The political situation was further complicated when I wrote an op-ed in the Wall Street Journal in the fall of 1990, which pointed out that EPA's own data showed that fuel oxygenation made no sense compared to the repair of broken cars. I strongly recommend this method of publication. The article was submitted on a Monday, accepted by Wednesday, rewritten and all the numbers and references checked by their excellent editorial staff on Friday, published the following Monday, and I got paid! The scientific journals in which our studies are normally published do not meet any of those criteria.

This article, which included the fact that we could identify those vehicles needing repair without inconveniencing the owners of the others, was read by Texas Congressman Joseph Barton (R-Ennis, Texas, one of a handful who never bounced a check). As a member of the House Commerce and Environment Committee, he introduced an amendment to the 1990 Clean Air Act Amendments that included the phrase "... including on-road emission monitors" at two locations where mandatory emission testing was legislated. These eight words in a 750 page act apparently did not please the USEPA, which tried to remove them with a felt pen as a technical amendment (as if they were a misprint). They were caught by the congressman's aide's girlfriend.

After the act was signed into law (President Bush, Nov. 1990), the EPA decided to removed the words by "interpretation." Interpretation proceeded in two steps. In the first step, on-road monitoring was defined as roadside pull-overs and tailpipe tests, or remote sensing, or on-board emission monitors on the passenger seat and plugged into the tailpipe, or any other "on-road" program the states could devise, and in any case they would get no credit from the EPA computer model.