168 
their separation from each other cannot be ef 
fected at all, or but with difficulty. In such 
cases, their joint weight, or that of the combina- 
tion in which they are separated, is ascertained, 
and then different methods resorted to for deter- 
mining their relative quantity, as, for instance, 
from their different physical properties. This is 
often resorted to in practical life, where, in mix- 
tures of only certain substances, the value de- 
pends on their relative quantity. The specific 
gravity of such mixtures affords, in many cases, 
a mean of determining their relative quantity, 
which is often made use of, particularly where 
one of the substances is water. Alcoholometry, 
or the method of determining the quantity of 
alcohol, in mixtures of alcohol and water, depends 
entirely on the difference of their specific gravity. 
All the various uses'of hydrometers, for ascertain- 
ing the strength of substances, depend on the 
same principle. As soon, therefore, as other sub- 
| stances are added, besides those for which it was 
constructed, it ceases to be applicable to this pur- 
pose. The richness of many ores may be deter- 
mined approximately by their specific gravity. 
Besides specific gravity, other physical properties 
may be employed, as different fusibilities; thus, 
the relative quantity of two different fatty mat- 
ters has been determined from the point of fusion 
of their mixture. Other methods are derived 
from their relation to light or heat; thus, in 
mixtures of chloride of sodium and potassium, 
_ their relative quantity has been ascertained by 
the degree of cold which they produce by dis- 
solving them in water. 
In cases where two such substances have strong 
chemical, but similar affinities, and, therefore, 
are difficult to separate, but their combining 
weights are different, another method is fre- 
quently made use of in analytical chemistry, for 
which the rather improper name of éndirect ana- 
lysis has been proposed by Poggendorff. This 
method consists in separating them both toge- 
ther, by combining them with a third substance, 
and ascertaining the weight of this combination 
accurately ; we then determine, in any other ana- 
lytical way, the quantity of this third substance, 
which, subtracted from the former weight of their 
combination with it, gives the quantity of the 
mixture of them both; calculating, then, from 
their respective combining weights, the quantity 
which this third substance would have yielded 
with either of them alone, we are able to calcu- 
late, from the quantity actually obtained, the 
relative quantity in which they exist. 
In some cases, where we are unable to separate 
an ingredient in any state or combination suit- 
able for the estimation of its quantity, this can 
only be obtained by estimating the weight, suc- 
cessively, of all the other ingredients, and sub- 
tracting the joint weight of these from the weight 
of the whole substance; but it is evident that, 
in this case, any inaccuracy in the estimation of 
the other ingredients will fall upon it. 
ANALYSIS. 
The preparation of a substance for analysis, 
and the subsequent separation of the ingredients, 
either for their recognition, or, more particularly, 
for their quantitative estimation, requires many 
mechanical operations and manipulations, such 
as pulverization, solution, precipitation, pouring, 
washing, filtration, ignition, weighing, &c., the 
perfect performance of which is of the greatest 
importance to the success of the analysis, but is 
often very tedious, and can only be learned by 
practice in a laboratory; yet a thorough know- 
ledge of the principles on which they depend, 
will highly facilitate their acquisition. They are 
the rudiments of analytical chemistry, on which 
its performance depends; but as they often seem 
of a trifling nature, it requires some exertion of 
mind to submit to the drudgery of acquiring 
them. As analytical chemistry is the base of all 
chemical investigation, the want of them forms 
a great obstacle to the pursuit of chemistry, and 
they should therefore be early attended to. The 
spilling of a drop in pouring, or the spirting dur- 
ing evaporation, will often spoil long-anticipated 
results, and a single moment’s inattention be the 
cause of losing weeks of labour, at a point when 
the result was nearly within grasp. The inability 
or unwillingness to go through the laborious and 
tedious process of an analysis, or to repeat it ||_ 
when unsuccessful, is also a continual temptation 
to scientific mendacity, by anticipating and pro- 
nouncing analytical results which have never been 
obtained. But truth cannot be obtained without 
labour, and is inseparably connected with hon- 
esty ; and although a false ambition may thereby 
obtain, temporarily, a desired object, merited dis- 
credit usually returns, in the end, upon its origi- 
nator. Asarule, no results should ever be ad- 
mitted in science, unless accompanied by the 
details of the experiments from which they are 
derived. 
Notwithstanding the perfection of the methods 
and the utmost care, some small loss of the in- 
gredients in the different operations of quanti- 
tative analysis cannot be avoided. In well-per- 
formed analyses, of ordinary difficulty, this ought 
not to exceed one per cent., and in most cases be 
less. Beginners are apt to obtain more than the 
true weight, which arises from imperfect wash- 
ing of the precipitates, or from hygroscopic mois- 
ture, attracted during the weighing, &c. In 
analysis, for practical purposes, the loss or excess 
is often distributed proportionally on all the in- 
eredients, so as to make up exactly 100 parts; 
but in all scientific investigations, it is an estab- 
lished principle to give the results exactly as they 
are obtained. 
As all chemical compounds are formed by the 
combination of their ingredients, in certain and 
fixed proportions, which are multiples of their 
atomic weights, it becomes a strong proof of the 
correctness of our analysis of a pure chemical 
compound, if the percentic composition corre- 
spond with multiples of the atomic weights; and 
