Wednesday, February 27, 2008

The Cetaceous Kings of the Lizards (and the Other King too)

Over two weeks without posting anything, what a shame. I by all means haven't stopped blogging, but for some reason I just decided to write four quasi-complete posts. Well, here's post number two:

Having already talked about the most attenuated living cetaceans, I figured I might as well bring up the basilosaurids. These were some of the last archaeocetes with some Basilosaurus specimens apparently just crossing into the Oligocene (Manning, 2003). In some classifications the family (Basilosauridae) is paraphyletic and includes the sub-family Dorudontinae as well as their close relatives, modern whales (Autoceta - see here). Names used more recently (e. g. Naish 2004) and occasionally in the past seem to indicate that Dorudon and relatives got bumped up to family status (Dorudontidae) which modern whales are either the sister group of or part of. Uhen 1999 does have a distinct Pontogeneus/Basilosaurus/Dorudon clade and a Zyghorhiza/Autoceta clade, but let's not get into this. The point is that while basilosaurids are fairly close to the mysticete/odontocete split and are often used as an example of whale evolution (mostly for the hind legs) it seems that they were up to something a bit...different. Superficially the dorudontids resembled modern whales proportion-wise, but basilosaurids were far more elongated. The generally unsatisfying reconstructions of Basilosaurus often are too fat or resemble rorquals down to the pleats, and mostly just look like stretched out modern cetaceans. I wonder what a reconstruction of Lissodelphis would look like if we only had the skeleton (incidentally, I've never seen a good image of this). Could we predict the tiny flukes, lack of a dorsal fin/ridge and remarkable swimming speed. Well, eventually we would, but probably not from a superficial glance. When basilosaurids are given a closer look, it seems that they are very strange indeed...

While everybody knows that Basilosaurus is the "king of the lizards" despite being cetaceous, the "other king" Basiloterus hussaini is seldom mentioned outside of one paper (Gingerich et al., 1997). Well, okay, it is only known from two lumbar vertebrae from Pakistan and perhaps another from England, but Basiloterus is still fairly interesting. The ~40 million year old vertebrae are pretty large (~20 cm/4 in, if complete) and elongated, but not as extreme as Basilosaurus. It appears that either this is the most basal basilosaurid or their next nearest relative. The dorudontid Pontogeneus was fairly large (4/3 the size of Dorudon) and also comparable in morphology to Basilosaurus too, and Uhen 1998 and Uhen 1999 place them in a clade together. Basiloterus was excluded, but would presumably be even more Basilosaurus-like. Unfortunately I can't find illustrations of the vertebrae from that species, but Gingerich et al. 1997 presents both Basiloterus and Basilosaurus drazindai vertebrae:

Basiloterus is on the left and B. drazindai is on the right. I unwittingly drew these to scale (Basiloterus is ~20 cm and B. drazindai is ~30 cm). The metapophyses are at quite different angles, in B. drazindai they are nearly perpendicular to the neural spine, but in Basiloterus they are much more upwards angled. Also noteworthy is the more rounded neural spine of B. drazindai and the much more elongated centrum.

Basilosaurus drazindai is another basilosaurid seldom seen outside of Gingerich et al. 1997, and was from the same time and location as Basiloterus. And yes, it too has another vertebrae from England that may be assigned to it. The single described vertebrae was probably either a lumbar or an anterior caudal and is of similar size to B. cetoides and B. isis, although it is notable for having a larger neural arch and metapophysis than the latter species.

Gingerich et al. use the discussion of vertebrae to segue into a discussion of one of the more problematic aspects of Basilosaurus: how it moved. They implied that with the relatively short gap in between B. drazindai and B. isis and the subsequent reduction in the neural arch and metapophyses had something to do with a fundamental change in locomotion. But, what could it have been changing to? The authors (well, beforehand) discuss previous theories; such as Howell's in 1930 that since the vertebrae did not interlock and the processes were short it could not power a fluke, but instead a "continuous symmetrically-placed bilateral finfold running the length of the tail". However, Buchholz 2001 points out that the shrinking vertebrae at the end of the tail point to the presence of flukes which would only take up around 3% of the total length. Next up was Kellogg (1936) who noted that since the vertebrae were connected only by the ends of their centra, it would have been quite maneuverable. Buchholtz also regarded this as a highly maneuverable species, and regarded the thick ribs as a feature which would have allowed it to level off. Slijper in 1946 interpreted the large metapophyses and low neural spine in some of the posterior lumbars and caudals as a sign of it moving by horizontal undulations. Gingerich et al note that there appears to be more going on than simple enlargement, and raise the possibility that increasing centrum height and length would mean a great deal of the vertebrae would have been filled with fatty marrow in life. They interpret this as meaning Basilosaurus lived predominantly on the sea surface, which is a horizontal plane...

