pmc logo imageJournal ListSearchpmc logo image
Logo of pnasPNAS Home page.Reference to the article.PNAS Info for AuthorsPNAS SubscriptionsPNAS About
Proc Natl Acad Sci U S A. 1999 November 23; 96(24): 13857–13862.
PMCID: PMC24155
Ecology
The delayed resurgence of equatorial forests after the Permian–Triassic ecologic crisis
C. V. Looy,* W. A. Brugman,* D. L. Dilcher, and H. Visscher*
*Laboratory of Palaeobotany and Palynology, Utrecht University, Budapestlaan 4, 3584 CD Utrecht, The Netherlands; and Paleobotany Laboratory, Florida Museum of Natural History, University of Florida, Gainesville, FL 32611
To whom reprint requests should be addressed. E-mail: c.v.looy/at/bio.uu.nl.
Contributed by D. L. Dilcher
Accepted September 22, 1999.
Abstract
In conjunction with the Permian–Triassic ecologic crisis ≈250 million years ago, massive dieback of coniferous vegetation resulted in a degradation of terrestrial ecosystems in Europe. A 4- to 5-million-year period of lycopsid dominance followed, and renewed proliferation of conifers did not occur before the transition between Early and Middle Triassic. We document this delayed re-establishment of equatorial forests on the basis of palynological data. The reconstructed pattern of vegetational change suggests that habitat restoration, migration, and evolutionary processes acted synergistically, setting the stage for successional replacement of lycopsid dominants by conifers within a period of ≈0.5 million years.
 
Understanding of patterns and processes of past ecologic crises is no longer a matter of purely academic interest. Studies of biotic and biogeochemical change related to mass extinction events may substantially contribute to a prediction of the long-term consequences of current man-induced environmental deterioration and its associated biodiversity decline (1, 2). In making such a contribution, we should focus on the potential of high-resolution analysis of distinctive survival and recovery phases after episodes of global ecosystem collapse (3). At present, conceptual models of survival and recovery are notably constrained by a variety of marine paleoecologic records (36). However, because of its role in global biomass storage and its sensitivity to environmental change, land vegetation is one of the most obvious biota to be investigated for patterns of survival and recovery. Notably, the aftermath of the unrivaled Permian–Triassic (P–Tr) crisis (7, 8) should serve as a valuable field of botanical inquiry.
At the very end of the Permian, dieback of woody vegetation dramatically affected terrestrial ecosystems (912). In the northern and southern humid climatic zones of Pangaea, it resulted in the widespread disappearance of peat forests (13, 14). In the semi-arid equatorial region, the dominant conifer taxa of the Late Permian Euramerican floral realm became extinct at, or close to, the P–Tr junction (15). Among the surviving plants, lycopsids played a central role in repopulating the initial ecological desert. Subsequent radiation and expansion among Isoetales (16) resulted in dense populations of the succulent quillwort Pleuromeia sternbergii (17).
Throughout Europe, plant megafossil information indicates that ecosystem recovery to precrisis levels of structure and function did not occur before the transition between the Early and Middle Triassic. A replacement of the Pleuromeia vegetation by coniferous forests is evidenced by the Voltzia-dominant communities that characterize the early part of the Middle Triassic (1719). In contrast to the discontinuous and qualitative plant megafossil record, successive pollen and spore assemblages can provide the sample size and stratigraphic spacing needed to resolve the temporal pathway of plant succession during this recovery phase. Here, we document and discuss the delayed return of woody plants in Europe on the basis of quantitative palynological data from an Early–Middle Triassic transition sequence in Hungary.
In current chronostratigraphic classification, the Early–Middle Triassic junction corresponds to the boundary between the Olenekian and Anisian Stages. In Europe, a continuous shallow marine Olenekian–Anisian transition sequence is particularly well developed in the Transdanubian Central Range in Hungary (20, 21). The boundary approximates the transition between the marls and limestones of the Csopak Marl Formation and the lagoonal dolomites and marls of the Asófő Dolomite Formation. In outcrops, the Csopak Marl Formation yields the Tirolites ammonoid fauna, indicative of the Spathian Substage of the Olenekian. Continuously cored exploration wells have enabled detailed biostratigraphic analysis of marine and land-derived palynomorphs (2023). Quantitative distribution patterns of spore and pollen types (Fig. 1) are obtained from samples of borehole Bakonyszűcs-3. Information on the botanical affinity of the types is summarized in Table 1.
