The Quaternary Glaciation of Western New England with Correlations to Surrounding Areas

The Quaternary glaciation of western New England with correlations to surrounding areas

John C. Ridge

Department of Geology, Tufts University, Medford, MA 02155, U.S.A. E-mail: [email protected]

Abstract bottom 14C ages. A comparison of terrestrial 14C ages for ice margins in the Merrimack Valley with marine 14C ages Over the last 20 years there have been major improvements for contemporaneous marine ice margins in southern Maine to our understanding of the Quaternary glaciation of New suggest that a 600-1300-yr marine reservoir correction England. Numerous Accelerator Mass Spectrometer (AMS) should be applied to the marine chronology. Correlations 14C ages of terrestrial plant fossils have eliminated some of between New England and New York have been formu - the errors associated with 14C ages of lake-bottom bulk lated with the matching of independent palaeomagnetic de- sediment samples and marine fossils. The rebirth of Antevs’ clination records from both areas and the connection of ice- New England varve chronology and its 14C calibration have front positions that appear to align geographically. There added precision to the Late Wisconsinan deglaciation appears to be a contempo-raneity of readvances in both chronology and a means of precisely testing correlations of regions that also match cold intervals on Greenland ice core deglacial events. Palaeomagnetic studies of both declination records. These observations support a rapid response of the and polarity have allowed precise correlation between New last ice sheet’s ablation system to cold events. Following England and adjacent areas. deglaciation a delayed isostatic tilting occurred across cen- The exact chronology of pre-Late Wisconsinan glacia- tral to northern New England with water planes dipping tion is still poorly constrained because of difficulties in 0.85-0.94 m/ km towards the south-south-east. The non- determining numerical ages. The terrestrial glacial record of glacial remnants of glacial lakes in the upper Connecticut New Jersey dictates that New England was glaciated at Valley persisted until at least 10.5 14C ka BP (12.5 cal ka least twice in pre-Wisconsinan time, at least once when the BP) and may have been seen by the first humans in the geomagnetic field had a reversed polarity (pre-Illinoian, area. While isostatic uplift was well underway in southern Marine Isotope Stage (MIS) 22 or older, >850 ka) and again and central New England recession of ice in Canada during the Illinoian (160-180 ka BP, MIS 6). Controversy allowed the invasion of marine water into the Champlain exists concerning pre-Late Wisconsinan till in central and Basin at about 11.0 14C ka BP (13.0 cal ka BP). This age northern New England with its weathering supporting an estimate is consistent with previous studies that attempted 14 Illinoian age and amino acid ages from reworked marine to remove a reservoir error from marine C ages of Champ - fossils in Boston Harbor supporting an Early Wisconsinan lain Sea fossils. age. In the Connecticut Valley of southern New Hampshire and Vermont the advance of Late Wisconsinan ice buried 1. Introduction advance outwash and lake beds were deposited in tributaries impounded by advancing ice. Late Wisconsinan ice This paper is a review of the Quaternary glacial history of reached its limit on islands along the southern coast at 24.0- New England except for Maine, which is covered by 20.0 14C ka BP (28.0-23.7 cal ka BP). The overall pattern of Thompson et al. (this volume) in detail. Correlations and deglaciation was one of spurts of ice recession punctuated comparisons will be given to adjacent New Jersey, New by readvances and end moraine building, and acceleration York and Maine when they are relevant to determining the of ice recession over time. The early chronology of degla- ages of pre-Wisconsinan glaciation, evaluating the syn- ciation (20.0-15.0 14C ka BP, 23.7-18.0 cal ka BP) has been chroneity of Late Wisconsinan glacial readvances, and inferred from correlations to Greenland ice core records and evaluating differences between terrestrial and marine 14C varve sequences have been used to crudely constrain the ages. This review will focus heavily on developments over chronology. The later part of deglaciation (15.0-11.5 14C ka the last 20 years that have greatly changed our understand- BP, 18.0-13.4 cal ka BP) has been determined precisely ing and resolution of New England’s glacial record. where basal ice-proximal varves are matched to the 14C- Until recently the glacial chronology of New England calibrated New England varve chronology in the Connecti- was largely based on spot occurrences of 14C ages and cut, Merrimack, Passumpsic, and Winooski Valleys. The interpretations tied to weathering characteristics. Over most resulting precise chronology of deglaciation in central and of New England traceable end moraines are scarce as northern New England is about 1500 years younger than in compared to regions further west making morphostrati- previous models largely based on bulk sediment lake- graphical correlation difficult. The use of morphostratigra- 164 John C. Ridge

Fig. 1. Southern and central New England showing locations relevant to pre-Late Wisconsinan glaciation and the advance of Late Wisconsinan ice. Also shown are the limits of Late Wisconsinan, Illinoian and pre-Illinoian glaciation in New Jersey.

phy has largely been limited to stratified deposits with the the difficulties in studying deposits that were mostly ob- detailed mapping of morphosequences, a mapping tech- literated by Late Wisconsinan glaciation. Pre-late Wiscon- nique employed by the United States Geological Survey sinan deposits are exposed at only scattered localities be- over the last half century (Jahns & Willard, 1942; Koteff, neath younger deposits making them difficult to correlate. 1974; Koteff & Pessl, 1981). Several stratigraphical In addition they were sometimes subjected to large-scale techniques have evolved to greatly increase the accuracy glaciotectonic deformation by later ice advances. Remnants and precision of glacial chronology in New England, of pre-Late Wisconsinan deposits on the continental shelf especially for the Late Wisconsinan. The resurrection of of southern New England (cf. review by Stone & Borns, varve chronology has provided a means of determining the 1986) have been altered in many places by later glaciation sequence of deglaciation with unprecedented resolution. A and sea-level change making them difficult to study and major break through has been the 14C calibration of varve correlate in cores and seismic records. Further difficulties sequences for a span of about 5000 14C years (6000 varve arise when trying to determine numerical ages beyond the or calibrated years), a period covering Late Wisconsinan limits of 14C dating. Many infinite 14C ages or ages at the deglaciation from central Connecticut into Canada. This limit of 14C dating have been obtained on deposits overly- calibration has also allowed a comparison between ing pre-Late Wisconsinan glacial units, giving a minimum terrestrial and marine 14C ages that may eventually produce age to the upper boundary of pre-Late Wisconsinan units an accurate reservoir correction for marine 14C ages. (Shafer & Hartshorn, 1965; B. Stone & Borns, 1986). A Combined with the varve chronology is also a record of single U-Th age and amino acid racemisation ages on ma- palaeomagnetic declination that has allowed correlation rine fossils from two areas have placed further constraints with glacial lake beds and events in New York. on the ages of pre-Late Wisconsinan glacial units along the Palaeomagnetic polarity also represents a promising para- coast of New England (Belknap, 1979, 1980; Oldale, 1982; meter along with cosmogenic surface dating for deciphering Colgan & Newman, 1999). However, marine fossils do not the history of pre-Wisconsinan glaciation. occur in glacial deposits further inland and there are few numerical constraints in central and northern New England. Other techniques are just now being applied to the glacia- 2. Pre-Late Wisconsinan glaciation tion of New England. Palaeomagnetic records have recently added age constraints to the pre-Wisconsinan stratigraphy The full chronology of pre-Late Wisconsinan glaciation in of other parts of the north-eastern U.S. (Braun et al., 1994; New England remains largely uncertain because of some of Gardner et al., 1994). Cosmogenic surface dating is just New England 165 beginning to produce numerical ages that can be applied to Boston Harbor (Kaye, 1961; Humphrey & Hager, 1991; both late and pre-Late Wisconsinan glaciation (Larsen et Newman et al., 1993) has not yet been attempted. al., 1995; Bierman et al., 2000; Stanford et al., 2000).

2.2. The penultimate glaciation across New England 2.1. The Pre-Wisconsinan of New Jersey Perhaps the most contested aspect of pre-Late Wisconsinan The exact number and ages of pre-Wisconsinan glaciations glaciation is the age of the penultimate glaciation across in New England have yet to be determined but adjacent New England (Oldale & Colman, 1992). In coastal strati- areas of New Jersey represent a terrestrial environment graphical sections, the cores of drumlins, and at isolated beyond the limits of the Last Glaciation where extensive inland localities, there are multiple till sections. While there pre-Wisconsinan deposits have been studied (Fig. 1). is general agreement that the uppermost till is Late Wis - Detailed superficial mapping of pre-Wisconsinan deposits consinan, ages for the lower units have ranged from Illinoi- in New Jersey indicate that New England was glaciated at an to Late Wisconsinan. A pre-Late Wisconsinan (Early least twice during pre-Wisconsinan time (B. Stone et al., Wisconsinan or Illinoian) age, representing a penultimate 2002). Extending no more than 15 km beyond the Late glaciation, has been favoured for lower tills at most Wisconsinan limit is till defining the limit of Illinoian locations because the weathering of these units is generally glaciation. Based on the preservation of glacial landforms thought to have occurred prior to the Last Glaciation. The (end moraines and terraces) and the degree to which the dispute over the age of lower or drumlin till units in New deposits are weathered this area has been correlated (B. England has generally become known as the ‘two till prob- Stone et al., 2002) with proposed late Illinoian (180-160 ka lem’ (for reviews see: Shafer & Hartshorn, 1965; Koteff & BP, MIS 6) deposits in surrounding areas of Pennsylvania Pessl, 1985; B. Stone & Borns, 1986; Weddle et al., 1989; (Braun, 1994, this volume). The character of weathering of Newman et al., 1990; and Oldale & Coleman, 1992). tills on Long Island (Sirkin, 1982; Cadwell, 1989) and in Perhaps the best-constrained two-till stratigraphy is that southern New England (Oldale, 1982; Oldale et al., 1982; of Nantucket on the southern coast of New England. Here Oldale & Es kenasy, 1983; Melvin et al., 1992; Oldale & Late Wisconsinan till and stratified deposits overlie an Colman, 1992; J. Stone et al., 1998b) suggests the exten- interglacial shallow marine to littoral sand and gravel unit sion of Illinoian ice beyond the Late Wisconsinan limit. with shells and coral. The interglacial units have yielded A second older area of pre-Wisconsinan glaciation in amino acid ages and a single U-Th age of 140-120 ka BP New Jersey extends up to 40 km beyond the Late (MISubstage 5e; Oldale, 1982; Oldale et al., 1982; Oldale Wisconsinan limit (Fig. 1) and contains scattered highly & Coleman, 1992). Till immediately beneath the inter- weathered occurrences of glacial sediment (Stanford, glacial marine sequence must therefore be no younger than 1993a; B. Stone et al., 2002). The poor preservation of Illinoian (MIS 6). Correlation of this unit to Martha’s deposits and landforms in this region does not allow the Vineyard (Oldale, 1982), Block Island, the Montauk Till of limit of ice advance to be mapped precisely. This region is Long Island (Muller, 1965; Shafer & Hartshorn, 1965; assumed to represent an area of pre-Illinoian glaciation and Sirkin, 1982, 1986; Veeger et al., 1994) and deposits it may have the deposits of more than one glacial stage. offshore (Oldale & Eskenasy, 1983) conflicted with the Correlation of these deposits to similar units in original Early Wisconsinan age designation for till in these Pennsylvania (B. Stone et al., 2002), some of which have other areas (Oldale, 1982; Oldale & Colman, 1992). reversed palaeomagnetic polarities (Braun, 1994; Gardner The two till stratigraphy of Boston Harbor drumlins et al., 1994), indicate an age greater than 790 ka BP. Based (Fig. 1) is typical of the two-till stratigraphy seen in on their reversed magnetism the deposits in Pennsylvania drumlins across much of southern and central New England have been tentatively assigned to MIS 22 (Braun, 1994; 850 (Shafer & Hartshorn, 1965; B. Stone & Borns, 1986; ka BP, pre -Illinoian ‘G’ of Fullerton & Richmond, 1986). Weddle et al., 1989; Newman et al., 1990; 1993: Oldale & Heavily oxidised pre-Illinoian stratified silt and clay near Colman, 1992; Newman & Mickelson, 1994). Based on a Bloomsbury, N.J. have yielded reversed magnetic consistent pattern of weathering, best documented in the directions (J. Ridge & R. Witte, unpublished). Magnetism Boston Harbor drumlins (Newman et al., 1990, 1993; in these deposits may be carried by iron oxide or hydroxide Newman & Mickelson, 1994), the lower or “drumlin” till is minerals formed by weathering suggesting an age generally perceived as a single pre-Late Wisconsinan unit potentially much older than MIS 22. Given the extent of throughout the region. It also became clear that the Boston pre-Illinoian glaciation in New Jersey it is reasonable to Harbor drumlins were formed by ice moving with a conclude that New England was glaciated at least once different direction than during the Last Glaciation. Surface during pre-Illinoian time. Correlation of the New Jersey bedrock striations across the Boston area indicate an ice deposits to possible equivalents on islands along the flow direction of 155-160° during the Last Glaciation. Till southern New England coast (Fuller, 1914; Woodsworth & fabrics, drumlin orientations, and striations on bedrock Wigglesworth, 1934; Kaye, 1964; Shafer & Hartshorn, overlain by the lower till in Boston Harbor indicate an ice 1965; Belknap, 1979; Oldale, 1982), on the continental flow direction of 110° (Newman et al., 1993; Newman & shelf (cf. review by Stone & Borns, 1986), and beneath Mickelson, 1994).

166 John C. Ridge

Oldale & Colman, 1992). Amino-acid ‘ages’ of 214 and 200 ka BP on Mercenaria shell fragments from the lower till in Boston Harbor (Belknap, 1979) were also compatible with an Illinoian age. These ages were later reevaluated to a Sangamonian (MIS 5) age (Belknap, 1980) but these revisions have largely been ignored in assigning an age to the lower till based on its weathering. Early Wisconsinan glaciation nearly as extensive as the Late Wisconsinan limit seems incompatible with global sea level and ice volume determined from marine 18O records and dated tropical coral terraces (Oldale & Coleman, 1992). Still, outside Boston Harbor, the lower till in central New England, representing a penultimate glaciation, has not been directly- dated with a numerical dating technique. Creating further controversy, Colgan & Newman (1999) obtained amino- acid ‘ages’ on 27 Mercenaria shells from the lower till in Boston Harbor that cluster into ranges of 135 + 5 ka BP, 100 + 15 ka BP, and 70 + 20 ka BP. These new ages suggest an Early Wisconsinan advance of ice that constructed drumlins across Boston Harbor. If these amino- acid ‘ages’ are accurate it implies that there are at least two ages for the lower till units, MIS 6 on Nantucket and MIS 4 in Boston Harbor. In New Hampshire, thinking on the two-till problem developed much as it did to the south. A preferred Illinoian interpretation of the lower till in stratigraphical sections is primarily based on weathering developed prior to the Late Wisconsinan. Emphasised in the study of several sections has been not only lower till weathering but also the glaciotectonic deformation, and incorporation of weathered lower till into Late Wisconsinan deposits (Koteff & Pessl, 1985). Unfortunately the problem is more difficult in northern New England where there is a complete absence of materials for numerical dating. Also, it is more difficult to make the argument that Early Wisconsinan ice volume was insufficient to cover northern New England. In at least one place in northern New England multiple till stratigraphy has been found without any evidence of interglacial weathering. At the Comerford Dam in the Connecticut Valley just west of Littleton, New Hampshire (Fig. 2) is a three-till section Fig. 2. Western New England and adjacent areas showing the in which the lowest till exhibits no weathering (Crosby, approximate geographical positions and numbers of long 1934b; Ridge et al., 1996; Thompson et al., 1999). The age measured varve sequences in the northeastern United States and some of the glacial lakes from which varves were measured. NE yr of the lowest till in this sequence could be anything from 2701-6352 and 6601-7750 represent the original New England late to pre-Wisconsinan in the absence of any numerical varve chronology of Antevs (1922, 1928). Varve numbers in ages that can be applied to the stratigraphy. brackets are varve sequences that have not been matched to the New England varve chronology. The ice margin along the southern coast of New England is the limit of Late Wisconsinan 3. Late Wisconsinan glaciation glaciation. 3.1. Recent advances in Late Wisconsinan chronology From the 1960’s and into the early 1980’s the lower till in central New England was generally accepted to be an Over the last 50 years 14C ages from organic sediment in Early Wisconsinan unit, equivalent to the Montauk Till on proglacial, glacial and post-glacial deposits have been used Long Island (Shafer & Hartshorn, 1965; B. Stone & Borns, to determine chronologies for the Last Glaciation in New 1986). During the 1980’s numerical ages from Nantucket England (Shafer & Hartshorn, 1965; Davis et al., 1980; and correlations across southern New England lead to an Davis & Jacobson, 1985; Hughes et al., 1985; B. Stone & interpretation of southern New England lower tills, Borns, 1986; Dyke & Prest, 1987). However, few of the 14C including the Montauk till, as Illinoian (Oldale, 1982; ages were precisely tied to glacial events and many had New England 167

