New publication (Reassessment of historic ‘microsaurs’ from Joggins, Nova Scotia, reveals hidden diversity in the earliest amniote ecosystem; Mann et al., 2020, Papers in Palaeontology)

Title: Reassessment of historic ‘microsaurs’ from Joggins, Nova Scotia, reveals hidden diversity in the earliest amniote ecosystem
​Authors: A. Mann; B.M. Gee, J.D. Pardo, D. Marjanović, G.R. Adams, A.S. Calthorpe, H.C. Maddin, J.S. Anderson
Journal: ​Papers in Palaeontology
DOI to paper: 10.1002/spp2.1316

General summary: In what is my biggest collaboration to date with nearly all of the early tetrapod workers in Canada, here's probably my last paper of 2020: a very thorough revision of the 'microsaur' Asaphestera from Joggins, Nova Scotia! Joggins is one of the oldest sites with a well-preserved late Carboniferous tetrapod community (315-318 Ma), and it preserves a unique forest swamp environment in which animals are frequently preserved inside of logs (whether or not they lived in the logs is another question). Because of its age, Joggins is critical for exploring the early origins of many groups, as it often preserves the oldest members of many Paleozoic clades in North America, if not globally. For example, the temnospondyl Dendrerpeton from Joggins is the oldest known temnospondyl on the continent. There's a rich suite of 'microsaurs,' the ever controversial group of tetrapods of unclear affinities to either crown reptiles or to crown amphibians, that have been reported from Joggins, but many of the original studies are dated such that the interpretations are of limited utility for modern workers trying to elucidate phylogenetic relationships and other macroevolutionary trends. The frequent disarticulation and fragmentation of specimens also makes referrals of specimens to the same taxon somewhat suspect, with the potential for purportedly well-known taxa to be represented by a chimeric blend of multiple taxa. We spent a lot of time looking at historic specimens of one such 'microsaur,' Asaphestera, because we were particularly interested in exploring some of these very poorly known Carboniferous 'microsaurs' that are really important for understanding the evolution and relationships of the much better known early Permian taxa that everybody on this paper has studied to various degrees. Through this, we identified that the various specimens lumped under Asaphestera intermedia are actually a hodge-podge of different taxa, at least one of which is not even a 'microsaur,' but as we propose, is perhaps the earliest definitive synapsid! 

The oldest synapsid?

The oldest synapsid, extremely controversially so, is Protoclepsydrops, also reported from Joggins. As can be seen from Carroll's (1964) illustration of the holotype, the material, mostly postcranial in nature, is perhaps a little lacking for some people, which is why it's remained controversial as to whether it is in fact a synapsid. More unequivocal early synapsids include Archaeothyris and Echinerpeton, which are known from Florence, another locality in Nova Scotia that's a few million years younger than Joggins. You can find the original description of Archaeothyris ​by my advisor here, and some recent work on Echinerpeton by my friend and the leader author of this study, Arjan Mann, here and here. Long story short, like most other tetrapod groups, identifying unequivocal early members of most of these groups is very difficult. 

Holotype of Asaphestera platyris, which we refer to Synapsida here.

The specimen to the left, a former referred specimen of Asaphestera intermedia that was originally the holotype of a separate species called 'Asaphestera platyris​,' is what we're identifying as a less dubious synapsid and thus restoring as a valid taxon once again, but under a totally new identity. The specimen is admittedly not the nicest - these are the pancakes that we get at Joggins - but a few features indicate synapsid affinities that we listed in the paper: "large facial process of the maxilla, a single, narrowly triangular, small median postparietal flanked by the tabulars, which are restricted to the occipital surface of the skull, and substantial contact between the postorbital and parietal." 

Among synapsids, this specimen most closely resembles the eothyridids, although it shares a number of features with acleistorhinid parareptiles, which were often confused for eothyridids in their earlier history of study (perhaps they still are eothyridids?). Whether or not it could be the same taxon as Protoclepsydrops can't be resolved here because the newly resurrected Asaphestera platyris has no known postcrania at present. Asaphestera would have precedent though, having been named in 1934 versus Protoclepsydrops ​in 1964. 

We didn't intend to search for scant evidence of early synapsids, but surprising things can emerge from detailed surveys of early amniotes, most of which would look like small lizard-y things to the average person.

