RIFLEMAN. 465 



group. This parallellism, however, is not more remarkable or inexplicable than the 

 well-known isomorphism of some marsupials and placental mammals, and 'tenui- 

 rostral,' or ' conirostral ' Clamatores are not stranger than marsupial ' Carnivora ' or 

 ' Rodentia.' 



It was not until 1882 that the XENICID^E were shown to be mesomyodic, and, 

 consequently, were taken from the Oscines, having formerly been placed alternately 

 in the families Certhiidas, Sittiche, Troglodytidae, and Sylviidae. In support of their 

 present position amongst the clamatorial birds, I quote the following summary of their 

 characters as given by the late Professor Forbes, the discoverer of their true relation- 

 ship : - 



" Xenicus and Acanthi sitta are true mesomyodian forms, and therefore in no 

 intimate degree related to such Oscines as Sitta, Kitella, or other ' Certhiidae,' since 

 the syrinx has more of the complex nature of that organ in the Oscines, and there is 

 no other intrinsic syringeal muscle than the thin lateral tracheal, the position of 

 which is that characteristic of all the mesomyodian Passeres, resembling somewhat that 

 of Todus (haplophone). Externally the non-oscinine structure of their wings, which 

 have a 'first' (tenth) primary nearly as long as the following (preceding) one, and of 

 the tarsus, which is non-bilaminate, is at once apparent. Ten rectrices. No trace 

 of plantar vinculum. Sternum has a single pair of posterior notches. Holorhinal. 

 Vomer broad and deeply emarginated anteriorly, the maxillo-palatines slender and 

 recurved. Perhaps nearest to Pittidaa, though differing in many respects." 



The present family is restricted to New Zealand, the genus Acanthi sitta, consisting 

 of one or possibly two species, showing some resemblance to the creepers, except for 

 the extremely short tail, while the two species composing the genus Xenicus most 

 remarkably personate our small short-tailed wrens, but their colors show considerable 

 admixture of green. X. longipes is strictly arboreal in its habits, while X. gilviven- 

 tris, according to Dr. Haast, is found in the high Southern Alps of New Zealand, ex- 

 clusively amongst the large taluses of debris high on the mountain sides. Instead of 

 flying away when frightened, or when stones are thrown at it, or even when shot at, 

 it hides itself among the angular debris. Acanthisitta chloris, known as the ' rifle- 

 man,' especially in its habits resembles the creeper, for it is said to be generally 

 seen running up the boles of the larger trees, often ascending spirally, prying into 

 every chink or crevice, and moving about with such celerity that it is rather difficult 

 for the collector to obtain a shot. 



From New Zealand to Madagascar is not so long a step, ornithologically speaking. 

 The PHILEPITTID^E form one of the most interesting forms of the latter island. They 

 have been knocked considerably about in the system, and Mr. R. B. Sharpe even pro- 

 posed to make them a sub-family under the birds-of-Paradise, while A. Milne-Edwards 

 placed them next to the Nectariniida3, but Forbes showed, a few years ago, that they 

 are mesomyodian with broncho-tracheal syrinx, though differing enough in detail to 

 justify their position as a separate family. The most remarkable feature in regard to 

 the intrinsic muscles is that at the lower insertion they are peculiarly expanded, 

 though not fixed to the ends of the bronchial semi-rings, which are peculiarly modified. 

 Of external characters we shall only call attention to the taxaspidean tarsi, the rather 

 long wing-coverts, and the short tail, a peculiarity which the Pfiilepittce have in com- 

 mon with all the Old World forms of the present super-family. The male has a large 

 lobed and naked caruncle above the eyes, and both sexes possess a peculiar penicil- 

 lated tongue. Only two species of the genus Philepitta are known P. castanea and 

 VOL. iv. 30 



