Content uploaded by Trevor H Worthy
Author content
All content in this area was uploaded by Trevor H Worthy on Mar 15, 2015
Content may be subject to copyright.
New Zealand Journal of Zoology, 1997, Vol. 24: 69-121
0301-4223/2401-069 $2.50/0 © The Royal Society of New Zealand 1997
69
A reappraisal of the late Quaternary fossil vertebrates of
Pyramid Valley Swamp, North Canterbury, New Zealand
R. N. HOLDAWAY
Department of Zoology
University of Canterbury
Private Bag 4800
Christchurch, New Zealand*
*Present address: Palaecol Research, 167 Springs
Road, Christchurch 8004, New Zealand.
T. H. WORTHY
Palaeofaunal Surveys
43 The Ridgeway
Nelson, New Zealand
Abstract The late Quaternary fossil vertebrate
fauna from deposits at Pyramid Valley, North Can-
terbury, New Zealand is reassessed. The faunal com-
position as contained in previous lists is updated, and
minimum numbers of individuals represented are
given. Measures of faunal diversity are presented and
compared with values for present New Zealand sys-
tems and other fossil sites. The revised faunal list in-
cludes a tuatara, one gecko, at least 46 species of
bird, and one species of bat. The avifauna was domi-
nated by moas and waterfowl. The fossil record was
biased towards large taxa because of the taphonomic
properties of the site. Deposition was slow: individu-
als were added a few per year and not as a result of
catastrophes. Bird diversity, by various measures,
was high for a New Zealand site. Diversity indices
and rarefraction curves suggest that a site, or group
of comparable smaller sites, must contain more than
300 individuals before adequate estimates of the
species richness of the sampled fauna are possible.
Ten diurnal vertebrate guilds are recognised, several
of which are now extinct; all guilds have lost at least
50%
of their former constituent species. Sample
sizes of two moa taxa were sufficient to allow
Received 9 January 1996; accepted 8 August 1996
preliminary analysis of age and sex ratios: in
Dinornis giganteus the sex ratio was skewed
strongly in favour of the putative female individu-
als;
in Emeus crassus the ratio was nearly equal.
Consequences of the differences are discussed. Evi-
dence is presented that Gallirallus minor, reported
from fossil deposits, was based on individuals of
Gallirallus australis at the lower end of the size
range. Evidence of predation of adult moas by
Haast's eagle is recorded, and the predator-prey ra-
tio for a system based on birds is given. The chro-
nology, environment, and components of
the
fauna
are discussed.
Keywords New Zealand; avifauna; fossil birds;
Pyramid Valley; diversity; systematics; paleo-
ecology
INTRODUCTION
The fossil site known as Pyramid Valley Swamp,
near Waikari in North Canterbury, New Zealand
(Fig. 1; NZMS 260 M33/772038; 42°58'26.8"S,
172°35'51.5"E), was brought to the attention of the
scientific community in the early 1940s (Allan 1941;
Duff
1941;
Falla 1941a, b; Percival 1941). The first
reports were based on examination of the strati-
graphy and the contents of
24
pits.
The pits were dug
at various points in what was then a swampy pad-
dock, where probing with steel rods indicated the
presence of bone. Work proceeded sporadically dur-
ing summer weekends from 1939 to 1941 (Duff
1949).
The early excavations revealed that Pyramid Val-
ley was especially important because the moa bones
were often preserved in discrete groups that usually
represented individual
birds.
In other swamp or lake
deposits bones from many individuals, and often
several taxa, were mixed together. The individual
skeletons from Pyramid Valley allowed the previ-
ous assignments of skeletal elements to species to
be checked, especially the association of crania with
postcranial elements.
70
New Zealand Journal of Zoology, 1997, Vol. 24
NEW ZEALAND
Honeycomb Hill
Cave
Fig. 1 Location of Pyramid Val-
ley, North Canterbury, and of Lake
Poukawa and Honeycomb Hill
Cave fossil sites.
44 S
172 E
173 E
Scattered bones and some part skeletons of many
other birds were found with the moa skeletons. As-
sociated microfossils allowed reconstructions of the
depositional and surrounding environments of the
lake.
At the time, however, there were no techniques
for dating the remains; radiocarbon dating was not
developed until the late 1940s.
Excavations at the site were curtailed by the Sec-
ond World War. Work did not resume until late
1947,
when a party from Canterbury University
College visited the swamp with
R.
Cushman Murphy
of the American Museum of Natural History
(AMNH). A larger excavation was arranged in early
1948 in association with Canterbury College, who
held the excavation rights, to obtain specimens for
AMNH and for Canterbury Museum (R. Cushman
Murphy, manuscript diary, and letters). Cushman
Murphy was also involved with the major excava-
tion programme that began in February and March
1949 (Duff 1949,1951,1955;Eyles 1955). This was
more systematic than the earlier digs. A series of
12 ft (3.66 m) squares was laid out and excavated,
mainly by J. R. Eyles and R. J. Scarlett. Each square
in turn was cleared to the sterile basal blue clay, a
process that took about 1 day (Eyles 1955). The
emphasis was, as before, on retrieval of moa skel-
etons,
but the grid allowed the position of each speci-
men to be recorded with some accuracy. The vertical
face of each succeeding square in the sequence was
excavated using shovels, and by hand. The spoil was
discarded into the trench behind the excavators. The
width of the squares increased the speed and
Holdaway
&
Worthy—Pyramid Valley fossil fauna
71
121
120
119
109
110
118
107
108
117 11«
94
93
B2
115
<
84
88 A82
65 [63
80
61
78
79
76
77
74
76
1113
73
71
100
96
69
96
96
67
99
97
65
101
103
62
63
107
108
60
•1
105
106
se
59
111
S6
57
90
64
55
89
52
S3
sa
60
61
89
112
B
48.7
\
XX
XIII
J
XXII
48.1
7
A
XXI -
XVI
^
XVII
,
/
VII
7
/
XVIIlTj
n
u
50 metres
Fig. 2 Excavation
at
Pyramid
Valley.
A,
Excavation
layout,
1949 and
subsequent.
B,
Pre-1949
excavations,
in relation
to later layout (after survey and site
plans,
courtesy
M.
&
J.
Hodgen). 50-122 are excavated squares; specimens were
usually catalogued with square
of
origin, sometimes depth
as
well. Roman numerals refer
to
individual skeletons
excavated in late 1930s and early
1940s.
Arabic numerals
(e.g.,
48.1) refer
to
pits opened in 1947 and
1948,
before the
systematic excavations
began.
Eyles
(195
5,
fig.
1) provided a
basemap
for the proposed excavation
grid,
and the
location
of previous pits.
convenience of excavation (Duff
1955;
Eyles 1955),
and Eyles (1955) estimated that one
moa
skeleton
took
an
average
of
about
75 min to
remove.
Al-
though
it
allowed
a
greater area
to be
excavated
in
the time available, the procedure would have made
it difficult
to
detect many
of
the small bones
in
the
sediments.
Further excavations
in
1952, 1954, 1955,
1956,
1957,
1963,
and 1965
(Moar 1970; Gregg
1972)
expanded the coverage
of
the
1949 grid.
A
mistake
in identifying
a
marker
peg
when
the
third line
of
squares was laid out led to the later series being
off-
set from the first (Fig. 2). The last excavation to date
was
in
December 1973, when
a
small area
40 m
south-west
of
the main grid was opened up
for the
5th INQUA Conference (Burrows
et
al. 1981).
Results
of
the excavations have been published
sporadically since 1949, when
a
popular account
(Duff
1949,
reprinted 1951) was produced. Few de-
tails have been published
on the
vertebrates,
although
the
site
is
mentioned
in
most accounts
of
New Zealand's extinct fauna. Scarlett (1955a)
re-
vised the initial list
of
birds
(Falla 1941b);
the
new
list included two new taxa,
an
extinct harrier hawk
(Scarlett 1953), and an extinct rail (Scarlett 1955b).
Scarlett worked
on the
fauna
for
many years;
his
most recent list (Scarlett 1969) included tuatara
Sphenodon
sp.,
short-tailed
bat
(Mystacina
sp.),
long-tailed
bat
{Chalinolobus sp.),
and a
fish
(pos-
sibly
an
eel Anguilla sp.),
but no
specimen details
were recorded.
Some of the names and taxa included by Scarlett
(1955a, 1969) have been changed as more informa-
tion and material has become available. For exam-
ple,
Olson (1975) reviewed the extinct rails of New
Zealand and transferred Rallus hodgeni to the genus
Gallinula, later emending
its
specific epithet
to
hodgenorum (Olson 1986). Olson (1977) also
de-
scribed
as a
new species specimens previously
at-
tributed
to the
Australian pink-eared duck
72
New Zealand Journal of Zoology, 1997, Vol. 24
Malacorhynchus
membranaceus.
Worthy (1995) has
redescribed the taxon using material located during
this study. The confusion between the large parrots
is discussed in Holdaway & Worthy (1993).
No complete summary of the vertebrate fossils
from Pyramid Valley has been published.
As
the type
locality for three species of extinct bird, as well as
several invertebrates, and as a classic site on which
to base reconstructions of the pre-human environ-
ment, Pyramid Valley is an historical nexus for Qua-
ternary studies. Although richer sites such as
Poukawa (Horn 1983) and Honeycomb Hill (Wor-
thy 1993a) have been discovered since, the verte-
brate fauna of Pyramid Valley is still pivotal to
studies of the extinct fauna.
The analysis presented here is not based on new
excavations. It is a reassessment and re-analysis of
the existing collections. The list of taxa and speci-
mens is the first consolidated and independently
verified list of the material from Pyramid Valley,
based on current understanding of New Zealand
Quaternary taxa. It forms part of a wider study of the
late Quaternary vertebrate faunas of the South Island
(Worthy & Holdaway 1993, 1994a, 1996a, b). Our
discussion of guilds in the extinct fauna and of ver-
tebrate diversity in the North Canterbury region
draws on data in Worthy & Holdaway (1996a),
where the Pyramid Valley fauna is briefly summa-
rised. The site is of such biological and historical
importance, as the basis of many previous discus-
sions of New Zealand's extinct fauna, that it is
treated in depth here.
Review of previous work
Duff
(1949,
reprinted 1951) gave a popular account
of the fauna. Inaccuracies and inconsistencies intro-
duced with this publication have persisted in the lit-
erature (e.g., Trotter & McCulloch 1984). Most
recent work involving the Pyramid Valley material
has relied on the published lists, using them as bases
for comparisons of regional faunas (e.g., Atkinson
& Millener 1991). The well preserved, associated
moa specimens have featured in taxonomic (Cracraft
1976a, b) and functional analyses (Atkinson &
Greenwood 1989). Plant remains preserved with the
skeletons were used in an analysis of the diets and
habits of major taxa (Burrows 1980a, b; 1989; Bur-
rows et al. 1981; Trotter & McCulloeh 1984;
Batcheler 1989), although much was destroyed in
error before it could be analysed (Burrows et al.
1984).
The paleoenvironment has received more pub-
lished attention than the vertebrates. Knowledge of
the Pyramid Valley environment during deposition
of the bones is based on analyses of a variety of
microfossils preserved with the macrofauna. Percival
(1941) discussed the implications of changes in the
ostracod fauna over time and collected a new spe-
cies of
Limnocy'there.
Hornibrook (1955) expanded
Percival's observations and summarised the diver-
sity of the ostracods. He described a new species of
Limnocythere, and discussed the environmental im-
plications of
the
ostracod fauna and changes in the
assemblages.
Deevey (1955) studied the sediments and
microfossils from cores collected by R. C. Murphy
in 1948. His interpretations of the changes in
cladoceran, ostracod, chironomid, mollusc, and dia-
tom assemblages suggested that the lake had passed
through "a normal lake ontogeny" (Deevey 1955).
He also discussed the depositional environments and
concluded that the Pyramid Valley lake had been
small and shallow throughout its history. Although
the first radiocarbon dates for the site had a very large
scatter and proved difficult to interpret, Deevey
(1955) was inclined to accept that moa were still
being trapped at a very late date (c. 670 years b.p.).
The vegetation surrounding the former lake has
been inferred from pollen samples (Harris 1955;
Moar 1970) and from plant macrofossils in the sedi-
ment (Burrows 1989) and in moa gizzards (Mason
in Falla 1941b; Mason in Gregg 1972; Burrows
1980a, b; Burrows et al. 1981). Harris (1955) pro-
duced a pollen profile of the upper levels of the de-
posit, and related the results to the pattern and
composition of the recent vegetation in Canterbury.
The pollen record was later extended to cover the
whole 5000-year duration of the deposit (Moar
1970).
The new profile was supported by more ra-
diocarbon dates. The new dates indicated that, al-
though the deposit spanned the late Holocene, it did
not include layers as recent as Deevey (1955) sug-
gested. Moar (1970) reported changes in the com-
position of the local flora that culminated in
deforestation by fire within the last 1000 years.
Further radiocarbon dates on moa bone and plant
material were published by Gregg (1966,1972), and
the site dates were summarised by McCulloch &
Trotter
(1979).
Gregg (1972) suggested that the dates
showed Pyramid Valley lake to have existed from
c. 4000 years to 2000 years b.p.
In the systematic investigations at Pyramid Val-
ley from 1949 onwards, 60 (of 71 marked out) 3.66 x
3.66 m squares (12 x 12 ft) and two half-squares (12
x 6 ft) were excavated to depths of
1.22—1.83
m
(980-1470 m
3
). A representative section (Sq. 82)
Holdaway & Worthy—Pyramid Valley fossil fauna
73
was presented by Eyles (1955, pi. 3). An additional
330^490 m
3
had been excavated before. In total,
between 1300 and 1960 m
3
of sediment were re-
moved to recover 183 moa skeletons or part skel-
etons and 2385 bones of other birds. Therefore, one
moa was removed for every 7.14—10.71 m
3
of sedi-
ment, and 0.55—0.82 m
3
of sediment for every bone
of other species. The rate of return for other birds
was probably much lower than this crude estimate
suggests, because many elements were found as as-
sociated skeletons.
The collection
The material from the first excavations was identi-
fied, in the main, by R. A. Falla. Later R. S. Duff
and others, such as W. R. B. Oliver (for the moas),
also identified material. The bulk of the collection
was,
however, identified by R. J. Scarlett, who also
catalogued it. The present analysis relies largely on
the excavations and cataloguing done by Scarlett
from 1947 to 1965. Unfortunately time, and addi-
tions to and rehousing of the collection, have resulted
in some of the collections becoming misplaced. This
material will remain unavailable for study until the
whole collection has been checked, sorted, validated,
and rehoused, a task that is now under way.
The 'lost' material accounts for only a very small
part of the original Pyramid Valley collection. A
more serious problem for workers using the cata-
logue and collection at present is that the early iden-
tifications were done without access to a
comprehensive collection of comparative material.
Consequently the normal size range for elements of
some extant taxa could not be appreciated. Speci-
mens that were larger or smaller than a perceived
"norm" were sometimes assigned to new taxa that
cannot now be supported. The problem was, of
course, even more serious for extinct species, for
which often only a few bones were held.
Lack of access to adequate comparative material,
and failure to recognise diagnostic morphological
and morphometric characters, resulted in a signifi-
cant level of misidentification. The Pyramid Valley
list has been "sanitised" as far as possible within the
bounds of current knowledge.
MATERIALS AND METHODS
Abbreviations: AIM, Auckland Institute and Mu-
seum; AMNH, American Museum of Natural His-
tory, New York; CM, Canterbury Museum,
Christchurch; MNZ, Museum of New Zealand Te
Papa Tongarewa, Wellington.
Almost all the material collected during the many
excavations at Pyramid Valley is held at the Canter-
bury Museum, Christchurch. Several moas, one kiwi
(Apteryx sp.), and probably one adzebill (Aptornis
sp.) have been exchanged with or donated to other
institutions. In addition six partial moa skeletons,
other moa bones, and an adzebill were acquired by
AMNH
as
part of an agreement negotiated before the
1947-48 excavations (AMNH accession cards).
Collections from Pyramid Valley for radiocarbon
dating and other studies have been registered in the
Fossil Record system of the Geological Society of
New Zealand. The numbers and details are as fol-
lows:
NZMS260: M33/fll-fl7 (all Sq. 85,
microfauna, material held by Institute of Geologi-
cal
&
Nuclear Sciences, Lower
Hurt).
NZMS 1: S61 /
f642 seeds, bark, twigs, Sq. 119, 680-730 mm be-
low surface,
D.
R. Gregg,
25.1.1965;
S61/f643 seeds,
bark, twigs, Sq. 119, 810-860 mm, D. R. Gregg,
25.1.1965;
S61/f644 seeds, bark, twigs, 970-
990 mm, Sq. 119, D. R. Gregg, 26.1.1965; S61/f645
seeds,
bark, twigs, 1160-1230 mm, Sq. 119, D. R.
Gregg, 26.1.1965; S61/f646 seeds, bark, twigs,
1340-1390 mm, Sq. 119, D. R. Gregg, 26.1.1965;
S61/f647 lower peat, 1600-1700 mm, Sq.
119,
D. R.
Gregg, 26.1.1965; S61/f648 moa gizzard {Dinornis
121D?), 1300 mm, Sq. 121,2.2.1965;S61/f649 skel-
eton and gizzard Emeus
crassus
121B, Sq.
121,
Can-
terbury Museum party, -.1.1965; S61/f650
gastropods, 9 ft, auger hole in saddle south-west of
swamp, D. R. Gregg, J. E. Cox, 21.2.1965 (samples
to N. Moar and E. Deevey); S61/f651 moa gizzard
{Eury apteryx geranoides 121
A),
Sq. 121, Canter-
bury Museum party, -.1.1965; S61/f655 moa skel-
eton and gizzard {Dinornis), Sq. 122, gizzard held
by C. J. Burrows, INQUA excavation, -.12.1973;
S61/f656 moa skeleton and gizzard {Dinornis),
INQUA excavation, -.12.1973.
All
bone material from Pyramid Valley that could
be found in New Zealand (Canterbury Museum,
Auckland Institute and Museum, and Museum of
New Zealand Te Papa Tongarewa collections) and
at AMNH (THW) was checked and re-identified if
necessary. A full list is given in Appendix 1. Mate-
rial recorded in the museum register but not located
during this study is listed as "not found".
Material was identified by comparison with re-
cent specimens or, for extinct taxa, with voucher
specimens in the collection at Canterbury Museum.
Moa taxa, anatids, the owlet-nightjar {Aegotheles
74
New Zealand Journal of Zoology, 1997, Vol. 24
novaezealandiae), parakeets (Cyanoramphus spp.),
and extinct gruiforms were checked by
THW;
Nestor
parrots and passerines were examined jointly; all
remaining material was checked by RNH. All bone
measurements were made with vernier calipers to
0.01 mm and rounded to the nearest
0.1
mm. Meas-
urements of moa bones were made to the nearest
millimetre as described in Worthy (1987b).
Minimum numbers of individuals (MNI) were
computed by counting the number of ipsilateral rep-
resentatives of the most abundant element, except
where otherwise noted. The potential mobility of
material within the site precluded association of
miscellaneous material for moas with particular par-
tial skeletons. The minimum numbers for moa taxa
were taken as the largest number of directly associ-
ated material sets. Movement of smaller
bones
seems
to have been common; minimum numbers for most
smaller taxa were assessed as if each bone was an
individual sample, except where a single individual
was obviously involved, when the final MNI was
adjusted accordingly. Adult and juvenile material
was summed separately.
The higher classification used is that of Benton
(1993).
The avian systematic list is based on Sibley
& Ahlquist (1990), as modified in Holdaway (1988,
1991 a)
and further discussed in Worthy & Holdaway
(1993,
1994b). In addition, we follow Atkinson &
Millener
(1991),
Turbott
(1990),
and Millener (1991)
in not recognising
Gallirallus
minor
Hamilton,
1893.
We present mensural data to support its synonymy
with Gallirallus australis, as suggested by Olson
(1975).
We do, however, recognise the North Island
and South Island takahe as distinct species,
Porphyrio mantelli and
P.
hochstetteri, respectively,
contra the current checklist (Turbott 1990).
The phrase "in the vicinity of Pyramid Valley"
or similar, used here, implies a time frame of
2000-
4000 years
b.p.
(the period of deposition at the site),
unless otherwise stated. When the deposit was found,
the lagoon had been drained and the site was a
swamp. References to the excavation and in the lit-
erature usually refer to Pyramid Valley Swamp or
the swamp deposits. There is evidence that much of
the deposition took place when the depression was
filled by a more or less permanent pond or lake.
References here to the site at the time of deposition
therefore normally refer to it as a lake. The lake al-
most certainly dried out in hot summers, at which
time deposition would occur in a swamp environ-
ment. The distinction is probably irrelevant to the
broad picture presented here, except insofar as it
affected the bogging of large birds. Later in the his-
tory of the deposit, some individuals undoubtedly
broke through the later peat surface into the boggy
lacustrine sediments below.
RESULTS
General
The Pyramid Valley fossil avifauna included a mini-
mum of 46 species (Table 1), representing 23 fami-
lies (Fig. 3). The estimate of species number is
conservative because species of parakeet and kiwi
are difficult or impossible to differentiate with
present knowledge (Worthy & Holdaway 1994b;
authors' unpubl.
data).
Therefore, it is uncertain how
many of each are represented in the collection. The
taxa in the fossil fauna are listed below, with number
of specimens and MNI (summarised in Table
1).
The
fauna as known at present is compared with inter-
pretations summarised in three earlier lists in Ta-
ble 2.
Diversity indices
The Pyramid Valley fauna was similar in species
richness to those of other major sites such as Lake
Poukawa (Holocene) and Honeycomb Hill Cave
(Otiran Glaciation to Holocene), but it differed in
species composition (Table 3). The differences were
related to the location (South Island), immediate
environment (a small, isolated lake among forested
hills),
and age (late Holocene).
The fossil fauna recovered from Pyramid Valley
was biased towards large taxa, most of which were
flightless. The principal exceptions—the New Zea-
land pigeon Hemiphaga novaeseelandiae and the
large parrots—are among the largest volant
taxa.
The
pigeon was the most abundant non-moa taxon. One
of the largest taxa (Dinornis struthoides) and the
smallest, such as the small passerines, were repre-
sented by one specimen or just a few.
Bird diversity, as measured by species richness
and by Shannon's Index, Hill's numbers, and
rarefraction curves (Ludwig & Reynolds 1988) was
high (Table 4; Fig. 4). The most species-rich groups
were the passerines (12 species), waterfowl
(8),
moas
(5),
parrots (4 or 5), and rails (4). The rarefraction
curves also show that sites with fewer than about 300
individuals represented give significant underesti-
mates of the fauna around the site during deposition
(Fig. 4A,
B).
For all three major sites for which data
are given, the rarefraction curves do not begin to
level out before sample sizes of about 300 are
reached. At 50 individuals (a normal sample size)
Holdaway & Worthy—Pyramid Valley fossil fauna 75
Table 1 Numbers of elements and minimum number of individuals of adults and subadults (or juveniles) of taxa in
Pyramid Valley fossil avifauna.
%1,
percentage of total individuals of all taxa; %2, percentage of total individuals
excluding moa sample. Includes only material to taxon. See text (p. 96) for explanation of number of eagles (four used
in percent calculations).
Taxon
Apteiyx sp. cf. australis
Pachyornis elephantopus
Emeus crassus
Euryapteryx geranoides
Dinornis struthoides
Dinornis giganteus
Coturnix novaezelandiae
Cnemiornis calcitrans
Euryanas finschi
Tadorna variegata
Malacorhynchus scarletti
Anas chlorotis
Anas gracilis
Anas superciliosa
Aythya novaeseelandiae
Cyanoramphus sp.
Nestor meridionalis
Nestor notabilis
Strigops habroptilus
Ninox novaeseelandiae
Sceloglaux albifacies
Aegotheles novaezealandiae
Hemiphaga novaeseelandiae
Gallirallus australis
Porphyrio hochstetteri
Fulica prisca
Gallinula hodgenorum
Aptornis defossor
Himantopus novaezelandiae
Circus eylesi
Harpagornis moorei
Falco novaeseelandiae
Poliocephalus rufopectus
Egretta alba
Xenicus sp.
Traversia lyalli
Anthornis melanura
Adult
25
1
27
19
31
21
87
2
18
17
30
137
89
22
2
18
14
633
177++
7
37
29
517++
13
147+
22
6
5
1
1
1
3
Prosthemadera novaeseelandiae45
Petroica australis
Petroica macrocephala
Mohoua novaeseelandiae
Corvus moriorum
Callaeas cinerea
Philesturnus carunculatus
Turnagra capensis
Bowdleria punctata
Number of taxa
Total
Mean/taxon
SD
olj
mean
8
1
3
5
48
9
11
3
41
2292
55.90
126.52
19.76
No.
of elements
Subadult
3
3
2
9
18
2
2
8
1
1
1
1
10
3
4
7
6
1
2
4
4
1
22
92
4.18
4.13
0.88
Total
28
1
27
22
33
30
105
4
20
25
30
138
89
23
2
19
15
643
180++
11
44
35
517++
14
149+
22
6
9
1
1
1
3
49
8
1
3
5
49
9
11
3
41
2385
58.17
127.56
19.92
%
1.17
0.04
1.13
0.92
1.38
1.26
4.40
0.17
0.84
1.05
1.26
5.79
3.73
0.96
0.08
0.80
0.63
26.96
7.55
0.46
1.84
1.47
21.68
0.59
6.25
0.92
0.25
0.38
0.04
0.04
0.04
0.13
2.05
0.34
0.04
0.13
0.21
2.05
0.38
0.46
0.13
Minimum no. of individuals
Adult Subadult
3
1
1
4
4
4
8
1
2
2
7
11
9
4
1
3
4
67
16
2
5
7
10
1
5
2
1
1
1
1
1
2
9
3
1
2
1
7
2
2
2
41
220
5.37
10.44
1.63
1
1
1
3
3
2
1
3
1
1
1
1
2
1
1
3
2
1
2
1
2
1
22
38
1.59
0.80
0.17
Total
4
17
81
21
1
63
1
1
5
5
7
11
3
3
5
7
12
9
5
1
4
5
69
17
3
8
9
10
2
7
2(4)
1
2
1
1
1
2
11
3
1
2
1
8
2
2
2
46
440
9.57
17.17
2.532
%1
0.91
3.88
18.26
4.79
0.23
14.16
0.23
0.23
1.14
1.14
1.60
2.51
0.68
0.68
1.14
1.60
2.74
2.05
1.14
0.23
0.91
1.14
15.75
3.88
0.68
1.83
2.05
2.28
0.46
1.60
0.91
0.23
0.46
0.23
0.23
0.23
0.46
2.51
0.69
0.23
0.46
0.23
1.83
0.46
0.46
0.46
%2
1.56
0.39
0.39
1.95
1.95
2.72
4.28
1.17
1.17
1.95
2.72
4.67
3.50
1.95
0.39
1.56
1.95
26.85
6.61
1.17
3.11
3.50
3.89
0.78
2.72
1.56
0.39
0.78
0.39
0.39
0.39
0.78
4.28
1.17
0.39
0.78
0.39
3.11
0.78
0.78
0.78
76
New Zealand Journal of
Zoology,
1997, Vol. 24
Table 2 Comparison of species lists of birds for Pyramid Valley.
Falla (1941)
Scarlett (1955a) Scarlett (1969) This paper
Apteryx sp. cf. australis
Dinornis maximus
1
Dinornis robustus
1
Pachyornis elephantopus
Emeus crassus
Emeus huttoni
2
Euryapteryx gravis
1
Cnemiornis calcitrans
Casarca sp.
1
Malacorhynchus sp.
Anas chlorotis
Anas sp.
Anas superciliosa
Fuligula sp.
1
Nestor meridionalis
Hemiphaga sp.
Gallirallus sp.
Rallus spp.
Fulica sp.
Aptornis defossor
1
Prosthemadera
novaeseelandiae
Corvus sp.
Calleas sp. (sic)
Apteryx sp. cf. australis
[Dinornis maximus]'
[Dinornis robustus]'
Dinornis torosus
1
[Pachyornis elephantopus]
[Emeus crassus]
[Emeus huttoni]
2
[Euryapteryx gravis]'
Cnemiornis calcitrans
Euryanas jinschi
Tadorna sp.
Malacorhynchus sp.
Anas castanea
1
Anas gibberifrons
i
Anas superciliosa
Ay thy a sp.
Cyanoramphus
novaeseelandiae
Cyanoramphus sp.
Nestor sp.
Nestor meridionalis
Strigops habroptilus
Ninox novaeseelandiae
Sceloglaux albifacies
Hemiphaga novaeseelandiae
Gallirallus sp.
Gallirallus
minor
2
Notornis mantelli
1
Palaeolimnas
chathamensis^
Rallus hodgenfi
Aptornis otidiformis
Himantopus himantopus
Circus eylesi
Circus approximans
Harpagornis moorei
Falco novaeseelandiae
Podiceps
rufopectus^
Prosthemadera
novaeseelandiae
Petroica (Miro) australis
Petroica sp.
Palaeocorax moriorum}
Callaeas cinerea
Philesturnus carunculatus
Turnagra capensis
Apteryx sp. cf. australis
Dinornis maximus
1
Dinornis
robustus^
Dinornis torosus
1
Pachyornis elephantopus
Emeus crassus
Emeus huttoni
2
Euryapteryx gravis
1
Coturnix novaezelandiae
Cnemiornis calcitrans
Euryanas finschi
Tadorna variegata
Malacorhynchus sp. cf.
membranaceus
Anas chlorotis
Anas gibberifrons
1
Anas superciliosa
Ay thy a sp.
Cyanoramphus
novaeseelandiae (sic)
Cyanoramphus sp.
Nestor sp.
Nestor meridionalis
Strigops habroptilus
Ninox novaeseelandiae
Sceloglaux albifacies
Hemiphaga novaeseelandiae
Gallirallus australis
Gallirallus
minor
2
Notornis mantelli
1
Nesophalaris chathamensis
1
Pyramida hodgenfi
Aptornis otidiformis
Himantopus himantopus
Circus eylesi
Circus approximans
Harpagornis moorei
Falco novaeseelandiae
Podiceps
rufopectus*
Prosthemadera
novaeseelandiae
Petroica (Miro) australis
Petroica sp.
Palaeocorax
moriorum^
Calleas cinerea (sic)
Philesturnus carunculatus
Turnagra capensis
Apteryx sp. cf. australis
Dinornis giganteus
Dinornis struthoides
Pachyornis elephantopus
Emeus crassus
Euryapteryx geranoides
Coturnix novaezelandiae
Cnemiornis calcitrans
Euryanas finschi
Tadorna variegata
Malacorhynchus scarletti
Anas chlorotis
Anas gracilis
Anas superciliosa
Aythya novaeseelandiae
Cyanoramphus sp.
Cyanoramphus sp.
Nestor meridionalis
Nestor notabilis
Strigops habroptilus
Ninox novaeseelandiae
Sceloglaux albifacies
Aegotheles
novaezealandiae*
3
Hemiphaga novaeseelandiae
Gallirallus australis
Porphyrio hochstetteri
Fulica prisca
Gallinula hodgenorum
Aptornis defossor
Himantopus novaezelandiae
Circus eylesi
Harpagornis moorei
Falco novaeseelandiae
Poliocephalus rufopectus
Egretta alba*
Xenicus sp.
Traversia lyalli
Anthornis melanura*
Prosthemadera
novaeseelandiae
Petroica australis
Petroica macrocephala
Mohoua novaeseelandiae
Corvus moriorum
Callaeas cinerea
Philesturnus carunculatus
Turnagra capensis
Bowdleria punctata*
*Labelled Pyramid Valley in collection, but not included in previous lists; 'name changes;
2
taxon no longer
recognised; ^locality listed in Scarlett (1968) and Rich & Scarlett (1977). [ ], not formally listed, noted as same as
previous list.
Holdaway & Worthy—Pyramid Valley fossil fauna
77
the curves are still rising steeply. In the absence of
a large site such as Pyramid Valley, a series of less
rich sites with different taphonomies is necessary to
give a satisfactory estimate of regional vertebrate
diversity.
The rarefraction curve was steeper for Pyramid
Valley than for the other sites above a sample size
of
50
(Fig. 4B) because more taxa were represented
by a few individuals. Poukawa, with its richer water-
bird fauna, had a high species richness, but the abun-
dance of some taxa meant that the curve was less
steep.
The faunas from Honeycomb Hill Cave (Ea-
gle's Roost and Graveyard) were depauperate in
waterfowl. All sites had similar increments of taxa
per number of individuals at lower sample sizes, so
it was not possible to predict total species richness
for a fauna at sample sizes below about 50—100 in-
dividuals.
The avifauna of Pyramid Valley was dominated
by a few abundant taxa, as shown by the steep ini-
tial slope of a plot of log relative abundance against
taxon rank (Fig. 5). Even with moas excluded, other
large taxa (such as pigeons) were disproportionately
represented, although it is not known whether that
is an artefact of
size
or of
local
abundance resulting
from some other factor such as presence of suitable
food supply. Taxa in Hawke's Cave, a predator de-
posit (Worthy & Holdaway 1994b, 1996b), were
more evenly represented, as they were in lake
deposits at Poukawa and in the accumulations at
Honeycomb Hill Cave. The unevenness of
the
rep-
resentation at Pyramid Valley may reflect significant
size biasses related to conspicuousness during exca-
vation as much as tendency for inclusion and pres-
ervation in the deposit.
Species not present
In a rich site it might be expected that most, if not
all,
species present in the area should have been rep-
resented in the fossil fauna. Species not recorded
might therefore indicate aspects of the local environ-
ment not apparent from the least of those present.
Forest, shrubland, and freshwater birds found in
the South Island at the time of human colonisation
that were not identified in the collection included
Apteryx haastii, Podiceps cristatus, Phalacrocorax
melanoleucos, Phalacrocorax
carbo,
Pelecanus sp.,
Botaurus poiciloptilus, Ixobrychus minutus, Cygnus
sumnerensis, Anas rhynchotis, Biziura delautouri,
Circus approximans, Gallirallus philippensis,
Porzana pusilla, Porzana tabuensis, Porphyrio
melanotus, Charadrius bicinctus, Thinornis novae-
seelandiae, Anarhynchus frontalis, Coenocorypha
cf. aucklandica, Himantopus himantopus, Larus
bulleri, Larus dominicanus, Sterna albostriqta,
Sterna caspia, Chrysococcyx lucidus, Eudynamys
taitensis, Halcyon sancta, Acanthisitta chloris,
Pachyplichas yaldwyni, Dendroscansor decurviro-
stris, Anthus novaeseelandiae, Mohoua ochro-
cephala, and Rhipidura fuliginosa (Potts 1882;
Oliver 1955; Worthy & Holdaway 1995).
Guilds
Of the 11 diurnal vertebrate guilds recognised at
Takaka Hill (Holdaway & Worthy 1996), 10 were
represented at Pyramid Valley (Table 5), and an-
other—aerial insectivore—was represented in other
sites nearby by the fantail Rhipidura fuliginosa. Four
of the five nocturnal or nocturnal/crepuscular guilds
had representatives at Pyramid Valley (Table
5).
The
nocturnal aerial insectivore niche was almost cer-
tainly filled by the long-tailed bat {Chalinolobus
tuberculatus) as at Takaka Hill, but none of the sites
or taphonomies in the area was suitable for the en-
trapment or preservation of this species. The guilds
include many taxa that are now locally or globally
threatened or extinct, the most depleted being the
terrestrial herbivores and omnivores and the preda-
tors (Table 5). The extreme depletion of the local
vertebrate fauna is illustrated by the percentage loss
of taxa from each guild; losses ranged from 50 to
100%
(Table 6).
Rate of entrapment
The area excavated so far at Pyramid Valley is
c. 7.5% of the area of the whole swamp. Duff (1955)
estimated that the density of moas was 750-800 per
acre (c. 1800/ha) in the area of swamp excavated
until 1952. Moas are unlikely to have been evenly
distributed throughout the deposit, and some sections
may be barren (the area excavated was thought to
be the richest, as tested by probing). For the
c. 1087 m
2
excavated so far, the 183 individual moas
represent an absolute maximum of
1683
birds for the
1.5 ha deposit.
Burrows (1989) suggested that bird remains were
added
to
the lake deposit for
a
period of perhaps 1300
years.
It
is
possible, though unlikely, that the site may
have trapped
birds
over
a
much shorter
time
than that
occupied by the lake phase at the site
(c.
4000-2000
years ago). The gyttja sediment accumulated at
c. 0.45 mm/yr (Gregg 1972) and would have taken
some time to develop to a depth sufficient to effec-
tively trap moas. Burrows (1989) pointed out that
most of the
14
C dates from the lake deposits cover a
period of 250 radiocarbon years, between 3750 and
78
New Zealand Journal of Zoology, 1997, Vol. 24
Table 3 Comparison of species composition of fossil avifaunas from Pyramid Valley, Lake Poukawa (layer 3), and
Honeycomb Hill Cave (HHC) (Graveyard layers 1 & 2 and Eagle's Roost). N, North Island; S, South Island; ?,
presence not confirmed, or numbers not given; (), represented by different species in North Island, but the present
checklist maintains only subspecific difference (Turbott 1990); -, indicates presence in the island fauna but absence
from the deposit; blank indicates absent from that island. Data for Lake Poukawa from Horn (1983) and Worthy &
Holdaway (1993); for Honeycomb Hill Cave- Eagle's Roost, from Worthy & Mildenhall (1989).
Taxon I
Apteryx sp. cf. australis
Anomalopteryx didiformis
Megalapteryx didinus
Pachyornis elephantopus
Pachyornis mappini
Emeus crassus
Euryapteryx geranoides
Euryapteiyx curtus
Dinornis struthoides
Dinornis novaezealandiae
Dinornis giganteus
Coturnix novaezelandiae
Cnemiornis calcitrans
Cnemiornis gracilis
Cygnus sumnerensis
Euryanas finschi
Tadorna variegata
Anas chlorotis
Anas gracilis
Anas superciliosa
Anas rhynchotis
Hymenolaimus
malacorhynchos
Aythya novaeseelandiae
Biziura delautouri
Malacorhynchus scarletti
Cyanoramphus sp.
Nestor meridionalis
Nestor notabilis
Strigops habroptilus
Ninox novaeseelandiae
Sceloglaux albifacies
sland
NS
NS
S
s
N
S
NS
N
NS
NS
NS
NS
S
N
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
N?S
NS
NS
N?S
NS
NS
NS
Aegotheles novaezealandiaeHS
Hemiphaga novaeseelandiae
Gallirallus australis
Porphyrio hochstetteri
Fulica prisca
•NS
NS
(N)S
NS
Minimum number of
individuals in sample
Pyramid Lake
Valley Poukawa
4
—
-
17
80
21
1
_
62
1
1
-
5
5
11
3
3
—
—
5
_
7
7
12
9
8
1
4
5
69
17
3
8
10
14
-
52
—
3
6
3
3
2
-
17
17
2
125
91*
107
12*
_
53
2
_
7
18
_
29
2
2
—
16
76
26
4
HHC
4
2
7
-
—
_
1
—
3
1
-
48
_
1
_
_
—
1
_
_
—
10
35
11
_
4
8
36
19
—
—
Taxon
Gallinula hodgenorum
Capellirallus karamu
Porzana sp.
Aptornis otidiformis
Aptornis defossor
Coenocorypha cf.
aucklandica
Himantopus novaezelandiae
Circus eylesi
Harpagornis moorei
Falco novaeseelandiae
Poliocephalus rufopectus
Podiceps cristatus
Phalacrocorax carbo
Phalacrocorax varius
Phalacrocorax
melanoleucos
Egretta alba
Acanthisitta chloris
Xenicus sp.
Traversia lyalli
Dendroscansor
decurvirostris
Anthornis melanura
Prosthemadera
novaeseelandiae
Gerygone igata
Petroica australis
Petroica macrocephala
Mohoua novaeseelandiae
Mohoua ochrocephala
Rhipidura fuliginosa
Corvus moriorum
Callaeas cinerea
Island
NS
N
NS
N
S
NS
NS
NS
N?S
NS
NS
NS
NS
NS
NS
N?S
NS
NS
NS
N?S
NS
NS
NS
NS
NS
S
S
NS
NS
(N)S
Philesturnus carunculatus (N)S
Turnagra capensis
Bowdleria punctata
(N)S
NS
Minimum number of
individuals in sample
Pyramid
Lake
Valley Poukawa
9
-
10
—
2
7
4
1
2
—
-
—
1
-
1
1
-
2
11
—
3
1
2
-
-
1
8
2
2
2
15
22
3
-
2
-
12
-
4
5
2
3
2
5
3(4)
-
-
-
—
1
-
—
—
-
3
13
2
—
—
HHC
3
-
2
5
-
1
2
—
-
1
—
-
-
21
48
19
1
2
2
1
52
10
5
6
2
4
61
10
1
_
NOTES:
The collection from Lake Poukawa has not been reassessed. The identifications of Oxyura australis and
Ardea pacifica from Lake Poukawa (Horn 1983) are problematical. Ardea paciflca would be the first fossil record of
what is at present a rare vagrant, and requires further study; it is not included here. The anatid bones from Lake
Poukawa require further examination in the light of improved knowledge—the early record of Anas rhynchotis would,
if
true,
confirm its presence in New Zealand before human occupation, which its absence from Pyramid Valley argues
against; Oxyura australis material
is
probably referable to Malacorhynchus scarletti or Aythya (authors' unpubl. data);
Millener (1991) placed the material in Aythya. Nestor notabilis and Harpagornis moorei have been reported from the
Otiran in the North Island (Holdaway 1991b; Holdaway & Worthy 1993) but they have not been reliably reported from
Holocene deposits.
Holdaway & Worthy—Pyramid Valley fossil fauna
79
Fig. 3 Representation by major
faunal groups (orders and fami-
lies)
of
birds in the
Pyramid Valley
fauna: A, all groups;
B,
excluding
moas.
Taxa listed clockwise from
top
of figures.
Taxa between
song-
birds and rails as in A.
B
3500 years
b.p.
As some moas definitely entered the
deposit through the overlying peat, and bones were
found in the peat, the shorter period is likely to he
an underestimate. The bounds implied
are,
however,
useful to apply absolute limits to the rates of depo-
sition. The rate of entrapment of moas would range
between 0.139/yr(one every 7.18 years) and 10.8/
yr (low extreme, 0.065/yr) for
a
period of deposition
lying between 250 and 1300 years (extreme 2500),
and the number of moas ranging from the number
excavated to the highest possible number.
One gauge of the possible bias resulting from
bogging as opposed to random incorporation is a
comparison of raw rates of deposition for a large,
preferably volant, non-moa taxon in different entrap-
ment regimes. Rates for the well-dated deposits at
Honeycomb Hill Cave and in Pyramid Valley sug-
gest that lakes or swamps trapped eagles at a higher
rate than did caves. In the Graveyard deposit in
Honeycomb Hill Cave the
14
C dates suggest that
eagle remains in Layer 3 were deposited over a pe-
riod of at least 4200 years, lasting from c. 20 600
years b.p. to 16 200 years
b.p.
and perhaps for 6600
radiocarbon years, until 14 000 years b.p. (Worthy
1987a). The corresponding rates for the five eagles
in the Graveyard were 0.0012/yr or 0.00091/yr, de-
pending on the time involved. On average, a bird
died in the cave every 840—1100 years (Holdaway
1991b).
If the four eagles identified in Pyramid Valley
died there over
a
period of
1300
years, the minimum
rate of entrapment would be 0.0031/yr, or one bird
every 325 years,
2.5—3
times the rate at Honeycomb
Hill Cave. Even if eagles were trapped over 2500
years,
which is improbable because entrapment was
more likely in the lake sediments than in the peat
after the lake had been vegetated, the rate would still
be more than for the cave. If only 20% of
the
birds
80
New Zealand Journal of Zoology, 1997, Vol. 24
20
Pyramid Valley
Eagle's Roost
Graveyard 1&2
Eagle's Roost + Graveyard 1&2
oukawa Layer 3
Poukawa Layer 3 + all moas
I
10
I
20
Sample
I
30
size (n)
i
40
I
50
en,
of
50 —i
40 -
30 -
20 -
10 -
B
-•-Pyramid Valley
—*—Eagle's Roost
—o—
Graveyard 1&2
—•—Eagle's Roost + Graveyard 112
-o-Poukawa Layer 3
—o—
Poukawa Layer 3 + all moas
1
100
1 ' I '
200 300
Sample size
i
400
(n)
i
500
I
600
Fig. 4 Comparison of species di-
versity by rarefraction curves for
three major fossil avifaunas (Pyra-
mid Valley, Honeycomb Hill Cave,
Poukawa Swamp): A, fossil sam-
ple sizes up to 50 individuals; B,
fossil sample sizes up to 600 indi-
viduals. Data for sites at Honey-
comb Hill Cave (Eagle's Roost;
Graveyard, layers 1 and 2) from
Worthy (1993a) and Worthy &
Mildenhall (1989). Data from
Poukawa from Horn (1983), with
additional information by THW
(unpubl. data).
that entered each trap were preserved and found, then
an eagle might have been trapped in Pyramid Val-
ley Swamp every 65 years (Holdaway 1991b).
The rate for Pyramid Valley is probably under-
estimated because only part of
the
deposit has been
excavated—that containing the highest concentra-
tion of moas. For the shorter period suggested by
Burrows (1989), eagles would have been added at
one per 60 years (0.0167/yr), or one every 12 years
(0.0835/yr), allowing for attrition. Only 7.5% of the
site has been excavated, so it is possible, if unlikely,
that up
to
40 eagles may be represented, which would
make a maximum rate of incorporation of 0.16-0.8
eagles/yr (Holdaway 1991b). Therefore, eagles seem
to have been added
to
the deposit at intervals of 1.25—
325 years, most likely at the upper limit of the range.
Moas were trapped at 10-40 times the rate for ea-
gles.
The most likely explanation of the higher rate of
deposition (not, strictly speaking, entrapment) of
eagles at Pyramid Valley than in Honeycomb Hill
Cave is that, in addition to fatal injuries incurred
Holdaway & Worthy—Pyramid Valley fossil fauna
81
during predation attempts, eagles died and fell or
were washed into the lake in the normal course of
events, as with most of the other volant species. In-
clusion by random death would be much more com-
mon over the surface area of a lake or swamp than
over a small cave entrance.
Predator/prey ratios
The ratio of minimum numbers of eagles to those of
moas was 4:183 (2.19%). Including other likely prey
species (goose, takahe, adzebill) the ratio was 4:197
(2.03%).
For Eyles's harrier (Circus eylesi), assum-
ing that the bird took diurnal species ranging in size
from the larger songbirds to takahe, the ratio was
7:211 (3.32%).
SYSTEMATIC PALEONTOLOGY
The number of bones listed for each taxon is for
confirmed identifications of that taxon; MNI were
calculated from this sample. Catalogued material not
found is listed separately in Appendix
1.
The number
of elements was not available for a few specimens
in collections outside New Zealand; the extra
elements are indicated by +. The record offish can-
not be confirmed as the material was not located.
Series AMNIOTA
Class REPTILIA
Order SPHENODONTIA
Family SPHENODONTIDAE
Genus Sphenodon Gray
Sphenodon sp. (tuatara)
MATERIAL: 27/4
Daugherty et
al.
(1990) showed that there are at least
two living species of tuatara, and that several
populations of the more widespread Sphenodon
punctatus are genetically distinctive. They noted that
the living populations of
S.
punctatus on islands in
western Cook Strait differed both from northern
populations of the same species and from the sole
known remaining population of
S.
guntheri on The
Brothers islands at the eastern entrance to Cook
Strait. No osteological differences between the spe-
cies and populations have yet been discerned (Wor-
thy & Holdaway 1995), therefore we have not
referred material from the extinct eastern South Is-
land populations to species.
Table 4 Diversity measures for Pyramid Valley and other New Zealand fossil faunas of comparable age
and size. ER, Eagle's Roost; Q1&2, Graveyard layers
1
& 2 combined. Poukawa Swamp:
L3,
Layer
3
data
only (see note 6); L3+moas, including all moas as being in layer 3 (note 2).
Diversity Index
Hill's numbers
NO
Nl
N2
Hill's Evenness
Shannon indices
H'
TT
**max
Pielou's Evenness
J
Simpson's Index
MNI
Pyramid
Valley
46
19.727
11.423
0.557
2.982
3.829
0.779
0.0875
438
Honeycomb Hill
ER
33
15.737
11.152
0.689
2.756
3.497
0.788
0.0897
319
Gl
&2
25
6.699
5.435
0.510
2.272
3.219
0.706
0.184
133
Cave
ER+
G1&2
41
19.688
14.368
0.715
2.980
3.714
0.802
0.0696
452
Poukawa Swamp
L3
37
16.297
10.870
0.645
2.791
3.664
0.762
0.0920
716
L3+moas
43
18.878
12.624
0.650
2.938
3.761
0.781
0.0792
797
NOTES:
(1) seabirds omitted; (2) Poukawa moa data from Holdaway & Worthy (1993), all included as
from layer
3;
(3) the Honeycomb Hill sites are late Glacial to Holocene; (4) Poukawa identifications have
not been checked; (5) data for Honeycomb Hill Cave faunas from Worthy & Mildenhall (1989), for
Poukawa from Horn (1983); (6) data from layer 3 only for Poukawa included because of contamination
from post-Polynesian arrival taxa in other layers.
82
New Zealand Journal of Zoology, 1997, Vol. 24
o
c
a
•a
c
3
.2
c
1 1 1
1
III
-
10 -
I
III 1 1
ITT-1
I
' i
—•—Pyramid Valley
—O-Pyramid Valley
(-moas)
—*— Poukawa Swamp
—*—Honeycomb Hill Cave
—•—Hawkes Cave
hssss
• • i i
Fig. 5 Diversity of bird species
recovered from Pyramid Valley and
from two other major sites, and a
predator
site,
given as logarithm of
relative abundance against rank.
10 20 30 40
Rank of taxon (decreasing abundance)
50
Order SQUAMATA
Family GEKKONIDAE
Genus Hoplodactylus Fitzinger
Hoplodactylus sp. (small gecko)
MATERIAL: 1/1
Small geckoes are very rare in lacustrine or swamp
deposits. They are, however, abundant in predator
deposits near Pyramid Valley (Worthy & Holdaway
1996a).
Class AVES
Order STRUTHIONIFORMES
Suborder CASUARII
Family APTERYGIDAE
Genus Apteryx Shaw & Nodder
Kiwi bones are morphologically conservative and so
present a problem for paleontologists. Adult bones
of the little spotted kiwi Apteryx owenii are, how-
ever, significantly shorter and more gracile than
those of either brown
kiwi
A.
australis or great spot-
ted kiwi A. haastii (Worthy 1997), and so can be
separated from adult bones of the larger species.
Bones of the larger species cannot be separated on
form or measurements on present knowledge; al-
though A. haastii does reach a larger size than A.
australis, dimensions of adult bones of the two spe-
cies overlap widely (THW unpubl. data).
Relatively few kiwi bones were found at Pyramid
Valley. Most were either immature or were non-
diagnostic elements. No bones of
A.
owenii were
found. Only one set of associated bones of an indi-
vidual was found, a specimen gifted to Monash
University, Victoria, Australia in the 1970s. Meas-
urements of these bones (ex CM
Av5841,
measured
by RNH) are comparable to those of CM Av5843
and Av 15060. All fossils were stout for their length,
as illustrated by the shaft widths (range 9—10 mm,
as against 6.4-8.7, 5.3-8.0, and
6.1-8.0
for femora,
tibiotarsi, and tarsometatarsi, respectively) in com-
parison with
A.
owenii (THW unpubl. data). There-
fore,
despite the limb bones being slightly shorter or
just longer than those of
the
largest A. owenii, they
are too stout to be of this species. In this respect they
are similar to A. australis, and we refer them to a
large kiwi cf. A. australis. However, although simi-
lar in stoutness to those of the
A.
australis group, its
bones were much shorter than those of the 17 South
Island brown kiwi, including Stewart Island birds,
available for comparison. The possible implications
of these differences will be discussed elsewhere,
using the larger samples of similar large kiwi and
abundant little spotted kiwi from South Canterbury
(Worthy 1997).
Apteryx sp. cf. A. australis (eastern brown kiwi)
MATERIAL: Adults
25/3;
juveniles
3/1.
Total 28/4.
Apteryx sp. (immature)
MATERIAL: Juveniles
22/3.
Total 22/3.
Table 5 Broad guilds of vertebrates at Pyramid Valley during the mid Holocene. Bold type = globally extinct; *, locally extinct; t, reduced numbers locally; gp,
species group; [ ], taxa almost certainly present, but excluded because of taphonomic bias.
DIURNAL
Aquatic Terrestrial Arboreal
Arboreal-terrestrial
Aerial
Predator
X
o
o
3.
I
<
Malacorhynchus scarletti
Anas gracilis^
Anas superciliosa^
Aythya novaeseelandiae^
Himantopus novaezelandiae*
Poliocephalus rufopectus*
Egretta alba\
Herbivore
Emeus crassus
Pachyornis elephantopus
Euryapteryx geranoides
Dinornis struthoides
Dinornis giganteus
Cnemiornis calcitrans
Porphyrio hochstetteri*
Omnivore
Tadorna variegata
Anas chlorotis*
Euryanas finschi
Coturnix novaezelandiae
Gallirallus australis*
Insectivore
Mohoua novaeseelandiae
Bowdleria punctata*
[Acanthisitta
chloris]
[Gerygone
igata]
Insectivore
Xenicus sp.*
Traversia lyalli
Petroica macrocephala
Petroica australis^
Philesturnus carunculatus*
Insectivore
[Rhipidura fuliginosa]
Circus eylesi
Harpagornis moorei
Falco novaeseelandiae^
Insectivore-nectarivore
Anthornis melanura
Prosthemadera
novaeseelandiae
Frugivore-folivore
Frugivore-folivore
Cyanoramphus novaezelandiae*
Cyanoramphus auriceps^
Strigops habroptilus*
Fulica prisca Hemiphaga novaeseelandiae
Gallinula hodgenorum Callaeas cinerea*
Aptornis defossor Nestor meridionalis\
Omnivore
Nestor notabilis
Turnagra capensis
Corvus moriorum
Terrestrial
NOCTURNAL OR NOCTURNAL/CREPUSCULAR
Arboreal-terrestrial Aerial-arboreal-terrestrial Aerial Predator
Omnivore
Sphenodon sp.*
Insectivore
Apteryx sp.*
Insectivore-nectarivore Insectivore-nectarivore- Insectivore
frugivore
Hoplodactylus gp. Mystacina tuberculata* [Chalinolobus tuberculatus] Aegotheles
novaezealandiae
Sceloglaux albifacies
Ninox novaeseelandiae
c
S
00
84
New Zealand Journal of Zoology, 1997, Vol.
24
Suborder DINORNITHI
Family EMEIDAE
Subfamily ANOMALOPTERYGINAE
Genus Pachyornis Lydekker
Pachyornis elephantopus (Owen, 1856) (heavy-
footed moa)
MATERIAL:
17
part skeletons, plus miscellaneous
elements.
Subfamily EMEINAE
Genus Emeus Reichenbach
Emeus crassus (Owen, 1<846) (eastern moa)
MATERIAL:
81
part skeletons, plus miscellaneous
elements.
Genus Euryapteryx Haast
Euryapteryx geranoides (Owen, 1848) (stout-
legged moa)
MATERIAL:
21
part skeletons, plus miscellaneous
elements.
Family DINORNITHIDAE
Genus Dinornis Owen
Dinornis struthoides Owen, 1844 (slender moa)
MATERIAL:
One individual.
Dinornis giganteus Owen, 1844 (giant moa)
MATERIAL:
Partial skeletons of at least 63 individu-
als,
including 9 juveniles.
Dinornis
sp.
MATERIAL:
4
bones, from
a
minimum of 2 individu-
als,
one
of
which was a juvenile.
Moa sp. indet.
MATERIAL:
See Appendix 1.
Species composition
of
the moa fauna
Five
of
the nine South Island species were repre-
sented.
All
were typical members
of
eastern South
Island faunas
in
the Holocene. Nineteen more indi-
vidual moas were recorded here (including those
in
AMNH, seen THW) than the 164 listed in Anderson
Table
6
Changes
in
species richness (S)
for
broad guilds
of
vertebrates
at
Pyramid Valley from mid-Holocene
to
present faunas. H, herbivore; O, omnivore; I, insectivore; I-N, insectivore-nectarivore; F-F, frugivore-folivore; I-N-F,
insectivore-nectarivore-frugivore; SD, specialist dabbler; D/S, dabbler/shore feeder; W, wader; D, diver. "Nocturnal"
includes crepuscular/nocturnal; *, taxon replaced by self-introduced taxon (W,
Himantopus
leucocephalus;
predator,
Circus
approximans);
replacements not included in extinction calculation. Taxa introduced by humans not included.
Main activity height
Aquatic
Terrestrial
Arboreal
Arboreal-terrestrial
Aerial
3
Aerial-arboreal-terrestrial
Predator
Total
Food
1
SD
D/S
W
D
H
O
I
I
I-N
F-F
I
F-F
O
I-N
I
I-N-F
Pre-human
1
2
2
2
7
8
3
2
3
5
3
3
3
44
Diurnal
Present
0
1
0*
0
0
1
0
1
I
2
0
0
0
j2*
5
%
loss
100
50
100
100
100
87.5
100
50
67
100
100
100
67
87
Species richness
(S)
Pre-human
2
1
1
3
7
Nocturnal
Present
0
0
0
I
2
1
%
loss
100
100
100
67
86
'Feeding method/site for aquatic taxa; Occasional to very rare visitor, not resident near Pyramid Valley
(Hemiphaga
novaeseelandiae);
3
other taxa {Rhipidura
fuliginosa,
Chalinolobus tuberculatus) known from other fossil sites near
Pyramid Valley
(authors'
unpubl.
data).
Holdaway & Worthy—Pyramid Valley fossil fauna
85
Fig. 6 Size frequency histogram
of tibiotarsus lengths of individu-
als of Dinornis giganteus from
Pyramid Valley.
12-1
10 -
£ 8-
E 4"
720-759 I 800-839 I 880-919 I 960-999
680-719 760-799 840-879 920-959
Size classes (length in mm)
(1989).
We consider
the
dinornithids separately from
the emeids because of the marked morphological
differences between the two groups, and the prob-
able differences in their habitat use.
Emeids The relative frequency of emeid taxa was
very similar to that at Glenmark (Worthy 1990).
Pachyornis and Euryapteryx were in roughly equal
numbers, together comprising about half the total
emeids, Emeus crassus making up the balance.
Glenmark is c. 20 km away and is the nearest large
moa site to Pyramid Valley. The similar proportions
of emeid species at the sites suggest similar environ-
ments. Cheviot Swamp, farther to the north, had
more
P.
elephantopus (17) than either Emeus
crassus
(10) or Euryapteryx geranoides (3).
Pachyornis elephantopus seems to have been the
commonest moa in open landscapes of tussock grass-
land and low shrubs; it dominates loess faunas (Wor-
thy 1993b). Emeus and Euryapteryx appear to have
preferred lowland (<200 m), flat and upland
(>200 m), hilly areas, respectively (Worthy 1990),
in agreement with the higher proportion of Emeus
than Euryapteryx at Glenmark (85± m a.s.l.), Pyra-
mid Valley (335 m), and Cheviot. The higher pro-
portion of Pachyornis elephantopus at Cheviot
suggests that there was more shrubland around this
site than at either Pyramid Valley or Glenmark.
Dinornithids Pyramid Valley has a uniquely high
proportion of dinornithids in comparison to other
swamp and lake sites (Worthy 1990). The anomaly
in representation probably resulted from the unique
conditions of entrapment rather than the composi-
tion of the source fauna around the site. The very
large Dinornis giganteus appear to have been very
susceptible to miring in lake sediments. The remains
of gizzard contents (Burrows et al. 1981) contained
fruits that support the contention that many of the
birds,
if not most, were trapped in summer. Summer
drought is normal in the Pyramid Valley area today.
The present lake fluctuates in extent seasonally.
Moas probably ventured out onto the exposed lake
bed to drink at the remaining pools.
Dinornis giganteus was by far the most abundant
dinornithid in Pyramid Valley. Crania of dinornithids
can be distinguished on morphology (Worthy 1994);
all the crania from Pyramid Valley were referable
to D. giganteus. Earlier lists included D. novae-
zealandiae, or its equivalent
D.
robustus, on the ba-
sis of the presence of the smaller individuals
Av8418,
8475, 8495, and 15028 (Falla 1941b;
Scarlett 1955a, 1969). The individual XVIIA
(=Av8461) listed by Falla is
a
juvenile.
The lengths of the tibiotarsi of all part skeletons
of
D.
giganteus were normally distributed (Fig. 6),
despite the smallest being well within the range for
D.
novaezealandiae
from other areas (Worthy 1994).
The summary statistics for the pooled samples for
all limb bones are (mean ± SD, SE
mean
, CV(%), n,
range):
femur 392.0 ± 30.16, 4.71, 7.69, 41, 328-
447;
tibiotarsus 838.3 ± 80.86,
12.33,
9.65,43,681-
992;
tarsometatarsus 452.4 ± 46.48, 7.01, 10.28,
345—531.
Although the extension of size at the lower
end of
the
range goes beyond the usually accepted
limits for
D.
giganteus, the coefficients of variation
for the pooled samples range from 7
to
just over 10,
which is normal for a single, sexually dimorphic
species (Holdaway 1990a).
Most previous attempts at "defining" species of
Dinornis have involved arbitrarily constraining "spe-
cies"
to size ranges considered by the worker to be
acceptable (e.g., Archey
1941;
Hutton 1892; Oliver
New Zealand Journal of Zoology, 1997, Vol. 24
1949).
This method was partly a response to the
small samples available, and partly an inability to
perceive the extent of normal variation possible rela-
tive to the huge absolute size of
the
bones.
Where a large sample was available, as at
Makirikiri in the North Island, three size groups
could be discerned; their intermediate points on his-
tograms could be taken as the species limits (Wor-
thy 1989). The large sample from Pyramid Valley
allowed a better estimate of the full size range of
D.
giganteus than had been possible before. It would
be unacceptable to remove the smaller individuals
from the bottom of the size range, as that would re-
move the lower tail of the normal distribution.
Hence, between 2000 and 4000 years ago the popu-
lation of D. giganteus near Pyramid Valley had
tibiotarsi from 680 mm to nearly 1000 mm long.
The ratio of femur length to tarsometatarsus
length may be a useful means of discriminating be-
tween D. giganteus and D. struthoides. For a series
of the smallest individuals attributed to
D.
giganteus
on element length, the ratio was always <1; for the
individual attributed to
D.
struthoides (Av8415), the
ratio was >1.
Population structure
Emeids The large sample of Emeus crassus allowed
an examination of
its
population structure. Of the 81
part skeletons recorded from the deposit, nine are
outside New Zealand, one was used for dating, and
four could not be located, leaving 67. The nine that
were exchanged or donated outside New Zealand
were probably adult specimens. Of the 67 remain-
ing, nine were juvenile, and were more than half
adult size, but with patellae unfused to proximal
tibiotarsi, and tarsals unfused to distal tibiotarsi and
proximal tarsometatarsi. No young smaller than half
the mean adult size were recorded. Three individu-
als could be classed as subadult; their bones had
reached adult size and shape, but the fused elements
were still incompletely ossified. This stage is the
equivalent of a 4 year old brown kiwi Apteryx
australis (Beale 1991).
In total, therefore, 82% were adult (more if the
exchanged specimens were included), more ad-
vanced in ontological age than
a
4 year old kiwi, and
up to
18%
were subadult or juvenile. The proportion
of immature birds was rather lower than for
Megalapteryx
didinus
(56%)
or Pachyornis australis
(30%) at the Graveyard, Honeycomb Hill Cave
(Worthy & Mildenhall 1989). Proportions of
juve-
niles are known from two other
sites:
at Makirikiri—
Anomalopteryx
didiformis
14%,
Pachyornis mappini
29%
(Worthy 1989); and at Cheviot—Emeus
crassus 30% of 10, Pachyornis elephantopus 12%
of 17(THWunpubl. data).
No other large samples have been studied yet, but
if these values reflect actual population age struc-
tures they suggest that production of young differed
markedly between species. The smaller upland moa
Megalapteryx didinus seems to have had a higher
reproductive potential than did its lowland counter-
part, the little bush moa Anomalopteryx didiformis.
The Pachyornis species had similar proportions of
juveniles. The pattern appears to parallel the degree
of seasonality in the preferred habitat: moas in
strongly seasonal environments, such as the
subalpine zone, may have had a higher reproductive
potential than those in areas with more equable cli-
mates such as wet lowland forest.
Dinornithids Because of the number of individuals
sent to other institutions, only 55 of the 63 Dinornis
giganteus could
be
aged, assuming that the chick was
of this species. Forty-six (84%) were adult, eight
(14%) juvenile, and one (2%) subadult. The only
other site where dinornithids formed a significant
proportion of the moa fauna, and which could there-
fore be meaningfully compared with Pyramid Val-
ley, was Makirikiri. There, the proportions of
juveniles were D. novaezealandiae 13%, D.
struthoides
40%,
and
D.
giganteus 9%. In both sites
D.
giganteus had a low percentage of juveniles,
which suggests that it had lower reproductive rate
and so may have been a K-selected species.
Sex ratios
As noted above, the distribution of tibiotarsus length
for the sample of Dinornis giganteus was unimodal
(Fig. 6). The tibiotarsus length normalised to body
size (as the ratio tibiotarsus length to femur length)
was,
however, bimodally distributed (Fig. 7A). This
suggests that one sex had a relatively shorter
tibiotarsus. There were also fewer of the sex with the
relatively longer tibiotarsi (Fig. 7A). If the difference
was sexual, then these results suggest that the sex
ratio was skewed or that one sex preferred to live
near the lake or was preferentially attracted to it. If
the birds with the longer tibiotarsi were males, as is
likely for Emeus (see below),
D.
giganteus probably
did not have a monogamous mating system. No liv-
ing ratite is known in which the sexes have differ-
ent habitat preferences.
A plot of normalised tarsometatarsus length
against normalised tibiotarsus length (as ratios of
femur length) for identified individuals of Dinornis
species from other sites and specimens from Pyramid
Holdaway & Worthy—Pyramid Valley fossil fauna
87
Fig.
7
Frequency histograms for
ratios of tibiotarsus length to femur
length.
A, All
individuals
of
Dinornis from Pyramid Valley;
individual
in
0.520-0.524 class
referred
to
Dinornis struthoides.
B,
All
individuals
of
Emeus
crassus.
E.
crassus individual pre-
served with egg under body cavity
is presumed
to be
female. Indi-
viduals with higher ratios have
shorter tibiotarsi relative
to
body
size.
0.450-0.454
I
0.470-0.474 0.490-0.494
I
0.610-0.514
0.440-0.444 0.460-0.464 0.400-0.484 0.500-0.504 0.520-0.524
Ratio classes
10 -i
~
5-
Adult with egg
Adults
Subadults
I
0.540-0
S4<
0530-O53*
1
\
11
Ratio classes
Valley (Fig. 8A), supported the presence
of
sexual
dimorphism
in
D. giganteus and the identification
of the sole D. struthoides from Pyramid Valley.
There was a small cluster of individuals at the lower
end of the distribution for
D.
giganteus, for both re-
ferred and confirmed (with crania) individuals from
Pyramid Valley and elsewhere. There was less evi-
dence for
a
similar separation of sexes among the few
confirmed
D.
novaezealandiae, which had leg bones
with similar proportions of length relative (to body
size) to those of
D.
giganteus. D. struthoides, how-
ever, had significantly shorter tarsometatarsi relative
to their body size. The ratio of tarsometatarsus length
to femur length for
D.
struthoides was over 1.00 for
all confirmed individuals and for the Dinornis from
Pyramid Valley attributed
to D.
struthoides
(Fig. 8A).
Two species have been included in Emeus in the
past (Archey
1941;
Oliver 1949). These were lumped
later into one sexually dimorphic species because no
shape differences were detected
by
multivariate
analysis; the pooled samples from both "species" had
coefficients of variation (CVs) within the range
for
sexually dimorphic species (Cracraft 1976b).
If
is.
crassus were truly dimorphic,
the
size-frequency
histograms for the major elements should be bimo-
dal,
as
they
are for
Euryapteryx curtus
and
Pachyomis mappini (Worthy 1987b). The lengths of
the large sample of leg bones from Pyramid Valley
were
not
obviously bimodal,
but the
femur
and
tibiotarsus
had
markedly skewed distributions
(Fig. 9). The CVs (6-7)
for all
three elements can
be explained best by
a
moderate degree of sexual size
dimorphism. The normalised tibiotarsus length had
a more normal distribution (Fig. 7A) than that of
D.
giganteus (Fig. 7A), but there was some indication
of bimodality.
A
female—an individual associated
with an unlaid egg (Av8309)—was toward the up-
per end of the distribution, and hence had
a
relatively
short tibiotarsus (Fig. 7B).
If
the relatively shorter
o
'•5 0.95 H
E
E 0.85-
." * •
0.475 0.525
Fem/tbt length ratio
B
New Zealand Journal of Zoology, 1997, Vol. 24
Fig. 8 Ratio of tarsometatarsus
length to tibiotarsus length, both
normalised to body size
as
ratios of
femur length. A, Dinornis
spp.:
•,
Pyramid Valley individual referred
to Dinornis giganteus; +, Pyramid
Valley individual confirmed as D.
giganteus; A, D. giganteus con-
firmed at other sites; •, Dinornis
novaezealandiae confirmed at
other
sites;
•, Dinornis struthoides
confirmed at other sites; *, D.
struthoides, referred, Pyramid Val-
ley.
B,
Emeus
crassus:
•. adult; A,
subadult; +, adult with egg.
("Con-
firmed", individual with associated
cranium and identified by cranial
characters; "referred", individual
without cranium, identified by limb
bone lengths or length proportions.
"Adult", individual with ossifica-
tion consistent with that of adult
(>4 year old) kiwi; "subadult", in-
dividual with unfused or partly
fused epiphyses or patellae.)
0 56 0 58 0.6
Fem/tbt ratio
tibiotarsus (and hence lower body height in
a
heavier
bird) indicates some degree of sexual dimorphism,
the female was again the heavier sex, as in kiwis and
other ratites. As in D. giganteus, the heavier birds
with relatively shorter tibiotarsi in relation to body
size (as measured by femur length) appear to have
been females.
The normalised lower leg proportions of the
Emeus individuals (Fig. 8B) showed the presence of
two roughly equal groups in the population, with the
presumed female (bird preserved with egg) in the
group with the relatively shorter tibiotarsus. The
roughly equal groups suggest that the sex ratio was
similarly closer to equality than in D. giganteus.
Because of the relatively brief period of deposi-
tion (in evolutionary terms), it is unlikely that the
differences in proportion observed were the result of
evolutionary changes. They are more likely to have
been characteristics of
a
single population.
Infraclass NEOAVES
Order GALLIFORMES
Family PHASIANIDAE
Genus Coturnix Bonnaterre
Coturnix novaezelandiae Quoy & Gaimard,
1830 (New Zealand quail)
MATERIAL: Adult 1/1.
The New Zealand quail was apparently most abun-
dant in open grasslands in European times (Oliver
1955),
but it also seems to have inhabited dry
shrubland (authors' unpubl. data). The nearest suit-
able habitat was probably on the exposed ridges
around Pyramid Valley.
Holdaway & Worthy—Pyramid Valley fossil fauna 89
Fig.
9
Size
frequency histograms
for lengths
of (A)
femur,
(B)
tibiotarsus, (C) tarsometatarsus of
individuals of Emeus crassus from
Pyramid Valley.
220-229 230-239 240-249 250-259 260-269 270-279
Size classes (length
in mm)
20-i
350-369 370-389 390-409 410-429 430-449 450-469
Size classes (length
in
mm)
470-489 490-509
25-1
1S0-189 190-199 200-209 210-219
Size classes (length
in mm)
90
New Zealand Journal of Zoology, 1997, Vol. 24
Order ANSERIFORMES
Family ANATIDAE
Subfamily ANSERINAE
Genus Cnetniornis Owen
Cnemiornis calcitrans Owen, 1865 (South Island
extinct goose)
MATERIAL: Adults
27/1;
juveniles 0/0. Total
27/1.
The goose was the rarest anseriform at Pyramid
Valley. Given the size of goose bones (comparable
to those of the adzebill), the rarity of remains is prob-
ably a good indication of its relative abundance in
the living assemblage. If
so,
forest was not its pre-
ferred habitat, even when open water was available.
The goose may have been almost exclusively terres-
trial, as were other island geese (e.g., Hawaiian goose
Branta sandvichensis)
and the moa-nalos (goose-like
ducks of
Hawaii;
Olson & James 1992). The goose
is one of the few carinate species recorded from loess
deposits (Worthy 1993b). The closest ecological
equivalent in Australasia was probably its relative
the Cape Barren goose Cereopsis novaehollandiae
(THW unpubl. data), which is now confined to small
islands off southern Australia in habitats of grass-
land and shrubland
<1
m high (Marchant & Higgins
1990).
Subfamily TADORNINAE
Genus Tadorna Lorenz von Oken
Tadorna variegata (Gmelin, 1789) (paradise
shelduck)
MATERIAL: Adults
31/4;
juveniles
2/1.
Total 33/5.
The paradise shelduck is one of the few species
found fossil at Pyramid Valley that still breeds at the
lake (pers. obs.). Their presence on a forest-fringed
lake some distance from large riverbeds, in a largely
forested area (Moar 1970), supports the suggestion
that they could survive in different habitats than
those occupied today (Holdaway 1989).
Subfamily ANATINAE
Genus Euryanas Oliver
Euryanas finschi (Van Beneden, 1875) (Finsch's
duck)
MATERIAL: Adults
19/4;
juveniles 3/1. Total 22/5.
Finsch's duck was more common in drier
areas,
such
as the east of the North and South Islands in the
Holocene, and northwest Nelson and Takaka during
the Otiran glaciation (Worthy & Holdaway 1993,
1994b).
Genus Malacorhynchus Swainson
Malacorhynchus scarletti Olson, 1977 (Scarlett's
pink-eared duck)
MATERIAL: Adults
21/4;
juveniles 9/3. Total 30/7.
Pyramid Valley is the type locality for
Malacorhynchus scarletti Olson, 1977. Further ma-
terial found in the Canterbury Museum collection
catalogued under other taxa has been used to
redescribe the species (Worthy 1995). M. scarletti
was much larger than the extant M. membranaceus
of Australia. The Australian bird is
a
highly nomadic,
specialised filter-feeder. The bill of M. scarletti had
a similar shape to that of M. membranaceus; it was
also a filter-feeder.
Genus Anas Linnaeus
Anas chlorotis Gray, 1845 (brown teal)
MATERIAL: Adults
87/8;
juveniles 18/3. Total 105/
11.
Including specimens listed as cf. chlorotis:
Adults
88/8;
juveniles
21/3.
Total
109/11.
The brown teal was the most abundant duck at Pyra-
mid Valley. Elsewhere it is common in pitfall traps
away from swamps or wetlands (Worthy &
Holdaway 1993, 1995). Pyramid Valley was prob-
ably very like the habitats in which the species was
observed in early European times, the margins of
a
shallow forest pool. The brown teal forages mainly
at night, when there are few human observers, and
it was probably widespread in North Canterbury, in
a variety of habitats (Williams 1974).
Anas cf. chlorotis
MATERIAL: Adults
1/1;
juveniles
3/1.
Total 4/2.
Anatid material in the correct size range for A.
chlorotis,
but without diagnostic morphological fea-
tures.
Anas gracilis Buller, 1869 (grey teal)
MATERIAL: Adults
2/1;
juveniles 2/2. Total 4/3.
The femora (Av6190, 6191, both right) are shorter
(37.64 and 36.36 mm, respectively) and more slen-
der (proximal width 7.60, 6.76; shaft width 3.20,
2.94; distal width 8.35, not measurable) than those
of
A.
chlorotis, but are within the range for
A.
gra-
cilis (Worthy & Holdaway 1994b, tables 1-3). Hu-
meri (Av5948 left, L 61.04, P 13.56, S 5.20, D 9.50;
Av5949, right L 67.26, P 14.27, S 5.17, D 10.62)
differ from those of
A.
chlorotis as follows: (1) the
entepicondyle is not as rotated caudally, so that the
entepicondyle faces c. 45° to the palmar aspect; (2)
in humeri of A.gracilis the marked tuberosity be-
tween the shaft and the bicipital fossa
in
A chlorotis
Holdaway & Worthy—Pyramid Valley fossil fauna 91
is absent or weak; (3) the proximal end of
the
bone
is relatively smaller in
A.
gracilis.
The juveniles show that this small duck bred at
Pyramid Valley and proves that the grey teal is not
a recent immigrant to New Zealand, as has been sug-
gested (e.g., Turbott 1990). Its reported abundance
at Lake Poukawa (Horn 1983) has yet to be con-
firmed by re-examination of
the
material.
Anas superciliosa Gmelin, 1789 (grey duck)
MATERIAL: Adults
18/2;
juveniles
2/1.
Total 20/3.
The grey duck is still widespread in Canterbury (Bull
et al. 1985), but is less common than formerly and
has been displaced from settled districts by the in-
troduced mallard Anas platyrhynchos (Falla et al.
1979).
Anas sp.
MATERIAL: Adults
2/1;
juveniles 15/3.
Genus Aythya Boie
Aythya novaeseelandiae (Gmelin, 1789) (New
Zealand scaup)
MATERIAL: Adults
17/2;
juveniles 8/3. Total 25/5.
The scaup is still present on many lakes and ponds
in the Canterbury foothills, and on some ponds on
the plains, even near Christchurch. Potts (1882) re-
corded it as being abundant in the 1870s, but there
was apparently a major decline after European set-
tlement and the introduction of cats, rats, and
mustelids.
Anatidae sp. indet.
MATERIAL: Adults 8/1; juveniles 22/6.
Order PSITTACIFORMES
Family PSITTACIDAE
Genus Cyanoramphus Bonaparte
Cyanoramphus novaezelandiae (Sparrman, 1787)
(red-crowned parakeet)
Cyanoramphus auriceps (Kuhl, 1820) (yellow-
crowned parakeet)
Cyanoramphus malherbi Souance, 1857
(orange-fronted parakeet)
MATERIAL: Adults
30/7;
juveniles 0/0. Total 30/7.
There is, as yet, no satisfactory method of separat-
ing these taxa as fossils, either morphometrically or
using morphological characters (Worthy &
Holdaway 1993, 1994b). They are therefore treated
together here. C. malherbi is maintained as a sepa-
rate species on the basis of biochemical studies
(Triggs & Daugherty in press), although its specific
status was not supported by earlier breeding experi-
ments (Taylor et al. 1986).
Yellow-crowned parakeets are now restricted on
the mainland to large tracts of native forest. The red-
crowned is nearly extinct on the mainland (Bull et
al.
1985; Elliott et al. 1996); the orange-fronted is
very rare in a few subalpine valleys between Arthur's
Pass and Lake Rotoiti in Nelson Lakes National
Park. Parakeets of one or more species were com-
mon at Pyramid Valley.
Genus Nestor Lesson
Nestor meridionalis (Gmelin, 1788) (South
Island kaka)
MATERIAL:
Adults
137/11;
juveniles 1/1. Total 138/
12.
The two subspecies of kaka can be separated on
morphometrics (Holdaway & Worthy 1993). All the
material included here lies within the range of the
living South Island kaka. Previously it was included
in lists as Nestor n. sp., but all the material is refer-
able to Nestor meridionalis (Holdaway & Worthy
1993).
In turn, the material previously attributed to
South Island kaka is actually referable to the kea
Nestor
notabilis.
Kaka are still present in large tracts
of native forest along the main ranges (Bull et al.
1985;
O'Donnell & Dilks 1986) but are declining
under pressure of predation by introduced stoats
Mustela erminea and other factors (Beggs & Wilson
1991).
Nestor notabilis Gould, 1856 (kea)
MATERIAL: Adults
89/9;
juveniles 0/0. Total 89/9.
This is the first record of kea from Pyramid Valley.
The material has previously been referred to the
South Island kaka (see above).
The kea is generally considered to favour high-
altitude vegetation (Falla et al. 1979; Turbott 1990),
although O'Donnell & Dilks (1986) found it to be
widespread in lowland forest in Westland. It appears
to have been about as common as kaka in the mid-
Holocene forest around Pyramid Valley, but some
birds may have been attracted to feed on carcases.
Several moa pelves from the deposit display evi-
dence not only of predation by Haast's eagle but also
damage consistent with consumption of internal or-
gans (i.e. kidneys) by either eagles or keas or both
(authors' unpubl. data). Typical of
the
damage was
for slots
c.
25 mm wide and 75—100 mm long to have
been cut back from the cranial margin of one or both
anterior iliac plates. Distinct beak-point marks are
visible in the margins of the slots in some pelves.
Similar damage has been recorded in moa pelves
92
New Zealand Journal of Zoology, 1997, Vol. 24
excavated from Glencrieff Swamp near Pyramid
Valley, which also contained kea and eagle remains
(Worthy & Holdaway 1996a).
The evidence, although not conclusive, suggests
that kea may have regularly scavenged moa carcases
and attended eagle kills. The damage is consistent
with the mode of attack attributed to modern keas
attacking sheep, in which the bird concentrates its
attentions on the hind-quarters and the fat surround-
ing the kidney. Sheep may therefore have replaced
moas as an opportunistic item in the diet of some
keas.
Nestor sp. indet. (kea or kaka)
MATERIAL: 9/2.
A residue of Nestor elements lacked diagnostic fea-
tures.
Genus Strigops Gray
Strigops habroptilus Gray, 1845 (kakapo)
MATERIAL: Adults
22/4;
juveniles 1/1. Total 23/5.
The kakapo, if it survives on the mainland, does so
as isolated, aged males in the south-west of the South
Island (Powlesland et al. 1995). A population for-
merly on Stewart Island has been translocated to
other offshore islands (Powlesland et al. 1992). The
species was widespread in the South Island during
the Holocene (Worthy & Holdaway 1993, 1994b).
Order STRIGIFORMES
Family STRIGIDAE
Genus Ninox Hodgson
Ninox novaeseelandiae (Gmelin, 1788)
(morepork)
MATERIAL: Adults
2/1;
juveniles 0/0. Total 2/1.
The morepork, the smaller of the two New Zealand
owls,
is still common in and near native forests in
Canterbury. It is occasionally heard near Pyramid
Valley (M. Hodgen pers. comm.) but is no longer a
common resident.
Genus Sceloglaux Kaup
Sceloglaux albifacies (Gray, 1844) (laughing
owl)
MATERIAL: Adults
18/3;
juveniles 1/1. Total 19/4.
The laughing owl was widespread in Canterbury
until the late nineteenth century (Potts 1882). It was
still breeding near Waikari after the 1860s (Worthy
& Holdaway 1996a). It was a major nocturnal preda-
tor of small to medium-sized vertebrates (Holdaway
& Worthy 1996).
Family AEGOTHELIDAE
Genus Aegotheles Vigors & Horsfield
Aegotheles novaezealandiae (Scarlett, 1968)
(New Zealand owlet-nightjar)
MATERIAL: Adults
14/4;
juveniles 1/1. Total 15/5.
The collection includes paratypic material (Scarlett
1968).
Owlet-nightjars are nocturnal insectivores
that also take small vertebrates such as lizards and
frogs.
The New Zealand owlet-nightjar was among
the largest of the genus, and was relatively abundant
near Pyramid Valley. Its remains occur frequently
in laughing owl prey deposits in the area (Worthy
& Holdaway 1996a). In Atkinson & Millener
(1991,
fig. 13) the owlet-nightjar is shown sitting length-
wise on a branch, after the manner of a true night-
jar; owlet-nightjars sit across branches (Pizzey &
Doyle 1980).
Order COLUMBIFORMES
Family COLUMBIDAE
Genus Hemiphaga Bonaparte
Hemiphaga novaeseelandiae (Gmelin, 1789)
(New Zealand pigeon)
MATERIAL:
Adults
633/67;
juveniles 10/2. Total 643/
69.
The New Zealand pigeon was the most abundant bird
in the deposits after
Emeus
crassus.
It may have been
attracted to the edge of the lake by fruiting matai
(Prumnopitys taxifolia) trees. There is abundant
macrofossil and microfossil evidence that matai was
common around the lake (Moar 1970; Burrows
1989).
New Zealand pigeons appear to have fa-
voured drier eastern forests and shrublands over the
dense wet forests of western New Zealand, as they
are also very abundant in the Lake Poukawa fauna
(Horn 1983) but less common in collections from
Waitomo (western North Island), near Punakaiki
(West Coast, South Island), and near Takaka (north-
western South Island) (THW unpubl. data; Worthy
& Holdaway 1993, 1994b).
Order GRUIFORMES
Suborder RALLI
Family RALLIDAE
Genus Gallirallus Lafresnaye
Gallirallus australis (Sparrman, 1786) (weka)
MATERIAL: Adults 177++/16; juveniles 3/1. Total
180++/17.
Holdaway & Worthy—Pyramid Valley fossil fauna
93
The weka was the most abundant of the four rails at
Pyramid Valley. The CM collection contains much
Gallirallus material labelled either "Gallirallus sp."
or "Gallirallus minor". No morphological features
were found by which these specimens could
be
sepa-
rated from recent or fossil specimens of
Gallirallus
australis.
The length of all measurable major leg bones and
humeri of
Gallirallus
australis sensu lato in the Can-
terbury Museum was measured, but including only
one (usually the left) of any pair identifiable as from
a single individual. Summary statistics for the sam-
ple,
with and without the Pyramid Valley compo-
nents,
are given in Table 7. Except for the small
sample of humeri, the addition of Pyramid Valley
specimens to the whole-country samples did not al-
ter the range, but reduced standard deviations (and
hence also standard errors of the mean) and coeffi-
cients of variation, as would be expected with a
larger sample from a single-source population. In no
instance did the addition of Pyramid Valley data sig-
nificantly alter the whole-country mean.
The lengths in this heterogeneous sample were
normally distributed (Fig. 10), and the coefficients
of variation for all three were within the range ex-
pected for a sexually size-dimorphic species
(Holdaway 1990a). A degree of dimorphism was
apparent in tibiotarsus length but was not marked in
the other leg elements or the small humerus sample
(Fig. 10).
Ranges of length for North, South, and Stewart
Island series referred to G. australis, and for those
referred to G. minor given in Scarlett (1972), are
depicted as lines in Fig. 10. The ranges for all were
drawn as subsets from a continuous distribution. No
gaps are apparent between elements referred to G.
minor and those referred to
G.
australis. The ranges
given for North, South, and Stewart Island sub-
samples by Scarlett (1972) overlap largely or com-
pletely, and no morphometric bases were found for
the elements surveyed by which the races could be
separated (Fig. 10).
Although the type material was at the lower end
of the pooled range (Fig. 10), the data do not sup-
port retaining the nominal species Gallirallus minor
Hamilton,
1893.
Its description was an artefact of the
small sample sizes available to Hamilton (Olson
1975).
In the absence of diagnostic morphological
or morphometric characters, all Gallirallus material
is here assigned to G. australis. The sample from
Pyramid Valley was generally towards the upper end
of each range, suggesting that there may be some
clinal or habitat-related differences in size within the
species.
The tibiotarsus length distribution exhibited some
degree of bimodality. Although insufficient associ-
ated individuals were available for measurement, this
flightless species may also display a sexual differ-
ence in relative length of the tibiotarsus, as in the
much larger and unrelated moas.
Table 7 Summary statistics for element length (millimetres) for femur,
tibiotarus, tarsometatarus, and humerus of mainland Gallirallus specimens in
Canterbury Museum. Mean + SD/SE
mean
, «/CV(%), range. Comparative values
for Gallinula hodgenorum from Olson (1975); SE
n
data given.
, and CV calculated from
Length
Gallirallus australis
Gallinula hodgenorum
Without
Pyramid Valley
Including
Pyramid Valley
Humerus
Femur
Tibiotarsus
Tarsometatarsus
50.94 ± 2.62
0.699, 14
5.13,47.4-55.9
71.1+5.04
0.787,41
7.08,60.1-82.4
100.43 ± 7.80
1.126,48
7.76,85.5-114.7
60.18 + 3.80
0.521,53
6.31,52.4-66.5
51.40 + 2.81
0.680, 17
5.46, 47.4-57.2
71.60 ±4.97
0.702, 50
6.93,60.1-82.4
101.03 ±7.66
0.981,61
7.58,85.5-114.7
60.11 ±3.83
0.483,
63
6.37, 52.4-66.5
40.4 ± 2.52
0.460, 30
6.24, 32.4-43.2
57.9 ±2.56
0.342, 56
4.42,52.1-63.6
73.5 ±2.02
0.341,35
2.75,
69.2-77.9
42.9 ±2.10
0.297, 50
4.90,39.1^8.0
94
New Zealand Journal of
Zoology,
1997, Vol. 24
% Pyramid Valley sample
46.0-47.9 48.0-49.9 50.0-51.9 52.0-53.9 54.0-55.9 56.0-57.9
Size classes (length in mm)
• Whole sample
10-1
Pyramid Valley sample
2 -
• RJS North Island
RJS South Island
RJS
'G.minor'
I 62.0-6
RJS Stewart Island
B
X
62.0-63.9 66.0-67 9 70.0-71.9 74.0-75.9 I 76.0-79 9 ' 82 0-83.9
60.0-61.9 64.0-65.9 68.0-69.9 72.0-73.9 76.0-77.9 80.0-81.9
Fig. 10 Frequency histograms for
lengths of major limb elements of
Gallirallus australis: A, humerus;
B,
femur; C, tibiotarsus; D,
tarsometatarsus. Distributions are
given for both the whole sample
measured (solid bars) and for the
sample from Pyramid Valley
(hatched bars). Ranges of meas-
urements given in Scarlett (1972)
for North, South, and Stewart Is-
land populations, and for the puta-
tive
' Gallirallus
minor', are shown
by bars over the appropriate sec-
tions.
Range of lengths for G.
australis from Castle Rocks (type
locality for
G.
minor)
is
from Olson
(1975).
Size classes (length in mm)
Summary statistics for lengths of corresponding
limb bones of Gallinula hodgenorum arc given in
Table 7, for ease of comparison with those of
Gallirallus australis. Coefficients of variation were
lower than for the pooled samples for Gallirallus
australis, indicating a lower degree of sexual dimor-
phism in the smaller species.
Although weka were the most numerous of the
four rails at Pyramid Valley, they were not—given
their relatively large size and flightlessness—excep-
tionally common in the deposits.
Genus Porphyrio Brisson
Porphyrio hochstetteri (Meyer, 1883) (South
Island takahe)
MATERIAL: Adults
7/2;
juveniles
4/1.
Total 11/3.
Takahe are today restricted to a few valleys in
Fiordland, apart from those translocated to offshore
islands as part of a management programme. The
North Island species (Porphyrio mantelli) survived
until the late nineteenth century, but is now extinct.
The presence of takahe at Pyramid Valley conflicts
with the hypothesis of Mills et
al.
(1984) that the bird
was an obligate grassland species that declined as
forests spread after the end of the Otiran glaciation.
The habitat at Pyramid Valley was valley-bottom
podocarp-hardwood forest and mixed shrubland sur-
rounding a small lake. The nearest grassland was in
the mixed shrubland-grassland community on the
adjacent hilltops. Beauchamp & Worthy (1988)
noted that takahe were widespread in New Zealand
during the Holocene, before human colonisation,
usually in sites that were some distance from open
grasslands. The preferred habitat was probably the
floristically rich and structurally dense ecotone be-
tween forests and shrublands. The juvenile shows
that takahe bred at Pyramid Valley.
Genus Fulica Linnaeus
Fulica prisca Hamilton, 1893 (New Zealand
coot)
MATERIAL: Adults
37/5;
juveniles 7/3. Total 44/8.
Holdaway & Worthy—Pyramid Valley fossil fauna
95
Fig. 10 (continued)
Whole sample § Pyramid Valley sample
RJS North Island
8 -I
"4-
J
2
-I
RJS 'G.minor'
RJS South Island
I 87 0-88
9 I
910-92
9 I
95 0-96
9
99.0-100.9 103 0-104.9 107 0-108
9
1110-112
9
1
85 0-86
9
S9.0-90.9 93.0-94.9 97 0-98
9
1010-102
9
105.0-106.9 109 0-110
9
113.0-114.9
Whole sample
Size classes (length in mm)
I; Pyramid Valley sample
14-
12 -
10-
8 -
6-
4 -
2 -
Castle Rocks (type locality G. minor)
(U=56, Olson (1975))
RJS '6.
minor'
i
RJS North Island
RJS South Island
D
52.0-53.9 I 56 0-57.9 60.0-61.9 64.0-65.9 f
50-51.9 54.0-55.9 58.0-59.9 62.0-63.9 66 0-67.9
Size classes (length in mm)
The large, flightless New Zealand coot was not con-
fined to the vicinity of standing or running water
(Worthy & Holdaway 1994b, 1996a) and was prob-
ably widespread in the forests and shrublands around
Pyramid Valley. It bred at Pyramid Valley.
Genus Gallinula Brisson
Gallinula hodgenorum (Scarlett, 1955)
(Hodgens' waterhen)
MATERIAL: Adults
29/7;
juveniles 6/2. Total 35/9.
Pyramid Valley is the type locality for this small
flightless rail. Much of the material in the collection
was included in the type series. As with the coot,
Hodgens' waterhen was not always associated with
streams or ponds (Worthy & Holdaway 1994b). It
may have been similar to the Tasmanian waterhen
Gallinula mortierii in preferring terrestrial habitats.
In the absence of extensive grassy swards, the New
Zealand bird may have lived on more open sections
of
the
forest floor.
Rallidae sp. indet.
MATERIAL: Adults
0/0;
juveniles 1/1. Total 1/1.
Suborder APTORNITHI incertae sedis
Family APTORNITHIDAE
Genus Aptornis Owen
Aptornis defossor Owen, 1871 (South Island
adzebill)
MATERIAL: Adults 517++/10; juveniles 0/0. Total
517++/10.
The relative abundance of this large gruiform at
Pyramid Valley contrasts both with its absence on
the West Coast (Worthy & Holdaway 1993) and its
rarity at Takaka (Worthy & Holdaway 1994b). In-
deed, the Pyramid Valley sample was nearly twice
as large as the whole sample from Takaka, where all
the material came from sites of Otiran age. In the
Holocene it apparently favoured the drier mosaic of
forests and shrubland in the eastern South Island.
96 New Zealand Journal of Zoology, 1997, Vol. 24
Order CICONIIFORMES
Suborder CHARADRII
Family CHARADRIIDAE
Genus Himantopus Brisson
Himantopus novaezelandiae Gould, 1841 (black
stilt)
MATERIAL: Adults
13/1;
juveniles 0/0. Total 13/1.
Including Himantopus sp. juvenile below: 13/1; 1/
1.
14/2.
The re-identification of the stilt material as the en-
demic black stilt constitutes the first fossil record for
the species (Holdaway
1995).
No other wading birds
were present, although several other species appear
in predator sites near adjacent rivers (Worthy &
Holdaway 1996a). The proportions of the fossil stilt
differed from those of both the recent black stilt and
the pied stilt (Himantopus himantopus leuco-
cephalus),
indicating that introgression between the
two taxa may not have commenced at the time of
deposition (Holdaway 1995).
Himantopus sp.
MATERIAL: Adults
0/0;
juveniles 1/1. Total 1/1.
A
juvenile coracoid is almost certainly referable to
the black stilt; if valid, it would confirm breeding at
Pyramid Valley.
Suborder CICONII
Family ACCIPITRIDAE
Genus Circus Lacepede
Circus eylesi Scarlett, 1953 (Eyles's harrier)
MATERIAL: Adults
147+/5;
juveniles
2/2.
Total 149+/
7.
The material includes the holotype and paratypes.
None of the Circus material from Pyramid Valley
could be ascribed to the present harrier in New Zea-
land, Circus approximans. Diagnostic elements, or
parts of elements, were all attributable to Circus
eylesi. The size distributions of the elements previ-
ously attributed to C. eylesi did not conform to that
of a sexually dimorphic species, nor did those for the
supposed
C.
approximans, which is size dimorphic
(RNH unpubl. data). In effect, as the sample was
previously identified, both species would have been
represented in the deposits by one sex only—females
for
C.
approximans and unknown for
C.
eylesi. The
morphology, geographical and temporal variation,
diet, flight, and evolution of
this
huge harrier are at
present under study (RNH unpubl. data).
