ALIMENTARY TRACT OF CERTAIN BIRDS. 85 



short loops, fi-om which it follows that the two long loops inter- 

 communicate directly. These facts, so far, are correctly shown in 

 Dr. Mitchell's figure referred to. Conurus leucotis is precisely 

 similar to Ara ararauna. 



Melopsittacus undulatus is still further reduced. The proximal 

 complex is, as in the last species, reduced to two, but there is only 

 one of the two distal loops left and that is shortened. 



On the other hand, Eclectus pectoralis is more complex than the 

 forms already considered. Among this group of birds the liga- 

 mentum ileo-duodenale is r^ather short and the pancreas extends 

 to the very end of the duodenal loop, or in some cases nearly so. 

 The ileic loop is often considerably longer than the duodenal. It 

 seemed to me' to be' particularly long in Stringaps hdbroptilus, 

 where it measured fully 8 inches in length. 



Among the Pigeons I have examined one or two forms not 

 studied by Dr. Mitchell, This group contrasts, for instance, with 

 the Rallidfe in the variety of intestinal patterns which it exhibits. 

 This, it will be noticed, is in accord with variations in the group 

 in other anatomical char'acters *• I propose, however', to deal with 

 two genera in which the intestine has become shortened in relation 

 to the fruit-eating habit, leaving other forms ttside until I have 

 been able to make a moi^e comprehensive study. In one of these, 

 Ptilopus hellus, the reduction in length, coupled with great increase 

 in calibre, has not go'ne so far as in the genus Carpophdga, with 

 which I shall deal later. In Ftilopiis belhcs the duodenal loop is 

 quite well mai-ked, though short and rather wide' ; the pancreas 

 extends to its very end. The jejunal region which follo\vs is also 

 distinct from it ftnd fro'm the ensuing ileic loop. The jejunal 

 region consists of a single loop only, the' two limbs of which lie in 

 close apposition, but can be considerably separated without tearing 

 any membranes. This region of the gut is therefore, in consisting 

 of a single loop without further complications, precisely like that 

 of the Bustards and Cariamas. Tlie ileic loop is also well marked 

 and rather wide ; it is much longer than the duodenal. The ileo- 

 duodenal ligament is present, but not very e'xtensive. There is 

 also a ligament uniting the jejunal loop' to the entering limb of 

 the ileic loop for about halfway down the' latter. The plan of 

 the intestinal tract in this bird is therefore a i-athef primitive one, 

 reminding us of that of Otis and Gliunga and of the Ficopass6res. 

 In fact, the degeneration of the intestine has resulted in the throw- 

 back to a presumsibly earliei- state of afFairsv 



The genus Carpojuhaga shows a further degerieration of the 

 intestinal tract, which is well known to fee very short in 

 this genus of Pigeons f. The shortening is allso accompanied by 

 widening. 



I have examined CJ. ceuea and C. concinmcc- In both of these 

 the duodenal loop has vanished and is represented perhaps by the 



* Gan-od, " On some Points in the Anatomy of the Columhte," P'. Z. S. 1874, p. 249. 

 t Cf., c. (/., GarroU, "Notes oU tht Oizzaid &*, of Caryoj/hrt^a latrans," P. Z. iS. 

 1878, "p: lOk 



