IMAGE EVALUATION TEST TARGET (MT-S) v.. 1.0 l.r 11.25 |50 ^^" ■■■ Ui liii ■lutl M 11.6 O S^ <% A // w /A ^J* # JV ^^ ^\ „ ^ >^. V"^' CIHM/ICMH Microfiche Series. CIHM/ICMH Collection de microfiches. Canadian Institute for Historical Microreproductions Institut Canadian de microreproductions historiques 1980 Technical Notes / Notes techniques The Institute has attempted to obtain the best original copy available for filming. Physical features of this copy which may alter any of the images in the reproduction are checked below. EZ] D D D Coloured covers/ Couvertures de couleur Coloured maps/ Cartes g6ographiques en couleur Pages discoloured, stained or foxed/ Pages Jtcoior^es, tachet^es ou piqu6es Tight binding (may cause shadows or distortion along interior margin)/ Reliure serrd (peut causer de I'ombre ou de la distortion le long de la marge intdrieure) L'Institut a microfilm6 le meilleur exemplaire qu'il lui a 6t6 possible de se procurer. Ce:tains dtfauts susceptibles de nuire A la quality de la reproduction sont notte ci-dessous. D D D D Coloured pages/ Pages de couleur Coloured plates/ Planches «n couleur Show through/ Transparence Pages damaged/ Pages endommagdes Th po of fill Th CO or ap Th fill ini IVIi in up bo fol D Additional comments/ Commentaires suppl6mentaires Bibliographic Notes / Notes bibliographiques D D Only edition available/ Seule Edition disponible Bound with other material/ Reli6 avec d'autras documents □ Pagination incorrect/ Erreurs de pagination Pages missing/ Des pages manquent D Cover title missing/ Le titre de couverture manque D M&ps missing/ Des cartes g6ographiques manquent D Plates missing/ Des planches manquent D Additional comments/ Cc mmentaires suppKlmentaires IS la The images appearing here are the best quality possible considering the condition and legibility of the original copy and in keeping with the filming contract specifications. Les images suivantes ont M reproduites avec le plus grand soin, compte tenu de la condition at de la nett<9t« dj I'exempialre film6, et en conformity avec les conditions du contrat de filmage. The last recorded frame on each microfiche shall contain the symbol —^ (meaning CONTINUED"), or the symbol V (meaning "END"), whichever applies. Un des symboles suivants apparaTtra sur la der- nidre image de cheque microfiche, selon le cas: le symbols — ^ signifie "A SUIVRE", le symbols y signifie "FIN". The original copy was borrowed from, and filmed with, the Icind consent of the following institution: L'exemplaire filmA fut reproduit grAce it la g6n«rosit6 de l'6tablissement prAteur suivant : Library of the Public Archives of Canada La bibliothique des Archives publiqu^s du Canada ' .<.■>':. Maps or plates too large to be entirely included in one exposure are filmed beginning in the upper left hand corner, left to right and top to bottom, as many frames as required. The following diagrams illustrate the method: Les cartes ou les planches trop grandes pour Atre reproduites en un seul clich6 sont fiimies A partir de Tangle supArieure gauche, de gauche A droite et de haut en bas, en prenant le nombre d'images nAcessaire. Le diagramme suivant illustre la mAthode : ■■:'*'■;:':'■ :; ¥ '"'■ ■■'- ; /'' \V ■ ' '', ■/•■' ' "'.' ■ V- 1 2 3 ■'■.. '■ ''■■ -■ '■■^:.' '".'■- ■''■■ ' ' ' 'iv ■* - ■'"'■ ■ ^- ■ '''' 1 2 • 3 1 2 3 4 5 6 m iplpf mmm BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA Vol. 6. pp. 117-146 ^S GEOLOGICAL NOTES ON SOME OP THE COASTS AND ISLANDS OF BERING SEA AND VICINITY Br GEORGE M. DAWSON ASSISTANT DIRECTOR OF THE GEOLOGICAL BURVBy OF CANADA ROCHESTER PUBIJSHED BY THE SOCIETY February, 1894 ^«i;ujWHHli;«<'ji |(, I ii,iiiij|i|ip)ipiHMpipiippp BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA Vol. 5, pp. ii7-i4« February 2, i894 GEOLOGICAL NOTES ON SOME OF THE COASTS AND ISLANDS OF BERING SEA AND VICINITY BY GKORGe" M. DAWSON ' , ASSISTANT DIRECTOR OP THE GEOLOGICAL SURVEY OP CANADA (Read before the Society December 27, 189S) -^ , . CONTENTS P»ge Introduction 117 Aleutian Inlands 119 Akutan lHlanurpose of their determination. ' • • *Biill. U. S. GeoluKical Survey, no. 84, 1892, p. 234 et aeq. H": »'5IL,. AKUTAN ISLAND. 110 Aleutian Islands. Ahitan Maud. — The south side of this iHland was tho firat part of tlie Alputian rliain sijjhtod by us in ai)i)r()a('bing Bering Hca. It is character- ized l)y rugged and bold cliff's, broken into stacks and pinnacles at the })oints, but between retiring into coves and bays, from some of which rather wide valleys run inland. The varied and often strongly contrast- ing coloring of the weathered rocks in these cliffs, together with the abso- lute treelcssness of the land and the vivid green of the sward and herbage with which it is covered when;ver not too rocky or too elevated for any growth, were the most striking features. These, however, are almost equally found in all the islands of the Aleutian chain. Steej) and irregular hills and ridges rising from the shores culminate in the central part of the island in mountains sufficiently higl: to carry J Itic/.ijiwi Beds FletnJcs of Old f.'arui. FiQVRi 1.— Diagram il FlMUBeds irv Cliff^s. :■ (7 the Structure of the northern Part o/Akittan Iiland. much snow in these latitu.iv • the year round. The north side of this island was afterward seen under favorable (^oiiditions of weather, and the island as a whole apyjears to represent the denuded remnants of a single great volcanic center. The original focus of eruption seems to have been situated to the west of the middle of the island. Somewhat nearer the actual center of the island a little cloud of steam still issues from one of the higher points, and occasional small eruptions have been noted.* The eastern portion of the island shows j)art of the lower slope of the original great volcanic cone, the l)eds flattening out gradually to the ea.st- ward in conformity with the decreasing slope of the surface. The antiquity of the volcanic action to which the island as a whole owes its origin, is shown by the amount of the subsequent effect of denuda- tion upon it. The deep valleys have evidently been cut out by ordinary subaerial erosion during a prolonged period of waste, and many of the * Alaska and its ReBourees: DnII, pp. 467, 470. w>i^^-.'