TNT command line tutorial

This post is as much for me as others, this should go over nearly everything you could need to do to run a simple parsimony analysis on the command line version of TNT. I am writing this as I install and open TNT for the second time, just to ensure I complete every step. Though I am using a Mac for this, it could also be used to help with the command line versions on other systems.

TNT is free to download from, choosing whatever computer you are using. What you do now is open the folder, and move it to wherever you want it to be. Double click on tnt.command. It should say that TNT is from an unidentified developer. To bypass this simply Control select it, and choose Open. If this is your first time, it will likely say it doesn’t have an application to open on, for that choose Terminal.

This next step was bypassed for me because I have already opened TNT before, so this is off what others have already said online. Now, type in cd (wherever you placed the tent folder), so that Terminal can find the tnt.command document. Now, type in ./tnt.command, so Terminal will open the TNT command script. It will probably appear with a large page showing the Terms and Conditions, and it tells you to type something (I believe it says type yes), follow its instructions, and wait a moment for the next page of T&C to appear, and follow its instructions until you get to a page where the cursor is following tnt*>. If it has already opened the Terms and Conditions, skip the above step and simply type yes to progress. This is good so far.

Now all you need is a dataset. Making a dataset is fairly easy, all it needs to be for TNT to run it is a nexus file. To find one online, simply google Nexus file, or to make your own I would recommend the programs Paup* or Mesquite.

Some settings might need to be changed depending on the size of the matrix. For something larger, type mxram 200 to a larger memory, states num for numerical matrices, and nstates NOGAPS so gaps aren’t treated as another state (it seems the second command has been deprecated, but doesn’t affect processing).

To run the dataset in TNT, the file needs to be in the same location as the tnt.command executable. In Terminal, type proc (nexus filename). The file name has to be exactly what you see or you will get an error message. If your file is in the same location, and TNT doesn’t read it, recheck that the current directory (cd) is pointing to the correct location, the folder both tnt.command and the matrix are in. This can be done by typing cd and returning, and then typing anything when cdir> appears.

Now it may be best to save your cladograms to a log file, as Terminal doesn’t show all of it if over a certain amount of rows. To do this type log (file name, does not have to be a pre-existing file). Whenever you retype the above, the file will be wiped clean and filled with anything typed after the Log processing.

Some fun stuff is next. You actually get to see your trees. For the first thing, you should find the strict consensus, type qnelsen ;. Your trees should appear, but if not type tplot ;. As far as I know, the only other tree you can complete this easily is a majority, by typing majority ;.

I hope this guide can help you to have some fun with phylogenies. If you have any trouble with the above, just comment below and I will try to help.

Posted in Phylogeny | Tagged | 3 Comments

Phylogeny and gaps

It’s always been a peeve of mine when a group is always on the edge of a phylogenetic analysis, but never the focus. This is very obvious in Eusauropoda, where taxa like Shunosaurus, Mamenchisaurus and Omeisaurus are always included as basal stems, but never are in the middle of an analysis (Wilson 2002 and follow-ups come close but still only contain ~10 stem eusauropods). Luckily for us Mamenchisauridae is getting an overhaul sometime within the next decade, but this might not spill over to the badly-analysed british taxa like Cetiosauriscus or ex-Pelorosaurus. Another bad group I’ve come to notice is Averostra. Very in-depth analyses like Carrano et al 2012, Rauhut & Pol 2012 and Nesbitt et al 2011 include taxa on either side of this three-way division, but only some badly-formed analysis like Smith et al 2007 actually include multiple taxa from Coelophysoidea, Ceratosauria and Tetanurae. I think I may end up solving this specific issue, as it covers a taxon I feel I want to analyse. Great matrices like those of Mortimer and Cau would cover this group though, if they get published eventually. This is just a short post, so I’ll include a skeletal diagram of Wuerhosaurus, and its very odd proportions. One pixel is one milimeter scaled to W. ordosensis, with white material from that species and grey from W. homheni. I’m not certain they are the same genus but that is another issue for another time.


Posted in Phylogeny, Sauropoda, Theropoda | Tagged , , | Leave a comment

Skeletal making

I really haven’t been able to think of a topic this year that I can disclose the contents of to others, so here I am writing up a short piece on a new skill I’ve acquired. Be warned, there are skeletal diagrams here of several dinosaur taxa; All are CC-BY 4.0.

