(….reprinted  from Bulletin of The Archeological Society of New Jersey, No. 22, Nov. 1965)

 

AN ABORIGINAL QUARTZ QUARRY
AT SAMP MORTAR RESERVOIR, FAIRFIELD, CONNECTICUT

By

 Bernard W. Powell

 

PREHISTORIC quartz quarries and chipping stations are not commonly described from this area.  One such is here described, and the nature and extent of the workings are identified and defined.  Techniques of the ancient quarriers are inferred from the excavated evidence, and a chronological placement is advanced.       Steatite, schist, and graphite quarries presumably dating to very ancient times have been identified by numerous researchers in the Northeast (R. P. Bullen, 1940; W. S. Fowler, 1943; W. J. Howes, 1944; and G. C. Dunn, 1945).  These materials were widely sought and traded by Amerinds, apparently in preceramic times.       Less frequently cited are quartz quarries, probably because float quartz and other durable silicates are widely dispersed in the glacial drift of southern New England.  Nevertheless, especially attractive outcrops of quartz were worked by the aborigines from time to time.  I here describe one such site, and the nature and extent of its workings.

THE SITE

      Near 41' 11' 18” N. lat. and 73' 15' 48” W. Lon., north of Fairfield, Fairfield County, Connecticut, on the west side of Samp Mortar Reservoir (Fig. 2), is a small aboriginal quartz quarry.  The site lies about 1,500 feet south-southwest of the bridge where Black Rock Turnpike crosses the upper end of the reservoir. A prominent rock escarpment strikes roughly southwest here.  At one spot along its edge occurs a group of glacial potholes; erroneous attribution of these phenomena to the Indians may account for the area name. Just east of the escarpment rises a small knoll at about 120 feet above sea level.  In heavy underbrush near the top of this knoll is an outcrop of clean, massive, white quartz.  Apparently, a major intrusion reaches the surface here and was discovered and exploited by the aborigines long ago.  Local country rock is contorted gneisses and schists, presumably of the Hartland Formation (Rodgers, Gates, et al.,1956).  Detailed topographic relations can be ascertained from the Westport Quadrangle of the U. S. Geological Survey 7.5 Minute Series.

      It is not wholly clear whether the intrusion penetrated a preexistent fracture-and-fault system.  However, many of the quartz pieces are quite flat-sided, and it is postulated that at least some of the contacts in the veins were along flat-faced breaks.  Further, the intrusion is riven with smooth, flat cracks.  Some of these are roughly parallel and a foot or more apart.   The flat surfaces are coated with a very thin layer (2mm. or less) of massive, black tourmaline.  A mineralogist 1), has said this is quite common where vein quartz cuts gneisses and schists. In my opinion, the flat-sided nature of the quartz here was one of its most attractive properties to the aborigines, as will be explained.

      South and slightly west of the outcrop are several large quartz boulders, some up to 6 feet through (Fig. 2).  They lie at different radii to the outcrop, but their orientations and distributions (none to east or north) strongly   suggest glacial action as the force which wrenched them loose from the outcrop and skidded them to their present positions. Some are slightly embedded in soil; the largest and most distant sits about 75 feet away on an exposed ledge.  Though weathered and stained, these erratics show signs of battering by the   quarriers of long ago.

      Concentrations of chips noted as weathering from the surface in the vicinity of these boulders prompted several test pits (indicated parenthetically as 1, 2, 3, 4, and 5 in Fig. 2).  These pits confirmed presence of chipping refuse and a few partially worked pieces at some depth.  This material, plus the battering, is unmistakable evidence of attention by the early quarriers at this hillside site, and the boulders are therefore defined as part of the complex.  But it is unquestionably at the outcrop that the major evidences of aboriginal quarrying abound.

TRENCH A

      Chipping refuse and tailings littered the ground here.  Accordingly, a test trench, Trench A, was run through this area and into the knoll (Fig. 2).  This disclosed the major quarrying station at this site. Both plan and profile were plotted (Fig. 3).  All materials hereinafter described come either from Trench A (and the expanded headward extension of it as shown), or Trench B just to the west.  This subsidiary trench confirmed the general impression and relationships noted in Trench A, and no plan or profile was made.

At the surface, faint outlines of an ancient ditch or hollow were detected at the northern end of Trench A. As can be seen, Trench A extends southward into the outcrop itself; it is postulated that the faint outline relates to activities of the ancient quarriers.  In this respect, the depression is reminiscent of the ancient quarries at renowned Flint Ridge, Licking County, Ohio, where I have traced similar hollows in the vicinity of the outcrops.

