Recovering Phylogenetic Signal from DNA Sequences

Recovering Phylogenetic Signal from DNA Sequences:
Relationships within the Corvine Assemblage (Class Aves)
as Inferred from Complete Sequences of the Mitochondrial
DNA Cytochrome-b Gene ’
Kathleen Helm-Bychowski * and Joel Cracraft “f
*Department
of Chemistry, DePaul University;
American Museum of Natural History
and t Department
of Ornithology,
Phylogenetic analysis of cytochrome-b sequences and cranial osteological characters
for nine genera of corvine passerine birds supports the hypothesis that the two
major groups of birds of paradise, the manucodines
and paradisaeinines,
constitute
a monophyletic
group and that their postulated sister group is the Corvidae (crows,
jays, and allies). The data are also consistent with the hypothesis that the bowerbirds
are not closely related to the birds of paradise but instead lie near the base of the
corvine assemblage. The corvine radiation exemplifies a case of multiple star phylogenies embedded within a major clade, with the branching pattern characterized
by very short internodal divergence times. Such histories are difficult to resolve no
matter what type of data is employed, because little change accumulates between
branching events. With respect to sequence data, reconstructed
tree topologies are
sensitive to the choice of outgroup and to the method of analysis (e.g., transversion
vs. global parsimony).
In such cases, assessing the “reliability” of a best-fit or mostparsimonious
tree inferred from any particular data set becomes problematic. Statistical tests of tree topologies that depend on random sampling of characters will
generally be inconclusive
in that all cladistic components
will tend to be poorly
supported because relatively few character-state
changes will be recorded between
branching events. It is suggested, on the other hand, that congruence in cladistic
signal across different data sets may be a potentially more useful method for evaluating the reliability of the signal of any one data set. Resolution of star phylogenies
will probably be possible only if DNA sequence and morphological
characters are
combined in a single analysis.
Introduction
With the rise of molecular
systematics over the past decade, there have been
expectations
that molecular data, especially DNA sequences, will solve most of the
long-standing
problems about phylogenetic relationships.
Although those expectations
are still widespread, and although many of the problems indeed seem nearer resolution,
the use of molecular
data has not automatically
led to increased understanding
of
phylogenetic
history for many groups. One reason for this is that recovering relationships from DNA sequences is a complex problem lacking a straightforward
solution:
there are many methods of phylogenetic
inference that can be applied to sequence
analysis, and, within the context of each, numerous
models and assumptions
about
1. Key words: mitochondrial
lonorhynchidae,
Corvidae.
DNA, cytochrome-b
gene, parsimony
analysis,
Aves, Paradisaeidae,
Address for correspondence
and reprints: Joel Cracrafi, Department of Ornithology,
of Natural History, Central Park West at 79th Street, New York, New York 10024.
Mol. Bid. Evol. 1O(6): 1196- 12 14. 1993.
0 1993 by The University of Chicago. All rights reserved.
0137~4038/93/1006-0005$02.00
1196
American
Pti-
Museum
Corvine
Phylogeny
from Mitochondrial
Cytochrome-b
Genes
1197
sequence change might be adopted, thereby potentially
altering any inferred phylogenetic hypothesis. Methods employing a parsimony criterion of minimal characterstate change to sequence data offer distinct theoretical and methodological
advantages
in that assumptions
can be minimized
and/or made explicit, thus permitting a more
detailed evaluation
of their influence (Cedergren et al. 1988; Smith 1989; Swofford
and Olsen 1990; Cracraft and Helm-Bychowski
199 1; Hillis 199 1). Just as important,
currently available algorithms are sufficiently powerful to enable the discovery of exact
solutions under a variety of assumptions
about sequence change, except in cases involving large numbers of taxa.
Although the best-fit or most-parsimonious
tree explains the data better than do
less optimal trees, this does not guarantee that those solutions will represent the true
phylogenetic history, and the failure to find that history may be due to the investigator’s
choice of algorithmic
method, to the assumptions
underlying the analysis (e.g., how
sequence changes might be weighted), or, more likely, to the quality and quantity of
the data. Difficulties in resolving phylogenetic
history may also be a function of the
history itself, in that attempts to reconstruct
the cladistic structure of lineages that
have diverged closely in time-so-called
star phylogenies-can
yield many conflicting
best-fit solutions, depending on the method of analysis.
We have two goals in this paper. The first is to explore how sequence data, in
this case those derived from mitochondrial
DNA (mtDNA),
might be used to resolve
phylogenetic
signal within a radiation containing
multiple star phylogenies.
We will
suggest that the sequence data are not adequate, by themselves, to specify a strongly
defined cladistic signal but rather that the latter becomes more interpretable
when
other kinds of data are taken into consideration.
The second goal is to present additional data that help resolve relationships within
a major radiation of songbirds. The birds of paradise (Paradisaeidae)
and bowerbirds
(Ptilonorhynchidae),
both endemic to the Australopapuan
region, have long been
considered
members of the corvine assemblage, a large group of family-level
taxa
whose interrelationships
are still incompletely
understood.
Although paradisaeids and
ptilonorhynchids
have been thought to be related on the basis of morphological
similarities (Bock 1963)) DNA hybridization
data support the removal of the bowerbirds
from the vicinity of the birds of paradise and place them near the base of the corvine
assemblage (Sibley and Ahlquist
1990). Edwards et al. ( 199 1) used 924 bp of the
mitochondrial
cytochrome-b
gene to examine relationships
among 14 passeriform
taxa including
one paradisaeid
and one ptilonorhynchid.
Their analysis separated
these two taxa relative to one another but did not provide a clear resolution of their
relationships
within the oscine songbirds. We use sequence variation for the entire
cytochrome-b
gene ( 1143 bp) to address the following phylogenetic questions: ( 1) Are
the two major groups currently included in the Paradisaeidae-the
manucodines
and
paradisaeinines
-monophyletic?
(2) What is the sister group of the paradisaeids? (3)
Are the bowerbirds closely related to the birds of paradise, or are they positioned near
the base of the corvine assemblage?
Material and Methods
Taxa Examined and Source of DNA
Eleven species, representing a spectrum of corvine higher taxa and two outgroups,
were investigated in this study: four paradisaeids-including
a manucodine,
Manucodia
( = Phonygammus) keraudrenii, and three paradisaeinines,
Epimachus fastuosus, Ptiloris paradiseus, and Diphyllodes magn$cus; two species of the Ptilonorhynchidae
(Ptilonorhynchus violaceus and Ailuroedus melanotus); a member of the Corvidae
1198
Helm-Bychowski
and Cracraft
(crows, jays, and allies; Cyanocitta cristata); a true shrike (Lanius ludovicianus; Laniidae); a vireo (Vireo olivaceus; Vireonidae);
and two noncorvines,
including a thrush
(Catharus guttatus; Turdidae) and a New World suboscine passerine, the least flycatcher (Empidonax minimus; Tyrannidae).
Total genomic DNA was isolated from
frozen tissue (muscle, liver, and heart), using standard methods (Lansman et al. 198 1).
Data Collection
Fragments
of the cytochrome-b
gene were amplified by the polymerase
chain
reaction (PCR) using standard reaction conditions (Kocher et al. 1989). Combinations
of eight oligonucleotide
primers were used for amplification
and sequencing of overlapping fragments in both directions, including those that follow [“L” and “H” refer
to the light strand and heavy strand, respectively; numbers refer to the chicken sequence
(Desjardins
and Morais 1990)] : ( 1) L14827 (ND5),
5’-CCACACTCCACACAGGCCTAATTAA-3
‘; ( 2 ) L 149905 ‘-CCATCCAACATCTCAGCATGATGAAA-3
‘;
(3) Ll5068,5’-ACTAGCAATACACTACACAGCAGA-3’;
(4) HI 5 104,5’-GAGTCAGCCATATTGGACGTCTCGGC-3
‘; ( 5 ) L 15236,5 ‘-TACCTAAACAAAGAAA(7) H15298,
CCTGAAA-3’;
(6) L15311,5’-CTACCATGAGGACAAATATC-3’;
5 ‘GCCCCTCAGAATGATATTTGTCCTCA-3
‘; ( 8 ) L 15506, 5 ‘-CTCACCTTCCTACACGAAACAGG-3’;
(9) H 15505, 5’-CTGCATGAATTCCTATTGGGTTGTTTGATCC-3’;
(10) L15656, 5’-AACCTACTAGGAGACCCAGA-3’;
(11) H15710,
5 ‘-GTAGGCGAATAGGAAGTATC-3
’ ; ( 12 ) L 15920, 5 ‘-ACATGAGTCGGAAGCCAACC-3’;
( 13) H15914,5’-GGTTGTTCTACTGGTTGGC-3’;
and ( 14) H16065
(tRNA-Thr),
5’-GGAGTCTTCAGTCTCTGGTTTACAAGAC-3
‘. (Primers 2 and 7
were modified according to the method of Kocher et al. 1989, and primers 5, 6, and
14 are courtesy of P. Arctander, S. Edwards, and S. Paabo of the A. C. Wilson laboratory, Berkeley; all others were designed in our laboratory).
