338 Willis. — Further Evidence for Age and Area; 
easterly or south-westerly direction, i. e. at right angles to the general 
direction of the group. 
Taking for each member of the flora the number of islands upon which 
it has been found, one obtains the following result : 
Table II. 
No. of Islands. 
Wides. 
Endemics. 
All 
97 (74) 
4 i 
6 
— 
8 
5 
1 
11 
4 
5(4) 
55 
3 
7 
80 
2 
17 
113 
I 
22 
273 
Total 
149 (125) 
581 
Rarity (in figures from i to 7) 
2 “7 (3-o) 
5* 6 
In great probability, however (see 5, Introduction, p. xvi), some of 
these wides, viz. twenty-three in those reaching all islands, and one reaching 
four, were introduced in prehistoric times, and as of the wides only reaching 
one island or two, most reach Oahu or Hawaii (the largest and most 
important islands, the former having a great port of call— Honolulu), 
probably many of the last two classes were recently introduced. This 
reduces the list of wides to 125, and brings the rarity to 3*0, which, how- 
ever, makes no difference to the argument. 
From this table it is evident that on the whole the wides arrived in the 
group at an early period, and had already most of them spread over the 
whole of it before it broke up into the existing islands, which stand upon 
a submarine plateau. The way in which the wides reached the group does 
not matter to the present discussion, whether overland, or by other means. 
Only fifty-one of them arrived too late to spread over the whole of the 
islands before they became broken up, and of these there is good reason to 
suppose that many are comparatively very recent arrivals. 
A considerable number of endemics were also evolved in time to reach 
the outer boundaries of the present archipelago before it broke up. It must 
be remembered that upon my hypothesis all plants, given long enough time 
and absence of serious barriers, will ultimately cover the whole area (cf. 11, 
p. 20, Table XIII, columns from G onwards). In the Hawaiian islands the 
whole area is mountainous, so that plants of all elevations can probably 
spread with comparatively equal facility, though it would seem probable 
that as the suitable areas of high elevation are of necessity smaller, in most 
cases, the plants of high altitudes would most often tend to spread less 
rapidly than those of low. This, therefore, must be added to the list of 
causes that may modify the action of age and area, as given in a previous 
paper (13, p. 206). 
