Full text of "Records of the Auckland Institute and Museum"

See other formats

AUCKLAND

Part
Part
Part
Part
Part

Part

(pp

( pp.

( pp.
( pp.
(pp.
(pp.

RECORDS

OF THE

AND MUSEUM

VOL. 4.

Published by Order of the Council:
Gilbert Archey, Director

Edited by: A. W. B. Powell

.1-92) + - -
93-144) - - -
145-192) - - -
193-262 ) -

263-308 ) i 3
309-382 ) - -

Unity Press Ltd.

c issued 20th December, 1950.
: issued 19th December, 1951.
- issued 22nd December, 1952.
issued 12th February, 1954.
_ issued 20th December, 1954.

N issued 25th October, 1956.

, Printers, Auckland.

a
ie tes B ~~
f Vive, & ee re: i 2 be
Mei AY ‘eck
ee oo ¢ At Bex ;
Y 7
oo ME i 1 fn

INSTITUTE

ee Bo
=a athe
™“ aa - - Ry =

LP a

—_ i

forp:

Ae
pes,

7 ‘ :
ra
wai

Coe
‘a =
4

: “nh ia * =
or ie Te ee e
i 1 Mes airy i

at} 5 ' _
¥. EOP yy la
ca ot aes
a

The
le

" i i a - i
‘ns an © Lite” - ' i bel 7 "Bi as :
i ae * So ia A Se PRT in
0m 7 wr ee
Te hee ae Ai
7 e fiew ‘*\-

CONTENTS

BOTANY AND ECOLOGY.

Algae of the Three Kings Islands.
By V. J. Chapman, Auckland University College
A New Species of Microdictyon Decaisne in New Zealand.
By Vivienne Dellow, Auckland University College ..
Effect of Goats on Great Island, Three Kings; The Permanent
Quadrats Resurveyed
By M. Holdsworth, University of Otago Es Pel
Elingamita (Myrsinacea), a New Monotypic genus from West Island,
Three Kings.
By G. T. S. Baylis, University of Otago -
Incipient Forest Regeneration on Great Island, Three Kings Group.
By G. T. S. Baylis, University of Otago
New Plant Localities in the Auckland Province.
By Ruth Mason, Neville T. Moar, Botany Division, D.5.I-R.,
Wellington, and Robert Cooper, Auckland Museum
Pohutukawa X Rata.
By R. C, Cooper, Auckland Museum a; +4
The Flora of the Three Kings Islands: Additional Notes: with Note
on Suttonia.
By W. R. B. Oliver, Wellington

The ngs Kings Cabbage Tree.
By W. R. B. Oliver, Wellington

Variation in Hebe (Scrophiulariaceac) at Huia and Blockhouse tise

New Zealand.
By R. C. Cooper, Auckland Museum

ETHNOLOGY.

Human Forms in the Art of Melanesia.
By Paul S. Wingert, Columbia Untversity
Tau Thu: The Maori Canoe Prow.
By Gilbert Archey, Director, Auckland Museum

GEOLOGY.

A. Note on the Geology of the Albatross Point District, Kawhia.
By A. P. Mason, Associate Geologist, Auckland Museum ..

Notes to Accompany a Topographical Map and a Provisional Geologi-
cal Map of Great Island, Three Kings Group.
By M. H. Battey, Auckland Museum

Observations on the Structure of Far Northern New Zealand,
By M. H. Battey, Auckland Museum

The Geology of Rangiawhia Peninsula, Doubtless Bay, North Auck-
land.
By M. H. Battey, Auckland Museum

The Occurrence of Babingtonite in Spilite from Three Kings erate
By M. H. Battey, Auckland Museum i

The Torehine Beds of Coromandel Peninsula.
By R. N. Brothers, Auckland maverathy PoneEs) and A. P.
Mason, Auckland Museum

ZOOLOGY.

A Bird Census and some Recent Observations on Birds on Great
Island, Three Kings Group.
By E. G. Turbott, Auckland Museum, and P. C. Bull, see
Ecology Section, D.S.1LR.

Page

Page .

Page

1 YO

113

103

205

111

38]

365

153

6
Crs

309

263

193

A New Rail from Cave Deposits in the North Island of New Zealand.
By R. A. Falla, Dominion Museum, Wellington '

A New Species of Cermatulus Dallas from the Three Kings fefauds.
New Zealand (Hetcroptera:; Pentatomidac).
By T. E. Woodward, Auckland University College

A Note on the Occurrence of Chelisoches morio (Fabricius) on Pit-
cairn Island, South East Pacific Ocean (Dermaptera: Labiduri-
dac).

By E. T. Giles, Auckland

Aquatic Insects = Little Barrier Island.
By K. A, J. Wise, Plant Diseases Division, Auckland

Four New Species of New Zealand Land Snails and the Systematic
Position of Seats cordelia Hutton.
By A. W. B. Powell, Auckland Museum

Land Mollusca atte Bahk Islands of the Three Kings Group: with
Descriptions of Three New Species.
sy A. W. B. Powell, Auckland Museum

Life History of Austrosuccinea arche yt, an Annual Snail, and its
Value as a Post-Glacial Climatic Indicator.
By A. W. B. Powell, Auckland Museum

Melolonthinae (Coleoptera) from the Three Kings [slands.

By Bb. B. Given, Entomological Research Station, Nelson ..
Mollusca from the Continental Shelf, Eastern Otago.

By A. W. B. Powell, Auckland Museum

New Records and Descriptions of Hemiptera-Heteroptera aa the
Three Kings Islands.
By T. E. Woodward, University of Queensland

New Records of Miridae (Heteroptera) from New Zealand, with
Descriptions of a New Genus and Four New Species.
By T. E. Woodward, Auckland University College

New Zealand Molluscan Systematics, with Descriptions of New
Species. Part 1
By A. W. B. Powell, Auckland Museum hs
New Zealand Molluscan Systematics, with Descriptions of New
Species. Part 2.
By A. W. B. Powell, Auckland Museum

Notes on the Birds of the Three Kings Islands.
By E, G. Turbott, Auckland Museum i; <>
Notes on the Plumages and Breeding Cycle of the Spotted Shag
Phalacrocoray (Stictocarbo) pwictatus punctatus (Sparrman,
1786).
By E. G. Turbott, Auckland Museum

On Further Colonies of Placosty/us Land Snails from Nani! omost
New Zealand. |
| By A. W. B. Powell, Auckland Museum

Some Coleoptera from the Noises Islands, Hauraki Gulf.
By J. C. Watt, Papatoetoe by
Some Stray Tropical and Sub-Tropical Sea Birds in eae Zealand.
. . By E. G. Turbott, Auckland Museum
spiders from the Three Kings Islands.
By B. J. Marples, University of Otago
Succinea archeyi Powell.
By H. FE. Quick, Reading, England
The Melluscan Land Operculate Genus Liarea.
By A. W. B. Powell, Auckland Museum

The Genus Rhopalimorpha Dallas (Heniiptera- Fea on th with a
Description of a New Species.
By J. G. Pendererast, Auckland

Vhe Genus ENE pha Dallas (lHeimiptera-Pentatomidae).
By J. G Pendergrast, Auckland Be i.

Page

Page
Page

Page

: 241

61
267

73
215
9
169
239
141
340

134

GENERAL

Albatross Point, Geology of
elllodiscus fallax un. sp.
clllodiscus spiritus n. sp. ..
Aquatic Insects of Little

Island .. “ a
Archey, Gilbert.

Tau Ihu: The Maori Canoe Prow
Arenimitra arenosa; exasperata
Armigera turbotit n. gen. & sp.

Art of Melanesia. Human Forms in
lustrodiaphana maunganitica n. sp.
Austromitra brunneacuicta n. sp.
Austromiira gajra mn. sp... 6+ 4s
Austrosuccinea archeyi; Lite history
and value as Post-Glacial cli-
atic indicator a4 ve

Anatomy (see Quick)
3attey, M. H.

Notes to Accompany a Topographi-
cal Map and a Provisional Geo-
logical Map of Great Island,
Three Kings Group

Observations on the Structure of
Far Northern New Zealand

The Geology of Rangiawhia Penin-
sula, Doubtless Bay, North
Auckland rr ne Les

The Occurrence of Babingtonite in
Spilite from Three Kings Islands

Baylis, G, T. 8.

Elingamita (Myrsinaceae) a New
Monotypic Genus from West
Island, Three Kings te

Incipient Forest Regeneration on
Great Island, Three Kings Group

Brothers, R. N., and Mason, A, P.
The Torehine Beds of Coromandel
Peninsula ca is

Stiin PC.. Turbott, E. “G. and

Cabbage tree, Cordyline kaspar 1. sp.

Cabestana (C winatilesta) otagoensis
n. Sp. a A m

Capeiltralius werney nn. wen. & S).

Cermatulus nasalis ..

Cermatulus turbottt n. sp.

Chapman, V. J.

Aleae of The Three Kings Islands,
New Zealand ..

Charonia capax; ribicunda

Chelisoches morio 4%

Chinamiris inuohlenbeckiag il.
Ar Sp. ae .

Chiracanthium insulare N. Sp.

Chlamwvs delicatula—lusitriton Coy
SIUMTEY ee anal Lat cee ee

Chlamys radiala ' ra is
Chiaiiy's (Mimachlamys) taiaroad isp.
Chlamy's Coase aie) delicatula
Coleoptera trom Noises Islands
Coluzea mariae n. Sp.

Barrier

gen.

153
163
164

321

305
17-4
330
145
1&5
185
183

99
103
193
245
381

236

; 242

26
2.4

199
175
213

10
332

74
75
169
70
317
180

INDEX
Cominella virgata brookest n. subsp.
Conuber conica 22 SS Eee,
Cooper, R. C.

Pohutukawa X Rata. Variation in
Metrosideros (Myrtaceae) on
Rangitoto Island, New Zealand

Variation in Hebe (Scrophularia-
ceac) at Huia and Blockhouse
Bay, New Zealand ..

Cooper, R. C. (see Mason, Ruth, ete)

Cordyliic kaspar n. sp...

Coromandel — Peninsula ; Tarehine
3eds ap 4 Se

Cymus novacsclaindiac n. sp.

Cyrtorhinus cumberi i, sp. -.

Dellow, Vivienne.

A New Species of Microdictyon

Decaisne in New Zealand ni
Delonagapia n. ras for Gerontia cor-

dcelia 3 , i .
Deracocorts maoricus n. sp. -
Elingamita johnsoni n, gen. & sp...
fmosamia licind ei
Engyytatus nicotianac ‘i
Ewnaticina cincta .. «eee
Falla, R. A.
A New Rail From Cave Deposits
in the North Island of New
Zealand
Far Northern New Zealand. Observa-

tions on the Structure of .. sh
Gasparia nebulosa n, gen. & sp.
Geology of Albatross Point .. ..
Geology of Rangiawhia Peninsula ..
Gerontia cordelia, see Delouagapia n.

gen. ees aS Se 1?
Giles, E. T.

A Note on the Occurrence of Che-
lisoches morio (Fabricius) on
Pitcairn Island, South East Paci-
fic Ocean (Dermaptera: Labi-
duridae ) ny heh gd

Given, B. B.

Melolonthinae (Coleoptera) from

the Three Kings Islands
Haliotis. virginea crispata A
Hfalticus tibialis
ITinea brasthana .,
Holdsworth, M.

Effect of Goats on Great Island,
Three- Kings: The Permanent
Quadrats Resurveyed

Iredalina aurantia n. sp. -
Iredalina mirabilis 81,
Fiareg aupourta D. Sp. on. «0 a:
Liarea aupouria tara n. subsp.
Liarea eyea tessellata n, subsp...
Liarea hochstetteri alta uu. subsp.
Liarea ornata n, sp. Be

Liareca turriculata partula un. subsp. ;

181
174

205

295
381

hy
224
“16

309
334
153

167

267
. 172

174

113
239
239
286
287
280
280
282
291

Marples, B. J.

Spiders from the Three eye
Islands

Mason, A. P.

A Note on the Geology of the
Albatross Point District, Kawhia

Mason, A. P., Brothers, R. N. and

The Torehine Beds of Coromandel
_ Peninsula

Mason, Ruth; Moar, Neville qt and

Cooper, Robert
New Plant Localities in the Auck-

New Plant Localities in the Anciand

Liarea turriculata waipoua n. subsp. 291
Lithyphantes regius n. sp. . 337
Maori Canoe Prow; Tau Thu ¢ 865

- 329

153

193

land Province .. ee te
Mayena australasia blacki n. subsp. 237
Mayena australasia vossi n. subsp. .. 176
Metrosideros. Variation, Rangitoto

Island . 205
Microdictyon mutabile n. ‘sp. 3
Microvoluta obconica n. sp. .. - 183
Moar, ere T. (see Mason, Ruth,

etc
Monoplex australasiae ) 436
Murdochia hirsutissima n. sp. xs
Nassarius particeps 182

Province '
Noises Islands. Senne “Coleoptera

from the .. : A) eat 7
Odontria carinata n. sp. . 269
Oliver, W. R. B.

The Flora of the Three Kings
Islands: Additional Notes: with
Note on Suttonia iyo

The Three Kings Cabbage Tree .. Jol

Pachymelon (Palomelon) smithin. sp. 81
Panopea smithae n. sp. 73
Panopea wanganuica n. sp. 8()
Panopea zelandica .. i. @8
Paralaoma buddlei n. sp. a os

Pendergrast, J. G.

The Genus Rhopalimorpha Dallas
(Hemiptera-Heteroptera) with a

Description of a New Species .. 31
The Genus Rhopal:morpha Dallas
(Heteroptera, Pentatomidae) 159
Perisporochnus regalis n. gen. & sp. 202
Phalacrocorax punctatus punctatus .. 343
Phrixgnathus blacki n. sp. ree 5%
Placostylus, ambagiosus gardneri n.
ibs, Stk. Pee TR Oy Vy EGS
Placostylus ambagiosus hancoxi n.
subsp. St a et ae. 156
Placostylus ambagiosus michiei n.
SUDSTE tnt. lee ee es) | thse
Placostylus, ambagiosus pandora n.
Bnei +t wae Gh Ue uno
Placostylus ambagiosus paraspiritus
n. subsp. . +5 Ov
Placostylus dpsbagiosus Wares n.
subsp. . 135

Placostylus aa oa

Propesinum
Proxiuber hulmei n. sp. ..

Quick, H. E.

Tohunanwa trigonia n. sp. .
Tau Thu: The Maori Canoe Prow
Uhalassohelix prouset n. sp.
Theridion longicrure ni. sp.
Three Kings Islands

bollonsi
Form A e. .

Powell, A. W. B.

Four New Species of New Zealand
Land Snails and the systematic
position of Gerontia cordelia
Hutton

Land Mollusca from Four ‘Islands
of the Three Kings Group: with

Descriptions of Three New
Species
Life History of A ustrosuccinea

archeyi, an Annual Snail, and its
Value as a Post-Glacial Climatic
Indicator Ys .

Mollusca from the
Shelf, Eastern Otago .

New Zealand Molluscan System-
atics, with Descriptions of New
Species. Pact #1

New Zealand Molluscan ‘System-
atics, with Descriptions of New
Species. Part 2

On Further Colonies of Plagostilus
Land Snails from Northernmost
New Zealand ..

The Molluscan Land Operculate
Genus Liarea

umbilicatum

Continental

Succinea archeyi Powell

Rhopalimorpha humeralis
Rhopalimorpha lineolaris n. sp.
Rhopalimorpha obscura ..

Rhytida forsythi n. sp.

River Canoe Prow

Septifer cf. bilocularis ..

Spiders from the Three Kings Islands
Spotted Shag, Phalacrocorax punc-

tatus.
Cycle
“thenarus myersi n, Sp.

NESS and Epestine

Stray Tropical Sea Birds in New

Zealand <.
Teahunanua alata n. gen. & Sp.

Algae of
Babingtonite in Saibite fron
B-rd Census, Great Island
Birds of, Notes on
Cermatulus (Heteroptera) feb by
Effect of Goats and Permanent
Quadrats Resurveyed
Elingamita johnsoni n. gen. & sp.
Flora of ; Additional notes .
Geological and topographical maps
Hemiptera-Heteroptera; New Rec-
ords
Incipient Forest Regeneration

, 131

. 163

127

61
73

134
her

. 174
*-239

. 123

- 187

.. 170
aera
i BOS
.. 164

. ooo

ye LO?
. 263
. 245
Toor ki

99
eLark
93

biweao
. 103

Land Mollusca from Four Islands of
Melolonthinae (C EE from
Spiders from

Lonna dolinum ,

Tropical and Sub- Tropical Sea Birds

Turbott, E. G.

Notes on the Birds of the Three
Kings Islands ..

Notes on the Plumages ‘and ‘Breed-
ing Cycle of the Spotted Shag,
Phalacrocorax (Stictocarbo)
punctatus punctatus (Sparrman,
1786) Pee se PE ete y he

Some Stray Tropical and Sub-
Tropical Sea Birds in New Zea-
land

Turbott, E. G. ‘and Bull, P. tH

A Bird Census and Some Recent
Observations on Birds on Great
Island, Three Kings Group

('ifleya marwicki n. sp. -.

Utileyva williamsi n. sp.

Variation in Hebe ..

Venustas blackin. sp. .. «.. «

Venustas punctulata multigemmata
n. subsp. ..

War Canoe Prow ..
Watt, J. C.

Some Coleoptera from the Noises

Islands, Hauraki Gulf s
Wingert, Paul S.

Human Forms in the Art of Mela-

nesia =i
Wise, K. A. J.

Aquatic Insects of Little Barrier

Island .
Woodward, T. ‘E.

A New Species of Cermatulus Dal-
las from the Three Kings Islands,
New Zealand (Heteroptera: Pen-
tatomidae ) re.

New Records and Descriptions of
Hemiptera-Heteroptera from the
Three Kings Islands a

New Records of Muiridae (H Bigs
roptera) from New Zealand, with
Descriptions of a New Genus and
Four New Species wp

Xenophalium sca matat n.
sp. +
Zeadmete barkeri n. Sp.

a Ce al aS

ee aa
\
ys

an
ath

> #
Py mt Pu
1p ” Shang f Foy
EF i, EP mak:
Sal § é 3 ~ Fads
teen. a F
va #
lt BLY & 4
“SP OCS? »
Se Gy
— 4 j ?

RECORDS

AUCKLAND INSTITUTE
AND MUSEUM

VoL. 4. No. 1

Published by Order of the Council:
Gilbert Archey, Director

Edited by: A. W. B. Powell,

Assistant Director

20TH DECEMBER, i950

Unity Press Ltd., Printers, Auckland

CONTENTS

VOL: .4,. NO... 7.

A New Species of Microdictyon Decaisne in New Zealand.

By Vivienne Dellow, Auckland University College

New Records of Miridae (Heteroptera) from New Zealand, with
Descriptions of a New Genus and Four New Species.
By T. E. Woodward, Auckland University College

A. New Species of Cermatulus Dallas from the Three Kings Islands,
New Zealand (Heteroptera: Pentatomidae ).

By T. E. Woodward, Auckland University College

The Genus FRhopalimorpha Dallas (Hemiptera-Heteroptera) with a
Description of a New Species.
By J. G. Pendergrast, Auckland

The Geology of Rangiawhia Peninsula, Doubtless Bay, North
Auckland.

By M. H. Battey, Geologist, Auckland Museum

Life Histery of Austrosuccinea archeyi, an Annual Snail, and its Value
as a Post-Glacial Climatic Indicator.
By A. W. B. Powell, Assistant Director, Auckland Museum

Mollusca from the Continental Shelf, Kastern Otago.

By A. W. B. Powell, Assistant Director, Auckland Museum

New Plant Localities in the Auckland Province.

By Ruth Mason, Neville T. Moar, Botany Division, D.S.1.R..

Wellington; and Robert Cooper, Botanist, Auckland Museum

Page

Soe

A New Species of Microdictyon Decaisne
in New Zealand.

By VIVIENNE DELLOW,
Botany Department, Auckland University College.

Abstract.

_ A revision of the marine Chlorophyceae of New Zealand has involved a study
of the plant which has previously been distributed as Wicrodictyon wmbilicatian
(Velley) Zanard. It soon became evident that the New Zealand species was not
identical with M. umbilicatum, and was sufficiently different to merit the rank of
a new species. A description of this species follows, together with a discussion ot
its affinities,

Microdictyon mutabile sp. nov. Figs. 1-4.

Thalo recenter carpto colore viridissimo, siccato atrovirescente
forma pulvini vel rosulato; lato 2-7cem., alto 1-Scm. ; marginibus
frondium rosulatorum irregulariter lobatis; ramis distichis vel impariter
dispositis, saepe ad peripherum in compluribus planis ; flabellatis vel recte
angulatis; venatione non conspicua; flamentis primis 3-5 eminentibus ;
ramis plerumque acuto angulo, ad peripherum frondis latescentibus ;
thallo adfixo angustis tenuibusque cellulis rhizoideis saepe elongatis ;
septis praesentibus vel absentibus; cellulis rhizoideis 300-400 longis,
simplicibus aut irregulariter bifurcis, ortis ab ima parte filamentorum
principum aut deorsum ab articulo substrato paene recte linea adjacente ;
segmentis secundis anastomosis vel liberis; anastomose per anulos
annularios in cacumine, non nunquam secundum muros adjacentes seg-
mentorum; si quod textum, triquetrum plerumque, maxima latitudine
0,2-0.5mm.; segmentis primis 300-600» longis, 100-160, latis, a
superiore parte paulo tumidis; cellulis plerisque add inarginem frondis
decrescentibus; secundis segmentis coalescentibus 190-360% longis,
50-90» latis; minoribus diametro a termino ultimo segmenti; apicibus
segimentorum obtusis ; muris tenuibus ; muris lateralibus 1.0-2.0p crassis ;
muris terminalibus 3.0-6.54 crassis; cellulis genitalibus imsignibus
projectibus conicis ad apicem cuiusque cellulae, umlatere dispositis.

Thallus bright green when fresh, dull blackish-green when dried,
cushion-like or rosulate ; 2-7em. broad, 1-5em. high; margins of rosulate
fronds irregularly lobed ; branching distichous to irregular, often 11 more
than one plane at periphery, flabellate or rectangular; venation not con-
SpIcuous ; 3-5 primary radiating filaments; angle of branching usually
acute, becoming wider towards periphery of frond; attachment by
narrow thin-walled, often elongated rhizoidal cells ; septa present or lack-
ing; rhizoidal cells 300-400, long, simple or irregularly forked, growing
either from base of main filaments or downwards from a joint lying

4 DELLow.

more or less parallel and close to substratum; secondary segments
anastomosing or free; anastomosis by annular rings at tips or occasion-
ally along adjacent walls of segments; mesh when present usually
triangular, 0,.2-0.5mm. in longest diameter; primary segments, 300-600

long and 100-160» broad, slightly swollen at upper ends; cells in general
becoming smaller towards margin of frond; secondary anastomosing
segments 190-360n long, 50-90n broad, smaller in diameter at distal end
of segment; apices of terminal segments obtuse; walls thin; lateral walls
1.0-2.0 thick, terminal walls 3.0-6.5 thick; reproductive cells distin-
guished by conical projections near apex of each cell, unilaterally
arranged.

Habitat; Locally abundant in dense or scattered clumps on Corallina
officinalis (LL) De Toni, between M.L.W.N. and M.L.W.S., flourishing
on rocks of gentle slope in sheltered water, especially where there is a
thin layer of silt or mud deposited on the coralline turf—a subordinate
member of the Corallina-Hormosira association.

Type specimen No. 618 Herbarium U. V. Dellow (in Herbarium,
Auckland Institute and Museuin) ; No. 307, Fasc. XIII Herbarium V.
W. Lindauer as Microdictyon wmnbilicatuin (Velley) Zanard?

Distribution: Endemic. So far, AZ. mutabile has been recorded on
the east coast of the North Island as far south as Mayor Island
(37°20°S, Latitude). On the west coast the only record is from
Anawhata (almost due west of Auckland). Urupukapuka Island
(Otehe1 Bay, Bay of Islands) ; No. 10962 Herb. Lindauer, S. A. Rose:
Herb. Auck. Museum. Anawhata; S. A. Rose, Herb. V. J. Chapman.
Taranga Island; L. M. Cranwell, Herb. Auck. Museum. Opo Bay,
Mayor Island; R. P. Bell, Herb. Auck. Museum. Leigh; No. 618 Herb.
U.V.D. Whangaparaoa Peninsula; No. 469 Herb. U.V.D. Long Bay,
Hauraki Gulf; No. 658 Herb. U.V.D. Narrow Neck; No. 693 Herb.
LLU 2s |

In most localities where this species occurs, two distinct growth
forms can be found: one markedly spongiose, forming a firm, compact
cushion, the other displaying the more typical features of Microdictyon,
with flat, expanded laminae aggregated into a rosulate thallus. Fewer
anastomoses occur in the spongiose growth form. ‘There is also less
variation in the number of primary radiating filaments, three acute-

*
<a

angled branches usually being present at each joint. In addition,
cells at the base of primary filaments are slightly larger than
those at the base of reticulate fronds. The distinction between the two
forms is by no means clearly defined, since plane reticulate fronds have
been found growing out from the base of spongiose cushions. On the
other hand, when reticulate fronds have been grown in culture for
several months, the whole thallus has been found to develop into a lax

series of filaments 1n which there 1s not a single anastomosis.

The plant has been named “mutabile” on account of its extreme
plasticity both in culture and in its natural habitat.

AFFINITIES (a) Generic

Plants of the cushion-hke growth form show a certain affinity to
Rhipidiphyllon in their acute-angled, flabelliform branching, together

New Species of Microdictyon. 5

with a relatively low percentage of segments anastomosing by annual
rings, but they are separated unquestionably from this genus by their
much greater size (Ihipidiphyllon is only 2-4mm. broad—Boergesen,
1924, p. 250), the occurrence of septa in the thin-walled rhizoidal cells,
and also by the fact that branching occurs in more than one plane.

AN
= HO

NS [} 4

_ (a) IMerodictyon mutabile—portion of periphery of reticulate frond (type

specimen). x 17. ,

(b) Two secondary anastomosing segments, showing annular form of
attachment. x 130.

_—a<
pas ©
72
.
—

Pig. 2. AZ. muttabile—basal portion of type specimen, x 11,
Fig. 3. MW. anutabile—young plant of cushion form, from material collected at

Whangaparaca Heads, Hauraki Gulf (No. 469, Herb. U.V.D.). Note
absence of anastomoses and polystichous branching. x 17.

Tig. 4. AY, mutabile—branch with small conical projections through which
liberation of swarmers has taken place. x 17

A DELLOW,

A likeness to Boodlea is seen in the spongiose habit and cell size.
Apart from these two points of resemblance, Boodlea differs from
M icrodictyon in general in its lack of true distichous branching, and in
its incurving ramuli; and from this species in particular in its tenaculoid
anastomosing segments.

The occurrence of occasional adhesions of parallel segments along
the longer axis of the cells (Fig. 2a) is reminiscent of Anadyomene.

(b) Specific

The importance of the nature of the structures of attachment in
subdividing the different species of Microdictyon has been emphasized
both by Reinbold (1913) and by Setchell (1929). Setchell groups the
species primarily according to their mode of attachment, and secondarily
according to the nature of ramification and resultant meshwork.
Microdictyon mutabile fits most readily into the Calodictyon section of
the genus, as defined by Setchell (1929, p. 502). Like other species in
this section, it anastomoses by annual rings born on the tips of unmodi-
fied segments, it lacks regular venation, the branching is chiefly flabellate
and acute-angled with meshes tending towards triangular, and there is a
limited area of attachment.

Further evidence in favour of placing M. mutabile in this group is
furnished by the known geographic distribution of the Section Calo-
dictyon (Setchell, 1929, pp. 502, 580-584), which is represented by a
greater number of species in the Pacific Ocean than in any other part of
the world, Within the Calodictyon group the closest relationship of
M. mutabile seems to be in the sub-group Atrovirescentes with
M. calodictyon itself, which it resembles in cell size, in the nature
of the reticulate frond, and in the lack of distinct venation.
Characters which separate M. mutabile from M. calodictyon are the
spongiose growth form, the branching in more than one plane with
fewer anastomoses, the lack of a truly umbilicate base, and the presence
of branched rhizoidal cells.

M. mutabile differs strikingly from other members of this
group, and indeed from most of the known species of Microdictyon,
in the spongiose form with its cushion-like habit and branching in several
planes, together with the lack of a truly umbilicate base; but Setchell
notes in connection with M4. montagnei of the Boodleioides section
(op. cit. p. 577), that specimens from the South Pacific “show a tend-
ency to depart from the strict plane characteristic of the genus.” This
nay be a homoplastic response to the set of environmental conditions
peculiar to the above-mentioned geographical region, although it is just
as likely to be an ecological response by the plants to exposure during
low spring-tidal periods. The majority of hitherto described species
appear to grow relatively deep down in the sublittoral region (Setchell,
op. cit. pp. 471-473). It is of interest to note that the Microdictyon
spongtolum of Berthold, which grows near the lowest level of the inter-
tidal region at Naples and which shows polystichous branching, was held
by Bitter (1900) to be identical with the true deep-water Microdictyon
of that locality. He regards the change in habit and mode of branching
to be due to the increase both in temperature and in light intensity.

New Species of Microdictyon. /

M. mutabile differs from M. wmbilicatian in possessing the tollow-
ing features: (a) a much smaller frond; (1) predominantly flabellate
branching; (c) smaller primary segments, which are 3-4 times as long
as broad; (d) smaller secondary segments; (¢) thin, unstratified walls.

In the herbarium of the Auckland Institute and Museum there are
a number of dried specimens assigned to Microdictyon which have been
collected in different parts of the North Island of New Zealand. Micro-
scopic examination shows that the majority belong without doubt to the
M. mutabile assemblage. They include plants collected at Otehei Bay,
Bay of Islands, by Miss S. A. Rose and Mr. V, W. Lindauer, and at Opo
Bay, Mayor Island, by R. P. Bell. In some of the latter specimens, cells
of the peripheral branches are shorter and broader in relation to their
length than is customary for MW. mutabile—100-140p long and 1100p
broad in many cases. The same holds for the minute (and probably
incomplete) specimen found at Taranga Island by Miss L. M. Cranwell.
This plant is notable for the abnormally great number of small,
peripheral segments anastomosing both terminally and along lateral
walls, the whole giving an appearance rather like Anadyomene.
However, a portion of the reticulate material of the type specimen of
M. mutabile when grown in culture for several weeks produced a simular
close network with an abnormally high percentage of short anastomosing
segments. It is probable, therefore, that Miss Cranwell’s plant is an
ecological variant of the typical reticulate frond with a more open mesh.

ACKNOWLEDGMENTS.

The writer is indebted to Miss R. F. de Berg, M.Sc., who originally
started this work, for many valuable suggestions about its continuance.
She would also like to thank the following: Mr. R. C. Cooper, for
making available material in the Auckland War Memorial Museum;
Mr. V. W. Lindauer, for the loan of specimens from his herbarium
and for helpful criticism of the text; Professor V. J]. Chapman, under
whose guidance the work was carried out ; and, lastly, Mr. K. J. Dellow,
M.A., for patient assistance with the Latin translation.

REFERENCES.

BITTER, GEORG., 1900.—Zir morphologie und physiologie von JMicrodielyon
uinbilicatum. Pringsh. Jahrb., vol. 34, pp. 199-235, pl. 7.
ROERGESEN, F., 1924.——Marine algae from Easter Island, In “Natural History
of Juan Fernandez and Easter Island.” Ed. Carl Skottsberg. Vol.
Il, pp. 251-253, figs. 3-4.

REINBOLD, TH., 1913—Microdictyon, im A. Weber van Bosse: Liste des aleues
du Siboga. Siboga Exped. Mon. 59a, pp. 06-68.

SETCHELL, W. A., 1929.—The Genus Microdictyou., Univ. of Calif. Publ. in
Bot. Vol. 14, No. 20, pp. 453-588.

4
‘4
+

2 Ree,

ie bag
Gc se

New Records of Miridae (Heteroptera) from
New Zealand, with Descriptions of a New
Genus and Four New Species.

By T. E. WOODWARD, Misc., Ph.D., DUC, FREES.

Department of Zoology, Auckland Untyersity College.

Absiract.

The Miridae of New Zealand are poorly known, only thirteen species having
been recorded, of which three, or possibly four, are introduced, This ae adds
six species, four of them new and two introduced. It is hoped to deal with other
members of the family in later papers, The Miridae comprise one of the largest
families of the Heteroptera, and it seems likely that many more species remain to

be described, although, as 1s the case with most other fatnilies of this sub-order, the

total will probably prove low in comparison with that in other regions of similar

abCa.

The holotype and allotype of cach new species and specimens of
the introduced species have been deposited in the collections of the
Auckland War Memorial Museum and paratypes in the Dominion
Museum, Wellington.

In all proportionate measurements, 1 unit = 0.025 nim,

ACKNOWLEDGMENT.

The writer is greatly indebted to Dr. W, Ic. China, of the British
Museum (Natural History ), for the identification of the tntroduced
species and for his other invaluable assistance, bana? In generic
determinations. without which, work on this extensive and olten difficult
eroup would, mm New Zealand, have been impossible or much more
uncertain.

SUB-FAMILY MIRINAE,

Genus CHINAMIRIS gen. nov.

Body oval, dorsally with a covering of pale, deciduous pubescence.
Head, str -ongly declivous in front; eyes contiguous with and exserted
heyond anterior margin ot pronotum ; vertex w ith complete transverse
carina between eyes; antennae rather slender, with the first segment
about as long as head and the second segment at least twice as long as
first: rostrum reaching hind coxae, Peneotcen shortly trapeziform,
with prominent anterior collar; calli well developed ; sides sinuate ; base
shallowly emarginate, exposing mesoscutum ; dise without punctures but
distinctly transversely rugose. Ostiolar peritreme large. Cuneus and
membrane deflected, the latter mottled and with two “cells, Posterior
femora incrassated ; tibiae with dark spines.

10 W oopWARD.

Genotype: Chinamiris muchlenbeckiae sp. nov.

Near Pocciloscytus Fieber, 1858, from which it can be distinguished
by the form of the pronotum: disc not strongly convex and depressed in
front, impunctate, transversely rugose, the rugae rather widely sepa-
rated, not closely interconnected to approach a punctate condition: sides
sinuate.

As a basis for future wider comparison, certain differences are
noted between the male terminalia of Chinamiris muchlenbeckiae and of
the type species of Poeciloscytus, P. unifasciatus (Fabricius, 1794), the
only species of this genus at present available to me for study. The
former is distinguished by the prominent, backwardly projecting lobe
on the left margin of the terminal abdominal sternum. The left clasper
is similar in size and genera! form in the two species (large, with the
apical portion strongly curved forward and the extreme apex sharply
pointed and down-bent), but in C. muehlenbeckiae does not have the
espa portion twisted and flattened, nor is there a prominent subapical

lation.

Chinamiris muehlenbeckiae sp. nov., figs. 1, 2.

;

Length, 4.4mm. Width across hemelytra, 2.0mm. Rather broadly
oval. Head, pronotum, scutellum, and hemelytra except for membrane
clothed with a mixture of short, fine, recumbent, dark hairs and pale,
deciduous, scale-like hairs; ventral surface with pale pubescence.
Ostiolar peritreme large, pale.

Colour: Dark brown, ground-colour testaceous heavily infuscated
with black or brownish black mottlings. On vertex, pronotum, and
scutellum a more or less defined narrow, median, pale testaceous line;
mesial part of pronotal disc mostly testaceous, lateral regions mostly
brownish black.

Head: Vertex behind nearly flat between eyes, with a prominent,
rounded, posterior carina extending the whole width between the eyes.
Head in front slightly convex, declivous, with the dark hairs suberect.
Tylus strongly convex. Jugum pale at upper and lower margins, black
in middle. Lorum with two pale spots near anterior margin. Bucculae
pale except at base. Eyes large, brown or brownish black, narrowly
margined with pale testaceous ; touching pronotum and extending beyond
its anterior angles; from above, each eye nearly # as wide as interocular
space (13.5: 19). Rostrum reaching hind coxae.

Antennae: Nearly as long as whole body (169: 176). First seg-
iment cylindrical, moderately thickened, fuscous, clothed with very short,
fine, dark bristles, length twice width of an eye and almost equal to width
of collar; other segments finely pubescent; second segment testaceous,
infuscated for about apical third and usually also shortly near base,
extreme base pale; third and fourth segments fuscous. Proportionate
lengths of segments I-IV in male, 27: 68: 42: 32 (in female segment
Il is somewhat shorter (62).

Prothorax: Collar well defined, posteror margin convex; longest
in middle, where it is nearly 4 total length of pronotum (6: 31). Pro-
notum short; sides sinuate; calli well developed, confluent in middle;

New Species of Muiridae. 1]

disc behind calli only moderately convex, impunctate, transversely
rugose ; posterior angles rounded, dark; base very shallowly emarginate
in middle, width 14 times width across anterior rounded angles behind

collar, twice width of collar, twice total median length, and 4 as wide

—

oc a

a
i ee eee

fies, 1-2. Chinamiris muehlenbeckiae sp. nov. 1, 6; 2, left clasper of ¢.

Figs. 3-6. Deraeocoris maoricus sp. Nov. 3,2; 4, 2; 5, left clasper; 6, right clasper.
lig. 7, Engytatus nicotianae (Koningsberger), .

}? WOopDWARD,

again as head across eves (60: 40: 30: 31:46). Nyphus and propleura
pale-maregined,

Scutellum: Convex, transversely rugose, anterior region completely
or mottled with testaceous, posterior region darker except for pale
median line and pale apex. Mesoscutum exposed.

Hemelytra: Mottled testaceous and fuscous Costal margin
broadly convex, nearly straight in middle, incurvinge near base and apex.
Clavus and membrane decli ivous. Clavus pale, almost colourless to hight
amber, with apical and inner angles black and basal margin more or less
distinctly reddish; towards inner margin more or less inv: ne by darker
mottlings. Outer margins of corium and cuneus in ratio 80; 28. Mem-
brane fuscous, large cell black, all with pale mottlings ; veins pale.

Legs: Trochanters paie. Coxae and femora dark brown to black
vith pale mottlings. Tibiae with suberect dark spines; pale with four
dark bands, the narrow, subbasal band often not clearly separated from
the second in the hind tibiae; fourth band apical. Tarsi fuscous, black
at apex,

Male Terminalia: From the left-hand postero-lateral margin of
the terminal abdominal sternum there projects backwards above the base
of the left clasper, ni distinct from it, a prominent lobe, conical or
nearly cylindrical in form and with the apex somewhat narrowed and
bluntly rounded. The cor responding process on the right side is a very
much smaller, inconspicuous, subtriangular lobe, marked off ventrally
by a notch in the sternal margin. Left clasper large, wide at base:
proximal % stout, broadly curved, but not twisted or flattened; apical
third caneed and tapering, strongly curved forward, with extreme
apex finely pointed. Light clasper small.

Localitics: Woletype ¢@, allotype 2, paratypes: 2 64, 2 99, 17 other $2
and 19 other Y, collected at Foxton, Manawatu, Noi th Tand 8/1/50; 2 ¢ 4,
1 9 at Paiaka, Manawatu, 5/1/50; all by beating MWuehlenbechia australis Meissn.,
trom which alse a series of nymphs was obtained,

SUB-FAMILY DERAEOCORINAE
Genus DERAEOCORIS Kirschbaum.

Deraeocoris Iairsechbaum, 1855, Jahrb. Per. naturk. Nassau, 10. Distant, 1904,
Fauna Brit. India, Rhynch., 2, 406.
Capsus Fieber, 1858, ]Mien. ent. AWonat., 2, 307, not of Fabricius, 1803.

Macracapsus Reuter, 1875, Petites Nouv, ent., 1 (137), 547. 1879, Ofvers.
Pinska letensk.-Soac. forlt., 21, 55.

Type: Ctmer olvaccus Vabricius. 1776, = Capsus nicdius Kirschbaum, 1855,
fixed by Distant.

Deraeocoris maoricus sp. nov., figs. 3-6.

Female: Broadly oval. L ength, 3.2mm, Width across hemelytra,
1.9mm, Dorsal surface shining, strongly convex, almost entirely bare.
srowmsh black, with legs vellowish brown,

New Species of Miridae. 13

Head: limpunctate; markedly declivous in frent. Width across
eyes a little greater than leneth to apex of tvlus (37: 33). Iyes brown-
ish black. From above, each eve slightly less hit > as wide as inter-
ocular space (10: 17). Vertex ochreous between eyes; posterior trans-
verse carina, tylus, juga, and lorum black. Tylus and juga with sparse,
fine hairs: tylus strongly convex, juga short, convex. Bucculae not
prominent. Rostrum reaching to hind coxae.

Antennae: 3 as long as whole body (99: 130) and $ as long again
as posterior width of pronotum (99; 60); clothed w eh, fine, suberect
hairs. First segment somewhat swollen, black ; second segment ochreous
in middie, black at each end, the dark apical portion notably dilated ;
third and fourth segments black, with hase of third ochreous ; propor-
tionate lengths of segments I-IV, 16: 42: 27: 14.

Prenotum: Disc moderately and fairly evenly convex, ochreous,
with brownish black infuscations and coarse black punctures; calli and
anterior collar impunetate; calli prominent, confluent; collar pale ochre-
ous, narrowly margined with black; sides nearly straight; base broadly
convex, only very shehtly sinuate; width at basal angles 25 times that
across collar and rather less than twice length, ‘nehudlinie collar

(60; 24: 34).

Scutellum: Only slightly convex and moderately raised, inpunc-
tate, with very fine transverse rugulae; ochreous, w ith apex and often

more or less of median region dark, t the dark area sometimes almost
covering clisc.

Hemelytra: Strongly convex, with cuneus and membrane strongly
depressed; costal margins broadly convex; clavus with coarse, dark
punctures; corium with rather finer punctures set more widely apart;
cuneus very finely and sparsely punctate; clavus ochreous with margins
narrowly or widely dark; corium blackish brown, often with more or
less of costal and claval borders ochreous ; cuneus brownish black ; mem-
brane small, infuscate towards apical border, leneth (from apex of
clavus to apex of membrane ) ; ereatest width :: 46: 31; veins dark
brown; length of costal margins of corium and cuneus, 66: 20.

Legs: Slender; yellowish brown, femora, apex of tarsi, and band
near middle of tibiae darker; fore coxae ochreous, mid and hind coxae
brownish black. Clothed with very short hairs; spines of tibiae incon-
spicuous, very fine and short; femora with a few very long and slender
pale erect hairs on posterior margin,

Abdomen: Venter shining brownish black, impunctate, finely
pubescent.

Male: There is a marked sexual dimorplism, the ¢ differing from
the @ in the following respects:

Elongate oval, only moderately convex above. Length, 41nm.

Width, 1.7 mm.

Head: Width of head, interocular space, and eye in ratio
34.5; 16.5: 9.

14 W ooDWARD.

Antennae: 3% as long as body (105: 160) and 3 as long again as
posterior width of pronotum (105: 59), First segment ochreous in
middle, black at base and apex, with ochreous region almost obscured
in dark specimens. Second segment longer than in @ and with basal
black region shorter; slightly thickened towards apex, more gradually
so than in @. Length of segments I-IV, 16: 46: 27: 16.

Pronotum: Colour as in 2, except that disc and calli are often
largely black; convexity between basal angles and declivity behind it
usually more pronounced than in @ ; posterior width, width of collar,
length, in ratio 59; 22: 36. |

Scutellum: Entirely brownish black or black in all speciinens seen ;
rugulae more pronounced than in @.

Hemelytra: Only moderately convex, with cuneus and membrane
scarcely depressed; extending well beyond apex of abdomen; clavus
black ; corium and cuneus more or less uniformly dark brown: incision
at claval suture more obvious than in @ ; costal margins only weakly
and gradually convex; corium and cuneus much longer than in @, their
costal margins in ratio 80: 30; membrane large, length (measured as in
2) to greatest width, 50: 80, apical infuscation altogether more pro-
nounced than in @, being both darker and much broader, extending as
a wide brown band around entire margin except for a narrow pale strip
next to apex of cuneus, and leaving a large central area and the main
cell clear; veins brown, margined by an infuscated zone of membrane.

Legs: Femora lighter than in @ , often with a reddish brown tinge.

Genitalia: Left clasper with basal lobe black, conical: apical pro-
cess rather long and slender, curving first upwards and to right and
then forwards and to left, apex finely pointed. Right clasper very small.

Close to D. biréi Poppius, 1915 (from New Guinea), but differing
in the impunctate scutellum, the colour of the antennae, the longer
hemelytra of the ¢, with the large, clear central area of the membrane.

Localities: Holotype °, allotype ¢, Botanical Reserve, Nelson, 11/12/49.
Paratypes: Auckland, 22/5/49 (4), 13/3/49 (2). Others: Nelson, 10-11/12/49
(9 64, 1 2); Auckland, 22/12/38 (4 92), 25/10/44 (1 9%), 13/3/49 (3 99),
18/3/50 (1 9); Paihia, Bay of Islands, North Auckland, 10/2/49 (2 99).

SUB-FAMILY MACROLOPHINAE (DICYPHINAE).
Genus ENGYTATUS Reuter.

Engvtatus Reuter, 1876, Ofvers. Kongl. Vetensk.-Akad. Forh,, 32 (9), 82.
1910, Act. Soc. Sct. Penn,, 37 (1), 151. Cyrtopeltis Fieber, Reuter,
1909, ihid., 36 (2), 62.

Type: E. geniculatus Reuter, 1876,

Engytatus nicotianae (Koningsberger), fig. 7.

Leptoterna nicotianae Koningsberger, 1903, Mededeel’s Lands Plantent., 64, 32,
pl. 4, fig. 8 Cyrtopeltis (2?) nicotianae Kirkaldy, 1908, Proc. Linn.
Yoc. N.S. Wales, 33, 377 (as new species). ?Dicyphus tabaci Frog-
gatt, 1920, Agri. Gas. N.S. Wales, 31, 715-716 (possible synonym;
description very brief). Dicyphus nicotianae (Konings.) Fulmek,

New Spectes of Miridae. L5

1925, Deli Proefstat., Bull, 25, 2, not of Horvath, 1922. /ingytatus

lems Reuter, China, 1938 (partin), Ann. Mag. nat. Hist. (11) 1,
604-607. Engviatus nicotianae (Konings.) Usinger, 1946, 6b. P. Bishop
Afus., Bull, 189, 72-74, fie. 17.

This species, described from Java, has a wide distribution—the
Malay Archipelago, Australia, the Pacific (Guaim, New Caledonia, F1j1).
Its presence in New Zealand 1s of interest as a potential pest of tobacco,
apparently a preferred host, and perhaps of other solanaceous species.
Owing to the mainly tropical distribution of this Mirid, there is the
possibility that heavier infestations may be looked for on plants growing
under glass,

Dr. W. I. China states (im Ii?.) that this species is probably
eenerically distinct from the genotype /:. geniculatus Reuter.

The following redescription was made from New Zealand spect-
mens :—

Male: [Elongate oblong. Length, 3.7-4.0 mm. Width, 1.1 mm.
It appears that there may be local variations in size. Usinger (1946)
writes: “The length is given as 4 mm. in the original description, whereas
my series is uniformly about 3.5mm.” Clothed dorsally with short,
fine, dark hairs. ead, pronotum, and scutellum yellow, often with
more or less of a greenish tinge.

Head: Small, subglobular, usually with a more or less well defined
median dark line; frons and tylus strongly convex, the latter black at
apex ; eyes small, brownish black, not reaching pronotum. each 3 as wide
as interocular space (6: 9); rostrum reaching posterior end of middle
coxae.

Antennae: Finely pubescent ; about 4 as long as body (124: 100) ;
first segment only slightly thickened, black in middle, pale at base and
apex; second segment pale in middle, black at base and apex; third and

fourth brown, with extreme base of third paler; proportionate lengths
of seginents I-IV, 15: 44: 45: 20.

Pronotum: Anterior collar sharply constricted, nearly 1-7 total
median length of pronotum (3: 22); calli pronounced, convex, with
disc shallowly grooved behind them; anterior margin nearly straight;
sides only slightly sinuate ; posterior angles broadly rounded; base deeply
and widely emarginate, largely exposing mesoscutum, just over twice
as wide as across collar (36: 17) and about 3 as wide again as median
length (36: 22); disc shining, remotely and very finely and shallowly
punctate,

Seutellum: Disc moderately raised, smooth, with fine hairs; apex
acute, black; base to length in proportion Eve iS:

Hemelytra: [:xtending well beyond abdomen; costal margins of
corium straight, nearly parallel, almost three times length of cuneus
(75: 26); corium and clavus pale, translucent, straw-yellow; apex of
corium dark-margined and with a dark spot before cuneal fracture and a
smaller, less well defined dark spot just beyond apex of clavus; claval
suture narrowly dark; cuneus translucent, almost colourless except for
dark apex; membrane finely rugulose, colourless except lor the very

16 Woopw arp.

narrowly infuseate margins and the veins, which are dark except towards
hase; outer cell very small. inner cell oblong.

Legs: Slender, pale stramineous: finely pubescent, tibiae with fine,
dark spines and with a small dorsal dark spot at extreme base; tarsi
brown, dark at apex.

Ventral surface: Coxae and ventral thorax shining yellow or
yellowish green; abdomen green or yellow to grey, with fine, pale
pubescence.

Terminalia: ‘The left clasper and the terminal abdominal segment
of the ¢ are peculiar and highly distinctive in form. The left clasper
comprises a stout, curved ventral lobe from the inner margin of which,
near the middie projects upwards a long, flattened, blade-like chitinous
arm. The apex of the terminal abdominal segment forms a bilobed.,
upturned process, the apical arm ef which is short and blunt and lies
to the right of the ventral lobe of the left clasper, while the other arm
is considerably longer, broad basally and apically curved to the left as a
slender process behind the blade of the clasper and above the broad lobes

ol the clasper and abdomen. These strictures are figured by Fulimek
(1925) and Usinger (1946) (see above).

Female: Resembles ¢ except in the following respects: Head and
interocular space slightly wider; eye: interocular space :: 6.25: 10.
Antennae, especially second and third segments, considerably shorter
(13: 31: 33: 19), and only about 3 as long as body (96: 160). Pro-
ROL slightly wider ; width of base: wide of collar: median leneth

: 39: 18: 22. Costal margins of corium and cuneus, 80: 26.

Localities: Pathia, Bay of Islands, N, Auckland, 20-25/3/49, 5 @ @, 2 29, (Dr.
R. A. Cumber). Auckland, 3/49, 1 ¢@, 1 2 (T.E.W.). ees sae ae i11
lieht trap, 25-28/1/50, 4 9°, 2 65 (Mr. E. T. Giles}. Determmed by Dr. W.
I). China,

SUB-FAMILY CYLLECGRINAE (ORTHOTYLINAE).
Genus CYRTORHINUS Fieber.

Cyrtorhinus Ficher, 1858, Wien, ent. Monatschr., 2, 313. Tytthus Weber,
1864, ibid., 8 82. Sphyracephalus (partim) Douglas and Scott, 1865,
Brit. Hem., 1, 349. Cyrtorhtins Reuter, 1884, Act. Soe. Set. Fenn.
13. 379 (emendation). Periscopis Breddin, 1896, Deutsch. ent. Zeil,,
1896. 106 (not of Fitzinger, 1843). Breddinicssa Kirkaldy, 1903,
Wien. crt. Zeil., 22. 130 (orm. pro Pertse Opus Breddin).

