SEP 4 fg5j X 6 - ,oax New York State Museum Bulletin Published by The University of the State of New York m NE' e: No. 296 ALBANY, N. Y. December 1934 NEW YORK STATE MUSEUM Charles C. Adams, Director GEOLOGY AND MINERAL RESOURCES OF THE HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES By A. F. Buddington Temporary Geologist, New York State Museum WITH A CHAPTER ON THE PALEOZOIC ROCKS OF THE LOWVILLE QUADRANGLE By Rudolf Ruedemann State Paleontologist, New York State Museum CONTENTS PAGE Introduction 7 Summary of geology 9 Physiography 20 Physiographic provinces 20 Topographic types 23 Pleistocene temporary lakes and high level deltas 30 Recessional and ground moraines. 39 Lakes 46 Direction of movement of Pleisto- cene glacier 47 Recent and preglacial drainage. ... 51 Postglacial upwarp 54 Igneous rocks 56 Metamorphic and mixed rocks. ... 106 Structural geology 138 Historical introduction. 138 Structural geology of Grenville belt 142 Structure of the main igneous complex 162 Structure of Paleozoic beds 172 PAGE Pre-Potsdam weathering of Pre- cambrian rocks 173 Paleozoic formations (Hammond and Antwerp quadrangles) 177 Paleozoic rocks of the Lowville quadrangle (By Rudolf Ruede- mann) 183 Economic geology 194 Hematite deposits 194 Galena veins 202 Pyrite and pyrrhotite deposits. . .209 Gold sands 214 Graphite 217 Crystalline limestone, dolomite and marble 218 Sandstone 220 Limestone 220 Gravel and sand 220 Mineral specimen localities 221 Bibliography 223 Index 247 ALBANY THE UNIVERSITY OF THE STATE OF NEW YORK 1934 M296r-Je32-2ooo THE UNIVERSITY OF THE STATE OF NEW YORK Regents of the University With years when terms expire 1944 James Byrne B.A., LL.B., LL.D., Chancellor New York 1943 Thomas J. Mangan M.A., LL.D., Vice Chancellor Binghamton 1945 William J. Wallin M.A., LL.D. ----- Yonkers 1935 William Bondy M.A., LL.B., Ph.D., D.C.L. - New York 1941 Robert W. Higbie M.A., LL.D. ----- Jamaica 1938 Roland B. Woodward M.A., LL.D. - - - - Rochester 1937 Mrs Herbert Lee Pratt L.H.D. ----- New York 1939 Wm Leland Thompson B.A., LL.D.- - - - Troy 1936 John Lord O’Brian B.A., LL.B., LL.D. - - Buffalo 1940 Grant C. Madill M.D., LL.D. ----- Ogdensburg 1942 George Hopkins Bond Ph.M., LL.B., LL.D. - Syracuse 1946 Owen D. Young B.A., LL.B., D.C.S., LL.D. - New York President of the University and Commissioner of Education Frank P. Graves Ph.D., Litt.D., L.H.D. , LL.D. Deputy Commissioner and Counsel Ernest E. Cole LL.B., Pd.D., LL.D. Assistant Commissioner for Higher Education Harlan H. Horner M.A., Pd.D., LL.D. Assistant Commissioner for Secondary Education George M. Wiley M.A., Pd.D., L.H.D., LL.D. Assistant Commissioner for Elementary Education J. Cayce Morrison M.A., Ph.D., LL.D. Assistant Commissioner for Vocational and Extension Education Lewis A. Wilson D.Sc., LL.D. Assistant Commissioner for Finance Alfred D. Simpson M.A., Ph.D. Assistant Commissioner for Administration Lloyd L. Cheney B.A., Pd.D. Assistant Commissioner for Teacher Education and Certification Hermann Cooper M.A., Ph.D. Director of State Library James I. Wyer M.L.S., Pd.D. Director of Science and State Museum Charles C. Adams M.S., Ph.D., D.Sc. Directors of Divisions Archives and History, Alexander C. Flick M.A., Litt.D., Ph.D.. LL.D. Attendance and Child Accounting, Charles L. Mosher Ph.M. Educational Research, Warren W. Coxe B.S., Ph.D. Examinations and Inspections, Avery W. Skinner B.A., Pd.D. Health and Physical Education, Law, Charles A. Brind jr B.A., LL.B. Library Extension, Frank L. Tolman Ph.B., Pd.D. Motion Picture, Irwin Esmond Ph.B., LL.B. Professional Licensure, Charles B. Heisler B.A. Rehabilitation, Riley M. Little B.S., B.D. Rural Education, Ray P. Snyder School Buildings and Grounds, Joseph FI. Hixson M.A. Visual Instruction, New Y ork State Museum Bulletin Published by The University of the State of New York No. 296 ALBANY, N. Y. December 1934 NEW YORK STATE MUSEUM Charles C. Adams, Director GEOLOGY AND MINERAL RESOURCES OF THE HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES By A. F. Buddington Temporary Geologist, New York State Museum WITH A CHAPTER ON THE PALEOZOIC ROCKS OF THE LOWVILLE QUADRANGLE By Rudolf Ruedemann State Paleontologist, New York State Museum CONTENTS PAGE Introduction 7 Summary of geology 9 Physiography 20 Physiographic provinces 20 Topographic types 23 Pleistocene temporary lakes and high level deltas 30 Recessional and ground moraines. 39 Lakes 46 Direction of movement of Pleisto- cene glacier 47 Recent and preglacial drainage. ... 51 Postglacial upwarp 54 Igneous rocks 56 Metamorphic and mixed rocks. ... 106 Structural geology 138 Historical introduction 138 Structural geology of Grenville belt 142 Structure of the main igneous complex 162 Structure of Paleozoic beds 172 PAGE Pre-Potsdam weathering of Pre- cambrian rocks 173 Paleozoic formations (Hammond and Antwerp quadrangles) 177 Paleozoic rocks of the Lowville quadrangle (By Rudolf Ruede- mann) 183 Economic geology 194 Hematite deposits 194 Galena veins 202 Pyrite and pyrrhotite deposits . . . 209 Gold sands 214 Graphite 217 Crystalline limestone, dolomite and marble 218 Sandstone 220 Limestone 220 Gravel and sand 220 Mineral specimen localities 221 Bibliography 223 Index 247 ALBANY THE UNIVERSITY OF THE STATE OF NEW YORK 1934 Digitized by the Internet Archive in 2017 with funding from I MLS. LG-70-15-0138-15 https://archive.org/details/newyorkstatemuse2961newy ILLUSTRATIONS All figures reproduced from drawings and photographs by A. F. Buddington, except figure 17- page Figure I Index map, showing location of quadrangles ... .. 8 Figure 2 Diagrammatic section to show relations of Adirondack dome of Precambrian igneous and metamorphic rocks to the over- lying Paleozoic beds and supposed peneplains 19 Figure 3 Block diagram showing assumed conditions at one stage in the history of Pleistocene glaciation in the Black River valley. . 32 Figure 4 Section along line from just south of Lowville through Crystal Lake School, showing relations of overlying Pleistocene deposits to bed rock 41 Figure 5 Map showing recessional moraines and direction of ice move- ment. Modified from map by Frank B. Taylor, 1924 48 Figure 6 Chatter marks on Potsdam sandstone, on road running west from schoolhouse about two and one-half miles south-southwest of Hammond 49 Figure 7 Hornblende syenite cutting across foliation of granite and terminating in a line of fragments 103 Figure 8 Dike of hornblende syenite intruding granite, and itself brecciated and veined by granite 103 Figure 9 Generalized trend of foliation in the northwest Adirondacks. ... 143 Figure 10 Location of identified axes of folds in Hammond quadrangle. . . 145 Figure 11 Lewisburg syncline, Antwerp and Lake Bonaparte quadrangles 146 Figure 12 Generalized structure section across Sherman Lake syncline and Somerville anticline, Hammond quadrangle 147 Figure 13 Generalized structure section of Rossie isoclinally folded com- plex, Hammond quadrangle 150 Figure 14 Inferred structure section across California phacolith, Lake Bonaparte quadrangle (northeastward extension of Dority Pond granite mass, Antwerp Quadrangle) 158 Figure 15 Generalized structure section across Croghan intrusive complex, from Croghan to southeast corner of Lowville quadrangle. ... 165 Figure 16 Drag folded hyperite sill in granosyenite within mylonite band, one and one-fifth miles southeast of Fargo, Antwerp quadrangle 168 Figure 17 Sketch map showing the tangential master streams of the Adirondack massif and the escarpments that control their course. Drawing by Rudolf Ruedemann 185 Figure 18 Generalized structure section showing relation of iron ore and pyrite deposits to the country rock, Keene-Antwerp belt. 201 Figure 19 Cliffs of Potsdam sandstone, northwest of Chapel Corners, Hammond quadrangle 229 Figure 20 Quarry at north edge of town of Lowville. Pamelia limestone at base, Lowville limestone at top 229 Figure 21 Typical scarp formed by Theresa sandy dolomite. Edge of table-land two and one-half miles west of Natural Bridge, Antwerp quadrangle 230 Figure 22 Table-land on Potsdam sandstone, looking north-northwest from drumlin one-half mile east of Sterlingville, Antwerp quadrangle 230 Figure 23 Rolling hill or knob country underlain by granosyenite, south- west of Mount Tom, Carthage quadrangle 231 Figure 24 Low rolling limestone country with lineal ridges and depressions. Northeast of South Woods School, Hammond quadrangle. ... 231 Figure 25 Topography developed parallel to curving beds at end of synclinal structure. Hill at right is granite, flat silt-floored valley is in quartzitic and calcareous beds, and curving slope at left is in limestone. Syncline one mile northwest of Brasie Comers, Hammond quadrangle 322 4 N EW YORK STATE MUSEUM Figure 26 Figure 27 Figure 28 Figure 29 Figure 30 Figure 31 Figure 32 Figure 33 Figure 34 Figure 35 Figure 36 Figure 37 Figure 38 Figure 39 Figure 40 Figure 41 Figure 42 Figure 43 Figure 44 Figure 45 Figure 46 Figure 47 Figure 48 Figure 49 Figure 50 PAGE Gabbro hill surrounded by limestone lowland, west side of Beaver creek, south of state road, Hammond quadrangle .... Surface of peneplain of Precambrian age on granite and gneiss, one mile east of North Wilna, Antwerp quadrangle Surface of peneplain on quartz diorite, slightly dissected by erosion, three and one-quarter miles east of Antwerp Precambrian peneplain on limestone, three miles northeast of Somerville, at South Gouvemeur. Slight bedrock irregu- larities are obscured by silt deposits. Hammond quadrangle Precambrian peneplain on granite much dissected by post-Paleo- zoic erosion. Village of Rossie on Indian river, Hammond quadrangle Surface of Pine Plains delta (built in glacial Lake Iroquois) .... Silt plain (built in glacial Lake Iroquois) two and one-half miles south of Philadelphia, Antwerp quadrangle Croghan recessional moraine, one mile south of Croghan, Low- ville quadrangle Croghan west recessional moraine, one mile southwest of Croghan, Lowville quadrangle Devoice moraine, southwest of Natural Bridge, Antwerp quadrangle Drumlin and bowldery ground moraine, one-half mile east of Sterlingville, Antwerp quadrangle Pleasant Lake, at north end looking south; hill at left is gabbro, lake basin is in limestone; Hammond quadrangle Lake of the Woods; walls are of Potsdam sandstone; Hammond quadrangle Seven-foot boulder of anorthosite resting on Potsdam sandstone, one-half mile northwest of village of Black Lake, Hammond quadrangle. Boulder probably was carried by ice 100 miles or more from its source Postglacial gorge of Black river in Lowville limestone, just below Great Bend, Antwerp quadrangle Injection gneiss or arteritic migmatite, plicated. North end of Lewisburg syncline Pyroxenic gneiss brecciated by granite, an agmatite, Hammond quadrangle. Photograph of specimen on exhibition in New York State Museum, Albany. Length of slab is six feet Interlayered limestone and pyroxene-scapolite-feldspar granu- lite crumpled and crenulated with steep pitch. Three- quarters of a mile northwest of Scotch Settlement School, Hammond quadrangle Granite with included layer of folded amphibolite. Hyde School, Hammond quadrangle Pegmatite vein broken and displaced by flowage of limestone, one mile northwest of Elmdale, Hammond quadrangle Photomicrograph of hyperite, showing texture characteristic of normal crystallization from a magma. Two miles northwest of Indian River, Lowville quadrangle Photomicrograph of hyperite, partly mashed ; from belt showing cataclastic structure Photomicrograph of hyperite, from mylonite band ; one mile south- east of Fargo, Antwerp quadrangle Photomicrograph of granosyenite, showing texture characteristic of normal crystallization from a magma, and no evidences of crushing. One-half mile east of Bushes Landing, Lowville quadrangle Photomicrograph of granosyenite showing protoclastic structure; feldspars granulated, quartz in massive leaves. One and one- half miles north of Croghan 232 233 233 234 234 235 235 236 236 237 237 238 238 239 239 240 240 241 241 24,2 242 243 243 244 244 ILLUSTKATIONS 5 Figure 51 Figure 52 Figure 53 Figure 54 Map 1 Map 2 Map 3 Map 4 PAGE Photomicrograph of granosyenite showing cataclastic structure, one-half mile northeast of Texas Road School, Lowville quadrangle 245 Photomicrograph of granosyenite from mylonite band, Mount McQuillen, Carthage quadrangle 245 Schoolhouse built in 1817 of Potsdam sandstone, two and one- half miles southwest of Hammond 246 Stratified sand and gravel in kame south of Strickland Comers, Antwerp quadrangle 246 Geologic map of Hammond quadrangle ( In pocket at end) Geologic map of Antwerp quadrangle (In pocket at end) Geologic map of Lowville quadrangle (In pocket at end ) Geologic map of the Precambrian area of the Carthage quadrangle (In pocket at end ) , . GEOLOGY AND MINERAL RESOURCES OF THE HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES By A. F. Buddington Temporary Geologist, New York State Museum WITH A CHAPTER ON THE PALEOZOIC ROCKS OF THE LOWVILLE QUADRANGLE By Rudolf Ruedemann State Paleontologist, New York State Museum INTRODUCTION The areas of the Hammond, Antwerp and Lowville quadrangles' lie along the western and northwestern border of the Adirondack mountains, predominantly within the foothill or lowland belt. The adjoining quadrangles of the Adirondack province have all been mapped. For this reason and because the three quadrangles form a practical unit for discussion, their description is here combined in one report. The northeastern corner of the Carthage quadrangle is also included, as it is underlaid by Precambrian rocks of the Adirondack province. It is believed that practically all types of geology char- acteristic of the western and northwestern Adirondacks are found within these areas. Each quadrangle covers about 214 square miles, and the whole area comprises over 700 square miles. The location of the areas described in this report are shown in Figure 1. The author spent a total of about 37 weeks in geologic mapping of this area during the summers of 1917, 1920, 1927, 1928 and 1929, doing successively the Lowville, Hammond, Antwerp and a part of the Carthage quadrangles, and he is responsible for this bulletin with the exception of the chapter on the Paleozoic rocks of the southwest corner of the Lowville sheet, which were mapped and studied in detail by Dr Rudolf Ruedemann, whose report on them is incorporated as a separate chapter. The writer is indebted to C. H. Smyth jr for critical reading of the manuscript of this report, to N. C. Dale for a sketch of the trend lines of the foliation on the Oswegatchie quadrangle, to W. J. \7] 8 NEW YORK STATE MUSEUM Figure I Index map showing location of quadrangles HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 9 Miller for permission to use the strikes and dips given by him on the map of the Russell quadrangle, and to A. J. Eardley, who drew the block diagram, figure 3. In 1920 Dr W. M. Agar was with the author for a few weeks while making an intensive study of cer- tain mineral localities, and the writer has profited from his results. The areas along the west and northwest border of the Adirondacks are much more accessible and are cleared of forest to a far greater degree than the inner higher parts constituting the core. For this reason the complex geology may be seen much more clearly and may be worked out in greater detail than in the adjoining heavily forested country, where so much of the bedrock is buried beneath a cover of forest litter. Probably a larger percentage of land is cleared and utilized for agricultural purposes in the Hammond quadrangle than in any other equal area of the northwest Adirondacks. For this reason and because of the intricacy of the geology, some 17 weeks or nearly half the total time was spent on this area alone. This region was formerly heavily forested but has been lumbered off and now only a minor part is in woodland and most of this is second growth. Dairying constitutes the chief industry and most of the cleared land is used for pasture. Hay and oats are the principal crops. The falls of the Beaver and Black rivers have been utilized to a large extent for the development of power, and large paper mills are located at Beaver Falls, Carthage, Philadelphia, Great Bend and Deferiet. Some lumbering is still carried on along the extreme eastern border of the Fowville quadrangle. Abandoned farmhouses and remnants of foundations testify to the far greater number of farms formerly existing in the region. On the Antwerp quadrangle all the farmhouses along certain short stretches of connecting road have been abandoned and the roads closed. In many cases there has been consolidation of farms and the best fields are still worked and the best pastures still used, but in other cases the old pastures have been permitted to grow up to underbrush. Most of the territory is served by good roads. The entire area of Pine Plains in the southern part of the Antwerp area is a United States Government Reservation used for training purposes by the Army. SUMMARY OF GEOLOGY INTRODUCTION The bedrock underlying this region comprises all of the three major classes of rocks that form the earth’s crust— sedimentary, metamorphic and igneous. The metamorphic rocks, known as the 10 NEW YORK STATE MUSEUM Grenville series, are the oldest and include quartzite, gneiss, schist and crystalline limestone and dolomite. The igneous rocks are those that have consolidated and crystallized from a liquid solution, com- posed predominantly of molten silicates. Such solutions of molten silicates with a small percentage of water and other volatile com- pounds and perhaps with suspended crystals arising from partial crystallization, are called magmas. The magmas that were succes- sively intruded in this region differed in composition, and the varieties of rocks that they formed comprise: gabbro diorite, quartz diorite syenite, quartz syenite or granosyenite and granite. These rocks are younger than the metamorphic Grenville group for they have crystallized from molten solutions that were forced up into the Grenville rocks from deep within the earth. The sedimentary rocks are younger than the preceding two groups and include the Potsdam sandstone; Theresa sandy dolomite; the Pamelia, Lowville, Water- town and Leray limestones; and the sands, silts and clays of the valley floors and glacial lake deltas. The Grenville metamorphic series and the igneous rocks are of Precambrian age, the sandstone and limestone beds are of Paleozoic age and the sands and silts of the valley floors are of Pleistocene and Recent age. The metamorphic rocks of the Grenville series have been formed by recrystallization from sandstones, limestones and shales ; the quartzite being derived from sandstone, the crystalline limestone and marble from limestone, the gneiss from quartzite and shale and the schist from shale. Since the metamorphic rocks are derived from sediments and since sediments must have been derived from erosion of preexisting older rocks, the problem arises as to where the older rocks are from which the original sedimentary rocks (now changed to the Grenville metamorphic group) were derived. The answer to this problem is unknown. The oldest rocks exposed are the Gren- ville beds themselves. It seems very probable, however, that if we could go into the earth somewhere beneath the Grenville series, still older rocks would be found. The Grenville beds, however, may be of the order of age of a billion years. GRENVILLE SERIES The Grenville series of rocks is named after the township of Gren- ville in Ontario where they are well exposed. The metamorphic rocks of the Adirondack region are also called the Grenville series because they are believed to be of the same general age and represent the same group of rocks as those exposed in the area where they were originally named and described. The Grenville beds before meta- HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES II morphism consisted of sandstone, limestone and shale to a thickness of perhaps the order of magnitude of 15,000 to 20,000 feet in this region. The change of these original rocks to quartzite, crystalline limestone or marble and gneiss and schist has been brought about by a combination of strong mountain-making stresses acting along northwest-southeast lines and by effects concomitant with and attend- ant upon the intrusion of large masses of molten rock squeezed up into the beds from a source many miles deep within the crust of the earth. During the latter part of the period of intrusion of these molten masses or magmas, the earth stresses folded the sedimentary beds and some of the already intruded masses of magma that may have been wholly or only partly consolidated. One result of this folding is well shown in ground plan by the festooned shape of the granite and quartz diorite sheets northeast of Rossie. Another result is that the originally essentially horizontal sedimentary beds now in their metamorphic condition stand on end over much of the area. The intruded molten magmas had many volatile compounds in solution, of which water was by far predominant. As the magmas cooled and crystallized they gave off highly heated vapors and solu- tions that permeated widely through the bordering sedimentary rocks and together with the effect of the stresses that folded the rocks, effected a complete transformation and recrystallization of the sedimentary rocks into their metamorphic equivalents that constitute the Grenville series. The solutions and vapors given off by the mag- mas in many cases carried some silicates and other relatively insol- uble materials in solution and deposited them in the country rocks, thus materially adding to their changed and altered character. The last portions of a magma to crystallize are usually much richer in water and other volatile compounds than the original magma and therefore have a much greater fluidity. Such residual, more liquid, mobile parts were squeezed into the surrounding Grenville rocks where they form the almost omnipresent granite pegmatite veins or lenses. IGNEOUS ROCKS AND THEIR MODE OP INTRUSION Within the areas underlaid by Precambrian rocks, the percentage of those of igneous origin increases toward the southeast. In the Hammond quadrangle one-third of the area is underlaid by igneous rocks; of the Precambrian rocks on the Antwerp sheet two-thirds are of igneous origin; and in the Lowville sheet more than nine- tenths of the Precambrian rocks are igneous. 12 NEW YORK STATE MUSEUM There are at least two series of intrusive rocks; the oldest includes gabbro, diorite, and quartz diorites, and the younger groups include syenite, quartz syenite, and granite. All these rocks are cut by coarse-grained veins of granitic composition called pegmatite. All the magmas whose crystallization and consolidation formed the igneous rocks were perhaps intruded during one epoch, but if so, this period must have been very long and the intrusions at successive intervals. The magmas in the Antwerp and Hammond areas were for the most part forced in along the bedding planes of the Gren- ville sediments, spreading them apart, disrupting and breaking the more brittle members, stretching the more plastic layers, locally shredding and disintegrating the Grenville beds, and in part reacting with, dissolving and replacing them. Intrusive bodies that are thus for the most part conformable with the structural planes of the Grenville beds may be called sills or sheets. The granite (Alexandria type) magma is here interpreted as having come into the Grenville beds when the latter were already partly folded. It rose beneath the arches of the folds and formed lens-shaped masses concave down- wards, called phacoliths. Both the sheets and phacoliths were affected subsequent to their emplacement by intense stresses along northwest-southeast lines A great complex of igneous rock, varying from syenite to granite, forms the bulk of the Adirondack mountain area, exclusive of the anorthosite core in the eastern part. This main igneous complex forms the bedrock of the Precambrian area of the Lowville quadrangle and the southeastern part of the Antwerp sheet. Only relics of the Grenville series are found within this igneous complex, and the problem arises as to whether so great a body of igneous rock originated through intrusion in the same fashion but on a much grander scale than that exemplified in the belt where the Grenville formations predominate, or whether it has had quite a different mode of emplacement. Augen gneisses and porphyritic rocks form a border belt to the main igneous complex in the southeast part of the Antwerp area and the northwest part of the Lowville quadrangle, and extend northeast across the Lake Bonaparte quadrangle, where the body of which they form a part has been previously called by the writer the Diana complex. This complex has a stratiform banded structure and is tentatively interpreted as part of an overturned close folded sheet. The rocks, previously called the Croghan complex, underlying the southeast part of the Lowville quadrangle, however, appear to be part of a much larger unit which extends to the south HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 13 and east and which, so far as its size is concerned, might be termed a batholith. Discussion of the mechanism of its emplacement must await further knowledge. The igneous masses, in part as a result of their efforts to shoulder aside the beds of the Grenville series as they were squeezed in along the bedding planes, and in part as a result of the action of stresses acting along northwest-southeast lines during or following the period of their intrusion, acquired a structure resulting in a tendency to part parallel to certain planes. This structure we call foliation, and it is often very well developed in the igneous rocks. MOUNTAIN BUILDING IN PRECAMBRIAN The whole complex of metamorphic and igneous rocks, with the possible exception of the Croghan complex, has been subjected to intense earth stresses that folded it much in the fashion that a pile of paper is crumpled when pressed from each end. The beds and igneous sheets now stand nearly vertical in many cases and in others the folds were pushed beyond the vertical and overturned toward the southeast. At some time after the conclusion of this epoch of disturbance, this region was probably the site of a great mountain range with a northeast-southwest trend. At an undetermined date, perhaps long after the epoch of mag- matic intrusion and folding, a few fissures were opened up, through which came a basic magma that consolidated to form black diabase dikes. These preceded the deposition of the gravel and sand that went to form the Potsdam formation. They cross the folding and are the youngest intrusive rocks in the area. They are too small to show on the map and were seen only in the northwest part of the Hammond quadrangle. PERIOD OF PRE-POTSDAM PENEPLANATION Following the building of the mountain range, a long period of weathering and erosion set in. The mountains were lowered and leveled to a gently rolling country, essentially the same as in the foot- hill belt of the Antwerp and Hammond quadrangles today, except that the surface was covered with a deeply weathered, thoroughly leached residual soil instead of with a veneer of glacial drift as now. This surface is called the “Precambrian peneplain,” for we know it was formed in Precambrian time since it extends beneath the oldest Paleozoic beds. 14 NEW YORK STATE MUSEUM PALEOZOIC DEPOSITS Introduction. Following the formation of the Precambrian peneplain a series of sedimentary deposits were laid down upon its surface within this region. At first a downwarp appears to have taken place along the general location of the present St Lawrence valley and in this trough were deposited the sands and sandy dolo- mitic muds that now make up the Potsdam sandstone and the Theresa sandy dolomite respectively. The exact conditions under which the lower part of the Potsdam sandstone was deposited are not known but the upper part and the Theresa formation were formed in marine waters, for they carry marine fossils. At some time after the deposition of the Potsdam and Theresa formations there was a succession of gentle warpings of the north- west Adirondack region that many times brought this whole region beneath the ocean for shorter or longer periods. At such times cal- careous muds were deposited, which on subsequent compaction to solid rock came to form a series of limestone formations. These limestones, so far as shown within the area mapped, embraced the Pamelia, Lowville, Leray, Watertown, Trenton and Cobourg lime- stones. They form the steep slopes and benches of the Tug Hill plateau south and west of the Black river on the Lowville, Carthage and Antwerp quadrangles. Potsdam and Theresa formations. These beds are relatively flat-lying and formerly covered the whole area of the Hammond and Antwerp quadrangles as one continuous blanket. They rest uncon- formably on the underlying Precambrian rocks, and conglomerate lenses in the Potsdam beds occasionally contain pebbles derived from the immediately underlying rocks, although this is not common. Practically all the residual soil was washed away before the deposition of the Potsdam sands but beneath the loose soil, except in the lime- stone areas, there was a deep zone of weathered and altered rock, which was not everywhere removed but is locally still preserved at many places beneath the cap of Potsdam sandstone, monuments to the fact that weathering processes went on in much the same fashion over a half billion years ago as they do today. The Potsdam sandstone is so named from the town of Potsdam, New York, where it is well exposed. The sandstone beds on the west side of Black Lake valley have been traced continuously north- east to Potsdam and beyond. They form a fringe around the bor- ders of the whole northern and northwestern Adirondack area. The sandstone areas are usually terminated by sharp cliffs yielding good exposures of the beds (figure 19). The Potsdam sandstone con- HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 15 sists of layers or beds of sand that have been cemented by silica through a deposit of secondary quartz around each of the original grains. Ground water derived from the rainfall dissolves a little silica at the surface of the earth and while seeping through the porous sand beds deposits the silica until the sand beds are cemented to a coherent consolidated rock. Few fossils have been found in the Potsdam sandstone in this area but they have been found elsewhere in this formation and indicate that the sands were laid down in Upper Cambrian (Ozarkian) time, hundreds of million years, perhaps a half-billion years ago. The lower beds are thought to have been formed as river sands washed down from hills to the southeast and spread out over a lowland belt where the wind might aid in working them over and rounding "the grains. The finer more clayey material was swept out to more distant points. The upper beds of the Pots- dam formation, however, grade into the known marine fossiliferous Theresa beds and are thought to have been laid down on the floor of the ocean, which at that ancient time lapped around and gradually encroached west upon the borders of the northwestern Adirondack area of crystalline rocks. The Theresa beds must have been laid down on the floor of a shallow sea for they carry marine fossils. At the time of their deposition not only was the sand carried in by the rivers spread over the ocean bottom, but calcareous muds were pre- cipitated, probably in part by organisms and accumulated with the sand to form calcareous or dolomitic sandstones Carbonates are easily recrystallized and rearranged by ground water solutions and the Theresa beds thus became coherent and solid rock subsequent to their deposition as loose sandy carbonate muds. To the westward there are a series of sedimentary beds younger than the Theresa beds but still belonging to the Paleozoic system. These may, in part, have originally been present over the Hammond and Antwerp areas, but if so, they have since been wholly eroded. The variation in thickness of the Potsdam sandstones whereby they thin towards the west and south and thicken toward the east has led to the belief (Cushing To, p. 15) that the deposits of sand began forming first in the Champlain region and gradually spread westward, being deposited in a broad shallow depression, trough or basin whose axis roughly coincided with the modern St Lawrence axis, so that hereabouts we find simply a thinned part of the forma- tion near the limit of its western extent. Its thickness is about equal to the difference in altitude between the ridge crests and valley bot- toms of the pre-Potsdam topography, upon which it was deposited, and hence varies rapidly in thickness from place to place and was 1 6 NEW YORK STATE MUSEUM but scantily deposited upon the elevations. The Theresa beds were laid down in a similar trough. Pamelia limestone. Along the extreme southern border of the Antwerp quadrangle and in the Carthage and Lowville areas the Potsdam and Theresa formations are absent and a formation known as the Pamelia limestone, of younger age than the Theresa, lies with unconformable relations directly upon the Precambrian rocks. The Pamelia limestone is named from the town of Pamelia in Jefferson county. About 18 feet of the upper beds are well exposed in this region in the gorge of the Black river between Deferiet and the dam above Herring (Antwerp sheet), in the quarry at Herring, and in the bed of Mill brook, on the Lowville quadrangle. The Lowville limestone, however, forms the surface bedrock between Deferiet and Herring. The beds in the lower part of the state road-metal quarry at Lowville are also part of the Pamelia formation (figure 20). The Pamelia formation is less than 60 feet thick on the east border of the Theresa quadrangle and is 72 feet thick on the Port Leyden quadrangle at Roaring brook near East Martinsburg. The thickness is probably somewhat less in the Lowville and Antwerp areas. The Pamelia limestone is formed from lime muds deposited in marine waters, which through subsequent compaction have formed solid rock. The calcium carbonate deposits were laid down on a surface of the Precambrian rocks that sloped southwest, or opposite to that of the surface to the north on which the Potsdam and Theresa deposits were made. Lowville limestone. Overlying the Pamelia formation there is another series of limestone beds called the Lowville limestone, which takes its name because of good exposures near the town of Lowville. It is very similar in character to the underlying Pamelia limestone but is separated from it by a calcareous conglomerate or coarse sand- stone bed a few feet thick, and the fossiliferous beds carry a different fauna from that of the Pamelia. The Lowville limestone is well exposed along the gorge of the Black river below Great Bend, in the upper part of the quarry at Lowville (figure 20) and in the bed of Mill brook (Lowville quadrangle). Like the Pamelia, it was deposited as lime muds in marine waters. The basal calcareous sandstone and conglomerate indicate that the shore line probably did not lie far to the north or east and that coarse sand or fine pebbles were spread out over the sea bottom at this time. Along the Black river above Deferiet and along the canal at Her- ring, the base of the Lowville formation is marked by a buff weather- ing sandy limestone bed about two feet thick. On the south bank of HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 17 the river about one-quarter mile below Great Bend, where an aban- doned pulpwood plant stands, there are about 20 feet of the upper division of the Lowville beds exposed. They consist mostly of com- pact dove limestone weathering in layers an inch to several inches thick with locally thicker beds. There are many highly fossiliferous layers, of which the most abundant and conspicuous are the Tetra- dium coral beds. Phytopsis and coiled gastropods are common, and locally thin strata are full of brachiopods, lamellibranchs etc. Trilobites also are found. Occasionally splendidly preserved large cone-shaped orthoceratites are clearly visible on the weathered bed- ding planes. A massive Tetradium coral bed forms the top of the section. At the first low scarp back from the river the Leray lime- stone is exposed. It is a more uniform character and weathers with a checked or flakey character. A few black chert nodules and lenses may be seen on the flat, bare, exposed surfaces on the bench above the scarp, and an occasional orthoceratite. The Lowville limestone on the Antwerp quadrangle is probably not more than 40 feet thick. Ruedemann measured a thickness of about 38 feet on Mill creek at Lowville. A thin section of the basal sandstone of the Lowville formation from the canal east of Deferiet shows it to consist of quartz with a little associated feldspar in a calcareous groundmass. The large grains of quartz are well rounded. The smaller ones are subangular. The feldspars include both plagioclase and microperthite. Occasional zircon and magnetite grains are present. Leray limestone. The Leray limestone overlies the Lowville limestone and together with it is grouped to form the Lowville for- mation. The Leray limestone on Mill creek at Lowville as measured by Ruedemann has a thickness of 13 feet. It is characterized by the presence of a number of thin black layers. It is exposed in many of the low scarps in the extreme southwest corner of the Antwerp quadrangle, in the southwest corner of the Lowville sheet and in the steep slopes southwest of the Black river on the Carthage sheet. Over much of the country underlain by the Leray limestone the latter forms but a thin cap not more than four or five feet thick overlying the Lowville limestone. The lower bed of the Leray is quite massive and about four feet thick, and thus differs from the thinner bedded Lowville beds beneath. It is exposed near the abandoned pulp mill southeast of Great Bend. Watertown limestone. The Watertown limestone overlies the Leray limestone member of the Lowville formation and is about 15 feet thick. It is exposed in the southwest corner of the Antwerp i8 NEW YORK STATE MUSEUM and Lowville quadrangles, and in the slope southwest of the Black river on the Carthage quadrangle. It forms a solid bed of dark blue- gray to black limestone and usually breaks up into small blocks. Large conical-shaped fossils (cephalopods) are often seen on the flat weathered surface of the beds. A few chert nodules occur in the lower beds. Trenton and Cobourg limestones. The Trenton limestone is a relatively thick formation but occurs for the most part outside this region. It is found only in the extreme southwest corner of the Lowville quadrangle and in the unmapped part of the Carthage area. A description by Ruedemann is given later in this report. The Cobourg limestone is also found only in the unmapped part of the Carthage quadrangle and in the extreme southwest tip of the Lowville quadrangle. A description by Ruedemann is given later in this report. POST-PALEOZOIC UPWARP AND EROSION ERA The Adirondack area must have been upwarped as a dome at least as early as the end of the Paleozoic era and probably several times earlier, for we find no evidence of marine deposits after the early Paleozoic except for a very brief interval in the Pleistocene and Recent periods and then only in a thin strip around the outer fringe. It seems highly probable that the same earth forces that folded the Paleozoic beds of the Appalachian mountain province to the southeast warped up the Adirondack dome. Since that time the region has been subjected to the processes of weather- ing and erosion for hundreds of millions of years. The result has been that the blanket of Paleozoic beds so far as they extended over the Adirondack area has been completely stripped off except for residual patches around the outer fringes, and erosion has eaten deeply into the underlying crystalline rocks of the Precambrian except along the foothill belt of the northwest Adirondacks, where the cover of Potsdam sandstone has been too recently removed to permit erosion seriously to modify the old original pre-Potsdam topography. It is the current conception that upwarp of the dome took place several times, was not uniform and that erosion did not proceed at a uniform rate over the Adirondack area but that there were periods of upwarp when erosion was intensified and the rivers bit deeply into the rocks, followed by very long intervals of stability when erosion produced wide areas of relatively flat or gently rolling country called peneplains, The major features of the present topography of the HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 19 northwest Adirondacks are usually interpreted in terms of two such major periods of peneplanation, with perhaps intervening minor epochs of alternating vertical movements and stability. It is thought that the evidences for two such former plain surfaces can be seen in the uniformity with which certain hilltops over a wide area rise to approximately the level of two such sloping plains located at different altitudes. A diagrammatic section is shown in figure 2 to illustrate the Figure 2 Diagrammatic section to show relations of Adirondack dome of Precambrian igneous and metamorphic rocks to the overlying Paleozoic beds and the supposed peneplains. Modified after Chadwick supposed relations of the peneplains to the Adirondack dome and its border of Paleozoic sedimentary rocks. The pre-Potsdam peneplain is represented as coming out from beneath the blanket of Paleozoic beds and rising into the air with the cross section of a dome. This peneplain has been destroyed by prolonged erosion except for a zone near the edge of the sedimentary beds and beneath the latter. The tops of the higher hills in the Adirondack mountains are thought to suggest the existence of a former peneplain (indicated in cross section by the dashed line), that has been so upwarped and so deeply eroded that only the relics are represented by certain hilltops. A third peneplain is thought to be indicated by the erosion surface of the St Lawrence lowlands, into which the present streams are now cutting and which is in turn in process of being destroyed although the work has not gone far. These latter two peneplains are, respectively, referred to a Cretaceous and Tertiary age although this is open to question and the history of the present topographic surface is known to be much more complex than indicated by this simple outline. PLEISTOCENE AND RECENT TIME Following this prolonged era of successive epochs of gentle uplift, doming and erosions, came the time of cooler climate. At several centers in Canada great ice caps formed and spread outwards in all directions. Part of the ice from the Labradorean center came down along the axis of the St Lawrence valley and spread southward across the Antwerp and Hammond quadrangles and was deflected by the north end of the Tug Hill plateau to the southeast up the valley of the Black river in the Lowville area. The ice at this time buried the 20 NEW YORK STATE MUSEUM whole of the topography of this region and advanced to the south part way across Pennsylvania and New Jersey before it halted. Evidence elsewhere indicates that there were several such advances of the ice with intervening periods when the ice melted away and the front retreated to the north, but only the effects of the last ice advance have been identified in this region. During the advance and retreat of the ice glacial drift in part stratified and in part unstratified was deposited directly in connection therewith. During the retreat of the ice the north flowing rivers and their tributaries were for long or short periods dammed by the ice front and a suc- cession of temporary lakes was formed. At this time too the land stood much lower, so that when the ice melted away from the St Lawrence valley, marine waters are thought by Fairchild (’19) to have extended as an arm of the sea up the valley and in these waters further deposits of stratified sands, silts and muds were laid down, partly or completely burying the underlying glacial drift. This arm of the sea, called Gilbert gulf, was destroyed and the marine waters were withdrawn as a result of a subsequent slow uptilt of the land, amounting to some 460 to 520 feet within the area of the Hammond quadrangle according to Fairchild’s estimates. Others, however (Wright, ’23, p. 44; and Goldring, ’22, p. 153-87), question whether the marine submergence was as great as supposed by Fair- child, in which case the beach deposits below the level of the Lake Iroquois deltas would be ascribed to formation in temporary glacial dammed lakes similar in origin to Lake Iroquois itself. The marine waters certainly advanced as far as Brockville, Ontario, and marine fossils are found at an altitude of 30 feet above the river at Ogdens- burg, N. Y. PHYSIOGRAPHY PHYSIOGRAPHIC PROVINCES Taking into consideration the geology, the topography and the history of the origin of the present surface features, the region lies within three different physiographic provinces, although belonging to the foothill belt of the Adirondack mountains. The other two physiographic provinces represented are the St Lawrence Valley Lowlands and the Tug Hill Plateau section. St Lawrence lowlands. In the northwest corner of the Ham- mond quadrangle is a relatively flat country that starts at the edge of a line of cliffs or scarps along the west side of Black lake and Black Creek swamp. This line of cliffs is formed by sandstone (Pots- dam sandstone), and much of the plains country to the northwest is HAMMOND, ANTWERr AND LOWVlLLE QUADRANGLES 21 underlain by flat-lying beds of sandstone or sandy dolomite (Theresa) of Paleozoic age (except for a local belt along the valley of Chip- pewa creek, where erosion has stripped off the sedimentary beds and exposed the Precambrian granite beneath). This portion of the Hammond area (about 28 square miles) belongs to the St Lawrence Lowlands physiographic province of the United States. The cliffs that lie just to the northeast and southwest of North Hammond are formed by the Potsdam sandstone and the overlying Theresa sandy dolomite, which forms the flat upland of the extreme northwest cor- ner of the quadrangle. Similarly, on the Antwerp quadrangle, much of the country south of the Indian river is underlain by flat-lying Paleozoic beds (Pots- dam sandstone and Theresa sandy dolomite), and belongs to the St Lawrence Valley lowlands. Grenville foothill belt of Adirondack Highland section. The geology of the Adirondack Highland section of the United States, like the topography, is in marked contrast to that of the St Lawrence Lowlands section. Practically all of the Adirondack area is under- lain by Precambrian crystalline rocks. The major part of the Adiron- dack section consists of igneous rocks such as anorthosite (Mount Marcy and Keene Valley region), gabbro, syenite, quartz syenite or granosyenite, and granite, with not more than 25 per cent of in- cluded bands of gneiss, quartzite, and crystalline limestone. The Lowville area east of the Black river is characterized by just such a complex of rocks except that there is no anorthosite. Along the northwest part of the Adirondack massif, however, there is a belt underlain by a complex of rocks in which the Gren- ville formations predominate. This belt is exposed for a length of about 60 miles and a width of 30 miles, extending from Potsdam on the northeast to Butterfield lake and Lewisburg on the south- west. It is bordered on the northeast, northwest and southwest by the Paleozoic formations of the St Lawrence Valley lowlands. The rocks are of the Grenville formations consisting of crystalline lime- stone, quartzite, gneiss and schist with about 30 per cent of intrusive bodies of granite and a little associated syenite and gabbro. This complex of rocks is more easily eroded than the more uniform igneous rock of the main body of the Adirondacks and therefore very probably was the site of a major broad lowland in the Precambrian peneplain. The peneplain has not been warped up in post-Cambrian time to the same degree within this belt as within the main Adiron- dacks to the southeast. As a result of both the foregoing facts this belt forms a lowland or low foothill belt relative to the higher more 22 NEW YORK STATE MUSEUM rugged country to the southeast. The Hammond and Antwerp quadrangles lie within this foothill belt except for the two areas belonging to the St Lawrence lowlands, as previously described. In contrast to the flat plains and table-lands of the St Lawrence lowlands, the Northwest Adirondack foothill belt comprises a series of alternating narrow ridges or elongate hills and flat-bottomed val- leys and depressions, and is rather rough, although on a small scale and in a mild fashion. Within the Grenville foothill belt of both the Hammond and Antwerp areas, however, there are numerous isolated patches of flat-lying Potsdam sandstone which give rise to flat- topped hills and table-lands, but they form only a small part of the region. The difference in height between the tops of the hills and the bot- toms of the valleys in the Hammond quadrangle is usually not much more than ioo feet, rarely over 150 feet, and the maximum differ- ence is about 260 feet at the southeast end of Yellow lake, where the lake level is 342 feet above sea level and the high hill at the south- east end is a trifle over 600 feet in altitude. The highest point in the quadrangle is at the south side, south of Spragueville at an altitude of 620 feet. The difference in altitude between the tops of the hills and the bottoms of the valleys is less in that part of the Antwerp quadrangle underlain by Precambrian rocks than in the Hammond area; it is everywhere less than 100 feet. The various kinds of rocks of the Antwerp area have more nearly the same resistance to erosion than those of the Hammond quadrangle and have therefore yielded a much gentler relief. The surface of the Lowville quadrangle, east of the Black river and west of the great delta sand plain along the west border, has a rounded hilly character of moderate relief, with altitudes ranging from 750 feet on the west to somewhat over 1350 feet on the east. The Number 4 quadrangle lies to the east of the Lowville area and has a more rugged mountainous aspect with altitudes ranging from 1200 to 2200 feet although the structure, kind, and age of the under- lying rocks is similar. The difference in topography is simply one of degree and the hills of the Lowville area are the foothills of the main Adirondack massif. Tug Hill plateau. West of the Black river, on the Lowville and Carthage quadrangles, there is a complete change in the type of topography and underlying bedrock from that east and north of the river. The rocks west of the river, except for a narrow band of Precambrian rocks, which skirt the base of the steep slope or escarp- HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 23 ment, are sedimentary beds of Paleozoic age with a very gentle dip. Topographically the area belongs to a flat-topped highland, which slopes away towards Lake Ontario and is known as the Tug Hill plateau. Frontenac axis. The St Lawrence lowlands extend along both sides of the St Lawrence river, from Chippewa bay to Quebec, and along the southeast side of the St Lawrence river to Lake Ontario from a point half way between Alexandria bay and Clayton on the American side, and along Lake Ontario west of Kingston on the Canadian side. The Thousand Islands themselves, however, are for the most part of Precambrian metamorphic and igneous rocks and constitute a connecting link between the similar rocks of the Adiron- dacks on the east and those of the Canadian highlands or Laurentian plateau to the northwest. This belt of Precambrian rocks, which thus separates the St Lawrence lowlands into a northern and south- ern division, and includes the area in the vicinity of Butterfield lake, Alexandria bay, the Thousand Islands, and the Canadian mainland between Kingston and Brockville, is known as the Frontenac axis. TOPOGRAPHIC TYPES Introduction. Six distinct major types of topography, each with its own characteristic features, are present within these quadrangles. These types comprise (1) table-lands, (2) remnants of a Precam- brian peneplain on the igneous and Grenville metamorphic rocks, (3) hill country, (4) flat-floored silt and clay-filled valleys, and depressions, (5) Pleistocene high-level extinct glacial lake deltas, and (6) recessional moraines. Other elements of the topography, although occupying smaller areas, are large domical rock hills com- pletely covered with a veneer of ground moraine, glacial outwash sand plains on the far side of the recessional moraines, and the flood plains of the rivers and creeks. One of the most striking features of these quadrangles is the exceedingly intimate relation between the topography and the kind and structure of the underlying rocks. The grain of the topography conforms in extraordinarily detailed fashion to the grain of the underlying rocks. The general trend of the rock formations, is northeast and in keeping with this, the general trend of the topog- raphy is northeast, although on the Lowville and Carthage quadrangles there is a belt of rock with a northerly to westerly strike and a corresponding parallel alignment of the ridges and valleys. The hills and the valleys that characterize the present topographic surface are the result of processes of erosion acting on rocks of 24 NEW YORK STATE MUSEUM unequal resistance and different structure. Hills have not been split apart one from the other, or raised up as separate blocks or domes. On the contrary the intervening valleys were formerly of rock reaching to the same general level as the adjoining hilltops but being of more easily eroded material have been worn out to form the lowlands. The processes of weathering and the activities of rivers have been the predominant factors in producing the topog- raphy as we find it. During the Pleistocene or “Great Ice Age,” the continental glacier, which moved down from the north, rounded off the hills, removed the soil, slightly eroded the valleys, left a thin veneer of bouldery drift on the hills and of stratified drift in the valleys and resulted in considerable disturbance of the drainage. In general, however, it merely modified the topography which it found, and did not produce its major features. The valleys and hills, except those formed of glacial deposits were all present much as we see them now before tbe glacier came over them. Processes of weathering, such as the expansive breaking and wedg- ing power of freezing water, the alternate expansion and contraction of rocks due to heating and cooling, the solvent and chemical attack of rain water, the expansion due to chemical alteration, and the action of growing vegetation itself are all aiding to break up the rock or dissolve it at the present time. In the pyritous, rusty- weathering gneisses there is the additional agency of sulphuric acid and ferrous sulphate, which are formed by the weathering of the pyrite. The result of all these agencies is the production of mechanically separated blocks, fragments and grains of rock ; of chemically altered and rotted rock ; and of material in solution in the ground water. The particles of rock and soil are then washed into the rivers and transported to the sea. Given time, the present topography will be modeled to much lower relief than the present, as has the present topography been eroded from a very different topography of the past, and by similar agencies. All of the rocks are affected by systems of fractures or cracks (either potential or actual) called joints. Such fractures constitute an important factor in guiding the manner and the location in which the process of weathering and erosion work. Usually there is one set of joints parallel to the foliation or bedding of the rocks, and another at approximately at right angles. The steep cliff faces of many of the granite hills are such because processes of weathering have caused great slabs or blocks to break away along a joint sur- HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 25 face. The usual rectangular character of granite blocks is due to an intersecting set of joints. The limestones disintegrate and break down to a granular sand, which often has a rusty iron-colored aspect, so conspicuous on steep hill slopes and in pits for road materials. Solution wells and narrow troughs are also developed in limestone along joint planes through the action of ground water dissolving the limestone. Pulpit rock on the road south of Oxbow (Hammond sheet) is the result of the removal by solution of a limestone inclusion originally present in the granite here. Included lenses of limestone in partial stages of solution occur in the granite, at the top of the hill above. The granites in weathering break up in both large and small blocks along intersecting systems of fractures; and they also break down through curved scales and flakes peeling off about parallel to the surface. This is particularly conspicuous on the bare hilltops where the scars of fresh rock and the flakes which have peeled off may be seen together. Table-lands. The table-lands are for the most part underlain by the flat-lying Paleozoic sediments ; the Potsdam sandstone, the Theresa sandy dolomite and the Leray cherty limestone forming the topmost layer of rock over wide areas. The table-lands are flat because their surfaces conform approximately to the bedding planes of the sediments and they are marked off from the valleys that have been cut into them by bare cliff faces or very steep slopes, or occa- sionally by a series of steps or benches. Their surface is diversified only by low scarps or steep slopes marking the rise from the top of one bed to the top of one considerably higher. These scarps are often extraordinarily persistent and may be traced for miles. Such scarps are almost invariably formed at the contact of the Potsdam sand- stone and the Theresa dolomitic beds. The Theresa beds rest upon the sandstone and are usually fringed by a line of low cliffs or steep slopes (figure 21). The Potsdam sandstone similarly presents a line of steep cliff faces or slopes at its borders where ever it rises from the surface of the Precambrian rocks. The table-lands thus every- where stand a little above the adjacent country. Except locally, where recessional moraine occurs as in the southwest corner of the Antwerp quadrangle, the table-lands are covered with only a thin veneer of ground moraine and lake silts, usually not more than a few feet thick. Often there are many spots here and there where the soil has been worn off and bare rock exposed. Boulders are lit- tered here and there over the table-lands except where the fields have 26 NEW YORK STATE MUSEUM been cleared for cultivation. At places bare rock for acres in area is exposed, especially on the Potsdam sandstone. Locally the high- ways are on such bare rock areas for considerable distances. On the Hammond quadrangle the table-lands are well developed in the northwest corner, northwest of Black creek and Black lake ; in the vicinity of Grass lake and Lake of the Woods, and within the radius of a few miles of Somerville. There are other local patches of table-land wherever the Potsdam sandstone is exposed over any considerable area. On the Antwerp quadrangle they are excellently developed between Black creek on the south and the Indian river on the north (figure 22). In the southwest corner of the Antwerp quadrangle south of the Black river there is an area of table-land for which the cherty Leray limestone forms most of the surface bed rock and smaller areas are immediately underlain by the Lowville or Pamelia lime- stone. On the Lowville quadrangle table-lands occur only in the extreme southwest corner. Hill or knob country. Hill or knob country (figure 23) dominates the foothill belt of the northwest Adirondacks. The character and orientation