A Trip to the Races
The name Dale Serjeantson has a journalistic ring to it — like an American crime-writer or a fictional detective. It’s a name that should be as well known to ornithologists as that of Howard Carter is known to archaeologists. Dale is also an archaeologist. She is a world authority on, among other things, bird bones. So, in 2004, when she unearthed the remains of Pterodroma petrels at three sites in the Western and Northern Isles of Scotland, she was not joking [1]. It turns out that more remains had been found in Scandinavia and Holland. Hold it right there. Although Pterodroma sounds quite like Pterodactyl, Pterodroma petrels are not extinct, although all their North Atlantic populations are perilously close to becoming so. They are the Holy Grail of seabirds.
In 2014 Dale pressed the nuclear button when, along with others, she published an article [2] discussing the ‘range collapse in a former British seabird species.’ The seabird in question was, as we call it today, Fea’s Petrel Pterodroma feae. The truth of the matter is that none of the 15 ‘Fea’s type’ bones so far discovered are especially ancient. Indeed some date from between AD 300 and AD 800. In other words, Pterodroma petrels were, presumably, once a regular sight around the coasts of Ireland, Britain and Western Europe and — so it seems — they probably even bred in certain areas.
As a group of seabirds, Pterodroma petrels are classed as ‘gadfly’ petrels. The derivation is a nod to the birds’ amazing flying skills. In shape, all species resemble shearwaters but with boomerang-shaped wings and a tapered rear. The set of the wings — pressed forward yet hooked back — seems to liberate the bird from conventional twisting, gliding and flapping. Pterodroma petrels zip by with a rolling action that pitches the flier higher above the sea than a shearwater. Like pride before a fall, each dizzy ascent is followed by a steep descent. Watching one passing among lines of Manx Shearwaters, I was struck by its weightless, arcing flight. A seabird made of balsa-wood.
The ancestral home of the world’s Pterodroma petrels is the Southern Oceans. Globally, there are around 32 species, of which a select handful breeds in the North Atlantic. As with many other seabirds, coming ashore to nest is their Achilles’ heel. Grounded and looking for a burrow or cavity in which to lay a single egg, they require predator-free locations. Mostly, that means remote islands. And, belt and braces, arrivals and departures are conducted in darkness. When those criteria are met, many thousands can call one island home. Talk about putting all your eggs in one basket. Yet, until humans came along, the system worked perfectly.
Despite the limited number of breeding retreats, evidence shows that, just 500 years ago, there were millions of Pterodroma petrels flying around the North Atlantic. In the western hemisphere, Black-capped Petrel P. hasitata was abundant, with breeding concentrations on several Caribbean islands. Nowadays the species is drastically reduced and breeds only on steep, forested mountains on Hispaniola. At least Black-capped Petrels are still with us. Jamaica Petrel P. caribbaea was wiped out. Both species suffered a cocktail of woes from habitat loss, hunting and introduced predators, the worst of which were cats, rats and — on Jamaica — the Indian Mongoose. From an Irish perspective, neither species is likely to have shown up in our seas. Or did they?
The possibility is debatable. Pterodroma petrels, blessed with almost ‘super avian’ globetrotting powers, travel vast distances. They do so, not only after the breeding season, but during it. Geo-locator tracking studies from extant species in the North Atlantic — such as Bermuda Petrel P. cahow, Zino’s Petrel P. madeira and Fea’s Petrel P. feae — show that foraging trips dedicated to rearing a chick normally exceed 2,000 km and take more than a week to complete. Longer trips extend to 16,000 km and last 22 days [3]. Bermuda Petrel, that nests only on a few islets off the coast of Bermuda and whose current world population stands at around 300 individuals, has reached Irish waters (photographed 270 km west of Co. Kerry) as recently as May 2014 [4].
Bermuda Petrel provides a classic example of a once common bird that we never knew. According to Bob Flood’s harrowing account [3] of the bird’s near-total demise, there were possibly more than a million Bermuda Petrels when Christopher Columbus sailed from Spain and landed in the New World in 1492. Lying in the path of Spanish galleons, Bermuda was discovered not much later. Not only did the Spanish sailors see the land, offshore at night they could hear the sound of swarms of petrels milling over the archipelago. The birds’ hiccuping sound, rendered ‘Cahow’, stuck its English name until fairly recently. The sailors released hogs to serve as a future food source. The hogs dug up and destroyed the petrels’ nesting burrows. But that was just the start. Finding the bird to be ‘incredibly abundant’, settlers collected eggs and ate adults and young. With people came rats that joined in the feast. Rats also ate the settlers’ food and crops. So cats were introduced to control the rats. That too failed — and another predator was added to the list of killers. By 1616 the bird was so rare that Bermuda’s governor prohibited further collecting — unfortunately local rats and cats could not read — but it was too late. Miraculously a few, estimated at 17 or 18 breeding pairs, survived on small islets, out of reach of people and predators. For more than 400 years the birds clung on until discovered and protected in the 1950s. Recovery has been painstakingly slow because, for one reason, seabirds do not breed until they are several years old and only rear one youngster per annum.
A little over 40 years ago, Pterodroma petrels were not on the radar of any birdwatchers in Ireland or Britain. Unless, as I did, you chanced upon a mention in The Handbook of British Birds [5] of old records of three different Pterodroma species (two from the Southern Hemisphere) from the nineteenth and early twentieth century in England, you would never have heard of the genus. Moreover, there was a whiff of suspicion about some of the English records. Where did the specimens originate? Then, in 1973, a European field guide appeared [6] that showed a breeding dot for Soft-plumaged Petrel Pterodroma mollis on Madeira, with a highlighted zone around the island indicating the bird’s at-sea distribution. The species was illustrated in a plate containing Fulmars. Not to put too fine a point on it, it looked like a Fulmar’s ugly sister. In 1977 the first brick-heavy volume of The Birds of the Western Palaearctic (BWP) was published. Text in the latter indicated that the breeding distribution of Soft-plumaged Petrel spanned Madeira and the Desertas (three islands a short distance east of Madeira) and, 2000 km further south, the archipelago of Cape Verde. Overall, the population was tiny, especially on Madeira, and — a familiar story — ‘some hundreds’ breeding at Cape Verde suffered ‘considerable human predation’ because they were slaughtered for their body fat that was used ‘medicinally’, a euphemism that amounted to superstition.
Nonetheless, over a global breeding range scattered across islands in the Roaring Forties, Soft-plumaged Petrel was common. In effect, this was a southern hemisphere seabird with a breeding foothold in the northern hemisphere. Under ‘Geographical variation’ BWP distinguished the three populations as different races of one species (a synonym of ‘race’ is ‘subspecies’, which has superseded the term race in recent literature). In taxonomic terms, southern hemisphere birds were classified as Pterodroma mollis mollis. The third word, known as a trinomial, designates an individual’s race or, as I prefer to use from this point on, its subspecies. What criteria formed a basis upon which to classify each type of subspecies? The rationale, as outlined in BWP, rested on size. Plumage was deemed ‘very similar … in all three races’ but size mattered. The largest of the three, in terms of wing length and bill size, was the population breeding on Cape Verde — P. m. feae. At the other end of the scale, the birds breeding on Madeira — P. m. madeira — were the smallest.
Hindsight, as everyone knows, is a wonderful thing. It is easy to be taken in by sumptuous authoritative tomes which, like BWP, proclaim credentials based on ‘large editorial teams, many experts … and co-operation with the editors of specialised works’. However, the books were wrong. The plates of BWP depict Soft-plumaged Petrel alongside the largest shearwaters that occur in the North Atlantic, such as Great Shearwater. In terms of size, Soft-plumaged Petrel — irrespective of whatever subspecies — was shown to be similar. If you read the text, however, you discover otherwise. In fact, Soft-plumaged Petrel is much smaller and equivalent in size, although not build, to Manx Shearwater. Then, during the afternoon of 5th September 1974, some kind of Soft-plumaged Petrel passed the tip of Cape Clear Island, Co. Cork [7]. A ghost no longer.
Back in 1974 news travelled slowly. Details of the bird’s appearance did not make it into print until 25 years later. In a way that did not matter because no one would have been much the wiser as to where it came from. At the time, informed guesses favoured a waif from one of the Soft-plumaged Petrel colonies in the southern oceans. So little was known about the bird’s status in the North Atlantic that debate seemed pointless. Could it have been a long-lost survivor from the former breeding sites identified through bone remains discovered by Dale Serjeantson? Imagine that! If Bermuda’s Cahows, presumed extinct for over 400 years, had survived, then perhaps some European Fea’s-type petrels might have done the same? Alas, this is not the case. Dale Sergeantson et al. concluded [2] that the ‘northern colonies’ that probably existed in Scotland, Scandinavia and Holland, were lost due to ‘extirpation … possibly associated with human-mediated invasive mammal introduction across northern Europe.’
Once again, rats, probably Black Rat Rattus rattus, also known as the Ship Rat, were singled out as the likely angel of death. Recent research on predation by rats on seabirds — using ingenious methods of live-sampling the rats to determine their diet from isotopes of prey in the rats’ tissues and gut — concluded that small colonies of burrow-nesting seabirds are at greatest risk of being eradicated [8]. But the hard facts will never be known for sure. Indeed, it is not too outlandish to speculate that populations of Fea’s-type petrels might once have bred on some of the great seabird cities used nowadays by other nocturnal breeders such as Manx Shearwaters. A similar notion is expressed by Dale Serjeantson: ‘our results suggest that surviving Macaronesian populations of Fea’s Petrels represent the final remnant of a wider Holocene [our current epoch, taking the end of the last ice age, roughly 10,000 years ago, as a time reference] distribution of this species complex, with the Madeira and Cape Verde archipelagos constituting the final sanctuary of Pterodroma in the North Atlantic.’
