Willis. — The Distribution of Species in New Zealand . 449 
This table bears out the prediction that was made, in the most com- 
plete manner, 1 showing a decrease in the range from 691 to 272 miles, 
followed by an increase again to 753. A coincidence is perhaps worth 
noting — that the species whose range actually begins at the North Cape 
(o miles) have also an average range of just 753 miles, the same as those 
which begin in Stewart Island. 
Passing on to deal with other cases of prediction and verification, there 
are many other things which, if my hypothesis is correct, we shall expect to 
find in the numerical arrangement of the ‘wides’ and endemics of New 
Zealand. In the first place, as New Zealand is separated from the nearest 
land area of important size by an immense stretch of water, it is evident 
that few-* wides * can have arrived there in recent times (geologically speak- 
ing), though no doubt an occasional one or two have done so, and in very 
recent times, with the advent of the Maoris and of the white man, others 
have come there. The vast flora of introduced weeds may be left out of 
account, for there is not the least evidence to show that they would have 
spread had not foreign conditions , or disturbance of the native conditions, 
been also introduced. We shall therefore expect to find that if the flora be 
divided into groups, as in the case of Ceylon, according to rarity (i. e. according 
to area occupied), the ‘ wides ’ are, as in Ceylon, much more common than 
the endemics, and that both groups are arranged in graduated series, the 
wides being most numerous at the top, the endemics at the bottom, of the 
classification, since few wides are now coming in, whilst endemics probably 
continue to evolve. 
This is exactly what we do find, as a glance at the table will show. 
Table VIII. 
Range in New Zealand? 
1. 1,001-1,080 miles 
2. 
881-1,000 
»» 
3 . 
761-880 
4 . 
641-760 
>> 
5 . 
521-640 
»> 
6. 
401-520 
>> 
7 . 
281-400 
» 
8. 
161-280 
>> 
9 . 
41-160 
» 
10. 
1-40 
>> 
W ides. 
Endemics. 
122 
52 
79 
60 
39 
59 
38 
61 
26 
79 
27 
105 3 
18 
97 
20 
93 
9 
128 3 
21 3 
168 3 
399 90 3 
Rarity in figures from i to io, cor- 
responding to this classification . 3*5 6.5 
Equal, in figures from 1 to 6, as in 
Ceylon, to 2*1 3*9 
(Mean rarity of whole flora by this table, 5-6.) 
1 This result also shows with individual families and genera. 
2 I have used ten divisions (equal, with the exception of the first and last) here to simplify 
calculation. Six as a number has no special virtue beyond local convenience in Ceylon, and to have 
used it in New Zealand would have made more difficult the classification of the species by their 
longitudinal range. And, further, it would be impossible to decide what divisions exactly corre- 
sponded to those used in Ceylon. 3 Explanation given below. 
