A new Rhamphorhynchus with soft tissue: TMP 2008.41.001

A new PeerJ paper by Hone, Henderson, Therrien and Habib (2015) reports on a new complete, articulated (with a crushed and scattered torso) Rhamphorhynchus specimen, TMP 2008.41.001, the Tyrrell specimen (Fig. 1).

Figure 1. The new Tyrrell specimen of Rhamphorhynchus.

Figure 1. The new Tyrrell specimen of Rhamphorhynchus.

One species
Hone et al report, “Here we follow Bennett (1995) in considering all Solnhofen specimens of Rhamphorhynchus to belong to a single species, R. muensteri.” This is wrong and lazy. Phylogenetic analysis (Fig. 2), which Hone et al do not attempt, divides this genus into several clades. Even the feet have distinct pedal proportions. The Tyrrell specimen nests at the base of the JME SOS 4785 (Darkwing specimen) clade and is similar in size to other clade members.

Figure 2. Cladogram of Rhamphorhynchus.

Figure 2. Cladogram of Rhamphorhynchus. See, they’re not all one species. And phylogenetic miniaturization occurred at the genesis of this genus.

Juveniles and subadults?
Hone et al. report, “The genus has previously been split into a dozen or more species but these have convincingly been shown to consist of juveniles and subadults of a single species (see Bennett, 1995 for a review).” This is also wrong. We know from several single genus bone beds that hatchlings and juveniles of all tested pterosaurs had adult proportions. We know from phylogenetic analysis that a juvenile Rhamphorhynchus was recovered in phylogenetic analysis because it scored identical to an adult but was less than half as tall.

The specimen used to be in a private collection
of the quarry owners. It was discovered in 1965 and recently sold to the Tyrrell. It is preserved in ventral view with light impressions of wing membranes and a trapezoidal tail vane.

The skull
Hone et al. report, “Some sutures in the skull can be tentatively identified but these are mostly not clear, either because they are being obliterated as a result of cranial fusion during ontogeny, or owing to crushing of elements.” Here (Fig. 3). DGS colorizes the skull bones. I did not notice any obliteration in the sutures.

Figure 3. Rhamphorhynchus Tyrrell specimen after DGS colorizing of the bones.

Figure 3. Rhamphorhynchus Tyrrell specimen after DGS colorizing of the bones.

The teeth
Hone et al. considered the tooth count (twelve uppers, ten lowers) “higher than normal” for Rhamphorhynchus (ten uppers, seven lowers), but the extras appear to be incipient teeth or tooth tips from the right side of the skull.

Sacrum
Hone et al. identify four sacrals (Fig. 5), not counting the anterior vertebrae that lie between the ilia and sends out processes to the anterior ilia.

Caudals
Hone et al. report, “The divisions between the vertebrae are difficult to distinguish along the majority of the length of the tail and parts are covered by the left pes, so a vertebral count is not possible.” I had less of an issue while applying DGS (Fig. 4). But then I had only a jpeg, not the real thing. The photo looks good. Is this a case where DGS trumps first hand observation? See figure 6 for comparison.

Figure 4. Rhamphorhynchus, Tyrrell specimen, caudals. They are distinct from one another contra Hone et al. 2015.

Figure 4. Rhamphorhynchus, Tyrrell specimen, caudals. They are distinct from one another contra Hone et al. 2015. Click to enlarge.

Dorsal ribs
Hone et al. report, “Numerous dorsal ribs and gastralia are preserved on the specimen but a count is not possible given that many elements overlap one another.” This is exactly what DGS does best (Fig. 5) because the eye get overwhelmed by the chaos and colors segregate and ultimately simplify the issue.

Figure 5. Torso of Rhamphorhynchus from Hone et al. 2015. Above as originally interpreted. Below using DGS. What Hone et al. identify as a mc (metacarpal) is the radius + ulna. Scale bar = 2 cm. One rib is actually a prepubis. An extra sacral rib is identified here. The coracoids are in light blue. The light gray areas maybe an egg. A smaller second possible egg is also in gray. The sternal complex (not just the sternum) appears to be broken into several parts. Fibula parts are identified along with a second ischium.

