in the face, but this could not be regarded as an indication of their
general well-being. It was admitted that some girls left through
illness. Employees returned to their homes for January and February,
when the factory was closed down; there was also three days' holiday
in June. In the dormitory I noticed that each girl had the space of
one mat only (6 ft. by 3 ft.). Twenty-two girls slept in each
dormitory. The men connected with this factory were low-looking and
shifty-eyed.

An agricultural expert who was well acquainted with the conditions of
silk manufacture and of the district and was in a disinterested
position told me after my visit to this factory how the foremen
scoured the country for girl labour during January and February. The
success of the _kemban_ or girl collector was due to the poverty of
the people, who were glad "to be relieved of the cost of a daughter's
food." Occasionally the _kemban_ had sub-agents. The mill proprietors
were in competition for skilled girls, and money was given by a
_kemban_ intent on stealing another factory's hand.

The novices had no contract. The contract of a skilled girl provided
that she should serve at the factory for a specified period and that
if she failed to do so, she should pay back twenty times the 5 yen or
whatever sum had been advanced to her. Obviously 100 yen would be a
prohibitive sum for a peasant's daughter to find. The amount of the
workers' pay was not specified in the contract. The document was
plainly one-sided and would be regarded in an English court as against
public policy and unenforceable. Married women might take an infant
with them to the factory. In more than one factory I saw several
thin-faced babies.