Huxley and the fossil gaps

The Berlin specimen of Archaeopteryx lithographica, a historically important fossil which helped to establish birds as a component of the reptile family tree

The term Sauropsida ("lizard faces") has a long history, and hails back to Thomas Henry Huxley, and his opinion that birds had risen from the dinosaurs. He based this chiefly on the fossils of Hesperornis and Archaeopteryx, that were starting to become known at the time.^[3] In the Hunterian lectures delivered at the Royal College of Surgeons in 1863, Huxley grouped the vertebrate classes informally into mammals, sauroids, and ichthyoids (the latter containing the anamniotes), based on the gaps in physiological traits and lack of transitional fossils that seemed to exist between the three groups. Early in the following year he proposed the names Sauropsida and Ichthyopsida for the two latter.^[2] Huxley did however include groups on the mammalian line (synapsids) like Dicynodon among the sauropsids. Thus, under the original definition, Sauropsida contained not only the groups usually associated with it today, but also several groups that today are known to be in the mammalian side of the tree.^[4]

Sauropsids redefined (Goodrich, 1916)

By the early 20th century, the fossils of Permian synapsids from South Africa had become well known, allowing palaeontologists to trace synapsid evolution in much greater detail. The term Sauropsida was taken up by E.S. Goodrich in 1916 much like Huxley's, to include lizards, birds and their relatives. He distinguished them from mammals and their extinct relatives, which he included in the sister group Theropsida (now usually replaced with the name Synapsida). Goodrich's classification thus differs somewhat from Huxley's, in which the non-mammalian synapsids (or at least the dicynodontians) fell under the sauropsids. Goodrich supported this division by the nature of the hearts and blood vessels in each group, and other features such as the structure of the forebrain. According to Goodrich, both lineages evolved from an earlier stem group, the Protosauria ("first lizards"), which included some Paleozoic amphibians as well as early reptiles predating the sauropsid/synapsid split (and thus not true sauropsids). His concept differed from modern classifications in that he considered a modified fifth metatarsal to be an apomorphy of the group, leading him to place Sauropterygia, Mesosauria and possibly Ichthyosauria and Araeoscelida in the Theropsida. ^[4]

Detailing the reptile family tree

In 1956, D.M.S. Watson observed that sauropsids and synapsids diverged very early in the reptilian evolutionary history, and so he divided Goodrich's Protosauria between the two groups. He also reinterpreted the Sauropsida and Theropsida to exclude birds and mammals respectively, making them paraphyletic, unlike Goodrich's definition. Thus his Sauropsida included Procolophonia, Eosuchia, Protorosauria, Millerosauria, Chelonia (turtles), Squamata (lizards and snakes), Rhynchocephalia, Rhynchosauria, Choristodera, Thalattosauria, Crocodilia, "thecodonts" (paraphyletic basal Archosauria), non-avian dinosaurs, pterosaurs and sauropyterygians. However, his concept differed from the modern one in that reptiles without an otic notch, such as araeoscelids and captorhinids, were believed to be theropsids. ^[5]

This classification supplemented, but was never as popular as, the classification of the reptiles (according to Romer's classic Vertebrate Paleontology^[6]) into four subclasses according to the positioning of temporal fenestrae, openings in the sides of the skull behind the eyes. Since the advent of phylogenetic nomenclature, the term Reptilia has fallen out of favor with many taxonomists, who have used Sauropsida in its place to include a monophyletic group containing the traditional reptiles and the birds.

Cladistic definitions

Sauropsida and the 19th-20th century conception of the class Reptilia. Both are superimposed on a cladogram of Tetrapods, showing the difference in coverage.

