Banding One Millions Birds

A Gambel's White-crowned Sparrow (Zonotrichia leucophrys gambelii) and an Eastern White-crowned Sparrow (Z. l. leucophrys).

Photo: Long Point Bird Observatory

Here is a recent article I wrote for BirdWatch Canada, a publication of Bird Studies Canada. This spring, Long Point Bird Observatory, for which I am the Program Coordinator, banded its one millionth bird, an milestone that no other Western Hemisphere bird observatory has yet reached. This article was part of a lead up to that event.

Long Point Bird Observatory 2016 Year End Report

The first Barred Owl (Strix varia) ever banded at Long Point Bird Observatory was captured in 2016.

Photo: Terren (Wikimedia Commons)

In 2016 Long Point Bird Observatory (LPBO), the oldest bird observatory in the western hemisphere, completed its 58th migration monitoring season. LBPO banded 44,612 birds last year. Additionally 5,419 recaptures of previously banded birds were processed. The first Barred Owl ever banded at LPBO was captured during the fall season. Other notable banding records included the eighth ever Broad-winged Hawk (the first since 2006), the fourth ever Painted Bunting and record high banding totals for:

· Cliff Swallow, 22 (previous record was 13 in 1982);
· Oregon Junco, three (single birds banded in six previous years)
· Red-eyed Vireo, 496 (previous record was 490 in 2012);
· Summer Tanager, four (tied with 2009);
· Tufted Titmouse, five (previous record was four in 2005);
· Warbling Vireo, 162 (previous record was 143 in 2014); and
· Yellow Palm Warbler, six (tied with 2005).

Despite these highlights, LPBO banded the fewest birds since 2004. Last year, LPBO banded 15.5% (4,498 individuals) fewer birds than the previous 10-year average. The story was no better for the number of species and forms, with 2016 again being the lowest total since 2004. The 141 species and forms captured in 2016 was about 10% (16 species) below the previous 10-year average.

The complete 2016 Year End Report can be found here.

Naturescape: Red-headed Weaver

A Red-headed Weaver (Anaplectes rubriceps) contorts its neck while tying knots into its complex nest.

Photo: Mark Conboy

Naturescape: Sanderling

A Sanderling (Calidris alba) runs ahead of the surf on a California beach.

Photo: Mark Conboy

Naturescape: Curve-billed Thrasher at Dawn

A Curve-billed Thrasher (Toxostoma curvirostre) lords over Organ Pipe National Monument at sunrise.

Photo: Mark Conboy

Naturescape: Crested Barbet

A Crested Barbet (Trachyphonus vaillantii) exploring my campsite in northern Botswana.

Photo: Mark Conboy

BioBrevia: Comprehending Cuckoos

A victim of brood parasitism, an Eurasian Reed Warbler (Acrocephalus scirpaceus) feeds a Common Cuckoo (Cuculus canorus) nestling. The decline of Common Cuckoo means that this behaviour is becoming less common across Europe.

Photo: Per Harald Olsen (Wikimedia Commons)

A new paper in Nature Communications, Population Decline is Linked to Migration Route in the Common Cuckoo, by Hewson et al shows that Common Cuckoos (Cuculus canorus) take two different routes from Europe to their African wintering grounds during autumn migration. The researchers used satellite telemetry to track birds in virtually real time as they made their way south. The data showed that birds which took the shorter route experienced higher mortality rates than birds that took the longer route. This differential mortality correlated to population declines on the British breeding grounds. It's further evidence that effective conservation of migratory birds must take into account migration ecology as much as breeding and wintering site ecology.

Understanding Molt: A Bander's Perspective

The broad, round-tipped, thick flight feathers, sharply contrasting black and white primaries and alula, nearly unifromly white secondaries, and pure white primary coverts all indicate this Snow Bunting (Plectrophenax nivalis) is a male in definitive basic plumage, meaning that it hatched in at least 2014 (this bird was captured in January 2016). Molts and plumages are among the most important features used for aging birds during the banding process.

Photo: Mark Conboy

This primer on molt in birds is read by all new volunteers and staff at Long Point Bird Observatory (LPBO). This is the most basic level of understanding that banders are expected to have before they can begin work in the banding lab. This primer assumes that the reader has at least a basic understanding of feather terminology (i.e., what are the greater coverts versus median coverts).

In bird banding, a comprehensive understanding of the strategies, timings, and mechanics of molt is necessary for aging birds with accuracy and precision. Molt is a cyclic pattern of feather replacement, and is highly variable between species, and exceptions occur to almost all of the "rules" that follow, but in general terms:

• Every species at LPBO molts all of its feathers at least once a year, usually after the breeding season.
• Some species have an additional annual molt before the breeding season.
• In most species, young birds molt differently than adults. For example, some species have two molts during their first 12 months of life, but only have a single molt every 12 month after that for the rest of their lives.

Note: The term molt is best reserved only for feather replacement that is part of a cyclic and predicable process, so that the growth of new feathers to replace randomly lost or damaged ones is not considered to be a true molt, but should instead be referred to as adventitious feather replacement.

