rer aod pe a” hie bebe Ut bes Ariat ipl Hy AA mt Tipo fy as 14 prin pel i oH at f ue rae ‘4 syed pe “at tars at nites S ht BS pitas Yi it ne oe Hata Se OTe Reh wines ae bat Ki 4 HPO ie ire a Hite i ae tial ! thy ah => ae atidee ae ae ae oe ie a (bts et i _ - ne i 8 i ee i Nani} 4 LY, i Taig cersaesbihiee tthe iat - a a ’ is a ud es us ni an oy ie ‘ if i ait bang snemaapo Co ise - Hike Nin yat ate ae Bee os i ee i ua i} 44 ut ut a ny ante i) rirtlate ray pe aso, = = ee Soa eres aol . niga Nh ete = a fERS 24 = ate th | i ss faa ay ie ite i ue ty a 2s ae ae ne i Sa F- ar = ane it +164 i He jen i ‘ _ Stise se. saab’ Peso : Rae Sd eS. ss ; S555 = ‘+ SSeS SSS e tS Sata nees = rel as: pets ae setae: . Same StS ine Faz. o> ri Ts Pre Fs a a ee 3 = Soe pLpsseretybeTy Lat ay 4 Aan et A o Mate i = Scesae sie eree: eo = at FRESE Sessa. eae ress aes as, aoe Sor ss =e = = a = 5 os pea ae Loos STSSE. HH atin Ait itil ur et ne ! i) | y ia : He ie cae bie ; in itty it * i iy oe Arran ihe ; tae Pee a a a i i ea i sae : Me ; . a i i Hi a i eeaitnl ase Mt atte tf ‘a Hi it ee eae a Hae stay HG a \ Hi Hees uaedst i a etettht 12: Lae NG a ? Me a ae i ie ean i aay ity it ae saya aay i 4 Higa ce i if ha H / ue 7 a i : a Hien ce a eae a 3 f p4 i is) if r ; iy " My ae nut at i ia at ria { i 1 ii it i if recip sty ii (i iit “if it iy ait Y Dees uh i AN Bit f atthe pe 4) Nii re fist an MaaM ui Dsante tae f } : rt Wh sintnt bh ee a a ey Aah Ht ) I alien ihdlh bith a ae : ae es a aie eae nt iy : a Ht Le i shat a . HH Wel i ht i oo BHR ae i nei HM ith! es i ¥ a Ke i f F | a iif } tH : i yun a A ve hl a f i ve on Hi mtr fi ae a i # Mt o ys Wes } ne na net yay: HA igs aah ; 4h Hod r K ii 7 rf Veer) Uns i ny AA te ay in Aa) i) fi a Hh iit tH ca teh i H ee ne SER i ai . aie t aon ne ie oe ot thn ont Rett Bia pele thi re “lh i irate PAU MH Rube a ue aa AE it La Mie i iy uN My mney HM ‘ if) Hi un neler ft Hota ae co oe a eae: a f ae snes hy i ann dh i r RAH abe rah AVALON ety Pant) } i) cH ‘ rere innit Ta reir aaa ea ney aie MN ‘et ne np * he TA HY iy : if i) Blew ah Hi : HJ Na t i ie Hea set fel ar ay iH ait rapa ete robe ah ie Hh te! Riad a a ha sate ce $i : Beka He ation CH | FOR DERE PEOPLE FOR EDVGATION BOR S CIE NGE LIBRARY OF THE AMERICAN MUSEUM OF NATURAL HISTORY ea ae mn BOSTON: ENED ee kev ly Ee) Oi CuuH ayes 1880. wale NAME AL ATCT» “i oe ie he Acc ss CONTRIBUTION S TO THE Goh O LO G. ¥ OF EASTERN MASSACHUSETTS. BY WILLIAM O. CROSBY. ne ee BOSTON: BOSTON SOCIETY OF NATURAL HISTORY. 1880. Lx, PUBLISHING COMMITTEE. S. H. Scupperr, | Axrpueus Hyatt, S. L. Assot, M.D., J. A. ALLEN, Epwarp BuRGESS. Press of Rockwell and Churchill, 39 Arch Street, Boston. TABLE OF INTRODUCTION . Topography and General Outlines Eozoic ForRMATIONS Naugus Head Series . Huronian . Granite Petrosilex Details of the Petrosilex Relations of the. Batrociios fs ihe Shanmuts Gioan : : : 47 CONTENTS. Petrosilex in Newbury . 55 Concretionary Structure 0 57 Petrosilex on Marblehead Neck ahd tte Aaiicout FTelends c 63 Petrosilex of the Lynn and Medford Area . 0 . : 74 Petrosilex in Needham ¢ SMe 9 79 Petrosilex in Dover, Medfield, ana edhe 3 : 80 Petrosilex in West Roxbury, Hyde Park, Milton, an ihe Blue Hill Region . C 5 ¢ 82 Petrosilex in Hingham and emortih c 5 ° : 91 Relations of the Petrosilex and Granite 5 5 0 ene tc 95 Diorite . 9 : 101 Hornblendic Gneiss, fh op — Stratified ‘Greim ‘ 6 : 9 105 Limestone . : : : . . 112 General Relations of the Baronial Rooks 0 : 0 0 . 2 116 Hozoic FormMATIoNS, — continued 123 Montalban 123 Gneiss 135 Mica Slate . 136 Argillite 137 Limestone . 0 : c : ° 139 General Relations of ie Montaliws Onke c pee bys : : 140 Stratigraphy of the Nashua and Merrimac Valleys . : : 154 General Relations of the Older Crystalline Formations of Eastern ieee chusetts ° 9 161 Shawmut Group : 163 Details of the Shawmut Goa 166 iii iv TABLE ‘OF CONTENTS. Shawmut Group in Newbury Shawmut Group in the Marblehead etn Shawmut Group in the Lynn and Medford Region ‘Shawmut Group in Brighton, Newton, and Needham . Shawmut Group in Dedham, Hyde Park, Dorchester, and Milton, Shawmut Group in the South Shore District : c General Relations of the Shawmut Group PaLEeozo1c FoRMATIONS Primordial ° © 4 a 9 : Boston Basin Stratigraphy of the Boston eaei South Shore District Quincy and Milton The Hyde Park, Mattapan, ard Sian Belt The West Roxbury and Dorchester Synclinal The Brookline and Roxbury Conglomerate Belt The Upper Falls, Chestnut Hill, and Boston Slate Belt . The Conglomerate Bordering the Newton Lower Falls and Brighton Band of Amygdaloid . 0 Conglomerate and Slate in Needham and South Natick The Broad Slate Belt, between the Boston & Albany Railroad and the Crystallines of Waltham, Arlington, Medford, and Malden : : : . : 5 Nahant , Marblehead Neck . Volume of the Conglomerate and Slate : Relations of the Conglomerate and Slate to the Orystallines Basin of the River Parker . Mineralogical Notes Pinite . Kaolinite Devonian ? Carboniferous Supplementary Note. Page 166 167 169 170 174 175 1i7 181 183 183 195 197 208 216 223 229 235 245. 250 253 1 S'¥ 261 263 265 266 267 269 269 oom wise 275 276 PREFACE. Harty in the summer of 1875 I began a systematic study of the rocks in this vicinity, with the view of collecting such data as, added to my previous knowledge, would enable me to prepare a thesis for graduation in the department of Natural History of the Massachusetts Institute of Technology, on the “* Geology of the Environs of Boston.” In the beginning of the following winter I was charged by Prof. A. Hyatt, under the direction of the Massachusetts Commission to the Centen- nial Exhibition, with the task of preparing a geological map of the State, for exhibition at Philadelphia. Opportunity was afforded me by the Commission to visit all parts of the State. The Boston Society of Natural History generously permitted the use of my time as assistant in their Museum for this purpose, and, as that winter was unusually favorable for ex- ploration, I was in the field almost constantly. The Centen- nial map and the report on the same were completed in the following spring. The wider range of observation which this work for the State had aftorded seemed to justify me in ex- (v) vl PREFACE. tending the scope of my thesis: so as to include the whole of Eastern Massachusetts, and it was accordingly finally written, under the title of this paper. The thesis was accepted for publication by this Society, and the Government of the Massa- chusetts Institute of Technology made an appropriation to pay for printing the map. I resolved, however, partly on the suggestion of Prof. Hyatt, to test my conclusions by another year of field work, and various circumstances have ‘conspired to extend this time to nearly two years. By this means my original observations have been doubled, and, although the general plan of the paper remains the same, it far exceeds its former limits, and many of the geological boundaries on the map have been brought nearer to an expression of the truth. In deference to the view of Dr. T. Sterry Hunt, exipteneell to me personally, and before this Society, I have given the rocks, which in the original paper were designated as the Norian system, a name having no chronological signification. The Huronian petrosilex and felsite are no longer regarded as synchronous with the breccias ; but the latter rocks, together with the amygdaloids, which appear to belong to the same horizon, are set apart as a distinct formation under the name of the Shawmut group. The conglomerates about Boston have been found to underlie the slates, and hence are now regarded as Primordial instead of Carboniferous. For assistance received my acknowledgments are due, first of all, to Prof. Alpheus Hyatt, who generously placed at my disposal his original notes, sketches, and collection, represent- ing observations made during several years on the geology of Essex County; and I am also indebted to him for many opportunities for investigation in the field and laboratory. PREFACE. : Vil To Dr. T. Sterry Hunt I am indebted for much valuable instruction, advice, and criticism, while a student in the Mas- sachusetts Institute of Technology and since, without which the performance of this work by me would have been impossible. Though placed in a position most favorable to a thorough acquaintance with Dr. Hunt’s views concerning the origin of crystalline rocks and their relations to the more recent, fossiliferous rocks, I entered the field with a general disbelief in their applicability to the rocks of this region, which my first observations only strengthened ; and any conclusions in har- mony with these views which may be found in this paper have been reached in opposition to preconceived ideas wholly at variance with them. For valuable data, freely contributed, my acknowledgments are also due to Mr. L. S. Burbank, of Woburn; Mr. T. T. Bouvé, President of the Boston Society of Natural History ; Mr. F. W. Very, of Hyde Park; Prof. W. H. Niles, of the Massachusetts Institute of Technology; Mr. W. W. Dodge, of Cambridge, and others. The chemical analyses recorded in this paper were made in the Woman’s Laboratory of the Massachusetts Institute of Technology, and for these I am under special obligations to Mrs. Alice B. Crosby, Mrs. Ellen H. Richards, Miss E. M. Walton, Miss H. A. Walker, and Miss Jennie M. Arms. The President and Directors of the Eastern Railroad Com- pany have assisted very materially in the exploration of Essex County by granting free passes over their line. Boston, August, 1878. th pau + iy INTRODUCTION. THE vast region extending from Long Island Sound, north- easterly, and nearly insulated by the St. Lawrence, Richelieu, and Hudson Rivers, and Lake Champlain, is well marked off, geographically, from the rest of the continent. Dr. Ezekiel Holmes,! recognizing the geographic distinctness of this most eastern member of the continent, and its almost com- plete isolation by water, an isolation that appears to have been perfect during the first part of the Paleozoic era, has correctly termed it a peninsula. Newfoundland is a detached portion of this geographic unit. In its geologic relations, also, the region indicated is readily separable from the adjoining terri- tory. ‘The rocks are mainly crystallines, and, except in the lower part of the valley of the Hudson, cannot be connected, along their strikes, with the rocks of other regions.2 The -? Ann. Report on the Agriculture and Geology of Maine, 1861, p. 100. ? The Paleozoic sediments of this region, it is true, are continuous through the broad interval between the Adirondacks and the Highlands of the Hudson, and were probably connected, at one time, through the valley of the St. Lawrence, with the more exten- sive deposits of the same age to the westward. Yet the crystallines are to so great an extent the predominant rocks of this large territory, and, as the writer conceives, have had, as a whole, an origin so vastly more remote than the most ancient of the uncrys- talline sediments (these newer formations being regarded as merely superficial deposits, often obscuring, but not essentially altering, the main structural features of the region), as to warrant leaving the latter out of view in these general remarks. OCCAS. PAPERS B. 8S. N. H.—III. 1 2 north-easterly strike prevalent north-east of a line extending, in a general way, from Boston, via the White Mountains, to the northern end of Lake Champlain, would, if continued south-westerly, carry the crystallies of New England over New York, and the Middle States generally. But beyond the line indicated the formations bend to the south, and through- out south-western New England are characterized by approxi- mately north and south strikes, being cut off at right angles by the waters of Long Island Sound. Along the New York boundary, in southern Connecticut, however, the north-east strike is resumed, and the rocks of that district sweep across Manhattan Island and the Hudson. The breadth of New England strata subjected to this westerly deflection is prob- ably not so great as at first sight appears. The crystallines of the Highlands of the Hudson are Laurentian, and are rec- ognized by all geologists as older than those farther south, in the vicinity of New York. It is the latter only that can be traced easterly into New England, or that are here considered as really belonging to this great geological province. The Laurentian strata of the Highlands have a north-north-east strike, varying toward north and south; and it is probable that in their northerly extension, although soon dipping beneath more recent formations, they sweep around to a course parallel with the one hundred and fifty miles of newer crystallines on the east, and reappear in Washington and Saratoga Counties in New York; thus connecting the Lauren- tian of the Highlands and New Jersey on the south with the great Adirondack area on the north, and constituting at all points a barrier against which abut the western edges of the crystalline series proper to New England. The complete absence, so far as has been definitely ascertaied, of Lauren- tian rocks from western New England corroborates this view. In short, the crystalline formations of the northern half of the Atlantic seaboard exhibit, in their strikes, a grand double curve. Beginning in the north-east with a trend parallel with the St. Lawrence and the Laurentide + 3 Mountains, they bend to the south in southern New Eng- land, and conform with the Laurentian barrier just pointed out; while the remnant that is not cut off in the southern extension, by the Atlantic, finally regains the former strike, and spreads across southern New York and New Jersey. This great Laurentian wall, which is doubtless continuous, under the Paleozoic deposits, from Labrador to Virginia, or beyond, formed the western shore of the ocean, in which the newer crystallines of the Atlantic slope were deposited; and, subsequently, in its southern half, it became the eastern border of the great Paleozoic sea. It has been a primary axis of continental oscillations, a centre of disturbance — itself undis- turbed. The importance of this ancient Laurentian axis, as a structural feature of the continent, makes still clearer the isolation and unity of what may be called the New England or North-Eastern geological province. It is not intended to deny the existence in New England of the older crystalline formations, for Hitchcock has shown that they certainly occur in the White Mountain region; but I may safely assert that their occurrence in the western and central portions of this province is always something exceptional. They are not characteristic of any considerable areas.. This north-eastern province is naturally divisible, in a geo- graphical sense, at least, into two parts, —a south-eastern and a north-western. The division is marked by a line of valleys parallel with, but subordinate to and less distinct than, that isolating the province on the north and west. ‘This axis of depression follows the valley of the Connecticut, from Long Island Sound, to near its source; then, curving north-easterly, and skirting the northern base of the White Mountains, it is marked by the valley of Lake Umbagog, and the great lake region of Northern Maine, passing to the northward of the Katahdin Range. Entering New Brunswick, it is continued, with the same trend, through the valley of the Restigouché River and Bay of Chaleur, to the Gulf of St. Lawrence. This secondary line of depression, like the first, is marked 4 throughout the greater part of its extent by considerable deposits of uncrystalline sediments. These are, however, almost wholly of early Paleozoic age, and show that these valleys are of very great antiquity. The rocks forming the subject of this paper occur wholly in the more south-eastern of the two divisions, or sub-provinces, above indicated. This division embraces that great deflection of our Atlantic coast-line known as the Gulf of Maine. The head of the Gulf of Maine is at Portland; and its apparently but slightly contracted mouth is guarded by the two salient angles, Cape Sable and Cape Cod; while the Provinces of New Brunswick and Nova Scotia enclose its north-eastern end as Eastern Massachusetts does its south-western. The line joining the head of the Bay of Fundy and Plymouth, in Massachusetts, may be regarded as the axis of the gulf, and is about three hundred and seventy-five miles long, with a trend approximating north-east and south-west. This line is of fundamental importance in the structure of this region, a base line for both geography and geology. The maximum breadth of the gulf at Portland is one-third of its length, or one hundred and twenty-five miles. The prevailing and all but universal line of strike throughout this entire region, from Rhode Island to the Gulf of St. Lawrence, as already stated, is north-east and south-west, — parallel with the axis of the Gulf of Maine. The rocks which meet the water of this large gulf are, for the most part, ancient crystallines ; and, where- ever these are exposed to observation, the indications are plain that the gulf is the product of erosion, — has been carved from the ancient strata by which it is bordered, — and is not the result of a synclinal or other depression of this portion of the earth’s surface. The agent of erosion appears, at first sight, to have always been the waves of the Atlantic, and to have acted uniformly from the east toward the west. A study of the true contours of this body of water, however, brings to light facts which seem to militate against this view. An examination of a Coast Survey chart of this region shows that the water does 5 not become constantly deeper as we proceed seaward from the shore, anywhere between Portsmouth, N.H., and