The Americas comprise 55 states and territories with permanent populations, from Alaska to Tierra del Fuego. How many still have a total fertility rate above replacement? Four: Haiti, Guyana, French Guiana, and Saint Martin. These four hold about 13 million people. The Americas hold about 1.05 billion. So fewer than 1.3% of the hemisphere's population lives in areas where fertility is still above replacement. The data for Haiti, Guyana, French Guiana, and Saint Martin are not good. Still, most observers working from recent figures put their TFR between 2.4 and 2.6. My guess is that all four will fall below replacement before 2035. Among the region’s demographic giants, the U.S. now has the highest TFR, ahead of Brazil, Mexico, Colombia, and Argentina. This is not a Hispanic effect, though Hispanic fertility is relatively high. The non-Hispanic white TFR in the U.S. exceeds the TFR of Brazil, Mexico, Colombia, and Argentina. A few observations. First, none of the four holdouts speaks Spanish or Portuguese. Three of them, Haiti, French Guiana, and Saint Martin, are francophone, even if local creoles are more common than standard French. The fourth, Guyana, is anglophone. I stress this because many observers have not noticed that all of Central America, Guatemala and Honduras included, is already below replacement. A quick-and-dirty calculation puts the current TFR of the main language groups in the Americas at Portuguese: 1.5, Spanish: 1.55, English: 1.55, and French: 2.0. Second, at this rate, the Americas will begin recording more deaths than births around 2040, perhaps sooner. From that point, all population growth will depend on immigration from outside the continent. Finally, a few caveats. The data for many countries, especially in the Caribbean, are awful. I doubt better data would change any of these conclusions, but I want to be careful. And given how fast the TFR has fallen over the past decade, part of the drop may be a tempo effect. I lean toward those who think tempo will not matter much here, but I may be wrong. And, as always, whenever I can, I use data from each country’s National Institute of Statistics rather than the UN WPP (or databases like the World Bank’s, which rely on it). On a few occasions, to update time series, I built the TFR myself, using past TFR, CBR, and population data as input. While this creates some noise, none of it is particularly important to my arguments.