Fig. 3. The six major varve sequences in the north-eastern United States (>200 yr long) that have been measured showing their relative ages. Varve records within each grid have their own numbering system. Varve records from separate geographical areas within each grid have been matched. The overlap of grids shows the possible overlap of different varve sequences that have not been matched. inherent uncertainties when used to interpret glacial the ages of glacial events and test 14C ages for their internal stratigraphy. 14C ages from bulk or whole -sediment samples consistency with high precision. and marine fossils were used verbatim without correction In areas of southern New England the 14C-calibrated because they were the only ages available and correction varve stratigraphy and palaeomagnetic records cannot yet factors had not been formulated. Bulk sediment ages of be applied to deglaciation chronology and there is still a lake -bottom sediment are now known to have potentially reliance on 14C ages of limited accuracy. Another approach large errors (Shotton, 1972; Oeschger et al., 1985; Andrée has been to determine ages for moraines based on et al., 1986; Cwynar & Watts, 1989; Törnqvist et al., 1992; correlations (Boothroyd et al., 1998) to successive cold Wohlfarth et al., 1993, 1995; Lowe et al., 1995; Abbott & phas es depicted by oxygen isotope records of the GISP2 ice Stafford, 1996; Wohlfarth, 1996; Björck et al., 1998a; core from Greenland (Cuffey et al., 1995; Stuiver et al., Turney et al., 2000). In the north-eastern U.S. and adjacent 1995). Although the accuracy of these correlations requires Quebec errors are up to 10 kyr, with 1-2 kyr errors being further testing, they appear to be consistent with the common (Mott, 1975, 1981; Karrow & Anderson, 1975; combined 14C-calibrated varve and palaeomagnetic Miller & Thompson, 1979; Davis & Davis, 1980; Davis et chronologies for central and northern New England and al., 1995; Lini et al., 1995; Ridge et al., 1999). Glacial existing local 14C ages. Oxygen isotope correlations are marine reservoir errors have also been well documented consistent with the rapid response of glacial ablation to (Mangerud, 1972; Hjolt, 1973; Mangerud & Gullikson, periods of cold that trigger ice advance or stabilisation 1975; Sutherland, 1986; Bard, 1988; Southon et al., 1990; (Lowell et al., 1998). Bard et al., 1994; Austin et al., 1996; Birks et al., 1996, Stocker & Wright, 1998; Bondevik et al., 1999). Late Pleistocene glacial marine environments and modern 3.2. Varve chronology and palaeomagnetic records in New marine waters in and around New England are estimated to England and adjacent New York have reservoir errors of 500-2000 yr (Anderson, 1988; Hillaire-Marcel, 1988; Rodrigues, 1988, 1992; Kasgarian, 3.2.1. Major varve sequences in the north-eastern U.S. 1992; Ridge et al., 2001; Ridge, 2003). Over the last ten years accelerator mass spectrometer Since the early 1990’s, the rebirth of the New England 14C dating of terrestrial plant fossils and the application of varve chronology, and its associated 14C calibration and minimum marine reservoir corrections (400-500 yr) have palaeomagnetic records, have provided a tool unmatched in eliminated some ambiguities regarding 14C ages (Thompson precision for determining the ages and correlations of et al., 1996; Dorion, 1997; Borns, 1998; Ridge et al., 1999; glacial events across much of New England and New York. Dorion et al., 2001; Ridge, 2003). Calibration of the 14C Varve chronology in the United States began in earnest in time scale (Stuiver & Reimer, 1993; Stuiver et al., 1998), 1920 with an expedition led by Gerard De Geer, who was the rebirth of varve chronology in the north-eastern United accompanied by three of his students, Ragnor Lidén, Ebba States, and palaeomagnetic records (Ridge & Larsen, 1990; Hult De Geer, and Ernst Antevs. However, it was Antevs Ridge et al., 1990, 1996, 1999, 2001; Ridge, 2003) now (1922, 1928) who almost single-handedly assembled the six provide an opportunity to make calendar year estimates of major varve sequences (length >200 yr) that exist in the 168 John C. Ridge north-eastern United States (Fig. 2). Antevs (1925, 1931, eastern New Hampshire, the Ashuelot Valley of south- 1951) also assembled long varve sequences in Canada. The western New Hampshire, and the Hudson Valley near six American sequences (Fig. 3) represent the measurement Newburgh, New York and from Catskill to near Cohoes, and compilation of varve records from many different New York. The Hudson Valley records are important, not exposures and glacial lakes into averaged or normalised only because correlations can be documented across a wide sequences that depict regional variability of varve thickness region and between hydrologically separate basins, but also from year to year. the Hudson Valley records from Catskill to Cohoes filled a The oldest varve sequences (Figs 2 and 3) are from gap in records from New England (varves 5600-5687). glacial lakes in New Jersey (Stanford & Harper, 1991; Recently varves in this time period were recovered in a core Stanford, 1993b; B. Stone et al., 2002). A 200-yr sequence in the Connecticut Valley at Amherst, Massachusetts (Fig. from Lake Passaic at Little Falls (Antevs, 1928) and a 2; Rittenour, 1999). This led to a 10-yr correction to the 2532-yr sequence from the Hackensack River valley lower Connecticut varve sequence where exceedingly thick (numbers –1097 to 1434) do not appear to overlap1. The flood units in the Hudson Valley (varves 5669-5678) had Hackensack sequence was assembled by Reeds (1926) and been misidentified as annual layers (measurements by G. partly replicated by Antevs (1928). The upper 1434 varves De Geer reported in Antevs, 1922). of this sequence represent deposition of varves in a non- The upper Connecticut varves (6601-7750; Antevs, glacial2 lake making them difficult to match with 1922, 1928) were formulated from matching records in the contemporaneous glacial varve sequences further north. Connecticut Valley north of Claremont, New Hampshire, The remaining four major American sequences are the Passumpsic Valley in northern Vermont, and the based at least partly on varve measurements in New Winooski Valley of western Vermont (Figs 2 and 3). At one England. The oldest of these sequences is 700 years long varve exposure at Newbury, Vermont in the upper and represents a correlation of varve records from glacial Connecticut Valley Antevs (1922) counted an additional lakes in the Hudson Valley near Haverstraw, New York and 750 varves to varve 8500. The additional 750 varves were the Quinnipiac Valley near New Haven, Connecticut (Figs . too thin for Antevs to measure in the field. Recently this 2 and 3; Antevs, 1928). This sequence might overlap the section was collected in overlapping cores and measured varves in the Hackensack Valley in New Jersey but it is with the aid of a computer and magnified digital images entirely older than varve sequences further north. (Ridge & Toll, 1999), extending measurements to varve The most useful varve sequences, for deciphering the 8679 (+35/-20). Plus/minus values were applied to varves glacial history of New England, were compiled by Antevs in the upper part of the section to account for uncertainties (1922, 1928) from mostly varves in the Connecticut Valley in the identification of annual layers. Beginning at varve and have become known as the lower and upper 7470 the upper Connecticut varves in the Connecticut and Connecticut varve sequences (Figs 2 and 3). These two Passumpsic Valleys abruptly become thin (<5 mm), a sequences are the only two that have been calibrated with change that appears to represent a greatly reduced volume 14C ages. They share a numbering system and together are of glacially derived sediment. While the varves above varve known as the New England varve chronology. The 7470 may record a climate record and annual weather numbering system arbitrarily begins with varve 2701, an patterns (Ridge & Toll, 1999) the change to non-glacial attempt by Antevs to accommodate the future addition of varves has made correlations to areas with glacial varves older varves to the south. The lower Connecticut varves problematic (Ridge, 2003). (2701-6352; Antevs, 1922, 1928) represent a correlation of Antevs (1922) could not find a match of the lower and varves from the Connecticut Valley (Hartford, Connecticut upper Connecticut sequences. He concluded that the lower to Claremont, New Hampshire), the Merrimack Valley of sequence was entirely older and deposited while areas north of Claremont remained ice covered indicating a delay in ice recession in the Claremont area (Fig. 2). Numbering of the 1 The terms ‘match’ and ‘overlap’ have distinct meanings. upper Connecticut varves was arbitrarily started at 6601 Matching varve sequences have similar annual thickness (Antevs, 1922) with a 249-yr separation between the two records that allow a definitive annual correlation. Over- Connecticut sequences that has become known as the lapping varve sequences cover the same time span but their Claremont gap (Fig. 3). A close examination of the two exact annual correlation is not known. sequences in the mid -1990’s by the author resulted in finding a crude but possible match of the two sequences 2 The term glacial varve is used where the delivery of (varve 6012 = 6601) that would have eliminated the sediment by glacial meltwater controls annual layer Claremont Gap and implied that deglaciation of the thickness. A non-glacial varve is deposited in a lake where Claremont area was rapid and continuous (Ridge et al., delivery of sediment from meteoric (runoff) sources 1996, 1999). Elimination of the Claremont Gap appeared to 14 dominates or is the only source, usually in periglacial and be compatible with existing C ages and palaeomagnetic paraglacial environments. It is difficult to match con- records (discussed later) from the two sequences and an temporaneous glacial and non-glacial varve records because apparent lack of any ice-marginal deposits in the Claremont annual thickness is controlled by different mechanisms in area that would indicate a prolonged halt or readvance of each case (Ridge & Toll, 1999; Ridge, 2003). the receding ice sheet. However, more recent detailed

New England 169

Table 1. 14C ages from varves in the New England varve chronology

Laboratory Age d13C NE varve Material dated Reference number (14C yr BP) PDB(‰)

Canoe Brook, Dummerston, Vermont GX-25735 12,660 + 50 -28.9 5858 Woody twigs and Dryas leaves Ridge et al., 2001 GX-14231 12,355 ± 75 -27.2 6150 Bulk sample of silt and clay with Ridge & Larsen, 1990 non-aquatic leaves and twigs GX-14780 12,455 ± 360 -27.6 6150 Handpicked non-aquatic leaves Ridge & Larsen, 1990 and twigs, mostly Dryas and Salix CAMS-2667 12,350 ± 90 --- 6150 Salix twig N. Miller, pers. comm. 1993 GX-14781 12,915 ± 175 -27.1 6156 Bulk sample of silt and clay with Ridge & Larsen, 1990 fragments of peat and gyttja Amherst, Massachusetts Beta-124780 12,370 + 120 -27.1 5761-5768 Plant fragments from core Rittenour, 1999

Newbury, Vermont GX-23765 11,530 ± 95 -27.0 7435-7452 Woody twig Ridge et al., 1999 GX-23766 11,045 ± 70 -27.5 8206 Woody twig Ridge et al., 1999 GX-23640 10,940 ± 70 -26.8 8357 Woody twig Ridge et al., 1999 GX-23641 10,080 ± 580 -26.7 8498-8500 Woody twig Ridge et al., 1999 GX-23767 10,685 ± 70 -26.3 8504 Woody twig Ridge et al., 1999 GX-23642 10,040 ± 230 -26.5 8542-8544 Chunk of wood Ridge et al., 1999 GX-23643 10,440 ± 520 -26.8 8652-8662 2 woody twigs Ridge et al., 1999

East Barnet, Passumpsic River valley, Vermont GX-26456 11,220 + 50 -27.1 7754 Woody twig Wilson, 2000

Columbia Bridge, Vermont WIS-961 11,540 ± 110 -29.0 (>7400) Wood fragments Miller & Thompson, 1979

WIS-919 11,390 ± 115 -27.5 (>7400) Wood fragments Miller & Thompson, 1979 Missing d13C values are ages for which no value was recorded. mapping has revealed three sets of small end moraines all (Figs 2 and 3). This sequence is 345 years long and associated with minor readvances in the Claremont area probably overlaps the upper Connecticut varves in time (Ridge, 1999, 2001) and eight new varve sections between (between varves 7500-7750) although definitive matching Charlestown and Claremont (Fig. 2) that do not match of the two sequences is not likely because varves above either the lower or upper Connecticut varves. The new 7470 in the upper Connecticut Valley are non-glacial. In varve sequences are short (<100 yr) and do not provide the addition to the six major varve sequences there are many complete sequence needed to cover the Claremont gap. other shorter sequences of less than 100 varves, scattered These observations along with a new 14C age and across New England from Connecticut, Rhode Island, and calibration of the lower Connecticut varve sequence south-eastern Massachusetts to northern Vermont and into indicate a gap between the lower and upper Connecticut Québec (Antevs, 1922, 1928) that have not been matched to varves of about 350 years (Ridge, 2003). any of the major sequences. Antevs (1928) assembled a sixth major varve sequence Beginning in the 1930’s varve chronology lost in the Champlain Valley near Essex Junction, Vermont credibility with most Quaternary geologists in North 170 John C. Ridge

America as a valid chronological tool. First, Flint (1929, 1999; Ridge, 2003) using 14C ages from terrestrial plant 1930, 1932, 1933) and later Goldthwait (1938) erroneously fossils (Table 1; Fig. 4). Key 14C ages, screened for their interpreted the last deglaciation of New England as being precis ion and potential accuracy, were calibrated to U-Th the result of regional stagnation. Regional stagnation was years using the CALIB4.3 computer programme and incompatible with systematic south to north recession of an INTCAL98 data set (Stuiver & Reimer, 1993; Stuiver et al., active ice margin (Lougee, 1940) and the general south to 1998) that take advantage of parallel 14C and U-Th ages of north onlapping of varves in the wake of the receding ice tropical corals with well-constrained marine reservoir 14C sheet (Antevs, 1922, 1928, 1939). Flint (1930) initiated corrections. A calibrated time scale was then applied to the doubts as to the accuracy of varve chronology but recanted varves by simply adding or subtracting varve years from the some of his early criticisms in later publications (Flint, calibrated ages assuming that a varve year and a calibrated 1932, 1933). It was also during this time that Gerard De year represent an equal time span. Once a calibrated (U-Th) Geer (1921, 1926, 1927, 1929, 1940) and later Ebba Hult time scale was established a corresponding 14C time scale De Geer (1951a, 1951b, 1954) proposed transatlantic and was calculated using the CALIB4.3 programme. inter-hemispheric correlations based on the annual 14C ages ranging from 12.5-12.3 14C ka BP, from varve matching of individual varve records. Most geologists then 6150 at Canoe Brook in south-eastern Vermont (Table 1), and today, including Antevs (1931, 1935, 1953, 1954), had were first used to calibrate the lower Connecticut varves serious reservations about global correlations, which led to (Ridge & Larsen, 1990; Ridge et al., 1996, 1999). further scepticism by North America geologists. Later, Flint Unfortunately these 14C ages do not provide a very precise (1956) used erroneous interpretations of 14C sample sites to calibration because they overlap a plateau in the 14C time reject the length and validity of the New England varve scale (12.6-12.4 14C ka BP) when about 200 14C years span chronology despite protests by Antevs (1962). All about 900 calibrated years (Fig. 4). A new 14C age from references to Antevs’ (1922, 1928) work on varves in New varve 5858 at Canoe Brook (12,660+50 14C yr BP = 15,300 England were omitted from the last two editions of the cal yr BP; Table 1), that pre-dates the 14C plateau, has been three glacial and Quaternary geology textbooks authored by used to revise the calibration of the lower Connecticut Flint (1947, 1957, 1971). These textbooks were used by varves (Ridge, 2003). This new calibration increases the almost every North American university student in the field calibrated age of the lower Connecticut varves by about 700 until about 1980. yr over the previous version. The new calibration (Ridge, Periodically after the 1930’s geophysicists were able to 2003) implies that the Claremont gap, between the lower exactly match new varve sections to the New England and upper Connecticut varves, is about 350 calibrated or varve chronology (McNish & Johnson, 1938; Johnson et varve years long as compared to the 249-yr gap (varves al., 1948; Verosub, 1979a; b; Thomas, 1984). Detailed 6352-6601, Fig. 4) predicted by Antevs’ (1922). surficial mapping across New England by the United States An average of five 14C ages collected over a range of Geological Survey (USGS) has also revealed a systematic 1200 varves (1100 14C yr) and with precisions less than 100 south to north recession of an active ice front during the last yr have been used to calibrate the upper Connecticut varves deglaciation (Koteff & Pessl, 1981); a model consistent (Table 1; Fig. 4). Despite using five 14C ages over a range with Antevs’ work. In the late 1980’s Ridge & Larsen of 1200 varves there is still some uncertainty in the (1990) were able to match exactly a new exposure of over calibration of the upper Connecticut varves on the order of 530 varves at Canoe Brook, Vermont (Fig. 2) to the New +200 14C years. Of the five 14C ages used to calibrate the England varve chronology. The Canoe Brook section sequence the youngest three 14C ages are older than the covered a span of the varve chronology that Antevs (1922) calibrated time scale by about 200 14C years while the assembled by matching and averaging varve measurements oldest two 14C ages are younger than the calibration by 100- from eleven separate exposures. Since 1990 many new 200 14C years. These discrepancies are smaller when sections have been matched to the New England varve viewed in terms of calibrated years. chronology (Ridge et al., 1996, 2001; Levy, 1998; Rittenour, 1999; Wilson, 2000; Larsen et al., 2001) and there appears to be no doubt about the accuracy of the varve 3.2.3. Palaeomagnetic declination records from New chronology as a correlation tool. Mathematical analysis of England and New York the varve chronology has also been used to determine the periodicity of sub-decade to century-scale climate Varve chronology by itself has not yet allowed a correlation oscillations that occurred during Late Wisconsinan of glacial events beyond New England except to a few deglaciation (Rittenour et al., 2000). places in the Hudson Valley. More regional correlations have been accomplished through palaeomagnetic declination stratigraphy recorded by remanent magnetisation in 3.2.2. 14C calibration of the New England varve chronology laminated muddy glaciolacustrine sediments. Palaeo- magnetism is a chronological correlation tool that extends 14C and calibrated (U-Th) time scales for the lower and the 14C and calibrated age chronologies of the New England upper Connecticut varve sequences have been formulated varve chronology to a wider region, especially to the and revised (Ridge & Larsen, 1990; Ridge et al., 1996, Mohawk Valley of central New York and the Champlain

