Seth is 2 1/2, a cheerful, friendly and appealing child. Bright, extraordinarily verbal, observant and curious, he appears enchanted in a playroom filled with toys and puzzles -- some significantly more complicated than a child his age might be expected to master, but no problem to his nimble fingers and quick mind.

He leaves his mother briefly to go off with an adult companion to find a rocking horse, and does so with equanimity, apparently secure in the knowledge that the separation is temporary, that she'll be there when he gets back.

To a casual observer he looks like just the kind of kid anybody could hope to have -- self-possessed, self-sufficient, affectionate and well behaved.

So far, so good. But Seth's mother and her psychiatrist and some other experts are watching Seth around the clock -- for the least change in patterns of behavior, the slightest sign of emotional distress. They've been watching since shortly after he was born and will continue to for some time to come.

They watch because Seth has about a 70 percent chance of becoming a severely manic-depressed person, or of having some other serious mood disorder. Seth's mother, herself manic-depressive most of her life, is "frightened to death," she says. But she is beginning to be optimistic. Cautious, careful, but hopeful.

The toy-filled room in which Seth is playing is in an apartment in the Old Stone House on the campus of the National Institutes of Health. Nobody lives in the apartment, but in the last few years, dozens of infants and children and their mothers and their playmates have spent hours there. Each of the rooms has a couple of mirrors. Each of the "mirrors" is actually a one-way glass through which the activites inside are videotaped and monitored by experts.

The apartment is a laboratory of the Child Rearing Project at the National Institute of Mental Health. It is here that researchers record, dissect and analyze the ways mothers interact with their children and infants. The researchers want to tease out the differences between the ways mothers who are biologically depressed interact with their children and the way normal mothers interact with theirs. Eventually, they hope to identify the things that trigger depressions or abnormal mood swings in young children and how to avoid those triggers.

Recent research has demonstrated a clear genetic factor in major mood disorders, especially in the children of depressed parents, who stand about a three times greater chance of developing a mood disorder.

In some cases, signs of behavioral dysfunctioning can be identified in infants. Some of the Rearing Project's early work with infants, which already has been published in the scientific literature, demonstrated inappropriately excessive reactions like prolonged rages when the mother leaves the room and other abnormal behaviors in infants of depressed mothers, markedly different from those in more normal environments.

But "those findings still do not tell us what the connection with the mother might be," says Dr. Donald McKnew, a research psychiatrist for the project and an authority on depression in children. "What does the mother do that might cause these?

"Are the behaviors we start to see in mood disordered 5- and 6-year-olds just genetic? Does a wire go off in your head and your norepinephrine levels change too little of this brain chemical is linked to depression and you become an altered child whether or not your mother is wonderful or terrible, whether she loves you or hates you? Or is it mostly the mother, and that little change of the wiring in your head makes just a slight difference?"

McKnew, who with colleague Dr. Leon Cytryn in 1983 published "Why Isn't Johnny Crying? Coping With Depression in Children," believes that the "truth lies somewhere in the middle. We're almost sure you can't have a mood disorder without the genetic defect, even a mild mood disorder. So you've got to have the altered chemistry, but it's pretty clear that your environment has got to sock it to you in some kind of way . . . We know that the environment can change the specific chemistry. Norepinephrine levels can be altered over a given day's time depending on whether you've had a good day or a bad day. So it's an interplay . . . The Rearing Project is the first attempt to glue it all together."

Susan, Seth's mother, is also in a sense his environment. She is acutely aware that she might be the one to "sock it" to Seth. She is also willing -- eager, in fact -- to talk about her own disorder, how it might affect Seth and how McKnew and the Rearing Project are helping her lessen his chances of "catching" her depressive illness. "They are," McKnew has said of these genetic depressions, "as contagious as German measles" when the risk is there.

Susan, 39, who has requested that her surname not be used, was raised in a well-to-do, intellectually enriched household. Her father was in and out of mental hospitals, but if she even mentioned it, she recalls, she was quickly shushed by her mother.

She began having migraine headaches at about 4 1/2, had a difficult time relating to classmates and "played sick a lot so I didn't have to go to school . . . There was a lot of loneliness." She saw her first psychiatrist at age 10.

By the time she was 18 she was a full-fledged "bipolar depressive -- type one." Type ones have manic periods that can last for weeks or longer followed by deep, suicidal depressions.

Susan was only 18 the first time she tried to kill herself. Her life became a series of roller-coaster mood swings, psychosomatic and neurological illnesses -- she had Bell's palsy five times, for example -- and visits to psychiatrists, psychologists, therapists, sometimes more than one at a time, in different cities. Withal, she managed to get through college, do a bit of work on a master's degree, get married and even work as a nursery school teacher.

She got pregnant -- wrongly assured by a doctor that her mood disorder was not genetic -- but became deeply depressed and sought out McKnew to help her "keep Sethy from having to go through that blackness."

Says McKnew, "I had to break the news that Seth was indeed at risk, and when I first met Susan her moods were totally out of control. She was unable to get up, unable to come to sessions, unable to do anything, much less take care of a child."

McKnew diagnosed Susan's illness precisely -- the first time in her life that had been done -- and she responded well to the psychoactive drug Lithium, which regulates the brain chemicals involved in her manic stages, and to Elavil, which controls her depression.

She misses her manic stages: "They were fun," she says. That, says McKnew, is a common reason for manic-depressive patients to stop taking medicine. But Susan's concern for Seth overwhelmed anything else, and, says McKnew to her, "You don't give yourself enough credit. You've changed. You're taking care of yourself."

After two or three four-hour sessions in the apartment, volunteer parents and children receive a thorough psychiatric analysis and, after the experts have watched and rated the tapes, receive what is essentially "a family evaluation," which measures the child against the norms.

Susan may "hover too much," the raters suggested, but otherwise they saw nothing amiss. "We were born with a missing chemical," says Susan. "We have to live the best we can, and here is a program that can give us the edge, thank God."

"Eventually," says McKnew, "we may really be able to give an added edge. If the study produces the results we hope, we may find the certain common threads that run through certain mothers with certain kinds of ailments we don't see in normal mothers." More Information ------

Write Rearing Project, Laboratory of Developmental Psychology, NIMH, Bldg. 15K, Bethesda, Md. 20205, or call 496-4431.