Eastport, ‘ Maine; but, on the contrary, the water is usually as deep at twenty to fifty miles from shore as at any greater distance less than two hundred and fifty miles, beyond which the abysses of the ocean begin, and the bottom falls away rapidly to twelve hundred and fourteen hundred fathoms; and not un- frequently the depth is- actually greater at the lesser distance from shore, so that the water shallows seaward. Thus, the greatest depth recorded in the region of the Gulf of Maine, one hundred and sixty fathoms, is only fifty miles from Ports- mouth; and thirty-five miles from Mt. Desert gives us one hundred and forty-five fathoms: while beyond this the water shallows sensibly, and an equal depth is not found nearer than one hundred and fifty miles from shore. Extending easterly from Cape Cod, for nearly two hundred miles, is a broad expanse of shallow water, with depths rang- ing from five to fifty fathoms, and including George’s Bank, which reaches the surface one hundred and twenty miles from land. A similar but less distinctly marked area of shallow water stretches southward from Nova Scotia, with a breadth of sixty miles, and a maximum depth of seventy fathoms. Sepa- rating these two areas of moderate depths is a strait of deeper water, eighty to one hundred and fifty-five fathoms. Following the fifty-fathom line, this strait is, perhaps, fifty miles wide ; while for the one-hundred-fathom line it is reduced to thirty- five miles. It is plain that a broad submarine ridge or plateau extends over nearly nine-tenths of the distance between Nova Scotia and South-eastern Massachusetts, forming a nearly com- _ plete barrier between the comparatively deep water of the Gulf of Maine and the greater depths of the ocean beyond. If the sea bottom were elevated fifty fathoms, the Gulf of Maine, although still three hundred miles long, and having a maximum depth of one hundred and ten fathoms, would be changed from a broad-mouthed bay to an almost completely land-locked gulf. The glaciating agent operated powerfully to obliterate that 6 feature of the Gulf of Maine, upon which I am here insisting, viz., its nearly continuous eastern rim; which, though no less real than the western border of the gulf, escapes recognition through its submergence beneath the waters ofthe Atlantic. One of the most important effects of glacial action was the transportation of immense quantities of detrital materials from the north and north-west toward the south and south-east. According to Prof. Hitchcock, the sheet of drift covering South-eastern Massachusetts is at least three hundred feet thick; and any one familiar with the geology of this region, who notes the characters of the pebbles and boulders which this drift contains, cannot doubt that it has been largely derived from the land to the northward; nor will it seem improbable that Northern Massachusetts, and even New Hamp- shire, may have contributed something to the result. We have no reason to think the action of the ice-cap in this instance was anything exceptional; it probably swept Northern as well as Southern New England, and in even a more thorough manner. Where now is the vast accumulation of debris that, according to this supposition, must have been removed from portions of Maine and New Hampshire? I believe it has been shoved, in large part, beyond the present limits of the land, and is now spread over the bottom of the Gulf of Maine; contributing largely to the elevation of this bottom to the level of the eastern rim, and thereby tending to obliterate the latter as such. Jeffrey’s Bank and Cashe’s Ledge, lying off the mouths of the Kennebec and Penobscot Rivers, sixty to seventy miles long, two-thirds as broad, rising from a depth of one — hundred and sixty fathoms, and approaching at many points within five to fifty fathoms of the surface, should, apparently, be regarded as a huge pile of glacial detritus, which, although everywhere distinct from the eastern rim of the gulf, greatly obscures its basin-like character. That the glacial action, while tending to fill up the gulf, has contributed little or nothing to the formation of the ridge or plateau constituting its eastern border becomes sufficiently obvious when we note 7 the remoteness of this barrier from the land, and observe that a considerable breadth of deep water intervenes at all points between it and Jeffrey's Bank. If, then, the debris were removed from the bottom of the Gulf of Maine, it would doubtless appear much more clearly than now to be a genuine depression, or basin, shut off from the outside ocean by a broad barrier ; which, whether we consider it to have been, at the time when this gulf was eroded, above the sea, and backed, may be, by an Atlantis, or as having been always submerged, constitutes a serious objection to the view that the erosion of the Gulf of Maine has been effected wholly or mainly by the waves of the Atlantic. Without intending or desiring to raise the question of the existence in past geologic time of an Atlan- tic continent, it will be shown in the sequel that there are other reasons than those here advanced for believing that the Gulf of Maine was, at least during the period of its formation, land- bordered on the east as well as on the west; that it was eroded mainly, if not entirely, before the beginning of Pale- zoic time; and that the rocks bordering this gulf, in Eastern Massachusetts at least, are so disposed about the line which I have denominated the axis of the gulf as to indicate that this line is really the axis of a great anticlinal or series of anti- clinals, the erosion of which, after the usual manner of anti- clinal erosion, has, I conceive, produced the depression in question. . It has long been known that the rocks bordering the Gulf of Maine exhibit a general uniformity along the strike. This likeness is usually believed to extend not only to the Paleozoic and more recent sediments and their relations to the crystal- lines, but also to the crystallines themselves, which are con- ceived to be arranged in broad irregular belts, parallel with the axis of the Gulf of Maine, and each of which is, in a general way, of the same age and composition throughout its extent. Careful comparisons of the rocks, especially the crystallines, of these various districts, have never been instituted; and the prevalent opinion here alluded to, though probably correct, is ae) based upon general impressions gathered from hasty, bird’s-eye- view surveys, rather than detailed and accurate knowledge. During the last decade, various members of the Canadian Geological Survey, and especially Mr. G. F. Matthew and Prof. L. W. Bailey, have greatly extended our knowledge of Acadian geology. Their labors, however, have been mainly confined to New Brunswick; and Nova Scotia is yet, as regards an accurate knowledge of the composition and relations - of the crystalline formations of the province, well-nigh an unknown land. Our knowledge of the crystallines of Maine, also, is wholly imadequate for the purposes of comparative geographic study. In the study of the rocks of Eastern Massachusetts presented here, a detailed comparison of this region with those to the north-east has not been attempted ; _ our knowledge, as just stated, not being considered ripe for accurate generalization over so wide an area. Yet, although the end kept steadily in view has been merely to advance in some degree our knowledge of the character, distribution, and relations of the various rocks occurring in a limited district, it is hoped the way has been, in a measure, prepared for the higher study hinted at. TOPOGRAPHY AND GENERAL OUTLINES. The extent of territory coming within the scope of this paper is intended to be shown by the smaller of the two accompany- ing maps. It includes in a general way all that part of the State east of the north-south range of highlands of which Mt. Wachusett is the culminating point, and north of the Rhode Island boundary ; comprising the whole of Essex, Middlesex, Norfolk, and Suffolk Counties, the eastern half of Worcester County, and the northern portions of Bristol and Plymouth Counties. As shown by the map, the length and breadth of the district are approximately equal, and the area is not far from three thousand square miles, or about three-sevenths of the State. i) The topographic descriptions given us by the elder Hitchcock and other writers, and the general evenness of the surface, obviate the necessity for extended remarks under this head here ; and, furthermore, the trends of the principal water-courses and shore lines are sufficiently obvious on the map. I would call attention, however, to the general relation of the topographic features to the geologic structure, —a point likely to be over- looked in a level and drift-covered region like this. The prev- alent line of strike in Massachusetts, as is well known, is north and south, and in the western half of the State there are no exceptions of importance to be noted. ast of the Nashua valley, however, a north-east and south-west strike prevails, especially in Essex and Middlesex Counties; a comparatively limited area in the south-eastern part of Worcester County and the adjacent portion of Rhode Island exhibits a strike at right angles to this, or north-west and south-east ; while among the Paleozoic strata a nearly east and west strike is most com- mon. That ‘‘ geology is revealed in topography” is a trite axiom of the science, which is well exemplified, even in this extensively glaciated region. In a recent paper in the Ameri- can Naturalist,! ‘‘On the Surface Geology of Eastern Massa- chusetts,” I have discussed at some length the relations of the topography of this district to the glaciating agent and to the geologic structure. The general conclusion reached is, ‘‘ that, comparatively speaking, the ice-cap rested lightly upon the land, and that the topographic features having a skeleton or framework of rock are, as a rule, of pre-glacial origin.” In other words, to quote further, ‘‘ it appears probable that if the present mantle of drift were entirely removed from the face of the country, leaving a surface of naked rock, we should have in all important respects a restoration of the pre-glacial contours. And this ancient topography having been, as I conceive, fashioned mainly by agents more subtle than an ice-cap, and hence taking a deeper hold on geologic structure, would, if thus undisguised, reveal a closer correspondence with the structure 1 Vol. xi., pp. 577, et seq. 10 lines of the subjacent rocks than we are able to detect in the existing hills and valleys considered as a whole.” One of the most remarkable facts in the distribution of glacial detritus, or drift, in Massachusetts, as I have already observed, is the com- paratively great depth to which it has been accumulated over the south-eastern portion of the State. There is a marked paucity of rock outcrops in the southern half of Plymouth County ; south of Plymouth and east of Middleborough they are rarely met with; and Barnstable County is absolutely des- titute of them. It is not improbable that the solid rocks in this region are so deeply buried by the unconsolidated superficial deposits, that, if the latter were removed, the whole of Barn- stable County and a considerable part of Plymouth County would be invaded and covered by the sea. Certainly here, if anywhere, we may expect lake basins and river valleys to ex- hibit in their forms and trends a complete independence of the underlying rocks; and, as I have shown in the paper above cited, this expectation is fully justified by the facts. In Worcester County, and north of this region of excessive drift, however, the dependence of the surface lineaments on the geo- logic structure is very marked. This is especially noticeable in the case of the larger features, such as the Blackstone, Nashua, Merrimac, Parker, Ipswich, Charles, Neponset, and other rivers ; the Wachusett range of highlands; the parallel range forming the eastern rim of the Nashua valley ; the somewhat irregular belt of hills extending from Cape Ann to Beverly ; the well-known range sweeping with bold front from Swampscott to Waltham; and the Blue Hill range in Milton and Quincy. The last three lines of hills, being principally composed of un- stratified rocks, are less regular and distinct than the others ; yet they no less clearly reveal the structure of the rocks com- posing them ; for exotic rocks, being, in a certain sense, struc- tureless, only conform with the general law in giving rise to a systemless topography. The rocks of Eastern Massachusetts admit of a convenient, and, it is believed, a chronologic, division into two great ft ; groups, — the crystallines and the uncrystallines. The former predominate, constituting the surface rocks over fully nine- _ tenths of the area to which this paper particularly relates. The chronologic distinctness of these groups will probably not pass unquestioned. Yet it is true that, with the exception of the ‘argillites of the Nashua valley, which Prof. C. H. Hitchcock ! has correlated with the Primordial slates about Boston, none of the crystallines of the eastern part of the State have, of late years, been regarded as of Paleozoic or more recent age, by any geologist familiar with the rocks of this region. I have studied the relations of our Primordial strata to the erystallines at every accessible point within my knowledge, and in all cases, where the evidence is of a decisive nature, it affords unequivocal support to the view, that between the crystallines and the oldest uncrystallines there is a great chronologic break, a ‘lost interval” of immense duration; for the unconform- ability is perfect, and the imperfect lithologic resemblance sometimes observable is due in every case to derivation or to very local alteration. Furthermore, it will be shown in the sequel, mainly on stratigraphic grounds, that the oldest rocks in the State are on its eastern border, facing the Atlantic, and - that the various crystalline series appear to be so arranged geographically as to become successively newer as we proceed from Massachusetts Bay to the Berkshire Hills; so that, even if there were crystallines of Paleozoic age in Massachusetts, they would not probably be found in the eastern portion of the State. I cannot but regard the distinctness of these two great classes of rocks in this region as the most certain of all the conclusions reached in this paper. Considering the Primordial horizon as the base of the Paleozoic system, then it is true that, in this region at least, the profound break between the crys- tallines and the uncrystallines is coincident with that recognized generally between the two grandest chronologic divisions, the 1 Geological Map of Massachusetts, in Walling’s Atlas, 1871. Subsequently, in the ~ final report upon the Geology of New Hampshire, Prof. Hitchcock has referred these argillites to the ancient Huronian system. . 12 Eozoic and the three succeeding eras, — Paleozoic, Mesozoic, and Cenozoic, —for which, taken collectively, geologists have failed to provide a name. Several divisions have been established among the crystallines of Eastern Massachusetts, based, for the most part, on litho- logic and stratigraphic characters. The groups thus formed are believed to have chronologic value, to be chronologically dis- tinct. It is not without hesitation that, in describing these series as both lithological and chronological, I raise the vexed question of the value of purely physical characters as tests of age among crystalline rocks; for I began my study of the erystallines of this State with a wholesome distrust of the value of lithelogical data in establishing chronological divisions, and am not yet wholly prepared to apply this principle to wider regions. Within my limited field of observation, however, I am satisfied that its application is safe. It adapts itself so per- fectly to all the facts, so far as I know them, that I do not hesitate to assert that the lithological characters of the divisions which have been worked out among the crystallines of Eastern Mas- sachusetts — the chronological and geographical distinctness of which can scarcely be doubted —are as unlike as the faune of any two successive geological formations. ee ee ee 13 EOZOIC FORMATIONS. The oldest of these crystalline series is the smallest in extent.. I propose to designate it, provisionally, as the Naugus Head series, —in allusion to the promontory of this name on the Marblehead shore, where the rocks of this series are best developed and were first observed. Succeeding the Naugus Head series, in the order stated, are the Huronian and Montal- ban systems, and a series of semi-crystalline rocks, of very dissimilar composition, but appearing to be identical in age and geognostic relations, which, having their greatest development in the vicinity of Boston, I have named collectively the Shaw- mut group,—Shawmut being the ancient Indian name of Boston. In employing the names ‘‘ Huronian,” ‘‘ Montalban,” Bia. terms having chronological signification, to designate the crys- talline divisions of Eastern Massachusetts, I mean to assert the general lithological and stratigraphical resemblance, but not necessarily the chronological equivalence, of the groups thus designated to the great and generally recognized EKozoic divisions of other regions. My views on this subject cannot be justly characterized as merely speculative ; for at all impor- tant points they rest on the solid ground of facts. My position with respect to our crystalline or Eozoic formations, briefly stated, is this: There exists in Eastern Massachusetts a great series of crystalline schists agreeing in all important respects with the Montalban system as defined by Dr. T. Sterry Hunt. Farther east is another vast crystalline formation, differing widely from the first, and possessing all the essential char- acters of the Huronian. ‘This second series is plainly older than, and underlies, the first ; and it is underlaid in its turn by a still more ancient-looking terrane which, consisting prin- cipally of coarsely crystalline, and frequently exotic, basic rocks, presents a general resemblance to the Upper Laurentian or Norian series of the Canadian geologists. This, as above 14 stated, I have named, provisionally, the Naugus Head series. In other words, the crystalline divisions of Eastern Massachu- setts agree in composition and sequence with those established in neighboring regions ; that this implies chronological equiva- lence might fairly be questioned, and I do not assert it. NAUGUS HEAD SERIES. Several small areas of the rocks referred to this series have been marked on the accompanying map.' Of these, two are of especial importance ; the first includes the city of Salem, Salem Neck, the islands of Great Misery and Little Misery, Baker’s Island, Naugus Head on the north end of Marblehead, various small islands between Marblehead and Great Misery, -and several narrow strips along the Beverly shore; while the second area embraces all but the seaward end of Large Nahant. A glance at the map will show that these areas lie contiguous to the sea-shore ; and this fact — since these shores are for the most part rocky —has enabled me to trace the boundaries with considerable accuracy. In Salem and Peabody, however, the rocks marked as belonging to this series seem to have been more or less fluent, and are blended with the diorites of the Huronian formation in inextricable confusion; which, to- gether with the absence of outcrops north of North river, has rendered the determination of the western boundary of this area difficult ; and the line as laid down on the map is partly hypothetical. . The rocks of this series, though frequently stratified, appear in general to have been somewhat fluent, and usually exhibit more or less extravasation; but doubtless in some cases the metamorphic action has stopped short of this extreme term, though destroying all traces of bedding. In many places, as notably on Winter Island and Great and Little Misery, the entire formation seems to have been plastic, and the extravasa- tion has been so extensive that the character of the rock 1 See post., pe 23. 