The Holocene 11,1 (2001) pp. 65–80 The palaeoecological history of the Praz- Rodet bog (Swiss Jura) based on pollen, plant macrofossils and testate amoebae (Protozoa) E.A.D. Mitchell,1 W.O. van der Knaap,2 J.F.N. van Leeuwen,2 A. Buttler,3 B.G. Warner4 and J.-M. Gobat1 (1Botany Institute, University of Neuchaˆtel, Rue Emile Argand 11, CH-2007 Neuchaˆtel, Switzerland; 2Geobotanical Institute, University of Bern, Altenbergrain 21, CH-3013 Bern, Switzerland; 3Laboratoire de Chrono-e´cologie, UMR 6565 CNRS, UFR des Sciences et Techniques, 16 route de Gray, Universite´ de Franche-Comte´, 25030 Besanc¸on, France; Swiss Federal Institute for Forest, Snow and Landscape Research (WSL/FNP), Antenne Romande c/o EPEL, 1015 Lausanne, Switzerland; 4Department of Geography and Wetlands Research Center, University of Waterloo, Ontario, Canada N2L 3G1) Received 1 December 1999; revised manuscript accepted 1 May 2000 Abstract: Stratigraphy, radiocarbon dating and analyses of pollen, plant macrofossils and testate amoebae were used to reconstruct the development and ecology of a small raised bog in a karst-dominated landscape in the Swiss Jura Mountains. Special focus was on past vegetation and on the history of Pinus rotundata in relation to anthropogenic and climatic influences. Testate amoebae were used to reconstruct past local soil pH and water-table depth. The inferred development of the Praz-Rodet bog typifies a classic hydroseral terrestrialization of a small basin. Two features are specific for this site. First, the bog was much wetter than today for a long period; according to our hypothesis, this only changed as a consequence of human activities. Second, two hiatuses are present at the coring location (Younger Dryas–early Preboreal, and 4700–2800 cal. yr BP), the former probably caused by low lake productivity due to cold temperatures and the latter by the erosional activity of the adjacent small river. The date of 2800 cal. yr BP for renewed peat accumulation may be related to climatic change (Subboreal–Subatlantic transition). Pollen indicators failed to show one hiatus: an apparently complete pollen sequence is therefore no guarantee of an uninterrupted sediment accumulation. Evidence of early minor human impact on the vegetation in the Joux Valley dates back to c. 6850 calendar years, congruous with the early Neolithic in the Jura Mountains. The history of Pinus rotundata appears to be more complex than previously believed. Human activity is clearly responsible for the present abundance of this species, but the tree was naturally present on the bog long before the first evidence of important human disturbance of the site (1500 cal. yr BP). It is suggested that, in karst-dominated landscapes, dense forests growing on mineral soils around raised bogs may significantly reduce summer evapotranspiration by acting as windbreaks. Forest clearance results in increased evapotranspiration, causing a lowering of the water table on the bog and a modification of the vegetation cover. This hypothesis has implications for the management of similar small raised bogs in karst-dominated landscape. Key words: pollen, plant macrofossils, testate amoebae, Lateglacial, Holocene, peat bog, human impact, Pinus rotundata, Switzerland.  Arnold 2001 0959-6836(00)HL447RP 66 The Holocene 11 (2001) Introduction Ombrogenous peatlands are not common in central and southern Europe. Continental and subcontinental climates with prolonged warm and dry summers are not suitable for widespread peat for- mation and bog development. This holds also for the Jura Moun- tains, where in addition the hydrogeomorphic conditions (karst) and the calcareous bedrock are unfavourable. Human activities have further eliminated or severely degraded bog remnants. The Joux Valley in western Switzerland and adjacent France neverthe- less preserves a number of living bogs, which are among the most southern in Europe. In this region, it is not clear to what degree humans have had an influence on shaping existing bog ecosystems. Is it possible that some modern bog plant communities are natural and have Figure 1 Map of Switzerland, showing the coring locality. not been affected by humans? The origins of Pinus rotundata (= P. uncinata = P. mugo s.l.) and the dominant vegetation com- munity on the bogs today (Pino mugo-Sphagnetum) have been 1500 mm) and cool mean temperatures (annual average ,5° C) topics of special interest. Some studies suggest that this com- in this region have been important for sustaining bog conditions. munity is natural and has remained intact since deglaciation The central and highest part is characterized by a raised bog (Chastain, 1952; Richard, 1961; Matthey, 1964; Royer et al., community with pine (Piceo-Vaccinienion uliginosi), an open 1978; Julve, 1983; Sandoz, 1987). Others have speculated that P. raised bog community (Sphagnion magellanici), and a bog pool rotundata was introduced onto the bogs in the last two or three community (Caricion lasiocarpae). A Sphagno-Piceetum com- centuries and is not a postglacial relict (Feldmeyer-Christe, 1990; munity exists around the periphery. A small stream cuts through Reille, 1991). the southeast edge. Open pasture (Cynosurion) surrounds the bog The latitudinal position of the Jura Mountains near the southern at the northeastern to northwestern side and a small river (Orbe) limit of bogs in Europe suggests that bogs in this region are sensi- marks the limit of the mire at the southeastern side (Figure 2). tive to environmental changes, both natural and human-induced. Plant community nomenclature follows Delarze et al. (1998). In this region, bogs suffer from summer droughts and are there- fore likely to respond to global warming and changes in the pat- tern of summer precipitation. Furthermore, calcareous terrains with minerogenous waters and with well-developed karst rep- resent unusual conditions for peatlands. In an attempt to shed some light on these questions and to com- pare past communities unaffected by human activities against modern bog communities, a detailed palaeoecological investi- gation was conducted on a representative bog remnant. A record spanning the last 11 000 years was found to exist in the Praz- Rodet bog, one of the few remaining little-disturbed bogs in the Jura Mountains. Stratigraphic relationships, radiocarbon dating and analyses of pollen, plant macrofossils and testate amoebae were undertaken on a sediment core. While pollen represents both a regional and a local palaeoecological signal, this study focuses on local indicators from the pollen, plant macrofossils and testate amoebae to reconstruct a detailed history of local bog communi- ties. This study is the first well-dated postglacial multiproxy Figure 2 Site map of the Praz-Rodet bog (redrawn from Gallandat, 1982). palaeoenvironmental record from the Jura Mountains and contrib- Vegetation types are: 1 = open raised bog community (Sphagnion utes to an understanding of the age and development of these rare magellanici); 2 = bog pool community (Caricion lasiocarpae); 3 = raised ecosystems in this part of Europe. bog community with pine (Piceo-Vaccinion uliginosi); 4 = Betulion pubescentis raised bog community; 5 = small sedge fen (Caricion davalli- anae and Caricion fuscae); 6 = tall nitrophilous grass and willow com- munities (Filipendulion); 7 = pasture (Cynosurion); 8 = wet meadow sub- Study site and regional setting ject to natural nutrient input (Calthion). The circled star indicates the coring location. The Orbe river is outlined in bold. The dashed bold line Praz-Rodet is a small bog about 5 ha in size situated at the south- shows the hypothetical past position of the river based on the vegetation western end of the Joux Valley near the western border of Switz- pattern (Gallandat, 1982) of which the present Caricion lasiocarpae com- erland (46°349000 N lat, 6°109250 E long; 1035 m a.s.l.; Figure munity (2) is believed to be a remnant. The small stream running at the southwestern side of the mire (shown in bold) is believed to represent the 1). The bedrock is pure calcareous sedimentary rock with well- remnant of another hypothetical past meander of the river. The hypotheti- developed karst. Praz-Rodet occupies a basin behind a terminal cal ancient river bank is situated at the limit between vegetation types 3 moraine from the last glaciation (Aubert, 1943; Bruckert and and 5 (Gobat, 1984). The horizontal line at the southwestern end of the Gaiffe, 1980). Soligenous calcareous water flows along the edges mire indicates the location of the transect study of Gobat (1984). Small of the bog basin. Dolines (karst holes) drain the periphery of the arrows indicate the flow of water flowing through the wet meadow and mire. The surface at the centre is raised and is influenced only into the dolines. Open arrows = calcareous water; closed arrows = bog by ombrogenous water. Abundant precipitation (annual average water.  E.A.D. Mitchell et al.: Palaeoecological history of Praz-Rodet bog, Swiss Jura 67 Human activities in the Joux Valley date back to prehistoric counted per sample. Other invertebrate fossils encountered with times (Guignard, 1972). The valley was important as a trade route the testate amoebae were recorded. and for glass manufacturing during Roman times (Rod, 1953; All diagrams were generated using TILIA and TILIAGRAPH Guignard, 1984; 1985; 1987). The first permanent settlements, (Grimm, 1991). traced to the tenth and eleventh centuries, were at the northeast end of the valley about 15 km from the study site. In the Jura Numerical analyses Mountains, bogs were a minor source of fuel peat since the six- For numerical analyses, logarithmic transformations were applied teenth century (Chastain, 1952) but this changed in the eighteenth to pollen and testate amoebae percentage data in order to reduce century when wood became scarce and was substituted by peat as the influence of dominant taxa. For testate-amoebae data, only the primary source of fuel. Up to this time, bogs were widespread samples with 50 or more shells were included in the analyses. throughout the Jura Mountains. Subsequently, many bogs were Presence-absence data were used for the plant macrofossil data. drained, converted to agriculture land, and used for pasture by For zonation, pollen (P), plant macrofossil (M) and testate amo- livestock. The bog studied here, however, has remained ebae (T) data were submitted to a depth-constrained clustering untouched. using the program CHRONO from the R package release 3.0 (Legendre et al., 1985; Legendre and Vaudor, 1991). This method is non-hierarchical and the algorithm used is agglomerative with a proportional link. Each fusion between groups was tested for Material and methods significance with a Monte-Carlo permutation. Poller zones were determined separately for regional taxa (regional zones, Pr) and Field coring peatland taxa (local zones, Pl). Canonical correspondence analysis A core was taken in 1993 near the centre and highest part of the (CCA) was performed on the fossil testate amoebae data and on bog (Figure 2) with a large diameter (8 cm) piston corer (Wright a set of samples of modern testate amoebae from an independent et al., 1984). The upper 1 m was collected in 1994 using a War- study (Mitchell et al., 1999) using the program CANOCO (ter denaar sampler (10 3 10 cm; Wardenaar, 1987). The 1993 core Braak, 1988–92). Only taxa that occurred in both data sets were was split longitudinally. One half was used for detailed strati- included. The ordination space was defined by modern samples graphic analyses, pollen subsampling and radiocarbon dating, and and associated environmental variables, with the fossil samples the other half was used for testate amoebae and plant macrofossil added as passive samples. Samples are grouped according to the subsampling. The surface 1 m core was used for 210Pb and geo- depth-constrained clustering results on the CCA scatter diagram. chemical analyses. Two additional cores, one near the north edge Past pH and water-table depths were inferred using a two-way and another near the south edge of the bog, were taken for strati- weighted averaging procedure in which the modern data are used graphic analyses and comparison with the central core. They sug- to estimate the optima of the species. These optima were then gest that the topography of the underlying impermeable layer is used to infer modern values (sensu Birks et al., 1990). This double relatively flat, which was confirmed by a radio-magneto-telluric averaging resulted in a shrinkage of the inferred values that was study (Mitchell, 1995). corrected by an inverse linear deshrinking procedure where the observed values are regressed on the initial estimated value. The Dating and palaeoecological analyses ecological optima of the species and the performance of the trans- A total of 18 levels were radiocarbon dated. Radiocarbon ages fer functions (using slightly different methods) are given in Mitch- were calibrated to calendar years (cal. yr BP, present taken as ell et al. (1999). The performance of the model we used is as 1950) using CALIB version 3.0.3c (Stuiver and Reimer, 1993). follows: water-table depth: RMSEP-WA(tol) –20.28 cm, r2 = The 210Pb ages have been published elsewhere (Appleby et al., 0.58, mean difference between predicted and observed values 1997; Shotyk et al., 1997; van der Knaap et al., 2000). −14.8 cm; pH: RMSEP-WA = 0.28, r2 = 0.64, mean difference Samples for pollen analysis were prepared using standard pro- between predicted and observed values = 0.25. cedures (Faegri et al., 1989). Exotic Lycopodium clavatum spores were added to samples from 0 to 1 m for the determination of pollen accumulation rates (Stockmarr, 1971). Pollen and spores Results and interpretation were identified using Punt (1976), Punt and Clarke (1980; 1981; 1984), Punt and Blackmore (1991), Punt et al. (1988; 1995), and Stratigraphy and age Reille (1992). All pollen percentages were calculated on the basis Details on the sediment stratigraphy are given in Table 1. Of