330 Willis . — The Distribution of the Plants of 
regarded from the ‘ age and area ’ point of view — it is almost always found to 
be one about whose identification or true nativity there is a doubt. This 
was well shown in the case of the exceptional number of 31 species found in 
the last class of the flora in place of the expected 5 or 6 (1. c., p. 452). 
In the next column, the plants reaching the Chathams only, there is 
a conspicuous exception in class 9, Pomaderris ape t ala , which is of such 
interest that I have devoted to it the last paragraph of this paper. The 
16 species in class 10 in the last column have already been dealt with in the 
preceding paper, as mentioned above. Nos. 2, 3, 4, 9, and 14 of that list 
range to the islands and are omitted here. 
Now if we omit these exceptional cases, the order of rarity comes 
exactly into line with my hypothesis, the lowest being for the plants of two 
or three islands and New Zealand, the next for the plants of one group and 
New Zealand, and the highest for New Zealand only. But in any case, the 
plants which go to the islands are far commoner (more widespread) in New 
Zealand than the plants which do not. Now there is no conceivable reason 
why ranging also to a few little islands should make a species more wide- 
spread in New Zealand, unless it be age, which has given them time to 
spread in New Zealand to the maximum degree. The reverse hypothesis, 
that dispersal goes with youth, will be rather hardly pressed to explain why 
youth should ensure that a species should reach more islands. Nor, to go 
back to yet older views, is there any reason why great dispersal in New 
Zealand should ensure reaching the islands. If it were so, why were only 
45 wides of class 1 selected, and the other 35 left behind, and why 19 of 
class 2 (instead of any of these 35), leaving 39 behind ? It is evident that 
those which reached the islands were on the whole the first comers to New 
Zealand. The intermediate position of the species which range only to one 
group of islands renders the older explanations impossible. 
Further consideration of what has been said brings out a very important 
point which may easily be lost sight of. Twenty species ranged the entire 
length of New Zealand, and got to two or more island groups ; 25 to one 
island group ; and 35 did not reach the islands at all. It is therefore 
evident that many of these last 60 once ranged to greater or less distances 
across land which is now submerged. In other words, submergence may 
overtake spread , and greatly reduce, even to extinction, the area occupied 
by a species. New species, therefore, have probably the best chance of 
survival and wide dispersal if they arise in the middle of a large continental 
area, and those that have the best chance of long survival are those which 
have been fortunate enough to disperse into areas so large that the chance 
of extermination by submergence or other catastrophe is least. 
Other points of interest come out from an examination of Table I. 
The number of wides which reach two or more islands is 25, reaching one 
group only is 53, and not reaching any is 213. This indicates age as the 
