Selective Breeding / Home Page / Colony Breeding / Hybridization/ Excess Birds / Primary Mutations / Combination Mutations / Nestling Attributes /


by Michael L. P. Retter

 Avian hybrid zones are not newly discovered phenomena.  Indeed, over one
hundred years ago, they had been identified for at least two North American complexes
(Harrison 1993).  As time progresses, humans’ curiosity about these intriguing natural
laboratories also increases.  In the past few decades there have been great advances in the
understanding of hybrid zones, but inevitably, more and more questions have surfaced.
This essay will discuss avian hybrid zones and their implications for the evolutionary
process; in particular, it will focus on their effects on classical species concepts.  It will
also propose a revised species concept, and discuss how it would apply to a few selected
avian complexes, Fox Sparrow being covered the most thoroughly.

 What is a hybrid zone?  First, one must understand what constitutes a hybrid.
Various definitions have been used over the past two hundred years.  The widespread
nineteenth century definition stated that hybrids were the sterile offspring of different
species, while mongrels were the fertile offspring of varieties of the same species
(Harrison 1993).  A later definition by Mayr (1963) described hybridization as “the
crossing of individuals belonging to two unlike natural populations that have secondarily
come into contact” (Harrison 1993).  Interestingly, in his highly anticipated book, David
Allen Sibley states that “Hybrids are the offspring of parents of two different species.
The offspring of parents of two different subspecies (of the same species) are know as
intergrades rather than hybrids . . .” (Sibley 2000)  In evolutionary biology (and the
discussion of hybrid zones), the definition of hybrid is expanded to include more than
just the first (F1) generation, and Sibley goes on to include this in his definition.  As the
reader will discover, biologists do not all agree on what constitutes a species, so it is the
author’s opinion that using such specific terminology (species) would only further
complicate an already perplexing matter.  To step around this problem, for the purposes
of this essay, “hybrid” will refer to any individual with mixed ancestry (that is, with
parentage going back to two or, rarely, more unique populations).  There is no
consideration as to whether the two hybridizing populations are considered distinct
species, and therefore, under this definition, intergrades are classified as hybrids.

 Now that the definition of a hybrid has been established, the discussion of hybrid
zones may resume.  Perhaps the best definition is “[where] genetically distinct groups of
individuals meet and mate” (Harrison 1993).  This definition nicely works with the
definition of hybrids as outlined by Mayr above, but on the other had it is rather
ambiguous.  It allows for the inclusion of hybrids between both sympatric (residing in the
same area) and largely parapatric (residing in different, adjacent areas) populations.  All
of the hybrid zones discussed in this essay are of the latter type.  One important step in
recognizing a hybrid zone is verifying that it is, in fact, not a steep cline.  A cline is a
gradual change in (often) morphology from one area to another.  In the case of a cline,
intermediate individuals are not hybrids, but rather part of the continuum from one
extreme to the other.  The North American population of Yellow Warbler (Dendroica
petechia) is a good example; as one goes north and west within its breeding range, the
birds get larger, duller (more greenish-gray), and as one goes southeast, the birds become
smaller, and brighter yellow.  Aiding the biologist in the decision of cline vs. hybrid zone
is the fact that in many hybrid zones, the populations on either side are relatively
homogeneous, which implies that a cline is not involved.  In a cline, the population
would likely continue to change on either side of the zone in question.

 If one leaves the definition of hybrid to include the word “species,” then he must
establish what a species is.  Many ideas have circulated, but two concepts are widely
accepted.  The older of the two, and the more accepted one is called the Biological
Species Concept (hereafter, BSC).

 A recent definition of the BSC states that true species are “biological . . . taxa
which are reproductively isolated; [the] degree of isolation required for species status
varies . . .” (Zink 1994).  Strictly applied, the BSC implies that hybridism is not allowed
between different species where the hybrids would be fertile.  If this were the case, gene
flow from one population into the other (and vice versa) would be possible.  Another
problem with the BSC is that it relies on sex to define species.  What then becomes of an
asexual species?  Is it not a species?  Is each individual lineage its own species?  Clearly,
this is an ambiguity of the BSC which should be corrected.

