r per cent out of 493 migrants whose occupations were
ascertained, were doing what may be called semiskilled
or skilled work, as puddlers, mold-setters, painters and
carpenters. On the other hand, in the South 59 out of 529
claimed to have been engaged in skilled labor, while a large
number were rural workers.

The following table shows the occupations of migrants in Pittsburgh as
compared with statements of occupations in the South:

Occupations In Pittsburgh % In South %

Common laborer 468 95 286 54
Skilled
or semiskilled 20 4 59 11
Farmer -- -- 81 15
Miner -- -- 36 7
Sawmill workers -- -- 9 2
Ran own farm or
father's farm -- -- 22 5
Ran farm on crop
sharing basis -- -- 33 6
Other occupations 5 1 0 0

It seems clear that most of the migrants were engaged in unskilled
labor. The reason given by the manufacturers in accounting for this
disparity were that the migrants are inefficient and unstable, and
that the opposition to them on the part of the white labor prohibits
their use on skilled jobs.[1] Ninety-five per cent of the negro
workers in the steel mills were unskilled laborers. "In the bigger
plants," says the investigator, "where many hundreds of negroes are
employed, almost one hundred per cent are doing common labor, while
in the smaller plants, a few might be found doing labor which required
some skill." Epstein believes that this idea is often due to the
prejudice of the heads of departments and other labor employers. A
sympathetic superintendent of one of the large steel plants said that
in many instances it was the superintendents and managers themselves
who are not alive to their own advantage, and so oppose the negroes in
doing the better classes of work. The same superintendent said that
he had employed negroes for many years; that a number of them had been
connected with his company for several years; that they are just
as efficient as the white people. More than half of the twenty-five
negroes in his plant were doing semiskilled and even skilled work. He
had one or two negro foremen over negro gangs, and cited an instance
of a bl