Cleal, C.J., Thomas, B.A., Batten, D.J. & Collinson, M.E. 2001. Mesozoic and Tertiary Palaeobotany of Great Britain. Geological Conservation Review Series No. 22, JNCC, Peterborough, ISBN 1 86107 489 1. The original source material for these web pages has been made available by the JNCC under the Open Government Licence 3.0. Full details in the JNCC Open Data Policy
Lake
Introduction
This is one of only two sites still yielding fossil plants from lower Eocene beds traditionally called the 'Dorset Pipe Clays'. These deposits are of about the same age as the London Clay, but are fluvial rather than marine deposits and thus largely lack the mangrove-palm-dominated vegetation seen in the London Clay. Lake yields the most diverse of the Dorset Pipe Clay floras, containing 69 species of mainly angiosperms and ferns, 30 of which (and 3 genera) are unique to the site.
In the western part of the Hampshire Basin, the London Clay Formation sensu stricto is mostly missing, and is replaced by fluviatile sediments. These sediments have included a number of historically important palaeobotanical sites, including Alum Bay on the Isle of Wight (see La Harpe and Salter (in Bristow, 1862), Gardner (in Reid and Strahan, 1889) and Crane, 1977, 1978) and Corfe and Studland in Dorset (Brodie, 1853; Gardner, 1877, 1886a; Gardner and von Ettingshausen, 1879; Chandler, 1962). These leaf beds now yield very little material, but in recent years it has been discovered that there are also fruit and seed deposits, which are much more productive. The site at Lake, on the shore of Poole Harbour, was discovered in 1938 by E. St J. Burton and most of what we know about the flora here is based on his collection (now stored at the Natural History Museum, London). Chandler (1955, 1962) has described the fossils from here, and some were briefly mentioned by Collinson (1980a), who also illustrated a Ficus fruit (Collinson, 1989). Fruits of the dogwood family from here were also used in a chemical investigation of fossil resins (van Aarssen et al., 1994).
Description
Stratigraphy
The low cliffs here
Palaeobotany
Lake has yielded nearly 70 species of plant fossil, the full assemblage being shown in
The enigmatic 'Scirpus' lakensis (see
Interpretation
Lake has produced by far the most diverse and well-preserved fossil flora from the Dorset Pipe Clays. The only comparable site is Arne, but that has yielded only just over half the number of species. The Arne fossils also often suffer from the coarse overgrowth of pyrite crystals, which can mask the morphology of the fruits and seeds.
The dominant plant fossils at Lake are what Chandler (1962) identified as isolated fruits of Scirpus. However, essentially identical specimens in the Eocene strata at Messel occur within receptacular fruiting heads, showing that they are seeds and thus totally unlike Scirpus but more similar to modern Cyclanthaceae (Panama hat family) (Collinson, 1982b, 1988, 1996b). Collinson (1996b) showed that these fossils came from plants that grew along rivers and lakeshores. Further evidence of aquatic conditions is the presence of the water lily, Palaeonymphaea (Collinson, 1980a). The relatively high proportion of icacina, moonseed and grape family members probably reflects a dense swathe of lianas, growing on trees surrounding the rivers and lakes, whose fruits dropped directly into the water.
The Dorset Pipe Clay flora at Lake is important because it complements the more diverse London Clay flora, such as found at Sheppey, Bognor Regis and Herne Bay. Whereas the Nypa–Ceriops mangrove vegetation that fringed the coast dominates the London Clay flora, the Dorset Pipe Clay flora represents more inland vegetation. Only one example of a Nypa fruit has been recorded from the Dorset Pipe Clays, from a temporary exposure near Stoborough, Dorset (Collinson, 1993, 1996b). There are nevertheless many similarities between the floras, especially at the family level. Putting the dominant diagnostic elements of this interval aside (Nypa in the London Clay and 'Scirpus' for the Dorset Pipe Clays — Collinson, 2000a), both assemblages include species of the dillenia, dogbane, dogwood, flacourtia, grape, icacina, moonseed, sapodilla, spurge, sumac, sweetleaf and tea families. There are groups present at Lake and not in the London Clay, including the caper, elder, ebony and styrax families. Nevertheless, it seems that behind the effects of the 'local' mangrove versus lake or river margin vegetation, both floras are representing basically similar, paratropical forests.
