1889.J CONVOLUTIONS I\ BTRDS. 30" 



enumeration and description of the intestinal convolutions as they 

 occur in the numerous orders and families of birds, because this will 

 be done elsewhere. 



The Table (pp. 308, 309) contains, in a condensed form, an account 

 of the principal modifications of the intestinal folds, and the diagram 

 (Plate XXXI 1.) shows the Rffinities, or, to speak more cautiously, the 

 convergent similarities, of all the principal families, as they are 

 suggested merely by a study of their intestinal arrangements. " The 

 birds will he discussed only from this point of view in order to test, 

 and to draw attention to, the taxonomic value of those characters 

 which are exhibited by the modes in which the mid-gut isstowedaway 

 in the abdominal cavity. 



Many of tnese similarities are perhaps merely coincidences, and 

 in this case can have no taxonomic significance ; but if these simi- 

 larities coincide with those of several other organic characters, they 

 are entitled to a higher rank as indicating not convergence but 

 common descent of those birds in which they persistently occur. 



There seems to be a sort of belief prevailing that the intestinal 

 convolutions are very variable and unreliable in the same species, 

 that they are a matter of accident ; Ijut, on the contrary, I have found 

 them constant to an astonishing extent, not only in the same species 

 but in many large families. Of course secondary shortening and 

 widening of the gut (owing to the assumption of ("rugivorous habits) 

 may reduce the number of loops, and may render the original 

 arrangement quite untraceable, as in, c. g., C'arpophaga, Rhamphastus, 

 Inanucodia. When a bird has acquired strictly piscivorous habits, 

 the gut is considerably lengthened and narrowed, and mav, just as in 

 Fundion and in Haliaetus, render the old formation quite unrecog- 

 nizable. These are, however, exceptions, which are not numerous ; 

 as a rule the lengthening of the pre-existing loops and the additional 

 intercalation of new ones does not disturb the typical formation, but 

 rather throws interesting lights upon the lines of new departure along 

 which certain birds have become developed, e. g., the Alcedinidse 

 from a Coraciine stock, now modified through the acquisition of 

 carnivorous and piscivorous habits. 



In the following Table the order adopted is one of mere con- 

 venience, without necessarily indicating near relationship. The 

 second column contains the number of principal loops ; this can 

 best be ascertained by spreading the intestine out on the table with- 

 out tearing the mesenteric connections. The next three columns 

 refer to the 2nd, 3rd, and -4th principal loops : r means that the loop 

 in question is a right-handed one, like the duodenum ; / that it is 

 retrograde, or left-handed ; o signifies that the loop is open ; cl that 

 it is closed. The last column indicates in a few words the type of 

 formation. 



The diagram (Plate XXXII.) requires some explanation. All 

 the birds of which the names are written inside the inner of the 

 two concentric circles are on the whole orthocoelous, whilst those 

 placed between the two concentric circles are cyclocoelous ; of the 

 latt'T, the underlined families are telogyrous, the others mesoi'vrous. 



