Molecular Ecology (2003) 12, 9971006

Molecular phylogeography of the Amazonian Chagas

disease vectors Rhodnius prolixus and R. robustus

Blackwell Science, Ltd

F E R N A N D O A . M O N T E I R O ,* T O B Y V . B A R R E T T , S I N E A D F I T Z P A T R I C K ,
C E L I A C O R D O N - R O S A L E S , D O R A F E L I C I A N G E L I and C H A R L E S B . B E A R D *
*Division of Parasitic Diseases, Centers for Disease Control and Prevention, Atlanta, USA, 4770 Buford Hwy, Mail Stop F-22,
Chamblee, Atlanta, GA 303413724, USA, Instituto Nacional de Pesquisas da Amazonia, C.P. 478, Manaus, AM, 69011970, Brazil,
Pathogen Molecular Biology and Biochemistry Unit, London School of Hygiene and Tropical Medicine, London, UK, Medical
Entomology Research and Training Unit/Guatemala, CDC and Center for Health Studies, Universidad del Valle de Guatemala,
Guatemala City, Guatemala, Centro Nacional de Referencia de Flebotomos, Seccion de Entomologia Medica, Universidad de Carabobo,
Maracay, Venezuela

Abstract
The phylogeographical structure of the closely related species Rhodnius prolixus and R.
robustus is presented based on a 663-base pair (bp) fragment of the mitochondrial cytochrome b gene. Twenty haplotypes were recovered from 84 samples examined, representing 26 populations from seven Latin American countries. The resulting phylogenetic tree
is composed of five major reciprocally monophyletic clades, one representing R. prolixus
and four representing R. robustus. While R. prolixus is a very homogeneous assemblage, R.
robustus has deeper clades and is paraphyletic, with the clade comprising R. robustus from
Venezuela (Orinoco region) more closely related to the R. prolixus clade than to the other R.
robustus populations from the Amazon region. The R. robustus paraphyly was supported further by the analysis of a nuclear gene (D2 region of the 28S RNA) for a subset of specimens.
The data support the view that R. robustus represents a species complex. Levels of sequence
divergence between clades within each region are compatible with a Pleistocene origin.
Nucleotide diversity () for all R. prolixus populations was extremely low (0.0008), suggesting
that this species went through a recent bottleneck, and was subsequently dispersed by man.
Keywords: Amazonia, cytochrome b, mtDNA, phylogeography, Rhodnius, speciation
Received 30 August 2002; revision received 6 January 2003; accepted 7 January 2003

Introduction
The Triatominae (Hemiptera: Reduviidae) comprise a
subfamily of haematophagous insects that are vectors
of the Chagas disease agent, Trypanosoma cruzi. Chagas
disease (or American trypanosomiasis) is the most serious
parasitic disease of the Americas (World Bank 1993),
affecting approximately 12 million people throughout this
region, with an additional 90 million estimated to be at risk
(Schmunis 1999). Most species of triatomines are potential
vectors of T. cruzi, but only a few have become well
adapted to living in human habitations and are therefore of
Correspondence and present address: Fernando Monteiro,
Laboratrio de Doenas Parasitrias, Departamento de Medicina
Tropical, Instituto Oswaldo Cruz, Avenida Brasil 4365, Rio de
Janeiro, RJ, Brasil, 21045 900. E-mail: fermonte@globo.com
2003 Blackwell Publishing Ltd

greater epidemiological importance. One of these species

is R. prolixus, the main Chagas disease vector in Venezuela,
Colombia and Central America (Lent & Wygodzinsky
1979). R. prolixus is thought to be an exclusively domestic
species throughout most of its range, but in Venezuela
there has been some confusion due to the presence of the
closely related and morphologically similar sylvatic species
R. robustus. The latter was reported first in Venezuela by
Lent & Valderrama (1973) with bugs collected from Attalea
maracaibensis (= A. butyracea) palm trees. However, the
existence of a sylvatic R. prolixus has long been known by
Venezuelan researchers (Gamboa 1963, 1973; Gomz-Nez
1963) but its status, to date, remains unclear (Schofield
2000). The distribution of these two species overlaps in
most of northern South America (Colombia, Venezuela,
Bolivia, Ecuador, French Guiana and north Brazil) with R.
prolixus occurring further north through Central America to

998 F . A . M O N T E I R O E T A L .
southern Mexico (Lent & Wygodzinsky 1979). Considerable
overlap of key morphological characteristics used to separate
these species (Hurtado-Guerrero 1992; Harry 1993a; Harry
1994), however, has led to occasional misidentifications,
thus making the precise limits of their distributions difficult to establish (Solano et al. 1996).
R. prolixus and R. robustus are two of the four members
of the R. prolixus species group that also includes R. neglectus
and R. nasutus, from drier habitats in central and northeastern Brazil, respectively. Although it is relatively easy to
distinguish R. neglectus and R. nasutus from each other and
from the rest of the group based on morphology (Harry
1994; Dujardin et al. 1999), the separation of R. prolixus
from R. robustus can be problematic (Harry 1993a). This
observation, together with the lack of diagnostic allozyme
loci between these two species (Harry et al. 1992), led to the
suggestion that R. robustus was not a valid taxon (Harry
1993b; Barrett 1995). It was shown later, based on two mitochondrial DNA (mtDNA) fragments (cytochrome b and
large subunit ribosomal RNA), that these two species
belonged to different evolutionary lineages and should
thus be regarded as separate (Lyman et al. 1999). The analysis of additional samples from the Amazon region, however, uncovered a more interesting pattern, showing that
R. robustus is quite heterogeneous (Monteiro et al. 2000)
and could conceal more than one incipient species. In addition, the relationship between R. prolixus and R. robustus in
the Venezuelan Orinoco basin remains unclear.
Therefore, we present here a comprehensive mtDNA
phylogeographical analysis of this controversial pair of
species, which includes samples representing 25 populations from seven Latin American countries. Samples of R.
robustus include specimens from geographical locations
very close to the type localities for this species. [R. robustus
was described by Larrousse in 1927 based on two female
specimens from two different geographical locations.
Curiously, Larrousse did not designate a holotype (or
type locality) for the new species. See Discussion for more
information.] Results are discussed in terms of taxonomy,
biogeography and epidemiological significance.

Materials and methods

Samples and gene fragment used
A total of 84 specimens representing 12 populations of R.
prolixus and 14 populations of R. robustus were sequenced
for a 663-base pair (bp) fragment of the mitochondrial
cytochrome b (cyt b) gene. The closely related R. nasutus
was used as the outgroup. Detailed information on the
samples used in this study is given in Table 1.
Insects were identified on the basis of size and colour differences on the hind tibia of later stage nymphs (Lent &
Wygodzinsky 1979).

DNA extraction, amplification and sequencing

Genomic DNA was extracted from individual bug legs as
described in Lyman et al. (1999). Standard polymerase chain
reaction (PCR) techniques were used to amplify the gene
fragments using the following primers: CYTB7432F, 5GGACG(AT)GG(AT)ATTTATTATGGATC, and CYTB7433R,
5-GC(AT)CCAATTCA(AG)GTTA(AG)TAA. Primer design
was based on conserved regions of the cyt b gene of
Triatoma dimidiata (Dotson & Beard 2001) after comparison
with other insect cyt b sequences. Amplified PCR fragments were sequenced, edited and aligned as described
in Monteiro et al. (2000).
In order to validate the results based on the mitochondrial cyt b fragments, and to exclude the possibility of
mtDNA introgression between the R. prolixus and R. robustus lineages, we also present an alignment of the D2 variable region of the 28S RNA (D2) for a randomly selected
subset of samples which represent the five main cyt b
clades observed (and the outgroup). Four D2 sequences
were already available (Monteiro et al. 2000), and three
others were sequenced here with the same primers, D2F,
5-GCGAGTCGTGTTGCTTGATAGTGCAG and D2R, 5TTGGTCCGTGTTTCAAGACGGG (Porter & Collins 1996).

Phylogenetic analysis
Sequences were subjected to parsimony and distance
analyses. When more than one individual yielded the same
sequence, only a single haplotype was included in the analysis. However, for illustrative purposes, same haplotypes
from different populations were incorporated to the tips of
the tree a posteriori. Maximum parsimony (MP) trees were
constructed using paup* version 4.0b8 (Swofford 1999).
The shortest trees were found via a heuristic search with
stepwise addition of taxa, using 100 random input orders
and tree bisectionreconnection (TBR) branch swapping.
We examined the effects of several different character
weighting schemes on tree topology: equal weight across
sites, applying a 1:15 transition/transversion ratio (TS:TV),
and giving first, second and third codon positions weights
of 4-8-1, respectively. In all cases characters were considered unordered. The neighbour-joining (NJ; Saitou &
Nei 1987) algorithm was used to construct a tree based on
a Kimura 2-parameter (K2-p, Kimura 1980) distance matrix.
The K2-p method was used to produce the distance matrix
because the Jukes & Cantor (1969) estimate of the number
of nucleotide substitutions per site between haplotypes
was smaller than 0.3, and the TS:TV ratio was higher than
2 (Nei 1996). Statistical support for clades in the MP and NJ
phylogenetic trees was assessed by the bootstrap method
(Felsenstein 1985) with 1000 replications. For MP bootstrap
analysis, simple stepwise addition of taxa and TBR branch
swapping options were used. Basic statistics from aligned
2003 Blackwell Publishing Ltd, Molecular Ecology, 12, 9971006

P H Y L O G E O G R A P H Y O F R H O D N I U S P R O L I X U S / R . R O B U S T U S 999
Table 1 List of samples used in the study

Species

Collection site

Haplotype

Field/
colony

Domestic/
sylvatic

Code

Date collected

R. prolixus

Orica, Francisco Morazan, Honduras

Las Palmas, Guatemala
Tituque, Guatemala
Tuticopote, Guatemala
Modesto Loaiza, Coyaima, Colombia
Ibague, Colombia
Pampanito, Trujillo, Venezuela
Pampan, Trujillo, Venezuela
Pampanito, Trujillo, Venezuela
San Jos Tiznados, Gurico, Venezuela
Ortiz, Gurico, Venezuela*
Cojedes, Venezuela

7
6
9
3
1
1
3
1
2
3
4
1

a
b
b
b
b
b
c
c
c
b
b
b

F
F
F
F
C
C
C
C
C
C
F
C

D
D
D
D
D
D
D
D
D
S
S
D

prHO
prGU 1
prGU 2
prGU 3
prCO 1
prCO 2
prVE 1
prVE 2
prVE 3
prVE 4
prVE 5
prVE 6

1999
June 1995
June 1995
June 1995
September 1996
February 1995
1997
1987
1960
1988
July 2001
1995

R. robustus

Pampanito, Trujillo, Venezuela

Candelaria, Trujillo, Venezuela
Napo, Ecuador
Carauar, Amazonas, Brazil
Porto Velho, Rondonia, Brazil
Apu, Amazonas, Brazil
Itupiranga, Par, Brazil
Purupur, Amazonas, Brazil
Novo Repartimento, Par, Brazil
Barcarena, Par, Brazil
Cayenne, French Guiana
Balbina, Amazonas, Brazil
UHE Paredo, Roraima, Brazil
Rio Mapuera, Par, Brazil

4
3
2
4
1
4
1
4
5
5
1
1
5
3

d
e,f
g
h,i,j
k
l,m
n
n
o
p
g
r
s
t

C
C
C
F
C
C
C
C
C
C
F
C
C
C

S
S
S
S
S
S
S
S
S
S
S
S
S
S

roVE 1
roVE 2
roEC
roBR 1
roBR 2
roBR 3
roBR 4
roBR 5
roBR 6
roBR 7
roFR
roBR 8
roBR 9
roBR 10

1997
1988

February 2000
1985
September 1996
1984
December 1995
August 1998
1996
March 2000
November 1983
March 1987
June 1986

R. nasutus

Teresina, Piau, Brazil

naBR

*Three of the four samples were sequenced for 415 bp only.

sequences, K2-p distances between haplotypes, and nucleotide diversity (: equation 10.5 in Nei 1987) were computed
using mega (Kumar et al. 1994).

Tituque where from two and six bugs were sampled, in

order to detect within-house polymorphism. Samples from
Orica, Honduras represented a pool from several houses.

Genetic variation in R. prolixus

Results

In order to assess the levels of genetic diversity in natural

populations of R. prolixus, we compared field-collected
samples from three villages in Guatemala Tituque,
Tuticopote and Las Palmas with samples from Orica,
a village in the neighbouring Central American country
Honduras. These areas have not been treated previously
with insecticides for Chagas disease control. Tituque,
Tuticopote and Las Palmas are small villages (approximately
250 small adobe houses each) in a mountainous area in east
Guatemala. While Tituque is separated from Tuticopote by
a valley of 1 km in breadth, these two villages are 4 km apart
from Las Palmas. The three villages are approximately
250 km from Orica, in Honduras. Three houses were sampled
from each Tituque and Tuticopote while six houses were
sampled from Las Palmas. Sampled houses were separated
by at least 200 meters from each other. One bug was
analysed per house, with the exception of two houses in
2003 Blackwell Publishing Ltd, Molecular Ecology, 12, 9971006

Sequence variation and phylogenetic analysis

Analysis of the 86 sequences (including the outgroup)
for the cyt b gene fragments revealed the existence of
21 unique haplotypes (Table 2). Phylogenetic analyses
of these haplotypes yielded 111 variable characters, of
which 40 were autapomorphies and 71 were parsimony
informative. MP analysis of these characters produced four
equally parsimonious trees of 150 steps (CI = 0.753, RI =
0.888, RC = 0.669), which differed only in some fine-scale
arrangement of terminal taxa within otherwise stable
clades. No significant change was observed with the use of
alternative weighing schemes. One of the four MP trees
was recovered using the NJ method with Kimura 2parameters distances (Fig. 1). The saturation of transitions
was not a concern because all uncorrected (p, Nei 1987)
pairwise distances were below the 910% saturation

1000 F . A . M O N T E I R O E T A L .
Table 2 Polymorphic sites observed in 20 R. prolixus and R. robustus haplotypes. Hapotypes af are from the Orinoco basin, haplotypes g
t are from the Amazon region. Sites 171, 222 and 423 are transversions

Haplotype
a
b
c
d
e
f
g
h
i
j
k
l
m
n
o
p
q
r
s
t
Consensus

0000000000011111111111111222222222222223333333344444444444444455555555566666666666
0112456667800122334566779223344447778891245679900112223455667812334466902223333445
3388840138759119087425148252535890298942751626925140365769581302146948401271469451
* *
*
** ** *
*
* *
*
GAG....C.A.TC.........C....A.T.....TCCACC.AT...G.T.CC..GC..CGTAT...TT.T...........
G.G....C.A.TC.........C....A.T.....TCCACC.AT...G.T.CC..GC..CGTAT...TT.T...........
G.G....C.A.TC.........C....A.T.....TCCACC.AT.A.G.T.CC..GC..CGTAT...TT.T...........
G.G.CCGCGA.T...T....G.C..A...T...G.T..ACC.ATG..G.T.CCG..C.CC.TATGG.TT.T..A....T...
G.G.CCGCGA.T...T....G.C..A...T...G.T..ACC.ATG..G.T.CCG.GC.CC.TATGG.TT.T..A....T...
G.G.CCGCGA.T...T....G.C..A...T...G.T..ACC.ATG..G.T.CCG.GCTCC.TATGG.TT.T..A....T...
...............TC.CCG.....T.........C.............................A...TTC.......TG
...............TCCCCG.....T.........C.............................A....TC.......TG
...............TC..C......T.........CC............................A....TC.......TG
...............TC..CG.....TA........CC............................A....TC.......TG
...............TC..CGT....T.........CC..........G.G....................TC...C.T.T.
.............G.TC..CGT.C..T..............A........................A....TC.......T.
...............TC..CGT.C..T.......................................A....TC.......T.
...A.....AG.............T.....C.T...C.........G..T.....G.....T...........AC....T..
...A.....AG.............T.....C.T.............G..T.....G.....T...........AC.......
........................T......CT...C.........G...G...T..................A.G..TT..
..............A.........T.......T...C.............G...T..................A.G.GT...
........................T.......T.............G.......T..................A.G.GTTT.
.......................CT...A...T.T...........G.......T..............A...A.G.GTT..
........................T.......T.............G.......T......TA..........A.G.GTTT.
AGAGTTATAGACTTGCTTTTACATCCCGGCTTCACCTTGTTGGCAGAAACATAACATCTTACGCAAGCCGCCTGTATACCCA

*Nonsilent substitutions. Sequences representing the variation observed (haplotypes a, d, g, n and p), have been submitted to Genebank,
Accession nos: AF421339, AF421340, AF421341, AF421342, and AF421343.

Fig. 1 Maximum parsimony tree (one of four

shortest trees with 150 steps and CI = 0.753)
of 21 unique 663 bp cyt b haplotypes. Note
that while R. prolixus is monophyletic and
homogeneous, R. robustus is a paraphyletic
assemblage of four different lineages from
two different regions. *The Amazon region
includes the samples from the Amazon and
Tocantins basins, and French Guiana. Several
terminal branches represent more than one
individual sharing the same haplotype.

2003 Blackwell Publishing Ltd, Molecular Ecology, 12, 9971006

P H Y L O G E O G R A P H Y O F R H O D N I U S P R O L I X U S / R . R O B U S T U S 1001
threshold for the cyt b gene (7.8% within the ingroup, and
9.7% with inclusion of the outgroup; Meyer 1994; Griffith
1997). As expected for a protein-coding gene, third codon
positions were the most variable (75.0%) followed by
first (19.6%) and second (5.4%). Estimated nucleotide frequencies were A, 0.313; C, 0.213; G, 0.138; and T, 0.336.

Tree topology and sequence divergence

The tree is composed of five major clades, one representing
R. prolixus and four representing R. robustus. While R. prolixus
is a very homogeneous and monophyletic assemblage, R.
robustus has deeper clades and is paraphyletic, since the
clade composed of the R. robustus from Venezuela (or the
Orinoco region) is related more closely to the R. prolixus
clade than to the other R. robustus clades from the Amazon
region (Figs 1 and 2). The tree is well supported with the
majority of bootstrap values being 99 or higher. Sequence
divergence within clades is usually small (maximum K-2p
corrected = 1.5%); whereas, between clades in each region is
on average 6 times higher. The sequence divergence of haplotypes between regions can be as high as 8.5% (Table 3).
The alignment of the seven 630 bp nuclear D2 sequences
obtained for samples naBR, roBR4, roEC, roBR8, roVE2,
prCO1 and prVE5, is presented in Table 4. These samples
are a random selection of members of the five main clades
seen in the cyt b tree (Fig. 1), plus the outgroup (naBR). The
ingroup shares six synapomorphies (four substitutions at
positions 6, 97, 274, and 429 and two insertions: 189190
and 195204). However, the most important observation is
that R. prolixus (prCO1 and prVE5) and R. robustus from
the Orinoco region (roVE2) share a derived C in position
360, therefore further supporting the R. robustus paraphyly
observed in the cyt b tree.

Fig. 2 Geographic location of the five monophyletic cytochrome b

haplotype clades of R. prolixus (dotted line) and R robustus IIV.
Numbers 1 and 2 indicate collection sites of R. robustus samples
that are very close to the type localities given by Larrousse (1927)
in the description of the species: mouth of the Tef river, in Brazil,
and French Guiana, respectively. Levels of sequence divergence (Kimura 2-parameter) between clades within the Amazon
and Orinoco regions are indicated. It should be noted, however,
that R. prolixus has never been recorded from Panama or southern
Costa Rica.

Comparisons

Sequence
divergence (%)

Estimated
separation (Myr)

1. Within clades
R. prolixus
R. robustus I
R. robustus II
R. robustus III
R. robustus IV

0.2 (0.2 0.3)

0.2 (0.2 0.3)
1.0 (0.3 1.5)
0.2 (0.2 0.2)
1.0 (0.3 1.4)

0.4

0.4

3.3 (3.0 3.3)

1.4

3. Between clades in Amazon region

R. robustus II vs. R. robustus III
R. robustus II vs. R. robustus IV
R. robustus III vs. R. robustus IV

4.0 (3.6 4.4)

3.4 (3.0 3.9)
2.3 (2.0 2.8)

1.7
1.5
1.0

Between 2 and 3

7.2 (5.6 8.5)

3.1

2. Between R. prolixus and R. robustus

I from Orinoco region

2003 Blackwell Publishing Ltd, Molecular Ecology, 12, 9971006

Table 3 Mean cytochrome b sequence

divergence levels (Kimura 2-parameter)
for within and between clade comparisons
(range in parenthesis) and estimated time
of separation from common ancestor.
Separation time is estimated assuming a
rate of 2.3% pairwise sequence divergence
per million years (Myr). Note how all
between clade comparisons in both
regions (numbers 2 and 3) give estimates
that fall within the Pleistocene (before 2
My ago). Some estimates are not given
because of insufficient sampling or
possible lack of resolution

1002 F . A . M O N T E I R O E T A L .
Table 4 Alignment of 450 bp fragments (of the 630 bp sequenced) showing the variation observed in the D2 region of the 28S RNA for a
random selection of members of the five main clades seen in the cyt b tree, plus the outgroup (naBR). Note how the derived C in position
360 is shared by R. prolixus (prCO1 and pr VE5), and R. robustus from the Orinoco region (roVE2), further indicating that R. robustus is
paraphyletic. GenBank Accession nos AF435856 AF435862

naBR
roBR4
roEC
roBR8
roVE2
prVE5
prCO1
Consensus
naBR
roBR4
roEC
roBR8
roVE2
prVE5
prCO1
Consensus
naBR
roBR4
roEC
roBR8
roVE2
prVE5
prCO1
Consensus
naBR
roBR4
roEC
roBR8
roVE2
prVE5
prCO1
Consensus
naBR
roBR4
roEC
roBR8
roVE2
prVE5
prCO1
Consensus

1
..........
.....C....
.....C....
.....C....
.....C....
.....C....
.....C....
TTGCTTAACT
91
..........
......T...
......T...
......T...
......T...
......T...
......T...
CAAGGGCAAT
181
........-..........
..........
..........
..........
..........
..........
AAGTTATACC
271
..........
...A......
...A......
...A.A....
...A......
...A......
...A......
CACTTGAAAT
361
..........
..........
..........
..........
..........
..........
..........
AATTCGGATT

..........
..........
..........
..........
..........
..........
..........
TTTAAATGAT

..........
..........
..........
..........
..........
..........
..........
TTGAGATGGC

..........
..........
..........
..........
..........
..........
..........
CTCTCGCCCT

..........
..........
..........
..........
..........
..........
..........
ATTCAGTGTA

..........
..........
..........
..........
..........
..........
..........
ACAGCTGTGA

..........
..........
..........
..........
..........
..........
..........
TAGTGGGTTT

..........
..........
..........
..........
..........
..........
..........
GGTCGCTCTC

..........
..........
..........
..........
..........
..........
..........
GGTGGACCGC

..........
..........
..........
..........
..........
..........
..........
ACTTCTCCCT

..........
..........
..........
..........
..........
..........
..........
TAGTAGGACG

..........
..........
..........
..........
..........
..........
..........
TTGTGACCTG

..........
..........
..........
..........
..........
..........
..........
TCAATAAATA

..........
..........
..........
..........
..........
..........
..........
TTCTAAGTAT

..........
..........
..........
..........
..........
..........
..........
TTGGCTATTA

....-----..........
..........
..........
..........
..........
..........
GTTAAGGTAT

----......
..........
..........
..........
..........
..........
..........
TTTCTTTAAA

..........
..........
..........
..........
..........
..........
..........
ACAGTTTTAG

..........
..........
..........
..........
..........
..........
..........
CCGTTTTATA

..........
..........
..........
..........
..........
..........
..........
TACTGGATAA

..........
..........
..........
..........
..........
..........
..........
AATTGACAGT

..........
..........
..........
..........
..........
..........
..........
AACGAATTAT

..........
..........
..........
..........
..........
..........
..........
TATATATATG

..........
..........
..........
..........
..........
..........
..........
TAAAAATATA

..........
..........
..........
..........
..........
..........
..........
TATAATGGAA

..........
..........
..........
..........
..........
..........
..........
AGTGTCCTAA

..........
..........
..........
..........
..........
..........
..........
AATATGGCTG

..........
..........
..........
..........
..........
..........
..........
TTTGCAAGTG

..........
..........
..........
..........
..........
..........
..........
GGTTGGTAAA

..........
..........
..........
..........
..........
..........
..........
TTTAACCGGT

..........
..........
..........
..........
..........
..........
..........
TAACTATTCC

..........
..........
..........
..........
..........
..........
..........
GCCTACTGTT

..........
..........
..........
..........
..........
..........
..........
GGTAAACTGT

..........
..........
..........
..........
..........
..........
..........
TCCTAGGACT

..........
........T.
........T.
........T.
........T.
........T.
........T.
GTGCTTATAA

..........
..........
..........
..........
..........
..........
..........
TCACCGGTCG

Genetic variation in R. prolixus

The 18 field-collected R. prolixus samples from Tituque,
Tuticopote and Las Palmas, Guatemala were surprisingly
similar, presenting a single haplotype. A single haplotype
was also detected in the seven samples from Orica,
Honduras, which differed from the former by a nonsilent
transversion. Therefore, nucleotide diversity () was
very low: zero for populations of each country, and

90
..........
....G.....
....G.....
..........
....G.....
....G.....
....G.....
ATTTAAATAG
180
..........
........T.
..........
..........
..........
..........
..........
TGTCTGTTCT
270
..........
..........
..........
..........
..........
..........
..........
GGTGTTGAGC
360
..........
..........
..........
..........
.........C
.........C
.........C
AAATAGTTTT
450
..........
..........
..........
..........
..........
..........
..........
GCAGCGATTC

0.0006 when these were combined. Colony samples

derived from six populations in Venezuela and two
populations in Colombia revealed the same lack of
variation, and were remarkably similar to the haplotypes
found in Guatemala and Honduras. Their inclusion in
the analysis (one haplotype of each population) only
slightly increased (0.0008). In summary, only three
haplotypes were detected for all R. prolixus samples
analysed.
2003 Blackwell Publishing Ltd, Molecular Ecology, 12, 9971006

P H Y L O G E O G R A P H Y O F R H O D N I U S P R O L I X U S / R . R O B U S T U S 1003

Discussion
This study supports the idea that the closely related R.
prolixus and R. robustus are separate taxa, but at the same
time reveals that R. robustus is a paraphyletic species
complex. Surprisingly, R. robustus I (from the Orinoco
basin) is related more closely to R. prolixus than to the other
R. robustus (IIIV) from the Amazon region, based on both
mitochondrial and nuclear gene fragments (Figs 1 and 2,
Table 4). The phylogeography of these two species reveals
strong geographical structuring, with R. prolixus forming a
northern genetically depauperate clade, while R. robustus
is an artificial assemblage comprised of a clade from the
Orinoco basin plus three reciprocally monophyletic and
parapatric Amazonian clades.
R. prolixus was described first from rural houses in the region
of La Guayra, Venezuela by Stl (1859). Following Carlos
Chagas discoveries in Brazil (Chagas 1909) it was implicated
as a major vector of Chagas disease to humans, and reported
progressively from various localities in Venezuela, Colombia
and some parts of Central America and southern Mexico
(Dias 1952). Throughout most of its distribution it appears
to be an exclusively domestic species, and shows very little
variation either by mtDNA sequences (Fig. 1, Table 2), or
allozymes (Chavez et al. 1999; Dujardin et al. 1999).
By contrast, R. robustus was described by Larrousse in
1927, based on two female specimens considered larger
and darker than a reference series of domestic R. prolixus.
One of these originated from the region of Cayenne,
French Guiana, and the other from the mouth of the Tef
river, in the Brazilian Amazon (Fig. 1, nos 1 and 2),
although neither was formally designated as holotype. In
spite of this poor description, the validity of the taxon was
acknowledged by Lent & Jurberg (1969) and Lent &
Wygodzinsky (1979) on the basis of colour differences on
the hind tibia of later stage nymphs, and slight differences
in the shape of the basal plate struts of the male genitalia.
R. robustus samples from the Amazon region have been
shown recently to be genetically different from R. prolixus
(Lyman et al. 1999; Monteiro et al. 2000); however, the relationship of these taxa in Venezuela is controversial.
In that country, R. prolixus was thought to be an exclusively domestic species until 1961, after several unsuccessful attempts to detect its sylvatic foci. However, it was
shown later to occur in palm trees in a tropical semihumid
environment in Gurico State (Gamboa 1963). Studies
comparing R. prolixus from palms and from huts revealed
some differences between the two, the most distinctive one
being related to the larger size of the domestic insects
(Gomz-Nez 1963; Gamboa 1973). These observations
introduced the idea of the possible existence of two fully
fertile (Gomz-Nez 1963; Gamboa 1973), albeit ecologically different, forms of R. prolixus: a domestic form in huts
and a sylvatic form in palm trees. Lent & Valderrama
2003 Blackwell Publishing Ltd, Molecular Ecology, 12, 9971006

(1973) later recognized the existence of the morphologically similar, and essentially sylvatic, R. robustus in
Venezuela. However, the absence of diagnostic allozyme
loci between R. prolixus and R. robustus from that country
led to the idea that R. robustus was not a valid taxon (Harry
1993b). Lack of allozyme differences is not per se proof
of cospecificity, and high allozyme similarity in spite of
significant divergence of mitochondrial genes has been
observed in other arthropods (e.g. Langor & Sperling 1997;
Gusmao et al. 2000), indicating that allozymes might sometimes be too conserved to detect recent speciation events,
particularly when levels of gene variation are low.
According to our data, the uniqueness of the R. robustus
I haplotypes and the condition of sympatry between these
two taxa in Pampanito (3.3% sequence divergence) argue
in favour of the idea that R. prolixus and R. robustus I are
indeed separate taxa in Venezuela. Moreover, the results
agree with cross-mating experiments between prVE4 and
roVE2, which show a decrease in fecundity (GalndezGirn et al. 1994), and with wing shape differences between
these two species recently detected by geometric morphometrics (Villegas et al. 2002). Therefore our results, added
to previous observations, favour the following scenario for
the distribution of R. prolixus and R. robustus I, in Venezuela:
R. robustus I is sylvatic and found in palm trees, whereas
R. prolixus is primarily domestic but can also be found in
the sylvatic habitat, as indicated by the samples from San
Jos Tiznados and Ortiz, in Gurico state (see origin of
Venezuelan samples in Table 1).

The taxonomic status of R. robustus

We analysed field-collected samples from sites that are
very close to where the two specimens used by Larrousse,
to describe R. robustus, originated. If the samples we
analysed from the roBR1 and roFR geographic regions
(Fig. 1, nos 1 and 2) are representative of Larrousses
original specimens, it becomes clear that the original type
specimens belong to different evolutionary lineages.
R. robustus clades II and IV differ on average by 3.4% of
sequence divergence, which is about the same level of
differentiation seen between sympatric R. prolixus and R.
robustus I, in Venezuela. However, even though the three
Amazonian R. robustus clades are reciprocally monophyletic
and present regional coherence, at present they lack
diagnosable characters other than the molecular ones that
define them, and thus do not meet all three criteria of the
operational procedure used by da Silva & Patton (1998) for
identifying allopatric species forms in Amazonian mammals.
Nonetheless, it has been shown that fertility between
members of the three R. robustus clades from the Amazon
region (roBR2, roBR4 and roBR8) is greatly reduced in
cross-mating experiments (Barrett 1995), which seems to
indicate the existence of some form of barrier to gene flow.

1004 F . A . M O N T E I R O E T A L .
Similar observations have been made for some Heliconius
erato butterfly races in northern South America, which
presented comparable levels of sequence divergence (Brower
1996). Therefore, we believe that Larrousses description of
R. robustus was based on insects that belong to two of the
(possibly) four different cryptic R. robustus taxa.
The delineation of the different prolixus/robustus evolutionary units will now facilitate the search for distinguishable
characters to allow for their morphological identification.
To resolve the R. robustus paraphyly, we suggest that
groups II, III and IV of R. robustus from the Amazon region
keep their present name, according to the law of priority
(but still acknowledging the existence of the three subgroups), while giving R. robustus I a new name. However,
we strongly recommend any nomenclatural change to take
place only after a thorough morphological and morphometric characterization has been completed for all four R.
robustus lineages revealed in this study.

Biogeography, times of divergence and speciation

The traditional view that diversification of the Amazonian
biota was caused by glaciation cycles during the
Pleistocene was introduced by Haffer (1969), based on
distributional data for birds, and became known as the
refugium theory. The theory attempts to explain the latest
of the series of differentiation events beginning in the
Cenozoic that contributed to the development of the
modern biota of the Amazon basin (Simpson & Haffer
1978). In short, it is based on the premise that climatic
changes during the Pleistocene caused rain forests to
contract into isolated pockets separated by savannah. This
would have confined small populations and favoured their
divergence by genetic drift, which would have facilitated
allopatric speciation. The increasing availability of molecular data has made it possible to test whether sister taxa
from the Amazon (as well as in other tropical rain forests) are actually compatible with a Pleistocene origin
(Moritz et al. 2000). Even though the refugium theory
seems to account adequately for the pattern seen in some
species of neotropical bats (Ditchfield 2000), it fails to
explain the diversification seen in several other Amazonian
mammals such as didelphid marsupials, echimyd rodents
or pithecine monkeys, all of which appear to predate the
Pleistocene (da Silva & Patton 1998; Boubli & Ditchfield
2000). These estimates of the time of divergence between
lineages are derived from rates of sequence divergence
that are calibrated generally based on known vicariant
events. Such rates have been calculated either for the entire
mitochondrial genome for recently diverged arthropod
taxa (within the last 3 Myr, 2.3% per Myr; Brower 1994), or
for the cytochrome oxidase I gene (2.4 2.6% per Myr) for
beetles (Caccone & Sbordoni 2001). Assuming that such
values are applicable to triatomines (and to the cyt b

fragment we used), even if the slightly slower rate of 2.3%

of sequence divergence per million years is used, all
estimates between the clades within both Amazon and
Orinoco regions are compatible with a Pleistocene origin
(Table 3). Because such a pattern of phylogenetic discontinuity with geographical orientation of haplotypes most
probably involves longterm biogeographical barriers to
gene flow (Avise et al. 1987), it could well be explained by
the refugium theory.
Group I of R. robustus, from the Orinoco region of Venezuela, forms a major clade with R. prolixus, while groups II,
III and IV form a second major clade representing those
from the Amazon forest region (Fig. 2). The vicariant event
that would have separated the ancestors of these two major
clades is older (2.43.7 My) and dates within the Pliocene.

Hypothesis on the origin and spread of R. prolixus

Our data indicate that R. prolixus probably originated at
around 1.4 million years ago in the Orinoco lowland
forests, when an ancestral prolixus/robustus I stock was
separated in different refugia, giving rise to R. prolixus and
R. robustus I. Both species lived in association with birds
and mammals, especially those that nest on palm trees.
They encountered man some time after his arrival in South
America and R. prolixus developed the ability to colonize
human habitations. It was probably very recently, after
European colonization, that R. prolixus became an increasingly
domestic species (Schofield & Dujardin 1999). Demographic
growth and increased mobility of human populations
caused its dissemination to regions ecologically different
from its putative centre of origin (Dujardin et al. 2000).
Thus, for most of its present distribution, it seems unable to
move back from huts and colonize the new sylvatic
environments. The fact that all domestic R. prolixus analysed
in this study are genetically depauperate (overall = 0.0008)
seems to be an indication of a recent bottleneck. However,
whether this bottleneck is a reflection of the adaptation to
domesticity remains to be demonstrated by the analysis of
more field-collected sylvatic R. prolixus populations.

Implications for vector-control

Rhodnius prolixus is the primary Chagas disease vector in
Venezuela, Colombia and parts of Central America. The
main reason why it is such a good vector is because it is
essentially domestic throughout most of its range, for
reasons discussed above. Thus, once a village is treated
with insecticide and becomes triatomine-free, there will be
no great risk of reinvasion of treated premises from sylvatic
foci. Hence, it is believed that it should be a feasible target
for eradication in much the same way as with the domestic
forms of Triatoma infestans in the Southern Cone region in
South America (Schofield & Dujardin 1997). There is no
2003 Blackwell Publishing Ltd, Molecular Ecology, 12, 9971006

P H Y L O G E O G R A P H Y O F R H O D N I U S P R O L I X U S / R . R O B U S T U S 1005
doubt that in the past some of the sylvatic R. prolixus
populations in Venezuela were misidentified R. robustus I,
and that reports of sylvatic R. prolixus from the Amazon
region were, in the same way, misidentifications of R.
robustus IIIV (Monteiro et al. 2000). Moreover, it is the
most probable explanation for the observation in Venezuela
of huts long inhabited by men and triatomine-free, although
surrounded by R. prolixus infested palms (Gomz-Nez
1963). However, it should be considered that at least in
areas in Venezuela where true sylvatic R. prolixus populations
seem to occur (as prVE4 and 5, from Gurico) recolonization
of domiciles by sylvatic insect populations might be a
concern. On the other hand, all four R. robustus clades appear
to represent entirely sylvatic species, and it is not clear why
these populations have been unable to make the transition
to domestic environments as R. prolixus. Although there
are no reports, to date, of R. robustus colonizing houses, it
can be found occasionally in human habitations, where it
flies in from neighbouring palms attracted by light (Tonn
et al. 1976; Naiff et al. 1998; Feliciangeli et al. 2002). The
epidemiological significance of these accounts is negligible
in comparison with disease transmission mediated by R.
prolixus, but localized cases are likely to occur in some
areas (Miles et al. 1983; Feliciangeli et al. 2002).

Acknowledgements
We thank J.P. Dujardin, I. Galndez-Girn and T. Lehmann for
helpful discussion and comments. The authors are also grateful to C. Aznar, I. Galndez-Girn, J. Jurberg, R. Carcavallo,
J.S. Patterson, M. Yeo, M.A. Miles, S.A.S. Valente, C. Ponce and
C.J. Schofield, for kindly providing specimens. We also thank
B. Holloway and the staff of the NCID Biotechnology Core Facility
for synthesis of the oligonucleotide primers. This work benefited from international cooperation through the ECLAT research
network, and from the Pilot Programme for Protection of Brazilian
Rainforests/PPD G-7. The use of trade names does not constitute
endorsement by the US Public Health Service or the Centers for
Disease Control and Prevention.

References
Avise JC, Arnold J, Ball RM et al. (1987) Intraspecific phylogeography: the mitochondrial DNA bridge between population
genetics and systematics. Annual Review in Ecology and Systematics,
18, 489522.
Barrett TV (1995) Species interfertility and crossing experiments in
Triatomine systematics. In: Proceedings of the International Workshop on Population Genetics and Control of Triatominae (eds Schofield
CJ, Dujardin JP, Jurberg J), pp. 7277. INDRE Press, Mexico.
Boubli JP, Ditchfield AD (2000) The time of divergence between
the two species of Uacari monkeys: Cacajao calvus and Cacajao
melanocefalus. Folia Primatologica, 71, 387391.
Brower AVZ (1994) Rapid morphological radiation and convergence in the butterfly, Heliconius erato, inferred from patterns
of mitochondrial DNA evolution. Proceedings of the National
Academy of Sciences USA, 91, 6491 6495.
2003 Blackwell Publishing Ltd, Molecular Ecology, 12, 9971006

Brower AVZ (1996) Parallel race formation and the evolution of

mimicry Heliconius butterflies: a phylogenetic hypothesis from
mitochondrial DNA sequences. Evolution, 50, 195221.
Caccone A, Sbordoni V (2001) Molecular biogeography of cave
life: a study using mitochondrial DNA from bathysciine beetles.
Evolution, 55, 122130.
Chagas C (1909) Nova tripanozomiaze humana. Estudos sobre
a morfolojia e o ciclo evolutivo de Schizotrypanum cruzi n.
General n. sp., agente etiolojico de nova entidade morbida
do homem. Memrias do Instituto Oswaldo Cruz, 1, 159218.
Chavez T, Moreno J, Dujardin JP (1999) Isoenzyme electrophoresis
of Rhodnius species: a phenetic approach to relationships within
the genus. Annals of Tropical Medicine and Parasitology, 93, 229307.
Dias E (1952) Doena de Chagas nas Amricas. III. Amrica Central.
Revista Brasileira de Malariologia E Doenas Tropicais, 4, 75 84.
Ditchfield AD (2000) The comparative phylogeography of neotropical mammals: patterns of intraspecific mitochondrial DNA
variation among bats contrasted to nonvolant small mammals.
Molecular Ecology, 9, 13071318.
Dotson EM, Beard CB (2001) Sequence and organization of the
mitochondrial genome of the Chagas disease vector, Triatoma
dimidiata. Insect Molecular Biology, 10, 205215.
Dujardin JP, Chavez T, Moreno JM, Machane M, Noireau F,
Schofield CJ (1999) Comparison of isoenzyme electrophoresis
and morphometric analysis for phylogenetic reconstruction of
the Rhodniini (Hemiptera: Reduviidae: Triatominae). Journal of
Medical Entomology, 36, 653659.
Dujardin JP, Schofield CJ, Panzera F (2000) Les vecteurs e la maladie
de Chagas. Recherches taxinomiques, biologiques et gntiques. NS 24
(5). Acadmie Royale des Sciences dOutre-Mer, Classe des Sciences naturelles et mdicales.
Feliciangeli MD, Dujardin JP, Bastrenta B et al. (2002) Is Rhodnius
robustus (Hemiptera: Reduviidae) responsible for Chagas disease transmission in Western Venezuela? Tropical Medicine and
International Health, 7, 280287.
Felsenstein J (1985) Confidence limits on phylogenies: an
approach using bootstrap. Evolution, 39, 783791.
Galndez-Girn I, Barazarte R, Marquez J et al. (1994) Relaciones
reproductivas entre Rhodnius prolixus Stal y Rhodnius robustus
Larrousse (Hemiptera, Reduviidae, Triatominae) bajo condiciones de laboratorio. Entomologia y Vectores, 1, 314.
Gamboa CJ (1963) Comprobacin de Rhodnius prolixus extradomiciliario en Venezuela. Boletin de la Oficina Sanitaria Panamericana,
54, 1825.
Gamboa CJ (1973) La poblacin silvestre de Rhodnius prolixus en
Venezuela. Archivos Venezolanos de Medicina Tropical y Parasitologia Medica, 5, 321352.
Gomz-Nez JC (1963) Notas sobre la ecologa del Rhodnius
prolixus. Boletn Informativo de la Direccin de Malariologia y
Saneamiento Ambiental, 3, 262272.
Griffiths CS (1997) Correlation of functional domains and rates of
nucleotide substitution in cytochrome b. Molecular Phylogenetics
and Evolution, 7, 352365.
Gusmao J, Lazoski C, Sole-Cava AM (2000) A new species of
Penaeus (Crustacea: Penaeidae) revealed by allozyme and
cytochrome oxidase I analyses. Marine Biology, 137, 435446.
Haffer J (1969) Speciation in Amazonian forest birds. Science, 165,
131137.
Harry M (1993a) Use of the median process of the pygophore in
the identification of Rhodnius nasutus, R. neglectus, R. prolixus
and R. robustus (Hemiptera: Reduviidae). Annals of Tropical
Medicine and Parasitology, 87, 277282.

