Seeking to discredit Microsoft Corp.'s final witness at its antitrust trial, the government yesterday called attention to his $800 hourly fee, errors in a key chart he displayed and his lack of familiarity with topics on which he testified.

The full-scale credibility attack on Richard L. Schmalensee, the dean of the Sloan School of Management at the Massachusetts Institute of Technology, was intended to undermine his testimony earlier this week that Microsoft's business behavior has not violated the nation's antitrust laws.

Government attorney David Boies began a confrontational cross-examination of Schmalensee by inquiring how much he has been paid by Microsoft to appear as an expert witness. Schmalensee replied that he could not provide an exact figure. "I simply don't have the records," he said. "I haven't addressed the subject in my mind."

But upon further prodding, Schmalensee estimated that he has billed Microsoft for at least $250,000 in consulting services over the last two years -- at $800 an hour.

Earlier in the proceedings, a Microsoft attorney argued in court that the government's chief economic expert, MIT professor Franklin M. Fisher, stands to profit if the government wins the case because several private firms that are considering suing Microsoft have retained him as a consultant.

"If there is an incentive to [tailor] testimony because of payments, this witness has a lot more incentive than Professor Fisher," who was paid $500 an hour by the government, Boies said outside the courthouse.

Boies also sought to dispute Schmalensee's contention that the Netscape Internet browser continues to be a hearty competitor to Microsoft's Internet Explorer product.

Schmalensee had displayed two charts to show that usage of the Netscape browser continues to grow. The first one showed 6 million to 7 million people had obtained the Netscape product with computers between spring 1998 and spring 1999. The second chart, however, showed that the average number of people who obtained Netscape browsers during the same period was 8 million.

"A comparison with the one on the right suggests there is a difficulty," Schmalensee conceded.

"There's more than a difficulty," Boies shot back. "They cannot be reconciled."

Boies also argued that Schmalensee neglected to employ the same market-research data used internally by Microsoft because it did not support his conclusions.