New England 171

Fig. 4. Atmospheric 14C and calibrated (U-Th) time scales applied to the New England varve chronology. Some individual 14C ages from the varve chronology (Table 1) used to calibrate the chronology are plotted for comparison.

Valley of north-eastern New York and western Vermont. two records must in some way overlap is justified given the Correlation based on declination records takes advantage of time covered by the New England record (15.2-10.5 14C ka the secular variation of geomagnetic declination that BP, 18.2-12.5 cal ka BP). This is the time span generally produces a pattern of eastward and westward declination recognised as the period of deglaciation in central to maxima separated by periods of rapid transition (1º/10-20 northern New York (Connally & Sirkin, 1973; Muller & yr). The maxima have different amplitudes and in some Prest, 1985; Muller & Calkin, 1993). In both declination cases are unique over the time of deglaciation. records is a western declination maximum of 320-295º that Remanent declination records have been obtained from is unique during the period of Late Wisconsinan varves of the New England varve chronology and muddy deglaciation in both regions (Fig. 5). As a test of this lacustrine units in New York (Fig. 5). The Connecticut correlation independent 14C ages associated with both Valley record was first compiled from measurements in records are similar where the palaeomagnetic records varves matched to the New England varve chronology by match. 14C ages from Canoe Brook indicate an age of 12.7- McNish & Johnson (1938) and Johnson et al. (1948). 12.5 14C ka BP when declination approached the extreme Verosub (1979a, 1979b) later verified these results in most western maximum in varves of the Connecticut Valley. In of the lower Connecticut Valley varves (varves 3000-5350) the Ontario Basin the western maximum itself occurs in using modern laboratory techniques. Over the last decade sediment deposited at about the time of the initiation of declination records from nearly all of the remaining lower Lake Iroquois (Brennan et al., 1984). The early sediment of and upper Connecticut varves, including the extension of Lake Iroquois has an age of 12.5-12.1 14C ka BP based on the chronology to varve 8500 have been verified (Ridge et 14C ages of spruce twigs from Lake Iroquois deposits al., 1996, 1999, 2001; Ridge, 2003). (Muller & Prest, 1985; Muller & Calkin, 1993). The In central New York to northern Vermont lacustrine declination records also match at approximately equivalent sediment used to construct a declination record has not been 14C ages when the Champlain Sea invaded the Champlain matched to the New England varve chronology (Fig. 5). Basin (about 11.0 14C ka BP). At the time of this event a The time scale of this record is simply relative age and it is declination of 0º is recorded during a shift from western to non-linear. Relative ages for the western Mohawk Valley eastern declination on both records. samples are based on the superposition of samples from Four major readvances in central New York, represent- single outcrops and the positions of units within a complex ed by diamicton units in the Poland Formation and later de- stratigraphical sequence (Ridge et al., 1990, 1991; Ridge & posits of the Ontario Basin and Mohawk Valley, are tied to Franzi, 1992; Ridge, 1997). In other regions the succession the New York declination record (Ridge et al., 1990). The of deglaciation and glacial lakes in New York and Vermont readvances include the Salisbury, Hinckley-St. Johnsville, has been used to determine relative ages of palaeomagnetic Barneveld-Little Falls, and Rome Readvances (Fig. 5; Mul- samples (Brennan et al., 1984; Pair et al., 1994; Ridge et ler et al., 1986; Ridge et al., 1990; Ridge & Franzi, 1992; al., 1999). Muller & Calkin, 1993; Ridge, 1997). Palaeomagnetic Matching of declination maxima of similar amplitude is corre-lation to the New England varve chronology has used to correlate the New England and New York allowed the application of numerical ages to New York declination records. Also, the general assumption that the events. 172 John C. Ridge

Fig. 5. Correlation of Late Wisconsinan palaeomagnetic declination records from New England and New York. Declinations in New England are plotted versus their position in the New England varve chronology. Declinations in New York are plotted on a non-linear relative age scale as determined by superposition at single outcrops, the pattern of deglaciation, and the sequence of glacial lakes. Error bars on some data points are α95 (cone of confidence) values. Data sources in the Connecticut Valley are Johnson et al. (1948); dots with tie line), Verosub (1979a); open circles and envelope) and Ridge et al. (1999); solid circles with error bars). Data from the Merrimack Valley are from Ridge et al. (2001); open squares with error bars). Data from New York are from Ridge et al. (1990); Mohawk Valley, dark circles with error bars), Brennan et al. (1984); open circles from Ontario Basin and Mohawk Valley), Pair et al. (1994); open circles from St. Lawrence Basin) and Ridge et al. (1999); solid circles from Champlain Basin). The abbreviations of glacial readvances in New York (Table 2) are given at the right side of the diagram.

John C. Ridge 173

Table 2. Abbreviations of moraines, ice margins, and glacial of the first arrival of ice in various places. There are few readvances in Figures 5, 6, and 7. 14C ages that document the first arrival of the last ice sheet and most of these 14C ages are at the limit of the 14C dating BB Buzzards Bay Moraine technique on the southern coast of New England, making BF Bloomfield ice margin BL Barneveld-Little Falls Readvance interpretations of the rest of the region highly speculative. BP Bridport Readvance In the absence of numerical ages there is currently no CA Carthage-Harrisonville ice margin precise chronology for the first arrival of Late Wisconsinan CC Cassville-Cooperstown Moraine ice across New England. CH Chicopee Readvance In the Connecticut Valley of southern Vermont and CL Claremont moraines New Hampshire it has been possible to make some CM Charlestown Moraine interpretations of the character of the land surface, CO Covey Hill ice margin especially in valleys, when the Late Wisconsinan ice first CP Camp Meeting Cutting ice margin arrived. Detailed superficial mapping has revealed a variety DM Delmar Readvance EF Enosburg Falls ice margin of weathered and unweathered deposits beneath till of the FI Fishers Island Moraine Last Glaciation. In the Connecticut Valley at West FP Fresh Pond Moraine Lebanon, New Hampshire (Fig. 1) a non-weathered fluvial HA Lake Hackensack ice margin gravel (interpreted to be advance outwash) has been found HH Harbor Hill Moraine along the valley side beneath Late Wisconsinan till (Larsen, HS Hinckley-St. Johnsville Readvance 1987a). The fluvial unit as well as well records with a HV Haverstraw ice margin similar extensive sub-till gravel unit to the south (Ridge, LB Littleton-Bethlehem Readvance 1990) suggest that during ice advance the valley was LC Lake Charles ice margin occupied by a large braided river to an elevation at least 35 LH Lake Hitchcock dam ice margin LZ Luzerne Readvance m above the modern Connecticut River. The fluvial MB Middleburg Readvance deposits cannot be the product of deposition during Late MS Middlesex Readvance Wisconsinan ice recession and burial by till of a readvance OC Ogdensburg-Culvers Gap Moraines because a deep glacial lake occupied the Connecticut OS Old Saybrook Moraine Valley during final ice recession. PI Pellets Island Moraine A very common situation in the Connecticut Valley of PR Pineo Ridge Moraine southern Vermont and New Hampshire is highly weathered QN Quinnipiac ice margin schist and micaceous gneiss, colluvium, and bouldery RH Red Hook Moraine stream deposits beneath unweathered glacial lake beds that RK Ronkonkama Moraine RM Rome Readvance are varved and contain ice-rafted debris. All these sub-till RO Rosendale Readvance units occur as a package beneath Late Wisconsinan till SA Salisbury Readvance (Ridge, 1988, 1999, 2001). These deposits are preserved in SH Shenandoah Moraine east-west trending tributaries of the Connecticut River from SL Star Lake Moraine Claremont to Walpole, New Hampshire (Fig. 1; Cold, Little SM Sandwich Moraine Sugar, and Sugar River valleys of New Hampshire and the SP Sands Point Moraine Williams River valley of Vermont) where they were SS Sussex Moraine protected from the scour of southward advancing Late TCC L. Taunton-Cape Cod ice margin Wisconsinan ice. The unweathered lake beds in these TM Late Wisconsinan Terminal Moraine valleys appear to represent lakes impounded in tributaries VH Valley Heads Moraines WC West Canada Readvance by the southward advance of a lobate Late Wisconsinan ice WR Wolf Rock Moraine front in the Connecticut Valley. Along the Cold River near Walpole a saprolite and overlying colluviated soil and rock debris are preserved beneath lake beds. This section has 3.3 Initial advance of Late Wisconsinan ice been interpreted (Ridge, 1988, 1999) as a sequence produced by Sangamonian (MIS 5) weathering that was Relatively little is known about the period of advance of followed by mass wastage during the Early to Middle Late Wisconsinan ice across New England because it is Wisconsinan (MIS 3 and 4), and lake ponding during the often difficult to distinguish sub-till deposits put down advance of Late Wisconsinan ice (MIS 2). This is the during the advance of the last ice sheet from deposits of simplest explanation but stands as a hypothesis untested by earlier glaciations. Pre -glacial weathering of sub-till numerical dating techniques. In all the river valleys deposits has been the primary way of identifying deposits mentioned above, sub-till lake beds have been found from earlier glaciations. A lack of weathering in sub-till extending below modern stream levels indicating a greater deposits, however, leaves one guessing although it has been degree of valley dissection just prior to lake ponding than temptingto interpret non-weathered fluvial and lacustrine exists today. deposits as representing the advance of Late Wisconsinan In some of the valleys mentioned above it is difficult ice. Another difficulty has been determining the chronology without detailed mapping to determine whether some of the 174 John C. Ridge

Fig. 6. The late Wisconsinan deglaciation chronology of the northeastern United States in atmospheric 14C ka BP. Arrows indicate ice front positions that are the limits of glacial readvances. The abbreviations of end moraines, ice margins and glacial readvances are given in Table 2.

unweathered sub-till lake beds are associated with local re- Along the perimeter of the ice sheet there are very few 14C advances during Late Wisconsinan ice recession. Along the ages from terrestrial plant fossils indicating when ice Sugar River near Claremont (Fig. 1; Ridge, 2001) palaeo- reached its maximum and chronologies have had to rely magnetism has provided a test of a possible Late Wiscon- heavily on 14C ages from bulk sediment samples and marine sinan age. The lacustrine unit in the Sugar River valley is shells. 14C ages from wood, bulk sediment and marine over 20 m thick and is buried beneath over 20 m of Late fossils in southern New England and on Long Island Wisconsinan till. As is the case with most sub-till lake beds indicate that the ice reached its maximum at about 24.0- the top of the Sugar River unit is highly deformed by 20.0 14C ka BP (28.0-23.7 cal ka BP; B. Stone & Borns, overriding ice. However, the basal 10 m of the unit escaped 1986). The 14C age of 24.0 ka BP is beyond the reach of the deformation by overriding ice other than compaction. INTCAL98 data set (Stuiver et al., 1998) and the age of Orientated palaeomagnetic samples of laminated silt and 28.0 cal ka BP is given as an approximate projection clay collected from the base of the unit yielded a remanent beyond the limits of precise calibration. declination of 23.0º (a 95 = 2.4º, n = 16). Lacustrine units Ice did not reach its maximum extent simultaneously deposited during Late Wisconsinan deglaciation of the across southern New England and into New Jersey (Figs 6 Claremont area have a distinctly different remanent and 7). On eastern Long Island the Ronkonkama Moraine declination of 320-340º. marks the terminal position. This moraine is truncated to the west by the Harbor Hill Moraine (Sirkin, 1982, 1986), 3.4 The Late Wisconsinan maximum which appears to be continuous with the terminal moraine in New Jersey (Salisbury, 1902; Cotter et al., 1986; Along the southern coast of New England large end Stanford & Harper, 1991; Stanford, 1993b; B. Stone et al., moraines represent the terminal position of the last ice sheet 2002). If the interpretation of moraines on Long Island is from Long Island eastwards onto the continental shelf correct, it implies that the terminal position of south-eastern south-east of Cape Cod (Figs 6 and 7; Oldale, 1982; Sirkin, New England is somewhat older than the terminal position 1982, 1986; B. Stone & Borns, 1986; Uchupi et al., 2001). in New Jersey. New England 175

Fig. 7. The late Wisconsinan deglaciation chronology of the northeastern United States in calibrated (U-Th) ka BP. Arrows indicate ice front positions that are the limits of glacial readvances. The abbreviations of end moraines, ice margins, and glacial readvances are given on Table 2.

