482 MR. A. H. G. DORAN ON THE MORPHOLOGY 



Between these extremes will be found : — first, the stapes with straight crura and a 

 moderate aperture, as in the Carnivora and Muridse, including the Cetacean form, 

 where the opening is filled more or less by the thickness of the crura ; then the stapes, 

 with thin straight crura and a very small aperture, as in Bradypus; and lastly the 

 variety seen in Midas, Didelphys, Tolypeutes, and Macropus, where the crura are fused 

 for a considerable way from the head, diverging near the base, the separation beginning 

 much earlier in the first two than in the last two genera just named, so that the former 

 possess a subvariety of higher type ; from this last we find examples, in several Orders, 

 of insensible gradations towards the still higher straight-limbed type, with a free aper- 

 ture, as in the Carnivora. 



In Didelphys opossum all the ossicula are of very similar character ; the lamina of the 

 malleus is even wider ; a'nd there is a little tubercle on the inner side of the neck, 

 representing the processus muscularis, very feebly indicated in D. virginiana. In both 

 species the processus gracilis is held in a chink between two forks of the tympanic ring 

 during lifetime, as in most of the Marsupialia. 



In the remaining families of marsupials the divergence of the crura is either completely 

 absent in the stapes or exceptional and inconstant. 



In the ossicula of the Tasmanian Wolf (Thylacinus cynocephalus) (PI. LXIV. fig. 17), 

 so like a Canis in its external appearance, there is hardly one point to be found in which 

 they resemble the ear-bones of a dog. 



The head is "very ill-developed, though not much flattened laterally, and distinctly 

 convex upwards. The articular surface is deeply cut, with prominent margins ; the lower 

 facet is very narrow. The neck is very thin, long, and narrow, forming a very prominent 

 humplike bend midway between the head and the handle. The lamina is very broad and 

 prolonged upon the processus gracilis, to form with it an extremely long process, which 

 is very narrow, except near the club-shaped, curved, and flattened extremity. The pro- 

 cessus muscularis is indicated by a very faint elevation on the inner side of the neck. 

 The manubrium is slender, and very similar in character to that of Macropus *. 



Altogether, even in the character of the lamina, the malleus of the Tasmanian Wolf 

 more resembles that of certain Insectivora' than that of any Carnivorous mammal. 



The incus has an ill-developed body with a very short processus brevis ; the stapedial 

 crus is long and very divergent ; it bears a small elliptical Sylvian apophysis on a rather 

 long pedicle. 



The stapes (figs. 17 & 43) is absolutely columelliform. The head is very small, and 

 the column long, and rather broader near the base ; there is not a vestige of an aperture. 

 The base is wide horizontally, but shallow vertically, with the upper border well curved, 

 contrasting with the almost circular base of the Ornithodelphia and Sauropsida, and 

 constituting a higher type of the columelliform stapes, in that the footplate assumes the 

 same form as is observed in the higher mammals, where that ossicle is bicrurate. 



* The front and lower part of the head ends in a point, and is not prolonged on to the lamina joining the pro- 

 cessus gracilis in the Carnivora &c. With regard to the manubrium in Dasyurus, consult my observations on Myr- 

 mecobins further on. 



