In October 1975, the Environmental Protection Agency (EPA) awarded a contract to the District of Columbia to conduct a study of harmful elements that may have been in the city's water. The contract was terminated about three years later. No final report was submitted.

In his evaluation of the project on Jan. 17, 1979, EPA's Elbert Dage wrote: "The contractor made no attempt to follow standard scientific protocols or rationale . . . Monthly progress reports were not submitted . . . No financial reports were submitted. The draft final report was completely unacceptable, not understandable, illegible, grammatically incorrect, filled with irrelevant information . . . and devoid of required illustrations or graphics."

By then, $95,706 had been paid out for the study.

This is not an unusual occurrence in the world of government contracting, where $1 million is spent for every hour in the year. The federal system of awarding work to outside consultants and researchers -- and then monitoring that work -- has spawned an epidemic of conflicts of interest, favoritism and waste.

Today, in the third article in a series on government contracting, The Washington Post presents more case studies in waste by the federal bureaucracy, beginning with an example concerning the drinking water in the nation's capital.

The study of elements in the city's water supply was done by the D.C. Department of Environmental Services, with assistance from the National Bureau of Standards.

In an interview, Bailus Walker, who was the project officer for the District of Columbia, said: "There were some problems with the contract, but they don't come to mind at the time. I don't think that it was the highest quality. I truly can't recall the specifics. I wasn't that close to it."

The contract was riddled with problems from the start, Dage, the EPA official who wrote the scathing evaluation, said in an interview that the EPA was somewhat to blame for the poor performance.

"There was something like six different EPA project officers on the contract at different times," he said. "It unfortunately got tossed around there [D.C.] as much as it did here [EPA]."

Arnold Clark, bureau chief of the D.C. consumer health division, said that "this whole thing was screwed up."

Clark noted that two of the city's food sanitarians -- Carl Diener and George Wall -- were chosen to work on the study. He said they were both very good workers but he couldn't understand why they were put on this particular contract.

"They didn't know anything about that sort of thing [metal levels in water]," Clark said. "They were food sanitarians . . . I think they were chosen just to impress the federal government because they had good credentials."

The preliminary report indicated that levels of cadmium, mercury and lead in the D.C. area were "within a 'safe' range."

Dage called that conclusion "highly questionable" in a Dec. 12, 1978, letter. "Information from existing literature and local poisoning prevention programs contradicts this conclusion. Lead poisoning in children is fairly common in certain areas of this city," wrote Dage.

In the end, the government was left with no reliable information about lead, mercury and cadmium in D.C. water. And this from a project that was original intended to serve as a model for other cities in testing metal levels in water.

Dage said he did receive a request from a New England town to use the D.C. study as a model. "What I told them," he recalled, was that they should use the study as an example of how NOT to do it." 'Like Murphy's Law'

On Nov. 14, 1977, the Department of Housing and Urban Development (HUD) kicked off what was to be a major effort to reduce crime rates in public housing. As part of that effort, the government awarded an $89,224 contract to BDM Corp., a McLean consulting firm.

The contract ended without its central mission ever being undertaken.

"Everything that could go wrong did. It was like "Murphy's law,'" recalled HUD's program officer, Richard Burk.

William Young, BDM's project officer, did not dispute that assessment. "Clearly the purpose of the study was aborted," he said in an interview. "I've been in the business 12 years. It's not something that comes as a surprise to me."

For HUD, there was more at stake than an academic interest in crime deterrence. A survey completed before the award of the BDM contract found the incidence of robberies at selected public housing sites were three times that of the national rate, and that sexual assaults were twice the national rate.

The plan was to conduct surveys of break-ins and burglaries at selected public housing sites before and after the installation of special security doors, windows and locks. The surveys were to be conducted in Washington, Baltimore and Los Angeles.

The results would be used to determine how many and what kinds of security features should be purchased nationwide for public housing complexes.

The project got off to a slow start. For months after the contract began, the National Bureau of Standards which coordinated the project along with HUD, was still testing the security equipment that was to be used.

On Oct. 18, 1978, nearly a year after the study was supposed to have begun, HUD decided to delete Los Angeles from the program, to save money. Shortly after that the public housing site in Washington chose not to buy security equipment, and also dropped out of the survey.

Thus, a year into the study, two of the three cites involved were eliminated from the survey. Only Baltimore was left.

"They then concluded that the information obtained from the study of one site (Baltimore, Md.) would be insufficient to reach any general valid conclusions," wrote contract specialist Norm Zima, in an internal National Bureau of Standards memo.

Tenants and crime victims in Baltimore and Washington had been interviewed for the study -- but none of the security features was ready to be installed.

A year after the study began, the government terminated the contract. Federal officials said the results could not be used reliably.

"I was shocked," said John Stroik, a project leader for the National Bureau of Standards. The government had spent $55,083 without answering the basic question: Would the security features reduce crime?

Without benefit of the study, public housing sites around the nation may spend more for security features than they need, or may shy away from them completely, having no evidence that they work, say government program officers and BDM's Young.

"Some tenants are kind of jaded by now," Stroik said. "They see this thing as being just another one of those government actions that would not do anything for them personally.

"It's hard for them to see what collecting data has to do with their lives. I think this one would have been different. But we'll never know, because it was never completed." 'The Phantom Contract'

At EPA, there is a project they used to call "the phantom contract."

At the firm with the contract -- Optimal Data Corp. in Huntsville, Ala. -- the company owner, Lincoln Teng, said of the EPA project: "It's all kind of pretend."

