New Jersey presented a more complex problem. Here were Welsh and Swedes, Finns and Danes, as well as French, Dutch, Scotch, Irish, and English. A careful analysis was made of lists of freeholders, and other available sources, in the various counties. The results of these computations in the States from which no schedules of the First Census survive are given in Table B printed on page 28. The calculations for the entire country in 1790, based upon the census schedules of the States from which reports are still available and upon estimates for the others are summed up in the following manner: _Number and per cent distribution of the white population, 1790:_ _Nationality_ _Number_ _Per Cent_ All Nationalities 3,172,444 100.0 English 2,605,699 82.1 Scotch 221,562 7.0 Irish 61,534 1.9 Dutch 78,959 2.5 French 17,619 0.6 German 176,407 5.6 All others 10,664 0.3 To this method of estimating nationality, it will at once be objected that undue prominence is given to the derivation of the surname, an objection fully understood by those who made the estimate and one which deprives their conclusions of strict scientific verity. In a new country, where the population is in a constant flux and where members of community composed of one race easily migrate to another part of the country and fall in with people of another race, it is very easy to modify the name to suit new circumstances. We know, for instance that Isaac Isaacks of Pennsylvania was not a Jew, that the Van Buskirks of New Jersey were German, not Dutch, that D'Aubigné was early shortened into Dabny and Aulnay into Olney. So also many a Brown had been Braun, and several Blacks had once been only Schwartz. Even the universal Smith had absorbed more than one original Schmidt. These rather exceptional cases, however, probably, do not vitiate the general conclusion here made as to the British and non-British element in the population of America, for the Dutch, the German, the French, and the Swedish cognomens are characteristically different from the British. But the differentiation between Irish, Welsh, Scotch, Scotch-Irish, and English names is infinitely more difficult. The Scotch-Irish particularly have challenged the conclusions reached by the Census Bureau. They claim a much larger proportion of the original bulk of our population than the seven per cent included under the heading Scotch. Henry Jones Ford considers the conclusions as far as they pertain to the Scotch-Irish as "fallacious and untrustworthy." "Many Ulster names," he says,[5] "are also common English names.... Names classed as Scotch or Irish were probably mostly those of Scotch-Irish families.... The probability is that the English