In general, molt is usually symmetrical, meaning, for example, that a feather actively molting on the left wing is simultaneously molting on the right wing. Similarly, if the outermost tail feather is actively molting on the left side of the tail, the outermost tail feather on the right side should be molting simultaneously.

Molt usually occurs in a predictable sequence and that sequence is reflected in the numbering system we apply to the wing and tail feathers. All of the wing’s flight feathers and the tail feathers are numbered in the order which they typically molt.

Wing and tail feather terminology and numbering.

Illustration: Adapted from the North American Banders' Study Guide (North American Banding Council)

In most species, the primary feathers molt from the innermost to the outermost, and the secondary feathers molt from the outermost to the innermost. The tail feathers, properly called rectrices, molt from the innermost to the outermost.

In order to age birds you will need to know what patterns of molt each species uses. Read the molt section in Pyle for each species as you encounter them. Contained within the molt section is almost all of the information you need to age a bird. Pyle tells you what kind of molt(s) each species undergoes and when it occurs. Pyle also notes exceptional (eccentric) kinds of molts that don't fit the general patterns discussed below.

There are three types of molts that you will regularly encounter at LPBO: preformative, prebasic and prealternate. These types of molts can occur in several different arrangements or molt strategies. The majority of LPBO birds follow two particular molt strategies:

1. Complex Basic Strategy

Natal Down

ê

Prejuvenal Molt à Juvenal/First Basic Plumage

ê

Preformative Molt à Formative Plumage

ê

Definitive Prebasic Molt à Definitive Basic Plumage

2. Complex Alternate Strategy

Natal Down

ê

Prejuvenal Molt à Juvenal/First Basic Plumage

ê

Preformative Molt à Formative Plumage

ê

First Prealternate Molt à First Alternate Plumage

ê

Definitive Prebasic Molt à Definitive Basic Plumage

ê

Definitive Prealternate Molt à Definitive Alternate Plumage

ê

Definitive Prebasic Molt à Definitive Basic Plumage

Note: There are two additional molt strategies that are rarely used by species encountered at LPBO, so won’t be addressed in detail here: Simple Basic Strategy and Simple Alternate Strategy.

Though these diagrams may seem nebulous, especially if the terminology is new to you, they will become more clear as we examine each molt and each plumage in turn, below. Fundamental to this, is understanding that molts preceed plumages and plumages only arise from molts. A bird cannot change plumages without molting!

Note: Molts are named with the prefix “pre”; so that the molt which leads to a bird’s juvenal plumage is called the prejuvenal molt. There are other molt and plumage terminology systems in use, but the one presented here is the best and is the one in most widespread use among banders today.

We will examine all of the molts and plumages and outline, briefly, how they can be used to determine the age of a bird, beginning with natal down. Natal down is the first kind of feather that nestlings grow. Some species hatch fully covered in down (precocial), others hatch naked and subsequently grow down (altricial). The development of natal down is not considered a true molt. True molts actually begin with the next stage of feather growth.

The prejuvenal molt occurs in the nest. It leads to the development of juvenal plumage. The prejuvenal molt is sometimes also referred to as the first prebasic molt and juvenal plumage is sometimes called first basic plumage. At LPBO we tend use the more intuitive terms prejuvenal molt and juvenal plumage. The prejuvenal molt is unusual in that all feathers are grown at the same time, thus it is a simultaneous molt. All subsequent molts in a bird's life are sequential molts, meaning feathers are replaced a few at a time, not all at once.

Juvenal plumage is grown in the nest. Its primary function is to make young birds mobile so they can leave the nest, reducing the risk of depredation and giving them the ability to follow their parents while they forage. Juvenal plumage is short-lived, lasting for only a few weeks in most LPBO species, before being replaced by formative plumage. Juvenal plumage, aside form usually looking very different from all subsequent plumages that a bird will have, has particular characteristics that make it distinctive. Understanding these characteristics is imperative to aging birds in non-juvenal plumages (i.e., formative, basic, alternate):

• Body feathers are whispy, with widely spaced barbs; they have the appearance of being low-quality feathers.
• Primary feathers are usually narrow with a somewhat pointed tip.
• Retrices are usually narrow and come to a noticeably pointed tip.
• Flight feathers are usually dull coloured, with limited or no coloured edging.
• Flight feathers and rectrices are usually thin, sometimes to the point of being translucent.
• Growth bars line up across the wing and tail feathers.
• Fault bars, when present, line up across the wing and tail feathers.

Recognizing the characteristics of juvenile feathers takes practice. In fact, all of these characteristics should be thought of in relative terms, so that juvenal feathers are more pointed than non-juvenal feathers. After handling many individuals of the same species you’ll begin to understand the relative differences between juvenal and non-juvenal feathers.

