Printer Friendly

Impact of the Younger Dryas cooling event upon lowland vegetation of Maritime Canada.

INTRODUCTION

The switch from the last glaciation to the present interglaciation is known as the late-glacial period, lasting from [approximately equals] 14000 to 10000 years before present (BP). This period of transition was a time of marked climate instability, characterized by several abrupt climate oscillations. The most severe and longest lasting was the Younger Dryas cold period, [approximately equals] 10800-10000 [sup.14]C yr BP, when there was a return to almost full-glacial conditions in the circum-North Atlantic region (e.g., Watts 1977, Lowe et al. 1980, Mott et al. 1986, Wright 1989, Peteet et al. 1990, Mayle et al. 1993a). The magnitude of the Younger Dryas cooling was very great; geological evidence indicates renewed glacial activity in parts of northwest Europe, such as the highlands of Scotland (Sissons 1979), while fossil beetle studies in Britain (Atkinson et al. 1987) and fossil midge studies in Atlantic Canada (Walker et al. 1991, Levesque et al. 1993, Wilson et al. 1993) indicate a drop in maximum summer temperatures of 6 [degrees] - 7 [degrees] C. Furthermore, the Younger Dryas began and ended extremely rapidly. Recent ice-core evidence from Greenland indicates that both its onset and termination occurred over timespans perhaps as short as 10 yr or less (Alley et al. 1993, Taylor et al. 1993), even faster than the expected rate of warming due to increases in greenhouse gases today. Paleoecological studies on the impact of the Younger Dryas upon the vegetation can therefore provide important insights into ecological questions such as the nature of vegetation response to extremely rapid and severe climate cooling and warming.

Numerous late-glacial pollen profiles from sites in Atlantic Canada (e.g., Livingstone and Livingstone 1958, Anderson 1983, Mott et al. 1986, Jette and Mott 1989, Stea and Mott 1989, Mayle et al. 1993a) have shown that during the late-glacial period the vegetation of Atlantic Canada was very sensitive to climatic change because the boundary between forest and tundra lay in this region. As a result, vegetation changes caused by the Younger Dryas cooling varied considerably over relatively short distances. Sites near the southern coast of New Brunswick and in central mainland Nova Scotia show a change from closed spruce forest or woodland to shrub-tundra at the onset of the Younger Dryas, while sites in central New Brunswick and Cape Breton Island show a change from shrubtundra to herb-tundra instead.

The ecological information that has been gleaned from these vegetation reconstructions, however, is rather limited, and our knowledge of the constituent species of the late-glacial flora of the region is still very poor. This is due to a number of factors. (1) Sampling resolution has generally been low, limiting the potential temporal resolution. (2) Pollen taxa that may potentially yield important ecological information (e.g., pine [Pinus] and alder [Alnus]) have, at most sites, not been identified to species level, and many pollen diagrams have been published in summary form only. (3) In the same vein, even with more refined pollen identification, a number of important pollen taxa cannot be identified below family or genus. For example, in the late-glacial period does birch pollen represent tree or arctic shrub species? Is Cupressaceae pollen derived from temperate eastern white cedar (Thuja occidentalis) or the more northerly ranging common juniper (Juniperus communis)? What taxa are represented by pollen of the ericad family, important shrubs in arctic/boreal regions? This level of detailed vegetation reconstruction is possible by studying plant macrofossils. In addition, macrofossils can be used to determine categorically what species were locally present at a site.

The purpose of this research is to attempt to answer the following ecological questions. (1) What species were contributing to the late-glacial flora? (2) Were the changes in vegetation brought about by the Younger Dryas qualitative (changes in taxa), or quantitative (similar taxa, but changes in abundance)? (3) What was the rate of vegetation change as a result of the abrupt initial cooling? Did some species respond more quickly than others? Were there significant lags in vegetation response? (4) Similarly, how did the vegetation respond to the subsequent rapid warming?

In this paper we will draw upon detailed sediment, pollen, and macrofossil studies from sediment cores of six lake sites in New Brunswick and Nova Scotia in an attempt to answer these questions.

STUDY AREA

Physiography and deglaciation

Atlantic Canada belongs mainly to the Appalachian physiographic province (Grant and King 1984). It is comprised of two types of terrain: (1) broad rolling lowlands within 100 m of sea level, developed on folded Carboniferous rocks dissected by deep valleys; and (2) fault-bounded pre-Carboniferous crystalline massifs rising abruptly from lower terrains, forming steep-sided, deeply dissected uplands and highlands with summits [is greater than] 400 m. The region is bounded to the north by the Precambrian Shield, and fringed by a broad shelf underlain mainly by flat-lying Mesozoic and Cenozoic strata. Glacial deposits are thin or absent on the resistant uplands, but nonglacial deposits are locally thick along the flanks. Thicker Quaternary deposits cover the lowlands, especially over shale and karst topography. Surface features, however, give little evidence of the nature of Quaternary sequences that lie beneath. Stratigraphic studies are thus primarily dependent on sections resulting from fluvial dissection, coastal retreat, and artificial excavations for gypsum, limestone, and coal.

Although geologists previously interpreted the pattern of deglaciation of southwestern New Brunswick as an ice-marginal recession toward the northwest (e.g., Rampton et al. 1984), more recent evidence suggests that regional stagnation of ice masses is a more likely interpretation (Seaman et al. 1991). Earliest deglaciation in the region is generally considered to have occurred adjacent to the Bay of Fundy (Stea and Wightman 1987, Stea and Mott 1989). Ice flowed southwestward out of Chignecto Bay and westward out of Minas Basin into the head of the Bay of Fundy (Fig. 1), probably as a response to marine incursion into the isostatically depressed Bay of Fundy. Accelerator-mass-spectrometer (AMS) [sup.14]C dates obtained from the bottomset beds of a glaciomarine delta at Spencers Island, Nova Scotia, near the head of the Bay of Fundy, provide evidence that the Bay of Fundy became virtually ice free by [approximately equals] 14000 yr BP (Stea and Wightman 1987). However, AMS [sup.14]C dates on plant macrofossils from basal organic sediment from Splan Pond and Mayflower Lake in southern New Brunswick and from Lac a Magie in southern Nova Scotia (Mayle et al. 1993a; Figs. 1, 3, 6 and 10) suggest that it is unlikely that the ice margin retreated much beyond the present-day coastline of the Bay of Fundy until shortly before 12 000 yr BP We acknowledge though, that dates from the basal organics of lake sediments represent minimal age estimates for deglaciation, since there was obviously a lag of unknown duration between deglaciation of a site and deposition of organic lake sediments or plant macrofossils in a lake basin.

[Figure 1 ILLUSTRATION OMITTED]

The manner of deglaciation of Nova Scotia is poorly known. The following account is taken from Stea and Mott (1989), based upon chronological, palynological, and sedimentological data. Separation of local centers of ice outflow in northern Nova Scotia and the South Mountain region probably occurred between 14500 and 13000 yr BP. Retreat continued up the Minas Basin, and ice-front retreat along the Atlantic coast may have reached the present-day coastline by that time. Separation of ice masses over northeastern and southwestern mainland Nova Scotia may have occurred by 13 000 yr BP or slightly later. Deglaciation continued toward local ice centers, so that by 12 000 yr BP, or shortly thereafter, glaciers were restricted to the Cobequid and Antigonish Highlands, most of western Cape Breton Island, possibly part of Prince Edward Island, and the South Mountain. Deglaciation of mainland Nova Scotia may have been virtually complete by 11 000 yr BP, but small remnant or stagnant ice masses may have persisted in the Antigonish Highlands and the north side of the Cobequid Highlands. Although Railton (1972) argues that an icecap covered the South Mountain at this time, based on extrapolation from dates younger than 10000 yr BP at Oakhill Lake and Minard Lake, the Younger Dryas oscillation can be recognized from the lithological changes in the undated sequence toward the base of the lake cores, indicating that the South Mountain was deglaciated prior to 11 000 yr BP (the lower Younger Dryas boundary). The pattern of deglaciation in Cape. Breton Island is not well understood due, in part, to the scarcity of lake sites that have been reliably dated. Although a basal organic AMS [sup.14]C date of 14010 [+ or -] 100 yr BP from Chase Pond (Mayle et al. 1993a, Figs. 1 and 13) suggests that deglaciation may have been earliest at Cape Breton Island, it is likely that dates from this site are anomalously old (see discussions in Levesque et al. 1993 and Mayle et al. 1993a). The lack of typical late-glacial lithological changes from some lake cores in central and northern Cape Breton Island suggests that a remnant icecap persisted in the Cape Breton Highlands until at least 10000 yr BP (Stea and Mott 1990). Likewise, absence of organic sites on Prince Edward Island older than 10000 yr BP, together with evidence from mapping of ice-flow directions and raised marine features, points to the persistence of a residual ice mass over part of the island and Northumberland Strait until the end of the late-glacial period (Stea and Mott 1990).

[Figure 13 ILLUSTRATION OMITTED]

Elevation isolines of raised marine features (Stea 1987) show that late-glacial shorelines were [is less than or equal to] 80 m above present-day mean sea level in the Bay of Fundy region, but were below present-day sea level along the Atlantic coast of Nova Scotia.

Modern vegetation

The present-day vegetation of the Maritime Provinces (Nova Scotia, New Brunswick, and Prince Edward Island) has been classified by Rowe (1972) as the Acadian Forest Region. The forest is closely related to the Great Lakes-St. Lawrence Forest Region and, in some respects, to the Boreal Forest Region. It therefore has affinities with both the boreal forest and the temperate deciduous forest. Boreal species include black spruce (Picea mariana), white spruce (P. glauca), balsam-fir (Abies balsamea), balsam poplar (Populus balsamifera), trembling aspen (P. tremuloides), tamarack (Larix laricina), and paper birch (Betula papyrifera). Temperate deciduous species include eastern white pine (Pinus strobus), sugar maple (Acer saccharum), yellow birch (B. alleghaniensis), eastern hemlock (Tsuga canadensis), beech (Fagus grandifolia), American elm (Ulmus americana), and black ash (Fraxinus nigra). Red spruce (Picea rubra), red pine (Pinus resinosa), red oak (Quercus rubra), and red maple (A. rubrum) are also common constituents. Eastern white cedar (Thuja occidentalis), although common in New Brunswick, is uncommon in other provinces of the region, and jack pine (P. banksiana) also has a limited distribution.

