152 Mr. D. G. Elliot on the Trocl.ilidie. 



almost the entire portion of South America. The habitats of 

 the A'arious species appear tolerably well defined, and do not, 

 as a rule, encroach upon each other. 



Before commencing the review of the species separately, I 

 will here say a word about the colour of the plumage. In all 

 birds whose feathers are metallic, there is more or less varia- 

 tion to be observed in the colours they reflect ; and while, 

 happily, in the majority of cases the hues are generally the 

 same, so as to enable naturalists to determine and recognize 

 a species, yet it not infrequently happens that specimens of 

 the same species exhibit different shades of the same colour, 

 or else, as in the case of the green Humming-birds, exhibit 

 a tendency to clothe themselves in feathers of a more or less 

 red or coppery tint in the places that should properly show 

 only a purely brilliant green. Amongst no class of feathered 

 creatures, perhaps, is this change more observable than among 

 the Trochilidse, and in no group of that great family more 

 than in the members of the genus Chlorostilbon. Sometimes 

 the plumage reflects, in place of green, golden hues of different 

 degrees of intensity ; and it is not uncommon to find some in- 

 dividuals of a species possessing brilliant metallic crowns, 

 while those of others are either dull green, or else show 

 only a very slight brilliancy. It is this changeableness of 

 colour in their plumage, together with their variableness in size, 

 both of the body and its members, that has given rise to so 

 much confusion in this group, and to the descriptions of spe- 

 cimens that have really no specific value, and which, when 

 seriously investigated, with ample materials, we are obliged to 

 eject from a position they improperly occupy, and consign 

 them to that ever-growing bete noire of all naturalists, the 

 synonymical dust-bin. Alas that there is no legalized 

 dustman to cart it away and all its contents ! Having ex- 

 amined, with one or two exceptions, all the knoAvn species 

 of this genus, I have given at the head of this paper an 

 analytical table, by which, in spite of the difficulties just 

 mentioned, all the species, here acknowledged as such, can be 

 recognized without any great difficulty ; but if variations in 

 size or tint of plumage are to be considered sufficient in them- 



