154 ANNUAL REPORT 
used very satisfactorily. Careful tests by Professor Edward Orton, Jr., 
of the dissolving action of alcohol upon cement showed that practically 
nothing was brought into solution. 
The specific gravity of the alcohol used fluctuated around 0.811. In 
determining the sizes of the various sediments a well-known brand of 
Portland cement was used and the following results were obtained: 
Sediment. Masiaumsize Minimum size in inches. eer size in inches. 
F | 
1 minute settling | 0.008 | 0.0002 0.00200 
3 minutes settling 0.002 0.0002 | 0.00072 
9 minutes settling 0.00075 0.00007 | 0.00030 
27 minutes settling 0.00034 Too small to measure 0.00020 
a 
The average size is obtained from 50 measurements. 
Under the head of physical testing of Portland cement attention will 
be called to the importance of the mechanical analysis of cement. 
Analysis by the Elutriation Method.—This fine method depends 
On. the fact. that sthe \carmyine power Olu a) Culienty sonmrcamy, 
liquid is a function of the velocity. Knowing the velocity of a cur- 
rent of water, for instance, we can determine the size of the particles 
removed by it, and constructing an apparatus in which a sample of clay 
or soil is subjected to washing by an upward current of water, by start- 
ing with a given low velocity we can remove the fine grains, clay sub- 
stance and dust completely, so that with this velocity after awhile the 
flowing water is perfectly clear. If, now, the velocity is increased the 
larger particles will be carried off, and so on. 
The law governing this separation is governed by Newton's law of 
gravity and is modified as follows: 
3Z 
d=v’?————_ 
4g (s-l) : 
Where d is the diameter of the particle, assuming that all grains 
are spherical, v the velocity of the water current, g the acceleration of 
gravity, 9.81 m, s the specific gravity of the particle and z a constant de- 
pending upon the surface conditions, its mean value being 2.65. ‘The 
expression, by substituting the above values, becomes: 
d=v*o.0000255 millimeters. 
This formula assumes that all particles are of equal hydraulic value 
and that they are spheres of uniform diameter. 
The result of many analyses have shown that the above formula must 
be corrected to 
d=v"™/! 0.0314 millimeters. 