The relative abundance of
Circus
eylesi at Pyra-
mid Valley, and the absence of
C.
approximans,
suggests that C. approximans was either very rare
in New Zealand or absent before human-induced
environmental changes began (600-800 years ago;
McGlone 1989).
C.
approximans is also absent from
the pre-Polynesian layers at Poukawa, in which C.
eylesi is common (RNH unpubl. data).
Circus sp. indet.
MATERIAL: Adults 2/1.
Genus Harpagornis Haast
Harpagomis moorei Haast, 1872 (Haast's eagle)
MATERIAL:
Adults
22/2;
juveniles 0/0. Total 22/2(4).
The raw MNI of
2
was adjusted to 4 by an analysis
of measurements and calculated intermembral pro-
portions (Holdaway 1991b). The difference may be
taken as a measure of the relationship of MNI to
actual numbers of individuals for other taxa, but in-
sufficient data were available to apply mensural cri-
teria more generally.
Haast's eagle was the largest predator in the Pyra-
mid Valley fauna. Eagles killed at least 12 of the
moas preserved in the deposit, judging from dam-
age to pelves, particularly of Emeus
crassus.
Haast's
eagle also killed adult Dinornis giganteus (Worthy
& Holdaway 1996a).
Family FALCONIDAE
Genus Falco Linnaeus
Falco novaeseelandiae Gmelin, 1788 (New
Zealand falcon)
MATERIAL: Adults
6/1;
juveniles 0/0. Total 6/1.
The smallest of the diurnal raptors in the Pyramid
Valley deposits, the falcon is the only one extant, and
breeds in the hills to the west and south of Pyramid
Valley (Fox 1978). Analysis of prey deposits has
shown that falcons take many invertebrates as well
as mainly small diurnal vertebrates such as skinks
and open-country passerines (Worthy & Holdaway
1996a). The falcon also occasionally took larger
birds,
but Eyles's harrier and Haast's eagle were the
principal large predators in pre-human New Zealand.
Family PODICIPEDIDAE
Genus Poliocephalus Selby
Poliocephalus rufopectus (Gray, 1843) (New
Zealand dabchick)
MATERIAL: Adults
5/1;
juveniles
4/1.
Total 9/2.
The New Zealand dabchick was present in the South
Island during the first years of European settlement
but declined in the nineteenth century (Potts 1870).
Holdaway & Worthy—Pyramid Valley fossil fauna
97
It last bred south of Cook Strait in the early 1940s,
and since then has been confined to the North Island
(Turbott
1990).
The Pyramid Valley lake would have
provided a rich supply of small invertebrates and
aquatic vegetation for this or other small grebes.
Family ARDEIDAE
Genus Egretta T. Forster
Egretta alba (Linnaeus, 1758) (white heron)
MATERIAL: Adults
1/1;
juveniles 0/0. Total 1/1.
This record has not been published before, although
the specimen had been correctly identified. The
heron is the only large piscivorous bird recorded
from Pyramid Valley, but its diet is varied (includ-
ing crustaceans, insects, and birds, including
ducklings; Oliver 1955), so its presence does not nec-
essarily imply an abundance of fish. White herons
are nomadic outside the breeding season.
Order PASSERIFORMES
Suborder TYRANNI
Infraorder ACANTHISITTIDES
Family ACANTHISITTIDAE
Genus Xenicus Gray
Xenicus sp.
MATERIAL: Adults
1/1;
juveniles 0/0. Total 1/1.
This is a new record for Pyramid Valley. Xenicus
wrens have been recorded from predator deposits
nearby (Worthy & Holdaway 1996a). The very small
bones would have been difficult to detect during the
excavations. Only
X.
gilviventris is extant; it is now
confined to subalpine and alpine shrublands and
rockfields of the main ranges.
X.
longipes was found
in Canterbury in the nineteenth century, and some
of
the
few nests found were reported there by Potts
(1882).
It has apparently been extinct on the main-
land since the early twentieth century, and globally
since 1972 (Turbott 1990).
Genus Traversia Rothschild
Traversia lyalli Rothschild, 1894 (Lyall's wren)
MATERIAL: Adults
1/1;
juveniles 0/0. Total 1/1.
This is a new record for Pyramid Valley. The spe-
cies is common in predator deposits nearby (Wor-
thy & Holdaway 1996a). It survived into the
European period only on Stephens Island in Cook
Strait, where it became extinct as a result of habitat
destruction and predation in or shortly after 1894.
The species is common in fossil deposits on both
main islands (Worthy & Mildenhall 1989; Worthy
& Holdaway 1993, 1994b).
Suborder PASSERI
Parvorder CORVIDA
Family MELIPHAGIDAE
Genus Anthornis Gray
Anthornis melanura (Sparrman, 1786)
(bellbird)
MATERIAL: Adults
3/2;
juveniles 0/0. Total 3/2.
A new record for Pyramid Valley, the bellbird is still
common in the shrubland and forest remnants of
North Canterbury and visits wooded gardens near the
site.
It is the only honeyeater common east of the
ranges in Canterbury (Bull et al. 1985).
Genus Prosthemadera Gray
Prosthemadera novaeseelandiae (Gmelin, 1788)
(tui)
MATERIAL: Adults
39/7;
juveniles 3/2. Total 42/9.
The tui was abundant in the deposit. This may re-
flect its comparatively large size, but there were
fewer of the larger raven (Corvus moriorum) in the
deposit, so the numbers may reflect those in the liv-
ing fauna more closely than preservation and collec-
tion biases would suggest. Tuis are now very rare in
the eastern South Island (Bull et al. 1985) (see note
added in proof
p.
121).
Family PETROICIDAE
Genus Petroica Swainson
Petroica australis (Sparrman, 1788) (New
Zealand robin)
MATERIAL: Adults
8/3;
juveniles 0/0. Total 8/3.
Robins are now rare in North Canterbury, the clos-
est populations to Pyramid Valley being in the beech
forests to the north and west (Bull et al. 1985). The
presence of three individuals of such a small species
suggests that it was abundant near the lake. Robins
can live in tall forest or in marginal and low scrub.
Petroica macrocephala (Gmelin, 1789) (New
Zealand tomtit)
MATERIAL: Adults
1/1;
juveniles 0/0. Total 1/1.
This is a new record for Pyramid Valley, the speci-
mens having previously been listed as Petroica sp.
The tomtit is one of the most abundant and wide-
spread of
the
small endemic forest birds remaining
in the South Island (Bull et al. 1985).
98
New Zealand Journal of
Zoology,
1997, Vol. 24
Family CORVIDAE
Genus Mohoua Lesson
Mohoua novaeseelandiae (Gmelin, 1789)
(brown creeper)
MATERIAL: Adults
3/2;
juveniles 0/0. Total 3/2.
This is a new record. Brown creeper are still wide-
spread in forests and shrubland to the west of Pyra-
mid Valley and elsewhere in Canterbury (Bull et al.
1985),
but not near the site
itself.
Its bones are very
small; the presence of two individuals indicates that
it was probably common near the lake.
Genus Corvus Linnaeus
Corvus moriorum Forbes, 1892 (New Zealand
raven)
MATERIAL: Adults
5/1;
juveniles 0/0. Total 5/1.
Only one individual was represented, which suggests
that New Zealand's largest passerine was uncommon
to rare near Pyramid Valley. Although widespread,
the raven is seldom a major component of fossil fau-
nas (e.g., Worthy &.Holdaway 1993, 1994b).
Family CALLAEATIDAE
Genus Callaeas J. R. Forster
Callaeas cinerea (Gmelin, 1788) (South Island
kokako)
MATERIAL: Adults
49/7;
juveniles 1/1. Total 50/8.
South Island kokako are now extremely rare, if not
extinct. The species was abundant in Canterbury
forests until the 1860s (Potts 1882), and seems to
have been common at Pyramid Valley. Kokako have
also been recorded from predator sites nearby (Wor-
thy & Holdaway 1996a).
Genus Philesturnus Geoffrey St.-Hilaire
Philesturnus carunculatus (Gmelin, 1789)
(South Island saddleback)
MATERIAL: Adults
14/4;
juveniles 1/1. Total 15/5.
As with the kokako, the saddleback survived in good
numbers in Canterbury until the 1860s, especially in
the podocarp forests on Banks Peninsula (Potts
1882).
It was probably common at Pyramid Valley;
it is also present in predator sites near Pyramid Val-
ley (Worthy & Holdaway 1996a) (see note added in
proof
p.
121).
Family TURNAGRIDAE incertae sedis
Genus Turnagra Lesson
Turnagra capensis (Sparrman, 1787) (South
Island piopio)
MATERIAL: Adults
11/2;
juveniles 0/0. Total 11/2.
The South Island piopio occurs frequently in preda-
tor deposits nearby (Worthy & Holdaway 1996a),
and seems to have been reasonably abundant in the
area. Indeed, the piopio seems to have been more
common in the eastern South Island than elsewhere.
Its favoured habitat may have been in the mixed for-
est/shrubland vegetation, if its distribution among the
predator sites indicates more than individual prey
preferences (Worthy & Holdaway 1996a).
Family SYLVIIDAE
Genus Bowdleria Rothschild
Bowdleria punctata (Quoy & Gaimard, 1830)
(fernbird)
MATERIAL: Adults
3/2;
juveniles 0/0. Total 3/2.
This is the first record of the fernbird for Pyramid
Valley. The three bones were also the first fossils of
it to be collected. The fernbird has since been found
in several predator deposits, and the first published
record is in Worthy & Holdaway (1995). Fernbirds
are usually associated with low shrubland and
fernland (Oliver 1955), but they are also recorded
from tall forest (O'Donnell & Dilks 1986).
Class MAMMALIA
Order CHIROPTERA
Family MYSTACINIDAE
Genus Mystacina Gray
Mystacina robusta Dwyer, 1962 (greater short-
tailed bat)
MATERIAL: 6/2.
Although Scarlett (1969) recorded long-tailed bats,
all the material examined was referable to the greater
short-tailed bat (Worthy et al. 1996). Greater short-
tailed bats are more abundant than Mystacina
tuberculata in fossil deposits in the eastern South
Island (Worthy & Holdaway 1995, 1996a). In the
South Island areas surveyed to date, M. tuberculata
outnumbers M. robusta only in the deposits on
Takaka Hill, in northwest Nelson, which were laid
down under a vegetation dominated by Nothofagus
beech forest (Worthy & Holdaway 1993, 1994b).
DISCUSSION
General
The Pyramid Valley fauna is important because it
provides a benchmark for the species richness of
large vertebrates in the downlands of the eastern
Holdaway & Worthy—Pyramid Valley fossil fauna
99
South Island before the impact of human colonisa-
tion. The deposit was laid down in a relatively short
period of
the
late Holocene, and the unusual condi-
tions favoured the preservation of many individual
skeletons, especially of
moas.
The remains of
juve-
niles provide breeding records for over 22 species
in the area. Although excavation techniques were not
optimised for recovering smaller
bones,
enough were
collected to give a fair sample of most taxa over
100 g. The fossil avifauna had a high species rich-
ness but low numbers of individuals of most taxa.
Waterfowl and large ground birds dominated both
the species list and the number of individuals, be-
cause the glutinous lake bed preferentially trapped
heavy animals, and because of collection biases.
The collection is of taxonomic importance: Pyra-
mid Valley is the type locality for three species, and
type material for another was collected there. The
relatively large samples of species such as
Gallirallus australis and Malacorhynchus scarletti
have allowed further analysis of
the
range of varia-
tion in New Zealand taxa. The number of individual
moa first facilitated some major progress in under-
standing the diversity of the group (Cracraft 1976a,
b),
and now has allowed insights to be gained into
intraspecific variation, including sex and age distri-
butions. From the results, inferences can be made
about the breeding systems of at least two species
from different subfamilies.
The species list grew over time and with excava-
tion effort. Apart from the effects of sample size,
important reasons for changes were the reduction in
number of moa species recognised since the first lists
were compiled, the description of new species, and
the merging of nominal species such as Gallirallus
minor. The present list includes species identified in
the collections but never included in a published list,
others that had been misidentified and listed under
another name, and some that had remained uniden-
tified. Sampling biases are illustrated by the goose,
of which a specimen was found during the first se-
ries of excavations but none thereafter. The increase
in the species list with time and collecting effort in-
dicates that faunas known from small samples or a
limited excavation in a large site are unlikely to be
representative of the regional fauna.
The absence of some groups such as obligate
piscivores—cormorants, or the duck Mergus—sug-
gests that conditions did not favour fish-eating birds.
At least one species offish, possibly an eel, has been
reported (Scarlett 1969). The waterbirds present in-
cluded dabbling and sifting ducks, a small grebe, a
stilt, a heron, and a gallinule, which would have
exploited the abundant
mollusc,
ostracod, and insect
faunas reported by Deevey (1955).
Chronology
Gregg (1972) suggested that the deposition of the
moas in the gyttja antedated the layers immediately
above because the stratification was apparently un-
disturbed. However, the photograph in Gregg (1972,
fig. 2) does not show any bone where stratification
is most obvious. The presence of bones in the over-
lying peat indicates, however, that entrapment was
not confined to the lake period.
As noted by Burrows (1989), the range of
age
of
deposition based on bone dates and those from as-
sociated gizzards seems to be much less than that
required for the development of the sediments based
on dating of peat and plant macrofossils. A short
period of bone deposition would also mean that the
deposit was derived from a single fauna, rather than
from a changing assemblage. Major changes in
faunal composition would be less likely over 2000
years of relatively stable climate in the late Holocene
than over the same period during the glacial transi-
tion. Relatively few dates have been obtained, so the
actual period of deposition remains uncertain. It is
certain, however, that the deposit formed during the
forested period of the late Holocene (6000-1000
radiocarbon years b.p.) (Moar 1970; Worthy &
Holdaway 1996a).
It is assumed that the available dates on moas re-
flect the period of deposition for the rest of the fauna.
However, no other taxa from the fauna, especially
smaller taxa that would have died and fallen into the
lake,
have been dated directly. Williams et
al.
(1993)
suggested that there was considerable bioturbation
of the deposit by struggling moas, and that there is
evidence that the area of the deposit was forested.
They do not cite sources in support of their conten-
tion, but a photograph of the stratigraphy, presum-
ably taken during the 1973 excavation, was
presented. No other authors have mentioned
bioturbation, and none is obvious in transparencies
of the walls of test pits (Cushman Murphy 1947-49,
AMNH).
Vegetation
The Pyramid Valley Swamp deposit sampled the
fauna of a valley-bottom, lakeside ecosystem. Pyra-
mid Valley is in a seasonally dry region of North
Canterbury which is exposed to strong winds and ex-
tremes of temperature—hot, dry north-westerlies in
summer, and cold southerlies that bring periodic
snowfalls in winter. Burrows et al. (1981) and
100
New Zealand Journal of Zoology, 1997, Vol. 24
Burrows (1989) suggested that the lake was sur-
rounded by a mixed podocarp forest dominated by
matai (Prumnopitys taxifolia), with many
broadleaved trees in the canopy and subcanopy.
Burrows et al.
(1981:
321) quoted Moar (1970)
in support of the view that the Nothofagus forest
implied by the abundant fuscospora-type pollen
"could have been distant from the site". Moar (1970:
460) noted, however, that Nothofagus forest was the
dominant vegetation in the region, and that "the for-
ests were similar to the remnant Nothofagus forests
... on the foothill ranges adjacent to the Canterbury
Plains", which are the only forests available for com-
parison today. He noted that
N.
menziesii need have
been no closer than the existing stands c. 32 km to
the west, but the reference to trees producing
fuscospora-type pollen implies that Nothofagus was
common in the forest around Pyramid Valley.
Matai may well have been the dominant tree in
valley bottoms, but it is likely that there was pro-
nounced altitudinal stratification in forest type, and
that beech was dominant higher on the valley slopes
and in more exposed sites. Although the climate at
Pyramid Valley was probably milder 3000^4000
years ago (Burrows et al. 1981), seasonal droughts
and periodic southerly storms would have confined
tall forest to valley bottoms. On the most exposed
ridges and south-facing slopes there were probably
patches of open shrubland and grassland. The
Coturnix in the deposit supports the pollen-based
vegetation reconstruction that suggests some
shrubland nearby (Moar 1970; Burrow et al. 1981;
Burrows 1989), but its rarity suggests that the
shrubland around the lake margins was not exten-
sive.
It may have been confined to exposed ridges.
A reconstruction of Quaternary vegetation (Harris
1969) is too coarse grained to permit meaningful
interpolation of the vegetation around Pyramid Val-
ley, but does emphasise that the patterns and com-
position of
the
vegetation were far from static. The
sections provided by Molloy (1969) suggest that the
valley was at or near an ecotone between lowland
podocarp forest and submontane podocarp-beech
forest. Complexity of the vegetation near the site was
probably a result of the dissected terrain to the south,
and of exposure to drought and strong winds. As a
result, there was almost certainly marked zonation
of vegetation in response to altitude, shelter, and
moisture availability. These differences over very
short geographical distances are reflected in the pres-
ence of
a
variety of vertebrate taxa typical of differ-
ent habitats.
Faunal diversity
The measures of diversity employed in this study
show that the Pyramid Valley fauna was one of the
most species rich of the large fossil sites known at
present.
Diversity indices
The use of diversity indices allows characteristics of
the fossil faunas such as richness, relative dominance
by taxa or groups, and different taphonomies to be
compared directly more easily than by verbal de-
scriptions. Taphonomic factors mean that indices—
apart from species richness itself-—for the faunas are
not strictly comparable to data from living faunas
(Holdaway
1990b),
but the indices are useful in com-
parisons of faunas from different areas, and of dif-
ferent times, if the taphonomic effects are taken into
account.
Comparisons of the indices for several faunas, and
in particular the rarefraction curves, gave estimates
for the minimum number of specimens that must be
identified before the species richness of the living
fauna of an area at
a
particular period can be assessed
with any accuracy. MNI of 300 or more are neces-
sary, preferably from a single site. Pooling of data
from several sites, including several different
taphonomies, will, however, give comparable val-
ues.
Guilds
The guilds suggested by Atkinson & Millener (1991)
for North Canterbury include ground herbivores,
arboreal herbivores, frugivore/herbivores and
frugivore/"nectivores", "nectivores", aquatic insec-
tivores, ground insectivores, smaller arboreal insec-
tivores, larger arboreal insectivores, and major
predators of vertebrates. These are broadly similar
to those recognised here (Table 5) and in Holdaway
& Worthy
(1996),
but there are important differences
in detail.
Ground herbivores
Atkinson & Millener (1991) accepted Dinornis
novaezealandiae, D. struthoides, and Tadorna
variegata as members of the ground herbivore guild.
Of these, the first was not present in the fauna (Wor-
thy & Holdaway 1996a), the second was very rare,
and the third is included here as an omnivore. Takahe
have now been found at other sites in North Canter-
bury (a ledge on a cliff at Weka
Pass,
The Deans slip,
and at Waikari Cave; Worthy & Holdaway 1996a),
and there is now no reason to doubt that they used
Holdaway & Worthy—Pyramid Valley fossil fauna
101
the forests, and especially the dense shrublands,
rather than swampland. Hodgens' waterhen has a
similarly broad distribution in sites away from
wetlands such as the Pyramid Valley lake margin,
and probably ranged freely in the shrublands.
The goose was very rare in North Canterbury.
Only two specimens are known from Holocene de-
posits there (Worthy & Holdaway 1996a), so it was
not an important part of the terrestrial herbivore
guild. It appears to have preferred more open grass-
land or shrubland, and was more abundant farther
south in the Holocene (THW unpubl. data). The bird
found at Pyramid Valley may have been from a small
population in the low shrubland and patches of
grass-
land on the ridges to the south of the valley.
Arboreal herbivores—frugivores andfolivores
The parakeets are included here as frugivores and
folivores, but perhaps should be considered as om-
nivores as they include a quantity of animal mate-
rial in their diet (Nixon 1994). On the Chatham
Islands the diet of red-crowned and Forbes' para-
keets varied according to season, seeds being most
important in winter and late summer, and leaves
being very important in spring (Nixon 1994). Inver-
tebrates formed only a small part of the diet, mainly
in winter and spring. On Little Mangere Island,
Forbes' parakeet took a higher proportion of inver-
tebrates, whereas red-crowned parakeets on South
East Island favoured leaves (Taylor 1975). Parakeets
eat a variety of foods, but it appears that leaves are
a staple. In the presence of a range of insectivores
they may rely more on leaves.
We agree with Atkinson & Millener (1991) that
it is too early to judge the role of kokako in the South
Island. There are conflicting reports of where kokako
were most common last century in the South Island
(Worthy & Holdaway 1994b). The ecology of South
Island kokako may have differed from that of the
North Island species in ways that are still unclear.
The more rapid decline of South Island kokako, in
contrast to the situation with most other endangered
species whose main refuges were in the South Island,
also points to differences in habitat or behaviour that
may have made the southern populations more vul-
nerable.
Nectarivores
The nectarivore guild at Pyramid Valley included
both South Island species of honey eater. At present
the tui is extremely rare or absent over large areas
of the eastern South Island. The reduction in nectar-
bearing trees and shrubs with deforestation may have
lowered the availability of food below the economic
limit for the larger species. In addition, as Atkinson
& Millener (1991) point out, some of the important
northern nectar sources were absent from North
Canterbury.
Aquatic insectivores
Atkinson & Millener (1991) placed the brown teal
and Finsch's duck in this guild. We do not agree (see
below). The main aquatic insectivorous bird of New
Zealand streams is the blue duck, which was absent
from the North Canterbury deposits (Worthy &
Holdaway 1994a,
b).
It was not, however, restricted
to fast-flowing streams and rapids before humans
arrived (Worthy & Holdaway 1994b), and so other
reasons for its absence must be postulated. The wa-
terfowl fauna at Pyramid Valley is species rich (for
New Zealand) in the absence of blue duck, so an-
other species, or combination of species, may have
excluded it ecologically.
Arboreal/terrestrial insectivores and terrestrial
omnivores
These guilds include the
species
placed
in the
ground
insectivores of Atkinson & Millener (1991). The
brown teal was only loosely associated with open
water, and is included here as a terrestrial omnivore
(Worthy & Holdaway 1994b). As noted elsewhere,
the kiwi at Pyramid Valley may have been an
undescribed taxon allied
to
the brown
kiwi.
The great
spotted kiwi was certainly not in the ground insecti-
vore guild east of the beech forests of the main
ranges. The snipe was
rare,
as
were
the
thick-thighed
wren and Lyall's wren.
The rifleman was not present in the Pyramid Val-
ley deposit, but one of
the
Xenicus species was, and
both genera are present in owl deposits nearby (Wor-
thy & Holdaway 1996a). Both species of Petroica
were present near Pyramid Valley, as were both
Mohoua, but only
M.
novaeseelandiae was found in
Pyramid Valley
itself.
The absence of yellowhead
at Pyramid Valley may indicate that beech was ab-
sent from the lake fringe. Yellowheads were found
in owl sites at higher levels (Worthy & Holdaway
1996a), where beech was almost certainly present.
Larger arboreal insectivores
We do not find much evidence for a guild of large
arboreal insectivores. The absence of kingfisher re-
mains from Pyramid Valley, and their rarity in other
deposits, does, however, support the contention of
Atkinson & Millener (1991) that this species may be
one of the recent arrivals that became established
102
New Zealand Journal of Zoology, 1997, Vol. 24
inland only after major vegetation changes had oc-
curred. It is in coastal cave faunas from Kaikoura
(authors' unpubl. data), and may have been restricted
to coastal habitats in the presence of the complete
Holocene fauna.
Atkinson & Millener (1991) regarded the laugh-
ing owl as a larger arboreal insectivore. Worthy &
Holdaway
(1993,
1994a, b, 1996b) and Holdaway
& Worthy (1996) have demonstrated that the owl
was a major predator of small to medium-sized ver-
tebrates as well as larger flightless invertebrates. It
belongs in the nocturnal predator guild. The owl was
also primarily a bird of closed, drier forests, and
hunted very near or on the ground (Holdaway &
Worthy 1996).
The morepork is primarily an insectivore at
present, but is known to take larger vertebrates when
they are abundant (Anderson 1992); it too belongs
in the nocturnal predator guild. The owlet-nightjar
is included as a predator because, although its
smaller Australian relative feeds on large flying and
terrestrial insects, the New Zealand bird was much
larger and could have taken lizards and frogs as well.
The remaining large arboreal insectivore, the long-
tailed cuckoo, is rare as a fossil and is unknown from
deposits in North Canterbury. It is largely nocturnal.
The major predators of vertebrates were Haast's
eagle, Eyles's harrier, and the New Zealand falcon.
The adzebill probably fits better in the terrestrial
omnivore guild. Although the adzebill probably took
large invertebrates and small vertebrates up to the
size of tuatara and small petrels (Holdaway 1989),
it probably also consumed substantial amounts of
vegetable matter. The eagle and harrier were char-
acteristic of open forest/shrubland mosaics in the
Holocene and Otiran, and have not been found at any
site where the contemporaneous vegetation was wet,
dense forest typical of the present-day West Coast
or the North Island. Contra Atkinson & Millener
(1991),
the explanation for the eagle's absence from
the North Island in the Holocene (Holdaway 1991b)
is probably simply that there was insufficient of the
preferred habitat to support a self-sustaining eagle
population.
Taphonomy
In swamp and lake deposits in New Zealand the fos-
sil sample is typically biased towards large flight-
less taxa. A tendency to become bogged while
looking for food or water increases the likelihood of
being incorporated over the basic stochastic event of
individuals of any species dying in or over the site.
The lake shore habitat at Pyramid Valley was
therefore sampled in a different way to that of the
rich sites that developed in caves and tomos, which
act
as
random pit traps for flightless or ground-dwell-
ing taxa. Apart from eagles and other birds of prey
being attracted to trapped but living prey species,
caves generally do not attract their victims.
Most of the waterfowl would have been attracted
to the lake. Moa may have been attracted to the wa-
ter source as well as the vegetation. Gizzard analy-
ses might suggest that moas were commonest at
Pyramid Valley in late summer, when surface wa-
ter was probably scarce in the area. Alternatively,
they may have been attracted mainly by the abun-
dance of favoured fruits in the lakeside shrubland.
For large birds the primary "sampling" factor was
miring, and moas seeking water in the summer
drought period, when the lake would have been at
its lowest, would have been exposed to greatest risk,
the largest birds being mired preferentially.
Birds such as the pigeon, kaka, and songbirds
were preserved in the lake sediments after apparently
dying naturally over or near the site. Some may have
fallen or been washed in as carcases, or have been
trapped by sticky sediments while drinking at the
edge of pools during dry weather. No other mecha-
nism, such as catastrophes, is necessary to explain
their presence in the sediments.
The flying bird best represented in the Pyramid
Valley deposit was the New Zealand pigeon. It was
almost certainly attracted to the edge of the lake by
the abundance of matai fruit, rather than by the wa-
ter source. Water was important, however, because
the last water sources to fail would then be the foci
for many species. At the same time, the conditions
for entrapment would be ideal, with large margins
of glutinous sediment perhaps masked by a mat of
vegetation.
Waterfowl are to be expected in a lake deposit;
the larger taxa, with the largest requirements for
drinking water, would have been both the most regu-
lar visitors and the most likely to get bogged. The
bias towards larger individuals may have affected
within-species entrapment rates, resulting in artifi-
cially skewed sex ratios in the fossil sample given
that both sexes of each species would presumably
have been attracted to the water source equally. The
sex ratio of the largest taxon (Dinornis giganteus)
was skewed but to an extent that differences in pro-
pensity to mire were unlikely to be a major cause.
Haast (1872) and Duff (1949) suggested that ea-
gles were trapped while attempting to either kill a
trapped moa or while scavenging from a moa car-
case.
Archey (1941) suggested that the absence of
Holdaway & Worthy—Pyramid Valley fossil fauna
103
crania and cervical vertebrae from many of the moa
skeletons in swamps may result from their removal
by eagles. It is, however, at least as likely that these
body parts were the first to be disarticulated during
decay and were separated from the remainder of the
carcase (Eyles 1955).
There is clear evidence that eagles killed moas at
Pyramid Valley. Some, at least, of the eagles in the
deposit may have been trapped while killing a moa.
The eagle's talons are particularly long (Holdaway
1991 b) and may have been readily caught in tangled
vegetation around the lake margin. The birds may
have been injured during the struggle with a moa.
Large raptors are particularly abundant at Rancho La
Brea, southern California, where asphalt seepages
provided a treacherous surface for walking (Howard
1932).
The birds there were attracted by trapped liv-
ing or dead mammals. Drowning in open water is
unlikely to have been a common cause of death for
eagles at Pyramid Valley, because several taxa, in-
cluding harriers, can rise from a water surface even
while carrying a load (Oliver 1955; Brown &
Amadon 1968).
Damage during predation was not the only fac-
tor contributing to the presence of eagles at Pyramid
Valley. The principal difference between entrapment
of eagles in swamps or lakes and in caves is that
swamps and lakesides are normal habitats for eagles,
whereas caves are not. Living bait was almost cer-
tainly necessary for eagles to be trapped in most
caves,
but presence of eagles near swamps or lakes
is sufficient to explain their inclusion in the depos-
its.
The rates of deposition calculated for the larger
taxa, although admittedly crude, argue against any
catastrophic mode of entrapment and deposition. The
process may have been episodic or
cyclic,
with peaks
during the summer and especially in prolonged
droughts, but the incorporation of fossils was not the
result of cataclysmic events.
Features of the fauna
Reptiles
It is impossible at the moment to determine which
of the two species of tuatara currently recognised
was present in North Canterbury or, indeed, whether
the two forms were once sympatric. Similarly, it is
difficult to judge the former abundance of tuatara in
the area from the few remains present, and hence its
importance as a ground predator is unknown. Sur-
viving populations prey on large invertebrates and
small vertebrates, including adults and young of
burrow-nesting petrels. Records from other sites
(Worthy & Holdaway 1996a) show that large inver-
tebrates and several species of petrel were present
nearby. The range of prey was probably far greater
on the mainland than on the present island refuges.
Small geckoes seem to have been ubiquitous in
pre-human New Zealand. Their remnants are unu-
sual in lacustrine deposits. The record from Pyramid
Valley
is
supplemented substantially
by
that from the
many predator deposits in the area, in which liz-
ards—both geckoes and skinks—are abundant
(Worthy & Holdaway 1996a).
Kiwis
Kiwis were rare in the Pyramid Valley deposit. Only
one species, possibly one of the brown kiwis, was
identified. In the eastern South Island the larger kiwi
(as opposed to A. owenii) was significantly smaller
than its counterparts in the west or north of the South
Island. Kiwi abundance may have been limited by
the great variety of medium-sized ground birds, or
perhaps the area was not favoured kiwi habitat. The
absence of
snipe
in the lake deposits, and therefore
presumably around the lake
itself,
when they were
present nearby at higher altitudes, implies that the
habitat available may have been more finely parti-
tioned when there were more species in the fauna.
In addition, ground-probing species may have been
restricted by the absence of soft substrate during
summer droughts, which could have made it diffi-
cult to feed. Substrate firmness was shown not to
limit the feeding success of snipe on The Snares is-
lands south of New Zealand, but the substrate there
was never dry or dusty (Miskelly 1989). Other spe-
cies that depend on subsurface invertebrates, such
as the rook Corvusfrugilegus, are restricted by food
availability when the soil dries out in summer
(Murton 1969).
Moas
Numerically, the fossil fauna was dominated by the
moas.
Although five of the nine South Island taxa
were represented, Dinornis giganteus and Emeus
crassus dominated, and the other two emeids—
Euryapteryx geranoides and Pachyornis ele-
phantopus—were minor components, in about equal
proportions. The one specimen of D. struthoides
indicates its extreme rarity near Pyramid Valley.
Each species' representation is probably a conse-
quence of both the favoured habitat of each taxon
and its propensity for being trapped. The numerical
dominance of
D.
giganteus and£. crassus suggests
that the species-rich shrubland and forest around the
104
New Zealand Journal of Zoology, 1997, Vol. 24
lake was good habitat for
both.
They are the charac-
teristic moas of the eastern lowland South Island, and
were replaced elsewhere by E. geranoides and P.
elephantopus. E. geranoides was especially charac-
teristic of the steeper hill country around Pyramid
Valley, and not of the plains and valley floors (Wor-
thy & Holdaway 1996a).
