Figure 13.1 Diagrammatic reconstruction of the history of continental drift.
Chapter 13 FIG 1
Figure 13.2 Graphic depicting two very different approaches to understanding global patterns of species richness. The circle represents the globe, and
shades of color represent latitudinal zones with the latitude zero across the center. On the left, black points represent individual species, and clearly the
number of species is correlated with latitude; tropical habitats have more species than temperate habitats. Explanations for higher diversity in tropical
regions center on correlations between numbers of species and environmental variables such as temperature and moisture. On the right, evolutionary
relationships of species are shown (lines) with a hypothetical monophyletic clade represented. This graphic stresses a history of diversification indicating
that more clades originated in the tropics, and because of niche conservatism, few clades were able to evolve ecological traits allowing them to disperse to
temperate climates. Adapted from Wiens and Donoghue, 2004.
Chapter 13 FIG 2
Figure 13.3 Graphical model showing effects of niche conservatism and niche evolution on faunas in different regions of the world. Colored circles
represent different clades. Two clades (blue and green) have retained ancestral niche characteristics and have distributions restricted to tropical and
subtropical environments. A third clade (black) has evolved tolerance to lower temperatures and lower precipitation and dispersed, no longer occurring in
tropical regions and thus showing niche evolution. Adapted from Wiens and Donoghue, 2004.
Chapter 13 FIG 3
Figure 13.4 Temperature isotherms and the northern and southern limits of frog (30 species), salamander (26 species), and lizard (16 species)
distributions in eastern North America (Piedmont and Coastal Plain). Isotherms are mean annual temperature (°F); the integers in each zone are number
of species with ranges terminating in the interval (northern–southern termini). Isotherms from USDA, 1941; distributional data from Conant and Collins,
Chapter 13 FIG 4
Figure 13.5 Geographic variation in clutch size among populations of the snapping turtle (Chelydra serpentina) in North America. Integers indicate mean
clutch size at specific locality. Data from Fitch, 1985, and Iverson et al., 1997.
Chapter 13 FIG 5
Figure 13.6 Biogeographic realms of the world.
Chapter 13 FIG 6
Figure 13.7 Graphical model showing the difference between vicariance and vicariance with a single dispersal event in the construction of area
cladograms. At time 1 (T1) the ancestor lives on a single large continent. By time 2, continents have separated, creating the first vicariant event, and
additional speciation has occurred on the continent on the right. By time 3, four continents exist. On the left, no dispersal has taken place, and additional
speciation events have occurred on continents 1, 3, and 4. On the right, speciation has also occurred on continents 1, 3, and 4, but one of the species on
continent 3 has dispersed to continent 2 where, over time, that species has differentiated. Thus the species on continent 2 has one of the species from
continent 3 as its closest relative (sister taxon). Cladograms at the top show phylogenetic relationships between the species under each scenario.
Numbers across the top refer to present distribution of species on the four continents. By comparing a dated phylogeny with independently derived dates of
vicariant events (in this case, continental splitting), it is possible to falsify a vicariance hypothesis. The phylogeny on the left supports a vicariance
hypothesis, and the one on the right falsifies it for the species on continent 2, leaving dispersal as the only viable hypothesis for the origin of that species
on continent 2. Although our example uses continents, other barriers (mountains, rivers, ecological transitions) can result in vicariance. Adapted in part
from Futuyuma, 1998.
Chapter 13 FIG 7
Figure 13.8 Comparison of phylogenetic relationships of chelid turtles and their distributions to the tectonics of southern continents. Unfortunately, the
most recent cladogram (far right) fails to falsify either of the earlier hypotheses because Chelodina, South American chelids, and remaining Australian
chelids form an unresolved polytomy. Cladograms adapted from Gaffney, 1977; Seddon et al., 1997; and Georges et al., 1998.
Chapter 13 FIG 8
Figure 13.9 Although diversification of lizards in the Anolis nitens (formerly A. chrysolepis) complex has been used to support the Vanishing Refuge
Theory of Amazonian diversity, molecular analysis of the group clearly shows that diversification took place much earlier. Although the described
subspecies are each others' closest relatives, a deep split in haplotypes from the north (Roraima state in Brazil and Ecuador) and the south (Amazonas
and Acre States in Brazil) and relatively deep splits in more recent clades placed their origins before the existence of proposed refuges. The A. nitens
clade, at minimum, is > 15 million years old. Left, maximum parsimony bootstrap tree; right, maximum-likelihood tree. Adapted from Glor et al., 2001.
Chapter 13 FIG 9
Figure 13.10 Phylogenetic relationships of Amazonian sphaerodactyline geckos. Based on gene sequence data, major divergences occurred during the
Miocene–Pliocene, much before the Pleistocene, when Amazon refuges existed. Adapted from Gamble et al., 2008.
