370 Analyses of Books. [May, 
Cuap. 1.—Of the Method of classifying Inorganic Bodies accord- 
ing to ther Chemical Nature. 
As the object of chemical research is to discover the nature 
and properties of bodies, it is of great. importance to present 
them in the most methodical order. Hitherto bodies have been 
classed according to certain characters, and attempts have been 
made to define the classes thus formed. By this manner of 
proceeding, a certain degree of order has been introduced into 
chemistry, but by no means suited to the present state of the 
science. In fact, these definitions, according to which it has 
been attempted to class every thing, were established at a time 
when the science had. made but little progress, and when the 
object was merely to arrange bodies in certain isolated groupes. 
While matters continued in this state, it was easy to make the 
definition agree with the things, and the limits of the definitions: 
might be in some measure regarded as limits assigned by nature 
herself. But when new discoveries filled up the gaps which 
' existed between these groupes, these definitions required to be 
modified or extended ; and after all they could not be made to 
apply correctly. Some were abandoned altogether, and others 
were retained, which, however, are not more capable than the’ 
rest of withstanding a rigid examination. 
When Boerhaave and Stahl began their chemical career, only 
six perfect metals were known, and the remaining metallic bodies 
were excluded from the class on account of their brittleness. 
This distinction continued till the discovery of a variety of new’ 
metals, and filled'up the great gap between the ductility of gold 
and the brittleness of arsenic. It was then presumable that the 
intervals still remaining would be gradually filled up by new sub- 
stances, and consequently that the ductile and brittle metals 
could no longer constitute two separate classes. A great degree: 
of volatility was also considered as’sufficient to exelude various 
bodies from the class of metals. But at present when we know 
that even gold itself may be volatilized by electricity, by powerful 
burning glasses, and by the heat excited by means of oxygen 
gas, we cannot consider want of volatility as essential to the 
metallic nature of bodies. The temperature at which gold is 
volatilized is some hundred thermic metres above that at which 
arsenic or mercury is converted into'vapour. By thermic metre, 
Prof. GErsted understands the thermometric space comprehended 
between freezing and boiling water. This space he considers as 
unity, and thus uses a language that applies equally to all the 
different thermometrical scales which are employed in different 
countries. 
_Ifthe mean temperature of the earth were five thermic metres: 
higher. than it is, arsenic and’ mercury would be always in the 
state of vapour; and yet the ratio between the vapour point of: 