NAME GAME
During the 1980s, two sleeping giants awoke. The first was a renaissance in re-thinking the species relationships between southern hemisphere Soft-plumaged Petrels and their superficial equivalents in the North Atlantic. It started to emerge that more differences existed between them than had been realised. One plumage distinction, that only southern hemisphere populations had a band across the chest, established that the bird that passed the tip of Cape Clear was from one of the North Atlantic populations — because it lacked a band. Eventually, in 1983, Bill Bourne, a long-standing authority on seabirds, proposed treating ‘Soft-plumaged Petrel’ as three species, and not just one type of petrel with three varieties of subspecies [9].
In the early twenty-first century, ornithology has been overrun with scientists using molecular methods — the examination of genetic signatures — as an allegedly more objective way of establishing the evolutionary history of birds and the relationships between them. Although this new approach has been illuminating, it has also become increasingly impenetrable for a lay person to understand. Genetic history provides fascinating insights but it may not manifest itself in the outward appearance of the birds themselves. Luckily, in terms of ‘look and feel’, the genes of all three Pterodroma petrels back up their evolutionary history [2]. So the three-way split became, with a lot of name-changing along the way, Soft-plumaged Petrel Pterodroma mollis, Fea’s Petrel, P. feae and Zino’s Petrel P. madeira — named after a father and son, Alexander and Francis Zino, two inhabitants of Madeira, that have battled to save the bird from extinction [10].
The other force that awoke was small bands of birdwatchers staring for hours out to sea, forever hopeful that among the passing hordes, a winged rara avis might pass. Maybe, as sometimes happens, an albatross. For a lucky few, the reward was even better.
During the month of August and the middle of September in each of the four years between 1989 and 1992, no less than 11 ‘Soft-plumaged Petrels’ were seen from headlands located on the tips of Devon, Cornwall and Humberside and, in Ireland, Co. Cork and Co. Down [11]. No seabird creates a greater gulf between agony and ecstasy. You have to be on the spot when one passes. Pterodroma petrels do not hang about. If the bird is distant and pitches into a trough in heavy seas never to reappear, companions sitting next to a jubilant observer may be unable to share a once-in-a-lifetime experience. I have plumbed those depths. Nowadays, a few are seen every autumn. How did we miss them? Maybe the population has increased and more are passing? No, if anything, their numbers are still falling. Detection seems to be down to raised expectations and concentrated effort at what we now know to be the peak time — August. Superior optics, principally telescopes, has brought them into our ken. From an Irish perspective, most observers — myself included — were also looking in the wrong place. The plates in BWP created the impression that, being a large seabird, one would, most likely, be in the company of ‘big shearwaters’, meaning Great, Cory’s and Sooty Shearwater — species that pass at longer ranges. So that is where we looked. Nobody really thought to check among the endless traffic of Manx Shearwaters closer to shore. And that, quite often, is where they are.
Can you guess what is coming next? You do not have to be a genius to work out that, because human curiosity likes to identify things and ascribe a name, the hunt was on to determine the precise species of Pterodroma passing Ireland and Britain. By the new millennium photographs added more evidence. The upshot was that none — so far — hail from the southern hemisphere. Hence, it was no longer appropriate to refer to a sighting as a Soft-plumaged Petrel P. mollis. Any that were photographed were deemed to be Fea’s Petrels P. feae. Meaning that Zino’s Petrel P. madeira was safely excluded?
Ornithological authorities are not immune to the Health & Safety disease. In some quarters the new wave of records have been branded ‘Fea’s-type petrel’. This implies that the birds were ‘either-or.’ Although it is not impossible that a Zino’s Petrel could appear off our coasts — in fact tracking data show that some Zino’s Petrels are not too far west of Ireland during the birds’ breeding season [12] — the bird’s low population level reduces the odds considerably. And, of course, Zino’s Petrel ought to be distinguishable if seen under favourable conditions, especially close alongside a boat.
A gold rush of interest erupted in the late twentieth and early twenty-first century into the conservation and field identification of the North Atlantic’s Pterodroma petrels. It became feasible to take trips to Madeira where, one way or another, it was possible to see one of the fabled few from a headland or ferry. Numerous photographs and illustrations in journals and field guides — such as Lars Jonsson’s Birds of Europe (1992) — brought the birds’ appearance to wide notice. Then, in 2010, a landmark 36-page paper by Hadoram Shirihai, Vincent Bretagnolle and Francis Zino [13] nailed the identification at sea of all Pterodroma petrels breeding in the North Atlantic. Subsequently, a multi-media book written by Bob Flood and illustrated by Martin Elliott and Ashley Fisher [3] served as icing on a cake that has, by now, been exceedingly well baked.
MUHAMMAD GOES TO THE MOUNTAIN
Autumn is a time of wanting and waiting. Ever since I clapped eyes on what we now know is Fea’s Petrel — on 20th August 1991, off the Irish Sea of Co. Down — long waits have punctuated the gaps between the five others I have seen, all of them off Ireland in August. The last, in 2018, was on a seabird trip that lurched out to sea from Baltimore, Co. Cork. This yielded my closest Fea’s that was glimpsed during dire conditions. Paul Connaughton, the trip organiser, spotted it just before it crossed the stern and swept past like a Ferrari overtaking a tractor. Through a curtain of rain it felt like an apparition but my view was more than adequate to see its Boxer Dog face and meat-hook bill. Buoyed by having seen the ultimate seabird, no one minded when the trip was abandoned. For me, however, the fight was over. I thought about this all the way home. I decided that I had enough of brief encounters. I cracked and booked a week in Madeira in June 2019 with ‘Windbirds’ — professional ornithologists that take you to Pterodroma heaven by fast boat.
My dream was to drool over a selection of Fea’s Petrels rather than squander more years waiting for a lone shooting star. To allow for weather cancellations, you don’t book for just one trip with Windbirds, you sign up for three all-day trips spaced over five days. Zino’s Petrel, rather than Fea’s Petrel, is the big draw. But not for me. Having followed the elucidation of Fea’s Petrel for nearly thirty years, I had been conditioned to believe that, unless a closely-related — but much rarer — Zino’s Petrel was around long enough to be photographed in forensic detail, I would not be seeing a ‘new’ seabird: ‘distinguishing [Fea’s and Zino’s Petrels] at sea is well nigh impossible without extremely good views and high definition photographs.’ [13] I assumed that the scale of difference between Fea’s and Zino’s, if such a thing could be determined on a lone bird, would be like seeing a migrant Wheatear and, thinking it a bit on the big side, speculating that it was a ‘Greenland’ Wheatear. That is, a Wheatear drawn from a subspecies population that, because it performs a trans-oceanic migration, is larger-bodied and longer-winged than other Wheatears. But nonetheless, a Wheatear.
I was one of nine tour participants. The others were English guys who knew each other and, unlike me, several had at-sea experience of Pterodroma petrels. Having seen most of the A to Y of the world’s seabirds, they were here for Z. As often seems to be the case with a bunch of Brits, they were all called Pete or Steve. Anyway, I didn’t become chum and we were a happy crew. Windbirds is a two-person company. Hugo Romano and Catarina Correia-Fagundes have been running Pterodroma trips for 15 years. They also work on Madeira as part of the conservation effort to protect Zino’s Petrel. In other words, their expertise is unparalleled.
At the start of the first day, I was standing alongside Pete and Steve — yes, all eight of them — overlooking a marina that contained an impressive RIB (rigid-hulled inflatable boat). We reckoned, indeed hoped, that this was to be our vessel. As departure time neared there was no sign of Windbirds. Then with a great whoosh, an even bigger RIB appeared around a headland and sped straight at us. It was Miami Vice come to life. Our commander was Captain Catarina. Hugo was in charge of chum. I think Catarina enjoyed the matriarchal role — shepherd of a crowd of landlubbers on the high seas. Over the roar of the engines Hugo explained the day’s plan. Given that, in June, Fea’s and Zino’s Petrels are in the early stages of the breeding season, some are likely to be inshore during the day, prior to visiting breeding sites at night. This is not because they feed around the islands where they nest. Detailed tracking studies [12, 14] pinpoint the prime feeding grounds as a wide region of deep water located north and west of the Azores. Our hope was that, especially in the late afternoon, we might intercept a bird heading towards the coast. Or, we might encounter one or more loafing on the sea, parked next to flocks of other seabirds. Although we had chum, Hugo said that neither Fea’s nor Zino’s are particularly attracted to it. Perhaps, they are not hungry. But the smell can make them curious. We would be stationed near chum slicks for lengthy periods, hoping that a Pterodroma, piqued by the scent, might arrive.
All cruises have moments of tedium, but usually the passengers can relax on sunny decks or lounge by a pool. Each of us was strapped to a springy seat — to cushion the impact of bouncing waves — that felt like a vaulting horse. Although all eyes were supposed to be on duty, Catarina and Hugo were the chief spotters. When Catarina cut the engines, everybody was alert in a flash. What had been seen?
The sense of adventure was intoxicating. We slipped past sleepy flocks of Cory’s Shearwaters in siesta mode. Small clusters bobbed up and down on royal blue water. Too close and we caused a domino effect. One by one, the whole flock taxied into flight. Seen off an Irish headland, the slow steep-sided roll of a Cory’s Shearwater would be a source of great excitement. Around Madeira, they were the staple. A few Manx Shearwaters provided a useful benchmark. These, Hugo informed us, were local breeders and not migrants from cool northern latitudes. Although it was Nirvana to be among such splendour, I suspected that the main motive was to look for Pterodroma. The word was a metaphor for big game.