3.5 Deglaciation of southern New England (20.0-14.6 14C Jersey may be the Ogdensburg and Culvers Gap Moraines ka BP, 23.7-17.5 cal ka BP) (Minard, 1961; Sirkin & Minard, 1972) and the Bloomfield ice margin (Stanford & Harper, 1991; Stanford, 1993b; B. Early ice recession along the southern coast of New Stone et al., 2002). England was in large glacial lakes ponded on an A second large recessional position is marked by isostatically-depressed land surface between end moraines submarine moraines along the southern coast of Connecti- and high areas of the continental shelf to the south, and the cut (Figs 6 and 7) that come on land to form the Old receding ice sheet and land surface to the north. Evidence Saybrook Moraine in Connecticut (Goldsmith, 1982; J. of these lakes is found on the modern land surface of Stone et al., 1998a, 1998b) and the Wolf Rocks Moraine in southern Connecticut and Rhode Is land (Antevs, 1928; Rhode Island (Boothroyd et al., 1998). A contemporaneous Boothroyd et al., 1998; J. Stone et al., 1998b) and below position further east may be the Sandwich Moraine on Cape sea level between the coastal islands and mainland of Cod, which truncates the eastern extension of the Buzzards southern New England (Lewis & Stone, 1991; J. Stone et Bay Moraine (Oldale 1982). Boothroyd et al. (1998) project al., 1998b). Two major recessional positions also mark an age for these features of 17.1 14C ka BP (20.4 cal ka BP) early ice recession in southern New England. The Sand based on correlation to the GISP2 ice core. In northern New Point to Fishers Island Moraines on Long Island (Sirkin, Jersey contemporaneous features may be the Sussex 1982, 1986), which appear to truncate the Harbor Hill Moraine and an ice margin formed in northern glacial Lake Moraine on western Long Island, have been traced eastward Hackensack that extends through New York City (Stanford across the ocean floor to the Charlestown Moraine in Rhode & Harper 1991; Stanford 1993b; B. Stone et al., 2002). A Island (Schafer & Hartshorn, 1965; Boothroyd et al., 1998). prominent lithological break and provenance change occurs These features may be correlative with the Buzzards Bay 1097 varves above the bottom of the Hackensack varve Moraine on Cape Cod (Oldale, 1982). Boothroyd et al. sequence (at ‘0’ varve of Reeds, 1926). This boundary may (1998) have proposed an age of 17.9 14C ka BP (21.3 cal ka represent northward recession of ice into New York to a BP) for these features by correlation to the GISP2 ice core. point where level dropped as Sparkill Gap became the Approximately contemporaneous ice front positions in New northward-draining spillway for a non-glacial lake in the 176 John C. Ridge

Hackensack Valley (Stanford & Harper, 1991). If this is eventually become the drift dam for glacial Lake Hitchcock true ice recession from the Bloomfield ice margin in New (J. Stone et al., 1982, 1998b; Stone & Ashley, 1992; Stone, Jersey to just over the New York border took more than 1995). The Lake Hitchcock dam is older than the oldest 1100 years. measured varve behind it from Lake Hitchcock (varve 2868 In the Hudson Valley of southern New York varve of, Antevs, 1928; Figs 2 and 3). The latter was measured in records near Haverstraw and in the Newburgh area place surface exposures near Hartford at a site where there are an constraints on the age of deglaciation. The 700 varves unknown number of varves in the subsurface. The measured by Antevs (1928), near Haverstraw, do not calibrated age of varve 2868 implies a minimum age of appear to overlap the New England varve chronology 15.3 14C ka BP (18.3 cal ka BP) for the construction of the (lower Connecticut varves). Therefore, the oldest Haver- Lake Hitchcock dam at Rocky Hill. An estimated age of straw varve recorded by Antevs is at least 700 yr older than 16.0 14C ka BP (19.1 cal ka BP) for this event (J. Stone et varve 2701 of the lower Connecticut varves. The oldest al., 1998b) is compatible with ages of surrounding features measured varve at Haverstraw, which is not the base of the determined from ice-core correlations (Boothroyd et al. varve sequence at this locality, must then have an age of at 1998) and the 14C calibration of varves in the Connecticut least 16.0 14C ka BP (19.1 cal ka BP, Figs 6 and 7). This is Valley. the youngest possible age for deglaciation at Haverstraw. The Rosendale Readvance in the Hudson Valley (Figs 6 Near Newburgh Antevs (1928) measured the oldest and 7; Connally, 1968; Connally & Sirkin, 1973, 1986) is varve of the New England varve chronology (varve 2701) the youngest readvance in the Hudson Valley that pre-dates but he was not able to measure to the bottom of varve post-15.0 14C ka BP readvances in central New York that section in this area. The occurrence of thick varves (8-20 have been palaeomagnetically-correlated to the New cm) and bedrock not far beneath Antevs’ measured section England varve chronology (Fig. 5). East of the Hudson suggest that varve 2701 is not significantly younger than Valley the limit of the Rosendale Readvance appears to the age of deglaciation. Varve 2701 has an age of 15.4 14C match the Red Hook Moraine (Connally & Sirkin, 1986). ka BP (18.4 cal ka BP), which is also the youngest possible The limit of the Rosendale Readvance to the west follows age for deglaciation at Newburgh (Figs 6 and 7). This is the flank of the Catskill Mountains (Connally & Sirkin, approximately the position of the Pellets Island and 1973; Dineen, 1986) and is interpreted to be equivalent to Shenandoah Moraines (Connally & Sirkin, 1970, 1973, the Cassville-Cooperstown Moraine (Krall, 1977; Fleisher, 1986). 1986) and West Canada Readvance of central New York On Cape Cod and in south-eastern Massachusetts (Figs (Ridge et al., 1991; Ridge & Franzi, 1992; Ridge, 2003). 6 and 7) ice simultaneously ponded glacial Lakes Taunton Palaeomagnetic declinations of 20-30º from lake beds and Cape Cod (Larson, 1982; B. Stone & Peper, 1982). The associated with recession from the West Canada Readvance age of this ice margin can only be bracketed from the ages (upper Newport Beds of Ridge et al., 1990, 1991) do not of surrounding ice margins. Further west in Connecticut match any of the declination records from the lower two ice-front positions have been reconstructed that mark Connecticut varves. The West Canada Readvance therefore the limit of ice when the ice front crossed the Quinnipiac appears to pre-date varve 3000 of the New England varve Valley, near New Haven, and when the dam for glacial chronology giving the readvance and correlative features a Lake Hitchcock was constructed at Rocky Hill (J. Stone et minimum age of 15.2 14C ka BP (18.2 cal ka BP). Deposits al. 1998a, 1998b). These ice front positions appear to associated with the West Canada Readvance, in central slightly post-date the Haverstraw ice margin and pre-date New York, are truncated by a subaerial erosion surface, the the ice margin at Newburgh in the Hudson Valley. The Shed Brook Discontinuity (Ridge, 1997). Associated with Connecticut ice margins are approximately the same age as the unconformity is a fluvial gravel unit in the western the ice margin at Lakes Taunton and Cape Cod although Mohawk Valley (Little Falls Gravel) that pre-dates geographically the Quinnipiac ice margin may be a better lacustrine units and diamictons (Poland Formation of Ridge correlation. Varves measured by Antevs (1928) in the et al., 1990) associated with later glacial readvances (Ridge, Quinnipiac Valley do not overlap the New England varve 1997). These features indicate a period of ice recession in chronology. The oldest Quinnipiac varve measured by the Mohawk and Hudson Valleys that allowed unrestricted Antevs, which is not the bottom of the Quinnipiac varve river drainage eastwards from the Ontario Basin. section (Figs 2 and 3), is at least 500 yr older than varve 2701 of the lower Connecticut varves. This corresponds to a minimum age of 15.8 14C ka BP (18.9 cal ka BP) for 3.6 Deglaciation from central to northern New England deglaciation at the Quinnipiac ice margin. The estimated (14.6-12.0 14C BP, 17.5-14.1 cal BP) age of J. Stone et al. (1998a) for this ice margin (16.5 14C ka BP, 19.7 cal ka BP) is consistent with the varve-based The precise age of deglaciation along the Connecticut chronology and the interpreted ages of surrounding Valley, from southern Massachusetts to northern New features. Hampshire and Vermont (Fig. 8), along the axis of Lake As ice receded through central Connecticut (Figs 6 and Merrimack in eastern New Hampshire (Fig. 9) and in the 7) an ice contact delta was deposited in Lake Middletown Winooski Valley of north-central Vermont (Larsen, 1987b) in the Connecticut Valley at Rocky Hill that would is known from the calibrated ages of varves in the New

New England 177

Table 3. Rates of deglaciation from southern Massachusetts to northern New England in the Connecticut and Merrimack Valleys.

Segment location Distance (km) New England varve Approximate 14C- Approximate cal.- Rate of ice recession year numbers (span) year span (yr BP) year span (yr BP) (m/varve yr)

Recession from southern 37 <2868-3600 (>732) >15,300-14,650 >18,300-17,550 maximum of 20 Lake Hitchcock basin at Hartford, Conn. (site 139, Antevs, 1928) to Connecticut/ Massachusetts border

Recession from Chicopee 20 3900-4600 (700) 14,400-13,800 17,250-16,550 29 Readvance. Holyoke to Amherst, Massachusetts

Recession after delay or 30 4950-5250 (300) 13,500-13,200 16,200-15,900 100 Readvance at Camp Meeting Cutting, Massachusetts to New Hampshire/Massachusetts border

New Hampshire/ 60 5250-6000 (750) 13,200-12,600 15,900-15,150 80 Massachusetts border to Charlestown, New Hampshire

Merrimack Valley: Suncook, 30 5700-6000 (300) 12,800-12,600 15,400-15,100 100 New Hampshire to south of Franklin, New Hampshire

Recession following 125 6600-7075 (475) 12,400-11,900 14,425-13,950 263 Claremont readvances and moraines. Claremont, New Hampshire to St. Johnsbury, Vermont

Recession from Littleton- 60 7200-<7450 (<250) 11,800->11,550 13,800->13,550 minimum of 240 Bethlehem Moraine (Readvance) at Littleton, New Hampshire to Columbia Bridge, New Hampshire

Locations are given on Figures 2, 8 and 9. Approximate 14C and calibrated ages from Figure 4.

England varve chronology (Figs 6 and 7). Thick, sandy Massachusetts to north-central New England (Table 3). basal varves representing an ice-proximal environment, and Limiting rates of ice recession can also be determined for in some cases seen resting on till, bedrock or coarse ice- northern Connecticut and northernmost New Hampshire contact deposits at numerous sites along the axes of these and Vermont. Correlations have been formulated between valleys, have been matched to the New England varve ice-front positions in New York and New England by chronology (Antevs, 1922, 1928; Ridge et al., 1996, 1999, matching palaeomagnetic declination records (Fig. 5; Ridge 2001; Larsen et al., 2001; Ridge, 2003). Basal varves are et al., 1990; Ridge, 2003). approximately contemporaneous with ice recession and The Chicopee Readvance in southern Massachusetts depict a general northward onlap of the varve sequence in (Figs 6, 7 and 8A; Larsen & Hartshorn, 1982) has a the wake of receding ice. The basal varve stratigraphy and minimum age of 14.5 14C ka BP (17.4 cal ka BP) as its 14C calibration allow a recession chronology of the last indicated by the oldest recorded varves in the area of the ice sheet to be established with a high degree of precision. readvance. The readvance could be the same age as thick Some readvances can also be bracketed by the varve sandy varves (varves 3600-3750) just south of the sequence. The varve chronology also allows precise varve- readvance limit (Ridge, 2003). Based on palaeomagnetic year estimates of the rate of deglaciation for valley correlations (Ridge et al. 1990; Ridge, 2003) the Salisbury segments between readvances and still stands from southern Readvance of central New York (Ridge & Franzi, 1992; 178 John C. Ridge

Fig. 8. Time-distance plots (a-c) of the time spans of all measured varve sections in the Connecticut and Passumpsic Valleys from Connecticut to northern Vermont and New Hampshire. The distance axis on all plots is measured from an arbitrary position near Hartford, Connecticut. The time axes are New England varve chronology years. The dashed line marking the base of sections resting on till is the approximate age of deglaciation. Key to symbols is given on Fig. 8c. (a.) Connecticut to northern Massachusetts. New England 179

Fig. 8b. Time-distance plots (a-c) of the time spans of all measured varve sections in the Connecticut and Passumpsic Valleys from Connecticut to northern Vermont and New Hampshire (continued).

(b) Northern Massachusetts to central New Hampshire and Vermont. 180 John C. Ridge

Fig. 8c. Time-distance plots (a-c) of the time spans of all measured varve sections in the Connecticut and Passumpsic Valleys from Connecticut to northern Vermont and New Hampshire (continued).

(c.) Northern New Hampshire and Vermont.

Ridge, 2003) and its correlative Middleburg Readvance 1990; Ridge, 2003), the Barneveld-Little Falls Readvance (Fleisher, 1986; Dineen, 1986) in the Hudson Valley (14.5 (13.5 14C ka BP, 16.2 cal ka BP) in central New York 14C ka BP, 17.4 cal ka BP) and the Hinckley-St. Johnsville (Ridge & Franzi, 1992; Ridge, 2003) and its equivalent in Readvance (Ridge & Franzi, 1992; Ridge, 2003; 14.35 14C the Hudson Valley, the Delmar Readvance (Dineen, 1986), ka BP, 17.2 cal ka BP) appear to be approximately occurred at about the time ice was at the Camp Meeting contemporaneous with the Chicopee Readvance. Cutting margin in New England. In eastern Massachusetts The Camp Meeting Cutting Readvance, in the ice recession in Lake Charles (Stone & Peper, 1982) and Connecticut Valley (Emerson, 1898; Larsen & Hartshorn, the Cambridge Readvance (Shafer & Hartshorn, 1965), 1982), may be represented by a slight delay in ice recession which created the Fresh Pond Moraine (Chute, 1959; Kaye, between varves 4700-4900 (Fig. 8A) but it did not reach as 1961), occurred at about 14.6-13.5 14C ka BP (17.5-16.2 cal far south as Amherst (Antevs, 1928; Ridge, 2003) where ka BP) given their positions relative to other ice front readvancing ice would have interrupted a continuous positions (Figs 6 and 7). sequence of varves 4682-6027 (Rittenour, 1999). The age Calibration of the New England varve chronology has of the potential Camp Meeting Cutting Readvance (Figs 6 allowed the application of terrestrial 14C ages to basal and 7) is constrained to about 13.6 14C ka BP (16.3 cal ka varves (Fig. 9) and ice recession in the Merrimack Valley BP). Based on palaeomagnetic correlations (Ridge et al. (Figs 6 and 7). Nearly continuous tracing of ice-front New England 181 Fig. 9. Time-distance plot of time spans of all measured varve sections in the Merrimack Valley of New Hampshire. Symbols are the same as for Fig. 8.

positions from the Merrimack Valley across eastern New varves by Gerard De Geer (varves 5669-5678 of Antevs Hampshire (Koteff et al. 1993) to southern Maine (Smith 1922; Rittenour, 1999) appear to represent the catastrophic 1980, 1982, 1985; Thompson & Borns 1985; Smith & release of water into the Hudson Valley from a low level Hunter 1989) has allowed the tracing of terrestrial ice front Mohawk Valley glacial lake (Lake Amsterdam of LaFleur, positions to a region where the last ice sheet receded in 1979, 1983). The flood units have an age of 12.9 14C ka BP marine water (Fig. 10). Previously, marine 14C ages from (15.5 cal ka BP) and represent the final blockage of the southern Maine were used to propose the age of eastern Mohawk Valley by Hudson Valley ice (Figs 6 and deglaciation in southern New Hampshire (cf. Koteff et al., 7). Shortly after this was the Luzerne Readvance in the 1993 for review). There appears to be some inconsistency upper Hudson Valley (Connally & Sirkin, 1971, 1973). The among marine 14C ages from ice-proximal deposits of Luzerne Readvance here has been assigned an age of 12.6 southern Maine suggesting that the marine ages may have 14C ka BP (15.3-15.2 cal ka BP) based on what appear to be limited accuracy for determining the age of deglaciation geographically-equivalent readvance positions in the Con- (Fig. 10). All except two of the uncorrected marine 14C ages necticut Valley near Claremont (Ridge, 2001, 2003) and the are 600-1300 yr older than the ages for single ice front correlation of the Luzerne Readvance by Muller & Calkin positions inferred from varve stratigraphy calibrated with (1993) with the Rome Readvance to the west. The age terrestrial 14C ages (Ridge, 2003 as revised from Ridge et proposed here is in agreement with a 14C age of 12,400 + al., 2001). This disparity between marine and terrestrial 14C 200 yr BP (I-3199) obtained on wood fragments 15 cm chronologies suggests that marine reservoir errors are from the bottom of a kettle depression formed in outwash inconsistent across the region and through time (Retelle & of the Luzerne Readvance (Connally & Sirkin, 1971). A 14C Weddle, 2001) and generally higher than a standard age of 13,150 + 200 BP yr (I-4986) obtained 3 cm from the correction of 400 years. In this author’s view the bottom on twelve pooled samples of lacustrine sediment is application of a 400-yr reservoir correction to marine 14C older than the age presented here. ages in southern and eastern Maine (Dorion et al., 2001) is a move in the right direction but it does not completely resolve the discrepancy between the marine and terrestrial 3.7 Connecticut Valley lakes 14C chronologies. The recession of ice into the Connecticut Valley of New As mentioned previously deposition of a delta in Lake Hampshire and Vermont (Figs 6, 7 and 8B) began at about Middletown at Rocky Hill in southern Connecticut (Flint, 13.3 14C ka BP (16.0 cal ka BP) and continued to 1933, 1953; J. Stone et al., 1982, 1998b; Stone & Ashley, Charlestown, N.H. by about 12.6 14C ka BP (15.2 cal ka 1992) led to the ponding of Lake Hitchcock (Fig. 2; BP) where detailed mapping has revealed three end- Lougee, 1935a; 1939, 1957). A stable bedrock spillway moraine systems all associated with small readvances from eventually developed at New Britain, Connecticut Charlestown to Claremont (Ridge, 2001). Based on (Loughlin, 1905; Flint, 1933; Jahns & Willard, 1942; Stone, palaeomagnetic correlations (Fig. 5) the period of delayed 1995; J. Stone et al., 1998b) that allowed Lake Hitchcock to ice recession near Claremont appears to match the Rome persist until ice receded much further north. Recent Readvance of central New York (Muller et al., 1986; controversy has focused on the position of the receding ice Muller & Calkin, 1993; the Ninemile Readvance of Ridge front when the stable phase of the lake drained due to & Franzi, 1992). Continued rapid ice recession into the partial failure of the Rocky Hill dam. The beginning of northern Connecticut Valley from 12.5-12.0 14C ka BP differential isostatic uplift may have triggered this event (14.7-14.1 cal ka BP) is recorded by onlapping basal varves (Stone, 1995, 1999). that have been matched to the upper Connecticut varves of At the southern end of Lake Hitchcock organic-rich the New England varve chronology (Fig. 8C). sandy lake beds capping clayey varves appear to represent In the Hudson Valley non-annual flood units in the the sudden drainage of the stable phase of Lake Hitchcock varve stratigraphy that were originally misidentified as down to a shallow lake (Stone & Ashley, 1992; Stone, 182 John C. Ridge Fig. 10. Ice-front positions in southern New Hampshire and southern Maine from the Merrimack valley eastward across the coastal area of Maine where ice was fronted by marine water during recession. Shaded area in Maine is the region within the marine limit. This area had many islands projecting above glaciomarine submergence that are not shown on the map. Ice- front ages (14C ka BP) in the Merrimack Valley are based on the terrestrial 14C calibration of the New England varve chronology as applied to basal ice-proximal varves (Fig. 9). 14C ages (ka BP) in Maine and off the coast of New Hampshire are uncorrected ages from marine fossils that have been interpreted to closely approximate the time of deglaciation.

