the evening of 28 July 2007 I was on board MV Sapphire south of Scilly
with four other birders, Alan Hannington and John Higginson from
Scilly, Tony James from Newcastle, and Ken Adelsten Jensen from Norway.
The skipper was Joe Pender who also has an interest in seabirds. The
south-westerly wind had picked up from force 3 when we left harbour to
force 5 and was strengthening. There was 100% cloud cover but
precipitation from that afternoon had all but ceased, although there
was some sea spray that was a problem if looking into the wind. Light
and other conditions were good looking down wind. We were drifting and
chumming watching European Storm-petrels heading in from down wind,
searching for a Wilson’s Storm-petrel.
about 18.55 I was located port side of the cabin looking down wind.
Through binoculars I saw a long-winged black-brown storm-petrel rise
over the crest of a wave heading directly towards the boat. Immediate
impressions said it was not a Wilson’s. Compared to Wilson’s it was
larger, the arm relative to the hand was longer, and the wings were not
flattened but were held in a shallow m-shaped bow. Neither did those
first moments resonate with Leach’s, in particular because the jizz and
flight behaviour was steady and methodical rather than tern-like.
Madeiran flashed into the mind. I shouted to the other birders, “Get on
and despite these very first impressions, my next stage of thinking
over the following 10 seconds or so was akin to trying to force a
square peg into a round hole trying to make this storm-petrel a
Wilson’s. I guess this was because a Madeiran off Scilly was just too
incredible to believe. Also, in these early stages of interpretation,
there was some correspondence between Wilson’s and our Madeiran, albeit
weak, like relatively straight trailing edge to the wing and purposeful
travelling flight behaviour. And of course we were expecting to see
Wilson’s. Conversely, Leach’s was never a candidate.
30 metres away from the boat the storm-petrel banked to its right
though continued to approach but now was methodically quartering the
sea surface. I could see an upperwing covert bar but it did not reach
the leading edge (Leach’s reaches the leading edge) and was not strong
as on a typical Leach’s. The white rump was broader than long (Leach’s
is longer than broad) extending to the lateral undertail-coverts
(unlike Leach’s) and was still clearly visible when the storm-petrel
banked away showing its underside (Leach’s white rump is then
obscured). The ‘band rump’ seemingly always was visible as the
storm-petrel manoeuvred around (unlike Leach’s). The tail was short and
the rear carriage looked particularly short relative to the wing span,
whereas the rear carriage of Wilson’s looks relatively long even when
feet are retracted eliminating toe projection. There was no toe
projection beyond the end of the tail (Wilson’s mainly has a very
obvious toe projection). By now I was trying to accept that this
storm-petrel had to be a Madeiran, but could it really be?! I was aware
that the only British record involved one found dead at Milford Haven,
Hampshire in November 1911 (Category B). If this was a Madeiran, then
to all intents and purposes it was a first for Britain (Category A)!
Madeiran-like storm-petrel was approaching close to the stern and I
expected it to pass the stern and head into the slick off starboard
side and feed. Instead it banked to its left and flew a course parallel
to the port side at about 12 metres off. We had front row seats. Now I
could see the unique thickish bill that always looks too thick for a
storm-petrel (much thinner in Leach’s and Wilson’s). The body was
chunky (relatively slim in Leach’s and Wilson’s). The leading edge of
the wing was mildly angular at the carpal joint and the trailing edge
was gently angular (both sharply angular in Leach’s and the trailing
edge in Wilson’s is all but straight). The arm was broad (narrow in
Leach’s). The wing tips were moderately blunt (very pointed in Leach’s
underwings were evenly hued black-brown with the body all black-brown,
and these plumage features, amongst many other plumage and structural
features, eliminated both species of Fregetta storm-petrel,
White-bellied and Black-bellied, where underwing coverts are largely
white as is the belly (normally with a dark central stripe in
Black-bellied). European Storm-petrels alongside the Madeiran simply
jizz was consistently methodical, almost predictable, like a European
Nightjar, not buoyant and sometimes unpredictable as Leach’s, or
hirundine-like as with Wilson’s, or like a small bat as with European.
It rose up to three metres above the sea surface on several occasions
(very rare in Wilson’s and European that almost always keep within a
metre of the sea surface).
noted all relevant structural and plumage features, and flight
behaviour, and deliberated albeit briefly over other possible
storm-petrel species, I was now confident enough to call, “It’s a
Madeiran!” The Madeiran continued on its methodical way past the bow
and then, banking to its left, away from the boat and was lost to
sight. The whole event lasted about one minute. The exact location was
‘049 51.741 N, 006 08.781 W’ or approximately 7.5 miles south-east of
St Mary’s Quay.
was clear-cut and categorical given the close pass and excellent views
with all key field identification features seen, and so celebrations
immediately began. However, we did not have an experienced photographer
on board and none of us amateur photographers had a camera at hand
keeping them bagged because of sea spray. The lack of photographic
evidence was niggling at the back of my mind. After about 50 minutes
revelry had peaked and we had mostly fallen quiet as the facts sunk in.
about 19.45 I was amazed to see the Madeiran approaching the boat for a
second time. We were treated to an almost carbon copy performance and
an opportunity to run through once again all key field identification
features. This time I yelled to the skipper Joe Pender to take photos
and he had time to run into the cabin, turn on his camera and take a
photograph (which see). This fine record shot shows enough of the
features mentioned above to be identifiable as a Madeiran Storm-petrel.
The exact location of the second sighting was ‘049 51.590 N, 006 07.422
W’ or approximately 8.4 miles south-east of St Mary’s Quay.
Madeiran was clad in what looked like very fresh plumage like a fresh
juvenile or recently moulted immature or adult bird, and certainly was
not in wing moult. Different populations of Madeiran Storm-petrel in
the Atlantic breed at different times of the year, some time sharing
burrows on the Azores with each having non-overlapping breeding and
dispersal periods, and hence different moult timings. Thus, the state
of plumage of our Madeiran does not obviously point to the age or
origin of the bird.
it was extremely helpful though quite astonishing that this sighting
occurred just two weeks after publication of our article in British
Birds that deals with identification of ‘black-and-white’ storm-petrels
of the North Atlantic (Flood and Thomas 2007) and completion of our DVD
on flight behaviour of these storm-petrels (Flood and Fisher 2007).
MV Sapphire is about 16 metres in length and six metres across the
beam, eye-level is about four metres above the sea surface, and angle
of view is shallow making it; (a) relatively easy to see bill shape and
proportions, underwing-coverts and axillary feathers, the extent of
white on undertail-coverts and rear flanks, and the belly, and, if we
had not been so close to the bird, (b) relatively difficult to see
travelling wing shape, tail shape and toe projection, wing-covert bars,
white on rump and uppertail-coverts, and moult and wear of remiges and
Like a European Nightjar Caprimulgus europaeus. Chunky-bodied and
heavy-billed, warm-toned black-brown; in flight methodical and
predictable, overall rather plain.
length and wing span Medium-sized appearing much larger than
accompanying European Storm-petrels, with comments from other observers
including “its like a small shearwater”, but taking into account size
illusion between storm-petrel species at sea I would estimate the body
length and wing span ratios for European Storm-petrel versus our
Madeiran Storm-petrel to be in the region of the known ratios 1.27 and
This bird was not in moult thus there were no moult related issues that
affected structure. Wing shape Long, broad arms, blunt-ended wing-tips,
most of the time held straight out so leading and trailing edges were
only moderately angular at the carpal joint. Head on profile Slightly
bowed in shallow-M, arms medium length, hands long. Tail Shape Shallow
fork but appeared square-ended when fanned as the bird manoeuvred. Toe
projection None. Body build Chunky. Bill shape and proportions Thick,
bulky, and heavily hooked, unlike other storm-petrel species.
Eye-level above sea surface was about three metres thus angle of view
was shallow (also see ‘Vessel’). General colour and colour tones
Warm-toned black-brown and overall rather plain-looking. Upperwing
covert bars At range, quite uniform dark upperwings with fairly
indistinct covert bars. Close-up, greater coverts formed dull bars that
stopped short of the leading edge. Underwing coverts and axillary
feathers Evenly hued black-brown Rump, uppertail- and undertail-coverts
and rear flanks Narrow, rectangular, broader than long, extended to
lateral undertail-coverts and rear flanks, seemingly always visible.
Belly All-dark. Bill Black. Eyes Black. Legs Not seen.
behaviour In travelling flight steady and buoyant with short runs of
shallow wing beats, low banking turns, small shearwater-like glides,
occasionally rising to three metres above sea surface in these
manoeuvres. When searching for food progressing by weaving a regular
horizontal zigzag and quartering areas of the sea surface.
Above Madeiran Storm-petrel Oceanodroma castro c8.4 miles south-east
of St Mary’s quay, Isles of Scilly (Joe Pender). The combination of structure and
visible plumage shown in this image are unique to Madeiran. Note the long
outstretched wings, considerable length of the arm relative to the hand,
leading edge mildly angular at the carpal joint and trailing edge gently
angular, wing tips moderately blunt, a short tail relative to the wing span, a
‘band rump’ extending to the lateral undertail-coverts still clearly visible
when banking revealing the underside (as in this photo), and all dark
underwings and body. Also note the Madeiran is about three metres above the sea
surface. The photograph is a fine record shot that clearly shows several key
field features that combined are diagnostic of Madeiran Storm-petrel.
Sketch By Tony James.
Elimination of other species Comparisons are between our Madeiran and other possible storm-petrel species.
European Storm-petrel Hydrobates pelagicus
European is much smaller and those present were dwarfed by our
Madeiran. Knowledge of size illusion between storm-petrel species at
sea prevented wild comparisons such as ‘twice as big’ and the known
relative wingspan and body length of 1.17 and 1.27 respectively felt
correct. European flies like a bat with weak fluttery flight compared
to the stronger, methodical, Nightjar-like flight of our Madeiran.
European has short arms and hands and blunt-ended wings that always are
strongly angular at the carpal joint on both leading and trailing
edges, whereas the wings of our Madeiran were long with arms broad and
medium length and hand long, with wings held out quite straight so that
leading and trailing edges were only moderately angular. European tail
is short and gently rounded whereas the tail of our Madeiran had a
shallow fork. European has a compact short body whereas our Madeiran
was long and chunky. European bill is short, slim, and only slightly
hooked whereas our Madeiran had a bulky bill that was heavily hooked.
European exhibits a pale upperwing covert pencil line, whereas our
Madeiran showed covert bars. European has white underwing covert
panels, whereas the underwing of our Madeiran was evenly hued
Wilson’s Storm-petrel Oceanites oceanicus
Wilson’s is smaller and indeed we had seen fair numbers of them off
Scilly in the forerunning two months so the larger size of our Madeiran
when it first arrived was immediately apparent. Wilson’s jizz and
flight behaviour is well characterised by the hirundine simile so
different from the methodical, Nightjar-like flight of our Madeiran.
The wingspan of our Madeiran immediately was visibly longer than
Wilson’s based on recent observations of Wilson’s. Wilson’s compared to
our Madeiran has shorter wings and is more angular at the carpal joint
on the leading edge and straighter on the trailing edge. The head on
profile of Wilson’s shows wings straight and stiff, arms short, hands
medium length, whereas the wings of our Madeiran were slightly bowed in
a shallow-M, arms medium length, and hands long. Wilson’s tail is
medium length, slightly concave with rounded corners, whereas the tail
of our Madeiran had a shallow fork. The rear carriage of Wilson’s looks
quite long relative to the wing span, but the rear carriage of our
Madeiran looked particularly short compared to the wing span. Wilson’s
has long spindly legs that normally project beyond the tail in
travelling flight, unlike our Madeiran. Wilson’s has an evenly
proportioned body whereas our Madeiran was chunky. Wilson’s bill is
medium length and only slightly hooked, whereas our Madeiran had a
bulky bill that was heavily hooked. Wilson’s normally has obvious and
broad upperwing covert bars that start short of leading edge and extend
to body, which are far more obvious than the covert bars on our
Leach’s Storm-petrel Oceanodroma leucorhoa
Our Madeiran flew with methodical, Nightjar-like flight whereas Leach’s
flies with buoyant and graceful, deep, languid wingbeats, and may make
unpredictable speed and direction changes, darting, vertical leaping,
and bounding ahead. Leach’s wings are long and quite narrow with
decidedly pointed wing-tips, and leading and trailing edges are
strongly angular, whereas our Madeiran had long wings held outstretched
so that leading and trailing edges were only moderately angular, and
the wing tips were blunt. Our Madeiran looked longer winged than
Leach’s because mainly it held its wings outstretched, whereas Leach’s
tends to hold its wings swept back. The tail of Leach’s is deeply
forked and scooped, whereas the tail of our Madeiran had a shallow
fork. Leach’s body is rather long and slim, whereas our Madeiran had a
chunky body. Leach’s bill is relatively long, slender, and only
slightly hooked, whereas our Madeiran had a bulky bill that was heavily
hooked. The upperwing covert bars of Leach’s normally are striking
stretching across the wing coverts to the leading edge, whereas our
Madeiran had duller covert bars that ended short of the leading edge.
Leach’s white patch on the rump is dull and not gleaming white, barely
extends to undertail-coverts and rear flanks, is longer than it is
broad, and is hard to observe at sea especially when the bird banks
away, whereas our Madeiran had a narrow, rectangular white patch,
broader than it was long, that extended to the lateral
undertail-coverts and rear flanks, and seemingly was always visible
even when the bird banked away.
Black-bellied Storm-petrel Fregetta tropica and White-bellied Storm-petrel F. grallaria.
Fregetta storm-petrels are fat and compact, whereas our Madeiran was
quite long bodied but chunky. Fregetta storm-petrels fly like an
exhibition skateborder and tend to skim close to the sea surface,
whereas our Madeiran flew with methodical, Nightjar-like flight
occasionally rising to three metres above the sea surface. Fregetta
wing shape is broad with short arms, strongly curved leading and
straight trailing edges, tapering to pointed wing-tips, whereas our
Madeiran had long, broad arms, blunt-ended wing-tips, with wings held
straight out so leading and trailing edges were only moderately
angular. Fregetta head on profile shows wings slightly bowed in a
shallow-M with arms short, hands long, whereas our Madeiran also had
wings slightly bowed, but arms were medium length and hands were long.
Fregetta tail shape is short and square-ended, whereas our Madeiran was
shortish but with a shallow fork. Fregetta bill shape is short,
broad-based, and finely hooked, whereas the bill of our Madeiran was
bulky and quite heavily hooked. Fregetta upperwing covert bars vary
slightly between species but in both cases normally are less
conspicuous than exhibited by our Madeiran. Fregetta underwings have
obvious white covert panels, some outer greater underwing-coverts
dark-centred with white fringes, whereas our Madeiran underwing was
evenly hued black-brown. Both Fregetta species show extensive amounts
of white on the belly, though the pattern varies, however, our Madeiran
had an all dark belly.
RL and Thomas, B (2007) ‘Identification of ‘black-and-white’
storm-petrels of the North Atlantic’, British Birds, 100: 407-442.
RL and Fisher EA (2007) Flight Behaviour of ‘black-and-white’
storm-petrels of the North Atlantic, DVD, privately published via