15 changes at nearly every rod. One important fact should be noted here, viz., nowhere in this region does the Naugus-Head series appear to be cut by eruptives belonging to another for- mation ; for all the exotic rocks of this group may be easily re- ferred to, or shown to be derived from its stratified members. _ The stratified rocks occur chiefly in Marblehead, and on the Beverly shore west of Curtis Poimt. In Marblehead the strike s E.-W., with a vertical dip; while on the Beverly shore the strike varies from N.—S. to N.E.-S.W., and the dip is thirty degrees to vertical to the N.W. The average strike of the whole system is N.E.-S.W. More or less distinct bedding has also been observed on the north side of Great Misery, at several points on Salem Neck, on Coney Island, and at one point on the north-east shore of Nahant. The rocks of this series are composed chiefly of feldspar and pyroxene. These minerals occur mixed in very various pro- portions. Perhaps the most characteristic rock is one composed almost entirely of feldspar, containing little pyroxenic or hornblendic material, and frequently destitute of it. It is usually coarsely crystalline, this variety prodominating along the Beverly shore east of Curtis Point; and the crystallization is sometimes extremely coarse, as, notably, on the Beverly shore, near the western end of West Beach, and in Marble- head, especially about Dolliber’s Point. At the former locality the rock is evidently exotic, and some of the feldspar crystals are of immense size, ranging from three to six inches in length, and one to two inches in breadth: but at Dolliber’s Point it is distinctly bedded, and is interstratified with fine-grained pyrox- enicrocks. Another, but less abundant, variety, is fine-grained, presenting a very uniform texture and appearance, and ap- proaching the compactness of felsite. This occurs at many points, but is most largely developed on Baker’s Island, which is principally formed of it. A dark, heavy, usually coarse- grained diabase or norite, varying greatly in composition, and frequently occurring as a nearly pure pyroxene rock, is the most abundant rock of this series. It is the prevailing rock on 16 Nahant, where it is frequently very coarse and pyroxenic, with a high specific gravity. A more feldspathic variety exhibits traces of bedding on the north-east shore of this peninsula, as already noticed. This stratified rock is about midway between the Spouting Horn and Maolis Garden, and is of very limited extent. The lines of bedding trend E.—W. for one hundred feet or more, and then toward the west curve abruptly, but smoothly, around to the south, forming a right angle. Dip vertical. This pyroxenic rock, in its several varieties, and the feldspathic rock just noticed, occurring chiefly as eruptives, underlie the city of Salem, and form the peninsula of Salem Neck, and the islands of Great Misery, Little Misery, Pope’s Head, and Eagle Island. The stratified rocks of this series consist usually of interstratified, frequently alternating, pyrox- enic and feldspathic beds. They are of all textures, from the finest to the coarsest ; the stratified pyroxenic rocks, however, are generally fine-grained, schistose, and very distinctly bedded. These last are well developed on Marblehead, west of Dolliber’s Point, and especially about Naugus Head. It has been suggested that some of the coarse feldspar, so abundant in the rocks of this series, may be labradorite; but the analyses thus far made do not support such a view. Yet, as only two specimens have been submitted to analysis, the question cannot be regarded as settled. The first specimen analyzed was taken from the very coarsely crystalline, stratified, feldspar rock on Dolliber’s Point, Marblehead, and was placed in the hands of Mrs. E. H. Richards, of the Massachusetts Institute of Technology, who made the following report : — SiO yee f : : i : : ee : 66.639 TO ee Meyer: Maman Tre ar Oa mMneaNe ee ere 19.2575 TsO oben Magia dnb ib cu eee a le ae 4.500 Na,O . ; . : : 5 ‘ 3 d 10.011 CaO. eke 3 i : : ! : : traces EEO a. : : ; 4 : : : : traces 100.525 a eg ee nate 17 Such a chemical constitution belongs to no known species or variety of feldspar, though coming very near orthoclase; and, fearing there had been an error in the first analysis, a second was made from the same piece,—a clear, apparently un- weathered specimen, — with a like result. A careful inspection of a large number of specimens seemed to explain the anomaly, by showing that the feldspar, although coarsely and perfectly crystalline, is probably a mixture ; the crystals appearing to be formed of thin interlaminated plates of at least two different feldspars. This can be seen only on slightly weathered sur- faces, one of the feldspars decomposing more readily than the other. The constituent layers were not analyzed, as it seemed impracticable to separate them. But if, as appearances indicate, it is a mixture, and a mixture of two species only, they are probably orthoclase, and albite or oligoclase, the former pre- dominating. The color. of this feldspar is bluish-gray, weather- ing white, and its sp. gr. varies from 2.55 to 2.60. Dissemi- _nated through it are numerous small crystalline grains of magnetite. The coarse feldspar of the Beverly shore, similar in physical characters to the preceding, was analyzed by Mrs. Crosby, and found to have substantially the same chemical con- stitution, containing one-half of one per cent. of calcium oxide. The specimens just described, from the coarsely crystalline, nearly pure, feldspar rock, are certainly not labradorite, and yet that this species exists in some of the rocks of this series does not,. in my mind, admit of reasonable doubt. Much of the labradorite in the norite of Essex County, New York, and probably of other localities, is of a pale-green color and waxy lustre, and seemingly compact. In the coarse pyroxenic rock of Nahant; the feldspar is frequently physically identical with this, and yields by analysis a larger percentage of lime. The silica and calcium oxide only were determined, with the following result : — SiO, CaO Labradorite from Westport, N.Y. A . 54.60 4.85 Feldspar from Nahant % 5 . - 48.71 8.70 OCCAS. PAPERS B. 8. N. H.—III. 2 18 The same feldspar, apparently, occurs on Salem Neck, and at several other points in that region. In a formation so desti- tute of free silica as the Naugus Head series, it were natural to expect to find the feldspar mainly triclinic; and this expecta- tion is fully realized, for, save in the very coarse feldspar rock of the preceding paragraph, I have rarely failed to observe the striz indicative of plagioclase ; and in not afew instances, judg- ing by physical characters alone, this plagioclastic feldspar is most probably labradorite. A triclinic feldspar from the west end of Salem Neck, apparently little altered, afforded Mr. Geo. H. Barton fifty-seven per cent. of silica. Dark-colored mica, probably biotite, is common in the rocks of this formation, though seldom abundant. Pyroxene appears as a principal constituent, and hypersthene is believed to occur. The massive, coarsely crystalline diabase, or norite, at Nahant is often decidedly epidotic; and the epidote is particularly abundant on the south shore, east of the steamboat wharf. The most striking, and probably the most important minera- logical character of this series of rocks is, that all members of it are absolutely destitute of quartz. Laithologists will recognize, in the complete absence of quartz from this formation, a strong indication that the principal constituent, after feldspar, is pyroxene, and not hornblende; and it is probable that every basic rock of this series may be properly classified, in a gen- eral way at least, as either diabase, norite, or pyroxenite. The coarse feldspar rock, mentioned above, and consisting mainly of orthoclase, is probably often a true syenite. That this series of pyroxenic and feldspathic rocks, with its ‘associated minerals, — which is sometimes stratified, oftener eruptive, frequently very coarsely crystalline, and always quartzless, —is distinct from anything observed elsewhere in Massachusetts, cannot be doubted. Lithologically at least, it may be said to be swz generts in Massachusetts geology. The great disturbance which the Naugus Head series everywhere exhibits, and its thoroughly crystalline appearance, stamp it as older than the Huronian and Montalban formations ; and other 19 and more important considerations confirm this view. The northern or Salem area of this series is bounded everywhere either by the sea or by rocks belonging to the Huronian system. On the north, in Beverly, we have coarse Huronian granite, such as is quarried in Quincy and Rockport; while to the west and south, in Peabody, Salem, and Marblehead, are dioriteand fine-grained hornblendic granite, also Huronian. What, now, are the geognostical relations of the Naugus Head series to these Huronian terranes? It underlies them. Everywhere, along the boundaries of the Naugus Head areas, we find the various members of this series penetrating and cutting through the Huronian rocks. But the converse of this is never ob- served. Nowhere, so far as my observations extend, does the Naugus Head series appear to be cut by the adjoining Huronian rocks; nor by any member of the Huronian system; nor, in fact, by any rocks not easily referable, as already stated, to the stratified portions of this series itself. In short, the Naugus Head series appears to be, as it-were, at the bottom; and, while it has been extravasated extensively through superjacent formations, it is penetrated by nothing foreign to itself. The relations of the Naugus Head series and the Huronian forma- tion are best displayed in the cliffs along the shores of Beverly and Manchester. Here the Huronian granite, already men- tioned, is cut extensively by great dykes and eruptive masses of rocks, both pyroxenic and feldspathic, clearly belonging to the Naugus Head series ; the feldspathic dykes appearing, usually, to be older and larger than the pyroxenic. But this granite, although it has evidently been more or less fluent, is never found cutting any member of the Naugus Head series ; and we are thus forced to the conclusion that this series is older than ‘ the granite. : The Norian beds of Canada and Labrador were formerly regarded as forming: part of the Laurentian, bearing the name “¢ Upper Laurentian” ; and, wherever their relations to the un- derlying terranes have been observed, they lie directly upon the Laurentian, never upon the Huronian or Montalban. The 20 Huronian, also, has heretofore, wherever its foundations could be clearly made out, been found resting on a Laurentian floor. The Norian and Huronian have never been observed under con- ditions favorable for the accurate determination of their mutual stratigraphic relations, and hence these are not positively known. Prof. Bailey and Mr. Matthew, however (in the Geological Survey of Canada, Report of Progress for 1870-71, p. 41), report the occurrence, seven miles east of St. John’s, in the Province of New Brunswick, of a small area of crystalline, anorthic, and hypersthenic rocks, which are regarded as Norian by Dr. T. Sterry Hunt, after an examination in situ; and ocncerning the geognostical relations of which they say, ‘‘ On the north side these anortholite rocks are met by red gneiss and granite, similar to the Laurentian gneiss at Indiantown, and on the south are covered by conglomerates and diorites of the Huronian series.” This language is explicit, and indicates that here, at least, the normal position of the Norian is below the Huronian. Yet it seems doubtful if this can be regarded as proved by the facts observable at this place; for, according to Dr. Hunt, these Norian rocks are, apparently, devoid of stratification, and may belong to an extravasated mass; while the adjoining Huronian beds are vertical. The mere occurrence, however, of these Norian outcrops between ledges of Laurentian and Huronian rocks, suggests that this is their normal position, and if their geognostical relations to the Laurentian and Huronian beds are regarded as due to the action of eruptive agencies, it is hardly conceivable that they have come from any horizon above the Huronian. Although conclusive proof that the Norian system is older than the Huronian has not been afforded us by a study of their relations at the only poimt where they are known to occur together im situ, yet the bulk of the evidence points in that direction ; and the general facts, that the Norian rocks are usually more crystalline, have in every respect a more ancient aspect, and always exhibit less unconformability with the Laurentian than the Huronian, increase the probability 21 that this is the true sequence of these great crystalline forma- tions. This, it may be added, is the sequence deemed most probable by Dr. Hunt, viz., Laurentian, Norian, Huronian. The Naugus Head series is certainly distinct from, and (as I have already shown) probably underlies, the Huronian ; and, since it bears no likeness to the Laurentian system, we are brought to the conclusion, that, if it is to be correlated with any series already described, that series is the Norian. In short, the Naugus Head series does not resemble the Laurentian, and is, stratigraphically, where we should expect to find the Norian ; and these are, mainly, the considerations which led me to designate this series as the “ Norian,” on the Centennial Geo- logical Map of Massachusetts, and in my report on the same. Having been informed, however, by so good an authority as Dr. T. Sterry Hunt, whose opinion I had no opportunity to obtain before the publication of the report above referred to, that this series, chiefly on account of the supposed absence of labradorite, cannot be regarded as of Norian age, I have em- ployed here, and on the accompanying map, a provisional designation having no chronological signification. I am con- strained to believe, however, that, save in not holding labra- dorite as a principal constituent, if such proves to be the fact, the Naugus Head series presents a fair agreement, lithologically, — with the essential characters of the Norian, as the latter has been described by Dr. Hunt. It is proper to state, in this con- nection, that Dr. Hunt, in 1869," identified, as belonging to the Norian formation, a boulder found on Marblehead Neck, and possibly derived from the Naugus Head areas to the northward. _I know by personal observation that the Naugus Head rocks are scattered as erratics all over the town of Marblehead, including the Neck. Prof. A. Hyatt long ago recognized the rocks about the city of Salem as probably older than the petrosilex of Mar- blehead Neck; but he did not separate them from the Huronian diorites of Salem, Swampscott, and Marblehead. 1 Amer. Jour. Sci. (2) xx1x., 183, 398. 