      Stratigraphy seems to support this interpretation, for no obvious upsets or anomalies occur.  It should be noted, however, that there is a local rumor that some rock shelters (numerous along the escarpment) were dug years ago by "college students" - said to be from Yale.  A search of the literature fails to reveal any published material and a direct appeal to the Department of Anthropology at Yale has gone unanswered, so this rumor remains unchecked.  If, however, excavators were active in the vicinity thirty or more years ago, it is conceivable they may have dug here, and the soil today is reweathering back to a "natural" profile. This, however, is thought to be quite unlikely, in view of the numerous chipped and partially worked forms here, which surely would not have been left in situ by any excavators.

      The "topsoil" at the site was really a deep layer of duff, whose lowest portions only were sufficiently distintegrated to form a soil.  This material filled the hollow and lay over much of the outcrop.  In places it could be bodily lifted away, or rolled back like a rug, similar to a condition cited previously by me at a wooded site some miles west of here (Powell, 1959).   Beneath this material, in most spots, lay the mass of the outcrop.  However, in some places (Fig. 3) this "topsoil” was underlain by tan, yellowish subsoil, presumably of the B horizon.  This was most noticeable in the northern half   of Trench A, and at several spots in the southern half, where this subsoil filled crevices in the outcrop.

      Despite the profusion of chips observed at the surface, well over 90 percent of all material recovered came from the subsoil.  Much of this material was concentrated along the fissures in the outcrop.  As excavation proceeded, the modus operandi of the quarriers became apparent. They had concentrated their efforts along the natural fractures, and so, over the millennia, had widened and deepened their workings.  They worked on, over, and through their own refuse, apparently stopping now and then to blank out smaller pieces, losing some of them in the shuffle, breaking still others, and withal   continuing to pry out and remove large blocks along the fissures.  At least this is a composite guess based on the observed situation.

      Many blocks occur at random over the site; a number of them lie just northeast of the working face exposed in Trench A (Fig. 2) and were undoubtedly pitched there by the Indian miners.

      Chips, flakes, cores, tailings, blanks, and occasional worked pieces continued to the deepest sounding made, and showed no noteworthy variation in intensity or distribution.   The deepest spot reached in our excavation was 18 inches below the datum plane AE (Fig. 3) and 30 inches below the original surface, since the ground here slopes up to the top of the knoll. At this depth, many of the fissures pinched out and ran together. It seems highly probable that chipping refuse and other material occur to the bottom of every crack and cranny in the intrusion.  Difficulty of excavating along these narrow seams and low yield of meaningful data precluded further investigation, but I am confident our observations and inferences are substantially correct.

      A small ash lens was exposed on the east face of Trench A, not too far below the surface in the northern end (Fig. 3).  A badly fire-disintegrated piece of quartz lay in conjunction with it.  It is entirely possible, due to the shallowness of this feature plus the fact that the only modern artifact at the site (a man's corroded wrist-watch) was found at the surface here, that it dates from the modern era.  However, even if it dates to aboriginal times (no  sufficiently uncontaminated charcoal or organics could be obtained from it), it probably does not relate to the quarrying operations.  Use of fire as a quarrying technique of early men has a dubious reputation in the literature, and most authorities deny that it was ever used (Ellis, 1957).

      An apparent anomaly at the site is complete absence of any hammerstones or mauls.  The obvious intensity of the percussive activity here argues for frequent loss and breakage of such tools.  None were found, and there is no ready explanation for their absence.

Description of Recovered Materials

      The bulk of the material was strictly quarry trash and tailings. Nevertheless, each piece was examined for what it could reveal about the activities that once transpired here.  In all, we recovered and examined over 500 pieces.  The sheer mass of this material dictated cataloguing as it was removed, and leaving it in the field.  Despite the intensity of the industry here, few completely formed or finished pieces were found.  However, a number of blanks and a few rough artifacts permitting tentative identification were recovered, and are described below.

      Because of the flat-sided nature of the quartz at Samp Mortar Quarry, described earlier, the quarriers were able to secure a wide range of pieces having two roughly parallel sides.  These constituted natural cores, with ready-made striking and anvil platforms.  It was a simple matter to set such pieces down on a nearby ledge, and begin at once to detach flakes through either direct or indirect percussion.  Analysis of the refuse leaves little doubt that this was one of the main activities of the quarriers.  Again and again we noticed cores with marked ridges of percussion showing where the maximum number of flakes had been drawn from the striking platform. When such a core was inverted, it invariably showed one of the flat, tourmaline-coated faces downmost as the anvil platform.  On several of these exhausted cores, negative bulbs of percussion could be traced around the periphery of the anvil platform, caused by rebound forces transmitted back to the core when originally struck.