Type: Capsus clegantulus Mever-Dur, 1843 = Capsus caricts Vallen, 1807.

Cyrtorhinus cumberi sp. novy., figs. 8, 2.

Both brachypterous and macropterous forms occur. Four ¢ 24
were collected and all were macropterous. Of cleven 2 9, ten were
hrachypterous and only one was macropterous, and even in this specimen
the hemelytra, and particul itly the membrane, were considerably shorter
than in the ¢@ ¢@. Itis possible that wider collection will reveal a small
proportion of brachypterous ¢@ é.

New Species of Miridae. 17

Named after Dr, R. A. Cumber, of the Entomological Research
Station, D.S.I.R., Nelson, to whose hospitality and assistance is due the
collection of this and other interesting species of Hemiptera,

Macropterous Male: Ilongate, oblong. Length, 3.7 mm. Width,
lium. Clothed with very short, fine, pale, recumbent hairs.

Colour: Face, tvlus, juga, lorae, genae, antennal bases, and anterior
* of pronotum, including collar, black; eyes reddish or blackish brown ;
vertex. yellow-brown, sometimes tinged with greenish or reddish, pos-
teriorly with a more or less well defined brick-red band, sometimes
extending on to eyes, extreme posterior border black. First segment ot
antennae blackish brown, extreme apex pale; other segments black.
Scutellum and posterior 2 of pronotum yellow, the former sometimes
tore or less tinged with green, and the latter with orange. Legs yellow-
ish brown, femora sometimes tinged with green, apex of tarsi dark.
Hemelytra green; membrane transparent, iridescent, lightly infuscated
with brown or grey, especially near margins; veins yellow or light
brown, narrowly margined with darker brown. Abdomen yellow or
ereen.

Head: Shining, smooth except for microsculpture of minute
punctures and reticulations; face declivous and subvertical in front of
eyes; the black anterior region narrowing behind and ending in an acute
apex between eyes, the brown posterior region extending forward around
inner margin of each eye. Eyes nearly touching and extending beyond
anterior margin of pronotum; from above, each eye $ as wide as inter-
ocular space (8.5: 14). Tylus strongly convex; juga short, nearly flat ;
hucculae black, margins pale, fringed with a row of long, fine hairs.
Rostrum yellowish brown, black-tipped, reaching to middle coxae.,

Antennae: Slightly longer than body (160: 150); clothed with
fine pubescence. Hirst segment rather stout, slightly curved, with two
long hairs on apical half of inner margin, slightly longer than pronotum
(21: 20); second more slender than first, cylindrical; third and fourth
more slender than second. Proportionate lengths of segments I-LV,
212-6) ¥.532 25.

Pronotum: Shining; considerably widened posteriorly; with an
extremely short anterior collar; sides sinuate just behind muddle; base
widely cmarginate, largely exposing mesoscutunr; anterior angles rather
broadly rounded; posterior shoulders more angularly rounded, scarcely
raised; anterior black region in form of large, convex callus, with surface
minutely punctate-rugulose and with five shallow depressions, two on
each side of mid-line and one median and posterior ; posterior pale region
shallowly, remotely punctate and finely rugose; across posterior angles
slightly wider than head, twice as wide as collar, and ¢ as wide again as

long (35: 31: 17: 20).

Seutellum: Nearly flat; disc transversely rugulose, most strongly
so near base behind mesoscutum; the latter glabrous except for a single
row of sparse, short, fine hairs a short distance before posterior margin ;
scutellar triangle, including mesoscutum, about $ as long as wide at
base (21: 25).

Is WOODWARD.

Hemelytra: Well surpassing abdomen: costal margin nearly
straight, slightly and gradually convex in anterior third, scarcely incised
at cuneal fracture ; costa and subcosta nearly parallel throughout ; corm
and clavus convex, shining, shallowly punctate-ruguiose ; claval suture
deeply depressed ; length of costal inargins of corium and cuncus, 65: 25;
membrane well developed, extending far beyond apex of cuneus, les eth
(from apical margin to apex of clavus) about twice greatest width

(63: 30).

Hind Wings: Well developed, passing abdomen and nearly as long
as hemelytra.

Legs: Slender, with fine brown pubescence; tibiae with eee
fine, brown spines. Usinger (1939, Proc. Hazwai?. ent. Soc., 10 (2)
2/2) points out that the genus Cyrforhinus is anomalous among Mirids
in the variable nature of the pretarsal processes ; while some species have
fully developed arolia, in others these str uctures are replaced by a pair
of fine, parallel setae, C. cinberi resembles such species as fitlvuns
Knight and lividipennis Reuter in having the large, membranous, con-
vergent arolia characteristic of most members of the sub-family.

Terminalia: Left clasper with ochreous ventral lobe stout, curved,
convex below, concave above, the rounded and somewhat narrowed apex
upturned; the brown, sclerotised dorsal process coming off from dorsal
surface of ventral lobe towards base, curved upwards and to right,
slender, sinuous, very long, with fine apex. Venter of terminal abdomi-
nal segment produced as a broadly rounded lobe to right of left clasper.
The two closely apposed genital valves projecting backwards and upwards
from end of abdomen above base of left clasper. Right clasper: outer
process a large, ochreous, knob-like, backwardly projecting pyriform
lobe, beset with very long hairs; inner process a shorter, broad, laterally
flattened, backwardly directed plate, ochreous with apical margin brown-
ish black.

‘fhvansl tt paditn Female: Elongate oval. Length, 3.4-4.1 mm.
Width, 1.3mm. (The length is affected by the state of distension of
the aed abdomen. )

Colour: As for 6. except that first antennal segment is reddish
brown with pale apex and second segment often reddish brown in
middle, with base shortly and apex longly blacks.

Head: Jnterocular space rather wider than ind (eve: interocular
space :: 8.5: 16).

Antennae: { as long BS sin d (140: 160) and usually shorter than
body; seginents T-[V, 20: 35: 43: 22.

Hemelytra: Short, leaving the last two or three complete tergites
exposed; costal margin more convex than in é@ ; corium slyoate less
convex than in ¢, nee us nearly flat; cuneus very short, only 4 as long
as corium (13: 65); membrane very small, not or occasionally barely
extending beyond apex of cuneus, le neth (from apex of clavus to Apex

of cuneus) about three times greatest width (30.. 9 er 32 433).

Hind Wings: In form of short, rounded triangles, reaching usually
less than half way along abdomen.

New Species of Miridae. 19

Fies. 8-9. Cyrtorhinius, cumberi sp. noy, 8, macropterous ¢; 9, 6 terminalia, left
posterior aspect: L, left clasper ; R, right clasper; T, terminal lobe of
abdomen.

Figs. 10-11, Halticus tibialis Reuter. 10, %; Ll, ¢&.

Ties, 12-15. Sthenarus myerst sp. nov, 12, ¢ | 13, left clasper; 14, right clasper,
ventro-lateral aspect; 15, theca.

20) WoopDWaArb.

_ Macropterous Female: Length, 3.6mm. Width. 1.3 mm. (In
this form also, length will no doubt vary with state of distension of
abdomen.) Head and colour of antennae as in brachypterous 2.

_ Antennae: Much shorter than in ¢ and about as long as body.
Segments I-IV, 20: 58: 43: 25.

_ Hemelytra: Reaching but not entirely covering last complete ter-
gite. Asin brachypterous 2 except that cuneus is intermediate in length
between that of macropterous ¢ and of brachypterous @ (corium:
cuneus:: 70: 19), and membrane is larger, extending well beyond apex
of cuneus (length (apical margin to apex of clavus) ; ereatest width
924 23).

Hind Wings: Longer than in brachypterous @, but considerably
shorter than hemelytra and ending at third complete tergite from end of
abdomen. |
A Locality: Holotype $, allotype ?, 2 paratype 3 3, 2 paratype (%, 8 other
++, 1 other @, and 9 nymphs, below and in tufts of rushes and erasses, Paiaka,
Manawatu, 4/1/50.

Several species of Cyrtorhinus have been shown to be exclusively
predacious on the eggs of leaf-hoppers, particularly Delphacids, inserted
by the @ @ into the tissues of leaves and stems (vide Usinger, 1939,
op. cit., 271-273). C. mundulus (Breddin) has been used successfully
to control the sugar-cane leaf-hopper Perkinsiella saccharicida Kirkaldy.
Careful observations on the feeding habits of the New Zealand species
would be of considerable interest. Delphacids occurred abundantly with
it at the bases of the tufts of grass and rushes.

Genus HALTICUS Hahn.

FHalticus Hahn, 1832, Wansenart. Ins., 1 (3), 113, pl. 18. Astemma Latreille,
1829, im Cuvier, Réegue Anim., ed, 2, 5, 199, not of Lepeletier et
Serville, 1825. Hurycephala Laporte, 1832, Mag. de Zool., 2, Suppl.,
36. Halticocoris Douglas and Scott, 1865, Brit. Hemipt., 1, 478.

China (1943, The Generic Names of British Insects, pt. 8, p. 268) points out
that if the date of publication of Eurycephala Laporte can be proved to haye been
before November, 1832, this genus will replace Halticus Hahn.

Type: Acanthia pallicornis, Fabricius, 1794 = Cicada aptera Linnaeus, 1764.

Halticus tibialis Reuter, figs. 10, 11.
FHalticus tibialis Reuter, 1891, Revue d’Entomologic, 10, 135-136.

This small bug was described from Java and has since been widely
recorded from tropical Africa and Asia (including Ceylon, the Carolines,
Amboina and Macassar). Usinger (1946, of. cit. p. 86) quotes Esaki
as recording it as injurious to beans in the Carolines.

The following redescription, based on New Zealand specimens, is
given as an aid to identification.

Macropterous Male: Oval. Length, 2mm. Width across heme-
lytra, 1 mm.

Colour: Shining black; eyes brown; vertex with a narrow pale
yellow line along inner margin of eye; antennae pale yellowish brown,
with apex of second segment and third and fourth segments except at

New Species of Muiridae. 21

hase fuscous. Legs pale yellow-brown, with claws, apex of tarsi, and
all except apical quarter of the swollen hind femora black ; basal half of
hind tibiae more or less infuscated. Membrane of hemelytra fuscous.
A certain amount of colour variation can apparently be expected, Reuter
(1891) describes the rostrum except for the last segment and the apex
of the femora widely as pale yellow, and Usinger (1946) describes all
the femora as black except at the apex.

Head: Highly polished, nearly glabrous, minutely rugulose-
punctate; downwardly flexed, so that little of its length is visible from
dorsal aspect; almost as long to apex of tylus as wide across eyes
(25: 26); carinate posterior margin slightly overlapping front of pro-
notum; width across eyes: posterior width of pronotum :: 26: 30. Eyes
prominent, with greatest length set vertically; contiguous to pronotum
and extending beyond its anterior angles; from above, each eye just
over 4 as wide as interocular space (5.5: 15). Tylus convex. Rostrum
stout, reddish brown, reaching or just surpassing posterior margin of
hind coxae.

Antennae: Slender, three times as long as posterior width of
pronotum; first segment somewhat swollen, not reaching apex of head,
with two long, erect hairs in apical third; other segments slender, cylin-
drical. clothed with short, stiff hairs; second segment slightly longer
than costal margin of corium (35: 33); proportionate lengths of seg-
ments [-IIT, 9: 35: 25.

Pronotum: Short, trapeziform, weakly convex, shortly and gently
declivous towards base; disc minutely and shallowly punctured, obscurely
transversely rugulose ; sides straight ; posterior angles broadly rounded ;
hase only slightly emarginate; about % as wide across anterior angles as
across posterior (22: 30); posterior width twice median length.

Pronotum and hemelytra except membrane with rather sparse,
eolden, deciduous pubescence.

Seutellum: Small, scarcely raised, obscurely transversely rugulose,
minutely and sparsely punctate ; rather less than twice as wide at base as
long (12: 7).

Hemelytra: Corium and clavus convex ; claval suture deep ; costal
margin moderately convex, deeply incised at base and apex of cuneus;
membrane well developed, extending well beyond abdomen, with cell
complete, veins wide, narrowly dark-margined; cuneus and membrane

strongly deflexed; costal margin of cuneus slightly more than § that ot
corium (12: 33).

Legs: Clothed with fine, short hairs; hind tibiae with dark, erect
spines.

The short- and long-winged forms in this genus differ strikingly.
In general appearance the brachypterous forms superficially resemble the
small ‘“‘flea-beetles” (Halticidae).

Brachypterous Female: Shortly oval; dorsal surface broadly
convex. Length, 1.75 mm. Width, 1.25 mm. Width across eyes: pos-
terior width of pronotum :: 29: 32. From above, eye well over 4 as
wide as interocular space (6.5: 16), Antennae less than three times as
long as posterior width of pronotum; second segment shorter thanin ¢,
barely } length of costal margin of corium (29: 40). Width across
anterior angles of pronotum just over % that across posterior angles

23 WOoDWARD.

(23:32). Pronotum not declivous along base. Heimelytra broadly and
evenly conyex; clavus very small, claval suture inconspicuous, not
depressed ; costal margin of corium strongly and widely convex; cuneus
and membrane not deflexed - margin of cuneus marked off anteriorly
by angular incision but posteriorly by a gradual convexity ; costal margin
of cuneus only 4+ that of corium (10: 40) ; membrane very small,
luscous, without cells, projecting only shortly beyond cuneus and not
passing end of abdomen,

Localities: 1 macropterous ¢, 1 brachypterous 2, Russell, Bay of Islands,

®, Auckland, ee 1 macropterous ¢, Paihia, Bay of Islands, 13/2/49. Deter-
mined by Dr. W. E. China.

SUB-FAMILY PLAGIOGNATHINAE (PHYLINAE).

Genus Sthenarus Fieber.

Sthenarus Fieber, 1858, Wien. ent. \Tonatschr., 2, 321. Kirkaldy, 1906, Trans.
<lmer. eni. Soc., 32, 123. Phoenicocoris Reuter, 1875, Bih. svenska
letens-Akad. Forh., 3 (1), 55,

Type: Capsus rottermundi Scholtz, 1846, fixed hy Kirkaldy.
Sthenarus myersi sp. nov., figs. 12-15.

Male: Oblong oval. Length, 3mm. Width, 1.3mm. Dorsal
surface and thorax at sides clothed with fine, pale, deciduous hairs,
easily rubbed off. Ventral surface finely pubescent.

Colour: Black. [Eyes black or reddish black. Rostrum except at
apex, clavus at extreme apex, trochanters, femora at extreme apex,
ventral margin of fore and mid femora, tibiae, second segment of tarsus,

and claws ochreous; tibiae with extreme apex black, and banded with
conspicuous black spots.

Head: Strongly declivous; face subvertical. Vertex nearly flat,
shining, with miucrosculpture of minute, close punctures. Juga short,
flat; tylus scarcely raised above them. A complete, rounded posterior
carina between. eyes, with a single very fine hair on each side a short
distance from eye. Eyes touching and extending beyond sides of
anterior margin of pronotum; from above, each just over half as wide
as interocular space (8: 15). In front view, head # as long (to apex
of tvlus) as wide across eyes (21:31). Rostrum reaching to hind coxae.

Antennae: Black, very short, first segment appearing extremely
short from above, owing to flexion of HAC, Clothed with dark pubes-
cence; two bristles set close together at 4 from apex. Relative length of
segments I and II, 9: 36.

Pronotum: Short trapeziform; sides straight; anterior margin
neatly straight, very slightly convex, with an extremely short,
unsculptured anterior rim, set shehtly below surface of disc; posterior
inargin nearly straight; anterior “and posterior angles rounded. Dise
only moderately convex, transversely rugulose and with microsculpture
of minute punctures. length about equal to width across anterior
angles and 4 width at base (24: 49).

Scutellum: Nearly flat. Sculpture as for pronotum; the exposed
mesoescutum with micropunctures only. Basal width to length: scutellum
only, 23:19; with mesoscutum, 30: 23.

New Species of Microd iclvon. 23

Hemelytra: Punctate-rugulose. Well surpassing abdomen ; cuneus
and membrane deflexed. Costal margin of corium slightly convex, three
times as long as that of cuneus (60: 20) and equal in length to posterior
tibiae. Membrane black, a small pale ochreous spot at anterior end ot

vein and another at its posterior margin.

Legs: Hind femora very broad, somewhat flattened and curved ;
tibiae with fine ochreous setae arising from the black spots.

Genitalia: The ¢ claspers in the Plagiognathinae are very sinall, but
a preliminary study of five genera indicates that they may be used, when
necessary, as good taxonomic characters, showing both generic and
specific distinctions. In S. myersi the lett clasper 1s short, black, and
subconical, with the outer surface convex and the inner concave; tts
lateral spines unequal, one short, thorn-like, the other much longer, only
slightly curved, with the apical half much narrowed. Right clasper

glume-like; basal % wide, concavo-convex, distal ! narrowed and pro-
longed as an awn-like process. Theca with a slender acessory spine,
bent basally and thereafter nearly straight and parallel with main spine,
reaching about 4 way to its apex. (In this subfamily the theea (aedea-
eval sheath) is a prominent, dark, spine-like process with the outer margin
at its base attached to the abdomen just in front of, but separate from,
the right clasper, and is directed backwards and to the left. It is hollowed
or grooved along its inner side to ensheath the aedeagus, and fits against
the concave surtace of the left clasper, the two spines of which embrace
it. Both claspers with conspicuous basal peg inserted into abdomen. )

Female: Similar to 2, except in the following particulars. Length,
3.linm. Rather broader across hemelytra (1.9 mm).

Colour: Rather lighter above; dark brownish black. Base and
posterior angles of pronotum narrowly margined with ochreous. Anten-
nae with basal 2 of second segment and extreme apex of fourth ochreous,
Clavus tinged with reddish brown. Membrane tuscous, usually with

veins paler and with pale spot behind cuneus.

Antennae: Second segment slightly shorter than in ¢, Segments
Ve Pade Thies.
Localities: 2 & & 2 29, Foxton, Manawatu, 8/1/50, beating Muehlenbeckia

austratis. © @. 1S Te Paki, North Auckland, 21/1/50, Leptospermum, 06
$9, Neakengo Bay, Nerth Auckland, 27/1/50, Leptospermain,

The specimens from the far North are consistently smaller than
those from the Manawatu, described above, and the ? @ are lighter in
colour, with a distinetly reddish brown tinge, especially on head, pro-
notum, scutellum, cuneus, and anterior parts of corium and clayus (the
pronotum and clayus of one only are black). Length: @,2.5imm.; ¢,
2 6mim.-2.7 mm. Width: ¢, 13mm; 2, 14mm. Other dimensions
i proportion, @.g., in 3, width of eye: interocular space :: 7.5:14; head,
length: width :: 19: 29; length of corium: cuneus ;: 54: 18; pronotum,
length: width at base, :: 21:43 (in @, 21.5: 44). All structural features,
including the ¢ genitalia, are similar, and it is not proposed, on the basis
of the present material, to suggest subspecific categories.

This species is named after the late Mr. J. G. Myers, who added a
ereat deal to our knowledge of the New Zealand Henuptera.

‘A New Species of Cermatulus Dallas from the
Three Kings Islands, New Zealand
(Heteroptera: Pentatomidae).

By T. E. WOODWARD, Department of Zoology,

Auckland University College.

The specimens on which this species is based were collected by Mr.
E. G. Turbott, of the Auckland Museum, from Leptospermuin ericoides
shrubland on Great Island, Three Kings (Turbott, 1948, p. 261), and
put aside as differing in appearance from Cierimatulis nasalis (West-
wood).

[ wish to thank Mr. Turbott for drawing my attention to this inter-
esting material, and Dr. Archey, Director of the Auckland Museum,
for the opportunity of examining and describing it.

SUB-FAMILY ASOPINAE.

Genus CERMATULUS Dallas.

1851.—Dallas, List Specimens Hemipt. Ins, Coll. Brit. Aus. 1, p. 100.
Type: Aelia nasalis Westwood, 1837,

Cermatulus turbotti sp. nov. Fig. 1.

Length of the 3 9 @ seen: 11.5mm., 13.5 mm., about 15 mim.
(wings spread in this specimen). Total length about twice basal width
of pronotum, which is rather less than greatest width across abdomen
(5.75:6:6.5:7:7:8). Moderately convex above and strongly punctate,
punctures finer than in C. nasalis, particularly on pronotum ; connexivunl
closely, very finely punctate; finely punctate below, venter of abdomen
strongly convex, with punctures sparser and shallower at extreme mar-
gins, apex, and particularly in middle. Surface of body rather shining.

Colour: Dark above, with ground-colour of ochreous, punctured
and infuscated with black; with more or less distinct bronzy or ereentsh
reflections, either restricted to head or extending also over pronotum and
scutellum as a conspicuous metallic sheen. Dorsal surface of abdomen
black with bronze or coppery reflections. Ventral surface of abdomen
light ochreous ; punctures blackish brown; a black patch on either side
of mid-line at anterior margi of each sternite, the patches on last com-
plete sternite longer and more irregular than others; anterior and pos-
terior ventral marginal angles of each sternite black.

A New Species of Cermatulus. 25

Head: Nearly flat above, more finely punctate than pronotum.
Juga with outer margins straight and parallel in the middle, posteriorly
arcuately diverging to meet eyes; anterior angles widely rounded, though
less broad than in nasalis, with more of the curve lateral and less directly
anterior than in this species. Tylus with sides nearly straight and
parallel, converging only slightly in front; apical margin free, distinctly
convex, in all three specimens projecting slightly beyond juga. Whole
snout much less blunt at end than in nasalis. Tylus much more sparsely
punctate than in nasalis, punctures entirely or almost entirely restricted
to margins; distinctly transversely rugulose in front and behind. Juga
punctate and distinctly rugose.

Tylus black-margined, yellowish ochreous in mid-line. Juga black
except for a narrow, sublateral, yellowish ochreous line. A similarly
coloured, impunctate, triangular patch behind inner posterior margin
of each eye, and another similar but ovoid patch on each side otf base of
head behind ocellus. On each side, a shining, tmpunctate, black or
bronzy-black patch in front of ocellus, and another behind ocellus
internal to the basal ochreous spot. |

Greatest width across juga slightly over 4; width across eyes
(18: 35; 19: 37; 20: 39). Each eye somewhat less wide than in nasalis
in proportion to interocular space (only $:-7: 21; 7.25: 22.5; 8: 23).

Ventral surface pale stramineous; rostrum ochreous, reaching pos-
terior coxae, apical segment black, sides of first segment concave shortly
before apex, sparsely and shallowly punctate and obscurely striate.

Antennae: About half as long as body. Antenniferous tubercle
small, black above, with a short, pale, ventral spine. First segment very
short, black, with ventral ochreous streak ; second segment wholly brown-
ish black, or with apex black; the others with base reddish brown and
about apical 2 in fifth segment and apical 3 in third and fourth segments
black: the dark and paler regions clearly contrasted. Seg. III 3 to + as
long as II (22: 30; 20: 32; 18: 26); IV in these specimens about 4
longer than JII (26: 22; 24: 20; 22:18); V subequal to or rather shorter
than IV.

Pronotum: More finely punctate than in masalis, Anterior margin
deeply and widely excavated. Anterior angles with a very short, blunt.
outwardly and forwardly directed spine. Sides incurved and sinuate at
middle, posterior half projecting outwards at a pronounced angle and
not in line with anterior half, and posterior angles thus more prominent
and acute in appearance than in nasalis; anterior half only obscurely
crenulate, posterior half smooth. Posterior margin in front of scute]lum
straight; on each side, outside base of scutellum, with a triangular pro-
cess overlying base of clavus; postero-lateral margin outside this sinuate.
Posterior width 2.15 to 2.25 times length. Lateral margins pale ochreous
except on dark posterior shoulders. A more or less pronounced median
pale line, most definite anteriorly, and extending on to anterior part of
scutellum. Calli black or bronzy black.

Scutellum: Raised and broadly convex in front, remainder nearly
flat. Disc distinctly rugulose. Sides concave behind middle, straight

before and after concavity, the margins in front of it posteriorly con-

26 W oopwW ARD.

vergent and (with wings closed) incurved at basal angle ; margins behind
it only slightly convergent; apex broadly rounded, off-white or pale
creamy yellow, with only sparse, fine, shallow punctures. Median line
with punctures comparatively few and fine; except at base and apex
shining black, the black area extending in front to form a Y, with the
anterior median pale line enclosed between the arms. Inner margin of
fovea at basal angles straight, yellowish ochreous. |
Mesosternum: Disc black, with coppery reflections, impunctate,
transversely rugulose ; median carina a low, ochreous ridge.

Wings: Extending rather t urther beyond abdomen than in nasulis.
Corium with an impunctate, shining black patch in centre of posterior
two-thirds. Membrane of hemelytra brown, finely rugulose; veins dark.
Hind wings nearly colourless, faintly cinereous ; veins dark brown.

Legs: Ochreous; femora and tibiae spotted with reddish- or black-
ish-brown ; apex otf tibiae and of tarsal segments fuscous, tars! some-
times almost entirely black. Femora unarmed; with fine, rather long
and sparse hairs, A single black, apically directed spine on ventral sur-
face of fore tibiae only; tibiae otherwise unarmed, clothed with fine
pubescence; upper surface with shallow, longitudinal groove. Tarsi
clothed with fine, pale hairs, longest and erect towards apex of last
seginent.

Abdomen: Dorsal surface very finely punctate. Sides convex,
connexivum extending moderately beyond costal margins of closed
Wings, posterior angle in each segment projecting slightly beyond ante-
rior margin of the next, but not spined or backwardly produced. Con-
nexivum strongly marked with black at anterior and posterior margins
of each segment; orange or orange-brown between them. Anterior
abdominal spine very short, not reaching anterior margins of hind coxae,

Localitics: Collected by Mr. E. G. Turbott on Great L., Three Kings Is. 19,
Tasman Valley, 6/3/46: 2 22 near depot, 5 and 10/5/46; all on Leptospermuun
evicoides A, Rich. (kanuka).

Types: Holotype and 2 paratypes m Auckland Museum.

Close to Cermatulus nasalis (Westwood), but readily distinguished
by the differently shaped pronotum and juga, the markedly convex apex
and nearly impunctate dise of tylus, the finer punctation, particularly on
pronotum, the colour of antennae and apex of seutellum, and the metallic
bronzy reflections on some or all of the regions listed in the description.

The only other recorded species in the genus, C. pulcher Tryon,
1892. is from Fly River, British New Guinea.

For fuller comparison, a redescription of C. nasalis is added, giving
those features in which this species differs from C, turbotti, since most
of them are not covered in detail in the earlier descriptions. To avoid
recapitulation, features common to both species are omitted from this
account.

Cermatulus nasalis (Westwood). Fig. 2.
1837.—Aelia nasalis Westwood, Cat. Hope 1, p. 32.
1842—Asopus nunonularis Erichson, Arch. fur Naturg. 8, p. 276.
1844.—-4sopus nummularis Herrich-Schaettfer, Wanecnart. Ins. 7, p. 114, fig.
776 (as new species).
1831.—Cermatulus nasalis (Westwood) Dallas, List Hem. Ins. Coll. Brit. Mus.
p. 106, pl. 2, fig. 3.

A New Species of Cermatulus. 2/

1867.—Asopus binotatus Walker, Cat. Specimens, Het. Hem. Coll. Brit. Mus,
1, p. 144 (recorded from Brazil: in error, according to Kirkaldy, 1909).

1867.—Rhaphigaster pentatomoides Walker, ibid. 2, p. 370.

i)
MY ii,

if,
yy

lI, :

Fig. 1. Cermatulus turbottt sp. nov.; head and pronotum.

Fig. 2. Cermatiulus nasalis (Westwood); head and pronotum.

2& WoOoDWARD.

Length: @ ¢, 10.5-12.5 mm, (23 specimens); ¢ 6, 9-10 mm. (5
specimens). Width across abdomen: 9? 9, 5.75-6.75mm. @ @,
S-6mm, Width across pronotal shoulders: @ 2, 5-6mm.:; ¢@ é,
4.5-5.5 mm. Proportions as in turbotti. More coarsely punctate above
than in turbotti, especially on pronotum. Venter of thorax rather
coarsely and deeply punctate, venter of abdomen more finely and
shallowly punctate, particularly in middle, at sides and apex.

__ Colour: Ground-colour of testaceous or ochreous, punctured and
infuscated with brownish black. Considerable variation in the general
colouration of this species is given by differences in the shade of the
ground-colour (yellowish-, orange-, or reddish-brown) and the relative
darkness of the punctures and infuscations and the extent of the latter.
Without any bronzy metallic sheen. Dorsal surface of abdomen black:
ventral surface mottled ochreous and testaceous, rather variable, but
darker than in turbotti, with similar black markings, but last pair of
black patches usually extending to or near posterior end of sternite and
broadly confluent behind, sometimes entirely fused.

Head: Anterior angles of juga more broadly rounded than in
turbotti, with more of the curved margin facing anteriorly. Tylus with
free apex straight or very nearly so, at the most scarcely and very
broadly and bluntly convex; ending level with juga or very slightly
shorter or longer. Whole snout with end more bluntly rounded than in
turbotti. Each eye somewhat wider than in turbotti in proportion to
interocular space (2). Tylus punctate in middle as well as at margins.
Tylus and juga with or without rugulae. Vertex with an impunctate
black patch narrowly surrounding each ocellus and extending forwards
near inner posterior angle of eye. Sides of tylus with narrow black
margins, continued back to base of head as broader black bands more
or less confluent with the black patches described above. Impunctate,
ochreous spots as in furbotti. Rostrum ochreous or reddish brown,
apical segment brownish black or black, sides of first segment rarely
distinctly concave before apex.

Antennae: First segment ochreous, sometimes more or less infus-
cated ; other segments ochreous or reddish brown, second wholly so, third
and fourth with about apical 4 and fifth with about apical % fuscous or
dark reddish brown, the dark apices less heavily pigmented and less
strongly contrasted with the paler bases than in turbotti, the two regions
often scarcely differentiated. Seg. IV 4 to 3 as long again as III.
Other proportions with range as in turbottt.

Pronotum: Very coarsely punctate. Sides straight or nearly
straight, sometimes slightly sinuate at middle; anterior half more or less
distinctly crenulate, posterior half smooth. Posterior shoulders less
prominent and more bluntly rounded than in turbotti. Posterior width

2.3 to 2.45 times length. Lateral margins paler ochreous throughout.
Calli black.

(Fig. 2 illustrates about the maximum extent of the lateral sinua-
tion, to show that even where such occurs the condition is markedly
distinct from that in turbotti. Many specimens of nasalis have the
margins quite straight. ) )

A New Species of Cermatutlis. 29

Scutellum: Disc not distinctly rugulose. Apex yellowish- or
orange-brown. Mid-line, except at base and apex, black or brownish
black. Inner margin of fovea at basal angles ochreous or orange-brown.

Mesosternum: Black; median low ridge usually pale.

Wings: Shortly exceeding abdomen, [lind wings cinereous, with
green reflections; veins dark brown.

C. nasalis is a rather variable species (size, colour, proportionate
length of antennal segments, presence or absence ot a slight sinuation
in sides of pronotum). In respect of the first three features, C. turbottl
also shows variation; the length, indeed, in the three @ 2 examined has
a greater range than in 23 9 2 of nasalis collected at different times
from several widely separated localities. There seems little doubt that
the distinctive specific characters of furbofti have evolved as the result
of the continued isolation of a restricted population, initially sharing the
general variability in certain features and possibly at the same time
displaying small local peculiarities, later accentuated and added to.

~The apparently local origin and restricted insular distribution of
this species is of interest in view of the very wide distribution of
C. nasalis, which occurs in Australia and Tasmania and, within New
Zealand, has been recorded from a wide range of localities in both North

and South Islands. (See, e.g., Myers, 1926, p. 494.)

Factors involved in the isolation of the Three Kings fauna include
the continued separation of the islands from the mainland, now about
35 miles away, since about the early Tertiary, the effects of wind and
strong currents, and the sheer, rocky shores. (See, e.g., Buddle, 1948;
Oliver, 1948; Turbott and Buddle, 1948. )

The isolation of such a small population could easily have provided
suitable conditions (e.g., by periodic or even a single extreme reduction
‘nthe numbers of the effective breeding population in the area) to permit
action of the Sewall Wright effect, which increases rapidly with the
smallness of the population, in the differentiation by “drift” of a new
species in the Three Kings, while on the mainland the population has
remained conspecific with the Australian form.

While there is as yet little or no direct information on the present
numbers or biology of the species. it is possible to list tentatively a
number of factors which might have induced such a process as outlined
above: (1) The initial segregation and continued isolation of a popula-
tion extremely small in comparison with that of the mainland. (2) The
observed low population density of species of the predacious sub-fanuly
Asopinae, compared with that of many phytophagous insects, including
other Pentatomidae. (3) The long non-breeding period of C. nasalis
(as of most other Pentatomids) compared with ithe breeding period.
(Univoltine, with the breeding season restricted to a few weeks in the
warmer part of the vear.) There is thus ample opportunity for the
considerable reduction of the breeding population for any one yeat
(especially in view of the smallness of the area inhabited, which in-
creases the chances of factors acting more or less uniformly throughout
it). Two main factors which might be involved, either singly or together,

30 W oopWaARDb.

in such reduction: exceptionally adverse climatic or other ecological con-
ditions during the long non-breeding period (particularly over winter) ;
an extreme fluctation in (egg) parasite-host balance, involving a tem-
porary decrease in numbers of the latter, followed by a rapid building
up of the population from the survivors. (4) Judging from collections,
the apparently small proportion of males to females in C. nasalis,
materially reducing the effective size of the population. (Wright, 1940,

p. 170.)

The Asopinae are predacious. Thus any possible effects, on the
nology of the species, of the recent vegetational changes that have
occurred on Great Island through the depredation of goats (Baylis,
1948; Turbott, 1948), would presumably have been mainly or entirely
indirect, through their influence on the populations of insect prey. But
since the known range of prey of other species of the sub-family,
including C. nasalis, is very wide, such effects have probably been neg-
ligible or at least much less extensive than with imany of the purely
phytophagous insects, and the population has probably not suffered
adversely. It is possible that the development of a Leptospermum shrub-
land, typically with the insect and other arthropod fauna prolific in
numbers of individuals, has even increased the population. In any case,
as is general in this predacious group, specimens, even if fairly numerous
over an area, would be expected to be rather sparsely distributed within
it.

REFERENCES.

RAYLIS, G, T. S., 1948. Vegetation of Great Island, Three Kings Group, Rec.
Auck. Inst. Mus., Vol. 3, Nos. 4 and 5, 239-252.

BUDDLE, G. A., 1948. The Outlying Islands of the Three Kings Group, ibid.,
195-204.

MYERS, J. G., 1926, Biological Notes on New Zealand Heteroptera, Trans,
N.Z. Inst., Vol. 56, 449-511.

OLIVER, W. R. B., 1948. The Flora of the Three Kings Islands. Ree. Auch.
Just. Mus., Vol. 3, Nos. 4 and 5, 211-238.

TRYON, H., 1892. Zoology of British New Guinea, Part I1.—Heniptera, Ann.
Ouecnsland Mius., No. 2, 15-16.

TURBOTT, E. G., 1948. Effect of Goats on Great Island, Three Kings, with
descriptions of Vegetation Quadrats, Rec. Auck. Inst. Mus., Vol. 3,
Nos. 4 and 5, 253-272.

TURBOTT, E. G., and BUDDLE, G. A., 1948. Birds of the Three Kings Islands,
thid., 319-336.

WRIGHT, S., 1940. The Statistical Consequences of Mendelian Heredity in
Relation to Speciation, The New Systemalics (ed. J. Huxley),
Oxford, 161-183.

The Genus Rhopalimorpha Dallas (Hemiptera-
Heteroptera) with a Description of a New
Species.

By J. G. PENDERGRAST, St. Heliers.

In recent years the consensus of opinion has heen that the genus
Rhopalimorpha Dallas is represented by only two species, Iv. obscura
White (New Zealand and the Chatham Islands) and IR. Juaneralts
Walker (Queensland). R. similis Mayr has long heen regarded as a
synonym of R. obscura, Buchanan White (1878) wrote: “Je. similis
Mayr is, I feel pretty sure, the same as obscura White.’ Mayr (1866)
listed certain differential characters for distinguishing 7. similis trom
R. obscura, On inspection these are found to apply equally well to the
latter species. In his description of similis Mayr (1866) noted that
the scutellum had “eine feine Endspitze.” As will be seen below, the
possession of an acute apex to the scutellum is an important character-
istic of R. obscura. Similarly, R. ignota Hutton has proved to be
synonymous with obscure. After examination of the type Myers
(1924) stated: “The writer is of the firm opinion that R. ignota 1s
admissable not even as a constant colour variety.” And, 1 urther, “After
examining some hundreds of specimens from widely separated localities
the present writer feels sure that there 1s only one species, and that this
is surprisingly constant in structural characters.” Through the courtesy
of Mr. R. R- Forster, of the Canterbury Museum, the author has been
able to examine the type of R. ignota and agrees with Myers that this is
synonymous with RK. obscura.

While working on the anatomy and life history of R. obscura it
hecame apparent to the writer that two species were present in his
collections. Dr. W. E. China, of the British Museum, was kind enough
to examine the collection of that institution and agreed that there were
two species involved. Writing to Dr. T. E. Woodward of the Auckland
University College, he said: «there are two species involved.
These can be distinguished in both sexes most easily by the apex of the
ecutellum which is pointed in one and rounded in the other. The 9
type of KR. obscura White has the pointed apex of the scutellum and
this is the species which is most abundantly represented in our
collection.”

Bécause of the lack of a generic description (Dallas, 1851, provides
only an almost worthless key) and the inadequacy of the descriptions of
R. obscura and R. huimeralis, it is felt that the genus and these species
should be redescribed and it is hoped to make this the subject of a future

paper.

PENDERGRAST.

FAMILY PENTATOMIDAE.
SUB-FAMILY ACANTHOSOMATINAE.
Genus RHOPALIMORPHA Dallas, 1851.

1. Rhopalimorpha obscura A. White, 1851.

Rhopalimorpha stmnilis Mayr, 1864.
Rhopalomorpha stinilis Mayr, 1866.
Rhombocoris sunilis (Mayr) Hutton, 1874.
Rhopalunorpha igneta Hutten, 1898.

2. Rhopalimorpha humeralis Walker, 1867.

3. Rhopalimorpha lineolaris sp. nov. (Figs. 2, 3, 4).

Rhopalimorpha obscura White. Myers, 1926 (partinz).
Length: Female, 8.0-8.5mm. Male, 6.5-7.5mim.

General Colour: Dorsally usually mahogany-brown, frequently
green-brown, sometimes brick-red; ventrally lighter except as detailed.

Head: Dorsally, coarsely punctured with black except for smooth
band in mid-line marked with white posteriorly; tylus bordered with
deep black clefts, anterior extremity prominent and rounded; lateral
jugal margins raised and white. Ventrally smooth and glossy except
for gula and few punctures on gena; gula roughened and _ slightly
pubescent; maxillary plate with conspicuous tooth-like projection with
flat dorsal surface: rostrum reaching intermediate coxae; antennae
slender. reddish-amber, fifth and distal half fourth segment dark brown,
joint between second and third segments inconspicuous, second scarcely
longer than third (1.05: 1.00); eyes dark purple; ocelli bright red,

Thorax: Pronotum and scutellum coarsely punctured except on
callus areas and on smooth median band marked with light stripe;
pronotal margins white or buff; scutellum apex light coloured, broad,
non-acute; hemi-elytron fairly broad, corium green-brown, membrane
buff, neryures light brown; scent gland orifice bordered above by con-
spicuous dark brown plate marked with white dorsally; mesothoracic
carina small but more prominent than in obscura; temur dark brown
with black punctures, remainder of leg dark amber.

Abdomen: Somewhat swollen; connexivum inconspicuous, marked
with black in each segment; venter dark brown mesially, lighter towards
edges; ventral spine broad, extending almost to intermediate coxae.

Female: Sixth sternum marked with pair of conspicuous dark
circular setose patches; valves covering genital opening making up flat
circular area; seventh sternum with broad low median keel,

A New Species of Rhopalimor pha. 33

Male: Ninth segment or pygophor with slightly concave ventral
posterior margin bearing single median patch of long bristles.

Types: All collected Orakei, Auckland, 8/7/50. Host plant: Carex
longifolia. Holotype @ and allotype ¢ deposited in Auckland War
Memorial Museum. One pair (? and ¢) of paratypes deposited in
Dominion Museum, Wellington, Canterbury Museu, Christchurch, and
Otago Museum, Dunedin.

Fig. 1. Rhopalimorpha obscura White. Male. Ventral view posterior abdominal
segments.

Vie. 2. Rhopalimorpha lincolaris sp. nov. Male. Ventral view posterior abdominal

segments.

Rhopalimorpha lineolaris sp. Nov. Female. Ventral view posterior

abdoniunal segments.

Fig. 4. Rhopalimorpha lineolaris sp. nov. Apex of scutellum.

—

’

—
73

ioe)

REFERENCE LETTERING.

H.e—Hemi-elytron. 6-9 —Abdominal sterna.
S.A.—Setose area. 8y.—Valves enclosing vulva.

SJ}

\
.

This species can be distinguished from R. obscura White chiefly by
the non-acute apex of the scutellum, the genitalia and the somewhat
swollen abdomen lacking a definite connexivum. There are other less

34 PENDERGRAST.

obvious distinguishing characters, such as the punctured femora, the
large plate marking the orifice of the scent gland, the more conspicuous
tooth-like projections from the maxillary plate and the more definite
mid-dorsal stripe on head and thorax. In addition, certain differences
in the internal structure have been noted, chief of which is the number
of ovarioles in each ovary. In R#. lineolaris the number is four, while
in . obscura it is seven, as is usual in the Pentatomidae.

Preliminary investigations have shown that R. lineolaris probably
has a very similar distribution to that of R. obscura in New Zealand,
1.e., over the greater part of both main islands. Around Auckland,
individuals of both species are frequently found living together on the
same plant. In the Auckland area the main food plants are Carex longi-
folia; C, divulsa; Juncus effusus; Mariscus ustulatus ; and the Cocksfoot
grass, Dactylis glomerata.

The hte history of both species is at present being investigated and
differences in the nymphs of the two species have heen discovered. It
is intended to describe these in a later paper.

ACKNOWLEDGMENTS.

A\lention has already been made of the writer’s indebtedness to Dr.
China, of the British Museum and to Mr, Forster. of the Canterbury
Museum. In conclusion the writer would gratefully acknowledge the
valuable advice and helpful criticism of Dr. T. E. Woodward, of the
Zoology Department, Auckland University College, under whose super-
vision the work has been carried out.

REFERENCES.
DALLAS, W. S., 1851. List Hemipterous Insects in Brit, Mis., Part 1, pp.
193-197.
FIUTTON, F. W., 1847. Trans. N.Z. Inst., vol. 6, p. 170.
—__— 1898. Trans. N.Z. Inst., vol. 30, p. 159.
MAYR, G., 1864. Verh. Zool. Bot. Ges., Wien., 14, p. 912.

1866. Reise Osterreichischen Fregatte Novara wim die Erde 1857-1859,
Zool. Hemipt., pp. 74-76, Plate 2, fig. 14.
MYERS, J. G., 1924. Records Cant. Museum, II, No. 4, p. 171.
__—_—. 1926. Trans. N.Z. Inst., vol. 56, pp. 502-505.
WALKER, F, 1867. Cat. Specimens Het—Hemiptera in Collection of Brit. Mus.,
Ratt 21. arG:

WHITE, A., 1851. In Dallas; List Hemipterous Insects in Brit. Mus., Part 1,
p. 293.

WHITE, F. B., 1878. Ent. Mo. Mag., vol. 14, p. 277.

ee)
Cyr

The Geology of Rangiawhia Peninsula,
Doubtless Bay, North Auckland.

By M. H. BATTEY, Geologist, Auckland Museum.

Abstract.

3asic pillow lavas, submarine keratophyres and associated sediments (late
Palaeozoic or early Mesozoic) are overlain unconformably by breccia, conglomerate
and sandstone (perhaps upper Cretaceous). Folding about cast-west axes has bent
these sediments vertical. After the folding, tear faults striking north-west have
displaced the country north-east of the fractures north-westwards at least 24 miles.
With this movement, a dyke-like mass of basic and intermediate plutonic rocks was
intruded along the plane of weakness for over 70 miles, from North Cape to south
ot Whangaroa Harbour.

There is a marine terrace just over 100ft. above sea level and one at S0it.

INTRODUCTION.

The northernmost part of New Zealand is comprised of a group
of high, rocky tracts linked together and joined to the “mainland” by
helts of sand hills, of which the chief is that joining the North Cape-
Cape Reinga highland (itself consisting of several smaller blocks linked
by sand dunes). Te Waka o te Haua (The Bluff”), and Mt. Camel to
the high ground near Kaitaia. This region Bell and Clarke (1910)
designated the Aupouri Peninsula, from the name of the Maor tribe
that formerly flourished upon it (text-fig. 1). It is a region of distinctive
seeographical and geological character, comprising massifs that, though
strictly part of the mainland, present themselves irresistibly to the mind
in the character of islands—an effect to which the topography and their
isolation are contributory.

Rangiawhia Peninsula* is a land-tied island forming the promontory
that separates Doubtless Bay on the east from Kangaunu Bay on the
west. at the southern end of the great tombolo leading to North Cape.
The rocky part of the Peninsula (plate 5) rises to a height of 600 feet
above the sea and has rather the shape of a hand with one extended
finger pointing to the north. It is 7 miles long trom north to south and
5 miles from east to west, with an area of 15 square miles and is joined
to the mainland by a belt of sand dunes and swampy ground some 3
miles wide and 8 miles long, in which some of the dunes rise to heights
of more than 100 feet. This tombolo serves also to link together! the
conical rocky hill Puheke and the low-lying rocks around Rangiputa,
both to the west of the Rangiawhia block.

The geology of this area has received but cursory treatment from
earlier geologists, probably because of its relative inaccessibility. Hector
(1891, p. Ixxxi) visited a copper deposit on the south face of Knuckle

ee eee ee ee eS ee
* Sometimes called Cape Karikari Peninsula, from the name of tts northernmost
pronpontory

‘pueppny Yon uUsoyyiou jo deur Ayyeooy “[ “sY-}NoL,

4 a Or S OQ S$ ,
J 7°90 SIIIW JO sTEIS ee

ene WIOSNIN3d IuNOdnv

pivgiyy

U2AMItT UM PiLDyledy) _«o

(AMG 24L,) #1 ws BA
UNV YY 220 VYYMM AL ‘3
Yravag i PUT -AqVauryy,, PN

Avg
nunyvouvy UPL pbussvbu2esvs
0 DADINPVOY Ad T3V°D

24DY 14DY oe adv) Yyz40N

qq suabsag "oN

Geology of Rangiawhia Peninsula, 37

Point that had been prospected in 1847*, and published an interesting
sketch of the cliff-face there. On McKay’s map of 1894 the area is
inarked as composed of Palaeozoic sediments and igneous rocks, the
boundary between solid and drift being delineated. Bell and Clarke
(1910), following McKay (1894, pp. 72, 89, 90) and Marshall (1908,
p. 81), map the rocks as falling in their Older Igneous Group. Appar-
ently neither they, nor Marshall, actually visited Rangiawhia Peninsula.
: - . rocks of the*Peninsula may be grouped as follows (see also
piate 9):

4. Basic and Intermediate intrusives. Dykes, mainly of inter-
mediate nature, invade the older rocks throughout the Peninsula; but
they become more numerous and more massive to the east, giving place,
about Whangatupere Bay and in Cape Karikari to the north-west, to a
continuous mass of coarse-grained igneous rock of a banded nature
which includes thoroughly basic and even ultrabasic types.

3. Breecias, conglomerates, grits and sandstones, appearing as a
narrow strip along the south-east coast of the Peninsula, between
Whatuwhiwhi and Knuckle Point.

2. Hard black shales of uncertain stratigraphic position. They
have so far been found only in the same places as rocks of Group 4.
They lie to seaward of the breccias, conglomerates and sandstones and
are faulted against them, generally below high tide mark.

1. Acid and basic soda-rich submarine lavas, including pillow
lavas, with some associated sediments, underlying most of the south-
western part of the Rangiawhia Block.

GROUP 1.—THE LAVAS AND THEIR ASSOCIATED
SEDIMENTS.
Distribution.

Rocks of this Group underlie all that part of the Peninsula extend-
ing eastwards and north-eastwards from the northern end of Tokerau
Beach to beyond Brodie’s Creek, and inland to the northern side of the
helt of high ground called the Te Ari Ridge that runs east and west
between Whatuwhiwhi and the village of Merita, except for a narrow
coastal strip north-east of Whatuwhiwhi where beds of Groups 2 and 3
occur. Excellent outcrops occur along the coast, but the rocks along
the ridge are in general much weathered and the slopes of the ridge south
of Merita are mantled with sands to a height of about 200 feet. Rocks
of this group also occur along the crests of the ridges leading to Knuckle
Point and Pihakoa Point, where they form remnants of the roof of

* Hector writes that the prospecting was done in 1857, but the Rev. Wm. Puckey,
who was a missionary at Kaitaia, records in his journal having visited the
3rodies at Knuckle Point, whence they had gone to open a copper mine, on
22nd March, 1847.

38 BATTEY.

the intrusion of basic and intermediate rocks. Maitai Pa, in the centre
of Matai Bay*, is also composed largely of them, and they also outcrop,
surrounded by intrusive rocks, inside the north head of Matai Bay.

The rocks of the Group exhibit some variety, and the work so far
carried out permits the recognition of three subdivisions, namely:

(hb) Dark-coloured basic lavas often variolitic and sometimes
showing pillow structure ;

(a) Indurated sandstones and argillites.

(a) Argillites and greywackes. At the northern end of Tokerau
Beach, interstratified argillites and greywackes in beds one inch to eight
inches in thickness outcrop at the base of the cliffs and extend around
the headland to the beach Perehipey. Phe dip wavers in both direction
and amount, but in general it is east-north-east at about 30°. The beds
are disturbed by a number of small faults and broken by closely-spaced
joints running north-west by west and north-north-east. They are over-
lain with apparent conformity by dark lavas of subdivision (pb). A thin
sill-lke mass of lava was seen near the top of the siratified rocks.

In a bluff near the middle of the beach Perehipe thm-hbedded
argillite and greywacke 1s Tollowed upward by lava with quartz amyg-
dales up to 2em. across and dense greenish variolite 3 feet thick. Hard,
splintery greywacke and areillite follow in relatively thick beds much
broken by small dislocations; the strata are thinner (about 6 inches
thick) higher up in the succession. A small reversed fault dipping 43°
to the north-east displaces the beds about 12 feet.

At the eastern end of the beach there is a small outcrop of the
thin-bedded sediments, with wavy but generally horizontal stratification.
These beds meet white-weathering keratophyre on their northern side
in a vertical contact running east by north. The keratophyre sends small
apophyses amongst brecciated wisps of the sediment and has perhaps
slightly hardened it. On the southern side of the exposure the sediments
pass beneath a breccia of keratophyre, some greenish, streaky crushed
lava and argillite fragments. The way in which argillite is mingled with
keratophyre in the headland east of Perehipe will be described below.

The greywackes and argillites make their next appearance at
Knuckle Point, the easternmost point of the Peninsula. They occur
low in the cliffs on the northern side of the promontory, dipping south-
west at a moderate angle. Greywacke is here more prominent than
argillite, and though the rocks are in part thin-bedded, as at Tokerau
Beach, more massive strata also occur. The sediments are closely asso-
ciated with fine-grained dark-coloured lavas, but the relationship of the
two rock types is not well shown. By analogy with the sequence near
Tokerau Beach, it may be presumed that they are interstratified.

