Human judgment is notoriously fallible and perhaps seldom more so than in facile decisions that a character has no adaptive significance because we do not know the use of it.

1

Human judgment is notoriously fallible and
perhaps seldom more so than in facile decisions
that a character has no adaptive significance
because we do not know the use of it. George
Gaylord Simpson, 1953
Source Simpson 1953, The Major Features of
Evolution, University of Chicago Press
2
But is everything really an adaptation?

• A trait might not enhance survivorship or
  fecundity relative to alternative traits
• A trait might not have originated for its current
  function

3
Spandrels of San Marco
Source Gould Lewontin 1979, Proceedings of
the Royal Society 205, 581-598
4
Adaptation a character that arose by natural
selection for its present function
Exaptation a character that is adaptive but did
not evolve for its present function
Source Gould Vrba 1982, Paleobiology 8, 4-15
5
Keas use sharp beaks to rip through the skin of
sheep.
Because all parrots have a similar beak, it is
clearly an exaptation.
The Kea (Nestor notabilis)
6
Evidence of Adaptation
1. Place in an historical perspective 2. Make
sure selection has occurred 3. Identify
ecological basis for selection 4. Determine
heritability of the trait 5. Map gene flow
within the population
Source Brandon 1990, Adaptation and
Environment, Princeton University Press
7
Environment
E0
E1
E1
E1
E1
Trait
T0
T1
T0
T1
T1
F0
Function
F1
F0
F1
F1
A
E
B
C
D
Taxon
Evidence for Adaptation
Sources Coddington 1988, Cladistics 4,
3-22 Baum Larson 1991, Systematic Zoology 40,
1-18
8
Environment
E0
E1
E0
E1
E1
Trait
T0
T1
T1
T1
T1
F0
Function
F1
F1
F1
F1
A
E
B
C
D
Taxon
Evidence for Exaptation
Sources Coddington 1988, Cladistics 4,
3-22 Baum Larson 1991, Systematic Zoology 40,
1-18
9
Terminology
Cladea monophyletic group of taxa.Sister
Taxonthe taxon that is most closely related to
the taxon of interest.Outgroupthe taxon that
is most closely related to the clade of interest
(or the ingroup) an outgroup can be used to
establish the ancestral character state.
Parsimonythe assumption that true evolutionary
scenario involved the minimal number of
transitions to produce the observed pattern of
phenotypic variation.
10
Homologysimilarities of phenotype among taxa
because of common ancestry.Homoplasysimilaritie
s of phenotype among taxa because of
convergence.Pleisiomorphya primitive character
shared by some or all members of a
clade.Synapomorphya derived character shared
by some or all members of a clade. Autapomorphy
a derived character that is unique to a taxon
(i.e., not present in the outgroup)
11
Adaptation of Lizard Toes
Gecko Feet
Source Larson Losos 1996, pp. 187-220 in
Adaptation, Academic Press
12
The Comparative Method

• Examine phenotypes of taxa that inhabit different
  environments.
• Assumes
• 1) Current environments reflect past
  environments (at least for long enough period for
  phenotypes to reach an evolutionary equilibrium).
  
• 2) Genetic constraints have not prevented
  natural selection from producing the optimal
  phenotypes in each environment.

13
Latitudinal Variation in Body size Drosophila
subobscura
Huey et al. 2000, Science
14
Why one should compare gt2 species?

• A difference between two species could be caused
  by any number of reasons besides the one
  preferred by an investigator.
• For example, flies in Denmark could be larger
  than flies in Spain because their environments
  are colder or because their environments contain
  more food.

Source Garland Adolph 1994, Physiol. Zool.
67 797-828
15
The Problem of Non-independence

• Comparative data for populations or species are
  not independent, as assumed by parametric
  statistics.
• The true degrees of freedom are always fewer than
  the number of data points.

16
Latitudinal Size Clines in Sceloporus
? Bergmanns cline ? No cline
Source Angilletta et al. 2005, Am. Nat. 164 in
press
17
Phylogenetic Comparative Analyses

• When you do not know the phylogeny
• Partition variance by taxonomy to determine which
  levels will be useful for analyses of adaptation
• When you think you know the phylogeny
• Factor out variance described by phylogeny and
  examine residual variance
• Compute phylogenetically independent contrasts

Source Miles Dunham 1993, Annu. Rev. Ecol.
Syst. 24 587-619