the of a pollen sum that includes tree, shrub and herb pollen but 18 radiocarbon dates obtained (Table 2), the three basal dates excludes aquatic and peatland pollen. Subsamples of 20 cm3 were measured by displacement in a graduated cylinder and washed with warm water over a 300 mm Table 1 Lithostratigraphy of the Praz-Rodet core mesh sieve. The residue was sorted under a dissecting microscope and all identifiable plant remains were picked and stored in stop- Depth (cm) Lithostratigraphy pered vials. Six categories of remains were recorded semi-quanti- tatively according to Aaby and Berglund (1986). Plant remains 0–5 Living mosses and litter were identified to the lowest possible taxonomic level with the 5–47 Well-preserved Sphagnum-dominated peat aid of Le´vesque et al. (1988) and Grosse-Brauckmann (1972; 47–58 Strongly decomposed peat 1974) for vascular plants, and Janssens (1983) and Daniels and 58–179 Mixed Sphagnum and herbaceous peat 179–195 Transition Eddy (1990) for mosses. Pinus needles were identified to species 195–260 Less-decomposed Sphagnum-dominated peat with the aid of Sandoz (1987). Charcoal and insect remains were 260–320 Decomposed Sphagnum-dominated peat noted but not identified. Plant taxonomy follows Corley et al. 320–350 Transition (1981) for mosses and Tutin et al. (1964–80) for vascular plants. 350–405 Compact coarse-detritus gyttja Subsamples of 2 cm3 were analysed for testate amoebae follow- 405–650 Lake deposits: silt and clay ing Tolonen (1986) and Warner (1990). A minimum of 200 shells Below 650 Glacial deposits: mixed coarse gravel and finer was achieved for most samples except for some samples at the material (not studied) bottom of the core. Percentages are based on the total number 68 The Holocene 11 (2001) Table 2 Radiocarbon dates of the Praz-Rodet core Conventional Calibrated Depth Thickness Laboratory radiocarbon age BP (1s) (cm) (cm) number (1) Method age BP (2) d13C ‰ Material dated 53 0.5 UtC-4939 AMS 257 6 46 BP –28.9 Sphagnum magellanicum 213 6 213 97 0.5 UtC-4938 AMS 1144 6 34 BP –28.3 Andromeda polifolia twig 1 leaves 1026 6 40 130 4 B-6493 decay 1620 6 40 BP –24.2 Sphagnum magellanicum 1479 6 60 162 4 B-6494 decay 2130 6 30 BP –26.4 Sphagnum magellanicum 2095 6 41 180 1 UtC-4936 AMS 2589 6 35 BP –25.7 S. magellanicum 1 S. Sect. cuspidata, amorphous 2736 6 19 188 4 B-6495 decay 4220 6 40 BP –27.9 S. magellanicum 1 S. Sect cuspidata, 1 S. Sect. 4743 6 95 acutifolia 216 4 B-6496 decay 4670 6 80 BP –27.8 Sphagnum capillifolium 5435 6 133 260 4 B-6497 decay 5210 6 60 BP –27.1 S. magellanicum 1 S. capillifolium 6036 6 128 288 4 B-6498 decay 5800 6 80 BP –26.7 Eriophorum vaginatum 1 S. magellanicum 6609 6 104 318 1 UtC-4052 AMS 6270 6 60 BP –27.6 4.5 Potentilla palustris seeds 1 1 other seed 7128 6 93 350 1 UtC-4053 AMS 7330 6 90 BP –26.7 6 Carex seeds 1 4.5 Potentilla palustris seeds 8075 6 93 370 1 UtC-4937 AMS 7630 6 47 BP –27.8 amorphous peat 8385 6 30 380 1 UtC-4054 AMS 7400 6 70 BP –26.9 Cyperaceae leaves 8165 6 142 392 2 UtC-4055 AMS 9560 6 80 BP –26.3 seeds 10684 6 200 403 1 UtC-4056 AMS 9780 6 60 BP –26.3 herbs 10956 6 48 425 1 UtC-4057 AMS 4780 6 70 BP –26.8 one piece of charcoal 5465 6 129 453 1 UtC-4058 AMS 5560 6 50 BP –26.9 one piece of charcoal 6354 6 50 473.5 2 UtC-4059 AMS 7780 6 60 BP –24.1 leaf fragments 8492 6 61 (1) Laboratory designations are: UtC = R.J. Van de Graaff laboratory, Utrecht, The Netherlands; B = Radiocarbon-Laboratory, Bern. (2) Dates are corrected for isotopic fractionation. (UtC-4057, UTC-4058, UTC-4059) are too young based on the well-established pollen chronology for this region (Welten, 1982; Ammann and Lotter, 1989). These dates are on charcoal and charred vascular plant leaf remains. Dates on charcoal are often problematic because of possible absorption of young carbon. The radiocarbon dates indicate a hiatus in the sediment at 184 cm. The chrono-stratigraphy of the Lateglacial part of Praz-Rodet bog is based on biostratigraphic correlation of pollen zones from sites at Les Cruilles (1035 m a.s.l.), 15 km NE in the same valley, and Le Marais des Amburrex (1300 m a.s.l.), 5 km ESE in an adjacent valley (Wegmu¨ller, 1966). In the case of Le Marais des Amburnex, however, we correlate Wegmu¨ller’s Firbas zone IIa (first part of Allerød) with Firbas zone Ibc (Bølling), assuming synchrony of this biozone with the similar biozone Firbas zone Ibc in Les Cruilles. Firbas zone III, the Younger Dryas, is absent from Praz-Rodet bog, which suggests the occurrence of another hiatus at 406 cm. An depth-age curve was constructed for four separate sections (Figure 3, left), as follows. Figure 3 Depth-age relationship (left) and sediment-accumulation rates (1) 0–184 cm: ages for the surface 50 cm are based on the 210Pb (right) for the Praz-Rodet bog based on 18 calibrated radiocarbon dates age estimates. Ages and sediment-accumulation rates of the and two biostratigraphic dates. Ages are expressed as cal. yr BP = calendar years before ad 1950. Calibrated radiocarbon dates are depicted as large corresponding levels were calculated by linear interpolation. diamonds with standard deviations, biostratigraphic dates as open circles, The chronology of the section 50–184 cm was modelled with and pollen samples as small diamonds with confidence intervals. PSIMPOLL (Bennett, 1993), using a four-term polynomial function, and using all calibrated radiocarbon dates in this section and the 210Pb age estimate at 50 cm. 14 000 cal. yr BP at 430 cm). The ages of samples older than (2) 184–370 cm: PSIMPOLL was used with a three-term poly- Allerød time are extrapolated. nomial function on the calibrated radiocarbon dates for this core segment. The Firbas zones for the Holocene given in Table 3 are based on (3) 370–406 cm: PSIMPOLL was used to interpolate linearly biostratigraphic correlation of pollen zones with those in Les Cru- between the calibrated radiocarbon dates at 370 and 403 cm. illes and Le Marais des Amburnex. As a result, the ages of some The radiocarbon date at 380 cm appears too young. The of our biostratigraphic Firbas zones differ slightly from chronos- apparent Younger Dryas hiatus extends for about a half mil- tratigraphic Firbas zones as defined by Mangerud et al. (1974). lennium into the Preboreal period. Sediment-accumulation rates are low in the early Holocene, and (4) 406–472 cm: ages were linearly interpolated between the increased markedly at 8500 cal. yr BP. This is similar to average approximate ages used for the beginning and end of the rates in 25 lakes and mires studied for pollen in Switzerland, Allerød as established in the GISP2 Greenland ice core showing a marked increase c. 9000 cal. yr BP (van der Knaap and (Stuiver et al., 1995) (13 000 cal. yr BP at 406 cm; and Ammann, 1997).  