Chapter 13 FIG 10
Figure 13.11 The frog Physalaemus petersi is ideal for testing biogeographic hypotheses on origin of diversity in the Amazon River basin because it
occurs across an area divided by large rivers (Riverine Hypothesis) and where an elevational gradient exists associated with the Andes (Elevational
Gradient Hypothesis). The dotted outline shows the approximate distribution of P. petersi (which extends farther to the east than shown), and the dashed
lines show the locations of the Guianan (upper) and Brazilian (lower) Shields. Adapted from Funk et al., 2007.
Chapter 13 FIG 11
Figure 13.12 Geographic ranges of major clades of Physalaemus petersi based on haplotype divergences. Numbers indicate average percent corrected
sequence divergence. Adapted from Funk et al., 2007.
Chapter 13 FIG 12
Figure 13.13 History of diversification of modern amphibians. (A) Phylogeny of modern amphibians with geological timescale across the top. (B) Net
diversification rates for amphibian clades. Clade numbers refer to those in (A). Net diversification rates (d – b, where b = speciation rate and d is extinction
rate) per clade are shown under the lowest possible relative extinction rate (red, d:b = 0) and an extremely high possible rate (blue, d:b = 0.95). (C)
Comparison of proportion diversity of extant amphibian clades in the Late Cretaceous (left) and now (right). Adapted from Roelants et al., 2007.
Chapter 13 FIG 13
Figure 13.14 History of global patterns of amphibian net diversification. (A) Rate through time (RTT) plot derived from the time tree (Fig. 13.13) compared
with models varying in relative extinction rates from 0 to 0.95. (B) RTT plot of net diversification rates estimated under low extinction rates (red, d:b = 0)
and high extinction rates (blue, d:b = 0.95) for successive 20-million-year intervals (280–100 mybp) and 10 million year intervals (100–20 mybp). Circles
and asterisks indicate estimates that differ significantly from those expected under low extinction rates (d:b = 0) and high extinction rates (d:b = 0.95),
respectively. (C) Amphibian net extinction rates (blue) compared with amniote family origination (green) and extinction (red) rates based on the fossil
record. Note that the amphibian data (blue) are represented on a log scale, and thus differences are even more dramatic than shown. Adapted from
Roelants et al., 2007.
Chapter 13 FIG 14
Figure 13.15 Divergences that most affected global distribution of Microhylidae and Natanura occurred in the Cretaceous. (A) Molecular time tree
phylogeny showing divergence patterns. (B) Horizontal colored bars and lines at interval nodes (standard deviation and 95% credibility intervals) indicate
vicariance events as follows: orange: Australia <–> Indo-Madagascar; yellow: Africa <–> South America; blue: Africa <–> Indo-Madagascar; purple:
Madagascar <–> India (Seychelles); green: South America–Antarctica <–> Indo-Madagascar (with the Kerguelen Plateau involved). (B) Gondwana in the
Late Cretaceous. Abbreviations: AF = Africa, MA = Madagascar, In = India, EU = Eurasia, SA = South America, AN = Antarctica, AU = Australia–New
Guinea, KP = Kerguelen Plateau. Adapted from Van Bocxlaer et al., 2006.
Chapter 13 FIG 15
Figure 13.16 Biogeographic history of ranoid evolution. Dashed branches are lineages of uncertain phylogenetic position. Colored bars across the top of
the phylogeny indicate age of ranoid fossils from their respective continents: (1) undetermined ranoids from the Cenomanian of Africa, (2) Ranidae from
the Maastrichtian of India, (3) Raninae from the Late Eocene of Europe, and (4) Raninae from the Miocene of North America. Gray shading indicates an
apparent lack of dispersal between Africa and other biogeographic units (between nodes 6 and 17) for about 70 million years. The K–T (Cretaceous–
Tertiary) boundary is indicated by the vertical dashed line. Asterisks indicate calibration points. Adapted from Bossuyt et al., 2006.
Chapter 13 FIG 16
Figure 13.17 Dated phylogeny of ranoid frogs centering on the phylogenetic position of four families endemic to the Western Ghats of India and hills of Sri
Lanka, the Ranixalinae, Micrixalinae, Lankanectinae, and Nyctibatrachinae. The phylogeny demonstrates that these clades are outside (sister to) other
ranoids. Molecular dating places the origin of the clades containing these four subfamilies in the Cretaceous. Adapted from Roelants et al., 2004.
Chapter 13 FIG 17
Figure 13.18 The recently described frog Nasikabatrachus sahyadrensis is among the oldest of the Neobatrachia and ties the fauna of the Seychelles to
the fauna of India. Its ancestors must have been present on the Indo-Madagascan fragment of eastern Gondwana during Middle–Late Jurassic or Early
Cretaceous. Photograph by S. D. Biju.
Chapter 13 FIG 18
Figure 13.19 Early diversification of the Bufonidae occurred near the end of the Upper Cretaceous, failing to confirm a Gondwana origin of the family.
Diversification into modern genera occurred later, during the mid-Paleogene. Horizontal bars and shaded rectangles indicate 95% credibility intervals of
estimates of divergence times. Colors indicate geographical distributions of each lineage. Adapted from Pramuk et al., 2008.