The modern distribution of these taxa in the Maritimes is strongly influenced by proximity to the ocean, topography, and forest clearance since arrival of European settlers, as described by Loucks (1962). Coastal areas have strong boreal affinities as a consequence of the cold Labrador Current, especially around the Bay of Fundy where cool onshore winds prevail. Most sites support coniferous forests of white spruce (especially within the spray zone), black spruce, balsam fir, and tamarack. Tolerant hardwoods, such as beech, sugar maple, and yellow birch, are prominent only on hilltops. One notable feature of coastal forests is the minor presence of red spruce, which is otherwise common throughout the Maritimes. The original forest has mostly been cleared, leading to an abundance of trembling aspen, white birch, grey birch (B. populifolia), red maple, and white spruce on disturbed sites.

STUDY SITES

Six small lakes were chosen for study, two in New Brunswick, and four in Nova Scotia (Fig. 1). The rationale for choosing sites close to the Bay of Fundy and the Atlantic coast was based on evidence that the Younger Dryas cooling was most likely centered in the North Atlantic Ocean, so the severity of the cooling is expected to have been greatest at those sites where the climate was most strongly influenced by the North Atlantic Ocean. Following deglaciation, glacioisostatic crustal rebound of the region in combination with eustatic sea-level rise resulted in a complex history of relative sea-level change in the region. Maps plotting elevation isolines of raised marine features in the study area (Stea 1987) were therefore consulted prior to the initial lake survey to ensure that lakes chosen for study were at a high enough elevation to exclude the possibility that they may have been inundated by the sea during the late-glacial/early postglacial period.

Small (0.4-4.2 ha), preferably steep-sided lakes were chosen where there was a likelihood of macrofossils being focused toward the central, deeper parts of the basin. Lakes with inflowing streams were avoided, since the stream could have transported pollen and macrofossils from an unknown source area, which would complicate the interpretation of the pollen and macrofossil records. In addition, inflowing streams may disturb the surface lake sediment, thereby reducing the potential temporal resolution of the pollen record.

Splan Pond (45 [degrees] 15' 20"N, 67 [degrees] 19' 50"W; Fig. 1), formerly published as Basswood Road Lake (Most 1975, Mott et al. 1986) and Splan Lake (Rawlence 1988), is [approximately equals] 4 ha in area, at an elevation of [approximately equals] 106 m, and is located 7.2 km northwest of St. Stephen, New Brunswick. The maximum water depth is 10.6 m and the immediate topography is gently undulating, with steepest slopes along the eastern shore. There are no inflowing or outflowing streams. Surrounding vegetation consists of a mixture of conifers and hardwoods: paper birch, white pine, balsam fir, red maple, and spruce.

Mayflower Lake (45 [degrees] 18' 10"N, 66 [degrees] 04' 15"W; Fig. 1) is [approximately equals] 4.2 ha in area at an elevation of [approximately equals] 50 m in Rockwood Park, Saint John, New Brunswick. Although evidence for a raised marine delta (55 m elevation) in the Saint John area would suggest that this basin may have been subject to late-glacial marine transgression (Stea 1987), the late-glacial sequence from this site and other unpublished lakes in Rockwood Park (D. J. Rawlence, personal communication) indicates that the sea had retreated from this locality well before the onset of the Younger Dryas. Water depth at the center of the wider portion of the lake is 11 m; there is a small outflowing stream. Surrounding woodland is dominated by eastern white cedar and red spruce, with occasional paper birch, yellow birch, and red maple. The lake is surrounded by steep banks, rising to [approximately equals] 15 m in some places.

Little Lake (44 [degrees] 40' 5"N, 63 [degrees] 56' 20" W; Fig. 1) is [approximately equals] 0.5 ha in area at an elevation of [approximately equals] 40 m near Boutilier, St. Margaret's Bay, [approximately equals] 25 km west of Halifax, Nova Scotia. Maximum water depth is 5 m and there is a small outflowing stream. The basin is partially filled in with Sphagnum bog and black spruce surrounding the lake, which may originally have been [approximately equals] 2.2 ha in area. The lake basin is surrounded by steep slopes, ranging in height from [approximately equals] 5 to 25 m above the lake surface. The local vegetation is dominated by a dense red spruce forest with occasional red maple. Balsam fir saplings dominate the understory.

Lac a Magie (44 [degrees] 15' 50"N, 66 [degrees] 04' 45"W; Fig. 1) is [approximately equals] 0.6 ha in area, at an elevation of [approximately equals] 60 m, and is located near Church Point, southwest Nova Scotia. Water depth at the center of the lake is 2.2 m; there are no inflowing or outflowing streams. Part of the surrounding forest has been cleared and is dominated by herbs and shrubs such as Rubus and Myrica gale.

The local woodland is dominated by red maple and tamarack with occasional paper birch and spruce. The local topography is relatively flat.

Chase Pond (informal name) (45 [degrees] 39'5"N, 60 [degrees] 40'30"W; Fig. 1) is [approximately equals] 0.4 ha in area at an elevation of [approximately equals] 15 m near Grand River, southern Cape Breton Island, Nova Scotia. A narrow fringe of Sphagnum bog and shrubs occupies the lake margin, indicating that the pond was formerly larger in area, possibly 1 ha. Dead black spruce trunks in the Sphagnum bog and beaver dams in the small outflowing stream indicate that the water level has been raised a metre or so by beaver activity. Black spruce and tamarack surround the bog margin. The basin is surrounded by steep banks, [approximately equals] 10 m above lake level, covered with dense white spruce forest with occasional balsam fir. Maximum water depth is [approximately equals] 4 m.

Main-a-Dieu Pond (informal name) (45 [degrees] 58'30"N, 59 [degrees] 50'45"W; Fig. 1) is [approximately equals] 0.5 ha in area at an elevation of [approximately equals] 30 m and is located [approximately equals] 4 km from Main-a-Dieu near Main-a-Dieu Passage, eastern Cape Breton Island, Nova Scotia. The lake has a very small inflowing and outflowing stream. Much of the basin is occupied by black spruce/Sphagnum bog and associated shrubs such as juniper (Juniperus communis), ericads, and sweet gale (Myrica gale). The original lake area may have been [approximately equals] 1.5 ha, although the size of the original basin is difficult to estimate since the surrounding topography is very flat. Local vegetation is almost exclusively black spruce, with occasional tamarack and balsam fir.

METHODS

Field methods

All sites were cored from the frozen lake surface in winter using a 5 cm diameter modified Livingstone piston sampler (Wright 1967). At most sites cores were taken toward the steepest-sided shore (generally approximately equidistant between the lake center and the shore). The Younger Dryas sequence could generally be recognized in the field as a distinct lithological unit: a light-colored band of silty clay or organic silt bracketed by darker, more organic sediment. Observable lithological changes at Little Lake were more subtle and difficult to discern. Two to six replicate cores of the lowermost sediment spanning the presumed late-glacial and early Holocene sequence were taken [approximately equals] 30 cm from the location of the original core hole. The reason for coring closer to the shore (instead of the usual practice of coring at the lake center where the sediment sequence is expected to be longest) and taking replicate late-glacial sediment sequences was to increase the likelihood of obtaining sufficient amounts of terrestrial plant macrofossils for AMS [sup.14]C dating. Cores were wrapped in plastic film and aluminum foil in the field and transported to the laboratory where they were stored at 3 [degrees] C until subsampled. The sites were revisited in the summer to record the local vegetation.

Laboratory methods

Fresh lake-sediment samples for pollen and loss-on-ignition (0.5 mL for most samples, 1 mL for pollen samples of basal clays) were removed with a calibrated brass sampler (Birks 1976), generally at 1- or 2-cm intervals, taking precautions to avoid contamination by scraping away the surface sediment and using distilled water to clean the sampling instruments. Pollen samples were stored in distilled water at 3 [degrees] C until processed. Finer or coarser sampling resolution was employed where sediment sequences were shorter or longer respectively. Sampling resolution was highest across the Younger Dryas boundaries where pollen and loss-on-ignition changes were expected to be most abrupt.

The organic content of sediment was estimated by loss-on-ignition (LOI). Samples were dried for 12 h at 100 [degrees] C and combusted for 1 h at 550 [degrees] C (Dean 1974). Replicate cores were correlated lithologically and cut into 1- or 2-cm segments. The outer 1-3 mm were scraped off each segment to remove possible contaminant macrofossils. All non-horizontally oriented macrofossils on or near the outside of the core segment were removed for the same reason. Segments from replicate cores were then combined, provided there was perfect correlation. Subsequently the sediment was dispersed by gentle agitation in water and washed through 0.5 and 1.0 mm mesh screens to recover macrofossils. Stones or pebbles were also recovered. Identifiable terrestrial plant macrofossils (in most cases twigs) from selected stratigraphic levels (e.g., Younger Dryas boundaries) were then dried at 100 [degrees] C for 12 h, weighed, wrapped in aluminum foil, and submitted to the IsoTrace Laboratory of the University of Toronto for AMS [sup. 14]C dating. Other macrofossils were stored in glycerine.

Three tablets of Lycopodium clavatum spores (Stockmarr 1971) were added to each pollen sample to facilitate calculation of pollen concentration and accumulation rates. Samples were then processed by standard methods (Faegri and Iversen 1975), which included sieving the late-glacial samples through 7 ,[micro]m mesh screens to remove fine inorganic sediment (Cwynar et al. 1979). Processed pollen was stored in silicon oil, and prepared slides were routinely examined at x400 magnification. Critical grains were examined under anisole at x1000 magnification.