Possible differences between the sexes in abun-
dance support the hypothesis based on sexual size
dimorphism that at least some moas had non-mo-
nogamous mating systems. As noted above, the al-
ternative explanation—that the sexes had different
preferred habitats—can be dismissed because such
a system is absent in living ratites, and because the
sexes are likely to have had a similar large demand
for water.
If
the
inferences on which sex was the larger, or
at least which had the longer tibiotarsus, are correct
then it appears that females of/), giganteus outnum-
bered
males.
Species whose adult sex ratio is skewed
in favour of females are usually polygynous (e.g.,
ostrich Struthio camelus and emu Dromaius novae-
hollandiae (Marchant & Higgins 1990), which in-
habit vegetation no denser than savannah
woodlands). The forest-dwelling cassowaries
Casuarius are monogamous (Marchant & Higgins
1990).
It is likely that D. giganteus required more
open vegetation than some of the other lowland
moas.
Waterfowl
A comparison of the species composition in the fossil
fauna and the present waterfowl fauna indicates the
great changes that have occurred over the past 1000+
years,
and provides clues to the habitat requirements
of both living and extinct forms. The brown
teal
Anas
chlorotis, one of the rarest of the surviving ducks,
was the most abundant anatid at Pyramid Valley, as
it was elsewhere (Worthy & Holdaway 1994b). It
lived in terrestrial habitats as well as on lakes and
so would have been widespread in the forest and
shrubland around Pyramid Valley as well as on the
lake
itself.
The closely related Campbell Island teal
Anas aucklandica nesiotis survives on Dent Island
in wet vegetation on
a
peat and moss substrate, where
it forages for invertebrates (Marchant & Higgins
1990).
Early records suggest that the brown teal was
abundant in kahikatea Dacrycarpus dacrydioides
swamp-forest, and that it foraged at night (Oliver
1955).
Entrapment of birds found as fossils in caves
probably occurred mainly during its period of activ-
ity (i.e. during its night-time forays into the forest).
Most of its original habitat has been cleared or
drained, and the species has declined to a few rem-
nant populations (Marchant & Higgins 1990).
After brown
teal,
the other ducks in order of abun-
dance were Scarlett's pink-eared duck, Finsch's
duck, paradise shelduck, New Zealand scaup, grey
teal, and the grey duck. Scarlett's pink-eared duck
Malacorhynchus scarletti was a filter-feeder. The
present occupant of that niche, shoveler (Anas
rhynchotis), was not present, although limnological
conditions would have been ideal for it (i.e. fertile
hinterland, much organic input, abundant freshwa-
ter invertebrates such as ostracods). There is no rea-
son to believe that shoveler did not reach New
Zealand in the past at least as often as they do now
(Falla et
al.
1979). That the shoveler
was
not a breed-
ing resident suggests that new arrivals were out-
competed by the large Scarlett's pink-eared duck, at
least on small inland ponds. The living Australian
pink-eared duck is smaller than the shoveler and
prefers shallow water with abundant insects and
crustaceans for food (Marchant & Higgins 1990).
Shoveler prefer large permanent wetlands with
fringing vegetation and areas of open water
(Marchant & Higgins 1990). They occur with pink-
eared ducks in Australia, but open wetlands were rare
in pre-human New Zealand (mainly swamp forest
communities; M. McGlone pers. comm.), and there
may have been insufficient habitat for shoveler be-
fore the land was cleared. Although the shoveler may
have been present on saline coastal lagoons, which
are not favoured by pink-eared ducks, shoveler are
unlikely to have been present in New Zealand be-
fore human settlement. There are no confirmed
records older than 1000 years.
The grey duck is the most abundant small duck
in New Zealand
today.
Scaup live and breed in small
ponds today, so long as there is sufficient food and
cover, so it is not surprising that both species were
present in some numbers at Pyramid Valley. That
grey teal were also present and breeding
is,
however,
surprising. Their presence shows that the species is
not a recent colonist (Turbott 1990; Millener 1991),
and that this small nomadic duck
was
part of
the
New
Zealand fauna before humans arrived. The Pyramid
Valley juveniles are the earliest New Zealand breed-
ing record of grey teal.
The single goose Cnemiornis calcitrans indicates
that the species was probably rare near Pyramid
Valley. The bones of such a large bird should have
been readily preserved and more obvious to collec-
tors than those of smaller taxa, so the small sample
is probably significant.
The waterfowl in the Pyramid Valley fauna were,
Holdaway & Worthy—Pyramid Valley fossil fauna
105
so far
as
is known, primarily vegetarian or insectivo-
rous.
Three—brown teal, Finsch's duck, and extinct
goose—were predominantly terrestrial (Worthy &
Holdaway 1993,1994b). The others—New Zealand
scaup, pink-eared duck, grey teal—fed in the lake.
The grey duck probably exploited both land and
water resources.
None of the other three endemic ducks known
from the South Island—South Island merganser
(Mergus sp.), musk duck (Biziura delautouri), blue
duck {Hymenolaimus malacorhynchos}—was
present. The merganser was presumably mainly
piscivorous (Falla et al. 1979), as are extant species
(Delacour & Scott 1956). It may have been coastal,
like other mergansers. Mergansers were recorded
from streams and sheltered coastal waters at the
Auckland Islands (Falla et al. 1979). Elsewhere,
merganser bones have been recorded in coastal de-
posits at Grassmere, Tokerau Beach (Far North),
Delaware Bay (Nelson), and Native Island (Stewart
Island) (THW unpubl. data).
The extant Australian musk duck prefers swamps
with deep water and dense vegetation; the Pyramid
Valley lake would have been too small and open to
support a resident population.
That the blue duck was found in caves such as
Hobsons Tomo on Takaka Hill, kilometres from any
surface water (Worthy & Holdaway 1994b), sug-
gests that the habitats recorded in European times for
some taxa do not represent all habitats occupied be-
fore Polynesians arrived, 700—800 years ago. Blue
ducks were absent from Pyramid Valley, and have
not been recorded from coastal lagoon sites either,
but their bones may have been confused with those
of other ducks. Blue ducks do visit river mouths,
where early explorers found them (Beaglehole
1974).
In the early years of European settlement blue
ducks lived on lowland streams such as the Omaka,
near Blenheim (M. Newman pers. comm. to RNH)
and the Opihi, near Raincliff (THW unpubl. data).
It is probably significant that the streams at such sites
included riffles. The disappearance of the duck from
these and similar lowland sites was a result of hu-
man hunting and predation by cats, dogs, and prob-
ably mustelids. Only on the steeper mountain
streams are there now enough refuges from preda-
tors.
Parrots
There were nearly as many keas as there were kakas.
There were
no
juvenile keas, but this does not mean
that keas did not breed in lowland Canterbury since
only one juvenile kaka (an undisputed lowland
breeder) was represented among 138 bones.
The
Cyanoramphus
material could include any or
all of
the
species. Parakeet bones were minor com-
ponents of the collection, but their presence indicates
that most of the mainland parrots were sympatric at
Pyramid Valley.
Rails and adzebill
The small gallinule and the coot were the common-
est rails after the weka. Both bred at the lake. One
of the two takahe Porphyrio hochstetteri was a ju-
venile, indicating that, although it was not abundant,
the takahe definitely bred near Pyramid Valley in the
late Holocene. The pukeko
Porphyrio melanotus
was
absent. Pukeko are now so abundant and widespread
that its recent arrival was not previously suspected.
As with the shoveler, the pukeko was probably a
regular visitor to New Zealand, but could not estab-
lish until suitable habitat was produced after human
settlement and the other terrestrial rails were extinct.
A minimum of 10 adzebills suggests that this
large, flightless gruid was relatively common near
Pyramid Valley, in contrast to its absence on the
West Coast and around Takaka at the same time
(Worthy & Holdaway 1993, 1994b).
Predators
The birds of prey included two species of accipitrid
(eagle, harrier), a falcon, two owls, and an owlet-
nightjar. The extant
Circus
approximans was absent,
and seems to have colonised New Zealand after the
larger species went extinct.
Haast's eagle was the principal predator of the
large herbivores (moas, and probably adzebills,
takahe, and kakapo) present in the area. The goose
was probably too rare near the lake to be a signifi-
cant item of diet
there.
Interactions between Haast's
eagles and moas have been speculated upon for
many years. Pyramid Valley was an ideal site in
which to look for evidence for the predation of live
birds predicted by Holdaway (1991b). THW found
12 moa pelves with major lesions attributable to
damage by eagle
claws.
The damage was in the same
place and had the same disposition as predicted
(Holdaway 1991b). Similar damage to pelves was
also recognised on material that had been collected
from Kapua in the late nineteenth century, and on
material collected by us from Glencrieff Swamp, less
than 2 km from Pyramid Valley (Worthy &
Holdaway 1996a). The moa species concerned in-
cluded
Emeus
crassus, Euryapteryx geranoides, and
Dinornis
giganteus.
The damage—large perforations
and rents—was consistent with the eagle having
gripped the live moa. There would have been no need
to inflict wounds on a dead moa. The eagle was
106
New Zealand Journal of Zoology, 1997, Vol. 24
adapted to killing and feeding on very large prey
(Holdaway 1991b; RNH unpubl. data).
The other accipitrid, Eyles's harrier, was over
twice
as
heavy as the living harrier (Holdaway 1989)
and had modified wing proportions, convergent on
those of forest-hunting goshawks (Accipiter) (RNH
unpubl. data). It was large and powerful enough to
have taken the Nestor parrots, pigeons, waterfowl,
and the large rails, as well as large and medium-sized
songbirds.
The ratios of numbers of diurnal predators to their
prey in the deposit (2.05—3.32%) are within the range
for fossil predatory endotherms (1.5-9%; Bakker
1986;
Paul 1988). The relative representation of taxa
in the deposit therefore more or less reflects relative
abundances in the living assemblage at Pyramid
Valley. Modern mammalian (i.e. endotherm) com-
munities have lower predator/prey ratios
(0.2—1.5%;
Bakker 1986), perhaps because of disturbance by
humans. Bakker (1986) used ratios of numbers of
predators and prey species in fossil collections to
support physiological hypotheses for the predators,
but so many assumptions must be made in recon-
structing even a single community from one collec-
tion, such as that from Pyramid Valley, that further
extrapolations are fraught with considerable difficul-
ties.
The data may supply a partial answer to the
question of predatonprey ratios in the South Ameri-
can faunas dominated by predatory ground
birds
and
mammalian herbivores (Paul 1988: 416).
Passerines
The most abundant passerines were the compara-
tively large tui, the saddleback, and the South Is-
land kokako. Four other passerines, including the
largest (raven) and the smallest (both acanthisittid
wrens),
were each represented by a single individual.
Noteworthy is one of the few confirmed fossils of
the fernbird. Supposed records of this species at other
sites have proved to be erroneous. Material usually
identified as fernbird is usually referable to
Pachyplichas, one of the extinct acanthisittid wrens.
Comparisons with other local faunas
In comparison with those of other regions, such as
around Honeycomb Hill Cave (northwest Nelson)
and Lake Poukawa (eastern North Island), the Pyra-
mid Valley fauna has about the same overall species
richness but a different group of most-abundant taxa.
Most of the differences result from the different
depositional regimes, different excavation proce-
dures,
and, for Poukawa, different species available.
The Lake Poukawa fauna must be taken as a
provisional list, because knowledge of several
groups has increased considerably since the fauna
was first analysed (Horn 1983).
The similarity of the surrounding environments,
and the partial synchrony of Pyramid Valley and
Lake Poukawa, invite comparison of their avifaunas.
The moa faunas differed more than might be ex-
pected from the range of possible species. For ex-
ample, Dinornis giganteus was a conspicuous part
of the Pyramid Valley fauna but was very rare at
Poukawa (Horn 1983). Perhaps reflecting the differ-
ent moa fauna, eagles were absent from Poukawa
(Horn
1983),
although Eyles's harrier
was
abundant,
as at Pyramid Valley.
The presence of piscivores—including cormo-
rants,
pelicans, and the merganser—at Lake
Poukawa, reflects its richer fish fauna, perhaps a
function of its greater expanse of open water and
connection to streams. The larger area of open wa-
ter may also have attracted swans and the musk duck,
both absent from Pyramid Valley.
Material was deposited at Pyramid Valley over a
much shorter period than at Lake Poukawa. The
Pyramid Valley deposit is a "snapshot" in time, at
most a few millennia, and provides an important
baseline for interpreting the environment and fauna
of the eastern South Island before the advent of hu-
man influences. The major changes in faunal com-
position at Lake Poukawa occurred during the last
1000 years, a period not represented at Pyramid
Valley.
The late Holocene Pyramid Valley fauna was
typical of
the
fauna in the eastern South Island mo-
saic vegetation, and was very similar in composition
to that of the Otiran deposits in Honeycomb Hill
Cave and at Takaka (Worthy 1993 a; Worthy &
Mildenhall 1989; Worthy & Holdaway 1993,
1994b). The major difference was that the most
abundant moa species at Pyramid Valley {Dinornis
giganteus and Emeus crassus) were very rare or ab-
sent further west. Before humans arrived there was
a fauna characteristic of drier, more open mosaics
of forest, shrubland, and grassland. Several species
seem to be sure indicators for this fauna at a particu-
lar place and time, for example the eagle and
adzebill.
The broad patterns of bird distribution, and their
changes through time, that have become apparent
over the past decade or so (e.g., Anderson 1982,
1989;
Worthy & Mildenhall 1989; Worthy 1989,
1990,
1993a, b; Worthy & Holdaway 1993, 1994b,
1996a, b) may now be further refined. Evidence is
accumulating for finer grained distribution patterns
Holdaway & Worthy—Pyramid Valley fossil fauna 107
for many species. It is especially apparent in the
faunal assemblages from sites in North Canterbury,
including Pyramid Valley (Worthy & Holdaway
1996a). Altitudinal separation resulting from vegeta-
tion zoning seems to have controlled the distribution
of species such as the quail, and seasonal food short-
ages imposed on probing species by drought-hard-
ened ground may explain the distribution or rarity
of snipe and kiwi. There is also a suggestion that
moas such
as
Pachyornis
elephantopus,
Euryapteryx
geranoides, and Emeus crassus may have been af-
fected directly (according to physical demands) or
indirectly (via vegetation) by features of the terrain,
particularly steepness of hill slopes. The predators,
in turn, would be affected by both the abundance and
physical accessibility of prey.
A paleoenvironmental reconstruction for the area
including Pyramid Valley has been attempted before
(Atkinson & Millener 1991). In comparison to the
present reassessment, which has the advantage of
data from other sites in the area, Atkinson &
Millener's reconstruction was based on the faunal
lists and catalogue entries for Pyramid Valley and
other sites as first interpreted; the data were not
checked. The interpretations of vegetation and cli-
mate are similar, as might be expected from the simi-
lar sources of evidence, but their interpretation of the
faunas, especially the ecology of major species and
the composition of guilds, often differs radically
from ours. The differences may be traced to two
main sources: their reliance on the initial but un-
checked identifications, and their use of presence-
absence data. The first meant that the supposed
presence of some (e.g., Dinornis novaezelandiae,
Circus approximans) had to be accounted for; the
second gave unwarranted prominence to species
such as Dinornis struthoides, which were very rare
in the district (Worthy & Holdaway 1996a).
CONCLUSIONS
The vertebrate fossil fauna from Pyramid Valley has
been highly influential in the development of recon-
structions of New Zealand's Holocene vertebrate
fauna. It is, therefore, somewhat surprising that a
complete list of the species present and their rela-
tive abundance has not been presented before. The
initial appreciation of the site's importance may have
been—and the direct instigation of
its
scientific ex-
cavation certainly was—the fruit of
a
visit by Robert
Cushman Murphy of the American Museum of
Natural History. Subsequent research was confined
largely to increasing the collection and describing
new species.
Since the discovery of the wealth of New Zea-
land's cave deposits, and then of the remarkable se-
ries of deposits left by laughing owls and falcons,
the emphasis in New Zealand Quaternary paleon-
tology has perhaps moved away from swamp and
lake deposits. The Pyramid Valley collection has
several features that are unmatched in any more re-
cently discovered deposit: the number of associated
skeletons of moas and other species, such as
adzebills; the perfect preservation of even small el-
ements; the thoroughness of the curation; the rela-
tively short duration of deposition; and the number
of examples of gizzard contents from which vegeta-
tion and diet can be reconstructed. It is hoped that
the lists presented here will facilitate future research
on the Pyramid Valley material, and that reassess-
ment and analyses, in company with the other faunal
interpretations for North Canterbury (Worthy &
Holdaway 1996a), will provide abase for
more
work
on the distribution, systematics, and paleobiology of
the remarkable faunas of
the
region.
ACKNOWLEDGMENTS
The project of which this work is part was funded by a
grant from the New Zealand Foundation for Research,
Science and Technology to Palaeofaunal Surveys. Re-
assessment of the Canterbury Museum collections was
supported by grants from the Mason Foundation. Geoff
Tunnicliffe and Bev McCulloch kindly allowed access
to the collections in their care; the grant to RNH was
part of Geoff s ongoing programme on the recent
osteological collections. Amanda Freeman gave much
valuable assistance with locating material. Thanks are
also due to the Department of Zoology, University of
Canterbury, for providing facilities for RNH. The
American Museum of Natural History (New York) kindly
gave permission to examine and quote material from the
Robert Cushman Murphy diaries and correspondence,
and from their photographic archive. Allison Andors
(AMNH) kindly provided copies of accession cards for
the Pyramid Valley material. Mike and Jan Hodgen
(Pyramid Valley) were always most hospitable and kindly
allowed RNH to examine and use maps of Pyramid
Valley excavations. Special thanks are due to Brian
Reeve and Jim Eyles for permission to examine Jim's
excavation diaries and sketches from the 1949
programme. We thank Paul Sagar and Janet Wilmshurst
for providing helpful comments on various drafts of the
manuscript. Pyramid Valley, and many other sites, would
not be available for study, or at least the work would be
far more difficult, without the efforts of Ron Scarlett,
who assiduously and effectively documented every
specimen that fell into his hands. It could be said that he
108
New Zealand Journal of
Zoology,
1997, Vol. 24
cut his paleontological teeth on the Pyramid Valley
material. That paleoornithology exists as a discipline in
New Zealand today is a tribute to his endeavours over
nearly 40 years.
REFERENCES
Allan, R. S. 1941: Preliminary report on excavations at
Pyramid Valley Swamp, Waikari, North Canter-
bury: geological introduction. Records of the Can-
terbury Museum 4 (7): 326—327.
Anderson, A. J. 1982: Habitat preferences of moa in
central Otago, A.D. 1000-1500, according to
palaeobotanical and archaeological evidence.
Journal of The Royal Society of New Zealand
12:
321-336.
Anderson, A. J. 1989: Prodigious birds: moas and moa-
hunting in prehistoric New Zealand. Cambridge,
Cambridge University Press. 238 p.
Anderson, S. H. 1992: Shearwaters and saddlebacks as
prey at a morepork ground nest. Notornis 39 (1):
69-70.
Archey, G. 1941: The moa. Bulletin of the Auckland
Institute and Museum
1:
1-245.
Atkinson, I. A. E.; Greenwood, R. M. 1989: Relation-
ships between moas and plants. New Zealand
journal of ecology 12 (supplement): 67—96.
Atkinson, I. A. E.; Millener, P. R. 1991: An ornithologi-
cal glimpse into New Zealand's pre-human past.
Acta XX Congressus Internationalis Ornithologici
1:
127-192.
Bakker, R. T. 1986: The dinosaur heresies. New York,
William Morrow. 481 p.
Batcheler, C. L. 1989: Moa browsing and vegetation
formations, with particular reference to decidu-
ous and poisonous plants. New Zealand journal
of ecology 12 (supplement): 57—65.
Beaglehole, J. C. 1974: The life of Captain James Cook.
London, A & C Black.
Beale, A. 1991: The maturation of the skeleton of
a
kiwi
(Apteryx australis mantelli)—a ten year radio-
logical study. Journal of The Royal Society of
New Zealand
21:
219-220.
Beauchamp, A.; Worthy, T. H. 1988: The takahe's de-
cline in distribution—a re-examination. Journal
of The Royal Society of New Zealand 18: 103—
112.
Beggs, J. R.; Wilson, P. R. 1991: The kaka Nestor
meridionalis, a New Zealand parrot endangered
by introduced wasps and mammals. Biological
conservation 56: 23—38.
Benton, M. J. ed. 1993: The fossil record 2. London,
Chapman and Hall. 845 p.
Brown, L. H.; Amadon, D. 1968: Eagles, hawks & fal-
cons of the World. London, Country Life. 2 vols.
945 p.
Bull, P. C; Gaze, P. D.; Robertson, C. J. R. 1985: The
atlas of bird distribution in New Zealand. Wel-
lington, Ornithological Society of New Zealand.
296 p.
Burrows, C. J. 1980a: Diet of New Zealand Dino-
rnithiformes. Die Naturwissenschaften 67: 151—
153.
Burrows, C. J. 1980b: Some empirical information con-
cerning the diet of
moas.
New Zealand journal of
ecology
3:
125-130.
Burrows, C. J. 1989: Moa browsing: evidence from the
Pyramid Valley mire. New Zealand journal of
ecology 12 (supplement): 51—56.
Burrows, C. J.; McCulloch, B.; Trotter, M. M.
1981:
The
diet of moas based on gizzard contents samples
from Pyramid Valley, North Canterbury, and
Scaifes Lagoon, Lake Wanaka, Otago. Records
of the Canterbury Museum 9
(6):
309-336.
Burrows,
C.
J.; McSaveney, M. J.; Scarlett, R. J.; Turnbull,
B.
1984: Late Holocene forest horizons and a
Dinornis moa from an earthflow on North Dean,
North Canterbury. Records of the Canterbury
Museum 10(1): 1-8.
Cracraft, J. 1976a: The species of moas (Aves:
Dinornithidae). Smithsonian contributions to
paleobiology
27:
189-205.
Cracraft, J. 1976b: The hindlimb elements of moas (Aves:
Dinornithidae): a multivariate assessment of size
and shape. Journal of morphology
150:
495—526.
Cushman Murphy, R.
1947—49:
Manuscript diary, letters,
and photographs on file in the American Museum
of Natural History.
Daugherty, C. H.; Cree, A.; Hay, J. M.; Thompson, M. B.
1990:
Neglected taxonomy and continuing
extinctions of tuatara (Sphenodon). Nature 347:
177-179.
Deevey, E. S. 1955: Paleolimnology of the upper swamp
deposit, Pyramid Valley. Records of the Canter-
bury Museum 6
(4):
291-344.
Delacour, J.; Scott, P. 1956: Waterfowl of the World, vol.
2 (The dabbling ducks). London, Country Life.
Duff,
R. S. 1941: Preliminary report on excavations at
Pyramid Valley Swamp, Waikari, North Canter-
bury: notes on moa excavations at Pyramid Val-
ley, Waikari. Records of the Canterbury Museum
4
(7):
330-353, pi. XLIII-LIII.
Duff,
R. S. 1949: Pyramid Valley. Christchurch, Pegasus
Press for the Association of Friends of the Can-
terbury Museum. 48 p. [Reprinted 1952.]
Holdaway & Worthy—Pyramid Valley fossil fauna
109
Duff,
R. S. 1951: Moas and moa hunters. Wellington,
Government Printer.
Duff,
R. S. 1955: Further report on excavations at Pyra-
mid Valley Swamp, Waikari, North Canterbury:
Foreword. Records of the Canterbury Museum 6
(4):
253-255.
Elliott, G.; Dilks, P.; O'Donnell, C. F. J. 1996: The
ecology of yellow-crowned parakeets (Cyano-
ramphus auriceps) in Nothofagus forest in
Fiordland, New Zealand. New Zealand journal of
zoology
23:
249-265.
Eyles,
J. R.
1955:
Field notes on the excavations. Records
of the Canterbury Museum 6
(4):
257—260.
Falla, R. A. 1941a: Preliminary report on excavations at
Pyramid Valley Swamp, Waikari, North Canter-
bury: general introduction. Records of
the
Can-
terbury Museum 4
(7):
325-326.
Falla, R. A. 1941b: Preliminary report on excavations at
Pyramid Valley Swamp, Waikari, North Canter-
bury: the avian remains. Records of the Canter-
bury Museum 4
(7):
339-353.
Falla, R. A.; Sibson, R. B.; Turbott, E. G. 1979: The new
guide to the birds of New Zealand. Auckland and
London, Collins. 247 p.
Fox, N. C. 1978: The distribution and numbers of New
Zealand falcons (Falco novaeseelandiae).
Notornis 25(4):
317-331.
Gregg, D. R. 1966: New dates for Pyramid Valley moas.
New Zealand Archaeological Association news-
letter 9
(4):
155-159.
Gregg, D. R. 1972: Holocene stratigraphy and moas at
Pyramid Valley, North Canterbury, New Zea-
land. Records of
the
Canterbury Museum 9 (2):
151-158.
Haast, J. 1872: On Harpagornis, an extinct gigantic bird
of prey, containing description of femur, ungual
phalanges, and rib. Transactions and
proceed-
ings of the New Zealand Institute 4 (1871): 192-
196.
Harris, W. F. 1955: Progress report on pollen statistics
from Pyramid Valley
Swamp.
Records of the Can-
terbury Museum 6
(4):
279-290.
Harris, W. F. 1969: The Quaternary history of the vegeta-
tion. Pp. 334-339/n: Knox, G. A. ed. The natural
history of Canterbury. Wellington, A.H. & A.W.
Reed. 620 p.
Holdaway, R. N. 1988: The New Zealand passerine list:
what if Sibley & Ahlquist are right? Notornis 35
(1):
63-70.
Holdaway, R. N. 1989: New Zealand's pre-human
avifauna and its vulnerability. New Zealand jour-
nal of ecology 12 (supplement): 11—25.
Holdaway, R. N. 1990a: Harpagornis assimilis Haast,
1874,
a synonym of Harpagornis moorei Haast,
1872.
New Zealand natural sciences
17:
39-47:
Holdaway, R. N. 1990b: Changes in the diversity of New
Zealand forest birds. New Zealand journal ofzo-
ology
17:
309-321.
Holdaway, R. N. 1991a: Sibley et al.'s (1988) classifica-
tion of living birds applied to the New Zealand
list. Notornis 38
(2):
152-164.
Holdaway, R. N. 1991b: Systematics and palaeobiology
of Haast's eagle Harpagornis moorei Haast, 1872
(Aves: Accipitridae). Unpublished Ph.D. thesis,
Department of Zoology, University of Canter-
bury, Christchurch, New Zealand.
Holdaway, R. N. 1995: A fossil record of the black stilt
Himantopus novaezelandiae Gould, 1841. New
Zealand natural sciences
22:
69—74.
Holdaway, R. N.; Worthy, T. H. 1993: First North Island
fossil record of kea, and morphological and
morphometric comparison of kea and kaka.
Notornis 40
(2):
95-108.
Holdaway, R. N.; Worthy, T. H. 1996: Diet and biology
of the extinct owl Sceloglaux albifacies (Gray,
1844) (Aves: Strigidae) on Takaka Hill, Nelson,
New Zealand. Journal of zoology London 239:
545-572.
Horn,
P.
L. 1983: Subfossil avian deposits from Poukawa,
Hawke's Bay, and the first record of Oxyura
australis (blue-billed duck) from New Zealand.
Journal of The Royal Society of New Zealand
13:
67-78.
Hornibrook, N. de B. 1955: Ostracoda in the deposits of
Pyramid Valley Swamp. Records of the Canter-
bury Museum 6
(4):
267--277'.
Howard, H. 1932: Eagles and eagle-like vultures of the
Pleistocene of Rancho La Brea. Carnegie Insti-
tute of Washington publication 429 (contribu-
tions
to
palaeontology): 1—111.
Hutton, F. W. 1892: The moas of New Zealand. Transac-
tions and proceedings of the New Zealand Insti-
tute 15 (1891): 93-172.
Ludwig, J. A.; Reynolds, J. F. 1988: Statistical ecology: a
primer on methods and computing. New York,
John Wiley & Sons. 337 p.
McCulloch, B.; Trotter, M. M. 1979: Radiocarbon dates
for moa remains from natural sites. New Zealand
journal of geology and geophysics 22: 277—279.
McGlone, M. S. 1989: The Polynesian settlement of New
Zealand in relation to environmental and biotic
changes. New Zealand journal of ecology
12
(sup-
plement):
115-129.
110
New Zealand Journal of
Zoology,
1997, Vol. 24
Marchant, S.; Higgins, P. J. 1990: Handbook of Austral-
ian, New Zealand & Antarctic
birds.
Vol.
1
Ratites
to ducks. Part A. Ratites to petrels. Part B. Aus-
tralian pelican to ducks. Melbourne, Oxford Uni-
versity Press. Pp.
1-736,
737-1400.
Millener, P. R. 1991: The Quaternary avifauna of New
Zealand. In: Vickers-Rich, P.; Monaghan, J. M.;
Baird, R. F.; Rich, T. F. ed. Vertebrate palaeon-
tology of Australasia. Pioneer Design Studio and
Monash University Publications Committee.
Pp.
1317-1339.
Mills,
J. A.; Lavers, R. B.; Lee, W. G. 1984: The takahe—
a relict of the Pleistocene grassland avifauna of
New Zealand. New Zealand journal of ecology 7:
57-70.
Miskelly, C. M. 1989: Social and environmental con-
straints on breeding by the New Zealand snipe
Coenocorypha aucklandica. Unpublished Ph.D.
thesis,
Department of Zoology, University of Can-
terbury, Christchurch, New Zealand.
Moar, N. T. 1970: A new pollen diagram from Pyramid
Valley swamp. Records of the Canterbury Mu-
seum 8
(5):
455-461.
Molloy, B. P. J. 1969: Recent history of the vegetation.
In:
Knox, G. A. ed. The natural history of Canter-
bury. Wellington, A.H. & A.W. Reed. Pp. 340-
360.
Murton, R. K. 1969: Rook. In: Fisher, J.; Peterson, R. T.;
Cramp, S.; Hosking, E.; Warham, J.; Serventy,
V.; Coleman, B. ed. Birds of the world. London,
Purnell. Pp. 2845-2848 (in parts).
Nixon, A. J. 1994: Feeding ecology of hybridizing para-
keets on Mangere Island, Chatham Islands.
Notornis 41 (supplement): 5—18.
O'Donnell, C. F. J.; Dilks, P. J. 1986: Forest birds in
South Westland: status, distribution, and habitat
use.
New Zealand Wildlife Service occasional
publication 10. 179 p.
Oliver, W. R. B. 1949: The moas of New Zealand and
Australia. Dominion Museum bulletin
15:
1—206.
Oliver, W. R. B. 1955: New Zealand birds. Wellington,
A.H. & A.W. Reed. 661 p.
Olson, S. L. 1975: A review of the extinct rails of the
New Zealand region (Aves: Rallidae). National
Museum of New Zealand records 1 (3): 63—79.
Olson, S. L. 1977: Notes on subfossil Anatidae from New
Zealand, including a new species of pink-eared
duck Malacorhynchus. Emu
77:
132—135.
Olson, S. L. 1986: Emendation of the name of the fossil
rail Rallus hodgeni Scarlett. Notornis 33 (1): 32.
Olson, S. L.; James, H. F. 1992: Descriptions of thirty-
two new species of birds from the Hawaiian Is-
lands.
Part I. Non-passerines. Ornithological
monographs
46:
1—88.
Paul, G. S. 1988: Predatory dinosaurs of the world. New
York, Simon & Schuster. 464 p.
Percival, E. 1941: Preliminary report on excavations at
Pyramid Valley Swamp, Waikari, North Canter-
bury: notes on the deposit of the Pyramid Valley
Swamp. Records of the Canterbury Museum 4
(7):
327-329.
Pizzey, G.; Doyle, R. 1980: A field guide to the birds of
Australia. Sydney, Collins. 460 p.
Potts,
T. H. 1870: On the birds of New Zealand. Transac-
tions and proceedings of the New Zealand Insti-
tute
2:
40-78.
Potts,
T.
H.
1882:
Out in the open. Christchurch, Lyttelton
Times. 301 p.
Powlesland, R. G.; Lloyd, B. D.; Best, H. A.; Merton, D.
V. 1992: Breeding biology of the kakapo Strigops
habroptilus on Stewart Island, New Zealand. Ibis
134: 361-373.
Powlesland, R. G.; Roberts, A.; Lloyd, B. D.; Merton, D.
V.