 The other widely accepted system, the Phylogenetic Species Concept (hereafter
PSC), states that species are “minimally diagnosable clusters of individuals . . . which
may or may not be reproductively isolated” (Zink 1994).  Like most human attempts to
describe the natural world, this concept also has flaws; one problem many scientists have
with it is that subspecies are not recognized within the concept.  After all, subspecies are
presumably  not considered species themselves because they exhibit clinal variation.  So
the problem arises, if one does accept the concept of subspecies, where does one draw
the line between them?  This is a difficult question to answer, and perhaps the PSC is
correct in treating subspecies as “taxa non grata” (Howell 1995).

 The PSC has a few obvious advantages over the BSC.  First of all, it allows for
treatment of allopatric (occupying disjunct areas) populations as separate species,
whereas application of the BSC necessitates a knowledge of the two populations’ ability
to hybridize.  Since allopatric populations do not come into contact with one another, one
cannot observe a natural meeting between the two species, and the world is left to ponder
whether the two populations are separate species.  In this respect, the author favors the
PSC.  In defense of the BSC (and to the detriment of the PSC), “it is hard to argue against
the importance of reproductive isolation because its acquisition segregates a genetic
lineage such that subsequent evolution becomes independent of other lineages” (Able
2000).  Often cited as a problem with the PSC is its elimination of subspecies (discussed
earlier); this seems impractical, especially since biologists need some way to
communicate what organism they are referring to, often on as specific a level as possible.
Eaton described the situation well:  “subspecies is not just a step down from the level of
species to something of a smaller scope or distinctness, nor is it necessarily a population
that is starting to become a distinct species.  Rather, the term is applied simply when
specimens  in a given area have a close similarity to each other and are recognizably
different from those in one or more other areas, yet are not genetically separated from
those.  The amount of difference between members of different subspecies is not
important.  It varies greatly in different cases” (Eaton 1970).  Whereas the BSC may be
to conservative, the PSC is often too liberal:  for instance, the PSC would likely
recognize ten to twenty species of Song Sparrow (Melospiza melodia)!   It is important to
understand that anytime humans try to categorize and classify things in nature using their
own “fictional” criteria, nature will fight back.  The species problem is only one instance
of where science cannot answer all the questions.  Recent actions taken by the American
Ornithologists’ Union (hereafter AOU), including the splitting of the Sage Grouse, Scrub
Jay, Sharp-tailed Sparrow, Solitary Vireo, and Plain Titmouse show that the institution is
leaning more toward the PSC than the BSC.  It could even be that the PSC will become
the precedence used by the AOU in its future rulings on species limits.

 But must we choose one concept over the other?  In light of all these pros and
cons, the following is the proposal of a concept for the treatment of taxa whose ranges
end at hybrid zones:  A species is any diagnosable population of individuals that are
either relatively homogeneous over the bulk of their ranges, or show apparent clines, and
that exhibit gene flow across the bulk of the population; a species may or may not
include various subspecies.  Also, hybrid zones shall denote the barrier between two
species when any of the following is apparent.  1) offspring are not fertile. 2)
interbreeding is infrequent. 3) gene flow across the zone is minimized by reduced hybrid
fitness outside the zone. 4) the time necessary for the two taxa to extensively intergrade is
entirely too long to permit this action.

 From the 1970s to the mid 1990s, Baltimore (Icterus galbula) and Bullock’s
Orioles (I. bullockii) were considered one species, the Northern Oriole (I. galbula), by
the AOU.  Subsequent action taken by the AOU has restored full species status to these
two taxa.  Perhaps one of the reasons behind this latest decision is the apparent decrease
in hybrid fitness as one moves out of the hybrid zone (a form of ecological selection) -- a
concept that has implications for many pairs of taxa that meet at hybrid zones.  Rising
showed that Baltimores exhibit an increased metabolism when under heat stress. This in
turn should yield an increase in water loss which would then make the Baltimore (or any
descendent of a Baltimore) more unfit to live in the hotter, xeric habitats populated by
Bullock’s (Harrison 1993).  Even though “neutral” genes could possibly pass through the
hybrid zone, the integrity of the two parent populations should remain intact so long as
the precipitation in these areas remains somewhat stable (Harrison 1993).  As the hybrids
have decreased fitness outside the hybrid zone and the diagnostic characters of the taxa
are maintained, they would be treated as separate species under the author’s species
 The hybrid zone of Yellow-shafted (Colpates auratus auratus) and Red-shafted
(C. a. cafer) Flickers has been known for over a hundred years, and it still remains a topic
of fierce discussion among avian taxonomists.  Moore and Price showed that the flicker
situation had many things in common with that of the orioles.  Most importantly, they
concluded that the phenotypes of the two taxa would not change within any time period
which is relevant to evolution, because it would depend on the removal of the Rocky
Mountains (Harrison 1993).  Moore and Price suggested that the two flicker taxa are
distinct species, after all if this were not the case, shouldn’t all flickers be
orange-shafted?  Interestingly, the AOU has not adopted their position.  Among other
things, they did not prove that there is “any barrier to free interbreeding between the two”
(DeBenedictis 1997).  The author suggests following Moore and Price in this matter.

