According to this paper* there were estimates of ~600k remaining in 1800. 3.9million citizens in 1790 USA -> nontrivial number and effect on the mean centroid,

According to this paper* there were estimates of ~600k remaining in 1800. 3.9million citizens in 1790 USA -> nontrivial number and effect on the mean centroid, especially considering they are not uniformly spread. As for no discontinuity between 1840-1860, it does say 'living in the general population'. So maybe that excludes the remainder. The 'settled' areas being only a small fraction native of course depends on the time you measure it relative to the disease onset... which often comes after the first settler. I think at first the ratio is the other way. *: AMERICAN INDIAN MORTALITY IN THE LATE NINETEENTH CENTURY: THE IMPACT OF FEDERAL ASSIMILATION POLICIES ON A VULNERABLE POPULATION. http://cid.bcrp.gob.pe/biblio/Papers/NBER/2006/Octubre/w12572.pdf http://cid.bcrp.gob.pe/biblio/Papers/NBER/2006/Octubre/w1257...