty and for the most part light; hence the fair success attained shows excellent judgment in cultivating. As will be seen by the photographs at Cambridge, the large specimen trees would be a credit to any garden, and the groups of European florist-plants are about as good as one could expect to see any- where. It was said to me that these European groups are among Geology and Natural History. 261 the main attractions of the garden to the citizens. The citizens with whom I conversed were justly proud of the establishment. The Victoria regia house is one of the principal features of the garden, but the condition of the plants at the time of my visit was a disappointment. It seemed as if the method of heating by water from an open boiler might be at fault. It was not easy to see how the water could contain as much air as in the ordinary method of heating by pipes through the tank, and it appeared as - if this was at the bottom of the mischief. The other houses looked well for the season. The selections in them appeared judicious and many individual plants were of exceptionally good growth. Considerable prominence was given to horticultural, and, one may say, utilitarian aspects of vegeta- tion. This doubtless serves to augment the interest felt by the general public, from whom directly or indirectly all the funds come. A Museum for economic botany, well-arranged, and full of good illustrations of the subject, occupies a conspicuous place in the grounds. Its most attractive department is a collection of the leguminous plants which have proved pernicious to stock. The carpological series is good, and the products of the useful plants are well displayed. In a separate apartment was seen the herbarium of the director, Dr. R. Schomburgh, who was even then prostrated by illness which has since terminated fatally. Although confined to his room and a great sufferer, the ven- erable Director received me on two occasions and conversed freely about his plans, all of which looked in the direction of increasing the local interest in Botany and Horticulture. Very profitable botanical excursions can be made from Ade- laide. ‘The hand-book for the neighborhood is a recent Flora by Professor Tate of the University. It is handy and accurate. Visitors who may have time for botanical studies in any of the Australasian colonies should be reminded that in almost every large city there can be found a botanist or two well acquainted with the most desirable localities for herborizing. Judging from my Own experience in obtaining their advice, these local botanists are not easily wearied in well-doing. Some of the local collec- tions are enriched by notes taken on the spot, and possess great interest. Melbourne.—It was my good fortune to make the journey from Adelaide through Ballaarat to Melbourne in company with Mr. Samuel Dixon, of South Australia, who has acquired an ex- cellent knowledge of the flora, and has occupied himself with some of the more interesting industrial questions connected with the forage plants of the Colonies. The first and last part of this railway journey of about 500 miles was made in the afternoon and early morning, and gave a glimpse of high lands and of the dreary desert scrub, with here and there a view of good soil and rich growth, 262 Scientifie Intelligence. In passing, it may be noted that the railway journey northeast from Adelaide to the famous silver mines at Broken Hill in New South Wales brings before the tourist capital illustrations of true Australian deserts. In fact, the town of Broken Hill lies within sight of one of the spots where the great explorer Sturt was imprisoned by the lack of water. The scanty vegetation fur- nishes, aS so many Australian plants elsewhere do, striking instances of adaptation to a dry climate; the locality is so readily accessible that it should not be left unvisited. Mr. Dixon gave many facts relative to the utilization of deserts and of desert plants in that region, which he has incorporated in an instructive article published in the Proceedings of the Royal Society of South Australia (vol. viii). I had also the benefit of Mr. Dixon’s guidance, the following day, in my first informal visit to the Botanic Garden of Melbourne. The garden is about a mile south of the city itself, and lies on the narrow river, the Yarra, which flows through Melbourne. The situation is good, but the soil in some parts is far from the best. The Australian flora is represented by fine old specimens, if one can call anything old in such a new country, and by young plants which have been added in recent years. | The garden abounds in effective views which are much appre- ciated by the citizens. As in all new countries, for instance, our own, there is a good deal of pleasant rivalry between the larger places; in the botanical gardens of Australia there is found a convenient object for comparison. ‘The judicious visitor cannot go far wrong in his answers to anxious inquiries as to respective excellences of the gardens, where there is so much to please and so little that calls for unfavorable criticism. In the Melbourne Garden the most attractive groups of plants are (1) the Proteaceous, composing a striking mass of shrubs and small trees many of which happened to be in flower: (2) the Eucalypts, by no means all of them equally good as representa- tives, but most of them having distinctive characters recogniza- ble as a whole; (8) the Myrtacez, and the Acacias. The best photographs of the Melbourne garden are those which show the groups in question, and two in which the Governor’s residence is seen in the distance. The Director is much embar- rassed by the peculiar condition of the labor question in Australia. Somewhat similar difficulties arise wherever the one in charge cannot employ or dismiss the workmen for whose good work he is held responsible. Mr. W. R. Guilfoyle, the Director, with his capable assistants, is organizing a Museum of Economic Botany, and bringing together other appliances for the illustration of botany. Not very far from the Garden lives Baron. Ferdinand von Mueller, Government Botanist. He is surrounded by his Herba- rium and Library to which he has devoted his life and fortune. The correspondence which he carries on is incredibly voluminous, and it is understood to be conducted wholly with his own hand. Geology and Natural History. 263 His Handbook of Victorian Plants is easy to use after one be- comes accustomed to the dichotomous arrangement, and it is very helpful in the minuteness of its descriptions. The Baron has done hard work in economic botany as applied to Australia and in the endeavor to make the useful plants of the colonies better known in the Old World and in America. As everybody knows, such work is always a thankless and ungracious task, for the mistakes and failures in the introductions are likely to outnumber the successes. In everything concerning the advancement of the colonies, Baron Mueller has taken a hearty interest and is in every way identified with his adopted home. Besides the incipient economic Museum at the Botanic Garden and the large collections under the charge of Baron Mueller there is an interesting botanical department connected with the Melbourne Museum. This is under the direction of Mr. J. Cosmo Newbery, and is now being re-arranged previous to its final disposition in the new Museum building. The specimens which illustrate the cereals and their products were nearly ar- ranged at the time of my visit and indicated that the new system would be successful from an educational point of view. The environs of Melbourne include many municipalities which are commonly counted in with Melbourne proper, when the city is compared with its sister cities. Passing outside the circle of associated communities, the botanist comes upon very instructive © botanical ground. One does not have far to go by rail to stand before the giant specimens of Eucalyptus, and by boat to be in the presence of queer Australian plants, like Epacris and so on, growing wild. Sydney.—The third great Australian Garden is in New South Wales, about six hundred miles from Melbourne, Victoria. For beauty of situation it stands without a rival. It has a com- manding position on the shore of the harbor, and _ possesses remarkable elements for landscape treatment. The harbor of Sydney (Port Jackson) is one of the most cele- brated in the world, usually being associated with that of Rio de Janeiro, as the finest in existence. Like outstretched divergent fingers, promontories extend into this charming sheet of water. On parts of the slopes of two of these the Botanic Gardens, covering about forty acres, have been established. As was to be ‘expected, the representation of native plants is somewhat different from that in the other gardens, owing to difference in the climate. In certain directions, for instance, palms from the smaller islands of the Polynesian archipelago, the garden is exceptionally rich. The specimens are numerous and well grown. A good deal of attention has been paid also to economic plants. The most inter- esting photographs which I could secure were (1) Individual plants; (2) the Palms of Lord Howe’s Island, ete.; (3) the ceneral view from the brow of the hill. To Mr. Charles Moore, the director, I am indebted for many views of the garden, taken some time since. The collection of all these now at Cambridge Am. JouR. ScI.—THIRD SERIES, Vou. XLII, No. 249.—SxEpt., 1891. bila ct: = ————— Sete ee 264 Scientific Intelligence. illustrates fairly well the wide range of cultivation possible in this favored climate. Botany Bay of the early navigators lies within easy excursion distance of the city of Sydney. There and in the contiguous peninsulas, one can see growing wild the native plants which gave the place its appropriate name. In point of fact, the garden at Sydney was visited considerably later by me than those at Adelaide and Melbourne, a journey through Tasmania and New Zealand intervening. But it has seemed best to bring the three larger gardens together in a single sketch, reserving the visit to the economic museum in Sydney for a third communication. Before leaving the subject of these three gardens, it may not be out of place to call attention again to the deep interest and local pride felt by the people of the respective cities in these establishments. Every intelligent person with whom I conversed upon the subject appreciated the importance of such institutions in a country with undeveloped resources. It was also felt that, since these gardens, and the smaller ones, for that matter, keep in touch with Kew, the botanical interests of the colonies, particu- larly in their economic aspects, were receiving due attention. The Botanic gardens of the south do not appear to sustain any close connection with the Universities. They are, of course, available for purposes of investigation, but they are governmental and not academic instititions. It is frequently said that in the southern hemisphere everything is reversed from what is found in the northern. This is certainly not true of the budgets for botanical gardens. These institu- tions are everywhere very popular, but I did not find in any case that too much money was provided for the running expenses. In fact, I observed no instance where a somewhat lar ger income would not have improved the condition of affairs. But the directors and superintendents of the larger gardens, and the curators of the smaller ones made the best use of the rather scanty funds placed at their disposal. . The position of government botanist (in Victoria), filled by the distinguished von “Mueller, seems at first anomalous. But when it appears that, as matter of fact, this position has left its in- cumbent far more free to elucidate botanical questions affecting all the colonies, than if he were burdened with administrative duties connected with the botanical garden in one colony, the establishing of the office has had happy results. It may not be out of place to say that on every hand in the colonies Baron von Mueller’s preéminence receives hearty recognition, even in quarters where the relations might naturally have been some- what strained. The willingness with which the government botanist comes to the assistance of young botanists and amateur collectors in the colonies may have had much to do with the general interest in botanical matters exhibited in the three most populous colonies. G. L. G gE ND L:X.. Art. XX VI.—WNotice of New Vertebrate Fossils; by O. C. Marsa. REcENT researches on a number of extinct animals have made it evident that several of them are new to science, and that others possess some characters of interest which have not hitherto been observed. In the present paper, some of the results of this investigation are placed on record, and others will be given in a later communication. CERATOPSIDA, Triceratops elatus, sp. nov. One of the largest members of the Ceratopside, representing a distinct species, is at present known from the skull only, which was secured during the past year. Although this skull is about six feet and a half in length, it belonged to an animal scarcely adult, as indicated by some of the cranial sutures. The rostral bone is not codssitied with the premaxillaries as in old animals, and the superior branch of the former bone has its extremity free. The nasal horn-core, however, is firmly codssified with the nasals. It is of moderate size, with an obtuse summit directed upwards. ‘The main horn-cores were quite long, with their extremities pointed and directed well forward. These horn-cores are compressed transversely, the section being oval in outline. One of the most striking features of the skull is the parietal crest, which was quite elongate, and much elevated, more so than in any of the species hitherto discovered, and this has suggested the specific name. The length of this skull from the front of the rostral bone to the back of the parietal crest was about seventy-eight inches, and the greatest transverse expanse of the posterior crest was about forty inches. The summit of one of the frontal horn- cores was about twenty-eight inches above the orbit, and fifty-three inches from the base of the quadrate. This interesting specimen was found in the Ceratops beds of the Laramie, in Wyoming, by Mr. J. B. Hatcher of the U.S. Geological Survey, whose previous discoveries are well-known. 266 O. C. Marsh—WNotice of New Vertebrate Fossils. Torosaurus latus, gen. et sp. nov. Another well-marked species of this group, which may be referred to a new genus, is represented by one skull, and parts of the skeleton, from nearly the same horizon as the specimen above described. One of the most striking features of the present species is seen in the posterior crest, which, instead of being complete as in the skulls hitherto found, is perforated by a pair of large openings. These are in the parietals, but they have the inner margin of each squamosal for their outer border. They are well behind the supra-temporal fossee, but doubtless were originally connected with them. They may be called the supra-temporal fontanelles. ‘The squamosal bones, moreover, are very long and slender, and distally only show near the ends sutures for union with the parietals. Another distinctive character is seen in the main horn-cores, which are placed well back of the orbit. The nasal horn-core is short, with the apex compressed, and directed forward. : This genus is of much interest, as it represents an earlier and less specialized form than either Ceratops or Triceratops, both of which have the posterior crest complete. The existing Chameleons show the other extreme, where the outline only of the parietal crest has been attained. Some of the principal dimensions of this skull are as follows: Length from apex of nasal horn-core to extremity of squamosal ic SoS ee 3. Be sen ee 80 inches. Distance from same apex to front of orbit.---.----- 7 Distance from same to front of parietal opening --.. 54 “ Width between posterior extremities of squamosals.-. 56 “ This important specimen was discovered by Mr. J. B. Hatcher, in the Laramie of Wyoming. Torosaurus gladius, sp. nov. A second species of apparently the same genus is represented by various portions of a skull in good preservation. In this specimen, the nasal horn-core is short and obtuse, and nearly upright. The main horn-cores are elongate, oval in outline, and in position resemble those of the skull above described. The most remarkable features in the present specimen are the squamosal bones, which are greatly elongated, and so attenuated as to have the general shape of the blade of a sword, thus suggesting the specific name. These bones, moreover, show but slight evidence at their distal extremity of union with the parietals, as the inner margin is rounded for nearly half the length. This feature will distinguish the present species from all others hitherto described. O. O. Marsh—Notice of New Vertebrate Fossils. 267 The following are some dimensions of portions of this specimen : : Length of horn-core from top of orbit to summit._-. 27 inches. Antero-posterior diameter of same horn-core at base. 8 “ iimmaverse diameter of same ..2.-..2..--1--/.----- eines Length of squamosal behind exoccipital groove ----- So eer © es SiSt PIUC UBIO IIe Re ae el eo tl, "na ee sae en Myadth atmiddle__- 2... -- sii i, [ict aad ae el aR a OP hise These interesting specimens were also found in the Laramie of Wyoming by Mr. J. B. Hatcher. ANCHISAURID A. Ammosaurus, gen. Nov. The Yale Museum has recently secured two interesting specimens of Dinosaurs from the Triassic sandstone of the Connecticut valley. In comparing these with the known species of Anchisawrus from this formation, the fact became evident that among them are two well-marked genera. One of the specimens, which is described below, cannot now be distinguished generically from the type of Anchisaurus, while the one described by the writer as Anchisaurus major is quite distinct, and hence a new genus is here established for its reception. ‘The distinctive characters are well marked in the pelvic arch. There are three vertebree in the sacrum, but they are not eoossified with each other, being free, as in the Crocodélia. The ilium is comparatively small, and has a slender pre-acetab- ular process. ‘The pubes are broad, elongate plates, perforate above, and not codssified with each other. In form, they resemble the corresponding bones in Zanclodon, where, how- ever, the two are coossified, and imperforate. The ischia meet the pubes by an extensive union. ‘Their distal ends are slender, directed backward, and closely adapted to each other. This species may now be known as Ammosaurus major. Anchisaurus colurus, sp. nov. The new species is represented by perhaps the most perfect Triassic Dinosaur yet discovered, as the skull and greater portion of the skeleton were found in place, and in fine preservation. It is smaller than the specimen above described, but similar in its general proportions, yet the two may be readily distinguished by the pelvic arch and posterior limbs. The pubes are distinct from each other, imperforate above, and the distal portions are only moderately expanded. The process that projects backward to meet the ischium is slender, © and the face for union with that bone is quite small. The sacrum and ischia resemble those of Ammosaurus above described. 268 =O. C. Marsh— Notice of New Vertebrate Fossils. The skull is of moderate size, and of delicate structure. In its general shape, it somewhat resembles the skull of Hatteria. The supra-temporal fosse are very large, and the orbits especially so. The quadrate is inclined forward, and the upper and lower temporal arches are slender. Compressed, cutting teeth are present both in the premaxillary and max- illary bones. The lower jaws have similar teeth, and the rami are not united to each other at the symphysis in front. The vertebree and limb bones are hollow, and the whole skeleton is lightly built. The neck is long, and the tail of moderate length. The scapula is elongate, and the coracoid very small and imperforate. The humerus has a strong radial crest, and the radius and ulna are nearly equal in size. There were five digits in the manus, the first, second, and third being armed with strong claws. The temur is longer than the tibia, and has a flattened head, somewhat like that of a crocodile. The tibia is short and stout, and the fibula well developed. The astragalus is not coossified with the tibia, and the caleaneum is distinct. There were five digits in the pes, but only four functional, the fifth being represented by the metatarsal alone. The skull of this reptile is about five and one-half inches long, and the lower jaw four and one-half inches. The scapula and humerus are of equal length, each about six inches long. The femur is about eight inches in length, and the tibia about six. The animal when alive was about five and one-half feet long. The present remains were found near Manchester, Conn. A more complete description of this interesting reptile, with illustrations, will soon be published. BRONTOTHERID &. Allops crassicornis, sp. nov. The present species is represented by the nearly perfect skull of an adult, but not old animal. The skull is of medium size, with the zygomatic arches moderately expanded. The nasal bones do not project beyond the premaxillaries. The horn-cores are very short and massive, with rounded summits, and thus form one of the striking features of the skull. The dentition is complete, and in fine preservation. The single incisor is quite small, and situated close to the canine. The latter is of moderate size, and projects but little above the rest of the dental series. There is no diastema between the canine and the first premolar, which is small, and has its inner face on a line between the canine and the second prémolar. The second, third, and fourth premolars are large, and have a strong inner basal ridge. The last molar has its anterior margin somewhat in advanee of the front border of the posterior nares. O. C. Marsh—Notice of New Vertebrate Hossils. 269 The length of this skull on the median line is about thirty inches, and the width across the zygomatic arches twenty-three inches. The width across the horn-cores is fourteen inches. The*extent of the superior dental series is sixteen inches. The type of this species was found in the Brontotherium _beds of South Dakota, by Mr. J. B. Hatcher. Brontops walidus, sp. nov. This: well-marked species is based upon a skull in fine preservation, which agrees in its main characters with the other species of this genus, but is particularly short and robust. The zygomatic arches are widely expanded, almost as much as in any skull of this group. The nasal bones have only a moderate extension in front, and do not reach the end _ of the premaxillaries. The free portion is broad and massive. The horn-cores are of moderate size, nearly round in section, and have their obtuse summits directed somewhat backward. The occipital crest slopes forward, and is expanded transversely. The length of this skull on the median line is about twenty-six inches. The greatest transverse diameter across the zygomatic arches is twenty-two inches, and across the summits of the horn-cores, fourteen inches. The type specimen of the present species is from the Brontotherium beds of South Dakota, where it was secured by Mr. J. B. Hatcher. Titanops medius, sp. nov. The present species is from nearly the same horizon as the type of the genus, but is of smaller size. It is represented by one skull in fair preservation, with the horn-cores and dentition complete. The free portion of the nasals is very small, and projects but slightly beyond the anterior line of the horn-cores. The latter are compressed antero-posteriorly, and project laterally nearly at right angles to the median line of the skull. The two incisors on each side are quite small, and separated from each other and from the canine. There isa slight diastema behind the canine. The first premolar is small, and triangular in outline. ‘The second premolar is of moderate size, and the third and fourth premolars have only an incomplete inner basal ridge. The width of this skull across the horn-cores is twenty-three inches, and the distance from the end of the nasals to the front of the posterior nares is sixteen inches. The extent of the upper dental series is seventeen inches. This specimen is from near the top of the Brontotherium beds of South Dakota, where it was discovered by Mr. J. B. Hatcher. New Haven, Conn., August 10th, 1891, ive Rie y ee By ‘ cPptys te NVR CE SUS ce & PV oatbigtaae 90) ae tae a had apes | : tah 7 g Wy PPA terres ee Ce i , aw" -_ ‘ & é LA - " [ ) P ' ee aoe Phd ta) is sh ‘aE, ey ba ioe _ © s ; + f ai i” i =a ty 3 : ¥ Tt +1 >) TA éree: i tyes as . rm ; © ; is SEGSTE ees hg + 3 ee 2) i 4 / he ee if A } re Phy, Lay ike eer 7 ty 1 PE wis IM # > a ; a Aik a y mee M af é 7 2 : % ‘ { 7 ld ‘ aye ; ‘ \ i 4 4 J of ; ; ee ne ¥ ; we - s / “te } 4 3 ; i i CRF ney ? 7 ¥i 7 } als | ¥ :S + ; r ‘ i) s » ’ , . ’ ? j \ vice ‘ , t, Ly l F { : : f : acd ’ «, r Te ' ¢ ‘ ' st y r Fe ’ rf & . s MINERALS THE LATEST ACCESSIONS TO OUR STOCK. mee’ __In General.— We have been aiming lately to strengthen our stock of “ad RARE SPECIES, and we believe we now have a larger number of species than has ever before been offered for sale in this country. _ We have also added largely to our stock of MINERALS FOR BLOW- ' PIPE ALALYSIS, in which line we take pride in claiming that our stock is unsurpassed in the world. We are now ready for the fall fresh supplies of first-class material. Of FINE CABINET SPECIMENS our stock is stronger than ever, three experienced collectors having been devoting their time to Securing the best for us. a So Art. XX VII. Agr, of the Possibilities of Hconomec Botany; by GEoRGE LINCOLN GOODALE. [Presidential address delivered before the American Association for the Advance- ment of Science, at Washington, August, 1891.] Our Association seaman of its president, on his retirement from office, some account of matters connected with the department of science in which he is engaged. But you will naturally expect that, before I enter upon the discharge of this duty, I should present a report respecting the mission with which you entrusted me last year. You desired me to attend the annual meeting of the Australasian Association for the Advancement of Science, and express your good wishes for its success. Compliance with your request did not necessitate any material change in plans formed long ago to visit the South Seas; some of the dates and the sequence of places had to be modified; otherwise the early plans were fully carried out. I can assure you that it seemed very strange to reverse the seasons, and find mid-summer in January. But in the meeting with our brethren of the southern hemisphere, nothing else was _reversed. The official welcome to your representative was as cordial, and the response by the members was as kindly as that which the people in the northern hemisphere would give to any fellow-worker coming from beyond the sea. The meeting to which I was commissioned was held in January last in the Cathedral city of Christchurch, New Lea- land, the seat of Canterbury College. Am. JouR. Scl.—THIkD SERIES, Vou. XLII, No. 250.—Octossr, 1891. 272 G. L. Goodale—Possibilities of Economic Botany. Considering the distance between the other colonies and New Zealand, the meeting was well attended. From Hobart, Tasmania, to the southern harbor, known as the Bluff, in New Zealand, the sea voyage is only a little short of one thousand miles of rough water. From Sydney in New South Wales to Auckland, New Zealand, it is over twelve hundred miles. If, therefore, one journeys from Adelaide in South Australia, to Christchurch, New Zealand, where the meeting was held he travels by land and by sea over two thousand miles. From Brisbane in Queensland, it is somewhat farther. Although certain concessions are made to the members of the Associa- tion, the fares by rail and by steamship are high, so that a journey from any one of the seats of learning in Australia proper to New Zealand is formidable on account of its cost. It is remarkable that so large a number of members should have met together under such circumstances, and it speaks well for the great strength and vigor of the Association. The Australasian Association is modelled rather more closely after the British Association than is our own. ‘The president delivers his address upon his inauguration. There are no gen- eral business meetings, but all the details are attended to by an executive committee answering to our council; none except the members and associates are invited to attend even the sectional meetings and there are some other differences be- tween the three associations. The secretaries stated to me their conviction that their organization and methods are better adapted to their surroundings than ours would be, and all of their arguments seemed cogent. Although the Association has been in existence but three years, it has accomplished great good. It has brought together workers in different fields for conference and mutual benefit; it has diminished misunder- standings, and has strengthened friendships. In short it is doing the same kind of good work that we believe ours is now doing, and in much the same way. Your message was delivered at the general evening session immediately before the induction of the new officers. The retir- ing president, Baron von Mueller, and the incoming president Sir James Hector, in welcoming your representative, expressed their pleasure that you should have seen fit to send personal ereetings. In replying to their welcome, I endeavored to convey your felicitations upon the pronounced success of the Association, and your best wishes for a prosperous future. In your name, I extended a cordial invitation to the members to gratify us by their presence at some of our annual meetings, and I have good reason to believe that this invitation will be accepted. I know it will be most thoroughly and hospitably honored by us. G. L. Goodale—Possibilities of Economic Botany. 273 On the morning of the session to which I refer, we received in the daily papers, a cable telegram relative to the Bering Sea difficulties (which were then in an acute stage). In your stead, I ventured to say, ‘In these days of disquieting dispatches, when there are rumors of trouble between Great Britain and the United States, it is pleasant to think that ‘ blood is thicker than water.’” This utterance was taken to mean that we are all English-speaking kinsmen, and even before I had finished, the old proverb was received with prolonged applause. The next meeting of the Australasian Association is to be held in Hobart, the capital of Tasmania, under the presidency of the governor, Sir Robert Hamilton. The energetic secre- taries Professor Liversidge, Professor Hutton and Mr. Morton, promise a cordial welcome to any of our members visiting the Association. Should you accept the invitation, you will enjoy every feature of the remarkable island, Tasmania, where the meeting is to be held. You will be delighted by Tasmanian scenery, vegetation and climate, but that which will give you the greatest enjoyment in this as in other English South Sea colonies, is the fact that you are among English-speaking friends half way around the world. You will find that their efficient Association is devoted to the advancement of science and the promotion of sound learning. In short you will be made to feel at home. The subject which I have selected for the valedictory address deals with certain industrial, commercial and economic questions: nevertheless it lies wholly within the domain of botany. I invite you to examine with me some of the possi- bilities of economic botany. Of course, when treating a topic which is so largely specu- lative as this, it is difficult and unwise to draw a hard and fast line between possibilities and probabilities. Nowadays, possi- bilities are so often realized rapidly that they become accom- plished facts before we are aware. In asking what are the possibilities that other plants than those we now use may be utilized we enter upon a many-sided inquiry."* Speculation is rife as to the coming man. May we not ask what plants the coming man will use ? There is an enormous disproportion between the total num- ber of species of plants known to botanical science and the number of those which are employed by man. The species of flowering plants already described and named are about one hundred and seven thousand. Acquisitions from unexplored or imperfectly explored regions may increase the * For references, notes, etc., see p. 300. 274 G. L. Goodale—Possibilities of Economic Botany. ageregate perhaps one-tenth, so that we are within very safe limits in taking the number of existing species to be somewhat above one hundred and ten thousand.’ Now if we should make a comprehensive list of all the flowering plants which are cultivated on what we may calla — fairly large scale at the present day, placing therein all food’ and forage plants, all those which are grown for timber and cabinet woods, for fibres and cordage, for tanning materials, dyes, resins, rubber, gums, oils, perfumes and medicines, we could bring together barely three hundred species. If we should add to this short catalogue all the species, which with- out cultivation, can be used by man, we should find it consid- erably lengthened. A great many products of the classes just referred to are derived in commerce from wild plants, but exactly how much their addition would extend the list, it is impossible in the present state of knowledge to determine. Every enumeration of this character is likely to contain errors from two sources: first, it would be sure to contain some species which have outlived their real usefulness, and, secondly, owing to the chaotic condition of the literature of the subject, omissions would occur. But after all proper exclusions and additions have been made, the total number of species of flowering plants utilized to any considerable extent by man in his civilized state does not exceed, in fact it does not quite reach, one per cent. The disproportion between the plants which are known and those which are used becomes much greater when we take into account the species of flowerless plants also. Of the five hundred ferns and their allies we employ for other than decorative purposes only five; the mosses and liverworts, roughly estimated at five hundred species, have only four which are directly used by man. There are comparatively few Algae, Fungi, or Lichens which have extended use. Therefore, when we take the flowering and flowerless to- gether, the percentage of utilized plants falls far below the estimate made for the flowering alone. Such a ratio between the number of species known and the number used justifies the inquiry which I have proposed for discussion at this time—namely, can the short list of useful plants be increased to advantage? If so, how % This is a practical question ; it is likewise a very old one. In one form or another, by one people or another, it has been asked from early times. In the dawn of civilization, mankind inherited from savage ancestors certain plants, which had been found amenable to simple cultivation, and the products of these plants supplemented the spoils of the chase and of the sea. The question which we ask now was asked then. Wild —_ el i ae Ee G. L. Goodale— Possibilities of Economic Botany. 275 plants were examined for new uses; primitive agriculture and horticulture extended their bounds in answer to this inquiry. Age after age has added slowly and ‘cautiously to the list of cultivable and utilizable plants, but the aggregate additions have been as we have seen, comparatively slight. The question has thus no charm of novelty, but it is as prac- tical to-day as in early ages In fact, at the present time, in view of all the appliances at the command of modern science and under the strong light cast by recent biological and tech- nological research, the inquiry which we propose assumes great importance. One phase of it is being attentively and syste- matically regarded in the great Experiment Stations, another phase is being studied in the laboratories of Chemistry and Pharmacy, while still another presents itself in the museums of Economie Botany. Our question may be put in other words, which are even more practical What present likelihood is there that our tables may, one of these days, have other vegetables, fruits and cereals, than those which we use now? What chance is there that new fibers may supplement or even replace those which we spin and weave, that woven fabrics may take on new veg- etable colors, that flowers and leaves may yield new perfumes and flavors? What probability is there that new remedial agents may be found among plants neglected or now wholly unknown? The answer which I shall attempt is not in the nature of a prophecy; it can claim no rank higher than that of a reasonable conjecture. At the outset it must be said that synthetic chemistry has made and is making some exceedingly short cuts across this field of research, giving us artificial dyes, odors, flavors, and medicinal substances. of such excellence that it sometimes seems as if before long the old-fashioned chemical processes in the plant itself would play only asubordinate part. But although there is no telling where the triumphs of chemical synthesis will end, it.is not probable that it will ever interfere essentially with certain classes of economic plants. It is impossible to conceive of a synthetic fiber or a synthetic fruit. Chemistry gives us fruit-ethers and fruit-acids, and after a while may pro- vide us with a true artificial sugar and amorphous starch ; but artificial fruits worth the eating or artificial fibers worth the spinning are not coming in our day. Despite the extraordinary achievements of synthetic chemis- try, the world must be content to accept for a long time to come, the results of the intelligent labor of the cultivator of the soil and the explorer of the forest. Improvement of the good plants we now utilize, and the discovery of new ones must remain the care of large numbers of diligent students 276 G. L. Goodale— Possibilities of Economic Botany. and assiduous workmen. So that, in fact, our question resolves itself into this: can these practical investigators hope to make any substantial advance ? It will be well to glance first at the manner in which our ~ wild and cultivated plants have been singled out for use. We shall, in the case of each class, allude to the methods by which the selected plants have been improved, or their products fully utilized. Thus looking the ground over, although not minutely, we can see what new plants are likely to be added to our list. Our illustrations can, at the best, be only fragmen- tary. We shall not have time to treat the different divisions of the subject in precisely the proportions: which would be de- manded by an exhaustive essay ; an address on an occasion like this must pass lightly over some matters which other oppor- tunities for discussion could properly examine with great ful- ness. Unfortunately, some of the minor topics which must be thus passed by, possess considerable popular interest; one of these is the first subordinate question introductory to our task, namely, how were our useful cultivated and wild plants se- lected for use ? A study of the early history of plants employed for cere- monial purposes, in religious solemnities, in incantations, and for medicinal uses, shows how slender has sometimes been the claim of certain plants to the possession of any real utility. But some of the plants which have been brought to notice in these ways have afterwards been found to be utilizable in some _ fashion or other. This is often seen in the cases of the plants which have been suggested for medicinal use through the absurd doctrine of signatures.° It seems clear that, except in modern times, useful plants have been selected almost wholly by chance, and it may well be said that a selection by accident is no selection at all. Now- adays, the new selections are based on analogy. One of the most striking illustrations of the modern method is afforded by the utilization of bamboo fiber for electric lamps. Some of the classes of useful plants must be passed by with- out present discussion; others alluded to slightly, while still other groups fairly representative of selection and improve- ment will be more fully described. In this latter class would naturally come, of course, the food-plants known as I. Tur CEREALS. Let us look first at these. The species of grasses which yield these seed-like fruits, or as we might call them for our purpose seeds, are numerous ;° twenty of them are cultivated largely in the Old World, but a ea ae ee G. L. Goodale— Possibilities of Economic Botany. 277 only six of them are likely to be very familiar to you, namely, wheat, rice, barley, oats, rye and maize. The last of these is of American origin, despite doubts which have been cast upon it. Tt was not known in the Old World until after the discovery of the New. It has probably been very long in cultivation. The others all belong to the Old World. Wheat and barley have been cultivated from the earliest times; according to De Candolle, the chief authority in these matters, about four thousand years. Later came rye and oats, both of which have been known in cultivation for at least two thousand years. Even the shorter of these periods gives time enough for wide variation, and as is to be expected there are numerous varieties - of themall. For instance, Vilmorin, in 1880, figured sixty-six varieties of wheat with plainly distinguishable characters.” If the Chinese records are to be trusted, rice has been culti- vated for a period much longer than that assigned by our history and traditions to the other cereals, and the varieties are correspondingly numerous. It is said that in Japan above three hundred varieties are grown on irrigated lands, and more than one hundred on uplands.* With the possible exception of rice, not one of the species of cereals is certainly known in the wild state.* Now and then specimens have been gathered in the East which can be re- ferred to the probable types from which our varieties have sprung, but doubt has been thrown upon everyone of these eases. It has been shown conclusively that it is easy fora plant to escape from cultivation and persist in its new home even for a long time in a near approximation to cultivated form. Hence, we are forced to receive all statements regarding the wild forms with caution. But it may be safely said that if all the varieties of cereals which we now cultivate were to be swept out of existence, we could hardly know where to turn for wild species with which to begin again. We could not know with certainty. To bring this fact a little more vividly to our minds, let us suppose a case. Let us imagine that a blight without parallel has brought to extinction all the forms of wheat, rice, rye, oats, barley and maize, now in cultivation, but without affect- ing the other grasses or any other form of vegetable food. Mankind would be obliged to subsist upon the other kindly fruits of the earth; upon root-crops, tubers, leguminous seeds, and soon. Some of the substitutions might be amusing in any other time than that of a threatened famine. Others would be far from appetizing under any condition, and only a few would be wholly satisfying even to the most pronounced vegetarian. In short, it would seem, from the first, that the cereals fill a place occupied by no other plants. The composition of the grains 278 G. L. Goodale—- Possibilities of Economie Botany. is theoretically and practically almost perfect as regards food ratio between the nitrogenous matters and the starch group ; and the food value, as it is termed, is high. But aside from these considerations, it would be seen that for safety of preser- vation through considerable periods, and for convenience of transportation, the cereals take highest rank. Pressure would come from every side to compel us to find equivalents for the lost grains. From this predicament I believe that the well- equipped Experiment Stations and the Agricultural Depart- ments in Europe and America would by and by extricate us. Continuing this hypothetical case, let us next inquire how the Stations would probably go to work in the up-hill task of making partially good a well-nigh irreparable loss. The whole group of relatives of the lost cereals would be passed in strict review. Size of grain, strength and vigor and plasticity of stock, adaptability to different surroundings, and flexibility in variation would be examined with scrupulous care. But the range of experiment would, under the circum- stances, extend far beyond the relatives of our present cereals. It would embrace an examination of the other grasses which are even now cultivated for their grains, but which are so little known outside of their own limit, that it is a surprise to hear about them. For example, the Millets, great and small; would be investigated. ‘These grains, so little known here, form an important crop in certain parts of the east. One of the leading authorities on the subject* states that the Millets constitute ‘a more important crop” in India “than either Rice or Wheat, and are grown more extensively, being raised from Madras in the south to Rajputana in the north. They occupy about eighty- three per cent of the food-grain area in Bombay and Sinde, forty-one per cent in the Punjab, thirty-nine per cent in the Central Provinces,” ‘in all about thirty million acres.” Having chosen proper subjects for experimenting, the culti- _ vators would make use of certain well-known principles. By simple selection of the more desirable seeds, strains would be secured to suit definite wants, and these strains would be kept as races, or attempts would be made to intensify wished-for characters. By skillful hybridizing of the first, second and higher orders, tendencies to wider variation would be obtained and the process of selection considerably expedited.’ It is out of our power to predict how much time would elapse before satisfactory substitutes for our cereals could be found. In the improvement of the grains of grasses other than those which have been very long under cultivation, experi- ments have been few, scattered and indecisive. Therefore we are as badly off for time-ratios as are the geologists and archeologists, in their statements of elapsed periods. It is G. L. Goodale— Possibilities of Economic Botany. 