How do you limit money if you falter in measuring it?

The president of the Federal Reserve Bank of Minneapolis, E. Gerald Corrigan, tells a story about talking recently to some 20 or so United Steelworkers union local presidents in Hibbing, Minn., and getting a surprise.

After the session was opened to questions, one of the Taconite iron ore mine workers asked: ''Why was M-1 down $3 billion last week?''

Mr. Corrigan might have expected such a query from a bond dealer in New York. But not in northern Minnesota. ''It just kind of confirmed that we do indeed have a generation of money supply junkies on our hands,'' Corrigan noted.

Oddly, however, just when the public is getting educated as to the importance of such money supply statistics as M-1 (the sum of transaction-type accounts in banks and currency in circulation), the ''Ms'' are coming under attack.

''The pace of financial innovation has led us to the point where any definition of the money supply must be arbitrary and unsatisfactory,'' Frank E. Morris, president of the Federal Reserve Bank of Boston, told a conference at the Atlanta Fed last month.

Mr. Morris admits that his opinion as to the unreliability of the money supply figures as a proper indicator for the nation's vitally important monetary policy remains ''very much of a minority view.'' More people agree with him outside the Fed than inside the central bank, he noted in a telephone interview.

The Minneapolis Fed's Corrigan, for instance, says he's not as troubled with the measurement problem of the M's as Morris. ''I still think we can work with these things.''

Most economists nowadays regard monetary policy as being at least as important as, and probably more important than, fiscal policy in steering the nation's economy. If the Fed creates too much money, the economy rides up on the shoals of inflation; with too little money, the economy founders on the reefs of recession.

But if the Fed can't measure money properly, how is it to know whether it is pumping out too much, too little, or just the right amount of new money?

Morris complains: ''. . . we can no longer measure the money supply with any kind of precision.''

He sees several factors as complicating the money supply picture:

1. The sharp rise in interest rates means fewer people will leave their money in noninterest-bearing bank demand deposits. They move the money to various bank certificates of deposit, cash management accounts offered by brokerage houses, money market mutual funds, or other financial instruments which pay interest.

2. Computers have reduced to minimal levels the cost of transferring such liquid assets as savings accounts or money market funds into a bank demand deposit, thus providing immediately usable money.

3. Regulations require banks to set aside a larger percentage of demand deposits or other ''transaction accounts'' as reserves. Those reserves earn no money for the bank. Says Morris: ''This gives an advantage to institutions not subject to reserve requirements to offer a similar financial service on a more advantageous basis.'' Money market funds, for instance, have no reserve requirements.

These factors, Morris argues, have created enormous incentives for innovation in the financial field. For example, market funds have grown enormously -- and some portion of those funds should be regarded as ''money,'' since it is used for transactions, he adds.

Another innovation likely to have the most impact on the monetary aggregates in the next few years, Morris says, is ''deposit-sweeping.'' With this attractive service, deposit balances over a specified amount are shifted automatically into an income-earning asset on a daily or weekly basis. For instance, a brokerage house cash management account will sweep surplus money into a money market fund. Banks are starting to do the same thing. The Fed will face the problem of having to decide whether the funds in the money market funds is really ''money'' or not.

Speaking of the school of economists that concentrates on the money supply as a guide and indicator of economic policy, Morris said: ''The monetarists must feel nervous about this. The foundation of their religion is eroding.''

One such monetarist, Allan H. Meltzer, of Carnegie-Mellon University, maintains that some of the problems of measuring money could be eased by various technical adjustments. The Fed could stop seasonally adjusting the figures. It could replace ''lagged reserve'' accounting with ''contemporaneous reserve'' accounting. At present, banks set aside reserves according to their deposits two weeks earlier. Dr. Meltzer would like reserve requirements set aside in accordance with current deposits. Further, the discount rate -- the interest rate the Fed charges on loans commercial banks take to meet their reserve needs -- could be allowed to float somewhat above the federal funds rate, the interest rate banks charge each other for money borrowed overnight to meet reserve requirements.

Mr. Morris's solution is for the Fed to set a goal of a specified amount of growth in ''total liquid assets'' -- a measure that includes money, broadly speaking, plus such short-term financial assets as Treasury bills, commercial paper, bankers' acceptances, and savings bonds. It has a close relationship to gross national product.

But to reach that target, Morris would control bank reserves, a much smaller figure and fully controllable by the Fed.

So, though he criticizes the monetarists, the Fed bank president comes out rather close to them. Dr. Meltzer, for instance, wants to target monetary policy by controlling the so-called ''monetary base,'' which includes bank reserves and currency.

What this argument boils down to is that those on both sides want to limit and control the amount of financial assets fed into the nation's economy so as to reduce inflation. But they can't quite agree on the technical details of how to do so.