E.A.D. Mitchell et al.: Palaeoecological history of Praz-Rodet bog, Swiss Jura 69 Table 3 Summary and interpretation of regional pollen diagram of the Praz-Rodet bog Pollen Plant macrofossils The vegetation is reconstructed in two ways. Pollen and spores A total of 11 zones were recognized based on plant macrofos- representing the regional vegetation are presented in Figure 4 and sils (Figure 6). Needles of Pinus rotundata were found at Table 3 and from local sources in Figure 5. Pollen of Cyperaceae 128 cm (1500 cal. yr BP) and 93 cm depth (720 cal. yr BP). can be from both regional and local sources. Also, some pollen Zones M-1 and M-2 contain fen indicators. Above 320 cm, in types can represent taxa that once grew outside the peatland basin zone M-3, Sphagnum magellanicum (present in most samples but may have migrated onto the peatland as the surface became above this level), Scheuchzeria palustris and Eriophorum dry, e.g., Pinus. These are presented in both figures. Pollen zones vaginatum suggest oligotrophic conditions. Indicators of identified in this study refer to this site only. ombrotrophic conditions (S. capillifolium and Vaccinium 70 The Holocene 11 (2001) Figure 4 Pollen percentage diagram of the Praz-Rodet bog showing regional pollen types. Zone boundaries are as follows: dashed line = hiatus; drawn line = other zone. *** = statistically significant zone boundary (P , 0.05). oxycoccos) appear in zone M-4. In zone M-6 bog pool species ophorum indicate wet conditions but these species disappear are recorded. Zone M-8 contains indicators of ombrotrophic in zone M-10 which suggests drier conditions. In zone M-11 bog (V. oxycoccos) and drier conditions (Andromeda polifolia). S. cf. rubellum and V. oxycoccos indicate moderately dry In zone M-9 S. cuspidatum, S. tenellum, S. palustris and Eri- ombrotrophic conditions. The final zone M-12 may represent  E.A.D. Mitchell et al.: Palaeoecological history of Praz-Rodet bog, Swiss Jura 71 Figure 4 Continued. 72 The Holocene 11 (2001) Figure 5 Pollen percentage diagram of the Praz-Rodet bog (Swiss Jura; 1040 m) showing local pollen types. Zone boundaries are as follows: dashed line = hiatus; drawn line = statistically significant zone boundary (P , 0.05). Analysis: Jacqueline F.N. van Leeuwen.  E.A.D. Mitchell et al.: Palaeoecological history of Praz-Rodet bog, Swiss Jura 73 Figure 6 Plant macrofossil diagram of the Praz-Rodet bog (Swiss Jura; 1040 m). Zone boundaries are as follows: dashed line = hiatus; drawn line = statistically significant zone boundary (P , 0.05). Analysis: Edward A.D. Mitchell. 74 The Holocene 11 (2001) a recent shift in the vegetation towards much drier conditions 2 in Figure 2) might be a remnant of an ancient meander of the as suggested by Calluna vulgaris. river (Gallandat, 1982). Similarly, soil chemical characteristics such as pH and calcium content show a clear vertical limit at the Testate amoebae southwestern margin of the bog (Figure 2) suggesting that the peat Results of the testate amoebae analysis are given in Figure 7 and was eroded by an ancient meander of the Orbe river (Gobat, Table 4. Amphitrema flavum and Hyalosphenia papilio are charac- 1984). The renewed peat accumulation at 2800 cal. yr BP may be teristic of oligotrophic wet habitats such as bog hollows. A. wrigh- related to another more recent shift in the course of the Orbe river. tianum is characteristic of wetter situations than A. flavum and H. This event is possibly connected to the rapid climatic change to papilio. Assulina seminulum is characteristic of ombrotrophic cooler and moister conditions discussed by van Geel and Renssen moist to dry habitats. Cryptodifflugia oviformis is characteristic of (1998) and van Geel et al. (1996; 1998a; 1998b), with which the relatively dry habitats (small hummocks). Trigonopyxis arcula is high water table inferred from testate amoebae would also agree. characteristic of dry habitats such as drained sites. Heleopera rosea, Nebela tincta, N. militaris and A. muscorum are present in Development of the Praz-Rodet bog all habitats within raised bogs and transitional mires (Tolonen, Sedimentation began in the centre of the Praz-Rodet basin at the 1986; Mitchell et al., 1999). end of the Older Dryas time. Initially the site was a lake with In the canonical correspondence analysis (CCA), four environ- little aquatic vegetation. By Allerød time, well-developed shore mental variables were found to be significant (Monte-Carlo test, vegetation with such species as Filipendula ulmaria surrounded 999 permutations) by the forward selection procedure. These the lake. Conditions during the Younger Dryas and early Pre- were, in order of importance, pH (P = 0.001), water-table depth boreal are not known because of a lack of sedimentary record. (P = 0.001), water content of fresh samples expressed as percent- When sediment accumulation resumed in the middle Preboreal, age of free volume (P = 0.002) and water content of the samples shallow open water existed as indicated by a number of aquatic after one hour drainage expressed as percentage of free volume plants typical of shallow lakes or wetlands with open water. This (P = 0.017). Together, these four variables explain 30.1% of the developed into a fen with less extensive open water between 9300 total variation in the modern testate amoebae data set. The first and 7250 cal. yr BP. During the latter part of this phase (7600– three ordination axes are significant (P # 0.05). Axes 1 and 2 7300 cal. yr BP) pollen suggests the presence of taxa typical of a are represented in Figure 8. Testate amoebae and other taxa are wet grassland. High proportions of Allium-type, Valeriana represented in Figure 8a. Figure 8b represents fossil types and officinalis-type and Caltha-type pollen suggest that these plants modern types from surface samples. Most zones are well separ- grew close to the coring site. This wet grassland community ated in the ordination diagram. The positions of fossil samples resembles riparian communities along the Orbe river (Gallandat, relative to modern samples used to define the ordination space 1982), which may have been much closer to the coring site than and relative to the environmental variables illustrate the ecological today. Plant macrofossils confirm the existence of a fen during history of the coring site. The overall main trends are towards a this period. wetter bog with minor changes (zones T-2 to T-6) followed by By about 7300 cal. yr BP, a Sphagnum mire began to form at dramatic change towards dry conditions (zones T-7 and T-8). the coring site replacing the earlier fen. The period 7300–6200 cal. yr BP is a transitional phase where the fen was gradually replaced by ombrogenous raised bog. This is especially clear in the testate Discussion amoebae and in the stratigraphic change from compact detritus peat to Sphagnum-dominated peat. The first evidence of human Chronostratigraphy and hiatuses in the stratigraphic activity occurred at around 6850 cal. yr BP as shown by the occur- record rence of Plantago lanceolata-type, which indicates grazing. This Our chronology is in general in agreement with Firbas zones date agrees with the time of the early Neolithic period in the Jura shown in Table 3. The close agreement is good if we consider Mountains (SPM II, 