279 impossible for us to ignore the fact that there appear to be ’ occasions in the life of a species when it seems to be peculiarly susceptible to the influences of its surroundings.’* A species, like a carefully laden ship, represents a balancing of forces within and without. Disturbance may come through variation from within, as from a shifting of the cargo, or, in some cases from without. We may suppose both forces to be active in producing variation, a change in the internal condition render- ing the plant more susceptible to any change in its surround- ings. Under the influence of any marked disturbance, a state of unstable equilibrium may be brought about, at which times the species as such is easily acted upon by very slight agencies. One of the most marked of these derangements is a conse- quent of cross-breeding within the extreme limits of varieties. The resultant forms in such cases can persist only by close breeding or by propagation from buds or the equivalents of buds. Disturbances like these arise unexpectedly in the ordi- nary course of nature, giving us sports of various kinds. These critical periods however, are not unwelcome, since skill- ful cultivators can take advantage of them. In this very field much has been accomplished. An attentive study of the sagacious work done by Thomas Andrew Knight shows to what extent this can be done.” But we must confess that it would be absolutely impossible to predict with certainty how long or how short would be the time before new cereals or acceptable equivalents for them would be provided. Upheld by the confidence which I have in the intelligence, ingenuity, and energy of our Experiment Stations, I may say that the time would not probably exceed that of two generations of our race, or half a century. In now laying aside our hypothetical illustration, I venture to ask why it is that our Experiment Stations and other insti- tutions dealing with plants and their improvement, do not undertake -investigations like those which I have sketched ? Why are not some of the grasses other than our present cereals studied with reference to their adoption as food grains ? One of these species will naturally suggest itself to you all, namely, the Wild Rice of the Lakes.” Observations have shown that, were it not for the difficulty of harvesting these grains which fall too easily when they are ripe, they might be utilized. But attentive search might find or educe some variety of Zezania, with a more persistent grain and a better yield. There are two of our sea-shore grasses which have excellent grains, but are of small yield. Why are not these, or better ones which might be suggested by observation, taken in hand ? 280 G. L. Goodale—Possibilities of Economic Botany. The reason is plain. We are all content to move along in lines of least resistance, and are disinclined to make a fresh start. It is merely leaving well enough alone, and so far as the cereals are concerned it is indeed well enough. The generous grains of modern varieties of wheat and barley com- pared with the well-preserved charred vestiges found in Greece by Schliemann,” and in the lake-dwellings,” are satisfactory in every respect. Improvements, however, are making in many directions; and in the cereals we now have, we possess far better and more satisfactory material for further improvement both in quality and as regards range of distribution than we could reasonably hope to have from other grasses. From the cereals we may turn to the interesting groups of plants comprised under the general term II. VEGETABLES. Under this term it will be convenient for us to include all plants which are employed for culinary purposes, or for table use such as salads and relishes. : The potato and sweet potato, the pumpkin and squash, the red or capsicum peppers, and the tomato, are of American origin. All the others are, most probably, natives of the Old World. Only one plant coming in this class has been derived from Southern Australasia, namely, New Zealand Spinach, (Zetra- ond. : Among the vegetables and salad-plants longest in cultivation we may enumerate the following—turnip, onion, cabbage, purslane, the large bean (Taba), chick-pea, lentil and one species of pea, garden pea. To these an antiquity of at least four thousand years is ascribed. Next to these, in point of age, come the radish, carrot, beet, garlic, garden-cress, and celery, lettuce, asparagus and the leek. Three or four leguminous seeds are to be placed in the same category, as are also the black peppers. Of more recent introduction the most prominent are, the parsnip, oyster plant, parsley, artichoke, endive and spinach. From these lists I have purposely omitted a few which belong exclusively to the tropics, such as certain yams. The number of varieties of these vegetables is astounding. It is, of course, impossible to discriminate between closely allied varieties which have been introduced by gardeners and seedsmen under different names, but which are essentially identical, and we must therefore have recourse to a conserva- tive authority, Vilmorin,’* from whose work a few examples have been selected. The varieties which he accepts are sut- G. L. Goodale—Possibilities of Economic Botany. 281 ficiently well distinguished to admit of description and in most instances of delineation, without any danger of confusion. The potato has, he says, innumerable varieties, of which he accepts forty as easily distinguishable and worthy of a place in a general list, but he adds also a list, comprising, of course, synonyms, of thirty-two French, twenty-six English, nineteen American and eighteen German varieties. The following numbers speak for themselves, all being selected in the same careful manner as those of the potato: celery more than twenty; carrot more than thirty; beet, radish and potato more than forty; lettuce and onion more than fifty; turnip more than seventy ; cabbage, kidney bean and garden pea more than one hundred. The amount of horticultural work which these numbers represent is enormous. Each variety established as a race (that is a variety which comes true to seed) has been evolved by the same sort of patient care and waiting which we have seen is necessary in the case of cereals. but the time of wait- ing has not been as a general thing so long. You will permit me to quote from Vilmorin” also an account of a common plant, which will show how wide is the range of variation and how obscure are the indications in the wild plant of its available possibilities. The example shows how com- pletely hidden are the potential variations useful to mankind. “ Cabbage, a plant which is indigenous in Europe and Western Asia, is one of the vegetables which has been cultivated from the earliest time. The ancients were well acquainted with it, and certainly possessed several varieties of the head-forming kinds. _ The great antiquity of its culture may be inferred from the im- mense number of varieties which are now in existence, and from the very important modifications which have been produced in the characteristics in the original or parent plant. The wild Cabbage, such as it now exists on the coasts of England and France, is a perennial plant with broad-lobed, undu- lated, thick, smooth leaves, covered with a glaucous bloom. The stem attains a height of from nearly two and a half to over three feet, and bears at the top a spike of yellow or sometimes white flowers. All the cultivated varieties present the same peculiarities in their inflorescence, but up to the time of flowering they exhibit most marked differences from each other and from the original wild plant. In most of the Cabbages, it is chiefly the leaves that are developed by cultivation; these for the most part become imbricated or overlap one another closely, so as to form a more or less compact head, the heart or interior of which is composed of the central undeveloped shoot and the younger leaves next it. The shape of the head is spherical, sometimes flattened, sometimes conical, All the varieties which form heads in this way are known by the general name of Cabbages, while other kinds with 282 G. L. Goodale—Possibilities of Esonomio Botany. large branching leaves which never form heads are distinguished by the name of Borecole or Kale. In some kinds, the flower-stems have been so moditied by cul- ture as to become transformed into a thick, fleshy tender mass, the growth and enlargement of which are produced at the expense of the flowers which are absorbed and rendered abortive. Such are the Broccolis and Cauliflowers.” But this plant has other transformations. “Tn other kinds, the leaves retain their ordinary dimensions, while the stem or principal root has been brought by cultivation to assume the shape of a large ball or turnip, as in the case of the plants known as Kohl-Rabi and Turnip-rooted Cabbage or Swedish Turnip. And lastly, there are varieties in which cultivation and selection have produced modifications in the ribs of the leaves, as in the Couve Tronchuda, or in the axillary shoots (as in Brussels sprouts), or in several organs together, as in the Marrow Kales, and the Neapolitan Curled Kale.” Here are important morphological changes like those to which Professor Bailey has called attention in the case of the tomato. Suppose we are strolling along the beach at some of the sea- side resorts of France, and should fall im with this coarse eru- ciferous plant, with its sprawling leaves and strong odor. Would there be anything in its appearance to lead us to search for its hidden merit as a food plant? What could we see in it which would give it a preference over a score of other plants at our feet? Again, suppose we are journeying in the high lands of Peru, and should meet with a strong-smelling plant of the Night-shade family, bearing a small irregular fruit, of sub- acid taste and of peculiar flavor. We will further imagine that the peculiar taste strikes our fancy, and we conceive that the plant has possibilities as a source of food. We should be led by our knowledge of the potato, probably a native of the same region, to think that this allied plant might be safely transferred to a northern climate, but would there be promise of enough future usefulness in such a case as this, to warrant our carrying the plant North as an article of food? Suppose, further, we should ascertain that the fruit in question was relished not only by the natives of its home, but that it had found favor among the tribes of South Mexico and Central America, and had been cultivated by them until it had attained a large size; should we be strengthened in our venture? Let us go one step further still. Suppose that having decided upon the introduction of the plant, and having urged everybody to try it, we should find it discarded as a fruit, but taking a place in gardens as a curiosity under an absurd name, or as a basis G. L. Goodale—Possibilities of Economie Botany. 283 for preserves and pickles; should we not look upon our experi- ment in the introduction of this new plant as a failure? This is not a hypothetical case. The Tomato,” the plant in question, was cultivated in Europe as long ago as 1554;”° it was known in Virginia in 1781 and in the Northern States in 1785; but it found its way into favor slowly, even in this land of its origin. A credible witness states that in Salem it was almost impossible to induce people to eat or even taste of the fruit. And yet, as you are well aware, its present cultivation on an enormous scale in Europe and this country is scarcely sufficient to meet the increasing demand. A plant which belongs to the family of the tomato has been known to the public under the name of the strawberry tomato. The juicy yellow or orange-colored fruit is enclosed in a papery calyx of large size. The descriptions which were published when the plant was placed on the market were attractive, and were not exaggerated to a misleading extent. But, as you all know, the plant never gained any popularity. If we. look at these two cases carefully we shall see that what appears to be caprice on the part of the public is at bottom common sense. The cases illustrate as well as any which are at command, the diticulties which surround the whole subject of the introduc- tion of new foods. Before asking specifically in what direction we shall look for new vegetables I must be pardoned for calling attention, in passing, to a very few of the many which are already in limited use in Kurope and this country, but which merit a wider em- ployment. Cardon, or Cardoon; Celeriac, or turnip-rooted celery; Fetticus, or corn-salad; Martynia; Salsify; Sea-kale ; and numerous small salads, are examples of neglected treasures of the vegetable garden. The following which are even less known may be mentioned as fairly promising.” (1) Arracacia esculenta, called Arracacha, belonging to the Parsley family. It is extensively cultivated in some of the northern States of South America. The stems are swollen near the base, and produce tuberous enlargements filled with an excellent starch. Although the plant is of comparatively easy cultivation, efforts to introduce it into Europe have not been successful, but it is said to have found favor in both the Indies, and may prove useful in our Southern States. (2) Ollucus or Ollucus, another tuberous-rooted plant from nearly the same region, but belonging to the Beet or Spinach | family. It has produced tubers of good size in England, but they are too waxy in consistence to dispute the place of the better tubers of the potato. The plant is worth investigating for our hot dry lands. ! 284 G. L. Goodale—Possibilities of Heonomic Botany. (83) A tuber-bearing relative of our common Hedge-nettle, or Stachys, is now cultivated on'a large scale at Crosnes in France, for the Paris market. Its name in Paris is taken from the locality where it is now grown for use. Although its native country is Japan, it is called by some seedsmen Chinese Artichoke. At the present stage of cultivation, the tubers are small and are rather hard to keep, but it is thought “that both of these defects can be overcome or evaded.” Experiments indicate that we have in this species a valuable addition to our vegetables. We must next look at certain other neglected possibilities. Dr. Edward Palmer,” whose energy as a collector and acute- ness as an observer are known to you all, has brought together very interesting facts relative to the food-plants of our North American aborigines. Among the plants described by him there are a few which merit careful investigation. Against all of them, however, there lie the objections mentioned before, namely : (1) The long time required for their improvement, and (2) The difficulty of making them acceptable to the commu- nity, involving (3) The risk of total and mortifying failure. In the notes to this address the more prominent of these are enumerated. In 1854 the late Professor Gray called attention to the re- markable relations which exist between the plants of Japan and those of our Eastern coast. You will remember that he not only proved that the plants of the two regions had a common origin, but also emphasized the fact that many species of the two countries are almost identical. It is to that country which has yielded us so many useful and beautiful plants that we turn for new vegetables to supplement our present food-resources. One of these plants, namely, Stachys, has already been men- tioned as rather promising. There are others which are worth examination and perhaps acquisition. One of the most convenient places for a preliminary exami- nation of the vegetables of Japan is at the railroad stations on the longer lines, for instance, that running from Tokio to Kobe. For native consumption there are prepared luncheon boxes of two or three stories, provided with the simple and yet embar- rassing chop-sticks. It is worth the shock it causes one’s nerves to invest in these boxes and try the vegetable contents. The bits of fish, flesh and fowl which one finds therein can be easily separated and discarded, upon which there will remain a few delicacies. The pervading odor of the box is that of aromatic vinegar. The generous portion of boiled rice is of excellent quality with every grain weil softened and distinct, and this G. L. Goodale— Possibilities of Economie Botany. 285 without anything else would suffice for a tolerable meal. In the boxes which have fallen under my observation there were sundry boiled roots, shoots and seeds which were not recog- nizable by me in their cooked form. Professor Georgeson,” formerly of Japan, has kindly identified some of these for me, but he says ‘“ there are doubtless many others used occasionally.” One may find sliced Lotus roots, roots of large Burdock, Lily bulbs, shoots of Ginger, pickled green Plums, beans of many sorts, boiled Chestnuts, nuts of the Gingko tree, pickled greens of various kinds, dried cucumbers, and several kinds of seaweeds. Some of the leaves and roots are cooked in much the same manner as beet-roots and beet-leaves are by us, and the general effect is not unappetizing. The boiled shoots are suggestive of only the tougher ends of asparagus. On the whole, I do not look back on Japanese railway luncheons with any longing which would compel me to advocate the indis- criminate introduction of the constituent vegetables here. But when the same vegetables are served in native inns, under more favorable culinary conditions, without the flavor of vinegar and of the pine wood of the luncheon boxes, they appear to be worthy of a trial in our horticulture, and I there- fore deal with one or two in greater detail. Professor Georgeson, whose advantages for acquiring a knowledge of the useful plants of Japan have been unusually good, has placed me under great obligations by communicating certain facts regarding some of the more promising plants of Japan which are not now used here. It should be said that several of these plants have already attracted the notice of the Agricultural Department in this country. ’ The Soy Bean (Glycine hispida). This species is known here to some extent, but we do not have the early and best varieties. ‘These beans replace meat in the diet of the common eople. : Mucuna (Mucuna capitata) and Dolichos (Dolichos cultra- tus) are pole beans possessing merit. Dioscoreu ; there are several varieties with palatable roots. Years ago one of these was spoken of by the late Dr. Gray, as possessing “‘excellent roots, if one could only dig them.” Colocasia antiquorum has tuberous roots, which are nutri- tious. Conophallus Konjak has a large bulbous root, which is sliced, dried and beaten to a powder. It is an ingredient in cakes. Aralia cordata is cultivated for the shoots, and used as we use Asparagus. @nanthe stolonifera and Cryptotenia Canadensis are pala- table salad plants, the former being used also as greens. 286 G. L. Goodale—Possibilities of Heonomic Botany. There is little hope, if any, that we shall obtain from the hotter climates for our southern territory new species, of merit. The native markets in the tropical cities, like Colombo, Batavia, Singapore and Saigon, are rich in fruits, but outside of the native plants bearing these, nearly all the plants appear to be wholly in established lines of cultivation, such, for instance, as members of the Gourd and Night-shade families. | Before we leave the subject of our coming vegetables, it will be well to note a naive-caution enjoined by Vilmorin in his work, Les Plantes Potageres.™ “Finally,” he says, “we conclude the article devoted to each plant with a few remarks on the uses to which it may be ap- plied and on the parts of the plants which are to be so used. In many cases such remarks may be looked upon as idle words, and yet it would sometimes have been useful to have them when new plants were cultivated by us for the first time. For instance, the giant edible Burdock of Japan (Zappa edulis) was for a long time served up on our tables only as a wretchedly poor Spinach, because people would cook the leaves, whereas, in its native country, it is only cultivated for its tender fleshy roots.” I trust you are not discouraged at this outlook for our coming vegetables. Two groups of improvable food-plants may be referred to before we pass to the next class, namely, edible fungi and the beverage plants. All botanists who have given attention to the matter agree with the late Dr. Curtis of North Carolina that we have in the unutilized mushrooms an immense amount of available nutriment of a delicious quality. It is not improbable that other fungi than our common “edible mushroom” will by and by be subjected to careful selection. The principal beverage-plants, Tea, Coffee and Chocolate, are all attracting the assiduous attention of cultivators. The first of these plants is extending its range at.a marvellous rate of rapidity through India and Ceylon; the second is threatened by the pests which have almost exterminated it in Ceylon, but anew species, with crosses therefrom, is promising to resist them successfully ; the third, Chocolate, is every year passing into lands farther from its original home. To these have been added the Kola, of a value as yet not wholly determined, and others are to augment the short list. Ill. Frorirs. Botanically speaking, the cereal grains of which we have spoken, are true fruits, that is to say, are ripened ovaries, but for all practical purposes they may be regarded as seeds. The G. L. Goodale— Possibitities of Economic Botany. 287 fruits, of which mention is now to be made, are those com- monly spoken of in our markets, as fruits. First of all, attention must be called to the extraordinary changes in the commercial relations of fruits by two direct causes, (1) The canning industry, and (2) Swift transportatiou by steamers and railroads. The effects of these two agencies are too well known to require more than this passing mention. By them the fruits of the best fruit-growing countries are carried to distant lands in quantities which surprise all who see the statistics for the first time. The ratio of increase is very startling. Take for instance, the figures given by Mr. Morris at the time of the great Colonial and Indian Exhibition, in London. Compare double decades of years. | 1845, £886,888. 1865, £3,185,984. - 1885, £7,587,523. In the Colonial Exhibition at London, in 1886, fruits from the remote colonies were exhibited under conditions which proved that, before long, it may be possible to place such delicacies as the Cherimoyer, the Sweet-cup, Sweet-sop, Ram- butan, Mango and Mangosteen, at even our most northern sea- ports. Furthermore, it seems to me likely that with an in- crease in our knowledge with regard to the microbes which produce decay, we may be able to protect the delicate fruits from injury for any reasonable period. Methods which will supplement refrigeration are sure to come in the very near future, so that even in a country so vast as our own, the most perishable fruits will be transported through its length and breadth without harm. The canning industry and swift transportation are likely to diminish zeal in searching for new fruits, since, as we have seen in the case of the cereals, we are prone to move in lines of least resistance and leave well enough alone. To what extent are our present fruits likely to be improved ? Even those who have watched the improvement in the quality" of some of our fruits, like oranges, can hardly realize how great has been the improvement within historic times in the character of certain pears, apples, and so on. The term historic is used advisedly, for there are pre-historic fruits which might serve as a point of departure in the consid- eration of the question. In the ruins of the lake-dwellings in Switzerland,” charred apples have been found, which are Am. Jour. Sci.—THIRD SERIES, Vou. XLIJ, No. 250.—Ocroper, 1891. 20 288 G. L. Goodale—Possibilities of Heonomic Botany. in some cases, plainly of small size, hardly equalling ordinary crab apples. But, as Dr. Sturtevant has shown, in certain directions, there has been no marked change of type, the change is in quality. In comparing the earlier descriptions of fruits with modern accounts it is well to remember that the high standards by which fruits are now judged are of recent establishment. Fruits which would once have been esteemed excellent, would to-day be passed by as unworthy of regard It seems probable that the list of seedless fruits will be materially lengthened, provided our experimental horticultur- ists make use of the material at their command. The com- mon fruits which have very few or no seeds are the banana, pineapple and certain oranges. Others mentioned by Mr. Darwin as well knuwn are the bread-fruit, pomegranate, azarole or Neapolitan medlar, and date palms. In commenting upon these fruits, Mr. Darwin” says that most horticulturists “look at the great size and anomalous development of the fruit as the cause and sterility as the result,’ but he holds the opposite view as more probable, that is, that the sterility, com- ing about gradually, leaves free for other growth the abundant supply of building material which the forming seed would otherwise have. He admits, however, that “there is an antag- onism between the two forms of reproduction, by seeds and by buds when either is carried to an extreme degree which is independent of any incipient sterility.” Most plant-hybrids are relatively infertile, but by no means wholly sterile. With this sterility there is generally aug- mented vegetative vigor, as shown by Nageli. Partial or com- plete sterility and corresponding luxuriance of root, stem, leaves and flower, may come about in other obscure ways, and such cases are familiar to botanists." Now it seems highly probable that either by hybridizing directed to this special end, or by careful selection of forms indicating this tendency to the correlated changes, we may succeed in obtaining impor- tant additions to our seedless or nearly seedless plants. Whether the ultimate profit would be large enough to pay for the time and labor involved is a question which we need not enter into; there appears to me no reasonable doubt that such efforts would be successful. There is no reason in the nature of things why we should not have strawberries without the so-called seeds; blackberries and raspberries, with only deli- cious pulp; and large grapes as free from seeds as the small ones which we call “currants”? but which are really grapes from Corinth. These and the coreless apples and pears of the future, the stoneless cherries and plums, like the common fruits before G. L. Goodale—Possibilities of Economic Botany. 289 mentioned must be propagated by bud division, and be open to the tendency to diminished strength said to be the conse- quence of continued bud-propagation. But this bridge need not be crossed until we come to it. Bananas have been per- petuated in this way for many centuries, and pineapples since the discovery of America, so that the borrowed trouble alluded to is not threatening. First we must catch our seedless fruits. Which of our wild fruits are promising subjects for selec- tion and cultivation ? _ Mr. Crozier of Michigan has pointed out* the direction in which this research may prove most profitable. He enumer- ates many of our small fruits and nuts which can be improved. Another of our most careful and successful horticulturists believes that the common blueberry and its allies are very suitable for this purpose and offer good material for experi- menting. The sugar-plum, or so-called shadbush, has been im- proved in many particulars, and others can be added to this list. But again we turn very naturally to Japan, the country from which our gardens have received many treasures. Referring once more to Professor Georgeson’s studies,” we must mention the varieties of Japanese apples, pears, peaches, plums, cherries and persimmons. ‘The persimmons are already well-known in some parts of our country, under the name “kaki” and they will doubtless make rapid progress in popular favor. The following are less familiar: Animals and Plants under Domestication (Am. Edn.), vol. ii, p. 205-209. 26 American Garden, N. Y. 1890-91. 27 American Garden, N. Y. 1891. 28 Useful Native Plants of Australia, by J. H. MAIDEN, Sydney. *9 The Flowers of Japan and the Art of Floral Arrangement. By JosiaAH Con- DER, F.R.I.B.A., Architect to the Imperial Japanese Government. Yokohama, 1891. See also two other works by the same author: Theory of Japanese Flower- arrangements, and Art of Landscape-gardening in Japan. (1886.) 30 Mr. SAMUEL DIxoN’s list is in vol. viii (for 1884-85) of the Transactions and Proceedings and Report of the Royal Society of South Australia. Adelaide, G. Rob- ertson, 1886. Bursaria spinosa, ‘a good stand-by,” after the grasses dry up. Pomaderris racemosa, ‘‘ stands stocking well.” Pittosporum phyllaeroides, *‘ sheep exceedingly partial to its foliage.”’ Casuarina quadrivalvis, ‘‘tenderness of fiber, wool would be represented by it in our finer wool districts.” Acacias, The Wattles. ‘ Value as an astringent, very great,” being curative of a malady often caused by eating frozen grass. Acacia aneura (mulga). ‘‘Must be very nutritious to all animals eating it.” This is the plant which is such a terror to the stockmen who have to ride through the ‘‘ scrub.” Cassia, some of the species with good pods and leaves for sheep. The foregoing are found in districts which are not wholly arid. The following are, more properly, ‘‘ dry” plants. Sida petrophila, ‘‘as much liked by sheep as by marsupials.” Dodonaea viscosa, Native Hop-bush. ‘“ Likes warm, red, sandy ground,” Lycium australe, ‘‘ Drought never seems to affect it.” Kochia aphylla: ‘‘ All kinds of stock are often largely dependent on it during protracted droughts.” Rhagodia parabolica: ‘‘ Produces a good deal of foliage.” Atriplex vesicaria: ‘‘Can be readily grown wherever the climate is not too wet,” I have transferred only those which Mr. Dixon thinks most worthy of trial. Compare also Dr. VAsey’s valuable studies of the plants of our dry lands, espec- ially Grasses and Forage plants (1878), Grasses of the arid districts of Kansas, Nebraska, and Colorado (1886), Grasses of the South (1887. 31 The weeds of German gardens and agricultural lands are mostly from Medi- terranean regions, but the invasions in the uncultivated districts are chiefly from - America, (such as Oenothera, Mimulus, Rudbeckia). Handbuch der Pflanzengeo- graphie, von Dr. OSCAR DRUDE, (Stuttgart), 1890, p. 97. 33 The list of economic plants published by the Department in Washington is remarkably full, and is in every way creditable to those in charge. Am. Jour. Sor1.—THIkD SERIES, Vou. XLII, No. 250-—OctTossrr, 1891. 21 304 T. Holm— Vitality of some Annual Plants. Art. XX VIII.—On the Vitality of some Annual Plants ; by TuHEo. Hotm. (With Plate X.) THE curious fact, that there may be individuals of annual species of which the life-time is not limited merely to one year, has been noticed by several botanists. Exceptions of that kind often seem to be nearly accidental, but in most cases they are, however, to be considered as due to certain external fac- tors, as for instance climate and soil or cultivation. Indeed, the number is not very small of the species in which a fluctua- tion has been observed from being annual to perennial or at least biennial in a modified sense, as well as of those for which similar intergradation-forms have been recorded between the biennial on the one side and the annual and perennial ones on the other. . Irmisch** mentions for instance, that Echinospermum Lap- pula, which is usually biennial may occur as annual, having al- ready developed the flowers in the first year; he observed that the same is true in the case of Hyoscyamus niger, of which even the annual form has been described as a proper species (H. agrestis) since it is very different in habit from the biennial type. The same author has also observed, that Hypericum humifusum and MMalva neglecta may occur as both annual and perennial.’ Sedum annwum, which has been described as annual by Hartman, Areschoug and Blytt, was observed by Warmingt to be biennial, and even that this seemed to be the normal for this plant. Hildebrandt enumerates several species, which, although they occur under widely different conditions in both hemispheres, nevertheless seem to be constant as to their life-duration and habit; such species for example are the annual Polygonum aviculare, Erigeron Canadensis, Papaver Rheas, ete., while of perennials Zhymus serpyllum, Verbena officinalis, Urtica dioica, ete. He enumerates on the other hand several other species, which show a tendency to vary from annual to bien- nial, among which are many Crucifere; Composite and culti- vated Graminec, besides some usually biennial species, which may occur as perennials, as for instance some Crucifere, Um- bellifere, Papilionacec, etc. Similar aberrant forms’ are eyi- dently far from rare in this country, though the author has not * Thilo Irmisch: Zur Morphologie der monocot. Knollen und Zwiebelge- wachse, 1850, p. 211. | Eug. Warming: Om Skudbygning, Overvintring og Foryngelse. (Naturhist. Forenings Festskrift, 1884, p. 16.) + Fr. Hildebrand: Die Lebensdauer und Vegetationsweise der Pflanzen, ihre Ursachen und ihre Entwickelung. (Engler’s Botanische Jahrbicher, vol. ii, 1881, p. 51.) a eee eee TF. Holm— Vitality of some Annual Plants. 305 succeeded in finding any special observations in the literature, and as he has had the opportunity of observing a few cases of that kind, they seem to likely to be of some interest, at least locally. eee nudicaule Walt. (1. Sarothra Michx.) is un- doubtedly typically annual, but a few individuals were, how- ever, collected late in the fall, which seemed to prove an exception. The base of one of these specimens is figured on plate X, fig. 1, and we see here two densely leaved branches proceeding from the lower part of the stem, and probably developed in the axils of the cotyledons. These shoots with numerous imbricate leaves might be supposed to be able to winter over and in the following year to give rise to flowers; the root-system in these individuals was unusually strong, show- ing not only a primary root, but also a few, and rather strong, lateral ones. a Another example is Delphinium consolida L., which as far as known to the author has not been recorded as otherwise than annual in Europe and here. The plant is rather rare in the vicinity of Washington, but occurs as escaped from cultivation on the banks of the Potomac. It was observed here to vary from annual to perennial, although both forms were growing together, and the perennial form showed a strong, persistent primary root and several flowering stems, besides a cluster of buds for the following year. This kind of variation was also observed in Cyperus jfiaves- cens L., a plant, which is rather common in wet places around Washington, and I have sometimes met with individuals, which certainly seemed to be perennial. One of these is fig- -ured on plate X, fig. 6, and shows in contrast to the annual type a distinct rhizome with creeping stolons, from the nodes of which several strong roots proceed. It must be admitted that this specimen shows the general aspect of a perennial, stoloniferous Cyperus, able to give rise to new individuals by a vegetative propagation. Whether this species has been recorded as perennial also in other countries, I do not know, but Lange* mentions, however, the fact that he has collected the plant in France, where some specimens showed “tuberous stolons,’ and therefore questions its character as annual. Grenier and Godront have, nevertheless, described the plant as merely annual. It might be mentioned here, that a similar variation also exists in Carex cyperoides L., which under normal conditions is truly annual; Lange states (1. ¢. p. 118) that the periodical disappearance of this plant in several parts of Europe has been explained by the fact, that it is able to veg- * Joh. Lange: Haandbog i den danske Flora, 1886-88, p. 116. + Grenier et Godron: Flore de France, vol. iii, 1855. 306 T. Holm— Vitality of some Annual Plants. etate for several years without flowering, if the locality becomes inundated.* It seems then, that this ability to vary among Cyperacee is easy to explain, although it may not be a necessary consequence in all cases; our specimens of Cyperus were observed as occur- ring together with numerous annual individuals and under the _ very same conditions. The Graminee may undoubtedly include several other species, besides the cultivated ones, mentioned by Hildebrand (I. ec.) in which this same variation may take place. Zragus race- mosus Hall, represents at least the same case as that of Cyperus. It has always been considered as annual, lately by Dr. Vasey,t although some specimens in U. 8. National herbarium prove, that it can also occur as perennial. These specimens have long stolons above-ground with abundant formation of leafy shoots at each node, from where long roots are also developed. A part of a stolon has been illustrated in plate X, fig. 2, where we see not only the crowded shoots, but also a secon- dary formation stolons (S). All these shoots were developed in the axils of leaves, belonging to the stolons, and they showed besides the proper leaves also the characteristic pro- phyllum (fig. 3). This small leaf had a rather unusual shape than otherwise observed ; it was not only distinctly bicarinate, but showed at its apex two long teeth, corresponding to those of an ordinary palet (fig. 4). A transverse section (fig. 5) of the prophyllum shows the prominent keels and the very thin margins, besides the presence of not only two, but even six nerves, those of the keels being the strongest. This manner of growth seems to be rather common in North American Gramine, especially those, which ramity, « and there is no doubt that under favorable conditions they might change their habit from annual to perennial or at least biennial. ) 7 The Crucifere, which include representatives of all the three types of growth, are, as mentioned above, quite apt to vary in a similar manner. The genus Avadis is very instructive in this regard, as for instance A. dentata Torr. et Gr., of which the lower part of a perennial specimen has been figured on Plate X, fig. 7, which shows the character of a biennial in a perennial form. We see here a leafy shoot terminating the main axis, from the leaves of which flowering stems will be developed next year. We see further, that the base of this shoot is surrounded by now faded leaves, merely indicated by the petioles, and from the axils of which proceed the ascend- ing flowering stems of the year. * Compare: Bull. de la Société botan. de France, 1860, p, 186. + Geo. Vasey: Grasses of the Southwest, Part I, 1890. (Bull. of U.S, Dept. of Agriculture). a T. Holm— Vitality of some Annual Plants. 307 The plant illustrates then at once the two stages of a bien- nial growth: the leafy roset, which will winter over, and the flowering stems from a similar roset of the preceding year. The age of this specimen is at least three years, for there is a distinct stem-part (S’) to be seen between the primary root and the now faded roset of leaves. This part of the stem is in contrast to the upper part which is horizontal and fixed to the ground by strong roots in addition to the persistent primary one. The plant has undoubtedly merely developed a leafy roset in its first year, when the seed germinated, probably flowered the second year and produced the stem-part S’, flow- ered again in the third year, producing contemporarily a shoot that will flower next year, until finally an inflorescence will terminate the main axis and the entire individual die away after the ripening of the fruits. The closely allied A. lyrata L. has been mentioned by Hil- debrand (1. c.) as being annual or biennial. It occurs, however, also as perennial, and most commonly so in the Southern United States. JI have collected several specimens in the vicinity of Washington, which were all decidedly perennial, and most of the individuals in the National Herbarium from other parts of North America showed the same fact. Some specimens from Japan were, however, annual, with flowers and leaves devel- oped in the first year and at the same time. The perennial form shows nearly the same habit as above described for A. dentata, but commonly with a profuse development of lateral leafy shoots together with the inflorescences, so that the life is secured for a considerably longer time than in the pre-, ceding species. That also Arabis lewigata Poir. may occur as perennial has “been recorded by Hildebrand (1. c.), but without data; Gray has considered this species as truly biennial, in which form it occurs exclusively around Washington, where it is very abun- dant. Washington, D. C., July 22d, 1891. EXPLANATION OF PLATE X. Fig. 1.—AHypericum nudicaule. The base of the stem, showing two densely leaved shoots. Natural size. Fig. 2.—Tragus racemosus. Part of a stolon A—A, with secondary branches S and several crowded shoots. Natural size. Fig. 3.—Same. Part of a stolon, showing two leafy shoots in the axils of two leaves, belonging to the stolon. P, the prophyllum. Slightly enlarged. Fie. 4.—Same. The prophyllum; a, side, and b, back view. Fig. 5.--Same. Transverse section of the prophyllum. Fig. 6.—Cyperus flavescens. The base of a perennial specimen, showing the stolons. F, the base of the flowering stem. Natural size. Fig. 7.—Arabis dentata. The base of the plant, showing the primary root (R) and some secondary ones (r). L, L' and L? indicate the leafy rosets of Ist, 2d and 3d year. F, the lower part of flowering stems. Natural size. * 308 Gooch and Danner— Method for the Art. XXIX.—A Method for the Separation of Antimony From Arsenic by the Simultaneous Action of Hydrochloric and Hydriodie Acids; by F. A. Goocu and EH. W. DANNER. | [Contributions from the Kent Chemical Laboratory of Yale College—-IX.] A METHOD for the separation of arsenic from antimony _ based upon the difference in volatility of the lower chlorides was introduced originally by Fischer.* This method of treat- ment consisted in the reduction of the chlorides by means of ferrous chloride and the volatilization of the arsenic by repeated distillations of the mixture with hydrochloric acid of twenty per cent strength added in successive portions. The process has been subsequently modified by Hufschmidt+ by the substitution of gaseous hydrochloric acid, introduced in continuous current into the distilling mixture, for the aqueous acid, and later changed further and improved by Classen and Ludwig,t who employ ferrous sulphate, or ammonio-ferrous sulphate, in place of the less easily prepared ferrous chloride. In its latest form the method is exceedingly exact, but the con- ditions are such that the antimony in the residue must be determined gravimetrically. It has been our endeavor to so arrange the process that the determination of the antimony may be made by a rapid volumetric method, and this we have attempted to accomplish by substituting for the iron salt, which utterly precludes the direct volumetric estimation of the antimony, another reducer—hydriodic acid—which can inter- fere in no way with the subsequent determination of the antimony by the well known iodometric method. It has been shown in previous work in this laboratory that arsenic§ and antimony| may both be reduced by the action of hydriodic acid applied under appropriate conditions. In those processes, however, it was essential that the arsenic should not volatilize, and the conditions were adjusted accordingly. In the present case we have to test the reducing action of hydrio- dic acid in the presence of strong hydrochloric acid and at the boiling temperature of the solution—conditions arranged to bring about the volatilization of the arsenic as rapidly as pos- sible. Certain preliminary experiments gave indication that a half gram of arsenic oxide could be completely volatilized by the action of potassium iodide in excess in the manner described, and that an equivalent weight of antimonious oxide * Ann. Chem. u. Pharm. 208, 182. + Ber. d. deutsch. chem. Ges., xvii, 2245. ¢ Ber. d. deutsch. chem. Ges, xviii, 1110. § Gooch and Browning, this Journal, vol. xl, p. 66. || Gooch and Gruener, this Journal, vol. xlii, p. 213. | t Separation of Antimony from Arsenic. 309 (added in the form of tartar emetic) was retained entirely in the residue under similar conditions of treatment. Moreover, it appeared that this action could be brought about in solutions measuring no more than 100 cm.°* at the beginning and no less than 50 em.* at the end of the distillation, so that a very considerable saving of time over that demanded by the process of distillation as left by Classen and Ludwig could be effected. Accordingly we proceeded to test the action of the hydriodic acid quantitatively, following the same general lines. The distillation-apparatus consisted of a flask of 250 cm.* capacity, provided with a hollow glass stopper tightly fitted in a ground joint, the stopper itself being sealed upon a large glass tube bent suitably to connect the interior of the flask with an upright condenser, while through the hollow stopper, and sealed into it, passed a smaller glass tube reaching nearly to the bottom of the flask. The arrangement was such that a current of gas entering the smaller tube would pass nearly to the bottom of the flask and then out through the hollow stop- per into the condenser without meeting joints of rubber or cork. Into this flask was weighed, for the experiments of Table I, about a half gram of carefully recrystallized tartar emetic, and a half gram of pure di-hydrogen potassium arseniate and a gram of potassium iodide were added in concentrated solution, the volume of liquid being made up to 100 em.* by the addition of strong hydrochloric acid. A brisk current of hydrochloric acid gas was passed into the solution through the tube sealed into the glass stopper of the flask until complete saturation was effected, and then the liquid was heated and distilled in the continuous current of hydrochloric acid gas until the volume of 30 cm.* was reached. Iodine was evolved as soon as the liquid became warm and the greater part of it passed into the distillate with the first 10 em.’ When the final concentration was reached the solution was in each case colorless, but on cooling there appeared in one of the two experiments of this set a pale yellow tint which van- ished with the dilution involved in the transfer and washing from the flask previous to titration. The addition of starch to the cooled and diluted liquid developed no color. To the liquid were added 1 grm. of tartaric acid, to keep the antimony in solution during subsequent treatment, sodium hydrate nearly to neutrality, and hydrogen sodium carbonate in excess amounting to about 20 em.* of the saturated solution; and the antimonious oxide in solution was titrated by decinormal iodine standardized against tartar emetic. The details of these exper- iments are given in Table I. In Table II are comprised the accounts of experiments similar in general to those of Table I, excepting that the final i 310 Gooch and Danner—Method for the volume after concentration was a little more, and the antimony was in every case oxidized in alkaline solution by standard iodine previous to the introduction of hydrochloric acid and distillation. Table III includes the records of experiments similar in every respect to those of Table Il excepting that as starch showed a slight color in the cooled liquid after distillation, the solution was treated with an excess of sulphurous acid which was subsequently oxidized exactly by standard iodine previous to neutralization and the final titration. ii H.KAsO, TRL Volume Color Sb203 . Sb203 Error taken. | taken.! initial.| final. on cooling.|with starch.) taken. | found. : Sri... | Sma Aewa.e a| enn erm. grm. grm. 0°5 TSO 108 30 pale yellow none 0:2282| 0°2271| 0:0011— 0°5 1:0 100 30 none none 02283) 0°2266| 0:0017— | ie 0°5 U5 120 50 none none 0°2258) 0°2235 0-:0023— 0°5 0°5 100 | 50: |pale yellow; none 02952 0°2235) 00017 — 0°5 0°5 100 50 pale yellow; none | 0°2178, 0:2163) 0°0015— 0°5 0°5 100 50 trace none 02231) 0°2231) 0:0000 0°5 0°5 100 40 trace none | 0°2261! 0°2235) 0°0026— HOt: 0°5 0°5 100 50 |pale eT faint | 0°2268) 0 2265) 0:0003— 0°5 0°5 100 50 pale yellow taint | 0°2306) 0:2300, 0:0006— 0°5 0°5 100 50 Gale ole faint | 0°2272 02264 | 0:0008— The same general phenomena were observed in all these experiments, and deficiencies in the amounts of antimony indicated, whether the element was present in the lower or higher degree of oxidation before distillation, appear in all, but most notably in the results of Tables land Il. These losses cannot be attributed, entirely at least, to mechanical transfer in the process, inasmuch as the greater losses are not associated with the greater concentrations; and, furthermore, according to our qualitative experiments made under the con- ditions of these determinations, no antimony, so far as we could observe, passes into the distillate. If the coloration of the liquid on cooling were due to the liberation of iodine by the action of air upon the hydriodic acid the iodine thus set free might be counted upon to oxidize a corresponding portion of the antimony in the neutralization, and so to occasion a deficiency in the indications of titration. Against this sup- position, however, we have the evidence of experiment that the greatest losses are not found in those cases in which color was developed in the cooling liquid. Moreover, in all cases, Separation of Antimony from Arsenic. 811 excepting those of Table III, starch gave no test for free iodine in the diluted liquid, though it must not be overlooked that the presence of a considerable amount of hydrochloric acid tends to impair the delicacy of the test. If, on the other hand, the color is not due altogether to free iodine it is difficult to account for its development unless it is caused by the for- mation of antimonious iodide as the solution of strong hydro- chloric containing also hydriodic acid cools. On the whole, we are inclined to attribute at least a part of the apparent deficiency to the presence, at the time of neutralization, of a small amount of iodine chloride, which, in accordance with what is known of its modes of formation, might be formed by the oxidizing effect of the antimonic and arsenic oxides upon . the mixed acids. At all events, it is evident that if iodine chloride were present we should expect to note the phenomena which we do see; it would give, in small quantity, little or no color to the liquid, would not show the starch reaction for free iodine in the acid solution, and would be destroyed with the formation hydrochloric and hydriodic acids by the addition of sulphurous acid to the still acid liquid, leaving the antimony unchanged and determinable iodometrically in alkaline solution after the exact oxidation of the excess of sulphurous acid by iodine in acid solution; on the other hand, it would act in alka- line solution like the free halogens and tend to diminish the antimony indicated by titration. Whatever the real cause or causes of the deficiency may be, it appears in the results of Table III that the treatment with sulphurous acid affects the indications favorably. The mean error of three closely agree- ing determinations is 0°0006 grm.—and this is plainly within the limits of allowable variation in iodometric work with decinormal solutions. It appears, therefore, that hydriodic acid may be made to serve satisfactorily as a substitute for the ferrous chloride of Fischer’s original method, or for the ferrous sulphate of the modification of Classen and Ludwig, the determination of the residual antimony being perfectly practicable. The method of proceeding which we advocate is briefly summarized in the following statement: To the solution of the oxides of arsenic and antimony, taken in amounts not exceeding 0°5 grm. of each, potassium iodide is to be added in a little more than the equivalent quantity, and enough strong hydrochloric acid to raise the entire volume of the solution to 100 cm.’ Hydro- chloric acid gas is passed into the liquid to saturation as well as during the distillation to follow, and the distillation is carried on until the volume of the liquid decreases to 50 cm.* or a little less. The liquid is cooled rapidly, treated first with an excess of sulphurous acid and then with iodine to the exact 312 M. C. Lea—Allotropic Silver. oxidation of the former reagent; and, after the addition of 1 grm. of tartaric acid to every 0°2 grm. of antimonious oxide, the acid present is nearly neutralized with sodium hydrate, the neutralization being completed by hydrogen sodium carbonate _ added in excess to an amount corresponding to 10 em.* of the saturated solution for every 0'1 grm. of antimonious oxide present. ‘Titration with decinormal iodine standardized against tartar emetic gives the antimony quickly and with a fair degree of accuracy. The whole process requires about an hour and a half for completion. Art. XXX.—Wotes on Allotropic Silver ; by M. Carry LEA. Lelations of the Yellow to the Blue Forms.—The gold-and copper-colored forms on the one hand, and the blue, bluish- green and steel-gray on the other hand stand in close relations to each other. In previous papers there has been described a erystalline state intermediate between these active forms and ordinary silver, which intermediate condition, while retaining the bright yellow color of the active form is nearly as indif- ferent to reagents as ordinary silver. Into this intermediate state both the yellow and blue forms are capable of passing, and apparently the intermediate states of both kinds of allo- tropic silver are identical: the intermediate form of blue silver as yellow. Thus when lumps of blue silver are heated in a test tube to about 180° C. they assume a gold color and luster. The same change takes place at the same temperature when films of blue silver are placed in a hot air bath. But relations much closer than these exist. Blue silver can be converted into yellow at ordinary temperatures and conse- quently with retention of its active properties. This is accom- plished through the agency of sulphuric acid. When a solution of silver is obtained by the action of sodium hy- droxide and dextrine on silver nitrate* it appears to contain the blue variety, for if allowed to precipitate spontaneously by long standing, or if precipitated by acetic acid, dilute nitric acid, or by many neutral substances, it gives a form of silver which is dark red while moist and dries with a blue surface * Forty grams each of sodium hydroxide and of yellow or brown dextrine (not white) are dissolved in two litres of water and 28 grams of silver nitrate in solu- tion are added in small quantities at a time, with frequent stirring, so that several hours shall elapse before the last portion is added. The solution is always slightly turbid when viewed by reflected light, by which it shows a beautiful deep green color. By transmitted light it is deep red, and when diluted, abso- lutely transparent. By diminishing the proportion of silver nitrate to one-half, a solution nearly er quite clear by reflected as well as by transmitted light is ob- tained, M. C. Lea—Allotropic Silver. 313 color. (It is always a little difficult to characterize these sub- stances by their colors since the surface color which they show when dry—either in mass or in films—is mostly complement- ary to their color when wet. As the surface color is much the more characteristic, I have adopted the course of naming them by that.) The behavior of the red solution obtained by soda and dextrine with dilute sulphuric is very interesting and instruc- tive. When 100 c.c. of solution are poured into 100 cc. of water to which 3 ce. of sulphuric acid have been previously added, a dark red precipitate falls, which, when dry, especially in films, is blue. The mixed liquid from which the precipi- tate is formed is acid. Increasing the proportion of acid to 4, 5 and 6c. successively, the substance obtained has a green surface color becoming more yellowish green in proportion as . the acid is increased in quantity. With 7$c¢.c. the substance no longer dries green but yellow. Increased proportions of acid produce substances drying with a coppery shade. It will be seen that from a single solution, and using one substance only as a precipitant, we can obtain the whole range of different forms of. allotropic silver, by simply varying the proportions of the precipitant. That these forms of silver should subsist in the presence of sulphuric acid in excess is remarkable. For the most part the presence of this acid tends to quickly convert allotropic to ordinary silver. For example, bright yellow allotropic silver obtained with ferrous tartrate was washed on a filter with’ water containing 1/500 its volume of sulphuric acid: in two or three-hours the entire mass was converted into gray ordinary silver. It is observable that the substances precipitated with the least acid, have a very splendid luster, and that this luster diminishes steadily as the proportion of acid is increased. Up to 6 e.c. to 100 the effect is hardly noticeable, after that it be- comes more marked. But we can also obtain the converse of this reaction. Just as the solution which naturally would yield the blue product, can be made to yield the yellow by the presence of excess of strong acid, so the solution which normally yields the yellow substance, may be made to produce blue (or rather green) silver by adding alkali. Thus a mixture of dilute solutions of ferrous sulphate and of Rochelle salt added to mixed solutions of silver nitrate and of Rochelle salt, results in the formation of gold-colored silver. But if we add a little sodium hydrox- ide, either to the iron solution or the silver mixture, we shall get a bluish green product, whose properties show that it belongs to the blue class and not to the yellow. Evenif a 314 M. C. Lea—Allotropic Silver. solution of the hydroxide is added immediately after the iron solution has been poured into the silver, the result is the same. _ There is therefore a well marked tendency of acids to give rise to the formation of the yellow product and of alkalies to the blue. But this is a tendency only. Both substances can ~ be produced from neutral solutions, and slight changes are _ sufficient to alter the product formed. Thus, ferrous tartrate, in dilute solution acting on silver tartrate gives rise to the for- mation of the gold-colored substance, but when citrates are substituted, the blue substance is obtained. Production of Allotropic Silver by Inorganic Substances.— For reasons which will be mentioned presently, the reduction of silver must take place gradually to produce the allotropic form, and for a time it seemed an invariable condition that an _ organic substance of some sort should be present. This, how- ever, proves not to be essential. In a paper presented to the American Academy and kindly read for me by Professor Remsen at the meeting in April last, I alluded briefly to having found a reaction depending upon inorganic agents only. It is as follows: Sodium hypophosphite added to silver nitrate does not effect reduction, but when hypophosphorous acid is set free by the addition of phosphoric acid, a red coloration appears, indicating the presence of allotropic silver. The coloration is transitory, no doubt because of the strong ten- dency of free mineral acids to convert allotropic to normal silver, but red and blue stains form on the sides of the vessel. Phosphorous acid gives similar results, though perhaps less well marked. Action of Light on Blue Silver.—This action differs with different varieties: it was more especially examined with the form that is obtained from the soda dextrine silver solution already described by pouring the solution into an equal bulk of water to which sulphuric acid had been added in the propor- tion of 4 ¢.c. to each 100 ¢.c. of water. This form was selected because it is easy to obtain with great constancy of result, and because it is one of the forms of blue silver most sensitive to light. ee to light, this substance first becomes more distinctly blue, losing a slight greenish shade. With continued exposure it passes to a yellow-brown shade, and finally to a perfectly pure golden-yellow of great brilliancy and luster. The last is the intermediate or crystalline form. The action of light on this form of silver is remarkable in this respect, that its first effect is to ¢ncrease the sensitiveness to reagents. This result was so unexpected and a priord so improbable, that it was subjected to the most careful verification before M. C. Lea—Allotropic Silver. 315 being accepted. The action is very easily shown by exposing a film of the substance to light, covering part of the surface with an opaque screen. After twenty or thirty minutes of exposure to strong summer sunshine, the film may be plunged into a one per cent solution of potassium ferridcyanide, when the part exposed colors much sooner and more strongly than that which was covered. The effect is shown still better by placing the film in a frame, covering part with paper rendered absolutely opaque by coating it with thick tin foil, part with translucent paper (thick white writing paper or very thin brown paper) and leaving part wholly exposed. After four or five hours action of strong summer sunshine, the film is to be treated with, weak ferridcyanide. The part wholly exposed having passed into the gold-colored crystalline condition (if the exposure has been sufficient) is wholly unacted upon, the part covered by the translucent paper is rapidly attacked, that wholly protected is attacked slowly. So that the portion moderately acted on by light has very markedly increased in sensitiveness thereby. It follows that upon this form of silver light has a reversing action, first exalting its sensitiveness, then completely destroy- ing it. : It is impossible to overlook the analogy which exists between this action of light, and that which light exerts on silver bromide. The latter substance though incomparably more sensitive to light, is subject to the same reversing action, first gaining in sensitiveness to reducing agents and then, by continued expos- ure, becoming less sensitive than originally, a change com- monly known as solarization. Causes determining whether in the reduction of Silver, the Allotropic or the Normal form shall be produced.—i have examined the phenomena connected with the reduction of silver under a great variety of conditions. These for the most part do not deserve particular mention but seem to lead up to this generalization: that the reduction of silver may be direct or indirect, direct when it passes from the condition of the normal salt or oxide to that of the metal, indirect when the change is first to sub-oxide or to a corresponding sub-salt. So far as my observation has gone when the reduction is direct the reduced silver always appears in its ordinary form. But when the reduction is ¢ndirect the silver presents itself in one of its allotropic states. The following reactions support this view. Three of the principal modes of formation of allotropic silver are: (1) reduction of silver citrate or tartrate by ferrous citrate or tartrate; (2) acting on silver nitrate or oxide hy 316 M. C. Lea—Allotropie Silver. dextrine and fixed alkaline hydroxide; (8) acting on silver nitrate or carbonate by tannin and fixed alkaline carbonate. Now, if in either of these three cases we interrupt the action before it is complete by adding an excess of dilute hydro- chloric acid we shall obtain a dark chestnut-brown or some- times purple-brown substance which on examination proves to be a mixture of silver subchloride and photochloride. When, after complete removal of the excess of hydrochloric acid by thorough washing or better by boiling with distilled water, the substance is treated with cold dilute nitric acid that portion of the sub-chloride which is not combined with the normal chloride is broken up and there remains photochloride of a very rich and intense rose-color.* The production of silver sub-chloride in all these cases would seem to indicate that the reduction when the acid was added was incomplete, and that in case (1) a sub-salt, and in cases (2) and (3) a sub-oxide was first formed as an intermediate step before complete reduction. Hither of these substances would of course give rise to the formation of subchloride when treated with hydrochloric acid. It is important to ob- serve that this result is to be obtained only by interrupting the reaction before it is complete. When, for example, allotropic silver in solution is produced by the action of sodic hydroxide and dextrine and after complete reduction, hydrochloric acid is added, the liquid becomes filled with gray normal silver, which presently collects to a cake. When this cake is well washed and boiled with water, and then treated with dilute nitric acid, solution takes place: a trace of photochloride is left behind. It has been mentioned elsewhere that hydrochloric acid, though without action on ordinary silver, is capable of form- ing a variable quantity of protochloride when placed in contact with allotropie silver. I have not met with any exception to this general principle that when a reaction leading to the formation of allotropic silver is interrupted by the addition of hydrochloric acid, subchloride is abundantly formed as one of the products. In all such cases the reduction is evidently indirect. The silver does not lose at once the whole of its oxygen, but apparently passes through an intermediate form, probably Ag,O, the reduction of which tends to the formation of allo- tropic silver. These facts lead directly up to the question: does silver exist in its subsalts in the allotropic form? There are some * This is a very beautiful reaction and degerves more particular mention than can be given here. It is perhaps the best means for obtaining silver photochlo- ride, for which purpose I have often employed it, both on account of its facility and certainty, and the very beautiful color of the product. ’ a V -@ | the ~ Fo tt em, HI. L. Smyth— Geology of Steep Rock Lake, Ont. 317 facts that would support this view, especially the very rich and varied coloration of the subsalts corresponding to the almost infinite variety of color of allotropic silver, while normal salts of silver when formed with colorless acids are mostly colorless. On the other hand, the greater activity of allotropic silver and its less specific gravity would seem to indicate a simpler molecular constitution than that of normal silver. Art. XXXI.—Structural Geology of Steep Rock Lake, On- tario; by HENRY Luoyp SmytH. With Plate XI. GEOGRAPHY. StEEP Rock LAKE is situated in the Province of Ontario, Canada, northwest of Lake Superior and south of the Can- adian Pacific Railway. It lies about twenty-five miles east of the center of the rough quadrilateral formed by the Canadian Pacific Railway on the northeast, the Lake of the Woods on the northwest, the United States boundary on the southwest, and the shore of Lake Superior from Pigeon River to Port Arthur, on the southeast. As given on the map of the Prov- ince, scale 1 inch = 30 miles, published by the Crown Lands Department, Toronto, 1884, the geographical position of the extreme southern point of the lake, where it receives the waters of the Aticokan River from the east, is about lat. 48° 52’ N. and long. 91° 80’ W. from Greenwich; or, it is about halfway upon the map between Lac des Milles Lacs and Rainy Lake. TOPOGRAPHY. Steep Rock Lake is one of the many that, collectively, con- nected by longer or shorter links of river make up the River Seine from Island Falls as far as, and doubtless below, the Aticokan River. In shape the lake resembles, as shown on the map, an irregular and slightly distorted letter M, of which the western or left arm, (looking north), runs north and south, and the eastern or right arm northwest and southeast. This peculiar form is closely related to the character and to the structure of the rocks in which the lake lies as will be seen in what follows. The Seine River, after a beautiful fall, two hundred feet across, and forty to fifty feet high, over granite, some three hundred yards northeast of the lake, flows into it at a point about a mile southeast of the northw est extremity of the eastern arm. It leaves the lake at the extreme southern end of the western arm. Between the points of entrance and Am. Jour. Sci1.—Tuirp Serizes, Vou. XLII, No. 250.—Octoser, 1891. 318 FI. L. Smyth—Geology of Steep Rock Lake, Ont. exit there is no current observable by the eye and the differ- ence in level must be exceedingly small. As regards dimensions:—from the southern extremity of the eastern arm to the mouth of the river near the Falls is about 84 miles; from the Falls to the Elbow, 3 miles; from the Elbow to the Upper Narrows, 14 miles; and from the Upper Narrows south to the Aticokan River 34 miles. The entire lake including the portion of Lake Margaret shown on the map could be inscribed within a rectangle 6 miles from | east to west, and 54 miles from north to south, or within an area of 383 square miles. As the name implies the lake has bold rocky shores, which, in places rise 150 feet from the water in nearly perpendicular cliffs. The total length of shore line is approximately 28 miles not counting the smaller bays and indentations. Not less than # of this length shows rock in place either at the water’s edge or within a few hundred feet of it, and of this perhaps 4 may be studied without leay- ing the boat. The contour of the water line shows a very beautiful depen- dence upon structural conditions. The eastern arm follows the general strike of the rocks from the bay north of Lake Margaret portage northwest to Falls Bay. On the northeast- ern shore of this arm the lower limestone makes several bold headlands that rise abruptly from 60 to 100 feet above the water. In the bays between these headlands the basement granites intersected by a large number of greenstone dikes form the shore, and rise more gently into the broken hum- mocky hills that generally characterize the granitic areas of the region. A few hundred feet back from the southwestern shore the great trap intrusions or flows of Horizon IV make a continuous ridge, which is estimated to reach a height of 250 feet above the water. This ridge runs, without interruption, the top showing only a few minor sags, from the shore west of the portage into Lake Margaret, for 8 miles along the strike in a northwest direction as far as the wide expansion of Falls Bay. The eneissic phases of the granites, and associated irruptives compose the west shore of Falls Bay, from the great limestone exposure at the head, sonth to the Elbow. As far south as Wiegand’s Point, the granite cliffs are high and very steep and are broken across only in two or three places. On the south shore of Falls Bay from Trap Point to Jack Pine Pt. the shore line cuts the strike of the rocks nearly at right angles, and from Jack Pine Pt. south to Pine Beach obliquely at a less angle. The ridges descend rather gradually to the lake along this shore, the harder rocks making little headlands separated by sand and shingle beaches. ae ee = H. L. Smyth—Geowogy of Steep Rock Lake, Ont. 319 From the southern point of Pine Beach the shore again follows the strike of the rocks, as it sweeps round the south pitching axis of the middle anticlinal ; and in the stretch from Bowlder Point to the Upper Narrows, in which the strike locally varies between N. 2° W. and N. 18° E. the water line minutely corresponds to the minor deflections. This shore _ shows continuous rock exposure, and the cliffs reach a height in places of 40 or 50 feet above the water, having perpendicu- lar faces along cleavage surfaces. The highest land about the lake is north and west of Con- glomerate Bay, and in the peninsula between it and Northwest Bay. Except for a fringe of the Conglomerate and lower limestone (Formations I and II) along Northwest Bay, and of the upper horizons of the series east of the fault line that crosses the southeast extremity, the peninsula is composed entirely of the basement gneisses and granites, which rise from the water west of Conglomerate Bay in a steep cliff (along a surface of faulting) 100 feet or mere high. The hills in the northern part of the peninsula may reach a height of over 300 feet, but this as well as other figures concerning elevations is an eye-estimate only, and not a measurement. GEOLOGY. The rocks exposed around the shores of Steep Rock Lake and of Lake Margaret, are divisible into three principal groups. The lower division consists of granites and gneisses, which typically are medium grained, hornblendic, and granitoid, with faint foliation. Locally they present considerable variations in composition and very great variations in structure. Resting upon these basement rocks is a series of rocks show- ing a thickness of about 5000 feet in exposure along the shores of the lake. Upper members that are not seen upon the lake probably exist in the trough of the eastern synclinal, southeast of Jack Pine Pt. The series is divisable into nine formation which as far as known are perfectly persistent along the strike throughout the area studied. It offers many impor- tant points of difference, lithologically, in structure, and in its relations to the underlying granites, from any series of rocks previously described in western Ontario. Leaving the ques- tion of correlation to be discussed after the series has been deseribed it will be called for purposes of description in this paper, the Steep Rock Series. At the southeastern extremity of the eastern arm, at the north end of the portage into Lake Margaret, lying across the edges of the Steep Rock Series, begins a succession of later granite porphyries, and massive hornblende rocks, striking 320 H. L. Smyth—Geology of Steep Rock Lake, Ont. N. 55 to 65 E., which pass upward, in going south across the strike, into the schists of the Aticokan River. These will be termed the Aticokan Series. Basement Complex.-—The granites were very hastily exam- ined in the narrow fringe in which they are exposed along certain shores of the lake. They were studied mainly with reference to their distribution and structural relations to the — overlying Steep Rock series; no attempt can now be made to separate geographically the various kinds of rocks which are included in the basement series, or to indicate their relations to one another. The predominant rock in the basement series is a hornblende muscovite granite of medium grain, composed of clear to bluish quartz, feldspar, a green hornblendic mineral, and muscovite. The color on the weathered surface is white, slightly tinged with green, and on the fresh fracture a darker well marked green. This is the usual type. Occasionally a red granite carrying biotite 1s seen, which owes its color to flesh-colored feldspar. True gneisses are rare, but they are occasionally found as at locality 50 on the ridge north of the mouth of the creek emptying into the bay north of Lake Margaret portage, and in the peninsula east of North- west Bay, at locality 125. At both localities the rock is a coarse hornblende gneiss, exhibiting a parallel arrangement of the constitutent minerals, and pegmatization. At locality 125 the coarse gneiss carries angular inclusions, which are finer grained and darker than the mass of the rock, but similar in composition. Distinguishable from these gneisses in which the origin of the gneissic foliation is unknown is a great body of chloritie gneisses which have unquestionably been derived from the hornblende granite by crushing. These are found at and near the turn of folds; for example, at the head of Falls Bay, north of the Elbow, and along the north and west shores of Northwest.Bay. Good examples are seen on the west shore from the head of Falls Bay to the Elbow where the whole series has been forced round through an angle of more than 120°. All stages in the process are seen. At one end of the series is found the typical hornblende granite, traversed by little wavy fissures, generally parallel to the regional direction of cleavage, N. 43° E., along which part of the hornblende is represented by thin leaves of fresh chlorite. At the other end of the series the quartz and feldspar are greatly granulated, and the hornblende has entirely disappeared; the chlorite is arranged in parallel bands, and the rock has developed in it a highly perfect schistose structure. : H. L. Smyth—Geolegy of Steep Rock Lake, Ont. 321 All these granitic rocks are traversed by an immense number of dikes of greenstone, and more rarely of quartz porphyry, all of which for structural reasons are conveniently considered with the basement complex. These belong to three eras of irruption. (1) Those which supplied pebbles to the conglomerate at the base of the Steep Rock series. (2) Those which are seen to traverse both the granitic and Steep Rock series, and to have participated in the folding. (3) A single massive dike of porphyrite (?), which cuts through the most schistose phase of the granite at the turn of the sharp fold at the head of N.W. Bay, and is clearly subsequent to the latest period of folding of the region. The dikes of class 2 are best seen along the N.E. shore of East Bay. They are rudely parallel, the walls are straight and nearly vertical, trending from N. 45° to N. 65° E., and in a general way cutting the granites and the lower horizons of the S. R. series in a direction normal to the contact and strike. They vary in width from one to two feet up to 70 feet, and clearly were the chimneys through which passed up the mate- rial for the great mass of interbedded traps on the south side of the same arm. The contact phenomena with the country rock are uniformly as follows: When the dike is less than 6 or 8 feet in width it is fine grained, without crystalline structure, and throughout is highly schistose in the general regional direction. The wider dikes have massive and crystalline interiors, but are fine grained and schistose in a direction parallel to the induced regional cleavage, for a distance of 24 or 3 feet from the wall. The country rock also is schistose next to the wall of the dike, the belt affected being narrower where the adjacent rock is granite than where it is limestone. | With regard to distribution, it may be said in a general way that all shore lines north of the water, from Lake Margaret to Northwest Bay, are, with few exceptions, made up of the rocks of the basement complex. The exceptions are the head- lands in East Bay and at the Elbow which are of limestone, the great mass of limestone at the head of Falls Bay, and the limestone between Camp Bay and Conglomerate Bay, and the Conglomerate along the latter. Steep Lock Series.— The Steep Rock series consist of 9 well marked and persistent horizons exposed about the lake. It is very probable that other higher members exist in the land area southeast of the shore from Jack Pine Pt. to Pine Beach. The 9 formations are given in the table below which reads upward in ascending order: 322 H. L. Smyth—Geology of Steep Rock Lake, Ont. IX. Dark Gray clay slate. VIII. Agglomerate. VII. Greenstones and greenstone schists. VI. Upper conglomerate. V. Upper calcareous green schist. IV. Interbedded crystalline traps. Ill. Ferruginous formation. IT. Lower limestone. I. Conglomerate. In the foregoing description of the basement complex the northern limit of the Steep Rock series has roughly been indi- cated. Formations I and II occur in isolated patches north of the water on shores that are otherwise occupied only by the granites. Along the whole course of the lake they dip, at very steep angles, ranging from 60° to 80° away from the basement rocks, upon which they hang as a time-worn fringe, having no extension inland. The shore line lies sometimes in the granites and sometimes in the Steep Rock series, but in a general way follows closely in direction the boundary between them. It is only along these northern shores that formations I and II are seen at all, and as they are usually found together, separated from the higher members by intervening water, it will be convenient to keep them apart from the rest for pur- poses of description. The basal member of the Steep Rock series, which is generally found between the granites and the lower limestone, is a bed having a maximum thickness of nearly 100 feet, presenting the various phases of a conglomer- ate, coarse and fine, a quartzite and a quartz schist with feldspar. The formation occurs as a coarse conglomerate at the eastern end of Conglomerate Bay. The lowest member exposed at the water consists of rounded and water-worn pebbles of quartz and greenstone, of considerable size, the largest seen being a foot in diameter, imbedded in a green schistose matrix. The strike of the rock is about N. and S., as indicated both by the alignment of the pebbles, and the lines of junction of layers carrying no pebbles. No granitic pebbles were found at this locality. In the higher portion of the bed pebbles become smaller and relatively fewer, and the rock passes into a green schist, with small clastic grains of quartz. On the northern end of the island in Northwest Bay forma- tion (I) is represented by a fine conglomerate consisting of closely-packed small quartz grains (128) with little cementing material, holding occasional pebbles up to 3 or 4 inches in diameter, of rounded and water-worn quartz, bluish, milky- white and dark. A layer of limestone of uncertain thickness is also interbedded. The lowest formation does not usually occur as a coarse conglomerate in East Bay. It there consists, ~~ H. L. Smyth— Geology of Steep Rock Lake, Ont. 3238 as a rule, of beds of quartz pebbles, none larger than buckshot, with little cement, alternating with layers of massive quartzite. In crossing from the base of the limestone to the granite it becomes at first slightly and then more and more feldspathic as the latter is approached. Near the junction both rocks are very similar in composition, so that it is quite impossible to draw the line between them from considerations of composition. There is an apparent transition from one rock into the other. This transition zone, which is from 20 to 30 feet in width, is uniformly highly schistose, in the regional direction N. 43° E., which, in the eastern arm, where the strikes are from N. 50° to 60° W., crosses the courses of the contact and the bedding nearly at right angles. The schistose structure is traceable through the transition zone into the undoubted granite in which it dies out gradually, being represented a few feet away by little discontinuous wavy cracks, along which chlorite is usually developed, and by a faint foliation. From these facts it appears certain that the granitic com- plex supplied by erosion the materials for Formation I, and that the contact is therefore one of unconformability. There is no unconformability of structure; for the only normal struc- ture possessed by the basement rocks, that of schistosity, was demonstrably imposed upon them at a time long subsequent to the accumulation of the various materials which now compose the Steep Rock series. The absence of a sharp line of de- markation between the complex and Formation I, which may seem to be a difficulty in the way of accepting the existence of an unconformability, is believed to be capable of a satisfactory explanation. In considering the orotechnic history of the region the transition zone will be shown to represent, not a transition in time, but a mechanical transition in composition, dating from a time subsequent to the accumulation of the rocks of the Steep Rock series. Formation II, the lower limestone, lies above Formation I, with which it is seen in direct contact at a number of localities. The rock is very uniform in character wherever exposed. It is a dark to light bluish gray limestone, not at all highly erys- talline, often banded with layers of lighter color, along planes of original bedding. The light bands vary in width from a fine line up to 6 or 8 inches. Bedding planes are also often marked by thin cherty seams. Basal portions are frequently massive and siliceous, and in some localities are highly charged with pyrites the decomposition of which causes the rock to weather brown. The upper part of the formation is a breccia, composed of fragments of limestone, showing original struc- ture, and of trap, imbedded in a matrix that seems to be mostly made up of a consolidated calcareous flour. It is widely 324 HI. L. Smyth—Geology of Steep Rock Lake, Ont. distributed, and is nearly if not quite coextensive with the limestone. The total thickness of the lower limestone cannot be determined precisely, as it is nowhere seen in contact with the overlying Formation III, but quite surely is not less than 500 nor more than 700 feet. A much greater thickness is ex- posed in the north side of Conglomerate Bay, in part resulting from duplication by faulting. On the eastern arm, where the rocks are comparatively undisturbed the limits indicated are those given above. Formation III is found only on the south shore of the eastern arm which it fringes from the Point N.W. of Lake Margaret Portage to Falls Bay, in much the same way that the limestone fringes the north shore. As a whole the forma- tion consists of an extremely soft, fissile dull green, very pyrit- iferous rock, which carries in some localities many pebbles of limestone and a few of trap. In the lower parts of the formation the limestone fragments, which are identical with the rovk of Formation II, are rather numerous, and some are large, one, angular in shape, being over two feet in diameter. Others are apparently rounded and waterworn. In other localities pebbles are not fonnd at all. For the most part the only structure observable is the regional cleavage which is very perfectly developed. At two localities a fine banding | parallel to the strike of the rock was observed. At loe. 27, south of trap bluff this banding is very prettily shown. The rock carries a few rounded pebbles of limestone. The banded structure is thrown into little compressed 8.W. pitching folds the tangent plane to which is parallel to the plane of the dip in the limestone across the bay. Apart from the limestone inclusions and this banding the rock shows no trace of sedi- mentary origin. Under the microscope it is seen to contain no clastic material and all the evidence, which is not however conclusive, seems to point to its having been originally a vol- canic ash. At two localities a bed of banded jasper and iron ore, generally magnetite, is found, which belongs near the base of the horizon. A high bluff of trap in East Bay, probably a lenticular intrusion, must also be included in it. Except on this south shore of the eastern arm Formation III is everywhere covered by the waters of the lake, within the area studied. Sufficient evidence of its continuity is afforded however by the presence of bowlders from the characteristic iron ore hori- zon at several widely distant points. The thickness varies considerably but may be taken at a maximum of 600 feet. Formation IV consists of interbedded eruptives, which may reach a maximum thickness of 1000 feet. The rock is very uniform in character, the variations occurring being mainly in texture. It is normally a massive, rather coarsely crystalline A. L. Smyth— Geology of Steep Rock Lake, Ont. 325 greenish-gray rock, made up of plagioclase and hornblende, and is probably a diorite. Locally it includes layers of green schist which are to be regarded as mechanical derivatives, analogous to the chloritic gneissic phases of the basement granite. The formation is best seen south of the eastern arm, where it forms a long ridge running from Lake Margaret por- tage N.W. to Falls Bay, and in the two natural sections made by the shore line at both the northern and southern ends. Small patches are exposed at a number of localities about the lake. It is uncertain whether this is an intrusive sheet or a flow. Formation V is a very calcareous green schist containing thin seams of limestone. The included layers are quite pure finely crystalline limestone. Thin sections of the less cal- careous portion show that the rock consists of 50 to 60 per eent of calcite, and for the rest, of argillaceous material and secondary quartz. Originally it was probably in the main a very calcareous shale, with thin beds of limestone. The thick- ness is about 600 feet. Immediately above the limestone comes Formation VI, a conglomerate, having a maximum thickness of about 100 feet. It varies in habit, from a hydromica schist, carrying many grains of quartz, the clastic character of which is evident in thin section, to a rather coarse conglomerate, the pebbles in which seem to be entirely of quartz and granite. The locality in which it may be best seen is on Falls Bay, on the shore east of Jack Pine Point. Formation VII. The type rock of this horizon is a light greenish gray, massive, close textured rock, which weathers a light brown. In thin section it appears to be of eruptive origin, but owing to the complete alteration of the bisilicate it is uncertain whether it was originally a diabase or a diorite. Departing from this as a type, on the one hand are found members which show crystalline structure macroscopically, and on the other banded green schists which to the eye have every appearance of being altered sediments. Under the microscope however they show no trace of sedimentary origin. This banding is of two kinds. (1), a fine banding due to an alternation of thin seams differing slightly from one another in color and in texture. (2), a coarser, due to the interbedding of layers of the massive varieties with the schistose. These layers are of all thicknesses, from a few inches up to several feet. The structure, of both varieties, appears to be antece- dent to the last folding of the series, since it is often greatly contorted, and frequently intersected by the regional cleavage, and in general is parallel to the true strike. A graphitic schist, twenty feet or more in thickness, is also included. There is an 326 «Hf. L. Smyth—Geology of Steep Rock Lake, Ont. evident stratigraphical succession in the various members, the banded schists predominating towards the top. The peculiar gray green color and close texture are characteristic of the rocks of this horizon. The thickness is about 1400 feet. Formation VIII. The agglomerate is best seen at Jack Pine Point and at locality 79 to the south of it. At locality 79, it consists of pebble-like inclusions, greatly resembling the type rock of Formation VI, imbedded in a light gray-green fissile : matrix which is bright on the SSD wp cleavage surfaces. The inelu- REA emp — sions vary in size from a frac- a GF tion of an inch up to 5 or 6 Vio which is the same as the mat- oe rix apparently, differing from SS I Ne Y it only in lacking the schistose NO UD og WL, structure. They are hardly | distinguishable from the mat- Ch ee of inches ‘Xx in color, on the fresh frae- tuie, but on the weathered ¢ z 4? surface the inclusions stand Fig. 1.—Agelomerate, Loc. Theshaded out, and weather a lighter inclusions in nature are lighter in color hyown. Under the micro- than the enclosing reenachiet, The VA"- scone the inclusions are seen to consist of an eruptive rock entirely similar to the massive form of VII. On Jack Pine Point the agglomerate is beautifully plicated and the inelu- sions follow the little folds. (Fig. 1.) Formation IX. Above the agglomerate, at locality 81 a fine grained clay slate is found, which besides a perfect slaty cleavage in the regional direction, shows alternating light and dark bands, which probably represent planes of deposition. Structure.