Figure 1Figure 1
Quantitative distribution patterns of selected spore and pollen types from the Early to Middle Triassic (Olenekian–Anisian) transition sequence in the Transdanubian Central Range, Hungary (borehole Bakonyszűcs-3, 6 km east of the town (more ...)
Table 1Table 1
Botanical affinity of principal Late Permian and Early/Middle Triassic spore/pollen categories from Europe
Immediately apparent is the overwhelming abundance of Densoisporites nejburgii in the lower part of the Csopak Marl Formation, indicative of the Pleuromeia monocultures known from Olenekian megafossil records in Europe. Selaginellales and other pteridophytes are subordinate elements. In the upper part of the Csopak Marl Formation, gradual shifts in relative abundances give a straightforward picture of the sequence of the gymnosperm invasions. Initial invasion was by the conifer Yuccites, as reflected by the coming and proliferation of Voltziaceaesporites heteromorphus. Additional elements followed, including the possibly ullmanniaceous conifer that produced Jugasporites conmilvinus. The lycopsid component declines drastically and disappears from the record in the basal part of the Asófő Dolomite Formation. The decrease in Voltziaceaesporites and Jugasporites, and a simultaneous increase in Triadispora and other coniferous pollen, mark the onset of the Voltzia-dominant climax vegetation of the Anisian.
The general trend observed in compositional change between Densoisporites-dominant and Triadispora-dominant assemblages is consistent with coeval (semi) quantitative palynological records from outcrops and boreholes in many parts of Europe, northern Africa, and Israel (Fig. 2). The pollen and spore assemblages originate from both terrestrial (lacustrine, fluviatile) and shallow marine sequences. They are invariably dominated by allochthonous elements, thus reflecting changes in regional vegetation. Because of long-distance (wind, river) dispersal, the assemblages include a regionally derived mixture of pollen and spore types originating from both lowland and upland communities (24). The most complete comparative records are from boreholes in Poland (25, 26).
Figure 2Figure 2
Geographic distribution of Early to Middle Triassic palynological records from Europe and adjacent parts of Africa and Asia, reflecting part of a successional change from a Pleuromeia vegetation to a Voltzia-dominant climax vegetation. 1, Britain (61 (more ...)
Despite the difference in temporal scale, the unidirectional Early to Middle Triassic vegetation change in Europe mimics patterns of plant succession that occurred in response to the climate amelioration at the beginning of the Holocene approximately 11,000 years ago. In both cases, palynological data indicate that the re-establishment of climax forests proceeds through phases, in which herbs and shrubs are important functional members of the community. However, although rapid Holocene plant succession is notably regulated by habitat restoration and migration (27), it is conceivable that the long-term pattern of Early to Middle Triassic succession is also influenced by evolutionary processes. Significant influences of climate change are unlikely. Throughout the Late Permian, Early Triassic and Middle Triassic, terrestrial sedimentation in Europe is consistently dominated by the “Buntsandstein” redbed facies, indicative of warm, semi-arid conditions.
Under the prevailing semi-arid conditions, massive dieback of woody vegetation in Europe is likely to have resulted in extensive soil erosion. Early Triassic palynological data (Fig. 3) confirm the pioneering role for surviving lycopsids. Evidenced by the often frequent occurrence of Scythiana, bryophytes also appear to have been successful elements in the colonization of depopulated terrain (28). By analogy with many of their recent representatives, both lycopsids and bryophytes could have been preadapted to oligotrophic conditions. Earliest lycopsid pioneers may have included both Selaginellales and Isoetales. Because of their successful radiation and expansion after P–Tr extinctions, the Isoetales may be classified as crisis progenitors (36), specifically adapted to stressed environments of the Early Triassic survival phase. At least in Europe and other semi-arid regions, such as northern China (29), the resulting widespread vegetation, characterized by the relatively large succulent Pleuromeia sternbergii, remained stable for a longer period of time. This prolonged ecologic success of heterosporous lycopsids is remarkable and needs further investigation. Because of the unisexual gametophytes, free-sporing heterospory cannot be regarded as an effective long-distance dispersal strategy in the water-limited environments of Europe (30, 31).
Figure 3Figure 3
Generalized stratigraphic distribution of selected latest Permian and Early/Middle Triassic spore and pollen types in Europe. For information on the botanical affinity of the spores and pollen, see Table 1. The ecologic crisis and associated conifer (more ...)
