bull semen collected in normal artificial insemination practice
and compared with the expert's subjective assessment of their
motilities. Three determinations of recovery time were made on
each sample. Motility ratings were assigned according to the
nsual practice of inseminators and for the purpose of analysis
these ratings were regarded simply as scores on a linear scale

of increasing vigour from 1 to 10. The very long recovery times
associated with samples of poor motility suggested comparing the
logarithms of the recovery times with the motility scores. . Linear
regression relating these two quantities was then fitted. Devia-
tions from regression were not significant indicating that the
relationship could be represented by a straight line. The slope
of the line was significantly different from zero again indicating
a definite relationship between log (recovery time) and motility
score. Differences in the mean values of log (recovery time)
between ratings were also significant.

There was considerable scatter of the readings of the log
(recovery times) about the regression line. An attempt to find
the source of this scatter was made by experimental measurement
of the repeatability of the recovery times. On a number of sub-
samples from the same ejaculate the standard deviation of the
times was 0.03 (log units). In comparison with this, the standard
deviation of the sample means about the regression line was about
ten times larger, viz 0.3 (log units) which suggested that the
variability was largely in the subjective ratings.

In the evaluation of the instrument it was ascertained that
the variation of sperm concentration encountered in practice did
not invalidate the method. Within a concentration range of 1 : 5,
such as normally occurs, it was found that the recovery times were
independent of concentration. The method was found suitable for
the samples of low motility and low concentration and in this
respect it was more useful than the impedance change frequency
method of Rothschild? which was reliable only on samples of high
motility and high concentration.

Future developments. Attempts to work at dilutions up to 1 : 20
showed that measurements of recovery time could just be made, i.e.
‘the photocell just detected changes in light intensity during and
after shearing. This is of interest in the growing practice of
preserving spermatozoa by freezing which involves dilution of
semen at this rate with glycerol/egg yolk/citrate. Application
of this method of measuring motility awaits the preparation of
optically clear egg yolk.

The use of the oscilloscope described here for measuring

1046