dence suggests that most immigrants tend to acquire English language skills over time.

Today, after three or more generations of descendants of the original immigrants, offspring of European groups are virtually indistinguishable in terms of education, income, occupation, and residence. Because of extensive intermarriage and the changing patterns of ethnic identification among descendants of European immigrants, the boundaries between different national-origin and ethnic groups-Italians, Irish, Polish, and Jewish, for example-are increasingly blurred. If population projections had been done for groups of European origin at the beginning of the twentieth century, they would have failed to predict the voluntary choices of ethnic ancestries of the present U.S. population.

Under high rates of ethnic intermarriage, ethnic identity becomes quite varied and increasingly a matter of choice. In recent years, ethnic and racial intermarriage has been increasing in this country and is increasingly common among children and grandchildren of Asian and Hispanic immigrants. Current population projections of the future ethnic composition of the U.S. population are especially hazardous because future patterns of intermarriage and the meanings of race and ethnicity are uncertain.

American public attitudes about immigration have long been equivocal. The United States has had periods of large-scale immigration, with considerable public support and welcome, and periods of great distrust and antagonism toward immigrants. In the past 50 years, public opinion polls have allowed us to chart more clearly how the American public views immigration and regards immigrants. Americans have increased their opposition to immigration in recent decades, in part, it appears, because of economic concerns. These attitudes vary greatly, however. College graduates have more positive attitudes toward immigration. Black, Hispanic, and Asian Americans tend to have more favorable attitudes toward immigrants than do non-Hispanic whites.

Public concerns with immigration are centered on illegal immigration, although the average resident greatly overestimates the proportion of immigrants who are illegal. Over two-thirds of respondents believe that most recent immigrants are illegal, whereas the proportion of illegals among total immigrants is closer to 20 to 30 percent.

The scant available data on crime do not allow us to say much about its relationship to immigration. It is hard to draw firm conclusions from the currently scarce information. The crime rate increased from the 1960s until about 1990, then has declined noticeably for the past six-years. There is no apparent association in these temporal trends with immigration. From available studies, it appears that overall crime rates have been associated more with other factors, including the changing demographics of the country (with shifts in the number of young men), fluctuations in drug use, and changes in the effectiveness of the police and criminal justice system in reducing local crime. The problems of data of the criminal justice system make it very difficult to reach empirical conclu-