of the topography with the exception of the veneer of glacial deposits conforms in detail to the kind and structure of the underlying bedrock. The low hill country of the Hammond and Antwerp quadrangles and the valley of the Black river in the Low- ville area might be called a resurrected topography for in large part it was formed in pre-Potsdam time and has been reexposed by erosion of the overlying Paleozoic beds. The crystalline limestones are relatively soft and soluble rocks, and are therefore found to underlie most of the valleys and low areas and practically all of the lake basins on the Hammond quadrangle. The rain water seeps through the limestone very slowly, but dissolves it and then carries it away in solution to the rivers, and thus aids in lowering the level of the limestone areas. Where broad belts of crystalline limestone are present the topog- raphy usually comprises long, narrow, rounded, lineal ridges with intervening depressions (figure 24). The ridges are parallel to the bedding of the limestone. Often the reason why certain bands of the limestone belt constitute ridges, as compared with adjoining bands, is plain to be seen. In many cases a pegmatite or granite sheet, par- allel to the bedding, forms a rib that strengthened a particular band and caused it to weather less rapidly. Indeed, in some portions of HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 2J the limestone country, each hill means a big white pegmatite dike. In other cases the limestone is strengthened by the presence of siliceous lenses, quartz veins, quartzitic layers or quartzite beds. In other cases, the reason why one band forms a ridge and the other a valley is not evident. The limestone is usually crossed by a series of fractures or joints at about right angles to the bedding. These fractures are frequently widened by the action of rain water that dissolves the limestone and produces long, narrow, rounded open clefts or small elliptical-shaped wells. The great broad lowland from Gouverneur to Somerville lying southeast of the Oswegatchie river is underlain by a belt of limestone. On the Antwerp quadrangle narrow, lineal valleys underlain by limestone are found in the north- west corner, north of Antwerp, and southeast of Lewisburg. The quartzitic gneisses, quartzites and injection gneisses often form hills and ridges that trend parallel to their bedding or foliation, as the case may be. If much limestone is present as intercalated layers or beds in the rusty gneisses, however, a valley or depression may form on them. In some cases even the hardest quartzite beds form valleys, and when layers of quartzitic rock are inclosed in granite they usually form the locus of a valley or depression. Often the quartzites are intersected by a close set system of fractures, in which case they break down and weather rapidly to form valleys. The granite masses uniformly form hills. The only exception is where the granite is strongly fractured or jointed, or has a pronounced platy parting parallel to the foliation. The faithfulness with which the trend of the ridges reflects the structure of the rocks which form them is splendidly exemplified near Rossie. If the topographic map of this area is studied, it is seen that the ridges appear as though arranged in a series of festoons. The arrangement of these hills corresponds in detail to the curves resulting from the synclinal and anticlinal structure of the rocks (figure 25). On the borders of the granite masses or where the granite occurs in narrow sheets, it has a marked banded or laminated structure. This foliation guides the development of the topography and we have a series of lineal ridges parallel to the foliation, with gentle topo- graphic slopes in the direction of dip of the foliation and steep slopes in the opposite direction. In the cores of the larger granite masses, however, the foliation becomes inconspicuous, and in the place of lineal ridges we have flat-topped hills, such as are well shown about one mile northeast of Robb School, one mile northwest of Scotch Settlement, on the top of the granite mass west of Payne lake etc. 28 NEW YORK STATE MUSEUM The major factor giving rise to such flat-topped hills in the granites is a pronounced, almost horizontal jointing or sheeting structure. The granite tends to break off in great flat slabs parallel to this structure. The mass of granite in the vicinity of California School (Ham- mond quadrangle) exhibits extraordinarily well the effect of the foliation or banding of the granite in influencing the type of topog- raphy which is developed. The California road, after leaving the flat valley of Birch creek, climbs up the steep slope formed by the resistant granite mass ; and then, about one-quarter of a mile farther on, it drops down over a steep slope to a relatively broad flat area, which extends to and slightly beyond the schoolhouse. This broad, relatively flat area is surrounded on all sides by a series of lineal ridges with steep faces toward the schoolhouse, gentler slopes away on the opposite side, and flats between the successive ridges. Care- ful examination will show that the foliation and banding of the granite is almost horizontal over the broad flat area, but dips gently away on all sides from it, the foliation thus having the form of a dome or inverted basin. A large proportion of the flats on top of the dome and between the encircling hogbacks is cultivated. The other varieties of igneous rock also tend to form hills especially when they occur in limestone (figure 26). On the Lowville quadrangle there is a greater uniformity in the kinds of rock underlying the hill country, since most of it is igneous rock. The alignment of the ridges and longer axes of the hills is parallel to the banding and foliation. This is well shown in the semi- circular belt of rock on the west half of the quadrangle, where the trend of both the rock structure and topography changes from south- west, south of New Bremen to north-south, south of Croghan, to northwest, northwest of Croghan. Included bands of schist have localized many of the depressions, and an intensive jointed structure has been the cause of numerous others. Even within the various types of igneous rocks themselves, however, there is a variation in the degree of resistance and the weaker bands have suffered rela- tively more erosion. For example, on the Lowville quadrangle there is a relatively flat lowland extending northward from Crystal creek between the two recessional moraines to the Beaver river and then extending northwest to Swiss creek. This is probably underlain by a belt rich in schist inclusions. Remnants of Precambrian peneplain. One of the great out- standing features of the Hammond and Antwerp quadrangles is the fact that the present topography on the Precambrian rocks is for the most part of ancient origin, is pre-Potsdam age. HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 29 The reason why this topography of the past, hundreds of millions of years old, has been preserved for us, is that it was warped beneath the sea and buried under a great thickness of sediments of Paleozoic age. The events of all subsequent time were insufficient wholly to remove this protecting cover of rock strata. Remnants of Potsdam sandstone are still found in the ancient pre-Potsdam valleys of the Hammond and Antwerp areas, and in much of the region to the west of these quadrangles the extension of this topography is still buried beneath its cap of sandstone, yet to be exposed to the light of a future time. There has been some slight modification of the ancient topography by erosive agents and by the glaciation of Pleisto- cene time where the protecting cover of Potsdam sandstone has been eroded. The altitude of the district is so low, however, and the time is so short since most of the protecting blanket of strata was removed, that the topography of the present is substantially the topography of a remote past age. It was, and is, a surface of moderate relief, where the rocks varied greatly in resistance as in the Hammond quadrangle, and of very flat topography where the rocks are uniform in character. Farther to the southeast, in the Lowville quadrangle east of the Black River valley, and in the core of the Adirondacks, the protect- ing sheet of Paleozoic sediments was removed long ago ; and the present topography there is largely the result of erosion during a more recent era. Yet we find a similarity in kind of topography and in the relation of altitude to kind of rock. On the Antwerp quadrangle, forming a fringe to the outer north- ern and eastern border of the Paleozoic beds, there is a band a mile or two wide of very flat topography developed on the granite, syenite and gneiss of the Precambrian. It is well shown in the loop of the Indian river between Lewisburg and Woods Mill (figure 27) north of the Indian river from Woods Mill to a mile beyond Nauvoo and within the great loop of the Indian river between Nauvoo, Antwerp and Philadelphia (figure 28). Considerable areas are also found north and west of Philadelphia, in the extreme southeast corner of the Antwerp quadrangle, south and southwest of North Croghan and south and southwest of Fargo. On the Lowville quadrangle similar areas of flat topography developed on Precambrian granites and fringing the Paleozoic beds are exposed south of Dadville and New Bremen and southeast of Lowville, although in this area the overburden of silts obscures much of the surface. On the Ham- mond quadrangle, flat granite areas are again found where Chippewa creek has removed a strip of the formerly overlying Paleozoic sedi- 30 NEW YORK STATE MUSEUM mentary beds, and similar flat areas on syenite or granite are exposed along the borders of the Potsdam sandstone northeast of Pope Mills, north of Hyde School. On the Hammond and Antwerp quadrangles between Natural Dam and Antwerp there are broad relatively flat areas developed on the belt of crystalline Grenville limestone within the general vicinity of the areas of Potsdam sandstone (figure 29). Other small local flats are found developed on Precambrian rocks adjacent to small areas of Potsdam sandstone throughout the Ant- werp and Hammond quadrangles. The constant feature that these flat areas on the Precambrian all have, is that they border areas of Paleozoic sediments and where exposures are favorable may be seen to pass beneath the overlying sediments. As the flats are followed out away from the borders of the Paleozoic beds they are found to become more and more etched by the processes of erosion (figures 28-30) and finally merge into the hilly or mountainous country typi- cally developed on the Precambrian rocks of the eastern part of the Lowville area and the core of the Adirondacks. Like the table-lands the flat Precambrian peneplain surfaces are as a rule covered with but a thin veneer of drift and much bare rock is exposed. This Pre- cambrian peneplain will be described in more detail in a later chapter. PLEISTOCENE TEMPORARY LAKES AND HIGH LEVEL DELTAS At many localities in the Antwerp, Carthage and Lowville quadrangles there are relatively flat areas of sand which spread out fan-shaped from a point on a river or creek as a center and have steep slopes to the lower bed rock level on the outer edge. These sand plains all lie well above the bed of the present rivers or creeks to which they appear to be directly related. They are confidently interpreted as deltas that were built out by the creek or river into lakes temporarily formed by the front of the ice acting as a dam to drainage during the retreat of the glacial ice cap in Pleistocene time. The great ice cap during its retreat first melted away from the hill tops and for considerable periods of time the valleys of the Adirondacks must have been partly occupied by tongues of ice, whose top surface lay well below the tops of the hills on either side. In a valley such as that of the Black river or the St Lawrence, such a tongue of ice would form a dam to the river and result in a tempo- rary lake. As the ice tongue melted down and back its dissipation was stayed for considerable lengths of time at certain positions as indicated by the recessional moraines and at other times it retreated rather rapidly to a new position of temporary halt. As lower and HAMMOND. ANTWERP AND LOWVILLE QUADRANGLES 31 lower outlets were opened for the waters and as the ice front retreated intermittently backwards, lakes at successively lower levels, and of successively different location, shape or size were formed. Into these lakes came streams flowing from the mountains, which owing to their recent exposure from beneath the ice cap, were covered with a veneer of drift readily subject to erosion. The streams were therefore heavily laden and built forward deltas into the temporary lakes. Successive lowering and final draining of these glacial lakes has thus left a series of deltas at different altitudes perched on the sides of the hills along many of the streams and rivers. The sur- face of much of the deltas is so sandy and porous that but little of their area is in cultivation. Some formerly under cultivation have been abandoned because of the movement of sand by the wind sub- sequent to the removal of the forest litter which formerly held it. Wind work has locally built up many sand dunes on the deltas and scoured corresponding basins. A good description of the history of these glacial lake waters has been given by Fairchild (’12). Lake Port Leyden and deltas of Independence, Beaver, and Oswegatchie rivers. The valley of the Black river was at one period occupied by a tongue of ice that sufficed to dam back the river and result in the formation of a lake extending from Honnedaga (Rem- sen sheet) to north of the northeast corner of the Lowville quad- rangle, a length of 44 miles (figure 3). This lake has been called Lake Port Leyden by Fairchild (’12). Some data have been given by Miller, who writes, The highest water level in this lake was apparently something over 1300 feet at which time an outlet probably crossed the Black River West Canada Creek divide near Honnedaga (Remsen sheet) and flowed southward toward Trenton Falls. Further retreat of the ice lobe down the Black River valley certainly opened an outlet south- westward past Boonville and down Lansing Kill toward Rome, caus- ing deposition of the great delta deposits north of the latter place. The preglacial divide was doubtless near Hurlbutville. The lake stood approximately at the 1250 foot level when it started over this divide and it cut down the divide rapidly until the 1140 foot level in the lake was reached. By this time the ice tongue had so far melted as to allow an escape of the water northerly and north- westerly along the west side of the ice tongue and into Lake Iroquois near Watertown. Along the eastern border of the Lowville quadrangle there are two great lobate areas of sand lying between the altitudes of 1280 to 1350 feet that completely cover the underlying rock except for an 32 NEW YORK STATE MUSEUM BLACK RIVER VALLEY ICC YONOUC Figure 3 Block diagram showing assumed conditions at one stage in the history of Pleistocene glaciation in the Black River valley. Shows the mountainous topography on the Precambrian crystalline rocks of the Adirondacks, the scarps and rock benches of the Tug Hill plateau, and an ice tongue occupying the Black River valley obstructing the north-flowing drainage and forming a lake (Port Leyden) that drains south. Deltas have been built forward into the lake and overlap on the submerged edge of the ice HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 33 islandlike hill here and there which tops the surrounding flats. This high level sand plain has a practically flat surface pitted by numerous depressions or basins, many of which form the sites of ponds and lakes whereas others are dry. This sand slopes upward from the western edge and was built into Lake Port Leyden largely by the Independence, Beaver and west branch of the Oswegatchie rivers. These rivers have since cut through the delta and locally exposed the underlying ground moraine or bedrock, revealing the fact that along the western edge the sands are up to 150 feet in thickness, thinning to zero at the eastern edge where they abut against the mountains. The lower beds of the delta are clay, the upper beds are stratified sand and some gravel. The basins on the surface of the plain have steep sides, often from 40 to 60 feet deep. The origin of the pitted character of the sand plain is probably to be found in the relation of the delta to the ice tongue. The delta sands of the east border of the Lowville area similarly extend southward across the Port Leyden quadrangle on to the Remsen sheet. The delta surface is pitted throughout this length. At the present time such a surface is known to result from the consequences attendant upon the burial of the uneven irregular surface border of an ice tongue and from the stranding and burial of icebergs. In the first case the depressions arise from the subsequent melting of the ice and slumping in of the overlying sands, and in the second case from the melting of the ice block. Some of the depressions are very large to be ascribed to stranded ice blocks. It seems very probable that the delta sands were built forward over a stagnant longitudinal border zone of the ice tongue that was acting as a dam to the river. The general border of the ice tongue might well be similar to that of the ice front at the time of formation of the Albion moraine. Stagnant, partly buried ice sheets such as are here postulated are found today in front of some of the glaciers of Alaska. Lake Glenfield and deltas. Following the stage of Lake Port Leyden the ice front retreated so as to open an outlet for the lake waters around the north end of the Tug Hill plateau into the Ontario basin. As Fairchild writes (’12, p. 16) “This lake did not have a permanent level but a series of falling levels as new outlets were opened on the north-facing slopes southwest of Carthage. The primitive and highest level was something over 1200 feet, the present altitude of the bottom of the highest channel, a mile northwest of Copenhagen. The lowest level was a blending into the waters of Lake Iroquois at about 740 feet. At the beginning of this stage 34 NEW YORK STATE MUSEUM the waters extended up the valley to Boonville, but the area dimin- ished as the surface fell by the opening of successively lower outlets until only the lower or northern part of the valley was flooded, probably only the stretch from the ice front to Glenfield. At a high stage of Lake Glenfield, deltas at Sweet School (Low- ville quadrangle), Copenhagen and southeast of Harrisburg (Carth- age quadrangle) were built and are now at an altitude of 1200 to 1220 feet. At a lower stage deltas were built at Kirschnerville and Belfort on the Lowville sheet and at Kings Falls (Carthage sheet) and are now at altitudes of 900 to 960 feet. At a third lower stage deltas were formed at Ossoit School and Croghan and are now at an altitude of 840 to 880 feet, and at a fourth lower stage deltas were built at Natural Bridge and along the immediate valley of the Black river. They are now at altitudes of 760 to 820 feet. Sweet School and Copenhagen deltas. South of Sweet School (Lowville sheet) there is a broad flat sand plain at an altitude of a little over 1200 feet. On the east is a steep slope of the front of a higher delta (1345 feet) built in Lake Port Leyden, and on the west is the steep front sloping down to bed rock. This delta was built at the higher stage of Lake Glenfield by the Beaver river. A small part also occurs on the south side of the Beaver river. The delta depos- its of the Independence river made at the same time are confused with a morainic character due to the presence of the ice front there. South of Copenhagen (Carthage sheet) there is a delta at an alti- tude of about 1200 feet along the Deer river, especially well shown along the east side of the valley. A delta at about 1220 feet, a mile in diameter, occurs along Beaver river (Carthage quadrangle) about two miles a little south of east of Harrisburg. Belfort and Kirschnerville deltas. On the Lowville quadrangle west of Belfort along the Beaver river is a delta sand plain with a radius of about two miles extending as far west as Balsam creek, High Falls and Henry School. At the west its altitude is about 920 feet sloping upwards to 980 feet at Belfort. This level, 960'-]-, marks an intermittent lowering. Near and at Kirschnerville, Mur- mur creek and Sandy creek built deltas into this lake. The sand plains extend for a mile and a half north of Kirschnerville, where they do not go above 940 feet, and east of Kirschnerville they do not reach above 960 feet. The slightly greater altitude attained by the Belfort delta is due to subsequent uptilting. Croghan and Ossoit deltas. At a date later than that of the formation of the Belfort and Kirschnerville deltas Lake Glenfield HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 35 was lowered and a temporary lake established at a level which is now about 860 feet. The town of Croghan is built on delta sands, which form a flat extending up Black Creek valley. The delta has a gentle slope up the creek valley from 840 feet to 880 feet. There is a steep drop from 840 to 820 feet which marks the front of the delta at Croghan. South of Black creek the sharp drop is from 860 to 840 feet. This delta was built by Black creek, which has now cut a valley in the delta and locally exposed bed rock and ground moraine. Beaver river must also have built out a delta into this lake but it has since scoured it out so completely that only relics are left. A remnant of a clay terrace at an altitude of 840 feet occurs just at the forks where the Harrisville road starts about three- quarters of a mile north-northwest of Croghan. At Ossoit School there is a delta sand plain along Crystal creek with a slope up stream from 840 to 880 feet. Kings Falls delta. On the west bank of the Black river (Carth- age quadrangle) two miles west of Deer river there is a small terrace at an altitude of 900 feet, formed by Deer creek just below Kings Falls. Natural Bridge delta. On the Antwerp quadrangle, spreading out from Natural Bridge (Lake Bonaparte sheet) as a focus is a delta sand plain with a radius of two and one-half to two and three- quarters miles. The higher parts are at an altitude of 820 feet but most of it is about 800 feet. Natural Bridge is built on it. This delta is remarkable for the semicircular flat basin on the north side where the delta is absent and a swamp is present. The slopes of the delta facing the swamp are steep everywhere. The best explan- ation for this seems to be that the end of a tongue of ice lay here in the valley of Indian river and that the delta was built out across it. When the ice melted, the vacant space was left. On the Low- ville area delta sands up to an altitude of 780 feet are prominent along the east side of the Black River valley where side streams such as the Beaver river, Crystal creek, Harvey creek enter and on the Port Leyden quadrangle where the Independence river enters. Mil- ler (To, p. 54) reports delta sands at about 800 feet on the Port Leyden quadrangle along the Black river. Probably all these deltas were built into the same lake as that at Natural Bridge, the northern deltas being higher because of subsequent uptilting at the north- northeast. Lake Iroquois and Pine Plains delta. After a further stage of recession of the ice cap came another temporary halt of the ice front in -such a position that the St Lawrence River drainage remained 36 NEW YORK STATE MUSEUM dammed and a great lake occupied the present basin of Lake Ontario together with a broad belt around its borders. This lake is known as Glacial Lake Iroquois and drained into the Mohawk river at Rome. The waters of Lake Iroquois completely covered the area of the Hammond and Antwerp quadrangles except for the south- east part of the Antwerp sheet. The great sand plain on the Ant- werp area, fanning out from Herring with a radius of six or seven miles, is the old delta of the Black river built out into Lake Iroquois. A tongue of this delta extends northeast up the valley of Indian river to an altitude of 700 feet at the east end of the sheet. Indian rivet did not build a delta into Lake Iroquois because of Indian lake (Lake Bonaparte sheet), which acted as a settling basin at a higher altitude. The Pine Plains delta extends back up the valley of the Black river across the Carthage quadrangle on to the Lowville sheet, where it merges gently into sand plains formed in the temporary lakes of higher levels already described. Just northwest of Fargo the delta sands appear to reach an altitude of about 740 feet but much of the higher part of the delta is but little above 720 feet and the great flat area of Pine Plains (figure 31) is 680 to 700 feet. Along the south side of the Black river it is difficult to map the delta sands. The bedrock with a veneer of ground moraine and kames is so near the surface that the hills of gravel and boulders often project through the sand and the sand is often eroded locally so as to expose the ground moraine. The sands here do not reach to 720 feet and are mostly at 700 feet or less. A thin veneer of sand occurs at higher levels but it is difficult to tell whether it is associated with the glacial drift or with the sand plain. The area north of the Black river between Deferiet and Herring is a puzzle. It was in part above the level of the delta and in part the sands must have been swept off by the river when it was at a higher level. Southwest of Great Bend the altitude of the delta does not quite reach 700 feet. It seems probable that the lake level may have stood for some time at about 700 feet, although doubtless some deposits were made when the lake was at higher levels. East of Hubbard Crossing there is a great area of sand dunes and one large depression scoured out by the wind. Pebbles and cobbles polished by the wind blown sand are strewn over the surface. The front of the delta sands at the north and west is a steep slope 100 to 120 feet high but has a gentle slope to the northeast up the valley of Indian river. In front of the delta deposits, however, there extends a great flat underlain by clay or silt for several miles farther north to the Indian river southwest of Philadelphia. HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 37 The delta consists of clay at the base and sand at the top. The clay beds are splendidly exposed at the base of the bluff banks cut by the Black river one and one-quarter miles north of Deferiet. The beds of clay are again exposed all around the north edge of the steep front of the delta sands and directly underlie the country between the sand of Pine Plains and the Indian river west of Black creek (figure 32). A thin veneer of clay and silt also covers the table- lands north of the delta sands, east of Black creek, and forms deeper deposits occupying the valleys between the hills. At several localities morainal hills form islets above the level of the clays and, on the other hand, the Black river, Indian river, and Trout brook have sunk through the clay beds and exposed the underlying ground moraine, so that their valleys are strewn with boulders. The silts make good farm land. The surface of the clay belt rises from 480'+ along the Indian river southwest of Philadelphia to 535 feet at Strickland Corners, 580 feet a half mile west of Reedville and a mile south of Sterlingville, and 650 feet at Great Bend. This makes a rise of 55 feet in the first three miles, 100 feet in the first four and one-quarter miles, and 170 feet in eight miles, or an average slope of 21 feet a mile in a north-south direction. The maximum thickness of the sand beds that overlie the clay is less than 100 feet beneath Pine Plains, and the clay must be locally 30 to 40 feet around the Great Bend of the Black river, but on the average the clay is probably not much more than 15 feet thick. The bedding of the delta deposits slopes outward from the Black river and all the underground water drains in a similar direction and not into the river. Numerous springs are thus found on the outer border of the sands, such as Knapp Spring, Cold Spring etc. Delta deposits at lower levels. Deltas at lower levels than that of the Pine Plains delta are inconspicuous on the Hammond and Antwerp quadrangles. There are local sand and silt plains, however, here and there that suggest deposits in temporary lakes. North and west of Antwerp at the head of Hawkins creek there are terraces at an altitude of 500 to 520 feet. Similarly, along the valley of Shingle creek there are broad flats at about 510 feet, and south of Spragueville there is a flat-topped hill with what appears to be a dissected terrace at 540 to 560 feet, and southwest of the Spragueville cemetery the altitude of a sand plain is 540 feet. In the southeastern part of the Hammond quadrangle the old shore line of a temporary water level appears to be indicated by deposits of sand along the north side of the Gouverneur moraine at 38 NEW YORK STATE MUSEUM Spragueville and on the Potsdam sandstone west of Malterner creek and along the creek at an altitude of about 460 to 480 feet and the Hats along Matoon creek from 440 to 480 feet. In the northwest part of the Antwerp area terrace deposits up to 440 feet are found three-fourths of a mile north of Bentley Corners, and flats occur at 440 feet altitude along Otter creek, and 420 feet at the mouth of Halls creek. There is extensive evidence that this region together with all of northern New York was at a much lower altitude during Pleistocene time than it is now. Marine shells have been found in beds 30 feet above the river at Ogdensburg and at altitudes of 335 to 360 feet at Norwood, showing that there has been postglacial uplift of at least this amount and probably more. The arm of the sea that then occupied the St Lawrence valley consequent upon the withdrawal of the ice is known as Gilbert gulf. No marine fossils have been found in either the Hammond or Antwerp quadrangles, however, and it is not certain whether the deposits at the altitudes of 420 to 480 feet were made in a temporary fresh water lake or in the marine waters of Gilbert gulf. Flat silt and clay-floored valleys. In each valley and depression between the ridges of the hill country and among the table-lands there are deposits of silt, clay or sand that yield a more or less broad flat even floor. The larger rock hills rise abruptly out of flat plains. With only the exception of some of the higher hills every square foot of the territory of the Hammond, Antwerp and Lowville quadrangles has been for long or short periods beneath temporary lakes and therefore has received deposits of material from such lakes, usually of silt or clay. Similarly, because of the suc- cessive intermittent or gradual lowering of water levels, every square mile has passed through the zone of wave action and the veneer of glacial drift subjected to erosion, transportation and deposition. In consequence the steep slopes and rock knobs have been more or less swept clear of their original cover of glacial debris, especially of the finer materials. The sediment thus eroded has been shifted to lower levels and accumulated in the depressions and valleys of the then lake bottoms. Silt is the characteristic material of many of the valleys and depressions. Sand bars are found locally at higher alti- tudes. About two miles southeast of Natural Dam, on the north bank of the Oswegatchie river, a bed of well-stratified banded clay is exposed. The finer materials that the rivers brought into the tem- porary lakes were in considerable part swept far beyond the mouths of the rivers and spread out in front of the high level sand plains, HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 39 as is so conspicuously true in the case of the Sterlingville clay belt in front of the Pine Plains delta on the Antwerp sheet. The major streams and creeks, as well as some of the minor trib- utaries following the retreat of the ice, the draining of the temporary ice-dammed lakes and the uplift of the land reoccupied the valleys and sunk their courses more or less deeply into the silts and clays which partly fill them, so that bluffs mark their courses. A terrace is conspicuous along both sides of the valley of Chippewa creek (Hammond quadrangle) northeast of the road. The terrace lies largely between the 260 and 280-foot contours but extends up to 285 or 290 feet. It rises to at least 30 feet above the present creek level. Along the Indian river there are relics of terraces or sandbars. At the north side of the hill north of the north end of Muskalonge lake there is a remnant of a sand terrace about 30 feet above the river. About one mile north of the outlet of Red lake on the west side of the river there is a sand terrace perched on top of a hill 40 feet above the river level and with a sand bar in the lee of the hill. The Oswegatchie from Natural Dam to the Wegatchie (Hammond quadrangle) is cut 40 to 50 feet below the level of the sand and silt-filled valley. Along the northwest arm local sand plains rise higher above the river. Southwest of Elmdale there is a hill 60 feet above the river level covered with sand and near the eastern edge of the Hammond quadrangle relics of sand plains are found 80 feet above the river. The Black river has similarly cut its immediate valley through the clays and sands of Lakes Port Leyden and Glenfield, so that bluff banks 15 to 20 feet high line its course. Similar conditions exist along its tributaries where they cross the valley flats. On the Antwerp quadrangle similar conditions are found along the Indian river from east of Antwerp to beyond the west border of the sheet, and along Hawkins creek north of Antwerp. RECESSIONAL MORAINES AND GROUND MORAINES Introduction. While the Pleistocene ice cap was in the process of withdrawal, through melting away, there were periods when the front of the ice remained within a relatively narrow belt for a con- siderable period of time and other periods when the ice front was gradually retreating toward the north through more rapid melting. As the ice melted a large part of the debris which it carried was deposited at the margin. Hence, when the ice front was gradually retreating this debris would be spread out as a relatively thin sheet 40 NEW YORK STATE MUSEUM of “ground moraine,” whereas when it remained relatively stationary for a considerable period of time, the debris would accumulate as a thick belt of submarginal drift, known as a “recessional moraine.” Such a moraine usually consists of hillocks and hollows following one another in rapid succession without order in their arrangement, except that they lie within the belt of the terminal moraine. This irregularity in the thickness of deposit results from the uneven dis- tribution of debris within the ice, from the local fluctuating move- ments of the ice and the irregularities of the ice front and from local burial by debris of portions of the ice front with subsequent melting of the ice and irregular slumping of the overlying drift. In part the recessional moraines consist of embankmentlike ridges. At least 12 such recessional moraines are found in these quad- rangles. The moraines are named from the town nearest to which they pass. In as much as the moraines of these quadrangles were practically all deposited when and where the ice front was bordered by temporary lake waters, they tend to have a less marked and con- trasted a relief and to consist of stratified material in contrast to typical till. Their surfaces in contrast to the usual land-laid moraines are littered with only a few boulders. On the Lowville sheet five such recessional moraines are well shown. They were formed by an ice tongue which lay in the Black River valley while the hills on either side were free of ice. Along the west side of the valley conditions were unfavorable for the accumu- lation and preservation of bands of recessional moraine on the steep slopes and only inconspicuous evidence of them is found here. The country to the east of the Black river is hilly but in general with a much more gradual slope upwards towards the east, and on this broad slope the recessional moraines are well marked. A sketch showing the relations of several recessional moraines and of a high level delta to the bedrock is given in figure 4. Dicob moraine. The Dicob recessional moraine is exposed inter- mittently along a line extending south-southwest through Dicob School from about three-fourths of a mile east of Bush’s Corners to the south border of the sheet. At the northeast end it is buried beneath the delta sands built out by Beaver river into glacial Lake Port Ley- den. Most of this moraine is not higher than 1200 feet, but about one and one-half miles northwest of Sperry ville there is a hill of glacial drift rising to an altitude of 1300 feet, or 200 above the surrounding country. Most of the belt adjoining the west steep slope of the sand plain at the south part of sheet has a gentle undulatory surface HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 41 with here and there hills covered with boulders rising above the general level. The material in general is stratified sand and gravel with a few boulders. The equivalent of this moraine on the west side of the valley is not certain. On the Carthage quadrangle a moraine lies at the foot of a scarp slope west of West Martinsburg LD : Sand deposits of extinct glacial lake, and recent alluvium KM : Recessional moraines DD: Delta sand deposits in extinct glacial lake, Port Leyden PC: Precambrian igneous and metamorphic rocks Paleozoic : Bedded limestones Figure 4 Section along line running from just south of Lowville through Crystal Lake School, overlapping on Carthage and Number Four quadrangles, showing the relations of the overlying Pleistocene deposits to bed rock. Vertical scale is 15 times the horizontal where it rises to an altitude of 1320 feet. Farther northwest its altitude is given as 1250 feet. The altitude is thus somewhat higher than that of the Dicob moraine and it is not certain whether it is the equivalent of the Dicob belt or of a belt farther east assumed to lie beneath the cover of delta sands. Northwest of West Lowville (Carthage quadrangle) there are small narrow patches of kamelike material at an altitude of about 1240 feet. Crystal Dale moraine. The Crystal Dale moraine extends north from the south edge of the Lowville sheet through Puffer school and just west of Crystal Dale to within a half mile of Strifts school, where it takes a bend to the east and then extends two and one-half miles farther north, passing about a mile east of Kirschnerville, At the south end this moraine appears to merge with the Dicob moraine. All except one small hill is below 1180 feet and usually the higher hills are 1120 to x 160 feet. Croghan moraine. The Croghan moraine extends as a well- defined ridge from just west of Petries Corners nearly to Croghan (figure 33). North, south and southwest of Monnatt School there is a group or line of gravel hills a couple of miles long which prob- ably belongs to the Croghan moraine. The altitude of the higher hills is 940 to 1000 feet. 42 NEW YORK STATE MUSEUM Croghan West moraine. The Croghan West moraine lies about a mile to the west of the Croghan belt and parallel to it from two miles east of Bush’s Landing as far as the Beaver river (figure 34). It can be traced intermittently for six miles farther to the northwest of Croghan. Its altitude southeast of Crystal creek is 940 to 1000 feet but is only about 900 feet at the northwest. Along the limestone escarpment on the west side of the valley there are intermittent patches of moraine at an altitude of 960 to 980 feet. It shows in the little creek valley near the south edge of the quadrangle, in a road cut one and one-half miles from the south border, and in Mill Creek valley. Back of the Lowville Fair Grounds and at intervals for three-quarters of a mile north, kame morainic material consisting of stratified sand and gravel with a sprinkling of boulders is banked up against the escarpment and locally forms a narrow bench at 960 to 980 feet altitude, just below the top of the steep slope. These patches of moraine were doubtless formed by the west side of the ice tongue at the same time the Croghan and Croghan West moraines were being formed on the east side. Beaver Falls moraine. This moraine is not very well defined. It extends south from Beaver Falls for a couple of miles and a northward extension is indicated by a moraine embankment of gravel a mile long alongside the west side of the road between two and three miles north-northwest of Beaver Falls. The altitude of the higher hills is 840 to 860 feet. On the west side of the valley, at the south edge of the Lowville area, there is a much dis- sected kame terrace or moraine that rises from 820 to 840 feet at its outer edge to 880 feet at the inner edge. North of Lowville there are bouldery gravels forming a terrace at an altitude of 860 to 880 feet, although in part this terrace is on bedrock. These gravels appear to have been formed in part by deposition from the edge of the ice and in part by streams from the escarpments to the west at the time the east edge of the ice tongue was forming the Beaver Falls moraine. Deer River moraine. On the Carthage quadrangle a group of hills forms a belt about four miles long and a mile wide extending east from Deer river right across the Black River valley. These hills are rounded, 40 to 80 feet high and composed of bouldery till or gravel. There are many big boulders in this moraine. The altitude of the higher hills is 800 feet. Carthage, Devoice and Fargo moraines. In the southern part of the Antwerp quadrangle and the north-central part of the Carthage HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 43 area there is a curving belt about three miles wide in which morainal ridges or bands of morainic hills are found. This belt lies south of the Black river and extends southeast from Great Bend to Carthage, where it is crossed by the Black river and turns and strikes north- east towards Natural Bridge. West of Natural Bridge the belt of moraine topography is buried beneath the sand plain delta of Indian river. The belt thus projects up the Black River valley for a few miles. The front of the ice at the time of formation of the morainic bands must have had a tongue which lay in the valley of the river and projected somewhat beyond the general line of the ice front for that time. East of the Black river three bands of recessional moraine may be distinguished. The best individualized band is that followed by the road southwest from Natural Bridge towards Devoice Corners (figure 35). Much of this moraine is a single high ridge rising 30 to 60 feet above the lowland on either side with an altitude of 800 to 840 feet. To the southeast of this band, paralleling it and separated from it only by a narrow valley or low flat, is a broad band of moraine one-half to one mile wide. The Carthage moraine does not form a uniform high ridge but usually comprises a low hummocky topog- raphy. In the southwestern and northeastern part of this moraine belt none of the kames are 20 feet high and the moraine in general does not reach above 800 to 820 feet. It is broader, lower and more hummocky than the ridge moraine from Natural Bridge to Devoice Corners. Three-quarters of a mile south of Devoice Corners, how- ever, there is a group of kames that do reach 20 feet in height and an altitude of 850 feet. This moraine extends southwest on to the Carthage quadrangle. The road northeast to Rogers follows it. Here the moraine comprises a broad gently sloping veneer of drift on the bedrock without the characteristic hummocky character. About one and one-half miles southwest of North Croghan there is a space where the moraine is missing and bare rock is exposed, because the height of the moraine here did not reach to that of the bedrock. One and a half miles south of Carthage there is a hill 80 feet high composed of bouldery drift of morainal origin. The third band of the moraine belt of the Fargo moraine extends north- east and southwest through Fargo. The Fargo band is a broad belt about a mile wide. For about a mile southwest of the Indian River delta sands the band has a very conspicuous morainic character. Just southwest of the schoolhouse three and one-half miles north- east of Fargo there is a splendid group of high kames. The morainic character also shows up splendidly along the west side of the road 44 NEW YORK STATE MUSEUM from Fargo to Natural Bridge, where there are long steep-sided ridges 20 to 40 feet high. The kame morainic topography is mag- nificently displayed for about a mile and a half beyond a point a mile and a half from the forks at Fargo. Much of the moraine here is marked by many large slabs of Potsdam sandstone at the surface. Farther southwest the moraine has, in general, a smaller relief and a more hummocky character. Between Devoice Corners and Fargo the kame morainic character is not well shown above an altitude of about 760 feet where the surfaces above this consist of broad flats or gentle slopes covered with sand and scattered boulders. In the Fargo band of moraine there are small isolated conical hills perched on top of the Theresa table-lands, large round isolated hills, groups of hills, long ridges and low rolling hummocky topography. West of the Black river on the Antwerp quadrangle a splendidly developed moraine extends for a couple of miles southeast from Great Bend between the two roads. At the south end there are gravel pits. To the southwest of the moraine is a broad flat table- land with a thm veneer of sand that may in considerable part represent an outwash plain. Another band of moraine extends for two and one-half miles south from the south side of the Black river just southwest of Herring on the Antwerp quadrangle. On the south side of the Black river at the Herring dam there is a bouldery moraine, and the large hill a mile south has a sandy surface with boulders strewn sparsely to moderately over its surface, which may represent another band of recessional moraine. Along the road from Deferiet to Great Bend and beyond there are low hills that here and there rise above the plain. The road and railroad cuts show these hills to be composed of bouldery drift and the boulders are nearly all of limestone. On the west bank of the Black river at Deferiet Bridge the river has cut into high gravel banks that are of kame morainic character. Watertown moraine. Traces of a belt of moraine believed by Taylor (’24) to be part of a band well developed near Water- town are found on the Antwerp quadrangle about a mile west of North Wilna, southwest of Woods Mill and a half mile south of Dority pond. West of North Wilna the moraine is marked by low hills composed of bouldery till. The boulders are predominantly slabs of Theresa and Potsdam beds. The slabs are at all angles and some are very large. Philadelphia moraine. The Philadelphia moraine on the Ant- werp quadrangle extends along a line parallel to the Indian river HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 45 and a mile southeast of it from near Philadelphia to a mile east of Antwerp, where it crosses the river and is inconspicuous for several miles. At the north side of the Antwerp quadrangle about two miles west of the east border the moraine again becomes con- spicuous for a mile and a half and is well developed in the south- east corner of the Hammond quadrangle for a mile northeast of Spragueville, where it comprises a line of hills a hundred feet or more high composed of coarse gravel with a few boulders and a sandy matrix. There are many pits in the Philadelphia moraine both on the Antwerp and Hammond quadrangles that are worked for gravel. Only traces of a third moraine, the Dekalb moraine, which is so prominently developed south of Dekalb Junction, are found in the Hammond area. The hill of sand and gravel at the north end of Pleasant Lake may well represent a part of the Dekalb moraine. Similarly about one-half mile to a mile and a half north of South Hammond there is a slope covered with bouldery drift east of the railroad, and a gravel kame or terrace west of the railroad. Relation to moraines outside region. The moraines of the St Lawrence valley as a whole have been discussed by F. B. Taylor (1924). The Albion moraine lies at the foot of the second scarp on the Carthage quadrangle at 1250 to 1300 feet altitude, and may be continuous with the Dicob moraine on the east side of the valley or with a moraine buried beneath the delta sands. The Oswego moraine is continuous with one or more of the bands of the Carthage, Devoice and Fargo belt of moraine. The Watertown, Philadelphia and Gouverneur moraines have already been referred to. Ground moraine. The ground moraine comprises deposits of unassorted material called till consisting of mixed boulders, sand and clay deposited beneath the ice both during its advance over the area and its withdrawal through melting away. On the Hammond quadrangle except on a few hilltops above 460 to 480 feet at the south and 500 feet in the central portion the ground moraine is obscured or completely buried by the silts or sands subsequently deposited from the waters of Lake Iroquois and Gilbert gulf. The till is therefore not a conspicuous element at the surface of this area. On the broad flat plains underlain by the flat lying sandstone or limestone of the northwest corner of the quadrangle evidence of the presence of the drift sheet is found in the boulders scattered here and there over the surface and often consisting predominantly of igneous rock, which must necessarily have been transported from a distance. The 46 NEW YORK STATE MUSEUM boulders have been cleared off the cultivated fields and thus show mostly in the pastureland. On the higher hills throughout the Ham- mond quadrangle boulder-littered areas are common, because the thickness of sediment deposited there from the waters of Lake Iroquois and Gilbert gulf were insufficient to bury them, as is the case in the valleys. A conspicuous but not quantitatively large element of the topog- raphy are large domical hills with no rock exposed but consisting exclusively of bouldery ground moraine at the surface. These hills, however, have the rolling topography characteristic of recessional moraines and are thought to be merely rock cores covered with a veneer of drift. Examples in the Lowville area are Beech Hill and the great hill north of Indian river along the Croghan-Harrisville road, and on the Antwerp area the flat hilltops about Fargo above 760 feet are of this type. Elliptical-shaped hills of till or unstratified drift oriented parallel to the direction of movement of the glacier are called drumlins. Drumlins were identified at only one locality on these quadrangles. One-half mile east of Sterlingville (Antwerp quadrangle) there is a hill about one-half mile long, oriented with a north-south direction, that appears to have the characters of a drumlin, built on the surface of the Potsdam table-land (figure 36). LAKES A number of lakes are found within the Hammond quadrangle. All of these are the result of the irregular deposition of glacial drift in preglacial valleys or of the sands and clays in the temporary glacial dammed Lake Iroquois and Gilbert gulf. Yellow lake and Payne lake are the result of a slightly greater deposition of stratified sands and clays in the preglacial valley at Oxbow than along the valleys now occupied by Payne and Yellow lakes. They lie in elongate depressions on the surface of the valley deposits. Red lake was formed because of the obstruction of a preglacial valley by a deposit of sand beneath the glacier. The sand hills on each side of the outlet of Red lake represent parts of the obstructing barrier. Pleasant lake lies in a valley eroded in a relatively soluble weak limestone band between gabbro on the east and granite on the west (figure 37). The limestone is exposed at each end of the lake. The valley occupied by the lake is thought to have drained north- ward to Birch creek in preglacial time. During the retreat of the ice sheet a thick ridge of drift was formed at the edge of the ice when the margin stood for a long period of time just north of the HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 47 north end of the lake. The sandy deposits slope down to the lake from the north and are believed to represent part of the Dekalb recessional moraine. Hickory lake also lies in a preglacial valley. At the south end there is a bar of sandy drift. Lake of the Woods (figure 38), Grass lake and Moon lake belong to the type whose borders are partly formed by the Potsdam sand- stone. Where this is the case, the walls are usually steep or cliffed. The basins of these lakes, however, lie in the Precambrian limestone. The amphitheater-like character of these basins has arisen through retreat of the walls of Potsdam sandstone by a process of sapping. The weak and soluble limestone and weak calcareous and fissile beds at the base of the Potsdam are more easily eroded than the over- lying resistant sandstone. This results in undermining and slumping of the unsupported beds and the retreat of the scarp face. Still better examples of this type of valley amphitheater-head (Hinds, ’25, p. 816-18) are found in the lakes near Redwood (Alexandria quadrangle) . In the eastern part of the Lowville area there are a number of small lakes and depressions in the surface of the great high-level border delta. Some of the basins have steep sides from 40 to 60 feet high. The deltas are thought to have been in part built forward over the side of a lobe of ice that lay in the Black River valley at this time. This edge must have had a very irregular surface and its subsequent burial and melting permitted the overlying deposits to slump in where the high parts of the ice surface had been. Lakes of such origin are called pit lakes. DIRECTION OF MOVEMENT OF PLEISTOCENE GLACIER The direction of movement (figure 5) of the last advance of the ice cap of Pleistocene time is readily ascertained from a study of cer- tain local peculiar and characteristic markings on the surface of the rocks. Such features are best seen on surfaces of fine-grained rocks such as the Potsdam sandstone, granite (Alexandria type), and quartzite where local areas polished by the sand and silt in the base of the ice have not yet been weathered away since the retreat of the ice. They are found sometimes on rocks of other types. Characteristic ice markings on such surfaces are striations, gouges and grooves, chatter marks and crescentic gouges. The striations and gouges are made by rock fragments held firmly in the base 4s NEW YORK STATE MUSEUM Figure 5 Sketch map showing recessional moraines and direction of ice movement. Modified from map by Frank B. Taylor, 1924 HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 49 of the ice and dragged over the underlying rock surface in the direc- tion of motion of the glacier. These are the most common markings. The chatter marks comprise a series of concentric curved vertical fractures with their axis aligned along the direction of motion of the glacier and their concave sides turned in the direction of motion of the moving ice (figure 6). There may be as many as a score or Figure 6 Chatter marks on Potsdam sandstone on road running west from schoolhouse about two and one-half miles south-south- west of Hammond. Arrow shows direction of ice motion more of such fractures arranged on a single axis and the diameter varies from an inch to a foot. The crescentic gouges are the result of the breaking out of a thin rock wedge along two fractures, one an almost vertical crescentic-shaped fracture, the other a flat fracture dipping gently in towards the vertical one. The crescentic gouges or lunar fractures are oriented with their concave surface away from the direction of motion of the ice and therefore oriented in the opposite direction from the chatter marks. The three types of glacial markings are often found in association and are particularly well shown at many localities on the Potsdam sandstone as in the vicinity of Robb School and the schoolhouse about two and one- half miles south-southwest of Hammond. Many observations on the direction of motion of the ice as deduced from a study of the markings indicate that in the Hammond quad- rangle it moved in general almost due south across the area under- lain by the Precambrian rocks and a little west of south in the area underlain by Potsdam sandstone in the northwest corner of the 50 NEW YORK STATE MUSEUM quadrangle. The direction of motion in this northwest corner is about S. 20° W., and in the rest of the Hammond area it varies between S. 50 W. and S. 150 W. The direction of ice flow which is thus indicated is much more to the south than the alignment of the topography and cuts across the topographic trend at a consider- able angle. North of Carthage the striae have a direction south-southeast but to the southeast, east of the Black river, they gradually swing more easterly until in the vicinity of Naumburg they are east-southeast. On the Lowville quadrangle the striae in the valley of the Black river for a width of four miles or more from the base of the escarp- ment on the west side of the river, all are oriented in general about east-southeast, whereas in the eastern half of the quadrangle they strike north-south. The striae in the valley are not parallel to the valley but trend considerably more easterly. This may be explained by assuming that at the time of formation of the striae in the valley the surface of the ice then had a lower level than the ice cap on the Tug Hill plateau to the southwest and that movement of the ice from this higher level at right angles into the valley tended to deflect the direction of ice movement into a more easterly direction. On Antwerp quadrangle the striae are divergent. In the north- west half of the area the striae average about S. 20° W. whereas in the southeast part, southeast of Woods Mill and Fargo they strike S. 30° E. South and southeast of Fargo they strike almost due south. The divergence of orientation of the striae is obviously related to the Rutland promontory, the high north end of Tug Hill plateau. The basal currents of the moving ice were split so that part flowed south-southwest around the west side of the plateau and part flowed south-southeast around the east side of the plateau and up the valley of the Black river. All the hilltops must have been covered and the valleys filled with ice at the time of formation of the striae. The ice main- tained its own general direction of flow irrespective of the trend of the hills except where the north end of the high Tug Hill plateau split the direction of motion. Most interesting results and sidelights are thrown on the direction of movement of the ice cap if the boulders of the glacial drift are studied and compared with their most probable sources. A recon- naissance of this problem in northern New York has been made by J. H. C. Martens (’25, p. 81-116). He finds that boulders of anorthosite are found scattered here and there all along the south side of the St Lawrence valley, whose most probable source is a HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 51 great body of rock, called the Morin anorthosite mass, north of Montreal in Canada. The writer has found such boulders on the Hammond quadrangle and along the valley of the Black river as far as several miles to the south of Lowville. In a pasture about two and one-half miles northwest of the village of Black Lake (Hammond sheet) there is a boulder of anorthosite seven feet in diameter resting on Potsdam sandstone (figure 39). This boulder was very probably transported by ice for a hundred miles from its source in the Morin anorthosite mass in Canada, and it may have taken several hundred years for it to make the journey. It is rounded and faceted by the wear which it