^ ...... ■,j!t \i:x 120 (i. M. T»AWS(>N^"OAST8 OK JUCKING SKA AND VICINITY. lower rid^ri'H siikI hill.s owe their ])rc'Hont t'orni.s entirely to wueh action The iteripherai parts of the island have during the same period been much reduced and cut back by the sea. The ditts along the south shore and those altout the middle of the north sbore dis])lay l>edding, wbieli in the planes of section ap}tears nearly horizontal. The individual beds are for the most part thick, and some of them show a i)rononnced basaltic structure. Unalaska Maud. — Respecting the island of Unalaska little can be added to the notes lately brouglit together by Dr Dall.* A good view was ob- tained of that i)art of the north coast between ^lakushin and Captain's bays, behind which rises Makusbin volcano, the highest mountain (5,474 feet) on the island. The upper parts of this mountain and the bigh ridges about it were deeply covered with snow in August, and small, irregular glacier masses, more or less crevassed, were observed lierc and there. . Makusbin is not a tj'pical volcanic cone, but an irregular, lumpy mountain mass witb some rather bold spurs and crests even about its upi)er parts, which seem to evidence considerable waste by denudation since tbe latest important eruptions. A cloud of white steam still, how- ever, hangs about its summit. The natural processes of waste have not acted sufficiently long upon this island to produce rounded forms or light slopes. The hills are generally shari) edged, peaked and bold, but often covered with herbage nearly to their tops. The shore-cliffs, from abreast of Makusbin to Captain's harbor, show horizontal or slightly inclined bedding, which is rather too fine and uni- form to be explained as that of sui)erpo3ed lava flows. The beds ex- posed probably consist of scoria or volcanic ash deposited under water- Dall notes the occurrence of marine Miocene dej)osits in Makusbin bfty.f The slopes and hills about Captain's harbor were all closely scanned for evidence of old beachlines, but none were seen. There appears to have been no notable upward movement of this land since the denuda- tion which produced its present form took place. The little flat about the village of Unalaska is composed of gravel covered with black soil, and is about twenty feet above high-water mark. Its occurrence might almost be ex{)lained as a beach deposit at the present storm level, but it more probably indicates a ver}'^ slight rise of the land. Atka Island. — A call was made at Nazan bay, on the east side of Atka island. This island, with others near it which were more or less imper- fectly seen, appears to be practically entirely composed of volcanic rocks. Some interesting notes on Atka are given by Dall in the publication * Op. cit., p. 242. t Op. clt., p. 24;i. In .iddition to tho voloanin rocks of vnrioH.s kinds, Dall mentions thn occur- rence of granite or syenite [gabbro ?] in the interior of this island. Op. elt., pp. 233, 242. ATKA ISLAND AND OTHKRS TO THE WESTWARD. 121 already (juotod (pagcH 248-244), from wliich it appeaix that in Karovinski bay, on tho west side of the island, jtienes of fossil wood, sometimes ailici- fied, as well as marine Miocene fossils in tuttaceous volcanic deposits, are found. The little islands in Nazan haj' and the low projecting i)<)ints ahout it have a basaltic appearancjc and rather irregular columnar structure. The rock showing on the beach, at the village, is a greenish gray. Hue grained material, very hard and in j)lace8 containing many grains of iron-|)yrites, probably clastic and apparently a diabase, though not microscoi)ically determined. It is also traversed by small drusy veins of quartz, and large loose masses of quartz were found which had evidently not travelled far. In a small brook, which has cut out a little ravine near the village, a considerable dei)th of superficial earthy material is exposed. This ex- hibits a certain amount of stratification jiarallel to the slope of the ground. It is brownish or reddish brown in color, and ai>peared to be composed of volcanic detritus, which has either been deposited in the sea when the land was at a s(micwhat lower level, or perhaps more ])robably merely wa.shed down the slopes while in an incoherent and fresh state. Material of the same kind was recognized elsewhere on this and other islands of the Aleutian chain, sometimes in rather notable tjuantity. No indications of old terrace levels were observed about Nazan bay, but around the base of the mountain which forms a projecting i)oint to the north of the entrance of the V)ay, three or more successive, indistinct terrace-like markings occur, the highest being at an elevation of about 1,000 feet above the present sea level. These markings may represent old beachlines im])ressed in soft material, but from their indefinite character this remains uncertain. Great Sitkin Island. — After jiassing Atka island a fine general view of Great Sitkin island was obtained. This evidently consists of a single large volcanic cone, which, according to the charts, is still 5,033 feet high. Its upi)er portion was heavily covered with snow. Biildir Island. — Buldir island was passed sufficiently near to enable it to be well seen. Its eastern end is most* elevated, and high cliffo there show a series of flows or beds of volcanic material, dipping rather steeply we.'arings. Tliey are separated at their nearest points by a dist^mce of 26 nautical miles. Copper island, which lies farthest to the eastward, is divided by 1!)0 miles of deep ocean from Attn, the easternmost of the Aleutian chain, while Bering island is distant some \)r> miles from the nearest part of the peninsula of Kamchatka. The high volcanic mountains of the peninsula may in clear weather be seen from Bering island, l)Ut the latter is proliably never imder any circum.stances visible from the mainland. In view of the fact ihat the Aleutian islands were, when discovered, rather thickly iidiabited, and that evidences exist on them of inhabitants long antecedent to hi,storic times for the region,* it is somewhat remark- able that the Commander islands ap|)car never to have been visited by man ))revious to their di.scovery by the Uu.'