I'd be curious if there actually is some sort of connection between living at the surface and lateral movement. Phocid seals typically move by lateral undulations of the spine (Adam, 2004), so I wonder if there were similar pressures at some point in the past. Despite their locomotion in the water pinnipeds locomote on land by flexing their spines vertically, and elephant seals are apparently flexible enough to literally bend over backwards. Could both forms of locomotion be possible to a degree? Buchholtz 2001 simply stated that the "torso undulated during locomotion" (clever) with a higher wave amplitude than any other whale, but due to the surprisingly low vertebral count* it couldn't have multiple undulations. She concluded that it was a slow swimming specialist in maneuvering, presumably meaning a lot of flexibility both laterally and vertically. Gingerich et al. 1990 noted that other faunal evidence pointed to Basilosaurus isis inhabiting shallow mangroves or seagrass. It is also worth pointing out that Gingerich had described 243 partial Basilosaurus skeletons as of 1990 (vs. 77 for other archaeocetes) in Zeuglodon Valley, so they do seem to have been rather common. Exactly what niche Basilosaurus occupied doesn't seem too clear, but whatever it did it seemed to have been successful for a while.

*She gave a figure of 60, but it may be 70 (see below). As mentioned previously, Lissodelphis has 86-89 and Lagenorhynchus albirostris has 91.

The size of these species is also something of a confusing matter. Basiloterus cetoides is normally stated to be somewhat larger than B. isis (Uhen, 1998), which is stated to be 16 m (52 feet) long (Gingerich et al. 1990). Judging by Kellogg 1936's statement that the largest heads were 1.5 meters long, the whale in his illustration should have measured 18 m (60 feet) in length, although sometimes a figure of 20 m (65 feet) is given (Naish 2004).

Basilosaurus cetoides, from Kellogg 1936. This illustration is still used in publications (e.g. Naish 2004), although appears to be a problem with it.

Gingerich et al. 1990 mention that B. isis has 7 more (rather long) vertebrae than B. cetoides and Zalmout et al. 2000 indicate three more vertebrae were present. Kellogg 1936 gave a count of 7 cervical, 15 thoracic, 15 sacral/lumbar and 21 caudal; Zalmout et al. gave it as 7 cervical, 18 thoracic, 20 sacral/lumbar and 25 caudal vertebrae. Kellogg based his assumption on non-overlapping vertebrae from two specimens, and the Gingerich studies based theirs off of three overlapping specimens. Zalmout et al. suggest that this was the formula for both of the later Basilosaurus species, so I modified Kellogg's illustration subsequently:

Fortunately, this bears a strong resemblance to the illustration of B. isis* from Gingerich et al. 1990 (which I saw afterwards), except for the more elongated thoracic region due to the presence of three more very long vertebrae. This version of B. cetoides is about 25% longer than before, going from 18 meters to 23 (75 feet). Then there are of course vague indications of 20 meter individuals or even vaguer references to 25 meter ones which apparently did not take the increased vertebral count into affect. There still is the possibility that B. cetoides actually was shorter, we just can't know for certain at the moment, but either way it was an incredibly large mammal. As for the weight, the fat Basilosaurus from the BBC's Walking with Beasts was said to be 18 meters long and 60 tonnes; Gingerich et al. 1990 pegged it at roughly 6 tonnes (3-12 in the confidence interval), which is far more bearable.

*Zalmout et al. mention that Ca-14 in B. cetoides is 32% longer the corresponding one in the B. isis specimen, but also 20% wider. Did very large Basilosaurids get even thinner proportionally?