—The Steep Rock Series is folded into an eastern synclinal, a middle anticlinal and a western synelinal, which is faulted across the axis near the sharp turn. A line drawn from the limestone exposure at the head of Falls Bay to Jack Pine Point indicates approximately the position of the axial plane of the eastern synclinal. ast of this line the various members uniformly strike to the west of north. The dips are high to the S.W. ranging from vertical to - 60, and on the average may be taken at 70. From the agglomerate at Jack Pine Point southward to the point north of Pine Beach, where the upper part of Formation IV is ex- Oo 3 is ey H. L. Smyth— Geology of Steep Rock Lake, Ont. 327 posed, the shore line again crosses the intermediate formations in descending order, the strikes bending round gradually to N. 20° E. A line drawn a little west of south through the eastern point of the limestone of the Elbow marks the intersection of the axial plane of the middle anticlinal with a horizontal plane. West of this line the limestone of the Elbow, Formation VI, and the various members of Formation VII, which alone are exposed on the southern and western shore, strike again to the northwest, gradually bending round along the latter to the east of north. The limestone at Conglomerate Bay abuts against a cliff of the basement granite, the line of separation marking the position of a fault. About 6000 feet S.W. along the line of this fault, which is well marked by a breccia in the peninsula N.W. of the upper narrows, Formations [ and II are found again on the opposite side, striking N.W., and farther north on the large island in N.W. Bay, bending round again to the southwest. On the west shore of the lake west of N.W. Bay green slates, probably belonging to Formation VII, are found west of Formation II, and again on Birch Point indicat- ing another fault which trends to the west of north. These two faults so complicate the structure of the western part of the lake, that the relations of the rocks, which are all recog- nizable as belonging to one or another of the LX formations of the series, could not be worked out in the time available. There are two points in the general structure of the rocks of the lake which are especially noteworthy and significant. They are: 1st. The high pitch of the axes of the great folds. At the turn of the middle antieclinal at the Elbow, dips, which are well marked in the limestone, range from Vertical to 75° to the south, (Section II.) “At the turn of the western synclinal in Northwest Bay the dip is about 60° in the same direction. We have here, then, folds with very high south-pitching axes, the pitch in the case of the anticlinal being nearly 9U° and in the case of the synclinal at a lower angle. In the case of the eastern synclinal the pitch is also high, though apparently considerably less than 90°, as indicated by the greater thick- . hess of the series measured along the axis than across it. The Steep Rock series therefore dips away from the granites, at the turns of folds, at angles which do not differ materially from those observed in the long straight stretches; as, for example, that in Hast Bay. 2nd. ‘The regional cleavage. Throughout the whole area is observed a regional cleavage, which has a tolerably uniform direction between the limits N. 38° and N. 48° E., and trav- erses all the rocks of the region, both the eruptive and sedi- mentary members of the Steep Rock Series, and the rocks of basement complex. It has largely obliterated the original 328 «=. L. Smyth—Geology of Steep Rock Lake, Ont. lamination of the sediments and banded schists of the Steep Rock Series, and is now the dominant structure. In inferring the orotechnic history of the region the origin of the N.E. cleavage must be ascribed to a force acting per- pendicular to it, or ina N.W. and 8.E. direction; and since this cleavage runs through, and in many cases masks all pre- vious structure, the force which produced it must have been the last force which has left its marks upon the rocks of the lake. To this force also must be ascribed the action which left the Steep Rock series in its present folded attitude. What was the position of the rocks just before this cleavage- producing force acted upon them? It could not have been | horizontal, for in the long stretch in East Bay, where the strike runs nearly straight for four miles, and in the limestone exposures N.W. of the Elbow, the plane of the dip nearly coincides with the direction of this force, and the cleavage planes intersect it nearly at right angles.’ A N.W. and S.E. force acting parallel to the present strike in the plane of the dip, could not have tilted these portions of the Steep Rock Series into their present nearly vertical position. It seems necessary to suppose, therefore, that before the cleavage-pro- ducing force acted, that part of the Steep Rock Series that we know, existed asa N.W. and 8.E. striking monocline, having a high dip to the S.W. as the result of previous folding by a N.E. and 8.W. force. ee y} Fig. 2.-—Diagram showing the attitude Fig. 3.—Diagram showing atti- of the Steep Rock Series just previous to tude of Steep Rock Series after the second folding. second folding. The folding of the Steep Rock Series indicates therefore two periods of orotechnic action. In the first period, the force acted ina N.E.-S.W. direction, and folded the series about horizontal axes, having a N.W.-S.E. direction. That part of the series now exposed about the lake was left as a N.W— striking monocline, with a high dip towards the S.W. In the second period, the cleavage-producing force acted in a N.W.— S.E. direction upon this monocline and produced upon it two effects. 1st, it caused it to yield as a whole, not by vertical arching over horizontal axes, to which the nearly vertically H. L. Smyth—G@eology of Steep Rock Lake, Ont. 329 standing leaves of the series would oppose their maximum rigidity, but by horizontal buckling about nearly vertical axes, to which the opposed rigidity would be a minimum. Figures 2 and 3 illustrate this point. 2nd, consequent upon the regional movements attending the folding were produced mi- nute fissures, and a rearrangement of particles along planes perpendicular to its direction ; or, in other words, the regional cleavage. : These two periods of orotechnic action explain also the schistose dikes and the transition zones between Formation I, and the granitic complex. The dike at locality 41 will serve as an example of the class. It is from 60 to 70 feet in width and cuts the granitic complex, the southeast wall running N. 48° KE. For a distance of three feet from the wall the dike is very schistose and fine grained. ‘The interior is massive and erystalline. Under the microscope the interior is seen to be a quartz diorite, consisting of quartz, plagioclose, hornblende,— in places altered to chlorite and epidote, magnetite and apatite. The feldspars are nearly all saussurized. A slide from the schistose portion shows it to be a hydromica schist, with a great deal of normal chlorite. The quartzes are strained and broken, and the magnetite is granulated and drawn out into fissured “augen.” Some epidote is arranged along planes of foliation. The rock shows shearing and crushing in an eminent degree. Previous to the first orotechnic period we may suppose that the Steep Rock series lay in a horizontal position upon the basement complex. The parallel dikes, of which 41 is an example, which supplied the materials for the interbedded eruptives of the series, constituted a system of thin vertical beds running through and binding together the grantitic com- plex, and the sediments and interbedded eruptives of the upper series. The effect of the first force was to arch the series about horizontal axes parallel to the present strikes in East Bay, and perpendicular to the course of the chimney series of dikes. It acted parallel to the direction of these dikes and therefore opposed to their greatest rigidity. As the upper series bent under the action of the force, there must have been a difference in the rate of yielding of the bedded sediments and thin horizontal eruptives, on the one hand, and the verti- cal dikes and massive granitic complex on the other. This difference in rate of yielding must have produced grinding: Ist, of the basal sediments on the granitic complex; 2nd, between the vertical dikes and the rocks through which they passed, whether sediments, interleaved eruptives, or of the complex. The grinding would result in shearing and comminution of both rocks in zones adjacent and parallel to the contacts. The 3380 FZ. L. Smyth— Geology of Steep Lock Lake, Ont. transition zone at the junction of the basement complex and Formation I, represents the depth to which the granites, previously weakened by disintegration, were affected by the grinding. In this zone there was also doubtless a certain intermingling of particles produced by the action of gravity. The later orotechnic force has imposed schistose structure in these zones of comminution just as in the sedimentary members of the upper series. Behavior of the complec.—How have the crystalline rocks of the basement complex yielded to the tremendous stresses which produced the buckled folds in the upper series? This most interesting question must be left incompletely answered. A few points, however, are clear. Great relief was afforded by the fault across the northern end of the lake. The dip of the fault plane, unfortunately, was not observed, but it is con- ceived that the movement was essentially horizontal, and that the 6000 feet of observed throw is nearly the full measure of its amount. In the zone adjacent to the Steep Rock Series, in which alone the granites were studied, the presence of numerous folded dikes, in the localities in which the whole series has been specially folded, seems to show that the granites yielded by bend- ing and not by faulting. One of these is shown in figure: “ieee one of many seen in the stretch of shore north of Wie- gand’s Point. This bending in the granites implies, ,, - Of course, accord- 77 ing to Heim’s law, a tremendous load upon them. Thickness.—The thickness of the Fig. 4.—Plan of folded Dike. a=Greenstone members of the Dike. 6=Chloritic gneiss, autoclastic from Steep Rock series granite. The surface dips gently to the North, civen in the de- 15". = North ig Scale of Feel“ 5 10 »o ° scription of the separate formations, is that measured in the section across East Bay, and is undoubtedly too great. This part of the series underwent tremendous longitudinal compression by the cleay- age-producing force; the thinly bedded members have been thrown into little compressed folds, and the more massive members have been contorted on a larger seale. In both ways Cy Sey” H. L. Smyth— Geology of Steep Rock Lake, Ont. 381 the thickness of the series measured in section has been largely but indeterminately increased. In the stretch from the Elbow north to the head of Falls Bay, west of the axial plane, the series is much thinner, probably as the result of three causes: (1) A general stretching in this direction. (2) A possible faulting along the axis. This has not been observed, however. (3) A probable thinning out of the trap horizon in going west. For these reasons a measurement across the series there would probably be at least as much less than the true thick- ness as the East Bay section is greater. A mean between them, or 4500 feet, may be taken as an approximation to the true thickness. General Considerations.—The study of the Steep Rock series shows some results both positive and negative which have a general interest in connection with the geology of the region N.W. of Lake Superior. 1st. The contact of Formation I with the basement complex is one of erosion. Zd. The complex at the time of the deposition of the Steep Rock series was made up of consolidated crystalline rocks, and there is no evidence whatever that it has since undergone fusion, or recurred to the condition of a magma. 3d. The rocks of the Steep Rock series have been subjected at two periods, more or less distant, from one another, to the action of great orotechnic forces, which acted—the first in a N.E. and 8.W. direction, and the second in a N.W. and 8.E. 4th. The latter force has imposed upon all the rocks of the region a N.E. structure, which has largely, but not entirely, obliterated preéxisting lamination in the sediments and schists of the Steep Rock series. 5th. The two orotechnic actions have produced great devel- opments of autoclastic* schists, both in the granites and in the rocks of the Steep Rock series; the present structure in which was induced and determined in direction by the last force. The consideration of the Aticokan series, with a more general discussion of the mutual relations of the three series of rocks, and an attempt at correlation, must be deferred to another paper. | The author wishes to express his great obligations to Prof. Raphael Pumpelly for many valuable suggestions. Mr. C. Livy Whittle, of Cambridge, Mass., has kindly examined a number of thin sections from the Steep Rock series, and the results of his study are incorporated in the above description of formations. Port Arthur, Ontario, June, 1891. * That is, schists formed in place from massive rocks by crushing and squeez- ing, without intervening processes of disintegration or erosion, removal and depo- sition. = 332 B. J. Harrington—So-called Amber of CW, | Tf. Art. XX XII.—On the so-called Amber of Cedar Lake, North Saskatchewan, Canada; by B. J. Harrineron, McGill College, Montreal. , THE occurrence of mineral resins in some of the coals and lignites of the Northwest and British Columbia has been known for many years, and the results of a partial examina- tion of specimens from three localities were published by the writer in the report of the Geological Survey for 1876-77, p. 471. The conclusion then arrived at was that none of the specimens could be referred to amber or succinite, though in some respects closely resembling that substance. Attention was also called to the statement of Goeppert that he knew of no instance of true amber being found in the brown coals of northern Germany, the substance occurring in those beds being “retinite.” During the summer of 1890, Mr. J. B. Tyrrell, M.A., of the Geological Survey of Canada, visited a locality on the west shore of Cedar Lake, near the mouth of the North Saskatch- ewan, where a mineral resin resembling amber in appear- ance has been found in large quantity. With regard to it Mr. Tyrrell says: “It occurs mixed with sand and many fragments of partly decayed wood, on a low beach behind a gradually shelving shore and along the face of a deep, wet, spruce swamp. The pieces were, for the most part, smaller than a pea, but could be readily seen glittering among the sand and vegetable debris. Some pieces were found as large as a robin’s ego, and Mr. King [of the Hudson’s Bay Company] informed me that he had collected pieces very much larger. It has evi- dently been washed up on the shore by the waves, but its exact age has not been positively determined. “The first place at which it was seen was in a small bay behind a limestone point projecting towards the north, but the most extensive deposit is more than a mile south of this point, where a rounded beach stretches across the margin of a low swamp. This beach is about a mile in length and from eighty to one hundred and twenty feet in breadth. The amber is found most plentifully along its ridge, where it constitutes between five and ten per cent by volume of the sand and vege- table debris, and holes dug to a depth of two feet show no diminution in its quantity. Towards the edge of the lake, however, the sand is freer, both from fragments of wood and amber. It is difficult to make an accurate estimate of the quantity of amber on this mile of beach, but it may confidently be said to be found throughout the distance in a band thirty feet wide, with a minimum depth of two feet.”* * Summary report of the Geological Survey Department for 1890, p. 22 Cedar Lake, North Saskatchewan, Canada. 333 The writer is indebted to Mr. Tyrrell for specimens of this so-called amber from Cedar Lake, and the results of their examination, as far as completed, will now be given. The substance was in pieces, for the most part very irregular in shape, some being more or less angular, others approximately spherical, and others flattened, discoid or lenticular. Some of the pieces were smaller than a pea, but they ranged from this up to the size of an ordinary bean (about 2 centimeters long). In color they varied from pale yellow to dark brown, and many, when examined by transmitted light, appeared clouded or banded from the presence of black carbonaceous matter. Superficially they were generally dull, owing, perhaps, to chemical change, but on fresh surfaces the luster was resinous. The fracture was conchoidal. Though electric on friction, they appeared to be less strongly so than ordinary amber. Light-colored fragments, free from black carbonaceous mat- ter, were selected for examination, and any superficial crust carefully removed by scraping. The hardness of these selected pieces was fully 24. The specific gravity, as obtained with a quantity of material in the specific gravity bottle, was 1-055 (at 20° C.), and a single fragment gave by suspension with a hair 1:0548 (20° C). The material for analysis was finely powdered and dried over sulphuric acid 7 vacuo. The combustions were made with lead chromate in the usual way, and the ash determined with a separate portion in a platinum crucible. The following are the results obtained : rr EF Mean. Gurpon SLO Yi lata! BS S| 79°96 iyropen |. 10°37 10°55 10°46 Oregon 2 9°53 9°45 9°49 Pash bel. O00 0°09 0°09 100°00 100°00 100°00 Excluding the ash the results become: pi Il. Mean. Marvontiiits 222: 80°08 79°98 80°03 Hydrogen 724...) 1038 10°56 10°47 Dyson: sen... 9°54 9°46 9°50 100°00 100°00 100°00 The ash was brick-red in color and found to contain silica, alumina, iron, lime, and magnesia. The only solvents whose action upon the resin has been tried as yet are absolute alcohol and absolute ether, and the effect of these was ascertained as follows: One gram of the finely pow- Am. Jour. Sci1.—THIRD SERIES, Vou. XLII, No. 250.—OctTosrr, 1891. 334 B. J. Harrington—So-called Amber of dered resin was mixed with ten grams of pure quartz sand in a cylinder of filter paper and extracted in Soxhlet’s apparatus, in the case of the alcohol for three and a half hours (24 siphon- ings) and in the case of the ether for two hours (24 siphonings). | In each case the sand and filter paper were previously ex- tracted by the special solvent for several hours. The extract from the resin was evaporated in a weighed platinum dish and the residue dried at 100° C. The results obtained were as follows : Dissolved by absolute alcohol -- ---- 21-01 per cent. 66 66 66 ether ahs Peed 94°84 ce The effect of more prolonged action of. the solvents has not as yet been ascertained. .The alcoholic extract after drying was brownish in color, while that obtained with ether was only faintly yellow. When small fragments of the resin were heated in closed tubes it was found that they began to soften at about 150° C., the point of softening being roughly ascertained by pressure with a platinum rod. At 180-190° C. the fragments were sufficiently yielding to be pressed into one mass by the plati- num rod. Heated up to 300° OC. the resin did not melt into a flowing liquid, put became soft and elastic, and had darkened a good deal from partial decomposition. Fragments of genuine amber behaved in a similar manner, but began to soften at about 140° C. At 180° they could be readily pressed into one mass, and in the one experiment tried they seemed to darken more readily than the Cedar Lake resin when heated up to 280°-300° C. The ordinary statement that amber fuses at 287° C. is certainly misleading, the fact being that it begins to soften at a very much lower temperature, gradually getting softer and softer as the temperature rises, but not becoming a flowing liquid until decomposition takes place. : : On heating the Cedar Lake resin in a test tube or retort no” erystals of succinic acid were obtained, although they were readily obtained from true amber by similar treatment. It is customary to assign to amber the formula O,,H,,O,, which gives: carbon 78°94, hydrogen 10°53, oxygen 10°53; but this is apparently based upon very insufficient data—so far as the writer is aware, upon the single analysis of Schrotter (carbon 78°82, hydrogen 10:23, oxygen 10°95), which really corresponds much more closely to C,,H,,O,. Such a substance as amber, too, coming from a variety of localities and originally derived from very different plants can scarcely be expected to agree closely in composition with one definite formula. fies ee § 4 Cedar Lake, North Saskatchewan, Canada. 335 The Cedar Lake resin contains more carbon than the amber analyzed by Schrotter and less oxygen, and in this respect comes nearer to Walchowite and to some of the recent copals from India. The relations of some of these bodies will be made plain by the following tables: Ratio of C, H, Ratio of C, H, and O atoms, Carbon. Hydrogen, Oxygen. and O atoms. taking C =40. Le. Sit LT: ei A 4e-82 = 10-23. 10°95 9°60:14°95:1 40:62°29:4:16 1YS Weanivine S22 2 = = - foes Oa) Wiss 2251S list 4056305 :-3°91 III. Cedar Lake Resin 80°03 10°47 9°50 11°23:17-63:1 40: 62-79: 3°56 IV. Copal (Bombay)-- 79°70 10°40 97905; 10715: 1683.21 »40:62°62: 3°72 VY. Copal (Caleutta)__ 80°34 10°32 9:34? 11°46:17°67:1 40:61°67: 3.49 I, Phillips’ Mineralogy (1852), p. 630. Anal. by Schrotter. Il. Dana’s Mineralogy (1869), p. 741. Anal. by Landolt. IV. Watts’s Dictionary of Chemistry (ed. 1.), vol. ii, p. 19. Anal. by Filhol. V. Watts’s Dictionary of Chemistry (ed. i), vol. 1, pea9: Anal. by Filhol.* Though resembling amber in some of its characters, the Cedar Lake resin may here be classed provisionally as “ re- tinite,” on account of its differmg from amber in its deport- ment with solvents,t in not yielding crystals of succinic acid on distillation, and in. having a somewhat different ultimate composition. The name retinite as used by some mineralogists is a convenient general term to include such substances as Walchowite, Krantzite, Jaulingite, Rosthornite and the Cedar Lake resin, which last, by way of distinguishing it from other retinites, may be called Chemawinite (from Chemahawin or Chemayin, the Indian name of a Hudson Bay post, not far from where the resin occurs). Though the origin of this substance is not certainly known, there can be little doubt that it has been derived from one of the Tertiary or Cretaceous lignites occurring on the Saskatche- wan. Some of these are known to contain resins, one of which, examined by the writer, was not essentially very different from the Cedar Lake material. It behaved similarly on heating, had a hardness of over 2, a specific gravity of 1:066, and dis- solved in absolute alcohol to the extent of 29-30 per cent. Some of the larger pieces of the Cedar Lake resin might, perhaps, be employed for ornamental purposes (beads, etc.), and possibly the material might be utilized by the varnish- maker. ‘This question will be discussed when the examination of the resin is completed. * In the last analysis, as given by Watts, there is an error. The total is given as 100, whereas it is really only 99°80. It is here assumed that the error is on the oxygen—the constituent determined by difference: A similar error occurs in Schrotter’s analysis of amber, as given by Dana. + The statements in works on mineralogy with regard to fossil resins are often vague and sometimes conflicting. Thus, in speaking of the action of such solvents as alcohol or ether, we are told nothing as to the strength of the solvent, the dura- tion of its action, etc., and the results given are, therefore, often of little value. 336 O. C. Marsh— Geological Horizons as ry 1 Vi « Art. XXXIIT. — Geological Horizons as determined by Vertebrate Fossils ;* by O. C. Marsa. With Plate XII. — In 1877, the author endeavored to bring together some results of his researches in the Rocky mountain region and in other parts of the country, relating to the succession of vertebrate life.+t This led to a comparison of the relative value of the three different groups of fossils; plants, invertebrates, and vertebrates, in marking geological time. In examining the subject with some care, the author found that, for this purpose, plants are not satisfactory witnesses ; that invertebrate animals are much better; but that vertebrates afford the most reliable evidence of climatic and other geological changes. The subdivisions of the latter group, and, in fact, all forms of animal life, are of value in this respect, mainly according to the perfection of their organization, or zodlogical rank. Fishes, for example, are but slightly affected by changes that would destroy Reptiles or Birds, and the higher Mammais succumb under influences that the lower forms pass through im safety. The special applications of this general law, and its value in geology, readily suggest themselves. In accordance with this principle, the author next attempted to define the principal geological horizons in the West which he had personally investigated, and then taking in each the largest and most dominant vertebrate form which characterized it, used the name for the horizon. In the same way, some of the principal horizons of the East were named, and the whole brought together in a section to illustrate vertebrate life in America.t The names thus given to various horizons were not intended to replace those already applied, but merely to supplement them, and by new evidence, to clear up those in doubt. The same principle had long before been found to work admirably in Europe, where certain characteristic invertebrate fossils, especially Ammonites, had served to mark definitely various subdivisions of a single formation. The wider application of the principle to vertebrate fossils, from their earliest known appearance to the present time, has already helped to complete the record of vertebrate life in America, and rendered an equal service to systematic geology. Since this method of defining geological horizons by vertebrate fossils was first used by the author in 1877, many important * Abstract of Communication made to the International Geological Congress Washington, D. C., August 28th, 1891. + Introduction and Succession of Vertebrate Life in America. Address before the American Association for the Advancement of Science, Nashville, Tenn., August 30, 1877. { The same address, Frontispiece. determined by Vertebrate Fossils. 337 discoveries have been made, especially in the West, and much information bearing on the subject has been obtained from various quarters. In 1884, the author revised and extended the first section for his monograph on the Dinoverata, and it seems fitting on the present occasion to bring together once more some of the later evidence, and place on record the more -important horizons now known to the author by personal exploration, or by other investigations which he has verified. The accompanying section, Plate XII, is designed to represent in outline, in their geological order, the successive horizons at present known with certainty from characteristic vertebrate fossils. The correlation of these horizons with those deter- mined on other evidence is important, and considerable progress in this direction has already been made, but the results cannot be presented here. In-comparing the present section with the one first published by the author, it will be noticed that no vertebrates are yet known in the Archean or Cambrian, but a single fortunate discovery in Colorado has recently carried back the first known appearance of Fishes, from the lower Devonian to the lower Silurian, or more specifically, from the Schoharie Grit to the - Trenton. The next point of importance is in the Triassic, in the horizon of the Connecticut river sandstone where so many foot-prints have been found, and attributed to Birds. Recent discoveries in these beds have shown that at least three distinct forms of carnivorous Dinosaurian reptiles, all of moderate size, lived at that period, and doubtless did their share in leaving foot-prints behind them. In two of the skeletons secured, the bones of the hind feet are still in position, and in life could have made some of the foot-prints previously discovered. Near the base of the Jurassic, a new horizon may now be defined as the Hallopus beds, as here alone remains of the remarkable reptile named by the author //allopus victor have been found. Another diminutive Dinosaur, anosaurus, occurs in the same strata. This horizon is believed to be lower than the Baptanodon beds, although the two have not been found together. The Hallopus beds now known are in Colorado, below the Atlantosaurus beds, but quite distinct from them. The Baptanodon beds have been found at many localities, in Dakota, Wyoming, and northern Utah, everywhere beneath the Atlantosaurus beds, and having below them, at various localities, a series of red beds, which may, perhaps, contain the Hallopus horizon, but are generally regarded as Triassic. Beside the two species of Baptanodon described by the author, the next vertebrate in importance, in the same horizon, 338 Scientific Intelligence. is a small Plesiosaur, which may be called Parasaurus striatus. One specimen only has been found in northern Wyoming. The Atlantosaurus beds of the upper Jurassic are now known to be one of the best marked horizons yet discovered. They have been traced for more than four hundred miles along the eastern flank of the Rocky mountains, and nearly everywhere contain great numbers of fossil vertebrates, espe- cially gigantic Dinosaurs and other reptiles, as well as many diminutive mammals of primitive types. The same deposits have been found on the western slope, with the Baptanodon beds beneath them. — The most remarkable of the new horizons recently deter- mined are the Ceratops beds in the Laramie series, at the top of the Cretaceous. This horizon is as strongly marked as that of the Atlantosaurus beds, and has now been traced for nearly eight hundred miles along the eastern base of the Rocky mountains. Toward the north, it is underlaid by marine Cretaceous strata containing Fox Hill fossils, but further south, various older formations are found immediately beneath it. The overlying strata, when present, are usually of Tertiary age. The Fort Union Eocene beds on the upper Missouri, the Brontotherium beds of the Miocene in Wyoming, and further south in Colorado the Phohippus beds of the Pliocene, may be seen immediately above. The vertebrate fauna of the Ceratops beds is remarkably rich and varied. The gigantic horned Dinosaurs named by the author the Ceratopside espe- cially abound, and determine the horizon with accuracy. Other Dinosaurs are numerous; and a few Birds, and various Mammals of Mesozoic types have also been secured. In the various horizons of the Tertiary, as repeated in the present section, no changes of importance have been required, as more recent discoveries fully confirm their value and accurate determination. SCIENTIFIC INTELLIGENCE I. CHEMISTRY AND PHysIcs. 1. On the Absorption Spectrum of Liquid Oxygen.—In a pre- liminary examination of the absorption spectrum of liquified oxygen, Orszewsk1 observed four bands corresponding to the wave lengths 628, 577, 535 and 480, these bands being the same as those noted by Liveing and Dewar in the spectrum of gaseous oxygen at high pressures, in addition to two bands in the ultra red agreeing with the Fraunhofer lines A and B. More recently Olszewski has prepared liquid oxygen in larger quantity and has examined its absorption spectrum more critically. Using a layer Chemistry and Physics. 339 30 mm. thick and 50 mm. high, contained in a thin glass vessel surrounded by three glass beakers to protect it from outside heat, it was found possible to maintain it at atmospheric pressure at its boiling point —181°4° for half an hour or more; and thus to submit it to observation for that time. The four absorption bands above mentioned were observed, and in addition a fifth band corresponding to the Fraunhofer line A, more intense than the band of wave length 535 but less so than the others. No ab- sorption band corresponding to the line B was seen. In 1883 liquid oxygen was described as colorless ; but with larger quan- tities, the author has noticed that in a layer of greater thickness than 15 mm., it has a distinct blue color by transmitted light. Since special care was taken in the purification of the gas, and since ozone was proved to be absent, the author believes this color to be characteristic of liquid oxygen. Moreover, he suggests that the blue color of the sky may be due to the oxygen in the atmosphere.—Ann. Phys. Chem., Il, xlii, 663; J. Chem. Soc., lx, 773, July, 1891.. } GiB: 2. On the Production of Ozone in Rapid Combustion.—The statement of Inosvay that ozone is not produced in rapid com- bustion having been questioned, he has reéxamined the matter and concludes that the tests by which the presence of the ozone was established by Loew and Cundall were not satisfactory. He finds that neither in the products of combustion nor in the air taken from around a flame is any substance present which (1) gives the odor of ozone, (2) renders thallous oxide paper brown, or (3) permanently decolorizes a solution of sulphophenyl-azo-a- naphthylamine so that naphthylamine no longer restores the color. By carefully depriving the gas used of sulphur compounds, he obtained in only a single experiment a reaction with thallous oxide paper ; and this after about seven hours. Taking special precautions to keep the temperature of the flame low, how eve and employing a special collecting apparatus, he obtained the thallous oxide reaction in about 4 to 5 minutes and the other re- action in 10 to 15 minutes. Examined in this way the author finds the flame of methane to give less, the flames of hydrogen and carbon monoxide more ozone than that of illuminating gas. Moreover it appears that the relative amounts of nitrous acid and ozone formed by a flame depend upon its temperature and upon its surface ; the ozone formation being favored by a low temper- ature. Oxyg en did not give as good results as air. Even if the oxygen is partially converted into ozone by blowing a current of this gas or of air on a flame, this fact the author thinks does not contradict his statement that ozone is not formed during rapid combustion. These results agree with those of Dewar and those of Elster and Geitel. The former chemist ozonized oxygen by passing it over white hot platinum. Since therefore the condi- tions essential to the production of ozone are not present in ordi- nary combination, this cannot be the source of the ozone of the atmosphere.— Bull. Soc. Chim., Ill, iv, 707; 4. Chem. Soc., |x, 798, July, 1891. G. F. B. 340 Scientific Intelligence. 3. On Sulphuryl Peroxide.—By the action of the silent electric discharge upon a mixture of either sulphuric oxide or sulphurous oxide and oxygen, Berthelot obtained several years ago a erystal- lized compound tc which he gave the name persulphuric acid and the formula $,0,. The same substance was also obtained by Berthelot by the electrolysis of a 40 per cent sulphuric acid. Shortly afterward Trause made a preliminary examination of this compound and concluded that its formula was SO, and not S,0,; and further that it was not an acid oxide as Berthelot sup- posed, but a neutral substance, sulphuryl peroxide. The present paper deals with the analytical results of his investigation. Al- though he has not succeeded in isolating the peroxide, he has obtained it free from the 40 per cent sulphuric acid in which it was dissolved. ‘This was done by diluting the solution with 2 to 4 times its volume of water and adding to it freshly prepared barium phosphate. The sulphuric acid is thrown down as barium sulphate and the filtrate contains the peroxide dissolved in phos- phoric acid with some barium phosphate. It does not seem capa- ble of existing in solution in pure water. As the peroxide easily evolves oxygen and is reduced to sulphuric oxide, the composition of the dissolved compound was ascertained by taking a known vol- ume of the solution, determining first the active oxygen therein by a known solution of ferrous sulphate, titermg back with perman- ganate, and then the sulphuric acid as barium sulphate. In two experiments, the active oxygen was found to be 9°62 and 35°96 milligrams, the SO, present being 49°5 and 178'5 ; giving the ratio 1:5°1 and 1:5 in the two cases. Hence 16 parts active oxygen are combined with 80 parts of SO,; i. e., SO,+O = SO, ; or per- haps (SO,) +O, = 8,O,. In order to determine its neutral char- acter, the electrolyzed sulphuric acid after dilution with one to two parts of water and cooling to —10°, was saturated with dilute alkali ; a process which did not ‘affect the SO,. On boiling the neutral solution thus obtained for a half hour, until a drop gave no blue coloration with zinc-iodide-starch solution, the active oxygen was expelled and the solution became intensely acid. Evidently if the peroxide had been an acid oxide and had formed an acid with the water present, a salt K,SO, would have resulted from the saturation of this acid by potassium hydroxide. And this on giving up oxygen would have produced K,SO, still neu- tral. This evidence of neutrality on the part of the peroxide was confirmed by quantitative data. The ratio of the active oxygen in the solution before boiling to the sulphuric acid produced by the boiling was determined ; the acid by titration with sodium or potassium hydroxide, using rosolic. acid as an indicator ; and the active oxygen either by ferrous solution and permanganate or by potassium iodide and sodium thiosulphate. The ratio varied from 1:4°56 to 1:5°10; giving 1: 4°85 asamean. The active oxygen as determined by the iron method was somewhat higher than that given by the iodine method, owing to the presence of acid car- bonate of the alkali in the solution, which decreased the free acid Chemistry and Physics. 341 in this solution and also decreased the quantity of the active oxygen as determined by the iodine method. The author regards this compound either as sulphuric oxide in which a single atom of oxygen is replaced by a double one, SO,(O,) or as hydrogen peroxide in which the hydrogen is replaced by SO,, corresponding _ to Brodie’s class of neutral peroxides. Berthelot’s 8,0, he regards as SO,+S0O,. Since in not too dilute sulphuric acid, it dissolves without evolution of oxygen, the equation $,O,+ H,O = H,SO,+ ‘SO, shows the identity of the product thus obtained with that produced by electrolysis.—Ler. Berl. Chem. Ges., xxiv, 1764, June, 1891. G. F. B. 4, A Dictionary of Applied Chemistry ; by T. E. Tuorpx assisted by eminent contributors. Vol. I, 714 pp., 8vo. Lon- don, 1891, (Longmans, Green & Co.)—The first volume of the. Dictionary of Applied Chemistry—the successor, on the techni- cal side, to Watts’s Dictionary of Chemistry—was noticed in volume xxxix of this Journal, (p. 519). The second volume has now appeared and carries the work on from Eau to Nux, and hence the completion of the whole may be looked for at an early date. Some of the more important subjects discussed at length, and in many cases with liberal illustrations, are the following : Explosives, by W. H. Deering ; fermentation, by P. F. Frank- land; fuel, by B. H. Brough; coal gas, by Lewis Wright ; gold, by E. J. Ball; india rubber, by C. A. Burghardt ; iron by Thomas Turner ; lead nye Bik? Bedson ; matches, by E. G. Clayton ; napthalene, by W. P. Wynne. The same thorough but concise treatment before noted characterizes this volume and makes the _ work as a whole indispensable to all interested in any of the many departments of technical chemistry. 5. Meusurement of time of Rotation.—The ordinary methods of determining the time of very rapid rotation depend in general upon the contact of a stylus on the prong of a tuning fork with a rotating wheel or cylinder, or on the use of the electric spark with a pendulum to indicate the time of rotation. K. Pryrz departs from both of these methods and employs a falling body upon or against which the rapidly rotating body spirts a fine jet of coloring matter. In this way retardation of contacts is pre- vented, and the time is referred directly to the time of a falling body. The author gives in detail the method of holding the small glass tube containing the coloring matter, and the method of obtaining the records. Examples of determination of time by this method are given and the author claims that the time of a complete revolution of his apparatus could be determined to 0°00002 of a second.—Ann. der Physik und Chemie, No. vii, 1891, pp. 638-651. Bre 6. Method of determining Specific Heat by means of the Electrical Current.—Yhe method of determining specific heats by the use of Joule’s law has not proved useful, except in non- conducting liquids. Professor PraunpLEeR has obviated the difficulty of conduction through the liquid by employing glass 342 Scientific Intelligence. spirals filled with mercury. These spirals were placed in a Wheatstone’s bridge in order to control the ratio of the resist- ance during the flowing of the current and to keep it constant.— Wiener Berichte, April 9, 1891. Js) th. 7. Optical relation of Organic Dyes.—K. VocEt discusses the sensitizing power of the various compounds of eosine and gives charts of the sensitizing power. He recommends for ortho- chromatic photography that ordinary dry plates containing a weak amount of iodide of silver should be bathed in the follow- ing: 25°™ solution of the coloring matter, erythrosine, in water, (1:1000); 0°5°™ solution nitrate of silver (1:20 water); 2°™ ammonia, spec. grav. 0°94; 75°™ distilled water. The author finds that the eosine dyes which are the most strongly fluorescent are the poorest sensitizers. Among the other conclusions of the writer we find the following: The sensitizing power of the eosine dyes depends: 1. On the sensitiveness to light of the dye. 2. On the proportion of light rays that is absorbed in other than chem- ical work. The more energy of the latter that is consumed in other than chemical work the smaller the chemical action.— Ann. der Physik und Chemie, No. vii, 1891, pp. 449-472. J. de 8. Maxim’s Flying Machine.—It is stated that Mr. Maxim is now constructing a flying machine at Crayford, which is nearly ready for launching. “It will be propelled by a light screw making 2500 revolutions per minute. The motive power is said to be a petroleum condensing engine weighing eighteen hundred pounds, and capable of raising a forty thousand pound load. The real suspending power will lie in an enormous kite measuring 110 feet long and 40 feet wide.”— Nature, July 30, 1891. 4. 7. 9. Small Hlectrometers.—At a meeting of the Physical Society held in London, June 26, Protessor Boys described small portable electrometers of his design. In one of these the needle was cross-shaped and made of zine and platinum, and reliance was placed upon contact electricity to keep the needle at different potentials. Tbe instrument was very sensitive-—Vuture, July 16, Leon: suit 10. Influence of brightness upon phenomenu of interference of light.— According to Michelson and Morley the red hydrogen line is a Close double. They found that if the hght of this line was employed to obtain interference bands, that these bands disap- peared with a difference of path of 15,000 wave-lengths, and also with a difference of path of 45,000 wave-lengths, and from a similar phenomena produced by the double sodium line it was concluded that the hydrogen line H, consisted of two compo- nents at a distance apart of 4, of that of the sodium lines. Exertr concludes from his investigation of this subject that this difference in the position of minima is not connected with du- plicity of the line but depends upon particular conditions of the source of light ; and believes that peculiarities in the appearances of the hydrogen line in stars of certain types depends also upon character of emission of the light. The method of high interfer- an Geology. 343 ences promises to give an insight into the relation between the character of the light and the distribution of light in its spectral lines.—Ann. der Physik und Chemie, Ne 790-807, no. 8, 1891. oui 11. Thought transfer ence.—Professor Basen. President of the section of Mathematics and Physics at the late meeting of the British Association, used the following language: ‘‘ May there not also be an immaterial (perhaps an etherial) medium of com- munication? Is it possible that an idea can be transferred from one person to another by a process such as we have not yet grown accustomed to, and know practically nothing about? In this case I have evidence. I assert that I have seen it done and am perfectly convinced of the fact.”—Wature, Aug. 20, 1891, p. 386. Pe 2. E. Il. Gro.uoey. 1. Lifth Triennial Meeting of the International Congress of Geologists.—The International Congress commenced its sessions at Washington, on Wednesday, the 26th of August. The meet- ing was called to order by Prof. T. McKenney Hughes, of Cam- bridge, England. After the election of officers, in which Dr. J. S. Newberry was chosen President, the chair, in the absence of Dr. Newberry, was taken by Prof. Joseph LeConte, one of the Vice-Presidents. The principal subjects discussed during the sessions are the following: The Ice-period in America and north- ern Europe and the classification of pleistocene formations, which was opened by President T. C. Chamberlin and occupied Thurs- day ; the Correlation of European and American geological for- mations, opened by Prof. G. K. Gilbert, occupying Friday ; the Graphic system used in geological work, opened by Major Powell, on the forenoon of Monday. The afternoon of that day was given up to discussions relating to the geology of the regions to be visited by the Western excursion. On the subject of Correlation, the value of the effects of physi- cal events or conditions and of relations in flora, in fauna, in In- vertebrate species and Vertebrate species was variously discussed. The weight of opinion appeared to favor the view that Verte- brate .species, when present, afforded the best evidence as to chronological relations. Prof. Zittel gave the highest place to Vertebrates. An abstract of Prof. Marsh’s remarks is contained in the paper on page 336. The next meeting of the Congress, or that of 1894, will be held in Switzerland, probably at Berne; and, on special invitation re- ceived from the Geological Survey of Russia, the following, in 1897, will probably be held in St. Petersburg. The party for the excursion to the Yellowstone Park, Colorado, etc., included about eighty members of the Congress, of which more than half were those from abroad. The following ladies were of the number: Miss Mary Forster of London, Mrs. Mary 344 Scientific Intelligence. Caroline Hughes of Cambridge, England, Madame Marie Pavlow of Moscow, Madame Henriette Sihleano of Bucharest, Roumania, Madame Maria G. Stefanescu of Bucharest, Mrs. 8S. F. Emmons, Miss Mary G. Markoe and Miss C. A. Smith, of Washington, and Mrs. H. 8. Williams of Ithaca. Through the Park the party has the special guidance of Messrs. Hague and Iddings. The excur- sion will occupy 25 days ; or for those who go also to the Colorado Cafion, an additional ten days. The latter trip is under the di- rection of Major Powell. $20.00. - + a Pacific R. R. Survey, 13 vols. $12.50. * fe American Journal of Science and Arts, 138 vols. > $975 00.) ae Hayden, U.S. Geological Survey. Monographs, 12 vols., 4to. _ ahs eae Bas ea Wheeler’s U.S. Geological Survey, 8 vols., 4to. $20.00. U. S. Geological Surveys. Annual Reports, 10 vols.,; 410; Sg $12.50. cik Baird, Cassin & Laurence. Birds of North America, 2 vols. ~ $5.00. ie Holbrook, North American Herpetology, 4 vols. Pes ye | cae Agassiz. Contributions to Natural History of U.S., 4 vols., 4to. $25.00. es Popular Science Monthly. Complete set to 1890. ae $35.00. — Humphreys & Abbot. Physics and Hydraulics of the Mississippi, 20 plates, % i Se 4to. “$3.50, 0 | American Naturalist. Complete set to July, 1891. $60.00. Elliott, Botany of South Carolina and Georgia, 2 vols., 1824. $7.50. Nature. Complete set to 1890. . $50200." =) a Pennsylvania Geological Survey, 100 vols. | $35.00. Baird, Brewer & Ridgway. North American Birds, 3 vols., 64 plates, 593:\1 BS illustrations, Ato, 1874. $20.00. >. 