Note: Growth Bars are feather pigment bands that, when visible, can be useful for aging birds. Growth bars develop because daytime and nighttime metabolic processes take place at different rates in birds, so that pigment deposition is more concentrated during the day and less concentrated at night. In some species or individuals, growth bars can be difficult to see, but in other species or individuals they can be easily detected. Because the prejuvenal molt is simultaneous, juvenal plumage growth bars line up across the wing feathers and rectrices. Conversely, all other molts are sequential, so growth bars don't line up across the feathers that have grown during the preformative, definitive prebasic or prealternate molts. Growth bars in the rectrices need to be used with caution because it is possible for a bird in definitive basic plumage that accidentally lost its tail to adventitiously regrow those feathers simultaneously, in which case it may have growth bars similar to a bird in juvenal plumage. Fault bars are growth bars that developed during periods of stress, so that pigmentation is particularly weak. They are usually very obvious, but their occurrence is somewhat infrequent. In addition to being weakly pigmented, fault bars are also usually structurally weak, meaning they are prone to breaking, leaving the feathers looking unnaturally truncate. The same characteristics and cautions apply to fault bars as to normal growth bars.

Growth bars in Carolina Wrens (Thryothorus ludovicianus) and other wrens show up as dark bands on the wing feathers and rectrices. Growth bars are usually far less obvious in other birds, but the principle is the same. The top photo is of a bird in definitive basic plumage, showing dark growth bars slightly staggered on the wing feathers. The bottom photo is of a bird in formative plumage, showing dark growth bars that line up across the inner greater coverts and across the primaries and secondaries.

Photos: Marie-Anne Hudson (Piranga)

The preformative molt replaces the juvenal plumage with the formative plumage. The most important thing about the preformative molt is that it is typically a partial molt. That means most or all body feathers, but usually no primaries, secondaries, tertials or retrices are replaced (in some species a few tertials or the central rectrices may be replaced).

Note: Part 1 of Pyle doesn’t use the term preformative molt, but instead uses the term first basic molt. This volume was written before a new (and more accurate) terminology came into use. When Part 1 of Pyle says first prebasic molt, it really means preformative molt. When Part 1 of Pyle says first basic plumage, it really means formative plumage. You should attempt to always use the new terminology and not the terminology used in Part 1 of Pyle. Part 2 of Pyle, having been more recently published, uses the new terminology.

Formative plumage is roughly analogous to what most field guides call immature plumage. Understanding that formative plumage follows the partial preformative molt is crucial to aging LPBO birds! So, once again: typically, the preformative molt is partial, involving the replacement of only some juvenal plumage with formative plumage, while retaining other parts of the juvenal plumage. Here, the concept of molt limits becomes important. A molt limit is the contrast between different generations of feathers. In formative plumage a bird will show contrast between retained juvenal and replaced formative feathers. Detecting molt limts is the key to aging birds by plumage. Typically, molt limits occur in the wing feathers. In many LPBO species, juvenile primaries, secondaries, tertials and primary coverts are retained, while the greater, median and lesser coverts are replaced (fully or partly) by formative feathers. Thus there may be a visible molt limit between the retained juvenile feathers and replaced formative feathers in birds that have completed their preformative molt. Formative feathers are usually brighter in colour, rounder-tipped, broader, and thicker. In general, they look very much like definitive basic feathers (see below). When we detect a molt limit that's the result of a preformative molt, we know that the bird hatched in the most recent breeding season.

Myrtle Warbler (Setophaga coronata) in formative plumage. The primaries, secondaries, primary coverts and alula are retained juvenal feathers, while the greater and median coverts are replaced formative feathers. This molt limit, between the duller juvenal feathers and brighter formative feathers could only occur on a bird that hatched in the most recent breeding season.

Photo: Marcel Gahbauer (Piranga)

Note: When looking for molt limits, it is important to be aware of pseudolimits in some species. Pseudolimits are apparent contrasts between feathers, giving the impression of a true molt limit, but they are merely inherent differences between feathers of the same generation, thus are not indicative of a bird's age. Only a handful of LPBO species regularly have confusing pseudolimits, including many sparrows and Brown Thrasher (Toxostoma rufum).

Brown Thrasher (Toxostoma rufum) in definitive basic plumage. The apparent contrast between the inner and outer greater coverts is normal for Brown Thrashers in all plumages; it is a pseudolimit and is not useful for aging.

Photo: Peter Pyle (Piranga)

The definitive prebasic molt replaces formative plumage with definitive basic plumage. In contrast to the preformative molt, the definitive prebasic molt is almost always a complete molt in LPBO birds. Every single feather on a bird’s body is replaced during a complete molt: every body feather, every flight feather, and every rectrix. For LPBO birds, the definitive prebasic molt usually occurs on the breeding grounds, right after breeding is completed, though exceptions do occur. The reason the molt is called definitive is because it occurs every year of the bird’s life, except for its first year, when the preformative molt takes place at roughly the same time. The preformative molt occurs only once, replacing the juvenal plumage, while the definitive prebasic molt occurs every single year thereafter.