Figure 5. Torso of Rhamphorhynchus from Hone et al. 2015. Above as originally interpreted. Below using DGS. What Hone et al. identify as a mc (metacarpal) is the radius + ulna. Scale bar = 2 cm. One rib is actually a prepubis. It is much more robust then even the anterior ribs. A fifth acral rib is identified here. The coracoids are in light blue. The light gray areas maybe an egg. A smaller second possible egg is also in gray. The sternal complex (not just the sternum) appears to be broken into several parts. Fibula parts are identified along with a second ischium.

Sternal complex
Hone et al. refer to the sternal complex as the sternum. That’s inexact. They know it’s not just a sternum, but also includes the clavicles and interclavicle. Nesbitt (2011) assumed these latter elements were missing from pterosaurs in his analysis, so such deletions have real world consequences in cladograms.

Figure 6. Rhamphorhynchus Tyrrell specimen wing GIF movie showing vane and wing tip ungual visible in high contrast.

Figure 6. Rhamphorhynchus Tyrrell specimen right wing GIF movie showing vane and wing tip ungual visible in high contrast. Note the lack of differentiated caudal vertebrae. Click to enlarge.

Wings and their membranes
Hone et al. identify an ulna where an ulna + radius is present, as described in their text. In prior works these authors have supported the deep chord wing membrane false hypothesis, despite all evidence demonstrating otherwise. Here again is another narrow chord wing membrane with a direct connection to the elbow. That the knees are drawn up does not negate this observation, which is universal in pterosaurs.

FIgure 9. Rhamphorhynchus wing GIF movie (click to enlarge) showing radius + ulna, pteroid and standard narrow chord wing membrane.

FIgure 9. Rhamphorhynchus left wing GIF movie (click to enlarge) showing radius + ulna, pteroid and standard narrow chord wing membrane.

Wing tip
Hone et al. note that both wings terminate in a squared off tip. They were not present when this specimen was prepared 50 years ago. I agree that no wing tip ungual is readily apparent here, as opposed to the many seen on several specimens previously. If you bump up the contrast on the matrix, several ungual candidates appear (Fig. 10). The “squared-off tip” described by Hone et al. looks like any other articular surface, as in the other interphalangeal joints on the wing. This should have been noted.

Figure 10. Right wing tip of Tyrrell specimen of Rhamphorhynchus showing blunt tip and, with higher contrast, several ungual candidate impressions.

Figure 10. Right wing tip of Tyrrell specimen of Rhamphorhynchus showing blunt tip and, with higher contrast, several ungual candidate impressions.

 

Figure 11. Pelvic elements of Rhamphorhynchus, Tyrrell specimen, replaced to their in vivo positions in lateral view along with the two possible egg candidates for comparison to the pelvic opening. Seems like a good fit.

Figure 11. Pelvic elements of Rhamphorhynchus, Tyrrell specimen, replaced to their in vivo positions in lateral view along with the two possible egg candidates for comparison to the pelvic opening. Seems like a good fit. The prepubis, originally identified as a rib, has no counterparts among the ribs. It is more robust and straighter.

Pelvis
Hone et al. report, “The pelvis is partially disarticulated and some elements appear to have been lost.” The ilia are both easy to see. Hone et al. report, “The proximal part of the right pubis is articulated with the right ilium, but only the articular end is visible and the rest appears to be hidden below other elements.” I did not see that. I did see both pubes scattered in the mix (Fig. 5). They are not readily apparent. Hone et al. report, “Only one ischium (?right) can be identified.” I found both (Fig. 5) parallel to each other. Hone et al. report, “Both prepubes are preserved but are in poor condition and covered by other elements. They are in close association but are not articulated with one another and lie posterior and ventral to the sacrum.” The authors did not identify the prepubes in their tracings. In ventral view the prepubes should not be covered by other elements (which elements?). I found one prepubis, misidentified as a rib by Hone et al. and the other one where they said it was. I don’t think they realize how large the prepubes are in this species of Rhamphorhynchus, which is a ‘chubbier’ pterosaur than most others owing to its long ribs, gastralia and deep prepubes. No other ribs are robust like the prepubis. And all of the anterior ribs, those likely to be more robust, but are not in this species, are accounted for. Plus it matches the darkling prepubis (Fig. 12).