22 The only rocks in Massachusetts that have been observed passing below the Huronian system, or cutting through its lower members, are those composing the Naugus Head series; and this, together with its crystalline character and immense dis- turbance, convinces me that this series is the oldest in the State. In the light of our present knowledge the conclusion cannot be avoided, that the Naugus Head series is the real base of the geological column of Massachusetts. The coarse-grained, readily disintegrating exotic diabase, so extensively quarried in Medford, —and also occurring in Somer- ville, Brookline, and, probably, other places, — bears a strong resemblance to certain members of the Naugus Head series, es- pecially to the coarse pyroxenic rock of Nahant and Salem Neck; and the idea is natural that they are extruded portions. of this series, which may, I think, be regarded as the probable seat of many of the eruptive masses cutting the newer formations of this region. For the accurate identification of these exotics in Medford, Somerville, Brighton, Brookline, etc., we are indebted to Mr. M. E. Wadsworth.! This observer has found that the principal dark constituent of these rocks, in the unaltered condition, is always pyroxene, never hornblende ; a fact which accords well with the general conclusion already stated with respect to the’ composition of the basic rocks of the Naugus Head series. The _ forms and general relations of these masses will be more fully described in connection with the uncrystallines which they intersect. The small area of basic exotics exposed near the head of Washimgton avenue, in Chelsea, has been doubtfully referred to the Naugus Head series. . North of Wenuchus Lake, in Lynn, is a hill composed of a nearly pure feldspar rock, — a coarse, whitish fieldspar, apparently the same as that so well developed on the Beverly shore. The rock is massive, having the aspect of an exotic; and around the base of the hill it can t Proc. B. S. N. H., xrx., 217. 23 be seen penetrating the Huronian diorite. It is undoubtedly safe to regard this as an outlier of the series in question. In what precedes I have described all the areas of the Naugus Head series marked on the map; but recent observations have con- vinced me that others exist. The hills immediately north of Wadsworth’s Station on the New York and New England Rail- road, in Franklin, appear to be entirely composed of a rock very similar to the prevailing type on Salem Neck, — quite destitute of quartz, and consisting chiefly of a coarse, triclinic feldspar, of bluish and grayish colors, with some mica, dark-colored, and often bronze-like, a green mineral that may be hypersthene, and a very little hornblende or pyroxene. No stratification is visible ; and the boundaries of this area are entirely unknown, save that it does not appear to extend much, if any, south of the railroad. The high hills in Sharon, near Sharon Centre, and on either side of the Boston and Providence Railroad, appear to afford another area of these rocks. ‘These hills are near the centre of the large area marked on the map as Huronian diorite ; and some observations made by Mr. F. W. Very, in Foxboro’, in connec- tion with my own, lead me to suspect that, on the geological map of the future, the Naugus Head series will demand a con- siderable portion of the territory here assigned to a newer for- mation. Among the rocks occurring here I have recognized the most of those found in the Salem and Beverly areas. They are of all textures, and,some varieties hold abundant grains of magnetite or menaccanite. In passing over the road leading north-easterly from Reading Village, and about one mile from the Boston and Maine Railroad, I have observed several ledges of a coarse, apparently exotic, dioritic rock, the chief constitu- ent of which is a coarsely crystalline plagioclase, which I am strongly inclined to believe is labradorite; the rock, in that event, probably being a norite. It is in a region where out- crops are few and ‘far between, and I could learn nothing of its extent. 24 HURONIAN. The rocks that may be referred to this system in Massachu- setts, like those of the Naugus Head series, are believed to occur only in the eastern portion of the State. They cover a wide area; and, except where the Naugus Head series, Shaw- mut group, and rocks of Paleozoic age face the Atlantic, they form the seashore from the New Hampshire line to Plymouth. This formation is bounded on the north and north-west by a line running south-westerly from Salisbury through Essex and Mid- dlesex Counties to Concord. Here, after giving off a long and narrow deflection, which continues nearly twenty miles farther to the south-west, the line bends to the south, and continues through Framingham, Holliston, Medway, and Bellingham, to the north-east corner of Rhode Island. On the south, it is met mainly by the carboniferous rocks of Bristol and Plymouth Counties. . The Huronian area has an extreme length, measured from the New Hampshire line in Salisbury to Manomet Hill in Plymouth, of sixty-five miles ; and a maximum breadth of forty miles across the southern end, not counting the narrow band stretching from Concord to Westborough. It is almost completely divided near the middle by the Primordial, and, possibly, more recent rocks, which lie about the shores of Boston Harbor, extending westerly to Natick, and south-westerly to Rhode Island. On the accom- panying maps the Huronian series comprises the areas marked as ‘‘ syenite,” ‘* porphyry,” and ‘‘ hornblende slate,” on the geo- logical map of Massachusetts, prepared by Professor Edward Hitcheock.! The existence in Eastern Massachusetts of rocks of Huro- nian age was first announced by Dr. T. Sterry Hunt, in 1871. In a paper on ‘‘ Granites and Granitic Vein-stones”? he speaks of the ‘‘ felsites” and ‘‘ felsite porphyries,” or ‘‘ orthophyres,” oc- * T have marked the Huronian boundary on the map with a heavy line, for the sake of greater distinctness. * Chemical and Geological Essays, p. 187. 25 curring in Lynn, Saugus, Marblehead, and Newbury ; and says in this connection, ‘‘ These rocks are, throughout this region, dis- tinctly stratified, and are closely associated with dioritic, chlo- ritic, and epidotic strata. They appear to belong, like these, to the great Huronian system.” Dr. Hunt has included here all the rocks which it is proposed to refer, in this paper, to the Huronian series, save the binary and hornblendic granites (which are so characteristic of: Hastern Massachusetts) and the limestones. In consequence of finding the Hozoon canadense in the serpentinic limestone of Newbury, Dr. Hunt, in 1870,! referred this limestone and the associated rocks, as well as the more crystalline and less serpentinic limestones of Chelmsford and Bolton, and the gneiss in which they are included, to the Laurentian system. He made no mention, in this connection, of the serpentinic limestone in Lynnfield, which is probably of the same age as the Newbury deposit, since the associated rocks appear to be the same. Prof. C. H. Hitchcock, also, in 1871,? referred to the Triassic period, certain diorites in the vicinity of Salem and Boston, which are here referred in part to the Nau- ous Head series, and in part regarded as of Huronian age. And more recently, in his late report (1875) on the geology of New Hampshire, he has applied the term ‘‘ Labrador” to this broad Huronian area, although these rocks have scarcely a single character in common with the Labrador or Norian series as defined by Dr. T. Sterry Hunt and the Canadian Geological Survey.? It will be shown in the sequel, however, that all the rocks within the area described probably belong to one and the same lithological and stratigraphical series, the general characters of which stamp it as Huronian. A glance at the maps will show that the attempt to delineate this formation lithologically, 7. e., to show the distribution of its various lithological members, has been attended by moderate 1 Amer. Jour. Sci. (2), xurx., 75. 2 Geological Map in Walling’s Atlas. 8 Still later, in the atlas of the New Hampshire Survey (1878), Prof. Hitchcock recognizes the existence of several limited Huronian areas in Hssex County. 26 success. More might have been accomplished with a longer time for exploration; yet much must ever remain undone, on account of the great extent to which the rocks are, in some dis- tricts, concealed by superficial deposits. A special color has been used to represent the general Huronian formation wherever the particular lithological representative is not known; but the probabilities are great that the rock, whatever it is, belongs to this age.! The Huronian system in this region, like the Naugus Head series, though in a somewhat less degree, exhibits great disturb- ance. Distinctly bedded rocks are the exception ; and, although many apparently structureless rocks are probably really strati- fied, it is undoubtedly true that a large part, perhaps the greater part, of the formation has been more or less fluent, and ex- travasation may be set down'as a characteristic structural feature. The extent to which some of the rocks of this series, in Eastern Massachusetts, are characterized by a condition ap- proaching chaos, can be fully appreciated only by those who have studied them somewhat in detail in the field. The strati- fied portions of this series have usually a N. E.—S. W., vary- ing to K.—W.., strike ; and the areas of unstratified and extrava- sated rocks generally exhibit in their trends a tendency to parallelism with the strike of the stratified rocks. The latter usually dip steeply to the north-west. The Huronian series of Eastern Massachusetts is principally composed of the following rocks, or, rather, groups of rocks :—_ Granite (hornblendic and binary). Petrosilex (passing into felsite and quartzite). Diorite (unstratified and largely exotic). Hornblendic Gneiss, Stratified Diorite, etc. Limestone. ‘ Oo BR OF bo ee 1 In adopting the plan, wherever practicable, throughout this work, of mapping the various formations lithologically rather than geologically, I have done much to keep my facts and theories separate. 27 Although so connected lithologically and stratigraphically as to be clearly members of one great series, yet these various ' groups are, on the whole, well separated, occurring mainly in large masses. The stratigraphic distinctness would be much more striking but for the wide-spread extravasation which some of the divisions have experienced. Their general separateness implies that they are, for the most part, of different ages, are chronologically distinct; and may each be regarded as a sort of sub-formation. | Whence it follows that, in all their more special relations, they admit of separate description ; and that is the plan adopted here. The different groups will be taken up in regular order, beginning with the oldest. The true sequence, excluding some of the limestone, is expressed in the foregoing classification. GRANITE. The typical granite of this region, as shown at the quarries in Quincy and other places, is a coarsely crystalline ageregate of orthoclase, quartz, and hornblende. Orthoclase is the pre- dominant mineral, and in its abundance constitutes the leading character of the rock; this is pre-eminently a feldspathic gran- ite. The hornblende is usually small in amount, and the rock frequently passes, through the disappearance of hornblende, into binary granite. All the Huronian granites of this region, and especially this typical variety, are remarkably firm and coherent, being strongly contrasted, in this respect, with the most of the granite of the Montalban and newer formations. In Topsfield, in and near the village, the granite is locally rather loose and friable ; but this is the only instance of the kind within my knowledge which is not clearly the result of atmospheric action. And it may be said that these granites, as a rule, resist chemi- cal quite as well as mechanical forces. The colors of the gran- ite are mainly due to the feldspar, the hornblende seldom being sufficiently abundant to sensibly darken the tint of the aggre- gate. The feldspar is usually grayish or bluish, though differ- 28 ent shades of red and pink are very common, and green and other tints are frequently met with. Feldspars of several dis- tinct colors are sometimes commingled in the same hand _ speci- men. A beautiful granite occurs in West Dedham and Dover, for example, in which the feldspar is chiefly of a light-green color, but contains interspersed through it numerous crystals of a flesh-red feldspar. Such instances are common, especially among the granites in Dedham and Dover, and along the Swampscott shore. The occurrence of masses of red and white granite on the south-western end of Marblehead Neck, in close juxtaposition, but not blended, is well known, and has been noted by Dr. Hunt, Prof. Hyatt, and other observers. ‘The - hornblende of the granites presents a variety of aspects, fre- quently having a greenish-slaty, or chloritic appearance ; and it is probable that our Huronian granites are often, locally, of the variety called “protogine” by the French geologists.' It is worthy of note that the Huronian granites of Kastern Massa- chusetts are usually destitute of mica, the micaceous granite of Rockport and Gloucester, to which attention has been specially directed by Mr. M. E. Wadsworth,” probably constituting the single notable exception. These granites, as every observer well knows, are subject to extensive variation in both texture and composition, and the va- rieties are so numerous, pass into each other so frequently, and are so perfectly blended, that their distinction on the map would be, at least in the present state of our knowledge, a hopeless task, and I have not attempted it. Since, however, there is such a deficiency of definite knowledge concerning the geographic relations of the different varieties of granite, even a general sketch of the distribution of the leading types may prove useful as a basis for ‘future observations. For this purpose I will consider the granites as embracing three principal varieties. These are, (1) the coarsely crystalline, little horn- 1 According to Prof. Haughton, the green mineral of protogine is never tale, but usually chlorite, or some kindred mineral. 2 Proc. Bost. Soc. Nat. Hist., x1x., 309. 29 blendic, typical granite, already mentioned ; (2) a finer-grained, less hornblendic, euritic variety, sometimes approaching petro- . silex; (3) the very hornblendic, commonly fine-grained granite, which is usually poor in quartz and frequently passes into diorite. 1. The coarse, typical granite above described, well known in Quincy, where it is so extensively quarried, may be traced the entire length of the Blue Hill range; but with greatly di- minished breadth towards the west. Westerly, it is limited to the northern side of this group of hills, and at no point does it appear to reach their southern border. A fine example of the typical granite forms the northern shore of Cohasset Harbor ; and it is the prevailing variety in Cohasset and Hingham. In- deed, judging from the observations of myself and others, and the specimens collected by Prof. Hitchcock, and now in the State cabinet, I conclude that the broad area of granite which sweeps through northern Plymouth and Bristol and southern Norfolk Counties, is mainly, and towards its eastern end almost entirely, composed of this variety. In the northern part of Plymouth County, south-east of a line drawn from Scituate to Abington, rock outcrops are few and far between, and the granite marked on the map as occurring in Kingston, Duxbury, Marshfield, etc., is largely conjectural. I have seen no indica- tions, however, of the existence here of any other rock; and hence I have ventured to color the entire area as granitic, in- stead of employing the general Huronian color, which would, perhaps, have been the wiser course. I suppose that this granite is mainly of the typical variety. Ihave observed this kind in sttu in Marshfield ; and Mr. Elmer Faunce reports an exposure of the same rock on Jones River in Kingston. At some points in this southern region the coarse granite is porphyritic with feldspar crystals. Prof. Hitchcock! mentions several localities, and I have observed a good example on Thax- ter Street in Hingham. It is always of local extent. The 1 Final Report Geol. of Mass., p. 669. 30 granite in the western part of Wrentham is mainly of the typi- cal variety ; and the same may be said of the granite in West Roxbury, and of that in the immediate vicinity of Dedham Village, and southward to the New York and New England Railroad. The large granitic area in Dover, Medfield, Natick, and Sherborn is mainly composed of the variety under considera- tion, which is here largely characterized by a greenish feldspar. _ In the northern part of Needham, also, it asserts itself, nearly to the exclusion of all other varieties ; but ceases to be the pre- vailing rock as we pass into the southern part of Weston. The eranites of these western towns —Dedham, Norwood, Dover, Sherborn, Natick, Needham, etc. — are almost universally of a reddish or pink color, sometimes greenish, but very rarely dark gray or bluish, as in Quincy, Rockport, Peabody, and other towns to the east and north-east. The granite about the southern end of Spot Pond, in Medford and Stoneham, is, so far as I know, all of the coarse typical variety ; and the same is true, with unimportant exceptions, of the areas in the northern part of Stoneham, and in Wakefield and Reading. Well- marked typical granite occurs in Saugus, south of North Saugus, and in all parts of the large granitic area in Peabody, Lynn-. field, and Lynn, northof Wyoma Lake in the latter town. The so-called ‘‘ Peabody granite” cannot be distinguished from that quarried in Quincy. The granite in eastern Lynn, on the Swampscott shore, and on Marblehead Neck and the adjacent islands, though subject to considerable variations, may benearly all referred to the typical variety. The granite composing the small area of this rock in the southern part of Marblehead con- tains little or no hornblende, is coarsely crystalline, and would belong to the typical variety but for the fact that it is nearly destitute of quartz. A somewhat similar rock, in which there is an entire absence of quartz, forms a large hill immediately north of the eastern arm of Wenuchus Lake, in Lynn. It does 1 I am indebted to Mr. J. S. Diller for knowledge of a small area of granite, not marked on the map, which exists in Malden, north of Salem Street and between Prospect Hill and Maplewood. 