Chapter 13 FIG 19
Figure 13.20 These maps illustrate the key geological events associated with diversification in the Bufonidae. (A) Bufonids originated in South America
about 88 mybp, after the breakup of Gondwana. At some point, bufonids dispersed into the Old World and diversified into the Eurasian and African clades,
likely across Beringia. (B) Approximately 43 mybp, during the Eocene, bufonids dispersed back into the New World. Although at least three possible routes
existed (Berengia, DeGeer, and Thulean land bridges), the Thulean land bridge is most likely because it provided a much more mild climatic regime.
Adapted from Pramuk et al., 2008.
Chapter 13 FIG 20
Figure 13.21 Phylogenetic relationships of eleutherodactyline frogs showing geographical distribution for each clade. Adapted from Heinicke et al., 2007.
Chapter 13 FIG 21
Figure 13.22 The origins of Middle American and Caribbean clades of eleutherodactyline frogs can be modeled based on the timing of divergences. (A)
Dispersal over water from their South American origin probably occurred during the Middle Eocene (49-37 mybp), resulting in the formation of the Middle
American clade (MAC) and the Caribbean clade (CC). (B) Higher sea levels led to isolation of a western Caribbean clade (WCC) on Cuba and an eastern
Caribbean clade on Hispaniola and Puerto Rico during the Early Oligocene (approximately 30 mybp). (C) Dispersal from Cuba to the mainland led to the
radiation of the subgenus Syrrhopus in southern North America during the Early Miocene (approximately 20 mybp). Concurrently, members of the ECC
and South American clade (SAC) colonized the Lesser Antilles. (D) The closing of the Isthmus of Panama during the Pliocene (approximately 3 mybp)
resulted in overland dispersal of MAC species to South America and SAC species to Middle America. Adapted from Heinicke et al., 2007.
Chapter 13 FIG 22
Figure 13.23 The first major divergence event in the history of caecilians occurred on Gondwana approximately 178 mybp with several other Gondwana
divergences. This deep divergence accounts for the presence of caecilians on most southern continents today. However, an added twist is the much more
recent dispersal (40-53 mybp) of members of the Ichthyophiidae into Southeast Asia. Adapted from Pough et al., 2003, and Wilkinson et al., 2002b.
Chapter 13 FIG 23
Figure 13.24 Divergence of African caecilians cannot be tied to a single biogeographical event. (A) Left, phylogeny based on 12S and 16S gene
sequences; right, uncorrected lognormal molecular clock showing divergence times. (B through G) West (C, E, G) and East (B, D, F) African caecilians.
(H) Map of Africa showing current non-overlapping distributions of West and East African caecilians. Adapted from Loader et al., 2007.
Chapter 13 FIG 24
Figure 13.25 Although it has been assumed that small fossorial amphibians and reptiles would not be able to disperse across oceans, it appears that
amphisbaenians have done just that. Based on the dated phylogeny and the position of landmasses at the time, the only supportable hypothesis for
dispersal of Amphisbaenidae ancestors to the New World is transatlantic during the Eocene (arrow 1, upper left). The most likely hypothesis for dispersal
of cadeids is transatlantic during the Eocene (solid arrow 2, upper left), although a complex terrestrial dispersal cannot be ruled out (dashed arrow 2, upper
left). Adapted from Vidal et al., 2008.
Chapter 13 FIG 25
Figure 13.26 Subaerial (surface) connections between Madagascar and Antarctica likely existed approximately 90 mybp. Dark shading indicates
submerged areas. Mad = Madagascar, KP = Kerguelen Plateau, GR = Gunnerus Ridge, and EB = Enderby Basin. Adapted from Noonen and Chippindale,
Chapter 13 FIG 26
Figure 13.27 Phylogenetic relationships of three reptile clades: (A) pleurodont iguanid lizards, (B) boine snakes (including Ungaliophiidae and erycine
genera Eryx, Charina, Calabraria, and Lichanura), and (C) Podocnemid turtles. Clades of interest are indicated by thick branches, and colors correspond to
shaded geographical distributions. Adapted from Noonen and Chippindale, 2006.
Chapter 13 FIG 27
Figure 13.28 Distribution of the nine species of Crotaphytus. Circles indicate sampling localities for phylogenetic analysis. Adapted from McGuire et al.,
Chapter 13 FIG 28
Figure 13.29 Phylogenetic relationships of crotaphytid lizards based on mtDNA sequence analysis. Note that species identified on the basis of morphology
(names) do not sort out on the gene tree. Rather, some species (e.g., C. bicintores) are nested in clades with other species. For a more detailed
phylogeny, see original article. Adapted from McGuire et al., 2007.
Chapter 13 FIG 29
Figure 13.30 Model showing the mechanism ("introgression conveyor") that resulted in Crotaphytus bicintores populations in southwest Arizona acquiring
mitochondrial haplotypes from adjacent populations of C. collaris. Timing for introgression events 1, 2, and 3 are estimated at 3 mybp, 1 mybp, and recent,
respectively. Adapted from McGuire et al., 2007.
Chapter 13 FIG 30