244 Barbets, and tlieir Distribution. 



from branch to branch. Their flight is short, but rapid ; their 

 course consisting of a series of numerous undulations. I 

 never saw any of the species of this group on the ground. I 

 am not acquainted with the mode of propagation of these 

 birds, except that Trachyphonus margaritatus builds in holes of 

 trees, and lays white eggs, usually from four to six in number. 

 In the months of October and November, I have often seen 

 half-fledged young ones of this species clustering together, in 

 the peculiar way that may be observed in some of the Euro- 

 pean genera [Pants, for instance), and sitting on the smooth 

 side of the small branches, chirping as they await their 

 parents. With raw flesh and hard and soft-boiled eggs, I have 

 kept some of them a long time in confinement/'' 



" The Gapilonidce of North-eastern Africa are not exactly 

 migratory, though they appear at the time Avhen the Sycamores 

 (Ficus sycamorvs) are ripe in countries where they are not 

 generally met with." 



On crossing over the Atlantic to tropical America we again 

 meet with Barbets of organization and habits nearly similar to 

 those of the Old World, but neither so widely distributed nor 

 so abundant in species as their brethren of Africa and Asia. 

 Only two forms of American Barbets are yet known to science, 

 namely the genera Gapito and Tetragonops, the former with 

 fourteen and the latter with two species.* The latter is the 

 most aberrant form of the whole family — that is, that which 

 departs farthest from the general character of the group — and 

 shows manypoints of resemblance tothe Hill-toucans (Andigcnce). 

 There can be no question, indeed, that the Barbets are closely 

 allied to the Toucans (Ehamphastidce) , and should be placed next 

 to them in the natural series ; and to those who hold the 

 doctrine of the derivative origin of species, it is interesting to 

 note that the most Toucan-like Barbets inhabit the same region 

 as the most Barbet-like Toucans. 



It appears, therefore, that in the case of the Barbets, we 

 have an instance of members of the same natural family of birds 

 being met with in the tropics of both the Old and the New 

 World. And this is a very noteworthy fact, for it must be 

 recollected that, as a general rule, the avi-faunas of these two 

 regions — that is, the general series of the birds which inhabit 

 them respectively — arc perfectly distinct from one another. 

 Not only are the species — if we except some few wandering 

 forms of nearly universal distribution — invariably distinct from 

 one another, but in nearly every case these species are referable 

 to different genera, and, as a general rule, it may even bo 

 stated that the most characteristic birds of these two regions 



* Seethe Author's articles on the American Barbets, in the "Ibis,"18Gl, 

 p. 182 ; 1862, p. 1 ; and 186 I, p. 370. 