31 violence to the rules of lithologic nomenclature to call this rock eranite. It closely resembles a rock of the Naugus Head series on the Beverly shore; and it is probably an outlier of that formation. It occurs as an eruptive, cutting the adjoining diorites ; and I have given it the Naugus Head color on the map. Continuing northward, we find the granite of the large area between Beverly and Rockport characteristically coarse, and, as a rule, little hornblendic, especially towards the north-east, where, as already noticed, the hornblende is sometimes replaced by mica. The granite in Essex, Ipswich, Hamilton, and Tops- field, so far as observed, appears to belong here. In the latter town it disintegrates readily, and is known as the ‘‘ rotten rock.” Coarse, little hornblendic granite occurs in Rowley, along the railroad, north of the station, and at several points in Newbury, north of the river Parker. In both towns it becomes finer- grained towards the west. We have traced this type over a large area, and have found it everywhere presenting substantially the same characteristics. Though the limits of local variation in texture and composi- tion are wide, yet to the general view it remains essentially un- changed ; and the observer feels that, from Plymouth to the Merrimac, from Rhode Island to Cape Ann, it must be, in its origin and petrologic relations, one and the same rock. 2. The fine-grained, little hornblendic granite contains even less hornblende than the typical variety ; and, like that, it is frequently a binary granite, — composed entirely of orthoclase and quartz. ‘This variety not unfrequently appears, when con- taining little or no hornblende, to pass through eurite into petro- silex. Jam unable to point to any particular locality as afford- ing incontestable evidence of this transition ; yet indications of such a passage are constantly presented to the observer. He meets on every hand partial proofs; and, whether studying granite or petrosilex, he is ever finding rocks which are neither, but both, — until the gradation is complete, and the conclusion is forced upon the mind that these rocks must be unequally metamorphosed portions of one and the same original rock. This BY) is the view held by Prof. Edward Hitchcock,' after many years of observation ; and the same conclusion has also been reached by Mr. T. T. Bouvé.? Such specific evidence as can be adduced on this point will be presented after the description of the petro- silex. The euritic, or fine-grained binary granite of this region is as wide-spread, but not so abundant, as the type. The fine- grained granite in the vicinity of the trilobite quarry on Quincy Neck, along the east side of Payson’s Hill, the west and north side of Weymouth Fore River, and south and west of the Braintree station on the Old Colony Railroad, may be referred to this variety. In the western part of Braintree there can be little doubt that the rock in question passes into the Blue Hill petrosilex. The granite adjacent to the slates at Mill Cove, in North Weymouth, is identical with that adjoining the Para- doxides quarry on the opposite side of the river. ‘These ‘euritic granites differ from the type only in texture ; they are undoubt- edly identical with it in origin, and the impracticability of a geographic separation becomes apparent as we study’ their dis- tribution ; for the two varieties are everywhere found intimately associated. We have just seen that the Quincy range of gran- ite, though mainly composed of the typical variety, is bordered, about its eastern end, by a fine-grained, sub-crystalline species ; and in traversing the broad areas composed chiefly of typical granite, in southern Norfolk and northern Bristol and Plymouth Counties, and in Essex County, between Beverly and Rockport, the observer is constantly encountering small patches of this rock, which is usually of a pale pink or reddish color. I have seen it in Cohasset, along the Old Colony Railroad in North Easton, on the New York and New England Railroad in Wal- pole and Norfolk, in the vicinity of Pride’s Crossing in Beverly, and in the neighborhood of Rockport on Cape Ann. On Marblehead Neck there is a rock which is closely associated 1 Final Report Geol. of Mass., p. 667. * Proc. Bost. Soc. Nat. Hist., xvit., 220. 33 with the typical granite on the one hand, and with the petrosilex on the other; and it has been regarded by Prof. Hyatt as transitional between them. Although this is undoubt- edly its true position, yet it probably belongs on the petrosilex side of the imaginary line separating these two rocks. A some- what coarser, but otherwise identical, rock in the vicinity of Woburn and Purchase Streets, in West Medford, appears to be similarly related to the granite and petrosilex of that town. In the immediate vicinity of the Clarendon Hills Station, on the Boston and Providence Railroad, are outcrops of a rock which, though marked on the map as petrosilex, and unquestionably closely associated with the petrosilex of that region, probably belongs, in a lithologic sense, with the fine-grained granite. The coarse granite about Dedham Village passes on the north and west into a finer-grained variety; which becomes finer the farther we trace it from the type, and appears to cover all the northern and north-western parts of the large area in Dedham, marked as granitic on the maps. This rock, quarried in West Dedham, is the principal material used in the construction of Trinity Church in Boston. Beyond the Charles River, in Needham, it becomes so fine-grained and eryptocrystalline as to appear at some points inseparable from the adjoming granitoid petrosilex, or quartz porphyry. Here, as elsewhere, a sharp line of demarcation between these rocks cannot be avoided on the maps, but it probably does not exist in nature. The rocks in Needham, however, are exten- sively drift-covered ; and this inference as to the relations of the granite and petrosilex of this town might not be sustained by the facts, could they be fully known. The granite in the southern part of Dover is similar to that just described; and there are abundant indications of a passage into the petrosilex on the east, the evidence in this instance being clearer than in the other. This variety of granite occurs in the western part of Newbury, along the river Parker, and its characters are es- pecially distinct south of this stream, on the road leading south from Byfield Parish. The granite bordering either side of OCCAS. PAPERS B. 8S, N. H.—IIl. 3 34 the Merrimac, in Newburyport and Salisbury, appears to be in- termediate between this variety and the preceding. Portions of it, however, approximate to the third variety in the pro- portion of hornblende they contain. 3. The more hornblendic of the Huronian granites are usu- ally fine-grained, and the increase of the hornblende is com- monly attended by a diminution of the quartz; so that the rock exhibits, through the entire absence of quartz, and, probably, a concomitant change from orthoclase to a triclinic feldspar, frequent passages into diorite. - This variety is subject to equal or greater variation in com- position and texture than those already described, and _ its boundaries are even more difficult to trace. In northern Lynn and Saugus, between the petrosilex on the south and the typical granite of Lynnfield and Peabody on the north, is a belt of this rock nearly a mile wide and more than three miles long. Geographically, it sometimes occurs, as in this instance, be- tween petrosilex and granite; but, as a rule, it is associated with diorite, and this association is usually very intimate. This diversity of petrologic relations is due to the fact that, as re- gards its origin, we are dealing with more than one rock. Those portions only of this variety that are associated in their distribution with either the typical or euritic varieties, or with petrosilex, are marked as granite on the maps ; when connected with diorite alone, it has been (for reasons that will appear farther on) regarded as diorite, and has not received a dis- tinctive color. This rock is always finer-grained and more hornblendic when accompanying diorite than when connected with petrosilex, or with other granites; and, in the former association, it rarely approximates to either the typical or euritic varieties, or to petrosilex; in fact it is doubtful if in such cases the feldspar is orthoclase, plagioclase being more probable, —in which event the rock becomes, of course, sumply a quartzose diorite. The Huronian granites of this region sometimes exhibit traces of stratification. This structure has been observed in the fine- 35 grained hornblendic variety north of Linmere in the southern part of Peabody, and very locally at several points in Salem, Swampscott, and Marblehead. A stratified hornblendic granite, or gneiss, occurs to alimited extent among the other stratified rocks in Reading, conforming with them in dip and strike. In Peabody, on the Newburyport turnpike north of Locust Dale and near the Danvers line, Prof. Hyatt has observed a granitoid gneiss, apparently conformably interstratified with the horn- blende slate occurring there. This is probably a true gneiss. The typical granite along the Hastern Railroad in Rowley, and in Newbury south of the river Parker, presents a coarse but im- perfect gneissic structure at several points. I observed the strike in one place to be N. 60° W.., dip vertical. Portions of the granite on the west side of Hospital Point, in Beverly, have a peculiar schistose structure; the constituent minerals, es- pecially the quartz, appearing not as grains or crystals, but as small lenticular sheets, lying in parallel planes. This appears to be the granulite of the books; strike, north-west. Granite possessing a more normal gneissic structure also occurs in this locality. In Arlington, north-west of Spy Pond, there is some stratified granite, or granitoid gneiss; the bedding is quite distinct at some points; strike, north-east. A rock similar to the stratified granite of Rowley is exposed in the railroad cut at Natick, east of the station, and appears, in part, to be inter- calated with the hornblende slate and diorite. In the north- eastern part of Dover, near Mill Brook, Mr. F. W. Very has observed an interesting series of stratified hornblendic and petrosilicious rocks, some of the latter of which approximate to the euritic variety of granite. ‘The average strike is north- south. Most of the rock on the Cohasset shore, between Little Harbor and Black Rock, is a little quartzose gneiss, which, to the general view at least, does not seem to be wholly foreign to the surrounding granite, but has the appear- ance, rather, of being a portion of the granite which still retains a trace of the structure that originally characterized the whole. ‘The strike is quite constant, and averages about north- 36 west. Sofar as my observation extends, the hornblendic material of the stratified granite is never distinctly crystalline, but always presents the slaty or chlorite appearance already noticed (ante, p: 28) as frequently characterizing the hornblende of our Huro- nian granites. This incipient or bastard hornblende, although characteristic of the stratified granites, is not peculiar to them. The above are all the instances of any importance that have come under my notice, of the occurrence of stratification, or a structure resembling stratification, in the granites of this formation. Many examples of well-marked bedding have prob- ably escaped observation; and there can be little doubt that the granites in some cases really possess a gneissic structure where the rock is too coarse and massive to enable the eye to detect it. Yet I do not hesitate to assert that such phenomena must, wherever occurring, be very local; for it can be proved beyond a doubt that the Huronian granites of this region are mainly exotic. We have seldom far to look to find, in the form ef enclosed, angular fragments of clearly-stratified rocks, evi- dence of their extravasation; and near the boundaries of the granites we often observe them cutting the adjoining rocks, es- pecially if these are stratified, in a manner incompatible with any theory that would regard them, in their present condition, as chiefly indigenous. It is not meant that the extravasation has been universal: it may not have been even general, but certainly a sufficient amount of the granite bears evidence of a fereign birth to warrant the belief that the greater part of this rock has at least been fluent. Prof. N.S. Shaler, in 1869, announced the discovery, in the Quincy granite, of structure lines, which he regarded as traces of stratification ; and he was led, in consequence, to suggest the original sedimentary condi- tion of all the granite of that range. The evidence of bed- ding here is certainly much less clear and unequivocal than in the instances cited above ; and although believing with Prof. Shaler, and also with Mr. T. T. Bouyé, that all these granites 1 Proc. Bost. Soc. Nat. Hist., xi1r., 173. 37 are metamorphosed sediments, I conceive that the peculiar planes of separation (joint structure) referred to by Prof. Shaler demand a different interpretation from that proposed by him, for evidence is not wanting of the extravasation of the granite at many points along the Blue Hill or Quincy and Milton range. Following are the clearest cases with which I am acquainted. It is well known that the Quincy granite is met along its northern border by conglomerate and slate. Inter- posed between these uncrystalline sediments and the granite at some points is a compact, greenish, slaty rock, which is occa- sionally felsitic, but oftener heavy and semi-crystalline, ap- proaching diorite. It exhibits obscure traces of bedding in a few places. The actual contact of this rock with the granite is seldom observed ; it is displayed, however, at a place about one-half mile west of the Old Colony R.R., and immediately north of a quarry situated a few rods south of the southern- most part of Adams Street, on the west side of the private road that runs from Adams Street to the quarries. The con- tact line is extremely irregular; and the relation of the granite to the semi-crystalline rock is unquestionably that: of an exotic. Some three miles to the south-west, near the centre of this large granite range, about three-fourths of a mile east of Randolph Turnpike, and near the reéntrant angle in the western boundary of Quincy, is an island of this same dark- colored, slaty rock, but more distinctly argillaceous, and the traces of bedding a little less obscure. It is very distinctly cut by dykes and irregular strings of the underlying and surrounding granite. According to Prof. W. H. Niles, the _ relations of the granite and slate on Weymouth Fore River, near the trilobite quarry, afford equally conclusive evidence that at least a portion of the granite has experienced some extravasation since the deposition of the slate. ‘The slates on the South Shore R.R., immediately east of the station at Weymouth Landing, are in contact with the granite, which euts through, and overlies them in a manner possible, appar- ently, only with an exotic; and at the contact of the granite 38 and. slate south-west of the station, Prof. Niles has observed angular fragments of slate actually enclosed in the granite, _ though lying only a few inches from their original positions in the parent bed. The induration, as if by heat, of the slate and conglomerate at most points where they adjoin the granite, and the frequent development of amygdaloidal characters in the slate in those places, are also facts which tell strongly in favor of the former igneous condition of the granite. The evidence of the extravasation of the granite afforded by a study of its relations to the uncrystalline rocks appears to be sufficiently conclusive as regards the portions of granite imme- diately involved. But since we know that the extrusion of this rock occurred mainly at a period anterior to the depo- sition of these uncrystallines, it is to the relations of the different varieties of granite to each other and to the other erystallines that we must look for proof of the exotic nature of the granite taken as a whole; and here the evidence is both abundant and convincing. Following are the more striking of the many proofs of the extravasation of the granite that have come under my notice. At Hospital Point on the Bey- erly shore, near the water’s edge south of the lighthouse, is a considerable mass of a distinct mica-slate enclosed in the coarse, structureless granite. A smaller mass of a similar stratified schist is enclosed in the. granite near the northern end of the railroad-cut in Beverly. On Marblehead Neck the relations of the granite to the fine-grained, distinctly stratified schist occurring there, are such as to leave no doubt that the granite is exotic. Along the shore, at the south-western end of the neck, the exposures are magnificent, and one can see, especially at low tide, numerous angular, ragged, contorted masses of the schist, of various sizes, enveloped by the granite. Farther east, at the northern limit of the granite on the eastern shore of the neck, a large dyke of this rock cuts through the adjoining petrosilex. The granite composing the small area of this rock on the west