7 Bischoff, Chemical and Physical Geology, 3 vols. 1854-1859. $10.00.) am Lowe and Howard. Beautiful Leaved Plants. 60 colored plates, 1872, S580. Sowerby, Recent and Fossil Shells. 264 plates, 1825. $15.00. Torrey, Botany of New York. 161 plates, 4to. $7.50. : Shaw & Stephens. Zoology, 28 vols. 1200 plates, fine eit $20.00. Hayden, Geological Atlas of Colorado. Folio, 1878. $3.50. King, Exploration of 40th Parallel, 9 vols., Ato, and folio complete. $40.00. — _Pursh, Plants of North America. 24 colored plates. $10.00. Coffin, Winds of the Globe, 4to. $5.00. Lowe, British and Exotic Ferns, 8 vols., 479 colored plates, 1880, $20.00. CHRISTMAS PRESENTS. The systematic collections that we put up are very attractive as well as instruc- + tive. The hard wood boxes add much to the desirability of the collections for - es presents to young people. Many a child might have its tastes turned to natural — Riepal history by even the 50c. collection which is very pretty. coe oe i NUMBER OF SPECIMENS. | in cae fa ey free 100 | 200 300 te ey ~ he saat oe : > a of & % Crystals and fragments, 34in.......... $ 50 | $1 00 $2 00 $1 00 | $2 00 $4 00 mat. Student’s size, larger, ve x gin ele | 1 50 3 00 6 00 | 5 00 10 00 20.00," 7 AMATCUE SiSiZOr AveRe eA 8.5.46, Scaide bone niare tale aoe chai cee eerie 10 00 20 00 45 00 ae High School, or Academy size, 24% x 376 | diss Shelf Specimens........ VAs 25,00" 50 00 125 00 ae College size, 34% x 6in., Shelf Specimens Selah seualelleveiate beta tetat Oita Mbaee Cie roe | 50 00 100 00 250 00 we, tie Petrified wood (described last month), garnets from Alaska, Salida and many ——t™s other localities, opals from Mexico and elsewhere and many beautiful Species for: presents for collectors can be supplied. Meteoric Iron from Cafon Diablo in complete pieces from 25c. upward. This is the cheapest and most interesting meteoric iron ever sold. oa . Matlockites, Phosgenites from England. We have received from the cblen: Ais tion of a gentleman recently deceased ‘at. the locality the finest specimens ever offered for sale in this country. They are priced lower than the price paid the men at the locality. Amnglesites from the same locality and sme Te es: = Fresno, Utah, just determined. : ? A; EY FOODE, 4116 Elm avenue, Philadelphia, Pa., v. s. A ! Mhib-wed, UalesYy, THE “AMERICAN JOURNAL OF SCIENCE [THIRD SERIES] Oe Art. XX XIV.—The Solution of Vulcanized India Rubber ; by Cart Barus. 1. InrRopuctrory.—In my work* on the solubility of glass in water, I showed that in proportion as the state of dissocia- tion or the molecular instability of glass is increased with rise of temperature, the solvent action of water increases at an enormously rapid rate; that inasmuch as the solution takes place between a solid and a liquid, sufficient pressure must be ~ applied to keep the fluid in the liquid state, whenever the vapor tension at the temperature in question exceeds the atmospheric pressure. Thus, at 100°, the action of liquid water on glass is nearly negligible; but even at 185° solution occurs at so rapid a rate that capillary tubes may become filled with solid hydrated silicate, in place of water, in an hour. Here, however, about 10 atm. must be applied to keep the solvent in the liquid state essential to speedy reaction. 2. The present application.—Uaving attempted to apply the same principle to the actual solution of vulcanized India rub- ber, I obtained confirmatory results at once. To my knowl- edge this material has not heretofore been advantageously dis- solved in a volatile reagent, or in any reagent by which it is speedily and copiously taken into solution, and from which it may be conveniently obtained. Of. $4, note. It follows by analogy from § 1, that the rubber must be hot enough to be in a state of dissociation, i. e., that the coherence of the rubber-sulphur molecules must show an instability in regard to whatever solvent may be used. It follows, more- * This Jour., xxxviii, p. 408, 1889. Ibid, xli, p. 110, 1891. Am. JoUR. Sci.—THirRD Series, VoL. XLII, No. 251.—NOVEMBER, 1891. 25 ae sas a ae +. 360 C. Barus—Solution uf Vulcanized India Rubber. over, that the system of rubber and solvent is to be kept under pressure suflicient to insure the liquid state of the solvent. It follows obviously that this temperature must only be so high, ceet. par., as to change in the least degree possible, the useful character of the rubber eventually to be deposited from solu- tion. Hence, I act on vulcanized India rubber at the lowest convenient temperature facilitating the solvent action, and at a pressure preferably exceeding the vapor tension of the solvent at the given temperature. Whatever other favorable action pres- sure may exert (such, for instance, of forcing the fluid into the physical pores of the semi-solid by a principle akin to Henry’s law) is clear gain. In my machine* it is rather more conven- ient to act under 100 atm., or more, than at lower pressures. Hence I did not scruple to use pressures as large as, or above, this, testing the adequacy of low pressures, however, by special experiment. §§ 18, 20. The samples of vulcanized rubber acted on were five in number, and their character may be detailed as follows: a. Very elastict sheet rubber, usually not pigmented, trans- lucent in thin films, brownish in color, used for rubber bands and sheeting, chemical rubber tubing, etc. b. Less elastic and harder rubber, pigmented gray, opaque, largely used for rubber tubing, etc. c. Non-elastic, pigmented rubber, flexible, opaque gray, used for low class rubber tubing and low class merchandise in general. d. Ebonite. e. Same as a, rotted by age and exposure. 3. Solution in carbon disulphide.—F rom experiments made at 100° and 160°, it appears that elastic sheet rubber (q@), is not fully soluble in OS, in a reasonable time, if at all. It is quite soluble at 185°, and soluble to a remarkable degree and at a remarkably rapid rate at 210°. Hence the pressure under which solution is to take place, should here be greater than * See Proc. Am. Acad., xxv, p. 93, 1890, or Phil. Mag, October, 1890, p. 338. The present method of work is simple: Glass tubes 10°™ to 15°™ long, and -3°™ or ‘4°™ in diameter, closed at one end, and drawn out to a capillary canal with three enlargements at the other, were filled with a charge of vulcanized rubber and solvent, and then introduced into the steel piezometer tube. JI made use of the temperatures of boiling turpentine (i60°), aniline (185°), naphthalene (210°), and diphenylamine (310°). To separate the charge from the oil of the piezometer which transmits pressure, I first employed a thread of mercury inserted into the capillary canal. Finding, however, § 17, that this metal acted on the charge, I replaced it by a thread of water, or contiguous threads of gasolene and water, Charges were usuzlly introduced in the ratio of one part by volume of rubber to three or more of solvent, § 18. About 1°¢ to 2°¢ of solution were obtained per heating. I made considerably over 150 experiments, most of them at 210° and 100 or 200 atm. Experiments on a larger scale were also made in great number, chiefly with the object of studying the product deposited from solution, § 20. + By elastic I mean extensible with resumption of the original shape when the pull ceases. C. Barus—Solution of Vulcanized India Rubber. 361 15 atm., but need not exceed 30 or 40 atm. Inasmuch as CS, thus unites with rnbber in any proportions, clear brown solu- tions of any viscosity may be obtained. Diluting such (thick) solutions with cold CS,, the solvent is first greedily absorbed ; but the final complete solution of the unagitated syrupy rubber takes places very slowly. Finally, by exposing any of the solutions to air, the OS, evaporates, and the dissolved vulcan- ized rubber is regained without sacrifice of its original non- viscid quality. Similarly fissured brittle sheet rubber or tub- ing (e), which has become useless for practical purposes by age, is quite soluble in OS, at 200°, so far at least as its undecom- posed portion is concerned. Elastic gray rubber (6), dissolves completely to a gray liquid, in which the pigment is suspended. $16. The concentrated solution hardens at once on exposure to air, reproducing a rubber of nearly the qualities (6). The same is true of the non-elastic sample (c). Treatment at 310° resulted in a decomposition of the rubber. Commercial ebonite (@) is first partially devuleanized at 200°, (excess of rubber) and eventually dissolves in excess of solvent. The partially devulcanized product is elastic on drying, but finally hardens to a tough solid having a leathery quality. The solution leaves a black stain, with free sulphur apparent after evaporation. § 15. Gases are frequently evolved during solution of highly vulcanized rubber in OS,. §$17 and 19. As a whole my experiments show that excess of sulphur is first removed by the solvent, after which the vulcanized rub- ber itself passes into solution. § 15. 4. Solution in liquids of the paraffine series.—The elastic rubbers (a and ¢) dissolve easily in liquid mineral oils, at 200°. The pressure necessary will, of course, vary with the boiling point of the oil used, and may be as high as 50 atm. in the very volatile gasolenes. Commercial gasolene, though a good solvent of the rubbers « and ¢, is less powerful in ease of } and c, unless excess of solvent be used. On exposure to air, the gasolene evaporates, leaving a residue which soon hardens. Mineral oils of a higher carbon order than gasolene, petroleum,* for instance, dissolves the rubbers @ and ¢ even more easily. The solution, however, dries only after much time and proba- bly only in thin films. Solubility seems to increase as the oil lies higher in the carbon series. §§ 12, 18. * Looking up the literature of the subject, I found that John J. Montgomery (Cf. Letters Patent No. 308,189, November, 1884, U.S. Patent Office), describes a process for the solution of vulcanized rubber. His statement of the tempera- ture and pressure necessary are substantially correct, although he confines his experiments to a petroleum oil boiling at 200° or higher. The oil is subsequently driven off by injections of steam. This is the nearest approach to an available and true solution (a solution which does not remain permanently sticky like the turpentine and other solutions) which I have found. The essential peculiarity of the methods in the above text is solution in volatile solvents. 362. =O. Barus—Solution of Vulcanized India Rubber. 5. Solution in turpentine.—In case of the elastic rubber (a), complete solution is at once effected at 200°, whereas at ordi- nary temperatures the time necessary is enormous, if indeed the solutions in the two cases be the same. The syrupy liquid obtained at 200° seems to dry in very thin films. Special experiments made at 160° showed that no reasonably speedy solution takes place even in liquid turpentine at this tempera- ture, thus corroborating the inferences of §§ 1, 2, 38. Gray rubber (0) is acted on with greater difficulty at 210°. The solution leaves a white glossy stain which hardens. Pressure need not exceed 5 atm. 6. Solution in chloroform and carbon tetrachloride.—Elastie sheet rubber (a) dissolves at once in liquid CHCl, at 210°. Pressure should exceed 15 atm. and need not be larger than 25 or 30 atm. Solutions of any degree of viscosity seem to be obtainable. They dry at once on exposure to air, leaving a hard residue relatively dark in color. Possibly this was due to the presence of sulphur in the chloroform. §15. Gray rub- ber (0) is attacked with decomposition of the solvent and evo- lution of gas. 7. Solution in aniline.—Solution in the liquid at 200° takes place at once, in case of elastic rubbers (a). Pressure need not exceed a few atmospheres. ‘Thin films apparently dry on long exposure. 8. Solution in animal oils.—Neither in the case of sperm oil, nor of lard oil was the elastic rubber (qa) dissolved on removing from the piezometer. Both distintegrated on stand- ing, to a solution, often with slow evolution of gas. 9. Treatment with glycerin.—At 200° no solution occurs. Glycolic alcohols were not examined. Cf. § 12. 10. Solution in benzol and higher aromatic hydrocarbons. —The elastic sheet rubber (a) dissolves at once in liquid C,H, at 200°. Pressure should exceed 7 atmospheres, but need never be higher than 30 atm. The solution exposed to air hardens rapidly. Solution of gray rubber (6) is less easy. Solution of elastic rubber (@) in liquid toluol at 200° also takes place with great ease. The liquid dries slowly. Pres- sures of less than 10 atm. suffice. 11. Solution in ethylic and higher ethers.—Elastie sheet rubber (a) dissolves at once in liquid ethylic ether at 200°. Pressure should exceed 25 atm., but need not be greater than 40 or 50 atm. The solution hardens immediately on exposure to air. Gray rubber (4) is attacked with difficulty. 12. Treatment with alcohols—At 200° india rubber (@) is not dissolved in liquid methyl or in liquid ethyl alcohol, and only slightly so in liquid amyl alcohol. Thus, again the solu- bility seems to increase with the molecular weight of the sol- vent. § 18. _C. Barus—Solution of Vulcanized India Rubber. 363 13. Treatment with ketones.—India rubber (a) treated with liquid acetone at 200°, is converted into a sticky paste from which it hardens at once on exposure to air. Pressure should exceed 15 atm., but need not be greater than 30 or 40 atm. 14. Treatment with water and mineral acids.—In no ease was there a trace of true solution at 210°. Water probably enters the physical pores of the elastic rubber (qa), as this sub- stance becomes superficially rough and warty on drying in steam at 200°, after being treated with liquid water at the same temperature. It does not melt. $18. Strong hydro- chloric acid (1:2) has no obvious effect, while strong sulphuric acid (1:3) seems only to char the rubber. Treating gray rub- ber (6), with HCl, I found its solubility in OS,, C,H, and gas- olene to have decreased. 15. Treatment for vulcanization. Liquid ebonite.—Liquid ammonic polysulphide at 185° or 200° does not change the appearance of gray rubber (6) markedly; but the sample loses its elasticity and shows a semi-plastic consistency. This I be- heve to be due to additional vulcanization induced by the poly- sulphide. If now the sample be treated with liquid CS, at 200°, the solvent is decomposed with the evolution of much gas, and the rubber restored to its original elastic quality. The gas is liberated throughout the mass of the rubber, and the sample, when taken out of the tube, has the form of an enor- mously inflated cellular sack, which issues from the glass tube explosively, but soon collapses on exposure to air. As a whole these results agree with the behavior found for ebonite in § 3. In both cases it is possible to pass from a more vulcanized to a less vulcanized solvent by treating an excess of rubber. It will be shown below, § 19, that the gas evolved is probably due to the double decomposition of water and CS, More interesting is the direct vulcanization of a rubber solu- tion, to liquid ebonite, by aid of a solution of sulphur. In case of elastic sheet rubber (a), this even begins at 160°; but it is more complete at 185° and 210°. In ease of pure (non-vulean-' ized) rubber dissolved in CS, with excess of sulphur, scarcely any change of the flesh color is observed at 160°, and the sul- phur crystallizes out of the solvent in aeedles, on exposure. At 185° and 210°, however, the charge turns black, showing complete vulcanization. If equal masses of vulcanized rubber (a) and sulphur be treated, the product, after heating to 210°, is not dissolved nor soluble, until the excess of sulphur is re- moved. §§ 3,15. Gasis often evolved. §§ 17,19. In pro- portion as less sulphur is used relatively to the rubber, the product becomes more immediately soluble and less gas is evolved. Adding about 20 per cent of dissolved sulphur to the elastic rubber (a), I obtained serviceable solutions of ebo- 364. CC, Barus—Solution of Vulcanized India Rubber. , nite, on treating at 200° either in CS, alone, or in mixtures, § 16, of this liquid with gasolene, benzol, ete. In most cases these harden very quickly to a jet-black enamel. With less sulphur the color is brown in thin films. 16. Solution in mixtures of solvents, and solution of mixed gums.—By acting on vuleanized rubbers with mixed solvents of the above kind, I obtained very satisfactory results. All] the rubbers mentioned (@ to e), ebonite excepted, pass easily into true solution by such treatment. Thus the gray elastic rubber (6) dissolves at once in a mixture of OS, with gasolene, or benzol, or ether, etc.; or of benzol and toluol; or less _ easily in mixtures of benzol and gasolene; etc. Ebonite is partially devuleanized, and would probably be dissolved in _large excess of solvent. § 38. No gas was evolved in any ease, $19, which is an advantage of this method. Im all cases the solutions hardened rapidly on exposure to air, yield- ing the pigmented rubber if the solution be shaken, or a purer rubber, if the sediment be removed by subsidence and decantation. | Equally feasible is the solution of mixed gums in a suitable solvent at 200°. Thus I made solutions of mixed vulcanized rubber and gutta percha in CS,, which dried at once on ex- posure to air; mixtures of rubber and shellac dissolved in CS,, drying more slowly; mixtures of vulcanized rubber and rosin dissolved in CS, and in gasolene, which dried in thin films only after long exposure; ete. 17. Direct devulcanization.—When, by any of the above methods a solution of vulcanized rubber is obtainable, direct devulcanization may be attempted by mixing the charge with some sulphur absorbent. Such material must be chosen which at 200° acts neither on the rubber nor the solvent. Metallic filings do not appear to be available. Treating ebonite with CS,, C,H,, or gasolene, to which copper filings had been added, I found the charge, after exposing to 200°, to be disintegrated, while an enormous amount of gas was evolved. Scarcely any of the solvent was left in the tube. The direct action of copper or of sulphur, on OS,, ete., at 200° is insufficient to account for this reaction. $19. The gas must, therefore, be produced at the expense of the ebonite, or of the reagent in presence of ebonite; and since all the solvents used behave alike, at the expense of the ebonite. This may furnish some clue as to the chemical character of the rubber as related to the gases evolved. Gaseous decomposition frequently sets in on exposure of highly vulcanized rubber solutions even to ordinary room temperatures, whereas at 0° and under slight pressure (1 or 2 atm.) the gas remains in combination. Bright steel is —— 0. Barus—Solution of Vulcanized India Rubber. 365 searcely attacked.* In fusing impregnated india rubber, § 18 I frequently noticed that the colder ends of the mass were opaquely discolored. Possibly, therefore, the sulphur at 200° may be gradually segregated by diffusion or evaporation. My experiments on this subject failed. 18. Fusion of impregnated rubber.—If vulcanized india rubber be impregnated or saturated by digesting it with the cold reagent (any solvent of pure rubber), for a suitable time (a few minutes to many hours), the swelled mass not only shows a relatively low melting point, but it remains liquid after cooling, provided the solvent is not allowed to escape. This is an observation of practical importance, since the re- tortst can thus be charged with solid or dry rubber, a minimum of solvent be used in treating or lost by evaporation, and con- centrated solutions be obtained often fit to be used at once. The rubber so melted hardens on exposure. Finally the pres- sure necessary in this case is the smallest possible, and may be below the data given for the divers solvents above. The quantity of solvent retained by solid rubber is very large: Thus elastic sheet rubber will hold 7 or 8 times its weight of CS,, or 1 to 2 times its weight of naphtha. Gray rubber (elastic) absorbs more than its weight of naphtha; ete. Experiments may be cited as follows: Non-impregnated vulcanized rubbers (@ to e) do not melt if exposed in a closed tube at 210°. Only in the case of very slightly vulcanized pure rubber gum is there a trace of fusion perceptible at the edges, and here it may even be due to a stain of dirt (oil) acci- dentally left there. Gray rubbers (6, ¢) with a superficial coat- ing of exuded suJphur, turn black from the formation of a film of ebonite. All the india rubbers (@ to e) fuse at 210°, when previously saturated, or nearly so, with cold carbon disulphide, and exposed in a close-fitting glass tube. Jf the pressure be reduced by a capillary aperture at one end of the otherwise closed glass tube, or if the tube be only partially filled and the empty end kept cool, the impregnating solvent is merely distilled off, and no fusion takes place. Whereas at 160° fusion scarcely occurs, melting seems to be complete in the well impregnated elastic rubber (a) at 175°. There is therefore an approximate coinci- dence of the thermal data in the present and in the above paragraphs. * Fortunately, therefore, steel apparatus is available on a large scale. An interesting question occurs as to what becomes of the carbon, in the case where sodium, mercury. copper, etc., are attacked by hot liquid CS. and not by cold C&p. + The present experiments were made in closed glass tubes. nearly filled with the impregnated rubber. After fusion the mass frequently appeared to have shrunk. Cf. § 2. 366 OC. Barus—Solution of Vulcanized India Rubber. Similar results were obtained with benzol, with gasolene and higher petroleum oils, etc. Fusion is absent or only incipient at_ 160°, and more than complete at 210°, provided the gasolene be not too volatile. §§4,12. In general the gray rubbers (6, c) fuse to a more viscous mass than the gum rubbers (q@), the consistency of cold solutions in the latter case about that of treacle. The occurrences of this paragraph therefore would resemble the fusion of a salt in its water of crystallization, but for the exceptional behavior that impregnated vulcanized rubber after fusion retains a consistency which is liquid relatively to the original non-impregnated charge. The analogy with the solu- tion of starch, or of gluten, is thus more close and immediate. In all these cases the solid swells up when impregnated with the solvent, and fuses to a relatively less viscous consistency, or to a thin solution, when a certain temperature (below 100° in ease of starch and gluten and above 160° in case of vuleanized india rubber) has been reached. Hence it is not unreasonable to suspect that even ordinary dry wood, or woody tissue, which swells to a marked degree when impregnated with water, may pass into actual solution if the temperature at which the water acts is sufficiently high, and the pressure above the vapor tension of water at that temperature.* I mention finally that the reduction of melting point pro- duced in vulcanized india rubber by the impregnating reagents may perhaps advantageously be discussed in accordance with Raoult’s law; but owing to the difficulty of defining the melt- ing point of the unimpregnated rubber, and the close prox- imity of the melting points after impregnation with different reagents (CS,, C,H,, gasolene) my views on this subject have not taken shape. It is known that in general that the melting point produced by a dissolved colloid is relatively very small, from which an exceedingly large molecular weight of the colloid has been inferred. The above results show that in the converse experiment, where the melting point of the colloid is lowered by a solvent, the effects will probably be normal and pronounced. Nevertheless [ doubt whether the thin rubber fluids obtained are true solutions, i.e. represent a case in which the division of the solid has actually reached a definite molecule; for on * T have since tested this surmise at some length, but found in every case that cellulose is decomposed before solution in water takes place. In spite of the presence of water under pressure, the phenomenon seems to be a dry distillation. [ may here refer to the remarkably close analogies in the thermal behavior of rubber and gelatine which have recently been discovered by Bjerken (Wied. Ann., xii, p. 817, 1891). The author has reason to believe that moist gelatines are heterogeneous mixtures of solid and liquid. The behavior of rubber, as discussed above, is characterized at low temperatures by a fixed maximum of absorbed sol- vent. The term mixture is scarcely applicable at once. OC. Barus—Solution of Vulcanized India Rubber. 367 long standing in sealed vessels a gradual thickening of the liquid with final coagulation seems to be the invariable result. Thus there must be a gradual growing together of the indi- vidual particles, until tinally the whole solution forms one co- herent gelatinous mass. To summarize: Suppose the coherence of rubber to be due to (cohesive) affinities, capable of being saturated like ordinary affinities. Then in case of impregnation with a solvent, a part of these combine with the similar affinities of the solvent. The result is the decided decrease of tenacity (observed). To liquefy the impregnated sample, the residual cohesive forces of the rubber must be withdrawn, and this can be done by heat. The liquid so obtained, I do not conceive to be a true solution, but rather a suspension of particles, the exceeding fineness of which is determined by conditions discussed elsewhere.* Dit- fusion is thus an excessively slow process, and hence the liquid on cooling need not become solid again. In proportion as the individual particles unite however, coagulation gradually sets in, and its structure is probably that of a fine sponge hold- ing solvent in its interstices. If the coagulated solution be reheated (under pressure), a thin viscid solution is again ob- tained, which in its turn coagulates. 19. Behavior of reagents.—The frequent occurrence of gaseous products in the above experiments made special ex- periments on the decomposition of reagents necessary. Beuzol and gasolene were found stable at 210°, and often above this temperature, both in the presence of water, or of sulphur. $21. Carbon disulphide, however, in addition to relatively slight decompositions producible by sodium, or mercury, or copper ($17), at 210°, is doubly decomposed by water at this temperature, with the evolution of much gas, presumably H,S and CO,. CS, remains stable in the presence of zine white (a common rubber pigment), or of sulphur, or of bright steel, at 210°. §17. Hence a thread of mercury to shut off the ex- * perimental tubes, § 2, is generally objectionable, as is also a thread of water in case of CS,. Moreover the absence of gase- ous reaction in case of mixed solvents, § 16, is to be attributed to the fact that CS, and the water are intentionally separated by layers of benzol or gasolene. An interesting question is suggested here, as to whether it be possible to express affinity on a scale of temperatures. Let it be required to determine the affinity of a metal for sulphur. At ordinary temperatures not even sodium decomposes C%,, whereas such decomposition occurs if the temperature be sufi- ciently high. Hence the temperature at which the decompo- sition definitely sets in (for copper sooner than for iron, etc.) is * Barus: this Journal, xxxvii, pp. 126-128, 1889. 368 0. Barus—Solution of Vulcanized India Rubber. a reciprocal expression of the affinity of the given metal for sulphur,—bearing always in mind that the stability of the solid metallic molecule also enters into the consideration. The arbitrary reagent OS,, in its relations to all the metals to be examined, fulfills a similar purpose to an arbitrary spring balance in measuring gravitational forces. § 21. 20. Summary of the results.—In the above paragraphs I have therefore indicated a method by which vulcanized india rubber of any quality or character whatever, as well as the undecomposed or reclaimable part of rubber waste, may be dissolved or liquified in a reasonably short time ;* the solutions possessing any desirable degree of viscosity or diluteness, from which india rubber may be regained on evaporation of the solvent. I shall elsewhere describe divers forms of apparatus by which the above operations may be carried out on a larger scale. They are of no interest here; but I mention them since it is only from such work that a full insight into the quality of the rubber deposited from any given solution may be obtained. Experiments made in bulk in this way showed the material deposited from solution to be considerably inferior to the orig- inal rubber, both as regards tenacity and elasticity. Its chief value in the physical laboratory will therefore be that of fur- nishing an air-tight cement or an acid-proof varnish, capable of withstanding more than 200° centigrade. Rubber newly deposited from any of the above solutions presents a very curious case of slowly reacting elasticity. Ifa thread, say 0-1 em. thick, be twisted and then let go on a frictionless surface, it will squirm like a live worm for some minutes. If it be stretched, the original length is regained with visible slowness. Throughout my work the approximate constancy of the dis- solution temperature irrespective of the solvent has been the marked feature. Thus in case of CS,, of turpentine, of the ~ vulcanization of dissolved pure india rubber, etc., no action took place below 160°. Even this temperature is higher than is needful for vuleanization effected in the dry way, where 110° to 140° are deemed sufficient. Moreover the solution of vuleanized rubber in OS,, for instance, takes place quite as easily under 700 atm. as under, say 20 atm., as is particularly manifest from the fusion of impregnated rubber, and in special high pressure experiments. In my work on the compressibility of liquidst I showed that compressibility is essentially asso- ciated with the extra-molecular forces whereas the molecule remains relatively incompressible. Temperature, however, has immediate access to the molecule; and thus it follows that * Practically at once, if the material is not too bulky. + This Journal, xxxix, p. 510, 1890. A. M. Edwards—Infusorial Earths of Pacific Coast. 369 whereas the effect of temperature in experiments like the above is manifest, the effect of pressures of the order applied is relatively inappreciable. 21. Digression.—From the above (eens I infer that the difficulty encountéred in endeavoring to dissolve carbon is probably attributable to a relatively high dissociation tempera- ture of the solid carbon molecule. I made many experiments to test this view, in all of which I failed to obtain solution even at low red heat and 600 atm. of pressure. My work thus corroborates the negative results of Hannay* on the direct so- lution of carbon. My tests were made with gasolene, water, benzol and carbon disulphide, usually at 500° and 500 atm. In ease of gasolene I observed at higher temperatures. Usu- ally the reagents were decomposed (particularly Cs,, C,H, and OCHCl,) with the evolution of much gas, while the carbon re- mained appreciably unaffected. Decomposition by metals (copper corroded by CS,, and gasolene acted on by palladium) showed sooty deposits only. Art. XXX V.—Leport of the Examination by Means of the Microscope of Specimens of LInfusorial Karths of the Pacific Coast of the United States; by ARTHUR M. ED- warps, M.D. | SOME time since I had transmitted to me by Mr. George Gibbs, the geologist of the Northwest Boundary Expedition, a collection of earths gathered at different points on the Pacific coast of the United States in the states of Washington, Oregon and California, as well as British Columbia, with a request that I would make an examination of them by means of the microscope, the more particularly for the purpose of determin- ing the characters of the organic remains to be found in them. Through this means I have been enabled to study and record the discovery of several deposits of minute organisms, and at the same time very materially assist in unraveling the geology of some points of the country hitherto found to be somewhat difficult of comprehension. At the time these examinations were made, that is to say, in the latter part of the year 1861, very little was known concern- ing many points in the veology of our Pacific Coast, and my own experience in the study of such earths had been rather slight. Hence, when I made my report in 1862, I was unable * Hannay: Proc. Roy Soc., Ixxx, p. 188, 1880; Chem. News, xli, p. 106, 1880. Cf. Hannay and Hogarth: Chem.’ News, xli, p. 103, 1880; Mallet and Hannay: Nature, xxii, p. 192, 1880. 870 A. M. Edwards—Infusorial Earths of Pacific Coast. to go very fully into the subject of the evident mode of forma- tion of the strata containing the microscopic organisms. Since that time the Geological Survey of the State of California has been undertaken and a much more extended suite of gather- ings has come into my hands. Through the knowledge acquired from the examination of these collections made at various points from Puget Sound to the southernmost border of California, I have been enabled to furnish such information that the history of both the marine and fresh water, so called, Infusorial deposits of that portion of the country has been pretty thoroughly worked out. With regard to the marine strata very little has, as yet, been published. The results arrived at concerning one class of the fresh water strata has been made known in a communication of Professor Whitney’s read before the California Academy of Natural Sciences, February 4th, 1867. (Proc. Cal. Academy, vol. iii, p- 319.) These he has therein shown to be the beds of enor- mous extinet lakes or inland seas, the material of which has been altered in character by the superposition upon it at differ- ent periods of lava or sand and gravel or ashes and pumice. In this way it can be readily understood that, as the volcanic action ceases, anew growth of microscopic organisms might take place over the erupted material lying upon the older deposits and, in fact, that many such layers might accumulate one over the other. Such has been the case at various points upon the Pacific Coast from Puget Sound to Lake Mono in HKastern California, which is the most southern point from which I have received such material. At some future time I may have more to say with regard to this class of deposits, for I have examined many of them during the progress of the Geological Survey of California and, when my report thereon is published, I shall be enabled to go more fully into the sub- ject. As I have several specimens from strata of this charac- ter to describe in the present report I shall, for the time being, indicate them as sub-Plutonic, which is the most distine- tive appellation I can now find for them. | Among the specimens which I have examined in connection with this survey, and aside from those which do not contain any organic remains, and hence will be treated of separately, I have then, first, the sub-Plutonic, which I have just alluded to, and which are always of fresh water origin; second, the fresh water deposits of more recent formation and, in fact, which are now under process of growth all over the world beneath ponds and lakes, and which I have hitherto been in the habit of calling sub- Peat, but I have lately preferred to designate as Lacustrine Sedimentary, as I consider that they are better so indicated. Besides these two classes of deposits, which differ 3 A. M. Edwards—Infusorial Earths of Pacific Coast. 371 from each other only in time and in the fact that in the most recent a certain amount of organic matter usually remains and the material is light and readily pulverulent. We have in the older one, on account of the volcanic heat added to, or with- out aqueous action, the material has had all of its organic mat- ter removed. And it has become a less or more hard rocky mass of a light color. Hence we have strata of a totally differ- ent character. These are of marine origin and of an age sup- posed to be coincident with the Miocene Tertiary. At ail events they are much older than the most ancient fresh water stratum containing Diatomacese as yet discovered. Of the mode of formation of these last mentioned strata I shall not now pause to treat, as | have already thrown out some hints respecting my opinions upon this point in some remarks made before the Essex Institute, Salem, Mass., January 4th, 1869, an abstract of which will be found in the Bulletin of that asso- ciation, vol. i, page 11. I have treated of the same subject in a paper read before the American Association for the Advance- ment of Science, at the Salem meeting, August 25th, 1869. Hereafter I will treat fully of this subject in my report on the microscopic material of the Geological Survey of California now in preparation. It will be readily perceived that it is fortunate that my report on the matter herein treated of was not published at the time it was sent in, and I feel that I can congratulate myself that Mr. Gibbs has again submitted the matter to me for revision, for at the present time I can do more justice to it and throw light upon some points which, at the time, I was unable to fully comprehend. The constantly recurring records of the discovery of fossilif- erous deposits containing the remains of such minute organ- isms as the Diatomacez, Radiolaria and Rhizopoda, constitut- ing the well known ‘Infusorial earths’ of most geologists reveal the fact that these atomies play a very important part in the world’s future; and while almost every newly found specimen exhibits one, if not mrore, of what have been consid- ered new species, it proves, at the same time, the cosmopolitan character of many already known forms, which are thus seen to oceur spread over the globe in great profusion from the equator to the poles. In some eases these widely-spread species will not vary appreciably, be their dwelling place under the burning sun of the tropics, the more equable climate of the temperate zone or the frozen fields of the poles. Other forms, however, on the contrary, appear to vary to so great an extent with every few degrees of latitude that specimens gathered at the equator and in localities a very little removed therefrom, either north or south, might be supposed, on superticial exam 872 A. MW. Hdwards—Infusorial Earths of Pacific Coast. ination, to be distinct. So markedly is this the case that we not unfrequently find that hasty observers have so classed them and even made use of locality for the determination of specific distinctions. That the Diatomacez, which are the organisms with which I shall most particularly treat in this paper, are extremely cosmopolitan in their habits; in fact, perhaps more so than any other group, would seem to be already established, but the imperfect state of our knowledge of them and their life-history at present, leaves us a great deal in the dark as to the full extent of their variation during the. lapse of time or through local distribution. Much has yet to be done in this field of investigation and large and widely extended collections made of both the recent and extinet forms, before we can assert that we know anything very certain with regard to their position in the chain of being, their habits, history, or range of variations in time or space. I do not, at the present time, desire to go more fully into this branch of the subject, merely confining myself to a thorough report upon the specimens submitted to me by Mr. Gibbs. The student who desires to follow researches in a field which will yield profitable returns cannot choose for himself one in which less is known, perhaps, than this, and when its applications to geology are considered, for my part, I can hardly imagine one more enticing. Below I give a list of the specimens sent to me for examina- tion by Mr. Gibbs and which were collected by him during the prosecution of the Northwest Boundary Survey. * Hot spring, Harrison’s lake, British Columbia. Nahchess river, Washington. Alkaline deposit, Similkamen river, Washington. ‘ * Steilacoom creek, No. 1, Washington. *k 66 6 No. 2. 3 * Point Roberts, * Bluff west of Camp Simiahmoo, Washington. * Camp Simiahmoo, No. 1, Washington. * 6¢ 66 No y) ee . 9 * 66 “ No. 3, 6c Winass river, ks * Point Ludlow, es Those localities marked with asterisk (*), are from the west or coast side of the mountain range, while the others are from the eastern slope. This is a point of importance and to be borne in mind as will be shown farther on. The principal point to be decided in examining these specimens was whether they contained any traces of organic remains by means of which their marine or fresh water could be determined. There- fore they were first superficially examined so as to note if any —— A. M. Edwards—Infusorial Earths of Pacific Coast. 373 such remains appeared and those that showed signs of yielding definite results were set aside for further study after they had been properly prepared. In this way all of the specimens sub- mitted to me were examined. Most of them were found to contain no traces of organic remains by means of which might be ascertained their origin, as desired. The presence of the siliceous skeletons of Diato- maces in any earth, or deposit of any kind reveals at once the fact that such a deposit has formed beneath the surface of water or, if the remains are not evenly distributed throughout its mass, it may have been overflowed by water having Dia- tomaceze living in it. Besides this, it may be also ascertained as to whether it has been thrown down from fresh water or in the ocean. Although this branch of the subject has not received the attention that its importance deserves yet we can with some considerable degree of certainty even determine as to whether the water from which such a deposit was thrown down was a lake, a bog or marsh, an estuary or the open ocean. As the matter comes to be more fully studied and the knowl- edge of facts is increased we shall doubtless be able to deter- mine these and similar points with a greater degree of accuracy. The indestructible nature of these skeletons, on account of their consisting mainly if not entirely of silica, deposited dur- ing the life of the plant in its tissues, preserves for the student of nature a record of former aqueous submergence, and, as their distinctive characters are not very difficult of recognition by careful students we thus have typical forms of organisms to use for the purpose of determining the marine or fresh water origin of any specimen under examination. At the same time it must be remarked that by far the greatest portion of the time that has been spent by most observers on the Diato- macez has been evidently mainly for the purpose of discover- ing new forms rather than ascertaining the life-history or even the distinctive characters of already known species. So that our lists have become but a heterogeneous mass of mere names applied to often accidental, sometimes distorted or even frac- tured specimens. I[ can not too earnestly enter my protest against the recognition of the species-monger as a naturalist ; such observations and records as those [ allude to do not only not advance our knowledge but certainly retard its progress by placing new obstacles in the path of the student of nature. Elsewhere I have spoken more fully on this subject and shown how it is that this branch of biology has fallen undeservedly _into disrepute among scientific naturalists; at the present time I will refrain from saying more than I have already put upon record. 874. A. WM. Edwards—Infusorial Earths of Pacifie Coast. From what has been said with regard to distinguishing the origin of a deposit by means of the minute remains present in it, it will be readily understood that we can thus determine to a certain extent its age, as to whether the overlying water has been fresh, brackish or salt. In the last case we shall find present such oceanic genera as Triceratium, Coscinodiscus, Aulacodiseus, or Actinocyclus. If the source of the deposit has been the shallow water along shore we should expect to find littoral species among which would be some of the Pleu- rosigma or Amphiprora; often, of course mixed with deeper water forms or even fresh water varieties accidentally mixed by being washed down from elevated stations. On the other hand if we find the genera Tabellaria, Cocconema or Himan- tidium to be present, the fresh-water origin of the gathering is established. So a group of mixed marine and fresh-water species would indicate the formation of such a deposit under very peculiar circumstances, but such mixtures are extremely rare. One of the few of this character which I have seen being a gathering of living specimens from the St. Johns river in Florida, which on account of its course being nearly North and South, is so affected by the tides that the marine species of Diatomacege at least are carried up almost to its head. At some future time when the life history of these minute forms is better understood observers will doubtless be able to ascertain from the examination of gatherings of the siliceous skeletons whether they have grown and been deposited in a lake, river or brook, near the level of the sea or at high alti- tudes as well as the fact of the fresh or salt character of the water. In fact [ feel convinced that a time will come when this mode of study applied to deposits generally will reveal many circumstances connected with the formation of most of the strata constituting the available mass of the earth. At the . present time so little is known of certainty with regard to the life-history of the Diatomaceze; the attention of observers having been mainly turned towards the finding of new forms and manufacturing them, when found, into so-called species, that little can be stated definitely with regard to their distribu- tion or habit. For years [ have been engaged in gathering material to illustrate this point and [am in hopes that, as facilities for collection increase valuable information will be accumulated. With regard to the mixture of forms considered peculiar to fresh or salt water respectively, a case of supposed mixing of species in a lake into which the ocean had access at certain periods of high tides is recorded by Dr. Gregory in the - celebrated ‘Glenshira sand,’ as it has been called, and such may have been the circumstances under which this deposit was thrown down for we have an example of a similar phenomenon A. M. Edwards—Infusorial Earths of Pacific Coast. 