Pollen and macrofossil identifications

Identification of fossil pollen was based on published keys (Kapp 1969, McAndrews et al. 1973, Moore and Webb 1978, Faegri and Iversen 1989) and pollen reference slides. Nomenclature largely follows Fernald (1991), although Hinds (1986) was followed for treatment of family names. A minimum sum of 300 pollen grains and spores of terrestrial taxa was tallied for each sample and used to calculate pollen percentages. The sum of all pollen and spores (including aquatic taxa) was used to calculate percentages of aquatic taxa. Pollen diagrams were zoned by inspection and these zonations were transposed onto the macrofossil diagrams.

Determination of spruce pollen to species level was not attempted because red spruce, which is common in Maritime Canada, cannot be reliably distinguished from black or white spruce. Pinus pollen can be attributed to either Haploxylon (soft or white pine) or Diploxylon (hard pine) type. In Atlantic Canada these types correspond to P. strobus and P. banksiana/resinosa respectively. Pinus pollen was recorded either as "Pinus undifferentiated" (including degraded, deformed, or crumpled grains), P. strobus, or P. banksiana/resinosa. The ratio of Haploxylon: Diploxylon pine pollen was based on a minimum count of 25 identifiable grains per sample, if the percentage of total pine pollen was [is less than] 10%, and a minimum of 50 identifiable grains if the percentage of total pine pollen was [is greater than] 10%. In order to determine this ratio, counts were supplemented when necessary by counting outside the sum. This ratio was then used to assign the "total pine pollen" counted within the pollen sum to either Haploxylon or Diploxylon type. (It was assumed that the undifferentiated component is composed of Diploxylon and Haploxylon pine in the same proportion as the identified Pinus grains.) Pine pollen in the lowermost basal clays was generally too degraded to allow its identification to species. Since pine pollen percentages at Chase Pond and Main-a-Dieu Pond were low "throughout the sequence (generally 510%), Pinus was not identified to species at these sites.

There are three native species of alder (Alnus) commonly found in eastern North America: green alder (A. crispa), speckled alder (A. rugosa), and smooth alder (A. serrulata). Pollen from the latter two species cannot be distinguished and is here referred to as A. rugosa-type. Refer to Mayle et al. (1993b) and Richard (1970) for morphological criteria used to distinguish A. crispa from A. rugosa-type pollen in this study.

All macrofossils were identified by comparison with modern reference collections, counted and tallied. However, identified twig fragments were not enumerated, but were instead recorded as present or absent since their quantification is more problematical. Betulaceae fruits and catkin scales were identified to species by comparison with reference material of A. crispa (green alder), A. rugosa (speckled alder), A. serrulata (smooth alder), Betula glandulosa (dwarf birch), B. papyrifera (white birch), B. populifolia (grey birch), B. alleghaniensis (yellow birch), B. Ienta (black birch), and B. pumila (swamp birch). Morphological criteria of fruits used for identification of these taxa are: shape and size of body, shape and size of wing, ratio between maximum body diameter and maximum wing diameter, angle of attachment at junction between wing and body, and degree of pubescence on body (Fig. 2). The most reliable criterion for distinguishing the fruits of B. papyrifera from those of B. populifolia is the degree of pubescence. Pubescence on fruits of B. populifolia, if any, is restricted to a few hairs at the base of the style, while the upper half or third of the body of B. papyrifera fruits is densely pubescent. Morphological criteria for the identification of catkin scales are: shape, size, and degree of pubescence (Fig. 2). Although the catkin scales of B. populifolia and B. papyrifera are similar in shape, the convex surface of B. populifolia catkin scales is much more pubescent than those of B. papyrifera. Needles from coniferous taxa such as spruce and tamarack were more commonly found as fragments, such as tips and bases, rather than whole needles. Minimum estimates of needle number were therefore obtained by adding the highest number of needle tips or bases to the number of whole needles. Similarly, minimum numbers of Dryas integrifolia leaves were obtained by adding the number of identifiable basal leaf fragments to the number of whole leaves. Dryas leaves were identified as D. integrifolia after comparison between leaves of D. alaskensis, D. crenulata, D. drummondii, D. integrifolia, D. octopetala, D. punctata, D. sylvatica, and Ledum decumbens. Criteria used for separation of D. integrifolia from the other taxa are: size, shape of leaf margin, and angle of attachment at junction between lamina and petiole (Fig. 2). Juniperus communis needles were distinguished from other members of the genus and other coniferous taxa by comparison of needle shape, and arrangement of epidermal cells and stomata (Fig. 2). Diagnostic features used for identification of Pinus strobus needles were the serrated needle margins and triangular cross section (Fig. 2). Some taxa are represented by different types of macrofossils, e.g., flowers, anthers fruits, seeds, twigs. Generally, however, only the macrofossil type showing the greatest stratigraphic range, or most continuous sequence, was plotted in the macrofossil diagram.

[Figure 2 ILLUSTRATION OMITTED]

No macrofossil diagram was plotted for Main-aDieu Pond because, even after pooling seven replicate cores, the sediment was virtually devoid of macrofossils.

RESULTS

Sediment lithology

Table 1 summarizes the sediment lithology of the six sites. The carbonate component of the sediment cores from all of the sites was minimal (since reaction with 10% HCI, prior to processing for pollen, was negligible), except for the basal clays of Little Lake, Main-a-Dieu Pond, and Chase Pond, which reacted strongly with 10% HCI. At most of these sites there is a sharp contact between gyttja and silt or clay at the base of the Younger Dryas, whereas the transition from silt or clay to gyttja near the end of the Younger Dryas spans several centimetres. In addition, stones [equal to or less than] 1 cm found in the less organic Younger Dryas sediment are more prevalent toward the base of the sediment unit.

TABLE 1. Summary of sediment lithology of the six sites, shown in stratigraphic sequence, with increasing sediment depth toward the bottom of the table.
                         Depth below sediment/water interface (cm)
        Sediment
        lithology        Splan Pond         Mayflower Lake

Dark brown gyttja,
   grading to organic
   silt toward bottom          0-468.5             0-350
Light brown organic
   silt                    468.5-472.5             ...
Light gray/pink clay       472.5-530.5(*)          ...
Light gray silt                ...               350-361.5(*)
Brown organic silt         530.5-556           361.5-381.5
Clay                         556-565           381.5-390
Clay/sand                    565-595               ...
Gravel [+ or -] clay/sand    595-632               ...

                         Depth below sediment/water interface (cm)
        Sediment
        lithology        Little Lake

Dark brown gyttja,
   grading to organic
   silt toward bottom             0-703
Light brown organic
   silt                         703-[approximately equals] 729
Light gray/pink clay             . . .
Light gray silt
                                [approximately equals] 729-752
Brown organic silt              752-864
Clay                            864-898
Clay/sand                       898-901.5
Gravel [+ or -] clay/sand     901.5-941

                         Depth below sediment/water interface (cm)
        Sediment
        lithology        Lac a Magie

Dark brown gyttja,
   grading to organic
   silt toward bottom             0-839.5
Light brown organic
   silt                           ...
Light gray/pink clay              ...
Light gray silt               839.5-858
Brown organic silt              858-892.5
Clay                          892.5-898.5
Clay/sand                         ...
Gravel [+ or -] clay/sand         ...

                         Depth below sediment/water interface (cm)
        Sediment
        lithology        Chase Pond

Dark brown gyttja,
   grading to organic
   silt toward bottom             0-517
Light brown organic
   silt                           ...
Light gray/pink clay              ...
Light gray silt                 517-533.5(*)
Brown organic silt            533.5-590.5
Clay                          590.5-608.5
Clay/sand                     608.5-620.5
Gravel [+ or -] clay/sand         ...

                         Depth below sediment/water interface (cm)
        Sediment
        lithology        Main-a-Dieu

Dark brown gyttja,
   grading to organic
   silt toward bottom             0-488
Light brown organic
   silt                         488-493.5
Light gray/pink clay          493.5-502
Light gray silt                   ...
Brown organic silt              502-530
Clay                            530-572
Clay/sand                         ...
Gravel [+ or -] clay/sand         ...




(*) Occasional stones > 0.5 cm.

The above stratigraphic changes are more clearly revealed in the percent loss-on-ignition (LOI) curves (Figs. 3, 6, 8, 10, 13, and 15), showing changes in the organic content of sediment. Changes in the percent LOI curves have been used to define the precise Younger Dryas boundaries; we have taken the point where the percent LOI begins to drop as the lower boundary, and the point where it returns to that value as the upper boundary (see Mayle et al. 1993a for discussion of these criteria). At all of the sites, changes in percent LOI precede the onset of observable lithological transitions, described above, by several centimetres.

[Figure 3, 6, 8, 10, 15 ILLUSTRATION OMITTED]

Although percent LOI curves show changing organic content of sediment, they do not reveal whether a drop in percent LOI is due to a decrease in lake productivity or an increase in erosion of mineral matter from the surrounding lake catchment into the lake. Chronologies for Splan Pond and Lac a Magie have allowed curves to be plotted for influx of mineral and organic matter into these lakes (Figs. 4 and 11). It is clear at both these sites that the decrease in percent LOI during the Younger Dryas is largely due to increased influx of mineral matter, since the latter rises approximately 10-fold, while organic influx remains relatively constant throughout the sequence.

[Figure 4, 11 ILLUSTRATION OMITTED]

Chronology

A late-glacial chronology was developed from a series of accelerator-mass-spectrometer (AMS ) [sup.14]C dates (Table 2). Younger Dryas boundary age estimates were obtained by extrapolation from dates close to the boundaries within the Younger Dryas sediment. In the case of Splan Pond, however, a linear regression was plotted through the Younger Dryas dates, using the equation:

y = 4392.2 + 12.26x (y = age, x = depth).

[TABULAR DATA 2 NOT REPRODUCIBLE IN ASCII]

This linear regression was then extrapolated to the Younger Dryas boundaries. Refer to Mayle et al. (1993a) for a full discussion of the dates and age estimates for the Younger Dryas boundaries.

