Geological Survey of the State of New York. 97 
work, in supplying the material for laying the numerous locks 
and aqueducts throughout its whole length. | 
Analysis and experiments with natural and artificial water ce- 
ments, have thrown such light on the subject that the arts need 
not suffer from ignorance in their application, and analysis should 
be held necessary, and perhaps actual trial, to determine the val- 
ue of a supposed water limestone. Dumas (Chimie Appliqué 
aux Arts) gives the following general results concerning limes, as 
ascertained by Berthier, Vicat, and John. 
1, That limestones almost pure always produce a lime that like 
clay, forms a paste with water; and 2. Limestones of complex 
composition, but containing no argile, produce a lime that does not 
form a paste with water, and is not hydranlic. _When a lime- 
stone of the first class contains 90 per cent. of pure lime, 10 per 
cent. of magnesia, silica, alumina or oxide of iron together, hard- 
ly alter its properties. When the magnesia rises to 20 or 25 per 
cent., the lime acts with water like class No. 2, and is not partic- 
ularly hydraulic. Vicat asserts, that when it rises to 30 or 40 per 
cent. it communicates the hydraulic property. The hydraulic 
property has at one time been attributed to the ox. iron, at an- 
other to the ox. of manganese, and to each of the fomeige earthy 
ingredients. Berthier finds by analysis of hyd. limes that one 
with 9 to 10 per cent. of argile is moderately hydraulic, and emi- 
nently with 20 to 30 per cent., but he thinks equal parts of alu- 
mina and silica are best. Finally: 1. Alumina alone is no better 
than magnesia alone. 2. Silica is the principal element, as no 
compound without this is hydraulic. 3. Ox. iron and manganese 
are usually without influence. 4. The union of silica, lime, and 
magnesia, or alumina, forms the best hydraulic limes. — 
The analysis of several hydraulic limestones by Dr. Beck, 
gives from 12 to 18 per cent. of magnesia. The profitable man- 
ufacture of magnesian salts is suggested from the serpentine and 
magnesian minerals of Staten Island, of some of the river coun- 
ties above New York, and from the deposits in St. Lawrence 
County. 
Native iron is recorded from two localities, viz. Burlington, 
Otsego County,'a specimen of which is in the museum of the 
Albany Institute, and Penn Yan, Yates County, a specimen of 
which contained a minute portion of carbon, but no nickel, or 
cobalt. 
Vol. xxx1x, No. 1—April-June, 1840. 13 
