74 THE ROYAL SOCIETY OF CANADA 
arriving at a satisfactory. system of calculation for these Narrows, 
would constitute a valuable contribution to this subject. 
In dealing with a series of passes on the Pacific coast, which carry 
a heavy traffic, it was not practicable to publish tables of slack water 
for them all. Two of them were selected as standard passes, and the 
others are referred to these; as the difference in the time of slack water 
in neighbouring passes is found to be very constant. . For the standard 
passes themselves, the method of reference to two tidal stations in 
the opposite directions for high-water and low-water slack respectively, 
has recently been adopted with marked inprovement. The only 
departure from constancy is for annual variation; and this can be 
sufficiently allowed for, in the calculation of slack water, by a series of 
differences for successive calendar months. 
The possibility of correlating the turn of the current with the 
time of the tide may be limited, where an extreme complexity is 
reached, because of two distinct types of tide.at the two ends of a 
strait. This is the case in the Gut of Canso; and its currents have 
baffled hydrographic surveyors as well as navigators. At times, there 
is a diurnal tide at one end while the ordinary semi-diurnal tide con- 
tinues at the other. As a consequence, the current may turn only 
once in the day, while at other times it turns twice, as tidal streams 
usually do. It was found, after careful investigation, that this change 
in behaviour is in complete accord with the moon’s change in declina- 
tion. There is a further complication however, due to dominant flow 
in one direction through the strait. Even when the moon is on the 
equator, and the duration of the flow should become equal each way, 
the current runs longer in the dominant direction. Under the con- 
ditions obtaining, it is only possible for the current to have a definite 
relation to the time of the tide when the moon is near the equator; 
that is, during two groups of about three days each which occur 
twice in the declination-month of 27} days. 
Narrow inlets to large areas —The problem in this case, would be 
reduced to its simplest conditions if a tide of uniform amplitude rose 
and fell rhythmically before a narrow entrance to an infinite area; 
that is, an area so large that there was practically no change of level 
within it. Although the current might be violent at high and low 
water, the time of ‘slack water would correspond closely with half 
tide, when the level inside and out becomes momentarily the same. 
In this class of inlet, if the current is reasonably moderate, there 
is a considerable choice of method admissible. (1) As slack water 
must be about the time of half tide, a comparison may be made with 
either the previous or following high or low water, in the endeavour 
to find the most constant time-interval between the two. (2) Instead 