Schools of dolphins splashed by and Sperm Whales lolled on the surface, generating a glassy wake over which Bulwer’s Petrels swooped like satanic spirits. Captain Catarina was the first to find quarry. I have no idea how. She clocked a Pterodroma one-armed with binoculars while we tumbled over wave crests; men holding on with both hands. Details were scant — dark underwings, white body and a dark, cowled head. The élan flight was a clincher: fast, effortless, cavalier. The methodical momentum of Cory’s was eclipsed by a freestyle Top Gun. As for Manx — they seemed magnetized to the surface, obliged to fly low, their wing-beats rickety compared to the trapeze of a Pterodroma.
‘Fea’s’ was the one word that came from behind the tiller when we swung around to wonder if Catarina had identified the bird to species. It certainly reminded me of those I had seen before. The stocky build, fat face and ‘blackened-in’ Fulmar bill were déjà vu. Although Fea’s is not much bigger than a Manx Shearwater, the wings are more rangy and Spitfire-like and the foreparts are bull-necked; a bird whose physique outstrips its size. I don’t recall who found the next but by the fourth of the afternoon everyone’s search image had been trained. Fea’s Petrels may be a lot smaller than a Cory’s Shearwater but their flippant upswing integrated with a gravity-defying descent meant that they were ‘pro’ fliers among a bunch of Sunday drivers. Windbirds indeed.
Two moored dots — too small for Cory’s — diverted the boat into a wall of disorderly waves. At such times binoculars are best tucked away and pre-emptively wiped in readiness for action. Everyone waited for Catarina to hit the breaks. When she did I beheld something I had never seen before — a Fea’s Petrel sitting alongside a Bulwer’s Petrel. It was darker-headed than I expected, and its grey back was subsumed into lengthy wings that extended way beyond the tail. Its pose was surprisingly upright and, as ever, the bill stood out. A voice in my head said ‘Mediterranean Gull’: a subconscious likeness suggested by the latter’s combination of (in non-breeding and immature plumage) a black facial mask and heavy drooping bill.
Tea and sandwiches were handed out. Usually this is a cue for a bird-of-the-day to show up and create chaos. But we were left in peace to enjoy the refuelling stop and rest our eyes, apart from watching Cory’s Shearwaters gliding close by, their bill colour changing from yellow to pink as the sun dipped and bathed everything in a rosy radiance. The moon appeared and hung as a white dot in a pastel sky. Tea over, the engines rumbled back into life and we were back at work.
I had grown accustomed to my place at the back of the boat. I had room there. But if anything appeared off the bow, I was last in the queue. When a greater than usual commotion gripped the front, I was all ears. There was shouting and Hugo’s name was yelled for help — with identification. The hullabaloo was because the latest Pterodroma, zipping along dead ahead, was a titch. I tried to see it. Catarina gave chase. The suspect was not alone. A kettle of birds was in the area — Cory’s, Manx and Bulwer’s. There was even another Pterodroma that was ‘dismissed’ as a mere common-or-garden Fea’s. Pterodroma petrels do not obey the rules of gravity. A pivot to one side does not mean the bird will continue in that direction. Although the boat was capable of keeping up with the quarry, it could pop up anywhere. Near the stern, I saw the head-on silhouette of a medium-sized shearwater swing low into a trough. It was a Manx Shearwater — just a scare. All quarters of the sea were frantically scanned. I decided to pick one tangent and stick to it. My Manx Shearwater planed up from a canyon of water. This time it had its lights on. ZINO’S! It shot past the bow. Everybody seemed to be on it. Shutters whirred. Catarina locked-on and managed to synchronise the boat with the bird’s swish.
Unlike my previous Pterodroma experiences in the eastern North Atlantic, this individual could be watched relatively easily. For a few blissful minutes, we became part of its world. Our speeds matched. Although it was hard not to take photographs and just look, it was around long enough to leave a permanent record that transmitted from the retina along the optic nerve to the brain. I can still see it in my mind’s eye. Everyone on board was aware of the difficulties of distinguishing between Zino’s and Fea’s Petrels. Hugo had reiterated the nature of the challenge. Now what? Those that managed to get the best images converged at the helm; Hugo was cast in the role of King Solomon. Hold it right there. My brothers-in-arms were not rookies. Anything but. I kept my own counsel, but I was curious to know if a consensus was going to emerge. Hugo and Catarina’s trips are the only bird tours that offer a chance of seeing Zino’s Petrel. The last thing Windbirds is going to do is drop the ball and endorse a claim that is less than bullet-proof. Fingers zoomed in on images that revealed the relative size and shape of the bird’s undeniably small bill. There seemed general agreement. Based on bill structure, the bird would pass as a Zino’s Petrel.
PARADISE LOST
Few people that travel to Madeira will think of the destination as once part of a moist, subtropical paradise. Yet, before humans arrived, forest covered most of the island from sea level up to around 1,500 metres. Most of it is gone, cleared by Spanish and Portuguese settlers. Madeira was not inhabited until 1419 and a written account [15] describes how the island’s conquistador, Zarco, sent men into the forest to ascertain if there were any wild animals, but none were found: ‘only birds of various kinds, which were taken by hand because they were not used to seeing people.’ Large areas of the forest were razed in a fire that lasted seven years. Populations of breeding seabirds were devastated by the introduction of rats and cats. To this day, feral cats are the biggest threat to the few remaining Zino’s Petrels. Between six and ten cats are trapped on the breeding ledges every year [3]. We will never know how common Fea’s and Zino’s Petrels once were but a visiting German archaeologist, Harald Pieper, recorded the bones of both in caves used, presumably, for nesting [16]. In the same way that Bermuda Petrels were eradicated from the main islands of Bermuda but survived on tiny uninhabited enclaves, maybe that is why Madeira’s Pterodroma petrels occur only on a few mountain ledges or on the barren, uninhabited desert island of Bugio? In this sad tale, it is easy to overlook a salient detail. Based on differences in bone size alone, Harald Pieper established that two species were involved.
If I had not seen a Zino’s Petrel with my own eyes I would have thought that the bird was a mirage. Why? Because it did not fit the image, perpetuated by at least two ostensibly reliable seabird references [17, 18], of what Zino’s Petrel looks like. Maybe this singleton was an exception — an example of an end-point in variation with the smallest possible bill? Harald Pieper did not, however, find a range of skeletal remains that amounted to variation in one species. He was able to distinguish two species without seeing a single feather.
My trip to Madeira was devoted to Fea’s Petrel. I knew that Zino’s Petrel might materialise too, but I had been conditioned to believe that the separation of the two species was Mission Impossible. I was less than excited at the prospect of seeing a seabird for which I needed a DNA sample to reach a diagnosis.
If bill proportions are the only means by which Zino’s Petrel can be told from Fea’s Petrel then I’m a rabbit. I suspect that you, dear reader, are anticipating a long list of caveats that will reduce the mooted identification to some kind of nit-picking? I can only speak for myself and — of course — I am open to the accusation that anything I say is based on just one Zino’s Petrel. Not so fast! We are still at Day One in this account. In the course of two more days at sea, other Zino’s were seen and some were photographed. Hugo became a happy bunny and, at the end of the trip, awarded each of us a little lapel badge; reserved for those lucky enough to see the seabird behind Windbirds trips. Trip reports on the Windbirds website provide a treasure trove of images of Zino’s Petrels. The way to access this resource is to search ‘madeirabirds.com/trip_reports’. Not all reports contain images but those that are listed under ‘Zino’s Petrel Pelagic Expedition’ often do, such as ‘Zino’s Petrel Pelagic Expedition 17–19 June 2014.’
I finished up with no qualms concerning the possibility — not impossibility — of the identification challenge. Just because, most of the time, you will not be able to be certain, should not mean that the task is hopeless. If I was sea-watching in autumn in Ireland and saw a lone juvenile skua at, say, one kilometre range, I would be loathe to decide if it was a Pomarine, Arctic or Long-tailed Skua. But if I was on a boat and saw the same bird at 20 metres, it would be a straightforward process to determine its species, especially if it was photographed. To me, Zino’s and Fea’s can be afforded a similar degree of ‘identification feasibility’. A vast amount of time and words has been invested in this topic by, among others, Hadoram Shirihai [13] and Bob Flood [3]. Their wisdom is peerless and there is no doubt that, given the exigencies of distance and ‘size creep’ between, say, the smallest Fea’s and the largest Zino’s, only the best candidates will make it through the screening process. But that is a far cry from writing off the lot — or creating a myth that amounts to the same thing [17, 18].
In his exhaustive exposition in Birding World [13], H Shirihai worked with trapped individuals at breeding sites and measured everything measurable. Let’s look at one set of biometrics — wingspan. For a total of 43 Fea’s Petrels (13 from Bugio, 30 from Cape Verde) the wingspan varied from 860 mm to 943 mm. Just as an aside, the shortest and longest measurements both came from Cape Verde Fea’s. For a total of 6 Zino’s Petrels, wingspan varied from 800 to 843 mm. Hence, by a margin of 17 mm, there was no overlap between the longest Zino’s and the shortest Fea’s. Although 17 mm is not a huge difference, it amounts to a tectonic fault-line dividing two species. More telling is the difference between the average wingspan of Fea’s (irrespective of whether the measurements are from Bugio or Cape Verde) of 908 mm, versus, for Zino’s, 829 mm. That is 79 mm or, in old money, more than three inches. No wonder I considered Zino’s Petrel to be smaller than Fea’s Petrel with shorter wings.