Family | Species | Lake | Arne | Studland |
Pteridaceae | Acrostichum lanzaeanum (Visiani) Chandler | × | × | |
Schizaeaceae | Lygodium kaulfussii Heer emend. Gardner and Ettingshausen | × | ||
L. poolensis Chandler | × | |||
Anemia poolensis Chandler | × | × | ||
Ruffordia subcretacea (Saporta) Barthel, 19761 | × | |||
Taxodiaceae | Taxodium lakensis Chandler | × | × | |
Sequoia couttsiae Heer2 | × | |||
Actinidiaceae | Saurauia crassisperma (Chandler) Mai3 | × | ||
S. poolensis (Chandler) Mai, 19704 | × | |||
Anacardiaceae | Dracontocarya glandulosa Chandler | × | ||
?Lannea sp. | × | |||
Rhus lakensis Chandler | × | |||
R. spp. | × | |||
Apocynaceae | Apocynospermum acutiforme Chandler5 | × | ||
A. lakense Chandler5 | × | |||
Arecaceae | Calamus daemonorops (Unger) Chandler | × | ||
?cabal sp. | × | |||
Boraginaceae | Ehretia lakensis Chandler | × | ||
Burseraceae | Palaeobursera lakensis Chandler | × | ||
Capparaceae | Burtonella emarginata Chandler | × | × | × |
Palaeocleome lakensis Chandler | × | |||
Capparidispermum eocenicum Chandler | × | |||
Caprifoliaceae | Sambucus parvula Chandler | × | ||
Cornaceae (including Mastixiaceae) | Dunstania lakensis Chandler6 | × | ||
Eomastixia rugosa (Zenker) Chandler (see Mai, 1993) | × | × | ||
E. urceolata Chandler | × | |||
?Mastixia cantiensis Reid and Chandler7 | × | |||
Mastixicarpum crassum Chandler (see Mai, 1993) | × | |||
Swida quadrilocularis (Chandler) Mai, 19998 | × | |||
Cucurbitaceae | Cucurbitospermum lakense Chandler | × | ||
C. obliquum Chandler | × | |||
Cyperaceae | 'Scirpus' lakensis Chandler | × | × | |
?Scirpus sp. | × | |||
Caricoidea arnei Chandler | × | |||
C. obscura Chandler | × | |||
?Caricoidea sp. | × | |||
Cladiocarva minima (Chandler) Mai in Mai and Walther, 19789 | × | |||
Ebenaceae | Diospyros headonensis Chandler | × | ||
Euphorbiaceae | Euphorbiotheca lakensis Chandler | × | ||
E. platysperma Chandler | × | |||
E. tuberculata Chandler | × | |||
E. digitata Chandler | × | |||
Euphorbiospermum punctatum Chandler | × | |||
Wetherellia variabilis Bowcrbank | × | |||
Flacourtiaceae | Oncoba rugosa Chandler | × | ||
Hamamelidaceae | Steinhauera subglobosa Presl10 | × | ||
Icacinaceae | Iodes acutifornzis Chandler | × | × | |
Natsiatum eocenicum Chandler11 | × | |||
?Palaeophytocrene foveolata Reid and Chandler | × | |||
kacinicarya inornata Chandler | × | × | ||
Lauraceae | Laurocatpum spp. | × | ||
Lythraceae | Anzmannia lakensis Chandler | × | ||
Alatospermum lakense Chandler | × | |||
Menispermaceae | Tinospora arnensis Chandler | × | × | |
Palaeococculus lakensis Chandler | × | × | ||
Wardensheppeya poolensis (Chandler) Eyde, 1970 | × | |||
Moraceae | Ficus lucidus Chandler (see Collinson, 1989) | × | ||
F. sp . | × | |||
?Moraceae | Ovicarpum reticulatum Chandler (see Collinson, 1989) | × | ||
Nymphaeaceae | Palaeonymphaea eocenica Chandler (see Collinson 1980a) | × | ||
Nyssaceae | Nyssoidea eocenicum Chandler | × | × | |
Rosaceae | Rubus acutiformis Chandler | × | ||
Rutaceae | Phellodendron costatum Chandler | × | ||
Rutaspermum excavatum Chandler | × | |||
R. glabrum Chandler | × | |||
R. magnificum Chandler | × | |||
R. striatum Chandler | × | |||
Sabiaceae | ?Meliosma sheppeyensis Reid and Chandler | × | ||
Sapotaceae | ?Sapoticarpum sp. | × | ||
Solanaceae
- |
Solanunz arnense Chandler | × | ||
Solanisperrnum reniforme Chandler | × | |||
Styracaceae | Styrax elegans Chandler | × | ||
Symplocaceae | Symplocos headonensis Chandler | × | ||
S. lakensis Chandler | × | × | ||
Theaceae | Cleyera? obliqua Chandler | × | ||
?Gordonia sp. | × | |||
Thymelaeaceae | Thymelaeaspermum lakense Chandler | × | × | |
T? sulcatum Chandler | × | |||
Vitaceae | Vitis ambigua Chandler | × | ||
V. arnensis Chandler | × | |||
V. cuneata Chandler | × | |||
V. excavata Chandler | × | |||
V. lakensis Chandler | × | |||
V. lusatica Czeczott and Skirgiello12 | × | × | ||
V. platysperma Chandler | × | × | ||
V. poolensis Chandler | × | |||
V. pygmaea Chandler | × | × | ||
V. goodhartii Chandler | × | × | ||
V. symmetrica Chandler | × | |||
V. triangularis Chandler | × | |||
Tetrastigma acuminata Chandler | × | |||
?T lobata Chandler | × | |||