Picking a taxon to make a skeletal diagram of is probably the most difficult of all choices for me. I am not very good at making decisions in any aspect of life, skeletals are no different.Veterupristisaurus

One of the first topics I tried to skeletal was simply known bones pasted onto the outline of a separate taxon, illustrated by my Veterupristisaurus skeletal above. The silhouette is from Scott Hartman’s Acrocanthosaurus, and thus the silhouette is copyrighted to him. The bones themselves are free for your enjoyment, such as to add onto a separate silhouette.

Irritator skeleton

Next, I tried to create full digital skeletals, like the Irritator above. This was my first full skeletal, for a project I have yet to complete. Full skeletals have shown me more about the way bones interact, with some advice from friends on vertebral counts and pectoral articulations.


I have also at times attempted to create traditional skeletals, and then transform them into digital, like this Haestasaurus you see above. This is probably one of the hardest for myself to cope with because being a perfectionist I always try to get the bones exact, and with most traditional methods that’s an impossibility.


Sometimes to choose I just pick a taxon that badly needs a skeletal, or a better one, like this Peteinosaurus. This one was very difficult because of the lack of online data, apparently coming from the fact that Italy has a law against the distribution of photographs of Italian specimens, which in fact may be part of an issue discussed on SVPOW how the line between private and public specimens may be more blurred than people perceive. Hopefully this skeletal gets more use than the abomination found on David Peters website (not linking, I would not wish to draw readers here over to his pages).

As time has gone on and my skill and techniques have progressed, I’ve started to realize that I really can make skeletal diagrams that are both very accurate (in my own opinion) and display the known material correctly to the fullest extent. My second most recent skeletal diagram, a full one of the new taxon Europatitan eastwoodi Fernández-Baldor et al 2017, can be seen below.Europatitan


Welcome back to images galore. 🙂

Fernández-Baldor, F.T.; Canudo, J.I.; Huerta, P.; Moreno-Azanza, M.; Montero, D. (2017). Europatitan eastwoodi, a new sauropod from the lower Cretaceous of Iberia in the initial radiation of somphospondylans in Laurasia”PeerJ5: e3409. doi:10.7717/peerj.3409.

Posted in Anatomy, Palaeoart, Sauropoda, Theropoda | Tagged , , , | Leave a comment

Tyrant reptile King? Which king?

Yes, my last post was a decently long time ago, so I thought I’d pick up the slack by writing about everyones “favourite” theropod, Tyrannosaurus rex. It is by far the most completely known of any dinosaur, with several famous and complete skeletons such as Sue, Stan, Trix, Tristan, AMNH 5027, and many more. The entire skeleton is known and represented, except, very funnily, for the feet *. But, oddly enough, such “completeness” may not be present, as the great Tyrannosaurus rex may be multiple species.


Tyrannosaurus “rex” specimens AMNH 5027 (above) and CM 9380 (below; holotype). Both images copyright Scott Hartman, used here simply to illustrate an idea.

Look at the image above. As per priority, the name Tyrannosaurus rex must stay with the holotype specimen, CM 9380. CM 9380 is relatively complete, as far as dinosaur specimens go, even though it lacks a tail and neck. However, there are a few minor differences between the specimens shown above, but simply not enough to change the consensus that both are T. rex. This may change with the next specimens I’ll show though.


Many Tyrannosaurus skeletons, including most of the ones mentioned here. Top to bottom: “Dynamosaurus imperosus“, CM 9380, AMNH 5027, Stan and Sue. Skeletals by and copyright GetAwayTrike

The oddest of all specimens is Stan. With much larger legs, a much smaller skull, and very different features that the other specimens, Stan is a definite candidate for a new species. Unlike T. rex, Stan has a very odd pelvis, particularly the ilium, and a very thin femur. The skull has larger fenestrae of different shapes. While the mandible is relatively similar, the cranium has some major differences.


The AMNH 5027 skull, as illustrated by, and copyright Tracy Ford. From Paleofile.


To keep the sources consistent, these Stan skulls are also by and copyright Tracy Ford.

The proportions of Stan’s skull are surprisingly different. While it would be expected that individuals of the same species are nearly identical, there are several glaring issues with that. Stan has a higher skull overall, with vertical orbits, a less rectangular antorbital fenestra, and a larger maxillary fenestra. The infratemporal fenestra has a much larger opening ventral to the constriction, the orbit doesn’t have a posterior protrusion, the lachrimal is more robust, the naris is more anterior-facing, the quadrate is more gracile, the dorsal maxilla expansion isn’t as long, the postorbital is smaller, the squamosals don’t extend as far behind the skull, the jugal fenestra beneath the antorbital fenestra is smaller, the lachrimal fossa is larger, and there is no ventral expansion at the posterior end of the maxilla. Thats around 15 features just in a moderate examination of the skull, which may even go up to 30 if I examine the skull more in-depth. With very different proportions, and many discrete characters, Stan is odd enough for me to consider it a separate species from AMNH 5027 and CM 9380, which are T. rex by default.


