t-one per cent protests, and the most necessary of these the
Bureau investigates. Perhaps ten or a dozen in the entire country get a
recount. The Bureau doesn't officially recognize some of them but sends
an inspector to look over the ground, and see if everything was done
right. That's what we're going to do in New York."

"All right," said Hamilton briefly.

"You'll be on that train?"

"Yes, Mr. Burns," the boy answered. "Eleven-fifty P.M., Tuesday."

The opportunity was one which Hamilton had been coveting, for he felt
that if he only had a chance to get at the city methods he would have
covered almost the entire ground of the field-work of the Decennial
Census, and while he was sorry to leave his Porto Rican friend, still
the novelty appealed to him greatly, and in spite of his former chief's
mathematical conversation, Hamilton was genuinely fond of him.

"I've been wondering, Mr. Burns," the boy said, as they stood in the
great concourse of the Union Station at Washington, "whether there would
not be a very large number of protests about census figures,--people
always seem to have such an exaggerated idea of the size of their own
towns."

"There is to some extent," Burns replied. "I think something like a
hundred places filed protests in this last census."

"Then I read something, too, about census frauds," Hamilton said, "soon
after the taking of the census, in which it was suggested that some
enumerators--who were paid per capita--had bolstered up the figures in
order to get more out of it."

"There was a little of that," the Inspector said, "but by far the
greatest amount of fraud was due to the desire on the part of the
inhabitants of a town or city to make the place appear larger and more
important. Tacoma, Washington, was the most flagrant example of this,
why, they padded 32,527 names there, and even when the Census had made a
recount they tried to repeat the same performance, complaining of the
results and demanding a second recount."

"Was this granted?"

"It was," the Inspector replied, "largely in order that the Census
Bureau itself might have an opportunity to check the correctness of its
methods. The second recount was performed by expert statisticians and
with extreme care."

"And how did it come out?" the boy asked.

"It substantiated the first recount in every way. It was, indeed, a
wonderful object lesson in showing how small is the margin of error in
the United States Census."

"B