Myrtle Warbler (Setophaga coronata) in definitive basic plumage. All of the feathers of the wing are broad, round-tipped, and thick. There are no molt limits, meaning that this bird hatched not in the most recent breeding season, but at least in the season before that.

Photo: Marcel Gahbauer (Piranga)

Definitive basic plumage is roughly analogous to what field guides call adult plumage. It differs from formative plumage in that no juvenal plumage is mixed in with the basic feathers. When a bird is found to have only definitive basic plumage we know that it is one that hatched not in the most recent breeding season, but at least in the season before that. It is impossible to age most LPBO birds with any more precision than that because the definitive basic plumage occurs in birds that are one year old, two years old, three years old, etc; it is the same every year following the preformative molt. Definitive basic plumage has particular characteristics that make it distinctive from juvenal and formative plumages. Understanding these characteristics is imperative to aging birds:

• Body feathers are tight and crisp with dense barbs; they have a high-quality appearance.
• Primary feathers are usually broad and a fairly rounded at the tip.
• Retrices are usually broad and come to a noticeably truncate end.
• The feathers of the wing are usually colourful, usually with thick coloured edging.
• Flight feathers and rectrices are usually thick.
• Growth bars do not line up across the wing and tail feathers.
• Fault bars, rarely present, don't line up across the wing and tail feathers.

Like with juvenal feathers, recognizing the characteristics of definitive basic feathers takes practice, and should be thought of in relative terms, so that definitive basic feathers are less pointed than juvenal feathers, for example. After handling many individuals of the same species you’ll begin to understand the relative differences between definitive basic and juvenal feathers.

The complex alternate molt strategy has two additional types of molt that are not part of the complex basic strategy. These are called prealternate molts and they result in the development of alternate plumages, which are roughly analogous to the breeding plumages illustrated in field guides. Less than half of LPBO species have a prealternate molt and in all cases it's a partial or incomplete molt. Prealternate molts usually take place on the wintering grounds either in winter or early spring.

Note: In banding terminology there is an important distinction between partial and incomplete molts. As discussed earlier, partial molts include most or all body feathers, but usually no primaries, secondaries, tertials or retrices are replaced (in some cases a few tertials or the central rectrices may be replaced). By contrast, incomplete molts usually include the replacement of all body feathers as well as some primaries, secondaries, tertials and/or rectrices. Partial prealternate molts are far more common among LPBO species than are incomplete prealternate molts.

The first prealternate molt is partial (or in some cases incomplete) and results in the first alternate plumage. The first prealternate molt replaces some feathers of the formative plumage. The first prealternate molt usually is limited to feathers of the head as well as the greater, median and/or lesser coverts in many LPBO species. In other species, such as American Goldfinch (Spinus tristis), almost all of the body feathers will also be replaced, but such extensive molts are not the norm for LPBO species. Incomplete molts may also include some primaries, secondaries, tertials and/or rectrices. Birds in first alternate plumage may thus show molt limits between three generations of feathers in their wings (e.g., juvenal primaries, secondaries, tertials and primary coverts, formative outer greater coverts, and first alternate inner greater coverts, as seen in some Myrtle Warblers [Setophaga coronata]). When we detect molt limits that are the result of a first prealternate molt, we know that the bird hatched in the most recent breeding season.

Myrtle Warbler (Setophaga coronata) in first alternate plumage. The primaries, secondaries and primary coverts are retained juvenal feathers. The second outermost greater covert is a formative feather, while the rest of the greater coverts and median coverts are first alternate feathers. These molt limits, between three generations of feathers, duller juvenal feathers, a slightly brighter formative feather, and brightest first alternate feathers, could only occur on a bird that hatched in the most recent breeding season.

Photo: Marcel Gahbauer (Piranga)

Like the first prealternate molt, the definitive prealternate molt is partial (or in some cases incomplete); it results in the definitive alternate plumage and replaces some feathers of the definitive basic plumage. The definitive prealternate molt is usually also limited to feathers of the head as well as the greater, median and/or lesser coverts in many LPBO species. But there are cases when body feathers, primaries, secondaries, tertials and/or rectrices are replaced. Birds in definitive alternate plumage show molt limits between two generations of feathers in their wings (e.g., definitive basic primaries, secondaries, tertials, primary coverts and outer greater coverts, and definitive alternate inner greater coverts). When we detect molt limits that are the result of a definitive prealternate molt, we know that the bird hatched not in the most recent breeding season but at least in the season before that.

Myrtle Warbler (Setophaga coronata) in definitive alternate plumage. All of the feathers of the wing are broad, round-tipped, and thick. The primaries, secondaries and primary coverts are definitive basic feathers, while the greater coverts and median coverts are definitive alternate feathers. A molt limit of this kind means this bird hatched not in the most recent breeding season, but at least in the season before that.