Figure 12. The darkling specimen of Rhamphorhynchus, very similar to the Tyrrell specimen, showing the depth of the gastralia and prepubis.

Figure 12. The darkling specimen of Rhamphorhynchus, very similar to the Tyrrell specimen, showing the depth of the gastralia and prepubis.

The foot
traits alone nested the Tyrrell specimen within its clade as this is the only clade with penultimate pedal phalanges longer than the others (Fig. 13). Click here to see others.

Figure 13. Pes of the Tyrrell specimen of Rhamphorhynchus.

Figure 13. Pes of the Tyrrell specimen of Rhamphorhynchus.

The wing membrane
Hone et al. report, “Each wing has a more narrow chord along most of its length than seen in some specimens of Rhamphorhynchus (e.g., BSPG 1938 I 503a, the ‘DarkWing’ specimen—Frey et al., 2003) suggesting some postmortem shrinkage of the membranes (Elgin, Hone & Frey, 2011).”

There is no shrinkage!
Hone et all are refusing to face the facts. They are making up scenarios to avoid the narrow-wing morphology (Peters 2002). This pterosaur, like all others, has a narrow chord wing membrane. Hone et al acknowledge that. And so does the dark-wing specimen, as documented earlier and shown below (Fig. 14). When the wing is outstretched, as if in flight, the membrane goes with the wing finger and it is stretched between the elbow and wing tip. Any other attachment points needlessly complicate matters. Any other scenarios are excuses and just-so stories.

Figure 1. The darkwing specimen of Rhamphorhynchus. Top: in situ. Middle: Soft tissues highlighted. Bottom: Neck and forelimb restored.

Figure 15. The darkwing specimen of Rhamphorhynchus. Top: in situ. Middle: Soft tissues highlighted. Bottom: Neck and forelimb restored with wings outstretched. This is another narrow chord wing membrane when the parts are restored to their in flight position. The arrows show how much the wing would have to stretch to attach to the ankle. But there’s no muscle and bone to stretch it. Remember, in flight the tibia stick almost straight out laterally,

Biut wait… there’s more!
Hone et al. report, “Proximal to the elbow, the right tenopatagium (Fig. 6) is rather less clearly preserved than the left actinopatagium (Fig. 5), but does appear to meet the left ankle as is considered common, or even ubiquitous, for pterosaur wing membranes (Elgin,Hone & Frey, 2011).” Yeah, right… This is really reaching. This is why these guys keep rejecting my papers and why I don’t attend pterosaur symposia. They are adamant about rejecting anything I have published on. Evidently, I have (figuratively) poisoned the well. And that’s a sorry state of affairs. They will never say, “well, I guess Peters 2002 was right about the narrow chord wing membrane. It’s right here in front of us.”

You should know
Hone et al. report, “Furthermore, at least some parts of the wings have been covered with some form of transparent preservative and brush marks (e.g., swirls) are clearly visible in places on the matrix.”

Uropatagia are preserved
But due to the extreme bending of the knees, their shape cannot be determined. Hone et al. provided an extreme closeup of fibrils in a uropatagium (their figure 7, but note the singular state here as they falsely believe, based on the Sordes error, that one membrane extended from leg to leg). They reported the element on the right is the right tibia, but the right tibia is devoid of tissue, as far as I can tell. I was unable to match the extreme closeup to any other wider view shots. There does appear to be soft tissue between the left femur and tibia (remember the specimen is on its back so left is right and right is left). Their figure 8 has a wider view and represents the left tibia. Still the fibrils are close to the tibia and they provide no evidence that these are not separate uropatagia, as in all other pterosaurs.