The class Reptilia has been known to be an evolutionary grade rather than a clade for as long as evolution has been recognised. Reclassifying reptiles has been among the key aims of phylogenetic nomenclature.^[7] The term Sauropsida had from the mid 20th century been used to denote a branch-based clade containing all amniote species which are not on the synapsid side of the split between reptiles and mammals. This group encompasses all now-living reptiles as well as birds, and as such is comparable to Goodrich's classification. The main difference is that better resolution of the early amniote tree has split up most of Goodrich's "Protosauria", though definitions of Sauropsida essentially identical to Huxley's (i.e. including the mammal-like reptiles) are also forwarded.^[8][9] Some later cladistic work has used Sauropsida more restrictively, to signify the crown group, i.e. all descendants of the last common ancestor of extant reptiles and birds. A number of phylogenetic stem, node and crown definitions have been published, anchored in a variety of fossil and extant organisms, thus there is currently no consensus of the actual definition (and thus content) of Sauropsida as a phylogenetic unit.^[10]

Some taxonomists, such as Benton (2004), have co-opted the term to fit into traditional rank-based classifications, making Sauropsida and Synapsida class-level taxa to replace the traditional Class Reptilia, while Modesto and Anderson (2004), using the PhyloCode standard, have suggested replacing the name Sauropsida with their redefinition of Reptilia, arguing that the latter is by far better known and should have priority.^[10]

Cladistic definitions of Sauropsida include:

• Sauropsida as the total group of reptiles: “Reptiles plus all other amniotes more closely related to them than they are to mammals” (Gauthier, 1994).^[1] This is a branch-based total group definition. Gauthier (1994) considered turtles to be descended from parareptiles, thus defining Reptilia as a more restricted crown group encompassing diapsids and parareptiles (apart from mesosaurs, which he considered to be the most basal branch of sauropsids).
  • Sauropsida as a total group, synonymous with Reptilia sensu lato: “The most inclusive clade containing Lacerta agilis and Crocodylus niloticus, but not Homo sapiens” (Modesto & Anderson, 2004).^[10] This total group definition leaves the question of turtle ancestry unresolved.
• Sauropsida as a broad node-based group: “The last common ancestor of mesosaurs, testudines and diapsids, and all its descendants” (Laurin & Reisz, 1995).^[11] Though formulated differently, this grouping was similar in scope and intention to the definition provided by Gauthier (1994).

Thermal and secretion

The early synapsids inherited abundant glands on their skins from their amphibian ancestors. Those glands evolved into sweat glands in synapsids, which granted them the ability to maintain constant body temperature but made them unable to save water from evaporation. Moreover, the way synapsids discharge nitrogenous waste is through urea, which is toxic and must be dissolved in water to be secreted. Unfortunately, the upcoming Permian and Triassic periods were arid periods. As a result, only a small percent of early synapsids survived in the land from South Africa to Antarctica in today's geography. Unlike synapsids, sauropsids do not have those glands on the skin; their way of nitrogenous waste emission is through uric acid which does not require water and can be excreted with feces. As a result, sauropsids were able to expand to all environments and reach their pinnacle. Even today, most vertebrates that live in arid environments are sauropsids, snakes and desert lizards for example.

Brain structure

Different from how synapsids have their cortex in six different layers of neurons which is called neocortex, the cerebrum of Sauropsida has a completely different structure. For the corresponding structure of the cerebrum in the classic view, the neocortex of synapsids is homology with only the Archicortex of the avian brain. However, in the modern view appeared since the 1960s, behavioral studies suggested that avian neostriatum and hyperstriatum can receive signals of vision, hearing, and body sensations, which means they act just like the neocortex. Comparing an avian brain to that to a mammal, nuclear-to-layered hypothesis proposed by Karten (1969), suggested that the cells which form layers in synapsids' neocortex, gather individually by type and form several nuclei. For synapsids, when one new function is adapted in evolution it will be assigned to a separate area of cortex, so for each function, synapsids will have to develop a separate area of cortex, and damage to that specific cortex may cause disability.^[23] However, for Sauropsida functions are disassembled and assigned to all nuclei. In this case, brain function is highly flexible for Sauropsida, even with a small brain, many Sauropsida can still have a relatively high intelligence compared to mammals, for example, Corvidae. So, it is possible that dinosaurs, like T-rex which we know have tiny brains compared to their enormous body size, are much more intelligent than what people have imagined.^[24]