      With a little practice, one could pick up examples of this type of core very readily in the tailings.  The smallest in a series of thirty catalogued in the field was 2 inches long; the largest was over 30 inches and required two men to lift it.

      The adhering black tourmaline layer proved a boon in retrieving deliberately struck flakes from among the trash.  Many of these were quite large; some showed angular detachment from the cores as shallow as 30 degrees, suggesting that very oblique blows must have been delivered by the early knappers.

      A few of the more finished items, and pieces where two or more examples occur and are thought to represent conceptual tool patterns, were photographed and are shown in Plates II and III.

      Plate II A, a and b, illustrate a repetitive type of fist-size chunk with rough percussion-chipped edges around three-quarters of the perimeter and slight indenting or notching on the lower edge.  Specimen c has a nicely chipped   curvate cutting edge running up onto a prominent spur on the left edge.  Other examples of this form were noted in the field.  Specimens d and e are, in some sense, not unlike a form called the "Triangular Tailing-Breaker" described by Fowler (1963) for steatite quarries in Massachusetts and Rhode Island.  He believes it relates to Late Archaic times there. Many examples of the schist "blades" f and g, some perhaps natural and not man-made, occurred at Samp Mortar. These were the only artifacts noted that were not made of quartz. Since many ledges of schist abound nearby, the quarriers may well have spent some time roughing out these blades.  One can only speculate as to their use, however, since the soft micaceous schist would not stand up to much hard service.  The polls of h and i are broken off, but the worked bits along their lower edges indicate they were crude celts of the plano-convex variety.  Their breakage and loss at the site might suggest they were actually quarry tools, but this seems uncertain.

      In Plate II B are three large blades or blade-blanks.  Numerous examples of this form were noted at the quarry.  Probably one attractive aspect to a quarry source for quartz is that large, uniform pieces of stock could be obtained. This was not possible with a regional pebble industry, which nonetheless may have been contemporary with the quarry industry. Doubtless such pieces were later modified into large hide-scrapers, choppers, and blades of all types, and perhaps into hoes and agricultural tools of later peoples too, but no finished items were found here.  All are bifacially flaked from both edges, which left them with medial ridges of percussion.  Two, b and c, may show incipient stem formation.   In Plate III, a may be a hand-held pick, possibly a quarry pick. The specimens b-h suggest projectile-point blanks, but some may be purely fortuitous, since all are very roughly percussion struck.  Two, band d, may be blanks for triangular forms; c does suggest specifically a side-notched blank, and has a marked similarity to the illustrated type Brewerton Side-Notched, described by Ritchie (1961) mainly for sites of the Middle Archaic horizon in New York. The relatively narrow-bladed blanks e-h may be Lamokoid in concept, and might also be generally thought to relate to Archaic times.   The semilunate knives (or scrapers) in the series i-m seem a well-enough defined group. They are like many such forms described from sites, generally Archaic, in the Northeast. Fowler (1963) describes a "Stemless Knife" whose general criteria match those of this series; he assigns it mainly to the Late Archaic in Massachusetts. A drawing in a letter 2 illustrates a strikingly similar artifact from the Christiana stone bowl quarry in Pennsylvania. Specimen i in this series comes from Bitter Rock Shelter about 10 miles west of Samp Mortar Quarry (Powell, MS).  It occurred near the bottom of a relatively deep stratigraphic deposit there, and is interpreted by me to be Archaic in that context. Its similarity to the Samp Mortar series is quite apparent. Specimens n and o are rather undistinguished and perhaps deserve no special comment.

Discussion

      At some point in postglacial time, Amerinds discovered and began to work a quartz outcrop and associated boulders several miles inland from the southern Connecticut seacoast.  Sharp cultural or temporal diagnostics are wanting, but several tentatively identified artifact forms fall generally in the Mid-to-late Archaic Period, so it is postulated that initial quarrying here began perhaps 4,000-5,000 years B.P.