Age-depth relationships for Splan Pond and Lac a Magie were determined by (1) interpolation between the early Holocene date and the upper Younger Dryas boundary age estimate; (2) interpolation between both Younger Dryas boundary age estimates; and (3) interpolation between the lower Younger Dryas boundary age estimate and the date for the basal organics, and extrapolation of the curve to the base of the core. When constructing the pollen accumulation rate diagram for Splan Pond (Fig. 4), levels 532, 531, and 470 were omitted, since extrapolation from dates across sharp sediment boundaries resulted in artificially high influx values at these levels (confirmed by comparison with concentration data). Although seven dates were obtained for the Chase Pond core, the lateglacial dates are considered anomalously old (Levesque et al. 1993, Mayle et al. 1993a); therefore, no pollen accumulation rate diagram was plotted for this site.

Pollen assemblage zones and macrofossils

The six sites can be divided into two groups: those that were forested prior to the Younger Dryas (Splan Pond [Figs. 3-5], Mayflower Lake [Figs. 6 and 7], and Little Lake [Figs. 8 and 9]), and those that were not (Lac a Magie [Figs. 10-12], Chase Pond [Figs. 13 and 14], and Main-a-Dieu Pond [Fig. 15]). Within these two groups, pollen zones can be identified that are common to each of the three sites. Pollen zones are numbered and prefixed by "F" and "T" to indicate "forest" or "tundra" sites respectively.

[Figures 5, 7 ILLUSTRATION OMITTED]

Forested sites (Figs. 3-9). --Zone F1 is characterized by a diverse assemblage of herbaceous taxa, most of which are either confined to this zone, or reach maximum percentages here: Cyperaceae (10-20%), Poaceae (5%), Artemisia (5%), Plantago, Dryas, Leguminosae, Rumex/Oxyria, Caryophyllaceae, Tubuliflorae, and Selaginella rupestris-type (ranging from [is less than] 5 to 10%). The following arboreal taxa are also abundant: Picea (10-25%), Pinus (25-40%, decreasing further up), and Quercus (5-10%). Other common taxa are: Betula (Splan Pond [Fig. 3], 25% at the base, decreasing upwards), Salix (10 and 20% at Mayflower Lake [Fig. 6] and Little Lake [Fig. 8], respectively), and Dryopteris (Little Lake, rising sharply to 40% in the middle of the zone). However, pollen accumulation rates (PARs) for all taxa at Splan Pond (Fig. 4) are very low. Macrofossil assemblages in this zone are dominated by Cyperaceae, Dryas integrifolia, and at Little Lake, Caryophyllaceae. At Little Lake macrofossils of shrub taxa are also common: Salix herbacea, Vaccinium uliginosum, Empetrum, Arctostaphylos, Cassiope hypnoides, and Kalmia.

Zone F2 is characterized by high percentages of Betula (20-45%), Populus (5-45%), and Cupressaceae (5-10%), all of which increase toward the upper boundary, whereas the proportion of herbs declines. Salix and Cyperaceae are common near the base, but decrease further up to [is less than]3% and [is less than]15%, respectively. In addition, Artemisia at Mayflower Lake and Myrica at Little Lake reach maximum percentages during this zone of 20 and 35%, respectively. PARs for Splan Pond show similar changes, although PARs show that Cyperaceae increases rather than decreases further up the zone. The macrofossil assemblages are characterized by Betula glandulosa, Salix, Vaccinium uliginosum, Myrica gale, Empetrum, Arctostaphylos, Cyperaceae, and Dryas integrifolia.

Zone F3 is dominated by maximum percentages of Picea (35-60%). The lower boundary is placed at the point in the Picea curve corresponding to 10%. Cupressaceae, Betula, and Populus all decline toward the top, although at Little Lake these taxa fluctuate considerably through the zone. Other common taxa at Little Lake are: Myrica (15-25%), Lycopodium (10%), and Dryopteris (10%). Cyperaceae is common at Splan Pond and Mayflower Lake (15-25% at the base, decreasing to [is less than]10% toward the top). These percentage changes are reflected by the PARs at Splan Pond. Macrofossil assemblages in this zone are characterized by Picea, Juniperus, Betula glandulosa, Myrica gale, Salix, Vaccinium, Empetrum, Arctostaphylos, and, at Splan Pond (Fig. 5), Cyperaceae.

Zone F4 is characterized by a sharp drop in percentages of Picea to 2-15%, a rise in Alnus crispa from negligible values to 3-10%, and a large increase in pollen of herbs. Dominant taxa are: Pinus (10-30%), Betula (15-40%), Salix (5%), Cyperaceae (10-30%), and Dryopteris-type (10-[is greater than]50%). Cupressaceae, Populus, and Quercus disappear or decrease to negligible values. Myrica also decreases during this zone, values ranging from 5% to negligible. Pinus strobus increases from trace values at the base of the zone to 5-10%. Poaceae, Tubuliflorae, and Caryophyllaceae increase at Splan Pond (2-5%), although similar patterns for these taxa are not evident at Mayflower Lake or Little Lake. At Splan Pond and Mayflower Lake, percentages of indeterminable grains increase to [is greater than]5%. PARs for Splan Pond closely match the percentage changes, although influx for total Pinus rises only slightly and generally remains below 300 [grains.cm.sup.-2].[yr.sup.-1] throughout most of the zone. Macrofossil assemblages are characterized by Picea, Betula glandulosa, Salix herbacea, Vaccinium uliginosum, Cyperaceae, Juncus, and Dryas integrifolia. Concentrations and distributions of these taxa through the zone vary widely between sites. At Splan Pond and Little Lake, concentrations of Picea are greatest in the lower third of the zone, and are low further up, while at Mayflower Lake (Fig. 7), Picea is present only near the zone boundaries. At Splan Pond, several herbaceous taxa appear: Potentilla-type, Asteraceae, Caryophyllaceae, and Monarda-type. Unique to this site are isolated occurrences of Pinus strobus macrofossils within this zone.

Zone F5 is characterized by increases in Picea (20-40%), Cupressaceae (2-10%), Populus (2-10%), Quercus ([is greater than]5-[is greater than]10%), and Myrica ([is greater than]5-[is greater than]15%). The lower boundary is placed at the point in,the Picea curve corresponding to 10%. Larix and Abies first appear in the middle of the zone and increase toward the top. Pinus strobus increases throughout the zone at Splan Pond and Little Lake to [is greater than]10% and [is greater than]15%, respectively, at the upper boundary. At Mayflower Lake P. strobus is common throughout at [approximately equals]10%. P. banksiana/resinosatype decreases from 5% at the base to [is less than]2% at the top of the zone. The behavior of Betula varies between the sites. It decreases to [is less than]15% at Splan Pond, but at Mayflower Lake and Little Lake it increases to [is greater than]30% and 45%, respectively. Other common taxa are Lycopodium (5-15%) and Dryopteris (10-15%). These percentage changes closely parallel the PAR changes for Splan Pond. Macrofossil assemblages vary considerably between sites. All three sites are characterized by Picea, Myrica gale, and Cyperaceae. Other taxa include: Larix, (Splan Pond and Mayflower Lake), Abies (Splan Pond), and at Little Lake, Pinus strobus (near the upper boundary), B. papyrifera, and B. populifolia. Salix is only present at or near the lower boundary.

Zone F6 is marked by significant variation between the three sites, possibly due to the fact that the upper boundary of the zone is the top of the spectrum and is therefore undefined. The zone is characterized by decreases in Picea ([is less than]7%), Cupressaceae (negligible values), Populus ([is less than]3%), Lycopodium ([is less than]2%), and Dryopteris-type ([is less than]3%). At Splan Pond and Little Lake, Larix and Abies also decrease to 3% or less. Pinus strobus rises to 15-35% and Betula rises to [is greater than]40% at Splan Pond and Mayflower Lake. At Little Lake Betula decreases to [is less than]30% at the upper boundary. In addition, Tsuga canadensis rises above negligible values to 4% at Splan Pond. Quercus and Myrica remain common at 5-10%, and Alnus crispa rises to 10% at Splan Pond and Little Lake. These pollen changes are closely reflected in the PAR diagram for Splan Pond (Fig. 4). Macrofossil assemblages are characterized by Picea, Larix, Abies, Pinus strobus, B. papyrifera, B. populifolia (Little Lake), Alnus crispa (Little Lake), Myrica gale, and Cyperaceae.

Tundra sites (Figs. 10-15).-Zone T1 is dominated by Pinus (10-20%), Betula (10%), Salix (10%), Ericales (5-15%), and Cyperaceae (10-20%). Also common are: Shepherdia canadensis (5-10%) and Dryopteris-type (5-10%) at Chase Pond (Fig. 13) and Main-a-Dieu Pond (Fig. 15), and Selaginella rupestristype (3-5%) at all three sites. The PARs for Lac a Magie (Fig. 11), however, are low for all taxa during this zone. The sparse macrofossil assemblages (Figs. 12 and 14) are characterized by Salix, Vaccinium, Cyperaceae, and Dryas integrifolia.

[Figure 9 ILLUSTRATION OMITTED]

Zone T2 is characterized by high percentages of Betula ([is greater than]15 45%), low ([is less than]5%) but consistent values of Populus, and peaks in Cupressaceae and Myrica (2025% at Lac a Magie [Fig. 10] and 2-[is greater than] 5% at Chase Pond and Main-a-Dieu Pond). Changes in other taxa vary between sites. Other common taxa at Lac a Magie are: Picea (10%), Pinus banksiana/resinosa (10%), Quercus ([is greater than]5%), Poaceae ([is greater than]10%), Tubuliflorae (5%), Lycopodium ([is greater than]5%), and Dryopteris-type ([is greater than]10%). At Chase Pond common taxa are: Pinus ([is less than] 10%), Salix (510%), Ericales (5%), Cyperaceae (10-25%), Poaceae ([is less than]10%), Artemisia (10%), Tubuliflorae (5%), and Lycopodium ([is less than]5%). Common taxa at Main-a-Dieu Pond are: Ericales (10-20%), Cyperaceae (20-25%), Poaceae (5%), Lycopodium (10%), and Dryopteris-type (10%). The patterns in percentages are reflected in the PAR diagram for Lac a Magie (Fig. 11), although the PARs for most taxa are [is less than]300 [grains.cm.sup.-2].[yr.sup.-1] Macrofossil assemblages are characterized by Juniperus communis, Betula glandulosa, Myrica gale, Salix, Vaccinium, Empetrum, Arctostaphylos, Cyperaceae, and Dryas integrifolia.

