Diapsids are by far the most speciose group of amniotes, with about 14 600 extant species (Goin et al., 1978). They have invaded all major habitats, from the polar circles (many migratory birds) to deserts (many lizards) and even the ocean (sea snakes, sauropterygians). Diapsids include most flying vertebrates (birds) and most poisonous chordates (snakes and the Gila Monster).
The early history of diapsids is poorly documented. Until the late seventies, the oldest known diapsids were the Upper Permian (250 Myr old) younginiforms from South Africa and Madagascar (Harris and Carroll, 1977; Currie, 1980, 1981, and 1982), and a few other contemporaneous diapsids of uncertain affinities (Carroll, 1976a and b). However, recent work (Reisz, 1977) has extended the fossil record of diapsids to the Pennsylvanian (about 300 Myr ago), and greatly increased our knowledge of the diversity of early diapsids (Brinkman et al., 1984; Reisz et al., 1984; Laurin, 1991; deBraga and Reisz, 1995). The oldest known crown-diapsids (saurians) date from the Late Upper Permian (Carroll, 1975; Evans, 1987).
Extant diapsids are classified into either lepidosaurs (lizards and Sphenodon) or archosaurs (birds and crocodiles). Both of these clades are very successful and speciose (Fig. 1), and archosaurs include some of the most fascinating vertebrates that ever lived, such as the pterosaurs (flying reptiles of the Mesozoic) and the many extinct groups of dinosaurs. Indeed, the present diversity of archosaurs, even though it compares favourably with many other clades, is a mere shadow of what it was in the mesozoic.
Diapsida was named after the two fenestrae (holes) found in the temporal region of the skull of most early and some extant diapsids. The lower temporal fenestra is between the jugal, postorbital, squamosal, and quadratojugal. The upper temporal fenestra is between the postorbital, parietal, and squamosal. Some diapsids have lost the lower fenestra (lizards) or even both fenestrae (snakes, amphisbaenids), but their early ancestors had both fenestrae.
Despite numerous studies on diapsid phylogeny and classification, diapsids taxonomy still suffers from a lack of consensus. However, according to the principles of priority suggested by de Queiroz and Gauthier (1990, 1992, 1994), several diapsid taxa have been formally defined phylogenetically, and these definitions should be respected if diapsid taxonomy is ever to be standardized. These taxa include:
Diapsida: The most recent common ancestor of araeoscelidians, lepidosaurs, and archosaurs, and all its descendants (Laurin, 1991).
Araeoscelidia: The most recent common ancestor of Araeoscelis and Petrolacosaurus, and all its descendants (Laurin, 1991).
Neodiapsida: Sauria plus all other diapsids that are closer to saurians than they are to araeoscelidians (Gauthier et al., 1988b).
Eosuchia: The most recent common ancestor of Coelurosauravus, Apsisaurus, younginiforms, lepidosaurs, and archosaurs and all its descendants (Laurin, 1991).
Younginiformes: The most recent common ancestor of Youngina, Acerosodontosaurus, and Hovasaurus and all its descendants (Laurin, 1991).
Sauria: All the descendants of the most recent common ancestor of birds, crocodiles, squamates, and Sphenodon (Gauthier, 1984; Gauthier et al., 1988b).
Lepidosauromorpha: Extant lepidosaurs and all extinct saurians that are closer to them than they are to extant archosaurs (Gauthier, 1984; Gauthier et al., 1988a).
Archosauromorpha: Extant archosaurs and all extinct saurians that are closer to them than they are to extant lepidosaurs (Gauthier, 1984, 1994).
Some of these taxa have been given multiple definitions, and some of these definitions have been applied to more than one taxon, but the definitions of taxa given above appear to have priority (de Queiroz and Gauthier, 1990, 1992, and 1994).
This classification, as well as most recent classifications of diapsids, does not recognize Lacertilia as a formal taxon (Lacertilia included lizards but not snakes and amphisbaenids) because this group is not monophyletic (it does not include all the descendants of a common ancestor). Indeed, snakes (Estes et al., 1988; Rieppel, 1988) and amphisbaenids (Wu et al., 1993) are the direct descendants of some lizards, so in this page, the word "lizards" includes these taxa as well.
Only skeletal autapomorphies (unique derived characters) can be confidently assigned to Diapsida because all the basal lineages of diapsids are extinct. In addition to the presence of an upper and a lower lateral temporal fenestra, these include:
A suborbital fenestra (Fig. 2). This is a relatively large hole in the palate that is located between the palatine, the ectopterygoid, and the maxilla. Other taxa may have a foramen (a small hole just large enough to allow passage for a nerve or a blood vessel) in this region, but there is usually no fenestra.
A long radius, measuring between 70% and 90% of the length of the humerus. This ratio is variable in extant diapsids, but early diapsids consistently had a longer radius than other taxa, in which it was typically less than 70% of the humeral length.
Skeletal autapomorphies can unambiguously be attributed to Sauria, and other characters (soft anatomical, physiological, etc.) present in saurians but not in other extant amniotes are also usually attributed to Sauria, although they may have been present in other diapsids and even in their close relatives ("Protorothyridids" and captorhinids). Saurian autapomorphies include:
A low concentration of urea in the blood plasma resulting from a loss or suppression of the urea cycle. Lissamphibians, mammals, and turtles have a higher concentration of urea in their plasma.
