Sunday, December 28, 2008

Strange Little Spirula

We can safely say that Spirula is an unusual coleoid cephalopod with a ventrally curved ("endogastric") planispiral shell, a vestigial radula, a photophore on the tip of the mantle, and oegopsid (cornea lacking) eyes (Warnke and Keupp 2005, Warnke 2007, Young 1996). That's about the limit of what can be said concisely...

The internal planispiral shell of Spirula, taken from scamazine's flickr. While this shell looks similar to those of the extinct ammonites, it curves ventrally rather than dorsally. This illustration shows how the shell fits into the animal.

Spirula was formerly considered to contain several species (see here) but now only S. spirula is valid. Spirula has two disjunct populations in the Atlantic and Indo-West Pacific Oceans; a molecular study of intraspecific variation suggests that individuals from Fuerteventura and New Caledonia are distinguished to a degree typical of separate species (Warnke 2007). The lifespan of the species is only ~20 months, making gene flow appear to be unlikely between the populations, but individuals closer geographically have not been investigated yet (Warnke 2007). With the sparse molecular and morphological data on Spirula variation, much more work needs to be done in order to (re-)establish multiple species.

So now that we haven't resolved the number of Spirula species, what exactly is it? Spirula certainly is a unique looking cephalopod and many sources (e.g. Wikipedia) place it in a monotypic order. Warnke and Keupp (2005) suggested that Spirula is the most basal decapodiform and further suggest that it can be used to study ammonite development! There are prominent morphological distinctions between the groups - ammonites tend to have shells with exogastric coils (heteromorphs are of course exceptions) and four prismatic layers while Spirula has a shell with endogastric coiling and two prismatic layers (Warnke and Keupp 2005). However, the initial chambers (protoconch) of ammonites and Spirula show morphological similarities and the mode of mineralization appears to be the same (Warnke and Keupp 2005). Warnke and Keupp (2005) cite prior preliminary molecular evidence to support the notion of Spirula being the most basal decapodiform and imply that similarities are plesiomorphies. Later molecular studies contradict the placement used by Warnke and Keupp (2005) but the issue of Spirula/ammonite similarities being homologies or homoplasies is still unclear. Spirula development will certainly be an interesting topic to investigate regardless.

More recent molecular analyses suggests that Decapodiformes consists of two orders (rather than 4) - Sepioidea (containing Sepiidae, Myopsida, Sepiolidae, Idiosepiidae, Spirula) and Teuthoidea (Strugnell et al. 2006). Spirula consistently grouped with Sepiidae and it should be noted that the proposed clade has the synapomorphies of sperm placement in females and the structure of the statoliths and tentacle clubs (Lindgren and Daly 2007, Strugnell et al. 2005). Their next nearest relative is either Myopsida or Sepiolidae, but either way Spirula is fairly nestled within the Decapodiformes (Lindgren and Daly 2007, Strugnell et al. 2005). The earliest probable member of Sepioidea is the early Carboniferous Shimanskya which is either a member of the Spirula-lineage or another taxa that convergently lost the nacreous layer (Strugnell et al. 2006). This fossil and the molecular clocks suggest that many clades are far older than previously thought, e.g. Sepiidae was previously thought to have originated in the Oligocene but apparently diverged around a couple hundred million years earlier (Strugnell et al. 2006)! Supporters of the Spirula-is-basal camp could argue that this incredible revision suggests that Spirula is not placed correctly, but other evidence in Strugnell et al. (2006) implies that many decapodiform lineages are similarly ancient. Undoubtedly this will undergo further revision, but it seems very unlikely that Spirula is the most basal of the Decapodiformes.

Despite apparently existing in large numbers, much of the life history of Spirula remains poorly known (Lukeneder et al. 2008). Study of shell isotopes suggests that the juveniles are born in waters >1000 m deep, migrate to warmer waters from 400-600 m as adults, and then migrate back down into deep and cold waters (Lukeneder et at. 2008). It's worth pointing out that Lukeneder et al. (2008) suggest that future studies of the Spirula life history can be applied to ammonites...


Lindgren, Annie R. and Daly, Marymegan. 2007. The impact of length-variable data and alignment criterion on the phylogeny of Decapodiformes (Mollusca: Cephalopoda). Cladistics 23, 464-476.

Lukeneder, Alexander et al. 2008. Stable isotopes (18O and 13C) in Spirula spirula shells from three major oceans indicate developmental changes paralleling depth distribution. Marine Biology 154, 175-182.

Strugnell, Jan et al. 2006. Divergence time estimates for major cephalopod groups: evidence from multiple genes. Cladistics 22, 89-96.

Strugnell, Jan et al. 2005. Molecular phylogeny of coleoid cephalopods (Mollusca:Cephalopoda) using a multigene approach; the effect of data partitioning on resolving phylogenies in a Bayesian framework. Molecular Phylogenetics and Evolution 37, 426-441.

Warnke, Kerstin. 2007. On the Species Status of Spirula spirula (Linne, 1758) (Cephalopoda): A New Approach Based on Divergence of Amino Acid Sequences Between the Canaries and New Caledonia. IN: N. H. Landman et al. (eds.). 2007. Cephalopods Present and Past: New Insights and Fresh Perspectives, 144-155. Springer.

Warnke, Kerstin and Keupp, Helmut. 2005. Spirula—a window to the embryonic development of ammonoids? Morphological and molecular indications for a palaeontological hypothesis. Facies 51, 60-65.

Young, Richard E. 1996. Spirulida Haeckel, 1896, Spirulidae Owen, 1836. Spirula spirula Linnaeus, 1758. Version 01 January 1996. in The Tree of Life Web Project,

Spirula can also be seen bearing gifts on my Christmas card. Unusually among cephalopods, the buoyant shell puts the animal in a head-down orientation (presumably because it had bottom-based ancestors?). I accidentally depicted it without skin of the mantle.

Saturday, November 29, 2008

The Leopard Seal

It was twelve feet long, of slender and graceful build, with a cruel, thin-lipped muzzle and formidable fangs in the front of the jaws; a curiously prehistoric looking beast, despite its beautiful coat. That it was a dangerous animal was proved by the fact that we found in its stomach large balls of hair, three inches in diameter - the remains of crab-eating seals that it had devoured. As these balls were of hair and not fur, it was evident that the sea-leopard's victims were not mere babies. I can well imagine that such a beast would give rise to sea-serpent stories, for, seen from a little distance rearing two or three feet out of the water, shooting forwards and retracting its head (a characteristic movement) it would resemble a serpent more than a mammal.

The sporadic media appearances of Hydrurga leptonyx cast them as fearsome, penguin-eviscerating, reptilian monsters. I don't believe in monsters - behind reputations like these are animals which are not fully or properly understood. Hydrurga and other Antarctic pack-ice seals happen to be the most poorly known phocids for obvious logistic reasons (van den Hoff et al. 2005). I wouldn't say that there's a dearth of information on Hydrurga compared with some taxa, but it is a rather cryptic species and a lot of its basic biology needs clarification.

I feel obliged to add how annoyed I am by lazy statements like "such and such taxa is behind sea serpent reports" - I can't recall any southern hemisphere "sea serpents" sounding anything like Hydrurga.

A Creative Commons photo of a leopard seal, from Crouchy69's Flickr

Hydrurga is a phocid ("earless seal") and a member of the clade Monachinae; molecular evidence suggest that monk seals (Monachini) are basal members of the clade which also includes a sister group of elephant seals (Miroungini) and other southern seals (Lobodontini) (Arnason et al. 2006, Higdon et al. 2007). The relations within Lobodontini are more contentious, cranial morphology (e.g. those bizarre teeth) suggests a close relation between Hydrurga and Lobodon ("crabeater" seals) but molecular evidence suggests Hydrurga is allied with Leptonychotes (Weddell seal) instead (Davis et al. 2004, Arnason et al. 2006, Higdon et al. 2007). Lobodontini seems to have undergone a rapid radiation, perhaps a mere 7 million years ago, which may have caused this phylogenetic confusion (Higdon et al. 2007, Davis et al. 2004). With fossils taken into account, it appears that the bizarre late Miocene/early Pliocene Acrophoca is the closest relative of Hydrurga (Walsh and Naish 2002) - more information on the bizarre "swan-necked seal" can be found on Darren's old blog.