Photo: Marie-Anne Hudson (Piranga)

Note: There is another type of molt that is rather rare among North American songbirds, which I haven't yet addressed: the presupplimental molt, which leads to the supplimental plumage. Presupplimental molts can vary widely between species. For example, one species may molt only wing feathers, while another species may molt only body feathers; thus, presupplimental molt is sort of a catch-all term which is applied to any molt, whether it be partial, incomplete or complete, that occurs in addition to the prejuvenal, preformative, prebasic and prealternate molts. Only two LPBO species are known to undergo presupplimental molts of any consequence: Northern Cardinal (Cardinalis cardinalis) and Indigo Bunting (Passerina cyanea). Refer to Pyle for details about aging these species.

When aging birds by molt, put all of the above information together to decide what age to give the bird:

1. Read the molt section in the species account in Pyle so you know what to look for at a given time of year.
2. Eliminate all plumages you are not likely to encounter at the time of year.
3. Open a wing and look at the primaries, secondaries, tertials, primary coverts and greater coverts in particular, for molt limits. Determine molt limits by comparing the shape, colour, thickness and amount of wear between feathers.
4. If molt limits are present, determine if they are the result of a preformative molt or a prealternate molt. If molt limits are not present determine if the feathers are uniformly juvenal or non-juvenal.
5. Open the tail and look for molt limits. Determine molt limits by comparing the shape, colour, thickness and amount of wear between feathers. Pay particular attention to the central and outer rectrices.
6. If molt limits are present, determine if they are the result of a preformative molt or a prealternate molt. If molt limits are not present determine if the feathers are uniformly juvenal or non-juvenal.
7. Look at the body feathers. Determine if the body feathers are juvenal or non-juvenal. Use the plumage criteria given in Pyle to see if additional clues about age can be found in the colours of body, wing feathers and rectrices.
8. Look at other characteristics such as colour of the eyes, orbital skin, legs, beak and/or mouth lining. Determine if the bird has a gape, is in breeding condition, and/or if the skull is completely ossified (look for these last three criteria at the appropriate time of year).
9. Come to a final decision about the bird's age based on a combination of criteria.

If there is conflicting evidence for age, it may be necessary to make a decision based on the strength of one criterion over another. In general, the most reliable plumage features (for a suitably experienced bander) are the wing feathers. Adventitious replacement or extreme wear of rectrices can result in confusion, so I prefer to use rectrics in combination with other plumage criteria, namely the wing feathers, when possible. Body feathers can be deceiving to inexperienced banders who expect males to always be brighter than females and visa versa - this is not necessarily the case, especially with the warblers! The use of non-plumage characteristics should not be overlooked, especially skull ossification, which with practice, can be the most accurate way to determine a bird's age.

There are plenty of additional resources about molt:

Piranga (NatureInstruct)

North American Banders' Study Guide (North American Banding Council)

The Variety of Molt Strategies (Steve N.G. Howell)

Finding Order Amid the Chaos (Steve N.G. Howell)

Aging North American Landbirds by Molt Limits and Plumage Criteria (Dan Froehlich)

As you read through these documents, and as you begin to work in LPBO's banding lab, you will see that aging birds by molt is far more nuanced and fraught with exceptions than I've presented here. This makes banding a source of perpetual learning and challenge, something that any serious bander will embrace, rather than dismay.

Naturescape: Birds of Rio Lagartos

American Flamingos (Phoenicopterus ruber) preen in front of a mixed flock of American White Pelicans (Pelecanus erythrorhynchos), Brown Pelicans (Pelecanus occidentalis) and Laughing Gulls (Leucophaeus atricilla) near the salt-producing town of Los Colorados at the northern tip of the Yucatan Peninsula, Mexico.

Photo: Mark Conboy

Isla de Pequeños Carnívoros: Cozumel

Splendid Toadfish (Sanopus splendidus) are endemic to the coastal reefs of Isla Cozumel

Photo: Randall McNeely

Rough and wild was the 30 minute ferry crossing from the Mexican mainland to Isla Cozumel. The boat, big enough to hold a couple hundred passengers, lurched its way across the 19 km wide channel, tossing about on the waves as though it was a mere canoe. Suspecting, based on the crew's cavalier attitude, that this was par for the course, I sat back and enjoyed the dramatic heave-to of the waves and veils of spray that blasted from the bow. The ride reminded me more of the North Atlantic than the western Caribbean. Despite the waves, our capable Mexican captain brought us dockside without incident. Half-domesticated Brown Pelicans (Pelecanus occidentalis) and skeins of undomesticated tourists lined the wharf. The pelicans were a welcome sight, and so were the Ruddy Turnstones (Arenaria interpres) that foraged among the feet of passersby, like some kind of maritime pigeons. I had stilled myself for tourists, but the sight of half a dozen massive cruise ships anchored nearby caused me to recalibrate my expectations. Cozumel, like Playa del Carmen, the mainland port from which I sailed, is a tourist trap, attracting sun-seekers from all over the north to hotels and resorts, including some that (disappointingly, but not surprisingly) have their own private pods of captive dolphins! But, I wasn't on the island for a luxury vacation, and I certainly wasn't heading to play with the caged cetaceans. Instead, I was in search of something far more interesting and far more worthwhile: Cozumel's unique endemic species.