Zone T3 is characterized by decreases in Cupressaceae, Populus, and Quercus, as Alnus crispa (410%), Cyperaceae ([is greater than]5-35%), and herbs in general, rise. At Lac a Magie total Pinus remains approximately constant through this zone, although P. banksiana/resinosa-type decreases to 6%, while P. strobus first appears above negligible values at the base and increases to 10% at the top of the zone. Other changes at this site are: increases in Betula (15-20%), Ericales (4%), and Dryopteris-type (25%), and decreases in Picea (5%) and Myrica (10%). PARs for these taxa closely match the percentage changes at this site. At Chase Pond and Main-a-Dieu Pond, increases occur in: Pinus (15%), Salix (4-7%), several herbs (Artemisia, Ambrosia, Rumex/Oxyria, Caryophyllaceae, Epilobium, Thalictrum, and Dryopteris-type [Chase Pond], ranging from 1-8%), and indeterminable grains (5-10%). Decreases occur in Betula ([is less than] 15%), Myrica ([is less than] 2%), and Ericales (2%), while at Main-a-Dieu Pond Lycopodium and Dryopteris-type also decrease to [is less than] 5%. Macrofossil assemblages are characterized by Betula glandulosa, Salix, Vaccinium, Empetrum, Cassiope hypnoides, Cyperaceae, Juncus, Dryas integrifolia, and Rumex.

Zone T4 varies considerably between the three sites, due in part, to the fact that the upper boundary is the top of the pollen spectrum and is therefore undefined. The zone is characterized at all three sites by increasing percentages of Picea (10%), Cupressaceae ([is less than] 5%), Betula (30-65%), and Quercus ([is less than] 5% at Chase Pond and Main-a-Dieu Pond, 15% at Lac a Magie), sharp decreases in Alnus crispa and Cyperaceae to [is less than] 2% throughout most of the zone, and a decrease in Salix from [is less than] 5% to negligible values by the middle of the zone. At Chase Pond and Main-a-Dieu Pond the Betula curve has two peaks, one near the lower boundary, and a second, larger peak toward the top of the zone. Changes in other taxa are site-specific: at Lac a Magie, Larix appears at the base of the zone, reaching 4% before the middle, and then declining gradually to 2% at the top. Abies increases toward the middle of the zone, rising to 7% at the top. P. banksiana/resinosa-type gradually declines through the zone, while P. strobus increases to [is greater than] 15% at the top. All shrubs and herbs decrease to [is less than] 5% by the middle of the zone, and [is less than] 2% by the upper boundary. The PAR changes in this zone for Lac a Magie match the percentage changes very well. At Chase Pond, Larix rises above negligible values to 2% at the top of the zone, and Populus occurs consistently at 2% in the upper half. Myrica peaks at 40% in the lower half, but then decreases to 5% at the upper boundary. At Main-a-Dieu Pond, however, Myrica rises to 20% near the base of the zone, and remains relatively constant to the upper boundary. Changes in other taxa are common to both Chase Pond and Main-a-Dieu Pond: Ericales and Poaceae rise to 5% and [is greater than] 10% near the base of the zone, but subsequently decline to [is less than] 2%. Lycopodium and Dryopteris-type reach maximum percentages of 15-43% and 8-25%, respectively. Tubuliflorae occurs at [is less than] 5%, but other herbs decline at the base of the zone to negligible values or disappear completely. The macrofossil assemblage at Lac a Magie is characterized by Picea, Larix, Abies, Pinus strobus, and Betula papyrifera. The Chase Pond assemblage is characterized in the lower half of the zone by B. glandulosa, Salix, Vaccinium, Empetrum, and Dryas integrifolia. The upper half of the zone is characterized by Picea, Larix, Abies, and B. papyrifera. Cyperaceae and Myrica gale occur throughout.

DISCUSSION

Since the purpose of this paper is to determine the response of vegetation to the cooling marking the onset of the Younger Dryas and warming marking the termination of this event, vegetation reconstructions will be limited to the Younger Dryas pollen assemblage zone, and the zones immediately preceding it and following it. In addition, although Levesque et al. (1993) have presented pollen evidence for a shorter-term climatic event (the Killarney Oscillation) preceding the Younger Dryas of most of these sites, this event is not always apparent in the pollen spectra of this study due to the coarser sampling resolution in pre-Younger Dryas sediment.

Impact of the Younger Dryas on boreal forest or woodland

The high percentages (35-60%) and PARs (1600 [grains.cm.sup.-2].[yr.sup.-1]) of spruce pollen in the upper half of zone F3, together with the presence of spruce needles, at Splan Pond, Mayflower Lake, and Little Lake, suggest that closed spruce forest or woodland was present locally at these sites just prior to the onset of the Younger Dryas, i.e., shortly before [approximately equal] 10770 yr BP (Mayle et al. 1993a). Evidence for widespread distribution of boreal forest or woodland in southern New Brunswick and central mainland Nova Scotia at this time comes from similarly high spruce pollen percentages from the following sites: Little Lake, southern New Brunswick (Most 1975, approximately equidistant between Splan Pond and Mayflower Lake); Chance Harbour Lake, near New Glasgow, Nova Scotia (Jette and Mott 1989); Shubenacadie and Lantz, central mainland Nova Scotia (Stea and Mott 1989); and Canoran Lake, south/central mainland Nova Scotia (Railton 1972, Mott 1992). Macrofossil assemblages of this zone shed light on the probable identity of several pollen taxa. The presence of Betula glandulosa (dwarf birch) macrofossils, and the fact that neither B. papyrifera or B. populifolia macrofossils appear until zones F5 and F6, indicates that birch pollen in this zone is most likely B. glandulosa. Also, the occurrence of Juniperus communis (ground juniper) macrofossils, coupled with the absence of Thuja occidentalis macrofossils, points to Juniperus communis as the source of Cupressaceae pollen. Macrofossil evidence suggests that Myrica pollen can probably be attributed to Myrica gale. Although total pine pollen exceeds 10% in this zone, the low PAR values for P. banksiana/resinosa at Splan Pond (200 [grains.cm.sup.-2].[yr.sup.-1]) suggest that it is due to long-distance transport. Similarly, the low percentages of oak pollen in zone F3 ([is less than] 4%) in combination with very low PARs at Splan Pond ([is less than] 100 [grains.cm.sup.-2].[yr.sup.-1]) also indicate that oak is unlikely to have been present in Atlantic Canada at this time.

These sites were near the range limit of expanding spruce forest at this time, since low spruce pollen percentages ([is less than] 10%) and the absence of spruce needles indicate that spruce forest was absent from more northerly sites such as Joe Lake (Mayle et al. 1993a), Collins Pond (Stea and Mott 1989), Gillis Lake (Livingstone and Livingstone 1958, Mott 1992), Chase Pond, and Main-a-Dieu Pond, prior to the early Holocene (10 000 yr BP). Open areas in the spruce woodland or forest were likely occupied by shrubs and herbs such as dwarf birch, ground juniper, willow, ericads such as Vaccinium uliginosum, Empetrum, and Arctostaphylos, and sedge. Sweet gale grew around the lake shore. The association of spruce with arctic/boreal species such as Vaccinium uliginosum, dwarf birch, and ground juniper suggests a flora similar to that of the modern northern boreal forest in sub-arctic North America.

At Splan Pond the 60% spruce pollen peak in zone F3 corresponds with the 20% peak in LOI, the latter marking the lower Younger Dryas boundary. The subsequent sharp decline in percent LOI and spruce pollen percentages signals the Younger Dryas and the transition between zones F3 and F4, as argued by Mott et al. (1986). At this site, the lithological change from organic silt to Younger Dryas clay (drop in LOI from 19 to 4% across only 1.5 cm) occurs in only 25 yr. Spruce pollen percentages across the zone boundary drop from [is greater than] 55% to [is less than] 15% in only 50 yr, and spruce PARs drop eightfold from [is greater than] 1600 [grains.cm.sup.-2].[yr.sup.-1] to 200 [grains.cm.sup.-2].[yr.sup.-1] over the same time period. The absence of complete chronologies for Mayflower Lake and Little Lake precludes estimates of rates of sediment and spruce percentage changes for these sites, but the similarity of the LOI and pollen profiles suggests that they probably occurred over a similar time frame. At all three sites, the onset of the spruce pollen decline coincides with the decline in LOI values marking the lower Younger Dryas boundary. Other taxa, however, do not show percentage changes until a few centimetres above the lower Younger Dryas boundary (e.g., increases in Betula, Salix, Cyperaceae [Little Lake], and Dryopteris-type [Little Lake]), suggesting that these taxa showed a lag in response. The fact that certain taxa, such as Cyperaceae, do not show lags at every site may indicate that establishment of a taxon is dependent on local site conditions, such as changing soil factors or competition from other species, rather than any intrinsic response lag that is characteristic of the taxon.

The pollen and macrofossil assemblages of zone F4 are indicative of a shrub-tundra or forest-tundra. The strongest evidence for vegetation response to a severe climatic cooling comes from the abundance of arctic/ alpine or boreal species in the macrofossil record, such as Vaccinium uliginosum, Salix herbacea, and Dryas integrifolia. These taxa are restricted to the tundra pollen assemblage zones F1, F2, and F4, and are generally found in tundra or forest-tundra today (Porsild and Cody 1980), although V. uliginosum occurs in bogs within the southern boreal and mixed forests. Although macrofossil evidence demonstrates that dwarf birch was present during zone F3, the pollen data indicate that populations increased gradually through the Younger Dryas, peaking toward the top of zone F4. The coincident sharp declines in spruce pollen percentages and PARs at the onset of the Younger Dryas indicates that the climate became too cold for spruce, although persistence of Picea needles throughout the Younger Dryas sediment of Splan Pond and Little Lake suggests that scattered, localized populations of spruce probably survived, e.g., around the shores of these sheltered lake basins. The high concentrations of spruce needles in the lower third of zone F4 appear to contradict the pollen record. It is likely, however, that these high concentrations are due to increased erosion of soils and associated litter from the surrounding lake catchment, rather than an increase in local spruce populations. A likely explanation for this increased erosion is reduced soil stability caused by an opening up of the vegetation cover (due mainly to a decline in spruce), possibly in combination with renewed freeze-thaw processes such as solifluction; there is evidence from mainland Nova Scotia (Stea and Mott 1989) and possibly New Brunswick (Lamothe, unpublished data) for renewed glaciation during the Younger Dryas. The presence of numerous angular stones [is greater than] 0.5 cm in the Younger Dryas sediment attests to the magnitude of erosion.

