How an ancient sea creature grows up. (Stalking Larvae).

Researchers in Japan have finally observed living larvae of a sea lily, a marine invertebrate with a 500-million-year fossil record. The finding supports a century-old hypothesis about how the vertebrate nervous system evolved.

Until now, marine scientists could only speculate about the larvae of this most-primitive echinoderm, a group that also includes starfish and sea urchins. Because echinoderm larvae and vertebrate embryos share some features, such as the formation of mouth structures, echinoderm larvae have been of particular interest to developmental biologists.

However, most of the 80 species of sea lilies found worldwide live on inaccessible, rocky slopes in the deep ocean, some as deep as 8,000 meters. Although sea lilies are animals, they look like plants because they attach to the sea floor by a stalk.

One species, Metacrinus rotundus, lives at relatively shallow depths of 100 to 150 m off the coast of Japan. Shonan Amemiya of the University of Tokyo in Kashiwa was developing ways to keep M. rotundus adults healthy in aquariums when he discovered that their sperm and egg cells always mature in September. After pinpointing this spawning period, Amemiya and his colleagues collected adult males and females from the ocean as they were just about to reproduce. In the laboratory, the scientists succeeded in raising the resulting larvae for 10 days. The researchers describe their observations in the Jan. 9 Nature.

After 3 1/2 days, the 0.6-millimeter-long, free-floating larvae have bands of tiny hairs, or cilia, running sinuously down their fronts and backs. In other echinoderms, this larval form--called an auricularia--has interested scientists since they first discovered it over a century ago. In 1894, marine biologist Walter Garstang suggested that the primitive ciliary band, which harbors neurons, evolved into the dorsal nerve cord of vertebrates.

"We were most excited when we found that the sea lily embryos developed into auricularia-type larvae," says Amemiya. Since sea lilies have such ancient roots, the observations reveal what the echinoderm's ancestral larvae probably looked like and support Garstang's hypothesis, says Amemiya.

Identifying the sea lily's larva is a "major advance," says Richard J. Mooi of the California Academy of Sciences in San Francisco. But he cautions against making broad conclusions about echinoderm evolution. M. rotundus may not even represent all members of the stalked crinoids, the class to which the species belongs.

The new findings, however, may help ecologists learn how sea lilies reproduce and disperse. "It's tough to work on these animals," says Charles G. Messing of the Nova Southeastern University Oceanographic Center in Dania Beach, Fla.