2 PROCEEDIISrGS OF THE NATIOISTAL MUSEUM vol.77 



Beebe (2) seems to have been the first to record a centripetal type 

 of tail molt (that is, one starting with the outermost pair of rec- 

 trices and proceeding inward and ending with the middle pair, 

 in other words, just the opposite of the centrifugal type), and, 

 indeed, it is in his work that the two terms centripetal and centrif- 

 ugal were first applied to the present subject. Beebe found that 

 some pheasants started the caudal molt with the middle pair of 

 rectrices, while others began with the outermost pair. Later, in 

 another paper (3) he made a hasty survey of immediately available 

 material and found that a woodpecker {Geleus species) also had a 

 centripetal tail molt. All the true pheasants (subfamily Phasianinae 

 containing, according to Beebe, the genera Lophophorus, GhM- 

 cophasis, Acomiis, Lophura, DiardigaUus, LoMophasis, Grossop- 

 tilon, Germaeus^ Gatreus^ Pucrasia, Symiaticus, Galoplmsis, Phasi- 

 anus, Ghrysolophus, and G alius) have a centripetal tail molt, while 

 the Perdicinae (genera Perdix, Gotv/rnixy Gaccahis, FrancolinuSy Pter- 

 nisfes, etc.) have a centrifugal type. In the peafowl {Pavo) the 

 molt begins with the second from the outermost pair and 



* * * there follows a regular progression inward, the outer pair being 

 molte(l just before the inner ones. This sequence is invariable, both in the 

 10 pairs of rectrices of the cock and the 9 pairs of the peahen. 



In the argus pheasants and their allies (subfamily Argusianinae 

 containing the genera Polyplectron^ GhalcuruSy Argusianus^ and 

 Bheinardius) Beebe finds the molt to begin with the third from the 

 central pair and to proceed outward and inward, the second and 

 first pairs (inner) falling, respectively, between the fourth and the 

 fifth and the fifth and the sixth pairs. 



In his life history studies of the Panamanian toucan, Rhamphastos 

 hrevicari7iatus, Van Tyne (4) writes that — 



* * * in their method of tail molt toucans are nearly unique among 

 birds. Instead of molting the restrices in regular order, beginning with the 

 central pair and progressing outward, they exactly reverse this and molt the 

 tail from the outer toward the central feathers. Beebe * * * first de- 

 scribed this and called it the " centripetal type " of tail molt. He also recorded 

 this type * * * in a tropical woodpecker {Celeus) and in certain pheasants. 



I am not aware of its occurrence outside of these groups. 



The fact that a centripetal type of tail molt had been found in a 

 woodpecker and in a toucan suggested the thought that it might be 

 fairly widely distributed among coraciine and picarian birds. Con- 

 sequently, while studying the extensive series of species of these 

 two and related orders collected in Africa by the late Edgar A. 

 Mearns, I made a point of examing their molts in detail. Later I 

 made a rather hasty survey of Neotropical and Asiatic groups not 

 found in Africa to get a somewhat broader picture of the distribution 

 of the centripetal tail molt. 



