sal get a tet No
Lg WET a Ra ee elle pte Ae
a on er
¥ ea

Madera oh Vy sie Pied a Bia de ae LD!

ete hat ACER bet NS aft willy Beteaiin tae tla?
iweb t atlas’ anal italia at

Fon Ao iag leh ak panel MEE A

Dee er ek Ed car en ae
ON Mig att eV ot ely Mica nt Ady

gel BLE A he
a talin Rat eee rete Bey ofa MY *8G
A ha WP eS ell) hE VOU YeeA Nel WE Wide tre ela atatels
Late
arte 3 Distnet
caper seed meu
anit eta Min 2M A aH
Sea Sae eta Had yl so

Sind agin Bed eta asia Oe

Mgt eT at wel ee
pee Tact ate

aE BO ew 4
wt Eee ¢
spi wb id EAP AM Dat NE wp Med ete pins Pate Eee

8 ye aha Patt

anata
Ne Vat
eee Ret

tae Se
NE Be Med Ee
Cyeretnll

Pees Dear re
Piet at alia
aa eae Na

eriety

Hite (Fair ideo te tier yrergeey

w

rE paieay

Aes AU AAA QA

ender Oe NC AG

Sue vi VWebo ue Gerad luke lager

Ly Bud OR Tog GAN ORS
Le dypt eee at

Pe OE Sth a
VP andy

¢

aad eee ee
rude ane
Vee eh ae
Perna seen
anette, yt
Pea

CaN OE INT
ape ed odes PB
Pat tp Ms ae 2

Trane
ea deds

genes
fens dndve

ad TOE PW
pide fe ahee et

wn ere oe
weep tet
‘

eae
Sv TAIN
we UshGe
rerenerros
PR Daan

nw Te eee

pay

Reha eee ei
Wed aged eres thats ESS pute wea

Oe
Mp ee

Pre tne ees
Pape eted gy tee rates feat

aus ety
er ee ee
ere Teh W een Se

LEDER ONG

yd PEEVE RSET OMEN Lh

atv nw
awe sealet
ave
yey th

uv
Poe Ca
2b vagtthe ty

renee

OreeeLeD

Der SVv han tewe ¢
ateeeey

tN 7M ean

PE LEE
Spb ah pees gaa eh Gh ww wt geeey

eres.

eke rig end eee

Ty OTe aT e DE
ey uiiee es sony
UR Dd ey MP MES
ree ee eee

one are rere tat ea)

eae SPE AUNT an INANE

tpadige dy ends an PS ody CE AEE UE EE
$ tas voywe dv ane
srorn

fered)
Seype es

sone

er Pier ree tes

ews
Wide nity

Pacey

apse emeharéhe
err eit as Wont
LP a gh TA MD

creer eee ea

BNI DM at NY

"

' pea peek
Cra edete IT aN PVG Bete sh Tb Et EO SESE aN GETS
Fe agen eg IN geared yg by AD
pare
Vipake
ee eT I forens
DPN eed ph Oa ae tm gn AY See yee gen sees
Pe Aree, Wir eae oe Ao
Venere ey
Perea ee nay
LP ee
Serer ere
Pn ee
ay
ere)

See ee Dare CSE ig
SAUTE Laden ed
ne eee) or
erty

Vary

ayb ony antes dye yoke

VY bath eg ye WAS VE

hes VAN

PE Ae ON PL Duet gun

Hererrerh a7 aries

see

sea

Lares tay ove rwe
et

evs ang bed sae

Deawes heh
ath he
Ay ate

pre.

“4 vs 2 ey

“nN

cere
”

POON VINA UTNE

Surg Bee ety
vie INS
Reet ees

WS Seger te

tere f

es eleds

igh Z ot ae mag By

6

cn fal her ye ev NER Sh A A MNS ee ee Ea oe BM

eB OR a Ae et ER ae 6 et Be PEW VERS w ea rey

ee) Peete Re eae

A phat sd has DA Meye at eA dled Me Sao te

afl beh Bat A OTL OG

~ RM dtd CAEN a

SAN tS test Aube rebels (8 > EDD APTA

A) ee eee Ne Grw OE Usha boa athe me Ve 5 oe!
Eee gy ba SRS R Ne ERRNO) Pee cork eB DST AT ARRAN,

eageaemey

<- pres feb he Ee

Satie eee
EN

vee?

hate t
BHF
ere

othe Fae
oe ON a le ee Lae Net a

wees wh tease
she eters

LVAD eee Se

tant
ro ERs
ieee
as
Tot. ee
i
f

~
boar
,
.
———
~
~
k
x
ree
]

——-

Scottish Birds

The Journal of the Scottish Ornithologists’ Club

Editor: Anne-Marie Smout

Assisted by: Professor D. Jenkins, Dr J.B. Nelson and Professor P.J.B. Slater
Business Editor: The Secretary, S.O.C., 21 Regent Terrace, Edinburgh EH7 5BT
(tel. 0131-556 6042).

Scottish Birds, the official journal of the Scottish Ornithologists’ Club, publishes original material
relating to ornithology in Scotland. Papers and notes should be sent to The Editor, Scottish Birds, 21
Regent Terrace, Edinburgh EH7 5BT.

Two issues of Scottish Birds are published each year, in June and in December. Scottish Birds is
issued free to members of the Scottish Ornithologists’ Club, who also receive the quarterly
newsletter Scottish Bird News and the annual Scottish Bird Report. These are available to non-
members at a subscription rate (1992) of £30.

The Scottish Ornithologists’ Club was formed in 1936 to encourage all aspects of ornithology in
Scotland. It has local branches which meet in Aberdeen, Ayr, the Borders, Dumfries, Dundee,
Edinburgh, Glasgow, Inverness, New Galloway, Orkney, St Andrews, Stirling, Stranraer and Thurso,
each with its own programme of field meetings and winter lectures. The Waterston Library at the
Club’s headquarters at 21 Regent Terrace, Edinburgh EH7 5BT is the most comprehensive
ornithological library in Scotland, and is available for reference during office hours (Monday to
Friday, 9.00 - 17.00 hrs). (But please phone beforehand). A comprehensive stock of Scottish local
bird reports is held at headquarters and may be purchased by mail order.

SOC annual membership subscription rates

AGU ltt saas dvssccecsesconenntecendeccdsinceeneas eb dencdeeiiewdats nella ts aeetGch eee ner ee ocean Fl5:00
Family (husband & wife, & childrem under 18) <ccscccccccccceeescoceeeenceeeeeeeee see eee eee £22.00
Junior (under 18, or student UNder 25))...25.:..cccceececndeceeenceeceenenecsnaseecete 2 eeeee ee eee £6.00
Pensioner/Unwaged .......s.00.ccsessve-ecieit.ncdass eedsnconeanenoceeereemechGandescockeunee=eee ee ee ar £8.50
Pensioner Farmlly’ &. cic. csosaccsccus este ceppuetoens oa vnne acs oneee coesueeceneeeaeteeaneeseeseeee £12.00
LRG: acbsedlecbonncdamenuashaebtincwtirvcsssudctiemnaedeautcat ona: sete a aRee Ra nene eee eee te £300.00
Lo: Fann Y sascesae acthncte'daciacsocsedaetaeasaden Aonvate ascec weno mete e gente dee emer eer £450.00

All subscriptions may be paid by Banker’s Order and Covenanted. Subscriptions paid this
way will greatly assist the Club.

Published by the Scottish Ornithologists’ Club, 21 Regent Terrace, Edinburgh EH7 5BT.
Printed by Milne, Tannahill & Methven Ltd., 113-119 Glover Street, Perth PH2 OJE.

Scottish Birds (1995) 18: 1-10

The status of the Eider on the Isle of May and other Forth

Islands.

J. CALLADINE, M.P. HARRIS, S. TAYLOR & S. WANLESS

Eiders have bred on the islands in the Firth of Forth for centuries, for
instance in the early 17th century they were said to be among the
commonest birds found on the Isle of May (Rintoul & Baxter 1935). In
recent years, counts of nests on those Forth Islands managed as nature
reserves (Isle of May, Inchmickery and Fidra) have suggested a marked
increase. This paper documents the past and current breeding status of the
species on the Forth islands, including a systematic survey in 1994. Many
Eiders used to breed on the Fife mainland; up to 1800 were thought to nest
in the Tentsmuir and St Andrews areas in the early 1970s (Smout 1986).

Very few breed there now (Fife Bird Reports), suggesting a eae’ shift in

the breeding population.

Introduction.

Amongst the Forth islands, the largest
concentration of Eider Somateria mollissima
nests has always been on the Isle of May, yet
very few young other than those recently
hatched are ever seen on or around the
island. The number of well grown,
approaching fledging, young seen per year
varies from 0-15 despite there being up to
1000 nests (Isle of May wardens’ reports).
Over many years a small number of females
with young have been seen leaving the north
end of the Isle of May. In addition, fishermen
and other boat users reported occasional
groups of ducklings crossing the eight
kilometres of open water between the island
and adjacent Fife coast. In 1994, we attempted
to establish whether Isle of May ducklings
reached the Fife coast and determine roughly
the extent of that dispersal.

Methods

Past nest counts come from published ana
unpublished sources. In 1994 all the Forth
islands (except the Bass Rock) were
systematically searched for Eider nests during
the last week of May. The timing and
methodology were consistent with annual
counts on the nature reserves. Where
appropriate, the islands were divided into
sections, each section being covered by a
team of four or five counters walking transects
3-4m apart and eventually covering the entire
section. Afew areas were missed because of
inaccessibility or to prevent disturbance to
other breeding birds.

On the Isle of May, prior to the counts in 1983-
1994, arandom sample of nests (usually 100)
were marked with short, unobtrusive canes.
During the count each nest found was

2 J.Calladine et al.

recorded as either marked or unmarked. The
proportion of marked nests found was used
as a measure of nest finding efficiency. For
example, if 85 of the 100 marked nests were
found it was assumed that 85% of all nests on
the island were found. The actual nest court
was accordingly corrected to give a more
accurate estimate of the actual number of
nests. Asimilaraccuracy check and correction
was not possible on the other islands, so
these counts are probably serious
underestimates of the true breeding
population.

Between 20 May and 15 June 1994, 120
incubating female Eider on the Isle of May
were marked (under licence) with 4cm-long
pieces of electricians’ tape on the back of the
head. The tags were attached using a three
metre cane to avoid a close approach to the
birds and adhered to feathers with superglue.
Only birds which were not excessively agitated
at the approach of the marking cane were
tagged. No birds deserted as a result of
marking.

From early June to mid-July (after the main
hatching period), the adjacent Fife mainland
coast from St Andrews to Lower Largo was
systematically searched by S T. In addition,
the coast from St Andrews to Tayport, and
from Lower Largo to Kirkcaldy was less
intensively search (Table 1). For ease of
observation, and hence efficiency, most
searches were made within two hours either
side of high tide. All Eiders, including
ducklings, were recorded and the females
checked for tags.

SB 18 (1)

Table 1. Dates of searches of the
Fife coast for Eiders, 1994.

Date Area searched
June 1 St. Andrews - Eden Estuary
4 Kinkeil Braes
8 Crail - Cellardyke
9 St. Monance - Elie
10 Earlshall
12 St. Monance
16 Crail
UW Kingsbarns - Fife Ness
22 Methill - Ruddons Point
29 Ruddons Point - Earslferry
25 Elie - Crail
27 Crail - St. Andrews
28 St. Andrews - Eden Estuary
29 Guardbridge - Tayport
July 1 St. Monance - Elie
2 St. Andrews - Kinkeil Braes
4 Eden Estuary
7 Crail - Anstruther
9 Kirkcaldy - Lower Largo
12 Ruddons Point
2 Earlsferry - Elie
15 St. Monance - Fifeness
16 Eden Estuary
Results

A total of 2331 Eider nests were found on
eightislands in 1994. Comparable counts are
available for seven of those islands from 1987
(Table 2).

1995
Table 2
1987
Inchgarvie no count
Inchcolm 128
Inchmickery 118
Inchkeith 187
The Lamb 3
Fidra 163
Craigleith 182
Isle of May 635
Total 1416+

Status of Eider on Isle of May & Forth Islands 3

Counts of Eider nests on Forth Islands 1987 and 1994.

1994

60
258
313
323

6
220
153
998

2331

Note: Isle of May nest counts are corrected for nest finding efficiency (see methods)

a. Isle of May

As already mentioned, inthe early 17th century
Eiders were said to be common and were
recorded through the 18th and 19th centuries
(Rintoul & Baxter 1935). Persecution was
apparently responsible for a decline in the
late 19th century. For the first 60 years of the
20th century, the number of nests varied
between 20-100, with perhaps an increase in
the late 1950s (Eggeling 1960). In the 1960s
and early 1970s, between 100-200 Eiders
nested each year, followed by a marked
increase from the mid-1970s, to the present.
Since 1978, when systematic counts began,
the number of nests has increased by an
average of 8.5% per year (Fig. 1). The annual
fluctuations of nest counts are not unexpected
because most recruitment of Eiders into the
breeding population is probably by birds
hatched in a relatively few, particularly ‘good’
years (Mendenhall & Milne 1985). Also the
proportion of adults with breeding experience

nesting in any one year can be highly variable:
45%-100% has been recorded (Coulson
1984).

2. Inchkeith

Eiders nested in the 17th century (Rintoul &
Baxter 1935), butthen declined, disappeared,
or perhaps simply went unrecorded for many
years. A count of 18 nests in 1960 was then
considered the highest ever (Smith 1974).
Concurrent counts of ducklings (69 in 1962
and 75in 1963) suggested many more actually
nested. In 1969, 120 adult females, 60 young
and an additional 20 nests with eggs were
found (Smith 1974). Acount of 187 nests was
made in 1987 (Nature Conservancy Council).
In 1994, 323 nests were counted, suggesting
an increase at a similar rate to that which
occurred on the Isle of May.

4 J.Calladine et al.

3. Inchmickery

Eider eggs were taken in the mid-1800s; a
single nest was found in 1934, suggesting a
decline in the late 19th and early 20th centuries
(Rintoul & Baxter 1935). Up to 100 females
nested in the 1970s and early 1980s, but
there has been a marked and sustained
increase (11.3% per year) since 1987 (Fig. 1).

SB18(1)

4. Inchcolm

In common with some other Forth islands,
Eiders bred in the 18th century and had either
declined or disappeared by the late 19th
century (Rintoul & Baxter 1935). Systematic
counts found 128 nests in 1987 and 120 in
1988 (NCC counts). The population appears
to have doubled since that time to 258 nests
in 1994 (Table 2), equivalent to a similar rate
of increase to that which occurred on
Inchmickery.

Figure 1. Counts of Eider nests on the Isle of May, Inchmichery and Fidra, 1978-

1994.

Notes: 1) Nest counts are plotted on a log. scale
2) Rate of increase on the Isle of May = 8.5% per annum (17 = 0.68)
3) Rate of increase on Inchmickery = 11.3% per anum (17 =\0:83)
4) Regression not significant for Fidra

1000

@ Isle of May
O Inchmickery
+ Fidra

Nest Counts

100
18” 79" 8031" 82 "85" $4

85 86
Year

Isle of May
|

Inchmickery

87 88.89 90. OL. O27 ame

1995

Table 3. Counts of Eider ducklings on mainland Fife coast.

Area
Eden Estuary

St. Andrews

St. Andrews-Boarhills

Boarhills-Kingsbarns

Kingsbarns-Cambo

Cambo-Fife Ness

Fife Ness-Crail

Crail

Crail-Cellardyke

Cellardyke

Cellardyke-Anstruther

Anstruther-Pittenweem

Date
4 July

1 June
28 June
2 July
23 July

8 June
27 June

27 June

17 June
27 June

17 June
27 June
15 July

15 July

8 June
16 June
7 July
15 July

8 June
25 June

8 June
25 June
7 July
15 July

12 June
25 June
15 July

25 June
15 July

Status of Eider on Isle of May & Forth Islands

No. of ducklings

— NM ON

6 J.Calladine et al.

Area

Pittenweem-St. Monance

St. Monance-Elie

Elie-Earlsferry

Earlsferry-Ruddons Point

Ruddons Point-Lundin Links

West Wemyss-East Wemyss

Isle of May

SB18(1)

Date No. of ducklings
12 June 4
15 July 3
25 June 16
15 July 1
8 June 4
25 June 7
13 June 11
22 June 3
25 June 20
22 June 6
25 June 16
9 July 36
9 July 5
9 June 69
17 June 15
5 July 1
20 July 5

Note: Maximum number of ducklings seen in each area is shown in bold.

5. Craigleith and Fidra

Eiders nested on these and neighbouring
islands in the late 19th and early 20th centuries
(Rintoul & Baxter 1935). In contrast to some
other Forth islands, annual counts on Fidra
between 1987-1994 suggested a stable
population of between 142-220 nests (Fig. 1).
Craigleith was the only island where the 1994
count was actually less than that in 1987,
although the difference was relatively small
(Table 2).

6. Post-hatching dispersal from the
Isle of May

Eider ducklings were seen on the Isle of May
and on the Fife coast from the Eden Estuary
in the north to West Wemyss in the south
(Table 3). Amaximum of 445 ducklings were
found within the area searched. Twenty-four
tagged females were seen attending newly
hatched young on and around the Isle of May
between 24 May-20 June (casual
observations only). Thirteen tagged females

1995

Status of Eider on Isle of May & Forth Islands 7

Table 4. Sightings of tagged female Eiders on mainland Fife coast.

Location Distance Date

from Isle

of May

(km)
St. Andrews 26 1-26/6
Cambo 16 17/6
Cellardyke 9 8/6
Cellardyke 9 25/6
Anstruther 9 12-25/6
St. Monance 13 12/6
Ardross 15 25/6
Elie 17 25/6
near Elie 17 25/6
Kincraig Point 20 13/6
Lower Largo 25 22/6

Number of | Number of Number of
tagged untagged ducklings
females females in being
attendance attended

2 0 3

1 5 25

1 5 6

1 2 3

2 0 5

1 3 4

1 9 2

1 4 3

1 0 1

1 1 11

1 44 6

Note: Distances from the Isle of May are minimum distrances travelled assuming no

overland crossings are made.

were seen off the Fife mainland coast, all
attending young (Table 4). Female Eiders
commence moulting their head and neck
feathers in July (Cramp & Simmons 1977) so
some marked females will have lost their tags
before being seen away from the nest, through
moult and also, no doubt, through inadequate
adhesion of tags to feathers. The 13 birds
seen therefore represented the absolute
minimum number of the original 120 marked
birds which potentially took ducklings to the
mainland.

Discussion

Counts from the Isle of May and other islands
in the Firth of Forth show an increase in the
number of breeding Eiders in recent years.
Between 1987-1994, the combined
populations of the Isle of May, Fidra, Craigleith,

Inchmickery, Inchcolm and The Lamb
increased by about 62% (Table 1). The 1994
survey coincided with record counts from the
regularly monitored islands. The combined
total of 2331 nests for the Forth Islands
represents 7% of a recent estimate of the
British breeding population (Gibbons et al.
1993). The Isle of May alone held 3% of the
British population, while Inchmickery’s 313
nests (1% ofthe estimate) on just 1.1 hectares
was no less impressive.

On the Isle of May, nest finding efficiency has
averaged 78% over the past seven years. In
addition, a small proportion of birds would not
have commenced laying at the time of the
count; on Coquet Island (Northumberland)
5% of females had notlaid by 28 May (Coulson
1984) and some failed nests could have been
overlooked. No doubt, therefore, the counts

8 J.Calladine et al.

of nests on the Forth islands were
underestimates; probably near 3000 female
Eiders nested in 1994.

On the Isle of May, the increase in nesting
Eiders probably began shortly after the
departure of the last dog in 1973 (following
the evacuation of the lighthouse keepers’
families in 1972) and coincided with a period
of reduced breeding gull numbers, from 17,000
- 20,000 pairs in 1972 to about 3000 by 1986
(Isle of May wardens’ reports). It has been
implied that gulls are important predators of
Eider ducklings (Munro & Bedard 1977,
Mendenhall & Milne 1985). The effect however
of such predation on the overall survival of
ducklings and eventual recruitment into the
breeding population is questionable (Swennen
1989). Swennen found that female Eiders
were effective at defending ducklings from
gull attacks and those ducklings that were
taken did not react fast enough to alarm calls
uttered by the females; it was suggested they
could be unfit and unlikely to have survived to
maturity anyway. This could be the case on
the Isle of May and other Forth islands. The
recent increases in the Eider populations of
Inchkeith, Inchmickery and Inchcolm have
coincided with substantial increases in their
breeding gull numbers, suggesting the relative
unimportance of gull predation. On Inchcolm,
from 1987-1994 the combined population of
Herring Gulls Larus argentatus and Lesser
Black-backed Gulls L. fuscus increased by
86% to about 3300 nests (NCC/Scottish
Natural Heritage counts), while on Inchmickery
over the same period the increase was by
77% to 216 nests (Royal Society for the
Protection of Birds counts). An ‘animal
sanctuary’ including cats, dogs and a pig was
kept on Inchkeith between 1987-1991. The
effect of the animals on ground nesting birds,
including Eiders, can only be presumed but it
is perhaps not coincidental that Inchkeith’s

SB18(1)

Eider population increased at a lesser rate
than on neighbouring Inchcolm and
Inchmickery during that period; the dogs and
pig were seen to eat Eider eggs. Inchkeith
holds large numbers of breeding gulls; a
combined total of 7600 nests of Herring and
Lesser Black-backed gulls were counted in
1994 (SNH). It would be interesting to monitor
any changes in its Eider population, following
the abandonment of the animal sanctuary, in
the face of a large and increasing gull
population (30% since 1987). Arepeatcensus
in a further five years should be made and
might help to clarify the relative importance of
the effects of ground predators and gull
predation on Eider populations.

On the Isle of May monitoring of hatching
success suggested that the island is a safe
place for Eiders to nest; 70-95% of nests
successfully hatched eggs in each of the
years 1984-1994 (Isle of May wardens’
reports). The paucity of records of well-grown
ducklings around the island implies that there
is insufficient accessible food (for ducklings,
attendant females, or both) within its
immediate vicinity to maintain the population
which use the island as a safe nesting area.
Observations of Isle of May tagged birds on
the mainland Fife coast in 1994 (Table 3)
suggested that a proportion of the breeding
Eiders have to leave the island in order to rear
young.

Exchange of ducklings and aitendant females
within Eider creches is common (Swennen
1989) but many females manage to keep
their own ducklings with their creches (Bustnes
& Erikstad 1991). Those female Eiders which
attend and defend young tend also to be
those which successfully hatched eggs, even
if they initially abandoned their own broods
and later attended a créche (Bustnes &
Erikstad 1991). It is reasonable to assume

1995

that at least a proportion of those ducklings
seen with Isle of May marked adults originated
from the Isle of May. Considering the potential
output of ducklings from the Isle of May (998
nests, each with 3-5 eggs of which over 70%
hatch) and the mere handful of Eiders now
suspected of nesting on the Fife mainland
(Fife Bird Reports), most of the ducklings
found between St Andrews and Lower Largo
(the most distant sightings of marked adults)
must come from the Isle of May. As some of
the ducklings travelled at least 26 kilometres
within 7-10 days of hatching (Table 4) including
a minimum of eight kilometres of open sea, it
could be possible for ducklings from the Isle
of May to cross the 15 kilometres to the
Lothian coast. Further work would be required
to determine how many ducklings are
successfully reared and where they are reared.
In any case, it appears that duckling survival
must be greater than was suspected from
observations on the Isle of May itself.

Considering the national importance of the
Firth of Forth’s breeding Eiders, and those of
the Isle of May in particular, any conservation
management should consider their
requirements. It is essential that the island is
protected from ground predators, including
even casual dogs or other animals. The
impact of the increasing pressure from human
visitors should also be assessed. Human
disturbance, albeit usually inadvertent, leads
to predation of both eggs and young by gulls
(pers. obs.) and this is likely to occur
irrespective of duckling fitness and hence
individual fledging potential. The potential
impact of spillages from the concentrated
shipping in the Firth of Forth must also be
considered. Certainly monitoring the Forth’s
breeding Eiders is desirable, but the wide
dispersal of ducklings from sites such as the
Isle of May increases the difficulty of measuring
productivity.

Status of Eider on Isle of May & Forth Islands 9

Acknowledgements

For help with the 1994 counts we thank the
staff of SNH, Melanie Findlay andthe Scottish
Wildlife Trust survey teams, Dave Fairlamb
andthe RSPB (also for providing some earlier
nest counts), Ronnie Morris, Su Bryan, John
Logie, Douglas McLean, Chris Wernham and
those who carried out earlier counts. Keith
Ferry assisted with marking of incubating
Eiders. We are grateful to Historic Scotland
and Mr. T. Farmer for permission to visit
Inchcolm and Inchkeith respectively.

References

Bustnes, J.O. & Erikstad, K.E. 1991. The role
of failed nesters and brood abandoning
females in the creching system of the
Common Eider Somateria mollissima.
Ornis. Scand. 22: 335-339

Coulson, J.C. 1984. The population dynamics
of the Eider Duck Somateria mollissima
and evidence of extensive non-
breeding by adult ducks. /bis. 126:
525-543

Cramp, S. & Simmons, K.E.L. (eds) 1977.
The Birds of the Western Palearctic
VOI i @nO re

Eggeling, W.J. 1960 The Isle of May. Oliver
and Boyd.

Gibbons, D.W., Reid, J.B. & Chapman, R.A.
The New Atlas of Breeding Birds in
Britain and Ireland 1988-1991.Poyser.

Mendenhall, V.M. & Milne, H. 1985. Factors
affecting duckling survival of Eiders
Somateria mollissima in north-east
Scotland. /bis 127: 148-158

Munro, J. & Bedard, J. 1977. Gull predation
and creching behaviour inthe Common
Eider. J. Anim. Ecol. 46: 799-810

Rintoul, L.J. & Baxter, E.V. 1935. A Vertebrate
Fauna of Forth. Oliver and Boyd.

10 J.Calladine et al. SB18(1)

Smith, R.W.J. 1974. Seabirds of Inchkeith. Swennen, C. 1989. Gull predation upon

Edinburgh Ringing Group Report Eider Somateria mollissimaducklings.

1974: 7-12 Destruction or elimination of the unfit?
Smout, A-M. 1986. The Birds of Fife. John Ardea 77: 21-45

Donald

J. Calladine, Scottish Natural Heritage, 46 Crossgate,
Cupar, Fife KY15 5HS

M.P. Harris, Institute of Terrestrial Ecology, Hill of Brathens,
Banchory, Kincardineshire AB31 4BY

S. Taylor, 7 Moir Crescent, St Andrews, Fife KY16 8XN

S. Wanless, Institute of Terrestrial Ecology, Hill of Brathens,
Kincardineshire AB31 4BY

Revised typescript received 28 December 1994.

Scottish Birds 18: 11-19

17

Lek habitats of Capercaillie at Abernethy Forest, Strathspey

R. W. SUMMERS, R. PROCTOR &S. 1. H. SHAH

The habitat around 26 stances at five Capercaillie leks at Abernethy Forest
in 1993 was described. Leks occurred in old open semi-natural pinewood
and an old plantation (60 years old). Stances were on slopes, ridges and
flat ground, and for those on slopes, aspects were generally northerly. The
tree density at Abernethy Forest was lower than at leks in Scandinavia and
Estonia. The field layer around the stances was dominated by tall heather
and Vaccinium, though the stances themselves had short vegetation. Two
of the leks were close to tracks so, in the future, human disturbance may

pose a problem.

Introduction

The Capercaillie Tetrao urogallus is a Red
Data bird in Britain because of its localised
distribution and recent decline in population
size (Batten et al. 1990, Gibbons etal. 1993).
Male Capercaillie display collectively at tradi-
tional sites called leks. Alek is normally used
by several males, each of which displays
primarily on afew square metres of ground
(its stance). Hjorth (1982) has likened the lek
stance of each male to the innermost small
portion of a wedge-like home range extend-
ing outwards. This is known as the ‘piece of
cake’ model.

Lek sites in Scandinavia are often in old
woodland on elevated areas in forests where
the under story is open and visibility is more
than 30m (Hjorth 1985, Rolstad & Wegge
1987a). In Estonia, leks are mainly found in
old pine forest on bogs with no undergrowth
(Vint 1991). In a study in the Austrian Alps,
Capercaillie preferred a south-south-east
aspect for lekking, suggesting a preference
for warmer hillsides in the early morning
(Spitzer 1985). Given the preference for old
forest and certain topographical features, it
seems likely that leks will occur in only certain

places, so that modern forestry is potentially
deleterious if it changes the characteristics of
these places. In Scandinavia it was found
that moderate logging at lek sites was
acceptable as long as tree density was not
reduced below 500 trees per ha and clear
cuts were not greater than 50m in diameter
(Rolstad 1989). Large clearfells (greater than
20ha) and heavy thinning (leaving less that
400 trees per ha) reduced the numbers of
birds using lek sites (Rolstad & Wegge 1989a),
as did general fragmentation of the forest
because the number of leks in an area
depended on the size of patches of old forest
(Rolstad & Wegge 1989b). The smallest
occupied forest was c.50ha, and the number
of leks increased with one lek added for each
2.5-3km@ increase in forest size (Rolstad &
Wegge 1987b). In Scotland, Picozzi et al.
(1992) found that Capercaillie leks were also
associated with old forests, which were often
semi-natural. Despite this and other studies,
habitat descriptions of leks have tended to be
general. This study sets out to describe the
habitat around stances at Capercaillie leks in
Abernethy Forest, one of the largest pieces of
semi-natural forest left in Scotland (Steven &
Carlisle 1959, Bain & Bainbridge 1988).

12 R.W. Summers et al. SB 18(1)

Figure 1. Diagrammatic representations of trees within difference forest structure
types, based on their “granny” and “plantation” scores. Stipple - canopy, black -
canopy overlap with neighbours, white - dead branches. Semi-natural pinewood
typically falls within types 2-10. Adapted from Picozzi et. al. (1992).

Granny Score

Plantation Score

1995

Study area and methods

Abemethy Forest (57°15'N, 3°40’W) covers
about 30km? of the northern slopes of the
Caimgorm Mountains and extends into Strath-
spey. Old semi-natural woodland comprises
about two-thirds of the forest, the rest being
plantations of different ages. Scots pine
Pinus sylvestris is the dominant tree. The
semi-natural forest has had selective felling,
leaving clearings with some trees which act
as seed sources for natural regeneration.
However, because the red deer Cervus
elaphus population has been high for about
100 years and deer browse young trees,
there has been little regeneration in the wood-
land, leading to an old (greater than 100
years) open forest with a park-like appear-
ance in some areas (Steven & Carlisle 1959,
O’Sullivan 1973, Heard 1988).

In order to describe the woodland at the leks,
we made the following measuremenis from
the ten closest trees around each lek stance:
the tree height, diameter at breast height
(DBH) and canopy depth (percentage of the
trunk that had foliage). The density of at least
ten trees greater than 1m tall within a 4-25m
radius of the stance was measured. The
woodland structure in terms of two principal
component scores (a ‘granny’ score and
‘plantation’ score) was described (Picozzi ef
al. 1992, Fig. 1). Increasing ‘granny’ scores
indicate a tendency towards open semi-natu-
ral pinewood whilst increasing ‘plantation’
scores indicate a trend towards evenly-spaced
dense woodland with no ground vegetation.
Lek habitats were related to available habi-
tats based on the areas of each woodland
type, described in terms of their ‘granny’ and
‘plantation’ scores (Raistrick 1995). The field
and ground layers of vegetation were as-
sessed by pacing a grid, 5 x 10 paces, and
recording the plant species at each step.

Lek habitats of Capercaillie at Abernethy Forest 13

Also, the heights of heather Calluna vulgaris,
Vaccinium myrtillus and V. vitis-idaea were
recorded using a sliding polystyrene disk
(diameter 24cm and mass 30g) on a gradu-
ated stick whose end rested on the ground.
The slopes and aspects of stances, and dis-
tances from stances to the nearest track or
path were also measured. To determine
whether Capercaillie select slopes with a
particular aspect, systematic sampling of the
whole forest was carried out to assess the
availability of different aspects. Using 1:25,000
Ordnance Survey maps, the presence of a
slope and its aspect was recorded at the
intersections of a 0.5km grid. A similar proce-
dure was adopted to determine whether birds
tended to lek away from tracks which are a
potential source of disturbance. The dis-
tances from the systematically chosen points
to ihe nearest track were measured and re-
lated to distances between leks and tracks.

Results

Six leks were known at Abernethy Forest and
habitat data were obtained from 26 stances at
five of these. The average distance between
the nearest leks was 1.9km (standard
deviation 0.9km) and the lek density was one
per 5km of forest. The average distance
between adjacent stances within a lek was
108m (S.D. 62m). Six of the stances were on
ridges, 15 on slopes and five on flat ground.
For 14 (93%) of those stances on slopes, the
aspect was northerly (between NW and NE).
Ninety-five of 196 systematic sample points
throughout the forest were on slopes and
60% of them faced between NW and NE.
Only 18% were between SE and SW. Thus,
the preponderance of north-facing slopes in
the forest can explain why most stances face
north. The habitat at lek stances varied
markedly between the five leks. The median
tree density ranged from 140 - 790 trees per

SB 18(1)

14 R.W. Summers et al.

68-02 LS OSS-SzcS OSS - 0Sz 006-0S8 006 OSz2-SZ SLL (wi) 49eJ} 0} BoUR}sIq
EL-0 L Ze-v kh 8z ZZ-0 9¢ 9€-22 vz OZ-8¢ GS (%) 18A09 B|paeN
66-78 G6 LLZ-Ly Z9 96-82 29 8/-19 Hl 29-92 Or (%) J@A00 SSO
JaAe] PUNOIH)
; - 22-0 8 cL-OL LL O€-S OL 81-0 9 (%) 4@A09 SSBI5)
09-72 O€ 8c-v Gl Zl-c /) vb-zl ve Ov-2z Se (%) J8A09 WNIUIDIe A
19-0 12 Gl-Ze ZS 9Z-0S €9 ve-9l 62 ve-0 Gl (%) J8@A00 JeyyeaH
EL-6 LL co-E GL = = ve-6l 61 SL-EL rot (ww) yyBley WwiniujaoeA
GE-6L UG 8p-8E vy v&-0F Ze LS-Or 7 €p-1Z Ze (Wo) }YyBiey seyyeay
JahKe} pjal4
9G-Le 6E Gl-vS v9 08-EZ hi) GZ-0S OZ Gl-SS 99 (%) Adoueo 9011
LS-LY LV Ly-L\ Ze LS-Zl ve gS-Ee 6€ 96-Sz 92 (WO) HG ‘Jejeweip aes]
O2-Z1 SL Z1-6 OL GL-S OL Li-vl OL QL-EL vl (w) jyBley ees]
OvS-0St S6r Orv-091L 062 O91-OEL Srl Ovz-Orl Ovl O092L-06r 06Z (ey Jed ‘ou) Ajisuep 8e11
SoMeNO ueIpsy| salen uelpey) seen uelpsyy seen ueipey\ sejiueno ueipey\
(=u) G 487 (=u) y81 = (=u) € 481 (g=u) 2481 ~(OL=U) L 497

‘$]94¥9E1g UI UMOUS SI ya] yOea Je Sadue]s Jo Jaquinu ay]
Jsaso4 Ayjausaqy je aljjiediadeg jo saduejs ya] je soysiiajaeseyo JeyigeH “| FJTIGVL

1995

ha (Table 1). Median tree heights were 10 -
18m, and diameters at breast height 26 -
47cm. The canopy depth was related to tree
density, with values up to 77% where tree
density was low. The forest structure at
stances, in terms of ‘granny’ and ‘plantation’
scores is shownin Table 2. Fourleks (numbers
2-5) had stances with high ‘granny’ scores
(forest types 2-10, Plate 1), whilst lek 1 was
mainly in a 60 year-old plantation (Plate 2).
One stance at lek 4 was amongst very young
trees (low ‘plantation’ and ‘granny’ scores).
The forest types at the leks generally reflected
the main forest types found at Abernethy
Forest, in that 46% of the stances were in 46%
of the forest that was old and open (types 2-
10), and 27% in 12% of the forest that was old

Lek habitats of Capercaillie at Abernethy Forest 15

plantation (type 11). However, stances tended
not to occur in the young forest (types 15, 16
and 22). The field layer around the stances
was dominated by either heather (15 - 63%)
or Vaccinium (7 - 35%) (Table 1). The
heather was taller (27 - 44cm) that the
Vaccinium (11 - 19cm). The ground layer
was composed primarily of moss (40 - 95%)
or dead needles (1 - 55%). The vegetation at
the stances themselves was composed of
either needles, moss or very short Vaccinium,
heather or Empetrum nigrum. The median
vegetation height at the stances was 4.5cm
(quartiles 2.6-6.4cm). Stances varied greatly
in their distance from tracks, ranging from 51
- 900m (Table 1). The distances from the
centre of each lek to tracks were 50, 170,

TABLE 2. The frequency with which lek stances occurred in different types of forest
structure and the percentage area of these types in Abernethy Forest. See Fig. 1 for

diagrams of the forest types.

Forest Lek number
type Percentage
1 2 8) 4 5 All area

2 1 1 2 7.4
3 1 1 6 18.6
4 1 1 ike
9 1 1 2.6
10 1 1 2 Bal
11 5 1 1 7 12.0
15 0 5.6
16 1 1 8.1
7 1 1 Wes!
18 3 3 0.9
19 1 1 0.4
22 1 1 8.4
Others 0 7.0

16 R.W. Summers et al.

200, 300 and 700m. In comparison, the
median distance that systematic points were
from tracks was 200m (quartiles 50-300m)
(Fig. 2). While the sample size for leks was
too small for statistical comparison with the
systematic distances, it does show that there
was no tendency for leks to be situated well
away from tracks.

Discussion

The average distance between leks at Aber-
nethy Forest was similar to that in other parts
of Scotland (2.1km) (Picozzi et al. 1992), in
Scandinavia (2.0-2.1km) (Wegge & Rolstad
1986) and in Estonia (1.9-2.4km) (Viht 1991).
In a review of the habitat requirements of
Capercaillie in Scandinavia, Rolstad & Wegge
(1989c) indicated how the habitat provided
the birds’ primary needs. The simplest expla-

SB 18(1)

nation for their liking of old forests was through
their requirements for food (primarily pine
needles and blaeberry Vaccinium myrtillus
leaves and berries) and ability to move through
the forest (to stand on branches large enough
to support their mass, and fly between trees).
Thus, as male Capercaillie are sedentary and
rarely move more that 400m from their lek
(Gjerde & Wegge 19839), itis possible that the
lek habitat is determined partly by the birds’
basic requirements as well as the intrinsic
qualities of the lek (Rolstad & Wegge 1987a).
Capercaillie display in the trees around the
stance as well as the stance itself so require
strong horizontal branches to stand on. The
old ‘granny’ trees at Abernethy Forest pro-
vide this. Old open forest also gives all-round
visibility, which can be enhanced by display-
ing on a ridge (Rolstand & Wegge 1987a). At
Abernethy Forest, we found features of leks

Plate 1. A lek stance (beside the large tree on the left) in old open forest with
a high “granny” score.

1995

Lek habitats of Capercaillie at Abernethy Forest 7A

Plate 2. A lek stance (in the foreground) in old plantation.

which were dissimilar to leks elsewhere. Two
of the leks occurred in park-like landscape
where the tree densities were only 140 and
145 per ha. These tree densities were well
below the acceptable densities (over 500
trees per ha) found in Norway (Rolstad 1989).
Also, leks in Estonia usually occurred amongst
trees at 600-700 per ha (Viht 1991). Spitzer
(1985) noted that leks tended to have a SSE
aspect in the Alps, suggesting that the birds
benefited from the warmth of the morning
sun. Incontrast, the leks at Abernethy Forest
tended to have anortherly aspect. Abernethy
Forest is on the northern side of the Cairn-
gorm Mountains so the majority of the slopes
faced north. However, there are small hills in
the forest which would allow Capercaillie to
choose south-facing slopes. Perhaps it is
more important to select warmer hillsides in
the Alps where it is colder than Abernethy
Forest at the times when Capercaillie display.

One lek in Abernethy is believed to have been
recently abandoned (in 1979) due to human
disturbance (S. Taylor pers. comm.). Thus,
given the apparently shy nature of Capercaillie,
it was surprising to find that there was no
particular avoidance of tracks and one of the
leks was close to a track well-used by vehi-
cles, cyclists and pedestrians. However, the
stances closest to the track are no longer
used, suggesting that the lek has moved back
from the track. As the Capercaillie population
in Scotland is small (Batten et a/. 1990), it is
important that lek sites are protected, both
from disturbance and from destruction through
forestry. Unfortunately, the rarity of the bird
and the attractiveness of its spectacular dis-
play leads bird-watchers to seek out
Capercaillie in spring. This increasing pres-
sure may lead to disruption of leks.

18 R.W. Summers et al.

SB 18(1)

Figure 2. The frequency distribution of distances between systematic points

and nearest tracks in Abernethy Forest.

30

Number

0 250

Acknowledgements

Staff at Abernethy Reserve provided logistic
support. In particular, D. Dugan gave much
help. M. Avery, |. P. Bainbridge, D. Dugan,
P. Mayhew and S. Taylor commented on the
draft.

References

Bain, C. & Bainbridge, |. 1988. Abetter future
for our native pinewoods. RSPB Con
servation Review 2: 50-53.

Batten, L.A., Bibby, C.J., Clement, P. Elliott,
G.D. & Porter, R.F. 1990. Red Data
Birds in Britain. Poyser, London.

Gibbons, D.W., Reid, J.B. & Chapman, R.A.
1993. The New Atlas of Breeding
Birds in Britain and Ireland: 1988-
1991. Poyser, London.

Gjerde, |. & Wegge, P. 1989. Spacing pat
tern, habitat use and survival of
Capercaillie in a fragmented winter
habitat. Ornis Scand. 20: 219-225.

Heard, S.C. 1988. A study of past land use

500 750

Distance (m)

and present forest structure. MSc
Thesis, Univ. Coll., London Univer
sity.

Hjorth, |. 1982. Attributes of Capercaillie
display grounds and the influence of
forestry - a progress report. Proc.
intern. Symp. Grouse 2: 26-35.

Hjorth, |. 1985. The distribution of Capercaillie
males on leks in relation to the forest
structure on the recruiting area. Proc.
intern. Symp. Grouse 3: 217-235.

O'Sullivan, P.E. 1973. Land use changes in
the forest of Abernethy, Inverness-
shire (1750-1900 AD). Scott. Geog.
Mag. 89: 95-106.

Picozzi, N., Catt, D.C. & Moss, R. 1992.
Evaluation of Capercaillie habitat. J.
appl. Ecol. 29: 751-762

Raistrick, P.J. 1995. An evaluation of the
habitat structure and management
policies of Abernethy Forest. BA The
sis, Glasgow Univ.

Rolstad, J. 1989. Effects of logging on
Capercaillie (Tetrao urogallus) leks.

1995

1. Cutting experiments in southcentral
Norway. Scand. J. For. Res. 4: 99-
109.

Rolstad J. & Wegge, P. 1987a. Habitat char
acteristics of Capercaillie Tetrao
urogallus display grounds in south
eastern Norway. Holarctic Ecol. 10:
219-229.

Rolstad J. & Wegge, P. 1987b. Distribution
and size of Capercaillie leks in relation
to oldforest fragmentation. Oecologia
72: 389-394.

Rolstad, J. & Wegge, P. 1989a. Effects of
logging on Capercaillie (Tetrao
urogallus) leks. Il. Cutting experi
ments in southeastern Norway.
Scand. J. For. Res. 4: 111-127.

Rolstad, J. & Wegge, P. 1989b. Effects of
logging on Capercaillie (Tetrao
urogallus) leks. Ill. Extinction and
recolonisation of lek populations in
relation to clearfelling and fragmenta

Lek habitats of Capercaillie at Abernethy Forest 19

tion of old forest. Scand. J. For. Res.
4: 129-135.

Rolstad, J. & Wegge, P. 1989c. Capercaillie
habitat: a critical assessment of the
role of old forest. Proc. intern. Symp.
Grouse 4: 235-250.

Spitzer, G. 1985. Population structure and
distribution in a retrograde Capercaillie
(Tetrao urogallus) population -
subpopulaiion of displaying cocks.
Proc. intern. Symp. Grouse 3: 354-
362.

Steven, H.M. & Carlisle, A. 1959. The Native
Pinewoods of Scotland. Oliver & Boyd,
Edinburgh.

Viht, E. 1991. Size and forest structure of

display grounds of Capercaillie Tetrao

urogallus pleskei at Alutaguse, north-east

Estonia in 1979-89. Ornis Scand. 22: 294.

Wegge, P. & Rolstad, J. 1986. Size and

spacing of Capercaillie leks in relation to

social behaviour and habitat. Behav. Ecol.

Sociobiol. 19: 401-408.

R. W. Summers & R. Proctor, RSPB, Etive House,

Beechwood Park, Inverness

S.l.H. Shah, D.F.O. Wildlife, Kakul Road, Abbottabad, Pakistan

Revised typrescript received 16 February 1995

20 Scottish Birds 18: 20-23

SB 18(1)

Nest reliefs and feeding rates of Merlins

R. C. DICKSON

This paper summarises observations on nest reliefs and feeding rate of
Merlins in Galloway at three and six nests respectively between 1971-92. It
shows that males did about a third of the daytime incubation. Food items
brought were found to be less frequent in the pre-laying and incubation
periods and greater when the young hatched and fledged; there was no
difference in feeding rates between different periods of the day. Males
provided most of the food throughout the greater part of the breeding cycle.

Introduction

During long-term studies of Merlins Falco
columbariusin Galloway, the number of times
males relieved incubating females were
recorded at three ground nests in 1972-77.
Between 1971-92, their feeding rates at six
nests were also recorded from the pre-laying
stage through to post-fledging and the results
pooled from all nests (Table 1). One nest site
was situated on a cliff face with newly planted
conifer plantations below and open hill above;
two were tree nests, 38kms apart, both at the
edge of conifer plantations surrounded by
open moorland; and three were at ground
nests, 3-5kms apart, onheather slopes above
the edge of conifer plantations. The breeding
cycle was categorised into four stages: pre-
laying; brooding; nestling and post-fledging.

Nest relief and behaviour

Both sexes incubated. In 46 hours
observations from hidden positions at some
distance, males (only blue-backed males were
present at all nests) relieved females on 14
occasions (0.30 reliefs/hour). Trimble (1975)
and Green (1977) recorded males spending
up to two hours incubating but the longest

relief in Galloway was 1.8 hours and the
shortest 8+ min. (mean 40 min.). Five (86%)
of the reliefs were after males brought prey to
the sitting females and she fed while he
incubated. In the other nine instances when
reliefs were watched, females either preened,
or once chased an unidentified passerine in
the breeding area, once recovered cached
prey, and once attacked two Carrion Crows
Corvus corone until the male left incubation
duties and took over the attack; the female
immediately returned to the nest in a
spectacular stoop.

On relieving incubating males, females
soared, tailand wings spread, above nests on
three occasions before dropping vertically in
spectacular stoops to the nests. Conversely,
males were only once seen to soar on being
relieved; otherwise they either preened or
flew away directly. On all occasions, however,
nest relief was preceded by both sexes calling
beforehand, usually from a perch.

In two all-night watches at the same nest site
between 2030-0800hrs and 1915-0900hrs
on31 May/1 June 1972 and 1973 respectively,
females incubated throughout, commencing
at 2052 and 2155; the first and only relief by

1995 Nest relief and feeding rates of Merlin 21

Table 1. Feeding rates of Merlins in relation to the breeding cycle in Galloway, 1971-

92.
Period No.food No.hrs No.prey
items observ.Del./hr. Morning (hrs)
brought No. No.
Pre-laying 12 47 0.26 7 (22)
Brooding 11 71 0.15 2 (23)
Nesitling 37 69 0.54 é (23)
Post fledge 39 79 0.49 7 (19)

Notes : (a) x2 = 0.56, 1 d.f.,N.S. (b) x2 =0.94, 2d.

2 d.f., N.S.

amale was at0702 on 1 June 1973. However,
Rowan (1921-22) found that a male incubated
throughout one all-night observation which
he stated could merely have been an individual
idiosyncrasy.

Feeding rates

The total number of prey brought overall was
99 items during 266 hrs observations (0.37
deliveries/hour). However, the number of
food items brought was nearly three times
less in the pre-laying and brooding stages
(0.19 deliveries/nour) than in the nestling and
post-fledging stages (0.51 deliveries/hour).
The total number of deliveries recorded overall
in the pre-laying stage was 12 deliveries in 47
hours observations (0.26 deliveries/hour).
Females largely stopped hunting in the pre-
laying stage and became less active towards
the egg-laying stage, possibly to avoid the
risks involved in hunting then (‘egg-laying

Periods of day (deliveries)
Afternoon (hrs) Evening (hrs)

No. No. No. No.
4 (20.5) 1 (4.5) (a)
4 Zi} Pes (27) (b)
18 (25) 12 (21) (c)

11 (29) 21 (31) (d)

f.,N.S. (c) x2-3.88, 2d-f.,N.S. (d) x2 =3.50,

lethargy’ Newton 1979), and relied more on
males to provide for them. Interestingly,
though, prey deliveries were greater in the
pre-laying (0.37 deliveries/hour) than in the
egg-laying stage (0.20 deliveries/hour). This
could be related to the requirements of females
to put on body fat prior to the commencement
of laying and/or that it may be essential to
successful pair-bonding in that females could
choose good male providers. Inthe brooding
stage, when females were inactive, prey
deliveries were reduced by nearly a half (0.15
deliveries/nour). The number of prey
deliveries subsequently increased threefold
inthe nestling (0.54 deliveries/hour) and post-
fledging (0.49 deliveries/hour) periods when
the young probably needed large amounts of
energy for growth and development (Table1).
Thus there is a significant difference in feeding
rates between stages with the highest
occurring in the nestling and post-fledging
periods (Chi-squared = 18.82, 3d.f., P<0.01).

22 R.C.Dickson

Males alone did most of the hunting and
provisioning of food until late in the nestling
period when they were eventually helped by
females. Then the number of prey deliveries
increased three times as much as males
alone provided during the brooding period.
This higher rate continued until the young left
the breeding area, but males still contributed
more deliveries than females in the ratio of
lee

Some of the older publications state that
Merlins are fed at certain times of the day
(Selous 1913, Taylor 1914). In this study,
however, there were no significant differences
in feeding rates in any of the separate stages
of breeding cycle between periods of the day
(Table 1).

SB 18(1)

Discussion

By noting which sex was flushed from nests in
the wild during routine inspections in
Northumbria and Newfoundland, Newton et
al. (1978, 1986) and Temple (1972) concluded
that male Merlins took a share of incubation.
The findings in this paper agree with previous
conclusions that males did abouta third of the
daytime incubation. Rowan heard no signals
between sexes at changeover in Yorkshire
and he considered that males left when so
inclined and his mate then took over. No
regular rituals preceded nest reliefin Galloway
either but both birds called before relief,
probably indicating that they were ready to
relieve and be relieved. Males brought prey

Table 2. Comparison of food deliveries at Merlin’s nests in Galloway, Norway,

Europe and Iceland.

Period

Galloway
Pre-laying 0.26
Brooding 0.15
Nestling 0.54
Post fledging 0.49
Notes:

a) Data from Sperber & Sperber1963

Nos of prey deliveries/hour

Norway Europe Iceland
(a) (b) (Cc)
0.28 - 0.14
0.38 0.64 -
0.46 - -

b) Data from Chislett 1933; Macintyre 1947; Hard & Enemar 1980.

C) Data from Selous 1913.

1995

on only 36% of reliefs suggesting that
commencement of incubation duties by males
was not triggered by the presentation of food.

There are few studies of feeding rates of
Merlins throughout the breeding season apart
from the studies by Sperber & Sperber (1963)
in Norway and several for the nestling period
(Chislett 1933; Macintyre 1947; Hard &
Enemar 1980). Sperber & Sperber provides
the only direct comparison and then feeding
rates are fairly similar (Table 2). Other
references to prey deliveries range from 0.36
- 0.82 deliveries/hour with a higher average
of 0.64 in the nestling period than recorded
either by Sperber & Sperber or by this study
(Table 2). Pertinently, Selous’s (1913)
recordings during the brooding stage of 0.14
deliveries/hour is identical to the findings in
the study (Table 2).

Acknowledgements

| thank the Forestry Commission (now Forest
Enterprise), South Scotland for permission to
enter their land. Dr Pat Monaghan of the
Applied Ornithology Unit at Glasgow
University kindly carried out the statistical
analyses.

References

Chislett, R. 1933. Northward Ho! for Birds.
London

Nest relief and feeding rates of Merlin 23

Green, D. 1977. Moorland Falcons. Wildlife
19 314-319

Hard, |.& Enemar, A. 1980. Stenfalken Fa/co
columbarius bytesval och matnig un
der ungarnas botid. Var Fagelvarid
39: 25-34.

Macintyre, D. 1947. Highland Naturalist
Again. Seeley, Service & Co. Ltd.,
London.

Newton, |. 1979. Population Ecology of
Raptors. Poyser, Berkhamsted.

Newton, I., Meek, E. & Little, B. 1978. Breed
ing ecology of the Merlin in Northum
berland. Brit. Birds 71: 376-398.

Newton, |., Meek, E. & Little, B. 1986. Popu
lation and breeding of Northumbrian
Merlins. Brit. Birds 79: 155-170.

Rowan, W. 1921-22. Observations of the
breeding habits of the Merlin. Brit.
Birds 15: 194-202, 222-231.

Selous, E. 1913. A diary of ornithological
observations made in Iceland. Zoolo
gist 17: 57-66; 169-174, 301-313.

Sperber, |. & Sperber, C. 1963. Notes on the
food consumption of Merlins.
Zoologika Bidrag fran Uppsala 35:
263-268

Taylor, E.W. 1914. The habits of the Merlin.
Naturalist 1914: 115-120.

Temple, S.A. 1972. Sex and age character
istics of North American Merlins. Bird
Banding 43: 191-196.

Trimble, S.A. 1975. Habitat management
series for unique or endangered spe
cies: Merlin. U.S. Department of Inte
rior Report No. 15.

R.C. Dickson, Lismore, New Luce, Newton Stewart DG8 OAJ.

Revised typescript received 22 December 1994.

24 Scoitish Birds 18: 24-50

SB 18(1)

The Status and distribution of wintering Pink-footed and Greylag

Geese in east central Scotland

M.V. BELL & S.F. NEWTON

The numbers of Pink-footed and Greylag geese wintering in Perth &
Kinross District of Tayside Region and in Central Region were monitored
through the winters of 1986/87 to 1993/94 by counts at roosts and of birds
on the feeding areas. Pinkfeet were present from late September to early
May with peak numbers in autumn when c.65,000 were usually present in
early to mid-October with 86,868 in mid-October 1991 the largest number
recorded. Numbers fell sharply thereafter with 20,000-25,000 usually present
from December to February with a small increase in March and April as
birds returned north. Migrant Greylag were present from mid-October to
mid-April and numbers were relatively stable with 9000-14,000 present
throughout the winter. This represented a large decrease from 10-30 years
earlier when this was the major wintering area in Britain. Pinkfeet regularly
roosted at 21 sites and were found feeding in 466km? while the more
dispersed population of Greylag roosted on at least 55 waters and fed in
491km?. In total at least 60 sites were used as roosts and 751km’ for
feeding. The change in status of both species is discussed in relation to
recent changes in agricultural cropping regime and shooting practice.

Introduction

In the first decade of the Wildfowl Trust’s
November census of wintering Pink-footed
Anser brachyrhynchus and Greylag Geese
A. anserin Britain, initiated in 1960, Perth and
Kinross was the most important area for both
species, with major roosts of Pinkfeet at Loch
Leven, Dupplin Loch and the Carsebreck
Lochs and of Greylag at the Blairgowrie Lochs,
Drummond Loch and the Carsebreck Lochs
(Boyd & Ogilvie 1969; 1972). The area totals
inearly November varied from 12,237 - 27,375
Pinkfeet and from 14,202 - 33,109 Greylag,
accounting for up to 35.9% and 57.0% of the
respective UK populations (Boyd & Ogilvie
1969; 1972). The general winter ecology and
feeding distribution of geese in this area was

described by Newton et al. (1973), while a
detailed study of feeding ecology was
undertaken at Loch Leven (Newton &
Campbell 1973). The importance of the area
was maintained through the 1970s with up to
54.3% of the Pinkfeet and 47.4% ofthe Greylag
populations found in November, but since the
early 1980s the area has become relatively
less importantfor geese (Thom 1986; Wildfowl
& Wetlands Trust annual reports on the grey
goose counts).

Most of the major goose roosts in Britain have
been protected by Site of Special Scientific
Interest (SSSI) designation for a number of
years and more recently many have also
been given Special Protection Area (SPA)

1995

and Ramsar Site status. Many of these
designations were based on the annual
November counts which give a rather crude
measure of site importance but are often the
only data available. In many areas little is
known about the number of geese using
particular roosts throughout the winter, or
about the use geese make of the farmland
round each roost for feeding, with the
exception of the early study by Newton &
Campbell (1973) at Loch Leven and more
recent studies in Lancashire (Forshaw 1983)
and north-east Scotland (Bell ef a/. 1988; Bell
1988). Changes in agricultural support have
resulted in large changes in cropping patterns
in many areas used by wintering geese over
the last ten to twenty years. In addition there
is now much more disturbance of geese on
the feeding areas by farmers wishing to protect
autumn sown crops and also by organised
shooting parties, especially in the Kinross
Basin, Strathmore and Strathearn. During the
course of this study several areas which
regularly held large numbers of geese in the
late 1980s were virtually unoccupied a few
winters later. These changes make it vital to
understand the feeding behaviour of geese at
each roost.

The Central Scotland Goose Group was
formed in the autumn of 1988 to bring together
allthe local goose enthusiasts and co-ordinate
their efforts to this end. This paper is the result
of their labours and describes the present
status and feeding areas used by both species
throughout the winter in the old counties of
Perthshire, Kinross-shire, Stirlingshire and
Clackmannanshire, which now form Perth
and Kinross District of Tayside Region and
Central Region (Fig. 1). A more detailed
account of the feeding ecology of geese in
parts of this area will be presented elsewhere.

Satus of wintering Pink-footed & Greylag Geese in central Scotland 25

Methods

Geese were usually counted leaving the roosts
at dawn by experienced observers familiar
with the behaviour of the birds at their particular
sites. At some roosts it was necessary to
deploy several counters because of the size
and topography of the site. It was possible to
obtain satisfactory counts of Pinkfeet at some
sites on the dusk flight shortly after they
arrived in autumn and again in late April just
before they departed. In some areas it was
possible only to count geese on the feeding
areas by systematic searches from roads.
The date, time, farm name, 6 figure grid
reference and food crop of all feeding flocks
were recorded.

Allthe main roosts were counted in November,
and more recently also in October, for the
National Goose Count organised by the
Wildfowl & Wetlands Trust. In addition to
these national co-ordinated counts, counts
were undertaken at most sites throughout the
winter from the autumn of 1988 onwards. At
one of the main roosts almost weekly counts
were obtained as part of a detailed study on
the feeding ecology of the two species.

Data are presented in several ways depending
onthe coverage achieved at the various sites.
For most roosts only one count was obtained
in each month but at those where more than
one count was available, the largest has been
used. This includes all the data for the
Carsebreck Lochs/Strathallan, early autumn
counts at Dupplin Loch, and some months at
Kercock and in Strathearn for Greylag. The
three largest monthly counts were averaged
to give the ‘regular’ count as defined for ducks
(Atkinson-Willes 1963). The average count
for each month gives a monthly index of site
usage.

26 Bell & Newton. SB 18(1)

Figure 1. Map of the study area. Major rivers and 10km square grid lines are shown.

10 20 30
70

2 A

Almond

estuary

YW

L.Leven

|
!
)
|

1995

Results & Discussion

Pink-footed Geese regularly roosted at 21
sites and were recorded feeding in 466km*°.
They were mainly confined to the lower-lying
straths with large flocks in the Forth valley,
Strathallan, Strathearn and the Kinross basin
(Fig. 2). In contrast, Greylag Geese were
much more dispersed, roosting ona minimum
of 55 sites and feeding in 491km? of which 284
were not occupied by Pinkfeet. As well as
occurring in the main valleys they were found
also far into the hills, most notably in the upper
Tay/Tummel catchment (Fig. 3). In total
751km? were used by feeding geese.
However, many of these squares contained
only a few suitable fields so the true foraging
area was very much less than 751km*.

Both species flighted in much better light at
dawn than at dusk throughout the winter. At
the Carsebreck Lochs where both species
were regularly observed, Pinkfeet tended to
flight a few minutes before Greylag at dawn.
As the autumn progressed departures became
earlier with the main flight usually 25-30 min.
before sunrise in November and 45 min.
before sunrise in December and January.
With increasing day length they departed
slightly later, 25-40 min. before sunrise, for
the remainder of the winter. Cloud cover
made little difference to departure times,
consequently the departures on gloomy
dawns in mid-winter could not always be
seen. Most very early departures (up to 76
min. before sunrise) occurred on fine frosty
mornings. In early autumn Pinkfeet returned
30-60 min. after sunset and by late autumn
45-75 min. after sunset was typical. People
living near roosts often reported geese flying
at night but the extent to which night feeding
occurred was difficult to determine. One of us
lived under the flightline of Pinkfeet commuting
between roosts in Strathallan and the Forth
valley and geese often returned well after

Satus of wintering Pink-footed & Greylag Geese in central Scotland 27

dark (up to 7hrs 20min. after sunset) on
completely overcast nights. We have no
reason to believe this behaviour was unique
to this roost. Geese were disoriented during
fog and were often unable to locate the roosts
if visibility was less than c.400m, circling the
street lights of local towns and villages all
night, behaviour previously noted in north-east
Scotland (Bell et a/. 1988).

The following account details the present and
past status of each species by area.

1.FORTH VALLEY
a) Skinflats to Alloa

There was a well established roost of Pinkfeet
on the estuarine mud flats at Skinflats in the
1930s but this was deserted during the
1939-45 war because of disturbance from
aircraft. The roost was re-occupied in the
1950s with regular autumn numbers of 1100
and a peak of 1700, somewhat smaller
numbers than before (Atkinson-Willes 1963).
Several hundred Greylag also used the site at
this time. It was suggested that the roost was
deserted at some stage during the 1970s or
early 1980s (Owen etal. 1986) butitis unclear
whether this assertion was based on
systematic counts at the roost or on casual
observations. The mudflats east of
Grangemouth have not been used by roosting
geese for many years.

At present Skinflats holds c.2000 Pinkfeet in
autumn (Table 1) with generally lower
numbers later in the winter, though 4930 in
January 1992 was the highest total recorded
during this study (Table 1). Apart from the
October and November counts most of the
figures in Table 1 were based on field surveys.
After the shooting season most geese moved
up river and roosted in the Alloa/Tullibody

28 Bell & Newton. SB 18(1)

Figure 2. The distribution of feeding areas for Pink-footed Geese by one km square. The
Forth and Tay estuaries and rivers Forth, Allan, Earn, Tay and Tummel are shown also.

Figure 3. The distribution of feeding areas for Greylag Geese by one km square. The
Forth and Tay estuaries and rivers Forth, Allan, Earn, Tay and Tummel are shown also.

Ke)

Inch area where substantial flocks remained
to late April in recent years with 3500 in late
April 1991. The geese fed over the alluvial
plain from Skinflats village west towards
Stirling on the south side of the estuary and
also favoured the Kennetpans/Clackmannan
area on the north shore and were recorded in
27km? (Fig. 2). Much of the farmland was
intensively arable and geese were
unwelcome, there being much disturbance to
protect crops in addition to wildfowling along
the estuary.

In late October and early November 1990
large numbers of geese appeared at Gartmorn
Dam near Alloa with 1600 Greylag and 450
Pinkfeet present on 8 November. The Pinkfeet
quickly moved on but hundreds of Greylag
remained in the area into the New Year, also
roosting on the Forth at Cambus on some
nights. Large numbers often flighted east out
of Gartmorn towards Powmill but the exact
feeding areas were not located. Since the
1990/91 winter Pinkfeet have been very
occasional in small numbers and most flocks
of Greylag have been under 200 birds which
fed close to the roost. The neighbouring
Peppermill Dam (in Fife) held 110 Pinkfeet on
the mid-November count in 1993.

Small numbers of Greylag, usually under 200
birds, sometimes roosted on the estuary at
Skinflats or further upstream around Alloa
Inch. Six hundred were at the latter site on 31
January 1993. Greylag were recorded feeding
in 11km? along the estuary and in the
Clackmannan area (Fig. 3).

b) Lake of Menteith and the Carse of Stirling

Historically the Flanders Moss/Lake of
Menteith area was one of the few inland
haunts of Pinkfeet in Scotland with several
thousand present in the earlier decades of

Satus of wintering Pink-footed & Greylag Geese in central Scotland 29

this century (Berry 1939). Greylag first
appeared in the early 1900s and by the 1920s
numbered up to 2000 birds (Berry 1939).
Through the 1960s and 1970s the November
counts usually found 500-2500 Pinkfeet,
though none were located in several autumns,
and 500-1000 Greylag, with a peak of 2235 in
1975 (Boyd & Ogilvie 1969, 1972). Lake of
Menteith and Flanders Moss were the main
Pinkfoot roosts at that time; the Greylag
preferred Lake of Menteith, though up to 500
regularly used Loch Rusky in the early 1960s.
This site was then deserted following
disturbance from fishing and boating (Boyd &
Ogilvie 1972) but was occupied again in the
late 1970s.

The Carse of Stirling was a difficult area to
cover since geese from a number of roosts
fed there and too few observers were available
to cover all these roosts simultaneously.
Consequently many of the data were obtained
by field searches of the area west of the M9 to
the periphery of Flanders Moss in addition to
counts of geese roosting at Lake of Menteith.
Field searches were problematical in the areas
south-west of Lake of Menteith but the geese
rarely flighted in that direction. Visibility was
restricted by the topography and trees on the
undulating ground between the Carse and
the Teith Valley, but this area was more
important in autumn when the geese fed on
cereal stubbles and at this time most counts
were of birds at the roosts. Pinkfeet known to
have commuted into the Forth valley from
roosts in Strathallan were subtracted from the
field counts shown in Table 1, so that these
figures represent the numbers of geese
roosting in the Forth valley.

Pinkfeet arrived at Lake of Menteith in late
September and in recent autumns the peak
was between 1000 - 3300 birds in either
October or November (Table 1). The

SB 18(1)

30 Bell & Newton.

“SYJUOW

asay} jo Led Ajuo J0j Juaseid alam asaab se uoyejndod sejnBas ay) ejejnzjeo 0] pasn JOU a4aM sjunoo Aep puke Jaquiajdas

vy96 - O6r8 7609 60Lr LES vS6Z 9688 6SLOL - obeiane
0889 O6LL O0EZ 0zz9 O06E GZ6r OScr =-—dOS Zr OZLZ 000¢ 76/661
O6E6 OSsL O8zZit 0008 Ose 08s9 0688 0882 OreZ 006S €6/Z661
69/6 OOOOL OOvOlL S¥r68 0802 OvOZ 0Z06 0S96 Ovz6 OOLEL 726/LG66L
9688 OS9S O6L6 oess OLEE O00rZ O8zZ O¢E9 0266 OS6SL 16/0661
26901 OZ6 OSS6 00zr O89L OOLr Oveés O0€8 OOZzrl 0026 06/6861
Z9SS\ 0929 0088 Orgs 0618 0Sz9 OOS OOV9L OOOZI OOF 68/8861
606 002s 00SS 0Sz9 O€br OSsE 009r O9rlt OOrlL OO0ES 88/Z861
LLE9 0S8z 006S 0S99 00zS OZ8L - 0019 000S OOZE 28/9861
Jeinbas Kew idy JEN qe uer 92q AON 190 das
‘SIOJUIM P6/EGE6L - Z8/986L UBI[EYIe4S Ul asaa+y pajooj-yUld JO sjuNnod AjyjuowW Winwxey *~ JVGWL
‘abeiane jnjbuiueew e ajejnajeo oO} ajgeyjieake ejep JUBIOWJNSU! = -
0002 906 - v96 +69 8r8 O90! s9g_ - Buys jo esyeg ‘beAe15
O8zZ 69vE PSer OStG CONCewoeele GeZle tee = sleec Bulns Jo esued “JeepUld
LeS9 - - - - - - 6cSc_ «Ler yoreye|| YOO] ‘Jea}4Uld
O€6r LOZ 6ZSL OOLL vZ8 LE8L 299 EZ8L E€8El Asenys3 yo ‘}9e)4UId
xew Jein6be Jdy JEIN qo uer 09q AON OO

“SIOJUIM P6/E66L - 68/886L AaljeA yoy ay} ul siosas bulsn asaab jo sjunod Ajyjuow abeseay *| AIGVL

1995

November totals for 1991, 1992 and 1993 are
known to be low as geese roosting at Flanders
Moss and Loch Rusky were missed,
depressing the average for that month. Inthe
1993/94 winter an SNH/RSPB survey team
confirmed that Loch Rusky and the lochan on
Flanders Moss were more important roosis
for both Pinkfeet and Greylag than Lake of
Menteith (Thompson & Harding in prep.).
There were no records of either species
roosting at Loch Watston near Doune during
this study though over 1000 Pinkfeet were
found there on the November count in 1979
and 1980. Numbers of Pinkfeet feeding on
the Carse were usually low from late
November through December and January,
largely due to disturbance on the feeding
areas, but geese returned to the area in
February once the shooting season was over
(Table 1). The Forth Valley frequently
remained free of snow when sites further
north in Strathallan were snow covered and
geese displaced from there or from further
afield then boosted numbers as happened in
January and February 1990 when almost
5000 were present.

When the Pinkfeetfirst arrived in autumn they
fed close to Lake of Meniteith, with the farms
on the periphery of Flanders Moss being
particularly favoured. Later in the winter they
moved further east to the Blairdrummond and
Gargunnock areas. There were only two
records of Pinkfeet feeding in the Buchlyvie to
Balfron areas during this study. In March
1991 Pinkfoot feathers were found at Mill
Dam above Kippen. Most Pinkfeet feeding at
the eastern end of the Carse came from
roosts in Strathallan but birds also commuted
to the Thornhill areas at times, mixing with the
geese from the roosts further west. Up to
5900 Pinkfeet from roosts in Strathallan were
recorded feeding in the Forth Valley. Why
geese feeding near Thornhill should choose

Satus of wintering Pink-footed & Greylag Geese in central Scotland 31

to fly 21km to roost in Strathallan rather than
3-8km to the roosts west of Thornhill is unclear
but was presumeably related to disturbance
and roost security. On 2 April 1994, 8050
were found feeding on the Carse and 7250 of
them roosted locally, the largest number
recorded during this study. Pinkfeet were
recorded feeding in 80km? in the Carse of
Stirling.

The number of Greylag using the Carse of
Stirling fluctuated considerably over recent
winters. On occasion only a few hundred
were located, butin the 1991/92, 1992/93 and
1993/94 winters over 1000 were present for
long periods, particularly in the Gargunnock
and Drip areas. The areas west towards
Kippen and Thornhill used to be the main
Greylag feeding sites, with smaller numbers
north and west of Thornhill. The river Forth
meanders through the Carse several meters
below the level of the surrounding farmland
and Greylag often fed along the sloping grassy
banks where they were difficult to see. Some
low counts may be due to birds being missed
along the river. Up to 400 were noted in fields
by the Teith south of Callendar on occasions.
The feeding distribution of Greylag was slightly
more restricted than that of Pinkfeet, being
recorded in 43km?*

The roosting behaviour of Greylag was even
more complicated than that of Pinkfeet. In
autumn geese feeding at the east end of the
Carse came from Loch Coulter, some 8km to
the south in the Gargunnock hills, on several
occasions but this roost was not counted
regularly. None were recorded at North Third
reservoir which is closer to the Carse. Later in
the winter these birds probably roosted along
the river Forth. The Greylag further west may
also have used the river or flighted west to
roost at Loch Rusky or Lake of Menteith.
Greylag from Strathallan very rarely flighted

32 Bell & Newton.

into the Forth valley during the first four winters
of this study but more recently small flocks of
up to 50 birds have commuted to feed here. It
seems probable that Greylag roosting at Loch
Coulter may also flight east to feed on farmland
north of Stenhousemuir which is equi-distant
to the Carse of Stirling. A flock of up to 300
Greylag has recently appeared on the
Slammannan plateau south of Falkirk, sharing
the area with Scotland’s only flock of Bean
Geese A. fabilis (J. Simpson pers. comm.).

c) Loch Mahaick

Loch Mahaick on the Braes of Doune is
somewhat of an enigma. Pinkfeet were first
recorded there in autumn during the 1950s
(Atkinson-Willes 1963), but the site was not
checked regularly. C.1700 birds were present
in November 1963 - 1964 and from 1974-1983
between 0 -2500 were found on the November
count. In November 1988, 6530 were recorded
and in the following two autumns large
numbers were also present with rapid build
ups over a few weeks during October leading
to transient peaks and rapid falls through
November (Table 1). Few birds were found
later in the winter and the site did not appear
to be used in spring. Since then autumn
numbers have fallen back to a few hundred.
The geese flighted out between SSW and
SSE with small flocks very occasionally going
east into Strathallan. The main feeding areas
appeared to be on barley stubbles on the
undulating farmland south of the Teith, an
area where viewing is difficult. Greylag were
only noted at Loch Mahaick on three occasions
with 178 the most.

SB 18(1)

2. STRATHALLAN
a) Pink-footed Geese

Pink-footed Geese first appeared at the
Carsebreck Lochs in the 1930s, initially only
in autumn, but by the 1960s were present all
winter (Berry 1939; Atkinson-Willes 1963).
The number of Pinkfeet found on the
November count remained fairly steady at
between 1000-4000 through the 1960s, 1970s
and early 1980s with a peak count of 6270 in
1974 but showed a marked increase in the
late 1980s.

The first large arrival of Pinkfeet usually
occurred in the last week of September with
marked autumn peaks in 1987-1991 before
numbers fell away in mid-winter. In 1992 and
1993 autumn numbers were much lower than
in the previous five seasons (Table 2). There
was aspring peak in March or April with major
departures in the last week of April or first
week of May. Late spring departures were a
feature of this site; 10,000 were still present
on 3 May 1992. Autumn numbers reflected
the amount of grain shed before or at harvest
(Bell et a/. in prep.). In autumns with clean
harvests the peak occurred in the last week of
September (1990, 1991) or the first week of
October (1989), whereas following dirty
harvests (1987, 1988) the autumn peak
occurred later in October or in November and
large numbers remained for longer than in
years with clean harvests (Table 2). Changes
in cropping regime from spring barley to
oilseeds and long-term set-aside led to a
large drop in autumn numbers from 1991
onwards (Table 2).