side of Wenuchus Lake, in Lynn, is mingled with the adjoining petrosilex in such 39 inextricable confusion as to force the conclusion, that at least one of these rocks has been fluent; and near the lake shore, enclosed in the granite, are angular fragments of a schist resembling that on Marblehead Neck. The same phenomenon has been observed by Prof. Hyatt and myself at many points in Peabody and Lynnfield, especially in the quarries about the village of Peabody, and along the western border of this large granitic area. In North Saugus, on the high hills to the west of the private road running north from Forest Street and Central Brook to Water Street we have, perhaps, the finest example of the extravasation of the granite yet observed in this region. The exposures of the rock here are re- markably good ; and the granite is coarse and sharply defined, where it penetrates the adjoining petrosilex and hornblende slate im irregular dykes, or envelops isolated masses of these rocks that have been wrested from the parent beds. The relation of the granite and stratified rocks along the rail- road in Natick appears to approximate to this in part. In the western part of Wrentham, about one-half mile from the western boundary of the town, and the same distance north of the Rhode Island line, what appears to be Huronian granite has been observed cutting fine-grained schists that I refer to the Montalban age. To the north of this, about Wads- worth’s Station on the New York and New England Railroad, and generally throughout this southern region, especially in Foxborough, Sharon, Canton, Franklin, etc., the granite and diorite are so related that one must be exotic, probably both. A glance at the map suffices to show that the geographical distribution of the granite is in harmony with the theory of its extravasation ; for it is seen to occur among the other members of the Huronian series in a manner decidedly irregular. It does not form continuous, well-defined areas; but is found in isolated patches, with boundaries difficult to define, and such as can be readily accounted for only on the supposition of a former plastic or fluent condition of this rock. The large area marked as granite between Boston Harbor and the Carbonif- 40 erous formation of Plymouth and Bristol Counties would appear to be an exception to the above statement; but it should be said of this area that it is largely covered by drift material, especially toward the southern border, and compara- tively few observations of the rocks have been made. Slates are said to occur in Abington; and it is known that there are occasional small masses of diorite included in or cutting through the granite. These appear to be very irregular in outline, and are most numerous toward the west, where we approach the large area colored as diorite on the map. The accurate mapping of these limited patches is practically im- possible ; and, since we know, from such observations as have been made, that the diorites can form but a small proportion of the whole, I have ignored them on the map, marking the entire region as granitic. A more detailed map, such as it is hoped the student of the near future will construct, would show the boundaries of the granitic areas to be much more complex than they are here represented, where we have hardly an approximation to the intricacy of nature. A farther inspection of the map will make it evident, how- ever, that the granite is not without some system in its dis- tribution. The granite areas north of the Boston and Albany Railroad, and south of the Essex branch of the Eastern Rail- road, are arranged along a line having a direction about N. 50° E.; forming a discontinuous band of variable width, which extends from Natick to Rockport, on Cape Ann, a dis- tance of forty-five miles; and we readily find in this wall of granite, probably the most resistant rock of this region, a sufli- cient cause for the existence of this prominent headland. The slate and conglomerate in Newton and Watertown are proba- bly underlaid by granite, which would account for the discon- tinuity of the range here. The granite in Marblehead and Swampscott, though considerably out of the line of this belt, is evidently connected with it: and, if the water were re- moved from those shores, we should probably find that these isolated patches are really united, forming an irregular Al offshoot or spur of the large granitic area of Lynn and Peabody. Similarly, the granite in Reading and Wakefield may be regarded, not as an isolated mass, but as connected, under the drift, with the larger tract of granite on the east. This connection is indicated on the map. I have sought in vain for a surface connection between the Peabody granite and that in Beverly. Between this Natick and Rockport granite belt and the large area of granite, already noticed, on the south, are several considerable masses of this rock, of which the Blue Hill range is one; and it is probable that part at least of the territory marked on the map as general Huronian, in Medway, Medfield, and Franklin, now extensively drift-cov- ered, is underlaid by granite. The attempt to detect any system in the arrangement of these masses has been attended with poor success ; and I can only say that their present complicated and apparently systemless disposition appears to be due to two forces operating to arrange the granite along two distinct lines; one line coinciding in direction with the Natick and Rockport belt, north-east and south-west, and the other having an east and west trend. It will be readily observed that the Dover, Dedham, and Blue Hill granite masses are in line with each other, and appear to form an east-west band of this rock, connected with the Natick and Rockport band, in Natick, and with the broad southern area, in Braintree. But this may be illusory, a more probable view, perhaps, being that these masses are portions of independent north-east and south-west bands parallel with that already described, the south-western extensions of which are concealed by newer rocks or drift. North of the Natick and Rockport granite range, in Essex County, we find a parallel but shorter belt stretching through southern Ipswich, Hamilton, and Topsfield. The granite in Essex connects this ,band with the longer range. Another band runs with nearly the same course through Rowley and Newbury, south of the river Parker, branching westerly ; and north of this stream a narrow band of variable width reaches 42 from Byfield, where it appears to connect with the area last mentioned, along the north side of Kent’s Island and Old Town Hill, probably, under the drift, to Plum Island River. Still farther north is the Salisbury and Newburyport band, which extends from the western boundary of the formation, along either side of the Merrimac to its mouth. This is the shortest of all the bands, and its boundaries are the least perfectly known.' These four granite ranges— the Ipswich, Rowley, Newbury, and Salisbury ranges — are essentially parallel with each other, and with the Natick and Rockport zone on the south. They all narrow rapidly westward; so that only one, the Salisbury range, appears to reach the western limit of the formation. In the Rowley and Newbury bands, notably, the granite is coarsest toward the east. Toward the northern end of Essex County, in Newbury and West Newbury, granite is known to exist outside of the areas indicated above. It occurs in narrow bands and irregular patches, parallel, in a general way, with the large areas between which they le, and too small for delineation on the map. The number and boun- daries of these small areas remain to be determined. The granite described in the preceding pages is not known to occur in Massachusetts beyond the limits here assigned to the Huronian system. ‘Two areas of ‘‘ syenite,” it is true, are rep- resented on the geological map prepared by Prof. Hitchcock as occurring in the Connecticut Valley; one in Belchertown and Ludlow, and the other in Hatfield and Whately. But these rocks are really, for the most part, hornblendo-micaceous granites, which have a Montalban rather than a Huronian aspect, and are intimately associated with the Montalban rocks 1 The northern half of Essex County is extensively drift covered; and, notwith- standing this county was regarded by Prof. Hitchcock as, on the whole, the most rocky in the State, outcrops are wanting over most of the area north of the Beverly and Cape Ann range of hills. My opportunities for exploration in that region, also, have been rather limited. And hence the boundaries of some of these granitic areas, as they appear on the map, are, within certain limits, largely conjectural. Where out- crops were wholly wanting, I have been guided by the character of the boulders contained in the drift. 43 of that region. The relations of the granite here referred to the Huronian system, to the Naugus Head series, have been dis- - ~ cussed (ante, p. 19). It can scarcely be doubted that it overlies this ancient series, and belongs to a newer system. It is associated in some part of its distribution with every member of the Huronian series, and in all cases the indications are plain that the granite is the older rock. It has been set down by all observers as the oldest rock in this region, and this view is abundantly justified (except as regards the Naugus Head series), not only by its generally coarsely crystalline aspect and the great disturbance and almost complete absence of stratifica- tion which it everywhere exhibits, but also by the general fact that it cuts, as an exotic, all the other members of the Huronian system. In fact, it pierces, in its well-nigh universal extray- asation, every rock in this region, save the Naugus Head series and the newer uncrystallines. In its geographical dis- tribution we have a strong indication that the granite belongs to the Huronian system ; for it is co-extensive with that system, and does not occur beyond its limits. And it will be shown farther on that its lithological relations point indubitably to the same conclusion. But its petrology makes it clear that, if the granite is referred to the Huronian series, it must be regarded as the lowest, and hence the oldest member of that series. It appears, in fact, to be the foundation of the Huronian system in Massachusetts. It will not have escaped observation that this so-called Hu- ronian granite has, as a whole, a decidedly Laurentian aspect ; and some geologists would probably refer it to that great funda- mental system. Indeed, the granite of the Newbury belt, de- scribed above, is one of the rocks associated with the serpentinic limestone of that town, and referred by Dr. T. Sterry Hunt to the Laurentian, as noticed (ante, p. 25). The most character- istic rock of the Laurentian system, it is well known, is a frm, coarse, granitoid, and, to the eye, frequently un- stratified gneiss, composed chiefly of orthoclase and quartz, with little hornblende and less mica, —a description which t4 applies very well to the granites in question, save that they are generally exotic. ‘There are, however, two important considerations which, in my estimation, throw the weight of evidence against the view that these granites are Laurentian. These have already been stated. First, the relations of the granite to the petrosilex make it necessary to regard these two rocks as members of the same series, so that if one is Lauren- tian, the other must be also. Now, the petrosilex is taken as the most typical Huronian rock in this region; and, if it is included in the Laurentian system, there is really no reason why the other rocks, here referred to the Huronian age, should not be disposed of in the same way; for they appear to be as intimately associated with the petrosilex on one side as the granite is on the other. Second, this granite overlies the Naugus Head series, and hence, if the former is Laurentian, the latter must be a distinct formation below the Laurentian ; which is, to say the least, highly improbable. It might be urged by some that we have here an inversion of the natural order; that the granite, though belonging to an older system, overlies the Naugus Head series in consequence of some gigantic fault, overturn fold, or igneous overflow; the extravasation of the Naugus Head series through the granite having occurred sub- sequently to the inversion. ‘This violent hypothesis, however, is entirely unsupported by facts, except that evidences of stratification are more frequently met with in the Naugus Head series than among the granites, indicating that the latter are the older rocks ; and this indirect evidence is offset by the fact that the Naugus Head rocks, especially the feldspathic varieties, are, as a rule, much more perfectly and coarsely crystalline than the granite. Of course, that the granite occupies the geologic horizon here claimed for it can be proved directly and conclusively only with respect to those portions of this rock showing either a passage to petrosilex, or intersection by rocks of the Naugus Head series, —a small part of the entire granite formation. Hence the way may be considered open for the introduction of evidence 45 pointing to a division of the granite between two distinct formations, the Huronian and something older ; but I have seen -no evidence of this character; all the facts, so far as I have observed, going to show that, as respects its origin at least, the granite is a single rock. Although well satisfied that a large proportion of the granite has been in a state of igneous plas- ticity, yet its relations to the stratified petrosilex and the many traces of bedding which it still retains forbid me to believe that the mass of this rock has been elevated from any vast depth; it seems rather like an extensive stratified formation which has been softened zm stu, and then to a greater or less extent forced out of its normal position by the pressure of surrounding and overlying terranes. PETROSILEX. e The determination of the relations of the group of rocks here included under the general name of petrosilex is the most difficult problem in the geology of Eastern Massachusetts. No other rock in this region is subject to such great variations in composition and structure, or is more puzzling in its petrologic relations. This diversity is not apparent at first view. On the contrary, these rocks are likely to strike the casual observer as comparatively uniform throughout their distribution and ‘simple in their relations ; and it is only by a careful study of the whole field that the mind is finally divested of this idea, and the con- clusion reached that these rocks are not one, but many. Porphyry is a much-abused word, which, like syenite (by syenite is not meant the aggregate of orthoclase and hornblende to which this term is now applied), has outlived its usefulness, and should, in the opinion of the best geologists, be allowed to become obsolete in its substantive use; as a geological term. Much of the rock in this region to which the term porphyry is applied is not even porphyritic; showing into what logical errors we are led when we choose for the basis of a lithological name a property common to many rocks, and which cannot be 46 ‘correlated with any particular chemical constitution. A char- acter so superficial as the one in question, which is common to rocks so unlike chemically as a hornblendic diorite and a quart- zose petrosilex, and which is constant in no rock, appearing to be indifferently present or absent in rocks which, saving this property, are identical; which is, in fact, as inconstant and fleeting as color; is certainly valueless for the purposes of litho- logical classification. Those geologists who favor a reform in lithological nomenclature usually employ ‘‘ petrosilex” and << fel- site” to designate the rocks under consideration, but make further reform necessary by treating these terms as synonyms and ap- plying them indiscriminately. These rocks have a wide range in mineralogical and chemical composition, and include two types that should probably be regarded as specifically distinct : one consisting of an intimate mixture of orthoclase and quartz, or of orthoclase alone ; and the other composed essentially of a plagioclase feldspar, and rarely containing much quartz, fre- quently destitute of it, but often appearing, when poor in quartz, to contain some hornblendic material. It seems best to agree with Prof. Phillips in restricting ‘‘petrosilex” to the former rock, and ‘‘felsite” to the latter. Many lithologists apply the term ‘‘ porphyrite” to the basic division or species, but this name appears to be quite as objectionable as por- phyry. In many cases it is hardly possible to determine, without a chemical, or at least a microscopical, analysis, to which type a. rock may belong; for they agree closely in most of their physical characters; but this cannot be urged as an objection to the proposed limitation of the meaning of the terms, nor to their intelligent use where possible, any more than in the case of diorite and dolerite, or of any compact rocks haying a strong physical resemblance. It is probably true that in this region these rocks pass into each other insensibly, constituting a perfect series, to the two extremes of which only, the terms pe- trosilex and felsite, as here defined, are strictly applicable. But this is after the manner of all rocks, and is as true of granite and diorite, slate and conglomerate, as of those in question. 