Microsaur math

So we've already turned part of one 'microsaur' into a synapsid at this point. Net: no change. But the holotype of Asaphestera platyris was not the holotype of Asaphestera intermedia, so the latter was still valid as a 'microsaur' at this point... 

The next specimen on the list is one that allegedly had three parts, all of different taxa, but only one was ever figured, the lectotype of what was named Hylerpeton intermedium way back in the day by Dawson (1894). Hylerpeton is its own distinct taxon (a possible gymnarthrid), with Hylerpeton dawsoni as the type species, so when Steen (1934) identified the differences from that species, she then created Asaphestera to accommodate 'Hylerpeton' intermedium. The species was still valid but not part of Hylerpeton, and the lectotype, among other specimens, were packaged into Asaphestera platyris. This is probably really confusing at this point. Seems like a good time to introduce the very helpful schematic created by one of the co-authors, Jason Pardo, below. The synapsid that we identify is specimen RM 2-1192. It was always the holotype of Asaphestera platyris, but not the holotype Asaphestera intermedia, which was the new combination that Carroll (1963) came up with to combine a bunch of new and old specimens, taking Dawson's species name (modified for gender) and Steen's genus name. 

Holotype of Hylerpeton intermedium, a nomen dubium after our analysis.

​RM 2-1131 was the original holotype of Hylerpeton intermedium and became the holotype of the new Asaphestera intermedia. Here it is above on the right. If the figure does not look that impressive...that's because the specimen is not very impressive. We couldn't identify it to a particular tetrapod clade, let alone diagnose it as a unique taxon of whatever clade that might be, and we designated it as a nomen dubium ("doubtful name"), which is the fancy way of saying "this is not valid because it is not clearly unique." So at this point, we've killed one 'microsaur.' Net: -1 microsaur

But there are a bunch of other specimens that were referred to Asaphestera intermedia (so not the holotype) by Carroll. Now if the holotype of a species is so incomplete that it can't be diagnosed, it usually doesn't matter if you think that you have a much better newer specimen. Even if the holotype was a rib, and you find a complete skeleton that you think is the same thing, the original holotype carries the name. If the holotype is suspect, then the name goes with it. There are instances where you can create "other types," such as neotypes, lectotypes, etc. Wikipedia has a surprisingly good summary (paleontologists don't use all the ones that are listed). Neotypes, for example, are necessitated when the holotype is destroyed and you have a good referred specimen on hand that can become the new bearer of the name. But usually if the holotype is bad, then it becomes  a nomen dubium (which has sometimes led to the loss of some otherwise very good names). 

Holotype of the new 'microsaur' taxon Steenerpeton silvae.

In any event, we then set about examining the referred specimens of "Asaphestera intermedia," at this point a sunk taxon, because some of them are much nicer. The one above was sufficiently complete to be diagnosed and differentiated from other taxa, and we made it a holotype of a new 'microsaur' taxon: Steenerpeton silvae, which we named for Margaret Steen, (the same Steen I keep mentioning throughout this post), and for 'of the forest,' referring to Joggins original paleoenvironment. So now we've added back one 'microsaur!' Net: +0 microsaurs

Steenerpeton appears fairly derived already, in contrast to traditional interpretations of Asaphestera of any flavour, and we suggest that it may be a recumbirostran, the derived group of 'microsaurs' that host a suite of fossorial adaptations. This actually lines up well with the documented presence of other derived 'microsaurs' like possible pantylids and gymnarthrids at Joggins and provides more evidence that the origin and diversification of recumbirostrans begin earlier than has been historically conceived and probably occurred in an interval without substantial body fossils. The former point in particular is a real running theme at this point for most of the research that all of us do. 

There were some other poorly known taxa that we wanted to get a look at because they have also been interpreted as relatively primitive 'microsaurs.' One, Ricnodon, is better known from Europe. Reports of Ricnodon at Joggins are very fragmentary and seem dubious. The taxon is still valid, but it doesn't clearly exist at Joggins. Net change: -1 'microsaur'

A second one is Archerpeton anthracos, a taxon named by Carroll in 1964. It was revised in 1996 by Reisz and Modesto who identified it as a 'microsaur,' rather than as a reptile; of course, with the hypothesis that 'microsaurs' are reptiles, this is  a moot point really. In any event, in spite of the nice illustrations of the holotype of Archerpeton, it doesn't preserve much helpful information. In fact, we couldn't diagnose it. The previous diagnosis listed a number of features that aren't truly unique to this taxon. Nomen dubium. Net change: -2 'microsaurs'

Holotype of the indeterminate tetrapod Archerpeton anthracos.