^sians in 1741. The climate of these islands is humid and cool, insuring a luxuriant growth of grasses and herbaceous jdants of various speciej wherever there is sufficient soil, but though less rigorous than that of the lands in similar latitudes on the eastern side of Bering sea, no trees or shrubs are anywhere found upon them. •Dull in ContribiitlotiB to North American Ethnology, vol. i. 'f 124 a. M. DAWSON — COASTS OF niCUIXO SKA AND VICINITY. I': Bering Maud. — leering island is about 50 miles in extreme length, with a width of nearly 2U miles at its northern and wider end. From this it narrows gradually, hut irregularly, to caj)e Maniti, its southeastern ex- tremity. The northern half of the island is low, with a rolling or nearly fiat surface, much of which is described as consisting of "tundra" land. It includes one large lake, which discharges on tlie north shore. The southern half is higher, and appears, as seen from the sea, to consist of a mass of rounded hills, varying in height from several hundred to perhaps 1,000 feet. There arc no harbors alxmt the island, l)ut a fair anchorage, with off-shore winds, may be found in a bay at Niivolski, on the west coast of the island, about ten miles from its northern end. The only permanent settlement, with the head(iuarters of the Russiiin government of the islands, is situated at this place. The shores of the higher southern i)ortion of the island are ^'enerally bordered by cliffs or steep scarped banks, with narrow and V-8li''^P6d valleys breaking through them to the sea. On the east side of cape Maniti, and for some miles northward, regu- larly stratified rocks in rather thin l)eds of pale brownish colors were observed, dipping regularly northward at an angle of about 15°. Farther to the northwestward, along the same eastern coast, at cape Tolstoi (thir- teen miles from cape Maniti) paler fawn-colored or cream-colored beds were seen, dipping away from the shore at low angles. They are crum- bling and incoherent in character, and produce long slopes of debris in some places between the bluffs and the sea. Similar rocks apparently continue from cape Tolstoi to Stareya bay, at a further distance of nine- teen miles, but the cliffs become lower and the scarped banks are less steep. A landing Avas effected at Stareya bay, when it was found that the scarped slopes, which often resemble sand from a distance and are so described in sailing directions, are in reality composed of angular and rubbly fragments of whitish, yellowish and gra}' argillites or shales, with crumbling sandstones and argillaceous, fine grained gray limestones. All these rocks are well bedded, and on some surfaces small carbona- ceous plant fragment^ were observed, though none of these were deter- minable. The material of the Ijeach is composed almost entirely of the debris of similar rocks, and it is prol)ablc that the whole northeastern coast of the island, at least this ftxr, consists of moderately indurated sediments of Tertiary age, regularly bedded and present in considerable or great thickness. The browner beds of tlie vicinity of cape Maniti may, however, be tuifaceous volcanic material. While it is not improb- able that basaltic or other volcanic rocks may also occur, as some such were found upon the shore, they were not actually seen in place. No BERING ISLAND. 125 , with 11 this rn ex- iiearly ' land. , The ist of a erhaps lorage, le west e only rnment merally ■shaped d, regu- )rs were Farther ,oi (thir- red beds re cruni- lebria in pavently of nine- are less that the are so jlar and es, with iiestones. carbona- re deter- y of the th eastern ndurated siderable Maniti iniprob- )iiie such aoe. No >c crystalline or other evidently foreign rocks were found upon the beaches. The soil in the valleys and on the lower slopes of the hills is a reddish, fine grained material, doubtless formed by the disintegration of the rocks above described. No satisfactory general views of the coastline of the northern and lower jiart of Bering island were obtained on account of foggy weather. A landing was, however, made on the north shore at capo Yushin, where tbe " north rookery " is situated. Tiie shore is here rocky, and Avide, low reefs run out from it, entirely comi)osed of volcanic rocks. One of these is a dark brown melaphyre, containing plagiocUise, augite and olivine crystals, with some magnetite, embedded in a groundmass of the same constituents. There is also a fragmental rock of somewhat peculiar apjieai'ance, which seems not to ])e a true agglomerate, but an eruptive material charged with fragments of dissimilar rocks. The basis is some- what amygdaloidal, and may ver\' ])robably have the same composition as tbe rock first noted. \\'ell formed jjyroxene crystals are abundant in some parts of the mass. The rocks are much shattered, and it was not easy to determine the precise relations of the two varieties here associated. No trace of sedimentary rocks like those of Htareya was seen. At Nikolski, on the west side of the island, the point south of tbe little ])iiy is composed of hard, fine grained, gray, augite-j)or[)hyrite, composed of ])lagioclase, augite, and a light brown biotite, considerably altered to chlorite, apatite and magnetite. It is homogeneous in texture and ap- ))arently massive. Here and there tliis rock is curiously spotted witii Mesh-colored chalcedony, which occurs in it in small kernels not dis- tinctly amygdaloidal. The relation which this rock may bear to the stratified sediments of other jiarts of the island remains uncertain, as no sedimentary rocks were seen here. Basaltic rocks are, however, ])robably abundant in the northern part of the islaml, for fragments of such rocks are common on the beaches. The shores about Nikolski in some |)la(>es show a well marked low terrace, at twonty to thirty feet aV)(>ve high-water mark, which evidently indicates an elevation of about that amount, as there is a second still lower Hat just above the actual beach, which may be accounted for by tbe accumulation of storm-wash under the present conditions or very nearly so. This lower fiat is no doubt that in which the Rhytlna bones were found to be most abundant by Nordenskjold. With these excep- tions no terracing was observed in Bering island. According to Mr N. (irebnitsky, the governor of the Commander islands, some fossil shells and plants have been found in the rocks of Bering island, which, f)n transmission by him to Saint Petersburg, were referred to the Miocene Tertiary. Lignite is also found on the island, but in inconsiderable XVII— BuLi,. Gkol. Sof. Am., Vol. .'