Animalian size is an inevitable topic for this blog and I feel obliged to note that Hydrurga is probably* the largest lobodontin and is certainly among the larger pinnipeds. It is common for sources to claim that Hydrurga is sexually dimorphic and typical figures for maximum size are 3.4 m (11') and 450 kg (990 lbs) for males and 3.6 m (12') and 590 kg (1300 lbs)** for females (Reeves et al. 2002). I'm of the opinion that maximum measurements are highly misleading outside the context of averages and cannot justifiably be used except in the most dire of data-deficient situation, e.g. with some ziphiids. Furthermore, sexual dimorphism was not observed in the 77 individuals measured by van der Hoff et al. (2005) and the previously reported "dimorphism" is probably insignificant enough to have been an artifact. It's unfortunate that van der Hoff et al. (2005) do not establish an average STL with their data - however we do know that Cave and Bonner (1987) consider a 2.52 m (8'3") STL specimen immature and Visser et al. (2008) use 3 m (~10' - EL?) as an average. van der Hoff et al. (2005) do establish equations and methods for assessing body condition and mass of Hydrurga, which comes in handy since this species has a rather high mortality rate when anesthetized.

* van der Hoff et al. (2005) state that among Antarctic seals only Mirounga reaches larger sizes, but I'm not completely sure if Leptonychotes weighs more on average.
** Bizarrely, these maximum figures are understatements rather than exaggerations - Higgens et al. (2005) state that the maximum weight is 650 kg (1432 lbs). Muir et al. (2006) note that the leopard seal which killed Kirsty Brown in 2003 was estimated to be a staggering 4.5 m (14'9") in length. Assuming that the STL was something like 4 m, the seal would have been ~825 kg (1800 lbs) in good condition. Muir et al. (2006) assume this animal was female, but there is no strong evidence for sexual dimorphism.

Skull of "Ogmorhinus" (= Hydrurga) leptonyx from The Royal Natural History by Richard Lydekker and Philip Lutley Sclater.

As illustrated above, Hydrurga has post-canine teeth with cusp complexity second only to Lobodon. While the cusp development may not be homologous*, both genera use their teeth to strain krill from the water (Nowak 1999). While Nowak (1999) stated that krill takes up 45% of Hydrurga's prey biomass, this has been contradicted; Walker (1998) found krill from only one scat in 45, Hall-Aspland and Rogers (2004) reported krill in 4% of scats and some male leopard seal stomachs, unpublished isotope data from Hall-Aspland further suggested krill is not a major food item and diving data from Kuhn et al. (2006) suggests that krill is not consumed by juveniles in winter (as was previously thought). It doesn't make sense for Hydrurga to have highly specialized dental morphology for capturing krill on rare occasions, so there must be instances where euphausiids play a critical role in the diet of the species...

* Since neither Leptonychotes or Acrophoca has these accessory cusps. Alternately, the cusps could have been lost several times.

Leopard seals are well known for taking large, homeothermic and occasionally human-sympathetic prey. In one recorded instance, Hydrurga exhibited predatory behavior on a human who was ultimately killed (Muir et al. 2006). Evidence of Hydrurga predation can be found physically in the form of scars on many adult Lobodon (Nowak 1999) - the number of predation attempts must be staggering since there are millions of Lobodon. Hall-Aspland and Rogers (2004) looked at the summer diet of east Antarctic Hydrurga and found that it mostly consisted of Adelie penguins (Pygoscelis adeliae) with Lobodon, fish, amphipods and krill as supplements. Prior studies cited by that paper of Antarctic peninsula Hydrurga found that gentoo penguins (Pygoscelis papua), macaroni penguins (Eudyptes chrysolophus), Antarctic fur seals (Arctocephalus gazella), fish, squid and krill were taken during summer but only krill and fish were consumed during winter. So it looks like krill is important for at least some populations, or individuals in certain locations. Even with a dentition having specializations for taking krill, Hydrurga is capable of an impressively broad diet which probably capitalizes on local abundances. One article cited by Forcada and Robinson (2006) hypothesized that the catholic diet of the leopard seals allow them to have a flexible breeding strategy with less seasonality and a shorter breeding cycle.

While Hydrurga occupies a rather high trophic position and can even prey on other pinnipeds (normally as juveniles), it is not immune from predation. One abstract mentions a tiger shark (Galeocerdo cuvier) caught off Rio de Janeiro with a Hydrurga in its stomach, but an individual from that far north was probably a juvenile in poor condition*. Authors in the past have suggested that Orcinus can prey on Hydrurga and this has been observed by Visser et al. (2007). Since Hydrurga is a rather cryptic species that spends far less time on ice floes than Lobodon, it seems likely that this is a rather rare event.

* Hydrurga observed in N. Argentina were juveniles 2-2.5 m in length and were in poor conditions, if not as corpses (Rodriguez et al. 2003). The presence of juveniles in these areas may be due to food competition from adults in winter (Rodriguez et al. 2003).

Understanding the ecological significance of Hydrurga is complicated by the difficulty of estimating how many individuals there are. Recent aerial surveys between 64 E and 150 E off east Antarctica observed only 29 Hydrurga individuals and extrapolated a populations of 3700-23,400 (Southwell et al. 2008). A previous survey in a somewhat comparable area (90 E to 160 E) estimated 68,000 animals and Southwell et al. (2008) noted a concurrent acoustic survey which detected Hydrurga in 98% of 54 acoustic sites yet only saw in 2%. Southwell et al. (2008) appear to conclude that Hydrurga is cryptic rather than very rare and it looks like a more robust survey method is needed. Somehow, prior sources came up with population figures in the hundreds of thousands, e.g. 400,000 in Nowak 1999. Hydrurga in Tasmania were assumed to be vagrants in the past, but sightings between July and November (i.e. winter and spring) have included animals in good condition so they're a natural part of the fauna (Rounsevell and Pemberton 1994). Other temperate areas should be investigated to determine if Hydrurga is a cryptic resident rather than an occasional straggler.

I think that's about it for Hydrurga; while this isn't a poorly known species, many basic aspects of its biology need clarification. The species does lend itself to some interesting videos:


Arnason, Ulfur et al. 2006. Pinniped phylogeny and a new hypothesis for their origin and dispersal. Molecular Phylogenetics and Evolution 41, 45–354

Cave, A. J. E. and Bonner, W. N. 1987. Facial asymmetry in a leopard seal. Br. Antarct. Surr. Bull. 75, 67-71. Available

Davis, Corey S. et al. 2004. A phylogeny of the extant Phocidae inferred from complete mitochondrial DNA coding regions. Molecular Phylogenetics and Evolution 33, 363–377

Forcada, Jaume and Robinson, Sarah L. 2006. Population abundance, structure and turnover estimates for leopard seals during winter dispersal combining tagging and photo-identification data. Polar Biol 29, 1052–1062

Hall-Aspland, S. A. and Rogers T. L. 2004. Summer diet of leopard seals (Hydrurga leptonyx) in Prydz Bay, Eastern Antarctica. Polar Biol 27, 729–734

Higdon, Jeff W et al. 2007. Phylogeny and divergence of the pinnipeds (Carnivora: Mammalia) assessed using a multigene dataset. BMC Evolutionary Biology 7:216

van den Hoff, John et al. 2005. Estimating body mass and condition of leopard seals by allometrics. Journal of Wildlife Management 69, 1015-1023

Kuhn, Carey E. et al. 2006. Diving physiology and winter foraging behavior of a juvenile leopard seal (Hydrurga leptonyx). Polar Biol 29, 303–307

Muir, Shona F. et al. 2006. Interactions between humans and leopard seals. Antarctic Science 18, 61-74.

Nowak, Ronald M. 1999. Walker's Mammals of the World, Sixth Edition. John Hopkins University Press.

Reeves, Randall R. et al. 2002. National Audubon Society Guide to Marine Mammals of the World. Alfred A. Knopf, New York.

Rodriguez, Diego et al. 2003. Occurrence of Leopard Seals in Northern Argentina. LJAM 2, 51-54

Rounsevell, D. and Pemberton, D. 1994. The Status and Seasonal Occurrences of Leopard Seals, Hydrurga leptonyx, in Tasmanian waters. In: Australian Mammalogy

Southwell, Colin et al. 2008. Uncommon or cryptic? Challenges in estimating leopard seal abundance by conventional but state-of-the-art methods. Deep-Sea Research I 55, 519–531

Visser, Ingrid N. et al. 2008. Antarctic peninsula killer whales (Orcinus orca) hunt seals and a penguin on floating ice. Marine Mammal Science 24, 225–234

Walker, T. R. et al. 1998. Seasonal occurrence and diet of leopard seals (Hydrurga leptonyx) at Bird Island, South Georgia. Antarctic Science 10, 75-81

Walsh, Stig and Naish, Darren. 2002. Fossil seals from the late Neogene deposits in South America: A new Pinniped (Carnivora, Mammalia) assemblage from Chile. Palaeontology 45, 821-842

No wonder this took me so long to write...

Saturday, November 15, 2008

What is Gigantosaurus?

It is not Giganotosaurus, a late Cretaceous carcharodontosaurid theropod rivaling Tyrannosaurus for size. As Google shows us, the confusion is widespread.