375 in the case of the Mystic Pond, near Boston, Mass. Here the bed of the pond is much below that of the river which serves as its outlet, so that at the time of high tides the salt water, which on account of its superior density creeps up beneath the fresh water, runs over the bar at the entrance and flows down into the pond, thus mixing the forms of life found therein. An account of this locality with a list of the forms of Diato- maceze observed in the mud brought up from the bottom of the pond by Messrs. Greenleaf and Stodder will be found in the Proceedings of the Boston Society of Natural History, vol. viii, page 119. So also I have examined a locality of a like kind upon Phillips’ Beach between Swampscott and Marblehead, Mass. Herea small mass of fresh water fed at uncertain periods by intermitting streams, by drainage or by infiltration of water through the beach sand, by which the salt is removed, lies a short distance within and at a lower level than the shore and in it grow many fresh water plants and are found several fresh water animals. Yet at times of high tide or during storms the salt water must find egress, for in it I observed marine species of Diatomaceze in the mud taken from the bottom and, in fact, some few were noticed living in the water of the pond. The microscope thus applied to geology, in the hands of experienced and competent observers, besides the above, re- veals the fact as to whether a gathering under examination be of recent origin, deposited in a pond, Jake, river, marsh, bay, or ocean in existence at the time; or contain mostly extinct forms or be situated in time below the alluvial, and hence to be classed among the truly fossil strata. So that by means of such an examination we come to classify specimens containing Diatomaceze according to the age or mode of occurrence of these forms, and I have provisionally grouped my gatherings into, first: Recent, both marine and fresh water; second, Lacustrine sedimentary, now forming, although in many cases dating their period or origin as far back as the Post-Glacial. The recorded occurrences of similiar deposits of fresh water forms in the Tertiary I consider extremely doubtful; third, we have then the deposits to which I have given the distine- tive title of snb-Plutonic and the mode of occurrence of which I have alluded to above; fourth, thereafter and lastly we have the true Marine Fossiliferous strata which, as far as recorded, have been found only in the Lower Miocene Tertiary. A subdivision of some of these groups is convenient; as, for instance, the recent gatherings may be so arranged as to indi- cate the peculiar habitat of the species contained in it; the so-called ‘natural leathers” and “paper ;’ the soundings from Am. Jour. Sc1.—TuHirp Series, Vou. XLII, No. 251.—NovemBeEr, 1891. 26 376 A. M. Hdwards—Infusorial Earths of Pacifie Coast. the sea-bottom or shell cleanings, as well as harbor muds, the contents of the intestines of marine and fresh water ani- mals and the like be indicated. However, I think that nearly all gatherings may be fairly grouped under the four heads I have adopted. As the tendency of most persons who have turned their attention to the Diatomacez, which are the organisms I shall more particularly consider in this report, has been towards looking for differences where similitudes should have been searched after, | must be permitted to say a few words on that point. The progress of time, the more especially if it be very much extended, may, and in fact will, so change the apparent characters of. all living organisms that they can hardly, in the present condition of our knowledge, be distinguished one from another; but they will most assuredly revert to the parent type, even if the modifying influence be continued in power, so strong, so persistent, so fundamental is the inherent germ- force implanted in the individual. The Diatomaceze are not so liable to be influenced by outward circumstances, apparently, as some other groups ; but, at the same time, understood energies do affect them very materially, so as to change their outline, for instance, leaving their main characters of sculpture intact. I very much doubt if time has as great or as lasting an effect in causing such modifications as locality and, therefore, must consider the use of this point as a basis for distinguishing species to be unscientific and unjustified, at least with regard to these organisms. Among the specimens I have to report upon herein, we have examples of all of the four groups I have adopted, as Recent, Lacustrine, Sedimentary, and sub-Plutonic, under which head are to be placed the tripolis of commerce and Marine Fossil strata. The first Lacustrine Sedimentary deposit discovered in this country was that found by the late Prof. J. W. Bailey at West Point, N. Y., and was described by him in volume xxxv of this Journal. Since that time similar deposits have been discovered at many widely separated points in this country and in Europe, which bears out the opinion expressed by Prof. Bailey that strata resembling the West Point earth in general characters would be found under every bog in the country. In Europe such has been the case, as the Lough Mourne, Premnay, Peterhead, Toome Bridge and Mull in Great Britain and others on the Continent bear testimony. After receiving from Mr. Gibbs the collection of specimens I have already given a list of, he also sent three more, and these I shall include here, as they come from the same portion of country as the first. They are marked as below: A. M. Edwards—Infusorial Earths of Pacific Coast. 377 * Shookum Chuck, a branch of the Chihalis river which flows into Gray’s Harbor, Washington. * Colseed Bay, Hood’s Canal, Washington. Pit River, eight miles from Fort Crook, Cal. These may all be supposed to belong to the western or coast slope of the mountain range, although Mr. Gibbs says that that from Pit river, the eastern branch of the Sacramento, may belong to either side. Of the fifteen earths but seven were found to contain the remains of Diatomacez. * Hot Spring, Harrison’s lake, B. C. This consists of a saline mass evidently deposited by the hot spring, but contains no organic remains. Nahchess river, Washington. No organic remains. Alkaline deposits, Similkamen river, Washington. This specimen is of very much the same general character as the first. * Steilacoom Creek, No. 1, Washington. * Steilacoom Creek, No. 2, : * Point Roberts, Washington. * Bluff west of Camp Simiahmoo, Washington. No organic remains. * Camp Simiahoo, No. 1, Washington. A lacustrine sedimentry deposit, containing: Amphiprora navicularis. Gomphonema intricatum. Himan- tidiura bidens. Himantidium gracile. Melosira varians. Pin- nularia major. Pinnularia viridis. Pinnularia mesolepta. Stau- roneis anceps. Camp Similkamen, No. 2, Washington. No organic remains. * Camp Similkamen, No. 3, Washington. A lacustrine sedimentary deposit, containing : Amphiprora navicularis. Cocconema leptoceros. Cocconema lanceolatum. Cymbella?. C.?. Gomphonema (Pinnularia) amphioxys. Gomphouema olivaceum, Himantidium arcus. Hi- mantidium biceps. Himantidium bidens. Himantidium ?. Navicula elliptica. Navicula cuspidata. Navicula amphigom- phus. Nitzschia (Synedra) spectabilis. Orthosira distans. Pinnularia gigas. Pinnularia dactilus. Pinnularia nobilis. Pin- nularia mesclepta. Pinnularia viridis. Pinnularia tabellaria. Pinnularia Johnsonii. Pinnularia ?. Stauronies phcenicenteron. Surirella craticula. 3878 A. MW. Kdwards—Infusorial Earths of Pacific Coast. Amphiprora navicularis is the one Ehrenberg has given that name to and is quite common in lacustrine sedimentary de- posits in this country although I do not remember ever to have seen it anywhere else. The form I have called Watzschia spectabilis evidently belongs to that genus and appears to be identical with Synedra spectabilis C. E. Wenish (Syn. Brit. Diat. 1853, 1389), who describes a form as WVetzeschia scalaris W.S.., thus claiming the authorship, although he gives Synedra scalaris as the original form and Kiitzing as the founder. The fact is that Synedra scalaris was founded by Ehrenberg (Amer. 137, IJ, ii, 18) and his form was from freshwater at Surinam, and Andover, Conn. Llano Co., Texas. Ville- Johanb- . : b. Marietta neuve P. georgen- Hidden and South Quebec, stadt, Hillebrand Mackintosh. Carolina. Canada. Saxony. UO; 44-17 46°75 83-955 41°06 59°30 UO, 20°89 19°89 34°67 22°33 Tho, - 6°69 757 1:65 6°41 ) ZrO» 0-34. 0-20 we | CeO, 0°34 0-19 ‘40 & none. La group. 2°36 2°05 Detsl ts Y group. 9°46! Ie 2 6°16 2577 J CaO 0°32 0°41 aa) 1:00 PbO 10°08 10°16 Be te! 127 » 6:39 H.O 1-48 2-54(ign.) undet. 1-47 3-17 N 0°54 a 0°86 0°02 $i0, 0°46? 0-20 0-19 0°50 Insol. 1-478 1°22 0:13 Fe,03 0°14 0°58 tr. 0°10 0°21 x ye 0-099 570310 98°74 39°93 98°39 100°72 97°95 Sp.G. 8-29 8-01 6°89 1At. wght. 111°4. 2From thorogummite. 3Mainly fergusonite. 4 At. wght. 1242. .5AsU;0;,. 6Atomic weight 113°6. The oxalates of this group were white, not pink like those in Ia, but the color of the ignited oxides was the same in both cases and very light. 7 Atomic weight l1l°2 approximately. § MgO,Na,.O. 9 Bi.O3. 10 Includes: Al,03(?) 0°20, Bi.O; 0°75, CuO 0°17, MnO 0:09, MgO 0°17, Na,.O 0°31, P.O; 0°06, As,O; 2°34, V,.05, WOs, MoO;(?) 0°75, SO; 0°19=5°08. No. Ia isa re-analysis of nivenite from Llano County, Texas, the material for which was kindly given by Mr. W. E. Hidden. It agrees in the main with the original analysis of this variety by Hidden and Mackintosh,* which is reproduced under Id, and it confirms the presence of nitrogen, suspected but not proven by them. A small remnant of their original powdered sample gave me 0°52 per cent of nitrogen. In a the earths appear in slightly greater total amount than in 6 and they are more subdivided into groups and elements, which accounts fully for the difference between the atomic weights of the metals of the yttrium group of the two analyses. It was rendered certain by a second test that a group of earths whose | * This Journ., III, xxxviii, 1889, p. 481. AM. JOuR. SCI.—THIRD SERIES, Vou. XLII, No. 251.—NOVEMBER, 1891. 27 392 W. EF. Hillebrand—Analyses of Uraninite. sulphates are insoluble in potassium sulphate other than those of Th, Zr, and Ce is present.* A very satisfactory turmeric paper reaction for zirconia was obtained in this analysis as also in that next following, which would go to show that the hypo- thetical ZrO, of several of my earlier analyses was probably in fact zirconia. The cause of the considerable loss shown by the analysis is not known. It may be mentioned that nivenite is more soluble than any uraninite heretofore examined by me, not even excepting cleveite. One hour sufficed for complete decomposition in very dilute sulphuric acid (1H,SO, to 6H,O) at the temperature of boiling water. | No. II is from a new locality, Marietta, Greenville Co.,. South Carolina, and the total amount found, a few small frag- ments, was kindly given by Mr. W. E. Hidden for examina- tion. It was impossible to free the least altered portions from the yellow and orange alteration products with which they were intimately commingled, therefore the analysis represents the composition of a mixture. Unfortunately also the portion in which UO, and N were to be estimated was lost, but it was seen that the mineral was very soluble and gave oft consider- able gas. From the preponderance of the yttrium group over the other rare earths the mineral is to be classed with nivenite and cleveite rather than with those varieties rich in thoria, a a conclusion already foreshadowed by its ready solubility. No. III is an analysis of uraninite from the Villeneuve mica mine, Township of Villeneuve, Ottawa County, Province of Quebee, Canada. To Mr. G. C. Hoffmann, of the Canadian Geological Survey, who first recognized and reported this occeur- rence,t I am indebted, for the material analyzed. It was evidently somewhat altered and was accompanied by oxidized alteration products. Hoffmann gives the density of a piece as 9:055. Crystalline form was lacking, but it unquestionably belongs to the crystallized uraninites, being found like most if not all of them in coarse granite (pegmatite). No. IV represents the composition of a specimen from Johanngeorgenstadt in Saxony, received from Mr. A. Lésch, of St. Petersburg, through Mr. E. A. Schneider of the U. 8S. Geological Survey. Notwithstanding the altered and crumb- ling character of the specimen it is proper to publish the analysis, since the only one previously made that has come under my observation, by Pfaff in 1822, is very incomplete. By panning, a very fair article as regards visible impurity was * It may be here remarked that the subdivision of the earths into the groups indicated by (La,Di),O, and (Y, Hr).Os3 in all my former analyses should not be taken too literally. By the former is meant those earths insoluble in potassium sulphate and by the latter those soluble in that reagent. + Annual Report Can. Geol. Sur., vol. ii, 1886. Report T, p. 10. W. F. Hillebrand—Analyses of Uraninite. 393 obtained. It is not known wherein the loss is to be sought. Like the great mass of the Bohemian mineral this showed no evidence of ever having- been crystallized, and as in that also rare earths are absent, and also nitrogen except for an uncertain trace. From the analyses of uraninite thus far made it appears that the species may be broadly divided into two groups, the one of which is characterized by the presence of rare earths, the other by their absence. With the former group nitrogen appears to be invariably associated, while in the latter it is present, if at all, only in minute quantity. Besides these chemical differences there is one of another kind, for probably all varieties of the first group occur in more or less well defined crystals, while the members of the second group are generally, if not altogether, massive and free from crystalline form. These differences suggest naturally a dissimilarity of origin and envi- ronment. Examination shows that the manner of occurrence and the association of other minerals is different and in such a way as to render an unlike immediate origin probable. All of the rare earth uraninites, with exception of the zirconiferous variety from Black Hawk, Colorado, occur as an apparently original constituent of coarse oranites (chiefly pegmatitic), while the others are evidently of secondary formation, as evi- denced by their presence in metalliferous veins in more or less intimate association with numerous sulphides of silver, lead, cobalt, nickel, iron, zinc, copper, etc. The Colorado variety occupies an anomalous position as regards the two groups. I prefer to regard it provisionally as a member of the second group, where its mode of occurrence and want of crystalline form as well as small percentage of nitrogen seem to place it, although its zirconia and traces of other earths would admit it to the first. Attention is called to the above points merely to show that the chemical and physical differences of the two groups may be susceptible of more simple explanation than would appear from the face of the analyses. Laboratory U. 8. Geological Survey, Washinton, D. C., June, 3894. Rk. &. Cull—Silicified Woods of Eastern Arkansas. Art. XL.—TZhe Tertiary Silicified Woods of Eastern Arkansas; by R. ELLSwortH CALL. (Published by permission of the State Geologist of Arkansas.) THE occurrence of silicified wood in the sands and gravels of the Tertiary of the Lower Mississippi Valley has long been known. Aside, however, from the numerous localities men- tioned by Hilgard,* nearly all of which are in the state of - Mississippi, little attention has been given it. Numerous geolo- gists have spoken of it or incidentally studied it in connection with other investigations, but hitherto no attempt has been made to recognize the species and fix their taxonomic value, if, indeed, they possess any such value. Among those who have investigated the Orange Sands and other Tertiary deposits of the Mississippi Valley and who have added to our informa- . tion as to the occurrence of these fossils are Hilgard,+ Pen- rose,t and Knowlton.§ i The last named has made the only microscopic study of these fossils which is on record. Since his investigations are based upon material which, for the most part, was collected by the writer, it is thought that it will be useful to place on record in this form, a more detailed statement of the conditions of the occurrence of the silicitied woods, their peculiarities, their structural relations and their stratigraphical position, in the hope that it may eventually prove to be of use in correlat- ing the deposits in which they are found. These fossil woods occur throughout the area covered by Tertiary sands and gravels in the State of Arkansas. When in large masses they are apparently rarely far removed from beds of Tertiary lignite; if in small masses or in small frag- ments they occur in the gravels of nearly all the region and in the beds of the streams and brooks of the area covered by the Tertiary. Occasionally whole trunks of trees are found, often partially buried in the sands or deeply imbedded in the gravels which cover the flood plains of the creeks and ravines within the Tertiary area and especially along Crowley’s Ridge, from Helena to the Missouri line. Specimens have been obtained from logs or stumps 27 se¢w and in undisturbed Ter- tiary beds at the following points: Hope, Hempstead county ; * Agriculture and Geology of Mississippi, 1860, pp. 20, 21, e¢ seg. + Agriculture and Geology of Mississippi, 1860, pp. 20, 21, e¢ seg. First Annual Report of the Geological Survey of Texas, 1889; “ A Prelimi- nary Report on the Geology of the Gulf Tertiary of Texas from Red River to the Rio Grande.” By R. A. F. Penrose, Jr., pp. 1-101. : § See Annual Report of the Arkansas Geological Survey for 1889, vol. ii, pp. 249-267, Plates IX—XI. R. FE. Cali—Silicified Woods of Hastern Arkansas. 395 Camden, Ouachita county; near Red Land, Cleveland county ; at Red Bluff, Jefferson county; at Helena, Forrest City, Witts- burg, Wynne, Harrisburg, Jonesboro, Gainesville, Boydsville, and St. Francis in the country traversed by Crowley’s Ridge in the eastern part of the State. All of these localities have furnished examples of silicified wood from large logs or stumps in place and always imbedded in Tertiary sands or gravels. It is a remarkable fact that hitherto, in Arkansas, silicified woods have been seen but very rarely in the Tertiary clays. At all the localities mentioned above, except one, the wood is found only in gravels or sands, 27 s2tw, or in redeposited gravels and sands in the low valleys. The geological section of the Crowley’s Ridge region, to which area this paper especially refers, shows the following sequence, seen in the generalized section in St. Francis county which is characteristic for the southern portion. Generalized Southern Section on Little Crow Creek. 1, A loess soil, with enough sand to render it decidedly siliceous. This is the surface member and is usually of but little depth. Typical loess, varying in depth from thirty to ninety feet, eroding rapidly, and presenting a characteristic loess topog- raphy. ‘This member caps the ridge even at its highest points. 3. A clayey, pebble-bearing, bluish or otherwise dark colored loess clay which forms the base of the typical loess deposits and probably marks the first stage in the loess deposition. This member varies somewhat in different localities, being often quite thin and is even sometimes wanting. The peb- bles are most abundant in the lowermost portion. 4, Orange-colored gravels, irregular in thickness, rudely stratified, sometimes well assorted so that only coarse gravels, or vice versa, are seen; there are occasional pockets or lenses of sand derived from the underlying member. In rare instances this bed lies directly upon the clays. Silicified coniferous wood often occurs in this member. 5. Party-colored sands, of variable fineness, often quite irregu- larly stratified, sometimes overlying the pebble bed but usually occurring underneath it. The sand grains are well rounded. ‘There are occasional masses or pockets of red, drab, white, or yellow pipe clay. 6. Blue, black or drab clays, horizontally stratified, with small sometimes large pieces of coniferous lignite. This member constitutes the greater portion of the body of the ridge. Along its margin it is to be seen only in the deepest ravines, or along the St. Francis and such of its small tributaries as flow from the ridge. It is often penetrated in deep wells, as at Forrest City, and underlies the whoie region. The lower exposed portion is fossiliferous, the fossils are marine, and Claibornian in age. ‘The clays are therefore Eocene Tertiary. bo 396 Rk. &. Call—Silicified Woods of Hastern Arkansas. Slight differences in the section appear in various portions of the Ridge but are not worthy of remark in this connection. The generalized section for the northern portion of the Ridge, made at a point seventy-five miles north of St. Francis county shows the following sequence: Generalized Northern Section near Gainesville, Greene County. 1. A humus, largely siliceous, or a soil mainly sand. At the highest hilltops this soil contains gravel or may be entirely replaced by waterworn gravel. 2. Gravel bed, commonly removed by erosion. 3. Sands of Tertiary age, false bedded, party-colored, coarse or fine, banded often with drab, red or white pipe clay, or the last may be in pockets or lenses. These sands are generally loose, but in certain localities they have metamorphosed into avery hard, glassy quartzite. The areas of metamorphism are linearly distributed over many square miles but are con- fined chiefly to the west side of the ridge. Silicified woods are found in this member at many localities, but none has yet been discovered in the metamorphosed portions. 4, Drab, blue and black clays of Hocene Tertiary age, horizon- tally stratified, occasionally fossiliferous, the fossils being chiefly the leaves of deciduous trees. These clays contain rare beds of lignite of small extent and erratic vertical dis- tribution. Moreover, the clays are commonly gypsiferous and are further characterized by abundant small plates of muscovite in the cleavage planes. Silicified wood was seen at a single locality, on Cache River. The absence of fossils mn nearly all the members of the Arkansas Tertiary renders necessary their distinction upon lithological and structural data. The large masses of silicified wood in the upper members of the series are the only organic forms known above the Eocene clays. If in any way these silicified woods may be genetically connected with the lgnite beds a means of correlation will not certainly be had but the fact may sometime possess taxonomic value. Studies made in eastern Arkansas seem to show that all or nearly all of the silicified woods of the Tertiary sands and gravel beds are derived in some manner from the underlying beds of lignite. In many places whole tree trunks, stumps standing in place, or large fragments of silicified wood occur so related to lignite deposits as to show that they are derived therefrom. In the northwestern portion of Greene county, on the west side of Crowley’s Ridge, are masses of wood partly in the form of lig- nite and partly silicitied. The lignitized part is buried in Eocene clays; the silicified ends are buried in Eocene Tertiary sands. It would appear that in this case, before the sands R. FE. Call—Silicified Woods of Eastern Arkansas. 397 were eroded away, the portion of the trunk which had been buried therein was subjected to the action of waters contain- ing silica in solution and the lignitic matter was replaced by silica. The silica is, of course, all present as secondary quartz, is often massive but, also, frequently crystallized. Especially is holoecrystalline quartz abundant in specimens of wood that were partially decayed when the older lignification process began. In the drusy cavities of such lignite are found large numbers of perfect and rather large quartz crystals. These are often, in some specimens always, characterized by a uniform dark or brownish color which is due to inclusions of limonite.* Professor F. H. Knowlton, of the U.S. Geological Survey, has studied microscopically both the lignite and silicified woods found in eastern Arkansas. The results of his work may be found in vol. ii of the Arkansas Geological Survey, Reports for 1889. His studies have developed the interesting fact that the woods belong to both dicotyledonous and coniferous types. This occurrence is the first known dicotyledonous wood found in this country in rocks older than Pleistocene and is the first dicotyledonous form determined by internal structure. If, therefore, examinations of both lignites and silicified woods are made and it results that the same form or forms are repre- sented in both, a strong reason exists for genetically connect- ing the silicified woods with the lignites. | Unfortunately for taxonomic purposes all the forms described by Prof. Knowlton are new, but some otherwise valuable results have been reached. In the first place he finds, among the four new species studied, two forms which are clearly dicot-* yledonous, and two others distinctly coniferous in relationship. The species are: © . Coniferous. Dicotyledonous. Cupressinoxylon Arkansanum, Laurinoxylon Branneri, Cupressinoxylon Calli, Laurinoxylon Lesquereuxiana. There was also a single additional specimen whose affinities appeared to be dicotyledonous and to belong to Laurinoxylon ; the condition of the material would not admit of closer: deter- mination. The specimens found indicate comparatively few species but these few must have existed in great numbers. One of the most valuable and pertinent facts in this connec- tion is the finding of the dicotyledonous Lawrinoxylon Bran- nerz in the lignite bed of Bolivar Creek, as lignite, deeply buried in Eocene clays in massive form. * An especially fine example of this nature was taken from a section in Ter- tiary sands 13 miles southeast of the town of Camden on the line of the Camden and Alexandria railroad. Of the many thousands of quartz crystals which this specimen exhibits not one has been seen which is free from inclusions of limonite. 398 =f. & Call—Silicified Woods of Hastern Arkansas. Thus far sufficient distributional facts to give a taxonomic value to the fossil woods have not been discovered. Until extensive collections throughout the whole region of the south- ern Tertiary have been made it will not be possible to use these forms for purposes of differentiation or of correlation. It is be lieved, however, that since in the Tertiary sands of Arkansas, Louisiana, Texas and Mississippi the same relations of silici- fied woods to lignites have been observed, it may be possible to coordinate the divisions recognized in those states by geologists and devise a system of nomenclature that will explain the relationships of the various beds to each other, though it ean- not be done at present. During the progress of the study of the region by the writer it became more and more clear that the silicified wood had some intimate relation to the pockets or beds of lignite which are scattered throughout the ridge. It was early noticed that no lignite occurs in the sands or gravels above the clays, and that no detached masses of silicified wood occur entirely in‘the clays. As the investigation proceeded it became a favor- ite hypothesis that the silicified wood was transformed ‘lignite, and that careful microscopic study would probably prove the hypothesis to be correct. Professor Knowlton’s investigations appear to verify the hypothesis. The opinion that the silicified wood was, in some way, to be connected with the lignites of the beds underlying the sands was suggested by Hilgard* many years ago. Speaking of the occurrence of fossils in the Orange Sands he says: “.. . The closest scrutiny I have bestowed on hundreds of extensive exposures, has failed to detect any fossil apparently peculiar to the formation as such. This might seem paradoxical enough to any one acquainted with the fre equent occurrence of silicified wood in these strata, but it soon becomes quite obvious to an attentive observer that the regions of the frequent occurrence of this fossil in the Orange Sand are coextensive with those in which fossil wood, either silicified—when imbedded in siliceous sands—or lignitized, occurs in the underlying lignitiferous Cretaceous or Tertiary strata. It is not unusual to find trunks of silicified wood imbedded partly in the unchanged lignitie strata, partly in the Orange Sand; the portion contained in the latter. being nearly or w holly deprived of carbon, while the part imbedded i in the lignitie material is, if at all silicified, of an ebony tint and often “contains pyrites.” Again, “I am con- vinced that the great part, if not all of this fossil wood is derived from the underlying strata and will be represented in their flora.” * This Journal, IT, vol. xli, p. 313, 1866. R. EF. Call—Silicified Woods of Hastern Arkansas. 399 There can be little question, therefore, that the process of silicification has occurred, in some cases at least, since these masses were torn from the underlying beds by the waters which deposited the sands above the clays.* As ordinarily understood the process is purely a chemical one and perhaps very slow. It consists in the replacement, particle by particle. of the carbon of the lignite by silicic acid, or silicon dioxide. It is by no means essential that the organic matter be unchanged when the process begins. If the belief that this wood repre- sents what was once lignite be a correct one, then the process of silicification can occur in the case of organic matter which has already undergone a partial change. Where found in clays in a silicified condition, it has probably resulted from the same processes that are seen to obtain in the highly siliceous sands or gravels which overlie them. Though the impervious nature of most clays renders the percolation ‘of of silica-charged waters a matter of great difficulty such perco- lation certainly occurs in them. The silicified masses of wood are often far too large to have been removed from the clays and deposited in the overlying gravels by an ordinary wave or current action for they sometimes weigh tons. In the form of lignite the same masses could have been transported by cur- rents but since very large pieces have been rarely, if ever, found far from lignite deposits even that proposition has very little weight. The vertical distribution of the silicified woods of the Arkan- sas Tertiary is limited by the line of contact between the sands and clays which constitute the Arkansas series. Below this line the silicified wood never occurs, with the single exception above,t so far as observations have yet extended. Aboveit no * Dr. R. A. F. Penrose, Jr. (op cit., pp. 24, 26, 50, et seg.), has placed on record the numerous occurrences of silicified wood in the Tertiary of Texas; he finds it in both sands and clays. In his description of the Sabine River beds he says: ‘ Silicified wood is of very frequent occurrence in these strata; sometimes occur- ring as small fragments; and at other times as large trunks of trees. On the Brazos River, in the northern part of Milam County, was seen a trunk one anda half feet in diameter, protruding from a clay bed. Ten feet of it were exposed, while the rest was imbedded in the clay. In manv places such fragments are collected in great quantities, but it is especially plentiful in the lower part of the _ Fayette beds. It is generally dark brown or black inside, and weathers gray or buff color on the outside. Sometimes it occurs partly lignitized and partly silici- fied. It frequently shows shrinkage cracks which are filled with quartz or chalcedony, and are often lined with quartz crystals.” In this case stratification was but partial or was still in progress and since there is exposed in the face of the bluff a log which was partially lignitized and partly silicified it proves all but conclusively that, even in the Texan Tertiaries, the lignitic precedes the siliceous condition of these woods. + In this case the stumps are still standing, the roots, also silicified, ramifying in all directions in Eocene blue clays. Less than one hundred feet east, however, the line of contact between the sand beds and the clays was disclosed 1n a ver- tical cut in a hillside. ‘This line was at or near the elevation of the stumps. It 400 Lt. EF. Call—Silicified Woods of Hastern Arkansas. lignites have ever been found. The vertical range is therefore limited by the thickness of the sand and gravel bed which is commonly, in Arkansas, between fifty and eighty feet. There is a marked difference in the vertical range of this fossil in the Tertiary of Arkansas and the Tertiary of California. In the latter State the vertical range is often many hundreds, even several thousands, of feet. Whitney says:* “It will be proper to add to some of the most important facts gathered during the investigation of the gravel deposits in regard to the mode of occurrence of the fossil plants of the Pliocene epoch. The vertical range of these has been alluded to, and it may be more distinctly stated that either fossil wood or leaves have been found at every elevation, from the lowest to the highest, where gravels occur. Even as high as Silver Moun- tain City, at 7,000 feet of elevation, large masses of fossil wood are found in the voleanic deposits ; and in Plumas county the same occurrence has been noted on several of the highest mountains in the region, as Penman’s Peak and Clermont, peaks from 7,000 to 8,000 feet high ...... Fragments and often large masses of wood are found, both in the gravels and the associated clayey and tufaceous beds. In the gravel they frequently bear the marks of transportation from a dis- tance, as would be expected.” In the California Tertiary the most completely silicified and best preserved specimens of wood occur in connection with deposits of a voleanic character, sometimes a rhyolitic ash.t It is suggested by Whitney that these relationships have some- thing to do with the process of silicification. For that region Whitney believes that not only were the woods silicified after their imbedding in white pulverulent volcanic ash but ‘ the lava itself exhibits signs of having been acted on by silicifying agents after its deposition.” That the greater part of the series of beds included in the gravel formation has been thoroughly permeated with waters holding silica in solution and that chemical changes induced thereby are sufficient to explain the phenomena appears quite probable. The relations which the phenomena sustain to the facts of voleanism so abundant in that region are set forth and the conclusion is drawn that that rela- tion explains silicification in these woods. In California it becomes a subordinate problem under volcanism. The chemical processes which obtained in the case of the Arkansas gravels were not codrdinate with those of California, was clear that, if the stumps did not actually project into the overlying sands, they were but a short distance below and under conditions to favor silicification from waters percolating through the clay to them. * Auriferous Gravels of the Sierra Nevada, pp. 235, 236. See also this Jour- nal, II, vol. xli, p. 359, 1866. + Op. cit., pp. 327-329. Weed and Pirsson—Sulphur, Orpiment, Realgar, etc. 401 for there is no evidence of volcanism or any similar phenomena associated with their silicification. The silica in the eastern locality must be sought in the accompanying sand beds and was probably brought into solution by the action upon it of organic acids. The study of the Arkansas Tertiary silicified woods appears to justify the following conclusions: 1. The silicified woods of eastern Arkansas are all of Ter- tiary age. 2. They are derived from the beds of Eocene clays that underlie the sands and gravels in which they commonly occur. 3. They are silicified lignite; the process of silicification has occurred either while they were still in clays or most often after they were removed and buried in the sands or gravels. 4. They possess as yet no taxonomic value in determining the relative ages of the members of the Tertiary series. Geological Survey, Little Rock, Ark,, July 15, 1891. Art. XLIL—Occurrence of Sulphur, Orpiment and Realgar in the Yellowstone National Park ; by WALTER H. WEED and Louis V. Pirsson. : SULPHUR. In the Yellowstone National Park there are besides the well known geyser basins, many small hot spring areas and localities where fumeroles and solfataras are still active. At most of these places deposits of sulphur occur, in and around the vents from which sulphurous vapors issue. At the Highland Hot Springs and at Crater Hills these vents are quite abundant and large deposits of sulphur are found frequently having most beautiful clusters of delicate crystals. The latter locality, from which the specimens herein described were obtained, is a small group of hills whose white and steaming slopes form a prominent feature of the eastern part of Hayden Valley, the open grassy country traversed en route from the Grand Canyon of the Yellowstone to the Lake or the Firehole geysers. The hills, often called Sulphur Mountain, rise about 150 feet above the surrounding level, and are formed of fragmentary material wholly rhyolitic, decomposed and cemented by the vapors that rise at innumerable points through the hills. There are but few springs at this locality; the most prominent and most active is an ever-splashing bowl of green sulphurous waters known as the Chrome Spring. Behind this basin the slopes are light colored, chalky white, rose-pink and dull yellows . 402 Weed and Pirsson—Sulphur, Orpiment and being the predominant tints Large masses of rough clinker- like rock he scattered about the slopes, resting upon small pieces of the same cemented material or upon the smoother slope of white pulverulent silica resulting from the complete decomposition of the rhyolitic material by the acid waters and vapors. Several parts of the slopes show the dull yellow color of sul- phur, such places usually being further marked by many steam- ing orifices a few inches across. These vents are generally lined with a layer of radially fibrous sulphur, whose surface is thickly set with delicate frost-like clusters of crystals. Many of the vents are partly closed by the sulphur and others com- pletely sealed but filled with hot vapor which is copiously emitted when the roof is broken. No temperatures above 200° F. were obtained from any of the vents. It is im these closed vents that the largest and most beautiful crystal clusters have been found. Upon taking such.a specimen from one of these vents it is a deep orange tint, and of course quite hot; as it cools the crystals loosen with a loud and continuous crackling so that a slight jar is sufficient to cause many of the clusters to fall to pieces when cooled. These crystal clusters are of interest as sulphur in the crystal form has been described from but one American locality—Nevada.* In examining a mass of these crystals it is seen that the crystalline mass is of great brittleness, owing largely to the fact that the crystals are gen- eyally hollow. Often a mere skeleton of what would otherwise be a good sized crystal is present. One that would be an inch high and proportionately broad and wide consists only of nar- row strips preserving the edges of the pyramids on each other ; this frame work is then filled with other crystals and parallel growths, also of hollow delicate material. In general the mass is made up of confused erystal aggregates closely united below and toward the top branching into arborescent forms. Often small solid crystals occur attached to the mass. Several of these were selected for measurement. They proved to be of the usual orthorhombic sym- metry. The habit is strongly pyramidal. One of them that is typical of the series is shown in the figure. The forms which were identified on j, this crystal are: c, 001, O. e, 101, 1-2. y, 112, +. a, 183, 1-3. im, VO, L.: mn, O11, 1-%, s, 113, + gq, 131, 3-3. h, 130, 2-3, reek blag é, 115, 4. The identification of these forms is shown by the following tables of caleulated and measured angles. For * KH. 8. Dana, this Journal, xxxii, p. 389, 1886. . a Realgar in the Yellowstone National Park. 403 the calculated angles the axes of Kokscharow* have been taken in which, ae WavreG = UsSls0o le 90359 Forms. / Calculated. Measurement. . pap Wall Oar 5a g4P ol, 94? AN?” 94" 40? pap Winall 36 404 (SOLAS USSG" 44z pam 111.110 iB 204 pile 200 lS ecoo8. peel) ~ 101 | 36 41 3b, 45,30.» DO mam 111 011 | 47 26 is eres nau” ua ae 111.133 | Di 29 ia Dad. > dome vleaites ene © DFT 30 mag 111131 | 29 104 D9) AB 206517 May AV11,. 112 | 15 11% 15 134 Wins Jill. 113 26 293 26/ 284, 26 304 wer AIT x 115 | 40 33+ 40 34 pac 111.001 71 393 wl) 414 In general the planes, even though minute, gave fair reflections owing to the brillianey of their surfaces. The only exception to this was the plane 2-3 1380, from which no satisfactory reflec- tions could be obtained. It is easily identified however since it lies in the zones 110,110 and 001,181. Though no tests were made the material is apparently of great purity. It is very homogeneous and is of a delicate sulphur-yellow. ORPIMENT AND REALGAR. The presence of arsenic in the hot spring waters of the Yel- lowstone Park, and the deposition of the hydrous arseniate of iron, scorodite, by them, has already been noticed in this Jour- nal.* While studymg the hydro-thermal phenomena of the region for the U. 8. Geological Survey, under the direction of Mr. Arnold Hague, a careful search for deposits of the arseni- eal sulphides was rewarded by the discovery of realgar and orpi- ment, at the Norris Geyser Basin. This locality, though pos- sessing few geysers worthy of comparison with those of the Firehole basins, is peculiarly interesting by reason of the new- ness of its geysers and the great variety of its chemical deposits. It covers an area of some six square miles situated amid the forests of the great rhyolite plateau of the Park whose gradual slopes rise on every side. The area of present activity is included between a loop of the Gibbon river and two spurs of Gibbon Hill, an eminence of rhyolite that rises above the gen- eral level of the country to the south. The multitude of * Min. Russl., vi, p. 368, 1874. * A. Hague, this Journal, vol. xxxiv, Sept., 1887. 404 Weed and Pirsson—Sulphur, Orpiment, Reulgar, ete. vividly colored pools and equally bright tinted waterways, the white sinter flats, and the creamy rose and yellow shades of the decomposed rhyolite, the whole surrounded by a setting of dark green pines, presents a strange picture not easily for- gotten. The specimens of realgar and orpiment come from the western part of this basin, between the 100 spring plain and the Gibbon river. The deposits of siliceous sinter so abundant elsewhere in the basin are here quite scanty and form a thin coating upon rock composed of small angular fragments of pearlite, obsidian, and other forms of rhyolite—generally more or less decomposed and mixed with quartz grains, the whole compactly cemented by silica deposited by the hot spring: waters. Several small outflows of clear and hot acid water issue from this cement rock, their united overflow forming a small stream which flows through a shallow gutter in the rock and joins the Gibbon river a few hundred yards beyond. Near the vents the channels are lined with an incoherent deposit of milky sulphur which frequently coats and obscures a growth of alge. The gray surface of the rock shows no trace of the brilliantly colored arsenical sulphides, and it is only upon breaking this rock about the vents and prying up the plates at the margins of the channels that the realgar and orpiment are noticed. Plates of rock thus obtained show a brilliant red and yellow surface of the mixed sulphides and large pieces of rock from about the vents are penetrated and filled with deposits of the yellow orpiments, the dark red realgar, and the mixture of the two. Many of the specimens show layers of waxy dark red translucent realgar an eighth of an inch thick, covered by incoherent amorphous orpiment and alternating layers often occur. The orpiment generally possesses a tangled filamentous structure upon the surfaces of the plates as if deposited upon alow threads, and where the surface of the plate is covered with realgar as is frequently the case, it too possesses this curious form, the mineral being in stalagmitic aggregates with a general fibrous matted structure. | In the channels the rocks show no surface coloring from the deposition of the sulphides but many small pebble-like masses occur, lying in the bottom of the stream, which consist almost entirely of dark red translucent realgar. This appears to be the most promising material collected for mineralogical exami- nation, but unfortunately none of the specimens proved to con- tain any crystals which could be measured and identified under the microscope it proved to be in rounded stalagmitie growths consisting of a confused crystalline aggregate. Such light reflecting surfaces as were seen proved to be small cleavage planes. In the closed tube the substance melts and then forms L. V. Pirsson—Mineralogical Notes. 