•., 189:1. m'i!Skl',iAiM^f^^'iU','':M 126 O. M. DAWSOX — COASTS OF BERING SEA AND VICINITY. ;!; quantity. Nordenskjcild gives some general notes on Bering island, and quotes a statement made to liim by Mr Grebnitsky of a character .similar to the ahove.^ To the south of Xikolski the western sliore of Bering island was not seen. Copper Island. — ("opper or Medni island is a])out thirty miles in length, with a greatest width of nltout five miles, to the south of the middle of the island. It forms a single mountainous ridge, of which the highest })arts probably attain an elevation of ;^,00() feet, and is much bolder in t>utline than Bering island. Its surface is exceedingly irregular, and comprises very little Hat land of any kind, whih; its shore is often bor- dered by high and rugged seacliffs, particularly along tlie southeastern side. The shoreline of this side is sinuous, while that of the northeastern side is deeply indented by several considerable l)ays, butaft'ords no good harbors for large vessels. There are three small settlements on the north- east coast — (ilinka, Karabelny and I'reoJjajenski — the last named being the most northern and the only one continuously oc('Ui)ied during the winter months. The island ajjpears to be almost entirely comj)osed of volcanic rocks of some antiiiuity. No volcanic cones or craters were ol)serveil, but, on the contrary, the (existing relief is evidently the result of oridnary denudation. The slopes are generally steep and are sometimes sur- mounted by rocky crests, but are usually more or less coni])letelj' grass- covered from base to sumnut. The hills in their form and general appear- ance much resemble those of the higher parts of Saint Matthew island. Copper island was crossed near its southeastern end from Glinka (Pestchanni of sonu; charts). At (Hinka the rocks seem to have a general southeasterly dip, and both here and on the other side of the island are for the most part gray and l)rowni.sh ])or})hyrites (augite-porphyrite?), with some massive beds of coarse agglomerate. Where the scarped slopes of the southwest side of the island were first reached, a bed ten to twenty feet in thickness of a soft i)ale tuflaceous rock was found. This consists of small fragments and fine amorphous material, all apparently volcanic in origin, and contains embedded pieces of tree-trunks, sometimes more or less silicified, but more usually in the form of lignit ). The tuff' was observed in some instances to fill what had originally been hollows in the rotten wood. Below this is a bed ten feet or more in thickness of coarse conglomerate with well rounded stones, which also contains lig- nitized fragments of trees. The pebbles from the conglomerate consist of volcanic rocks similar to those common in the vicinity, and the whole of the water-bedded intercalation apjjcared to l)e referable to the temporary • Voyage of the Vesft, vol. ii, pp. 280, 201. COPPEU ISLAND, 127 occurrui.fe of l)0!U'h conditions during a stage of the period of volcanic activity tc which the rocks of the island generally are due. The dij) at this place is southeastward at an average angle of 15°. In the valley behind Glinka village pretty definite evidences of terrac- ing were observed at several different levels. The horizontal lines are, however, liut faintly impressed. The highest of these was estimated to be 600 or 700 feet above the present aealevel. At Preobajenski, near the northwestern end of the island, the rocks seen were chiefly greenish and purplish i)ori)hyritic materials, of which no specimens were brought back. The rocks which form a high clitT to the north of the village at this place were not examined. They are well stratifieserved particularly about cape Japounski or Tshipunski, where it gives form to the end of the promontory, and spreads along the bases of the higher hills sometimes with a width of a mile or more. At cape Japounski (estimating from the heights given on the charts) this plane is, in its higher parts, 7(X) to 800 feet above the present sealevel, but declines gradually to its seaward edge, where it is about 600 feet in height. Traces of the same or a similar plane, though at a somewhat lower level, were again seen in the immediate neighbor- hood of Avacha ]>ay. At cape Japounski this tiat bordering land or narrow plateau has itself been since cut through by narrow V-^haped valleys which run from the inland hilly tract to the sea.* The excavation of the later valleys seems to have occurred while the land stood some fifty or one hundred feet below its present level, for the valleys are not cut down to the sea, but terminate seaward at such heights above the waterline. The coast cliff may l)e re[)resented diagramatically thus : ("', ' FiouHK .i.— Diagram illustrating the Profile of the Coast Cliffs at Cape Japounski, That the plateau of cape Jajiounski is not one of deposition, but subse- quently impressed, is shown by the fact that the underlying rocks are seen in the seacliffs, particularly near the extremity of the cape, to be well stratified and to be inclined at various angles, which are sometimes rather high and are entirely independent of the level contour of the * The general appenrance of cape Japouuskl is very well illustrated in view no. 3 on chart no, 84,, U. S. Hydrographio offloo. KAMCHATKA. 