My interest in Gigantosaurus stemmed from this illustration:

From the 1932 edition of the Meyers Blitz-Lexikon

It is one of the more ridiculous I've seen, just look at those hind limbs! But hey, at least it isn't wallowing in water. I'm not sure how long that locomotive is (I'll have to ask my dad)* but if we assume the human is 1.8 m tall, the dinosaur's head is 4 m long (13'), the length of the forefoot is 2.3 m (7.5'), the highest part of the body is 12 m (40') off the ground and the total length is probably over 60 meters (200'). This is even bigger than Amphicoelias fragillimus according to Carpenter (2006) and it could be the largest sauropod ever depicted. From the original German source I was able to discern that this was an East African sauropod and it was apparently larger than Diplodocus.

*About 36.5 feet long and 15 feet high. This puts the guy in the 1.7's somewhere and makes the dinosaur somewhat smaller but, it's still crazily oversized!

There is currently no valid dinosaur species named "Gigantosaurus", so what does this illustration depict? The mention of East Africa directs us to the German Tendaguru expedition of 1909-1913 where Fraas described Gigantosaurus africanus and G. robustus. "Gigantosaurus" was argued to be occupied by a later author who moved the dinosaurs into the genus Tornieria and changed their specific names (T. africana, T. robusta); later T. africana got moved to the genus Barosaurus and T. robusta became Janenschia robusta (Remes 2006). A third member of the genus (G. dixeyi) was named 20 years after the others and now occupies the genus Malawisaurus (Jacobs et al. 1993). The illustration did not mention a species but I don't think it's so much a composite of the different genera as it is a product of active imagination and limited understanding of sauropod morphology. And probably scaling as well.

A 1912 (cropped?) drawing of "Gigantosaurus" by Heinrich Harder, taken from here. Note the tail stretching into the horizon and the very wrong hands. That link does specifically address Janenschia, by the way

As for the former members of "Gigantosaurus", Tornieria is now a valid genus of diplodocine diplodocid diplodocoid once again, it appears to be the sister group to Barosaurus + Diplodocus (Remes 2006). Contrary to the scale the drawing suggests this was not a mega-sauropod or super-sauropod, femoral length suggests a similar length as Diplodocus (Remes 2006). As far as the shape of the animal, it had elongated cervicals like Barosaurus but retained the trait of proportionally short hindlimbs with a tibia:femur ratio similar to Apatosaurus (Remes 2006). From a biogeographical perspective, Tornieria is unusual since it is the only diplodocid from the upper Jurassic of Gondwanaland (aside from other "Barosaurus" remains) and suggests that other species remain to be discovered (Remes 2006).

A 1930 illustration of "Gigantosaurus africanus" (= Tornieria africana) and Diplodocus. Note that in real life, the two dinosaurs are actually about the same size! The Hairy Museum of Natural History alerted me to the existence of this photo, which was put online by LIFE and Google.

The other two genera are quite distinctive from the diplodocid as they appear to be titanosaurids. Janenschia is from either the late Jurassic or early Cretaceous and while it has been classified as a basal titanosaurid in the past (and sometimes camarasaurid), it has not been subjected to recent phylogenetic analyses (O'Leary et al. 2004). Whatever it is, Janenschia appears to grow at a blazing 663-993 kg/year (titanosaurids seem to have been fast growers) and could have reached its adult size of ~14 tonnes in 20-30 years (Lehman and Woodward 2008). That's probably slightly more than Diplodocus (or Tornieria?) weighed, but come on, terrestrial mammals have attained similar sizes. Malawisaurus is from the late Cretaceous and is clearly a titanosaurid (based on strongly procoelous anterior caudals), and is in fact a relatively basal one (Jacobs et al 1993, Wilson 2006). One cervical vertebrae as described by Jacobs et al. (1993) was 41 cm tall (with spine and all) and I'm not thinking that belonged to a very large sauropod. Some pages claim a length of 9 m, but didn't explain where they got their figure from.

So what is the original "Gigantosaurus" which caused the name changing? Gigantosaurus megalonyx was described by Seeley in 1869, quite early in our knowledge of sauropods since Astrodon was first described in 1865. From the scraps of information I've gathered, it was described from sacrals, a radius, tibia and fibula and unsurprisingly was not given a sufficient description. It is currently listed as indeterminant and probably the same as other sauropods from the area. So while it was unfortunate that a nomen dubium did not allow for the use of Gigantosaurus, the genus would have split up anyways - plus, it wasn't the most appropriate name in the world.

I also found that the name "Gigantosaurus" has been attached to this illustration:

It was taken from the Copyright Expired page and was apparently drawn by Vincent Lynch in 1914 and published in Scientific American. It looks like a scene from the Lost World, which was allegedly filmed over a decade after this was drawn. Could these be an early instance of the dinosaur-running-amok-in-a-city theme? Whatever the case, it also seems to be drawn horribly out of scale. Interestingly, this image (and its less exciting cousin) has also been labeled as the brachiosaurid Pelorosaurus. That genus has no known synonymy with Gigantosaurus, and is presumably either an informal nickname or a mistake of the website author.

This post originated from a comment at The World We Don't Live In. It was originally intended to be a "picture of the day"-type post but quickly spun out of control.


Carpenter, Kenneth. 2006. Biggest of the big: A critical re-evaluation of the mega-sauropod Amphicoelias fragillimus Cope, 1878. Paleontology and Geology of the Upper Jurassic Morrison Formation. New Mexico Museum of Natural History and Science Bulletin 36, 131-138. Available

Jacobs, Louis L. et al. 1993. New material of an early Cretaceous titanosaurid sauropod dinosaur from Malawi. Palaeontology 36, 523-524.

Lehman, Thomas M. and Woodward, Holly N. 2008. Modeling growth rates for sauropod dinosaurs. Paleobiology 34, 264-281.

O'Leary, Maureen A. et al. 2004. Titanosaurian (Dinosauria: Sauropoda) remains from the continental intercalaire" of Malawi. Journal of Vertebrate Paleontology 24, 923-930.

Remes, Kristian. 2006. Revision of the Tendaguru sauropod Tornieria africana (Fraas) and its relevance for sauropod paleobiogeography. Journal of Vertebrate Paleontology 26, 651–669

Wilson, J. A. 2006. An Overview of Titanosaur Evolution and Phylogeny. Actas de las III Jornadas sobre Dinosaurios y su Entorno. 169-190. Available

Monday, November 10, 2008


One of my classes assigns write-ups on invertebrate peer-reviewed literature, so I figured that I might as well cannibalize and extend my original report to post it here. I also gave a short talk on Remipedia that a "lucky" few were subjected to - let's hope this is a more coherent format.

I have an inexplicable intellectual attraction to relictual organisms, making remipedes fascinating to me despite my relative unfamiliarity with Crustacea. I should point out that the popular conception of a "crustacean" is essentially synonymous with Decapoda (shrimp, lobsters, crabs, etc.) and excludes all of the other various clades of mandible-bearing arthropods with two pairs of antennae, two maxillae pairs and division into tagma (e.g. cephalothorax, pereon, pleon). Remipedes are the exception to the latter trait and simply possess a head with a long, homonomously segmented trunk somewhat reminiscent of myriapods (especially chilopods = centipedes), onychophorans and some polychaete annelids. It was long assumed that arthropods evolved from a long-bodied annelid-like ancestor so early remipede morphological workers (i.e. those in the 80's and 90's!) assumed that remipedes lay at the base of the crustacean phylogenetic tree; one worker even used remipedes to root the crustacean phylogenetic tree and placed them in a more basal position than Burgess Shale arthropods (Odaraia, Canadaspis) (Schram and Koenemann 2004). As we'll see later, the only modern consensus about remipede relations is that they aren't at the base of the "crustacean" family tree...

The unpigmented, eyeless, marine cave-dwelling members of class Remipedia were first discovered off the Bahamas in 1979* and were described by Yeager (1981). Remipedes were later found in other Caribbean locales such as the Yucatan peninsula, the Turks and Caicos, Cuba, and the Canary islands - incredibly they've also been found in roughly antipodal Western Australia as well (Yager and Humphreys 1996). Yager (1981) suggested that a remipede had in fact been described before: the enigmatic Carboniferous Tesnusocaris (described in 1955) which also possessed homonomous segments with paddle-like appendages. Emerson and Schram (1990) re-described Tesnusocaris with both a pair of uniramous ventrolateral appendages used for swimming and a midventral pair used for sculling; the authors hypothesized that this two appendage pair state is a "missing link" between uniramous and biramous appendages. Koenemann et al. (2007a) found some aspects of the authors' reconstruction questionable** (the two limb pairs per segment?) but still used their description of Tesnusocaris as an outgroup for their phylogeny of modern remipedes (Nectiopoda) - despite potential weirdness Tesnusocaris still did have distinctively remipedian head appendages. The Mazon Creek assemblage of Illinois (home of Pohlsepia and Tullimonstrum) yielded the remipede Cryptocaris which also has the three pairs of prehensile cephalic appendages (maxillule, maxilla and maxilliped), but was not complete enough for analysis by Koenemann et al. (2007a). So, it looks like Remipedia has an even worse ghost lineage syndrome than octopuses.

