the manufacturer. The data given in columns 4, 6 and 8 was calculated, the water horsepower being obtained from the formula:

W.H.P. total head (in feet) X capacity (G.P.M.)

=

3,895

the constant 3,895 being used as the water was salt, this constant being obtained by a careful determination of the density of the water in use.

All data given in Table No. 2 is plotted on Figure 3 and an inspection of the curves shows that these curves are in general the same as those generally given for circulating pumps running on condensers, that is, the capacity curve is a straight line, the capacity varying directly as the revolutions; the head varying as the square of the revolutions and the horsepower approximately as the cube of the revolutions. The motor efficiency and set efficiency curves are each plotted as two curves, the motor being run on either 120 or 240 volts, the 120-volt connection being used on speeds up to about 360 r.p.m., and the 240-volt connection for speeds above that. It will be noted that there is a change of curvature in the pumpefficiency curve between 400 and 500 r.p.m., but this same curvature was found on two test runs, and it will be noticed in a description of the tests of the circulating pumps on the Nevada and North Dakota as given in this publication in February, 1918, that this same change occurred in the efficiency curves of these pumps.

The New Mexico's condensers, the details of which are shown in Figure 4, are of the Bureau of Steam Engineering's standard circular type, and the arrangement of pumps, condenser and piping are shown in Figure 1 for the starboard main turbo-generator. The circulating pumps were built by the Alberger Pump & Condenser Company and the driving motors by the General Electric Company, the unit tested bearing the following data on the name plates: