reparation for the failures already charged. Of the 1,657
examinations of this kind to satisfy for failures, 30.7 per cent result
in success. The boys are more successful than the girls by 4.5 per
cent. This particular procedure is not employed by more than two of the
eight schools. The other form of school examination employed for
disposing of failures is the special examination, usually following
some definite preparation, and given at the discretion of the teacher
or department head. Its employment seems also to be limited pretty much
to two of the schools, because for most of the subjects the Regents'
examinations tend to displace it in the schools of the New York State
and City systems. As only the successes were sure of being recorded in
these tests we do not know the percentage of success attributable to
this plan of removing failures. It probably deserves to be credited
with a fairly high degree of success, for relatively few pupils (less
than 200) utilize it, and then frequently after some extra preparation
or study--such as summer school courses or tutoring. These two forms of
school examinations jointly yield 37.5 per cent of successes on the
number attempted, so far as such are recorded.

4. THE SERVICE RENDERED BY THE REGENTS' EXAMINATIONS IN NEW YORK STATE

Whatever may be the merits or demerits of the Regents' examination
system in general for academic school subjects, these tests certainly
perform a saving function for the failing pupils, by promptly
rectifying so many of their school failures and thus rescuing them from
the burden of expensive repetition. A pupil's success in the Regents'
examination has the immediate effect of satisfying the school failure
charged to him. At the same time, it is possible, as is sometimes
asserted, that the anticipation of these tests inclines some teachers
to a more gratuitous distribution of failing marks as a spur to their
pupils to brace up and perform well in reference to the Regents'
questions. However, there is no trace of that policy found so far as
the schools included in this study are concerned. For the three New
Jersey schools considered jointly have a higher percentage of failing
pupils, and a slightly higher average in the number of failures for
each failing pupil than have the three New York State schools.

But it is more probable that the attitude referred to operates to
exclude the failing pupils from being freely permitted to enter the
Regents' tes