* Another class of possibly primitive crustaceans, Cephalocarida, was discovered off Long Island Sound in 1953. Granted, remipedes are ~1.5-4.5 cm in length and cephalocarids are 4 mm, but discovering new classes still sounds like a major surprise.
** They cite a later paper from the same authors in 1991 from the Proceedings of the San Diego Society of Natural History. I'm assuming that it's a more thorough version of their Science article.

Not much is known about the reproduction, life history and behavior of remipedes; what is known is fascinating, if somewhat contradictory. Being blind, remipedes have a gigantic olfactory apparatus and use their second pair of antennae to drive currents past the "fields of aesthetascs" on their first antennae pair; their ability to detect low odor concentrations has been confirmed by observations of their quick attraction to dead fish (Fanenbruck et al. 2004). Remipedes have been observed to be slow swimmers but they are not strict scavengers, they have raptorial mouthparts including a fang-like first maxillae and have been observed engaging in predatory behavior (Kohlhage and Yager 1994, Fanenbruck et al. 2004). The maxillule fangs connect to glands which are presumably involved in injecting prey; while empirical evidence of injection doesn't exist, the probable mechanics of injection have been worked out (van der Ham and Felgenhauer 2007). The potentially injected substance is apparently an oxygen-carrying respiratory pigment (!) - another substance capable of turning the hemocyanin-like compound into a harmful phenoloxidase has yet to be discovered (van der Ham and Felgenhauer 2007). In case you're diving in obscure marine caves, don't worry about remipede bites as they apparently have no adverse affect on people (Koenemann et al. 2007b). While the aforementioned evidence seemingly suggests that remipedes are sluggish marine centipede analogues - lab observations of Speleonectes indicate that they spend almost all of their time (>99%) filter feeding (Koenemann et al. 2007b). Koenemann's website has a video summary (warning, 50 megabytes) of 2-3 months of observed behavior, including some of the rare instances of predation (3!). From the video, it can be noted that the thoracic appendages are always moving (even at rest) and remipedes are capable of a fast "snake-like" strikes and coiling (Koenemann et al. 2007b). Even though these remipedes aren't very big animals (~4 cm), I would hesitate in calling them sluggish (note that some parts of the video are at 5x). It seems likely that remipedes spend most of their time filter feeding in the wild as well and engage in facultative predation/scavenging whenever something comes their way - the lack of these behaviors in the lab could be artifacts due to the environment and/or the (relatively) high abundance of potential prey.

So, what are remipedes?

Despite looking like hypothetical ancestral arthropods, remipedes are obviously quite specialized. If we ignore their confusing mosaic of morphological traits for the time being, molecular evidence gives us a wide range of opinions on their placement. Regier et al. (2005) suggested close kinship with cephalocarids and a somewhat more distant relation with branchiopods (both viewed as morphologically "primitive"), oh and all of those groups were in a clade containing hexapods, i.e. insects and kin! The authors note that all of the members probably had ancestors either near-shore or in marginal (read: really weird) marine habitats, possibly the result of competition from other crustaceans, myriapods and chelicerates (Regier 2005). Cook et al. (2005) used mtDNA to place remipedes in a derived clade with Collembola (hexapods!) - mind you in thus study both hexapods and crustaceans were paraphyletic! Since none of the other studies reach any consistent placement, let's look at morphology.

Unexpectedly, remipedes have an order of magnitude more neurons than other taxa like branchiopods and maxillopods and their complex brains resemble those of malacostracans and hexapods (Fanenbruck et al. 2004). Also unexpected are the recently discovered larvae of remipedes, which happen to be non-feeding (lecithotrophic) in a manner similar to malacostracans such as euphausiaceans and dendrobranchiates (Koenemann et al. 2007c). Remipedes larvae share many traits with the lecithotrophic malacostracans but differ in having three pairs of uniramous cephalic limbs, biramous trunk limbs and caudal rami developing on an anal somite rather than a teslon (Koenemann et al. 2007c). The first trait is especially odd since you would expect a maxilliped to resemble a trunk appendage during development, but this somehow is not the case. Convergence can't be ruled out of course, but the coincidence of similar brain morphology and occasionally similar development is interesting (unless the two are somehow connected). Koenemann et al. (2007c) echo a previous study which tenuously concluded that remipedes, cephalocarids and "most of the maxillopodans and malacostracans" form a clade to the exclusion of other crustaceans.

The previous study Koenemann et al. (2007c) are referencing (Schram and Koenemann 2004) used extinct and extant arthropods (including Tesnusocaris) in their analysis. One of the characters united remipedes with Eucrustacea was gonopore placement on the 6th through 8th thoracic segments - I'm not sure how this was coded in for remipedes. Interestingly, even this analysis found insects to lay within the group traditionally known as "crustaceans" - could it be that molecular and morphological camps are finally starting to agree?

This of course isn't everything on remipedes, but it should at least give an idea of the pioneering work being done on this fascinating group. Well, this took far too long to write, I've got obligations to fulfill like crazy...


Cook, Charles E. et al. 2005. Mitochondrial genomes suggest that hexapods and crustaceans are mutually paraphyletic. Proc Biol Sci. 272, 1295–1304

Emerson, Michael J. and Schram, Frederick R. 1990. The Origin of Crustacean Biramous Appendages and the Evolution of Arthropoda. Science 250, 667-669

Fanenbruck, Martin et al. 2004. The brain of the Remipedia (Crustacea) and an alternative hypothesis on their phylogenetic relationships. PNAS 101, 3868-3873.

van der Ham, Joris L. and Felgenhauer, Bruce E. 2007. The functional morphology of the putative injecting apparatus of Speleonectes tanumekes (Remipedia). Journal of Crustacean Biology 27, 1-9

Koenemann, Stefan et al. 2007a. Phylogenetic analysis of Remipedia (Crustacea). Diversity & Evolution 7, 33–51

Koenemann, Stefan et al. 2007b. Behavior of Remipedia in the Laboratory, with supporting Field Observations. 2007. Journal of Crustacean Biology 27, 534-542

Koenemann, Stefan et al. 2007c. Post-embryonic development of remipede crustaceans. Evolution & Development 9, 117-121

Kohlhage, Klaus and Yager, Jill. 1994. An Analysis of Swimming in Remipede Crustaceans. Philosophical Transactions: Biological Sciences 346, 213-221

Regier, Jerome C. et al. 2005. Pancrustacean phylogeny: hexapods are terrestrial crustaceans and maxillopods are not monophyletic. Proc Biol Sci. 272, 395-401.

Ruppert, Edward E. et al. 2004. Invertebrate Zoology: A Functional Evolutionary Approach. Seventh Edition. Thomson, Brooks/Cole, United States.

Schram, F. R. and Koenemann, S. 2004. Are crustaceans monophyletic? In J. Cracraft and M. J. Donaghue (eds.). Assembling the Tree of Life. Oxford University Press, Oxford, pp. 319-329.

Yager, Jill and Humphreys, W. F. 1996. Lasionectes exleyi, sp. nov., the First Remipede Crustacean Recorded from Australia and the Indian Ocean, with a Key to the World Species. Invertebrate Taxonomy 10, 171-187.

Yager, Jill. 1981. Remipedia, a new class of Crustacea from a marine cave in the Bahamas. Journal of Crustacean Biology 1, 328-333.

Wednesday, October 22, 2008

The Sperm Whale's Jaw; or Sorry, Gerald Wood

In length, the Sperm Whale’s skeleton at Tranque measured seventy-two Feet; so that when fully invested and extended in life, he must have been ninety feet long; for in the whale, the skeleton loses about one fifth in length compared with the living body. Of this seventy two feet, his skull and jaw comprised some twenty feet, leaving some fifty feet of plain back-bone. Attached to this back-bone, for something less than a third of its length, was the mighty circular basket of ribs which once enclosed his vitals.

- Excerpt from "Measurement of The Whale's Skeleton" - Chapter 103 of Moby-Dick; or The Whale by Herman Melville

Earlier in the chapter, Melville/Ishmael tells us that a Sperm Whale of the largest magnitude measures between eighty-five and ninety feet long and by his reckoning weighs at least 90 tons, or as much as a village of 1,100. By my own reckoning in this previous post, I pegged a 19.5 meter (67'11") bull at around 90 tonnes (~100 tons) and I estimated a 27.5 m (90 foot) bull to weigh a stupefying 210 tonnes/230 tons. If the whale of Tranque was as big as Melville/Ishmael claimed*, it would have rivaled the largest blue whales (Balaenoptera musculus) for size.