Single-stranded
template for sequencing was produced by asymmetric PCR using
either unbalanced
primer ratios (Gyllensten and Erlich 1988) or single primers (Allard
et al. 199 1). Some double-stranded
fragments were sequenced after recovery from
low-melting temperature
agarose ( Kusukawa et al. 1990; Qian and Wilkinson
199 1).
The products of dideoxy chain-termination
sequencing reactions (Sequenase system;
United States Biochemical)
were subjected to denaturing
gel electrophoresis
and autoradiography.
Sequences were aligned by eye, with the aid of the ESEE editor program
(Cabot and Beckenbach
1989). All nucleotides were sequenced in each direction.
Morphological
Data
In addition to the mitochondrial
sequence data, phylogenetically
informative
variation from 14 cranial osteological characters was also analyzed, using a subsample
of data from a larger morphological
study of corvine relationships
(J. Cracraft, unpublished data). A description
of the characters and character states, along with a
data matrix, is presented in the Appendix.
Polarization
of character-state
transformations was postulated by using outgroup comparison.
Phylogenetic
Analysis
The results presented here are part of an ongoing study of corvine and paradisaeid
relationships.
Phylogenetic analyses were undertaken with the program PAUP, version
3.1 ( Swofford 1993 ) . All sequence transformations
were analyzed unordered; all morphological character-state
transformations
were ordered (see Appendix).
The branchand-bound
option was used to guarantee discovery of the most parsimonious
solution.
Various methods exist for identifying cladistic signal and evaluating its strength
Corvine
Phylogeny
from Mitochondrial
Cytochrome-b
Genes
1199
in any given analysis (Swofford and Olsen 1990; Cracraft and Helm-Bychowski
199 1;
Hillis 199 1) . Here cladistic signal was identified by searching for most-parsimonious
trees. When multiple equally parsimonious
trees were found, the common signal was
identified by the use of strict consensus trees. Within the context of a parsimony
analysis, the point of departure for evaluating the strength of cladistic signal is a comparison of branch lengths, or the amount of synapomorphous
character change postulated to support each clade. For any given data set, however, character change might
be optimized in a variety of ways. To establish a consistent method across analyses,
branch lengths of most-parsimonious
trees were optimized by using the DELTRAN
option in PAUP, which distributes character change toward the tips of the tree. In
several instances, relative strength of signal was also evaluated by, first, examining the
stability of the cladistic components
in near-most
parsimonious
trees (Bremer 1988;
Cracraft and Helm-Bychowski
199 1; Kallersjii et al. 1992) and, second, by the bootstrap
resampling technique (Felsenstein
1985). Results from the latter method are difficult
to interpret statistically, because its underlying assumptions
are rarely, if ever, satisfied
by data such as coding DNA (Jones et al. 1993); consequently,
the bootstrap is used
here solely as a heuristic guide to the strength of signal.
Analysis of distances was accomplished
by using tree-building
routines (FITCH,
KITSCH ) in the PHYLIP program package (Felsenstein
199 1). Taxa were added to
each analysis randomly, and replicate analyses were undertaken
to search for best-fit
trees. Sequence divergences were estimated by using the DNADIST routine in PHYLIP.
Results
Nucleotide Sequences,
Acid Replacements
Transition
/ Transversion
Ratios, and Amino
Nucleotide sequences ( 1,143 bp) for the 11 taxa of this study are aligned in figure
1; the alignment is colinear with that of the chicken cytochrome-b
gene (Gallus gallus;
Desjardins and Morais 1990).
A matrix of transition / transversion
differences and their ratios is shown in table
1. We infer that the suboscine flycatcher Ernpidonax exhibits saturation of transition
substitutions
when compared to the other taxa, and the noncorvine
thrush Catharus
is nearing saturation.
Within the corvines, the bowerbirds Ailuroedus and Ptilonorhynchus are also nearing saturation relative to the other corvines, as is Vireo and, to
a lesser extent, Lanius. The three paradisaeinines
exhibit the fewest transversion
differences among themselves ( 14-22). The remaining taxa form four distinct clusters
with respect to the numbers of transversion
differences they show relative to the paradisaeinines (see brackets in table 1), as follows: (a) Manucodia and the jay Cyanocitta
( 5 1-57 transversion
differences),
(b) the shrike Lanius and the vireo (67-73 transversion differences),
(c) the two bowerbirds and the thrush (79-85 transversion
differences), and, finally, (d) the flycatcher ( 113-12 1 differences).
The significance of
this pattern will be discussed later.
Estimated total sequence divergence corrected for multiple hits (Jukes and Cantor
1969; Kimura
1980) is presented in table 2. Divergences
among the three birds of
paradise are approximately
lo%- 1l%, and they, in turn, exhibit a 15%- 18% difference
from Manucodia and a 17%-19% difference from the corvid Cyanocitta. The vireo
and shrike show divergences from the paradisaeinines
roughly similar to those of the
manucode
and jay. The bowerbirds, on the other hand, exhibit a 20%-22% sequence
divergence from the birds of paradise, a value similar to that shown by the thrush
Catharus. The flycatcher Empidonax is decidedly more distant, at 28%-30%.
Of the 38 1 codons, 297 (78%) show no variation across the 11 taxa studied.
Ptiloris
Epimachus
Diphyllodes
Manucodia
Cyanocitta
Lanius
Vir.20
Ailuroedus
Ptilonorhynchus
Catharus
Empidonax
ATGGCA~U\ATCIACGT~~A~~A~~T~T~~A~C~T~TCGAC~C~A~C~T~CAT~~TCTGATG~~~~ATCC~T~A~MTCTGC~MTCACA~GATTATCACA~C~A
... ..C.....C..................T
......................................
..A ..................................
..A..T.......T..............A...G
........ ..G
... ..T.....C ................. C .......... ..G...........C....T...........A.................T...........C.....T........C..T........T.....A.....T
...... ..G
... ..T ..................
..T..~.........C.......A..G..C..........C..T..C...............G~
.......... .C..T..T..A.....C..T...T..........A..C
............
... ..C..A.....................T................T..T..TC....T........T..................G~
...... ..T..C.....T..A.....C............GTG..A..C
............
... ..c ............... ..T.....CA.C......C.........TG.T..............T....................T........T..C......T.A.....C..T
........ ..TG..A..C.C...G..TT
..
... ..C.....C.................C..C.........G.A.........C....T.....A..A..C................C..................A..C.....T.........G.......A...G.T
.........
............
. ..A.C.C...C ..... G........T..C...A....................C..G....T.....A..C.....C..........C..........C.......A..C...............G.G.TC..A..C
... ..CA....C.....A.....T......A.U\..G..G.......T...G..C....T........T..C...........T...G......................C.....C..T......G
.A ... ..A..C.....C..T ..C
T .. ..C........GG.T.....A..CG....T
... ..C........T .............. A.....C....C.........TG.TC..........A.....A........T.....T.~...........C..G..A..A
......