1995). that the Firbas zones were established by Wegmu¨ller (1966) as From about 6200 cal. yr BP to the second hiatus starting at biozones rather than chronozones in a time when few radiocarbon 4700 cal. yr BP, pollen, testate amoebae and the peat stratigraphy dated pollen diagrams with a similar biostratigraphy were avail- indicate a raised bog. Plant macrofossils show the presence of able. raised-bog vegetation by 5530 cal. yr BP. Testate amoebae sug- The present record at Praz-Rodet is the best dated bog sequence gest a relatively high water table until 5200 cal. yr BP. This was in the Jura. The age-depth model indicates a complex history of followed by a phase with lower water table, which may have been sediment and peat accumulation. One hiatus in the stratigraphic the cause for the disappearance suggested by pollen of the typical record (Younger Dryas and early Preboreal; 406 cm) is obvious raised-bog species Andromeda polifolia at 5550 cal. yr BP. This in the pollen stratigraphy and coincides with a change in lithology. drier period coincides with modest human activity around the The lack of a record may be due to very little sediment accumu- basin in the form of forest grazing, presence of some pastures and lation in a highly unproductive lake resulting from the generally for the first time some arable fields. Early signs of forest clearance low temperatures in this period. A similar situation has been and soil erosion were also recorded by Shotyk et al. (1998), in observed by Wegmu¨ller (1966) at Les Cruilles 15 km to the nor- another peatland of the Jura Mountains, as increased deposition theast, in which the Younger Dryas is represented by a thin layer of atmospheric lead and scandium suggesting enhanced rates of of clay and lake marl. soil erosion. The other hiatus, dated 4700–2800 cal. yr BP (184 cm), is After 2800 cal. yr BP, Scheuchzeria and Drosera pollen and detected on radiocarbon dating only. This observation suggests testate amoebae such as Amphitrema flavum, A. wrightianum and that hiatuses in peat sequences may go undetected if detailed radi- Difflugia globulosa suggest the presence of hollows on a raised ocarbon dating is not available. There is no reason to believe that bog. Plantago lanceolata-type and Gramineae indicate that human this hiatus is related to human activities. The most probable expla- impact outside the mire increased more strongly than before. nation is that the Orbe River eroded some peat when it was at or The three lines of evidence show a shift towards dryness on very close to the coring site at some time in the past. Two inde- the bog surface at different temporal and spatial scales. The first pendent lines of evidence indeed suggest that the position of the is represented by the appearance of Pinus rotundata macrofossils Orbe river shifted in the past. The present vegetation pattern sug- that confirm the presence of this tree on the bog since 1500 cal. gests that the large Caricion lasiocarpae pool community (code yr BP, which coincides with an inferred drying trend based on  E.A.D. Mitchell et al.: Palaeoecological history of Praz-Rodet bog, Swiss Jura 75 Figure 7 Testate amoebae percentage diagram of the Praz-Rodet bog (Swiss Jura; 1040 m) with indication of inferred peat pH and water-table depths. Zone boundaries are as follows: dashed line = hiatus; drawn line = statistically significant zone boundary (P , 0.05). Analysis: Edward A.D. Mitchell. 76 The Holocene 11 (2001) Table 4 Summary and interpretation of testate amoebae diagram from the Praz-Rodet bog other fossil indicators. The second shift occurred after 860 cal. yr explained by the heterogeneity of the bog surface and differences BP, when Ericaceae (Calluna, Vaccinum) pollen suggests a dry in spatial scale. Testate amoebae and plant macrofossils are raised bog on which Pinus started to expand. Testate amoebae extremely local in origin whereas pollen is derived from more and macrofossils (Sphagnum cuspidatum, S. tenellum), however, distant sources. The combination of results suggests that wet hol- indicate that the coring location remained wet up to c. 160 cal. yr lows existed within an otherwise dry raised bog, a situation fre- BP and 240 cal. yr BP, respectively. The coring location probably quently encountered in the Jura bogs today. The third shift to remained a hollow or pool. However, differences in environmental drier conditions occurred after c. 50–65 cal. yr BP. Two lines of inferences based on testate amoebae and on pollen may be evidence, however, differ in detail. Testate amoebae suggest that  E.A.D. Mitchell et al.: Palaeoecological history of Praz-Rodet bog, Swiss Jura 77 limits for testate amoebae (zone T-7, 160–65 cal. yr BP) differ from those of pollen (zone P1–11, 100–50 cal. yr BP). This might be attributed to a timelag in the pollen signal of 15 to 60 years (3 to 5 cm in the peat) compared to testate amoebae. As soil organisms, testate amoebae are sensitive to changes in soil moist- ure and chemical conditions (Gobat et al., 1998; Mitchell et al., 2000a; 2000b) and may respond faster to changes than plants (Buttler et al., 1996; Gilbert et al., 1998). Canonical correspondence analysis of modern and fossil testate amoebae data The ordination technique of canonical correspondence analysis (CCA) is used to complement the palaeoecological interpretation on two points: (1) to quantify the proportion of the variance in species data explained by each environmental variable and to test the significance of these variables; and (2) to visualize in two dimensions the development of the site in relation to well-docu- mented modern sites and to environmental variables measured in these sites (Figure 8b). The three significant environmental variables explained 30% of the variance. The results show that most sample groups (zones) are well separated from both the preceding and following group. Interestingly, although a clear trend towards drier conditions appears, even in the period of driest conditions (zones T-8 to T- 10) the coring site hardly reached the driest conditions encoun- tered in natural open peat bogs today such as Sphagnum fuscum hummocks. Also, after the hypothesized erosion phase of the bog by the Orbe river (zone T-5), conditions remained closer to oligo- trophic wet conditions comparable with present-day Sphagnum cuspidatum pools than to present-day Sphagnum-dominated tran- sitional minerotrophic mires. CCA can also be used as a tool for predicting past values of environmental variables based on modern samples. However, for precise inference, a different analysis should be done for each variable separately. History of human impact and origin of Pinus Figure 8 Canonical correspondence analysis (CCA) scatter diagram of rotundata testate amoebae samples from the Praz-Rodet core. The ordination space Pollen data (Table 3) suggest that humans were present in the is a combination of ecological variables derived from a previous study on valley long before the first known permanent settlements of the modern