I should probably discuss the notion at the Palaeos page that Basilosaurus wasn't a sideline, but a "mechanically necessary intermediate form". They seem to be working by the theory treating "Eocene Cetacea essentially as a single population moving through phylospace along a smooth morphocline" - which I find quite problematic. Since when does evolution happen like this? The existence of oddballs like the trunked mid-Eocene Makaracetus should be enough to discount this idea. Basilosaurus is said to represent a grade between Georgiacetus and Dorudon, although the page neglects the fact that Dorudontids and Basilosaurids co-existed. There are also lots of missing species, some of which (Zyghorhiza, Chrysocetus) make more plausible "transitional" forms. The fact that cetaceans supposedly went through a stage when they had very elongated non-interlocking vertebrae and then quickly reversed this seems problematic and pointless. I think with more information on Eocene whales, we don't have to use mental gymnastics to make something like Basilosaurus a transitional form.

There are some interesting points brought up aside from the dubious phylogeny. The page raises the possibility of pike-like locomotion, that is, using a long body as a rudder and implying that it may not have been very slow after all. The "rigid" thoracic region is curious, but I think in light of the extremely elongated vertebrae and the presence of only one wave in the body at the time it makes sense. The possibility that the forelimbs were used in locomotion is raised, but they seem proportionally very small to the body size. While these ideas are interesting, their appearance outside of peer-reviewed literature and logical oddities means that we should by no means discount Buchholz.

While probably not directly related to modern whales, Basilosaurus and Basiloterus represent a fascinating evolutionary experiment. Exactly how a 60+ foot long serpentiform whale worked, especially with such strange vertebrae, is still a fascinating matter. Since even living species such as Lissodelphis have bauplans that are strange and effective (i.e. poorly understood), working out how oddballs with no modern counterparts work must be exceedingly difficult. Our knowledge of fossil whales seems to be increasing quite rapidly, and we probably haven't heard the last of old Basilosaurus.



Adam, Peter J. 2004. Monachus tropicalis. Mammalian Species. No. 747, p. 1-9.

Buchholtz, Emily A. 2001. Vertebrae osteology and swimming style in living and fossil whales (Order: Cetacea). J. Zool. Lond. 253, 175-190.

Gingerich, P. D., B. H. Smith, and E. L. Simons. 1990. Hind limbs of Eocene Basilosaurus isis: evidence of feet in whales. Science, 249: 154-157.

Gingerich, Phillip D. et al. 1997. Basilosaurus drazindai and Basiloterus hussaini, new Archaeoceti (Mammalia: Cetacea) from the middle Eocene Drazinda formation, with a revised interpretation of ages of whale-bearing strata in the Kirthar group of the Sulaiman range, Punjab (Pakistan). Contributions from the Museum of Paleontology, Vol. 30, No. 2, pp. 55-81.

Kellogg, Remington. A Review of the Archaeoceti. Carnegie Institution of Washington, Washington 1936.

Manning, Earl M. The Eocene/Oligocene Transition in Marine Vertebrates of the Gulf Coast Plain. From Greenhouse to Icehouse: The Marine Eocene/Oligocene Transition. Columbia University Press, 2003. Partially available: Here (p. 377)

Naish, Darren. 2004. Fossils explained 46: Ancient toothed whales. Geology Today, Vol 20, No 2.

Uhen, Mark H. 1998. Middle to Late Eocene Basilosaurines and Dorudontines. The Emergence of Whales. Partially available: Here

Uhen, M. D. 1999. New Species of Protocetid Archaeocete Whale, Eocetus wardii (Mammalia: Cetacea) from the middle Eocene of North Carolina. Journal of Paleontology Vol. 73, No. 3 pp. 512-528.

Zalmout, Iyad S. 2000. Priabonian Basilosaurus isis (Cetacea) from the Wadi Esh-Shallala formation: First Marine Mammal from the Eocene of Jordan. Journal of Vertebrate Paleontology 20(1):201–204.

Possible life appearance of an advanced basilosaurid (B. isis, maybe B. cetoides). The rounded flukes and dorsal/anal structures are speculative. The fineness ratio is 14 and the head could be up to 5 feet long.