suffered in its travels. Probably all except a few per cent of the material of the glacial drift, however, has come from the area south of the St Lawrence river. The striations on the rocks of the Hammond, Gouverneur, Lake Bonaparte and Antwerp quadrangles all indicate that the ice was here flowing in general in a southerly direction ; yet a few of the boulders of the drift indicate that they have been brought from far to the northeast. The explanation of this discrepancy is found in the fact that along the St Lawrence river the striations indicate a south- westerly flow of the ice about parallel to the valley and from this general axis the ice spread to the south across the Hammond quad- rangle and up the valley of the Black river. The boulders from Canada were thus transported southwest along the valley of the St Lawrence and then southerly and, at the south, somewhat south- easterly. A general picture of the movement of the ice in the country surrounding the Hammond quadrangle is given in figure 5. It is also possible that when the ice was at its maximum extent and thickness the direction of movement may have been in a general southwesterly direction across the Adirondacks, and that the striae that we now find and take into account were made only during the waning stages of the period of glaciation when the direction of motion may have been somewhat different. RECENT AND PREGLACIAL DRAINAGE After the retreat of the ice accompanied by the successive for- mation and destruction of a series of temporary lakes, and during the uptilting of the land the drainage once more established itself over this land surface. In many cases, however, the rivers and streams did not follow exactly the original valleys that they had occupied before the sub- 52 NEW YORK STATE MUSEUM mergence. In some cases the larger rivers found for themselves a wholly new course to the St Lawrence made up of a patchwork of preglacial valleys of other streams and postglacial courses in part across low divides. OSWEGATCHIE RIVER The Oswegatchie River is a fine example of a postglacial drainage different from the preglacial. Its course across the Gouverneur and Hammond quadrangles on its way to the St Lawrence river is a most extraordinary illogical route and can not be its preglacial valley. In going from Gouverneur to Peabody Bridge three and one-half miles north it takes a loop 263/2 miles in length by way of Oxbow, and throughout this entire length it follows preglacial valleys of smaller streams. It is a curious fact, however, that from Elmdale northeast to the border of the quadrangle the Oswegatchie is flowing in an ancient stream channel of pre-Potsdam age as indicated by the relics of Potsdam sandstone along the river banks. INDEPENDENCE AND BEAVER RIVERS On the Lowville quadrangle most of the drainage east of the Black river has been established since the glacial period. On the east side of the area the streams from the mountains, namely the Beaver and Independence rivers, debouch upon the flat surface of the high level deltas and have assumed a course consequent upon the slope and irregularities of the surface of the sand plain. There is therefore no reason why their present valleys should coincide with their preglacial valleys across this area and it is probable that they are different. Certainly these rivers in cutting down through the sands have at places found themselves superimposed upon the crests of what, if the overlying sands could be wholly stripped away, would be revealed as hills and ridges. As a result many falls are found along the present courses of these rivers such as those of Elmer falls, High falls and Beaver falls on the Beaver river and that at Sperryville on the Independence river. The headwaters of many of the smaller streams have just begun or are in the process of cutting back into the steep west-facing slope of the delta sands. BLACK RIVER Across the Lowville and Carthage quadrangles as far as Carthage the Black river meanders with a very gentle gradient across the old lake deposits, swinging from one side to the other and trenched between banks of clay and sand. At Carthage the river has cut HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 53 through the alluvial deposits and exposed a rock ridge, through which it forms a rapids with a drop of over 40 feet. From Carthage to Herring the river again flows through a wide flat flood plain of its own making and then enters a rock gorge in the Paleozoic lime- stones, which it follows as far as Deferiet. From Deferiet to just above the dam at Great Bend the river is again flowing on a flat flood plain of its own construction. Just above the dam at Great Bend the river reenters a rock gorge (figure 40) which it follows to the west border of the Antwerp sheet. The Pine Plains delta built into Lake Iroquois by the Black river buried its old valley. Consequently when Lake Iroquois was drained the river flowed out upon its delta and assumed a course consequent upon the slope of the delta surface which took it to the westward. The area inside the great loop from Deferiet to Great Bend is the old flood plain of the river across which the meander has successfully worked to the north. No bedrock is exposed here and it appears as though this area from Deferiet to Great Bend must be underlain by part of the preglacial Black River valley but from Herring to Deferiet and from Great Bend west the present channel is in a postglacial rock gorge. The former valley of Black river must have extended about northwest from Great Bend, for at Leraysville two and one-half miles northwest of Great Bend, the altitude of the lake in the valley of Pleasant creek is only 51 1 feet, whereas the altitude of the present valley two and one-half miles west of the same place is about 560 feet, or higher if the postglacial erosion is taken into consideration. Only the delta sands and clays prevent the Black river today from flowing northwest from Great Bend. Fairchild has argued (Cushing et al To, p. 141-45) that the Black river during Tertiary times flowed north and eastward around the northwestern border of the Adirondacks between the Precambrian rocks and the edge of the Paleozoics and that preceding the latest ice invasion it probably flowed northward rather than westward from the general vicinity of Carthage and crossed the present divide for north and south drainage. Cushing, on the other hand, suggests that the Black river in preglacial time followed a course to the St Law- rence, keeping west of the present divide. The preglacial course of the Black river is not obvious but it is certainly in part at least not the same as that of the present stream. INDIAN RIVER The preglacial drainage of the southeastern part of the Antwerp quadrangle appears to have flowed in general southwest into the 54 NEW YORK STATE MUSEUM Black river instead of into the Indian river as now, thus draining southwest into the Black river in preglacial time. It seems probable that the preglacial valley of the Indian river after leaving Natural Bridge (Lake Bonaparte sheet) continued southwest through the swamp north of Devoice Corners and probably westward into the Black river. The change in drainage took place after the draining of the temporary lake into which the river had built the Natural Bridge delta ; when the river flowed out across the delta which had buried its old channel and it found a new channel through the valley to the northeast. From the east border of the sheet to North Wilna the Indian river is in a narrow gorgelike channel, of a size quite out of adjust- ment to that of the river. There is no doubt that this course of the river is of postglacial origin and that it is following the former course of Bonaparte creek, which continued southwest past North Wilna to a junction with the Black river somewhere beneath what is now the Pine Plains delta. The zigzag course of the Indian river from Natural Bridge to Black lake is certainly a patchwork of several preglacial channels of smaller streams and of postglacial channels and gorges of its own making, although the actual details have not been studied. POSTGLACIAL UPWARP Accompanying and following the later stages of withdrawal of the ice sheet this region was subjected to an upwarp, greater at the northeast than at the southwest. This is well exemplified by the differential altitudes of the surface of the great compound delta of glacial Lake Port Leyden along the east side of the Black River valley. In the southeast corner of the Port Leyden quadrangle about four miles a little north of east from Boonville broad flat surfaces of the delta are at an altitude of about 1180 feet; in the southeast corner of the Lowville area similar surfaces are at about 1280 feet; and in the northeast corner of the Lowville quadrangle the altitude of broad flat surfaces of the delta is about 1350 feet, or a rise in altitude of 170 feet in 35 miles from south to north. In as much as these flat surfaces were originally formed about at the water level of temporary Lake Port Leyden, the present northward rise must have resulted from uptilting at a rate of about five feet a mile along a north-south line. All the deltas at lower levels show similar uptilting along a N. N. E.-S. S. W. line. The flats at about 770 feet south of Beaver HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 55 Falls may be compared with the surface of the Natural Bridge delta at 820 feet altitude 14 miles a little west of north. The deltas built out into Lake Iroquois and Gilbert gulf show the north-northeast upwarp very well. These have been specially studied by Fairchild (’19), from whom the following data are taken. For the deltas and beaches built in Lake Iroquois, he states that the present altitude of the Iroquois beach at Hamilton, Ontario is 362 feet, of the delta at Adams (Watertown sheet) 622 feet, of the Pine Plains delta built by the Black river near Great Bend (Antwerp sheet) 700 feet, of the Fullerville delta built by the Oswegatchie (Gouverneur sheet) 780 feet, of the Raquette river (Potsdam quadrangle) 900 feet, and the Iroquois beach at Covey hill 1030 feet, or a total post-Iroquois differential upwarp of 668 feet between Hamilton, Ontario and Covey hill in a direction about N. 20° E. The north boundary of the State was therefore as much as 668 feet lower than it is today at the time Lake Iroquois was extinguished and in addition as much lower than 668 feet by what- ever amount Hamilton has risen since the death of Iroquois. The post-Iroquois uplift of Hamilton is estimated at 72 feet so that the total post-Iroquois uplift of Covey hill is 740 feet. This means a bodily uplift of 72 feet for the whole region of northern New York in addition to the greater differential uptilting at the north. At the time of retreat of the ice front so that Lake Iroquois was drained the land stood so low that an arm of the sea extended up the St Lawrence valley. The exact distance to which the marine waters extended is still a question. Fairchild considers that they reached clear over most of the present site of Lake Ontario. Some geologists, however, question whether they extend this far. Certainly they reached beyond Ogdensburg, for marine shells have been found 30 feet above the river at Ogdensburg and at altitudes of 335 to 360 feet at Norwood (Cushing, T6, p. 60). No marine shells have been found in the Hammond or Antwerp areas beneath the levels assumed by Fairchild to have been covered by marine waters but this might be explained by the assumption that due to a great inflow of fresh water the gulf waters in this part were too brackish to support marine life. Based upon the data given by Fairchild, the altitude of the old shoreline of Gilbert gulf would be at about 550 feet in the northeast corner of the Hammond quadrangle, 480 feet in the southeast corner and 460 feet in the southwest corner of the Hammond sheet, and 410 feet at Great Bend on the Antwerp sheet. Fairchild (’19, p. 60) finds the following evidence of a shore line of Gilbert gulf in adjoining regions, a cobble bar at an altitude of 450 feet 56 NEW YORK STATE MUSEUM two miles southeast of Redwood, the delta of the Oswegatchie at Hailesboro two miles southeast of Gouverneur and cobble bars at two miles south of Richville by the Cole School at an altitude of 516 feet. At Covey hill the altitude of the latest Lake Iroquois beach is 1130 feet and of the marine beach of Gilbert gulf 740 feet, giving a difference in amount of 290 feet. Lake Iroquois is assumed to have been drained within such a short period that little or no uplift took place during its destruction. Consequently, this would make the altitude of the level of Lake Iroquois 290 feet and then there would be a constant interval of this amount between the late beaches of Lake Iroquois and the marine beaches of Gilbert gulf. If the altitude of Lake Iroquois at the time of its extinction was 290 feet then the post-Iroquois uplift at Great Bend would be 410 feet. Fair- child estimates that in addition there was a 6o-foot uplift at Great Bend, or a total uplift of 470 feet since the removal of the ice from that locality. IGNEOUS ROCKS METAGABBRO AND AMPHIBOLITE Practically all of the gabbro in the northwest Adirondacks has been so altered that it may appropriately be called metagabbro. The original pyroxene has been, in part or in whole, changed to hornblende. The result is a rock that in all its characters may resemble an amphibolite such as is believed to have resulted from the replacement of limestone or the recrystallization of impure lime- stones. Hence unless some evidences of intrusive relations to the country rock or some definite igneous textures, such as phenocrysts, are found, it may be impossible to determine whether a given rock is to be classed as metagabbro or as amphibolite resulting from recrystallization or replacement of sediments. On the Hammond sheet most of the bodies in question occur within the Grenville formations with apparent intrusive relations and are therefore meta- gabbro. On the Antwerp and Lowville sheets, however, where the rock occurs within larger bodies of younger intrusives, it is in many cases difficult or impossible to decide definitely whether a given mass is metagabbro derived from an igneous rock or amphibolite derived from a member of the Grenville series. Because of the difficulty or impossibility of making a positive decision as to the precise mode of origin of the amphibolite bands they have all been mapped together under the same symbol. In the Lowville quadrangle long narrow bands or lenses of meta- gabbro or amphibolite occur in an arc that sweeps in general from HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 57 Bush’s Landing, west of Kirschnerville, to southwest of the town of Indian River. A band composed of relics of amphibolite in augite syenite also extends northeast to Bush’s Corners from Petrie’s Corners. There are many lenses of such rock included in the granite that are too small to show on the map. They are typically exposed between Strifts School and Croghan and northwest of Beech hill. They vary in thickness from several feet to a hundred feet but can not be traced far along the strike. The relations indicate that granite and syenitic magma intruded a large belt of gabbro or amphibolite and dismembered it into many separate bands and lenses. Where the granite is full of such amphibolite layers, pegmatite veins, often slashed with quartz, are especially common. A considerable part of these ferromagnesian-rich bands show a sparsely porphyritic texture with large labradorite phenocrysts, such as the lenses south of Henry School, and are definitely metagabbro. A part of the bands show no features that enable one to say positively whether they were derived from gabbro or Grenville formations. On the Hammond quadrangle the largest single area of metagabbro is south and east of Pleasant lake. Elsewhere the metagabbro and amphibolite form long narrow bands or small lenses. One band lies along the east side of Yellow lake, where it is shredded and disin- tegrated by granite and extends northeast, passing three-tenths of a mile northwest of Griffith School. Another belt extends south- west from Scotch Settlement School. Several small lenses occur near Wegatchie, southwest of Farley School, and near Laidlaw School. The metagabbro, in general is a gneissoid, medium-grained, dark rock. The plagioclase varies from andesine to labradorite and the rock from olivine gabbro to gabbro-diorite. In the Lowville area part of the gabbro carries sparse phenocrysts of labradorite averag- ing an inch in diameter. These are oriented parallel to the foliation. The bands with the phenocrysts appear to grade into bands without such crystals. The most common metagabbro consists of about half ferromagnesian minerals and half plagioclase. The ferromagnesian mineral was in most cases originally predominantly pyroxene, although olivine is locally present and brown biotite may form up to 8 per cent or so of the rock, apparently as a primary mineral. In most of the metagabbro, however, the pyroxene has been partly or completely altered to hornblende. The plagioclase commonly shows a few antiperthitic rods of orthoclase. Ilmenite, magnetite and apatite are common as accessory minerals. Where the rock is much altered by late hydrothermal solutions, biotite may be found as a 58 NEW YORK STATE MUSEUM replacement of hornblende or plagioclase. The gabbro about one- third of a mile south of Henry School on the Lowville quadrangle consists of sodic labradorite, augite, hypersthene, olivine and horn- blende with accessory biotite, ilmenitic magnetite and apatite. The hornblende is secondary after pyroxene. The metagabbro just south- east of Pleasant lake on the Hammond quadrangle consists of about 45 per cent sodic labradorite, and 50 per cent pyroxene almost completely altered to hornblende, with accessory ilmenitic magnetite, apatite and biotite. The structure of the gabbro masses is exceedingly variable. Some of the small masses intruded in limestone have been completely pro- tected by the more easily deformed limestone and have a texture resulting wholly from normal processes of crystallization without a trace of deformation or of evidence for crystallization under stress. The gabbro masses of the Lowville sheet are predominantly massive with a polygonal shape to the constituent grains, a xenomorphic granular texture, probably of crystalloblastic origin. The Pleasant Lake (Hammond sheet) metagabbro mass consti- tutes a portion of a sill in the trough of a syncline. It has been markedly brecciated throughout by pegmatite dikes but the individual blocks themselves commonly show no results of crushing. Their texture, when examined with the microscope, is found to be that resulting from normal processes of crystallization and there is little or no evidence of deformation. As a rule thin sheets of the metagabbro and in particular the outer borders of the metagabbro masses and narrow bands included in granite are severely crushed and injected by granitic pegmatite so as to form an arteritic gneiss or migmatite. The belt of metagabbro west of the north arm of the Oswegatchie river consists of a hornblende gneiss with arteritic injection of a gra- nitic material. The hornblende gneiss has resulted from a most intense cataclastic mashing of the metagabbro with practically no chemical alteration except the development of a little biotite. Anorthosite, gabbro-anorthosite, and gabbro. At the south border of the Antwerp quadrangle, about two miles south of Fargo, there is a small area of greenish rock that varies from anorthosite to gabbroic-anorthosite, and locally contains narrow sills of gabbro rich in pyroxene and ilmenite magnetite. The anorthosite consists of coarse green plagioclase crystals. Much of it shows large crystals of plagioclase from an inch up to six inches in length in a finer grained groundmass. Locally the weathered surface has a very coarse porphyritic aspect. In the fields southwest HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 59 of the first schoolhouse north of Carthage on the Gates Corners road the anorthosite shows a very coarse breccia structure consisting of subangular to subrounded blocks of almost pure greenish white anorthosite, and rarely of gabbro, from one to five feet in diameter in a groundmass of gabbroic-anorthosite. The latter is much coarser, with crystals of plagioclase one to four inches in diameter, and with more interstitial pyroxene than the included blocks. The gabbroic- anorthosite appears here to bear an intrusive relation to the included blocks. A typical specimen of the coarse anorthosite consists of about 90 per cent andesine-labradorite with associated augite and hypersthene, and accessory titanite, magnetite, and apatite. The plagioclase shows an antiperthitic intergrowth of orthoclase rods, particularly on the borders of the clear plagioclase cores. Another specimen showed plagioclase as calcic as Ab 43 An 57. The gabbroic- anorthosite facies is similar except for an increase in the amount of pyroxene. Several bands of gabbro up to 250 feet or more in width, in part very rich in pyroxene with associated ilmenitic magnetite, occur as intrusions in the anorthosite and gabbroic-anorthosite parallel to the foliation. A specimen of one such pyroxene-rich band consists of pyroxene and a little plagioclase (andesine) with considerable ilmeni- tic magnetite reticulating as fine veinlets through the pyroxene, and to a lesser extent through the plagioclase. Apatite is an abundant accessory mineral, and there is a little biotite. The pyroxene includes the monoclinic and orthorhombic varieties. Locally small portions of the gabbroic facies of the anorthosite contain abundant grains of titanite, are quite irregular in texture, and suggest the assimilation of included limestone blocks. The whole mass is cut by numerous narrow granite dikes, and thin sections show many planes of cataclastic crushing. DIORITE AND QUARTZ DIORITE (HAMMOND SHEET) Introducton. Within a radius of four and one-half miles of Rossie (Hammond sheet) there are many masses of gray medium to fine-grained igneous rocks that range in character from diorite to quartz diorite with local variations towards monzodiorite. Another belt of dioritic rocks occurs in the vicinity of Moon lake ; and along the east side of Black lake there are many small lenses of diorite. A band of biotite-oligoclase-quartz diorite runs across the central part of Antwerp quadrangle. 6o NEW YORK STATE MUSEUM In general these masses appear to have been intruded in irregular sill or lenselike form conformable to the bedding of the Grenville formations, although locally with transgressive relations, and to have been subsequently subjected to stresses or folding along with the inclosing country rock. The dioritic intrusives are definitely younger than the gabbroic masses and are older than the syenitic and the granitic intrusives. About one mile southeast of Rossie is a gabbro mass that is markedly intruded by fine-grained gray diorite. The intrusive relations of the syenitic and fine-grained aplitic granitic rocks to the diorite are shown in the mass of diorite northwest of Grass lake and the intrusive relations of the porphyritic granite to the diorite are exhibited in the diorite mass on the extreme west border of the quadrangle one and three-quarters miles southwest of South Hammond, and in the large dioritic masses near the village of Rossie. The term “dioritic rocks” will be used here to include both the true diorites, quartz diorite and potassic feldspar-bearing facies (monzodiorite). The dioritic rocks of the Hammond quadrangle are gray and medium grained. On the fresh surface the plagioclase has commonly a greenish hue and the rock often resembles in general aspect the augite syenites of the country southeast of this quadrangle. In many of the masses the presence of biotite is a conspicuous and characteristic feature. In many cases it is difficult or impossible in the field to distinguish between the diorites and the gabbros on the one hand and between quartz diorite and granite on the other. Since most of each mass has not been checked by microscopic examination it is probable that some mistakes in mapping have been made. The mineral composition of representative types of the dioritic rocks of the Hammond area is given in the table on page 65. The plagioclase in all these rocks is characterized by a small amount of microscopic spindles or grains of orthoclase in antiperthitic inter- growth. In general, as the percentage of quartz increases there is a decrease in the amount of hornblende or pyroxene and a relative increase of biotite. In the most siliceous facies biotite alone is present to the exclusion of pyroxene and hornblende. Apatite and magnetite are constant accessory minerals in all the various facies. Zircon is present in the quartz diorites but is very rare in the diorites. Locally in some of the smaller masses of the Hammond area biotite is found partly replacing the pyroxene and in much mashed rocks secondary biotite is a conspicuous development replacing any HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 6l or all of the other minerals. It is usually the case also that in deformed facies of the dioritic rocks the hornblende and plagioclase form a symplectic intergrowth for a narrow zone along their mutual borders due to granulation and recrystallization under stress. In some of the quartz diorite the biotite has been partly altered to chlorite with the separation of magnetite, which now occurs dis- seminated through the chlorite. The diorites of the Hammond quadrangle appear to have been originally biotite -pyroxene diorites. In most cases the pyroxene is now, however, partly or completely altered to hornblende. In some of the isolated masses of diorite that have not suffered crushing and alteration there is no hornblende, as in the case of the mass one and one-half miles southwest of Conger island in Black lake, whereas in all the specimens of diorite that were examined from the Rossie mass there is no pyroxene, and in many of the diorite masses both pyroxene and hornblende are present. Where both pyroxene and hornblende are present the hornblende is in part definitely replacing the pyroxene and it seems probable that much of the horn- blende has had this origin. In the case of the diorite facies of the Rossie mass it can not be stated positively that the hornblende has had such a derivation as no residual pyroxene was found, but the extent of alteration of the pyroxene to hornblende is in general very definitely related to the degree of crushing and deformation which the rock has undergone. The greater the crushing the more com- plete the alteration of the pyroxene to hornblende. The Rossie diorites are much mashed and hence any pyroxene originally present might have been expected to be altered to hornblende. The alteration of the pyroxene to hornblende has released magnetite, which usually occurs as rims locally on the borders of the hornblende grains or along the cleavages of the hornblende. Rossie mass. Within the vicinity of the village of Rossie there is a considerable mass of gray igneous rock that extends to the north- east as two folded sheets with some minor associated bodies. The mass is of composite nature and comprises hornblende diorite, biotite-quartz diorite, and medium-grained granite associated in such fashion that to map them separately would take much more detailed study than has been given to it. Southwest of Rossie the mass is largely granite, and is described with the granites (Hermon type), except the eastern border, which is quartz diorite. Diorite forms much of the mass north of Rossie between the Indian river and Farley School, and quartz diorite forms most of the sill between 62 NEW YORK STATE MUSEUM Brasie Corners and South Woods School, although there are sheets of granite interbanded with the quartz diorite here. Diorite, quartz diorite and granite are all found in the portion southwest of Pleasant lake. About a mile east of Rossie there is a mass of metagabbro intruded by diorite. The two rocks were not mapped separately in the field and hence the relations shown on the map merely indicate the intrusive relations of the diorite to the gabbro. Near Bostwick creek there is a folded sill of aplitic granite (Alexandria type). This intrudes the dioritic rocks. All the igneous rocks are seamed with irregular veinings of granitic pegmatite that often carry small segre- gations of quartz and tourmaline. There is a tendency for the peg- matite veins to string out parallel to the gneissic structure although in a very irregular fashion. The plagioclase of a typical specimen of hornblende-biotite diorite three-quarters of a mile north of Farley School and of a typical specimen of biotite-quartz diorite, one mile northeast of Rossie is andesine-oligoclase or sodic andesine. South Hammond mass (quartzose diorite). East of South Ham- mond on the east side of Black Creek swamp there is a mass of quartz diorite of considerable size. The weathered surface is usually gray with an indistinct purplish or pink hue, locally greenish or even almost white near the borders of the mass. It is criss-crossed by granitic pegmatite veins, which are not usually over a few feet apart although generally only a few inches thick. Pegmatite veins larger than a foot in width are rare. Many of them carry a little black tourmaline. Aplite veins are also present. This diorite mass is conspicuous for its lack of inclusions of the country rock. They are rare, and when present are rarely over a foot in length. The rock of this mass appears to have been originally a biotite- pyroxene quartz diorite. As a result of secondary alteration a part or all of the pyroxene has been changed to hornblende. The average mineral composition of four specimens is given in the table, in column 5. Quartz and orthoclase are interstitial to the other minerals and except on the borders of the mass the quartz is not in spindle form. Apatite is always present as an accessory mineral although never present in noteworthy amount. Zircons are present but rare. The plagioclase has a very slight amount of orthoclase in antiperthetic intergrowth. In a typical specimen it is Ab72An28. The hornblende is secondary after pyroxene and the grains are usually accompanied by a rim of secondary magnetite resulting from the alteration. Moon Lake diorite. The belt of rocks between the Red Lake granite and syenite on the northwest and the limestone of Vrooman HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 63 Creek valley on the southeast is very difficult to map accurately because of the intimate intermingling of many different kinds of rock and because of mixed and variational facies amongst the intru- sives themselves. In general it consists of a mingling of impure crystalline nodular limestones, diorite, quartz diorite and granite with one or the other predominant. There is probably also some meta- gabbro present. The dioritic rocks are intruded and locally per- meated by granitic material giving rise to mixed gneisses. South of the large Potsdam sandstone area one mile northwest of Hunt School the predominant rock is a feldspathic diorite varying in com- position from 70 to 80 per cent plagioclase, o to 9 per cent quartz, 2 to 15 per cent hornblende, 1 to 7 per cent biotite, a trace to 6 per cent magnetite, o to 10 per cent microcline, and one per cent or more of apatite. The unusually high percentage of magnetite and apatite together with the quartz and microcline were probably introduced in part by solutions arising from the intrusive granitic magmas. Locally a little pyroxene is present. The diorite of the Moon Lake belt is considerably more feldspathic than the average. A specimen of fine-grained, gray granitic rock forming much of the peninsula on the south side of Moon lake is a quartz diorite consisting of about 45 per cent plagioclase, 35 per cent biotite, and 20 per cent quartz. The foliation of the diorite arises from the segregation of the ferromagnesian minerals in seams and the dimensional orientation of the minerals. Usually there is no cataclastic or protoclastic structure due to protection afforded by limestone layers. Locally, however, there is distinct cataclastic structure, shown especially well in some of the pegmatitic injection. Smaller intrusive masses. The small intrusive masses belonging with the dioritic suite of rocks are usually diorites. No quartz diorite was found in the specimens examined from the small lenses. Two specimens of monzodiorite were found, one from the north end of the mass west-northwest of Grass lake and the other at the extreme west edge of the quadrangle one and three-quarters miles southwest of South Hammond. The folded diorite sill northeast of Brasie Corners is a hornblende-biotite monzodiorite with about 12 per cent orthoclase. The diorite mass about one mile northwest of Rossie is famous for the contact metamorphic minerals developed along its borders and has been described by Smyth (’96, p. 260-70) and Agar (’2 3, p. 139-48). The titanite must be due to the effect of dis- 64 NEW YORK STATE MUSEUM solved lime from the adjoining limestone. The plagioclase varies from andesine to oligoclase-andesine. The percentage of accessory minerals such as apatite and mag- netite is often much greater in the small intrusive bodies than in the larger and may be due in part to introduction from without by solutions given off by deeper seated portions of the magma from which it itself was derived. Microstructure. The diorite and quartz diorite of the Rossie mass has a foliation that is due largely to a parallel dimensional orientation of the mineral constituents and to some extent to a partial segregation of each kind of mineral into parallel seams. In much of the quartz diorite the quartz is in irregular amoeboid- shaped grains and not in flat leaves such as are so common in the protoclastic granite (Hermon type) gneisses. In most of the rock the feldspars show a little peripheral granulation due to crushing and the quartz shows a wavy ribboned structure but protoclastic structure is practically absent and pronounced cataclastic structure is confined to long narrow crush zones. The diorite northeast of Moon Lake is a sill, compressed and some- what folded, but there is no protoclastic structure and but very little cataclastic effects, although strain shadows may be present in both the feldspars and quartz. The diorite mass east of South Hammond shows pronounced cataclastic structure in the northern part but in the southern part is quite massive without any granulation of the minerals. It seems likely that the large uniform mass of the southern part acted as a unit in resistance to the stresses and did not suffer internal crushing. Most of the small separate lenses in the Grenville gneisses show moderate to pronounced cataclastic crushing. There is no evidence whatever of protoclastic structure in any of the dioritic rocks. The gneissic banding appears to have originated wholly as a result of flowage within the intruding magma and to have been but slightly affected during their period of intrusion and consolidation by external forces. The orogenic pressures that folded and induced some crushing in the dioritic intrusives were subsequent lo the intrusion of the syenitic magmas and came at a time when the dioritic rocks were probably completely consolidated. HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 65 Dioritic and quartz dioritic rocks, Hammond quadrangle 1 2 3 4 5 6 7 8 Plagioclase 56 66.5 62 74 64 57-5 60 58 Microperthite or orthoclase 1 1 2 3 5 3 Quartz 3 5 • 5 3 9 19 21 i . 5 Pyroxene 9 12.5 4 2-5 Hornblende 13-5 21 10 1 1 6 24 Biotite 8-5 7 16.5 4 12 8-5 15 14 Magnetite 1 2.5 3-5 2 1 i-5 1 Present Apatite 1 1 Titanite 2-5 1 Average composition of samples from two small masses of diorite (monzodiorite) near west border of quadrangle 2 Average of three samples of diorite from the Rossie mass 3 Biotite-pyroxene diorite from mass one and one-half miles southwest of Conger island in Black lake 4 Average of five specimens of diorite from Moon Lake belt 5 Quartzose hornblende-diorite 6 Average of four specimens from South Hammond quartz diorite mass 7 Average of five specimens of quartz diorite from Rossie mass 8 Small mass of diorite one mile northwest of Rossie BIOTITE-OLIGOCLASE QUARTZ DIORITE (ANTWERP QUADRANGLE) A belt of gray biotite-oligoclase quartz diorite, a little more than two miles wide, extends across the northeast part of the Antwerp quadrangle. It may be traced to the northeast across the northwest corner of the Lake Bonaparte quadrangle, and part way across the Gouverneur sheet. As thus exposed, it has a length of about 18 miles and a width of one-half to three and one-half miles. On the Gouverneur quadrangle it has been mapped by Cushing as a grani- tized facies of amphibolite.1 The rock contains a few inclusions of amphibolite up to several feet thick. It is cut out at each end by granite (Hermon type), so that its original extension may have been greater. In the Antwerp quadrangle it is intruded by granite sheets and, intensively, by pegmatite veins. Where there is the smallest amount of granitic injection the rock is very dark gray to almost black at the surface, and in its uniformity and homogeneity appears to be certainly an igneous rock. In pro- portion as the granitic injection increases, the rock is lighter colored. The pegmatite veins usually vary from paper thinness to a few inches in width. The thicker veins are more common where very close Other geologists have suggested that it may be a metamorphosed sediment. 66 NEW YORK STATE MUSEUM to the bordering granite mass. The pegmatite veins form lenses where parallel to the foliation, but where they cross the struc- ture they have a characteristic zigzag or crenelate pattern or are in complicated puckers. Little pegmatitic and drusy seams are common, and carry quartz, tourmaline, muscovite and feldspar. In thin section the rock is seen to be composed of oligoclase, quartz, biotite and a little potash feldspar, with accessory muscovite, apatite and zircon. The biotite is strongly pleochroic from a deep greenish brown to a pale yellow-green, and is quite unlike the biotite of any other rocks in the district. A little muscovite is in parallel intergrowth with it. Biotite may form 15 to 20 per cent of the clean facies. Little veinlets of microcline and quartz, with associated muscovite and tourmaline, however, are common in most of the rock. The quartz diorite of the Antwerp quadrangle is very similar to that forming part of the Rossie mass on the Hammond quadrangle, and probably belongs to the same series of intrusives. SYENITE— GENERAL SUMMARY The term “syenite” is here used for rocks composed essentially of alkalic feldspars with one or more of the common mafic minerals, and they may contain quartz up to a maximum of 10 per cent. ; Locally, facies with more than 10 per cent quartz have been included with the syenites in geologic mapping where it has been impracticable to map them separately. The syenitic rocks as thus defined grade into quartz syenite or granosyenite with which they are associated. Syenites predominate in a belt including the southeast corner of the Antwerp quadrangle, the northeast corner of the Carthage area and the northwest part of the Lowville sheet. These all have the appear- ance of augen gneiss, in part with a coarse porphyritic texture, and cataclastic or protoclastic structure. These syenitic facies range from green augite-hypersthene syenite tot pink quartzose hornblende sye- nite. Syenite of an equigranular texture, occurs in the Lowville area within a radius of a few miles of the village of Indian River and in the southeast part of the quadrangle. The syenite of this type varies from a green augite-hypersthene type to a pink quartzose horn- 1 blende syenite, and the texture is probably crystalloblastic. Both the porphyritic and equigranular types of syenitic rocks show an interbanding of the different closely related variant facies. In all cases the foliation and banding are parallel. Within the broad belt of predominantly Grenville beds of the Hammond and Antwerp sheets there are also outlying sheets and masses of syenite. These are probably equivalent to the Alexandria HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 6 7 syenite described by Cushing (’io, p. 182-83) from the Alexandria quadrangle. The syenite bands of the main igneous complex are in part similar to the outlying syenitic sheets in the Grenville series of the Hammond and Antwerp areas but there are certain differences. Rock similar to the augite syenite of the main igneous complex is not found in the outlying intrusive sills, with the single known exception of a small mass near Theresa described by Cushing (To, p. 38). The pink syenite of the main igneous complex and the outlying intrusives also vary in the character of their feldspars and ferromagnesian minerals. In the outlying masses the feldspars are usually differentiated into microcline and plagioclase, whereas in the. syenite bodies of the main complex the feldspars exist for the most part as a perthitic intergrowth of the two minerals. Biotite is rare in the syenite and much of the granosyenite of the main complex and scarcely appears in rocks with less than 68 per cent SiC>2, whereas it is common to predominant in the larger masses in the Grenville beds even where the SiC>2 is as low as 56 per cent. Outlying syenite sheets in Grenville series; Hammond and Antwerp quadrangles. On the Antwerp quadrangle there are three masses of the syenitic rock — a narrow sill north of and parallel to Deerlick creek, a sheet about one and one-quarter miles south of Philadelphia parallel to the Indian river and a mass on the western border of the area about two miles west of Philadelphia. Six sills of pink to red syenitic rock are found in the north- western half of the Hammond quadrangle — northeast of Red lake ; on the west side of the Oswegatchie river about three-fifths of a mile east of Yellow lake; just east of Pleasant lake; north of Pope Mills; southeast of Lake of the Woods; and a mile west of Grass lake. Another band comes in from the Gouverneur quadrangle along the southeast side of Beaver creek. The rock appears to be predominantly a syenite augen gneiss with a considerable percentage of plagioclase, but locally in small masses it is predominantly potash feldspar. Locally, also, where the intrusive mass is small the rock is equigranular without any evi- dence of porphyritic structure. In the larger bands the ferromagnesian mineral is often biotite, whereas in the smaller and some of the larger masses both hornblende and biotite are present and in the smallest bodies hornblende alone occurs. The accessory minerals include quartz, magnetite, apatite, titanite and zircon. Apatite is in general more abundant and titanite is conspicuously more common than in 68 NEW YORK STATE MUSEUM the dioritic rocks or in the fine-grained granites. Much of the syenite has a messy heterogeneous appearance, with abundant small interstitial patches or lenticles richer in hornblende, biotite or locally scapolite. These darker portions are believed to be relics of shreds of schist resulting from the almost complete disintegration of bands of metamorphosed Grenville. The hypothesis that the syenitic rocks are in part the result of reaction between the granite (Hermon type) and metamorphosed Grenville sediments has been given serious con- sideration by the author, but although there is much in favor of this idea, some of the evidence seems to be against it. Several small isolated lenses of syenite are found in the limestones. One occurs about one-half mile southwest of Rapids School and another near Black lake about one mile west of Stark School.. The latter consists of about 50 per cent microperthite, 32 per cent plagioclase and 17 per cent hornblende. Quartz, magnetite, apatite, zircon and titanite are accessory minerals. Both these masses show cataclastic structure. Red Lake belt. The syenite northeast of Red lake is a moderately coarse porphyritic gneissoid rock with many masses and narrow inclusions of amphibolite. Along the northwest border is a band of amphibolite with intrusive sheets of both porphyritic syenite and porphyritic granite. Also along the northwest border some por- phyritic granite intimately associated with the syenite has been mapped with the latter. Along the southeast border the syenite con- tains a small amount of quartz. Seams and small veins of pegma- tite, often with black tourmaline, are common in the syenite. Narrow sheets of aplitic granite also occur in the syenite. The rock in general is a syenite augen gneiss. The mineral composition of a specimen from the north end of the mass consists of 43 per cent oligoclase, 35 per cent microcline, 9 per cent quartz, 8 per cent biotite, 2 per cent each of titanite and apatite and one per cent of magnetite. There are also a few minute zircons. The biotite is considerably chloritized. The structure is pronouncedly protoclastic but no cata- clastic effects are evident. Southeast of Lake of the Woods. The syenite mass southeast of Lake of the Woods is intimately associated with granite in such a fashion that the relations of one to the other has not been defi- nitely determined. The eastern band of the syenite is full of inclu- sions of gray gneiss and itself is banded gray and pink. The syenite is a coarse pink augen gneiss with a very pronounced cataclastic structure. The augen are microcline and oligoclase. Hornblende HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 69 may form up to 15 per cent of the rock and it is considerably altered and chloritized. Biotite flakes of secondary origin are common along the foliation planes. Some of the syenite contains a little quartz. Accessory minerals are abundant apatite grains and zircon, titanite and magnetite. Oswegatchie band. The syenite west of the Oswegatchie river forms a narrow band intruded in limestone. Its relations to the granite on the west were not ascertained due to the presence of an intervening limestone band and to covered areas. The southern portion of the mass is equigranular and wholly unmashed, whereas the northern half is severely crushed and has the appearance of a gneiss. The weathered surface of the syenite is pink but in a road cut whert the fresh rock is exposed it is green. One specimen examined from the southern portion proved to be a syenite with about 25 per cent oligoclase, 5 per cent chlorite, 2 per cent quartz and accessory magnetite, apatite and zircon. Bands of syenite also occur in the mixed gneiss between Yellow lake and the Oswegatchie river. Pleasant Lake mass. Just east of Pleasant lake there is a small mass of pink syenite intruded into metagabbro. This syenite weathers with a peculiar pitted surface. The syenite is relatively clear of inclusions of the metagabbro but is crossed by tourmaline- bearing pegmatite veins. The rock is an augen gneiss consisting of pink to white crystals of oligoclase and microcline averaging about one-half inch in length with a dark interstitial matrix of hornblende and biotite with associated accessory minerals including apatite, magnetite and titanite. Quartz is also present locally as an accessory mineral. The structure appears to be protoclastic. Pope Mills mass. Around the borders of the swamp northeast of Pope Mills there is a belt of pink to red syenite. Along the southern border the syenite is intrusive into a belt of pyroxenic amphibolite with which it forms a migmatite. There are also many bands of gray migmatite on the northern border and within the body of the pink syenitic rock ; and rarely a mass of pyroxenite repre- senting the metamorphosed limestone is present. Where the syenite occurs in large masses unmixed with the country rock, it is of very uniform character and has practically no pegmatite or quartz veins. The migmatite bands, however, are cut by granitic pegmatite and aplite veins. The syenite is so intensively mashed in all the specimens studied that it can not be satisfactorily studied with the microscope. Exam- ;o NEW YORK STATE MUSEUM ination of several sections suggests that it varies from a monzonite to syenite with microcline microperthite and plagioclase both forming small residual augen in a granulated and powdered matrix of the same minerals. The amount of ferromagnesian minerals is usually not more than 5 to 10 per cent and consists of biotite and some chloritized mineral which can not be identified but may have been biotite. Quartz and magnetite are the most common accessory min- erals, although apatite, zircon and titanite are common. The syenite is very intensively mashed as seen under the micro- scope and faulting on a small scale is conspicuous in the field. The structure, however, is in considerable part probably protoclastic. The rock is often of such a uniform color that the augen character is indistinct on the fresh surface and is more apparent under the microscope. West of Grass lake. The narrow band of red syenite augen gneiss about a mile west of Grass lake contains many included bands of gray Grenville gneiss, and itself is intruded by sheets of aplite. At the northeast end the syenite is intrusive into the band of diorite. Adjacent to the western border is a band of relic injection gneiss. Beaver Creek mass. A belt of syenitic rock extends southwest from the Gouverneur area along the southeast side of Beaver creek. This sill contains abundant layers and relics of schist high in biotite or hornblende. Locally a ghostlike breccia structure is observed in the syenite. It is as though the syenitic magma had literally shredded a belt of Grenville and disintegrated it in large part to a very fine degree. The result is a heterogeneous looking rock which might have been mapped as a mixed rock, although the syenite does predominate. There are also included bands of Grenville rock thin layered with limestone, quartzite and gneiss. The porphyritic aspect of the syenite appears to have arisen through a process of permeation of Grenville schists by magmatic solutions and resulting reaction and replacement. Tourmaline-bearing pegmatite veins are found here and there. The rock of Beaver creek mass varies from a syenite high in microcline to a monzonite or monzodiorite very high in plagioclase. The average rock has about equal parts of plagioclase and microcline. Locally granitic facies are present. Biotite is the most common and abundant ferromagnesian mineral and varies from o to 20 per cent. Hornblende may accompany the biotite but locally is the only ferro- magnesian mineral. Occasionally microperthite is the predominant feldspar. Some of the syenite is remarkable for the presence of scapolite and calcite of late primary crystallization forming interstitial HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 71 grains and making up several per cent of the rock. The calcite has probably resulted from solution and recrystallization of a little Grenville limestone and the scapolite from reaction with limestone. All the rocks carry abundant titanite, apatite and magnetite as accessory minerals. Titanite is very abundant and may form a few per cent of the rock. Zircons are sparse. Epidote is also common in small grains as an accessory mineral. Locally the micro- structure is massive without evidences of crushing but most of it shows a slight to moderate granulation of the feldspars along the borders of the grains. Sheets on Antwerp quadrangle. The belt about one and one- quarter miles south of Philadelphia is a quarter of a mile wide and exposed for several miles in length. It lies along the north border of a belt of porphyritic granite. Some bands of the syenite are quartzose and intermediate in composition between the normal granite and normal syenite. The rock is coarsely porphyritic with eye-shaped granulated feldspar phenocrysts up to an inch in length. It con- sists of about equal parts of plagioclase and microcline with a little quartz and biotite and accessory titanite, apatite, magnetite and zircon. The syenite west of Philadelphia is medium grained. The band north of Deerlick creek consists of coarse pink porphyritic syenite. All the syenite in the Grenville series of the Antwerp quadrangle shows protoclastic structure. SYENITE IN MAIN IGNEOUS COMPLEX Porphyritic augite syenite. Porphyritic augite syenite gneiss out- crops in narrow belts in the northwestern part of the Lowville quad- rangle and the adjoining parts of the Carthage and Antwerp sheets. It is a typical coarse augen gneiss in character, and is continuous with or similar in every way to the corresponding rock of the Lake Bonaparte quadrangle. It grades imperceptibly into the porphyritic hornblende syenite gneiss, and boundary lines can only be drawn approximately. The augite phase is usually green whereas the horn- blendic phase is usually red but this is not uniformly true. The dominant mineral is microperthitic feldspar, often inclosing a core of plagioclase or associated with plagioclase, the latter usually oligo- clase. The ferromagnesian mineral is a deep green, nonpleochroic monoclinic pyroxene locally altered to secondary hornblende. In some facies hypersthene is also present. Primary hornblende increases rapidly in the transition zones into the hornblende syenite. Magnetite and quartz are present as major accessory minerals with 72 NEW YORK STATE MUSEUM apatite, zircon, and locally pyrite and titanite as minor accessories. The apatite and magnetite tend to be associated almost wholly with the pyroxene. Porphyritic hornblende syenite. Like the porphyritic augite syenite this rock forms narrow belts in the area bordering the junction of the Lowville, Carthage and Antwerp sheets and is con- tinuous with or similar to the corresponding rock of the Lake Bona- parte quadrangle. As in the porphyritic augite syenite gneiss, microperthite forms the dominant mineral and occasionally the only feldspar. Plagioclase, usually oligoclase, is present in variable amounts and often forms a core at the heart of the microperthite feldspars. Hornblende is always the dominant ferromagnesian min- eral except at the border of the augite syenite masses ; and in typical specimens is present to the exclusion of pyroxene. Quartz is always present and varies from occasional grains up to io per cent, usually near the upper limit. Augite of a character similar to that in the augitic phase is present, locally, as an accessory mineral. Magnetite, apatite and zircon are omnipresent accessory minerals, and in places quite abundant. Locally an occasional grain of horn- blende is partly altered to biotite. Equigranular augite syenite (Lowville sheet). This rock con- stitutes narrow bands north and northeast of Monnatt school, a narrow band running northeast from Kirschnerville, and two bands in the southeastern part of the Lowville sheet. The syenite of the extreme southeastern corner is a medium- grained rock, green on fresh surface, weathering brown; a gneissic structure is usually very strongly developed and increases in its intensity going southeast on to the Number Four quadrangle, although at the edge of the sand plain it is practically massive. This gneissoid appearance is due largely to the segregation of different minerals into thin lenticular sheets of alternating ferromagnesian and feldspathic character. This gives the rock a characteristically finely streaked appearance. Narrow granitic pegmatite veins carrying a large percentage of coarse magnetite grains are common, cutting the syenite. In thin section the rock is seen to be typically granitoid in texture without any traces of crushing. An estimation of the mineral com- position by the Rosiwal method is given later. The augite syenite that forms the band running northeast from Tucker School through Crystal Dale, and which after an interruption by granosyenite reappears along the central portion of HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 73 the eastern side of the sheet, is very similar to that just described. In its gneissic structure the rock varies from very highly gneissoid to almost massive. In the field it appears very similar in character to the hornblende syenite that borders it on the northeast and the boundary line indicated on the map is only a very rough approxi- mation. The hornblendic rock is finer grained and is interleaved with reddish granite and granosyenite bands. Microperthite forms from 70 to 80 per cent of the typical augite syenite, augite from 4 to 10 per cent and quartz about 10 per cent. Hornblende is one of the principal minerals in some of the rock but much of it is often secondary after the augite. Magnetite, apatite and zircon are always present as accessory minerals and occasionally plagioclase (oligoclase). The mineral composition of two representative facies is given in the accompanying table (columns 1 and 2). The augite is a deep nonpleochroic grass green to emerald green variety, frequently twinned parallel to 010. It usually forms very irregular-shaped grains often with frayed borders and many embay- ments, typical results of resorption. Quartz occurs interstitial to the other minerals and also forms small circular inclusions in the feld- spars. The grain of the rocks vary from 1 to 2 mm in diameter. The augite syenite of the band east of Indian river contains a little hypersthene in addition to augite, considerable plagioclase and microperthite. Equigranular hornblende syenite (Lowville sheet). Northeast and southeast of Indian river (Lowville sheet) there is a belt of medium-grained hornblende syenite. It is predominantly pink but has local bands of a greenish hue. The typical rock as seen in the hand specimen is conspicuously streaked,' consisting of highly elongated narrow lenses (one-eighth to one-fourth of an inch wide) of feldspar wrapped around by thin intermittent sheets of black hornblende. The feldspar is mostly pink in color but some is pale olive-green. In thin section the rock is seen to consist predominantly of microperthite with associated hornblende and oligoclase and magnetite, biotite, apatite and zircon as accessory minerals. The mineral composition of two representa- tive facies is given in the accompanying table. The hornblende and oligoclase rich facies (4) is associated with much included amphibolite and it seems most probable that the high percentage of these two minerals is the result of assimilation and reaction with the included shreds and disintegrated layers of country rock. 74 NEW YORK STATE MUSEUM Mineral composition of representative facies of the equigranular syenite (Lowville quadrangle) I 2 3 4 Microperthite 76 80.5 56 25 Oligoclase i-5 23-5 51 Augite 9- 4-5 Hornblende 0.5 3 8 15 Biotite 1-5 1.5 Quartz 1 1 10.5 9 6 Magnetite 1-5 1 1-5 I Accessories 0-5 0-5 0-5 0.5 1 Augite syenite, equigranular texture, average of two sections, southeast part of Lowville quadrangle 2 Augite syenite, equigranular texture, average of two sections, southeast part of Lowville quadrangle 3 Hornblende syenite, equigranular texture, average of three sections, between Monnatt School and Indian river 4 Hornblende syenite, equigranular texture, hornblende rich facies, average of three sections, between Monnatt School and Indian river GRANOSYENITE AND GRANITE Introduction. In the main igneous complex there are rocks with quartz ranging from io to 20 per cent that are intermediate between granite and typical equigranular hornblende syenite on the one hand and between a porphyritic granite and the typical porphyritic horn- blende syenite on the other. The first is an equigranular grano- syenite and the second a porphyritic granosyenite. Granosyenite (equigranular). The equigranular granosyenite is found on the Lowville quadrangle in small areas along the eastern and northern borders. The equigranular granosyenite shows a less siliceous and slightly more basic border facies adjoining the syenites and grades into granite (Lowville type). Boundary lines between these latter two facies are arbitrary and only approximate at best. Where the quartz in the rock over