Although P. strobus pollen first appears at the base of this zone, low percentages ([is less than] 8%) and PARs ([is less than] 300 grains.[cm.sup.-2].[yr.sup.-1]) would suggest that this is attributable to long-distance transport from populations in New England. Isolated occurrences of macrofossils, however (Figs. 2, 5), indicate that P. strobus was present locally. The absence of P. strobus pollen or macrofossils in preceding zones eliminates the possibility that they may have been reworked from older late-glacial deposits. The local presence of P. strobus together with arctic/boreal species such as Salix herbacea and Dryas integrifolia is difficult to explain ecologically, since the modern northern range of white pine does not extend beyond the boreal forest of southern Ontario, southern Quebec, and Newfoundland. The notion that a thermophilous species such as white pine could migrate into the region at low densities at a time of rapid cooling when spruce populations were crashing is admittedly difficult to accept, but it should be borne in mind, though, that according to the Milankovitch astronomic theory of climate change, the Younger Dryas period was a time of maximum summer insolation (Berger 1978), with northern hemisphere summers receiving 8% more solar radiation than at present. This enhanced seasonality may account for the unusual association of arctic/boreal species with more thermophilous taxa such as white pine. Plant macrofossil and pollen influx studies by Richard and Larouche (1989) also provide evidence for unusual vegetation assemblages in early successional sequences from sites in Temiscamingue, Quebec. The authors state that "the initial vegetation was dominated by Picea and Larix and included Quercus (oak), Ostrya (ironwood), Ulmus (elm), and Fraxinus (ash) along with plants having a present-day mainly arctic distribution such as Dryas integrifolia and Silene acaulis (moss-campion)."

Although Alnus crispa pollen increases at all three sites during the Younger Dryas, the-low percentages (3-10%) and PARs ([is less than] 200 grains.[cm.sup.-2].[yr.sup.-1]), and the absence of its macrofossils from any late-glacial sequences in Atlantic Canada, have led us to conclude that A. crispa pollen in zone F4 results from long-distance transport from populations in New England, especially in light of the fact that Alnus is a prolific pollen producer. See Mayle et al. (1993b) for a detailed discussion of the significance of the Younger Dryas alder peak.

The sharp increases in percentages and PARs of spruce and poplar and the decline in Cyperaceae over an increment of only 2-4 cm, estimated to have taken only 50 yr at Splan Pond, indicate that the vegetation response to the climate warming at the end of the Younger Dryas (the transition between zones F4 and F5) was also rapid. The pollen and macrofossil data indicate that spruce/poplar forest or woodland rapidly replaced shrub-tundra, although a brief phase of dwarf birch expansion precedes the increase in trees. Juniper/ cedar was probably common in the woodland also. This rapid establishment of forest and reduction in solifluction resulted in a stabilization of the surrounding slopes, causing an abrupt drop in erosion and surface runoff, as indicated by the sharp decline in mineral influx at Splan Pond and the return to organic sediment evident at all three sites. The subsequent forest succession appears to have been rapid at Splan Pond: in only 140 yr, spruce PARs declined from 1800 to 800 grains.[cm.sup.-2].[yr.sup.-1], after which macrofossil and pollen evidence indicates the local arrival and population expansion of tamarack and fir. Increasing pollen values indicate that white pine and oak may also have arrived at the site, although no macrofossils of these taxa have been found in this zone. At Little Lake, the rise in tamarack and fir at the expense of spruce in the middle of zone F5 is also accompanied by a sharp rise in birch pollen percentages, which, according to the macrofossil evidence, can be attributed to the arrival of white birch and grey birch at the site. The increasing pollen percentages and macrofossil data also indicate that white pine had arrived by the end of zone F5.

Given the geological and paleoecological evidence for a Younger Dryas cooling throughout Atlantic Canada, the rapidity with which arboreal taxa replaced the Younger Dryas shrub-tundra, and the speed of subsequent forest succession by invasion of new arrivals such as tamarack, fir, tree birch, and white pine, is remarkable. Although spruce probably expanded from sparse populations persisting through the Younger Dryas, we would have expected that the rate of forest succession following the end of the Younger Dryas would have taken considerably longer, since taxa such as white birch, tamarack, and fir would have required time to migrate northward from New England, where, according to Davis and Jacobson (1985), their northern range limits are presumed to have been during the Younger Dryas. The fact that these taxa increased in abundance so rapidly at these three sites in Maritime Canada suggests that they may have been migrating into the region during the Younger Dryas at very low densities, surviving in favorable microclimates or sheltered habitats. This scenario may be plausible for white pine, needles of which were recovered from Younger Dryas sediment.

We acknowledge, though, that use of [sup.14]C dating to estimate reliable rates of vegetation and climate change at the end of the Younger Dryas period is problematic in the light of a detailed AMS [sup.14]C chronology from a site in Switzerland (totter et al. 1992). The high resolution of dates reveals a [sup.14]C age plateau at 10000 yr BP, spanning the upper third of the Younger Dryas pollen assemblage zone III and the lower half of the Preboreal pollen assemblage zone IV. If this plateau is due to a change in the atmospheric [sup.12]C:[sup.14]C ratio around this time, it would have been a global phenomenon, affecting dates from Atlantic Canada as well as from Switzerland. Clearly, other dating methods, such as verve or dendrochronology, are required to circumvent this problem.

Impact of the Younger Dryas on shrub-tundra

The late-glacial pollen zones of Chase Pond and Main-a-Dieu Pond closely resemble those of Gillis Lake, located near to Chase Pond (Livingstone and Livingstone 1958, Mott 1992), while the sequence at Lac a Magie is similar to that of nearby Brier Island Bog Lake, Digby Neck (Wilson et al. 1993).

The pollen and macrofossil assemblages of zone T2, immediately prior to the Younger Dryas, at Chase Pond and Main-a-Dieu Pond indicate that shrub-tundra dominated by B. glandulosa, Salix, Vaccinium uliginosum, Empetrum, and Arctostaphylos preceded the Younger Dryas event. Low-lying woody plants such as Dryas integrifolia, and herbs such as Cyperaceae, Poaceae, Artemisia, Tubuliflorae, and at Main-a-Dieu Pond, Lycopodium and Dryopteris-type, were probably also common. Relative proportions of these taxa at Lac a Magie differ somewhat. Pollen data indicate that B. glandulosa was probably less abundant than at the other two sites, and was probably co-dominant with ground juniper. Dominant herbs were Poaceae, Lycopodium, and Dryopteris-type, while Cyperaceae and Tubuliflorae were probably minor components of the vegetation. The ecological significance of high percentages of Dryopteris-type spores in this zone is unclear. Pine pollen percentages below 10-15% at all three sites, and PARs for P. banksiana/resinosa below 500 grains.[cm.sup.-2].[yr.sup.-1] at Lac a Magie, probably reflect long-distance transport, rather than regional or local presence of populations. Even lower percentage and PAR values for oak indicate that this taxon is also unlikely to have been present. Pollen percentages of [approximately equals] 5% indicate that spruce (a less prolific pollen producer than pine) was absent from Chase Pond and Main-a-Dieu Pond at this time. However, at Lac a Magie, spruce PARs of 300-400 grains.[cm.sup.-2].[yr.sup.-1] near the upper boundary of zone T2 are comparable with those from recent lake sediments from forest-tundra of northwestern Canada and Alaska (e.g., Anderson 1985, Ritchie 1977), suggesting that isolated stands or pockets of spruce may have been present in the shrub-tundra of southwestern Nova Scotia at this time. Evidence that spruce pollen percentages [is less than] 20% may indicate local presence of spruce comes from studies of fossil beetles (Coleoptera) by Miller and Morgan (1991). These authors found fossil remains of the beetle Phloeotribus piceae (an indicator of boreal trees) below Younger Dryas sediment at Lismore (a buried organic site near Antigonish, northern mainland Nova Scotia), although palynological studies of the same stratigraphic unit (Stea and Mott 1989) show spruce pollen percentages [is less than] 20%.

At Lac a Magie, the decrease in LOI values from 17 to 7%, which begins 3 cm below the boundary between pollen zones T2 and T3, occurs in [approximately equals] 110 yr and signifies the onset of the Younger Dryas cooling. The increase in mineral influx responsible for this decline in LOI is attributed to increased erosion due to destabilization of the surrounding slopes, as for the forested sites. At Lac a Magie, PARs show that rates of response and duration of lags, if any, vary between taxa. PARs of herbaceous taxa begin to increase coincidentally with the beginning of the LOI decline (the onset of declining LOI values is used to define the lower Younger Dryas boundary), influx of total herbs rising from 1400 to 2900 grains.[cm.sup.-2].[yr.sup.-1] in 130 yr. Herbaceous taxa, therefore, appear to have shown an immediate response to the climate cooling, but took [is greater than] 100 yr to reach maximum pollen values in zone T3. The rise in dwarf birch PARs, on the other hand, lags the onset of the Younger Dryas by 80 yr (3 cm) and takes a further 80 yr to reach maximum values in this zone. Similar lags of 3 and 5 cm occur at Main-a-Dieu Pond and Chase Pond respectively.