Isla Cozumel amounts to only about 10% of Quintana Roo's land area, but it holds an estimated 40% of the state's animal diversity; and a great deal of that diversity is found only on Cozumel.

The Cozumel Harvest Mouse (Reithrodontomys spectabilis), Pygmy Raccoon (Procyon pygmaeus), and Dwarf Coati (Nasua nelsoni) are all endemic. So is the enigmatic Cozumel Fox (Urocyon sp.), a very rare species, presumably similar to its mainland counterpart the Grey Fox (Urocyon cinereoargenteus), but it has never actually been scientifically described. Both the Cozumel Emerald (Chlorostilbon forficatus) and the Cozumel Vireo (Vireo bairdi) are endemic. A third endemic bird, the Cozumel Thrasher (Toxostoma guttatum) is exceedingly rare, indeed almost extinct. The Cozumel Whiptail (Aspidoscelis cozumela) is the only endemic reptile. The coral reefs which fringe the island's shores are home to the endemic, Splendid Toadfish (Sanopus splendidus). There is also the very unusual cave-dwelling sea star Copidaster cavernicola, and at least three endemic species of crustaceans: Agostocaris bozanici, Yagerocaris cozumel, and Bahadzia setodactylus.

The fun doesn't stop there. Isla Cozumel is also home to endemic subspecies of Common Opossum (Didelphis marsupialis cozumelae), Coues' Rice Rat (Oryzomys couesi cozumelae), White-footed Mouse (Peromyscus leucopus cozumelae), Collared Peccary (Pecari tajacu nanus), Great Curassow (Crax rubra griscomi), House Wren (Troglodytes aedon beani), Blue-grey Gnatcatcher (Polioptila caerulea cozumelae), Black Catbird (Dumetella glabrirostris cozumelana), Yucatan Woodpecker (Melanerpes pygmaeus pygmaeus), Golden-fronted Woodpecker (Melanerpes aurifrons leei), Yucatan Flycatcher (Myiarchus yucatanensis lanyoni), Brown-crested Flycatcher (Myiarchus tyrannulus cozumelae), Bright-rumped Attila (Attila spadiceus cozumelae), Rufous-browed Peppershrike (Cyclarhis gujanensis insularis), Yellow Warbler (Setophaga petechia rufivertex), Rose-throated Tanager (Piranga roseogularis cozumelae), Western Spindalis (Spindalis zena benedicti) and Northern Cardinal (Cardinalis cardinalis saturata). In addition, three subspecies are near-endemic: Both the Roadside Hawk (Buteo magnirostris gracilis) and the Yellow-faced Grassquit (Tiaris olivacea intermedius) are found on Cozumel as well as Holbox Island, off the Yucatan's north coast. The Bananaquit (Coereba flaveola caboti) is found on Cozumel and some other islands off the Yucatan Peninsula.

Why does Cozumel have so many endemic species and subspecies? What makes this relatively dry, rocky, hurricane swept, Caribbean Thatch Palm (Thrinax radiate) clothed, 486 sq km island such a hotspot of biodiversity? I've always wondered...

It's not unusual for islands to possess unique fauna. It's by virtue of their isolation that islands tend be relatively depauperate in total species, but of the ones that do occur there, a good many may be endemic. Madagascar and only Madagascar boasts lemurs. Tasmania has its eponymous Devil (Sarcophilus harrisii). Eil Malk has a lake teeming with Stingless Golden Jellyfish (Mastigias papua etpisoni). Isla Socorro has the Socorro Mockingbird (Mimus graysoni). Santa Cruz Island has the Island Scrub Jay (Aphelocoma insularis). The main islands of New Zealand have the Lesser Short-tailed Bat (Mystacina tuberculata). Indeed, most of these islands have pantheons of endemic species. From Galapagos to Borneo, from Sri Lanka to South Georgia, islands are hotbeds of endemicity. But many (not all, but many) islands that have particularly rich endemic diversity are rather isolated. Oceanic or microcontinental islands, those disconnected from the nearest continental shelf, are so isolated that when a wayward bird or reptile comes ashore, they are not likely to be joined by others of their kind. When there's no gene flow between an island and the mainland, a host of evolutionary processes like founder effect, genetic drift, and good old fashion natural selection, cause island colonists to diverge in form and behaviour from their continental ancestors.

Cozumel, as far as islands go, is not very isolated from the mainland. Though only 19 km wide, the channel that separates Cozumel from continental Quintana Roo, is also some 900 m deep. That's deep enough to ensure that ever since the island first rose out of the sea some 200,000 years ago, it has never had a physical connection (a land bridge, if you like) to the mainland. Cozumel has been completely submerged by the ocean during times of high water (it's highest point is only about 10 m above sea level), but it's never been connected to the mainland, not even during periods of low sea levels, such as during the last ice age. It's not nearly as isolated as a typical oceanic island, but then again, some of the most diverse islands on Earth are not particularly isolated either: Borneo, Sumatra and New Guinea, for example. Evidentially, it doesn't take extreme distance, just a certain degree of isolation, to promote island endemicity.