Myrtle (Dendroica coronata coronata) and Audubon’s Warblers (D. c. auduboni)
interbreed in British Columbia along the eastern edge of the Canadian Rockies.  Here,
there are two small hybrid zones involved, and “modeling . . . indicated that it would take
over six million years for [the taxa] to merge” (Able 2000); the interesting thing to
consider here is that Barrowclough, Zink, and McKittrick all place the average lifespan
of passerine species well below that figure (Able 2000).  In addition to the two above
taxa of Yellow-rumped Warbler, there exists a third taxon, Goldman’s Warbler (D. c.
goldmani), which inhabits the highlands of eastern Chiapas and Guatemala and is
allopatric with respect to both of the previously mentioned D. coronata taxa.  The AOU
currently treats all three taxa as subspecies within Yellow-rumped Warbler.  The author
suggests a three-way split, and especially suggests full species status for Goldman’s

 Ornithologists have long been intrigued by the intricacies of the genus Passerella.
Recent discussions have centered on exactly how many species are actually contained
within the genus. Early naturalists agreed that four species made up the genus; however;
a “strict application” (Zink 1994) of the biological species concept by the AOU lumped
them into one superspecies:  Fox Sparrow.  The remainder of this essay comprises an
introduction to the four forms of the Fox Sparrow and a discussion of evidence that
should show why at least one of the three accepted classification schemes is invalid.

 The Red Fox-Sparrow complex, iliaca, is the most brightly colored taxon in the
genus. Some diagnostic plumage characteristics include a gray eye-line, chestnut
ear-coverts, reddish breast streaks, a gray rump, and bright rufous tail.  Geographic
variation in the iliaca complex is very minuscule (Zink 1994).  The Yukon Fox-Sparrow,
zaboria, ranges from Alaska to Manitoba.  It differs from the nominate subspecies, the
Eastern Fox-Sparrow, only in having a grayer head and browner malar stripe.  The
distinction between the subspecies is not pronounced, and therefore identification within
the Red Fox-Sparrow complex is only possible with the bird in hand. (Rising 1996)  This
complex breeds in a wide band that stretches from Newfoundland to northern Alaska.
Their preferred breeding habitats are dense willow and alder thickets as well as spruce
and fir bogs.  The author describes the Red Fox-Sparrow’s call note as a loud check,
recalling the call note of a Brown Thrasher.  Sibley (2000) describes it as “a loud smack
like Brown Thrasher.”

 The Sooty Fox-Sparrow complex, unalaschcensis, varies clinally in intensity of
color. The upperparts and head are a variable shade of brown with streaks on the
underparts of the same color.  Sooties can be told from all other Passerella by a complete
lack of gray and rufous in the plumage.  The northernmost birds are a sandy brown while
southernmost birds are a dark coffee-like color.  Six subspecies are recognized in the
Sooty Fox-Sparrow complex, ranging from unalaschensis in the Aleutians to fuliginosa
in extreme northwestern Washington state.  Sooties prefer to breed in willows and alders
at the edge of wet habitats. Rising (1996) describes their call note as a sharp zitt, while
Sibley (2000) says it is like that of Red Fox-Sparrows.

