WHEN JIM Springer and Jim Lewis, identical twins adopted at birth, met for the first time at age 39, they felt like long-lost buddies. Among their many similarities, both had had law-enforcement training and worked part-time as deputy sheriffs. Another pair of separated twins, Oskar and Jack, were raised respectively in Germany and the Caribbean. Both, however, had hot tempers, a weakness for sweets, and an abundance of common idiosyncrasies such as sneezing in crowded elevators and flushing the toilet before using it.

These pairs are among the subjects of a study of identical twins reared apart being conducted at the University of Minnesota. What this and other research is showing is that our genes seem to have a far stronger role in shaping personality and behavior than has been commonly supposed.

These studies are adding a dramatic new chapter to the nature-vs.-nurture debate that has occupied sages for generations. Theories supporting biological determinism were fashionable in the early part of this century, contributing to the climate which produced the horrors of Nazi "eugenics." In postwar years, the pendulum swung in the opposite direction, toward the idea that all behavioral differences could be explained by environmental influences. Now, there is renewed scientific interest in the biological base of behavior, spurred in part by the recent identification of genetic and biochemical factors in mental illness.

What researchers have been finding, with practically any behavioral characteristic they care to measure, is that genes seem to account for roughly 50 percent of the variance in a trait. {See box.} That is, genetic and environmental influences are approximately equal in their effect on personality.

Such findings could demand a radical rethinking of many conventional ideas about child-rearing, since research indicates that the environment shared by members of the same family (assuming it is relatively "normal") has little measurable effect in shaping their personality characteristics. Psychologist Robert Plomin of Pennsylvania State University calls this "the single most important finding in behavioral genetics in the last decade." It seems that the environmental influences that do shape personality tend to be idiosyncratic and unique to the individual.

Biology, Parents & Politics

The idea that humans may inherit a predilection for certain behavioral characteristics has not been a popular one in modern America. It runs against the country's egalitarian grain and, at worst, seems to open the door to obnoxious notions about innate racial differences. Many social scientists have been swept up by the idea that, while physical characteristics operate within parameters established by the genes, human behavior is infinitely plastic. "Anyone who grew up in the Roosevelt era believes genetics only determines whether you're a turtle or a human being," says psychiatrist Sarnoff Mednick of the University of Southern California.

But now it appears that genes have a hand in shaping just about any type of behavior that can be measured, cognitive or emotional, normal or pathological. Such characteristics include "activity level, alcoholism, anxiety, criminality, dominance, extraversion, intelligence, locus of control (personal autonomy), manic-depressive psychosis, political attitudes, schizophrenia, sexuality, sociability, values and vocational interests," as enumerated by psychologist J. Philippe Rushton of the University of Western Ontario.

By studying twins and adopted children, scientists have ascertained that genes have a marked effect on cognitive abilities, although this finding is still not accepted by many social scientists. Most behavioral geneticists now agree that the heritability of IQ is somewhere between 50 and 70 percent. But research on the genetics of personality is a more slippery area than IQ because measurements are not as old and well-validated, and personality characteristics are not as stable as IQ. Nonetheless, some provocative findings have emerged.

One of the most recent studies comes from University of Minnesota psychologist Thomas Bouchard's ongoing study of identical twins reared apart. A personality questionnaire was administered to 348 pairs of twins -- identical and fraternal, reared together and apart -- including 44 pairs of identical twins reared apart.

In this study, the median correlation for 11 personality traits measured was about .23 for fraternal twins raised together. That is, they have about 23 percent of their personality characteristics in common. The correlation was .49 for identical twins reared apart and .52 for those reared together, approximately the proportion that would be expected if estimates were made based on genes alone. The overall heritability (the degree to which differences can be attributed to genes) of the traits measured came out at about .50.

One aspect of the research, which the investigators call "both remarkable and puzzling," showed that the correlations for the identical twins reared apart and those reared together are "remarkably alike" on all 11 factors. This indicated that shared family environment -- the only factor on which the two groups systematically differed -- must have contributed very little to the similarity between the twins reared together.

Perhaps the most surprising findings have been in the area of political and social attitudes. It is generally assumed that a person picks up those attitudes from his environment; but now it is beginning to look as though they may be conditioned by a combination of temperamental traits and cognitive style. In the Minnesota study the heritability estimate for "traditionalism," which is related to respect for discipline and authority and general conservatism, was .45.

This phenomenon also showed up in a study conducted in the late 1970s by Sandra Scarr of the University of Virginia and Richard A. Weinberg of the University of Minnesota. They measured indicators of "authoritarianism" in families with adopted adolescent children and then compared them to attitudes in a group of families with biological children. They found substantial correlations among related family members. But in most cases, correlations between adopted children and between adoptees and their parents were negligible or nonexistent. Apparently "moral decisions and authoritarian views are not learned in rote fashion from one's associates," Scarr writes.

