0198566123.pdf

VARIATION IN INSULAR ENDEMISM BETWEEN TAXA 241

monocots are endemic, and 205 of the 351 dicots are
endemic (Porter 1979, 1984). The low proportion of
endemic ferns may be explained as a function of
their better dispersability. This is indicated by the
data in Fig. 9.10 which illustrate not only the tiny
size of fern spores, but that seed sizes of grasses
and sedges (monocots) are intermediate in size
compared to the classic island endemics in the
family Asteraceae. The proportion of endemism in
the Hawaiian flora, is considerably higher than for
the Galápagos flora, but again is greater in angio-
sperms (91% endemic) than in ferns and their allies
(70% endemism) (Table 9.2).
Dispersal traits thus provide one key element of
the evolutionary syndromes characterizing remote
island biotas. This is again illustrated by the
Hawaiian flora, for which Carlquist (1974) has
attempted to determine the most likely means of
colonization of each of the ancestral colonists of the
current native flora (Table 9.8) (see also Keast and
Miller 1996). Species dispersed by terrestrial
mammals, for example, are not part of the game.
Porter (1984) has calculated that the Galápagos
flora could be accounted for by 413 natural
colonists. On this basis and assuming the age of the
archipelago to lie between 3.3 and 5 million years
(Simkin 1984), the present native vascular plant
flora of the islands could be accounted for by one
successful introduction every 7990–12 107 years.
The error margins on such estimates may be large,

but they serve to indicate that successful dispersal

events to such a remote location are very infre-

quent, judged on ecological timescales. Studies of

less remote islands (e.g. Krakatau) find that the

immigration rates vary through time, declining as

the most dispersive species colonize and the rich-

ness of the flora builds up. It is hard to know if the

same variation in rates occurs in very remote situ-

ations such as the Galápagos, but it seems likely

(Box 9.3). In any event, the above data indicate that

remote islands receive new colonists sufficiently

infrequently as to provide long periods for those

species that do colonize an archipelago early on to

100

80

60

40

20

0

0.01 0.1 110

Diaspore magnitude (mm)

Percent endemics

Ferns and allies

Grasses and sedges

Composites

Mascarenes

Galapagos

Key

Figure 9.10Percentage of endemism for

selected plant taxa of the Galápagos and the

Mascarenes, against diaspore magnitude (the

rough length of the dispersal element). (Adapted

from Adsersen 1995, Fig. 2.2.)

Table 9.8Most probable means by which the original flowering

plant colonists dispersed to Hawaii (data from Carlquist 1974)

Dispersal mode Percentage of colonists

Birds

Mechanically attached 12.8

Eaten and carried internally 38.9

Embedded in mud on feet 12.8

Attached to feathers by viscid 10.3

substance

Oceanic drift

Frequent (able to float for prolonged 14.3

periods)

Rare (most likely to have arrived) 8.5

by rafting

Air flotation 1.4