1006 F . A . M O N T E I R O E T A L .
Harry M (1993b) Isozymic data question the specific status of
some blood-sucking bugs of the genus Rhodnius, vectors of
Chagas disease. Transactions of the Royal Society of Hygiene and
Tropical Medicine, 87, 492493.
Harry M (1994) Morphometric variability in the Chagas disease
vector Rhodnius prolixus. Japanese Journal of Genetics, 96, 233250.
Harry M, Galindez I, Cariou ML (1992) Isozyme variability and
differentiation between Rhodnius prolixus, R. robustus and R.
pictipes, vectors of Chagas disease in Venezuela. Medical and
Veterinary Entomology, 6, 37 43.
Hurtado-Guerrero JC (1992) Anlise Morfomtrica de 14 Colnias
de Rhodnius prolixus e espcies afins (Hemiptera: Reduviidae:
Triatominae). MSc thesis, Universidade do Amazonas, Manaus.
Jukes TH, Cantor CR (1969) Evolution of protein molecules. In:
Mammalian Protein Metabolism (ed. Munro HN), pp. 21132.
Academic Press, New York.
Kimura M (1980) A simple method for estimating evolutionary
rate of base substitutions through comparative studies of nucleotide sequences. Journal of Molecular Evolution, 16, 111120.
Kumar S, Tamura K, Nei M (1994) mega: molecular evolutionary
genetics analysis software for microcomputers. Computation
Applied to Biosciences, 10, 189191.
Langor DW, Sperling FAH (1997) Mitochondrial DNA sequence
divergence in weevils of the Pissodes strobi complex (Coleoptera:
Curculionidae). Insect Molecular Biology, 6, 225 265.
Larrousse F (1927) Etude biologique et systmatique du genre Rhodnius
Stal (Hmiptres, Reduviidae). Annales de Parasitologie, 5, 6388.
Lent H, Jurberg J (1969) O gnero Rhodnius Stl, 1859, com um
estudo sobre a genitlia das espcies (Hemiptera, Reduviidae,
Triatominae). Revista Brasileira de Biologia, 29, 487560.
Lent H, Valderrama A (1973) Hallazgo en Venezuela del triatomino Rhodnius robustus Larrousse, 1927 en la palma Attalea
maracaibensis Martius (Hemiptera, Reduviidae). Boletin de la
Oficina Sanitaria Panamericana, 13, 175 179.
Lent H, Wygodzinsky P (1979) Revision of the Triatominae (Hemiptera:
Reduviidae) and their significance as vectors of Chagas disease.
Bulletin of the American Museum of Natural History, 163, 123520.
Lyman DF, Monteiro FA, Escalante AA et al. (1999) Mitochondrial
DNA sequence variation among triatomine vectors of Chagas disease.
American Journal of Tropical Medicine and Hygiene, 60, 377386.
Meyer A (1994) Shortcomings of the cytochrome b gene as a molecular marker. Trends in Ecology and Evolution, 9, 278 280.
Miles MA, Arias JR, Souza AA (1983) Chagas disease in the
Amazon Basin. V. Periurban palms as habitats of Rhodnius robustus
and Rhodnius pictipes triatomine vectors of Chagas disease.
Memrias do Instituto Oswaldo Cruz, 78, 391398.
Monteiro FA, Wesson DM, Dotson EM, Schofield CJ, Beard CB
(2000) Phylogeny and molecular taxonomy of the Rhodniini
derived from mitochondrial and nuclear DNA sequences.
American Journal of Tropical Medicine and Hygiene, 62, 460465.
Moritz C, Patton JL, Schneider CJ, Smith TB (2000) Diversification
of rainforest faunas: an integrated molecular approach. Annual
Review in Ecology and Systematics, 31, 533 563.
Naiff MF, Naiff RD, Barrett TV (1998) Wild vectors of Chagas disease in an urban area of Manaus (AM): flying activity during
dry and rainy seasons. Revista da Sociedade Brasileira de Medicina
Tropical, 31, 103105.
Nei M (1987) Molecular evolutionary genetics. Columbia University
Press, New York.
Nei M (1996) Phylogenetic analysis in molecular evolutionary
genetics. Annual Review in Genetics, 30, 371 403.
Porter CH, Collins FH (1996) Phylogeny of Neartic members of the

Anopheles maculipennis species group derived from the D2 variable region of 28S ribosomal RNA. Molecular Phylogenetics and
Evolution, 6, 178188.
Saitou N, Nei M (1987) The neighbor-joining method: a new
method for reconstructing phylogenetic trees. Molecular Biology
and Evolution, 4, 406425.
Schmunis GA (1999) Iniciativa del Cono Sur. In: Proceedings of the Second
International Workshop on Population Biology and Control of Triatominae (eds Schofield CJ, Ponce C), pp. 2631. INDRE, Mexico City.
Schofield CJ (2000) Area-wide control of Chagas disease vectors in
Latin America. In: Area-wide control of fruit flies and other insect pests
(ed. Tan KH), pp. 131133. Penerbit Universiti Sains Malaysia,
Penang.
Schofield CJ, Dujardin JP (1997) Chagas disease vector control in
Central America. Parasitology Today, 13, 141144.
Schofield CJ, Dujardin JP (1999) Theories on the evolution of
Rhodnius. Actualidades Biologicas, 21, 183197.
da Silva MNF, Patton JL (1998) Molecular phylogeography and
the evolution and conservation of Amazonian mammals. Molecular Ecology, 7, 475486.
Simpson BB, Haffer J (1978) Speciation patterns in the Amazonian
forest biota. Annual Review of Ecology and Systematics, 9, 497518.
Solano P, Dujardin JP, Schofield CJ, Romana C, Tibayrenc M
(1996) Isoenzymes as a tool for identification of Rhodnius
species. Research and Reviews in Parasitology, 56, 4147.
Stl C (1859) Monographie der Gattung Conorhinus und Verwandten. Berliner Entomologische Zeitschrift, 3, 99117.
Swofford DL (1999) PAUP*. phylogenetic analysis using parsimony
(*and other methods), version 4. Sinauer Associates, Sunderland, MA.
Tonn RJ, CarcavalloRU, Ortega R (1976) Notas sobre la biologa,
ecologa y distribucin geogrfica de Rhodnius robustus Larrousse,
1927 (Hemiptera: Reduviidae). Boletn de la Direccin de Malariologia Y Saneamiento Ambiental, 16, 158162.
Villegas J, Feliciangeli MD, Dujardin JP (2002) Wing shape divergence between Rhodnius prolixus from Cojedes (Venezuela) and
Rhodnius robustus from Mrida (Venezuela). Infection, Genetics
and Evolution, 2, 121128.
World Bank (1993) World development report 1993. Investing in
health. Oxford University Press, New York.

Fernando Monteiro is a research scientist at the Oswaldo Cruz

Institute (FIOCRUZ) in Rio de Janeiro, Brazil. He is interested
in the systematics, phylogeography and population genetics of
Chagas disease vectors. Toby Barrett is a research scientist on the
staff of the National Institute of Amazonian Research (INPA) in
Manaus, Brazil, with interests in the ecology and vector biology of
trypanosomiases and leishmaniases. Sinead Fitzpatrick is a graduate
student at the London School of Hygiene and Tropical Medicine,
where she is working on population genetics of R. prolixus in
Venezuela. Celia Cordon-Rosales is a research biologist at the
Medical Entomology Research and Training Unit/Guatemala, CDC,
and the Center for Health Studies, Universidad Valle, Guatemala,
with interest on the biology of Chagas and malaria disease control
and prevention. Dora Feliciangeli is Head of the section of Medical
Entomology and of the National Sandfly Reference Center at the
Universidad de Carabobo, Venezuela, where she works on the
epidemiology and control of Leishmaniasis and Chagas disease.
Ben Beard is Chief of the Vector Genetics Section in the Division
of Parasitic Disases at CDC and works in molecular epidemiology
and genetics of parasitic diseases and their insect vectors.

2003 Blackwell Publishing Ltd, Molecular Ecology, 12, 9971006