Take the carnivores of Cozumel, for instance. There's the endemic Pygmy Raccoon and the Dwarf Coati. As their names suggest, they're small compared to their mainland relatives. Dwarf Coatis, for example, are only about 75% the size of mainland White-nosed Coatis (Nasua narica). The very rare (apparently not a single museum specimen exists) Cozumel Fox is also a dwarf, being essentially a reduced version of the mainland's Grey Fox. Why does the Cozumel carnivore fauna have a decidedly dwarfish aspect? In fact, dwarfism extends beyond the carnivores; the island's Collared Peccaries, Great Curassows, and Cozumel Thrashers are all miniaturized when compared to their mainland counterparts. So what's the deal, why evolve towards smallness?

A general pattern among island fauna the world over is that big creatures get smaller on islands, while little creatures get bigger. Biogeographers call this Foster's Rule. Think of the Komodo Dragon (Varanus komodoensis), a supersized monitor lizard. Or consider the diminutive White-tailed Deer (Odocoileus virginianus clavium), the so-called Key Deer, of the Florida Keys. Big things get small, small things get big. I'd love to call it a law of nature, but it's not: as far as rules go, Foster's is one that's fraught with exceptions.  As David Quammen tells us in his wonderful exploration of island biogeography, Song of the Dodo, "Many kinds of animal are likely to grow larger on islands, yes, except under exceptional circumstances, which instead make them grow smaller. But to the exceptional circumstances there are other exceptions, which might again make them grow larger or, on the other hand, smaller". But even setting aside those exceptions, and their exceptions too, it can be difficult to say exactly what leads to dwarfism (or gigantism for that matter).

In general terms, it seems that on islands animals shrink when resources are scare (exceptions abound). Cozumel is relatively dry, having only localized permanent fresh surface water. It's relatively rocky. And it's prone to catastrophic disturbance in the form of hurricanes. Potential prey for predatory raccoons, coatis and foxes would also be rather small in size - birds, whiptails, insects, and seashore creatures like crabs. No need to be large to subdue small prey. Perhaps these factors could lead to dwarfism, if being small made coping with island life more efficient. By way of interest, there is a fourth carnivore (in this case a carnivore that eats mostly fruit, go figure) on Cozumel, but it's not endemic: it's the Kinkajou (Potos flavus), a species that is widespread in the neotropics, though reportedly becoming rather rare on Cozumel. The providence of the Kinkajou is questionable, with some suggesting that it was only recently introduced to the island by humans. Kinkajous on Cozumel are not dwarfs.

Western Spindalis (Spindalis zena)

Photo: Laura Gooch

Most of Cozumel's endemic species and subspecies seem to have an ancestral affinity with the Yucatan Peninsula. One particular exception is the very striking Western Spindalis, a tanager-like songbird that, along with a suite of similar congeners, occurs across the Greater Antilles. I found a few Western Spindalises along the overgrown roads of an abandoned subdivision project on a pleasantly overcast morning. Well, it was pleasant right up until two highly aggressive, but thankfully also very stupid, feral dogs put the run on me for the better part of a kilometre. Nonetheless, I saw the birds, and was rather happy to do so, partly because of their unique distribution (it's the only place in Mexico where they regularly occur) and because of the interesting taxonomic quandary they present. All spindalis species, there are four of them, were once classified as conspecific. Now they've been split, with separate species on Puerto Rico, Hispaniola and Jamaica, in addition to the widespread Western Spindalis. As for their general placement among the other passerines, there's still some debate. I'm intrigued by incertae sedis, species whose place in the taxonomic order is confused at best, or just simply unknown. For many years, the spindalises were considered to be tanagers, indeed the whole complex of species and subspecies was called the Stripe-headed Tanager. But genetic and traditional comparative taxonomic approaches tell us that spindalises are not tanagers. We don't yet know where to place them instead though. If they're not tanagers, what are they? Time will tell, I'm sure, but for the present I was quite content to stare at a mystery, until I heard those damned dogs coming for me!

Feral dogs aren't just a problem for birders, but they're also a problem for the endemic island wildlife. Introduced species threaten Cozumel's biodiversity. There are the usual culprits, that afflict island ecosystems all over the world: Domestic Dogs (Canis familiaris) and Domestic Cats (Felis catus), as well as House Mice (Mus musculus) and rats. The newest threat on Cozumel though, seems to be the Boa Constrictor (Boa constrictor). Boas are found on the mainland, in fact not far from the port of Playa del Carmen I saw a pair of Northern Caracaras (Caracara cheriway) ripping apart a massive road killed Boa Constrictor. But boas never made it Cozumel on their own. They were apparently released from the set of some B-rated movie about 40 years ago. The boas reproduced quickly, feeding on the island's birds, laying waste to Cozumel's once abundant, and not uncommon, Yellow-lored Parrots (Amazona xantholora).

Another island bird that has declined precipitously is the Cozumel Thrasher, but the degree to which boas are to blame is uncertain in this case. The thrasher, once an iconic Cozumel bird, was locally common until Hurricane Gilbert came ashore in 1988. After that, the thrashers virtually disappeared. Researchers searched throughout the 90's, seeing only a handful of thrashers and even capturing some of the last survivors. Subsequent storms seemed to push the already small population even closer to the brink. The last definitive sight record was of a single bird in 2006, but since then there have been no confirmed observations. If not totally extinct, the thrasher is most certainly functionally extinct. That is to say, even if there are a few remaining survivors, they are unlikely to ever re-establish a viable breeding population.

Many biologists and naturalists have asked, why did Hurricane Gilbert and subsequent storms, such as 1995's Hurricane Roxanne knock back the thrasher population so severely? After all, didn't this endemic species evolve to deal with the catastrophic habitat alterations that result from the hurricanes and tropical storms which sweep the island periodically? Perhaps, on pristine Cozumel, before the introduction of cats, mice, rats and boas, the thrasher population would have been able to recover from a devastating hurricane. Just maybe, the toll taken by so many non-native predators in addition to the effects of hurricanes (not to mention other possible adverse factors such as anthropogenic habitat changes, or even an unidentified invasive disease), was too much for the thrasher to endure.

Those feral dogs that ruined my spindalis watching, were something of a blessing in disguise. They forced me to relocate, and it just so happened that I came upon a cenote, and one that was guarded by a rather large and statuesque American Crocodile (Crocodylus acutus) to boot. Cenotes are water-filled sinkholes, and they're often part of complex subterranean karst (cave) networks. The cenotes on Cozumel are sort of like islands within the island, because they are really the only permanent sources of surface freshwater. Cozumel certainly isn't a desert island, it's covered in vegetation, and rainfall is frequent, if not sometimes torrential. But the limestone bedrock and thin soils drain rainwater very rapidly, making cenotes the only reliable surface waters. Most of Cozumel's cenotes, including Aerolito, the one I'd stumbled upon, connect to one another through a series of erosion-carved tunnels. The cave system also connects to the ocean, meaning that most of Cozumel's cenotes are anchialine in nature: they contain both fresh- and saltwater. Because freshwater is less dense than saltwater, the lower reaches of Cozumel's cenotes are salty, while the surface waters are fresh.

I wandered around the Red Mangroves (Rhizophora mangle) which fringed the cenote, watching for more crocodiles and hoping to find the Pygmy Raccoons that left their tracks in the mud. Sure enough, after a little stealthy tracking and mud up to my knees, I spotted one endemic raccoon among the tangle of strut-like mangrove roots. It recalled a slightly smaller, slightly greyer Northern Raccoon (Procyon lotor), the species with which I am familiar back home. Happy with that, I turned my attention to the Great-tailed Grackles (Quiscalus mexicanus), those large and gregarious blackbirds, as they worked their way through the trees, and within centimetres of the basking croc. Brazen or calculating, I wondered? Small fish swam in the clear cenote waters, colourful and plentiful. When a Mexican couple appeared, I left them to enjoy the cenote and its guardian crocodile. It wasn't until after I returned to town and began reading, that I began to understand just how impressive this cenote actually was.

American Crocodile (Crocodylus acutus)

Photo: Mark Conboy

Without some pretty serious dive training there aren't too many options for exploring cenotes. Aerolito, as one of the largest cenotes on Cozumel attracts the attention of cave divers, who can travel more than a kilometre underground through chambers and tunnels decorated with stalagmites and stalactites. Luckily YouTube provides a glimpse of what the cenote looks like deep underground. The décor is nice, but the video shows only a single example of the supposed abundance of organisms that apparently inhabit Aerolito. Including the endemic sea star, Copidaster cavernicola (unfortunately not the species featured ever so briefly at 2:03 in the video). Endemic crustaceans live here too. Both freshwater and brackish water fishes swim here. Aerolito is sort of like an underground estuary, with its mixing of freshwater and saltwater ecosystems.

Don't let the tourist trap reputation of Cozumel turn you off from the island's wondrous natural history. The island's diving is noteworthy (unfortunately I didn't have time to get offshore on this trip), but the lesser known facets of Cozumel, the dwarfed carnivores, the endemic species, and the deep cenotes, are all worth putting up with the crush of Hawaiian shirts and Bermuda shorts. Cozumel is a surprise, waiting to be discovered. Just watch out for the dogs.

BioBrevia: Garbage Birds

White Stork (Ciconia ciconia)

Photo: Mark Conboy

Birders often use the term "garbage bird" to describe common, uninteresting birds. White Storks (Ciconia ciconia), not often thought of as garbage birds, are redefining themselves as such in some parts of their range. The Canadian Broadcasting Corporation's As It Happens reports on White Storks wintering at garbage dumps on the Iberian Peninsula.