Gut contents
Hone et al report gut contents of an indeterminate vertebrate. “most of these are distorted and difficult to identify though their overall shape appears to be that of squat cylinders. Their exact identity cannot be determined as they are incomplete and partially covered by other elements, and much of the chest cavity has calcite crystal buildup. –– These bones may represent fish or tetrapod elements, but are not part of the pterosaur as they match none of the dissociated or missing material (ribs, gastralia, sternal ribs, pteroids, pelvic elements) but instead are a sub rectangular series and associated subcircular elements that collectively may be vertebrae (Fig. 3).” Rhamphorhynchus is typically considered a fish eater as fish have been found within certain specimens. ‘Hooklets’ [= simple spikes and hook-like shapes] are found by the thousands in the coprolites. Hone et al. report, “If the diagnosis is correct, this is the first recorded coprolite for any pterosaur.”

Odd that the torso should be so upset, but the soft coprolites untouched.

Hone et al. did not consider the possibility
of an internal immature egg. The item has an oval outline (Fig. 11). And there may be a second smaller, even more immature egg in the mix (Fig. 11). Hard to tell in all that chaos.

Ontogeny
Hone et al. are correct in stating the Tyrrell specimen is adult or nearly so. But sutures are not a reliable indicator of ontogeny. Several clades fuse early and others never fuse, patterns common to lepidosaurs, not archosaurs.

Found typos
Perhaps these can be corrected since they are online:

  1. several specimens seen to have consumed fish”
  2. The uropatagium has become displaced relative to the bones even in some exceptionally preserved specimens (e.g., Sordes PIN 2585-33). The holotype is PIN 2585-3). I find no record for #33 on the Internet.

References
Hone D, Henderson DM, Therrien F and Habib MB 2015. A specimen of Rhamphorhynchus with soft tissue preservation, stomach contents and a putative coprolite. PeerJ 3:e1191; DOI 10.7717/peerj.1191

The rise and fall of the pterosaur tail – part 3

Earlier and yesterday we looked at parts 1 and 2 of the evolution of the pterosaur tail, the reduction of the caudofemoralis, the rotation of the chevrons (attenuation of the entire tail), the development of the tail vane and the appearance of caudal rods.

This series was inspired by a guest post by Scott Persons at Pterosaur.net.blospot on Jan 2.

Today we’ll finish up by looking at Dorygnathus, its ancestors, its variety and descendants (Fig. 1). Previously unrecognized for its importance, Dorygnathus lies at the center of all later pterosaur evolution.

The Dorygnathus clade and the pterosaurs AND their tails that descended from it.

Figure 1. The Dorygnathus clade and the pterosaurs AND their tails that descended from it.

The story of Dorygnathus tails begins with a small Eudimorphodon specimen, Bsp 1994. (Fig. 1) that had a tail of unknown length with small chevrons and no caudal rods.

Sordes was derived from a sister to it and had a relatively short tail with short caudal rods.

The many species within the genus Dorygnathus were larger and had a longer tail supported by caudal rods, supporting our earlier hypothesis that size was a major factor, along with phylogeny, in the ossification of caudal rods.

One Dorygnathus specimen, SMNS 50164, gave rise to azhdarchids and their kin starting with TM10341. It was tiny, the tail was shorter and not stiffened by caudal rods or elongated chevrons. From this point on, elongated caudals and caudal rods no longer appeared in pterosaurs in this lineage.

Another Dorygnathus, R156, lies at the base of the much smaller short-tailed Ctenochasma and kin. From this point on, elongated caudals and caudal rods no longer appeared in pterosaurs in this lineage.

