CH. VIII] INVASIONS 81 



the minimum at the same. The more rapid fall of the endemic 

 curve is to be attributed (on the hypothesis of Age and Area) 

 to the fact that they are in general younger, and so have not 

 had time to sj^read so far. 



Treating the southern invasion in the same way, one obtains 

 the curves on p. 80, showing both endemics and Avides falling 

 off towards the north. The latter are shown with a double curve; 

 the upper shows the grand total of wides, but many begin at 

 the north and do not occur in the far south, showing that they 

 probably really belong to the northern invasion. Subtracting 

 these gives the lower curve, and the diminishing distance be- 

 tween these two curves shows the way in which these species 

 diminish southwards. The endemics, being more numerous, are 

 split into two curves, one endemic to New Zealand only, one 

 endemic to New Zealand and the outlying islands (Kermadecs, 

 Chathams, Aucklands). 



These curves provide a very formidable argument against the 

 supposition that endemics are djang out, for if so, why does 

 their number show its maximum with that of the wides, and 

 fall off to a minimum at the same point with the latter? 



They also illustrate various other points. For example, from 

 the much steeper curves of the southern invasion, one may 

 probably infer that it was much younger than the northern, 

 both wides and endemics having had less time to spread widely 

 in New Zealand. This is confirmed by the fact that both northern 

 curves, and that for southern wides, show no break of any kind 

 between 500 and 600, where Cook's Strait lies, while that for 

 southern endemics shoAvs a marked drop there, indicating that 

 when this group (the youngest of all, by hypothesis) came along, 

 the strait was at any rate beginning to be formed. The same 

 feature shows in a much more marked way at Foveaux Strait, 

 between the last two figures in the curves; even the northern 

 "wides" show a drop here, and the southern endemics an enor- 

 mous one. 



The greater age of the northern invasion may also be inferred 

 from the fact that in it the mmiber of the endemics at any zone 

 is always at least twice as great ns that of the wides, while in 

 the southern invasion the ciirve for endemics goes below that 

 for wides at both ends, or adding the endemics of islands, below 

 that for Avides at the northern end. 



The greater youth of the southern invasion is also emphasised 

 by the fact that it is composed to the extent of 83 per cent, of 