..... ..C...CT....A...C.T.....A..TT.......AG.A..TA....TC.C........C..T..A.....T..T......GCT..............C.....A
............. ..TCA........C.....C
... ..C
11501
11501
I1501
I1501
L1501
[1501
[I501
cl501
I1501
[I501
11501
Ptiloris
Epimachus
Diphyllodes
Manucodia
Cyanocitta
Lanius
Vireo
Ailuroedus
Ptilonorhynchus
Catharus
Empidonax
~G~AGCAGCACATTACCAGCAGA~C~C~AGC~T~G~C~TAGCTCACATAT~CGAGACGTCC~TTCGGAT~~~TCCG~C~GCATGC~C~A~CTCCA~ATT~C~TTGCATCTAC~ACATATC~C
..A......A....C..............T.............A.........C.....G......A..........T.................T..A..C...........T...C.......TA.C..............C
......
....... ..A....C ........................
..T........................A.T
...........................
..A..C...........T..AC........A
.......................
...... ..CAT...C...........T..T........A....C...T...T.......T......A..................................C..............TT.C......A.C...........T
.........
..AT....CAT...C..............T...T.........CA.....G.....T..G......A.................T.............C..C
... ..T ...... ..TT.C..T..TA................C..T
...
....... ..AT..................T........T..............C..T..T...........A........C........T........C.................AT.T......A
.......................
..A......AT...C.....T.................A....C...T.....C....CC......A....G.....Y....................A..C...............T.Y.....TA.C...........T..C
......
..A.....CA..........T.....T...AA.
....... ..GC...A.....A..T..T...........A.....T..C...T..........T..A.....G........T..AT.C
... ..TA.C...........G..C
... ..A
... ..T..CA.......T .......... ..M..........GCT........C..T.C...T..GA....A..............C........TT.A
............... ..AT.C......A.C.....T.....G..C
... ..A
..T.....CATG..C ........................
..T.A...A..C..C..T..G..T...A..........T..C
............ ..T..C.................AT.C......A.C...........C..C
......
..C......AT...C.....T..G........G........T.TA..A..T..C....C......CA.............C.....C.....G.....C
................. .T.C.....TA.C..............C
... ..A
[3001
[3001
I3001
I3001
13001
[3001
[3001
I3001
13001
[3001
I3001
Ptiloris
Epimachus
Diphyllodes
Manucodia
Cyanocitta
Lanius
Vireo
Ailuroedus
Ptilonorhynchus
Catharus
Empidonax
CGA~T~A~ACGGCCATA~~~G~C~G~~TTGGAGT~TC~A~CCTM~~MTAG~CAG~T~GTCGGATACGTC~~TGA~A~TATCC~~GA~T~TACA~CATTA~~C~A
... ..A.....T.....T.....T
.....................
..C...........T...T...................C...........T.....C..C
.................................
..c .........
..T.....................A.C...........C.......................C.....A
... ..A ...............
.........
..C..A..............A........C.....C
...............
T..T.....A........G...........C.....G.................C
.......
... ..M ..........
.T..C.....A..C.....A..A..C..............A..T.....A...........C.....T
...
... ..A..T.................A..................A.C......C.G..T..AT..G.............T..C..T..............A..A..............T....................C..C
... ..T
... ..AC ............... ..A.A..T..G..G..A.....TA.C...A.T.............TA...........C........A..G..T...T.A..........................A........T..C
...... ..T
... ..AC.A.................A....................A.....C.....C..A.....T...........T........A..G
......... ..C..............A........A...........C..A
......
... ..A..A ........................
..G.........A
...............
.A..G........G..G..T..C.....A..T........C..C..............G........G..A...........C
......
... ..M.T..T..T.....C...T.A..T..............T..C...........C........T...........T..C.....A...........C
............... ..A..T.....C..............A
......
..... M ................. ..G.....G .......... ..A.C..G..T.....C.........T..........C..C
... ..G..C.....A.....C..............A........G........A..C..C
......
... ..A..T.....T..A..C...T.AT.T.
............ ..ACA..C..C..T..T..TT.......T........T..C.....T..C..T.....A..A..............A..T..............A..C..C
......
14501
[4501
[4501
14501
I4501
r4501
[4501
[4501
[4501
[4501
I4501
Ptiloris
Epimachus
Diphyllcdes
Manucodia
Cyanocitta
Lanius
Vireo
Ailuroedus
Ptilonorhynchus
Catharus
Empidonax
CPCTCAGCMTTCCATACATCGGGCAAACCCTAGTAG~TGAG~GAGGAGGAT~T~GTAGAC~CC~A~~MCCCGATT~TTG~CTCCA~TC~C~TCCA~C~~TT~A~C~GACA~A~TCAC~GACA~C
..C...........T...........T...........T..............A...........C.C...A..A........C...T.A
......
T ................. ..T .................................
T.T....................C.
.. ..T.......................C..C........T.....C
.......................
A.....T..T.....C...C....C.C......A...T....C
............
T.......C..C..C........C.....A..............C...........C..............C.......................A........A.....T
...... ..C.CC..A......T....A.....C..T
...
.. ..T..C...T.T........A.....T........C.C......A...T....C.....A..C
...
T ........ ..C...........A ............ ..G...~C.....G....................C.....G..T.
T.......C..C..C.....T..T.....A
.........................
.C.....C........A.....C..T........C..T..T...........A..T...C....C.C...A..A...T....C.....T..T
...
T..........C...........A.....A....................C.....C..............A.......................A........A
..A ......... ..C.C...A..A..............C..C
...
...... ..C..C........T..C.....A
..T ...... ..G..A.....C.....C...........T..A
............... ..C.....G..T.....T..C........C..C.C
.... ..C..G.....C.....C..G
...
T.......C..C.....T.....A.....A..T...........A.......................T..A
............ ..T..CA....A........T..C........C..C.C......C........A..T..C..C
...
T..........C..T........C.....A..G...........C
....... .C..C.....T........C........A........C.....T
..................
..C..C.C......C..............C..C
..T
...... ..A.....C..........................T........TA....C.C......C......A.C........C
..T
T ........ ..C..C........C.....A..C.......................C..T..C
16001
[6001
[6001
(6001
I6001
I6001
[6001
I6001
I6001
16001
[6001
Ptiloris
Epimachus
Diphyllcdes
Manucodia
Cyanocitta
Lanius
Vireo
Ailuroedus
Ptilonorhynchus
Catharus
Empidonax
~ACACGAAACAGGATC~~~~~~G~T~CATCAGA~G~A~~C~TTCCACC~TACTACT~ATC~GACATC~AGGATTCG~~~TA~MCC~G~A~C~~A~A~ATTCTCC~~C~A
... ..T....................C........T........C..T........C...........A
............... ..T.....G.....T...............ACA..C...A.TC
....................
..C
........................
..C........T........C...........C...........A....................T......................TTTCA..CA.TA..C...................T
..C
... ..T.....G..C...........C....................T.................T..A........T
.............................
.A.C.TT..A..T.T...TC.G...T.............T
..C
..T .....................
..C..A........C.....C.....T
..................................
..C.AT.......T.............T..CA..CAT.T..C
....................
..C
............ ..C..............A..T...........C.....T.....C.............................T..............C..........T...A.....G~.C..........T.........A
.G
..........................
C..A.....T..C.....C..T........C...........A........T
......... ..T.....C..T..C.........G.ATCA....T....C....C
..................
T................C
........ A..A..C..T........C..T....................G........A.~.................T...T.T...A.C~.T.A..C.TA...A.G..C........A
...... ..G
................. C........A..A..C...........C.......................A......A...CA..G.......................G..T...T.A....T...TA.G..C.....T
............
... ..T...........T..T.....A..A........~....C.....T........T........A..........~.............................T.T...C..CATTT.TC....C..G........C...A
..
..C..T ..................
..~.A.....~.C.....C..T........C..G
........... ..T.....CA..G...........T...ATCA.CC....TCT..CA...ATM..C....TT.......A..T...T
.G
I7501
[7501
[7501
(7501
[7501
I7501
I7501
[7501
I7501
I7501
I7501
L
0
Ptiloris
Epimachus
Diphyllodes
Manucodia
Cyanocitta
Lanius
Vireo
Ailuroedus
Ptilonorhynchus
Catharus
Empidonax
TTAGGAGACCCAGAAAA~TCACACCACCAGC~TCC~T~C~CACCCCCTCATATC~C~G~TGATATTTC~ATTTG~TACG~ATCCTCCGAT~ATCCCC~C~~A~A~AGTC~A~~A~T~TCAGTTTTA
[9001
C.......T........T.....C..G.....C..T...G...............................C
.................C...........A..........................G..C...........C..CC
19001
..
