Monday, April 21, 2008

Are there limits to terrestrial mammals?

Body mass can provide critical information about species, from function to behavior to ecology (Reynolds 2002). Megafauna is frequently a subject of this blog, although a lot of it occurred early on and I don't feel that it covered the implications of large size very well. Just exactly how some species managed to function boggles the mind. "Mega-sauropods" were undoubtedly the largest terrestrial animals of all time with species such as Argentinosaurus and Puertasaurus apparently approaching 100 tons and others possibly exceeding it...I'm not getting into this but you can read about it here and here. Despite already being exceedingly large, the size of some megafaunal species are often exaggerated. Earlier in the blog I mentioned how the bear Arctodus simus was treated as if the weight of a hypothetically outsized male was average (this is obviously wrong visually) - and it was possibly made into a much more predatory animal. Reynolds 2002 cut the "black bear-sized beaver" Castoroides* down to 60-100 kg since previous estimates were based on a proportionally large skull (his used femurs). Reynolds noted that the femoral size of Castoroides was not much larger than a capybara - demonstrating how comparisons with species that aren't too closely related can be valuable for a rough comparison.

* This is some of the most preposterous exaggeration I have ever seen. Castoroides had a head/body length of 1.5 m (~5 feet). The living beaver Castor is 80-90 cm average and weighs 15-20 kg. Reynolds uses this obvious but hitherto overlooked comparison to get a figure of 52-60 kg, which would have tipped off earlier workers that something was very wrong with their estimate.

Paraceratherium (= Baluchitherium) in a copyright expired appearance. It looks fairly correct.

Indricotheres are hornless stem-rhinoceroses in the subfamily Indricotheriinae, family Hyracodontidae and superfamily Rhinocerotoidea and are now (mostly) regarded as having only the genus Paraceratherium (Antoine et al. 2008). The indricotheres are widely cited as being the largest land mammals of all time (Antoine et al. 2008) although this is not a very supportable contention (Fortelius and Kappelman 1993). It was called the world's largest land mammal while being introduced to the public and was later estimated at 20 or even 30+ tons (Fortelius and Kappelman 1993), a tonnage considerable for a sauropod. The original calculation for size was highly questionable as it assumed the fossils had the proportions of a "normal" rhinoceros (!) and scaled up fossils from different "grades" to the largest grade by apparently arbitrary values (Fortelius and Kappelman 1993). Widely circulated mass estimates were based on those sizes, showing how important it is to look into these factoids before repeating them mindlessly. Fortelius and Kappelman used various calculations (body length, head size, limb diameters) to get a mean of about 11 tons for Paraceratherium, although it may need to be raised somewhat due to sexual dimorphism. The authors concluded that the largest bulls may get 15-20 tons* - but I will admit that I'm not a big fan of speculating on freak specimens. People tend to latch on to the freakishly big specimens and assume it is the average - anacondas provide a particularly heinous example of this. However, there may be some constraints involved so I'll grudgingly let it slide.

*Gingerich 1990 calculated 9 tons for one "grade", but noted that individuals in the biggest "grade" may weigh 14-15 tons. He also predicts a maximum of 20 tons (I'll get to that).

The biggest problem with calling Paraceratherium the largest ever land mammal is that Proboscidians are apparently not considered. I'd hate to use another outsized freak, but the largest elephant ever measured was 4.2 meters from the foot to the shoulder lying down and was estimated to weigh 12 tonnes. Many sources act as these are precise measurements, which they're not. Wood though the bull would have stood 4 meters tall at the shoulder while standing and Christiansen cites an equation to calculate bull Loxodonta mass:

Body mass (kg) = 5.07 x 10^(-4) x (Shoulder height in cm)^2.803

This should yield a mass of around 10 tonnes, which is still close to the same size as a normal Paraceratherium. There still is the distinct possibility that the specimen had unusually long legs and some other measure (e.g. femur cross section) should have been used. Many species of proboscidian were smaller than today's elephant species and several were the same size, but on occasion there were species that got much larger on average (Christiansen 2004). Christiansen noted that the tall headed Elephas recki had a specimen that stood 3.75 meters at the shoulder in the flesh and various estimates gave masses approaching or exceeding 10 tonnes (Christiansen 2004). This species was a close relative of the living Asian elephant (E. maximus) and also seemed to share its more robust build than Loxodonta. The strait-tusked elephant E. antiquus seems to have been even larger with some specimens exceeding 11 tonnes and one partial femur hinting at an 14.5 tonnes specimen. Christiansen noted that fossils are normally assumed to be average size, although in that case it could have been an outsized specimen. Mammuthus species had a tall build similar to Loxodonta, but M. meridionalis was over 4 m at the shoulder and still weighed 7-10 tonnes.

It is not clear which species of proboscidian is the largest*. The mammoth M. trogontherii appears to have been 4.5 meters (14.7 feet) at the shoulder and weighed 13-15+ tonnes. Deinotherium giganteum was a non-elephant proboscidian (there is an excellent post here) that appeared to equal it in size. The author doesn't give any precise estimates, but the implication of these ballpark figures is quite staggering. I normally don't like freaks, but Christiansen states that such average figures would imply some specimens breaking the 20 tonne barrier. How could such a mammal get by?

*Wikipedia states that E. antiquus is 3.7 m at the shoulder, E. recki is 4.5, M. meridionalis is 4.5, M. trogontherii is 4.7 m and Deinotherium was up to 5 m. Once again, outsizes are given instead of much more useful averages. It also claims M. sungari was a 5.1 m 20 tonne mammoth, and since nobody else has heard of this species it could either be an outsized known specimen or a bunch of crap.

There still are a lot of open questions on why normal elephants are able to get so big in the first place. Clauss et al. 2003 note that elephants have a very low ingestion passage rate and digestibility rate, and Christiansen gives a figure of 40% for digestive efficiency. Elephants normally feed from 12-14 or even up to 19 hours a day, but Christiansen notes that a lot of this is low intensity feeding and consumption rates have been overestimated. Ruminants, even with more efficient digestive systems, appear not to have an advantage over hindgut fermenters at large sizes and have not evolved very large forms (Clauss et al. 2003). So maybe the inefficient digestive system of elephants is all that is possible at such sizes. Large size does have advantages such as requiring proportionally less food to maintain metabolism, avoiding predation, the ability to compete effective against smaller animals and to eat much less nutritious food (Clauss et al. 2003). Christiansen concludes that really giant proboscidians were a rare phenomenon since their dependence on large amounts of food (and increased movement) would have decreased population size and increased extinction rate. Often cited is a paper I can't find by Economos (1981) who concluded that the metabolic costs of gravity would limit large mammals to 20 tons (or tonnes?), and this was accepted by Clauss et al. (they ignored large proboscidians and mis-cited Fortelius and Kappelman claiming Paraceratherium was the largest land mammal). I unfortunately can't comment on that paper, but it does seem odd that at least three species averaged near that hypothetical limit and could theoretically have reached or exceeded it.

So to answer the initial question: It seems that ruminants are limited by competition from hindgut fermenters at large sizes and the largest of those species are probably limited by the environment. I'm skeptical of any proclamation of physical limitations, particularly one so relatively low and old, since there always could be ways of getting around them. 15+ tonnes isn't much for a sauropod, but for relatively recent herbivores with fast metabolisms it is quite a feat.


Antoine, Pierre-Olivier et al. 2008. A giant rhinocerotoid (Mammalia, Perissodactyla) from the Late Oligocene of north-central Anatolia (Turkey). Zoological Journal of the Linnean Society, 152, 581–592.

Christiansen, Per. 2004. Body size in proboscideans, with notes on elephant metabolism. Zoological Journal of the Linnean Society 140, 523–549.

Clauss, M. et al. 2003. The maximum attainable body size of herbivorous mammals: morphophysiological constraints on foregut, and adaptations of hindgut fermenters. Oecologia 136:14–27

Fortelius, Mikael and Kappelman, John. 1993. The largest land mammals ever imagined. Zoological Journal of the Linnean Society. 107: 85-101.

Gingerich, Phillip D. 1990. Predictions of body mass in mammalian species from long bone lengths and diameters. Contributions from the Museum of Paleontology, University of Michigan, 28: 79-92.

Reynolds, P. S. 2002. How big is a giant? The importance of method in estimating body size of extinct mammals. Journal of Mammalogy, 83(2):321–332.

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

Sunday, April 13, 2008

The Stoplight Loosejaw

In the book "Astonishing Animals" there is an incredible painting of a fish with jaws that look like something out of an H. R. Giger painting. The authors of the book noted that one of the animals in the book was an invention, and I began to wonder if the stoplight loosejaw was too bizarre even for a deep sea fish. A quick search in Google revealed the phony (the entry before it) and the existence of this incredibly bizarre fish. Even more troubling is the internal anatomy, which is some ways is even more strange.

The genus Malacosteus belongs to the order Stomiiformes and the family Stomiidae, which includes barbeled dragonfish. I can't resist sharing the larvae in the genus Idiacanthus. Anyways, the genus Malacosteus is usually stated to be monotypic (with M. niger) although it is possible that the one record of M. indicus does represent a distinct species and not a junior synonym (Sutton 2005) [see Post-script]. Malacosteus forms the subfamily Malacosteinae with the genus Aristostomias (Herring et al. 2005) and they do appear to be roughly similar in appearance (see Fishbase for the 200+ other species in Stomiidae). So while the basic appearance of the stoplight loosejaw isn't totally unprecedented, there still are some rather remarkable features. The head is capable of being "thrown back" thanks to unossified vertebrae, the jaws take up a quarter of the fish's length and subsequently this species has one of the relatively largest gapes of any fish (Sutton 2005). There are elongated teeth on the jaws, recurved gullet teeth and a lack of gill rakers, gill teeth or an ethmoid membrane (the "floor" of the mouth (Sutton 2005). Seeing Shouten's painting in Astonishing Animals makes this anatomy a lot more apparent and this photo was the best I could find online. I'll get back to the jaws and feeding because there are other strange features that deserve mention.

The "stoplight" portion of the name is in reference to a large red postorbital photophore. The production of red light is highly unusual amongst fishes and is only present in the closely related Aristostomias and the more distantly related barbeled dragonfish Pachystomias (Douglas et al. 2000). Apparently amongst animals only the beetle Phrixothrix is also capable of this (Sutton 2005). Even more amazing is that this feature seems to have evolved individually in the different fish genera, and the structure of the photophore in Aristostomias and Malacosteus is radically different (Herring et al. 2005). The photophore of Malacosteus seems to be clearly derived from the ones that operate in the more conventional blue-green spectra, and unlike the other species the red flashes are not visible to the human eye (Herring et al. 2005). Deep sea fish are not capable of seeing wavelengths longer than blue, excluding of course these species (Douglas et al. 2000).

Aristostomias and Pachystomias converge more so in possessing very long-wavelength sensitive pigments that have shifted towards red (Douglas et al. 2000); Malacosteus of course does something different. For one thing it uses a form of chlorophyll as a pigment, more specifically bacteriochlorophyll derivatives (Douglas et al. 2000). Since vertebrates aren't known to secrete chlorophyll derivatives, Douglas et al. theorize that it is somehow taken from green photosynthetic bacteria which are not known from the open ocean. The diet of Malacosteus is highly unusual and the stomach content was noted by the authors as possibly containing the pigment. I've always wondered how complex molecules are sent through the digestive system and incorporated in a specific part of the body, but it does seem reasonable.

Sutton 2005 further comments on all these unusual systems and ties it in with the unusual diet. The feeding morphology suggests that it feeds on very large prey, but M. niger feeds heavily on copepods, even at around the maximum size (24 cm or just under 10 inches). The feeding of this species has not been observed and Sutton comments that the lack of an ethmoid membrane and gill rakers seem problematic for eating small prey. Malacosteus may even be successful enough to select its prey (assuming there isn't a retention artifact) and it seems likely that there is an as-yet unknown feeding mechanism involved. Sutton also pointed out that the eyes are binocular-facing, quite unlike any other dragonfish, (see fig. 4B) and it may be related to the feeding habits. The stoplight loosejaw has apparently abandoned the vertical migration and feast-or-famine strategy of its family by maintaining itself on copepods and the rare large food item at one depth. Since the copepods may also provide the pigment, this, in Sutton's words, is a "chicken-or-the-egg" scenario.

It seems that there is a great deal to learn from this species, or genus*, and I'm still wondering exactly how something with jaws quite separate from the rest of the head eats copepods. I'm not entirely sure if it does get the pigment it uses to see copepods from the copepods it eats, but for now there just doesn't seem to be another plausible mechanism.


Douglas, R. H. et al. 2000. Long-wave sensitivity in deep-sea stomiid dragonŽ sh with far-red bioluminescence: evidence for a dietary origin of the chlorophyll-derived retinal photosensitizer of Malacosteus niger. Phil.Trans. R. Soc. Lond. B 355, pp. 1269-1272

Herring, Peter J. et al. 2005. Red bioluminescence in fishes: on the suborbital photophores of Malacosteus, Pachystomias and Aristostomias. Marine Biology 148: 383–394

Sutton, Tracey T. 2005. Trophic ecology of the deep-sea fish Malacosteus niger (Pisces: Stomiidae): An enigmatic feeding ecology to facilitate a unique visual system? Deep-Sea Research I 52, 2065–2076


Not one minute after I published this post, I stumbled on revised Malacosteus taxonomy:

Kenaley, Christopher P. 2007. Revision of the Stoplight Loosejaw Genus Malacosteus (Teleostei: Stomiidae: Malacosteinae), with Description of a New Species from the Temperate Southern Hemisphere and Indian Ocean. Copeia (4), pp. 886–900

I am not sure how I looked over this with Google scholar. This is the first taxonomic revision in nearly 75 years. Three species are currently recognized, M. indicus (from the Challenger expedition) and M. danae (after the research vessel). M. choristodactylus was also named and is apparently a synonym for M. niger. Kenaley examined 450 specimens of Malacosteus to resolve these taxonomic uncertainties. Apparently various specimens of M. indicus, M. danae and M. niger all describe one taxa, M. australis or the Southern stoplight loosejaw. It appears to have the same bizarre photophore and chlorophyll-related derivations. The paper very extensively described the anatomy of the species and is recommended for anyone further interested in that bizarre jaw.

Wednesday, April 2, 2008

Why Saber Teeth?

So why should a mammalian carnivore have saber teeth? Modern large carnivores seem to function perfectly well without them, although sabertoothed cats did go extinct fairly recently in the Late Pleistocene. Superficially, it would appear that saber teeth would be damage prone (and unrepairable) and cumbersome to use, but their presence in at least four* separate mammalian lineages indicates they can be advantageous in some situations. Even the non-mammalian Gorgonopsians have been called "sabre toothed reptiles" by at least one person (Jenkin 2001), despite the fact that they're not reptiles and elongated canines aren't exclusive to them in Synapsids. Enlarged canines are present in other groups such as the herbivorous Dinocephalians and Dinoceratans as well as molluscivorous walruses- but I'm not getting into that right now.

*And this is not considering the possible nimravid/barbourofelid split.

I carefully had to specify that no large Carnivorans have saber teeth because one genus of medium-sized felid has some surprisingly sabertooth-like traits. This really seems like a great opportunity to shed light on why saber teeth would evolve. The felids in question are from the genus Neofelis (There are two species now - Kitchener et al. 2006), also known as clouded leopards, and their long canines are often cited in popular articles. Christiansen noted that the morphology of the species was not well studied previously and was the first to properly describe the sabertooth-like characters - in 2006.

The upper canines are far out of the range for living felids, the canine length is proportionally close to sabertoothed cats such as Homotherium and nimravids (feliform sabertooths) such as Dinictis. The lower canines are small compared to the upper pair, although their overall length is still proportionally larger than any living cat and apparently most sabertooth cat genera except for the primitive Paramachairodus. Large canines necessitate a large gape, a it appears that Neofelis is capable of the widest gape (~90 degrees) of any living Carnivoran and within the range of sabertoothed carnivores. Relevant features related to such a gape include a ventrally deflected jaw joint, a posteriorly rotated facial portion of the skull and a low symphysial angle; all of these features are out of the normal felid range and within the sabertooth carnivore range. Other features such as zygomatic arch thickness, origin/insertion of the temporalis and carnassial-jaw joint distance, however, are within the range of modern felids.

So why would Neofelis evolve these characteristics? Genetic evidence indicates it is a sister taxa to the pantherine cats which don't have any of these traits, and the fossil record doesn't provide any evidence of when the traits evolved. Christiansen noted that the ecology of Neofelis is also poorly known, but cited observations of it killing prey exceeding it in size. Neofelis accomplishes this by a bite on the nape, whereas large cats kill large prey with a suffocating bite. Christiansen believes that this method for a faster kill may have been due to significant competition in the Plio-Pleistocene. There isn't any hard data yet, but it does seem to be a reasonable hypothesis. Members of Carnivora are well known for re-evolving ecomorphologies (Valkenburgh 2007) and I can't help but wonder if a sabertooth guild is a inevitable part of feline-like diversification and competition. This doesn't mean that Neofelis will evolve into sabertooths, and why it only inhabits one locality cannot be known at this time. Convergent evolution and Elvis taxa always bend my mind.

Christiansen further states that we really need a lot more data on this genus. He brings up the possibility that early sabertoothed cats and nimravids (and Neofelis?) may be functionally quite different from the derived species, and may not even be considered "sabertooths". While Neofelis isn't perfectly analogous to sabertoothed carnivores, the revelation of it sharing so many derived characters should give fascinating possibilities for the study of sabertooths. I haven't seen any articles building on this yet, but it is still pretty soon.

This article turned out a bit shorter than I'd like, and I do still have those other lineages of sabertoothed carnivores to talk about...


Christiansen, Per. 2006. Sabertooth Characters in the Clouded Leopard (Neofelis nebulosa Griffiths 1821). Journal of Morphology, 267:1186–1198 (2006).

Jenkin, Ian. 2001. Fossils Explained 33: Palaeozoic Carnivorous Reptiles. Geology Today, No. 17 V. 1 pp. 36-39.

Kitchener, Andrew C. et al. 2006. Geographical Variation in the Clouded Leopard, Neofelis nebulosa, Reveals Two Species. Vol. 16 Is. 23 pp. 2377-2383

Morlo, Michel et al. 2004. A new species of Prosansanosmilus: implications for the systematic relationships of the family Barbourofelidae new rank (Carnivora, Mammalia). Zoological Journal of the Linnean Society, Vol.140 No.1 pp. 43-61.

Valkenburgh, Blaire van. 2007. Deja vu: the evolution of feeding morphologies in the Carnivora. Integrative and Comparative Biology, volume 47, number 1, pp. 147–163

Tuesday, April 1, 2008

The Frontiers of Vertebrate Zoology

The following is an abstract of a peer-reviewed article which goes to press today:

Newly Described Unique Secondarily Aquatic Synapsid From The Anthropocene Of Rhode Island And Providence Plantations (USA): Homage To Heuvelmans

C. A. McCormick, E. Derby Upton, Nathanial D. Pickman, W. C. Webb & Juan Romero

The rate of discovery for large vertebrates, especially marine ones, has not ceased in the 21st century. The hypothesis of surviving marine synapsids from anecdotal reports was confirmed by a skeletal find in R.I. The skull showed mosaic traits of both advanced and primitive synapsids, along with numerous apomorphic traits. This, coupled with a unique thoracic structure (SJSV) indicate the species has an incredibly broad diet. There are numerous vertebral apomorphies, indicating different "modes" of swimming, possibly including slow cruising, fast thunniform swimming, and overland locomotion. The caudal region also shows hitherto unknown structures and muscle groups. Anecdotal reports suggest a huge range of behaviors and habitat and indicate this is a species fully capable of dealing with human planetary change. Phylogenetic analyses proves inconclusive, and future papers will shed light on that and other anomalies found in this remarkable species.

Unfortunately, legal issues do not permit a more extensive "media release" at this current time. The full article is available upon request and may also be found here:

Article Found Here

Enlargement of Fig. 5

April is the cruelest month, as far as blog output is concerned, but I have a few things in the pipeline.