Most Pinkfeet roosted atthe Carsebreck Lochs
but at times up to 3000 roosted at two other
sites in the valley, while upper Glendevon
reservoir 6km SE held large numbers (up to

Keith Brockle

us
3
=
<
S
he
=|
°
&
ov
len
2

Male Capercaill

This Turtle Dove Streptopelia turtur wintered at Lundin Links 1994/95.
Anne-Marie Smou

Hawfinch Coccothraustes coccothraustes at St Andrews, October 1994.
Mary Macintyre

The long-staying Grey-tailed Tattler Heteroscelus brevipes at Burghead, said to
have caused 90 arrests for speeding on the A9 when it was first reported in
November 1994. Mary Macintyre

Magpie’s nest — with a difference

The nest of a Magpie Pica pica was recently donated to Glasgow Museums. It had
been removed from about 15-20 feet up in a Cupressus tree, which was being
partially cut down in a garden in Rutherglen, Glasgow. The cup of the nest was like
a normal magpie’s nest, made from twigs, bound together with mud and lined with
fine roots. Underneath the nest, however, the magpies had constructed a platform
mainly using S-shaped wire wall ties from a nearby building site. Approximately
100 ties were used, together with the miscellaneous pieces of wire and even an old
aerial from a car! | always thought that magpies collected shiny objects and put
them in their nest. This is the first time | have heard of them actually making their
nest out of the objects which they have stolen.

Richard Sutcliffe, Curator, Science Department, Glasgow Museums.

Platform of wire wall ties for magpie’s nest from Rutherglen. | Glasgow Museums

1995

7360) in autumn 1989, 1990 & 1991. Several
temporary floodwater pools were used by
Pinkfeet also.

Pinkfeet from the roosts in Strathallan were
recorded feeding in 143 km*. The main feeding
areas were in Strathallan between Blackford
and Kinbuck up to the limit of cultivation, but
Pinkfeet often flighted NE into Strathearn and
SW into the Forth valley to feed between the
Carse of Lecropt and Blairdrummond Moss
(Fig. 2). At times they flighted as far as
Thornhill. Pinkfeet used several sites for
loafing. These offered a refuge close to feeding
areas from disturbance and were situated in
the direction of all the main flightlines at 300m
to 6.0km from the main roosts. Four were
fields in the valley while the other four were
fields just below the muir line. All were sites
with minimal disturbance and good visibility.

b) Greylag Geese

Greylag were first recorded at Carsebreck
around the turn of the century when the lochs
were first dammed and were numerous by the
1920s when 2000 were often present (Berry
1939; Atkinson-Willes 1963). When the
November counts of grey geese were initiated
Greylag were the slightly more numerous
here with 2500-6000 usually present and a
peak count of 8280 in November 1968.
Numbers declined sharply in the mid-1970s
to c.1500, with a brief recovery to 2000-4000
in the early 1980s, but these numbers have
not been maintained and 500-1000 is now
normalin November. Thenumbers of Greylag
showed a different pattern to those of Pinkfeet
and were generally less predictable. They
arrived in mid-October and numbers increased
steadily to peak in mid-November in two
autumns (1988 and 1989) or plateaued from
late November through December (1990 and
1991) (Table 3). In autumn 1992 there was an

Satus of wintering Pink-footed & Greylag Geese in central Scotland 33

early peak in late October within a week of the
birds arriving, while in the 1986/87 and 1990/
91 winters the peak numbers were present in
February (Table 3). Greylag departed in early
to mid-April showing no spring passage. In
autumn they were much less dependent on
cereal stubbles than Pinkfeet, feeding on
grass most of the winter (Bell et a/. in prep.).

Greylag roosted at the main lochs and on an
adjacent pool. After the shooting season these
roosts were largely abandoned in favour of
the river downstream from Greenloaning and
several floodwater pools. Where the two
species shared roosts they tended to roost
apart and flighted separately at dawn unless
disturbed by people.

Greylag had a much more restricted feeding
range than Pinkfeet and were recorded in 66
km? (Fig. 3). They favoured fields in the valley
floor and on the northern slopes of the valley,
rarely feeding on the Ochils side of the valley
or flighting into Strathearn or the Forth Valley.
By the end of this study, following changes in
the farming regime and increased disturbance
from shooting they were confined to just a few
farms. Greylag did not use specific loafing
areas, resting in the fields where they fed.
When Greylag and Pinkfeet fed in the same
field they tended to feed in discrete flocks.
However, inthe early years of this study when
Greylag were more numerous, the two species
sometimes fed in large randomly mixed flocks
and flighted together at dusk. This very unusual
behaviour was often noted here in the 1960s
and 1970s (M.A. Ogilvie pers comm.) and
was very occasionally noted in the Kinkell
Bridge area of Strathearn and in the Forth
Valley during this study.

SB 18(1)

34 Bell & Newton.

‘YJUOL BY] JO
ped Joj Juasaid AjuO a4am spiig souls uoKejndod sejnBas ay} ajejnzjed 0} pasn jou a4aM sjuNnod JEequia}dag pajunod JOU OU

808Z1 v69S vcS9 8L8L Sc6S GSIZ S6OLL GZILGE = obeiane
S886l ou GLEZ O0CV 089S ou Ovscl OOO6E ou v6/E66L
OS8SlL GL8E OSE9 0009 ou OO8V OOZSL 00SS¢ 0000¢°9 €6/c661L
Geoace SZOZ ou OOES 00¢6 ou OOZOL OO0687 OOSZS c6/L661L
ZS6Z1 OOES OOES Occ S9EL O09 OZVS 000d 000Z¢c'9 16/0661
ZOZS1L 0062 0¢c8 O9L9 OSZS OS8Z OSIZ OOOLE (OOZL) 06/6861
OOLEC O8EV OOZ9 0089 0028 OOSTL OO8rl OO00r OOO06E 68/8861
€6SSl ou 0889 069S O8E9 0802 OOLLL O0O000€'9 0006¢ 88/Z861L
Cevrl ou OOVY OSl¢ OO OOEZ OOOL | 000S¢°9 ou Z8/9861L
Jejnbo idy Je qe uer 9eq AON 190 das

‘SIOIUIM H6/E66L - 28/9861 ‘UseaYjeS Ul aSa—ab pajooj-yUuld Jo sjunod Ajyjuow wWnwIxey “y ATAWL

cvLl €c9 €v6 Lc9OL 97cl O8rl S8Sl 9cOl ebeiene
662 LL9 S6V 029 OOZ OZ6 9cl Coo v6/E66L
LEG O8€ OLZ 08S SEs Odcl SOLL Gc9l €6/c661
OvEL OSS Z801 S6EL S6cl OOEL GcEl S8e c6/1L 661
S891 O8€ S6cl OdS¢c OL6 Ovcl S16 08S L6/O66 1
Z691L SVIL Oclt O901 OcOl OSI OSl¢ OLVL 06/6861
SO00¢c O801 S8cl Ov6 Ov6 O8rl OLVe G90¢ 68/8861
€v0c Ze SS6 O67 lL OSS OOl¢e O8ve 088 88/Z861
Lc0€ : 009 O9EP Ocle 0002 OSV IL Ov6 28/9861
Jein6es idy Je qe uer 99eq AON 190

"SJOIUIM H6/EG6L - Z8/986L ‘UeI/eYIesS UI asaay bejAas5 Jo sjunoo Ajyjuow wnuixeyy *¢ F1VAVL

1995

3. STRATHEARN
a) Pink-footed Geese

Pink-footed Geese first appeared at Dupplin
Loch in the 1930s as a result of disturbance
from shooting on the Tay estuary to the east
(Berry 1939). Dupplin was the mostimportant
roost in Britain in the two decades following
the initiation of the Wildfowl Trust’s autumn
census of grey geese in 1960 (Boyd & Ogilvie
1969) with a peak of 27,500 in November
1973 comprising 33% of the population.
Newton et al. (1973) commented on the
unusual nature of Dupplin Loch as a Pinkfoot
roost: relatively small (c.30ha) with mature
mixed woodland to the water’s edge. The site
is only tenable because of the absolute
protection from disturbance given by the
owners since the roost was established over
60 years ago.

The November counts averaged over long
periods have remained relatively constant at
Dupplin over the last 34 years with means of
10,258 in 1961-69, 12,011 in 1970-79 and
8781 in 1980-89, with the average since
recovering to over 10,000 again. These figures
hide some considerable fluctuations with only
just over 3000 present in some years, but
more typically well over 10,000. More recently
mid-November numbers have fluctuated
between 5000 - 15,700. The largest numbers
occurred here some 4-6 weeks earlier as the
geese arrived in Britain for the winter. Fora
few weeks Dupplin held huge flocks (Table 4)
with peaks of over 30,000 in late September
or early October of six of the last eight autumns
and 57,500 on 28 September 1991. During
the arrival period in most years there is a
constant coming and going of geese over the
Perth area in all directions. It would be
interesting to know just how many birds pass
through Dupplin each autumn; in some years

Satus of wintering Pink-footed & Greylag Geese in central Scotland 35

the initial arrival of Pinkfeet in Aberdeenshire
was from the south (Bell pers. obs.). twas the
realisation that huge numbers of Pinkfeet
were concentrated at just a few sites in
Scotland in early to mid-October which
prompted the suggestion that this would be a
better time for censusing this population
(Newton et a/. 1990).

There was insufficient food available to sustain
such huge flocks for long and in autumns with
clean harvests all the stubbles within c.10km
of the roost were cleared within 1-2 weeks
and numbers fell very rapidly, e.g. from 42,000
to 8200 in 14 days from 7 October in 1990 and
from 39,000 to 10,700 in eight days from 9
October in 1993. Itis not known if Dupplin was
always such an important arrival site for
Pinkfeet since regular counts were not made
in the early autumn period. The change from
spring sown barley to winter wheat and barley
and oilseed rape means that the area is now
unable to support large flocks later into
November. It seems probable that in those
autumns when the site held over 10,000 birds
onthe November count that more would have
been present a month or so earlier as is still
the case. In only one recent winter, 1988/89,
was a five figure count maintained into
December following the very dirty harvest
that autumn. Typically 4000-8000 remained
from December onwards (Table 4) unless
snow moved the birds on. Numbers increased
slightly in March and early April but the area
was not important for staging and most birds
departed by the end of April.

Large flocks of Pinkfeet sometimes roosted
on floodwater, particularly after the shooting
season. The regular flood on the Pow at
Gorthy was particularly favoured and held up
to 4000 birds while floodwater by the Earn
between Dalreoch and Aberdalgie was also
used with up to 4300 birds noted. Numbers

36 Bell & Newton.

using Dupplin fell to 1500 at such times. The
totals in Table 4 includes geese roosting on
floodwater.

Most Pinkfeet roosting at Dupplin fed within
5km of the roost but some flighted out at least
11km at times. They were found feeding in
142km? (Fig. 2). The area to the north and
north-east round Tibbermore was particularly
favoured. Rather few birds flighted ESE down
the Earn; geese feeding at the mouth of the
Earn were believed to have roosted on the
Tay estuary but may have come from Loch
Leven. Small numbers regularly flignted SSE
out of Strathearn towards Glenfarg. Feeding
flocks were difficult to locate in this area
because of the topography. It is possible that
such flocks could have been leaving the area
for Loch Leven or further afield. The fields in
Strathearn between Forteviot-Dunning and
Aberuthven were frequently disturbed by
shooting parties and were little used during
the shooting season, especially in the last few
years of this study. To the SW Pinkfeet foraged
as far as the Kinkell Bridge area. There was
considerable overlap with Greylag from
Drummond Loch and occasionally with
Pinkfeet from Carsebreck here; on one day in
December 1988 geese from all three roosts
fed in the same potato field here. To the NW
Pinkfeet foraged as far as Abercairney and
also moved out of the valley to feed in the
Buchanty area in late winter and spring.
Visibility was restricted north of Methven by
undulating ground and trees and it was
probable that this area was used more
frequently than records indicated. It was also
thought that birds sometimes flighted out NE
over Perth to feed on the north slope of the
Sidlaws but overlaps with geese from the Tay
estuary made interpretation of the few feeding
flocks noted in this area difficult. Most of the
longer feeding flights occurred in late autumn
and early winter.

SB 18(1)

Pinkfeet used several areas for loafing andas
refuges from disturbance onthe feeding areas.
Cleavage Muir to the south, the Pow flood
and the Fornaught-Buchanty area were
particularly favoured but a number of farms
closer to the roost where the geese were
generally tolerated were also used. All these
sites offered good visibility of approaching
danger and some wereremote from roads
and normal human access. Methven Moss,
which was used many years ago (Newton et
al. 1973), was too overgrown to hold geese.

Pinkfeet were first recorded in numbers at
Drummond Loch in December 1988 when
2900 were present on the 18th, with over
1000 remaining into the first week of January.
The flock was feeding with Greylag in an
unharvested field of barley south of Muthill
and were almost certainly from Carsebreck
where the total fell coincidentally by a similar
number. None were recorded at Drummond
the following winter but large numbers were
again present in October 1990 with 3550 on
21st; none remained by mid-November. This
pattern has been repeated each autumn since,
with birds appearing in early October and
staying for just a few weeks. The peak counts
were 3400 in October 1991, c.3000 in 1992
and c.2500 in 1993. Their appearance has
coincided with lower numbers in Strathallan
and it seems probable that Pinkfeet displaced
by changing agricultural practice are now
feeding on cereal stubbles in the Crieff area
and roosting at Drummond. By clearing the
local stubbles of spilt grain before the Greylag
arrive the Pinkfeet may have contributed to
the drop in numbers of Greylag at Drummond
in recent autumns. However, Greylag were
less dependent on cereal stubbles than
Pinkfeet (Bell et a/. in prep.).

1995

b) Greylag

During the first decade of the November
counts, Drummond Loch was the single most
important roost for Greylag in Britain (Boyd &
Ogilvie 1972) with an average count of 5897
and a maximum of 10,242 in 1961-69. This
position was maintained through the 1970s
but there was a sharp fall from 1981 onwards
with 3500-5000 becoming the norm with a
peak of 7500 and a low of 1180 since 1981.
Some doubt has been caston the accuracy of
the large counts recorded in the late 1970s
(Thom 1986) but even with this proviso there
has undoubtedly been a substantial fall in
numbers over the last 10-15 years. In the
1960s Greylag also roosted regularly at the
neighbouring Loch Balloch and on three sites
by the Earn between Kinkell Bridge and
Innerdunning (Newton et al. 1973). Pitcairnie
Loch at Dupplin was also used by 500-1500
Greylag in November up to a few years ago
with a peak of 3990 in 1972. They are now
scarce there.

Greylag arrived in the second half of October
and in recent winters numbers peaked in late
October or early November (Table 5).
Numbers then dropped slightly through
December and January before increasing in
late February and March as birds moved
through on passage. Overall numbers tended
to be very constant through the winter (Table
5). There was an influx in December 1988,
thought to be associated with snow further
north, which gave the highest total recorded
during this study of 7200.

When the Greylag first arrived, they roosted
at Drummond but as the winter progressed
the flock dispersed onto many smaller waters
in Strathearn. In some winters this dispersal
occurred just before the mid-November count
and this caused considerable logistical

Satus of wintering Pink-footed & Greylag Geese in central Scotland 37

difficulties in obtaining a coordinated count in
November. Up to 3000 geese may have been
missed in November 1987, 1989 and 1991
before the full extent of this problem was
appreciated. Apart from Drummond Loch,
Greylag feeding in Strathearn roosted at Loch
Balloch, Mains of Drummond pond, Loch
Cowdens, the ponds at Abercairney House,
Loch Meallbrodden, East Fordun Dam and
on at least six semi-permanent floodwater
pools in the Kinkell Bridge area. Although the
smallest of these pools was only c.0.5ha, up
to 800 geese were accommodated at such
sites. Unless these pools were frozen, which
occurred very infrequently during recent mild
winters, a complete count of Greylag in
Strathearn necessitated either a complete
search of fields for feeding flocks or a
combination of roost and field counts, either
option taking about five hours. This was only
possible in reasonable weather.

Greylag were recorded feeding in 134km? in
Strathearn overthe 1986/87 - 1993/94 winters.
The most important feeding sites during the
first part of this study were in the Kinkell
Bridge area and the majority of birds fed here
from December onwards. However, in the
last three winters regular disturbance from
organised shooting parties has made this
area much less attractive and at times it has
been deserted. Shooting probably accounted
for the low numbers found in Strathearn in
November 1992 and 1993. Feeding areas
further downstream have largely been
abandoned. In the 1986/87 - 1987/88 winters
over 1000 Greylag fed regularly in the
Innerdunning to Dalreoch area but it is now
very unusual to find Greylag feeding south of
the A9 there. The local keepers believe regular
shooting to be the cause. Greylag feeding in
this area probably roosted at Pitcairnie Loch
or on the semi-permanent pools by the river,
some of which have now been drained.

‘ebelane ue ajejnojeo oO} ayep jUsIOWNSUI - - “AJUO S]UNOD PayeUIPJOOI JO SI SYIOT EUMOBbUIE|g 4O{ /2}O |

6092 ZcZl Ecol = Zpel = seer Lesh vBLzZ GrZl L6Z jauuuun | /Ae | seddr ‘Behe
GlvZ 6OLY ou L68 Agta Gcgoz 8lZZ €60r €G8e Syo07] auMmoburelg |e}0} BejAes5
Ovoe 86 - SLI 6 vS G6z CEE LEEL SeMo7 JO 7 ‘He|Aeu5
O00r 8ZZ 66S 9S2 Z8S QGP 926 L28 ZOl alun|D 7 ‘Behar
OSZL 669 cGy «SL ELE ZSL Zc9 G99 LSP 7 sapey\ ‘Hefei
000d - 5 ; . F 0S8 5 "| JUOWWOIS ‘He|Aaid
961LE ZEQL 8l9 9g¢ v6El 696 | LO0@z LyLL 869 yoooey ‘Be|Aes5
OZse SZEL g98l +199 g vel €6EL 929 GGL YOODEY ‘JEO/4Uld
oses 1Zze v8L decl vZ6l gS9 Grv97 €06E 8LSr UOIINI 9B UOYEGAy|N | 7 ‘Yapfuld
O€Zl 868 oss =e: ELS GOOL SLOL OLZ LZ8 cL 9u0dS ‘BejAaJ5
xe Jejn6bes Idy/ Je\\\ ge uer 09d AON po

‘SIQIUIM P6/E66L - 68/961 ‘AagieA ACL ay} Ul S}SOO4 Je aseeD bejAas5 pue pajooj-yUld jo sjunod Ajyjuow abesaay “9 3A L

‘WIEBUYIENS Ul SJSOOL JE BPNIOUI Sje}O} EYL

O9vr 88S¢ O9VE €Z6c OvZce CSEV 88Cr ZLSE obeiane
SLZE ou OcZE 00) 74 GEE ou OOEL O00r v6/E66 1
8807 VILLE OOSE G86c vSle Ou Ov0d 0SOS €6/c66 1
OELS OcSv G99E OOEr ou O09 GODS GES c6/L661
€89E GLoCo Ore GEES OLS¢2 Ovs¢ OCLV OLSE L6/0661
880r OZ61L OvOS O8SZc O8¢e OOVE GcC8e 00S¢9 06/6861
LE?S 0082 O6CE OLY O00 00¢Z O8Er 086¢c 68/886 L
O8Srv G8S 0262 OSZ¢2 O6SE Ov6éce O06S OScr 88/Z861
ESLS OS¢cc G40, = O0E?Z OSOE OVLS OZcZL OOOl 28/9861
Jein6ba idy JEW qe uer 99q AON 0

‘SIQJUIM P6/E66L - Z8/986L ‘UJeayjeS ul asaady be/Aas5 Jo sjunos Ajyjuow winuyxeyy “G ATV L

38 Bell & Newton.

1995

Smaller numbers of Greylag fed to the
south-east, south and south-west of
Drummond. Fields to the north-west between
Crieff and Comrie were little used during the
early years of this study but were more
favoured inthe last three winters when several
hundred birds regularly fed here. In late
October 1992, c.3500 fed on the Carse of
Lennoch, the largest number for some years.
Loch Balloch was frequently used by geese
feeding west of Crieff and Loch Cowdens was
used by those birds feeding at the Comrie end
of the Carse. At times there was considerable
mixing with Pinkfeet and Greylag from
Carsebreck in the area south-west of Muthill.

In the late 1980s, 100-200 Greylag fed in the
Buchanty area in late winter and spring. The
roost for these birds was not discovered but
Loch Meallbrodden or Loch Buchanty were
the most likely sites. In the 1990/91 winter
Greylag were particularly numerous in the
Buchanty area with up to 900 roosting at Loch
Buchanty from mid-November to
mid-December and smaller numbers to
mid-January when shooting dispersed the
flock. Some also roosted at Loch
Meallbrodden on occasion and two small hill
lochs in Logiealmond were also believed to
have been used. Large numbers were also
present the following winter with 1250 at Loch
Meallbrodden on 25 October (when there
were also 5050 at Drummond) and 2000 were
feeding in the Buchanty area by early
November. Very few Greylag were noted in
Glen Almond in the 1992/93 and 1993/94
winters. Since the appearance of Greylag in
autumn in Glen Almond small flocks up to 210
birds have been noted regularly at Loch
Freuchie near Amulree. These birds often
flignted to Glen Almond.

Satus of wintering Pink-footed & Greylag Geese in central Scotland 39

4. TAY VALLEY
a) River Tay, Scone.

The Greylag roost on the Tay at Scone held
its largest numbers from November to
February with a regular winter count of c.1000
(Table 6). The peak recorded during this
study was 1730 in February 1991. Some of
the monthly totals from which the average
counts were derived were of birds feeding in
the Scone and/or Luncarty areas and may
therefore be low. This was another awkward
roost to count because the geese could be
spread over at least 1km of bank and some
flightlines were obscured by trees. There was
always the worry that some birds roosted
further upstream by the mouth of the Almond
and along the racecourse. This roost became
untenable during spates and it was not clear
where the birds then moved to but on
occasions large numbers of droppings and
feathers were found on the North Inch in
Perth. The ponds at Dunsinnan House and
South Friarton near St Martins have been
reported to hold geese in late winter though
we have no direct evidence that these sites
were used as night roosts during this study.

The main feeding areas were within 4-5km of
the river to the east and north-east in the
Scone area but some birds also fed in the
Balbeggie-St Martins area some 6km distant.
Greylag also flew upstream to feed in the
Luncarty-Stanley area and occasional flocks
were found in the Moneydie-Pitcairngreen
area which may also come from the Tay.

b) Lochs Tullybelton & Mullion and the
Bankfoot area.

Loch Tullybelton and Loch Mullion are small
hilllochs which lie 14-16 km WSW of Kercock.
Pinkfeet first roosted at Loch Tullybelton in
the late 1960s when barley replaced oats as

40 Bell & Newton.

the main cereal crop on the farmland to the
east. The local estate believed that up to 5000
geese used the loch in autumn with few
present in mid-winter and small numbers in
spring. Counts since the 1989/90 winter
showed this to be an accurate assessment.
Pinkfeet were first noted at Loch Mullion in
October 1991; there were none here in
November 1989. Pinkfeet at both lochs flighted
out up to 12km between ENE and SSE to feed
in the fields towards Luncarty and Airntully,
the area round Bankfoot being particularly
favoured, with a few birds sometimes flighting
south into Glen Almond.

In the 1987/88-1988/89 winters, geese were
systematically counted on the fields in the
Luncarty-Airntully-Bankfoot area. At first it
was thought that many if not all the Pinkfeet
here were commuting from Dupplin Loch, 15
km SSW, but subsequently large flocks were
found on several days when it was known that
none had left Dupplin on this flightline. Pinkfeet
were not observed at Kercock in the 1988/89
winter. In spring of 1989 Pinkfeet were
observed loafing at Loch Tullybelton during
the day and permission was obtained to visit
the loch to count the geese at dawn the
following winter. The field counts may have
overestimated the use of Loch Tullybelton in
mid-winter since roost counts subsequently
showed few birds to use it at that time. Loch
Tullybelton has held 3000-5800 Pinkfeet in
autumn, peaking in either October or
November. Numbers usually fall sharply
thereafter and remain low until March or April
(Table 6); 2050 in April 1990 was the largest
spring count noted during this study. The loch
was often frozen for long periods from
December to February and then held no
birds. King’s Myre near Airntully was used by
200 Pinkfeet in October 1990 and was
probably used on other occasions. The river
Tay seems the only other possibile roost.

SB 18(1)

Loch Mullion held Pinkfeet in October and
November of the last three winters with 2550
in mid-October 1992 the most, giving a
combined total of 8350 for the two lochs. Loch
Mullion was not checked systematically in the
spring but 2000 roosted there in February
1994.

Greylag were not recorded at Loch Tullybelton
but small flocks of up to 100 birds were found
at Loch Mullion on three out of five autumn
counts. They probably fed to the south in Glen
Almond. Greylag were always found in the
fields in the Luncarty, Stanley and Bankfoot
areas with 500-1000 regular and a peak of
1530. These birds could have come from a
number of roosts, the Tay at Scone or Kercock
being the most likely. Several hundred
regularly fed north of Bankfoot in late winter
and these might have roosted on the nearby
Stare Dam or flighted north to Loch of Lowes
some 7km away. A small pool near Airntully
was used by Greylag 20 years ago (Newton
et al. 1973) but was not checked during this
study, though birds sometimes fed close by.
Kings Myre was another possible roost for
Greylag feeding in this area. Cairnleith Moss
at Bankfoot was a loafing area for both species,
especially Pinkfeet, and it is possible that
small numbers of Greylag might have roosted
in the marsh there.

c) Blairgowrie Lochs and river Tay,
Kercock

This is a complex series of roosts comprising
six lochs separated by up to 16km situated
between Blairgowrie and Dunkeld, and the
river Tay at Kercock-Meikleour four km to the
south (Fig. 1). These roosts were formerly
one of the most important sites in Britain for
Greylag but numbers have fallen substantially
over the last 10-15 years from regular

1995

November totals of 7000-15,000 to 3000-5000
(Table 6). Unfortunately two of ihe November
counts during this study (1989 & 1990) were
badly affected by fog and many birds were
missed; these two seasons were omitted in
calculating the November average. Numbers
usually peaked in November and December
falling back later in the winter with 1700-3000
present from January onwards (Table 6).
Numbers dropped sharply in March, but this
was probably due io birds dispersing onto
smaller pools since Greylag at other sites did
not leave the area this early. The lochs were
not used for staging by Greylag in spring. The
largest count recorded during this study was
7415 on 17 October 1992 and the only other
counts over 6000 birds were in October 1990
and November 1988. This suggesis that the
area can no longer sustain such numbers
through the winter and that many move on
shortly afier arrival. In recent winters c.200
birds using Marlee Loch were believed to be
feral.

The numbers of Greylag using the different
roosts varied considerably between and within
winters depending on disturbance, availability
of food and weather conditions. Loch of Lowes
held the largest numbers shortly after the
birds arrived in October through into November
with birds moving on quickly thereafter,
200-300 was a good total for the New Year
(Table 6). This site was often frozen for long
periods in mid-winter. There used to be
frequent interchange between geese roosting
at Loch of Lowes and Loch of Butterstone but
the latter site has been deserted since a fish
farm was built there. Loch Clunie also tended
to hold its largest numbers in the autumn but
rather more birds used this site later in the
winter (Table 6). Numbers at Marilee Loch
were rather variable between winters and the
largest counts occurred from November to
January. This roost seems to have become

Satus of wintering Pink-footed & Greylag Geese in central Scotland 41

more important in the last few winters, its
larger size perhaps offering a better refuge
from shooting. Stormont Loch was only
counted regularly in November and appears
to have become much less important than a
decade ago, birds apparently switching to
Marlee. The adjacent Fingask/White Loch
only held substantial numbers (646) once
during this study in February 1989. Monk
Myre, the eastern-most of the lochs and one
of the most important roosis a decade ago
was not used.

At Kercock and Meikleour on the Tay the
geese roosted on the shingle banks along
c.4km of river. This site recorded over 1800
Greylag in every month from October to
February with a regular count of c.1500 (Table
6) and was particularly important in mid-winter
when the lochs were frozen. However, it was
untenable when the river was in spate.
Numbers here were also much lower in March
and April (Table 6), possibly due to disturbance
from fishermen.

In autumn 1992 itwas discovered that several
hundred Greylag were flighting from the
Blairgowrie area to roost on lochs in the hills
to the north-west. A small lochan on Cochrage
Muir was used by up to 300 birds and up to
150 geese appeared to be flighting to Loch
Charles some 10km north of Loch Clunie.
Loch Benachally may also be used as a roost
at times. A further complication in assessing
numbers in this area can be floodwater and it
is possible that many of the lower counts
obtained later in the winter during mild weather,
and especially after the shooting season,
missed birds roosting on floodwater or on the
hill lochans.

The geese generally fed on the mixed farmland
within c.4km of the roosts but those birds
which flighted east and north-east from

42 Bell & Newton.

Kercock might have foraged further afield.
The decrease in numbers and change in
distribution of roosting birds reflected changing
agricultural practice to an autumn sown cereal
and oilseed regime, particularly east of
Blairgowrie into Strathmore. This has led to
increased disturbance associated with crop
protection and there is also much more
commercial shooting. This accounted for the
drop in numbers roosting at Monk Myre and
Stormont Loch. Few birds flighted east into
Strathmore, an area where syndicate goose
shooting has been particularly prevalent, and
the associated disturbance severe.

Pinkfeet first started to roost at Kercock during
the spring inthe late 1980s with 2100 recorded
on 1 April 1988. Since then they have become
regular with records throughout the winter
(Table 6) though the largest flocks still tend to
occur in autumn and late spring. The Pinkfeet
tended to roost separately from the Greylag
further upstream. They usually flighted out to
the south and west and there was probably
considerable overlap on the feeding areas
with birds from Loch Tullybelton. Indeed the
Tay at Kercock was probably used as an
alternative roost when Loch Tullybelton was
frozen.

Flocks of Pinkfeet have become increasingly
frequent at the Blairgowrie Lochs though they
rarely stay for long. There were 900 at
Stormont Loch on 13 November 1988 and
150 there on 21 October 1990 when there
were also 350 at Loch Clunie, while 1233
were at Loch of Lowes on 30 September
1990. Small numbers have been recorded
there and at Marlee on several occasions.

d) Upper Tay and Tummel Valleys

Small numbers of Greylag were found on the
November counts in Strathtay in the 1960s

SB 18(1)

with 1200 in 1967 the most. However numbers
appeared to decline sharply after 1972 with
none located between 1973 and 1975.
Numbers have since more than recovered
and by the end of the 1980s the flocks were
the largest recorded with a regular wintering
population of c.1900 birds (Table 6). The
geese arrived in the second half of October
and peaked in November and December,
2603 in November 1992 being the highest
total recorded. Numbers then fell back slightly
later in the winter but 1000-1500 usually
remained throughout unless snow moved the
birds on. There was aslight increase in March
and April as passage birds moved through
the area. Sizeable flocks often remained to
mid-April, later than at most other Greylag
sites in the area; 1885 were present on 14
April 1991.

This was a difficult area to count. The roosts
of most flocks were unknown and therefore all
counts were of geese feeding in the fields. A
drive of c.150 miles was required to cover all
the sites which took one observer about six
hours. The short day-length in mid-winter
made this impossible on anything other than
a fine day. During this study the main flocks
were located on the riverside haughs south
and west of Balliniuig and between Aberfeldy
and Dull, in the Moulin area of Pitlochry and at
Tulliemet near Ballinluig. Smaller flocks were
regularly found at Loch Tummel, Dunalastair
reservoir and Loch Tay with occasional flocks
at Glen Fincastle, Loch Rannoch and Loch
Moraig, in a total of 86km? (Fig. 3). A
complicating factor was the presence of c.200
feral birds which were usually found at Loch
Tummel. However, these birds were not
always discrete and mixed with the migrant
Icelandic birds at times.

Greylag feeding in the Ballinluig area may
have roosted on river shingle but also probably

1995

used Lochs Ordie, Benachally and Broom;
geese feeding in the Pitlochry area may have
also used the latter site as well as flying
upriver to roost at Loch Tummel or Loch
Moraig. The birds feeding in the Appin of Dull
used Loch Glassie and perhaps also Loch
Tay at times. Those geese feeding at Lochs
Tay, Rannoch and Tummel and at Dunalastair
reservoir roosted on these waters. There
were also a number of other hill lochs which
may have been used as roosis.

5. TAY ESTUARY

Only the western end of the inner Tay estuary
isin Perthshire, the remainder divided between
Angus and Fife. This is a very difficult site to
count, almost 20km in length from the mouth
of the Earn to Invergowrie Bay at Dundee,
with geese roosting anywhere along this length
depending on disturbance. A minimum of at
least six counters is required to cover the site
properly. Unfortunately several attempts at
coordinated counts during this study were
severely hampered by bad weather and the
status of both species isstill very unclear,
especially in late winter and spring. Acount of
c.10,000 geese, almost certainly Pinkfeet,
near Mugdrum in mid-March 1992 suggested
the area could be very important in spring.
Geese found feeding at Rhynds by the mouth
of the Earn almost certainly roosted on the
estuary. Pinkfeet also flighted NW over the
Sidlaws and south into Fife but these feeding
areas were not located.

6. LOCH LEVEN AND THE KINROSS
BASIN

_ a) Loch Leven

_ The populations of waterfowl at Loch Leven

Satus of wintering Pink-footed & Greylag Geese in central Scotland 43

are the best documented in Scotland, see
Allison etal. (1974). Pinkfeet were using Loch
Leven by the turn of the century with autumn
peaks of 2000-3000 birds while Greylag were
scarce. Numbers of Pinkfeet increased after
1920 and Greylag only colonised in numbers
after 1940. By the 1950s the site was a major
arrival point for Pinkfeet and wintering site for
both species.

Loch Leven was the first site in Britain where
a detailed study of the winter ecology of these
species was undertaken (Newton & Campbell
1973). Inthe late 1960s and early 1970s there
was an early autumn peak of 10,000-15,000
Pinkfeet at the end of September or early
October, then numbers fell away through the
autumn with 3000-5000 usually present from
December to April. There was no peak during
the period of spring passage. The build up of
Greylag was much slower peaking at around
3000-5000 in late autumn or mid-winter and
again no noticeable spring peak (Newton &
Campbell 1973). This general picture
pertained until the mid-1980s with a trend of
slightly larger numbers of Pinkfeet remaining
through the winter. Since then the numbers of
Pinkfeet have increased markedly with over
20,000 recorded in several recent Octobers
and 18,000 in November 1989 (Table 7).
Typically 5000-8000 now remained through
the winter decreasing in April as birds moved
north. Regular counts showed thatthe autumn
peak was uSually around the second or third
weeks of October a week or two after the peak
at Dupplin Loch. In contrast numbers of
Greylag fell sharply in the 1990/91 winter
(Table 8) and all but disappeared. Small
numbers of feral birds confuse the picture, but
a Greylag ringed in a catch of 20 on 15th
March 1993 was shot in Iceland on 25 April
1993 (A. Lauder pers. comm.) so migrant
Greylag were still using the site in that winter.

SB 18(1)

44 Bell & Newton.

QZI =

6 ou
062 ou
O61 0
ZSvl ou
SZ8 ou
L9L¢C OOV
2902 ou
LOLI ou
Jeinboa1 idy

Je

qes

V6/E6EL O} Z6/L661 40J S| BBeJaAY ‘pajunod jou = ou

L8

uer

SLL
9€
Zl
SVL
0081
Ou
OOZI
OSce
OOS|

9eq

cl
OZ

AON

L8
0€
6€
GZ
/L Sit
Ou
O0cl
OcZ
v9Ol

190

ebesene

v6/E66 1
€6/c661
c6/166 1
16/0661
06/6861
68/8861
88/2861
28/9861

‘SJOJUIM P6/E66L - 28/9861 ‘UaNa7 YI07 Je asaay BejAas5H Jo sjunos Ajyjuow winwixeyy “8 ATAWL

LOLSL :
0S96 ou
EvLEel ou
VVCEL 00SS
OOSIL ou
LOCEL Ou
CEC6 OOZV
CEvs ou
00cZ ou
Jeinbe: udy

L9E9
G88e
0029
c808
ou

0089
OOLS
OOcE
O06E

JEN

9ES9
0002
OSZZ
GL89
O8SV
ou

0009
0OSS
OOLS

qes

‘SDIEMUO 06/686 JO) Si abeIAAW ‘pajUNoD jou = OU

2669
2669
0092
OLLZ
00SZ
ou

0089
0069
O08

uer

O088S
OS¢S
0198
098¢
0089
Ou

00¢S
OOZZ
OOSYV

99q

cLVib I
O98ll
O€Z9
OZZ6
OOOl I
00081
0006
0026
0OSOl

AON

80cZ1
06001
OLOEC
O88le
00091
000S|I
00ccl
0062

Oc617

190

oebeione

v6/E66L
€6/c66 |
c6/166 |
L6/066 |
06/686 |
68/8861
88/286 |
28/986 |

‘SI9]UIM P6/E66L - 28/986 UBAAT YIOT Je asSaaty pajooj-yUld Jo s}uNOD Ajyjuow winwixeyy “2 AVGVL

eee)

Inthe early 1970s most Pinkfeet and Greylag
fed within a few km of the loch within the area
bordered by the perimeter roads (Newton &
Campbell 1973). Twenty years later it was
difficult to see a goose in this area away from
the RSPB reserve at Vane Farm. The Pinkfeet
flighted many km west towardsPowmill,
north-easttowards Strathearn, the Tay estuary
and Cupar and east towards Glenrothes.
Many feeding sites in these latter areas were
not found during this study and figures 2 and
3 are apoor representation of the extent of the
feeding area used by geese from Loch Leven.
To the north there was overlap with Pinkfeet
roosting at Glenfarg Reservoir, Dupplin Loch
and the Tay estuary.

Incommon with allthe arable areas of eastern
Scotland used by wintering geese large
changes have occurred inthe cropping regime
over the last 10-20 years which have had a
dramatic effect on the distribution of feeding
geese. Also, almost all the ground within the
perimeter roads which was formerly the main
feeding area is now shot over regularly. From
the top of Benarty Hill on a clear day it was
possible to observe the flocks of Pinkfeet
being constantly moved around the Kinross
basin by shooting parties (J. Burrow pers.
obs.). Disturbance from shooting was a major
determinant of the number of birds and their
distribution within the area. The RSPB is now
returning about 120ha of arable farmland at
Vane Farm to wet grassland for breeding
waders. This will also benefit wintering geese
by offering some of the birds a secure feeding
area and should help to ease the conflict with
local farmers. :

b) Glenfarg Reservoir
Glenfarg Reservoir is situated at the eastern

end of the Ochil Hills some 8km NNW of Loch
Leven. Up to 1500 Pinkfeet were recorded

Satus of wintering Pink-footed & Greylag Geese in central Scotland 45

here on three November counts in the 1960s
but were rarely found in the 1970s. During this
study large numbers were found on the
mid-October counts with 3990 in 1990 and
4800 in 1993. Numbers dropped rapidly in
November because of disturbance from
shooting atthe reservoir and inthe surrounding
area. The site did not appear to be used by
roosting geese in mid-winter or spring though
birds from Loch Leven and Dupplin sometimes
fed in the area. In the 1970s up to 1500
Greylag were regularly found at Glenfarg on
the November counts but numbers have been
much lower since and 226 was the largest
flock found during this study.

Conclusions

Large changes in the status and distribution
of both species were found compared to
20-25 years ago (Newton et al/.1973; Newton
& Campbell 1973). Major concentrations of
Pinkfeet (regularly over 10,000 birds) were
present atthree sites, Loch Leven, Strathearn
and Strathallan, though Pinkfeet roosted
regularly at21 sites and were recorded feeding
in 466 km?. A number of new roosts were
occupied inthe last decade and the expansion
in range Is continuing. The November counts
are the only long-term index of goosenumbers
available. During the 1960s and 1970s the
numbers of Pinkfeet in the area in November
reflected the national trend and showed a
steady increase but the large expansion of
the population since then has not been
reflected here and it appears that the area
has reached its carrying capacity (Fig.4). The
feeding area is much more extensive than
that described by Newton et al. (1973).

The largest numbers of Pinkfeet were present
in early to mid-October with a total of 86,868
on 12/13 October 1991. More typicaly c.65,000

46 Bell & Newton. SB 18(1)

Figure 4. The November total of Pink-footed Geese and their proportion of the total
population 1961-1993.

50 100
Q regional total % of UK total ~
< 40 80 6
= Y
= 30 eo *S
carat oO
& 0)
2 20 40 &
ae ati20 5
ja @) 0
1961 1971 1981 1991

Year

Figure 5. Average monthly counts of Pink-footed Geese at the main roosts: a =
Skingflats, b= Carse of Stirling, c = Strathallan, d = Strathearn, e = Lochs Tullybelton &
Mullion and Kercock, f = Loch Leven. Note different scales. x = insufficient data.

Thousands
oO
‘@)- ()

sen
OFN™D S0ae Nite
Pinkfeet

1995

were present at this time falling to about half
this number by mid-November with
20,000-25,000 remaining throughout the
winter. Most sites held their peak numbers in
October or November (Fig. 5). Dupplin Loch
was identified as a key site when the geese
first arrive in Britain, regularly achieving peaks
of over 30,000 in late September and early
October with 57,500 on 28 September 1991
accounting for a quarter of the population in
that year. Strathallan was the only site to
show a marked spring peak with large numbers
remaining here to late April or early May (Fig.
5):

The wintering population of Greylag was much
smaller and more dispersed than that of
Pinkfeet, roosting on at least 55 sites and
feeding in 491km?. Numbers were relatively
steady throughout the winter with little
discernible autumn or spring passage (Fig.
6), reflecting the fact that the area is on the
southern edge of the wintering range. The
population has undergone a large decline
and the November totals for the region are
now the lowest since counts began in 1960
(Fig. 7) though the birds are spread over more
sites. All the major roosts have lost birds and
some have almost been deserted, e.g. Loch
Leven. This has made it very difficult to obtain
a coordinated count of Greylag in November
and the only way to census flocks in several
places is to systematically search the feeding
areas which is very time-consuming and
dependent on reasonable visibility.

Many areas are nowless suitable for wintering
geese. Changes to autumn sown cereals and
oilseeds in the 1980s left less stubble available
in autumn and early winter and brought conflict
with farmers. Rotational set-aside may reverse
this change by leaving stubbles in areas
where autumn sowing is the norm. However,
long-term set-aside, particularly in areas

Satus of wintering Pink-footed & Greylag Geese in central Scotland 47

where spring ploughing and sowing was usual
as in Strathallan, has eliminated many suitable
fields and probably caused the recent
decrease innumbers here (Bell eta/. in prep.).

Observers who have been counting geese
since the 1970s can testify to the great
increase in disturbance to feeding geese from
shooting which has occurred over the last 20
or so years. There is also much anecdotal
evidence from people living in the countryside
to this effect. In theStrathallan area the
incidence of goose shooting on the fields has
increased at least three fold since 1988 (Bell
etal.inprep.). Many farms which were formerly
important feeding sites for both species have
become untenable because of disturbance
from shooting, particularly in Strathmore and
Strathearn. Greylag appeared much more
vulnerable than Pinkfeet to shooting on the
feeding areas and seemed unwilling to move
to new feeding areas in response to shooting.
This was partly a reflection of the fact that
most Greylag fed in relatively small areas
close to their roosts, in marked contrast to the
situation found in north-east Scotland where
Greylag flighted further than Pinkfeet (Bell
1988). Pinkfeet were more wary than Greylag
but, because they fed in larger flocks, were
better able to take advantage of a good food
supply before being disturbed. They were
much more mobile and less field-faithful on a
day to day basis than Greylag and disturbance
had less effect on them. However, being
more wary, they were likely to desert a site if
disturbed at the roost.

Disturbance of geese by shooting atthe roosts
was minimal at most sites during this study.
All the major roosts are SSSIs and in many
cases SPAs and Ramsar sites and one is a
NNR. The roosts are thus protected by the
highest British and European conservation
designations. In spite of this the numbers of
Greylag have fallen substantially to the point

48 Bell & Newton. SB 18(1)

Figure 6. Average monthly counts of Greylag Geese at the main roosts: a = Carse of
Stirling, b = Strathallan, c = Strathearn, d = Scone, e = Blairgowrie Lochs & Kercock, f
= upper Tay/Tummel. x = insufficient data.

NO
x
x
sad)

ONONPONOND FON O
Q.

Thousands

ee
OLN. ID) sie Nin
Greylag

Figure 7. The November total of Greylag Geese and their proportion of the total
population 1961-1993.

100

O1
O

regional total % of UK total

Regional total (thousands)
Percentage of UK total

OO

961 1971 1981 1991
Year

1995

of disappearance at some formerly major
roosts. The average November total for Loch
Leven, Dupplin Loch, Drummond Loch and
the Carsebreck Lochs was 3132 for the three
autumns 1991-93 compared with 11,542 for
the corresponding three autumns a decade
earlier and 19,847 in 1971-73. The need for
secure feeding areas is now a priority if the
present wintering range of Greylag is to be
maintained. Protecting the roosts has failed
to safeguard this population. Set-aside has
been a huge wasted opportunity to solve the
conflict between geese and farmers and
indeed in some areas has exacerbated the
problem. We believe many farmers and the
major estates would welcome an option to
manage set-aside land for geese if an
appropriate scheme was put in place by
SOAFD or an equivalent body.

Acknowledgements

We are very grateful to those landowners who
granted permission for access to roosts. The
vagaries of the Scottish climate can make
counting geese extremely unpleasant or
exhilarating, sometimes on the same morning.
We are very grateful to the following who
contributed to this study in their own time and
at their own expense, and especially to those
who counted sites regularly and drove many
miles looking for feeding flocks: B.Anderson,
J.Babbs, A.Barclay, D.Bell, L.Borradaile,
W.R.Brackenridge, R.A.Broad, A.Brown,
D.M.Bryant, J.Burrow, M.Callan, E.Cameron,
J.Evans, D.Fairlamb, J.Gallagher, S.Hacker,
N.Harding, S.Hewitt, J.Kirk, S.Laing,
A.Lauder, J.Law, W.Mattingley,P.McPhail,
G.McShane, A.V.Newton, |.Pirie, G.Powell,
K.Ringland, A.Robertson, S.& K.Sankey,
C.Scotland, M.& S.Shimeld, G.Smith,
R.Smith, E.& M.Still, M.Taylor, N.Taylor,
R.E.Thomas, K.Thomson, K.Thompson,

Satus of wintering Pink-footed & Greylag Geese in central Scotland 49

V.Underwood, J.Wallace, G.Wright,
R.E.Youngman. We also thank the BTO fora
small grant award made to the Central
Scotland Goose Group at the start of this
study, Carl Mitchell for supplying some
November counts from the late 1980s which
were missing from our files, and A.V. Newton
for help with the figures. We thank E. Cameron,
A. Lauder, W. Mattingley and A.V. Newton for
reading a draft of this manuscript.

References

Allison, A., Newton, |. & Campbell, C. (1974).
Loch Leven National Nature Reserve.
WAGBI.

Atkinson-Willes, G.L. 1963. Wildfowl in Great
Britain. Monographs of the Nature
Conservancy Number 3. HMSO.

Bell, M.V. 1988. Feeding behaviour of
Pink-footed and Greylag geese in
north-east Scotland. Wildfow 39:
43-53.

Bell, M.V., Dunbar, J. & Parkin, J. 1988.
Numbers of wintering Pink-footed and
Greylag Geese in north-east Scotland.
Scott. Birds 15: 49-60.

Berry, J. 1939. The Status and Distribution of
Wild Geese and Wild Duck in Scotland.
International Wildfowl Inquiry: Volume
//. Cambridge University Press,
Cambridge.

Boyd, H. & Ogilvie, M.A. 1969. Changes in
the British-wintering population of the
Pink-footed Goose from 1950 to 1975.
Wildfowl 20: 33-46.

Boyd, H. & Ogilvie, M.A. 1972. Icelandic
Greylag Geese wintering in Britain in
1960-1971. Wildfowl 23: 64-82.

Forshaw, W.D. 1983. Numbers, distribution
and behaviour of Pink-footed Geese
in Lancashire. Wildfowl 34: 64-76.

Newton, |. & Campbell, C.R.G. 1973. Feeding

50 Bell & Newton.

of geese on farmland in east-central
Scotland. J.Appl. Ecol. 10: 781-801.
Newton, |., Thom, V.M. & Brotherston, W.
1973. Behaviour and distribution of
wild geese in south-east Scotland.
Wildfowl 24: 111-121.
Owen, M., Atkinson-Willes, G.L. & Salmon,

SB 18(1)

D.G. 1986. Wildfowl in Great Britain
(2nd Edn). Cambridge University
Press, Cambridge.

Thom, V.M. 1986. Birds in Scotland. T.& A.D.
Poyser.

Thompson, K. & Harding, N. (in prep.) RSPB
report to SNH.

M.V. Bell & S.F. Newton Central

Scotland Goose Group

c/o 48 Newton Crescent Dunblane Perthshire FK15 ODZ

Revised typescript received 9 February 1995

Scottish Birds (1995) 18: 51-60

51

Short Notes

Diurnal and tidal activity patterns of Purple Sandpipers on the Isle of May, Fife.

In many studies of the winter feeding ecology
of waders, nocturnal activities of the birds are
neglected. Waders equipped with touch-
sensitive bills can locate prey in soft substrates
and potentially can feed at night. Whilst this
has been shown for some sandpipers feeding
on mudflats (McNeil 1991. Proc. Int. Orn.
Congr. 20: 1098-1104), there seems less
scope for tactile-foraging on rocky shores.
The Purple Sandpiper Calidris maritima is
unusual in that it winters on rocky shores and
occurs further north than any other wader,
experiencing long nights in winter. The aim of
this study was to obtain a better understanding
of the night-time activities of Purple
Sandpipers.

a te

The study was done from 14 - 17 April 1993
and 9 - 16 April 1994 on the Isle of May, Fife
(Plate 1). Four Purple Sandpipers were
dazzled-netted at night, one on 14 April 1993,
and three on 9 April 1994 on the east side of
Rona (Fig. 1). Transmitters, weighing 1g,
were glued to the back feathers and the birds
were released at the site of capture.

During two 24-hour periods, the radio signals
from the transmitters were listened to for a
minute, each half hour. The signal was classed
as ‘varying’ or‘steady’. Varying signal strength
indicated that the bird was active, due to the
changing alignments of the transmitting
antennae relative to the receiving antenna,

Plate 1. The Low Light and the north end of the Isle of May where the study was carried

out.

52 Short Notes

SB 18(1)

Figure 1. The Isle of May showing sites where the radio-tagged Purpe Sandpipers were
subsequently relocated. The numbers give the frequency of records. They were all
captured on the south-east part of Rona.

which has held still. Steady signals indicated
that the bird was standing still. The birds were
inactive only around high tide and were active
at low tide regardless of whether it was day or
night (Fig. 2). The fact that they were active
at night suggests that they were feeding. Itis
not known how they detect prey on rocky
shores at night, though Purple Sandpipers do
have atouch and taste-sensitive bill (Gerritsen
& Sevenster 1985. Fortschritte der Zoologie
30: 237-239) and do not need to rely on visual
cues to detect prey.

Information on the areas occupied by the
birds was obtained by triangulation. The

locations of birds were monitored at irregular
intervals but only those at intervals greater
than 3h were plotted. All the birds were
located within small areas on the island (Fig. 1),
usually close to where they were captured.
Bird 1 remained between east Rona and
Colm’s hole, spending most of the time on the
Tarbet rocks (Fig. 1). Bird 2, which was
located every time it was searched for, hada
restricted range over six days.

Birds 1, 3 and 4 went missing on occasions.
At dawn on 17 April 1993, bird 1 disappeared
from the area where it spent the previous day
and could not be found anywhere on the

1995

Short Notes 53

Figure 2. The activity patterns ( —_ active, HEM inactive) of three radio tagged Puple
Sandpipers in relation to tide and day/night ((__|night).

15-16April High tide

11 - 12 April

Day

Bidisgmememiomm oho | ia are eee a

Gk? |: [Set Eti me Ee Tee 6a!

Low tide

High tide

Low tide

Night

18) 420" 22> 24° 02. 04

Time (hours)

island. Later that day, the bird was located on
the mainland at Fife Ness, 8km from the
island. The bird had flown there in a force 4-
5 wind. Bird 3 had no set pattern to its arrivals
and departures. It remained on the island
during 11 April and departed early on 12 April.
It then returned for the night of 12-13 April but
was notlocated again until 2030hrs (GMT) on
15 April. It was still on the island at O630hrs
on 16 April when observations stopped. Bird
4 spent no daylight hours on the island, arriving
between 1930hrs and 2055hrs when it was
almost dark and departing between 0300hrs
and 0330hrs when daylight was just breaking
(sunrise was about 0515hrs). This bird was
also located at Fife Ness on 16 April 1994.
The weather was very windy (force 7,
northerly) during some of the times when the
bird commuted to and from the island.

Colour-ringed Purple Sandpipers marked on
the Isle of May have been sighted at many
localities in the outer Firth of Forth and north
along the Fife coast to St Andrews, as well as
on the island itself (Atkinson et a/. 1981. Ornis
Scand. 12: 18-27). This information together
with the data on the departures and arrivals of
radio-tagged birds indicate that the island is
used in different ways by Purple Sandpipers.
There are those that appear to be resident on
the island where they occupy small areas.
Also, there is evidence that some birds use
the island irregularly and there is turnover in
the Isle of May population. Then, there are
birds which use the island at night and
commute to the mainland for the day. In
earlier studies on the Isle of May, Summers et
al. (1975. Scott. Birds 8: 299-308) suspected

54 Short Notes SB 18(1)

that Purple Sandpipers commuted from and __ roosting place, perhaps because there are no
to the island, using the island as a safe _ rats on the island.

R.W. Summers, Lismore, Mill Crescent, North Kessock, Inverness IV1 1XY.

Plate 2. A Purple Sandpiper at a roost. Photographs by the author.

1995

Short Notes 55

Little Ringed Plovers attempt breeding in Dumfriesshire

As BTO representative for Dumfriesshire, |
was contacted by a bird-watcher who had
founda pair of Little Ringed Plovers Charadrius
dubius on 27 May 1992. Oncondition that the
site remained confidential we visited the
_ locality together on 31 May and watched an
incubating bird with its mate. When we
returned on 7 June both birds were feeding at
the water's edge; the scrape was empty and
the birds had failed. When we returned on 12
June a bird was incubating a clutch 60m away
from the original scrape. The birds were twice
observed chasing off Ringed Plovers
Charadrius hiaticula which approached their
nest. On 24 June and 4 July the birds were
still incubating this clutch and the original
observer returned on 7 July hoping to see the
pair with small young, but the adults had
disappeared. The site was searched again
on 13 July but no Little Ringed Plovers were
seen and we concluded that the pair had
failed at around the time of hatch.

In 1993 Little Ringed Plovers were first seen
on 9 May. There were two birds, one of which
performed a ‘butterfly’ type song flight over
the territory of the previous year, then
performed a scrape-display. On 11 May two
male Little Ringed Plovers were in song-flight
and a third bird was also seen. Hopes of
successful breeding were dashed by the

weather. Snow followed by heavy rain on 13-
15 May caused flooding and all suitable
nesting habitat disappeared under water for
at least two days. Although the site was
regularly visited throughout the season the
birds were not seen again. They were not
seen at all in 1994 despite frequent searches
at the appropriate times.

There are two former records of Little Ringed
Plovers breeding in Scotland; in Lanark in
1968 (Stalker, D. Scott. Birds 5: 282-3) and
at a disused gravel pit in Fife in 1989 (Oliver,
D.W. Scott. Birds 16: 42-3) but the site was
afterwards destroyed. Adisplaying male was
seen at another gravel pit in Fife in 1990, but
no female appeared and this site also was
subsequently filled in. The New Atlas of
Breeding Birds in Britain and Ireland: 1988-
19971 (1993, Gibbons, Reid & Chapman)
shows the Fife breeding record and two other
‘sight records’ for Scotland. The Atlas also
indicates a gradual spread of Little Ringed
Plovers into northern England, with the nearest
breeding birds being near the Solway, in
Cumbria. With so much recent gravel
extraction in Dumfriesshire for the upgrading
of the A74 to motorway standard it is possible
that other pairs may have attempted to breed
or may do so in future years.

Richard Mearns, Connansknowe, Kirkton, Dumfries DG1 1SX.

56 Short Notes

SB 18(1)

Numbers and breeding success of Manx Shearwaters on the Isle of Canna,

1973-94.

In April 1973 and 1974, all suitable Manx
Shearwater Puffinus puffinus sites on the
island were visited and the number of burrows
with shearwater droppings at their entrance
counted. Observation burrows were then
used to calculate the number of burrows
subsequently laid in. This gave a figure of
1000-1500 breeding pairs on Canna (Swann
& Ramsay 1976 Seabird Report 5:38-41).

In April 1993 the process was repeated. 253
burrows with droppings were located along
the crags by the Tarbert Road, on the south
side of the island, and none elsewhere. Of 35
study burrows subsequently checked, which

had had droppings at their entrance, only 17
(49%) were laidin. Two others were occupied
by single birds and a third contained a dead
adult. These figures suggest that there were
only 124-144 pairs of shearwaters nesting on
Cannain 1993. In April 1994 only 117 burrows
with droppings were located though
exceptionally heavy rain may have led to this
being an under estimate. These figures
indicate a massive decline in the number of
shearwaters nesting on Canna over a 20 year
period. The cause of this decline could be a
reduction in breeding success or increased
mortality rates or a combination of these two
factors.

TABLE 1. Number and percentage of occupied Manx Shearwater burrows on
Canna and breeding success as a percentage of eggs laid that fledged young.

1976 1977 1978 1979 1980 1981 1982 1983 1984. 13e5si3e6

no.occupied burrows 35 45 49 46 41 38. 42 38. 46 ~AA waz
% observation burrows

occupied 73% 72% 75% 78% 77% 66% 73% 82% 81% 67% 66%

% young fledged per

egg laid 51% 82% 73% 61% 76% 25% 2% 68% 59% 66% 65%
1987 1988 1989 1990 1991 1992 1993 1994

no. occupied burrows 41 31 13 21 25° 26 ti 16
% observation burrows

occupied 66% 65% 24% 42% 43% 45% 33% 32%
% young fledged per
egg laid 68% 42% 31% 52% 52% 46% 47% 75%

1995

Short Notes 57

Breeding success has been monitored on
Canna since 1976 (Table 1) using up to 60
observation burrows. The percentage of
occupied burrows was fairly high from 1976-
1984 (Table 1) but began to fall after 1984 and
dropped even further from 1989-1994.
Breeding success has tended to be variable.
Low success in 1976 and 1982 was attributed
to high levels of rat predation in the colony but
applications of rat poison in subsequent years
appeared to cure the problem (Swann &
Ramsay 1984. Scott. Birds 13:40-47). To
ascertain whether rat predation was still a
problem, chew sticks (spatulas dipped in
moiten margarine) were placed throughout
the colony in April 1994 to detect the presence
of rats. The sticks were collected three days
later. None had been chewed.

Excluding these ‘poor years, breedings
success tended to be about 0.59-0.82 young
per egg laid (1976-87). This compares witha
mean of 0.7 on Skokholm (1973-76) (Brooke
1990. The Manx Shearwater, Poyser,
London.). Between 1988-1993 breeding
success was markedly lower with only 0.31-
0.52 young per egg laid. The cause of this
decline in breeding success is not known.
Since 1982, however, other seabird species

R. L. Swann, 14 St.

Buzzard’s interaction with stoat

On 14 February 1994 at a floodwater pool in
a field at Cullerlie, near Echt, Grampian we
watched a Buzzard Buteo buteo feeding ona
dead young Herring Gull Larus argentatus.
As the Buzzard fed, a stoat Mustela erminea
appeared a few metres away. The Buzzard
flew 5m to afence post. Within a few seconds
the stoat appeared under the fence, directly

below the Buzzard. The stoat ran around the

on Canna have shown declines in numbers
and breeding success. Kittiwakes Alissa
tridactyla have steadily declined since 1982,
Shags Phalacrocorax aristotelis have
undergone a major decline in numbers and
breeding success since 1984, whilst Guillemot
Uria aalge and Razorbill Alca torda numbers
peaked in 1983 and 1985 respectively and
have since shown slight declines (Swann
1993, UNCC Report No. 181). The causes of
these declines are unknown buta reduction in
the availability of sandeels has been
suggested as a contributory factor in the
decline of the breeding success of Shags
(Swann et al. 1994 Seabird 16: 50-56) and
may also have had an effect on shearwater
breeding success.

Acknowledgements

Many thanks to everyone who helped gather
this data over a 19 year period. |am indebted
to Dr and Mrs J L Campbell and the National
Trust for Scotland for giving permission to
conduct our seabird studies on the island.
The work done on Canna is currently
supported by the JNCC Seabird Monitoring
Programme.

Vincent Road, Tain, Ross-shire 1V19 1JR

foot of the post, jumping up at the Buzzard
from time to time, at one point almost landing
on the fence. For about 10 mins the Buzzard
flew from post to post while the stoat ran after
itand continued to leap up at it now and again.

The Buzzard flew to the edge of the floodwater
about 15m from the fence. Again, within 30
secs the stoat reappeared 7-8m from the

58 Short Notes

SB 18(1)

Buzzard. The stoat walked, then began to run
straight for the Buzzard until it was within
0.5m of it before retreating. This behaviour
continued for about 1 min. Oniits final run, the
stoat went right up to the Buzzard’s legs and
actually nudged the bird. The Buzzard flapped
its wings and seemed almost to lose its
balance. The stoat then retreated about 1m
and began rolling onto its back. This behaviour
was followed by it runningin very small circles,
somersaulting, twisting and pulling itself along
on its back. The Buzzard appeared to be
watching it and moved a little closer.

King (1989. The Natural History of Weasels
and Stoats. Helm) refers to an observer, J. H.
Lawton, who watched almost identical
behaviour of a stoat ‘dancing’ in front of
Moorhens Gallinula chloropus where it
occasionally attempted to catch them.
‘Dancing’ stoats are well documented (Corbet
& Southern 1964. Handbook of British
Mammals). \t is still debated whether it is a
deliberate hunting technique or an involuntary
response to parasitic worms inside the skull
(King 1989).

It seems possible in this case that the stoat
was ‘mobbing’ the larger predator, or may
even have been hunting it. R. C. Dickson |
(1980 Scott. Birds 11: 119) observed an otter
Lutra lutra apparently attacking a Hen Harrier
Circus cyaneus on a fence post. It seems
unlikely that a stoat would attack prey as large
as a Buzzard but ‘in relation to their size, —
stoats and weasels kill larger prey than do
any other British carnivores’ (Hewson &
Healing 1971. The Stoatandits prey. J. Zool. |
Lond. 164: 239-244). Also Anderson (1966. |
The Countryman 122) saw a Buzzard swoop, |
pick up a weasel and fly off with it, but the
Buzzard fell to the ground, dead, covered with |
blood and with the weasel gripping its breast.
Svendsen (1982. The Wild Mammals of
North America) reports a dead eagle, which |
was found to have a bleached weasel skull
fixed to its neck.

In our case, the Buzzard flew back to the —
fence. The stoat also returned to the fence,
once again below the Buzzard. After a few
minutes the stoat disappeared and was not
seen again, although the Buzzard was still |
present when we left 15 min. later. |

Judy M. Duncan, 86 Broadfold Drive, Bridge of Don,Aberdeen AB23 8PP
Gillian F. A. Cooper, 14 Abbotshall Gardens, Cults, Aberdeen AB1 9LA

Spread-wing posture by a juvenile Peregrine Falcon in winter

On 2 February 1992 from 1136 to 1208hrs, a
dry day of weak sun and a moderate to fresh
wind, | saw a juvenile Peregrine Falcon Falco
peregrinus on an exposed fence post facing
into the wind in farmland on low ground in
west Galloway. On and off for over 15 min, it
sat with its wings extended at right angles

(like the spread-wing posture of a Cormorant
Phalacrocorax carbo), occasionally preened
and twice shook its feathers. During this time
it twice jumped down into rough grass and lay
down with its wings spread-eagled, facing
into the weak sun, for short periods. _ It
eventually flew away.

1995

On 9 February 1992 at 1344hrs, after heavy
rain in the morning and sunny spells in the
afternoon, | again saw a juvenile Peregrine
Falcon (probably the same bird) sitting on the
| same fence posts for over an hour andcarrying
out similar activities to those seen a week
‘earlier. It again jumped down into rough
| grass andtwice lay down with its wings spread-

eagled, facing the sun, and then sat with its
wings fully extended on a fence post. At
1450hrs it chased a flock of Sky Larks Alauda
arvensis.

Feather care used by Peregrines in winter
includes water bathing (Baker 1967. The
Peregrine Falcon. Collins, London), preening,
shaking and wing stretching but the above
behaviour is probably seldom seen or

Magpie’s nest - with a difference

The nest of a Magpie Pica pica, pictured inthe
colour plate was recently donated to Glasgow
Museums. It had been removed from about
15-20 feet up ina Cupressus tree, which was
being partially cut down in a garden in
Rutherglen, Glasgow. The cup of the nest
| was like a normal Magpie’s nest, made from
twigs, bound together with mud and lined with
fine roots. Underneath the nest, however, the
‘Magpies had constructed a platform mainly

;
|
\

Short Notes 59

recordedin the wild. Some falcons will spread
their wings fully while sunning in the prone
position on the ground but apparently only in
captivity (Simmons K.E.L. 1986. The Sunning
Behaviour of Birds. Bristol). Simmons (in litt.)
points out that interpretation of such incidents
is always difficult but that the birds could have
been ‘sun-drying’ and ‘sun-basking’ probably
after a bath on one or both occasions. The
Sparrowhawk Accipiternisus uses a standing
loose-spread wing posture for drying and
sunning after a bath and this posture has also
been seen in sunning Peregrines (see
Simmons for details ) butthe full-spread open
wing standing posture (See above) has not
been recorded in the Peregrine although it
has been seen in wild Eleonora’s Falcon
Falco eleonorae (Simmons 1986.)

R.C. Dickson, Lismore, New Luce, Newton Stewart DG8 OAJ

using S-shaped wire wall ties from a nearby
building site. Approximately 100 ties were
used, together with other miscellaneous
pieces of wire and even an old aerial from a
car! | always thought that Magpies collected
shiny objects and put them in their nest. This
is the first time | have heard of them actually
making their nest out of the objects which
they have stolen.

Richard Sutcliffe, Curator, Science Department, Glasgow Museums

60 Short Notes

Wheatears’ mountain top roost

The wide plateau on top of Ben Avon, Britain’s
16th highest mountain, in the eastern
Ciarngorms, is scattered with huge tors of
rock, the largest protruding some 20m
upwards out of the summit plateau. The tors
are formed of huge masses of granite with
linear cracks weathering into them, some
forming deep fissures.

It is pleasing to note these tors are of use to
birds. On 17 July 1994 | passed Clach
Choutsaich, a prominent tor at 1110m asl,
just prior to dusk. A group of seven or eight
Wheatears Oenanthe oenanthe were
perching about the tor and sitting in the mouths
of certain deep horizontal crevices, calling
perhaps to each other or perhaps in mild
alarm to me. | passed back again at 2215h,
now in very dim light. The Wheatears were

SB 18(1)

out of sight, but a soft call or two revealed that
they were deeper within the tor’s fissures,
settling inside to roost. Whether this was a
family group that may also have nested in the
tor is now known. During the day, Wheatears
were common on the stony parts of the summit
plateau.

Roosting is an aspect of bird behaviour often —
given scant treatment in species accounts. |
There is no mention of birds using these tors _
by Seton Gordon (1939, The Cairngorm Hills |
of Scotland, Cassell, London) nor by |
Nethersole-Thompson & Watson (1981, The |
Cairngorms, Melven Press, Perth). BWPVol |
5 (1988) does not detail roosting but records |
that outside Britain, Wheatears inhabit |
mountain-top plateaux up to 3000minEurope, |
above 3000m in Iran, and to 3500m in the

former USSR. |

Harry E. M. Dott, 70 Findhorn Place, Edinburgh EH9 2NW

61 Scottish Birds (1995) 18: 61-63

items of Scottish interest

SB 18(1)

Most of the following papers and reports on birds in Scotland are available in the Waterston
Library at 21 Regent Terrace for reference, and include all that have come to notice in the

period October 1994 to February 1995. The librarian would be glad to learn of anything that

has been missed, and to receive reprints or copies of papers on any aspect of ornithology or
natural history. Bird reports marked with an asterisk are available from the SOC at the prices
quoted, but please add 50p per order for postage and packing.

Scientific papers

Baines, D. 1993. Recent trends in Black
Grouse Tetrao tetrixnumbers. Game
Conservancy Ann. Rev. no 24: 133-
135.

Barton, T.R., Barton, C., Webb, A. & Carter,
1.C. (1994). Seabird distribution in
inshore waters of the western United
Kingdom between Wick and St David’s
Head from aerial surveys in 1987-
1991. Joint Nature Conserv. Cttee
Rep. 183: 53pp.

Catt, D. & Baker, P. 1993. The distribution
and estimation of Capercaillie Tetrao
urogallus numbers. Game Conserv.
Ann. Rev. no 24: 136.

Cresswell, W. 1994. Age-dependent choice
of Redshank Tringa totanus feeding
location: Profitability or risk? J. Anim.
Ecol. 63: 589-600. A study of attacks
by Sparrowhawks Accipiter nisus.

Cresswell, W. 1994. Song as a pursuit-
deterrent signal, and its occurrence
relative to other anti-predation
behaviours of Skylark A/auda arvensis
on attack by Merlins Falco
columbarius. Behav. Ecol. Sociobiol.
34: 217-223. A study in Scotland.

Cresswell, W. & Whitfield, D.P. 1994. The
effects of raptor predation on wintering
wader populations at the Tyninghame
estuary, south-east Scotland. /bis
2106: 223-232.

Daunt, F.H.J. 1994. The use ofa geographical

information system to study the effects
of forestry practice on Golden Eagles
Aguila chrysaetos reproductive
success. J. Raptor Res. 28: 55,
abstract only.

Dickson, R.C. (1995). Great Northern
Divers Gavia immer in Wigtownshire.
Trans. Dumfries & Gall. N.H. & Antiq.
Soc. 67: 1992: 78-79. Records cover
the period 1978-91.

Donald, P.F. & Evans, A.D. 1994. Habitat
selection by Corn Buntings Miliaria
calandra in winter. Bird Study 41:
199-210. Results from a BTO survey
in winter 1992-93 which included
Scotland.

Fox, A.D. & Bell, M.C. 1994. Breeding bird
communities and environmental
variable correlates of Scottish peatland
wetlands. Aquatic birds in the trophic
web of lakes. Proceedings of a
symposium held in Sacville, New
Brunswick, Canada in August 1991.
J.J. Kerekes (ed), Hydrobiologia 279/
280: 297-307.

Fox, A.D., Mitchell, C., Stewart, A., Fletcher,
J.D., Turner, J.N.V., Boyd, H.,
Shimmings, P., Salmon, D.G., Haines,
W.G. & Tomlinson, C. 1994. Winter
movements and site-fidelity of Pink-
footed Geese Anser brachyrhynchus
ringed in Britain, with particular
emphasis on those marked in
Lancashire. Bird Study 41: 221-234.

Francis, J., Stewart, B. & Duncan R. 1994.

62 Scottish Birds (1995) 17: 61-63

Wildfowl at Loirston Loch, Grampian
1960-93. North-east Scotland Bird
Report 1993: 75-79.

Green, R.E. & Williams, G. 1994. The
ecology of the Corncrake Crex crex
and action for its conservation in Britain
and Ireland. In Bignal, E. & Curtis,
D.J. (eds). Proc. 3rd euro. Forum on
Birds and Pastoralism. Scottish
Chough Study Group, Paisley,
Scotland.

Harris, M.P. 1994. Isle of May seabird studies
in 1994. Joint Nature Conserv. Cttee
Rep. 212: 30pp.

Harris, M.P., Buckland, S.T., Russell, S.M. &
Wanless, S. 1994. Post fledging
survival to breeding age of Shags
Phalacrocorax aristotelis in relation to
year, date of fledging and brood size.
J. Avian Biol. 25: 268-274.

Harris, M.P., Buckland, S.T., Russell, S.M. &
Wanless, S. 1994. Year and age-
related variation in the survival of adult
European Shags Phalacrocorax
aristotelis over a 24-year period.
Condor96: 600-605. Based on studies
on the Isle of May.

Harris, M.P., Halley, D.J. & Swann, R.L. 1994.
Age of first breeding in Common
Murres Uria aalge. Auk 111:207-209.
A study on the Common Guillemot in
Scotland. Hartley, I.R. & Shepherd,
M. 1994. Nesting success in relation
to timing of breeding in the Corn
Bunting Miliaria calandraon North Uist.
Ardea 82: 173-184.

Hartley, |.R. & Shepherd, M. 1994. Female
reproductive success, provisioning of
nestlings and polygyny in Corn
Buntings Miliaria calandra.

Heppleston, P.B. 1994. The nesting of the
Curlew Numenius arquata in Britain:
An analysis of records 1946-1985.
Scot. Nat. 106: 39-57. Based on BTO

SB18 (1)

nests records cards, the majority from
Scotland and the north of England.

Hewson, R. & Talbot, T.R. 1994. Factors
affecting nesting success of Hooded
Crows Corvus corone cornix in west
and north-west Scotland. Scot. Nat.
106: 31-38.

Holling, M. & McGarry, O. 1994. A review of ©
the changing status of Buzzards Buteo
buteo in the Scottish Borders. Border
Bird Report no. 14: 76-82. |

Innes, M.!. 1994. The movement of gulls and
terns off Peterhead, Grampian. North-
east Scotland Bird Report 1993: 64-
Tihs

Jenks, P., Stammers, B., Leitch, A.J. & Lewis,
T. (1993). Fair |sle seabird monitoring
scheme: Report to the Joint Nature
Conservancy Committee of the 8th
season's work in 1993. Joint Nature
Conserv. Cttee Rep. 211: 54pp.

Jenson, H., Tasker, M.L., Coull, K. & Emslie,
D. (1994). Acomparison of distribution
of seabirds and prey fish stocks in the —
North Sea and adjacent areas. Joint
Nature Conserv. Cttee Rep.207:11pp.

Knowler, J.T. 1995. Barrow’s Goldeneye
Bucephala islandica in Strathclyde:
new to Britain and Ireland. Brit. Birds
88: 104-106. A 1979 record now
accepted for the British list.

McGrady, M.J. & Grant, J.R. 1994. Amodified
power snare to catch breeding Golden
Eagles Aquila chrysaetos. J. Raptor
Res. 28: 61, abstract only. Based on
work in Argyll.

Marquiss, M. & Carss, D.N. 1994. Fish-
eating birds: Assessing their impact
on freshwater fisheries. /nst. Terr.
Ecol. R. & D. Rep. 15: 19pp.

