Ichthyostega and Acanthostega 

 Kejser Franz Joseph Fjord, Greenland 



Tiktaalik 

 Ellesmere 

 Canada 



Elginerpeton 

 Scat Craig, 

 Scotland 



Elpistostege and 

 Eusthenopteron 

 Miguasha, Quebec 



Hynerpeton and 

 Densignathus 

 Hyner (Red Hi 

 Pennsylvania 



eton and Jakubsonia 

 id Livny, Russia 



Ventastega 

 Pavari, Latvia 



Livoniana 

 Ligatne, Latvia 



hthyostegid 



Continents and oceans of the Late Devonian epoch (between 385 million 

 and 359 million years ago) were situated as shown on the map. The regions 

 that later became modern-day countries and continents are also outlined. In 

 general, landmasses in the Late Devonian were farther south and more con- 

 centrated around the equator than they are today. Tetrapods and their clos- 

 est relatives have been discovered at the sites indicated by colored dots, 

 and also occur in China and Australia (not shown). 



have forced paleontologists to rethink virtually every 

 piece of "settled" knowledge about the origin of 

 four-legged animals. Not only is there a new un- 

 derstanding of the anatomical innovations embod- 

 ied by Ichthyostega and Acanthostega. But more, pale- 

 ontologists have assembled a picture of the global 

 distribution of Devonian tetrapods, as well as the ex- 

 tent of their diversity, that was unthinkable even a 

 decade ago. One of the most exciting new fossils is 

 Tiktaalik, a fish-tetrapod "missing link" whose dis- 

 covery in the Canadian Arctic by paleontologists 

 from the National Academy of Natural Science in 

 Philadelphia and from the University of Chicago 

 made front-page news this past April. Although no 

 single fossil can fully explain a complex evolution- 

 ary event, Tiktaalik is a true intermediate form, and 

 it provides vital clues to the when, where, and how 

 of the transition from water to land. 



B 



ack in my laboratory at the University Museum 

 of" Zoology in Cambridge, England, we exam- 

 ined our haul of Acanthostega fossils. Acanthostega 

 turned out to be not terrestrial, but aquatic. It had 

 short ribs, uniformly shaped vertebrae, and a tail with 



an oar-shaped fin supported by 

 bony fin rays. It also breathed 

 with its gills, like a fish. Acantho- 

 stega's two-foot-long body 

 could readily bend from side to 

 side as it propelled itself through 

 the water with lateral move- 

 ments of its tail. 



One big surprise was that 

 Acanthostega had not five digits 

 on each of its four limbs, but 

 eight. So there was no original 

 template for rive fingers or toes. 

 The story that land animals 

 evolved when fishes such as Eu- 

 sthenopteron emerged from the 

 water to crawl over land and lat- 

 er developed limbs wasjust that: 

 a story. Instead, we suggested, 

 limbs may have evolved first as 

 paddles, which were used to 

 swim, to spread the animal's 

 weight across a soft, muddy 

 riverbed, or to push through 

 the swampy, weed-choked wa- 

 ters of streams and lake margins. 

 Only later did each limb lose a 

 few toes, become able to bear 

 weight, and turn into a leg for 

 walking on land. 



Ichthyostega, too, had a tale to 

 tell. The first specimens of Ichthyostega to emerge 

 from Greenland were found by Danes but were stud- 

 ied mainly by Swedes. Erik Jarvik, one of the most 

 influential of the Swedish paleontologists in the mid- 

 twentieth century, published a reconstruction of 

 Ichthyostega that remained the basis of popular and 

 scientific conceptions of early tetrapods for decades. 

 His version of Ichthyostega was a three-foot-long 

 quadruped with a complement of specialized fea- 

 tures, including broad, overlapping ribs and a per- 

 manently bent elbow, which seemed at odds with 

 Ichthyostega's role as a very early tetrapod. The tail, 

 though, was finned like that of a fish and marked 

 Ichthyostega as primitive. 



The specimens from our 1 987 expedition gave 

 us the first clue that something was wrong with 

 the earlier image of Ichthyostega. Once back in the 

 lab, my colleagues and I prepared the Ichthyostega 

 bones by removing their surrounding rock with 

 tools such as a dental mallet and a handled needle, 

 all under a high-power binocular microscope. We 

 found that instead of five toes, or eight, Ichthyostega 

 had seven: three tiny ones bunched together at the 



38 



NATURAL HISTORY July/August 200b 



