236 
THE DESCENT OF MAN. 
Part !• 
ing the varying offspring ; bnt most of our races have 
been formed, not intentionally from a selected paib 
but unconsciously by the preservation of many indi- 
viduals which have varied, however slightly, in some 
useful or desired manner. If in one country stronger 
and heavier horses, and in another country lighter and 
fleeter horses, were habitually preferred, we may feel 
sure that two distinct sub-breeds would, in the course 
of time, be produced, without any particular pairs 
or individuals having been separated and bred from 
in either country. Many races have been thus formed; 
and their manner of formation is closely analogous with 
that of natural species. We know, also, that the 
horses which have been brought to the Falkland 
Islands have become, during successive generations) 
smaller and weaker, whilst those which have run wild 
on the Pampas have acquired larger and coarser 
heads ; and such changes are manifestly due, not to 
any one pair, but to all the individuals having been 
subjected to the same conditions, aided, perhaps, by 
the principle of reversion. The new sub-breeds m 
none of these cases are descended from any single 
pair, but from many individuals which have varied in 
different degrees, but in the same general manner; 
and we may conclude that the races of man have been 
similarly produced, the modifications being either the 
direct result of exposure to different conditions, or the 
indirect result of some form of selection. But to this 
latter subject we shall presently return. 
On the Extinction of the Races of Man . — The partial 
and complete extinction of many races and sub-races 
of man are historically known events. Humboldt saw 
in South America a parrot which was the sole living 
creature that could speak the language of a lost tribe. 