any considerable area runs consist- ently less than 25 per cent and over 10 per cent it is mapped as granosyenite. Typically the rock averages about 20 per cent. North- east of Beach Millpond the granosyenite is one of the components of the granosyenite-hornblende syenite mixed gneisses, where it occurs intersheeted with the hornblende syenite and appears to bear intrusive relations to it. Bands of granosyenite frequently occur parallel to the foliation of the granite (Lowville) but are too indefinite to map. HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 75 The granosyenite of the northeastern corner is continuous with that of the Lake Bonaparte quadrangle and full details of the transi- tion to granite and of its mineralogical character will be found in that report. In thin section the rock is found to consist predominantly of microperthite with varying amounts of quartz, plagioclase and horn- blende. Biotite, magnetite, zircon and apatite occur as accessory minerals. Granosyenite (porphyritic) . Hornblende granosyenite with a porphyritic or augen gneiss structure forms a belt northwest of Croghan on the Lowville quadrangle, and underlies most of the area in the northeast corner of the Carthage and southeast corner of the Antwerp quadrangles. It grades in composition into granite on the one hand and hornblende syenite on the other and forms a mem- ber of the Diana complex. It contains many shreds or included bands of dark-colored pyroxene-biotite gneiss, and is intruded by a few hyperite sheets. The rock has a reddish hue. Quartz varies from io to 25 per cent and usually forms 15 to 20 per cent of the rock. In the hand specimen, however, it appears inconspicuous and one is prone in field examination to underestimate its quantity. Plagioclase and microperthite constitute the eyes or residual cores resulting from the crushing to which the rock has been subjected. The quartz is in flat leaves, which in thin sections show a granular or cataclastic structure. The ground mass usually consists of micro- cline and plagioclase. The plagioclase on the average is oligoclase. The microcline occasionally shows a little microperthitic intergrowth of plagioclase. The ferromagnesian mineral, where least altered, is hornblende. It does not form more than a few per cent of the rock. For the most part the original ferromagnesian minerals have been changed to biotite or to carbonate aggregates or chlorite. Magnetite, apatite and zircon are common accessory minerals. In part the biotite may be primary. Locally the granosyenite carries a little augite, evidently, in part at least derived from the disintegration of inclu- sions of pyroxene-biotite schist. On the Lowville quadrangle in the New Bremen area is an area of granite, unfortunately, so covered by overlying sands and glacial drift that little can be made of its relations to the surrounding rocks and its areal extent. The boundaries of the main mass outlined on the map are only approximate. The rock apparently forms the eastern end of a belt that starts just northeast of New Bremen and extends underneath the Black River valley reappearing at Dadville 7 6 NEW YORK STATE MUSEUM and passing westward beneath the overlying Paleozoic beds. Good exposures are found just south of the villages of New Bremen and Dadville. The rock is a very coarse porphyritic red gneiss of the augen type. The principal minerals seen in the hand specimen are feldspar, quartz and hornblende. The feldspar occur mostly as phenocrysts and, as such, form by far the major portion of the rock. They are for the most part granulated and flattened out into lenses parallel to the foliation. The average of eight measurements of such phenocrysts in a typical sample of the gneiss made in a plane at right angles to the strike showed the length along the dip to be three inches and the width one inch. Measured on the surface they range from one to two inches and average perhaps one and one-half inches in length. Where two or more such lenses overlap one another in echelon fashion they give rise to an appearance as though actual lenticular veins of feldspar had been injected parallel to the foliae. Many of the phenocrysts are fresh and uncrushed, but most of them have become granulated by crushing. Quartz usually occurs as a meshwork surrounding the phenocrysts. Together with some feld- spar it formed the relatively small amount of more or less equi- granular groundmass. Ferromagnesian minerals are trifling in amount. Locally a little magnetite is common. The rock prevail- ingly exhibits a strongly marked gneissic appearance but locally the phenocrysts are arranged haphazard. In thin section the rock is found to consist of microperthite more or less completely granulated, quartz in long leaflike forms with only a few fractures and exhibiting no such crush phenomena as the feldspars, and some granulated horn- blende, the latter never exceeding a few per cent. Magnetite, apatite and zircon are constantly present as accessory minerals. The plagio- clase portion of the microperthite is oligoclase. The grain of the rock is too coarse to use the Rosiwal method for the determination of the mineral percentages. The results of several estimates made in the field, however, show it to average quite consistently from 18 to 20 per cent quartz. The estimates were made by measuring the total width of quartz appearing along a line at right angles to the foliation and finding its ratio to the total length of the line. The belt northwest of Croghan is of finer grain than that in the New Bremen area, and the porphyritic lenticular structure although present is indistinct. The lenses average about one-half inch in length and are wrapped around by thin leaves of quartz. There is very little ferromagnesian mineral. In the fresher facies hornblende may form 2 per cent and biotite is a minor accessory. The predominant feldspar is usually microperthite and plagioclase is variable in amount. HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 77 In one specimen examined plagioclase forms 50 per cent of the rock and is associated with microcline. Quartz constitutes 20 to 25 per cent. There are numerous hyperite sheets, usually not more than 20 feet wide, in the granite parallel to the foliation. The granite is also cut by aplitielike material resembling the adjoining Alexan- dria granite. The porphyritic granite appears to bear intrusive rela- tions toward the hornblende syenite along its northern border. It contains numerous inclusions of schist and might be interpreted as a relatively quicker chilled facies. The rock of the small lense that outcrops on the Texas road three miles east of Texas School is not so coarse as that of the New Bremen district nor usually so conspicuously porphyritic, but in its composition it is more or less the same. The granite mass east of Texas Road Schoolhouse differs from the normal type in that augite forms the ferromagnesian mineral instead of hornblende. The rock is red and shows a cataclastic microstructure. Some of the phenocrysts have a plagioclase core surrounded by microperthite. The rock is obviously a facies of the adjoining porphyritic granosyenitic mass. SYENITE-GRANITE AND GRANOSYENITE MIXED ROCK A band of mixed gneiss about one-half mile wide extends from northeast of Bush’s Corners to southwest of Crystal Dale (Lowville sheet). A block of similar gneiss occurs north of Beach Millpond and east of Crystal lake. This combination gneiss consists prevailingly of a fine-grained, greenish hornblende syenite gneiss interleaved with sheets of reddish granite parallel to its foliation. In part this granite is a coarse red rock and in part it is a medium-grained quartzose gneiss whose borders are in places in sharp contrast to the syenite gneiss and in places seem to be but part and parcel of the syenitic gneiss. Quartzose pegmatite sheets an inch or so wide are abundant locally in the gneiss but they are barely distinguishable as discrete veins because of indistinct borders. Occasionally,, veins of peg- matite from 1 inch to 8 inches wide occur in the syenite which are extraordinarily rich in coarse magnetite grains. The pegmatite veins and granite sheets increase in number along the northwest border. The granite here also contains lenticular sheets of the syenite which appear to occur as long narrow bedlike inclusions. The syenite is a fine to medium-grained quartzose gneissoid horn- blendic rock. It is green on fresh surface but weathers in part to a reddish hue and in part to a gray. In thin section the rock shows a typical granitoid texture without signs of crushing. Microperthite 7« NEW YORK STATE MUSEUM is the predominant mineral forming from 75 to 85 per cent of the rock. Quartz and hornblende are the only other principal minerals. Quartz forms from 7 to 15 per cent and hornblende from 6 to 10 per cent of the rock. Magnetite, zircon and apatite are constant accessory minerals and are locally accompanied by biotite. The principal difference between this rock and the augite syenite is the absence of augite, its place being taken by hornblende, and a trifle more quartzose appearance on the weathered surface. The granite and granosyenite sheets are similar to the rocks of the main masses previously described and description will not be repeated here. GRANITES Introduction. There are at least two distinct types — Hermon and Alexandria — of granite represented within the Grenville belt of the Antwerp and Hammond quadrangles and throughout the northwest Adirondacks. The distinction between them is based upon both the texture and composition of the rocks and upon their structural relations and characters. Two similar lithologic types of granite, although in part perhaps of different ages, are found within the main igneous complex of the Antwerp, Lowville and Carthage sheets. Within the main igneous complex, on the Low- ville sheet, however, there is a third type — Lowville — whose relations to the two mentioned are not known with certainty and which has a uniform character over a wide area and may deserve to be ranked as a third distinct type. For these reasons it is mapped separately. One type — Hermon— -except where in small masses or along the contacts with the country rock, is moderately coarsely porphyritic with phenocrysts of feldspar, has a coarse to medium-grained ground- mass, much of it has about 20 per cent quartz (65 to 67 per cent Si02) and characteristically occurs as long narrow sheets parallel to the foliation or bedding of the country rock. The mass southeast of Spragueville (Hammond sheet) is but a minute portion of a great sheet of such granite that extends southwest, on the Antwerp sheet, to beyond Philadelphia, where it passes beneath the overlying Potsdam sandstone, and northeast across the Gouverneur quadrangle on to the Russell sheet. It has a minimum length of over 35 miles and a width of from one to three miles. This great sheet has been called the Hermon sill by Cushing, and granites similar in lithology to it will be referred to as the Hermon type. This sill is dominantly porphyritic but locally there are coarse-grained more siliceous equi- granular facies like that of the Lowville granite. HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 79 The second type — Alexandria-— is nonporphyritic and medium to fine-grained. Structurally it has almost exclusively two modes of occurrence: (i) as great elongated so-called batholiths such as the Alexandria batholith of the Alexandria quadrangle or in the west half of the Canton quadrangle, and (2) elliptical-shaped domical masses, such as the Payne Lake, Dodds Creek, Hyde School and Hickory Lake masses on the Hammond quadrangle, the masses west and east (Gouverneur and Reservoir hill) of Gouverneur on the Gouverneur quadrangle, and the Clark Pond and California bodies of the Lake Bonaparte quadrangle. Wherever the elliptical-shaped masses are found they are characterized by (a) a domical foliation locally asymmetric or even overturned, (b) local development of a characteristic thin shell of garnet rock and (c) pegmatite veins which carry accessory magnetite and rarely tourmaline. On the Alexandria sheet, to the west of the Hammond quadrangle, Cushing has mapped a long belt of fine to medium-grained granite extending along the south side of Wells island and the south side of the St Lawrence river. This belt of granite he has referred to as the Alexandria batholith. It is continuous with the granite along Chippewa creek on the Hammond quadrangle and extends to the northeast on to the Ogdensburg sheet, where it disappears beneath the overlying Potsdam sandstone. It has a minimum length of 26 miles. The westward portion of this batholith lies in Canada, where it has been mapped by Wright as the Mallorytown granite. Granite similar lithologically to the granite of the Alexandria batholith will be referred to as the Alexandria type. Petrographically the Hermon type where typically developed is characterized by its coarser grain, porphyritic texture, higher content of ferromagnesian minerals, smaller percentage of quartz (commonly a granosyenite) greater percentage of plagioclase and more common presence of the accessory mineral titanite. The Alexandria type in its typical development is characterized by a medium to fine grain, a higher percentage of quartz and potash feldspar (microcline, ortho- clase or microperthite) and a smaller amount of biotite. In the latter rock ferromagnesian minerals are often inconspicuous or wanting. The Lowville type is coarse to medium-grained equigranular in character over wide areas and is often high in quartz. It is, however, very similar to the equigranular border and chill facies of the Hermon type, and to the coarser facies of the Alexandria type. The relations of these two types is not certain. 8o NEW YORK STATE MUSEUM Granite and granosyenite (Hermon type) in Grenville belt. Masses of porphyritic granite varying to granosyenite occur in several belts and small bodies within the Grenville formations of the Antwerp and Hammond quadrangles. In small bodies and locally in narrow portions of large bodies the granite is medium-grained and non- porphyritic and more siliceous but in the main body of large masses it always has a porphyritic character. Along some belts the granite has come in along the foliation planes of amphibolite to such an extent as to constitute a mixed gneiss. In other masses there are included bands of Grenville formation. The latter is usually thin layered lime- stone and gneiss or limestone full of quartz veinlets. In small dikes in the limestones and locally near the borders of granite with lime- stone the granite assumes a light greenish gray hue. The typical rock consists largely of pink to light-colored microcline phenocrysts one-half inch or more in length in a gray groundmass. To a minor extent plagioclase also occurs as phenocrysts and locally orthoclase. There is always a distinct gneissic structure and as a result of crushing the phenocrysts are often lenticular in shape and the rock is a typical augen gneiss. Border facies of the larger masses are often finely porphyritic or medium-grained. The ground- mass of the porphyritic rock when examined microscopically is found to consist almost wholly of oligoclase and quartz with a little associated microcline, biotite and accessory apatite, titanite, magnetite and zircon crystals. There is locally very little microcline in the groundmass of the porphyritic rocks. Biotite is the most common dark mineral and usually does not form more than 5 per cent of the rock. Locally hornblende is present in addition to biotite. The biotite and hornblende are often altered to chlorite. The microcline has usually sparse microscopic spindles of plagioclase intergrown and a little myrmekitic intergrowth of plagioclase, and quartz is usually present. Quartz usually varies from 20 to 25 per cent, but in the medium-grained masses and dikes may be greater. Plagioclase probably exceeds or is equal in amount to microcline, and the rock might therefore be called an oligoclase quartz monzonite. The structure of the granite is exceedingly variable. Many dikes and small bodies in the limestones are massive with textures and structure characteristic of crystallization under quiet conditions ; most of the rock has a more or less pronounced protoclastic structure ; and the granite masses along the west side of Black Creek valley south- west of Hammond are extremely mashed and have a most marked cataclastic structure. HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 8l South of Oxbow is the northeast termination of a tongue of granite that extends far to the southwest on the Antwerp quad- rangle, where it joins with the belt extending southwest from Spragueville. This granite consists of microcline phenocrysts from one-half inch to an inch in length in a medium-grained groundmass of quartz, oligoclase and biotite with a little microcline. Locally hornblende is present in addition to biotite. The hornblende and biotite are locally altered to chlorite. Apatite, magnetite, zircon and a little titanite are accessory minerals. Except locally at the very borders of the mass south of Oxbow, however, the granite shows a protoclastic structure, and the rock is a typical augen gneiss. The quartz is in flat lenses that may be in massive leaves or consist of grains with a denticulate texture, each of which has a ribboned wavy extinction as a result of stress. Thin microscopic shear planes, along which the quartz and feldspars are powdered and secondary biotite developed, are common and such zones are often recrystallized. Some of the grains of microcline resulting from protoclastic granulation are locally replaced by a little myrmekitic intergrowth of plagioclase and quartz. The granite, however, is moderately but by no means severely mashed, being very much less mashed than are the porphyritic granites along the west side of Black Creek valley. At the extreme northeast end of the Sherman granite mass near the contact with the limestone the quartz is in flat lenses but there is only a little or no protoclastic structure and the quartz shows only wavy extinction as a result of stress. East of Red lake is the northeast end of a mass of granite that has a pronounced gneissic structure with small phenocrysts of plagioclase and microcline one-eighth to one-quarter inch in diameter. The microcline is predominant. It is fine-grained on the borders and coarse in the core. In thin section the rock is found to have a very marked protoclastic structure without a trace, however, of cataclastic effects. Only a very little ferromagnesian mineral is pres- ent, consisting of chloritized biotite. Between Scotch Settlement and the Indian river there is a con- siderable mass of granite, the foliation of which parallels its borders. The northeastern tongue is intimately involved with Grenville for- mation and bands of Grenville and amphibolite also occur in the western portion. Most of the mass is medium to coarse-grained with sparse to abundant phenocrysts of microcline. The mass one mile north of Laidlow School is medium-grained nonporphyritic but the body to the north and east is uniformly coarsely porphyritic. 8 2 NEW YORK STATE MUSEUM Medium-grained nonporphyritic facies do occur within the main mass. The southern border is intensively mashed and thin granite dikes in the adjoining limestone are plicated along with the latter. Most of the intrusive mass southwest of Rossie is a medium- grained gray gneissoid rock with very locally a porphyritic aspect. The phenocrysts, however, are never large, usually not more than one-quarter of an inch. The gneissic structure is pronounced around the borders and gives rise to lineal topographic ridges, whereas in the core the gneissic structure is indistinct and a pronounced more or less horizontal sheeting gives rise to broad areas of flat topography. The rock is seamed with irregular veinings of pegmatite, which often carry little segregations of quartz and tourmaline. There are locally bands of a pinker more coarsely porphyritic granite that may be a facies of the gray granite or may be a little later. The relations of the gray granite and the quartz diorite have not been ascertained. The average granite southwest of Rossie consists of 43 per cent of plagioclase, 25 per cent microcline, 28 per cent quartz and 3.5 per cent biotite. The microcline usually has a slight amount of spindles of plagioclase. No protoclastic or cataclastic structures were observed except locally though the quartz shows pronounced strain shadows. The biotite is oriented parallel to the gneissic structure, and some of the quartz is in elongated blebs. Rocks similar in character and structural relations to the por- phyritic granites of the Hammond quadrangle are also found else- where throughout the northwest Adirondacks and in Canada. The Brockville granite described by Wright (’23) appears to be their equivalent in Canada. Mineral composition of granite (Hermon type) in Grenville series I 2 3 Oligoclase 38 30 20 36.5 31 29 21 . 1 Microcline and orthoclase 39-6* 22.7 Quartz. Hornblende. 8 0-9 Biotite. 2 . S I3*4 2 . 6 Accessories. • 4 I 1 Estimated composition of typical porphyritic granite (Hermon type), Hammond quadrangle 2 Average of four medium-grained facies of granite (Hermon type), Hammond quadrangle 3 Brockville granite;' Brockville-Mallorytown area, Canada, J. F. Wright (’23, p. 25-26). ’“Potash feldspar HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 83 In the Antwerp quadrangle the porphyritic granite is found to consist of a groundmass of quartz, microcline and plagioclase, with accessory biotite and locally a little hornblende, surrounding pheno- crysts of microcline, orthoclase or occasionally microperthite and plagioclase. The medium-grained granite, interpreted as a facies of the porphyritic type, is more siliceous, hornblende is usually lacking, and biotite and a little muscovite are the predominant minor minerals. Accessory minerals in both facies of the granite comprise magnetite, apatite and zircon. Locally there is some titanite, perhaps in part due to effects of assimilation. Locally the biotite is partly or com- pletely altered to chlorite, and the feldspars are flecked with a little sericite. The structure of the granite is predominantly protoclastic. The quartz is in massive leaves with a wavy extinction. The feld- spar of the groundmass is a granular aggregate resulting from crushing, but not to such a fine degree as usually characterizes cata- clastic structure. There is, however, almost uniformly a little fine- grained material in the groundmass that has very definitely the appearance of recrystallized crushed stuff. In part it resembles a late stage myrmekite. Locally there is a little distinct cataclastic crushing. In the vicinity of Philadelphia there is considerable gray granite that contains inclusions of amphibolite and gneiss and appears to be cut parallel to the foliation by bands of normal reddish granite or aplite. The darker granite may have resulted from reaction with included bands of the Grenville gneiss, but this is not positive. Granite (Lowville type). A band of granite averaging about four miles in width strikes northeast-southwest across the central part of the Lowville area. A narrow offshoot extends off to the west of Croghan and another smaller one to the northwest of Monnatt School. About three-fifths of a mile west of Strifts School the granite incloses many sheets of amphibolite parallel to the foliation, and is somewhat coarser in texture. The amphibolite lenses vary in thick- ness from several feet up to a hundred feet. Pegmatite veins often gashed with white quartz lenses are more common than where the granite is free of inclusions. South and southeast of Kirschnerville there is a large area of practically clean granite with only a few pegmatite veins. Sheets of syenitic type are often associated with the amphibolite and together with it are inclosed within the granite. There is no gradation between the granite and syenite; the contacts are absolutely sharp. Near the southeastern border, the granite contains sheets of hornblende syenite which it has pried off from 84 NEW YORK STATE MUSEUM the adjoining mass. A mass of granite forms the conspicuous hills northwest of Puffer School, and incloses the block of Grenville gneiss near Pine Grove. Locally particularly along the borders with other rocks, the granite passes over into a granosyenitic facies, previously described, through a decrease in quartz, in which case boundaries can be mapped only approximately. The granite is a continuation of the same rock mapped by Miller on the Port Leyden sheet as granitic syenite. The granite is a medium to coarse-grained (6 mm) reddish equi- granular highly gneissoid rock of the typical leaf gneiss type. Owing to the feldspar being more readily decomposed than the quartz, the weathered surface of the gneiss has a finely jagged surface. Asso- ciated with the granite there are frequently abundant very coarse- grained pegmatite veins parallel to the foliation. Many of these veins exhibit a flattened meshwork of white vitreous quartz veins from one-half inch to several inches wide inclosing lenses of feldspar from several inches to a foot in length. They give to the rock a decidedly gashed or splayed appearance. The quartz of the granite itself occurs in flattened lenticular leaves resulting in the highly gneissoid appearance. Locally, especially along the southeastern border, there is a tendency for the granite to assume a porphyritic appearance and to resemble the porphyritic hornblende granosyenite of New Bremen. Sheets of porphyritic granite occur interleaved with slightly porphyritic granite. As a result of many estimates made in the field, the more siliceous facies of the granite was found to average about 40 per cent quartz, but it varies toward granosyenite with 20 per cent quartz. The predominant feldspar is microperthite, which varies from 50 to 70 per cent. Oligoclase varies from 2 to 20 per cent but usually does not form more than 10 per cent. In most of the rock the ferromagnesian minerals form only a few per cent or less. Biotite is the usual mineral, but hornblende is common in local facies. In part the horn- blende appears to be a relic from the disintegration of amphibolite bands by the granite. Magnetite, apatite and zircon are always pres- ent as accessory minerals. The average specific gravity of the granite is about 2.640. South and northwest of Croghan the granite is heterogeneous. The predominant rock is medium-grained with 20 to 30 per cent quartz, but it contains interbands of more coarsely crystalline and quartzose granite, coarsely porphyritic granite, granosyenitic variants, aplitic granite and many inclusions of pyroxene-biotite schist that HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 85 are dismembered fragments of a member of the Grenville formations. The granite becomes greenish locally adjacent to the schist inclusions and may form a green-gray assimilation gneiss with relict shreds of schist. In thin section the typical granite is found to consist of 40 to 60 per cent microcline with a little perthitic intergrowth of plagioclase, io to 25 per cent plagioclase, 20 to 30 per cent quartz and variable amounts of hornblende with accessory magnetite, biotite, apatite and zircon. Hornblende often forms about 5 to 6 per cent of the rock but may be absent and only a little biotite present. Granite (Alexandria type) in Grenville belt. The fine to medium-grained granites consist predominantly of microperthite, microcline or both, quartz and plagioclase. A trace of biotite is usu- ally present and hornblende or biotite up to a few per cent where the granite has disintegrated or reacted with included layers of dark Grenville gneiss. Magnetite is the most common accessory mineral and apatite and zircon are usually present but sparse. Allanite and tourmaline grains are found in the mass along Chippewa creek (Hammond quadrangle). Rutile needles of microscopic size are often found in the quartz. Titanite is very rare or absent. A trace of muscovite occurs occasionally. The granites have a primary gneissic banding, due in part to the lens-shaped structure of the quartz leaves and in part to variation in composition in alternate thin laminae. The bands vary from pink to lighter hues. Quartz varies in general from 25 to 40 per cent, plagioclase from o to 35 per cent and in rare cases up to 60 per cent. Microperthite from o to 60 per cent and microcline from 30 to 55 per cent. In the Dodd and Payne granite masses (Hammond sheet) the potassic feldspar is microperthite, in the Hyde mass, microperthite and microcline are both present and in the Chippewa mass, Payne sill and the sill one and three-quarters miles west of Natural Bridge microcline alone is the potassic feldspar. The aplites are wholly microcline and quartz. As might be expected, the ratio of potassic feldspar to plagioclase is less where the potassic feldspar is micro- cline than where it is microperthite. There is probably not much difference in the chemical composition of the rocks, except in the case of the aplites, which are more quartzose and potassic. The average composition of the microcline and microperthitic granites respectively are given below. 86 NEW YORK STATE MUSEUM Mineral Facies of Alexandria Granite 1 2 Microcline 41 26 3i-5 0-75 0-75 5i 34-5 0-75 1 .0 Plagioclase Microperthite Quartz Hornblende Biotite Magnetite 1 Average of eight microcline facies from Alexandria granite on Hammond quadrangle 2 Average of eight microperthite facies from Alexandria granite on Hammond quadrangle The Chippewa granite (Hammond sheet) differs from the other fine-grained granite masses in that the potassic feldspar is almost wholly microcline instead of microperthite. Otherwise it is similar. The average composition is given in the table. The biotite of this mass is altered to chlorite. The feldspar is in polygonal grains with leaves of quartz and might be interpreted as representing either protoclastic or crystalloblastic structure. The Hyde granite mass (Hammond sheet) forms the area south- west of Hickory lake in the bend between Hickory creek. The granite is a fine-grained pink rock with a marked gneissic structure. Layers of dark pyroxenic, hornblendic and biotitic gneiss are common. These usually vary from a fraction of an inch up to several inches thick. They are conspicuous because of their color but usually form less than 5 per cent of the body of the mass, although locally the granite may contain up to 20 per cent of included layers. The dark gneisses 'consist usually of hornblende, biotite and plagioclase with often associated monoclinic pyroxene. The hornblendic layers are more common near the borders and the biotitic bands are more common in the core. They are always parallel to the gneissic structure. Coarse pink pegmatite veins are common in the granite and many of them are parallel to the foliation. Magnetite is a common accessory mineral, although never in large amount. Tourmaline has not been seen in the pegmatite veins of the main body of the granite. A trace of biotite is usually present in the pegmatites. Quartz veins by themselves are rare, although quartz veinlets in the big pegmatite veinlets are common. In the granite itself quartz may vary from 20 to 50 per cent, potash feldspars from 10 to 60 per cent and plagio- HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 87 clase from 12 to 60 per cent. There is often a marked thin banding- in the granite resulting from a rapid variation in mineral compo- sition. In much of the rock ferromagnesian minerals are practically absent; in other portions hornblende or biotite may be present up to the extent of a few per cent. The plagioclase is often antiperthitic with sparse intergrowths of potassic feldspar. Magnetite usually forms xl/2 per cent and apatite is always present. The quartz often contains many microscopic rutile needles. Zircon and pyrite are other rare accessory minerals. The potassic feldspars are usually microperthite or microcline. The microcline appears in part as larger grains and in part as peripheral borders to the larger micro- perthite grains. The microcline of the groundmass may have formed during granulation of the larger grains. At the borders of the mass the granite usually becomes coarser grained, takes on a greenish hue and is much less quartzose. Plagio- clase is present to the exclusion of potash feldspar. In one specimen pyroxene forms about 6 per cent of the rock as skeletal remnants partially replaced by plagioclase. On the west side of Birch creek at the southwest end of the Hyde granite mass there is a sill of green granite in hornblende gneiss. The composition of this granite is given in the accompanying table and is the result of reaction with the gneiss. Usually 1 per cent of apatite is present in such border facies and a little allanite. The microtexture of the Hyde granite predominantly shows a small amount of intergranular material which may have resulted from crushing and recrystallization. Locally, there is a considerable amount of such material ; and locally also, the texture as a whole is that of interlocking normal crystallization. In the vicinity of the lower half of Bostwick creek (Hammond sheet) and about three-quarters of a mile north of Farley School there is a narrow sheet of pink aplite that bears intrusive relations toward the Rossie quartz diorite and diorite mass and the Grenville formations in the same region. All the rocks including both the Rossie and Bostwick intrusive sheets and the metamorphic country rocks have been folded. The pink aplite is a fine-grained rock with a gneissic banding and so pronounced a platy parting that it may readily be mistaken for a feldspathic quartzite, especially so since it is folded together with the Grenville formation and is approximately parallel to the structure. It is cut by pegmatite seams, some of them carrying tourmaline. The hills south of the alluvium-covered flat south of 88 NEW YORK STATE MUSEUM South Woods School comprise a complex series of rocks consisting of Grenville limestone and gneisses intruded by the Rossie dioritic rock and all intruded by pink porphyritic granite and pink aplitic granite sheets. The aplite resembles the Hyde granite in appearance except that it is much finer grained. The aplite consists, usually, almost wholly of the two minerals — microcline and quartz, the microcline carrying sparse microscopic spindle intergrowths of plagioclase. The quartz usually varies from 35 to 50 per cent and microcline from 50 to 60 per cent. One specimen consisted of 45 per cent microcline, 18 per cent plagioclase, 28 per cent quartz and 9 per cent biotite. The composition and structural relations indicate that the aplitic granite is genetically associated with the Hyde granite mass. In the Bostwick Creek aplite sheet the quartz is usually in flat narrow lenses or spindles and the microcline in grains that are in part polygonal-shaped with straight line sides suggesting protoclastic structure and in part in irregular grains suggesting primary crystall- ization. The evidence for protoclastic structure, however, is some- what indecisive and there is no evidence of cataclastic structure. The Dodd’s granite mass (Hammond sheet) is an elongated body lying between Dodds creek ard the Indian river, about half in St Lawrence county and half in Jefferson county. It is a fine to medium-grained pink granite similar to that of the Payne, Hyde, Hickory and Butterfield Lake masses. The approximate mineral com- position of two specimens is given in the table. Both show accessory magnetite, apatite, zircon and chlorite. In the specimen from the core, the microperthite preceded the plagioclase in order of crystalli- zation and the latter often forms the periphery to microperthite crystals. Mineral composition of Dodd’s granite mass 1 2 Microperthite 52 25 20 60 Quartz 40 Plagioclase. Accessories 3 1 Center of Dodd’s mass on county line 2 Southwest end of Dodd’s granite mass HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 89 The whole mass is very much mashed, the quartz is in spindles and is granulated, and the feldspars are finely crushed on their borders and along shear planes. The feldspars are much finer and more intensively mashed than the quartz. Locally a finely por- phyritic augen gneiss has been produced as a result of crushing. Evidently we are dealing here with a cataclastic gneiss. The Payne Lake sill (Hammond sheet) locally shows a granulitic structure but without notable cataclastic effects. The rock averages about 45 per cent microcline, 40 per cent quartz, 14 per cent plagio- clase and 1 per cent of accessories. The accessory minerals including only two specimens from the main mass of the Payne Lake granite were examined, and these from the south end. Both specimens show a normal texture of crystallization without protoclastic but with a little evidence of cataclastic effects. The rock averages about 50 per cent microperthite, 45 per cent quartz, 3 per cent plagioclase and 2 per cent accessory minerals, including the same minerals as those in the sill. The sill thus differs in that the microperthite has crystallized out as microcline and plagioclase. At the extreme south end adjacent to the garnet rock the granite has two-thirds per cent of allanite. Granite [Alexandria (?) type] in main igneous complex. On the Lowville and Carthage quadrangles there is a narrow belt of fine-grained granite that extends northwest from Croghan to Mount Tom and thence west to the Black river. South of Croghan it is interbanded with coarse-grained equigranular granite and with porphyritic granite. Throughout the belt there are abundant included fragments of pyroxene-biotite gneiss. In thin section the rock is found to consist predominantly of microperthite or microcline or both with some associated plagioclase and about 30 per cent quartz. Ferromagnesian minerals do not form more than 1 per cent except where the granite has disintegrated a shred of the pyroxene-biotite gneiss. The basic minerals are usually chloritized. Magnetite, zircon and apatite occur as accessory minerals. The rock appears even grained at the surface, but in thin section much of it is found to show larger residual cores of microperthite in a fine-grained groundmass. It it not evident whether this structure is the result wholly of crushing or whether it is indicative of an original por- phyritic texture. It is therefore not certain whether this granite is a chilled facies of the porphyritic granite or whether it belongs to the Alexandria type, although it has been mapped with the latter. The granite mass northeast of Blodgett Landing on the Carthage quadrangle is interbanded with and apparently intrusive into syenite. 90 NEW YORK STATE MUSEUM Here again, however, it is not certain whether the granite should be grouped with the Alexandria type or as a facies of the porphyritic granosyenite. The Mount Tom granite band and that north- east of Blodgett Landing are connected by a number of bands of granite in the syenite south of Swiss creek. Pegmatite and granite dikes. Throughout the area of the Gren- ville formations there are dikes and small masses of granite which may be either locally quite abundant or in separate isolated masses. In most cases it has not been determined with which type of granite these masses are genetically related. It is the author’s opinion, how- ever, that in most cases the granite dikes are offshoots of the por- phyritic granites which through quicker cooling have not developed the porphyritic character. They vary from red to green-gray in color. Most of the pegmatite veins likewise appear to be genetically related to the porphyritic granites. Pegmatite veins and lenses are widely variable in quantity but are omnipresent throughout all the varieties of the Grenville formation rocks of the quadrangle. There is no area of any considerable size anywhere that does not contain at least small veinlets of pegmatite. Only the larger dikes and lenses of granite and pegmatite are indicated on the map. There are a multitude of others not mapped. The largest masses of pegmatite seen are those forming the north- ward extension of the medium-grained granite southwest of the Payne Lake granite mass. Many of the small granite dikes and masses show a local develop- ment of pegmatitic character near their borders or at their termina- tions or are crossed by veinings of pegmatite. The pegmatites in a similar fashion locally show a development of quartz or quartz-tourmaline rock along their borders or at their terminations or are crossed by such veins. Feldspathic quartz veins are also common. Often successive belts are found in which the predominant dikes or veins are successively granite, pegmatite and quartz, with pro- nounced mineralization of the country rock accompanying the pegma- tite or quartz, or occurring separately as a late stage replacement. Agar (’23, p. 108-9) has described the transition from a belt of small granite masses through a zone of pegmatite dikes to one in which nodules and disseminations of minerals occur as replacements in the limestones which take place from northwest to southeast in the area between Indian river and the road from Oxbow to Scotch Settlement, and which probably indicates increasing dis- HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 91 tances above a main body of granite that underlies at depth. His description follows : The northern part of the area under discussion, just south of the granite boundary, contains numerous dikes of granite. These are, as a rule, slightly more acid than the main mass of the rock and are composed chiefly of microcline and granulated quartz, with small amounts of hornblende and mica drawn out into stringers, with titanite and occasionally tourmaline as accessories. They are every- where accompanied by a certain number of pegmatite dikes. Southward the zone of dominant granite dikes passes into another one of mixed granite and coarse quartz-feldspar pegmatite dikes and sills and, still farther south, the latter occur to the complete exclusion of the granite, some of the larger masses show a good transition between the granite and the pegmatite. One in particular shows a fine-grained tourmaline-bearing granite sill about 100 yards long, grading at several points into a coarse quartz-feldspar aggregate similar to the pegmatite dikes and sills. South of here the pegmatites continue to be very plentiful and about half way down the area, mineralized zones are found. These increase in number until the southern part is reached and the pegma- tite intrusives thin out. Throughout the mineralized zone, pockets and disseminations of minerals are found in otherwise clear lime- stone at a distance from the intrusives as well as along the borders and in the intrusives themselves. Not infrequently the minerals are found beyond the exposed end of a sill or dike of pegmatite, or connected with a quartz vein parallel to, or at the end of such a pegmatite. So from north to south within this area we find granite, then limestone cut by predominating granite dikes, then mixed granite and pegmatite dikes, large quantities of pegmatites with maximum min- eralization and finally a dying out of the pegmatites with a consider- able mineralization. West of the southern part of this area there are large bosses of porphyritic granite and south of it another mass of granite with its accompanying dikes and sills, and, in the central part of the area, there is a granite sill several hundred yards long and fifty yards or more wide; still the relations outlined above are clear enough when taken as a whole. The pegmatites have several different modes of occurrence depen- dent upon the character of the country rocks. In the limestones the pegmatite dikes are usually white and very rarely form an arteritic injection, whereas in the gneisses and amphibolite the pegma- tites are pink to red and often form a migrnatite of arteritic injection type. In the limestones the pegmatites usually form sheets or lenses from five to 25 feet wide, although strings of very small pegmatite nodules are also found. The pegmatites are usually light-colored and 92 NEW YORK STATE MUSEUM weather white. Many of the small nodules have a thin border of brown tourmaline-bearing quartz developed at their contact with the limestone. In the belts of limestone with pegmatite and quartz vein- ings there are all gradations between quartz, feldspar and tourmaline occurring in association as disseminated crystals in the limestones to tourmaline-bearing pegmatite veins of replacement origin. There are a few bands in which pegmatitic veinings and replacements are of similar character and in similar quantity to the arteritic injection gneisses. In the hornblende gneisses resulting from the mashing of the metagabbro the pegmatite occurs as arteritic injections for the most part but in the more massive metagabbro masses such as the Pleasant Lake mass and the mass about a mile south of Pine Hill School the metagabbro is brecciated and the pegmatite veins crisscross the mass. The pegmatite veins are often very large, up to 50 feet in width, and very coarse grained. A graphic intergrowth of quartz and feld- spar is often present locally, as well as segregations of graphically intergrown black tourmaline and quartz. Veinlets of tourmaline- bearing quartz are also found here and there, and the big pegmatite veins are always gashed with glassy quartz. The pegmatite veins that occur within the alaskitic type of granite masses often carry a little magnetite, and tourmaline is rare or wholly absent, whereas the pegmatite veins in the porphyritic granites very commonly carry black tourmaline. Tourmaline-bearing veins, however, usually cut the contact metamorphic shell of garnet rock which locally adjoins the alaskite granites (Alexandria type) and are therefore known to be in part younger than the latter. One pegma- tite vein carrying plates of ilmenite was found cutting a sheet of amphibolite in the syenite east of Red lake. HYPERITE DIKES The granite-syenite series (Diana complex) of rocks in the Low- ville area and the southeastern corner of the Antwerp sheet are cut by black dikes of hyper ite or diabase and by the gabbro rare narrow dikes of beerbachite. The hyperite dikes might be called metadiabase. Hyperite in this region is the name given to a rock that is interme- diate in composition between an olivine gabbro and norite, and hence typically contains both hypersthene and olivine in addition to other components. The rock varies from typical hyperite to diabase. The rock forming the stock on the northern border of the Lowville quad- rangle is typical hyperite. This is the largest mass of this character outcropping in this region, and it is noteworthy that dikes and sheets HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 93 of hyperite are most abundant within a ten-mile radius of this stock. The hyperite of the stock is a very tough rock, weathering dark gray, with a conspicuous diabasic texture. In thin section it is found to consist of interlocking laths of labradorite with pyroxene filling the intervening spaces, or of pyroxene and labradorite in ophitic inter- growth. Most of the pyroxene is diallage and the remainder is hypersthene. Olivine partly or completely altered to pyroxene is present and a considerable portion of the pyroxene is altered to an aggregate of hornblende and brown biotite. A small portion of the feldspar and pyroxene is granulated. The contacts of the hyperite with the inclosing gneisses are covered so that a study of their inter- relations could not be made, and the rock may be an older gabbro. In addition to the stock just described several other small lenses 'occur in the band of augite syenite gneiss south of Indian river, and innumerable thin sheets and narrow dikes usually not more than 20 feet wide occur in the syenite-granite complex, particularly in the porphyritic facies to the west, northwest and southwest. In the dikes olivine is absent and hypersthene may or may not be present. The absence of olivine in the smaller dikes is possibly accounted for by their more rapid rate of cooling and consequent lower temperature of beginning of crystallization. In the larger masses olivine crystallized out first and as far as possible reacted with the liquid at a lower temperature range to form the pyroxene, whereas in the smaller dikes pyroxene crystallized directly at the lower temperature. Locally the hyperite carries small to large phenocrysts of plagioclase and the rock has a porphyritic aspect. The magnetite grains are locally surrounded by an aureole of brown biotite. The beerbachite or fine-grained gabbro is sparse and rarely forms dikes over a few feet wide. They cut the gabbro and older rocks. They consist of a fine granular aggregate of polygonal-shaped grains, comprising about equal amounts of pyroxene and labradorite with accessory brown biotite and magnetite. Some dikes show up to 8 per cent of orthoclase or microperthite, and in some the pyroxene is partly altered to hornblende. The structure of the hyperite varies from wholly massive to almost completely pulverized depending upon its geographic location. DIABASE DIKES Within the immediate vicinity of the St Lawrence river diabase dikes are occasionally found cutting the Precambrian rocks (Cushing, ’25, p. 47), but they die out to the southeast and are very rare in the Hammond and Antwerp quadrangles. One such black dike is 94 NEW YORK STATE MUSEUM found about one and one-half miles northeast of Macomb, where it cuts the limestone and is approximately parallel to the structure. These dikes are believed to be the youngest of the Precambrian rocks in the region. AGE RELATIONS (GRENVILLE BELT) Metagabbro. The oldest known igneous rock in this region is the metagabbro of the Grenville belt. Except at the cores of con- siderable masses or in small masses isolated in limestone, this rock is always mashed and injected by younger intrusives or by pegmatite veins. East of Yellow lake (Hammond sheet) it is intimately intruded by the porphyritic granite (Hermon type) and east of Red lake and of Pleasant lake it is intruded by syenite. Anorthosite. The age relations of the small mass of anorthosite ■ and gabbroic-anorthosite on the Antwerp and Carthage quadrangles to the adjoining syenite and granosyenite is not known. The anortho- sitic rocks are cut by granite dikes, but no positive evidence was obtained on their relation to the syenitic masses, due to covered contact zones. Dioritic group. One and three-fourths of a mile southwest of Farley School (Hammond sheet) the diorite is intrusive into the metagabbro. One-half mile west of the north end of Grass lake a mass of dark gray diorite (monzodiorite) is intruded by red syenite and by aplitic granite believed to be offshoots from the bodies of Alexandria type to the west. Throughout the diorite and quartz diorite masses within several miles of Rossie there are locally many sheets of pink porphyritic granite that appear to have come in along the foliation planes. The biotite-oligoclase quartz diorite of the Antwerp quadrangle is intruded by granite (Hermon type). On the Lake Bonaparte quadrangle diorite is cut by quartz syenite of the Diana complex. Syenite (Grenville belt). As noted above, syenite is found intru- sive into diorite. The relations of the syenite to the porphyritic granite (Hermon type) are doubtful. On the Alexandria sheet Cushing (’io, p. 184) found the porphyritic granite in close associa- tion with a syenite and so related that he gained the impression that the whole represented a single intrusion. In the Hammond quadrangle the writer found the evidence con- fusing. East of Chapel Corners (Hammond area) there seems to be a gradation between the porphyritic syenite and a coarse red granite. At many localities the syenite is so closely associated with porphyritic granite and amphibolite that one is tempted to ascribe HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 95 its origin in part to a reaction between porphyritic granite and amphibolite. Opposed to both these ideas is the fact that syenite also occurs in small intrusive masses and narrow bands without any apparent direct association with either porphyritic granite or amphib- olite. In the belt northeast of Red lake (Hammond sheet) there are narrow bands of porphyritic granite parallel to the foliation of the syenite which might be interpreted either as intrusions or gradations as far as the writer was able to ascertain. In the main igneous com- plex of the Lowville sheet the porphyritic syenite similarly appears both to grade even though rather abruptly, into a granosyenite and also to be cut by porphyritic granitic and granosyenitic veinings parallel to the foliation. The granite (Alexandria type) and the syenite do not show grada- tional relations to each other as far as the writer has observed. Cushing (’25, p. 45-46, 69-70) has discussed at length evidence that he believes proves that the Reservoir Hill granite mass (Alex- andria type) shows gradation into syenite. At the south end of the Reservoir hill, east of Gouverneur, he finds that at the contact of the granite with the limestone the rock is green, is much less quartzose, carries pyroxene and is a quite characteristic sample of the acid phase of the ordinary green syenite of the Adirondack region. Cush- ing called the granite a member of the porphyritic group but the writer has examined the same locality and finds that the structural relations, local contact garnet rock, associated magnetite-bearing peg- matite veins and uniform fine to medium-grained nonporphyritic character together proclaim it to belong to the Alexandria and not the Hermon type of granite. The supposed syenite is the usual contact facies that domes of the Alexandria type of granite locally tend to develop adjacent to the country rock and is not syenite but more typically an acid quartz diorite. Wright (’23, p. 28) also states that the Mallorytown (Alexandria type) granite grades into granodiorite and into syenite. In the Hammond, Carthage, Lowville and Lake Bonaparte quad- rangles the writer has observed no gradation or direct genetic rela- tion between the syenites and the Alexandria type of granite. Granites (Grenville belt). The number and ages of the granitic intrusions in the northwest Adirondacks are subjects of controversy among geologists who have studied the problem. The biotite- oligoclase quartz diorite of the Antwerp quadrangle is excepted from the following discussion. C. H. Smyth jr was the first geologist to do systematic recon- naissance and detailed work in the northwest Adirondacks. In 1899 96 NEW YORK STATE MUSEUM he published a report on the geology of the crystalline rocks in the vicinity of the St Lawrence river. In this report he mapped two varieties of granite, one of which is called simply granite and the other granite gneiss. The former was subsequently named the Pic- ton granite and the latter the Laurentian granite by Cushing. As regards the age relations Smyth writes : As a whole, the granite is the more massive, and thus a younger looking rock than the gneiss, but when the former becomes foliated, as it often does, it is practically identical with the gneiss Certainly the writer could find no instances of the granite cutting the granite-gneiss, or of inclusions of the latter in the former, though inclusions of quartzite and schist are abundant. Cushing followed Smyth and did detailed mapping in the Thousand Island region (To, p. 36-43), where he postulated an older Lauren- tian granite (Alexandria type) and a younger granite, which he called the Picton. He also recognized the presence of a porphyritic granite associated with the Alexandria syenite, which he described as younger than the Laurentian granite, but with an unknown age rela- tion to the Picton, although he believed the Picton granite held inclusions of a porphyritic granite of similar type. Later (T6, p. 36-43), he did the geology of the Brier Hill and Ogdensburg quad- rangles and mapped a number of sills of porphyritic granite that he regarded as most certainly younger than the Laurentian and of the same type as the porphyritic granites that occur widely as differen- tiation facies of the syenite bodies in northern New York. His last work was the geology of the Gouverneur quadrangle (’25, p. 38-39), in which he summarizes his views regarding this problem and affirms the presence and relations of the three granites in the northwestern Adirondacks as follows: (1) An old granite classed as Laurentian (Alexandria type), always thoroughly gneissoid, gen- erally fine and even grained, usually with very small mica content, and containing amphibolite inclusions, often in great number, but with few or no other kinds of inclusions; (2) porphyritic granites (Her- mon type), which are regarded as younger than the Laurentian since on the Alexandria quadrangle the porphyritic granite is associated with a syenite mass that is in contact locally with the Laurentian granite and cuts it out across the strike and holds inclusions of it; locally this granite is nonporphyritic and while holding amphibolite inclusions in abundance also contains many inclusions of the various Grenville rocks; and (3) the Picton granite, which is described as a rather coarse-grained, nonporphyritic, red granite, which shows little sign of mashing or of metamorphism and in these respects as HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 97 well as in its demonstrable field relations to the granite-gneisses regarded as Laurentian shows itself to be unmistakably younger, as well as also younger than the porphyritic granites since it holds inclusions of a similar type. Cushing found none of the Picton granite on either the Brier Hill, Ogdensburg or Gouverneur quad- rangles. Miller worked extensively in the southern half of the Adirondacks and came to the conclusion that in that region there was evidence for only one period of great magmatic invasion. In 1916 M. B. Baker published a report on the geology of Kingston and vicinity in Canada. He postulated two granites, called respec- tively the Laurentian and Algoman. The Laurentian is described as a fine-grained pink to gray granite, always gneissic in structure and with syenitic facies. The Algoman is described as a coarse-grained granite with some syenite, which usually lacks the gneissic structure of the Laurentian. It is not found in contact with the Laurentian but contains large blocks of a Grenville-Laurentian combination as inclusions. He correlates the Algoman granite of the Kingston region with the Picton granite of Grindstone island. I11 1922 he again asserts the existence of the Algoman and Laurentian granites on the north of the St Lawrence river, and states that the Lauren- tian is cut by the Algoman. In 1923 J. F. Wright published a report on the geology of the Brockville-Mallorytown Map- Area, which lies to the northwest of the Hammond quadrangle across the St Lawrence river and covers in more detail a part of the same area mapped by Baker. Wright maps two granites, one of which he calls the Brockville and the other the Mallorytown. The Mallorytown is described as a medium to fine-grained pink to gray granite, both massive and slightly foliated, and more acid than the Brockville granite. Wright states that the granite grades into granodiorite and syenite, but that its relations to the Brockville granite could not be determined. The Brockville granite is described as a pink granite, mostly coarse-grained, with a slightly porphyritic aspect in a few outcrops. Wright found no evidence that the granites were of two ages. He traced the granite of the area (Mallorytown?) and found it to be the equivalent of both Cushing’s Laurentian and Baker’s Algoman. He further states that “neither Cushing nor Baker has positive evidence of two periods of granitic intrusion in this region, and both have exaggerated cer- tain lithological and structural distinctions in their mapping . . . . . in the Brockville-Mallorytown area in no case was granite or syenite