1995). A range of 14C ages from plant fossils in this Lake Hitchcock deltas in this valley were dissected by deposit and from the bottoms of pingo scars at the surface spillway drainage from Lake Winooski when water levels of the lake bed (Stone & Ashley, 1992; J. Stone et al., fell in the Connecticut Valley (Larsen, 1984; 1987a, 1987b) 1998b) suggest a lowering of the lake as early as 13.5 14C indicating that Lake Winooski was in existence during the ka BP (16.2 cal ka BP). The receding ice front would have lowering of Lake Hitchcock. In his original assessment of been in northern Massachusetts at this time (Figs 6 and 7). Lake Hitchcock, Lougee (1939, 1957) traced stable phase Ridge et al. (1999) have suggested that the drainage event deltas as far north as Lyme, New Hampshire where he may have been later if the oldest 14C ages were from plant thought the ice front stood when the lake drained to a much fossils that had been reworked by valley incision following lower level (his Lake Upham). Finally, Koteff & Larsen the lowering of lake level. Deltas reported in New (1989) recorded a string of stable phase Lake Hitchcock Hampshire and Vermont, which fall on a projected plane deltas from Massachusetts to north of St. Johnsbury, for stable Lake Hitchcock, suggest a later time for drainage Vermont, but some of these deltas in the northern of the stable phase of Lake Hitchcock (Koteff & Larsen, Connecticut Valley may be deposits formed in local 1989). Detailed superficial mapping in southern New tributary lakes. In addition drainage features associated Hampshire and Vermont has identified stable phase deltas with the margin of the Littleton-Bethlehem Readvance in as far north as Claremont (Ridge, 1990, 1999, 2001). Stable the Connecticut Valley appear to be lower than the phase deltas have also been traced up a long embayment of projected stable phase of Lake Hitchcock. Detailed Lake Hitchcock in the White River valley (Larsen, 1987a). mapping will be needed in the upper Connecticut Valley to New England 183 Fig. 11. The GISP2 (Greenland Ice Sheet Project 2) ice-core record of oxygen isotopes (Cuffey et al. 1995; Stuiver et al. 1995) plotted with the history of deglaciation in New York and New England presented here. The positions of peaks plotted for the early period of glaciation and deglaciation (RK to WC-CC- RO) are proposed correlations to the ice core record by Boothroyd et al. (1998) with some additions new to this paper. The positions of peaks on the later deglaciation record (following Shed Bk. Disc.) were determined using the precise deglaciation model of the calibrated New England varve chronology. The abbreviations of end moraines, ice margins and glacial readvances are given on Table 2. The time scale for the oxygen isotope record is based on annual layer counts in the GISP2 ice core. The time scale for the deglaciation chronology is the calibrated (U-Th) time scale of events in New York and New England discussed in text. A detailed discussion of the correspondence of the glacial and post-glacial varve stratigraphy of the upper Connecticut Valley (extending to 12.4 cal ka BP) to the ice core record is given in Ridge & Toll (1999).

resolve the occurrence of stable phase deltas from Lake Bethlehem Readvance that is bracketed by the New Hitchcock. England varve chronology (Fig. 8C; Antevs, 1922; Lougee, Regardless of where the ice front was when Lake 1935b; Ridge et al., 1996, 1999; Ridge, 2003). A large Hitchcock’s stable phase drained, lower lake levels have moraine system was deposited at the margin of the been recorded in the valley north of Massachusetts (Koteff readvance from Bethlehem, N.H. to St. Johnsbury, Vt. & Larsen, 1989; Cold River stage of Ridge, 1990, 1999, (Goldthwait, 1916; Crosby, 1934a) at 11.9-11.8 14C ka BP 2001). In the Connecticut Valley from Lyme northward a (13.9-13.8 cal ka BP; Ridge et al., 1996, 1999; Thompson low level lake persisted until at least 10.5 14C ka BP (12.4 et al., 1996, 1999). The Littleton-Bethlehem Readvance cal ka BP) as recorded by the long varve section at New- occurs at the time of the Older Dryas event (Fig. 11; bury (Antevs, 1922; Ridge & Toll, 1999; Ridge et al., 1999; Thompson, 1998; Ridge & Toll 1999; Thompson et al., Ridge, 2003). The first humans entering the Connecticut 1999) in ice core records from Greenland and adds to Valley may have witnessed this water body (Ridge, 2003). existing evidence of synchronous cooling in the North Recession of ice from the Littleton-Bethlehem Readvance Atlantic region (Dansgaard et al., 1989, 1993; Stuiver et al., in the Connecticut Valley (see following section) was in 1995; Ingólfsson et al., 1997; Björck et al., 1996, 1998b). glacial Lakes Coos (Lougee, 1939) and Colebrook (Fig. 2; In north-central Vermont the Middlesex Readvance buried Ridge et al., 1996; W. Thompson, personal communication; wood with a 14C age of 11,900 + 50 yr B.P. (GX-26457; Ridge, 2003). Both of these water bodies are above the Larsen, 2001). projected level of any stage of Lake Hitchcock but little In the Champlain Valley the Bridport Readvance work has been done to establish the exact elevation, lake (Connally & Sirkin, 1973) seems to have an advance limit level history, or possible connection of these two water that joins the Middlesex Readvance limit to the east (Figs 6 bodies. and 7). The position of the Bridport Readvance relative to

varve sequences (younger than varves 6601-7000, older 3.8 Final deglaciation and the Champlain Sea (after 12.0 than Essex varve sequence of Antevs, 1928) and Champlain 14C ka BP, 14.1 cal ka BP) Valley palaeomagnetic measurements (Fig. 5; Ridge et al., 1999) indicate that it is close in age to readvance events to The final phase of deglaciation in northern Vermont and the east. On the north-west side of the Adirondacks, the New Hampshire (Figs 6 and 7) began with the Littleton- Carthage-Harrisonville ice margin (Pair & Rodrigues, 184 John C. Ridge

1993) and Star Lake Moraine (Clark & Davis, 1988) may 3.9 Post-glacial isostatic adjustment be approximately the same age as the Bridport Readvance. 14C ages that have been used to estimate the age of features Detailed leveling surveys of glacial and meteoric deltas in north-east of the Adirondacks (about 12.5 14C ka BP) are the Connecticut Valley (Jahns & Willard, 1942; Koteff & from bulk sediment samples of lake-bottom sediment and Larsen, 1989) have revealed the character of post-glacial are incompatible with the assigned ages of features to the isostatic tilting across western New England. Koteff & south in central New York. The bulk sediment ages are the Larsen (1989) defined a flat, tilted (0.889 m/km, 159º) same as or older than 14C ages from spruce twigs marking water plane showing the alignment of both ice-contact and the initiation of Lake Iroquois in the Ontario Basin, which meteoric deltas from the stable phase of Lake Hitchcock pre-dates moraine deposition north-west of the Adiron- (especially Cold River stage of Ridge, 1990, 1999, 2001). dacks. A younger age for the moraines in the north-western This water plane is also parallel to planes defined by lakes Adirondacks is projected here as 12.2-11.8 14C ka BP (14.1- that preceded and succeeded the stable phase of Lake 13.8 cal ka BP). Hitchcock indicating that isostatic tilting did not commence until all these later features were formed. Koteff & Larsen In eastern Maine ice readvanced to form the Pineo (1989) have used this data to document a delay in isostatic Ridge delta and moraine complex (Figs 6 and 7), which has adjustment, until ice receded to northern New Hampshire been given an age of about 13.0 14C ka BP based on and Vermont. A similar pattern of uplift has also been uncorrected marine 14C ages (Stuiver & Borns, 1975; documented for glacial Lakes Sudbury and Nashua in Thompson & Borns, 1985; Kaplan, 1999). If a marine eastern Massachusetts (0.90-0.94 m/km) and Lake reservoir correction of about 1000 yr (discussed previously Merrimack in New Hampshire (about 0.9 m/km; Koteff, for southern Maine) is applied to the marine 14C age of the 1982; Koteff et al., 1993). Elevations of glaciomarine deltas Pineo Ridge Readvance it would change its age to about on the southern coast of Maine, features contemporaneous 12.0 14C ka BP (14.05 cal ka BP), or approximately equal to with Lake Merrimack deltas, support the delayed rebound the Littleton-Bethlehem Readvance in northern New hypothesis because they were submerged by sea-level rise Hampshire and Vermont and the Older Dryas Event. At this that out paced isostatic uplift during ice recession (Koteff et point such a projection is speculative and based solely on al., 1993). Lake Vermont (Fig. 2) and the Champlain Sea in the comparison of terrestrial and marine 14C ages in the Champlain Valley have almost parallel water planes southern Maine (Fig. 10). with a tilt (0.85-0.86 m/km, about 165°, Parent & Occhietti, The final recession of ice from the Connecticut Valley 1988) similar to older lakes further south in New England. into Canada was accomplished before 11.5 14C ka BP (13.5 A projection of the earliest Champlain Sea water plane cal ka BP), which is the age of terrestrial plant macrofossils (12.0-11.0 14C ka BP, 14.1-13.0 cal ka BP) to the south in varves, deposited after deglaciation at Columbia Bridge shore of Long Island on the southern coast of New England (Miller & Thompson, 1979), near the Canadian border (Fig. indicates that tilting in southern and central New England 2). Rapid recession of ice into Canada occurred at about the started much earlier than in the Champlain Valley. The age time that water was released from Lake Iroquois at Covey range used here covers the whole range of uncorrected Hill into Lake Vermont in the Champlain Valley (Parent & marine to terrestrial 14C ages for the Champlain Sea Occhietti, 1988; Pair & Rodrigues, 1993). An age of 11.7- invasion. A southward projection of this water plane at 14 11.4 C ka BP (13.7-13.4 cal ka BP; Figs 6 and 7) has been 12.0-11.0 14C ka BP has to meet sea level on the south assigned to this ice margin based on its relationship to shore of Long Island –90 to –65 m below modern sea level varves at Enosburg Falls in the Missisquoi Valley of (Fairbanks, 1989; Bard et al., 1990). If isostatic recovery on northern Vermont (Ridge et al., 1999; Ridge, 2003). Soon the southern coast of New England was not yet complete after the recession of ice into Canada, lakes in the St. the Champlain Sea’s projected water plane would have to Lawrence and Champlain Valleys were lowered to global end higher than eustatic sea level at 12.0-11.0 14C ka BP sea level as marine water of the Champlain Sea invaded because of lingering isostatic submergence. The elevation these isostatically-depressed basins. Palaeomagnetic mea- of the highest Champlain Sea strandline at the Vermont surements from sediment recording the transition from /Quebec border is 155 m and the distance from this position lacustrine to marine deposits in these basins (Fig. 5) show a to the south shore of Long Island in the direction of declination of 0º at the time of a transition from a western isostatic tilt (165º) is 460 km. For the 155-m strandline to to eastern declination (Pair et al., 1994; Ridge et al., 1999). join sea level at -90 to -65 m south of Long Island the Palaeomagnetic correlation of this event to the New average tilt of the projected Champlain Sea water plane is England varve chronology (Ridge et al., 1999; Ridge, 0.53-0.47 m/km. The southern end of this curved projection 2003) indicates an age range of 11.1-10.6 14C ka BP (13.1- has an even gentler tilt to accommodate the northern end of 12.7 cal ka BP). This age estimate is crude given the quality the projection that starts with a tilt of 0.86 m/km (Parent & of the palaeomagnetic data and compression of the 14C time Occhietti, 1988). The average tilt (0.53-0.47 m/km) is much scale at this time but it overlaps age estimates for the gentler than water planes for earlier lakes in central to Champlain Sea invasion that have attempted to eliminate southern New England (0.88-0.94 m/km, Koteff et al., 14 marine C reservoir errors (Anderson, 1988; Rodrigues, 1993) and New Jersey (0.66 m/km, Stanford & Harper, 1988, 1992). 1991). The divergence of strandline projections indicates

New England 185 that they did not experience the same amount of post- Davis et al., 1980; Davis & Jacobson, 1985; Hughes et al., glacial tilt and tilting was well underway in southern and 1985; B. Stone & Borns, 1986; Dyke & Prest, 1987). central New England by the time the Champlain Sea The early part of Late Wisconsinan deglaciation was invaded the Champlain Valley at 12.0-11.0 14C ka BP. relatively slow from the Late Wisconsinan limit through Connecticut, Rhode Island to south-eastern Massachusetts. Later deglaciation of Massachusetts and northern New 4. Conclusions England was very rapid and covered a greater distance (Table 3). Overall, deglaciation accelerated from south to 4.1. Pre-Late Wisconsinan glaciation north and occurred in rapid phases punctuated by readvances and end-moraine construction. This pattern is Based on the degree of weathering and extent of pre- borne-out by matching the chronology of the later half of Wisconsinan glacial deposits in areas south-west of New deglaciation (18,000-13,000 cal BP) with the pattern of England in New Jersey (B. Stone et al., 2002), it appears oxygen isotope variations (Dansgaard et al., 1989, 1993; that New England was glaciated at least twice during pre- Cuffey et al., 1995; Stuiver et al., 1995) seen in Greenland Wisconsinan time. Till beneath MIS 5-equivalent ice cores (Fig. 11; Ridge, 2003). The two records can be interglacial deposits on Nantucket and the apparently compared by aligning their independent time scales. A equivalent Montauk Till of Long Island appear to record an matching of these records is expected if the ice core records Illinoian (MIS 6) glaciation (Oldale, 1982; Oldale & represent a proxy for climate change around the North Colman, 1992). So far no deposits in New England have Atlantic region (Björck et al., 1998b). The earlier part of been correlated to pre-Illinoian deposits of New Jersey (B. the deglaciation record in New England (prior to 18,000 cal Stone et al., 2002) although equivalents may occur on BP) cannot yet be independently matched to the ice-core Martha’s Vineyard, beneath Boston Harbor and on the record because the numerical ages of events in southern continental shelf. Further inland the picture is not so clear. New England are not independently known with high Weathered till beneath Late Wisconsinan deposits in precision. The early deglaciation record shown in Fig. 11 is Massachusetts and in northern New England has been composed of previously proposed correlations in southern treated as a single unit and assigned an Illinoian age (Koteff New England (OS-WR and FI-CM according to Boothroyd & Pessl, 1985; Weddle et al., 1989; Newman et al., 1990; et al., 1998), with other events added. The existing Oldale & Colman, 1992; Newman et al., 1993; Newman & constraints on the ages of early deglaciation events in New Mickelson, 1994). This age conflicts with amino-acid England, although not very strict, are consistent with this ‘ages’ for shells in the lower till of Boston Harbor drumlins correlation. that support an Early Wisconsinan age (Belknap, 1980; Colgan & Newman, 1999). References

4.2. Late Wisconsinan glaciation Abbott, M.B. & Stafford, T.W. (1996). Radiocarbon geochemistry of modern and ancient arctic lake systems, The advance of Late Wisconsinan ice is recorded by fluvial Baffin Island, Canada. Quaternary Research, 45, 300- deposits in the Connecticut Valley of southern New 311. Hampshire and the fine-grained deposits of proglacial lakes Adkins, J.F., Cheng, H., Boyle, E.A., Druffel, E.R.M. & impounded by the advancing ice sheet in Connecticut Edwards, R.L. (1998). Deep-sea coral evidence for rapid Valley tributaries of southern New Hampshire and Vermont change in ventilation of the deep North Atlantic 15,400 (Larsen, 1987a; Ridge, 1988, 1999, 2001). The advance of years ago. Science, 280, 725-728. Late Wisconsinan ice reached the southern coast of New Anderson, T.W. (1988). Late Quaternary pollen 14 England as early as 24.0 C ka BP (23.7 cal ka BP) and stratigraphy of the Ottawa Valley-Lake Ontario region was complete by 20.0 14C ka BP (23.7 cal ka BP; B. Stone and its application in dating the Champlain Sea. In: Gadd, & Borns, 1986). N.R. (ed.), The Late Quaternary Development of the The record of Late Wisconsinan ice recession has seen Champlain Sea Basin. Geological Association of Canada, dramatic refinement with the rebirth of the New England Special Paper, 35, 205-224. varve chronology, new AMS 14C ages from terrestrial plant Andrée, M., Oeschger, H., Siegenthaler, U., Riesen, T., fossils and palaeomagnetic records that have allowed Möll, M., Ammann, B. & Tobolski, K. (1986). 14C dating correlation across the region (Ridge et al., 1990, 1996, of plant macrofossils in lake sediment. Radiocarbon, 28, 1999; Thompson et al., 1996, 1999; Ridge, 2003). A 411-416. 14 revised chronology, based on the new varve, C and Antevs, E. (1922). The Recession of the Last Ice Sheet in palaeomagnetic data depicts the deglaciation of central and New England. American Geographical Society Research northern New England 1000-2000 years later than in Series, 11, 120 pp. previous deglaciation chronologies that were largely based Antevs, E. (1925). Retreat of the Last Ice-sheet in Eastern 14 on scattered C ages from lake-bottom bulk sediment Canada. Geological Survey of Canada, Memoir, 146, 142 samples and marine fossil (Shafer & Hartshorn, 1965; pp.

186 John C. Ridge

Antevs, E. (1928). The Last Glaciation with Special T.L., Wohlfarth, B., Hammer, C.U. & Spurk, M. (1996). Reference to the Ice Sheet in North America. American Synchronized terrestrial-atmospheric deglacial records Geographical Society, Research Series, 17, 292 pp. around the North Atlantic. Science, 274, 1155-1160. Antevs, E. (1931). Late-glacial correlations and ice Björck, S., Bennike, O., Possnert, G., Wohlfarth, B. & recession in Manitoba. Geological Survey of Canada, Digerfeldt, G. (1998a). A high-resolution 14C dated Memoir, 168, 76 pp. sediment sequence from southwest Sweden: age Antevs, E. (1935). Telecorrelations of varve curves. comparisons between different components of the Geologiska Foreningens i Stockholm Förhandlingar, 57, sediment. Journal of Quaternary Science, 13, 85-89. 47-58. Björck, S., Walker, M.J.C., Cwynar, L.C., Johnsen, S., Antevs, E. (1939). Modes of retreat of the Pleistocene ice Knudsen, K.-L., Lowe, J.J., Wohlfarth, B. & INTIMATE sheets. Journal of Geology, 47, 503-508. members, (1998b). An event stratigraphy for the last Antevs, E. (1951). Glacial clays in Steep Rock Lake, Termination in the North Atlantic region based on the Ontario, Canada. Geological Society of America Bulletin, Greenland ice-core record: a proposal by the INTIMATE 62, 1223-1262. group. Journal of Quaternary Science, 13, 283-292. Antevs, A. (1953). Geochronology of the deglacial and Bondevik, S., Birks, H.H., Gulliksen, S. & Mangerud, J. neothermal ages. Journal of Geology, 61, 195-230. 14 Antevs, E. (1954). Geochronology of the deglacial and (1999). Late Weichselian marine C reservoir ages at the neothermal ages: a reply. Journal of Geology, 62, 516- western coast of Norway. Quaternary Research, 52, 104-114. 521. Antevs, E. (1962). Transatlantic climatic agreement versus Boothroyd, J., Freedman, J.H., Brenner, H.B. & Stone, J.R. 14C dates. Journal of Geology, 70, 194-205. (1998). The glacial geology of southern Rhode Island. In: Austin, W.E., Bard, E., Hunt, J.B., Kroon, R. & Peacock, Murray, D.P. (ed.), Guidebook to Field Trips in Rhode J.D. (1996). The Icelandic Vedde ash and the Younger Island and Adjacent Regions of Connecticut and Massa- th Dryas marine reservoir factor. Radiocarbon, 37, 53-62. chusetts. 90 New England Intercollegiate Geological Bard, E. (1988). Correction of accelerator mass spectro- Conference, Kingston, Rhode Island, pp. C5: 1-25. metry 14C ages measured in planktonic foraminifera: Borns, H.W., Jr. (1998). The progress of radiocarbon paleoceanographic implications. Paleoceanography, 3, dating. Geological Society of America, Abstracts with 635-645. Programs, 30(1), 6. Bard, E., Hamelin, B., Fairbanks, R.G. & Zindler, A. Braun, D.D., Ciolkosz, E.J., Inners, J.D., Epstein, J.B., 14 (1990). Calibration of the C time scale over the past Clark, G.M., Sasowsky, I.D. & Koberle, R. (1994). Late 30,000 years using mass spectrometric U-Th ages from Wisconsinan to Pre-Illinoian(?) Glacial and Periglacial Barbados corals. Nature, 345, 405-410. Events in Eastern Pennsylvania. Guidebook for 57th Bard, E., Arnold, M., Fairbanks, R.G. & Hamelin, B. Friends of the Pleistocene Field Conference, Bloomsburg 230 234 14 (1993). Th- U and C ages obtained by mass spectro- University of Pennsylvania, Bloomsburg, Pennsylvania, metry on corals. Radiocarbon, 35, 191-199. 118 pp. Bard, E.B., Arnold, M., Mangerud, J., Paterne, M., Brennan, W.J., Hamilton, M., Kilbury, R., Reeves, R.L. & Labeyrie, L., Duprat, J., Mélières, M-A., Sønstegaard, E. Covert, L. (1984). Late Quaternary secular variation of & Duplessy, J-C. (1994). The North Atlantic atmosphere- 14 geomagnetic declination in western New York. Earth and sea surface C gradient during the Younger Dryas Planetary Science Letters, 70, 363-372. climatic event. Earth and Planetary Science Letters, 126, Cadwell, D.H. (ed.), (1989). Surficial geologic map of New 275-287. York, lower Hudson sheet (1:250,000 scale). New York Belknap, D.K. (1979). Application of Amino Acid State Museum, Geological Survey Map and Chart Series, Geochronology to Stratigraphy of Late Cenozoic Marine 40. Units of the Atlantic Coastal Plain. Unpublished Ph.D. Thesis, Univ. of Delaware, Newark, 550 pp. Chute, N.E. (1959). Glacial geology of the Mystic lakes- Fresh Pond area, Massachusetts. United States Geolo- Belknap, D.K. (1980). Amino acid geochronology and the gical Survey Bulletin, 1061-F, 187-216. Quaternary of New England and Long Island. American Quaternary Association, Abstracts and Program, 16-17. Clark, P.U. & Davis, P.T. (1988). Deglacial chronology of Bierman, P.R., Davis, P.T. & Caffee, M.W. (2000). Old the northwestern Adirondack Mountains. Geological surfaces on New England summits imply thin Laurentide Society of America, Abstracts with Programs, 20, 12. ice. Geological Society of America, Abstracts with Colgan, P.M. & Newman, W.A. (1999). Amino acid Programs, 32(7), A-330. evidence for the age of the lower till in Boston Harbor, Birks, H.H., Gulliksen, S., Haflidason, H., Mangerud, J. & Massachusetts. Geological Society of America, Abstracts Possnert, G. (1996). New radiocarbon dates for the Vedde with Programs, 31(2), A-10. Ash and the Saksunarvatn Ash from western Norway. Connally, G.G. (1968). The Rosendale Readvance in the Quaternary Research, 45, 119-127. lower Wallkill Valley, New York. In: National Björck, S., Kromer, B., Johnsen, S., Bennike, O., Association of Geology Teachers Guidebook, Eastern Hammarlund, D., Lemdahl, G., Possnert, G., Rasmussen, Section, 22-28. New York, New Paltz.

New England 187

Connally, G.G. & Sirkin, L.A. (1970). Late glacial history adjacent areas of Canada. Quaternary Research, 23, 341- of the upper Wallkill Valley, New York. Geological 368. Society of America Bulletin, 81, 3297-3306. De Geer, E. (1951a). De Geer’s chronology confirmed by 14 Connally, G.G. & Sirkin, L.A. (1971). The Luzerne read- radiocarbon, C . Geologiska Foreningens i Stockholm vance near Glens Falls, New York. Geological Society of Förhandlingar, 73, 517-518. America Bulletin, 82, 989-1008. De Geer, E. (1951b). Conclusions from C14 and De Geer’s Connally, G.G. & Sirkin, L.A. (1973). Wisconsinan history chronology: Dani-Gotiglacial, with datings. Geologiska of the Hudson-Champlain lobe. In: Black, R.F., Foreningens i Stockholm Förhandlingar, 73, 557-570. Goldthwait, R.P. & Willman, H.B. (eds.), The Wiscon- De Geer, E. (1954). De Geer’s continuous chronology – or sinan Stage, Geological Society of America Memoir, 136, a stretched one with interpretations. Journal of Geology, 47-69. 62, 514-516. Connally, G.G. & Sirkin, L.A. (1986). Woodfordian ice De Geer, G. (1921). Correlation of late glacial clay varves margins, recessional events, pollen stratigraphy of the in North America with the Swedish time scale. mid-Hudson Valley. In: Cadwell, D.H. (ed.), The Geologiska Foreningens i Stockholm Förhandlingar, 43, Wisconsinan Stage of the First Geological District, 70-73. Eastern New York. New York State Museum Bulletin, De Geer, G. (1926). On the solar curve as dating the ice 455, 50-72. age, the New York moraine, and Niagara Falls through Cotter J.F.P., Ridge, J.C., Evenson, E.B., Sevon, W.D., the Swedish time scale. Geografiska Annaler, 8, 253-284. Sirkin, L. & Stuckenrath, R. (1986). The Wisconsinan De Geer, G. (1927). Late glacial clay varves in Argentina history of the Great Valley, Pennsylvania and New measured by D. Carl Caldenius, dated and connected with Jersey, and the age of the “Terminal Moraine”. In: the solar curve through the Swedish time scale. Cadwell, D.H. (ed.), The Wisconsinan Stage of the First Geografiska Annaler, 9, 1-8. Geological District, Eastern New York. New York State De Geer, G. (1929). Gotiglacial clay-varves in southern Museum Bulletin, 455, 22-49. Chile measured by Dr. Carl Caldenius, identified with Crosby, I.B. (1934a). Extension of the Bethlehem, New synchronous varves in Sweden, Finland, and U.S.A. Hampshire, moraine. Journal of Geology, 42, 411-421. Geografiska Annaler, 11, 247-256. Crosby, I.B. (1934b). Geology of Fifteen Mile Falls De Geer, G. (1940). Geochronologia Suecica principles. development. Civil Engineering, 4, 21-24. Kungl. Svenska Vetenskapsakademiens Handlingar, III Cuffey, K.M., Clow, G.D., Alley, R.B., Stuiver, M., Series, 18 (6), 367 pp. Waddington, E.D. & Saltus, R.W. (1995). Large arctic Dineen, R.J. (1986). Deglaciation of the Hudson Valley temperature change at the Wisconsin-Holocene glacial between Hyde Park and Albany, New York. In: Cadwell, transition. Nature, 270, 455-458. D.H. (ed.), The Wisconsinan Stage of the First Geological District, Eastern New York. New York State Cwynar, L.C. & Watts, W.A. (1989). Accelerator mass Museum Bulletin, 455, 89-108. spectrometer ages for late-glacial events in Ballybetagh, Ireland. Quaternary Research, 31, 377-380. Dorion, C.C. (1997). An updated high resolution Dansgaard, W., White, J.W.C. & Johnsen, S.J. (1989). The chronology of deglaciation and accompanying marine abrupt termination of the Younger Dryas climate event. transgression in Maine. Unpublished M.S. Thesis, Nature, 339, 532-534. University of Maine, Orono, 147 pp. Dansgaard, W., Johnsen, S.J., Clausen, H.B., Dahl-Jensen, Dorion, C.C., Balco, G.A., Kaplan, M.R., Kreutz, K.J., D., Gundestrup, N.S., Hammer, C.U., Hvidberg, C.S., Wright, J.D. & Borns, H.W., Jr. (2001). Stratigraphy, Steffensen, J.P., Sveinbjörnsdottir, A.E., Jouzel, J. & paleoceanography, chronology, and environment during Bond, G. (1993). Evidence for general instability of past deglaciation of eastern Maine. In: Weddle, T.K. & climate from a 250-kyr ice-core record. Nature, 364, 218- Retelle, M.J. (eds), Deglacial History and Relative Sea- 220. Level Changes, Northern New England and Adjacent Canada, Geological Society of America, Special Paper, Davis, M.B., Spear, R.W. & Shane, L.C. (1980). Holocene 351, 215-242. climate of New England. Quaternary Research, 14, 240- 250. Dyke, A.S. & Prest, V.K. (1987). Late Wisconsinan and Davis, P.T. & Davis, R.B. (1980). Interpretation of Holocene retreat of the Laurentide ice sheet. Géographie minimum-limiting radiocarbon dates for deglaciation of physique et Quaternaire, 41, 237-263. Mount Katahdin area, Maine. Geology, 8, 396-400. Emerson, B.K. (1898). Geology of Old Hampshire County, Davis, P.T., Dethier, D.P. & Nickmann, R. (1995). Massachusetts, comprising Franklin, Hampshire, and Deglaciation chronology and late Quaternary pollen Hampden Counties. United States Geological Survey, records from Woodford Bog, Bennington County, Monograph, 29, 790 pp. Vermont. Geological Society of America, Abstracts with Fairbanks, R.G. (1989). A 17,000-year glacio-eustatic sea Programs, 27(1), 38. level record: nfluence i of glacial melting rates on the Davis, R.B. & Jacobson, G.L., Jr. (1985). Late glacial and Younger Dryas event and deep-ocean circulation. Nature, early Holocene landscapes in northern New England and 342, 637-642.

188 John C. Ridge

Fleisher, P.J. (1986). Glacial geology and late Wisconsinan Hughes, T., Borns, H.W., Jr., Fastook, J.L., Hyland, M.R., stratigraphy, upper Susquehanna drainage basin, New Kite, J.S. & Lowell, T.V. (1985). Models of glacial re- York. In: Cadwell, D.H. (ed.), The Wisconsinan Stage of construction and deglaciation applied to Maritime Canada the First Geological District, Eastern New York. New and New England. In: Borns, H.W., Jr., LaSalle, P. & York State Museum Bulletin, 455, 121-142. Thompson, W.B. (eds), Late Pleistocene History of Flint, R.F. (1929). The stagnation and dissipation of the last Northeastern New England and Adjacent Quebec, Geolo- ice sheet. Geographical Review, 19, 256-289. gical Society of America, Special Paper, 197, 139-150. Flint, R.F. (1930). The glacial geology of Connecticut. Humphrey, J.T. & Hager, J.L. (1991). The Quaternary of Connecticut Geological Survey Bulletin, 47, 294 pp. the Boston Basin: a new look at some “old” geology. Flint, R.F. (1932). Deglaciation of the Connecticut Valley. Geological Society of America, Abstracts with Programs, American Journal of Science, 24, 152-156. 23 (7), 48. Flint, R.F. (1933). Late-Pleistocene sequence in the Ingólfsson, Ó., Björck, S., Haflidason, H. & Rundgren, M. Connecticut Valley. Geological Society of America (1997). Glacial and climatic events in Iceland reflecting Bulletin, 44, 965-988. regional North Atlantic climatic shifts during the Flint, R.F. (1947). Glacial geology and the Pleistocene Pleistocene-Holocene transition. Quaternary Science epoch. New York, John Wiley and Sons, 589 pp. Reviews, 16, 1135-1144. Flint, R.F. (1953). Probable Wisconsin substages and late Jahns, R.H. & Willard, M. (1942). The Pleistocene and Wisconsin events in northeastern United States and recent deposits in the Connecticut Valley, Massachusetts. southeastern Canada. Geological Society of America American Journal of Science, 240, 161-191 and 265-287. Bulletin, 64, 897-919. Johnson, E.A., Murphy, T., Torreson, O.W. (1948). Pre- Flint, R.F. (1956). New radiocarbon dates and late- history of the earth’s magnetic field. Terrestrial Pleistocene stratigraphy. American Journal of Science, Magnetism and Atmospheric Electricity (now Journal of 254, 265-287. Geophysical Research), 53, 349-372. Flint, R.F. (1957). Glacial and Pleistocene geology. New Kaplan, M.R. (1999). Retreat of a tidewater margin of the York, John Wiley and Sons, 553 pp. Laurentide ice sheet in eastern coastal Maine between ca. Flint, R.F. (1971). Glacial and Quaternary geology. New 14000 and 13000 14C yr B.P. Geological Society of York, John Wiley and Sons, 892 pp. America Bulletin, 111, 620-632. Fuller, M.L. (1914). The geology of Long Island, New Karrow, P.F. & Anderson, T.W. (1975). Palynological York. United States Geological Survey, Professional studies of lake sediment profiles from SW New Paper, 82, 231 pp. Brunswick: discussion. Canadian Journal of Earth Fullerton, D.S. & Richmond, G.M. (1986). Comparison of Science, 12, 1808-1812. the marine oxygen isotope record, the eustatic sea level Kasgarian, M. (1992). Radiocarbon in coastal waters off the record, and the chronology of glaciation in the United eastern United States. Ph.D. Thesis, Yale University. States of America. Quaternary Science Reviews, 5, University Microfilms, Ann Arbor, Michigan. 197-200. Kaye, C.A. (1961). Pleistocene stratigraphy of Boston, Gardner, T.W., Sasowsky, I.D. & Schmidt, V.A. (1994). Massachusetts. United States Geological Survey, Reversed-polarity glacial sediments and revised glacial Professional Paper, 424-B, 73-76. chronology, West Branch Susquehanna River valley, Kaye, C.A. (1964). Outline of Pleistocene geology of central Pennsylvania. Quaternary Research, 42, 131-135. Martha’s Vineyard, Massachusetts. United States Goldsmith, R. (1982). Recessional moraines and ice retreat Geological Survey, Professional Paper, 501-C, 134-139. in southeastern Connecticut. In: Larson, G.J. & Stone, Koteff, C. (1974). The morphologic sequence concept and B.D. (eds.), Late Wisconsinan Glaciation of New deglaciation of southern New England. In: Coates, D.R. England. 61-76. Dubuque, Iowa, Kendall/Hunt (ed.), Glacial Geomorphology. Publications in Geomor- Publishing. phology, State University of New York, Binghamton, 121- Goldthwait, J.W. (1916). Glaciation in the White 144. Mountains of New Hampshire. Geological Society of Koteff, C. (1982). Deglacial history of glacial Lake Nashua, America Bulletin, 27, 263-294. east-central Massachusetts. In: Larson, G.J., Stone, B.D. Goldthwait, J.W. (1938). The uncovering of New (eds.), Late Wisconsinan Glaciation of New England, Hampshire by the last ice sheet. American Journal of 129-143. Dubuque, Iowa, Kendall/Hunt Publishing. Science, 36, 345-372. Koteff, C. & Larsen, F.D. (1989). Postglacial uplift in 18 13 Hillaire-Marcel, C. (1988). Isotope composition ( O, C, western New England: geologic evidence for delayed 14C) of biogenic carbonates in Champlain Sea sediments. rebound. In: Gregerson, S. & Basham, P.W. (eds), Earth- In: Gadd, N.R. (ed.), The Late Quaternary Development quakes at North Atlantic Passive Margins: Neotectonics of the Champlain Sea Basin, Geological Association of and Postglacial Rebound, 105-123. Norwell, Massa- Canada, Special Paper, 35, 177-194. chusetts, Kluwer. Hjolt, C. (1973). A sea correction for East Greenland: Koteff, C. & Pessl, F., Jr. (1981). Systematic ice retreat in Geologiska Föreningens i Stockholm Förhandlingar, 95, New England. United States Geological Survey, 132-134. Professional Paper, 1179, 20 pp.

New England 189

Koteff, C. & Pessl, F., Jr. (1985). Till stratigraphy in New England. 101-114. Dubuque, Iowa, Kendall/Hunt Hampshire: correlations with adjacent New England and Publishing. Quebec. In: Borns, H.W., Jr., LaSalle, P., Thompson, Larsen, P.L., Bierman, P.R. & Caffee, M. (1995). W.B. (eds.), Late Pleistocene History of Northeastern Cosmogenic 26Al chronology of the late Wisconsinan New England and Adjacent Quebec. Geological Society glacial maximum in north-central New Jersey. Geological of America, Special Paper, 197, 1-12. Society of America, Abstracts with Programs, 27(1), 63. Koteff, C., Robinson, G.R., Goldsmith, R. & Thompson, Levy, L.B. (1998). Interpreting the carbonate concretions of W.B. (1993). Delayed postglacial uplift and synglacial glacial Lake Hitchcock. Unpublished B.A. Thesis, Mount sea levels in coastal central New England. Quaternary Holyoke College, Mount Holyoke, Massachusetts, 126 Research, 40, 46-54. pp. Krall, D.B. (1977). Late Wisconsinan ice recession in east- Lewis, R.S. & Stone, J.R. (1991). Late Quaternary central New York. Geological Society of America stratigraphy and depositional history of the Long Island Bulletin, 88, 1697-1710. Sound basin: Connecticut and New York. Journal of LaFleur, R.G. (1979). Deglacial events in the eastern Coastal Research, Special Issue, 11, 1-23. Mohawk – northern Hudson lowland. In: Friedman, G.M. Lini, A., Bierman, P.R. & Lin, L., Davis, P.T. (1995). (ed.), Guidebook for Field Trips. Joint Annual Meeting, Stable carbon isotopes in post-glacial lake sediments: a New York State Geological Association and New England technique for timing the onset of primary productivity Intercollegiate Geological Conference, 326-350. Troy, and verifying AMS 14-C dates. Geological Society of New York, Rensselaer Polytechnic Institute. America, Abstracts with Programs, 27(6), 58. LaFleur, R.G. (1983). Mohawk Valley episodic discharges Lougee, R.J. (1935a). Hanover submerged. Dartmouth – the geomorphic and glacial sedimentary record. In: Alumni Magazine, 27(8), 3-6. Friedman, G.M. (ed.), Eastern Section Guidebook for Lougee, R.J. (1935b). Time measurements of an ice Field Trips. National Association of Geology Teachers, readvance at Littleton, N.H. Proceedings of the National Eastern Section, Spring Meeting, 45-68. Troy, New York, Academy of Sciences, 21, 36-41. Rensselaer Polytechnic Institute. Lougee, R.J. (1939). Geology of the Connecticut Larsen, F.D. (1984). On the relative age of glacial Lake watershed. In: Warfel, H.E. (ed.), Biological Survey of Hitchcock, glacial Lake Winooski, and the Champlain the Connecticut Watershed. New Hampshire Fish and Sea. Geological Society of America, Abstracts with Game Department, Concord, Report, 4, 131-149. Programs, 16(1), 45. Lougee, R.J. (1940). Deglaciation of New England. Journal Larsen, F.D. (1987a). Glacial Lake Hitchcock in the valleys of Geomorphology, 3, 189-217. of the White and Ottauqueche Rivers, east-central Lougee, R.J. (1957). Hanover in the ice age. Dartmouth Vermont. In: Westerman, D.S. (ed.), Guidebook for Field Alumni Magazine, 50(2), 24-29. Trips in Vermont, Volume 2. 79th New England Intercollegiate Geological Conference, Montpelier, Loughlin, G.F. (1905). The clays and clay industries of Vermont, 29-52. Connecticut. Connecticut Geological and Natural History Survey Bulletin, 4, 121 pp. Larsen, F.D. (1987b). History of glacial lakes in the Dog River Valley, central Vermont. In: Westerman, D.S. Lowell, T.V., Hayward, R.K. & Denton, G.H. (1998). Role (ed.), Guidebook for Field Trips in Vermont, Volume 2. of climate oscillations in determining ice-margin 79th New England Intercollegiate Geological Conference, position: hypothesis, examples, and implications. In: Montpelier, Vermont, 213-236. Mickelson, D.M. & Attig, J.W. (eds.), Glacial Processes Past and Present. Geological Society of America, Special Larsen, F.D. (2001). The Middlesex readvance of the late- Paper, 337, 193-203. Wisconsinan ice sheet in central Vermont at 11,900 14C years BP. Geological Society of America, Abstracts with Mangerud, J. (1972). Radiocarbon dating of marine shells Programs, 33(1), A-15. including a discussion of apparent age of recent shells from Norway. Boreas, 1, 143-172. Larsen, F.D. & Hartshorn, J.H. (1982). Deglaciation of the southern portion of the Connecticut Valley of Mangerud, J. & Gulliksen, S. (1975). Apparent radiocarbon Massachusetts. In: Larson, G.J., Stone, B.D. (eds), Late ages of recent marine shells from Norway, Spitsbergen, Wisconsinan Glaciation of New England, 115-128. and Arctic Canada. Quaternary Research, 5, 263-274. Dubuque, Iowa, Kendall/Hunt Publishing. McNish, A.G. & Johnson, E.A. (1938). Magnetization of Larsen, F.D., Ridge, J.C. & Wright, S.F. (2001). unmetamorphosed varves and marine sediments. Correlation of varves of glacial lake Winooski, north- Terrestrial Magnetism and Atmospheric Electricity (now central Vermont. Geological Society of America, Journal of Geophysical Research), 43, 401-407. Abstracts with Programs, 33(1), A-66. Melvin, R.L., De Lima, V. & Stone, B.D. (1992). The Larson, G.J. (1982). Nonsynchronous retreat of ice lobes stratigraphy and hydraulic properties of tills in southern from southeastern Massachusetts. In: Larson, G.J. & New England. United States Geological Survey, Open- Stone, B.D. (eds), Late Wisconsinan Glaciation of New file Report, 91-481, 49 pp.

190 John C. Ridge

Miller, N.G. & Thompson, G.G. (1979). Boreal and western Wisconsinan Glaciation of New England, 1-34. Dubuque, North American plants in the Pleistocene of Vermont. Iowa, Kendall/Hunt Publishing. Journal of the Arnold Arboretum, 60, 167-218. Oldale, R.N. & Colman, S.M. (1992). On the age of the Minard, J.P. (1961). End moraines on Kittatinny Mountain, penultimate glaciation of New England. In: Clark, P.U. & Sussex County, New Jersey. United States Geological Lea, P.D. (eds), The Last Interglacial-Glacial Transition Survey, Professional Paper, 424-C, C61-C64. in North America. Geological Society of America, Special Mott, R.J. (1981). Appendix 4, palynology of southeastern Paper, 270, 163-170. Quebec. In: Shilts, W.W., Surficial Geology of the Lac Oldale, R.N. & Eskenasy, D.M. (1983). Regional Mégantic Area, Québec. Geological Survey of Canada, significance of pre-Wisconsinan till from Nantucket Memoir, 397, 99-102. Island, Massachusetts. Quaternary Research, 19, 302- Mott, R.J., Grant, D.R., Stea, R. & Occhietti, S. (1986). 311. Late-glacial climatic oscillation in Atlantic Canada Oldale, R.N., Cronin, T.M., Valentine, P.C., Spiker, E.C., equivalent to the Allerød/Younger Dryas event. Nature, Blackwelder, B.W., Belknap, D.F., Wehmiller, J.F. & 323, 247-250. Szabo, B.J. (1982). The stratigraphy, absolute age, and Muller, E.H. (1965). Quaternary geology of New York. In: paleontology of the upper Pleistocene deposits at Sankaty Wright, H.E., Jr., Frey, D.G. (eds), The Quaternary of the Head, Nantucket Island, Massachusetts. Geology, 10, United States, 99-112. Princeton, New Jersey, Princeton 246-252. University Press. Pair, D.L. & Rodrigues, C.G. (1993). Late Quaternary deglaciation of the southwestern St. Lawrence Lowland, Muller, E.H. & Calkin, P.E. (1993). Timing of Pleistocene glacial events in New York State. Canadian Journal of New York and Ontario. Geological Society of America Bulletin, 105, 1151-1164. Earth Science, 30, 1829-1845. Pair, D.L., Muller, E.H. & Plumley, P.W. (1994). Muller, E.H. & Prest, V.K. (1985). Glacial lakes in the Correlation of late Pleistocene glaciolacustrine and Ontario Basin. In: Karrow, P.F. & Calkin, P.E. (eds), marine deposits by means of geomagnetic secular Quaternary Evolution of the Great Lakes. Geological variation, with examples from northern New York and Association of Canada, Special Paper, 30, 212-229. southern Ontario. Quaternary Research, 42, 277-287. Muller, E.H., Franzi, D.A. & Ridge, J.C. (1986). Parent, M. & Occhietti, S. (1988). Late Wisconsinan Pleistocene geology of the western Mohawk Valley, New deglaciation and Champlain Sea invasion in the St. York. In: Cadwell, D.H. (ed.), The Wisconsinan Stage of Lawrence Valley, Quebec. Géographie physique et the First Geological District, Eastern New York. New Quaternaire, 42, 215-246. York State Museum Bulletin, 455, 143-157. Reeds, C.A. (1926). The varved clays at Little Ferry, New Newman, W.A. & Mickelson, D.M. (1994). Genesis of Jersey. American Museum of Natural History, New York, Boston Harbor drumlins, Massachusetts. In: Dardis, G.F. American Museum Novitates, 209, 1-16. & McCabe, A.M. (eds.), Subglacial Processes, Sediments Retelle, M.J. & Weddle, T.K. (2001). Deglaciation and and Landforms. Sedimentary Geology, 91, 333-343. relative sea-level chronology, Casco Bay lowland and Newman, W.A., Berg, R.C., Rosen, P.S. & Glass, H.D. lower Androscoggin River Valley, Maine. In: Weddle, (1990). Pleistocene stratigraphy of Boston Harbor T.K. & Retelle, M.J. (eds), Deglacial History and drumlins, Massachusetts. Quaternary Research, 34, 148- Relative Sea-Level Changes, Northern New England and 159. Adjacent Canada. Geological Society of America, Special Newman, W.A., Mickelson, D.M., Berg, R.C., Rendigs, Paper, 351, 191-214. R.D., Oldale, R.N. & Bailey, R.H. (1993). Pleistocene Ridge, J.C. (1988). The Quaternary geology of the upper geology of the Boston Basin and its adjacent Ashuelot, lower Cold River, and Warren Brook valleys of surroundings. In: Cheney, J.T., Hepburn, J.C. (eds), Field southwestern New Hampshire. In: Bothner, W.A. (ed.), Trip Guidebook for the Northeastern United States: 1993 Guidebook for Field Trips in Southwestern New Hamp- Boston GSA, Volume 2. Dept. of Geology and shire, Southeastern Vermont, and North-Central Massa- Geography, University of Massachusetts, Amherst, chusetts. 80th New England Intercollegiate Geological Contribution, 67, pp. U: 1-24. Conference, Keene, New Hampshire, 176-208. Oeschger, H., Andrée, M., Moell, M., Riesen, T., Ridge, J.C. (1990). Surficial Geologic Map of the Walpole, Siegenthaler, U., Ammann, B., Tobolski, K., Bonani, B., N.H. (7.5 x 15 minute) Quadrangle (1:24,000 scale). New Hofmann, H.J., Morenzoni, E., Nessi, M., Suter, M. & Hampshire Geological Survey, Open-file Report, 2 sheets Wölfli, W. (1985). Radiocarbon chronology of and explanation. Lobsigensee. Comparison of materials and methods. In: Ridge, J.C. (1997). Shed Brook Discontinuity and Little Lang, G. (ed.), Swiss Lake and Mire Environments Falls Gravel: Evidence for the Erie interstade in central During the Last 15,000 Years. Dissertationes Botanicae, New York. Geological Society of America Bulletin, 109, 87, 135-139. 652-665. Oldale, R.N. (1982). Pleistocene stratigraphy of Nantucket, Ridge, J.C. (1999). Surficial geologic map of the Bellows Martha’s Vineyard, the Elizabeth Islands, and Cape Cod, Falls Quadrangle, Cheshire and Sullivan Counties, N.H. Massachusetts. In: Larson, G.J., Stone, B.D. (eds), Late and Windham and Windsor Counties, Vt. (1:24,000

New England 191

scale). New Hampshire Geological Survey, Open-file and Relative Sea-level Changes, Northern New England Report, Concord, N.H., 3 sheets. and Adjacent Canada. Geological Society of America, Ridge, J.C. (2001). Surficial geologic map of part of the Special Paper, 351, 173-191. Springfield Quadrangle, Sullivan County, N.H. and Rittenour, T.M. (1999). Drainage history of glacial Lake Windsor County, Vt. (1:24,000 scale). New Hampshire Hitchcock, northeastern USA. Unpublished M.S. Thesis, Geological Survey, Open-file Report, Concord, N.H., 3 Univ. of Massachusetts, Amherst, 179 pp. sheets. Rittenour, T.M., Brigham-Grette, J. & Mann, M.E. (2000). Ridge, J.C. (2003). The last deglaciation of the northeastern El Niño-like climate teleconnections in New England United States: a combined varve, paleomagnetic, and during the late Pleistocene. Science, 288, 1039-1042. calibrated 14C chronology. In: Hart, J.P. & Cremeens Rodrigues, C.G. (1988). Late Quaternary invertebrate D.L. (eds), Geoarchaeology of Landscapes in the faunal associations and chronology of the western Glaciated Northeast U.S. New York State Museum Champlain Sea basin. In: Gadd, N.R. (ed.), The Late Bulletin, 497, in press. Quaternary Development of the Champlain Sea Basin. Ridge, J.C. & Franzi, D.A. (1992). Late Wisconsinan Geological Association of Canada, Special Paper, 35, glacial lakes of the western Mohawk Valley region of 155-176. central New York. In: April, R.H. (ed.), New York State Rodrigues, C.G. (1992). Successions of invertebrate Geological Association Field Trip Guidebook . 64th microfossils and the late Quaternary deglaciation of the Annual Meeting, Colgate University, Hamilton, New central St. Lawrence Lowland, Canada and United States. York, 97-120. Quaternary Science Reviews, 11, 503-534. Ridge, J.C. & Larsen, F.D. (1990). Re-evaluation of Salisbury, R.S. (1902). The glacial geology of New Jersey. Antevs’ New England Varve Chronology and new New Jersey Geological Survey, Final Report, 5, 802 pp. radiocarbon dates of sediments from glacial Lake Shafer, J.P. & Hartshorn, J.H. (1965). The Quaternary of Hitchcock. Geological Society of America Bulletin, 102, New England. In: Wright, H.E., Jr. & Frey, D.G. (eds), 889-899. The Quaternary of the United States, 113-127. Princeton, Ridge, J.C. & Toll, N.J. (1999). Are late-glacial climate os- New Jersey, Princeton University Press. cillations recorded in varves of the upper Connecticut Shotton, F.W. (1972). An example of hard water error in Valley, northeastern United States? Geologiska Fore- radiocarbon dating of vegetable matter. Nature, 240, 460- ningens i Stockholm Förhandlingar, 121(3), 187-193. 461. Ridge, J.C., Brennan, W.J. & Muller, E.H. (1990). The use Sirkin, L.A. (1982). Wisconsinan glaciation of Long Island, of paleomagnetic declination to test correlations of late New York to Block Island. In: Larson, G.J., Stone, B.D. Wisconsinan glaciolacustrine sediments in central New (eds.), Late Wisconsinan Glaciation of New England, 35- York. Geological Society of America Bulletin, 102, 26- 59. Dubuque, Iowa, Kendall/Hunt Publishing. 44. Sirkin, L.A. (1986). Pleistocene stratigraphy of Long Ridge, J.C., Franzi, D.A. & Muller, E.H. (1991). Late Island, New York. In: Cadwell, D.H. (ed.), The Wisconsinan, pre-Valley Heads glaciation in the western Wisconsinan Stage of the First Geological District, Mohawk Valley, central New York, and its regional Eastern New York. New York State Museum Bulletin, implications. Geological Society of America Bulletin, 455, 6-21. 103, 1032-1048. Sirkin, L.A. & Minard, J.P. (1972). Late Pleistocene Ridge, J.C., Thompson, W.B., Brochu, M., Brown, S. & glaciation and pollen stratigraphy in northwestern New Fowler, B. (1996). Glacial geology of the upper Jersey. United States Geological Survey, Professional Connecticut Valley in the vicinity of the lower Paper, 800-D, D51-D56. Ammonoosuc and Passumpsic Valleys of New Smith, G.W. (1980). End moraines and glaciofluvial Hampshire and Vermont. In: Van Baalen, M.R. (ed.), deposits, Cumberland and York Counties, Maine. Maine Guidebook to Field Trips in Northern New Hampshire Geological Survey, Augusta, Maine, scale 1:125,000, 1 and Adjacent Regions of Maine and Vermont. 88th New sheet. England Intercollegiate Geological Conference, Harvard Smith, G.W. (1982). End moraines and the pattern of last University, Cambridge, Massachusetts, 309-340. ice retreat from central and south coastal Maine. In: Ridge, J.C., Besonen, M.R., Brochu, M., Brown, S.L., Larson, G.J., Stone, B.D. (ed.), Late Wisconsinan Callahan, J.W., Cook, G.J., Nicholson, R.S. & Toll, N.J. Glaciation of New England, 195-209. Dubuque, Iowa, (1999). Varve, paleomagnetic, and 14C chronologies for Kendall/Hunt Publishing. late Pleistocene events in New Hampshire and Vermont Smith, G.W. (1985). Chronology of late Wisconsinan (U.S.A.). Géographie physique et Quaternaire, 53, 79- deglaciation of coastal Maine. In: Borns, H.W., Jr., 107. LaSalle, P. & Thompson, W.B. (eds), Late Pleistocene Ridge, J.C., Canwell, B.A., Kelly, M.A. & Kelley, S.Z. History of Northeastern New England and Adjacent (2001). Atmospheric 14C chronology for late Wisconsinan Quebec. Geological Society of America, Special Paper, deglaciation and sea-level change in eastern New 197, 29-44. England using varve and paleomagnetic records. In: Smith, G.W. & Hunter, L.E. (1989). Late Wisconsinan Weddle, T.K. & Retelle, M.J. (eds), Deglacial History deglaciation of coastal Maine. In: Tucker, R.D. &

192 John C. Ridge

Marvinney, R.G. (eds.), Studies in Maine Geology, England Intercollegiate Geological Conference, Storrs, Volume 6: Quaternary Geology. 13-32. Augusta, Maine, Connecticut, 5-29. Maine Geological Survey. Stone, J.R., DiGiacomo -Cohen, M., Lewis, R.S. & Southon, J.R., Nelson, D.E. & Vogel, J.S. (1990). A record Goldsmith, R. (1998a). Recessional moraines and the of past ocean-atmosphere radiocarbon differences from associated deglacial record of southeastern Connecticut. the northeast Pacific. Paleoceanography, 5, 197-206. In: Murray, D.P. (ed.), Guidebook to Field Trips in Stanford, S.D. (1993a). Late Cenozoic surficial deposits Rhode Island and Adjacent Regions of Connecticut and th and valley evolution of unglaciated northern New Jersey. Massachusetts. 90 New England Intercollegiate Geomorphology, 7, 267-288. Geological Conference, Kingston, Rhode Island, pp. B5: Stanford, S.D. (1993b). Late Wisconsinan glacial geology 1-20. of the New Jersey Highlands. Northeastern Geology, 15, Stone, J.R., Schafer, J.P., London, E.H., Lewis, R.L., 210-223. DiGiacomo-Cohen, M.L. & Thompson, W.B. (1998b). Stanford, S.D. & Harper, D.P. (1991). Glacial lakes of the Quaternary geologic map of Connecticut and Long Island lower Passaic, Hackensack, and lower Hudson Valleys, Sound Basin (scale 1:175,000). United States Geological New Jersey and New York. Northeastern Geology, 13, Survey, Open-file Report, 98-371, 1 sheet, 77 pp. 271-286. Stuiver, M. & Borns, H.W., Jr. (1975). Late Quaternary Stanford, S.D., Seidl, M.A. & Ashley, G.M. (2000). marine invasion in Maine: its chronology and associated Exposure age and erosional history of an upland crustal movement. Geological Society of America planation surface in the US Atlantic piedmont. Earth Bulletin, 86, 99-104. Surface Processes and Landforms, 25, 939-950. Stuiver, M. & Reimer, P.J. (1993). Extended 14C data base Stocker, T.F. & Wright, D.G. (1998). The effect of a and revised CALIB 3.0 14C age calibration program. succession of ocean ventilation changes on 14C. Radiocarbon, 35, 215-230. Radiocarbon, 40, 359-366. Stuiver, M., Grootes, P.M. & Braziunas, T.F. (1995). The Stone, B. & Borns, H.W., Jr. (1986). Pleistocene glacial and GISP2 δ18O climate record of the past 16,500 years and interglacial stratigraphy of New England, Long Island, the role of the sun, ocean, and volcanoes. Quaternary and adjacent Georges Bank and Gulf of Maine. Research, 44, 341-354. Quaternary Science Reviews, 5, 39-52. Stuiver, M., Reimer, P.J., Bard, E., Beck, J.W., Burr, G.S., Stone, B.D. & Peper, J.D. (1982). Topographic control of Hughen, K.A., Kromer, B., McCormac, F.G., van der the deglaciation of eastern Massachusetts: ice lobation Plicht, J. & Spark, M. (1998). INTCAL98 radiocarbon and the marine incursion. In: Larson, G.J. & Stone, B.D. age calibration 24,000 – 0 cal BP. Radiocarbon, 40, (eds), Late Wisconsinan Glaciation of New England,145- 1041-1083. 166. Dubuque, Iowa, Kendall/Hunt Publishing. Sutherland, D.G. (1986). A review of Scottish marine shell Stone, B., Stanford, S.D. & Witte, R.W. (2002). Surficial geologic map of northern New Jersey (1:100,000 scale). radiocarbon dates, their standardization and interpretation. Scottish Journal of Geology, 22, 145-164. United States Geological Survey, Miscellaneous Investigation Series, Map, I-2540-C, 3 sheets and 1 Thomas, G.M. (1984). A comparison of the paleomagnetic pamphlet, 41 pp. character of some varves and tills from the Connecticut Stone, J.R. (1995). Timing and mechanisms of glacial Lake Valley. Unpublished M.S. Thesis, Univ. of Massa- Hitchcock drainage. Geological Society of America, chusetts, Amherst, 136 pp. Abstracts with Programs, 27(1), 85. Thompson, W.B. (1998). Deglaciation of western Maine Stone, J.R. (1999). Effects of glacio-isostasy and relative and the northern White Mountains. Geological Society of sea level on late-glacial and postglacial water levels in the America, Abstracts with Programs, 30(1), 79. Connecticut River valley and Long Island Sound. Thompson, W.B. & Borns, H.W., Jr. (eds.), (1985). Geological Society of America, Abstracts with Programs, Surficial geologic map of Maine (scale 1:500,000). Maine 31(2), 71. Geological Survey, Augusta, Maine, 1 sheet. Stone, J.R. & Ashley, G.M. (1992). Ice-wedge casts, pingo Thompson, W.B., Fowler, B.K., Flanagan, S.M. & Dorion, scars, and the drainage of Lake Hitchcock. In: Robinson, C.C. (1996). Recession of the Late Wisconsinan ice sheet P. & Brady, J.B. (eds), Guidebook for Field Trips in the from the northwestern White Mountains, N.H. In: Van Connecticut Valley Region of Massachusetts and Baalen, M.R. (ed.), Guidebook to Field Trips in Northern Adjacent States. 84th New England Intercollegiate New Hampshire and Adjacent Regions of Maine and Geological Conference, Amherst, Massachusetts, 305- Vermont. 88th New England Intercollegiate Geological 331. Conference, Harvard University, Cambridge, Massa- Stone, J.R., Schafer, J.P. & London, E.H. (1982). The chusetts, 203-234. surficial geologic maps of Connecticut illustrated by a Thompson, W.B., Fowler, B.K. & Dorion, C.C. (1999). De- field trip in central Connecticut. In: Joesten, R. & glaciation of the northwestern White Mountains, New Quarrier, S.S. (eds), Guidebook for Field Trips in Hampshire. Géographie physique et Quaternaire, 53, Connecticut and South Central Massachusetts. 74th New 59-78.

New England 193

Törnqvist, T.E., de Jong, A.F.M. & van der Borg, K. (1992). Accurate dating of organic deposits by AMS 14C measurement of macrofossils. Radiocarbon, 34, 566-577. Turney, C.S.M., Coope, G.R., Harkness, D.D., Lowe, J.J. & Walker, M.J.C. (2000). Implications for the dating of Wisconsinan (Weichselian) late-glacial events of systematic radiocarbon age differences between terrestrial plant macrofossils from a site in SW Ireland. Quaternary Research, 53, 114-121. Uchupi, E., Driscoll, N., Ballard, R.D. & Bolmer, S.T. (2001). Drainage of late Wisconsinan glacial lakes and the morphology and late Quaternary stratigraphy of the New Jersey – southern New England continental shelf. Marine Geology, 172, 117-145. Veeger, A.I., Johnston, H.E., Stone, B.D. & Sirkin, L.A. (1994). Hydrogeology and water resources of Block Island, Rhode Island. United States Geological Survey, Water Resources Investigations, Report, 94-4096, 76 pp. Verosub, K.L. (1979a). Paleomagnetism of varved sediments from western New England: Secular variation. Geophysical Research Letters, 6, 245-248. Verosub, K.L. (1979b). Paleomagnetism of varved sediments from western New England: Variability of the recorder. Geophysical Research Letters, 6, 241-244. Weddle, T.K., Stone, B.D., Thompson, W.B. & Retelle, M.J. (1989). Illinoian and late Wisconsinan tills in eastern New England: A transect from northeastern Massachusetts to west-central Maine. In: Berry, A.W., Jr. (ed.), Guidebook for field trips in southern and west- central Maine. 81st New England Intercollegiate geological Conference, Farmington, Maine, 25-85. Wilson, B.R. (2000). A chronology and environmental interpretation of glacial to non-glacial lacustrine varves in the Passumpsic Valley, Barnet, Vermont. Unpublished B.S. Thesis, Tufts University, Medford, Massachusetts, 83 pp. Wohlfarth, B. (1996). The chronology of the last termination: a review of radiocarbon-dated, high-resolution terrestrial stratigraphies. Quaternary Science Reviews, 15, 267-284. Woodworth, J.B. & Wigglesworth, E. (1934). Geography and geology of the region including Cape Cod, the Elizabeth Islands, Nantucket, Martha’s Vineyard, No Mans Land, and Block Island. Harvard College Museum of Comparative Zoology Memoirs, 52, 322 pp.