* This is commonly spoken of as Merita Bay. The rocky knoll in the centre of the
bay is known to the Maori people as Maitai, and this name, corrupted to Matai
Bay, is applied to the whole bay on all the available maps and charts.

+ For convenience of description a number of Maori place-names, kindly furnished
by local residents, will be used. The localities so named are marked on the
maps.

Geology of Rangiawhia Peninsula, a9

Along the crest of the Knuckle Point ridge and high on its southern
slope, thin-bedded greywackes and argillites are weathered to pink and
white banded clays and dip south-west and west at angles of about oy
These must be the “pink tuffstones” shown in Heetor’s sketch (1891,
p. Ixxxi), for they are conspicuous above and to the east of the green
coloration on the cliff face, which presumably marks the site of the
copper prospect. I did not see the sharp synclinal fold shown in the
sketch, but this may well be visible from seaward.

The attitude of the beds at Knuckle Point seems to be due to move-
tents associated with the intrusion of closely-spaced dykes, trending
north-westward and connected with the Whangatupere Bay intrusion.

At Pihakoa Point, and south of it, there are small areas of indurated
fine and very fine sandstone with splintery or conchoidal fracture show-
ing small scale lamination on fresh surfaces. They are in more massive
beds than the rocks at Tokerau Beach but resemble some of those at
Knuckle Point. As at Knuckle Point, their relations to associated lavas
are not clear. A poorly-defined lamination strikes north-west and dips
at about 50° to the north-east. The influence of the numerous dykes
sent off from the Whangatupere Bay intrusive is again manifested in
the attitude and distribution of the strata. The roof rocks of the
intrusive have a strong north-west gram.

(b) Basic lavas. At the northern end of Tokerau Beach, and in
the bluff near the middle of Perehipe Beach, basic lavas, exposed for
only 18 feet above their base, succeed the argillites and ereywackes with
apparent conformity. In two places the lavas and sediments are inter-
stratified over a small part of the sequence at their junction. The lavas
are dark-green and variolitic with quartz-filled amygdales up to 2cm.
across in parts and patches of cavernous rusty calcite near the base.
Glassy material is not conspicuous, but pillow structure can be recog-
nized. on close examination, in the cliff face at the western end of
Perehipe Beach: the individual pillows are flattened and dip north-east
at about 20° in approximate conformity with the underlying sediments.

Some 34 miles to the north-east, greenish to dark-grey variolitic
lavas with associated black perlitic glass outcrop on both sides of the
entrance to Brodie’s Creek. They show pillow form on the eastern side
of the inlet and along the coast between 300 and 600 yards west-south-
west of its entrance. The pillows are well preserved at the westernmost
part of this exposure, where their elongation indicates that the mass has
heen tilted 66° to south by west: the mutually accommodating curves of
the pillows suggest no overturning (plate 1, fig. 1).

The lower part of the succession is best studied on the eastern side
of Brodie’s Creek. Dark, fine-grained lava appears just within the
mouth of the inlet and is followed to the south by 18 feet of fine-grained
dove-grey sandstone with current bedding, 12 inches of dark lava and
further laminated sandstones. ‘The contacts between lava and sediment
dip 76° north-west by north. The sandstone, which is crossed by
microscopic veins of albitic plagioclase, passes south into a grey-white
flinty rock with an orange-hrown to reddish weathering crust, crossed
by narrow quartz veins and cracks filled with epidote. The lamination
of the sediment gradually disappears as it assumes its new character.

40 SATTEY.

The resulting rock, which extends to the opposite (western) head of
the inlet, has lost almost all signs of clastic origin and is composed of
a mosaic of feldspathic material with scattered quartz grains and little
flakes of biotite. It appears to be a kind of adinole. —

The section at Brodie’s Creek is then interrupted for 50 yards by
a little beach. south-south-west of which sediment gives place to lava
along a plane dipping 52° north-north-west. The lava, which shows
good pillow form, continues for about 75 yards and is succeeded south-
south-eastwards, along a plane dipping 70° north-north-east, by a body
of white-weathering lava which extends for at least 100 yards farther.
Beyond this the coast has not vet been examined.

I think that this section, with its steep northward dips, may be
overturned. No definite evidence on the point was obtained, though the
current bedding in the sandstone may be expected to yield it to more
minute study. Nevertheless, the succession west-south-west of Brodie’s
Creek, which will be described below, where the beds dip south or south-
east, suggests that the beds south of the pillow lavas are the younger,
and at Tokerau Beach interstratification of lavas and sediments char-
acterizes the base of the succession.

About 450 or 500 feet of lava and sediment is exposed in the section
on the eastern side of Brodie’s Creek, which is terminated northward
by a mass of later intrusive rock. A little over 400 feet of lava appears
where the coast cuts across the strike 600 yards west-south-west of the
mouth of the creek, but this second section probably does not extend
so far down in the succession as the first, for the thick bands of sediment
do not appear in it north of the pillow lavas.

There are some points of difference between the succession imcor-
porating the basic lavas at Brodie’s Creek and that at Tokerau Beach.
The thin-bedded argillites and greywackes exposed at Tokerau are not
to be seen at Brodie’s, where intrusive rock limits the section on the
north. As has been mentioned, however, such sediments reappear at
Knuckle Point. The sandstones interstratified with the lavas at Brodie’s
Creek are unlike any exposed at Tokerau Beach. At Brodie’s Creek the
lavas themselves seem much fresher and the pillow form is much better
preserved than at Tokerau Beach. In spite of these differences, however,
it seems justifiable to correlate the basic lavas of these two localities,
not only on lithological grounds but also on structural evidence to be
given below in connection with the rocks of Group 3.

(c) Keratophyres. The keratophyres and quartz-keratophyres are
ereen to light-grey, weathering white, often with amygdales of chal-
cedony and quartz and commonly porphyritic, though the phenocrysts
are sparse. The silica of the amygdales is sometimes stained red, while
the plagioclase phenocrysts may be pink, and these, set in the green base
of a fresh lava, make a handsome rock. The amygdales often stand out
prominently on weathered surfaces, but otherwise the weathered rock
is buff or dirty-white, perhaps with flecks of green, and often earthy in
appearance. Carious erosion of surfaces of quite fresh rock 1s common
along the coast.

Throughout the keratophyres are discontinuous, often ill-defined
bands of light-coloured porphyritic rock streaked with bright-green
chlorite in flattened wisps and ragged ribbons, arranged with their broad

Geology of Rangiawhia Peninsula. 4]

surfaces parallel so as to impart a foliation to the rock, along which it
shows a rude fissility, The largest mass of this green streaky rock seen
is in the cliffs below the schoolmaster’s house at Whatuwhiwhi, where it
outcrops for over 200 yards in the cliffs. The flowing wisps of chlorite
give a lively impression of movement, and I believe that the bands of
green streaky rock represent crush zones along which adjustment has
taken place during the strong folding movements that have affected the
relatively rigid lavas. So far, no order has been discovered in their
distribution and disposition.

As has been mentioned (p. 38), quartz-keratophyre meets thin-
hedded argillite and greywacke in a partly intrusive contact trending
east by north at the eastern end of Perehipe Beach. Thence the kerato-
phyres in their different varieties outcrop eastwards along the coast to
Waiari; beyond this, they appear discontinuously in the cliffs north-
eastwards for about a mile, their exposure being interrupted by the
development of breccia, conglomerate and sandstone beds of Group 3.

There is a good deal of argillite interstratified with keratophyre and
ereenish crush rock between tide marks at the tip of the headland east of
Perehipe Beach, its amount decreasing towards high-water mark. The
planes of junction between sediment and lava dip north-north-east at
70°, Farther east, argillite is enclosed in the lava in such a way as to
give the appearance of a breccia. In places the argillite fragments are
only an inch or so across, but some large masses occur. The planes ot
junction between the larger argillite masses and the lava strike,
generally, east and west, with varying dips, sometimes southerly and
sometimes northerly. Not very much reliance can be placed on the
attitudes of the argillite masses, for it is quite likely that they were
displaced during the extrusion and possibly partly intrusive emplacement
of the keratophyre. The relationships here are reminiscent of those
described by Cox (1915, p. 308) from Pembrokeshire, where, he
believed, the rocks ‘‘represent what were practically lavas which
burrowed among the mud of the sea-flood” and that “at times, the
escaping vapours completely fractured the almost solidified rock, con-
verting it into a breccia.” He found intermingling of sediment and
lava at both bottoms and tops of flows. At Rangiawhia the presence of
the argillite probably indicates either the top or the bottom of a flow, but
the exposure is too limited to show which it 1s.

At Waiari a hard breccia of dark, angular keratophyre fragments,
up to 14 inches across, set tn a grey-white matrix of micro-crystalline
quartz and muddy material, forms a low bluff to the west of the stream
that discharges there. This is regarded as an agglomerate formed by
explosions connected with the effusion of the keratophyre, It is suc-
ceeded southward by a sheeted zone with fractures striking N.60°W.,
in which greenish crush rock penetrates the breccia 11 so intimate a
fashion as to suggest that it represents the alteration product of kerato-
phyre, here intrusive into the breccia before its cementation. Chloritic
keratophyre showing conspicuous carious weathering succeeds the
sheeted zone to seaward. This belt of fracture is probably connected
with the zone of green crush rock below the schoolmaster’s house at
Whatuwhiwhi, which lies on the line of strike of the fractures and 1s
itself broken by master joints trending N.GO°W. and dipping south-
westward at 85°,

42 BATTEY,

The dark-coloured pillow lavas 600 yards west-south-west ol
Brodie’s Creek give place southwards along the coast to light-coloured
porphyritic amygdaloidal keratophyres, which have a prominent wide-
spaced parting, taken as bedding, that dips south-eastwards at about
40°. Though quite distinct from the usual type of pillow-forming lava,
these rocks have shreds and streaks of glass associated with them and
are definitely pillowy in places. They outcrop continuously, save for
interruption by dykes, for about 700 yards along the coast, to their
contact with breccia and conglomerate beds in a headland 1,100 yards
south-west of the entrance to Brodie’s Creek. |

The apparent thickness of the keratophyres in shis section 1s a little
more than 1,000 feet, but this value is possibly greater than the true
thickness, for, as will be shown later, tear faults striking north-west
have allowed north-westward movement of the country on the north-
eastern sides of the fractures at short intervals along this coast. These
faults are difficult to detect in the lavas and have only been demonstrated
where the stratified rocks of Group 3 are exposed, but they undoubtedly
transect the lavas just as frequently, and may have caused repetition
of part of the keratophyre sequence. |

Inland, keratophyres outcrop in the Te Ari Ridge running east and
west between Whatuwhiwhi and Merita; in general the rocks are much
weathered, but fresh amygdaloidal keratophyre outcrops in the knoll
east of the Whatuwhiwhi-Merita Track where it crosses the ridge.

The lavas that form remnants of the roof of the intrusive mass
of Whangatupere Bay, along the ridge between Koware trig, station
and Pihakoa Point, prove under the microscope to be thermally meta-
morphosed representatives of the keratophyre group. In hand specimen
they are quite different from the rocks of the southern coast of the
Rangiawhia Block, being dense, dark-grey and splintery, and to be
distinguished only with difficulty, under the lens, from the indurated
ereywackes with which they are associated. The same ts true of the
lavas surrounded by intrusive rocks in the north-western corner of
Matai Bay.

Structure, correlation and age.

On the whole, little direct evidence on the structure bas been
eleaned from the outcrops of the rocks of Group |. Observations on
their contact with argillite and greywacke and on planes taken to
represent bedding in. the lavas suggest that, -between Verelupe and
Waiari, they strike roughly cast and west. ‘The dip is variable, either
northward or southward, steep or moderate. Around Brodie’s Creek,
too, the strike varies about the east-and-west direction, though in the
keratophyres south-west of the inlet it swings to north-east. The beds
dip steeply northward east of the inlet and southward or south-eastward
at steep to moderate angles south-west of it,

From the study of the breccias, conglomerates and sandstones of
Group 3, it is clear that the country has been subjected to vigorous

folding about east-and-west axes and, in so far as they were capable of
vielding by flexure, the lavas must have partaken of this deformation,

Geology of Rangiawhia Peninsula, 43

Since these rocks are rather rigid, probably some of the adjustment took
place by fracture and crushing. It is not possible to attempt to recon-
struct the attitude of the lavas before this Tolding.

Since the folding, the area has been broken by tear faults striking
north-west. the results of which will be more conyeniently considered
below in connection with the structure of the beds of Group 3. Here
it may be mentioned that there is some evidence that the pillow lavas
near Tokerau Beach and those near Brodie’s Creek are at the same
horizon and once formed a belt striking about east-and-west that has
heen disrupted and displaced by the tear faults (text-hg. 3, p. 49).

So far. no fossils have been found in Rangiawhia Peninsula to fix
the age of any of the rocks there, The nature and sequence of the lavas
invite direct comparison with those discovered on Great Island, in the
Three Kings Group, by Bartrum (1936a&b). He found spilitic pillow
lavas 40 feet thick at the base of the cliffs in North West Bay, associated
with greywackes showing fine lamination in places and hard black shales,
while albite porphyry*, estimated as 60 feet thick, outcrops high above
in the cliffs, and quartz-keratophyre lava (perhaps partly tuffaceous )
is recorded from surface blocks above the porphyry.

We have here a section strictly comparable, even in detail, with that
at Rangiawhia Peninsula. Lanvination of the greys rackes is character-
‘stie of both areas, while Bartrum’s description of the pillow lavas 1s
applicable in all points to those near Tokerau Beach. I have not been
able to visit these pillow lavas near sea level at the north-west landing
on Great Island, but the overlying keratophyres, whether they be
intrusive or extrusive, are comparable, both im appearance in the field
and under the microscope, with types from Rangiawhia Peninsula. It
may be remarked that the albite porphyry of Great Island is. richly
amygdaloidal in places, while little, if any, sedimentary rock intervenes
between its outcrop and that of the rocks deseribed by Bartrum as
lavas higher up the cliffs, though unfortunately talus obscures the
accessible ground. Its character as a sill can therefore not be regarded
as above question; it may well be part of a group of lava flows. It 1s
remarkable that even the tuffaceous rock (Bartrum, 1936a, p. +16) finds
its counterpart in the brecciated keratophyre at Watatt.

Phere can be little doubt that the rocks of these two areas, 85 miles
apart, belong to the same formation.

Bartrum believed that the rocks of Great Island belong “to the
Hokonui System . . . approximately mid-Mesozvic in age od
(1936a, p. 415) and correlated them with lavas interbedded in the
Waipapa Series of Bell and Clarke (1909) around Whangarea, The
recent discovery by the Geological Surveyt of fusilines and corals in a
marble associated with pillow lavas, in the Waipapa Series near Whanga-
roa, proves that this part of the Series is of Permian age, so that possibly
hoth the Great Island rocks and those of Group 1 in Rangiawhia
Peninsula are also as old as this.

* Tntrusive keratophyre in the terminology of Wells (1922).

* Personal communication from) Dr. J. Marwick.

4 BATrey.

Bartrum specifically observed that the Great Island igneous rocks
are not to be regarded as the magmatic associates of the Upper Creta-
ceous pillow lavas near Cape Maria van Diemen, but are older: “their
period of eruption is separated from that of the lavas of the mainland
by one of the greatest unconforimities yet established in New Zealand”
(1936a, p. 422). Specimens of lava collected froni South West Island
and Hole-in-the-Wall Rock (one of the Princes Tslets) by Buddle and
Johnston in 1947 were classified by Bartrum as quartz andesites resem-
bling members of the Upper Cretaceous (Rahia) Series of the area
around Cape Maria van Diemen, to which formation he assigned them
(Bartrum, 1948, p. 206). Since the rocks of the North Cape-Cape
Reinga area known as the Older Volcanic (or Whangakea) Series (Bell
and Clarke, 1910; Bartrum and Turner, 1928) have been transferred
to the Upper Cretaceous (or Rahia) Series (Bartrum, 1934), Great
Island has been the sole remaining bastion of supposedly’ pre-Cretaceous
rocks in the Far North.

To return to Doubtless Bay, there is a belt of basic submarine lavas
with pillow structure which forms the headlands along the southern
shore of the Bay, 8 miles south of the Rangiawhia Block, from Mango-
nut Township at least to the promontory west of Taipa Beach. Inland
of these, in Taipa Estuary, are Cretaceous sediments striking east-and-
west and dipping northwards at moderate to steep angles, which closely
approach the lavas on the east side of the Estuary, but are not to be
seen actually in contact with them. Green cherts occupy the interstices
between the pillows, and discontinuous bands of red and green chert
are caught up in the lavas. The largest mass of sediments seen in the
lava is at the western end of Taipa Beach, where the included and partly
invaded sedimentary beds strike south-west by west. Because of the
close association of the sedimentary beds with the lavas and their general
conformity of strike, it seems very likely that these lavas are part of
the Upper Cretaceous succession here. Besides, pillow lavas are well
known to be associated with Cretaceous and Eocene beds in many other
parts of North Auckland.

Vhere are no keratophyres between Mangonui and Vaipa, however,
and the sediments associated with the pillow lavas are quite different
from those in Rangiawhia Peninsula. Moreover, while the lavas
between Mangonut and Vatpa are freely invaded by a medium to coarse-
erained albite diorite, which is probably as abundant as the pillow lavas
in the coastal outcrops, no such intrusive appears in Rangiawhia
Peninsula.

lor these reasons, admittedly not highly satisfactory, | am inclined
to think that the Javas of Group 1 at Rangiawhia are not properly to be
correlated with those of the southern shore of Doubtless Bay, but are
more probably pre-Cretaceous.

The question whether they are coeval with the Palaeozoic members
af the Waipapa Series at Whangaroa, mentioned above, must be left
in abevance until fossils are found, or until more is known of the suc-
cessions and rock types in areas where there is fossil evidence of age,

Geology of Rangiawhia Penmsula. 45
GROUPS 2 & 3.—SHALE GROUP AND BRECCIA,

CONGLOMERATE AND SANDSTONE GROUP.

These two groups of rocks are exposed in close association along
the same stretch of coast and it is convenient to consider them together,
with the aid of a large scale map (plate +).

Distribution.

Breccias, conglomerates and sandstones of Group 3 outcrop as a
narrow, interrupted strip along the coast between Waiari and Brodie’s
Creek and again, apparently, at the old copper workings on the southern
side of Knuckle Point. JI have not yet visited this last locality, but
Hector (1891, p. Ixxxi) describes the deposit, which he regarded as a
volcanic agglomerate, as being the country of the copper-ore and his
cketch of the cliff shows that an excellent exposure of the junction
between the breccia and the older lavas is there available. This contact
has been examined on the coast at a point 1,100 yards south-west of the
mouth of Brodie’s Creek as well as in the area deseribed in detail helow.

The accompanying map (plate 4+) represents the relationships of
the formation for a distance of about a mile east-north-eastward from
Waiari. The strip of rocks of Group 3 is nowhere exposed for a width
of more than 150 yards and its greatest inferred breadth 1s only 300
yards. Marine erosion has to proceed but little farther to remove the
members of the Group altogether. The waves have, in many places,
worn away the land until the resistant lavas have been reached, so that
the rocks of Group 3 are exposed chiefly in the low tide platform,
though also in the cliffs in some places (plate 2, figs. 1,2). They’ strike
steadily east and west (save for a slight systematic curving due to
faulting), everywhere stand very steeply and are often overturned to
the south (plate 2, fig. 2).

Close to their contact with the lavas the breccias are very coarse
and contain masses, several feet across, of greenish crushed lava, normal
keratophyre and dark grey, indurated, fine sandstone. On the eastern
side of the point at Waiari they contain large boulders of graphic
eranodiorite (plate 2, fig. 3) of which no parent mass is now exposed*,
The majority of the smaller blocks and pebbles are of keratophyre.
Other constituents are spotted greywacke that has undergone mild
thermal metamorphism, coarse sandstone and pieces of an older con-
elomerate. Much of the coarser material 1s angular, but a fair propor-
tion of well-rounded boulders and pebbles is present. Generally
speaking, the material of the breccias and conglomerates becomes finer
upwards in the sequence and gives place to fine conglomerates, grits
and sandstones.

These characteristics of the beds and their distribution suggest that
the group represents the basal part of a normal marine sequence

nnn nEEE nnn nnn

“Tt may be recalled that smaller pebbles of granitic rocks occur in basal conglom-
erates of the Kaeo formation in the Whangaroa district (Bell and Clarke,
1909, p. 50).

415 BATTEY.,

deposited unconformably upon the lavas of Group 1, rather than a
volcanic agglomerate as Hector supposed. The great size and angular
form of many of the boulders make it clear that the land from which
the fragments were derived was very close to the area where the
deposits lie—some of the breccias, in fact, look like cliff-foot talus
accumulations.

The basal contact of the breccias with the lavas is well exposed in a
few places, but, since the beds stand at high angles and are overturned
in many places at the contact with the relatively unyielding lavas, the
Junction may be expected te have been a plane of movement and, as
such, is not capable of afferding much information about the original
nature of the surface on which deposition took place. Later faulting
has also occurred along lines crossing the contact. The possibility of
the generation of independent zones of breccia by both kinds of move-
ment cannot be disregarded, although the actual lines of such brecciation
would be difficult to detect in the field. The magma of the injected
series (Group 4+) has taken advantage of the contact as an easy route
of uprise in a number of places.

Between half a mile and a mile north-east of Waiari the beds of
Group 3 are limited to seaward by a fault striking parallel to the coast,
generally between tide marks, which brings the breccias, conglomerates
and sandstones into contact with hard, coal-black slaty shales, the parting
in which strikes slightly oblique to the fault-line and generally dips
south-east at about 60°. These shales reappear at the extreme edge of
the low-tide platform south-south-east of the cliffs at Walari. Small
pebbles of shale along the beach just east of the schoolmaster’s house at
Whatuwhiwhi suggest that the shale is not far off shore farther to the
south-west.

Several sills intrude the shales and they are crossed also by dykes,
both kinds of intrusion being representatives of Group 4 (plate 3,
figs. 2, 3).

Only for a distance of 40 yards about 200 yards north-east of
Homarere, have the shales been seen in contact with the lavas of Group
1. Here, though the contact is poorly exposed, the shale seems to be
faulted against the lavas along a plane striking roughly parallel to the
fault separating the shales from Group 3 beds elsewhere, and dipping
north-west at 63°.

Structure.

As the beds are followed north-eastwards from Watari, the east-
and-west trending contact of the breccias with the lavas 1s found to be
progressively displaced to the north by a series of tear faults striking
north-north-west. The movement of the country to the north-north-west
or north-west on the north-eastern sides of these faults is well displayed
by the displacements of the fault plane separating the black shales from
the breccias, conglomerates and sandstones, and as can be seen from the
mapping, the effects of drag along the faults is clearly exhibited in the
curving of the strike of the beds of the breccia-sandstone group as well
as in that of the black shales at Homarere.

Geology of Rangiawhia Peninsula. 47

In only two cases 1s the apparent relative displacement along the
faults in the opposite sense and in only one case is this appearance truth-
ful—that is, in the case of a reverse shift of 20 yards near Anaputa.
The second apparent case, where the shales meet the lavas 200 yards
north-east of Homarere, is due to a different cause. It will be seen that
the fault that separates the shales (Group 2) from the breccias, etc., of
Group 3 strikes obliquely to the east-and-west contact between beds of
Group 3 and lavas of Group 1. If we restore the beds to their original
position, before the north-north-west shifting took place, we find that the
trend of the fault must have caused it to approach ever nearer to the
breccia-lava contact as this contact was followed eastward, until it
crossed the boundary at a point that now lies 200 yards north-east of
Homarere (Grid position 92219838). Movement along the fault
promptly died out as the plane passed into the infrangible lavas and the
displacement was taken up by another plane of the same trend in the
more readily broken beds of Group 3 (text fig. 2). The trend of the

Breccia conglom- SS
Lovee Ea] Sete glen, Eazy Shales
TE erate & sandstone = =

Text-fie, 2, Diagram showing relations of shales to breccia group and lava group
north-east of Waiart.

new fault plane, however, still brings it continually nearer the breccia-
lava contact eastwards, and it must encounter the lavas again at a point
that now lies immediately north-east of the area mapped in detail. I
hope that its behaviour on doing so will be observable here when the
mapping is extended farther north-eastwards, though igneous intrusion,
the trend of the coast and the rugged cliffs together seem to militate

against it.

Before the wider implications of this tectonic pattern are discussed,
another point in connection with the map (plate 4) may be mentioned.
Faults are difficult to detect in the lavas and in the jumbled mass of the
lower part of the breccia. Four faults not observed seem necessary to
explain the distribution of the beds. Three of these have been drawn
‘accordance with the north-west or north-north-west grain of the
country so clearly seen in the disposition of dykes and joints, other
faults, and in the coastal morphology, The fourth, at Homarere, has
heen drawn with a north-north-eastward trend, on the basis of the
positions of lava outcrops protruding through the beach gravels, and ot
a mass of shale exposed a few dozen yards off-shore at low water,
supported by the presence of a prominent cleft in the lavas in the cliffs
and the position of a small outcrop of breccia. Although this trend 1s

48 BATTERY.

tectonically inharmonious, [ have preferred not to multiply assumptions
in order to dispense with it, for the observed fault plane at Waiari shows
that divergences from the general trend do occur. Obvious difficulties
arise, however, in the area indicated by the question-mark. Further
observation will probably show some deficiency in the mapping here.

_ When the style of tectonics found north-east of Waiari is applied
in explanation of the distribution of all the outcrops of Group 3 beds
so far known, including that described by Hector at the copper mine, we
find that, across a belt of country of a width of nearly 34 miles at right
angles to the planes of movement (taken as trending about N.40°W.),
the base of the breccias has been shifted north-westwards a total dis-
tance of over 24 miles. In the outcrops that I have myself examined,
of which the most north-casterly is that 1,100 yards south-west of
Brodie’s Creek, there has been a north-westward shift of 1,650 yards
distributed over a distance of 2,270 yards normal to the direction of
movement”,

This displacement 1s probably connected with the intrusion of the
Whangatupere Bay basic igneous mass, or rather, the intrusion, which
appears to connect with simiuar rocks at North Cape and perhaps around
Whangaroa, ts probably connected with the displacement. If this is so,
it seems fair to assume that the amount of disturbance and relative
north-westward shift will decrease south-westwards, away from the
intrusive mass and the supposed main plane of weakness. In other
words, the curve of shift against distance from the intrusive will flatten
out. We have not enough data to find out just how rapid this flattening
out may be, but some estimate of it may be attempted. The average
eradient of the curve over the whole distance between Waiari and the
copper minef is 77%. Between Waiari and the base of the breccia
south-west of Brodie’s Creek it is 72.5%, while between this last point
and the copper mine it is 80%.

This flattening of the curve of shift against distance from the
intrusion becomes of interest when the distribution of the rocks of
Group | is examined in the light of the tear fault movements. Let us
assuine that, over a further 2.400 yards south-westwards of Wauari, the
average gradient of the curve falls to 70%. This means that the breccia
base at Waiari must be shifted 1,680 vards to S.40°E., relative to the
outcrop of the pillow lavas near the northern end of Tokerau Beach.
Then, restoring the breccia base north-east of Watari to its original
east-and-west disposition, and plotting the position of the pillow lavas
700 yards north of the breccia base south-west of Brodie’s Creek, we
find that a line joining the pillow lavas at Tokerau Beach and the new

*Tn calculating the amounts of these displacements the strike of the base of the
breccia has been taken as being east-and-west. There is a suggestion, in the
trend of the relatively long section of the basal contact exposed south-west of
Homarere, that it may have been originally more nearly east by south, the
effect of drag on the tear faults having affected, to some extent, the whole of
the short sections of the beds between them. If this is so the gradient of shift
must be greater.

+ It must be realized that the position of the contact at the copper mine is only
approximately known at present.

Geology of Rangtawhia Peninsula. AY

position of those of Brodie’s Creek trends east by south (text-fig. 3).
It will be recalled that the few data gained from the outcrops of the
Group 1 lavas suggested a more or less east-and-west strike, with
which the results of the present reconstruction harmonize quite well.
Further comment on the significance of this zone of tear faulting is
eiven in discussion of the associated intrusive rock (p. 55).

A Knuckle Point

Breccia base ae ee bree_, se U x
Pillow lava outcrops seo a = et

Inferred posttion of pillow lavas ooo rons 9 woo yards
aes Ss) ee

Dipin lavas ‘angle in aegrees ¥ 50

Vext-fig. 3. Diagram showing the effects of tear faulting along south-east coast
of Rangiawhia Peninsula.

Correlation of the pillow lavas at Tokerau Beach and Brodie's
Creek implies the persistence of basic submarine lava-flows at roughly
the same horizon for 7,000 yards along the strike. This demand will
be the more readily granted for the [act that, as has already been
mentioned, between Taipa Beach and Mangonu Township, on the
opposite shore of Doubtless Bay, basic variolitic lavas, showing pillow
structure and similar in many respects to those around Whatuwhiwhi,
form a continuous east-and-west belt for at least 8.250 yards along the
coast.

The strong folding about east-and-west axes recorded by the rocks
of Group 3 supports the view that this fold-direction represents a trend
of fundamental importance in the Far North.

Structural data on the region north of the Ivactea aid Maungit-
taniwha Ranges, which extend west-south-west from south of Whanga-
roa Harbour to south of Kaitaia, are not numerous. Concerning the
northern flank of the Maungataniwha Range, Mclay (1894, p. 72)
remarks that “the Cretaceo-tertiary rocks sweep round the older rocks
in all directions and, except where contact is made along lines of fracture
by faulting, are usually ‘nclined at moderate angles.” He records dips
to the west and north-west at Peria (1894, p. 85).

Marshall, who made only a rapid traverse of the country, states
(1908. p. 8b) that a north or north-north-east strike “appears to be

50 BATTEY.

represented in the hills between Mangonui and Oruru Valley and in the
shales that are occasionally displayed in the range extending from Reef
Point to Raetea.”’

In the Whangaroa district, Bell and Clarke (1909, p, 42) found
that their Waipapa Series, which they regarded as Palaeozoic or early
Mesozoic in age, is folded into an anticlinorium that trends west-north-
west and east-south-cast on the eastern seaboard and east-north-east and
west-south-west in the inland area. Dips range from 30° to vertical.
The rocks of their Kaeo Series, of Cretaceous and Tertiary age, they
found to have a very variable strike, but they concluded that in the
southern area around Lake Omapere the strike is east-north-east and
west-south-west, while in the area around Whangaroa it is meridional
(p. 47).

Ongley (in Morgan, 1919, pp. 39-40) has supplied data on the
attitudes of limestones in the Victoria Valley and near Kaitaia. His
observations suggest a dominant south-eastward strike and moderate
south-westward dips. One almost east-and-west strike with a steep
southward dip and one north-east strike and north-westward dip are
recorded.

To these observations must be added the fact that Cretaceous rocks
strike east-and-west and dip northwards at angles from 30° to 70° for
a mile across the strike in Taipa Estuary.

In the North Cape-Cape Reinga area more detailed information is
available in the work of Hector (1872), McKay (1894), Bell and Clarke
(1910), and Bartrum and Turner (1928). From this work it is clear
that the lavas and sediments of the Rahia (Upper Cretaceous) Series,
with which the Whangakea Series is now united (Bartrum, 1934),
strike east-and-west or west-north-west and east-south-east with
moderate to steep dips. In the lowest division of the Coal Point (Mid
and Upper Tertiary) Beds Bartrum and Turner record prevailing
southward dips, with which McKay’s statement (1894, p. 72) agrees.
but in the middle and upper divisions the dominant direction of dip is
south-westward at angles of from 15° to 35°.

These observations seem to justify the conclusion that there exists
an important development of post-Cretaceous folding about east-and-
west axes in the region north of the latitude of Reef Point and the Bay
of Islands which has not so far received much attention save from Bell
and Clarke (1909) in their statements on the Waipapa Series and the
southern area of the Kaeo Series.

To move into a more speculative field, it seems possible that there
is some connection between this east-and-west folding and the trends of
the Raetea and Maungatantwha Ranges; the high ground on Rangi-
awhia Peninsula, including Puheke; the North Cape-Cape Reinga block
and the line of summits represented by the Three Kings Islands,
Generally speaking, these areas have cores of hard rocks with Cretaceous
and later sediments lying off them with, in many cases, northward and
southward dips. When McKay (1894, p. 72) writes that ‘the Cretaceo-
tertiary rocks sweep round the older rocks in all directions” it is possible
that he is describing the effects on the attitude of the strata of axial
culminations and depressions in a system of east-west folds.

Geology of Rangiawhia Peninsula, 5]

It should be noted that Bartrum and Turner (1928, pp. 130-7)
concluded that the evidence from North Cape supports the view that
“fold-axes open out to the north and north-west of the main structural
axis of New Zealand and pass through North Auckland as an integral
part of |Suess’s| ‘Third Australian Arc.” In reaching this conclusion
they were influenced by the north-westward trend oi the foliation in the
eabbro of North Cape, The present study proves, however, that this
gabbro. which must certainly be part of the same mass as that exposed
in, Rangiawhia Peninsula, was intruded under the influence of powerful
north-westwardly directed shearing forces, to which its foliation may be
aseribed. and that the intrusion and shearing were both later than the
folding movements.

The mechanics of tear faulting have been dealt with by E. A.
Anderson (1942, pp. 13-14 and Ch. V; see also Kennedy, 1946,
pp. 67-70). Tear faults (transcurrent faults ) theoretically should occur
in two sets inclined at somewhat less than +5° to the direction of
ereatest pressure, the movements on one set being sinistral and on the
other dextral. One or other of the two sets may in some cases he
suppressed.

The trend (N.40°W.) of the tear faults in our area and the fact
that the displacement on them is sinistral means that, to accord with
Anderson's theory, the direction of greatest pressure must have been
west by north and east by south. We are not dealing, therefore, with a
system of folds and faults related to one direction of compressive force,
like that so beautifully shown in South Wales (Anderson, 1942, p. 56).
On the contrary, we must infer a change from north-south compression
to east-west compression between the period of folding and the period
of tear faulting. That such a change should take place within a relatively
short span of time is surprising, but on the evidence available it seems
that the idea must be entertained provisionally. The meridional strike
found by Bell and Clarke (1909) in the northern area of their Kaeo
Series may perhaps support the idea that such a change did take place.

There have been two main ideas about the structure of North
Auckland. (1) The trend of the North Auckland Peninsula is not
connected with the direction of fold axes, but 1s the result of a later
system of fractures. “This was the early view of Suess (Tf, p. 146) and
was adhered to by Marshall (1908), Bell and Clarke (1909), Park
(1921) and Benson (1924, pp. 128-132), though these geologists were
not at one about the directions of fold axes. (2) The trend of the
North Auckland Peninsula is the expression of a system of north-west
trending folds. This was Suess’s later view (IV, p. 318) and was held
tentatively by Ferrar (1925, pp. 18, 33), more definitely by Bartrum
and Turner (1928, pp. 136-7), and with considerable confidence by
Macpherson ( 1940).

The study of Rangiawhia Peninsula supports the first of these
conceptions and, more specifically, shows that the early folding there
was about east-and-west axes and was presumably due to north-south
compression, while it suggests that the later north-west fractures were
due to east-west compression.

ep BATTEY.

[Enough is not known of the kind of displacement on the fault that
brings black shales of Group 2 against beds of Group 3 for it to be
definitely related to either of these two inferred systems of forces. The
faulting took place before the north-westward tear faulting and probably
after the east-west folding, for if the Group 3 beds are first unfolded
an improbable flat-lying thrust must be invoked to explain the relations
of the shales to the breccia group. The relative ages of the two sets of
beds are not known with certainty, but from the appearance and degree
of induration of the shale as compared with the breccias, conglomerates
and sandstones, there can be little hesitation in pronouncing the shales
the older. Where it meets the lavas of Group 1 near Homarere the
fault hades 27° to the north-north-west. Jévidence has already’ been
given for the belief that the movement in the area mapped took place
on two faults en echelon. If these are normal faults, a third system of
forces is required to explain them. It seems more likely that they are
related to one of the other sets of movements, in which case they are
probably tear faults, and their behaviour where one of the planes meets
the Group 1 lavas would suggest sinistral displacement. which would
relate them to the epoch of north-south compression,

Age.

Nothing is known of the age of the shales of Group 2. From their
indurated nature they are considered, without much doubt, to be older
than the Group 3 beds.

In the absence of fossils and of descriptions of comparable becls
elsewhere the age of the beds of Group 3 must also remain uncertain.
Reasons have been given for the belief that the underlying rocks may be
pre-Cretaceous, and it was there mentioned that the breccias, sandstones
and conglomerates do not have a Tertiary aspect.

Conglomerates at the base of the Kaeo Series (Cretaceous and
Tertiary) in the Whangaroa district contain pebbles up to 2 inches across
of graphic granitic rocks (Bell and Clarke, 1909, p. 50), and pebbles of
graphic granodiorite occur in Upper Cretaceous conglomerates on the
southern shore of Hokianga Harbour (Mason, 1948). The large size
of the masses of graphic granodiorite at Waiari, five or six feet across
(plate 2, fig. 3), indicate that the terrain from which they came was
close at hand. The presence of these boulders, together with the degree
of consolidation and general field appearance of the grits and sand-
stones accompanying the conglomerates are virtually all the justification
that exists for assigning to them an Upper Cretaceous age and regarding
them as basal beds of the post-Hokonuian transgression. As there will
be occasion to notice later, Bartrum (1934) correlates the intrusive of
North Cape with the supposedly early Tertiary intrusives of Silverdale,
near Auckland, and since the intrusives of Rangiawhia represent another
portion of the North Cape mass, and cut the beds of Group 3, we have
a tentative upper limit to the stratigraphical position of the latter and
to the date of their folding.

GROUP 4.—INTRUSIVE IGNEOUS ROCKS.

Distribution.
As has already been mentioned, dykes of medium- to course-grained
intermediate and basic igneous rock intrude the other solid formations

Geology of Rangiawhia Peninsula, 53

in all parts of Rangiawhia Peninsula, but become more numerous and
more massive eastwards, until they give place, around Whangatupere
Bay, and in Cape Karikari, to a continuous mass of coarse-grained
igneous rock.

The average trend of dykes is north-west by north, though those of
a group in the headland east of Perehipe run more nearly north, With
the exception of two dykes running east-and-west at the northern end
of Tokerau Beach, the only other trend represented is between north-
east by north and north-east by east near the contact of breccias and
conglomerates of Group 3 with the lavas of Group 1 between Waiari
and Brodie’s Creek, and in the black shales, in which the intrusions are
parallel with the parting.

The dykes are easily weathered and the sea has usually eroded them
away to form north-west trending clefts and inlets along the coast. It
may clearly be seen, as one proceeds eastward, how this grain gradually
overmasters the general north-eastward trend of the south-eastern coast
of the Peninsula and eventually becomes completely dominant and
dictates the course of the coast-line at Whangatupere Bay and in Cape
Karikari. Its influence, in reaction with the stubborn relics of the
invaded country, produces the bipartite form of Matai Bay, an effect
seen to even better advantage in the coastal morphology on both sides of
Bergen’s Point, at the south head of Doubtless Bay ( Provisional One
Mile Sheet N.7), and particularly in Taimaru Bay and the charming
little trefoil indentation of Waimahana Bay.

In the deep, narrow bay 1,000 yards south-west of Brodie's Creek
the pinnacles of rock arranged along north-west lines, which break the
surface of the water, are masses of country’ rock resisting the sea after
the intrusive rock that formerly enclosed them has been driven back a
furlong to the north-west.

The south-western shore of Whangatupere Bay is composed almost
wholly of intrusive rock with only small amounts of included hornfelsic
country rock. In the ridges leading to the headlands of Knuckle Point
and Pihakoa remnants of the roof of the intrusive are preserved in the
form of metamorphosed lavas and sediments of Group 1, invaded by a
swarin of dykes springing from the intrusive mass below. Quite large
masses of country rock continue down to sea level at the north-western
and south-eastern ends of Whangatupere Bay.

The western margin of the intrusion cannot be investigated along
much of its length for lack of outcrops, but it probably does not possess
a mappable boundary, passing rather, by increase in the amount of

‘ncluded rock, into a swarm of dykes, as for example in Maitai Pa*
(plate 3, fig. 1) and the north-western corner ot Matai Bay.

The rock composing the intrusion is petrographically very variable.
It ranges from leucocratic types well seen in the two bluffs on the
southern shore of Matai Bay to melanocratic rocks near the north-
western end of Whangatupere Bay and in Cape Karikari. The various
types have not yet been petrographically examined, but in the hand-
specimen they are seen to include light- and dark-coloured diorites,

——

Se ee ee
* The cusp in the centre of Matai Bay.

5.4. BATTEY.

gabbros, pyroxenites and greenish, porphyritic, uralitized or chloritized
andesites which appear to invade the other rocks in irregular dyke-like
masses. Narrow veins of a fine-grained rock with a very small propor-
tion of dark constituents traverse the complex. The contacts between
the different rock types are sometimes sharp, sometimes gradational ;
they run straight for only short distances in most cases, but their general
trend is between north and north-west.

The dykes west of the main intrusive mass are mostly andesites and
porphyritic micro-diorites.

In the main mass at Whangatupere Bay there is a well developed
joint pattern with one set of vertical joints, along which there has been
intense shearing, running N.30°W. to N.40°W., and another set roughly
at right angles to this, while a more or less horizontal sheeting char-
acterizes the rocks exposed along the coast. The joints cross all types
of rock in the complex indifferently.

The Rangiawhia intrusion strikes directly towards the massif of
ultrabasic and basic rocks that forms North Cape and Kerr Point, in
which there is a foliation trending steadily north-west and south-east in
conformity with the strike of the south-western margin of the mass
(Bartrum and Turner, 1928, p. 123 and map, p. 99). The gabbro at
Whangatupere Bay and Cape Karikari is very similar in hand specimen
and field cecurrence to that exposed on the shore of the bay on the
southern side of North Cape, and due west of the lighthouse, where
leucocratic veins and hornfelsic inclusions lke those at Whangatupere
Bay are to be found. No peridotites, however, have been recognized in
Rangiawhia Peninsula. Since Bartrum and Turner (1928, p. 121)
remark that gabbroic rocks intrude the ultrabasics at North Cape par-
ticularly on the western margin of the massif, it is possible that peri-
dotites lie off shore north-cast of the coast at Rangiawhia Peninsula.
On the other hand, it seems that there is a greater proportion of
leucocratic rock-types exposed in Rangiawhia Peninsula, while biotite,
which is not recorded in the descriptions of North Cape rocks, is con-
spicuous in the dicrites and some of the gabbros at Rangiawhia. This,
together with the preservation of part of the roof of the intrusion there,
suggests that in the south-east we have exposed the upper part of the
plutonic mass, which rose to a higher level in the crust at North Cape,
where its deeper parts are now visible. The fact that much coarser-
erained rocks (with crystals up to four inches across) are described
trom North Cape than any yet found at Rangiawhia may support this
idea. Petrographic study of the rocks should throw more light on the
question.

South-eastward, the Rangiawhia mass strikes towards Bergen’s
Point and thence to the intrusive mass of gabbro, diabase, diorite and
andesite of post-Waipapa (Palaeozoic-early Mesozoic) and probably
post-Kaeo (Upper Cretaceous-Tertiary ) age mapped by Bell and Clarke
(1909, pp. 76-7) around Haunga trig. station south-south-west of Kaeo.
Little is recorded of the grain of the intrusive body here, but the rock
types are the same as those to the north-west, and the copper mineraliza-
tion associated with them provides a link with the Rangiawhia rocks.

Geology of Rangiawhia Peninsula, 55

It seems very likely that they should be regarded as part of the same
great dyke-like mass”.

The correlation of the intrusions along the line between North Cape
and Haunga implies that the tectonic environment and style of faulting
so intimately linked with it at Rangiawhia probably prevails over the
came distance. Bartrum and Turner (1928, p. 104) infer that the
south-west side of the North Cape mass is faulted, and Bell and Clarke
(1909, p. 80. etc.) emphasize that the country of the Pupuke copper

tnines around Haunga is much faulted.

This belt of intrusion and inferred faulting, from North Cape to
Haunga, is probably part of a very’ much larger tectonic element, for,
when its line is continued south-east of Haunga beneath the Kerikeri
basalt flows, it emerges to run accurately along the south-western margin
of the basement rock (here mapped as greywacke and argillite) from
near Puketona to the western side of the Waiomio depressiony. Recent
discoveries by the Geological Survey have invested the line along the
south-western margin of the greywacke from Watomio to Whangarei
(where it joins the Harbour Fault of Ferrar, 1925, p. 18) with very
ereat tectonic importancet. The details of this work are eagerly
awaited, for it now seems that a line from North Cape to Whangarei
Harbour mouth may be expected to prove of great significance in North
Auckland structural interpretation.

Age of the intrusion and associated faulting.

Bartrum (1934), after incorporating the Whangakea Series of the
older maps of the North Cape area in the Upper Cretaceous Rahia Series,
pointed out that the North Cape gabbro and ultrabasic rocks must con-
sequently be regarded as having been intruded during the orogeny that
closed upper Cretaceous sedimentation in that area. He correlated them
with the supposedly early Tertiary serpentine intrusions in the Silverdale
district near Auckland.

It does not now seem necessary to suppose that the intrusion was
connected with the folding movements that closed the Cretaceous and
and early Tertiary sedimentation ; rather seems it to have been connected
with a quite different and later scheme of movements. ‘When more 1s
known of the age of the rocks of the Kaeo Series invaded by gabbros
and diorites near Whangaroa Harbour it will be possible to give a more
satisfactory maximum age for the intrusion and faulting. As for its
minimum age, there are intrusions in the lowest division of the Tertiary
beds near North Cape that might not unreasonably’ be supposed to be a
product of the North Cape intrusive episode, though Bartrum and
Turner who described them (1928, p. 127) did not regard them as such.
If they are connected with the North Cape mass the period of intrusion
and faulting is at least as late as lower Miocene.

= There are similar rocks between Tupo Bay and Taupo Bay north-west of the
entrance to Whangaroa Harbour and five miles north-east of the strike line
through North Cape, Cape Karikari and Haunga, which may be genetically
connected with the same intrusive system.

+ A narrow tongue of basement rock crosses this line on the northern side of Nga-
pipito Valley owing to the upthrow on the northern side of the Kawakawa
Fault.

~ Personal communication from Mr. R. Hay.

50 JATTEY.

All this remains speculative at present. All that can be said with
certainty 1s that the whole episode of intrusion and faulting was post-
Cretaceous.

..—DRIFTS.
(a) Klevated marine sands.

A flat bench some 250 yards broad, a trifle under 100 feet above
sea level and sloping inland slightly, fronts the coast between Waiari
and Brodie’s Creek. Subaerial erosion by a swampy streamlet that runs
south-westward along its inland margin and collects the run-off from
the high ground to the north has caused it to decline towards Waiari.
The same surface may be seen around the Hall at Whatuwhjwh} and
again north of the school-house, where it is 120 feet above sea level.
Farther west it is represented in the flat tops of the spurs between Patia
Beach and Perehipe (135 feet) and between Perehipe and Tokerau
Beach (150 feet). Thence it skirts the western end of the Te Ari Ridge
above the Tokerau Beach-Merita Road. In all these places it is under-
lain by sands which, to judge from the cuttings along the road, are about
70 feet thick on the spur between Tokerau Beach and Perehipe, though
they are much thinner to the east, for only about 10 feet of them can
be seen from the shore in the edge of the bench north-east of Waiari.

The schoolmaster’s house at Whatuwhiwhi starids on a remnant of
a subordinate flat bench 50 feet above sea level.

Sands cover the surface around the eastern end of the Te Ari
Ridge, between Brodie’s Creek and Merita, to a height of 195 feet, and
they sweep up the northern flank of the ridge to about the same height
along its whole length. The tops of the spurs are fairly flat for 1,500
yards south of Merita village—that is, to just below the 200-foot contour
—where the ascent of the main east-and-west ridge begins and the sandy
soil gives place to clay, but the break of slope is not very sharp and the
upper limit of the sands can be fixed only roughly.

East and north-east of Merita the sands probably extend to about
the 200-foot contour also (plate 3, fig. 1). The neck of land between
Matai Bay and Karikari Bay is composed of them and they lap on to the
slopes of the high ground along the eastern side of Cape Karikari.

Behind the little bay on the western side of the Cape, two terraces
are well developed; the more prominent one is a little more than 100 feet
above sea level, backed by low rolling downs to 200 feet, and the other
is at about half this height. These two surfaces accord with those
developed around Whatuwhiwhi. The 100-foot level is also represented
above the cliffs of consolidated sand at the eastern end of Karikari
Beach.

The sands vary in thickness. At the eastern end of Karikari Beach
they extend to sea level, or below, while between the road and Matai
Bay they are exposed almost to sea level and enclose well-preserved logs
of wood and stumps of small trees. Northward, however, the sand
capping the bluffs at the back of the northern half of Matai Bay can be
seen feathering out against the slopes of the hill at the northern head of
the bay. .

Geology of Rangtawhia Peninsula. 57

It seems from the gradual decrease in the height of the coastal
bench east of Tokerau Beach that a shght tilting diavennn ards to the east
accompamed, or followed, the change of sea- level. To explain the facts
that the sands extend higher on the northern face of the Te Art Ridge
than on the south and that there is no definite break of slope at their
upper linit, it inmay he suggested that more sand was supplied to the
north-western coast and that wind-borne sand was swept up the shore
above tide-mark there, whereas the well-marked bench on ihe southern
and south-eastern coast shows the true level of the strand, as does the
sand surface deposited just off the old shore around the eastern end of
Karikart Beach and on the western side of Cape Karikari. The rolling
surface, rising te 200 fect between the upper bench in this last area and

the higher ground to the cast, may represent old sand dunes behind the
100- foot shore.

The “L00-foot™ level* h las not previously been recorded in the Far
north (Henderson, 1924, pp. 582-3) but is well known along other parts
of the New Zealand coast ees the Auckland district southward and
belongs to a period of erosion called by Henderson the Awakino Cycle
(1924. p. 989). The 50-foot level represents a later, less prolonged
period of steady sea level and may, perhaps, be linked with an erosion
level of about this height near Whangarei (ibid., p. 582).

(b) Sand dunes and (c) Alhiviume.

The shifting and partly fixed sand dunes that form belts along the
shores of the tombolo joining Rangiawhia Block to the mainland, the
hxed dunes of its central part and the swamps impounded by the active
dunes against the fixed ones will not be considered here.

On Rangiawlia Block itself small foredunes lie behind the beaches
at Perehipe, P atia, \Whakarara and the little bay west of Cape Karikari.
A small all hivial bay-head filling lies behind the dunes of Patia and
Whakarara, joining the headland at Patia Pa to the land on the north,
anda sinall area of alluvium occupies the valley behind the dunes at
Cape WKarikari. These flats form an insignificant proportion of the land
surface but are important in the farming economy of the Maori popu-
lation,

SUMMARY.

The following conclusions can be drawn from the geological map-
ping of Rangiaw hia Peninsula to date:

lgneous rocks of the spilitie suite, including basic pillow lavas and
‘O
keratophyres, were poured out on the sea floor, and partly intruded into
ph}

the muds and muddy sandstones accumulating there, probably in late
Palaeozoic or early Mesozoic times. Such eruptions. it seems, took
place over a wide area, the sequence of events at the Three King's
Islands being closely comparable with that at Rangiawhia.