1995:
Number, fate, and distribution of kakapo
(Strigops habroptilus) found on Stewart Island,
New Zealand, 1979—92. New Zealand journal of
zoology
22:
239-248.
Rich, P. V.; Scarlett, R. J. 1977: Another look at
Megaegotheles, a large owlet-nightjar from New
Zealand. Emu 77: 1—8.
Scarlett, R. J. 1953: A sub-fossil hawk from New Zea-
land. Records of
the
Canterbury Museum 6 (3):
245-252.
Scarlett, R. J. 1955a: Further report on bird remains from
Pyramid Valley. Records of the Canterbury Mu-
seum 6
(4):
261-264.
Scarlett, R. J. 1955b: A new rail from South Island swamps
in New Zealand. Records of the Canterbury Mu-
seum 6
(4):
265-266.
Scarlett, R. J. 1968: An owlet-nightjar from New Zea-
land. Notornis 15 (4): 254-266.
Scarlett, R. J. 1969: Moas and other birds. Pp. 565-568
in:
Knox, G. A. ed. The natural history of Canter-
bury. Wellington, A.H. & A.W. Reed. 620 p.
Scarlett, R. J. 1972: Bones for the New Zealand archae-
ologist. Canterbury Museum bulletin 4. 29 p., 33
pi.
Sibley, C. G.; Ahlquist, J. E. 1990: Phylogeny and classi-
fication of
birds:
a study in molecular evolution.
New Haven and London, Yale University Press.
976 p.
Taylor, R. H. 1975: Some ideas on the speciation of New
Zealand parakeets. Notornis
22:
110-121.
Taylor, R. H.; Heatherbell, E. G.; Heatherbell, E. M.
1986:
The orange-crowned parakeet (Cyano-
ramphus malherbi) is a colour morph of the yel-
low-crowned parakeet
(C.
auriceps). Notornis 33
(1):
17-22.
Holdaway & Worthy—Pyramid Valley fossil fauna
111
Triggs, S. J.; Daugherty, C. H. in
press:
Conservation and
genetics of New Zealand parakeets. In: Bell, B.
D.
ed. Management of endangered species. Inter-
national Council for Bird Preservation bulletin.
Trotter, M. M.; McCulloch, B. 1984: Moas, men, and
middens. In: Martin, P. S.; Klein, R. G. ed. Qua-
ternary extinctions: a prehistoric revolution.
Tucson, University of Arizona Press. Pp. 708-
727.
Turbott, E. G. conv. 1990: Checklist of the birds of New
Zealand and the Ross Dependency, Antarctica.
3rd ed. Auckland, Random Century and the Orni-
thological Society of New Zealand. 96 p.
Williams, G. R. 1974: Extinction and the Anatidae of
New Zealand. Wildfowl Trust 15th annual report
for 1962-63.
Williams, M. A. J.; Dunckerley, D. L.; De Decker, P.;
Kershaw, A. P.; Stokes, T. 1993: Quaternary en-
vironments. London, Edward Arnold. 329 p.
Worthy, T. H. 1987a: The subfossil fauna of Honeycomb
Hill Cave. I. A study of Otira deposits in the
Graveyard. II. A reassessment of Megalapteryx
benhami Archey. Unpublished report, National
Museum of New Zealand. 59 p.
Worthy, T. H. 1987b: Sexual dimorphism and temporal
variation in the North Island moa species
Euryapteryx curtus (Owen) and Pachyornis
mappini Archey. National Museum of New Zea-
land records 3 (6): 59-70.
Worthy, T. H. 1989: An analysis of moa bones (Aves:
Dinornithiformes) from three lowland North Is-
land swamp sites: Makirikiri, Riverlands and
Takapau Road. Journal of The Royal Society of
New Zealand
19:
419-432.
Worthy, T. H. 1990: An analysis of the distribution and
relative abundance of moa species (Aves:
Dinornithiformes). New Zealand journal of zool-
ogy 17:213-241.
Worthy, T. H. 1993a: Fossils of Honeycomb Hill. Mu-
seum of New Zealand Te Papa Tongarewa. 56 p.
Worthy, T. H. 1993b: A review of fossil bird bones from
loess deposits in eastern South Island, New Zea-
land. Records of the Canterbury Museum 10 (8):
95-100.
Worthy, T. H. 1994: Reappraisal of Dinornis (Aves:
Dinornithiformes) species—a morphometric
analysis. New Zealand journal of zoology 21:
113-134.
Worthy, T. H. 1995: Description of some post-cranial
bones of Malacorhynchus scarletti, a large -ex-
tinct pink-eared duck from New Zealand. Emu
95:
13-22.
Worthy, T. H. 1997: The Quaternary fossil fauna of
South Canterbury, South Island, New Zealand.
Journal of The Royal Society of New Zealand
27(1):
67-162.
Worthy, T. H.; Holdaway, R. N. 1993: Quaternary fossil
faunas from caves in the Punakaiki area, West
Coast, South Island, New Zealand. Journal of
The Royal Society of New Zealand 23 (3): 147—
254.
Worthy, T. H.; Holdaway, R. N. 1994a: Scraps from an
owl's table—predator activity as a significant
taphonomic process newly recognised from New
Zealand Quaternary
deposits.
Alcheringa
18:229—
245.
Worthy, T. H.; Holdaway, R. N. 1994b: Quaternary fossil
faunas from caves in Takaka Valley and on Takaka
Hill, northwest Nelson, South Island, New Zea-
land. Journal of
The
Royal Society of New Zea-
land
24:
297-391.
Worthy, T. H.; Holdaway, R. N. 1995: Quaternary fossil
faunas from caves on Mt Cookson, North Canter-
bury, New Zealand. Journal of
The
Royal Society
of New Zealand
25:
333-370.
Worthy, T. H.; Holdaway, R. N. 1996a: Quaternary fossil
faunas, overlapping taphonomies, and palaeo-
faunal reconstruction in North Canterbury, South
Island, New Zealand. Journal of The Royal Soci-
ety of New Zealand
26:
275—361.
Worthy, T. H.; Holdaway, R. N. 1996b: Taphonomy of
two Holocene microvertebrate deposits, Takaka
Hill, Nelson, New Zealand, and identification of
the avian predator
responsible.
Historical biology
12:
1-24.
Worthy, T. H.; Mildenhall, D. C. 1989: A late Otiran-
Holocene paleoenvironment reconstruction based
on cave excavations in northwest Nelson, New
Zealand. New Zealand journal of geology and
geophysics 32 243—253.
Worthy, T. H.; Daniel, M. J.; Hill, J. E. 1996: An analysis
of skeletal size variation in Mystacina robusta
Dwyer, 1962 (Chiroptera: Mystacinidae). New
Zealand journal of zoology
23:
99—110.
Appendix 1 overleaf
112
New Zealand Journal
of
Zoology,
1997, Vol.
24
APPENDIX
1
LIST
OF
MATERIAL COLLECTED FROM
PYRAMID VALLEY, 1939-73.
Specimens
in
italics
are
listed
in
Scarlett (1953), Scarlett
(1955a),
or
Scarlett (1955b).
All
specimens catalogued
in
CMNZ. All CMNZ avian accession numbers are prefixed
by
Av,
omitted
in the
list
for
brevity. Abbreviations:
L,
left;
R,
right;
pt,
part;
sh,
shaft; ant, anterior; prox, proxi-
mal part; dist, distal part;
imm,
immature; juv, juvenile;
subad, subadult;
br,
broken; frag, fragment(s);
era, cra-
nium; prem, premaxilla; man, mandible; sea, scapula;
cor,
coracoid;
ste,
sternum; pel, pelvis; innom, innominate;
v,
vertebra;
cv,
cervical vertebra;
vt,
dorsal
or
thoracic
ver-
tebra; caudv, caudal vertebra;
hum,
humerus;
uln,
ulna;
rad, radius; cpm, carpometacarpus; carp, carpal; fern,
fe-
mur; tbt, tibiotarsus; tmt, tarsometatarsus. x/y, number
of
elements/MNI. Roman numerals denote excavation pits
from 1939-42 field seasons.
'Sq
no.' denotes position
on
excavation grid used from 1949 season onwards.
Radiocarbon dates are from Gregg (1972) and Burrows
etal. (1981).
Material listed from the American Museum of Natural
History (AMNH)
was
checked
by
THW, June
1996.
Genus Sphenodon Gray
Sphenodon
sp.
MATERIAL:
CM Rep
213/1;
Sq 60; CM Rep 25 3/1; Sq
58;
CM Rep 26
7/1;
Sq 62; CM Rep 27
6+/1;
Sq 66
and
58;
CM Rep 47 1/1;
Sq
109; CM Rep 134
3+/1;
Sq
121;
CM Rep 135 1/1;
Sq
120; CM Rep
136
1/1;
Sq
120;
CM
Rep
138
1/1;
Sq
121;
CM Rep
139
1/1;
Sq
121;
CM Rep
140
1/1; Sq
121;
CM
Rep 458 1/1;
in
spoil.
Genus Hoplodactylus Fitzinger
Hoplodactylus
sp.
MATERIAL: CM Rep 137 1/1; Lfem, Sq 121.
Genus Apteryx Shaw & Nodder
Apteryx sp.
cf.
australis (large)
MATERIAL:
5835 3/1; LRfem Rtbt (imm),
Sq
58; 5840
9/
1;
LRtmt [pelpt crapt 4caudv man—not found],VII;
5 841
14/1;
2fem 2tbt 2tmt 3v
fib
era prem man qua (to
P.
V.
Rich, Monash University 10.1978), VII; 5843 1/1; Lfem;
6090 1/1; Rtmt, Sq 84; 15060 1/1; Rfem,
Sq
108; 10275
1/
l;phal; 33541 1/1;
vt.
Apteryx sp.
(juv. or
undiagnostic)
MATERIAL:
5734 1/1; Ltbtimm, Sq52-53; 5837 l/l;pelpt
(L acetab, imm),
Sq
60-61;
6091 1/1;
Ltmtimm,
Sq 84;
6156 1/1; Ltbt,
Sq
84; 9941 1/1;
v; 10219
1/1; Lpalatine;
14361 1/1; Lmanpt,
VII;
15041 7/1;
pel 5vt
rib,
Sq 105;
20092 1/1; Lacetab, Sq
121;
20093
2/1;
Ltbtimm Rtmt,
Sq
121.
Not found: 5867 1/1; tbtimm; 9962 1/1;
v;
12657
1/1;
manfrag,
Sq 89.
Genus Pachyornis Lydekker
Pachyornis elephantopus (Owen,
1856)
MATERIAL:
17
part skeletons plus miscellaneous elements.
Part skeletons: 8315
["to
AIM
10.12"];
8348; 8380(see
below); 8381; 8382; 8383; 8384, immature; 8385; 8386;
8387;
8388; 8389; 8716; 15029; 15036.
Other material: 8427,
era,
ribs,
v;
8679,
era,
prem;
8684,
era, prem; 8595, man; 8707,
era;
8710, rem; 9230,
man; 9313, ste; 9389,48 v,
13
caudv; 9399, thyroid; 9403,
quadrate (see
below);
9404, quadrate; 9480, part
era;
9482,
era imm;
9502,
jug proc.
AMNH: 7305 (Field
no.
48.5A) era prem man "almost
complete skeleton"; 7307 (Field no. 48.6F) "partial skel-
eton without skull";
7313 (ex
8380, Field no.
72-C) era
prem
man jug 13v
2caudv ribs
pel ste
2fem
tbt
2tmt
2hallphal "etc."; 7318
(ex
9403)
qua.
Genus Emeus Reichenbach
Emeus crassus (Owen,
1846)
MATERIAL:
81 part skeletons, plus miscellaneous elements.
Part skeletons: 8300; 8301 A[to MNZ
S469];
83O1B
[to
AMNH,
see
below]; 8302; 8303
[to
AMNH,
see
below];
8304;
8305; 8306; 8307; 8308; 8309 [associated with
re-
constructed
egg
(Falla 1941b)]; 8310; 8311; 8312; 8313;
8314 [to MNZ
S470];
8316; 8317; 8319; 8320; 8321/9370;
8322;
8323; 8324; 8325; 8326; 8327; 8328; 8329; 8330;
8331 [Rtmt, 9ribs
to
Milwaukee Mus]; 8332; 8333 [legs
and feet
to
South Africa]; 8334; 8335
[to
Field Museum,
Chicago];
8336; 8337; 8338
[to AIM
B6094 (Rtmt,
v)];
8339;
8340; 8341
[to AIM
B6092]; 8342; 8343
[to Sin-
gapore Museum]; 8344; 8345
[to
Perth Museum]; 8346;
8347;
8349; 8350;
8351 [to
U.S.A.]; 8352; 8353; 8354;
8355;
8356; 8357; 8358; 8359
[to
University
of
Califor-
nia];
8360; 8361/8664; 8362; 8363/8529; 8364; 8365;
8366/9369; 8367; 8368; 8369; 12644-5; 13774; 13775;
13776;
14548; 15030;
15031;
15032;
18931;
20119 [to
Institute
of
Nuclear Sciences—dates NZ610 3600145
(bone) NZ625 3740+72 (gizzard)]; 28433 [exchanged
Hancock Museum, England].
Other
material:
7888, quadrate; 8318, era, prem; 8515,
pel;
8516, pel; 8536, era, prem,
LR
quadrates
[to
MNZ
S
460];
8553, ste; 8554, ste; 8570, tmt;
8571,
ste;
8621,
tmt;
8662,
prem;
8666,
prem; 8700, ste [to AMNH
7314];
8701,
ste;
8703,
era,
prem,
man;
8704,
era,
prem,
man;
8705,
prem; 8706, prem; 8714,
era;
8715,
ste;
9234, tracheal
rings;
9235, tracheal rings; 9356-7, tracheal rings; 9380-
1,
tracheal rings; 9382, eggshell; 9385, prem; 9387, 59
v;
9400,
man; 9401, quadrate; 9402, quadrate; 9404, quad-
rate;
9405, quadrate; 9406, quadrate; 9460-4, 5
pts
man;
9465,
tracheal rings;
9475,
pt
man;
9481,
pt
era;
9483,
era;
9487-8,
2
atlas; 9490-4,
5 era;
9495, axis; 9501,
10
jug
proc;
9506, tracheal ring; 9512,
pt man; 9514,
gizzard
stones;
10761,
v; 17280, man;
17281,
man, prem; 17286,
Rtbt; 17287, Lfib; 20125, Lfem; 20127, Rtmt [to National
Science Museum, Ueno Park, Tokyo]; 25566, Lfem.
AMNH: 7306 (Field no. 48.6D) partial skull
and
skel-
eton ("young"); 7308 (Field no. 48.6C)
era man
"partial
skeleton"
but
with too many
v
and phal—atlas
not
fitting
era;
7310 (ex
8303)
era
prem
man
thyroid 2qua
18r pel
ste
24v
4caudv 2fem 2tbt 2fib 2tmt 19phal 5hallphal
4uncproc
160±
trachrings;
7312 (as E.
huttoni,
ex 8301
VIIIA) prem man trachrings 22+v "several" ribs pel 2fem
2tbt 2fib tmt 4hallphal uncproc "etc."; 7314
(as E.
huttoni,
ex 8700)
ste.
Holdaway & Worthy—Pyramid Valley fossil fauna
113
Genus Euryapteryx Haast
Euryapteryx geranoides (Owen, 1848)
MATERIAL:
21
part skeletons plus miscellaneous elements.
Part skeletons: 8370 [to MNZ S 471]; 8371; 8372;
8373;
8374; 8375; 8376; 8377; 8378; 8390; 8622; 13773
["to Aust Mus" in Catalogue, but in CM]; 15033 [to Uni-
versity of
California];
15034; 15035; 20113 [to Aust Mus;
NZ623
3450+71
(gizzard)];
20114; 20115; 20116.
Other material: 8539, pel;
8633,
Rtmt; 8697, prem,
man; 8708, prem, man; 9069, Ltmt; 9340, half egg; 9388,
2 v; 9394, tracheal rings; 9407, quadrate; 9466, thyroid;
9467,
scapulocoracoid; 9489, era;
9503,
12 jug proc;
13777,
pel, ste, 7 v, 7 rib; 20102, eggshell; 20129, L quad-
rate,
prem, v; 25567, Ltmt.
AMNH: 7304 (Field no. 48.6E) "part skeleton"; 7309
pt skeleton, (era "found sev. ft. from skel. but in direction
of drift'" is Emeus) prem manpt qua trachrings ste pel
LRtbt LRfib Rtmt phals atlas 19r.
Genus Dinornis Owen
Dinornis struthoides Owen, 1844
MATERIAL:
8415, part skeleton [as D. torosus, Scarlett
(1955a)];
8702, prem, man.
Dinornis giganteus Owen, 1844
MATERIAL: 63 part skeletons (9 imm).
Part skeletons: 8416; 8417; 8418; 8419; 8420 [to
Taranaki Museum]; 8421; 8422;
8423;
8461, imm; 8462
[now MNZ S34088]; 8464; 8465; 8465; 8466; 8467, imm;
8468;
8469; 8470; 8471;
8473;
8474, imm; 8475; 8476;
8477;
8478; 8479; 8480;
8481,
imm; 8482, imm; 8483 [to
Chicago Museum]; 8484; 8485 [to AIM
460];
8486; 8487
(see below); 8488; 8489; 8490; 8491; 8492;
8493;
8494;
8495;
8547 plus leg bones; 10899 [to University of Cali-
fornia];
12643; 13778; 13779; 13899 imm; 13900 imm
[leg bones too juvenile; cranium does not
belong];
14448;
14449;
14450 [to Milwaukee Museum];
14451;
14549;
15024; 15025; 15026; 15028; 20117 imm; 20118
[NZ624 3640±72 (gizzard)]; 28434, (122B) imm
[NZ3936 3480+80 (bone); NZ3937 3590±60 (gizzard)].
Other material: 7562; 8530 era man; 8531 prem man;
8532 prem; 8533 man; 8534 era; 8595-6 era; 8665 tracheal
rings; 8696 era; 8709 era (see below) [now MNZ S34089];
8711 era; 8712 prem man quadrate (see below); 8713 era;
8717 era; 9293 5 tracheal rings; 9298 49 tracheal rings;
9301-3 tracheal rings; 9311 30 tracheal rings; 9314 LRfib;
9315 prem era; 9316 era; 9317 92 tracheal rings; 9346 50
tracheal rings; 9347 32 v; 9348 15 ribs; 9355 era (imm);
9378-9 tracheal rings (9378 see below); 9392 tracheal
rings; 9395 prem; 9396 2 quadrates (see below); 9397 2
quadrates; 9398 2 quadrates; 9449 10 phalanges; 9450 x
phalanges; 9451 era [to Taranaki Museum]; 9452 pt man;
9453 pt man; 9454 pt man; 9455 pt man; 9469 11
phalanges; 9470 11 phalanges, 3 hall; 9473 6 uncinate
proc;
9484 atlas; 9497 vomer; 9498 prefr; 9499 prefr; 9500
crafrag; 9504 prefr; 9505 tracheal rings; 10754 era; 10900
man; 11468 Rquadratojugal; 15027 fibula; 15037 era;
15226 pt man; 17436 prem; 17437 era; 17438 pel; 17440
Rfem; 20120 Lphal; 20121 era; 20122 pt man; 20123
Rfem; 20124 Lfem; 20126 2 cv; 20128 2 phalanges.
AMNH: 7303 (Field no. RCM.3) manart
"port.pt.skeleton"; 7315 (ex 8696) era; 7316 (ex 8712)
prem man "supposed to be a quadrate, but this not found";
7317 (ex 9396) 2qua; 7319 (ex 9378) 100+ trachrings.
Dinornis
giganteus:
7311 (ex 8487 Field
no.
70-B) pt skel
ste 26v 2fem tbt fib tmt 35pedphal ribs; 7301 (Field no.
48.3A) era prem man "partial skeleton"; 7302 (Field no.
48.3B) skull.
Dinornis sp.
MATERIAL:
9063 LRtmt
juv;
10262
vomer;
10264
crafrag.
Moa sp. indet.
MATERIAL:
7490 sternal ribs; 9323 eggshell; 9325 vomer;
9328 eggshell; 9329 thyroid; 9384 eggshell; 9390 54 ribs;
9471 63 phalanges; 9473 20 uncinate proc; 9476 vomer;
9477 vomer; 9478-9 brain casts; 9586 atlas; 9507 tracheal
ring; 9510 thyroid; 10263 vomer.
Genus Coturnix Bonnaterre
Coturnix novaezelandiae Quoy & Gaimard, 1830
MATERIAL:
20155
1/1;
Ltbtpsh,
Sq 121.
Not found: 14469 1/1; steant cf 15896 (this bone is
from inland Hawke's Bay and has not, been found yet,
either), Sq 96;
Genus Cnemiornis Owen
Cnemiornis calcitrans Owen, 1865
MATERIAL: 5406
25/1
;Lhum Lfem Rtbtpelbr 10rib4phal
4v pyg Rfib scabr, XX; 7447 1/1; Rrib, VII; 101111/1;
rib.
Not found: 5754 1/1; LI/3; 6780 frag(pel?); 7438 1/
1;
rib, VII; 7525 1/1; ribimm; 7526 1/1; hyoid, Sq 72;
9841 1/1; vt, Sq 68.
Genus Euryanas Oliver
Euryanas finschi (Van Beneden, 1875)
MATERIAL:
5708 1/1; Rhum, Sq 54-55; 5785 2/1; Rcor
pel;
5819 1/1; Ltbt, Sq 56-57; 5869 1/1; fur; 5993 1/1;
Lfem, Sq 56; 6151 1/1; Rhum, Sq 72; 7345 1/1; pel; Sq
63;
9622 1/1; Rhumimm; 9625 1/1; Lcor, XXII; 10644 1/
1;
Ltbt; 10645 1/1; Rfem; 13770 1/1; Rcpm, Sq 92;
18917 1/1; pel, Sq 111 4-5'; 18918 2/1; LRtbt (slightly
subad),
Sq 111 (5'depth); 20091 1/1; Rtbt, Sq 125;
20108 1/1; Ltbt, Sq 117 lower peat; 20109 1/1; Ltbt, Sq
117 lower
peat;
20111 1
/1;
syn (3pcs), Sq 117 lower peat;
20144 1/1; steant, Sq
121;
20145 1/1; Lhumdsh, Sq 120.
Not found: 7345 1/1; syn; 18919 1/1; Rpel(ischil
plate),
Sq 111 4-5' depth; 20080 1/1; pel(4pcs), Sq 120;
Genus Tadorna Lorenz von Oken
Tadorna variegata (Gnielin, 1789)
MATERIAL:
5756 11/2; LRuln LRptbt LRhum Rcpm
2LRtmt ste, XI and XIX; 5786 3/1; LRcor, XIX; 5815 1/
1;
pel; 5820 1/1; Ltbt, Sq
52-53;
5823 1/1; Ltbtp, Sq 56-
57;
5857 1/1; Lhum; 5872 1/1; Linnom, Sq 53; 5910 1/1;
Lcpm, Sq 62; 5937 1/1; Ltbtp (shaft defective?), Sq 56;
5942 1/1; Lulnptsh, Sq
62;
6099 1/1; pel; 6100 1/1; stebr,
Sq
66;
6123 1/1; Rtmtimm; 6147 1/1; Rhumbr, Sq 56-57;
114
New Zealand Journal of Zoology, 1997, Vol. 24
6148 1/1; Lhum, Sq 62; 6149 1/1; Luln; 7311 1/1;
Rhumimm, Sq
63;
7332 1/1; Lfemd, Sq
52-53;
7333 1/1;
Lfemd, Sq
71;
7354 1/1; Rfib,Sq 78-80; 7512 l/l;Rradp;
10647 1/1; Rfem; 20072 1/1; pelpt(2pcs), Sq 120.
Genus Malacorhynchus Swainson
Malacorhynchus scarletti Olson, 1977
MATERIAL:
5788 2/2; 2Rhum; 5821 1/1; Ltmtimm, Sq 56-
57;
5845 1/1; Luln; 5855 l/l;prempt
[Holotype];
5870 1/
1;
era;
61611/1;
Rtbtimm, Sq 63-65; 6186 1/1;
Rfemsubad, Sq
52-53;
6187 1/1; Rfemimm, Sq 72;
7321 1/1; Lmanpt (prox left ramus just caudal to tip);
7480 1/1; Ruin, Sq 72; 7492 1/1; Rulnimm, Sq 78-80;
9959 1/1; manpt, Sq 67or
69;
12645 1/1; Rfemsubad, Sq
90;
14697 1/1; Rfem; 15044 6/1; Rhum furpt Rcor Rsca
Ruin radpt, Sq 105; 15071 1/1; Rhum, Sq 107 in black;
15112 1/1; Lsca, Sq 107; 16121 1/1; Rtbtsubad, Sq 97;
20150 1/1; Lhum, Sq
121;
20176 1/1; RmanII/1 imm, Sq
121;
36035
2/1;
eraprempt, Sq
60;
36038 1/1; Lhumsubad
(only R so far); 36285 1/1; Rcpm.
Genus Anas Linnaeus
Anas chlorotis Gray, 1845
MATERIAL:
5709
1/1; pel, Sq 59; 5715 1/1; era, Sq 55;
5784
1/I;cra(2pts)
(found occipital region);
5791
1/1; pel,
Sq 52 or 53; 5793 1/1; Lsca; 5806 1/1; pelpt, 5807 6/3;
Rtbt R2Lhum Rcpm fur; 5808 1/1; prem, VII; 5809 1/1;
prem,Sq60or62;5863 1/1; steant, Sq59; 5864 l/l;Rtmt,
Sq 62; 5865 1/1; Rtmt; 5866 1/1: fur, XX; 5876
4/1;
Rtmt
Rhumpt Rtbt (imm), Sq
54;
5.8813/1; LRtbt Rhum, Sq 56;
5947 1/1; Lhum, Sq
59;
5982 1/1; Rtbt, Sq
60-61;
5983 1/
1;
Rtbtsubad; 5984 1/1; Rtbt, Sq 62; 5986 1/1; Rtbtimm;
6074 1/1; era, Sq 66; 6084 1/1; Ltmt, Sq 64; 6101 1/1;
stebr, Sq 64; 6153 1/1; Lhum, Sq 72; 6154 1/1; Lhum;
6158 1/1; Rtbt, Sq 68; 6160
2/1;
LRtbt, Sq 66; 6162 1/1;
Ltbt; 6163 1/1; Ltbtimm; 6173 1/1; pelbr, Sq 70; 6174 1/
1;
Linn(imm), Sq 70; 6175 1/1; syn, Sq 64; 6183 1/1;
Lfem, Sq 59; 6184 1/1; Lfem, Sq68; 6185 l/l;Lfem, Sq
64;
6188 1/1; Rfemimm, Sq 64; 6224 1/1; Rcor, Sq 59?;
6226 1/1; Rcor, Sq 64; 6230 1/1; Lcorbr, Sq 66; 6232 1/
1;
Lcor, Sq 64; 7259 1/1; Lhum juv, Sq 72; 7261 1/1;
Rcpm, Sq 71; 7262 1/1; Rcpm, Sq 69; 7264 1/1;
Lcpmimm;7281 1/1;Rfemimm, Sq59; 7315 l/l;basicra;
7328 1/1; Lmanfrag; 7329 1/1; Lmanpt, Sq 62; 7352 1/1;
Lmanp, Sq 79-80; 7353 1/1; Lman, Sq 78-80; 7415 1/1;
Lmanramus, Sq
67;
7417 1/1; Lcor, Sq
72;
7419
1/1;
Rcor,
Sq 60; 7422 1/1; Lcor, Sq 64; 7425 1/1; Lcorimm, Sq 63;
7455 1/1; Rsca; 7466 1/1; Rsca, Sq72; 7475
1/1/Luln,
Sq
78-80;
7482 1/1; Luln, Sq 62; 7486 1/1; Luln; 7487 1/1;
Ruin, Sq 59; 7488 1/1; Rulnimm; 7493 1/1; Lulnimm, Sq
78-80;
7494 1/1; Luln; 7504 1/1; Lrad, Sq 64; 7505 1/1;
Lrad; 7506 1/1; Lrad, Sq 59; 7507 1/1; Lrad; 7544 1/1;
Rmanfrag, Sq 62; 7547 1/1; Lfem-p; 9626 1/1; Lmanp,
XXII; 10118 1/1; crabasesubad, Sq 54-55 peat; 10681 1/
1;
Lcor; 10682 1/1; Rhum; 10683 2/1; LRhumimm;
10685 1/1; Rhum; 10686 1/1; Ltbt, XXII; 10687 1/1; Rtbt;
10688 1/1; Ltbt; 10876 1/1; stebr, Sq
85;
12655 1/1; Rfem,
Sq 90; 12660 1/1; Luln, Sq 89; 13767
2/1;
Lcorimm
Ltmtimm, Sq 92; 14378 1/1; fur, XIII; 14379 1/1; Rcor,
XIV; 14460 1/1; Rhumsubad, Sq 97; 15064 1/1; pel, Sq
108;
15074 1/1; Rhum, Sq 109; 18920 1/1; Lhum, Sq 112
in yellow; 20140 1/1; era; 20385 1/1; Rmanp(ramus), spoil
heap;
20406 1/1; Rmanp(ramus), XI; 23586 1/1; Rsca.
Not found: 5691
1/1;
era, Sq
60;
6189
1/1;
Rfemimm,
Sq
59;
10668 1/1; tbt; 10684 1/1; humimm, Sq 54 or 55.
Anas cf. chlorotis
MATERIAL:
5946 1/1; Rtmtimm; 5999 1/1; stefrag, Sq 60;
7291 1/1; Rfemimm; 18914 1/1; Rcpmimm, Sq 111 4'.
Anas gracilis Buller, 1869
MATERIAL:
5948 1/1; Lhum, Sq 62; 5949 1/1; Rhum, Sq
60-61;
6190 1/1; Rfemsubad; 6191 1/1; Rfemsubad, Sq
65.
Not found: 5858 1/1; ste.
Anas superciliosa Gniclin, 1789
MATERIAL:
5786 1/1; Rsca, XIX; 5792 1/1; Rhum; 5818 1/
1;
Lcor (subad), Sq 54-55; 5883 3/1; fur 2v, XVIII;
5980
2/1;
Rhum Ltbt, Sq 58; 5981 1/1; Lsca; 5992 1/1;
Rsca, Sq
56;
6112 1/1; steant, Sq
77;
6231 1/1; Rcorpfrag,
Sq 55; 7322 1/1; Lsca, XVIII; 10680 1/1; Rcor; 14462 1/
1;
premtip, Sq 94; 14463 1/1; Rtbt, Sq 97; 15077 1/1;
cra(subad), Sq 107; 20098 1/1; Ruin, Sq 119; 20104 1/1;
Rrad, Sq 121; 20148 1/1; Rtbt, Sq 121.
Not found: 5822 1/1; Ltmtimm, Sq 58.
Anas sp.
MATERIAL: 5856 1/1; Lhum(imm); 5871 1/1; Ltmtimm;
5950 l/l;Lhum(imm?);6113 l/l;crafrag, Sq64; 6235 1/
1;
fibpt, Sq 84; 7216 1/1; Ltmtimm, Sq 64; 7217 1/1;
Rtmtimm, Sq 63; 7273 1/1; Lfemimm, Sq 63; 7457 1/1;
Lsca, Sq
59;
7462 1/1; Lscaimm; 7538 1/1; syn; 10110 1/
1;
crafragimm (other part of 10113); 15098 2/1; Rcor
Lhum imm, Sq 107; 20095 2/1; Rhumimm(119)
Lhumimm(121), Sq
119/121;
20151 1/1; Lischprocpub,
Sq 121.
Not found: 5859 1/1; ste, Sq
60-61;
5862 1/1; Rtbt;
5874 1/1; fur, Sq 60; 7276 1/1; femimm, Sq 56; 9893 1/
1;
Lfem, XIV; 10113 1/1; crfrag (other part of 10110);
10147 1/1; vimm, Sq 68; 13769 1/1; Lhumd, Sq 92.
