ALIMENTARY- TRACT OF CERTAIN BIRDS. 61 



I have not met with many divergences among the Picopasseres 

 from the typical structure. 



One of the most abnormal types — if not the most abnormal — 

 among the Picopasseres is the Ground Hornbill, Bucorvus ahys- 

 sinicus. The duodenal loop is longish and the pancreas extends 

 nearly to its end. It is perfectly free fi-om the ileic loop, which 

 is longer than it. The ileic loop, moreov^er, is indented at its free 

 extremity and tlius shows signs of being bent over upon itself. 

 It is also considerably longer than the duodensil loop. As in 

 other Picopasseres, the jejunal loop is more or less divided into 

 two, and the distal loop of these two is attached to the outgoing 

 limb of the ileic loop, which on its way to the colon is looped once 

 in a way precisely like that shown funong tlie Accipitres and in 

 some other birds. These facts are particularly intei-esting, because 

 they confirm current opinion as to the anatomical likenesses 

 between the Hornbills and the Hoopoe. It is plain from 

 Dr. Mitchell's figure * of the intestinal tract of that bird that 

 Upupa epops agrees with Bucorvus in a number of the characters 

 to which I have referred above. He figures the two loops of the 

 middle part of the intestine and the small '"supracpecal" loop, 

 which latter is so characteristic a feature of Bucorvus as compai'ed 

 with other Picarian birds. He does not, however, advert to this 

 loop by that name or compare it with the " kink" which he found 

 in the Accipitres of both the Old and New World. Nor does he 

 indicate a mesenteric attachment between the jejunal and ileic loops 

 in Ujnipa such as I find in Bucomus. It is impossible, moreover, 

 to be certain from Mitchell's figure how far the ileic and duodenal 

 loops are connected. Theii' entire mutual freedom in Bucorvtts is 

 an uncommon feature. Although Dr. Mitchell happens, as I think, 

 to be wrong in remarking that the cha,racter of the gut does not 

 unite the Hoopoes and Hornbills closelj^ he was perfectly right 

 in making that statement from the facts before him. This is 

 a further example of the difficulty of arriving at sound cla^si- 

 ficatory conclusions without an exhaustive knowledge of the 

 facts. 



I have lately had the opportunity of examining the gut of 

 Upupa, and can add something to the account given by Mitchell. 

 It is a rather move abnormal member of the Picopasserine group 

 than I had supposed. In my specimen there was no supracpecal 

 kink. The duodenal loop was very wide (as Mitchell has 

 remarked) and rather irregular in outline at its end, suggesting, 

 therefore, a commencing spiral as in irypagus and Cathartes 

 — a fact which may be of some significance. The duodenal 

 loop is larger tha.n the ileic — precisely the reverse condition 

 obtaining in Bucorvus. And while in Bucorvus there is no 

 ileo-duodenal ligament, there is a short one in Upupa not 

 nearly so extensive as in Picopasseres generally, and thus 

 bridging over the gap between Bucorvus and its allies. 



* Trans. Linn. Soc. i.e. p. 247, fig. 65. 