* Probably a little more than 100 feet on the average.

58 BATTEY.

Little is known of the structure of these rocks, or of the earth-
movements that followed their formation, but an east-and-west strike
appears to prevail in them now. If the basic pillow lavas be accepted as
of one horizon, their distribution offers some hope for the future elucida-
tion of this phase. Verhaps granodiorites were intruded then. Whether
they are of this age or older, they were exposed by erosion of the land
that existed after the movements.

Succeeding submergence of a steep coastline crossing the area led
to the formation of a great boulder beach over which grits and sand-
stones and probably a pile of other sediments were laid down as the
waters deepened. This submergence possibly took place in Cretaceous
times.

Violent folding about east-and-west axes followed this sedimenta-
tion and bent the strata to a vertical attitude. After the folding—
perhaps as a later phase of the same movement, or long after—the crust
broke along a fracture system running east-north-east along the south-
eastern margin of the present Peninsula and the area to the south was
displaced relatively to that on the north, so that shales of some lower
formation were brought against the boulder beds and sandstones. The
failure to find fossils to establish the age of the formations involved in
this phase restricts severely the interpretation of the geological history,

Later still, in post-Cretaceous times, that part of the crust lying
north-east of the area was wrenched north-westward for a distance of
some miles, part of the displacement being accounted for by movements
along a system of closely-spaced tear faults that runs north-westward
across the present Peninsula and probably extends a long way to the
north-west, through North Cape and to the south-cast past Whangaroa,
beyond which it may link with an important line of displacement that
continues to Whangare1 Harbour.

At the same time a long belt of intrusive basic and intermediate
igneous rocks with numerous small offshoots was injected along the
plane of movement over a distance of some 70 miles.

Subsequent small movements of sea level in comparatively recent
times are recorded by elevated marine terraces and sands.

ACKNOWLEDGMENTS.

[ am most grateful to Mr. and Mrs. D. G. Forsyth, now of Kaitaia,
for the generous hospitality which they bestowed on me at Whatu-
whiwhi during two field excursions. My thanks are also extended to
Mr. R, A. Johnston, of Kaitaia, for his kindness in placing his cottage
at Merita Bay at my disposal. Many courtestes were proffered to me
also by the residents of the Peninsula.

This paper represents part of the results of work on Rangiawhia
Peninsula which is being carried out with the aid of a grant from the
Hutton Memorial Fund of the Royal Society of New Zealand.

Geology of Rangiawhia Peninsula. 59

REFERENCES.

tes a F M., 1942, The dynamics of faulting. Edinburgh, Oliver and
poya,

BARTRUM, J. A., 1934. The pillow-lavas and associated rocks of the North Cape
area, New Zealand. N.Z. Journ. Sci. & Tech., vol. 16, pp. 158-159.

2ARTRUM, J. A., 1936a. Spilitic rocks in New Zealand. Geol. May, vol. 73,
pp. 414-423.

BARTRUM, J. A., 1936b. Notes on the geology of the Three Kings and other
outlying islands of northern New Zealand. 2. Journ. Sci. & Tech,
vol. 18, pp. 520-530. |

BARTRUM, J. A., 1948. Report on rocks collected by Mr. G. .\. Buddle from
islands of the Three Kings Group. Ree. duck, Inst. Mus., vol. 3,
pp. 205-206.

BARTRUM, J. A.. and TURNER, F. J., 1928. Pillow-lavas, peridotites and
associated rocks of northernmost New Zealand. Trans. N.Z. /nst.,
vol. 59 (1929), pp. 98-138.

BELL, J. M., and CLARKE, E. de C., 1909. The geology of the Whangaroa
Subdivision, Hokianga Division. N.Z. Geol. Surv., Bull. No. 8 (n.s.).

BELL, J. M., and CLARKE, F. de C.. 1910. A eeological reconnaissance of north-
ernmost New Zealand. Trans. N.Z. Inst., vol. 42 (1909), pp. 613-624.

BRIENSON. W. N., 1924. The structural features of the margin ot Australasia.
Trans. N.Z. Inst., vol. 55, pp. 99-137.

COX. A. H., 1915. The geology of the district between Abereiddy and Abercastle
(Pembrokeshire). O.J.G.S., vol. 71, pp. 273-342.

FERRAR. H. T.. and others, 1925. The geology of the Whangarei-Bay of Islands
Subdivision, Kaipara Division, N.Z. Geol, Surv., Bull. No. 27 (n.s.).

HECTOR, J., 1872. Report on the coal seams at Wangaroa and Mongomu, Auck-
land. Repts. Geol.’ Explor. during 1871-72 (No, 7), pp. 153-158.

HECTOR, J., 1891. Progress Report. /epts. Geol, k.xvplor. during 1890-91

(No. 21), p. \Xxxi.

HTLENDERSON, J., 1924. The post-Tertiary history of New Zealand. Trans,
N.Z. Inst., vol. 55, pp. 580-599.

KENNEDY, W. Q., 1946. The Great Glen Fault. OU .GS., 102, pp. 41-72.

McKAY. A,, 1894. On the geology of Hokianga and Mongonui Counties, Northern
Auckland. Repts. Geol. Explor. during 1892-93 (No. 22), pp. 70-90.

“M4ACPHERSON, E. O., 1946. 9 An outline of late Cretaceous and Tertiary
diastrophism in New Zealand. N.Z. Dept. Sci. & Industr. Res.,
Geological Memoir, No. 6.

MARSHALL, P., 1908. Geology of centre and north of North Island. 7 rass.
N.Z. Inst., vol. 40 (1907), pp. 79-98.

MASON, A. P., 1948. The geology of the central portion of Hokianga County.
Thesis, Auckland University College Library.

MORGAN, P. G., 1919. The limestone and phosphate resources of New Zealand.
Part I, Limestone. N.Z. Geol, Surv. Bull, No, 22 (its. ).

SUESS. E., 1906-9. The Face of the Earth, vols, HI, TV (English translation).
Oxford, The Clarendon Press.

’ ‘
e - 7 3 " © e* - = e -
; ae oi ‘ if hd i
Wray 7 x a
E < “ i : r : F \ ‘ Fr
. “ * 3 *
é : E
i ‘ it
= 2 2 . *
: iJ tie on ~ :
- . ry ° * : .
+ ” yf ‘ > ‘
> fi , » ¥
‘ . ; |
’ ; Re se aa we * ~
a z i , 4 ca
3
7 é 7 2 Pe : ‘
? * os e » é .
> Pie “ad s ‘J
r, :+- ' ‘
Fi n ; a F
i : : r ats > . | <
> : ‘ 1 + ~: Ar
s . eo Sack : y
a » s
ae ¥ Je ' F .
va? ' : bs r
t ’ 7 . > oa
> F . a : :
> és ‘< :
2 - i :
a ode ~ 4 * ‘4 Ps Cf
A a - ; k : A 7 o 3
= : ‘ r ;
KX ; x ae - ;
4 ” . .
, a> ful \ .
nae. : : A z z 1
* : ’ #,-z x
¢ : : F
: 7 b, Pa.”
. . .
‘
: , : Ay
‘
F '
’ ~ F a
reas A : ; r :
Ae: '
> ' é : : ; . “2 ‘
-¢ be * ih .
7 ie :
: 2 4 ®
Aad ce ; :
- F eX
¢ ; : a q
P '
> Ao - ,
- h Mt ‘ - *
- a i 7
ae y . . ” ay '
ayn 4 A 2 4 f ; 4 ‘ ¥
7 E « se : "
* ¥ J - * : *
7 . " re
} . es > a 3 ,
co oe ’ - ¥
. , AH : ‘ on =}
Mi _ : ‘ F +
€ : .
Fa . ’, .
« “

PEA Te et,

Gig, 1. Pillow lavas 600 yards SSW. of Brodie’s Creek. The formation dips
66° to the left (S. by W.).

lig. 2. Closer view of pillow lavas SSW. of Brodie’s Creek. There is hardly any

n the pillows here: the iiterstitial rock 1s

sedimentary material betwee
c)

glassy lava.

“

”

Pages.

Big: 2.

PLATE 2.

‘ine conglomerates, grits and sandstones at Waiari. Shales outcrop at
cxtreme low tide mark at the upper left corner of the photograph.

Breccia-conglomerates overturned and cut by a slightly transgressive sill
236 yards NE. of Homarere. Shales are faulted against them 18 yards
to left of bluff.

Boulders of graphic granodiorite weathered from conglomerate 100 yards
N. of headland at Waiari. Beds containing them are 150 feet below those
: sale

of £12 1.

rea Ths

”

Tt 5
J ¢ phate pee
= oh
~ + a"
= a 4
. = “a; Sham Sas
as = aah
+ : te
. waren ee Fin
"s ee TT) E a
- ~s :
- ~* ami ox
. ca
a eee eee aa
On be wr et ee
i = . t
«
o. e-te-F i. a
P. " 4
- - '
‘ =
P ai
& +
eae ta «
“
‘
a .
i ’
4
‘
’

‘

PLATE 3.

lag. 1. Matai Bay looking S.

' NW. grain in Maitai Pa (centre) due to dyke
intrusion and shearing. 200 feet sand surface behind right end of Pa.

rig. Z. Black shales at Homarere. Hammer rests on sill.

Keratophyvres in cliffs.
Between the two is a patch of breccia beyond right edge of photo. Flat
top of cliffs 1s 100 feet marine terrace.

Fig. 3. Dyke cutting black shales 550 yards NE. of Homarere.

emibinep eel

*
o

emibinep eel

*
o

PLATE 4.

GEOLOGICAL MAP

OF PART OF THE

COAST EAST OF WHATUWHIWHI /.
RANGIAWHIA PEN“, DOUBTLESS BAY~.. ._ :

oe ooo z6 pi49

Scale
oO 100 200 300 yards
———S—————

EXPLANATION

Lavas (chiefly keratophyres) 4" Dip — angle oe
us Breccia, conglomerate and in degrees 57
we sandstone (lines indicate strike) Vertical
oes Shales (lines indicate strike) AAA strata ~~ 2
a Marine sands (elevated) (| Approximate Low 3
Dykes and Sills ~rl = water mark 5

Geological boundaries ca) observed ae High water mark
(b) inferred ---7~ a
Faults (ay) observed —~~ _ Gravelly beaches .“
(b) inferred —~

CE:

Moturoa ls. _

\
|

Coast not
examined

Marae white

GEOLOGICAL MAP OF PART OF THE

Explanation
Alluvium
Sand dunes
Elevated sands
Gabbro, dtorite etc,
Dykes
Breccia, conglomerate € sandstone
Shale
Feratophyre
Basic pillow lava
Argillite $ greywacke

Roads § tracks |.-*

GY
~~
< .
>
O
oQ
So
ry
ra)
E

Drifts

Cafe Karikari

Coast not
examined

iy dee %, yocale

1000 Sco oO
Pihak

ic00

oa Foint

z.
“Oe ra

RANGIAWHIA PENINSULA, DOUBTLESS BAY

sta cea
3 04 ex
Sa
& [KAA]
ars
3 EZA

2 fii
te

Bice

Contour interval tooft.
1 Mites
2z0co Yards

SS 5> Knuckle

A Point

Coast not
examined

Coast not examined

See detailed map

‘¢ ALV Td

vibe

Gt. sey
Ut

2+ sce.

Life History of Austrosuccinea archeyi,
an Annual Snail, and its Value as a
Post-Glacial Climatic Indicator.

By A. W. BL POWELL, Auckland Museum.

Abstract.

A native snail, lustrosuccinea archey?, first discovered in consolidated sand
ot post-Pleistocene age at Doubtless Ray, Northland, is now shown to he a still
living species on coastal dunes, with a range from Spirits Bay to Mount Maunga-
nui and as a fossil to as far south as Cape Kidnappers. This snail is found to have
en annual evcle which correlates with rainfall. The extremely specialised habitat
and narrow tolerance of these srails makes the forsil occurrences useful indicators
Got past xerophytic phases in respect to post-Pleistccene climate.

In 1933 I deseribed a new species of land snail (Suecined archeyt)
obtained from post-Pleistocene consolidated sands associated with “moa”
remains (Euryvapteryx geranoides and curtis) at Tokerau Beach, Doubt-
less Bay. The New Zealand snail is closely allied to the Recent South
Australian Succinea australis Ferussac, 1821, which species Iredale
(1937, p. 307) made the type of his genus Austrosuccinea.

In June, 1947, Mr, C. W. Devonshire, then of Lake Olna Native
School, reported that living Austrosuccinea were abundant during winter
and spring just behind the fore-dune at the “planks,” road access to
Tokerau Beach, but that living exainples entirely disappeared during
the summer months. A personal visit to the locality in June, 1947 con-
firmed the presence of living Austrosuccmea,

ACKNOWLEDGMENTS.

Tam greatly indebted to Mr. C. W. Devonshire for his original
report and subsequent observations, to Mr. A. Hancox for monthly
records and growth series taken over a period of from two to three years,
and to Mr. D. G. Forsyth for similar observations whilst he was
stationed near the locality. For useful data relating to other occurrences
| am indebted to Messrs. J, D. H. Buchanan, of Havelock North, k. K.
Dell, Dominion Museum, Wellington, and G. Williams, Mt. Maunganui.
| The rainfall records were eenerously made available by Dr. M. A.
I’. Barnett, director of Meteor ological Services, Air Department, aes
ington, and the botanical determinations by Mrs. Allen (nee Miss L.

Molesworth), Miss U. V. Dellow and Mr. Robert C. Cooper.

The Generic Position of archeyi

Iredale s proposition of Austrosuccinea (1937, p, 307) is seat
primarily upon Quick's (1933, p. 312) statement one Vieoran spec
mens (australis) resemble arenaria in jaw, radula and genitaha. Since
then Boettger, 1939, proposed Quickella for the English-European
arenaria, and Quick (1936, p. 42) has claimed considerable resemblance

62 POWELL.

between the South African striata and australis, special reference being
made to the characters in common of less than twice as many marginal
as lateral teeth and the large median pro jection on the jaw. In this later
paper Quick disassociated australis and arenaria on the grounds that a
penis sheath is present in the former but absent in the latter.

. Pilsbr) (1948, pp. 771-847) recognised the genera Ouaylonia,
Suceined and Quickella, but reduced Austrosuccinca to a synonym of
Succined on the grounds that the few diagnostic remarks made by Iredale
were misleading. Nevertheless, there seems to he a eood case for the
recognition of Austrosuccinea in the light of Quick's (1936) remarks.

sexl figure Al. clustrosuccinea archeyi. a, Jaw; b. radula, showing central
tooth, first and tenth laterals and number 17, a marginal. c, shows the
mantle markings of an animal removed from its shell.

sununarised, the differentiating criteria for the several Suecinid
genera would appear to be as follows:

Penis provided with a sheatn
Penis with a narrow appendix
Jaw with weak median projection only
Marginal teeth with long tapered bases
our to five times as many marginals as laterals 2. Owylonu

Life History of Austrosuccinea. 63

Penis without an appendix
Jaw with strong median projection and (usually)
lateral folds
Marginal teeth with long bases
Twice as many marginals as laterale 205 reas ye OHCCIBCa

Jaw with strong median projection only
Marginal teeth with broad shallow bases
One and a half as many marginals as laterals Austrosuccinea

Penis without a sheath
Jaw with strong median projection only
Marginal teeth with broad shallow bases
Marginals and laterals few and of equal number .. Quickella

I have no preserved material of australis, but from Quick's excellent
account of that species (1936, pp. 36-39) the New Zealand arche
appears to be very similar.

The radula of archeyi shows a slight difference from that of
australis in that the cones are shorter. The radula formula 15 -+ 10 +
1+ 10 + 15 & 90 is almost identical with that of @ustralis, which
according to Ouick is 16 + 10+ 1+ 10 + 16 X 90. It appears also
that australis is not restricted to a sand-dune habitat, but occurs inland
cso.

Ouick (1936, p. 37) described very different mantle markings for
australis from those found in archeyt. He described the mantle in
australis as dark and opaque with a faint indication of mottling at the
periphery and with a few yellowish-white chalky spots. In archeyi there
1S a sparse pattern of several dark intermittent narrow axial streaks
with a large dark rectangular patch near the lower front margin of the
mantle (7 ext fig, Ac.).

It is of interest also that Quick (1933, p. 310) records the English
arenaria as nee on damp circular depressions in sand dunes and that
Pilsbry (1948, p. 843) describes the habitat for the eastern North
American Pee ere of this genus (Quickella) as living in small thickets
in the sandy shore zone.

That an annual cycle is not peculiar to the New Zealand species 1s
shown by a note under Oayloma decampi gouldi Pilsbry (1948, p. 782),
a New England Succinea that frequents the aquatic vegetation of muddy
ponds, river margins and ditches. “Probably most of the large indi-
viduals die by the end of summer, as I have often found only hali-grown
shells in autunin where large ones were found earlier in the season.”

Habitat

The habitat of Austrosuccinea archeyi is extremely specialised, for
the species occurs on a substratum of fine textured loose sand (fine
quartz-sand at most places) only on or in the vicinity of the first and
second dunes, parallel to the beach, usually in a narrow belt not more
than 100-200 yards wide, and only where the original rather sparse native
plant cover remains intact, That is the ‘“Sand-grass Dune Community”
to the “Scrub Dune Community” of Cockayne (1928, pp. 92-93).

O44. POWELL.

The natural shade or shelter plants of the Austrosuccinca habitat
are Cassia retorta A. Cunn., Spinifex hirsutus Labill. and more uncom-
monly Coprosma acerosa A. Cunn. and Muehlenbeckia com plexa Meissn.
Two introduced plants add to the cover in certain areas (Taipa Beach) :
iupin, Lupinus arboreus Linn. and a Mediterranean erass, hare’s tail,
Lagurus ovatus Linn,

The food plants essential to the snails are a blue-green alga
Anabaena variabilis Kutz (Cyanophyceac), which is available only dur-
ing the wet months, and the outer tissue and finer roots of the Spinifer
during the dry months, when the Anabaena dries and disappears.

Observations by Mr. Hancox show that 4nabaena is in a lush state
from May to August, is more or less dessicated during September to
early October, dries up and is dispersed by winds from October to March
and new growth appears again with the rains during March or April.

LIFE HISTORY

Food, or its availability. which is governed by moisture, controls
the growth and life span of Austrosuccinea archeyi. These snails hatch
irom between June and August and reach maturity, just under 13 mm.
in height, within twelve months, all adults dying between August and
October of the year following their birth.

Young snails tide over the dry months of late spring and summer
by partial aestivation, during which they keep alive with difficultvy—but
add little to their growth. They lie dormant in the sand around the roots
of Spinifer and sealed by an epiphragm during dry periods, but will
emerge and feed during rains. The alga has by this time disappeared,
and until the new growth appears in March or April the young snails
have only the food afforded by the Spinifea. Since hatchings occur
over a period of two months in the spring and heavy mortality results
from abnormal dry periods the size ranges vary from year to year for
this reason. Fully grown living adults, however, can only be obtained
from July to the end of September.

A prolonged dry summer results in a very late commencement of
adult growth, but under such conditions growth seems to be accelerated
when the rains commence and full adult size obtains before the end of
August, the regular time for the dying off of the adult population, Jt
was noticed that even during an abnormally wet spring the adult mor-
tality took place in spite of the fact that the algal food was still in good
supply. The life cycle of the snail is thus shown to be geared to the
average conditions,

| have not yet managed to visit any of the colonies of the snails
during the breeding months, but Mr. A. Hancox noted copulation and
eggs on 28th April, 1949, and Mr. C. W. Devonshire recorded the
deposition of eggs in clusters of up to fifteen on 16th July, 1947. He
described the individual eggs as spherical and approximately 1 mm, in
diameter and that they were attached to the thallus of the alga.

An egg associated with snails preserved in alcohol and collected by
Mr. Hancox on 28th April, 1949, had a diameter of 1.75 mm. It was
spherical with a large yolk which was slightly yellowish with a fine
irregular network of faint lines, not reticulated. The yolk was sur-
rounded evenly by a covering of clear jelly.

‘9 ife / listo r\! of A lf SHrOSUccuned. 69

Monthly size log for Austrosuccinea

smallest Average Largest Locality Remarks
Jan. 2, 1949 2.50 5.00 6.75 To.
20), 1948 4.00. 6.00 8.50 To.
Feb. 12, 1949 245 * 5.00 7.50 To.
Mch. 21, 1949 6.50 7.25 8.30 S: —
April 25, 1950 6.50 &.00 9.00 hier
28, 1949 6.00 9.50 10.50 To.
May 25, 1950 5,50 §.50) 9.25 (iF
27, 1949 7.00 10.50 11.75 Ta
27, 1949 7.25 9.00 10.75 To.
June 6, 1948 7.50) 10.00 11.50 To.
10, 1949 Tid 9.50 10.30 To. (40% dead)
19, 1947 7.50 5.00 10.50 To.
: 25, 1950 7.20 &.75 9.00 To.
July 14, 1949 7.10 8.50) 1T75 To. (80% dead)
Aug. 2, 1947 8.25 11.00 1225 To. (maximum )*
3, 1948 7.50 10.00 10.75 To.
25, 1948 6.75 10.00 11.50 To.
Sept. 19, 1948 8.50 9.00 10,50 To. (mostly dead)
Oct. — —- —- -— (all adults dead)
Noy. 3, 1948 2.00 5.50 7.75 To. (new generation)
Dec. 9, 1949 2.50 4.50 7.00 La
12, 1948 3.50 5.50 7.09 To.

(To. = Tokerau Beach, Ta. = Taipa Beach, S. = Spirits Bay,
+ = abnormally wet season. )

Maximum sized dead example 13.00 mm. (To. 3/8/1948).

*Maximum sized living example, 12.25 mm.

Although the above records are rather intermittent the rapid winter
erowth curve, which corresponds with the availability of Anabaena, the
adult mortality period from August to the end of September and the
appearance of the new generation in November, are all very clearly
shown,

The monthly size ranges show variation from year to year, but this
is resultant from a variable rainfall, year to year, as shown by the follow-
ing table and in text figure B.

Rainfall in inches at Mangonui
Jan. Feb. Mar. Apl. May June July Aug. Sept. Oct. Nov. Dec. Annual

1°47 .94 65 2.78 7.89 527 7.61 10.46 5.21 612 3.91 6.09 6.20 63.13
1948 340 187 S770. 5.87 S00. S88. 12:08 B86 5.05 B21 421 3.62 67.46
1949 Oo “S.0S S.c2> 4:93. 828 FI B25 5.01 3,60 “Sislo 62) 2A R72
Average

64 years 3.72 3.39 3.57 464 5.66 579 5.92 5.76 4.57 3.96 3.28 2.98 53.24

These records were kindly supplied by Dr. M. A. F, Barnett,
Director of Meteorological Services, Air Department, Meteorological
Branch, Wellington.

It was remarked that over the period of 64 years at Mangonui,
a rainfall of 12.08 inches (July, 1948) has only been exceeded once in
July, and that was in July, 1946.

POWELL.

D430 AON 190 43S DAW Jnr Nar

75! ete sia ja tM

avad sLindgW WW

SYALAWITTIW NI STIVNS

qvadSLindy %08

‘pourluexea spdures pojep YIVI IOF 3zZ1s aSeIIAe DY} eq SsosO 94} Pue ApIAadsoI
Sozts TUNUWIXeI PUR LUINUITTUT 94} JWosotdet syuiod MOITE YIM saul] [eoysA sy TL ‘DUIDDU
Bole oY} fo aduasqe JO soURpUNqe sAe[aI 9Yy} pue MOTaq UMOYsS st Suse] 88a yo porsad ou
‘steak $Q AO} OAIND [[P}UIeT ISeIOAP OY} O} UOT}LJ9I UI UMOYs sIPaA JURADTOI au IO} [peyures JSUIESE
pa}jojd sievak OM} JOAO SUOl}eIaUas PautDNsoIsny JO YMOIS 9y} Suljuasoidos ydescy ‘7 aan 4x07

%o OF

qwad SLINAY

Ss)

676!

on, 4 Pt Pay nt

=) 7 TE SS NARA OTS NS

6+6]

QvVaqg SLINGAY 1

Srei

WW ddW YWW 834 NWF 33d AON 490 daS ONY Int Nar AWW wd¥ WW 834 NWP

‘ech Ps, ea

sae Seo . Se ee a a ee
6—_;;;,6
$O903

Lv6I Crs!

YNAVGUNY

<A ueer” an oie
hie -\
LST TTA Socks

-
-
ete eee

avadq SLINAy LSOW

SAHINI NI T1IWSNIW¥ 2\

Life History of clustrosnecined. G7

Living Colonies of Austrosuccinea

Spirits Bay, 1-35 miles S.W. from Kapowairua, A, Hancox, R.A. and
H, S. Prouse and A.W.B.P., 21/3/1949. The alga (Anabaena)
was abundant and fully developed as in mid-winter, owing to an
unusually wet summer season. The maximum size was 8.30 mim.,
which 1s large for this time of the year but resultant from a phe-
nominally early growth of Anabaena. The colonies of snails, which
were quite prolific, were under the shelter of Coprosma accrosa
A. Cunn. and Muehlenbeckia complexa Meissn.

Vokerau Beach, Doubtless Bay, C. W. Devonshire, D. G. Forsyth, A.
Hancox, A. C. O'Connor and A.W.B.P., 1947-1950, Distributed
intermittently between the first and second coastal dunes over the
entire length of the beach. The most accessible colony is at the
“planixs” or road access to the beach. Very extensive fires during
the summer of 1950 have taken heavy toll of the colonies. The
shelter plants are Cassima retorta A. Cunn., Spinifexr hirsutus
Labill, and more uncommonly Coprosma acerosa A. Cunn, and
Mucehlenbeckia complera Meissn. This paper deals largely with
the material collected and observations made at this locality over
the period from June, 1947, to July, 1950.

Vaipa Beach, Doubtless Bay, A. Hancox and A.W.B.P., 1949-1950, A
flourishing colony occupies the fore-dune over the western half of
the beach. There is rather more shelter than in other more natural
locations owing to the addition of two vigorous introductions, the
hare’s tail grass, Lagurus ovatus Linn. and the lupin, Lupinus
arborceus Linn.

One nule north of Tauranga Kawau Point, between Whananaki and
Mimiwhangata, A.W. B.P., 27/10/1947, The remains of what had
been a living colony of snails a few months earlier occupied only a
few ES yards of the foredune. ‘The shelter plants were very
sparse, only Cassinia retorta and Spinifer hirsutus, the latter being
quite dead. The <lnabacna was still moist and fleshy where it was
under cover. Grazing sheep had so reduced the plant cover that
it is doubtful if this colony still exists.

Mount Maunganui, Bay of Plenty, in coastal dunes several miles east-
ward along the beach, Living examples were taken by Mr. G.
Wiiliams in 1947, but a recent search (.\ug., 1950) failed to locate

the species either alive or dead.

Extinet or subfossil colonies of Austrosuccinea

1. On rounded, almost detached headland one mile south of Cape Maria
van Diemen (mainland). R. Michie and A.W.B.P., 17/11/1948.
This headland has a thick cover of 15 to 20 feet of consolidated
wind-blown sand on a base of hard rock. At present most of the
top of the headland is in rough grass fringed with flax (Phornuumne)
and a stunted area of coastal Berub down the almost perpendicular
seaward face. On the landward side there is only drifting sand
and bare rock with no vestige of plant cover. Erosion of the con-
solidated sand in several places reveals fossil elusfrosucemnea in a

POWELL.

band one to ten inches below the surface, and scattered for some
feet deeper are fossils of Rhytida duplicata and a Placostylus
aucestral to a living new subspecies which occupies the scrub area
of the seaward cliff.

The sequence revealed by these occurrences suggests an carly
cover of coastal scrub with Placostylus and Rhvtida followed by a
dry period when the scrub disappeared from all but the seaward
cliff, and the headland was overwhelmed by blown sand. Then the
process of consolidation under more moist conditions, the develop-
ment of the Sand-grass Community allowing advent of Ausiro-
succinea and now the development of the Shrub-dune Community
modified by grazing stock. The now complete cover of the sand
on the headland excludes Austrosuccinea, which apparently only
exists under xerophytic conditions.

2. Ona small island directly south of the most northern headland block

at Cape Maria van Diemen mainland, A. Hancox and A.W.B.)..
18/11/1948. This island has been completely burned off and is now
covered by a rank growth of “spiny clover.”

Remains of Austrosuccinea and Placostylus ainmbagiosus
worthy: Powell are abundant in the surface layer of ash and humus.
The sequence here has heen from coastal scrub to a Sand-grass dune
Community, but the cover of blown sand is slight compared with
the first locality. Succession from the xerophytic phase has been
obscured by recent fires, but the charred remains of stunted Metro-
sideros and Coprosma indicate a normal development to the Sand-
scrub Community.

3. The “Placostylus ainbagiosus priscis block’ occupying the high dune

and “bad-lands” area lying about 1 mile to the south of the N.W.
Cape Maria van Diemen mainland headland and the Werahi Stream
and including the high dune of Herangi, 700 feet. The south-west
corner of this block is of consolidated light brownish sand and
produces fossil remains of Paryvphanta qwatti, Rhytida duplicata,
Serpho kitt an arboreal snail, and Liarea n. sp. as well as vast
numbers of Placosivlus ambagiosus priscus. The presence of the
former indicates a coutemporary condition of dense coastal rain
forest. The mode of occurrence of the snails at this locality (Bar-
trum & Turner, 1928, pl. 21, fig. 7) is exactly as those found living
at Unuwhao, 700-900 feet, between Spirits Bay and Tom Bowling
tay. Meandering lines of Placos/\lus as they occurred nestled in
along roots and under sistelia, Liarea in circular patches suggestive
of their association with Corer clumps, and the sporadic occurrence
of Paryphanta watt are all indicative of a natural forest community
suddenly overwhelmed by advancing sand dunes. That the under-

lying consolidated sand was capable of supporting a coastal rain

forest is thus indisputable, as also is the inference that the climate
must have been much more moist than now. Now the vegetation
in the block is restricted to a sparse covering of scrub with flax
(Phorminm) and toetoe (lrundo kakaho) only on the top of

‘a

Herangi, Here again the deterioration to xerophytic conditions is

Life [History of Austrosuceimea. GY

shown hy marginal occurrences of fossil Justrosuccinea. Burning
of the area by the former Maori occupants and subsequently by
settlers has undoubtedly hastened the destruction of the vegetation
and allowed mass movements of sand.

It is interesting to note that a map of the area prepared by
T. I<. Thompson in 1895 and now in the Lands and Survey Office,
Auckland, shows a covering of “fern, manuka and scrub” between
Herangi and the West Coast (1.¢., No. 1 locality). This traverses
the middle of the priscus colony where the heavy rain forest
formerly stood. Further, a photograph taken by W. H. Winkel-
mann in 1902 from Cape Maria van Diemen Island shows the whole
of this area to the upper slope of Herangi to be devoid of vegetation
as it 1s today.

North-western cnd of Tokeratt Beach, Doubtless Bay, G. Archey
and A.W,B.P., February, 1932. Type locality for the species,
which was found abundantly as a fossil with “moa” remains. Other
fossil land snails in the consolidated dunes, notably R/ytida dunniae
and Phenacharopa novoseclandica, indicate a former coastal forest
cover. Living Phenocharopa elsewhere in the north is found only
in the very damp innermost recesses of tall nikau (Rhopalostylis)
stands.

sand dunes 14 miles south of Neunguru, A.W.B.P., 1934. Bleached
sheils are encountered in loose sand drifts. A Shrub-dune area
xists nearer to the Ngunguru river and the alga Anabacna is

abundant, but no living Austrosuccinca were located,

]
4
i

Ocean Beach, Whangarei Heads, R. K. Dell, Jan., 1938, bleached
shelis in coastal dunes.

In consohdated sand at the back of the dunes, Oneroa, Waiheke
Island, Auckland. I<. Hipkins, 1946. The outcrop has since been
destroved by building activities.

Ocean Beach, five to sts miles south of Cape Kidnappers in fixed
dunes, about 109 yards from the beach, together with fossil R/ytida

spelace Powell: J. D. TL. Buchanan.

Age significance of the subfossil occurrences

It will be noted from the above list and chronological table (follow-

ing) that the same climatic sequence 1s shown m each locality. That
is. a former moist climate allowing of a coastal rain forest on beach flats,
followed by a presumed dry period when the coastal forests died and
allowed the formation of drifting dunes and the development of a Sand-
ETAss Community with ustrosuccinea and ultimately the elimination of
that species under natural conditions with the succession to the fully
developed Shrub-dune Conmnunity. Tn most cases, however, the advent
of grazing annals, fires and other human interference has intervened
between the development of these two communities.

7) POWELL.

The dating or correlation of the New Zealand periods of climatic
change with those of the mcre fully known European sequemce is con-
jectural.

Raeside (1948, pp, 153-171) dated the Canterbury warm forest
period from between the seventh and the fourteenth centuries A.D. He
also drew attention to the Report of the Committee on Glaciers for 1945
(Trans. Amer. Geophys. Union, 27, p. 219), in which it was postulated
that the causative climatic variations affected both hemispheres simul-
taneously and not in alternation and therefore it is reasonable to suppose
that the same pronounced Post-Pleistocene variations and the major
Pleistocene variations were also synchronous in the two henmuspheres.

One other line of evidence has a direct bearing upon the Northland
climatic sequence, and that is the occurrence in raised beaches of the
bivalve mollusc Anadara trapesia.

This is a gregarious species living partially buried on inter-tidal
mudflats. It occurs living in the subtropical waters of Queensland, New
South Wales and the Great Australian Bight. and as a fossil in South
Australia, Victoria, Tasmania and Northern New Zealand. Its present
restricted distribution 1s considered to have been due to refrigeration.
the habit of living partially exposed to the atmosphere making it an easy
victim to a sudden drop in temperature.

It is worthy of note that odd worn valves of this species occur In
slightly raised beds of coarse shelly shingle underlying the foredunes
both at Spirits Bay and at Tokerau Beach. These shingle beds must .
have been deposited during the first post-glacial period, when they
received their covering of wind-blown sand that later consolidated during
the warm, humid, second period, allowing of the growth of coastal rain-
forest. Crocker and Cotton (1946, pp. 64-82) discuss Australian
Anadara occurrences at some length and conclude, quoting Professor
David. that the last Australian cold phase was from three to ten thousand
years past and that the period of maximum aridity in South Australia
may have been as late as three thousand years ago.

I have just received a paper by W. F. Harris, “Climatic Relations
of Fossil and Recent Floras” (1950), pp. 53-65). in which the following
chronology is suggested:

Recent or Holocene

Period 1. First post-glacial period to about S000 B.C. (== Northern
Zoreal). Increasing warmth and comparative aridity.

Period Period of maximum warmth, 4500-500 B.C. Greater por-
tion of this period humid (= Atlantic), latter part dry (=

Sub-Boreal).

Period 3. Cooler and moister climates lasting to the present day.
Rainfall higher than now until a few centuries ago. (==

Sub-Atlantic. )

Life History of Austrasucetiea. 7 |

Apparent sequence for four Northland Austrosuccinea sites

| 2 a 4
Headlandlm. Pl. a priscus .W. headland Spirits Bay
S. of Cape M. block, 1 m. S. Cape M.v. D. dunes 1-43 m.

v. Diemen Cape M. v. (mainland ) S.W. from
Diemen. Kapowairua.
Sh-dune with Drifting sand Sh.-dune Se.-dune with Present.
Pu i. overwhelming A.
Sh.-dune at

Herangt,

== —_ —— —— —— So

Se.-dune with Sp.-dune with S¢.-dunewith Sg-dune with Period III:
A. Sh.-dune A, Sh.-dune A. Sh.-dune A. as now.
with PI. 1. with Pl. p. with Pl. « A.D. to Present:
more moist.

Coastal f. Rain f. with Coastal f. Sh.-dune - Period Lf: dry.

with Pl. 2. Ply p., ta. with Pl. W. Se.-dune with 4500-500 B.C.:
Res. a Le, Pl. 3 (dwart) warm humid,

coastal f.

Formation of older consolidated dunes on * * Period I:
Anadara in comparative
raised shelly aridity.

ROCK FOUNDATION beach under- -5000 B.C:
(Older volcanic Whangakea Series? lying dunes. increasing

Trias Jura) warmth.

Pleistocene.

( Abbreviations: Sg.-dune = Sand-Grass Dune Community; sh.-dune = Shrub-
Dune Community; f. = forest; A = Austrosuccinea; L = Liarea n.sp.; Pll =
a living new subspecies of Placostylus ambagiosus; Pl.2 = a fossil ancestral form
of Pl.1; Pl.3 = a new fossil subspecies or form of the ambagiosus spiritus group.
It is dwarfed and obviously lived under xerophytic conditions; Pa. = Paryphanta
watt: Plc. = Placostylus a. consobrinus; Pl. p. = a. priscus; Pl.s, = a. spiritus,;
Pi.w. = a. worthyi; R. = Rhytida duplicata and S. = Serpho kiv.)

Note.—The asterisks in the column for locality 4 denote the absence
of consolidated sands at this site. It is presumed that the spiritus
colonies were originally associated with consolidated sands, but these
have since broken down and become loose and drifting with the fossil
snails lying on or near the surface.

Study of the above table reveals an alternative possibility, and that
is the assumption that the Northland formerly rain-forested coastal
dunes were contemporary with the Canterbury warm forest period,
referred by Raeside to Period IT] between the seventh and fourtheenth
centuries A.D.

However, it is much more likely that these consolidated dunes had
their origin in the comparatively arid Period I and received their rain
forest cover during the succeeding warm humid period of 4500-1000
B.C., for the Northland sequence seems to indicate that the moist early
centuries of Period ITI had a rainfall only sufficient to support a
Shrub-dune Community.

NI
rae)

POWELL.

REFERENCHS.

PARTRUM, J. A., & TURNER, F. J., 1928. Pillow-Lavas, Peridotites and
Associated Rocks of Northernmost New Zealand. Trans. N.Z, J/nst.
59, pp. 98-138.

COCKAYNE, L., 1928. Vegetation of New Zealand, Die Vegetation der Erde
XIV, Leipzig, pp. 92-93.
CROCKER, R. L., & Cotton, B.C., 1946. Some Raised Beaches of the Lower

South-east of South Australia and their Significance. Trans. Moy.
Soc. S. Aust., vol. 70 (1), pp. 64-82.

HARRIS, W. F., 1950. Climate Relations of Fossil and Recent Floras. Tuatara,
vol. 3, No. 2, pp. 53-66.

IREDALE, 1937. A Basic List of the Land Mollusca of Australia. Aust. Zool.
8, p. 307.

PILSBRY, H. A., 1948. Land Mollusca of North America. Monog. No. 3, Acad.
Nat. Sci. Philad., vol. 2, pt. 2.

POWELL, A. W. B., 1933. Two New Land Snails from New Zealand, Proc.
Malac. Soc. Lond., vol. 20, pp. 191-194, text figs. 4 and 5.

POWELL, A. W. B., 1947. Distribution of Placostylus Land Snails in Northern-
most New Zealand. Rec. Auck. Inst. Mus., vol. 3 (3), pp. 173-188.

QUICK, H. E., 1933. The Anatomy of British Succineae. Proc. Malac. Soc.
Lond., vol. 20, pp. 295--318.

QUICK, H. E., 1936. The Anatomy of some African Succineae, and of Succinea
hungarica Hazay and S. australis Férussac for comparison. Ann.
Natal Mus., vol. 8, pp. 19-45.

RAESIDE, J. D., 1948. Some Post-Glacial Climatic Changes in Canterbury and
their effect on Soil Formation. Trans. Roy. Soc. N.Z., vol. 77, pp.
153-171.

PLATE 6.

Composite photograph of Taipa Beach, looking nortua-west. Austrosuccined

lives under Spinifer Cleft foreground), Pimelia (right foreground) and Cassia

(middle right margin), but only on and in the vicinity of the foredune. The scat-

tered dark patches on the sand are the dried-up remains of the alga Anabaena and

the pale spots the flower heads of the introduced hare’s tail grass, Lagurus ovatus.
20th January, 1950.

Mollusca from the Continental Shelf,
Eastern Otago.

By A, W. B. POWELL, Auckland Museum.

Abstract.

A hard-bottom fauna from off the Continental shelf of Eastern Otago is named
the Chlanys delicatula-usiiriton community, It is compared with a similar North
Vest American hard bottom community. The following new species of mollusea
are described: I’enustas blacki, Pachymelon (Palomelon) smithi, and Panopea
amithae; also a new Pliocene ancestor to the latter, Panapea qwariganiicd,

The material was obtained by Captain A. Black, of the trawler
“Tairoa,” in from fifty to seventy fathoms along the continental shelt
of Eastern Otago, from the Nuggets north to the mouth of the Waitaki
River.

I am much indebted to Captain Black for his excellent work in
collecting the material and to the late Mrs. J. G, Smith and Mr. Smuth
for the care they have taken with the preparation of the specimens, the
accoinpanying notes and the generous gift of type specimens and repre-
sentative series for the Museum collections.

An interesting outcome of study of this collection is the revealed
presence of an extensive hard bottom community occupying the outer
edge of the eastern Otago shelf. The dominants of this community,
Chlamys delicatula and [usitriton laudandum, were previously consid-
ered to rank amongst our very rare molluscs.

This community immediately recalls a northern hemisphere anala-
gous one in the Strongylocentrotus-Argobuccinum formation from the
vicinity of Puget Sound, Washington (Shelford and Towler, 1925),
The dominants there are the green sea-urchin Strongylocentrolus
drobachiensis and the mollusc Fusitriton oregonensis (i.e., Argobucci-
num of Shelford and Towler). Other dominants are a starfish, crab,
several barnacles, two scallops, Chlamys hindsti and C. hericius, and a
easteropod, Calliostoma costatum. Among the sub-dominants are the

mussel Modiolus modiolus and the brachiopod Terebrataha transversd.

The Washington community is stated by Shelford and Towler to
be best developed on rock bottom, but the same association of animals
occurs on shelly and even. hard-sand bottom. It ranges to a depth otf
at least 150 metres, but the upper limit is not clearly stated. Algae is
of little significance and occurs only in places.

Except that sea-urchins are minor inclusions in the Otago com-
munity there is an almost exact parallel between the faunal composition
of the New Zealand and Washington communities.

The Otago community, which I here designate the Chlamys delt-
catula-Fusitriton community, develops in from 50 to 70 fathoms on a
hard-sand, shelly or gravel bottom under the influence of strong tidal

7 | POWELL.

currents and is stenothermaliy confined to between 46° S, and 54°42’ S.
with its optimum development in the vicinity of 46° S. It would seem
that Fusitriton loses dominance quickly to the south, but does oecur
sporadically far to the north of the northern Recent limit for delicatula,
which is Otago. /usitriton laudanduim has been trawled in 40-50
fathoms off Cape Campbell, Marlborough, and |] have a specimen from
the Ninety Mile Beach, Northland. The furthest south record for
laudanduin is Auckland Islands (Cape Expedition).

At the southern extremity of the delicatula range the dominance
is shared with the venerid bivalve Tawera mawsoni, and this may be
termed the Chlamvs delicatula-Tawera community.

A more precise evaluation would be formation and association
respectively for these communities, but the term formation conflicts with
the geological usage of that term and causes confusion, especially in
paleoecological work, so the non-committal term community is preferred
tor the present.

The Chlamys delicatula-Fusitriton Community

a, Dominants

Chlamnys (Zygochlaniys) delicatula (Hutton), Pusitriton laudan-
dum Finlay, Astraea heliotrepiutm (Martyn), Alcithoe swainsoeni
Marwick, Venericardia purpirata Deshayes, Atrina selandica (Gray),
Modiolus areolatis (Gould), Arca novaecgelandiae Smith (Moll.), and
Neothyris lentieularis Desh.) (Brach.).

hb. Subdominants

Longimacitra elongata QO. & G., Lima zealandica Sowerby, Ostrea
charlottae Finlay, Glycymerts laticostata Q. & G.. Argobuccinum
fumdum (Dunker), Verconclla fairfieldae Powell, Alcithoe calva
Powell, Panopea smithae n. sp., Chlamys radiatus Hutton, C. celator
Finlay (Moll.), and Ferebratella sanguinea Leach (Brach.).

c. Secondary species

Tredalina mirabilis Finlay, Charoenia capax euclioides Finlay, V enus-
tas tigris (Martyn), I’. pellucida forsteriana Dell, ’. punctulata aimpla
Powell, F’. foveaurana Dell, ’. blacki n. sp., lV’. cunningham pagoda
Oliver*, Cardita aoteana Finlay, Wonta gelandica (Gray), Pachymelon
(Palomelon) smithi n. sp.. Austrofusus glans agrestior Finlay, Maori-
colpus roseus QO. & G., Maoricrypta (Zeacrypta) monoxryla (Lesson),
Cominella (Eucominia) otakauica Powell, and Xenophalium (Xenoga-
lec) fintavi Tredale* (Moll. ).

Mr. J. G, Smith mentioned the occasional presence of sea-urchins
in the trawls, but I have not seen material. From other dredgings in
the vicinity I have Pseudechinus huttoni Benham and Gontocidar?s
umbraculum Hutton.

—

* These two species are uncommon 1 the collection and probably belong to a more
shallow fine textured sandy bottom. Finlay recorded both from 20 to 30
fathoms off Otago Heads and off Waikouaiti.

Mollusca, Eastern Otago 75
Owing to the large mesh of the commercial trawls there is an almost
complete absence of the smaller and micro species in the collection,
However, several bottom samples in from 50-70 fathoms off Otago
Heads, obtained by a naturalists’ dredge, reveal a rich microfauna and
the following quite abundant small molluses: Micrelenchius sanguincus
caelatus (Hutton), Thoristella chathamensis benthicola Finlay, and
Myadora novaeselandiac Siuith.

Chiamys radiata (Hutton)

Five examples in the collection are large and three of them match
Stewart Island topotypes in every respect. That is. they are finely and
evenly scaly-ribbed and are of either wniform reddish-purple or uni-
formly orange colour. These are the only colours encountered at Stewart
Island. The remaining two Otago specimens have slightly coarser and
more sparse ribs and a diffused pattern of reddish-brown on a buff to
pinkish-white ground.

I have not seen typical rudiata from north of Otago. Cook Strait
examples are mostly the smaller vari-coloured more spiny and sparsely
ribbed gemmulata. A small vari-coloured radiata is found at Cook
Strait, but it never reaches the size of adult topotypes. Hybridization
between radiata and gemmnulata seems to occur. However, a form of
equal size to fully adult topotypes of radiata occurs ca, 70 fathoms off
Kapiti Island. These were dredged in great numbers by the R.R.5.
Discovery IT, but only odd valves were taken and it may be that these
shells are from a fossil deposit marking a cool phase in the post-
Pleistocene, contemporary with the Cape Campbell, 70 fathoms deposits
that yielded Chlamys dehcaiula and Eucominia marlboroughensts
(Powell, 1946, Rec. Auck. Inst. Mus. 3 (2), p. 144).

The Kapiti shells, however, are much more finely ribbed than any
other radiata form and merit ‘subspecific designation. The following
table is based upon a radial rib count per centimeter at a point 30 mm.
from the umbo in each case,

Lefivalve. Right valve.

radiata (Topotypes. Steward Island) 20-24 18422
, Otago Heads (typical) 20-23 15-20
Otago Heads (gemmulata pattern) 19-20) 16-16

Kai lwi (Up Pliocene) 21-25 20-23
radiata N. subsp.2 (Kapiti Island) 26-35 22-28
genuniuala (Topotypes? Cook Strait) 14-23 14-20
consociata Hauraki Gulf, 20 fath, 8-24 16-20

Huaraki Gulf examples are vari-coloured and small. They are
covered by Smith's name consociata and may be regarded as a northern
subspecies of radiata rather than a form of the coarser sculptured
geminulata.

My suggested arrangement of these shells is as follows:

radiata: Forsterian, with its normal limit North Otago; Upper

Pliocene but never reaching the maximum adult size of Recent
shells.

radiata consociata: Hauraki Gulf and Northland.
gemmulala: Cook Strait and as an influence to the south.

76 POWELL.

Chlamys (Zygochlamys) delicatula (Hutton)

1873.—Pecten delicatulus Hutton Cat. Tert. Moll. p. 20.

1873.—Pecten diflira Hutton Cat. Tert. Moll. p. 31.

site wil 4 2 sitbantarctica Hedley Aust. Ant. Exped. Ser. C4 (1), Moll.

p.
1924.—-Chlamys, cambellicus Odhner, N.Z. Moll. Pap. Mort. Pacific Ixped.
p. Ol.

There seems little doubt that the names covered in the above syn-
onomy refer to one species with a time range from Nukumaruan
(Middle Pliocene) to the present and a Recent geographic range from
Macquarie Island 54°42’ S. to Eastern Otago, 46° S.

Genus PANOPEA Ménard de la Groye, 1807.

The bringing together of material from a wide range o! localities
has revealed the presence of a second Recent Panopea in New Zealand
waters and a new species ancestral to it from the Pliocene.

The two Recent members are easily separable by the depth of the
pallial sinus, which feature is coupled with a distinctive shell outline for
each species. In concise terms, selandica has a shallow sinus which
extends only half the distance from the posterior end to the umbo and
the shell outline is rectangular, broadly rounded anteriorly but square-
ended posteriorly. The new Recent species has an extremely deep sinus
which extends to beneath the umbo and the shell outline is more or less
rhomboidal, cut away at the lower anterior margin and obliquely pro-
tractively arcuate at the posterior end. The Pliocene new species has a
very deep sinus also, but coupled with a further distinctive shell shape.
The posterior gape is considerably greater for the Recent new species
which, on this evidence, coupled with that of the very deep sinus, is
indicative of a much deeper burrowing habit than for selandtied.

The distribution of the two Recent species presents a contused
picture :

(1) Neither species is stenothermic—selandica is more commonly found
in northern shallow waters, and the new species in deeper southern
waters. 20-70 fathoms, Eastern Otago, Stewart Island and Chathain
Islands. However, there are zclandica records (shallow water)
from The Spit, Otago Heads, and Stewart Island as well as n. sp.
records from the Hauraki Gulf im 26-30 Tathoms.

(2) Both selundicu and a species ancestral to the Recent deep-stnused

~ one occur in the Wanganui series, ie., se/andica typical from Land-
guard Bluff and Castlecliff and a deep sinused species from Land-
enard Bluff, Castlecliff, Kai Iwi, Nukumaru sands, Waipipt and
Waihi Beach, Hawera. I am not certain about the nature of the
sinus in the latter two occurrences, since the material is too fragile
to allow excavation, but externally they conform with the Castle-
cliffian deep sinused species. Specimens of Panopea orbita from
Mount Harris (Otaian-Hutchinsonian) mid-upper Oligocene, are
deep sinused.

——

a a a re ret

Vinlay (1926, T.N.Z.1. 57, p. 473) was incorrect in adopting the vernacular render-
ing, Le, “Panope,” as the genus name. The original proposition (April,
1807) was Panopea.

(3)

(4)

(5)

(6)

Mollusca, Eastern Otago 77

The habitat of ¢elandica is coastal in shallow-water in a substratum
of fine textured, often micaceous sand. The 60 fathoms Hick’s
Bay record is based upon an odd valve that may have washed down
from shallow water.