C....G .................C........C..A...G.......T..C
....................C.................C...........A....................G
.......................
cc ..
19001
C............................T..C..T...GT..............................C..T...........T..C...........A....................T..TT....~....C..C.....AC
.G
19001
C....G.....C.....T.....G..C.....C.....CG..........A..C..T..............C
.................C..T..T..G.....T..A..............G..T.....C.....C..C.....CC
..
I9001
C................T...G.G..G.....C..A...G....C..T..G.....T.....C........C.....G..C...........T..A.....A.......................A.....CT.......C...A.C
...
I9001
C......G....C..A..A...........T
C ...............................
.................C....................A.......................A..G..C.....C..C..CA.CC
..
[go01
C.......T........T.....C........C..AT..T.......T..A.....T.....G..G.....C........C........A.....A.....T........T.....T..............C.....C..A..C..AC
..
[go01
c........c.....~....G..A..C..C..T.....C.................C...........T..A.....A.......................A.....C.....C..A.....CC
C ......................
..
[9001
C ......................C.....T..C......G..........A..C........C........C........C..T..T..C.................A.................A.....C.....C..C..C..CC
[9001
..
C....C.....C........T.....C.....C..A...GTA..T..A..C
C..T..T..C..C.....A...T.A.....T.....A........T..T...........G..C..C..C..C..CC
....................
..
igo01
Ptiloris
Epimachus
Diphyllodes
Manucodia
Cyanocitta
Lanius
Vireo
Ailuroedus
Ptilonorhynchus
Catharus
Empidonax
~CCTATTCCTCATTCCTCFGCTTCACACATC~C~CGATC~T~CTTT~GAC~CTAT~C~TC~ATTCTG~~~AGT~~~C~ACT~TT~~CATGAGTA~CAGCC~~AGTCG~~C~A~CATTATT
A..........T...........C..........................C..............G........G........C....................C..G
...........T.................T........C..C
A................C..A.....T..G.............................T.......................C
.......................G...........T..........................C
..C
...T.............AT.A..C.
...A............C........C........TT..........T..........~......G.C..T.....CG..T..........G.................G...........C
...
...........TG.C..AT.C..A..TGTC...........C.....G...........T.....C
C ......CG.AG.T.....C..C...........C..A...........A........G
................
........C
....................A........C.....C...
.........T.A..C.....A..C....A............T.................T..G..T...............G~......G.C...A.T..C.....G..C
...........A..A..A..A........C..............C.....C.....T........C
..C
........C......G.G..G..T..C...G....C..C.....T...A.C..............T.....G........C
..A..G.....G..C..C.......................................
..T.....C................CA.....TG
.............C..............A...........A........G.....C
..C
..A.........C.A..C..A..A..........................C
.....C.....C..............T...GCC.....CT.G...A.C.....C..G...........A...........G..G
...............
..C
............T.......A..C....A...T........C........C..............C........C
.......CCT....TG.A.....CT..G.G..C..C........A...........A..............C
...........TGCC..AT.C..C....T...A........CA
.......C..T..T.....T..C...C.A..........CC.......G.C
........CG....T......A.C..A........T..A..............C
..C
Ptiloris
Epimachus
Diphyllcdes
Manucodia
Cysnocitta
Lanius
Vireo
Ailuroedus
Ptilonorhynchus
Catharus
Empidonax
ATCGGACAATTAGC~CATT~C~A~T~T~TTGTT~AGT~TATT~C~~GT~GTGTGffAG~C~~A~C~CC~T~
..T.....GC.............T.......C....A
.C............C.G.C..A........T...T....T.....C
.......
...
.........C........C............C...cA.........C.......A..C.....C..A
...........................
............A.......AA.....
.........G...........A........C.....A........C.......T....C......A
..T......C....T..C..TG.........C...CA.C...A................G....CA............A..........C
...
..T..T...C.............
T......C....CC........C.......T..C.C...C..A............A.......A
......
.........C....T...C..AG........C...CA.........C.......T..C.C.G....A............A....A..A
......
.....C...C........A....A.......C...CA.C...A.TC.G........C.CCGCA.CC............A.C.....A..C
...
........GC........C..........T.C...U\.C...T..C.G.....A....~G~.~............A....G.....C...
.....c...c........A....T.......C...C........C........C...C.GC...A........T...A....G..A..C
...
..T..C............C..A.........CC..CF.A..~.............CA.CG.GA~.....G..T........A..AT.C
..G
FIG. 1.-Aligned
cytochrome-b
gene sequences
lll431
111431
[11431
[11431
[11431
I11431
[11431
[11431
[11431
[11431
[11431
for the 11 species used in the present study
[10501
[10501
[10501
[10501
[10501
[10501
[10501
[10501
[10501
[10501
[10501
1202
Table 2
Distance Matrix (Percent Sequence Divergence) for Cytochrome-b Gene for 11 Taxa
Taxon
;3
Es
1.
2.
3.
4.
5.
6.
7.
Ptiloris .........
Epimachus ......
Diphyllodes .....
Manucodia ......
Cyanocitta ......
Lank
.........
Vireo ..........
8. Ailuroedus ......
9. Ptilonorhynchus
10. Catharus .......
II. Empidonax .....
NOTE-Values
transition/transversion
.
above diagonal
ratio.
1
2
3
4
5
6
7
8
9
10
11
...
10.9
11.3
10.7
16.6
19.3
18.5
16.5
21.6
20.1
20.9
29.9
...
10.3
10.2
10.6
18.0
19.4
19.5
15.8
20.8
22.1
19.1
29.1
...
15.3
16.6
14.7
15.9
18.3
19.0
15.4
21.2
21.0
19.5
28.4
...
17.8
17.8
16.8
18.3
20.3
19.2
17.5
24.7
24.1
19.4
29.8
16.8
17.6
17.1
17.4
19.5
21.9
20.7
25.2
25.4
21.6
28.3
15.0
13.9
14.0
15.8
18.5
18.1
...
19.1
18.5
18.8
21.4
21.5
22.0
17.4
19.7
19.7
19.3
26.0
...
18.6
19.5
18.7
21.0
21.9
20.6
17.4
16.8
18.6
22.2
31.1
...
18.5
17.0
17.4
17.2
19.0
19.1
17.0
19.4
19.3
22.0
32.0
24.9
24.2
23.6
24.5
23.8
26.4
21.7
25.5
25.6
22.5
26.8
are corrected
by using Jukes and Cantor’s
(1969) method;
...
ii.;
25.4
23.7
21.6
32.2
values below diagnoal
are corrected by using Kimura’s
(1980) two-parameter
. ..
model with a 10: I
1204
Helm-Bychowski
and Cracraft
Blocks of conserved codons roughly match the postulated positions of redox centers,
whereas more variable regions match transmembrane
domains (Howell 1989). This
is similar to the overall pattern of variation that has been observed for mammalian
b (Irwin et al. 199 1 ), although the avian sequences presented here are
cytochrome
somewhat more conserved.
Patterns of base composition
are similar to those found for passeriform birds by
Edwards et al. ( 199 1, fig. 3). All four bases are in approximately
equal frequencies
for first codon positions; T (40.7%) predominates
at second positions, whereas G
( 12.9%) is underrepresented;
and at third positions, G (3.5%) and T ( 13.9%) are both
in low frequency. Overall frequencies were G ( 13.0%)) A ( 29.3%)) T (25.0%)) and
C (32.1%).
Parsimony
Analysis
of Corvine
Interrelationships
mtDNA Sequences
Two parallel analyses were undertaken,
the first including
all substitutional
changes (global parsimony)
and the second including
only transversion
changes
(transversion
parsimony).
Because transition/ transversion differences (table 1) among
the taxa suggest that transition
substitutions
are approaching
saturation
in some of
the more distantly related taxa, phylogenetic
relationships
of these deep branches are
more likely to be revealed by using transversion
parsimony
analysis (Swofford and
Olsen 1990; Cracraft and Helm-Bychowski
199 1; Kraus and Miyamoto
199 1). In
contrast, transition substitutions
should become more informative as younger branching events are examined.
Parsimony
analysis of all 11 taxa yields a single most-parsimonious
tree for all
11 taxa of 1,135 steps when all variation is considered (fig. 2A) and a single tree of
447 steps when only transversions
are examined (fig. 2F). Using the the New World
flycatcher Empidunax as the root, both trees identify two strongly supported clades,
one with the three genera constituting
the paradisaeinines
and the other with the two
bowerbirds.
Otherwise, the two trees share few similarities.
The noncorvine
thrush
(Catharus) lies at the base of the tree in the transversion analysis, which is consistent
with DNA hybridization
distances (Sibley and Ahlquist 1990, pp. 832-834).
In both
topologies, Empidonax has a very long branch compared with all others on the tree,
which suggests the possibility that such a distant outgroup may be influencing
the
resulting topology (Wheeler 1990).
When Empidonax is deleted from the analysis, a single tree of 970 steps was
found for the global parsimony
analysis (fig. 2B), whereas 10 equally parsimonious
trees resulted from the transversion
parsimony
analysis (fig. 2G). When the results
of both figure 2F and DNA hybridization
(Sibley and Ahlquist 1990, pp. 832-834)
are followed, the noncorvine
Catharus can be used to root the tree. With Empidonax
eliminated,
both trees show more congruence.
The paradisaeinines
and ptilonorhynchids are still strongly supported,
but, in both analyses, the latter clade is now the
sister group to all other taxa, a position that is supported by DNA hybridization
distances (Sibley and Ahlquist 1990, p. 859). In the transversion
analysis of figure
2G, moreover, all 10 most-parsimonious
trees identified the bowerbirds as the sister
group of the other corvines. In the global analysis Vireo lies outside the shrike Lank,
the corvid Cyanocitta, and the four paradisaeids, but the transversion
parsimony analysis does not clearly resolve these relationships.
The anomalous
placement together
of Cyanocitta and Lanius in the global analysis is likely the result of both having very
long branches ( see below ) .
Very similar patterns of relationships
are maintained
when Catharus is deleted
GLOBAL PARSIMONY
TRANSVERSION
PARSIMONY
Ptiloris
Epimachus
Diphyllodes
Manucodia
Cyanocitta
LMius
Vireo
Ailuroedus
Ptilonorhyrtchus
CathLttW
D
Ptiloris
Epimachus
Diphyllodes
Cyanocitta
Mamcodia
Lmiu.9
FIG. 2.-Global
(A-E) and transversion (F-J) parsimony analyses of the cytochrome-b
data set, with
outgroups successively deleted (see text). Lengths of branches for all most-parsimonious
trees are shown
above branch, and percentages of 500 bootstrap replicates are shown (in brackets) below the branch; branch
lengths are not shown on strict consensus trees. Tree statistics (tree length, consistency index for informative
characters, and retention index) are as follows: A = 1135, 0.502, and 0.320; B = 970, 0.522, and 0.341; C
= 862,0.544, and 0.358; D = 628, 0.585, and 0.337; E = 528,0.626, and 0.378; F = 447,0.430, and 0.375;
G = 361, 0.462, and 0.408 (strict consensus of 10 trees); H = 305, 0.513, and 0.442; 1 = 199, 0.574, and
0.490 (strict consensus of 2 trees); and J = 155, 0.637, and 0.554.
1206 Helm-Bychowski and Cracraft
from the analysis and the bowerbirds are used as the root of the trees (fig. 2C and H).
A single most-parsimonious
tree of 862 steps was found in a global analysis (fig. 2C),
and a single tree of 305 steps resulted from the transversion
parsimony analysis (fig.
2H). The topologies of the global analyses shown in figures 2B and C are identical,
and, in the transversion
analysis, the tree of figure 2H is identical to one of the mostparsimonious
trees summarized
in the consensus of figure 2G.
Transversion
distances (table 1) indicate that Vireo and Lanius are much more
distant from the birds of paradise than are either Cyanocitta or Manucodia. Thus,
when the bowerbirds are deleted, and when Vireo is chosen as the root of the tree
(Vireo generally lies more distant from the paradisaeinines
than does Lank; also see
table 1)) the resulting topology of a global analysis (fig. 2D) is the same as that of
figure 2C. In the solution of figure 2D, Lanius and Cyanocitta again join, but other
evidence does not support their close relationship
(see below); on this tree both have
very long branches relative to the other taxa. A transversion
analysis in which the
bowerbirds have been deleted (fig. 21) produces the same topology as the tree of figure
2H, if both trees are rooted on Vireo. Two equally parsimonious
trees of 199 steps
result, with either Vireo or Lanius rooting the tree at the same position (fig. 21). The
positions of Manucodia and Cyanocitta are anomalous relative to the birds of paradise
in that other evidence, to be described in the next section, strongly supports a closer
relationship
of manucodines
and paradisaeinines.
If Vireo is deleted and Lanius is used as the root, a global analysis yields a monophyletic Paradisaeidae
with the corvid Cyanocitta as its sister group (fig. 2E). A transversion analysis, on the other hand, places Manucodia outside Cyanocitta (fig. 25).
On summarizing
the above, parsimony analysis of cytochrome-b
sequences provides support for the following phylogenetic
hypotheses: ( 1) the monophyly
of the
core birds of paradise, the paradisaeinines,
(2) the monophyly
of the bowerbirds, (3)
the placement
of the noncorvine
thrush Catharus outside the corvines, and (4) the
placement
of the bowerbirds
at the base of the corvines. The relationships
of the
remaining
four corvine groupsthe paradisaeid
manucodes
(Manucodia ) , corvids
not consistently
resolved.
( Cyanocitta), vireos ( Vireo), and shrikes (Lanius)-are
This is especially true when more “distant”
roots are used (e.g., Empidonax, Catharus,
and the bowerbirds; see fig. 2A-C and F-H). When these more distant taxa are deleted
from the analysis, then unrooted transversion
parsimony
networks place Cyanocitta
and Manucodia closer to the birds of paradise than are either Vireo or Lanius (fig.
21-J)) which is consistent with transversion distances (table 1)) DNA hybridization
distances ( Sibley and Ahlquist 1990)) and a combined data set of morphological
and
sequence characters (see below).
mtDNA Sequences and Morphology
The interrelationships
of the corvines were also investigated
by combining
the
1,143 bp of sequence with 14 characters from a study of cranial morphology (J. Cracraft,
unpublished
data; see Appendix). This combined data set helps resolve several cladistic
components.
As in the previous section, two parallel analyses were undertaken,
one
using global parsimony
(fig. 3A-E) and the other using transversion
parsimony (fig.
3F-J); in both, the morphological
characters were ordered (see Appendix).
With the addition of the morphological
characters, Manucodia is united with the
paradisaeinines
rather consistently in both global and transversion
parsimony analyses
(7 of 10 analyses in fig. 3). Two global analyses (fig. 3C-D) place Manucodia with
Cyanocitta and Lanius; yet, in the analysis in which ptilonorhynchids
are the root
(fig. 3C), the next most-parsimonious
tree (one step longer) unites ibfanucodia with
GLOBAL PARSIMONY
TRANSVERSION PARSIMONY
Empidonax
LVireo
Ptiloris
Epimachus
Diphyllodes
Manucodia
Cyanocitta
L.&us
Vireo
W
32
37
30+34
Ailuroedus
Ptilonorhynch~
Catharus
H
16
Vireo
93
Ailuroedus
Ptilonorhynchus
Ptiloris
+
a7
Manucodia
h
L
I
Vireo
(,,
56
Ptiloris
Vireo
Ptiloris
J
UW
Epimachus
Diphyllodes
Manucodia
76
-
122
Lanius
Cyanocitta
Manucodia
Cyanocitta
Lanius
FIG. 3.-Global (A-E) and transversion (F-J) parsimony analyses parallel to those of fig. 2 but with
the addition of 14 ordered morphological characters (see Appendix). Lengths of branches for all mostparsimonious trees are shown above branch, and percentages of 200 bootstrap replicates are shown (in
brackets) below branch; branch lengths are not shown on strict consensus trees. Statistics for trees (tree
length, consistency index for informative characters, and retention index) are as follows: A = 1167, 0.509,
and 0.349. B = 1001,0.530, and 0.372 (strict consensus of two trees); C = 892,0.552, and 0.390; D = 655,
0.592 and 0.373; E = 544, 0.644, and 0.437; F = 477, 0.455, and 0.430 (strict consensus of four trees); G
= 38;, 0.497, and 0.480; H = 332, 0.544, and 0.508; I = 222, 0.615, and 0.574; and J = 177, 0.657, and
0.588 (strict consensus of two trees).
1208
Helm-Bychowski
and Cracraft
the paradisaeinines,
and, in the analysis in which Vireo is the root (fig. 3D), a tree
two steps longer does likewise. Thus, the analyses of figure 3 lend support to the
monophyly of the Paradisaeidae. One-way DNA hybridization
values are also consistent
with a sister-group relationship between manucodines
and paradisaeinines
(Sibley and
Ahlquist 1987).
When the morphological
characters and transversion
changes are analyzed together (fig. 3F-J), the jay Cyanocitta is resolved as the sister group of the paradisaeids
(in particular,
see fig. 3G-I). In the global analyses, however, Cyanocitta is united
consistently
with the shrike Lank.
The most likely explanation
for this anomalous
cladistic pattern is the fact that both Cyanocitta and Lanius have very long branches
relative to the other taxa (especially see fig. 3C-E; in the two trees forming the consensus
tree of fig. 3B, both taxa also have long branches).
In the transversion
analyses, in
contrast, the branch lengths of the various lineages are more nearly equal, and in none
of these do Lanius and Cyanocitta come together. Instead, Cyanocitta is united with
the paradisaeids.
In summary, the addition of morphological
characters provides important evidence for resolving the relationships
among Cyanocitta, ikfanucodia, and
the paradisaeinines.
Discussion
Corvine Interrelationships:
The Importance
of Different
Kinds
of Data
Three kinds of data-mtDNA
sequences, cranial osteological, and DNA hybridization -can
be used to examine relationships
among the corvine families. None of
the three data sets provide a clear resolution of the relationships,
when used separately.
Figure 2 shows that cladistic patterns inferred from cytochrome-b sequences can change
substantially,
depending on the class of substitutions
examined and the taxa included
in the analysis. This behavior apparently results from short internodal
distances and
from the inclusion of distant outgroups that introduce noise into an analysis because
of multiple hits (homoplasy),
as well as from the tendency for long branches to attract
one another, again because of homoplasious
similarities (Penny 1988; Miyamoto and
Boyle 1989). Much the same pattern can be observed when the DNA hybridization
data are analyzed in detail. In this case short internodal distances can lead to different
cladistic results when those are produced by algorithms having only slight variation
in assumptions
(results not shown). Finally, the morphological
data alone are also
incapable of providing a satisfactory resolution, because too few characters are shared
among lineages.
Detailed analysis of the mtDNA sequences by themselves (fig. 2), and especially
when combined
with a small morphological
data set (fig. 3), supports the following
hypotheses of relationship:
( 1) monophyly
of the paradisaeinines,
(2) monophyly
of
the paradisaeids,
( 3 ) a sister-group relationship
between corvids and paradisaeids, (4)
monophyly
of the ptilonorhynchids
and their placement
as the sister group of the
corvine taxa included in this study, (5) the placement of the vireos and shrikes between
the bowerbirds
and the corvid + paradisaeid
clade, and (6) the placement
of the
thrushes outside the corvines. These conclusions are consistent with DNA hybridization
distances, although published trees based on the latter are sometimes in conflict with
one another (e.g., see Sibley and Ahlquist 1990, pp. 832-833 and 86 l-863).
The results presented in figures 2 and 3 differ in a number of respects from those
of Edwards et al. ( 199 1). In particular, the placement of the thrush Catharus within
the corvines in the latter study, rather than outside the corvines as in the present study
(also see Sibley and Ahlquist 1990)) was apparently due to cross-contamination
of a
PCR product that resulted in a composite sequence for Catharus as reported in Edwards
Corvine
Phylogeny
from Mitochondrial
Cytochrome-b
Genes
1209
et al. ( 199 1) . Other conflicts in relationships
are the result of differences in methodology. The present study investigated relationships by using sequentially smaller subsets
of the taxa and elimination
of distant outgroups. This procedure led to a seemingly
more reliable resolution
in that the results are more congruent with those of DNA
hybridization
and morphology.
A Relative
Metric
for Corvine
Phylogeny?
Previous studies have suggested a linear relationship between the age of divergence
of two taxa and the accumulation
of transversion
differences in their mtDNA (Brown
et al. 1982; deSalle et al. 1987; Miyamoto and Boyle 1989; Irwin et al. 199 1) . Absolute
ages of divergence are unknown
for corvine taxa, yet two arguments can be used to
adopt a preliminary
hypothesis that transversion
differences are roughly correlated
with time in these birds. First, there is a general observation that the pattern of transition
and transversion substitutions in corvines (table 1) is similar to that for the cytochromeb gene of mammals in which a linear relationship between transversion
differences
and ages of divergence has been proposed and is better supported by using fossil dating
(Irwin et al. 199 1). This relationship,
moreover, seems characteristic
of all the mitochondrial
genes that have been investigated (Brown et al. 1982; deSalle et al. 1987;
Miyamoto
and Boyle 1989), despite the fact that individual
genes, including cytochrome b, may sometimes evolve at different rates in different lineages (e.g., see Martin
et al. 1992; Martin and Palumbi 1993). Second, the pattern of transversion
differences
among corvines roughly corresponds
to their relative ages of divergence as inferred
from their phylogenetic
branching pattern (fig. 4), even though the number of differences, per se, was not used to establish that pattern.
Adoption of the hypothesis that age of divergence is roughly linear with transversion differences allows us to propose a preliminary
metric for some patterns of
diversification
within the corvines. A more detailed picture of corvine diversification
will come only when we better understand
the relationships
of corvine families not
included in this study.
Divergence among the three paradisaeinines
on the basis of 14-22 transversion
differences is - l%-2% (table 1). Because the three genera represent major clades
within the paradisaeinines
(J. Cracraft, unpublished
data), the transversion differences
can be interpreted to mean that most of the remarkable radiation of birds of paradise
occurred within a span of ~5 Myr (when rates of cytochrome-b
evolution comparable
to those seen in mammals
are assumed; Irwin et al. 199 1). On the other hand, a
transversion
sequence divergence between the manucodines
and paradisaeinines
of
-4.7%
(55 transversion
differences)
indicates that the family itself originated well
before the radiation of the diverse paradisaeinine
lineages, perhaps as much as 20 Mya
(a similar tempo for paradisaeid
evolution
is implied by DNA hybridization
data;
Sibley and Ahlquist 1987).
An important
finding of this study is that the corvid Cyanocitta has transversion
divergences from the paradisaeinines
essentially no different than those of Manucodia.
This implies that the large, worldwide corvid radiation has the same time frame as
does the less diverse paradisaeid radiation restricted to the Australopapuan
region. If
sequence evolution of corvine cytochrome
b is similar to that of mammals, then the
age of divergence of the corvid and paradisaeid lineages would be -20 Myr.
At approximately
67-73 transversion
differences (6%) from the paradisaeinines
are the laniids and vireonids. A similar rate of transversion
divergence in birds and
mammals would indicate that these lineages originated -24-25
Mya.
divergence from the paradisaeinines,
Finally, bowerbirds show - 7% transversion
,
12 10
Helm-Bychowski
and Cracraft
paradisaeinines
manucodines
Corvidae
Laniidae
Vireonidae
Ptilonorhynchidae
I
Turdidae
6.9-7.4%
10.4%
suboscine outgroup
FIG. 4.-Summary
hypothesis for the relationships of seven corvine higher taxa derived from analyses
shown in fig. 2-3. See text. Three, possibly four, rapid radiations (star phylogenies) are embedded within
the corvine radiation, as follows: ( 1) the basal paradisaeinine
lineages which show I%-2% transversion
differences, (2 ) the corvid-manucodine-paradisaeinine
lineages which show 4.5%-5.5% transversion differences
from the paradisaeinines,
and (3) the vireonid-laniid-higher
corvine lineages which show 5.9%-6.4% transversion differences from the paradisaeinines.
The bowerbirds lie at the base of the corvines and may have
diverged soon after the corvines diverged from other major passerine lineages.
which suggests a possible age of origin for them and the other corvines of 28 Myr.
Bowerbirds,
like paradisaeids,
are restricted to the Australopapuan
region, but the
present-day
diversity of bowerbirds is very much less. The two genera of bowerbirds
included in the present study represent the two major lineages, and they have a transversion divergence of slightly >6%. Both these lineages, therefore, apparently
have
had a more ancient history in Australopapua
than have the paradisaeids.
Star Phylogenies,
Parsimony
Analysis,
and Congruence
If one accepts the hypothesis that transversion
differences among taxa are more
or less linear to their divergence times, then the above results have identified several
instances of star phylogenies
embedded
within the corvine radiation
(fig. 4). Star
phylogenies
involve the rapid radiation of three or more lineages over a relatively
short period of time; thus, cladistic reconstructions
will generally be characterized
by
short internodes. Short internodes create difficulties for recovering phylogenetic signal,
and those difficulties will generally extend to all kinds of data and methods of phylogenetic reconstruction.
It is arguable that one of the major reasons numerous groups
of organisms have resisted resolution of their phylogenetic relationships is the existence
of short internodes.
Star phylogenies thus seem to be common.
Star phylogenies
have important
implications
for molecular
systematics.
Our
results suggest that a simple search for the most-parsimonious
or best-fit tree may be
insufficient
when attempting
to resolve cladistic signal within a star phylogeny. A
parsimony analysis of the entire data set (fig. 2A or fig. 3A), for example, yields a tree
Corvine
Phylogeny
from Mitochondrial
Cytochrome-b
Genes
12 11
strikingly different from the final estimate of the phylogeny (fig. 4). Such a result,
moreover, is independent
of method: analysis of the cytochrome-b data set with distance
methods (FITCH, KITSCH, or NEIGHBOR
JOINING,
using different corrections
for multiple hits and back mutations;
table 2) all yield cladistic patterns similar to
those of figure 2 (results not shown). This implies that a single cladistic resultwhether derived from discrete data or distances-may
not necessarily yield a satisfactory
description
of the phylogenetic
signal in the data.
Given that star phylogenies involve short internodal distances, it might be expected
that the results of statistical tests will tell us little about the strength of the phylogenetic
signal other than that we have too few data (witness the human-chimp-gorilla
problem).
It is likely, for instance, that “true” cladistic relationships will often fail to be supported
by high bootstrap values (unless one has much more data than is typically used in
molecular systematic studies), even though that cladistic component
is frequently, or
consistently,
present in various analyses. One example in this study is the sister-group
relationship
between Manucodia and the paradisaeinines
when the combined data set
(fig. 3) in which their branch generally has low bootstrap support is used (only in fig.
3E does support attain a value that might be termed “significant”).
The cladistic patterns derived from the cytochrome-b
sequences are complex,
but the relationships
that one infers from them are largely congruent with patterns
seen in the DNA hybridization
distances. Perhaps the most striking example of the
importance
of congruence
in resolving star phylogenies is the resolution of the paradisaeinine-manucodine-corvid
trichotomy.
Recognition
of a monophyletic
Paradisaeidae is not apparent in most analyses of the cytochrome-b
sequences (fig. 2) but
does emerge in a global parsimony analysis when a less distant outgroup is used (fig.
2E). This latter result, however, could not be judged to be “reliable”
without the
addition of morphological
data (fig. 3) that further corroborate
this cladistic component. The “reliability”
of paradisaeid monophyly,
moreover, is strengthened
further
by congruence
with a noncombinable
data set, DNA hybridization.
This study illustrates the power of parsimony analysis of discrete characters as a
method for revealing the phylogenetic
signal contained in sequence data. In cases of
star phylogenies,
we cannot expect that a given set of data will produce a robust
estimate of the actual pattern of relationships,
and we cannot appeal to those data
alone when evaluating the “reliability”
of the best-fit or most-parsimonious
tree. We
must turn to congruence
analysis, either by examining whether additional data corroborate the original cladistic pattern or by examining whether independent,
noncombinable data sets identify the same cladistic components
(Cracraft and Mindell 1989;
Miyamoto and Cracraft 199 1) . Indeed the whole notion of “reliability”
may be elusive,
since neither statistical analysis nor any other method would logically lead us to reject
the most-parsimonious
(best-fit) tree in favor of a less-parsimonious
result (Carpenter 1992).
Sequence Availability
Sequences have been deposited
x74251-x74261).
in EMBL and GENBANK
(accession
numbers
Acknowledgments
We wish to thank Drs. Robert Fleischer, Lloyd Kiff, Ross Crozier, and Robert
Zink; N. W. Longmore and the Queensland Museum (Brisbane); San Diego Zoological
Society; and the American Museum of Natural History for providing tissue used in
this study. Drs. Thomas Quinn, W. Kelly Thomas, Scott Edwards, and A. C. Wilson
12 12 Helm-Bychowski and Cracraft
contributed
in numerous
ways to the initial phases of our work. We are grateful to
Drs. David Minded, Robert Zink, Mike Miyamoto, and Frank Gill, as well as to Dr.
Walter Fitch and two anonymous
referees, for their helpful comments on the manuscript. Our research has been supported by the Department
of Anatomy and Cell
Biology, University
of Illinois; the American
Museum of Natural History; and by
National Science Foundation
grants BSR-8805957 and BSR-9007652.
APPENDIX
Table Al
Morphological Data Set of 14 Characters, Showing Phylogenetically
Variation for 11 Taxa
Informative
CHARACTER=
lb
TAXON
Ptiloris . .
Diphyllodes .
Epimachus
Manucodia . . .
Cyanocitta
Lanius
..
Vireo . . .
.
Ailuroedus
PtilonorhynchuJ
Catharus . .
Empidonax
.
2”
3*
4e
5f
6*
01111031131
0111103113
011110311312
01210020120
00010010110
000?0040110
0000000004
10?10150120
10?10150120
0001000015?
0001010010
7’
8’
9j
10k
11’
12”
13”
2
11
11
11
12
0
0
10
0
0
0
0
0
0
0
0
0
0
0
0
0
0
14’
1
0
0
0
0
0
0
0
NOTE.-Data
are from a larger study of corvine morphology (J. Cracraft, unpublished
data).
a All characters are ordered from primitive (0) to derived (I-2) except characters 7 and IO which have the characterstate tree (4,5,1(2(3)))0 (see Swofford 1993, p. 79). Character state “?” = lacking or inapplicable.
b Quadrate, lateral condyle. 0 = protrudes posteromedially;
and I = reduced or absent.
’ Quadrate, orbital process. 0 = expansion at distal end; and I = lacking.
d Ectethmoid. 0 = broadly fused to frontal; I = narrowly fused; and 2 = narrowly fused, braces lateral nasal bar.
’ Lacrimal. 0 = absent; and I = present.
f Lacrimal. 0 = foot not expanded: and I = foot expanded.
g Lacrimal. 0 = head not expanded; and 1 = head expanded.
h Vomer. 0 = delicate: 1 = robust, does not brace nasal septum; 2 = robust, braces nasal septum; 3 = robust, braces
nasal septum and maxillopalatines;
4 = robust, lateromedially compressed; and 5 = robust, distally simplified.
’ Maxillopalatines. 0 = large: and 1 = reduced or vestigial.
j Postorbital process. 0 = well-developed; and 1 = reduced or absent.
’ Zygomatic process and adductor musculature. 0 = process not separated from jaw articulation, musculature passes
between zygomatic and postorbital processes; I = process separated from articulation,
much of the musculature passes
under process; 2 = ventral surface of process excavated, musculature passes between zygomatic and postorbital processes;
3 = ventral surface of process excavated but process reduced in size, musculature passes between zygomatic and postorbital
processes: 4 = ventral surface of process excavated but excavation extends along dorsal margin of ear opening, musculature
passes between zygomatic and postorbital processes; and 5 = process not separated from jaw articulation, but ventral muscle
scar extends posterodorsally to ear opening.
’ Internasal septum. 0 = does not extend far posteriorly; and I = extends to interorbital septum.
m Mediopalatine
process of premaxilla. 0 = absent; I = moderately developed, not fused extensively to nasal septum;
and 2 = well-developed, fused extensively.
” Attachment for adductor musculature. 0 = upper compartment small, ventral compartment not expanded; and 1
= both compartments expanded.
o Lateral nasal bar. 0 = narrow; and 1 = broad dorsally.
LITERATURE
CITED
M., D. L. ELLSWORTH,
DNA suitable for sequencing
ALLARD,
and R. HONEYCUTT.
using the polymerase
199 1. The production of single-stranded
chain reaction. Biotechniques
10:24-26.
Corvine
Phylogeny
from Mitochondrial
Cytochrome-b
Genes
12 13
JOCK, W. J. 1963. Relationships
between the birds of paradise and the bowerbirds. Condor 65:
91-125.
JREMER, K. 1988. The limits of amino-acid sequence data in angiosperm phylogenetic reconstruction. Evolution 42:795-803.
JROWN, W. M., E. M. PRAGER, A. WANG, and A. C. WILSON. 1982. Mitochondrial
DNA
sequences of primates: tempo and mode of evolution. J. Mol. Evol. l&225-239.
CABOT, E., and A. T. BECKENBACH. 1989. Simultaneous-editing
of multiple nucleic acid sequences with ESEE. Comput. Appl. Biosci. 5:233-234.
CARPENTER, J. M. 1992. Random cladistics. Cladistics 8: 147- 153.
~EDERGREN, R., M. W. GRAY, Y. ABEL, and D. SANKOFF. 1988. The evolutionary relationships
among known life forms. J. Mol. Evol. 28:98-l 12.
JRACRAFT, J., and K. HELM-BYCHOWSKI. 199 1. Parsimony and phylogenetic inference using
DNA sequences: some methodological
strategies. Pp. 184-220 in M. M. MIYAMOTO and J.
CRACRAFT, eds. Phylogenetic
analysis of DNA sequences. Oxford University Press, New
York.
SRACRAFT, J., and D. P. MINDELL. 1989. The early history of modern birds: a comparison of
molecular and morphological
evidence. Pp. 389-403 in B. FERNHOLM, K. BREMER, and H.
JORNVALL, eds. The hierarchy of life. Elsevier, Amsterdam.
DESALLE, R., T. FREEDMAN, E. M. PRAGER, and A. C. WILSON. 1987. Tempo and mode of
sequence evolution in mitochondrial
DNA of Hawaiian Drosophila. J. Mol. Evol. 26: 157164.
DESJARDINS, P., and R. MORAIS. 1990. Sequence and gene organization of the chicken mitochondrial genome: a novel gene order in higher vertebrates. J. Mol. Biol. 212:599-634.
EDWARDS, S. V., P. ARCTANDER, and A. C. WILSON. 199 1. Mitochondrial
resolution of a deep
branch in the genealogical tree for perching birds. Proc. R. Sot. Lond. [B] 243:99- 107.
FELSENSTEIN, J. 1985. Confidence
limits on phylogenies: an approach using the bootstrap.
Evolution 39:783-79 1.
. 199 1. PHYLIP, version 3.4. University of Washington Press, Seattle.
~YLLENSTEN, U. B., and H. A. ERLICH. 1988. Generation
of single-stranded
DNA by the
polymerase chain reaction and its application to direct sequencing of the HLA-DQA locus.
Proc. Natl. Acad. Sci. USA 85:7652-7656.
HILLIS,D. M. 199 1. Discriminating
between phylogenetic
signal and random noise in DNA
sequences. Pp. 278-294 in M. M. MIYAMOTO and J. CRACRAFT, eds. Phylogenetic analysis
of DNA sequences. Oxford University Press, New York.
HOWELL, N. 1989. Evolutionary conservation of protein regions in the proton-motive
cytochrome
b and their possible roles in redox catalysis. J. Mol. Evol. 29:157-169.
IRWIN,D. M., T. D. KOCHER, and A. C. WILSON. 1991. Evolution of the cytochrome b gene
of mammals. J. Mol. Evol. 32:128-144.
JONES, T. R., A. G. KLUGE, and A. J. WOLF. 1993. When theories and methodologies
clash: a
phylogenetic reanalysis of the North American ambystomatid
salamanders (Caudata: Ambystomatidae).
Syst. Zool. 42:92- 102.
JUKES, T. H., and C. R. CANTOR. 1969. Evolution of protein molecules. Pp. 2 l- 132 in H. N.
MUNRO, ed. Mammalian
protein metabolism. Academic Press, New York.
K~~LLERSJO,M., J. S. FARRIS, A. G. KLUGE, and C. BULT. 1992. Skewness and permutation.
Cladistics 8:275-287.
KIMURA, M. 1980. A simple method for estimating evolutionary
rates of base substitutions
through comparative
studies of nucleotide sequences. J. Mol. Evol. 16: 11 l-120.
KOCHER, T. D., W. K. THOMAS, A. MEYER, S. V. EDWARDS, S. PAABO, F. X. VILLABLANCA,
and A. C. WILSON. 1989. Dynamics of mitochondrial
DNA evolution in animals: amplification and sequencing with conserved primers. Proc. Natl. Acad. Sci. USA 86:6 196-6200.
KRAUS, F., and M. M. MIYAMOTO. 1991. Rapid cladogenesis among the pecoran ruminants:
evidence from mitochondrial
DNA sequences. Syst. Zool. 40: 117- 130.
KUSUKAWA, N., T. UMEMORI, K. ASADA, and I. KATO. 1990. Rapid and reliable protocol for
12 14 Helm-Bychowski and Cracraft
direct sequencing
72.
of material
amplified
by polymerase
chain reaction.
Biotechniques
9:66-
LANSMAN,R. A., R. 0. SHADE, J. F. SHAPIRA,and J. C. AVISE. 198 1. The use of restriction
endonucleases to measure mitochondrial DNA sequence relatedness in natural populations.
J. Mol. Evol. 17:214-226.
MARTIN, A. P., G. J. P. NAYLOR, and S. R. PALUMBI. 1992. Rates of mitochondrial DNA
evolution in sharks are slow compared with mammals. Nature 357:153-155.
MARTIN, A. P., and S. R. PALUMBI. 1993. Body size, metabolic rate, generation time, and the
molecular clock. Proc. Natl. Acad. Sci. USA 90:4087-409 1.
MIYAMOTO,M. M., and S. M. BOYLE. 1989. The potential importance of mitochondrial DNA
sequence data to eutherian phylogeny. Pp. 437-450 in B. FERNHOLM,K. BREMER,and H.
JORNVALL,eds. The hierarchy of life. Elsevier, Amsterdam.
MIYAMOTO,M. M., and J. CRACRAFT. 199 1. Phylogenetic inference, DNA sequence analysis,
and the future of molecular systematics. Pp. 3-17 in M. M. MIYAMOTOand J. CRACRAFT,
eds. Phylogenetic analysis of DNA sequences. Oxford University Press, New York.
PENNY, D. 1988. What was the first living cell? Nature 331: 11 l-l 12.
QIAN, L., and M. WILKINSON. 199 1. DNA fragment purification: removal of agarose 10 minutes
after electrophoresis. Biotechniques 10:736-738.
SIBLEY,C. G., and J. AHLQUIST. 1987. The lesser melampitta is a bird of paradise. Emu 87:
63-66.
-.
1990. Phylogeny and classification of birds. Yale University Press, New Haven.
SMITH, A. B. 1989. RNA sequence data in phylogenetic reconstruction: testing the limits of its
resolution. Cladistics 5:32 l-344.
SWOFFORD, D. L. 1993. PAUP: phylogenetic analysis using parsimony, version 3.1 s. Illinois
Natural History Survey, Champaign.
SWOFFORD,D. L., and G. J. OLSEN. 1990. Phylogeny reconstruction. Pp. 4 1 l-50 1 in D. HILLIS
and C. MORITZ, eds. Molecular systematics. Sinauer, Sunderland, Mass.
WHEELER, W. C. 1990. Nucleic acid sequence phylogeny and random outgroups. Cladistics 6:
363-367.
WALTER M. FITCH, reviewing
Received
December
Accepted
July 2 1, 1993
editor
3, 1992; revision
received
July 2 1, 1993