found to cut each other granite grades into syenite or grano- 98 NEW YORK STATE MUSEUM diorite and foliated gneiss grades into massive granite and all the structural relations of the region can be explained on the basis of only one period of extensive granitic intrusion.” In 1926 the report by Buddington and Smyth on the Lake Bona- parte quadrangle was published. In this report the authors state that two granites are present but that their interrelations are unsolved. They suggest, however, that the fine-grained granite (Alexandria type) may be younger than the syenitic rocks, as dikes of fine-grained granite are found in the latter. In summary, we find that lithologically three types of granite are present in the Grenville belt of the Northwest Adirondacks — a coarse porphyritic variety (Hermon type), a coarse to medium equigranular granite and a fine to medium-grained type (Alexandria). A great divergence of opinion arises with respect to the age relations of these granites. Cushing and Baker believe that the granites are of widely different ages and that the medium to fine-grained granite of the Alexandria type is the older and of Laurentian age. Miller and Wright, on the other hand, believe that all the granites belong to but one great period of magmatic invasion. Smyth and Buddington were not able to determine decisively the relations of the granites and were noncommittal. In connection with a study of the mode of intrusion of the Alex- andria type of granite in the northwest Adirondacks, the writer spent two days in a study of the evidence cited by Cushing to prove the Alexandria type of granite older than the Picton and Hermon types. Cushing states that the Alexandria syenite cuts out the Laurentian granite along its strike and holds inclusions of it. To the writer the evidence is equally good that the granite cuts out the syenite along its northwest contact instead of the reverse as stated by Cushing. The writer saw no evidence of the syenite containing inclusions of granite or of syenite cutting the granite. He did find a number of dikes of aplitic granite from two inches to ten feet wide cutting the syenite. These may be significant of a younger age as no such dikes were seen cutting the main so-called Laurentian mass. The writer could find no dikes of syenite in the narrow tongue of granite in the north central portion of the Alexandria syenite. It has the appear- ance of a dike parallel to the foliation and finally the syenite has a pronounced cataclastic structure whereas the granite does not show similar cataclastic effects. Through the courtesy of Professor Smyth, the writer was enabled to examine a number of thin sections from many localities of the Picton granite on both Wells and Grindstone islands and also of the HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 99 so-called Laurentian granite on the south side of Wells island. In the Picton granite the borders of the feldspars have been completely pulverized by crushing and the quartz has been granulated as a result of mashing. The larger residual grains of feldspar are also tra- versed by veins of crushed material. The rock shows a very marked cataclastic structure. In the so-called Laurentian granite, on the other hand, the quartz is in lenses but shows no similar cataclasis. The feldspar is in grains of about equal size and many of the grains are polygonal in shape but there is no evidence of powdered edges or veins of crush material. The rock does not in any sense have a cataclastic structure. Although it may be interpreted as having a protoclastic structure. This structure is characteristic of the whole Alexandria batholith. Each rock is consistent in its characters throughout its extent and the contrast is definite and marked. Taken by itself alone it might suggest either that the so-called Laurentian granite instead of being older is actually younger than the Picton and deformed while still hot, in contrast to the older Picton granite, or that the granite (Laurentian) was actually older but was not affected by the cataclasis which affected the Picton type. It may be that the Picton granite is mashed not because it is older but because it was in a crush zone. Further study of the so-called Laurentian granite on the north side of Wells island and its relation to the Picton is needed. The texture of the Alexandria mass may be crystalloblastic. Along the west side of Black Creek valley and at several other localities coarse-grained and porphyritic granite masses are cut by dikes of granite that are very similar to the Alexandria type, but again these may merely be aplite dikes genetically related to the granite they are cutting. Cushing (To, p. 183) has shown that the Picton granite is so similar chemically and mineralogically to the porphyritic granites “as to be almost grotesque,” but he believes nevertheless that the Picton is younger. The writer, on the other hand, believes that the Picton is probably a nonporphyritic or rather an indistinctly porphyritic I facies of the porphyritic granites (Hermon type). This is based upon the similarity in structural relations and upon the fact that Cushing was unable to find other masses of the Picton granite any- where in the Brier Hill, Ogdensburg, Alexandria Bay or Governeur I quadrangles where the porphyritic masses are common. As a result of his studies in the Lake Bonaparte, Hammond, Ant- werp and Lowville quadrangles and a review of Cushing’s data, the writer concluded that at the present time there is no body of well- established evidence contrary to the hypothesis that in the Grenville TOO NEW YORK STATE MUSEUM belt of the northwest Adirondacks the granites all belong to the same cycle of magmatic activity ; that the porphyritic granosyenite granite (Hermon type) is the older; that the medium to coarse-grained granite, which forms small masses or long narrow bands, particularly on the Antwerp quadrangle, is a biotite more siliceous, facies of the porphyritic type with which it is intimately associated; and that the fine to medium-grained granite (Alexandria type) is similarly an alaskitic-like facies somewhat younger than the porphy- ritic types. Subsequent to the submission of the manuscript of this report to the printer, the writer studied the northeast end of the Diana com- plex on the Oswegatchie quadrangle ; and at one locality three-fifths of a mile west-southwest of Red School, and at another on the road to Kalurah, one-half mile southeast of where it enters the Oswe- gatchie quadrangle from the west, he found positive evidence of syenite (quartzose chill facies of the Diana complex) intruding and containing fragments of a pink fine-grained syenitic aplite. The aplite contains some disseminations, streaks, and veinings of ferro- magnesian mineral resulting from assimilation of limestone which occurs near-by on the west and consists of microcline, orthoclase, albite and a little accessory quartz. It is possibly related to granitic aplite and modified to a syenitic facies consequent upon intrusion into and reaction with the limestone. Syenite belonging to the Diana complex is elsewhere positively cut by granite, so that there is little doubt of at least two ages of acid rocks in the northwest Adirondacks. This raises again the question as to whether all the Alexandria type of granite may not belong to this older period of intrusion as sug- gested by Cushing and Baker for a number of masses of this type. Certain sill-like and injection gneiss complexes of fine-grained aplite associated with pyroxene gneisses and amphibolite, such as form the northwest border of the Diana complex northeast of Harris- ville and a narrow band east of North Croghan on the Lake Bona- parte quadrangle are certainly older than the syenite. Similar nar- row bands of aplite and mixed rock, such as that south and south- west of Fargo on the Antwerp quadrangle and the belt through Mount Tom on the Carthage and Lowville quadrangles, may be older than the syenite and granosyenite, although satisfactory proof is lacking. Some of the masses of Alexandria type of granite show evidences of intense deformation, others show but little. The writer has found no satisfactory direct evidence of the age relationships to other intru- sives of any of the major masses (Alexandria type) in the area occu- HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES IOI pied by the Grenville series, and is not prepared positively to assign them to any definite age or ages. The available evidence is of such character that they might be either pre-syenite or post-syenite, or part one and part the other and with similar indeterminate relations to the Hermon granite. AGE RELATIONS (MAIN IGNEOUS COMPLEX) Introduction. Two igneous complexes may be distinguished within the main igneous portion of the Adirondack massif as exposed in this region. They have been previously designated as the Diana and Croghan masses or complexes respectively by the writer. The rocks of the Diana complex are for the most part characterized by strong cataclastic or protoclastic structures, and by a porphyritic or lenticular texture. The rocks of the Croghan complex are characterized by an aclastic or locally slightly protoclastic struc- ture and an equigranular texture. The equigranular syenite north and north-northeast of Monnatt School on the Lowville quadrangle, however, may belong to the Diana complex, as also that on the south- east corner. All the rocks of both complexes' usually exhibit equally well a more or less well defined foliation. As seen in plan on the areal map of the Lowville quadrangle, the extension of the structural trends of the Diana complex appear locally to be crosscut by the belt of granite and granosyenite of the Croghan complex. For this reason, and others previously discussed, the Lowville granite and associated granosyenite are considered intrusive into the rocks of the Diana complex. Sills of granosyenite presumably belonging to the Croghan complex are interleaved with the syenite and porphyritic granosyenite of the Diana complex along its southeast border. The Croghan complex as exposed on the Low- ville sheet is but a small portion of a very large body which extends to the south, east and northeast. Discussion of the mode of intru- sion of this mass and its structural form must await further study. Diana complex. From evidence on the Lake Bonaparte quad- rangle, it appears to the writer that the Diana body may be tentatively interpreted as an overturned, isoclinally folded, differentiated strati- form sheet. The sheet was differentiated previous to deformation and consists in the lower portion of a chill facies represented by a quartz- ose augite syenite followed successively upward by basic green augite or augite-hypersthene syenite (for example, near Natural Bridge) which grade successively upward through green quartzose augite syenite and red quartzose hornblende syenite into hornblende granosyenite or granite in the upper part (for example, near North 102 NEW YORK STATE MUSEUM Croghap). Locally thin shonkinite bands and lenses occur in the basic augite syenite on the Lake Bonaparte quadrangle. Pyroxene gneiss inclusions, which occur along the line from Mount McQuillen, Fargo, North Croghan and east, suggest the possibility that they represent metamorphosed roof remnants in a closely com- pressed synclinal axis. Another synclinal axis appears to be indi- cated running northwest from Croghan and again by the mass of porphyritic granosyenite south of New Bremen. The thickness of the sheet along a line from Natural Bridge to North Croghan may be of the order of magnitude of about 12,000 feet. The undiffer- entiated magma is thought to have been of a syenitic composition, possibly a quartz-bearing augite syenite, such as forms the facies at the base and is well shown on the Lake Bonaparte quadrangle near Harrisville. Croghan complex. The mutual relations of the individual rock types of the Croghan igneous complex, like that of the Diana, offer many perplexing features. In passing across the strike, alternating bands of two or more types of rock are frequently crossed in rapid succession ; yet, whereas the change from one rock to another is very abrupt, distinct evidences of intrusion are usually lacking for the bands maintain a uniform parallelism and only here and there can positive evidence of crosscutting relations be obtained, whereas frequent examples of sharp transitions are found. The rocks mapped as mixed gneisses afford the most striking example. The change from granite or granosyenite to hornblende or altered augite syenite is so abrupt as to suggest that the syenite had been solid and the granitic material liquid at the time of intrusion ; yet the contact, on the other hand, is not a clear cut line and crosscutting contacts are practically absent. Quite anomalous cases of apparently mutual intrusion show in repeated instances at one locality in this district. Fine-grained green equigranular hornblende syenite, consisting predominantly of micro- perthite, oligoclase and hornblende, has the appearance of dikes in the granite and yet is itself cut by the selfsame granite. The gra- nitic dikes are not pegmatitic nor aplitic but are similar to the main mass in every way. Two such instances are illustrated in the accom- panying diagrams (figures 7 and 8). In another case the hornblende syenite contains inclusions of pyroxene-biotite gneiss and is itself intruded by granite. In still another case a dike of this green horn- blende syenite is found intruding a red porphyritic hornblende sye- nite and is itself crossed by dikes of the red syenite. It is a possible HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 103 IE3 e3 r v— Ja cm a eatra c3 E3 Q C3 Q CT3 Porphyrit i c Grano- Syenite Granite mm Hornblende Syenite Figure 7 Hornblende syenite cutting across foliation of granite and a sheet of porphyritic granosyenife, but terminating in a line of frag- ments in the granite Figure 8 Dike of hornblende syenite intruding granite, and itself brecciated and veined by granite 104 NEW YORK STATE MUSEUM point of significance that these freak features occur in a zone of very sharp curvature of the bands of gneiss themselves, in the vicin- ity of Croghan.1 The augitic syenite bands occurring within the Croghan complex may be interpreted as an earlier facies of the Croghan complex or as included remnants of the older Diana intrusives. The band of sye- nite northeast of Kirschnerville appears as though isolated and included by intrusion of the granite. The quartz-bearing syenite in the southeastern part of the Lowville quadrangle is of such a minera- logical character that it might be interpreted as a chill facies of the Diana intrusives. Further data are necessary, however, for an adequate discussion of this problem. As previously mentioned, the relative age of the gabbro with large labradorite phenocrysts and the Lowville granite is not abso- lutely certain. Locally granite and gabbro interfinger in such an intimate fashion parallel to the foliation as to defy attempts to make certain which is younger. The gabbro, however, is cut by granite pegmatite veins. The hyperite dikes and sheets cut all varieties of the Diana complex, as well as the granosyenite which appears to be a border facies of the Croghan complex along its northwest edge. At one locality hyperite appears to be cut by granite pegmatite veins which may be related to the Lowville granitic rocks, and at another locality the hyperite is cut by granite related to a granite of the Grenville belt. There are at least two ages of basic intrusions in this region, a pre-syenite gabbro and post-syenite hyperite. MAGMATIC STEMS AND LINES OF DESCENT Two. distinct stems (Goldschmidt, ’22) or lines of descent through magmatic differentiation are represented among the igneous rocks of the areas covered by this report. The first and most important one, quantitatively, is the syenitic series, whose members constitute the great main igneous complex of the core of the Adirondacks. A suite of intrusives which are a variant of this line of descent comprise the syenitic and granitic rocks of the belt marked by the predominance of the Grenville series in the Northwest Adirondacks. The second distinct line of descent is represented by the dioritic rocks near Rossie and the quartz diorite belt of the Antwerp quadrangle. 1 These relationships would be interpreted by some geologists as a case of palingenesis; a refusion of the granite with some movement and intrusion into the dike rocks, still solid because of higher melting intervals. The writer is not prepared to discuss this possibility further at this time. HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 105 The writer has discussed the origin of the Adirondack magmatic stem comprising the intrusives of the Adirondacks as a whole, on a physicochemical basis, in another paper (Buddington, ’31). A comparison with igneous rocks of the charnockite series of India, the Alaska Coast Range and Sierra Nevada batholiths of western North America, and varied members of the Mull (Scot- land) intrusives, shows that a characteristic feature of the Adiron- dack syenitic intrusives is the relatively prolonged delay in the for- mation and separation of quartz. The main Adirondack line of descent is believed to be characterized by the separation of pyrox- ene instead of hornblende, and is interpreted as the result of differ- entiation in much drier magmas in the early and early intermediate stages of its history when compared with the Alaska Coast Range and Sierra Nevada lines of descent. Within the belt characterized by the predominance of the Gren- ville series (Antwerp and Hammond quadrangles), there is a series of intrusives with similar chemical compositions and variations to the syenitic series of the main igneous complex, but differing from them in some mineralogical details. Hornblende and biotite take the place of pyroxene in these outlying masses at much earlier stages of differentiation, microperthite crystallized separately as microcline or orthoclase and plagioclase to a greater degree, and muscovite appears in sparse amounts in some of the granite. These phenomena suggest formation from a series of magmas with a relatively higher volatile content. It might be deduced that the higher volatile con- tent of the magmas associated with the Grenville beds was brought about through accession from the sediments. Another interpreta- tion would be that only the more fluid (because of higher water content) facies of the syenitic magmas were able to be forced freely far into the Grenville sediments, analogous to the fact that many of the smaller offshooting dikes and lenses within this same belt are of pegmatitic character, the most quartzose facies, or accompanied by phenomena indicating a higher content of volatiles. Or, more probably they are a facies of the Lowville granite, modified as a result of intimate intrusion in the Grenville. The Rossie line of descent is markedly different from that of the Adirondack type. The most siliceous rock is a biotite quartz diorite with insignificant potash feldspar ; whereas in the Adirondack series it is a biotite granite rich in potash feldspar. The intermediate members of the Rossie series are types of diorite and quartz diorite instead of syenites and granosyenites, as in the Adirondack stem. The Rossie series is characterized by the association of biotite and io6 NEW YORK STATE MUSEUM pyroxene in the same rock in gabbro, diorite and quartz diorite. In the more siliceous biotite quartz diorite a little muscovite accom- panies the biotite. The percentage of quartz increases rapidly as biotite takes the place of hornblende, and the plagioclase becomes in general slightly more alhitic. The Adirondack syenitic lines of descent are compared with the Rossie dioritic suite and the Alaska Coast Range intrusives in the following table, using as a basis a similar plagioclase. It is obvious that the Rossie line of descent is characterized by the relatively early formation of biotite and quartz. The Rossie series of rocks is similar to the Trondjhem suite of intrusives described by Gold- schmidt in Norway and ascribed by him to an origin in magmas high in content of volatiles. Comparison of several series of intrusives based on equivalent composition of plagioclase PLAGIOCLASE Coast Range Intrusives (Buddington, 1929 b ) Rossie Dioritic Series Adirondack Syenitic Series (Grenville Belt) Adirondack Syenitic Series (Diana Complex) Bergen-Jotun Mangerite Series (After Goldschmidt) ALBITE Biotite Granite and Quartz Diorite Biotite Granite Diopside and Aegerine Granites OLIGOCLASE Biotite Quartz Monzonite Biotite Quartz Diorite (a little muscovite) Hornblende- Bio tite Granosyenite and Biotite Granite (a trace of muscovite) Hornblende Granosyenite Diopside or Hypersthene Granite Hornblende- Biotite Quartz Monzonite Pyroxene or Hornblende- Biotite Quartz Diorite Hornblende- Biotite Syenite Augite or Hornblende Syenite Quartzose Syenite Diopside- Hypersthene Syenite and Pyroxene Mangerite ANDESINE Hornblende- Bio tite Granodiorite and Quartz Diorite, and Hornblende Diorite Pyroxene- Biotite or Hornblende- Biotite Diorite and Quartz Diorite Basic Augite- Hypersthene Syenite Augite- Hypersthene Mangerite LABRADORITE Gabbro Gabbro Gabbro METAMORPHIC AND MIXED ROCKS INTRODUCTION The rocks of the Grenville series underlie the larger part of the Hammond and Antwerp areas, and consist exclusively of meta- morphic, highly crystalline types. Predominantly the Grenville gneisses and schists are so intimately permeated by and interleaved with granitic pegmatite or aplite, parallel to the foliation, as to con- HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 107 stitute injection gneiss or arteritic migmatite (figure 41). More homogeneous masses with a granulose texture, such as some of the pyroxenic and hornblendic gneisses southwest of Nelson Corners (Hammond quadrangle), are cut by a reticulating network of peg- matitic veinlets so as to constitute a breccia (figure 42) or agmatite (Sederholm, ’26). The gabbro mass east of Pleasant lake is simi- larly broken on a coarse scale by pegmatite veinlets. Locally amphi- bolite belts are interleaved on a coarse scale by granite or syenite, so as to constitute mixed rocks. The crystalline limestones have for the most part been to some extent thoroughly permeated by the more mobile magmatic solutions, and are locally thin-banded with veins of quartz or pegmatitic nodules, or with layers partly to completely replaced by disseminated silicates ; but there is no fine scale peg- matitic injection either of the arteritic or agmatitic type. In local belts the igneous rocks, such as gabbro, syenite or granite, are so interleaved with included bands of limestone as to constitute mixed rocks. In many cases it is difficult or impossible to determine the extent to which the present rocks are the result of recrystallization, on the one hand, or of additions from and replacement or alterations by magmatic solutions, on the other. There are also gradations between intrusive rock with only an included shred of Grenville rocks, on the one hand, to Grenville material with only a little peg- matitic injection or impregnation on the other. The mapping and discussion of the rocks and the determination of exact boundary lines for different kinds are therefore in many cases subject to con- siderable personal judgment and in part open to other interpretations than the one selected. PYROXENIC GNEISS BELT BETWEEN BLACK AND GRASS CREEKS (HAMMOND SHEET) The mixed rocks between the Paleozoic beds on the west and Grass creek and the arm at the head of Black lake on the east con- sist predominantly of gneisses and crystalline feldspathic schists more or less thoroughly permeated and injected by pegmatite and aplitic veins parallel to the foliation and intruded by many masses of alaskitic granite, diorite and quartz diorite and by abundant sheets of pegmatite. The pegmatite veins are often so abundant and stand out in such relief in the topography that the whole belt when viewed from a distance appears to consist of them. Con- siderable white quartzite and some quartz-veined limestone beds are also found interbedded with the gneisses. Throughout the belt io8 NEW YORK STATE MUSEUM southwest of the Oakvale quartz diorite mass the gneisses and granite are often so intermingled that it is practically impossible to distinguish between areas of granite with inclusions of gneiss and gneiss with intrusions of granite, there being so much gradation from one to the other. Pyroxenic gneisses are the predominant rocks, although biotite- feldspar schists, usually carrying sillimanite, form a major mem- ber of the series. Hornblendic gneisses occur associated with the pyroxene-bearing varieties. Biotitic orthoclase or microperthite gneisses with more or less quartz are also found. Locally there are metamorphic replacement masses of a granu- lose texture consisting of pyroxene, microcline, and calcite ; or of pyroxene and scapolite ; or of pyroxene with about 30 per cent plagioclase. Titanite is a common accessory in these rocks. All the gneisses with the exception of a few specimens show a markedly cataclastic structure. The feldspars, scapolite and pyrox- ene are all mylonitized and the hornblende is sliced and strung out along the foliation. The quartz is in flat leaves parallel to the folia- tion and is likewise granulated. Pyroxene with twinning bands and microcline formed from orthoclase are both common and both have resulted from pressure. BIOTITIC GNEISS AND QUARTZITE BELT EAST OF BLACK LAKE AND GRASS CREEK East of Grass creek and Black lake there is a wide belt of gneisses and white quartzite with interbedded limestone, the whole intruded by aplite sheets, abundant pegmatite veins and the Rossie diorite and quartz diorite masses. The predominant rocks of this belt are biotitic feldspar gneisses with associated white quartzites. The gneisses form the greatest bulk of the rocks and in them the pre- dominant feldspar is usually a potash variety, either orthoclase, microcline or microperthite, although plagioclase gneisses are also common. Quartz varies in amount from a minor accessory in many of the gneisses to the major constituent in the quartzites. Pyrox- enic gneisses, however, are also found in this belt particularly near limestone beds and granulose pyroxene — scapolite beds are also common. The prevalence of biotitic gneisses, however, distinguishes this belt from that to the southwest in which pyroxenic gneisses predominate. Only a few thin sections from the belt east of Black lake and Grass creek have been examined. These indicate a very variable structure. ro Crystalline Lim«eten« Gneiss Figure 13 Generalized structure section of Rossie isoclinally folded instrusive complex, Hammond quadrangle The syenite mass may in turn be a relic of a sill intruded in the upper portion of the gabbro sheet. The foliation of the granite together with the included bands of Grenville around the south end of the syncline, west of Scotch Settlement School have a very steep to vertical dip. The limestone bordering the south edge of the granite similarly has a vertical to steep (70 degrees) northerly dip. The evidence for a syncline at this end is therefore not as good as at the north but the foliation and the Grenville inclusions change their strike as around the end of a fold. The axial plane of the syncline dips 80 degrees or more to the northwest. Within the long granite masses north -northeast and southwest of Laidlow School there are several localities where the arrangement of the strike and the directions of dip suggest that we are dealing here with several sills of granite which have been folded on an HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 151 anticlinal axis with locally a northeast pitch of 45 to 50 degrees. About three-quarters of a mile northwest of Scotch Settlement School there is a band of thin-layered limestone and gneiss closely compressed, puckered and plicated (figure 43). The anticline one-half mile east of South Woods School is splen- didly exposed in every detail. It strikes about north-south and pitches 40 to 45 degrees south. The foliation of the quartz diorite and granite conform to the anticlinal structure and have every appearance of representing a folded composite sill comprising quartz diorite, granite and aplite. This structure has been studied in detail and the writer believes it gives a clue to the type of structure charac- terizing the Rossie complex. Adjoining the anticline on the east the igneous sill shows a very definite synclinal structure that may represent the northwestward extension of the Pleasant Lake syn- clinal axis. A synclinal structure pitching 50 to 60 degrees southwest is very clearly shown in the Grenville formations about one-half mile south of Stark School, but its southern extension as shown in figure 10 is based largely on inference and probabilities. The structure northwest of South Woods School has not been accurately determined. The gneiss between Bostwick creek and Black lake, however, appears to be exposed on an anticlinal core pitching southwest. Between Rapids School and Farley School critical areas are covered and the anticlinal axis is largely based on inference. The gabbro masses appear to comprise sills and lenses intruded at a different horizon in the Grenville than the Pleasant Lake gabbro. The anticline southwest of Rossie is also very well exposed, and the southern half is especially well brought out by the layer of Grenville that is included within the composite sill of granite and quartz diorite. The foliation of the igneous rock is conformable with the bedding of the Grenville. The anticline pitches about 50 degrees south at the southern end. The structure of the igneous rock at its northern termination a mile west of Rossie is puzzling. The strike appears to conform to the boundary and the dip to be vertical. The writer is unable to interpret the significance of this. North of Rossie between the Indian river and the granite mass there are gneisses locally thin-layered with limestone. The pitch of corrugations on the bedding surfaces of these gneisses is 60 to 70 degrees south. Between Mile Arm bay and the laccolith of biotite pyroxene quartz diorite on the west there is a well-marked anticlinal axis in the Gren- 152 NEW YORK STATE MUSEUM ville formations which here consist of interbedded white quartzite and pyroxenitic gneisses with intercalated beds of limestone. Both limbs of the fold dip 70 to 80 degrees west and the anticline pitches about 50 degrees south. Pegmatites form an important part of the formations here and locally must form 75 per cent or more of the mass. They have been affected by the pressure that aided in folding the Grenville for the foliation of big (10 to 15 feet wide) dikes that cross the bedding is parallel to the foliation of the Grenville. The extension of this anticlinal axis to the northeast towards Lonesome bay is largely hypothetical. The microstructure of the gabbro, diorite and quartz diorite of the Rossie complex varies from highly mashed cataclastic rock along long narrow crush zones to massive material ; the latter with no evidence whatever of crushing or stress except a ribboned wavy extinction in the quartz. Mashing is occasionally very prominent along the border zones of the igneous masses. The foliation is of primary magmatic origin and not related to the degree of cataclastic crushing. In some of the rock the foliation is quite indistinct. The gabbro mass of the Pleasant Lake syncline is intensively broken up and brecciated by granite pegmatite veins, but the microstructure of much of it shows no cataclastic phenomena. Much of the diorite and quartz diorite show evidences of cataclastic effects in the periph- eral granulation of the constituent grains as observed with the microscope. Mosaic faulting on a small scale was noted at several localities. No protoclastic structure was seen in any of these rocks. The writer believes that the gabbro, diorite, quartz diorite and syenite sill complex was folded after complete consolidation. Most of the granite directly associated with the dioritic rocks is equigranular and medium to coarse-grained indicating a more quickly cooled facies than the coarse porphyritic granite that forms the eastern and southern portion of the Pleasant Lake syncline and the large sheets to the southeast. The foliation of the granite is of primary magmatic origin, for there is no crushing or alteration in most of it. Locally, however, there are zones where the granite has been severely crushed. The amount of crushing and faulting in the granite is so small that it seems as if the folding of the granite sills must have occurred before their complete consolidation. Overturned folds dipping southeast. The structure of the rocks between Lake of the Woods and South Hammond, south of the quartz diorite laccolith, is so complicated that it has not been unrav- eled. The foliation of both the granite and of the Grenville in gen- HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 153 eral strikes northeast and dips southeast and this taken together with the presence of a few observable small folds suggest the proba- bility of isoclinal folding overturned towards the northwest. Along the western border of the quadrangle from the Potsdam sandstone south to the Grass Lake belt of limestone the dip is almost uniformly about 60 degrees southeast; farther to the northeast, west of Nelson Corners the dip is usually about 70 degrees southeast. Just west of the area of Potsdam sandstone, a mile northwest of Nelson Corners there is some evidence of an overturned synclinal structure with axial plane dipping 70 degrees east-southeast but the data at hand are not decisive. Along a narrow belt just west of Grass lake and Grass creek to north of Nelson Corners the beds dip 80 degrees southeast or vertical. In the vicinity of Pope Mills the dip is also in general uniformly to the southeast and the general plan of the rocks suggest a number of folds overturned towards the northwest. A synclinal axis has been indicated in figure 10 passing west of Bigelow School. The strikes and dins about one-half mile northeast of Bigelow School seem to indicate a synclinal structure with a steep pitch northeast, although the beds have not been systematically mapped in sufficient detail to prove this. For a width of a mile across the strike the beds have a dip of about 35 to 45 degrees southeast, strongly suggesting folds overturned toward the northwest. The syenite sill north of Pope Mills also shows evidences of iso' clinal folds overturned towards the northwest, with the dip of the foliation varying from 55 to 70 degrees southeast, across a width of three miles at right angles to the strike. On the Brier Hill quadrangle to the north the foliation of the Macomb granite mass similarly has a dip of 60 to 80 degrees almost uniformly to the southeast. Continuous with this belt of southeast dipping folds, there is to the southwest on the Alexandria Bay quadrangle (Cushing,' To, p. 109) a belt about four miles wide in the vicinity of Butterfield lake in which the dip of the bedding or foliation is uniformly about 60 degrees southeast. This includes the Alexandria syenite mass. There is thus a belt about 25 miles long and two and one-half to four miles wide, as exposed, stretching northeast-southwest across the Brier Hill, Hammond and Alexandria Bay quadrangles, in which the bed- ding or foliation dips almost uniformly to the southeast and in which folds overturned toward the northwest have been observed. Within this general belt there is but one fold which has a northwest dip. 154 NEW YORK STATE MUSEUM This lies between Mile Arm bay and the quartz diorite laccolith on the west, and its steep northwest dip, unusual for its location, is doubtless due to the effect of the resistant igneous mass. Secondary folds. The limbs of the major folds often show secondary folds and crumples, which may be interpreted as due to the result of differential movement of one member with respect to another with the consequent formation of drag folds in the material between the two moving sheets or to local crumpling of the rocks where they have been oriented parallel to the orogenic stress as a result of rising anticlines or domes. The question may well be raised as to whether even folds on such a scale as those of the Rossie complex and the Pope Mills syenite sheet are not due to differential movement of a northwest plate with respect to one on the southeast, producing drag folds within the intervening mass. The limestone could act as a sufficiently mobile and plastic mass to afford the necessary accommodation for such movements. On the Gouverneur quadrangle, also, a fine example of a secondary fold is shown northeast of Jonesville School, where limestone, garnet gneiss and quartzite, granite (Hermon type) and amphibolite are all involved in a sharp cross fold pitching 40 to 45 degrees north-north- east. Secondary folds may play an important part in the structure of this region. Folded sills. With the exception of the granite (Alexandria type) phacoliths, practically all the intrusive igneous masses in the Grenville belt are here interpreted as having essentially a sill or sheet structure, in part locally transgressing the bedding at a small angle; but in general parallel to the bedding and in large masses with great extension parallel to the strike as compared with the width. Three small granite (Alexandria type) sills are also present. One of them, at the north end of the Payne Lake granite phacolith, is obvi- ously a subsidiary intrusion connected with the intrusion of the main granite body. The sill of aplitic granite north of Farley School is folded in a very conspicuous manner, and it is probable that the curve at the southwest end of the sill two miles west of Natural Dam also indicates folding. The folding of other igneous sills has already been discussed in connection with a description of the folds within the Grenville formations and the Rossie complex. The sheet or sill structure of the gabbro and granite masses (Hermon type) has long been known and the folding of the gabbro sills has been conclusively proven by Martin (T6, p. 96-108) but the idea that the granite (Hermon HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 155 type) sheets have in part similarly been folded is new (Buddington, ’29a) and further confirmatory evidence is necessary. The evidence is not absolutely conclusive as to whether the granite (Hermon type) sills have been intruded into a series of previously folded beds in approximate conformity with their structure, or whether they were intruded contemporaneously with the folding and themselves subjected to orogenic pressures and folded along with the Grenville rocks. The granite is continuous around the nose of the Somerville anti- cline northwest of Philadelphia and around the end of the Sherman Lake syncline south of Oxbow. Three miles northwest of Antwerp, the granite shows a peculiar accordion pattern where it occurs at the end of a northeastward pitching minor syncline. The same type of structure is exhibited at the northeast end of the Lewisburg syn- cline (figure 11 ). On the Gouverneur quadrangle, northeast of Jonesville School the Moss Ridge granite sheet is continuous with the Battle Hill sheet and shows a secondary syncline and sharp anticline pitching 40 to 45 degrees north northeast. A lineal structure with similar pitch also characterizes the granite itself. It is conceiv- able that these structural relations might arise from control exerted by preexisting structures. It is certain, however, that the granite has been subjected to strong pressure, as indicated by the pro- nounced protoclastic structure throughout, and therefore probable that intrusion was contemporaneous with at least the later stages of folding. In view of the abundant evidence that the gabbro, diorite, quartz diorite and syenite sills have been folded, and that the granite (Hermon type) shows in part similar phenomena, the writer believes that the balance of evidence favors the conception that the granite (Hermon type) was intruded into beds that were folded and dis- turbed to a moderate extent and subsequently subjected to continued folding along with them. Granite (Alexandria type) phacoliths. The Payne Lake, Dodd’s Creek, Hyde School and Hickory Lake granite masses all show a foliation parallel to their borders and the latter two bodies exhibit a typical domical foliation. A study of these and similar granite masses throughout the northwest Adirondacks has led the writer (’29a) to propose the hypothesis that they are phacoliths. The granite bodies on this quadrangle show only part of the phenomena that have led to the interpretation of their structural relations. These phenomena are similar, however, to those found in the masses with structural relations better exposed in other quadrangles of the northwest Adirondacks. NEW YORK STATE MUSEUM 156 The Hyde School granite mass shows very well the domical char- acter of the foliation and the domical arrangement of included beds of amphibolite (metasomatically replaced limestone). The major structure of the body is a dome elongate in a general northeast-south- west direction with the west limb dipping 30 degrees west to north- west and the east limb dipping 45 degrees southeast to east on the edge of Birch Creek valley, which completely encircles all but the north end. The steeper dip on the southeast limb is in accord with the hypothesis that the major stress came from the northwest. Along the axis of the dome there are several minor gentle cross folds oriented with a northwest-southeast direction. One such fold is very well shown in the vicinity of California School and results in a minor domical structure to the foliation superimposed on the major dome or anticline. The granite west of California School shows a marked lineal structure or pitch best brought out by the quartz with a strike north 350 west or in the general direction of the axis of the cross fold. The foliation at the southwest end dips about 30 to 45 degrees southwest. The foliation on the crest of the major dome is flat except on the limbs of the minor cross folds. Throughout the granite there are thin included layers of amphi- bolite that conform to the foliation of the granite. Locally these are folded back on themselves as the result of the interaction of the forces that produced the major and minor folds (figure 44). Unfortunately the valley of Birch creek is partly filled with strati- fied drift that covers the contact of the granite and the country rock. The Grenville formations, however, on the outer side of Birch Creek valley all dip away from the contact zone in conformity with the structure of the granite. A zone of interbedded limestone, quartzite, gneiss and schist appears to swing com- pletely around the southern part but around the north end the structure is obscured by the Potsdam sandstone and areas covered by silts and swamp. The microstructure of the Hyde granite is remarkable for the absolute absence of any evidence whatever of cataclastic structure, and the relatively rare occurrence of evidences of protoclastic structure. The texture as a whole is that due solely to primary interlocking crystallization from a magma. The Hickory Lake granite mass similarly has a domical foliation with included layers of amphibolite conformable to the structure. At the south end there is much minor folding of the amphibolite layers with the foliation of the granite in conformity. The axes of these folds is parallel to the elongation of the dome and the crumples pitch HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 157 south. Some of the pegmatite dikes cross these folds and are them- selves unaffected by the crumpling. The Dodds Creek granite mass has an anticlinal structure as indi- cated by the foliation but the limbs dip almost vertical and the anti- clinal character is shown clearly only at the ends, where the strike of the foliation swings around each nose and dips about 60 degrees north at the northeast and 45 degrees west at the southwest extrem- ities respectively. The west limb dips 80 to 90 degrees west and the east limb about the same. The relations of the granite to the surrounding country rock are completely obscured. The Dodd’s Creek granite, unlike the Hyde granite, shows a pronounced cata- clastic microstructure. The whole mass is severely mashed. The structure of the Payne Lake granite mass has not been care- fully studied. As in the other masses the strike of the foliation is peripheral to the borders. The few observations available suggest an overturned fold with the axis dipping northwest. The series of granite masses thus described, as far as their folia- tion and included layers of amphibolite are concerned, seem to illus- trate anticlinal structures increasingly overturned towards the south- east. The foliation of the Hyde mass has a relatively broad open asymmetrical anticlinal form with a gentle dip 30 degrees on the northwest limb and a steeper dip (45 degrees) on the southeast limb. The foliation of the Dodd’s Creek mass has the character of a closely compressed anticline with dips of 80 to 90 degrees on the limbs. Finally the foliation of the Payne Lake mass has the form of a fold overturned toward the southeast with its axial plane dip- ping northwest. It will be further noted that the structure of the foliation of the Payne Lake and Dodd’s Creek granite masses is not independent of the general structural features of the country rock but is in conformity with them. Thus the Payne Lake mass lies in a zone of folds strongly overturned towards the southeast and the Dodd’s Creek mass in a zone of closely compressed folds with approximately vertical axes intermediate between the region of folds overturned toward the southeast and a zone to the northwest where the folds are overturned toward the northwest, and where the folia- tion of the granite mass along Black creek dips uniformly southeast. These facts indicate very strongly that the foliation and its struc- tural arrangement in the granite bodies were controlled by the same orogenic forces that governed the folds within the Grenville forma- tions. The question arises as to whether the folding took place after the intrusion of the granite masses or in part or in whole con- 158 NEW YORK STATE MUSEUM temporaneous with the period of intrusion. The granite masses east of Dodd’s creek and Black creek respectively are severely mashed and crushed but the granite of the Hyde School and Hickory Lake bodies show practically no signs of active pressure subsequent to their consolidation from the magma. This might be thought to imply that the granite masses in the southwest were much older than those in the northeast, but there is no substantiating evidence for this. All the granite masses of this type (Alexandria type) are of similar lithologic character and bear similar structural relations to all the other types of country rock. It seems more probable to the writer that the granite masses were all intruded contemporaneous with the latter part of the period of orogenic pressure and folding and that the foliation of the granite is of primary magmatic origin. The period of orogenic pressure continued after the consolidation of the granite masses and locally produced great crush zones. The Dodd’s creek and Black creek granite masses are assumed to lie within such zones of yielding. The writer has interpreted these granite bodies as phacoliths (figure 14). The term “phacolith” was originally proposed by Harker (’09, p. 76-78) for concordant intrusions in which preexist- ing structures in the country rocks have exerted a directing influence in connection with the mountain building forces. In cross section Granite 1 1 r -J. LqT 1 ■ J_L _X77E T Crystalline Limes'tone Garnet* gneiss Granite cartel S$ Figure 14 Inferred structure section across California phacolith, Lake Bona- parte quadrangle (northeastward extension of Dority Pond granite mass, Antwerp quadrangle) HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 159 he states that it represents typically a meniscus or sometimes a doubly convex form, usually has a long diameter in the direction of the axis of folding, and the ideal phacolith is subject to many modifica- tions, in accordance with varying mechanical conditions of intrusion. A phacolith must therefore have both a roof and a base. There is no evidence of a base for any of the granite phacoliths on the Ham- mond quadrangle, but the writer believes that he has found evi- dence for a base to bodies of granite with similar structural relations on the Gouverneur and Canton quadrangles (Buddington, ’29 a) and has therefore interpreted those of the Hammond area as being of similar form. It seems probable that along the borders of the pha- coliths there may have been some fracturing of the limestone so that transgressive contacts of small amount occur locally, and the foliation of the granite is locally steeper than that of the general dip of the adjoining limestone. At the western border of the quadrangle, east of Black creek there is a mass consisting of granite bodies and gneiss which interlock together so closely as to form a composite unit. Contacts between the granite and gneiss are blurred and indefinite. It is possible that the granite magma was at the start intruded in general along the bedding as an incipient phacolith but due to the more brittle char- acter of the gneiss as contrasted with crystalline limestone, the beds broke during folding and the granite spread as a series of separate bodies throughout the brecciated mass. Crush zones. A belt of rock exposed for a length of 25 miles and a width of two and one-half to four miles along the valley of Black creek and Black lake, in which the foliation dips almost uniformly to the southeast, has previously been described. This belt coincides with a zone of intensive crushing. The Grenville gneisses, the Alexandria syenite mass on the Alexandria quadrangle ; the Black Creek granite mass, the South Hammond quartz diorite and the Pope Mills syenite sheet, on the Hammond quadrangle, and the Macomb granite (Alexandria type) mass on the Brier Hill quad- rangle are all intensively mashed and show pronounced cataclastic microstructure. Small sheets of granite (Hermon type) in this belt are also mashed. Orogenic forces must therefore have been active subsequent even to the latest period of magmatic intrusion and consolidation. Local narrow crush zones within the Rossie complex have previously been referred to. The Dodd’s creek granite mass is intensively mashed. The geology to the west is covered by a cap of Potsdam sandstone and the explanation of why this granite body shows cataclastic structure is not clear to the writer. l6o NEW YORK STATE MUSEUM A crush zone will be described later as constituting the peripheral border of the main igneous complex for a length of at least 30 miles and a width of three to ten miles and involving all the varieties of igneous rocks and included bands of Grenville that form this com- posite mass. Why such intensive crushing should have been localized along these two belts and what they mean with respect to the type of movement involved is not obvious. Clearer understanding must await further knowledge of the location and relations of similar crush zones elsewhere within the Adirondacks and the Precambrian north of the St Lawrence. The relatively weak Grenville beds must have been subjected to tremendous stresses under great resistant pressures from all sides to have permitted the crushing and myloni- tization of such strong rocks as granite and syenite over a wide belt. In view of these phenomena it is no wonder that the Grenville lime- stone has flowed like a molten igneous rock, has thinned and thick- ened locally to a very marked degree, and has dismembered inter- calated schist layers and small dikes and pegmatite veins into frag- ments and carried the latter far apart from each other. It was upon the basis of such evidence that Emmons (’42, p. 37-59) originally thought that the Grenville limestone must be of igneous origin. Other structural features. Throughout the Grenville beds peg- matite veins with a zigzag or crenulate structure are common. These appear to be in large part the result of crumpling and plication arising from distortion and flowage of the inclosing beds resulting from com- pression and from differential movements between different beds. The pegmatite veins in the limestones show two different types of structural phenomena. Locally throughout the limestone areas there are abundant pegmatite veinlets and dikes that have been crumpled and isoclinally folded along with the limestone, both on a large and a small scale. On the other hand, there are also locally many pegma- tite veins which have been broken up and pulled apart by the flowage of the limestone (figure 45). Such broken fragments are usually rectangular in shape. The breaking and disruption of the pegmatite veins into a group of angular fragments must have taken place after consolidation, but the folding of the pegmatite veins very probably took place before their complete crystallization. The character of the texture and structure of the crumpled peg- matite dikes that are locally so abundant in the limestone deserves a systematic careful study. Unfortunately material was not. collected for such a purpose. The writer studied thin sections from the axes of close crumples in three different pegmatite veinlets at widely HAMMXJND, ANTWERP AND LOWVILLE QUADRANGLES l6l separated localities. One, about a mile west of the south end of Yellow lake, was taken from near the center of the axis of a crumple in a dike two feet wide. It showed a texture due to normal crystal- lization. Another, from about two miles west of Scotch Settlement School, showed a cataclastic-protoclastic structure; and a third, from a locality near Black lake about one mile west of Stark School, showed a well-developed protoclastic structure with spindles of clear uncrushed quartz and augen of feldspar severely granulated on their borders. In the garnet-biotite gneiss the pegmatite veins often similarly show a plicated structure. The amphibolite layers intercalated in the gneiss are often broken apart and displaced a few inches to several feet by the flowage of the inclosing gneiss, or are crumpled with the gneiss. In the absence of mylonitization, these phenomena are believed to have been brought about while the great volume of injected pegmatitic magmatic solutions were still incompletely crystallized. The pegmatite veinlets in the quartz diorite belt of the Antwerp sheet often show complicated zigzag patterns. It is not certain whether these veins were plicated while both they and the inclosing rock were still partly liquid and thereby constitute a case of “ptyg- matic folding” or whether they have come in along intersecting fractures (Read, ’28), or whether some other explanation is the correct one. A crenelate type of fracture seems probable. The biotite-garnet gneiss around the Somerville anticline and Sherman Lake syncline shows a structure consistent with the degree of crushing and recrystallization represented by the associated granitic intrusives which exhibit a protoclastic structure. In the gneiss, the quartz occurs for the most part as discrete grains, but in part is in leaves consisting of an aggregate of granules. The garnet-gneiss of the Lewisburg syncline, however, shows evidences of a pronounced cataclastic structure, accompanied by some recrystal- lization and there are many fault surfaces within the dolomite. This is likewise consistent with its nearness to the highly crushed border zone of the main igneous complex. Joints. The joint systems within the rocks have not been studied with sufficient detail to warrant any positive conclusions as to their orientation or origin. A few observations were made in the Rossie intrusive mass south of Rossie, where joints with directions within 10 degrees of east-west, N. 450 to 6o° W. and N. 55° E. were noticed. Sixteen galena-calcite veins were noted at widely separated locali- NEW YORK STATE MUSEUM 162 ties within a central belt of the Hammond quadrangle in both intru- sive masses and limestone. Their predominant strike is N. 6o° to 70° W. These veins are believed to fill fault fractures of post-Potsdam age. STRUCTURE OF THE MAIN IGNEOUS COMPLEX Introduction. The main igneous complex within this region has previously been referred to as comprising two units — the Diana and Croghan masses respectively. The writer tentatively suggests that the banding of the Diana complex may have arisen in a sill-like intru- sion as a result of gravity stratification during the period of emplace- ment or consolidation, accompanied by some intrusion of residual or later magmas into the earlier facies. The foliation of both the Crog- han and Diana complexes is fundamentally of primary origin formed during consolidation. The rocks of the Diana complex, however, were subjected to orogenic stresses during the later stages of solidi- fication, yielding a protoclastic structure ; and upon this, subsequent to complete consolidation, was superimposed a cataclastic mashing in a belt three to ten miles wide and at least 30 miles in length, extending across the Oswegatchie, Lake Bonaparte and Antwerp quadrangles. Within the cataclastic belt is a narrow band of ultramylonite. Diana complex, Lowville anticline (?). A major structure of the Diana complex is found on the Lowville, Carthage and the southeast corner of the Antwerp quadrangles, and is here referred to as the Lowville anticline( ?). A trend line of the structure may be traced by the belt of fragments of pyroxene-biotite gneiss that has resulted from the metamorphism and dismemberment of a bed of the Grenville series by the intrusives. This belt may be followed around an arc from New Bremen by way of Croghan and Mount Tom to the Black river. The foliation of the intrusives is conform- able to the strike and dip of the inclusions. The axis of the structure, as indicated by the lineal structure or pitch, strikes N. 60 to 65° E., and the distance across the structure at right angles to the strike is more than 15 miles. Unfortunately the western part of the anti- cline (?) is buried beneath the cover of Paleozoic beds. The Pre- cambrian rocks of the western half of the Lowville quadrangle, the southeastern corner of the Antwerp sheet and the northeast part of the Carthage area are all embraced in this structure. It is here interpreted as a great composite phacolith or folded sheet on the axis of an anticlinal fold strongly overturned toward the southeast. The north limb dips about 30 to 40 degrees north-northwest, and HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 163 the fold pitches about 40 ° to 45 0 along a north line and 25 0 to 30 0 along an east line. The vertical attitude of the foliation near New Bremen is appropriate and necessary for such a fold. In the northern half of this great arc, north of the Beaver river, south of Natural Bridge, and west of Croghan, Monnatt School and Aldrich School (Lake Bonaparte quadrangle) the rocks all show a most pronounced pitch, which often is so prominent that it is difficult to distinguish the primary banding or foliation. The pitch is indi- cated by the direction of maximum elongation of the individual minerals or mineral groups in the plane of foliation, and by striae, groovings and flutings along the walls of quartz veins and dikes. It is particularly marked in the rocks with cataclastic structure and becomes indistinct in the belt of rocks showing protoclastic structure, going toward the southeast. It affects all the many varieties of rocks in the area. Except for one small local band, it has a uniform strike of N. 6o° to 65° E. In many cases where aplitic or granitic dikes cross the foliation of the country rock their own foliation is parallel to the direction of pitch, and is quite distinct. The structure that we call pitch must have been formed subsequent to the complete consoli- dation of at least most of the rock for it is later than the quartz veins whose groovings and flutings have been molded by cataclastic crushing as shown by the microstructure and it is later than the youngest dikes since the pitch structure crosses them no matter what their orientation. The direction of the pitch is not influenced by the strike of the primary foliation. In the northeast corner of the Carthage quadrangle and the southeast corner of the Antwerp sheet (figure 9) the north limb of the Lowville anticline shows a marked pinch with a north-south axis. The direction of pitch on the east side of this pinch structure is N. 8o° to 90° E., which seems to indicate that this limb has been rotated 20 to 25 degrees subsequent to the formation of the lineal structure. Locally minor folds may be observed in the intrusives. One excellent example was observed in a rock cut on the main road one and one-half miles south of Fargo. The fold here is outlined by remnants of a disintegrated anticline of dark Grenville gneiss, but the banding of the foliation of the igneous rock is parallel to the folded Grenville. The rocks between Natural Bridge and Karter appear to be best interpreted as belonging to the southeast limb of a great overturned fold whose axis runs northeast toward Lake Bonaparte. 