The true habitat of the new species is problematic. Dead
shells, often with conjoined valves, are commonly dredged from
shelly beds in Foveaux Strait and off the Eastern Otago continental
shelf. Also the two Hauraki Gulf records are based upon odd
valves dredged from coarse shelly deposits. This would seem to
confirm that the new species lives buried deeply in shelly deposits,
but on the other hand it may be either a mud dweller from an
adjacent soft bottom area or it may occupy a gritty transitional zone
between the hard and soft bottoms. However, the fact that many
of the Otago shells were conjoined pairs indicates that they had
not been transported far if at all from the shelly deposits from
which the shells had been dredged.

The Pliocene records only serve to confuse the issue still Turther.
Admittedly the Pliocene deep-sinused species 1s not specifically
identical with the Recent new species, so it is not essential that it
should have lived under the same ecological conditions.

Pliocene records of selandica confirm the habitat preterence
shown by the Recent occurrences of that species. It is found in the
shallow-water loose yellow quartz mica sands ut Landguard Bluff,
that is “LG. 1” of Fleming (1947 Trans. Roy. Soc. N.Z.. 76, p.
324), and also at Castlecliff in [leming’s Zethalia-Amphidesma
Sands, “CU. 5,” which is a shallow water deposit also, of loose,
coarse, current-bedded micaceous sand. |

The Pliocene records of the deep-sinused species, however, are
conflicting, especially as all the following records are of coujoined
specimens that must have lived in or near to the substratum in which
they were buried. he occurrences are (a) Landguard Biuff,
“TG. 1,” in loose yellow quartz mica sands, ( b) Castlechtf, prob-
ably “CU. 2," muddy medium micaceous sandstone, (c) Kai Iwi,
“CL. 10d,” muddy sandstone, ( d) Nukumaru Brown Sands,
“NU. 2,” loose fine to coarse sands of shallow water occurrence,
(e) Waipipi, mudstone, and (1) Waihi Beach, Hawera, mudstone.

From the above it will be noted that sclandica is always associated
with a fine textured sandy bottom in shallow water on an open coast,
that the deep-sinused Recent species favours deep water, coarse
shelly deposits, and that the Pliocene deep-sinused species appar-
ently ranged through sandy and muddy substrata and oceurred
from shallow to deep water.

- .

Notwithstanding the confused ecological data, the tact remains that
two easily recognised species of Recent Panopea occur in New
7ealand waters and that a Pliocene one requires nomination as a
third species ancestral to the deep-sinused Recent one. The tact
‘s established that all specimens examined have either a shallow or
a deep pallial sinus, and that the latter occurs quite irrespective of
depth or bottom materials. An intermediate development of the
sinus is nowhere evident.

ie POWELL.

(7) lvidently the shallow sinus is a late development, since the earliest
appearance of this feature is in selandica from Castlecliff. That
species may represent a divergence from the main line to fit a
definite ecological niche, 1.e., a fine textured sandy substratum in
shallow water off an ocean beach.

Panopea zelandica Quoy & Gaimard

1835.—Panopaca selandica Q. & G. Voy. Astrol. 3, p. 547, pl. 83, f. 7-9.

1843.—Panopea solaindri Gray, Dieff. N.Z., p. 255.

1913.—Panopea selandica: Suter, Man. N.Z. Moil. p. 1013, pl. 61, figs. 10, a.

1926.—FPanope selandica: Finlay, Trans. N.Z. Inst. 57, p. 473.

1946.—Panope selandica: Powell, Shellfish of N.Z. 2nd ed., pl. 11, £. 22.

Types: New Zealand; Mus, Hist. Nat., Paris (sclandica): Tauranga, British
Mus. (solandrz).

The exact locality of Quoy and Gainard’s zelaudica was not stated
but it was most likely Tasman Bay, Nelson. Their figures do not show
the sinus, but the outline of the valves coincides with gelandica of
equivalent small size, 1e.. 73 x 52mm. Only individuals of larger size
than this show the medial flattening of the posterior end. The dorsal
view shows a‘degree of inflation more in accord with sclandica than with
the deep-sinused species. Gray's solandri from Tauranga 1s undoubtedly
a synonym of selandica.

Length. Height. Inflation, Antertor gape. Sinus. depth. Locality.
128.0 mm. 79.0 mm. = —. 41.0 mm. C
121.0 mm. 68.5 mm. 52.0 mm. 38.0 mr. 36.0 mm. R

92.5 mm. 57.0 mm. 37.0 mm. 26.0 nium. 36.0 min. O

65.0 mm. 37.5 mim. 23.5 mum. 17.0 mm. 26.0 min. M

(C = Collingwood Beach, West Nelson; R = Rona Bay, Wellington;

O = Opctiki; and Mi = Mt. Maunganw, Bay of Plenty.)

Localities: Tokerau Beach, Doubtless Bay (A.W.B.P., ‘uae 1950) ; Chelten-
ham Beach (\W, H. Webster) and Takapuna Beach, Auckland (C. R. Laws);
Mt. Maunganui and Opotiki, Bay of Plenty; off Hick’s Bay, 60 fath. (S. V oss):
Rona Bay and Lowry Bay, Wellington Har bour: Collingwood Beach, West Nelson
(/A.W.B.P.):; New Brighton, Canterbury; Warrington, Otago (Finlay coll. Auck.
Mus.); The Spit, Otaga Heads (C. R. Laws coll, Auck. Mus. ); Stewart Island
(Auck. Mus.).

Note.—The 60 fathoms Hick’s Bay record is a single valve that
may have washed down from shallow water. I know of no other deep
water records for gsclandica.

Panopea smithae n. sp. Pl. 7, fig. 5 and text figs. 4-6.

Shell large, solid, inaequilateral, eaping at both ends, but much more
posteriorly. “Outline distinctly rhomboidal, anterior end the shorter,
narrowly rounded above and obliquely cut away towards the ventral
imargin; posterior end broadly rounded to flattened and inclined pos-
teriorly.

Compared with selandica the new species is more solid, is of rhom-
boidal rather than rectangular outline, has a much deeper pallial sinus
which terminates level w ith the umbo and a ereater posterior gape of
the valves. Hinge and ligament are similar in both species, but the
posterior adductor scar is relatively larger and more circular in smithae.

Mollusca, Eastern Otago 79

Length. Height. injlation. Posterior gape. Sinus depth. Locality.
130.0mm. 76.0 mm. os = 79.0 mm. S
112.0mm. 74.5 mm. 49.0 mm. 42.0 mm. 63.0 mm. O
105.0mm. 67.0 mm. 50.0 mm. 43.0 mim. 650mm. OCH)

940mm. 63.0 mm. 42.0 mm. 37.0 mm. 57.0 mm. O

60.5 mm. 41,0 mm. = = 35.0 mm. O

(S = Stewart Island; O = off Eastern Otago ca. 70 fathoms; H = holotype.)

Localities: Off Eastern Otago ca. 70 fathoms (type); Stewart Island (Mrs.
R. H. Harrison) ; Foveaux Strait, 17 fathoms (Finlay coll. Auck. Mus.) ; Owenga,
Chatham Islands (A.W.B.P. Auck. Mus.) ; Wellington Harbour dredgings; Lowry
Bay and Lyall Bay, Wellington (Dominion Mus.) ; (Since the above was written
Mr. R. K. Dell has forwarded to me the Panopea material in the Dominion Museum
collection. Both zelandica and smithae occur in two of the dredgings, but there 1s
0 indication of the bottom materials. Following is a note of the localities and the
species represented: Wellington Harbour (suction dredgings), 3 smithae and 2
=-iandica: Falcon Shoal, 6 fathoms, Wellington Harbour, 8 swithae and 2 selandica;
Lyall Bay (cast ashore), 2 smithae.); 1 mile off Cape Rodney, Hauraki Gulf, 26
fathoms (one valve, Govt. Trawler “Ikatere” Stn. 30); } mile off Wellington Head,
Great Barrier Island, 30 fathoms (one valve “Ikatere” Stn. 35).

All the above records are from a coarse shelly bottom.

Fenopea selandica Q. & G. Fig 1, Collingwood; Fig. 2, Opotiki; Fig. 3, Mount
Maunganui. Panopea smithae n. sp. Fig. 4, Stewart Island; Fig. 5, 70
fathoms, Eastern Otago; Fig. 6, 72 fathoms, Otago Heads. Panopea
wanganuica n. sp. Fig. 7, Holotype, Kai Iwi, Upper Pliocene; Fig. 8,
Nukumaru, Middle Pliocene.

(All figures to uniform scale of 2-5th natural size. )

SO POWELL.

Panopea wanganuica n. sp. Pl. 7, fig. 6 and text figs. 7 sa

Shell large, relatively thin, inaequilateral, gaping at both ends but
very much more posteriorly. Outline ovate-cylindrical, relatively
straight dorsally and ventrally and with shell height slightly ereater
towards the anterior end. Both ends narrowly rounded above middle
height and arcuately cut away in broadly rounded sweeps to the flattened
to slightly concave ventral margin. Pallial sinus very deep, extending
to directly under, or even a little anterior to. the umbo. | i

Compared with smithaz, the Pliocene species differs at sight in
having both ends converging to the ventral margin. The posterior
adductor scar is intermediate in size between that of smithae and that
of selandica. A feature common to both simithae and wanganuica is the
very straight hinge line.

Length. Height. Inflation. Posterior gape. Sinus depth. Locality.
(100mm. 57.5 mm. 42.0 mmm. 41.0 inm. 724) mm K ( H)
98.5mm. 60.5 mm. 424) mm. 38.0 mim. 62.0 mm. N
78.0 mm. 41.0 mim. 30.5 min, 26.0 mn. 53.0 mm. L
(i, = Kai Iwi; N = Nukumaru; L = Landguard Bluff: H = holotype. )

[Tolotype: Auckland Museum.

Localities: Landguard Bluff, LGl (A.W.B.P. coll.) ; Kai Lwi, CL10d (holo-
type); Wanganui Castleclifian (Upper Pliocene) and tentatively Nukumaru Brown
Sands, NU2 (N.Z. Geol. Surv.) ; Nukumaruan (middle Pliocene); Waipipi and
\Wathi Beach, Hawera, Waitotaran (lower Pliocene).

The Nukumaru specimen is relatively shorter and much higher
towards the anterior end, resulting in proportions nearer to those o!
selandica, but the sinus is very deep and the hinge line straight as in
qanganitica.

The pallial sinus is not visible in either the Waipipi or Waihi,
Hawera, specimens (N.Z. Geol. Surv. coll.), so their status is undeter-
mined. Both are nearer to wanganuica in shape, but are relatively
shorter.

Venustas blackin. sp. Pl. 7, figs 3 and 4.

Shell conical, moderately solid, imperforate, of flesh to bright pink
colour, sculptured with numerous regularly granulated spiral ribs and an
interstitial pattern of from two to five crisp spiral threads. On the base
the primary spirals are narrower, weaker and wider-spaced and the
interstitial threads rather more prominent, but margining the umbilical
callus pad there are two closely spaced granulated spirals equal in
strength to those of the spire-whorls. The spirals on the upper surface
are dark pink with the granules pale pink to almost white. The smooth
basal spirals have an alternation of pale and dark pink which resolves
into an irregular radial pattern. The two granulated spirals bordering
the umbilical callus have the granules pale on a dark pink ground as on
the spire. The primary spirals are three on the second and third post-
nuclear whorls, five on the ante-penultimate and eight on the penultimate.
There are eight or nine smooth primary spirals on the base.

Height, 41.0 mm.; diameter, 41.0mm. Spire angle, 81°.

Holotype: Auckland Museum, presented by the late Mrs. J. G.
Sniuth.

Mollusca, Eastern Otago x1

Locality: Eastern Otago, ca. 70 fathoms (Holotype and paratype
only ).

The species seems to be nearest to Tonilin’s megaloprepes (19 LS,
B.A.N.Z. Ant. Res, Exped. vol. 5 (5), p. 225) from Macquarie Island
which has the same kind of basal sculpture, including the paired stronger
spirals margining the umbilical callus.

Pachymelon (Palomelon) smithi n. sp. Pl. 7, figs. 4 and 2.

Shell large, solid, narrowly fusiform, sculptured with vertical axial
folds, 14-15 per whorl, which reach from suture to suture but are obso-
lete on the last whorl, strong within the aperture but not fully visible
from without. Spire two-thirds height of aperture. Post-nuclear
whorls six: nucleus eroded. Columella straight with four very oblique
plaits. Basal notch moderately broad and very shallow. Colour unt-

formly pinkish-buff without colour markings.
Height, 118.0 mm. ; diamenter, 45.5 mm.
Holotype: Auckland Museum, presented by MLr. J. G. Smith.

Locality: Eastern Otago, ca. 70 fathoms (The holotype, a dead
shell, only ).

The species is nearest allied to the Chatham Islands wilsonac
(Powell, 1933, Rec. Auck. Inst. Mus. 1 (4), p. 204) from which it
differs in its narrower outline, taller spire, much less prominent columel-
lar plaits, and absence of axials from the whole of the body-whorl. The

absence of colour pattern in the holotype may not be a constant feature.

Iredalina mirabilis Finlay.
1926.—I redalina mirabilis Finlay Proc. Malac. Soe. 17, pp. 59-62.

The holotype, which until the present has heen unique, was trawled
in 40 fathoms off Otago Heads.

Four dead shells in the present collection add a little to our know-
ledge of the species, which is shown to have a smooth glazed surface
and pale salmon coloration without colour pattern, The protoconch is

small. conical and symmetrically coiled, but is not well enough preserved
in any of the specimens to furnish any further observations.

Finlay compared his genus with the plaitless Kerguelen Island
Provocator and Guivillea, but favoured derivation of Iredalina from
early Ericusoid stock. This Australian genus, Hricusa, however, has well
developed plaits, a relatively large nucleus with an oblique apex and a
distinetive colour pattern.

My impression is that /redalina is much nearer to Provocator, not
only on account of the plaitless pillar but also the glazed surface without
colour pattern, the small erect nucleus and the deeply retrocurrent trend
of the outer lip to the suture.

The holotype of /redalina mirabilis is evidently abnormally elon-
cated. The present specimens have a shorter spire and relatively more
inflated body-whorl: i.e., height, 101 mm. ; diameter, +2 mim.

PLATE 7.

Figs. 1 & 2: Pachymelon (Palomelon) smithi n. sp. (Holotype).

Figs. 3 & 4: Venustas biacki n. sp. (Holotype).
Fig. 5: Panopea smithae n. sp. (Holotype).

Fig. 6

Panopea wanganuica n. sp. (Holotype).

New Plant Localities in the Auckland Province.

By RUTH MASON, NEVILLE T, MOAR,
Botany Division, D.S.1.1., Wellington,

and ROBERT COOPER. Auckland Museum,

Most of the plants recorded in this paper were collected on a series
of field trips in the Auckland Province. Details of the localities visited
are given in the list at the end of the paper.

The material collected on the trip made by the three authors in
November-December, 1949, was shared in the field as far as possible.
The numbers following the initials of R. Mason and N. T. Moar are
their collecting numbers and their specimens are deposited in the
herbarium of the Botany Division, D.S.LR., Wellington, Other num-
bers following the initials of R. Cooper or the persona] names ol H.
Powell and F. Browne refer to the herbarium sheets in the Auckland
Museum.

Plants recorded in New Zealand for the first ume are indicated hy

an asterisk (*).

The Cyperaceae in the Auckland Museum collection, referred to in
this paper, have been identified by Mr. V. J. Cook, to whom thanks are
due. We are also very grateful to Mr. and Mrs. Hayden, Matihetihe
Native School. and to Mr, and Mrs. R. H. Michie, Kaitaia, for hospi-
tality and assistance in the field.

NATIVE PLANTS.

Atriplex erystallina Wook. tf.

Karikari Bay, R.M. & N.T.M., 286; Sandy beach, north end of
Waikuku Beach, North Cape tonibolo, H. I. Powell, 26430.
Callitriche verna L.

Medland’s Beach, Great Barrier Island, Ik. M|., +4.

Culochilus sp.

Specimens of Calochilus with yellow flowers and dark red fimbriae
on the labellum were collected at Aranga swamp (K.M. & N,T.M., 88).
Mr. Edwin D. Hatch considers it may be a hybrid between C. cai pestris
and C. robertsonii and that it is practically identical with those specimens
from Ahipara, probably of hybrid origin, which he records under €,
campestris (1949; 248), except that the Aranga specimens have well-
developed fimbriae.

Carex gaudichandiana Kunth,
Swamp alongside Keri Keri inlet road, R.C., 35801.

oat Mason, Moar, Cooper.

Carex lucida Boott.

Medland’s Beach, Great Barrier Island, R.M., 464.
Carex pseudocyperus L.

Kaitoke, Great Barrier Island, R.M., 505.

Carex stellulata Good.
Otoroa, R.C. & N.T.M.
Cladium huttoni T. Kirk.

Kaitoke, Great Barrier Island, R.M., 424; Muir's Lake, Waiuku.
N.T.M., 462; Thompson's Lake, Waiuku, N.T.M., 475; Ototoa. Kaipara
Harbour, N.T.M., 562; North Head, Kaipara Harbour, N.T.M., 606
Danthonia bromoitdes Hook. f.

Seacliffs at Leigh, R.M., 583,

Drosera pygmaea D.C.

Karikari Bay, R.M. & N.T.M., 274; Lake Ohia, R.M. & N.T.M..
304, R.C., 35778. :
Eleocharis acuta R. Br. var. tenuis Carse.

Kerikeri swamp, R.M. & N.T.M., 365, R.C., 35867; Otekairangi,
K.M. & N.T.M., 410; Oruru Stream near Taipa, R.C., 35996.
Eleocharis neo-selandica C, B. Clarke ex T. Kirk.

Waihopai Stream, R.M, & N.T.M., 178.

Gleichenia flabellata R. Br.
Whangaparapara Valley, Great Barrier Island, R.M., 500.

Glossostigima elatinoides Benth.

Medland’s Beach, Great Barrier Island, R.M., 446.
[T\'datella inconspicua Cheesem.

L. Rotokawau, North Head, Kaipara Harbour, N.V.M., 584; L.
Pouto, North Head, Kaipara Harbour, N.T.M., 571; L. Waiparera,
near Waiharara, H, [. Powell, 26434.

Juncus holoschoenus R. Br, var. multiflorus Carse.

Tokerau Beach road, R.M. & N.T.M., 301.
Juncus pallidus R. Br.

Kaiangaroa, R,M. & N.T.M., 322; Merita, R.C., 36160,
Korthalsella salicornioides (A, Cunn.) Van Tiegh.

Parasitic on Leptospermui ericoides at Kapowairua, Spirits Bay,
R.C., 24435.

Masus pumulio R. Br.

Waihopai Stream, R.M. & N.T.M., 140.
Metrosideros albiflora Sol. ex Gaertn.

Warawara Kauri Forest, R.C., 35577.

CA
ty

New Plant Localities.

Myriophyllum robustuim Hook f.

Whareana Creek, North Cape, H. E. Powell, 26439; Poutu, Kai-
para Harbour, N.T.M., 610.

Myriophyllum votschit Schindler.

Ahipara Hill, R.M. & N.T.M., 327, R.C., 35738; Lake Negatu,
R.M. & N.T.M., 265, R.C. 35775; Karikari Bay, R.M. & N.T.M., 297 ;
Lake Kanono, N.T.M., 588.

Nertera setulosa Hook. f.

! Waipoua Forest, K. W. Allison; Kaihu River, R.M, & N.T.M., &4.
R.C., 35504: Warawara State Forest, R.C., 35620, 35624; Waihopai
Stream, R.M. & N.T.M., 165, 171: Wairoa Stream, R.M. & N.T.M..,
193; Otekairangi, R.C., 35914; Oruru Stream, Taipa, R.C., 36018;
Pekerau, north of Taipa, H. E. Powell, 26438; Lake Waiparera, near
Waiharara, H. E. Powell, 26432.

Nothopanax anomalum (Hook. f.) Seem.

Near summit of Ounuwhao, 950’, Spirits Bay, R.C., 24477.

Pittosporum pimeleoides R, Cunn. ex A. Cunn.

In remnant of kauri forest on south bank of Oruru River, one mile
west of Taipa, R.C., 36009; H. E. Powell, 26437.
Pomaderris edgerleyt Hook f.

Near Babylon, R.M. & N.T.M., 64, R.C., 35483.

Potamogeton cheesemanit A. Bennett.

Medland’s Beach, Great Barrier Island, R.M., 445.
Pseudowintera axillaris (J. R. & G. Forst.) Dandy.

Near summit of Ounuwhao, 950’, Spirits Bay, R.C., 24475,
Schoenus apogon Roem. et Schult.

Warawata State Forest, R.C., 35571; Waihopai Stream, R.M. &
N.T.M., 149, 156; Ahipara Hill, R.C., 35707; between Whangatupere
Bay and Brodie’s inlet, R.C., 35792; Otoroa, R.C., 35841.

Schoenus carsei Cheesem,

Ahipara Hill, R.M. & N.T.M., 211, R.C., 35704; Kerikeri swamp,

R.M. & N.T.M., 373.
Schoenus nitens (R. Br.) Poir.
North Kaipara Head, N.T.M., 605.

Schoenus nitens (R. Br.) Poir. var. concinnus (Hook. f.) Cheesem.
Karikari Bay, R.M. & N.T.M., 308.
Scirpus caldwellti V. J. Cook.
Tokatoka, R.M. & N.T.M., 50, R.C., 35466a.
Scirpus medianus V. J. Cook.
Tokatoka, R.M. & N.T.M., 49, R.C., 35466b; Tryphena, Great
Barrier Is., R.M., 461; Lake Kanono, North Kaipara Head, N.T.M.,
591.

RG MASON, MoAR, CoopeEr.

Scirpus perviridis V. J. Cook.
Omapere, R.M. & N.T.M., 118, R.C., 35545: Lake Tongonge,

R.M. & N.T.M., 323: north of Lake Pokoroa, Waiuku, N.T.M., 493. \

Scirpus sulcatus Thouars var. distigmatosa C. B. Clarke ex Cheesem.

In crevices among wet boulders under Keri Keri Falls. R.C.. 35932.
Solanum aviculare Forst. f. var. albiflora Cheesem.

Kaiarara Bay, Great Barrier Island.

Utricularia delicatula Cheesem.

Aranga swamp, R.M. & N.T.M., 105: Ahipara Hill, R.M. &
N.T.M., 215, R.C., 35709; Otoroa swamp, R.M. & N.T.Mz, 332.
(tricularia novae-selandiae Hook. f.

Lake Ohia, R.M. & N.T.M., 318, R.C.. 35806: Lake Pouto.
NCTM, 583.

(tricularia protrusa Hook. f.

Lake Rotokawau, Opoe S.D., R.M. & N.T.M., 251, R.C., 35766,
29/80; Warikari Bay, R.M. & N.T.M., 282: Wilson’s Lake, N.T.M..
913; Pouto, North Kaipara Head, N.T.M., 582.

ZoNnsta matrella (1L.), Merr.

Kaitoke, Great Barrier Island. R.M.. 423.

INTRODUCED PLANTS.
Alisma plantago-aquatica 1.

Kaitaia, end of Bonnett’s Road, R.M. & N.T.M , 249, R.C., 35754.
This plant, without flowering stalks, was also seen in the stream running’
behind the motor camp at Kaitaia. According to Mr. R. H. Michie it
has come into the district within the last fifteen years. The leaves of the
Kaitaia plants are subcordate at the base, whereas the leaves of plants
from Wellington southwards have cuneate bases. The W ellington plants
are also larger in size. In view of the differences between the northern
and southern plants, and the distance between the occurrences, it is
possible that they have come from separate introductions.

Anacharis canadensis (Michx.) Planch.

Muir's Lake, Waiuku, N.T.M., 444. In the neighbourhood of
Auckland this is known only from the Waikato River and Western
Springs, and was not seen at all north of Auckland. The specimens
collected by k. H. Matthews and recorded by Cheeseman (1925: 1054
as Elodea) seem to be the only ones so far collected in North Auckland.
Anthemis nobilis L.

Otekairangi, R.M. & N.T.M., 405, R.C., 35900.

Apium tenuifolium (Moench) Thellung. |
Medland’s Beach, Great Barrier Island, R.M., 450.
Arunt sp., either Arum maciulatum L. or a closely allied species.

Near site of old house, Okupu Bay, Great Barrier Island.

New Plant Localities. 87

Aster subulatus Michx.
Some plants were seen at the head of Whangaparapara Harbour,
(sreat Barrier Island.
Blackstonia perfoliata (L.) Huds.
Karikati Bay, R.M. & N.T.M.., 291.
Cardamine pratensis L.
! Woodhill, R.M. & N.T.M., 10, 17; Kaitaia motor camp, RM. &
N.T.M., 206, R.C., 35698; Otekairangi, R.M. & N.T.M., 385.
Carex divulsa Good.
Whangarei station, R.M. & N.T.M., 413, R.C., 35915; Leigh,
R.M., 516.
Carthamus lanatus L.
Mangawai, Miss F. Browne, 36183.

*Cyperus sanguinolentus Vahl.

Waihopai stream, k.M. & N.T.M., 147, 160. These specimens
were compared with material from the National Herbarium of N.S.W..,
Sydney, and seem to agree well, except that the New Zealand specimens
have smalier inflorescences with no rays and fewer spikes.

Eichhornia crassipes Schlecht.

Muir’s Lake, Waiuku, N.T.M., 445. This was deliberately planted
in the lake.

Erechtites atkinsonae F. Muell.

Whangaparapara Harbour, Great Barrier Is., R.M., 501. Novem-
ber was too early to find this in flower and no specimens were collected,
but young plants were seen frequently wherever we went north of
Auckland.

Erechtites valerianacfolia D.C.
Kaitoke Valley, Great Barrier Island, R.M., 507, 512.

Erica baecans L.
Te Ahumata, Great Barrier Island, R.M., 471; ridge between Kai-
toke Valley and Awana Valley, Great Barrier Island, R.M., 508.

Eupatorium adenophorum Spreng.
In damaged bush near Otekairangi swamp, R.M. & N.T.M., 403,
R.C., 35899.

Festuca arundinacea Schreb.
Great Barrier Island.

Galiuim. palustre L.

Woodhill, R.M. & N.T.M., 12, R.C., 35429; Kaihu River, R.M. &
N.T.M.. 79, R.C., 35494; Lake Tongonge, R.M. & N.T.M., 5c Al a
35758: between Taumarere and Kawakawa, R.M. & N.T.M., 353.

Galium parisiense L.

Road between Medland’s Beach and Tryphena Harbour, Great
3arrier Island, R.M., 453,

X& \LASON, Moar, Cooper.

Glyceria sp.

In the gully near the wharf in Blind Bay, Great Barrier Island,
there is a patch of Glyceria, probably G. fluitans.
Hakea acicularis R. Br.

Kaitoke, Great Barrier Island, R.M., 418. This is widespread in
the scrub country on Great Barrier Island. It was, however, more
abundant to the south of Port Fitzroy Harbour than south of Mt.
Hobson and Mt. Young.

Flakea pubescens Schrad.

Kaitoke, R.M., 514. This is also widespread in the scrub country
on the Great Barrier Island, but appears to be much the most abundant
species south of Mt. Hobson and Mt. Young.

*Hedychium oblongum K. Schum.

Kaeo, a large patch by the roadside in the town, R.M, & N.T.M..
333, R.C., 35831. This plant fits the description of H., oblonguim K.
Schumann (1904: 48) quite well, except that the bracts are rather hairy
at the upper end and almost tomentose at the middle of the extreme tip.
Kyllinga brevifolia Rottb.

Waihopai Stream, R.M. & N.T.M., 146, 157: Kaitoke, Great
Barrier Island, R.M., 481; Wilson's Lake, NOP ML, B25:

Lavatera arborea L.
Rawene, R.M. & N.T.M., 122, R.C., 35538. On waste land near

the wharf there were several plants of a variegated form of L. arborca,
smaller than the normal green form. L. arborea was not noticed at any
other coastal localities.

*Ligustrum sinense Lour,

Kaeo, R.M. & N.T.M., 334, R.C., 35832. This plant is common on
roadsides and in waste spaces about Auckland City. It agrees with the
description of L. sinense but there is no authentic material for compari-
son in the Botany Division and Auckland Museum herbaria.

Lotus angustissimus L.
Great Barrier Island.

Lotus uliginosus Schkuhr.
Great Barrier Island.

Ludwigia palustris (L.) EI.

Near Babylon, R.M. & N.T.M., 80; Kerikeri swamp, R.M. &
N.T.M., 356, 376; stream alongside Puketona-Paihia road, one mile
above Haruru Falls, R.C., 35856; Otekairangi, R.M. & N.T.M., 408,
R.C., 35902; Kaitoke, Great Barrier Island, R.M., 441; Muir's Lake,
N.T.M., 456; Wilson’s Lake, N.T.M.

Mariscus congestus (Vahl.) C. B. Clarke.

Karikari Bay, R.M. & N.T.M., 294; Sandy track, Merita, R.C.,
36076; Pekerau, north of Taipa, H. E. Powell, 26435; Muir’s Lake,
N.T.M., 450; Thompson's Lake, N.T.M., 478.

New Plant Localities. ao

*Nymrphaea alba L.

Stream alongside Puketona-Paihia road about a mile above Haruru
Falls, RM. & N.T.M., 355, R.C., 35855; Muir's Lake, N.T.M., 446,

This may have been planted.

Opuntia monacantha Haw.

Okupu Bay, Great Barrier Island. Several plants not far from site
of old house.
Panicum lindheimeri Nash.

Kaitoke Valley, Great Barrier Island, at head of valley near eastern
hot springs. R.M., 506, 508.
Paspalum dilatatuim Poir.

Great Barrier Island.
*Passiflora edulis Sims.

Mangapiko Saddle, Great Barrier Island, R.M., 497. There were

a number of fruiting plants growing wild among short Leptospermui
scrub near an old hut where the tram tracks pass between Mangapiko
and Mt. Young.

Pennisetum clandestinum Hockst.
Roadside, Kaitoke Valley, Great Barrier Island.

Physalis peruviana L.
In bush gully between Whangatupere Bay and Brodie’s Inlet,

RAG 35795.

*Phytolacca americana L.

Matai Bay, R.C., 35783. These specimens fit the description ot
P. americana by H. Walter (1909: 52-55) much better than that of any
other species, but have 12 stamens instead of 10 and sometimes the base
of the raceme is paniculate. However, according to P. Wilson (1932:
263), the number of stamens may be Y to 12.

Phytolacca octandra L.
Medland’s Beach, Great Barrier Island, R.M., 452.

Pinus sp.

A species of Pinus, probably P. iar ‘iia, is spreading in the scrub
in the valley behind the post office and wharf at T ryphena, Great Barrier
Island. In the serub on the open hill on the south side of Kaiarara Bay
there are also small pine trees scattered here and there.

Polygala myrtifolia L.
Kaitoke, Great Barrier Island, R.M., 503. odd plants under Lepto-
spermum ericoides. It was also seen near Cooper's wharf, Port Fitzroy.

Polypogon liuttosus (Poir.) Hitehe.

Omapere, R.M. & N.T.M., 119, R.C., 35541; Koware Trig, above
Matai Bay, R.C., 35786; track from Brodie’s inlet to Merita, R.C..
36066.

90) Mason, Moar, Cooper.

Potamogeton crispus L,

Thompson's Lake, Waiuku, N.T.M., 480.
Psoralea pinnata L,

Tryphena Harbour, Great Barrier Island, R.M. 463. This was
common near the shore at the north end of the harbour. One or two
plants were noticed on the road leading from Blind Bay.

Ranunculus flanimnula LL,

Kerikeri Swamp, R.M. & N.T.M., 357, R.C., 35857: between Tau-
marere and Kawakawa, R.M. & N.T.M., 352, R.C., 35850.
Ranunculus fluitans Lam.

Wilson's Lake, N.T.M., 515. This was recorded by Cheeseman
(1925: 1064, as Rk. aquatilis) “from the Waikato southwards,” and by
Allan (1940: 48) as “frequent in streams and ponds in both Islands.”
It was not seen north of Auckland city and no specimens seem to have
been collected there.

*Rubus sp.

Near the junction of the Pawarenga-Broadwood and Broadwood-
Herekino roads, R.M. & N.T.M, 205, R.C., 35695. This is a plant with
an upright habit, somewhat like that of a raspberry but of denser
growth; the leaves are palmately divided. It was the only introduced
species of Aubus noted from the car as we drove from the Broadwood
turn-off through the Herekino gorge until Kaitaia was reached. It was
common and often growing in dense masses by the roadside and in damp
paddocks,

Rubus sp.

A few poorly developed canes of blackberry were noticed in the
Kaitoke valley, Great Barrier Island, mostly on farmland.
Rumex conglomeratus Murray.

Medland’s Beach, Great Barrier Island, R.M., 451.

*Rumex hydrolapathum Huds.

Helensville, N.T.M. & R.M., 21, R.C., 35435. These plants were
scattered in a patch about eight yards long on a low stopbank on the town
side of the bridge across the Kaipara River.

Scleropoa rigida (1.) Griseb.

Woodhill, R.M. & N.T.M., 11, R.C., 35427; North end of Taipa
seach, R.C., 30043.
Sisyrinchium micranthum Cay.

Ahipara Hill, R.M, & N.T.M., 246, R.C., 35772.
Soliva anthemifolia ( Juss.) R.Br.

End of Bonnett’s Road, Kaitaia, R.M. & N.T.M., 258, R.C., 35761.

Stenotaphrum secundatum (Walt.) Kuntze.

Great Barrier Island.
Tradescantia fluminensis Vell,

Okupu Bay, Great Barrier Island, near wharl—a small patch at
cde of stream near site of old house.

New Plant Localities. 9]

Trifolium glomeratum L.
Great Barrier Island.

Tropaeolum majus L.
Okupu Bay, Great Barrier Is., near site of old house near whart.

Ulex europaeus L.

This was recorded by Kirk (1869: 154) as occurring on the Great
Barrier Island, but no further details were given. At present it 1s grow-
ing abundantly about Port Fitzroy, but it was not seen at all in the south
of the island and local residents hope that its spread southwards will not
take place,

Vitis vinifera L. ?
Road between Medland’s Beach and Tryphena, Great Barrier

Island, R.M., 454. Grape vines are growing over several cuttings on
this road and are well established.

*Wahlenbergia sp.

Otekairangi, R.M. & N.T.M., 406, R.C., 35901. This plant fits
the description of W’. tadgellii Lothian (1946: 228-229), but there is no
authentic material with which to make a comparison in the Botany Divi-
sion and Auckland Museum herbaria.

Iatsonia bulbilliifera Matthews et Bolus.

Woodhill, R.M. & N.T.M., 13, R.C., 35430; Kaitaia motor camp,
R.C., 35701. This plant is a common wayside weed north of Auckland
and was noted near Warkworth, at Wellsford, in and about Dargaville,
Waimamaku, Omapere, Pangaru, and Mangonui, and it was abundant
at Kaeo, where it was first recorded.

Watsonia meriana Mall.

Henderson, R.M. & N.T.M., 1, 2, R.C., 35418, 35420; Ruawai,
R.M. & N.T.M., 44, R.C., 35463. The specimens collected at Henderson
included both white (Nos. 1 and 35418) and mauve (Nos. 2 and 35420)
flowered forms.

Zantesdeschia aethiopica Spreng.

Mitimiti sand dunes, R.C., 35654; Lake Tongonge, R.M. & N.T.M.,
255.
Zizania latifolia Turez.

Wilson's Lake, South Kaipara Head, N.T.M., 531.

LOCALITIES COLLECTED.
Most of the plants recorded in this paper were collected on the
following field trips: Mangonui, Spirits Bay, Kerr Point, Kaitaia, in the
North Auckland Botanical District. (February, 1949. R. Cooper.)

Auckland. Henderson, Woodhill, Helensville, coast road to Tauhcoa.,
Warkworth, Topuni, Maungaturoto, Ruawai, Tokatoka, in the South
Auckland Botanical District, Dargaville, Babylon, Maitahi, Kahu,
Aranga swamp, Waipoua Forest, Omapere, Rawene, Narrows Landing,
Pangaru, Mitimiti, Warawara State Forest, down coast to Waihopat
stream and swamp in upper part of valley, down Wairoa stream from
road near Reena. Runaruna, Herekino Gorge, Kaitaia, Ahipara Gum-
lands Road, Lake Tongonge from end of Bonnett’s Road, Lake Ngatu,

9? Mason, Moar, Cooper.

Lake Rotokawau in Opoe $.D., Karikari Bay and swamp by way of
Wairahoraho stream, Matai Bay, Brodie’s Inlet, Tokerau Beach road,
Lake Ohia, Kaiangaroa swamp, Mangonui, Kaeo, Otoroa swamp, swamp
on Kerikeri inlet road, Paihia, Kawakawa, Whangarei and Otekairangi
swamp, in the North Auckland Botanical District. (November-Decem-
ber, 1949. R. Mason, N. T. Moar, R. Cooper.)

Great Barrier Island, in the Thames Botanical District: Okupu,
Blind Bay to Kaitoke Vailey, Te Ahumata (Whitecliffs), Whanga-
parapara, Port Fitzroy by way of ridge between Kaitoke and
Whangaparapara valleys, foot of Mount Young and Mangapiko
to road and main wharf, up valley from Kaiarara Bay and
old track to saddle ‘between Mount Hobson and Mount Young,
landing at Cooper’s, Port Fitzroy; road from Kaitoke to Tryphena
Efarbour and old bridle path over to Blind Bay, hot springs in Kaitoke
Valley and over old track to ridge and down into Awana Valley.
(Christmas, 1949. R. Mason.) Since Kirk (1869; 144-154) published
an account of the botany of the Great Barrier, which included a list both
of native and of a considerable number of introduced plants, there have
been only occasional new records from the island. A good many intro-
duced plants that grow on the mainland have reached the Barrier and a
number of new records, mainly of introduced weeds, are included here.
Where no collecting number is given, no specimen was collected. No
attempt was made to make a complete list of introduced, escaped and
naturalized plants, and no doubt many more new records will be found.

Mangonui, Taipa, Brodie’s inlet, Matai Bay, Merita, Whatuwhiwhi,
in the North Auckland Botanical District (January, 1950. R. Cooper).

South Auckland Botanical District—Franklin County : Thompson’s
Lake, Map N51 195E O92N; Muir’s Lake, Map N51 210E O6/N ;
Lake Pokoroa, Map N46 15OF 19ON; Lake Okaihou (Houghton’s
Lake), near Muriwai. South Kaipara Head: Wilson’s Lake, Map N37
1SOE 920N; Lake Kuwakatai, Lake Ototoa. North Kaipara Head:
Swan Lake (Lake Pouto), Lake Rotokawau, Lake Waingata, Lake
Kanono, Lake Humu Humu. North Auckland Botanical District —
Mangonui County: Lake Ngatu, Lake Rotokawau in Opoe 5.D., Lake
Waiparera. Map references are given for lakes for which no names
were found on a map. (January-February, 1950. N. T. Moar.) |

REFERENCES.
ALLAN, H. H., 1940. A Handbook of the Naturalized Flora of New Zealand.
N.Z. Dept. S. and i.R. Bull. 83.
CHEESEMAN, T. F., 1925. Manual of the N.Z. Flora. 2nd ed.

HATCH, E. D., 1949. The New Zealand Forms of Calochilus R.Br. Trans.
Roy. Soc. N.Z. 77: 247-249.

KiRK, T., 1869. On the Botany of the Great Barrier Island. Trans N.Z. Inst.
1: 144-154.

LOTHIAN, N., 1946. Critical Notes on the Genus Wahlenbergia Schrader ; with
Descriptions of New Species in the Australian Region. Proc. Linn.
Soc. N.S.W. 71: 201-235.

SCHUMANN, K., 1904. Zingiberaceae. Pflanzenreich 20 (IV 46): 1-458.
WALTER, H., 1909. Phytolaccaceae. Pflanzenreich 39 (IV 83): 1-154.
WILSON, P., 1932. Phytolaccaceae. N. Amer. FI. 21: 257-266.

RECORDS

OF THE

AUCKLAND INSTITUTE
AND MUSEUM

Tein fe. INGOs ok

Published by Order of the Council :
Gilbert Archey, Director

Edited by: A. W. B. Powell

Assistant Director

19TH DECEMBER, 1951

Unity Press Ltd., Printers, Auckland.

CONTENTS

VOL. 4. No. 2.

Notes to Accompany a Topographical Map and a Provisional Geological
Map of Great Island, Three Kings Group.

By M. H. Battey, Geologist, Auckland Museum ..

{lingamita (Myrsinacea), a new Monotypic genus from West Island,
Three Kings.
By G. T. S. Baylis, University of Otago

Incipient Forest Regeneration on Great Island, Three Kings Group.
By G. T. S. Baylis, University of Otago

The Flora of the Three Kings Islands: Additional Notes: with Note
on Suttonia.
By W. R.. B. Oliver, Wellington

Effect of Goats on Great Island, Three Kings: The Permanent
Quadrats Resurveyed.
By M. Holdsworth, University of Otago

Succinea archeyi Powell.
By H. E. Quick, M.B., B.Sc., F.R.C.S., Reading, England

Land Mollusca from Four Islands of the Three Kings Group: with
Descriptions of Three New Species.
By A. W. B. Powell, Auckland Museum

On Further Colonies of Placostylus Land Snails from Northernmost
New Zealand.
By A. W. B. Powell, Auckland Museum

Notes on the Birds of the Three Kings Island.
By E. G. Turbott, Auckland Museum

Page

103

ELT

127

141

Notes to Accompany a Topographical Map
and a Provisional Geological Map of
Great Island, Three Kings Group.

By M. H. BATTEY, Auckland Museum.

The maps presented herewith have been compiled from surveys
carried out during a visit of five days’ duration to Great Island as a
member of an expedition conducted by the Auckland War Memorial
Museum, under the leadership of Dr. Gilbert Archey, in January, 1951.
Some observations on the geology of the narrow waist of the island had
ee made during an earlier visit in 1948, when four hours were spent
ashore.

TOPOGRAPHIC WORK.

The bases of previous maps of the Three Kings Islands have been
a sketch map of Great Island made by S. Percy Smith in 1887 (Cheese-
man, 1891) and the Admiralty Chart of the whole group at the scale of
1:100,000 revised in 1911. Mr. E. G. Turbott, who spent five weeks
on the island in 1946, took a number of compass bearings which he used
to correct existing maps. His final map of the whole group (Rec. Auck,
Inst. & Mus., vol. 3, p. 190, 1948) unfortunately lacks a scale.

The horizontal control of the map now presented is a triangulation
by prismatic compass bearings taken to prominent hill features and
coastal promontories. Mr. Turbott, in his observations, used some of
the same natural features as points of bearing and his readings have
been checked with those of the later survey and show satisfactory agree-
inent. Those taken from the conspicuous Rocky Knoll (750°) at the
head of Castaway Valley west of the narrow waist of the island were
particularly useful.

The map depends for scale upon a short and roughly measured
base between Bald Hill (601’) and the spot height 577’ on the crest of
a rocky bluff to the south-east. A longer and more carefully measured
base is the first requirement in any further work. Sites for such a base
are few, but the open ridge north-north-west and south-east of Bald:
Hill offers scope for a much improved base-line, which could be
expanded by observations to Rocky Knoll (750°) and to the highest
point of the island (920’) on the cliffs above North West Bay, west of
the landing place. A bearing picket on this high point would be helpful,
as it is poorly defined from points of view lying to the south and south-
east.

Vertical control was established by aneroid barometer corrected as
far as possible for diurnal barometric changes by repeated readings.
Weather conditions throughout the survey were exceptionally’ steady.
The camp site was used as a base for the daily traverses, its height being
taken as 315ft., the approximate mean of five pairs of reading at the
camp and at sea level taken with the least possible lapse of time between
them.

94 BATTEY.

_ A great amount of detail was plotted with the aid of a large collec-
tion of excellent photographs of the island (particularly its coastline )
taken by many different visitors and assembled by Mr, A. W. B. Powell,
Assistant Director of the Auckland Museum.

GEOLOGY.

In 1936 the late Professor J. A. Bartrum described spilite lavas
and keratophyres from Great Island, this being the first record of the
occurrence of rocks of the Spilitic Suite in New Zealand. The spilites
were seen to exhibit pillow structure at a point which Professor Bartrum
described to me in conversation as lying near sea level perhaps a quarter
of a mile west of the landing place in North West Bay. Beyond the
conclusion, based on microscopic texture, that the albitic porphyry that
he described is a flat-lying sill and that probable keratophyre tuff 1s
present, Professor Bartrum did not deal with the field relations of the
rocks. He regarded the greater part of the island as being composed
of greywacke.

In the course of the recent work only greywacke was found in the
eastern part of the island, but very little sediment outcrops in the western
part, to which the succeeding remarks on stratigraphy and structure
chiefly apply.

The igneous rocks found in the western part of the island are of
two kinds, apparently broadly the same as the two types described by
Professor Bartrum. The basic (probably spilitic) lavas are seen to
ereat advantage in the huge blocks that litter the shore east of the
Landing Place in North West Bay. They are dark green when fresh,
weathering yellow-brown. They are never conspicuously porphyritic,
but in some cases long, slender feldspar laths are clearly visible to the
unaided eye. The felted texture of fine feldspar microlites can gene-
rally be seen with a lens, even in weathered chips. Green chloritic spots
are commonly seen, but coarser amygdaloidal structure is an inconstant
feature. It may possibly be related in distribution to the tops of flows,
for it seems to become more conspicuous as an overlying band of
keratophyre is approached.

Pillow form was never observed (the classic locality could not be
reached without a dinghy), but a tachylytic breccia forms the top of
the crag (603’) above Hapuka Point and the steep walls of this crag
are worthy of further examination.

The keratophyres are characterized by their light weathering colours
and a tendency to form low bluffs in the inland area. They exhibit in
places a structure interpreted as flow-banding (see below). On the
fresh surface they are of a vitreous or greasy lustre, dark to pale green-
ish or pale yellow, with a streaky inhomogeneous appearance and, often,
conspicuous pink oblong feldspar phenocrysts in a groundmass that is
structureless even under a lens.

A brecciated phase was found in four places, and three of these are
believed to be of one horizon. It appears in hand specimen to be of the
nature of a flow-breccia and not a consolidated pyroclastic deposit. This
view is perhaps borne out by the uncertainty expressed by Professor

Maps of Great Island. 95

Bartrum whether the rock was a tuff or a flow. At a few places, for
example near South Point, the rock takes on a nodular appearance, the
nodules being the size of small marbles.

A detailed laboratory examination of the rocks collected on Great
Island has not yet been made.

The courses of the traverses made during the recent survey are
indicated on the geological map by the positions of outcrop symbols.
It should be understood that those in the upper basin of Tasman Stream,
south of the cliffs above North West Bay, are only approximately
located, as the low forest prevents satisfactory compass bearings.

Observations of dip shown in the keratophyres in the basin of
Castaway Stream and along the cliff top north-north-west ot the camp
were taken upon a planar structure in the rock which 1s interpreted as
flow-banding. It is best displayed upon weathered surfaces, which
exhibit discontinuous parallel grooves and ridges reaching about an inch
in thickness and afford in places quite satisfactory readings. Laminated
sediments intercalated in the basic lavas provided evidence of attitude
in the northern cliffs and in the bed of Tasman Stream. Orientation
of elongated amygdales in some of the lavas proved to be somewhat
erratic and of little apparent value in suggesting the attitude.

On the basis of these observations and upon the distribution of the
different lithological types the contacts between beds of basic lava and
keratophyre have been extrapolated from points where they were found,
so that they follow courses appropriate to the topography upon the
assumption of a regular easterly dip of 25 degrees.

Important in this interpretation is the presence of a conspicuous
and characteristic keratophyre breccia near spot height 823ft. on the
crest of the northern cliff and in boulders at the Landing Place in North
West Bay, but not eastward of the Landing Place. This is interpreted
as indicating that the breccia reaches sea level near the Landing Place,
although it is there masked by screes, and further examination is needed
to prove its actual position. Similar breccia occurs at the same inferred
horizon near South Point. The distribution of richly amygdaloidal basic
lavas, interpreted as the tops of flows, appears to conform with the
observed outcrop pattern, though the value of this indication is perhaps
open to question.

The presence of a reddened zone at the contact of one type of lava
with another has been noted at several places and, being conspicuous
from a distance, serves as a useful guide to places where contacts may
be sought in areas of bare ground. The contacts seen were all poorly
displayed. Clayey weathered material usually masked the actual junc-
tion of the two rock types. A readily accessible contact, better than
some in this respect, between basic lava below and weathered kerato-
phyre above, may be seen on the cliff top on the north side of the narrow
ridge joining the two parts of the island, a few yards west of its
narrowest and lowest point. A contact between fairly fresh rocks in
the northern cliff at 660{t. above sea level below spot height 823ft.
shows a very irregular line of junction which, however, cannot be traced
far.

96 BATTEY.

The agreement between the observed rock distribution and the
inferred boundaries on the map 1s satisfactory, as far as it goes, except
that no keratophyre band was picked up on a traverse made on a fixed
bearing westward from Rocky Knoll (750’) to the proximal end of
Hapuka Point spur (917). Outcrops in the place where keratophyre
should appear are, however, very sparse and obscure.

While it will be clear that much more remains to be done to check
the correctness of the outcrop pattern, it is hoped that the accompanying
maps will serve as a guide to future investigators. In particular, they
provide a base upon which the positions of further rock samples can be
fixed by any visitor who is not a geologist, and in this way advances in
our knowledge may be made from time to time.

A few simple tests of the stratigraphic picture may be mentioned.
The keratophyre encircling the head of Castaway Valley forms con-
spicuous white weathering bluffs and lines of boulders. The course of
the inferred contact between the keratophyre and the underlying basic
lava can easily be checked by any visiting naturalists.

A suite of specimens taken at intervals up the lower course of
Tasman Stream, where it flows in a rocky channel, would be of 1mme-
diate service, as would samples from the spur of white-weathering rock
west of the mouth and lower part of Tasman Stream.

Careful work on the northern cliffs would also serve to verify the
relationships. Unfortunately, screes obscure much of the accessible
part; nevertheless, accessible outcrops are plentiful. A climb from the
Landing Place to the cliff top, west of the route usually followed,
appears quite feasible and would afford much information.

The relationship of the greywacke of the eastern part of the island
to the igneous rocks is not known. The slopes around the head of South
East Bay may afford information on this question. The material en
the crest of the ridge linking the two parts of the island is weathered
to clay, and the nature of the contact between greywacke and kerato-
phyre cannot be seen. The extension of the greywacke to the cliff tops
west of South East Bay is entirely hypothetical.

A large part of the eastern section of the island has a smooth,
sloping surface underlain by sandy soil and, at East Point and the
Point south of it, well-rounded pebbles that have weathered out of this
sandy covering litter the ground. This surface, varying in height from
250ft. south of East Point to 325ft. where it merges into the higher
ground along the cliffs east of North West Bay and into the ridge
running south from these near their highest point (602), is regarded
as an elevated plain of marine erosion and deposition, with shore pebbles
and a sandy cover upon it.

A cavernous buff sand-rock, often quite hard, is conspicuous in
blocks in many parts of the island on both the eastern and western
portions. A little east of the highest point (602’) of the eastern part it
overlies greywacke and encloses pebbles of igneous rock and greywacke
with characteristic red weathering crusts upon them. Outcrops of this
cavernous buff deposit also cap the cliffs near East Point and south of
‘++ Masses of it have been found in the western portion of the island
also. sometimes at heights greater than that of the plain of marine

Maps of Great Island. 97

erosion in the east, and with them pieces of igneous rock with a reddish
weathering-crust. In general, it would seem that pieces of rock with
this red weathering-crust should be regarded with deep suspicion,
wherever they are found, although, in areas of poor outcrops in the
upper Tasman Valley, it is tempting to accept them as float from the
bed rock, This reddened crust on loose blocks is quite distinct from
the reddening at stratigraphic contacts noted above.