Genus Aythya Boie
Aythya novaeseelandiae (Gmelin, 1789)
MATERIAL:
5787 1/1; Lcorimm; 5860 1/1; prem(tip);
5861 1/1; Lfem; 5868 1/1; fur, Sq 59; 5893 1/1; fur, Sq
61;
5987 1/1; Rtbt, Sq 62; 6164 1/1; Rtbtimm, Sq 59;
6225 1/1; Lcor, Sq 62; 6229 1/1; Rcorsubad; 7319 1/1;
LRmanfrag, Sq 59; 7320 1/1; Lmanfrag, Sq 58; 7335 1/
1;
Lhumdsh; 9818 1/1; Rfem; 15073 3/1; Rhum Rrad
Ruin, Sq 109; 20070 3/1?; LRtmtimm Rfem, Sq 120;
20106
2/1;
Rtbtimm furpt, Sq 120; 20147 1/1; Lste, Sq
120;
20149 1/1; Rtbtsubad, Sq 121; 20159 1/1; Luln, Sq
121;
20166 1/1; Ltmt, Sq 120 or 121. [20388 1/1; Rsca,
XXII-THW to check]
Not found: 6152 1/1; humimm, Sq 77.
Anatidae sp. indet.
MATERIAL:
5945 1/1; Ltmtimm; 6065 1/1; Ltbtimm, Sq 54
or 55; 7050 eggfrag, Sq 64; 7053 4 egg frag; 7278 1/1;
Holdaway & Worthy—Pyramid Valley fossil fauna
115
Lfemimm, Sq 58; 7279 1/1; Lfemimm, Sq 58; 7282 2/1;
LRfemimm, Sq 56; 7283 1/1; Rfemimm, Sq 62; 7285 1/
1;
Lfemimm, Sq 69; 7287 1/1; Lhumimm, Sq66; 7289 1/
1;
Lhumimm, Sq 84; 7292
2/1;
Lfemimm, Sq
56;
7312 1/
1;
Lhumimm, Sq 59; 8280 3/1; Lcor Rtmtimm Lulnimm,
Sq 66; 10128 1/1; v; 10129 1/1; v; 10131 1/1; v, Sq 78-
80;
10132 1/1; v, Sq 78; 10133 1/1; v, Sq
51;
10157 1/1;
vimm, XX; 10235 1/1; ischpt; 10775 1/1; Rischium, Sq
80;
15126 1/1; Rscajuv, Sq 104; 20172
1/1;
Lfemimm, Sq
119;
20175 1/1; Lscaimm, Sq 121; 20177 1/1; Rtbtimm,
Sq 121; 20179
2/1;
LRfemimm, Sq 120.
Genus Cyanoramphus Bonaparte
Cyanorhamphus novaezealandiae (Sparrman, 1787)
Cyanoramphus auriceps (Kuhl, 1820)
Cyanoramphus malherbi Souance, 1857
MATERIAL:
5434 1/1; Lcor; 5929
3/1;
Ltbt Ruin Lhum, Sq
58;
5931 1/1; Luln; 5932 1/1; Rhum, Sq 60; 5933 1/1;
Lhum, Sq
52-53;
5988 1/1; Rcor; 5990 1/1; stept, Sq 60;
6083 1/1; prem, Sq 84; 6142 1/1; Ruin, Sq 70; 7054 1/1;
man, Sq 63; 7222 1/1; Rtbt, Sq 72; 7230 1/1; Rtbtd;
7268 1/1; Ruin, Sq 61; 7301 1/1; Ruin, Sq 66; 7314 1/1;
Ruin, Sq 60; 9901 1/1; tmt; 10780 1/1; Luln; 15084 1/1;
Rhum, Sq 107; 15095
2/1;
steant Rtbt, Sq 107; 18923 2/
1;
Ltbt Ruin, Sq 112 (4
1
); 20083 1/1; Rtmt, Sq 121;
20097
2/1;
Ltbt Rhum, Sq
119;
20164 1
/1;
Lman+sym, Sq
120 or 121 (spoil); 20167 1/1; Ruin, Sq 120; 20171 1/1;
syn, Sq 119; 20389 1/1; Lsca, Sq 107.
Not found: 5934 1/1; cor; 5971 1/1 ste.
Genus
Nestor
Lesson
Nestor meridionalis (Gmelin, 1788)
MATERIAL:
5595 7/1; prem man Lhum Luln Ltbt Lcpm
caudv, Sq 53; 5710 1/1; Rtmt, Sq 53; 5711 8/1; LRuln
Rrad LRtbt Ltmt Lung Lsca; 5712
2/1;
prem era, Sq 72;
5713 1/1; prem, Sq 70; 5724 1/1; Rtmt, Sq 59; 5731 1/1;
Luln, Sq
60-61;
5735 1/1; Ruin, Sq 53; 5737 1/1; Rtbtd,
Sq
52-53;
5738 13/1; man LRcpm Ltmt Lhum RpLuln
LRtbt Lfemp Rsca v
RII/1,
Sq 57; 5739 1/1; Ltbt, Sq 60-
61;
5760 1/1; prem; 5766 1/1; Rtmt; 5774 1/1; steant, Sq
62;
5775 1/1; steant, Sq 82; 5778
2/1;
Ruin Rrad, Sq 83;
5779 1/1; prem, Sq 78-80; 5783 1/1; steant, Sq
62;
5816 1/
1;
pel, Sq
53;
5832
1/1
;pelbr but complete, Sq
53;
5836 1/
1;
Rhum (transferred to 5758); 5839 1/1; Lcpm, Sq 56;
5849 1/1; steant; 5900 1/1; Ruin, Sq 53; 5907 1/1; Ltbt,
Sq 82; 5951
2/1;
LRhum, Sq 56-57/66; 5953 1/1; Rhum;
5954
1/1
;Rtbtd(br); 5955
1/1;Rtbt;
5959 2/l;LulnLrad;
5960 1/1; Lrad,Sq 54-55; 5961 1/1; Rsca, Sq
58;
5962 1/
1;
fur, Sq 62; 5965 1/1; Lhum, Sq 82; 5966 1/1; crapt, Sq
62;
5967 1/1; steant; 5996 1/1; pelbr; 6068
2/1;
stebrRtbt,
Sq 72; 6069 1/1; steant, Sq 65; 6071 2/1; 2x l/2man, Sq
64;6072 1/1; prem, Sq 84; 6073 1/1; man, Sq
78;
6075 2/
1;
Rhum Rtbt, Sq 66; 6077 1/1; Lcpm, Sq 60; 6078 2/1;
Rtmt Rhumpfrag, Sq 63; 6079 1/1; Rsca, Sq 69; 6080 1/
1;
Rsca; 6111 1/1; steant, Sq 60; 7058 1/1; syn, Sq 63;
7336 1/1; Rhumdsh; 7426 1/1; Rcor, Sq 72; 7427 1/1;
Rcor, Sq 56-57; 7430 1/1; RmanII/1, Sq 53; 7435 1/1;
Lmanll/1,
Sq 5
3;
7467 1
/1;
Rsca; 7473 1
/1;
Lsca, Sq 62-
63;
7476 1/1; Ruin, Sq 54-55; 7477 1/1; Luln, Sq 67;
7484 1/1; Ruin, XXII; 7495 1/1; Rulnl/2, Sq 59; 7501 1/
1;
Lrad, Sq
63;
7502 1/1; Rrad, Sq
64;
7508 1/1; Rradimm,
Sq 66; 7563 1/1; Lpal, Sq 62; 9898 1/1; Luln; 10121 1/1;
Lqua; 10149 1/1; v, Sq 56; 10878 1/1; Lpal; 10879 1/1;
Rtbt, Sq 85; 10880 1/1; Ruin, Sq 85; 10881 1/1; Lhum,
Sq 85; 10882 1/1; pel, Sq 85; 10886 1/1; Rrad, Sq 85?;
12654 1/1; Rhum, Sq 89; 13760
2/1;
pel stefrag, Sq 92;
13761 1/1; prem, Sq 92; 13768 1/1; syn, Sq 92; 14360 1/
1;
Rpal, Sq 68; 14377 1/1; steantfrag, XIV; 15049 1/1;
Ltbt, Sq 110; 15051 3/1; ste LRdshtbt, Sq 109; 15052 4/
1;
Rhum man LRpal, Sq 108;; 15068 1/1; era, Sq 107 (in
black);
15082 1/1; Lhum, Sq 101; 15107 1/1; Rrad, Sq
110;
15124 1/1; Lsca, Sq 102; 16105 1/1; Lcpm, Sq 59;
20069 l/l;Lfem, SqlO9adj 120—121;20081 1/1; Rhum,
Sq 121; 20082 1/1; Rtmt, Sq 120; 20084 1/1; Lhum, Sq
118;
20132 1/1; era, Sq 121 (yellow); 20133 1/1; Rtbt, Sq
120 (yellow); 20156 1/1; Rman, Sq 109; 20173 1/1;
Lacetpel, Sq 120/121? in spoil; 20193
2/1;
Lcor Lcpm, Sq
120;
20380 1/1; Rtbt, spoil heap; 23467 1/1; prem, Sq 62;
36983 1/1; Rhum, Sq
52-53.
Not found: "Nestor n. sp." 5725 1/1; prem, Sq 62;
5736 1/1; Ruin; 5749 2/x; 2uln; 57511/1; Rcpm;
5780 1/1; Rhum; 5956 1/1; Ltbt, Sq 78-80; 5963 1/1;
fur, XIII; 7420 1/1; cor, Sq
83;
7429 1/1; Rcor; 7431 1/
1;
carpphal, Sq
56;
7459 1/1; Rsca, Sq
72;
7503 1/1; rad,
Sq 69; 10115 1/1; pelpt, Sq 65; 10887 1/1; halphal, Sq
85;
12568 1/1; pal, Sq 84; 20401 1/1; manfrag.
Nestor notabilis Gould, 1856
MATERIAL:
5740
2/1;
Lhum, Sq
56-57;
5758
1/1;
pel, XA;
5759 1/1; pel(br but only slightly); 5761 1/1; l/2man;
5762 1/1; l/2man; 5765 2/1; Lhum Lcpm; 5773 4/1; ste
Rtmt pedphal frag, Sq
62; 5831 3/1;
LRcpm uln (missing);
5964 1/1; Lhum, Sq 62; 5969 1/1; Rcor; 6006 1/1; Ruin,
Sq 58; 6014 1/1; Rsca, Sq 58 or 59; 6070 1/1; ste, Sq 63-
65;
6076 1/1; pel, Sq 63-65; 6082 1/1; Rtbt; 6110 1/1;
steant, Sq 77; 6180 1/1; Rhum, Sq 50; 6181 1/1; Lcpm,
Sq 72; 6182 1/1; Lcpm, Sq 58; 6227 1/1; Rcor, Sq 72;
7054 1/1; syn, Sq
60;
7056 1/1; syn, Sq
65;
7057 1/1; syn,
Sq 81; 7059 1/1; Rfem; 7341 1/1; syn; 7452 1/1; Rsca;
7458 1/1; Rsca, Sq 72; 7465 1/1; Rsca, Sq 72; 7485 1/1;
Luln, Sq72; 7497 l/l;Rrad;7564 1/1; Lpal;9961
2/1;
2v,
Sq 51; 10223 1/1; man, Sq 63-65; 10228 1/1; Rhum;
10229 1/1; Rhum; 12132 1/1; Lcpm; 14165 1/1; Lqua,
XIII; 14375 1/1; Rhum, X; 14452
22/1;
LRsca furpt4rib
v Lcor LRrad pel ste LRtbt LRtmt man LRhum, LRuln,
Sq
96-97;
14474 1/1; Lrad, Sq
97;
14550 1/1; Rtbt, Sq 99;
14753 1/1; Rrad; 15047 10/1; pel man Rfem Rtbt Ltmt
Ruin Lrad Rcpm
rib,
Sq 101 (102Ltmt); 15050
6/1;
LRtbt
Rfem LRuln Lhum, Sq 105; 15056 1/1; man (small ind),
Sq 102; 15898 1/1; Lcor; 20099 1/1; vt, Sq 119; 20137 1/
1;
Ltbt, Sq 121; 20212 1/1; Ruin; 20404 1/1; Lulnd, 48.
Not found: 5713 1/1; era, Sq 70; 5763 1/1; fern;
5817 1/1; cor, Sq 52-53; 5826 1/1; hum, Sq 53; 5827 1/
1;
hum, Sq 51?; 6081 4/2; LRtmt Rtbt + cpm of an-
other, Sq 63 and 63-65; 7079 1/1; rad, Sq 72; 13901 5/
1;
fur v carpdig pal man (formerly 5761).
Nestor sp. indet.
MATERIAL:
5764 1/1; crapt, XVII; 5968 1/1; stefrags, Sq
55 and 56; 7416 1/1; crafrfrag; 7553 1/1; rib; 8277 1/1;
crafrag, Sq
68;
10138 1/1; v; 14182 1/1; Rpelfrag, Sq 51;
116
New Zealand Journal of
Zoology,
1997, Vol. 24
14183 1/1; Lpelfrag, Sq 81; 15085 1/1; Linnom, Sq 107
(in black).
Genus Strigops Gray
Strigops habroptilus Gray, 1845
MATERIAL:
5755
1/1;
Lfem; 5790 1/1; ste; 5935 l/l;Rtbt,
Sq
53;
5936 1/1; Rtbt-d, Sq 56-57; 5957 1/1; Ltbt; 5958 1/
1;
Rfemp, Sq 54, 59 or
61;
5995 1/1; pel3frags[= 1], Sq
53;
6015 1/1; man2frags, Sq
60-61;
6037 1/1; era, Sq 63;
6138 l/l;steimm,Sq62;7464
1/1;
Lsca-blade, Sq 63-65;
8279 1/1; Lfemp+sh, Sq 66; 14465 1/1; Rrad, Sq 94;
14551 3/1; man Rtmt vt, Sq 99; 15042 1/1; Ltbt, Sq 108;
15057 1/1; pel, Sq 108; 15058
1/1;
pelpt, Sq 107; 19101 1/
1;
Ltmt, Sq 116; 20076 1/1; Lhum, Sq 119; 20105 1/1;
Rsca, Sq 121; 20136 1/1; pel, Sq 120; 23585 1/1; Luln.
Not found: 7256 1/1; ?hum, Sq 63-65; 7257 1/1;
?hum; 7523 1/1; Lrad; 18912 1/1; Ltbt, Sq 113 (2
1
depth).
Genus Ninox Hodgson
Ninox novaeseelandiae (Gmelin, 1788)
MATERIAL:
6208 1/1; prem, Sq 63; 7299 1/1; tbt, Sq 60.
Genus Sceloglaux Kaup
Sceloglaux albifacies (Gray, 1844)
MATERIAL: 5989 1/1; cor; 6233 1/1; ste, Sq 77; 7469 1/1;
Lsca-blade; 7483 1/1; Luln, Sq 66; 13771 1/1; Lcor, Sq
92;
14453 4/1; Rrad Rhum Ltmt steant, Sq 98-99;
14470 1/1; Luln, Sq 97; 14471 1/1; Ruin; 15045 4/1;
LRfem Lrad Ltmt, Sq 102; 15055 1/1; Rtmt subad, Sq 105;
15083 2/2; LRuln (different sizes), Sq 102; 15087 1/1;
Ltbtpt, Sq 104 near lower edge.
Genus Aegotheles Vigors & Horsfield
Aegotheles novaezealandiae (Scarlett, 1968)
MATERIAL:
7213 1/1; Ltmt, Sq
84;
7214 1/1; Ltmt, Sq 82;
7215 1/1; Rtmtsubad; 7220 1/1; Ltbtdsh, Sq 82; 7236 1/
1;
Rtbtdsh, Sq 69; 7241 1/1; Rhum, Sq 82; 7242 1/1;
Lhum, Sq
64;
7252 1/1; Rfem, Sq
63;
7269 1/1; Rfempsh,
Sq 62; 7626 1/1; Rfempsh, Sq 62; 13772 1/1; Rhum, Sq
92;
14467 1/1; Rhum, Sq 97; 14472 1/1; Ltbt, Sq 97;
15118 1/1; Ltbt, Sq 104; 20100 1/1; Rfem, Sq 119.
Genus Hemiphaga Bonaparte
Hemiphaga novaeseelandiae (Gmelin, 1789)
MATERIAL:
55912/1;
Rcor Rhum, Sq 53; 5592 5/2;
2LRhum Rtmt ste, Sq 55; 5593 6/5; 5R(ld)Lhum, Sq 54
and 55; 5633 1/1; Lsca, Sq 58 or 59; 5686
4/1;
era prem
palatines, Sq
60;
5693 1/1; Lhum, Sq
60;
5694 1/1; Lhum,
Sq 60; 5695 1/1; pel, Sq
60-61;
5697
3/1;
LRhum(lbr)
Lcor, Sq
60-61;
5698 1/1; ste; 5699 1/1; pel, Sq 59;
5700 1/1; Lcor, Sq
52-53;
5701 1/1; stebr, Sq 54-55;
5702 1/1; stebr, Sq
54-55;
5704 1/1; Rhum, Sq
62;
5705 1/
1;
Rhum, Sq 62; 5706 1/1; Lhum, Sq 59; 5707 2/1;
LRhum, Sq 62; 5714 1/1; Rsca, Sq 58; 5719 3/2; Lfem
Lhum Lhumd, Sq 56; 5720
2/1;
Ruin 3fusedv, Sq
52-53;
5721 6/2; Rfem 2L2Rhum 3fusedv, Sq 56-57; 5722 17/
3;
pelpt L2Rcor Lhum Lhump Rrad stefrag L2Rcpm syn
rib Rtmt, LRuln Rhumd; LdLuln not numbered but listed
on card, Sq
59;
5723
14/3;
3Lsca 3fusedv 2Lhum 2Rcpm
2L2Ruln Lfem Lcor, Sq 62; 5729
10/1;
era LRhum LRtbt
LRcor rad(not found) Ruin steant, Sq 53; 5730 1/1; ste
perfect, Sq 62; 5731 1/1; Luln, Sq
60-61;
5740 l/l;Ruln,
Sq
56-57;
5741
1/1; ste, XX; 5742
1/1;
era; 5743
1/1;
pel;
5744
19/1;
pel LRhum LRsca LRtbt LRfem LRcor Ruin
LRtmt Rcpm Rrib sterib, caudv
MII/1;
5745 1/1; ste,
XVIB;
5746 1/1; ste, XVIB; 5747 7/7; 7ste(frags);
5748 11/6; 5L6Rhum; 5749 4/2; 2L2Ruln; 5750 6/4;
4L2Rfem; 5751 6/4; 4L2Rcpm; 5752 4/4; 4Rcor; 5753 4/
1;
LRtmt Rsca Ltbt; 5757 1/1; ste, XX; 5767 1/1; Rtmt;
5771
3/1;
Rhum Lcpm Lrad; 5776
4/1;
Rcpm Rsca-blade
Ruin Lfem, Sq 82; 5777 5/2; LRuln Lcpm 2Rcor(limm),
Sq
52-53;
5778
3/1;
Ruin Rrad ste, Sq
83;
5780
5/1;
Rfem
RcorLRuln(lbr) Rsca, Sq 78-80; 5782 1/1; Rcor; 5795 1/
1;
ste; 5796 4/4; 4steant; 5797
11/1;
pelfrags Lrad (prob-
ably modern bone, included by accident), LRcpm Rfem
2Lsca Rtmt Rhum Luln 3fusedv; 5798 1/1; ste; 5805 2/1;
Rcpm Rhum; 5825 7/2; Rhum 2LRcor Lsca Rcpm Rrad,
Sq 51?; 5828 6/2; 2Rhum Luln Lfem Rcor Lrad, Sq 56;
5838 6/2; L2Ruln Lfem 2Lcpm, Sq
60-61;
5844 6/3;
3Ruln(ppd) 3fusedv LRsca; 5845 3/1; Luln Rcpm Lfem;
5846 3/1; Lfemp Rulnd Rfem, Sq 54-55; 5847 11/3;
L2Rfem 3Lrad Lhum Lcor LRcpm Ruin, Sq
60;
5850 13/
1;
Ruin Lrad fur Lcor 3fusedv 4v 3sterib 2pedphal Rhum,
XX; 5880
2/1;
Rtmt stekeel, Sq
60;
5881
2/1;
Rrad Rcpm,
Sq 56; 5896 1/1; Rtmt, Sq 61; 5897 1/1; Rhum, Sq 61;
5898 1/1;Lhum-p; 5901
1/1;Ruin,
Sq59; 5902 l/l;Luln,
Sq 58; 5903 1/1; Ruin; 5905 1/1; Ruin; 5906 1/1; Luln;
5909 1/1; 3fusedv,Sq
60;
5911 1/1; Lcpm, Sq
52;
5913 1/
l;Lcpm-minor, Sq 62; 5914 1/1; Rcpm, Sq
53;
5920 1/1;
Rtmtd, Sq 51; 5925 1/1; Lfem, VII; 5926 1/1; Lfem;
5927 1/1; Rcor; 5970 5/3; 5stefrag(3ant+2); 5997 1/1;
MII/1,
Sq 54-55; 5998 2/1;
LRMII/1;
6001 1/1; pelfrag;
6002 1/1; pelfrag; 6003 1/1; syn; 6004 1/1; synsac, Sq 60;
6005 1/1; syn, Sq 53; 6007 1/1; Rsca, Sq 60; 6008 1/1;
Rsca-blade, Sq
58;
6009 1/1; Lsca, Sq
56;
6010 1/1; Rsca-
blade, Sq 62; 6011 1/1; 3fusedvbr; 6013 1/1; pelfrag;
6042 16/2; pel ste 2L2Rhum Rtmt MII/1 2LRsca Lcpm
3Lfem
Ltbt,
Sq
72;
6043
8/1;
Rhum Rtmt LRsca(br) Rcpm
Rfem Lcor Ltbt, Sq 66; 6044 5/1; stebr pel Lhum Lsca
Lcpmp, Sq 65; 6045 6/1; pel LRhum ste Lsca Rfem, Sq
68;
6046 6/2; ste Lhum Rsca Lfem Lcor Ltbtimm, Sq 64;
6047 1/1; ste, Sq
81;
6048 1/1; ste, Sq
81;
6049
3/1;
stebr
Rtmtbr Rcor, Sq 81; 6050 19/2; 2stebr 4fusedv
3LRhum(lbr) LRsca LRuln
MII/1
LRrad LRcor, LRcpm
Lfem, Sq
63;
6051 5/2; ste 2Rtmt LRcpm, Sq 69; 6052 1/
1;
Rhum, Sq 50; 6053 7/2; stebr Rtbt pelfrag Rsca 2Lcor
Rfem, Sq 74; 6054
2/1;
LRsca, Sq 65; 6055 1/1;
MII/1,
Sq
80;
6056
5/1;
Rcpm Rhum LRfem Lcor, Sq
84;
6057 9/
4;
Lsca 4L3Rcor Lhum, Sq 72; 6058 10/4; 2L2Rfem
2Rcor4Ltbt; 6059
2/1;
Rsca
Ltbtp,
Sq
62;
6060
2/1;
LRtbt,
Sq 59; 6061 1/1; Rtbt; 6062 1/1; Ltbt, VII; 6063
2/1;
Rtbt
Rradd, Sq
60;
6064 1/1; Ltbt, Sq 56; 6102
3/1;
Lrad stebr
Ruin, Sq 63; 6103 1/1; ste, Sq
81;
6104
2/1;
Ruin steant,
Sq 69; 6105
2/1;
Lrad steant, Sq 64; 6106 1/1; steant, Sq
72;
6107 4/2; Lrad 2Luln stefrag, Sq 72; 6108
2/1;
Rfem
steant, Sq 63; 6109 1/1; stefrag, Sq 65; 6116
4/1;
LRuln
Lrad pLcpm, Sq 67; 6117
4/3;
3LRrad, Sq 84; 6118 3/2;
2RulnRrad,Sq66;6119 1/1; Luln,
Sq61;
6120 l/l;Lrad,
Holdaway & Worthy—Pyramid Valley fossil fauna
117
Sq 74; 6121 3/2; 2LRrad; 6122 1/1; Rulnimm; 6176 2/1;
Lfem Luln, Sq 84; 6203
2/1;
Luln Lhum; 6204 1/1; Rfem,
Sq 77; 6205 1/1; Rfem, Sq 68; 6222 1/1; Rhumd, Sq 72;
6223 1/1; Rhumd, Sq 63; 6228 1/1; Rcor; 7071 1/1;
Rulnimm, Sq 66; 7072 1/1; Rsca; 7073 1/1; Lsca, Sq66;
7074 1/1; Rsca, Sq 77; 7075 1/1; scant, Sq 77; 7076 1/1;
Rrad, Sq 69; 7077 1/1; Luln, Sq 72; 7078 1/1; Lsca;
7218 1/1; Rtmtd, Sq 65; 7298 1/1; Rtbt, Sq 68; 7309 1/1;
Rtbt, Sq
52-53;
7330 1/1; Ltbtp, Sq84; 7342 1/1; syn, Sq
56-57;
7343 1/1; syn, Sq
72;
7344 1/1; syn, Sq
63;
7413 1/
1;
Lfib; 7414 1/1; Lmanramp; 7418 1/1; Lcor, Sq 81;
7421 1/1; Rcor, Sq 64; 7432 1/1; Rcor; 7463 1/1; Lsca;
7471 1/1; Rsca-tip; 7478 1/1; Ruin; 7496 1/1; Rulnd, Sq
72;
7498 1/1; Rrad; 7499 1/1; Rrad, Sq
77;
7500 1/1; Rrad;
7515 1/1; Lradd; 7520 1/1; Lradd; 7527 1/1; Rcorbr, Sq
62;
7539 1/1; Rtbtd, Sq
51;
7540 1/1; Rtbtdfrag; 7546 1/
1;
Ltbtp, Sq 58; 7549 1/1; Ltbtdimm, Sq 60; 7560 1/1;
furbr, Sq 54-55; 8278
2/1;
Rhum 3fusedv, Sq
66;
9558 1/
1;
pel, XX; 9559 1/1; pel, Sq
48.1;
9611 1/1; ste; 9623 1/
1;
Lcor; 9624 1/1; Lcor, XX; 9628 1/1; Rtbtpt, XXII;
9846 1/1; Ruin; 9847 1/1; Luln; 9848 1/1; stefrag, XX;
9849 1/1; pelfrag, XXII; 9853 1/1; Rcpm, XXII; 9892 1/
1;
Rtbt; 9899 1/1; Rrad; 9900 1/1; Lsca, XXII; 10126 1/
1;
Rhump, Sq 65; 10140 1/1; vimm, Sq 78-80; 10150 1/
1;
vimm (5th c); 10153 1/1; vimm dorsal, Sq 54-55;
10156 1/1; vimm (5th
c);
10231 l/l;pelpt,Sq74; 10232 1/
1;
steant, Sq 84; 10241 1/1; pelfrag, Sq 63-65; 10243 1/
1;
stefrag, Sq
81;
10247 1/1; pelfrag, Sq 56-57; 10248 1/
1;
pelfrag, VII; 10249 1/1; rib; 10250 1/1; stefrag;
10251 1/1; stefrag; 10318 1/1; Rhumd; 10872 1/1; pel, Sq
85;
10874 1/1; Ltmt, Sq 85; 10875 1/1; Rcpm, Sq 85;
10877 1/1; Lfem, Sq 85; 10885 1/1; Ruin, Sq 85; 12133 1/
1;
Rscabr; 12656 1/1; pelbr, Sq 89; 12658 1/1; Lsca, Sq
88;
12659
2/1;
Lsca-bladeRrad, Sq86; 12871 l/l;Lfemp,
Sq 48.6; 13762 8/3; pel 3ste(pts) Lhum Luln Lfem (F)
Lcor(13762H) Lcpm Rtbt, Sq 91-92; 14376 1/1; Rhump,
XX; 14455 1/1; pel, Sq 94; 14456 l/l;pel(3pcs=most),Sq
94
;
14457 i/i; Rhump, Sq 94; 14458 1/1; Luln, Sq 94;
14473 1/1; Lsca-blade, Sq 97 (in black soil); 14475 1/1;
Rhump, Sq
97;
15061
7/1;
ste Rtmt 2fusedvt LRuln RMII/
1,
Sq 108; 15065 1/1; Rhum, Sq 110; 15066 1/1; Lfem,
Sq 105; 15067
4/1;
Rtmt Lsca Lcor Ltmtp, Sq 107;
15070 6/1; ste LRcor pelpt Lsca Luln, Sq 107 see note;
15078 1/1; Rhum, Sq 109; 15079 1/1; steant, Sq 109;
15080 3/1; Lcor Luln Rhump, Sq 110; 15099 1/1; Rrad,
Sq 110; 15100 1/1; Rrad, Sq 109; 15101 1/1; Lrad, Sq 110;
15102 1/1; Lrad, Sq 104; 15103 1/1; Lrad, Sq 107;
15104 1/1; Rrad, Sq 109; 15106 1/1; Lhump, Sq 107 (in
black);
15120 1/1; Rtbtp, Sq 107 (in black); 15121 1/1;
Lulnp, Sq 104; 15123 1/1; Rsca, Sq 104 (outside);
15125 l/l;Lscaimm, Sq 107; 15127 l/l;Rscap, Sq 109;
16104 1/1; synsac; 16150 1/1; Rtbtd, Sq 84 black peat
layer; 18913 6/2; stept pelpt Rcorant LRtmt(different
lengths) rib, Sq 111 4-5'; 18915 1/1; pel, Sq 113 4';
20058 4/2; 2Rcor Rcpm rib, Sq 119 yellow; 20059 3/2;
Lhum Lhumd Lfem, Sq 121; 20060 1/1; Rtmt, Sq 120;
20064 1/1; stept(ant, L+keel), Sq 121; 20065 1/1; stept(-
Llat+ptkeel), Sq
121;
20066 1/1; steant, in spoil; 20067 1/
1;
Rcpm, Sq 109; 20068
6/1;
Rmanp 2phal ribpt Lqua cun,
Sq 119 120 cm; 20071 1/1; Rcpm, Sq 109; 20074 1/1;
Lhumd, Sq 120; 20075 3/1; Rtbt Ruin Rsca, Sq 119;
20077 1/1; Rrad, Sq 119; 20087 6/1; pelpt(-left) Lcpm
3fusedvbr ste2frag Rcpmp Rulnp, Sq 121; 20088 2/1;
Lcor-pLfem, Sq
121;
20101 1/1; Rhum, Sq
121;
20110 4/
1;
Lcpm-minor 2v Rtbtd, Sq 117 lower peat; 20139 1/1;
Lrad, Sq
120;
20141 1/1; Lcor, Sq 120 in yellow; 20142 3/
1;
Ltmt Rcor Rulnpt, Sq 121 black; 201436/1; Rhum 3vt
Lsca pelfrag, Sq 121 yellow; 20157 1/1; steant, in spoil;
20158 1/1; Rfem, Sq 110 (in old spoil); 20160
2/1;
Rsca
Lcor, Sq
121
in yellow; 20402 1/1; Rscap-d, VIII on bone;
20405 1/1; Rtbtd, Sq 92; 23495 1/1; Ruin, Sq 121;
28795 1/1; cv, Sq 102; 28796 5/1; 2cv caudv phal
ste(Rproc), Sq 107.
Not found: 5829 3/1; 3fusedv Rsca Ltbt, Sq 62;
5912 3/2; cor 2Rcpm, Sq 58; 7496 1/1; tmt, Sq 61;
10873 1/1; v, Sq 85; 13780 1/1; Rhum, Sq 92; 14380 1/
1;
Rrad. "62" Lsca, presumably Sq 62, number unknown.