In addition to lycopsid pioneers, the parent taxa of some of the accessory Early Triassic spore and pollen types could be survivors of the P–Tr crisis. Notably, multitaeniate pollen and Cycadopites may sometimes show more or less continuous Late Permian to Middle Triassic occurrences, but relative abundances vary considerably. These pollen types may represent lineages of small, stress-tolerant pteridosperms and cycads that survived as severely fragmented populations in locally favorable sites. There are no conclusive palynological data supporting a concomitant survival of conifers in Europe.
Temporal analysis of Early Triassic paleosol development (17) suggests a reduced rate of soil formation up to Late Olenekian times. By developing soil resources, the Pleuromeia-dominant vegetation probably contributed to the restoration of habitats necessary for the documented successional replacement by conifer-dominant vegetation (Fig. 1). There is no evidence, from either fossil wood or root horizons, that early successional conifers, such as Yuccites, could classify as trees. They probably formed densely populated shrubland. Part of the accessory gymnosperms may even represent herbaceous plants, as exemplified by Aetophyllum. This conifer genus, very rarely reflected in the Hungarian pollen record but common elsewhere in Europe, lacks secondary xylem tissue and had a ruderal life strategy (18, 19, 32). In contrast, the size of recovered fossil wood supports the idea that the late successional species of Voltzia include genuine trees (33). Root structures (17) indicate that the Voltzia-dominant climax vegetation had the character of a relatively open forest.
The successional dominants represent conifers that are unknown from the extensive Late Permian megafloral and palynological records. They are immigrants that invaded European habitats after their Early Triassic origination elsewhere. Dominant Late Permian conifers in Europe comprise mainly the families Walchiaceae, Ullmanniaceae, and Majonicaceae (3436). The long-ranging Carboniferous–Permian Walchiaceae became extinct close to the P–Tr boundary (15). This family is characterized by the presence of prepollen, i.e., pollen of zoidogamous gymnosperms that had not yet developed the capacity to produce nonflagellated sperm delivered by a pollen tube (15, 37). In effect, after the P–Tr crisis, the prepollen condition never re-appeared in gymnosperm evolution. On the other hand, the temporary re-appearance of Jugasporites could indicate that the Ullmanniaceae survived outside Europe. The ancestry of Yuccites is unknown, but Voltzia has a close phylogenetic relationship with the widespread Late Permian Majonicaceae (35, 36). It should be noted that Voltzia is only distantly related to Voltziopsis (36), a dominant conifer in Early Triassic vegetation of Gondwana (10, 12).
It may be hypothesized that the source areas of the Early Triassic immigrants in Europe had a refugial character. After the P–Tr crisis, extreme fragmentation and range contraction of ancestral populations could have resulted in small, vicariant populations segregated into isolated refugia, where suitable habitats continued to exist for extensive periods. It is thought that such refugia are bounded by ecotonal regions where, despite loss of original genetic variation, selection pressure is active for the fixation of genotypic and phenotypic novelties and enhanced adaptive radiation (38). There is no direct evidence where refugional source areas for the immigrants could have been located. At least for the Late Permian, accumulating information from Africa and Arabia supports the existence of a wide area with mixed Gondwana and Euramerican floras (39, 40) that may have served as a consistent source for northward migrating gymnosperm species. Co-occurrence of Densoisporites and Voltziaceaesporites starts earlier in the P–Tr section of the Salt Range in Pakistan than in Europe (41), suggesting a southern refugium for the ancestral lineage of Yuccites. However, because of the surprising discovery of a variety of Mesozoic-type gymnosperms in the Lower Permian of Texas (42), North America also must be taken into consideration as a potential source area for European Triassic conifers.
Radiometric age determinations for rocks associated with the Olenekian-Anisian transition are still lacking, so that constraints on the duration of survival and recovery remain highly speculative. Analysis of 40Ar/39Ar data from a late Anisian tuff in New Zealand yielded a date of 242.8 ± 0.6 million years (Ma) before the present (43), whereas a mean single-grain zircon U/Pb date of 241.0 ± 0.5 Ma was obtained from the basal Ladinian of northern Italy (44, 45). Considering the U/Pb age of 251.4 ± 0.3 Ma for the P–Tr boundary in South China (46), and assuming similar duration of the Early Triassic (Induan through Olenekian) and the Anisian, an interpolated age of about 246 Ma might be a realistic estimate for the Olenekian–Anisian boundary. A duration of 4 to 5 Ma for the process of Early Triassic repopulation can thus be inferred. Assuming approximately equal extent for the four Early Triassic substages, the documented late Spathian change from a Pleuromeia-dominant vegetation to the Voltzia-dominant climax vegetation could have taken place within ≈0.5 Ma.
The recovery of equatorial, semi-arid conifer forest in Europe is approximately coeval with the delayed recovery of humid peat forest in higher latitudes, as evidenced by the Early Triassic “coal gap” (13, 14). In general, the progress of terrestrial ecosystem recovery after the P–Tr crisis appears to be in harmony with the situation in the marine biosphere. Diversity analysis and paleoecologic study of benthic invertebrate faunas has for some time indicated, most notably by the interruption of reef building (48), that shallow marine ecosystems too did not fully recover before the Middle Triassic (7, 47). Records of radiolarians suggest a similar delay in the return to precrisis levels of zooplankton productivity (49). Curiously, in comparison to all other major ecologic crises in Earth history, repopulation after the P–Tr crisis proceeded exceptionally slowly (6). A coherent scenario explaining the strongly delayed pathways leading to terrestrial and marine ecosystem recovery still needs to be established. Yet, the reported pattern of forest recovery in Europe exemplifies that an evaluation of vegetation change may contribute significantly to an understanding of the synergistic action of postcrisis habitat restoration, and migration and evolutionary processes.
Acknowledgments
We thank the Geological Institute of Hungary for making available sample material. We are grateful to Doug Erwin and other participants in the International Geological Correlation Program Project (no. 335), Biotic Recoveries from Mass Extinctions, for inspiration, helpful discussion, and comments on the topic. We thank Bill DiMichele, Robert Gastaldo, and Greg Retallack for offering constructive critiques. This work was supported by the Council for Earth and Life Sciences with financial aid from The Netherlands Organization for Scientific Research. This paper is Netherlands Research School of Sedimentary Geology publication no. 990705, and University of Florida Contribution to Paleobiology publication no. 514).
Abbreviations
P–TrPermian–Triassic
Mamillion years

Footnotes
See commentary on page 13597.
References
1.
Buffetaut, E. Palaeogeogr Palaeoclimatol Palaeoecol. 1990;82:169–174.
2.
Myers, N. Palaeogeogr Palaeoclimatol Palaeoecol. 1990;82:175–185.
3.
Kauffman, E G; Erwin, D H. Geotimes. 1995;14(3):14–17.
4.
Kauffman, E G; Harries, P J. Geol Soc Sp Pap. 1996;102:15–39.
5.
Harries, P J; Kauffman, E G; Hansen, T A. Geol Soc Sp Pap. 1996;102:41–60.
6.
Erwin, D H. Trends Ecol Evol. 1998;13:344–349.
7.
Erwin, D H. The Great Paleozoic Crisis. New York: Columbia Univ. Press; 1993.
8.
Hallam, A; Wignall, P B. Mass Extinctions and Their Aftermath. Oxford: Oxford Univ. Press; 1997.
9.
Eshet, Y; Rampino, M R; Visscher, H. Geology. 1995;23:967–970.
10.
Retallack, G J. Science. 1995;267:77–80.
11.
Visscher, H; Brinkhuis, H; Dilcher, D L; Elsik, W C; Eshet, Y; Looy, C V; Rampino, M R; Traverse, A. Proc Natl Acad Sci USA. 1996;93:2155–2158. [PubMed]
12.
Retallack, G J. Geol Soc Am Bull. 1999;111:52–70.
13.
Veevers, J J; Conagham, P J; Shaw, S E. Geol Soc Am Sp Pap. 1994;288:187–196.
14.
Retallack, G J; Veevers, J J; Morante, R. Geol Soc Am Bull. 1996;108:195–207.
15.
Poort, R; Clement-Westerhof, J A; Looy, C V; Visscher, H. Rev Palaeobot Palynol. 1997;97:9–39.
16.
Retallack, G J. J Paleontol. 1997;71:500–521.
17.
Mader, D. Palaeoecology of the Flora in Buntsandstein and Keuper in the Triassic of Middle Europe. Vol. 1. Stuttgart: Fischer; 1990.
18.
Grauvogel-Stamm, L. Univ Louis Pasteur Strasbourg Inst Géol, Sci Géol Mém. 1978;50:1–225. [PubMed]
19.
Gall, J C; Grauvogel-Stamm, L; Nel, A; Papier, F. C R Acad Sci Paris, Earth Planet Sci. 1998;326:1–12.