As a yardstick, the wing lengths for Iceland Gull and Glaucous Gull, two species that are almost identical apart from size and structure, have a wing-length overlap — although, on average, there is a difference between the two [19]. Bob Flood’s text [3, page 162] states: ‘Zino’s, structure: medium-small and particularly lightly built — a miniature version of Fea’s Petrel. Some birds are strikingly small, slim-line, and petite in most proportions (probably females and juveniles).’ To give you another feel for a well-seen Zino’s Petrel versus a Fea’s Petrel, I likened Fea’s to a Ringed Plover (bull-necked with a stubby fat bill) compared to, for Zino’s, a Kentish Plover-like smaller head and slim bill.
Madeira sits in the blustery track of the northeast trade winds. Every day brought slightly different weather but there was always wind. Except one. It looked like the third and final trip would be conducted in conditions reminiscent of a Lake District cruise. Not so. We were cancelled. No wind, no Pterodroma. I have to admit, I would not have minded the chance to check. However — ‘good news’ — the next day was forecast to be blowy. As the boat picked up speed and headed into a steeplechase of white-capped waves, I decided not to unpack the camera but, instead, put on every scrap of wet gear I had. Even if a front seat were available, I wouldn’t want it. Minutes later I regretted those decisions. A Fea’s Petrel appeared out of nowhere and seemed to think it was a dolphin, riding the bow-wave so close that you could see the metallic gleam of its bulbous black bill. But not for long. As we tried to keep up, the sea proved too big for high-speed pursuit. Drenched but happy, stalwarts abandoned pole positions. I decided that, even if too risky to use a non-waterproof camera, I would be better off as the sea’s dancing partner. I grabbed a vacant front pew. It quickly became clear that there was action ahead. Arcing shapes contained at least one Pterodroma. Even though we had just left port and were en route to a more distant hot-spot, Hugo cut the engines and released chum. We hit the jackpot. For the one and only time, Fea’s Petrels came straight to us. Like moths to a flame. At times three were together and, apparently, a total of four was around us. The birds were hungry as they swooped and pitter-pattered on the sea, frantically gobbling up scraps of fish. Their feeding actions suggested a Fulmar that had been to ballet school. One swung past at arm’s length. A few lazybones Cory’s departed. For a while, the only birds alongside were Fea’s Petrels. An erstwhile fantasy was dangling right in front of my nose. Finally, I had a vivid idea of what Fea’s Petrel looked like. Even though the livery is composed of shades of grey, the contrasts were exploding in what felt like full, sunny colour. I could have stayed with them all day but Windbirds wanted more Zino’s. So, we left. The wind abated. The rest of the voyage passed like an ascent into heaven. A day when naked-eye views beat binoculars.
HOW TO FAKE A SPECIES
Pete Morris, a professional bird tour leader, had some bad news. But more of that shortly. Pete was the one person among us that had kept up with the latest round of ‘lumping and splitting’. What is that? Scientists are able to examine the DNA of all living organisms — and even some long-dead ones — and, on that basis, reconstruct evolutionary history or ‘phylogeny’. The point is to reveal relationships between species, especially from which branch of the ‘phylogenetic tree’ a species has descended. Ancestors can therefore be traced, and the lineage of species established. Such techniques are a powerful tool that allows life on Earth to be ordered or ‘classified’. And that is called taxonomy.
Where does lumping and splitting come in? Maybe using an example would be the best way to explain. On the Aleutian Islands off Alaska there lives a big, swarthy species called Song Sparrow. The same species occurs throughout North America’s various geographical regions and, like the disparate zones where it lives, the bird’s appearance varies widely. Hence, in contrast to the largest and darkest population on the Aleutian Islands, desert-dwelling populations in southern California are only half the size and their plumage is light-coloured to match the habitat. If you held one individual from each population in your hands, you would be convinced that you had two different species. In reality, you are looking at extremes in a range of variation. In fact, there are 29 distinct varieties — subspecies — of Song Sparrow in North America [20]. Even today, the whole lot is still regarded as one species consisting of geographic populations, each only slightly different from its nearest neighbour. In other words, the various types are lumped together. Although this unifying approach makes sense, the evolutionary history might tell a different story. I don’t know if it does. If it did, ‘splitters’ could look at the array of types and divide them up into possibly as many as 29 species, each connected but with small degrees of difference.
Although this has not happened for Song Sparrow, it has for many other birds. Before taxonomists homed-in on genetics as the be-all and end-all to determine species, other pointers were used in the past. Such as whether interbreeding occurred between two dissimilar-looking populations; in which case they were regarded as one type, albeit with more than one standard appearance. On the other hand, if their breeding ranges overlapped but they never interbred, they were said to behave as ‘good species’. In which case, they were deemed to have grown apart over time — evolved — and developed into separate species.
The field of taxonomy is infamous among biologists for its intractable arguments. Right from the eighteenth century, when Linnaeus came up with the ‘binomial’ system that we still use today, biologists from all sides — botanists, zoologists, ornithologists — have debated how to order the natural world. They are still at it. The only difference is that, nowadays, the tail that wags the dog is DNA and its new pup, phylogeny. As a criterion, DNA is an irrefutable marker; biological litmus paper that answers the question, who do you think you are? Importantly, unlike previous pronouncements made by ‘authorities’, DNA-based investigations are capable of being tested. However, just because DNA may differ — by a small amount — between two discrete, isolated populations, the living birds may be blissfully unaware that they might be on an evolutionary path which, in thousands of years, may result in one group not recognising the other ‘as family’. Such scenarios often stem from the fact that two birds can closely resemble each other — either because they are descended from a common ancestor or because, although not closely related, they have each adapted to the same environment and lifestyle [21]. Put another way, species are best defined by a combination of an organism’s evolutionary history and its ecology.
Maybe an example is called for? White-faced Storm-Petrel Pelagodroma marina is a small seabird — much smaller than Pterodroma — that occurs across the cool waters of the South Pacific and South Atlantic. It is also found in parts of the North Atlantic. Its presence there is believed to have come about when, as global climate altered after the last Ice Age (roughly 12,000 years ago), cool currents off West Africa became stronger and spanned the equator — thereby creating suitable feeding conditions in parts of the North Atlantic for certain seabirds that extended their range into it from the South Atlantic [22]. When a taxonomic spotlight fell on the global reach of the species, it became clear that populations varied across oceanic regions and could be differentiated.
Here is a paraphrased summary, ‘populations breeding in high latitudes are dark and heavily marked, while those breeding in lower latitudes where cool water wells up to the surface, are progressively paler and have larger bills, wings, tarsi, toes, and shorter tails with a smaller gradation. … The series demonstrates the sexual dimorphism [difference in size between sexes] and the progressive loss of pigment, increase in size, and shortness of the tail in lower latitudes. … The birds from [the South Atlantic] and New Zealand appear indistinguishable since they have a similar natural history and exploit a continuous zone of surface water along the subtropical convergence, [whereas] the birds from Macaronesia [Tenerife and the nearby Salvage Islands] are smaller and darker than those from the tropical Cape Verde Islands to the south.’
Not surprisingly, given the birds’ attachment to areas of the world’s oceans where food supply is seasonal, the various populations also breed at different times of the year. So, even today, the taxonomic approach is to treat White-faced Storm-Petrel as one species, consisting of six recognisable subspecies that differ, in small measure, in size and appearance [23]. I expect that, because White-faced Storm-Petrels breed on remote islands with each sub-population remaining loyal to its own ‘ocean’ and breeding site, measurable genetic differences will, over time, have arisen between each ‘tribe’. Thus far, no relevant genetic studies have been carried out on White-faced Storm-Petrel [24].
But if the theory that White-faced Storm-Petrel spread from the southern hemisphere into the northern hemisphere is true, then there is a good chance that the process might be corroborated from genetic evidence — because DNA provides an insight into what is called ‘evolutionary radiation’. In this way, researchers can connect the outward appearance of a bird to the hidden history preserved in its genes. Also underlying this process is something called ‘genetic drift’; an historical record preserved in DNA that shows the rate at which differences arise between populations which, for a variety of reasons, start to go their own way. This phenomenon has been well-researched in other seabirds [25] and in some landbirds. An opportunity to jump the gun and study genetic drift in action was provided for researchers that studied the rate of genetic divergence that has arisen between populations of Japanese Bush-warblers in Hawaii. This species was introduced to just one Hawaiian island (Oahu) in 1929 but, by the 1990s, had colonised all the other main islands. Researchers found that genetic drift had occurred and was most marked — but only at a genetic level — between the populations on the islands furthest from the founder population on Oahu [26].
