ZOOLOGY

Twice bittenMark W. Westneat

The toothy visage of a moray eel is a fearsome sight. The discovery that morays can thrust a second pair of jaws out from their throat to wolf down prey whole increases their predatory reputation still further.

Many animals swallow their prey whole. Snakes come to mind, of course; but amphibians, liz-ards, birds and thousands of fish species can also attack, and gulp down, prey nearly as large as their head — imagine swallowing a peanut-butter sandwich or a salmon whole and you get some idea of how remarkable a feat that is. The methods by which animals do this vary from the suction feeding of fishes1 to the ‘unhinged-jaw’ mechanism of snakes2 and the inertial feed-ing of lizards and birds3. (For inertial feeding, picture a pelican thrusting its head upward to use gravity to choke down a large fish.)

These gulping mechanisms, along with most other vertebrate feeding habits that involve killing, dismembering and/or swallowing other animals, have generally been thoroughly investigated. On page 79 of this issue, however, Mehta and Wainwright4 document yet another tactic. They show how moray eels — elongated snake-like fishes that inhabit coral reefs and rocky intertidal habitats worldwide — drag a large item of prey into their gullet by using a second set of grasping jaws that they thrust forward from deep in their throat.

Accessory jaws positioned in the throat

are known as pharyngeal jaws, and are quite common among fishes. In many species, some of the bones that support the gills, called the branchial arches, have been modified into feeding tools that can filter prey from the water, crush and grind hard food such as snails or clams, and even grasp and tear softer prey before it is swallowed5.

Perhaps the most widely known pharyn-geal jaws are found in freshwater cichlids and marine wrasses and parrotfishes. These fish families possess hard, toothy pharyngeal plates that are thought to have allowed a wide range of feeding habits to develop and promoted their evolutionary diversification6. But Mehta and Wainwright reveal4 an additional class of pharyngeal-jaw mechanism. They aptly term this the ‘raptorial pharyngeal jaw’, for its ability to reach far forward from its resting position in the pharynx, and grab the prey to transport it back towards the stomach.

The discovery of this mechanism in the retic-ulated moray eel (Muraena retifera) is notable in several respects. First, it is a classic example of discovery-based science, stemming from an inspirational “oh wow!” moment. Such moments

are crucial to the study of living organisms, for they complement the approach of testing a priori hypotheses with statistical analyses of large data sets. In this case, Mehta and Wainwright combined intellectual curiosity and visualization technology to reveal the moray eel’s unusual behaviour. They had previously found7 that several types of eel do not use suction at all during feeding. This led them to search for alternative ways in which these predators could transport their prey into the oesophagus. By recording high-speed videos of eel feeding events in the laboratory, the mecha-nism became clear: the videos show the pharyn-geal jaws projecting far forward into the mouth cavity to latch onto the food. Stills from one of the videos are shown as Figure 1, overleaf.

A second notable aspect of this research is the way the authors backed up their primary video records with a detailed anatomical study of the pharyngeal bones and muscles. They also performed X-ray fluoroscopic analysis (the same technique used in a ‘barium swallow’ to obtain pictures of the human gut) to examine the range of movement of the pharyngeal jaws. These data sets allowed a detailed and intrigu-ing look at a novel feeding mechanism in this diverse and important group of animals.

As with all fascinating discoveries in animal behaviour and function, unexplained facets of the story remain. First, Mehta and Wainwright generalize their results obtained in a single species more broadly among eels. Although this may be valid for certain groups of morays within the 200-odd species in the moray family Muraenidae alone (and another 400 species in other eel groups), a more extensive comparative

The great Sumatra–Andaman earthquake of 26 December 2004, and the tsunami that it triggered, also shook the geological community. Much scientific effort has since focused on the possibility of further calamitous events in the Bay of Bengal, and especially on understanding the southern stretch of the fault that was responsible for the earthquake, and which lies to the west of the island of Sumatra.