On June 28, 1974, the EPA awarded Optimal Data a "support service" contract paying the firm $43,442 for its services in compiling various data.

"We had engineers ready to work on the contract," Teng said. "It was supposed to be for $350,000 . . . but they never asked us to do anything. They just gave us a big contract, and then they asked us to recompile data they already compiled. It was really nothing."

Teng said he suspects the EPA gave his firm the contract simply because Optimal Data is a minority-owned firm. The government requires a certain number of contracts go to such firms.

Joseph McSorley, an EPA project officer, said in an interview that this is not so. EPA stopped the contract after $43,442 he said, simply because the EPA couldn't find the company.

"I don't know if they exist any more," McSorley said. "We tried to track them down for some voucher information, but the contracting officers couldn't find them.

McSorley said his main questions were: What was their address? Who was working for them? And did they still exist? "They go for seven or eight or nine months without billing us at all," he said. "And when they did bill, they wouldn't list any names of who was doing the work."

Teng, who lives in Huntsville and teaches at the University of Alabama, said in an interview: "We had an office in Atlanta . . . maybe they tried to reach us there . . . That address is no good anymore . . . Optimal Data completely stopped in 1976, the little EPA project was our last contract. . . . Legally, the company still exists, but we don't have an office and we don't have any activity going." The 68-mph Trucks

In the summer of 1974, the Environmental Protection Agency decided to measure the truck pollution emission levels in the country's two largest cities -- New York and Los Angeles.

The contract was given to a private firm, Wilbur Smith and Associates, and monitored within the EPA by project officer Roy Higdon. Some of the first data that came in stunned Higdon. It showed that trucks were driving through Manhattan at an average speed of 68 miles per hour.

"I couldn't believe that tractor-trailers were running around downtown New York City at that speed," Higdon recalled. Bewildered, he traveled down to the consulting firm's office in Columbia, S.C., and examined the raw data. The recording machinery used in the tests was all out of whack, he discovered -- so much so that trucks that were standing still were recorded as going 25 miles per hour.

Wilbur Smith and Associates was awarded the $124,022 contract in the first place, Higdon explained, because it had been the low bidder. But the contract ended up costing far more than the original bid -- there was a 200 percent cost overrun. And even with that, when the contract ended, less than half the assigned work had been completed.

There is general agreement that the project accomplished little of what it was supposed to, but there is dispute as to who was at fault.

Jack Cosby, the project director for Wilbur Smith, said in an interview that the major problem in the study -- and the reason for the massive overrun -- was that the testing equipment was faulty. He noted that the equipment was provided by EPA.

"If you're reading garbage, it's garage," said Cosby. "And nobody can interpret it."

Higdon took a different view. "I was constantly commuting to New York to do the work we were paying someone else to do," he said. "Fifty percent of what they did in New York was useless. I had to redo the work myself."

After $360,982 was spent in New York, the contract ended. Wilbur Smith never tested a single truck in Los Angeles. Instead, Higdon went to Los Angeles with a staff of EPA engineers and conducted the entire West Coast truck pollution study without the help of an outside contractor.

"You can't be proud of this one," said Wilbur Smith's Cosby. "You don't hit a home run every time you're at bat."

In hindsight, said EPA's Higdon, "we could have gotten the work done faster and cheaper if we never awarded a contract."

Higdon resigned from EPA in 1977, after eight frustrating years on the job. "I, being a taxpayer, watched thousands of dollars go down the tubes and it teed me off," he said. "But the government is notorious for contracting out what it could do in-house . . . these contracts, sometimes they are just plain ridiculous.' 'Never Again'

Late in first year as secretary of commerce, Juanita Kreps decided to measure how well corporations live up to their social, economic and environmental responsibilities. For Kreps, the subject had the highest priority. The contract memorandums were appropriately stamped "Urgent."

In a three-day period in October 19777, the Department of Commerce awarded two contracts totaling $223,462 to Human Resources Network, a Philadelphia consulting firm.

The firm was to develop a way to measure corporate responsibility and was to sponsor 12 conferences around the nation to bring government and corporate representatives together.

The Commerce Department was in such a hurry to begin the contracts that it waived the standard precontract audits. The consultants -- seven in all -- were each to be paid $400 a day for their work.

But six weeks after the contract started, a House subcommittee headed by the late Rep. John M. Slack (D-W.Va.) questioned the legitimacy of the program. It came under sharp attack at an Appropriations subcommittee hearing on Jan. 23, 1978, when several congressional suggested that the issue was better left to the corporations.

Within days of the hearing, both contracts were terminated by the Commerce Department.

"We were stunned," recalled Steven Nowlan, the firm's project manager. "We didn't think we were dealing with a full deck of cards in Washington."

The young Philadelphia firm only weeks earlier had hired new staff members, written letters to corporate clients asking for support, and begun preparing for the study and the conferences.

Now, it could only submit a brief report and pack up its notes.

"Some of the stuff we threw away. Some of it sat around a couple of years. Some of it still resides in a big file marked "Department of Commerce," Nowlan said. "We were made fools of in public. We never want a federal contract again."

For the nation's taxpayers, the tab was $72,811.

On April 10 of last year the Internal Revenue Service announced that it had awarded a $114,676 contract to CSR Inc. The purpose of the contract was "to conduct a general taxpayer opinion survey."

Some 60 questions posed at 5,000 households around the country probed taxpayers' opinions on how well the IRS collects taxes.

None of the questions sought to learn whether the taxpayers were happy about how the federal government is spending their money.