164 NEW YORK STATE MUSEUM Croghan complex. The complex of the Croghan mass exposed on the Lowville quadrangle appears to be but a small part of a much larger unit that extends at least across the Port Leyden quadrangle to the south, the Number Four sheet to the east, and the southern half of the Oswegatchie quadrangle to the northeast. The rocks of this unit are all equigranular and, although having a foliation, show little or no evidence of crushing except near the borders. It is difficult or impossible to work out accurately the structure because of the large areas overlain by Pleistocene deposits. A most interesting feature, however, is the alternation in the direction of the dip of the foliation from southeast to northwest and then back to southeast as one passes in a direction at right angles across the strike. The change seems to take place in the synclines (?) in part, not through the dips becoming flatter and changing to the opposite direction through the horizontal, but through their becoming steeper and passing to the opposite direction of dip through the vertical (figure 15) as in the Grenville beds on the Hammond sheet. In the southwest corner of the number 4 quadrangle and the southeast corner of the Lowville area the strike of the foliation swings around a semicircle as though around the nose of an anticline whose axis at the northeast plunges 30 degrees northeast and whose limbs dip 30 to 40 degrees northwest and southeast respectively. The southwest nose of this anticlinal structure may be seen near Glenfield on the Port Leyden quadrangle. A similar anticlinal and synclinal structure for the dip of the foliation is described by Miller (To, p. 36-37) in the intrusives of the Port Leyden quadrangle. The dips of the Bel- fort and Crystal Dale-Bush’s Corners bands of amphibolite or metagabbro are toward each other and seem to indicate a synclinal structure, and east of Petries Corners the nose of an anticline appears to be indicated by the opposed dips and by the manner of junction of the amphibolite or metagabbro bands. The manner of intrusion of this complex can not be adequately discussed until further data from adjoining quadrangles are available for the unit as a whole. Belts of cataclastic, protoclastic and aclastic structure. As a result of a detailed study of the characters of the internal structure and texture exhibited by the syenite-granite rocks of the Lake Bona- parte quadrangle it was found that the rocks could be grouped into three distinct belts : ( 1 ) a belt at least ten miles wide on the south- east in which there is no sign of crushing (aclastic) and in which the texture is essentially typical of that resulting from the normal crystallization of a magma, except that the minerals are more or HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 1 65 166 NEW YORK STATE MUSEUM less segregated in individualized streaks and may have a dimen- sional orientation (figures 46 and 49) ; (2) a central belt three to three and one-half miles wide in which the rocks exhibit a banding similar to that of the first belt but in addition exhibit internally a protoclastic texture wherein the feldspars are partially crushed and in polygonal grained aggregates whereas the quartz is in massive leaves, (figure 50) ; (3) a belt of crushed rock five to ten miles wide on the northwest in which the rocks exhibit a similar character of banding to that of the central belt but in addition possess a cataclastic structure wherein the feldspars are partially pulverized and the quartz is coarsely crushed or granulated (figures 47, 48, 51 and 52). An exactly similar state of affairs is found in the main igneous complex of the Lowville, Antwerp and Carthage quadrangles, and the borders of these separate belts are indicated on the map. The cataclastic and protoclastic gneisses are embraced within the Diana complex and the aclastic gneiss belongs to the Croghan complex. The border line between the cataclastic and protoclastic gneiss is independent both of the strike of the folia- tion and the trend of the belts of different gneiss ; for, as can be seen northeast of Naumburg, it cuts at a large angle directly across both the foliation and banding of the intrusives. The border line between the protoclastic gneiss on the one hand and rock with texture of interlocking crystallization on the other conforms with one possible exception, to both the strike of the foliation and the trend of the bands, and marks the line of contact between the porphyritic gneisses of the Diana mass, and the equigranular gneisses of the Croghan mass. Near Croghan this line appears to cut directly across a belt of granite. The pyroxene-biotite gneiss of the Carthage and western border of tbe Lowville sheet shows a cataclastic structure. In the belt of igneous rock showing protoclastic structure the Grenville gneiss consists of a granular mosaic which increases in size of grain towards the east. The hornblende grains form a skeleton mesh- work inclosing the granules of feldspar. In the belt of igneous rocks with foliation of primary crystallization the Grenville gneiss consists of a granular mosaic with the coarsest grain of any part of the belt of inclusions. Ultramylonite band. Within the peripheral crush zone of the main igneous complex on the Antwerp and Carthage quadrangles there is a narrow band within which the rocks are so intensively mashed as to constitute a readily distinguishable ultramylonite zone. HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES l6 J This band parallels the road on the east side from Mount McQuillen to about a mile south of Fargo, where it turns east and passes near North Croghan. Southwest of Mount McQuillen it runs south- west to the Black river. As thus outlined it is about eight miles long and one-quarter to one-half mile wide. The rock involved is predominantly granosyenite with included shreds of Grenville gneiss. It has been so crushed that for the most part the texture is as dense as that of a felsite, with only a very few small residual eyes of feldspar left. A peculiar feature is that much of the rock contains abundant disseminated titanite and deep green pyroxene grains that are much larger than the average feldspar and quartz particles. Titanite and such deep green pyroxene are usually asso- ciated with the effects of assimilation and disintegration of Gren- ville calcareous beds in this region and this is very probably the case here. The rock is not only crushed but is usually intensively fractured so that fragments larger than a few inches in diameter are rare. This is well shown on Mount McQuillen. Another phe- nomenon characterizing the band is the common occurrence of drag folds. A major one is shown just south of Fargo, where the boun- dary between the granosyenite and anorthosite outlines a sigmoid fold. The strike of the foliation is north-northeast and the dip about vertical. There are many subsidiary minor drag folds here, which are brought out wherever hyperite sheets or included bands of Grenville are found (figure 16). Similar structures may be seen on Mount McQuillen. The foliation usually has a dip of 40 to 60 degrees. The drag folds have a steep north-northeast pitch and the move- ments that produced the mylonitization, fracturing and drag fold- ing must have had a greater component in the horizontal direction than upwards. The northwestern block appears to have moved northeasterly relative to that on the southeast. Factors in origin of foliation and microstructure. Many fac- tors undoubtedly enter into the mechanism of the process whereby the rocks acquired their foliation and just as surely these same factors must play roles of unequal importance in giving the gneiss- oid structure to the rocks in the different belts. The following are suggested as the more important agencies. A segregation of the different minerals into individualized streaks rich respectively in feldspathic or in ferromagnesian minerals, and crystallization into individual leaves as in the case of quartz, are very important factors in the foliation of all the gneisses. Such S c a. t e NEW YORK STATE MUSEUM 1 68 Figure 16 Sketch of drag folded hyperite sill in granosyenite ; from mylonite band, one and one-fifth miles southeast of Fargo, Antwerp quadrangle HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 169 segregation appears in part to be genetically connected with dif- ferential flowage movements in the magma during intrusion and consolidation. Another factor is the dimensional orientation of previously crystallized minerals, as a result of flowage in the magma. This is of minor importance in the equigranular rocks of the southeast or aclastic belt, but is of relatively greater significance in the por- phyritic rocks of the Diana mass, where the phenocrysts, often of themselves, give the rock a gneissoid appearance, having been given a rough parallel orientation by the stream lines or currents within the magma and by the force of gravity. A third factor is the crushing and granulation of crystallized mineral grains, the particles being rolled out along the foliation planes. As far as present appearances show, this has not been a factor in the foliation of most of the rocks of the Croghan complex for little or no signs of crushing are visible there. In the rocks of the belts showing protoclastic and cataclastic structure, however, this factor has in varying degrees accentuated a primary foliation due to other causes. There is often a marked variation in the degree of crushing from band to band and the changes may be quite abrupt. A fourth factor is a foliation inherited from the injection, replace- ment, shredding and disintegration of blocks and beds of the Gren- ville series whose foliation planes have controlled the movements of the magmatic solutions. This is the most important factor in many local bands but a minor factor in much of the rock. A fifth element comprises pegmatitic veinings and pegmatitic crystallization parallel to the foliation due to movements of residual liquid during consolidation and to local variation in volatile content. Another very probable factor is the crystallization with dimen- sional orientation of some of the minerals as a result of stress acting on the viscous crystallizing mass. It is thought that the magma in many cases may have been sufficiently viscous to have carried strains of small amount, certainly in the later stages of consolidation. Recrystallization subsequent to the complete cessation of /the magmatic consolidation is considered of limited importance in pro- ducing the foliation of these rocks, although it is a major factor in producing the massive quartz leaves at a very late stage in the formation of the protoclastic gneisses, and in controlling their microstructure. Part of the crushing of the minerals of the rocks with proto- clastic structure may have taken place before complete consolidation and after the magma had so far crystallized that the crystallized 170 NEW YORK STATE MUSEUM particles constituted an interlocking meshwork with interstitial liquid. These rocks, however, must also have been subjected to strong deforming stresses after their essentially complete consolidation, for the pegmatite veins cutting them likewise show a protoclastic struc- ture, the intruded hyperite sheets and inclosed Grenville gneisses are granulated, and stresses of great magnitude are thought to have occurred long after the emplacement of the syenite and granosyenite members that show the protoclastic structure. Since the cataclastic crushing affects the quartz veins as well as the igneous country rock, and is associated with the development of secondary chlorite and carbonate, it must have taken place at rela- tively low temperatures. It seems probable that the rocks showing cataclastic structure were subjected to greater strains at lower tem- peratures than those with protoclastic structure. In crystalline schists it is often found that, subsequent to the period of intense deformation, there is a relaxation of the stress and metacrysts grow under conditions of uniform pressure while the temperature is still high. Doubtless such a baking, under lessened stress conditions and at moderate temperatures, also succeeded the crushing of the rocks with protoclastic structure, and of the rocks with cataclastic structure, but of the latter at somewhat lower temperatures. It may be of interest to compare these conditions and phenomena with the results of experimental investigations on the effects of strain and temperature on grain growth in metals (Jeffries and Archer, ’24, p. 85-145). It is found that there is a critical range of strain that is favorable to subsequent recrystallization and grain growth on baking at a moderately high temperature. Metals sub- jected to strain in excess of this critical range, and reduced thereby to a fine grain within the critical range of strain, and subsequently given the same heat treatment, yields a recrystallized coarse grained aggregate. In general the higher the temperature, the more favor- able the conditions for recrystallization. By analogy, then, the cataclastic rocks may be interpreted as hav- ing been subjected to such excessive strain that grain growth did not take place after the stresses relaxed, particularly as the temperature may have been relatively low. The rocks with protoclastic struc- ture, on the other hand, are, as far as stresses after essentially com- plete consolidation are concerned, interpreted as having been sub- jected to smaller strain of a magnitude probably within the critical range, and therefore to have recrystallized with grain growth due to the high to moderately high temperature during and subsequent HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 171 to their deformation. It seems probable that a little vapor and dilute siliceous solution may also have been still permeating the rock and accelerated the grain growth and recrystallization. The polygonal granular aggregate developed in the feldspars, as con- trasted with the massive quartz leaves, is attributed to the differ- ences in reaction of these different minerals to the factors involved. Evidence that the forces tending to induce the foliation came in part from the drag of the magma against its walls and also in part from outside regional pressures is well exemplified in the rocks of the Lowville anticline (?). In the extreme northeast corner of the Carthage quadrangle there are shreds of Grenville schist with a northeast orientation included in granosyenite with a northwest folia- tion and all are crossed by the lineal structure with an east-northeast strike. The lineal structure crosses dikes, quartz veins, and foliation alike and is certainly in part later than the complete consolidation of the intrusives as indicated by the crushing and grooving of the quartz veins. Another fact indicating that there were strong regional stresses is that the dikes and country rock exhibit a similar degree of crush- ing. In the belt of cataclastic structure the hyperite dikes are severely mashed, whereas in the belt showing protoclastic structure they are crushed but on a much coarser scale, and in the belt with interlocking texture of crystallization they are more or less massive. Similar phenomena are true of the included fragments of Grenville pyroxene-biotite gneiss on the Lowville quadrangle and of the quartz veins. The quartz of the veins in the rocks of the belt showing cataclastic structure is much mashed whereas that of the veins in the rocks in the belt showing normal interlocking crystallization is massive. Joints. Of about ioo observations made on the directions of the joint planes, 33 per cent lie between N. 750 E. and N. 750 W., averaging about E.-W., 25 per cent between N. 40° E. and N. 6o° E., averaging about N. 50° E. ; 25 per cent between N. 30° W. and N. 6o° W. averaging about N.-W., and 13 per cent between N.-S. and N. 25 0 W., with three directions lying outside these ranges. These observations would indicate the presence of two distinct systems, each comprising two sets of joints approximately at right angles, one system of joints having a northwest and northeast strike respectively, and the other an east-west and slightly west of north-south strike respectively. Only two joint planes were observed with strikes between north-south and N. 35 0 E. 172 NEW YORK STATE MUSEUM STRUCTURE OF PALEOZOIC BEDS Antwerp and Hammond quadrangles. The Paleozoic beds of the Hammond and Antwerp quadrangles are in general flat lying or with gentle dips. The basement on which the Potsdam sand- stone rests slopes on the average about 14 feet to a mile to the northwest over the Hammond and the northwest half of the Antwerp quadrangle. In the southeastern half of the Antwerp quadrangle and on the Lake Bonaparte quadrangle the slope is steeper, averag- ing more than 20 feet to a mile, in accord with the more rapid upslope of the core of the Adirondack massif. The low gently rolling hilly nature of the Precambrian surface upon which the sandstone remnants rest has been previously described. The granite inkers about one and one-half miles north- west of Hammond and one mile northwest of Cedars represent the tops of hills projecting through the sandstone. In the valley of Chippewa creek, the granite is exposed as a result of erosion of the Paleozoic beds. The dip of the sandstone is a little more northerly than the general dip of the surface of the Precambrian rocks and is probably the result of the gentle post-Paleozoic upwarp that constitutes the Frontenac axis. The sandstone northwest of Black lake and Black creek has a general dip to the north-northwest or northwest, but with broad very gentle warpings. In adjoining areas to the north, west and southwest, Cushing (T6, p. 52-53) describes the beds as very gently folded so as to produce a series of low domes and basins. South- east of Black creek and Black lake, on the Hammond quadrangle, many of the sandstone outliers have a synclinal structure and the large mass such as that near Gross lake shows gentle warpings and folds. Locally, dips as steep as 45 degrees have been observed and folds with limbs dipping 30 degrees are not uncommon. The beds are in steeper folds where there was a considerable difference in the relief of the surface upon which they were deposited, as where limestone and granite are associated. The sandstone caps on the broad granite masses are flat lying or with very gentle folds. The remnants of the former continuous sandstone formation, are now largely preserved in the limestone valleys. The synclinal structure and steeper dips of the beds must thus be in part primarily due to original deposition on steep hill slopes and in valleys. In part, how- ever, the steeper dips of the folds in the sandstone southeast of Black Creek valley must be really due to more active yielding to forces tending to produce folding. The alternation of resistant HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES I73 masses of igneous rock and relatively yielding limestone is favor- able to differential movement under stress. Tug Hill plateau. The Paleozoic beds of the Tug Hill plateau dip gently to the west and southwest instead of to the northwest. At Calcium on the Theresa quadrangle the altitude of the contact between the Pamelia and Lowville limestones appears to be at about 480 feet, and at the canal east of Deferiet the contact is at 660 feet, giving an average slope to the west of about 22 feet a mile for eight miles. On the Theresa quadrangle the slope is only about 16 feet to a mile to the west and to the east of Deferiet the beds rise more steeply at about 25 to 30 feet to a mile. The slope thus steepens towards the east and the core of the Adirondacks. Joints. No special study was made of the joint systems. The few observations made are in accord with the conclusions drawn by Cushing (To, p. 107 and 108) for the similar beds to the west. There are two main systems of joints one varying between N. 6o° to 70° E. and the other between N. 40° to 70° W. A minor set strike about N. 20° to 40° E. and N. 30° W. It would seem prob- able that the east-northeast joints were controlled in their develop- ment by the strike of the axes of the minor folds and warpings which in turn were controlled by the trend of the structure of the Precambrian basement complex. PRE-POTSDAM WEATHERING OF THE PRECAM- BRIAN ROCKS One of the outstanding features of the contact zone between the Potsdam formation and the Precambrian rocks in the Ham- mond and Antwerp quadrangles is the deeply weathered condition of many of the rocks, except the limestone, immediately beneath the caps of Potsdam sandstone. Similar phenomena have been described by Smyth (’01, p. 98-101) on Wells island in the St Lawrence river and by Chadwick (’20, p. 17-20) from the Canton quadrangle. Cushing (To, p. 160), on the other hand, in his description of the Thousand Islands region writes : “Except for the local accumulations of a very scanty amount of residual material in small pockets in the depressions (and this almost exclusively quartzose) the Precambric surface, as it passes under the Paleo- zoics, is remarkably free from surface decay, even the weak rocks being astoundingly fresh.” The writer at no place has observed a residual soil at the contact between the Potsdam and the Grenville limestones, and this fact 174 NEW YORK STATE MUSEUM doubtless bulked large in leading Cushing to make the statement he did. The limestone in pre-Potsdam time, however, had suffered marked solution as indicated by the deep vertical fissures filled with Potsdam sandstone that are common in the Somerville-Gouverneur limestone belt, and which were described by Cushing. The granites, gneisses, quartzites and schists, however, where still covered by Potsdam sandstone all usually show a much more deeply weathered character than the same rocks where not so preserved. The hematite ores of the Keene-Antwerp belt and of Mill-Site Lake, the flat Potsdam rests on the eroded edges of a much decomposed ferruginous schist”; “on Wells Island about one mile east of Thousand Island Park, the Potsdam rests on decom- posed granite-gneiss and schist and the lower 5 feet of the former is sandstone.” Chadwick writes of the Canton quadrangle as fol- lows : “There are even places where through some 10 to 20 feet of beds a complete gradation exists between ordinary fresh Grenville rocks beneath and the normal stratified red Potsdam sandstone above through a Grenville regolith constituting a basal Potsdam breccia.” In the course of his work on the Hammond and Antwerp quad- rangles the writer has observed many instances of preservation of the results of pre-Potsdam weathering. Many of the pits from which material is obtained to fix the country roads are in such weathered zones beneath caps of Potsdam sandstone. On the west side of Lake of the Woods about one-half of a mile from the foot a contact is well exposed in the walls of the lake. Thin-layered plicated quartzites of the Precambrian are overlain by Potsdam sandstone. The quartzites are thoroughly oxidized, reddened, kaolinized and porous for a depth of 15 feet. A two-foot bed of reddish sandstone lies directly on the eroded edges of the rotted rock and is succeeded by a thin layer of conglomerate. Another splendid example is shown just west of Rogers School (Hammond quad- rangle) where a big cut for road material has been made in rotted rock just beneath a capping of Potsdam sandstone. The Grenville rock is predominantly quartzite, stained reddish, and with its small amount of kaolinized feldspar. A few pegmatite veins are present and the feldspar of these is completely altered to kaolin. Sparse gneiss layers are similarly oxidized and weathered. At a number of localities granite so rotted and weathered that it could be dug out with a pick for road metal has been observed beneath a veneer of Potsdam sandstone or at a locality where the sandstone over- burden had been but recently eroded. HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 175 At a number of localities in the quadrangle there are ferruginous or ferruginous jasperoid replacements in local pockets in the lime- stone beneath the base of the sandstone. These are interpreted as iron and silica rich accumulations formed at or near the surface of depressions in the limestone in pre-Potsdam time that failed to be removed before the deposition of the Potsdam sands and were thus preserved. The hematite ores of the? Keene-Antwerp belt and of many local spots elsewhere in the quadrangle occur as replacements of the Precambrian granite or Grenville gneisses, and are interpreted (Smyth, ’94) as having been formed at or near the surface in pre-Potsdam time by iron-rich solutions derived from the weather- ing and oxidation of pyritic zones in the Grenville gneisses. Their preservation is therefore attributed to the failure of erosion to remove them before they were buried by the Potsdam sands and thus preserved to be re-exposed by post-Paleozoic' erosion. These ore bodies and ferruginous replacements are thus confirmative of a deep and thorough weathering in pre-Potsdam time, the results of which were not wholly removed before the deposition of the Potsdam formation. The pre-Potsdam age of these weathered rocks and ferruginous replacement deposits is attested by the fact that the sandstones are composed only of the minerals most resistant to weathering (quartz and accessory tourmaline, zircon, apatite) and the conglomerates only of the rock fragments most resistant to weathering (quartzite, quartz and sandstone) ; and that where the basal beds overlie the iron rich deposits, a part of the hematite has been eroded and incorporated in some of the overlying sandstone beds. Smyth (’01, p. 99) in describing the relations of a bed of Potsdam conglomerate to the underlying rocks states : “Not a pebble was found of granite gneiss or schist though these rocks form a large element of the Precambrian formations and several of the conglomerate outcrops rest on such rocks. From this character of the Potsdam beds the conclusion has been drawn that the Precambrian formations had been subject to thorough weathering, that the feldspars, ferromag- nesian minerals etc. had been decomposed and that in general the only coarse fragments available for building conglomerates or breccias were purely quartzose.” Relics of rotted rock formed as the deeper parts of a weathered zone during pre-Potsdam time and preserved beneath a cap of Pots- dam sandstone have been observed by the writer on the Potsdam, Canton, Gouverneur, Hammond, Alexandria Bay and Grindstone quadrangles. Although there is thus convincing evidence of deep 176 NEW YORK STATE MUSEUM thorough chemical weathering preceding the deposition of the Pots- dam sandstone, it is equally apparent that most of the residual soil and in many cases the deeper lying rotted rock had been swept off by erosion before the Potsdam sands were laid down, for they rest either on fresh rock, as in the case of the limestones and locally on the higher elevations of the granite or gneiss, or they lie on rock that is much weathered, leached and rotted but still coherent. It is interesting to note that similar topography and residual soils are found at the base of the Potsdam sandstone in Wisconsin but that only relics are found in Ontario. The following statements are taken from a description of the contact zone in Wisconsin by S. Weidmann (’03, p. 289) : From the foregoing it has been concluded that the pre-Cambrian land was a worn-down country, a peneplain of erosion, before the Potsdam was deposited upon it. . . . Lying at the contact of the gently sloping pre-Cambrian and the Potsdam, apparently every- where except about the pre-Cambrian monadnocks, is a widespread formation of partly decomposed crystalline rock and clay. . . . The clay varies considerably in thickness, but generally has a depth of 10 to 20 feet, though in places it is known to reach the unusual thickness of 40 feet. . . . It is believed by the writer that the wide-spread occurrence of the thick deposit of clay as a subjacent formation of the Potsdam obviously points to its origin and its presence there before the sandstone was deposited upon it. J. F. Wright (’23, p. 10) has described the contact zone between the Potsdam sandstone and the Precambrian in the Brockville- Mallorytown area, Ontario. He writes : Also, from a study of the contact between the Pre-Cambrian and the late Cambrian or Potsdam sandstone, and from the character of the Potsdam sediments, it is evident that the country was low and flat in Potsdam time and that a long period of erosion preceded the advance of the Upper Cambrian seas over the area. The Pre-Cam- brian rock beneath this contact is fresh and unweathered, just as it is in the recently glaciated surfaces beneath the boulder drift. The only evidence Wright (p. 33) found of pre-Potsdam weath- ering was in a few diabase dikes. He writes : The surface forms a yellowish rusty colored gossan which is so soft and decomposed that it can be easily removed with a pick and shovel. On a few dikes this gossan is at least six feet thick but in most places it is less than two feet. . . . The decomposed capping is preglacial or older for dykes that were exposed to intense glacial scour do not have such a capping. . . . This deep weathering may possibly be pre-early Paleozoic. HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 1 77 Burling and Kindle (’15. p. 7) have described the contact between the Paleozoic and the Precambrian in Ontario. They write: The fresh character of the Pre-Cambrian surface and the absence of residuary clays at its contact with the overlying Paleozoic are in marked contrast to the conditions reported in Wisconsin where deposits of clay io to 20 feet thick and consisting of decomposed Pre-Cambrian rocks separate the Cambrian from the Pre-Cambrian. On the Lowville quadrangle and the southwest corner of the Antwerp sheet the writer has nowhere seen a contact between the Pamelia limestone and the Precambrian so that the exact nature of the surface of the underlying rock is unknown. PALEOZOIC FORMATIONS (Hammond and Antwerp sheets) INTRODUCTION Only two of the Paleozoic formations of the St Lawrence Low- lands outcrop on the Hammond quadrangle, and on the Antwerp sheet except in the extreme southwest corner, the Potsdam sand- stones and the Theresa sandy dolomite. The Theresa formation is restricted to the northwest corner of the Hammond quadrangle and to the southern half of the Antwerp sheet, where it is underlain by the Potsdam sandstone. The Potsdam sandstone, however, occurs as innumerable patches or large areas throughout both quadrangles, representing remnants of a once continuous sheet that formerly covered the area, and which have not yet been entirely removed by erosion. There are innumerable patches of sandstone that are not shown on the geologic map due to their small size. In the south- west corner of the Antwerp sheet the Pamelia, Lowville, Leray and Watertown limestones are exposed in the border of the Tug Hill plateau. The Paleozoic formations have not been studied in detail by the writer and only a general discussion will be given. POTSDAM SANDSTONE Practically all of the Potsdam sandstone is a white, coarse, even- grained sandstone thoroughly cemented, usually by quartz but in certain beds by calcite. The upper beds in particular are calcareous where they pass upwards into the Theresa formation and the lower beds are often calcareous where they rest on the Grenville crystalline limestone. Locally, although rarely, there are red beds at the base and in the section along the west side of Black lake and Black 178 NEW YORK STATE MUSEUM Creek swamp (Hammond sheet) there is intercalated in the white sandstones a bed alternately striped with thin bands of red and white. Occasionally the sandstone layers have beautifully rippled- marked surfaces. Crossbedding is not prominently developed although present. Locally there are beds of coarse conglomerate, in part intercalated as lenses in the lower beds of the formation, in part occurring as a basal conglomerate. The pebbles of the con- glomerate are predominantly quartz and quartzite. Red sandstone is abundant in some lenses. Fragments of granite or gneiss were not observed. Locally at the base of the Potsdam sandstones there are deep- colored red beds, with conglomerate lenses. Near the north end of Lake of the Woods on the west side there are 15 feet of red sand- stone with conglomerate layers near the base overlying unconform- ably deeply weathered and oxidized Precambrian thin-layered quartzite. The fragments in the conglomerate are largely quartzite. Similar red sandstone is found at the base of the sandstone beds one and three-quarters miles south of Pope Mills, and a mile southwest of Farley School. The Potsdam formation is very well exposed in the cliffs along the west side of Chippewa Creek valley. The base, however, is usually covered. There appears to be a maximum thickness of 50 feet exposed here and locally there is probably not more than 30 feet of Potsdam sandstone between the underlying granite and the over- lying Theresa. In addition to the normal even-grained white sand- stone there are here several beds of peculiar character. One such bed near North Hammond is about two feet thick and has a vertical tubular structure throughout. The tubes are such as might result from annelid borings subsequently filled with sand and the struc- ture preserved. Beneath the base of the Theresa siliceous dolomite there are about five or six feet of thin-layered sandstone, in part calcareous, underlain by two feet of sandstone with pronounced intraformational slumping structure. The bedding is disturbed, crumpled and brecciated, but the movement must have taken place before consolidation and is doubtless to be explained as a result of subaqueous slumping. This bed has been traced for two miles to the southwest from North Hammond on to the Alexandria quad- rangle. The sandstone bed with the tubular structure evidently also has a wide expanse for it is found again two miles away on the south side of Chippewa Creek valley at the western side of the quadrangle. HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 179 The Potsdam is again well exposed in the cliffs along the west side of Black lake and Black Creek swamp. A number of quarries were opened in the sandstone here but none are working now. The sandstone in this section is from 60 to 80 feet thick. The upper ten feet of the formation is calcareous and characteristically weathers with a pock-marked surface or with a brown circle-spotted surface. A bed of red sandstone with stripes of white is present in the sec- tion and is well exposed as a three-foot bed in the quarry about three-quarters of a mile southwest of Oakvale. When examined with the microscope the predominant white sandstone is found to be a very clean quartz rock consisting of very well-rounded quartz grains cemented by a secondary growth of quartz. Occasionally a rounded tourmaline or zircon crystal is present and some bands are calcareous. In the reddish sandstone from the quarry about one mile southwest of South Hammond the quartz grains are coated with a thin film of hematite. The hema- tite is a primary deposit for the secondary silica surrounds the hema- tite coated grains. In the basal part of the Potsdam other types of sandstone are found locally, particularly where they rest on ferruginous deposits in the underlying Grenville. Many of these beds are ferruginous, usually red from hematite, but occasionally green from a chloritic or iron rich mineral. Thin sections of these rocks are all particu- larly characterized by the presence of a number of well-rounded grains of tourmaline (commonly green), zircon and apatite, named in the order of abundance. One titanite grain was noted. Where the beds overlie ferruginous replacements of limestone they are usually red from hematite and the sandstone may be calcareous. In the calcareous rocks the hematite may be associated with the calcite. In the red silica-cemented sandstones the hematite coats the rounded quartz grains and the silica cement is later. At the old ore pit about one mile southeast of Brasie Corners there is a bed of green sand- stone consisting of well-rounded quartz in a groundmass composed of a microcrystalline chloritic (?) mineral. On the Antwerp quadrangle the Potsdam and Theresa forma- tions are thinner than on the Hammond quadrangle and are wholly absent in the southern part where the Pamelia limestone rests directly on the Precambrian igneous rocks. South of the Indian river the Potsdam sandstone apparently varies in thickness from 15 to 25 feet in the larger areas of exposure, and usually consists of inter- bedded white sandstone in layers six inches to two or three feet ISO NEW YORK STATE MUSEUM thick and gray weathering calcareous sandstone in beds of similar thickness. The upper beds may carry dolomite crystals like the Theresa, and black organic markings are also found in these upper beds. A good exposure of the top bed of the Potsdam sandstone may be seen in the bottom of the quarry opening a mile south of Philadelphia. About eight feet of white sandstone, only in part calcareous, is exposed in the quarry on the bank of Black creek, one and seven-eighths miles south-southeast of North Wilna, and ten feet of the Potsdam sandstone are shown in the quarry three and one-quarter miles northeast of Fargo on the Natural Bridge road. The white sandstones often weather rusty with little spherical holes that give a characteristic pitted surface. Crossbedding on a small scale is common. Basal quartzite breccias are found in the valley of Deerlick creek. Cushing studied the Potsdam formation throughout the north- west Adirondacks and gives a brief summary (Cushing, ’25, p. 49- 50) of its relations, which is quoted here for its bearing on the Potsdam of the Hammond and Antwerp areas : From Potsdam west to Clayton the Potsdam formation is rather thin, and in places where the Precambrian floor was highest, on the granites and garnet-gneisses for example, it may entirely fail, and the Theresa formation lie on the Precambrian. East of Pots- dam the formation thickens rapidly by the successive addition of beds at the base. In other words, the thin formation present from Potsdam to Clayton represents only the summit of the forma- tion as it occurs to the eastward. This upper portion consists largely of somewhat calcareous, non extra-resistant sandstone, brown or white in color, carries locally marine fossils sparingly and seems an unquestionably marine formation. The sandstone found in residual patches on the Gouverneur quad- rangle differs from the above in being very vitreous and hard, and noncalcareous except locally where in immediate contact with limestone. Angular blocks of hard red sandstone are also found as pebbles in the basal conglomerates. The phenomena are quite as they are in the Thousand Islands region and suggest that this lower hard sandstone is materially older than the ordinary Potsdam of the region, although it is thought not to be older than the basal portion of the Potsdam sandstone of Clinton County. It also seems to be a nonmarine accumulation. . . . The writer is of the opinion that the older sand here is the approximate equivalent of the basal part of the formation as it appears in Clinton County, and that the other in like manner is equivalent to the summit, and that there is nothing here equivalent to the middle division of the formation in Clinton County. It is quite possible, however, that there is a break between the lower and middle divisions of the formation there, and the whole formation is very thick, at least from 1290 to 1500 feet. There is at hand today very little direct evidence in substan- HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES l8l tiation of the above opinion. The general Potsdam of this north- western area is very thin and is the unquestionable equivalent of the extreme summit of the formation in Clinton county. The pre- cise age of the basal sands of these hollows is an entirely open question. THERESA BEDS The Theresa beds give rise to a low but steep scarp or very steep slope at the edges of their area of outcrop, which is a marked topographic feature, on the level plain formed by the underlying Potsdam. Bluffs with rock exposures are occasionally found but usually the slope is covered. The boundary of the Theresa beds has therefore been mapped mostly on the basis of the topography. The Theresa beds are only occasionally exposed on the surface of the plain which they form. As exposed near North Hammond the Theresa beds comprise blue-gray, thin-layered, fine-grained sandy dolomite with sparse intercalated thin layers of sugary white sand- stone of a character similar to the Potsdam. Many of the bedding surfaces are covered with organic (fucoid ?) markings. A peculiar feature of the fresh surface of much of the rock is the flashing cleavage surfaces of calcite crystals that have grown in place and inclosed the quartz grains. They range from one-quarter of an inch to an inch in diameter. The beds are reported to be dolomitic. The rock is blue-gray on the fresh surface but weathers to a porous crumbly rusty brown “rotten rock.’’ The maximum thickness of the beds exposed on the Hammond quadrangle seems to be about 50 feet. A typical specimen of the Theresa dolomitic sandstone of the Hammond sheet where examined with the microscope is found to consist of quartz, dolomite and feldspar. The quartz is for the most part in fine subangular grains but well-rounded larger grains exactly like those of the Potsdam are common. Small grains of microcline, microperthite, and plagioclase are very common. On the Antwerp sheet the Theresa beds are characterized by a dirt brown weathered surface, thin-layered (one inch to a few inches) character throughout, organic markings or fossils on many weathered surfaces such as spiral flat coiled gastropods and curved and branching cylindrical structures with a diameter of one-eighth, to one-fourth inch, sandy calcareous character often with calcite sand rrystals. The Theresa, in part, appears to start sharply but con- formably above white sandstone although one or more beds of sand- stone, a few inches to a foot thick, are usually found in the lower ten feet. At other localities the lower beds of the Theresa are very NEW YORK STATE MUSEUM l82 sandy and include much sandstone similar to that of the Potsdam and it is then exceedingly difficult to decide exactly where the boundary lies between the Potsdam and Theresa formations. Good exposures of the Theresa may be seen at Reedville Falls, the quarry one and one-fourth miles northwest of North Wilna, in the road cuts three to five miles south of Antwerp in the Carthage road, the upper beds of the quarry a mile south of Philadelphia, and in many of the scarps formed by this formation. Much of the Theresa dolo- mite contains seams of calcite parallel to the bedding and rarely an oolitic bed occurs. At North Wilna a two-foot sandstone bed and other sandy layers occur well above the base and may represent a horizon similar to that of the Heuvelton sandstone farther north. About 30 to 35 feet of the Theresa formation are exposed a mile northwest of Woods Mill. The writer found no fossils in the Theresa formation, and the following quotation is taken from Cushing (T6, p. 28) : A series of “passage beds” of alternating sandstone and dolomite beds overlies the Potsdam everywhere in the circum-Adirondack region. To these beds we have been applying, for mapping pur- poses, the name of the Theresa formation. In the eastern sections these beds have large thickness, 150-200 feet, and are followed by the Little Falls dolomite, the three together forming the upper Cambrian (Ozarkian) series of northern New York. Deposition was seemingly continuous between these formations, and they grade into one another, without sharp boundaries, so that their separation from one another is largely a matter of convention, though they constitute three contrasted lithologic units. PAMELIA, LOWVILLE, LERAY, AND WATERTOWN LIMESTONES The lower beds of the Tug Hill plateau occupy the southwestern corner of the Antwerp quadrangle and are formed by a series of limestone formations. These beds have been briefly discussed at the beginning of this report and are continuous with the forma- tions on the Lowville quadrangle, which are described in detail in the following chapter by Ruedemann and therefore will not be dis- cussed here. At Devoice Corners there is a very small exposure of a green shaly limestone that may be the basal bed of the Pamelia formation at this locality resting unconformably directly on the Pre- cambrian igneous rocks. HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 183 PALEOZOIC ROCKS OF THE LOWVILLE QUAD- RANGLE By Rudolf Ruedemann INTRODUCTION The Paleozoic rocks occupy but a small area, altogether about seven square miles, in the southwest corner of the quadrangle. This area is a triangular sector of the Tug Hill plateau, which rises in a series of steep escarpments west of the Black river and surmounts the Precambrian rocks of the Lowville quadrangle to the east by nearly a thousand feet. The first of these scarps rising from the 8oo-foot line to near the 900-foot line is formed by the hard Lowville-Leray -Watertown limestone series. The top of the Leray- Watertown group forms a relatively narrow shelf rising to the 900- foot level, where another steep escarpment begins that rises to the i ioo-foot contour. This is formed by the Trenton limestones. Some 40 to 60 feet below the 1 200- foot contour begins the last escarpment shown on the L'ow- ville map (in the southwest corner). This is formed by the hard Cobourg limestone. A still higher scarp, 400 to 500 feet high, is seen about three miles west of the southwest corner of the map. This is composed of the great pile of shales and sandstones of the Utica and Lorraine formations. There are about four scarps and terraces distinguishable in the Paleozoic rocks west of the Black river, the more or less level ter- races being formed by harder sets of beds, as the Watertown and Leray limestones which underlie the first terrace about a quarter of a mile wide and on which a part of the village of Lowville is built. The most conspicuous and widest terrace on the map is that on top of the Trenton limestone, which is about a mile wide. The smaller escarpment of the Cobourg limestone sets off another terrace, also about a mile wide. A small portion of this is seen in the south- west corner of the map. From it rises the great scarp of the Utica and Lorraine formations, which leads to the Tug Hill plateau. Farther south the plateau is capped by the hard Silurian Oswego sandstone, which again rises in an escarpment, as for instance a mile west of Welch hill, eight miles south of the Lowville quadrangle. This series of terraces is a small section of a very pronounced feature of Adirondack physiography, namely, of the ring of escarp- ments of sedimentary rocks surrounding the Adirondacks on all sides. All these escarpments are eroded edges due to the recession 184 NEW YORK STATE MUSEUM by weathering of the piles of Paleozoic sedimentary rocks that, amounting to at least 1500 feet, once covered the Precambrian rocks of the Adirondacks to a varying depth. Erosion going on from the end of Devonian time has removed not only this enormous mass of material but also a great thickness of Precambrian rocks, as is proved by the greater dip of the surface of the Precambrian rocks under the sedimentary cover than in the exposed area (see Miller, ’10). This cuesta rim is especially distinct on the west side, where it forms the escarpment series west of the Black river and of West Canada creek, leading up to the Tug Hill plateau, and on the south side, where the Canajoharie and Utica shale escarpments rise in places more than a thousand feet above the river. There are lower rims on the north side, mainly north of the St Lawrence river. On the east side the faulting of the Champlain basin has largely masked the rim, although fragments or at least suggestions of it may still be found at the northeast and southeast corners of the massif. It is, in the east, mainly replaced by the Appalachian folding approach- ing the Adirondack Precambrian massif. A remarkable corollary of the strong development of these rims is the tangential direction, instead of a radial one, of the master streams of the drainage, so that the Black river, St Lawrence river, Lake Champlain, Hudson river and the Mohawk river form a quad- rangular frame of the Adirondacks (see sketch map figure 17), toward which the smaller tributaries diverge from the central por- tion of the Adirondack plateau. This strange arrangement of the principal water courses arose gradually, as the surrounding concentric rim of the Adirondacks began to rise higher due to the increasing erosion of the thick sedi- mentary formations, most of which were progressively overlapping on the Adirondack massif to a varying degree and increase in thick- ness as they drop away from the Precambrian mass. To this as a con- tributing factor is added the general dip of the old surface of the Pre- cambrian rocks away from the center of the Adirondacks. The result has been, that these tangential master streams have been formed along these escarpments, probably by stream piracy of the branches at the foot of the escarpments. They characteristically hug the foot of the escarpments, as is most typically seen in the Black river. It is also well recognizable in the Mohawk river and becomes there especially distinct where the fault blocks have brought the Pre- cambrian basement complex and early Paleozoic rocks to the sur- face. Also along the St Lawrence the Chazy-Black River-Trenton HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 185 series of limestones follows the river course to considerable extent on its north side, thus forming an outside rim. On the east side the Appalachian folds of the Green mountains and its dependent fault system have largely taken up the function of the exterior rim, holding the main drainage line to the foot of the Adirondacks. Figure 17 Sketch map showing the tangential master streams of the Adirondack massif and the escarpments that control their course. The out- line of the Precambrian area is shown by a double line of dashes. Drawing by Rudolf Ruedemann. With the great masses of Paleozoic sediments piled up in the concentric rim, and the master streams of the Adirondack drainage hugging the foot of the rim, it may truly be claimed that these streams on the whole have been gradually slipping down with their whole courses on the Precambrian surface by working against the inface and especially the foot of the surrounding rim. Typically then, the plateau and mountain system of the Adirondack Precambrian massif NEW YORK STATE MUSEUM 1 86 is surrounded by a concentric cuesta of sedimentary rocks with an inner lowland of varying width. Along the Black river the Precambrian is generally kept close to the west side of the river, and the boundary is a clear-cut line. On the other sides, notably the north and south sides, patches of the Pots- dam sandstone and Beekmantown dolomite have to a varying degree succeeded in remaining on the inner lowland. This is due to their natural hardness and the relative softness or greater solubility of the overlying shales and limestones. As an eroding agent the Mohawk river is mainly working against the Trenton (Canajoharie) and Utica shales on the south side, and while there is a major cuesta of the shales south of the river, there are minor escarpments of the lower Paleozoic rocks (much inter- rupted by the development of the fault blocks) on the north side, along the edge of the Precambrian rocks. In a similar way, the cuesta north of the St Lawrence river is formed by the Chazy-Trenton group of limestones, while the Potsdam and Beekmantown formations extend more or less into the inner lowland. There they are buried mostly under drift, but must have also formed in preglacial time small escarpments facing south. As we have seen, the Black River escarpment is composed of four secondary ones. The escarpments are hence all of a more or less complex character, resulting from the differences in hardness and solubility of the beds composing the retreating Tug Hill plateau. STRATIGRAPHY OF THE PALEOZOIC ROCKS The Paleozoic rocks exposed on the quadrangle are, in ascending order, the Pamelia, Lowville, Leray, Watertown, Trenton and Cobourg limestones. The mutual relationship, correlation and petro- graphic characters, as well as fossil contents of these formations have been very fully discussed by H. P. Cushing and the writer in the Geology of the Thousand Islands region (To) and more briefly by W. J. Miller in the Geology of the Port Leyden Quadrangle, Lewis county, N. Y. (To). There is, therefore, but a short note required in this place, especially, also, since these Paleozoic forma- tions cover but a very restricted area of the quadrangle. There is but one good section of the Paleozoic rocks found on the quadrangle. That is the one along Mill creek, passing through the village of Lowville, which affords the most perfect section through the Lowville (and Leray) limestones. It was for that reason that the name of the village was proposed in 1903 by Clarke and Schuchert, at the writer’s suggestion, for the limestone that since Eaton’s day (1824) had been known as the “Birdseye limestone.” HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 187 When the work of Cushing, Ulrich and Ruedemann in the Thou- sand Island region showed that the great mass of limestone that had hitherto passed as “Birdseye limestone” contained at the bottom a formation of Chazy age, this was separated as the “Pamelia lime- stone,” named after the town of Pamelia in Jefferson county (Cush- ing, ’08) ; and when it was further found that the Lowville is litho- logically and faunistically divisible into two divisions, the charac- teristic “Birdseye limestone” was retained as Lowville limestone and the upper, cherty division separated as Leray limestone (Ruede- mann, To), the name being taken from the town of Leray, Jeffer- son county. The Watertown limestone (Ruedemann, To) was originally known as the “seven-foot tier” and later as the Black River limestone (Hall, ’47), Since the Watertown limestone and the underlying Leray and Lowville limestones are closely united by their faunas and lithology, the name “Black River” has been extended to comprise this group of formations ; and thereby the term has been given its original conception, as set forth by several of the geologists of the First Geological Survey. The Trenton has not been subdivided in either the bulletin on the Thousand Island region or the Port Leyden quadrangle, because at the time these were written no detailed stratigraphic study had yet been made of our Trenton that would lead to a recognition of its principal subdivisions. This has meanwhile been done by Raymond (’12) and T. H. Clark (’19). Raymond divided the Trenton about Watertown into two principal divisions — the Trenton proper and the Cobourg limestone above it (first called Picton limestone). The lower division of the Cobourg limestone is reached in a waterfall in a ravine at the margin of the quadrangle, below the 1200-foot con- tour. The Atwater Creek, Deer River, Whetstone Gulf and Pulaski shales are all exposed within a few miles of the southern and western edges of the quadrangle, three of these with their type localities. The Roaring Brook section, at East Martinsburg, the best section for the limestone portion from the Pamelia to the top of the Trenton, measured by Cushing (Miller, To) Raymond (’12) and Clark (’19) is located only a mile south of the Lowville quadrangle, and the Whetstone Gulf section through the overlying shales, described by the writer (’25), is only four miles south of the quadrangle. Low- ville, situated between these standard sections to the south and the important Watertown sections to the north, has thus become an important stopping place for parties of geologists studying the Paleozoic stratigraphy, and it was already well known to paleon- tologists through the collecting zeal of Mr Hough, a generous con- 1 88 NEW YORK; STATE MUSEUM tributor to Hall’s original collections that were used in the prepara- tion of volume I of the Palaeontology of New York. This charmingly situated, prosperous town has thus considerable historic interest to American stratigraphers and paleontologists. PAMELIA LIMESTONE The lowest of the sedimentary formations is the Pamelia limestone. It consists typically of black and dove limestones alternating with gray and white earthy limestones. Its most characteristic fossil is a branching form of Tetradium in which the corallites remain free, and which fills certain beds of the black limestone. It was cited by Cushing and Ruedemann (’io, p. 72, 84) as Tetradium syringopo- roides Ulrich MS. Besides this there occur forms of Bathyurus that differ from the B. extans Hall of the overlying Lowville and have been described by Raymond (’13) as B. johnstoni and B. superbus. Besides these one finds gastropods, cephalopods, lamelli- branchs, other trilobites, corals and sponges. The Pamelia limestone is, according to Ulrich ('ll), of Chazy age, and fills an interval between the middle Chazy (Crown Point limestone) and upper Chazy (Valcour limestone) of the Champlain basin. It is then considerably older than the Lowville formation, and the entire Blount division comprising the Holston, Athens, Tel- lico and Ottosee formations of the Appalachian region is missing between the two formations on the west side of the Adirondacks, the hiatus being partly filled by the lower Normanskill graptolite shale in the eastern slate belt of New York. Its thickness is, according to Cushing (To, p. 69), 150 feet or more in the western portion of the Theresa quadrangle. It thins eastward toward the Adirondacks and southeastward, as is shown by the fact that in the wonderfully complete section at Martinsburg it is but 71 feet 8 inches thick, and Miller found that it dis- appears before the Remsen quadrangle is reached. In the Mill Creek section of Lowville only about 12 feet of Pamelia are ex- posed at the base of the continuous outcrop in the gorge. A couple of strata of dove-gray limestone with crystalline specks appear far- ther down in the creek bed just above the lowest bridge. Since the contact with the Precambrian is not exposed in this section, an ac- curate measurement of the Pamelia can not be obtained here. It is, however, obvious that its thickness is not less than at Martinsburg, nor much more. The lower black limestone and the basal sandy beds are entirely hidden by glacial drift and alluvial deposits. A few feet of Pamelia are also exposed at the base of the section at the town quarry, located at the top of the first platform west of HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 189 Dadville. Another small outcrop is in an abandoned quarry along the railroad half a mile north of the town quarry. In both cases only the uppermost beds are seen below typical Lowville with Tetradium cellulosum. Both these outcrops indicate a smaller thick- ness for the Pamelia than was observed at Roaring brook and sug- gest considerable local variation of the thickness owing to the irregular surface of the Precambrian, as is the case with the Potsdam sandstone farther north, where it forms the base of the sedimentary column. The following is the detailed section made by Cushing along Roar- ing brook and published by W. J. Miller (To, p. 22-23) : Feet In. Four inches of blue-gray calcareous shale above and nine inches of same beneath with a four-inch layer of mottled blue limestone like that beneath 1 5 A massive bed of blue granular limestone, mottled and laminated and with pebbles of blue dove limestone 1 6 Hard, gray-white limestone with spots of crystallized calcite 9 Bluish gray, thin-bedded limestone, with ripple marks and ostracods.. 