129 plane of denudation. Tliese rocks may very proljably be similar to thoae of Petrapavlovsk, l)ut they were not examined. Dia<:;ramatically the general structure of this part of the Kamcliatka peninsula may be represented as at i o^i ;, but it must be understood that the illustration is not an actual drawin the Hiore. jiirt is exist- at the I Hhal- of the le 1()0- ! it are ud. the ice im- ove the ve been 3 island ussians, as been cient to he same lutchin- nei- time which is -Brown »Ho islands The tvwk 4and, and Ak there from the me of pueh icnticates he also mains on ilmid led liim •d ft sufficient I'liahiska island. It will bo ol)servod that Mr Stanlo^'-Hrown does r')t question the fiiidinjf of the mammoth remains on the I'ribilof islands, and I do not attach the same significance to the absence of extraneous earthy matter in the soil to which he refers. This, in fact, appears to afford fnrther reason to believe that the bones could not have been carried thither in any adventitious manner, and to render it as nearly as ])os- aible certain that the animals to which they belonged must have found their way to the islands at a time when they were, connected with the American continent by means of a wide plain, such as Mr Stanley-Brown himself exjdains in one of the i)aragrai)hs of his paj)er,t would be made if an elevation of 200 feet should now take place in Bering sea. The absence of old sea-margins on the I*ril)ilof islands may be accepted as showing that since the time of their original elevation above the sea they have not been again submerged, but there is no evidence whatever to show that they may not have stood at higher levels. My observations agree with those of Mr Stanley-Brown in regard to the abs'cnce of erratics above the present sea-margin, but it may be added that not infrecpient pebbles and small bowlders of granitic rocks occur upon the actual beaches in association with local debris. These have in all probability been brought hither either by the Hoe-ice, which fills this part of Bering sea in winter, or attached to the roots of drift tree- trunks, which are often washed ashore. NuNiVAK Island. The form of Nunivak island is very imperfectly represented on the charts. It was apj)roached by us on the 7th of August on its south- western side, where a landing was effected. On the following daj' the western and northern shores were coasted at a distance as small as ap- peared to be compatible with safety, and the next night was spent at anchor in Eteolin harbor, at the northeastern extremity of the island. The island is throughout grass-covered, but entirely devoid of trees, though a few stunted shrubs are found in some of the valleys. Its coasts are usually rather low, but vertical cliffs of 100 to 150 feet in height appear at the points and projecting headlands, while shelving rocky shores, with occasional sandbeaches and sanddunes, characterize the various open bays. Tlie cliffs show several superposed and horizon- tal layers of basaltic rock, and in the low hills of the interior of the island similar but overlying massive flows of the same kind may be traced. These hills are all more or less plateau-like in form, and might readily be mistaken in some places for old marine terraces. The highest ♦Op. cit., p. 49«. XVIII— Bull. Geoi.. Soc. Am., Vol. r>, 1893. / 134 a. M. DAWSON — COASTS OV BERINO SKA AND VICINITV. ':' I ii! parts ol'tlu' isliind were OMtiniatcd at about 50() tW't. At F'^toolin liarl)()r tho rock is aj^ray oliviiio-dialiase, vory jjohjus and ocillular, and neparatcd into layers which siniilate liorizontal heddinjr, hut which are due to How Htrncture. I'art of the eant coast of the island was suhse(|nently seen from a dis- tance, and its ap[)earance is so similar to tliat of tiie otlu!r coasts tl)at there can be little doubt that the island is entirely composed of nearly horizontal basaltic Hows. Tiie basalts examincid are all fresh lookinjf and unaltered, like those of Saint Paul island. 'Die much altered sand- stones reiH)rted by Dall at Eteolin harbor were not found, nor was I able to identify any volcanic cones upon this island.* No erratics or traces of glaciation were observcnl on tho parts of Nuni- vak island visited. Cape Vancouver. Cape Vancouver, twenty-five miles distant from the eastern coast of Nunivak island, is a projecting point of Nelson island, Avhich is to all intents a portion of the adjacent Alaskan mainland. It is a bold and high promontory, which, thougli scarcely to be characterized as moun- tainous, rises to a height of probably 1,000 or IpOO feet. It evidentl}' forms one of several or many projections of higher land along this part of the .\laskan coast, which are connected by broad, low, level tracts. The north shore of the cape, which alone was examined, forms scarped blutfs or clitl's, rising from the edge of the sea, and presenting fine ex- ])()sures of sandstones and sandy shales, well bedded and dipping south- ward, at low and undulating angles. At the extremity of tlii^ capejthe.se beds ai)i)eared to be horizontal, and on the south side, though imper- fectly seen from a distance, they seem to lie at higher and more irregular inclinations. The sandstones, where examined, are grav, bluish and brownish in (H>lor, rather soft, and sometimes nodular. They contain a few very thin and dirty seams of coal or lignite, of which the thickest seen was only a few inches. There are also in the sandstones numerous carbonace<)ns fragments and occasional fossil leaves, of which a couple were collected. These have been submitted to Sir J. \\'illiam Dawson, who sup])lies the following note upon them ; "No. I. Jiiglnns (irinninatti, K. Briiun, Heor, Flora Fo.^Hilis Alaskana, ISOO, page 38. Ih., Flora Fns.silis Arctica, vol. I. Ih., Contributions to Fossil Flora of N. (ireenland. Trans. Koyal Society, 18(59. " This species is stated by Ileer to occur in sandstone at En^lisb bay, Alaska. It is also found at Atanekerdluk in Greenland, and is said to occur in the European »Cf. Dnll, op. cit., p. 245. CAPE VANCOUVKU. 135 larlit)!" !irut(Ml () How 1 a tlis- ,rt that nearly ookinji; I Hand- ! I able [■ Nuni- coast of is to all )ol(l and s nioun- vidently tluH part L'l tracts. ! scarped r fine ex- 1!