A satisfactory explanation of the origin and distribution of the
cavernous buff rock has not yet been found. The balance of probability
seems to favour its origin as a superficial accumulation of some earlter
higher stand of the sea. I believe that it should not be confounded with
a compact buff clayey rock, generally soft, that occurs, for example, jusi
below the lip of the cliff immediately north of the Castaways’ Depot, as
well as at other places, and seems to be sediment associated with the
keratophyres.

REFERENCES.

A chronological list of investigations into the natural history of the Three
Kings Group appears in Records of the Auckland Institute and Museum, Vol. 3,
Nos. 4 and 5, 1948, pp. 191-193, and contains references to most scientific papers
relating to the group. Two papers only need be mentioned here, the first of which
has unaccountably been omitted from the above-mentioned list.

BARTRUM, J. A., 1936. Spilitic Rocks sn New Zealand. Geol. Mag. 73, 414-423.

CHEESEMAN, T. F., 1891. Further Notes on the Three Kings Islands. Trans.
N.Z. Inst., 23, 408-424 (with map by S. Percy Smith).

PLATE 8.

MAP OF

GREAT ISLAND, THREE KINGS GROUP

made by M.H Battey
incorporaling observations by £.G.Turbott & detail from
numerous photographs.

Scale ue
yards 440 400 300 200 100 © Ya mile
a WE 3 LE eo es Sse ri
tt Castaway Depot G\ so ee \ yg —> ez
a Camp 4 vA , 242° 4 East Point
259 / ge
> GP. ,
@ “ial (~— et 5 met a
Qv ; “ O ry A \ 33 *
S 4 32
. Re” “h yy a ie N
Y ra toz * : Th58
“ Ved SA Bap (02 \
‘4 ~ tk = \ \ > 246 ANS
v oa Sy a I
eo WH 55° = z
i so ioe tse ee ain Lg Farmer
<7) 7, ” J ees ays ‘\ Ay 5 ier
\ Hix ve ney oe: 200 nt Ms) North East Arch gata
A Vv) Rh DD OCcKS
: . on oo
i 6 wt ‘ (Z Soe .) vf fy IT) s, SI. mm’;
; } W WY oy “600 AGM aa Z| ty
‘ \ ( | SN | Ma MK is va git 2 SB -
ANA Za SE
C7

a
on Son a: SS .
7 sss ‘ — -
} ‘ \ ) eos
o3 \ \ Sy 5 Peas ts: 245
N\ 56 * te
ae ae

{ er S
‘\ fon “ a
\ ( \ <- Tasn ,

‘e608 Bald Hill »
SOL! 5714) 475 2), 4,
SS: [3 425 a C4
at SS 400
Lee ~ SS ¥ Rr OSE

The Cove
a 4 cs f
A @455 t, SS
South Point @.~—=3% SE

January /49s!

— EE eee

H
,
4 ’
V4
ea
. “s
roan
t
*
.
,
vi
rn
’
:
‘
.
’
= o
he
+ oD
‘
2 *

— EE eee

H
,
4 ’
V4
ea
. “s
roan
t
*
.
,
vi
rn
’
:
‘
.
’
= o
he
+ oD
‘
2 *

PLATE 9.

| a
GEOLOGICAL MAP j: BDO
OF
= Smooth o*
\
GREAT ISLAND, THREE KINGS GROUP sand - covered &
Scale fe sloping surface S
yards 440 O Yq mile 3 325-250 Ft, re
fs +—_—_- — (5
ao Greywacke
B° b Y
a
a sandstone
ages plocks (oe :
ae eS
le e —_- 250
eae SiS
+ e ° aad Messer Ta Sot
. * + e Ki B e _—~ ee
: + Sin ee
— Greywacke » >/600ft
e e +." <
e SS) os
e + Y ° 25-30 = h massive ° 45 fi
oe 4 ps : Keratophyre brece/ated a *225TE.
° ‘
: e + fe ——. Rey naeke Basic variolite, probably < 2soft.
‘ ® Se < ==, (Aypotheti- spilitic
¥ a = cal) Keratophyre a got,
©
° Spilite; pillowy in massive + >r700f
places amygdal. x

Geological boundaries (a) located ie a

(bi snferred Bye es
Dip - angle in degrees WA
37

DOA x ¢
preeg 4

Elingamita (Myrsinaceae) a New Monotypic
Genus from West Island, Three Kings.

By G. T. S. BAYLIS, University of Otago.

In January, 1950, Major M. E. Johnson succeeded after several
unsuccesstul attempts in landing on West Island at a point from which
it was possible for him to climb the cliffs to the vegetation on the upper
slopes. There he made a comprehensive plant collection, which he
handed to me. It established that this small island though steep and
windswept has a considerable flora and is of much ecological interest.
In January, 1951, thanks to the skilled boat work of Mr. E. Beaver, of
Whangaroa, I accompanied Major Johnson on a second visit. A sub-
sequent paper will give an account of the vegetation of West Island
together with that of the other outliers of the Three Kings Group.
he present purpose is to describe a new tree first collected in fruit by
Johnson and found to be both flowering and fruiting in the following
January. Meanwhile, fruits had been submitted to the Herbarium,
Royal Botanic Gardens, Kew, whence Dr. Melville reported that they
appeared to be referable to the Myrsinaceae. The floral structure has
confirmed that the tree belongs to this family but the monograph by
Mez (1902), which remains the standard treatment, requires the erec-
tion of a new genus for its reception. The flowers have now been
examined at Kew also, and I am indebted to the Director, Sir Edward
Salisbury, for a report that “they do not seem to fit any genus of
Myrsinaceae hitherto described.”

ELINGAMITA n. gen.

Flores hermaphroditi, 4-6 meri. Sepala valvata, punctata, libera.
Corolla brevior sepalis, tubulosa, truncata vel margine obscure lobulato.
Filamenta corolla fere omnino libera et eadem in maturitate bene
excedenta. Anthera elliptica dorso affixa, tota longitudine dehiscenta.
Ovarium ovoideum in stylum crassiusculum attenuatum, stigmate punc-
tiformi. Ovula perpauca in parte superiore placentae uniseriatim
immersa. Fructus drupaceus, globosus, apice stylo persistento mucrona-
tus, endocarpio crustaceo, monospermus. Albumen sphaeroideum,
corneum, album, pulvino lato e placenta formato sedens. Embryo
cylindraceus, obliquus vel transversus, paululo cervatus.

Arbor foliis alternis, exstipulatis, simplicibus, punctis. Inflores-
centiae terminales, paniculatae, primo obtectae bracteis latis caducis.

A ceteris generis distat corolla deminuta sepalis breviore, omnino
vel fere omnino tubulosa, Typus &. johnsoni n. sp.

Flowers hermaphrodite, 4-6 partite. Sepals valvate, punctate free.
Corolla shorter than the sepals, tubular, truncate or with an obscurely
lobed mouth. Filaments almost wholly free from the corolla and in
mature flowers much exceeding it. Anthers elliptical, dorsifixed, splitting
down their whole length. Ovary ovoid narrowed into a rather stout

100 BAYLIS.

style with a stigmatic pit at the apex. Ovules very few immersed at
one level in the upper half of the placenta. Fruit a drupe, globose,
crowned by a peristent style, one-seeded with a brittle endocarp. [ndo-
sperm rounded, horny, white, seated on a broad cushion of placental
tissue. Embryo cylindrical, oblique or transverse, almost straight.

A tree with alternate, exstipulate, simple, gland-dotted leaves.
Flowers 1m terminal panicles concealed in bud by broad deciduous bracts.

Differs from all other genera in possessing a reduced corolla shorter
than the calyx and wholly or almost wholly tubular. Type FE. johnson
Nn. SP.

Elingamita johnsoni n. sp. PI. 10, figs. 1, 2. Text figs. 1-6.

Arbor glaber cortico leve, foliis coriaceis, integris, utrimque mani-
leste pinnate venosis, obovatis, 1n petiolos brevissimios eradatim con-
tractos, c,100-180 mm. longis, ¢.45-90 mm. latis, Paniculae florales
lutescentae, c. 50mm. longae ex aequo latae, fructiferae c.100 mm.
Flores pedicellis c.5 1mm. longis, corollis ¢.2.5 mm. latis, Ovula 2-4.
Fructus ruber, c.17 mim.

West Island, Three Kings, New Zealand.

A glabrous tree with smooth bark. Leaves leathery, entire, on both
surfaces strongly pinnate-veined, obovate, about 4-7 inches long and
1-33 inches wide, narrowed gradually into very short petioles. Panicles
in flower yellowish about 2 inches in length and breadth, in frmt 4
inches. Flowers on stalks about + inch long, corolla about 1-10th inch
diameter. Ovules 2-4. Fruit ree about 2- 3rd inch diameter.

The very reduced tubular corolla lacking well defined free lobes
and shorter than the calyx appears to be unique in the AM yrsinaceae. The
placenta has the uniseriate ovules of the tribe Myrsineae but the terminal
paniculate inflorescence, non capitate stigma and well developed filaments
separate Elingamita substantially from Myrsine, Suttoma and Rapanea,
the genera to which the other members of the family indigenous in New
Zealand have been, at one time or another, referred (Hosaka 1940,
Allan 1947, Oliver 1951). Closer allies are presumably to be sought in
the Pacific Fenda and Malaya, but both the genera of this region with
the same form of inflorescence, placenta and “stigma (Labisia and Tet-
rardisia) are peculiar monotypes. Geographically, t the nearest genus
with a paniculate inflorescence is Tapeinosperma, which is well developed
in New Caledonia and which has moreover a punctiform stigma. How-
ever, these resemblances are offset by a pacenta less like that of
Elingamita than is that of M/yrsine. Unfortunately, the vegetative
anatomy of the family does not appear to be sufficiently well known for
it to be employed as guide to the relationships of the new genus.

Elingamita johnsont 1s represented by perhaps a dozen trees on
West Island, but they are members of a windswept forest scrub in
which the true habit cannot be seen. Like the other Three Kings mono-
typic genus Plectomirtha Oliver it has the general aspect of a karaka
(Cor ynocar pus laevigata). The cream-coloured inflorescences are not
very conspicuous, but the ripe fruit is brilhant red and produced in
large bunches. The tree thus appears well worthy of pilere ation, and it

Fig,

ig.
lig,
Fig,
Tig.

Big,

tsi

Elingamita New Genus. 101

Flower x 10.—a, bud; b, mature; c, portion of calyx, corolla and
androecium.

a, entire; b, in L.S. of ovary.

Placenta x 44.
Fruit x 1.5.—a, entire; b, L.S. (persistent calyx not shown).
Embryo x 2a, in situ in the endosperm; b, dissected out.
Young inflorescence showing caducous bracts x 1.75.

Underside of leaf x 0.5.

102 BAYLIS.

is a pleasure to dedicate the species to its discoverer, Major Johnson, on
whose enterprise, skill and persistence the exploration of the smaller
islands of the Group has greatly depended. The generic name com-
memorates the tragic wreck in 1902 of the inter-colonial passenger
steamer Elingamite beneath the cliffs on which the tree grows.

I have to thank the late Major G. A. Buddle for the negative of
Plate 10 and Dr. M. Holdsworth for all the drawings in this paper.

REFERENCES.
ALLAN, H. H., 1947. Notes on N.Z. floristic botany—No. 8, Trans. Roy. Soc.
N.Z., 76, 596.
GUILLAUMIN, A., 1948. Flore de la Nouvelle-Calédonie—Phanérogames.

HOSAKA, E. Y., 1940. A revision of the Hawaiian species of Myrsine ( Suitonia,
Rapanea), (Myrsinaceae), B.P. Bishop Mus. Occ. Pap. 16, 25-76.

MEZ, C., 1902. Myrsinaceae, Das Pflansenreich, 9, IV, 236

OLIVER, W. R. B., 1951. The Flora of the Three Kings Islands: Additional
Notes: with note on Suttonia. This issue.

PLATE 10

Herbarium material: 1, in flower; 2, in immature fruit.
Elingamita johnsont n, sp.

oa -_" ‘nhs

a -

a
=

103

Incipient Forest Regeneration on Great Island,
Three Kings Group.

By G. T. S. BAYLIS, University of Otago.

A previous paper (Baylis, 1948) attempted to trace the history of
the vegetation on Great Island and reached the following conclusions.
The bulk of the island was originally covered by mixed coastal forest.
This was destroyed by a long period of Maori occupation, through which
most of the component species probably persisted as single specimens
and small groups. <A period of regeneration, probably impeded to a
small extent by goats, followed the departure of the Maoris about 1840,
so that by 1889 a variety of shrubs and small trees mingled with a
general covering of Leptosperimum. Most ofthese failed to survive
when the goat population grew large, and probably for about half a
century prior to their complete destruction in May, 1946, these animals
50 thoroughly searched the island for food that trees other than two
unpalatable species of Leptospermum were rarely, if ever, able to re-
establish themselves from seed. Latterly L. ericoides (kanuka) had
been more successful in this respect than L. scoparium (smanuka), so
that when the goats were destroyed, kanuka covered almost the entire
island, the principal exception being a piece of Zoisia grassland. The
trees of the mixed coastal forest persisted only as scattered individuals
and small groups among which more than half the species were repre-
sented by five trees or less. At least two of the presumed components
of this forest (Meryta sinclairii and Elingamita johnsoni) had dis-
appeared entirely.

In 1946 most of the kanuka on Great Island was senescent and
open. Since its seedlings are very intolerant of shade its re-establish-
ment never occurred until the parent plants had died, and not always
promptly then, sometimes through the presence of sedge undergrowth,
sometimes—notably on the eastern plateau—because the soil was by this
time windswept and eroded. The opportunities for entry of shade
tolerant seedlings and even for light-demanding seedlings hardier than
kanuka were thus extensive as soon as the goats ceased to make their
establishment impossible. It did, however, appear that on the eastern
plateau and elsewhere (e.g., the southern scrub area) the soil might have
deteriorated overmuch for other trees, both through poor quality of
Leptospermum humus (Scott Thomson and Simpson, 1937) and because
of erosion in the replacement phases.

An account of a brief survey of Great Island in December, 1947,
has already been published (Baylis, 1948). This paper sets forth
observations I made in January, 1951, while encamped for six days on
the island with a party organised for the Auckland Museum by its
Director, Dr. Archey, and Ornithologist and Entomologist, Mr. Turbott,
to whom I am much indebted for this opportunity. A re-charting of
Turbott’s (1946) quadrats was simultaneously undertaken. These
charts are presented and discussed in an accompanying paper (Holds

104 BAYLIS.

worth, 1951). Some results of a short visit that I was able to pay in
January, 1950, through the kindness of Major M,. E. Johnson are
included here. Expenses have been met by a grant from the Research
Fund of the University of New Zealand.

SEEDLING ESTABLISHMENT UP TO JANUARY, 1951.
Trees with seedlings intolerant of shade.

Only the species of Leptospermuim and Metrosideros excelsa (pohutu-
kawa) appear to fall in this class.

(a) Leptospermuin

Shade tolerant tree-seedlings are not yet sufficiently widespread to
cause a general halt in the cycle of death and re-growth of kanuka
(L. ericoides). Nevertheless, over much of the island it no longer
continues. In all moist places herbaceous undergrowth 1s tall and dense
and promises to deny the ground to kanuka seedlings over much wider
areas than hitherto, so that, though open patches of rank sedge or
Colensoa are still few and small, they seem certain to increase (figs.

1 and 2).

The eastern plateau is dry, but here kanuka 1s encountering compe-
tition from manuka (L. scoparium). This was the area in which the
latter species was in 1946 more plentiful than in any other part of the
island. Nevertheless, it was much less in amount than kanuka, which
appeared to be replacing it. It was here also that soil deterioration was
most obviously retarding re-establishment when old Leplospermaum died,
so that there were many open places. Since removal of the goats there
has been no obvious increase in the rate of colonisation of these spaces
by kanuka, but many of them have filled with manuka seedlings (fig. 3).
This suggests that manuka is the species better adapted to maintain
itself on the drier parts of the island but that in the seedling stage it
was the more liable to damage by goats. There remains little reason to
doubt that considerable replacement of manuka by kanuka accompanied
growth of the goat population between 1887 and 1945, and that the
reverse change is now under way.

The Zoisia area is also a dry one and the grass has not grown too
long to prevent a general invasion of the sward by kanuka seedlings
(fig. 4). This is in accordance with the familiar invasion of mainland
pasture by Leptosperinum which occurs when stocking is inadequate
(e.g., Levy, 1949), and the fact that L. scopariuim is not present here
also is accounted for by the remoteness of any seed source.

(b) Meatrosideros excelsa

The pohutukawa is demonstrating its ability to invade a grass turt
on part of the Zoisia sward, over about a quarter acre of which its seed-
lings promise to compete with kanuka for dominance (fig. 4). A few
dozen plants have also established themselves on bare soil between
kanuka bushes at one place on the eastern plateau (fig. 5). To a tree
of this colonising power Bald Hill, the entire Zoista area, the grassy
interspaces of the adjacent kanuka scrub and all the open strips of the
eastern plateau should be available; nevertheless, most of the pohutu-
kawa standing on or close to these areas have as yet no seedlings
associated with them.

Forest Regeneration, Groat Island. 105

Trees with shade-tolerant seedings.

The remaining trees have seedlings that are in some degree shade
tolerant and can establish themselves under a mature or thinning canopy
of kanuka—in fact, none are found making their initial growth wholly
without such protection. These species are components of mixed
coastal forest which is believed to be the climax vegetation.

(a) Cordyline australis

Cabbage trees were well dispersed over the island in 1945 and were
flowering abundantly. Seedlings are now common in kanuka forest and
shrubland, but since they remain unbranched for a long time and develop

|
only a tufted crown ‘this species may not exercise much controlling
influence on others.

(b) Aeryta sinclair

Puka seedlings are now conimon on the slopes which face 5.E. Bay
east of the depot, i.e., in the vicinity of Quadrat I (fig. 6) and adjacently
in the lower part of Castaway Valley. They grow rapidly and cast a
heavy shade, so that it is possible that a puka canopy will develop over
some of this area. Elsewhere under comparable conditions of soil and
shelter this species is seen only occasionally. It is rare in Tasman
Valley, and there are a few young plants under Hapuka Point and in
the shelter of pohutukawa below Bald Hill.

Even Cheeseman did not record Meryta on Great I., which means
that it is over 50 years since seeding trees grew there. If the possibility
of buried seeds lying dormant for over half a century cannot be dis-
missed (Crocker, 1938), at least some unusual soil disturbance would
be necessary to break the dormancy of such substantial numbers. It
seems beyond reasonable doubt that puka forests on North-East T. and
South-West I., particularly the former because of its proximity, are
the main seed source and red-billed gulls (Larus novaehollandiac),
whose droppings on Great I. have been found to contain Meryta seed,
are the principal carriers. Many nest on the island’s coast line and they
are constantly to be seen hovering about the forest roof, a habit ascribed
by Turbott (1951) to their feeding on cicadas. Admittedly, the part
of Great I. closest to North-East I. is virtually devoid of Meryta seed-
lings. However, the depleted aspect of this eastern plateau area has
already been mentioned, and the poor growth of the only young Meryta
found upon it confirms the impression that it is at present inhospitable
to that tree.

The two moribund groves, and the few single trees along the
western side of South East Bay are the nuclei of thickets of seedlings
up to 6ft. high. That from the eastern grove extends for about 5Oyds.
up the slope, that from the western grove for about 100 yards—in
both cases beneath old kanuka (fig. 7). Occasional seedlings are seen
much further from these sources, particularly on the northern side ot
South East Bay, but the species has yet to enter Tasman Valley.

(d) Pittosporum fairchildn

The large specimen close to the western Brachyglotiis grove has
several seedlings in its vicinity (fig, 8),.and though it 1s dispersed by

106 BAYLIS.

birds and the Brachyglottis by wind, occasional young plants of both
are met with comparable frequency on the high lying land between the
parent trees and the summit of the island. The remaining Pitlosporum
trees are close to Quadrat I and from these there is little spread as yet.

(e) Cyathea medullaris, Paratrophis simithii, Melicytus ramiflorus,
Coprosma macrocarpa, Litsaca calicaris, Melicope ternata.

Old trees grow at intervals among the Tasman Stream and young
plants of all, and particularly of Cyathea (fig. 9), are now common in
the valley bottom. A similar spread of Litsaca, Paratrophis (fig. 11),
Melicope and Melicytus centres round the specimens in the valleys east
of the depot. So far Litsaca alone 1s spreading actively from the clump
of mixed trees in Castaway Valley, and here its exceptional shade toler-
ance is apparent, seedlings occurring beneath kanuka too dense to allow
of any other undergrowth. Cyalhea has not yet spread to Castaway
Valley, but a few seedlings of Coprosma macrocarpa have appeared
there.

(f{) Htemerliodendron brunoniana, Planchonella costata var, austro-
montana (Sideroxylon novoselandicuin), Olea apetala, Cory-
nocarpus laevigata, Hedycarya arborea, Alectryon grandis,
Vitex lucens

These are all species with very few parent trees on the island and
seedlings at present confined to the immediate vicinity of one or more
of the specimens. The minimum increase is shown by Vitex. The two
trees fruit well, but so far only’ one seedling has appeared. No seed-
lings of Alectryon were found by either of the trees on the 1945 map,
but in its preparation a small gulley on the northern side of the northern
headland of Tasman Bay was overlooked, and this proves to contain two
further trees of Alectryon together with two Planchonella and six
Brachyglottis. Alectryon seedlings are plentiful for about 50 yards
along the gulley bottom. Planchonella has spread rather less, and
Brachyglottis somewhat further.

(g) Rapanea (Suttonia) dentata, Plectomirtha baylisiana

Seedlings were not seen beside any of the existing trees of Rapancea.
It has only been observed fruiting at Hapuka Point, where there were
two trees together, one of which is now dead. Probably it is, like several
of the genus, dioecious. Surprisingly, however, a few seedlings which
seem to be of this species occur about the camp site in Castaway Valley
—a full quarter mile from any mature tree.

The sole known tree of Plectomirtha has failed to establish any
seedlings on the stony ground which surrounds it, although the adjacent
Olea has done so. The nature of the ripe fruit is stil unknown, and
it may be that none has matured. Certainly many of the pamicles decay
soon after flowering.

Lianes.

~ Old vines of Clematis paniculata (C. indivisa) and Tetrapathea
felrandra in the valleys east of the depot have made rampant growth,
and produced exceptionally large leaves, This applies especially to
Tetrapathaea, which has also established abundant szedlings there.

Forest Regeneration, Great Island. 10/7

Parsonia heterophylla and Clematis in the Tasman Valley have not as
yet spread conspicuously, and there is no evidence that the single
Tecomanithe speciosa vine has seeded though it is thriving and has layered
itself at one point well removed from the old base.

Soft wooded species.

Three quickly maturing soft-wooded species have become con-
spicuous—Entelea arborescens, which is a small tree, Solanum aviculare
var.albiflora, which is a shrub attaining a maximum height of about
ie)

& feet, and Colensoa physaloides, which is a large herb forming

hydrangea-like clumps up to 5 feet high (figs. 2,9, 13). Colensoa was
abundant in. 1887 (Cheeseman, 1888) and a few plants persisted in
damp places in 1945. Now it forms large patches both in the Tasman
Valley and on the slopes which face S.E. Bay, east of the depot. The
Solanum is most plentiful over the area in which Meryta is most abund-
ant, and it seems likely that its seed came by the same means though its
presence on North East I. is not established. In addition, it occurs by
the western Brachyglottis grove (fig. 8). Entelea was collected on
Great I. as late as 1934. Transport of its large burred fruits from
adjacent islands is hard to imagine, and in the light of published work
(Millener, 1949) it may be that the odd plants still appearing come from
dormant seed, while the small thickets—one in the Tasman Valley and
one by the western Brachyglottis grove—are the product ot precociously
fruiting plants.

Other shrubs and herbs.

When goats browsed beneath the kanuka in most places there was a
turf in which Gnaphalium collinum, Lagenophora pumila, Cotula
ausiralis, Haloragis procumbens and Centella asiatica were prominent,
These creeping plants have been displaced by a vigorous growth of the
accompanying grasses coupled with a large increase in the frequency ot
some hitherto not plentiful, ic, Agropyron kirku, Microlaena stipoides,
Poa anceps, Poa seticulmis. Quail are common and may have aided
their dispersal. Two liliaceous herbs, Dianella intermedia and Arthro-
podium cirrhatum, have also become widespread. Arthropodimum was
previously restricted to inaccessible cliffs and Dianella, now specially
abundant on the eastern slopes of Tasman Valley, was found chiefly on
the eastern plateau.

A list of herbs which colonised bare soil and the finer scree material
on cliffs soon after removal of the goats has already been given (Baylis,
1948). It is apparent that on sea-cliffs the prostrate ecotype of
Myoporum laetum is gaining possession, but on Bald Hill (fig. 10) the
only woody plant so far established is the endemic Hebe insularis. No
form of Myoporum capable of developing into a tree is present on Great
Island. Seedlings often appear on the hills, but their growth is spindly
and they soon die. |

Additions to the species list since December, 1947.

In December, 1947 (Baylis, 1948), 18 species were collected that
had not been observed while goats were plentiful, though the possibility
of their being present in out of the way places could not be precluded.

108 BAYLIS,

The following are now added with the same reservation. The date
(1887) or (1889) means that the species was recorded by Cheeseman
(Oliver, 1948), though it is possible that 1889 records relate not to
Great I. but to South-West I.

Anagallis arvensis L.

Arthropteris tenella (Forst. f.) Sm. (1889).
Bromus catharticus Vahl.

Calystegia soldanella R. Br.

Calystegia tuguriorum (Forst. f.) R. Br. (1887).
Erechtites atkinsoniae F. Muell.

Erigeron canadensis L.

Flierochloe redolens Kk. Br.

Paspaluin scrobiculatwmn L. (1889).

Polystichum richardi (Hook.) Sm. (1887).
Sarcochilus adversus (Hook. f.).
*Solanum aviculare Forst. f. var. albiflora Cheesem. (1889).
Trifolium glomeratum L.

Uncinia uncinata (L) Kirk (1887).

Veronica plebeia R. Br.

DISCUSSION.

All species that were expected to occupy more ground after removal
of the goats have begun to do so except the two rarest, Tecomanthe and
Plectonurtha, each of which is still represented by a single plant that
may not have fruited.; The short life cycle of the herbs has enabled
some to spread extensively, the most striking examples being Colensoa,
Arthropodium, Dianella and two grasses, Agropyron kirkii and Micro-
lacna stipoides, both first noticed in 1947 and now widely diffused. The
distribution of woody species, however, remains in general closely related
to the seed source. Even Meryia and Solanuin which seem to have
come from a distance show this. Though birds appear to have carried
these species from other islands they have failed as yet to bring in
Macropiper in comparable quantity. The only markedly discontinuous
dispersal observed upon Great I. itself was the presence of Kapanea and
Coprosma macrocarpa at the camp site in Castaway Valley. As the
headquarters of collectors this place is suspect. Striking examples of
persistent localisation are Cyathea and Tetrapathaca, both still restricted
to the valleys containing the parent plants, but spreading abundantly
there.

Chance factors of distribution promise to exercise a major influence
on the composition of the new forest. These include the position of the
old forest relics; how far the ground immediately about each tree was

“Seedlings seen in 1947 were identified as the type, but the flowers prove to be
paler, the habit more slender, and the leaves unlobed except in the seedling
stage. Cheeseman’s record is simply S. aviculare, but he did not erect the
variety until 1920.

+Mr. J, Horton has successfully propagated Tecomanthe trom cuttings at the Plant
Diseases Division, Auckland. Further attempts will be made to establish
Plectomirtha by this means,

Forest Regeneration, Great Island. 109

suitable for seedlings and thereby for an early increase in seed produc-
tion close to the original centre; whether or not in the first few years
any seedlings arose at a distance to become separate centres of seed
supply.

A character which should assist some species is precocious fruiting.
The oldest seedlings of the fast growing trees Brachyglottis, Meryta and
Hiemerliodendron (fig. 12) are already themselves in flower, the first
two sometimes bearing an inflorescence when only two feet high.
Another distributional accident which may prove significant is the extent
to which ground adjacent to parent trees came early within the orbit of
aggressive temporary occupants, In this role Solanum aviculare var.
albiflora, Colensoa and Entelea are becoming conspicuous. Even a dense
sedge growth, since it is capable of excluding kanuka, can in the absence
of other seedlings cause local deforestation.

It will be interesting to observe which tree first secures all stations
suited to it. The progress of Meryta in view of its abundance on North-
East I. and South-West I. may be rapid when the seed supply is
augmented by a substantial seed crop on Great I. itself. However, the
exploration of these lesser islands is affording evidence that pure stands
of this araliad are not, as was earlier suggested (Baylis, 1948), a stable
climax even close to the sea.

REFERENCES.

BAYLIS, G. T. S., 1948. Vegetation of Great Island, Three Kings Group, Wee.
Auck. Inst. Mus. 3, 239-252.

CHEESEMAN, T. F., 1888. Notes on the Three Kings Islands, Trans. N.Z.
Inet., 20, 141-159.

CROCKER, W., 1938. Life-span of seeds, Bot. Kev, 4, 235-274.

FOLDSWORTH, M., 1951. Effect of goats on Great Island, Three Kings: The
Permanent Quadrats Resurveyed. This issue.

LEVY, E. B., 1947. The Conversion of Rain Forest to Grassland in N.Z., Tuatara,
2, Ofel -

MILLENER, L. H., 1947. A Study of Entelea arborescens, R. Br. (‘“Whau’).
Part I. Ecology, Trans. Roy. Soc, N.Z., 76, 267-288.

OLIVER, W. R. B., 1948. The Flora of the Three Kings Islands, Rec. Auck.
Inst. Mus., 3, 211-238.

OLIVER, W. R. B., 1951. The Flora of the Three Kings Islands: Additional
notes: with Note on Suttonia. This issue.

SCOTT THOMSON, J., and SIMPSON, G., 1937. Notes on Hydrogen-ion
Concentration of Forest Soils in the Vicinity of Dunedin, New
Zealand. Trans. Roy. Soc. N.Z., 66, 192-200.

TURBOTT. E. G., 1948. Effect of Goats on Great Island, Three Kings, with
Descriptions of Vegetation Quadrats. ec. Auck. Inst. Mus., 3,
253-272.

TURBOTT, E. G.. 1951. Notes on the Birds of the Three Kings Islands. This
issue.

= a taf Oy

ser 2 ee e - H
Mo a ee
“tel < 4 “ ryt eer. a
; is = ee

aries
ze,

+4
pet
fri. Shy
»

a - Ea
tls: ae Seer ee yan e
Pepe ths Soe a : A ;

<2 lege

ee,

ae
oe | lee
eee

ha at or ane

a el eee

Lar aS 2
ie

rl Sj
a

care)

111

The Flora of the Three Kings Islands:
Additional Notes: with Note on Suttonia.

By W. R. B. OLIVER, Wellington.

The following notes are intended to be read with the paper I pub-
lished on The Flora of the Three Kings Islands in Records of the
Auckland Museum, vol. 3, pp. 211-238, 1948.

Alectryon grandis Cheesem. (Oliver, l.c., p. 226).

Mature fruit similar to that of A. excelsum but larger, globose,
brown, pubescent, a lateral longitudinal flange extending to but decreas-
ing in prominence towards the apex; length 18mm,, diameter 12 mm.
Seeds black, shining. The fruits, usually, perhaps normally, are joined
together in twos on the flanged sides for practically the whole length.
The lateral flanges, except at the apex, are accordingly suppressed in the
joined fruits.

Clematis paniculata Gmelin, in Linn. Syst. Nat. ed. 13, 1791, replaces
C. indivisa Willd., 1800.

Rehder, Jour, Arn. Arb., 26, 70, 1945, has shown that both these
names were founded on C. integrifolia of Forster (not Linnaeus, 1753),
and as Gmelin’s name is earlier than Willdenow’s it should be accepted
for our species.

Planchonella costata (DC) Lam, var. austro-montana Lam, blumaa,
5, (1), 5, 1942.

This name should replace Sideroxylon novo-selandicum of my
former paper (J.c., p. 231). Lam quotes Pierre as the authority for the
combination Planchonella costata but gives what appears to be only a
herbarium reference. Var, austro-montana is the typical variety of the
species.

Rapanea dentata (Oliver) n. comb. for Suttonia dentata Oliver, Rec.
Auck. Mus., 3, 320, 1948.

The petals are united at the base, not free as I described them;
accordingly, the species falls into the genus Kapanea. Although there
is a Myrsine dentata Spreng., it is apparently a synonym of a species of
Rapanea, but it has never been transferred to that genus so the combina-
tion Rapanea dentata is not preoccupied.

NOTE ON SUTTONIA.

The genus Suttonia A. Rich,, if’ based on the free petals, cannot be
maintained because of 16 species in the Hawaiian Islands, undoubtedly
forming a natural group of Kapanea facies, which have the petals free
or united. In several cases both conditions are found in the same

112 OLIveER.

species. Hosaka (Occ. Pap. Bishop Mus., 16, 28, 1940) argues for the
inclusion of both Rapanea and Suttonia with Myrsine, but M. africana,
its type species, has the filaments of the stamens united into a flange on
the inside of the corolla, a structure not at all like any seen in either
Rapanea or Suttoma. I would agree with Hosaka in uniting Rapanea
and Suttonia, but not in merging the combined genus in Myrsine. The
change here recommended involves the following new combinations :

Rapanea australis (A. Rich.) Oliver, n. comb. for Suttonia australis
A. Rich., Fl. Nouv. Zel., 349, pl. 38, 1832.

Rapanea chathamica (Muell.) Oliver, n. comb. for Myrsine chathamica
Muell., Veg. Chatham Is., 38, pl. 7, 1864.

Rapanea coxu (Ckne.) Oliver, n. comb. for Myrsine corti Ckne., Trans.
N.Z. Inst., 34, 318, 1902.

Rapanea montana (Hook. f.) Oliver, n. comb. for Myrsine montana
Hook. f., Handb. N.Z. Fl., 184, 1864.

Rapanea divaricata (Hook. f.) Oliver, n. comb. for Myrsine divaricata
A. Cunn., Ann. Nat. Hist., 1, (2), 47, 1838.

Rapanea nummularia (Hook. f.) Oliver, n. comb. for Myrsine num-
mularia Hook. f., Handb. N.Z. Fl., 184, 1864.

PLATE 11.

Carex

1. Re-establishment of kanuka checked by sedges (Scirpus nodosius,
testacea)—Jan., 1950.

2 Re-establishment of kanuka checked by Colensoa. The Meryta seedling right
of centre was in flower—head of S.E. Bay, January, 1950.

Manuka seedlings, E. plateau. All surrounding bushes are kanuka—Jan., 1951.

Cs)

Vik tai Me An is ee
' ne at yi at i
a 1d ren, meal
: i pie au rad wet
Ve Poae

= +3
fae | 7 bs
a ir

tee

_ am vf ‘

5 et, ee
~ e s

;

21K
ee
‘ar

;

eae Je
[a
pee

' al glege
o oy
eis
=
WF
)

7 hon
eee ee

ty : me
a a a
hoe IL or
A, ae re
ea Sat jt tei

cyt

Kanuka and pohutukawa seedlings on the Zoisia sward——Jan., 1951.

Pohutukawa2 seedlings on bare soil

E. plateau, Jan., 1951.

Meryta, Melicytus and Pteris comans—edge of Quadrat J, Jan., 1950.

Brachyglottis beneath kanuka near eastern grove

Jan., 1951.

PLATE 13.

S Pittosporum, Solanum and Brachygiottis (right) beneath dead kanuka—near

Jan., 1951.

Colensoa beneath kanuka —Cordyline canopy—Tasman Valley
bottom—Jan., 1951.

G. Scirpus nodosus and Cyperus tussocks, mats of Disphyma and Gnaphalium
Intco-albuam—Bald Hill, Jan., 1951.

western grove

G. Cyathea and

PLATE 14.

11. Litsaea (foreground) and Paratrophis (right and rear) seedlings in Tasman
streambed—Jan., 1951.

—s
IO

The figure stands behind a flowering seedling ot Hiemerliodendron and points
out one of the largest Litsaea seedlings for comparison—near Quadrat I,

ati, 1931

+3 Colensoa colonies and a single Entelea (behind the figure)—Tasman valley
bottom, Jan., 1951.

113

Effect of Goats on Great Island, Three Kings:
The Permanent Quadrats Resurveyed.

By M. HOLDSWORTH, University of Otago.

The Permanent Quadrats in 1948.

On 6th October, 1948, a landing on Great Island was made from
the launch “Alert” but only four hours were spent ashore. The party
included E. G. Turbott and L. C. Bell.

Insufficient time was available before dark to take a complete series
of photographs corresponding to those of 1946, but representative ones
were secured on each of the quadrats—Plates 15-17, figs. 1-6.

The following observations are a condensation of Turbott’s field
notes on Quadrats I and II with the addition of information supplied
by Bell on Quadrat IIT.

Quadrat I

The most striking addition to the cover of Quadrat I was drifts of
Colensoa physaloides, a plant not represented at all on this plot before
the goats were exterminated. Tctrapathea tetrandra was another addi-
tion noted as widely distributed. Abundant new shoots had put out
from the lower parts of the trunks of Cordyline australis (especially),
Melicytus ramiflorus and Litsaea calicaris. Clematis indivisa seedlings
were noted adjacent to the groups of parent plants recorded in 1946.
[All the lianes of Clematis appeared to be of about the same age in
1951 and the groups did not appear to have extended their territory, so
these seedlings probably have not survived.]| Seedlings up to 3” ot
Meryta sinclairii and Brachyglottis arborescens were frequent, especially
about the position marked “P” in Pl. 24, fig. 18.

The four marked seedlings (see Turbott, 1948, p. 267) were
remeasured:

No. l. Melicope ternata 2’ 10” high
No. 2. Melicytus ramiflorus

(thought to be Litsaea calicaris in 1946) lV’ 4” high

No. 3. Melicope ternata 2’ 6” high
No. 4. Tetrapathea tetrandra

(thought to be Litsaea calicaris in 1946) 2’ 0” high

[In 1951, Nos. 2 and 4 could not be identified and probably have
not survived. The nearest plant to 2 was a Melicytus (1’ 9”), but this
was on the side facing away from the number. There was no seedling
at all adjacent to No. 4.|

Quadrat II

Changes on this plot were much less remarkable than on I, but
seedlings of Meryta sinclairti, Cordyline australis, Tetrapathea tetrandra
and Clematis indivisa were observed. [The last named did not survive
until 1951. |

114 Hortpsworte.

Among the herbs, Dianella intermedia had already established
several clumps and there were a few of Arthro podium cirrhatum. Both
these plants were not present in the quadrat in 1946. It was recorded
that the herbs, generally, were more flourishing than they had been
during the goat occupation. |

Quadrat III

Invasion of this grassland area by kanuka seedlings had begun and
the sward itself was longer than in 1946.

The Permanent Quadrats in 1951.

The opportunity was taken during the 1951 Auckland Museum
Expedition to the Three Kings Islands to re-map the permanent quadrats
laid down by E. G. Turbott in 1946 (Rec. Auck. Inst. Mus., 1948, q.v.).
Observers in the meantime (Baylis, 1951) have remarked the obvious
and rapid changes which have occurred in the vegetation of Great Island
since the extermination of the goats, but a remapping of the quadrats
establishes these changes in numerical terms—an interval of about five
years since the last census seems appropriate and it is hoped that it will
he possible to take subsequent censuses at the same interval.

Methods

When Turbott made the original observations on these plots, the
vegetation on each was so open that mapping could be accomplished by
sighting on to flag markers set up on the side lines. On Quadrat I, this
is no longer possible. The vegetation is already so dense that vision is
limited to a few metres and it was found necessary to lay a erid of
strings (two-metre squares were adopted) over the whole plot. Turbott
himself gave assistance in finding the boundaries of this quadrat, but
even so it is evident from a comparison of his Plate 50, fig. 20, with
Plate 23, fig. 17, that there are discrepancies in the positions of indi-
vidual trees between the two records. Errors thus introduced have
been allowed for in the discussion that follows.

Quadrat I

In Turhott’s photographs of this quadrat, taken in 1946, the vegeta-
tion looks old and decrepit for the trees are overaged and there is no
new growth below. Now, in 1951, although the condition of the kanuka
trees has still further declined, the forest looks quite flourishing, for
the upspringing of large numbers of tree seedlings has added a fresh
greenness to the plant cover. In some places the appearance of reju-
venescence has been intensified by vigorous growth of the passion vine,
and in others by the spread of Colensoa.

Along the western boundary, however, there is relatively littte
change in the appearance of the forest (Cf. particularly fig. 7 with Tur-
hott’s fig. 3). Roughly, this zone corresponds with a belt of cabbage
trees as shown in Turboit’s Pl. 50, fig. 20, and Pl. 23, fig. 17, here.
Pl. 24, fig. 18, in which the young trees are plotted, shows also a wide
band free of seedlings down the right half of the diagram. But this
area has nevertheless changed in general appearance since 1946, for
here has appeared a dense swathe of Colensoa along both branches of
the watercourse shown 1n Turbott’s diagrams.

Fig 1. Quadrat I, 6th October, 1948.
1948.) The sedges are more luxuriant than in 1946.

PLATE 15.

(Corresponds to Fig. 3A in Turbott,

Quadrat I, 6th October, 1948. (Fig. 4B in Turbott.) Vigorous
new shoots have grown from the bases of the cabbage trees. The
fern in the foreground is Pteris comans and the bush to its right

is Melicytus ramiflorus.

Photos: E. G. Turbott.

tft

= _— ae Sa oe | Ses a. — ah i! i a a
a a ee i Bt , c filial Me = ee ee a <i a + i ae ai ee one oe Sete See
etagre er ty hs — * a) a : .a - a a nels =o, 7 _ Eat . . Phy | - z “ oy ra i. Sheng Meet sy —"\8 it -
Spier 5 a SE * + ‘ ell oils : - ee ee ape ; =a i anes a ; ve el - et Cea as ae
ore irate, Ronee aoe ates

ae oy ae as oo - —

(= an id ane, ie ies ee Sa

ae ae ee tl

ro Ca ae - rt omg | i

“ey = eee

ae
a

iy
‘i os
apse re ay -

se an) Sa = ae cee
kl a ee - |
ee ce aie RR ecco bee oa a ie : : Saat oss ig: oe
L a Bae rn - ; a re aa 7 rt TP ci ae 2 ~ - - foe ie a Grea _— 4 ee 7 oe a ‘2 ime a con pees =a
cape > BEG mace ie eae ere oe ee tee, Oe OR aR es koa NM fhe ety OT Ftd
oh aa a rg cee Ca 3 | a euer, : | scans oe 2. Sena ge Ohh ee

%,
*
=
ere
a]
ch Ez
y i
Ry

bof ae 2. i. Lo ed i
att i Pe ee ee oe a
ao (lL r x i : 7 7 7 = oan hited a i
tS lar | - J ; nm =n, ee 7 4 gl ae — ae ee ee
Ws ace : ee a Par pu 4 : ; 7 “« ' ee ™ ae ee wi se rd) on?
tae ce — aah a 4 \ i a! -_ = ra 7 — = ne ta = = ree Ao . moms oe = er ag) gene a) 7 a ee - a = a. nar ean
7 aa I - as i a a ik A ba - = z thine oie _ = he Py - 7 og ae Coy | oe _ a = a - call - a!) ba “00
a = as ar Ciel ma ; Te 4 a _ - io ff ee cd : 2 = ih aes See as se re 4 r - ; aol, cs ee = Sree : 2 aK 7 . ean _ 47 a ame Pe —_
a el i ie — = > ee el cyte at 7 . he —! - a ais NR ee ee ag Se Be ee ee i ed ee
aan ee a a ell = ea a si’ iz a ey Ee =i) cian e ” a = _ - ae ce —~ gues. pe eS | pe 7 - a ri > ee “y ei: - a7 it ime =
es Agee 7 4 on — ae ris ae Pll ; fue he oe ha . ee Mee eo Peas od wy = Pa hk oe et i
ie en ee a Ry eat a = Pa ed ee oe Pe Le ee ae an oe Pel a oat ie a a ee ot ee ge ee ce Oe ee
or i reg ta es, - = ae 7 rol -_ eee ee aad i el lia che i Bt ft i : eur a 4 a i oe ees Sates — oe os ley ag ao ie ae aes 7 na a =é are oo a
>. a Siw . Z ee ae = ogy oi i he og _ on = , Tae ae 2. ~~ a a ae ‘| © ae ee Be a _ ioe os Se, = nc a a i LPL
ge a > a ae J a i a a —_ a hae he ae i er aes ee - a ee S| -- 7 imi oie a = ie
Cie lila, CET eE zz at ae ES 7 _ a ax = a fe a ae. a* Pa . es _ =) = coe ao = Ale be i i — ie t ie ee Se ee nok ert “= - —s Ea | ol oo
=. a ia . i = 7 i : = Jase a ae = or = ie! a =i =. a ies — Lae we vn oo ae ae Seles ee eh ie ed a Le a aa

a af po

np _
7 Fila
r" -
7
iti

pe
eee Sha

x
ae ame

ft Hom ee

: “ cae

a ‘. = a - = i :
nae 2 he i... er pee ; | = mak: a : . ike a

a
te

Ce ike open
aes Aes ae

es ee

_ cs
Soe i ws le
\ eae =
a 7 ,
'_ f
ma eee
a ae -
di a a. i :

7 a mi. =
eae au
eee ae |

. : Ca

eras a ii te a
oor on
tes i” ;
ee pall
peel are | Sia aes Pa tae = ‘hee
ee See a ee ee ee ,

Resurvey of Vegetation Quadrats, 115

The canopy

Analysis of Turbott’s diagram shows that, in 1946, the canopy was
constituted by 92 mature trees rooted inside the quadrat, of which 50
were kanuka and 29 cabbage trees, other species contributing but a few
specimens each. By 1951, the total number had fallen to 80. The
hgures are not exactly comparable, for there are some discrepancies
between the two records in the inclusion of trees on the boundary lines.
However, from their positions it is possible to identify 91 of the trees
present in 1946 in PI. 23, fig. 17, Of these, 19 (all kanuka) are repre-
sented now by dead boles, 1.e., in five years more than a third of the
mature kanuka trees have died.

Most of the gaps in the stippling representing the canopy in PI. 23,
hig. 17, can be related to the positions of the dead trees, the rest are
gaps already present in 1946, probably marking the positions of trees
which had died previously to Turbott’s census.

The undergrowth

Though the total number of trees forming the canopy has declined
by a fifth, the loss has been more than compensated for by the appear-
ance of seedlings. These are not, however, kanukas. On this quadrat
there are in fact no kanuka seedlings that can be said with certainty to
be new. It is true that the total of them is now 73 compared with 43 in
Turbott’s diagram, but there are still none outside the two thickets
shown there and the difference is almost certainly due to the difficulty
of defining a single pant in a thicket.

The other young trees and bushes recorded by Turbott were:
Melicope ternata, 15 seedlings; Coprosina rhamnoides, 9; and Myo-
porum laetum, 7. The present complement of this layer of vegetation
is shown in Pl. 24, fig. 18, and summarised in Table 1: It can be seen
that first place has now passed to mahoe, with almost as many seedlings

as the rest put together. Kanuka has fallen to second place and ngaio
has actually declined.

Geniostoma, Brachyglottis and Entelea were not present on the
plot in 1946, and Meryta, as far as can be known, was not present any-
where on Great Island,

Table 1.
QUADRAT I,

Trees forming the canopy:

1951 1946 (from

E.G.T.)
Leptospermum ericoides A. Rich. 34 50

Cordyline australis (Forst. £.) Hook. f. 33 29
Melicope ternata Forst.

Litsaea calicaris (A. Cunn.) Hook. f.
Paratrophis smith Cheesem,
Melicytus ranuflorus Forst.

at -— 3s hh ch Chi

Pittosporum fairchildu Cheesem.

116 HoLpswortH.

Young trees and shrubs:

Melicytus ramiflorus Forst. 238 0)
Leptospermum ericoides A. Rich. 73 43
Litsaea calicaris (A. Cunn.) Hook. f. 52 ()
Melicope ternata Forst. 51 15
Coprosma rhamnoides A. Cunn. 27 y)
Meryta sinclairti (Hook. f.) Seem. 15 0
Cordyline australis (Forst. f.) Hook. f. 7 ()
Myoporum laetim Forst. 5 7
Paratrophis smithit Cheesem. 5 ()
Pittosporum fairchildii Cheesem. 4 ()
Entelea arborescens R. Br, 3 )
Gemostoma ligustrifolium A. Cunn. 3 0)
Brachyglottis arborescens Oliver Z 0)

No significant numbers can of course be quoted for the individual
plants of the lianes Muehlenbeckia complexa, Tctrapathea tetrandra anc
Clematis indivisa, but the first is a new arrival on this plot with about
5 established colonies; Tetrapathea has certainly increased since 1946
(about 40 colonies) ; whereas Clematis is confined to the same 3 colonies
marked by Turbott. Established seedlings around the parent plants
were recorded by Turbott in 1948 but these do not seem to have sur-
vived.

The Herbs

Most of the ground herbs mapped in Pl. 25, fig. 19, could not be
recorded as individuals; furthermore, the stippling used to represent the
grasses and sedges records only whether they were present at all in the
squares of the grid, not the actual area covered.

The most conspicuous event in the ground layer has been the
invasion of the damper parts of the quadrat by Colensoa physaloides.
The areas affected are almost pure stands of this plant—the only herb
which has survived being engulfed by Colensoa is Pteris comans. With
the exception of 5 remaining tussocks, Colensoa has completely cleared
the watercourses of the Carex shown in Turbott’s Pl. 51, fig, 21.

The areas bare of ground cover are approximately the same as
indicated by Turbott—a large space in the S.W. quarter of the plot, two
spaces under the young kanuka trees and patches along the W. boundary.
Flowever, the large area in the S.W. quarter, truly bare in 1946, is now
a shrubbery of tree seedlings (cf. Pl. 24, fig. 18, and Pl. 25, fig. 19),

No attempt, on this occasion, was made to sort out the composition
of the areas occupied by “grasses.” Casual survey, however, showed
that the principal components are still Oplismenus undulatifolius and
Echinopogon ovatus, as recorded by Turbott. The squares marked as
containing “sedges” included very little Carew virgata, which has prob-
ably declined in favour of C. testacea since 1946.

Two new arrivals among the plot’s complement of herbs are
Erigeron canadensts and Haloragis erecta, The first was not recorded
anywhere on the island by Baylis in 1946 (see list, p. 247 et seq. Baylis,
1948), It is now found sporadically everywhere in the open kanuka
forest, but is not frequent on the plot itself and has not been recorded
in Pl. 24, fig. 19. Gnaphalinum collinum and Ovxalis corniculata are
probably now absent from this plot, but no opinion can be given about
the other small herbs mentioned by Turbott as they were not specifically
looked for in 1951.

PLATE 16.