The dentary of Dynamosaurus, from the original description by Osborn in 1905.


The dentary of Sue in multiple views. From the Everythingdinosaurs blog, pictures by John Weinstein (Copyrighter uncertain).

Now for the third distinct section, the “Dynamosaurus” morphs. In fact, since the holotype of Dynamosaurus clearly fits into this group, I am tempted to classify them as Tyrannosaurus imperosus. These specimens, Dynamosaurus and Sue, are alike in multiple ways. From their very large size to their robust dentaries and other bones, Tyrannosaurus imperosus, while not as odd as Stan, should be a separate species in my opinion. Dynamosaurus and Sue both have a very robust dentary, with a more flattened ventral edge and a more prominent anterior dorsal expansion. The dentaries are nearly identical in shape, with a similar foramen near the anterior ventral edge, a very large and similarly shaped internal mandibular fenestra, and sutures and ridges identical in every detail as far as I can tell. From this, it can just be assumed, probably accurately, that they represent the same taxon, and that taxon is as unique as Sue, a large, robust dinosaur with odd features and a large size. Sue themself (I am not going into gender arguments) seems to have a few shared features, with the overall skull being more similar to AMNH 5027 than Stan, and the pelvis having some odd lumps and depressions like Stan, but still quite different. Trix is the only questionable specimen that may result in Sue being the same species as Stan, but if so the binomial Tyrannosaurus imperosus will remain.

* As a side note, every single specimen of Tyrannosaurus either has very fragmentary and intermediate pedal remains, or is missing 1 foot, with the other near completely preserved.

Posted in Anatomy, Phylogeny, Theropoda | Leave a comment

What ever happened to Loricosaurus?

More sauropods, but deviating slightly from what I was planning to post today. Working through all the titanosaurs known from South America, I came across an old taxon I learned a little about several years back. Loricosaurus is a genus of Lithostrotian originally named in 1929 by Friedrich von Huene, for osteoderms he considered to belong to ankylosaurs, which would be the first ankylosaurs from South America. Of course, that idea did not hold up, and now the holotype scutes are clearly considered to be from a titanosaur. However, this is not a tale of that, but instead of the species of Loricosaurus.

Screen Shot 2016-08-16 at 10.13.03 AM

Holotype scutes of Loricosaurus scutatus (Figs 2-18), from Huene (1929) Pl. XLIII. Fig. 1 is an “acanthopholid” spine.

A little history on this: Originally, in 2013, I came across a mention of Loricosaurus somewhere, as a genus that is potentially a synonym of Neuquensaurus, but not certainly. Thus, I went to wikipedia, where I basically just compiled all the information I could find into the article, so I could learn a little about the taxon. Among the information I found, from I don’t know where, was that Loricosaurus noricus, named by Huene in 1929, was the type species, and that the other species, Loricosaurus scutatus, was also named by Huene in 1929. The website where I apparently identified Loricosaurus noricus was the PaleobiologyDatabase, but a few years back that site went defunct, and was replaced with Fossilworks. Apparently, with the change of site also came a revision of information, as any mention to Loricosaurus noricus was completely replaced. In addition, any search for “Loricosaurus noricus” in Google or Google Scholar comes up with no non-wiki-mirror results. The final bullet: I have gone through all of Huene (1929), and the only mentions of Loricosaurus are of the species Loricosaurus scutatus, which Huene wonderfully labels as “gen. et sp. nov.”. Thus, as far as I can tell, Loricosaurus noricus does not exist.


  • von Huene, F. 1929. Los saurisquios y ornitisquios del Cretacéo Argentino. Anales del Museo de La Plata(series 3) 3: 1–196.
Posted in Sauropoda | Tagged | 1 Comment

Every discrete anatomical feature you need to know (vertebrae)

I have been (still) working on the titanosaur phylogeny you have seen in my last few posts. This time, because inspiration struck, I have decided to go beyond the recent minimal-word, main picture posts of late, and instead I am going to list off and describe basically every discrete (non-obvious) anatomical feature you need to know on, who else, sauropods. Also, thanks to SV-POW for convincing me sauropods are far better than theropods, and thanks to all those long-dead palaeontologists who described such scanty remains as to give me lots of abandoned material to work on without needing to worry about others.