Harking back to Charles Darwin’s On the Origin of Species in 1849, evolution of living forms — birds, animals, plants, fungi, micro-organisms and even Homo sapiens — is a fact, not a theory. Hereditary traits link all life-forms to the past and, if reproduction occurs, the next generation. It is, however, important to draw attention to another facet of evolution. This is the part that Darwin called ‘natural selection’, meaning the differential survival and successful reproduction of some combinations of hereditary traits. In this way, tiny changes can occur and — ultimately — drive populations apart and create new species. Drawing the line between divergence and ‘species-hood’ is not an exact science. It is opinion. Science proceeds by testing hypotheses and by conducting experiments that may or may not provide evidence that supports the original hypothesis. Or so you would like to think. Heaven forbid that a new generation of jargon-ridden splitters is, ‘not only excessively dividing up the living world but clogging the taxonomic literature with unjustified and useless names, and all to satisfy their own egos.’[27]
What was Pete Morris’ bad news? It was that, on the basis of genetic research [28] and also the discovery of vocal differences between breeding populations [18], the British Ornithologists’ Union — guardians of the official list of all species recorded in Britain — had followed a recommendation made by the International Ornithological Congress (IOC) and had decided to ‘split’ Fea’s Petrel into two species:
‘Fea’s Petrel Pterodroma feae is being split into Pterodroma feae and Pterodroma deserta [Desertas Petrel], following published studies (Robb et al. 2008, Jesus et al. 2009), in particular highlighting differences between these taxa in breeding seasonality and vocals.’ [29]
The implications for all field observers in Britain and Ireland (because Ireland also follows taxonomic decisions made by the IOC) are, in the BOU’s own words: ‘Fea’s Petrel [Pterodroma feae] and Desertas Petrel Pterodroma deserta are now treated as separate species, whereas before they were considered subspecies of P. feae. As none of the six British records separated P. feae from P. deserta, P. feae is removed from the British List. [30]
In other words, even if you see, in ‘old money’, a Fea’s Petrel at point-blank range and photograph it, you will no longer be able to establish if the bird is, in ‘new money’ a Fea’s Petrel or a Desertas Petrel. This is because the appearance and measurements of the two newly-declared ‘species’ overlap [13]. But their genes differ — just a little. Numerous other scientists regard the ‘split’ as dubious. Here is how a team of researchers that worked for ten years with Fea’s Petrels on the Cape Verde Islands mentioned the decision in 2017: ‘… a split between [Cape Verde Fea’s Petrel] and [Desertas Fea’s Petrel] remains more controversial. For example, biometric differences between [Cape Verde Fea’s and Desertas Fea’s Petrel] are unclear (13), have not been formally tested and may be obscured by sexual size dimorphism, since no study reported biometric measures of males and females separately.’ [31] I could, if I bothered, bamboozle you with the opinions of other scientists and cite more genetic studies that fail to support the split.
Instead, let’s conclude the discussion about the genetic basis of the split by remembering the role of Dale Serjeantson who, you might recall, considered today’s North Atlantic Pterodroma petrels as constituting mere remnants of a once common class of seabird. Dale Serjeantson, with four colleagues — Selina Brace, Ian Barnes, Andrew Kitchener and Samuel Turvey — published, in 2014, results from DNA analysis that sought to elaborate the genetic relationship between the ‘extinct Scottish Pterodroma’, whose DNA was extracted from bone remains, and the still living Pterodroma petrels of Macaronesia (Fea’s Petrel from the Desertas, Fea’s Petrels from Cape Verde, and Zino’s Petrels from Madeira). Their paper concluded by saying: ‘The Macaronesian populations [of Fea’s Petrels] can be identified as distinct phylogenetic lineages but are not considered to have undergone sufficient morphological divergence to unambiguously represent distinct species, and the very low level of sequence divergence shown by the extinct Scottish Pterodroma indicates that these living and extinct taxa [a group that forms a cohesive unit] are best interpreted as comprising conspecific [belonging to the same species] populations within a wider Pterodroma complex that was formerly distributed across much of the north-east Atlantic region.’[2]
Let’s introduce another problem that complicates understanding the science behind molecular methods that enable genetic signatures in DNA to be analysed and, thereby, reveal relationships between species. The problem is a lack of plain English. I, for one, cannot understand the detail. I hope this is not because I am stupid. I have endeavoured to ‘keep up’ with ornithological research. On more than one occasion, when confronted with detail that I cannot understand, I contacted the lead author and asked for, in essence, a translation of their work from jargon into English. Anyone I contacted was pleased to help and even agreed with me that the language they were obliged to use was ‘horrible’ (Nora V. Carlson, Max Plank Institute of Ornithology, Radolfzell, Germany). So it is not just me. Professor Ian Newton, one of Britain’s best ornithologists, emailed this to me: ‘I agree with you that much of the scientific literature has over the years become increasingly impenetrable. As you know, I have made several attempts in recent years to summarise some of it in book form, but despite spending my life as a scientist, I find it real hard work working through scientific journals, and trying to translate the stuff into simpler English. One certainly doesn’t write books for the money, but I feel it is worthwhile if it allows non-scientists to learn about new discoveries in bird biology.’
But let’s get back to the split of Fea’s Petrels. Although other genetic studies of Pterodroma petrels in the Northeast Atlantic did not find evidence in favour of splitting Fea’s Petrel into two species, such as [2], the article by Jesus et al. [28] came down on the side of separating the breeding populations of Madeira (on Bugio) and Cape Verde into two species. Although I did my best to comprehend the data in Jesus et al., I decided that a better option was to seek help from Professor Desmond Higgins. Des Higgins is based at University College Dublin and is a world authority on genetic research. A 1994 research paper co-authored by him set the international standard for DNA sequence analysis. Indeed, Jesus et al. even used ‘the Higgins test’ in their research.
This is what Des Higgins made of the article by Jesus et al.:
‘The real question is not if the results are fishy but how you interpret those results, in terms of what defines a species. The definition has become fluid and there is no universally accepted formula to say what should be split and what should be lumped. The main results for me are in [Table 1 and Figure 4]. Figure 4 gives a visual overview and Table 1 gives some numbers. The numbers are ‘mean number of estimated substitutions per nucleotide site between each pair of groups or within’. If you multiply these by 100, you get the percentage difference. [The authors] quote percent differences of around 1% or 2%. These show that the three groups [Madeira Petrel P. madeira on Madeira, ‘Desertas’ Fea’s Petrel P. feae deserta on Bugio, Cape Verde Fea’s Petrel P. feae feae on Cape Verde] are very homogeneous within each group and have a divergence between the groups that is comparable to other species splits. They used a crude clock calculation that says the [birds] could have split well over a million years ago — but I don’t know how reliable that is. So, is that a species split or not? They give their arguments in the discussion and that is where you can argue about what it means or does not mean. They argue that the splits are comparable to other species splits that are widely accepted. They point out the apparent lack of any evidence in the sequences for hybridization (but the samples are small and this is for mitochrondrial DNA which is female specific; it says nothing about males). They also point to wide geographic separation and differences in breeding habits and especially timing which would help keep them apart (reproductive isolation). They say: “molecular analysis of DNA sequences allows the identification of management units and evolutionary significant units that are fundamental in conservation. Modern conservation theory requires the definition of Evolutionary Significant Units (ESU) rather than inferences on fitness and diversity of populations and limits between species (Moritz 1994) [32]. Our data for Bugio and Fogo [one of the islands in Cape Verde] suggest the existence of two ESUs.” The question of a split or not may have a lot to do with politics and conservation. These could be ‘cryptic species’ which means they are effectively separate species — but you cannot tell them apart in the field, away from breeding sites. You do not have to have morphological differences or field characters to [qualify as] different taxa [a group that forms a cohesive unit] but it makes field identification a nightmare. These taxa do warrant separate distinctions — but at what level, I cannot say.’
Bill Bourne, who, over many years, has made significant contributions to the elucidation of the taxonomic status of Pterodroma petrels in the North Atlantic [9,11] expressed his view on the proposal to split Fea’s as follows:
‘I must have seen most of the museum specimens of gadfly petrels of the Fea’s Petrel group, and birds at sea off the Cape Verde Islands and Desertas, and I could see little difference between them; it seems mainly theoretical. If it is wished to separate birds on their breeding seasons, what about the Blackcaps of the Cape Verde Islands or — on their voices and DNA — what about Chaffinches? The main argument for treating the Bugio petrels as distinct seems to be that this raises their conservation status — and is this honest?’ [33]
Basically, the argument in favour of splitting Fea’s Petrel into two species would result in one of the newly-declared species, that is, the Fea’s Petrels breeding on Bugio, becoming a world-endangered seabird with the entire population breeding on just one small crumbling plateau on a small island east of Madeira. Such a perilous predicament would, it is hoped, attract conservation attention — and enhance funding for conservation efforts and research. Even though the plight of all the world’s Pterodroma petrels is heart-breaking — and efforts to preserve them all are to be applauded — using rarity as a means to manipulate taxonomy in the interest of conservation and research funding is to discredit scientific decisions relating to naming species. Is this what is really going on?
UNSOUND APPROACH
If the genetic basis for advocating the split of Fea’s Petrel into two species is questionable, what about the other main reason marshaled in support of differentiating Fea’s Petrel from, as it is being called, Desertas Petrel? As a reminder, this was based on differences in the bioacoustics between Fea’s Petrels on the island of Bugio, one of the three Desertas Islands (therefore, ‘Desertas Petrel’) and Fea’s Petrels breeding much further south, on the islands of Cape Verde. Let’s dispense with the highfalutin jargon — ‘bioacoustics’ means voice.
The sole source asserting that vocal differences can be used to differentiate the two populations of Fea’s Petrels is contained on pp. 36–42 of the book Petrels, night and day, written by Magnus Robb, who also made most of the sound recordings that accompany the book on two CDs [18]. Robb’s claim is not published anywhere else — such as a reputable, peer-reviewed journal — and the all-important recordings, some of which were made by Alexander Zino, cannot be heard except by purchasing the book.