47 In geology, as in nature generally, hard and fast lines of de- marcation are the rare exception. We must rely mainly upon chemical and microscopical analysis for the data necessary for the separation of the pet- rosilex and felsite. But few analyses have been made for Eastern Massachusetts; so that, although well satisfied of the existence of both these rocks in this region, I have not attempted to trace the distribution of each separately on the maps. There can be no doubt that the petrosilex is much more abundant than the felsite, and hence on the maps, and, frequently, in these pages, I have, for obvious reasons, in- _ cluded both rocks under the former term. Relations of the petrosilex to the Shawmut Group. — But a much greater difficulty now confronts us. The petrosilex of this region is overlaid at many points by a group of rocks, including the well-known petrosilex breccia, which appear to be in every case merely the more or less thoroughly reconsolidated mechanical débris of petrosilex itself. This second group of petrosilicious rocks constitutes one member or division of a formation much newer than the Huronian, for which I have proposed, provisionally (ante, p. 13), the name Shawmut group: a semi-crystalline series which, as will appear in the sequel, underlies the Primordial slate and conglomerate of Eastern Massachusetts, coming between these oldest Paleozoic sediments and the Huronian beds, and appearing to have been formed toward the close of Eozoic time. The petrosilicious portion of the Shawmut group includes rocks of all textures, from a coarse breccia to a compact, homogeneous rock which the naked eye cannot distinguish from the parent petrosilex. They are proved to be of more recent origin than the Huronian petrosilex, not only by their petrological relations, since they everywhere overlie the Huronian, but also and most conclusively by the fact, already stated, that they are composed mainly of the débris of petrosilex, which, where the material is coarse, can be plainly seen to be identical with that which may be re- ferred with certainty to the Huronian system. In short, the 48 so-called porphyries of this region are separable chronologic- ally into two kinds, which are genetically related, and the real petrosilex, which belongs exclusively to the Huronian formation, is the oldest. The non-petrosilicious members of the Shawmut group are in the main lithologically dis- tinct from anything in the Huronian system; and the same is true of the petrosilicious portions of this group, where the breccia or conglomerate character is well developed. But where, as sometimes occurs, the recomposed rock has been formed as a fine-grained, compact, flinty-looking sandstone or slate, or, as is also perhaps not infrequently the case, has reached this condition by the loss of the breccia structure through subsequent metamorphic processes, the separation of the Shawmut rock from the Huronian petrosilex becomes a matter of extreme difficulty. The parent and offspring are. then chemically and, to the naked eye, lithologically indistin- guishable, and we are obliged to rely upon their petrological relations and microscopical characters for the means of dis- guishing them. And since, throughout its distribution, the new, pseudo petrosilex lies for the most part directly upon the old, sometimes in such a manner that the finer-graimed, more truly petrosilicious portions are brought in contact with the vastly more ancient parent rock, from which the whole were derived, and which they so closely resemble; and since there is not wanting abundant evidence that all these rocks have _been subject in a great degree to the action of disturbing and metamorphic agents subsequently to the deposition of the Shawmut group, not only increasing their strong lithological resemblance, but adding immensely to the complexity of their petrological relations, —it will be readily seen how puzzling a problem it is to determine, for the purpose of accurate delinea- tion on the map, or for detailed study, the limits of the Huro- nian petrosilex. So intimate is the association of the Huronian petrosilex and the petrosilicious rocks of the Shawmut group, and so complete, apparently, is the gradation existing be- tween the typical varieties of the, former rock and the normal 49 or unaltered petrosilex breccia, —a gradation due partly to the conditions of deposition of the Shawmut rocks, and partly to the unequal and in some cases extreme metamorphism of the breccia, whereby have been reproduced in this newer formation several types of structure characteristic of the parent beds, — that most observers have failed to distinguish them, considering the whole as of approximately the same age. ‘This manner of regarding the phenomena in question leads naturally to views concerning the origin of these rocks which find their culmina- tion in that proposed by Prof. Alpheus Hyatt and Mr. T. T. Bouyé,! according to which all these rocks, including the homo- geneous, structureless petrosilex, had originally a conglomerate structure, the present variety being mainly due to the unequal and unlike metamorphism of different portions of the primitive conglomerate. Substantially the same idea was subsequently expressed in my report on the Geological Map of Massachusetts, with the important exception that I there dissented from the conglomerate . origin of the homogeneous petrosilex and the evenly banded petrosilex, claiming that these were deposited as fine sedi- ments bearing the relation to the other rocks of slate to conglomerate. A more extended and careful study of these rocks than was possible before the publication of that report, however, has _ greatly increased my knowledge of their relations and distribu- tion, and led me to the adoption of the conclusion stated on the preceding page; which, in its essential features, — as regards the separation of the petrosilicious rocks here referred to the Shawmut group from the Huronian petrosilex, and the mutual relations in time and space of the two series thus formed, — is identical with the view previously held by Dr. T. Sterry Hunt.? The difficulty of distinguishing the Huronian petrosilex from 1 Proc. Bost. Soc. Nat. Hist., xviii, 217. 2 President Hitchcock, also, with his usual sagacity in such matters, expressed the suine idea, with singular clearness, nearly forty years ago. OCCAS. PAPERS B. 8, N. H.— In. 4 50 the more recent rocks resembling it is not diminished by the fact that a portion at least of the ancient-looking petrosilex of undoubted Huronian age contains petrosilex pebbles. These pebbles are usually somewhat angular, often of the same color as the enclosing rock or matrix, and although, as a rule, sharply outlined, yet frequently very inconspicuous, requiring weathering to make them visible. The inconspicuousness of part of the pebbles seems to arise in some eases from their identity with the enclosing rock, and sometimes from the partial blending of the pebbles and paste, whereby the outlines of the former are destroyed, and the pebbles, even when differing in color from the paste, appear as mere ill-defined spots or blotches in the latter. The pebbles seldom appear to be abundant or evenly distributed, the rock rarely presenting the aspect of a normal conglomerate or breccia. Concerning the true signifi- cation of these pebbles three hypotheses naturally suggest themselves: First, that they represent traces of a conglomerate . structure originally characterizing, perhaps, considerable por- tions of the Huronian petrosilex. Second, that they have re- sulted from the local disturbance and internal movement and crushing of the petrosilex, and the subsequent partial cementa- tion (lithologic regelation), under peculiar conditions of heat, moisture, and pressure, of the fragments thus produced ; a sup- position which, according to Prof. Pumpelly,' accounts for the breccia structure of portions of the petrosilex of Pilot Knob, Missouri. Third, that the pebbles have become imbedded in the petrosilex while the latter was in a liquid or plastic state, perhaps undergoing extravasation. That the Huronian petrosilex is now for the most part a stratified rock, and was originally wholly so, I cannot doubt; and it appears most probable that the conditions presiding over its deposition differed in degree only, if at all, from those that have obtained in more recent geologic times. Extensive beds of Huronian conglomerate, it is well 1 Geological Survey of Missouri: Preliminary Report on the Iron ores and Coal-fields, 1872. 51 known, occur among the other rocks of that age in the Lake Superior region; while in New Brunswick Messrs. Bailey and Matthew report’ the occurrence of well-marked conglomerate or breccia in each of the three groups which they have established among the Huronian rocks of that province; hence I conclude there is, on general principles, no a@ priore improbability in the supposition that the pebbles in our Hu- ronian petrosilex were deposited cotemporaneously with the other materials composing this rock. There are many instances where the second of the hypotheses above stated is an adequate explanation of the phenomena in question. The pebbles in these cases are usually obscure, often escaping observation on the first view of a fresh or polished surface. They pos- sess a special interest at some points, as throwing light on the origin of a peculiar variety of banded petrosilex, and will be noticed again farther on. Limiting our attention now to the more conspicuous of the included pebbles, we find that, as a rule, they are either too rounded, or differ too much in color and texture from the enclosing paste, to be brought within the scope of the regelation theory; and although the first hypothesis is beset by no special difficulty, save that it in- volves the existence of some older petrosilex from which the pebbles were derived, I am constrained to believe that we must look to the third supposition for the true explanation. For, so far as I am aware, those limited portions of the Huronian petrosilex containing distinctly marked pebbles occur chiefly in the immediate vicinity of the petrosilex breccia of the Shawmut group; and in nearly every instance the evidence is plain that the petrosilex has experienced some, extravasation, and has, consequently, been in a more or less fluent state. The point where the enclosure of pebbles in Huronian petrosilex is least open to doubt -is on the eastern shore of Marblehead Neck. The rock exposures here are exceptionably good, and afford ample facilities for the study of the mutual relations of the Shawmut group and the Huronian system. The former re- 1 Geological Survey of Canada; Report of Progress, 1870-71. 52 poses directly upon the latter; and both series have evidently suffered great disturbance. The light-colored Shawmut breccia is cut and torn in every direction by tortuous dykes of the black petrosilex, which itself very clearly holds angular pebbles of different varieties of petrosilex. Many of these pebbles are of the same general character as the enclosing rock, often ap- pearing to be identical with it, and as a consequence very inconspicuous ; but those which first arrest the observer’s eye, including a large proportion of the whole, are a highly crys- talline, whitish variety of petrosilex which is precisely similar to that forming large ledges a few rods to the northward. ‘The main point to be proved in this connection is brought out at this locality with especial clearness; viz., that the very same petrosilex which holds pebbles underlies, and is in part eruptive through, the breccia, forcing the conclusion that, in spite of a cer- tain superficial resemblance, these are distinct formations. This interesting locality is more fully described on a subsequent page. There are other localities where the evidence is nearly as plain, and poimts to the same conclusion. Among these are the vicinity of Cliftondale Station on the Saugus Branch of the Eastern Railroad, West Dedham, and Hyde Park. At Clifton- dale and in Hyde Park, the majority of the included fragments are of precisely the same character as the enclosing petrosilex, and are sufficiently accounted for by the internal crushing and regelation theory ; but other blocks—of quartzite, diorite, and granite —must have been introduced from without, and imply some plasticity in the petrosilex at the time of their enclosure. To summarize, the facts observable at the places named, and elsewhere, compel us to suppose that the petrosilex, while suf- fering great disturbance, has been locally crushed and brecciated, and that certain portions of this rock, perhaps as a consequence of enormous friction, have been softened to: an extent that would permit the envelopment of extraneous masses. In say- ing this I am conscious that such inferences must be drawn with the utmost caution, if we would avoid mistaking petrosi- liceous breccia for brecciated petrosilex. — J ea D3 Returning now to the question of the geographic separation of the Huronian petrosilex and the petrosilicious rocks of the Shawmut group, it is hardly necessary to observe that the dis- tinctness and simplicity of their common boundaries, as repre- sented on the map, are in most cases a very inadequate expres- sion of the complexity observable in the field. Along the - River Parker, in Newbury, the Shawmut rocks are non-petro- silicious, or, when petrosilicious, so far as I have observed, they are never compact, but have a distinct breccia structure ; and hence they may be readily distinguished from the adjoining petrosilex. On Marblehead Neck there are several limited areas of breccia. These are mere superficial patches lying ir- regularly upon the petrosilex ; and they are much smaller and more numerous than indicated on the map. They are repre- sented as occurring chiefly along the shore; but it is not im- probable that more or less breccia exists in the central portion of the Neck, covered by the soil. The islands north-east of Marblehead Neck marked on the map as composed wholly of petrosilex, nearly all comprise more or less Shawmut rock. It is represented as occurring on Lowell’s Island, and Ihave also observed it on Marblehead Rock, Halt-Way Rock, and South Gooseberry Island. The Shawmut breccia appears at Red Rock on the Lynn shore, but does not, apparently, extend far inland. The petrosilex in the immediate vicinity of Dungeon Rock, in Lynn, is distinctly stratified, and may belong to the Shawmut group; yet I think it is more probably Huronian. As already indicated, the breccia occurs at many points, but not continu- ously, through southern Saugus; and I have observed it as far north as the Pirate’s Glen, where the pebbles are small and the rock more compact than south of the railroad. In Malden, east of Maplewood, the petrosilex is skirted at intervals along its southern border, following Salem Street, by breccia. ‘This breccia is also less distinct, and more like petrosilex, than that in the vicinity of Cliftondale and East Saugus. Following the road north from Maplewood, we find nothing but breccia be- tween the Saugus Branch R.R. and the granite in Melrose. d4 West of the Boston and Maine R.R., in Malden and Med- ford, the Shawmut petrosilicious rock covers a considerable area. It is a well-marked breccia in the vicinity of Salem Street, but becomes smaller grained toward the north and west; and near the southern boundaries of Stoneham and Melrose it is often practically indistinguishable, with the naked eye, from Huro- nian petrosilex. The Shawmut rock composing the wedge- shaped area in West Medford, along the north side of the - Mystic River, is, in the main, impalpably fine, and easily mistaken for a Huronian rock, yet portions of it have a sandy or slaty texture, and determine the age of the whole. The region between the Malden Highlands and Maplewood has been but slightly examined. It embraces some petrosilex; but breccia is probably the prevailing rock. The petrosilicious rocks in northern Saugus and Melrose, and southern Wakefield, belong largely to the newer series. A considerable area of Huronian petrosilex, however, occurs in the vicinity of Crystal Lake, and the Greenwood Station, in Wakefield ; and it crops out at many points in Saugus and Melrose, a few only of which are indicated on the map. The relative distribution of these two series in this region, —Saugus, Melrose, Malden, and Medford, —as shown on the map, can be regarded as accurate only in the most general sense. I have traced the boundaries of none of the areas; and in some places, especially in North Saugus, the Huronian and Shawmut rocks are so difficult to distinguish lithologically, and are so closely involved petrologi- cally, that it is doubtful if their limits ever can be satisfactorily determined. I believe, however, that I have succeeded in separating roughly the areas in which petrosilex predominates from the areas in which breccia is the prevailing rock. The Shawmut rock in Needham, not being petrosilicious, is easily separated from the petrosilex. Although the petrosilex of West Dedham and Dover is without doubt mainly Huronian, I am not. confident that it all belongs to that ancient system. The area marked as breccia on the map includes all of that rock and a portion of the petrosilex.. The petrosilex west of the 5d Boston and Providence Railroad, in Hyde Park and West Rox- bury, is certainly mainly of the ancient type. Proceeding eastward, as we mount the bold ridge east of the Clarendon Hills Station, breccia again makes its appearance. Further east it occurs in large patches, but not continuously, from Calvary Cemetery on the north, nearly to Hyde Park Station on the south. It increases rapidly eastward, and covers the most of the area between Back Street and the New York and New England Railroad, the petrosilex appearing at intervals. East of this railroad the petrosilicious character of the Shawmut rocks appears to die out, and they are no longer difficult to dis- tinguish from the Huronian beds. The same is true south of the