Sunday, February 10, 2008

The Puma Lineage

As I've been known to do, let's start this post off with something of a mystery:

This certainly is not a normal looking puma (Puma concolor) - superficially it appears to be much more gracile than usual. The carcass was collected from Mexico in 1986 and was connected to something of a Mesoamerican legend called the onza, although at least one individual speculated that the legendary cat may have had a different basis. Regardless, this does not appear to have been a wholly isolated incident and the presence of other specimens (skulls) has been alluded to. Apparently there are some dental differences in addition to the long legs, large ears, faint stripes on the forearms, and (semi?) non-retractable claws. While at least one author prone to making outrageous theories speculated that this was an American "cheetah" (Miracinonyx), a 1996 genetic analysis in the ironically cryptic journal Cryptozoology identified it as being "indistinguishable from a western North American Puma". Over a decade later, absolutely nothing has been heard since.

These are interesting times we live in; the possibility of new large South American cat species has been seriously raised, making the idea of a unique puma population (sub-species?) seem more bearable. I haven't seen the genetic test, so I'm not sure it is consistent with that idea or if it is old enough to require another look. It would also be nice to see information on the skulls aside from a few vague remarks. The onza seems to be in dire need of a critical and comprehensive look, something far outside the scope of this article. But, oh, I do believe that Darren Naish at Tet Zoo did hint at some future coverage...

While I've already exhausted the onza as a viable topic, I think it can act as a sort of gateway to the rest of the puma lineage. First up are the American "cheetahs", Miracinonyx. M. inexpectatus was the larger of the two species, had a build between a cheetah and puma, and had fully retractable claws (Krausman & Morales, 2005). The claws of the cheetah (Acinonyx jubatus) claws are variously described as being either "weakly retractile" or "nonretractable" even in the same paper (Krausman & Morales, 2005) and it is implied that M. trumani (and some pumas?) had a similar condition. It has been quite overlooked, even by the previously mentioned paper, that the large dewclaw in living cheetahs plays a large role in predation by "anchoring" on to prey animals trying to escape (Londei, 2000). The enlarging of the dewclaw was due to the specialization in the other claws in locomotion, and it is curious to note that M. inexpectatus had a similar adaptation. Londei notes the additional curiousness of the puma having dewclaws much larger than expected with no known function, and noted that they might have been a vestige of sorts. So, what is the relation between the puma and American "cheetah"? Before the post-cranial skeleton was well known, M. trumani was thought to be an ancestral or modern puma "P. trumani" (Barnett et al., 2005); but subsequently it was placed in either the cheetah genus Acinonyx or even Felis (Krausman & Morales, 2005). It appears than most of the traits used to connect Acinonyx and Miracinonyx were associated with a cursorial lifestyle, but recently both morphological and genetic evidence suggests that M. trumani and Puma form a clade. They appear to have differentiated around 3.2 mya, perhaps due to increasing prairie and predation on pronghorns (Barnett et al., 2005). The very fast running speed of the unique pronghorns (last member of its family) would of course hint at a performance comparable with the cheetah. There is some interest in re-introducing cheetahs to the Americas, but as the article suggests, the individuals do not seem to be aware of the considerable separation between the Acinonyx and Miracinonyx. Regardless of the parallel evolution, I don't think meddling by introducing species will produce the expected results. Say, there is a local species that occasionally shows cheetah-like characteristics...

I don't mean to imply that they're terribly distant relatives, genetic evidence suggests that they diverged as recently as 4.92 mya (Johnson et al., 2006). The entire puma lineage itself was thought to have arisen 6.7 mya, but exactly where or what the cats looked like is not clear. Johnson et al. propose that the lineage is North American in origin, with Felidae itself originating in Eurasia only a few million years before. This model seems problematic of fossils from 3.5 mya of A. pardinensis, and Barnett et al. suggest that the New World was invaded about 6 mya by a puma-like cat. This is just barely within the confidence interval of Johnson et al., 2006. A. pardinensis was present throughout the Old World and was larger and less cursorial than the modern cheetah; and is noted for existing before molecular evidence assumed possible (Krausman & Morales, 2005). Judging by the supplemental data for Barnett et al, it appears that the oldest Miracinonyx is nearly as old. I haven't heard about any fossils of puma-like cats in the Old World [edit: See the comments], and it seems that the early evolution in this group is quite mysterious [edit: This still seems fairly true].