 Found from the interior of northwest British Columbia to Nevada and eastern
California is the schistacea complex, the Slate-colored Fox-Sparrow (Rising 1996).  It is
a tiny-billed bird with a gray head and mantle, brown wings, brown breast streaks, and a
russet tail.  It shows no geographic variation (Zink 1994).  According to Rising, “five or
six subspecies of Slate-colored Fox Sparrows [here meaning a schistacea-megarhyncha
complex] have been recognized, but most of these are poorly differentiated” (Rising
1996).  A two subspecies arrangement recognizing the rather distinctive altivagans (the
northern form) as its own subspecies, and the rest of the population in the southern
portion of the range as the nominate race might be more reasonable.  Altivagans looks
somewhat like one would expect a schistacea-iliaca intergrade to look; it is in fact so
distinct from the other Slate-coloreds that it “has been put with the Red Fox-Sparrows,
but vocal and biochemical evidence indicates that belongs [in the schistacea complex]”
(Rising 1996).  As with Reds and Sooties, Slate-coloreds also prefer to build their nests
on the edges of wet habitats but are much less picky about in which plant they build.
Their call note is sharp klink according to Rising (1996), or “a sharp smack, like Sooty
and Red populations” according to Sibley (2000).

 Along a narrow contact zone from southern Oregon to western Nevada,
schistacea interbreeds with megarhyncha, the Thick-billed Fox-Sparrow. Megarhyncha
is almost identical in plumage to schistacea but has a more extensive blue-gray hood and
less rust in the tail.  The most striking feature of this bird is its enormous beak which can
appear to be three times as large as that of schistacea.  “Megarhyncha” comes from the
Greek mega-, meaning 'great,' and rhunchos, meaning 'snout.'  A Thick-billed's beak also
differs in color from that of the Slate-colored.  Although the culmens of both groups are
grayish brown, Slate-coloreds have yellow lower mandibles instead of the steel blue of
the Thick-billeds’. (Rising 1996)  The megarhyncha complex breeds in mountains from
southern Oregon to southern California east to the Sierra Nevadas and shows little
geographic variation.  Sibley (2000) indicates this group has the most diagnostic call
note, “a high, flat squeak (sic) teep like California Towhee.”

 So how should the genus be treated?  Three basic arrangements exist, the first of
which is to leave the birds under the all-encompassing name of Fox Sparrow.  This would
be the valid course of action if one applies the BSC as strictly as the AOU has.  So, since
all Fox Sparrow groups hybridize with each other at some degree, if the BSC is a valid
theory, the current treatment of Fox Sparrows should be sufficient.  Less strictly applied,
the concept would favor full species status for iliaca and unalaschcensis while lumping
schistacea and megarhyncha (Zink 1994).  This second accepted arrangement had been
gaining widespread support.  For instance in The Sparrows of the United States and
Canada, Rising treats the Fox Sparrow as three species.  With a quick glance, one may
agree that the three-species arrangement is valid.  After all, Slate-coloreds and
Thick-billeds regularly interbreed, but mingling between Reds and Sooties is more
restricted (Zink 1994).  Plus, the only really noticeable difference between them is that
the Thick-billed has a larger beak.  Even though this grouping seems logical, mtDNA
evidence indicates that unalaschcensis and schistacea are actually sister taxa!  In fact,
genetically, megarhyncha is three times as distant from schistacea as unalaschcensis is
(Zink 1994).  In other words, the Slate-colored Fox Sparrow is much more closely related
to the Sooty than it is to the Thick-billed.  From this, one would conclude that contact
between schistacea and megarhyncha is secondary, clear evidence against “parapatric
differentiation” (Zink 1994).  Now that it has been proven that hybridization is not
restricted to sister taxa, recognizing a schistacea-megarhyncha complex “would
misrepresent the recovered pattern of evolutionary history” (Zink 1994).  Adoption of the
three-species concept would be to the following fictional scenario, albeit far-fetched.  Let
us assume that the Black Rhinoceros has been found breeding with the African Elephant,
so they are lumped together as the same species.  However, the Asian Elephant which has
no opportunity to breed with its southern counterpart, would remain a separate species.
The two elephants are clearly more closely related than the African Elephant is to the
rhino, and strict application of the BSC in this instance would grossly misrepresent
evolutionary history, defeating the purpose of the BSC.  More realistically, consider
Blue-winged and Golden-winged Warblers.  The two species interbreed freely and
produce fertile offspring.  According to the BSC, they are then one species (which to the
author, is clearly not so).   So, does the BSC work after all?

