226 DR. p. CHALMERS MITCHELL ON THE 



(Beddard, 1908, text-fig. 116); the term sinistra belongs to a 

 more distal region of the gut, and Tullberg's definitions, desci'ip- 

 tions, and figures make this point quite clear. These ansoi 

 dextrce may be definite and narrow, spirally twisted separately or 

 together. The three loops C.L. 1, 2, 3 in the diagram of Dipus 

 (text-fig. 21) represent a paractecal loop and two ansca dextrce. 

 Distad of the highest point of the recuri'ent limb, and to the 

 left of the equivalent of the transverse colon, there may be 

 another region of expansion. When this subsidiary loop is 

 simple and narrow, Tullberg recognises it as distinct and calls it 

 the ansa sinistra ; when it is thrown into irregular minor folds, 

 he leaves it undesignated. As I have already pointed out, I 

 have found both conditions of this expansion in different examples 

 of the same species, and therefore do not attach much importance 

 to it. But, definite or irregular, if it be named at all, ansa 

 sinistra is the correct name. It is absent in Dipus. 



The gut-pattern of Rodents, then, displays usually a separate 

 duodenum, a well-defined Meckel's tract, a cfecum frequently 

 spirally twisted, and an elongated hind-gut, variable in the 

 number and nature of the subsidiary loops which mav be 

 developed. 



Order Insectivora. Macroscelides species ? (text-fig. 22). 

 Talpa europcea (text-fig. 23). 



Taking the examples of animals grouped together as Insectivora 



that I had been able to examine when I wrote before (Mitchell, 



1905) and those that I have seen since, I cannot make up a 



series approaching completeness. Putting together my own 



observations with what I am able to gather from other writers, 



I think that three types of difierent degrees of simplicity 



can be distinguished among the gut-patterns of Insectivora. 



In Macroscelides (text-fig. 22) the duodenal legion cannot be 



recognised as separate from Meckel's tract. The latter is 



thrown into rather simple short loops round the whole of the 



descending limb and a small portion of the recurrent limb of 



the pendant loop ; then follows a long, nearly straight portion 



running up towards the dorsal line. The ctecum is enormous, 



and is attached high up on the recurrent limb of the pendant 



loop. The distal portion of the pendant loop is expanded to 



form a very large nearly closed colic loop, thrown into a number 



of minor loops. In the undisturbed condition this lies folded 



against the mesentery of Meckel's tract, but I found no secondary 



connection. The recurrent loop then bends round to form the 



straight rectum. The superficial resemblance between this 



pattern and that presented by some of the smaller Dipiotodont 



Marsupials, such as Fhalangista i^idjnna (Mitchell, 1905, lig. 5) 



is extremely close. When I had finished the drawing I thought 



that it had a familiar look, and on hunting through my notes, 



I found that, from the point of view of this memoir, it would 