405 a reddish translucent sublimate. In the open tube with a good eurrent of air volatilizes and deposits in the upper part small glittering octahedrons, which under the microscope in polar- ized light proved isometric (As,O,). Before the blowpipe on charcoal gives characteristic odors for sulphur and _ arsenic. Fused with carbonate of soda gives the reactions on dissolving for sulphur and arsenic. The association of the realgar and orpiment is such that no definite statement can be made as to which forms first, but realgar is certainly the last formed upon many of the speci- mens. Whether it is formed through a conversion of the orpiment or as a separate deposition is uncertain, but the specimens seem to indicate that the latter is the case. Sili- ceous sinter is the only other mineral occurring with these arsenical sulphides. Art. XLIL.—WMineralogical Notes; by L. V. Presson. 1. Cerussite.—Some specimens of cerussite, obtained through Messrs. English & Co., of New York, from the Red Cloud Mine in Yuma Co., Arizona, contain twin crystals in which the twinning plane is the uncommon form 7-3, 130. Since, moreover, they show some unusual developments in their erys- tal form, it has been thought that a description of them would be of interest. The specimens are in the cabinet of Prof. George J. Brush. The greater number of those observed have the form shown in fig. 1, which presents them in a basal projection. This gives a much better idea of these crystals, shaped like arrow- heads, than an orthographic projection. They are of various sizes up to half an inch in length, the largest observed. They 406 L. V. Pirsson—Mineralogical Notes. resemble the example of this method of twinning given by Kokscharow* in the development of the brachydomes. The figure shows them in ideal symmetry; they are generally at- tached by the barb-shaped end; sometimes one barb or indi- vidual is free with the faces developed as in the figure; in other forms both individuals are attached and the barb-like part is wanting. They occur on the specimen seen, with cerargyrite and wulfenite. The forms observed on these erystals are: r, 1-3, 130; k, 1-%, 011; », 3-%, 031; a, 4-%, 012; m,I, 110; p. 1, 111. Also the pinacoids 2-2, 100 and 2-2, 010, the pyramid 3, 112 and the brachy-diagonal pyramid 2-2, 121 have been identified in zones on the reflecting goniometer and measured with some accuracy, but they are too minutely developed to give any character to the crystals and are hence omitted in the figures. The crystals are well suited for measurement, the faces gener- ally giving good reflections of the signal. Fig. 2 shows a crystal from the same place with a different habit, the large $ brachydome & and great development of the brachy-prism 7 having resulted in a long spindle-shape form. Fig. 8 shows the same individual in basal projection. It oce- curred, attached at the end with the re-entrant angle, which in consequence was somewhat broken. Otherwise the crystal is quite perfect and in size about an inch long. The zone of brachydomes is somewhat striated, causing a rounding off toward the point. The signals of the prominent faces are, however, very distinct and give good measurements. On this crystal the forms observed were: b, 4-7, 010; c, O, 001; 7, 4-3, 130; &, 1-4, 011; a, 4-4, 012. The following table shows the identification of the forms by calculated and measured (supplementary) angles. Calculated. Measured. kap O1lLAIIL 43° 50” 40” 43° 517 map LO ALL sya 25) 0) 30 42, oD 50, 35 49 kxGr “OLD ~ 130 a9 03 10 59 00 Di el le ie 49 59 30 49 58 WAT VO i30 29 OT 730 30 00, 30 01 pat) Wade 45 19 55 45 16, 45 22 vad.) O81,0081 49 30 49 27 kaw Ooo 15. 59. 30 16 00, 16 04 Kak SOLU 108 16 108 12, 108 14 raw -lBOn 012 (ie. Bish Blo) nwo cae. 00012 L9' 52, BO 19 48 rab 130.2010 ze 39 30 28 37, 28 31 knb 0114010 54 08 54.17 001,112 34 46 34 54 1214121 85 59 86 00 * Min, Russl., vi, pp. 106, 1870. Also Atlas Taf. Ixxx, figs. 20 and 20 bis. L. V. Pirsson—Mineralogical Notes. 407 2. Hematite and Cassiterite.—Interesting specimens of hem- atite and cassiterite intimately associated and ‘crystallized have been forwarded by Prof. F. A. Genth for crystallographie ex- amination. The place from which they come is Mina del Diablo, Durango, Mexico. Among them a number of un- doubted pseudomorphs of cassiterite after hematite were ob- served. They are too small and lusterless for the forms to be determined, but the whole grouping is precisely the same as that of hematite in the familiar “ EKisenrose” habit, consisting of radiating plates. Often the central portion of these plates consists of a piece of hematite, the outer surrounding part of tin oxide. There were also seen pseudomorphs of cassiterite after some octahedral mineral, probably magnetite. These occurrences of cassiterite and hematite have already been described by Genth and vom Rath,* but in these specimens lately examined an additional point of interest was noted. This is the presence of cellular crys- tals of hematite filled with cassiterite. ~ One of these is shown in fig. 4. The forms present on this crystal are G, 0001, O- a, 1120, 7-2: 7, 1011, R; 7, O111,-1; s, 0221, -2; n, 2243, 3-2; d, 1012, 4; ©, 2021, 2. Those chiefly developed are the prisms and the basal pinacoids which give the crystals its habit. Im reality these faces are present to a considerable degree only along the edges, the re- maining parts being sunken and filled with a roughened surfag¢e _ of cassiterite which runs on through the crystal. The polish and luster of as much of the face as is present is, however, very fine and brilliant. A thin section of this dividual was prepared parallel to the unit prism to ascertain if the cassiterite had here also any definite orientation in regard to the hematite. None was ob- served. The two minerals, both appearing fresh and unaltered, were present in irregularly mingled masses. The cassiterite was formed of an intimate crystal aggregation as shown by its sight but uniform action on polarized light. From these facts it would seem as if the two minerals had been formed simultaneously and the hematite having a greater tendency to crystallize than the cassiterite had assumed its erystal boundaries without regard to the latter. 3. Gypsum.—The erystals which are illustrated in fig. 5 are from Girgenti, in Sicily, and are now in the cabinet of Prof. Geo. J. Brush. They are twinned according to the usual method, * Proceedings Am. Phil. Soc:, 1887, xxiv, 23. Am. Jour. So1.—TH1RD SERIES, VoL. XLII, No. 251.—NovEMBER, 1891. 408 L. V. Pirsson—Mineralogical Notes. the twinning axis a normal to 100 and the following simple forms are present, m, J, 110; 6, 24, 010; Ds l, -1, 111; and e, 4, 1038. Since the or- thodome 4-2, lacks only about two degrees of forming a right angle with the orthopin- acoid, the two domes in twin position prac- tically present the appearance of a basal plane and the whole crystal that of hemi- morphic orthorhombic rather than of mon- oclinic symmetry. This pseudo-basal plane is rough and oscil- latory and the very slight salient angle cannot be detected. The crystals are of good size and very symmetrical except at the end where the twin pyramids are; they are attached by this point and are as a consequence broken and disturbed at this place. es 4. Pennine.—The crystal form and optical properties of the violet chrome pennine or kimmererite from Texas, Penn., were originally described by Cooke* from ‘specimens from the cabinet of Prof. Brush. At the request of Prof. HE. S. Dana I have recently studied the suite of specimens in Prof. Brush’s collection from this locality and thanks to the present perfec- tion of apparatus for crystallographic investigation, I have been able to make out several forms observed by Prof. Cooke but which he was not then able to determine. The forms ob- served on these crystals are as follows: O, 0001; R, 1011; y, 2, 2025; 9, #%, 40-413; 2, 4, 10135 sp eae %, 1-2, 1122. The last three are new. The habit is shown in figs. 6 and 7. Fig. 6 is very similar to that given by Cooke, omitting the pyramids of the second order. As observed by him the erys- tals are generally twins. The planes forming the re-entrant angle are nearly always more or less striated, moreover this re- entrant angle is, so far as observed, invariably formed by the unit rhombohedron R (1011). Fig. 7 shows a case where there was practically no striation and the angle could be measured. The presence of pyramids of the second order is a noteworthy characteristic, on no crystal out of a large number was it ever wanting. With one or two exceptions, on very small crystals, they were invariably striated as shown in fig. 7. These stria- tions all lie in one zone, and this having been determined, it was possible to measure from the base along this zone, the re- flections of various pyramids which stood out in the band of light connecting them. a A In this way the presence of 4-2 (1126), 4-2 (2245), =%-2 (9°9°18-20) and 2-2 (7-7°14°8) was determined by moderately * This Journal, vol. xliv, pp. 201, 1867. L. V. Pirsson—Mineralogical Notes. 409 good measurements, some of which were repeated on several erystals. The following table shows the identification of the forms by calculated and measured angles. ments of Cooke are taken, in which pet. 3- 4951. 6. (0001 ,1011)=76° 05’. CAT cad CAY CAZ Cap Zap CaAX Forms. Calc. 0001 .1011 103° 55’ 0001 . 4:0°4:13 51 09 0001 ~ 2025 58 13 0001 . 1013 53 224 0001 . 1124 60 134 1013 . 1124 25. 5A 0001 . 1122 14 05 For the calculated the ele- Meas. 104° 51° 067 58 05 53 48 60 35 25 22 74 10 The form 1-2 (1122) was also identified by the fact that it lies in the zone between 4 (1018) and 7, R, (1011). 5. Mordenite.—The author desires to correct a small error which crept into the determination of the constants of this mineral as given in this Journal, xl, page 236. read instead of the figures there given, which are Mineralogical Laboratory, Sheffield Scientific School, a: =>) t a:b:e¢ \ New Haven, Nov., 1890. These should b:¢:; 0:40099: 1: 0°42792 angle B—88° 29’ 46" :10°40101: 1; 0°42623 angle 6=88° 30’ 30’. 410 J. Fi Kemp—Peridotite Dikes near Ithaca, NV. Y. Art. XLIII.—Peridotite Dikes in the Portage Sandstones near Ithaca, VN. ¥.; by J. F. Kemp. In the valuable paper on the peridotite* at Syracuse, N. Y. which appeared in this journal in August, 1887, the follow- ing statement is made and it doubtless expresses a very wide- spread and general impression. “This rock is interesting as being the only known instance of igneous intrusion in the unaltered and undisturbed Palseozoic strata of New York” (p. 144). Since the writer’s first residence in Ithaca (1886), the occurrence of trap dikes in tbe vicinity has been a sub- ject of frequent discussion in the geological laboratory of Cornell University. Conversations with alumni who were students under the instruction of Professor C. F. Hartt (1868-1878), revealed the fact that he made frequent men- tion of them and created the impression that. they were well recognized phenomena in two of the neighboring gorges. They do not appear to have become a matter of record except in two cases. Professor O. A. Derby (now in Brazil) in a short paper in the Cornell Review (which was the student publication of that date), vol. i, p. 70, 1874, entitled “ Hints to Geological Students,” mentions a number of loeali- ties involving in all four dikes. Three of these are in Casca- dilla Creek on the confines of the university and the fourth, said to be the best for study, is in Six-Mile Creek, two or three miles distant. Professor F. W. Simonds (now of Texas) pub- lished in 1877, a short article in the American Naturalist (vol. xi, p. 49) on the Geology of Ithaca, N. Y., and vicinity. The Six Mile Creek dike is again mentioned and described as filling a crack in the sides of the gorge but as pinching out before it reached the surface. Long before this, however, in 1842, in the Report on the 3rd District, N. Y. State Survey, p. 169, Vanuxem recorded four narrow dikes in the Genesee slate near Ludlowville, which is ten miles north of Ithaca. The locality has been recently visited by the writer but only the two dikes near the upper falls of Vanuxem could be found. They are each about an inch wide and only show over a short space as they disappear above and below. ‘They were inaccessible and from the distance of a few feet their igneous nature was not conclusively shown. The other two could not be found. Van- uxem also mentions another dike (1. e. pp. 207-208) at Manheim Bridge east of Little Falls, N. Y., more than one hundred miles *G. H. Williams: The Serpentine (Peridotite) occurring in the Onondaga Salt Group at Syracuse, N, Y., this Journal, August, 1887, p. 144. See also Proc. Geol. Soc. Amer., vol. i, pp. 533, 534. J. FF. Kemp—Peridotite Dikes near Ithaca, N. ¥. 411 northeast of Ithaca and seventy-five miles from Syracuse, but what its character is or whether it is indeed igneous is unknown to the writer. The dike in Six Mile Creek near Ithaca was re-located in 1887, and slides were at once prepared. It proved to be a thin mass 14 to 2 inches wide and fills one of the numerous, parallel, north and south joints which are extremely abundant in the shaly sandstones of the region. It crosses the stream like a narrow ribbon and pinches out a few feet above the sur- face of the water. It has a light brown or drab color with darker spots scattered through and is provided with numerous seales of a reddish mica. It effervesced and in the slides showed a mass of alteration products with very strong sug- gestions of an eruptive structure, but as the material was so decomposed it was decided to be too meagre, to deserve men- tion. It was subsequently submitted to Messrs. G. H. Wil- hams, Diller and Derby and the last two were strongly of the opinion that it was igneous and suggested blasting. Later dis- coveries make this procedure hardly necessary and prove the specimen to be undoubtedly an eruptive rock in advanced decomposition. During a visit from Professor Derby, the past autumn, the subject of dikes was again brought up and the probable loca- tion of one in Cascadilla Creek was indicated. The point is under the discharge raceway of the reservoir forming Willow pond, just east of the entrance to the Cornell Campus. In a receut drouth it became accessible. The dike is about three ‘feet in width and strikes north and south right across the course of the creek. It is in a recess formed by its weather- ing and a corresponding recess appears on the opposite side, filled however with dirt. It is covered with sand in the creek. The rock itself is very dark green to black. Its surface is mottled by black protuberances which look very much as if they were pebbles. But they readily crumble under the fingers to a black dirt. The fresher portions have a porphy- ritie aspect and suggest a peridotite at once, and this is verified by the slides. In the sections the rock is seen to be highly altered. The black masses prove to be the remains of large olivine and enstatite or bronzite crystals. The latter show very generally the striated appearance so characteristic of these pyroxenes but the silicate itself has changed to serpentine and carbonates. The crystals are 8-5™™ in diameter. The recognizable olivine is in smaller crystals than the pyroxene as a general thing, but appears in no incon- siderable amount. It is very probable that the larger, unstri- ated alteration masses were also olivine. The characteristic reddish biotite of the peridotites is distributed through the 412. S. F. Kemp—FPeridotite Dikes near Ithaca, N. Y. rock, and is still quite fresh. The crystals run about 0-2-0:3™ long. Their distribution imdicates at times an excellent flow structure. Magnetite is abundant both in irregular grains and rude crystals. A small amount of a reddish brown mineral, of high index but not entirely isotropic, is also present. It is probably perofskite. The groundmass consists in large part of numerous small acicular crystals of highly inclined extinct- ion which are augite. The groundmass seems originally to have been glassy. An analysis which was kindly made in the chemical laboratory at Cornell by Mr. W. H. Morrison, grad- uate student in chemistry yielded the following results. SiO; Fe,0, Al,O, Ca0Q MeO K,O Na.O | igssye aan 2/2 8444 11°92 © 28°60 5-45) 1-97 1-02 0-9 a ee The analysis shows at once the advanced stage of the decom- position and yet indicates a very basic rock. Qualitative tests failed to show chromium. The rock resembles very closely the Syracuse serpentine described by Dr. G. H. Williams in its general macroscopie ap- pearance and in many of its microscopic characters. The large phenocrysts are the same but the olivine is less fresh than at Syracuse. The reddish mica is present in both. It has also been compared with the peridotite from Elliott Co., Ky.,* and that of Pike Co., Ark.,} and evidently belongs to the same fam- ilv of rocks but as might be expected it resembles the Syracuse rock most closely. The occurrence is interesting because it shows the further distribution of igneous rocks in a region supposed to be free from them. Ithaca is some seventy-five or eighty miles south of Syracuse and much higher in the geological scale. The local rocks are shaly sandstones of the Portage stage and are extensively seamed by a series of north and south joints and another series, of west-northwest strike. The dikes in all cases follow the northerly series. It is not improbable that they belong to the same eruptive outbreak that found a larger mani- festation at Syracuse. If so the intrusion is put at a date later than the Upper Devonian, but beyond this no further deter- mination can be made with the data at hand. Geological Laboratory, Columbia College, New York City. * J, S. Diller: The Peridotite of Elliott Co., Ky. Science, Jan. 23, 1885, p. 65. Bulletin No. 38, U. S. G. S., 1887. + Branner and Brackett: Peridotite of Pike Co., Ark. This Journal, July, 1889, p. 50. A. E. Foote—Meteorie Lron of Cation Diablo. 413 Arr. XLIV.—A Wew Locality for Meteoric Iron with a Preliminary Notice of the Discovery of Diamonds in the Tron';* by A. E. Foote. With Plates XIV, XV. Historical sketch of the discovery.—tin the latter part of March, 1891, the mining firm of N. B. Booth & Co., of Albu- querque, New Mexico, received a letter from a prospector in Arizona informing them he had found a vein of metallic iron near Cafion Diablo, sending them at the same time a piece with the request for an assay. Sometime in April this piece was examined by a Colorado assayer who reported ‘76:8 per cent of iron, 1°8 per cent lead, 4 oz. silver, and a trace of gold. From its appearance we should take it to be a furnace product.” + This result was naturally not satisfactory to the mining firm and a mass weighing forty pounds was broken into several fragments with a trip hammer. One of these was sent to the President of the Santa Fe Railroad, and another to Gen. Wil- lhamson, the land commissioner of the Atlantic and Pacific Railroad Co., nm Chicago. Gen. Williamson consulted me as to the probable value of the so-called mine of “pure metallic iron,” stating on the authority of the prospector that the vein had been traced for a distance of about two miles, that it was forty yards wide in places, finally disappearing into a mountain and that a car load could be taken from the surface and ship- ped with but little trouble. A glance at the peculiar pitted appearance of the surface and the remarkable crystalline structure of the fractured por- tion convinced me that the fragment was part of a meteoric mass, and that the stories of the immense quantity were such as usually accompany the discovery of so-called native iron mines, or even meteoric stones. As soon as possible, in June, I made a visit to the locality and found that the quantity had, as usual, been greatly exaggerated. There were some remarkable mineralogical and geological features which, together with the character of the iron itself, would allow of a good deal of self deception in a man who wanted to sell a mine. Description of Locality.—Nearly all of the small fragments were found at a point about ten miles southeast from Cafion * Read before the American Association for the Advancement of Science, August 20th, 1891. + This assay was of such a remarkable character that I took the trouble to stop at the city where it was made and ask how such extraordinary results were obtained. I was informed that the lead, silver and gold were probably the results of the materials used in making the assay. . 414 A. H. Hoote—Meteoric Lron of Cation Diablo. Diablo near the base of a nearly circular elevation which is known locally as “Orater Mountain.” I believe this is the same as Sunset Knoll figured on the topographical sheets of the U.S. Geological Survey. This is 185 miles (297-72 kilo- meters), due north from Tucson and about 250 miles (402°34 kilometers) west of Albuquerque. The elevation, according to the survey, rises 432 feet (131°67 meters) above the plain. Its center is occupied by a cavity nearly three quarters of a mile (1:2 kilometers) in diameter, the sides of which are so steep that animals that have de- scended into it have been unable to escape and have left their bleached bones at the bottom. The bottom seemed to be from fifty to one hundred feet (15:24 to 30°48 kilometers) below the surrounding plain. The rocks which form the rim of the so- called ‘‘crater” are sandstones and limestones and are uplifted on all sides at an almost uniform angle of from thirty-five to forty degrees. A careful search, however, failed to reveal any lava, obsidian or other volcanic products. I am therefore unable to explain the cause of this remarkable geological phe- nomenon. I also regret that a severe gallop across the plain had put my photographic apparatus out of order so that the plates I made were of no value. About two miles (3°22 kilometers) from the point at the base of the “crater” in a nearly southeasterly direction, and almost exactly in a line with the longest dimensions of the area over which the fragments were found, two large masses were dis- covered within about eighty feet (24°38 meters) of each other. The area over which the small masses were scattered was about one-third of a mile (0°53 kilometer) in length and one hundred and twenty feet (86°57 meters) in its widest part. The longer dimension extended northwest and southeast. Description of the specimens.—The largest mass discovered weighs 201 pounds (91:171 kilos,) and as the photograph shows, Plate XIV, has a somewhat flattened rectangular shape showing extraordinarily deep and large pits, three of which pass entirely through the iron. The most remarkable example of such perforation is the Signet Iron from near Tucson, Ari- zona, now in the National Museum and figured in Prof. F. W. Clarke’s Catalogue.* One other large mass was found weighing 154 pounds (69°853 kilos?) This is also deeply pitted. A mass weighing approxi- mately 40 pounds (18°144 kilos) was broken in pieces with a trip hammer and it was in cutting one of the fragments of this mass that diamonds were discovered. Plate XV. * The Signet Iron was discovered about 30 miles (48°28 kilometers) from Tue- son. Dr. Geo. H. Horn states that 25 years ago he was told by the Spaniards that plenty of iron could be found on a range of hills extending northwest and southeast half way hetween Albuquerque and Tucson. A. E.. Foote—Meteorie Iron of Cation Diablo. 415 Besides these masses of considerable size.a careful search made by myself with the assistance of five men was rewarded by the discovery of 108 smaller masses. Twenty-three others were also discovered making a total of 131 small masses rang- ing in weight from ,'; of an oz. (1°79 grm.) to 6 lbs. 10 oz. {3006 kilos.)* A brownish white slightly botryoidal coating found on a number of the meteorites, is probably aragonite. A thorough examination of many miles of the plain proved that the car load of iron existed only in imagination. Accom- panying the pieces found at the base of the “crater” were’ oxidized and sulphuretted fragments which a preliminary examination has shown are undoubtedly of meteoric origin. About 200 pounds (90-718 kilos) of these were secured, from minute fragments up to 3 pounds 14 oz. or (1°757 kilos.) These fragments are mostly quite angular in character, and a very few show a greenish stain, resulting probably from the oxidation of the nickel. This oxidized material is identical in appearance with an incrustation which covers some of the iron masses and partially fills some of the pits. Composition.—After obtaining the meteorite I was unable to return to Philadelphia for sometime, and, therefore, sent a fragment of the 40 pound mass (18:144 kilos) to Prof. G. A. Koenig for examination. Prof. Koenig was compelled to leave town before this examination was completed. I take the fol- lowing, therefore, from his letters to me and from an account furnished the daily Public Ledger by Dr. E. J. Nolan, Secre- tary of the Academy of Sciences, of a preliminary notice made by Prof. Koenig, June 23rd, before the Academy of Natural Sciences of Philadelphia. In this account he says: “In cutting the meteoric iron for study it had been found of a extraordinary hardness, the section taking a day and a half, and a number of chisels having been destroyed in the process. When the mass, which on the exterior was not distinguished from other pieces of meteoric iron, was divided, it was found that the cutting apparatus had fortunately gone through a cavity. In the attempt to polish the surface so as to bring out the char. acteristic Widmannstattian figures, Dr. Koenig received word that the emery wheel in use had been ruined. On examination, he then found that the exposed cavities contained diamonds which cut throngh polished corundum as easily as a knife will cut through gypsum. The diamonds exposed were small, black, and, of course, of but little com- * Oct. 18th.—During September I received three additional large masses weigh- ing respectively 632, 506 and 145 pounds (or 286°678, 229°516 and 65°771 kilos.) The two latter were each perforated with three holes. A number of smaller masses up to 7 pounds, (3°175 kilos.) were discovered by digging. The three large masses and one of 23 pounds, (10°432 kilos) were covered with grass and earth.—A. E, F. 416 A. EL. Foote—Meteoric Iron of Cation Diablo. mercial value, but, mineralogically, they are of the greatest interest, the presence of such in meteorites having been unknown until 1887, when two Russian mineralogists dis- covered traces of diamond in a meteoric mixture of olivine and bronzite. Granules of amorphous carbon were also found in the cavity, and a small quantity of this treated with acid had revealed a minute white diamond of one-half a millimeter, or about =!, of an inch in diameter. In manipulation, unfortu- nately, this specimen was lost, but others will doubtless be obtained in the course of investigation. The minerals, troilite and daubréelite, were also found in the cavities. The propor- tion of nickel in the general mass is three per cent, and the speaker was not as yet able to account for the extraordinary hardness apart from the presence of the diamonds in the cavities.” Prof. Koenig in a letter to me gives the following points as definitely known. “(1.) Diamonds, black and white established by hardness and indifference to chemical agents. (2.) Carbon in the form of a pulverulent iron carbide occurring in the same cavity with the diamonds. The precise nature of this carbide, whether containing hydrogen and nitrogen is not ascertamed except in so far that after extracting all iron by nitro-hydrochlorie acid, the black residue goes into solution with deep brown color upon treating it with potassium or sodium hydrate. From this solution acids do not precipitate anything. (38.) Sulphur is not contained in the tough malleable portion of the mete- orite but in the pulverulent portion. (4.) Phosphorus is con- tained in the latter, and not in the former. (5.) Wickel and Cobalt in the proportion of 2:1 are contained in both parts nearly equally. (7.) Sedécon is only present in the pulverulent portion. (8.) The Widmannstittian figures are not regular. (9.) The iron is associated with a black hydroxide containing Fe, Ni, Co, P, in the ratio of the metallic part and therefore presumably derived by a process of oxidation and hydration of the latter.” Conclusions.—As this meteoric iron contains only 3 per cent of nickel while that from the Santa Catarina Mountains, 30 miles (48°28 kilometers) southeast of Tucson and 215 miles (346 kilometers) from this locality, contains from 8 per cent to 9 per cent, according to the analysis of Brush and Smith, they are quite distinct although somewhat alike in external appear- ance. They also somewhat resemble the Glorietta meteoric irons from about 3800 miles (482°8 kilometers) to the east north- east, in New Mexico. These contain 11:15 per cent of nickel. i em 1... > a Wadsworth—Trap Range of Keweenawan Series. 417 The most interesting feature is the discovery for the first time of diamonds in meteoric iron.* This might have been predicted from the fact that all the constituents of meteoric iron have been found in meteoric stones, and wice versa, although in different proportions. The incrustation of what is probably aragonite shown by some of the masses has rarely been noticed (I find two records by J. Lawrence Smith which he states to be unique, and both of these were from regions south of this one). The incrusta- tion is especially interesting as showing that the meteoric irons must have been imbedded a long time, as the formation of aragonite would be exceedingly slow in this dry climate. The remarkable quantity of oxidized black fragmental material that was found at those points, where the greatest number of small fragments of meteoric iron were found, would seem to indicate that an extraordinarily large mass of probably ' 500 or 600 pounds (226°796 or 2727156 kilos) had become oxidized while passing through the air and was so weakened in its internal structure that it had burst into pieces not long before reaching the earth. Art. XLV.—The South Trap Range of the Keweenawan Serves; by M. E. Wapbsworrn, State Geologist of Michigan. In a former communication published in the August num- ber of this Journal, it was shown that the eastern or supposed Potsdam sandstone, east of the copper-bearing rocks, underlies, in an apparently conformable synclinal fold, a limestone of Trenton or of some adjacent Lower Silurian formation. It was then suggested that the contorted state of the sandstone might have some weight in deciding the relative age of the eastern sandstone and the adjacent copper-bearing rocks. In endeavoring to contribute something to the solution of the relation of these two series of rocks, a party under the charge of Mr. A. E. Seaman of the Michigan Geological Sur- vey was directed to go to “Silver Mountain,” and thence to study the “South Trap Range,” in order to ascertain, if pos- sible, the exact relations of the lava flows of that range and the eastern sandstone. Part of this work has been done, and, * Attention may be called to the discovery by Haidinger (1846) of cubic crystals of a graphitic carbon in the Arva meteoric iron, and also of somewhat similar erystals from the Youngdegin, W. Australia iron, described by Fletcher (1887) under the name of cliftonite. Both have been regarded as pseudomorphs after diamond. 418 Wadsworth—Trap Range of Keweenawan Series. although far from being as decisive as could be wished, yet the observations would appear to be of considerable interest and importance. “Silver Mt.” (See. 1, T. 49, R. 36 W.) was found to be com- posed of interbedded lava flows, of which at least ten flows were made out with more or less certainty. These flows dip to the northwest at an angle of from ten to sixteen degrees. No sandstone was found nearer than two miles. This has a slight dip to the northwest. On Sec. 29, T. 47, R. 37 W., a series of melaphyr flows were observed dipping at a low angle to the north or a little west of north. The angle of their greatest inclination being from 15° to 20°. These flows are interbedded with sandstone which holds fragments of the melaphyr. A felsite dike also cuts through the beds. Similar lava beds are found on Sec. 25, T. 47, R. 38 W., and Sec. 80, T. 47, R. 37 W., which lie at a low angle, 9° to 16° north or a little west of north; while on See. 1, T. 46, R. 39, the flows dip from 9° to 20°, the principal dip being to the north at an angle of 14°. Outerops of the same old basaltic rocks occur on Sec. 35, T. 47, R. 88 W. and See. 8, T. 46, R. 39 W., which show a very low inclination. The most important observations were made in Sees. 11, 13 and 14, T. 46, R. 41 W., where the sandstone was found over- laid by some of the lava flows. The sandstone is found in con- tact with schists presumably of the Archean or Azoie Age. The base of the sandstone is of a conglomeritic character com- posed of rolled pebbles of quartz cemented by an argillaceous matrix formed from the debris of the underlying steeply in- clined and contorted schists. This sandstone dips at a low angle of from 12° to 14° northerly, its strike being S. 60° E. The conglomerate passes into a coarse reddish sandstone which can be traced in pits and exposures northwesterly, where the same coarse red sandstone is seen to pass up into a fine grained indurated sandstone or quartzite, which in its turn passes into a fine-grained indurated argillaceous schist and chert. This indurated zone is in immediate contact with the over- lying lava flow of the south trap range. The structure here is apparently that of a series of flows arising along a pipe or fis- sure, and shows the remains of the solidified neck with the downward bent sandstone or schist strata, together with the strong induration produced by the overflowing Java. The str ucture is indicated in the accompanying figure. It may be remembered that this structure is similar to that observed by the present writer in 1879, on the Douglass Houghton and Hungarian Rivers, except that at the latter two places much decomposition has occurred, leaving it a disputed A. Cary— Geological facts on Grand River, Labrador. 419 question whether the superposition of the lava on the sand- stone is due to its having flowed over it, or to a reversed fault. In connection with the above it may be pointed out that the eastern sandstone on Traverse Island, in Keweenaw Bay, was found by the Michigan Geological Survey to dip westerly at an angle of from five to fourteen degrees, and that the present writer showed that the eastern sandstone in the vicinity of Torch Lake, generally dipped from five to twenty-three de- grees northwesterly toward the copper-bearing series, and that it actually passed under the lava flows. 1, lava flows; ¢, cherty bands; s, s, indurated sandstones. The above observations would go to show that the lava flows of the “South Trap Range,” east of Lake Gogebic do not dip at a high angle, as has been generally asserted, and further that the eastern sandstone is not horizontal, as has been generally stated, but that the two dip at a low angle, generally 5° to 20°. These observations also indicate that the eastern sandstone, and the lava flows of the South Trap Range are one formation, and are as conformable as eruptions of lava ean be with a contemporaneous sedimentary deposit. The study of the South Trap Range will be continued. Michigan Mining School, Houghton, Mich., October Ist, 1891. Art. XLVI.—Geological Facts noted on Grand River, Lab- rador ;* by AUSTIN CARY. THE map of Labrador shows on its eastern coast one deep indentation. This body of water, comprised of Hamilton inlet and Lake Melville, is 140 miles long in all, and washes at almost every point the Archean rock of the country. * Prof. Leslie A. Lee in planning the Bowdoin expedition to Labrador the past summer determined to send a party up the Grand River to investigate its falls _ and obtain such scientific information as might be possible. This paper embodies the geological facts noted by that party. ‘Their meagerness and lack of detail must be largely attributed to the hurried nature of the trip and the serious acci- dents met with. 420 A. Cary—Geological facts on Grand Riwer, Labrador. Lake Melville receives at its head three large rivers. One of these, the Grand or Hamilton river, the largest in the peninsula, prolongs for many miles the general westerly trend of the inlet. Not only this, but the valley in which it flows is a continuation of the basin of the inlet, largely similar in char- acter, direction and width. For sixty miles the river flows on loose sedimentary material, lying again between steep rocky walls nowhere less than six or eight miles apart. The conti- nuity of this valley, from this point to the open sea seems evi- dent. It is a wide trough, 200 miles long, cut into the edge of the Labrador plateau and through its outlying hills. Seldom does the river in this region touch the rocky wall, but at a point 25 miles from the mouth it has dug into the southern wall, and a remarkably round gneissic hill some 400 feet high has been formed. Here also a half mile of fall and rapid makes a drop in the river of 70 feet. At the bottom of the section of sedimentary material thus exposed, fossiliferous Champlain clays were found, the total height of the section at this point being something over 200 feet. Toward the river mouth, it gradually drops to the level of the stream, while 40 miles farther up terraced banks of sand rise to a height of 400 or 500 feet. At a point between 60 and 70 miles from Lake Melville the sides of the rocky valley approach till they are but about a mile apart. This is well within the Labrador plateau which in this region is tolerably level, so that from the deeply sunk river bed its edges have the appearance of high, steeply slop- ing ridges. Parallel, and from a half mile to three miles apart, they extend for more than 200 miles, their regularity broken only by the deep-worn valleys of the largest streams, and by occasional perpendicular bluffs. Changes in direction are generally slow and easy. Marks of former deposit and wear are everywhere. Sand ter- races border the river in quiet regions, while beaches of water- worn stones mount the sides of the valley to a great ‘height. Typical potholes were noted in one place 50 feet or more above the present river level. It is worth remark that while the general height of the plateau, as set by a former traveler* is 2000 feet; this altitude is not generally gained by a single slope. At many places when the bank rises by. a steep angle or a bluff to a height of 500 or 600 feet, the remaining height is gained by a much more eradual slope. The Grand River in this region flows through one large lake called Waminikapou. This is but a portion of the river val- ley 40 miles long from which the loose material has been * Holme, Proc. Roy. Geog. Soc., April, 1888. A. Cary—Geological facts on Grand River, Labrador. 421 cleared out. From one to three miles wide, it contracts to about a quarter mile at its outlet, where the current passes out between perpendicular rocky bluffs, the talus from which serves in part to dam up the water. Of the geological features observed on this river the great cafion at the head of our travel is judged the most remarkable. _ At the upper end of this structure the river, which above here has been flowing on the plateau level, makes an abrupt drop and flows off with many sharp turns, a succession of falls and rapids, between abrupt walls. These walls, without a single break, continue for 20 miles, during which they are very often absolutely perpendicular, and at few points so slop- ing that it is possible to reach the river’s bed on foot. About 100 yards wide at the bottom, the gorge at its head is 150 feet deep, at its foot as much as 800. Grand in dimensions and unique in character as is this gorge, it has never been appre- ciated by the few men who have seen it. We suggest for it the name “ Bowdoin Cajfion.” 3 At its mouth the cafion opens into the side of the river val- ley described above and at right angles to it. The difference in structure here is very marked. The broader valley extends both ways the same in direction and character; but while, as seems probable, the main drainage of the country flowed origi- nally through this channel, it now holds but a small stream compared with the volume pouring out of the cafion. Several interesting facts were noted at the fall which seem to determine its present position. The river above this point is flowing on a hard, moderately coarse syenite which is hori- zontally jointed. At the crown of the fall the jointing, as is shown by a very plain section, takes a gradual curve. This curve the water follows downward until having reached a very considerable angle, it takes a perpendicular drop. The walls of the basin into which the river falls, while inaccessible to close inspection, were intersected by what appeared to be two or three trap dikes; while just here also was a region of special jointing and seaming. Somewhat below the fall the rock was noted as having changed to a syenitic gneiss. Our party spent four days in travel on the plateau in the neighbor- hood of the fall and cafion. So far as observed the plateau surface is worn down to a pretty even general level with per- petual minor elevations and depressions. Almost its whole surface is covered with angular bowlders. One rounded hill, from 500 to 800 feet in height was ascended, by far the highest elevation in a radius of many miles. It was christened by the party Mount Hyde. Glacial markings and bowlders were found on its summit. 422 Scientific Intelligence. SCIENTIFIC INTELLIGENCE. I. CHEMISTRY AND PHYSICS. 1. On the so-called “ Black Sulphur” of Magnus.—Kwarr has examined the mode of formation and the properties of the so-called “black sulphur ” first described by Magnus. He finds that the best mode of preparing it is to heat suddenly a mixture of sulphur and oil ; such for example as is obtained by dissolving almond oil (0°2 to 0-4 gram.) in ether, mixing the solutien thor- oughly with 50 grams flowers of sulphur and then evaporating the ether. Ifa small portion of this mixture be dropped from the point of a knife on to the bottom of a red hot platinum cruci- ble, most of it is volatilized, but there is left a loose black residue, which after cooling is removed. By repeating the operation, the product may be increased indefinitely. Even with the greatest precautions, the yield is very small, only 0°685 gram having been obtained in this way from 100 grams of sulphur and 0°4 gram of oil, as a mean of 23 experiments. The oil treated alone in this way gave only’ 0-011 gram of residue. As thus obtained the “black sulphur” contains some yellow sulphur, and if the mate- rials were not pure, also some ash. Its density varies from 2°622 to 1843, this want of homogeneity being due to the difficulty of reculating the temperature. It is insoluble in hot and cold water, alcohol, ether, hydrochloric acid, nitric acid, aqua regia, ammonia, caustic alkalies even when concentrated, and potassium cyanide. On evaporating it with potassium hydroxide solution, it is at- tacked just as the hydroxide is becoming solid and dissolved to a humus-brown mass soluble in water with a deep brown color, ‘ yielding a solution in which acids give a brown precipitate. On heating the black sulphur in the air it undergoes no change other than the loss of moisture and yellow sulphur until the tempera- ture reaches 200°-300°, when sulphurous oxide is formed. Ata red heat the mass takes fire. In the absence of air, the black sulphur slowly loses weight and continues to do so for many hours even at ared heat, no vapor of sulphur or empyreumatic vapors being evolved. After the weight becomes constant, the residue burns away in the air entirely. The amount of this resi- due was 43°59 per cent. Its quantity did not seem to vary when the heating took place in carbon dioxide and no brown sulphur vapors appeared in the vessel, although considerable potassium sulphide was formed when the delivery tube from this vessel was conducted under potassium hydroxide solution. The sulphur in the “black sulphur” was estimated by evaporating a known portion with potassium hydroxide and potassium nitrate solutions and heat- ing, until the residue was white. Three determinations gave 53°77, 56°76 and 57:07 per cent of sulphur, respectively. The residue left at a red heat contained 22°78 per cent of sulphur. Thus 44-46 per Chemistry and Physics. 423 cent of the sulphur is expelled at a red heat, while the residue contains 10 per cent only of sulphur mixed with 33-34 per cent of carbonaceous matter. Only 13-14 per cent of the total sul- phur is dissolved by potassium dichromate and bromine. Hence the author concludes that the ‘black sulphur” of Magnus is not in itself a modification of sulphur, but consists of such a modifi- cation either adhering to-or condensed with, a carbonization pro- duct of the oil, itself containing sulphur. The new form of sul- phur does not vaporize below a temperature which is far above the boiling point of yellow sulphur. Moreover its vapor is color- less and not brown; and it evolves sulphurous oxide below a visibly red heat without combustion. ‘Though evidently unstable in the free state, it can exist readily in contact with any surface which affords a substratum for its development. This the author finds in the blue color developed by sulphur on platinum, silver and lead sulphide, and in the blue of ultramarine.—J/. pr. Ch., II, xliii, 305; J. Chem. Soc., 1x, 877, Aug. 1891. Gy FB. BE, 2. Ona new form of Silicon—W aRREN has described a new form of silicon crystallizing in well-defined oblique octahedrons, obtained by subjecting potassium silicofluoride to an intense heat in contact with impure aluminum. On separating the graphi- toidal silicon thus produced, by the aid of acids, the new variety was obtained though only in small amount. The yield was in- creased by proceeding as follows: Pieces of aluminum the size of a walnut were thrown into a clay crucible containing a mix- ture of 4 parts of potassium silicofluoride, one part potassium carbonate and 2 parts potassium chloride, in a state of fusion. After the violent action was over, the crucible was heated to whiteness for about five minutes. On cooling a button was ob- tained containing 80 per cent of silicon. This was placed in a plumbago crucible with 12 parts of aluminum and 2 parts of tin and the whole was covered with a layer of sodium silicate. After heating to the highest attainable temperature for two hours, the crucible was cooled and the aluminum button was broken. It contained the new modification of silicon in large perfect crystals, having a full metallic luster and resembling the erystals of cast iron seen on breaking a pig of this metal. The silicon crystals are infusible and insoluble in all acids except hydrofluoric.