2 2 Shaly, dirty white mud limestone beds, mud-cracked 1 6 Bluish, subgranular, thin-bedded limestone 9 Blue dove, hard, mottled limestone, shaly below, traces of fossils 1 4 Hard, gray, granular limestone, pinkish tinge, much calcite 6 Thin, ripple-marked, dirty white, mud limestone beds 1 Hard, blue-gray, subgranular limestone, sand grains and ostracods above 2 6 Hard, gray-white limestone, less earthy than usual; calcite spots 9 Subgranular, blue, laminated limestone; welded contact with above... 2 5 Very massive, gray-white, earthy limestone, irregular splitting 5 9 Massive, blue dove limestone in two layers ; laminated ; styliolites and traces of fossils 1 9 Thin and thick-bedded, blue, subgranular limestone ; massive basal layer; laminated; much mud-cracked 7 9 Earthy, impure, gray-white beds ; massive and irregular splitting ; lower 12 to 18 inches hard, gray-white, calcite spotted, bunchy sur- faced 3 11 Gray-white, impure, earthy, blocky beds 1 8 Impure, gray-white, reddish tinged, earthy limestone, with purer layer of reddish limestone at base; quite massive; summit less red 7 8 Bluish black, fine, hard limestone, like second beneath but unfossil.. 1 Gray to blue-gray, magnesian looking limestone in three to six inch beds 2 9 .Three heavy layers of hard, bluish black limestone; ringing; multitude of small fossils, lamellibranchs, cephalopods, one-celled Tetradium, etc 4 7 Red and green, impure, calcareous, sandy mudstone, large sand grains ; upper six inches thin-bedded, rest massive 7 Very sandy firm green limestone, full of quartz grains 1 3 Hard, gray-white, somewhat sandy limestone or dolomite ; calcite spots 1 2 Hard, ringing, green, calcareous sandstone 10 Red, sandy, rotten shale 1 6 More solid, red and green calcareous sandstone ; rots easily 4 Red and green, rotten, calcareous sandstone 9 Green and red arkose; perhaps calcareous 11 Conglomerate resting on Precambric 10 Total 71 8 190 NEW YORK STATE MUSEUM LOWVILLE, LERAY AND WATERTOWN LIMESTONES The Lowville limestone has been described by the writer (To, p. 80) as follows : Lowville limestone. The Lowville limestone, which is the “Birds- eye limestone” of the old Geological Survey reports, has its maxi- mum development in New York in the region of the lower Black river, or in the southern portion of the area here mapped. It reaches there about 60 feet in thickness. It consists typically of thick and thin bedded, fine grained dove limestone which shows a character- istic ashen-gray weathering and contains either numerous more or less vertical worm tubes denoted as Phytopsis and filled with cal- cite (producing the “birdseyes” in sections) or shows in profusion the horizontally spreading tabulate coral Tetradium cellnlosum and related species. Between these typical Lowville beds there are inter- calated others of subcrystalline dark to black limestone, or of oolitic or also of shaly whitish weathering limestone. These intercalations usually contain a larger fauna than the dove limestone and carry lamellibranchs, gastropods and cephalopods, as well as ostracods and trilobites. The basal bed is conglomeratic and of very variable thickness ; it is overlain by several feet of strata that contain quartz grains or grit bands and are more or less shaly, the shaly limestone gradually becoming more massive upward and assuming the characters of the typical rock. These more or less sandy beds comprise about 4 feet. The uppermost portion of the Lowville beds which has been men- tioned by the earlier authors as the “cherty beds” has been found by Professor Cushing and the writer to be quite distinct from the typical Lowville beds and separated from them by an unconformity. It has for that reason been distinguished as a subdivision under the name Leray limestone and will be described separately (see below). It appears that in this region the Lowville beds beneath the Leray member can be conveniently divided into an upper and lower divi: sion of nearly equal thickness, the upper division alone containing the abundant Tetradium cellulosum and larger Phytopsis, as well as the typical massive dove limestone strata, while in the lower division more dark or black subcrystalline limestones containing only smaller forms of Tetradium and Phytopsis and more thin bedded dove lime- stones abound. In this lower division also two or more horizons of Stromatocerium can be observed, which give the beds a very irregular concretionary appearance. HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES I9I Four species of fossils have to be considered as highly charac- teristic of the Lowville formation in the area here mapped, viz : Tetradium cellulosum (Hall) Orthoceras multicameratum ( Emmons ) Hall O. recticameratum Hall. Bathyurus extans (Hall) These species are not known to occur above or below the Low- ville limestone, and are common enough to occur in every exposure of the formation. The type section of the Lowville is at Lowville. It is well exposed there in a quarry at the railroad bridge over Mill creek, where in the creek bank the uppermost part of the Pamelia (about 12 feet) is shown and a continuous section into the Leray limestone can be obtained. The section (Ruedemann, To, p. 83) is as follows: Section at railroad bridge at Lowville (type section) Lowville section 6' Cherty beds. Dark blue, finely granular limestone, dirty white weathering. Columnaria horizon at base 1' 6" Bed full of horizontal worm tubes. Chert horizon at base 5' 9" Transitional bed from Leray to typical Lowville. In aspect like cherty beds with a few cherts, but contains also Tetra- dium cellulosum, besides Leperditia fabulites, Rafines- quina minnesotensis, and other brachiopods and bryozoans Base of Leray 3' Dove limestone, massive. Phytopsis tubulosum common near top ; a few Tetradium cells 3' 1" Compact dark dove limestone, full of fossils (Tetradium, gas- tropods, lamellibranchs) and of crystalline calcite 4' 9" Thin-bedded, dove limestone, full of Tetradiums (form with narrow, round tubes) 5" Dark, fine-grained impure limestone with argillaceous streaks, containing a small Monticulipora 2' 4" Lighter, massive bed of dove limestone with few Tetradiums. Lowest eight inches black, with thin seams of flint 4" -7" Stratum of granular, light gray limestone full of lamellibranchs and gastropods, their shells filled with calcite 1' 3" Black, massive, crystalline limestone, full of Tetradium 3' 4 " Black to dark gray thin-bedded dove limestone, containing a few Phytopsis 4" Same rock as above, but full of a narrow form of Phytopsis 4" Black, dove limestone stratum full of crystalline calcite T 7" Dark gray, granular limestone with many calcite crystals. Bottom of quarry 4’ Dark gray, thin-bedded dove limestone, weathering shaly 4' Harder, argillaceous limestone 3' 10" Shaly dove limestone, varies much, very shaly in middle, full of sand grains, contains a few lamellibranchs i’tt " Hard, oolitic blue limestone, full of quartz grains and pebbles 6"-io" Shaly bed with seams of quartz grains or grit bands o-3’+ Dark bluish gray limestone, full of pebbles, shale below. Very variable in thickness. Unconformity. 192 NEW YORK STATE MUSEUM Base of Lowville 1' 10" Gray and pinkish granular limestone, dove in parts 4" Thin bedded shaly limestone, sand grains near top T 10" Dove, dark mottled, fine-grained limestone, typical upper Pamelia T 7" Dove limestone with argillaceous reticulation, light pink in parts, weathering shaly 9" Bluish black flinty dove limestone 10" Gray, granular limestone with calcite and quartz grains, in parts conglomeratic, a few fossils. (Rafinesquina incrassata) T 5" Light dove limestone, somewhat argillaceous, coarsely lami- nated. Phvtopsis on top 2’ Grayish, bluish, blocky, subgranular limestone T Compact bed of harder, light bluish gray limestone T + Dove, light gray limestone with crystalline specks The Lowville beds are also finely exposed now on the south side of Mill creek in the state quarry, the Lowville rock being crushed there for road metal. The Lowville is of less variable thickness than the Pamelia. In the Lowville section it is 37 feet 8 inches thick. Miller records a fairly uniform thickness of 54 to 57 feet on the Leyden quadrangle. Cushing measured 54 feet 7 inches in the Roaring Brook section. These measurements include the Leray limestone and the beds above that were united by Cushing and Ruedemann with the Water- town limestone and they, with those subtractions, give a thickness similar to that observed at Lowville. Also below Watertown (Ruede- mann, To, p. 82) in the bank of the Black river a thickness of 37 feet was observed for the Lowville s. str. (56 feet from the 7 foot tier down). The Lowville limestone was deposited by a sea encroaching upon the land from the southwest. It therefore extends far in that direc- tion and can be recognized as an invasion from the Gulf of Mexico. In mapping the Thousand Island region, Cushing and Ruedemann found that in the field the upper portion of the Lowville limestone, as originally conceived, grades lithologically and faunistically into the Black River limestone (seven-foot tier) of the earlier authors to such a degree that it is impossible to map them separately. These upper beds consist of heavily bedded, dark gray to black dove limestone with numerous layers of black cherty nodules. Where weathered these beds together with the overlying Water- town limestone are readily distinguished from the thinner bedded Lowville beds by their breaking up into small cubic blocks the size of the fist. On account of the transition the boundary was drawn where the cherts begin and Tetradium cellulosum disappears. Both the Lowville limestone s. str. and the Leray limestone member form together the Lowville formation, which is a most persistent HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 193 formation over a large area in contrast to the succeeding Watertown limestone. It was further found that the uppermost beds of the Lowville formation, about six feet thick, contain the cephalopods and other fossils characteristic of the seven-foot tier and are lithologically bound to it while separated by an irregular surface from the underlying Leray. These beds were therefore united with the seven-foot tier into the Watertown limestone. To this same black limestone, which is much cpiarried because it furnishes large blocks, has finally been added a bed of black impure limestone, one and one-half to two feet thick, that was observed at Watertown overlying the seven-foot tier. The Leray limestone thus conceived is 13 feet thick and the Water- town limestone 15 feet thick. The two together, on account of their heavily bedded structure, usually form a fairly broad platform or terrace, as that extending northward from Lowville, between the Lowville and Trenton scarps. The Leray limestone is shown well in the upper parts of the two quarries on either side of Mill creek in lower Lowville and in the ravine below the falls. The Watertown limestone is exposed below and just above the bridge of Main street in Lowville. The character- istic erosion of pure limestone by running water into sharp narrow ridges between deep channels is seen in the bed of the creek above the bridge. Smaller outcrops are scattered over the first platform, while the Lowville outcrops are seen in the first escarpment below the platform, either in ravines or in small abandoned quarries. TRENTON LIMESTONE The Trenton limestone measures at the combined Roaring Brook and Atwater Creek sections 475 feet, according to Miller (To, p. 31). This is a greater thickness than is anywhere shown farther south in the State. The increasing thickness from the Trenton Falls region northward is mainly due, according to Raymond (’12), to the inter- calation at the top of a limestone formation that is mainly developed to the north of the St Lawrence and is termed the Cobourg lime- stone. This formation comprises the upper 85 feet of the section at Martinsburg (Clark, ’19, p. 9) leaving still 390 feet for the Tren- ton proper, as against 346 feet of the combined Trenton Falls-Rath- bone section. The Trenton s. str. consists of thin-bedded gray, bluish and dark to black fine-grained limestones, alternating in places with thicker, coarse-grained or coarsely crystalline beds. These beds contain the characteristic Trenton fossils in varying numbers. They have been divided by Raymond into five zones from 194 NEW YORK STATE MUSEUM their principal fossils and the overlying Cobourg limestone into two zones. These zones are : 7 Light grey, coarse-grained limestone, with Rafinesquina dcltoidea, Hor - motoma trentonensis , and other fossils. Thickness, 26 feet. 6 Thin-bedded blue limestone with shaly partings. This zone contains many fossils, Rafinesquina deltoidea being the common and characteristic one. Thickness, 92 feet. 5 Thin-bedded blue limestone with thick shaly partings. Prasopora simulatrix is common all through but especially abundant at the base. Thick- ness, 100 feet. 4 Thin and thick beds of limestone, mostly dark and fine-grained. Diplograptus amplexicaulis is a common fossil. Thickness, 35 feet. 3 Thin-bedded dark limestone, with Triplecia extans, Dalmanella r ogata, etc. Thickness, 20 feet. 2 Thin-bedded dark limestone, alternating with thick coarse-grained beds. Cryptolithus tessellatus is the common fossil. Trematis terminalis, Platy- strophia lynx, Calymene senaria, and many other fossils are present. Thick- ness, 41 feet. 1 Thin-bedded gray limestone with an abundance of Dalmanella rogata, and some other fossils. Thickness, 32 feet. The Cobourg limestone consists in the lower division (over 90 feet) of thin-bedded blue limestone with shaly partings, in the upper division (26 feet) of light gray, coarse-grained limestone. The Lowville area shows the lowest zone directly overlying the Watertown limestone in the bed of Mill creek just above the bridge of Main street. Thence upward in the creek and its branches beyond the limits of the quadrangle the other zones are well exposed, in part at least. The Prasopora beds, or Trenton proper, some 100 feet thick, are well shown with their fossils, especially also the semi- spherical zoaria of the bryozoan Prasopora simulatrix in an aban- doned quarry along the road to Champion where it leaves the quad- rangle. The beds of division 7 (lower Cobourg limestone) with Rafinesquina deltoidea are seen only in a ravine at the edge of the sheet near the southwest corner. The dark black shales described as Atwater Creek and Deer River formations as well as the Lorraine beds (Whetstone Gulf and Pulaski formations) are exposed in the higher escarpments west of the Lowville quadrangle. They have been described by Ruedemann (’25) in New York State Museum Bulletin 258. ECONOMIC GEOLOGY HEMATITE DEPOSITS Introduction. Hematite ores similar to those of the Hammond and Antwerp quadrangles are found locally in the Precambrian rocks, beneath or near beds of Potsdam sandstone, at many places throughout the foothill Grenville belt of the northwest Adirondacks. HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 195 They have been known since at least as early as 1812, and have been mined at intervals during the past 100 years, with a total produc- tion of around 2,500,000 tons. The competition of the Lake Superior ores, together with business depression, resulted in the final shutting down of the last working mine in 1910. In the early days the ores were smelted on the ground in charcoal furnaces, but later v/ere sent to various iron centers for mixture with more refractory ores. During the past 30 years the ore was shipped mainly to Pennsylvania. The early history of the utilization of these ores is referred to by Durant and Pierce (’78), from whom the following quotation is taken : The first blast-furnace in northern New York was erected at Rossie. It was built in the summer of 1813 and got into operation in 1815. The Caledonia iron mine, one mile and a half east of Somerville, began to be wrought about the same time or a year before. A part of the time previous to 1826 bog iron ore was used. The principal supply has been from the Caledonia mine in the town of Rossie, the Keene and Wicks mines in Antwerp, and a small open- ing adjoining the Kearney mine. The last blast at this furnace ended October 14, 1867. The Caledonia mine was estimated to have furnished 100,000 tons of ore previous to 1852. The important ore deposits within the area under discussion are practically confined to a narrow belt about four and one-half miles long between Keene and Antwerp, and to the Shirtliff mine three and one-half miles north-northwest of Philadelphia. The Keene- Antwerp belt includes the Pike, Clark, Kearney, Caledonia, Keene. Morgan or Pardee, Old Sterling, Wight, Dickson, Ward, and Col- burn properties. Minor quantities of ore have been found in a belt striking west-northwest from a mile southeast of Brasie Corners to half a mile west of South Wood’s School, and at the northeast end of Muskalonge lake, and a mile southwest of Sherman lake. Fer- ruginous deposits have been prospected at a number of other locali- ties. The ore deposits of the Keene- Antwerp belt are all located along the contact zone between the Grenville limestone and gneiss, and occur in part in the limestone and in part in the gneiss. A mile north of Antwerp the contact line turns west toward Halls Corners, and a pocket of hematite is reported to have been cut on the state road from Antwerp to Theresa about two miles west of Antwerp. The deposits have been described by Smock (’89, p. 44-48), Smyth (’94), and Newland (’21, p. 126-29). The mines had all been abandoned for many years before the writer’s visit, and the open- ings and shafts were filled with water. The following accounts of the mines are, therefore, taken largely from previous reports. ig6 NEW YORK STATE MUSEUM Clark and Pike mines. This property has also been described by Smock (’89, p. 48) as follows: One-fourth of a mile northeast of the Kearney mine the Gouver- neur Iron Ore Company has opened on adjoining properties what are known as the Clark and Pike mines. They are on opposite sides of a low ridge of sandstone (Potsdam) and in a synclinal fold of this formation. The three slopes of the Clark mine are at the south- west border of the hill and on the northeast dipping ore bed. The greatest depth reached is 120 feet — on the slope. The ore has been found to vary in thickness, because of the undulating footwall, the hanging wall being more nearly uniform in its dip. Some hematite occurs in the latter, in thin layers in the sandstone, and the workings are not carried to the ore limit in that direction. The three slopes are located within a length of 300 feet on the course of the ore bed. The average dip is stated to be 38°. The Pike shaft is not as deep as the Clark mine. The total output is reported by Mr Webb to be 45,000 tons. Kearney mine. The following description is taken from the report by Smock (’89, p. 47) : The Kearney mine consists of a large open pit and extensive underground workings on the hill north of the Caledonia. The ore bed dips to southeast and east, under sandstone strata (Potsdam). Near the surface and at the outcrop the cap of sandstone was replaced by earth and the ore quarried out in an open pit, to a depth of nearly 100 feet. In this mine the ore occurs associated with the serpentine (chlorite schist) in large irregular-shaped bodies. Caledonia mine. Newland (’21, p. 128) has described the Cale- donia mine as follows : The ore is soft hematite with 50 per cent or more of iron, and has a relatively high proportion of lime, beneficial to its smelting quali- ties. Actual analyses of ore shipments show 52 to 58 per cent iron, 7 to 8 per cent lime, and 0.5 per cent phosphorus. The deposit lies near the contact of the Grenville pyritic gneisses and crystalline limestone, and the ore is encountered in both. It is capped in part by Potsdam sandstone which rests on an uneven erosion surface of the crystallines. The gneiss in this place is a quartz schistose phase, carrying graphite and pyrite, more or less altered secondarily to chlorite and serpentinous products, as an accompaniment likely of the ore-forming process. The footwall which is followed in the main shaft consists of limestone, while the schist is seen on the sur- face to the east and also is encountered underground in the hanging wall and as horses within the ore. The limestone along the contact with the deposit is uneven and contains pockets and irregular cavi- ties filled with hematite. The hematite not infrequently bears marks of replacement in its simulation of the banding and structural char- acter of the gneiss. HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES I97 Keene mine. The Keene workings are located about two-fifths of a mile a little south of west of the railroad station at Keene. The ore here is overlain by Potsdam sandstone, with a coarse quartz breccia or conglomerate at the base. A quantity of heavily pyritic chlorite schist was cut in the workings. Smock (’89, p. 46) writes that the mine has been worked over 500 feet on the line of strike and is 400 feet deep, and the ore has been on an average 12 feet thick. Some compact crystalline hematite and siderite occurs in association with the soft hematite here. Old Stirling mine. The Old Stirling mine is located two and three-fourths miles north of Antwerp at the terminus of the old railroad spur. Smock (’89, p. 45) states: It was first opened in 1836. The open pit at the northeast is 115 feet deep and approximately 500 by 175 feet. The underground workings are south and southeast of it, and the ore has been fol- lowed for a distance of 900 feet, and to a depth of 185 feet. The ore stands up well, and, by leaving pillars with arched roof in the galleries and drifts, no timbering is necessary. The ore varies from a specular ore of metallic luster and steel-gray shade to amorphous, compact masses of deep red. The total output of the Dickson and Old Stirling ores is estimated by Mr E. B. Bulldey, president of the company, at 750,000 tons. The ore occurs in a so-called serpentine rock which Smyth (1894) proved to be a chloritic alteration and replacement of gneiss and granite. Newland (’21, p. 129) writes: Potsdam sandstone overlies the ore on the west side of the pit, and its contact is scarcely distinguishable owing to the alteration that has taken place in the lower beds. The ore is somewhat siliceous and carries from 40 to 55 per cent iron with about 0.1 per cent phosphorus. The last operations were in the period 1904-10. Dickson mine. The Dickson property is located in the township of Antwerp, about a mile and a half north of Antwerp. The ore occurs as an earthy hematite replacing chloritic schist bands within gneiss, which dip 30 to 50 degrees northwest. A heavily pyritized band is also present. Newland (’21, p. 129) states that the mine was opened by an incline which in 1906 had reached a depth of 160 feet on the dip, which was steep to the northwest. The thickness varied from io to 40 feet. Shirtliff mine. The Shirtliff mine lies three and one-half miles northwest of Philadelphia (Antwerp quadrangle) where a small pond is marked on the map. It was first opened in 1838, was 198 NEW YORK STATE MUSEUM worked up to 1880 or a little later, and was exhausted before the report by Smock (’89). Newland (’21, p. 129) writes: The ore was exploited by an open cut 1300 feet long and for a farther distance of 500 feet underground, while the workings reached a depth of 200 feet or more. The pit lies at the edge of a Potsdam sandstone cap. Cuts in altered chloride gneiss have also been made in search for ore at the east end of the sandstone area. The ore dips northwest. Country rocks. The most common type of rock in direct asso- ciation with the hematite ores is a green quartz-chlorite schist, locally with disseminated flakes of graphite. Locally, as near the Cale- donia and Kearney mines, the country rock is a quartz-sericite aggregate stained with hematite, but this is not common. The field relations show that these schists are the result of alteration and par- tial replacement of bands of the Grenville gneiss, or locally of granite. Quartz veins in the gneisses are the last to be replaced. Grenville limestone is occasionally the gangue rock of the ore, and locally the base of the Potsdam sandstone is ferruginous. The origin and nature of the chloride and sericitic rocks has been discussed in detail by Smyth (’94 a and b). He has shown that the so-called serpentine of the older authors is a chlorite and that it is a replacement of feldspar, and to a lesser extent of quartz. He gives an analysis of both the chloride and sericitic type of alteration which are quoted below, together with another analysis of the sericitic Shepard, also from the Caledonia mine. I II III Si02 29.70 46.90 46.70 ai2o3 17-03 35-73 31.01 FeO* 27-15 2.48 3-69 MgO 10.66 0.83 0.50 CaO 1.68 0-45 Trace Na20 0.56 0.48 Trace K20 0. 10 6.41 11.68 h2o 11.79 S-oo 5-30 98.63 98.28 98.88 I Greatly altered (chloritized) granite. Old Stirling mine, C. H. Smyth, analyst. II Greatly altered (sericitized) granite or gneiss, Caledonia mine, C. H. Smyth, analyst. Ill Greatly altered (sericitized) granite or gneiss, Caledonia mine, J. L. Smith and G. W. Brush, Amer. Jour. Sci., (2) XVI : p. 50. ^Calculated as ferrous iron, but there can be no doubt that some of it is in ferric condition. (Comment by Smyth.) The presence of these pyritic-chlorite schists raises the problem, however, as to whether the chloritic alteration was accomplished at HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 199 the time of formation of the iron ores, as suggested by Smyth, or at a much earlier date as an accompaniment of the introduction of the sulphides of iron, which are also so intimately associated with these schists, or is in part of one origin and in part of another. The writer finds that a chlorite is definitely replaced by hematite, and is everywhere found in association with strong sulphide of iron mineralization, even where hematite deposits are absent. It is there- fore certain that at least a part of the chloritization was effected by hydrothermal replacement at the time of introduction of the sul- phides of iron. Chlorite schists are also found locally, however, that do not contain pyrite, as along the west side of the Potsdam sandstone area two miles north-northwest of Antwerp, and in a band about two miles southwest of Fargo. Granite and syenite respectively are here altered to chloritic rocks, and it seems probable that this alteration was accomplished at the same time as the hematite replacements. The sericitic alteration is similar to that found at so many locali- ties in Grenville gneiss beneath caps of Potsdam sandstone, and is ascribed to the results of Precambrian weathering and subsequent recrystallization. The altered rocks are thus in part of hydrothermal origin and in part formed by Precambrian weathering and solutions accompanying this weathering. The writer has had an opportunity to study the thin sections used by Smyth, and in addition a number from specimens collected by himself. The evidence all confirms Smyth’s conclusions as to the origin of the chlorite and sericitic schists through alteration and replacement of gneiss or granite. Smyth (’94, a) has further pointed out the general association of bands of gneiss rich in sulphides of iron with the iron ore deposits. A study of the pyrite and pyrrhotite deposits in 1917 by the author has served to further confirm this. A band of chloritic schist aver- aging over 40 per cent pyrite (by weight), with a width of 10 to 20 feet, is exposed at practically every iron ore prospect in the Keene- Antwerp belt ; a pyritic-chloritic band is prominent south- west of Oxbow in association with iron ore ; and the ore in the belt through Brasie Corners is associated with pyritic gneiss. There is uniformly a cap of Potsdam sandstone over the ore, or so near by as to indicate that it had formed a cover over the ore until recent erosion had removed it. The basal part of the sandstone where it rests on ore is often ferruginous, and grades into the ore. Smyth (’94, a, p. 509) states that at the Old Stirling mine the 200 NEW YORK STATE MUSEUM sandstone just above the ore contains distinct pebbles of the latter. Locally the basal part of the sandstone is ferruginous where it over- lies limestone. Thin hematitic layers occur in the sandstone locally, and at the locality one mile southeast of Brasie Corners a bed of greenish sandstone with a chloritic mineral occurs intercalated near the base. Origin. An extraordinary variety of hypotheses has been pro- posed for their origin. Emmons (’42, p. 97) regarded the ores, together with the associated so-called serpentine, as eruptive; Van- uxem (’42, p. 267) considered them as rather superficial concentra- tions lying between the crystalline rocks and the Potsdam sandstone, and as clearly connected with the latter as with the former ; Brooks (’72, p. 22) examined the Rossie mines and concluded that they showed a continuous series of sedimentary deposits, including sand- stone, ore, magnesian rock (serpentine of Emmons), and crystalline limestone, together with some granite; Smock (’89, p. 10) states that there are probably two classes of ore deposits, the original sediments and secondary concentrations ; Kimball states that the ores are the result of replacement of limestone; N. H. Winchell (’93) examined two iron prospects in the vicinity of Canton and Potsdam, and described the iron deposits as the ferruginous base- ment portion of the Potsdam sandstone and derived from near-by hematite bodies in the underlying Precambrian formations ; Smyth (’94) found that the so-called serpentine was a chloritic schist resulting from the alteration and replacement of granite and gneiss, and that the ores were formed primarily by the replacement of lime- stone by iron derived from decomposing pyrite ; Crosby (’02, p. 233) studied conditions at the Old Stirling mine and concluded that it was most probable that the chloritic rock or greenstone was a highly altered, basic eruptive dike, and that the iron ore resulted from the oxidation of sulphides of iron segregated in the eruptive ; New- land (’21, p. 127- ) adopts Smyth’s hypothesis for the origin of the ores, but states that the hematite was formed probably largely in Postcambrian time, since the Potsdam in contact with the ore is itself heavily impregnated and altered, so as to be with difficulty distinguished from the normal ore. Ingall (’99, p. 69-80) ascribes similar deposits in eastern Ontario to merely casual aggregations of iron peroxide, probably resulting from the decomposition of the ferruginous dolomitic portions of the basal beds of the Paleozoic series. He believes the ferruginous basal beds of the Paleozoics resulted from deposition by solutions carrying iron carbonate infiltrating into the porous portions of the HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 201 sandstone, and the masses and veins of ore in the underlying Gren- ville limestone resulted from fillings of fractures and waterworn channels and solution caverns. The same explanation has recently been adopted by Baker (T6, p. 17) for deposits of this type in Leeds county. The writer would sum up the essential features of the occurrence of the iron ores as follows: (1) they are uniformly associated with belts of Grenville schist or gneiss which are strongly mineralized with pyrite ; (2) the ore occurs for the most part as replacement of Grenville chloritic schist and limestone, and in very minor part as a ferruginous basal portion of the Potsdam sandstone; (3) a cap of sandstone rests, or has rested until very recently, over all the ore deposits ; and (4) the Potsdam sandstone rests on a Precambrian peneplain surface, and at many places outside of where iron ore deposits occur, there are deeply weathered gneisses preserved beneath the sandstone. The relations are shown diagrammatically in figure 18. Figure 18 Generalized structure section showing relations of iron ore and pyrite deposits to the country rocks, Keene-Antwerp belt From this evidence the writer follows Smyth (’94, a) in drawing the conclusion that the iron deposits are the result of oxidation and weathering in Precambrian time of the associated pyritic schists to yield iron-rich solutions which were effective in replacing near-by limestone and the chloritic schists to depths up to 200 feet. At the time of deposition of the Potsdam sandstone, part of the iron ore is believed to have been eroded and incorporated in the basal por- tion of that formation as detrital hematite. Smyth reports (’94, a, p. 509) that at the Old Stirling mine the sandstone just above the ore contains distinct pebbles of the latter. Since the formation of the ore, circulating groundwaters have locally penetrated along fractures and formed veinlets of specularite, in part associated with quartz, calcite or siderite and vugs lined with quartz, siderite, and specularite. These occur both in the ore and in the ferruginous base of the overlying sandstone. There is no corrosion of the sand 202 NEW YORK STATE MUSEUM grains of the Potsdam sandstone by the hematite, and no positive evidence of an ore shoot of hematite crossing the bedding of the sandstone has been described ; so that the writer is inclined to con- sider the age of the ore deposits as for the most part pre-Potsdam. Future possibilities. Newland (’21) writes that: The hematites assay upward of 40 per cent iron, the richest aver- aging 55 to 58 Per cent, and a variable phosphorus content, too high for Bessemer ore. The main drawback seems to be a rather high percentage of silica, arising from unreplaced quartz and silicates of the gneiss and their alteration products. Little systematic exploration has been attempted in this section ; in fact, there is less known about the resources than of any other iron ore district in the state. Most of the deposits, so far opened, were exposed at the surface or showed indications of their presence by the character of the soil, which usually takes on a deep iron stain wherever in contact with the hematite below. The cover of drift on the uplands is thin, a few feet of sand and gravel with some bowlder clay constituting the glacial deposit in most places. There is no doubt that deposits are concealed under the Potsdam strata, through which the iron is not readily revealed, and others may lie under drift-filled valleys in which the glacial accumulation may have a thickness of 100 feet or more. A guide to exploration is to be found in the occurrence of pyrite which accompanies the hematite belts throughout their extent. In the Keene-Antwerp belt there seems to be a promising field for exploration southwest of the Cale- donia mine, along the contact of the limestone and schist, which is mainly hidden by a thin cover of sandstone and has not actually been tested in the immediate neighborhood of the Caledonia mine itself. It is along the contact that the iron-bearing solutions would find the easiest channels for their circulation and the limestone, being readily soluble, would seem to offer the most available ground for the accumulation of considerable bodies of hematite. i A number of areas are shown on the map where Potsdam sand- stone overlies Grenville gneiss bands which are more or less pyritic and consequently afford conditions for at least small pockets of hematite to have been formed and preserved beneath the sandstone caps. GALENA VEINS Structure and character. Lead deposits in the form of galena- bearing calcite veins are found grouped in three areas on the Ham- mond quadrangle ; west of Rapids School one and one-half to two and one-half miles southwest of Rossie ; half a mile south of Macomb or Pierce’s Corners ; and near Bigelow School. A few scattered veins are found elsewhere in the towns of Rossie and Macomb. HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 203 The Rossie group of veins lie in general along an east-west line two and one-half miles long; three of the Macomb group also lie along an east-west line half a mile long ; and the three veins of the Bigelow School belt similarly lie along an east-west line one and one-half miles long. Yet all but one of the 12 veins that lie along these three east-west lines themselves have a strike of N. 58° to 8o° W. and therefore make an angle with the trend of the belt in which they occur. The general strike of all the galena veins is N. 6o° to 70° W. Since the general trend of the bedrock structure is north- east, the veins cut across the structure at a high angle, and are found in all kinds of rock. The veins are of the fissure type, varying from a few inches to a few feet in width, and up to at least a thousand feet in length. They dip steeply. South of Rossie they are largely in granite and quartz diorite ; the Macomb group is in crystalline limestone, as are the Bigelow School veins, except that at Mineral Point which crosses a granite dike. The gangue minerals consist of calcite with a little fluorite, barite, and, in the veins in limestone, some dolomite. Galena is the pre- dominant sulphide associated with sphalerite, and a little pyrite and sparse chalcopyrite. Celestite and strontianite have also been found in the veins. The galena occurs disseminated through bands of the calcite, and as crystal aggregates or clusters. Much of the calcite in some of the veins is pink. Occasionally thin bands of very fine- grained quartz occur within the veins. The vein matter often shows a rudely banded or crustified structure, with the calcite in growths oriented at right angles to the walls. Locally, sphalerite-bearing veins are younger than the galena veins. Rossie group. The Coal Hill vein of the Rossie group is located one and one-half miles south of Rossie, nine-tenths of a mile west- northwest of Rapids School. It is easily located by the fact that a tall sandstone chimney on the property still stands and is visible from the road. The vein has been described in detail by Smyth (’03), from whose report the following description is taken. The words in brackets are inserted by the author. The veins of which there are four, occur in a flat-topped hill. . . . Two varieties of gneiss, both quite massive, are present. . . . The pink gneiss [granite] is younger than, and intrusive in, the gray gneiss [quartz diorite]. . . . The veins cut across the foliation of the gneisses, at a high angle, approach parallelism with one another, and have a strike of about N. 8o° W. . . . Whether or not the fracturing was attended by any large vertical displacement is not 204 NEW YORK STATE MUSEUM shown, although the vein fissures are evidently due to faulting. . . . The Coal Hill vein varies from two to six feet in width, is nearly vertical and sharply defined [Stoped for a length of over 200 yards]. As is well known, the gangue is coarsely crystalline calcite, carrying galena, with a little pyrite, and small amounts of sphalerite and chalcopyrite. Some parts of the vein, particularly along the margins, contain abundant inclusions of the wall rock, forming a breccia cemented by the vein-stuff. The fragments vary greatly in size, and are sometimes angular, while in other cases they are rounded and have clearly suffered much loss by attrition or solution, or both. Many fragments are quite unaltered, while others appear to be more or less thoroughly changed into a mass of alteration products, containing strings and scattered grains of calcite and galena. The alteration product is light grayish green and rather fine-grained. . . . Under the microscope, fragments that, to the unaided eye, appear thoroughly altered, show a large amount of material which, beyond crushing, shows no change. Perfectly fresh, angular or rounded fragments of the various minerals lie in calcite and galena, with no trace of corrosion or alteration, . . . but on the other hand, many grains of feldspar have been bleached and some of them have been partly replaced by calcite and galena. ... At first, the fissure, more or less blocked with breccia, was formed. Next, the breccia was cemented, perhaps first by chloritic alteration products but more probably direct by vein-stuff, cementa- tion being accompanied by some replacement. Then followed another movement along the fault, with the production of an open fissure, which, in the fourth and last stage, was filled by a further deposition of the vein minerals . . . while as stated above, there is good reason for believing that this process [replacement] has been a factor in the formation of a vein, judging from present data, it has played a very minor part. x Narrow, very fine-grained quartz bands are present in the veins, and appear to be largely iii the central part. The strike of the vein would carry it into limestone on the east which underlies the valley. Newland (’19, p. 140) states that 3,250,690 pounds of lead were produced from the Coal Hill vein, and that the average content of the furnace ore was 67 per cent. Beck (’42, p. 49) refers to the Jepson vein, a short distance from the Coal Hill vein towards the road leading from Rossie to Oxbow, which has a hearing of S. 88° E. and has produced a large quantity of ore. The Victoria vein is three-tenths of a mile northwest of Rapids School. It is in quartz diorite and strikes N. 8o° W. The vein has been trenched at the surface for 300 yards, and splits into two veins at the east end. The galena is disseminated through the cal- cite gangue, whereas the sphalerite occurs in narrow seams. The HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 205 sphalerite is locally broken by calcite veins. There is sparse pyrite and fluorite, and a little very fine-grained quartz banding. New- land (Cushing and Newland ’25, p. 84) states that this vein was worked quite extensively, the vein having been followed to a depth of 300 feet, and a mill and smelter erected for treating the output. The old sandstone chimney still stands here. Another vein lies one and four-fifths miles southwest of Rossie and one and one-half miles west of Rapids School. It strikes N. 70° E., and has been stoped for a length of about 30 yards. The vein occurs wholly within the granite, and cuts across its foliation. About four-fifths of a mile west of Rapids School is a vein with a strike of N. 6o° W., that has been trenched for a length of 135 yards. It lies within the quartz diorite, although its line of strike projected to the east would carry it into limestone. Half a mile north of Robb School, a short distance east and west of the road respectively, there is a narrow galena vein, each striking about N. 6o° W. These veins are in the limestone. Along the same line of strike as the Rossie group of veins, but three miles to the east of Rapids School, four-fifths of a mile south- west of Laidlaw School, there is a narrow galena-calcite vein striking N. 60 0 W., and crosscutting an impure crystalline limestone. The vein is only about six inches wide. An adit has been driven about 20 yards on the vein. Leo Mullin of Rossie states that about a quarter of a mile west of Rapids School there are narrow sphalerite veinlets with a north- west strike in the rock. Macomb group. Near Pierce’s Corners, or Macomb, there is a group of veins that lie within the Grenville limestone and cross the strike of the bedding or foliation at a high angle. For the most part these veins strike N. 70° to 750 W. and dip at high angles. The best known ones are on the Pennock, Jones and Gene Downing farms. The vein on the Pennock farm is located half a mile south of Macomb. Two vertical shafts, 150 yards apart, have been sunk on the vein, and a drift is reported to connect them. The vein strikes N. 70° W. and apparently cuts across both an anticlinal and syn- clinal axis in the steeply dipping limestones. The fractured zone is said to be about 15 feet wide. About 200 yards west of the west shaft, across a meadow on a small ridge, a prospect trench exposes a vein apparently in line of strike with that developed by the shafts. A little dolomite is present in the gangue matter of this vein. A 206 NEW YORK STATE MUSEUM quarter of a mile to the east is a vein that has been traced for over too feet by prospect pits. It strikes about west. An adit has been driven 30 feet on the vein. The vein is wide, but there appears to be only a little sulphide mineralization. The vein on the Jones farm is located half a mile east-southeast of Macomb at the end of a private road. The foundations of an abandoned mill easily identify the locality. The vein strikes N. 70° W., and may be traced for about 125 yards. There is a shaft 50 feet or more deep, filled with water. The vein here is two feet thick, consisting of calcite with disseminated galena near the borders. There is a little sphalerite and green and purple fluorite. The walls are slickensided. There is a little disseminated pyrite in the country rock. There is little to be seen on the Downing farm, which is located four-fifths of a mile south of Macomb. A shaft 25 to 30 feet deep has been sunk on the vein. Dolomite is conspicuous in the gangue of the vein matter thrown on the dump. The vein strikes N. 75 0 W. Bigelow School group. South of Rollway bay on Black lake there is a vein cutting across a granite lens in limestone. It consists of several stringers in a zone several feet wide, and strikes N. 6o° W. This is probably the vein known as the Mineral Point. North- east of Bigelow School, just back of the house of Fred Turner, a vein in limestone has been stripped for a length of a hundred feet. The vein is about two feet wide and strikes N. 55 0 W. About nine- tenths of a mile east of Bigelow School is another vein striking N. 6o° W., in limestone. The vein has been stoped for a length of a couple of hundred feet. The ruins of an old mine building are still present here. Other veins. About nine-tenths of a mile west of Brasie Cor- ners there is a five-inch vein of galena-bearing calcite crossing quartz diorite. The vein strikes N. 70° W. W. L. Cumings of Bethlehem, Pa., has informed the writer that about one and one-fifth miles north of Rossie, east of the road and just north of Bostwick creek, there is an old prospect pit, and that a specimen from the weathered outcrop assayed 12 per cent zinc and 4 per cent copper, mostly in the form of carbonates. He states that it seems to accompany a quartz vein. Newland (’21, p. 142) has described a vein which is on the adjoining Alexandria quadrangle, but which is of sufficient interest to quote here : An interesting occurrence of galena is found about three-quarters of a mile north of the village of Redwood, town of Alexandria. It HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 20 7 is exposed on the sides of a low cut along the railroad north of the Redwood station. The outcrop of the vein is in Potsdam sandstone which forms a knob with an exposed thickness of from 35 to 40 feet. The sandstone along the sides of the vein is much fractured and shows vertical jointing. The main vein is about 4 feet wide and contains 2 feet of solid calcite in the middle with a foot or so of brecciated sandstone cemented by calcite on either side. The strike is N. 550 W. The vein can be traced from the railroad cut to the southeast across the adjoining field to a little ravine and west of the cut for about 200 feet over the top of the hill. It has been opened in several places. The vein does not carry any large amount of galena. It is apparent from the field relations that the sandstone overlies Grenville limestone as the latter is exposed in considerable force just north of the sandstone knob. The limestone itself is characterized by a small amount of galena in disseminated grains and crystals. Origin and comparison with similar deposits in Ontario. The origin of the vein-forming solutions is not wholly clear. Smyth (’03) discussed the problem and concluded that the veins were filled long after the cessation of igneous activity in the region, so that it is not possible to look to the granitic intrusives for a source of the solutions. He refers to the vein in the Potsdam sandstone north of Redwood, and considers it not improbable that this vein and the Rossie veins belong to the same period of mineralization. Smyth suggests that the vein minerals came from below, and that perhaps they were derived from the Grenville limestones by circulating water and deposited during their ascent through the fissures. To the writer, the lack of chemical alteration and minor amount of replacement of the wall rock, the breccia character of some of the veins, the occurrence of seams of very fine-grained quartz, the rudely banded or crustified character locally in some of the veins, and the very low silver content of the galena, all suggest that the deposits were formed in open fractures by relatively cold solutions, chemically inactive. The strike and nature of the fissures definitely allies them with fault fractures which have afifected Ordovician beds throughout southeastern Ontario. A vein of post-Ordovician age in the northwest Adirondacks has also been described by Beck (’42, p. 47) as follows: Near the village of Martinsburgh, galena is found associated with iron pyrites in narrow veins, traversing the Trenton limestone. At one of these, about three thousand dollars have been expended in excavations, which extend for nearly two hundred feet along the surface, and at one point are fifty feet in depth. 208 NEW YORK STATE MUSEUM Similar veins are found in the Precambrian and Paleozoic beds north of the St Lawrence in Ontario. Wilson (’24) has described one of the major deposits — the Kingdon — and the following notes are abstracted from his report. The Kingdon deposit lies five miles east of the town of Arnprior in Ontario. The principal vein occupies a fissure that cuts across Grenville limestone, diorite, granite, and pegmatite, with a N. 50° W. strike and an 80-degree dip. It has been followed by underground workings for 1350 feet and to a depth of 525 feet, with an average width of five feet. The vein matter consists chiefly of calcite inter- laminated with galena or, especially along the wall, with galena and sphalerite. Fluorite, barite, selenite and hematite are present but relatively uncommon. The proportion of galena varies greatly, but averages 5 per cent. The veins occupy fault fissures, as is demon- strated in the case of the north vein where overlying Beekmantown dolomite has been down faulted into the Grenville along the line now occupied by the vein. Wilson discusses two hypotheses for the origin of the veins, first, that they have been derived from the Paleozoic sediments through the agency of the groundwater circu- lation; and second, that they have been brought up from a deep- seated magmatic source by ascending heated waters, perhaps asso- ciated with deep-seated Devonian intrusives. He concludes that a definite decision between these alternative hypotheses is not war- ranted because of the conflicting character of the evidence. Uglow (T6, p. 17-18) has described a similar vein located about 18 miles north of Kingston, Ontario. The vein fills a fault fissure in the Precambrian, strikes N. 30° to 70° W., dips 750 to 8o°, and is composed of calcite with galena and a little sphalerite and minor quantities of pyrite. There is a well crustified structure. The vein has been definitely traced for about 1100 feet for each of two sec- tions, and may be a mile long. Shafts are sunk from 90 to 270 feet in depth. Many other veins of similar character and WNW strike, in southeastern Ontario, are described in the same report. Uglow discusses the probable origin, and concludes that their char- acteristics suggest an origin by waters of purely meteoric origin. On further reflection, however, in view of the fact that several faults in the vicinity of Montreal and Ottawa show a relative ver- tical displacement of upwards of 2000 feet, and that the fault fissures might extend to depths where they would tap the circulation of hot solutions carrying materials obtained ultimately from magmatic waters, he concluded that the vein material represents a combina- tion of material derived from both meteoric and magmatic waters. HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 209 History. A brief history of the development of these deposits is given by Durant and Pierce (’78) from which the following notes are taken. Galena was first discovered near Rossie, and work was com- menced in 1836. In the same year a vein of lead, zinc blende, and spar was discovered near the shore of Black lake, at a place named Mineral Point. A shaft was sunk 76 feet which yielded 1100 pounds of lead. In the branch of the Union vein two shafts were sunk, the western 55 feet and the eastern 50 feet in depth. The average yield of ore was 67 per cent or 10 >4 tons a fathom. Work was discontinued in 1840, and resumed between 1852 and 1855, but ceased because the lessees were unable to pay the one-twelfth royalty demanded by Mr Parish. They were again worked between 1864 and 1868. Smyth (’03, p. 429) states: The Rossie mines were worked during the Civil War by the Min- eral Point Lead Mining Company, which also had mines at Mineral Point, on the south shore of Black Lake. At that time the high price of lead had started up at least two other mines in the region, one near the village of Macomb, and the other in the town of Gou- verneur, near Beaver Creek. It is probable that no mining has been done since the close of the war. Smyth also states that “The Coal Hill vein near Rossie village produced 1625 tons of metallic lead in the years 1837 and 1838, according to the books of the smelting company.” Newland (’25, p. 84) writes that “The Victoria or Pardee mine, in the same vicinity, also was worked quite extensively, the vein having been followed to a depth of 300 feet and a mill and smelter having been erected for treating the output.” PYRITE AND PYRRHOTITE DEPOSITS Introduction. Concentrations of pyrite alone, or of pyrite in association with pyrrhotite, are found at a number of localities in the Hammond and Antwerp quadrangles. They are in large part confined to the inner border zone of the garnet-biotite gneiss belt where it adjoins the limestone around the great Somerville over- turned isoclinal anticline, as at the Keene, Morgan, Caledonia, Wight, Dickson and Laidlaw prospects. A narrow band of gneiss about a mile long, with pyritic concentrations, lies intercalated in limestone about two to three miles southwest of Oxbow. Another pyritic concentration is found about half a mile west of Grass lake south of the County line. These deposits have been previously 210 NEW YORK STATE MUSEUM described by the author (’17). and the following notes are largely taken from this report. Further details of specific prospects will be found in that publication. Pyrite deposits of similar character have been mined at the Stella mines (Hermon quadrangle), Pyrites (Canton quadrangles), and Cole mine (Gouverneur sheet) ; but following the World War, the mines shut down, and there is no production at the present time. Further references to pyrite in the northwestern Adirondacks may be found in publications by Brinsmade (’06), Smyth (’12), New- land (’17), Vogel (’18), and Miller (’26). Country rock. There are many varieties of gneisses in the Gren- ville series of this region, but the pyritic deposits are practically con- fined to one type. This type is the biotitic quartzose gneiss which is a mixed-rock or injection gneiss, consisting of recrystallized argillaceous and quartzose sediments injected by granitic pegmatite veinlets and permeated by minerals deposited by the granitic solu- tions. The predominant minerals are quartz, feldspar, a little bio- tite, occasionally garnet, rarely sillimanite, and the usual minor accessory minerals, such as apatite, zircon and magnetite. Structure and character of veins. The pyrite and pyrrhotite deposits represent zones or lenses, within which there is a relatively high concentration of iron sulphides as compared with the belt of gneiss with disseminated sulphides in which they occur and into which they grade. They are of a tabular or lenticular character, and conform in their orientation to that of the foliation of the inclos- ing chloritic schist and gneiss (figure 18). They are of replacement origin, and have the gross structural relations of “bedded veins.” A number of similar deposits are found throughout the Grenville belt of the Northwest Adirondacks. All the data indicate that the stronger veins, although they pinch and swell, are in general quite persistent both along the strike and down the dip. The deposits at the Stella mines in the town of Hermon were formerly worked extensively. The Stella vein was developed for a length of 1200 feet, and the Anna veins were mined for a length of 1800 feet and are known to have a total length of half a mile. The veins at the Anna shaft were opened up to a depth of about 600 feet down the dip, and the Stella vein was mined 900 feet down the dip, without showing any change in character. The Stella vein averaged 10 to 12 feet thick; and two veins worked from the Anna shaft, 18 feet each. The veins in the Keene-Antwerp belt average 10 to 20 feet in width. The pyrite veins locally show sharp flexures, pinches and HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 2 1 1 crenulations. There seems to be some local pyrite enrichment in such zones of disturbance. The pyrite veins show two rather different types : the most usual is that in which the pyrite is distributed more or less uniformly through the gangue as veinlets, disseminations, or both ; and the other, a sheeted type, in which thin layers of rich pyritic schist alternate with leaner bands. The typical ore consists of fine granular pyrite disseminated quite uniformly through the gangue, and of medium granular pyrite likewise disseminated through the gangue or bunched together in small irregular clot-shaped pockets or joined by a close woven network of veinlets. The concentrations at the Keene, Morgan, Laidlaw properties and the veins three-quarters of a mile northeast of the Bent prospect are of this type. In the veins of the southwestern half of the Keene- Antwerp belt the pyrite occurs in a system of veinlets of coarse character, more sharply differen- tiated from the gangue than usual and to a large extent are of coarser grain. Crystals up to an inch or more are common, although, as usual, there is a considerable pyrite of fine grain. At the Laid- law prospect seven feet of pyrrhotite occurs in the footwall of the pyrite. The gangue is a chloritized, partly replaced, facies of the gneiss, consisting of quartz, feldspar, chlorite, mica and graphite, the latter mineral finely disseminated in small amounts. The rusty gneisses which are quite common often carry 5 per cent or more of pyrite. The average sulphur content of the ore as mined at the Stella mines (Hermon sheet) was about 20 per cent; and the lump ore as shipped from the Cole Mine (Gouverneur quadrangle) contained 30 per cent or more of sulphur. The veins of sulphide concentration vary from 15 to 40 per cent sulphur, and as a rule the richer ore is asso- ciated with an increase of graphite and a greater chloritization. Surface outcrop of veins. Pyrite and pyrrhotite break down auite rapidly, on exposure to weathering processes, into iron hydrates or limonite, which gives the yellow and reddish brown colors of iron rust to the outcrop, and into solutions of iron sulphates with some free sulphuric acid, which in part pass away in the ground water. The acid solutions attack the silicate minerals of the rock and, together with the decomposition and removal of the sulphides, destroy the cohesion of the rock and cause its disintegration in situ. Most of the gneiss with disseminated sulphides near the surface is so weathered that it may be crumbled in the fingers to a granular aggregate. 