47 



The Bee-eater burrows for itself a hole, lays white eggs, but the male 

 is not kuown to box up his wife during the period of incubation. He 

 seems to behave very like a Kingfisher. Therefore we may surely link 

 the Ujmpa with the Bucerotes, as has before been suggested by many 

 naturalists, and place them near the Kingfishers and the Bee-eaters. 

 What, then, becomes of the value of the plantar tendons ? The Ujiupce 

 have Passerine plantars, but a Galline episternal process. The Bucerotes 

 and Meropes have heteromorphous plantar tendons, but have also a 

 Galline episternal process. Mr. Seebohm will require a diagnosis for 

 these groups, and also for his Mimogypes, which I shall bring back 

 once more as a suborder of the Falconiformes. I give the characters 

 after his own tabular form : — 



MIMOGYPES. 



A. Hallux present and connected with ih.Q flexor perforansdigitorum. [This excludes 



all other Suborders except the Salcyones, Coracicp, and Bucerotes.'] 



B. Episternal process not perforated to receive the feet of the coracoids. [This 



excludes the above-named Suborders.] 



UPUP^. 



A. Deep plantar tendons free. [This excludes the Gallince, as well as the MeroiKS 



and the Bucerotes.] 



B. Episternal process perforated to receive the feet of the coracoids. [This excludes 



the Oscines and Herodiones, which possess character A.] 



MEROPES. 



A. Episternal process perforated to receive the feet of the coracoids. [This excludes 



the Mimogypes and all Suborders but the Upupce, Bucerotes, and GidluKe.] 



B. HaUux present and connected Avith the Jle.ror perforans digitorum. [This excludes 



the Gallince and the Upiq^ce.] 



C. Spinal feather-tract well defined on the neck, but forked on the upper back. 



[This excludes the Bucerotes.] 



BUCEROTES. 



A. Episternal process perforated to receive the feet of the coracoids. [This excludes 



the Mimoggjyes and all Suborders but the Gcdlina', the Meropes, and the 

 Upiipce.] 



B. HaUux present and connected with theJJexorperforans digitorum. [This excludes 



the Gcdlinoi and the Z^pupce.] 



C. Spinal feather-tract not defined on the neck. [This excludes the Meropes.] 



In order to make Mr. Seebohm^s arrangement clearer to my hearers, 

 I have drawn up the following map (Plate VIII.) of his scheme for 

 comparison with that of Professor Fiirbringer and of my own which 

 follows later. 