Jianchangnathus preserves only a short portion of its gracile tail, but this taxon gave rise to Scaphognathus. Two specimens demonstrate a reduction in overall size and a reduction in the tail, but n110 shows that elongated caudal “threads” were present stiffening the tail. This clade gave rise to all later pterosaurs, (Pterodactylus, Germanodactylus and their descendants and kin) all of which had a short weak tail. Following Scaphognathus, elongated caudals and caudal rods no longer appeared in pterosaurs in this lineage.

Jianchangnathus also gave rise to a dead-end clade, the darwinopterids, all of which had a medium-length stiff tail, except Pterorhynchus (Fig. 1), which had an elongated tail with segmented vanes running along most of its length. Quite unique, so far as we know, given the general lack of soft tissue preservation.

So the story of caudal rod development was not a simple one. Not all basal pterosaurs ossified them. They do appear primarily in larger forms, but Scaphognathus n110 is the exception on that matter. Caudal rods are not associated with flight, but are associated with vane development and size. They add bulk to the tail, so cannot be weight-saving devices, contra traditional opinion.

And tail vanes act like arrow vanes or weather vanes, keeping the tip of the tail close to the parasaggital plane and in line with the airstream passively.  If a pterosaur wanted to turn it had to bank and to make a coordinated turn, that requires a rudder of spoilers in an airplane, a pterosaur could do all of that with just its wings, like a modern flying wing airplane.

Basal pterosaurs had such a thin-as-a-whisker tail that mass and balance were of little concern. Later large pterosaurs thickened the tail for their own romantic purposes. It may be no coincidence that head crests appeared about the time that long tails with vanes disappeared. That’s fashion for you. You’re either in or out.

Update: Notably, the metronome hypothesis places pterosaurs on the ground when they do “their thang” with their highly ossified tails. Notably, dromaeosaurids were grounded Archaeopteryx descendants. So, caudal supports in both cases were NOT for aerodynamics but terra-dynamics. (Contra Persons and Currie (2012) who reported, “the unique caudal morphologies of dromaeosaurids and rhamphorhynchids were both adaptations for an aerial lifestyle.”

As always, I encourage readers to see specimens, make observations and come to your own conclusions. Test. Test. And test again.

Evidence and support in the form of nexus, pdf and jpeg files will be sent to all who request additional data.

References (published after posted)
Persons WS IV and Currie PJ 2012. Dragon Tails: Convergent Caudal Morphology in Winged Archosaurs. Acta Geologica Sinica – English Edition 86 (6): 1402–1412. DOI: 10.1111/1755-6724.12009. http://onlinelibrary.wiley.com/doi/10.1111/1755-6724.12009/abstract

The rise and fall of the pterosaur tail – part 2

Yesterday we looked at the attenuation of the pterosaur tail, from tritosaur and fenestrasaur precursors like Huehuecuetzpalli and Cosesaurus to the basal pterosaur, MPUM 6009. Today we’ll look at how that hyper-attenuated tail evolved within the Pterosauria. This series was inspired by a January 2 guest blog by Scott Persons at pterosaur-net.blogspot.com in which Persons compared the tail of Rhamphorhynchus to those of dromaeosaurid dinosaurs.

The large pterosaur tree indicates that basal pterosaurs evolvedin two basic directions, 1) toward the Austriadactylus, Raeticodactylus and Dimorphodontia, which led to the Anurognathidae and 2a) toward the Eudimorphodontia, which led toward  Campylognathoides and Rhamphorhynchus on the one hand, and 2b) toward Dorygnathus and the rest of the pterosaurs on the other.

Basal pterosaurs in the Austriadactylus/Dimorphodon line. The blue areas represent the extent of the tiny caudofemoralis. The blue arrow points to the reduced distal tail vertebrae in Peteinosaurus, a basal protoanurognathid.

Figure 1. Basal pterosaurs in the Austriadactylus/Dimorphodon line. The blue areas represent the extent of the tiny caudofemoralis. The blue arrow points to the reduced distal tail vertebrae in Peteinosaurus, a basal protoanurognathid. That’s really the only news from this clade.

