You Little Devil

Skeleton of Scleromochlus taylori in multiple views.

A gratuitous helping of specimens from the Lossiemouth Sandstone Formation (?Carnian, Late Triassic) of Lossiemouth, near Elgin, northeastern Scotland, reveals nearly the entire skeleton of Scleromochlus taylori. Named by Alfred Woodward in 1907, it has languished despite the briefest of interested attentions due to the fact that the specimens are all natural casts, and none represent true bone. Moreover, they are so small and the grain size of the beds they are preserved in so large that it is difficult, if not impossible, to discern fine features. Above, I have attempted to compose the skeletons into four views, based on the excellent multi-frame preservation available (in dorsal, ventral, and lateral sides). Extrapolations of various elements also afforded the ability to produce a caudal view, although this does not necessarily derive from a particular specimen.

Dermal ossicles were preserved along the dorsal midline and the ventral midline, forming narrow and thin plates above the spine and broad, long plates below the belly and essentially covering the gastralia. In the above skeleton, I have omitted these dermal ossifications, and because of them chose not to depict all possible gastralia, as I could not infer their potential to change shape as they came near the pelvis. This is important because Scleromochlus taylori features in the evolutionary relationships of pterosaurs and dinosaurs, and is something of an ur-ornithodiran (Benton, 1999), providing further evidence of a link between common ancestry of dinosaurs and pterosaurs within archosaurs and exclusive of crocodilians or lacertilians. Thus, they are stem-avian reptiles, as arrayed on the grand reptilian crown tree:

Crown groups in "modern" Reptilia (rather than using the "classic" definition for only cold-blooded animals, exclusive of Aves). Reptilia includes all non-mammalian amniote crowns, and it might be best to simply qualify it as such. A represents the basic crown tree, while B shows the relative positions of Pterosauria, Scleromochlus, and the clade Ornithodira which contains Pterosauria and Dinosauria (here depicted as the lineage including the stem of crown-Aves). Alternative topologies are not shown, and the relative position of Testudines is not debated, but may be more basal, or less basal, than Lepidosauria (which itself may be paraphyletic).

Note that the phylogeny above is not based on analysis, but is horribly generalized for the sake of explanation. The taxa used are considered crown-based clades (defined as the most recent common ancestor of all living members and their descendants). This means Lepidosauria includes lizards, snakes, tuataras, mosasaurs, etc.; Testudines includes turtles; Crocodilia includes crocs, sphenosuchians, aetosaurs, phytosaurs, rauisuchians, notosuchians, and a host of other forms; and Aves includes only the birds. These are then useful to define nodes for the stems, where Archosauria is a node-based clade containing Aves and Crocodilia, and Reptilia contains Lepidosauria, Testudines, Crocodilia and Aves (“classically,” it contains only the lizards, snakes, turtles, and crocs, but as these form a monophyletic clade including birds, it would even if formulated in a restricted manner, by extension include Aves).

The Crowning Moment

I should further qualify that these are effectively my use-definitions for these names, and I suspect those I’d prefer as they reflect functional and historical use in the broad sense. While classical definitions have varied, especially in the use of Reptilia, long-standing usage of Reptilia implies it may continue to be useful. The alternative is to use taxa such as Sauropsida which lack the “baggage” of Reptilia, and so on, but this is uncommon and typically only of a relatively small set of researchers. For the purposes of communication, a gradual enlightening of the principle of phylogeny and the application of nomenclature using the three “basic” types of clade (stem, node, crown). Useage of explicit species- and specimen-based definitions would require picking ideal representatives for the major groups, and several have suggested or asserted that the earliest taxa named by Carolus Linnaeus now contained in each would be preferred, and I agree with this. Crown-clades may be qualified on these grounds with stem-based definitions excluding each other representative of the other crowns (so each of the crowns shown above would have one internal and three external specifiers), and the clades that are internal among them (e.g., Archosauria) would contain variations of these four taxa (two internal, two external, in this case), and so on. Thus, a name for the crown containing birds, crocs and turtles can be useful, excluding only lizards/snakes, even if not supported, may become necessary given recent studies that have placed Testudines closer to archosaurs than Lepidosauria; the inverse would also be true for the position of turtles outside of a lizard(croc+bird) arrangement.

Similarly, I have no taste for the Pan-[name] convention that has been otherwise offered, since even the use of nomenclature like “Pan-Testudines” with any form of mandatory structure (for crown clades) (PhyloCode, Art.10.3, note use of form “total” clade for the classic phrasing “crown clade”), becomes a step into the direction of special taxonomic categories, in the sense of ranks; that it, types of clades with preferred hierarchical relationships. It lacks imagination, and provides that those names that can build upon it or diverge from it linguistically, rather than just putting affices on the front or back, will be used less. A definition for a clade gives it its structure, and we shouldn’t have to look at a name to know which type it is: going there makes we wonder why we try to leave Linnaean Systematics in the first place. This is a slippery slope argument, and we should beware it.

It’s a devil of a detail, but one I think many people who are wary of the PhyloCode see as a problem, while also eschewing Linnaean Systematics: That in our rejection of an inadequate and non-functional system of nomenclature, we are trending to replace it with yet another. In the end, the PhyloCode merely eschews recognition of ranks, the ICZN/ICBN only support them up to the Family (Familia), noneukaryotic taxa might as well be form taxa when it comes to applying any sort of mandatory clade structure to them (to say nothing of form taxa in the form of parataxonomy in ichnites and other trace fossils [ichnotaxonomy] and eggs [ootaxonomy]). We are still free to play Linnaeus’ drawer game, or using John Wilkin’s analogy, a toy collector. The value is in collecting definitions, or associating taxa by their definitions, rather than treating taxa as equivalent regardless of content (ostensibly the objective of eschewing Linnaean Systematics), and thus we are still pretending that non-species taxa have some extrinsic value depending on what kind they are.

Scleromochlus, for me, is the devil in this machine (or gremlin, to be consistent with the analogy), as it helped push the problem of Reptilia around, dismantling core beliefs; I tend to de-perch the urvogel, Archaeopteryx, as an historical form leading up to the issue. I think the question of the origin of birds is settled as far as phylogenetics are concerned, but when it comes to Scleromochlus taylori, there is still a lot of play to be had.

Benton, M. J. 1999. Scleromochlus taylori and the origin of the pterosaurs. Philosophical Transactions of the Royal Society B: Biological Sciences 354:1423–1446.
Woodward, A. S. 1907. On a new dinosaurian reptile (Scleromochlus taylori, gen. et sp. nov.) from the Trias of Lossiemouth, Elgin. Quarterly Journal of the Geological Society 63:140–144.