But the fault’s northernmost limit, which extends along the coast of Myanmar (Burma) to Chittagong in Bangladesh and faces the densely populated Ganges delta (pictured), has received relatively little attention. Phil Cummins’ conclusion that an active zone still exists off Myanmar, as he reports on

page 75 of this issue, thus makes for disquieting reading (P. R. Cummins Nature 449, 75–78; 2007).

The Sumatra–Andaman earthquake was triggered when the Indian tectonic plate was thrust violently under the southeast Asian plate off the northwestern coast of Sumatra. Although this subduction zone was known to extend farther north through the Andaman Islands, its location and nature in the Myanmar region were less clear. It had been thought that subduction was no longer active in this area, and that the plate boundary extended, not under the sea, but on the land through Myanmar.

Cummins bases his alternative hypothesis on previous geological studies and recent geodetic

measurements, as well as on historical accounts detailing the effects of past earthquakes. He couples these observations with the fact that the floor of the Bay of Bengal has a thick layer of sediments, up to 20 kilometres deep, fanning out from the mouths of the Ganges and Brahmaputra rivers. This ‘Bengal fan’ insulates the underlying rock, creating thermal conditions more suited to generating earthquakes. Taken together, the implication is that a long, submarine subduction zone stretches for some 900 kilometres from the northern Andamans to the west of Chittagong.

The existence of such a fault requires a thorough re-evaluation of the potential for deadly

tsunamis in the northern Bay of Bengal. Cummins’ simulation of a large earthquake off the coast of Myanmar and the resultant tsunami shows the devastating effect it could have, and underscores the need for further study and monitoring of rock deformation in this region. Ninad BondreNinad Bondre is an associate editor at Nature Geoscience.

EARTHQUAKES

Burma’s fault

USG

S ER

OS

DA

TA C

ENT

ER

33

NATURE|Vol 449|6 September 2007 NEWS & VIEWS

������������� ��� ��������� ����

and evolutionary analysis of alternative strate-gies for prey capture and processing is needed to explain the origin and specialization of the raptorial pharyngeal-jaw mechanism.

The authors also compare the moray-eel transport mechanism to that of snakes. This is fair enough, as both groups are elongate, sinu-ous predators that can swallow their prey whole. But the ability of snakes to unhinge their jaws and alternately ‘walk’ the left and right jaws over a dead rodent to swallow it8 bears only a pass-ing resemblance to the moray’s ability to pull the prey down the gullet using a second set of jaws in the throat. These are different answers to a similar functional challenge, although further exploration of possible anatomical or

motor-pattern convergences in the two distantly related groups would certainly be worth while.

The prey-transport mechanism uncovered by Mehta and Wainwright4 will enhance the moray eel’s fearsome reputation. But it also pro-vides biologists with an important instance of a novel function in the skull of fishes that can be used to study the evolution of complexity. The spectacular assemblages of fishes on coral reefs, including eels, are among the greatest treasures of biodiversity. Now, when we snorkel or dive on a reef and see the fierce, toothy visage of a moray eel thrusting out of a crevice, we know that their raptorial pharyngeal jaws may be a primary reason for their success and diversity. ■

Mark W. Westneat is in the Department of

Zoology and Biodiversity Synthesis Center, Field Museum of Natural History, 1400 South Lakeshore Drive, Chicago, Illinois 60605-2496, USA.e-mail: mwestneat@fieldmuseum.org

1. Westneat, M. W. in Fish Biomechanics (eds Shadwick, R. & Lauder, G. V.) 29–76 (Academic, New York, 2006).

2. Kley, N. J. Am. Zool. 41, 1321–1337 (2001).3. Smith, K. K. J. Morphol. 187, 261–287 (1986).4. Mehta, R. S. & Wainwright, P. C. Nature 449, 79–82 (2007).5. Wainwright, P. C. in Fish Biomechanics (eds Shadwick, R. &

Lauder, G. V.) 77–101 (Academic, New York, 2006). 6. Mabuchi, K., Miya, M., Azuma, Y. & Nishida, M. BMC Evol.

