Facts can be uncomfortable things -- when you're trying to justify legislation that would vastly increase legal immigration.

On May 15, I released a study predicting the Senate's immigration bill (S.2611) would bring 103 million legal immigrants into the U.S. over the next 20 years. On the same day, I participated in a news conference held by Sen. Jeff Sessions, Alabama Republican, who predicted similar numbers. The Senate promptly amended the bill, scaling back the legal immigration rate. (Even with amendments, the bill still would produce 55 million to 60 million legal immigrants over the next 20 years.)

Since the publication of my original study, Alan Reynolds of the Cato Institute has issued two columns charging that my study was "inane nonsense" and a "cheap parlor trick." Although Mr. Reynolds is an economist of merit, I must regrettably note he has seriously distorted my findings.

My original paper showed S.2611 would increase legal immigration through many channels, including family chain migration. One important channel was a new "guest worker" (H-2C) program. (These workers would have the right to permanent residence and citizenship.) The bill would allow 325,000 "guest workers" to enter in the first year and would increase the number by up to 20 percent per year after that, based on employer interest.

My paper predicted the H-2C "guest worker" would grow 10 percent yearly, well below the legal limit. When combined with other provisions in the bill, I estimated this would result in granting 103 million immigrants legal status over the next 20 years. Most of the paper described this 103 million immigrant estimate in detail, breaking future immigration into eight separate categories.

The paper also provided a second estimate showing what would happen if the "guest worker" program grew at the maximum legal rate of 20 percent yearly. This would have resulted in 193 million immigrants over 20 years. This higher "legal maximum" estimate was mentioned in only one paragraph and was clearly intended to illustrate my preferred estimate of 103 million was not a theoretical "worst case" scenario, but, in fact, well below the bill's legal ceiling.

Alan Reynolds has attacked my study by pretending I predicted the worst-case scenario of 20 percent growth in the "guest worker" program. It takes considerable chutzpah to allege that my paper, titled "Senate Immigration Bill Would Allow 100 Million New Legal Immigrants Over the Next 20 Years," predicted more than 200 million immigrants, but that's exactly what Mr. Reynolds does.

Nearly every number Mr. Reynolds cites involves an "estimate" I never made. For example, he charges I predicted an annual inflow of 25 million immigrants by 2026. I predicted nothing of the sort. Mr. Reynolds charges I predicted 10.4 million guest workers entering the country in 2026; the actual number in the paper is 2.1 million.

Mr. Reynolds further asserts my estimate of future immigration under S.2611 comes from a "cheap trick" based on the "magic of compound interest." But "compounding" had little effect on the actual estimate. In doing research for the original paper, I produced some 20 models of immigration growth under S.2611, varying growth in the "guest worker" program and other factors. One model assumed zero growth in the "guest worker" program but still yielded 72 million immigrants over 20 years, more than threefold the current law.

Other models assumed linear growth rather than "compounding" but yielded very similar results to those in the published paper. In one model, I assumed the H-2C program started at 325,000 entrants and grew at a fixed linear rate with incoming workers increasing by roughly 75,000 each year. Though there was no "magic of compound interest" in this model, the result was 104 million immigrants over 20 years.

I did not publish all these estimates, since releasing numerous similar predictions would merely have confused the message; however, I think the examples just given clearly illustrate my conclusions were in no way based on a compounding "trick."

Mr. Reynolds further charges my estimate of 103 million legal immigrants exceeds the population of Mexico, and "most guest workers are expected to come from Mexico." But my estimate of 103 million immigrants gaining legal status included many categories of immigrants besides guest workers; moreover, the bill is very clear the H-2C "guest worker" program is designed to bring in workers from all over the world, not just from Mexico.

By glaringly misrepresenting and distorting my research, Mr. Reynolds' criticisms detract from a meaningful discussion of immigration issues.

Robert Rector is a Senior Research Fellow in Domestic Policy Studies at the Heritage Foundation.