Avengers assemble!

This is the biggest collaboration I've ever worked on (some might say the most ambitious crossover ever). It was born nearly 2.5 years ago at the SVP Annual Meeting in Calgary, which is where Arjan and I both gave talks on different 'microsaurs' (Arjan's turned into his published paper on the new Mazon Creek 'microsaur' Diabloroter, and mine turned into my published paper on new material of Llistrofus from Richards Spur). Jason Pardo (and Jason Anderson) had just had a high-profile paper come out that summer that recovered 'microsaurs' as crown amniotes, a paper I didn't stop hearing about from my advisor even though I spent the whole summer doing fieldwork in Arizona. We all ended up talking about future directions for 'microsaur' work and agreed on needing to reevaluate a lot of the earliest records of what are called "tuditanomorphs" to better characterize them for big tetrapod analyses to continue to sort out different relationships. This in turn spawned what's become a dynamic working group among the three of us and eventually led to us jokingly assigning ourselves as different members of the Avengers as this team of early career researchers doing early tetrapod work on basically every single clade that's out there. If you add up everything that we've collectively published, we account for actinopterygian and sarcopterygian fish, aïstopods (stem tetrapods), temnospondyls, parareptiles, seymouriamorphs, embolomeres, synapsids, and 'microsaurs.' You can see some of the other fruits of this working group in Pardo & Mann (2018), Mann, Pardo & Maddin (2019) and Mann & Gee (2020), with plenty of future work slated for the immediate future (or once things sort of go back to normal)! I will leave it to the readers' imagination as to which members we have self-assigned as...

The final sum

The Joggins 'microsaurs' are the only taxa from Carboniferous localities that have experienced these sorts of "lump/split" boom-bust cycles, going from being remarkably taxonomically diverse to remarkably intraspecific variable and back again. The classic Joggins temnospondyl, Dendrerpeton, is another great example of this; the type species, D. acadianum, includes several defunct genera such as Smilerpeton (great name) and Dendryazousa. There are numerous taxa that have been named that weren't even assigned to the correct clade! A lot of this has been because of historic interpretations based on the often fragmentary and disarticulated specimens at Joggins; we don't get many 3-D specimens, and the distortions can affect interpretations of the anatomy. This means that there's a lot of room to go back and look at specimens that were often collected more than 120 years ago and that haven't been re-examined in detail in more than 60, 70, 80... years to assess the original interpretations. Particularly in light of new discoveries at other localities, updated taxonomy, and the recognition that a lot of clades have deeper origins than was often hypothesized in the early 20th century, exploring one of the oldest (crown) tetrapod assemblages in the world is essential for unraveling a lot of the mysteries of the origin of the modern tetrapod clades.

When we started this project, there were seven 'microsaurs' recognized from Joggins: Archerpeton, Asaphestera, Hylerpeton, Leiocephalikon, Ricnodon, and Trachystegos. We added Steenerpeton (total: 8), but deep-sixed Archerpeton, recognized Asaphestera proper as a synapsid, and cast doubt on the Joggins records of Ricnodon (final count: 5). We've got plans for some of those other ones that didn't get addressed here...stay tuned for plenty more studies coming out on Joggins material and other Carboniferous faunas! 

---

Refs

• Carroll RL. 1963. A microsaur from the Pennsyvanian of Joggins, Nova Scotia. Natural History Papers of the National Museum of Canada 22:1-13.
• Carroll RL. 1964. The earliest reptiles. Zoological Journal of the Linnean Society 45(304): 61-83. DOI: 10.1111/j.1096-3642.1964.tb00488.x
• Dawson JW. 1894. Preliminary note on recent discoveries of batrachians and other air-breathers in teh coal formation of Nova Scotia. The Canadian Record of Science 6:1-7.
  
• Reisz RR, Modesto SP. 1996. Archerpeton anthracos from the Joggins Formation of Nova Scotia: a microsaur, not a reptile. Canadian Journal of Earth Sciences 33(5): 703-709. DOI: 10.1139/e96-053
• ​Steen, MC. 1934. The amphibian fauna from the South Joggins. Nova Scotia. Proceedings of the Zoological Society of London 104(3): 465-504. DOI: 10.1111/j.1096-3642.1934.tb01644.x