* How can a 72 foot skeleton possibly correspond to a 90 foot whale? A ~20 foot mandible would be at a 1:3.6 ratio to the skeleton length, which agrees well with the 1:3.8 m estimate below. The modified Gore et al. (2007) formula (see further below) estimates the whale's length at 21.4 meters (70 feet) using a skull length of 6 meters - could this have been based off a real specimen?

Skeletons of gigantic sperm whales (Physeter macrocephalus sometimes P. catodon) do not currently exist in museums as far as I know (if they ever did) and the only possible evidence of colossal bulls rests in preserved mandibles. In my previous post I discussed a 5 meter (16'4.75") mandible in the British Museum which Wood (1982) estimated to have come from a 25.5 meter bull; I extended Wood's reasoning to estimate a 5.5 meter mandible from the Nantucket Whaling Museum to correspond with a 27.5 meter bull. Wood (1982) noted that a 14.7 m whale had a mandible:body length ratio of 1:6.2 and a 16.28 m whale had a ratio of 1:5.4; he extended the graph to estimate a 1:5.1 ratio for his monstrous mandible and I extended it further to 1:5 for the even bigger mandible. I'm now quite certain that all of these estimated lengths and ratios are, in fact, hogwash.

The book, The Marine Mammals in the Anatomical Museum of the University of Edinburgh (courteously digitized by Google) notes that a 15'10" (4.83 m) mandible probably corresponded with a sperm whale 60 feet (18.3 m) in length - let's assume that this estimate is accurate. This whale would have had a mandible:body length ratio of around 1:3.8; if we assume there is little allometric change, the 5 m mandible would thus correspond to a 19 meter bull and the 5.5 meter mandible would correspond to a 21 meter bull. Given that the record length sperm whale was 20.7 meters in length, I'd say that these seem like plausible figures. But is this accurate?

Unfortunately, there is a lack of data on sperm whale mandible length:body length. However, if we assume that the length of the mandible is nearly the same as that of the skull, this gives us more data to work with. I'll give a range of estimates just to be safe, the first will assume a 1:1 mandible:skull ratio and the second will assume that the mandible is roughly 90% of the skull's length. Data on head length should be avoided since soft tissue can make it considerably longer. Gordon (1991) used data on sperm whale spermaceti length in comparison to body length to come up with a formula which Gore et al. (2007) in turn used to estimate body length from skull length. It is as follows:

Total body length = 9.75 − 0.521 (SL) + 0.068 (SL^2) + 0.057 (SL^3)

Where SL = skull length

This is quite a bit more complicated than what I was using as it appears to take allometry into account (where I previously sorta ignored it). Here is what happens when we plug in numbers:

5 meter mandible:
5 meter skull = 15.97 meters (52 feet)
5.5 meter skull = 18.42 meters (60 feet)

Wood's 25.5 meter estimate was probably off by 138% to 160%.
Using a 1:3.8 ratio gives estimates that are off by 114% to 119%

5.5 meter mandible:
5.5 meter skull = 18.42 meters (60 feet)
6.0 meter skull = 21.384 meters (70 feet)

My 27.5 meter estimate (based off of Wood's allometry) is probably off by 128% to 149%
Using a 1:3.8 ratio gives estimates that are off by 113% and 106%

A 6 meter skull was also present in the (fictional?) Tranque whale, and Melville/Ishmael's estimate for skeletal length (72 feet) is off by 103%. The estimation for body length in the flesh is off by 129%.

Hard data is needed on the mandible:body length ratio in sperm whales so these equations can be refined. I'd say that as is, the data suggests that the 5 and 5.5 meter mandibles do not correspond with bulls greatly exceeding the known 20.7 meter record. It could be possible that these mandibles are the product of abnormal growth analagous to acromegaly in humans (as suggest by Alan Hazen). There is currently no non-anecdotal evidence of bulls greatly exceeding 20.7 meters and while specimens somewhat larger than this probably existed in the past due to a larger average length, sperm whales do not appear to rival blue whales for the title of the largest animal ever to have lived.

I was planning to make this an addendum, but it quickly got out of hand...

For information on sperm whale jaw oddities, see the post at Tetrapod Zoology


Gordon, Jonathan C. D. 1991. Evaluation of a method for determining the length of sperm whale (Physeter catodon) from their vocalizations. Journal of Zoology 224, 301-314

Gore, M. A. et al. 2007. Sperm whale, Physeter macrocephalus, stranding on the Pakistani coast. J. Mar. Biol. Ass. 87, 363–364

Wood, Gerald. 1982. Guinness book of Animal Facts and Feats. Guinness Superlatives, Middlesex.

Sunday, October 19, 2008

Tales of the Incredibly Outsized

Outsized animals have been a mainstay of this blog in the past, notable posts can be found here, here and here. While those posts appear to have been relatively popular*, I don't intend to write any more moderately fluffy posts rattling off trivia. I think it would be more worthwhile to determine if animals can reach the lengths claimed for them, see if there are widespread patterns in how outsized animals can be and suggest widespread implications. And I'll probably talk about a lot of other stuff, I never know where these things wind up.

*I guess this isn't saying much, the Nature Blog Network tells me this blog is abysmally unpopular. Oh well...

Unfortunately, there are a lot of issues that need to be brought up before we can start having fun (sic). I'm sure everyone is aware of scaling, but just for the heck of it: when Animal A is twice the length (2^1) of Animal B it will have four times the surface area (2^2) and eight times the mass (2^3). Muscle and bone strength depend on area, so we can anticipate animals being proportionally bulkier at larger sizes - among other allometric changes. Quickly and generally, larger animals exist at lower densities, have larger home ranges, are subjected to less predation, have lower metabolic rates, change temperatures more slowly, are more efficient and faster locomotors, et cetera (Peters 1983). It's an absolutely fundamental measurement and one source stated that "[b]ody size is the single most important axis of biodiversity" (Brown et al. 2007). This post isn't looking at body size in the context of biodiversity, but it seems reasonable to suggest that body size can be very important from an intraspecific perspective.

Despite size being such a useful, nay, pivotal measurement, "fish story" data plagues many sources that should have been substantially more critical. As I discussed previously, the wels catfish (Siluris glanis) is frequently stated to reach 5 meters and ~300 kg* (e.g. Stone 2007); in reality, specimens over 2 meters are noteworthy and the largest accurately measured/well supported specimen was 2.78 m and 144 kg (see here). Statements such as "the wels grows up to 5 m long" and "the wels averages 1.4-1.6 m long" (wiki) create vastly different impressions of the wels' size - the former should outweigh the latter by about 40 times, for one thing. In the past I have also discussed the size of the green anaconda (Eunectes murinus); I can't help but note that some sources state that the anaconda is 6-10 m (~10-33 feet) long (e.g. Burnie and Wilson 2001). A survey of 1000 anacondas did not turn up any snakes longer than 17 feet (5.18 m) and there is still a reward out for any snake longer than 30 feet (see article). How do you report extreme figures as if they are the norm of the species? Unless there is an extreme lack of data (e.g. some ziphiids) there's no reason to give a maximum length instead of the average adult length and mass.

* I'm not sure where this figure originated, it may have been rounded from a 4.6 m/327 kg claim made by Kessler in 1856. The weight is abnormally low (a wels that length should be over twice as heavy), which strongly indicates a fabrication. Alternately, the length could have come from a sturgeon (apparently they can be confused for wels - see here) and the weight could have been a crude estimate (perhaps a linear scaling).

So... why look at outsized animals at all? By definition they will not be numerous, but they may have exaggerated ecological impacts. Larger and older black rockfish (Sebastes melanops) produce greater numbers of larvae over broader periods of time than smaller individuals; their larvae will also be larger, faster growing and more starvation resistant (Birkeland 2005). This will probably not be the norm amongst fish species which reproduce only once or twice in their life or have high turnover rates (Birkeland 2005). Within species there may be significant ecological differences between size classes, e.g. large parrotfish will excavate substratum while smaller ones will not (Birkeland 2005). That's an extreme situation of course, but if an animal is much larger than average it will have to be doing something different in order to sustain itself. Outsized individuals will have the ability to feed on larger prey items and if some individuals get as massive as claimed (i.e. several times heavier than average) they may even be at a distinct trophic level. While the impacts of the average individual are important, looking at all the individuals on the bell curve can give a more complete understanding.

Human exploitation tends to "favor" large individuals and this can cause the average size of an organism to decrease. If there is good historical data of individuals much larger than average, then it may indicate that modern individuals aren't at previous levels of diversity. This is assuming that outsized individuals have a genetic basis since humans with pituitary gigantism tend to have many other medical issues and rather short lifespans. Anyways, let's look at sperm whales (Physeter macrocephalus), which I mentioned in the past.

Wood (1982) states that the average adult bull* is 14.93 m (49 feet) in length, but another source (mentioned here) states that the average length is 16 m (52.5 feet). The difference doesn't look like much, but a 107% difference in length means at least a 123% difference in weight. That's probably a difference of thousands of tons of prey annually. Photographic measurement of 13 bachelor males off New Zealand ranged from 11.7-15.8 m (med. 14.2 m) and 7 males off California ranged from 14.7-18.2 m (med. 16.4) (Jaquet 2006). The sample size is too small to make a fuss over the figures (probably slight overestimates) but there do appear to be regional differences in sperm whale size (and behavior, proportions, markings, etc) (Jaquet 2006). This complicates things since the largest male may not have been as outsized as predicted if it came from a rather large population - a rough estimate will have to do. The record was a 20.7 m (67'11") bull caught off South Georgia in the '48-'49 season, by the way (Wood 1982). Taking the high road (16 m avg.), the record specimen was was about 129% longer than average and at least 217% more massive. So how massive are sperm whales? Wood (1982) estimates his average bull to weigh 36 tonnes (~40 tons) (I have no idea where this figure derives from) and mentions large bulls which were weighed in full at 18 m/53 tonnes (59 feet/58 tons) and 16 m/40 tonnes (52'6"/44 tons). Using these figures, the record bull would weigh at least 96, 81 or 87 tonnes. Wood (1982) also mentioned an 18 m bull which weighed out piecemeal at 53.37 tonnes, hinting that the whale was proportionally heavier in life (maybe 10% due to fluid loss ~ 58 tonnes/64 tons) given an estimate of 81-88 tonnes. Wood's estimate of 80 tonnes for the record bull may be a bit light and I'm not entirely sure how he arrived at this figure in the first place**.

*I should point out that the total length of a whale is taken from the snout to the notch in the tail fluke, some less than scrupulous authorities in the past got larger figures by measuring the flukes or even taking a measurement along the curve of the body.
**The largest sperm whale in the southern hemisphere was 19.5 m, and if you scale up Wood's 14.93 m/36 tonne animal to this size the estimated weight is 80.2 tonnes.

There's a world of difference between a ~40 tonne 16 m whale and a ~90 tonne 19.5 m whale, and incredibly much larger animals have been reported in the past. People are generally full of crap when it comes to extreme animal sizes and I'm extremely distrustful of any claims before the era of photography, but the case for giant sperm whales is unique. Wood (1982) mentions that the British Museum has a mandible measuring 5 m (16'4.75") in length, however there is apparently a 5.5 m (18') mandible in the Nantucket Whaling Museum (see here and here). The sperm whale jaw seems rather short when viewed externally (see here) but it seems much longer proportionally when the animal is viewed as a skeleton (see here). Why bring thus up? Well, Wood (1982) cites a source that claims a mandible:body ratio of 1:6.2 in a 14.7 m whale and a 1:5.4 ratio in an old 16.28 m whale. These may be measurements of the externally visible jaw, but I'm in desperate need of a peer-reviewed information on sperm whale mandible:body proportion. Wood (1982) used a 1:5.1 proportion for the animal to derive a figure of, gulp, 25.5 meters (83'8") and a 1:5 estimate for a 5.5 m jaw gives a length of 27.5 meters (90 feet). So let's say that Wood's estimate for sperm whale mandible:body length proportion is wrong and it is something like 1:4 - in this case the 5.5 meter mandible would still belong to a 22 m (72 foot) whale 106% longer and 120% heavier than the record specimen and close to triple the weight of the average sperm whale at something like 110 tonnes/120 tons. The record sperm whale was caught in 1950 and the mandibles are probably from much earlier (the 5 m one was from 1851) so it could be possible that the average shrank substantially since then. Wood's 25.5 m (83'8") sperm whale would be something like 170 tonnes/185 tons and if there was ever a 27.5 m sperm whale (90 feet) it would weigh something like 210 tonnes/230 tons. I'm not going to seriously entertain the notion that sperm whales rivaling blue whales for the largest animal actually existed, but it does seem like the average Physeter macrocephalus was considerably larger before heavy exploitation.

So are there patterns as to how large the largest outsized animals can be? Maurice Burton apparently stated that outsized specimens are 68% longer than the average size and 40% longer than the unusual large (Heuvelmans 1968 - from Burton's Living Fossils?). An animal 168% longer than average would also weigh around 5 times more - it would be a marvel for the bauplan of a species to expand to such extravagant sizes and work well enough for the animal to survive in nature. I'm not really a big believer in any self-proclaimed zoological "laws" or "rules" (I guess "Zoological Loose Guideline" doesn't have the same ring), but it is worth noting out that a population of sperm whales with 16 m long bulls could theoretically have an outsized specimen 26.88 m long. This pattern does not emerge with record-sized land mammals - the largest bush elephant (Loxodonta africana) was ~125% taller than average (~4 m vs. 3.2 m) and the largest Masai giraffe (Giraffa tippelskirchi) was a mere 111% taller than average (5.87 m vs. 5.3 m) assuming it was measured correctly (i.e. without the horns) (data from Wood 1982). Perhaps aquatic animals can tolerate outsized specimens more - the aforementioned record Wels catfish (2.78 m) was about 185% longer than average (~1.5 m).

Topics such as this one are certainly not going to be resolved in one blog post, so I'll end it here to get on to other things. There is a great deal of data out there, particularly on state record fish, that might be interesting to go through, but I think the rough picture here is sufficient. While incredible size may offer advantages such as resistance to predation and increased offspring health, tradeoffs will exist. Increased food acquisition comes to mind and there could be morphological issues as well. There are reasons that the outsize isn't the norm, but if situations change it could be adopted as a new norm. Australian giant feral cats are a possible example - emphasis on possible. Outsized animals aren't critical to understanding evolution and ecology but they can make for a potentially very interesting topic.


Birkeland, Charles and Dayton, Paul K. 2005. The importance in fishery management of leaving the big ones. TRENDS in Ecology and Evolution 20, 356-358

Brown, James H. et al. 2007. The metabolic theory of ecology and the role of body size in marine and freshwater ecosystems. In:I Hildrew, A. G. et al. 2007 (Ed.) Body Size: The Structure and Function of Aquatic Ecosystems. Cambridge University Press. pp. 1-15.

Burnie, David and Wilson, Don E. 2001. Animal: The Definitive Visual Guide. Dorling Kindersley, London and New York.

Heuvelmans, Bernard. 1968. In the Wake of the Sea-Serpents. Hill and Wang, New York.

Hutchings, Peter A. 2005. Life history consequences of overexploitation to population recovery in Northwest Atlantic cod (Gadus morhua). Can. J. Fish. Aquat. Sci. 62, 824–832

Jaquet, Nathalie. 2006. A simple photogrammetric technique to measure sperm whales at sea. Marine Mammal Science 22, 862-879.

Peters, Robert Henry. 1983. The Ecological Implications of Body Size. Cambridge studies in Ecology.

Stone, Richard. 2007. The Last of the Leviathans. Science 316, 1684-1688

Wood, Gerald. 1982. Guinness book of Animal Facts and Feats. Guinness Superlatives, Middlesex.

Sunday, October 5, 2008

Anomalous Cephalopod Appendages

Relative to other cephalopods, octopuses are reported with abnormal appendages* much more frequently. Symmetry and streamlining do not appear to be issues for animals that jet around infrequently and spend most of their time as benthic crawlers. It isn't fair to assume that every octopus is a benthic crawler, but as far as I can tell no pelagic octopuses were reported with abnormalities. Octopuses also seem capable of coping with arm morphologies quite divergent from the norm, one author reported seeing octopuses in the wild with the greater portion of seven arms missing (Toll and Binger 1991).

* The appendages are known as arms, regardless of how they're used. One recent study, which has not yet been peer reviewed as far as I can tell, reports that octopuses use arm pair IV as "legs" - but they primarily studied Octopus vulgaris and it is perfectly conceivable that different species have specialized their arms for different purposes.

Octopuses with subnumerary arm counts unrelated to natural losses have been reported. An article cited by Toll and Binger (1991) (Gleadall 1989 - see further reading) documented an Octopus sp. with seven arms*. Earlier this year it was claimed that the first "hexapus" was discovered, but somebody didn't bother checking the literature because I quickly found a description of a hexapodous specimen by Toll and Binger (1991). The aberrant Pteroctopus tetracirrhus entirely lacked arm pair II and there was a single adoral (near mouth) sucker located on the web between arms LI and LIII (Toll and Binger 1991). Arm pair I was observed with several other abnormalities: LI was narrower and shorter than RI, three of the first five suckers on LI were abnormally small, in the first ten rows of suckers on RI one third of them were abnormally small and there was widespread misalignment - some areas were even roughly tri-serial (Toll and Binger 1991). The authors reported that otherwise the specimen was normal and the abnormal appendages did not show signs of regeneration.