 After considering the evidence, one is left with two choices, to leave the Fox
Sparrow lumped together, or to split it completely.  The above argument has established
that the BSC may be ultimately flawed, so why should we continue usage of a concept
which has been shown to be defective?  Since the first arrangement was (depending on
your opinion) based on what has been proven to be an imperfect concept, that
arrangement may be discarded.  Furthermore, mtDNA evidence reveals that four distinct
genetic groups in Passerella are present which correspond exactly to the four species first
recognized by early naturalists (Zink 1994).  Now only the third and final arrangement is

 This final arrangement treats the genus as it originally was, as four distinct
species, and uses the PSC as its basis rather than the BSC.  A four species grouping fits
the PSC exactly.  Why should genetically distinct populations which can be told apart in
the field not be given full species status?  Indeed, Zink said in a later paper, “It seems
straightforward to assume that groups which look different and have different DNA may
be different species” (Zink and Kessen 1999).  Since the four species grouping seems to
be the only credible choice left, evidence seems to point out that the PSC may be the
better choice in this particular instance.

 Besides being genetically distinct, the Red, Sooty, Slate-colored, and Thick-billed
Fox-Sparrows differ in appearance, habitat preference, behavior, and voice.   Three
different species arrangements have been proposed, but two, including the current
arrangement, have been found to be flawed.  Perhaps Zink summarizes the situation best:
“Divergence has occurred and [sic] the limits of evolutionary units are apparent from
mtDNA and morphology.  The four groups maintain their (mtDNA) integrity over most
of their ranges and should be treated as species that have retained to varying degrees the
primitive ability to hybridize” (Zink 1994).  The author suggests that the AOU reconsider
the Passerella problem and overturn its current treatment thereby adopting a four species

Figure 1 - ranges of the 4 Passerella complexes.

 Hybrid zones are immensely interesting phenomena, and they have forced humans
to reevaluate some of their evolutionary concepts.  Must we choose one of the accepted
species concepts over the other?  In the author’s opinion, each situation should be
examined carefully and individually, and then a treated using a revised species concept.
He advocates a revised, more conservative PSC where subspecies are regognized,
including the following addendum:  A species is any diagnosable population of
individuals that are either relatively homogeneous over the bulk of their ranges, or show
apparent clines, and that exhibit gene flow across the bulk of the population; a species
may or may not include various subspecies.  Also, hybrid zones shall denote the barrier
between two species when any of the following is apparent.  1) offspring are not fertile.
2) interbreeding is infrequent. 3) gene flow across the zone is minimized by reduced
hybrid fitness outside the zone. 4) the time necessary for the two taxa to extensively
intergrade is entirely too long to permit this action.  In particular, the author suggests
that the AOU reevaluate the status of the Yellow-rumped Warbler, Northern Flicker, and
Fox Sparrow complexes.  Ultimately, the studies and further research into these
complexes may likely yield an increased understanding of hybrid zones, species
concepts, and the evolutionary process in general.


Able, Kenneth P.  “Sage Grouse Futures.”  Birding, 32 (2000), 306-216.

DeBenedictis, Paul A.  “Flicker Futures.”  Birding, XXXIX (1997), 420-424.

Eaton, Theodore H., Jr.  Evolution.  New York:  W.W. Norton & Co, 1970.

Harrison, Richard G.  Hybrid Zones and the Evolutionary Process.  New York: Oxford
University Press, 1993.

Howell, Steve N.G and Sophie Webb.  A Guide to the Birds of Mexico and Northern
Central America.  New York:  Oxford University Press, 1995.

Rising, J.D.  The Sparrows of the United States and Canada.  San Diego:  Academic
Press, 1996.

Sibley, David Allen.  The Sibley Guide to Birds.  New York:  Chanticleer Press, Inc.,

Zink, Robert M.  “The Geography of Mitochondrial DNA Variation, Population
Structure, hybridization, and Species Limits in the Fox Sparrow (Passerella iliaca).”
Evolution, 48 (1994), 96-111.

Zink, Robert M. and Ann E. Kessen.  “Species Limits in the Fox Sparrow.”  Birding, 31
(1999), 508-517.