PROCEEDINGS FOR 1902 XXI 
It has become evident that from a tidal point of view, the St. 
Lawrence may be divided into two regions. From the head of tide- 
water at Lake St. Peter to the Traverse, which is a little below Orleans 
Island, the tides and currents can best be referred to Quebec. Below 
this, in the open estuary as far as Anticosti, and in Chaleur Bay, the 
tides are more advantageously referred to the principal tidal station 
at Father Point. This accords with the natural features of the estu- 
ary, as the deep water runs up as far as the Traverse; and Quebec 
must be considered as in the river, about the true head of the estuary. 
It is also just below Orleans Island that the tide has its maximum 
range and the currents their greatest strength. 
To carry out this sub-division practically, it is evidently necessary 
to have tide tables for Father Point itself. With this object, the 
difference in the time of the tide between Father Point and Quebec, 
as given by two complete years of simultaneous observations, was 
carefully examined into. The observations gave the difference in 
time for 1260 consecutive tides; and it was found that this difference 
varied during the course of the month, the variation being greater 
in the case of low water. This variation was of a double character, 
firstly, in the period of the synodic month with the moon’s phases; 
and secondly, in the period of the anomalistic month with the moon’s 
distance. The total amount of the variation for the low waters, was 
41 minutes more or less than the mean value. In the endeavour 
to obtain a tidal difference with less variation than this, trial com- 
parisons were made with Wilhelmshaven, Germany; Harwich, on the 
North Sea; and Portsmouth, on the English Channel. The tide in 
each of these harbours is similar to Father Point, in having nearly 
the same range; and it might therefore be expected that one of these 
differences would prove to be more nearly constant than the difference 
with Quebec. This was not the case, however; but by an analysis 
of the difference in terms of the two periods above mentioned, a 
double series of variable differences were obtained by which the tide 
at Father Point can be correctly calculated from the tide tables for 
Quebec. The series used are given in full in the present report. This 
is the best method available until this Survey can afford the cost of 
determining tidal constants for Father Point itself, from the record 
which has already been secured there. 
This investigation is of interest as an example of the use of vari- 
able tidal differences. These have been much employed by this Survey 
in the calculation of local tide tables from ports of reference. It may 
never be practicable to secure direct astronomical data for every port in 
a country for which tide tables are needed; but by means of tidal differ- 
ences which vary in the period of one of the astronomical months, 