At Lac a Magie, the pollen percentage, PAR, and macrofossil data in zone T3 show changes in the floral composition of the shrub-tundra. Ground juniper virtually disappeared, while shrubs (B. glandulosa, Salix, Vaccinium, and Empetrum) and herbs (Cyperaceae, Artemisia, Tubuliflorae, Lycopodium, and Dryopteristype) increased. Ground juniper is the most thermophilous taxon of the shrub-tundra assemblage of zone T2, since today it extends only slightly beyond the northern limit of trees, and then only at favorable microhabitats such as south-facing river bluffs (Porsild and Cody 1980). Its disappearance from zone T3 and increases in more northerly ranging taxa such as dwarf birch, which today grows as far north as 70 [degrees] N latitude, and total herbs demonstrate a marked vegetation response to a colder climate. At Chase Pond and Main-a-Dieu Pond, the pollen percentage and macrofossil data indicate that the Younger Dryas caused the replacement of shrub-tundra by herb-tundra at the transition between zones T2 and T3. The shrubs B. glandulosa and Myrica decrease. Although Vaccinium twigs are present throughout zone T3 at Chase Pond, pollen data suggest that the abundance of most ericads declined. Macrofossils of the arctic/alpine dwarf shrubs Cassiope hypnoides (moss heather) and Salix herbacea at the top of this zone indicate a herb-tundra environment in the late Younger Dryas. The Salix pollen peak in this zone is probably attributable to increasing populations of S. herbacea. Increases in a number of herbaceous taxa, such as Cyperaceae, Dryas integrifolia, Artemisia, and Caryophyllaceae, are also indicative of a herb-tundra environment. Ambrosia, Rumex/Oxyria, Epilobium, and Thalictrum are herbaceous taxa unique to this assemblage, although some of these are present in the herb-tundra pollen assemblage zone T1, immediately following deglaciation. The presence of a Rumex seed in zone T3 of Chase Pond indicates that at least some of the Rumex/Oxyria pollen may be attributable to Rumex. Although Pinus pollen increases to 15% at both of these sites, influx values of [approximately equals] 400 grains.[cm.sup.-2].[yr.sup.-1], corresponding to similar percentages at Lac a Magie and Splan Pond, show that pine pollen in this zone is probably due to long-distance transport. Further support for this interpretation comes from the fact that the sharp increase in pine PARs and first appearance of macrofossils (indicating local presence of pine) do not occur until the base of zone F6 at Splan Pond (the most southerly of the six sites). As with the forested sites, the peak in Alnus crispa pollen is attributed to long-distance transport (see discussion in Mayle et al. 1993b).

The latitudinal difference and the associated temperature gradient between Cape Breton Island and southern Nova Scotia probably account for the differences in late-glacial vegetation changes between Lac a Magie and the more northerly Chase Pond and Maina-Dieu Pond. Except for the absence or scarcity of spruce at Lac a Magie, the late-glacial vegetation around this site appears to have been very similar to that of the three sites that were forested prior to the Younger Dryas, suggesting that the late-glacial climate was similar across southern New Brunswick and southern and central mainland Nova Scotia. The climate prior to the Younger Dryas at Lac a Magie may have been suitable for spruce forest, but spreading spruce populations had probably not yet reached the area. Pollen percentage data indicate that by 11 000 yr BP, spruce forest had reached southern New Brunswick (Most 1975) and central mainland Nova Scotia (Stea and Mott 1989, Mott 1992). The absence of spruce forest from southern and northern Nova Scotia by this time (the onset of the Younger Dryas) suggests that the most likely route of spruce migration into Atlantic Canada was via Maine, spreading along the southern coast of New Brunswick, across the isthmus and into central mainland Nova Scotia, whence it began to expand northward into Cape Breton Island and southward toward Lac a Magie. This pattern of spread differs from the route proposed by Green (1987), who postulated two simultaneous waves of spruce migration: one passing from the USA across Georges Bank and Browns Bank (which would have been at least partially above sea level at this time) of the Gulf of Maine to southern Nova Scotia by 11900 yr BP, and subsequently migrating northward; and the other spreading from Maine across southern New Brunswick and through the Chignecto Isthmus to converge with the other wave by 10000 yr BR Green's proposed pattern of migration is based on pollen records from sites with bulk-sediment [sup.14]C chronologies. Comparison with AMS [sup.14]C dates from other sites in the region (Mayle et al. 1993a) shows that many of these bulk-sediment dates, and the resulting estimates of time of spruce arrival in Nova Scotia, are most likely anomalously old.

As with Splan Pond, Mayflower Lake, and Little Lake, vegetation succession in response to climate warming at the end of the Younger Dryas at Lac a Magie appears to have been very rapid. The increases in PARs for arboreal taxa such as spruce, white pine, and oak, at the expense of shrubs and herbs, such as Ericales and Cyperaceae, occurred in 50-100 yr. Macrofossil and pollen data indicate that by 9870 yr BP the lake was surrounded by a woodland dominated by spruce, tamarack, and possibly oak. By [approximately equals] 9700 yr BP, populations of fir and white birch had arrived. White pine may also have arrived by this time, and macrofossil evidence shows that it was certainly present by 9180 yr BR As discussed above, the rapid arrival of tamarack and fir in the early Holocene may be accounted for by their migration into the region at low densities during the Younger Dryas, surviving in sheltered habitats. The early Holocene vegetation at Chase Pond and Main-a-Dieu Pond is quite different from that of Lac a Magie, no doubt due to the higher latitudes of these two sites and their greater distance from the migrating tree populations. At these two sites, the end of the Younger Dryas is marked by a return to shrub-tundra, shown by a rise in dwarf birch, sweet gale, and Ericales (Chase Pond), at the expense of most of the herbaceous taxa. Poaceae and Lycopodium also increase, but the ecological significance of this is unclear. Chase Pond macrofossil and pollen data indicate that by the middle of zone T4, the shrub-tundra was replaced by woodland or forest dominated by spruce. Tamarack and white birch appear shortly afterwards. Concentrations of tamarack needles and percentages of birch pollen (macrofossil data indicates that this is most likely white birch) subsequently increase rapidly, while spruce needle concentrations and pollen percentages decrease, so that by the top of zone T4 the local forest is dominated by tamarack and white birch. Fir also arrived by this time. Although a macrofossil record is absent from Main-a-Dieu Pond, a similar pollen assemblage for zone T4 indicates that a similar successional sequence probably occurred at that site.

CONCLUSIONS

In relation to our objectives, the following conclusions can be drawn. (1) The vegetation changes brought about by the Younger Dryas were both quantitative and qualitative in nature; e.g., spruce and Cyperaceae decreased and increased respectively in response to the

cooling, while subshrubs and herbs such as Dryas integrifolia, Rumex/Oxyria, Ambrosia, and Thalictrum were, at most sites, confined to the Younger Dryas period. (2) The rate of vegetation response to the onset of the Younger Dryas varied between taxa and between sites. At sites that were forested prior to the Younger Dryas, the onset of the spruce response coincides exactly with the onset of sedimentological changes, and the drop in spruce pollen percentages and PARs to Younger Dryas levels took only 50 yr at Splan Pond. The duration of lags and rates of change of other taxa such as Betula, Salix, Cyperaceae, and Dryopteris-type also vary between sites, suggesting that they may be controlled by site-specific factors such as competition and edaphic conditions, rather than any intrinsic lag in response that is characteristic of the taxon. At Splan Pond and Lac a Magie, lags in the response of vegetation to the onset of the Younger Dryas cooling (inferred from the sedimentological changes) vary from negligible to [approximately equals] 80 yr. (3) The rate at which vegetation responded to the subsequent rapid warming at the end of the Younger Dryas also appears to have been rapid, with forest-tundra or shrub-tundra changing to woodland or forest in only 50-100 yr. Other paleoecological studies provide evidence that spruce can respond very fast to abrupt climate warming. Studies from the boreal tree line of central Canada (MacDonald et al. 1993) provide evidence that the transformation from tundra to forest-tundra, in response to a climate warming between 5000 and 4000 yr BP, took only 150 yr. We acknowledge, however, that our estimates of rate of response may be inaccurate due to the radiocarbon plateau at about 10000 yr BP (totter et al. 1992). Also, the rate of vegetation change at a site will depend on the distance of migrating populations from the site. (4) The macrofossil data have facilitated a detailed reconstruction of the late-glacial vegetation. From the macrofossil evidence, it is clear that the spruce forest immediately preceding the Younger Dryas at Splan Pond and Little Lake contained typical arctic/boreal species such as Vaccinium uliginosum, dwarf birch, and ground juniper, a floral assemblage similar to that of the modern northern boreal forest in subarctic North America. Macrofossil presence of arctic/alpine taxa such as Salix herbacea, Cassiope hypnoides, and Dryas integrifolia points to the severity of the Younger Dryas cooling. (5) It has not been clear from previous research whether the drop in organic content of sediment during the Younger Dryas results primarily from a decline in deposition of organic matter or an increase in deposition of inorganic matter. Influx analysis at two sites confirms the notion that the relative decline in organic content is due to an absolute increase in deposition of mineral matter into the lake, i.e., there was an increase in erosion rates during the Younger Dryas.

[Figure 15 ILLUSTRATION OMITTED]

ACKNOWLEDGMENT

We thank the following people: Tim Keenan and Lisa Doner for assistance with fieldwork; Andre Levesque, both for field work assistance and helpful discussions throughout the course of the research and preparation of the original manuscript; Pierre Richard and Robert Mott for reviewing an earlier draft of the manuscript. The date of 11 820 [+ or -] 90 yr BP was determined by Beta Analytic Inc.; all other dates were determined by IsoTrace at the University of Toronto, a facility supported by the Natural Science and Engineering Research Council of Canada (NSERCC). This research was supported by a NSERCC Research grant to Les Cwynar.

LITERATURE CITED

Alley, R. B., D. A. Meese, C. A. Shuman, A. J. Gow, K. C. Taylor, P. M. Grootes, J. W. C. White, M. Ram, E. D. Waddington, P. A. Mayewski, and G. A. Zielinski. 1993. Abrupt increase in Greenland snow accumulation at the end of the Younger Dryas event. Nature 362:527-529.

Anderson, P. M. 1985. Late Quaternary vegetational change in the Kotzebue Sound area, northwestern Alaska. Quaternary Research 24:307-321.