This interpretation has been reinforced by a recent study by Australian researcher Nicholas Martin of the Queensland Institute of Medical Research and colleagues in England and Virginia. A mail questionnaire sampling attitudes on a variety of socially and politically controversial matters was filled out by 3810 pairs of identical and fraternal twins. In most cases the correlations between the identical twins significantly exceeded those for the fraternal pairs. For example, male twins' opinions on the death penalty correlated at .52 for the identical twins and .31 for the fraternal twins. Eighteen other items, ranging from opinions on disarmament to computer music, showed strong support for the genetic component and little for cultural inheritance. On only three items -- opinions about straitjackets, pajama parties and coeducation -- did the influence of the common environment significantly exceed that of the genes.

One study that is attempting to explore the salient aspects of early environment is the Colorado Adoption Project, begun in 1975. This study covers 245 adoptees and their families compared with 245 nonadoptive families. Tests include measures of cognition, temperament, language development and behavior problems as well as physical development from infancy. Tests are particularly designed to illuminate one of the trickiest areas of uncertainty: the nature and extent of genotype-environment correlation.

The Environment of Growth

It is becoming increasingly clear that many ostensibly environmental influences are in fact genetically mediated. For example, investigators found that in a number of measures, the correlations between infant environment and development (that is, between "good" environments and "good" developmental outcomes) were higher in biological than adoptive families. For example, the "personal growth" factor of the Family Environment Scale, which involves familial cohesiveness and expressiveness, correlates with "infant soothability" more in biological than in adoptive families. This suggests that the behavior of a biological parent is actually matching the child's genetic proclivities rather shaping the child's temperament. Findings such as these, says Plomin, point to the need for "a dramatic reconceptualization of how developmental psychologists think about the environment."

If the common environment has so little effect, than what environmental influences do account for that 50 percent? At this point, speculation centers on factors unique to individuals -- ranging from parent-child interactions to peer influences to random events. Researchers are not saying that family environment has no influence on personality; rather, it does not influence it in a uniform (and therefore scientifically predictable) way. In other words, individuals with different genotypes respond to the same environment in different ways. The conclusion this leads to, says David Rowe of the University of Oklahoma, is that the environmental influences that actually have an effect on development seem to be the ones that differentiate family members, rather than the ones that act to make them more alike.

Ironically, although researchers have been accused by some of "biological determinism," Plomin points out that behavioral genetics offers the most promising route for determining just what environmental factors influence personality and exert positive influences on personal and cognitive development. This is a question that has provided often meaningless results in investigations that make no allowances for genetic factors.

Nature Versus Nurture

Many scientists remain skeptical, and the methodology now used in behavorial genetics research is often subjected to the same criticism heard during the heredity-IQ controversy of the 1970s. Psychologist Leon Kamin of Princeton, a highly visible critic in that debate, says that the same old problems remain: It is difficult to eliminate the effects of selective placement in adoption studies, and the effects to "twinness" artificially raise apparent heritabilities. Critics argue, for example, that dressing alike, sleeping in the same room and other common treatment tend to make twins more alike than they otherwise would be.

Researchers acknowledge that the subjects in their studies do not represent the while gamut of possible environments, but they say biases in selection or placement are not great enough to distort generalizations about the population groups they are studying. They say statistics fail to show that the different environments shared by identical as opposed to fraternal twins makes any difference in the behaviors being measured. They have also found that even when fraternal twins are treated as identical by their parents, psychometric measurements reflect the fact that they are not.

Critics of behavioral genetics often fear that such findings will serve to reinforce racism, particularly when it comes to discussions of crime, aggression and IQ. But researchers stress that their field is designed for the examination of individual, and not group, differences. They emphasize that differences between individuals are far greater than differences between racial or gender groups (although some significant sex differences have been identified in traits related to aggression and antisocial behavior). Study populations, moreover, have all been composed of whites whose rearing environments are in the normal range; the more homogeneous the environment, the stronger will appear the genetic contribution to behavior. Thus, heritabilities cannot be inferred for groups subjected to severely deprived environments.

The most impressive support for the methodology used by behavioral geneticists comes from biology. Elliot Gershon, a psychiatric geneticist at the National Institute of Mental Health, points out that family studies used to be the only tools investigators had to ascertain whether there was a genetic component to certain diseases. "The validity of such approaches has been confirmed," he says, by the identification of specific genes for many illnesses, including, most recently, manic depression.

As the role of biology in behavior becomes increasingly clear, behavioral genetics is likely to reach out in many directions. The biggest question confronting the field is: What are the mechanisms by which genes influence behavioral characteristics? At this point, according to Harvard sociobiologist E.O. Wilson, the best guess is that such genes regulate "the form and intensity of emotional responses, the thresholds of arousals, the readiness to learn certain stimuli as opposed to others, and the pattern of sensitivity" to environmental factors.

For behavioral geneticists, the pitting of "nature" versus "nurture" is an irrelevant and misleading way to state the issues under investigation. Rather, as Wilson indicates, "nature" is what determines how the individual responds to "nurture."