First up, of course (thanks SV-POW) are the vertebrae. Vertebrae are by far the most distinctive feature of sauropods, with their large size, unusual shape, and unique accumulation of laminae (detailed soon). However, as I found out fairly recently, these vertebrae are not as unique as you may think. Many an old sauropod (thanks Owen *sigh*) was described based on a chimaera of both true sauropod and “Iguanodon” caudals, just review the history of Pelorosaurus conybeari/brevis if you don’t believe me. Apparently these vertebrae are only truly different in the internal and discrete features (you can thank me later for posting this, making lives easier). Also, as I seem to be alone in thinking, since days of yore (thanks again Owen *sigh*), mosasaurs also have surprisingly similar caudal vertebrae, likely behind Owen’s interpretation of the relationships of Cetiosaurus. Based on personal observations of the vertebrae, mosasaurs, particularly Plioplatecarpus (seen below), differ mainly in the transverse process not being reduced.


Plioplatecarpus sp. procoelous caudal vertebrae. Image from (and copyright)

Now, enough with the similarities and such of vertebrae, it is time to go through the most commonly used terms and features in sauropod vertebrae, beginning with the anatomy.

The vertebrae itself is little more than a single link in a chain, like one loop on a string of chain mail, although the ways they connect are different. The most important terms, without any doubt, are the general terms of location, Anterior, Posterior, Dorsal and Ventral. They mean front, back, top and bottom respectively. In caudal vertebrae, the terms Anterior and Posterior are sometimes swapped with Proximal and Distal, respectively, but mean the same thing, so don’t get confused. Without any distinct features present (such as in heavily eroded vertebrae), it can be very hard to tell the difference between anterior and posterior, but in vertebrae, the dorsal side always hold the Neural arch, a small tunnel through which “flows” the spinal chord, the nerve “highway” of our bodies. The main part of the vertebrae is the Centrum, which is a ball-and-socket joint that allows the main flexibility in the spine, while keeping it strong. The anterior face is called the Condyle, and the posterior face is called the Cotyle. In the side of the centrum most often is a depressed, or straight up hole, which is called the Pleurocoel. On the top of the neural arch is the Neural spine, a generally tall, often flattened peak. Anterior to this spine if the Prezygapophysis (pre for anterior), and posterior is the Postzygapophysis (post for posterior). These Zygapophyses are the dorsal links of the vertebrae that help them to keep their shape and strength, and form interlocking connections that, while being flexible, can also be incredibly rigid. On the side of the Neural arch in all but the posterior caudal vertebrae is the Diapophysis, also called the Transverse process. This is the dorsal articulation for ribs, which are pretty important to keep an animal from, you know, having the skin unable to keep the internal organs internal. Ventral to the diapophysis in cervical vertebrae, but anteroventral in dorsal vertebrae, are another rib articulation, called the Parapophysis. On the centrum, there are a few different terms to describe the type of ball-and-socket joint present. When both the Condyle and Cotyle are flat, it is called Amphiplatayan. When both Condyle and Cotyle are concave, this is called Amphicoelous. With a ball in front, the vertebrae it Opisthocoelous. Only found in caudal vertebrae, the Cotyle can have the ball, which is called Procoelous, and in very derived taxa both ends can have a ball, called Biconvex. There is also the state of one face being concave, and the other being flat, alternately called Procoelous/Distoplatyan or Procoelous-Opisthoplatyan (See the first comment here for a discussion of the differences).


Giraffatitan cervical from (and copyright) Janensch (1950), but modified by the great guys at SVPOW. The neural arch (not labelled) spans from the top of the centrum to the base of the neural spine.

Laminae are, in there meaning, small upraised areas present in vertebrae, often, but not always, as ridges. As what I would have once called an extreme layman with regards to laminae, one single realization made me into a master of laminar terms (still dwarfed by true sauropod experts though). This realization was, in a case like the Posterior Centrodiapophyseal laminae (PCDL), the main thing to do is break up the name. In this case, the words strung together are Posterior, Centrum and Diapophysis. From this, we can now tell that the lamina spans from the Centrum to Diapophysis, and it is the posterior one of these lamina (often branching itself from the Posterior centrum; in all cases, with only one lamina to the diapophysis from the centrum, there is no PCDL). This method can be used for all lamina, the key is to know the names for the anatomical features detailed, which is why I go over them above. There are a very large number of laminae in derived taxa, such as Dongyangosaurus, and often these laminae can be used to distinguish taxa.


Dongyangosaurus dorsal vertebrae, from and copyright Lu et al 2008.

There really is a huge number of laminae, which are named based off of Whitlock (2011) Wilson (1999) Janensch (1950), unless you are looking at Harris (2006), which instead uses bird laminae. Here are the lamina from the above image (the L at the end of each always means laminae, apart from the Pleurocoel [PL]):

  • SPDL – Spinodiapophyseal lamina
  • SPRL – Spinoprezygapophyseal lamina
  • PODL – Postzygodiapophyseal lamina
  • SSPDL – Subsidiary spinodiapophyseal lamina
  • CPOL – Centropostzygapophyseal lamina
  • CPOL1 – Centropostzygapophyseal lamina 1
  • CPOL2 – Centropostzygapophyseal lamina 2
  • PRDL – Prezygapodiapophyseal lamina
  • PPDL – Parapodiapophyseal lamina
  • CPRL – Centroprezygapophyseal lamina
  • ACPL – Anterior centroparapophyseal lamina
  • PCPL- Posterior centroparapophyseal lamina
  • SCPL – Subsidiary centroparapophyseal lamina
  • PCDL – Posterior centrodiapophyseal lamina
  • TPRL – Intraprezygapophyseal lamina

A few of these laminae (CPOL 1 & 2, SSPDL) are relatively rare among taxa, but most of these are found across nearly every sauropod.

The final section of the discrete anatomical features in vertebrae is pneumaticity. Pneumaticity in vertebrae comes in the form of air pockets scattered throughout the inside of the vertebrae. When these pockets are large and cavernous, they are called Camerate, but when they are small and numerous they are called Camellate. Camerate taxa usually have very large, deep, and expansive pleurocoels, with the coel feeding this air pocket from the outside. Camellate vertebrae most often still have this pleurocoel, but in nearly every case it is small and restricted, with the rest of the vertebrae having the “spongy” texture. Often there are small external entrances to these camellae, which are called Fossae, and sometimes even the Pleurocoel can be found within a fossa.


Internal pneumaticity. Tornieria shows camerae, Malawisaurus has camellae in the neural arch only, and Saltasaurus has a camellate centrum. From Wedel & Taylor (2013).

This brings me to (what I think it) the end of this post, which AFAIK covers the main discrete anatomical terms and features you need to know, about Sauropod vertebrae.


  • Wedel MJ, Taylor MP (2013) Caudal Pneumaticity and Pneumatic Hiatuses in the Sauropod Dinosaurs Giraffatitan and Apatosaurus. PLoS ONE 8(10): e78213. doi:10.1371/journal.pone.0078213
  • Janensch, W. (1950). “The Skeleton Reconstruction of Brachiosaurus brancai.”: 97–103.
  • Lu Junchang; Yoichi Azuma; Chen Rongjun; Zheng Wenjie; Jin Xingsheng (2008). “A new titanosauriform sauropod from the early Late Cretaceous of Dongyang, Zhejiang Province”. Acta Geologica Sinica (English Edition) 82 (2): 225–235. doi:10.1111/j.1755-6724.2008.tb00572.x.
Posted in Anatomy, Sauropoda | Tagged | 9 Comments

Work on titanosaurs (aka phylogeny #2)

Now, from what I had in my last post I expanded the matrix coding for 50 taxa instead of 42, and the tree definitely changed. Unfortunately, I forgot to add the taxa names in the first portion so instead they are labelled as 43-50. But the below list is of the taxa I added, in order.

43 – Brontomerus; 44 – Astrophocaudia; 45 – Macrurosaurus; 46 – Gigantosaurus; 47 – “Brachiosaurus” nougaredi; 48 – Mongolosaurus; 49 – “Titanosaurus” nanus; 50 – Atacamatitan

Most of these taxa are highly unknown, and the tree clearly reflects that. Below is a tree before a single character correction for “B.” nougaredi (changing a 1 to ?), as well as changes for Antarctosaurus giganteus (included in my analysis on the previous post, but errors in coding resulted in it being a Shunosaurid). It is funny to note that, although not labelled, the entire topology has changed, with Argyrosauridae separated from Lognkosauria, Argentinosaurus becoming a lithostrotian, Brontomerus grouping with Gigantosaurus and Venenosaurus away from Cedarosaurus.

Screen Shot 2016-06-27 at 9.42.08 PM

The entire tree is not very strongly supported, in fact, the highest clades include Diplodocidae (not shown above), at only 99.92%, not even found in all trees.

Posted in Anatomy, Phylogeny, Sauropoda | Leave a comment