samples (Mitchell et al., 1999). (a) Position of testate amoebae species. (b) Position of samples. Ten zones resulting from the depth-con- tenth or eleventh centuries and before the Middle Ages when strained clustering are also represented. The fossil samples are projected humans used the valley for glass production. Pollen suggest that passively in this ordination showing their successive position. Numbers grazing in the forests and some of the first pastures could have indicate the depth (in cm) of the samples. Axes 1 and 2 represent together existed as early as 6850 cal. yr BP. Humans appear to have had 24.6% of the total variance. Symbols represent ecological groups of mod- little or no direct impact on the bog until the middle of the eight- ern samples: filled pentagons = bog centre, wet sites (e.g., Sphagnum cus- eenth century when the forest was cleared to create extensive pas- pidatum pools); open triangles = bog centre, hummocks (e.g., Sphagnum turelands. After this time, the bog surface became drier and Pinus fuscum); open circles = tree-covered bog (Pinus rotundata); filled triangles rotundata expanded on the bog. There is no evidence from exist- = minerotrophic sites either on the side of peat extraction ditches, in an ing studies of a change in summer climate at the times when the open forest, or open meadow; filled circles = minerotrophic and moist bog became drier. Increasing temperatures at the end of the ‘Little sites (transitional Sphagnum mire). As. = Assulina; As.m. = A. muscorum; B. = Bullinularia; Ce. = Centropyxis; Co. = Corythion; Cr. = Cryptodif- Ice Age’ may have contributed to the drying-out of the bog once flugia; Cy. = Cyclopyxis; Cy.1. = C. laevigata; D. = Difflugia; E. = Eugly- it was exposed to desiccating winds as a result of deforestation. pha; E.a. = E. alveolata; E.c. = E. ciliata; He. = Heleopera; He.p. = H. The presence of Pinus rotundata on Jura bogs has long been platystoma; Hy. = Hyalosphenia; N. = Nebela; Ps. = Pseudodifflugia; Test believed to be a natural relict from Lateglacial times. Furthermore, = unidentified testate amoebae genus. ecological studies have suggested that the Piceo-Vaccinion uligi- nosi community is the final successional stage in bog development (Richard, 1961; Matthey, 1964; Royer et al., 1978, Julve, 1983; the bog became drier between 160 and 65 cal. yr BP which was Sandoz, 1987). In contrast, a more recent hypothesis suggested followed by a short period of fluctuating dry conditions up to the that Pinus rotundata was introduced about two or three centuries present. Pollen suggest a dry bog surface with Calluna between ago in the Jura Mountains (Feldmeyer-Christe, 1990; Reille, 100 and 50 cal. yr BP, followed by a period with a slightly less 1991). However, ecological studies have shown that Pinus dry surface on which Pinus rotundata expanded. rotundata on the Jura bogs fills a specific ecological niche at the periphery of ombrogenous areas and in fens (Frele´choux, 1997). Apparent timelag between pollen and testate The growth of pine in the central parts of the peatlands seems to amoebae data have been enhanced by local forest clearance as confirmed by There appears to be a slight discrepancy in the timing of change dendroecological studies (Frele´choux et al., 2000). Indeed, Pinus indicated by pollen and by testate amoebae. The upper and lower can survive long periods under marginal conditions without 78 The Holocene 11 (2001) producing significant amounts of pollen (Edelman, 1985). Reille eering Research Council of Canada, to BGW, supported this pro- (1991) concluded that P. rotundata expanded on bogs in the Jura ject. Radiocarbon dating was financially supported by Swiss Mountains and elsewhere one or a few centuries ago as the result National Science Foundation project No. 31.37620–93. of recent introduction, even though he discusses macrofossil evi- dence showing much earlier presence of Pinus sp. on the same bogs probably as small patches or as isolated individuals. Accord- References ing to Be´geot and Richard (1996), P. rotundata was not con- sidered to be a valuable tree by foresters and therefore it was Aaby, B. and Berglund B.E. 1986: Characterisation of peat and lake never planted on the peat bog of Frasnes (French Jura). However, deposits. In Berglund, B.E., editor, Handbook of Holocene palaeoecology P. rotundata was planted in the Hautes Vosges (Alsace, France) and palaeohydrology, Chichester: John Wiley, 231–46. by German forest managers prior to 1914 (Be´geot and Richard, Ammann, B. and Lotter, A.F. 1989: Late-Glacial radiocarbon-and paly- nostratigraphy on the Swiss Plateau. Boreas 18, 109–26. 1996). The same authors further conclude that the recent increase Appleby, P.G., Shotyk, W. and Fankhauser, A. 1997: 210Pb age dating in P. rotundata on peat bogs is a consequence of draining and of three peat cores in the Jura Mountains, Switzerland. Water Air and Soil peat harvesting. Pollution 100, 223–31. Our palaeoecological results do not support any of these Aubert, D. 1943: Monographie ge´ologique de la valle´e de Joux (Jura hypotheses. Our findings of P. rotundata macrofossils in peat vaudois). Mate´riaux pour la Carte Ge´ologique de la Suisse 78, 134 pp. about 1500 cal. yr old show this species neither to be glacial relict Be´geot, C. and Richard, H. 1996: L’origine des peuplements de Pin a` nor an anthropogenic introduction on the bog. We therefore con- crochets (Pinus uncinata Miller ex Mirbel) sur la tourbie`re de Frasne et sider the presence of P. rotundata on the Praz-Rodet bog as natu- exploitation de la tourbe dans le Jura. Acta botanica Gallica 143, 47–53. ral, though not old enough to be a glacial relict. We hypothesise Bennett, K.D. 1993: PSIMPOLL 2.27: program for plotting pollen dia- that a dry bog surface and the present abundance of Pinus on the grams and analysing data. Cambridge: Department of Plant Sciences, Uni- versity of Cambridge. bog are due to the clearance of the forest on the mineral soil Birks, H.J.B., Line, J.M., Juggins, S., Stevenson, A.C. and ter Braak, around the bog. The forest had functioned as a windbreak reduc- C.J.F. 1990: Diatoms and pH reconstruction. Philosophical Transactions ing evapotranspiration on the bog, but there was (and is) no direct Royal Society London B327, 263–78. hydrological link in the soil between these two environments. Bruckert, S. and Gaiffe, M. 1980: Analyse des facteurs de formation et Such a protection must have been considerable, because the Praz- de distribution ds sols en pays calcaire glaciaire ou karstique. Plaine de Rodet bog is quite small. When the protection disappeared, the Frasne, Bonnevaux et Montagne du Laveron, Jura central. Annales scienti- mesoclimate was altered and the bog adapted to new conditions. fiques de l’Universite´ de Besanc¸on. Biologie ve´ge´tale 4, 19–67. In northwestern Europe and North America, forest clearance Buttler, A., Warner, B.G., Matthey, Y. and Grosvernier, Ph. 1996: has often induced paludification through raising groundwater Testate amoebae (Protozoa: Rhizopoda) and restoration of cut-over bogs tables (Warner et al., 1989). The case of Praz Rodet is opposite in the Jura, Switzerland. New Phytologist 134, 371–82. Chastain, A. 1952: Recherches e´cologiques et floristiques sur le ‘Pinetum’ from this situation because: (1) the hydrology of the mire is inde- de la haute tourbie`re de la Vraconnaz (Jura helve´tique vaudois). Recueil pendent from its surroundings as far as groundwater is concerned des travaux de l’institut de botanique de l’Universite´ de Montpellier because of the karstic character of the landscape; and (2) summer (Supple´ment 2), 176 pp. winds are much hotter and drier and desiccate more strongly than Clymo, R.S. and MacKay, D.M. 1987: Upwash and downwash of pollen in northwestern Europe due to (a) a more continental climate and and spores in the unsaturated layer of Sphagnum-dominated peat. New (b) a more southerly position. Indeed, the development of the Phytologist 105, 175–83. Praz-Rodet bog is unusual, but similar cases may exist elsewhere Corley, M.F.V., Crundwell, A.C., Du¨ll, R., Hill, O. and Smith, A.J.E. in analogous mesoclimatical and geomorphological conditions. 1981: Mosses of Europe and the Azores: an annotated list of species, with synonyms from the recent literature. Journal of Bryology 11, 609–89. Daniels, R.E. and Eddy, A. 1990: Handbook of European Sphagna. Hun- tingdon: Institute of Terrestrial Ecology (National environmental council), Acknowledgements 262 pp. Delarze, R., Gonseth, Y. and Galland, P. 1998: Guide des milieux natur- We thank B. Ammann, H.J.B. Birks, D. Charman and an unidenti- els de Suisse. E´cologie – menaces – espe`ces caracte´ristiques. Lausanne, fied reviewer for comments, P. Schoeneich for valuable dis- Switzerland: Delachaux et Niestle´, 415 pp. cussions on the Lateglacial pollen stratigraphy, and W. Shotyk, Edelman, H.J. 1985: Late glacial and Holocene vegetation development A. Badstuber, K. Ruch and Ch. Stauffer for technical assistance. of la Goutte Loiselot (Vosges, France). Utrech: Drukkerij Elinkwijk Bv, The preparation of the pollen samples for the top metre was 197 pp. done by Jan van Tongeren in Utrecht, the Netherlands (Laboratory Faegri, K., Kaland, P. and Krzyinski, J. 1989: Textbook of pollen analy- of Palaeobotany and Palynology). The preparation, processing and sis (fourth edition). Chichester: John Wiley. radiocarbon-age dating of the samples coded with ‘B−’ were car- Feldmeyer-Christe, E. 1990: Etude phyto-e´cologique des tourbie`res des Franches Montagnes (cantons du Jura et de Berne, Suisse). Mate´riaux pour ried out by the Radiocarbon Laboratory of the Physics Institute le leve´ ge´obotanique de la Suisse 66, 163 pp. of the University of Bern, Switzerland. Radiocarbon samples Frele´choux, F. 1997: Etude du boisement des tourbie`res hautes de la coded with ‘UtC-’ were prepared, processed and AMS dated in chaıˆne jurassienne: typologie et dynamique de la ve´ge´tation – approche the R.J. Van de Graaff laboratory, University of Utrecht, The dendroe´cologique et dendrodynamique des peuplements arborescents. Netherlands. Unpublished PhD thesis, University of Neuchaˆtel, 360 pp 1 appendices The responsibilities of the different authors of this paper are as 674 pp. follows: EADM carried out and interpreted the analysis of plant Frele´choux, F., Buttler, A., Scheingruber F.H. and Gobat, J.-M. 2000: macrofossils and testate amoebae, the latter supported by BGW; Stand structure, invasion and growth dynamics of bog pine (Pinus uncin- JFNvL analysed pollen; WOvdK interpreted the pollen diagram ata var. rotundata) in relation to peat cutting and drainage in the Jura mountains, Switzerland. Revue Canadienne de Recherche Forestie`re and established the chronology; AB supported the testate-amoebae (Canadian Journal of Forest Research), 30, 1114–26. analysis with special emphasis on numerical analysis; all authors Gallandat, J.-D. 1982: Prairies mare´cageuses du Haut-Jura (Molinietalia, contributed in the final integration of results. Scheuchzerio-Caricetea fuscae et Phragmitetea). Mate´riaux pour le leve´ Grants from the Swiss Federal Office for Education and Science ge´obotanique de la Suisse 58, 2 volumes. (Project No. 95.0415) and the Wu¨thrich Fund (University of Gilbert, D., Amblard, C., Bourdier, G. and Francez, A.-J. 1998: Short- Neuchaˆtel), to EADM, and the Swiss National Science Foundation term effect of nitrogen enrichment on the microbial communities of a (project No. 83NC-037026) and the Natural Sciences and Engin- peatland. Hydrobiologia 373/374, 111–19.  E.A.D. Mitchell et al.: Palaeoecological history of Praz-Rodet bog, Swiss Jura 79 Gobat, J.-M. 1984: Ecologie des contacts entre tourbie`res acides et et le Jura. Bulletin de la Socie´te´ botanique de France 138 (Lettres marais alcalins dans le Haut-Jura suisse. Unpublished PhD Thesis, Uni- botaniques), 123–48. versity of Neuchaˆtel, 255 pp. —— 1992: Pollen et spores d’Europe et de l’Afrique du Nord. Marseille: Gobat, J.-M., Aragno, M. and Matthey, W. 1998: Le sol vivant. Bases Laboratoire de Botanique historique et de Palynologie, 520 pp. de pe´dologie, biologie des sols. Lausanne: Presses polytechniques et univ- Richard, J.-L. 1961: Les foreˆts acidophiles du Jura. Mate´riaux pour le ersitaires romandes, 521 pp. leve´ ge´obotanique de la Suisse 38, 164 pp. Grimm, E. 1991: TILIA and TILIAGRAPH. Springfield: Illinois State Rod, E. 1953: AMH A 66/12 (unpublished). Monuments Historiques et Museum. Arche´ologiques du Canton de Vaud, Switzerland. Grosse-Brauckmann, G. 1972: U ¨ ber pflanzliche Makrofossilien mitteleu- Royer, J.-M., Vadam, J.C. and Gillet, F. 1978: Etude phytosociologique ropa¨ischer Torfe. I. Gewebereste krautiger Pflanzen und ihre Merkmale. des tourbie`res acides et alcalines du Haut-Doubs. Publication du Centre Telma 2, 19–55. Universitaire d’Etudes Re´gionales 2, 109–85. —— 1974: U ¨ ber pflanzliche Makrofossilien mitteleuropa¨ischer Torfe. II. Sandoz, H. 1987: Recherches taxonomiques, bioge´ographiques et phytoe´- Weitere Reste (Fru¨chte und Samen, Moose u.a.) und ihre Bestimmungs- cologiques sur les principaux conife`res subalpins des Alpes: me´le`ze d’Eur- mo¨glichkeiten. Telma 4, 51–117. ope, pin cembro, pin a` crochet et pin mugho. PhD thesis, Universite´ d’Aix- Guignard, J.