So far I've only talked about the members of the lineage that resembled pumas, cheetahs or something in between, but there is one remarkable oddball nestled within the group. Herpailurus yagouaroundi, also known as the jaguarundi, is one strange looking feline. Oh, I should note that it was the species possibly connected with some early "onza" legends, phew. Anyways, the species is said to have "few feline features", and is a small/moderate (~5.1 kg) species with short legs, short round ears, elongated body, and unmarked pelage at any size or age (de Oliveira, 1998). Only ocelots, jaguars, and pumas are larger and the name may translate as "fourth cat"...or "brown cat" (d. O.). The Johnson et al. paper remarkably puts this cat in the same genus as the puma (P. yagouaroundi), a move which a few sites (like wikipedia) and papers have also followed. I find this exceedingly odd not only due to the morphological differences, but also because Barnett et al. concluded that Puma concolor and Miracinonyx formed a clade. To my knowledge, this paraphyletic usage of "Puma" has not been formally addressed. Anyways, the supplemental material of Barnett et al. noted that there wasn't much support for the placement of the jaguarundi and that it had "nucleotide composition well outside the range of the other Felinae" and excluded it to "reduce the potential for phylogenetic artefacts and the influence of stochastic error". Johnson et al. did not cite this paper. The authors speculated that the jaguarundi may have diverged as much as 5 mya and evolved in South America.

This seems to about wrap it up for the puma lineage. Superficially none of the living species look alike at all, but I guess once we look closer it starts to make more sense. Molecular and morphological data actually seems to be in agreement about the existence of this group, although it does remain quite mysterious. Why is the jaguarundi so small and mustelid-like in a group prone to evolving a cursorial cheetah-like lifestyle? Are there any other oddballs that evolved? And then there is the inevitable string of questions regarding the onza.

Darn, it looks like I'll have to touch on a few members of this group in greater detail another time around.



Barnett, Ross et al. 2005. Evolution of the extinct Sabretooths and the American cheetah-like cat. Current Biology Vol 15 No 15.

de Oliveira, Tadeu G. 1998. Herpailurus yagouaroundi. Mammalian Species, No. 578 pp. 1-6.

Johnson, Warren E. et al. 2006. The Late Miocene Radiation of Modern Felidae: A Genetic Assessment. Science 311, 73.

Krausman, Paul R. & Morales, Susana M. 2005. Acinonyx jubatus. Mammalian Species. No. 771, pp. 1–6.

Londei, Tiziano. 2000. The cheetah (Acinonyx jubatus) dewclaw: specialization overlooked. J. Zool., Lond. 251, 535-547

Monday, February 4, 2008

Integumentary Intrigue: Scaled Squids

Convergence is one of my favorite phenomena in biology. Evolution doesn't have a driving force to any sort of goal (like humanity), but on many occurrences unrelated groups have evolved very similar structures or even appearances. While cephalopods and fish have a fundamentally different body plan, squid have a streamlined body with camera-like eyes, fins, and a lateral line (in a few species). The females of one species of octopus even have a swim bladder. With this in mind, squids with scales don't sound nearly as bizarre...

I'm not sure if Cranchia scabra can be regarded as scaled or not; older authorities (Person, 1969) did but more recently the structures in question have been referred to as tubercles. These are also present to some degree on the unrelated genera Histoteuthis (occasionally) and Magnapinna talismani. Other odd textures such as "warts" have been recorded from some squids of the genus Onykia. These are not homologous of course, but they could shed light on why odd textures evolved in the first place. Relatives of Cranchia, Liocranchia and Leachia also have tubercles to a much lesser degree and other cranchiids occasionally have them on their funnel and nuchal locking mechanisms. If the tubercles are homologous within the cranchiids, then it appears that the tubercles were an external structure that was modified for other uses. But what use are they? I haven't seen any information on how hard the cartilaginous tubercles are, but Cranchia does have a behavior where it tucks its head and appendages into the mantle cavity and inks itself to turn opaque. Perhaps this is a re-evolving of a nautilus or ammonite-like strategy...but that isn't the focus. I purposefully neglected to mention another occurrence of tubercles.