The Austriadactylus/Dimorphodon lineage did not create caudal rods. The chevrons were aligned with each caudal centrum and extended the length of a single centrum. In some taxa a forward extension further lengthened each chevron. There is no vane preserved, but then no other soft tissues were preserved post-cranially. When more or less completely preserved, the tails in this clade were very long and attenuated, except when you get to Peteinosaurus, which terminates the tail with tiny bead-like vertebrae. This condition is the first stept to further reduction of all the tail vertebrae in anurognathids, convergent with the tail reduction in other derived pterosaurs.

Quick point: both tails associated with Dimorphodon are not found with Dimorphodon. They could belong to another sort of pterosaur from the same formation. Images are here and here1. Sedwick Museum, Cambridge J.61175,  of the Whinborne Collection2. Natural History Museum, London.  Specimen number 41349. Even so, the caudal rods are not developed as much as in Campylognathoides. 

The reduction of the tail in anurognathids reduces the amount of mass and its moment posteriorly. One would expect a similar loss anteriorly to keep things balanced. Either that, or the toes (while standing) and/or wings (while flying) had to move anteriorly a little. I haven’t seen it yet.

he tail of Eudimorphodontidae, including Campylognathoides and Rhamphorhynchus.

Figure 2. The tail of Eudimorphodontidae, including Campylognathoides and Rhamphorhynchus to scale. Here the convergent development of ossified caudal rods appear most strongly in the large taxa. The pattern is interrupted in small transitional taxa and reaches its acme in the largest Rhamhorhynchus. So size is key here. Larger size = thicker tail and more caudal rods.

Ossified intertwining caudal rods first develop in Campylognathoides, disappear in tiny transitional taxa, like Bellubrunnus, then reappear in Rhamphorhynchus and reach an acme of development in the largest known Rhamphorhynchus (Fig. 2). This suggests that caudal rods might have been present, but unossified when not apparent in smaller fossil pterosaurs. The data also demonstrates that both phylogeny and size determine the development of caudal rods. This clade has the best record of vane preservation, so caudal rods and vanes are correlated and sex may also have something to do with it, as described earlier.

Notably, the metronome hypothesis places pterosaurs on the ground when they do “their thang” with their highly ossified tails. Notably, dromaeosaurids were grounded Archaeopteryx descendants. So, caudal supports in both cases were NOT for aerodynamics but terra-dynamics. (Contra Persons and Currie (2012) who reported, “the unique caudal morphologies of dromaeosaurids and rhamphorhynchids were both adaptations for an aerial lifestyle.”

Next Dorygnathus and the reduction of the tail in the rest of the pterosaurs.

As always, I encourage readers to see specimens, make observations and come to your own conclusions. Test. Test. And test again.

Evidence and support in the form of nexus, pdf and jpeg files will be sent to all who request additional data.

References (published after posted)
Persons WS IV and Currie PJ 2012. Dragon Tails: Convergent Caudal Morphology in Winged Archosaurs. Acta Geologica Sinica – English Edition 86 (6): 1402–1412. DOI: 10.1111/1755-6724.12009. http://onlinelibrary.wiley.com/doi/10.1111/1755-6724.12009/abstract

The Origin of the Pterosaur Tail Vane

The pterosaur tail vane is found in only a few clades of long-tailed pterosaurs. Campylognathoides + Rhamphorhynchus have traditional vanes (Fig. 1). Other basal pterosaurs have fibers, I suppose similar to pycnofibres, emanating from the tail, often concentrated at the posterior tip, sometimes very similar to a vane. Pterorhynchus creates its own pattern with a series of vanes on the posterior tail.

Vane Usage
Traditional workers, like Marsh (1882) have suggested a steering usage for the tail vane. The heretical view was expressed by Peters (2002), “The tail vane on a pterosaur would have acted passively, like a weather vane, keeping the tail near the parasagittal plane in turns and during gusts of wind. It would have made a poor steering device compared to simple banking of the wings.” An hypothesis suggesting the vane’s use as a secondary sexual signal was blogged earlier.