^ south- vpclthese 1 iniper- in-o>fular wnish in very thin was only )onacc'()US collected. )l)lioH the 1H()0, page 'loi-a of N. Alaska. It r,uroi)oau MicK'oiio at Oeninnfen and Hohc Blimen. Very Hiinilar HpccicH, if not mere varietal foriUH, an- cipditcd by I-cwiiii'iiMix and Ward to tlie Laramie and Tertiary of western Anu'ricu. Tlie plaiitn found witii tliiw Hi)ei'ieH at Kngiiwli Imy, Alawiva, and at Alanekerdluk, Ureenland, are ohwely allied to those of the upi)er Laramie of Canada, and I have been inclined to refer tlieni to tins ago rather than to the Miocene. " No. '2. Fra>;ment of a leaf of considerable size, but too inii)erfect for determina- tion. It may possibly have belonged to a species of (^lurcun or of a large CoryliiH, like ('. Mriinaini, but this is (juite uncertain." Accordinjj to the claHHification adopted l)y J)r Dall in his recent work, the beds at ea])e Vancouver would appear to fall under the Kenai {;ruui> of the Miocene, though the locality is a new one.* Upon the heach at cape Vancouver fragments of vesicular hasalt are al)undant, and the distant outline of the cape led me to suppose that the stratified rocks are capped by basaltic Hows in the higher hills a short distance inland from the extremity of the cape. A fairly distinct though rather narrow terrace of earthy materials was observed along the north shore of the cape at a height of 8U to 100 feet above the sea. Saint Matthkw, Hall and Pinnacle Islands. Saint Alatthew island, with Hall and Pinnacle islands near it, are situated in the center of the northern part of Bering sea. They are so remote from any other land that they ajjpear never tt) have been reached by the Eskimo, though polar bears are brought to them on the floe-ice of winter and remain during the summer. Saint Matthew island itself is long and narrow, extending in a northwest by southeast direction for about thirty miles. Hall island, some five miles in greatest length, lies near the northwest end of Saint Matthew, and Pinnacle island is situated at a distance of six or seven miles to the south of the main islands. The islands are very imperfectly delineated on the existing charts. Saint Matthew island may bo described as consisting of the unsub- merged portion of a range of bold rounded hills, some parts of which probably reach an elevation of about 1,500 feet. It is in reality formed of three isolated groups of hills of unecpial size which may originally have been separated by narrow straits, but are now united by tracts of low gravelly land washed \x\) by the action of the sea. These low lands in- clude several lagoons, into \vliieh streams fall and from which the water percolates through the gravel to the sea. Hall island is in every way similar to Saint Matthew, but happens to be divided from it by a still existing strait. * Op. cU., p. 234. 'f fru-t sed, and resend)le rocks met with in British Columbia, where the centers of eruption of Miocene date have been cut through or exposed by denudation. Tn following the north coast of Saint Matthew island from cape Upright to its deepest indentation, which forms an open bay, where we anchored, a stretch of low land with gravel beach is first i)assed. Cliffs then border th-^ S.P' a'Td are composed of rather massive rocks of dark color, resem- bling those above described. In rounding the most prominent point between cape Upright and the bay, however, a thick stratum of a grayish yellow color is ol)served in the elif!'. This rests with perfect regularity on the darker rocks below, but its upper surface appears to)have been plowed up by the passage over it of tho overlying material in a molten state. The general dip of the beds is southward at an angle of about 15°. The light colored material is probably tufi'or fine volcanic agglomerate. From the anchorage westward, the rocks of the north shore of Saint Matthew island were seen only from the sea. They appeared to be i-* '4 II .1 SAINT MATTHKW, HALL AND FINNA<"LK ISLANDS. 137 been )ti<)n, »f tho Hecni mula- ed l)y found ,n the erthe- 8 were era or uitions island, m and )ud the steeidy ire very lurplish clastic, re aaso- atcrial, iitains sc rocks iritish )een cut pright ichored, l)order rt'scni- iit point grayish (gularity ive been molten )OUt 15°. onieratc. of Sahit id to be Hinular to tlume last dcHcribt'd, and arc very prol»ably of alxmt the Haine horizon throughout. The pale colored stratuni referred to reappears at several places, and always witii a low southerly dip away from the sea. The soutii side of Saint Mattlunv island was clearly seen from the sea and closely examined through tiie telesco|io westward to al)out abreast of Pinnacle island. Its general featunss and the appearance of its rocks are in every way similar to those of the north side. The general structure of Hall island, in whi(!h the same rocks are con- tinued, is illustrated ity the subjoined diagram, skct(^hed along it« east coast. The principal dij) is here to the northward at low angles, and the rocks c<»nsist of a series of " porphyrites," with int(!rcalat,ed tulVaceous and agglonieritic beds. The rocks shown at a, d and / in the diagram, con- sist preponderantly of grayish purple, purple and gray porphyrites, nearly massive, but sometimes with a rude, irregular, columnar structure, particularly toward the north end of the ishmd, where the ro(!k is dis- tinctly an augite-porj)hyrite. Tliey have liecn considerably altered and FiuuHE o. — Diagramatie Section alonij the east Coast of Hall Island. decomposed, freciucntly largely silicified by subsequent solfataric ac ion, while they are often markedly rusty in irregular bands. C and c are pale gray in color, and consist of tutt", volcanic ash or fine agglomerate, which is evidently water-bedded, and in some places rather finely strati- fied. Tliis is i)articularly the case in respect to the bed c, which is from 2(X) to 300 feet tliick, and is underlain by a similar thickness (b) of coarse brownish and blackish agglomerate, holding some fragments from one to three yards in diameter. The section evidently represents the results of volcanic action which has lieen in part or altogetln.'r sul»mariue. The rocks are everywhere much fractured and jointed, giving rise under the action of the sea t^^:^^c^-*...^.