unreliability of basing general deductions upon data obtained
during a limited period only. At the time of the spring tides
at the beginning of June the conditions were not favourable to
big tides, as although the moon was approaching her perigee,
her declination had nearly reached its northern limit and the
declination of the sun was 22 IN The first quarter of the moon
coincided very closely with the moon's passage over the
equator, so that the neaps would be bigger than usual. At the
period of the spring: tides, about the middle of June, although
the time of full moon corresponded with her southernmost
declination, she was approaching her apogee, and the
declination of the sun was 23 16' N., so that the tides would
be lower than usual.

In order to ensure accurate observations, the position chosen
for the tide gauge should be in deep water in the immediate
vicinity of the locus in quo, but so that it is not affected by
the waves from passing vessels. Wave motion is most felt where
the float is in shallow water. A pier or quay wall will
probably be most convenient, but in order to obtain records of
the whole range of the tides it is of course necessary that the
float should not be left dry at low water. In some instances
the float is fixed in a well sunk above high water mark to such
a depth that the bottom of it is below the lowest low water
level, and a small pipe is then laid under the beach from the
well to, and below, low water, so that the water stands
continuously in the well at the same level as the sea.

The gauge should be fixed on bearers, about 3 ft 6 in from the
floor, in a wooden shed, similar to a watchman's box, but