ACADEMY OF SCIENCES. 83 
at least two gauges should be established—one at that side of the basin near- 
est ‘‘whence the prevailing winds come,’’ and the other nearest ‘‘ whither 
they go.’’? These gauges are not only checks on each other when the wind’s 
action is an insignificant element, but where the wind drives water from one 
portion of the basin and piles it up in another, they furnish data indispen- 
sable for harmonizing soundings taken on those and calmer days. 
In such a basin, when but one gauge is used, the proper place for it, theo- 
rectically speaking, is the center of the basin. These considerations attended 
to, each gauge is firmly fixed in a well sheltered spot, so that its zero shall be 
below low-water at neap, and its top above high-water at spring-tides. By 
proper circumspection for the site of each gauge, one will generally be found 
to answer for each station, but where the observation is made from shore two 
or more may at times become necessary—the observer following the tide from 
gauge to gauge as it goes out and retreating over the same path as it comes 
in. The kinds of tide gauges are as various as the circumstances demanding 
them. The one ordinarily used is of the simplest kind, a straight vertical 
post divided into feet and tenths, numbered from the bottom upwards; this 
is found generally to serve its purpose, inasmuch as when it is too windy to 
read the gauge correctly, it is blowing too much to sound accurately. A ver- 
tical tube with small holes at the bottom to admit the water which supports a 
float, is, however, susceptible of closer readings under all circumstances; and 
for getting off-shore tides, Mitchell’s gauge is admirably adapted; while as a 
self-registering gauge, Saxton’s stands without a parallel and leaves nothing 
to be desired. 
The zero of each gauge should be referred by means of a spirit-level, or 
otherwise, to a bench mark cut distinctly and durably on some permanent 
object (and the remark made in the book), so that, if displaced, it can be 
properly replaced in position. 
For the purpose of reducing the soundings, it is mainly essential that the 
tide-gauge and sounding-boat watches be together; but where the laws of the 
tides of the locality are also desired, it is best to keep either lunar or mean 
solar time. A series of observations of the tides on these gauges, made simule 
taneously with the soundings, furnish data for reducing each sounding to the 
reference plane—the mean of the lowest low waters. This plane is also given 
by these tidal observations. The frequency of the necessary readings of the 
gauge varies from every half-hour to every five minutes, according to the 
rapidity of the rise and fall of the tide. 
And now, if there be not on the shore permanent well defined objects that 
will serve as signals—such as spires, towers, flagstaffs, light-houses, or tall 
slender trees, fixed by triangulation—then the hydrographer erects the neces- 
sary signals; usually tripods boarded up, and painted white if projected on 
dark back-ground from the sounding-boat, or red (or black) if against the 
sky or a sandy back-ground. 
The tide-gauges and signals being erected, the next step is to determine 
carefully with a theodolite the relative position of these signals, and plot 
them by the computed sides of the triangles of which they are the vertices. 
It is, however, not imperative that the actual sizes of the triangles be at first 