Zingiberaceae | Alpinia arnense (Chandler) Mai in Mai and Walther, 198513 | × | ||
Incertae sedis | Rhamnospermum bilobatum Chandler | × | × | |
Carpolithus arnense Chandler | × | |||
Footnotes to |
||||
1 Ruffordia subcretacea (Saporta) Barthel (1976) is the name in current use for Anemia subcretacea (Saporta) Gardner and Ettingshausen. However, the fossil, known as an almost complete plant, is very similar to Anemia and certainly belongs in the Glade including Anemia (Collinson, in press a). | ||||
2 Sequoia couttsiae Heer = Athrotaxis couttsiae (Heer) Gardner, both Taxodiaceae. The former name has been applied to British material but the latter is used currently in continental Europe. However, as both these genera are modern genera, more than a mere nomenclatural decision is involved here. For the British material, Chandler (1925–1926 p.13) initially rejected Gardner's assignment to Athrotaxis. She reconfirmed and re-instated the affinity with Sequoia in full and made detailed studies of leaves, leafy shoots, twigs, cones and seeds (Chandler, 1962, 1963b, 1964), noting a marked similarity to Sequoia sempervirens (Chandler, 1964, 1978, p. 40, under discussion of Sequoiadendron fordit) and rejecting affinity with Sequoiadendron (Chandler, 1964, p. 104, 1978 pp. 40 and 41). Fowler et al. (1973) emphasized the difficulties of determining isolated foliage of Taxodiaceae, noting that Chandler (1964) had stated that the leaves of S. couttsiae were not identical with either Sequoia or Sequoiadendron. Ruffle (1976) also treated the species as a member of genus Sequoia. Subsequently workers in continental Europe (Mai and Walther, 1978, 1985, 1991; Mai, 1998; Knobloch et al., 1996) have assigned the species to Athrotaxis citing the work of Dorofeev and Sveshnikova (1963) as the basis for this assignment. However, Dorofeev and Sveshnikova (1963) combined a range of material into their recombination Athrotaxis taxiformis (Unger) Dorofeev and Sveshnikova (including S. couttsiae and A. couttsiae). Athrotaxis taxiformis in their sense included material later assigned to an extinct genus of Taxodiaceae Doliostrobus Marion by Kvaček (1971). Mai (1998) expressed some reservations as to the use of the genus Athrotaxis for the remaining material. Furthermore, Dorofeev and Sveshnikova (1963) were apparently unaware of, and did not cite, the work of Chandler (1962, 1963, 1964) and, to the best of our knowledge, they had not been able to study the British material. There is considerable variation in Taxodiaceae leaves resulting in similarities between those of Taxodium, Sequoia, Sequoiadendron and Glyptostrobus. S. couttsiae is associated with tree stumps with wood of the Glyptostroboxylon type (Fowler et al., 1973). It is therefore possible that, even though Chandler found a convincing affinity with Sequoia based on foliage, cones and seeds, a fully reconstructed plant bearing S. couttsiae foliage and cones might not resemble modern Sequoia in all features. Until such a plant can be reconstructed, ideally based on organic attachment, and included in a dadistic analysis of the Taxodiaceae, its relationships must remain slightly uncertain. Finally one must consider the relative unlikelihood of a relationship with modern Athrotaxis, which is endemic to Tasmania. There are numerous relationships between the Palaeogene floras of Europe, and those of the USA and Asia (Manchester, 1999). Numerous modern genera recorded in these Palaeogene floras now grow in south-eastern Asia and America (Manchester, 1999; Tiffney, 1994), some occur in Africa and South America (Tiffney, 1994) and a few occur in Australia (Tiffney, 1994). However, all those occurring in Australia also occur today in eastern Asia or in both eastern Asia and the New World; none are Australian endemics today. Some claims for records of Proteaceae and Cunoniaceae in the northern hemisphere Tertiary are all based on very old literature and all have been subsequently rejected (Mai, 1995). Thus, there is no evidence for floristic affinity between Australia and the European Palaeogene, and the occurrence of a modern Australian endemic in the European Palaeogene is judged to be extremely unlikely. For all these complex reasons we have retained the name in current use in Britain for British material i.e. Sequoia couttsiae. While this volume was in press, Kunzman (1999) reassessed the systematic affinity of S. couttsiae and judged that it represented an extinct member of the Taxodiaceae, which he assigned to the genus Quasisequoia as Q. couttsiae (Heer) Kunzman. | ||||