Biol. 7, 10 (2007).7. Mehta, R. S. & Wainwright, P. C. J. Exp. Biol. 210, 495–504

(2007). 8. Cundall, D. & Greene, H. W. in Feeding: Form, Function, and

Evolution in Tetrapod Vertebrates (ed. Schwenk, K.) 293–333 (Academic, San Diego, 2000).

Figure 1 | Video action. These stills of a feeding moray eel come from the supplementary video evidence1 that complements Mehta and Wainwright’s detailed anatomical and X-ray fluoroscopic analyses. The pharyngeal jaws are seen only indistinctly here, but in the third frame (arrow) they shoot forward from within the pharynx to grasp the food item, which is then swiftly swallowed whole.

PHYSICAL OCEANOGRAPHY

Super spin in the southern seasDean Roemmich

The southern oceans are generally considered as isolated systems, much like their northern counterparts. But a combination of historical data and new density profiles suggests that they may be connected on a global scale.

All Earth’s major oceans contain a subtropi-cal gyre, a vast circulation system span-ning the entire ocean basin at mid-latitudes. These gyres have the crucial role in the cli-mate system of exporting excess tropical heat polewards1. Writing in Geophysical Research Letters, Ridgway and Dunn2 analyse new and historical data from the Southern Hemi-sphere oceans and confirm a previous idea3,4 of an inter mediate-depth connection linking the subtropical gyre circulations of the south-ern oceans. This opposes the conventional picture5 that the Southern Hemisphere mid-latitude circulation consists of distinct separated gyres within the Indian, Pacific and Atlantic oceans.

Ocean gyres are driven by the combination of low-latitude easterly trade winds and high-latitude westerlies, and so rotate clockwise in the Northern Hemisphere and anticlockwise in the Southern. The western boundary cur-rents of these gyres, such as the Gulf Stream

in the North Atlantic, are among the strongest ocean currents in the world. Despite having the same basic form, there are pronounced differences between the northern and south-ern subtropical gyres. The North Atlantic and Pacific oceans are completely separated by the continents on their eastern and western sides, whereas the Southern Hemisphere is truly the ‘ocean hemisphere’. The surface area between the Equator and 60° S is 84% ocean, and land masses form only partial barriers to the ocean currents.

The geography of the southern oceans made connections between them seem entirely plausible. Nevertheless, the confirmation of the existence of a ‘supergyre’ (an outer gyre surrounding the individual ocean gyres) con-necting the Southern Hemisphere circula-tions represents a new understanding of the southern oceans. The centres of the South Pacific and South Indian gyres are located at 35° S–45° S at a depth of 1,000 m (Fig. 1a), with

the South Indian gyre extending far eastward south of Australia. New Zealand is effectively the western boundary of the South Pacific gyre between 35° and 45° S; without the islands there would be much more extensive connec-tivity between the Southern Pacific and Indian oceans. Despite this barrier, the relatively small amount of land mass in the Southern Hemi-sphere allows global winds to drive Southern Ocean circulation and hydrographic varia-bility6,7 with minimal interruption.

The presence of the supergyre creates path-ways for ocean-to-ocean mixing and exchange of water properties. The near-surface portion of the supergyre controls a significant amount of the planetary heat balance. For ocean surface waters, the westward limb of the South Pacific gyre (the South Equatorial Current) bifurcates at the coast of Australia, where part of the current turns northward across the Equator and feeds the Indonesian Throughflow (the flow from the Pacific to the Indian Ocean through the Indo-nesian archipelago). This large transport of warm water into the Indian Ocean, balanced by much colder eastward flow south of Australia, represents an immense heat exchange of about 4 × 1014 joules per second from the Pacific8,9.

At intermediate depth, the existence of the southern supergyre has been suggested by simple models of wind-driven ocean circu-lation7. An agreement between models and data increases confidence in modelled climate trends in the Southern Hemisphere. These

34

NATURE|Vol 449|6 September 2007NEWS & VIEWS

������������� �� ����������������