Robb’s account opens with the lines, ‘almost everything we know about the breeding biology of ‘Fea’s Petrel’ actually refers to Desertas Petrel, which breeds on the island of Bugio in the Desertas.’ Hmmm. If this is Robb’s depth of knowledge about the bird’s breeding biology and lifestyle then he needs to read the scientific literature more thoroughly, which elaborates on the habits of the Cape Verde population of Fea’s Petrels too. A few lines further he says, ‘it seems likely that Desertas and Fea’s have been reproductively isolated for some time. Indeed, Ratcliffe et al. (2000) [34] thought that they were “probably cryptic species”’. In fact, Ratcliffe et al. made this comment purely on the basis of the distance between the breeding sites being wide (2000 km apart) and continued, ‘further phylogenetic analyses are necessary to test this hypothesis and elucidate the position of P. fea.’ [34]
As for measurements, Robb states, ‘Desertas [Fea’s Petrel] averages larger in several measurements [than Cape Verde Fea’s Petrel] and its average bill depth is greater than [Cape Verde Fea’s Petrel]’. The reference cited for this comment is from a 1995 article in Ibis, the journal of the British Ornithologists’ Union [35]. Unfortunately, the BOU charge for access to articles published online in Ibis. I refuse to pay their pimp publisher — Wiley Online Library — for the PDF of the article cited by Robb, although I have read the abstract, which does not contain any measurements. But this does not matter — because a more recent and comprehensive set of measurements was published in 2010 by the same author (Vincent Bretagnolle) in an article that he co-wrote with Hadoram Shirihai and Francis Zino [13].
None of the measurements published in 2010 corroborate the claims made by Robb. Let’s start with bill measurements. Shirihai et al. give three types of bill measurements. These cover (1) Bill depth from last feather above nose tube. (2) Bill depth immediately in front of nose tube. (3) Bill depth at maxillary nail/hook to gonys. At the risk of being pedantic — but also because few readers will be able to access the original article that is not available online — here is the published data. In all cases, the measured sample size was 300 individuals for Desertas Fea’s Petrel and up to 40 Cape Verde Fea’s Petrels. For (1) Desertas size range was 12.50 to 17.50 mm; Cape Verde size range was 12.10 to 14.80 mm. For (2) Desertas size range was 8.40 to 12.70 mm; Cape Verde size range was 8.62 to 10.52 mm. For (3) Desertas size range was 10.70 to 14.20 mm; Cape Verde size range was 10.70 to 13.80mm. In conclusion, the only ‘tendency’ was for a minority of Desertas birds to be larger-billed (most likely adult males). Otherwise, overlaps negate the worth of bill size as a valid guide to telling the two subspecies (my word) apart. The same applies to comparisons of body length, wingspan, wing length and tail length. As an example, the range of total body length from live birds measured at breeding colonies was, for Desertas Fea’s Petrels, 330 to 380 mm; for Cape Verde Fea’s Petrels, 360 to 390 mm. In other words, contra Robb, in body length at least, some Cape Verde Fea’s Petrels are actually longer than even the longest-bodied Desertas Fea’s Petrels.
Robb’s asserts that, ‘the time has come for Desertas Petrel [Desertas Fea’s Petrel] to be treated as a species in its own right, given that it differs from Fea’s Petrel [Cape Verde Fea’s Petrel] in breeding phenology, size and sounds.’ By the way, ‘breeding phenology’ is a fancy way of saying ‘timing of breeding season’. Although the Fea’s Petrels nesting at Cape Verde start a few months earlier than the Fea’s Petrels on Bugio, the staggered timing presumably reflects optimum breeding conditions, such as the availability of sufficient food to rear offspring. The argument that staggered breeding seasons drive species to diverge is hard to believe. For example, Roseate Terns are migratory in several of the world’s oceans — such as both sides of the North Atlantic and in the western Pacific — but other populations are resident and breed in tropical seas. Consequently, the various populations stagger their breeding season according to latitude. As mentioned earlier, White-faced Storm-petrel provides a further example of a seabird whose populations breed at different times with respect to latitude.
Robb’s second reason for advocating the split — size differences between breeding populations on Cape Verde and Bugio — is not, as note above, sustained by the measurements in Shirihai et al. As an important aside, bone remains from three Pterodroma petrels from El Hierro [36], the westernmost of the Canary Islands, provide an insight into the former distribution of Pterodroma in Macaronesia and, as discussed by D Sergeantson et al., Northwest Europe [2]. It should come as no surprise that, as happened across all parts of Pterodroma petrels’ European breeding range, populations on other island archipelagos, such as the Azores and the Canary Islands, were also decimated following human settlement and the associated arrival of rats and cats. Three sets of bone remains from caves on Hierro prove the existence of breeding Pterodroma petrels on this Canary Island. The bones were found in caves that came into existence as part of volcanic activity around 50,000 years ago. The age of the bones is not (yet) known but they may date from 2,500 years ago because ‘the data show an extinction process that probably began with the aboriginal arrival about 2,500 years ago.’ [37] Exhaustive measurements of the remains are published in Rando [36] — but few of the measurements are directly comparable with measurements in Shirihai et al. Nonetheless, judged against a single specimen of Desertas Fea’s Petrel, the El Hierro Pterodroma was a larger in skull length, length of [bill] culmen, femur and — pertaining to the wing — the humerus and ulna. The scale of difference from Fea’s Petrel appears to similar to that between Fea’s Petrel and Zino’s Petrel and can be appreciated from figures (actually photographs) 3 and 4 in Rando [36]. Hence, by way of confirming the opinion expressed by D. Serjeantson — ‘a museum of diversity rather than a cradle of evolution for Atlantic Pterodroma’ — what we have today is a residue of survivors that, across their breeding islands, show a measure of genetic drift. Agreement with such a scenario comes from V Bretagnolle who stated: ‘petrels breed on isolated and remote oceanic islands that provide natural geographic isolation; moreover [they] are highly [faithful to breeding locations]. These two characteristics should promote genetic drift and thus geographic variation.’ [25]
What are we to make of the differences in vocalizations that Robb says make Fea’s Petrel two species? But, before examining them, could they be explained by geographic variation? James [38] found geographical variation in calls of Manx Shearwater and also in the purring song that male European Storm-petrels produce in their burrows. The song consists of rapid clicks that sound like a purring trill, followed by a flourish (known as a ‘breath note’). The purr is just over one second long and the breath note lasts around a quarter of a second; a series of song repetitions is delivered in quick succession with only a fraction of a second between them. James found significant differences among four sites in the click rate within the purr, in the length of the breath note, and in the gap between songs. The most marked differences were between Storm-petrels recorded on Puffin Island, Co. Kerry, and Mousa in the Shetland Islands, two sites some 1100 km apart.
Bretagnolle [25], in his discussion of petrel calls (including Pterodroma) addresses ‘evidence for species recognition and geographic variation.’ He states: ‘geographic variation is common in petrel vocalizations and has been found in all families and most genera.’ The quantity of research into geographic variation in petrels — in morphology, coloration, genetics and vocalizations — is considerable. So much so, that if I listed all the academic papers that I have read confirming its existence, the list would swell to a paragraph. However, by way of a typical example, Bretagnolle and Genevois recorded and analysed geographic variation in calls of Blue Petrels Halobaena caerulea, a seabird that occurs throughout the Roaring Forties, across five breeding archipelagos and seven islands within some archipelagos [39]. The research found both ‘macro- and micro-scale variations’ between the archipelagos (macro-scale) and within clusters of islands (micro-scale). The authors suggested that such minor differences have the same origin, namely ‘extreme philopatry’ (faithfulness to breeding site). The same behaviour — loyalty to breeding site — was responsible for genetic differences discovered in another seabird, the Fairy Prion Pachyptila turtur, in breeding populations only a few kilometres part [40].
TECHNO-BABBLE TAXONOMY
Because many — but not all — petrels attend their breeding sites at night, sound is an important means of communication in breeding activities such as courting and advertising a potential nest site. Ergo, ‘calls, owing to their role in species-specific recognition in these birds, should be given priority over other characters for species recognition.’ [25]. It is one thing to sound erudite but to satisfy the standards demanded by science, there needs to be proof. Put these words in order of importance: ‘theory’, ‘hypothesis’ and ‘fact’. In this particular field of seabird science, researchers, to corroborate a theory or hypothesis and turn it into a verifiable fact, use ‘playback experiments’ that take advantage of the unique colonial and nocturnal behaviour of petrels. Usually, two types of experiment are used to test response. In the first type, reactions are scored by catching or counting the non-breeders (breeders are silent) that respond to a loudspeaker by either flying over it or calling. In the second type of playback, birds are subjected to playback tests at the nest site (breeding burrow).
As far as I know, Robb’s data — one set of CDs in a book, Petrels night and day [18] — is all there is to go on: no playback experiments, no peer-reviewed article in a reputable publication, not even a link so that the recordings might be publicly available. Because I have a copy of Petrels I have listened many times to the CDs, in particular the recordings of Cape Verde Fea’s Petrel (made by M Robb) and Desertas Fea’s Petrel (made by Alexander Zino). The two populations are isolated and members of each are more than likely to be faithful to their natal area (in other words, ‘philopatric’). There is, therefore, an expectation that their voices are likely to be different in some — probably small — degree. For the sake of layperson understanding, you could describe such a discernible difference — if one was found to exist — as representing a dialect. Nonetheless, the content will be comparable even if the sound made by petrels from one population is not a carbon copy of the other. Given all that is known about petrel vocalizations, such an outcome would be expected. To my ear — and given that the quality of the two sets of recordings is not the same — the calls come across as ‘peas in a pod’ in their degree of similarity.
What kind of sound do the birds make? Certainly, not at all like a songbird. The chief call — because it is used primarily during courtship it is classified as a ‘major call’ [25] — is a low-pitched wail. To give you some idea of the quality, it would make a good sound effect in a ghost story. Both sexes use the same call and there is a small — but discernible — difference in pitch. It is believed that males are the higher-pitched sex [35]. Each wail tails off slightly at the end but sometimes the caller ‘rounds off’ the wail with a little rise in pitch at the very end — like a tiny hiccup. In Robb’s commentary (below), he uses ‘-wik’ to describe the hiccup and uses ‘moans’ for wails.