Fig. 3. Quadrat I, 6th October, 1948. (Fig. 5C in Turbott.) Young kanuka
in opening, new shoots arising from
australis. Colensoa physaloides in right foreground. The

Entelea arborescens in the left foreground had completely obscured
this view by 1951.

the trunk of Cordyline
young

Fig.4. Quadrat I, 6th October, 1948, (Fig. 8F in Turbott.) Sedges more
luxuriant than in 1946. Colensoa physaloides has appeared along
the water-courses (centre and right foreground). Young Melicope
ternata and Brachyglottis arborescens in r:ght foreground.

Photos: E. G. Turbott.

) an

aaa

4 ae 2 I ah)
pate 3
y ry

a 7

F 7
a : es

*)
;

PLATE 17

pide 3,

Quadrat II, 6th October, 1948. (Fig. 1OH in Turbott.) Sedges more
luxuriant than in 1946. Coprosia rhamnoides in centre and to
right. Young Cordyline australis and Dianella intermedia in line

$cotsat este Tenet, Photo: E. G. Turbott.

Fie. 6. Quadrat III, 6th October, 1948. (Fg. 13L in Turbott.) The
tussocks are Scirpus nodosus. The young kanuka seedlings shown
by 1951 (PI. 22, fig. 15) are not yet apparent. The wh-te lines are
the boundaries of the plot with the west corner in the toreground.

Photo: L. C. Bell.

Resurvey of Vegetation Quadrats, 11/7

Quadrat II

As can be seen by comparing Pl. 45, figs. 9 and 10, in Turbott’s
paper with PI. 21, figs. 13 and 14, superficially little change has occurred
in the general appearance of this plot, for the tree seedlings are still
too small to add an additional layer to the vegetation. However, the
areas of turf have mostly been replaced by sedges throughout the whole
plot and Dianella is a new and prominent component of the herb layer.

With the exception of the cabbage tree in the E. corner, Turbott
did not mark the positions of the trees on this quadrat, but records that
the canopy was constituted by 46 kanuka trees. The total number of
live trees shown in Pl. 26, fig. 20, for the same area is now only 21,
including this same cabbage tree. On this plot, too, then, mortality among
the kanuka trees has been high, but whereas there is no evidence of
regeneration since 1946 on Quadrat I, on this plot there are a few (five)
young bushes additional to the thicket shown by Turbott. Coprosma
rhamnoides, which Turbott mentions as “scattered over the quadrat,” is
still the dominant shrub and the most numerous tree seedlings are of
the cabbage tree. The full complement of tree seedlings is shown in
Table 2. The Melicope is the same specimen recorded by Turbott and
it has grown very little since. During the 1948 landing this tree was
noticed to have been severely damaged by cicadas.

| Table 2.
Shrubs and tree seedlings recorded on Quadrat II, January, 1951:

Number of Plants.

Coprosma rhamnoides A. Cunn. 75
Cordyline australis (Forst.. f.) Hook. f 17
Leptospermum cricotdes A. Rich. 12
Meryta sinclair (Hook. f.) Seem. 8+-
Litsaea calicaris (A. Cunn.) Hook f. 4+-
Paratrophis smuthit Cheesem. is
Melicytus ramiflorus Forst. 2-7
Mvyoporum laetimn Forst. a
Melicope ternata Forst. ]

Those marked -+- are new records since 1946.

Only the larger herbs have been recorded individually, and the
shading representing Doodia media and Carex (testacca) in Pl, 26,
fig. 20, merely indicates whether these were present in the grid squares.
There was no bare ground on this plot except beneath the kanuka bushes :
the squares shown blank in the diagram being actually occupied by turf.
The nature of this was not investigated carefully, but the principal
component was Oplismenus undulatifolius.

A list of the larger herbs present is given in Table 3.
Table 3.

Larger herbs recorded on Quadrat II, January, 1951:

Number of Plants.

Erigeron canadensis L. 30-+
Haloragis erecta Schindler 17+
Dianella intermedia Endl. 11+
Arthropodium cirrhatum (Forst. f.) R. Br. 1+
Pavalhia tasinant Cheesem. 1+

Those marked + are new records since 1946,

118 HoLpswortTH.

Quadrat III

More change has occurred in the general features of Quadrat II]
than is apparent from a comparison of Pl. 22, figs, 15 and 16, with
Turbott's Pl. 47, figs. 13 and 14. The greater part of the plot is still
occupied by a short turf of mixed grasses and other herbs, but the inter-
spersed tussocky growth which can be seen in Turbott’s photographs is
Scirpus nodosus, whereas in Pl. 22, figs. 15 and 16, almost the same
appearance is given to the photographs by the swarm of windswept
cushions of kanuka which has spread west from the original scrub in
the north corner. Most of this change has become apparent since 1948
(cf. L. C. Bell’s photograph, Pl. 17, fig. 6, with fig. 15), though both
jaylis (Baylis, 1948; Turbott, 1948) and Bell recorded the establish-
ment of new kanuka seedlings.

On the other hand, the amount of Scirpus has probably not altered
—the positions of this shown in Turbott’s Pl. 52, fig. 23, correspond
with those in Pl. 17, fig. 21 (in the latter it occupies 2.1% of the total
area). This means that the area in turf has declined, for the area
occupied by kanuka has certainly increased. The area now occupied by
young kanuka is 24.4% of the total, and though it is not possible to
extract a figure from Turbott’s data for comparison, the kanuka in his
diagram occurs in only two groups of bushes, one of which, that in the
N. corner, is now dead. Young trees are already established under
the dead branches.

The sedges which Turbott noted under this group of bushes still
persist. Leneath the dense cover of the new kanuka, on the other hand,
the ground is quite bare.

The composition of the grass sward is presumably much the same
as it was in 1946. The following table (Table 4) is an analysis of the
strip of metre squares against the S.W. boundary.

Table 4.
Frequency of herbs forming the turf of Quadrat ITT.

No. of squares (out of 15)
in which species occurred.

Ceitella asiatica (L.) Urban 15
Deyveuxia crinita (L.) Zotov 15
Aira caryophyllea L. 10
Vulpia dertonensis (All.) Volk. 10

Wahlenbergia gracilis (Forst. f.) Sechrad.
Aira praccox L.

Gnaphalium collinum Lab.
Sonchus oleraceus L.

Danthonia seniannularis R. Br.
Hypochoeris radicata Lab,
Oxalis corniculata L,

Zoisia matrella (L.) Merrill
Carex breviculmis, R. Br,

Cotula australis (Lieb.) Hook. f.
Dichondra repens Forst.

Doodia media R. Br.

Erigeron canadensis L.
Hydrocotyle novaecselandiae D.C,

— ee et 8 On Ct ON CO GG. NO

With the exception of Erigeron canadensis, it is doubtful whether
anv of these are new to the plot, for unless they are in flower it is
difficult to separate the grasses and probably even this list is not
exhaustive,

PLATE 18.

Fie. 7. Quadrat I, 14th January, 1951. (Corresponds to Fig. 3A in Turbott,
1948.) The white line across the upper part of the picture is the
string marking the W. boundary of the plot. The seedling obscuring
the r:ght foreground is Melicytus ramiflorus.

Quadrat I, 14th January, 1951. (Fig. 4B in Turbott.) New erowth
from base of Cordyline australis in centre and left. Ground now
covered with Carex. Young Melicytus in foreground and to right.
The fern is Pteris comans.

en nn nnn aang EERE

N.B.—The view corresponding to Turbott Fig. 5C was completely obscured
by a young Entclea arborescens.

sone Fe

Lad . a
eo eee ;
eee Pace ees
ea Ve

ee
pees

<b t

Aacd

ke cial

i i
an

f come ea”
itn La ,

on
Lee ee en
he

74
ee)

i. ; en

i.
wag a
en

7
=e

-—
on

PLATE 19.

Fig. 9. Quadrat I, 14th January, 1951. (Fig. 6D in Turbott.) Coprosma

x
+f° a. .
rhamnoides, in centre foreground. Colensoa physaloides at base
of tree on right. Young Leptospermum ericoides on left. The white

string is the N. boundary of the plot.

Fig. 10. Quadrat I, 14th January, 1951. (Fig. 7E in Turbott.) New growth
from base of Cordyline on left. Colensoa obscuring foreground.

Resurvey of Vegetation Quadrats, 119

DISCUSSION.
(a) Regeneration of kanuka forest

On neither Quadrat I nor II is there any evidence that the existing
kanuka forest is being replaced as the old trees die out. Thus on both
a change to some other type of forest can be forecast. On both the rate
of change is rapid: Since 1946, 39% of the kanukas on Quadrat I have
died and probably 67% on Quadrat IT; yet there is no good evidence
of the appearance of a single new kanuka seedling since 1946 on Quadrat
I and only a 19.2% replacement on II. On both quadrats there were
eroups of kanuka seedlings in 1946, and, at least on Quadrat I, it is
clear that these had appeared below the only breaks in the canopy, its
while the goats were in occupation, the only factor keeping back the
erowth of new kanukas was the shade cast by their parents, and the
appearance of a gap anywhere in the canopy immediately induced a
replacement crop more than sufficient to fill the gap.

With the goats removed the sequence is quite different. Kanuka
seedlings are still not tolerant of the parental shade, but they are not
able to exploit the open spaces either. This is not a question of com-
petition with other tree seedlings, for on Quadrat I the principal open
spaces are almost free of them (excepting, of course, the kanuka seed-
lings existing pre-1946) and on Quadrat II the association of other
tree seedlings is nowhere dense enough to ofter resistance to colonisation.
The reason is therefore probably the coverage provided by the herbs,
which is, in fact, most dense below gaps in the canopy. The principal
plants involved are Colensoa and the sedges (the grass is nowhere very
vigorous), the former of which was completely controlled by the goats
and the sedges though present before 1946 are now much taller and
denser (cf. in Pl. 18, fig 7, with Turbott, PI. 42, fig. 3).

Thus, through the control that by browsing and trampling they
exec’sed on the coarser herbs, the goats were responsible for the main-
tenance of the kanuka cycle in the forested part of the island. On the
other hand, they also suppressed the seedlings of other trees which are
the natural heirs of the kanuka. The degree of regeneration still taking
place on Quadrat IT reflects the general dryness and poverty of the soil
on this plot compared with Quadrat I: the ground herbs are less flour-
ishing and can offer less resistance to the entry of kanuka seedlings.

On Quadrat ITT, the situation is quite different. The hillside is too
windswept to maintain high kanuka forest; instead, the same species
here can only attain the status of a stunted and matted scrub. Baylis
(1948) has pointed out that the occupation of this area by grass was
probably the result of a fire a long time ago (probably even before
Cheeseman’s visit in 1889) and the goats have delayed its recolonisation
by kanuka by inadvertent erazing of seedlings in the sward.

Since the removal of the goats the recolonisation has been greatly
accelerated and the grass will presumably eventually be excluded
altogether, though in five years less than a quarter of the sward has
disappeared on this particular plot.

(b) Alternatives to the kanuka cycle

Though it 1s certain that the succession of kanuka trees that have
forested Quadrats I and I] has now come to an end, what 1s to replace
it is not yet defined.

120 HonpswortH.

On Quadrat I, mahoe forms nearly half of the generation of seed-
lings that has sprung up since the goats were eliminated (Table 1). Yet
this apparent dominance is probably a temporary phase due to the
initial advantage of a freely fruiting parent plant of the same species
on the plot. On Quadrat II, for example, mahoe occupies quite a low
place in the order of frequency, this plot being further from a. seed
source. Mahoe dominated forest is occasionally met with (e.g., as a sub-
association of rain forest—Thompson and Simpson, 1938), but mahoe
is better known as a sub-dominant—a position which it commonly fills
in the various types of coastal forest around New Zealand (Cranwell
and Moore, 1935; Oliver, 1925, 1944; Hamilton, 1936).

Litsaea and Melicope, which follow in the list for Quadrat I, are
again probably more numerous on this plot than would be expected in a
random sample of the island's vegetation, because of the proximity of
their respective seed sources. Both these species are handicappd in
the race tor succession by a very slow rate of growth. For example, two
of the seedlings numbered by Turbott on Quadrat I were of M elicope
(Turbott, 1948, p. 267) and were then respectively 6” and 4” high.
Now, nearly five yeares later, they are only 4’ 9” and 2’ 11”. For com-
parison, some of the Meryia seedlings which have appeared since 1946
are already over 8’. All the Litsaea seedlings seen also were small.

There are two species which have a fairly high proportion of seed-
lings on both plots—Cordyline australis and Meryta sinclairii. The first
ot these had a regular place in the forest before 1946, so that its frequent
occurrence as seedlings everywhere now that the goats have gone was
to be expected. Of Meryta, on the other hand, there was no seed source
at all on Great Island in 1946, yet this, too, made an early appearance
everywhere and on both quadrats it is already a prominent feature of
the vegetation.

Baylis (1951 q.v.) has given reasons for anticipating the develop-
ment ot Meryta forest as the next successor to the kanuka. It has a
very rapid rate of growth (exceeded here perhaps only by Entelea) and
its large, leathery leaves cast a shade in which few competitors may be
expected to struggle for long. Some of the seedlings on the plots will
shortly be fruiting themselves, and then, with a nearer seed source than
the outlying islands, the competition for dominance on these plots should
shortly be decided.

On Quadrat III, the dominance that has been forecast for Leptos-
permum ericoides may here, too, be only temporary, though there is no
evidence yet for this on the quadrat itself. However, pohutukawa
seedlings, seeded no doubt from the surviving trees along Tasman Bay,
have appeared in some numbers in exactly the same type of association
not very far away from the quadrat. Since kanuka is so very light
demanding, in competition with pohutukawa it may eventually be ousted
from this quadrat also.

It has already been assumed that Leptospermum seedlings cannot
grow in the shade of the canopy cast by the same species. This is an
extreme intolerance, for the canopy of mature kanuka is not particularly
dense and within its shade a great number of other species both of herbs

PLATE 20.

Fig. 11. Quadrat I, 14th January, 1951. (Fig. 8F in Turbott.) &ntelca
arborescens in background. Colensoa physaloides, filling water-
course from right background to left foreground. Haloragis
procumbens in right foreground. The sedges shown in ‘ig, 4

(1948) have now largely been suppressed. The white string is the
E. boundary of the plot.

‘ie. 12. Quadrat I, 14th January, 1951. From point P in Pl. 24, fig. 18.
Young Meryta sinclairii. Pteris comans in foreground.

Young
Melicytus ramtflorus in centre.

ied =
a ar oe
Pre a ie is

as
',
a
hn
ih
a
Z
"
a=
ha
a
_
ls
er

l
7 r* as iz ! i= ol - :
i Lal 7 a
7 7 , wig : . a Mat a 7 =. < :
i So oe oe eo 4 = f 9
: a 7 = 7 } ae el h i
ae Roget. Weeae Maer a) an
* - fp" ri L -
7 sy anes ae eo « ™ sr si ar ae ; ia , 7 af i x : a, fy San = a } 1
: Sf cn ; =. a eld . ; on _— 4 Sie al 5 Se sft = E i } a “an =e ee A ont
ae fr Py =a te Bie . are » i = ams eae uae, — = 7
au ve a oe : 7 7 Li Se : i at Bart ae le i a et i ek My Se 2 ae gh 7 ff Phd ri = 7. k
ae | = a i” 7 - Tet ae oi) tae SS “i on Te ee ae 7 a 7 7 3 : ie =i
ie ae ee Ce ey ed er .| et ot. ae " ant, Pa he
oe ! en 7 : jee 154 pall — ls me =o els s
= %. F We? AY cs ae ha : me am ‘ e ol vines) ae: a 7 7 _ «tee get, 5
= i i = a in al i
0 7 sity" an a : ri ry eek ae a | v? A is a ’
ier tm rm ee) a Meal a, in he fr gia : he «i kid Ss ary a a =
c a De cs 7 vee 7 a me, 5 ~~ 7 pl i.e inne a 7 | a 7
Lue Wels By : ys ae re : ¥ Pan epg po t fo a ‘<i oe ,
‘ se 7 ye ee Yip aT nl =e ie fe ots ‘ at a Z 7 sun 7 7 - al
Rater er Mera Sot etfs ae a hao eels Pe: oF nee, i :
; el _ J i Pa = 7 ‘
te rene eRe ek | iy ae,
a he! ovis BaP MS? a a pon. aeoe 2b: Aci a,
Pear me it ae pe ae te Han eit aa vy "ae if ee ae ea —
5 a ie =a : Ao) oh a , . ere ee | Pe is 7 : - 7 . ‘ i =
a ae 1 _ ¥ a ery ae ry ies ny i ‘sy - 4 any iA ™ fe ly : ms r at . 9
* o —_ gtd ae ' aa r 7 + a fora Gi i. | ra. : ae Pi a"
war = 0 aw adne it~ i rere Fe jv od i 7" i a. i
en : F ; - ley Si ee Pe i in of ahs : a a cir el “* i is ‘ ae
7. aoe - : rs \ inatcie * tr toa so a = Sine le ace AP ig: ei i’ j ni 7 :
rae a ; ' il Ta eae toe ce hk Lie a PS oa a ar |
oy _ ‘ 7 aed res, Ree i ae ok | Tt, !
A oa eee | Co ee Agila gee
im pn ‘ Lo ie a a an io i m wo atengt a fo jos a7
a oe 2 ou Z i i } =i 7 ' 7 ¥ =e |
rd as be re; = amie w ‘ :
ee Ps ia rs ' 1 ‘ ha ;:
ih -. sir" ire , he Pat dat i, ae’ i
= ; a, wt Say uh le 1 rept - pris eee a a Q bale : 1 *
4 iL 4 i" ay i 7 ; : ' se 7 fee a car nS | 7 : 1 Me ie ; J
a a ui 7: ot ae ‘7 Ke a 4 } is? : “a ;
oy = ae a s+. 2 se ’ a oe + Ps
z A ~a Bh ¥ 1 .
i eA 1 iL, a al = Zé a 7, 2 -- eo rhb, aS +
"ihe es 4 | ae : i ha AY an ak ey : 4 oe i I
al 7 : 7 : 7 : rk tie ahaa ik i +a. “ “ih i ry ‘ f 7 F i
Z a mo a ie “tig Pale 4 } = Sor 7 sills fnkpegt ruet ae * 3 Sims ™ r
a Fy ia j - ‘es opti ot mgt : ts Fa =. ae” cal "tng ee Sees re an 7, 1] é r
a a : * a - ie . e PF va oak, a a ‘ i ‘’ f ;
Pi fom i‘ =a a oe iy wn ike ve a a an. a = hah “4 i coe ‘as ae : sh oh be
a ya he eh kate een ky Mera BOs) oper |
7 AT Pn an ite or. a i == 4” te peli oe . A Yet 1 ms ' 1
Fa ; =! - ——_" . wie oer 3 —) -_ rule tC = pari a = ye i
i ae a ial Loe! = = a r a igh 7 F
; an oe fa He : ing
i » - TL a pa Xl ey seas ify, 4 : a i
- » ch Pe ad 7 =f Pe “ 4 id! “ .
/ Fi ; ;
i i

Py. mh
My

7 al 7

a
7

Se
ca oa is - a A
=a_® 7 _ alae on | a = as ™ 7 ie
r -; -. oe ry i ae ae Se i % , bs
- 7 a of i co a " i.
ae wa ' : al y = ey » A ™ ry i : . a a :
ra" | 7 _! r Lack Fd - ra is 7
Pr’ - 2 . f Lain ur = sy * on _m 7 ‘a -_ = 7 - s as i. Pe ik 5 ot hal
i ” Py a nil nM =] - ss F me ‘ 7
FA zs eee od 3 ate . a aie, Peres ri i, ' ‘Sig j rs i) : 4 Po
i ye st =“ ‘a Se oA ; 1 fs a iis re «ie <a
Piet” ee Oe ee Bre se re ieee eee, Fe Pe ie a.
a ee x wh 2 ee ee 4 - - : a ee ae =a =e : iis eo Bh
Me ia. avr a be i en 4 wT me et = = ru F , * / iy =. ‘
: Le 1 - - = F Le es: r 7 a oa io a) 7 hy - 1 a“ aa _—s ? .
| eal, A ts rit + sm a | - i e e+ sf uf —— ae al = ae a 7) ime is
= La - i} iz - we a we = aaa i : > i ‘ mi 7 eo a A yet - = 7
ball 1 i 7 t i - ' . z
fi 5 a - ane i i" " i a. lie yh Dif ita ae. © <P ey, Pe
cs ‘os ; eit eg a : re Mapmee a as a | ape? j A i - i at = eel 7 : “st ete
i. Si AP as ee Petal it lla eet bien ke el ere
a a ove tT i 7 i 7 = 7 oe ae La i a _ Ts re
fF ar er rie i ar fal a i ia of is ene oo ah pens een i :
ig = _ a a a Pe a) ae 7 ie 2 yy 7, Pa rut Keene a i ae _ ri 7
Fi 44) “ Y rd =a = va i‘ om ee : a (tat it Af ry i Ss = ; ° tae
a ae : el " he id | 7 7 bol, a Ci Life® nh; * :* - oe r 5 @ pth
5 re) oe re aes | 7 ae eee r i » ye ee 1 oe
Z ew - 7 ae” Ee fi aoe, Py be eee . or ae
Py iit 4 hee Pls et ale is ei ae ee f
Ca : 3 japsinss ay ast i ru a 7. ¥ i - 7
as mle ; =t mt ihe ni la ie cue ie 45 a? rs
=r iat A - is Be bgt | - tena! ' > i ™ eS team tas Le
a ie ier a 7 a a or i “ nil : : % : . 7 — J ie
a . a = 7 ae = - in al bl «= 4 7 i ‘ ri ; he
ee tore fe Pde ere With de Pi wit
a ive ivilg - id - 2 =a 7 ee Si i . P = 4, « a “ a. i 7 if a
Cre ar ee rie, aT oa | a
: ‘1 _ ie er cet mi r i *. te ee in a Pay? = i" ak oy s. 7
= iven : “ a" ae ae 7 a
os ~ «, =i ~ A - beh ' ll - 7 = oe el F . * F a a ;
a an _ Fe a Ti. ae 7 7 =e er. “2 ‘te ac ain
yeas ae a ant Ds ee ae r
wanes cet re © Whee ee Moe or A ee :
Pi \ ; - ‘ i 2 a ae ae a = = r i i
eo re ; _ ae ole dal oe ;
Ci muah te ee A ae le
ane Woks Bias Ps oe a or a. ay Nat es
a ‘| = al 7 1 i: 7 at i oe
oa ae by a t i ; x ry a a - f LZ i
i oo ntl e ¥ i PL el ie a =
: a il “ee “ie aie
iJ 5 he
7 7 =. eee

Pe gre 7 7
- i aT:

a _ aery =
ae, aes he :
7 | oe | = a
are Pi Sete 2 2 Se :
Py ip ee Soar ee ee, rs
e — at | i. er -_ . a= i.
a | a? - Fi. : :* teh a i 7 “al 7 | ‘ a 7 rh
De ates a ao ; a ee a — 7 oe es ae
ee ee 7 i - = el a. - 77 oA ee ; a + 1 _ ome ae
| 7 - ees ae ee zs = oe, r Sets an >,
7 *) ee ae ia an a = ;
a ie ee Fans ie oe 3
oo oo = arn =

ie 13, Odacat TL. Lith January, 1951. (Fig. 9G in Turbott. )
N.B—The direction of this photograph is as in Pl. 26, fig. 20;
direction is shown wrongly in Turbott, Pr 32, fee eZ,

‘ig. 14. Quadrat IH, 11th January, 1951. (Fig. 10H in Turbott.) Young
Cordyline at back right and a plant of Arthropodium cirrhatum

immediately in front of it.
Photos: E. G. Turbott.

Resurvey of Vegetation Quadrats, bi

and tree seedlings are flourishing, Among them are included Meryta,
Melicytus and Colensoa, etc., often themselves quoted as light demand-
ing. The shade cast by the cabbage trees along the W. boundary and
by the isolated trees of Paratrophis and Melicope is much deeper and
below these the ground is almost bare. The only tree seedlings that have
established themselves in this deep shade are all of Litsaea calicaris.
Abundant seedlings of this same species are also growing in the thickets
of young kanuka where otherwise the ground is absolutely bare. This
suggests that over a long period of time Litsaea, in spite of its slow
erowth, will become an important constituent of the forest as it will
not have to await the death of a canopy tree—whatever that canopy 1s
composed of—before establishing itself.

(d) Other barriers to forest regeneration

Reference to Pl. 24, fig. 18, shows that on Quadrat I the seedlings
of any species are numerous only in the S.W. quarter of the plot. This
area almost exactly coincides with the area shown as bare ground in
Turbott’s Pl. 51, fig. 21. It thus seems clear that the herb layer, already
considered as controlling kanuka, is also a barrier to colonisation by
other species, whether it has the assistance of shade from the canopy or
not. The large, coarse herbs Colensoa and Carex testacea are the most
important, of which Colensoa is the more aggressive and has successfully
competed with the Carex itself. All the area now occupied by Colensoa
is shown in Carex in Turbott’s diagram.

Colensoa is undoubtedly a brake on the rate of colonisation of Great
Island by new trees, but it is everywhere limited to the wetter stations—
along watercourses and at the base of damp cliffs. ‘Thus on Quadrat II,
which is dryer than I, it is absent altogether. Moreover, it has probably
already attained the limit of possible colonisation even on Quadrat I.
So the restraint which it can exercise on regeneration is limited.
Furthermore, it is not tolerant of extreme shade (e.g., it is absent from
the cabbage tree belt) and will not in any case survive the dominance
of Meryta.

REFERENCES.
BAYLIS, G. T. S., 1948. Vegetation of Great Island, Three Kings Group. ec.
Auck. Inst. Mus., 3, 239. 7

BAYLIS G. T. S., 1951. Incipient forest regeneration on Great Island, Three
Kings Group. This issue.

CRANWELL, L. M. MOORE, L.B., 1935. Botanical notes on the Hen and
Chicken Islands. Rec. Auck. Inst. Mus., 1, 301.

HAMILTON, W. M., 1936. The Little Barrier Island, Part Il N.Z. Jour.
Sct. Tech., 17, 717.

OLIVER, W. R. B., 1925. Vegetation of Poor Knights Islands. N.Z. Jour,
Sci. Tech., 7, 376.

OLIVER, W. R. B., 1944. The vegetation and flora of D’Urville and Stephen
Islands. Rec. Dom. Mus., 1, 193.

THOMPSON, G., & Simpson, J.S., 1938. The Dunedin sub-district and the
South Otago Botanical District. Trans. Proc, Roy. Soc. N.Z., 67, 430.

TURBOTT, E. G., 1948. Effect of goats on Great Island, Three Kings, with
descriptions of vegetation quadrats. Rec. Auck. Inst. Mus., 3, 253.

ake
Ces ¥

PLATE 22.

Fig. 15. Quadrat III, 10th January, Losi (ie 3k. an Tinbote.).
Leptospermum scrub on left, culms of Deyeuxia crinita can be
seen in front of them. The white lines are the boundaries of the
plot with the E. corner at left centre.

Fig. 16. Quadrat III, 10th January, 1951. (Fig. 14M in Turbott.) The
white lines are the boundaries of the plot towards the W. corner.

Photos: G. T. S. Baylis.

POSI
AND FYS

Extent o

ee @e@es¢@

@eeee
@ee2es

PLATE 23.

Fig. 17—QUADRAT I.

POSITION OF TREES CONTRIBUTING TO THE CANOPY
AND ITS APPROXIMATE DENSITY (stippling).

Extent of the canopy and the positions of the trees contributing to it.

ty —Cordyline australis,

/ -—Litsaca caltearis.
k -—Leptospermum cricotdes,

—Mehevius ramifloris,

——Meltcope ternata.

—Paratrophis smithiu.

p — Pittosporum fairehildii,

om i :
‘ ——Dead tree still standing (Lepfospermium),
Xue
ee ~~ . - . - .
—Clematis indivtsa (liane ).

The density has been plotted on the basis of two metre squares—-

indicates square entirely shaded.

@eeee
®eseeee
e8® eee
eee eas
eee8@es

*.’. square partly shadcd, by canopy.

The diameter of the boles shown is arbitrary.

——

i
i

POS
SEEDLING,

as Hab -ext

——

i
i

PLATE 24.

Fig. 1S—QUADRAT LI.
POSITION AND RELATIVE SIZE OF YOUNG TREES AND
SEEDLANGS.

Telrapathaea tetrandra,

Brachyylottis arboreseens.
-Cordyline australis.
—fntelea arborescens,
--Genostoma liaustrifalaim.
—lLeptospermim cricoiues.
—Litsaca calicaris.
~—Melicyins ramiflorus,

—Myoporum laetiuim,

-Pittosporum fairchildit,

Coprosma rhamnoides,

—Paratrophis siuiuthii,

o Melicope ternata.

—Mucehlenbeckia complexa.

Ken anyvy FJr~ *®O MO OG

—Meryta sinclarrii,

14335 —Seedlings (Melicope) marked by ‘Turbott.
(2), (4, Position of seedlings marked by Turbott which

have not suryived,

ee ss
er @€f
ee se

ee extent of canopy made by young Leptospermuim trees
evese (up to 15’ high).

eree

3°

ms.

POST I4

NN
a
NNN

a= > &

e e@ @

e* e@ @

3°

ms.

PLATE 235.

big, JI9—QUADRAT 1.

alert ~ oy ON
Stee PEE PTTL Pe lye Foy me GALAN shy ae Cae me Say CON eA ee ate SN er ube AD 2
ee ea” TURN RR NARS NAAN SAAS SAS SNAAN SS NR SNN YS Cerise SN NNN SES R262 Ara
Slee t bek gm) ah hire eh Re ie fh a ern EE aga Ri eg tees Sted Tatty Arte ee = RE a eng ey ae eee, STEN RE
5 eae ee ee SVS FANN AKNANSR NARS AAAASY he ee WO eer ne AG ~ qoheee tar Mb eit Nn Se ON oer . SY
ry S NNN RNAS AANA SSS SAAS ASN ANSSS NAS SANS Cope MAS faba TE Me Le ed
Le Ba ee NSRAKASSARAANAAN SAN SVANSBALEASN Be hii Bey Merge rat “Se5 28 ay il Maat
| SV NNASSSLSAAV LAS ~NS SANA AANN ASS pT Ge goatee ieee Mee nee ‘ 2
ee es SSSA SSeS wee Sh eg ee x ‘ ~~
eae tab ce Reta res, Sp heh SAANAAANN ON See pW
Non ie talee ae, erp ag tase WN SNNNNANSA Cee ee ret
+ eh SNe Sass ey i Fe, HS sN AN SRS te Hat, ».
~ wy SN a eT Pe iy Ne ee hea he oe ~AN SANS AAS veel WY
sue Sy ET NN ene ae ye SE NS SAWNAN SN AS Ter
Nuss SS AES OS NS Rm TS TO SSAA SN SNS apes -Qy
~ * NAN SSSVR AAA AN SR ARAN AN SN WNSN NAN SAAR cscs
SNL ALN NN AAS NAAN NARA SERNA AAA NNANNA NANA SA OSS
BAPE EERE OS RNIN SINS ASAE IN Ning AG Bn 0, NANANA SAS ASS OO
ao s« 6 Be ENTSUN St, Oe EAA AS SORA he tye Se nN NANA ASSN SANS CD
SLA SL MES SNS AA NS SARA AAS BAS ASN WRASSE SAS SRSA ANAK VSS SANS ;
“A; eS Py nse Pie hiet te tot SSN AN SNA Soe a A NOS Ree ey eA oe SR Ne
5 DA es * a ag ee ay Ia Th ie, Sects, eign th Se CN OMS ia MN As RO SAN Mt Mh ead LS dg
RON NN SRSA SS NTS gS BR RCRA SSRN SNR NEN SNS SASS S SN SALAS RSS AS ANAS < ve
& EBL A ea teys Sect ae ~N ~N ANA Na te a oh icctyeh git Poet NE I RL Eh La BY, < 5
ON NN Nhe oN Oey Nee EN SNS SSN NS SEN NS SIN EAL SN 5
My Nay eS SS OS wt tity te WS SS SNS SAN SSIS ASA ASN SS SNS SRN as 7 aft tisid, neat nem
~S SAN =a ht gO Betas Te chi ~SN a a ey oper, . SANA SN SNS SSNS ES NS EANN ASSN SS SSS ~N SNS vale et Pe De high Ma TS
ie Gey ee NOW NF) Boece eee ie bs NRA Te Dh a eS ee Sa bee we sSSANSS Eph eR ee ae PNY Vy tee ety tate’ e ea tat
Pn Sots rg Ngee sr Sa PLP CRT ee habe SON NNN SN 8 ole OS RN NNA SSN NARRATOR ANS SN ie Me at are eee ees MN e
~AS AS OS Ua Cr ie LO a ASANSNNAS Sa tete tet etatats UNS SSS RNNASNN SINAN SOE SS SS SSS NSS Ber Rca as ABT FE)
eee Sh MAO A ORO rh arte Pan se ae head. NN, Cale MON tee dhlne 9 telwga a ace MN Oy Ne Ae ON NCR, SA DY we ORE NON ah ny Cty ans ee tte
a isa eee ie Ua ela le nN cK A Tec a Se REL RECN Ga Tas See acter Se ne Be pies tte ieeuaee Stree,
eg ge oe ee oe S Ae Wega Bind edad brad ttt doa Eat a ra ie Ne SS ASC SONREANIS ON 8 SANS RON SE TEL ee
Se SELES Fee a or ee we wie eyes pw ehe Se A SSO NC SR MER RS SNS NIRA SEAS SOD SIS Lee Noiatie tokig
tis SALTS ae eee ETAT ATT NNSA R SAAN ARAN SAS SNS ARAN SESS SNES NS vs SS ANN
-=-—-* on == ee Meee: a ares eat ea Se RAN NN AN NNN AA NANA YAS AR ESALE SS NSS Re) eo aye
Vet tA DRT ALN REM ESA CAC Mn SONS WAAC RSS ALAR ASS NC NINERS A SNS SAAD SS SY
Ae Wis ae oe tach KN RA meen NA Me RAR RA Se Rue ANRRARA SN SSS SVS AS ss
~ Sp PES RTS AON BEN, SEIT ON TR NTRS RO NRCG RENE. OS NS WIRES SSS LEN, Se
TOOT OT RANATA RAK AN IINRANTA SAAR RNA RRR SARE RNS ANON NS Nin WW era at bate tet
~ TT WOMAN AINA ASN A AAIN NARA RR RANA AAAS AANA ONY Att RN NNN ENN RA AAS mao
, Lys NER ALSO LEN ANAS AA NRNRR RON AENAER AN OOD ofS A OA AN ASSIS OSs fas
eee VA NNARN TT OTT ONNNARR AAAS ANAS NANA Sere SEL EEN Na None hated oe nae
earls te we b Abe sete erat TNS ee age es BENS Sk Se Sa hE Qe A NON I LD pe bP ees Seta Camis NN NAO yeaa
spurt cegtt4 rt Wr 77 ATA A ANSI TOF Gesu thee Maes Sale oe tit ele S48
Sea perth sand Stee WANNA AN ANINNNNAIAANRS Mea hatae Roe eee eer Ne te Sal aod ge nah ae
Sen eel g aN Se a en Te AS TOCRRANATATANOATSATS S ee i Bae a Te oa A
Pre art te ke ine | By py OO rea Soh eg Se Se te ns 5 i Sl Oe ah re me? AT ee er ne rae leet e = fac
BO te ert LE be dah poy Be Pa ee Sie tl eat ae ~AN i Reewra ri tes ae hs Lae
Lhureg see Mie eee tant ohbe 2 SESS SESE Soo COCO IEL OS Pe 5 esa oe ces ant
~ ~ I NC i ae ee A ge De pe Me Ae Se oe Ba: thea Yer ae pee Pie rie err ace > SS
iy PoE ae at Seen Pt eon atin Sale eh tee Pai 67 seis an re ae ae eee = ss sNANN -- ——-—F* = =
Shee Re SRE Verne dee ey DR tele WISUTET eal a eds Meco et yo wee a cc ao REAR Me crea ool ety hase
athe her Med aoe ehh Te en ed Pee San kgf an te Sagal se PES hs ASSES AEE NE wg reas ees tal ieee aey, Bh
: hrc) oak oe res a AG ot Ly At eet a ee harks 3 tas op eee oa OND SAAN Sy Sts pene et Uae “NT
a En oh CNANA Le ae ta See SE ae oe Se eee Oe ST SSS Sh eR a ng 3 Cop gigs tt a
STANT ATAT VL AAAS Ape ehetaris Gan Later pebtat yeh tometer ee, alg), ut tele y OT NPE Big Fes Baya Fe PR a A 8 aa re
SVAN ATS” AN AALR oice re veeyhee el hye 2 het he cee cpu woe LS IN Sys Bi Py emp oe br pear = neg pene Nh
eet. See: i Mae Gutgh | EOLe SER ee one res
ins eS enw aR ey Ay aOR. Satis WiRisaeteetnt es SOR ase a ta rangi es ee
vi tabpe eesatal, eb : +s Se SOSSSE RSIS SS SR
=, oe a P= ead a.“ SS iy at ay awe Sake ap My A
ae ee = ere oats Per , PS =e Deas hea tc en ts RNIN ES A ns es Sy
bas le eg eer rr eee Md lial! ae ie a ant VN NNSS ANNA SANS SSN SAS A
~ Raa Ns a ee 3 pa, oe AN AAA SANA a, Mu i A Be, SSS SAAN
oe a ~ eel EO. SSSA NSNNAN SSS SENS SAS SASS SVAN
Nps bar bse Pte a rt = ae NNSA SN NNNAANIS SSS SS ANS SSNS SSAA
SAAN SM ey NAAN A ANAININA SSSA SSSA AS SNS ARN
6 Ra 8 CW LN SNR NS NNANN'SA SSS SSR SAN SSS SSNS
-_-«e-- a VS SALAS SAASAARSSVL ALAS SNS SN AASNNAN
conn = Tene SARS ANAS NAN TS VNNALN
So ad ase or ne &- RV VAN NS ARN SSAN NS RNAS RN SSS VANS
m-eee FFT TIN NAR AA NASN SNS SSS SSN SSNS SANA
sae=e- Te en OS A So NS TR te RO ON Ty eR GOREN
Ce te amt oh ae toe ae pl Net ay NA Tn ND SEN NGS DOB Iie
See SE eee Batt Se EN ANN SRS NN
OF eR ee ete oe Be DRESSES SENN BNR ISSN ASN ENON
Na A AAR Fee eee ne Nae SE SN NRO NIN ARIS NONE
CARRASN ASN NN AUN SAN SSN NSN NNSA
—---- ha hb Laon eS oe ee Ce TTF FT FR AN NANA NAN NS AANN SANG SASS SSSA ANS NRRL
----- oO See aS ae SD OS RNS SNS SN SSA SSANA SEQ NUNS SSNS NA ES SAN
wees = Mal Bo TN eee oe ao SSE Rey EON Te TT ENS ey VAAN SSS SSS SSN AN SASS
--+-- Sih oe ariek ae po RNIN RNIN AN SANSA RN RNSNNSRS AS NAAN SAIN
o=--> Ft te i sae ey oie eye ie hee er SAN A SSO CS 8 8 SAAR
----- spy Ss paw zt NAAN NS RSA SSAA NAN SNS SS UNAS OF SSNS SINANISNSAINN
----- ais a a a NAN NNN SENN ANNAN SS NAN SINAN ee ONS SINAN NNANNANA
ees ie Th aes ae ae ML NN NASSAR NERS NSA NSN SRN SSNS ES NAININAININANAA
ee ~ ‘a, se ae a NN NAN N ANN SSAA RAN SNS EE SINNER
=----- — Oe SD Sie re ee NNN NAN ASN RNA NNN NSS SN SNS NS NRINNENNNININNS
CA CERIN SENOS WSN NA NAN NS NSS NNN NANNY Sheet CNN Se Ne ae
NASAN DIT TIN NA AANA AS NINNAS “a LAN NN NNSA AN SSNS NSS SSNS SS Ss Ss SAS =e
NANNN TTT TUNA ANN ANNAN ANNAN ~ ~ SSRN RAEN SNS SANS SDS SEA DD NS oe et
SAAN TIO TN ANANSI AANA SEATS SS RENNES SoS RNAS OPN a cule tet oe oe
= = ef he Noe cs NV VPNNNAN NARS RS SALVE NSS SSNPS SSNS SASAN
SA TDW Tw AAS ASIN Bree ~ re eta ar Ey wy eatat eon Seach
NAO Poh OG. Gat den S oS ENA SOASOND RS, SISSIES NIN TE NESS TUN Ay aN SEN Ne oy tant Sey Tat p pete aN oh ig aan ~---22222
SIRI A. alas all nfs wR oat ad SNe VON RNS Nee NS NE NES NENSWEN SS) NN Nite NN eS SRN NS ANE, OT tly tg Cer ee es tat aa aa a a TaN eee ery
AN SNA c : DSS SN NSN SNS SEAS SSNS PEPE Rees ES ONT RE Nom Sy ge Re Pt neh ie kerte ~~ 2
SVNNS ~ ---- He ory NN ANN NRINIAINIA SAN SARIN SSNS NNN ee NNN NSN SS RERUNS tear t eo, Sy torpedo rate oh,
ser ere ee Me ee ee = = ined hot pee 2 as Dene + wie ee ae ey eS SANNA
an NANA SN SATNNS SAS NNW AN SN ANASEN AN Nr rE NYAS NA EOS ON
Wj J
_ y
- re)
ra te
2 :
4 _
+ =
_— —
ae wo
e
oe) -
~~ eB)
ned =
~ as
tx) o
O L
<= as .
“ a
joe - wm
pe z
rr + =
ly i) ~~
am = =
= 7
tr Ww "
ply v
Ss —
CF ~ ~~
= oe
fr) ~ 5
~ _—~
= = “ee Lan | 3
4 ~ = x, —
re tt a a) cd cS
\ ~~ = ra =
x) o = x — S
nm = : vu =
— ha Sy = S va us
O ‘ 4 S = S =,
’ ~ =k ~ _— ~—
= S my = e S
oo = anc) tr n
ms cS ~
— ed Par SS ~ n S
4 i a eg - wn >
<I = i S w Ps Su
3S 6 ~ me ue ay
Ppod = = a)
~ ~~ res *
| !

- -(
—Co

f
A

Ya NS - - = * * «
SVN - © «= eee

POSITION

Z ‘ + 6
i J -
a
a . ‘ y i a
5
t r
« ‘ i ' Z Ps
; 0 .
. ' ' 2 j
: ,
an a ; >
‘
. . -
> ry - ‘I ; - : ' i i
; 7 : -
P : : ;
> :
‘ ri é ’
: i
- '
: . . '
’ “ ¢ : . rr
- (=
- id ' i - LY
= ’ 4 -
1 4 sos : ~*~. ‘ ‘
‘ . ~
, &# |
’ < ‘
. ‘“e 7
« !
* ec .
« 5
s ” : . s
t
hone 1
. o * ;
: t
q -”
» ' is
*
F : :
i . 7
< ‘
. Fa i
a -
- i ’
: i . '
= 7 . ~~ , 7
. 4 .
7 i * ‘ x
‘ ‘ vs =
= ye i
. x ‘d
~~ 5 i ’
. é ; f
: bi
_ i G 4
* -
: b a
‘ : 7 t 4
’ 4 al
‘ 7 :
, i
all '
' .
‘ * sd
i
. : ' f
t
F a
q
‘4 . *
. * . ;
i, ‘ ®
f f 5 :
-_ P
o) i ee

‘ 7 9
Fs
js :
3 Z
. . ‘
5 ‘ .
‘ »
a : .
. ~ :
Po
z
Fi =
‘ “ ~ ;
+
5 bs “ .
¥
: ee
4 = 7 :
j * Z
' a 5
4 , 7, a a
. ‘ ;
‘
4
=e <
ee ~ —— —= — a
>
ie 4 '
4 *
Pa F. ‘ as ni
1
= ‘
fr pu ne i, i” ‘
‘ - ;
.s * ¥ s
i . ,

a
= s
=
t
:
,
a
“%
. 2 r
. :
Pn t
7 , ~ i
es ; ‘:
.
mls a?
nee t
’ .
ae *
1 =
<
x
‘
a ’ -
a nn
F . *
* t. x =j
= ;
a
q 4 i * .
rT a5 i, pee ? ‘é
3 < 4 on
arhe 4 - ="

26.

PLATE

STRIBU-

YD DI

4
I

ING

el
‘DI

A
kK

]

A
S

U
YD

)

20
S
A,

“

h
OF

—

c
4

1
[

4
y

I
x
*T

DF
ND

(

>
X

‘LO
G

~
4

POSI
ON ¢

> * , = a ; i ge : ay eR <.Pe 6 . ay ”

Arte eo vigip fe ate OER ETRE Om 000 eN eees A ET ee dee
ererteryr ~~ & sft ta ™ OL OO LOL OLLI OL OE. ae ee i ie ae
seceeeee LF CMO LT OLLI IL oe, & i eee ©) tn
tee se se | IPF IG EI DOLE OP PL, ri AR US =
creer QQ PORTE, CL LLL LILO LOLOL OODLE OOO ¢ COOL OOP OE 4
“+ 7 Cee Rs teen opt G Mee Pe LOLA DL OL I & reek
o tt les tC a ge CI Re ee he )
. Micky Medi ta il Fo la OO IES LO FF 4 lene 4¢
ral 46 : L SH FOF PELE ¢ Oe A IIE A I OLE OF pa %~% e FPR
COL LOO OOO OLEOEL GS eh sere
LPS Or 0 ied seer of eee
CaN are ee ad oN ae. oe cere
Ne POT OOO Oa LLL cf
“Y hl LL ML OIL I LLL OLE * or \
- STEAL CPA COL MEERA LOST L LL. “¢
CLE GAS I CAPES AE kik pCO INGE: ies COOOL OSI LOS So. steer
cas se CLL OOM FF ie cee CLL LE ILL LF re <= 4é
Ce eorrererr PEP OE ct Pet ‘ Corer eel et tt LOCC OAP POPE OLE se
eleereerr ae ae tS DE eS s ee a eae ®& a + 46
Ceeheerte M700 ALAA 5 V4 ALLS Le xk- cree
Ce eet ee ay ose eS oT ak if oe i ee Aa I Ae ar at “fre
SOFC POPPE LEE ES DOBLE ae OPAC ET Oo Sree sere 444ee
orete Pe PN ee ee pie A eee ae a ee e
crete i Pe A et Gray aha »¢ %. YQ «rere a
bap ed ¢ s . ¢ oeree Lt a
ore a a a ete
cre S44 FP RE C4
see seeererre sere
see COCR OOE eres |
see sete cete eter
“fe C444 shes
cre OVE s* i ed FPL ALE %)
see A Tal A Re so ees Q)
seeerer Sercerte Tseteer © ;
POOPPOOL FILO EIGS SL IILL J LLL AP LOOD OOLES
¢ EE Ae gy ry Eby ae LE Spee Oke ae
Re) “0 4 TRL Sage Pe a 8 Sa Nie OP LALLA ITIL ALLS ES, Q) :
We aes LP LeALe x ap ae a LL SILL Le A * ‘
or ae et dae 0704S OS. 1) 4S Ln ‘a Y i FOL LE DA PEE Lay iy
Sm) oe ee ate 40 SO SS. vo iW Qf 4272.4". ~~
é nat ae ep CFS ot PO fee rc) pI o — Pepi tae i ha
Lerrrrrer ly YE NA ot PS Dd Kae PEPE Libis poeta
+ a ae ee a ae Sl PPA I PIPL Saw ye Oe eae wy '
COCOPOOPEIOE LE PIGUET SOP OP es 14 ae ae N 59)
CP PLOLOOOLELLOE SE LAE CEL EOC E LAO T ™~
ee ie ae eed SPOOF ICE Serres
eereeee LOOT AITP AOA OAT LE errerertk.
seers a a ae a a ae Corer
“Seeere CPF IAP OIE BAEC OO LL a a a a a
lecerrer SOLA PEPAOP EELS rere tree
ah ad - a ee a eflve a An A ee
cere ¢ certreLorrcresr QRprreriee
¢ 7 COCO OO Ce eresere
s oy a frte C4 OFFA
‘ “7 Stare stesr 4¢ SOLO HAE OEE
‘ aa ae ee fae v4 matte om me deo of Se
A aC a Ay eae ee Pe “ft ff ae ede
oer CED sah te alo Ks 6004
NED Pe Mp hak gad Ne ten St AR AIM AS BUI a Sa EGe» afte DR Pg TEL ELLIO
FSA LO LE ID AA OR LL OA OA EEE OK a PCC FSO Pre ete 4ge4 FOOL oO OIL LO OF 4S L066 f 44
OP? LL IGI LEI OP FF. eee 7.3 AER Ee ORE 2 qX OL Sf ab ff oe ad Ct OO ge ae
CA a ee ae ae etre & Ot Fe
FELL OE F “eee -?.4 2447.
CLOSETS. vee awe
ALI OOOO ROL DAD ORO Love oe ae ee
COOOL ELE OER a OP Oe ee
COPPOPECEOCEEUE GE 4 Bree, AOOTLE,.
‘ LOTS.
oe PIII ISSR PIII SS Z 7 os
Oe 4e t e e Oe QELS ~ re!
a

a
cB)
-
+
v
a_i
S .
—
wu
I
4 he
~ vu
=
n S
Q
se re
w
> He
— nt
om, oO
S)
= co) :
¢ res ~
% a a)
—_ oh Y -
ma 2 TA
fal — =
a b= —
a c Cc E _
ae f
= e zy oe
Fe 2) oa %% ¢
re ‘ ~~
ff = = = <—
~ Y s = Ly
‘ U ~ a
—
— = ae ad
pe) 3°) 7 tae
el oo ‘Z ~
= oni ra eS
~
v — : <—
ny ww mr ~ —
3, a
. ~ tad _ = —
: — ra 5) = ~ 6
. t ed = io“
x 4S 6 = = x <t =
= ew] A — ~ SS ~ ca") ~ a)
>= . ns : a
= ae ee ey Re ot ans = = = © S _ Ss
5 - S 4 =
oT ee endow ke at ES = SS = fe v . a A) %
@ ~~ ~ ~ ~ 4 ¢ -
rt ~ t ~ ow Uv ory ~
~ +s vu w ant Rey a ~ aes on — 2 ?]
me ~ r ae > bag ~ e 5 4 ~ cB) 3 oy s
— wv er =~ ~ bs ~ =~ ening ‘ i) S . ~
~ ~ . ~ ~ ey ~ . ll = = ~~ ~ ™ eee oe “6 t pat mr in -
e ST eros ge. OST ere SS, ees ee egies: ce feel! gS er gD ct ml iss] = a] ws
+ ~ ~~ = hs = ‘~~ al a a ~ = > a) ~ haw So ‘en¥ Lat el rt = .
~ = ~~ <a} —— 4 ~ ra a ~ SS _— ~ —_— ae Te. = as mas.
¢ S ee as ae “= ae a OTS Se Bee les pow 79) ~ — ~ te — +
~ =) By ~ ~ ~ ad ra] tans ~ S ~ 6 Sa = S = ‘ = ) ~ 4 a —_—
— ~ e ~ 4 — ~— ~ ~ & ~ = ~ Lait ae ad na S ec te
bs < al iad rls, > = = S Sp eet aan ~ 3 t ~~ et iho) a c= ~ +
~ ~ ae os ~ ma paed us & 5 f aS ~ os ~~ ~~ = par « — ra
BAe Pe ey =P Pe ee ae a < Ss =) [ = = = S oe
. Ss ; tes - ~
Be cy Wears tN et Tae aed oY aa ee 2s 8 8 = ; 4) = c = a =
— 4 “ . £ ~ ~~ a oo =~. r = nD) fi s =
Si Pa SP Se ea pol ech pet eee Sg ee) is a q > S n
ee ho
Se eted eels eS ern Wea gs go nS ee Cyr ChS) Beis Pas ce s) & —
at “ ~— L =) wh = a, »~ “— oe Qo = Pd =a) nN S U ~ ie) nes ‘ bo
— =) = RS, ° ~ a — ~ R Sa Sy. Tats? Fe > = S CB) ‘= Mw S « ~ La =e
‘ ~~ = ~~ ox Ned * ~ ~ — =~ ~~ 4% o a * Wi
~ 3S ~ = Pex ti ae Tit ~ SS is ° ~ es ee = a oO ~ eel bate “hs hee S T)
= SE eo es eta ee ed i Le ma eS! ES eat = S > - ~ 7 me . 4
SG it eo ee gS Ce oe Bo ow Ss = = cc - = = ~ z
SO ad rd AE pee Pea SRS a Sa ari wees oS tS rb) = a ie) re pe =) 5S i
t ty - ry 5 4 4 Re ‘a : , . = ‘ ry oo r Fite -
ry oS he eA Ce eS SS YS ~ Ss 4 Ww) | / { ag —~ 4) en
| | | j { ' ! i ' | i] | | | j t | 1 ] | | ' 1
| | | | l | { ' | | ‘ { i } i ' !
2 OG@)d Of xwew&w)~ ECK WY SAAN Saad 7+- ony P
#7 --- aw “<7“-
ae x Ba SO
44 --- —_ 4 04u -.*
47 ai
AN pyres

; parsale
Puree Te
- a SK a He
We oe r :
ae et
OES lan west
PR Stee aie

Mook

123

Succinea archeyi Powell

By H.E-QUICK, MAB., BSc, FIR:G:S.. Reading, England.