Pholidoteuthis massyae was once grouped in its own genus and family, but was recently classified with P. adami in the family Pholidoteuthidae, itself part of the Lepidoteuthid families (O'Shea et al., 2007). Other members of the clade, the Octopoteuthidae and Lepidoteuthidae are distinguished by a lack of tentacles as adults. Well, the last one is distinguished by something else, but I'll get to that later. What makes P. massyae so curious is that it has papillate tubercles, some incredible pictures of it can be seen here. Curiously P. adami has structures, viewable here, that are termed "dermal cushions", a term O'Shea et al. synonymies with "scales". That paper discussed the confusing taxonomic history behind the two species (once each given a family, for instance), but (for?) now they seem to be part of the same genus. That would imply a close relation between tubercles and dermal cushions and makes the idea of a scaled squid seem a whole lot more plausible.

The structures on Pholidoteuthis look somewhat like scales, but the overlapping "dermal cushions" of Lepidoteuthis resemble fish scales to an uncanny degree (see here). Here at last is the Scaled Squid. Lepidoteuthis is a member of the Lepidoteuthid families of course, but possesses odd dermal structures like the Pholidoteuthidae and doesn't possess tentacles like the Octopoteuthidae. It seems like all of these families tend towards gigantism, Pholidoteuthids have a mantle length of ~0.7 meters, Lepidoteuthis has a 1 meter mantle and Taningia has one up to 1.7 m long. As if the notion of a near-giant tentacle-less scaled squid seems pretty bizarre, but the recently discovered males add yet another oddity. The first mention of a male is from Jackson & O'Shea 2003, and it is worth mentioning that this is a very cryptic species. The males are around half the size of the females and have an incredibly modified sucker ring on the dorso-lateral arm. The authors speculate that the saber-like hook could be used to "aid male purchase" by either hooking into the female's flesh or interlocking with the scales. Alternately, they also could have been used in male vs. male fighting - but either way the sudden modification of the hooks is noted as being without parallel in the cephalopod universe. Oh, and the males also have a penis half the length of their mantle...

Unfortunately, that's about it for the subject of squid scales. It would be fascinating if the overlapping scales of Lepidoteuthis were somehow related to the bizarre hook on the males - but it could very well be that both have uses more or less unrelated to mating. It still isn't really clear exactly what the tubercles/dermal cushions/scales do on members of the Lepidoteuthid families, and I'd be curious if the Octopoteuthids actually lost that characteristic somewhere along the road. I'm sure that with time we might have a few more surprises and maybe a better understanding of this exceptionally strange characteristic.

I really wish the Tree of Life was as complete for all groups as it is for cephalopods, it really is a remarkably valuable resource.



Jackson, George D. & O'Shea, S. 2003. Unique hooks in the male scaled squid Lepidoteuthis grimaldi. J. Mar. Biol. Ass. U.K., 83, 1099-1100

O'Shea, S. et al. 2007. The nomenclatural status, ontogeny and morphology of Pholidoteuthis massyae (Pfeffer, 1912) new comb (Cephalopoda: Pholidoteuthidae). Rev Fish Biol Fisheries, 17:425–435

Person, Phillip. 1969. Cartilaginous Dermal Scales in Cephalopods. Science, Vol. 164, No. 3886, pp. 1404-1405.

Vecchione, Michael and Young, Richard E. 2007. Pholidoteuthidae Adam, 1950. Pholidoteuthis Adam, 1950. Version 18 December 2007 (under construction). in The Tree of Life Web Project,

Young, Richard E. and Mangold (1922-2003), Katharina M. 2007. Cranchia Leach 1817.
Cranchia scabra Leach 1817. Version 14 June 2007. in The Tree of Life Web Project,

Young, Richard E. and Vecchione, Michael. 1998. Lepidoteuthidae Pfeffer 1912. Lepidoteuthis grimaldii Joubin, 1895. The scaled squid. Version 01 January 1998 (under construction). in The Tree of Life Web Project,

Oh the Sea Monks, what a bizarre bestiary entity. The squid in the middle is supposed to be Architeuthis, but if these were based on squids and scaled they could possibly be our friend Lepidoteuthis. Of course, medieval artists tended to unnecessarily add scales to things and it would be out of range (Lepidoteuthis is tropical to sub-tropical) - so that's why I'm mentioning this here as a curiosity and nothing more.