Here is what Robb said in Petrels. He makes three points that I have listed as 1, 2 and 3. His justification for using sounds to split Fea’s Petrel into two species raises a lot of questions in my mind.
1‘The [Desertas Fea’s Petrels] you can hear in CD1–10 were recorded [by Alexander Zino] on the night of 28/29 September 1967 … at this stage in the breeding cycle, the earliest nestlings would be a few days old. Most of the birds calling would have been non-breeders, performing their courtship flights in preparation for future breeding attempts. Most of these calls are moans, equivalent to the moans of [Cape Verde Fea’s Petrels] … a rise in pitch at the end of a call is a strong characteristic of most [Desertas Fea’s Petrel] calls. More than 75% of their moans, both high and low variants, end with a rapidly rising squeak, which typically sounds like –wik. Sometimes this inflection is less obvious, for example when the call has been recorded from a distance. In [Cape Verde Fea’s Petrels] only a minority of moans have –wik endings.’
Not true. In CD1–08 [Cape Verde Fea’s Petrel, 18 February 2007] there are several –wik endings. This track lasts 1 minute 54 seconds. Of 30 clear calls, at least 8 (possibly 10) have -wik endings. Expressed as a percentage, this is either 27% or 33%. In CD1–10, duration 2 minutes 10 seconds, there are 57 calls (more than one bird is calling and the calls probably come from both sexes) and my estimate of –wik endings is 20. That is 31% — almost the same percentage as in Cape Verde Fea’s Petrel. I’m not in a position to reproduce images from Petrels without infringing on copyright.
2 ‘Another key difference [between Desertas Fea’s Petrel and Cape Verde Fea’s Petrel] was that the moans of Desertas Petrel drop less in pitch over the main part of the call, as you can see [in a magnified portion of sonogram]. Compared to the [magnified portion of sonogram of Desertas Fea’s Petrel], the descent is steeper and more curving [in Cape Verde Fea’s Petrel]. To find out whether these differences were genuine, I analysed [calls] of 312 [Desertas Fea’s Petrels, Cape Verde Fea’s Petrels and Zino’s Petrels], both high type and low type moans. The much higher proportion of wik endings and the smaller drop in pitch over the main part of the call in Desertas were confirmed.’ [Robb’s sonograms are captioned to indicate, for Cape Verde Fea’s Petrel, high moans dropping in pitch by an average of 24% and low moans also dropping 24% in pitch. For Desertas Fea’s Petrel, the average drop in pitch for high moans is 17% and, for low moans, an average of 18%.]
Because the drops in pitch are expressed as averages, some moans of Cape Verde Fea’s Petrels are the same as moans made by Desertas Fea’s Petrels. Indeed, given the scale of difference — 24% versus 17% or 18% — there is such little difference that, were it not represented on a sonogram, the difference in sound could scarcely be appreciated by a human. Moreover, even where some sonograms show the hair-splitting difference in pitch, other sonograms (of individual moans) do not. A significant factor that has to be taken into account is that the recording of Desertas Fea’s Petrel is much better quality than any of the Cape Verde Fea’s Petrel recordings. Just as Robb says about –wik notes [he also calls them ‘inflections’] being ‘less obvious … for example when the call has been recorded from a distance’, the same seems to apply to establishing the nature of any drop in pitch in the moans.
3 ‘I also noticed that high and low moans of [Desertas Fea’s Petrel] differ more from each other in fundamental frequency.’
In other words, if we assume that high moans are made by males and low moans are made by females (which is not yet, to my knowledge, a proven fact) then the degree of difference spans a wider gap in pitch for Desertas Fea’s Petrel compared to Cape Verde Fea’s Petrel. Once again, you need to listen to the sounds and look at, not just the one-off ‘sound bite cum annotated sonogram’ figures that accompany the text, but the complete sonograms for the longest cuts (track 8 for Cape Verde Fea’s Petrel and track 10 for Desertas Fea’s Petrel). Here again, while there is a good clip for Desertas Fea’s Petrel, the same cannot be said for Cape Verde Fea’s Petrel. Therefore, valid comparisons are, in my view, not easy to make — or to stand over. Incidentally ‘fundamental frequency’ means, basically, the dominant part of the sound which is also usually the lowest frequency but certainly the loudest. At the end of this Robb concludes: ‘the time has come for [Desertas Fea’s Petrel] to be treated as a species in its own right, given that it differs [from Cape Verde Fea’s Petrel] in … sounds.’
I asked two non-ornithologists to listen to the best clips. Both are, in their different ways, experts in sound recordings because this is their profession. One is a senior BBC sound engineer; the other ran a ‘bio-acoustic’ company. Neither saw any merit in the uniqueness of one petrel over the other. In every attribute where a difference was claimed to exist, both petrels shared the so-called unique attribute. Both experts felt that the comparisons were hampered by the quality of the two sets of recordings (good for just Desertas Fea’s Petrels). Both were more than a little surprised when I told them that Robb’s recordings and his interpretation of them formed a basis for the two petrels being split as two species because they have different vocals. One contact said, ‘but the interpretations are nothing more than techno-babble — can nobody see that?’
BURDEN OF PROOF
Work by Vincent Bretagnolle, an acknowledged authority on petrels with numerous publications to his name, has been referenced more than once in this article. Strangely, Bretagnolle seems to draw the ire of Robb who, several times, criticises him and even accuses him of flawed research into petrel vocalisations which, had it ‘gained wide acceptance’, then ‘efforts to conserve Desertas, Fea’s and Zino’s might have been severely compromised’ [18, p.23]. Unlike Robb, Bretagnolle did not favour the split of Fea’s Petrel into two species. Also unlike Robb, Bretagnolle’s research includes playback experiments and his results were scrutinised and subject to academic rigour before publication. For example, his analysis of vocal variation in Blue Petrels involved sound recordings made from 504 different individuals [39]. Robb’s claim that, because of vocal differences discovered by him, Cape Verde Fea’s Petrel should be split from Desertas Fea’s Petrel, is based on just one recording (by Alexander Zino) of Desertas Fea’s Petrels and a handful of recordings of Cape Verde Fea’s Petrels. In terms of statistical analysis alone, such a claim is hard to validate.
Robb and The Sound Approach have a habit of dabbling in taxonomy and not looking before they leap — the Fox News of taxonomy. Elsewhere in Petrels, Robb declares that the Mediterranean subspecies of European Storm Petrel Hydrobates pelagicus melitensis is actually a species in its own right — ‘Mediterranean Storm-petrel’ H. melitenis. Once again, minor vocal differences lie at the heart of Robb’s contention. But why are the claimed differences not part of the regional variation in European Storm-petrels described by others, such as James [38]? Robb added larger bill size (for those few Mediterranean individuals he has seen) to his justification for treating the birds as a different species. However, he was premature in this claim because a European Storm-petrel trapped in Scotland (confirmed as belonging to the nominate subspecies of European Storm-petrel from its DNA) demonstrated that variation in bill size in European Storm-petrels overlaps with measurements of European Storm-petrels breeding in the Mediterranean (of the melitensis subspecies) [41].
Often Robb (and The Sound Approach in general) seem less interested in meaning and more interested in wanting to sound significant. In other works, such as Undiscovered Owls [42] you find the same thing. Referring to the book’s taxonomic order of various owls, the authors (principally Robb) state that the book’s taxonomy: ‘does not follow any existing authority.’ Little wonder, therefore, that The Sound Approach have committed various blunders, subsequently corrected by reputable researchers, as detailed in this review [43]. Moreover, The Sound Approach rule-of-thumb seems to be that, if vocal differences are found between populations of the same species, then each population deserves, for conservation reasons, to be treated as a different species: ‘whether a bird is one species or another can influence everything from the building of an airport to the fate of a forest. [Because] vocalisations can form a diagnostic feature for a species, undervaluing [them] can dramatically effect [presumably they mean ‘affect’?] biodiversity’ [44]. The latter mumbo-jumbo amounts to a splitter’s charter and also gives an insight into the ‘rarity justifies taxonomic revision’ argument that seems to underpin Robb’s motives for advocating the split of Fea’s Petrels. What, therefore, would The Sound Approach make of playback experiments conducted on two subspecies of Eurasian Stone-curlew (one in mainland Europe, the other geographically isolated in the Canary Islands) whose vocalisations differ but which respond to vocalisations of each other: ‘playback experiments indicated that individuals from the nominate subspecies responded in the same way to the playback of calls of individuals belonging to both subspecies.’ [45]
Let’s get back to Pterodroma petrels. If any readers are lucky enough to see one off the Irish, British or northwest European coast, is there a means of assigning the bird to species? Obviously great care is needed but the recent occurrence of an unequivocal Zino’s Petrel (well seen and photographed by expert observers) off the Isles of Scilly [46] shows that, in ideal conditions, the at-sea separation of Zino’s Petrel from Fea’s Petrel is possible — even thousands of kilometres away from natal seas. Having been to Madeira, I certainly feel that, given a good look, Zino’s Petrel is identifiable in the field. What about distinguishing between individuals drawn from the two widely separated breeding populations of Fea’s Petrels? Armed with a copy of Bob Flood and Ashley Fisher’s North Atlantic Seabirds — Pterodroma petrels, you have a slim chance of finding a ‘winning combination’ of certain characters which, if the percentages fall within a narrow range, may favour one subspecies over the other. However, even Bob Flood failed to determine the subspecies of a Fea’s Petrel well seen by him and others — and comprehensively photographed at close range [47].
