A fish with hips

March 25, 2016 • 11:30 am

by Greg Mayer

The four-legged land vertebrates– amphibians, reptiles, birds, and mammals, collectively known as tetrapods— get around, at least mostly, on their four legs. Their front and hind limbs are attached, respectively, to their pectoral and pelvic girdles, bones that attach or closely adhere to the axial skeleton. The pectoral girdle consists of the scapula, coracoid, clavicle, etc., while the pelvic girdle is comprised of the anteroventral pubis, the posteroventral ischium, and the dorsal ilium. The ilium is firmly attached to the vertebral column by a connection to the sacral ribs extending from the sacral vertebrae. Your “hip bones”, at the widest part of your lower body, are easily felt (or seen if you’re wearing the right– i.e. not much– clothing), and these are the crests of your ilia.

Most fish, by contrast, have an unattached pelvic girdle, floating more or less free in the ventral part of the body, and free to move, evolutionarily, along the venter. The primitive condition, as seen in the bowfin, for example, the pelvic girdle has a position akin to its place in tetrapods– toward the rear of the body:

A bowfin (Amia calva)-- note the paired pelviv fins well back on the body. (the posteriormost ventral fin is the unpairedanal fin. From nicholls.edu
A bowfin (Amia calva)– note the paired pelvic fins well back on the body. (The posteriormost ventral fin is the unpaired anal fin. From nicholls.edu

In advanced ray-finned fishes, however, the pelvic fins may move far forward. In fact, the pelvic fins may move in front of the pectoral fins– the hind limbs are in front of the fore limbs! This is possible because the pelvic girdle is not attached to the vertebral column.

Skeleton of a Nile Perch from Norman, 1947 (via the Australian Museum).
Skeleton of a Nile Perch from Norman, 1947 (via the Australian Museum). Note the pelvic bones below and slightly in front of the pectoral arch.

The reason for this disquisition on the pelvic girdles of tetrpaods and fish is that Brooke Flammang and colleagues have just published a description of a living fish with a pelvic girdle attached to the vertebral column. This is really astounding! The species of fish is Cryptotora, a rare cave-dwelling fish from Thailand, that climbs on the wet walls of the caves in which it lives. In the figure below, which is a a head-on view of the pelvic girdle skeleton, the vertebrae are green, the pubis and ischium brown, the fin itself blue, and the sacral ribs and ilium are dark purple. Note the complete bony ring encircling the body from backbone to belly.

Cave fish (Cryptotora) pelvis. Flammang et al. 2016.
Cave fish (Cryptotora) pelvis. Flammang et al. 2016.

Tiktaalik, the “fishapod”, had pubis and ilium, but no sacral attachment or ischium. Some early tetrapods only had a looser sacral attachment than later tetrapods. Cryptotora, an advanced ray-finned fish, is not at all close to the lobe-finned piscine ancestry of tetrapods, so this represents a quite independent evolutionary origin of an attached pelvic girdle.

Carl Zimmer in the NY Times notes the tetrapod-like way in which the fish “walks” up cave walls, with an alternating left-right motion, and also provides a brief gif of one of the fish walking. This motion is more tetrapod-like than those of other walking fish (e.g., walking catfish). The alternating left-right motion of primitive tetrapod limbs is exactly what you would expect from the lateral undulations of a swimming fish. An attached pelvis– “hips”– in a modern teleost, however, is a really neat, and not expected, finding.

Flammang, B.E., A. Suvarnaraksha, J. Markiewicz and D. Soares. 2016. Tetrapod-like pelvic girdle in a walking cavefish. Scientific  Reports 6(23711):1-8. pdf

Coelacanth genome sequenced

April 19, 2013 • 4:08 pm

by Greg Mayer

Coelacanths are one of the three surviving groups of sarcopterygian (lobe-finned) fishes, and along with lungfish, one of the two groups that have remained fish in the vernacular sense (we tetrapods, the third surviving group, have of course become legged). The coelacanths also have a tremendous back story: known in the fossil record from the Devonian (over 350 mya) till the end of the Mesozoic Era (about 65 mya) but not afterward, it came as a great shock when one popped up in South Africa in 1938.

Latimeria chalumnae, the coelacanth (model)
Latimeria chalumnae, the coelacanth (model). Notice lobed fins.

Marjorie Courtenay-Latimer, curator of the local museum, got it at the fishing wharf in East London, Cape Province, where a fishing captain had put it aside for her as an unusual specimen. The discovery of a living specimen of a fish long thought extinct, and one related to tetrapods to boot, was a worldwide sensation. Others eventually turned up at various points in the western Indian Ocean, and in 1998 a population was discovered in Indonesia at the opposite end of the Indian Ocean.

Yesterday, a group led by Chris Amemiya (and including friend-of-this-site Neil Shubin) published the genome sequence of the coelacanth in the journal Nature. It is open access, and remarkably long and detailed given Nature‘s cramped editorial style. To me, two things stand out after a quick look. First, the coelacanth is the next closest relative to the tetrapods, after the lungfish. This is what was expected, but the paper provides much stronger support for this, using a  large data set (251 judiciously chosen genes). Note that, in an especially nice touch, Homo sapiens is represented in the figure by Miss Courtenay-Latimer herself.

Phylogenetic tree using coelacanth genomic data (Fig. 1 from Amemiya etal. 2013).
Phylogenetic tree using coelacanth genomic data (Fig. 1 from Amemiya etal. 2013).

Second, although, slightly more distantly related genealogically to tetrapods than lungfish, the slowly evolving coelacanths provide better comparisons for inferring events in early tetrapod genomic evolution than do the highly genomically derived lungfish. As Ammemiya et al. put it,

The vertebrate land transition is one of the most important steps in our evolutionary history. We conclude that the closest living fish to the tetrapod ancestor is the lungfish, not the coelacanth. However, the coelacanth is critical to our understanding of this transition, as the lungfish have intractable genome sizes (estimated at 50–100Gb)47. Here we have examined vertebrate adaptation to land through coelacanth whole-genome analysis, and have shown the potential of focused analysis of specific gene families involved in this process. Further study of these changes between tetrapods and the coelacanth may provide important insights into how a complex organism like a vertebrate can markedly change its way of life.

This is a bit reminiscent to the situation in studies of vertebrate origins: it now seems clear that sea squirts (urochordates) are closer genealogically to vertebrates than lancelets (cephalochordates), yet lancelets provide better comparative material for investigating early vertebrate evolution than do the highly derived sea squirts.


Amemiya, C., et al. 2013. The African coelacanth genome provides insights into tetrapod evolution. Nature 496:311-316. (pdf)

Holland, P. 2006 My sister is a sea squirt? Heredity 96:424–425. (pdf)

Good new paper on the fish-tetrapod transition

March 19, 2009 • 2:17 pm

Thanks to Carl Zimmer for pointing out a new paper by Jenny Clack in Evolution: Education and Outreach: “The Fish-Tetrapod Transition: New Fossils and Interpretations.” This is a good paper for the non-scientist who wants to know more about the documentation of this important transition. In WEIT I wrote mostly about the Tiktaalik roseae transitional form, largely because a lot of work on that fossil was done by my colleague Neil Shubin. I was criticized by some for not mentioning the other important fossils in this sequence, and Clack’s article fills this gap very well. Highly recommended.fish