Notochord never extends in front of brain; cranium present; vertebrae usually present; cartilage or bone or both present; heart chambered; red blood corpuscles usually present; brain well developed; 10 to 12 pairs of cranial nerves; dorsal and ventral nerve roots usually united; nephridia absent; epidermis with several cell layers; endostyle only in larval lampreys (ammo-coetes) and transformed into thyroid tissue in all others; sensory capsules present; neural crest formation present. The neural crest is a vertebrate innovation from which the first vertebrate skeletal tissue appears to have arisen (e.g., probably dermal bones, teeth, anterior neurocranium, and visceral arches). Maisey (2001a) reviewed the structure and function of the craniate inner ear and identified 33 apomorphic characters of the membranous labyrinth and associated structures in craniates, gnathostomes, and elasmo-branchs.
The classification followed here is based on the cladogram and classification in Donoghue et al. (2000). These authors sequence the following taxa, using their terminology, in a successive sister-group relationship (i.e., each taxon not in the parenthetical comments is sister to, or forms a cladistic node with, all those that follow): Cephalochordata, Myxinoidea (I adopt the name Myxinomorphi, in part to avoid using the ending for superfamilies), Petromyzontida (I adopt the name Petromyzontomorphi), Conodonta, Pteraspidomorphi (with Astraspis, Arandaspida, and Heterostraci sequenced in that order), Anaspida, Thelodonti (represented by Loganellia), Eriptychius and its sister group, the jawed vertebrates (together forming their "Unknown group B"), and their plesion, unnamed group C (herein termed the Osteostracomorphi, with Osteostraci [the best known], Galeaspida, and Pituriaspida). The position of Eriptychius is particularly uncertain; it is not considered as sister to the jawed vertebrates in the following discussion (see under Astraspida below). It therefore follows that the sister group of the Gnathostomata (jawed vertebrates) is the Osteostracomorphi (the combined taxon is unnamed). The group that is sister to the Cephalochordata (in the above, Myxinomorphi-Gnathostomata) is called the Craniata, while the sister group to the Myxinomorphi (Petromyzontomorphi-Gnathostomata) is the Vertebrata. The other nodes are unnamed, and in the sequence from Myxini to Osteostracomorphi, I have given these sequenced and named higher taxa the rank of superclass (i.e., the Myxinomorphi, Petromyzontomorphi, Conodonta, Pteraspidomorphi, Anaspida, Thelodonti, and Osteostracomorphi), the same as that of the Gnathostomata. The order in which the main taxa are presented in Janvier (1996) differs in modest detail and is as follows (no sequencing sister-group relationships for successive taxa are implied and the terminology of Donoghue et al., 2000, is used with Janvier's names, if different, in parentheses): Myxinoidea (Hyperotreti), Arandaspida, Astraspis (Astraspida), Heterostraci, Anaspida, Petromyzontida (Hyperoartia), Osteostraci, Galeaspida, Pituriaspida, and Loganellia (Thelodonti).
The classification used in the previous edition (Nelson,1994), shown immediately below, has thus been considerably changed. The terms Craniata and Vertebrata are no longer used as synonyms (as in Nelson, 1994:23), but are employed, conventionally, at different levels, with Craniata used at the sub-phylum level and Vertebrata as an unranked taxon within the Craniata.
Pteraspidiformes, and Thelodontiformes) class Cephalaspidomorphi (including the Petromyzontiformes, Anaspidiformes, Galeaspidiformes, and the Cephalaspidiformes (=Osteostraci) superclass Gnathostomata
One speculative view of the affinities and time of divergence of the major groups of fishes. The approximate age between boundaries of periods is given in millions of years (based on The Geologic Time Scale, 2001, U.S. Geological Survey, Lyn Topinka). The Pennsylvanian and Mississippian (distinct in North America) are together equivalent to the Carboniferous outside North America. The Tertiary is subdivided into the Paleocene, Eocene, Oligocene, Miocene, and Pliocene. Fossils are often dated within the Cretaceous to the following ages in the Late Cretaceous as (oldest to youngest) Cenomanian, Turonian, Coniacian, Santonian, Campanian, and Maastrichtian, and in the Early Cretaceous (from the boundary of the Tithonian of the Jurassic) as Berriasian, Valanginian, Hauterivian, Barremian, Aptian, and Albian (borders the Cenomanian). The terms Late and Early refer to ages, whereas the terms Upper and Lower refer to stratigraphy.
Are agnathans (jawless fishes) monophyletic? This is a bothersome question that concerns one of several major conflicts in fish phylogeny between the results of molecular biology and morphological studies. The term "Agnatha" is no longer used as a taxon name, as it was in the 1994 edition; it applied in a conventional sense to all taxa from Myxini to Pituriaspida, recognized now as a paraphyletic group. However, the term "agnathan" is still a useful one that can be used to describe these jawless fishes. Two groups of jawless fishes that are also considered here to be paraphyletic groups are the cyclostomes and the ostracoderms. The term "cyclostome" is used for the livingjawless fishes (hag-fishes and lampreys); this group is considered by most paleontologists and others using morphological evidence to be a paraphyletic group, and is so recognized here. However, there is molecular evidence from Mallatt and Sullivan (1998), Mallatt et al. (2001), Kuraku et al. (1999), Delarbre et al. (2002), and Takezaki et al. (2003) that supports the monophyly of cyclostomes (an older idea, termed the "cyclostome hypothesis," and rejecting the hypothesis that lampreys are more closely related to gnathostomes than to hagfishes, termed the "vertebrate hypothesis"). This is a serious conflict with the phylogenetic ideas accepted here and one that must be resolved, ideally by obtaining independent results that are in agreement from both molecular biology and morphological studies, before we can be confident in the basic phylogeny of early craniates. Meyer and Zardoya (2003) suggested the desirability of having larger data sets with greater taxon sampling to better support either the cyclostome hypothesis or the vertebrate (lamprey-gnathostome) hypothesis. It is interesting, though, that lampreys are now placed in the next sequenced group up from hagfishes following Donoghue et al. (2000), rather than in the class Cephalaspidomorphi, and sister to the Anaspidiformes. This suggests that they could have diverged within a relatively short time span. The term "ostracoderm" is used for the fossil armored jawless fishes; this is agreed to be a paraphyletic group but phylogenetically closer to the jawed vertebrates than to either hagfishes or lampreys. Forey (1995) reviewed past theories of relationships of agnathans and gnathostomes and of character evolution.
The paraphyletic jawless fishes (agnathans) are characterized by the following characters: jaws that are derived from gill arches absent (a biting apparatus, not derived from gill arches, is present in some fossil forms and in hag-fishes); no pelvic fins; one or two vertical semicircular canals (one canal but two ampullae reported in myxiniforms, at least two in pteraspidiforms); vertebral centra never present (only the notochord); gills covered with endo-derm and directed internally; gill arch skeleton fused with neurocranium, external to gill lamellae; gills opening to surface through pores rather than through slits; bony exoskeleton in most.
There are about 17 genera and 108 extant species of extant jawless fishes in four families. The three major clades of craniates with living or extant species— hagfishes, lampreys, and gnathostomes—have a total of about 54,711 species.
Was this article helpful?