3 Formerly Hordwellia crassisperma (Chandler) Chandler and thought to belong to the Theaceae (see Mai and Walther, 1985). | ||||
4 Formerly Actinidia poolensis Chandler, 1963b. | ||||
5 According to Manchester (1999, p. 476), the genus Echitonium Unger has priority for apocynaceous seeds for which the generic affinity is unclear. | ||||
6 Dunstania has been assigned to Cornus by some authors (e.g. Eyde, 1988) (see discussion in Manchester, 1994, p. 42). | ||||
7 Mai (1993) synonymized this with Mastixiopsis (Kirchheimer). | ||||
8 Formerly Cornus quadrilocularis Chandler. | ||||
9 Formerly Caricoidea minima (Chandler) Chandler. | ||||
10 Includes Protaltingia hantonensis Chandler (see Mai and Walther, 1985). | ||||
11 See footnote 8 to |
||||
12 Includes Vitis glabra Chandler (see Mai and Walther, 1991). | ||||
13 Formerly Aracispermum arnense Chandler, then included within the Araceae. |
As suggested by Collinson (1983b), a superficial appraisal of the London Clay and Dorset Pipe Clay fruit and seed assemblages (compare Tables 8.1 and 8.2) suggests marked differences at the rank of species. This may, however, be as much to do with the differences in preservation as with differences in the composition of the original vegetation (Collinson, 1983b). For instance, Dunstania glandulosa, Iodes acutiformis, Icacinicarya inornata and Palaeobursera lakensis from the Dorset Pipe Clay may well be the same as D. multilocularis, I. comiculata, I. platycarpa and P bognorensis from the London Clay, respectively (Chandler, 1962). The classic London Clay palaeobotanical sites (e.g. Sheppey) produce pyrite petrifactions that often yield the type of anatomical detail not seen in the Lake fossils. Furthermore, they are less vulnerable to shrinkage during fossilization, which can produce apparently dramatic differences in size and morphology Work on the sites that yield carbonaceous London Clay-type fruits and seeds (e.g. Walton-on-the-Naze) might help resolve this problem (Collinson, 1983b).
Another difference between the floras is the apparently greater abundance of smaller fossils at Lake, such as Ficus lucidus, Ebretia lakensis, Palaeocleome lakense, Capparidispermum eocenicum and Alatospermum lakense. However, this is probably mainly because of tidal winnowing of the seeds at Sheppey removing the smaller fraction rather than any significant difference in the original vegetation. Collinson (1983b, p. 16) noted a comparable difference between surface-picked larger fruits and smaller fruits in sieved concentrates at Sheppey.
Associated with the angiosperm fruits are very small fragments of fern foliage. They show extremely fine preservation of the reproductive structures, from which Chandler (1955) was able to establish that they all belong to the now mainly tropical family, the Schizaeaceae. They demonstrate that Lygodium, Anemia and the Anemialike Ruffordia had a much wider geographical range during the Eocene Epoch than they have today (Collinson, 1996a, in press a), a situation similar to that which we observe in the angiosperms.
Conclusions
The coastal exposures at Lake have yielded the most diverse assemblages of fossil fruits and seeds from the Lower Eocene Dorset Pipe Clays, about 50 Ma old. It preserves evidence of vegetation probably surrounding lakes or rivers, together with paratropical forest trees and lianas. It complements the floras of the similar-aged London Clay, as it lacks evidence of the Nypa mangrove-palm vegetation fringing the coasts. In contrast, it is dominated by 'Scirpus' lakensis from a river or lake marginal plant. Nypa and 'S.' lakensis are together diagnostic for early and middle Eocene floras in Britain. It also includes remains of the fern family Schizaeaceae, far outside the geographical range of their living relatives.