Marquiss, M. & Duncan, K. 1994: Diurnal
activity patterns of Goosanders
Mergus merganseron a Scottish river
system. Wildfowl 45: 290-221.

fa20

Marquiss, M. & Duncan, K. 1994. Seasonal
switching between habitats and
changes in abundance of Goosanders
Mergus merganser within a Scottish
river system. Wildfow! 45: 198-208.

Marquiss, M. & Rae, R. 1994. Seasonal
trends in abundance, diet and breeding
of Common Crossbills Loxia
curvirostrain an area of mixed species
conifer plantation following the 1990
Crossbillirruption. Forestry67: 31-47.

Mitchell, J. 1994. The birds of the Endrick
Mouth, Loch Lomond: An update of
the Annotated Check-list up to 1
January 1990. Scot. Nat. 106: 31-38.

Moss, R., Watson, A., Trenholm, I.B. & Parr,
R. 1993. Caecal threadworms
Trichostrongylus tenuisin Red Grouse
Lagopus lagopus scoticus: Effects of
weather and host density upon
estimated worm burdens. Parasitology
107: 199-209.

Murray, S. & Wanless, S. (1992). A count of
the Noss gannetry in 1992, and
analysis of Gannet monitoring plots
on Noss National Nature Reserve
1975-91. Joint Nature Conserv. Cttee
Rep. 50: 54pp.

Murray, S. & Wanless, S. 1994. Census of
the St Kilda Gannetry in May 1994.
Unpublished report to SOC 26pp.

Murray, S., Wanless, S. & Harris, M.P. 1994.
The effects of fixed salmon nets on
Guillemot Uria aalge and Razorbill
Alca torda in north-east Scotland in
1992. Biol. Conserv. 70: 251-256.

Newton, |. 1994. European Sparrowhawks
Accipiter nisus in conifer plantations.
J. Raptor Res. 28: 48. Abstract only.
A study in Scotland.

Okill, J.D. 1994. Ringing recoveries of Red-
throated Divers Gavia stellatain Britain
and Ireland. Ringing & Migration 15:
107-118. A study of birds ringed in
Orkney and Shetland.

Owens, |.P.F., Burke, T. & Thompson, D.B.A.
1994. Extraordinary sex roles in the
Eurasian Dotterel Eudromas

Items of Scottish Interest 63

morinellus: Female moulting areas,
female-female competition, and
females mate choice. Amer. Nat. 144:
76-100. A study in Scotland.

Riddiford, N. 1995. The Common Gulls
Larus canussharing anest. Brit. Birds
88: 112-113. Occurred on Fair Isle in
the period 1981-88.

Rushton, S.P., Hill, D. & Carter, S.P. 1994.
The abundance of river corridor birds
in relation to their habitats: a modelling
approach. J. Appl. Ecol. 31: 313-328.
Based onthe analysis of the Waterway
Bird Survey data, which included
several rivers in central and southern
Scotland.

Smith, R.D. 1994. Snow Buntings
Plectrophenax nivalis breeding in the
Cairngorms: population dynamics and
the influence of human recreation.
SNH Rev. no 1: 65pp.

Smith, R.D. & Metcalf, N.B. 1994. Age, sex
and prior site experience have
independent effects on the foraging
success of wintering Snow Buntings
Plectrophenax nivalis. Behaviour 129:
99-111. Astudy on Cairn Gorm, north-
east Scotland.

Stewart, F.M., Thompson, D.R., Furness,
R.W.&Harrison,N. 1994. Seasonal
variation in heavy metal levels in
tissues of Common Guillemots Uria
aalgefrom north-west Scotland. Arch.
Envir. Contam. Toxicol. 27: 168-175.

Stone, C.J., Webb, A. & Tasker, M.L. 1994.
The distribution of Manx Shearwaters
Puffinus puffinus in north-west
European waters. Bird Study 41: 170-
180.

Swann, R.L. 1994. Canna seabird studies
1994. Joint Nature Conserv. Cttee
Rep. 213: 8pp.

Thompson, D.L. & Douglas-Home, H. 1993.
(1994). Longterm changes in weather
and in the breeding schedule of
Common Swifts Apus apus. Avocetta.

William G. Harper

64 Advice to Contributors

SB 18(1)

Advice to Contributors

Authors should bear in mind that only a small
proportion of the Scottish Birds readership is
science-trained, and should aim to present their
material concisely, interestingly and clearly.
Unfamiliar technical terms and symbols should be
avoided wherever possible and if deemed essential
should be explained. Supporting statistics should
be kept toa minimum. All papers and Short Notes
are accepted on the understanding that they have
not been offered for publication elsewhere and that
they will be subject to editing. Papers will be
acknowledged on receipt and will be reviewed by at
least two members of the editorial panel and in
some cases also by an independent referee before
being accepted. They will normally be published in
order of acceptance of fully revised manuscripts.
The editors will be happy to advise authors on the
preparation of papers.

Reference should be made to recent issues of
Scottish Birds for guidance on style of presentation,
use of capitals, form of references, etc. Papers
should be typed on one side of the paper only,
double-spaced and with wide margins; two
copies are required and the author should also

Errata:- Snow Buntings: Vol 17.4
p 223-234

retain one. Headings should NOT be underlined,
nor typed entirely in capitals. Scientific names in
italics should follow the first text reference to each

species and should follow Voous ‘List of Recent
Holarctic Bird Species’ as given in the The British —

Birds’ List of Birds of the Western Palearctic (1984).
Only single quotation marks should be used
throughout and numbers one to ten should be
written out whereas 11 and above should be written

AUGUSt TOOT cccsccuaare ct but on the 5th (ifthe name of
the month does not follow).
papers shorter than 700 words will be treated as
Short Notes where all references should be
incorporated into the text, and not listed at the end,
as in full articles.

Tables, maps and diagrams should be designed to
fit either a single column or the full page width.
Tables should be self-explanatory and headings
should be keptas simple as possible, with footnotes
used to provide extra details where necessary.
Each table should be on a separate sheet. Maps
and diagrams should be in Indian ink and be
camera ready, but drawn so as to permit reduction
to half their original size.

For details of writing Research Progress Reports,
please contact the editor in advance.

Figure 1. Estimated first egg dates (days from 1st May) of all Snow Bunting nests and broods found during
1987-1993, arranged by 5-day periods (10=days 8-12, 15=13-17, etc.). Hatched bars represent Area A,
solid bars Area B.

Figure 2. Estimated first egg dates (days from 1st May, to nearest 5-day interval) of first brood Snow
Bunting nests found on Area A in each of the six summers 1988-1993. Hatched bars show nests which
were ultimately thought to be successful, filled bars those those which failed. The arrows indicate the
occurrence of heavy snowfalls (more than 5cm or giving 95+% snow cover) and the figure below is the
number of days for which 50% cover persisted.

Figure 3. The annual percentage of female Snow Buntings which, having reared one brood successfuly,
went on to rear a second brood successfully, against the mean first egg date (days from 1st May) of
successful first broods.

Figure 4. The frequency of total contents (estimated clutch size: eggs plus young) of 115 Snow Bunting
nests found with eggs or small young (filled) and 76 nests with 5 to 9-day-old young (hatched).

Figure 5. The frequency distribution of brood sizes in Snow Bunting nests when the young were ringed
(mostly 6-8 days old) in first (hatched bars; n=127) and second broods (filled bars; n=77).

Please note that

NEOTROPICAL
BIRD CLUB

Neotropical bird club launched

A club has been launched to promote the study and conservation of
the birds of the Neotropics (South America, Central America and the
Caribbean). It is currently seeking founder members to help reach
the launch budget of £2000, which is required to get the club running
and to publish the two first issues of its intended journal ‘Continga’.
Founder members will be asked to pay a minimum of £25, and will
be formally acknowledged in the first issue of ‘Continga’. ‘Continga’
will provide a colourful and much needed forum for exchange of
information on the avifauna of this extremely rich and diverse area,
and will contain papers and features on the birds and their
conservation as well as news of recent observations and discoveries
(at present, new species are still being discovered at the rate of
more than two a year). It is hoped that in due course the club will be
able to provide direct funding and support for practical conservation
programmes.

For further details and membership forms,
please contact:

Rob Williams,

Publicity Officer,

Neotropical Bird Club,

c/o The Lodge,

Sandy,

Bedfordshire SG19 2DL

Scottish Birds

Volume 18 Part 1 June 1995

Contents

Main articles
The status of the Eider on the Isle of May and other Forth Islands.

J. Calladine, M.P. Harris, S. Taylor & S. Wanless 1
Lek habitats of Capercaillie at Abernethy Forest, Speyside.

R.W. Summers, R. Proctor & S.I.H. Shah 11
Nest reliefs and feeding rates of Merlins. R.C. Dickson 20

The Status and distribution of wintering Pink-footed and Greylag
Geese in east central Scotland. M.V. Bell & S.F. Newton

Short Notes
Diurnal and tidal activity patterns of Purple Sandpipers on the
Isle of May, Fife. 51

Little Ringed Plover attempt breeding in Dumfriesshire. R.Mearns 55

Numbers and breeding success of Manx Shearwaters on the
Isle of Canna, 1973-94. R.L. Swann 56

Buzzard’s interaction with stoat. J.M. Duncan & G.F.A. Cooper 5;

Spread-wing posture by a juvenile Peregrine Falcon in winter.
R.C. Dickson 58

Magpie’s nest — with a difference. R. Sutcliffe 58)
Wheatear’s mountain top roost. H.E.M. Dott

Items of Scottish Interest. W.G. Harper
Advice to Contributors

Erratum

Published by the Scottish Ornithologists’ Club,
21 Regent Terrace, Edinburgh EH7 5BT. © 1992

Printed by Milne, Tannahill and Methven Ltd., Perth

See Re ee ee

Scottish Birds

The Journal of the Scottish Ornithologists’ Club

Editor: Dr. S. da Prato ;

Assisted by: Professor D. Jenkins, Dr J.B. Nelson and Dr. |. Bainbridge
Business Editor: The Secretary, S.O.C., 21 Regent Terrace, Edinburgh EH7 5BT
(tel. 0131-556 6042).

Scottish Birds, the official journal of the Scottish Ornithologists’ Club, publishes original material
relating to ornithology in Scotland. Papers and notes should be sent to The Editor, Scottish Birds, 21
Regent Terrace, Edinburgh EH7 5BT.

Two issues of Scottish Birds are published each year, in June and in December. Scottish Birds is
issued free to members of the Scottish Ornithologists’ Club, who also receive the quarterly
newsletter Scottish Bird News and the annual Scottish Bird Report. These are available to non-
members at a subscription rate (1992) of £30.

The Scottish Ornithologists’ Club was formed in 1936 to encourage all aspects of ornithology in
Scotland. It has local branches which meet in Aberdeen, Ayr, the Borders, Dumfries, Dundee,
Edinburgh, Glasgow, Inverness, New Galloway, Orkney, St Andrews, Stirling, Stranraer and Thurso,
each with its own programme of field meetings and winter lectures. The Waterston Library at the
Club’s headquarters at 21 Regent Terrace, Edinburgh EH7 5BT is the most comprehensive
ornithological library in Scotland, and is available for reference during office hours (Monday to
Friday, 9.00 - 17.00 hrs). (But please phone beforehand). A comprehensive stock of Scottish local
bird reports is held at headquarters and may be purchased by mail order.

SOC annual membership subscription rates

AGUIIt siccnsaadennnosvtascoadsiieeeattiat as ocate meses terke unten dete beweuntedaeeten Siete ee £15.00
Family (husband & wife, & children UNGeEr 18)) -c.c.c.cccenccceceeneee@enc cece eetseeeee eee ee eee £22.00
Junior (under 18, or student Under 25).........cc..cckscctneresccteaneccesswewes saree £6.00
Pe@nsioner/UNnWaged iiivsicnicesacencceceneeete cece solavenes scemeaaecenunesaeene oonekGlesaneeen eee or £8.50
PENSIONEr FaMily .:...scceccccsoccnneeccpheueetetaerene sash eentecsesseereennenetsacsetemeetscees tt ee ee £12.00
[SIRE scoot tisaciet)dectazdde acenr cee damnepaeaadoecede hed aceeverseeatceteminnctar (Ue eee eee et £300.00
Life Fanny sce sic on saeadeeites essus etecwahaeeten megs scbeuedeeraatbeenbmadnensignucteseeeges to eee £450.00

All subscriptions may be paid by Banker’s Order and Covenanted. Subscriptions paid this
way will greatly assist the Club.

Published by the Scottish Ornithologists’ Club, 21 Regent Terrace, Edinburgh EH7 SBT.
Printed by Milne, Tannahill & Methven Ltd., 113-119 Glover Street, Perth PH2 QJE.

Scottish Birds (1995) 18: 65-70

65

Occupation patterns of lochs by Slavonian Grebes in Scotland

RW SUMMERS & R A MAVOR

The median arrival date of half the number of Slavonian Grebes on breeding
lochs in Scotland was 4 April in 1993, 2 months before the mean date of
laying of the first eggs. Lochs at higher altitudes were occupied later than
those at lower altitudes. Most lochs (61%) held only one pair and the
largest number (15 pairs) was at Loch Ruthven. Three lochs held 38% of the
Scottish population. Loch areas varied from 0.3 to 376 ha (median 4 ha).
The number of breeding pairs on a loch was correlated with the area of
dense Bottle Sedge, the main breeding habitat. At some lochs, birds
departed early in the season without nesting. On average, the last
fledglings left at the same time as the last adults but some left a month

before and others a month after the last adults. The medi,
departure of half the adults were 2 August 1992 and aH)
latest (19 October 1992) was from the main moulting site.

Introduction

The Slavonian Grebe Podiceps auritus is one
of Britain’s rarest breedings birds (Batten et
al. 1990). It colonised Scotland in 1909
(McGhie 1994) and since then has bred in
Highland, Grampian and Tayside Regions,
usually nesting on moderately sized lochs in
beds of Bottle Sedge Carex rostrata (Thom
1986). This study describes the timing of
arrival and departure from different lochs and
relates the number of grebes on the different
lochs to loch size and the amount of breeding
habitat.

Study area and methods

From the 40 lochs used by Slavonian Grebes
in1992 and 1993, 25 were selected for detailed
observations. The subset comprised lochs
from all parts of the current breeding range:
Strathspey, Morayshire and north and south

} Shane

8 1996)
LIBRARIES

of the Great Glen, including a range of altitudes
so that lowland lochs could be compared with
upland lochs. The subset also held the bulk
of the known Scottish breeding population,
65 out of 73 breeding pairs in 1993.

The fieldwork was carried out from 1 May to
31 October 1992 and 1 April to 31 October
1993. Study lochs were visited once or twice
a week in order the count the grebes, using
the methods of Crooke etal. (1993). Then, in
order to time arrival and departure, smoothed
lines were drawn through the series of counts
using the distance weighted least squares
method and the dates on which half the
number of birds had arrived at, or departed
from, a given loch were determined from the
smoothed line. Asimilar method could not be
used for fledglings since the loss of chicks
from the lochs was partly due to mortality as
well as emigration. Therefore, to get
comparable departure dates for fledglings

66 RW Summers & R A Mavor

SB 18 (2)

and adults, the dates on which the last
fledglings and adults left were taken as the
mid point of the period between the day of last
observation and the first day when no birds
were seen.

Loch sizes were measured by overlaying a
grid, divided into scaled 0.25 ha squares,
over a 1:25000 map of the lochs and counting
squares which contained over 50% water.
Altitudes of lochs were also taken from
1:25000 O.S. maps. The areas of sedge
beds were estimated from their lengths and
breadths, which were paced inthe field. Dense
sedge beds were those where the amount of
water visible when viewing the bed at 30°
(measured with a clinometer) was estimated
at less than 50%.

For the sake of confidentiality, the lochs other
than Loch Ruthven, which is a well-known
RSPB reserve, have been given numbers
ratherthannames. Thenames corresponding
to these numbers are held at the RSPB and
Scottish Natural Heritage offices at Inverness.

Results

The study did not start until 1 April in 1993 so
the arrivals which took place in March were
not timed. However, the median date at
which half the number had arrived at a given
loch could be calculated, at 4 April, and the
latest date was 28 April. The altitude of the
breeding lochs ranged from 17m to 739m
(Table 1). The median altitude of those lochs
with arrival dates before 1 April was 198m
(quartiles 40-215m, n=9), whilst those with
arrival dates after 1 April was 247.5m (quartiles
213-390m, n=14). This difference was
significant (Mann-Whitney U=25, P=0.016)
showing that lochs at higher altitudes were
occupied later than those at lower altitudes.

Some of the lochs showed peaks in spring
numbers. At Loch Ruthven, the population
peaked at 39 birds in early May 1993, before
falling to the breeding population of 15 pairs
(Fig. 1). At loch 43, where there was no
breeding in 1993, the spring peak was 12

Table 1. Characteristics of breeding lochs of Slavonian Grebes in 1993.
Altitudes and areas were taken from all breeding lochs whilst sedge bed

details refer only to study lochs.

n
Altitude (m) 33
Area (ha) SI)

Area of dense sedge (m2) PAS

Number of sedge beds 25

Median Quartiles Range
225 65-325 17-739
4 1.3-9 0.3-376
550 104-825 0-8706
6 2-8 0-13

1995

birds. Lochs 35 and 36 also had spring peaks
(Fig. 1). Thus, the grebes initially occupied
certain lochs preferentially before dispersing
to other lochs for breeding.

Table 2. The number of lochs with
different numbers of Slavonian
Grebes in 1993

Number of lochs Number of pairs

20 1
5 2
5 3
1 6
1 uh
1 15

The number of pairs occupying each loch
varied from 1 to 15 pairs in 1993 (Table 2).
Loch Ruthven held the largest number (15
pairs) and the three main lochs held 38% of
the entire Scottish population. Loch sizes
varied from 0.3 ha to 376 ha; the median
value was 4 ha (Table 1). The smallest loch
which held more than one pair was 4.9 ha.
Generally, the larger lochs held more pairs
but there was no significant correlation
between population size and the loch area
(r'g4=0.30, ns). However, there was a
correlation between the population size in
1993 and the amount of dense sedge beds
(fo3=0.82, P<0.001). On those lochs which
had more than one pair of grebes, there was
usually only one pair of grebes nesting in
each sedge bed. However, the sharing of

Slavonian Grebes in Scotland 67

sedge beds took place at a number of lochs
and the closest distance between two
occupied nests was 2m.

The most simple pattern of occupancy of
lochs was observed at those lochs where
there was only one pair. Here, the birds
arrived, bred and departed either together or
separately (Fig. 1, loch 5). However, onthose
lochs with more than one pair, there were
sometimes departures or arrivals of additional
pairs, some of whom would attempt to nest
(Fig. 1, loch 65).

Several lochs showed an autumn peak in
grebe numbers. This was particularly
noticeable at loch 35 where there was a large
moulting group (Fig. 1). In 1992, the maximum
number was 46 birds, while in 1993 the
maximum was 44. Four moulting birds also
occurred on loch 43 which had no breeding
birds. Itis also likely that some birds moulted
at their breeding lochs, given that they were
present during September (Fig. 1, loch 36).

The first birds to depart from a loch did so on
26 May 1993, from loch 37 where there was
no breeding attempt. The median departure
dates for half the adults from breeding lochs
were 2 August 1992 (quartiles 17 July-20
August) and 29 July 1993 (quartiles 10 July-
25 August), whilstthe median departure dates
for the last adults were 15 August 1992
(quartiles 30 July-15 September) and 14
August 1993 (quartiles 24 July-15 September).
The median departure dates for the last
fledglings were 14 August 1992 (quartiles 8
August-22 August) and 15 August 1993
(quartiles 4 August-12 September), andthese
dates were strongly correlated with the
departure dates of the last adults (r7=0.92,
P<0.001 for 1992 and rg=0.87, P= 0.001 for
1993). On average, chicks departed close to
the time of the adults, but the range of chick

68 RW Summers & R A Mavor SB18(2)

Figure 1 Numbers of Slavonian Grebes on selected lochs in 1993
Circles - adults, triangles - chicks

4 4 46 45
Loch 35 2 §
36
3 3
27
2 2
18
1 1
9
fe) 9) O fe)
| Apr | May | Jun | Jul | Aug | Sep | Oct | Apr | May | Jun | Jul | Aug | Sep | Oct

10
Loch 65

Number

R fe)
| Apr | May | Jun | Jul | Aug | Sep | Oct | Apr | May | Jun | Jul | Aug | Sep | Oct

Loch 43 Loch Ruthven
12 ® 12 ie)

0 0 6) :
| Apr | May | Jun | Jul | Aug | Sep | Oct | Apr | May | Jun | Jul | Aug | Sep | Oct

1995

departures was from 34 days before, to 29
days after, the last adult. In the latter case,
the chick fledged despite the long period of
desertion. The latest date of departure was
on 19 October 1992 at the main moulting site
(loch 35). The pattern of departure from lochs
was found to be unrelated to loch altitude or
the breeding success of the birds.

Discussion

In Norway and Iceland, Slavonian Grebes
have been found to arrive on the breeding
lochs between mid April and mid June. Most
of them were either paired, or mated soon
after arrival (Fjeldsa 1973a). These arrivals
were later than in Scotland where there is less
ice onlochs in spring. Arelationship between
occupancy and temperature is supported by
the observation that lochs at higher altitudes
were occupied after lowland lochs.
Presumably, feeding conditions are related to
temperature.

Although, on average, the lochs were occupied
by 4 April in 1993; laying did not start until 7
May 1992 and 4 May 1993, and the mean
date of first eggs was 5 June 1992 and 4 June
1993 (Summers et al. 1994). The
corresponding dates in Norway and Iceland
were 12 May (first egg) and 13 June (mean
first egg date) (Fjeldsa 1973a). Itis not known
why there is such a delay in nesting by
Slavonian Grebes, but it may be related to the
late growth of Bottle Sedge which provides
the main nesting habitat (Summers et al.
1994). However, early establishment of
territories is perhaps necessary to ensure a
breeding attempt.

The main variable determining the number of
Slavonian Grebes settling on a given loch
seems to be the amount of suitable nesting
habitat; beds of dense Bottle Sedge. Two of

Slavonian Grebes in Scotland 69

the lochs, lochs 35 and 43, appeared to be
suitable feeding sites for grebes in that they
had both spring peaks and moulting birds, but
few and no breeding birds during the summer,
respectively. The limited nesting habitat at
these lochs may be the factor restricting
breeding.

The pattern of departure from lochs was
variable and dependent partly on whether the
birds attempted to nest and whether there
was a moulting population at that site. Thus,
birds departed early if they did not attempt
nesting and birds were present as late as
October if moulting occurred.

The three main moulting sites, Loch Ruthven,
and lochs 43 and 35 are all large lochs (over
100 ha), have large Stickleback Gasterosteus
aculeatus populations (the main food of the
Slavonian Grebe, Fjeldsa 1973b) and clear
water (Summers et al. 1994). The latter two
qualities are related to the efficiency of
catching fish (Summers et al. 1994). so
perhaps these lochs are chosen because
they provide good feeding opportunities.

Acknowledgements

We are extremely grateful to Scottish Natural
Heritage who provided grants to help finance
the study. The following people helped with
data collection; G Allison, S Benn, K Clauss,
C Gervaise, M Harvey, B Knickmann, A
MacLennan, H Paget-Wilkes, G Prest, R
Proctor and L Winskill. L Wilson helped with
the analyses. Also, we are extremely grateful
to the landowners who gave permission to
work ontheirlochs. Drs M Avery, | Bainbridge
and N Buxton commented on the draft.

70 R W Summers & R A Mavor

SB18(2)

References

Batten LA, Bibby C J, Clement P, Elliott G D
& Porter R F 1990. Red Data Birds in Britain.
Poyser. London.

Crooke C, Dennis R, Harvey M& Summers R
1993. Population size and breeding success
of Slavonian Grebes in Scotland. Britain’s
Birds in 1990-1991: The conservation and
monitoring review. 135-138. BTO/JNCC.

Fjeldsa J 1973a. Territory and the regulation
of population density and recruitment in the
Horned Grebe Podiceps auritus arcticus Boje,
1822. Vidensk. Meddr dansk naturh. Foren.
136: 117-189.

FjeldsaJ 1973b. Feeding andhabitat selection
of the Horned Grebe, Podiceps auritus (Aves),
in the breeding season. Vidensk. Meddr
dansk naturh. Foren. 136: 57-95.

McGhie H A 1994. Discovery of the first
British clutch of Slavonian Grebe eggs in a
museum collection. Scottish Birds 17:166-
167.

Summers R W, Mavor R A & Hogg S 1994.
Factors affecting loch selection and breeding
success of Slavonian Grebes in Scotland.
RSPB report to SNH.

Thom V M 1986. Birds in Scotland. Poyser,
Calton.

RW Summers and RA Mavor,

Royal Society for the Protection of Birds,

Etive House, Beechwood Park, Inverness, 1V2 3BW.

Revised manuscript accepted April 1995

_.

Slavonian Grebe

«=
m.
a che

Ron Summers

Scottish Birds (1995) 18: 71-77

Fi

Winter diet of Ravens in Perthshire

W A MATTINGLEY

This paper records the analyses of 700 pellets from roosts of non territorial
Ravens in Glen Quaich, Perthshire, during the winters 1990-91, 1991-92 and
1992-93. Remains of lagomorphs were present in 94% of the pellets. Other
food items occurring less frequently and irregularly included birds, sheep,
small mammals, Rowan berries and moth cocoons.

Introduction

Ravens Corvus corax are scarce breeding
birds in eastern Scotland compared with the
west of the country (Gibbons ef al. 1993).
Most of the Perthshire breeding population of
c60 pairs occur in the area to the west and
south of the AQ road ( P Stirling-Aird, pers.
comm.). There has been a slight decline in
breeding pairs since the 1960’s but the
population has remained stable for the last 10
years, and maybe increasing. In addition to
the known breeding pairs, a roost of up to 80
Ravens was found in Glen Quaich, central
Perthshire, in 1990. It was occupied, at times
erratically, over 3 winters from October to
March. Roosts are assumed to hold mainly
non breeding pairs and young birds (Holyoak
& Ratcliffe 1968; Ratcliffe 1990; Goodwin
1986; Heinrich 1990) as territorial adults roost
at or near their nest sites throughout the year
(Coombes 1948; Dare 1986; Ratcliffe 1990;
Heinrich 1990).

The diet of Ravens in various parts of Britain
is well documented (Holyoak 1968; Marquiss
et al. 1978; Newton et a/. 1982; Marquiss &
Booth 1986; Ewins ef a/. 1986) except for
central Scotland where they are at the eastern
edge of their breeding range. The unexpected
find of a large number of roosting Ravens in
the area provided an opportunity to investigate
their diet by an analysis of pellets collected

from the roost.
Study area and methods

In winters 1990-91, 91-92, 92-93, 15 - 80
Ravens frequently roosted in a Birch Betula
sp wood centrally positioned in Glen Quaich.
A small mature conifer plantation 4km from
the birch wood at the lower end of the glen,
and crags at the upper end, 6km from the
birch wood, were also used as alternate roost
sites.

Cushing (1941 in Knight & Call 1974) observed
Ravens in California moving 60km daily from
their roost to feed. Stiehl tracked Ravens in
Oregon 40km from their roost (Heinrich 1990)
and in Alaska, radio tagged Ravens made a
100km round trip. In Britain, little is known
about distances travelled to and from roosts
for daily feeding. In Cornwall, Coombs (1978)
found an unusually large number of Ravens
in an area of c9km diameter within a roost of
upto 150 birds. Thus, the foraging area of this
Perthshire roost could potentially encompass
a vast area of the uplands of central Scotland
but without some form of marking, the
distances birds travel to feed remains
unknown. Nevertheless, many birds were
seen on the high ground of the area mainly
within 15km of the roost; few were seen on the
low ground.

72 WA Mattingley

SB18(2)

Within an arbitrarily selected radius of 20km
from the roosting glen, approximately 75% of
the area is upland above 200m and suitable
for foraging Ravens. Encompassed within
this area are Glens Lednock, and Almond,
Logiealmond, Glen Cochill, Upper Strathtay
and south Loch Tay. The predominant land
uses here are sheep farming, management
for Red Grouse Lagopus lagopus scoticus
and Red Deer Corvus elaphus, and coniferous
forestry. Large populations of Rabbits
Oryctolagus cuniculus and Mountain Hares
Lepus timidus exist, particularly in
Logiealmond, upper Glen Almond and Glen
Quaich.

Regular collections of pellets were made when
the roost was occupied over the 3 winters.
The pellets were dried and broken up and the
contents were identified using various guides
(Day 1965, Corbett & Southern 1977, Yalden
1977, Debrot 1982) andareference collection
of fur and feathers.

Results

A total of 700 pellets was examined (Table 1),
with fur.and/or bones of lagomorph the most
frequent component found in 94%. For each
sample of pellets analysed in the 3 winters
there was no significant difference between
the samples (x“=2.62; df = 11; P> 0.99)
(Table 2). Some of the lagomorph bones
could be identified to species and 31% of the
pellets contained Rabbit but only 6% of the
total showed signs of Hares.

The next most recorded constituent in 69% of
the pellets was very fine well digested
vegetable material presumed to be secondary
ingested from the guts and dung of mammals.
The proportion of pellets containing vegetable
matter varied from 32 to 95% among samples

but not consitently within any one year or
season (,°=32.81; df = 11; P< 0.001).

Where items occurred less frequently, some
samples were combined to test for variation
between seasons or years using the Chi-
square test. Thus, smaller pellet samples
from January 1991, February 1991, October
to December 1992 and January 1993 with
March 1993 were combined to provide 7
samples from different months/years for
comparison (Table 2). Stones and grit were
recorded in 16% of pellets overall but, in no
sample did they occur more or less than
expected (,°=7.71; df=6; 0.20<P<0.30).
Remains of sheep, voles, birds, Rowan berries
Sorbus acuparia, and moth cocoons were all
present but the frequency of occurrence
varied. At certain times, for each of these
groups, higher or lower frequencies were
found than expected.

In the large mammal category, sheep wool
was found in 10% of pellets (Table 1). In
October and November 1991 an unexpected
number (>20% for each month) contained
signs of sheep but the following year the
combined October to December samples had
<4%. Remains of the only other large
mammal, Red Deer Corvus elaphus was
present in <1% of pellets.

Of the small mammals, Short-tailed Field
Vole Microtus agrestis was the most
commonly found (9%). Samples from
February 1991 and January/March 1993 had
the highest occurrence of vole at 27% and
16% respectively. The only other rodent
identified was Wood Mouse Apodemus
sylvaticus. \Insectivores included Mole Talpa
europaea and Common Shrew Sorex
araneus; another 2% of pellets contained
signs of insectivores but these were not
identified to species level. ji

Winter diet of Ravens in Perthshire 73

1995

(ead
(@) oI
(1) 4 (ee
(a
(g) |
({i)) II (p) €
(ji)
({h)) (1)
(CG) (\l)) fh
(g) | (9) €
(al (€) |
(Lp) 02 (9) ze (28) Z (p) € (g) ¢ (ae
(2) I (eal (5) (A) (6) € (p) |
(yee (S) | (GS) | (EL) 6 (Ol) kL (bp) ob (8) awe
(\!)) Il
(91) € (yp)
(2) | (a) Ih (\i)) fh (py)
(9) §€ (al
(a) Ih (8) + (68) 92 (1S) 95 (OL) OL (Zk) € (OL) 8 (())
(@) (I 91) € (a) ee(Gaecic. (ONO as (8) (yp) €
(Gee @ (GS) p
((!)) fh ((\!)) 1!
(Gg) | (a) (g) -§ (Sy ac (6) |
(ial
(26) 2 (8) ¥ (A) fh (@) 1 (sg) | (€) 2 (9) Z (ae)irac: ae (ol ne (6) Z WA) ed
(61) OL (2k) 9 (8) ¥ (LL) 2 (GD ecierea aoa (AD) ae B)e tg (CL) OL (08) 6
(91) € (8) ¥ (Ge) Al (p) 2 (92) S (EL) 6 (GR ice SG) (9) g (f)
(€9) zt (28) Sp (ee) 91 (95) 82 =z) 9 (69) @ (25) 29 (G6) 246 (96) pe (IZ) 9S (LZ) tL
ND) (€L) Z (A) fh (8) ¥ (py) & (6) OF (9) 9 (Boe (py) §€ (6) ot
(aay a a(Gcecie 0c) 0b. (ORG (GS) 4 (Ov) 22 (Zp) 9p (Gz) G2 = (28) 8 (49) 16 (29) 0
(ec) OL (85) 08 (19) 06 (29) 16 (68) 2 (9h) Le = (2S) 1G = (G9) 99» (09) GL Ss (9S) be = (02) 9

61 es 6v 0s 61 L9 601 rai}! G¢ 62 oe
£6/€/9 C6/L/08 c6/2N/2b 26/Lb/pe 26/01/S2 L6/LL/ZE §=66/OL/9~—siN/2/peSs«é G/V2/E 16/1/02 = L6/L/9

sa (ies TS ler 9
1

66
06/21/2

SJPWLUELU ‘piuy

pn

S8u0g “piujy

SJIBY ADEI|

jeuayeW ‘pluy

V $]eQ
| Huly uolessey
4 spng
Z SaAea| WNIUIDIe/,
pb paas ulyjyeo youig
€ Sd9Yy9 Pilg
) 69 Sdlag UEMOY
bl Swa}s Apoom
62 9/}99g
G abequey
7 Jayjeo} UaAeY
9 ays 663
G 499Q pay
) SLL pulg
) OZ daays
LL dds a0aioasu|
0) p «Malus uowWwOg
iG 9/0l\
0) | 9SNO|| POOM
) 99 a]0/

9

LL) pd U00909 Y}O/\|
69) €8P sayew ba
9) Lb as
Le) Zb2 yqqey

1S) 86€ ydiowobey

"sisayjuased ul ase sabej]Uadtad “Wal poo sejnaizied yoea Hululejuog sjajjad Jo ssaquinu ay] Moys sainbi4
"ONYSYPId /e4JUII UI PajIajjOI sjajjad UBABY OOZ JO UOIISOdWOD ay], “| a1GeL

74 WA Mattingley

SB18(2)

Bird feathers with occasional legs and beaks
were found in <16% of pellets (Table 1) and
Galliformes, particularly Red Grouse, were
most prevalent, particularly in the October
and November 1991 samples. However,
only 4% of pellets in the October/November/
December 1992 sample contained signs of
bird and none were found in the January/
March 1993 sample. Raven feathers found
in a small number of pellets were presumed
to have been ingested during preening.

Up to 8 moth cocoons were discovered in a
single pellet but 1 to 3 were more commonly
found in 11% of pellets. 20% and 19% of the
sample contained cocoons in December 1990
and in October 1991. Small numbers of
pellets (4%) contained the legs, elytra and
exoskeletons of beetles Coleoptera. The
majority of these were of the Silphidae family,
the burying and carrion beetles. No other
invertebrates were found.

Table 2 Results of Chi 2analysis, testing for variation between samples in the
number of pellets containing particular items. For commonly occurring items
(lagomorph and vegetable matter) all 12 samples were compared (df=11)
whereas for less frequently occurring items small samples had to be combined
(see text) so that comparisions were between 7 samples (df=6).

Item Chi2statistic
Lagomorph 2.62
Vegetable matter 32.81
Stones and grit Rat
Sheep 50.10
Vole 27.04
Bird 130.92
Rowan 234.08
Moth Cocoons 29.40

Rowan berries were recorded in 9% of pellets
with 50% in October/November/December
1992 but not in the previous October and
November. Insome cases, the berries formed
complete pellets and were frequently whole
with skins and seeds intact, mucus being the
only other recognisable constituent. Other
vegetation in the form of seeds, leaves, buds
etc was recognisable in <4% of pellets
including the only other berry, Bird cherry
Prunus padus (Table 1).

df (n-1) Probability value
11 >0.99
11 <0.001
6 0.20<P<0.30
6 <0.001
6 <0.001
6 <0.001
6 <0.001
6 <0.001

Remaining items such as garbage and egg
shell were presentin only very small quantities.

Discussion

Comparing this study with others in Britain,
only in Orkney was lagomorph found to be the
predominant food outside the Raven breeding
season (Marquiss & Booth 1986). Ravens
there were regularly recorded scavenging
Rabbit road casualties in the winter (Hope

1995

Jones 1980). This also occurred in Shetland
(Ewins ef a/. 1986) and south Scotland
(Marquiss et al. 1978).

In and around the Glen Quaich roosts, there
are several large Rabbit warrens and
myxomatosis was frequent in the 1990-91
and 1991-92 winters. Ravens are known to
kill Rabbits, probably mostly when injured or
young (Holyoak 1968; in Heinrich 1990;
Nogales & Hernandez 1993; Marquiss & Booth
1986) and in the glen and surrounding area
Ravens were regularly recorded near Rabbit
warrens as well as feeding on carcasses. In
one of the young plantations, about 7km from
the roosting glen, pest controllers intentionally
left shot or trapped Rabbits for Ravens and
Common Buzzards Buteo buteo. Another
regular source of lagomorphs were casualties
on public roads crossing the higher ground
within the area.

In Virginia USA (Harlow, Hooper, Chamberlain
& Crawford 1975), medium sized mammals
particularly Virginia Opossum Didelphis
marsupialis and Cottontail Rabbit Sy/vilagus
Spp. were more frequently found in pellets
from the spring roost and nest sites than in
those from the winter roost when sheep was
the most common constituent due to a winter
lambing regime in January and February. In
the Perthshire study, sheep remains were
present in only 10% of pellets. This lower
than expected figure is not easily explained
when stocking rates in the winter were 0.5 to
3 sheep/ha, with an expected mortality of 5-
15%. Most sheep are kept inby in the winter,
with few carcasses being available to Ravens
(A & M Kennedy pers. comm.). The high
frequency of sheep wool in pellets in October
and November 1991 may be due to a ‘clean-
up’ of summer deaths, when stocking rates
were higher, but this does not explain the low
proportion of sheep remains in the following
year.

Winter diet of Ravens in Perthshire 75

There were fewer than expected signs of Red
Deer in the samples, perhaps because deer
hair was seldom ingested when the birds fed
on grallochs or opened carcasses. Ravens
rely on Foxes Vulpes vulpes, raptors or
humans to open up large mammal carcasses
such as deer before they can feed (Hewson
1981; Marquiss & Booth 1986; Heinrich 1990).

Small mammals, particularly voles, have been
an important food item in several studies.
There is much suitable vole habitat near Glen
Quaich and during the 1991-92 and 1992-93
winters, unusually large numbers of wintering
Hen Harriers Circus cyaneusand Short-eared
Owls Asio flammeus were in the area,
suggesting high vole numbers, but only the
February 1991 and March 1993 pellets
contained a high frequency of voles.

Ravens are opportunistic feeders. In the
autumn of 1992, the Perthshire Ravens took
advantage of a large crop of Rowan berries.
Inthe Canaries (Nogales & Hernandez 1993),
berries and fruit were found to be important in
the diet especially on the central and western
islands where there was more vegetation and
it was suggested that Ravens there were
providing an important function as seed
distributors. This may also occur in Perthshire.

Where vegetable matter is a significant
constituent in the diet, Soler et a/. (1993)
suggest that the amount of grit and stones
ingested increases. In their study area in
southern Spain, the diet was mainly cereal,
and grit was present in 97% of the pellets. In
the present study, where lagomorph was the
main food, grit and stones appeared in only
16% of pellets.

In December 1990 and October 1991, there
was a high occurrence of moth cocoons,
mainly of Northern Eggar Lasiocampa quercus
callunae and Emperor Pavonia pavonia

76 WA Mattingley

moths, in the Perthshire pellets. Only small
frequencies of insect remains (mainly
Coleoptera) were found in most other British
studies and in the Canaries. However, in
Idaho insects were the third most common
constituent by weight in the summer and
autumn diet. Only 2 studies (both in the USA)
cited in Engel and Young have shown insects
to be important in the diet of Ravens.

Garbage was found in very few pellets in
Orkney, south Scotland and also in this study
presumably because the distance to the
nearest known rubbish tip was 20km. In
Argyll, Ravens roost, throughout the year,
within a short distance of a town dump which
they scavenge daily, numbers varying from
c20 to up to 200 depending on the amount of
food exposed (G Scott pers. comm.).

Over the past 20 years, winter roosts have
been known at different sites in central and
western Perthshire with up to 80 birds being
recorded (cited in Perth & Kinross SOC
records) but recent counts suggest the winter
population may be as high as 200 birds.
Central Perthshire is therefore extremely
important for wintering Ravens where they
are relying heavily on Rabbits and Hares.
Further studies are required to discover the
movements of this population, the effects of
persecution and the possible impact of Rabbit
Haemorrhagic Disease (RHD).

Acknowledgements

| would like to thank Dr M Marquiss for the use
of his personal collection of hair and feather
samples. | am grateful to M Robinson for
identifying beetle remains and E Cameron, A
Robertson and K Thompson kindly assisted
with the collection of pellets. Dr M Marquiss
gave advice throughout and commented on
earlier drafts of this paper.

SB 18(2)

References

Coombs F 1978. The Crows. B T Batsford
Ltd. London.

Corbet G B & Southern H N 1977. The
Handbook of British Mammals, 2nd edn.
Blackwell Scientific Publications, Oxford.
Dare P J 1986. Raven Corvus corax
populations in two upland regions of north
Wales. Bird Study 33:179-189.

Day M G 1965. Identification of hair and
feather remains in the gut and faeces of
Stoats and Weasels. J.Zool. 148:201-217.
Debrot S, Mermod C, Fivaz G & Weber J-M
1982. Atlas des Poils de Mammiferes
D’Europe.

Engel KA & YoungL S 1989. Spatial and
temporal patterns in the diet of Common
Ravens in southwestern Idaho. Condor
91:372-378.

Ewins P J, Dymond J N & Marquiss M 1986.
The distribution, breeding and diet of Ravens
Corvus coraxin Shetland. Bird Study33:110-
116.

Gibbons D W, ReidJB & Chapman RA 1993.
The New Atlas of Breeding Birds in Britain
and Ireland: 1988-1991. T & A D Poyser,
London.

Goodwin D 1986. Crows of the world. St
Edmundsbury Press Ltd, Suffolk.

Harlow R F, Hooper R G, Chamberlain D R &
Crawford H S 1975. Some winter and nesting
season foods of the Common Raven in
Virginia. Auk 92:298-306.

Heinrich B 1990. Ravens in Winter. Barrie
and Jenkins, London.

Holyoak D 1968. A comparative study of the
food of some British Corvidae. Bird Study
15:147-153.

Holyoak D & Ratcliffe D A 1968. The
distribution of the Raven in Britain and Ireland.
Bird Study 15:191-197.

Hope Jones P 1980. Bird scavengers on
Orkney roads. Brit. Birds 73:561-568.

1995

Knight R L & Call. MW 1974. The Common
Raven U S Dept. Interior Tech. Note No 344.
Marquiss M & Booth C J 1986. The diet of
Ravens Corvus corax in Orkney. Bird Study
33:190-195.

Marquiss M, Newton | & Ratcliffe D A 1978.
The decline of the Raven Corvus corax in
relation to afforestation in southern Scotland
and northern England. J.app/.Eco/. 15:129-
144.

Newton I, Davis P E & Davis J E 1982.
Ravens and Buzzards in relation to sheep
farming and forestry in Wales. J.appl.Ecol.
19:681-706.

Winter diet of Ravens in Perthshire The

Nogales M & Hernandez E C 1993.
Interinsular variations in the spring and
summer diet of the Raven Corvus coraxinthe
Canary Islands. /bis 136:441-447.

Ratcliffe DA 1990. Bird life of mountain and
upland. Cambridge University Press,
Cambridge.

Soler J J, Soler M & Martinez J G 1993. Grit
ingestion and cereal consumption in five corvid
species. Ardea 81:143-149.

Yalden D W 1977. The Identification of
Remains in Owl Pellets. Mammal Society,
Reading.

Wendy A Maittingley, Cluny House, Aberfeldy, Perthshire PH15 2JT

Revised Manuscript accepted September 1995

Hg A yp wh A,

Raven

he ie tig i 7p ee ee

Keith Brockie

78 Scottish Birds (1995) 18:78-87

Seabirds of Handa Island

SB 18(2)

JG STONEMAN & N A WILLCOX

Handa Island has been an important site for breeding seabirds since
records began. This paper collates and presents the relevant data. Three
new species have colonised within the past 100 years and the populations
of most species have either increased or remained stable. However, Puffin,
Shag, Kittiwake and large gull populations have decreased and Black

Guillemots no longer breed.
Introduction

Handa Island has always been an important
breeding ground for seabirds (eg Harvey-
Brown and Buckley 1887, Harvey-Brown and
MacPherson 1904, Lumsden 1886). The
local minister, writing in the late 18th century,
describes Handa as being remarkable for
being the resort of vast numbers of sea fowl
of different kinds.. so great (in number) that
the whole face of these tremendous rocks,
and the sea in the neighbourhood, appear
covered with them” (Sinclair 1979). The 12
crofting families that inhabited the island until
1848 exploited the seabirds for food and
feathers (Tulloch 1845).

The island was made a wildlife reserve in
1961 by agreement with the owners, DrJ & Mr
J Balfour. It was managed by the RSPB until
1991 and subsequently by the Scottish Wildlife
Trust. The first detailed island count was
made by the RSPB in 1970 as part of Operation
Seafarer (Cramp et al. 1974). Since 1972
most seabird monitoring studies have been
recorded in seasonal wardens reports. This
paper is a summary of a detailed report
(Stoneman & Willcox 1995).

Methods

Unless otherwise referred to the data has
been obtained from warden’s reports. This
includes all island counts made as part of the
national Seabird Colony Register. The data
used to compile this report where possible
conforms to monitoring guidelines specified
inthe Seabird Monitoring Handbook of Britain
and Ireland (Walsh et al. 1995). Where
methods depart significantly from these
guidelines, notes are given in the results
section.

Results
1. Fulmar Fulmarus glacialis

Handa was first colonised by Fulmars in 1903,
the first such place in the northern Highlands
(Cramp et a/ 1974). Six all island counts of
apparently occupied sites (AOSs) have been
made, the first in 1970 (Fig.1). They peaked
in 1977, since when numbers have dropped
slightly and levelled off at around 3000 pairs.

A single monitoring plot was established in
1983; the number of AOSs has not changed
significantly. Productivity has been measured
in this monitoring plot since 1992. From
1992-1995 (in chronological order) 0.43, 0.40,
0.56 and 0.42 chicks fledged per AOS.

1995 Seabirds of Handa Island 79

Fig. 1 All-island counts of Fulmar AOSs 1970 - 1989

a1:
|
|
a
|
}
~

|

2000 ——_-—-—----—-— -+_——_—_- + t ==
Anime ro Fa f5 16 wf 7 79° SO> 8182 83 “84 85 86° 87 °88
Year

Fig. 2 All-island counts of Shag AONs 1970 - 1995

|
a

|
4

+

|
SS SSS eee ee + —— —__——j —__}-~___j

Vesa Se pe Se et

70 71 72 73 74 75 76 77 78 79 80 8i 82 83 84 85 86 87 88 89 390 91 92 93 94 95
Year

80 J G Stoneman et al. SB 18(2)

Fig. 3 AOTs and no:.s of breeding pairs of Arctic Skuas 1968 - 1995

—}*. _ Breedingjprs.

a AOS

a a a a ee
69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95
Year

7 Breeding prs.

|
=— 2, =3 AGIs |

— nl

71995

Seabirds of Handa Island 81

2. Shag Phalacrocorax aristotelis

There are records of all island, land based
counts from 1970. The population peaked in
1973 and 1974 when there were c400
apparently occupied nests (AONs). The
population had dropped by over half by 1979,
and, although it made a slight recovery,
numbers have now fallen to an all time low

(Fig.2).

This fall in numbers is reflected in monitoring
plot counts which show a decline from 45
AONSs in 1990 to just 18 in 1995.

3. Arctic Skua Stercorarius parasiticus

Arctic Skuas first nested on Handa in 1968,
the first place to be colonised in the NW

Highlands (Cramp etal. 1974, Furness 1986).
Numbers increased steadily from 1973 to
1985, after which the population stabilised at
around 30 apparently occupied territories
(AOTs) (Fig.3).

Productivity during the years 1989, 1990 and
1991 was assessed at 1.25, 1.19 and 1.36
chicks fledged per nest respectively (Baber
1989, Baber 1990 and Furness and Aitken
1992).

4. Great Skua Stercorarius skua

Great Skuas first nested on Handa in 1964
(Waterston 1965, Furness 1986). After a
slow start, numbers have increased steadily
to 115 AOTs in 1995 (Fig.4).

Fig. 5 Breeding pairs of large gull spp. 1970 - 1995

a
_
a
a
=
3
a
@
=
a
S
=

—*— Gt. b backed
——#—_ Herning

aan COMMON

A ey y e
ea Bal
et i A I i ree oe

70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95

Year

82 J G Stoneman et al. SB 18(2)

Fig. 6 Kitthwake peak AONs on the Great Stack 1983 - 1995 and for the four
monitoring plots combined 1990 - 1995

—"— Gt. Stack only

——-— All 4 plots combined

—*— 2 monitoring plots

——O— All four combined

No. individuals

a a ae ce) ar

sea cians | pe ee
75 6 @ 78 79 680 81 62 83 84 65 66 87 88 69) (90 (Sl N92 cSicaaees
Year

igoo

Since 1988 between 50 and 100 Great Skuas
have been recorded at the Hill Loch club (a
congregation of non breeding birds).

Productivity during the years 1989, 1990 and
1991 was assessed at 1.25, 1.20 and 1.28
chicks fledged per nest respectively (Baber
1989, Baber 1990 and Furness and Aitken
1992).

5. Common Gull Larus canus

Handa has a small population of fewer than
20 pairs of Common Gulls (Fig.5).

6. Lesser Black-backed Gull Larus
fuscus

Small numbers (0-6 pairs) of Lesser Black-
backed Gulls bred up to 1989.

Seabirds of Handa Island 83

7. Herring Gull Larus argentatus

Numbers of Herring Gull peaked at 418 pairs
in 1977 but have subsequently declined to
only 12 pairs in 1995 (Fig.5).

8. Great Black-backed Gull Larus
marinus

In 1970 there were 62 pairs of Great Black-
backed Gulls, but, since 1989, numbers have
declined to a stable population of 26-27 pairs
(Fig.5). Atone time there were 20-25 pairs on
the top of the Great Stack but human
disturbance in 1883 and 1886 caused them to
abandon the breeding site (Harvey-Brown
and MacPherson 1904, Gordon 1935); in
1995 2 pairs nested on top of the Great Stack.

Fig 8 Average Razorbill head count on Little Stack 1974 - 1995

g

=
o
S)
3
2
ao
=
fo}
=

oO
oO

0 +-—+—++— +} +--+ + + + Ht
82 83 84 85 86 87 88 89 90 91 92

(Aerio. wi 18) 79) “80! “81

pt tp at tt tt et

Year

84 JG Stoneman et al.

SB 18(2)

Fig. 9 Peak Puffin Counts 1977 - 1995

——— Gt. Stack
——O— All island

No. individuals

9. Kittiwake Rissa tridactyla

Total island counts of apparently occupied
nests were made in 1970 (8,328), 1977
(12,500), 1987 (10,732) and 1995 (7,418).
Monitoring plot counts made since 1983 for
the Great Stack, and since 1990 for a further
3 plots, indicate a largely stable population
over the period, although there was a peak in
numbers in 1991 (Fig.6).

Five productivity monitoring plots were
established in 1990; productivity has ranged
from 1.22 to 1.58 chicks fledged per pair.

10. Common Tern Sterna hirundo

Fewer than 8 pairs of Common Terns usually
breed on Handa.

11. Arctic Tern Sterna paradisaea

Until 1981 only one pair of Arctic Terns bred.
Since then there have been 1-5 pairs.

12. Guillemot Uria aalge

Island censuses conducted in 1970 (30,790
individuals), 1977 (25,000), 1987 (98,680)
and 1994 (119,589+) indicate a substantial
increase in numbers. In contrast, there has
been little change in numbers at the 2
monitoring plots that were established in 1975
(Fig.7). Two additional plots have been
monitored since 1990. There has been a
small increase in numbers in the four
monitoring plots combined (Fig.7).

Guillemot productivity studies have been
carried out since 1991 at 3 sites.
Approximately 0.70 chicks have fledged per
pair each year.

13. Razorbill Alca torda

All island counts made in 1970 (8,367), 1977
(9,000) and 1987 (16,390) indicate a
substantial increase in numbers. On the
other hand, counts made at a monitoring plot

1995

indicate a steady decline (Fig.8). Productivity
was measured in 1988 and 1990 with 0.64
and 0.84 chicks fledged per occupied site
respectively.

14. Black Guillemot Cepphus grylle

Harvey-Brown and Buckley (1887) reported
local knowledge that Black Guillemots bred at
one time on Handa but that “the rats had
managed to dislodge them”. Small numbers
(2-4 pairs) nested between 1970 and 1978,
but have not bred since.

15. Puffin Fratercula arctica

Harvey-Brown and MacPherson (1904)
reported that Puffins used to nest on the
slopes of the island, but when they visited
they found that “rats simply swarm over the
best areas, and the Puffins have been
forced...over the edge”. Puffins are now
found only on the Great Stack and on cliff
ledges.

Until 1977 the number of pairs present on the
island was merely estimated. Since then the
peak numbers of individuals (ie late July or
early August) have been recorded on a
monitoring plot on the Great Stack and on all
island censuses (Fig.9). Whole island counts
increased by almost 300% between 1980
and 1987 while numbers on the Great Stack
remained relatively stable.

Discussion

Handa Island has been an important site for
breeding seabirds since records began. It
currently holds 13 species in the breeding
season. Fulmars, Arctic Skuas and Great
Skuas have colonised the island this century
and the populations of most of the other
breeding species have either remained
constant or have increased.

Seabirds of Handa Island 85

The most important species on the island is
the Guillemot which breeds at an
internationally important level. Numbers have
increased substantially since 1970, reflecting
national trends (Lloyd et a/. 1991). Handa
Island hold the largest colony of Guillemots in
Britain and Ireland. Inthe 1987 census Handa
held 8.2% of the population of Britain and
lreland: almost double the number of birds in
the next largest colony.

Whole island censuses of Razorbills have
also shown a substantial increase in numbers,
consistent with national trends. In the 1987
census Handa was the largest colony in Britain
and Ireland with 9.1% of the population (Lloyd
et al. 1991).

The expansion of the Kittiwake population
between 1970 and 1987 also follows national
trends. The reason for the subsequent 30%
decline is uncertain. The very high productivity
figures recorded since 1990 suggestan ample
source of recruits for the breeding population.

The principal colony of Arctic Skuas in
Sutherland is on Handa (Lloyd et a/. 1991).
Although numbers of Arctic Skuas have been
increasing in other areas, the increase on
Handa between 1973 and 1985 was the most
pronounced compared to anywhere else in
Britain (Gibbons et a/. 1993). Handa is also
an important site for Great Skuas whose
numbers are still rising. Their breeding
success between 1989 and 1991 was
considered to be spectacularly and
consistently high (R Furness, pers. comm.).
The increase in the Arctic Skua population
levelled off in 1985, possibly because of
competition from Great Skuas (Furness 1987).

Some species have declined. Shag numbers
have plummeted since 1970, in contrast to
national trends. Lloyd eta/. 1991 relate Shag
declines in other parts of the country to poor

86 J G Stoneman et al.

SB 18(2)

food supply (particularly in early spring) and
marine pollution.

The Herring Gull has also shown a long term
decline consistent with national trends. Great
Black-backed Gull numbers have also
declined to a lesser extent.

Human interference is believed to have hada
lasting affect on the Puffin and the Black
Guillemot. Historical records link the decline
of these 2 species with predation by Brown
Rats (Harvey-Brown and Buckley 1887,
Harvey-Brown and MacPherson 1904).
Research carried out in 1994 and 1995
(Aragundi 1994, Warden‘s Reports 1994 and
1995) provides further evidence that rats are
limiting seabirds. Puffins are limited to
inaccessible areas such as the Great Stack
and Black Guillemots have been eliminated.
The eradication of rats on the island is one of
the most important future management
options for the reserve, opening up the
prospect of an increase in the numbers of
Puffins and a return of Black Guillemots. It
could also allow other species of burrow
nesting birds, such as the Manx Shearwater
Puffinus puffinus and the Storm Petrel
Hydrobates pelagicus, to colonise the island.

Acknowledgements

We are very grateful for the encouragement
of Dr and Mr Balfour, the owners. Wardens’
reports until 1991 were supplied by the RSPB.
Paul Walsh of JNCC provided missing seabird
data and Bob Furness of the Applied
Ornithology Unit at Glasgow provided data on
Skuas.

References

Aragundi S 1994 The effects of predation by
the Brown Rat on Skuas on Handa Island.

Unpublished MSc. Thesis, Aberdeen
University.

Baber | 1989 Breeding success of seabirds of
Handa Island, Sutherland 1989. NCC Chief
Scientist Directorate commissioned report no
922.

Baber | 1990 Breeding success of seabirds of
Handa Island, Sutherland 1990. NCC Chief
Scientist Directorate commissioned report no
1136.

Cramp S, Bourne W R P & Saunders D 1974
The Seabirds of Britain and Ireland Collins,
London.

Furness R W 1986 The Conservation of Arctic
and Great Skuas and their Impact on
Agriculture. Report to NCC pp 144.
Furness R W 1987 The Skuas. Poyser,
London.

Furness R W & Aitken A 1992 Breeding
success of seabirds on Handa Island,
Sutherland 1991. JNCC report to 48.
Gibbons D W, Reid J B & Chapman RA 1993
The New Atlas of Breeding Birds in Britain
and Ireland 1988-1991. Poyser, London.
Harvey-Brown J A & Buckley T E 1887 A
Vertebrate Fauna of Sutherland, Caithness
and West Cromartypp 38-40. David Douglas,
Edinburgh.

Harvey-Brown J A & MacPherson HA 1904 A
Fauna of the North West Highlands and Skye
Lloyd C, Tasker M L & Partridge K 1991 The
Status of Seabirds in Britain and Ireland.
Poyser, London.

Lumsden J 1886 The Island of Handa, with
notes on its birdlife. In:Natural History and
Archeological Society 1884 - 85 pp 36-43.
Sinclair J (Ed.) 1979 Parish of Edderachilles.
In: The Statistical Account Scotland 1791 -
1799. Vol XVIII pp 395-402.

StonemanJG & Willcox NA 199) The Seabirds
of Handa Island Wildlife Reserve.
Unpublished report, Scottish Wildlife Trust,
Edinburgh.

Tulloch G 1845 Parish of Edderachilles. In:

1995

New Statistical Account of ScotlandVol 15 pp
132 William Blackwood and Sons, Edinburgh
and London.

Walsh P M, Halley G J, Harris M P, del Nevo
A, Sim | M W & Tasker M L 1985 Seabird
Monitoring Handbook for Britain and Ireland.
JNCC.

Waterston G 1965 Great Skuas breeding in
North West Highlands. Scott. Birds 3:313.

Seabirds of Handa Island 87

Warden’s reports 1974-1995: (in chronological
order) E Evans, D Mower, C Wells, M
Trubridge, K Fairclough, R Thorpe, J Leece,
M Allison, M Walker, C Hurlford, P Fisher, R
Ashcroft, G Macdonald, C Self, | Lawrie, M
Self, A Aitken, J Breckenridge, | Gordon, N
Harding and J Stoneman.

J G Stoneman and N A Willcox, Scottish Wildlife Trust,
Kirk Cramond,Cramond Glebe Road, Edinburgh. EH4 6NS

Revised manuscript accepted October 1995

Guillemot

oy “en
%

”

|
2 3
F
a at

Don Smith

88 Scottish Birds (1995) 18: 88-94 SB 18(2)

An analysis of prey taken by Merlins within the Lammermuir
Hills, 1984-1994

A HEAVISIDES, | R POXTON & A W BARKER

The prey remains found at Merlin breeding sites within the Lammermuir Hills
during the period 1984-1994 have been analysed. A total of 2040 individual bird
prey items were recorded, representing 43 species. However, only 7 of these
species accounted for approximately 89% by number (or 75% by weight) of the
total items located. A small number of insects and a single vole were also
recorded. The relative importance of open country species, and in particular
the Meadow Pipit, was confirmed, although woodland species were found to be
significant early in the season. Fledglings and nestlings were frequently taken
during June. Although all nests were on prime grouse moorland only 1% of
prey were small chicks of this species.

Introduction

The Merlin Falco columbarius is well
recognised as the small, uncommon raptor of
the uplands of western and northern Britain.
Its habitat in the breeding season is typically
open moorland, and, in contrast to most other
raptors, appears to benefit from nesting on
well managed grouse moorland, with heather
as opposed to grass, the dominant vegetation
(Bibby and Nattrass 1986; Cramp and
Simmons 1980). Its diet consists almost
entirely of small birds of the open country,
and an abundance of such species is
presumably a prerequisite for successful
breeding.

As part of a comprehensive study of the
breeding biology of the Merlin in the
Lammermuir Hills in southeastern Scotland,
which was started in 1984, records have been
kept of the prey items found within the vicinity

of Merlin nests.This paper presents an
analysis of this aspect of the study. Of the
several other published studies on the prey of
Merlins, those most likely to be comparable
were undertaken in Northumberland by
Newton et a/(1978;1984;1986) and in Wales
by Bibby (1987).

Study area and Methods

The Lammermuir Hills form a distinct group at
the northeastern end of the Southern Uplands
of Scotland. They cover approximately 400
square kilometres, and around two thirds of
the landis managed primarily for sheep rearing
and Red Grouse Lagopus |. scoticusshooting.
Additionally some Pheasant Phasianus
colchicus rearing occurs around the edge of
the area in coniferous plantations and
remnants of old woodland.

1995

The breeding population of Merlins has been
studied and monitored systematically since
the early 1980s. All known sites are visited
annually and the area supports around 15
pairs. The habitat is remarkably similar
throughout the study area and all nests were
located on heather moorland managed for
sheep and grouse shooting, with some
additional hunting access to small areas of
woodland, river valley systems and some
more grassy, sheepwalk habitat but generally
well away from agricultural farmland.

During each site visit a search was usually
made of likely plucking areas. The most
frequently used areas were on fence lines,
burnt heather patches, bare areas, rocks and
grouse shooting butts. Remains were normally
identifiable in situ but if this was not possible
they were placed in labelled plastic bags and
removed for later identification. This was
undertaken with the aid of bird skins, including
specimens from the Royal Scottish Museum.

Results

A total of 2040 individual bird prey items were
found during the 11 year period. Additionally
a small number of insects and a single Field
Vole Microtus agrestis were recorded.
Remains of species normally associated with
woodland or scrub were found, in small
numbers only, at virtually all sites. A few sites
would appear to offer greater opportunities
for capture of such species but all were clearly
most heavily reliant upon birds of the open
country.

Bird prey numbers and the percentage of diet
by numbers and weight is outlined in Table 1.
Mean prey species weights were calculated
from data provided in Birds of the Western
Palearctic (Cramp and Simmons 1977-1994)
and are also included in Table 1. In the case

Prey taken by Merlins in the Lammermuir Hills 89

of wader and grouse chicks, none of which
were much more than about a quarter
developed, corrections for weight were made
based on this proportion. A total of 43 species
were recorded ranging in size from Goldcrest
(seeTable 1 for scientific names of prey
species) to Woodcock and in weight from
circa 6 to 300 grams. The commonest 3
species which were Meadow Pipit, Skylark
and Wheatear are in the range of 18-36
grams.

The problem of being sure that the larger prey
items were indeed killed by Merlins, rather
than other raptors such as the Peregrine
Falco peregrinus, is acknowledged. Larger
species kill remains were, however, not
attributed to the Merlin unless found in clearly
occupied territories on plucking posts and in
near association with passerine kills. The
larger species were not generally found during
the incubation period suggesting these were
only taken when the female was contributing
to the hunting.

In terms of numbers the most important
species were clearly Meadow Pipit (62.9 %),
Skylark (9.3%), Wheatear (6.2%), Chaffinch
(4.5%), Starling (2.5%), Snipe (2.3%) and
Fieldfare (1.4%).Clearly Meadow Pipits are
captured approximately 7 times more often
than Skylarks and 10 times more often than
Wheatears. On a weight basis the order of
importance changes slightly to: Meadow Pipit
(37.7%), Skylark (10.8%), Snipe (8.1%),
Starling (6.4 %), Wheatear (4.7%), Fieldfare
(4.5%) and Chaffinch (3.3%). These 7 species
contributed approximately 89% of all avian
food taken by number or 75% by weight. The
study only followed the Merlins through until
late July and it would seem possible that the
importance of the Starling might well increase
from that time until the young Merlins had
dispersed, as flocks of mainly juvenile

90 A Heavisides et al. SB 18(2)

Table 1 Bird prey species of Merlins in the Lammermuir Hills 1984-94

Numbers found in Percent of diet by Mean
Mar/Apr May June July total number weight weights
used gms
Red Grouse Lagopus |. scoticus 0+4* 0+14* 0+2* 20 1.0 4.8 150*
Golden Plover Pluvialis apricaria 2 2 4 0.2 1.3 210
Lapwing Vanellus vanellus 3 5+3* = 1+1* 13 0.6 3.5 220+50*
Snipe Gallinago gallinago 11 23 7+2* 3 46 23 8.1 115+30*
Woodcock Scolopax rusticola # 2 1 3 0.1 1.4 300
Redshank Tringa Totanus 1 1 <0.1 0.2 150
Common Sandpiper
Actitis hypoleucos 2 5 1 8 0.4 0.7 66
Feral/racing Pigeon i
Columbia livia var. 3 3 6 0.3 2.9 400
Skylark Alauda arvensis 10 62 104 14 190 9.3 10.8 36
Sand Martin Riparia riparia 1 1 <0.1 <0.1 13
Swallow Hirundo rustica 1 1 2 <0.1 <0.1 19
Meadow Pipit Anthus pratensis 132 574 479 98 1283 62.9 37.7 18.5
Pied Wagtail Motacilla alba 4 4 5 5 18 0.9 0.6 20
Wren Troglodytes troglodytes # 1 1 <0.1 <0.1 8.5
Dunnock Prunella modularis # 1 2 1 4 0.2 0.1 19
Robin Erithacus rubecula # 1 2 3 0.1 0.1 18
Whinchat Saxicola rubetra 5 5 2 12 0.6 0.3 AT
Wheatear Oenanthe oenanthe 3 U2 42 9 126 6.2 4.7 24
Ring Ouzel Turdus torquatus 1 3 1 5 0.2 0.8 100
Blackbird Turdus merula # 6 4 1 11 OS V7 96
Fieldfare Turdus pilaris 22 6 28 1.4 4.5 102
Song Thrush Turdus philomelos# 4 9 13 0.6 i 74
Redwing Turdus iliacus 5 2 t 0.3 0.7 62
Mistle Thrush Turdus viscivorus# 1 3 4 0.2 0.8 123
Whitethroat Sylvia communis # 1 1 <0.1 <0.1 15
Willow Warbler
Phylloscopus trochilus # 1 11 2 14 0.2 0.2 8.5
Goldcrest Regulus regulus # 10 1 11 0.5 0.1 6
Coal Tit Parus ater # 1 ite 1 9 0.4 0.1 9
Blue Tit Parus caeruleus # 4 2 6 0.3 0.1 Uns
Great Tit Parus major # 1 2 3 0.1 0.1 18
Treecreeper Certhia familiaris # 1 1 <0.1 <0.1 9
Starling Sturnus vulgaris 5 4 34 7 50 ENS, 6.4 80
House Sparrow Passer domesticus # 4 4 0.2 0.2 30
Chaffinch Fringilla coelebs # 28 49 14 91 4.5 3.3 22
Siskin Carduelis spinus # 1 5 6 0.3 0.1 14
Greenfinch Carduelis chloris # 2 5 1 1 9 0.4 0.4 28
Goldfinch Carduelis carduelis # 3 3 0.1 0.1 16
Bullfinch Pyrrhula pyrrhula # 1 1 <0.1 <0.1 30
Redpoll Carduelis flammea # 1 1 <0.1 <0.1 14
Linnet Carduelis cannabina # 1 3 2 3 9 0.4 0.3 19
Crossbill Loxia curvirostra # 1 1 <0.1 0.1 41
Yellowhammer Emberiza citrinella# 2 5 2 9 0.4 0.4 26
Snow Bunting Plectrophenax nivalis 2 2 0.1 0.1 38

#counted as woodland/scrub species in Table 3 *small chicks (weights estimated)

1995

Starlings were common on the moors from
late June onwards.

Table 2 shows the increasing importance of
young birds of some species as prey as the
breeding season progresses. All 31 nestlings
(1.5% of total prey) specifically identified were
Meadow Pipits. With such poorly developed
prey remains, it is possible that some may
have been wrongly assigned to this species
although none were suspected of being any
other. All Red Grouse chicks were small and
probably only just able to fly. Lapwing and
Snipe chicks were too small to have been
flying. Fledglings were defined as birds with
incompletely grown feathers, usually with
waxy sheaths still attached, but which were
presumed capable of flight when caught. In

Prey taken by Merlins in the Lammermuir Hills 91

the case of Starlings, these were of little
importance as prey until flocks of juveniles
appeared on the hills from around late June.
Juvenile plumage was identifiable from prey
remains. The overall proportion of birds of the
year identified amongst prey remains rose
from 1% in May to 31% in June, dropping to
21% in July.

Non bird prey identified consisted of 12
Emperor Moths Saturnia pavonia, 6 Northern
Eggar Moths Lapsiocampa quercus, one
Bumble Bee Bombus sp., many Ladybirds
Coccinellidae sp. and one Field Vole Microtus
agrestis. Inthe case of the moths, wings were
found on the plucking posts as were the wings
and body ‘fur of the bee. The vole’s tail was
foundin aregurgitated Merlin pellet. Ladybirds

Table 2 Merlin prey species which included identified chicks/fledglings/nestlings/
juveniles

Species Age
Red Grouse Adult
Chick
Lapwing Adult
Chick
Snipe Adult
Chick
Skylark Adult
Fledgling
Meadow Pipit Adult
Fledgling
Nestling
Pied Wagtail Adult
Fledgling
Wheatear Adult
Fledgling
Starling Adult

Juvenile

Number found in:

Mar/Apr May June July
4 14 2
3 5 1
3 1
11 23 Y 3
2
10 62 99 14
5
132 5f2 350 87
2 102 if
27 4
4 4 3
2 5
3 72 29 8
13 1
5 4 =
31 7

92 A Heavisides et al.

SB 18(2)

are not normally noted in this area but in 1986
were extremely common and many body
casings were foundin Merlin pellets. This was
the only year this was observed. It is clearly
possible that insect prey is much more
common than indicated from these few
remains but, interms of importance, is unlikely
to approach that of avian species.

Table 3 illustrates the relative importance of
birds associated mainly with woodland and
scrub during the early part of the breeding
season when the more open country species
are still scarce, and their declining importance
thereafter. Table 1 shows which species were
allocated into each category. Allocation of
species into the woodland/scrub or open-
country categories follows that used by
Newton eta/(1984). Easily the mostimportant
woodland prey species was the Chaffinch,
4.5% by number overall.

Discussion

The analysis of prey remains produced
findings more comparable to those found in
Northumberland during 1974-82 (Newton et
al. 1978, 1984 ) than to those in Wales 1981-
84 (Bibby 1987). The Northumberland study
covered a large area of upland similar in
nature to the Lammermuirs but also with
more open grassland and young forest

plantations. The Welsh nests were also
located predominantly on open moorland but
generally within 2 to 4 km of abundant
farmland. This feature was shown by Bibby
(1987) to be the strongest correlate of prey
diversity. Information on hunting distances
from nests is scarce but Merlins in Grampian
were proven to take prey from a distance of
5.6 km (Rebecca et a/ 1990). Within this
range most Lammermuir Merlin nests were
within reach of only a little (generally rough
grazing) farmland, and a few small woodland
shelterbelts. The available woodland/scrub
prey habitat is certainly less in this study area
than in either the Northumberland or, in
particular, the Welsh case. The totals of bird
prey species identified in the 3 localities are
respectively Wales 55, Northumberland 50
andLammermuirs 43. Most ofthe prey species
found in Wales, but not in the Lammermuirs,
were predominantly woodland or scrub
species and several are scarce or absent in
the latter area. A few species, predominantly
waders such as Golden Plover, Snipe and
Redshank, were scarce or absent in Wales
where they are uncommon moorland breeding
birds.

Despite a wider habitat variety in the
Northumberland study area than in the
Lammermuirs, the main prey species
proportions between these areas, both in

Table 3 Proportion of open country and woodland prey species in different months

Number (N) and percentage of prey species in:

April May

N % N
Open country species 144 63:2 726
Woodland species 70 31:8 - t16

June July
% N % N %
86.0 703 96.7 141) 99a
14.0 24 pe eo Je 4.8

1995

terms of numbers and percentage by weight,
are broadly similar. The Welsh study shows a
similar dependency on the Meadow Pipit,
easily the commonest upland species of that
size in all areas, Dut more divergence when
comparing the relative importance of other
bird prey.The proportion of Meadow Pipits
identified amongst prey remains was
respectively 63% in the Lammermuirs, 56%
in Northumberland and 60.5% in
Wales.Skylark was the second most frequenily
taken species in both Northumberland with
12% and in the Lammermuirs with 9%. In
Wales, however, Skylark was only the fifth
most abundant prey recorded accounting for
only 2.8% of the total. Chaffinch was the
second most common prey species in this
area at 8.3%. Wheatear was found to be the
third most often taken species in the
Lammermuirs with 6.2% and in Wales with
4.5% but only the fifth in Northumberland with
2.0%.

All 3 studies found that the proportion of the
total identified which consisted of woodland/
scrub species fell as the breeding season
progressed. In the Lammermuirs (Table 3) ,
the highest percentage of these species,
31.8% was recorded during April, compared
with 23.1% in Northumberland and 47.3% in
Wales, and fell to 4.1% in July with 7.0% and
15.5% respectively in the other study areas.
lt is clear that in all areas, as the available
open country prey increases with their arrival
on to the moors in spring, so the need to hunt
on the moorland edge habitat decreases.

The proportion of prey consisting of all juvenile
(ie.nestlings, fledglings and those identifiably
in juvenile plumage) birds compared with the
Welsh figures is interesting with 31% in each
area during June, increasing in Wales during
July to 68.4% but falling in the Lammermuirs
to 21%. The Northumberland figures included

Prey taken by Merlins in the Lammermuir Hills 93

full grown juveniles with adults and so are not
directly comparable but an increasing
proportion with time of fledgling birds (including
nestlings) was clear with a July maximum of
27.6%. The incidence of confirmed Meadow
Pipit nest robbing by Merlins in the
Lammermuirs with approximately 1.5% of
total prey identified specifically as nestlings
compares with approximately 0.6% in
Northumberland. Inthe Welsh study numbers
were calculated in a way not directly
comparable but would suggest perhaps an
even higher incidence. Earlier studies have
also identified this activity (Armitage 1932,
Hard and Enemar 1980, Kermot 1981, Sperber
and Sperber 1963, Roberts et al 1962).

All moorland bird species of suitable size
regularly observed in the study area were
recorded as prey items except for Dipper
Cinclus cinclus. This supports the conclusions
of other studies that Merlins will feed primarily
on all available small, mainly open-country
bird species available locally (Watson 1979,
Hodson 1978, Hard and Enemar 1980,
Oliphant and McTaggart 1977 ).

Insect prey may be much more common than
indicated from analysis of located remains.
Northern Eggar and Emperor Moths are seen
regularly and may well be normally eaten or
dissected where killed. The taking of ladybirds
in numbers during the one year when there
was an eruption shows that Merlins can take
advantage of whatever is suitable and
common.

Although the study area is entirely within well
managed grouse moorland, only 1% of prey
were small grouse chicks (an average of 2 per
year), clearly indicating that Merlins do not
make any impact on grouse stocks.

94 A Heavisides et al.

Acknowledgements

The co operation and support of all landowners
and gamekeepers within the Lammermuir
study area is fully acknowledged. From the
Start this has been excellent, and without it
this research would not have been possible.
During the study period many people kindly
informed the study team of sightings of Merlins
all of which were welcomed and subsequently
followed up. Ourthanks to all Such contributors
too many to mention by name. Bob McGowan
atthe Royal Scottish Museum was very helpful
in allowing access to study skins used in prey
identification. The Scottish Ornithologists Club
kindly aided the monitoring study financially
by way of a grant over a period of 6 years.

References

Armitage J 1932. Merlin taking young from
nests. Brit. Birds 25: 303-304.

Bibby C J 1987. Foods of breeding Merlins
Falco columbarius in Wales. Bird Study 34:
64-70.

Bibby C J and Nattrass M 1986. Breeding
status of the Merlin in Britain. Brit. Birds 79:
170-185.

Cramp S and Simmons K E L (Eds) 1977-
1994. The Birds of the Western Palearctic,
Oxford.

SB 18(2)

Hard | and Enemar A 1980. The choice of
prey and feeding by the Merlin Falco
columbarius during the nestling period. Var
Fagelvarld 39: 25-34.

Hodson K 1978. Prey utilised by Merlins
nesting in short grass prairies in southern
Alberta. Canad. Field Nat. 92:76-77.
Kermot L H 1981. Merlins as nest predators.
Raptor Res. 15:94-95.

Newton I, Meek E and Little B 1978. Breeding
ecology of the Merlin in Northumberland. Brit.
Birds 71: 376-398.

Newton |, Meek E and Little B 1984. Breeding
season foods of Merlins Falco columbariusin
Northumbria. Bird Study 31:49-56.

Newton 1, Meek E and Little B 1986. Population
and breeding of Northumbrian Merlins. Brit.
Birds 79: 155-170.

Oliphant L W and McTaggart S 1977. Prey
utilised by urban Merlins. Canad. Field Nat.
91: 190-192.

Rebecca G W, Cosnette B L, Duncan A,
Picozzi N and Catt D C 1990. Hunting distance
of breeding Merlins in Grampian indicated by
ringed wader chicks taken as prey. Scottish
Birds 16: 38-39.

Roberts E L 1962. Merlins taking newly
hatched passerines. Scottish Birds 2 : 245.
Sohdl N S, Oliphant L W, James P C,
Warkentin | G 1993. Merlin. The Birds of
North America. No.44.

Sperber | & Sperber C 1963. Notes on the
food consumption of Merlins. Zool. Birdrag.
Uppsala 35: 263-268.

Watson J 1979. Food of Merlins nesting in
young conifer forest. Bird Study 26: 253-258.

A Heavisides, 9 Addiston Crescent, Balerno, Edinburgh EH14 7DB
| R Poxton, 217 Newbattle Abbey Crescent, Eskbank, Dalkeith EH22 3LU
AW Barker, 12 Glenesk Crescent, Eskbank, Dalkeith EH22 3BL

Revised manuscript accepted October 1995

Scottish Birds (1995) 18: 95-100

Yellow-Legged Gulls in Scotland

95

R W FORRESTER

There is a strong possibility that Yellow-legged Gull may be treated as a full
species in Britain. With this possibility in mind the Scottish Birds Records
Committee has been reviewing Scottish records of Herrings Gulls with
yellow legs. To date 3 records of Yellow-legged Gull are regarded as ac-
ceptable. The identification of Yellow-legged and similar gulls is discussed.
Observers are asked to note the need for detailed descriptions.

Traditionally ithas been considered that there
are many races of Herring Gull Larus
argentatus stretching around the northern
hemisphere. The classification supported at
present by the British Ornithologists’ Union
has 6 races occurring in Europe; argenteus
breeding in Britain, Ireland, western France
and the North Sea coast to West Germany,
nominate argentatus from Denmark and
Scandinavia eastwards, michahellis in
western France, western Iberia, the
Mediterranean and Morocco, cachinnans
from the Black Sea eastwards, atlantis in
Azores, Madeira and Canary Islands, plus
the American race smithsonianus which has
now been recorded as a rare vagrant to
Britain.

Many authorities have in recent years
considered that the races of Herring Gull with
yellow legs occurring in southern Europe
(cachinnans, michahellis and atlantis ) should
correctly be treated as a separate species
which they have called Yellow-legged Gull
Larus cachinnans. Several countries,
including Holland and USA, have now given
Yellow-legged Gull full species status, but the
BOU, whichis responsible for maintaining the
Official list of British birds, adopts a more
conservative approach and does not at
present recognize it as a separate species.

One of the real difficulties is that, while the
situation in western Europe is relatively clear,
our knowledge of exactly what happens in
eastern Europe and beyond is inadequate. It
is, however, understoood that the BOU are
monitoring the situation and as this could at
some time in the future result in Yellow-
legged Gull being treated as a full species,
the Scottish Birds Records Committee (SBRC)
has set out to attempt to establish the status
in Scotland of this group of races. Whilst
SBRC has a policy of supporting the
classification adopted by the BOU, for
simplicity we will use the name Yellow-legged
Gull in the remainder of this paper to refer to
the group of races cachinnans, michahellis
and atlantis.

The breeding range of Yellow-legged Gulls of
the race michahellis has, in recent years,
expanded in a northerly direction, with new
colonies in France close to the English
Channel. Probably as a result of this range
expansion, Yellow-legged Gulls have started
to occur regularly in southern England,
although the first accepted British record was
only in 1971. Flocks of 100 or more birds are
now recorded annually in south east England
(West Sussex to Essex) in the late summer/
autumn period and a scattering of winter

96 RW Forrester

records are occurring throughout much of
southern England westward to Cornwall. The
species is, however, less common north of
the Midlands although records are now more
frequentin several areas, e.g. Tyneside, where
there is now a regular late summer influx.
Whilst observer awareness has undoubtedly
improved , there has certainly been a recent
increase in the number of birds reaching
England.

Status in Scotland

The status of Yellow-legged Gull in Scotland
remains far from clear and is complicated by
the fact that a proportion of Herring Gulls of
the race argentatus breeding around the
eastern Baltic have yellow legs, these being
generally referred to as “omissus “. The
status of “omissus’ birds is poorly understood
and, to complicate matters further, some
authorities consider “omissus’ to be aseparate
race of Herring Gull. Due to the breeding
range of “omissus” being primarily to the
north east of the British Isles, they appear
more likely to occur in Scotland than in
England, possibly arriving with other
argentatus birds which winter in Scotland in
large numbers. On the other hand,
michahellis, which now breeds in northern
France, is much more likely to occur in
southern England than in Scotland. _ It is,
however, not yet clear whether Yellow-legged
Gulls already occur in Scotland more
frequently than “omissus” although that is
likely soon to be the situation if Yellow-legged
Gulls continue their recent northward
expansion in western Europe.

Records of Herring Gulls with yellow legs
in Scotland prior to 1989

Priorto 1989 there were only 16 known records
in Scotland of Herring Gulls with yellow legs
and these were all either published without
further comment or described as “omissus”

SB 18(2)

birds. Baxter & Rintoul (1953) refer to the
Scandinavian Herring Gull Larus argentatus
omissus in The Birds of Scotland and quote 2
sight records: Fair Isle 28 September 1921
(with yellow legs); Orkney 14 August 1946,
and 3 recoveries in Scotland of Scandinavian
ringed birds. As they described the Scottish
breeding race as argentatus, whereas it is
now treated as argenteus, it would appear
that the Scandinavian ringed birds (no leg
colour mentioned) would more likely belong
to the nominate race argentatus with pink
legs. The Orkney record also has no mention
of leg colour and, therefore, may not be a
yellow legged bird. The Fair Isle record,
attributed to Dr Eagle Clarke and Admiral L
Stenhouse, must be considered to be the only
definite record of a yellow legged Herring
Gull published in that book. Valerie Thom
(1986) states that yellow legged Herring Gulls
are occasionally reported and these may
belong to either L.a. omissus_ or
L.a.michahellis but no specific records are
listed.

The first record detailed in Scottish Birds SB
1 :459 states “A Herring Gull was seen at
Seafield, Leith on 10 April 1961 with bright
yellow legs. Its size and pale silver mantle
were the same as ordinary Herring Gulls with
it and its race remains indeterminate (A G
Ablett).” Fromthe above description, although
very limited, the bird was most likely an
“omissus’.

The next reference in Scottish Birds SB 5:389
is interesting. “A yellow-legged Herring Gull
at Bridge of Don, Aberdeen, on 19 January
1969 with pure white, unstreaked head.
Identical to nearby Herring Gulls in size, build
and general behaviour. Bill appeared slightly
brighter yellow with more noticeable red spot.
Its back and wings were slightly darker than
those of other Herring Gulls. The legs and

1995

feet were dull yellow, not as bright as those of
a Lesser Black-backed Gull Larus fuscus, but
distinctly yellow and not merely pale flesh or
ivory asinmany Herring Gulls. | consider that
this bird showed the characteristics of a
Scandinavian Herring Gull L.a.omissus. (M
A MacDonald)” Whilst the observer
considered at the time that this was an
“omissus” bird, it actually showed some
characteristics of Yellow-legged Gull.

Eight records were published in Scottish Bird
Reports during the 1980s none of which was
believed to belong to the Yellow-legged Gull
group:

“yellow legged bird Tyninghame, EastLothian
4 September 1983 “ SBR 1983:26

“yellow legged birds: 1 Thurso Bay, Highland
9 and 11 September 1985; 1 Musselburgh,
Lothian 21 April 1985” SBR 1985:109
“yellow legged individuals, perhaps
Scandinavian “omissus” singles Strathclyde,
Oban 29 January and 4 February 1986,
Rigmuir 11 February 1986, Wellbrae 25
October 1986” SBR 1986:241
“Scandinavian L.a.omissus 1 Doonfoot,
Strathclyde 9 January 1987 and another
Hogganfield Loch, Strathclyde 24 January
1987” SBR 1987: 28.

The following records of “omissus’ also exist
for Ayrshire:

Adult in Ayr harbour 8 March 1974

1 Seamill 2 December 1980

1 Troon 19 November 1982

1 possible “omissus’ Doonfoot 23 October
1988.

As Scottish Birds did not publish these 4
records it may be that other records of yellow-
legged Herring Gulls have not been recorded.

Records of yellow legged Herring Gulls/
Yellow-legged Gulls in Scotland from 1989

Yellow-Legged Gulls in Scotland 97

The picture changed considerably from 1990
when it became clear that the southern
European races of Herring Gull might become
a separate species. Prior to that time there
was little interest in Herring Gulls with yellow
legs in Scotland and many records
undoubtedly went unreported. However,
Scottish Bird Reports for the 3 years 1990-
1992 included 19 records of yellow legged
birds, all but one being claimed as belonging
to the Yellow-legged Gull group. The
published records for 1990-1992 did not
include any information to assist with
identification and SBRC requested Local
Records to obtain and submitfull descriptions.

Although details of many records are still
awaited, ithas become apparent that many of
these published records were not fully
documented and the observers are not now
able to produce full details. Frominformation
submitted to SBRC we have been able to
accept the following records as showing the
characteristics of races belonging to the
Yellow-legged Gull group (michahellis,
cachinnans and atlantis):

1 adult Kirkwall, Orkney 20 April 1989 (N
Odin, M Gray, ER Meek). This record has
not previously been published.

1 adult Newton Shore, Ayr, Strathclyde 6
January - 12 July 1991 (R H Hogg et a/) and
what was presumed the same bird also at
Newton Shore, Ayr 3 April - 14 May 1994(A
A Murray eta/). This bird was evidently seen
in 1992 and 1993, but no details have been
submitted to SBRC.

1 adult/sub-adult (probably 3rd or 4th year)
Doonfoot, Strathclyde 1 August 1992 (A
Stevenson).

Several additional records have been
examined by SBRC but have been found to
be insufficiently documented to prove
identification conclusively. This has to a

98 RW Forrester

SB 18(2)

large extent been due to observers in the past
not being fully aware of the important
characteristics required to separate Yellow-
legged Gull from “omissus’, due partly to lack
of adequate identification criteria in field
guides.

There are, therefore, 3 Scottish records of
Yellow-legged Gull which are acceptable to
SBRC. We welcome details of any records
which have yet to be submitted to us.

Identification

Whilst there are slight differences, e.g. size
and outer primary pattern, no clear features
have yet been established for separating the
races michahellis, cachinnans and atlantis

from each other in the field and we do not |

believe this should be attempted by anyone
lacking experience of the different races. We
therefore concentrate comment on the
characteristics used to separate Yellow-
legged Gull from Herring Gull.

In adult plumage, Yellow-legged Gull has
yellow legs, a dark mantle (similar in shade to
Common Gull L. canus), long extensively
black wing tips (with only a small white
subterminal mirror), rather a rounded head,
red orbital rings and a bright yellow bill with a
prominent red spot (Sometimes with black
subterminal markings in winter). Note,
however, that bare part colours fade in winter
and the legs in particular often become a
pale, washed out," creamy yellow. Michahellis
is generally slightly larger and bulkier than the
argenteus race of Herring Gull, but Yellow-
legged Gulls from the west coast of Iberia
may be smaller and more similar to Lesser
Black-backed Gulls in size and proportion.
From late summer to early winter, most adults
have very subdued head streaking and often
look white headed from a distance, at a time

when argentatus and argenteus races of
Herring Gull have heavily streaked heads.
However, atcloser range, Yellow-legged Gulls
often show a small grey smudge over and
behind the eye and pale grey streaking over
the crown and nape. This difference in head
pattern usually breaks down from December
onwards as both species acquire summer
plumage. Yellow-legged Gull moults its
primaries on average a full month earlier than
Herring Gull and this can lead to autumn
individuals having longer primaries than the
local Herring Gulls.

First year birds resemble Lesser Black-backed
Gulls but, after their post juvenile moult, are
largely white headed and whiter on the
underparts, are pale above, chequered darker
(thus recalling first year Great Black-backed
Gull L. marinus), have acontrasting black bill,
pale edged solidly dark tertials (with
sometimes some subterminal mottling), a well
defined black tail band and a dark band along
the greater coverts. Note that, like Lesser
Black-backed Gulls, the leading underwing
coverts are dark chocolate brown, contrasting
with greyer under greater coverts and under
remiges (Herring Gull is more uniformly paler
brown on the underwings). First year birds
start to acquire grey feathering on the back
and scapulars as early as April and, by late
May, they have generally acquired a
completely dark “saddle” (back and scapulars)
and so are much easier to recognize.

Second year birds also show the dark grey
“saddle”, have a white head and underparts
and a black tail band which contrasts with the
white tail base and rump. The overall effect
thus recalls a giant first year Common Gull.
The underwing pattern is similar to first year
(second year Herring Gulls are usually whiter
in this area). The bill usually remains black
but gradually turns greeny at the base before

1995

becoming yellow with black subterminal
markings by second summer.

Third year birds resemble adults but with
traces of brown immature feathering on the
primary coverts and maybe the rest of the
wings and tail.

The callis distinctive, with male Yellow-legged
Gulls having avery loud and deep call, strongly
recalling Great Black-backed Gull.

Occasional individual gulls have been
recorded showing most of the characteristics
of adult Yellow-legged Gull but with pink legs.
These are probably either variant Yellow-
legged Gulls with pink legs, or Herring Gulls
showing plumage convergence with Yellow-
legged Gull. Until this position is clarified,
with our current state of knowledge, it is,
therefore, preferable that leg colour is clearly
seen by any observer claiming a record of this
species from an area where it is less than
common, e.g. Scotland.

Adult “omissus” birds have yellow legs and
red orbital rings, as have Yellow-legged Gulls.
They may also have more black on the
primaries than pink legged argentatus, thus
looking a little more like Yellow-legged Gulls.
Although “omissus” do have slightly darker
backs/upper wings, they are likely to resemble
argentatus Herring Gulls in all other
characteristics. With our present state of
knowledge, immatures are probably
indistinguishable from ordinary argentatus
Herring Gulls.

Discussion

Although 3 records have now been accepted
the status of Yellow-legged Gull in Scotlandis
not yet clear. This is largely as a result of
observers’ previous lack of understanding of
the existence of different populations of

Yellow-Legged Gulls in Scotland 99

Herring/Yellow-legged Gulls with yellow legs
and the knowledge of how to distinguish
them. This situation is now rapidly changing.
Whilst published information on identification
has until recently been inadequate, this
situation is improving.

SBRC has now included Yellow-legged Gull
in the list of species which require to have full
descriptions completed prior to acceptance.
lt is, therefore, most important that observers
are aware of the need to fully document any
claimed Scottish record. In addition, SBRC
would welcome full descriptions of any
omissus birds seen to enable us to better
understand the status in Scotland of the
different yellow legged forms. Observers
should endeavour to obtain as detailed a
description as possible.

Acknowledgements

We wish to thank Keith Vinicombe, Alan Dean
and Rob Hume who very kindly assisted us by
commenting on most of the documented
Scottish records, by providing details of key
features to use in the assessment of records
and by making very helpful suggestions which
improved an earlier draft of this paper.

References

Baxter E V & RintoulL J 1953 The Birds of
Scotland Oliver & Boyd, Edinburgh.

Cramp S & Simmons KEL 1983 The Birds
of the Western Palearctic Vol3 pp815-837,
OUP, Oxford.

Harris A, Tucker L & Vinicombe K 1989 The
Macmillan field guide to bird identification
Macmillan, London.

Heinzel H, Fitter R & ParslowJ 1995 Birds
of Britain & Europe Collins, London.

Thom VM 1986 Birds in Scotland Poyser,
Calton.

“Scottish Birds” and “Scottish Bird Reports”
“Ayrshire Bird Reports” 1980, 1982 & 1988.

100 R W Forrester SB 18(2)

Ronald W Forrester, Secretary, Scottish Birds Records Committee,
31 Argyle Terrace, Rothesay, Isle of Bute PA20 OBD

Revised manuscript accepted October 1995

Peregrine Don Smith

Se TE

Scottish Birds (1995) 18: 101-121

101

SHORT NOTES

Raptor evasion and defensive behaviour by Barnacle Geese

On 5 November 1994 at Caerlaverock
Wildfowl and Wetlands Trust Reserve,
Dumfries & Galloway, | was observing, from
a one-man hide, a flock of c.1500 Barnacle
Geese Branta leucopsis 6Ometres from me.
The birds were feeding in the usual dispersed
pattern on improved pasture, partly inundated
with large pools caused by heavy overnight
rain. A flock of c.40 Wigeon Anas penelope
was feeding in a similar pattern on one of the
flooded pools among the Barnacle Geese.

Suddenly an immature female Peregrine Falco
peregrinus appeared, and made alow banking
stoop in an attempt to catch a Wigeon, all of
which rapidly took flight to the nearest large
water body some c.40m away, leaving one
female on its own in the field. The Peregrine
banked round sharply at low level to dive on
this single bird. The Wigeon was decapitated.
The Barnacle Geese which had been closest
to the attack had left a radius of c.30-40m
around the Peregrine and its kill. The
Peregrine continued to eat its victim for about
10-15 minutes, after which it flew off.

The behaviour which followed | find most
extraordinary. The Peregrine, on leaving the
carcase, flew directly low over the heads of
the surrounding Barnacle Geese at a height
of less than 1m. The geese quickly formed
themselves into small circular groups of 20-
40 birds all facing inward toward the centre of
the circle, leaving large spaces between each
group. The Peregrine passed over very low,
without initially making any attempts at attack,
but returned about 3 seconds later at the
same height and made a vain attempt to
strike into 2 of these groups. The Barnacle

Geese had their necks at full stretch toward
the predator and on 3 occasions the geese
were also seen to stretch out their wings in
what seemed an effortto protect neighbouring
birds. This “attack” by the Peregrine on the
Barnacle Geese lasted 15 seconds. After this
the Peregrine flew off andthe Barnacle Geese
resumed feeding. Unfortunately | was
unaware if there were any juveniles in the
groups which came under aerial threat.

Asimilar incident occurred at the same site on
3 December 1994, probably involving the
same Peregrine. Whilst | was overlooking
c.3000 Barnacle Geese from the Avenue
Tower at Caerlaverock, an immature female
Peregrine spooked the flock into flight. It
seemed to single out its victim from the flock
and struck an adult Barnacle Goose in mid
air. This bird tumbled to the ground, somewhat
dazed, and stood alone in the field while the
Peregrine swung round and made 2 more
attempts at striking it. The goose faced the
flying Peregrine with its wings fully
outstretched and neck extended. This
defensive action was successful as the
Peregrine did not pursue the attack but sat on
a fence post about 30-40m away for about 5
minutes, watching the Barnacle Goose, which
did not attempt to fly away. After this the
Peregrine turnedits attention to other Barnacle
Geese which were feeding in a neighbouring
field. This flock of c70 included a family party
of 5 which was about 30m away from the main
body of geese. The Peregrine first made 2
attacks on the family party and these birds
adopted the circular grouping action
immediately, with the 2 adults, wings
outstretched, sheltering the 3 goslings

102 Short Notes

between them. After these unsuccessful
attacks the falcon turned its attentions to the
70 birds nearby. They immediately took up
the closed circular formation again, which
thwarted the attacks of the Peregrine.

After these incidents the Peregrine flew off
over the merse. It continued to chase other
birds including more Barnacle Geese, but

SB 18(2)

during my observations it was not successful.
This behaviour is very rarely seen at
Caerlaverock but has been observed prior to
1994 (P J Shimmings, pers. comm.). The
relevant volume of BWP does not describe
any behaviour similar to this in Barnacle
Geese.

| thank T W Dougall for commenting on the
first draft of this note.

David J Patterson, Wildfowl & Wetlands Trust,
Eastpark Farm, Caerlaverock, Dumfriesshire DG1 4RS.

*

| Wa?
;

Barnacle Geese

Accepted June 1995

Bobby Smith

WES)

Short Notes 103

Sparrowhawk

Sparrowhawk forced to land in sea

On Saturday 4 March 1995, at the mouth of
_the Cromarty Firth, we saw a male
_ Sparrowhawk Accipiter nisusflying low above

tne water and being harassed by up to 20
Common (Larus canus) and Herring § (L.
argentatus) Gulls. The hawk was forced to

_ land in the water half way across the Firth

which is cikm wide at this point.

Angus Hogg

The bird was not harassed by the gulls and
seemed to be buoyant while on the water. It
remained on the water for c. 2mins before
taking off but was then repeatedly forced to
land in the sea. It progressed by only a few
metres on each flight when it seemed to be
struggling to keep in the air. As the hawk
neared the shore, two Hooded Crows (Corvus
corone) joined the gulls in harassing it. Fifty
metres from the south shore of the Firth, the
Sparrowhawk took off one last time with great
difficulty and flew to the shore and into some
bushes with the crows still in pursuit.

Xavier Lambin, Julian Lambin and Beth Scott, Culterty Field Station,
Department of Zoology, University of Aberdeen, Newburgh,

Ellon, Aberdeenshire AB41 OAA

Accepted April 1995

104 Short Notes

SB 18(2)

Feeding rates of Hen Harriers in West Galloway

Despite a fairly large literature on Hen Harriers
Circus cyaneus in Britain, there are few
published observations of their feeding rates
throughout the breeding cycle apart from
details given on nestling feeding rates (Balfour
& MacDonald, 1970, Scott. Birds 6:157-166;
Watson, 1977, The Hen Harrier.
Berkhamsted; Picozzi, 1978, Ibis 120:498-
509). During observations of breeding Hen
Harriers in west Galloway their feeding rates
were recorded between 1968-88 at 15
moorland nests from the prelaying stage
through to post fledging and the results pooled
from all nests.

Table 1 gives details of feeding rates in four
stages: prelaying, incubation, nestling and
post fledging. Overall the total number of
prey deliveries was 129 times in 236.5hours
of observation (0.55 deliveries/hr) but prey
deliveries were less in the prelaying and
incubation stages (0.46 deliveries/hr) than in
the nestling and post fledging stages (0.69
deliveries/hr). In all stages, however, males
acted primarily as food providers and in the
prelaying stage (0.39 deliveries/hr) males
provided prey for the largely inactive female.
According to Simmons et al. (1987. Ornis
Scand.18:33-41) food atthis stage is probably
vital to successful pair bonding. Interestingly,

Table 1 Feeding rates of Hen Harriers in relation to the breeding cycle in

West Galloway, 1968-88.

Stage No. of No No. of
items hours prey
brought obs. deliveries/

hours

Prelaying 20 51 0.39

Incubation 51 101 0.50

Nestling 34 50:30) 0:67

Post fledging 24 34 0.70

Totals 129 236.30 0.55

Notes

(a) x2 = 0.67, 2d.f., NS
(d) x? = 0.64, 2 d.f., NS

Periods of days (deliveries)

Morning Afternoon Evening
No. hours No. hours No. hours

obs. obs. obs.
10 27.00 ¢% 14.00 3 10:00%a)
124s iss) 18: 37:00 - - 21 528%35eb)
13 19.30 2 9.00 19 22.00 (c)
6 11.00 1 830 11. 14.3075 (a)
41 92.55 34 68.30 54 75.05

(b) x2 = 5.08, 1d.f.,P<0.02 (c) x2=3.92,2df.,NS

1995

though, the rate increased by 28% during the
incubation stage to 0.50 deliveries/hr but the
difference in rates between prelaying and
incubation was not statistically significant and
could have arisen by chance (J. Watson in
litt.), although it may be an adaptive advantage
for males to feed females for as much as
possible before the young hatch so that a
greater proportion of the food can go to feed
the nestlings (see Cade. 1982. The Falcons
of the World, London). The rates continued to
increase in the nestling (0.67 deliveries/hr)
and post fledging stages (0.70 deliveries/hr)
when females assisted males in the provision
of food. Thus there seems to be a significant
difference in feeding rates between stages
with the highest occurring in the nestling and
post fledging stages (x2 =5.75, 1d.f. P<0.02).
Males continued to act as the main providers
in the nestling stage (no polygamous nests
were involved) in the ratio of 2:7 (74%). Inthe
post fledging stage, however, the roles were
reversed when females contributed more
deliveries than males in the ratio of 3:0 (75%).

Short Notes 105

Breckenridge (1935. Condor37:268-276) too
found that male Northern Harriers Circus
cyaneus cyaneus captured 80% of prey for
the young during the period of their greatest
growth and Picozzi found that prey brought to
nestlings by males ranged from 56-86%.

Table 1 also gives feeding rates in the separate
stages of the breeding cycle at different times
of the day but there seems to be no significant
difference in delivery rates except in the
incubation stage when there was more than
expected in the evening (x2 = 603, Ichi:
P<0.02).

In comparison with feeding rates in the nestling
stage in other localities, the rates recorded in
west Galloway are lower (Table 2). Simmons
et al. point out that feeding rates between
populations in North America and Europe are
not comparable as different sized prey are
involved. In general, though, feeding rates in
the incubation and nestling stages are lower
in Britain than in North America (Table 2).

_ Table 2 Comparison of feeding rates of Hen Harriers in Scotland and North

America

Locality Prelaying Incubation

Orkney - -

East Galloway - -
_ Kincardineshire - -

_ North America :

West Galloway —0.39 0.50

Nestling Post Fledging Source
0.82 - Balfour &
MacDonald, 1970
0.94 - Watson, 1977
0.91 - Picozzi, 1978
3.00 - Simmons et al
1987
0.67 0.70 This study

106 Short Notes

SB18(2)

Table 3 ‘Runs’ of prey deliveries to Hen Harriers nests in West Galloway

1968-87.
Date Nest No.of items
stage brought

Slitets: BE 2
19.6.70 N 3
e776 N 2

17 ff S N 3
29.7.84 N 3

72a N33) | 2
30.5.87 | 2

Interval Equivalent to
No. prey
of mins. deliveries/hour
15 8.0
30 6.0
30 4.0
30 6.0
20 9.0
30 4.0
iF 6.0

l= incubation stage, N = nestling stage, PF =post fledging stage

However, evidence of some high individual
delivery rates were obtained in west Galloway.
On 7 dates only, delivery rates at 6 nests were
particularly high, equivalent to 6-13 times
above the average rate (Table 3). This could
have indicated that there were ‘runs’ of prey
deliveries similar to that found by Knapton &
Sanderson (1985, Canadian Field Naturalist
99:375-377) in Canada when passerine
nestlings were brought to a Merlin’s Falco
columbarius brood in very short periods of
time. Nestling prey possibly did not figure

largely in prey deliveries to Hen Harrier nests
in west Galloway. On only one occasion
during these observations were three callow
nestling Sky Larks Alauda arvensis found at
a nest. Donald Watson (in /itt.) has also
recorded high delivery rates to nests in east
Galloway and, on three occasions, recorded
deliveries of 4 items in one hour and 2 items
twice in 27 min. This is equivalent to 4.0
deliveries/hr; about four times his average
rate.

| thank Donald and Dr Jeff Watson for their
useful comments.

R C Dickson, Lismore, New Luce,
Newton Stewart DG8 OAJ.

Revised manuscript accepted September 1995

1995

Short Notes 107

Aspects of the breeding biology of the Common Buzzard in Glen Roy

From 1986 to 1991, and also in 1994, |
studied territory distribution and occupancy,
nest sites, reproductive success and food of
Common Buzzards Buteo buteoin Glen Roy,
Highland. The main study area ran from Roy
Bridge to Brae Roy and covered 29km2. An
altitudinal boundary of 350 metres above sea
level was also used, as Buzzards did not
nest, and rarely hunted, above this height.
Within the study area, 5 territories were located

(Fig. 1).

In general, raptor territory size is related to
food density (Schoener 1966 Ecology49:123-
141). In Glen Roy, territory 3 covered nearly
twice the area of the other territories, and was
alone in having deer forest as the predominant
habitat type, suggesting that food was sparser
in deer forest. Occasional non breeding and
territory vacancies in territories 3,4, and 5
were not obviously related to ecological
factors, but territories 1 and 2, which were
both vacant from 1991, had seen noticeable
ecological change. Both were dominated
(65% and 70% of the area) by plantation
forestry planted in the late 1970s. After about
10 years, plantations in this area reach a
thicket stage in which Buzzards cannot hunt,
and it is likely that the development of these
Sitka spruce Picea sitchensis trees resulted
in the loss of the 2 Buzzard pairs.

-The maximum population density (1pair/
5.8km2) was apparently low for the west of
_ Scotland (Moore 1957 Brit. Birds50:173-197;
Brown L 1976 British Birds of Prey Collins,
_ London) and was probably determined by
nest site and/orfood availability (Newton 1979
Population Ecology of Raptors Poyser,
London). In territory 3 there were few nest

sites, and the 1.5m depth of one nest
suggested many years of use. Greater nest
site potential existed in the other territories,
with broad leaved woodland offering the best
opportunities for nest sites. However, the
wide range of nest site locations used
suggested that focd availability was probably
more important as a determinant of territory
density and distribution.

Twenty one nests were located in the 5
territories, an average of 4.5 nests per pair.
This is a similar figure to that reported for
Buzzards elsewhere (Tubbs 1967 Brit. Birds
60:381-394; Picozzi and Weir 1974 Brit.
Birds 69:199-210). Seventeen of the nests
were typical tree nests, with one each in
Scots Pine Pinus sylvestris, Alder Alnus
glutinosa, Birch Betula pubescensand Rowan
Sorbus aucuparia and the remainder in Oak
Quercus spp, the most abundant large trees
in the study area.

Of 26 pair years, 9 produced one young, 10
produced 2 young and 2 produced 3 young.
In each of 3 years, predation accounted for
the loss of 2 eggs, 2 eggs and 2 young, and
in 2 years a pair failed early in the season, all
in the lower glen. The average number of
young fledged per breeding pair/year was
1.35, less than the 1.7 found by Tubbs (1972
Bird Study 19:96-104) for pairs in the Scottish
Highlands, but similar to the national average
of 1.37 (Brown L 1976 British Birds of Prey
Collins, London). Pair 2 had the lowest
productivity, and pair 3 were consistent in
fledging 2 young in each year that they bred.
However, fledging success did not differ
significantly between either pairs (Kruscal
Wallis, H=5.0 adj. for ties, df=4, p=0.2888) or

108 Short Notes SB 18(2)

Figure 1 Map showing principal study area and Buzzard territory distribution.

<0

c
: -
ae
£ @|:

>I.
/ Fr

PEAN

2 Approximate boundaries of buzzard territories.
2 Numbering corresponds to that used in the text.

_—7 _ Altitude above sea level, in metres.
“100 Contour interval is 100 metres.

(SEE)

Short Notes 109

years (Kruscal Wallis, H=2.1 adj. for ties,
df=6, p=0.910). Reproductive output,

measured as the number of young fledged
per known laying pair, was negatively
correlated with total rainfall, measured at
nearby Fersit, in May in each year (r=-0.717,
df=6, p<0.05). This result is consistent with
the finding of Kostrzewa and Kostrzewa (1990
Ibis 132:550-559).

Absolute numbers of prey items recorded
were low, but the range of species identified,
mostly small birds and mammals, was similar
to that found by Swann and Etheridge (1995
Bird Study 42:37-43) in nearby Glen Urquhart.

Observations of Common Buzzard breeding
biology in Glen Roy are in keeping with results
from other parts of Britain, and demonstrate
the versatile nature of this species.

Dominic Sargent, ‘Cruach Innse’, Roy Bridge,

Ss

Revised manuscript accepted October 1995

Inverness-shire PH31 4AJ

110 Short Notes

SB 18(2)

Arctic Skua incubating own egg and Common Gull eggs in Common Gull

nest

The 75 hectare island of Eynhallow, Orkney,
has breeding populations of around 15 pairs
of Arctic Skuas Stercorarius parasiticus and
200 - 300 pairs of Common Gulls Larus
canus. Generally, the Common Gulls nest
near the shore while the Arctic Skuas nest on
short Heather Calluna vulgaris and lichen on
the inland slopes.

Arctic Skua nests are normally unlined
depressions in the heather with 2 eggs, so, in
1994, we were surprised to find a grass lined
nest, clearly that of a Common Gull with 2
Common Gull eggs in it as well as one Arctic
Skua egg. All were warm. Arctic Skua eggs
are similar in size to Common Gull eggs,
although they are less pointed in shape, and
much darker in colour, with a silky lustre as
opposed to the matt surface of the gull eggs.

We retreated and presently watched a dark
phase Arctic Skua land at the nest and begin
incubating all3 eggs. There wereno Common
Gulls in the immediate vicinity, although gulls
could be heard calling from a colony on the
nearest shore, around 200 metres away. The
skua was still seen to be incubating later inthe
day.

Furness (1987) stated that, due to skua nests
being well spaced and territories vigorously

defended, the chances of a foreign egg getting
into a skua’s nest are small. Both cases he
had observed himself related to Great Skua
Catharacta skua with 3 eggs, all from that
species. Crawford (1974) reported an incident
of a South Polar Skua Catharacta maccormicki
incubating the egg of an Adelie Penguin
Pygoscelis adeliae in its own nest; it was
suggested that the skua may have stolen this
egg to eat but, when at its nest, accidentally
treated it as if one of its own eggs. We thought
that on Eynhallow a Common Gull nested
away from the normal colony, inside the
territory occupied by a pair of Arctic Skuas,
and rather than a foreign egg getting into a
skua nest the opposite occurred. The fact
that the gull nest was lined with grass indicates
that the gull was the original owner of the nest
site. The skua perhaps originally intended to
predate the eggs but became confused on
alighting by the nest, eventually to the point of
laying one of its own eggs in it.

References

Crawford R D 1974 Incubation of an Adelie
Penguin egg by a South Polar Skua. Notornis
21: 262-263

Furness R W 1987 The Skuas, T & A. D.
Poyser, Staffs, England

Paul A T Doyle, Culterty Field Station, University of Aberdeen,

Newburgh, Ellon, Aberdeenshire AB41 OAA

Susan J Gough, British Trust for Ornithology,

The Nunnery, Thetford, Norfolk IP24 2PU

Accepted June 1995

1995

Roseate Terns return to the Isle of May

During the 1930s, 1940s and early 1950s up
to 15-20 pairs of Roseate Terns Sterna
dougallii bred on the Isle of May, Firth of
Forth, although in most years the total was
only 2-5 pairs (details in Eggeling, 1960, The
Isle of May). A single pair nested
unsuccessfully in 1956 and the species then
abandoned the island. This exodus coincided
with the disappearance of the colonies of
Common Tern S. hirundo, Arctic Tern S.
paradisaea and Sandwich Tern S.
sandvicensis. Common and Arctic Terns
recolonised the Isle of May in 1979 and 1984
respectively (Wanless, 1988, Scottish Birds
15:1-8) and numbers of both species have
increased substantially since. In 1995 about
800 pairs of “Comic” terns nested. A single
pair of Sandwich Terns hatched a chick in
1990, but breeding was probably unsuccessful
(Isle of May NNR Warden’s Report).

_ Since 1956 Roseate Terns have become a
_ very rare species in Scotland; an estimated
9 pairs bred in 1994 (Walsh, Brindley &
Heubeck, 1990. UK Conservation No. 18)
and until recently the species was recorded
only rarely on the Isle of May. However, since
1989 birds have occasionally visited the Isle
of May tern colony. The Isle of May NNR
Wardens’ Reports document the following

Short Notes 1117

records: a bird on the ground in June 1989,
1-2 birds on 4 dates in May 1990, 2 birds in
1991, 2 singles in 1992, birds sporadically,
and including a displaying pair, in June 1993
and up to 3 birds, again including a displaying
pair, in June 1994. The latter pair were seen
clearly and neither carried a ring (J Calladine
pers. comm.).

A Roseate Tern was seen in the tern colony
in June 1995 but apparently disappeared.
However, on 16 July, while walking up the
main track, we were mobbed by atern making
an unusual call. The bird landed nearby to
reveal itself as an adult Roseate Tern. Not
only that but it was ringed! A telescope was
set up in a (successful) attempi to read the
ring number, and, as we struggled with the
minute numbers, a small spiky downed
Roseate chick emerged from behind a rock
and pecked at the adult’s beak. Later a
second adult arrived and fed the chick. This
adult also carried a ring. Sadly, the chick was
not seen after it was about 2 weeks old.

We anticipated that these birds would have
originated in east Britain. However, both had
been ringed at Rockabill, Dublin (371km
away), whichis nowthe maincolony in Ireland
and Britain. One had been ringed as a chick
in 1989, the other as a chick in 1992.

M P Harris, Institute of Terrestrial Ecology,

Hill of Brathens, Banchory, AB31 4BY
N Harding, Scottish Natural Heritage,
46 Crossgate, Cupar, Fife KY15 5HS

Accepted September 1995

112 Short Notes

SB 18(2)

Presumed feeding movements of auks offshore from Portnahaven on the

southwest coast of Islay

The window of my house overlooks the lower
part of Eilean Mhic Coinnich (MacKenzie
Island) and the Atlantic where it meets the
North Channel. In December 1993, | noticed
large numbers of auks flying south nearly
every day. Rather than counting for long
periods, | counted for 5 minute periods
approximately every 30 minutes. By January
1994 it was apparent that birds were flying
south during the period of a rising tide and, in
order to verify this, | made numerous
observations throughout ebbing tides, when |
saw only occasional birds flying south.

Birds began to fly south just before Low Water
at Orsay Island at a rate of around 25 per
minute, increasing in number at Low Water to
around 50 per minute and continuing for a
further 4-5 hours. Peak numbers were usually

-
Rh a

Strong Race

\\
4
\

\

\\

at 1-3 hours after Low Water when they have
reached 50 per second. They have frequently
been 1000 per minute but have ranged from
100 to1000, regardless of wind direction or
speed. These numbers were obtained by
counting for 5 minute periods and dividing by
3

During one particularly strong movement |
asked Tristan ap Rheinallt, Bird Recorder for
Argyll, to observe the exceptional number of
birds from my window at Lagmore House.
After doing so, he went to Rubha na Faing
approximately 1 mile (2km) to the north from
where he carries out regular seawatches. He
noted that only few birds could be seen from
that point but, from a high vantage point at
Claddach half way between Rubha na Faing
and Lagmore, he could see that the birds

Rhinns of Islay

Orsay Rhinns Point

U3)

were flying in towards Orsay Island from well
offshore. We both also noted that birds could
not be seen from Port Wemyss overlooking
the southeastern point of Orsay Island and
Rhinns Point.

From mid March 1994 the number of birds
seen flying south decreased considerably
and from the end of April to December 1994
| was absent from Lagmore House. In
December 1994 my observations began again
andconfirmed the same pattern up to the time
of writing: mid March 1995.

As the total number of birds seen flying south
in one day has been so large (up to 100,000+)
| wondered if they were feeding in the strong
tidal stream which flows behind Orsay Island
up to the West Bank, some 4 miles (8km) to
the northwest. | also wondered whether |
might be seeing the same birds on more than
one occasion. On 4 March 19951 noticed that
auks were flying much closer to shore than
usual in a northwest wind so | went up to Port
Wemyss to see if they could be seen rounding
Orsay Island. On this occasion the birds were
Clearly visible and could be seen landing on
the water where the tidal stream begins.
Presumably they more usually alight slightly
further south of Orsay where they cannot be
seen from Port Wemyss.

Short Notes 113

At Spring Tides the tidal stream flows
northwest and then northnorthwest at about 6
knots (Gilbert MacLellan and other local
fishermen). Thus, presumably, the birds are
feeding on the rising tide in heavy overfalls
and when they reach the end of the tidal
stream near the West Bank, fly back to the
start again. Certainly large numbers of gulls,
Gannets Morus bassana and Fulmars
Fulmarus glacialis can be seen gathering
over this tidal stream after Low Water, which
suggests that it is a rich feeding area.

During the winter months, at Orsay, there is a
period of rising tide during the hours of daylight
on all but the 2 days around each Neap Tide.
On these days surface currents and tidal
streams will be at their minimum and the
height at High Water also at its minimum, so
that birds can feed during daylight and will not
be moved any great distance. The numbers
of birds observed at Neap Tides have been
less than 50 per minute. Onall ofthe remaining
24 or 25 days of the lunar cycle, birds have
been seen flying south in considerable
numbers with the exceptional numbers
usually, but not always, occurring during
Spring Tides when the tidal stream is at its
maximum rate.

| would be interested to know whether any
similar feeding movements have been
observed elsewhere.

Jane Dawson, Lagmore House,
Portnahaven, Isle of Islay

Accepted June 1995

114 Short Notes

SB 18(2)

Successful breeding of Bluethroat in Scotland

On 17 July 1995 a gamekeeper and his son,
who were repairing a grouse butt on an area
of heather moorland in Highland Region, heard
an unfamiliar call. This turned out to be from
a male Red-spotted Bluethroat Luscinia
svecica. |\t became apparent that the female
was also present and that they were carrying
food. After a brief search the nest was located
and found to contain 3 well feathered young.
Although the nest was only some 10 metres
from the butt, the birds seemed unconcerned
at the activity and continued to feed the young
throughout the rest of that day. Not being
sure who to contact, the gamekeeper told
another local man who got in touch with the
RSPB.

On 20 July | was taken to the site and, aftera
couple of minutes, the birds appeared. Having
seen Bluethroats on migration | was surprised
atthe male’s behaviour. When coming in with
food he would perch on top of a nearby bush
before flying down to the nest. The female
was always more circumspect and would
approach the nest at ground level. Inthe 15
minutes or so that we were there the young
were fed 4 times (twice by each parent: 2
crane-flies Tipulidae and 2 small caterpillars).
Two faecal sacs were removed, both by the
male. On a previous visit, white moths and
hairy caterpillars were also seen to be brought
in.

The site was revisited on 23 July. Two young
were found to have fledged and were being
fed by the parents. One young remained in
the nest. On 25 July all 3 young were being
fed out of the nest. | was unable to return until
1 August when | could find no sign of the birds
despite a fairly exhaustive search of the
vicinity.

The site seemed unusual in that it was located
on an east facing slope at an altitude of
c450m on an area of open Heather Calluna
vulgaris moorland with patches of Juniper
Juniperus communis; the nearest water was
some 200 metres distant. The nest itself was
constructed of woven grass stems and set in
a hollow under a small bank. Surprisingly it
was located in an area of c200m2 which had
been burnt for grouse management in 1994.
The only living vegetation within 5 metres of
the nest were some little patches of unburnt
heather and some regenerating blaeberry
Vaccinium myrtillus; a burnt juniper bush lay
close by.

This is the third proven breeding of Bluethroats
in the UK but only the second successful one
(Brit. Birds 61: 524-525; 81:118-119). Ason
the other occasions, it followed a spring when
above average numbers of migrant
Bluethroats were seen in Scotland.

| would like to thank the finders and the other
people who contacted me; unfortunately
confidentiality prevents me from naming them.

Stuart Benn, RSPB, Etive House, Beechwood Park, Inverness I1V2 3BW

Accepted September 1995

1995 Short Notes 115

Juvenile Bluethroat in Inverness-shire

_ At 9.00 hours on 7 August 1995 whilst mist
netting in an estuarine reedbed beside the
_ Beauly Firth at Lentran, Highland Region, |
_ was surprised to discover a Bluethroat
| Luscinia svecica in the net. This bird, which
was also seen and photographed by Graeme
| Prest, was identified as a first year male ina
| fairly advanced stage of post juvenile moult.
_ The wing feathers were unmoulted and most
ofthe juvenile body feathers had been moulted
with some new feathers still growing. Some
pale juvenile feathers on the crown had not
yet been moulted and new ones were still in
pin giving it a very streaky appearance.

In Birds in Scotland V M Thom (1986) states
that the earliest recorded date for Bluethroats
| on autumn passage is 29 August with the
_ main passage between mid September and
| mid October. The early date of this bird and
the fact that most passerines do not normally
_ migrate over long distances whilst in active
_ moult suggest that this bird may have been
_ reared locally.

Malcolm I Harvey, Clach Bhan,
Loaneckheim, Kiltarlity,
Beauly, Inverness-shire 1V4 7JQ

Bluethroat David Mitchell
Accepted October 1995

116 Short Notes

SB 18(2)

Polyterritoriality in a population of Corn Buntings on North Uist

Polyterritoriality is the maintenance of 2 (or
more) spatially separate territories atthe same
time and occurs in only a few species of birds.
Three instances of polyterritoriality were
recorded during a 5 year study of a colour
ringed population of Corn Buntings Miliaria
calandra on North Uist. Two males were
polyterritorial in very similar circumstances in
1988. In both these cases, the male initially
defended just one territory with a single
incubating female and then took over asecond
one, previously held by another male. The
fate of the previous territory owners and the
circumstances of the takeovers were not
known. In both cases, at take over, a single
female was already present there and had
just failed in a first nesting attempt. Both
females nested again on the territory shortly
after the takeovers. The 2 males continued to
be seen on both territories for the remainder
of the breeding season. The 2 territories held
by one of the males were 600m apart and
separated by 2 other territories. The female

=

AN

Corn Bunting feeding young

y
\ ee:
; Ail
SG SEEDY
“ Soa =e
VAD] KX Fe Zz

CORN BUN TING
EDING YOUNG

on this male’s first territory failed to fledge any
young, while the female on its second
subsequently fledged 3 young. The 2 held by
the other male were 400m apart and separated
by a single territory. The female on this
male’s first territory fledged 3 young but the
female on the second failed to fledge any
young. The third instance of polyterritoriality
occurred in 1990. The male again initially
held one territory and then for a period of
about 3 weeks simultaneously defended a
second, 1500m from the first. The second
territory was previously unoccupied and no
females nested on either.

Polyterritorial males were rare despite the
apparent potential for males to increase their
productivity by taking over territories of other
males. The takeovers may be possible only
when the previous owner dies or is too ill to
defend the territory which could account for
the rarity of the phenomenon. However, ifthis
is the case, itseems odd thatin the 2 takeovers

—

SE
Cm

E Pickard

GRAMPIAN 1979

en

1995

that were recorded, it was not an immediate
neighbour that had taken advantage of the
situation. It has been suggested that
polyterritoriality in Pied Flycatchers Ficedula
hypoleucais ameans by which males deceive
females into thinking they are unmated
(Lundberg & Alatalo, 1992, The Pied
Flycatcher, Poyser). In this species, males
establish a second territory once a female
has chosen to nest on theirfirst territory. This
increases their chances of acquiring a second
female since females would be unlikely to
settle on the male’s first territory knowing
there was already a female present. Wood
Warblers Phylloscopus sibilatrix are similarly
polyterritorial, though late nesting females of
this species were thought to have little choice

Short Notes Tai?

but to pair with polyterritorial males, because
the few unpaired monoterritorial males
available defended territories in unsuitable
habitat (Temrin, 1989, Anim. Behav., 37:579-
586). The situation is clearly different in the
present study since 2 of the polyterritorial
Corn Buntings took over territories that were
previously occupied by other males and
already had females settled on them. The
third polyterritorial male did establish its
second territory in a previously unoccupied
area, but did not have a female nesting on its
first territory. Although polyterritoriality is an
important part of the mating system of Pied
Flycatchers and Wood Warblers itis rare and,
apparently, more opportunistic in the Corn
Bunting.

Michael Shepherd, Scottish Natural Heritage,

2 Anderson Place, Edinburgh EH6 5NP

lan R Hartley, Dept of Zoology, University of Cambridge,

Downing Street, Cambridge CB2 3EJ

Accepted June 1995

Goldfinch feeding on seaweed

On 2 September 1994, my husband and |
were bird watching along the banks of the
_ River Forth at Culross. We saw a Goldfinch
_ Carduelis carduelis fly down to the shore
where it started to peck at a bright green
seaweed. We observed it more closely

through a telescope and confirmed that it was
indeed eating the seaweed. Small pieces
were taken in and not discarded. C J Henty
believes the seaweed to be a species of
Enteromorpha, probably E. intestinalis. BWP
mentions that the leaves of land plants are
eaten occasionally but quotes no records
involving seaweed.

Barbara Hay, 33 Springbank Gardens, Dunblane, Perthshire FK15 9JY

Accepted August 1995

118 Short Notes

SB 18(2)

Red backed Shrike breeding in Perthshire

InJune 1994 we were informed by a Perthshire
householder that he thought a Red-backed
Shrike Lanius collurio was present in his
garden. Identification was confirmed on 22
June by Alan Barclay, David Allan and Susan
White when 2 adult Red-backed Shrikes (male
and female) were seen.

With minimum disturbance a nest was located
on 1 July by DA who had observed the male
carrying food to the nest. The female took
flight as the nest was approached and 6
nestlings approximately 4 days old were noted.
The nest site was ina Broom Cytisus scoparius
3.5m in height intertwined with Bramble Rubus
fruticosus agg. with the nest 1.5m from the
ground. Habitat within the immediate vicinity
of the nest site was a dense thicket of Broom,
Blackthorn Prunus spinosa and Gorse Ulex
europaeus intertwined with Bramble. The
birds were observed hunting up to about
120m from the nest site in mainly scrub habitat
with a few scattered trees, although the area
also included several gardens and some short
grassland. The scrub in the hunting territory
was less tall and dense than at the nest site.

The 6 young were ringed on 5 July; no runts
were apparent and about 1-2mm of feathers
had emerged from the primary and secondary
pins of the young birds. The site was not
visited until 7 days after ringing.

Toconfirm fledging, observations were made
from a distance, on 9 days between 12 and 29
July. On 14 July 3 young birds were noted
perched ona shrub close to the nest site. At
different times on the same day both adults
were recorded, in each case followed by a
fledged young bird. A total of 3 young birds
was also observed on 18 and 29 July together

with the adult female. This was the maximum
number of fledglings recorded at any one
time. Two juveniles were noted on 26 July
hunting and feeding independently from a
fenceline perch; 29 July was the last date on
which a Red-backed Shrike was seen on
territory. The last sighting of the adult male
was on 14 July. Itis possible that he hunted
further afield, although searches were made
in suitable habitats outwith the nesting area.

On one occasion, mobbing of the shrikes by
Swallows Hirundo rustica was observed.
Repeated stoops were made by 6 or 7
Swallows at 3 fledged young shrikes perched
close together in the outer branches of a
Blackthorn. The young birds responded by
moving several cm closer towards the centre
of the bush. The Swallows then turned their
attention to the adult female which was perch
hunting from a nearby fence. The adult
shrike appeared to ignore the Swallows and
continued hunting by flighting down to the
ground and back to the fence. The whole
episode lasted for about 4 minutes.

The adults’ hunting behaviour usually involved
flighting to the ground from fences about 1m
in height, although tall trees and telegraph
wires were also utilised as perches. Prey
identification in the field was not possible
although on one occasion the adult female
was observedin hot pursuit of a Willow Warbler
(Phylloscopus trochilus). The outcome of
this chase was unknown as both birds dashed
into a dense shrub.

Prey species were identified by pellet
analyses. A single pellet was collected from
the nest rim at the time of ringing and a further

5 were collected from beneath the nest after —

=" ore aonwmnmnzszt wr wn =w

S- = “Ss ewes}. slU ere Ohl

ome

—

a =o

1995

Short Notes 119

the birds had left the area. The nest itself
contained food debris and was removed later
for analysis. The inherent dangers of pellet
analysis mean that some softer bodied prey
may be under estimated or missed. Insects
were identified to generic level. All 6 pellets
contained parts of Bumble bees Bombus spp.
2 pellets contained parts of Carabid beetles
Pterostichus sp, 1 pellet only contained parts
of Dung beetle Geotrupes sp. There were
small pieces of bone in 2 pellets; one also
contained aclaw which suggests a bird. The
nest also contained fragments of Bumble
bees and Carabid beetles as well as pieces of
Whirlygig Beetle (Gyrynus sp.) and a bone
from afrog Rana. Prey species identified do
not differ greatly from those observed by Ash
(J S 1970. Brit Birds 63:185-205, 225-239)
and (Bibby C 1973. Bird Study 20:103-110)
in southern England.

As aBritish breeding species the Red-backed
Shrike has been in decline for over 100 years.
Their last stronghold was in south east
England where they became extinct in 1989.
Rare Bird Breeding Panel Reports in British
Birds, however, record that breeding occurred
in at least one site in England in 1991 and in
1992.

Red-backed Shrikes have been seen in
suitable habitat in the breeding season in
Scotland between 1977 and 1992. In most
cases only single birds were noted or breeding
was not recorded. Breeding was proven in
Scotland, however, during the three summers
of 1977-79 and in 1992. It has been
speculated that these shrikes were birds of
Scandinavian origin that had drifted to the
‘wrong’ side of the North Sea during spring
migration.

Red-backed Shrikes did not return to this
(} study site, or to the surrounding area, in 1995.

We acknowledge with thanks: D. AllanandS.
White for field observations; J. Ralston (SNH)
and K. Baker (BTO) for swift processing of
Schedule 1 Licences; and, especially, those
understanding householders, who must
unfortunately remain anonymous, who
permitted ornithologists to view and study
their shrikes from underneath their shrubs
and behind their herbaceous borders!

A M Barclay, Scottish Wildlife Trust,
Loch of the Lowes,

Dunkeld, Perthshire PH8 OHH

BRS Morrison, 7 Dixon Terrace,
Pitlochry, Perthshire PH 16 5QX

Steven Brown

Red-backed Shrike

120 Short Notes

SB 18(2)

Willow Warbler feeding Mistle Thrush nestlings

On the afternoon of 7 July 1994, a Willow
Warbler Phylloscopus trochilus was seen by
Stuart Ross carrying food to a Mistle Thrush
Turdus viscivorus nestina Yew Tree Taxus
baccata in Glen Feshie. The tree was 25
metres from the front door. The nest was 20
metres from the base but, because of the
sloping ground, it was easy to see into the
nest with binoculars.

An adult Mistle Thrush was seen to land ina
clump of Birches Betula sp 100 metres from
the Yew from where it flew straight to the back
of the Yew and hopped down to the nest. At
this point the nestlings responded much more
energetically than for the Willow Warbler.
Their calls were unmistakable Mistle Thrush
‘churrings’ and the birds raised themselves
onto their feet. The adult deposited a large
food item into the nearest beak. It then flew
off towards a grain field. Eighteen minutes
later, the Mistle Thrush returned and was met
with the same enthusiasm as before. The
bird then flew off. During these 18 minutes,

the Willow Warbler had visited the nest 6
times, each time bringing food. It always
landed on the nest next to what seemed the
strongest nestling. However, the Willow
Warbler would then move along the line, and
it was often the second or third beaks which
received the food. On one occasion, it
removed a large faecal sac.

| watched the nest for a further hour, but the
adult Mistle Thrush did not return. Alarm
notes were heard once not far away, and an
adult was seen feeding in the field about 50
metres from the nest. However, during that
hour, the Willow Warbler continued its
attentions. It foraged mainly amongst the
leave of an Ash Fraxinus excelsior nearby.

On 8 July the nest was empty. On 9 July an
adult Mistle Thrush was seen on the ground
near the Birch clump. It was carrying food
when it flew off into a nearby wood, possibly
to feed fledged young.

P W Gossip, Briarhill, 7 Highfield Road, North Berwick, EH39 4BW

Willow Warbler

‘Stan da Prato |

——

1995

Short Notes / Correspondence

1217

Carrion Crows roosting on ground with Black-headed Gulls

While engaged in dawn and dusk surveillance
at Forfar Loch, Angus, intermittently
throughout March 1995, | noted that up to 47
Carrion Crows Corvus corone were nightly
sharing a ground roost site on a gravel
promontory in association with a roost of up
about 700 Black-headed Gulls Larus
ridibundus, about 350 Oystercatchers and
with a smaller number of waterfowl species
also in attendance.

Ground roosting by Carrion Crows is not in
itself uncommon. | have, for example, notes
on one of 243 birds roosting in heather on
South Vist and shore roosting by small groups
was regular in Fife and in the Solway area.

A significant aspect of the Forfar site was that
there are apparently many suitable roosting
trees including a closed canopy conifer stand
within c600 metres used by other species as

aroost. Clearly the ground roost was preferred
by the crows and there appearedto be aclose
association with the gulls. There was no
significant antagonism between the 3 main
species but it became obvious that the gulls
were much more alert and performed regular
minor ‘panic flights’, usually returning to their
original position within an average of 2.8
minutes. The exact causes of their alarm
could not be determined in all cases although
they were seen to respond to passing Heron
(3), Cormorant (1), Tawny Owl (1),
Sparrowhawk (2) and nearby factory noises
(17). Crows reacted to the gulls’ flighting with
increased calling by head raising and turning,
walking short distances, resuming a relaxed
posture when the gulls quietened.

It appeared that the crows found it
advantageous to roost in association with the
gulls.

John G Young, SNH, Area Officer, Angus and Dundee Districts
West Lodge, Lindertis, Airlie, by Kirriemuir, Angus DD8 5NP

Accepted June 1995

Correspondence

Roosting behaviour of Wheatear

| agree with Dott, (1995, Scott. Birds 18:60)
that “Roosting is an aspect of behaviour often
given scant treatment in species accounts.”

but it is not true that BWP Vol 5 (1988) “does
not detail roosting”. Outside the breeding
season “both sexes roost singly in grass or
heather tussock, burrow entrance, stone wall,
etc (P J Conder). Each bird uses several
roost sites Berk (1961)” (page 780).

Jeffrey Boswall, Biological Imaging, Derby University, Green Lane, Derby DE1 1RX

EEE ee eee eee ene

122 Scottish Birds (1995) 18: 122-127

OBITUARIES

George Mackenzie Dunnet

1928 - 1995

The Scottish Ornithologists’ Club made
George Dunnet an honorary member in June
1986. He was born in Caithness and went to
school at Peterhead. He was awarded first
class honours in zoology at Aberdeen
University, obtaining an excellent grounding
in knowledge of the whole animal kingdom.
He worked for his PhD degree at Aberdeen
on Starlings and then visited Oxford where he
attended seminars with Charles Elton and
Niko Tinbergen who were giants in the
formative years of the young zoologists of the
1950s and 1960s. The patterns of discussions
and seminars which evolved in Oxford spread
to Aberdeen as well as more widely to many
of the academic centres of the post war world.

George became Regius Professor of Natural
History at Aberdeen University, President of
the British Ecological Society, a long term
fellow of the Royal Society of Edinburgh, an
honorary Doctor of the University of Stirling,
and was awarded the OBE andCBE. Hewas
a distinguished academic, a notable teacher,
a gifted administrator and, above all, an
objective scientist. But it will be for other
things that his memory will linger long in the
minds of his fellow members of our club.
George was a friend to so many and this is
how we will remember him.

George Dunnet took first to Australia, and
then to Aberdeen, the informal yet critical

SB 18(2)

thinking of the Oxford era of the 1950s. Under
the guidance of Vero Wynne-Edwards, he
initiated studies of community ecology at
Culterty Field Station in the University of
Aberdeen at a time when the discipline was
burgeoning elsewhere. His approach was
characterised by the encouragement of young
zoologists working on their own projects to
develop a critical attitude to the development
of ideas and each other’s progress, criticizing
also their seniors’ ideas in a way that had not
always been encouraged by previous
generations of academic scientists. In this,
George’s role was humble; he criticised yet
advised, seldom forcing his opinions except
as a development of logical thinking. And in
all this, he took a great interest in the careers
of his staff and students, helping where he
could but always encouraging people to
improve their own thinking and enhance their
scientific development. Sociality was also
very important in this Culterty tradition, with
not only regular serious discussions but also
light hearted gatherings at the pub on Friday
evenings and frequent parties at Culterty
House, which was also George and Mom’s
(his wife) home. Mom’s kitchen at Culterty
provided not only physical warmth for visitors
from southern climes but also a welcome,
friendship and often useful advice, much
appreciated by newcomers faced with the
problems of a strange environment.

Stimulated initially by Wynne-Edwards and
Robert Carrick, George became increasingly
interested in seabirds and in marine
conservation. He will be remembered by
Scottish ornithologists as the first chairman of
the Seabird Group and for his co authorship
of 5 papers on Rooks, but mainly for his work
on Fulmars in Orkney on the uninhabited
island of Eynhallow. This programme was
low intensity, but very carefully planned to be
a long term study with aims and questions

123
David Jenkins

Us

Obituaries

~
D
=
=
=

Q
®
>
S
D

10)

124 Obituaries

SB 18(2)

which were scientifically important in their
own right but which also provided a basis for
introducing successive year classes of
students to the rigour of a well designed field
project. The study has continued in every
breeding season from 1950 to the present.
The initial aims were to determine for Fulmars
whether they were able to relay if the egg was
lost, whether they could breed in consecutive
years, and how long they lived. Later, their
general breeding ecology was studied, leading
finally to research on how new birds are
recruited to the breeding population and
whether Fulmars breed to the end of their
lives or become senescent. Many results
have been published in a series of papers that
are well Known. They include the facts that
Fulmars lay only one egg and do not relay in
the same year if the egg is lost, they are able
to breed in consecutive years, they begin
breeding at about 8 years old and breed for
over 30 years, usually pairing with the same
mate for life. This is one of the longest
running studies of seabirds in the world, with
data collected on over 1000 individuals at
over 1000 nest sites on Eynhallow.

George’s modest yet strongly supportive
leadership gained -him many friends and
Culterty became one of Britain’s foremost
ecological research stations. While it was
initially primarily ornithological, this role
changed to encompass work on fish,
mammals and invertebrates as well as the
modelling of community interactions. It was
in the primarily ornithological phase that the
SOC recognised George’s contribution to the
development of critical thinking among
Scotland’s young scientific ornithologists. It
was Characteristic of George’s approach to
ecology that, not only did he attract to Culterty
nearly 100 postgraduate students, about half
from overseas, but he also visited them
subsequently on their home ground and thus

acquired a worldwide perspective on
ecological and conservation problems. Partly
as a result, he also spent sabbatical years on
research and teaching in New Zealand and
Australia.

This international outlook on practical
ecological problems was combined with
considerable charm. He was always
supportive of friends and colleagues and was
greatly indemandas chairman of committees,
where he acquired a well deserved reputation
in guiding groups containing members likely
to maintain opposing views on difficult
questions. He was always objective and
much in demand within his university and in
the wider field of practical conservation. His
success derived from his ability to appreciate
the personal contribution which colleagues
could make to particular problems, to pick the
right people for each assessment, to evaluate
problems critically, always to be ready to list
to controversial views but to base policy
recommendations, particularly in wildlife
conservation, on the most objective evaluation
of the facts available.

In the late 1960s and 1970s George was
prominent in encouraging the oil industry to
promote an ecological awareness in the
developing offshore oil business. Early
problems were concerned with pipelines, and
he chaired the Aberdeen University
Environmental Liaison Group which, in 1977,
evolved into the Shetland Oil Terminal
Advisory Group which he was still chairing at
the time of his death. SOTEAG won the
confidence, not only of the oil companies and
Sullom Voe Terminal management, but also
of academics, leading to major symposia
involving the Royal Society of Edinburgh and
research on oil related problems near Sullom
Voe, including a major programme monitoring
seabirds throughout Shetland.

1995

George’s nonconfrontational style was widely
appreciated. Always approachable, he would
not shy away from difficult issues and his
_ strength was in finding a way forward which
_ left those concerned with protecting the
- environment, and those whose business it
_ was to create wealth by extracting oil from it,
both feeling that they could work togeiher.
His work for SOTEAG was specifically
mentioned in his OBE citation in 1986.

His experience in tackling difficult issues led
to his chairmanship of the Review Team on
_ Badgers and Bovine Tuberculosis, ofthe Fish
Farming Advisory Committee and of the
_ Salmon Advisory Committee. This last job
was particularly challenging and his success
ledtoaCBEin1994. Atthe time of his death
he was in Copenhagen, working with the
international group examining the
_ environmental impact of the proposed bridge
over the Kattegat between Denmark and
_ Sweden.

_George’s ability to apply his knowledge and
understanding of ecology to practical problems
was capitalised upon by the Nature
Conservancy andits successor organizations.
He served on the Scottish Committee of NC,
on NC’s Scientific Advisory Committee and
_ he was chairman of the Nature Conservancy
~Council’s Advisory Committee on Science.
He became a member of the Main Board of
Scottish Natural Heritage, chairman of its
Research Board and a member of the Joint
_Nature Conservation Committee. He always

Obituaries 125

tried to lead his committees to base their
policy recommendations on deductions and
predictions based on facts, wherever these
were available, and on probabilities when
facts were scarce or absent. In these
circumstances, he recognized the need for
intuitive approaches based on experience.
Nonetheless, throughout his career he had
always tried to ensure that scientific objectivity
informed debate about nature conservation,
even if, forsound pragmatic reasons, scientific
objectivity did not carry the day. Buthe was
the only scientist on the Main Board of SNH
and became increasingly concerned that the
voice of science was not heard in
circumstances when this would be reasonable
andihata nature conservation agency which,
in the past, had claimed to base its policies on
objective science was no longer doing so.
Accordingly, he gave notice of early retirement
from SNH in 1994 although his appointment
had more than another yeartorun. This was
a great disappointment to him, and the
decision was not reached lighily, involving a
considerable period of stress.

George's life was one of public service. This
was widely recognized and his personal
warmth, humanity and honesty, together with
his conviction that science based conservation
is an important ethic, led to much respect and
a very wide circle of friends. He is sadly
missed. On 22 September 1995, George’s
ashes were scattered on his beloved
Eynhallow, enriching forever this enchanting
island.

David Jenkins

126 Obituaries

SB18 (2)

William Gerald Harper
1915 - 1995

Bill Harper, who died in hospital in Edinburgh
on 14 September, will be remembered in
ornithological circles as Honorary Librarian of
the Waterston Library, a post he held with
considerable distinction and dedication for 18
years and for his pioneering work on radar
ornithology in the late 1950s.

Born in Brighton, Sussex, on 22 December
1915, during the First World War, he spent
the whole of his childhood in Brighton,
attending the local Grammar School with his
elder brother Stanley. They were both clever
boys, and both went on to Imperial College,
London, in Bill's case to study physics and
then optics for an M.Sc.

On leaving university he joined the Bermuda
Meteorological Service in 1937 and then,
during the war, he was in uniform and assisted
in providing weather services for the USAF. It
was in Bermuda that Bill met and later married
Hetty and where their two sons, John and
Geoffrey, were born in the latter war years.

The family returned to England in 1947, where
Bill had a short spell weather forecasting at
Heathrow, before being posted to Shetland to
be Director of the Meteorological Observatory
on the hill above Lerwick. The 5 years in
Shetland saw the awakening of Bill’s interest
in birds. Family holidays were spent on Fair
Isle, staying in the old naval huts then occupied
by the bird observatory.

In 1953, they left Shetland to settle in Luton,
Bedfordshire, for another 5 year period. Bill
was in charge of the British Meteorological

Office Radar Research Station which was
situated at East Hill, near Dunstable. It was
here that he was to conduct some important
work on radar ‘angels’.

Using radars much less powerful and
sophisticated than modern equipment, he
was able to observe ‘angel’ occurrences on
most operational days. Recordings of the
daytime density of ‘angels’ at this station had
been maintained since 1952, so that analysis
of these records permitted Bill to show that
the periods of intense ‘angel’ activity
corresponded to the times of spring and
autumn bird migration. Further conclusive
evidence was revealed by studying the
directions of movement of ‘angels’. This
research resulted in 2 important papers in
Nature (1957) and Proc. of the Royal Society
(1958). Unbeknown to him at the time, Ernst
Sutter had been studying the same mysterious
radar echoes in Switzerland and pipped him
to the post by publishing his results earlier in
1957.

The next move was to Malvern,
Worcestershire in 1958, where Bill worked at
the Royal Radar Establishment. He and
Hetty then spent 2 years in Aden and 6 in
Germany, finding time to make a number of
trips to Africa and in Arabia. In 1973 Bill’s
career meant a move to Edinburgh to be in
charge of the Meteorology Unit at
Corstorphine. He retiredin 1977 and thereafter
devoted much of his time to watching birds
and serving the SOC. He succeeded Irene
Waterston as Librarian in October 1977, a
post he held until this spring when ill health
forced him to resign.

Under his able custodianship the library
expanded greatly and reflected the large
increase in the publication of bird books over
the last 20 years. The number and range of

hone, Bisa

1995

journals were also considerably increased
and this section is without equal in Scotland.
With his extensive knowledge he was well
able to direct and assist any library user, be
they professional researcher or the club
member simply looking for information on the
birding opportunities of their chosen holiday
destination. His stand of second hand books,
sold for the benefit of library funds, was an
essential part of our conferences. He was

Bill Harper

Obituaries LAZ

always ably assisted by Hetty, who does
much valuable work in keeping the library in
good order and who did work in the Bird
Bookshop for many years.

His many friends will remember Bill as a man
of quiet good humour, sincere and honourable
and always willing to lend a hand when the
need arose. He was totally dependable and
he will be sadly missed.

David Clugston

Hetty Harper

128 Advice to Contributors

SB 18(2)

Advice to Contributors

Authors should bear in mind that only a small
proportion of the Scottish Birds readership are
scientists, and should aim to present their material
concisely, interestingly and clearly. Unfamiliar
technical terms and symbols should be avoided
wherever possible and, if deemed essential, should
be explained. Supporting statistics should be kept
to a minimum. All papers and Short Notes are
accepted on the understanding that they have not
been offered for publication elsewhere and that
they will be subject to editing. Papers will be
acknowledged on receipt and will be reviewed by at
least 2 members of the editorial panel and, insome
cases, by anindependentreferee. They willnormally
be published in order of acceptance of fully revised
manuscripts. The editor will be happy to advise
authors on the preparation of papers.

Reference should be made to the most recent
issues of Scottish Birds for guidance on style of
presentation, use of capitals, form of references,
etc. Papers should be typed on one side of the
paper only, double spaced and with wide
margins; 2 copies are required and the author
should also retain one. We are happy to accept
papers on Applemac computer discs. We cannot

handle other formats as both the SOC computers
and those at our printers are on the Apple system.
Please contact Sylvia Laing on 0131 556 6042 to
discuss this. Headings should not be underlined,
nor typed entirely in capitals. Scientific names in
italics should follow the first text reference to each
species. Names of birds should follow the official
Scottish list (Scottish Birds Vol 17 : 146-159). Only
single quotation marks should be used throughout.
Numbers should be written as numerals except for
one and the start of sentences. Avoid hyphens
except where essential eg in bird names. Dates
5 August
LIC eee ee but on the 5th (if the name of the
month does not follow). Please note that papers
shorter than c700 words will be treated as Short
Notes, where all references should be incorporated
into the text, and not listed at the end, as in full
articles.

Tables, maps and diagrams should be designed to
fit either a single column or the full page width.
Tables should be self explanatory and headings
should be kept as simple as possible, with footnotes
used to provide extra details where necessary.
Each table should be on a separate sheet. Maps
and diagrams should be in Indian ink and be
camera ready, but drawn so as to permit reduction
to half their original size.

Erratum - The status and distribution of wintering Pink-footed and Greylag
Geese in east central Scotland. Vol 18: 24-50

M V Bell & SF Newton, the authors, have asked us to point out that all references to km?
should be ‘one km square’ apart from p.27, paragraph one, last line. This change was made
at the proof stage. Unfortunately the authors did not see the proofs and consequently were
unable to correct this and other typographical errors. We very much regret this break down in
Our procedures.

Editorial changes

Scottish Birds has been edited by Mrs Anne-Marie Smout since 1991. This issue is the first
for which Dr Stan da Prato has been responsible. The new editor and the SOC as a whole wish
to thank Anne-Marie for her hard work and commitment to the journal. We also thank Dr Bryan
Nelson and Professors David Jenkins and Peter Slater for their valuable work on the Editorial
Panel. The new editor is particularly grateful that Bryan and David have agreed to stay on and
that Dr lan Bainbridge has agreed to join the Panel.

NEOTROPICAL
BIRD CLUB

Neotropical bird club launched

A club has been launched to promote the study and conservation of
the birds of the Neotropics (South America, Central America and the
Caribbean). It is currently seeking founder members to help reach
the launch budget of £2000, which is required to get the club running
and to publish the two first issues of its intended journal ‘Continga’.
Founder members will be asked to pay a minimum of £25, and will
be formally acknowledged in the first issue of ‘Continga’. ‘Continga’
will provide a colourful and much needed forum for exchange of
information on the avifauna of this extremely rich and diverse area,
and will contain papers and features on the birds and their
conservation as well as news of recent observations and discoveries
(at present, new species are still being discovered at the rate of
more than two a year). lt is hoped that in due course the club will be
able to provide direct funding and support for practical conservation
programmes.

For further details and membership forms,
please contact:

Rob Williams,

Publicity Officer,

Neotropical Bird Club,

c/o The Lodge,

Sandy,

Bedfordshire SG19 2DL

Scottish Birds
Volume 18
Contents

Main articles
Occupation patterns of lochs by Slavonian Grebes in Scotland.

Part 2 December 1995

R.W. Summers & R.A. Mavor 65
Winter diet of Ravens in Perthshire. W.A. Mattingley 71
Seabirds of Handa Island. J.G. Stoneman & N.A. Willcox 78
An analysis of prey taken by Merlins within the Lammermuir Hills,

1984-1994. A. Heavisides, |.R. Poxton & A.W. Barker 88
Yellow-Legged Gulls in Scotland. R.W. Forrester 95

Short Notes
Raptor evasion and defensive behaviour by Barnacle Geese. D.J. Patterson 101

Sparrowhawk forced to land in sea. X. Lambin, J. Lambin & B. Scott 103
Feeding rates of Hen Harriers in West Galloway. R.C. Dickson 104
Aspects of the breeding biology of the Common Buzzard in Glen Roy.

D. Sargent 107
Arctic Skua incubating own egg and Common Gull eggs in Common

Gull nest. P.A.T. Doyle & S.J. Gough 110
Roseate Terns return to the Isle of May. M.P. Harris & N. Harding a
Presumed feeding movements of auks offshore from Portnahaven

on the southwest coast of Islay. J. Dawson ti2
Successful breeding of Bluethroat in Scotland. S$. Benn 114
Juvenile Bluethroat in Inverness-shire. M./. Harvey LS
Polyterritoriality in a population of Corn Buntings on North Uist

M. Shepherd & |.R. Hartley 116
Goldfinch feeding on seaweed. B. Hay 1
Red-backed Shrike breeding in Perthshire. A.M. Barclay & B.R.S. Morrison 118
Willow Warbler feeding Mistle Thrush nestlings. P.W. Gossip 120

Carrion Crows roosting on ground with Black-headed Gulls. J.G. Young 121
Correspondence

Roosting behaviour of Wheatear. J. Boswall 12
Obituaries

G.M. Dunnet. D. Jenkins 122

W.G. Harper. D. Clugston 126
Advice to Contributors 128
Erratum

Wintering Pink-footed and Greylag Geese in east central Scotland 128

Editorial changes 128

Published by the Scottish Ornithologists’ Club,
21 Regent Terrace, Edinburgh EH7 5BT. © 1995

Printed by Milne, Tannahill and Methven Ltd., Perth

Scottish Birds

The Journal of the Scottish Ornithologists’ Club

Editor: Dr. S. da Prato

Assisted by: Professor D. Jenkins, Dr J.B. Nelson and Dr. |. Bainbridge
Business Editor: The Secretary, S.O.C., 21 Regent Terrace, Edinburgh EH7 5BT
(tel. 0131-556 6042).

Scottish Birds, the official journal of the Scottish Ornithologists’ Club, publishes original material
relating to ornithology in Scotland. Papers and notes should be sent to The Editor, Scottish Birds, 21
Regent Terrace, Edinburgh EH7 5BT.

Two issues of Scottish Birds are published each year, in June and in December. Scottish Birds is
issued free to members of the Scottish Ornithologists’ Club, who also receive the quarterly
newsletter Scottish Bird News and the annual Scottish Bird Report. These are available to non-
members at a subscription rate (1992) of £30.

The Scottish Ornithologists’ Club was formed in 1936 to encourage all aspects of ornithology in
Scotland. It has local branches which meet in Aberdeen, Ayr, the Borders, Dumfries, Dundee,
Edinburgh, Glasgow, Inverness, New Galloway, Orkney, St Andrews, Stirling, Stranraer and Thurso,
each with its own programme of field meetings and winter lectures. The Waterston Library at the
Club’s headquarters at 21 Regent Terrace, Edinburgh EH7 5BT is the most comprehensive
ornithological library in Scotland, and is available for reference during office hours (Monday to
Friday, 9.00 - 17.00 hrs). (But please phone beforehand). A comprehensive stock of Scottish local
bird reports is held at headquarters and may be purchased by mail order.

SOC annual membership subscription rates

AGUIE si cpiciccicsissanvert sacoanitensnaanveaddodgnwashnesswasenousceledeitadesansadbese: eee ean £15.00
Family (husband & wife, & children under 18) .2:..2.3:ernc.scc..0eneeeese2snes censor £22.00
Junior (under 18, or student Under 25)........:..:<.cccsssesedesccs-qc0snececse0s: ee £6.00
Pensiomer/UnWaged ..........ccscccccesssveeessasssandsdseseuumnens saunas osbaue de sqeeseegeeeee £8.50
Pensioner Family ............000cceccecontedbvonssisecdoussocadvucensusduesnosessvencests>e (arr £12.00
LAPS ose tages serie nnsnasdeh ve sass eaeond vsanbcdadeusaeeiceetardones ee scnciacue: eeeeeataa tee er £300.00
Life FAMly sec dcivivnstennessscchwansaceecevn cas teeeaceliagaxuon coe pemene saat tes tease ce ee eee £450.00

All subscriptions may be paid by Banker’s Order and Covenanted. Subscriptions paid this
way will greatly assist the Club.

Published by the Scottish Ornithologists’ Club, 21 Regent Terrace, Edinburgh EH7 5BT
Printed by Douglas Gauld & Co Limited, 22 Tannadice Street, Dundee DD3 7QH

Scottish Birds (1996) 18: 129-131

129

—

Amendments to the Scoitish list fire arene Mia.

of the Scottish List first published in 1994 This is the first subsequent
report of the Committee and contains additions, corrections and

modifications to the original list

Readers should refer to the introduction to
the full list (Scottish Birds 17:146-159) forthe
explanation of the principles used in
establishing it. These include following the
categories used by the British Ornithologists’
Union and also being guided by any decisions
which they make thatin turn affect the Scottish
List. In addition we will not add a species to
the Scottish List unless the record has first
been accepted by the appropriate assessment
committee.

Green-backed Heron Butorides striatus
add to Category A

Category D3 was incorrectly described in the
introduction to the Scottish List (Scottish Birds
17:146), and should correctly be defined as
covering “species which would otherwise
_ appear in category A or B except that they
_ have only even been found dead on the
_ tideline”.

There are 4 species which have only occurred
as corpses in Scotland: Green-backed Heron,
Red-necked Stint Calidris ruficollis, Cetti’s
Warbler Cettia cetti, and Black and White
Warbler Mniotilta varia. There is one record
for each species. Red-necked Stint, Cetti’s
_Warbler and Black and White Warbler were
all found away from the tideline and, therefore,
were placed in Category A. Green-backed
Heron was omitted from the Scottish List in
error. The bird was found near the tideline on

the shore at Tyninghame, Lothian on 25
October 1987, but was described as having
been found “freshly dead, probably killed bya
Fox, Vuipes vulpes (British Birds 81:542). It
apparently showed no signs of having been
washed ashore and therefore should be
placed in Category A.

Crested Lark Galerida cristata transfer
to Category B from Category A.

Buff-bellied Pipit Anthus rubescens transfer
to Category B from Category A.

In the original list we placed Crested Lark and
Buff-belled Pipit in Category A. There is only
one Scottish record of Crested Lark and 2
Buff-bellied Pipit records and all are prior to
1958. Both species should therefore appear
in Category B.

Hudsonian Godwit Limosahaemastica add
to Category A

A Hudsonian Godwit was seen near
Collieston, Grampian on 26 September 1988
(British Birds 83:463) and should therefore be
in Category A. This species was omitted in
error from the original list.

White Pelican Pelecanus onocrotalus delete
from Category D

This species was included in our original list

130 SBRC

SB 18 (3)

on the basis that there have been a few
published records e.g. two separate
wandering groups in May 1973 (Scottish Birds
8: 221). Records since 1958 require to be
accepted by the British Birds Rarities
Committee, who have now accepted 16
records, none of which are from Scotland. It
appears that no Scottish record has even
been submitted to them for acceptance and
therefore we require to delete this species
from Category D of the Scottish List. We
would urge observers of free flying birds in
Scotland to submit details to BBRC to enable
the species to be re-instated.

As a result of decisions in the Twentieth and
Twenty first Reports of the Records Committee
of the BOU (Ibis 136:253-256; 136:497), 3
amendments now require to be made.

Asian Brown Flycatcher Muscicapa
dauurica add to Category D1
Fair Isle, Shetland 1-2 July 1992

The BOU agreed to place the species in
category D1 pending further information on
patterns of vagrancy andits status in captivity.
Painted Bunting Passerina ciris delete
from Category D1

The BOU consider the species to be an
unlikely natural vagrant to Britain and the
species has been common incaptivity. They,
therefore, now do not consider it likely that
any have reached Britain naturally.

Daurian Starling Sturnus sturninus delete
from Category A and transfer to Category D1

The only accepted record was of a bird on
Fair Isle on 7-28 May 1985. The BOU now
considers that natural vagrancy to Britain is
not very likely and it is considered particularly
unlikely in spring. In the 1980s the species

occurred in captivity in Britain and on the
Continent, therefore escape would have been
possible.

There have recently been several first records
in Scotland of species previously recorded
within Britain, which have now been accepted
by the British Birds Records Committee, being
detailed in their annual reports which are
published in British Birds.
Lesser Scaup Aythya affinis add to
Category A

Male, Milton Loch, Dumfries & Galloway 29
December 1990 (British Birds 85:516). This
was the first of 3 accepted records in recent
years.

Red-necked Stint Calidris ruficollis add to
Category A

Juvenile recently dead, Fair Isle, Shetland 31
August 1994 (British Birds 88:512)

Grey-tailed Tattler Heteroscelus brevipes
add to Category A

Juvenile, Burghead, Grampian 27 November
- 27 December 1994 (British Birds 88:517).
Marmora’s Warbler Sylviasarda add to
Category A.

Male, in song, St Abb’s Head, Borders 23-27
May 1993 (British Birds 87:554).

Spanish Sparrow Passer hispaniolensis
add to Category A

Male, North Ronaldsay, Orkney 11-19 August
1993 (British Birds 87:561).

Pallid Harrier Circus macrourus § transfer to
Category A from Category B

The first of 2 1993 records was of a male,
Tayside (locality withheld) 5-7 May 1993
(British Birds 88:505). The only previously
acceptable Scottish record was of a male
shot on Fair Isle in 1931.

1996 Amendments to the Scottish List 131
ee a EE

The above changes bring the totals in the Among the decisions which are still awaited

various categories to: are those concerning Eastern Phoebe on St
Kilda and Blyth’s Pipit on Fair Isle, both
Category A 454 potential further additions to Category A.
Category B 15
Category C 6 The Scottish Birds Records Committee, which
is responsible for maintaining the Scottish
475 List, now consists of Ken Shaw (Chairman),
Category D 12 Peter Gordon, Eric Meek, Kevin Osborn, David
Clugston, Bruce Forrester and Ron Forrester
487 (Secretary).

Ronald W Forrester, Secretary,
Scottish Birds Records Committee
31 Argyle Terrace, Rothesay, Isle of Bute, PA20 0BD

Black-browed Albatross Malte Andersson

132 Scottish Birds (1996) 18: 132-133

SB 18 (3)

Records of species recorded in Scotland on 5 or fewer

occasions

PETER R GORDON & DAVID CLUGSTON

When compiling the Scottish List it became apparent that there was no
single source to which the Scottish Birds Records Committee could refer
for a summary of acceptable records of rare Scottish species. Following
the publication of the official Scottish List it therefore seems appropriate
that an annotated list should appear of those records which are considered
acceptable and on which the Scottish List is based

This first attempt at publishing a systematic
list of all occurrences of the rarest birds in
Scotland has been confined to those species
for which the SBRC can establish no more
than 5 acceptable records, published prior to
31st December 1995. E V Baxter & LJ
Rintoul 1953, The Birds of Scotland, Oliver &
Boyd, Edinburgh and V M Thom 1986, Birds
in Scotland, Poyser, Calton, did much
groundwork and documented most relevant
records, but the purpose of those publications
was rather different than to provide a list of
records. Inthis list SBRC has used the following
principles in deciding which records are
acceptable and which are not:

1 We have not accepted any record of a
species which does not appear on the British
List, which is compiled by the Records
Committee of The British Ornithologists’ Union
(BOU).

2 No record has been accepted by us which
would be the first British record unless it has
been accepted as such by the BOU.

3 Records since 1958 of species appearing
on the British Birds Rarities List must be
accepted by the British Birds Rarities
Committee (BBRC).

From this itcan be seen thatthe only decisions
which have been required on our part have
been in connection with pre 1958 records
and records since 1958 of birds that are rare
in Scotland but not dealt with by the BBRC.
For pre 1958 records, only those considered

acceptable at the time, and which have not

since been examined again and found

wanting, are included. There has been no |

further examination of records specifically for
this paper. These records have mostly been
well documenied, but we would welcome any
constructive criticism of any of the records
which we have accepted and also would be
pleased to learn of any records omitted for

which a strong case could be put forward for |
acceptance. All of the records since 1958 |

have been examined and accepted by the
BBRC, except for those concerning Lesser
Spotted Woodpecker, Cetti’s Warbler and
Dartford Warbler, each being represented by
a single acceptable record, plus Egyptian
Goose in the supplementary Category D list.

The categories are defined as follows:
Category A. Species which have been

recorded in an apparently wild state in
Scotland at least once since 1958.

age > -

1996

Category B. Species which were recorded in
an apparently wild state in Scotland at least
once up to 31 December 1957, but have not
been recorded subsequently.

Category C. Species which, although
originally introduced by man, have now
established a regular feral breeding stock
which apparently maintains itself without
recourse to further introduction.

Category D. Species which would otherwise
appear in Category A or B except that:

D1 there is a reasonable doubt that they
have ever occurred in a wild state, or

D2 they have certainly arrived with a
combination of ship or human assistance,
including provision of food and shelter, or
D3 __—sitthey have only ever been found dead
on the tideline, or

D4 species that would otherwise appear
in category C except that the feral population
may or may not be self supporting.

_ There are now only 15 species appearing in
_ Category B of the Scottish List. Of these
_ White-faced Petrel (1 record), Magnificent
| Frigatebird (1), Squacco Heron (4), Houbara

_ Bustard (1), Eskimo Curlew (3), Sooty Tern
| (2), Whiskered Tern (1), Black-billed Cuckoo
_ (2), Eagle Owl (4), Crested Lark (1), Buff-

| bellied Pipit (2) and Black and White Warbler
| (1) have been recorded on 5 or fewer

occasions and we therefore list all records for
| these species. There are more than 5 records
.) for Ruddy Shelduck, Capercaillie (the original

Species recorded in Scotland on 5 or fewer occasions 133

population became extinct during the
eighteenth century and the species was
established again from 1837, but there is no
record of any bird having reached this country
by natural means within the last 200 years
and the present Scottish population is
therefore Category C) and Great Auk.

References are generally given only for the
first record. In cases where the reference for
the first species does not contain a published
description, but a useful article describing a
subsequent occurrence is available, this is
cited as well. The taxonomy and
nomenclature used are as in the Scottish List
(Scottish Birds 17:146-159) which follows
that in the British Ornithologists’ Union’s
Checklist of Birds of Birtain and Ireland sixth
edition, 1992. This is based on K H Voous’s
List of recent holarctic bird species, 1977,
revised edition, BOU. We have, however, not
followed the BOU’s lead in introducing major
changes to the English names, inmost cases,
retaining those in more general usage.

Peter Gordon has been responsible for
researching and tabulating the records since
1958. David Clugston completed the same
task with the pre 1958 records. Now that this
initial compilation of records has been
published, SBRC intend, over a period of
time, to examine the pre 1958 records again
with a view to assessing their validity in more
detail. Itis hoped that this and future work will
lead to a more accurate understanding of the
status of rarer species in Scotland.

Ronald W Forrester, Secretary, Scottish Birds Records Committee

134 Scottish Birds (1996) 18: 134-143

SB 18 (3)

Systematic list

A Pied-billed Grebe Podilymbus
podiceps

1-8 October 1975, Carlingwark Loch,
Kirkcudbright, Dumfries & Galloway (Scottish
Birds 9:297-8)

9 January -27 March 1977, Loch of Strathbeg,
Grampian

8 June 1983 - 22 August 1985, Askernish,
South Uist, Western Isles

24-25 April 1987, Lochmaben, Dumfries &
Galloway

A Black-browed Albatross Diomedea
melanophris

A bird summered on the Bass Rock, East
Lothian 18 May - 28 September 1967, 13 April
- about 20 July 1968 and again 10 April - 3
May 1969 (British Birds 61:22-27)

Other sightings, probably relating to this bird,
are:

7 or 8 February 1968, at sea, 10-14 miles off
Eyemouth, Berwickshire

28 February 1969, 6 miles off St Abbs Head,
Berwickshire

23 August 1969, off Elie Ness, Fife

13 August 1969, Hoy, Orkney

What was probably the same bird was at
Hermaness, Shetland 21 July - 4 August
1972 andagain inthe summers of 1974-1994
(but not 1989 or 1990). Earliest and latest
dates there are 14 February and 20
September. Other sightings, the majority of
which are probably attributable to this bird,
are:

8 August 1972, Fife Ness, Fife (4 days after
last Hermaness sighting)

21 August 1975, adult, Scapa Flow, Orkney
(3 days after last Hermaness sighting)

27 July 1976, at sea, south of Lerwick,
Shetland

21 and 26 July 1978, Yell Sound, Shetland
16 March 1980, near Rockall, Western Isles
(the day after first Shetland sighting that
season)

7 August 1984, atsea, 10 km west of Corsewall
Point, Dumfries & Galloway (reported at
Hermaness from 27 February but no last date
given)

30 January 1990, sea area Viking 60°00’N
02°00’E

A Little Shearwater Puffinus assimilis
30 June 1974, off Islay, Argyll (Scottish Birds
9:380-1)

14 September 1985, Corsewall Point,
Dumfries & Galloway

29 July 1990, Butt of Lewis, Western Isles
9 December 1990, dead, Musselburgh,
Lothian

A Wilson’s Petrel Oceanites oceanicus
1 October 1891, caught alive, Jura, Argyll
(Annals of Scottish Natural History 1892:18)
30 August 1988, at sea, Little Minch, about 5 |
km SE of Renish Point, Harris, Western Isles
8 August 1993, at sea, about 50 km NW of
Ramna Stacks, Shetland

B White-faced (Frigate) Petrel
Pelagodroma marina

1 January 1897, young female, caught alive,
Colonsay, Argyll (Annals of Scottish Natural
History 1897:88)

B Magnificent Frigatebird Fregata
magnificens

10 July 1953, immature female, caught alive,
later died, Tiree, Argyll (British Birds 47:58)
Unidentified frigatebird off Forvie, Aberdeen
20 August 1960

Sa

1996

A Green-backed Heron Butorides
striatus

25 October 1987, first-winter, freshly dead,
Tyninghame, Lothian (Lothian Bird Report
1987:80)

B Squacco Heron Ardeola ralloides

9 October 1852, shot, near Stockton, Forth-
Clyde Canal, Lanark (D A Bannerman 1957
The Birds of the British Isles 6:97 Oliver and
Boyd, London)

pre 1877, shot, Dalmahoy, Midlothian

7 September 1896, adult male, shot, North
Ronaldsay, Orkney

5 June 1913 for over a week, Butt of Lewis,
Outer Hebrides

4 Cattle Egret Bubulcus ibis

10-19 May 1979, Loch of Kinnordy, Tayside
(British Birds 73:494)

14 October - 20 November 1986 and 6-14
December 1986, near Stranraer and
Crocketford, Dumfries & Galloway

A Red-breasted Goose Brantaruficollis
Winter 1956/57 Beauly Firth, Inverness-shire
(Scottish Naturalist 69:118)

One birdseen in Dumfries & Galloway winters
1991-92 - 1993-94 as follows:

2 October 1991 - 7 March 1992, Caerlaverock
and Mersehead; 5 October - 7 November

- 1992 Caerlaverock and Mersehead; 18

November - 31 December 1992 Caulkerbush
and Mersehead; 1 January - 11 February
1993 Southwick, Mersehead, Caulkerbush
and Caerlaverock; 25 September - 24
November 1993 Caerlaverock; 10 November

_ 1993 - 10 January 1994, near Dumfries
_ 5 December 1994, Endrick Mouth, Central/
r Strathclyde

A Black Duck Anas rubripes

22 -28 December 1979, adult male, Stanley
Dam, Paisley, Strathclyde (British Birds 73:

499)

Systematic List 135

11 October 1981 - 2 March 1982, male, North
Kessock, Inverness, Highland

9 February - 31 August 1985, female,
Tyninghame, Lothian and, again, 8 February
- 27 April 1986

26 November 1989 - 6 March 1990, first
winter male, Barr and Castle Semple Lochs,
Strathclyde

4 February - 29 April 1990, male, Loch of
Spiggie, Shetland

A Lesser Scaup Aythya affinis

29 December 1990, male, Milton Loch,
Dumfries & Galloway; same 31 December -
25 January 1991, Auchenreoch Loch,
Dumfries & Galloway (Birding World 4:29)
9-13 May 1993, first year male, Loch of
Spiggie, Shetland

A Bufflehead Bucephala albeola

14 March 1980 - at least 18th, male, West
Loch Bee, South Uist, Western Isles (British
Birds 74:464)

A Barrow’s Goldeneye Bucephala
islandica

4 November - 28 December 1979, adult male,
Irvine, Strathclyde (British Birds 88:104-106)

A Pallid Harrier Circus macrourus

24 April (at least)-8 May 1931, second year
male, shot, Fair Isle, Shetland (Scottish
Naturalist 1932:1)

5-7 May 1993, male, probably 2nd summer,
locality withheld, Tayside

15-16 September 1993, juvenile, Exnaboe
area, Shetland

A Lesser Kestrel Falco naumanni

25 October 1897, female, shot, Boyndlie,
Aberdeenshire (Annals of Scottish Natural
History 1898:51)

23 June 1987, male, Fair Isle, Shetland

136 SBRC

SB 18 (3)

A American Kestrel Falco sparverius
25-27 May 1976, male, Fair Isle, Shetland
(Scottish Birds 10:90, British Birds 74:199-
203)

A Eleonora’s Falcon Falco eleonorae
14 June 1985, male, Bornish, South Uist,
Western Isles (British Birds 83:459)

A Sora Porzana carolina

25 October 1901, male, shot, Tiree, Argyll
(Bulletin of the British Ornithologists’ Club
12:26)

12 November 1913, immature male, shot,
near Ness, Lewis, Outer Hebrides

30 October 1982, adult, Foula, Shetland

A Little Crake Porzana parva

March 1852, found dead, Thornton, Banff
(Zoologist 1860:6968)

29 March 1909, Girvan Harbour, Ayr, found
alive, later collected

29 September 1911, caught, Loch Scamadale,
Argyll

April 1959, male, dead, Uyeasound, Unst,
Shetland

11 May 1970, male, trapped, Fair Isle,
Shetland

A Sandhill Crane Grus canadensis
26-27 April 1981, first summer, Fair Isle,
Shetland (British Birds 76:105-109)

17-26 September 1991, first winter moulting
to second winter, Exnaboe and Sumburgh,
Shetland

B Houbara Bustard Chlamydotis
undulata

24 October 1898, immature female, shot, St
Fergus, Aberdeen (Annals of Scottish Natural
History 1899:51, 73)

A Cream-coloured Courser Cursorius
cursor
8 October 1868, male, shot, near Lanark,

Lanarkshire (Zoologist 1868:1459)

10 October 1949, 3 near Luggiebank, North
Lanark, Lanarkshire (Scottish Birds 5:28)
9-21 October 1965, Gullane Point, East
Lothian (Scottish Birds 4:230-232)

A Black-winged Pratincole Glareola
nordmanni

18-19 May 1927, female, exhausted, later
died, Fair Isle, Shetland (Scottish Naturalist
1927:111)

11 July 1976, Loch of Strathbeg, Grampian
(Scottish Birds 10:314-15)

A Greater Sand Plover Charadrius
leschenaultii

9 June 1979 - at least 14th, Sandside Bay,
Deerness, Orkney (British Birds 73:568-573)
24 June 1982, Aberlady Bay, Lothian

18-19 August 1991, adult or first summer,
Don Estuary, Grampian

A Caspian Plover Charadrius asiaticus
12-13 July 1988, adult, sex indeterminate,

Aberlady, Lothian (Lothian Bird Report |

1988:87-88)

A Sociable Plover Chettusia gregaria
3 November 1926, first winter female, shot,
North Ronaldsay, Orkney (Scottish Naturalist
1927:127)

Early December 1949, Isbister, Rendall,
Orkney

for about a week around 15 January 1969,
Eday, Orkney

A Great Knot Calidris tenuirostris

EEE

15 September 1989, adult, Scatness and

Pool of Virkie, Shetland (British Birds 85:426-
429)

A Semipalmated Sandpiper Calidris
pusilla

13-15 May 1992, Fairlsle, Shetland (Scottish

Birds 16:277-9)

oo — —
= —_ on a =

1996

7-8 June 1993, Stronsay, Orkney

A Western Sandpiper Calidris mauri
27 May - 3 June 1956, trapped, Fair Isle,
Shetland (British Birds 56:55-58)

25 September 1988, juvenile, Pool of Virkie,
Shetland

A Red-necked Stint Calidris ruficollis
31 August 1994, juvenile, recently dead, Fair
Isle, Shetland (Birding World 7:355-358)

A Least Sandpiper Calidris minutilla
14 August 1955, shot, Virkie, Shetland
(Scottish Naturalist 1957:170)

11-14 September 1965, Cadder, nearLenzie,
Lanark (Scottish Birds 4:506)

31 July - 4 August 1988, adult, Cove, near
Aberdeen, Grampian

A Sharp-tailed Sandpiper Calidris
acuminata

13-21 October 1956, Hamilton, Lanark
(Scottish Birds 1:94-96, 254, 256-7)

17 August 1985, adult, Aberlady Bay, Lothian
(Scottish Birds 14:124)

| 13-15 September 1993, adult, Scatness,
_ Shetland

OA Stilt Sandpiper
himantopus

18 April 1970, Dornoch, Sutherland (British
Birds 64:350)

11-18 September 1976,.Garth’s Loch,
Scatness, near Sumburgh, Shetland

Micropalama

B Eskimo Curlew Numenius borealis
6 September 1855, shot, Cairn Monearn,
near Stonehaven, Kincardine (The Naturalist
1855:265)

29 September 1878, shot, Slains, Aberdeen
| 21 September 1880, Forest of Birse,
| Kincardine

Systematic List 137

A Hudsonian Godwit Limosa
haemastica

26 September 1988, near Collieston,
Grampian (British Birds 83:463)
A Upland Sandpiper Bartramia
longicauda

13 October 1933, adult female, shot, Ruthwell,
Dumfries (British Birds 27:205-6)

5 October 1970, Fair Isle, Shetland

25 September 1975, Fair Isle, Shetland

24 April 1980, Hirta, St. Kilda, Western Isles
22 September - 6 October 1993, juvenile,
found dead on 18 November, Foula, Shetland

A Greater
melanoleuca
26-27 May 1953, Dunrossness, Shetland
(British Birds 48: 363)

25 October 1957, immature male moulting
into winter plumage, subsequently found dead,
Ythan Estuary, Aberdeen

14 August 1978, Peninerine, South Uist,
Western Isles

19 May 1985, Loch Sligachan, Skye, Highland
25 October 1985, Glenegedale, Islay,
Strathclyde

Yellowlegs’_ Tringa

A Solitary Sandpiper Tringa solitaria
A few years previous to 1870, shot, Banks of
Clyde, Lanark (Ibis 1870:292)

20 October 1990, juvenile, Malaclete, North
Uist, Western Isles

13 - 15 September 1992, juvenile, Fair Isle,
Shetland (Scottish Birds 17:62-64)

A Terek Sandpiper Xenus cinereus
20-21 June 1975, Whalsay, Shetland (Scottish
Birds 10:53-5)

5-12 June 1977, Sandside Bay, Highland
5-8 June 1987, Tankerness, Orkney

25 May 1991, male, song heard, Mid Yell
Voe, Shetland

138 SBRC

A Grey-tailed Tattler Heteroscelus
brevipes

27 November - 27 December 1994, juvenile,
Burghead, Grampian (Birding World 7:469-
472)

A Lesser Crested Tern Sterna
bengalensis

21-22 August 1987, Musselburgh, Lothian
(Lothian Bird Report 1987:79)

7-16 August 1989, adult, Musselburgh,
Lothian - presumed same bird; juvenile hybrid
x Sandwich tern S. _ sandvicensis
Musselburgh, Lothian 8-12 and 25 August
1989, same 2 September, Aberlady Bay,
Lothian

16 May 1993, adult, St Abbs Head, Borders -
presumed same as earlier bird

A Forster’s Tern Sterna forsteri

6 October - 9 November 1985, age uncertain,
Musselburgh and Granton Harbour, Lothian
(Scottish Birds 14:126)

A Bridled Tern Sterna anaethetus

6-7 August 1979, first summer, Stromness,
Orkney (British Birds 73:513)

2 August 1988, Sands of Forvie, Grampian
21 July 1993, Eigg, Highland

B Sooty Tern Sterna fuscata

End of May 1939, moulting into 2nd winter,
dead, Myatt Hill, near Denny, Stirling (British
Birds 33:197-8)

22 April 1954, Isbister, Rendall, Orkney

B Whiskered Tern Chlidonias hybridus
29 May 1894, adult male, shot, Nithsdale,
Dumfries (Annals of Scottish Natural History
1894:179)

A Rufous Turtle Dove Streptopelia
orientalis

31 October - 1 November 1974, first year, Fair
Isle, Shetland (Scottish Birds 10:55)

SB 18 (3)

A Great Spotted Cuckoo Clamator
glandarius

14-30 August 1959, immature, Rendall,
Orkney (Scottish Birds 1:152)

B Black-billed Cuckoo Coccyzus
erythrophthalmus

6 November 1950, first winter, dead,
Southend, Kintyre, Argyll (Scottish Naturalist
1951:131)

11 October 1953, died 12th, Foula, Shetland
(Scottish Naturalist 1953:196)

B Eagle Owl Bubo bubo

1830, shot, Sanday, Orkney (T E Buckley
and J A Harvie-Brown 1891 A Vertebrate
Fauna of the Orkney Islands p140 Douglas,
Edinburgh)

Autumn 1863, near Haroldswick, Unst,
Shetland

Early March 1871, Baltasound and Huney,
Unst, Shetland

February 1883, Duntrune, Kilmartin, Argyll

A Hawk Owl Surnia ulula

Winter 1860/61, shot, Skaw, Unst, Shetland
(H L Saxby 1894 The Birds of Shetland p 54-
55 McLaren and Stewart, Edinburgh)
December 1863, shot, Maryhill, Lanark
November 1868, shot, Greenock, Renfrew
21 November 1898, adult female, shot, Gight,
Aberdeen |
12-13 September 1983, near Lerwick,
Shetland and 20-21 September 1983, same
bird, Bressay, Shetland

A Common Nighthawk Chordeiles
minor

12 September 1978, trapped, near Kirkwall,
Orkney (Scottish Birds 11:85) |

A Chimney Swift Chaetura pelagica
8-10 November 1991, St Andrews, Fife —
(Scottish Birds 16:216-218) |

1996

A Needle-tailed Swift Hirundapus
caudacutus

11-12 June 1983, South Ronaldsay, Orkney
(Scottish Birds 13: Scottish Bird Report
1984:50)

25 May - 6 June 1984, Quendale, Shetland
28 May - 8 June 1988, Hoy, Orkney

11-14 June 1991, Noss, Shetland

A Little Swift Apus affinis

29 May 1985, St Andrews, Fife (Scottish
Birds 14:257-8)

1 November 1991, Fair Isle, Shetland

A Lesser Spotted Woodpecker
Dendrocopos minor

23 September 1968, Stirling/Perth boundary,
2 birds at site where suspected during
previous 2 years, not seen 1969, 3 seen
there early January 1970 (Scottish Birds
6:210-12)

Other records (such as those mentioned in V
M Thom 1986 Birds in Scotland p254 Poyser
Calton) are undocumented and therefore
unsatisfactory.

A Calandra Lark Melanocorypha
calandra

28 April 1978, Fair Isle, Shetland (Scottish
Birds 11:25-6)

21 September 1994, St Kilda, Western Isles

A Bimaculated Lark Melanocorypha
bimaculata
8 June 1976, Fairlsle, Shetland (British Birds
72:462-63)

B Crested Lark Galerida cristata
2 November 1952, Fairlsle, Shetland (British
Birds 46:211)

B Buff-bellied Pipit Anthus rubescens
30 September 1910, immature male, caught,
St Kilda, Outer Hebrides (Annals of Scottish

Systematic List 139

Natural History 1911:51-2)
17 September 1953, immature, Fair Isle,
Shetland

A Alpine Accentor Prunella collaris

6 October 1908, Fairlsle, Shetland (W E Clark
1912 Studies in Bird Migration Vol 2 p146
Gurney and Jackson, London)

27-28 June 1959, Fair Isle, Shetland

A Siberian RubythroatLusciniacalliope
9-11 October 1975, first winter male, trapped,
Fair Isle, Shetland (British Birds 72:89-94)

A Isabelline Wheatear Oenanthe
isabellina

At least 17 October - 10 November 1979,
trapped 23 October, Girdleness, Grampian
(British Birds 74:182-85)

20-21 September 1993, Whalsay, Shetland

A Rock Thrush Monticola saxatilis

17 May 1910, adult male, trapped, Pentland
Skerries, Orkney (Annals of Scottish Natural
History 1910:148-149)

8 November 1931, Fair Isle, Shetland

16 October 1936, Fair Isle, Shetland

17 June 1962, female, St Kilda

30 June 1970, first summer male, trapped,
Fair Isle, Shetland (Scottish Birds 6:336)

A Blue Rock Thrush Monticola solitarius
4-7 June 1985, found dead on 8 June, first
summer male, Skerryvore Lighthouse,
Strathclyde (British Birds 88:130-32)

A Siberian Thrush Zoothera sibirica
1-4 October 1954, adult male, Isle of May, Fife
(Scottish Birds 8:114, British Birds 48:21-25)
13 November 1984, male, South Ronaldsay,
Orkney

1-8 October 1992, trapped, first winter female,
North Ronaldsay, Orkney

140 SBRC

A Hermit Thrush Catharus guttatus
2 June 1975, Fair lsle, Shetland (British Birds
72:414-17)

A Swainson’s Thrush Catharus
ustulatus

25-29 October 1980, Scatness, Shetland
(British Birds 74:484)

30 September - 6 October 1990, Fair Isle,
Shetland

21 October 1993, trapped, Holm, Orkney

A Dusky Thrush Turdus naumanni
18-21 October 1961, first winter female,
trapped, Fair Isle, Shetland (British Birds
55:57)

24 September 1968, Whalsay, Shetland
(Scottish Birds 5:392-94)

6-13 November 1975, Firth, Shetland

16-17 October 1993, probably since 15th,
first winter male, Fair Isle with another first
winter male 17 October 1993.

A Cetti’s Warbler Cettia cetti
5 October 1993, freshly dead, Leith, Edinburgh
(Scottish Birds 17:176-7)

A Thick-billed Warbler Acrocephalus
aedon

6 October 1955, trapped, Fair Isle, Shetland
(British Birds 49:89-93)

23 September 1971, trapped, Whalsay,
Shetland, died Lerwick 25 September

A Olivaceous Warbler Hippolais pallida
24 - 26 September 1967, trapped, Isle of May,
Fife, killed by Great Grey Shrike Lanius
excubitor (Scottish Birds 10:24-5)

A Dartford Warbler Sylvia undata
18 May 1983, StAbbs Head, Borders (Scottish
Birds 13:52-3)

A Marmora’s Warbler Sylvia sarda
23-27 May 1993, in song, St Abbs Head,

SB 18 (3)

Borders (Birding World 6:182)

A Ruppell’s Warbler Sylvia rueppelli
13 August - 16 September 1977, male,
Dunrossness, Shetland (British Birds 74:279-
283)

3-19 October 1990, male, trapped 3rd,
Whalsay, Shetland

A Orphean Warbler Sylvia hortensis
10 October 1982, immature, trapped Seton
Park, Aberdeen (British Birds 77:552)

A Brown Shrike Lanius cristatus

30 September - 2 October 1985, adult,
Sumburgh, Shetland (British Birds 86:600-
604)

A Nutcracker Nucifraga caryocatactes
1 October 1868, shot, Sanday, (T E Buckley
& J A Harvey-Brown 1891 A Vertebrate
Fauna of the Orkney Islands p125 Douglas,
Edinburgh)

October 1868, shot, Invergarry, Inverness
early September 1891, Corsemaizie,
Wigtownshire

21-23 August 1968, Lerwick, Shetland

A Red-eyed Vireo Vireo olivaceus

13 - 16 October 1985, first winter, Wick,
Highland (British Birds 81:588)

1-7 October 1988, Newton Plantation, North
Uist, Western Isles

8 November 1988, Thurso, Highland

13-14 October 1991, Barns Ness, Lothian

3 October 1992, Arinagour, Coll, Strathclyde

A Trumpeter
githagineus |
8-9 June 1971, Handa, Sutherland (British
Birds 70:45-49) |
26-28 May 1981, male, Sanday, Orkney

4 June 1992, male, Balnakeil, near Durness,
Highland

Finch Bucanetes —

1996

A Pine Grosbeak Pinicola enucleator
8-9 November 1954, adult female, trapped,
Isle of May, Fife (British Birds 48:133-4)

25 March - at least 25 April 1992, male,
probably first year, in song, Lerwick, Shetland

A Evening Grosbeak Hesperiphona
vespertina

26 March 1969, male, St Kilda, Outer Hebrides
(British Birds 64:189-194)

10-25 March 1980, adult female, Nethy Bridge,
Highland

A Spanish Passer
hispaniolensis

11-19 August 1993, male, trapped, North
Ronaldsay, Orkney (Birding World 6:309-
311)

Sparrow

B Black-and-white Warbler Mniotilta
varia

Mid-October 1936, dead, Scalloway, Shetland
(British Birds 53:98)

A Tennessee Warbler Vermivora
peregrina

6-20 September 1975, immature, trapped,
Fair Isle, Shetland (British Birds 74:90-94)
24 September 1975, immature, trapped, Fair
Isle, Shetland (different bird)

5-7 September 1982, trapped, Holm, Orkney

A Yellow Warbler Dendroica petechia
3-4 November 1990, male, Helendale,
Lerwick, Shetland (British Birds 84:500;
86:530)

24 August 1992, first winter male, trapped,
North Ronaldsay, Orkney

A Chestnut-sided Warbler Dendroica
pensylvanica

20 September 1985, first year, Fetlar, Shetland
(British Birds 86:57-61)

Systematic List 141

A Blackburnian Warbler Dendroica
fusca

7 October 1988, first winter male, Fair Isle,
Shetland (British Birds 83:489)

A Cape May Warbler Dendroica tigrina
17 June 1977, singing male, Paisley Glen,
Renfrew, Strathclyde (British Birds 73:2-5)

A Yellow-rumped Warbler Dendroica
coronata

18 May 1977, male, Fairlsle, Shetland (British
Birds 71:526)

22-23 October 1982, Newton, North Uist,
Western Isles

A Blackpoll Warbler Dendroica striata
30 September - 3 October 1985, first winter,
trapped, Whalsay, Shetland (Scottish Birds
15 :Scottish Bird Report 1987:50)

6 October 1990, Sumburgh, Shetland

30 September 1991, Fair Isle, Shetland

A American Redstart Setophaga
ruticilla
1 November 1982, female/immature,
Portnahaven, Islay, Strathclyde (British Birds
76:525)

A Ovenbird Seiurus aurocapillus
7-8 October 1973, trapped, Out Skerries,
Shetland (British Birds 68:453-455)

A Common Yellowthroat Geothlypis
trichas

7-11 June 1984, male, Fetlar, Shetland (British
Birds 78:582, Scottish Birds 14:124)

A Hooded Warbler Wilsonia citrina

10 September 1992, Hirta, St Kilda, Western
Isles (British Birds 86:530, Birding World
5:380)

142 SBRC

SB 18 (3)

A Savannah Sparrow Ammodramus
sandwichensis

30 September - 1 October 1987, trapped, Fair
Isle, Shetland (British Birds 85:561-564)

A Song Sparrow Zonotrichia melodia

27 April - 10 May 1959, male, Fair Isle,
Shetland (British Birds 53:429)

17 April- 7 May 1979, male, trapped, FairIsle,
Shetland and 10 June 1979, (no sign of a
ring), Sumburgh, Shetland

11-26 April 1989, male, trapped, Fair Isle,
Shetland

A White-crowned Sparrow Zonotrichia
leucophrys

15-16 May 1977, trapped, Fair lsle, Shetland
(British Birds 73:466-470)

A Dark-eyed Junco Junco hyemalis

1 May 1966, Foula, Shetland (British Birds
60:332)

10 May 1967, Foula, Shetland

7 May 1969, Out Skerries, Shetland

19 May 1977, Glen Affric, Highland

3-4 May 1992, Hamilton, Strathclyde

A Cretzschmar’s Bunting Emberiza
caesia

10-20 June 1967, male, Fair Isle, Shetland
(British Birds 62:144-148)

9-10 June 1979, male, Fair Isle, Shetland

A Yellow-browed Bunting Emberiza
chrysophrys

12-23 October 1980, male, age uncertain,
Fair Isle, Shetland (British Birds 76:217-225)
22-23 September 1992, North Ronaldsay,
Orkney

A Pallas’s Reed Bunting Emberiza
pallasi

29 September- 11 October 1976, adultfemale,
Fair Isle, Shetland (British Birds 73:402-408)

17-18 September 1981, first year, trapped,
Fair Isle, Shetland

A Rose-breasted Grosbeak
Pheucticus ludovicianus

7 October 1983, first winter male, South Uist,
Western Isles, taken into care and died next
day (British Birds 77:560)

A Bobolink Dolichonyx oryzivorus

18 September 1975, Out Skerries, Shetland
(British Birds 70:222)

28 September 1986, St Kilda, Western Isles
29 September - 2 October 1986, Fair Isle,
Shetland

A Brown-headed Cowbird Molothrus
ater

24 April 1988, male, Ardnave, Islay,
Strathclyde (British Birds 87:284-288)

A Northern Oriole Icterus galbula
19-20 September 1974, immature, Fair Isle,
Shetland (Scottish Birds 10:58-9)

30 September - 3 October 1988, female or
first winter male, trapped, Benbecula, Western
Isles, probably present since about 22-24th

CATEGORY D

D1 Greater Flamingo Phoenicopterus
ruber

27-31 May 1988, Pool of Virkie, Shetland
(Shetland Bird Report 1988:13)

31 May - 28 June 1994, adult, Ythan estuary,
Grampian

These 2 records are the only ones, so far, to
be admitted to this Category by the BOU.

D1 Egyptian Goose Alopochen
aegyptiacus

16 March - 4 April 1977, Levenmouth -
Grahamston, Fife (Scottish Birds 10:124)
21-30 April 1977, Coulmore, Black Isle,
Highland

1996

March - May 1980 and 15-19 April 1981,
Linlithgow, Lothian
1-14 February 1992, 2, Torven, Highland

D1 Baikal Teal Anas formosa

5 February 1958, female, shot, Loch Spynie,
Morayshire (British Birds 74:321, 460)

19 February - 7 April 1973, male,
Caerlaverock, Dumfries

Di Saker Falcon Falco cherrug

1-5 October 1976, Out Skerries, Shetland
(British Birds 73:530)

27-29 May 1978, Fetlar, Shetland

23 October - 3 December 1986, first winter,
Fair Isle, Shetland

D2 Northern Flicker Colaptes auratus
July 1981, corpse found, probably died on
board ship and decomposedin dry conditions,
Caithness, Highland (British Birds 75:531)

Di Cedar Waxwing Bombycillacedrorum
25-26 June 1985, Noss, Shetland (Ibis
135:221, British Birds 86:538)

D1 Asian Brown Flycatcher Muscicapa
dauurica
1-2 July 1992, first summer, trapped, Fair

Systematic List 143

Isle, Shetland (Ibis 136:254, Birding World
5:252-255, British Birds 87:247-252)

D1 Daurian Starling Sturnus sturninus
7-28 May 1985, male, trapped, Fair Isle,
Shetland (British Birds 82:603-612, Ibis
136:497)

D1 Chestnut Bunting Emberiza rutila
9-13 July 1974, adult male, Foula, Shetland
(British Birds 70:444)

11 June 1985, first year female, trapped, Isle
of May, Fife

15-16 June 1986, first summer male, trapped,
Fair Isle, Shetland

2-5 September 1994, female, trapped, Out
Skerries, Shetland

D1 Blue Grosbeak Guriaca caerulea
mid - 26 August 1970, male, Out Skerries,
Shetland (Scottish Birds 8: Scottish Bird
Report 1970:397)

10-11 March 1972, Kiltarlity, Inverness

22 May 1977, male, dead, Ettrick, Borders

D1 Indigo Bunting Passerina cyanea
3-7 August 1964, male, Fair Isle, Shetland
(British Birds 73:531)

Peter R Gordon & David Clugston on behalf of the Scottish

Birds Records Committee.
Address for correspondence:
The Secretary, SBRC,

31 Argyle Terrace, Rothesay, Isle of Bute, PA 20 OBD

144 Scottish Birds (1996) 18: 144-151 SB 18 (3)

Changes since 1890 in the size of Pink-footed Geese wintering
in Britain

H BOYD & J BERRY —

Weights and linear measurements of Pink-footed Geese shot in Fife

between 1891 and 1953 are compared with measurements of geese shot in
1966-68 and caught for ringing in 1953-59 and 1987-94. Before 1948, adult
females tended to be heavier and longer-winged than those now living. In |
1891-1909, when mean weights were higher than in any recent period, they
were lowest in October and increased gradually until April, without a dip in
midwinter, as has been found in recent winters. More historical records are

needed to determine whether the pattern has really changed

Introduction

It has been shown recently that individuals of
Arctic nesting Snow Geese Anser
caerulescens are decreasing in size, the
decreases being associated with reductions
of preferred food plants due to heavy grazing
near breeding colonies, following large
increases in the total numbers of geese
(Jefferies et al 1979, Kerbes et al 1990, Cooch
et al 1991, 1993). This paper examines the
possibility that, before 1950, when the Pink-
footed Geese Anser brachyrhynchus breeding
in Iceland and east Greenland were much
scarcer (<40,000) than they are now
(>200,000), their sizes and seasonal changes
in weight might have differed from those of
geese now living.

Materials and methods

One of us (JB) weighed and measured most
of the geese which he shot in 1928-53, and
has records of weights of geese shot by
William Berry between 1891 and 1929. HB
helped weigh large numbers of Pink-footed
Geese caught in rocket nets between 1953
and 1959 (Beer & Boyd 1962). In 1966-68 C
R G Campbell weighted and measured geese

shot in Kinross and Perth. Mr Carl Mitchell
has provided us with preliminary results from
weighing and measuring Pink-footed Geese
caught by the Wildfowl and Wetlands Trust

for ringing at several sites in England and ©

Scotland in 1987-94. This make it possible to

compare the weights of Pink-footed Geese in —

5 different periods during the last 100 years;

wing lengths in 1929-48 with those in 1966- |
68 and 1987-94; and tarsus lengths in 1929- —
48 and 1987-94. Although very few geese ©

were handled in any one year between 1891

and 1953, the pooled samples are sufficient —
to demonstrate several differences in size |

over the last 100 years.

From 1891 to 1948, and in 1953, the weights

of geese were recorded in Imperial measure,
to the nearest half ounce (W Berry) or ounce.

These records have beenconverted metrically |

(to 10g). In 1957-59, 1966-68 and 1987-94,
metric balances were used. Because live

geese need to be restrained while being
weighed, and some persist in being active, —

the accuracy of live weighings may be less
than that achievable with dead birds. This
may help to account for the high variability in

the recorded weights of live geese discussed _

by Beer & Boyd (1962).

1996

Comparisons of dead and live weights might
be affected by different biases in the choice of
targets by catchers and shooters. Almost all
the geese shot by William and John Berry
were Pink-footed, so that they were not
selecting that species from mixed flocks by
size. Whether they were aiming at small
individuals (as Cree Indians would, knowing
that young geese are better to eat), or
exceptionally large ones (as trophy hunters
try to do), can only be determined now by
tests for skewness and kurtosis in the
distributions of the weights and measurements
about the period means. Though the linear
measurements show some right skewing,
this does not exceed that generally found in
sets of biological measurements (Simpson

Changes in Pink-footed Geese wintering in Britain 145

Catches for ringing are also potentially biased;
forexample, geese caught early in the morning
tend to be lighter, and to include more first
winter birds, than those taken later in the day,
because families with young geese tend to
arrive first at feeding sites. As the exact
records of times of catch in 1953-59 have
been lost this source of bias cannot be
explored.

Results
Linear measurements
Table 1 summarises the measurements of

total length, wing, exposed culmen and tarsus
made on geese taken in Fife between 1929

and Roe 1939), so that it is reasonable to
believe that selection by size was not
achieved, even if it had been attempted.

and 1948. A few taken in later years are not
included in Table 1. Not all measurements
were made on all individuals. Tables 2 and3

Table 1 Linear measurements of Pink-footed Geese in Fife, 1929-1948. n=sample
size (standard deviation in brackets)

measurement

adult adult 1st winter 1st winter
(mm) male female male female
n mean n mean n mean n mean
‘totallength 27 751 18isgq nest 72 pimp gin BEAR ne 694
(25.0) (34.4) (39.7) (56.7)
wing 21 441 30 434 4 402 10 420
(10.5) (18.5) Gh Adsl (16.4)
culmen 22 46.1 23 AA rove 40 14 44.1
(from feathers) (2.2) (2.2) (2.9)
tarsus 17 70.8 25 69.7 =a Ga:0'"*"10 67.9
(4.8) (5.8) (4.4) (5.9)

146 H Boyd & J Berry

SB 18 (3)

Table 2 Mean wing and tarsus lengths(mm) of adult Pink-footed Geese in Britain at
different periods, 1850-1994. (Standard deviation in brackets)

c.1850 1928-48
(1) (2)
Wing
males 444 441
(10.5)
females 419 434
(18.5)
Tarsus
males 64.5 70.8
(4.8)
females 59.3 69.7
(5.8)
Sources: (1) MacGillivray (1852); (2) this study;

recordcomparable measurements from other,
later samples and from several published
sources. The few first winter males in the Fife
sample were smaller than most of the first
winter females, an anomaly not found in any
other samples. Because all 4 of these linear
measurements are liable to be affected by the
detailed techniques used by the individuals
making them, caution is necessary when
comparing samples. For example, it seems
very probable that Macgillivray (1852)
measured the tarsus in a different way from
that used later. Nevertheless, his
measurements (of birds obtained from the
Edinburgh markets) show a large difference
between males and females, resembling those
found after 1950, and unlike the Fife sample.

Although the general impression from Tables
1 - 3 is that there have been no substantial

1966-68 1987-94

(2) (3)
452 449

(16.1) (17.2)
432 427

(11.6) (16.4)
: 74.3
(3.5)

: 71.3
(3.5)

(3) Wildfowl & Wetlands Trust

changes in body measurements, the mean
wing lengths of both sexes in 1929-48 were |
different from those measured more recently.
In later samples, the wings of males have
tended to be longer, and those of females —
shorter, than those of the geese taken in Fife. |
The mean tarsus length of males has also
been greater.

Bauer and Glutz (1968) claimed that the Pink-

footed Geese breeding in Spitzbergen and
wintering in western continental Europe are
smaller than those from Iceland and
Greenland, while Ogilvie (1978) noted that “It
has been claimed that the Svalbard birds are
a little larger than the others but insufficient |
measurements have been obtained to show ©
this conclusively”. Samples of adult males

and females wintering in Denmark in 1910-21 |
and in The Netherlands in the 1970s show no -

Males
Denmark Netherlands

Wing 442 443
(11.8) (10.8)

~ Culmen 47.1 47.2
(225) (2.1)

- Tarsus 74.2 76.4
(7.5) (2.5)

1996

Changes in Pink-footed Geese wintering in Britain

147

Table 3 Linear measurements (mm) of adult Pink-footed Geese from Spitzbergen
wintering in Denmark, 1910-1921, and in the Netherlands, 1970-1975. (Standard

deviations in brackets)

significant differences from each other in wing

_ and tarsus length (Table 3). The mean wing

length of these European females is much
less than those of the Fife birds and of the
large samples taken in Britain since 1987.
The mean length of males’ wings was similar
to those in Fife in 1928-48, though much
shorter than that of the recent British or the

_ Fife samples. The mean tarsus length of the
_ adult males is similar to that of recent British

birds and markedly longer than in those from

_ Fife.

Changes in wing and tarsus length do not
necessarily indicate changes in body size.
Body length does, but has considerable

_ practical disadvantages, since it can only be

measured with some consistency on dead

| birds, with difficulty on live birds and not at all

on museum skins of geese, the necks of

| which have usually been shortened or bent to
_ reduce the space taken up by the specimen.
} The Fife sample shows some correlation

_ between body length and wing length, much
_ Stronger in adult females (r=0.748, p<0.001,

n=28) than adult males (r=0.386, p=0.05,

Females
Denmark Netherlands
424 420
(4.8) (9.1)
44.3 43.2
(1.8) (1.4)
68.5 69.4
(3.9) (3.0)

n=25). The correlations were strong after
1940 (females r=0.786, p<0.001, n=15; males
r=0.804, p<0.001, n=10) but weak earlier
(females r=0.328, males r=0.349, neither
significant at the 10% level).

Weights

Although nothing is known of the relative
abundance of first winter and older birds, or of
females and males, among the geese handled
by William Berry in 1891-1909, their mean
weight seems to have been greater than that
of the geese shot in Fife in 1928-48 and near
Loch Leven in 1966-68, and of those caught
for ringing in 1953-59 and 1987-94 (Table 5).
There are considerable differences between
the means of the age and sex classes in the
3 more recent samples (Table 6), so that
there is nota simple pattern of decline through
these samples.

Although all the monthly samples of adults
from Fife are small, Table 7 suggests
substantial differences in the pattern of change
through the winter in the 4 periods. In 1891-

148 H Boyd & J Berry SB 18 (3)

Table 4 Differences in mean lengths (mm) of wing and tarsus of adult Pink-footed
Geese in Fife, 1929-48 from (a) British wintering adults in later periods and (b) adults
wintering in Denmark and the Netherlands

Statistical significance of t: *p <0.05, ** p<0.01, *** p<0.001

Interval Wing Tarsus

males females males females
(a) Fife,
1929-49 compared
to Britain
1950s +10*** -6** +5.6*** -0.3
1966-68 +11* -2 - -
1987-94 +8* Te 495°" +1.6
(b) Europe
compared to
Fife, 1929-48 -2 +12.4** -5.0** -0.05
Britain, 1987-94 +6.3* +5.4 -1.5 +1.55

Table 5 Mean weights (kg, to 0.01) of Pink-footed Geese wintering in Britain during
3 periods from 1891 to 1968. Because the records for the first period were not
classed by age and sex, the records within each of the 2 later periods are also
lumped

Period n mean SD significances of differences
1928- 1966- 1987-
1948 1968 1994
1891-1909 39 2.84 0.33 NS ee ae
1928-1948 71 2.75 0.40 at +e a
1966-1968 113 2.49 082 _ **

1987-1994 1932 2.59 0.34

“pr 0:0 1h p= 00011

|
;

————— ere

1996

Changes in Pink-footed Geese wintering in Britain

149

Table 6 Mean weights (kg) of Pink-footed Geese wintering in Britain during 3
periods since 1928, grouped by age and sex.

1928-1948
n mean SD n
Adult males 21 2 620:3 1 30
Adult females 21 Za (0:30 30
1st w. males + 252.044 22
1stw. females 6 22039 31

1909, the mean monthly weights of the geese
shot were rather higher in midwinter than in
autumn, before peaking in April. The April
means were also highest in those 2 later
periods in which some geese were weighed
late in the season.

Discussion

The recent pattern of weights increasing from
October (soon after migration) to November,
then falling to a midwinter low before rising to
a pre migration peak in April, seems to be
common in migratory waterfowl, so that the
apparent lack of a midwinter drop in 1891-
1909 is unexpected. Much of the benefit
which modern agricultural practices have
conferred on geese stems from improvements
in grass production, including the replacement
of some permanent grassland by short term
leys and the use of varieties that begin growth

at low temperatures early in the spring, as
_ well as heavy use of fertilisers to increase
. yields.

With none of those potential
advantages available at the end of the 19th

_ century, how might it have been possible for

geese wintering in Fife to maintain high

_ weights throughout the winter? Noimmediate

1966-1968 1987-1994
mean SD n mean SD
221 0.34 ST 1285 0.30
2.02 0.23 S116 2.56 0.27
2.43 0.39 452 2.54 0.29
223 O32 383 2.29 0.27

answer can be offered. Historical records of
weights and measurements from other parts
of the wintering range might show whether
this difference from today in the pattern of
weight change through the winter was
widespread, or simply the chance outcome of
dependence onsmallsamples. Investigations
into past variations in climate and agriculture
are also needed.

More generally, if we are to learn ecological
lessons from the past, rather than supposing
(unwisely) that only new data are of use in
dealing with current problems, there is an
urgent need to rescue manuscript and early
published records from oblivion, by
establishing suitable archives for their
collection and preservation. The Waterston
Library already serves this function for Scottish
ornithology, but the SOC lacks the resources
to search for additional records or to promote
the uses to which historical information can
be put. A sense of urgency in the search for
historical material has been heightened by
the advent of computers. Though admirable
for dealing with large volumes of data, they
tend to suggest a false discontinuity between
the present and the past, by discouraging

150 H Boyd & J Berry

Table 7 Monthly mean weights (kg) of Pink-footed Geese from October to April
during 4 periods, from 1891-1909 to 1987-1994 Adults and first winter birds pooled

Period Oct Nov Dec
LEST=1909 mean 2.16 un 2.78
SD 0.33%..0:24.. 0572
1928-1948 mean 2.77 2.84 258
SD 0:62-- 0.16 0.24
1966-1968 mean 2xah- 2.34 2.64
SD 0:33 0:45 0134
1987-1994 mean 25g 260 © 2.02
SD Oia2 “Oss” 20137

searches for information not already stored in
machine readable form: ecology existed long
before the 1970s.

Acknowledgements

We are indebted to C R G Campbell for
unpublished records, and to A D Fox, C
Mitchell and the Wildfowl & Wetlands Trust
for access to recent records, prior to
publication. An anonymous referee helped to
remove some obscurities.

References

Bauer K M & Glutz von Blotzheim U N (eds).
1968. Handbuch der Vogel Mitteleuropas.
Band 2 Anseriformes (1 Teil). Akademische
Verlag., Frankfurt am Main.

Beer J V & Boyd H 1962. Weights of Pink-
footed Geese in autumn. Bird Study 9: 91-99.
Cooch E G, Jefferies, R L, Rockwell R F &

SB 18 (3)

Jan Feb Mar Apr
2.86 2.86 - 2.94
0.41 0.26 0.34
2.84 2.95 2:65)" 2269
0.44 Ot 0.18 O89
2153 2.66 - -
0.33 0.40

2.40 Zor 2.05 2.09
0.31 0.33 0:30° “0°39

Cooke F 1993. Environmental change and
the cost of philopatry: an example in the
Lesser Snow Goose. Oecologia 93: 128-
hope

Cooch E G, Lank D B, Rockwell R F & Cooke ©
F 1991. Long-term decline in body size ina
Snow Goose population: evidence of
environmental degradation? Journal of Animal -
Ecology 60: 483-496.

Cramp S & Simmons K E L (eds). 1977.
Handbook of the Birds of Europe, the Middle
East and North Africa. Vol.3. Oxford University
Press, Oxford.

Jefferies RL, JensenA & Abraham KF 1979.
Vegetational development and the effect of
geese on vegetation at La Perouse Bay,
Manitoba. Canadian Journal of Botany 57:
1439-1450. |
Kerbes R H, Kotanan P M & Jefferies RL
1990. Destruction of wetland habitats by
Lesser Snow Geese: a keystone species on |
the west coast of Hudson Bay. J.appl.Ecol. |

1996 Changes in Pink-footed Geese wintering in Britain 151

27: 242-258. Owen M 1980. Wild Geese of the world.
Macgillivray W 1852. A History of British Batsford, London.
Birds Vol 4. Orr, London. Schigler E Lehn 1925. Danmarks Fugle.

Matthews G V T & Campbell C RG 1969. Bind |. Gyldendalske Boghandel, Nordisk
_ Weights andmeasurementsofGreylagGeese Forlag, Kobenhavn.

in Scotland. Wildfowl 20: 86-93. Simpson GG & Roe A 1939. Quantitative

Ogilvie MA 1978. Wild Geese. Poyser, Zoology. McGraw-Hill, New York & London.
_ Berkhamsted.

H Boyd, 1032 Pinewood Crescent, Ottawa, Ontario, Canada K2B 5Y5
J Berry, The Garden House, Tayfield, Newport-on-Tay, Fife DD6 8HA

Revised manuscript accepted October 1995

Pink-footed Goose Keith Brockie

152 Scottish Birds (1996) 18: 152-158

SB 18 (3)

A census of the St Kilda gannetry in May 1994

S MURRAY & S WANLESS

An aerial photographic survey of the 3 sub colonies of the St Kilda gannetry
was Carried out on 15 May 1994. The overall total for St Kilda was 60,428
apparently occupied sites (AOS), with 14,660 AOS on Stac Lee, 12,950 AOS
on Stac an Armin and 32,818 AOS on Boreray. Numbers had increased
since the last survey in 1985; totals were 8.4%, 9.3% and 33.0% higher on
Stac Lee, Stac an Armin and Boreray, respectively. The overall total was
20.7% higher. For Boreray some of this increase was almost certainly due
to differences in methodology between the 2 surveys (the 1985 count was
made mainly from land or sea based counts and the standard of
photographs was generally lower than in 1994). However, increases in
some sections were undoubtedly real, and in some cases, were associated
with an expansion of breeding distribution. Count methodology and
photographic standard were directly comparable for Stac Lee and Stac an
Armin and the increases of 0.9% pa and 1.0% pa can be regarded as
genuine. The average rate of increase of 3.2% pa for Boreray between 1985-
94 is likely to be an overestimate. A comparison with the 1959 Boreray
count suggests an average rate of increase of 0.8% pa between 1959-94.

We conclude that a slow but sustained increase of c0.9% pa has taken place
in the St Kilda gannetry over the last 35 years

Introduction

There is a long tradition of counting Gannets
Morus bassanus among British ornithologists
with 7 major censuses this century. During
this time the number of gannetries in the east
Atlantic has increased from 12 to 39 and the
number of occupied sites has risen from

c91,000 to c223,400 (Wanless 1987). The
largest gannetry Is on St Kilda, Western Isles
where, during the last full survey in 1985, the
colony contained 50,050 apparently occupied
sites which represented 22% of the total east
Atlantic population (Murray & Wanless 1986).
Another major census of east Atlantic
gannetries was completed between 1994-95
in which obtaining an accurate count of St

Kilda had high priority. The site is also
designated as a Special Protection Area (SPA)
by virtue of its importance to Gannets and |
other seabirds.

Undoubtedly St Kilda is the hardest gannetry |
in the world to count both because of its size
and the complexity of the cliffs. Censusing it
presents a formidable logistic challenge and, |
as a result, past efforts to count the colony
from the land and/or sea have been limited. |
The archipelago is remote and subject to
severe and unpredictable weather and an-
aerial survey offers the most efficient census
method. However, this approach is costly
and can be fraught with difficulty. Of the 6 ©
attempts to date, the mostsuccessfulinterms —

1996

of coverage and photographic quality have
been the first in 1959 (Boyd 1961) and the
present survey in 1994.

Methods
Aerial Survey

An aerial photographic survey of the St Kilda
gannetry was carried out on 15 May 1994 in
ideal weather conditions, with good to
excellent visibility. The aircraft was in the
vicinity of the colony for about 50 minutes;
much of this time was spent c5 km offshore
while the pilot was briefed on tne topography
of Borerayandthestacs. Four photographers
were on board, which enabled maximum
coverage to be achieved in the minimum
time. The coverage of both Stac Lee and Stac
an Armin was completed on one close circuit
of each stac. Boreray required level runs at
different altitudes along the east and west
cliffs separately. There was no disturbance
to breeding Gannets and it was clear from the
photographs that no nests were left
unattended. However, some non breeding
and club birds did leave the cliffs. Itwas also
intended to augment the aerial survey with
land and sea counts. However, poor weather
meant that few landings were possible in
1994. Fortuitously a photograph of section
89 (the only part of the colony for which there
were no usable aerial photographs) was taken
in late July.

Counting protocol

The count was made by projecting the colour
slides onto a sheet of paper and blocking out
the images with a felt tip pen. The counting
unit used was the apparently occupied site
(AOS: one or 2 Gannets occupying a site
suitable for breeding, irrespective of whether
any nest material is visible). The standard of

Census of St Kilda Gannetry 1994 153

the slides was sufficiently high to allow sites
to be counted directly and there was no need
to convert counts of individual birds to AOS by
the appropriate attendance ratio for pairs (cf
Nelson 1978). To facilitate comparisons with
previous counts, the colony was divided into
sections. Twenty six were used for Boreray;
these were amalgamations of the 89 originally
delimited by Boyd (1961) for his 1959 count
and have been used in the 1979 and 1985
surveys (Murray 1981, Murray & Wanless
1986). Sections for Stac Lee and Stac an
Armin were virtually identical to those used in
1985 (Murray & Wanless 1986) but did not
correspond to those adopted by Boyd.

Photographic coverage

Aerial coverage of Stac Lee and Stac an
Armin was 100%, and for both the standard of
photographs was extremely high. The one
exception was the west face of Stac an Armin
for which the slides, although perfectly
adequate for counting, were of only moderate
standard. Except for a small part of sections
80 and 83-88, which were hidden from the air,
complete aerial coverage was achieved for
sections 1-88 on Boreray. This represented
97.7% of the total count. Hidden areas were
estimated to represent 1.4%. No usable
aerial view of section 89 was obtained. This
section was counted from slides taken from
the sea and represented 0.9% of the total.

Many of the sections were photographed
from much closer range than has previously
been achieved and were of avery high quality.
Forty-eight slides were eventually selected
from which to make counts of Boreray, 9 and
13 were used for Stac Lee and Stac an Armin
respectively. Copies of the count slides and
prints showing the section boundaries are
lodged with the Seabird Colony Register at
JNCC Aberdeen.

154 S Murray & S Wanless SB 18 (3)

Table 1 Section by section comparison of counts of the Boreray gannetry in June
1985 and May 1994. Counts for 1985 are taken from Murray and Wanless (1986)

Occupied sites

Section 1985 1994 % change
12 106 71 -33.0
ani 1603 1892 18.0
13-15 984 1756 78.4
16-41 3922 5815 48.3
42-43 1070 1316 23.0
44 57 89 56.1
45 522 500 A
46 522 576 10.3
47 521 784 50.5
48 468 835 78.4
49 529 1063 100.9
50 49 34 -30.6
51-57 2974 4068 36.8
58-63 1865 2447 Sde2
64 411 549 33.6
65 308 455 47.7
66 435 463 6.4
67-72 1402 1828 30.4
73 326 426 30.7
74-78 +472 2622 48.0
79 413 381 i
80-821 1385 1723 24.4
83-88 | 2754 2831 2.8
89 279 294 5.4
Total 24676 32818 33.0

1996

Results
Boreray

Counts by the 2 observers gave totals of
31,855 and 32,882 AOS respectively. Taking
the average of these counts as the best
population estimate and adding on 100 AOS
and 349 AOS (based on direct counts in
1985) to allow for known under counts of
section 80 and 83-88 respectively, gives an
overall corrected total of 32,818 AOS on
Boreray in 1994. The standard of the count
was considered to be high. The 1994 total of
32,818 AOS was 33.0% higher than the last
count made in 1985 of 24,676 AOS and
almostall sections of the colony hadincreased
(Table 1). However, methodology for the 2
counts differed markedly with a much higher
percentage of aerial coverage in 1994 (98%)
compared with 1985 (34%). Moreover, the
quality of many of the photographs used to

Census of St Kilda Gannetry 1994 155

make the 1985 counts was much lower than
in 1994. These differences hinder the
interpretation of some of the apparent changes
shown in Table 1, but it appears that there
have been marked changes in colony extent
in sections 13-15, 42-43, 48, 49 and the
summit and east side of section 83. The
apparent increases in sections 16-41 and 51-
57 should be treated cautiously as they are
likely to be due to the improved quality of
aerial coverage.

Stac Lee

Counts by the 2 observers gave totals of
14,404 and 14,916 AOS respectively
indicating an average total of 14,660 AOS.
The standard of the count was considered to
be high. Comparisons between 1994 and
1985 were not hindered by differences in
survey methods or photographic quality. The
1994 total of 14,660 AOS was 8.5% higher

Table 2 Section by section comparison of counts of the gannetry on Stac Lee in
1985 and 1994. Counts for 1985 are taken from Murray and Wanless (1986)

Occupied sites

Section 1985
Top Table 7098
Casting Point 2148
Bothy Face 1063
North Face 3212
Total 13521

1994 % change
7141 0.6
2877 33.9
1525 43.4
3117 -2.9
14660 8.4

156 S Murray & S Wanless SB 18 (3)

Table 3 Section by section comparison of counts of the gannetry on Stac an Armin
in 1985 and 1994. Counts for 1985 are taken from Murray and Wanless (1986)

Occupied sites

Section 1985 1994 % change
East face and

south summit 8241 9089 10:3
Lower east face 1554 1398 -10.0
South centre and

lower south ledge 1043 1259 20:7
West face AOS 1204 18.8
Total 11851 12950 9.3

Gannets David Mitchell

1996

than the 1985 count of 13,521 AOS (Table 2).
Most of the increase had occurred on the
lower part of the south face ie on Bothy Face
and Casting Point. The prominent empty
area in front of and above the bothy adjacent
to Top Table was still unoccupied. Numbers
on Top Table and North Face were very
similar to those in 1985 and the very obvious
Gannet free ledge on the latter remained
uncolonised. There were no obvious changes
in distribution suggesting that the observed
increase was due to an increase in breeding
density.

Stac an Armin

Counts by the 2 observers gave totals of
12,609 and 13,291 AOS respectively
indicating an average total of 12,950 AOS.
The 1985 and 1994 counts were directly
comparable in terms of aerial coverage and
photographic quality. The 1994 total of 12,950
AOS was 9.3% higher than the 1985 count of
11,853 AOS (Table 3). The 3 main areas of
colony expansion identified in 1985 (details in
Wanless 1987) had expanded further by 1994
with the biggest increase occurring in the
south centre colony.

Discussion

The 1994 survey indicated that there were
32,818, 14,660 and 12,950 AOS on Boreray,
Stac Lee and Stac an Armin respectively.
This gives an overall total for St Kilda of
60,428 AOS. Taken at face value this
represents a 20.7% increase over the last
count of 50,050 AOS in 1985. However, as
has been stressed in the results, although
count methods and standards for Stac Lee
and Stac an Armin were very similar in 1985
and 1994, counting methods differed markedly
for Boreray with a much greater percentage
of aerial coverage and higher photographic

Census of St Kilda Gannetry 1994 157

quality in 1994 compared with 1985. Thus,
while we are confident that the observed
changes on Stac Lee and Stac an Armin
accurately reflect the population trend
between 1985 and 1994, changes on Boreray
have to be treated more cautiously.

Counts for Stac Lee and Stac an Armin indicate
average rates of increase of 0.9 and 1.0% pa
respectively between 1985 and 1994. These
values are remarkably similar to the rates of
increase of 0.9% and 0.8% pa calculated for
Stac Lee and Stac an Armin over the period
1959 to 1985 (Boyd 1961, Murray & Wanless
1986).

Counts for Boreray suggest that numbers
increased at 3.2% pa between 1985 and
1994. We now consider that the 1985 count
was probably too low and the calculated rate
of increases, therefore, too high. Conversely,
the increase between 1959 and 1985 was
probably bigger than previously suggested
(Murray & Wanless 1986). Comparing the
1994 count with that for 1959 (which was also
obtained almost entirely from an aerial survey),
gives an average rate of increase of 0.8% pa,
ie a very similar trend to that shown by Stac
Lee and Stac an Armin over the same period.
We conclude that a slow but sustained
increase has occurred on Boreray and the
Stacs over the last 35 years, with the St Kilda
gannetry increasing at an average rate of of
0.9% pa.

This survey has reaffirmed the importance of
the St Kilda gannetry and has also highlighted
the difficulties involved in censusing this large
and complex colony. Westrongly recommend
that future counts are based primarily on
aerial surveys, if possible, with concurrent
land and sea visits.

158 S Murray & S Wanless

Acknowledgements

This survey was a partnership project and
contributions towards the costs were received
fromthe Scottish Ornithologists’ Club, Scottish
Natural Heritage, Joint Nature Conservation
Committee, the British Trust for Ornithology,
the Seabird Group and the Royal Society for
the Protection of Birds. We thank all these
organisations for their support and Mark
Tasker, Emma Brindley, Jeremy Greenwood
and Greg Mudge for commenting on an earlier
version of the manuscript. The success of the
aerial survey was largely due to the flying
skills of Alan Mossman of Air Alba. His
enthusiasm and cooperation gave us a
memorable experience. We are also grateful
to Mary Harman, Digger Jackson, Kenny
Taylor and Jim Vaughan for backup
photography and Mike Harris for helpful
discussions.

SB 18 (3)

References

Boyd J M 1961. The gannetry of St Kilda.
Journal of Animal Ecology 30: 117-136.
Murray S 1981. A count of Gannets on
Boreray, St Kilda. Scottish Birds 11: 205-
Zt

Murray S & Wanless S 1986. The status of
the Gannet in Scotland 1984-85. Scottish
Birds 14: 74-85.

Nelson JB 1978. The Gannet.
Poyser, Berkhamsted.

Wanless S 1987. A survey of the numbers
and breeding distribution of the North Atlantic
Gannet Sula bassana and an assessment of
the changes which have occurred since
Operation Seafarer 1969/70. Research and
survey in nature conservation series No 4.

T&AD

Nature Conservancy Council, Peterborough.

S Murray, Craigie Dhu, Cardney, Dunkeld,

Perthshire PH8 OEY

S Wanless, Institute of Terrestrial Ecology,

Hill of Brathens, Banchory, Kincardineshire AB31 4BY

Revised manuscript accepted December 1995

er ee Se

Scottish Birds (1996) 18: 159-165

Les)

Ravens nesting on buildings in Orkney

C J BOOTH

Ravens have been recorded nesting on buildings in 5 territories on
Mainland Orkney between 1972 and 1995, as well as on buildings on other
islands in Orkney. Ruined buildings, as described below, provide good
substitute nest sites when cliffs and trees are absent. Ravens nesting on
buildings on Mainland were more successful than those at other sites
although there was no difference in brood size. Such use of buildings by
Ravens appears to be more widespread in Orkney than in other parts of
Britain. Similar choices elsewhere in the Western Palearctic and in north
America are noted. By using this type of nest site Ravens are able to nest
in areas where their more typical nest sites, cliffs and trees, are absent

Introduction

In the Western Palearctic the typical nesting
sites of the Raven Corvus corax are on cliffs,
both sea and inland, or in trees and are
usually difficult of access (Cramp and Perrins
1994). Occasionally man made structures
such as pylons or buildings are used. During
a study of Ravens on Mainland Orkney it was
noted that, although in the majority of Raven
territories the nests were on sea cliffs, there
were a number of territories in which nests
were regularly sited on ruined buildings.

There are historical references to Ravens
nesting on buildings in Orkney. Low (1813),
writing in 1770, mentioned that Ravens
sometimes used the steeples of churches
and ruined buildings. Buckley and Harvie-
Brown (1891) referred to Wolley’s Egg Book
where it was stated that, in 1849 when the
Kirkwall Cathedral was undergoing repairs,
the Ravens that always nested there had
moved to the old ruins of the Bishop’s palace.

The purpose of this paper is to detail the use
of buildings as Raven nesting sites in Orkney
during the period 1972 to 1995.

Study area and methods

The study area involved the whole of Mainland,
the largest of the Orkney Islands. It has an
area of approximately 490 sq km and a
coastline of 234kms, 50kms (21%) of which
consists of cliffs over 15 metres in height,
169kms (72%) of low rocky shore and 15kms
(6%) of beaches (Mather, Ritchie and Smith
1975). There are areas of moorland and hills
which rise to over 260 metres. The low lying
land is extensively cultivated, there are a
number of large lochs, some small plantations
and several quarries, some of which are in
regular use.

Annually, from 1972 to 1995, an effort was
made to locate all the nesting Ravens on
Mainland and, from 1983 to 1995, to record
the success of each breeding pair. In the
inland areas of the island, particularly where
there were no Suitable cliffs, quarries or trees
to provide nest sites, attention was paid to
derelict and disused buildings, especially if
they were situated at a distance from human
habitation. Additionally, during the study
period, any reports from other islands in
Orkney of Ravens nesting on buildings were

160 CJ Booth

noted and anumber of personal observations
were also made on some of these islands.

Results

The number of nesting pairs of Ravens on
Mainland recorded annually during the period
1972 to 1995 varied between 21 and 35, with
a total of 47 different territories occupied at
least once. The term ‘territory’ is used to
describe a locality such as a stretch of cliff or,
inland, an area of mainly agricultural land
where a pair of Ravens was nesting. There
were uSually several nest sites within a
territory. In some territories the same nest
site was used for several years in Succession
but in others an alternative site was chosen
the following year.

In one territory 3 types of nest site were used:
a building, a disused quarry and the steep
bank of a burn. In 29 (61.7%) territories the

Table 1 Types of nesting site used in 47
Raven territories from 1972 to 1995

Type of site No of territories

Sea cliff 15 metres and over 29
Sea cliff less than 15 metres 6
Building 5
Quarry (used and disused) 5
Tree 2
Inland cliff 1
Steep bank 1

nest sites were on sea cliffs 15 metres or over
in height, while 5 (10%) territories had nest
sites on buildings and in 4 of these it was the
only type of nest site used.

SB 18(3)

A nesting attempt was considered to have
taken place if a completed lined nest or anest
with eggs or young was found, and to have
been successful when it was known that one
or more young had fledged from the nest.
Brood sizes were of young definitely known to
have fledged and are therefore minima. Some
young may have died on fledging, or shortly
after, and before the visit to record the number
of young was made. Itis also possible that at
cliff sites some of the fledged young were
hidden from view.

As can be seen from Table 2, Ravens nesting
on buildings on Mainland were more
successful than those at other sites on the
island, although there was little difference in
brood size.

Details of nest sites on buildings were as
follows:

Territory 1 Nest site in the bell tower of a
ruined church. There are 2 openings in the
tower, one above the other. The upper
opening was used most frequently. The
churchis surrounded by fields and the nearest
occupied house about 200 metres away.

Territory 2 Four different buildings have
been used. On the first occasion the nest was
built on top of the east gable end of a ruined
and roofless chapel. The following year it was
built on girders under the galvanised roof of a
building on a disused wartime airfield. A local
farmer pulled the roof down just after the
young fledged, destroying the site. Next year
the birds moved to the top of a brick support
pillar on a nearby building. This site was
used, with varying success, over a period of
9 years. All the old wartime buildings were
demolished during the next winter and the
following year the pair built a nest on the floor
of a loft of a disused cottage but the young

1996 Ravens nesting in Orkney 161

Table 2 Breeding success of Ravens on Mainland, Orkney 1983-1995

No of No % Mean brood
attempts successful * successful size
Nests on
buildings 42 31 73.80 2.90
All other
nests 326 176 53.98 2.88

* Success means one or more young fledged from that nest

Raven Jim Young

162 CJ Booth

SB 18(3)

were shot when they were ready to fledge.
The birds then returned to the ruined chapel,
nesting about two thirds up on a gable end,
using both east and west ends in different
years. The farmer who owns the land where
the chapel is situated does not interfere with
the nests and young have been reared
successfully for the past 6 years. The chapel
is surrounded by fields and the nearest
habitation about 600 metres away.

Territory 3 Two ruined cottages have been
used, with 2 sites on each. On one ruin, the
nests were sited on the top of the chimney
stack and on the side of the gable end where
it meets the chimney. On the other the nest
was Situated at the base of the chimney stack
where some stones were missing, leaving a
sheltered recess. Unfortunately, after 7 years
the chimney began to lean and the pair moved
to the other side, then, shortly after the young
fledged, the chimney finally collapsed and the
site has not been used since.

Territory 4 At least 4 sites have been used
but, as the newly completed nests have been
destroyed at 2 sites, it is possible that there
have been attempts at other sites also and
that nests have been removed before | have
able to locate them. Of the known sites, one
was on top of a chimney stack of a disused
cottage, 2 were situated on the roofs of ruined
cottages and against the chimney stacks,
while the fourth was against the kiln end, on
the roof of an outbuilding of a disused cottage.
The nearest occupied houses at any of the
sites were about 200 metres away. Young
have been reared successfully at both the
first and last site.

Territory 5 This was an alternative site;
others used in the territory were in a disused
quarry and on the steep bank of aburn. The
nest was situated on collapsed wooden beams

and stone roof slabs between 2 buildings.
When these fell down further, the nest was
built ina gap under a roofing slab of one of the
buildings. The buildings were in a field about
600 metres from the nearest occupied house.

Nesting on buildings was also recorded from
some of the other islands:

North Ronaldsay Here a window opening in
the old lighthouse has been used.

Sanday Buildings have been used in 2
territories. In the south of the island a large
ruined house, which stands in the middle of a
field, has provided a number of nesting sites
over aperiod of at least 22 years. Nests have
been built on the window sills of the upper
floor and also against the chimney pots. The
other territory was in the north east of Sanday,
where the nest was on the roof of an
outbuilding of an unmanned lighthouse on a
small tidal island.

Stronsay The upper floor of a large ruined
building has been the site of nest over a
period of several years.

Egilsay The tower of a ruined church has
been used on at least one occasion.

Damsay Nesting has been attempted on a
ruined building on several occasions. Damsay
is a small low lying island only 800 metres
offshore and may be an alternative site for a
nearby territory on Mainland that is occupied
sporadically and where cliff nesting sites are
used.

1996

Discussion

In some inland areas of Mainland Orkney,
where there are no quarries, inland cliffs or
suitable trees, ruined buildings provide good
substitute nest sites. This is also the case on
several of the other Orkney Islands, especially
those that are low lying. Only 5% of the 96km
of coastline on Sanday and just 13% of the
62kms of Stronsay coastline consists of cliffs
over 15 metres in height (Mather et al 1975),
the type of site most frequently occupied on
Mainland. There are no cliffs over 15 metres
high on the islands of Damsay, Egilsay and
North Ronaldsay.

Breeding Ravens tend to be very wary, often
leaving the nest when humans appear in their
territories. A number of the buildings on
which Ravens nest are within 200 metres of
occupied houses and all on Mainland are
surrounded by agricultural land, so, in the
breeding season, there is usually some
disturbance, particularly from farming
activities.

Despite this, 73.8% (31 out of 42) of breeding
attempts on buildings on Mainland during the
years 1983 to 1995 were successful which
compared favourably with the success of
53.98% (176 out of 326) for breeding attempts
of Ravens at all other sites on Mainland for
the same period (Table 2).

The success of the nests on buildings
depended very much on the attitude of the
farmers on whose land the buildings were
situated. A few farmers destroyed nests
whenever they found them but others were
more tolerant, even taking the view that the
breeding Ravens may keep away birds such
as the Great Black-backed Gull Larus marinus
and Hooded Crow Corvus corone which they
considered to be a greater threat to their
sheep and lambs.

Ravens nesting in Orkney 163

In a study of the breeding success of Ravens
on Mainland Orkney from 1983 to 1995
(Booth 1985), human interference was noted
as the main known cause of nest failure.

Ravens have been recorded nesting on
buildings in other parts of Britain but this habit
does not appear to be as frequent or
widespread as it is in Orkney.

| have seen nests on a derelict mine building
in Cornwall and on an old wartime building in
Shetland. Nesting on ruined mine buildings
in Devon and derelict cottages in
Northumberland and Shetland was noted by
Holyoak and Ratcliffe (1968), while successful
nesting on the Guildhall, Swansea was
recorded by Hume (1975). On the Isle of Man
a building was used regularly as anestsite for
a number of years (Cullen 1978). Nesting has
also been reported on other man made
structures such as pylons and viaducts.

Nesting of Ravens on buildings is not confined
to Britain and there are records from elsewhere
inthe Western Palearctic (Cramp and Perrins
1994) and also North America (Bowles and
Decker 1930, Heinrich 1990). In a lowland
area of south west Iceland 48% of pairs used
man made structures, including abandoned
farm buildings (Skarphedinsson et al 1990),
although there were only scattered records of
these types of sites from other parts of Iceland.

Acknowledgments

| am very grateful to everyone who has, over
the years, supplied me with information on
Ravens nesting on buildings and particularly
the following: R G Adam, J Booth, M F
Cuthbert, J F Holloway, ER Meek, P Reynolds,
J B Ribbands, A Salisbury, R H Thorne, K F
Woodbridge. My special thanks go to P B
Heppleston for his comments on an earlier
draft of this paper, to C M Booth for help with

164 CJ Booth

translations and J Booth in preparing this
paper.

References

Booth C J 1985. The breeding success of
Ravens on Mainland Orkney 1983-1985.
Orkney Bird Report 1985: 59-62.

Bowles HJ & Decker F R 1930. The Ravens
of the State of Washington. Condor 32(4):
192-201.

Buckley T E & Harvie-Brown J A 1891. A
Vertebrate Fauna of the Orkney Islands
Edinburgh.

Cramp S & Perrins C M (eds) 1994. Birds of
the Western Palearctic Vol 8: 206-223 Oxford
University Press, Oxford.

Cullen P 1978. Acensus of breeding Ravens
in the Isle of Man. Peregrine 4: 264-273.

SB 18(3)

Heinrich B 1990. Ravens in Winter. Barrie &
Jenkins, London.

Holyoak D & Ratcliffe D A 1968. The
distribution of the Raven in Britain and Ireland.
Bird Study 15: 191-197.

Hume R A 1975. Successful breeding of
Ravens on city buildings. British Birds 68:
515-5 :1,6:

Low Rev G 1813. Fauna Orcadensis.
Published and edited by W Leach, Edinburgh.
Mather A S, Ritchie W & Smith JS 1975. An
introduction to the morphology of the Orkney
coastline. The Natural Environment of Orkney.
Nature Conservancy Council 1975.
Skarphedinsson K H, Neilson O K, Thorisson
S, Thorstensen S & Temple S A_ 1990.
Breeding biology, movements and
persecution of Ravens in Iceland. Acta
Naturalia Islandica 33: 1-45.

C J Booth, 34 High Street, Kirkwall, Orkney KW15 1AZ

Revised manuscript accepted December 1995

Hl

Raven nest in territory 5 Chris Booth

Scottish Birds (1996) 18: 165-169

165

The hunting behaviour of Merlins in winter in Galloway

R C DICKSON

Observations on the hunting behaviour used by Merlins in winter in west
Galloway between 1965-95 are summarised and categorised. Up to 8

methods were distinguished and details are given of these techniques. A
low level flight attack was the commonest method used by both male and

female/immature Merlins

introduction

In summer it is not clear exactly how Merlins
Falco columbarius do hunt (Bibby 1987)
although Newton et al (1984) state that they
usually chase and kill flying prey. In winter
and on migration, however, their methods are
relatively better known when it is easier to
study their foraging activities by direct
observation. Between 1965-95, 287 hunts by
Merlins were observed in various habitats in
west Galloway; these hunts were categorised
and the different methods used by both classes
of Merlins, blue (males) and brown (females
or juveniles) are shown in Table 1.

Results
Low flight attack

Merlins hunted from anything that served as
aperch, even the ground, and which provided
them with an adequate stance from which to
locate potential prey, often flocks of birds in
open country (‘still hunting’: see Cade 1982,
Dickson 1988). Fromthese positions Merlins,
both inland and on the coast, would launch
attacks using a low level approach less than
a metre above the ground. Such an attack
involved a very low fast, horizontal and direct
flight sometimes ending in a glide of 50-300 m
towards the intended prey which was usually

feeding on the ground. This caused the prey
to take flight ahead of the Merlin, which then
attempted to catch the bird on the wing close
to the ground. This method accounted for
64% of observed hunting flights. Both blue
and brown Merlins used this as their principal
method: 58% by blue (14.3% success), 65%
by brown (10.9% success). Dekker (1988)
also recorded surprise attacks on prey initially
on the ground by one of the North American
subspecies in 72.3% (9.4% success) of hunts
in Canada.

Height and stooping

Merlins would sometimes attack prey at a
height above 10 metres but, more often, it
was usually the end result of alow level chase
when the intended victim ‘towered’ apparently
trying to escape. The intended victim, whether
a wader or a passerine, seemed to try to keep
above the pursuing falcon, with both climbing,
sometimes out of sight (“exhaustion hunting’:
see Rudebeck 1951), but, as Meinertzhagen
(1959) points out, it is not clear whether the
prey is purposely driven up by the Merlin or
whether the prey deliberately gains height to
avoid a stoop. If the attack did not endina
long stoop, Merlins would get above the prey
and short, shallow stoops/swoops would
follow, sometimes ending close to the ground.
Merlins would also carry out short stoops if

SB 18 (3)

166 RC Dickson

Sa|IUeANI 10 Saye} = UMOIG ‘Sa;eW = aN|q :Se}ON

888] 555 9565650506655 0

L8¢ €¢ 6€¢ S 8YV s|e}O]
'0 | : : 0 | : : : ; Buiddoy jsod
O'l € : : oa € : : : ; Buyle1s
vt v : ; v0 | : : 79 € seyje6o} BuyunH
Lc 9 Cise C Gc 9 : 5 : - goundod
vc ZL. i = G2 9 - - tare} | soijoe} sounog
8°OL Le G9 4 (oer vz - - 971 I dooms
psemdn pue bulseyo jie]
SLI LS 8p g ZI ov iat | 8'°8I 6 Buidoojs pue yyb19}
L'v9 v8 GOL = ZI €°S9 9Sl e vt v €°8S 82d syoeye yyH1|} MO7
sjuny SUILO/\ SUILOY\ SUILO/\ SUILOY\
je}O] umoig Aq umosq Ag anjq Aq anjq Aq juny jo adA |
INJSSeDNNS sjuny JO ON InJssedonNS s}UNY JO ON

a a a a ee Ss SS Ss ae eee ee ee
SS a ee ee een eee

S6-S96I Aemoyjed }SaM Ul SUILJayW Aq pasn spoyjew jo ssao9ns Buljuny pue Aouanba.y uolersasqgo | e1qeL

——

1995

their victims sought cover and/or the prey
was crouching on the ground (see also ‘bounce
tactics’). It was the second commonest
method overall (17.8%) and was used by
both classes, 18.8% by blue (11.1% Success),
17.6% by brown (4.8% success). Buchanan
etal (1988) found itthe commonest method of
attack (45%, 7.3% success) on flocks of
Dunlins Calidris alpina in North America, but
the highest success rates there were during
low flight attacks (8.3%).

Tail chasing and upward swoop

If an initial attack failed, usually when the prey
apparently saw the falcon approaching atthe
last moment, Merlins would sometimes follow
very close behind the intended victim (‘tail
chasing’) persistently and doggedly pursuing
its victim’s every move, sometimes zig zagging
and terminating the pursuit either by swooping
up to try and catch the prey from below or by
a short stoop from above. It was possibly not
the most energy efficient way of hunting in
winter, especially if Merlins had to indulge in
prolonged, energy expending chases. This
method did not occur very often overall
(10.8%) and only 6.5% of these hunts were
successful.

Bounce tactics

In North America, ‘Merlins approach flight on
prey ended with a quick down thrust which
appeared to startle its prey into the air (Page
& Whitacre 1975). These ‘bounce tactics’
(Cramp & Simmons 1980) are probably used
to flush prey crouching on the ground but this
method (2.4%) was infrequently used in Merlin
attacks in west Galloway since most of the
intended prey had already taken flight before
Merlins had reached them. It did not account
for any successful attacks. Ash (1960),
however, recorded ‘bounce tactics’ as the

Hunting behaviour of Merlins in winter in Galloway

167

usual method of hunting Skylarks Alauda
arvensis in Hampshire.

Pounce (seizing prey on the ground)

Merlins would sometimes pounce on to the
ground if prey was not quick enough to get
airborne after a low flight attack, and, on a
very fewoccasions, Merlins would deliberately
pounce from a fence post after a very short
flight. This method was not often used or
seen in hunts in west Galloway (2.1%) but
accounted for 33.3% of successful hunts.
Other observers have recorded this more
frequently (eg Brennecke 1951, Rudebeck
1951).

Hunting together

Three hunts involved blue and brown Merlins
together and one huntinvolved 2 brown birds.
Hunts by 2 Merlins probably occurred by
chance and there was no indication that each
assisted the other; methods used were !ow
flight attacks, tail chasing and short stoops by
either bird. Hunting together was only seen
on 4 occasions (1.4%) but none ended in
success. Many other observers (eg St John
1888, Parker 1978) have recorded hunting
together and Bengison (1975) found that
31% of hunts in Iceland were by breeding
pairs.

Stalking

Merlins would sometimes approach prey
feeding on the ground by walking across the
ground keeping behind vegetation or contours
(Dickson 1979, 1988). The stalk on foot
usually ended witha shortflightand appeared
to be an extension or variant of the low flight
attack. It only occurred in 1.0% of hunts but
none ended in success. Stalks on foot were
usually only under 2 metres long although

168 RC Dickson

SB 18 (3)

Fleet (1993) saw a Merlin stalk Dunlin on foot
for over 30 metres in Germany.

Post hopping

Post hopping occurred when Merlins
approached flocks of passerines by hopping/
flying very short distances from one fence
post to another. This method may have been
a variant of stalking but was little used by
Merlins in these observations (0.4%) and
none were successful.

Discussion

Merlins employ a wide variety of hunting
techniques in winter facilitating exploitation of
a wide range of prey species (Dickson 1988).
Rudebeck (1951) identified at least 6 hunting
modes in Europe including 4 forms of stooping,
one horizontal attack and one ‘mobbing’. In
North America; Buchanan et al (1988)
distinguished 7 on one particular species
including 3 forms of stooping, 2 low flight
attacks, a flock chase and ‘ringing’. Up to 8
broad categories were distinguished in west
Galloway but hunts could, and did, commence
as one type of attack andendas acombination
of one of more methods. Nonetheless, in
attempts to catch their prey quickly, both blue
and brown Merlins consistently used short
distance low flight attacks. This method
probably also helped them take their prey by
surprise. Itseemedan efficient way of hunting;
of 28 successful hunts, 21 (75%) occurred
using low flight attacks. Most attacks (<80%)
lasted less than a minute.

Some published observations from the
breeding season provide direct comparisons.
Newton et al (1984) observed 2 methods of
hunting: fast low flying attacks and watching
and flying from a perch (‘still hunting’).
Bengtson (1975) also only recognised 2

methods in Iceland: surprise attacks (49%)
and persistent chases (51%). Radio tracking
of adults in Wales in summer suggested
Merlins may use a ‘sit and wait’ hunting
technique (‘still hunting’) ( Petty 1995). Sodhi
et al (1991), on the other hand, combined the
results of summer and winter hunts by radio
tagged urban Merlins in Canada, including
flights by 7 Merlins trained for falconry. They
only distinguished 2 main techniques: attack
from a perch (58% in summer, 95% in winter)
and cruising flights (87% in summer, 5% in
winter).

References

Ash J S 1960. Birds of prey numbers ina
Hampshire game preserve 1952-59. British
Birds 53: 285-300.

Bengtson S-A 1975. Jaktbeteende och
bytesval hos en islandosk population av
Stenfalk. Fauna och Flora Uppsala 70: 8-12.
Bibby C J 1978. Foods of breeding Merlins
Falco columbarius in Wales. Bird Study 34:
64-70.

Brennecke VEH 1951. Zur Jagdiveise des
Merlin-Falken. Vogelwelt 72: 82-84.
Buchanan J B, Schick C T, Brennan L A &
Harman S G 1988. Merlin predation on
wintering Dunlins. Wilson Bulletin 100: 108-
118.

Cade T J 1982. The Falcons of the World.
London.

Cramp S & Simmons K E L (eds) 1980. The
Birds of the Western Palearctic. Vol 2 Oxford.
Dekker D 1988. Peregrine Falcon and Merlin
predation on small shorebirds and passerines
in Alberta. Can. J. Zool. 66: 925-928.
Dickson RC 1979. Merlin apparently stalking
prey. Scottish Birds 10: 275.

Dickson R C 1988. Habitat preference and
prey of Merlins in winter. British Birds 81:
269-274.

Fleet D M 1993. Merlin stalking Dunlins on
foot. British Birds 86: 181.

1995

Meinertzhagen R 1959.
Predators. London.
Newton I, MeekER & Little B 1984. Breeding
season food of Merlins in Northumbria. Bird
Study 31: 49-56.

Page G & Whitacre D F 1975. Predation on
wintering shorebirds. Condor 77: 73-83.
Parker A 1978. Merlins hunting together
outside breeding season. British Birds 71:
3:

Petty SJ 1995. Merlins and Forestry. Forestry

Pirates and

Hunting behaviour of Merlins in winter in Galloway 169

Commission Research Information Note 254.
Rudebeck G 1951. The choice of prey and
modes of hunting of predatory birds with
special reference to their selective effect.
Oikos 3(II): 200-231.

Sodhi N S, Warkentin | G & Oliphant L W
1991. Hunting techniques and success rates
of urban Merlins. Journal of Raptor Research
25s 27-1Sile

St John C 1888. The Wild Sports of the
Scottish Highlands. London.

R C Dickson, Lismore, New Luce, Newton Stewart,

Wigtownshire DG8 OAJ.

Revised manuscript accepted December 1995

Merlins

Steven Brown

170 Scottish Birds (1996) 18: 170-181

SB 18(3)

Numbers, distribution and habitat associations of Corn
Buntings on the Outer Hebrides and Tiree in 1995

P F DONALD, P HINES, P JACKSON, J DUSTOW,

| HEPBURN, G F ALBON & S JERVIS

All known Corn Bunting populations on the Inner and Outer Hebrides were
visited in June and July 1995 to collect information on changes in numbers
since previous surveys, to relate these to any changes in habitat and to
measure current populations and distribution. The overall recorded
population on the Uists and Benbecula changed very little between 1983
(the year of the last fuil survey) and 1995, with an estimated population in
1995 of between 240 and 320 territories. Territory densities remain amongst
the highest in Britain. However, there was a considerable decline in
numbers on North Uist and an increase on South Uist between the 2
surveys. Territorial male Corn Buntings selected dry machair, particularly
cultivated dry machair, usually near coastal dunes. Territory density within
each count site was positively correlated with the proportion of the site
made up by cultivated dry machair, and on North Uist population declines
coincided with a 44% reduction in the area of this habitat. The population
increases on South Uist did not coincide with any measured gross habitat
change, suggesting that habitat quality had improved. On Tiree, only 6 to
10 Corn Bunting territories were found, compared with the 60-95 estimated
to be present during the 1970s. The importance and conservation of the
Hebridean populations of this species are discussed

This paper describes the results of a British
Trust for Ornithology (BTO) expedition to the
Uists, Benbecula and Tiree during June and
July 1995. The aims of the expedition were to
assess the distribution and population size of
Corn Buntings Miliaria calandra, to assess
recent population trends and to relate these
to changes in habitat.

The Corn Bunting is one of Britain’s most
rapidly declining bird species. Between 1970
and 1990 the British population declined by
over 70% (Marchant etal 1990) andthe range
contracted by over 30% (Gibbons et al 1993).

The declines are almost certain to be
connected with changes in farming practice,
although exactly which ones remains unclear
(Donald 1996). A historical review of regional
population and range changes has shown
that the current decline is the second to have
taken place this century, a similarly dramatic
decline having occurred during the 1920s
and 1930s (Donald et al 1994). During both
periods, declines were most severe in northern
and western regions of Britain, and particularly
in western Scotland. A national survey of the
species, undertaken by the BTO in 1993,
suggested that the British breeding population

1996

numbered around 20,000 territories in that
year (Donald & Evans 1995).

At the turn of the century, the population
density of Corn Buntings in the Inner and
Outer Hebrides was probably higher than
anywhere else in Scotland (Baxter & Rintoul
1953) and several contemporary accounts
describe the species’ extraordinary
abundance both on these isles and in coastal
regions of the adjacent mainland (eg Graham
1890; Harvie-Brown & Buckley 1892). The
species was even common on islands as
remote as St Kilda at the end of the last
century. Declines appear to have started
during the early years of the 20th century,
with the first island extinctions occurring during
the early 1930s (Donald etal 1994). Declines
have continued to such an extent that, by the
time of the second Atlas of Breeding Birds
(1988-91), breeding was recorded in western
Scotland only on the Uists and Benbecula, on
Tiree and at a few scattered localities on the
adjacent mainland.

In 1983, the Nature Conservancy Council
(NCC) undertook a survey of breeding Corn
Buntings on the Uists and Benbecula (Williams
etal 1986). They estimated a total population
of between 260 and 350 territories, the majority
(over 70%) on North Uist. At the time, fears
were expressed that the loss of traditional
strip cultivation on the machair (a narrow,
cultivable shell sand plain with a distinctive
flora) to improved grassland would cause
populations to decline (Williams et al 1986).
Such declines were observed on North Uist,
where the population fell from 117-149
territories in 1983 to around 100 in 1988 and
possibly as few as 60 in 1991 (Donald et al
1994). These declines coincided with an
increase in the area of silage, a decline in
traditional strip rotation cultivation and
increased fencing, allowing more intensive

Corn Buntings on the Outer Hebrides and Tiree 171

stocking regimes (R J Fuller pers comm).
Population trends on Benbecula and South
Uist are unrecorded.

Corn Buntings on Tiree were surveyed
regularly between 1969 and 1987 (Cadbury
1989). A population of 60-95 territories
between 1971 and 1979 declined sharply in
the early 1980s to less than 20, although this
number increased slightly during the late
1980s. Population trends thereafter are
unrecorded.

Methods

On the Uists and Benbecula, each of the 14
sites visited during the 1983 survey (Fig 1)
was revisited between 17 June and 29 July
1995, the period of peak song activity in this
late nesting species. The bulk of the fieldwork
was carried out in the last 2 weeks of June,
with follow up visits to selected sites (those
which were only visited once in June or those
which were covered in poor weather) in July.
6054 ha of machair, sand dunes, fen and
blackland (an intermediate grassland between
the machair strip and peatland) were covered,
an area accounting for 63.4% of the total area
of lowland habitats on the Isles (Williams et al
1986). Further details of the lowland habitats
on the Hebrides are given in Fuller et al
(1986). Between 2 and 4 visits were made to
each site. The weather throughout most of
the fieldwork period was hot, calm and sunny.
In such conditions, singing birds could be
detected at distances of up to 500m. On
Tiree, all Known previous breeding sites and
all apparently suitable areas were visited at
least once between 17 and 24 June 1995. A
territory mapping technique (Marchant et al
1990) was used. Birds were recorded on
1:10,000 maps using appropriate activity
codes to denote birds singing, calling etc and
to distinguish registrations relating to the same

172 P Donald et al SB 18(3)

Figure 1 The location of the 14 study sites where Corn Buntings were counted in
1983 and 1995 on the Uists and Benbecula. Numbering follows that of Williams et al
(1986)

16

South Uist

10km

1996

or to different individuals. Population sizes at
each site were calculated in the same way as
in the 1983 survey, using “minimum” counts
(the highest count recorded at each site on
any one visit) and “maximum” counts, which
used records from all visits, activity codes and
known nearest neighbour distances to
estimate the number of territories present.
The latter method is comparable with the
methods of the Common Birds Census
(Marchant et al 1990) although fewer visits
were made in the present survey than are
required by the CBC. The figures obtained for
the area surveyed were multiplied up to take
account of the 3501 ha of lowland habitats
which were not surveyed. In doing this, it was
assumed that territory densities outside the
study areas were the same as those within.
Some data collected outside the study areas
suggested that densities may have been
slightly lower, on average, than those within
the study areas, although habitats were
similar. This would lead to a slight
overestimation of the total population but
does not compromise comparisons of overall
population changes between the 2 years.
The BTO expedition and a Royal Society for
the Protection of Birds (RSPB) wader survey
team between them resurveyed all the habitats
measured in 1983 to assess habitat change.
Such changes could then be related to any
observed changes in Corn Bunting
populations.

As each singing male Corn Bunting was
located, the relative proportions of the
following habitats within a 100m radius of the
bird were assessed by eye: open shore, sand
dunes, uncultivated dry machair, cultivated
dry machair, uncultivated wet machair,
cultivated wet machair, fen, blackland and
open water. Recent fallow was classified as
uncultivated because the vegetation structure
was usually more similar to that habitat than

Corn Buntings on the Outer Hebrides and Tiree 173

to cultivated machair. The extent of wet and
dry machair was assumed to have been the
same as that measured onthe basis of floristics
for the 1983 survey of breeding waders and
used by Williams et al (1986) in their analyses
of Corn Bunting habitat selection. These data
on habitats within territories were only
collected on the first visit to each site, since if
data were collected for the same male 2 or 3
times, they would not be independent (ie
there would be problems __ of
pseudoreplication). Habitat data were
collected from a high proportion of all the
males present in the study areas (a minimum
of 63%) and were used to derive some
measure of habitat association. Because the
data collected were proportions and therefore
not independent (ie if the relative area of one
particular habitat type is high, the relative
areas of the others must be low), these data
were analysed using a simple measure of the
frequency with which each was found within
male Corn Bunting territories. These could
then be compared with the availability of each
habitat to derive a simple assessment of
habitat selection.

Results

Numbers and distribution on the Uists
and Benbecula

Table 1 gives the “minimum” and “maximum”
counts of territorial males in each of the 14
study areas shownin Fig 1. Multiplying up the
counts from the areas covered (to account for
the areas not covered) yielded a 1995
population estimate for the Uists and
Benbecula of between 240 (based on
minimum counts) and 320 (based on
maximum counts) territories compared with
the 260 and 350 estimated using the same
methods in 1983. As indicated in the Methods
above, the 1983 and 1995 estimates for the

174 P Donald et al SB 18(3)

Table 1 Calculated “minimum” and “maximum” Corn Bunting territories in
each of the study areas depicted in Fig 1 (see text for definitions). The total
area of the sites is slightly greater than the area actually surveyed since it
includes areas of unsuitable habitat (road, human habitation etc)

Site Area (ha) Minimum count Maximum count
South Uist

1 395 16 21
2 372 7 13
3 554 US) 22
4 674 4 8
5 575 0 0
6a 234 0 0
6b 554 26 35
Benbecula

10 496 10 10
14 520 9 14
North Uist

4 479 UL 18
9a 387 16 24
9b 168 0 0
16 STi 27 32
it, 630 5 6
Total 152 204

Table 2 Numbers of Corn Bunting territories recorded on 14 study sites in the Uists
and Benbecula in 1995. Figures from the 1983 survey (from Williams et al 1986) are
given in brackets

Minimum count Maximum count Area surveyed (ha)
North Uist 65 (117) 80 (149) 2106
Benbecula 19 (17) 24 (24) 940

South Uist 68 (22) 99 (35) 3008

1996

total Outer Hebridean population may be
slightly too high. The figures suggest a decline
in the overall population of less than 10%.
However, the proportions of the total
population recorded on North and South Uist
changed markedly (Table 2). Populations on
North Uist were found to have declined by
nearly 45%, whereas those on South Uist
increased by over 200%. Populations on
Benbecula were virtually unchanged. On
North Uist, population declines were apparent
in all study areas between 1983 and 1995,
although the decline may have stabilised
since 1991, when approximately the same
number of territories was found as in 1995
(Donald et al 1994). On South Uist,
populations increased most markedly in study
areas 1 and 6b, where the combined
“minimum” number of territories increased

Corn Buntings on the Outer Hebrides and Tiree 175

from 19 in 1983 to 41 in 1995. Not all
populations on South Uist increased; for
example, study area 6a, which held a
population of around 5territories in 1983, was
found to have been abandoned by 1995 (Table
1).

Recorded densities of singing male Corn
Buntings in occupied study areas (calculated
using “maximum” counts) ranged from 1.0
territories /km2 in study area 17 to 6.3
territories /km@ in study area 6b (mean = 3.8
/km2). On the machair strip alone (ie coast,
dunes, wet and dry machair without open
water, fen or blackland), densities in occupied
aie sites ranged from 1.3 to 7.3 territories
/km< (mean= 5.1 /km?). The highest density
recorded in a single 1km square approached
15 territories /km< (study area 16, which

Table 3 Areas (ha) of cultivated and uncultivated machair on study areas in the
Uists and Benbecula in 1983 and 1995. Measurements of the areas of other crop
types (such as hay and silage) were not available. Areas of non crop habitats
(blackland, dunes, open water and shore) were assumed not to have changed

1983
North Uist machairs
Cultivated dry 404.7
Uncultivated dry 938.4
Cultivated wet 65.7
Uncultivated wet 458.1
Benbecula machairs
Cultivated dry 56.3
Uncultivated dry 275.4
Cultivated wet 9.3
Uncultivated wet 176.5
South Uist machairs
Cultivated dry 222.7
Uncultivated dry 728.6
Cultivated wet 117.6
Uncultivated wet 1063.5

1995 % change
228.7 -43.5
1114.4 +18.8
11.6 -82.3
512.2 +11.8
78.6 +39.6
25S -8.1
tat +40.9
27 -2.2
201.8 -9.4
749.5 +2.9
75.4 -35.9
1105.7 +4.0

176 P Donald et al

included Balranald RSPB reserve).
Corn Buntings and habitat change

Changes in the areas of the main lowland
habitat types on North and South Uist and
Benbecula are shownin Table 3. No evidence
was found for changes in the areas of dunes,
fen, blackland or open water between 1983
and 1995 so the areas of these habitats were
not measured in 1995. The population
declines on North Uist coincided with a decline
in the area of cultivated dry machair of 43.5%.
Similarly, the extinction on South Uist of the
small population in study area 6a coincided
with a reduction in the area of cultivated dry
machair from 11.4 hato 2.1 ha. However, the
increase in the population of South Uist as a
whole did not coincide with changes in the
area of any measured habitats (Table 3).
Even in study areas 1 and 6b, where
population increases were greatest, there
were no substantial differences in the extent
of any of the habitat types measured between

SB 18(3)

the 2 years.

Habitat associations of Corn Buntings on
the Uists and Benbecula

Corn Bunting territory densities were positively
correlated with the proportion of each study
site made up by cultivated dry machair (Fig 2)
and total cultivation (r, =0.94, n= 14, P<0.001)
but not with cultivated wet machair (r, = 0.40,
n = 14, ns). There was also a significant
correlation between Corn Bunting territory
density and the proportion of the machair
strip alone (defined above) made up by the
cultivated dry machair (rg = 0.78, n=14,
P<0.001). Cultivated dry machair was also
found to be importantin the analysis of territory
habitat data (Fig 3). Of the 127 territories for
which data were available, 94% were found
on dry machair, and 78% of these (73% of all
127 territories) contained cultivated or recently
cultivated dry machair. Dunes or other coastal
habitats were recorded in 58% of the 127

Figure 2 Correlation between Corn Bunting territory density (calculated from
“maximum” counts) and the proportion of each study site made up by cultivated dry

machair (rg = 0.93, n = 14, P<0.0001)

Corn Bunting territory density (males per sq km)

p
0 0.02 004 006 008 0.1

Cultivated dry machair as a proportion of total area

0.12,.0:14 0.16 O18 Bie

® South Uist * North Uist ® Benbecula

1996

Corn Buntings on the Outer Hebrides and Tiree 177

Figure 3 Habitat availability and associations of territorial male Corn Buntings on
the Uists and Benbecula. Black bars show the availability of each habitat (% of total
combined area of the 9 habitats), grey bars show the percentage of territories (n=
127) holding each of the 9 habitats (at least 5% by area). Habitats are listed in
decreasing order of availability. BLK = blackland; UDM = uncultivated dry machair;
UWM = uncultivated wet machair; DUN = sand dunes; CDM = cultivated dry machair;
WAT = open fresh water; FEN= fenland; SHO = open shore; CWM = cultivated wet

machair

Habitat type

BLK
UDM|

CWM|

304.30 20,2015

% of total area

territories, yet these habitats made up only
16% of the total study area. Less than 4% of
territories contained wet machair, a habitat
occupying 72% of the area of dry machair (Fig
3). Across all study sites combined, cultivated
dry machair made up only 5.1% of the total
area, yet was found in over 70% of territories.
Cultivated dry machair was found within Corn
Bunting territories significantly more frequently
in relation to its availability than was cultivated
wet machair (adjusted x2 =815)'P<0:01):

Corn Buntings on Tiree

We estimated that between 6 and 10 widely
spaced territories were present on Tiree.
These birds were all present in areas where
the species previously bred in higher numbers

2 0 20 40 60 80 100

% of territories holding habitat

(Cadbury 1989). Territories appeared to be
more associated with grazed land and hay or
silage meadows than on the Uists and
Benbecula and less associated with coastal
habitats, although sample sizes were too
small for statistical comparison. The current
low numbers reflect a severe decline in the
population levels of the late 1970s.

Discussion

These results suggest that Corn Buntings
fared considerably better on the Outer
Hebrides between 1983 and 1995 than
populations elsewhere in Britain. The overall
population of the Outer Hebrides declined by
less than 10% compared with a decline of
over 50% recorded by the Common Bird

178 P Donald et al

SB 18 (3)

Census across Britain as a whole during the
same period. However, some populations
declined considerably, particularly on North
Uist, where declines were more similar to the
national trend. At least in some areas of
North Uist, these declines form part of a long
term trend which started well before 1983
(Donald et al 1994). At Balranald RSPB
reserve (study area 16), there were 63-66
territories in 1977, 56 in 1983 (Williams et al
1986) and 27-32 in 1995.

The declines on North Uist coincided with a
reduction in the area of cultivated dry machair.
However, marked increases in some
populations on South Uist were not correlated
with any measured habitat change and an
increase in the area of cultivated dry machair
on Benbecula (Table 3) did not result in an
increase in Corn Bunting populations there.
Population changes could, therefore, only be
partly explained by changes in the areas of
any of the gross habitat types measured.
From this, it seems likely that changes in
habitat quality were as important as changes
in habitat availability. The crude methods of
habitat categorisation used might have
masked a considerable change in habitat
structure or suitability. Donald & Forrest
(1995) also found that changes in Corn Bunting
densities correlated very poorly with gross
changes in habitat in a sample of farmland
Common Bird Census plots. A number of
changes in habitat quality could have affected
Corn Bunting populations. The availability of
food, particularly in winter, is likely to have
been important in declines of British Corn
Bunting populations (Donald & Evans 1994,
Donald 1996) and the same may apply to
some Hebridean populations. Contemporary
literature suggests that, at the turn of the
century, corn ricks and stack yards were very
important to Corn Buntings on the Hebrides
(eg Graham 1890) and recent observations
suggest that this is still the case (M Shepherd

& | R Hartley, unpublished data). However,
these are now less common features of
machair agriculture due to the increase in
silage production. The warming effects of the
Gulf Stream may have contributed to the
survival of the Hebridean populations by
reducing the birds’ feeding requirements in
winter. The tendency of declining populations
to survive longer or in higher numbers in
coastal areas has been pointed out previously
(Donald et al 1994).

Densities of singing male Corn Buntings on
the Outer Hebrides remain high relative to
populations studied elsewhere in Britain. Ina
study in the Fens of Lincolnshire, where
densities are amongst the highestin mainland
Britain, the highest density recorded was 9
territories /km2 (Gillings & Watts 1996)
compared with the highest density of 15
territories /km2 on North Uist. Average territory
density in the Fens was around 2.8 territories
/km“, compared with 3.4 territories /km@ in
lowland habitats on the Uists and Benbecula
as a whole and 5.1 territories /km< in the
machair strip alone. This suggests that the
low intensity agriculture of the machair is
favourable for this species, despite occurring
at the northernmost limit of its world range.

The habitat associations of Corn Buntings
recorded by the current survey accord well
with those of previous surveys. Williams et al
(1986), Hartley et al (1995) and Hartley &
Shepherd (1996) all noted the importance of
cultivated and uncultivated dry machair and
dunes and the avoidance of wet machair,
fens and blackland. This habitat selection
could arise either from preferences for the
types of vegetation growing on dry machair
and/or in the adjacent dunces (where many of
the nests were found, Hartley et al 1995) and
its associated food resources, or from
preferences for dry, sandy soils, which have
been noted elsewhere (eg Gillings & Watts

eo

1996

1996). Cultivated dry machair was found by
Hartley et al (1995) to be the most preferred
foraging habitat of female Corn Buntings when
provisioning nestlings but was avoided for
nesting. The avoidance of wet machair might
result from the positive association of birds
with dunes, since wet machair generally
occurs further inland, and therefore further
from the coastal dunes, than dry machair.

Conservation of the few remaining Corn
Bunting populations in the Hebrides is of
considerable importance to the future of this
species in Scotland. Historical evidence
clearly shows that the Hebrides, and the
coast of western Scotland, have the potential
to support extremely high densities of Corn
Buntings. Should the factors which are
causing the present decline be halted or
reversed, repopulation of abandoned islands
and the adjacent mainland would be more
likely to occur if Some populations still exist,
because the species appears to be fairly
sedentary. Although the ability of the
Hebridean population to recolonise distant
areas appears to be low at present (Shepherd
et al 1996), the species is capable of rapidly
recolonising areas close to its existing range
when conditions there become suitable
(Hustings 1996). The Hebridean populations
are now the only viable ones in western
Scotland, and indeed amongst the few
remaining in western Britain, and could act as
a source for the future recolonisation of
abandoned areas. The conservation of these
populations depends upon an understanding

__ of responses of birds to changes in land use.
_ The data presented above suggest that, so
_ far, the Hebridean populations have survived
_ better than most mainland populations.
_ However, the vulnerability of these isolated
_ populations is reflected in the severe declines
_ which have occurred on North Uist and Tiree.

The future of the Hebridean Corn Buntings

_ should not be assumed to be safe and

Corn Buntings on the Outer Hebrides and Tiree

19

populations need to be monitored at regular
intervals.

The inclusion in 1994 of most of the machair
of the Uists and Benbecula into the
Environmentally Sensitive Areas (ESA)
scheme could be of considerable benefit to
the conservation of Corn Buntings on the
Outer Hebrides. Amongst other things, the
scheme encourages farmers to reduce
pesticide inputs, which are already very low,
to maintain existing machair and to keep at
least 15% of ploughable machair in arable
crop or recent fallow (SOAFD 1994). Such
measures are also likely to benefit the
internationally important populations of
breeding waders on the machair (Fuller et al
1986). Geographical differences in the take
up of this voluntary scheme might explain the
differing population trends of Corn Buntings
on North and South Uist but such data are
confidential and were not available to test this
hypothesis. The introduction of a similar
scheme on Tiree might help the recovery of
the dwindling population there.

Acknowledgements

This project would not have been possible
without the help and encouragement of many
people. We would particularly like to thank
Des Thompson, Mike Shepherd and John
Love of Scottish Natural Heritage and Rob
Fuller of the BTO for their help and advice
throughout, and Craig Whyte, Steve Babbs,
Stuart Murray and Tony Cook of the RSPB for
collecting additional Corn Bunting data and
for sharing the work of collecting and collating
habitat data. We are also very grateful to
South Uist Estates Ltd., The Scottish Office
Agriculture and Fisheries Department
(SOAFD), North Uist Estate and The Church
of Scotland for granting access to their land

180 P Donald et al

SB 18(3)

and to Mark O’Brien, George Boobyer, Cath
Jeffs, Phil Benstead, Jim Reid, Gwen Evans,
Claire Forrest, SuGough andLt Col C Stoddart
for their help in various ways. We would also
like to thank the crofters of the Hebrides for
the hospitality they extended to us during our
visit. This work was funded by the Joint
Nature Conservation Committee on behalf of
Scottish Natural Heritage, English Nature,
the Countryside Council for Wales and the
Department of the Environment (Northern
lreland). Generous additional funding was
provided by the Rosemount Trust and the
Adrian Darby Charitable Trust. This paper
was greatly improved by comments from Rob
Fuller, Mike Shepherd, Des Thompson, the
Editor and an anonymous referee.

References

Baxter E V & RintoulL J 1953. The Birds of
Scotland. Edinburgh.

Cadbury C J 1989. Corncrake and Corn
Bunting status and habitats on Coll and Tiree.
In: The Birds of Coll and Tiree; Status, Habitats
and Conservation. ed by D A Stroud. NCC/
SOC, Edinburgh.

Donald P F 1996. The Corn Bunting Miliaria
calandra in Britain: a review of current status,
patterns of decline and possible causes. In:
Donald P F & Aebischer N J (Eds) The Ecology
and Conservation of Corn Buntings Miliaria
calandra. JNCC (UK Nature Conservation,
No 13) Peterborough.

Donald P F & Evans AD 1994. Habitat
selection by Corn Buntings Miliaria calandra
in winter. Bird Study 41:199-210.

Donald P F, WilsonJ D & Shepherd M 1994.
The decline of the Corn Bunting. British Birds
87:106-132.

Donald P F & Evans AD 1995. Habitat
selection and population size of Corn Buntings
Miliaria calandra breeding in Britain in 1993.
Bird Study 42:190-204.

Donald P F & Forrest C 1995. The effects of

agricultural change on population size of Corn
Buntings Miliaria calandra on individual farms.
Bird Study 42:205-215.

Fuller R J, Reed T M, Buxton N E, Webb A,
Williams T D & Pienkowski M W_ 1986.
Populations of breeding waders (Charadrii)
and their habitats on the crofting lands of the
Outer Hebrides, Scotland. Biol. Cons. 37:333-
361.

Gibbons DW, ReidJB & ChapmanRA 1993.
The New Atlas of Breeding Birds in Britain
and Ireland: 1988-1991. T & A D Poyser,
London.

Gillings S & Watts PN 1996. Habitat selection
and breeding ecology of Corn Buntings Miliaria
calandra in the Lincolnshire Fens. In: Donald
P F & Aebischer NJ (Eds) The Ecology and
Conservation of Corn Buntings Miliaria
calandra. JNCC (UK Nature Conservation
No 13) Peterborough.

Graham HD 1890. Birds of lona and Mull.
Edinburgh.

Hartley |R, Shepherd M & Thompson DBA
1995. Habitat selection and polygyny in
breeding Corn Buntings Miliaria calandra.
Ibis. 137: 508-514.

Hartley | R & Shepherd M 1996. The
behavioural ecology of breeding Corn
Buntings Miliaria calandra on North Uist. In:
Donald P F & Aebischer N J (Eds) The
Ecology and Conservation of Corn Buntings
Miliaria calandra. JNCC (UK Nature
Conservation No 13) Peterborough.
Harvie-Brown J A & Buckley TE 1892. A
vertebrate fauna of Argyll and the Inner
Hebrides. Edinburgh.

Hustings F 1996. The decline of the Corn
Bunting Miliaria calandra in The Netherlands.
In: Donald P F & Aebischer N J (Eds). The
Ecology and Conservation of Corn Buntings
Miliaria calandra. JNCC (UK Nature
Conservation No 13) Peterborough.
Marchant J H, Hudson R, Carter S P &
Whittington P 1990. Population Trends in
British Breeding Birds. BTO, Tring.

|
)

1996

Scottish Office Agriculture and Fisheries
Department (SOAFD) 1994. Machair of the
Uists and Benbecula, Barra and Vatersay
Environmentally Sensitive Area; explanatory
leaflet for crofters and farmers. SOAFD,
Edinburgh.

Shepherd M, Hartley | R & McGregor P K
1996. Natal philopatry and breeding site

Corn Buntings on the Outer Hebrides and Tiree 181

fidelity of Corn Buntings Miliaria calandra on
the Uists. In: Donald P F & Aebischer N J
(Eds) The Ecology and Conservation of Corn
Buntings Miliaria calandra. JNCC (UK Nature
Conservation No 13) Peterborough.
Williams T D, Reed TM & WebbA 1986.
Population size, distribution and habitat use
of the Corn Bunting inthe Uists and Benbecula.
Scottish Birds 14:57-60.

P F Donald, P Hines, P Jackson, J Dustow, | Hepburn, G F Albon &
S Jervis, British Trust for Ornithology, The Nunnery,
Thetford, Norfolk IP24 2PU

Revised manuscript accepted January 1996

4
\

mn

BN
N

Corn Bunting

; a i}
, ‘ile 4 y;’
N) /} fi*4

182 Scottish Birds (1996) 18: 182-186

SHORT NOTES

Hunting times of Hen Harriers in winter

The Hen Harrier Circus cyaneus is essentially
an open country species. In winter in west
Galloway Hen Harriers hunt during the day
over many habitats ranging from low lying
farmland to the higher moors (Dickson 1992,
The Birds in Wigtownshire Wigtown).
Although easy to observe in these habitats,
there is very little information on the times
they hunt during the course of a winter’s day.
| watched Hen Harriers, spending an equal
amount of observation time at different hours
of the day, between August and March 1965-
95. | timed and pooled every hunting foray,
categorised as flying at low speeds
systematically quartering the ground, whether
successful or not by both classes of harriers,
grey (adult males) or brown birds (females or
juveniles). No hunts are included, however,
if harriers were on breeding moors in August
or at their roosts in winter.

SB 18 (3)

Figure 1 gives the results of 795 hunts
recordedin west Galloway. During November
to February there was a late morning peak at
1100 hrs (GMT) and avery similar peak in the
afternoon at 1500 hrs (GMT) with a definite
lull in hunting activity at midday. A similar
pattern emerged between August and March
despite the longer daylight. The need to
replenish energy stores lost overnight is
suggested by the gradual build up of hunting
activities in the morning, reaching a peak of
activity some 2 to 3 hours after the birds had
left their roosts. Another peak of activity in
late afternoon suggests that there was also a
need to replenish energy stores before
roosting.

These findings agree with Hamerstrom &
Wilde (1973, Inland Bird Banding News
45:123-127) who found that most hunting by
Northern Harriers Circus cyaneus cyaneus in
North America is done during morning hours
with a midday lull, followed again by a late
afternoon period of hunting. Similar hunting

Figure 1 Hunting times of Hen Harriers in west Galloway, 1965-95

30,.—
80 —
imo
60)

50 — a

Hunting forays in 30 min. periods

7.00
8.00
9.00
10.00
11.00
12.00

a
ov
ot . --o-- Aug - Mar
! * —=— Nov - Feb
*
a 2
Q
re
5 q.
a ‘h
4 i=}
¢ Te par
b ~ Sk fe
‘p-a
oS i=] o o iJ oa tJ o
ioe] wr uw © ~ o bor) eo
— — — — = —= - N

Time of day (h)

patterns occur in other birds of prey in winter
such as Sparrowhawks Accipiter nisus
(Newton 1986, The Sparrowhawk, Poyser,
Calton) and Merlins Falco columbarius
(Dickson 1993, Scottish Birds 17:56-57), but

Short Notes 183

a different pattern was found in Peregrines
Falco peregrinus when peak hunting activities
occurred at midday with a lesser peak in the
late afternoon (Dickson, British Birds in press).

R C Dickson, Lismore, New Luce,
Newton Stewart, Wigtownshire DG8 OAJ.

Revised manuscript accepted December 1995

Oystercatchers nesting in clear cut
forestry plantations

During May 1994 a pair of Oystercatchers
Haematopus ostralegus was seen on several
occasions over a newly clear cut area within
a Forestry Commission conifer plantation in
the Cree Valley, Dumfries and Galloway. The
patch of ground was bounded on all 4 sides
by mature Sitka Spruce Picea sitchensis
although on one edge only a narrow strip of
trees divided the clear cut area from the river.
The nearest pasture was 0.4km distant. There
was no ground flora present on the clear cut,
only a litter of dead spruce needles,
interspersed by tree stumps and dead
branches. On 27 May | was surprised to
watch one of the birds return to a nest with 2
eggs at the foot of alow stump. On6 June the
scrape was empty, with no sign of the birds.

| was unaware of any previous nesting attempt
by Oystercatchers on clear cuts, but by chance
learned that Forest Officer Andy Walker had
discovered a nest, also in May 1994, ona
15ha clear cut of Japanese Larch, Larix

kaempferi, felled in the previous year, in the
valley of the Water of Ae, about 70km east of
the Cree Valley site. This clear cut was
bounded on 3 sides by mature plantation,
with the nest situated 40m from a grass
pasture on the fourth side. The nest, which
held 2 eggs on 13 May, was on compacted
brash. Incubation continued at least until 25
May; on 17 June the scrape contained one
unhatched egg.

Oystercatchers have been extending their
range through the river valleys of south west
Scotland since the late 19th century (E J M
Buxton. 1961. The inland breeding of the
Oystercatcher in Great Britain, 1958-59. Bird
Study 8: 194-209), and are widespread in the
catchments of both the Ae and the Cree.
Tilled land is relatively uncommon in these
pastoral areas and most nests are found on
river shingle or roadside gravel. Both
situations provide a dry, bare substrate which
reflects radiant heat, and these features are
shared by the clear cuts. The utilisation of
clear cuts for nesting presumably allows the
birds to exploit nearby feeding areas where
suitable nest sites are scarce.

Geoff Shaw, Forest Enterprise, South Scotland Region,

55 Moffat Road, Dumfries, DG1 1NP

Accepted March 1996

184 Short Notes

Oystercatcher apparently rearing
Lapwings

On 27 May 1995, at Colquhar, Leithen Water,
Borders, | noted a pair of Lapwings Vanellus
vanellus mating ona vegetated shingle island.
This may have been a second breeding
attempt since a single bird, which had been
noted sitting on a nest nearby on 13 May, was
absent on 20 May. On 4 June, aLapwing was
noted sitting on a nest. The nest was not
visited due to the danger of trampling by
sheep or predation by Jackdaws Corvus
monedula.

Oystercatcher David Mitchell
On 10 June, | noted that an Oystercatcher,
Haematopus ostralegus, was incubating a
clutch atthe same spot as the Lapwing having,
| assumed, displaced it from its nest. Again
this nest was not disturbed to save it from
potential damage by sheep or Jackdaws. An
Oystercatcher was also noted incubating on
18 and 24 June.

On 1 July, at 0600 GMT, | saw that there was
no Oystercatcher at the nest site, but there
was one in an adjacent field. On leaving my
car to check this bird it rose, giving alarm

SB 18 (3)

calls, and another bird lifted from a patch of
shingle at the side of the burn. As | made my
way to this spot with the intention of measuring
and ringing any Oystercatcher chicks, | was
astonished to see 3 recently hatched Lapwing
chicks. After measuring and ringing them, |
returned to my car and watched one of the
Oystercatcher pair return to brood them, while
the other stood guard nearby. | did not see
either adult provide any food for the chicks
and | left at 0630 GMT.

On 8 and 13 July (my last visit to the area) |
noted that both Oystercatchers were still
actively alarming. On both occasions |
retrapped the same chick. Its bill length (to
feathers) had increased from 11.5mm on 1
July to 18.4mm on 13 July, and it seemed in
reasonable condition.

It seems that the Oystercatcher took over a
Lapwing clutch during the early stages of
incubation, possibly because they had lost
theirown clutch nearby, hatched it successfully
and reared the brood. Lapwings were present
throughout this time, but is not known if they
were responsible for the clutch. Lapwing
chicks find their own food shortly after they
become mobile and adult Oystercatchers
carry food to their chicks. It would have been
interesting to see if the Lapwing chicks
accepted any food provided by the
Oystercatcher, but such observations were
not made.

| am grateful to Rosebery Estates for
permission to study birds on their land and |
particularly thank Mr Templeton, the shepherd
at Colquhar, for his tolerance of my activities.

Tom Dougall, 62 Leamington Terrace,
Edinburgh EH10 4JL

Accepted October 1995

1996

Short Notes 185

Dunnock

Dunnock feeding in Spruce canopy

Inthe autumn of 1993 the plantations of Sitka
Spruce Picea sitchensis in the Galloway
Forest Park, in south west Scotland bore a
heavy seed crop. In order to observe birds
feeding on the cones, | constructed a small
hide in the canopy of a 30 year old stand of
trees. Observations were made at a distance
of 2-20m on clusters of cones at a height of
12-17m in the spruce crowns near the forest
edge.

From midday on 23 November the cone bracts
were open in warm sunshine, and the
accessible seeds were being taken by mixed
flocks of birds, which includes 3 species of tits
Paridae and 4 finches Fringillidae. Most
species fedin an inverted position, by clinging
to the cone, but the 2 heavier finches present,
Chaffinch Fringilla coelebs and Brambling F
montifringilla, fed by perching on a branch
and reaching over to insert the bill between
the open bracts of the nearest cone. Fora
brief period around 1400 hrs asingle Dunnock
Prunella modularis appeared in the canopy
and took seeds directly from a cone, by

Brian Turner

adopting the ‘perch and reach’ method of the
larger finches.

Dunnocks are predominantly ground feeders,
although feeding in vegetation at heights of
up to 8m has been recorded in summer;
seeds can form up to 90% of the winter diet (G
Bishton in BWP vol 5). The species is
widespread in conifer plantations. (R J Fuller
1982. Bird habitats in Britain. Poyser, Calton;
A Bevington 1991. Habitat selection in the
Dunnock in northern England. Bird Study 38:
87-91), and it would be surprising if spruce
seeds were not taken from the ground, where
they are often abundant after a heavy crop
has been shed. However, BWP quotes only
a single published record, of seeds of the
Norway Spruce P abies , obtained in a series
of stomach samples from nestlings in a
Russian study.

Geoff Shaw, Forest Enterprise,

South Scotland Region,
55 Moffat Road, Dumfries DG1 1NP

Accepted March 1996

1996 Scottish Birds (1996) 18: 186-189

OBITUARIES

Dr Norah Armstrong

Died 1996

Dr Norah Armstrong was born in Hamilton
and was a daughter of the Very Rev Matthew
and Mrs Gladys Stewart. Afterher schooling
at Hamilton Academy she went to Glasgow
University where she studied and qualified in
medicine. Nature was always part of her life
and at one time she studied snails. It was,
however, not until she came to live at
Lochmaben that she was able to devote more
of her time to watching and studying nature,
especially birds. She was also a very good
botanist. In her capacity as a doctor she will
be remembered for her work in the Well
Woman Clinics where, for many years, she
worked at Lockerbie, Moffat, Kelloholm and
Dumfries.

Norah became a member of the Dumfries
branch of the SOC soon after coming to
Dumfriesshire. The interests of this branch
became very much part of her life. She was
secretary for many years and became its first
lady chairman. Norah was again acting as
secretary when her illness made it too difficult
for her to carry on and, very reluctantly, she
resigned the post. The welfare of speakers
was always paramount in her mind and many
of them enjoyed the hospitality of Deil’s Dike.
She was also very generous in her support of
money raising projects for the branch and
also supported other good causes. Among
the things she gave to the Dumfries branch
were the loudspeakers for the public address
system. Applegarth Wildlife Sanctuary has

also benefited greatly from her generosity
and knowledge.

She was an enthusiastic conservationist and
created a small wetland area, complete with
a permanent hide, at her home. Recently a
larger area was developed which includes a
pond, shrubs and several acres of woodland.
Norah was also a first class photographer
and was, for many years, a member of the
Nature Photographers Portfolio. Colour
transparency was her favourite medium which
allowed her to produce her own colour prints
in the darkroom. Some of her work has been
exhibited locally and at Leigh in Lancashire
but most of her prints were for the NPP
portfolio. Local clubs and societies often
benefited from her talks and slide shows on
medicine and nature. At the one day
conference at Galashiels in 1995 she took
candid shots of personalities for the SOC.
When it came to the main conference, she
declined to how her pictures, preferring to
relax and enjoy the company of her friends.

A well travelled person, the Highlands of
Scotland were her favourite with Norfolk taking
second place. It was while she was watching
a pair of Red-backed Shrikes that | saw her
medical skills come to the fore. A girl hada
terrible accident on her bicycle in front of her.
Norah suddenly changed from a carefree bird
watcher to a methodical, organised doctor.
She was, for a time, a member of the North
Solway Ringing Group and took part in the
ringing of many waders. Lapwing chicks
were among her favourites and there were
also some hair raising experiences on the
Scar Rocks which are probably best forgotten!

| have been extremely fortunate knowing
such a wonderful person and | am sure many
others feel the same. Norah was nursed at
home with great care and devotion by her

187

Obituaries

like to visit Deil’s Dike to see the sanctuary,

they would be most welcome to do so.

ltwas

Bobby Smith

husband and her daughter, who is also a
one of Norah's wishes that, if anyone would

doctor, until she died on 12 January .

1996

£
£
Y
>
2
feo)
Oo
aa)
SSA ao

oO)
(oy
2
7)
E
<x
A coms
fe
(e)
<

188 Obituaries

lan Munro

1910 - 1996

The death of lan (JHB) Munro on 15th March
after a mercifully fairly short spell in hospital
must have brought to his many friends a
conflict of great sadness - and deep sympathy
for Bep - and memories of very happy times
inhis company. He hada wealth of knowledge
of birds, a great fund of reminiscences of
army service during the War and a seemingly
inexhaustible collection of funny stories told
in an intriguingly “dead-pan” way so that a
holiday with him was enormously enjoyable.
(Having been with him for a week on the Isle
of May for 17 years and ona Zimbabwe safari
| can remember hearing the same story only
once - and how | wish that | could hear them
all again!)

It was in the early 1930s that lan, with a
coterie of friends from the Edinburgh Academy
and elsewhere, formed the Midlothian
Ornithological Club which soon grew into the
SOC. Life must have been fairly hectic in
those early days for a keenly gifted athlete
and budding CA yet he found time somehow
for two projects in which he was to be
profoundly involved. The Isle of May had
been visited often between 1907 and 1933 by
the almost legendary Misses Baxter and
Rintoul from Fife but in 1934 the MOC started
up the embryonic Observatory. They built the
first of the Heligoland traps enabling birds to
be caught - and ringed - instead of being shot
for identification, binoculars and cameras not
being what they are today. How thrilled lan
must have been to be, with Bep, in the party
which crossed in October 1994 to celebrate
the Observatory’s 60th birthday. lan’s entries

SB 18(3)

inthe Daily Log may nothave beenas hilarious
as those of the inimitable MFMM but they
were always full of insights. For many years
too he was Treasurer of the Observatory.

The second of lan’s major undertakings was
centred on the Pentland Hills which for him,
as with Robert Louis Stevenson, were “the
hills of home”. He must have scoured most of
the area covered by the Pentlands but part of
the research he set himself was to cross from
Nine Mile Burn to Bavelaw by the same 15
mile trail each week for a year between
February 1948 and January 1949 - whatever
the weather, 780 miles of it. He collected
records from every conceivable source -a
great deal of them from G L Sandeman - and
in 1988 was published his Birds of the Pentland
Hills. An amazingly scholarly book, no wonder
it took him so long, especially in view of the
fact that his working life with The Scotsman
can have been no sinecure under Lord
Thomson.

It was not until he was well into his forties that
lan married Bep, then on the nursing staff of
the Edinburgh Royal Infirmary. Fortunately
Bep too was keen on birds so there began a
wonderfully happy partnership.

lan played a round of not-too-serious golf
most weeks at Gullane. They both took part
in the Kirk and village life but it was clear that
they were just blissfully happy together. Their
home on the fringe of Colinton has a delightful
air to it and the garden was a place of peace
- after hard work - much favoured by birds.
And although birds continued to draw them
away for holidays to other parts of the country,
and to other counties - the Med, Africa, India
- it was surely to Colinton that they came
home to.

There must be many hundreds who, in one

189

Obituaries

1996

them; so, too, could they but express it, the
birds of Lothians and beyond have much to

be grateful for to him.

way or another, owe lan a great deal for the
fascination and joy that birds have brought

lan Balfour-Paul and John Arnott

John Arnott

lan Munro

190 Scottish Birds (1996) 18: 190-192

SB 18(3)

Correspondence

Density of breeding Crows in Argyll

The density of breeding Crows in Argyll quoted
in Dott (Scottish Birds 17:205-211) should be
2.56 pairs per km? (Hewson & Leitch 1982).
This is higher than the highest densities found
by Dott in urban and suburban south east
Scotland (1.75-2.50 pairs per km2) or on one
rural area in the Borders (0.50 pairs per km2),
and similar to Picozzi’s 2.3-3.0 pairs per km
on farmland and woods, excluding adjacent
moorland, which Crows used in north east
Scotland. The high density in Argyll was
probably due to a good food supply; Crows
foraged along the shoreline and on sheep
and lamb carcasses.

Using Dott’s method, 6 contiguous tetrads
within the main study area of 20.25 km@ of
moorland with coastal woodland and pasture
in Argyll had 0.5 to 1.5 pairs of Crows per km2
(one tetrad had none), compared with 0.25 to
0.50 pairs per km? forhill, moorand coniferous
plantations in south east Scotland.

R Hewson, Department of Zoology,
Aberdeen University, Aberdeen AB9 2TN

The figure of 2.56 pairs per km? from Hewson
& Leitch was taken as 2.6 for comparative
purpose in my paper. At some stage 2.6 was
changed to 26 pairs per km2. | apologise for
failing to notice this and thank Ray Hewson
for his observations.

Harry E M Dott, 70 Findhorn Place,
Edinburgh EHS 2NW

Red-backed Shrike breeding records
in Scotland

In addition to those during 1977-79 and in
1992 mentioned in the paper on the species
breeding in Perthshire in 1994 inthe December
1995 issue of Scottish Birds (SB 18: 118-119)
there was a successful breeding attempt on
Shetland in 1990 which produced one fledged
young bird (Shetland Bird Report 1990 p55).

Bob Proctor, 94 Reid Street, Bishopmill,
Elgin, Moray, 1V30 2HH.

Bill Paton

Carrion Crow

Yellow-legged Gulls in Scotland

Inthe paper “Yellow-legged Gulls in Scotland”
(Scottish Birds 18:96) R W Forrester refers to
a Fair Isle record “attributed to Dr Eagle
Clarke & Admiral Stenhouse”. The reference
should be to Surgeon Rear-Admiral John
Hutton Stenhouse, who worked with Dr Eagle
Clarke and visited Fair Isle many times
between 1920 and 1926. The mistake may
have originated with Baxter & Rintoul (1953)
in their book The Birds of Scotland which is
cited by R W Forrester.

lan J Stenhouse, c/o 31 Kinarvie Road,
Crookston, Glasgow G53 7EA.

1996

Correspondence 191

Cormorants on Handa

In the recent paper on Seabirds of Handa
(Stoneman & Willcox 1995, Seabirds of Handa
Island, Scottish Birds 18: 78-87) there is no
mention of Cormorant Phalacrocorax carbo.
Varying numbers of Cormorants breed on the
adjacent Badcall islands and there is
documentary evidence that they formerly bred
on Handa.

In a paper by Harvie-Brown (Extracts from a
Journal of a Nesting Tour in Sutherland in
1867, Zoologist 1867: 851-865) appears the
astonishing statement “Cormorants ...
abundant on Badcall islands and increasing
on Handa where they were introduced by Mr
Maclver, the Duke’s factor at Scourie”. This
seems incredible, and there is certainly no
mention of it in the Memoirs (Henderson G,
1905,Memoirs of a Highland Gentleman,
being the Reminiscences of Evander Maclver
of Scourie, Constable, Edinburgh). Maclver’s
main interest in Handa appears to have been
the acquisition of the grazing following the
removal of the tenants in 1848. Maclver did
not come to Scourie until 1845, yet in June
1849 John Wolley (Newton A, 1864-1907,
Ootheca Wolleyana ,Porter, London) shot a
Cormoranton the nest on Handa and took the
eggs “These were fresh but many of those
about were hatched”. This would appear to
make Harvie-Brown’s account all the more
incredible. Later, Harvie-Brown (Onthe birds
found breeding in Sutherlandshire Proc Nat
Hist Soc Glasgow 2 1869-75) says “not
numerous Handa, groups only of halfa dozen
on the highest parts of the cliffs”, and in 1884
(Harvie-Brown J A & Buckley T E, Appendix
to St John’s Tour in Sutherlandshire, second
edition, Edinburgh) “Rare on Handa, choosing
the highest ledges”. This is repeated in
Harvie-Brown & Buckley 1887 A Vertebrate
_ Fauna of Sutherland, Caithness and West

Cromarty David Douglas, Edinburgh). In
Harvie-Brown and MacPherson1904 (A
Fauna ofthe North-West Highlands and Skye
David Douglas, Edinburgh) “Colony still on
Handa, not increasing.”

| can find no later records of Cormorants
breeding on Handain references suchasCG
Connell (Bird Life on Handa in 1925,Scottish
Naturalist 1925: 143-147;) and Dickinson &
Harris (Handa Birds, July 1959,Fair Isle Bird
Observatory Bulletin 1960: 1-4). The
statement by Baxter and Rintoul (The Birds
of Scotland, Oliver and Boyd 1953) “Nests
quite plentifully ...on Handa” cannotbe correct.
There were none seen in 1954 when Dougal
Andrew and | stayed on Handa in June.
Cormorants have been recorded fishing
offshore from Handa in most months and in
July 1964 and June 1965 | received records
of adults with young on Handa but | was not
able to confirm that these had actually bred
on the island.

Two questions arise. When did Cormorants
cease to breed on Handa and why? To the
first the answer is anyone’s guess, sometime
between 1904 and 1925. To the second
“Decreasing and much persecuted” in 1875
provides a clue, but those on the Badcall
islands were equally persecuted and survived
in much greater numbers. Another reason
why Handa has not been successfully
colonised may be geological. Handa is
composed of Torridonian sandstone with
vertical cliffs. The horizontal bedding of the
rock provides narrow ledges with overhangs
and small caves more suitable for Shags.
However, Cormorants prefer open, sloping
sites as found on the rounded lumps of
Lewisian Gneiss which form the Badcall
islands.

lan D Pennie, Badcall,Scourie, Sutherland

192 Advice to Contributors

Advice to Contributors

Authors should bear in mind that only a small
proportion of the Scottish Birds readership are
scientists, and should aim to present their material
concisely, interestingly and clearly. Unfamiliar
technical terms and symbols should be avoided
wherever possible and, if deemed essential, should
be explained. Supporting statistics should be kept
to a minimum. All papers and Short Notes are
accepted on the understanding that they have not
been offered for publication elsewhere and that
they will be subject to editing. Papers will be
acknowledged on receipt and will be reviewed by at
least 2 members of the editorial panel and, in some
cases, by anindependentreferee. They will normally
be published in order of acceptance of fully revised
manuscripts. The editor will be happy to advise
authors on the preparation of papers.

Reference should be made to the most recent
issues of Scottish Birds for guidance on style of
presentation, use of capitals, form of references,
etc. Papers should be typed on one side of the
paper only, double spaced and with wide
margins; 2 copies are required and the author
should also retain one. We are happy to accept
papers on Applemac computer discs. We cannot
handle other formats as both the SOC computers

awe

Hooded Crow

and those at our printers are on the Apple system.
Please contact Sylvia Laing on 0131 556 6042 to
discuss this. Headings should not be underlined,
nor typed entirely in capitals. Scientific names in
italics should follow the first text reference to each
species. Names of birds should follow the official
Scottish list (Scottish Birds Vol 17: 146-159). Only
single quotation marks should be used throughout.
Numbers should be written as numerals except for
One and the start of sentences. Avoid hyphens
except where essential eg in bird names. Dates
5 August
WOOT es ae but on the 5th (if the name of the
month does not follow). Please note that papers
shorter than c700 words will be treated as Short
Notes, where all references should be incorporated
into the text, and not listed at the end, as in full
papers.

Tables, maps and diagrams should be designed to
fit either a single column or the full page width.
Tables should be self explanatory and headings
shouldbe kept as simple as possible, with footnotes
used to provide extra details where necessary.
Each table should be on a separate sheet. Maps
and diagrams should be in Indian ink and be
camera ready, but drawn so as to permit reduction
to half their original size.

NEOTROPICAL
BIRD CLUB

Neotropical bird club launched

A club has been launched to promote the study and conservation of
the birds of the Neotropics (South America, Central America and the
Caribbean). It is currently seeking founder members to help reach
the launch budget of £2000, which is required to get the club running
and to publish the two first issues of its intended journal ‘Continga’.
Founder members will be asked to pay a minimum of £25, and will
be formally acknowledged in the first issue of ‘Continga’. ‘Continga’
will provide a colourful and much needed forum for exchange of
information on the avifauna of this extremely rich and diverse area,
and will contain papers and features on the birds and their
conservation as well as news of recent observations and discoveries
(at present, new species are still being discovered at the rate of
more than two a year). It is hoped that in due course the club will be
able to provide direct funding and support for practical conservation
programmes.

For further details and membership forms,
please contact:

Rob Williams,

Publicity Officer,

Neotropical Bird Club,

c/o The Lodge,

Sandy,

Bedfordshire SG19 2DL

Scottish Birds

Volume 18 Part 3 June 1996

Contents

Main articles
Amendments to the Scottish List R W Forrester
Records of species recorded in Scotland on 5 or fewer occasions
P R Gordon & D Clugston
Changes since 1890 in the size of Pink-footed Geese wintering in Britain
H Boyd & J Berry
A census of the St Kilda gannetry in May 1994 S Murray & S Wanless
Ravens nesting on buildings in Orkney C J Booth 159
The hunting behaviour of Merlins in winter in Galloway R C Dickson
Numbers, distribution and habitat associations of Corn Buntings on the

Wee)

132

144
ere

165

Outer Hebrides and Tiree in 1995 PF Donald, P Hines, P Jackson, J Dustow,

| Hepburn, G F Albon & S Jervis

Short Notes
Hunting times of Hen Harriers in winter R C Dickson
Oystercatchers nesting in clear cut forestry plantations G Shaw
Oystercatcher apparently rearing Lapwings T Dougall
Dunnock feeding in Spruce canopy G Shaw

Obituaries
Dr Norah Armstrong RT Smith
lan Munro | Balfour-Paul & J Arnott

Correspondence
Density of breeding Crows in Argyll R Hewson
Density of breeding Crows in Argyll H Dott
Red-backed Shrike breeding records in Scotland R Proctor
Yellow-legged Gulls in Scotland | J Stenhouse
Cormorants on Handa | D Pennie

Advice to Contributors

Published by the Scottish Omithologist’s Club
21 Regent Terrace, Edinburgh EH7 5BT.© 1996

170

182
183
184
185

186
188

190
190
190
190
toa

192

QL
690
354343
BIRDS

Scottish Birds

The Journal of the Scottish Ornithologists’ Club

Editor: Dr. S. da Prato

Assisted by: Professor D. Jenkins, Dr J.B. Nelson and Dr. |. Bainbridge
Business Editor: The Secretary, S.O.C., 21 Regent Terrace, Edinburgh EH7 5BT
(tel. 0131-556 6042).

Scottish Birds, the official journal of the Scottish Ornithologists’ Club, publishes original material
relating to ornithology in Scotland. Papers and notes should be sent to The Editor, Scottish Birds, 21
Regent Terrace, Ecinburgh EH7 5BT.

Two issues of Scottish Birds are published each year, in June and in December. Scottish Birds is
issued free to members of the Scottish Ornithologists’ Club, who also receive the quarterly _
newsletter Scottish Bird News and the annual Scottish Bird Report. These are available to non- ~
members at a subscription rate (1992) of £30.

The Scottish Ornithologists’ Club was formed in 1936 to encourage all aspects of ornithology in
Scotland. It has local branches which meet in Aberdeen, Ayr, the Borders, Dumfries, Dundee,
Edinburgh, Glasgow, Inverness, New Galloway, Orkney, St Andrews, Stirling, Stranraer and Thurso,
each with its own programme of field meetings and winter lectures. The Waterston Library at the _
Club’s headquarters at 21 Regent Terrace, Edinburgh EH7 5BT is the most comprehensive ~
ornithological library in Scotland, and is available for reference during office hours (Monday to
Friday, 9.00 - 17.00 hrs). (But please phone beforehand). A comprehensive stock of Scottish local
bird reports is held at headquarters and may be purchased by mail order.

SOC annual membership subscription rates

Lite rattan oso. sten. falseece-cecdsees tation codew udccsesaats denonactemeveenace teeta tees oe ee eee £450.00

All subscriptions may be paid by Banker’s Order and Covenanted. Subscriptions paid this
way will greatly assist the Club.

Published by the Scottish Ornithologists’ Club, 21 Regent Terrace, Edinburgh EH7 5BT
Printed by Douglas Gauld & Co Limited, 22 Tannadice Street, Dundee DD3 7QH

Scottish Birds (1996) 18: 193-196

193

surveyed in one year. This paper brings together the most recent data on
Whimbrel in Shetland from several sources, including systematic bird
surveys listing breeding Whimbrel since 1983. Apparently the breeding
population of Whimbrel in Shetland has changed little since 1983-86 and is
fairly stable at 479 pairs which represents 98% of the UK breeding
population. However, due to the many variations in survey techniques and
the large number of different observers, conclusions are tentative. Because
surveys have involved several different techniques, a complete population
survey followed by periodica! monitoring is recommended.

Introduction

This paper brings together the most recent
data on Whimbrel Numenius phaeopus in
Shetland from a variety of sources. Most UK
Whimbrel breed in Shetland (Richardson
1990) and a number of Sites of Special
Scientific Interest (SSSIs) have been notified
by the Nature Conservancy Council, now
Scottish Natural Heritage, primarily or in part
to protect their breeding habitat. The Shetland
population of Whimbrel is small at 479 pairs,
0.1% of the Holarctic population (Cramp &
Simmons 1983), but, within the UK, the
Shetland population is important. However,
the entire breeding population of Whimbrel in
Shetland has never been completely surveyed

in one year.

_ Methods

_ This paper collates information on Whimbrel
collected during wader and skua surveys in

;

|

Shetland between 1983 and 1994. For 1989-
94 and 1983-86 data from 7 and 5 sources
were used. Sources not mentioned specifically

in the text are listed in Appendix 1. For each
of 2 time periods (1983-86 and 1989-94)
when duplicate surveys were done, the most
recent data are used. Richardson (1990)
incorporates information for 1982-86 from
several sources and we compare them with
later surveys (1989-94).

Both Richardson (1990) and Arnott etal (1992)
state that the optimum time for surveying
Whimbrelis 1-20 June after which the majority
of chicks have hatched and some of the
Whimbrel have left their nesting sites. All the
surveys on which this report is based were
carried out during this part of the breeding
season. However, it is now known that peak
laying dates for Whimbrel are around 10-15
June and that some Whimbrel begin to leave
their nesting areas after this period. It is now
recommended that Whimbrel surveys should
finish by 15 June (M Grant pers comm).

Count methods differed considerably between
surveys. RSPB standard upland survey
methods involved observers walking transects
250m apart and mapping the apparent centre

194 CP Dore et al

SB 18(4)

of territory of breeding pairs recorded
(Peacock et a/ 1985). NCC standard upland
survey techniques used transects 200m apart
(Rothwell et a/ 1988). RSPB skua surveys,
which also recorded Whimbrel, used different
distances between transects depending on
the topography of the land and the density of
nesting skuas. Attimes, these transects were
up to 500m apart (Bird et a/ 1986). The 1992-
94 SNH survey involved dividing sites into
0.5km2 (500 x 500m), walking for 25 minutes
in each square to scan the area and recording
the presence and behaviour of birds (Arnott et
al 1992, Bates et a/ 1994).

Difficulties also arise in the interpretation of
survey results. Recent surveys (Arnott et a/
1992, Bates et a/ 1994) have relied on the
Brown and Shepherd (1993) methodology,
which uses a Strict definition of breeding
behaviour, and may not be appropriate for
Whimbrel. It is considered that this has led to
lower numbers of breeding Whimbrel being
recorded.

Results

Table 1 shows the numbers of breeding -
Whimbrel in different areas of Shetland in
1989-94 and 1983-86. The total population

foundin 1989-94 was 479 pairs, approximately
98% of the estimated UK breeding population.

This was 21 pairs fewer than in 1983-86. |
Table 2 shows the numbers for similar areas |
found by Richardson (1990). Richardson’s

total is 8 fewer than that of 1989-94 and 29 -
fewer than that of 1983-86. |

——— ae

Richardson (1990) found the north isles of |
Unst and Fetlar to be the most important
areas, together holding 43% ofthe population. |
The data for 1983-86 and 1989-94 presented |
here give similar figures of 36% and 37% —
respectively for these 2islands. Richardson’s |
(1990) figure for Yell was 8%, similar to our —
data for 1983-86 and 1989-94. The west |
Mainland accounted for 18% of the population

in Richardson (1990); we found 19% in 1989- ©
94 and 22% in 1983-86. The next most |

Table 1 Estimates of the number of pairs of breeding Whimbrel in Shetland
in 1982-86 by Richardson (1990) and in 1983-86 and 1989-94

Area 1982-1986 %
Unst 110-115 25
Yell 31-37 8
Fetlar 73-79 18
Whalsay 18 a
North Mainland 41-48 10
Central Mainland 55-58 12
West Mainland 65-87 18
South Mainland 15-23 4
Bressay

Burra

Other Islands 5-6 1
Total 413-471

* estimate

1983-1986 % 1989-1994 %

102 20 106 228
39 8 38 8
81 16 70 15
18 4 19 4
61 12 57 129
65 13 59 12)
109 22 89 19
13 3 26 5 il
5* 1 51 |
7 1 10 2 aI
]
500 479 |

1996

important areas are the north and central
Mainland, where Richardson (1990) found
22%. In 1983-86, we found 25% and, in 1989-
94,24%. However, due tothe many variations
intechniques andthe large number of different
observers involved, this population estimate
requires further clarification, although it is
noteworthy how similar the proportions are
from survey to survey.

Discussion

Table 1 shows that there appears to have
been little change in the numbers of Whimbrel
breeding in Shetland since 1983 with 500 in
1983-86 and 479 in 1989-94. Table 1 also
shows that similar proportions of the total
population were recorded in different areas of
Shetland during the 3 surveys. Table 1 shows
that a small numerical decrease appears to
have occurred on Fetlar. However, no change
has occurred in the numbers of Whimbrel
breeding in areas outside the statutory bird
sanctuary over the last 10 years, where any
change would be expected to be most
pronounced (RSPB unpublished wardens’
reports). Inside the sanctuary Whimbrel breed
at high density and are difficult to count. The
apparent change may be due largely to
differences between surveyors. Itis believed
that the apparent differences between the
1983-1986 data and the total given by
Richardson (1990 Table 2) are due to his
‘omitting some of the 1986 data (M G
Richardson, pers comm).
j
| The total figure for Whimbrel of 479 breeding
pairs in 1989-94 is thought to be the most
reliable estimate of the breeding population
of Shetland Whimbrel currently available.
Because the Shetland breeding population of
“Whimbrel has never been surveyed in one
year and previous surveys have used different
“methods, it is recommended that a more
“accurate method of surveying breeding

|
t

Whimbrel breeding in Shetland 1989-95 195

Whimbrel should be developed and that a
complete population survey of Shetland and
other known breeding areas in Scotland be
carried out. Population monitoring should
then be introduced to determine long term
changes in the status of this important species.

Acknowledgements

The surveys quoted in this report were funded
by RSPB, Nature Conservancy Council and
Scottisn Natural Heritage. M O’Brien, M
Grant and G Mudge commented on an earlier
draft. We thank all those who supplied
additional data and advice, Richard Stillman,
lain Robertson and Mike Pennington, and all
those who surveyed uplands and skuas in
Shetland.

References

Arnott D A, Shotton J & Dowse A 1992. A survey
of moorland breeding birds in selected sites
throughout mainland Shetland. Unpublished report
to SNH.

Bates MA, Shepherd K B, Whitfield P & Arnott DA
1994. A breeding wader and upland bird survey of
selected sites in Shetland Unpublished report to
SNH.

Bird DR, PriorA, EllisP M& Ewins PJ 1986. RSPB
skua Surveys in Shetland, 1986. Unpublished report
to RSPB.

Brown A F & Shepherd K B 1993. A method for
censusing upland breeding waders. Bird Study
40:189-195.

Peacock M A, Beveridge F M &Campbell L H 1985.
Moorland Survey in Shetland, 1985. Unpublished
report to RSPB.

Cramp S & Simmons K EL 1983. The Birds of the
Western Palearctic Vol 3:484-496. Oxford
University Press.

Richardson M G 1990. The distribution and status
of Whimbrel Numenius phaeopus in Shetland and
Britain. Bird Study 37:61-68.

Rothwell A, Stroud D A & Shepherd K 1988.
Shetland moorland bird surveys 1986. Unpublished
report to NCC.

196 CP Dore et al

SB 18 (4)

C P Dore and P M Ellis, RSPB, East House, Sumburgh Head Lighthouse,

Virkie, Shetland

E M Stuart, Scottish Natural Heritage, 2-4 Alexandra Buildings,

The Esplanade, Lerwick, Shetland

Revised manuscript accepted February 1996

Appendix 1

Whimbrel surveys since 1982 not referred to in the
text

Fisher P & Richardson M G. 1983 Whimbrel and
Wader Distributions, Shetland, 1983. Unpublished
report to NCC.

Hawkins |. 1990 Fetlar 1990 report. Unpublished
report to RSPB.

Herfst M & Richardson M G. 1982 Whimbrel and
wader distributions on Unst, Shetland, 1982.
Unpublished report to NCC.

Smith S 1992. Black Park annual report, 1992.
Unpublished report to RSPB.

Wynde R M & Richardson M G 1985. Moorland
and coastal bird surveys, 1985. Unpublished
report to NCC.

Other sources of data not in report

1994. RSPB/SNH survey of a small number of
sites throughout Shetland.

1992. RSPB Shetland wide skua survey during
which any obvious Whimbrel territories were noted.
1989. Survey of Unst specifically for Whimbrel by
M G Pennington.

Wm Paton |

Scottish Birds (1996) 18: 197-204

197

Breeding waders in the Cairngorms Straths ESA in 1995

N PICOZZI, DC CATT & R P CUMMINS

Baseline counts of Lapwings, Oystercatchers, Redshanks and Curlews
were made in 1995 on 25 farms in the recently designated Cairngorms
Straths ESA. Preference indices for crop types showed that Lapwings
preferred bare ground and rushy pastures, Oystercatchers bare ground,
Redshanks old grasslands and rough grazings, and Curlews rough
grazings. ESA requirements which have potential benefit for these species
are described. Numbers of Redshanks were low, but those of Lapwings
and Oystercatchers were comparable to earlier counts on ‘key’ sites on

Scottish farmland.
Introduction

Modern intensive farming methods have been
identified by the BTO and the RSPB as a
major contribution to the decline of waders
and songbirds. Waders such as the Lapwing
Vanellus vanellus and Redshank Tringa
totanus have been particularly hard hit by the
drainage of wet areas, both large and small
(Galbraith et al 1984). Indeed the Lapwing,
once a familiar sight on British farmland, is
now a Red Data Book candidate, as are the
Redshank and Curlew Numenius arquata
(Batten et a/ 1990). The Oystercatcher
Haematopus ostralegus is much less
dependent on wet areas, and surveys have
shown anincrease in both range andnumbers
on farmland in recent years (Gibbons et al
1993).

Toencourage less intensive farming practices,
10 areas in Scotland have been designated
as Environmentally Sensitive Areas (ESA) by
the Scottish Office. The scheme has patential
benefit for wading birds. Under the scheme,
farmers within the boundaries of an ESA
have the option of registering for payments
which help compensate for loss in revenue
which may result from less intensive practices.

There are 2 tiers in the scheme. Tier 1 covers
all the farm and is compulsory. Tier 2, which
is optional, provides higher payments for
positive enhancement measures. The
standard requirement of Tier 1 is to avoid
damaging rough grazings, unimproved
pastures, reverted improved land, wetlands,
water margins, native woodland, amenity
woodland and scrub by land clearance,
ploughing, new drainage, modifying existing
drains, levelling, reseeding or cultivating (CS/
ESA 1 1993). Tier2 payments are additional
and vary depending on the measures to be
undertaken, some of which are mandatory,
others optional. They include encouraging
regeneration of native woodland by fencing,
conserving wetlands and herb rich
meadows, and leaving unsprayed headlands
around arable crops. An approved
Management Plan clearly defines for 10
years ahead just what measures will be
carried out each year (reviewable after 5
years).

In 1995, we visited 25 farms in the recently
designated (1993) Cairngorms Straths ESA
to count Lapwings, Oystercatchers,
Redshanks and Curlews. The aim was to
establish a baseline, before the prescriptions

198 N Picozzi et al

could take effect, against which future counts
can be compared.

Methods
Selection of farms

Five farms for which management plans had
been prepared were chosen at random in
each of the 5 main regions (upper Deeside,
upper Donside, Strath Avon, upper and lower
Speyside) to ensure that the whole ESA was
represented.

Counts of waders

Lapwings, Oystercatchers, Redshanks and
Curlews were counted on each farm on as
much of the inbye land, ie land which is, or
has been, cultivated and which is usually
enclosed by a fence or a dyke, as it was
possible for 2 observers to assess in 3-4
hours (at least 44 ha). In particular, any
wetland site that was designated under Tier
2 was included, as were the fields adjacent to
such sites. Hill grazings, usually dominated
by heather Calluna vulgaris, were not visited.
Three counts were made on the 25 farms.
The first covered the period when the birds
were establishing territories (11 April - 3
May), the second covered the main nesting
period (9 - 23 May) and the third was to
estimate breeding success (12 - 29 June).
To complete these counts, it was necessary
to visit 2 or 3 farms ina day. We therefore
chose to work on those species which could
be counted reliably at any time of day; we did
not include Snipe Gallinago gallinago
because they cannot be counted with any
degree of accuracy except in the early
morning or late evening. Their requirements
for the wettest ground are similar to those of
Redshank, for which we felt reliable counts
could be made as they usually flushed and

SB 18 (4)

called when an observer was still some
distance away.

The methodology of O’Brien and Smith (1992)
was adopted, to be consistent with earlier

studies. Where possible, counts were made —

from a vantage point, of displaying birds on
the first visit, and of nesting birds and birds

with chicks on the later visits. As it was not |

always possible to find a suitable vantage
point for counts and some birds could have
been missed, transects at 100 m intervals

were then walked through fields, whenever

conditions allowed, to flush birds. The total
number of Lapwings seen was halved to give
an estimate of the number of pairs present
(see O’Brien and Smith 1992).
Oystercatchers, Curlews and Redshanks
were usually seen in pairs but if seen singly
were assumed to represent pairs. The data
presented in this paper refer only to breeding
birds. Flocks of passage or non breeding
birds are excluded because of their itinerant
nature.

Definitions of land use
For the purposes of this study, the following

definitions for land class (based on those in
CS/ESA 1 1993) were adopted:

1. Bare ground landundertillageinspringfor °

cereals, grass, root crops or oil seed or forage
rape. This land was either newly ploughed or
had been used for root crops or rape the
previous year.

2. Hay/silage fields which were sometimes
grazed in the early spring but then fertilised
with dung, slurry or inorganic fertiliser for the
production of hay or silage.

3. Short term grass and leys fields used
primarily for grazing and lambing, which were

1996

in regular short term rotation.

4. Long term grass and permanent pasture
grazed grassland with a variety of low herbs,
neither in regular cultivation nor partof along
term rotation. This was usually on ground
that would have been difficult to cultivate.

5. Rough grazings land containing semi
natural vegetation including rough grassland,
used, or suitable for use, as grazing. Land,
often along stream sides, with more than
50% of rushes, usually in rank clumps, was
included here.

6. Rushy pasture |and, previously improved
by agricultural management, which now had
plants indicative of unimproved pasture. In
general this meant the presence of invasive
clumps of rushes which occupied less than
50% of a field. These fields were normally
still being grazed by sheep and occasionally
by cattle.

All the farm maps were digitized using the
ARC/INFO Geographical Information System
(1989). From this, the areas of each field

were readily obtained and the proportions of
each crop type calculated.

Habitat preference

_ Preference for a particular crop type by each
bird species within each farm was calculated
according to Duncan’s (1983) preference
index (Pi); values greater than 0.3 indicate
preference, and less than 0.3 avoidance:

Proportion (%) of birds using the crop type
Pi =logi9 ——_—_-————_ _ + 1
Proportion (%) of land available

Two sets of indices were calculated from the
first 2 counts for each species. The first
method (A) considered that, if a species was

Breeding waders in the Cairngorms Straths ESA in 1995 199

present on a farm, but absent from one or
more particular crops, the preference index
for those crops should be included in the
calculations as O (/e Pi = logy9 1). The
reasoning is that if, say, Lapwings were
present on a farm, they could in theory have
selected any of the crop types there; this
method takes into account both the crop
types that they did select, and those they did
not. A disadvantage of the method is that
there may have been more crop types than
birds, in which case not all crop types would
have an equal chance of selection. The
second method (B) overcomes this objection
by treating absence as a missing data value
and so looks at the relative preferences for
the habitats which the birds actually used.

Statistical analyses

The indices were not distributed normally
and so differences from the expected value
for crop preference by waders were tested
for significance by non parametric Wilcoxon
tests (SAS (1990) NPARIWAY
PROCEDURE). The preference indices for
each crop were first grouped by region. The
differences between regions were not
Statistically significant and so data for all
farms in the ESA were treated as a whole.

Results
Habitat preference

Results for the 4 species are summarized in
Figure 1, in which the mean statistically
expected rank order of preference for a crop
has been subtracted from the mean of the
observed ranked orders of preference. A
positive result indicates preference, a
negative result indicates that the proportion
of all the pairs on the farm that were on that
crop was less than would have been expected

200 N Picozzi et al

from the available amount of that crop type ie
it was apparently avoided. The results of the
statistical tests, indicated in the following text
by superscript, are given in the Appendix.

Lapwing Selection among crop types showed
similar patterns using methods A orB. Overall
differences in the selection of crop types
were not significant using method A’, but
were using method B2. Bare ground, rushy
pastures and rough grazings, were favoured
by nesting birds, but leys were avoided.

Fig 1 Selection for crops by Lapwings, Oystercatchers, Redshanks and Curlews on the first (solid)

SB 18 (4)

Oystercatcher Differences for selection
between crop types were significant using
method A? and method B4. The clearest
preference was for bare ground. Short term
grassland and leys were mainly avoided by
nesting birds.

Redshank There were too few birds (16 |

pairs on 8 fields on the first count, 13 on 6
fields on the second) for results to reach
significance. However, the results were

generally consistent. The main preference |

was for old grassland and rushy pastures.
Leys and hay/silage fields were avoided.

and second (hatched) counts using 2 methods (see text). A positive result indicates slection, a

negative result avoidance. Crop 1 = bare, 2=hay/silage, 3=leys, 4= old pasture, 5=rough grazing

6 = rushy pastures. Number of farms in the sample shown on the X axis

LAPWING (a)

1O; tO Se

CURLEW (a)

S Sy V2 de ve ve

Difference between mean observed and expected ranked perference indices for each crop

20

LAPWING (b)

REDSHANK (b)

Go, it

- ee

?
}

i 4

i

1996

Curlew Both methods showed a significant,
or near significant, preference among crops
on the first and second counts using method
A° or method B®. Method A indicated a
preference for rough grazings.

Breeding success and density of pairs

An extensive survey based on just 3 visits to
a farm over a 10 week period gave an
imprecise estimate of breeding success since
itwas not possible to assess the performance
of known pairs. The fields in which pairs were
seen with young may not have been those in
which the birds nested. Lapwings for example
often take their broods into fields with sheep,
cattle or ponies after 1-2 weeks, so the number
of broods in such fields does not necessarily
relate to the number present there in spring
as nesters. The data for breeding
performance given here are therefore only a
guide and based on the maximum number of
pairs of each species seen on each farm in
spring on the first or second visit, whichever
was the greater, and the number of broods,
or adults giving distraction displays, seen on
the third visit (Table 1).

Breeding waders in the Cairngorms Straths ESA in 1995

201

Discussion

Habitat preferences and the implications
of the ESA prescriptions for wading birds

The 2 methods of treating the preference
indices gave broadly similar results. However,
method B (based only on those crops that
birds occupied) showed clearer apparent
preferences than method A. It is possible
that differences between the 2 methods were
due partly to the coarse habitat definitions
and partly to the choice of a field as a unit for
describing habitat. Forexample, the presence
of wet patches was not taken into account
and this may have affected the choice of
field. Nevertheless, the main findings are in
broad agreement with previous studies of
waders on Scottish agricultural land (Galbraith
and Furness 1983; Galbraith et a/ 1984;
O’Brien 1995). These authors showed that
the greatest densities of most wading species
were on poorly drained rough grazings rather
than improved grazings or cereal fields.
Barrett and Barrett (1984) also showed that
unimproved pasture in east Sutherland held

Table 1 Number of pairs of Lapwings, Oystercatchers, Redshanks and Curlews seen
on 25 farms totalling 1710 ha in the Cairngorm Straths ESA in 1995 and the
proportion (%) of these pairs with well grown or fledged young in late May and June.
Comparative densities from an earlier study on Scottish agricultural land are shown

Species Pairsin Pairs with
spring young (%)
Lapwing 244 65
Oystercatcher 140 60
Redshank 16 82
Curlew 40 43

Mean Density at Density at
density/km2 __ key sites 2 random sites?
14 11 2
8 5 2
1 3 1
2 4 0

aAverage density/km? at ‘key’ wader sites and © on farmland sites selected at random on

Scottish farmland in 1992 (O’Brien 1995)

202 N Picozzi et al

SB 18 (4)

greater numbers of breeding waders than
other habitats. All land on the farms we
visited in the Cairngorms Straths ESA will be
covered by Tier 1 prescriptions, a standard
requirement of which is to avoid damaging
grasslands by new drainage or modifying old
drains (see Introduction). This requirement
alone is almost certain to safeguard, and
perhaps increase, the amount and suitability
of land for waders.

Reverted improved land (rushy pasture), rough
grazings and bare ground were the preferred
habitats of Lapwings in this survey. Short-
term grassland and leys were the least
preferred options for Lapwings as nest sites,
though leys with livestock were important
feeding areas for broods. The Tier 1
prescriptions are likely to be beneficial to
Lapwings as they will favour the retention of
damp pastures. Nests on bare ground were
often destroyed by farming operations, mainly
because they were not easily seen.

Oystercatchers showed a clear preference
for bare ground as nest sites. They could
benefit from the Tier 2 arable option, under
which cultivation is restricted to the period
between 28 February and 15 May, except for
root crops, where nests must be marked and
avoided during operations. We found that
farmers marked Oystercatcher nests and
avoided them wherever possible, even when
not entered for this option. Their nests were
much more conspicuous than those of
Lapwings.

The numbers of Redshanks were very low
with only 16 pairs found on 1710ha. Although
the sample was small, a preference for rushy
and older pastures was indicated. The
importance of wetlands, which can be subject
to the enhanced payments under Tier 2, is
more critical for this scarce species than for
the other 3-considered here.

Both sets of results for Curlews indicated a —
preference for the rough grazings. Leys and —

silage fields were used, butless than expected
from the area available.

The greatest —

proportion of records were from rough ©

grazings (11 of 24 observations of Curlews

on the first count, 9 of 23 on the second —

count). The preference indicated for bare
ground in Figure 1 using method B may be

misleading since it was based on a small ~

number of birds (1 out of 24 records on the
first count and 3 out of 23 on the second
count). The Tier 1 prescription to avoid
damaging rough grazings by poaching,
feeding practices, overgrazing and tracking
by vehicle should benefit Curlew and other
birds nesting there.

Breeding success

It would be unwise to read too much into the

results for breeding success and no other —
broad survey of this nature has attempted to —

give such data. Lapwings present particular
problems as they may lay up to 4 replacement

A AR sty

— ——

clutches following eggloss, sothatthereisa_ ,
wide range of ages in June, with the first | _

broods already having fledged and present

in flocks. The data given here suggest a —
reasonable breeding success for Lapwings. —
Galbraith (1988) calculated that a pairshould —

produce 0.8 young to sustain the population. _
Atleast 65% of the pairs inthe ESA produced | °

one or more young. The minimum
requirement therefore was probably met.

There are no comparable data for ~

Oystercatchers, Redshanks or Curlews.
Densities compared with other studies

The most recent survey of waders on Scottish

agricultural land (O’Brien 1995) countedbirds _
on 236 1 km sites selected at random, and —
171 other sites considered to be of ‘key’ —
importance for waders (only one of these, a —

1996

farm in Donside, was selected in the current
survey). Average densities per 1 km? for
Lapwings and Oystercatchers on the key
sites were slightly less than those on the
study farms (Table 1), although the farms
were selected at random. Our baseline counts
were the first part of a 10-year programme to
monitor the effects of the scheme on wader
populations. They suggest that, overall,
Lapwing and Oystercatcher populations on
the group of farms in this study were
comparable with the best recognised sites
on Scottish farmland. This is encouraging,
as the ESA designation and participation in
the scheme were too recent to have had any
effect. However, the densities of Redshanks
and Curlews on key sites were greater than
those found here, and further emphasise the
cause forconcern over Redshanks expressed
by Galbraith et al (1984). Curlews were more
likely to be present on the hill grazings which
were not part of our survey.

Acknowledgements

Nicola Penford and Alison Espie of FFWAG
greatly assisted us with their advice in setting
up the project. We are indebted to the
farmers for access to their land. Robert
Moss’s comments on the text and advice on
the statistical treatment of the data are
gratefully acknowledged. This survey, which
formed a small part of a wide ranging survey
into many aspects of the ESAs in Scotland,
was funded by SOAEFD.

Breeding waders in the Cairngorms Straths ESA in 1995

203

References

ARC/INFO Geographical Information System. 1989.
Environmental Systems Research Institute Inc.
USA.

Barrett J & BarrettC 1984. Aspects of censussing
breeding Lapwings. Wader Study Group Bulletin
42:45-47.

Batten L.A, Bibby J, Clement P, Elliott G D & Porter
RF 1990. Red Data Birds in Britain. Poyser,
London.

CS/ESA 1993. Cairngorms Straths Environmentally
Sensitive Area - Explanatory Leaflet for Farmers.
The Scottish Office, Agriculture and Fisheries
Department, Edinburgh.

Duncan P 1983. Determinants and the use of
habitat by horses in a Mediterranean wetland.
Journal of Animal Ecology 52:93-109.

Galbraith H 1988. Effects of agriculture on the
breeding ecology of Lapwings Vanellus vanellus.
Journal of Applied Ecology 25:487-503.
Galbraith H and Furness RW_ 1983. Breeding
waders on agricultural land. Scottish Birds 12:148-
153:

Galbraith H, Furness R W & Fuller R J 1984.
Habitats and distribution of waders breeding on
Scottish agricultural land. Scottish Birds 13:98-
107.

Gibbons D W, ReidJB & Chapman RA 1993. The
New Atlas of Birds in Britain and Ireland: 1988-
1991. Poyser, London.

O’Brien MG 1992. 1992/93 Survey of breeding
waders on Scottish agricultural land. Scottish Bird
News 28:1-2. Scottish Ornithologists’ Club,
Edinburgh.

O’Brien MG 1995. Survey of breeding waders on
Scottish Lowlands. Interim unpublished Report to
RSPB. RSPB, Edinburgh.

O’Brien MG & Smith K W 1992. Changes in the
status of waders breeding on wet lowland grasslands
in England and Wales between 1982 and 1989.
Bird Study 39:165-176.

Picozzi N & CattD C 1988. Waders of agricultural
land. Research report 2. Institute of Terrestrial
Ecology, Banchory.

SAS Institute Inc. 1990. SAS Users Guide, Version
6, 4th ed. NC:SAS Institute Inc, Cary.

204 N Picozzi et al SB 18 (4) ©

N Picozzi, D C Catt and R P Cummins,
Institute of Terrestrial Ecology, Banchory Research Station,
Hill of Brathens, Glassel, Banchory, Kincardineshire AB314BY

Revised manuscript accepted March 1996

Appendix Results of statistical tests; the reference indicates the superscript
figures given in the text. For details of methods A and B, see text

Reference Species Analytical Result
method
Count 1 Count 2
1 Lapwing A X5*=6.22 p=0.28 X5%=640 p=0.27
2 Lapwing B X5° = 13.72 p=0.02 X52 = 18.42 p=0.006

3 Oystercatcher A Xs52=13.29 p=0.02 Xs5%= 11.20 p=0.047
4 Oystercatcher B X5*=17.85 p=0.003 X5° = 16.25 p=0.006
5 Curlew A —- X52 = 23.05 p=0.001 X52 = 14.34 p=0.014

6 Curlew B X5° = 13.75 p=0.017 X52 =9.78 p=0.082

-

arab
:

oe

Lapwing Barry Larking

Scottish Birds (1996) 18: 205-213

205

Human induced increases of Carrion Crows and gulls on

Cairngorms plateaux

ADAM WATSON

Formerly, Carrion Crows were rare and gulls mostly scarce on plateaux in
the Cairngorms. Walkers increased on Cairn Gorm in the late 1950s
following easier access, and more Crows were seen, eating scraps. Since
1961, new roads and lifts led to many walkers on the nearby plateau. Crows
and gulls increased there, often eating scraps. Crows robbed many
Ptarmigan nests. Crows and gulls were scarce on parts of the plateau where
I saw nobody, and on plateaux with no public roads or lifts, where | saw far
fewer people than on Cairn Gorm plateau. Crows became scarcer since
1992, after more trapping on nearby low ground.

Introduction

Carrion Crows Corvus coroneand gulls Larus
haunt low land, usually avoiding alpine land
above 750 m. After a road was built to Cairn
Gorm ski area in 1960 and a chairlift in 1961,
many people went there and to the nearby
plateau (Morris et a/ 1974). More Crows and
gulls visited the ski area than nearby land with
few people, often eating food scraps (Watson
1979).

This paper records Crows and gulls on plateau
A and other alpine land in summer (1 May-10
October) 1943-88, until other bodies began
full time work. My hypotheses were that
numbers in different summers on A, on A’s
different parts, and on A and other plateaux,
would be unrelated to numbers of people in
these summers and areas. All 3 proved to be
rejected.

Study areas and methods
Area A is a 17.6 km plateau on Cairn Gorm

and Ben Macdui (Fig 1) the foremost UK site
for subarctic wildlife. Other plateaux in the

Cairngorms massif between Spey and Dee
(somein Fig 1, others andnumber of observer
visits in Watson 1991) had no public road or
lift. | also made notes on the Mounth from
Lochnagar to Cairnwell ski area.

Watson (1979,1991) described count
methods. Locations were noted to avoid
counting the same birds more than once a
day. | corrected for ties in rg tests. To save
space, P is < 0.001 in tests with no P value
stated, and ‘numbers’ means numbers seen
per Summer day per year.

Results
Numbers of people

Inthe late 1950s, walkers increased on Cairn
Gorm, following easier road access and other
developments (Watson 1984). By the late
1950s, numbers had risen on A too (Watson
1991; 1956-59 v. 1943-55 annual means.
n=4 & 13, Mann-Whitney U=8, P=0.05). After
the main developments began in 1960,
numbers rose successively each year bar
one, to a 1974 peak (Watson 1991).

206 Adam Watson

SB 18 (4)

to Inverness =c00
A
H
i
f .
Aviemore @ 2S BN
2 “Srom, 4 Glenmore Forest
Rd wooo, oses
30% @ %
ee PES, chairlift
ge Rothiemurchus
5 Estate Cairn Gorm

Calm Lochan —
Braerlach .

Lochan Buldhe

Ben Macdul a
Derry

Cairn Toul
Cairngorm

eeee Main road

= Downhill ski area

Mountain summit ry Cairn Gorm plateau

daa a’ Bhulrd

Balmoral é Aa oioerteen
\ e Geo
D eesid€..
Braemar C ee
Z Lochnagar

¢ .a01

.
s
2
a
.
2
.
.

%
chairlift 4

The Calmwell~ .”

Successive rises then ended, with 6 in 1974-
88 but 8 declines including a run of 4 in 1981-
85 after a subsidiary 1981 peak. However,
numbers stayed much higher than in the
1960s. The main change was that successive
year to year increases ended.

On other plateaux and other alpine land in the
Cairngorms massif, where there were no
public roads or lifts, people did increase,
reflecting the popularity of hill walking. The
increase on Awas far greater (Watson 1991),
associated with easy access by roads,
chairlifts, and paths.

Crows onA

In 1932-49, D Nethersole-Thompson (in /itt)
saw a Crow on less than one summer day in
10 on alpine land on the Spey side of the
Cairngorms, but saw Crows eat walkers’
scraps at Cairn Gorm and on A’s north part in
the late 1950s, and in 1959 take Ptarmigan

Lagopus mutus and Dotterel Charadrius
morinellus nests (Nethersole-Thompson &
Nethersole-Thompson 1961).

Before the developments, Table 1 shows no
year with Crows seen and 17 years without,
but afterwards 21 with and 6 without (Fisher
exact P). Annual mean numbers of Crows
and people in 1943-88 were related (n=46.
's=0.83). Watson (1981) noted that in 1971-
80 the percentage of summer days with Crow
sightings increased with year (rg=0.96,
P<0.01). The tendency to increase then
ended. Numbers in 1981-88 were lower than
in 1975-80 (n=8 & 6. Mann-Whitney U=4,
P=0.008, though higher than in 1961-74 (U=8,
P=0.01). The biggest totals seen in a day in
each year in the 1970s exceeded those in the
1960s or 1980s (Table 2).

Since 1970, | have seen Crows on most parts
of A south to Ben Macdui summit, but mainly
on north parts on and near Cairn Gorm. |

1996

Table 1 Mean number of Crows and gulls
seen per summer day per year and n of
observer day visits on Cairn Gorm plateau
before and after a public road and chairlift
were built at Cairn Gorm ski area in 1960-
61

Crows Gulls n
1943 0 0 1
1944 0 0 1
1945 0 0 1
1946 0 0 2
1947 0 3 7
1948 0 0.1 12
1949 0 0.2 6
1950 0 0.3 6
1951 0 0.1 8
1952 0 0.3 8
1953 0 0.3 4
1954 0 0.2 9
1955 0 0.1 7
1956 0 0.3 7
1957 0 0.4 5
1958 0 0.7 3
1959 0 0.3 $
1960 0 1.0 4

Cairn Lochan (0.7), Ben Macdui summit and
west of Cairn Gorm (0.6), andlower elsewhere.

After Lochan Buidhe hut was built in August
1967, people often stopped there (Watson
1991) and left scraps. The hut was removed
on 6-8 June 1975. The proportion of Crows
seen there in 1967-74 (28% of 143 sightings
on all of A) exceeded that in 1976-88 (5% of
322). As these would include birds seen
more than once, any statistical analysis
involving a P value would be invalid because
of inflated totals, but the point is clear without
it. As the part at the hut site covered only
0.16% of A, Crows obviously still favoured
that part after the hut’s removal. Even after,

Carrion Crows and gulls on Cairngorms plateaux

207
1961 0 2.0 +
1962 0 0.7 4
1963 0 1.0 +
1964 0 1a 3
1965 0 3.0 2
1966 0 2.0 2
1967 Les) 0.6 7
1968 0 0.8 6
1969 0.3 1.3 6
1970 P2249) 122 6
1971 1.3 1.0 12
1972 0.6 tat 18
1973 0.7 2.0 25
1974 5.3 0.9 16
1975 4.0 8.5 24
1976 25 3.4 17
HO 3.1 Ci 17
1978 3.7 3.1 11
1979 5.6 0.5 9
1980 2.6 0.3 9
1981 1.3 6.3 18
1982 hee het 14
1983 122 0.2 ih
1984 3 0.2 6
1985 1.4 0.1 7
1986 1.8 el 6
1987 4.0 O27 6
1988 1.0 0.7 6

people still stopped there, where 2 paths met
near an attractive lochan.

‘Crows on A favoured summits, huts, popular

camp sites, paths, human snow holes, and
grassland. At sunset they flew north to gather
at dusk at 600 m on the moor below Coire na
Ciste before flying to roost in Glenmore Forest,
and at dawn flew south. | saw none on A from
dusk to dawn. The biggest single flocks were
18 on several parts of A, 30 on subalpine land
(650-750 m) west of Coire Cas, and 35 on the
moor before roosting in December 1991.

After 1960 | saw Crows on A each month in
May-September, and, since 1970, each month

208 Adam Watson

SB 18 (4)

Table 2 The biggest totals of Crows and gulls seen in a day each year on Cairn
Gorm plateau

CROW
Median Range
1960s 0 0.8
1970s ks 6-22
1981-88 4 2-19

CROWS

1960s v 1970s
1970s v 1980s

Mann-Whitney

GULLS

1960s v 1970s
1970s v 1980s

GULL
Median Range
3 1-7
10 7-34
5 1-22

U=4, P<0.002
Y=127P<0:/05

U=5, P<0:002
U=14, P<0.05

1970s values exceeded 1960s and 1980s ones in these tests.
The smallest totals seen in a day were 0 each year for Crows and for gulls.

in October-April on days with little fresh snow.
All seen on A were full grown except for one
with down traces, begging from an adult (see
2 paragraphs below for another case). In
winters since the mid 1970s they were seen
daily around car parks and buildings on the ski
area, even in snowy periods. In 1989 a pair
raised 2 voung at 650 m in Coire na Ciste,
nesting on a pylon of a chairlift operated only
in winter. In 1995-96, afew single Crows gave
territorial calls beside car parks at Coire Cas
and Coire na Ciste.

Crows on A often ate craneflies, leather jackets
and moths on the ground, insects stranded on
snow, walkers’ scraps, and a few corpses of
birds, sheep and domestic Reindeer.
Sometimes | saw one fly slowly while looking
down, the usual behaviour when hunting for
eggs, and twice saw one Carry an egg.

| found Ptarmigan nests robbed by Crows on ©
all A’s parts north of Lochan Buidhe and ©
Cairn Gorm’s north slopes west to Creag an —
Leth-choin, and saw sucked eggs each year ~
since 1971 on these areas. Sucked eggs ©
were noted at up to 20 widely different places —
in a summer. Robbing north of Cairn Gorm —
was severe each year in 1975-82 (Watson
unpublished). | saw 3 failed attempts to take |
Ptarmigan chicks, and one chick taken. J
Porter (pers comm) saw a Crow kill 3 downy |
Ptarmigan chicks south of Cairn Gorm, hide |
them in a hole, and fly off, to return shortly |
with fledged young which begged for food
and were given the chicks. Ptarmigan on
Cairn Gorm bred worse since the |
developments than before, and worse than
on other massif plateaux where | saw no

1996

Crows (Watson 1981,1982). A Crow robbed
one of 10 Dotterel nests with fresh eggs onA
(Watson 1989).

Crows on other alpine land

Watson (1979) recorded Crows on the ski
area and other parts of Cairn Gorm’s north
slope. | noted sightings there each summerin
1979-88, on almost every visit. They often
frequented the Ptarmigan Restaurant and
were there on each summer evening after
people left, though not in late summer 1996
when the Restaurant was closed.

In1967,a Crow was at 900 mon Ben Macdui’s
south side. On Derry Cairngorm | saw one in
1967,1969 and 1972, and found a Ptarmigan
nest robbed by a Crow in each of these years,
but none since. Watson (1991) saw few
people on Derry Cairngorm, but it was near
popular walking routes, camp sites and huts
at Derry Lodge, where Crows lived at all
seasons in the 1960s and early 1970s. They
roosted in the wood in winter (eg 20 in 1966-
67). From 1971, gamekeepers greatly
increased their Crow killing as Mar Lodge
Estate became keener on grouse shooting.
On other plateaux and other alpine land in the
Cairngorms massif, such as Beinn a’ Bhuird,
| saw very few people (Watson 1991) and no
Crows.

At Cairnwell | saw no Crows each summer in
1946-61. | noted sightings each summer in
1962-78, after a chairlift was built in 1962 at
the ski area and people increased greatly
(Watson 1979). In 1967, | saw one on Glas
Maol, just east of the ski area’s east end. They
were commonest in 1971 on Cairnwell, with
up to 13 atatime, frequenting also the nearby
Meall Odhar, Carn Aosda and Carn a’
Gheoidh, and once one on Glas Maol. They
took many Ptarmigan eggs (Moss & Watson

Carrion Crows and gulls on Cairngorms plateaux

209

1984) and some of Red Grouse Lagopus
lagopus scoticus. After the estate learned
this, gamekeepers increased their Crow killing
from early 1972.1 saw only 1-2 each summer
in 1972-78, and none since 1979, apart from
one in 1986 which took a grouse egg and one
in 1989. None was seen on alpine land in
winter.

In May-June 1963 and 1964, the gamekeeper
and | saw a few Crows up to 1100 m on
Lochnagar, a popular hill made more
accessible by bulldozed tracks. We found 3
Ptarmigan nests robbed by Crows, and 20
more sucked eggs on the ground.
Gamekeepers then increased their Crow
killing and | saw none on annual visits after
1967.

Ravens on alpine land

Ravens Corvus corax are natural on
mountains abroad, but none breeds in the
Cairngorms massif, where carrion on alpine
land is scarce (Watson 1966). D Nethersole-
Thompson (in litt) ‘practically never saw a
Raven on alpine land on the Spey side of the
Cairngorms in 1933-64. Each year in 1978-
88, however | saw 1-3 at a time on A in July
and August, and watched them feed on dead
sheep and domestic Reindeer. |Saw none on
other Cairngorms plateaux, where sheep and
reindeer were scarce or absent.

An occasional Raven was seen on the
Cairnwell hills and Callater hills in a few
summers in 1954-86. On Glas Maol and west
to the Cairnwell | saw a pair each summer in
1990-94, and 4-7 in 1995-96. Sheep and Red
Deer Cervus elaphus summered at high
density, corpses of both occurred each
summer, and 5 Ravens were at a dead sheep
in 1995. So, Ravens on alpine land inthe area
in Fig | were associated with unnaturally
many dead sheep and deer.

210 Adam Watson

They were not entirely carrion eaters. On Ain
every summer in 1978-88 | saw 1-3 eating
walkers’ scraps at summits, and in several
years on Glas Maol likewise. In 1995 M
Marquiss (pers comm) saw one fly to
Lochnagar summitto eat plentiful scraps after
walkers left.

Gulls on A

Each summer in 1933-59, D Nethersole-
Thompson (in litt) saw a few Black-headed
Gulls Larus ridibundus. | saw some in each
summer since 1947, with Common Gulls Larus
canus annually since 1969. Out of 754
sightings, 91% were Black-headed, 8%
Common and 1% Lesser B lack-backed Larus
fuscus.

Before ski development, Table 1 shows 13
years with gulls seen and 4 years without, but
afterwards 27 with and O without (Fisher
exact, NS). Annual mean numbers of gulls
and people were related (n=46, r.=0.61).
Highest gull numbers were in 1975,1977 and

Herring Gull

SB 18 (4)

1981, all peak cranefly years (Tipula spp).
This may have been secondary, as high gull
numbers did not occur during cranefly peaks
in earlier years or since 1981. Gull numbers in
the 1980s have been lower (1980-88 v.1971-
79, n=9 & 9, U=13, P<0.02). Also, the biggest
gull totals in a day each year in the 1970s
exceeded those in the 1960s or 1980s (Table
2). This followed an end to successive yearly
increases of people, and coincided with some
declines of people

| saw no gulls on 7 of A’s 30 parts, and nobody
on 5 of these (the same ones as for Crows).
On 30 parts for all 46 years combined, gull
and people densities were related
(n=30,r,=0.78), and gull and Crow densities
(r~=0.71 ). The highest gull density was at
Lochan Buidhe (14.6 per ha in 1971-88),
followed by Fiacaill a’ Choire Chais (5.4),
south of Cairn Gorm and above Coire an t-
Sneachda (0.6). Ben Macdui and Cairn
Lochan (0.4), and low elsewhere.

Gulls on A favoured lochs, huts, paths,
summits south to Ben Macdui, and grassland.

David Mitchell

1996

They often caught insects on the ground, and
less frequently stranded on snow. | saw one
feed on a dead bird. Often they ate walkers’
scraps. Occasionally they followed people,
such as 69 flying with a party from Cairn Gorm
for 1 km on to A (D Gowans, pers comm).
When 51 marines camped for 3 days in June
1975 to remove Lochan Buidhe hut, 28
Common and 6 Lesser Black-backed Gulls
attended, eating copious scraps.

My first date for a gull on Cairn Gorm summit
was 17 April, but into A itself 25 May. | saw
few in August and the last on 25 August. Most
flew north to Glen More in the evening, but in
late July and Augusta few adults and juveniles
stayed up overnight. In August 1971 a Fox
Vulpes vulpes killed 3 Black-headed Gulls in
widely distant parts of A at night. | saw no gull
eating eggs or chicks on A, but sucked
Ptarmigan eggs were found at 2 alpine lochs
where Common Gulls nest (Gordon 1907;
Nethersole-Thompson & Watson 1974).

Gulls on other alpine land

Watson (1979) recorded numbers on Cairn
Gorm ski area and other parts of Cairn Gorm’s
north slope up to 1978. In 1979-88 they were
seen in every summer and on almost every
visit around Ptarmigan Restaurant, as well as
upper Coire na Ciste south to Cairn Gorm
summit. Like gulls on A, they flew north to
Glen More at dusk.

Some Common Gulls have nested at the
alpine Loch nan Cnapan and Loch nan
Stuirteag by Cairn Toul and Braeriach, anda
few Black-headed Gulls at Loch nan Stuirteag
in 1958 (Nethersole-Thompson & (Watson
1974). A few Black-headed Gulls were on
Braeriach each summer in 1933-59 (D
Nethersole-Thompson in litt), and | saw
several of both species eating insects on

Carrion Crows and gulls on Cairngorms plateaux

211

Cairn Toul and Braeriach each summer in
1943-88. | noted few people there and on
other plateaux in the Cairngorms massif
(Watson 1991), and on the latter no gulls.

Common Gulls have long nested at Loch nan
Eun west of Cairnwell, and Loch nan Eun
west of Lochnagar, and they and Black-
headed Gulls in nearby high glens. | saw both
on Lochnagar and Glas Maol each summer
after 1943, eating mainly craneflies. This
included early years with few people, and
later years with many ((Watson 1988). Flocks
were big in years of many craneflies. On 4
July 1977, Glas Maol had huge cranefly
numbers, up to 20 per m¢. It was hard to walk
without standing on them, and so many flew
off that their wings rustled. About 300 Common
and Black-headed Gulls with 12 Herring Gulls
Larus argentatus fed on them.

Discussion

Gordon (1912) found that Crows or gulls ate
every egg in many Ptarmigan nests deserted
after a severe early June snowstorm in the
Cairngorms massif, and saw a Crow pair on
alpine land at Cairn Toulon such an occasion.
In 1953, D Nethersole-Thompson (in /itt) found
many Ptarmigan eggs taken by a few Crows
on Sgoran Dubh in the west Cairngorms, after
scores of hens deserted their nests in a
severe snowfall on 3 June. Clearly, exceptional
numbers of deserted nests led to a few Crows
visiting alpine land temporarily, in years before
many people were there. However, the
sustained occurrence of many Crows
throughout each summer on alpine land inthe
Cairngorms is manifestly human induced.
This followed ski developments in the early
1960s, easier access, and more people
leading to more food scraps.

212 Adam Watson

SB 18 (4)

Many Crows have long bred on Speyside low
ground. Alpine land on Cairn Gorm is marginal,
as no Crows have bred there. One would
expect them to decline there if low ground
stocks were reduced. Crows on Cairn Gorm
in 1992 were scarcer than in 1987-91,
coinciding with increased trapping on
neighbouring Abernethy Estate (Cairngorms
Working Party 1993). Rothiemurchus Estate
greatly increased Crow trapping in 1992 (J
Grant, perscomm at Cairngorm Estate Liaison
Meetings) and Abernethy Estate likewise in
1992-94 (P Mayhew of RSPB at Liaison
Meetings). Crows on A were fewer in 1993,
and fewer still in 1994 (A Amphlett of RSPB
at Liaison Meetings). In summer 1995 | saw
only 2 in 8 days on A and in summer 1996
none in 7 days, the fewest since 1968. in
December 1995 only 9 gathered before
roosting, compared with 35 in December 1991.
KS Bryers (pers comm) saw 8 inthe Reindeer
Company’s fenced area in May 1996.

As gulls nested at alpine lochs in decades
with very few walkers, their presence on
alpine land was natural, eg Gordon (1925)
saw a few Black-headed Gulls on high land
after nesting, and (1963) both species feeding
on moths on Braeriach plateau in June.
However, on A, which had no colony nearby,
the sustained presence of many in summeris
clearly human induced.

Both sides at the Lurcher’s Gully Public Inquiry
in 1981 accepted that Crows and gulls were
unnaturally common on Cairn Gorm and A,
were typical of low ground and alien to alpine
land, and threatened hill birds. One could
lessen the cause - tourists’ discarded food
scraps - by education. Killing Crows and gulls
would ignore this, and so would have to be
repeated.

Acknowledements

| thank Dr D B A Thompson and Prof B W
Staines for comments.

References

Cairngorms Working Party 1993. Common Sense
and Sustainability: A Partnership forthe Cairngorms.
Scottish Office. Edinburgh.

Gordon SP 1907. Birds of the Loch and Mountain.
Arnold. London.

Gordon S 1912. The Charm of the Hills. Cassell.
London.

Gordon S 1925. The Cairngorms Hills of Scotland.
Cassell. London.

Gordon S 1963. Highland Days. Cassell. London.
Morris D, Hammond, EC & Kessler CDJ 1974.
Cairngorms National Nature Reserve :A Reporton
the Characteristics of Visitor Use. Nature
Conservancy Council. Aviemore.

Moss R & Watson A 1984. Maternal nutrition, egg
quality and breeding success of Scottish Ptarmigan
Lagopus mutus. Ibis 126:212-220.
Nethersole-Thompson D & C 1961. Breeding
behaviour of the Dotterel. In Bannerman D A 1953-
63, The Birds of the British IslesNol X: 246-253.
Nethersole-Thompson D & Watson A 1974. The
Cairngorms. Collins. London.

Watson A 1966. Hill birds of the Cairngorms
Scottish Birds 4:179-203.

Watson A 1979. Bird and mammal numbers in
relation to human impactat ski lifts on Scottish hills.
Journal of Applied Ecology 16:753-764.

Watson A 1981. Detailed Analysis. Evidence for
Lurcher’s Gully Public Inquiry.

Watson A 1982. Effects of human impact on
ptarmigan and red grouse near ski lifts in Scotland.
Annual Report of Institute of Terrestrial Ecology
1981:51

Watson A 1984. Paths and people in the
Cairngorms. Scottish Geographical Magazine 100:
151-160.

Watson A 1988. Dottere! Charadrius morinellus
numbers in relation to human impact in Scotland.
Biological Conservation 43:245-256.

Watson A 1989. Dotterel populations and spacing
on three Scottish areas in 1967-86. Ornis Fennica

1996 Carrion Crows and gulls on Cairngorms plateaux 213

Common Gulls Wm Paton
66:85-99.
WatsonA 1991. Increase of people on Cairn Gorm Manuscript accepted March 1996

plateau following easier access. Scottish
Geographical Magazine 107:99-105.

A Watson, c/o Institute of Terrestrial
Ecology, Banchory Research Station,
Hill of Brathens, Glassel,
Kincardineshire, AB31 4BY

214 Scottish Birds (1996) 18: 214-221

The Troup Head gannetry

SB 18(4)

S WANLESS, J MATTHEWS & W R P BOURNE

The gannetry on Troup Head was founded in 1988 and is currently the only
mainland gannetry in Scotland. This paper summarises changes in
numbers, breeding success and chronology between 1988 and 1995.
Numbers of individuals, apparently occupied sites, nests and club birds all
increased rapidly between 1988 and 1993, there was a general decline in
1994 but numbers recovered in 1995. The breeding population increased at
an average rate of 64% p a in 1988-95 and most of this increase must have
been due to immigration. Breeding success averaged 0.53 chicks per nest
and was generally lower than that at other British gannetries over the same
period. This was possibly due to the high proportion of birds breeding for
the first time in any year. There was no evidence that laying had become
earlier, or that breeding success had increased, as the colony grew.

Troup Head, 14 km east of Banff on the
southern side of the Moray Firth is the only
site where Gannets Morus bassanus breed
onthe mainland of Scotland. The first reported
sighting of prospecting Gannets was in 1986;
breeding was definitely recorded in 1988 but
could have occurred the year before
(Matthews & North 1989). Since then, the
colony has grown dramatically and this paper
draws together the counts and observations
which have been made up to 1995 in order to
document changes in the distribution,
numbers, breeding success and timing of
breeding of the colony. These results are
compared with those from the Bass Rock,
East Lothian, the nearest gannetry (Nelson
1978), and Bempton, Humberside, currently
the only other gannetry on the British mainland
(J Fairhurstin Nelson 1978) and other recently
colonised colonies in Scotland (Murray &
Wanless 1986).

Methods

During each season from 1988 to 1995,

several observers visited the colony, mainly
between May and August. The Gannets are
extremely difficult to countand observe. Some
of the viewpoints are far (0.5 km) from the
birds, while other sections can be viewed only
by scrambling down steep, unstable slopes
above the cliffs. A variety of different counting
units have been used by the various observers:
a) the number of individuals in the breeding
area and/or the number of apparently occupied
sites (one or 2 Gannets occupying a site
suitable for breeding, irrespective of whether —
any nest material is present). Both these
units give an indication of the total number of
birds associated with the colony; b) the
number of nests which provides an index of
the breeding population (though not all nests
have contents); and c) the number of chicks
which when expressed as a proportion of the
breeding population gives an index of breeding
output. Counts have also been made of the
number of birds attending clubs and the ©
proportion of these individuals in sub adult
plumage.

1996

Ai least one estimate of most of the above
units was available for each year 1988-95,
thus enabling annual changes to be assessed.
Where several estimates were available the
highest was used since we judged that lower
values were due either to the timing of the
countorto an observer's unfamiliarity with the
colony. Counts and annual estimates of
breeding success have previously been
published in JNCC Annual Reports on seabird
numbers and breeding success in Britain and
Ireland. In a few instances, figures in the
present paper differ slightly from these earlier
values, due to differences in interpretation. In
most years a sample of chicks was aged in
Augustusing plumage characteristics (Nelson
1978). Hatching dates of these chicks were
back calculated and converted to laying dates
by assuming an incubation period of 44 days
(Nelson 1978).

Troup Head Gannetry 215

Results and discussion
Distribution

The first area colonised by Gannets was the
stretch of cliff between Ignet Craig and
Mercurry Heugh (section X in Fig 1). In 1989,
birds moved onto the east side of Mercurry
Heugh (W) and in 1991 there was a further
eastern expansion with the colonisation of
the section of cliff between Mercurry Heugh
and Thirlet Point (V). The west face of Troup
Head (Ue) was first occupied in 1992.

Changes in numbers

The maximum number of birds recorded
during each season increased dramatically
between 1988 and 1993 and then remained
at around 1100 until 1995 (Table 1). No
precise figures for the number of apparently
occupied sites are available for 1988-1990,

Figure 1 Location of the Troup Head gannetry showing the counting divisions used

Collie Head

Thirlet Point

Troup Head

0 200 m

216 S Wanless et al

but subsequent counts showed an increase
up to 1993, a slight drop in 1994 anda
recovery to the 1992 level in 1995.

Between 1988 and 1993 numbers of nests
increased annually (Table 2) and the average
rate of increase was a spectacular 86% per
annum. This period of rapid increase ended
in 1994 when numbers fell by 29%. However,
the population recovered in 1995 to well above
the 1993 level.

Table 1 Peak counts of the number of
Gannets present an of apparently
occupied sites (AOS) at Troup Head
between 1988-95

Birds AOS
1988 25 5+
1989 100+ 19+
1990 395 38+
1991 488 200
1992 900 345
1993 1103 577
1994 No count 500
1995 1097 533

Evaluating changes where annual counts
have been made by different observers, often
using slightly different methods at different
times of the season, is difficult since it is
impossible to rule out observer and/or
sampling error as the source of the difference.
While this reservation should be borne in
mind, there do appear to be several reasons
for believing that the decline in 1994 was real.
First, counts made through the season show
that the number of chicks present in the
period before the first young should have
fledged decreased rather than increased;
second, the number of club birds associated
with the colony was also much lower than
normal (see later), and third, section Ur was
completely deserted by late August (Table 2).

SB 18 (4)

Ug is the most accessible part of the colony.
Foxes Vulpes vulpes have been seen in the
vicinity of Ignet Craig and there was an earth
with at least 2 cubs in a crack half way down
the cliff west of the gannetry in June 1994. It
is therefore possible that the desertion could
have been due to Fox predation. However,
since decreases were also recorded by
several observers independently in some of
the other less accessible areas, additional
unknown factors may have been involved.
The fact that in 1995 numbers recovered to
well above the 1993 level suggests that, for
some reason, many birds which had built
nests in 1993 did not do so in 1994 but they
and a substantial number of new recruits
nested in 1995. While the reason for this
temporary decrease is unknown the pattern
is reminiscent of the non breeding events
which have periodically been recorded in the
Shags Phalacrocorax aristotelis onthe Isle of
May (Aebischer 1986, Aebischer & Wanless
1992) and it is possible that unfavourable
feeding and/or weather conditions deterred
many Troup Head Gannets from nesting in
1994.

Even with the setback in 1994, the average
rate of increase in numbers of breeding pairs
over the first 7 years of the colony was 64%
pa. This rate was broadly similar to those
recorded for the 2 most recently founded
Shetland colonies, Fair Isle and Foula, after
their establishment, but considerably more
rapid than the 6% paincrease initially recorded
for the Flannan Isles, Western Isles (Murray
& Wanless 1986).

Timing of breeding

No direct observations of laying dates have
been made at Troup Head but calculated first
egg dates varied between 3-24 April (Table
3). Except for 1991, when the date of the
median egg was 20 May, the midpoint of

os oe

1996 Troup Head Gannetry 217

Table 2 Nest counts for sections of the Troup Head gannetry, 1988-95. The
locations of the sections are shown in Figure 1

Number of nests in section

Year Date UE V Ww X Total Count unit
1988 29 June 5 5 Total nests
1989 13 Aug a 15 7 Total nests
1990 10Aug 5 a2 37 Well built nests
1991 10Aug 18 oT 49 94 Well built nests
1992 8June 32 100 45-50 57-59 234-241 Well built nests
23 July 29 83 42-43 54 208-209 ‘Total nests
1993 28June 35 104 76 105 320 Well built nests ?
4 July 49-51 139-142 73-74 67-68 328-335 Total nests
1994 12 June 14 103 64 58 239 Well built nests
26 July 3 5 75 58 181 Well built nests
20 Aug 0 26 72 26 124 Well built nests
1995 4May 31 110 159 58 358 Total nests
31 May 49 167 175 139 530 Total nests

Table 3 Estimated breeding chronology of Gannets at Troup Head, 1988-95. No data
are available for 1993

Laying date of

Year n First Median Last

egg egg egg
1988 4 13 May 20 May 26 May
1989 8 14 April 28 April 2 June
1990 14 8 April 3 May 27 May
1991 51 18 April 20 May 6 June
1992 100 ~ <23 April 3 May >25 June
1994 106 24 April 10 May 29 May
1995 ? c. 3-10 April - .

Note: in many years last egg dates were undoubtedly later than indicated

218 S Wanless et al

SB 18 (4)

laying occurred in late April/early May. The
latest laying date was typically in late May/
early June. However, since in some years a
few birds were apparently still incubating
during the August checks and 3 chicks were
present in October 1992, actual last egg
dates were undoubtedly later than these
checks suggest.

First and median egg dates for Troup Head
were generally later than those for the Bass
Rock (2 March-5 April, 15-20 April
respectively), Bempton (25 March-early April,
17-23 April) and Hermaness (late March -
early April, late April ?) while the latest egg
dates were usually earlier (third week of June-
mid July, first half of June and June ? for the
Bass Rock, Bempton and Hermaness
respectively; Nelson 1978).

At Bempton, Nelson (1978) noted that, as
numbers increased, laying date became
earlier. Thus mean laying data advanced
from early May to mid April. In contrast, at
Troup Head there was no evidence of any
systematic advancement of laying date with
increasing colony size. However, Nelson
(1978) found that this temporal trend could be
reversed if the proportion of young,
inexperienced breeders in the colony
increased and this effect could, at least to
some extent, explain the lack of any change
at Troup Head.

Breeding success

Breeding success averaged 0.53 over the 8
years (Table 4), a value which is markedly
lower than the 0.75 recorded in an intensive
study during the 1960s on the Bass Rock
(Nelson 1978), and which is at the lower end
of the range obtained for other British
gannetries using similar methods during the
late 1980s and early 1990s (Walsh et al
1994).

Table 4 Breeding success of Gannets at
Troup Head 1988-95. Success was
calculated as the maximum number of
chicks recorded in August expressed asa
proportion of the peak nest count

Number of

Year Nests Chicks Breeding

success
1988 5 4 0.80
1989 1 / 8 0.47
1990 37 14 0.38
1991 94 52 0.55
1992 241 121 0.50
1993 335 190 0.57
1994 239 146 0.61
1995 530 184* 0.35*
Mean+SD 0.53+0.13

* underestimate

There was no progressive increase in breeding
success with increasing numbers, an effect
which had been noted at Bempton and in a
small isolated group on the Bass Rock (Nelson
1978). The lack of any relationship at Troup
Head may be associated with the high
proportion of birds breeding for the first time
in any year (see later), since this category
tends to have alow breeding success (Nelson
1978).

Non breeding club birds

In 1988, the Gannets used as a club a site
previously occupied by loafing Kittiwakes
Rissa tridactyla. The area was situated 10-20
m above and to the west of the ledge first
colonised by the Gannets. Counts of this club
show considerable variation within a season.

|

1996

Maximum numbers recorded increased
rapidly between 1988 and 1991 but
subsequently remained at 300-400 birds, the
exception being 1994 when the highest count
was only 60 (Table 5). A few birds in sub-
adult plumage were present in 1988. The

- proportion of immatures increased as the

| colony grew and in 1994 and 1995 this

category accounted for 20-25% of the total.
Nelson (1978) concluded that an effect of
increasing colony size was a progressive rise
in the proportion of immature plumaged birds.
Such an increase was evident at Troup Head
and may, in part, have reflected the return of
young birds to their natal colony.

_ Asecond club was recorded on Ignet Craig in

1992 (Section Z in Fig 1). The area was
definitely occupied in 1994 and 1995 but no

counts were reported in 1993. However,

whether this was due to lack of usage or to its
being overlooked by observers is not clear.
There was also a club above Thirlet Point in
1995.

Table 5 Maximum numbers of club birds
and the percentage of Gannets in sub
adult plumage in the Troup Head gannetry

1988-95
Year Number of % sub adults
club birds
1988 <24 some
1989 55 some
1990 59 some
1991 314 some
1992 394 9
1993 275 9
1994 60 25
1995 369 20+

Troup Head Gannetry 219

Population dynamics

Nelson (1978) estimated thatthe intrinsic rate
of increase of the Gannet was around 3% pa.
Like many othercolonies in the years following
their establishment, the increase in numbers
at Troup Head greatly exceeded this intrinsic
rate indicating that substantial immigration
had occurred. The level of immigration was
estimated using a simple population model in
which the observed number of females
breeding in any year (assumed to be
equivalent to the peak number of nests) was
compared with the number expected from the
previous year’s total, after allowing for losses
due to adult mortality (assumed to be the
same as on the Bass Rock ie 6%, Nelson
1978). The results show that in every year,
except 1994, the observed number of females
greatly exceeded the predicted value
indicating that a high percentage of the birds
present in the breeding population each
season were new recruits with estimates
varying from 32% in 1993 to 66% in 1990 (Fig
2). Furthermore, since Gannets do not
normally breed until they are 4 or 5 years old
(Nelson 1978), none of the increases up to
1992 were attributable to Troup Head bred
birds. A crude estimate of potential numbers
of natal recruits was calculated by applying
survival rates to 4 or 5 years of age (assumed
to be the same as those estimated for Bass
Rock birds eg 0.19 and 0.18, Nelson 1978),
to the peak chick count in any year.
Comparison of these predicted numbers with
the estimated number of new recruits each
year (Fig 2) shows that between 1992-95<2%
of recruits were likely to have been reared at
Troup Head. However, once the stronger
cohorts from 1992-95 start to recruit, this
percentage may increase, assuming that
these birds show ahigh level of colony fidelity.

220 S Wanless et al SB 18 (4)

Figure 2 A comparison of observed numbers of breeding female Gannets at Troup Head
1988-95 (-—-) and those predicted from the previous year’s total assuming an annual
adult survival rate of 0.94 (......). The difference between the observed and predicted
values indicates the number of recruits. The lower than predicted total in 1994 may have

been due to non breeding by experienced birds

ml Swe lets li =? oa 0 alk on

(7p)
@® 500
rr)
£
& 400
5)
=
mo]
S. ».d00
@
eee
oO
w= 200
(o)
ime
2 100
i=
—
= O
1988 1989 1990 1991

The origins of immigrants to the Troup Head
gannetry are unknown. Ringing studies at
several recently colonised gannetries in
Norway have shown that there is considerable
variation in the distances moved by immigrants
(Barrett & Folkestad 1996). Thus, although
some individuals come from nearby colonies,
in one case a bird reared at Les Etacs in the
Channel Islands recruited into a colony in
northern Norway, a movement of 3000 km.
To date, no attempt has been made to catch
or ring birds at Troup Head but such a study,
although technically challenging, might
provide interesting information on the identity
of Gannets using the colony.

Acknowledgements

Weare grateful to all those who have provided
counts and information about the Troup Head
gannetry, in particular Successive co
ordinaters of the JNCC Colony Site Register:

1992
Year

1993 1994 1995

Clare Lloyd, Paul Walsh and Kate Thompson
and also Mark Tasker, lan Francis, Stuart

|

Murray and Mike Harris. Bryan Nelson, Mark i

Tasker, Kate Thompson and Mike Harris also |

improved an earlier draft of the paper with
their comments.

References

Aebischer N J 1986. Retrospective investigation of

an ecological disaster in the Shag, Phalacrocorax

aristotelis: a general method based on long-term |
marking. Journal of Animal Ecology 55: 613-629. —
Aebischer N J & Wanless S 1992. Relationships —
between colony size, adult non breeding and ©

environmental conditions for Shags Phalacrocorax

aristotelis on the Isle of May, Scotland. Bird Study |

39: 43-52.

Barrett R T & Folkestad A O 1996. The status of the
North Atlantic Gannet Morus bassanus after 50
years in Norway. Seabird 18: 30-37.

Matthews J & North S 1989. Gannets breeding on
mainland Scotland. Scottish Birds 15: 132-133.

1996 Troup Head Gannetry = 221

Murray S & Wanless S 1986. The status of the WalsnPM, Brindley E & Heubeck M 1994. Seabird
Gannetin Scotland 1984-85. Scottish Birds14:74- | numbers and breeding success in Britain and

85. Ireland, 1993. UK Nature Conservation, No. 17,
Nelson JB 1978. The Gannet. T& ADPoyser, Joint Nature Conservation Committee,
Berkhamsted. Peterbororough

Sarah Wanless, Institute of Terrestrial Ecology, Hill of Brathens,
Banchory, Kincardineshire AB31 4BY

Jill Matthews, c/o Scottish Natural Heritage, Wynne-Edwards House,
17 Rubislaw Terrace, Aberdeen AB1 1XE

W RP Bourne, Department of Zoology, Aberdeen University,
Tillydrone Avenue, Aberdeen AB9 2TN

Revised manuscript accepted March 1996

Gannets David Mitchell

222 Scottish Birds (1996) 18:222-230 SB 18 (4

The feeding behaviour of Greylag and Pink-footed Geese
around the Moray Firth, 1992-93

| J STENHOUSE |

Feeding Greylag and Pink-footed Geese were studied on the coastal plain of

the Moray Firth. In autumn, Greylags selected cereal stubble and germinat-
ing winter cereals, and, in winter, cereal stubble, germinating winter cereals, —
and turnips. Greylag and Pink-footed Geese both selected improved grass-

land in spring, avoiding grass fields with sheep. Both species preferred
large fields, although fields used tended to be smaller in winter than in
autumn and spring. Both species used new fields at a similar steady rate
throughout the seasons, and were seen on most fields only once.

Introduction

Wild geese have used agricultural land in
Britain since early this century, with some
species relying almost entirely on arable
habitats for feeding for several decades (Kear
1963). However, there is much variation in
feeding behaviour between goose species,
and often even within a species, from one
region to another. Geese in one region may
take advantage of acrop whichis not available
in another eg Pink-footed Geese Anser
brachyrhynchus in Norfolk spent 75-80% of
foraging time on sugar beet tops (Gill 1993).
Several populations and regions have been
extensively studied (Newton & Campbell 1973,
Patterson eta/1989, Gill 1993). Management
schemes have been initiated in some areas.
However, managementwhich proves effective
in one region of the country may not be
applicable elsewhere. This study from October
1992 to May 1993 investigated the feeding
behaviour of migratory Greylag Anser anser
and Pink-footed Geese around the coastal
plain of the Moray Firth.

Study area and methods

The study area covered 122,000 ha of the
coastal plain of the Moray Firth, from the

Dornoch Firth in Ross-shire to Spey Bay in |
Morayshire (Fig 1). The Moray Firth area |
comprises amixture of coastal habitats backed |
by a wide, flat fringe of rich soils. The climate —
is relatively mild, and, as a result, cereals are —

the most important crop, particularly barley
which is sold for malting. The local wildfowl

roost on inland waters and estuaries, andthe |
study area included 15 goose roost sites |

(Stenhouse 1993).

The study area was divided into 6 regions (Fig

1). All feeding goose flocks in each region |
were counted every 2 weeks from roads |

which provided good viewing points without _

disturbing the birds.

Six road transects, each 5km long and 400m |
Crop type was ©
recorded 3 times (autumn, winter and spring) —
foreach field on both sides ofthe roadtransect. —

wide, were used (Fig 1).

The major crop type within 200m of the

1996

Greylag & Pink-footed Geese around the Moray Firth 1992-93

223

Fig 1 Study area showing count regions (1-6) and road transects (a-f)

yA
tf
FA oa
ig 1 ZA
<
get »
2 aa
4 S
gf
of
Po
Boe 2
77 Aaa
f 3
ef
Se J
a : “4
\ -~
eae G Tee =
\ iss 5 ee
A ae
4 a
\ S|
A 4+ Ss
ee en See

roadside was recorded and the length of each
_ crop along the transect was determined from
1: 25,000 maps. Since the transect was of a
fixed width, the area of each crop was assumed
to be proportional to the length of each crop
_ along the transect. The proportion of each
crop type available was calculated from the
total length along the transect of each crop
_ observed divided by the total length of all
_ transects (60km). This provided a sample of
the crops available to the birds through the
seasons.

Crop type, area, boundary type, andpresence
of livestock were recorded for fields visited by
geese, called ‘goose fields’. Similar data
_ were recorded at a nearby field (2 fields away

—_
- =
- =
_- =
-_ =

in arandom direction) not known to be visited
by geese, called ‘nongoose fields’. Data for
autumn, winter and spring were analysed
separately.

Indices of selection of crops were determined
using the equation:

Q=r(1-p)/p (1-r)

where ris the proportion of birds in a given
habitat in a given time period, and p is the
proportion of that habitat available on the 6
transects (Jacobs 1974).

Results

Two hundred and twenty-four observations
were made of feeding goose flocks in 9 crop

224 I J Stenhouse

types in 128 different fields. The total number
of birds observed in each crop type during
each 2 week count period was calculated.
Greylag Geese fed almost entirely on cereal
stubbles in autumn and took advantage of
newly sown grain in late November (Fig 2). A
few records also showed use of turnips during
the poorest weather in late December and
January. Improved grassland was used
throughout, but mostly from late December to
spring. Small amounts of germinated winter

SB 18(4)

cereal and rough grazing were also used in
spring.