Anderson, T. W. 1983. Preliminary evidence for late Wisconsinan climatic fluctuations from pollen stratigraphy in Burin peninsula, Newfoundland. Pages 185-188 in Current research. Part B. Geological Survey of Canada. Paper 83-1B. Canadian Government Publishing Centre, Ottawa, Canada.

Atkinson, T. C., K. R. Briffa, and G. R. Coope. 1987. Seasonal temperatures in Britain during the past 22,000 years, reconstructed using beetle remains. Nature 325:587-592.

Berger, A. L. 1978. Long term variations of daily insolation and Quaternary climatic change. Journal of Atmospheric Science 35:2362-2367.

Birks, H. J. B. 1976. Late-Wisconsinan vegetational history at Wolf Creek, central Minnesota. Ecological Monographs 46:395-429.

Cwynar, L. C., E. Burden, and J. H. McAndrews. 1979. An inexpensive sieving method for concentrating pollen and spores from fine-grained sediments. Canadian Journal of Earth Sciences 16:1115-1120.

Davis, R. B., and G. L. Jacobson, Jr. 1985. Late glacial and early Holocene landscapes in northern New England and adjacent areas of Canada. Quaternary Research 23:341368.

Dean, W. E. 1974. Determination of carbonate and organic matter in calcareous sediments and sedimentary rocks by loss on ignition: comparison with other methods. Journal of Sedimentary Petrology 44:242-248.

Faegri, K., and J. Iversen. 1975. Textbook of pollen analysis. Third edition. Munksgaard International Publishers, Copenhagen, Denmark.

Faegri, K., and J. Iversen. 1989. Textbook of pollen analysis. Fourth edition. John Wiley and Sons, Chichester, UK.

Fernald, M. L. 1991. Gray's manual of botany. Biosystematics, floristic and phylogeny series. Volume 2. Eighth edition. Dioscorides, Portland, Oregon, USA.

Grant, D. R., and L. H. King. 1984. A stratigraphic framework for the Quaternary history of the Atlantic Provinces, Canada. Pages 173-191 in R. J. Fulton, editor. Quaternary stratigraphy of Canada--a Canadian contribution to IGCP Project 24. Geological Survey of Canada. Paper 84-10. Canadian Government Publishing Centre, Ottawa, Canada.

Green, D. G. 1987. Pollen evidence for the postglacial origins of Nova Scotia's forests. Canadian Journal of Botany 65:1163-1179.

Hinds, H. R. 1986. The flora of New Brunswick. Primrose, Fredericton, New Brunswick, Canada.

Jette, H., and R. J. Mott. 1989. Palynostratigraphie du tardiglaciaire et de l'Holocene de la region du Lac Chance Harbour, Nouvelle-Ecosse. Geographie physique et Quaternaire 43:27-38.

Kapp, R. O. 1969. Pollen and spores Wm. C. Brown, Dubuque, Iowa, USA.

Levesque, A. J., F. E. Mayle, I. R. Walker, and L. C. Cwynar. 1993. A previously unrecognized late-glacial cold event in eastern North America. Nature 361:623-626.

Livingstone, D. A., and B. G. R. Livingstone. 1958. Lateglacial and postglacial vegetation from Gillis Lake in Richmond County, Cape Breton Island, Nova Scotia. American Journal of Science 256:341-359.

Lotter, A. F., B. Ammann, and M. Sturm. 1992. Rates of change and chronological problems during the late-glacial period. Climate Dynamics 6:233-239.

Loucks, O. L. 1962. A forest classification of the Maritime Provinces. Proceedings of the Nova Scotian Institute of Science 25:86-167.

Lowe, J. J., J. M. Gray, and J. E. Robinson, editors. 1980. Studies in the late-glacial of north-west Europe. Pergamon, Oxford, England.

MacDonald, G. M., T. W. D. Edwards, K. A. Moser, R. Pienitz, and J. P. Smol. 1993. Rapid response of treeline vegetation and lakes to past climate warming. Nature 361:243-246.

Mayle, F. E., A. J. Levesque, and L. C. Cwynar. 1993a. Accelerator-mass-spectrometer ages for the Younger Dryas event in Atlantic Canada. Quaternary Research 39:355360.

Mayle, F. E., A. J. Levesque, and L. C. Cwynar. 1993b. Alnus as an indicator taxon of the Younger Dryas cooling in eastern North America. Quaternary Science Reviews 12:295305.

McAndrews, J. H., A. H. Berti, and G. Norris. 1973. Key to the Quaternary pollen and spores of the Great Lakes region. Life Sciences Miscellaneous Publications, Royal Ontario Museum 61. University of Toronto Press, Toronto, Canada.

Miller, R. E, and A. V. Morgan. 1991. Late-glacial Coleoptera fauna from Lismore, Nova Scotia. Atlantic Geology 27: 193-197.

Moore, P. D., and J. A. Webb. 1978. An illustrated guide to pollen analysis. John Wiley and Sons, New York, New York, USA.

Mott, R. J. 1975. Palynological studies of lake sediment profiles from southwestern New Brunswick. Canadian Journal of Earth Sciences 12:273-288.

--. 1992. Postglacial climatic history of Nova Scotia from pollen and macrofossil evidence. Pages 1-20 in The climate of Nova Scotia. The Atmospheric Environment Service, Environment Canada. Canadian Government Publishing Centre, Ottawa, Canada.

Mott, R. J., D. R. Grant, R. Stea, and S. Occhietti. 1986. Late-glacial climatic oscillation in Atlantic Canada equivalent to the Allerod/Younger Dryas event. Nature 323:247250.

Peteet, D. M., J. S. Vogel, D. E. Nelson, J. R. Southon, R. J. Nickmann, and L. E. Heusser. 1990. Younger Dryas climatic reversal in northeastern USA? AMS ages for an old problem. Quaternary Research 33:219-230.

Porsild, A. E., and W. J. Cody. 1980. Vascular plants of continental Northwest Territories, Canada. National Museum of Natural Sciences, National Museums of Canada, Ottawa, Canada.

Railton, J. B. 1972. Vegetational and climatic history of southwestern Nova Scotia in relation to a South Mountain ice cap. Thesis. Dalhousie University, Halifax, Nova Scotia, Canada.

Rampton, V. N., R. C. Gauthier, J. Thibault, and A. A. Seaman. 1984. Quaternary geology of New Brunswick. Geological Survey of Canada, memoir 416. Canadian Government Publishing Centre, Ottawa, Canada.

Rawlence, D. J. 1988. The post-glacial diatom history of Splan Lake, New Brunswick. Journal of Paleolimnology 1: 51 -60.

Richard, P. 1970. Atlas pollinique des arbres et de quelques arbustes indigenes du Quebec. Le Naturaliste Canadien 97: 97-161.

Richard, P. J. H., and A. C. Larouche. 1989. La vegetation postglaciaire du Temiscamingue, Quebec, durant l'episode glaciolacustre Barlow. Canadian Journal of Botany 67:544558.

Ritchie, J. C. 1977. The modern and late-Quaternary vegetation of the Campbell-Dolomite uplands near Inuvik, N.W.T., Canada. Ecological Monographs 47:401 423.

Rowe, J. S. 1972. Forest regions of Canada. Department of the Environment, Canadian Forestry Service Publication No. 1300.

Seaman, A. A., B. E. Broster, L. C. Cwynar, M. Lamothe, R. F. Miller, and J. J. Thibault. 1991. Late glacial and postglacial events in coastal environments: southwestern New Brunswick. 1991 conference field excursion guide. Canadian Quaternary Association, Fredericton, New Brunswick, Canada.

Sissons, J. B. 1979. The Loch Lomond Stadial in the British Isles. Nature 280:199-203.

Stea, R. R. 1987. Quaternary glaciations, geomorphology and sea level changes: Bay of Fundy Region. International advanced course on late Quaternary sea level correlations and applications, Halifax, Nova Scotia, Canada. Symposium field trip guide in cooperation with IGCP project 200.

Stea, R. R., and R. J. Mott. 1989. Deglaciation environments and evidence for glaciers of Younger Dryas age in Nova Scotia, Canada. Boreas 18:169-187.

Stea, R. R., and R. J. Mott. 1990. Quaternary geology of Nova Scotia. Guidebook for field excursion. Nova Scotia Department of Mines and Energy, open file report 90-008.

Stea, R. R., and D. M. Wightman. 1987. Age of the Five Islands Formation, Nova Scotia, and the deglaciation of the Bay of Fundy. Quaternary Research 27:211-219.

Stockmarr, J. 1971. Tablets with spores used in absolute pollen analysis. Pollen et Spores 8:615-621.

Taylor, K. C., G. W. Lamorey, G. A. Doyle, R. B. Alley, P.M. Grootes, P. A. Mayewski, J. W. C. White, and L. K. Barlow. 1993. The "flickering switch" of late Pleistocene climate change. Nature 361:432-436.

Walker, I. R., R. J. Mott, and J. P. Smol. 1991. Allerod-Younger Dryas lake temperatures from midge fossils in Atlantic Canada. Science 253:1010-1012.

Watts, W. A. 1977. The Late Devensian vegetation of Ireland. Philosophical Transactions of the Royal Society of London B 280:273-293.

Wilson, S. E., I. R. Walker, R. J. Mott, and J. P. Smol. 1993. Climate and limnological changes associated with the Younger Dryas in Atlantic Canada. Climate Dynamics 8: 177-187.

Wright, H. E., Jr. 1967. A square-rod piston sampler for lake sediments. Journal of Sedimentary Petrology 27:957-976.

--. 1989. The Amphi-Atlantic distribution of the Younger Dryas paleoclimatic oscillation. Quaternary Science Reviews 8:295-306.
COPYRIGHT 1995 Ecological Society of America
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1995 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Mayle, Francis E.; Cwynar, Les C.
Publication:Ecological Monographs
Date:May 1, 1995
Words:13337
Next Article:Spatiotemporal variation in demographic transitions of a tropical understory herb: projection matrix analysis.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters