Environmental Biology of Fishes 32: 59-74, 1991. 0 1991 Kluwer Academic Publishers. Printed in the Netherlands. Diversity of extinct and living actinistian fishes (Sarcopterygii) Richard Cloutierl,* & Peter L. ForeyZ I Museum of Natural History, Department of Systematics and Ecology, The University of Kansas, Lawrence, KS 660452454, U. S. A. 2 Department of Palaeontology, British Museum (Natural History), Cromwell Road, London SW7 5BD, England Received 15.4.1990 Accepted 11.8.1990 Key words: Coelacanth, Fossil record, Species diversity Synopsis A total of 121actinistian species belonging to 47 genera and 17 undetermined actinistians is reported from the literature. There are 69 valid specieswith fair assessment of their phylogenetic position; 21 valid species with poor assessmentof their phylogenetic position; 31 actinistian incertae sedis; and 18 taxa that had been identified incorrectly as actinistians or are nomen nuda. The fossil record of the actinistians covers a history of approximately 380 million years. The greatest diversity occurred during the Scythian (Early Triassic). Introduction Actinistian systematic overview The Actinistia (= Coelacanthi, Coelacanthia, Coelacanthiformes, Coelacanthii, Coelacanthina), represented today by a single living species, Lati- meria chalumnae Smith, has a long fossil record beginning with the Middle Devonian Euporosteus eifelianus (Givetian) and ending with the Late Cre- taceous Macropoma lewesiensis (Santonian). Most of us are familiar with the past 65 million year stratigraphic gap (i.e., lack of fossils). Nevertheless the group had been known from fossils prior to the discovery of L. chalumnae for more than a century. The present contribution provides a list of actinis- tian taxa with our assessmentsof synonyms. In presenting this taxonomic list, we hope to provide some basis for assessment of species diversity, and also to give some idea of the basis of knowledge since it remains true that many species are known only by very fragmentary remains. The systematics of actinistians is poorly resolved. Prior to the discovery of Latimeria chalumnae in 1938, actinistians were thought to belong to a single crossopterygian family - the Coelacanthidae. Some authors (Berg 1940, Vorobyeva & Obruchev 1964, Lehman 1966, Romer 1966, Andrews et al. 1967, Thomson 1969, Moy-Thomas & Miles 1971, Andrews 1973, Carroll 1988) have addressed the classification of actinistians, but these attempts have resulted in the recognition of gradal and monotypic groups (Forey 1981, Cloutier 1990). No generally accepted phylogeny or higher taxonomic scheme is used for the Actinistia. However, the most commonly referred classifications are those of Berg (1940), Vorobyeva & Obruchev (1964), An- drews et al. (1967), Lund & Lund (1985), and Car- roll (1988). Moy-Thomas (1937)) Forey (1981)) and Cloutier (1990) have investigated the systematicsof Paleozoic actinistians and provided lists of Paleo- zoic taxa. However, the clarification of interrela-
16
Embed
Cloutier and Forey 1991 Diversity of Extinct and Living Actinistian Fishes (Sarcopterygii)
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Environmental Biology of Fishes 32: 59-74, 1991. 0 1991 Kluwer Academic Publishers. Printed in the Netherlands.
Diversity of extinct and living actinistian fishes (Sarcopterygii)
Richard Cloutierl,* & Peter L. ForeyZ I Museum of Natural History, Department of Systematics and Ecology, The University of Kansas, Lawrence, KS 660452454, U. S. A. 2 Department of Palaeontology, British Museum (Natural History), Cromwell Road, London SW7 5BD, England
Received 15.4.1990 Accepted 11.8.1990
Key words: Coelacanth, Fossil record, Species diversity
Synopsis
A total of 121 actinistian species belonging to 47 genera and 17 undetermined actinistians is reported from the literature. There are 69 valid species with fair assessment of their phylogenetic position; 21 valid species with poor assessment of their phylogenetic position; 31 actinistian incertae sedis; and 18 taxa that had been identified incorrectly as actinistians or are nomen nuda. The fossil record of the actinistians covers a history of approximately 380 million years. The greatest diversity occurred during the Scythian (Early Triassic).
Introduction Actinistian systematic overview
The Actinistia (= Coelacanthi, Coelacanthia, Coelacanthiformes, Coelacanthii, Coelacanthina), represented today by a single living species, Lati- meria chalumnae Smith, has a long fossil record beginning with the Middle Devonian Euporosteus eifelianus (Givetian) and ending with the Late Cre- taceous Macropoma lewesiensis (Santonian). Most of us are familiar with the past 65 million year stratigraphic gap (i.e., lack of fossils). Nevertheless the group had been known from fossils prior to the discovery of L. chalumnae for more than a century. The present contribution provides a list of actinis- tian taxa with our assessments of synonyms. In presenting this taxonomic list, we hope to provide some basis for assessment of species diversity, and also to give some idea of the basis of knowledge since it remains true that many species are known only by very fragmentary remains.
The systematics of actinistians is poorly resolved. Prior to the discovery of Latimeria chalumnae in 1938, actinistians were thought to belong to a single crossopterygian family - the Coelacanthidae. Some authors (Berg 1940, Vorobyeva & Obruchev 1964, Lehman 1966, Romer 1966, Andrews et al. 1967, Thomson 1969, Moy-Thomas & Miles 1971, Andrews 1973, Carroll 1988) have addressed the classification of actinistians, but these attempts have resulted in the recognition of gradal and monotypic groups (Forey 1981, Cloutier 1990). No generally accepted phylogeny or higher taxonomic scheme is used for the Actinistia. However, the most commonly referred classifications are those of Berg (1940), Vorobyeva & Obruchev (1964), An- drews et al. (1967), Lund & Lund (1985), and Car- roll (1988). Moy-Thomas (1937)) Forey (1981)) and Cloutier (1990) have investigated the systematics of Paleozoic actinistians and provided lists of Paleo- zoic taxa. However, the clarification of interrela-
60
tionships of post-Paleozoic actinistians, which cor- responds to most of the specific diversity, has only recently been attempted (Cloutier 1991, Forey 1991).
The compiled tabulation (Table 1) provides a list of all genera, species, and undetermined actinis- tians mentioned in the literature. This list is supple- mented by geological, stratigraphic, and geograph- ic information for each taxon. The primary author for each taxon is provided and included in the bibliography article (Forey & Cloutier 1991); how- ever, authors of subsequent taxonomic changes are not given in the list but are incorporated in the bibliography article. For each genus, the type spe- cies is listed first, followed by other referred species alphabetically and the undetermined species.
Assessments about synonymies of taxa have been followed only when it was justified in the literature or following work in progress by one or other of us. Many actinistian species were de- scribed originally under different genera; in order to indicate these nomenclatural changes, we follow the convention adopted in the Handbook of Pale- oichthyology (see e.g., Denison 1978) by refering to the original generic assignment as the first entry of synonymy. Spelling of species names is in agree- ment with the International Code of Zoological Nomenclature (Ride et al. 1985). Among the spe- cies that have been recognized herein, it remains likely that some are synonyms or not valid, but further systematic studies are needed and beyond the scope of this paper. For example, there are three species in the genus Rhipis, all of which are Late Cretaceous taxa known solely from scales, and two of them come from the same locality; thus it is possible that these forms represent a single species. For this reason a brief statement of parts of the animal known is given with each entry.
Our tabulation is divided into four categories reflecting our systematic interpretation of these taxa. For each category the genera are listed al- phabetically. Category ‘A’ deals with the genera and species that can be placed with some confi- dence within at least one of the cladistic classifica- tions given in Forey (1991) and Cloutier (1991). This category includes species that can be posi- tioned in reference to terminal taxa, nodes, re-
stricted clades, or a restricted range of nodes on the cladograms.
Category ‘B’ lists the genera and species that are considered to be valid species but for which little else may be said about their relationships (i.e., phylogenetic position). This category accounts for the Actinistia incertae sedis. These taxa are impor- tant for estimating total diversity but are of little use when estimating rates of clade evolution (morphological and taxonomic).
Category ‘C’ lists genera and species of those remains that can only be identified as actinistians. Their status as valid species is questioned. Finally, category ‘D’ includes the taxa that are undefined (nomen nuda), or those that have been cited as coelacanths but are referable to other fish groups.
Based on our tabulation, there are 69 valid spe- cies with fair assessment of their phylogenetic posi- tion (Category A); 21 valid species with poor as- sessment of their phylogenetic position (Category B); 31 actinistian incertae sedis (Category C); and 18 taxa that had been recognized in the literature as actinistians but are not (Category D). There are 47 recognized genera, 29 of which are monotypic and 17 undetermined actinistians (most of which are partial remains). Thus, there is a total of 90 valid species and 31 incertae sedis actinistians distributed from the Givetian (Middle Devonian) to the Re- cent; this total differs from that of Forey (1984) who reported approximately 70 species belonging to 28 actinistian genera.
Fossil actinistians have been found on most con- tinents with the exception of Antarctica and Aus- tralia. They are predominant in Europe (47 valid species and 17 incertae sedis), Africa (17 valid spe- cies and 6 incertae sedis), and North America (19 valid species and 5 incertae sedis); this probably reflects the relative bias of collecting rather than corresponding to some real biogeographic pat- terns.
Less than half (46%) of the actinistian taxa (cate- gories A-C) are known from entire skeletons var- ying in state of preservation. Of the remaining taxa, 36% are known almost exclusively from head or head fragments, 10% from scales solely, and 8% from partial trunk.
61
I = Category A
: = Category B i = Category C
Fig. 1. Diversity of actinistian species through geological time based on the listing of Table 1. Category A is represented by a straight bar, category B by a dotted line, and category C by an undulating line. Each bar corresponds to the number of species for a given biostratigraphic stage.
Patterns of diversity
Histograms of generic diversity through geological time have been provided by Schaeffer (1952a, 1953), Thomson (1977), andForey (1984,198). As pointed out by Smith & Patterson (1988), mono- typic and paraphyletic taxa are problematic in ge- neric-level studies because they provide historical- ly uninformative patterns. Among actinistians there is a high proportion of monotypic genera (see above) and others are known to be non-mono- phyletic (e.g., Rhabdoderma and Coelacanthus). However, temporal diversity studied at the specific level can be used to address historical questions (e.g., patterns and trends of species diversity).
Here we use Table 1 to provide some details of species diversity. We plot the number of species of categories A-C (above) against time (Fig. 1). The geological time scale of Palmer (1983) is used for the determination of age-boundaries and duration of geological units (i.e., age, period, and stage); in contrast to Palmer (1983), the Tertiary is divided into five epochs (Pliocene, Miocene, Oligocene,
Eocene, Paleocene; Holmes 1959) rather than 15 stages. The geological stage was used as the basic biostratigraphic unit for the count of taxa. In the diversity histograms, the bars were plotted accord- ing to the total diversity (i.e., number of taxa) occuring during a given stage. The bars were set on a proportional time axis (X-axis) at the mid-time between consecutive boundaries of each stage (i.e., [t&,1/2).
The proportional distribution of valid species (categories A and B) and known taxa (categories A-C; provided in parentheses) through geological era is as follows: 28.9% (33.05%) Paleozoic, 70% (66.12%) Mesozoic, 0% (0%) Cenozoic, and 1.1% (0.83%) Quaternary. There are four primary modes in the species diversity: (1) Frasnian with 6 species, (2) Namurian with 7 species (plus 2 in- certae sedis), (3) Scythian with 17 species (plus 4 incertae sedis), and (4) Tithonian with 10 species (plus 1 incerta sedis). There are two major geolog- ical gaps in the fossil record of the actinistians: (1) Middle Jurassic (gap of ca. 29 My) and (2) Tertiary (gap of ca. 66My).
62
Table 1. List of actinistian species with their respective geological, stratigraphic, geographical, and morphological information. Fm. = Formation, Gr. = Group, Mbr. = Member, Subgr. = Subgroup, Supergr. = Supergroup
‘Category A’ Alcoveria Beltan 1972 A. brevis Betlan 1972 (type species)
Middle Triassic; Ladinian. Muschelkalk. Spain.
Allenypterus Melton 1969 A. montanus Melton 1969 (type species)
Lower Carboniferous; Namurian A (E,B). Bear Gulch Limestone Mbr., Heath Fm., Big Snowy Gr. Montana, USA.
Axelia Stensiii 1921 A. robustu Stensio 1921 (type species)
Lower Triassic; Smitho-Spathian. Sticky Keep Fm., Sassendalen Gr. West Spitsbergen.
A. elegans Stensio 1921 Lower Triassic; Smitho-Spathian. Sticky Keep Fm., Sassendalen Gr. West Spitsbergen.
Axelrodichthys Maisey 1986 A. araripensis Maisey 1986 (type species)
Celacantideo of Richter 1985 Lower Permian. Irati Fm. Brazil.
Coelacanth bone of Orvig 1986 Palaeocene. Sweden.
Coelacanth remains of Gardiner 1966 Upper Devonian Alberta, Canada.
Coelacanthidae gen. et sp. indet. of Dziewa 1980 Early Triassic. Knocklofty Fm. Tasmania, Australia.
Coelacanthidae genus non det. of Woodward 1895 [‘Coelacanth’ of Schaeffer 19411 Upper Jurassic. Talbragar Beds. New South Wales, Australia.
Coelacanthinien genre indtt. of Casier 1961 Lower Cretaceous. Congo.
(not positively determinable)
(nomen nudum)
(nomen nudum)
(nomen nudum)
(nomen nudum)
(nomen nudum)
(nomen nudum, dipnoan Conchopoma gadiforme Kner 1868) (nomen nudum)
(nomen nudum)
(nomen nudum)
(not positively determinable)
(rhizodont)
(not positively determinable)
(not positively determinable)
(dipnoan)
(not positively determinable)
(actinopterygian)
(not positively determinable)
73
Trends of diversity
Schaeffer (1952b, p. 109) maintained that the orig- ination (speciation) rate of actinistians ‘. . . re- mained consistently low throughout their long his- tory’. Raup et al. (1973) have suggested that the pattern of actinistian species diversity is one which shows an initial burst of speciation followed by a drastic decrease in diversity ending in a single sur- viving species.
The variation of the diversity through the dura- tion of the group can be evaluated using non-para- metric Spearman’s rank-correlation coefficient. This coefficient is used to quantify the correlation between the number of taxa and the geological age. This is done in order to describe the trends of diversity through time. Rank-correlations were calculated for the two different units of time: geological period (rs period) and biostatigraphic stage (rs stage ). The rankings were attributed as follows: rank 1 equals Middle Devonian, rank 2 equals Late Devonian, rank 3 equals Early Carboniferous, . . . , and rank 20 equals Quaternary for the geological period; rank 1 equals Givetian, rank 2 equals Fras- nian, rank 3 equals Famennian, . . ., and rank 48 equals Holocene for the biostratigraphic stage. The number of species occurring in any one of the ge- ological periods and biostratigraphic stages are ranked numerically; for instance, the Early Triassic is the geological period and the Scythian is the biostratigraphic stage containing the most species and for the purpose of computation this is given the highest rank. For equal rankings of species num- bers, the computation procedure described by Dagnelie (1977) is invoked.
These rankings were then applied to the non- parametric Spearman’s rank-correlation coeffi- cient algorithm [see Cloutier (1991) for discussion of the method].
The history of the actinistians ranged over 20 (14 with actinistians) geological periods (from Middle Devonian to Quaternary) including 48 (31 with actinistians) biostratigraphic stages (from Givetian to Holocene) for a corresponding duration of 380 million years. The coefficients of rank-correlation calculated for the geological periods and the bio- stratigraphic stages show a low negative correlation
(for categories A and B: r, perjod = - 0.4459, Y s srage = - 0.2455; for categories A-C: r, pen’od = - 0.4944, r, stage = - 0.3551) between the diversi- ty-rank and time-rank. None of these correlations are significant. Therefore, these results support Schaeffer’s (1952b) conclusion that there has been no significant decreasing or increasing trends in species diversity through time,
Some of the peaks of actinistian diversity are coincident to major geological events. These events are not interpreted as causal effects on the diversity. They might have an influence on actual diversity because of vicariance events or habitat di- versification, or an effect on apparent diversity by increasing the likelihood of fossilisation owing to an increase of sedimentation, for example. The Frasnian peak succeeds the Caledonian orogeny which affected Europe and precedes the Acadian orogeny which affected North America; most of the Frasnian species are found in north-east Amer- ica and Europe. A second peak of species diversity, chiefly reflecting North American actinistian fau- na, occurred during the Namurian, stage associ- ated with the beginning of the Sudetian erogenic phase. The greatest diversity occurred during the Scythian which corresponds to a worldwide marine transgression (Schaeffer & Mangus 1976). The greatest actinistian diversity occurring during the Tithonian is documented mostly in Europe, and may be associated with the Late Kimmerian oro- genie phase. Finally, the Al&an peak may be broadly contemporaneous with the Subhercynian/ Austrian orogeny which influenced western Afri- ca; most of the Cretaceous actinistians are found in northern Africa and South America.
References cited
See ‘References cited’ in Forey & Cloutier (1991) for all taxo- nomic references.
Cloutier, R. 1991. Patterns, trends, and rates of evolution with- in the Actinistia. Env. Biol. Fish. 32: 23-58. (this volume)
Dagnelie, P. 1977. ThCorie et mkthodes statistiques. Applica- tions agronomiques. Vol. 1, 2nd ed. Presses Agronomiques de Gembloux, Gembloux. 378 pp.
Denison, R.H. 1978. Placodermi. pp. l-62. In: H.-P. Schultze
74
(ed.) Handbook of Paleoichthyology, Vol. 2.) Gustav. Fisch- er Verlag, Stuttgart.
Forey, P.L. 1991. Latimeria chalumnae and its pedigree. Env. Biol. Fish. 32: 75-97. (this volume)
Forey, P.L. & R. Cloutier. 1991. Literature relating to fossil coelacanths. Env. Biol. Fish. 32: 391-401. (this volume)
Holmes, A. 1959. A revised geological time scale. Trans. Edin- burgh Geol. Sot. 17: 183-216.
Lenoble, J. & Y. Le Grand. 1954. Le tapis de l’oeil du coela- canthe (Latimeria alzjouanae [Smith]). Bull. Mus. Hist. nat., Paris 26: 460-463.
Palmer, A.R. 1983. The decade of North American geology. 1983. Geologic time scale. Geology 11: 503-504.
Raup, D.M., S.J. Gould, T.J.M. Schoff & D.S. Simberloff. 1973. Stochastic models of phylogeny and the evolution of diversity. J. Geol. 81: 525-542.
Ride, W.D.L., C.W. Sabrosky, G. Bernardi & R.V. Melville (ed.). 1985. International Code of Zoological Nomenclature.
XX General Assembly of the International Union of Biolog- ical Sciences. International Trust for Zoological Nomencla- ture. H. Charlesworth & Co. Ltd., Huddersfield. 338 pp.
Schaeffer, B. & M. Mangus. 1976. An early Triassic fish as- semblage from British Columbia. Bull. Amer. Mus. Nat. Hist. 156: 517-563.
Smith, A.B. & C. Patterson. 1988. The influence of taxonomic method on the perception of patterns of evolution. Evol. Biol. 23: 127-216.
Thomson, K.S. 1977. The pattern of diversification among fish- es. pp. 377-404. In: A. Hallam (ed.) Patterns of Evolution, as Illustrated by the Fossil Record, Development in Palaeontol- ogy and Stratigraphy, Vol. 5, Elsevier, Amsterdam.
Vorobyeva, E.I. & D.V. Obruchev. 1967. Subclass Sarcoptery- gii. pp. 420-509. In: Y.A. Orlov (ed.) Fundamentals of Pale- ontology, Vol. XI Agnatha, Pisces, Israel Program for Scien- tific Translations, Jerusalem. (Russian original 1964).