212 NEW YORK STATE MUSEUM The heavy mineralized zones ar usually weathered to a depth of several feet ; and only on a shoulder or knob where erosion has proceeded faster than decomposition, or beneath a surface which has been protected by glacial polish, can fresh specimens be obtained. It is therefore to be expected that a considerable portion of the pyritic zones will underlie the valleys. The gossan of the surface outcrop is a cellular, porous, spongelike mass of siliceous material with little, if any, sulphide remaining, and usually colored a reddish brown or buff by the iron hydrates. Fresh erosion surfaces of the heavy mineralized zones — bands with a sheeted structure excepted— usually present a more massive homogeneous appearance than do their wall rocks, showing a coarse spheroidal weathering or rounded knobs. This more massive character is due to the greater homo- geneity of the veins resulting from the uniform alteration and replacement of the gneiss with more or less complete obliteration of the foliated structure. Locally, the glacial drift overlying the sulphide zones has been cemented with limonite, deposited from solutions arising from the weathered veins. Keene-Antwerp belt. There are five localities where pyrite deposits are exposed along this belt, which is five and one-half miles in length. The Caledonia is just east of the Old Iron Works in the town of Rossie ; the Keene is about one-third of a mile west of Keene’s Station (Hammond quadrangle) ; the Morgan is about one and one-half miles southwest of Keene’s Station ; the Wight is on the Antwerp quadrangle about two miles north of the village of Antwerp; and the Dickson is about one and one-half miles north of Antwerp. Hematite or iron ore beds occur in association with the pyrite, and the latter in most cases has been exposed incidental to the search for and extraction of the iron ore. A carload of pyrite ore was shipped from a prospect in this belt during the World War. The exposures of the veins are all situated on outcropping knobs of rock above the valley levels. The trend of the veins usually quickly carries them beneath lowlands filled with stratified drift or beneath remnants of Potsdam sandstone. The continuity in the length of the veins of this character, as demonstrated at the Stella mines and at Pyrites, suggests that the veins of this belt may well have greater extension than that actually exposed. The veins occur in a narrow belt of greenish schists lying between the normal bluish gray gneiss to the southeast and the crystalline limestone to the northwest. In general, they appear to be restricted to a zone close HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 213 to the contact with the limestone, and at the Morgan prospect lime- stone lies 350 feet above the pyrite vein. The schists in which the pyrite concentrations occur are for the most part of a green chlori- tic nature, the result of hydrothermal alteration and replacement of the gneiss. Locally, parts of the schist have been altered to, and replaced by, beds of hematite and quartz-hematite, particularly near to or beneath the Potsdam sandstone caps. The junction of the schists and gneisses is usually well marked by a line of hills formed by the more resistant gneiss, whereas the schists are at lower elevations. Between the Keene and Caledonia prospects the pyritic schist band probably lies beneath the Potsdam sandstone for most of the way. Between the Keene and Wight prospects it lies partly under sandstone and partly under alluvium. A meadow overlies the zone between the Wight and Old Sterling mines, and between the Wight and Dickson prospects. At only one locality are both the footwall and hanging wall of the veins exposed. This is at the Wight vein, where the thickness is from 18 to 20 feet. At the Keene, Morgan and Dickson veins, 9 to 10 feet, 8 feet, and 15 feet thickness respectively are exposed. The total width of these veins must be greater. The percentage of sulphur in grab samples taken at the Morgan and Wight veins ran 23.24 and 28.3 per cent respectively. The vein matter probably averages 20 to 25 per cent sulphur at these deposits. Laidlaw belt. The Laidlaw belt lies along the east side of the Sherman Lake granite mass on the Hammond quadrangle, south- east of Oxbow. Only two veins are exposed in the belt. The north- ern, or Laidlaw vein, exposes at its maximum width 20 feet of pyritized gneiss overlying seven to ten feet of pyrrhotite-bearing rock. The whole would probably average 20 per cent or better in sulphur. The vein can be traced for 200 yards. Another vein half a mile southeast is located near the junction of the gneiss and lime- stone. Seven feet of average pyritic schist is exposed. Oxbow Southwest belt. There are two veins exposed in a nar- row band of gneiss in limestone. The northern vein is six feet thick as exposed, and can be traced for 500 yards along the strike. The other vein (Bent vein) is two and one-half miles southwest of Oxbow. The exposure shows 25 feet across the strike of the vein consisting of medium granular pyrite in alternating leaner and richer bands. The vein can be traced for 400 feet. A grab sample yielded 22 per cent of sulphur. 214 NEW YORK STATE MUSEUM Origin. The origin of these deposits has been discussed in detail by Smyth (’12), and the writer is in agreement with his con- clusions, a summary of which follows. The veins showing relatively large concentration of pyrite are for the most part the result of replacement of sheared zones in the Grenville gneiss by magmatic solutions rich in hydrogen sulphide and iron. The sulphide ores were formed at a late period in the history of igneous activity, after the pegmatitic injection, and the source is ascribed to fluids given off by the consolidating granitic magmas. The formation of the chloritic alteration product, uni- formly associated with the gangue of the veins, is also ascribed to these solutions. The graphite of the veins represents in part a con- centration of organic material originally present in the sediments from which the Grenville gneiss was derived, and in part new material contributed directly by the magmatic solutions. The depo- sition of the sulphide minerals took place on cooling, when the necessary lower temperatures and pressures were reached, nearer, but still at a considerable depth beneath the surface. The deposition was probably genetically connected with complex chemical reactions involving the precipitation, in part simultaneously, of both graphite and pyrite. Smyth also assumes that there probably had been a primary precipitation of pyrite contemporaneously with the forma- tion of the sediments, a concentration of this pyrite by circulating ground waters, and a concentration by circulation of magmatic fluids; but that the major source of the pyrite was in the magmatic solutions. Miller (’26) studied the pyrite deposits at the Stella mines and concluded that the main source of the pyrite was the pyrite originally disseminated through the gneiss, and that it was simply taken into solution by the penetrating granitic and pegmatitic liquids or vapors, carried to higher levels, and there deposited in more concentrated form. Miller offers no new data, but places a relatively greater emphasis on a factor already postulated and discussed by Smyth. GOLD SANDS Repeated attempts have been made to obtain gold from the sand and gravel deposits that occur widely distributed throughout the Adirondacks. The widespread sand and gravel deposits of the Lowville quadrangle have naturally attracted the attention of pro- spectors in search of the precious metal. Two mills which were erected primarily to recover the gold, and other heavy metals if present, were noted in this quadrangle by the writer. One is HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 215 located where Sandy creek crosses the road about a mile north of Kirschnerville. The other is situated at the spot marked on the map by crossed hammers about three miles northeast of Kirschner- ville. It is reported locally that prospecting has been carried on at other localities. The author made no personal investigation as to the presence or absence of the precious metals in these sands as the general subject has been repeatedly touched upon in various reports of the New York State Geological Survey. Quotations of pertinent character from two of these articles follow (Newland, To, p. 16-20) : The basis of all this activity is the alluvial and glacial sands which occur in almost every stream valley. These sands are not derived from remote regions to the north though the opinion seems common that they have been transported by ice from Canada and even from as far away as Alaska — but are the result of erosive agents working upon the local rocks. They consist entirely of the minerals of the country formations which are chiefly granites, syenites, gabbros, and gneissoid rocks of very ancient (Precambrian) origin. Along with the lighter components, quartz and feldspar, there is a small proportion of heavy minerals like garnet, pyroxene, hornblende and magnetite which have sometimes been separated by water action into distinct layers and which are found as black sands along the shores of the Adirondack lakes. It is in these heavier concentrated portions of the sands that gold should be found if present anywhere in the region. Assays by reputable firms have sometimes shown a small quantity of gold, ranging from a mere trace to perhaps $1 a ton. The examination of innumerable samples under the microscope has failed to reveal any of the precious metals though, of course, such evidence is not conclusive as to their absolute presence or absence. On the other hand the claim has often been put forth that the common quartz sands in certain places carry from $2 to $20, which, if true, would under ordinary circumstances bring them into the zone of practical exploitation. The statement is commonly made by interested parties that the fire assay is unsuited to the determination of gold in these sands and that special analytical methods must be employed. To this it can only be said that the fire assay has stood the test of long prac- tice in all the mining regions of the world. No reasonable explana- tion for its failure in the present instance has been forthcoming, though by the use of such vague terms as “volatile gold,” “atomic condition,” “chemically combined gold,” and “gel gold,” there seems to be an implication that the gold is driven ofif or left untouched by the fire. As a rule the gold found by fire assay is several per cent more than the amount extractable by the most refined commer- cial processes. For the analyses of sands from Lewis county, which is the scene of present activity in mining, we are indebted to the Engineering 2l6 NEW YORK STATE MUSEUM and Mining Journal (March 19, 1910) through whose enterprise samples were recently collected and assayed. The samples were taken by B. J. Hatmaker who had previously experimented with sands from the same localities. The following particulars are from Mr Hatmaker’s letter transmitting them : The samples marked “A” are from an immense deposit along the Black river and represent three samples taken 300 feet apart. These samples gave me by fire, from $3.59 to $3.80 per ton. The samples marked “B” are from a deposit back in the hills which should run around $3. This particular sample was taken by Professor Locke of the Boston Institute of Technology, and myself. It represents the sand of which Dr N. S. Keith, of Philadelphia, has milled several tons and has reported $2.50 to $3 recovery, by amalgama- tion. My fire assays in this have run $1.50 and $2.75. Professor Locke was unable to get more than a trace. The report on the results of assays by the firm of Ricketts and Banks, as printed in The Engineering and Mining Journal, is as follows : The samples of sand marked “A” and “B,” received sealed under signa- ture of B. J. Hatmaker, submitted for assay contain : A B Fire assay : 0.005 oz. 0.005 oz. Wet assay 0.005 oz. 0.005 oz. gold per ton of 2000 pounds. Additional samples marked “A” and “B” were also submitted by The Engineering and Mining Journal to the firm of A. R. Ledoux & Co. who made the following report : The two samples of sand submitted to us on February 1, 1910, marked respectively “A” and “B,” and sealed with paper bands, bearing the signa- ture of B. J. Hatmaker, have been assayed by the usual fire assay method, yielding : “A” — Gold=o.oo25 oz. per ton=$o.os per ton. “B” — Gold=o.0O5 oz. per ton==$o.io per ton. This work was very carefully done, using large assay charges. In view of the statement that these sands are said to contain gold combined with some element, or elements, causing the gold to volatilize during the fire assay pro- cess, and that this method is not capable of detecting gold in these sands, we have repeated the assays by a wet method which involves digestion of the finely ground sands with aqua regia at a low temperature for a long time, filtering off the acid liquid, evaporating it to a small bulk and examining the concentrated solution for gold. By this method we obtained: In sample “A” — gold, trace In sample “B” — gold 0.003 oz. Per ton==$o.o6 per ton Supplementing these tests, a portion of each sample was concentrated by panning and the concentrates were examined both with a hand glass and also microscopically. Neither sample showed the presence of any visible gold or of any usual mineral or substance which might possibly carry gold. The con- centrates are principally magnetic iron particles mixed with some complex silicates of the garnet family. Portions of each sample contained in closed tubes of hard glass were heated in a blast lamp flame to the melting point of the glass. A quantity of com- bined water condensed on the cool parts of each tube but neither sample yielded any sublimate of volatile matter whatever. HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 217 From the above tests we conclude that these samples are ordinary siliceous sands and that they contain only traces of gold as are usually found in such sands. Traces of gold are frequently present in many rocks and sands, and it is not unusual to find gold values equivalent to a few cents per ton in ordinary rocks, such for instance, as granite paving blocks. These samples do not contain any extraordinary or unusual element or any substance which would cause the gold to volatilize in the ordinary process of assaying, nor in fact do they contain any volatile substance except combined water. The economic record of past enterprise in this field is certainly not reassuring to those intent on new ventures. Though it is impos- sible to give an accurate estimate of the outlay of capital represented by previous experiments, the total must amount to several hundred thousand dollars. An idea of the wide interest which the early enterprises aroused may be gained from the official records which show that over 4000 claims to gold and silver discoveries mainly within the Adirondacks, were filed in the year 1898. We know of no instance where the public has received any financial return for its investment. GRAPHITE Graphite is quite common as disseminated flakes in the Grenville limestone and gneisses, particularly the pyritic varieties of the latter. It often constitutes 2 or 3 per cent of the gneisses. It locally forms a few per cent of a bed of gneiss in the band at the pyrite prospect two miles southwest of Oxbow, and deposits have been prospected at localities one and one-half miles southeast of Pope Mills and three miles south of Rossie. A brief reference is made to the latter locality by Newland (’08, p. 30), which is quoted here: In St Lawrence county some attention has been given to a deposit occurring on the Indian River about 3 miles (probably south) from Rossie village. The graphite forms the principal constituent of a schist, through the body of which it is distributed richly in very small scaly particles. It is a crystalline graphite, but too fine in size to be easily separated. Trial shipments of the crude material were reported to have given satisfactory results when used for foundry purposes. The graphite deposit near Pope Mills has been described by Mills (’08, p. 396), from whom the following quotations are taken: The cut shows a finely laminated, graphitic quartz schist, com- plexly foliated and corrugated. The laminated deposit is crumpled, friable, quartzose, and contains a small proportion of iron (Pyrite). The property has been worked to a limited extent by a small com- pany under the name of the Macomb Graphite Company of Canton, N. Y. A section about 75 feet long and nearly 15 feet deep was cut into the face of one of the folds. The structure is uniform and rich in crystalline graphite. The company mined about 100 tons 2l8 NEW YORK STATE MUSEUM of the rock for experimental purposes. The deposit seems to grow richer with depth. Although the milling was done with rather crude equipment, the yield was from 15 to 20 per cent of graphite. Sev- eral tons of excellent concentrates were produced, averaging more than 90 per cent carbon. This product was distributed to various manufacturers, to determine its practical value for lubrication, foundry work, metallic paint, and other uses. The reports were highly satisfactory. For lack of capital the company has tem- porarily ceased operations. Ailing (’17, p. 1 12) says of this property: It would appear that this deposit is of very fine crystalline form that usually is referred to as amorphous. The material from the Macomb locality is well suited for certain forms of lubrication, foundry work, etc., but it is not crystalline enough to be used in the manufacture of crucibles. He gives the average diameter of the graphite flakes as 0.145 mm by 0.202 mm, with a minimum of 0.09 mm and a maximum of 0.42 mm. The graphite schist at this locality contains calcite, quartz, feldspar and a little apatite, with sparse pyrite as gangue mineral. It strikes about N. 250 E., dips steep to vertical, and is associated with quartzitic beds. A width of 12 feet is exposed in the face of cut. CRYSTALLINE LIMESTONE, DOLOMITE AND MARBLE The relatively clean limestone of the Grenville belts is a source for building and monumental stone, furnace flux, crushed stone, and for lime burning. Relics of old lime kilns used in the early days are found at several localities within the Hammond quad- rangle. A marble quarry, the Rylestone, not operated at present, is located on a low ridge nearly due west of Gouverneur and south of Natural Dam. The marble is bluish gray with a mixture of white and blue calcite. Much of the rock carries disseminated blades of tremolite. There is a little scapolite and, in local beds, pegmatite nodules with brown tourmaline and tremolite-tourmaline nodules. The limestone of the Gouverneur-Somerville belt is extensively quarried on the Gouverneur quadrangle ; and as the rock is similar throughout, analyses of rock from that quadrangle (Newland, ’25, p. 120) are included in the table below, together with partial analyses by Agar (’23, p. 132). HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 219 Analyses of crystalline limestones 1 2 3 4 5 6 7 8 Insoluble S1O2 3-55 1.26 1 .01 1.58 } -79 3-49 51-45 A12O3 •13 .08 •65 .29 • 23 ■ 63 2.38 5236 4.12 51-02 2.49 56.45 3- 80 52.24 Fe203 MgO MgCOj CaO 6.40 7-50 6.85 CaC03 87.06 1.68 00 4- 4^ CS^-J 88.94 i-74 h2o C02 42.56 •03 S •05 .02 .04 1 St Lawrence Marble Quarries, R. W. Jones, analyst. 2 Quarry of Gouverneur Marble Company, R. W. Jones, analyst. 3 Rylestone quarry, R. W. Jones, analyst. 4 Northern New York quarry, R. W. Jones, analyst. 5 White marble from a limestone-chondrodite zone, two and one-half miles northwest of Oxbow (Hammond quadrangle), W. M. Agar, analyst. 6 Limestone far from mineralization, two miles north of Oxbow (Hammond quadrangle), W. M. Agar, analyst. 7 White recrystallized limestone amongst silicates, Reese farm, two miles southwest of Richville (Gouverneur quadrangle), W. M. Agar, analyst. 8 Limestone, Reese farm, two miles southwest of Richville (Gouverneur quadrangle), W. M. Agar, analyst. Newland (Cushing and Newland ’25, p. 118-19) makes the fol- lowing comments on the limestone : The limestone from the Gouverneur area is susceptible to high polish, which together with its attractive luster and good texture has given it considerable favor as a monumental stone. For building purposes it is mainly used in rock face ashlar, of which the color is medium gray; on cut or hammered surfaces the color is consider- ably lighter. The limestone in general is medium to coarse crystal- line and white or light gray in color, but sometimes is a dark blue. It is normally a calcite limestone with a varying, though small, per- centage of magnesia. The carbonates amount to about 95 per cent of the whole mass, so that it ranks as one of the purer limestones to be found in the State. Some belts of relatively pure limestones in the Hammond quad- rangle have been previously referred to in the chapter on Meta- morphic Rocks under the heading of “Crystalline limestones.” Dolomite occurs in the belt east of Lewisburg, and is extensively quarried just east of the border of the Antwerp sheet on the Lake Bonaparte quadrangle. Dolomite occurs at Peabody Bridge on 220 NEW YORK STATE MUSEUM the Gouverneur quadrangle, and may extend southwest along the line of strike on to the Hammond quadrangle. SANDSTONE The Potsdam sandstone was formerly quarried for building material and paving stone on a small scale at many points along the scarp paralleling the railroad in the town of Hammond. No quarry- ing is at present in progress, except for crushed stone for road metal. As the sandstone crumbles readily under impact, however, it is ill adapted to this purpose. It does, however, make excellent paving stones, and is an attractive stone for building, as exemplified by the Presbyterian Church at Hammond. A photograph of one of the old schoolhouses built of this stone is shown in figure 53. LIMESTONE The Pamelia, Lowville and Black River limestones have been used locally for structural purposes, for lime-burning and for road metal. The limestones serve very well for road metal, and quarry- ing is actively carried on for this purpose in the Pamelia and Low- ville limestone at Lowville. The Theresa sandy dolomite is also used locally for road metal, and is much better adapted for this pur- pose than the Potsdam sandstone. GRAVEL AND SAND Most of the recessional moraines, particularly those on the Ant- werp quadrangle, afiford excellent sources for gravel (figure 54). The Dicob, Crystal Dale, Croghan West, Beaver Falls, Carthage, Devoice, Fargo and Philadelphia moraines may be mentioned as examples. The moraine west of Deferiet, North Wilna and Woods Mill, however, is bouldery with slabs of the Paleozoic limestones, as are parts of the Fargo moraine. Sand is available in quantity in the delta deposits which have been previously described. On the Hammond quadrangle pits have been dug in weathered, disintegrated crystalline limestone, and the weathered rock locally beneath the caps of Potsdam sandstone, for road repairing. The large high level deltas, such as that along the east border of the Lowville quadrangle and the Pine Plains delta of the Ant- werp sheet, afiford excellent pure supplies of ground water. The water supply for the city of Carthage is obtained from a reservoir on the Indian river about two miles below its source. The river here is a beautiful, crystal-clear cold stream fed by underground HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 221 water draining from the sand plain. There are no agencies for pol- lution in the catchment area above the reservoir. Croghan also obtains its water from the ground water of the sand plain. MINERAL SPECIMEN LOCALITIES The highly heated solutions and vapors given off by the succes- sive magmatic intrusions introduced many new minerals into the Grenville limestones, some of which occur in well crystallized form. There are several localities on the Hammond quadrangle that are well known for the fine minerals they have furnished, which have found their way into the collections of many museums. Many of these mineral localities have been studied and described by Agar (’21, ’23). Among the best known are the following: Near Rossie, roughly one and one-half miles north along the road from Rossie to Hammond, about one-quarter mile southwest of the last house to the north of the road before it crosses Grass creek, there is a cliff face of diorite covered with beautiful, well-crystallized minerals. This face was formerly in contact with limestone which has since been eroded away. Apatite, pyroxene, feldspar, scapo- lite, titanite and pink calcite occur here. Near the northwest end of Grass lake, there are a few old pits sunk in the limestone in search of minerals. Fine apatite crystals are found here. There are many granite pegmatite veins faced with rounded quartz crystals at contact with the limestone, and a bed of garnet gneiss. Well- crystallized feldspar and pyroxene are also found here. About one and four-fifths miles northeast of Macomb, on the east side of the creek, there are a number of old pits sunk in search of sunstone. Many fine specimens still lie on the dumps. The sunstone occurs in the veins of calcite and oligoclase crossing pegmatite dikes in limestone. Quartz crystals are also common, locally throughout the Macomb belt of limestone, in bands accompanying quartz vein and pegmatitic replacements. In other bands tremolite occurs in good disseminated crystals. Brown tourmaline is also common, but in sparse disseminations. Flakes of black graphite and brown phlogopite mica are present almost everywhere as disseminations in the crystalline limestones. Good crystals of phlogopite have been obtained near Payne lake. Yellow rounded grains of chondrodite are abundant in many of the beds of crystalline limestone north of Oxbow, west of Yellow lake. Locally, beautiful little lilac colored spinels are found associated with the chondrodite. A great variety of minerals is found in an area of crystalline limestone of about one square mile, two and one-half miles northwest of Oxbow 222 NEW YORK STATE MUSEUM and north of the road just north of the north end of the Payne Lake granite mass. Diopside, phlogopite, missonite, wernerite, marialite, apatite, tremolite, chondrodite, pink and lilac spinel, micro- perthite, orthoclase, albite, black and brown tourmaline, pyrrhotite, titanite, quartz, graphite and serpentine replacing chondrodite, are all found in nodules or as disseminations in the limestone. Magni- ficent groups of sea-green fluorite crystals were obtained in the early days along the northeast side of Muskalonge lake (Hammond quadrangle) but only small crystals are found now along the shore. A. small quantity for commercial use was mined here. At the old Stirling iron mine north of Antwerp, crystals of specu- lar hematite, stilpnomelane, chalcodite, siderite, ankerite, millerite, quartz, aragonite, sphalerite and pyrite are found in vugs in the ore. Beautiful crystals of calcite, galena and pyrite were formerly obtained from the Rossie lead mines. Small quantities of disseminated molybdenite are found in the pegmatitic veins in granite outcropping on the farm of William J. Aucter, three-quarters of a mile southeast of Bushes Corners (Low- ville quadrangle). The granite is green to pink on fresh surfaces, weathering to a reddish hue. Four narrow molybdenite bearing peg- matite veins were noted, each scarcely more than an inch wide, and exposed for a length of 25 feet. The flakes of molybdenite are relatively coarse and will average from a quarter of an inch to one inch in diameter. Traces of molybdenite were also found in pegma- tite veins in the coarse red granite, one-half mile south of Strift’s schoolhouse, west of the highway. It is possible that larger molyb- denite bearing pegmatite veins may occur in this locality but the present known occurrences are interesting from a mineralogical viewpoint only, as they constitute a new locality for this mineral. A little molybdenite was also seen in a pegmatite vein in the Hickory Lake (Hammond quadrangle) granite mass. ADDENDUM After the completion of the foregoing report, the writer intensively restudied the Diana complex as a whole and made broader regional studies of Adirondack geology. From this have resulted additional significant data. Where the term “protoclastic” is used in this report in describing the structure of the Diana complex, the writer followed current prevailing practice with respect to the Adirondack syenitic rocks. His subsequent studies, however, suggested the probability that this structure may be of crystalloblastic origin, formed during deforma- HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 223 tion and recrystallization in the solid state, rather than of protoclastic origin. Stronger and more positive indications have also been found that the Croghan complex (Lowville granite and its facies) is intrusive into and crosscuts the banding of the Diana complex. BIBLIOGRAPHY Adams, F. D. & Barlow, A. E. 1910 Geology of the Haliburton and Bancroft Areas, Province of Ontario. Geol. Surv. of Canada, Mem. 6. 419P., 70 pis. Agar, W. M. 1921 The Minerals of St Lawrence, Jefferson, and Lewis Counties, New York. Amer. Min., v. 6, no. 10:148-53; 11:158-64 1923 Contact Metamorphism in the Western Adirondacks. Amer. Phil. Soc., Proc., v. 62, no. 3 :95~I74 Ailing, H. L. 1917 The Adirondack Graphite Deposits. N. Y. State Mus. Bull. 199. 1 SOP- 1919 Some Problems of the Adirondack Pre-Cambrian. Amer. Jour. Sci., 4th ser., 48:47-68 1924 The Origin of the Foliation and Naming of Syntectic Rocks. Amer. Jour. Sci., 8:12-32 1925 Geology of the Ausable Quadrangle. N. Y. State Mus. Bull. 261. i26p., map 1927 Stratigraphy of the Grenville of the Eastern Adirondacks. Bull. Geol. Soc. Amer., v. 38, no. 4:795-804 Baker, M. B. 1916 The Geology of Kingston and Vicinity. Ann. Rep’t, Ontario Bureau of Mines, v. 25, pt 3. 36p. 1923 Geology and Minerals of the County of Leeds. Ontario Bureau of Mines, v. 31, pt 6. 26p. Balk, Robert 1930 Structural Survey of the Adirondack Anorthosite. Jour. Geol., 38 ^89-302 Beck, L. C. 1842 Mineralogy of New York. xxiv+ 536p., plates Bowen, N. L. 1917 The Problem of the Anorthosites. Jour. Geol., 25 :209-43 1928 The Evolution of the Igneous Rocks. 332p. Princeton Univ. Press Brinsmade, R. B. 1906 Pyrites Mining and Milling in St Lawrence County, New York. The Mineral Industry, 14:525-27 Brooks, T. B. 1872 On Certain Lower Silurian Rocks in St Lawrence County, New York. Amer. Jour. Sci. (3), 4:22-26 Buddington, A. F. 1917 Report on the Pyrite and Pyrrhotite Veins in Jefferson and St Law- rence Counties, New York. N. Y. State Defense Council, Bull, no. 1. 4op. 1919 Foliation of the Gneissoid Syenite-Granite Complex of Lewis County, New York. 14th Rep’t Director, N. Y. State Mus., p. 101-10 1929a Granite Phacoliths and Their Contact Zones in the Northwest Adirondacks. N. Y. State Mus. Bull. 281 :5i-io7 224 NEW YORK STATE MUSEUM 1929 b Geology and Mineral Deposits of Southeastern Alaska. U. S. Geol. Surv. Bull. 8oop. 1931 The Adirondack Magmatic Stem. Jour. Geol., 39:240-63 Burling, L. D. & Kindle, E. M. 1915 Structural Relations of the Pre-Cambrian and Paleozoic Rocks North of the Ottawa and St Lawrence Valleys. Geol. Surv. of Canada, Mus. Bull. 18. 23p., maps Chadwick, G. H. 1920 Paleozoic Rocks of the Canton Quadrangle. N. Y. State Mus. Bull. 217-18 Clark, T. H. 1919 A Section in the Trenton Limestone at Martinsburg, New York. Bull. Mus. Comp. Zool. at Harvard College, v. 63, no. 1. i8p., 1 pi. Crosby, W. O. 1902 Geological History of the Hematite Iron Ores of the Antwerp and Fowler Belt in New York. Amer. Geol., 29:233-42 Cushing, H. P. 1899 Augite-syenite Gneiss near Loon Lake, New York. Bull. Geol. Soc. Amer. 10:177-92 1905 Geology of the Northern Adirondack Region. N. Y. State Mus. Bull. 95. 453P- 1915 Age of the Igneous Rocks of the Adirondack Region. Amer. Jour. Sci., 4th ser., 39:288-94 1916 Geology of the Vicinity of Ogdensburg, [New York] (Brier Hill, Ogdensburg and Red Mills Quadrangles). N. Y. State Mus. Bull. 191. 649., map Cushing, H. P., Fairchild, H. L., Ruedemann, R. & Smyth, C. H. jr 1910 Geology of the Thousand Islands Region. N. Y. State Mus. Bull. 145. I94p., maps Cushing, H. P. & Newland, D. H. 1925 Geology of the Gouverneur Quadrangle. N. Y. State Mus. Bull. 259. I22p., map Daly, R. A. 1917 Metamorphism and Its Phases. Bull. Geol. Soc. Amer., 28 :375-4i8 Durant, S. W. & Pierce, H. W. 1878 History of St Lawrence County, New York. 5219. Emmons, Ebenezer 1842 Geology of New York, pt 2 (Surv. 2d Geol. Dist.) p. 335-427 Fairchild, H. L. 1912 The Glacial Waters in the Black and Mohawk Valleys. N. Y. State Mus. Bull. 160. 47p., maps 1919 Pleistocene Marine Submergence of the Hudson, Champlain, and St Lawrence Valleys. N. Y. State Mus. Bull. 209-10. 76p., maps Flint, R. F. 1929 The Stagnation and Dissipation of the Last Ice Sheet. Geog. Rev., 19:256-89 Gillson, J. L., Callahan, W. H. & Millar, W. B. 1928 Adirondack Studies : The Age of Certain of the Adirondack Gabbros, and the Origin of the Reaction Rims and Peculiar Border Phases Found in Them. Jour. Geol., 36:149-63 Goldring, Winifred 1922 The Champlain Sea. N. Y. State Mus. Bull. 239-40 :i53-94 HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 225 Goldschmidt, V. M. 1916 Geologische-Petrographische studien in Hochgebirge des Sudlichen Norwegens, Vid. Sk. I Mat.-Naturv. Klasse, no. 2. I40p. 1922 Stammestypen der Eruptivgesteine, Vid. Sk. I, Mat.-Naturv. Klasse, no. 10. I2p, Harker, Alfred 1909 The Natural History of Igneous Rocks. 384P. Hinds, N. E. A. 1925 Amphitheater Valley Heads. Jour. Geol., 33:816-18 Ingall, E. D. 1899 Report on the Iron Ore Deposits along the Kingston and Pembroke Railway in Eastern Ontario. Geol. Surv. of Canada. Ann. Rep’t 12, pt 1 : 69-80 Jeffries, Zay, & Archer, R. S. 1924 The Science of Metals. xvii+ 46op. McGraw-Hill, New York Kemp, J. F. & Ailing, H. L. 1925 Geology of the Ausable Quadrangle. N. Y. State Mus. Bull. 261. I26p., map Martens, J. H. C. 1925 Glacial Boulders in Eastern, Central and Northern New York. 20th Rep’t Director, N. Y. State Mus. Bull. 260:81-116 Martin, J. C. 1916 The Precambrian Rocks of the Canton Quadrangle. N. Y. State Mus. Bull. 185. 1 12p., map Miller, W. J. 1909 Geology of the Remsen Quadrangle. N. Y. State Mus. Bull. 126. Sip., map 1910 Geology of the Port Leyden Quadrangle, Lewis County, New York. N. Y. State Mus. Bull. 135. 6ip., 11 pis., map 1914 Magmatic Differentiation and Assimilation in the Adirondack Region. Geol. Soc. Amer. Bull. 25 ^43-64 1916 Origin of the Foliation in the Precambrian Rocks of Northern New York. Jour. Geol., 24:587-619 1917 The Adirondack Mountains. N. Y. State Mus. Bull. 193. 97p., 30 pis., 13 fi&s. 1918 Adirondack Anorthosite. Bull. Geol. Soc. Amer., 29:399-462 1918 Banded Structure of the Adirondack Syenite-granite Series. Science, N. S., 48:560-63 1919 Pegmatite, Silexite, and Aplite of Northern New York. Jour. Geol., 27:42-44 1923 Pre-Cambrian Folding in North America. Bull. Geol. Soc. Amer., 34 : 679-702 1926 Geology of the Lyon Mountain Quadrangle. N. Y. State Mus. Bull. 271 :i-ioi 1926 Origin of Pyrite Deposits of St Lawrence County, New York. Econ. Geol., 21 65-67 1929 Newly Found Adirondack Anorthosite. Amer. Jour. Sci., 5th Ser. 18 : 383-40 1 Mills, F. S. 1908 The Economic Geology of Northern New York. Eng. & Min. Jour., 85 :396-<)8 ; Feb. 22, 1908 Newland, D. H. 1906 The Mining and Quarry Industry of New York State. N. Y. State Mus. Bull. 112:37-40 226 NEW YORK STATE MUSEUM 1908 Mining and Quarry Industry of New York State during 1907. N. Y. State Mus. Bull. 120. 82p. 1910 The Mining and Quarry Industry of New York State. N. Y. State Mus. Bull. 142. 96p. 1917 Pyrite in Northern New York, Eng. & Min. Jour., 104:947-48. Dec. 1, 1917 1921 Mineral Resources of the State of New York. Bull. 223-24:126-29 Quirke, T. T. 1924 Correlation of Huronian and Grenville Rocks. Jour. Geol., 32:316-35 Raymond, P. E. 1914 The Trenton Group in Ontario and Quebec. Geol. Surv. of Canada, Summary Rep’t for 1912 ^42-50 Read, H. H. 1928 A note on “Ptygmatic Folding” in the Sutherland Granite-Complex. Geol. Survey of Scotland, Summary Rep’t for 1927 :3-8 Ruedemann, Rudolf 1925 Fundamental Lines of North American Geologic Structure. N. Y. State Mus. Bull. 260:71-80 1922 The Existence and Configuration of Precambrian Continents. N. Y. State Mus. Bull. 239-40:67-152 1931 The Tangential Master-streams of the Adirondack Drainage. Amer. Jour. Sci., 5th Ser. 22:431-40 Sederholm, J. J. 1926 On Migmatites and Associated Pre-Cambrian Rocks of Southwestern Finland. Bull. Comm. Geol. Finlande, No. 77. i43p., 8 pis., map Smock, J. C. 1889 Iron Mines and Iron-Ore Districts in the State of New York. N. Y. State Mus. Bull. 7. 7op. Smyth, C. H. jr 1894a Report on the Geology of Four Townships in St Lawrence and Jef- ferson Counties. N. Y. State Mus., 47th Ann. Rep't (for 1893) : 685-705 1894b On a Basic Rock Derived from Granite. Jour. Geol. v. 2, no. 7 : 667-79 . _ 1895 Crystalline Limestones and Associated Rocks of the Northwestern Adirondack Region. Bull. Geol. Soc. Amer. 6 :263-84 1896 The Genetic Relations of Certain Minerals in Northern New York. N. Y. Acad. Sci. Trans., v. 15, especially p. 268 1897 The Crystalline Rocks of St Lawrence County, N. Y. State Geol. Rep’t 15:479-97 1899 Crystalline Rocks of the Western Adirondack Region. N. Y. State Geol. Rep’t 17, especially p. 482-83 1901 Geology of the Crystalline Rocks in the Vicinity of the St Lawrence River. 19th Ann. Rep’t N. Y. State Geologist, p. 85-104 1903 The Rossie Lead Veins (St Lawrence County, New York). School of Mines Quarterly 24:421-29 1912 The Genesis of the Pyrite Deposits of St Lawrence County, New York. N. Y. State Mus. Bull. 158:143-82 1917 Genesis of the Zinc Ores of the Edwards District, St Lawrence County, New York. N. Y. State Mus. Bull. 201:1-32 Smyth, C. H. jr, & Buddington, A. F. 1926 Geology of the Lake Bonaparte Quadrangle. N. Y. State Mus. Bull. 269. io6p., map Taylor, F. B. 1924 Moraines of the St Lawrence Valley. Jour. Geol., v. 32, no. 8 :64i-67 HAMMOND, ANTWERP AND LOWVILLE QUADRANGLES 227 Uglow, W. L. 1916 Origin of Certain Ore-deposits. Econ. Geol., 2 187-9 2 1916a Lead and Zinc Deposits in Ontario and Eastern Canada. Ann. Rep’t Ont. Bur. of Mines, v. 25, pt 2. 56p., map Van Hise, C. R. 1896 Principles of Pre-Cambrian North American Geology. 16th Arm. Rep't U. S. Geol. Surv., pt 1 .-571-843 Vanuxem, L. 1842 Geology of New York, 3rd District. 3o6p. Vogel, F. A. 1908 Pyrites Mining and Milling in St Lawrence County, New York. The Mineral Industry, 16 : 845-51 Weidmann, Samuel 1903 The Pre-Potsdam Peneplain of the Pre-Cambrian of North Central Wisconsin. Jour. Geol., 11:289-313 Whitlock, H. P. 1903 New York Mineral Localities. N. Y. State Mus. Bull. 70:74-87 Wilson, M. E. 1924 Arnprior-Quyon and Maniwaki Areas, Ontario and Quebec. Geol. Surv. of Canada, Mem. 136 Winchell, N. H. 1893 Twenty-first Ann. Rep’t Geol. and Nat. Hist. Surv. of Minn., for 1892, p. 104-8 Woodworth, J. B. 1905 Ancient Water Levels of the Champlain and Hudson Valleys. N. Y, State Mus. Bull. 84. 2o6p. Wright, J. F. 1923 Brockville-Mallorytown Map-area, Ontario. Geol. Surv. of Canada, Mem. 134. 63P. ■ ,V I ■ ilMi ■ : • !>/ ;n: >’■ ;• > ii • . «.j.u i~ . ft IK . 1 J c, I •J ,msxiH./;V ■<:■ >K -U ,il ( v . io : . ii 1/ ' ! i <; i i . , . i ■ : fjnlHaO ifitoH K> ii. iv < !*l -Mjf ii. niiiJi; j; tt i ■; 37 ( > « •, .S .H : >IiiriW .MHwQrb^r rfwwCl .mtm/ i itv/iutl/' Ihu; t-spi .■t/ •• ■ ' ! . ■ . ..'{dor ,-ilA 1)1 [’< ) j* .V: m2 ti • .GnatnO r.»i», qcl/ ,! , ; : f< , Figure 19 Cliffs of Potsdam sandstone, northwest of Chapel Corners, Hammond quadrangle Figure 20 Quarry at north edge of town of Lowville. Pamelia lime- stone at base, Lowville limestone at top (above black ink line) [229] Figure 21 Typical scarp formed by Theresa sandy dolomite. Edge of tableland two and one-half miles west of Natural Bridge, Antwerp quadrangle Figure 22 Table-land on Potsdam sandstone, looking north-northwest from drumlin one-half mile east of Sterlingville, Antwerp quadrangle [230] Figure 23 Rolling hill or knob country underlain by granosyenite, southwest of Mount Tom, Carthage quadrangle _ Figure 24 Low rolling limestone country with lineal ridges and depres- sions. Northeast of South Woods School, Hammond quadrangle l23il Figure 25 Topography developed parallel to curving beds at end of synclinal structure. Hill at right is granite, flat silt-floored valley is in quartzitic and calcareous beds, and curving slope at left is in limestone. Syncline one mile northwest of Brasie Corners, Hammond quadrangle Figure 26 Gabbro hill sourrounded by limestone lowland, west side of Beaver Creek, south of state road, Hammond quadrangle 1 232] Figure 27 Surface of peneplain of Precambrian age on granite and gneiss, one mile east of North Wilna, Antwerp quadrangle Figure 28 Surface of peneplain on quartz diorite, slightly dissected by erosion, three and one-quarter miles east of Antwerp [233] Figure 29 Precambrian peneplain on limestone, three miles northeast of Somerville, at South Gouverneur. Slight bedrock irregularities are obscured by silt deposits, Hammond quadrangle Figure 30 Precambrian peneplain on granite much dissected by post- Paleozoic erosion. Village of Rossie on Indian river, Hammond quadrangle [2341 Figure 31 Surface of Pine Plains delta (built in glacial Lake Iroquois) two and one-half miles south of Philadelphia, Antwerp quadrangle Figure 32 Silt plain (built in glacial Lake Iroquois) two and one-half miles south of Philadelphia, Antwerp quadrangle [235] Figure 33 Croghan recessional moraine, one mile south of Croghan, Lowville quadrangle Figure 34 Croghan west recessional moraine, one mile southwest of Croghan, Lowville quadrangle [236I Figure 36 Drumlin and bouldery ground moraine, one-half mile east of Sterlingville, Antwerp quadrangle [237] Figure 37 Pleasant lake, at north end looking south; hill at left is gabbro, lake basin is in limestone ; Hammond quadrangle Figure 38 Lake of the Woods ; walls are of Potsdam sandstone ; Hammond quadrangle [23SI Figure 39 Seven-foot boulder of anorthosite resting on Potsdam sand- stone, one-half mile northwest of village of Black Lake, Hammond quadrangle. Boulder probably was carried one hundred miles or more from its source Figure 40 Postglacial gorge of Black river in Lowville limestone, just below Great Bend, Antwerp quadrangle 1 239 1 Figure 41 Injection gneiss or arteritic migmatite, plicated. North end of Lewisburg syncline Figure 42 Pyroxenic gneiss brecciated by granite, an agmatite, Ham- mond quadrangle. Photograph of specimen on exhibition in New York State Museum, Albany, New York. Length of slab is six feet [240] Figure 43 Interlayered limestone and pyroxene-scapolite-feldspar granu- lite crumpled and crenulated with steep pitch. Three-quarters of a mile northwest of Scotch Settlement School, Hammond quadrangle Figure 44 Granite with included layer of folded amphibolite. Hyde School, Hammond quadrangle [241 1 Figure 45 Pegmatite vein broken and displaced by flowage of limestone, one mile northwest of Elmdale, Hammond quadrangle Figure 46 Photomicrograph of hyperite, showing texture characteristic of normal crystallization from a magma. Two miles northwest of Indian River, Lowville quadrangle, x nicols, x 35 diameters [242] Figure 47 Photomicrograph of hyperite, partly mashed ; from belt show- ing cataclastic structure. Southeast corner of Antwerp quadrangle, x nicols, x 35 diameters Figure 48 Photomicrograph of hyperite, from mylonite band ; one mile southeast of Fargo, Antwerp quadrangle, x nicols, x 40 diameters [2431 Figure 49 Photomicrograph of granosyenite, showing texture character- istic of normal crystallization from a magma, and no evidences of crushing. One-half mile east of Bushes Landing, Lowville quadrangle, x nicols, x 35 diameters. Q = Quartz Figure 50 Photomicrograph of granosyenite showing protoclastic struc- ture, feldspars granulated, quartz (Q) in massive leaves. One and one-half miles north of Croghan, Lowville quadrangle, x nicols, x 35 diameters 1 244] Figure 51 Photomicrograph of hornblende grano-svenite gneiss show- ing cataclastic structure ; feldspar is pulverized, quartz is granulated, x nicols, x 35 diameters. One-half mile northeast of Texas Road School, Lowville quadrangle. Q = Quartz Figure 52 Photomicrograph of granosyenite from mylonite band. Mount McQuillen, Carthage quadrangle. Quartz (Q) and feldspars pulverized, x nicols, x 35 diameters r 245 ] Figure 53 Schoolhouse built in 1817 of Potsdam sandstone, two and one-half miles southwest of Hammond Figure 54 Stratified sand and gravel in kame south of Strickland Corners, Antwerp quadranale [246] INDEX Alexandria batholith, 79 Alexandria type magma, 12; granite, see granite Algoman granite, see granite Altitudes and relief, 22 Amphibolite, 56-58, 65, 80, 83, 85, 107, 1 16, 1 18, 120-26, 132, 133, 156, 157, 161, 164; see also metagabbro Annelid borings in Potsdam sandstone, 178 Anorthosite, 12, 50, 58, 59; age rela- tion of, 94 Aplite, 62, 68, 69, 70, 83, 85, 87, 88, 106, 108, 1 16, 131 ; syenitic, 100 Arteritic migmatite, see migmatite Assimilation, 68, 70-71, 73, 75, 76, 80, 83, 89, 100, 122, 123, 126, 132-33 Augen gneiss, 12, 66, 67, 70, 71, 76 Augite syenite, equigranular, 72 ; gradation of, 101 ; inclusions of, 104 ; porphyritic, 71-72 Batholiths, 140 Bathyurus, 191 Beaver Falls moraine, 42 Beaver river, 9, 31, 52 Beerbachite, 93 Belfort and Kirschnerville deltas, 34 Bent pyrite prospect, 21 1, 213 Bigelow School group of galena veins, 206-7 Biotitic gneiss, 86, 108, 109, 116-20, 127-30 Biotitic schist, see biotitic gneiss Birdseye limestone, 186, 187, 190 Black River, 9, 31, 52, 53; limestone, 187, 192 Boulders, 25, 36, 37, 46, 50, 51 Brockville granite, see granite Caledonia, hematite mine, 196, 198; pyrite at, 212, 213 Carthage moraine, 42-43 Cataclastic microstructure, 159, 161, 164-71 Celestite, 203 Cephalopods, 18, 188 Chalcopyrite, 203 Chatter marks, 47, 49 Chippewa granite, 85-86 Chlorite, 61, 69, 88, 126, 170, 198, 199, 214 Chloritic rock, origin of, 198-99, 200, 201, 211 Clark hematite mine, 196 Clay, 10, 36-39; floored valleys, 38-39 Coal Hill galena vein, 203-4, 209 Cobourg limestone, 18, 183, 187 Cole pyrite mine, 21 1 Conglomerate, 14, 16, 175, 178, 190 Contact aureole, 132 Contact metamorphism, 132-35 Copenhagen delta, 34 Corals, 188 Country rocks, of hematite deposits, 198-200 Crenelate structure, 160, 161 Crescentic gouges, 47, 49 Croghan and Ossoit deltas, 34 Croghan complex, 12; age relations of, 101-3 ; structure of, 164 Croghan moraine, 41 ; West moraine, 42 Crush zones, 159-60 Crystal Dale moraine, 41 Crystalline limestone, 1 1, 30, 107, 110- 14, 137, 138; economic aspects of, 218-20; topographic expression, 26- 27 Cuesta, 186 Deep River moraine, 42 Deltas, 30-38; of Independence, Beaver and Oswegatchie rivers, 31- 33 ; in Lake Glenfield, 33-34 ; post- glacial unwarp of, 54-56 Devoice moraine, 42-43 [247] 248 INDEX Diabase, 92-94 Diana complex, 12; age relations of, 101-2; structure of, 162-63 Dickson mine, 197 ; pyrite deposit at, 212, 213 Dicob moraine, 40-41 Diorite and quartz diorite, II, 59-66, 152; age relations of, 60, 94; dikes of, 149; microstructure of, 64 Dodds granite, 85, 88-89, 132, 155, 157, 158 Dolomite, 118, 144, 180, 219 Domical hills, 23, 46 ; foliation, 79, 155, 156 Downing Farm, galena vein on, 205-6 Drainage, recent and preglacial, 51-54 Drumlins, 46 Elmdale fold, 148-49; belt of garnet gneiss and quartzite, 1 17, 149 Eye schists, 127, 129 Fargo moraine, 42-43 Feldspathic gneisses, 107-9, 116-20, 127-32, 136 Feldspathic schists, see feldspathic gneisses Fluorite, 205, 206, 222 Fold axes, 144 Folds, 13, 139-64, 172-73; drag, 154, 167; secondary, 154 Foliation, 13, 138, 156; across dikes, 163; domical foliation, 79, 155, 156; in Grenville series, 141-42; in igne- ous rocks, 148, 167-71 ; origin of foliation in gneissoid intrusives, 139, 152; relation to topography, 27-28 Frontenac axis, 23 Gabbro, n, 58-59, 92, 93, 104, 107, 120, 123, 139, 150, 152, 154; see also metagabbro Galena veins, 202-9; history of devel- opment of, 209; origin and compari- son with deposits in Ontario, 207- 9; strike of, 161-62 Garnet-biotite schist, 123 Garnet gneiss, 116-19, 121-23, 1251-30, 136-38, 144 ; amphibolite aureoles, 132 Garnet rock, 132-33 Garnet-sillimanite gneiss, 118-19, 127- 30, 132, 133 Gastropods, 17, 181, 188 Gilbert Gulf, 20, 45, 55 ; upwarp of deltas in, 55-56 Glacial ice, 19-20, 24; direction of movement of, 47-51 Glacial lakes, 19-20, 30-39, 47; bould- ers, 50-51 ; striae, 47-51 Glacial lake deltas, 20, 30-39, 220 Gold sands, 214-17 Gossan, 212 Gouverneur limestone, 137, 138; see also crystalline limestone Granite, age relations of, 95-101 ; Alexandria type, 78-79, 85-90, 95, 98-100, 132, 140, 144, 154, 158; Algoman, 97 ; Brockville type, 82, 97 ; description of, 78-92; dikes, 90-94, 1 12, 1 14; Hermon type, 65, 78-83, 96-100, ill, 112, 122, 144, 154, 155; Laurentian, 96, 97, 98 ; Lowville type, 78, 79, 83-85, 95, 101 ; Mal- lorytown, 97; Picton, 96, 98, 99; Precambrian weathering of, 174; topographic expression, 27-28 Granite pegmatite, see pegmatite Granosyenite, equigranular, 74-75 ; gradation of, 101 ; Hermon type, 80-83 1 mixed rock, 77-78 ; porphy- ritic, 75-77 Graphite, 112, 115, 198, 211, 214, 221; deposits, 217-18 Grass lake, 47 Gravel, 39-45, 220 Gravity stratification, 162 Grenville foothill belt, 21-22 ; pyrite and pyrrhotite veins in, 210 Grenville series, 10-11; contact meta- morphism of, 132-35 ; description of, 106-32; folding of, 139-40; inclu- sions of, 70, 85, 107, 162; strati- graphy of, 135-38; structure of, 139-62 Ground moraine, 25, 39, 40, 45-46 INDEX 249 Hematite, 179 ; deposits, 194-202 ; essential features of occurrence of, 201 ; future possibilities of, 202 ; origin of, 175, 200-2 Hermon type granite, see granite Hickory lake, 47; granite, 155, 156, 158 Hill or knob country, 26 Hornblende syenite, equigranular, 73, 83 ; gradation of, 101 ; porphyritic, 72 Hornblendic gneiss, 58, 86, 131 ; see also amphibolite and metagabbro Hyde granite, 85, 86-88, 121, 155, 157, 158 Hyperite, 77, 92-93, 104 Igneous origin of certain gneisses, 138 Igneous rocks, description of, n-13, 56-94; mode of intrusion of, 11-13, 1 55-59, 162-64, see also sills; rela- tive age of, 94-104 Independence river, 31, 52 Indian river, 53-54 Iron ores, see hematite Jepson galena vein, 204 Joints in Grenville belt, 161-62; in main igneous complex, 171 ; in Paleozoic beds, 173; topographic effect, 28 Jones farm, galena vein on, 205-6 Kame moraine, 44 Keene-Antwerp belt, hematite deposits of, 195, 199; pyrite and pyrrhotite deposits of, 210, 21 1, 212-13 Keene hematite mine, 196, 197; pyrite vein at, 21 1, 213 Kingdon lead deposit, 208 Kings Falls delta, 35 Kirschnerville deltas, 34 Laccoliths, 149, 152 Laidlaw belt of pyritic deposits, 213 Laidlaw pyrite vein, 2x1 Laidlaw School anticline, 148 Lake Glenfield, 33-34 Lake Iroquois, 20, 35-37. 45, 53; up- warp of deltas built in, 55-56 Lake of the Woods, 47 Lake Port Leyden, 31-33 Lakes, present, 46-47 ; temporary, 20, 30-39 Lamellibranchs, 188 Laurentian granite, see granite Lead veins, see galena veins Leray limestone, 10, 17, 25, 26, 182, 183, 187, 190-93 Lewisburg syncline, 116, 142, 144-46, 155 Limestone, 10, 16, 17, 18, 25, 182, 186- 93; economic aspects of, 220; Pre- cambrian weathering of, 174; see also crystalline limestone Limonite, 21 1 Lineal structure, 162, 171 Lorraine formation, 183 Lowville anticline in Diana complex, 163, 171 Lowville type granite, see granite Lowville limestone, 10, 16-17, 26, 182, 187, 190-93 Lowville, section at railroad bridge at, 191-92 Macomb group of galena veins, 205-6 Magmatic differentiation, 101-7, 138 Magmatic stems and lines of descent, 104-6 Mallorytown granite, see granite Marble, 218-20 Marine fossils, 20, 38 Marine waters, 20, 38 Metagabbro, age relations of, 94; description of, 56-58 ; intrusions into, 69, 92 Metamorphic rocks, 11, 106-42 Migmatite, 58, 69, 91, 92, 100, 107, 1 16; of granite or syenite-amphi- bolite, 122-26; see also metamorphic rocks Mill Creek section, 188 Mineral Point galena vein, 206 Mineral specimens, localities for, 221- 22 Mixed rocks, 107, 119, 123-24; see also migmatite Molybdenite, 222 Monzodiorite, 59, 63, 70 250 INDEX Moon lake, 46, 62 Moon Lake diorite, 62-63 Moraines, 39-47 Morgan pyrite vein, 21 1, 212, 213 Muskalonge Lake syncline, 148 Myrmekite, 81, 83, 124 Natural Bridge delta, 34, 35 Old Stirling hematite mine, 197, 199, 200, 222 Orogenic stresses, 11, 13, 138, 139, 157, 158, 159, 171 Orthoceratite, 17, 191 Ossoit delta, 34 Oswegatchie river, 31, 52 Overturned folds, dipping northwest or overturned toward southeast, 140-52, 162; dipping southeast, 152- 54 Oxbow southwest belt, pyrite deposits of, 213 Paleozoic deposits, 14-18 ; formations of Hammond and Antwerp quad- rangles, 177-82; of Lowville quad- rangle, 183-94 Pamelia limestone, 10, 16, 26, 182, 187, 188-89 Payne lake, 46 Payne Lake granite, 85, 89, 132, 155, 157 Pegmatite, 1 1, 57, 58, 65, 69, 70, 72, 77, 79, 82, 83, 84, 86, 87, 90-92, 104, 106, 107, 109-14, 1 16, 121, 122, 152, 157, 169, 170; structure of, 160-61 Peneplains, 19 Pennock farm, galena vein on, 205 Phacoliths, 12, 132, 140, 144, 149, 154, 155-59 Philadelphia moraine, 44-45 Physiographic provinces, 20-23 Phytopsis, 17 Picton granite, see granite Pike hematite mine, 196 Pine Plains delta, 35, 36, 53 Pit lakes, 47 Pitch, 162, 163; see also structural geology Pleasant lake, 46; syncline, 149, 15 1, 152 Pleistocene, 10, 19-20, 24; recessional moraines and ground moraines, 39- 46 ; submergence, 38 ; direction of movement, 47-51 ; temporary lakes and high level deltas, 30-39 Pope Mills, graphite deposit near, 217 Porphyritic rock, of replacement origin, 70 Postglacial unwarp, 34, 54-56 Post-Paleozoic upwarp and erosion, 18-19, 21 Potsdam formation, 13, 14-15, 177-81 Potsdam sandstone, 10, 14, 18, 20, 21, 25, 26, 29, 47, 177-81, 199, 201 ; annelid borings in, 178; economic aspects of, 220 ; galena vein in, 207 ; slump structure of, 178; structure of, 172; tubular structure, 178 Precambrian peneplain, 13, 21, 23, 28- 30, 201; soil, 13, 14, 173, 176; weathering, 13, 1 73-77, 199, 201 Pre-Potsdam peneplain, see Precam- brian peneplain Pre-Potsdam stream channel, 52 Pre-Potsdam topography, 18, 28 Protoclastic structure, belts of, 164-71 Pyrite, 115, 199, 201, 203, 205, 206 Pyrite-chlorite schists, 198 Pyrite deposits, 209-17; country rock of, 210; origin of, 214; structure and character of, 210-12 Pyroxene-biotite gneiss, 119 Pyroxene-biotite schist, 75, 84 Pyroxene scapolite rocks, 130 Pyroxenic gneiss, 86, 102, 107, 108, 1 19-120, 130-32 Pyrrhotite, see pyrite Quartz, of replacement origin, 109, no, hi, 112, 113, 114-16, 133, 135 Quartz-diopside rock, 1 14-15 Quartz diorite, 59-66, 82, 152 Quartz-feldspar gneiss, 116 Quartz lenses, 83, 118 Quartz mesh limestone, m-15, 135 Quartz-pyroxene rocks, 114-16, 133 Quartz veined limestone, 107 Quartz veins, 57, 84, 92, 107, in, 115, 13 1, 170, 206 INDEX 251 Quartzite, 11, 107-10, 115, 117, 136, 137, 144 Recent deposits, 10, 20; uplift, 20, 38 Recent time, 19-20; upwarp in, 54-56 Recessional moraines, 23, 39-45 Red lake, 46 Relief in feet, 22 Roaring Brook section, 189 Rossie dioritic rocks, 59-66, 104-6 Rossie group of galena veins, 203-5 Rossie igneous complex, 149-52 Rusty gneisses, 114, 21 1 St Lawrence lowlands, 19, 20-21, 23 Sand, 31, 33, 35, 36, 220 Sand plain, 35, 36, 37, 38, 39 Sandstone, 10, 14, 15, 18, 20, 21, 25, 26, 29, 47, 175; economic aspects of, 220 Sapping of walls, 47 Scapolite, 70, 71, 108, no, 111-13, 120, 124-26, 130, 131, 221 Schist inclusions, 28 Sericitic rocks, origin of, 198-99 Serpentine, 113 Sheets, see sills Sherman Lake granite body, 81, 148; syncline, 116, 144, 146, 148, 155 Shirtliff mine, 197 Shonkinite, 102 Sillimanite, 1 19, 129; schists, 127; eyes, 127 Sills or sheets, 1 1, 12; anorthosite, 58; aplite, 70, 78, 83, 85, 87, 88, 89, 100, 101 ; Diana complex, 162 ; diorite and quartz diorite, 60-62, 63, 141 ; folded, 154; gabbro, 58, 139, 150; granite, 84, 144-48; syenite, 66-67, 70-71, 141, 150 Silt, 10, 36, 37 ; floored valleys, 38-39 Somerville anticline, 116, 144, 146, 148, 155 South Hammond quartzose diorite, 62 Sphalerite, 203, 205, 206, 222 Sponges, 188 Stella pyrite deposits, 210, 211, 212 Stratiform sheet, 1-01 Stratiform structure, 12 Stratigraphy, of Grenville, 135-38; of Paleozoic, 186-93 Striations, 47 Stromatocerium, 190 Strontianite, 203 Structural geology, 138-73; of Gren- ville belt, 142-62 ; of intrusives, 140- 41 ; of main igneous complex, 162- 71 ; of Paleozoic beds, 172-73 Sunstone, 221 Sweet School and Copenhagen deltas, 34 Syenite, 12, 66-74, 123, 150, 152; age relations of, 94-95 ; Alexandria, 96, 153; mixed rocks, 122-26; series, 104-6 Syenite-granite mixed rock, 77-78 Symplectic intergrowth, 61 Table-lands, 25-26 Tangential drainage around Adiron- dack massif, 184-86 Temporary lakes, see lakes Terrace, 35, 38, 39, 42 Tetradium, 17, 188, 189, 190, 191, 192 Theresa dolomite, 10, 14, 15, 16, 21, 25, 181-82; sandstone in, 181, 182 Till, 45-46 Topographic types, 23-30 Trenton limestone, 18, 183, 187, 193- 94 Trilobites, 188 Tug Hill plateau, 20, 22-23 1 structure of, 173 Turner farm, galena vein near, 206 Ultramylonite .band, 166-67 Uptilt or upwarp, post-Paleozoic, 18- 19; post-Pleistocene, 20, 38, 54-56 Utica formation, 183 Victoria galena vein, 204-5, 209 Watertown limestone, 10, 17-18, 182, 183, 187, 190-93; moraine, 144 Weathering, 13, 24-25 ; of pyrite veins, 21 1 ; Precambrian, 13, 199 Wight pyrite deposit, 212, 213 Yellow lake, 46 Zinc, 206; see also sphalerite k> •!/[ : ?: ■ -j v ' ■ *fl> : , '• ; , .Pi ,/fjnftl.joi 93n»tv/g,.l ■>(' ■ . ' „ '■ ' ‘ • : . > ."tin? ( I ■ ' :1 , ■ . - ■ - ■■ ■ . ' ■ u W) rt ja § 4-» c O ffi < CD •C 0 3 •s ■5 <4-4 O <4-4 <4-4 <4-4 O O O 04 cd CU O4 04 0 1-H cd a cd a cd a a bn 0 cd 0 0 0 0 1-4 -d *5b n •a 'o ‘So 0 r—4 0 *a 0 'o cd 0 < y s s « « o o o o co 04 o* rt RS 04 5- I a a 2 i 1 to ' r*3psr LEGEND dominantly ofulnclo- Inoustrlno orl«lu. Map- ped whore so widespread as to obscuro bed root. Ulorlto and quartz dlor- lntruHlvo sheets or gram- lte and apllte. m White quartzite, layered white quartzite and llmegtone, with lo- cal beds or «ruy quartz- ite and gneiss. Geology by A. F. Buddington, 1920-1927, UNIVERSITY OF THE STATE Topography by U. S. Geological Survey and State of New York, 1908-1910. GEOLOGIC MAP OF THE HAMMOND QUADRANGLE GRENVILLE SERIES, PRECAMBRIAN H BULLETIN NO. 296 GEOLOGIC MAP OF THE ANTWERP QUADRANGLE 1 V mm -4 GEOLOGIC MAP OF THE PRECAMBRXAN AREA, CARTHAGE QUADRANGLE LEGEND Heeesslonal moraines Delia sands formed In Glacial Lake Iroquois Qrunlte. fine to medium Gran 1 to. medium coarse porpbyrlr' - Grnnosyonlto, porphy- Hornblende syenite, Auglte syenite, porpliy- Anorthoslte. gnbbro-nn- ortbuslte and local gubbro. l^H Pyroxone-blotlte gneiss, occurring ns a belt oflu- Strlkc and dip of foliation Vertical foliation GEOLOGIC MAP OF THE LOWVJLLE QUADRANGLE ■ukk LEGEND SEDIMENTARY ROCKS :sSS®= jiil type") l' e;s m^ss3^sies& >- 8lr,Mo?1,POr IfSSHS ass