20.
Haas, J; Tóth Makk, Á; Oravecz Scheffer, A; Góczán, F; Oravecz, J; Szabó, I. Ann Inst Geol Publ Hung. 1988;65:1–356.
21.
Broglio Loriga, C; Góczán, F; Haas, J; Lenner, K; Neri, C; Oravecz Scheffer, A; Posenato, R; Szabó, I; Tóth Makk, Á. Mem Sci Geol. 1990;42:41–103.
22.
Brugman, W A. Ph.D. thesis. Utrecht, The Netherlands: Utrecht Univ.; 1986.
23.
Góczán, F; Oravecz-Scheffer, A; Szabó, I. Acta Geol Hung. 1986;29:233–259.
24.
Traverse, A. Paleopalynology. Boston: Unwin Hyman; 1988.
25.
Orłowska-Zwolińska, T. Acta Palaeontol Pol. 1984;29:161–194.
26.
Orłowska Zwolińska, T. Bull Pol Acad Sci Earth Sci. 1985;33:107–117.
27.
Delcourt, H L; Delcourt, P A. Quaternary Ecology - A Paleoecological Perspective. London: Chapman & Hall; 1991.
28.
Brinkhuis, H; Visscher, H. Lunar Planet Inst Contr. 1994;825:17.
29.
Wang, Z. Rev Palaeobot Palynol. 1996;91:121–142.
30.
DiMichele, W A; Davis, J I; Olmstead, R G. Taxon. 1989;38:1–11.
31.
Bateman, R M; DiMichele, W A. Biol Rev. 1994;69:345–417.
32.
Mapes, G; Rothwell, G W; Grauvogel-Stamm, L. Am J Bot. 1997;84:137.
33.
Mägdefrau, K. Paläobiologie der Pflanzen. 4th Ed. Jena, Germany: Fischer; 1968.
34.
Clement-Westerhof, J A. Rev Palaeobot Palynol. 1984;41:51–166.
35.
Clement-Westerhof, J A. Rev Palaeobot Palynol. 1987;52:375–402.
36.
Clement-Westerhof, J A. Origin and Evolution of Gymnosperms. Beck C B. , editor. New York: Columbia Univ.; 1988. pp. 298–337.
37.
Poort, R J; Visscher, H; Dilcher, D L. Proc Natl Acad Sci USA. 1996;93:11713–11717. [PubMed]
38.
Levin, D A; Wilson, J B. Structure and Functioning of Plant Populations. Freysen A H J, Woldendorp J W. , editors. Amsterdam: North-Holland; 1978. pp. 25–100.
39.
Broutin, J; Doubinger, J; El Hamet, M O; Lang, J. Rev Palaeobot Palynol. 1990;66:243–261.
40.
Broutin, J; Roger, J; Platel, J P; Anglioni, L; Baud, A; Bucher, H; Marcoux, J; Hasmi, A H. C R Acad Sci Paris, Sér IIa. 1990;321:1069–1086.
41.
Balme, B E. Univ. Kansas Dept. Geol. Spec. Publ. 4, Stratigraphic Boundary Problems: Permian and Triassic of West Pakistan. Kummel B, Teichert C. , editors. Lawrence, KS: Univ. Kansas Press; 1970. pp. 305–453.
42.
Chaney, D S; Mamay, S H; DiMichele, W A. Geol. Soc. Am. Abstr. Prog. 1998. , A-312.
43.
Retallack, G J; Renne, P R; Kimbrough, D L. New Mex Mus Nat Hist Sci Bull. 1993;3:415–418.
44.
Brack, P; Rieber, H; Mundil, R. Albertiana. 1995;15:45–56.
45.
Mundil, R; Brack, P; Meier, M; Rieber, H; Oberli, F. Earth Planet Sci Lett. 1996;141:137–151.
46.
Bowring, S A; Erwin, D H; Jin, Y G; Martin, M W; Davidek, K; Wang, W. Science. 1998;280:1039–1045. [PubMed]
47.
Schubert, J K; Bottjer, D J. Palaeogeogr Palaeoclimatol Palaeoecol. 1995;116:1–39.
48.
Flügel, E. Geol Soc Am Sp Pap. 1994;288:247–266.
49.
Kukawa, Y. Palaeogeogr Palaeoclimatol Palaeoecol. 1996;121:35–51.
50.
Brinkhuis, H; Schiøler, P. Geol Mijnbouw. 1996;75:193–213.
51.
Lundblad, B. Medd Dansk Geol Foren. 1948;11:351–363.
52.
Balme, B E. Rev Palaeobot Palynol. 1995;87:81–323.
53.
Yaroshenko, O P. Paleontol Zh. 1985;19:110–117.
54.
Potonié, R; Schweitzer, H J. Paläontol Z. 1960;34:27–39.
55.
Klaus, W. Erdöl Z. 1964;80:119–132.
56.
Gomankov, A V; Meyen, S V. Trans Inst Akad Nauk SSSR. 1986;401:1–173.
57.
Meyen, S V. Rev Palaeobot Palynol. 1997;96:351–447.
58.
Grauvogel-Stamm, L. Palaeontographica B. 1972;140:1–26.
59.
Grauvogel-Stamm, L; Álvarez Ramis, C. Cuad Geol Ibérica. 1996;20:229–243.
60.
Visscher, H; Brugman, W A. Rev Palaeobot Palynol. 1981;34:115–128.
61.
Warrington, G. Q J Geol Soc London. 1970;126:183–223.
62.
Warrington, G. Rev Palaeobot Palynol. 1974;17:133–147.
63.
Visscher, H. Acta Bot Neerl. 1966;15:316–375.
64.
Doubinger, J; Bühmann, D. Z Deut Geol Ges. 1981;132:421–449.
65.
Ecke, H H. Ph.D. thesis. Göttingen, Germany: Georg August Univ.; 1986.
66.
Reitz, E. Geol Abh Hessen. 1985;86:1–36.
67.
Reitz, E. Geol Abh Hessen. 1988;116:105–112.
68.
Reinhardt, P; Schön, M. Monatsber Deut Akad Wiss Berlin. 1967;9:747–758.
69.
Reinhardt, P; Schmitz, W. Freiberger Forschungsh C. 1965;182:19–26.
70.
Schulz, E. Abh Zentr Geol Inst. 1965;1:257–287.
71.
Schulz, E. Abh Zentr Geol Inst. 1966;8:3–20.
72.
Adloff, M C; Doubinger, J. Bull Serv Carte Géol Alsace Lorraine. 1969;22:131–148.
73.
Doubinger, J; López-Gómez, J; Arche, A. Rev Palaeobot Palynol. 1990;66:25–45.
74.
Diez, J B; Grauvogel-Stamm, L; Broutin, J; Ferrer, J; Gisbert, J; Linan, E. C R Acad Sci Paris, Sér IIa. 1996;323:341–347.
75.
Solé de Porta, N; Calvet, F; Torrentó, L. Cuad Geol Ibérica. 1987;11:237–254.
76.
Broutin, J; Doubinger, J; Gisbert, J; Satta-Pasini, S. C R Acad Sci Paris, Sér IIa. 1988;306:159–163.
77.
Ramos, A; Doubinger, J. C R Acad Sci Paris, Sér IIa. 1989;309:1089–1094.
78.
Pittau Demelia, P; Flaviani, A. Rev Palaeobot Palynol. 1982;37:329–343.
79.
Barabás-Stuhl, Á. Acta Geol Acad Sci Hung. 1981;24:49–97.
80.
Antonescu, E; Patrulius, D; Popescu, I. Inst Geol Geof Dări de Seamă 1976;62:3–30.
81.
Kilani-Mazraoui, F; Razgallah-Gargouri, S; Mannai-Tayech, B. Rev Palaeobot Palynol. 1990;66:273–291.
82.
Adloff, M C; Doubinger, J; Massa, D; Vachard, D. Rev Inst Franç Pétr. 1986;41:27–72.
83.
Brugman, W A; Visscher, H. Subsurface Palynostratigraphy of Northeast Libya. El-Arnauti A, Owens B, Thusu B. , editors. Benghazi, Libya: Garyounis Univ. Press; 1988. pp. 157–169.
84.
Eshet, Y. Geol Surv Isr Bull. 1990;81:1–57.
85.
Eshet, Y. Isr J Earth Sci. 1990;39:1–15.
86.
Visscher, H; Brugman, W A. Mem Soc Geol It. 1988;34:121–128.
87.
Yin, H; Zhang, K; Wu, S; Peng, Y. The Palaeozoic-Mesozoic Boundary. Yin H. , editor. Wuhan: China Univ. Geosci. Press; 1996. pp. 3–28.