The color of wakefulness is gray and smoky on the Patuxent River. The Turtleman of St. Mary's County tips a six-foot-tall, rectangular chicken-wire basket--its bottom and sides draped with slimy jellyfish--and out spill a variety of creatures into the bottom of the fiber glass boat.

Croakers flap with deathly gasps. Maryland blue crabs snap their claws and sound like a chorus of muffled castanets. The turtles, diamondback terrapins, desperately try to climb up the side of the boat.

But the Turtleman, in white rubberized overalls and boots, his big hands in yellow rubber gloves, grabs the turtles one at a time, sizes them up and calls out the numbers that identify the individual turtles to his student assistant.

He tosses the "keepers" (the crabs big enough to sell) into a bushelbasket, secures the turtles in net bags and pitches the rest of the creatures back into the water.

For the past 12 years, Willem M. Roosenburg, 40, a nationally known Ohio biologist, has been pursuing a scientific obsession with the life histories and sexual behavior of diamondback terrapins, the Maryland state turtle. His work is regarded highly enough that it has led to laws protecting the creatures.

Around Southern Maryland waterways, he is a familiar sight even if few people know his name or understand his work in ecology or evolution. Most just call him Turtleman.

Along a five-mile stretch of marshes, sandy beaches and tidal creeks of the Patuxent River, Roosenburg has made 20,000 captures of 8,000 individual turtles since 1987. He has recorded each capture, tagged each turtle and then set them all free. He can recognize individual turtles by their distinctive markings and by the identifier he has notched on their carapace.

"Turtles have an extremely peculiar manner of sex determination," Roosenburg explained and then launched into the dense, sometimes impenetrable language of research science.

With people and most creatures, gender is determined by sex chromosomes inherited from parents. The result is practically an even ratio of males to females.

"With turtles, we throw that picture out the door," Roosenburg said.

For the diamondback terrapins, the temperature of the nesting sand determines whether a turtle egg hatches into a male or female turtle. The resulting gender ratio is, Roosenburg said, "female biased," 2 to 1.

Roosenburg has found, for instance, that nesting beaches that have a northern (cooler) exposure are male terrapin beaches and that the southern (warmer) facing beaches are female turtle beaches.

"It's really quite puzzling to a biologist," said Jeff Lang, a research scientist at the University of North Dakota who is working on environmental sex determination among crocodiles and alligators. "It doesn't seem like a very rational system. Willem's work focuses on the why: Why does this make sense in terms of evolution and ecology?"

Even after 12 years, Roosenburg, an assistant professor at Ohio University's Department of Biological Sciences, has no clear answers to the questions. Which isn't surprising, he said laughing, considering that the evolutionary life of turtles is 250 million years. A decade, considered in that time span, is indeed just a pinprick of a beginning.

But his early findings from a general demographic study have resulted in some small measures of protection for the diamondback terrapin, a turtle common in the brackish waters of the Eastern Seaboard, from Cape Cod to the Texas Gulf area.

In 1992, he persuaded officials of the state Department of Fisheries to ban commercial capture of terrapins from May to July 31, during the terrapins' nesting season.

Roosenburg's research also resulted in new state regulations requiring the redesign of bank traps, the ingenious chicken-wire contraption watermen use to catch crabs along the shallow river shores. The redesigned bank traps, made of chicken wire over a wooden frame, are taller, allowing room for the turtles to surface for oxygen.

Most recently, Roosenburg recommended that residents add a simple wire attachment to their crab pots, another kind of trap, that excludes turtles but still catches regulation-sized crabs. It is a simple device that could prevent a significant number of terrapin deaths, he said.

Roosenburg is an affable, laid-back scientist who, with his Dutch-boy haircut, is youthful in appearance and manner. He grew up in Calvert County, where his father worked as a biologist at the Chesapeake Biological Laboratory.

Roosenburg's childhood was spent crabbing. During years when he doesn't have grants to pay for his research, Roosenburg supports his research with his earnings as a summertime waterman, selling his crabs to a local restaurant.

As an undergraduate, he spent years helping with research on sea turtles in Costa Rica. He married one of his first graduate assistants, Kate Kelley, a biologist researching tiger beetles.

He didn't encounter his first diamondback until he began his study of them in 1987.

As scientific concepts go, environmental sex determination is new--it was discovered in the early 1980s--and its big question is why. Roosenburg's passion is fueled by a goal: "I'd like to contribute to its understanding."

"The lineage of turtles is very old," he said. "The interesting question is how did turtles manage to survive periods of dramatic climate change while organisms such as dinosaurs became extinct?"

Also, turtles are accessible and lovable creatures, and talking about them is one way to engage people to discuss biodiversity and conservation, he said.

In Massachusetts, New Jersey, Rhode Island and Connecticut, the terrapins are endangered or protected. In other states, there are noted declines in their population. But in Maryland, 12,000 terrapins are harvested commercially every year for sale in Asian markets as food.

"My ultimate objective is to protect terrapins from commercial harvest," Roosenburg told a small audience during a recent presentation at the Calvert Marine Museum on Solomons Island.

Out in the water the next morning, Roosenburg said that the crush of development along the scenic waterfront of St. Mary's is threatening nesting beaches.

"Those rocks weren't there," he said, pointing to a beach at Washington Creek where rocks have been piled to prevent erosion. "That shoreline used to be wooded. Now it's planted with sea grass. It's going at an alarming rate."

On Kiawah Island off the South Carolina coast, a study of diamondback terrapins showed that a significant percentage of the turtle population was being killed every year in unattended recreational crab pots, said Whit Gibbons, a herpetologist who teaches ecology at the University of Georgia.

"The point is turtles are going to be gone from our area in a couple of generations if this continues," Gibbons said. "That's the value of Willem's work. If we want diamondback terrapins, we've got to change our behavior."

In the annals of turtle studies, Roosenburg's work ranks as one of the most comprehensive and largest of its kind--"right up there," evolutionary biologists say, with the so-called "Great Turtle Study" of Michigan that was started in the 1950s and continues today.

"It's outstanding, isn't it?" said Roosenburg's father, Willem Herman Roosenburg, 76, who was a shellfish biologist. "If I had lived to produce a biologist that's much better than I had dreamed I could ever have been, then my life is complete."

CAPTION: Willem M. Roosenburg and his assistant, Erin Casey, lift one of the traps set to catch the diamondback terrapins from the Patuxent River.

CAPTION: Roosenburg measures one of the diamondback terrapins.

CAPTION: Biologist Willem M. Roosenburg has studied diamond terrapins along the Patuxent River for 12 years.