- '•'^:\:v •: * '^^^^^^^^^^^^^^^^^^^^^^^^H — ■ *- — — •t^ — "t^ ■' V-W*. ' ,. „ -t^^V ^ - _J Vui\j\\v.r\— Shattered granitic Rocks, Cape Chihukak, Saint Lawrence Ixlnnii. off to seaward in clitTs or steep rocky slopes, with here and there pro- jecting pinnacles of fissured rock rising from them. A landing was made on the east side of cai)e Chihukak, where the plateau was found to consist of gray biotitc-granite of uniform color and texture and moderately coarse grain. The surface of the plateau is everywhere covered with irregular, angular, broken blocks of granite, much like those often found on high mountain summits, and doulitless the result of severe climatic conditions acting upon the naturally jointed rocks. No erratics were found upon the plateau nor any traces of glacial striation or smoothing, though if such at any time existed they may r 140 O. M. DAWSON' — OOAHTS OF BERING SKA AND VICTNITY. i m-' have been lost in consequence of the breaking up of the original rocky surface. The impression conveyed was, however, that this condition of the surface was of ancient date and had been i)reserved because of the exenii)tion of the region from the eftect of glaciating agents. Some stones were found on the i)rosent beach which did nt)t appear to belong to the actual vicinity, but if transported from any distance the aliundance of floe-ice known to occur in these seas in winter is quite sufficient to ac- count for this. Nothing whatever was found to favor the theory of an " oversweeping glacier," the sui)posed action of which has been particu- larly illustrated by Mr John Muir from the contours of hills and cliffs on this island.* The level contour of the plateau suggests that it may represent an ancient plane of marine denudation or peneplain. On tlie east side of the cape a fairly distinct terrace occurs at a height of about fifty feet above the sea. This has been out back in the granitic rocks as a narrow step, which is now encumbered with broken blocks from the old sea- clift' above. Pla:f«cni Terrace corered vntlifall^ii. rt>f/cAS. .-..^fe^ Sea J^evfl. FiuuBK l.~Diagramatic Section of the east Side of Cape Chibukak, Saint Lawrr.nce Island. VieMS of the island as a whole, from cape Chibukuk, and also from the sea to the westward with a remarkably clear atmos])here, failed to disclose any distinct volcanic cones or craters, but as both C'aptain Hooper and Rfr Muir speak very positively of the existence of such cones on the island, it is probable that they are to be found in its cen- tral or eastern parts.f The surface of the island, so far as seen, consists wholly of barren moorland, with grass and moss and often rock. There are no trees, and large masses of snow were found in some places along the bases of th,e clifi,s and down to the level of the sea. Plover Bay. From Saint Lawrence island we crossed to the Siberian coast at Plover bay, 80 named because H. M. S. Plover, Captain Moore, wintered there • Report of the cruise of the Corwin, 1881, pp. 137-140. t Ibid., pp. 3a, 140. PLOVKR BAY. 141 'lover there in 1848-'49, in connection with the F'ranlclin seiirch. It is situated in latitude 64° 'HY, and indents the southern ])art of the promontory sepa- rating Anadir hay from Bering strait. The weather was such as to give us while approaching it a good view of a long stretch of this part of the Siherian coast. The outlines of this coast are everywhere hold and mountainous, though none of the highest points in sight probahly ex- ceed 4,000 feet in elevation. It is entirely hare and treeless, brownish or gray, showing only here and there in the valleys the green color of heritage. Fiord-like inlets, and narrow straits of the same nature, characterize this part of the coast, but they are on a small scale as compared with those of British (,'olumbia and southeastern Alaska. Soundings given on the charts show that the water in these inlets and channels is deeper than that about their mouths, but the greatest depth actually recorded appears to be about 50 fathoms.* Plover bay is one of these small Hords, surrounded by steep, rocky mountains, notably covered everywhere on their slopes with talus ma- terial, consisting of broken angular rock, through which s{)ires and crags of solid rock often project, especially on the sides facing the sea. Gen- erally speaking, the mountains show ordinary denudation forms, with wide buttress-like projections and intervening stee|) valleys and ravines ; the shapes assumed resembling those commonly met with where the rocks are so much shattered and jointed as to crumble away under the weather with almost eipial facility in any direction. The ranges end along the coast in capes terminated by seacliffs. On the whole, the most pecidiar feature is the great abundance of loose angular material. It is doubtful to what extent this may be directly referred to rapid dis- integration due to the subarctic climate of the climate, or in how far it may be accepted as evidence of prolonged weathering uninterrupted ]\y glaciating agents. From cape Tchalpin (Indian point) to cape Nismenni, and thence as far as cape Tchukotski, the rocks as seen from the sea are generally gray in color and are in all i)robability granitic. Between the range ending seaward at cape Tchukotski and the valley containing lake IMooreof the chart is a smaller range composed near the sea of similar gray rocks, but about two miles inland assuming brownish and reddish (>olors. Brownish and reddish weathering rocks also compose most of the next rangt>, which separates lake Moore from Plover bay and includes mount Slavianka. From what was afterward seen in Plover l)ay, this differ- ence of coloration mav not indicate anv essential change incomiiosition. * Dull, howevor, siii'iiliH of ii (li'ptli of over KKi fathoms liiivinjt lipeii o))taInecl in thf oenter of Plover bny. AInsUa luul Its Rosoiirprs, pp. 4r).'>, nia. XIX— Biii.1., Okoi.. Soc. .