Genus Gallirallus Lafresnaye
Gallirallus australis (Sparrman, 1786)
MATERIAL: 5594 1/1; Rtbt, Sq 58; 5799 1/1; pel; 5824 1/
1;
pel; 5830
2/1;
LRtmt, Sq 62 and 63; 5833 2/1; LRtbt;
5834 1/1; Rtbtd, Sq 56-57; 5851 4/2; Ltmt Rfem LRtbt;
other tmt now 6839; 5853 2/2; Rfem Lfembr, Sq 56-57;
5873 1/1; pelbr; 5877 1/1; Ltbt, Sq
60-61;
5916 1/1; Rfem,
Sq 61; 5918 1/1; Lfemd; 5919 1/1; Rfemd+sh, Sq 62;
5938 1/1; Ltbt, Sq 58; 5939 1/1; Rtbtd, Sq 56; 5940 1/1;
Ltbtd, Sq
52-53;
5973 1/1; pel[=syn+plates]; 5974 1/1;
pel;
5975 1/1; Rcor, Sq
58;
5976 1/1; Rcor, Sq 62; 5994 1/
1;
Rtmt, Sq 59; 6085 1/1; Rtmt, Sq 63; 6086 1/1; Rtmt,
Sq 69; 6087 1/1; Ltmt, Sq 64; 6088 1/1; Ltmt, Sq 72;
6089 1/1; Ltmt, Sq
63;
6097
1/1;
pel, Sq
68;
6098 1/1; pel;
6114 1/1; era, Sq 66; 6130
2/1;
Rtbt Rfem, Sq 63-65;
6131 1/1; Ltbt, Sq 63; 6132 1/1; Ltbt, Sq 63; 6133 1/1;
Ltbt, Sq 66; 6134 1/1; Rtbt, Sq 72; 6135 1/1; Ltbt, Sq 66;
6136 1/1; Rfem; 6137 1/1; Lfem; 6139 18/1; ste LRtmt
Ltbt LRfem 12v, Sq 84; 6146 1/1; Lfem, Sq 72; 6157 1/
1;
Ltbtp+sh, Sq 64; 6172 1/1; Rtbtp, Sq 65; 6198 2/1;
LRhum; 6201 1/1; Lfem; 6202 1/1; era; 6644 1/1;
Rcorimm, Sq 81; 6645 1/1; Rcor; 6647 1/1; Ruin, Sq 51;
6839 l/l;Rtmt(wasptof5851);7060 1/1; Lhum, Sq 51;
7061 1/1; Lhum, Sq 60; 7062 1/1; Lhum, Sq 63-65;
7063A 3/1; 3phal; 7065 1/1; Lcpm, Sq 50; 7066 1/1;
Rcpm; 7067 1/1; Lcpm; 7068 1/1; Lcpm; 7069 1/1; Lcpm,
XI;
7070 1/1; Rrad; 7346 1/1; ptsyn; 7258 1/1; Lhum, Sq
60-61;
7318 1/1; Rtbtd, Sq 65; 7331 1/1; Rtbtd; 7334 1/
1;
Rtbtp, Sq
58
(by
date);
7340 1
/1;
syn, Sq
60-61;
7456 1 /
1;
Lsca, Sq 59; 7517 1/1; phalanx; 7565 32/1; 9phal fib
LRsca Rcor 13rib lstmet 3caudv digpha! [radfrag -, not
GAAU]manfrag,Sq62;9815 l/l;prem;9816 1/1; Lhum,
Sq 58; 10122 1/1; v; 10123 1/1; v; 10124 1/1; v, Sq 72;
10227 1/1; stept; 10137 1/1; v, Sq 78-80; 10270 1/1;
pelfrag; 10768 1/1; Rtbt, Sq 58; 10769 1/1; Ltbtd (pt of
5938),
Sq 58; 10770 1/1; Lfem, Sq
58;
10771
2/1;
LRtmt,
Sq 58; 10772 1/1; Rtmt, Sq 58; 14458 1/1; Rfem, Sq 97;
15044a 1/1; Ltbtp; 15059 5/1; Ltbt LRtmt Rfem pedphal
slightly subad, Sq 107; 15076 1/1; Ltbt, Sq 109; 15081 1/
1;
Lhum, Sq 101; 15086 eggfrags, Sq 108; 18922 1/1;
Rtmt, Sq 115 spoil; 18924 1/1; Rrad, Sq 112 (4' depth);
20062 3/1; LRtmt Lcpm, Sq 120; 20063 2/1; Rfem(DG)
Lfem(RJS), Sq 120; 20078
x/1;
most of skeleton LRtmt
118
New Zealand Journal of
Zoology,
1997, Vol. 24
LRtbt pel man Rfem LRhum ste Luln Lcpm, Rfib alphal
Lrad Lcor manyv manyribs manypedphal, Sq 120 in yel-
low; 20079 1/1; Ltmt, Sq 120; 20085 1/1; Lhum, Sq 121;
20152
1
/1;
Luln, Sq
121;
20161
2/1;
Rfemp Luln, in spoil;
20174 1/1; Lrib, Sq 120; 20180 1/1; Lmanp, Sq 120;
20387 1/1; Ltmt, Sq 84; 36286 1/1; Lscasubad.
Not found: "Gallirallus minor" 7313 1/1; Luln, Sq 64.
"Gallirallus sp." 6066 1/1; prem, Sq 72; 7444 1/1; Rrib;
7879 1/1; fembr, Sq 66; 10125 1/1; pelpt. "Gallirallus
australis 5728 1/1; ste, Sq 58; 5917 1/1; Lfemimm, Sq
51;
9619 1/1; Lfemimm; 20061 3/2; 2pelant Ltbt, Sq
120.
Genus Porphyrio Brisson
Porphyrio hochstetteri (Meyer, 1883)
MATERIAL:
5789 1/1; Ltmt; 5842 1/1; Lfem, IX; 5921 1/
1;
Rtmtimm; 5922 2/1; Rtbt Rfemp+Rfemd, Sq 56;
6041 2/1; premtip mantip [immature], Sq 83; 7453 1/1;
Rsca, Sq 59; 15039 1/1; Rfem, Sq 108; 18930 1/1;
Lulndsh, Sq 112 4'; 33833 Rmanpsubad.
Genus Fulica Linnaeus
Fulicaprisca Hamilton, 1893
MATERIAL:
5589
7/1;
pel LRfem Ltmt Ltbt ste Rtbt, Sq 59,
58,
57; 5590 1/1; Ltbt, Sq 56; 5768 1/1; Rtbt; 5769 1/1;
Lhum, VII; 5770 1/1; Ltbtp; 5875 2/1; LRhum, Sq 59;
5904 1/1; Luln, Sq 59; 5915 2/1; LRtmtimm; 5952 1/1;
Lhumd(subad); 6155 1/1; Lhumimm, Sq 72; 7263 1/1;
Lcpm, Sq 59; 7347 1/1; immLtbtd; 7428 1/1; Lcor, Sq 59;
7439 1/1; rib, Sq 59; 7440 1/1; rib, Sq 59; 7448 1/1; rib,
Sq 59; 7449 1/1; rib, Sq 59; 7450 1/1; rib, Sq 63-65;
7481 1/1; Luln, Sq 66; 9627 l/l;Rtbt(shbr), Sq
51;
9756 1/
1;
Rtbt; 9757 1/1; pel; 9759 1/1; Rtmt; 9760 1/1; Lfem;
9761 1/1; Lhumsubad; 9762 1/1; Lhum; 9763 1/1; Rcor;
9766 1/1; Rtmtjuv; 9767 1/1; Lfem; 10127 1/1; v, Sq 59;
10130 1/1; v, Sq 59; 10274 1/1; phal, Sq 59; 13759 1/1;
Rtbt, Sq 94; 20094 1/1; manant, Sq 119; 20135 1/1; pel,
Sq 121; 20403 l/l;Rtmtp.
Not found: 7429 1/1; Rcor; 9758 1/1; ste; 14458 1/1;
Luln, Sq 94. Fulica cf. atral 9627 1/1; Rtbt(shbr).
Genus Gallinula Brisson
Gallinula hodgenorum (Scarlett, 1955)
MATERIAL:
5803
2/1;
pel Rtmt [Scarlett
(1955b)
lists 5802
as tmt, 5803 as pel]; 5804
1/1;
pel [repeated as 5803 in
Scarlett (1955b)]; 5829
1/1;
Lfem, Sq 62; 5854
1/1;
Ltbt;
5943 1/1; Rtmt, Sq 58; 5977 1/1; Rtbt, Sq 62? (or 60);
5978
1/1;
Rtbt,
VII;
5979
1/1;
Ltbtimm, Sq
60-61;
5985 1/
1;
Rtbt, Sq 62; 6192
1/1;
Rfem, Sq
56;
6193
1/1;
Lfem, Sq
59; 6194 1/1; Rfem; 6195 1/1; Rfem, Sq 68; 6196 1/1;
Lfem,
Sq 58; 6197
1/1;
pel, Sq 72 [Holotype]; 6646 1/1;
Ruin,
Sq 54 [6647 in Scarlett (1955b)]; 7207 1/1; Lfem;
7238
1/1
[7328 Rhum in Scarlett
(1955b)];
Lhum;
7304 1 /
1;
Rtbtimm [L missing]; 7316 1/1; Ruin, Sq 80; 7548 1/
1;
Ltbtdimm, Sq 68; 8281 1/1; Ltbt (see 8282), Sq 66
[8291 in Scarlett (1955b)]; 8282 1/1; Ltmt (may be same
bird
as
8281),
Sq 66; 8284
1/1;
Ltmtimm (same individual
as 8285?), Sq 66; 8285 1/1; Lhumimm (same individual
as 8284?) [Rhum in Scarlett (1955b)], Sq 66; 9842 1/1;
Lfem,
XXII;
9852 1/1; Ltmtimm,
XXII;
10774 1/1; Ltbtd,
Sq
72;
10776 1/1; Rfemp, Sq 78-80; 15048
3/1;
Rtbt(102)
Ltbt(101)Lfem(103), SqlOl, 102, 103; 15075 1/1; Ltbt,
Sq 109; 15129 1/1; Rsca, Sq 107; 20056 2/1; Lfem Ltbt,
Sq 119 in yellow.
Not found: 6198 2/x; 2hum, Sq 84; 7305 1/1; tbtimm;
7550 1/1; humd; 8283 1/1; cor, Sq 66; 10255 l/l;Rfemd,
Sq 52, 53, 56 or 57. Scarlett (1955b) lists 9817 as a
tibiotarsus; thia has not yet been reconciled with any of
the material still in the collection.
Rallidae sp. indet.
MATERIAL:
9817 1/1;
tbtimm.
Genus Aptornis Owen
Aptornis defossor Owen, 1871
MATERIAL:
5383 1/1; Rtbt,
XVIB;
5384
2/1;
v Lhum, XIX;
6016
103+/1;
era prem man 2fem 2tbt 2tmt 2fib 2cor hum
15v 2caudv, ?halluxphal 31trachring larynx fusedtendons
15rib 24?pedphal, Sq 62; 6018
93+/1;
era man ste pel
Lhum 2tbt 2tmt 2fem 19v 4caudv 15rib, 23trachring 2fib
LRsca 2cor fusedtend 23pedphal 2halluxphal, Sq 65;
6019 8/1; Ltbt Ltmt Lsca Rfib 2phal ribpt rib, Sq 56, 58,
59;
6020 1/1; rib, Sq 60; 6021
51+/1;
pel LRtbt LRtmt
stebr man LRcor Rfib 2sca 15 v LRfem LRhum, fusedtend
20rib 12pedphal, 48.9A; 6025
94/1;
pel LRfem LRtbt ste
LRhum LRtmt Lcor LRfib 18pedphal, 35trachring larynx
LRsca 3hallphal digitphal
21
v, VIIA; 6028
2/1;
Lfib phal;
6029 1/1; Lsca(malformed), Sq 78; 6030 1/1; pel, XVIII;
6031 7/1; pel LRfem 2cor 2v, VIIB; 6032
111+/1;
pel
LRtbt 21 v 3caudv LRtmt LRfem 2hum LRfib 4digitphal
sea, ?vomer40trachring fusedtend 23pedphal 3hallphal ste
LR cor era man,
48.1 Oe;
6034 1/1; Rhum, Sq 66; 6035 2/
1;
pel Rfem, Sq 69; 6036
2/1;
Lhum digitphal, Sq 81;
6444 3/1; 2?digphal rib; 7541 1/1; sternal rib, VII;
11467 l/l;uncproc; 11469
2/1;
2rib(frags), Sq 59; 14454
nearly complete skeleton (-femora), Sq 96-97; 15038 1/
1;
Rtmt, Sq 109 (in black); 15043 1/1; rib, Sq 103;
16140
2/1;
cra(worn) prefr, Sq 94; 20073 1
/1;
rib3pcs, Sq
119 in grey; 20131 1/1; Lhum, Sq 121 1'7" in black;
20153 1/1; ribste, Sq 121.
Not found: 6033
91+/1;
pel man 2tbt 2tmt 2hum 15v
3caudv 16rib 25trachring sea, ?vomer 2digitphal 2fib
cor fusedtend 13pedphal hallphal era prem, ste (Sq 66
larynx, doubtfully associated), Sq 64; 7051 10/1;
lOtrachring, Sq 64; 7052 1/1; rib, VII; 7542 1/1; pubis;
9521 1/1; v, Sq 72; 9563 3/1; phal 2rings,
48.1;
9902 1/
1;
caudv; 10883 1/1; phal, Sq 85; 11470 1/1; pubproc,
Sq 59; 14181 1/1; Lcpm.
AMNH: 7300 1/1; "almost complete skeleton"—era
man 12v "pt
coccyx"
"most of shoulder
girdle"
ste pel "leg
and foot bones" 12r 2uncproc trachrings hyoid osstend
2sesamoids; R. C. Murphy, Feb 1948.
Genus Himantopus Brisson
Himantopus novaezelandiae Gould, 1841
MATERIAL: 6124 11/1; ste LRtmt Rtbt LRhum fur Luln
LradpedphalRcpm, Sq
69;
7290 1/1; Lfem, Sq
62;
7433 1/
1;
Lpelpt(ischpubetc); may belong with 6124 (Scarlett
1955a).
Not found: 7267 1/1; Rcpm, Sq 69.
Holdaway
&
Worthy—Pyramid Valley fossil fauna
119
Himantopus
sp.
K4ATERIAL:
15130 1/1;
Lcorimm,
Sq 107.
Genus Circus Lacepede
Circus eylesi Scarlett,
1953
MATERIAL:
5687 19/1; LRtbt LRhum Ltmt LRrad Luln
LRcor
man
4phal(lung)
4rib,
Sq
60;
5688
2/1;
RtbtRhum,
Sq 60; 5689
14/1;
LRtbt LRtmt
Ruin
LRfib v 6phal(4ung),
Sq 60; 5690 52/1; ste pelbr Lhum LRfem LRulnfurLsca
LRcor LRcpm LRrad LRtmt
Wv
(2,
4,
5,
10, 14, 15,
16,
17,
19) man,
pyg
2alphal 9rib 12phal(4ung),
Sq 62;
5692 1/1;
ste;
5703 8/1; Rtmt Lsca Rrad
fib
Lcpm 3rib,
Sq
62;
5716 5/1; Lhum (PARATYPE); Rhum Rtmt Lcor
Rrad
(NOT
paratypes),
Sq
56-57, 111;
5717 1/1;
pelbr,
Sq
53; 5718
1/1; Rfem,
Sq
58;
5732
1/1;
Rtbt,
Sq 56;
5733
111;
Ltbt,
Sq
58; 5848
111;
Rcpm,
Sq
60; 5908
111;
Rcpm,
Sq
60; 5941
1/1;
Rtbtpimm,
Sq
52-53;
6039
211;
era prem,
Sq 64
[Holotype]; 6040
211;
pel
Ltbt,
Sq 65;
6067
111;
Lfem,
Sq 63;
6145
111;
hum,
Sq
72; 6220
1/1;
Lfib;
7248
1/1;
Rulnd,
Sq 63
[was C.
approximans];
7310 1/1; Ltbtp,
Sq
72; 7348 1/1; Rtbtdimm, 52-53
[was
C.
approximans]; 7437
1/1; rib, Sq
51; 7442
1/1; rib, Sq
63-65;
7443
1/1; rib, Sq
59; 7460
1/1;
Rsca,
Sq 58 [was
C. approximans]; 7528
x/1;
steribs
cf.
5690,
5703,
7551,
Sq
62;
7529 1/1; rib; 7551 1/1;
steribef.
5690,
5703,
7528,
Sq 62; 7552 1/1; rib,
Sq
52-53;
7554 1/1; rib;
10116 1/1;
pelpt,
Sq
65; 10117 1/1;
furpt;
10119 1/1;
v, Sq
63;
10120
1/1;
v,
Sq 60;
10870
1/1;
fur, Sq
85;
13758 6/1;
Rtmt pelpt 3phal Rcpm,
Sq
92 (1—3ft, bl-y); 13758
7/1;
v
sea stefrag sterib 2rib pretip, Sq
94;
14191 1/1; Lfurp
(cf.
5687,
5688, 5689),
Sq 60 or
62; 15040
2/1;
era
steant,
Sq
109;
15046
1/1;
Luln,
Sq
110-108 border; 15069
1/1;
Lcpmpt,
Sq
107; 15105 1/1;
Rcarpal,
Sq
107 (in
black);
18926
1/1; rib,
Sq
115 (2
1
);
20138
1/1;
Rtbt+tbtfrag,
Sq
122 (black) [was
C.
approximans]; 28794 1/1; vtimm,
Sq
107.
Circus
sp.
indet.
MATERIAL:
6150
1/1; Lulnp,
Sq
63; 6234
1/1;
fibpt.
Genus Harpagornis Haast
Harpagornis moorei Haast,
1872
MATERIAL:
5684
7/1;
era prem prefr LRquj
LRpal?,
Sq 51;
5685 8/1;
era
prem LRqua
3v
man,
Sq
68; 5736 1/1;
max
proc;
6012
1/1; ste5frag,
Sq
54; 6177 1/1; stefrag assoc
with 6178,
Sq
84;
6178 1/1; stefrag associated with 6177,
Sq 66; 9845
1/1;
Lqua; 12355 1/1; man, VIH; 12356
1/1;
Lquaj;
28366
1/1;
Lfem,
Sq 60.
Genus Falco Linnaeus
Falco novaeseelandiae Gmelin,
1788
MATERIAL:
6216
1/1; Ltbt, Sq64; 6219
2/1;
Rhum Rtbtd,
Sq
66;
6762
1/1; era;
7479
1/1;
Luln,
Sq 68;
7555
1/1;
Rfemp,
Sq
63.
Genus Poliocephalus Selby
Poliocephalus rufopectus (Gray,
1843)
MATERIAL: 5894 1/1; Ltbt,
Sq
56-57; 5895 1/1; Rtbt,
Sq
54-55; 5895 1/1; Rcpm,
Sq 62;
5972 1/1; stebr; 6096
1/
1;
Rtbtimm,
Sq
68; 7246
1/1;
Ltbtimm,
Sq
72; 7260
1/1;
Lcpm; 7277
1/1;
Rfemsubad,
Sq 72;
14466
1/1;
Rhumimm
(cf.
7246),
Sq 97.
Genus Egretta
T.
Forster
Egretta alba Linnaeus,
1758
MATERIAL:
7510 l/l;Lradp, Sq51.
Genus Xenicus Gray
Xenicus
sp.
MATERIAL:
14468 1/1; Rfem, Sq 97.
Genus Traversia Rothschild
Traversia lyalli Rothschild,
1894
MATERIAL: 7225
1/1;
Ltmt,
Sq 62.
Genus Anthornis Gray
Anthornis melanura (Sparrman,
1786)
MATERIAL:
6206 1/1; ste, Sq
67;
7227 1/1; stebr; 15131
1/
1;
Rtbtimm,
Sq
107 (possible); 20382 1/1; Rtbtdsh, spoil
heap.
Genus Prosthemadera Gray
Prosthemadera novaeseelandiae (Gmelin,
1788)
MATERIAL:
5814 1/1; man: 5888
2/1:
LRtbt. XI1L 5890
1/
1;
Ltmt,
Sq
78-80; 5891
1/1;
Rhum; 5892
1/1;
Rcor,
Sq
57:
5899 1/1: Ruin. Sq
52-53:
6126 1/1; steant, Sq 63-65;
6128 2/l;LtbtLhum,Sq81;6129 1/1; Rtbt, Sq
69; 6141
1/
1;
Ltbt, Sq
70;
6144 1
/1;
Lhum; 6209 1
/1;
steant (subad),
Sq 66;
6210 1/1;
stefrag;
6214 1/1;
tmtbr; 6221
1/1; era,
IX (black layer);
7211
1/1;
Ltmt,
Sq 58;
7219 1/1;
Ltbtimm,
Sq
56:7235 1/1: Rtbt: 7244 1/1; Lhum,
Sq 60;
7245 1/1; Lhump,
Sq
63;
7250 1/1; Rfem,
Sq
62; 7251
1/
1;
Rfem,
Sq
54;
7253
1/1;
Rfempsh; 7300
1/1;
Rtbt;
7545
1/1;
Rtbtp,
Sq 70;
10773
1/1;
Ltbt,
Sq
63-65;
10884 1/1: pel.
Sq
85; 12661 1/1: Ltbtd.
Sq
86: 13765
1/
1;
Ruin,
Sq
92; 13766
1/1:
Ltbtp subad.
Sq
92; 14459
1/
1;
Rtbt,
Sq
94;
14461 1/1;
Ruin,
Sq
97: 15097
1/1;
Ltbt.
Sq 110; 15119 1/1; Ltbtd,
Sq
107; 15122 1/1; steant(imm),
Sq 104; 19102 3/1; Rtbt LRtmt,
Sq
116-117; 20089
1/1;
Rtbt,
Sq
121;
20096 3/1; Rtmt steant Lrad,
Sq 119;
20146 3/1; Rhum stebr Rtbtbrsh,
Sq
120;
20154 2/1;
LtbtdshRtmt,
Sq 121.
Not found:
6125
1/1; manbr,
Sq 69.
Genus Petroica Swainson
Petroica australis (Sparrman,
1788)
MATERIAL:
5928 1/1; Rtbt; 6207 1/1; Ltbt, Sq 66; 6212
1/
l;Rman+sym, Sq 64; 7247 1/1; Luln,
Sq
63-65; 7254
1/
1;
Rfemdsh; 7255
1/1;
Lfem,
Sq
60; 13764
1/1;
Ltbt,
Sq
92;
15113
1/1; Ltbt,
Sq 107.
Not found: 6213
1/i;
era, Sq 58.
Petroica macrocephala (Gmelin,
1789)
MATERIAL:
6211 1/1; Rtbtdsh, Sq 67.
Genus Mohoua Lesson
Mohoua novaeseelandiae (Gmelin,
1789)
MATERIAL:
7224
1/1;
Rtbt; 7226
1/1;
Ltbt,
Sq 66;
20170 1/1; Ltbt,
Sq 121.
120
New Zealand Journal of Zoology, 1997, Vol. 24
Genus Corvus Linnaeus
Corvus moriorum Forbes, 1892
MATERJAL:5811
1/1; Luln; 5812 1/1; Rtbt; 5882 1/1; hum,
Sq 56-57; 7423 1/1; cor, Sq
60-61;
7513 l/l;Lrad.
Not found. 7514 1/1; l/2radimm; 7530 1/1; furbr;
7536 1/1; rib, Sq 60; 9958 1/1; manpt.
Genus Callaeas J. R. Forster
Callaeas cinerea (Gmelin, 1788)
MATERIAL:
5813 1/1; ste; 5878 1/1; Ltbt; 5884 1/1; Ltbt;
5885 1/1; Ltbt, Sq
60-61;
5923 1/1; Rfem, Sq62; 5924 1/
1;
Rfem; 5991 1/1; Lsca; 6000 1/1; pelbr + ste (nonum);
6092
2/1
;Ltmt Rtbt, Sq
63;
6093 1/1; Ltmt, Sq68; 6094 1/
1;
Rtmt, Sq 77; 6095 1/1; Ltmt; 6115 1/1; era, Sq 72;
6159 1/1; Ltbtp, VII; 7537 1/1; Ltbtd; 9562 1/1; Ltbt;
10236 l/l;Linnom, Sq 56; 10237 1/1; Rtbtd; 10239 1/1;
pelfrag; 10871 1/1; Rtbt, Sq 85; 13763 2/1; pel Rcor, Sq
92;
15053 1/1; man, Sq 107; 15054
2/1;
Rtbt Lhum, Sq
105;
15062 1/1; man, Sq 108; 15063 1/1; mantip, Sq 102
(in black); 15109 1/1; stept, Sq 107 in black; 15128 1/1;
Lsca, Sq 107; 18921 2/1; Ltbt pelpt, Sq 115 4-5' depth;
19103 1/1; stept, Sq 117?; 20057 7/1; man [3pcs
Rartic+tip] Rhum Lsca LRfem Rtbt Ruin, Sq 119 in yel-
low; 20090 1/1; Ruin, Sq 121; 20134 3/1; Lfem Ltmt
Linnom, Sq 120; 20168 1/1; Lhumpsubad, Sq 121;
20178
2/1;
2phal, Sq
121;
20381 1/1; Rtbt, spoil heap.
Not found: 5810 1/1; prem; 7233 1/1; premtip, Sq 63;
7239 1/1; hum; 7240 1/1; hum, Sq 53.
Genus Philesturnus Geoffrey St.-Hilaire
Philesturnus carunculatus (Gmelin, 1789)
MATERIAL:
5930 1/1; Rfem, Sq 56; 6215 1/1; Ltmt;
7271 1/1; Lfem; 7474 1/1; Lsca; 10220 1/1; Ltmtp;
15096 1/1; mansym, Sq 110; 15110 1/1; Rilioisch plate,
Sq 107; 15111 1/1; ste, Sq 110; 18916 1/1; Ltbt, Sq 111
(4').
Not found: 7212 1/1; Rtmt.
Genus Turnagra Lesson
Turnagra capensis (Sparrman, 1787)
MATERIAL:
5886 1/1; Rtbt, Sq 59; 5887
2/1;
Ltbt Rfem,
Sq 62; 5889 1/1; Rtmt; 6127 1/1; Ltbt, Sq 84; 6143 1/1;
Rfem, Sq 74; 10888 1/1; Luln, Sq 85; 12662 1/1; Ruin,
Sq 86; 15093 1/1; Ltmt, Sq 105; 15094 1/1; era, Sq 110;
20383 1/1; Rcor.
Not found: 15092 1/1; Rtbt, Sq 108. A bone labelled
15092 is recent, from a turdid, probably Turdus merula,
and seems to have been included in the collection by mis-
take.
The "original" 15092 has not been found.
Genus Bowdleria Rothschild
Bowdleria punctata (Quoy & Gaimard, 1830)
MATERIAL:
7223 1/1; Ltbt, Sq 82; 7232 1/1; pel, Sq 57;
9894 1/1; Ltbt.
Sp.
indet. (all seen)
MATERIAL:
5459 1/1; rib; 5466 1/1; rib; 6365 1/1; rib;
7064 1/1; phal; 7221 1/1; Rtmtimm, Sq 82; 7306 3/1;
LRtbt cpm allimm; 7324 1/1; rib; 7325 1/1; rib, Sq 63-
65;
7326 1/1; rib, Sq 59; 7327 1/1; rib; 7350 1/1;
pelpt(2ischpubimm), XVIII; 7441 1/1; Rrib; 7445 1/1;
Rrib,
Sq 69; 7451 1/1; ribbr; 7452 1/1; rib; 7454 1/1; rib,
Sq
60-61;
7470 1/1; Rscabrimm, Sq 63; 7521 1/1; rib;
7532 1/1; rib; 7536 1/1; rib; 7556 1/1; rib, Sq
63;
7557 1/
1;
rib, Sq
52-53;
7559 1/1; rib;
7561
1/1; tbtimm?, Sq 52-
53;
9850 1/1; rib, Sq 67; 9851 1/1; rib; 9904 1/1; tbtimm,
Sq 66; 10275 1/1; phal; 11056 eggfrag; 11474 fib;
14464 1/1; egg (2/3 in pieces, 3' down in yellow), Sq 95;
18927 eggfrags, Sq 113; 18928 eggfrags, Sq 115;
19104 2eggfrags, Sq 116-117?; 20086
4/1;
4v, Sq 120;
20165 1/1; Ltbtimm, Sq 120 or 121; 20166 1/1;
Rcarpimm, Sq 121; 23543 eggfrags, Sq 110; 28797 1/1;
vt, Sq 109; 28798 1/1; vt, Sq 103; 33541 vtimm, spoil
heap;
33542 1/1; phal.
Unidentified (not all seen)
MATERIAL:6161
1/1;tbtimm,Sq63-65;7228 l/l;tbtimm,
Sq
58;
7229 1/1; tbtd, Sq60; 7231 1/1; tbtp; 7234 1/1; tbt,
Sq 84; 7243 1/1; hum, Sq 63; 7249 1/1; rad; 7265 1/1;
Lcpmimm; 7266 1/1; Lcpmimm; 7270 1/1; Rfempsh, Sq
62;
7272 1/1; cpm, Sq
60;
7274 1/1; Rcpm, Sq 82; 7275 1/
1;
Lcpmimm; 7278 1/1; Rfemsubad; 7280 1/1; Lfemimm,
Sq 59; 7284 1/1; Lfemimm, Sq 62; 7286 1/1; Lfemimm,
Sq
66;
7288 1/1; Lhumimm; 7293 1/1; tbtimm, Sq 54-55;
7294 1/1; tbtimm, Sq
67;
7295 1/1; tbtimm, Sq
56;
7296 1/
1;
tbtimm, Sq
63;
7297 1/1; tbtimm; 7302 1/1; tbtimm, Sq
57;
7303 1/1; tbtimm; 7307 1/1; tbtimm; 7317 1/1;
digphal; Y 7323 1/1; Rsca, Sq 65; 7337 1/1; syn, Sq 63;
7339 1/1; syn, Sq 50; 7349 1/1; tbtimm, Sq 77; 7351 1/1;
fib; 7355 1/1; fib, Sq
58;
7446 1/1; Rrib, Sq
72;
7458 1/1;
Lsca, Sq 72; 7461 1/1; Rsca; 7468 1/1; Rsca, Sq 69;
7472 1/1; Rsca-tip, Sq 84; 7489 1/1; Rulnimm, Sq 62;
7508 1/1; tbtimm, Sq 62; 7509 1/1; radimm, Sq 56-57;
7511 1/1; l/2rad; 7516 1/1; rad?imm?, Sq 63; 7518 1/1;
hallux phal; 7519 1/1; rad, Sq 78-80; 7522 1/1; fern;
7524 1/1; rad, Sq
65;
7531 1/1; rib, Sq 62; 7532.1 1/1; rib;
7533 1/1; l/2rad, Sq 67; 7534 1/1; l/2rad; 7535 1/1;
humimm, Sq
52-53;
7538 1/1; humimm, Sq 56; 7543 1/
1;
hyoid; 7558 1/1; phal; 9854 1/1; phal; 9895 1/1;
tmtimm; 9960 1/1; digit; 10114 1/1; Rqua; 10134 1/1; v,
Sq 61; 10135 1/1; vimm, XIII; 10136 1/1; v; 10139 1/1;
v,Sq62; 10141 l/l;v; 10142 1/1; vimm, VII; 10143 1/1;
v; 10144 1/1; v, Sq 73; 10145 1/1; v; 10148 1/1; v, XI;
10148 1/1; v; 10151 1/1; v; 10152 1/1; v; 10154 1/1; v;
10155 1/1; v; 10158 1/1; v; 10159 1/1; v; 10160 1/1; v;
10161 1/1; v; 10162 1/1; v; 10163 1/1; v; 10164 1/; v;
10165 l/l;v; 10244 l/l;rib, Sq
66;
10245 1/1; rib, Sq
61;
10246 1/1; rib; 10252 1/1; digphal; 10253 1/1; rib;
10254 1/1; rib; 10271 1/1; crafrag, Sq 68; 12663 1/1;
Ltbtimm, Sq 86; 12664 1/1; Luln, Sq 86; 14471 1/1; Ruin,
Sq 97; 17924 1/1; steant; 18929 eggfrags, Sq 111;
20107 1/1; Lsca, Sq 120; 20112 eggfrags, Sq
12;
20162 4
pes eggshell; 20163 3 pes eggshell, Sq 121;
20169 8eggfrag, Sq
121;
20384 1/1; Rtbtimm, spoil heap;
33511?, Sq 119; 36531 1/1; phal; 36532 Rmanp.
Holdaway & Worthy—Pyramid Valley fossil fauna
121
Genus Mystacina Gray
Mystacina robusta Dwyer, 1962
MATERIAL:
CM FMa 289 1/1;
Rhum,
Sq 103; CM FMa
290 1/1; Rhum, Sq 103; CM FMa 512 1/1; man, Sq 119;
CM FMa 513 1/1; Ruin, Sq 119; CM FMa 514 1/1; Ruin,
Sq 120; NZMa 512, 27.1.1965, Sq
119-120.
Note added in proof
Specimens underlined under Prosthemadera (p. 119) have
been re-examined. Av5899 is Callaeas
cinera;
all the oth-
ers are Philesturnus corunculatus. Totals in Table
1
have
not been adjusted. Philesturnus carunculatus appears to
have been rather more abundant than indicated in the spe-
cies account, and breeding is confirmed.
