to other Branches of Knowledge, 335 



The Australians constitute a third race in these regions. 

 Their dialects appear, from the researches of Captain Gray, 

 to fonn a particular family of languages, displaying great 

 varieties, but bearing traces of affinity among themselves, 

 and derived from a common origin. One of the most inter- 

 esting observations as yet made respecting them is the re- 

 mark of Mr Norris, who has discovered indications of con- 

 nection between the Australian dialects and the Tamulian, 

 spoken by the aboriginal inhabitants of the Dekhan. 



1 have endeavoured to explain what I believe to be the 

 principles on which ethnological research must be conduct- 

 ed, if we would maintain for that study and its results the 

 character of a really scientific and philosophical pursuit. 

 The only certainty that can be obtained in the formation of 

 groups and families of nations, must be founded mainly on 

 historical proofs. We must begin by establishing the histori- 

 cal fact of relationship or consanguinity between tribes of 

 people, before we venture to refer them to one race, or to 

 assert their diversity of origin. The deviations which are 

 known to have taken place within the limits of particular 

 families, are too great to allow us to assume diversity of 

 origin on the mere ground of physical difference ; and it is 

 equally obvious, that we cannot assume a near relationship 

 on the simple evidence of physical resemblance. 



It would carry me far beyond the limits of the subject of 

 this paper if I were to attempt to sum up any 'general re- 

 sults, or trace the bearing of facts on the great question of 

 the unity or diversity of human families ; but I may be al- 

 lowed to conclude with this remark, that the farther we ex- 

 plore the various paths of inquiry which lie open to our re- 

 searches, the greater reason do we find for believing that no 

 insurmountable line of separation exists between the now 

 diversified races of men ; and the greater the probability, judg- 

 ing alone from such data as we possess, that all mankind are 

 descended from one family. 