Science has given us a forensic ability to see how evolution operates. It is ironic that, just when we have reached a point where we can code the natural world by its DNA, human abuse of life on Earth is forcing us to choose between co-existence or co-extinction. If the modern approach to conserving nature is to remove our ability to tell species apart from what we see and hear, how might a latter-day Linnaeus or Darwin have distilled order out of observable differences? It may come as a surprise, but I rejoice in the fact that, in small ways, Cape Verde Fea’s Petrels and Desertas Fea’s Petrels are on a long journey towards being separate beings. They are not there yet. The same can be said of many other species, whose tribes consist of different and distinct subspecies, such as Dunlins, Herring Gulls and Eiders. Variety really is the spice of life. It should not be carved up into ever more tiny pieces that hide its visible achievements as a work in progress.
ACKNOWLEDGEMENTS
Unless I had gone to Madeira and ‘looked for myself’, none of this article could have been written. Windbirds (Hugo Romano and Catarina Correia-Fagundes) were phenomenal in their bird-finding skills and in their expertise in distinguishing between Zino’s and Fea’s Petrels at sea. Their opinion of the Fea’s/Cape Verde ‘split’? Let’s just say they do not agree with it. Pete Morris and everyone on the three boat trips were great company. Pete’s images were invaluable in illustrating the Zino’s Petrels we saw. Alexander Lees cast his expert eye over the text and gave it a thumbs-up. Eric Dempsey provided useful feedback and Michael O’Clery drew the delightful map. Dale Serjeantson, Bob Flood, Neville McKee and Julian Wyllie were a big assistance and didn’t mind being pestered with questions. Professor Des Higgins was an ace. Little did I know that the Lionel Messi of geneticists lived and worked in Ireland. Des ‘translated’ academic jargon for me and clarified the potential motives behind the reasons for splitting Fea’s Petrel into two species. As ever, I am indebted to Richella Duggan for her brilliant editing skills. She read every word and improved the content considerably.
REFERENCES
1 Serjeantson, D. 2005. Archaeological records of a gadfly petrel Pterodroma sp. from Scotland in the first millennium AD. Documenta Achaeobiologiae (proceedings of) vol. 3: 235–246.
2 Serjeantson, D. et al. 2014. Late Holocene range collapse in a former British seabird species. Journal of Biogeography vol. 41:1583–1589.
3 Flood, B. & Fisher, A. 2013. North Atlantic Seabirds, Pterodroma petrels. Page 26. Pelagic Birds & Birding Multimedia Identification Guides.
4 ‘Irish Birding’ website.
5 Witherby. H.F., Jourdain. F.C.R., Ticehurst, N.F. & Tucker, B.W. 1940. The Handbook of British Birds, vol. 4, pp. 62–68. H.F. & G. Witherby Ltd, London.
6 Heinzel, H., Fitter, R. & Parslow, J. 1973. The Birds of Britain and Europe. London.
7 Enticott, J. 1999. Britain and Ireland’s first ‘Soft-plumaged Petrel’ — an historical and personal perspective. British Birds vol. 92:504–518, October 1999.
8 Stapp, P. 2002. Stable isotopes reveal evidence of predation by Ship Rats on seabirds on the Shiant Islands, Scotland. Journal of Applied Ecology 39:831–840.
9 Bourne, W.R.P. 1983. The Soft-plumaged Petrel, the Gon-gon and the Freira, Pterodroma mollis, P. feae and P. madeira. Bull. BOC 103:52–58.
10 Zino, F., Oliveira, P., King, S. & Buckle, A. 2001. Conservation of Zino’s Petrel in the archipelago of Madeira. Oryx: 35 (issue 2, April 2001):128–136.
11 Bourne, W.R.P. Bourne. 1992. Debateable British and Irish seabirds. Birding World vol. 5 (October 1992):382–390.
12 Zino. F., Phillips, R. & Biscoito. M. 2011. Zino’s Petrel movements at sea — a preliminary analysis of data-logger results. Birding World 24 (June 2011): 216–219.
13 Shirihai, H., Bretagnolle, V. & Zino, F. 2010. Identification of Fea’s, Desertas and Zino’s Petrels at sea. Birding World 23 (July 2010): 239–275.
14 Ramirez, I. et al. 2013. Year-round distribution and habitat preferences of the Bugio Petrel. Marine Ecology Progress Series: 476:269–284.
15 Zino, F. & Zino, P.A. 1986. An account of the habitats, feeding habits, breeding and density of the population of the Long-toed Wood Pigeon. Bogagiana no. 98, pp.1–15.
16 Discoveries made by Harald Pieper are quoted in Bourne, W.R.P. 2000. Conservation, taxonomy, nomenclature and gadfly petrels. British Birds 93:571 (October 2000).
17 Svensson, L., Mullarney, K. & Zetterstrom, D. 2009. Collins Bird Guide (second edition), pp.72–73.
18 Robb. M., Mullarney, K. & The Sound Approach. 2008. Petrels night and day, p.25. Poole, Dorset.
19 Olsen, K. M. & Larsson, H. 2004. Gulls of Europe, Asia and North America. Helm. London.
20 Rising, J.D. 1996. A Guide to the Identification and Natural History of the Sparrows of the United states and Canada. Academic Press. San Diego.
21 Baptista, L. & Welty, J.C. 1998. The Life of Birds. Thomson. California.
22 Bourne, W.R.P. 1953. On the races of the Frigate Petrel Pelagodroma marina with a new race from the Cape Verde Islands. Bulletin of British Ornithologists’ Club vol.73, 79–82.
23 Wikipedia entry for White-faced Storm-petrel.
24 Silva, C.M, Duarte, M.A. & Coelho, M.M. 2011. Anonymous Nuclear Loci in the White-faced Storm-Petrel and their applicability to other procellariiform seabirds. The Journal of heredity 102(3):362–365
25 Bretagnolle, V, 1996. Acoustic communication in a group of non-passerine birds, the petrels. Chapter 9 (pp.160–177) in Ecology and evolution of acoustic communication in Birds.
26 Foster, J.T., Walker, F.M., Rannals, B.D. & Sanchez, D.E. 2018. Population genetics of an island invasion by Japanese Bush-warblers in Hawaii, USA. The Auk 135(2):171–180.
27 Yoon, C.K. 2009. Naming Nature. Norton. New York.
28 Jesus, J., Menezes, D., Gomes, S., Oliveira, P. Nogales, M & Brehm, A. 2009. Phylogenetic relationships of gadfly petrels Pterodroma spp. From the Northeastern Atlantic Ocean: molecular evidence for specific status of Bugio and Cape Verde petrels and implications for conservation. Bird Conservation International 19:199–214.
29 Gill, F., Donsker, D, & Rasmussen, P. 2020. IOC World Bird List (version 10.1).
30 British Ornithologists’ Union website.
31 Militao, T., Dinis, H.A., Zango. L., Calabuig, p., Stefan, L.M. & Gonzalez-Solis, J. 2017. Population size, breeding biology and on-land threats of Cape Verde petrel in Fogo Island, Cape Verde (12 pages).
32 Moritz, C. 1994. Applications of mitochrondrial DNA analysis in conservation: a critical review. Mol. Ecol. 3:401–411.
33 BirdLife’s Globally Threatened Bird Forums. Archived (online) 2014 discussion: Fea’s Petrel is being split: list P. deserta as Vulnerable and P. feae as Near Threatened?
34 Ratcliffe, N., Zino, F.J., Oliveira, P., Vasconcelas, A., Hazevoet, Costa Neves, H., Monteiro, L.R. & Zino, E.A. 2000. The status and distribution of Fea’s Petrel in the Cape Verde Islands. Atlantic Seabirds 2: 73–86.
35 Bretagnolle, V. 1995. Systematics of the Soft-plumaged Petrel Pterodroma mollis: new insight from the study of vocalisations. Ibis 137: 207–218.
36 Rando, J.C. 2002.New data of fossil birds from El Hierro (Canary Islands): probable causes of extinction and some biogeographical considerations. Ardeola 49:39–49.
37 Navarro, J.F., Martin, E. & Rodriguez, A. 1990. Las primeras etapas del programa de excavaciones en las Cuevas de San Juan y su aportacion a la diacronia de la Prehistoria de Canarias. Investigaciones Arqueologicas en Canarias. II:189–201.
38 James, P.C. 1985a. Geographical and temporal variation in the calls of the Manx Shearwater and European Storm-petrel. Journal of the Zoological Society of London 207:331–344.
39 Bretagnolle, V. & Genevois, F. 1997. Geographic variation in the call of the Blue Petrel: effects of sex and geographical scale. The Condor 99:985–989.
40 Ovenden, J.R., Wust-Saucy, A., Bywater, R., Brothers, N. & White, R.W.G. 1991. Genetic evidence for philopatry in a colonially nesting seabird, the Fairy Prion. Auk 108:688–694.
41 Menzie, S. 2019. An atypically large-billed nominate-race European Storm-petrel in Scotland. British Birds 112 (January 2019):44–60.
42 The Sound Approach. 2015. Undiscovered Owls. The Sound Approach. Poole. Dorset.
43 Bulletin of the British Ornithologists’ Club, vol. 135, no.4 (December 2015), p.281.
44 The Sound Approach. 2006. The Sound Approach to birding p.136. Poole. Dorset.
45 Dragonetti, M., Rodriguiez-Godoy, F., Caprara, M. & Barone. R. 2020. Acoustic analysis and playback experiments do not support revision of the Central and West Canary Islands subspecies of the Eurasian Stone-curlew. Located online at: researchgate.net/publication/340584017.
46 Account online by Bob Flood at: Rare Bird Alert, finders-in-the-field, Zino’s Petrel, Isles of Scilly, 30 July 2020.
47 Flood, R.L. 2015. Process for at-sea identification in the feae-complex applied to a petrel observed off the Isles of Scilly. Seabird 28: 78–88.
