The column was about a couple of tough housing cases. The authorities, hoping to maintain the properties as racially mixed and stable, had put a limit on the percentage of black families to be accepted. To operate on a first-come, first-served basis, they feared, would make the development increasingly black, as blacks had fewer housing options to begin with.

On the other hand, some blacks saw such benign quotas as a sort of reverse reverse discrimination, which, no matter its good intentions, worked to the disadvantage of blacks, precisely because they had fewer options. The column was about that dilemma.

But then I said something that has triggered a spate of letters from one end of the country to the other.

If the development becomes black enough, I said, whites would stop moving in; those already there would start moving out, and the property would become first all-black and then black slum.

The letter-writers wanted to know whether I really believed the inevitability of an all-black housing development's deteriorating into a slum.

The short answer is: No, I don't. I know too many all-black or predominantly black neighborhoods (including my own) that show no sign of deterioration.

But in the two cases I wrote about -- a public-housing development in Charlottesville, Va., and a federally subsidized middle-income development in Brooklyn -- there is a longer, less optimistic answer.

Apartment houses, developments or neighborhoods that are all-black because black people find them attractive tend to remain attractive. Neighborhoods of home-owners, black or white, tend to retain their value.

But there is another scenario that applies to developments like Brooklyn's Starrett City. When integrated neighborhoods -- particularly modest- income neighborhoods -- lose a critical mass of whites (the "tipping point," as it is called), the remaining whites tend to leave in an accelerating flood, creating a glut that depresses prices, thereby making the homes accessible to people of still-lower incomes, which reduces the neighborhood's attractiveness even for blacks who have options. There frequently follows a near panic- level exodus of people determined to get out while the getting is good. The result is a downward spiral that can quickly produce a slum.

There is a special fear of such a development in Starrett City, situated as it is between white working-class Canarsie and economically depressed, crime-ridden East New York. The critical factor is the absence of options, whether due to race or economics.

In the Charlottesville public-housing case, the white applicants, while officially poor, are, in general, less poor than the black applicants and for that reason, as well as for reasons of race, have somewhat greater options outside public housing.

But that isn't quite all of it. It is also true that, whether in public housing or not, white residents tend to get better treatment from landlords and more responsiveness from city agencies than do black residents. Many know at first hand of cases where apartment buildings that shift from mostly white to mostly black get less janitorial service, less security and less maintenance than formerly. And the usual result is that those residents with options -- those who won't take it anymore because they don't have to take it anymore -- move out, accelerating the downward spiral.

Does all-black equal slum? Of course not. Does housing occupied almost entirely by society's rejected tend to deteriorate into slums?

I believe it is possible, through wise policy, to avoid that eventuality; but for now, the answer, unfortunately, is: Yes.