Early and Recent Observations and Hypotheses on Tail Vanes
Marsh (1882) described the segments of the tail vane in Rhamphorynchus phyllurus as extensions of the neural spines and chevrons.

Dr. David Hone reported in his blog“Structurally, all vanes have transverse banding across them which is presumably some form of reinforcement, though where the vanes are composed entirely of skin and interstitial tissues or have perhaps cartilage or anything else involved in their composition is not known.” 

Tail Hairs Expand and Coalesce to Form a Vane
The evolution and first appearance of a tail vane has not been covered in the literature. Figure 1 portrays select fenestrasaur and pterosaur taxa with tail hairs and tail vanes. It’s clear that one evolved from the other. Hairs can become cornified. Pangolin scales and rhinoceros horns are examples of this. Below are descriptions of several taxa that demonstrate the evolution of tail vanes from tail hairs (pycnofibres).

pterosaur tail vane

Figure 1. Pterosaur tail vane evolution. Click to enlarge.

Tail Hairs and Tail Vanes in Select Fenestrasaurs (including Pterosaurs)
Cosesaurus – Slender hairs, first identified as proto-feather divisions by Ellenberger (1978, 1993), emerge from the entire length of the tail.
Longisquama – Difficult to determine, but possible hairs gathered at the tail tip form a primitive decoration convergent with the vane of later pterosaurs.
Anurognathus – Tail hairs restricted to posterior half of the tail vestige.
Batrachognathus – Only a few longer hairs tip the vestigial tail.
Campylognathoides – The tail hairs coalesce to become a vertical tail vane that acts as a secondary sexual signal and passively reorients the tail tip in line with the flight path like feathers on an arrow.
Rhamphorhynchus intermedius – Phylogenetic size reduction included tail vane reduction.
Rhamphorhynchus (Vienna specimen) – The vane assumes a diamond shape.
Sordes – An atypical expansion of the distal tail bones including just a few tail hairs forming a vane shape, but not a vane.
Dorygnathus (Donau specimen) – The tail hairs coalesce at the tip but do not form a vane.
Dorygnathus SMNS 50164 – More substantial tail hairs indicated.
Pterorhynchus – Several dozen mini-vanes extend down the posterior half of the tail. These were individually expanded tail hairs.
Scaphognathus (Maxburg specimen) – An ephemeral trinagular tail vane may have been present. Subsequent taxa appear to reduce this.

Summary
Tail vanes in pterosaurs were most prominent in the Campylognathoides and Rhamphorhynchus clade. The evolved from tail hairs that first appeared in Cosesaurus. Tail vanes were reduced and disappeared in pterosaurs with a reduced tail.

As always, I encourage readers to see specimens, make observations and come to your own conclusions. Test. Test. And test again.

Evidence and support in the form of nexus, pdf and jpeg files will be sent to all who request additional data.

References
Ellenberger P 1978. L’Origine des Oiseaux. Historique et méthodes nouvelles. Les problémes des Archaeornithes. La venue au jour de Cosesaurus aviceps (Muschelkalk supérieur) in Aspects Modernes des Recherches sur l’Evolution. In Bons, J. (ed.) Compt Ren. Coll. Montpellier12-16 Sept. 1977. Vol. 1. Montpellier, Mém. Trav. Ecole Prat. Hautes Etudes, De l’Institut de Montpellier 4: 89-117.
Ellenberger P 1993. Cosesaurus aviceps . Vertébré aviforme du Trias Moyen de Catalogne. Étude descriptive et comparative. Mémoire Avec le concours de l’École Pratique des Hautes Etudes. Laboratorie de Paléontologie des Vertébrés. Univ. Sci. Tech. Languedoc, Montpellier (France). Pp. 1-664.
Marsh 0C 1882. The wings of pterodactyles: American Journal of Science 23: 251-256.
Peters D 2002. A New Model for the Evolution of the Pterosaur Wing – with a twist. – Historical Biology 15: 277–301.