* Not to be confused with male Haliphron atlanticus

Octopuses with abnormally high counts of arms have also been reported. The Pink Tentacle posted on an octopus with 9 arms, which either has an extra RI or LIV (where's a hectocotylus when you need one?). The limb may be "split" off from another, I'll discuss this below. Kumph 1960 cited an earlier study (Parona 1900 - see further reading) which described an O. vulgaris with two thin arms in the place of LI and an Eledone cirrhosa* with an arm in between RII and RIII. Toll and Binger (1991) described the first decapodous octopus**, a lab-raised O. briareus which had a repeated arm pair IV. All of the arms were evenly spaced around the branchial crown, but the octopus still had a couple of other abnormalities. Male octopuses have a specialized arm for spermatophore transfer (RIII*** - the hectocotylized arm) and the Toll and Binger specimen had a supernumerary spermatophoric groove on LIII with no hectocotylus. A supernumerary diverticulum of the penis was also present.

* Toll and Binger 1991 cite this specimen as E. moschata, describes it as E. aldrovandi. Gonzalez and Guerra note that the species name has been changed.
** I can't help but mention the fact that there are octopodous decapods. Theirs' is a natural condition, however.
*** Some species have a hectocotylized LIII (Palacio 1973).

Abnormal hectocotylization is another abnormality recorded in octopuses. Palacio (1973) discussed some prior cases: hectocotylization was reported on RIII and LIII from an Eledone cirrosa (Appellof 1892 - see further reading) and on RIII and LII from Octopus briareus (Robson 1929 - see further reading). Palacio (1973) reported an O. vulgaris with hectocotylized RIII and RIV and an O. selene with hectocotylized RII and RIII. The extra hectocotylized arms all manifest with roughly the same morphology, they are not diminished in size and have proportional differences compared to the normal arms (Palacio 1973). Unlike the Toll and Binger decapodous octopus, none of these four specimens exhibited abnormalities with the genitalia (Palacio 1973). Palacio (1973) speculated that the extra hectocotylized arms were functional and were likely the result of mutations in sex-linked genes. It seems really odd that the mutation would occur on different arms each time, but how the hectocotylus wound up on RIII (for the most part) also seems rather odd. Another paper (Jereb et al. 1989) has discussed bilateral hectocotylization, but not exactly in a language that I can read.

So far I've only mildly alluded to the commonest and most bizarre condition that occurs in cephalopod arms - splitting. Bifurcation has been observed recently in the tentacle clubs of Onykia (= Moroteuthis) ingens by Gonzalez and Guerra (2007); polyfurcation was observed not-so-recently in the arms of a cuttlefish (Okada 1937 - see further reading). As can be predicted, octopuses are far more prone to demonstrating this trait - 7 were captured off Japan from 1884 to 1964 (Okada 1965a). Kumph (1960) described a bifurcation in LIII from O. briareus, both of the new arms had a biserial series of suckers and there was some webbing in between them. Kumph rejected the idea that it was the result of injury and interpreted it as a mutation or developmental oddity. This is a rather mild defect compared to some.

Around 8 years ago I found some diminutive quasi-cryptozoology book in a library (the downtown Naperville library) and I recall being extra dubious of the claim that octopuses have been found from extra "tentacles". Besides Toll and Binger (1991), the genesis of this post was a link my sister sent me to a post from Pink Tentacle which showed octopuses with up to 96 branching "armlets". The pattern of branching is quite complex: bifurcations have a continuous series of suckers, trifurcations have side branches with independently developed sucker series, then there are dorsal bifurcations, dorso-lateral mixed branching, dorsal branching and subcutaneous branching (with no suckers) (Okada 1965a, 1965b). Toll and Binger (1991) suggested that Okada's data indicates that the branching is not a random event. All of the specimens that Okada looked at were Octopus vulgaris from Japan, but it could be possible that this is due to sampling bias. If there is a regional tendency to be crazily polyfurcated I have no idea why.

Figures 4 and 5 from Okada 1965b. These are from the 1884 Uraga specimen with 90 branches total. Incredibly, RII did not have any branches. The left arm is LIII with five primary and thirteen secondary branches (21 total). The right arm is RIV with four primary and eighteen secondary branches (30 total). Ventral arms (III and IV) tend to have more furcations.

So what could possible cause these anomalous appendages? Gonzalez and Guerra (2007) note that in limb regeneration, an Apical Epidermal Crest covers the severed organ and Hox genes are activated to reproduce a limb as it was in the embryo. In their Onykia specimen, the tentacles appeared to have been regenerated but with clubs only (and bifurcate ones at that) - as such it had to get by with arms only (Gonzalez and Guerra 2007). Toll and Binger (1991) brought up amphibian regenerations with (surgically induced) supernumerary distal limb development but didn't reach firm conclusions on its relevance. These appendage abnormalities are likely due to developmental defects and/or regenerative difficulties, but the literature still is scant and I'm sure there's a lot more interesting work to be done.

Before I go, I should mention that five pairs of arms appears to be the ancestral conditions for cephalopods, including Nautilus. Pairs I and II turn into parts of the hood (!) and III-V turn into digital tentacles. I should probably mention the ocular and buccal tentacles in Nautilus which are not of homologous origin to these. Since there appears to have been a polyfurcation event in the past (albeit a less haphazard looking one), this could support the idea of some appendage anomalies being wholly genetic.


González, Ángel F. and Guerra, Ángel. First observation of a double tentacle bifurcation in cephalopods. JMBA Biodiversity Records - Published Online

Kumph, H. E. 1960. Arm abnormality in Octopus. Nature 185, 334-335.

Okada, Y. K. 1965a. On Japanese octopuses with branched arms, with special reference to their captures from 1884 to 1964. Proc Jap Acad 41, 618-623. Available

Okada, Y. K. 1965b. Rules of arm-branching in Japanese octopuses with branched arms. Proc Jap Acad 41, 624-629. Available

Palacio, F. J. 1973. On the Double Hectocolyziation of Octopods. The Nautilus 87, 99-102. Available (staring on page 124).

Shigeno, Shuichi et al. 2007. Evolution of the Cephalopod Head Complex by Assembly of Multiple Molluscan Body Parts: Evidence from Nautilus Embryonic Development. Journal of Morphology 269, 1-17

Toll, Richard B. and Binger, Lynetta C. 1991. Arm anomalies: cases of supernumerary development and bilateral angensis of arms pairs in Octopoda (Mollusca, Cephalopods). Zoomorphology 110, 313-316

Further Reading:
For those with better access to materials than me

Appellof, A. 1892. Teuthologische Beitrage IV. Uber einem Fall von doppelseitiger Hectokotylisation bei Eledone cirrosa (Lam.) d'Orb. Bergs Mus Aarsb, 14-15

Gleadall, Ian G. 1989. An octopus with only seven arms: anatomical details. Journal of Molluscan Studies 55, 479-487.

Jereb, P. et al. 1989. Sue due esemplari anomali di Scaergus unicirrhus (Mollusca, Cephalopoda). Oebalia 15, 807–809.

Okada, Y. K. 1937. An occurrence of branched arms in the decapod cephalopod, Sepia esculenta Hoyle. Annot Zool Japon 17, 93-94

Parona, C. 1900. Sulla dichotomia delle braccia nei Cefalopodi. Boll Mus Zool Anat Comp Univ Genova, vol 4 (No 96), 1-7.

Robson, G. C. 1929b. On a case of bilateral hectocotylization in Octopus rugosus. Pro Zool Soc Lond. 95-97.

Oddly enough, I have a somewhat relevant shirt featuring an octadecapodous octopus. I got it from yonder (I feel contractually obliged). I also feel obliged to say that this was the best of several pictures, yeesh.

Sunday, September 28, 2008

The Flexibility of Plesiosaur Necks

I can't believe that I've let this poor blog collect dust for over a month. Well, I sorta can. Unlike many other blogs, even some alleged Science Blogs, I don't (OK, no longer) have any interest in posting fluff on politics, religion, soon-to-be-stupid memes, my social life (sic) and blah blah blah. I'm hardly qualified to write about the subjects this blog frequents and I don't think the Internet needs to be any more polluted with preachin' to the choir commentary and knee-jerk rants. When I do get the urge to fluff, I'll make sure to put it at the top of posts so people will instinctively know to skip it.

