Exceptionally, this post is not about a reptile or a reptile ancestor,
but it’s about a fascinating Carboniferous taxon that continues to defy identity and affinity, Tullimonstrum gregarium (Richardson 1966; Early Carboniferous, Mazon Creek Formation; up to 35 cm in length; Fig. 1), popularly known as the ‘Tully monster.’

A recent paper
by Sallan et al. 2017 argued against the vertebrate affinities of Tullimonstrum earlier advanced by Clements et al. 2016 and McCoy et al. 2016. They both provided evidence that Tullimonstrum was a relative of lampreys or jawed fishes. The so-called ‘mouth’ was considered the single olfactory opening and the terminal pincers were considered jaws in their hypothesis. Gill pores were also identified.  That all made sense to the editors and referees at Nature, but Sallan et al. 2017 rallied against those arguments. Their instructive PDF is online here.

Figure 1. GIF movie of a Tullimonstrum specimen with matrix removed and specimen unfolded, traced and compared to the extant flatworm, Stenostomum.

As usual,
I knew next to nothing about Tullimonstrum before sitting down with it today. Sallan et al., did not advance their own hypothesis identifying the Tully monster. They simply knocked down the arguments advanced by Clements et al. and McCoy et al. for vertebrate affinities. I worked from online photographs, Wikipedia (link below) and the Sallan et al. paper.

Let’s start with some observations

1. Tullimonstrum can be easily folded (Fig. 1), so  it was probably flat, rather than plump, and soft-bodied throughout, rather than stiffened with a central notochord
2. Very few invertebrates are bilateral, have eye spots (refractive bodies), an elongate anterior and a ventral (non-terminal) mouth. However, Stenostomum (Fig. 1), an extant freshwater flatworm, is one such invertebrate.
3. The odd appearance and short duration of Tullimonstrum suggest a very specialized niche from a clade that doesn’t usually fossilize

Then let’s finish with some fresh speculations

1. Like Stenostomum, Tullimonstrum could have been a fresh water ectoparasite on larger Carboniferous vertebrates and/or invertebrates, finding them with its eyespots, swimming to them with its fins, attaching to them with its anterior gripper and feeding on whatever they fed on with its ventral mouth.
2. Perhaps Tullimonstrum preyed on clades that became extinct shortly after the Mazon Creek formation and went extinct with them
3. The flat body could have been appressed to the surface of its host, adding little drag to its host’s swimming speed.
4. Those little circles that dot the torso of Tullimonstrum could be homologous to the little circles that dot the torso of Stenostomum: egg cells.

Added August 27, 2020
The tiny annelid worm Stylaria lacustris (Fig. 2) has traits similar to those found in Tullimonstrom. The mouth and anus are terminal, distinct from Tullimonstrum. No eyes are present.

Figure 2. The extant flat annelid worm, Stylaria lacustris, has a slender proboscis and body segments similar to that of extinct Tullimonstrum.

References
Clements T et al. (5 other authors) 2016. The eyes of Tullimonstrum reveal a vertebrate affinity. Nature 532:500-503.
McCoy et al. (15 other authors) 2016. The ‘Tully monster’ is a vertebrate. Nature 352:496-499.
Richardson ES Jr 1966. Wormlike fossil from the Pennsylvanian of Illinois. Science 151:75-76.
Sallan L et al. (six other authors) 2017. The ‘Tully monster’ is not a vertebrate: characters, convergence and taphonomy in Palaeozoic problematic animals. Palaeontology 2017:1–9.

https://en.wikipedia.org/wiki/Tullimonstrum