Neponset, in Milton, except about the eastern end of the more eastern of the two petrosilex areas occurring there ; at this point the breccia reappears. . In the foregoing attempt to separate the Huronian petrosilex from the Shawmut group I have determined which rocks are Huronian mainly by the following criteria : — All compact, flinty rocks, having a clean fracture, and which (to the eye) are perfectly homogeneous (except where porphyritic or banded), showing no traces of a pebbly struc- ‘ture, were regarded as Huronian ; also all petrosilicious rocks, pebbles of which have been found in the Shawmut breccia. This last test is applicable only where the rock possesses pecu- liar characters by which it can be unfailingly identified. The separation is not, and is not claimed to be, exact, but simply the best that my data will allow. Many of my earlier observa- tions are rendered worthless for this purpose because I did not, at the time they were made, recognize the Shawmut rocks as | belonging to an age distinct from the Huronian. | DETAILS OF THE PETROSILEX. Petrosilex in Ne ewbury. — The Newbury belt of petrosilex extends along the valley of the River Parker from the mouth of this stream to Byfield Parish, a distance of five miles. The 56 general trend of the belt is E.N.E. Its width is very vari- able, a few rods to one and a half miles; this results mainly from the irregular manner in which the newer rocks — Shaw- mut and Primordial — are superimposed upon the petrosilex. Leaving these more recent rocks out of view, the petrosilex appears to be bounded on all sides by granite; lying between the Rowley granitic belt on the south and the Newbury belt on the north. This petrosilex is, on the whole, remarkable for its uniformity. The prevailing color is a deep red, or brownish- red; but it is sometimes purplish, pinkish, or grayish, and occasionally nearly white. It is never distinctly porphyritic, but nearly always has a well-marked banded structure. To the north of Old Town Hill and between that and Kent’s Island, however, much of the petrosilex is not banded. Its colors here are different shades of red and gray, varying from a dirty white to nearly black, but all weathering reddish. Some of it is dis- tinctly quartzose and has a granular or euritic aspect. The banded structure so characteristic of most of the Newbury petrosilex is produced by the interlamination of layers of quartzose and feldspathic materials. The quartzose or jas- pery layers are darker colored than the feldspathic, and the latter are prone to weather white ; so that the banding is most conspicuous on weathered surfaces. The thickness of the lami- nz usually varies from a fine line to one-sixteenth of an inch, and seldom exceeds one-eighth of an inch. The individual lami- nz are generally continuous for several feet or yards, except where broken by faults or other disturbance; and adjoining lamine, at least the larger ones, rarely coalesce or divide. As a rule, each layer maintains the same thickness through- out its length; the shorter ones, however, are sometimes lenticular, and these occasionally coalesce. The petrosilex has been immensely disturbed, and consequently the laminz strike and dip in every possible direction; yet there are a certain general strike and dip which predominate and are characteristic of the rock as a whole. ‘This prevailing strike ranges from N.E.-S.W. to E.-W., parallel with the general 57 trend of the belt; the corresponding dip is to the N.W. and N., and usually quite steep. This banding or lamination is frequently met with in the petrosilex and felsite of other regions, and is familiar to all geologists. In England and on the continent of Europe, where it is the universal belief that petrosilex and felsite are always and everywhere of igneous origin, where they are regarded as trappean rocks, and so classified by systematic lithologists, this banded structure is naturally and commonly interpreted as probably resulting, like the variously hued lines and bands in a slag from an iron furnace, from the motion of the mass when in a pasty and semifluid condition. In this country radically differ- ent views are gaining ground. Prof. Dana, in his ‘* Manual of Geology,” classes the petrosilicious group as metamorphic rocks; and Dr. Hunt has long taught their aqueous or sedi- mentary origin. This view is the only one in harmony with the facts observed in this region, and especially with the band- ing, or, as I prefer to call it, stratification, of the petrosilex. The bands in question are undoubtedly lines of bedding; and the lamination has substantially the same signification in petro- silex as in slate or sandstone. The petrology of the banded petrosilex at some points in Hastern Massachusetts yet to be described, absolutely forbids the acceptance of the motion-in-a- plastic-mass theory of the origin of the banded structure: The Newbury rock is, for the most part at least, a genuine petrosilex. A typical specimen of the banded variety from Kent’s Island contains 75.7 per cent. of silica. This indicates at least 25 per cent. of free quartz in the rock. Another speci- men from a point further west, near Dummer Academy, afforded 76.4 per cent. of silica. Concretionary structure. — South of the River Parker and near the road leading south from Old Town, at a place known as the Bartlett Mine, is a rock with a singular concretionary structure, which the miners call ‘‘toad stone.” It presents a spotted appearance, whence the name. This rock must, I think, be regarded as a variety of the petrosilex; in fact, a 58 part of this mass is a compact and, to the naked eye, structure- less petrosilex, of a greenish color. Chemical analysis, too, shows that the “toad stone” has the composition of a true petrosilex ; a shghtly weathered, but otherwise typical, specimen affording Miss E. M. Walton the following result : — SiO « 5 : : ; : ; : f 77.200 Al,O3; . ; : E 2 ‘ : 5 12.482 Fe,O; . ; i A j : , ; é 1.570 CaO ee jo3 : ; , 5 : 3 3 .800 KEOre : ; i # : 5 ; x 1.230 Na,O . ‘ : x , : : é 4.423 TBO) ‘ ; 5 ; : ‘ : z 2.004 MnO . : : : ! 3 ; : trace 99.709 As in the banded petrosilex, there is here a partial separation of the quartzose and feldspathic materials. Instead, however, of forming thin sheets or lamine, we have one of the con- stituents — the feldspar — segregated into small spheroidal, ellipsoidal, or almond-shaped masses; while the interstices between these are filled with the quartz, — not necessarily pure silica, but quartzose material,—-which is usually small in amount. The feldspathic aggregates sometimes appear of very irregular shapes and sizes; but usually they are regularly rounded, of uniform size in the same part of the rock, and have their longer axes parallel, giving the mass a faint appearance of stratification. At one point, too, they are arranged in lines, which are straight and parallel ; or, more exactly, the quartzose material is in thin layers, and each attenuated stratum has a layer of feldspathic nodules on either side of it. The nodules, thus arranged, sometimes coalesce, forming continuous lamin, and reproducing the banded structure. The strike and dip, as indicated by the parallel arrangement of the nodules, appears to be the same in all parts of the rock, and agrees with that recorded above for the petrosilex generally. The almond- shaped masses vary in size from mere specks to one-fourth 59 of an inch in diameter. ‘Their structure is concentric as the rule, and radiate as the rare exception. Each aggregate (except ~ the smallest, which are usually homogeneous to the eye) consists of a layer of white material enclosed concentrically between a dark-colored centre or nucleus and a thin layer of greenish- white, which comes between the white zone and the external quartzose material. In the unweathered state these different layers have a uniform hardness about equal to orthoclase ; but, when weathered, the dark nucleus becomes of a dull-green color and yields readily to the knife ; and both it and the white band yield the more readily the more the rock is weathered. Where the almond masses are large it can be plainly seen (and proba- bly all would show it under the microscope) that in each of them the general structure of the whole rock is reproduced on a smaller scale, each nodule containing a number of smaller masses of the same form and structure as itself. These smaller or secondary aggregates also have their major axes parallel. They are most conspicuous in the green nucleus. This rock. is susceptible of a tolerable polish. The question naturally arises as to whether the structure of the ‘toad stone” is original and general, or due to local alteration. I incline to the former view, although evidence supporting the latter is not wanting. This rock forms the south side and summit of a small hill rising from the tide-water marshes of Mud Creek, and is met on the north by a*petrosilex breccia. The junction of the two rocks is marked by a trap dyke, several feet wide, which dips steeply to the north. Now it is in the immediate vicinity of this dyke that the rock in question possesses the banded structure described on the preceding page ; beyond six inches or a foot from the eruptive mass the banding appears to be wholly wanting. ‘The bands are exactly parallel with the dyke. These facts might be considered as proving a causal connection between the dyke and the pseud-odlitic struc- ture of the adjoining rock. But I think this is illusory ; for in other parts of the petrosilex belt, on Kent’s Island, and to the westward, there is evidence showing that this peculiar 60 concretionary structure is really wide-spread, and is probably in every case an original structural feature of the petrosilex. Extended observation has convinced me that this structure, more or less perfectly developed, characterizes most of the petrosilex of this belt, especially the banded variety. It can usually be detected only on weathered surfaces, but is some- times so pronounced as to be observable in the mass of the rock. The concretions are all small, and, as in the ‘toad stone,” are confined to the feldspathic material. The nucleus can usually be seen, indicating the concentric structure; but I have failed to detect the greenish layer. The resemblance, though not complete, is probably sufficient to establish the essential identity of the two rocks. A small mass (probably an erratic) of qenoeiles breccia, found on Marblehead Neck by Prof. Hyatt, contains pebbles of a banded petrosilex similar to that in Newbury, which are characterized by a nodular structure, differmg somewhat, but probably not essentially, from that described above. In the banded petrosilex of Marblehead Neck, also, I have ob- served abundant traces of a similar structure. On the north- ern end of the Neck, near the light-house, the stratified petro- silex contains numerous intercalated layers, varying in thickness from an inch or less to a foot, which are largely, and, in some cases wholly, composed of concretions or nodules. The con- cretions are of all sizes, from those requiring a magnifier to make them visible, to two inches in diameter. The smaller concretions are all spherical, but those exceeding half an inch in diameter are more or less flattened on the under side, and the largest are disc-shaped. There is little appearance of separa- tion according to sizes, the large and small being usually inter- mingled. When crowded the larger nodules give the upper surface of the layer a botryoidal aspect. A concentric structure is sometimes observable in the concretions; but usually they appear quite structureless, save in this: the regular banding or stratification of the petrosilex is continuous through the nodules. The banding, both in the nodules and outside of them, is quite 61 independent of the forms of the nodules, showing no tendency to conform with their outlines. In short, these concretions are ~ essentially identical with those occurring in the most recent de- posits of sand or clay. ‘The concretions weather less readily than the matrix, and are usually quite inconspicuous on un- weathered surfaces. The larger nodules, however, are some- times partially decomposed exteriorly; and in these may be discerned traces of a secondary concretionary structure, or nod- ules within nodules. In this locality (near the light-house) the banding of portions of the petrosilex is unusually fine and even, yet much contorted; and some of the more feldspathic layers have a finely granular or brecciated appearance, precisely as if the layers had been crushed 7m s¢tw and reconsolidated. The minute fragments appear at first sight to be quite sharp and an- gular, —and so they are, as regards the fracture planes pro- ‘ducing them; but close observation shows that most of the grains are more or less rounded and concentric. In fact, they are concretions, and, like those just described, they weather less readily than the matrix. The concretionary structure appears to be wholly wanting in none of the banded petrosilex on Marblehead. Neck, and it may be occasionally detected in other varieties. In the petrosilex of West Dedham, west of Woodland Street, Mr. F. W. Very has observed a band or bed about three feet wide, strike N. 30° E., in which the spheru- litic structure is well developed. The rock shows traces of stratification, and consists of a greenish base enclosing occasional erystals of reddish-brown feldspar, and numerous spherical con- eretions of a dark-purple color, about a line in diameter, and without apparent structure. The spherules are more quartzose than the matrix, appearing in relief on weathered surfaces. The base itself is also concretionary, resembling the ‘‘ toad stone.” There is no visible quartz in this rock, although grains of this mineral occur in the bright-green petrosilex by which the stratum is bordered. On the line between Hyde Park and West Roxbury, near Muddy Pond, Mr. Very has found a dirty, greenish-white, stratified petrosilex with scattering spherical con- 62 cretions, the largest of which are about one-fifth of an inch in diameter. They have a radiate structure, and appear to consist of crystalline quartz. The greenish tinge of this rock is due to a soft, brownish-green mineral resembling serpentine, which forms layers in the rock. It is, however, probably aluminous rather than magnesian. The same observer also reports the occurrence of concretions in an otherwise structureless petro- silex in Hyde Park Village. The petrosilex ridge between Pine Garden and Calvary Cemetery, in Hyde Park, — especially towards the southern end, — affords the best example which I have yet observed of the concretionary structure in the petrosilex. This is a white, banded variety of petrosilex, and the concre- tions are usually spherical, varying in size from a pin-head to five inches in diameter. Occasionally they are so numerous that the rock is almost entirely composed of them, and the weathered surface has the aspect of a pudding-stone. The nod- ules are commonly purplish, weathering white, and showing but faint indications of concentricity. In some limited masses of the petrosilex of a bright-red color, the concretions are small, white, and distinctly stellate. On the east side of the ridge, toward Back Street, the spherical concretions are mostly want- ing, and are replaced by small, purple layers, which, though of very irregular shapes and sizes, always present botryoidal sur- faces and a concentric structure. ‘They coincide with the band- ing of the petrosilex; are rarely more than an inch thick, but often two feet or more in length. Abundant concretions have been observed in a portion of the petrosilex forming the rounded north-east corner of the Blue Hill area of petrosilex. They are mostly small, generally spherical and radiate, but some- times elongated and concentric. Small fragments of a feld- spathic rock containing concretions have been found near the trilobite quarry in Braintree. This rock has not been ob- served in situ. It holds scattering feldspar crystals, which are much fewer and several times larger than the rounded nodules ; each crystal, however, is enveloped in a whitish layer, giving it a concentric appearance. ‘The foregoing are all the instances 63 that have come under my notice where the concretionary or pearlstone structure in the petrosilex is well marked ; but indi- -eations are not wanting that this structure, or a tendency to- ward it, is very general in this rock. Petrosilex on Marblehead Neck and the adjacent islands. —There are several distinct varieties of petrosilex on Marble- head Neck. The most conspicuous and beautiful of these is the banded variety already alluded to. This has substantially the same characters asin Newbury. The bands, however, are gener- ally broader, — one-sixteenth to one-eighth, and sometimes one- fourth of an inch. The feldspathic layers are rarely reddish, usually gray, bluish, or purplish, but always weathering whitish. The banding is often very indistinct on a fresh surface. The bands are, perhaps, less regular and continuous than in the Newbury rock, coalescing and dividing more frequently. They exhibit greater disturbance, and. at many points are locally resolved into a condition approaching the chaotic, portions of the rock appearing to have been crushed and partially fused; yet, on the whole, a north-easterly strike appears to prevail, and this is no less clearly a stratified rock. Crystals of feldspar are common in this rock, but are rarely numerous enough to give it a porphyritic aspect. This variety of petrosilex is well developed on the north-west side of the Neck, and in the vicinity of the light-house at the northern end. Several small patches also occur on the east- ern shore near Castle Rock, and to the northward.