Sauropterygians are derived aquatic diapsids characterized by features such as an enlarged upper fenestra, the loss of a lower temporal arch, a rigid quadrate and others. I'm using "plesiosaur" as all of the members of the clade Plesiosauria, which is defined by the presence of limbs modified into hydrofoils, limb girdles forming large ventral plates, absent nasals, short tails, rigid trunks and others (O'Keefe 2002). This clade includes plesiosauroids such as cryptoclidids, elasmosaurids, plesiosaurids, polycotylids as well as pliosauroids such as brachaucheniids, pliosaurids, leptocleidids and rhomaleosaurids. Traditionally it was thought that "plesiosaurs"/plesiosauroids were long necked and "pliosaurs"/pliosauroids were short necked but this is not always the case (Smith 2008). The pliosaur morphotype has evolved on three separate* occasions (in polycotylids, pliosaurids and rhomaleosaurids) and variations in body type within clades appears to be very complicated (O'Keefe 2002).

*Interestingly, Williston suggested that short necks may have evolved on more than one occasion - in 1907 (O'Keefe 2002). His comments on the functions of long plesiosaur necks (below) also sound like they could be taken from a recent abstract.

From: David Thomas Ansted’s The Great Stone Book of Nature (George W. Childs, Philadelphia, 1863). The pterosaur is priceless.

As the headache-inducing illustration above shows, plesiosaurs were commonly depicted as using their long necks in a snake-like, swan-like or harpoon-like manner. Plesiosaur necks had astoundingly high counts of vertebrae (up to 72 in
Elasmosaurus) but the lengths of the vertebrae were also elongated and the articulations were close and rigid (Smith 2008). Is it me, or does there seem to be a pattern of elongated reptile necks not being particularly flexible (e.g. azhdarchids, Tanystropheus)? Anyways, Samuel W. Williston noted the lack of flexibility in 1914 and suggested that plesiosaurs generally held their necks out strait and captured prey by moving the anterior portion of their necks with downwards and lateral movements. Zammit et al. 2008 claims that as recently as 1993 a paper reconstructed elasmosaurs as having an "S-shaped" neck in the vertical plane, but the author of that paper describes such reconstructions as "fanciful" (Storrs 1993). While the idea of plesiosaurs with very flexible necks fell out of scientific favor long ago, popular culture still depicts them in illustrations such as these.

There is at least one plesiosaur with what appears to be an adaptation for vertical movement in its neck. A juvenile specimen (only 700 mm, 28 inches in length) of the lower Cretaceous Leptocleidus had steeply angled zygapophyses in its cervical vertebrae which suggest the capacity for vertical movement (Kear 2007). The degree of flexibility was not mentioned, but it was apparently different enough from other taxa to suggest different prey types (Kear 2007). I feel that I am obliged to mention that Leptocleidus is a rhomaleosaurid* with the pliosaur-type morphology, apparently unlike many of its relatives (O'Keefe 2002). Leptocleidus doesn't seem to be too terrible short necked (as depicted here), it still had >20 cervicals and Fig. 6 of Kear 2007 shows the neck to be about twice the length of the head. Steeply angled cervical zygapophyses are unique as far as I can tell (which isn't very far), but I think it would be interesting to see if other attempts at "pliosaurs" and/or other juvenile Leptocleidus had this feature. Australian fossils of freshwater plesiosaurs are not strongly informative of taxonomy for the most part, but interestingly some suggest Leptocleidus (Kear 2006). Maybe having an unusually flexible neck in the vertical plane is useful for living in shallow near shore marine, brackish and freshwater environments - the juvenile in question was from marine deposits. Also problematic is that freshwater plesiosaurs in Australia were apparently subjected to cold to near-freezing conditions according to Kear (2006) - I couldn't imagine a 28 inch juvenile managing that. Freshwater plesiosaurs are potentially very interesting, they've been found worldwide from the early mid-Jurassic to the Late Cretaceous by the way, and I'd be curious about any morphological adaptations.

* Edit: Or not. A more recent phylogeny showed that species assigned to Leptocleidus belong to the sister clades Polycotylidae and Leptocleididae. This clade, Leptocleidoidea is a sister group to Pliosauridae. A clade of those two groups is a sister clade to Rhomaleosauridae (Smith and Dyke 2008). Thanks to Darren Naish for pointing this out.

Back to plesiosaur necks, how flexible are they anyways? The genesis of this post was a paper by Zammit et al. (2008)
which rigorously examined just that in the elasmosaur Aphrosaurus. The authors created life-sized 2D models of the vertebrae in dorsal and lateral view and used the minimum and maximum amount of intervertebral cartilage to create a possible range (Zammit et al. 2008). Models were also made of a boid, snake-necked turtle and sea lion for comparison - these tended to produce slight underestimates (Zammit et al. 2008). It turns out that Aphrosaurus could bend its neck 87–155° in the dorsal plane - far from the 360°+ needed for a swan-like posture - and motion in the ventral plane (75–177°) and lateral plane (94–176°) appears to have been greater (Zammit et al. 2008). The authors mention an unpublished master's thesis which showed a similar pattern from Cryptoclidus and Muraenosaurus (both cryptoclidids) and noted that the vertebral centra in those genera had concave articular faces and rounded lateral margins, imply more vertebral movement (Zammit et al. 2008). Exact figures were not given, but the vertebral count (~40) was lower so the cryptoclidid necks are not necessarily more flexible overall.

Zammit et al. mention that cervical zygapophyses are inclined more posteriorly so the back of the neck has increased vertical flexibility at the expense of lateral flexibility; the amount of flexibility also decreases going towards the posterior end of the neck. Previous papers (which I can't access) mention a "tongue in groove" structure also in the posterior part of the neck may be analagous to zygantrum–zygosphene articulations in snakes, which reduce torsion (Zammit et al. 2008, Moon 1999). Elasmosaurs seem to lack a mid-neck increase in flexibility that appears to have been present in cryptoclidids (Zammit et al. 2008). As far as function, Zammit et al. conclude that a strait held neck combined with lateral and/or ventral movement to capture prey is plausible but arching and slight s-curves appear possible as well; these are consistent with models of elasmosaurs as benthic grazers, ambush predators, and active predators using snake/turtle-like strikes.

It is worth mentioning that plesiosaurs typically* had dorsolaterally oriented orbits (e.g. the Lower Jurassic elasmosaurid Occitanosaurus - Bardet et al. 1999), indicating that they could not see prey below them. It would seem very odd then that elasmosaurids have been known to consume benthic prey, but a recent reconstruction (McHenry et al. 2005) shows the plesiosaur with its head and neck at an angle where it could probably see what it was eating. Plesiosaurs are also known to be capable of underwater olfaction, but I'm guessing that it wasn't acute enough to make up for a considerable blind zone. I'd hate to be terribly bold here, but perhaps plesiosaurs captured prey with lateral and/or dorsal movement of the neck. Considerable ventral movement would still be important for the aforementioned grazing on benthic prey and perhaps for sneaking up on some animals. Just... throwing it out there.

* I haven't come across any exceptions

Well, I lost most of a night's sleep thanks to this post, but it was worth it.


Bardet, Nathalie et al. 1999. A new Elasmosaurid Plesiosaur from the Lower Jurassic of Southern France. Palaeontology 42, 927–952

Kear, Benjamin 2006. Plesiosaur remains from Cretaceous high-latitude non-marine deposits in Southeastern Australia. Journal of Vertebrate Paleontology 26, 196–199

Kear, Benjamin. 2007. A Juvenile Pliosauroid Plesiosaur Reptilia: Sauropterygia) from the Lower Cretaceous of South Australia.
J. Paleont. 81, 154–162

McHenry, C. R. et al. 2005. Bottom-Feeding Plesiosaurs. Science 310, 75

Moon, Brad R. 1999. Testing an Inference of Function From Structure: Snake Vertebrae Do the Twist.
Journal of Morphology 241, 217–225

O'Keefe, F. Robin. 2002. The evolution of plesiosaur and pliosaur morphotypes in the
Plesiosauria (Reptilia: Sauropterygia).
Paleobiology 28, 101–112

Smith, Adam Stuart. 2008. Fossils Explained 54: Plesiosaurs. Geology Today 24, 71-75

Smith, Adam S. and Dyke, Gareth J. 200 The skull of the giant predatory pliosaur Rhomaleosaurus cramptoni: implications for plesiosaur phylogenetics. Naturwissenschaften. Available

Storrs, Glen W. 1993. Function and Phylogeny in Sauropterygian Evolution. American Journal of Science, 293-A, 63-90

Zammit, Maria et al. 2008. Elasmosaur (Reptilia: Sauropterygia) neck flexibility: Implications for feeding strategies. Comparative Biochemistry and Physiology, Part A 150, 124-130