\ji., Vol.. r., 18!);l. 142 G. M. DAWSON — COASTS OF BERING SEA AND VICIMTV. Jiuld Head, which forms the seaward end of the ranj^e last referred to, shows in its cliffs several well marked intrusive dikes, weathering yel- lowish and blackish. The whole west side of Plover bay, as well as the outer coast beyond it for several miles, consists of gray rocks which are likewise probably granitic. The only locality in which the rocks were actually examined in this region, in the short time available, was the east side of Plover bay between port Providence and Ennna harbor. They are here, in the main, rather coarse grained, gray biotite-granite, much like that of the west end of Saint Lawrence island, l)Ut in places passing into a horn])lendic granite. There is also, however, a considerable proi)ortion of gray and reddish porphyritic rock, resembling a mica-syenite or minette, which is prob- ably later in date than the granite and intrusive iiv it. No strictly volcanic rocks of any kind were seen in this vicinity, nor were any strati- fied rocks observed. The general description of this part of the Siberian coast above given will show that superficial earthy deposits are not abundant, but there are in the vicinity of Plover bay some deposits of this kind Avhich attracted special attention. The point on the outer coast immediately east of the valley of lake Moore, terminates in an apron-like flat of land which breaks off seaward in a low cliff, apparently formed of hard clay, Aveathering to a fawn- color and thickly studded with large bowlders which lie more or less definitely in regular lines, giving a stratified ajipearance to the whole. Material of the same kind is preserved in the angle of the next bay nearest to Bold Head. It Avas also seen from a distance to form the coast along the bottom of the bay into which Reindeer river floAvs, on the ojjposite side of the entrance of Plover bay, and again occurs in tAvo places on Ennna harbor on the Avest side, a little Avithin the entrance to the harbor and at its southern bay. The last named locality Avas the only one actually examined on the ground. The deposit is here bluish gray Avhere freshly exposed, and is a rather hard clay Avith a considerable ]iroportion of coarse sand and gravel, containing many boAvlders of somewhat varied lithologic character. These are subatigular in form, but none Avere found which actually showed glacial scratching, neither Avere any shells found in the mass. The deposit, nevertheless, undoubtedly represents a species of boAvlder-clay. It does not anywhere take the form of definite terraces, but as it is noAvhere seen, or at least not in any considerable mass, at a greater height than about 200 feet above the sea, Avhile it is occasionally rather wide spread l)elow this level, it may be assumed as a whole to GENERAL REMARKS. 143 represent an ajiproximate terrace-level, havhig relation to a former de- preasion of the land of about the amount stated. In Plover bay, on August 16, large masses of snow occupied many of the hollows, sometimes quite down to the edge of the sea. Portions of these accunnilations undoubtedly last throughout the summer. No glaciated rock surfaces were actually observed, but this negative evidence is here of small value, erature and decrease the precij)itation on the adjacent lands. Evi- dence has, however, recently been obtained of a much more important factor in regard to late changes of climate in this region, in the observa- tions of Mr L ('. Russell, which show that the great mountain range of the Saint Elias alps must have been entirely formed in Pliocene or post- Pliocent! times.* The cruinjiling and upheaval of the beds which now form this range must have relieved a notsible and accumulating tangen- tial pressure of the earth's crust, the result of which it is yet dilHcult to trace ; but that it must have brought about extensive changes of level ♦ National Geographic Mugiiisine, Wnsliinglon, p. 174. Hull. U. S. Gool. Surv., no. 84, p. 259. 146 a. M. DAWSON — COASTS OP BERIKG SEA AND VIOINITY. I' ' I i throughout the region over which this pressure was exerted seems cer- tain, and I am inclined to 8U])pose that it may luive had much to do with the great later Pliocene uplift and subseciuent depression to which the British Columbian region ap])cars to have been subjected* One of the most remarkable features connected with the Bering sea region is the entire absence of any traces of a general glaciation. State- ments to the effect that Alaska, as a whole, showed no such traces were early made by Pall f and concurred in by Whitney. The result of mj' later investigations in British Columbia and along the adjacent coasts have been to show that such original statements were altogether too wide; that a great Cordilleran glacier did exist in the western part of the con- tinent, but that it formed no i)art of any hypothetical polar ice-cap, and that large portions of northwest America lay beyond its borders.J Statements made by Mr John Muir, in which he not only attributed every physical feature noted by him in Bering sea to the action of glacia- tion, but even expressed the opinion that Bering sea and strait repre- sented a hollow produced by glaciation, § remain altogether unsupported. It might be unnecessary even to refer to them but for the fact that they relate to a region for which the data on this subject from other sources are so small. No traces have been found of general glaciation by land ice in the region surrounding Bering sea, while the absence of erratics above the actual sealine show that it was never submerged for any length of time below ice-encumbered waters. These facts, moreover, connect themselves with similar ones relating to the northern parts of Siberia in a manner which will be at once obvious to any student of the glacial period. RcLpecting the latest changes in elevation of the land, it may be stated that in several widely separated places there is evidence of a recent slight general uplift. This was noted at Unalaska, Attu, Bering island. Saint Paul island and Saint Matthew island, but the amount of elevation indi- cated is small, being in fact from 10 to 30 feet only. ♦ Trans. Royal Soe. Cniinda, vol. vli, sec. iv, p. 54. t Alaska Coast Pilot, 1809, pp. 195, 19(1; Alaskii and its Resources, pp. iW, 461. IQuart. Journ. Gool. Soc, vol. xxxiv, p. 119; vol. xxxvii, p. 28.1; Report of Progress, Geol Siirv. Can., 1877-78, pp. 13(1 B, 1,51 B; Trans. Royal Soc. Ciinnda, vol. vii, soc. iv, plate ii, map 4. i Report of the Cruise of the Corwin, 1881, p. 147. I sw I ^1 / '» ■ . jnis cer- ; )h to do ' : which , ring sea '- State- jes were t of my . ' t coasts ' 00 wide ; .he coii- 3ap, and '■t tributed f ghvcia- it repre- pported. Iiat they sources by land erratics y length .'i t :'i