      Observations of the outcrop itself, and of the tailings, cores, and blanks support the inference that the ancient quarriers concentrated mainly on breaking out large pieces of quartz along natural cracks.  Rough flake scars and negative bulbs of percussion, many still traceable on working faces in situ at the quarry, suggest that percussive methods of extraction were paramount.  Presence of numerous cores with naturally formed striking and anvil platforms is advanced as one of two main reasons the aborigines chose to work this site.  The other is thought to be availability of large, flat, homogeneous pieces of quartz, typified by the "quarry blades" shown in Plate II B.

      Presence of some secondary chipping (Plate II A, c, h. i), plus immense quantities of chipping debris, indicate that primary extraction was not the sole aim of the quarriers.  However, no finely finished pieces, and no hammerstones were recovered.  This may mean the quarriers carefully removed their finished pieces and their tools elsewhere.  In the case of the hammerstones, it is not clear how such substantial percussive activity can have transpired without loss or breakage of hammerstones (and other recognizable quarry tools) in some quantity at the site.

      The quarry may have been visited and worked by people from distant areas who carried their blanked-out pieces away with them, a phenomenon cited not infrequently in the literature.  The quarry may also have been worked by local people, perhaps even occupants whose remains (by rumor) are said to have occurred in nearby rock shelters.  A. quick survey suggests that these shelters are too disturbed to yield much meaningful data. Nothing is known of any inventories assembled from these shelters, but it is said that ceramics were found in one instance.  Marine shell fragments also occur at one of these shelters. One small scrap of marine shell (Venus mercenaria) was found at the surface in Trench B at the quarry, but it is entirely unclear whether this is ancient or modern, and whether it was left here by the Indians, or perhaps carried up here by rodents, carnivores, or other scavengers who may have disturbed the accumulations in the rock shelters. Perhaps only the two trianguloid projectile point blanks Plate III ( b, d) relate to Indians of ceramic times; it seems quite possible that quarrying had ceased here prior to the Woodland Period.

      The tourmaline-coated, flat-base cores might be detectable at more remote sites, but their uniqueness as coming only from Samp Mortar Quarry seems challenged by expert opinion that vein quartz in this condition is relatively common in regions of gneissic rocks.

Acknowledgments

      I thank my co-worker at the site, Mr. Ted Jostrand of Norwalk, for his assistance in the field; also Mr. David M. Seaman, Department of Mineralogy, American Museum of Natural History, New York, for comment on samples of the quartz submitted to him; and finally to representatives of The Greenfield Development Company for permission to make our investigation.
 

Footnotes

1 David M. Seaman, personal communication.
2 William S. Fowler, personal communication.
 
 

Literature Cited

      BULLEN, RIPLEY P.
                       1940 The Dolly Bond Steatite Quarry.  Bulletin Vol. 2, No. 1, Massachusetts Archaeological Society.
                        Attleboro.

      DUNN, GERALD C.
                       1945 The Oaklawn Quarry, Cranston, R. I.  Bulletin Vol. 6, No. 4,Massachusetts Archaeological
                       Society.  Attleboro.

      ELLIS, H. HOLMES
                       1957 Flint-Working Techniques of the American Indians.  The Ohio Historical Society. Columbus.

      FOWLER, WILLIAM S.
                       1943 A Quarry in Westfield, Mass.  Bulletin Vol.4, No. 3, Massachusetts Archaeological Society.
                       Attleboro.

                       1963 Classification of Stone Implements of the Northeast.  Bulletin Vol. 25, No. 1, Massachusetts
                       Archaeological Society.  Attleboro.

      HOWES, WILLIAM J.
                       1944 Indian Soapstone Quarries of Western Mass.  Bulletin Vol. 5,No. 4, Massachusetts
                       Archaeological Society.  Attleboro.

      POWELL, B. W.
                       1959 A Ceramic Find at Hunting Ridge, Connecticut. Bulletin Vol. 20, No.3, Massachusetts
                       Archaeological Society.  Attleboro.

                       n.d.  Bitter Rock Shelter: A Stratified Connecticut Site.  Manuscript in press; Massachusetts
                       Archaeological Society, 1963.Attleboro.

      RITCHIE, WILLIAM A.
                       1961 A Typology and Nomenclature for New York Projectile Points. Bulletin Number 384, New
                       York State Museum and Science Service.  Albany.

      RODGERS, JOHN; ROBERT M. GATES, EUGENE N. CAMERON, AND RUBEN J. ROSS, JR.
                       1956 Preliminary Geological Map of Connecticut. Geological and Natural History Survey.  General
                       Drafting Co., Inc., Convent Station, N. J.

END