412 NATURAL HISTORY OF BIRDS. 



very remarkable. Mr. Robert Swinhoe has described well the peculiar way in which 

 the hoopoe produces its notes by puffing out the sides of the neck and hammering on 

 the ground at the production of each note, thereby exhausting the air at the end of 

 the series of three, which make up its song. " Before it repeats its call," he continues, 

 " it repeats the puffing of the neck, with a slight gurgling noise. When it is able to 

 strike its bill the sound is the correct ' hoo-hoo-hoo ; ' but when perched on a rope 

 and only jerking out the song with nods of the head, the notes more resemble the syl- 

 lables < hoh-hoh-hoh.* Mr. Darwin makes use of this last fact to show that some 

 birds have instrumental means to produce their music." 



Like the true hoopoes, the wood-hoopoes, IRRISOKID^E, are accused of emitting a 

 powerful and offensive smell. They are inhabitants of the forests, where they climb 

 on the trunks of trees much in the manner of woodpeckers. They are described as 

 very noisy and restless birds. The wood-hoopoes are restricted to the tropical and 

 southern parts of the African Continent. The species here figured, Irrisor erythro- 

 rhynchos, has coral-red feet and bill. 



The three words, homalogonatous, antiopelmous, zygodactylous, at once and trench- 

 antly distinguish the PICOIDEJS from all the other Picarians ; and we cannot help 

 thinking that this super-family is an eminently natural one notwithstanding the many 

 yet unfilled gaps between the separate families composing it. These may be tabulated 



synoptically as follows : 



)Bucconid(K ; gonys rounded ; ventral 

 tract without gular 

 branch 

 Galbulidce ; gonys angular ; ventral 

 tract with an inner 

 gular branch. 



vomer truncate Ramphastidw 



Oil gland 



tufted ; 



no caeca; 



1 carotid 



not saurognathous ; manu- \ S 10 rectrices. 



brial rostrum pointed ) __ Wft _ ata j Megalaimida ) 



Indicated* 1 12 rec t r ice s . 

 saurognathous ; manubrial rostrum bifurcate Picidce 



vomer bifurcate 



It will be seen that the super-family is divisible in two groups : one with nude oil 

 gland and well developed caeca, the other with no caeca, but a tufted oil gland ; hence 

 Garrod referred the former to his order Passeriformes, the latter being his typical 

 Piciformes. 



He seems to have been in some doubt, however, concerning the BUCCONID^E, as to 

 whether they possess caeca or not, having had no specimen of this family for dissec- 

 tion ; but I find that Professor Burmeister has noted two long caeca in this family, thus 

 confirming Garrod's supposition. This family has been almost inextricably confused 

 by older authors with the Megalaimidae, from which they are distinguished by many 

 important characters besides those mentioned in the table above. One of these 

 deserves to be treated of a little more in detail since, though being an external one, it 

 is usually overlooked. It deserves our attention the more, as it is a character, the 

 development of which can be traced from the young to the adult plumage, thus afford- 

 ing us a means of telling which condition is generalized and which one specialized. 



Professor Sundevall was the first author to draw attention to the difference of the 

 upper wing-coverts in Passeres and some Picarians as compared with the rest of the 

 birds, showing not only that the large secondary coverts in the former are shorter, 

 not covering more than the basal half of the secondaries, while in the latter more 

 than the half is concealed by them, but also that in the former the small coverts are 



