88 PEOr. -W. B. BENHAM ON THE [Feb. 7, 



of gradation from those which had white bellies to those which 

 exhibited the characters of perfectly' typical Mus musailus. 



I tliiuk. then, that we may safely conclude that Mus musculus is 

 of at least several hundred years' standing at St. Kilda. 



There is one extremely interesting point which should not be 

 forgotten in connection with these tu o (St. Kilda Mice, nanielv the 

 fact that we have here a clear opportunity of studying the effect 

 on two distinct species of the same genus of isolation side by side 

 on the same island. Here we liave on a circumscribed area two 

 species in the course of evolution, the progress of which may be 

 easily studied from time to time. The species having now been 

 described, we may be able in 20 or 30 years' time, by comparing 

 specimens taken then and now, to estimate the amount of change 

 which they will in that time have undergone. It is interesting to 

 note, however, that so far the effect of isolation on the island is not 

 similar in the case of the two species, since apparently the Mouse 

 which must be supposed to have been the longer time at St. Ivilda 

 is the very one which has varied in a lesser degree than that which 

 we must regard as an introduction. For Mus hirtensis, which 

 appears to have been on St. Ivilda since that island was in con- 

 nection with the mainland, is certainly not much more different 

 from Mus sylvaticus than is Mus muralis from Mus musculus, yet 

 Mils muralis can only be an introduced species of at most a few 

 hundred years' standing. Kothing can give stronger emphasis to 

 the fact that diiferent species possess different powers of variability 

 and follow a different course of e\olution, so that it seems that we 

 cannot predict what will happen under certain circumstances to 

 one species from our experience of what has happened to another. 

 Every species, it would appear, has its own modes of evolution and 

 development, which are peculiar to it and to it alone. 



EXPLANATION OF PLATE IX. 



Fig. 1. 3fns hirtensis, p. 81. 

 Fig. 2. Mus muralis, p. 81. 



4. Notes on the Internal Anatomy of Notornis. By W. 

 Blaxland Benham, D.Sc., M.A,, Professor of Biologj', 

 University of Otago, Dunedin^ New Zealand. 

 [Received December 7, 1898.] 



Early in August of the present year, 1898, 1 had the opportunity 

 of examining the anatomy of that rare tiightless Rail, Notornis 

 mantelU, of which only three previous specimens had been obtained 

 during the last 50 years, so that it has been i-egarded by European 

 zoologists as probably extinct. Thus Gadow says, in Bronn's 

 ' Thierreich ' : " kiirzlich ausgestorbeu " (Systematic part, p. 1S2). 



The previous specimens did not reach the hands of naturalists ia 

 a condition fit for examination, but this fourth one arrived in a 



