8. Phytogeographic Relationships and Regional Richness Patterns
163
entire Pacific coast in southern South America was seen to be very
low and equability high. This last feature has favored the retention of
Cretaceous and Tertiary tropical and subtropical elements at temperate
latitudes in many parts of the world (Axelrod, 1992). Relic genera and
other ancient angiosperm elements owe their presence in the South
American rainforest today , without much doubt, to this equable climate.
What is remarkable about the rainforest, considering that it is a residue of
an initially much larger forest province, is the comparatively little intrusion
by more recent floristic elements. In this sense, it has shown remarkable
stability and resilience as a floristic assemblage.
The arguments outlined above have stressed the importance of historical
factors in the diversity of the southern temperate rainforests. It is more
appropriate to rephrase the above, with equable climate serving as the
baseline condition for the maintenance of the many ancient floristic elements that converged on southern South America. However, ecological
factors must be operating to maintain this diversity. The role of largescale disturbance in South American rainforest needs to be examined
critically. For example, has it increased or reduced diversity over evolutionary time by favoring certain taxonomic groups? A great variety of
temperature environments cutting east-west across the two major northsouth gradients for temperature and precipitation has probably also been
important, in that species have wide ranges of conditions to chose from
and have evolved in close contact with many other species.
Our demonstration of Rapoport's rule for a gradient with no evident
increase in annual temperature range was unexpected in the context of
Steven's (1989) hypothesis. What explanation then can be offered for
Rapoport's rule in our case? Rickles (cited as personal communication
in Stevens, 1992) suggests that Rapoport's rule might reflect how species
at different latitudes are distributed in relation to elevation. The species
occurring in high latitudes can undergo strong altitudinal displacement
northward along north-south trending ranges. In contrast , many more
northerly, low-elevation species cannot find equivalent environmental
conditions south-ward. As a consequence, in the northern part of a
latitudinal gradient, a combination of species with short and long ranges
would be found ; whereas at higher latitudes , only species with long
latitudinal ranges would be seen, leading to lower average range size at
the lower latitudes. A detailed analysis of altitudinal ranges might show
this to be the case in our gradient. Moreover, in our case, the more
northerly, low-elevation taxa should be constrained from expanding their
potential ranges northward because of aridity. It is worthwhile pointing
out that three tree species in cool temperate rainforest (Lomatia hirsuta,
Maytenus magellanica, and Rhaphithamnus spinosus) can also be found
at high elevations in the northern tropical Andes and in the southern
Brazilian uplands with total latitudinal ranges of 39 to 42°. A more
appropriate test for Steven's (1989) explanation of Rapoport's rule would
163
entire Pacific coast in southern South America was seen to be very
low and equability high. This last feature has favored the retention of
Cretaceous and Tertiary tropical and subtropical elements at temperate
latitudes in many parts of the world (Axelrod, 1992). Relic genera and
other ancient angiosperm elements owe their presence in the South
American rainforest today , without much doubt, to this equable climate.
What is remarkable about the rainforest, considering that it is a residue of
an initially much larger forest province, is the comparatively little intrusion
by more recent floristic elements. In this sense, it has shown remarkable
stability and resilience as a floristic assemblage.
The arguments outlined above have stressed the importance of historical
factors in the diversity of the southern temperate rainforests. It is more
appropriate to rephrase the above, with equable climate serving as the
baseline condition for the maintenance of the many ancient floristic elements that converged on southern South America. However, ecological
factors must be operating to maintain this diversity. The role of largescale disturbance in South American rainforest needs to be examined
critically. For example, has it increased or reduced diversity over evolutionary time by favoring certain taxonomic groups? A great variety of
temperature environments cutting east-west across the two major northsouth gradients for temperature and precipitation has probably also been
important, in that species have wide ranges of conditions to chose from
and have evolved in close contact with many other species.
Our demonstration of Rapoport's rule for a gradient with no evident
increase in annual temperature range was unexpected in the context of
Steven's (1989) hypothesis. What explanation then can be offered for
Rapoport's rule in our case? Rickles (cited as personal communication
in Stevens, 1992) suggests that Rapoport's rule might reflect how species
at different latitudes are distributed in relation to elevation. The species
occurring in high latitudes can undergo strong altitudinal displacement
northward along north-south trending ranges. In contrast , many more
northerly, low-elevation species cannot find equivalent environmental
conditions south-ward. As a consequence, in the northern part of a
latitudinal gradient, a combination of species with short and long ranges
would be found ; whereas at higher latitudes , only species with long
latitudinal ranges would be seen, leading to lower average range size at
the lower latitudes. A detailed analysis of altitudinal ranges might show
this to be the case in our gradient. Moreover, in our case, the more
northerly, low-elevation taxa should be constrained from expanding their
potential ranges northward because of aridity. It is worthwhile pointing
out that three tree species in cool temperate rainforest (Lomatia hirsuta,
Maytenus magellanica, and Rhaphithamnus spinosus) can also be found
at high elevations in the northern tropical Andes and in the southern
Brazilian uplands with total latitudinal ranges of 39 to 42°. A more
appropriate test for Steven's (1989) explanation of Rapoport's rule would