Pink-footed Geese passed through the area

quickly in autumn. Improved grassland was |

used intensively by Pink-footed Geese on
their return to the area, and there was some

use of cereal stubbles in February (Fig 3). By |
late March, they began to feed on newly sown —
grain and the new growth of grass in rough ~

grazing.

Fig 2 Crop use by Greylag Geese, showing numbers of birds observed in each crop

type for each fortnightly period

14000
12000
10000

8000

6000
4000

Number of geese

2000

In autumn, alarger proportion of cereal stubble
was available than improved grassland, and
these 2 crop types provided most available
habitat (Table 1). A small proportion of
germinating winter cereal was available, and
very small proportions of turnips and oilseed
rape. In winter, the proportions of cereal
stubble available decreased, due to ploughing,
while improved grassland remained constant
and availability of germinating winter cereal
increased.

In spring, proportions of cereal stubble
remained very small and improved grassland
remained constant. Proportions of
germinating winter cereal available increased

BNe@oOO

Winter cereal - newly sown
Winter cereal - germinated

Rough grazing
Improved grass
Stubble

further, while newly sown spring cereal
increased considerably, and some spring

cereal germinated. Turnips were no longer |

available, while the small proportion of oilseed
rape remained constant (Table 1).

Because few Pink-footed Geese were present —
through autumn and winter, counts for both —
species have been combined for these»

.

seasons. Together, Greylag and Pink-footed —
Geese showed selection for stubble and —

germinated winter cereal in autumn, and for

stubble, germinated winter cereal or turnips —

in winter. However, the data refer largely to
Greylag Geese. In spring, Greylag and Pink-
footed Geese showed, independently,
selection for improved grass (Table 2).

1996 Greylag & Pink-footed Geese around the Moray Firth 1992-93 225

Fig 3 Crop use by Pink-footed Geese, showing numbers of birds observed in ech
crop type for each fortnightly period

[J Spring cereal - newly sown

E Winter cereal - germinated
8000 _

4] Rough grazing
f4 Improved grass

6000 Ea Stubble

4000

Number of geese

Table 1 Proportion of crops available along all transects in autumn 1992, winter
1992/93, and spring 1993

Crop type Autumn Winter Spring

| Improved grass 0.28 0.28 0.28
: Rough grazing 0.00 0.00 0.00
Winter cereal (newly sown) 0.00 0.00 0.00
Winter cereal (germinated) 0.10 0.14 0.18
Spring cereal (newly sown) 0.00 0.00 0.26
Spring cereal (germinated) 0.00 0.00 0.08
Cereal stubble 0.37 0.28 0.03
Carrots 0.00 0.00 0.00
Potatoes 0.00 0.00 0.00
Turnips 0.02 0.02 0.00
Set aside 0.00 0.00 0.00
Oilseed rape 0.02 0.02 0.02
Linseed 0.00 0.00 0.00

Other 0.21 0.26 OnldS

226 | J Stenhouse SB 18(4)

‘Goose fields’ and ‘nongoose fields’ were —_ grass, spring cereals and others (x25=5.64)
compared to explore further possible factors in spring. However, in each case the test
that could beimportantinfieldselection(Table statistic was close to the 5% probability
3). No significant differences were found threshold. No significant differences were
between the number of goose andnongoose _observedbetween goose and nongoose fields
fields under stubble and other crop types according to boundary type, percentage
(x2 4=3.67) in autumn, under grass, stubble —_ standing water and nearest occupied building.
and others (x2 4=4.42) in winter, and under

Table 2 Selection indices by grey geese for different crops. Values over one
indicate selection, under one avoidance

Crop Autumn Winter Spring

both spp both spp Greylag Pink-footed
Improved grass 0.31 1.00 126.00 11.71
Cereal stubble 3.02 2.02 - -
Winter cereal
(germinated) 2.68 1.38 0.04 0.29

Spring cereal
(newly sown) - -
Turnips - Zot - -

Table 3 The number of goose fields and nongoose fields according to crop
distribution

Crop type Goose fields Nongoose fields

autumn winter spring autumn winter spring

Improved grass 10 8
Rough grazing

Winter cereal (germinated)
Spring cereal (newly sown)
Spring cereal (germinated)
Stubble

Turnips

Oilseed rape

Plough

No
rOoSoheoeo © oh
NO=-AD0o0on0od
—_—h,
Qook 6 oo ore
coal,
y=-ON-AON=0

oOoo-'OoO°Oo

1996

There were significant differences between
the size of goose and nongoose fields, with
goose fields being larger (F4 4g9=3.0,
P=0.08). Also, there were significant seasonal
changes (Fo 4g9=7.9, P<0.001), with goose
fields tending to be smaller in winter compared
with autumn and spring (Table 4).

Greylag & Pink-footed Geese around the Moray Firth 1992-93

227

Selecting only grass fields, a significant
association was found between geese and
grass fields without sheep (Fisher Exact Test,
2 tailed, P=0.021). The presence of cattle in
the area was So rarely recorded that the data
were unsuitable for statistical analysis (Table
5)

Table 4 Field areas (ha) for goose fields and nongoose fields

Goose fields

Season mean SD range
Autumn bt8 5.2 5-25
Winter 8.0 3.5 3-19
Spring 12.2 6.2 5-32

Nongoose fields

mean SD range
9.5 3.2 4-17
S27 3.2 4-19

10.5 4.0 4-25

Table 5 The presence of livestock in grass fields used and not used by geese

Sheep Cattle
present not present present not present
Goose fields 2 22 0 24
Nongoose fields | vf 10 1 16

Greylag Geese were observed in 104 fields,
and Pink-footed Geese in 53. Both species
were recorded in 29 fields, butnot necessarily
at the same time. However, where they were
recorded at the same time, they usually formed
discrete flocks. A total of 128 different fields
were used by geese throughout the study
period (Fig 4). The number of fields used by
Greylag Geese steadily increased through
the study, suggesting that Greylags selected

new fields at a steady rate. The number of
new fields used did not appear to be density
dependent, since it remained steady through
the seasons, while the total number of Greylag
Geese decreased from autumn to winter.
The total number of fields used by Pink-
footed Geese did notincrease through autumn
and winter since only a few birds passed
through in autumn (Fig 4). However, when
Pink-footed Geese returned to feed in the

228 I J Stenhouse

area in February, the pattern was similar to
that of Greylags, with a steady increase in the
total number of fields used. Again, the number
of new fields used did not appear to be density
dependent, and remained steady, while the
total number of Pink-footed Geese increased

SB 18(4)

through the spring. Both Greylag and Pink-
footed Geese were recorded in very few fields
more than once (Fig 5). Onthe rare occasions
when geese were recorded on a field more
than 3 times, these fields were always
immediately adjacent to a roost site.

Fig 4 The cumulative number of fields used by Greylag and Pink-footed Geese over

the study period
110 =; £-- 90 Greylag
ss) 100 ——*—__Pink-footed
= 90
5 80
oS 70
E s.60
5 50
a
f=}
ae
10

OP TN Nee

F MM A A M

J F
Month

Fig 5 The number of times Greylag and Pink-footed Geese were observed in

individual fields

80
70
60

50
40

Number of fields

1 2 3 4

Greylag
— Pink-footed

5 6 if, 8 9 10

Number of times geese observed

1996

Discussion

The general pattern of goose feeding
behaviour around the Moray Firth showed
seasonal changes in the birds’ diet which
followed seasonal changes in food availability.
Around the Moray Firth, Greylag Geese
concentrated on stubble in autumn and early
winter, and both Greylag and Pink-footed
Geese concentrated on improved grass in
spring. As in previous studies (Newton &
Campbell 1973, Foshaw 1983, Bell 1988,
Patterson et a/ 1989), little use was made of
winter cereals.

Several wildfowl species are known to exploit
unusual food resources in response to short
term climatic extremes (Kear 1962), and
Greylag Geese were observed in turnips ona
few occasions in January, coinciding with a
cold period and heavy snowfall. Similarly,
turnips were eaten by Greylag Geese around
Loch Leven only when alternative foods were
scarce, after snowfalls (Newton & Campbell
1973). Carrots are recognised as a favoured
food of Pink-footed Geese in Lancashire and
of Greylag Geese in some parts of Scotland
(Owen 1990). Carrots were grown in very few
sites within the study area and Greylag Geese
were recorded feeding on waste roots left
behind after harvest only once.

The feeding behaviour of the 2 species
appeared to be similar in terms of consistency
of use of fields. Both used a large number of
fields briefly and concentrated their activity in
very few fields. This suggests that neither
species is particularly predictable in their field
use beyond a few fields immediately adjacent
to major roosts.

Field size and location are recognised as
major determinants in field choice (Newton &
Campbell 1973, Gill 1993). Field area was

Greylag & Pink-footed Geese around the Moray Firth 1992-93 229

found to be important around the Moray Firth,
where geese fed in larger fields throughout
the seasons, but used smaller fields in winter
than in autumn and spring. Large fields
provided a better opportunity to avoid field
edges, where geese are vulnerable to hunting
and disturbance (Owen 1973). The
acceptance of smaller fields in winter may
reflect short term food shortages forcing use
of suboptimal sites.

The comparison between goose fields and
transect data indicated distinct seasonal crop
preferences. These were not suggested by
the more rigorous paired field comparison,
perhaps because similar crops are grown
close together. However, the data showed
that geese avoided grass fields containing
sheep, mostlikely due to the disturbance and
break up of flock structure.

Although this study did not attempt to assess
the extent of damage, the results do have
implications for any future management
proposals. The feeding behaviour of grey
geese within the study area suggests that any
problem of crop damage will be relatively
short lived, as well as limited in area.
Therefore, the problem is not so much one of
repelling geese from farmland as a whole, but
of protecting valuable crops for relatively short
periods of time (as was previously also
suggested by Newton & Campbell 1973).

References

Bell M V 1988. Feeding behaviour of wintering
Pink-footed and Greylag Geese in north-east
Scotland. Wildfowl 39:43-53.

Forshaw W D 1983. Numbers, distribution and
behaviour of Pink-footed Geese in Lancashire.
Wildfowl 34:77-88.

Gill J 1993. Pink-footed geese in Norfolk. /n Proc
Conf on Current Issues in the Ecology and
Conservation Management of Arctic Breeding

230 | J Stenhouse

Geese. BOU, Sunderland.

Jacobs J 1974. Quantitative measurement of food
selection: A modification of the forage ration and
Ivlev’s Electivity Index. Oecologia 14:413-417.
Kear J 1962. The history of potato-eating by
wildfowl in Britain. Wildfowl 14:54-65

Kear J 1963. Wildfowl and agriculture. In Atkinson-
Willes G L (Ed), Wildfowl in Great Britain: pp 315-
328. HMSO, London.

Newton | & Campbell C 1973. Feeding of geese on
farmland in east-central Scotland. J. applied
Ecology 10:781-801.

Owen M 1973. Tne management of grassland
areas for wintering geese. Wildfow! 24:123-130.
OwenM 1990. The damage-conservation interface
illustrated by geese. /bis 132:238-252.
Patterson | J, Abdul Jalil S & East M L 1989.
Damage to winter cereals by Greylag and Pink-
footed geese in northeast Scotland. J. applied
Ecology 26:879-895.

Stenhouse | J 1993. Grey Geese in the Moray Firth.
Reportto Scottish Natural Heritage Inverness, pp58.

SB 18 (4)

Acknowledgements

| would like to thank Drs N E Buxton and R W
Summers for their guidance and support
throughout this project, with further thanks to
Dr R W Summers for useful comments on an
earlier draft of this paper. | am grateful to |
Brockway and R_ Evans of the Moray Firth
Goose Study Group, R L Swann of the
Highland Ringing Group, and C Mitchell of the
Wildfowl & Wetlands Trust for providing
valuable information, as well as their
enthusiasm and encouragement. This project
was funded by Scottish Natural Heritage as
part of their ongoing research in nature
conservation, and their support is gratefully
acknowledged.

lan J Stenhouse, Biopsychology Programme,
Memorial University of Newfoundland, St. John’s,

Newfoundland A1B 3X5, Canada

Revised manuscript accepted April 1996

Pink-footed Goose

N

IS

Keith Brockie

Scoitish Birds (1996) 18: 231-241

231

The numbers of breeding waders in lowland Scotland

MARK O’BRIEN

Populations of breeding waders on lowland areas of Scotland were
estimated using a stratified random sample of sites. 648 km? of land were
surveyed by SOC volunteers and paid fieldworkers. A total of 82,500 pairs
of Oystercatchers, 92,000 pairs of Lapwings, 41,000 pairs of Snipe, 35,500 to
55,000 pairs of Curlews and 12,000 pairs of Redshanks were estimated.
Numbers are compared with previous population estimates, and reasons for

differences discussed.

Introduction

Five species of breeding waders,
Oystercatcher Haematopus ostralegus,
Lapwing Vanellus vanelius, Snipe Gallinago
gallinago, Curlew Numenius arquata and
Redshank Tringa totanus, are commonly
associated with farmiand habitats in the United
Kingdom. Thesize of Scottish farmland wader
populations was estimated in the early 1980s
(Galbraith et a/ 1984). This work indicated
that considerable numbers of waders were
thinly distributed across much of Scotland’s
farmland. Estimates of Scottish populations
of Lapwings, Oystercatchers and Redshanks
were made, but the numbers of Snipe and
Curlews were considered likely to be high on
upland areas, so Scottish population estimates
for these 2 species were not attempted.
Farmland wader populations on both the
Northern and Western isles have been
surveyed in more detail, and show higher
densities compared with areas on the
mainiand (Campbell et a/1988a,b, Fuller et a/
1986, Galbraith et a/ 1984). This is confirmed
by the density maps in the New Breeding
Atlas (Gibbons ef a/ 1993).

Recent surveys elsewhere in the UK have
given more accurate estimates of wader
populations and indicated considerable

declines in numbers of some species, giving
cause for concern over wader populations on
farmland. In 1987, there were estimated to be
123,124 pairs of Lapwing in England and
Wales (Shrubb and Lack 1992) and numbers
were shown to be falling, especially in the
south. Estimates of the numbers of waders in
Wales and Northern England have recently
been attempted (Green efa/ 1994, Murray et
al 1994). Welsh populations, in particular,
are giving great cause for concern. Data on
both the numbers and trends in numbers are
also available for areas of lowland wet
grassland in England and Wales (Smith 1983,
O’Brien and Smith 1992) and Northern Ireland
(Partridge and Smith 1992, Partridge 1992).
These studies show substantial declines in
farmland wader populations over the last few
years in the areas surveyed. Similarly,
comparing the distribution of waders recorded
in the 2 breeding atlases suggests a marked
reduction in range, at least for some species
(Gibbons et a/ 1993).

The main aim of this project was to update
population estimates for waders in lowland
Scotland.

Methods

The only practical way of estimating breeding

232 M O’Brien

SB 18 (4)

wader populations in lowland Scotland is
randomly to select representative sites for
survey. This will miss many sites known to be
good for breeding waders. The best approach
is to use a Stratified random survey where
‘identified wader sites’ (IWS) and a random
sample of the remaining lowland farmland
are surveyed. This follows the method used
to estimate breeding wader populations in
Northern Ireland (Partridge and Smith 1992).
A population estimate is then calculated by
summing the estimates derived from the IWS
and the random sites.

Information on the location and importance of
key sites for lowland breeding waders in
Orkney, Shetland and the southern isles of
the Outer Hebrides has previously been
collected (Campbell et a/ 1988ab, Galbraith
et al 1984, Fuller et a/ 1986). Comparable
information for mainland Scotland was not
available so Scottish Ornithologists’ Club
(SOC) regional organisers, British Trust for
Ornithology (BTO) regional representatives,
RSPB conservation officers and Scottish
Natural Heritage (SNH) area officers were
asked to list sites that they considered to be
important for breeding waders within their
areas. This resulted in a list of about 230 key
sites for breeding waders throughout mainland
Scotland. Each of these key sites was marked
on to the relevant Ordnance Survey 1:25,000
map, fields on the map were numbered and
copies of the maps provided for use whilst
surveying breeding waders. Surveys of the
sites were undertaken primarily by volunteers.
SOC regional organisers and BTO regional
representatives provided much help and
support in ensuring coverage.

To determine the location of the random
lowland sites, we needed a definition of
‘lowlands’ and an estimate of its location and
total area in Scotland. This was done using
the Macaulay Land Capability for Agriculture

classification system (LCA, Soil Survey of
Scotland 1982b) which integrates information
on soil, climate and relief, splitting Scotland
into 7 land classes, each of which is split into
1-3 divisions. Acomparison of the distribution
of IWS and land classes indicated that most
were on land classed as between 1 and 5.3,
ie from prime agricultural land through to
improved grassland. For the purposes of the
study this range of classes and divisions was
taken to represent the lowlands in Scotland.
This definition of lowland covers approximately
36,000 km? of Scotland, or 46.7% of the total
area of Scotland. It must be stressed that this
is a land capability rather than a landuse
Classification.

Estimating the average density of breeding
waders in Scotland - the random squares
survey

This part of the survey was split into 5 sections.
In 1992, 4 teams of 2 surveyors were employed
to survey a random sample of sites on
mainland Scotland and the Inner Hebrides. In
1993, 3 teams of 2 surveyors surveyed a
random sample of lowland sites on Shetland,
Orkney and the Uists. In addition, randomly
selected sites were surveyed on Lewis and
Harris by RSPB staff based on the islands.

For the purposes of the 1992 survey alowland
site was defined as any one kilometre square
where 75% or more of the square was
described as lowland (as defined above),
using information available on 1:25,000 LCA
maps (Soil Survey of Scotland 1982a). All
lowland squares from mainland Scotland and
the Inner Hebrides, with the exception of
those that had already been defined as an
IWS, were available for selection in this study.
A randomly selected sample of 240 one
kilometre squares from this set provided
unbiased estimates of the average density of
breeding waders on lowland mainland

1996

Scotland, away from the WSs.

In 1993, a lowland site was defined as any
one kilometre square where the predominant
land class was described as lowland. This
differs from 1992 in that the predominant land
class may not cover 75% of the square, since
many of the units of lowland land classes on
the islands are small relative to the size of the
survey unit. Key sites in the islands were not
considered separately, so the randomly
selected squares can be considered to
represent the best estimates of the average
density of lowland breeding waders in the
islands.

Methods used to survey breeding
waders

We used the field by field method to survey
breeding waders, previously used inthe survey
of lowland wet grasslands in England and
Wales (O’Brien and Smith 1992) and farmland
sites in Northern Ireland (Partridge and Smith
1992). The main difference between our
survey and the lowland wet grassland survey
in England and Wales is the time of day that
visits were made. This survey aimed to visit
sites within 3 hours of dawn or dusk. These
are the times when highest counts of breeding
waders are obtained (Reed et a/ 1985) and it
is known that these provide the best estimate
of number of breeding Snipe (Green 1985).
The disadvantage of dawn and dusk surveys
is that the area that can be surveyed in the
time is limited to about 100 ha per session.

All surveyors were provided with 3 copies of
a map of the survey area together with 3 sets
of recording forms. Each map was marked
with the site boundary, within which all fields
were numbered. Previous studies indicate
that 3 evenly-spaced visits will provide an
overall picture of the number of waders within

Breeding waders in lowland Scotland

233

asite (Smith 1983). Surveyors were asked to
visit sites on 3 occasions, visit 1 between 18
April and 8 May, visit 2 between 9 May and 29
May, visit 3 between 30 May and 19 June.

Surveyors were asked to walk through, and
get to within 100 m of any point in, each field,
and to look 2-400 m ahead, scanning with
binoculars to note the distribution of all waders.
Allwader registrations were mapped, although
only birds considered to be breeding within
the survey area were counted (see below).

Standard techniques for interpreting the wader
data and calculating the number of breeding
waders per site were used (Bibby et a/ 1992,
O’Brien and Smith 1992, Partridge and Smith
1992). In detail these were:

a) Oystercatcher: the maximum number
of pairs on any one visit was used (Smith
1983).

b) Lapwing: halving the maximum
number of individuals recorded on the site
between mid April and late May provides the
best estimate of the total number of pairs
(Barrett and Barrett 1983).

C) Snipe: the maximum number of
drumming plus chipping birds on any one visit
was used. The number of drumming plus
chipping birds recorded within the study area
is then multiplied by 1.74 to provide the best
estimate of the number of breeding pairs in
the study area (Green 1986).

d) Curlew: the maximum number of pairs
on any one visit was used. This is recorded
as ‘Curlew, old’in Table 4. Thisis the standard
method for estimating the number of Curlew
and has been used for all surveys published
up to the present time (Smith 1982, Partridge
and Smith 1992, O’Brien and Smith 1992).
Detailed ongoing research, comparing the

234 M O’Brien

number of Curlews recorded using the
methods outlined above with the number of
nests found on a site, suggests that this may
over estimate the number of Curlews (M
Grant pers comm). An alternative estimate is
included in the tables as ‘Curlew, new’, based
on the average number of Curlews recorded
on visits 2 and 3. It is thought that this may
give a better overall estimate of actual Curlew
numbers, although research to determine the
best method is still ongoing. A small number
of sites was only visited on one occasion,
early in the season. The number of pairs of
Curlew seen on that one visit was used in both
the ‘Curlew, old’ and ‘Curlew, new’ estimates.

e) Redshank: the mean density of
individual birds (excluding flocks) counted
before the first nests hatch (about 20th May)
has been shown to be correlated with peak
nest density (Cadbury et a/ 1987). Anumber
of sites in the present survey were visited for
the secondtime after the first Redshank nests
had hatched. Redshank detectability is
considerably higher once the adults are caring
for chicks; both birds in the pair perform a
distinctive, noisy display. In these
circumstances a different method of estimating
the number of pairs is required. The number
of pairs of Redshank acting as if with young

SB 18(4)

are added to the number of individual
Redshanks not acting as if with young on the
site.

Extrapolating the survey data to all
remaining lowland areas

To estimate the total number of breeding
waders in the Scottish lowlands the numbers
estimated on the IWS has been added to the
numbers estimated through the random sites
survey on Scottish mainland lowlands. A
small number of IWS were not surveyed for
several reasons. To allow for this, numbers
estimated on those IWS not surveyed were
calculated by multiplying the area not surveyed
by the density of each species of wader on
the surveyed IWS. The overall population
estimate for |WS was calculated by summing
the numbers seen andthe numbers estimated.

The surveys of random sites provide estimates
of the average density of each species on
lowlands within each of the study areas. These
densities, when multiplied by the area of
lowland within the study area and then added
together, provide unbiased estimates of the
numbers of each of the wader species in
lowland Scotland away from IWS.

Table 1 Population, densities and numbers of pairs of breeding waders on identified

wader sites in lowland Scotland

Species of Average Maximum Total no. Total no.
wader dene densit surveyed estimated
(Pr km“*) (Pr km“*) (No of prs) (No of prs)

Oystercatcher 4.9 25:2 1096.5 1521
Lapwing 11.4 55.0 2559.5 3548
Snipe 5:5 60.6 1225:0 1693
Curlew (old) 3.5 20.7 790.0 1093
Curlew (new) 2.8 18.8 616.9 860
Redshank 4.0 28.7 893.5 1240

235

Breeding waders in lowland Scotland

1996

9S6S 6ZSV 88¢S 9¢ OOl 9'°9 99S Keuyo
00EC BSZI 6c0¢c vi S6 O'S LLE PUB}OYS
900SS 66c6€ Lc69V OL LS Vit voce pueuley|
(pjo) MeyND (Ip
GZ61 v96 88rl vas 06 9'°9 99 SUIEH/SIMO)
v8sEe 0cle c8ce Ze v6 MAIL vs9 S}SIN
6e0r 867 LeLe ve €8 o's gee Aeuwio
96ES Z88e 66SV SV Z6 ce SOZ PUB}OEYS
LZ6SE vS6sl €969¢ Ol 6h 80 981 pue|uley|
edius (9
00E¢ Lvv voc Go 08 v's vS SUJEH/SIMS 7)
€O0E9 S8Sv OLvS €8 96 Goo LZébt S}SIN
8299 coSv 6LvS 6e S6 Z'8 oss Aeuyo
8veCc S9S1l 8ré6l Sl c8 60 66¢c PUB)OYUS
GZ606 LvO8S 9ErrL Co €S CG vLS puejuley|
Buimdey (q
VSI € 89 A 0c €'0 € SWIC H/SIME)
VILLE Leeo 9cl?e 8c v6 Vit 89S S}SI/
OS80l SZ88 L986 6c 86 8'Sl 9SOL Keuyio
9EZLE €lL6c 88CE Lo Z6 €'8 vos PUB)OYS
G8E9ol pO8Ees €c0S9 el vS Gal 6PPV pue|uley|
4249}2919}SAQ (e

(sid Jo On) (5.wy 4d) SP1Iq UIA (5. 4d) (sid Jo ON)
wnwixew Wwnuwiuluw uonejndod isuep soienbs isuaep Asains
S}IWI] BDUBPIUOD % GE poayewiysy wnwixeyy JO JU9910g eBbeioay jE}O] jo valy

PUB]]OIS PURIMO] UI SajIs pajaajas AjWOpues UO ssapeEM Bulpsssg jo ssied Jo sIaquUINN Z e\qeL

SB 18(4)

sss

cles
v9le
OcEl
B9eZ

c6LV
6c91
vEe6vEe

winwixew

cOl
88rC
96E1
9S8
8Lvc

SLSE
Z8LL
cLSEC

winuwiuiw

S}IUI] BUEPYUOS %GG

236 M O’Brien

(sid Jo On)
uonejndod
poeyewnsy

Zt
Lh

(>.uy 4d)
isuep
WwNnwixey,y

SP4IG YIAA
soisenbs
JO JU9D18q

9
Ocl
8c
L?
LO

00
0'0
99
S’€
60

0'0
00

(5.wy 4d)
isuap
ebessay

91 SUJEH/SIME)
Lo9 SISIN
281 AOUHO
SOL puejyeys
ee puejueyy
yueyspay (6
0 SUJEH/SIME7)
0 S}siN
lyy AEUO
Sle Pue)}eYS
Zod pueule|
(meu) MayIND (IP
0 SUJEH/SIMO 7)
0 S}SIN
(sad Jo ON)
Aaains
[e}O] jo ealy

1996

We used the ‘bootstrap’ method to estimate
95% confidence limits around population
estimates (Greenwood 1991, also see
Appendix for further details). These are
presented in Table 2 as the minimum and
maximum confidence limits.

Breeding waders in lowland Scotland 237

Information on the total number of pairs of
breeding waders recorded in each of these
survey areas, together with the average
density, an extrapolated population estimate
and 95% confidence limits around that
estimate are shown in Table 2.

Results c) Overall estimates

The estimates of wader numbers on the
a) Estimates fromtheidentifiedwader Scottish mainlandform avery high proportion
sites (IWS) of the total number of waders in Scotland

The combined estimates of breeding wader
numbers for 186 surveyed IWS, covering
223.8 km@ and 46 IWS (86.7 km 2), which
were not surveyed, are shown in Table 1.

b) Estimates from the random sites
survey

Between 10% and 20% of lowland areas in
the Northern Isles and the Uists were surveyed
(50-60 randomly selected squares surveyed
in each of the island groups), a considerably
higher percentage than on mainland Scotland
(0.7%) and Lewis and Harris (4.4%).

(Table 3). Only Redshank numbers on the
islands approach 50% of all the population
estimates.

The revised (new) method of estimating
Curlew populations produces a figure which
is about 65% that of the original (old) method.
Both figures are substantially higher than
previous estimates of Curlew numbers in
Scotland, despite the fact that they refer solely
to the Jowland Scottish breeding population.

Discussion

Table 3 Estimates of the numbers of pairs of breeding waders in the lowlands of

Scotland

Oystercatcher Lapwing Snipe Curlew Curlew Redshank

(old) (new)

Mainland (key) 1521 3548 1693 1093 860 1240
-- (random) 65023 74436 26963 46921 29253 4779
Shetland 3288 1948 4599 2029 1400 1076
Orkney 9867 5419 3137 5288 4120 1747
Uists 2726 5410 3282 0 0 2885
Lewis/Harris 68 1204 1488 0 0 349
Total 82493 91965 41162 55331 35633 12076
Min. conf. limit 71309 75644 32856 47729 29764 9761
Max. conf.limit 93890 108329 50568 63571 41183 14791

238 M O’Brien

The population estimates for Oystercatcher
and Curlew numbers are considerably higher
than previous estimates (Thom 1986, Piersma
1986). Our Curlew estimate for the Scottish
lowlands is considerably higher than the
previous estimate for the UK as a whole,
whilst the Scottish lowland Oystercatcher
numbers are more than twice the previous UK
estimate (Piersma 1986). Even if we use the
more conservative alternative figure for
Curlew populations, it is apparent that
Scotland has considerably more breeding
Curlews than previously recognised, even
without considering the substantial
populations that are likely to occur in the
upiands. These increases may be explained
by the more rigorous method of survey used
for this project and, in the case of
Oystercatchers, the fact that populations have
been increasing in the UK and throughout
other European countries bordering the North
Sea during the last 20 years (Hotker et a/
1991, Marchant et a/ 1990).

Previous estimates of breeding wader
populations in Scotland (Galbraith et a/ 1994)
were based on farmland lying below 300 m
ASL and were extrapolated across an
estimated area of 19,909 km2, just over half
the area included in the present survey (Table
4). These population estimates are markedly
lower than those given here, a result to be
expected as the authors were extrapolating
to a much smaller area of land. When
population densities for rough grazing above
300 m in Galbraith et a/ (1984) are multiplied
by the difference in areas between the 2
survey it seems that Lapwing numbers were
consistent between the 2 surveys, whilst
estimates of both Oystercatcher and
Redshank numbers increased significantly
(Table 4). The numbers of Oystercatchers
breeding inland have increased since the late
19th century, both in the UK and on farmland

SB 18(4)

in all countries around the North Sea (Gibbons
et al 1993). The new Breeding Atlas does not
show any evidence of an increase in
Oystercatcher range because most of the
squares in Scotland were already occupied at
the time of the first atlas, but it does suggest
that, in mainland Scotland at least, the
distribution of Redshank has declined
markedly, although increases in range are
apparent in Shetland and on Lewis and Harris
(Gibbons et a/ 1993). The present survey
suggests that half the breeding Redshank in
Scotland are on the Northern and Western
Isles. Galbraith et al (1984) do not provide
population estimates for Curlew and Snipe,
as substantial proportions of these populations
occur above 300 m and in the uplands.

Methods and interpretation can greatly affect
estimates of wader populations. Surveys of
the type used in the present study have
usefully been compared with alternative
intensive methods such as nest finding for
Lapwing, Snipe and Redshank (Green 1986,
Barrett and Barrett 1983, Bibby et a/ 1992).
This process is presently being undertaken
for Curlew (M Grant pers comm). O’Brien and
Smith (1992), in comparing the interpretation
of wader numbers on lowland wet grasslands
in 1982, found that the method used in the
present survey increased Lapwing and
Redshank numbers by 34% and 35%
respectively, when compared with
interpretations of the original survey in 1982.
Table 4 (see next page)

Scottish farmland data from Galbraith et al
(1984). Extrapolations of Scottish farmland
data are based on the assumption that
difference in area between the 1983 survey
and the present is equivalent to the area of
rough grazing above 300 m. See text for
details. Uist machair data from Fuller et al
(1986). Shetland data from Campbell (1989).

Table 4 also shows previous population

ee

1996

Breeding waders in lowland Scotland 239

Table 4 Previous population estimates for lowland breeding waders in Scotland

Scottish Scottish farm- Uist Shetland

farmland land estimates machair inbye
Area of survey (km) 19906 35977 186 330
Oystercatcher 22700 29800 2700 1466
Lapwing 63800 99500 4300 1083
Snipe - - 1-2000 243
Curlew : - 0 672
Redshank 5000 6500 2600 287

estimates for wader populations on the
machair of the Uists and inbye land on
Shetland. Comparisons with the data in Table
2 indicate that present estimates for lowlands
on the Uists are similar to those obtained in
the 1983 survey (Fuller et a/ 1986), although
from a wider area. Direct comparisons of
changes in wader populations on the Uists
suggest that numbers of all species, with the
exception of Oystercatcher, have declined
since the 1983 survey (Whyte and O’Brien
1995). The previous estimates for wader
numbers on Shetland (Campbell 1989) appear
to be much lower than those from the present
survey. These estimates were made by
undertaking a preliminary survey of all areas
of inbye land on Shetland, repeat surveying
the best looking areas for waders, and applying
a conversion factor to the areas not repeat
surveyed, based on the difference in wader
numbers on the best areas between the first
and subsequentvisits. Whilst wader densities
on the best areas in Shetland are similar to, or
higher than, the densities recorded for the
present survey, with the exception of Snipe,
the extrapolation appears to have seriously
underestimated the importance of the other
areas, especially for Lapwing, Oystercatcher
and Curlew (Gill eta/1994). The new Breeding
Atlas suggests a marked increase in the
distribution of Redshank in Shetland (Gibbons
et al 1994). In addition, the abundance map

suggests that a high proportion of the tetrads
in those 10 km squares which have gained
Redshank are themselves occupied,
suggesting a considerable increase in
Redshank numbers in Shetland.

Acknowledgements

| would like to thank the SOC regional
organisers, BTO regional representatives,
RSPB and SNH regional conservation officers
for providing information on locations of key
sites in their areas, and for providing contacts
on local fieldworkers. | wish to particularly
thank the voluntary fieldworkers, too many to
mention, without whom the basic key site
data would not have been available. George
Dodds, Rob Yaxley and Paul Gill surveyed
the random sites in both years, and Craig
Whyte, Martin Gray, Arfon Williams, Dave
Hodson, Matt Self, Anesti Davenhill and Tim
Watling surveyed random sites in one of the
years. Many thanks also to Rhys Green for
providing advice on the design of the survey
and to Allan Brown and Sylvia Laing for
obtaining support from voluntary fieldworkers,
also to lan Bainbridge for encouragement
throughout the project and to lan Bainbridge,
Mark Avery, Stan da Prato and an anonymous
referee for commenting on, and improving

240 M O’Brien

SB 18(4)

drafts of this report. Finally, but most
importantly, to the many landowners and
farmers who allowed fieldworkers to survey
waders on their land. Funding for the project
was provided by the RSPB.

References

Barrett J & Barrett C 1984. Aspects of censusing
breeding Lapwings. Wader Study Group Bulletin
42:45-47.

Bibby C J, Burgess N J & Hill D A 1992. Bird
Census Techniques. Academic Press, London.
Cadbury C J, Green R E & Allport G 1987.
Redshanks and other breeding waders on British
saltmarshes. RSPB Conservation Review 1:37-
40.

Campbell L H 1989. The importance of breeding
waders on croft and farmland in Shetland. RSPB
Conservation Review 3:75-78.

Campbell L H, Bird D R, Suddaby D & Ellis P M
1988. Breeding waders on croft and farmland in
Shetland, 1987. Unpublished report, RSPB, Sandy.
Campbell L H, Christer W G, Stenning J M, Davey
PR & Meek E R 1988. Wetland and marginal
moorland in Orkney. An ornithological survey of
unprotected sites, 1986-87. Unpublished report,
RSPB, Sandy.

Dodds G, Yaxley R & O’Brien M 1995. Numbers of
breeding waders on lowland areas of the Outer
Hebrides, 1993. Unpublished report, RSPB Sandy.
Fuller R J, Reed T M, Buxton ND, Webb A, Williams
T D & Pienkowski M W 1986. Populations of
breeding waders Charadriiand their habitats on the
crofting lands of the Outer Hebrides, Scotland.
Biological Conservation 37:333-361.

Galbraith H, Furness R W & Fuller R J 1984.
Habitats and distribution of waders breeding on
Scottish agricultural land. Scottish Birds 13:98-
107.

Gibbons D W, Reid JB & Chapman RA 1993. The
New Atlas of Breeding Birds in Britain and Ireland:
1988-1991. T & AD Poyser, London.

Gill J P, Williams H A, O’Brien M & Ellis P E 1994.
Numbers of breeding waders on lowland areas of
Shetland, 1993. Unpublished report, RSPB, Sandy.
GreenMJ, HarrisAD & Williams!T 1994. Breeding
waders in Wales, 1993. Unpublished report, RSPB,
Sandy.

Green R E 1985. Estimating the abundance of
breeding Snipe. Bird Study 32:141-149.
GreenwoodJJD 1991. Estimating the total number
and its confidence limits. Appendix in Shrubb and
Lack, Eds. (1991).

Hotker H, DavidsonNC & Fleet DM 1991. Waders
breeding on wet grasslands. Wader Study Group
Bulletin, 61, Supplement April 1991.

Marchant J H, Hudson R, Carter S P & Whittington
P 1990. Population trends in British breeding birds.
British Trust for Ornithology, Tring.

Murray S, Hinds P, Davenhill A, Morris S & O’Brien
M 1995. Numbers of lowland breeding waders in
northern England, 1993. Unpublished report, RSPB,
Sandy.

O'Brien M G & Smith K W 1992. Changes in the
status of waders breeding on wet lowland grasslands
in England and Wales between 1982 and 1989.
Bird Study 89:165-176.

Partridge J K 1992. 1992 resurvey of Northern
Ireland breeding wader sites. Unpublished report,
RSPB, Sandy.

Partridge J K & Smith K W 1992. Breeding wader
populations in Northern Ireland, 1985-87. J/rish
Birds 4: 497-518.

Piersma T 1986. Breeding waders in Europe, a
review of population size estimates and a
bibliography of information sources. Wader Study
Group Bulletin 48: Supplement.

Reed T M, Barrett J, Barrett C, Hayhow S &
Minshull B 1985. Diurnal variability in the detection
of waders on their breeding grounds. Bird Study
32:71-74.

Shrubb M & Lack P C 1991. The numbers and
distribution of Lapwings V. vanellus nesting in
England and Wales in 1987. Bird Study 39:20-37.
Soil Survey of Scotland 1982a. Land Capability for
Agriculture 1: 25,000 maps. Macaulay Institute for
Soil Research, Aberdeen.

Soil Survey of Scotland 1982b. Soil and Land
Capability for Agriculture. Macaulay Institute for
Soil Research, Aberdeen.

Smith K W 1983. The status and distribution of
waders breeding on wet lowland grasslands in
England and Wales. Bird Study 30:177-192.
Thom V M 1986. Birds in Scotland. Scottish
Ornithologists’ Club, Calton, Edinburgh.

Whyte C, Gray M & O’Brien M 1995. Numbers of
breeding waders on lowland areas of Orkney, 1993.
Unpublished report, RSPB, Sandy.

Breeding waders in lowland Scotland 241

Whyte C & O’Brien 1995. Changesinthenumbers _ report, RSPB, Sandy.
of waders breeding on machair habitats on the

Uists and Benbecula, 1983 to 1995. Unpublished Mark O’Brien, RSPB, 17 Regent
Terrace,Edinburgh, EH7 5BH

Revised manuscript accepted May 1996

aS
WIE

Lapwing S C Porter

242 Scottish Birds (1996) 18: 242-246

SB 18(4)

A winter survey of Goldeneyes on the River Deveron, north

east Scotland

PETER J COSGROVE

A winter survey of Goldeneyes on the main stem of the River Deveron, in
north east Scotland, was carried out in mid January 1996. There were 301
Goldeneyes on the river, representing approximately 7 birds per 2 km of
river. About a third of the birds were adult males and the rest females or
yearlings. Goldeneyes were absent in the upper reaches and most
abundant in the lower to middle reaches of the river. The figure of 301
birds qualifies the River Deveron for consideration as a nationally important
site for the species, holding nearly 2% of the estimated population in Great
Britain. However, the national population has been estimated in the
absence of comprehensive river surveys and so the national figures need to
be reassessed to take account of all Goldeneye habitats.

Introduction

The aims of the study were to census the
winter population of Goldeneyes Bucephala
clangula on the main stem of the River
Deveron, from the sea at Banff to Cabrach,
compare the count with other relevant data
and to make recommendations for future
censusing of local Goldeneye populations.

Methods

The winter census of Goldeneyes was carried
out by means of a transect count along the
whole length (87 km) of the River Deveron.
The count was carried out in mid January,
over a period of 4 days (21-24 January,
1996), during a period of similarly mild weather
conditions to minimise the possible effects of
weather related movements of birds along
the river. Counts were usually made in the
morning but sometimes continued into the
afternoon because of the short day length.

The whole river was walked from Banff to
Cabrach, scanning thoroughly at every bend
to avoid flushing birds ahead of the field
worker. The survey was carried out by 2
teams of 2. Observers covered a stretch of
river by walking upstream to a predetermined
spot where the other observer had left a car
and began his/her walk. Whenever possible,
observers moved around birds so as not to
flush them. The location of each bird was
recorded from grid references on 1:50,000
maps. Goldeneyes were recorded and
categorised as either adult males or
‘brownheads’ ie adult females and first winter
birds, using plumage characteristics.

Goldeneye distribution was plotted with
respect to river elevation; the whole river
being subdivided into 10 km sections starting
from the mouth. The mean elevation was
estimated (m ASL) from 1:50,000 Ordnance
Survey maps.

1996

Results

Three hundred and one Goldeneyes were
counted on the main stem of the River Deveron
over the 4 day count, giving a density of
approximately 7 birds per 2 km of river.
Goldeneyes were absentin the upper reaches
of the river and most abundant in the lower
part of the river (Fig 1). There were
comparatively high densities of individuals in
the lower to middle reaches of the river until
the confluence of the River Isla, near Milltown
of Rothiemay 50 km upstream at 80 m ASL,
where there was asharp decline. There were
no birds recorded on the upper Deveron
between Inschtammack, near Huntly, and
Cabrach.

Goldeneyes on the R Deveron, north east Scotland 243

About one third (109) of the 301 Goldeneyes
were adult males and the rest ‘brownheads’.
The proportion of ‘brownheads to adult males
remained high in all but one 10 km section of
river (Figure 2). The 40-50 km (80 m ASL)
section held a greater proportion of adult
males than ‘brownheads’. Goldeneye were
recorded individually and in small flocks of up
to 15 birds. There appeared to be no particular
bias in the composition of the flocks; both
adult male and ‘brownhead’ only groups, as
well as mixed groups, were recorded. The
proportion of male to female Goldeneyes in
groups of different sizes differed little from the
expected sex ratio of the groups calculated
from the total population censused (Figure 3).

Figure 1. Goldeneye densities in relation to elevation on the
River Deveron in Jan 1996.

20

No’/ind/10km section of river
§

10

10 20 40 60

80 120 170 240 300

Elevation (m a.s.1.)

SB 18(4)

244 PJ Cosgrove

Figure 2. The proportion of adult male to 'brownhead' Goldeneye

seen on the River Deveron in Jan 1996.

brownheads'

males

VMUU12B.®—®~..,

uolpiodoig

Distance upstream from the sea (km)

Figure 3. Sex ratio within Goldeneye groups

”
a
=|
2
»
—
o
2
a
©
x<
oO
”
2
in}

ry
$3 &
o £
E 2
‘oS
ft ike
2 ©
BS
o“e
a &
N GZ
N Z

vt N oO ive)

sdnoib jo saquinyn

Wd

12 13 14 15

10. All

9
Group size (No’ of ind birds)

ee

1996

Goldeneyes on the R Deveron, north east Scotland 245

Discussion

Most Goldeneyes in Britain are reported from
either the coast or from still freshwater (Owen
et al 1986, Duncan and Marquiss 1993).
There have been very few estimates of whole
river populations of Goldeneyes; most have
been made on sections of rivers, such as on
parts of the Tweed inthe Borders. Estimates
of wintering populations of Goldeneye have
been compromised by a lack of systematic
counts on rivers. Duncan and Marquiss
(1993) documented the numbers of
Goldeneyes on rivers and standing waters in
northeast Scotland to show the importance of
rivers as wintering habitats. Their study,
however, did not cover the River Deveron and
there appear to be no other published data on
the importance of this river for wintering
Goldeneyes. Duncan and Marquiss (1993)
found astrong tendency fora higher proportion
of adult males at sites at lower elevations.
This was mainly due to differences between
habitats; the greatest proportion of adult
males occurred on the lower sections of rivers
and onthe sea. Attidal river mouth sections
and effluent discharges, food is concentrated,
and this attracts good numbers of Goldeneyes,
particularly males. The River Deveron is a
small, relatively clean Salmon Salmo salar
river where food is probably not very
concentrated. Consequently, the site is
attractive to females which do not have to
compete with the larger, adult males. It is
possible that the lack of a higher proportion of
adult males on the lowest section of river
could be attributed to lack of concentrated
feeding sites such as are found on some of
the other rivers in northeast of Scotland.
Goldeneye numbers in Britain remained stable
from the late 1970s to the mid 1980s, and
then gradually increased (Waters and
Cranswick 1993). Duncan and Marquiss
(1993) found high numbers of wintering
Goldeneyes on whole river counts in northeast
Scotland. From these counts, and others

they carried out on still waters in the area,
they estimated that 62% of the region’s
Goldeneyes were on rivers. Numbers varied
considerably between years but these
differences could not be attributed to weather
related movements of birds.

Duncan and Marquiss (1993) suggested that
the location of Britain’s wintering Goldeneye
populations had changed over time, from still
water sites and a very few coastal
concentrations, to still water sites and some
important river sites. This was associated
with reductions in effluents discharged into
shallow waters (eg Campbell 1984). The
figure of 301 Goldeneyes shows that the
River Deveron is a nationally important site,
holding nearly 2% of the estimated population
in Great Britain. The Deveron population,
along with those of the Rivers Don and Dee,
makes the north east of Scotland a nationally
important area for wintering Goldeneyes.
However, the national population has been
estimated without comprehensive river
surveys, so that the national figures and the
levels qualifying for national and international
significance need to be reassessed following
national surveys of all Goldeneye habitats.

The important Goldeneye rivers of northeast
Scotland are also important for Goosanders
Mergus merganser (Marquiss and Duncan
1994). These are shot under licence from the
Scottish Office Agriculture, Environment and
Fisheries Department (SOAEFD). In winter
the 2 duck species forage in the lower to
middle stretches of the river. Although
Goldeneyes are not shot under SOAEFD
licence, they do experience the disturbance
resulting from shooting Goosanders and
special consideration should be given to the
conservation status of co-located species
(MAFF 1995). It is not clear how this
disturbance affects Goldeneyes, but in the
present survey they were just as wary of

246 PJ Cosgrove

humans as Goosanders were on the Deveron.
Birds took flight at the sight of humans atleast
400 m away and moved long distances.

Other ducks, predominately Mallard Anas
platyrhynchos, but also Wigeon Anas
penelope, Teal Anas crecca and Mandarin
Aix galericulata, also use rivers in Britain, so
systematic whole river counts should be
undertaken along with counts on selected still
waters. It is only through more thorough
censusing that a complete picture can be
made of important wintering habitats not only
for Goldeneye but other waterbirds. The
Deveron is close to the main Scottish
Goldeneye breeding area of Speyside. It is
possible that, with increasing numbers of
birds wintering on north east rivers, numbers
subsequently summering, and potentially
breeding, will increase. | Comprehensive
monitoring of these sites throughout the year
will show whether these rivers also become
important breeding areas for Goldeneyes.

Acknowledgements

| am grateful to Scottish Natural Heritage who
provided funds to help finance the study.

SB 18(4)

Three other fieldworkers helped in the
collection of the data: Chas Emes, Ross
McGregor and Eleanor Shield. | would like to
thank Dr Mick Marquiss who gave helpful
advice and commented on the early draft of
this paper.

References

Campbell L H 1984. The impact of changes in
sewage treatment on sea ducks wintering in the
Firth of Forth, Scotland. Biological Conservation
28:173-180.

Duncan K & Marquiss M 1993. The sex/age ratio,
diving behaviour and habitat use of Goldeneye
Bucephala clangula_ wintering in north east
Scotland. Wildfowl 44111-120.

Marquiss M & Duncan K 1994. Seasonal switching
between habitats and changes in abundance of
Goosanders Mergus merganser within a Scottish
river system. Wildfowl 45:198-208.

Ministry of Agriculture, Fisheries and Food 1995.
Regional Service Centre Instructions: Licensing
procedures for the killing and taking of wild birds.
Owen M, Atkinson-Willes G L & Salmon D G 1986.
Wildfowl in Great Britain. Cambridge University
Press, Cambridge.

Waters R J & Cranswick P A 1993. The Wetland
Bird Survey 1992-93: Wildfowl and Wader Counts.
BTO/WWT/RSPB/JNCC, Slimbridge.

P J Cosgrove, University of Aberdeen, Department of Zoology,
Culterty Field Station, Newburgh, Ellon AB41 OAA

Revised manuscript accepted May 1996

Scottish Birds (1996) 18: 247-253

SHORT NOTES

Feeding association of Dunnock
and Blackbird in snow

In early February 1996, there was a fall of
about 10 cm of snow at Bridge of Allan,
Stirling. A picnic table outside my window
acted as a bird feeder. | cleared half the table
and left bread and apple scraps but these
were covered by more snow to a depth of 2 or
3cm. Several Blackbirds Turdus merula and
at least one Dunnock Accentor modularis
were accustomed to feed in the area; a
Blackbird would land on the table and start
clearing snow, using bill, and foot scratching,
so that food remnants were exposed. After
several days of casual observation, | realised
that the arrival of a Blackbird on the table was
often followed by the appearance of the
Dunnock which would follow the Blackbird
around and peck into the exposed surface.

247

Both birds often left together at some
disturbance. Although both species were
often within about 15 cm of each other, there
were no signs of antagonism between them.
A thaw occurred before | could make detailed
observations and the association ceased
immediately. There is no comment on this
behaviour in Birds of the Western Palearctic
and Bishton (1986, /bis 128:526-539) did not
mention any commensal interactions.
Simmons (1985 Brit Birds 78:508) noted a
Dunnock pecking at the faeces of a Blackbird
but this did not happen in my observations.
Several Robins Erithacus rubecula,
Chaffinches Fringilla coelebs, and Blue and
Great Tits Parus caeruleus and P.major also
visited the table regularly but not when
Blackbirds were present. Presumably the
behaviour recorded here only happens when
snow cover is too thick for Dunnocks to forage
but thin enough for Blackbirds to clear it.

C J Henty, University of Stirling, Stirling FK9 4LA

Accepted March 1996

248 Short Notes

SB 18(4)

Decline of Sutherland rookeries

In April 1996, DMB took part in the BTO
Rookery Survey. In his extensive allocation
of tetrads in central Sutherland, he found only
one rookery plus 2 others in adjacent areas.
He contacted DM who had carried out the last
count of Sutherland rookeries in 1975 (Castle
M. Scottish Birds 9:327-334). We decided to
carry out a repeat of the 1975 count, omitting
3 distant rookeries in the Tongue area of
north Sutherland, which have been omitted

from the following calculations. In 1975, 27
rookeries contained 1808 nests. In 1996
these were reduced to 23 rookeries and 1314
nests. Eight of the 1975 rookeries were either
abandoned or the trees had been cut down.
Four new rookeries were discovered. This
gives an overall reduction of 4 rookeries and
494 nests. We do not know whether the
decrease is still continuing and another count
isneededto answer this question. Incidentally,
the majority of the rookeries are sited along
the agricultural strip of the county’s east coast.

Donald M Bremner, 4 Drummuie Terrace, Golspie KW10 6FZ
Donnie Macdonald, Elmbank, Castle Street, Dornoch 1V25 3SN

Accepted May 1996

Probable first breeding record of
Brambling in Britain represented ina
museum collection

The first accepted breeding record for
Brambling Fringilla montifringilla in Britain
was in 1920, when a pair with a nest was
found at Altnaharra in Sutherland (Hodgkin &
Hodgkin 1920. The breeding of Brambling in
Scotland, Scottish Naturalist 1920 :181-182).
E T Booth claimed to have a found a nest with
3 eggs in Perthshire in 1866, but this has
never been accepted. Breeding was not
confirmed again until 1979, when a nest was
found in Grampian (Buckland & Knox 1980,
Brambling breeding in Scotland, British Birds
73:360-361) but, since then, 7 pairs have
been confirmed as breeding, with an additional
35 possible breeding records, by 1990 (C
Mead in Gibbons et a/ 1993, The New Atlas of
Breeding Birds in Britain and Ireland 1988-
1991. Poyser).

Major William Stirling of Fairburn (1858-1914)
put together a substantial egg collection
between 1895 and 1910, the eggs coming
mainly from his own estate in Easter Ross
and neighbouring districts. In 1983 Captain
Roderick Stirling, William Stirling’s grandson,
generously donated the substantial collection
of several thousand eggs, with good
supporting documentation, to Inverness
Museum and Art Gallery and HMcG worked
onavoluntary basis on this collection between
1992 and 1995. The collection contains the
first British clutch of Slavonian Grebe Podiceps
auritus eggs (McGhie 1994, Discovery of the
first British clutch of Slavonian Grebe eggs in
a museum collection, Scottish Birds 17:166-
167), and important series of Greenshank
Tringa nebularia, Siskin Carduelis spinusand
Scottish Crossbill Loxia scotica clutches.

|

1996

In the course of work on the collection we
found a record on the Brambling page in
William Stirling’s uncompleted manuscript
catalogue of the collection, with the following
information:

28 May 1899 HM
3 Orrin Side, Fairburn V

HM refers to one of the tenants of the area
between the Rivers Orrin and Conon. The
only Orrin Side in Fairnburn is Orrinside (NH
48 51), an area of Birch scrub 50 metres
above sea level which agrees well with
descriptions of Brambling nesting habitat (eg
Newton, Finches. Collins 1972). The date is
in accordance with the late May dates of the
1920 Sutherland nest andthe 1979 Grampian
nest. Brambling eggs are, unfortunately, very
similar to Chaffinch, Fringilla coelebs, and
there is broad overlap in size, colour and
markings. The 3 eggs, still present in the

Short Notes 249

collection and bearing the set mark ‘v’,
measure 21.25 x 15.15, 20.15 x 14.74 and
21.20 x 15.05mm (to nearest 0.05mm),
agreeing well with published measurements.
The background colour is greenish blue, with
none of the smudging typical of Chaffinch
eggs. The eggs are thus more typical of
Brambling than Chaffinch in colouration and
markings.

Given that the eggs were found in suitable
habitat, that colour and size are typical of
Brambling, and that William Stirling was a
very competent ornithologist, we believe that
this clutch represents the first recorded
breeding attempt by Brambling in Scotland.

We thank Captain Roderick Stirling for reading
a draft of this short note and for his donation
of his grandfather’s important collection to
Inverness Museum and Art Gallery.

Henry A McGhie, 8 Queen Street, Mytholmroyd, W. Yorks HX7 5HN
Stephen A Moran, Inverness Museum & Art Gallery,

Castle Wynd, Inverness 1V2 3ED

Revised manuscript accepted May 1996

Brambling

250 Short Notes

SB 18(4)

Mobbing of Waxwing by Chaffinches
and Blue Tits

On31 March 1996 our attention was attracted
by the alarm calls of Chaffinches Fringilla
coelebs and Blue Tits Parus caeruleus which
were perched in a young ash tree Fraxinus
excelsior next to a road. More birds were
perched in adjacent trees and a total of 30-35
Chaffinches, mostly males, and 5 Blue Tits
were counted. We noted that the topmost
bird in the tree with most fruits was a Waxwing
Bombycilla garrulus. Most of the alarm calls
came from the finches in this tree. Periodically,
the Waxwing made a flycatching flight,
returning either to the same tree or to one
nearby. The finches scattered noisily from
whichever tree the Waxwing entered. They
then gradually moved back, surrounded the
Waxwing and continued calling until the cycle

was repeated. This pattern continued for
about 6 cycles over a period of 10 min before
the Waxwing flew off to a tree about 100 m
away, followed by a few of the finches. The
finches and tits did not approach closer than
about 2 m to the Waxwing, which did not
appear to be perturbed by their activities.

Birds of the Western Palearctic (Vol 5) does
not mention Waxwings attracting the
attentions of finches and tits in this way,
although it does record the North American
Waxwing B. pallidiceps chasing tits Parus
and Bullfinch Pyrrhula pyrrhula. However,
we did not observe any aggressive behaviour
by the Waxwing towards the tits and finches
on this occasion.

JM & VA Harrison, 8 Ferryfield Drive, Connel, by Oban,Argyll PA371SP

Accepted May 1996

Waxwing 1st prize in 1996 SOC photographic
competition

Eric McCabe

1996

Nestling predation by Merlins

Merlins Falco columbarius usually chase and
kill flying prey in the breeding season (Newton
et al 1984, Bird Study 31:49-56). This is
borne out by 6 British studies where the bulk
of prey remains at nest sites have mostly
included flying prey. However, nestling
predation, which has been well documented,
could also be considered a common hunting
strategy then (for example Hard & Enemar
1980. Var Fagelvarld 39:25-34).

Short Notes 251

This seems a high delivery rate. It is over
twice as much as was brought to the nest
overall (1.39 deliveries/hour) and is 5 times
as much overall in the nestling stage as found
in Britain and Europe (Dickson 1995, Scottish
Birds 18:20-23) and in an urban population in
Canada (Sodhi et a/ 1992. Canadian Journal
of Zoology 7:1477-1483).

Nestling prey has also been recorded in
Galloway (Scottish Birds 2:245) but there
was little evidence that nestlings figured in
prey deliveries to nests to the extent recorded

Table 1 Peak individual hourly feeding rates (‘runs’) of Merlins in the nestling and
post fledging stages of the breeding cycle in Galloway, 1971-92

Year Nest Items Hours Times Equivalent to
Stage brought observations between prey

deliveries deliveries/h
(mins)

1971 N 1 0.5 - 2.0

1972 Pr 4 375 49, 41,6,53 1.06*

1973 PE 1 2.5 - 0.40

1974 N 2 4.0 95 0.50

1976 PF 2 125 80 1.30

1977 PF 3 4.0 63,80 0.75

1988 PF 2 2:25 42 0.80

1992 Er. 2 3.50 33 0.57

Notes

N = Nestling; PF = Post fledging

* includes 2 items in 6 minutes delivered by a male

Knapton & Sanderson (1985, Canadian Field
Naturalist 99:375-377) watched a Merlin’s
nest containing young in sub arctic Canada
and found that 3 small passerine nestlings
were brought to the nest in short periods of
time during a 2 hour watch (equivalent to 3.0
deliveries/hour) and similar ‘runs’ of prey
deliveries occurred on 4 other occasions.

in Canada. Peak individual hourly feeding
rates recorded in any one year in Galloway,
when fieldwork included the times when
nestling prey would be available to Merlins,
are shownin Table 1. From this itcan be seen
there were no ‘runs’ of prey deliveries except
2 items delivered in6 min on one occasion in
1972.

252 Short Notes

It has been suggested that nestling prey may
be more important than thought (Sperber &
Sperber 1963, Zool. Bidrag Uppsala 35:263-
268; Bibby 1987, Bird Study 34:64-70). This
has not been borne out by the 6 British studies
where the number of nestling prey remains
found at nest sites have only varied between

SB 18(4)
ce Aas SOOM

0% - 4.6% with a mean of 2.2%. My data
suggest that the majority view is correct and
that nestling prey do not have, overall, alarge
role to play in the food intake of breeding
Merlins. This agrees with Cramp & Simmons
(1980 Birds of the Western Palearctic, Vol. 2
Oxford 1980) assessment that raptors,
generally, seldom prey on nestlings.

R C Dickson, Lismore, New Luce, Newton Stewart,

Wigtownshire DG8 0AJ

Revised manuscript accepted May 1996

Jim Young

i
{

ee caecum

1996

Black-throated Diver feeding on
Common Frogs

On 3 May 1996, in Argyll, | watched a Black-
throated Diver Gavia arctica fishing for about
three quarters of an hour until it flew off. Inthe
49 min during which | watched the bird fishing,
it did so continuously, moving around the loch
in an anticlockwise direction. The loch is
about 450 mlong by 130 wide. | watched from
about midway along one of the long sides, so
that the diver was as close as 40 m and never
further than 200 m away. Between each dive,
it remained on the surface only long enough
to dispose of prey or take a breath before
diving again. | timed several of the periods on
the surface, and they averaged no more than
7 secs. The average time was not affected by
whether the dive had been successful or not.
When a dive had been successful, the diver
shook the prey in the water on surfacing and
then swallowed it. Sometimes it swallowed it
before raising its head, but usually it raised its
head so that the bill was above the horizontal
and then swallowed the prey with a gulp. This
allowed me to see what sort of prey had been
caught.

| observed 47 dives, of which 20 were
successful and one unknown. There were
probably a few more dives at the end of the
period when the diver moved into very dark
water at the far shore. The average length of
the dives was a maximum of 55 secs and a
minimum of 53 secs. Some successful dives
were very short, down to 10 secs. The
success of dives varied according to the
bird’s position in the loch. At the north end, 9
dives failed out of 10, while in the south end
17 failed out of 37.

All the prey items were of similar size. The
portion which hung from the closed bill was
about 60 to 90 mm long. Thus the prey itself

Short Notes 253

would have been from 70 to 100 mm long.
Some of the prey items appeared linear in
shape and could have been small fish, but
some most definitely had limbs, which flapped
when the prey was shaken. The majority of
such prey items came from a small area of the
loch at the south end, where the diver made
13 successful dives in succession.

On inspection, one 20 m stretch of mossy
bank opposite this area was seen to have
about 15 clusters of partly dead Frog spawn,
about 100 mm above the water. This
confirmed that the loch was occupied by
many Common Frogs Rana temporaria, and
it is almost certain that these made up the
majority of the prey taken by the diver. The
diver had been fishing for a long time before
| was in position and counting. It is thus likely
that this feeding session provided well above
the 20 prey items observed.

Cramp & Simmons 1977 (The Birds of the
Western Palearctic, Vol 1, Oxford) state that
Common Frogs are recorded as being taken
by both Black-throated and Red-throated
Gavia stellata Divers and give weights for
Black-throated Divers in summer of 3310-
3400 gms for males and 2037-2471 gms for
females. Frazer 1983 (Reptiles and
Amphibians in Britain, Collins New Naturalist)
gives sizes for Common Frog in Scotland
(snout to vent) of between 80 and 95 mm, and
weights of females ranging from 28 gr post
spawning to 72 gms, and 49 gms for a 4 year
old male. Taking an average of 38 gms, the
total weight of the observed prey would be
about 760 gms in my 49 min observation
period. This represents 34% of the average
female Black-throated Diver weight or 23% of
the average male weight.

TDH Merrie, Craigie House, Craigie,
Clunie, Blairgowrie, PH10 6RG

Revised manuscript accepted June 1996

254 Scottish Birds (1996) 18: 254-255

SB 18(4)

OBITUARY

Ritchie Seath
1907-1996

Ritchie Seath died on 20 May 1996. He was
89. At the family’s suggestion, friends and
relatives gave to the SOC instead of sending
flowers. As one of the founder members of
the SOC, inaugurated on 14 January 1937,
he had been proud to receive honorary life
membership and a commemorative scroll
only a few weeks before his death. Born in
Edinburgh on 3 April 1907, he lived over the
Forth at Aberdour from 1942. He was a
classical scholar at the Royal High School,
and graduated Bachelor of Commerce at
Edinburgh University in 1928. Among his
sporting activities, he was captain of rugby at
school, played for the university XV and
continued this interest as a referee in the
adult game. From university he went into the
family printing business, Howie & Seath,
founded by his father. He retired on his 70th
birthday. During World Warlihe also worked
for the Admiralty at Rosyth. This took him
around the country to such places as Scapa,
where he formed a close friendship with
George Arthur, master baker and guardian of
Orkney birds.

In 1935, he married Marguerite Edwards.
When she died in 1994 after along illness, he
lived on his own, alert and sharp of mind, if
physically slowed a bit by age, until a few
months before his short, final illness. He
leaves a daughter, a son, grandchildren and
great grandchildren.

Ritchie had a lifelong interest in birds. He
was knowledgeable and active. Especially in
his younger days he made many trips to the
Grampians, Sutherland, Wester Ross, Islay,

Colonsay, Tiree, Coll and other parts of
Scotland, but he did not travel abroad. He
was a keen hillwalker in search of birds.
There were particular friends he went with,
and later his family, but he was not widely
known to SOC members. In the 1930s he
contributed notes to the Scottish Naturaliston
Dotterel behaviour and terns and a paper on
birds round Kinloch Rannoch, but organised
birdwatching and recording were not his
scene. Like others of his generation he had
a meticulously prepared and documented
egg collection, but,when such things became
generally unacceptable, they were expunged.

His grand passion was fine bird books,
especially with spectacular hand coloured
plates. His house was an Aladdin’s cave to
the lucky enthusiast who got to see it; every
bookcase and cupboard revealed new
treasures, carefully acquired over the years:
Gould’s 5 volume Birds of Great Britain, his
Century of Birds from the Himalaya Mountains,
and a superb double volume of his
Monographs of the Toucans and the Trogons;
Dressers 9 volume Birds of Europe and
Monographs of the Bee-eaters and Rollers;
Jardine and Selby’s //lustrations of
Ornithology, and many more.

Then, in 1965, Howie & Seath had to leave its
premises in Swinton Row to make way for the
St James’s Centre. Rather than borrow to
finance the move he determined to sell his
natural history and classical travel books, and
would not be dissuaded from this. The
auction realised good prices at Hodgson’s in
London on 25-26 March 1965. With a grant
from the Russell Trust the SOC secured his
complete bound set of /bis for the library.
Ritchie was a member of the BOU from 1942
untilhe died. Afterthe sale he most generously
went on paying to bind the /bis annually forthe
SOC for the rest of his life.

1996

He was very pleased when the club recognised
his outstanding knowledge of bird books with
an invitation to become its honorary librarian.
He filled this role from 1965 to 1983 and then
continued on the library committee. As well
as taking a keen interest in the library and
advising the committee, he contributed a
series of reviews to Scottish Birds in the
1960s and 1970s, mainly of finely illustrated
bird books. His commentary revealed a
professional understanding of how varied
methods of book production affect public
perception of the qualities of the artist. Edwin
Alexander was a particular favourite forwhom
he expressed admiration, ranking him with
such masters as Joseph Wolf and Archibald
Thorburn.

The sale of his books was notably less
traumatic than expected. With the business
into its new premises, it was not long before
he was quietly and selectively picking up fine
books again, including some he had just sold.
He did not grieve for those he could not
replace. He had enjoyed owning them but did
not have to hoard them. In 1973, he wrote
that “you have to add something good
occasionally to keep up interest. Recently |
sold some travel books and bought
Monographs, even at recent prices, and forgot
the cost’. In the late 1980s, knowing his
family would be faced with selling his books,
and now over 80, he sold the valuable ones
from his second collection privately.

Ritchie Seath had an interested enquiring
mind, centred round birds but extending
widely. He appreciated excellence and took
pride in the quality of his printing work. He
was sensitive to the rampant printing errors
and loose ungrammatical writing that are now
common and he would mark neat printers’
corrections in the margins of his less valuable
books, in pencil of course.

Obituaries 255

He was rather a private person, as book
collectors may be; a bit of aloner and with no
wish to be a public figure. He did not havea
bookplate and disapproved of them. Buthe
was happy to talk about birds and books and
help an aspiring collector along the rightlines.
The writer treasures copies of Lord Lilford’s
Birds of the British Isles and Ewen Kennedy,
No 47 of 75 copies, illustrated by Edwin
Alexander, bought in the Seath sale, as well
as other volumes acquired with his guidance.
He was aman one was happy to have known.

Andrew T Macmillan

Ritchie Seath at 88 , with BOU tie, on the

family boat at Inverkip Mrs Thomson

256 Advice to Contributors

Advice to Contributors

Authors should bear in mind that only a small
proportion of the Scottish Birds readership are
scientists, and should aim to present their material
concisely, interestingly and clearly. Unfamiliar
technical terms and symbols should be avoided
wherever possible and, if deemed essential, should
be explained. Supporting statistics should be kept
to a minimum. All papers and Short Notes are
accepted on the understanding that they have not
been offered for publication elsewhere and that
they will be subject to editing. Papers will be
acknowledged on receipt and will be reviewed by at
least 2 members of the editorial panel and, in some
cases, by anindependentreferee. They willnormally
be published in order of acceptance of fully revised
manuscripts. The editor will be happy to advise
authors on the preparation of papers.

Reference should be made to the most recent
issues of Scottish Birds for guidance on style of
presentation, use of capitals, form of references,
etc. Papers should be typed on one side of the
paper only, double spaced and with wide
margins; 2 copies are required and the author
should also retain one. We are happy to accept
papers on Applemac computer discs. We cannot
handle other formats because both the SOC

SB 18(4)

computers and those at our printers are on the
Apple system. Please contact Sylvia Laing on
0131 556 6042 to discuss this. Headings should
not be underlined, nor typed entirely in capitals.
Scientific names in italics should follow the first text
reference to each species. Names of birds should
follow the official Scottish list (Scottish Birds Vol 17
: 146-159). Only single quotation marks should be
used throughout. Numbers should be written as
numerals except for one and the start of sentences.
Avoid hyphens except where essential eg in bird
names. Dates should be written........... on5 August
NOOSA... gees but on the 5th (if the name of the
month does not follow). Please note that papers
shorter than c700 words will be treated as Short
Notes, where all references should be incorporated
into the text, and not listed at the end, as in full
papers.

Tables, maps and diagrarns should be designed to
fit either a single column or the full page width.
Tables should be self explanatory and headings
should be keptas simple as possible, with footnotes
used to provide extra details where necessary.
Each table should be on a separate sheet. Maps
and diagrams should be in Indian ink and be
camera ready, but drawn so as to permit reduction
to half their original size.

Grey Heron

David Mitchell

NEOTROPICAL
BIRD CLUB

Be ropical el coll scmmaelhed

A club has been launched to promote the study and conservation of
the birds of the Neotropics (South America, Central America and the
Caribbean). It is currently seeking founder members to help reach
the launch budget of £2000, which is required to get the club running
and to publish the two first issues of its intended journal ‘Continga’.
Founder members will be asked to pay a minimum of £25, and will
be formally acknowledged in the first issue of ‘Continga’. ‘Continga’
will provide a colourful and much needed forum for exchange of
information on the avifauna of this extremely rich and diverse area,
and will contain papers and features on the birds and their
conservation as well as news of recent observations and discoveries
(at present, new species are still being discovered at the rate of
more than two a year). It is hoped that in due course the club will be
able to provide direct funding and support for practical conservation
programmes.

For further details and membership forms,
please contact:

Rob Williams,

Publicity Officer,

Neotropical Bird Club,

c/o The Lodge,

Sandy,

Bedfordshire SG19 2DL

Scottish Birds

Voiume 18 Part 4 December 1996

Contents

Main articles

Numbers of Whimbrel breeding in Shetland in 1989-1994 and previously
C P Dore, PM Ellis & E M Stuart 193
Breeding waders in the Cairngorms Straths ESA in 1995 WN Picozzi,
D C Catt & R P Cummins

197
Human induced increases of Carrion Crows and gulls on Cairngorms plateaux
Adam Watson 205
The Troup Head gannetry S Wanless, J Matthews & WR P Bourne 214
The feeding behaviour of Greylag and Pink-footed Geese around the
Moray Firth, 1992-93 / J Stenhouse 222
The numbers of breeding waders in lowland Scotland Mark O’Brien 231
A winter survey of Goldeneyes on the River Deveron, northeast Scotland
Peter J Cosgrove 242

Short Notes

Feeding association of Dunnock and Blackbird in snow CJ Henty 247
Decline of Sutherland rookeries DM Bremner & D Macdonald 248
Probable first breeding record of Brambling in Britain represented in a museum

collection H A McGhie & S A Moran 248
Mobbing of Waxwing by Chaffinches and Blue Tits J M & VA Harrison 250
Nestling predation by Merlins RC Dickson Zon
Black-throated Diver feeding on Common Frogs T DH Merrie 253

Obituary
Ritchie Seath Andrew T Macmillan 254

Advice to Contributors 256

Published by the Scottish Ornithologist’s Club
21 Regent Terrace, Edinburgh EH7 5BT.© 1996

‘\

APRIL 00

TR Do SA

iN

|

|

|

|

|

I

3 9088 00986 7797

It

Il

PE EM RN MF

ere re

se etea es

Raw fo se oe a

sees

cde Stage at

sera R i perere

Ces gp AmAN OE OE Sega SPE SUD,
esses

pete ey.

San MSE Baty CALE

ate ght AGAMA ITE ae they ap ar ares er
ven a7 peered

aN eke SRE INSEE EL Spee ca SA, pesacomeaet’

54 dee agtyh ene ves >
Se PR PR NG EA AIAD

ppverererrnverrorei erent aL ine) tert ae ee 2k eas eee ee en eee ee Ce a Si Co e

re cat wefan thd SS RSCAYRAFS pa Rye KGF* say 9 etna Uwe wh pete eden et tps RS hea ke ge RPMS CPN LAS E Oe Ww : °

Np de SRR A Aen gee pep ES Ye dg SAM Cee” ped adap ur? eQead ee ene PN Ee Pa Pre TT Se et) ee ee ee ee Ld mate ERE? CE Capi

. 4 x ROOF PA pay to Rue GOUe Ahh 7 op TONOUL US OP MiFR PL Araya Pras AOA RGSS OD eee ee Mig yp Ne

EN a Ree te PAL NE pee ten eS EP NY

Perr rs

WEES R OAS OS

Perey os Sg hiR LAO R = ‘
Ss he ae aes my EE VP SPS REE : ‘ Peer Set P epapaide Gn Sa E> ate RPA NA RSS AAPAR S “BD SANGRE AER LSN APS 94am AAU SK,
fi Rypre UNA Ste RTE NE SRE EN few - Onan - aout SATURN AIA
neki spas + Kepabe PUPP AREAS ~ oP P PLR arsth epee ices p AP AE TEA A STINE SEE TUF 105 hy RP EM OMA LA EMSS a pad aia Dyadstie SRR RGR
Sueanees Pe RANA Sop wins Wyeee SUPA Ue oa CPN ES RN he pert s rOd Mele 5 OEPAR YAW SWE PERSE UDCE VEY pR ARDY Aepsyee sth Qo SR RT RUBE ARITA NCTE TS
<a 3 i Lanta ci eyaiag AYR Pe py eA AN ae wea KE : “
LPR NST Ar SS ASA SOR ee eae LA MATE hte NPA Pie apy MOSERE MeO Phase pana seve ag 2
RES tes reid fo aie 2 aes gf Wipe ay te Ty ee en a te a ee bapa s nate tpi
eS artas pr ee ees wey wrt are ee , - bn ame peat ae tip ste et
. ey Teer es Apane ee St iol Ll pene : Tne WOT ENP An ae he 7,

AeSR ACRE SS

aRe

a hape!
REM sete Oy

SECU CLEARS OL

pee ke eRe epee eae ep Rtege

yeneary

ps4 an

Sofees

PUTA AAT EP RS
SUPER Senet

: pang mer

rebese eens
eae RNA ae eS

AS Pe OL te

oe ora Curae a estan sg ab

oh wat

PRN ee oD RSA e es