— Chem. News, Ixiii, 46; J. Chem. Soc., |x, 799, July, 1891. G. F. B. 3. On a new Alkaloid from Conium maculatum.—The new alkaloid which was observed in Coniwm maculatum by Merck of Darmstadt, has been submitted to a careful chemical examina- tion by LaprENnBuRGe and Apam. As received it was a white powder, easily soluble in water, alcohol, ether and benzene and forming salts with acids. The new base was purified in two ways: first by distillation and second by recrystallization from toluene. Both portions had the same melting point and both gave on analysis similar numbers: carbon 66°66 and 67°14, and hydrogen 12°33 and 12°35, the nitrogen being 9°88. This agrees Am. Jour. Scl.—Tuirp Serizs, Vou. XLII, No. 251.—NovemBer, 1891. 29 : 424 Scientific Intelligence. with the formula C,H,,NO which is that of conhydrine. Being quite similar to this base the authors call it pseudoconhydrine. It boils at 229°-231°, the distillate solidifying crystalline, and fusing at 100°-102°. It is optically active, ap = 4°30°. Its con- stitution has not been fixed.—Ber. Berl. Chem. Ges., xxiv, 1671, June, 1891. G. F. B. 4. On TIron-tetracarbonyl and Nickel-tetracarbonyl.—Monp and QurINcKE have succeeded in forming a compound of iron with carbon monoxide analogous to the compound of nickel with this gas described by them in conjunction with Langer. When very finely divided iron, obtained by reducing iron oxalate in a current of hydrogen at the lowest possible temperature, but little over 400°, and then cooled in the gas to 80° is treated with car- bon monoxide, the escaping gas imparts a yellow color to the flame of a Bunsen burner into which it is passed, this result con- tinuing even at ordinary temperature for several hours. On pass- ing the gas through a heated glass tube at 200° to 350° a metallic mirror is formed, while at higher temperatures black flakes are . produced. Upon dissolving these mirrors in acid, the solutions gave all the known reactions of iron in remarkably brilliant colors. A solution in nitric acid, precipitated by ammonia, weighed and reduced by hydrogen gave 70°48 and 69°94 per cent of iron in the oxide in two cases, the theoretical amount being 70°00. Even under the most favorable circumstances, however, the quantity of iron obtainable in this way is very small. On treating 12 grams of finely divided iron with carbon monoxide for six weeks, only about two grams of the metal volatilized. After a time the action diminished in intensity, and it was found necessary to heat the iron in a current of hydrogen at 400° for about twenty minutes every five or six hours. When 23 liters of carbon monoxide passed per hour over the iron, the issuing gas contained not more than 0:01 gram of iron per liter; equivalent to less than 2 c.c. of Fe(CO),. Sulphuric acid absorbs the gas completely, but the solution decomposes very rapidly. Benzine and heavy mineral oils partially absorb it, producing tawny col- ored solutions which decompose on exposure to the air with sepa- ration of iron hydroxide. The analysis of the gas was effected by saturating a mineral oil with it by passing it through the oil for 8 to 16 hours, raising the temperature to 100° under a pres- sure of 500 mm. of mercury to free it from air and dissolved OO, and then to 180°, at which temperature the iron compound was decomposed, the carbon monoxide being evolved and the iron separating in the metallic form. The ratio of the CO to the Fe was found to be in five experiments 4°144, 4:030, 4°150, 4°264, and 5042 ; thus rendering it probable that the substance is iron-tetra- carbonyl, corresponding to the nickel compound. ‘The authors suggest that this iron compound may play a part in the process of cementation.—J. Chem. Soc., lix, 604, Aug., 1891 ; Ber. Berl. Chem. Ges., xxiv, 2248, July, 1891. G. F. B. TS oy eda & Chemistry and Physics. 425 BeERtTHELOT also has observed the formation of this iron-car- bonyl compound. If finely divided iron, obtained by reducing the precipitated oxide at ‘the lowest possible temperature or by igniting the oxalate in hydrogen, be treated at 45° with carbon monoxide, the escaping gas contains iron and burns with a brighter, sometimes whiter, flame than the monoxide itself. This flame produces spots on a porcelain surface held in it, these spots consisting of iron and its oxide. On passing the gas through a heated tube a metallic ring is deposited consisting of iron containing carbon. With concentrated hydrochloric acid, the gas gives ferrous chloride. On standing over water contain- ing air, iron oxide is deposited. The quantity of the iron com- pound is very small. Berthelot has also examined nickel-tetra- carbonyl and finds that it is permanent and without marked dissociation-pressure at ordinary temperatures. When preserved under water it is not decomposed if air be absent. When heated suddenly to above 60° it explodes, producing carbon and carbon dioxide besides nickel and the monoxide ; according to the equa- tion C,0,Ni= (CO,),+C,+Ni. It is not soluble in water, dilute acids or alkalies or acid cuprous chloride ; though hydrocarbons, especially oil of turpentine dissolve it. Mixed with air or oxygen it detonates on ignition and sometimes spontaneously as when the dry gas mixed with oxygen is agitated over mercury. In pres- ence of water, a greenish-white gelatinous precipitate is formed which contains nickel, oxygen, water and combined carbon, and which deposits carbon on heating. In the liquid state, nickel- tetracarbonyl shows similar reactions. Concentrated sulphuric acid explodes it, while its vapor when mixed with nitrogen is gradually converted by this acid into four times it volume of carbon monoxide, the nickel going into solution. Ammonia does not act at once on the pure gas, while if oxygen be present white fumes appear immediately. Hydrogen sulphide gives a black sulphide, hydrogen phosphide a black mirror-like deposit. Nitro- gen dioxide when mixed with the liquid or its vapor, produces a blue cloud which gradually sinks to the bottom of the vessel ; the resulting gaseous mixture containing nitrogen dioxide, car- bon monoxide and a new nickel compound. The author calls at- tention to the analogy between carbon monoxide and the radicals contained in the so-called metallo-organic compounds.—C. R exil, 1343; Ber. Berl. Chem. Ges., xxiv, Ref. 593, July, 1891. G. F. B. 5. On a sensitive Reaction for Tartaric acid.—MouLER has observed that when crystals of tartaric acid are thrown on sul- phuric acid of 66° B., containing one per cent of resorcinal, and the whole is heated gradually, a fine red-violet coloration is pro- duced when the temperature approaches 125°; complete carboni- zation taking place at 190°. Water destroys the color. The coloring matter could not be isolated since it was not soluble in ether, amyl alcohol, acetone, chloroform, or benzine. Using other ‘phenols, similar colors are produced ; phloroglucinol giving a red 426 Scientific Intelligence. and pyrogallol a fine violet color. Since these reactions are not given by succinic, malic, citric or benzoic acids, tartaric acid may readily be detected when mixed with any of these acids. To de- tect 0°01 milligram, the author evaporates the solution to be tested to dryness, one c.c. of the resorcinal solution is added, and the whole is gradually heated to about 125°. At first reddish streaks appear and then the. sulphuric acid becomes colored | throughout. If organic substances which char with sulphuric acid are present, the tartaric acid is removed by precipitation as lead tartrate and then tested, nitrates and nitrites should not be present.— Bull. Soc. Chem., Il, iv, 728; J. Chem. Soe., lx, 867, July, 1891. G. F. B. 6. Photography of the Spectrum in natural color.—H. W. VoGEL gives a historical account of the photography of color and an explanation of the failures to accomplish it. It appears that Zenker, in 1868, indicated the method of depositing layers of silver of suitable thickness to produce by interference of light colored photographs, a method which Lippman has lately devel- oped. Lord Rayleigh’s (1886) explanation of the colors in pho- tographs produced by adjusting the layers of silver to wave lengths in order to produce colors by interference is a repetition of the explanation of Zenker. In the earlier processes Ag,Cl was used in the sensitive film and the fixing of the image produced in this film, by hyposulphite of soda, destroyed by separation of fine silver particles the regular layers which were necessary to pro- duce interference colors. Lippman uses pure bromide of silver which, under the operation of fixing, leaves the film in homoge- neous, regular layers suitable for producing interference colors.— Verhandl. d. Physik. Ges. Berlin, 10, p. 33, 1891; Photogr. Mittheil., 28, p. 7. a a 7. Discharge of Electricity through exhausted Tubes without Electrodes.—J. J. THomson points out that the oscillations of the discharge from a Leyden jar produce during the short time of their duration enormous currents in the wire connecting the coatings of the jar, and therefore produce by induction very great electromotive force in the neighborhood of the wire. He therefore investigates the discharge by induction in rarefied ves- sels by wrapping these vessels with the wire connecting the coat- ings of a Leyden jar; thus producing luminous discharges in . these vessels without the direct passage of electricity from metal- lic terminals in the gas. Professor Thomson points out that the phenomena bear upon his theory of tubes of electrostatic induction. He regards the distinction between electrostatic and electromag- netic electromotive forces as one introduced for convenience of analysis rather than as having any physical reality. “The only ~ difference which -could be made, from a physical point of view, would be to define those effects as electrostatic which are due to tubes of electrostatic induction having free ends, and to confine the term electromagnetic to the effects produced by closed endless tubes. It is only when the electromotive forces are produced ex- ~ Chemistry and Physics. 427 clusively by the motion of the magnets that all the tubes are closed : whenever batteries or condensers are used, open tubes are present in the field.” The bearing of Professor Thomson’s ex- periments on the aurora is extremely interesting. The most re- markable appearance was presented when the discharge passed through oxygen. In this gas the bright discharge is succeeded by a phosphorescent glow which lasts for a considerable time, some- times for more than a minute. The spectrum of the afterglow is a continuous one, without bright lines. The only gas besides oxy- gen which shows the afterglow is air. The spectrum of the air glow showed bright lines. “Professor Thomson is continuing his investigation.— Phil, Mag., Oct., 1891, pp. 323-336. doa 8. Latio of Electromagnetic to Electrostatic units —J. J. Tomson and G. T. C. Szarte have undertaken a redetermina- tion of the value of this ratio. A complete account of their ex- periments can be found in Phil. Trans., Lond., 181 A., pp. 583- 621, 1890. The value obtained is v=2°9955'10"" cm. sec.~’. HB 9. Hxpansion of Water.—A useful table of the expansion of water from temperatures 0° to 31° is given by W. Marek.— Ann. der Physik und Chemie, No. 9, 1891, p. 171. apy i 10. Hxperimenis in Aerodynamics, by 8. P. Lanetry. 115 pp. 4to, with 10 plates. Washington, 1891 (Smithsonian Contri- butions to Knowledge, 801).—When the investigation of a sub- ject like that of “flying machines ”—at once so stimulating to the popular imagination and yet almost an ignis fatuus in the view of sober minds-—is made the subject of careful scientific experi- ment in skillful hands the results are sure to be of unusual interest and value. This is eminently true of Prof. Langley’s investigations in aerodynamics which briefly demonstrate experi- mentally that mechanical flight under proper direction is prac- ticable and further that the support of heavy bodies in the air, combined with very great speeds is not only possible but within the reach of mechanical means now available. The experiments detailed in this memoir were carried on at Allegheny Observatory between 1887 and 1891. They describe in the first place the “suspended plane ”-—a thin brass plane a foot square weighing two pounds hung vertically by a spring - from a surrounding frame and capable of receiving rapid lateral motion. Briefly expressed the important result of the experi- ments is to prove that the downward pressure diminishes as the velocity increases, the spring contracting as the plane is carried forward. A second instrument served to show graphically the direction of the total resultant pressure on a square inclined plane and to roughly measure its amount—this is called the “resultant pressure recorder.” Still another instrument, the ‘“‘plane-dropper,” was used to demonstrate that a horizontal plane in lateral motion requires an increased time for its descent, and also to measure the time of fall for different planes and other related points, thus giving the soaring speeds of wind-planes 428 Scientific Intelligence. set at varying angles and making it possible to compute the work expended in their uniform horizontal flight. Thus it is proved that less work is required in the aerial motion of heavy inclined planes at higher speeds than at lower ones. In the quantitative experiments connected with this part of the subject, a “component pressure recorder” was used together with a ‘“‘dynamometer-chronograph” to record the speed, the resistance to forward motion at the instant of soaring and other attendant phenomena. Reference must be made to the memoir itself for the details of the methods and results of the experiments with the instruments, alluded to. It is interesting, however, to note the conclusion reached, that, “‘so far as the mere power to sustain heavy bodies in the air by mechanical flight goes, such mechanical flight is possible with engines we now possess, since effective steam engines have lately been built weighing less than 10 pounds to an horse- power, and the experiments show that if we multiply the small planes which have been actually used, or assume a larger plane to have approximately the properties of similar small ones, one horse power rightly applied can sustain over 200 pounds in the air at a horizontal velocity of over 20 meters per second (about 45 miles per hour) and still more at still higher velocities.” The author adds further that the experiments “afford assurance that we can transport (with fuel for a considerable journey and at speeds high enough to make us independent of ordinary winds) weights many times greater than that of a man.” He goes on to say (we quote the author’s words) that he has “not asserted without qualification that mechanical flight is practicably possi- ble since this involves questions as to the method of constructing the mechanism, of securing its safe ascent and descent and also of securing the indispensable condition for the economic use of the power I have shown to be at our disposal, the condition, I mean, of our ability to guide it the desired horizonal direction during transport—questions which in my opinion are only to be answered by experiment and which belong to the inchoate art or science of aerodromics on which I do not enter. I wish, how- ever, to put on record my belief that the time has come for these questions to engage the serious attention not only of engineers but of all interested in the possibly near practical solution of the problem, one of the most important in its consequences of any which has ever presented itself in mechanics ; for this solution it is here shown cannot longer be considered beyond our capacity to reach.” 11. The Chemical Analysis of Irnon.—A complete account of all the best known methods for the analysis of iron, steel, pig- iron, iron-ore, limestone, slag, clay, sand, coal, coke and furnace and producer gases by ANDREW ALEXANDER Brarr. Second edition. 314 pp. Philadelphia, 1891 (J. B. Lippincott Com- pany).—The first edition of this valuable and attractive work was noticed in volume xxxvi (p. 387) of this Journal. In the ss Geology and Mineralogy. 429 present edition some new analytical methods have been added, the table of atomic weights has been revised and the errors noted during its use for the past three years have been corrected. 12. Die Fortentwickelung der elektrischen Hisenbahn-Einrich- tungen, von L. Koutrtrsr. Vienna, 189 (A. Hartleben’s Ver- lag).—This volume is published in the same form as those of the ‘“‘ Hlektro-technische Bibliothek ” repeatedly noticed in this Journal. It is devoted to the various applications of electricity to railroad traffic, in the telegraph, telephone, signals, etc., and ‘gives much information on these practical subjects compressed into a small space. Il GroLnocy AND MINERALOGY. 1. Report of Hxploration of the Glacial Lake Agassiz in Manitoba; by Warren Urpnam. 156 pages 8vo, with two maps and a plate of sections; forming Part EK, Annual report of the Geological and Natural History Survey of Canada, vol. iv, for 1888-89.—The departure of the ice-sheet of the Glacial period is shown to have been attended with the formation of a vast lake in the basin of the Red River of the North and of Lake Winnipeg, held by the retreating ice-barrier. It exceeded in extent the combined areas of the great lakes tributary to the St. Lawrence, and had a maximum depth of about 600 feet. Seventeen shore-lines, marked by beach-ridges of gravel and sand, are found at successive levels upon the northern part of this lacustrine area which are referable to stages of the glacial lake while it outflowed southward by way of Lakes Traverse and Big Stone and the Minnesota River. At lower levels, eleven later shore-lines belong to stages of outflow northward, previous to the recession of the ice from the region crossed by the Nelson River, whereby Lake Agassiz was reduced to Lake Winiipeg. The earliest and highest beaches have a gradual ascent of about one foot to the mile northward along an explored extent of 400 miles from south to north ; but in the lower beaches there is.a gradual decrease of this ascent, and the latest and lowest beaches are very nearly level. It is thus known that the area of Lake Agassiz was undergoing a differential northward uplift during the time of the ice-departure, and that the uplift was nearly completed within that time. On the adjoining country of Min- nesota and North Dakota eleven distinct terminal and recessional moraines indicate the maximum extension of this ice-sheet and stages of halt or re-advance interrupting its general retreat ; and five of these moraines, namely, the Dovre, Fergus Falls, Leaf Hills, Itasca, and Mesabi moraines, were accumulated after Lake Agassiz began to exist in the Red River Valley. An appendix of this report gives a tabulation of glacial striz on the region of Lake Agassiz and the country northward to Hudson Bay and the Mackenzie; and another appendix notes altitudes determined by the Canadian Pacific railway surveys in Manitoba and westward to the Pacific. 430 Screntific Intelligence. 2. Geological Survey of Texas, 2nd Annual Report, 1890. K. T. DumB.ez, State Geologist. 756 pp. 8vo, with maps, plates and sections. Austin, Texas, 1891.—The introductory chapter of this second Annual Report by Mr. Dumble reviews the work of the year, and the subjects of metallic and other mineral and economical resources of the State. It is followed by an account of the geology and resources of the iron ore district of East Texas, by E. T. Dumble, Wm. Kenedy, J. H. Herndon and J. B. Walker; on the geology of northwestern Texas, by W. F. Cummins ; on the geology and resources of the central mineral” region of Texas, by T. B. Comstock ; and on the geology and mineral resources of Trans-Pecos, Texas, by W. H. von Steernu- witz, with a report on the Cretaceous rocks of the region by J. A. Taff. Mr. Cummins, in his account of the Permian—the lower divis- ion of the Red Beds,—makes them in places 5000 feet thick, and every where conformable with the Carboniferous. He divides the formation into the Wichita or Lower, consisting of sand- stones; the Clear Fork beds, limestones, shales and sandstones, and some gypsum; and the Double Mountain beds, including limestones, shales and thick beds of gypsum. The overlying Triassic commences with sandstones and conglomerates, which re- semble and are supposed to be the equivalent of the Shinarump conglomerate of Powell—made the beginning of the Trias by Mr. C. D. Walcott. The Permian series is not separable from the Triassic by any marked unconformability, yet it is evident, Mr. Cummins remarks, that there was not continuous sedimenta- tion between the two. 3. Preliminary Notice of a New Yitrium-Silicate; by W. EK. Hippen. (Communicated).—Associated with the huge crys- tals of gadolinite, with yttrialite and the other yttrium minerals, found in Llano County, Texas, two years ago, I have discovered a few masses of a new species that is exceedingly rich in the yttrium earths. A preliminary examination has shown its den- sity to be 4°515. Its color is pale drab-green when pure. In thin splinters it is perfectly transparent. Its alteration products are of a waxy brick-red color and quite easily distinguished from those of gadolinite and allanite. It is easily soluble in acids, leaving gelatinous silica. The following are the results of an unfinished analysis by the writer : SLO) cy een 8 ir |e ae 25°98 JO. ebt gmerees ore, 2. Sane 61°91 atomic weight =118. BPe® . 22) eee oy Sire, ales 4°69 DOk: © Si Seer ts ee 0°40 Cd os. cr be ea 0°19 Ign, - loss 2 Sees a. oe 2°01 No thoria is present and but very little of the cerium earths. The oxygen ratio of the bases found to the silica is 83°47 : 86°60, or pointing to 1:1 if the analysis had been completed. Its for- Geology and Mineralogy. 431 mula would then be 3 SiO,, 2R,O, or of a mineral quite distinct from the gadolinite and yttrialite with which it is found asso- ciated. For this silicate so remarkably rich in yttria, I propose the name of Rowuanpire, after Professor Henry A. Rowland, whose spectrographic work on the so-called “rare-earths” is so novel and important. As opportunity offers a more extended - description will be given of this very interesting new species. 4. Anatase from the Arvon Slate Quarries, Buckingham Co., Va; by Grorcre H. Wiriiams (communicated).—The rarity of American localities for anatase is a sufficient warrant for recording a recent discovery of this mineral in its original posi- tion, made by the writer during June last. In the course of a trip through central Virginia occasion was taken to visit the State quarries five miles south of Bremo Bluffs on the James River railroad in Buckingham County. The largest of these quarries, belonging to the Williams Brothers, is situated at the terminus of the short branch railroad, Arvon station. This con- tains-the best quality of slate, but it is proportionately devoid of anything of mineralogicai interest. About a mile west of this place, however, where the slate of this district was first opened in what is now known as the Robert’s quarry, the cleavage is less perfect and regular, while cross joints are of frequent occur- rence. ‘These irregularities, which detract so seriously from the economic value of the slate, make this quarry more interesting than the other to the geologist and mineralogist. Here beauti- fully crinkled varieties of slate occur, and one regularly mottled sort is quite abundant, which in the field was surmised to contain ottrelite, but was found on more careful examination to owe its knots (“‘knoten”) to small rhombohedrons of some carbonate which is but feebly transparent on account of the great number of inclusions, probably of carbonaceous matter, which it con- tains. Huge blocks of this imperfect or “ bastard slate” have been thrown aside as worthless, and it was on the end of one of these, cut off very evenly at right angles to the cleavage by a cross joint, that the anatase crystals were found. The surface presented by this joint plane was of large size and was completely covered with small quartz crystals, among which were scattered minute individuals of pyrite and the anatase. The latter was fairly abundant and closely resembles the black, metallic, steep pyramidal variety, so well known from the Tavetsch valley in Switzerland. Hardly any crystals were noticed over a millimeter in length, while most were less than this. No forms except the unit pyramid, 1 (111), and the base, 0 (001), were observed. The pyramidal faces are horizontally striated and often built up into little flights of steps by an oscil- latory combination, as is so frequently the case with the Swiss crystals. The faces have a high metallic luster but are broken by growth, irregularities, and vicinal planes, which makes the reflected images multiple and the measurements unsatisfactory. The best crystal gave : 432 Scientific Intelligence. (111) : (111) OB ati” 97° 51’ (cale, v. Kok.) (111) .a(ODIg AIL 36") ADL Ag” (101) SC 136° 20". Jase The smaller crystals when placed under the microscope are found to be translucent with a rather pale yellow color, metallic lustre, and high refractive index. These show parallel extinction and a uniaxial figure. I am indebted to my friend, Prof. W. G. Brown of Washington and Lee University, for chemically examining one of the crystals, which he found to be composed largely of titanic oxide. Baltimore, Aug., 1891. 5. Ilvaite ; by G. Cu. Horrmann (communicated).—Several specimens of what proved to be the rare mineral, ilvaite, were received for identification from a gentleman who described it as occurring in large irregular masses in a vein about twenty feet wide, near the head of Barclay Sound, Vancouver Island, British Columbia. ‘Portions of the material were fairly free from for- eign admixture containing only small quantities of a white trans- lucent, cleavable calcite, this, however, was in some fragments supplemented by inclusions of altered tremolite, and in others by a brownish-yellow andradite. It had a more or less closely com- pacted crystalline structure. The lateral faces of crystals were not infrequently striated longitudinally, and sometimes exhibited a slight iridescent tarnish. Color, iron-black ; streak, greenish- ene luster, sub-metallic, brittle ; fracture, uneven. Before the blow-pipe fuses quietly at about 2°5 to a black magnetic globule. Hardness, 5°5 ; specific gravity, 3°85. Readily decom- posed by hydrochloric acid, forming a yellow jelly. An analysis conducted upon very carefully selected and pre- pared material, dried at 100° C., afforded the following results : SiO, Al,O3 Fe.0; FeO MnO CaO MgO H.O 29°81 0°16 18°89 32°50 2°22 13°82 0°30 1°62:=> 89722 6. Synthese du Rubis, par EK. Fremy. 30 pp. 4to, with 28 colored plates. Paris, 1891, (Vve. Ch. Dunod.)—The synthetic formation of minerals in general is a subject of great interest and one in which French chemists have made remarkable progress of late years. The results reached by one of the laborers in this field, M. Fremy, in the artificial production of rubies are given in this beautiful volume. In the most successful method, the rubies were obtained in an earthen crucible by the reaction at a very high temperature of a mixture of alumina (with more or less potash) upon barium fluoride, with bichromate of potash as coloring matter. They are well crystallized, clear, of brilliant color and sometimes weigh one-third of a carat. The author claims for them usefulness both in jewelry and in watchmaking. A series of fine colored plates show sections of the crucibles with the rubies scattered through the gangue, also clear isolated rhom- bohedral crystals (magnified), and further, the rubies cut and mounted for ornament in various forms. Geology and M ineralogy. 433 7. Brief notices of some recently described minerals.—BRAND- titE. A hydrous arsenate of manganese and calcium, formula Ca,MnAs,0,+2H,O, found at the Harstig mine, near Pajsberg, ‘Sweden. It is analogous to roselite and fairfieldite in composition and closely similar to the former species in its triclinic crystals. The color is white with vitreous luster, hardness =5-5'5; sp. gravity =3°671. An analysis gave: MOO. | MnO CaO JPbO. FeO ~MeO Cl 4,0 insol. nO eens. td035 25:07 096 > 0:05 40°90. 0°04 -°8°09 0:04 = 99-71 - Named by Nordenskidld and later described by Lindstrém.— G. For. Forh., xiii, 123, 1891. GANOPHYLLITE.—A hydrous silicate of alumina and manganese from the Harstig mine, Sweden. It occurs in monoclinic crys- tals with perfect basal cleavage. The color is brown ; hardness =4-4'5 ; sp. grav. =2°84. An analysis gave: si0, t PRM a eure tec rates bos fen ; ; . A , 7 1) ‘F z ae oi une a oka dried: aed nk : is rite e Te th te ie i im: 4 i « + : ey ee ee SES spit G" if hee RAT aOR, | fer Ne ‘ott RE (Et Rad aS Hen tEN ier: ra i aeut het Y 2% rd : &% ge ' ; ; Sa as rer) iva ot eee ioe : R cartes | ee , 2 rol. aa \ae pen ae 4 ional? “rlyaltaes wie ; a | i a5 _ ~ oy, bet. Ager Se ivy ys ny > a a? ~ » oh . as? | Roe ® a Pi : wet g CaS bes ire 4 * * * Am. Jour. Sci. Vol. XLII, 1891. Plate |. oe ae ; ‘ ‘ é " 4 + a is E ie) : ‘ . - © - 4 ‘ v / , ! ; . f . j , ‘ ‘ e ; J ~ t * i 4 ® ( a i ‘ na - . j 2 , , ' . ‘ “ . ’ . ‘ a 4 ‘ - s ‘ m, -~ ty ™ \ : ‘ ‘ v ‘ 4 ‘ ’ * ™ 4 y a * Ms afl ‘ . AM. JOUR. SCI. VOL. PLATE I] ois Py j i ne e Al AM. JOUR. SCI. VOL. XL. Gilli ayy Z ale st B ey nik Wii 5a Mt NEW HAVEN REGION BEFORE 1640 1 Inch=3 miles Heights above mean tide level. Areas of trap colored red, — 17 \ = Wy WW = = _ . LVE HANS STATE ST. | Wir, i} ay alt OPAL UO Wudy pourasva 9 499F 008 (UVa ALIO VY MON) HOON LSVA ‘TOS "HOOr ‘WV I GALyId AM. JOUR. SCI. VOL. XLI. TNs \\ \ A \\ PLATE IIl. EAST ROCK NEW HAVEN (Now A OLTY PARE) Scale 1 inch =800 feet Heights reckoned from High Tide Ss - Sandstone outcrop 8’8” Sandstone in fragments A & Northern Trap-mass B BY East Rock Trap-mass C C Indian Head Trap-mass D D’ Snake Rock Trap-mass W H Whitney Ridge and Dike Present condition of the Rock at itsS S W Angle English Drive Farnam Drive Refreshment House Soldiers Monument QUARRY CORNER. 209 ae tok o's Pi ‘ ‘ “/ : a j . Y. a =) » af “4 i 32 ae ies be ; a a ra a "a 2 ts me 3 . fe a Bee a Pens, a % : i P ‘ +) ' oe) Ye g . : F Fa " ~ 3 4 ” ; 5 i , : iN na ee i f : ' hy eV: ‘oe } f = a tw 4 ie J 14 j i { ia ! a Ny C a ae mo y ' 5 Bly ; { * i Am ’ fe" we Cydrasojyoyd & W014) ‘ASP 109119 ISUBIO TBO ‘YSOMTINOS OY} WIOAF YOOY IVA JO Mora 1681 [1X ‘JOA “19S “NOP ‘wy ) " ~ 7 « \ » a 2 ‘ ty ges ie x: yal 2 - J! a cone yeh bial aig cl eto eet eae ou Am. Jour. Sci., Vol. XLII, 1891. Plate V. oe Mace i) Vn ULI. Profile view of columns, East Rock, near the house on the brow of the Rock in Plate IV (From a photograpb.) niin SSS iiiisiss« Restoration of STEGOSAURUS UNGULATUS, Marsh. Al t ae One-thirtieth natural size. Plate X. Am. Jour. Sci., Vol. XLII, 1891. Plate XII. Procamelus, uS. Tyracodon, itanops, Titano- Bilotherium. m, Palewosyops, Ss. ngulates, , Dryptosaurus. n, Selenacodon, opleryxz. 1ornis. us, Tylosaurus. urs. rus, Diplodccus, u7us, Mammals, tenacodon. MIMOSAULUS. dents), } Am. Jour. Sci., Vol. XLII, 1891. VEEP FALLS Plate’ XI. GEOLOGICAL Map STEEP-ROCK LAKE Scace oF FeeT. Le l= ty | a H - = ——- = Cc (‘om vwvyvM wm p SYMBOLS : BASEMENT COMPLEX = STeEEP-RocK SERIES bower \warrow ATIKOKAN Do = = ° Map sHowinG GEOGRAPHICAL Posit1ON OF SOUTH WEGT SteePp-Rock LAKE \BaY ao lee ft ‘ A 6 fa i SG \ > Scace or Mires PK nS o th ZZ Mesozoic. PALEOZOIC. Am. Jour. Sci., Vol. XLII, 1891. Recent. Quaternary. Plate XII. Bos, Equus, Megatheriwm, Mylodon. Equus Beds. Phohippus Beds. Pliocene. Miohippus Beds. Miocene. |Oreodon Beds. Brontotherium Beds is) is S SI S) A GE v Equus, Tapirus, Hlephas. § Pliohippus, Tapiravus, Mastodon, Pracametus, ( Aceratherium, Bos, Morothervm. | Miohippus, Diceratherium, Thinohyus. § Oreodon, Eporeodon, Hyenodon, Hyracodon, ¢« Voropus. § Brontotherium, Brontops, Allops, Titanops, Titano- ( therium, Protoceras, Mesohippus, Hiotherium. Diplacodon Beds. Dinoceras Beds. Bb a CI “™ ~~ iat (3) i Dipilacodon, Epihippus, Amynodon. § Dinoceras, Tinoceras, Uintatheriwn, Palwosyops, Subearboniferous, or Sauropus Beds. Eocene. ( Orohippus, Hyrachyus, Colonoceras. Heliobatis Beds. _| Heliobatis, Amia, Lepidosteus. z Coryphodon, Lohippus, Lemurs, Ungulates Coryphodon Beds. } Tinedonte, Rodents, Serpents. ” : pan 4 . |Ceratops, Triceratops, Hadrosaurus, Dryptosaurus. Laramie Series, or Mammuls, Cimolomys,. Dipriodon, Selenacodon, Ceratops Beds. _|_ Nanomys, Stagodon. Birds, Cimolopteryx. G Fox Hill group. retaceous. are Birds with Teeth, Hesperornis, Ichthyornis. jColorado Series, On Mosasaurs, Edestosaurus, Lestosaurus, Tylosaurus. Pteranodon Beds. Pterodactyls (Pteranodon). Plesiosaurs. Dakota Group. : Atlantosaurus Beds) 4 Dinosaurs, Brontosaurus, Morosaurus, Diplodccus, Jurassic. Baptanodon Beds. Stegosaurus, Cumptonotus, Alosaurus. Mammals, Hallopus Beds. Dryolestes, Stylacodon, Tinodon, Cienacodon. . First Mammais (Dromatherium). Triassic. ete OF Beds Dinosaur Footprints. Anchisaurus, Ammosaurus. (OMA ERUNIE) Sy BIC Crocodiles (Belodon). Permian. Nothodon Beds. Reptiles (Wothodon, Sphenacodon). 1 Coal Measures, or First Reptiles (2?) Losaurus. Carboniferous) Zosaurus Beds. First known Amphibians (Labyrinthodonts), Sauropus. Dinichthys Beds. 4 Devonian, : Lower Devonian. Dinichthys. Upper Silurian. Silurian. —— Lower Silurian. Cambrian. | Primordial. Ecne Archean. | Tanrentian. First known Fishes. No Vertebrates known SECTION TO ILLUSTRATE VERTEBRATE Lire IN AMERICA. : 7 m7 «4 fae Ye . 5 P Plate XIII. Am. Jour. Sci., Vol. XLII. 1891. . = ’ S\N WAT e ee : . ate et ee ee AUT | W'S 77 8 Geert SIZE. TONGANOXIE METEORITE. Fic. I].—Etched surface, reduced one-fifth. Fig. [.—Five-twelfths natural 4 ~ J ” " i me See ae reemapeeyy ae Am. Jour. Sci., Vol. XLII, 1891 Meteoric Ir Larger mas aime Jy hee yi A r BR ae gel Ue \ ay - a Ss y ay te eae 5 - r. Sth Nk 7 = - “ uf s t ~ Ween y ’ i ’ Y- 2 F a. x xs ‘* Me ae Sx 4 a oi Am. Jour. Sci., Vol. XLII, 1891, Plate XIV. Arizona, June, 1891. llected near Cafion Diablo, ; : Tae eae srahing 201 Ibs., completely perforated in three places. it a ed asain? tS iti eames » a Ee Speen oti gn Net a re treatm Plate XV. Aime Ourr oel., Vol AEM, 1eOil: 20 NTS URES AE SS ae I Oe a ar re Polished Surface of Meteoric Iron from Cafion Diablo, Arizona. showing Widmanstattian figures. A small black diamond is shown protruding near one side of the central black cavity, at D; a circle of scratches made by small loosened diamonds can be seen near this spot. 4 Sar ae sd dete A Ss a en remem a ¢ ~~ 2 tee Am. Jour. Sci., Vol. XLII, 1890. Plate XVI. ee gs = 3 fon | v4 , ia | can y a) Lie a) Serer i + « : 3 \ r ue cad ms " j i if - - ; * 2 "} “ y x : | ; I ; vm > ie ¥ . ies {is P a FE norte vi > c Se a = se : + a ‘ CP ae tal » ab Nd oy" pe’ '§ LIV s==4) which are given below. oS 0 ther publication, ‘Ifa cultured stranger from another world were to | himself in this one, and were to make a study of wrliterary advantages. he would be impressed espe- jally, we are confident, by the abundance, variety and i Pea cn quality of the contents of LITTELL’S G AGE.” — The Congregationalist, Boston. ‘Iti is nearly half a century since the first volume of s sterling publication came from the press, and to- ay it stands the most -perfect publication of its kind he world. . Thereis but one LIVING AGE, though ny have essayed imitations. While their intent as no doubt been worthy. they have lacked that rare iscriminating judgment, that fineness of acumen, and vat keen. appreciation of what constitutes true excel- ence, which make LITTELL’s LIVING AGE the incom- arable publication thatitis. . We know of no other publication that is so thorough an educator, for it _ touches all live subjects and gives the best thought of leading minds concerning them.” — Christian at ‘Work, o New York. a NO eclectic journal has ever deserved so well of epublic. . It contains nearly all the good literature the time.” — The Churchman, New York. Papi improves with age. It is a treasure-house of e best periodical literature in the language, and subseribers are easily enabled to keep themselves ac- quainted with the work of the most eminent writers the time.” — Standard of the Cross, Philadelphia. a “Tt maintains its leading position in spite of the multitude of aspirants for public favor. . He who ae for a few years to it gathers a choice ees. even though he may have no other books.” — Here York Observer. ses\8e ‘Indeed it may well be doubted whether there exists any more essential aid to cultivation of the mind among English-speaking people; and its importance pteone with the ever growing rush and hurry of - modern times. . Certain it is that no other magazine n take its place i in enabling the busy reader to keep with current literature,” — Episcopal Recorder, tladelphia. ae 7) r LIVING AGE, Ale eee THe Livine AGE ba so rruusbes its jubilee, itis interesting to recall Ww the prophecy made concerning it by MR. JUSTICE STORY upon read- ing the prospectus in April, 1844. He then said, ‘‘I entirely approve. the plan, If it. can obtain the public patronage long enough, it will contribute in an eminent degree to give a healthy tone, not only to | our literature, but to public opinion. “Tt will enable us to possess in a | moderate compass a select library of the best productions of the age.”’ | That THe Livine AGs has fully justified this forecast is proved by | the constant praises which, during all the years of its publication, | have been bestowed upon it by the. press; some of the more recent of EEKLY MAGAZINE of sixty-four pages, THE Livine AGE gives more than "Three and a Quarter Thousand = Bee Solanin octavo pages of reading-matter yearly, forming four large volumes, Tt esents in an inexpensive form, considering its great amount of matter, with fresh- less, OWing to its weekly issue, and with a satisfactory completeness attempted by no best Essays, Reviews, Criticisms, Tales, Sketches. of Travel and Discovery, Poetry, Scientific, Biographical, | Historical, and Political Information, from the entire body of Foreign Periodical Literature. Tt is therefore invaluable to every American reader, as the only satisfactorily fresh COMPLETE compilation of an indispensable current literature, —indispensable _ use it embraces the productions of THE ABLEST LIVING WRITERS in all Opinions. ‘*If has, in the half century of its existence, fur- nished its “host of readers with literature the best of the day, such as cannot fail to educate and stimulate the intellectual faculties, and create tastes and desires for loftier attainments.” — Presb’n Banner, Pittsburgh. ‘Tt is incomparably the finest literary pr oduction of modern times.”—Herald and Presbyter, Cincinnati. “For the man who tries to be truly conversant with the very best literature of this and other countries, it is indispensable.”— Central Baptist, St. Louis. “The subscription price is Jow for the abundance of excellent reading given.” — New-York Evangelist. “Tt would be cheap at almost any price.” — Califor- nia Christian Advocate, San Francisco. “Ttsaves much labor to a busy man who only wants to read the best.” — The Advance, Chicago. “Tt retains the characteristics of breadth, catho- licity and good taste which have always marked its editing. The fields of fiction, biography, travel, _ science, poetry, criticism, and social and religious .discussion all come within its domain and all are well represented.” — Boston Journal. _ “Itmay be truthfully and cordially said that.it never offers a dry or valueless page.” — New- York Tribune. “To read it is itself an education in the course of modern thought and literature.”—Bujfalo Commercial Advertiser. “Coming weekly, it has a great advantage over the monthly magazines and reviews.” — San- Francisco Chronicle. “Ttis one of the invaluables to those whose time is limited.” — Houston (Tex.) Post. ‘““No one who pretends to keep au courant with what is doing in science and literature can afford to dispense w ith it.’ *— Hartford Courant. “Tn giving a comprehensive view of the best current literature, the product of the best writers of the day, it stands unriyalled.”— Canada Presbyterian, Toronio. _ PUBLISHED Wire at $8.00 a year, free of postage. : Ie— TO NEW SUBSCRIBERS for the year 1892, remitting before Jan. 1, the weekly numbers of 1891 issued after the receipt of their subser iptions, will besentg gratis. ‘LUB PRICES FOR THE BEST HOME AND FOREIGN LITERATURE. [Possessed of LITTELL’S LIVING AGE, and of one or other of our vivacious American monthlies, a for will find himself in command of the whole situation.” — Philadelphia Evening Bulletin.] ‘ _ For $10.50, Tux Livine AGr and any one of the four-dollar monthly magazines } « or Harper’s Weekly or Bazar) will be sent for a year, postpaid; or, for $9.50, THE ae Lryine AGeE and Scribner’s Magazine, or Lippincott’s Magazine, or the St. Nicholas. Rates for clubbing Tur Livine AGE with more than one other periodical will be sent on application. Sample copies of THE Livine AGE 15 cents each. a: ie.) eee ADDRESS. “> -- LITTELL & CO., 31 Bedford St., Boston. CONTENTS. Page Art. XLVII.—Percival’s map of. the Jura-Trias trap-belts of Central Connecticut, with observations on the up- turning, or mountain-making disturbance, of the Forma- tion; by J. D. Dana. With a map, Plate XVI_____- 439 XLVIII. “The Detection and Determination of Potasstan Spectroscopically; by F. A. Goocn and T. §. Hart___ 448 XLIX.—The Ultra-Violet Spectrum of the Solar Promi- nences;, by G.° H. HALE y. 2 52.2025 6 oie ee L.—Phonics of Auditoriums; by E. Currer _._...-....._. 468 LI.—The Secular Variation of Latitudes; by G. C. Comstock 470 LII,—Capture of Comets by Planets, especially their Capture by Jupiter; by H. A.“ Newton -2---_.- 2. 482 LIII.—Distribution of Titanic Oxide upon the surface of the Barth; by F. P: DuNNINGTON -u.- --3.¢2 22 491 iy Wetes on a Missouri Barite; by C. LuepmKine and HL. Av WHEELER -. LoSl. pee ue oo ee LYV.—The Contraction of Molten Rock; by C. Barus -..-- 498 LVL—Notes on Michigan Minerals; by A. C. Lanz, H. F. Ke.ier and F. F. Smarpiess ._.-.---- ld er 499 SCIENTIFIC INTELLIGENCE. Chemistry and Physics—Two.new Modifications of Sulphur, ENG&L: Chemistry of the Carbon compounds or Organic Chemistry, von RicuTER, 509.—System of Inorganic Chemistry, Wm. Ramsay: An Introduction to the Mathematical The- ory of Electricity and Magnetism, W. T. H. Emragz, 510.—Chapters on Hlec- tricity, 8. SHELDON: Apparent change in electrochemical equivalent of copper. VANNI: Electrolytic generation of Gas in a closed space, CHABRY: Damping of electrical oscillations, BSERKNES: Velocity of Electrical waves in solid insulators, AVONS and RUBENS, 511. Geology—British Earthquakes of 1889, C. Davison, 512,—Formation of Graphite in Contact-metamorphism, Beck and Luzi, 514.—Geological Survey of Alabama, EK. A. SmirH: Geological Survey of Missouri, Bulletin No. 5, A. WINSLOW: Geological Survey of Georgia, L. W. SPENCER, 515.—Geological facts on Grand River, Labrador, A. Cary: Index to the known Fossil Insects of the World, S. H. ScuppEr: Stones for Building and Decoration, G. P. MERRILL: Manga- nese, its uses, ores and deposits, R. A. F: PENROSE, Jr., 516, Botany—Botanic Gardens in the Equatorial Belt and in the South Seas, 517. Miscellaneous Scientific Intelligence—Analysis of the water of the Salt Lake, Alia- paakai, on Oahu, Hawaiian Islands, Prof. Lyons, 522.—National Academy of Sciences: The Metal Worker, A. O. KITTREDGE, 523. Obituary—J. FRANCIS WILLIAMS. Ly id WL AT IML yet Nib bid fol | TT TTLET TLL TT LTT gb tel Lelelalad alateiaa! | ale alin Ay a . . a Lee aaa ccc UCC otces v TY x ; ABA | Aa Tileh 2 _— is ‘A d ~ abl anager ona ry eat ‘Ras Mt be ‘ ty," As Widete gy ne 2° alls, sebhas yyy ahinnk TT TYy nell Te tol delle es, SN - Waes4¥h oe : ants My pit — ih shy ar ub Whe: 7 UNE Wiad a "ee Lite aT ed Paraeimeaea:: 7. 18 6 ; 3 At, Baa 4. Ay. pier Se & -, Vapegtt14@ ets. . a ‘eae Atht } ee ibd ye Pag qt ad Ln Ba ¥ Ys, | é Ty Wy Ag, . Qe bbb ha | PN PTT tT rey rae a r a = ) OT ane . “ > aa Maye : Osa : '] v\f re Va et 7a Ya> Wwewnn Gua Puig 8” xs Wires +) Ree a* a, aa, are, Mal as 228 NASee i “a Wes sgn peta tne RTT Det yc kaa mares nein aed Y SN atl) YT Hi alate \ mie ey “a. an, é 4 7; - : Yan ay ~~ aahhaey NYY PT amlanblal acceetns he Nb fy -a, aghaes! * A ys if, oot vault “ PERT hind ae Ae oat al} | OOD a ait ankacl gue OE CS ee | Se eo yan, the 2 ARand vey er "Aaecr ; 1. ane & a¢;7- Wor? ‘eo? hat: toe? “oor bY 1 Aag, ns Rin". .) Hy itty Ha § i rw " “9 VEN ~OBh way é 1 Ihiy AL Wap natas Mg eee EINE rye ana uA a ny f A epee egich! Sy pein, Sen wee Tenn | LY Wiel i) 4 : . f Lae ; ware) YSuNm ‘ & ‘ ] . AR a NY ening ih yer Ay e A qv | in ‘NGece e 2% 5 7 ee ' Taina yrvveabr: TTT | et i ea dsehel Paaameatn th tI oy wal x ~~ ene! a Serenity na Aye eee 44a Ay Aw and buns AU. ~ 4, act OP hey »& 9 ! ah eS Bee oer Mel TPT nt PR ANIM anh a Maa 7 aay Raa. | | pew qe. HINO INN teas” a Nia a, Ay -. v f phh bie Waa TN Vio \. : sd bY, b / v pe Ag 4 AMAL, A 4 diy: wy . pe eee tyr Mseytean Ne vreove *4A a Wa maa | ! Wt tyy ban @ errs - * »Ry Pye ah die La Ll nt « AKA ANION PeoRep, a al Nan Pere Yaa Bet Om een CIE LNA * bya’ A iy TiN aie sar v wraseuaas cate a sMbA LAAN te a iia ah SERV! pal y3 | + tilt pinned” ® »409qe!is.. yy hia 4, go a UPR, byt dy t (n° heeee- a 00 Ry pes as) wy yyy” } near ee a). ar * Diab oeryy rat dnd a be ee , oth, . Or 1 veep) PoMetee 1 teat “NAA Phe. he Nw TTT tt) DPR y See. i ea Weel! lal. TPL ELT TIT ree = y a \ a Dane Aa, Aan ' Se ae ‘day a] — ae errs > a4 Ra; J AS re Ava x 4) “ues r A i a : i aae a ie A. PER AN WY ALY | fad am, MONS ARs: oe ates aa ‘ A PP yee Laine - Wars an Ne RasiRls ae a AMARA a aaa Saar ana an Mh ST ARs oa, PPR Veal Jere ee aman arn pose fabaaty Corned a PATTY pe atte | | fra pS A i ¥% “ PT : r rey a 4 . A z ~la A tie | ie Ni ytinons Tree Tweet HiT ARE pons sate sescall HERE aN wapAtaa’ SAN Sa atte as yas An aah . Aan a. Sf Ke iy as »¥. a rie | iy a : "a . tg NAS Ww WIN Aa 17 La) \ {i ah “Athy ; _ Na, a _ &An Va TA sAsQnan waa ‘ b WARS f { i a \ Reet AAB Rang Or-Be MAMAN prada oA? i ; Rn. 02t4QRan - ARARAN itttecas UR MU au 90 mere a typ mn af 3 rvY Vaan i. aoa, ; a , \ A PPM GQ NT easy Lhe . +o , W ea Ald ri RY ee vt qeha | ; Fsruiphl Tye, “Ae” abst eal | sna Aan’ aan, Wh \ "ihe >». . oe ~ my Wea sy ea 5 Nath Aaryy h& -pit ly Rasa yy e A) ~~ m ‘ phe _~ 4a @ BAe e pid... Mhida'h Shiekh | sali Nyy apn nanann aih-aib ya Arai VN sea ll TEAL ’y egiiairey yy CGY Wical } TY yy TRL Ft manne ATA4244r: # Ly Ys vie i im - ebhted ta tin | pF ahr at te EDAD ie >» N44 SY Te ee Nes NS0> ‘ (Y ; TIM 01298 5438