The loss of the apposition between the kidney and the adrenal gland. These two organs are juxtaposed in lissamphibians, mammals, and turtles.
The presence of Huxley's foramen. This foramen is a hole in the distal end of the extracolumella (a cartilaginous extension of the stapes that contacts the tympanum). Huxley's foramen is absent in lissamphibians, mammals, and turtles.
Temporal muscles originating on the dorsolateral surface of the skull table. In araeoscelidians and younginiforms, the temporal muscles originate from a fascia attached to the lateral edge of the skull table and from the ventral surface of the skull table.
Prefrontal-nasal suture anterolaterally oriented (Fig. 3D). This suture is parasagittal in araeoscelidians, Claudiosaurus, and younginiforms (Fig. 3A-C).
Squamosal confined to the dorsal half of the skull, except for a narrow ventral process supporting the quadrate. The squamosal of younginiforms, Apsisaurus, and araeoscelidians is broad ventrally.
Strong, broad contact between the paroccipital process and the cheek. This contact is weak and often cartilaginous in younginiforms and araeoscelidians.
Quadrate deeply emarginated posteriorly. The quadrate of saurians supports a tympanum in its deep posterior emargination. The quadrate of younginiforms has a very shallow emargination that probably did not support a tympanum. The quadrate of Apsisaurus and araeoscelidians is not emarginated.
Slender stapes. The stapes of saurians is modified to function as a middle ear ossicle to transmit high-frequency air-borne sounds from the tympanum to the inner ear. The stapes of younginiforms and araeoscelidians is more massive and it is not specialized as a middle ear ossicle.
Dorsal process of stapes with ossified connection to paroccipital process of opisthotic. In other taxa, this connection is cartilaginous (or it may consist of a tendon), when it is present.
Large retroarticular process. This is the insertion point for the muscles that open the lower jaw. The retroarticular process of araeoscelidians, Coelurosauravus, and younginiforms is much smaller.
Cleithrum absent. The cleithrum is a dermal bone located on the anterior edge of the scapula, dorsal to the clavicle. It is present in araeoscelidians, Coelurosauravus, and younginiforms.
Lateral manual centrale (a bone in the wrist) small or absent. The lateral and medial centralia are approximately of equal size in araeoscelidians and in some younginiforms primitively (in Acerosodontosaurus).
Fifth distal tarsal absent (this is a small bone in the ankle, proximal to the fifth toe). Araeoscelidians, Coelurosauravus, and youngina have five distal tarsals.
Fifth metatarsal hooked (this bone supports the fifth toe). This bone is straight in other diapsids.
Females larger in most turtles and snakes, but males larger in most lizards and crododilians; extreme Sexual Dimorphism in larger pythons where females may weigh more than 10 times as much as males; Sexual Dimorphism also in body shape, tail length, head size, color, scales, and crests.
Please note that some authors place the turtles within the Diapsida. Please refer to the accessory page on Diapsid Phylogeny and to the Discussion of Phylogenetic Relationships on the Amniota page for more information on this issue.
Diapsid phylogeny has been intensively studied in the last decade, but the affinities of many groups are still controversial. For instance, the affinities of the Upper Permian diapsids Galesphyrus (Carroll, 1976a), Heleosaurus (Carroll, 1976b), and Heleosuchus (Carroll, 1987) from South Africa are uncertain because these taxa are represented by fragmentary remains. However, the relationships of the other major groups of stem-diapsids are now reasonably well understood. The affinities of araeoscelidians were initially debated (Peabody, 1952; Vaughn, 1955; Reisz, 1977), but they are now universally believed to be the earliest known group of diapsids (Laurin, 1991; deBraga and Reisz, 1995). Younginiforms were previously believed to represent early lepidosauromorphs (Benton, 1985; Evans, 1988; Gauthier et al., 1988a), but they are now considered to be close relatives of Sauria (Laurin, 1991; Gauthier, 1994).
Some diapsid taxa like Ichthyosauria, Sauropterygia, Placodontia, and Choristodera remain problematic. Placodontia and Choristodera do not appear on the main tree because recent work suggests that they are either lepidosauromorphs or archosauromorphs (Rieppel, 1993, 1994; Rieppel and deBraga, 1996; deBraga and Rieppel, 1997; Merck, 1997). Only ichthyosaurs and sauropterygians appear because they may be either stem-diapsids, archosauromorphs, or lepidosauromorphs (Caldwell, 1996; Mercki, 1997; Motani et al., 1998). Turtles may also be saurians, either lepidosauromorphs or archosauromorphs, but for a discussion of this topic, please refer to the page on Amniota. For more information, see the linked page on Diapsid phylogeny.
Barcode of Life Data Systems (BOLD) Stats Specimen Records:12397 Specimens with Sequences:11664 Specimens with Barcodes:9341 Species:2405 Species With Barcodes:2071 Public Records:7423 Public Species:1134 Public BINs:2076