-P. 1972–87: Unpublished reports. Dossier communal Monu- Marseille III, 650 pp. ments Historiques et Arche´ologiques du Canton de Vaud, Switzerland. Shotyk, W., Cheburkin, A.K., Appleby, P.G., Fankhauser, A. and Kra- Heal, O.W. 1964: Observations on the seasonal and spatial distribution mers, J.D. 1997: Lead in three peat bog profiles, Jura Mountains, Switzer- of testaceans (Protozoa: Rhizopoda) in Sphagnum. Journal of Animal land: enrichment factors, isotopic composition, and chronology of atmos- Ecology 33, 395–412. pheric deposition. Water Air and Soil Pollution 100, 297–310. Janssens, J.A. 1983: A quantitative method for stratigraphic analysis of Shotyk, W., Weiss, D., Appleby, P.G., Cheburkin, A.K., Frei, R., bryophytes in Holocene peat. Journal of Ecology 71, 189–96. Gloor, M., Kramers, J.D., Reese, S. and van der Knaap, W.O. 1998: Julve, P. 1983: Les groupements des prairies humides et des bas-marais: History of atmospheric lead deposition since 12 370 14C yr BP from a peat e´tude re´gionale et essai de synthe`se a` l’e´chelle de l’Europe occidentale. bog, Jura Mountains, Switzerland. Science 281, 1635–40. PhD Thesis, Univ. Paris-Sud, 225 pp. SPM II (Sto¨ckli, W.E., Hiffeler, U. and Gross-Klee, E. (editors) 1995: Legendre, P. and Vaudor, A. 1991: Le Progiciel R – analyses multidi- Die Schweiz vom Pala¨olithikum bis zum fru¨hen Mittelalter. II Neolithikum. mensionnelles, analyse spatiale. De´partement de sciences biologiques, Basel, 358 pp. Universite´ de Montre´al, 144 pp. Stockmarr, J. 1971: Tablets with spores in absolute pollen analysis. Pol- Legendre, P., Dallot, S. and Legendre, L. 1985: Succession of species len et Spores 13, 615–21. within a community: chronological clustering, with applications to marine Stuiver, M. and Reimer, P.J. 1993: Extended 14C data base and revised and freshwater zooplankton. American Naturalist 125, 257–88. CALIB 3.0 14C age calibration program. Radiocarbon 35, 215–30. Le´vesque, P.E.M., Dinel, H. and Larouche, A. 1988: Guide illustre´ des Stuiver, M., Grootes, P.M. and Braziunas, T.F. 1995: The GISP2 d18O macrofossiles ve´ge´taux des tourbie`res du Canada. Centre de recherche climate record of the past 16 500 years and the role of the sun, ocean, and sur les terres Ottawa (Ontario) Publication 1817, 65 pp. volcanoes. Quaternary Research 44, 341–54. Mangerud, J., Andersen, S.T., Berglund, B.E. and Donner, J.J. 1974: ter Braak, C.J.F. 1988–92: CANOCO – a FORTRAN program for Quaternary stratigraphy of Norden, a proposal for terminology and classi- Canonical community Ordination (version 2.1). Ithaca, New York: fication. Boreas 3, 109–28. Microcomputer Power, 96 pp. Matthey, W. 1964: Observations e´cologiques sur la tourbie`re du Cachot. Tolonen, K. 1986: Rhizopod analysis. In Berglund B.E., editor, Handbook Bulletin de la Socie´te´ Neuchaˆteloise des Sciences Naturelles 87, 103–33. of Holocene palaeoecology and palaeohydrology, Chichester: John Wiley, Meisterfeld, R. 1977: Die horizontale und vertikale Verteilung der Testa- 645–66. ceen (Rhizopoda: Testacea) in Sphagnum. Archiv fu¨r Hydrobiologie 79, Tutin, T.G., Heywood, V.H., Burges, N.A., Valentine, D.H., Walters, 319–56. S.M. and Webb, D.A. 1964–80: Flora europaea (five volumes). Cam- Mitchell, E.A.D. 1995: The postglacial developmental history of the Praz- bridge: Cambridge University Press. Rodet bog, Valle´e de Joux, Swiss Jura. Masters thesis, Laboratoire d’Ecol- van der Knaap, W.O. and Ammann, B. 1997: Depth-age relationships ogie Ve´ge´tale et de Phytosociologie, Institut de Botanique, Universite´ de of 25 well-dated Swiss Holocene pollen sequences archived in the Alpine Neuchaˆtel, 112 pp. Palynological Database. Revue de Pale´obiologie 16, 433–80. Mitchell, E.A.D., Borcard, D., Buttler, A.J., Grosvernier, Ph., Gilbert, van der Knaap, W.O., van Leeuwen, J.F.N., Fankauser, A. and D. and Gobat, J.-M. 2000a: Horizontal distribution patterns of testate Ammann, B. 2000: Palyno-stratigraphy of the last centuries in Switzer- amoebae (Protozoa; Rhizopoda) in a Sphagnum magellanicum carpet. land based on 23 lake and mire deposits: chronostratigraphic pollen mark- Microbial Ecology, 39, 290–300. ers, regional patterns, and local histories. Review of Palaeobotany and Mitchell, E.A.D., Buttler, A., Grosvernier, Ph., Rydin, H., Albinsson, Palynology 108, 85–142. C., Greenup, A., Heijmans, M.P.P.D., Hoosbeek, M.R. and Saarinen, van Geel, B. and Renssen, H. 1998: Abrupt climate change around 2 650 T. 2000b: Relationships among testate amoebae (Protozoa), vegetation BP in north-west Europe: evidence for climatic teleconnections and a ten- and water chemistry in five Sphagnum-dominated peatlands in Europe. tative explanation. In Issar, A.S. and Brown, N., editors, Water, environ- New Phytologist 145, 95–106. ment and society in times of climatic change, Dordrecht: Kluwer, 21–41. Mitchell, E.A.D., Warner, B.G. Buttler, A. and Gobat J.-M. 1999: Eco- van Geel, B., Buurman, J. and Waterbolk, H.T. 1996: Archaeological logical patterns of testate amoebae (Protozoa) on peatlands in the Jura and palaeoecological indications of an abrupt climate change in The Mountains, Switzerland and France. Ecoscience 6(4), 565–76. Netherlands, and evidence for climatological teleconnections around 2650 Punt, W., editor 1976: Northwest European pollen flora, I. Amsterdam: BP. Journal of Quaternary Science, 11, 451–60. Elsevier, 145 pp. van Geel, B., van der Plicht, J., Kilian, M.R., Klaver, E.R., Kouwen- Punt, W. and Clarke, G.C.S. editors 1980: Northwest European pollen berg, J.H.M., Renssen, H., Reynaud-Farrera, I. and Waterbolk, H.T. flora, II. Amsterdam: Elsevier, 265 pp. 1998a: The sharp rise of D14C ca. 800 cal bc: possible causes, related —— editors 1981: Northwest European pollen flora, III. Amsterdam: climatic teleconnections and the impact on human environments. In Mook, Elsevier, 138 pp. W.G. and van der Plicht, J., editors, Proceedings of the 16th international 14 —— editors 1984: Northwest European pollen flora, IV. Amsterdam: C conference, Radiocarbon 40, 535–50. Elsevier, 369 pp. van Geel, B., van der Plicht, J. and Renssen, H. 1998b: Solar forcing Punt, W. and Blackmore, S. editors 1991: Northwest European pollen of abrupt climate change around 850 calendar years bc. In Peiser, B.J., flora, VI. Amsterdam: Elsevier, 275 pp. Palmer, T. and Bailey, M.E., editors, Natural catastrophes during Bronze Punt, W., Blackmore, S. and Clarke, G.C.S., editors 1988: Northwest Age civilisations, BAR International Series 728, 162–68. European pollen flora, V. Amsterdan: Elsevier, 154 pp. Wardenaar, E.C.P. 1987: A new hand tool for cutting peat profiles. Can- Punt, W., Blackmore, S. and Hoen, P.P., editors 1995: Northwest Euro- adian Journal of Botany 65, 1772–73. pean pollen flora, VII. Amsterdam: Elsevier, 275 pp. Warner, B.G. 1990: Testate amoebae (Protozoa). In Warner, B.G., editor, Reille, M. 1991: L’origine de la station de pin a` crochets de la tourbie`re Methods in Quaternary ecology, Geoscience Canada, Reprint Series 5, de Pinet (Aude) et de quelques stations isole´es de cet arbre dans les Vosges Geological Association of Canada, St John’s, 65–74. 80 The Holocene 11 (2001) Warner, B.G., Kubiw, H.J. and Hanf, K.I. 1989: An anthropogenic cause Welten, M. 1982: Vegetationsgeschichtliche Untersuchungen in den for quaking mire formation in southwestern Ontario. Nature 340, 380–84. westlichen Schweizer Alpen: Bern-Wallis. Denkschriften der Schweizer- Wegmu¨ller, S. 1966: U ¨ ber die spa¨t- und postglaziale Vegetationsgesch- ischen Naturforschenden Gesellschaft 95, DSNG 95, 104 pp. ichte des Su¨dwestlichen Jura. Mate´riaux pour le leve´ ge´obotanique de la Wright, H.E., Mann, D.H. and Glaser, P.H. 1984: Piston corers for peat Suisse 48, 1–144. and lake sediments. Journal of Ecology 65, 657–59. View publication stats