Abstract.

The anatomy of Suecinea (Austrosuccinea) archeyi Powell is described and
compared with some of the features of other members of the group. The geo-
eraphical range of Swuccinea s.s.. Oxyloma, Quickella, and Austrosuccinea is
discussed.

I am indebted to Capt. F. W. Short for recent shells and alcohol
specimens of Succinea archeyi from the sand dunes at Taipa, Doubtless
Bay, Northland, New Zealand, and to Mr. L. W. Stratton for subfossil
shells from Tokerau sand-dunes, Northland. For the original descrip-
tion and figure of the shell see Powell, 1933. As only subfossil shells
were known then, it may be added that recent shells are pale amber
coloured with a waxy lustre. Powell, 1950, gives an excellent account
of the ecology, and a figure of the mantle pattern and the radula,

Description: Alcohol specimens show the usual external and
internal features of the family. The pigmented bands on the mantle and
body whorl are placed one at the posterior border of the kidney, one at
its anterior border, and one, sometimes broken up into spots, near the
mantle margin (pigment pattern is often characteristic of species). The
genital orifice is an oblique groove terminating at the labial palp (fig. 5).
The only difference between Powell’s radulae and mine is the possession
in mine of a small endocone at the base of the mesocone in most or all
of the marginal teeth.

The ovo-testis is of the usual somewhat flattened raspberry shape
with a pigmented covering and numerous bifid white follicles. The
hermaphrodite duct is pale proximally, densely pigmented in the middle
dilated and convoluted middle part and pale distally. At its entry
towards the base of the tongue shaped albumen gland are the two pig-
inented seminal vesicles, united almost to their apices, with one slightly
longer than the other and sometimes again minutely bifid. The small
fertilisation pouch is as usual unpigmented, figure 1. The folded gela-
tinous oviduct narrows in its distal half, and the slender pigmented duct
of the small globular spermatheca dilates very slightly as it enters the
free oviduct so low down that the vagina is exceedingly short.

The compact prostate, also somewhat pigmented, is of the usual
bean shape characteristic of Succineidae, and the vas deferens runs
forwards under the right tentacular retractor muscle, and back along the
penis sheath to enter the epiphallus at its apex. The penis sheath, about
3 mm. long, is a little dilated at the apex and sharply bent over and bound
down by the retractor muscle of the penis which splits to enclose the
termination of the vas deferens. A narrow neck joins this apical portion
to the rest of the sheath. On opening the sheath, figure 2, the penis 1s
seen to have a slight spiral twist. The epiphallus when opened, figure
3. shows three or four circular rows of papillae in its proximal part,

‘*
Rly

and a series of alternating circular folds distally, while the penis has

124 QUICK.

feeble longitudinal folds. In S. australis the penis sheath is not bent
over at the apex nor dilated, and the epiphallus is relatively shorter and
lacks the circular folds, figure 4. S. striata Krauss from S, A‘frica, and
S. andecola Crawford, from Peru have essentially the same type of
penis and epiphallus.

In S. australis, figure 7, the pigment pattern of mantel and body
almost reverses that of archeyi, figure 6, leaving white streaks outlining
the kidney, and white spots elsewhere, while striata has a dark streak
over the centre of the kidney and a different pattern less pigmented over
the centre of the mantle, and a dark band near its margin (Quick,
1936), (Powell, 1950). The mantel pattern of S. andecola (see Quick,
1939) recalls that of Limnaea peregra.

The shells of S. australis, striata, andecola, caduca, norfolkensis,
and of solitaria and aperta from Western Australia, all show a curious
feather stitch or trellis-like structure, apparently in the substance of the
shell, between the lines of growth, figure 8. It is best seen under a
2-3rds. inch objective by transmitted light, looking down on the last
whorl through the mouth of the shell. This is not present in Succinea
s.s., Oxyloma, or in Quickella arenaria. Seeing that archeyit agrees so
closely with, at any rate, the first three of the above species it was
unexpected to find that this trellis structure was not apparent in its
shell. Fresh translucent shells must be used, as the structure is obscured
in weathered and sub-fossil specimens.

Discussion: Iredale (1937) proposed a new genus Austrosuccinea,
type Succinea australis Férussac for a group of nine or ten Australian
and Tasmanian Succineids as follows: “The Australian specimens dis-
sected by Quick proved unlike the typical Succinea, but were nearer a
form that is conchologically dissimilar. There may be three or four
groups in the Australian fauna and it will be necessary to examine them
anatomically. Austrosuccinea nov. Type Austrosuccinea australis Fér.
According to Quick, Victorian specimens resemble arenaria in jaw,
radula and genitalia, and of arenaria he wrote that it differs so much
from other British species that it will not fit into either of their sections,
and from its jaw it cannot be placed in Oryloma, so another name is
required for the type. The Australian type differs entirely from
arenaria, being more like elegans as figured by Quick.”

A few paragraphs before the one Iredale quotes (Quick, 1933),
Quick showed that in proposing Oxyloma, type Succinea hungarica
Hazay, Hazay was mistaken in supposing that the jaw had no median
projection, but as a matter of fact it has a well marked one except
perhaps in old worn jaws. Succinea hungarica belongs to the same
group as pferfferi, elegans and many exotic species forming the section
Oxyloma. This paragraph was interpolated after the main part of
Quick’s paper was written and seems to have escaped Iredale’s notice, as
it makes his reference to the jaw irrelevant. There is no suggestion in
the main part of the paper that the genitalia of arenaria resemble
australis, and indeed the absence of penis sheath and epiphalus is
described, so the statement towards the end of the paper that arenaria
resembles australis in its genitalia is a regrettable oversight by Quick
that escaped correction on proof-reading.

It is generally agreed that the shells of Succinea of differing
groups are often so much alike, and sometimes so variable in the same

Succinea archeyt. 125

l. Succinea (Ausirosuccinea) archeyt Powell. Genitalia x 14 circa.

2. Suecinea (Austrosuccinca) archeyi Powell. Penis sheath opened, showing
the penis and epiphallus within, x 14 circa.

3. Succotvea (Austrosuccinca) archeyt Powell. Epiphallus and proximal
portion of penis opened and lying within the sheath, x 16 circa.

4. Succinea (Austrosuccinea) australis Fer. Penis and epiphallus opened
and lying within the sheath, x 10 circa.

5. Suecinea (Austrosuccinea) archeyi. Part of alcohol specimen showing
the reproductive or:fice and other features, x 8 circa.

6. Succinea (Austrosuccinea) archeyi. Showing the mantle pattern, x 2.6
circa.

Succinea (Austrosuccinea) australis. To show how the mantle pattern
almost reverses that of archeyi, x 2.6 circa.

“I

8 Succinea (Austrosuccinea) australis. The “feather-stitch” or decussate
shell structure between the striae, x 75.

126 QUICK.

species, e.g., pfeifferi, that they do not form a reliable basis for classi-
fication. fredale, 1937, agrees that dissection is necessary to determine
the group to which a species belongs, so it is difficult fo see why so many
Australian succineids of which the anatomy is unknown are assigned to
definite groups. Again, he says (Iredale, 1939) that the Succineidae are
at present allowed a world-wide range, but that this is questionable, and
that recent researches into British forms show distinct groups in that
small compact area. He therefore proposes the genus Austrosuccinea,
southern or Australian type, and for a second peculiar group, the
generic name Arborcinea. It is, however, surely wrong to question the
world-wide range of the family Succineidae, for snails with the very
characteristic assemblage of external and internal features in tentacles,
foot, jaw, stomach and genitalia occur all over the world. Probably
Succineidae is a misprint for Succinga s.s. Succineidae cannot be
divided into southern and northern groups, for to take an instance,
Ox\loma ranges from Greenland to South Africa, and Succinea s.s.
from the northern hemisphere to S, Africa, and a European species,
arenaria, which Boettger, 1939, has raised to generic rank Quickella, is
more nearly related to the Pacific genus Ca/inella than, as far as 1s
known at present to any other European species, though members of
this group occur in North America also. Succinea australis is nearly
related to striata Krauss, South Africa, and to andecola Crawford,
Peru.

I have only been able to examine dried up bodies of S. norfolkensis
Sykes, after soaking, but it appears to belong to the same group as
australis, striata and andecola. If future dissection of norfolkensis con-
confirms this, Iredale’s proposal of Spirancinaa Iredale, 1945, as a new
genus will become unnecessary. Sykes, 1900, proposed Tapada for
norfolkensis, but Studer in 1820 named putris as his type, so the name
is not appropriate for norfolkensis.

The group Austrosuccinea is characterised by the presence of a
penis sheath, absence of a penial appendix, a very short epiphallus, low
entry of the spermatheca duct into the oviduct, genital orifice an open
groove, and microscopic feather-stitch shell structure, but the latter
jeature is absent in archevyi.

As S. archeyi differs from the other members of this group mainly
in its relatively somewhat larger epiphallus, and absence of feather-
stitch shel! structure, it may also well be included in Austrosuccinea.

REFERENCES.

BOETTGER, C. R., 1937. Zool. Ans., 127, 3/4, 49-52.
TREDALE, T., 1937. Australian Zool., Sydney, 8, 307-308.
IREDALE, T., 1939. Journ. Roy, Soc. W. Australia, 25; 12:
IREDALE, T., 1945. Australian Zool., Sydney, 11, 53.

5 POWELL, A. W. B., 1933. Proc Malac. Soc., 20, 191.

6 POWELL, A. W. B., 1950. Records Auckland Mus., 4, 61-72.
7. QUICK, H, E., 1936. Ann. Natal, Mus., 8, pl. 3, fig. 5.

8. QUICK, H. E., 1939. Proc. Malac. Soc., 23, 334.

4 SYKES, E. R., 1900. Proc. Malac. Soc., 4, 144, pl. 13, fig. 12.

pik
.

!
.

= cw
; +

127

Land Mollusca from Four Islands of the
Three Kings Group: With Descriptions
of Three New Species.

By A. W. B. Powell, Auckland Museum.

In this, my third contribution on the land mollusca of the Three
Kings Islands, a presumed moderately complete census of species, dis-
tributed on the four forested islands of the group, is presented.

Satisfactory samples of leaf mould were obtained from all four
islands and a fairly thorough search for macro-fauna was made on Great
Island and North East Island and to a lesser extent on West and South
West Islands.

The land molluscan fauna of Great Island is best known, since I
have personally collected there on three occasions, including one visit
of five days’ duration.

Including subspecies, the total of land molusca for the group now
stands at 24, and they are distributed thus:

No. of Endemic On one On two On all
4.

species, other Id. others.
Great Island 19 *** 7 y 1 a
North East Island 11 * () 6 2 3 *
South West Island 11 ** 1 5 5 3 *
West Island 8 * () 2 x 4 *

Only three of the twenty-four species and subspecies are found on
the mainland: Tornatellinops novoseelandica (Pfr.), Paralaoma lateum-
bilicata (Suter), and Delos cf. jeffreysiana (Pfr.). The asterisks
denote the number of these mainland species present.

The following table shows the number of species common to the
indicated brackets of islands. The asterisks have the same significance
as above.

Great Island + N.E. Q* North East Id. + W. 7 3
Great Island + S.W. ides North East Id. + S.W. 6 *
Great Island + W. 4* North East + W. + S.W. 4*
Great Island + N.E. + W. 4* South West + W. 3%
Great Island + N.E. + W. On all four islands 3 *

ee et

128 POWELL.

LIST OF LAND MOLLUSCA.

[GREAT ISLAND JN.E [SOUTH WEST 1p. [WEST ID,
a ade Seal ikl NM es One a a SS) ee

-
’
.

5 3 7 2]
Number_of specie A A OD OO OY DD
379

Number of specimens 84 188 | 348 | 183 | 762 | 261 | 649 25 858
(Microfatuna only)

* Widely distributed on the mainland also. T = type locality.

MACROFAUNA:

Placostylus b, ballonsi
Suter.
» O. bollonst
Form A.
» 6, arbutus Powell
» D. caperatus
Powell
Rhytida (Rhytidarex)
buddlei Powell
Rhytida (Rhytidarexr)
johnson: Powell
Allodiscus cassandra
(Hutton)
+, turbotti Powell

MICROFAUNA:
Murdochta annectens
Powell
,, filtcosta Powell
,, solitaria Powell
» hirsutisstma n. sp.
Therastella pectinifera
Powell
Egestula gaza (Suter)
Mocella manawatawhia
Powell
Laoma labyriuthica
Powell
Phrixgnathus subartel
Powell
» blacki n. sp.
*Paralaoma latcumbilicata
(Suter)
» regia Powell
,, turbotti Poweil
» buddlet n. sp.
Laomarer sericea
Powell
*Delos cf. jeffreystana
(Pfr.)
*Tornatellinops
novoseelandica (Pfr)

208

158

The numbers of specimens are inserted as a very rough guide to
the frequency of occurrence of the species, but the results are not
accurate enough to be termed quantitative. The samples varied in bulk
from a 25lb flour bag in samples D, E, and I to about one tenth that
amount in samples K. and L. Some samples, I for instance, consisted
largely of coarse unproductive debris such as whole puka leaves and
large twigs. On the other hand, relatively small samples of humus, F,
K and L for example, were phenomenally rich, |

LIST OF LEAF MOULD SAMPLES.

A. Great Island. South East Bay and landing slope to vicinity of
Provision Depot. A small sample of leaf mould, “Will Watch”
Expedition, A. W. B. Powell, Feb., 1934.

B. Great Island. Below rock face, N.E. of Provision Depot.
“Arbutus” Expedition, A. W. B. Powell, Dec.. 1945, and FE. G.
Turbott,

Land Mollusca of Three Kings. 12%

C. Great Island. Kanuka forest in depot stream valley. “Arbutus”
Expedition, A. W. B. Powell, Dec., 1945. Specimens collected
in situ. |

D. Great Island. North West Landing slope in stunted ngaio scrub
at ca. 500 feet. Site of Placostylus bollonsi caperatus colony.
“Ocean Star” Expedition, E. G. Turbott, 15:1:1951.

E. Great Island. South West coast, ca. 700 feet. Site of Placostylus
bollonsi arbutus colony. “Alert” Expedition, A. W. B. Powell,
6:10:1948. Leaf mould from between large boulders in an
undisturbed area under Paratrophis smithti and Brachyglottis
arborescens,

F. North East Island. Leaf mould from marginal areas of puka
forest, Major G. A. Buddle, Jan., 1948.

G. South West Island. Leaf mould from varied locations. Major
G. A. Buddle, March, 1949,

H. South West Island. Leaf mould from varied locations, Dr. G.
T. S. Baylis, March, 1950.

I. South West Island. Leaf mould from puka forest on summit
ridge, “Ocean Star’ Expedition, E. G. Turbott, 13 :1:1951.

J. South West Island. Half way up eastern slope under small
grove of puka and kanuka, surrounded by scrub. ‘Ocean
star” Expedition, E. G. Turbott, 13:1:1951.

K. West Island. Leaf mould from varied locations. Major M. E.
Johnson, Jan., 1950.

L. West Island: Leaf mould from varied locations. Major M. E,
Johnson, Jan,, 1951. |

ACKNOWLEDGMENTS.

[am deeply indebted to the gentlemen referred to above, who have
at the expense of valuable time that could have been devoted entirely
to their own interests, generously collected samples and made observa-
tions on my behalf.

DISTRIBUTION OF SPECIES.

The distributional results show a high degree of endemicism for
Great Island, which is to be expected since it is the largest and ‘highest
island of the group and the only one with permanent water.

The few endemics from the three smaller islands, none on North
Ikast and West Islands respectively and one on South West Island,
points to a fairly recent severing of these- islands from the larger mass,

Great Island.

From a manuscript by Dr. G. T. S. Baylis on the vegetation of
South West and West Islands, I have been given permission to quote
some of his observations that have a bearing upon the distribution of
snail communities. This may be summarised briefly:

Great Island. Vegetation greatly modified. Induced dominance
of kanuka (Leptosperimui ericoides) brought about by several centuries
of Maori occupation followed by a.considerable population of -goats for
over half a century. (See also Baylis and’ Turbott, Rec. Auck. Inst.

130 POWELL.

Mus. 3 (4 & 5), pp. 239-252 and 253-272.) The puka (Meryta sin-
clari), absent during the goat infestation, has reappeared since their
extermination in 1946,

_ North East Island. Induced dominance of the puka (Meryta
sinclairit), probably resultant from Maori agricultural activities, of

which there is plenty of evidence.

South West Island. The dominant puka canopy again suggests
that the forest cover is not natural but seral, although evidence of former
Maori occupation is not evident. Reduction in the occurrence of two
species of tea-tree (Leptospermum) to a single example of L. ericoides,
on the southern end, was noted. This is of significance in respect to the
occurrence of the snail Egestula gaza (see later).

West Island. The vegetation suggests a climax condition. It is
a mixed forest with puka as one of the rarer trees in the assemblage.
There is evidence of Maori visitation, but not occupation. Of the four
forested islands Dr. Baylis considers West Island alone to be in a
natural state,

THE MACROFAUNA.

Placostylus bollonsi. This and two subspecies occur on Great
Island. The typical species is abundant on North East Island and was
probably intentionally or accidentally transported there by the Maoris
from Great Island. No Placostylus has been located on South West
Island, but much of the island remains to be searched, particularly the
western slope. If an intensive search fails to reveal Placostylus on
South West Island it may be significant that this, the only island of the
four without Placostylus, is the one without evidence of Maori occupa-
tion or visitation, and a different solution of the puka dominance on that
island will require to be advanced. South West Island and North East
Island both have a dominant cover of puka, and conditions appear
similar except that South West is the dryer island. On West Island a
small form of Placostylus bollonsi is abundant. |

Rhytida (Rhytidarex) johnstoni and buddlei. The former occurs
on both North East and West Islands and the latter only on South West
Island. No Rhytida has been found on Great Island. If Placostylus
survived the modifications to the Great Island flora then Rhytida, if it
was ever present, should have survived also.

Allodiscus cassandra, Found abundantly on Great Island, North
East Island, and on West Island (one dead shell). It may have been
carried from one island to another by Maoris.

THE MICROFAUNA.

Egestula gasa. This is the abundant snail throughout the kanuka
scrub and forest of Great Island. A very few examples came from
leaf mould taken representatively over South West Island, but it is
apparently absent from both North East and West Islands.

It would appear that since Egestula gaza shows a marked prefer-
ence for the rather acid condition and poor humus associated with
kanuka, the present abundance of the species on Great Island is entirely
resultant from the induced dominance of kanuka occasioned by the com-
bined factors of human occupation and browsing activities of goats,

PLATE 27.

Figs. 1 & 2: VFParalaoma buddlet n. sp. 1.4mm. x 2.15 mm. South West
Island. Holotype.

figs. 3 & 4: Paralaoma regia Powell, 1948. 0.85 mm. x 1.7mm. North
East Island. Holotype.

Fig. 5: Phrirgnathus blacki n. sp. 1.9mm, x 1.8mm. Great Island
(Holotype ).
Fig. 6: Murdochia hirsutissina n. sp. 5.6mm. x 6.0mm. Great Island

( Holotype).

Land Mollusca of Three Kings. 131

which gave the unpalatable kanuka its chance to spread. This suggests
that Egestula gasa under natural conditions was a rare local on Great
Island just as it is today on South West Island.

It is of interest to note that the species is entirely lacking from the
puka forest of the summit ridge on South West Island and its occurrence
there must be in marginal scrub. Since Baylis infers a reduction of
Leptospermum on South West Island to a minimum since Cheeseman’s
visit in 1889, the scarcity of this snail there can be accounted for.

Future observation on Great Island should show a reduction in the
numbers of this snail as the kanuka loses dominance with the develop-
ment of mixed forest.

Mocella manawatawhia. Living examples of this snail were found
only under ngaio (Myoporum lactum) on the site of the Placostylus
bollonsi caperatus colony, North West landing slope, Great Island.
It is noteworthy that Egesiula gaza was absent from this location, prob-
ably because of extreme dryness since some kanuka was present.

Murdochia filicosta. This is one of the most abundant snails of the
puka forest on North East, West, and South West Islands. It has a
near relative in the scarce and extremely local 7, annectens from Great
Island. The latter was probably a dominant on Great Island when the
original cover included puka. The development of three scarce local
species of Murdochia on Great Island is probably resultant from isola-
tion in original forest remnants.

The remaining items of the microfauna show a curious haphazard
distribution among the four islands and no useful evidence is apparent
that suggests any particular linking sequence between the islands.

SYSTEMATIC.
Placostylus (Basileostylus) bollonsi Suter, 1908. Form A.

The race of bollonsi from West Island averages a slightly smaller
adult size than for the typical subspecies from the South East Landing,
Great Island. The epidermis is thinner, of a paler shade of light brown
and the aperture tends to be less capacious. These differences, however,
are too slight to warrant a new subspecific nomination for the West
Island form. The North East Island colony is indistinguishable from
the typical species. Since marked subspeciation in bollonst is apparent
only on Great Island, it is assumed that both the West Island and North
East Island occurrences are resultant from Maori visitations and that
in respect to the West Island colony a slightly stunted ecological variant
is developing in response to the more exposed nature of the habitat.

Dimensions and other features of ten examples of the West Island
form, for comparison with the tables given in my 1948 paper (Powell,

1948, pp. 286-287) :

Ht. (mm. ) Diam. (mm. ) Spire ratio. Spire angle.
77.50 35,00 2.28 30°
84.00 35.00 2.44 37°
84.50 35.00 2.36 34°
85.00 36.00 2.36 34°
85.00 36.00 2.43 Ag be
85.00 36.00 2.36 35°
85.00 36.00 2.43 36°
86.00 37.50 2.32 35°
87.00 39.00 2.37 Bo”

87,50 37.50 2.36 36°

132 POWELL.

The largest example measured 90.5mm. x 37.5 mm. with an
aperture to spire ratio of 2.43 and a spire angle of only 28°. This
specimen is abnormally elongated, due to an injury at the second post-
nuclear whorl, and has been ignored on this account.

_ The egg, as with those from Great Island, is exceedingly variable
in size and shape: 17.5mm. x 11.75mm.; 16.0mm. x 11.5 mm.. and
14.00 mm. x 11.75 mm.

The embryo shows an elongated tendency as in that of bollonsi
arbutus and bollonsi caperatus.

All other characteristics are identical with those of the typical
species.

Locality: West Island, Three Kings Group. Abundant over most wooded
parts of the Island. oO

Major Johnson, on his second visit in January, 1951, at my request,
endeavoured to include in his collecting the largest examples obtainable.
That the West Island race is constantly smaller is thus substantiated.

Average measurements of all the bollonsi colonies for comparison
with the West Island form:

bollonsi bollonsi Ht. (mm.) Diam. (mm. ) Spire Spire

ratio. angle.

S.E. landing, Great Island |
(type loc.) 91.65 37.15 2.33 36

N. of S.E, landing, Great Is. 94.75 38.70 2.35 36.8

North East Island 95.67 39.55 2.29 42.4

West Island (Form A) 84.65 36.30 2.37 35.2

bollonst caperatus
N.W. landing, Great Island

(type loc.) 90.20 34.60 2.52 35.3
bollonsi arbutius
S.W. of Great Is. (type loc.) 108.40 41.45 2.47 32.4
Hapuka Point 97.85 37.15 2.47 32.6

(The average is of ten examples in every case.)

Murdochia hirsutissima n. sp. Pl. 27, fig. 6.

Shell large for the genus, trochiform, umtbilicated, with tall,
narrowly conical early whorls but rapidly expanded over the last three
whorls, which bear complicated epidermal structures produced at the
periphery and the middle of the whorls into very long flexuous hair-like
processes. Whorls 73, including a small papillate protoconch: of two
smooth whorls. Following two whorls narrowly conical and sculptured
with numerous retractively arcuate somewhat irregular epidermal axial
folds. Remaining three whorls rapidly expanding and biangulate, one
angle at the middle of the whorls and the other sutural, which renders
the body-whorl sharply carinate. Both angles bear long. hirsute pro-
cesses, those on the middle angle being the longer and more erect than
those at the suture which on the spire whorls tend to lie flat against the
succeeding whorl. There are about 50 primary radials on the body
whorl and most of them bear processes although many are shed or
damaged. There are several weaker radials in each interspace that do
not develop processes. On the base, all of the primary radials bear
dense short backwardly directed bristles. Umbilicus deep, about one
seventh major diameter of base. Colour golden-brown, with the early
post-nuclear- whorls and the longer proeesses darker brown.

Land Mollusca of Three Kings. 134

Height, 5.6 mm.; diameter, 6.0mm. (holotype),

Locality: Great Island, South West coast, ca. 700 feet; site of Placostylus
ballonst arbutus colony, in leaf mould amongst large boulders, under Paratrophis
and Brachyglottis; A.W.B.P., “Alert” Expedition, 6:10 :1948.

Holotype and paratypes in Auckland Museum

Phrixgnathus blacki n. sp. Pl. 27, fig. 5.

Shell small, elevated-conic, perforate, with keeled periphery. Sur-
face smooth and shining with a distinct regular radial pattern in reddish-
brown on a light horn-coloured ground. Whorls 63, including a low
dome-shaped, smooth, colourless protoconch. Spire whorls lightly con-
vex with deeply channelled supra-margined suture. Body-whorl sharply
keeled at the periphery and bearing a narrow rounded spiral cord.
Surface sculpture of dense concavely arcuate radial riblets, about 30
per mm. Umbilicus narrow and deep, about one tenth diameter of
base which is sculptured with subobsolete dense radials and spirals.
Colour pattern of clear cut radials, 14-20 per whorl, with interspaces
approximately equal to the width of the radials. Base devoid of colour
pattern. |

Height, 1.9mm.; diameter, 1.8 mm.

Locahty: Great Island, South West coast, ca. 700 feet; site of Placostylus
hollonst arbutus colony, in leaf mould amongst large boulders under Paratrophis
and Brachyglottis; A.W.B.P., “Alert” Expedition, 6:10 :1948,

Holotype: In Auckland Museum.

The species is nearest allied to P. erigone Gray, from which it is at
once distinguished by the supra-margined suture, rib margined periphery
and clear cut reddish-brown radial pattern recalling that of Laoma
labyrinthica Powell, which it somewhat resembles in size and shape.

Paralaoma buddlei n. sp. Pl. 27, figs. 1, 2.

Shell small, depressed turbinate, widely umbilicated, finely radially
costate, thin, shining, dark horn coloured. Spire equal to height of
aperture. Body-whorl with narrowly rounded periphery very slightly
above the middle. Whorls 44, regularly and slowly increasing, includ-
ing smooth protoconch of 14 whorls. Suture deeply impressed, Post-
nuclear sculpture of numerous crisp retractively arcuate, somewhat
irregular, radial ribs, approximately 48 on the penultimate whorl and
60 on the last whorl. Interstices reticulated with microscopic radial
threads crossed by numerous spiral lirae. Umbilicus about one fourth
major diameter of the base.

Major diam., 2.15mm.; minimum diam., 1.9 mm.; height, 1.4 mm. (holotype).

Locahty: South West Island, in leaf mould in the puka forest on
ridge. KE. G. Turbott, 13:1:1951. “Ocean Star” Expedition.

Holotype and many paratypes in Auckland Museunn.

the summit

The species is similar to regia in sculptural detail, but has a higher
spire, more numerous axials, is darker coloured, and attains a larger size.
Named after the late Major G. A. Buddle, D.S.O., whose keen interest
in the fauna of the Three Kings Islands has resulted in the finding of
several new species of mollusca.

REFERENCES.

POWELL, A. W. B., 1935. Land Molluscs of the Three Kings Islands, New
Zealand, Proc. Malac. Soc. 21, pp. 243-248.

POWELL, A. W. B., 1948. Land Mollusca of the Three Kings Islands. Rec.
Auck. Inst, Mus. 3 (4 & 5), pp. 273-290.

13-4

On Further Colonies of Placostylus Land
Snails from Northernmost New Zealand.

By A. W. B,. POWELL, Auckland Museum.

Abstract.

_ This paper is supplementary to my 1947 “Distribution of Placostylus Land
Snails in Northernmost New Zealand.” Further field work has resulted in the
discovery of more living colonies and a number of additional subfossil sites. The
primary purpose of this paper is to describe sixteen new subspecies. On the com-
piction of the field work it is proposed to publish a distributional map and to
discuss more fully the significance of the distributional pattern.

Placostylus (Maoristylus) ambagiosus michiei n. subsp. Pl. 28,
fig. 1.

A lightly built shell of distinctive colouring, with obscure spiral
malleations and a simple peristome bearing a weakly developed basal
tubercle only. Spire slender, 36°-39°, Aperture 2.12 to 2.20 total height
of shell. Colour of epidermis Buckthorn-brown, and Dresden-brown,
Mars-brown and Chestnut-brown (Ridgway, Pls. 14 & 15); peristome
orange-rufous (Ridgway, Pl, 2) ; Interior of aperture with a slight bluish
grey smear of callus; brown epidermal colour showing through, where
not clouded by callus. A narrow white line submargins the suture.

Height. Diameter. Apertural ht. Spire angle.

70.0 mm. 32.5 mm. 33.5 mm. 39° (holotype)
70.5 mm, 31.0 mm. 33.0 mm, 38°

74.5 mm. 31.5 mm. 34.0 mm. 36°

Holotype and paratypes in Auckland Museum,

Locality: Kerr Point herbfield, North Cape block, under the matted aprons
oi stunted flax (Phormium) near the eastern margin of the herbfield and only
along the coastal ridge.

Associated plants in the vicinity of the snail occurrences are Hebe
speciosa brevifolia, Cheesem., Cassinia amoena Cheesem., and Leuco-
pogon richer R. Br.

The herbfield 1s sparse and the growth nowhere more than three
feet in height. It is on a hard pan and gets very dry in summer. The
food plant was not ascertained, but it is certainly not karaka, which is
absent from the herbfield. Strangely enough, the apparently better
conditions on the seaward cliff face below the herbfield, where a higher
growth of coastal scrub with karaka occurs, no examples of Placo-
stylus were found.

The subspecies michie: seems to have evolved with the herbfield,
which would appear therefore to be a community of some antiquity.

I am indebted to Mr. R. H. Michie, of Kaitaia who brought this
colony to my notice.

This subspecies and typical ambagiosus from Cape Maria van
Diemen Island are the only New Zealand members with a chestnut
brown epidermis and an orange apertural callus. The true relationship,

Placostylus Land Snails. 135

however, is with watti, from coastal forest along the eastern margin of
the North Cape block. Both watti and michiai exhibit spiral malleations,
a feature not observed in the other subspecies. From watti the sub-
species mitichiei is distinguished by its different colouring, thin shell and
the reduction of the apertural processes to one basal tubercle.

Placostylus (Maoristylus) ambagiosus watti Powell, 1947.

1947. —Placostylus (Maoristylus) ambagiosus watti Powell. Rec. Auck. Inst.
Mus. 3 (3), p. 187, pl. 22, f. 10, 11 (only).
_ Lf ype locality: Midway between Waikuku Beach and North Cape Lighthouse,
10-30 feet elevation and from 25 to 200 feet back from the boulder beach.

Only five examples were taken from the type locality, which has
since been destroyed by fire. However, a more extensive colony was
discovered by Mr. N. Gardner last year on the cliff face about half a
mile south of the North Cape.

I now find that I was mistaken in associating with watti the sub-
fossil colonies from Tom Bowling Bay and Waikuku Beach. The latter
have all five apertural processes strongly developed, especially the
parietal tubercle. This influenced me to regard the apertural formula
of qwatti, in the broad sense, to be 3, 3, 3, 3, 3, The norm of watti is
now shown to be a lightly built shell with an apertural formula of
Z, 2, 2, 0, 0, occasional examples, such as the holotype, show a very weak
parietal tubercle. Also, the epidermal colour of those living under
normal shaded conditions is shown to be darker, a rich dark-chocolate.

Placostylus (Maoristylus) ambagiosus gardneri n. subsp. PI. 28,
fig. 8

1947.—Placostylus (Maoristylus) ambagiosus watti Powell, Rec. Auck. Inst.
Mus. 3 (3), p. 187, pl. 22 (in part), f. 12, 13.

This is a heavier shell than watt: and has all five apertural pro-
cesses well developed, especially the parietal tubercle. It oceurs in the
consolidated dunes at Tom Bowling Bay and in the loose dunes at
Waikuku and Whareana. Its evolutionary descendant is undoubtedly
the recent ambagiosus whareana, described following.

Height. Diameter. Apertural ht. Spire angle.

74.0 mm. 35.0 mm. 36,0 mm. 48°

78.5 mm. 36.0 mm. 38.5 mm. 48° (holotype)
81.0 mm. 34.0 mm. 37.0 mm. 44°

Localities: Tom Bowling Bay, along almost the full length of the beach, with
Fhytida duplicata, weathering out in situ from the underlying consolidated dunes.
Remains of trunks and limbs of puriri and pohutukawa indicate the former presence
of coastal forest on the site during a post-Pleistocene moist period (type) ; Waikuku
Beach in leo-e dunes mainly along the middle section of the beach; Whareana
Reach in loose dunes extending S.E. from Wharekawa Point to about the middle of
the beach and approximately in front of the living colony of ambagiosus whareana.

Holotype and paratypes in Auckland Museum.

Placostylus (Maoristylus) ambagiosus whareana n. subsp. Pl. 28,
fig. 2

Evidently the recent descendant of the subfossil gardneri, trom

which it differs in the absence of the parietal tubercle. The apertural

formula is 3,3, 3,3,0, No other recent subspecies has the three pro-

cesses of the inside of the outer lip so heavily and constantly developed.

136 POWELL.

The coloration is the same as for annectens; that is, russet to mars
brown, diffused with warm-sepia and very narrow white subsutural line.
Aperture deep red-brown within and light ochraceous salmon on reflected
edge of peristome.

Height. Diameter. Apertural ht. Spire angle.

79.5 mm. 35.0 mm. 40.0 mm. 44° (holotype )
$3.0 mm. 35.0 mm. 40.0 mm. 42°
80.0 mm. 36.5 mm, 40.0 mm. 43°

Locality: Whareana, cast coast between Waikuku Beach and Parengarenga in
a steep valley to the north of Whareana Stream. Abundant in mixed forest, chiefly
under kohekohe (Dysorylum spectabile Hook). It does not seem to occur in the
considerable area of forest in the valley of the Whareana Stream. First discov-
ered by Mr. N. Gardner.

Holotwpe and paratypes in Auckland Museum.

Placostylus (Maoristylus) ambagiosus hancoxi n. subsp. Pl. 28,
ne. 3;

Although this subspecies has only a moderate development of the
apertural processes the peristome is heavily callused externally as in
watt, gardneri, whareana and annectens. It is of the size of whareana.,
but has much weaker inner lip processes and the addition of a weak
parietal tubercle. It is much smaller than aunectens, in which a parietal
tubercle is present also.

The shape and coloration is exactly as in whareana.

Height. Diameter. Apertural ht, Spire angle.

79.0 mm. 35.0 mm. 38.5 mm. 43° (holotype )
75.0 mm. 34,0 mm. 38.0 mm. 44°

79.0 mm. 34.5 mim. 38.5 mm. 4()°

Locality; Near crest of second coastal ridge N.W. of Maukin’s Nook, between
Waikuku Beach and Parengarenga Harbour. Only eight adults were taken,
apparently the last remnant of a former colony along this ridge, which has been
devastated by fires. A considerable area of forest down the southern slope did
not produce further material. (A. Hancox, R.A, Prouse, H. S. Prouse, R. H.
Michie, and A.W.B.P., April, 1950.)

It would appear that members of the ambagiosus series live chiefly
on the seaward cliff faces and rarely further inland, to a distance of
not more than a mile, but in such cases only on or near the crests of
ridges.

[Tolotype and paratypes in .\uckland Museum.

Placostylus (Maoristylus) ambagiosus spiritus Powell, 1947.
FPI,-28; fie. 6.

1947 —Placostylus (Maoristylus) ambagtosus spritus Powell. Rec. Auck. Inst.
Mus. 3 (3), p. 185.

The type locality for this subspecies is in loose post-Pleistocene
dunes two to three miles east of Pandora, Spirits Bay. I have since
located numerous former colonies ranging intermittently from the type
locality eastwards along the coastal dunes to within a few hundred
yards of Maungapiko, at Kapowairua, the eastern end of Spirits Bay
Beach.

There is considerable size variation in these subfossil colonies and
one, at 1 mile west of Maungapiko, on the first dune from the beach,
is the smallest member of the ambagiosus series yet discovered. Adults

Placostylus Land Snails. 137

range from 48-64 mm. in height, Subfossil Austrosuccinca archeyi were
associated, indicating dwarfing and then extinction of the colony by the
succession to a sand-grass dune community.

In my 1947 paper I attributed to spiritus the then recently discov-
cred Recent colony from the cliff face 3 mile west of Pandora. A visit
to the locality, affording study of ample material, shows that this Recent
colony is nearer to the dwarf colony referred to above than to spiritus.

It 1s significant that there is a gap of about three miles between the
Recent colony and the nearest subfossil occurrences of spiritus.

Table of dimensions of spiritus colonies.

Max.-size (mm.) Min, size (mm.) Average size (mm. )
(10 examples )

4m. w. of Maungapiko 77.0 x 34.0 — —

ym. w. of is 790 x 32.0 74.0 x 31.5 76.25 x 31,50
Im. w. of _ 77.0 x 315 69.0 x 31.0 73.50 x 31.15
2 m. w. of y 76.0 x 32.0 69.0 x 28.5 72.55 x 30.90
24m. w. of - 75.0 x 30.0 68.5 x 30.5 70.40 x 29.85
3 m. w. af 5 73.0 x 30.0 63.5 x 27.0 69.80 x 29,90
32m. w. of f 75.0 x 32.0 66.0 x 27.0 70.00 x 28.80
Type loc. 75.0 x 32.0 64.0 x 28.0 68.75 x 29.15

Placostylus (Maoristylus) ambagiosus pandora n. subsp. Pl. 28,
fig. 4
This subspecies is characterised by its small size, and very dark
epidermis. Only the basal tubercle is developed. The nearest approach
to the epidermal colour in Ridgway is warm blackish-brown (PI. 39),
aperture deep red-brown within and light ochraceous salmon on reflected
edge of peristome.

Locality: In a small remnant of coastal forest, half way up cliff face, { mile
west of Pandora. The terrain is boulder strewn covered with masses of Muchlen-
leckia. The snails occur under this growth as well as under boulders and around
the bases of Phormium, that is within the leaf fall area of several large karaka

(Corynecarpus, laevigata) (type): subfossil on first ccastal dune 1 mile west of
Kapowairua, Spirits Bay. (AW.B.P., 21:3:1949.) )

Height. Diameter. Apertural ht. Spire angle.
67.5 mm. 31.0 mm. 34 mm. 40° (holotype)
71.0 mm. 31.0 mm. 35 mm. 43°
61.5 mm, 29.06 mm. 31 mm. 43°
64.0 mm. 28.0 mm. 29 mm. 42° (1m. w. Maungapiko)
48.0 mm. 20.5 mm. 21 mm. 36° (1m. w. rf

Holotype and paratypes in Auckland Museum.

Placostylus (Maoristylus) ambagiosus paraspiritus n. subsp.
P28; fie. FT:

A living colony on a small, almost detached headland one mile south
of Cape Maria van Diemen and a few hundred yards west of the type
locality for priscus bears a strong resemblance to subfossil spiritus, from
the Spirits Bay dunes. The living colony, however, appears to have
existed in situ for a long time, as evidenced by subfossil Placostylus
three to four feet down in consolidated sand underlying the Recent
colony. Only one adult subfossil was taken and it is dwarfed and
senile, apparently abnormally short and inflated, 63 mm. x 30 mm.; with
a priscus style of aperture (Pl. 28, fig. 7. Associated with this sub-
fossil Placostylus were numerous /¢hytida duplicata, indicating con-

138 POWELL.

temporary coastal forest conditions, and near the surface Austrosuccinea
remains, pointing to a sand-grass community, now succeeded by a shrub-
dune community, resulting in the extinction of Austrosuccinea. During
the xerophytic sand-grass community phase the Placostylus probably
survived on the steep seaward cliff face.

Although spiritus and paraspiriius are very similar in size, form and
the apertural processes, it is evidently a case of near convergence, for
spiritus is a Post-Pleistocene fossil with a Recent derivative, pandora,
in its own area, which is situated no nearer than six miles from the
paraspiritus colony and with two subfossil subspecies, priscus and
lesleyae ranged in between. No kecent colony has been located between
paraspiritus and pandora, but the report of a “comparatively fresh speci-
men taken below the lighthouse at Cape Reinga remains to be investi-
vated. See locality map, Powell, 1947, Rec. Auck. Inst. Mus. 3 (3).
m. 188). Ph 232)

Compared with spiritus, paraspiritus differs in having the three
outer lip processes weakly developed and no thickening on the columella,
In spiritus the uppermost process of the outer lip is mostly quite as well
developed as the basal tubercle. Neither has a parietal tubercle. The
apertural formula for spiritus is 2, 2, 3, 1, O and for paraspiritus
1,1, 2,0,0. (See folder plate, Powell, 1947, l.c.p. 188, pl. 25.)

Coloration warm blackish-brown. Aperture deep red-brown within
and light ochraceous salmon on reflected edge of peristome.

Height. Diameter. Apertural ht. Spire angle.
72.25 mm. 32.0 mm. 35.5 mm 40° (holotype)
72.00 mm. 32.5 mim. | 36.0 mm. 41°
70.50 mn. 36.5 mm. 34.0 mm. 39°
Max. size. Min. size. Average size (10 examples ).
73.5 mm. x 31.0 mm. 66.5 x 29.5 mm. 70.80 mm x 31.75 mm.

Locality: On steep seaward face of small rounded headland about one mile
south of Cape Maria van Diemen. Under flax, small karaka and other stunted
coastal scrub. First discovered by Mr, R. H. Michie, of Kaitaia.

Holotype and paratypes in Auckland Museum.

Summary of distribution of Placostylus ambagiosus & subspecies.
(+ = new locality records)

Placostylus ambagiosus ambagiosus Suter, 1906.

Recent. Cape Maria van Diemen (Island), Small colonies in flax
(Phormiunt ) champs on the south-west cliff face. (Type) Restricted
to the Island. x (see footnote).

Placostylus ambagiosus hinemoa Powell, 19-47.

Subfossil. Cape Maria van Diemen (Island), in consolidated sands.
Pleistocene or early Post-Pleistocene (type) Restricted to the Island.

Placostylus ambagiosus worthyi Powell, 1947.

Subfessil. Cape Maria van Diemen (Mainland), on north-eastern
side of headland, formerly an island but now linked to the maintand by
a tombolo of consolidated and drifting sand (type); on a small island,
accessible by wading at low tide, half a mile south of type locality (A.
Hancox and A.W.B.P., 18:11 :1948) ; in loose sandy humus, with Aus-
frosuccinea remains? ; on a former islet $ mile south of type locality and
just off the western escarpment of the “priscus block,” in coarse
cemented shell sand7,

Placostylus Land Snails, 139

Placostylus ambagiosus consobrinus Powell, 1938.

Subfossil—Recent. Cape Maria van Diemen (Mainland) at type
locality for worthy: but in the overlying loose sand. A Recent example
with epidermis intact was found here last year by Mr. B. S. Bird. °

Placostylus ambagiosus priscus Powell, 1938.

Subfossil. Cape Maria van Diemen (Mainland) about three-
quarters of a mile east of worthyi type locality in consolidated dunes
(type) ; many former colonies on south and eastern slopes of Herangi, -
700 feet, down to Te Werahi Stream and Swamp; Twilight Beach
between Cape Maria van Diemen and Scott’s Point (N. Gardner)+; one
mile south of Te Paki Stream and one-quarter mile inland in consolidated
dunes (Miss Lesley Keene) f.

Placostylus ambagiosus paraspiritus n. subsp.

Recent. On headland one mile south of Cape Maria van Diemen
(Mainland) (type)f.

Placostylus ambagiosus lesleyae Powell, 1947.

Subfossil. Taputaputa Bay, east of Cape Te Reinga, in con-
solidated dunes (type); in consolidated dune on eastern side of Cape
Te Reinga (FE, Richardson) 7.

Placostylus ambagiosus spiritus Powell, 1947.

Subfessil. Spirits Bay, two to three miles east of Pandora in loose
dunes (type) ; eastward along coastal dunes to Kapowairuaf.

Placostylus ambagiosus pandora n. subsp.

Subrecent-Recent. On cliff face three-quarters of a mile west of
Pandora (type, Recent) ; first coastal dune, one mile west of Kapowai-
Tua, Spirits Bay (subrecent) f.

Placostylus ambagiosus keenorum Powell, 1947.

Recent. Maungapiko, 50-150 feet, Kapowairua, eastern end of
Spirits Bay.
Placostylus ambagiosus annectens Powell, 1938.

Recent. Unuwhao, 900 feet, on track between Spirits Bay and Tom
Bowling Bay (type) ; coastal ridge, north of Unuwhao, 850 feet}; The
Huka, 30-40 feet, above stream, east side of headlandf.

Placostylus ambagiosus watti Powell, 1947.

Recent. At base of cliff midway between North Cape and Waikuku
Beach (type) ; on cliff half a mile south of North Cape, 50 to 100 feet
(N. Gardner) f.

140 POWELL.

Placostylus ambagiosus michiei n: subsp.
Recent. Kerr Point herbfield, North Cape Block?.

Placostylus ambagiosus gardneri n. subsp.

_Subfossil. Tom Bowling Bay in consolidated coastal dunes (type) ;
Waikuku Beach in loose dunes; Whareana Beach in loose dunes}.

Placostylus ambagiosus whareana n. subsp.

Recent. Valley to north of Whareana Stream, between Waikuku
Beach and Parengarengaj.

Placostylus ambagiosus hancoxi n. subsp.

Recent. Crest of second coastal ridge north-west of Maukin’s
Nook, between Whareana and Parengarengay.

FOOTNOTE:

x Suter (1913, Man. N.Z. Moll., pp. 767-768) recorded Placostylus hong
from Kaitaia and Mangonui and hongii ambagiosus from Kaitaia. I have examined
the material in the Suter collection upon which these identifications were made and
find that they are all ambagiosus ambagiosus, the typical subspecies, which in the
Northernmost block is definitely restricted to Cape Maria van Diemen Island.

However, I am assured by Mr. Harry Matthews, of Kaitaia that his wife
collected the Kaitaia snails recorded by Suter from near the bank of a river
amongst karaka and flax on her father’s farm (Mr. Dunn) three miles south of
Kaitaia in about 1888.

Regarding the Mangonui record, Mr. A. Hancox collected six bleached, worn
shells of ambagiosus ambagiosus from the dunes at Aurere, southern end of the
‘Tokerau Beach, about 1934.

The Kaitaia locality is too far inland to have been a natural occurrence. A
fairly thorough search of the Tokerau dunes was made last year but no further
examples of Placosiylus were seen, although subfossil Rhytida dunniac and various
Charopidae were common.

Since both the Kaitaia and Mangonui localities are about seventy miles distant
from the Cape Maria van Diemen Island type locality and that this typical sub-
species is clearly the evolutionary product of insular isolation I can only regard
these two mainland occurrences as resultant from human transportation,

PLATE

Fig. 1: Placostylus (Maoristylus) ambagiosus michiei n. subsp., Kerr Point (holo-
type. Fig 2: P. (M.) ambagiosus whareana n. subsp., Whareana (holotype). Fig. 3:
P, (M.) ambagiosus hancoxt n. subsp., second ridge N.W. of Maukin’s Nook
(holotype): Fig. 4: P. (M.) ambagiosus pandora n. subsp., near Pandora (holo-
type). Fig. 5: P. (M.) ambagiosus spiritus Powell, 1947, Spirits Bay loose dunes,
subfossil (holotype). Fig. 6: P. (/.) ambagiosus paraspiritus n. subsp., one mile
5. of Cape Maria van Diemen (holotype). Fig. 7: P. (M/.) ambagiosus subsp. ?
3-4{t. below paraspiritus colony in consolidated sand (note: the parietal process is
an abnormality). Fig. 8: P. (M/.) ambagiosus gardneri n. subsp., Tom Bowling

Bay, subfossil, consolidated sand (holotype). (AIl to uniform scale.)

v7,

141

Notes on the Birds of the Three Kings Islands

By E. G. TURBOTT, Auckland Museum.

Since the account by Turbott and Buddle (1948), further observa-
tions on the birds of the Group have been made by Buddle (1948, 1949)
—-on South West Island, Stella Rock and Great Island—and Johnson
(1950), who landed on the previously unexplored West Island in early
January, 1950. Johnson found a colony of red-billed gulls on the
highest point of the latter island (cf. Turbott and Buddle, ibid), and
noted bell bird and red-fronted parakeet. He saw large numbers of
petrel burrows on the main bush-clad slope of the island, but only a
few could be examined as the visit was so brief. One burrow contained
a young fluttering shearwater.

The following brief observations on the birds of Great Island were
made on a visit by an Auckland Museum party in the Marine Depart-
ment’s M.V. “Ocean Star” from 10th-15th January, 1951,

Breeding of the black-winged petrel (Picrodroma hypoleuca
nigripennis ) .

The black-winged petrel was discovered on Great Island by P. C.
Bull on 3rd December, 1945, when some 24 birds were seen ashore at
night at the eastern point of the island. At the Kermadec Islands the
eggs are laid in late December and early January, so that it is probable
that breeding had not yet reached this stage on Great Island. No
further observations were possible at this time.

During the present visit, on 14th January, the black-winged petrel
was seen ashore at the same point. During the day, the bare, soft ground
beneath the scrub along the cliff-tops was examined, but no burrows of
this species could be found amongst the numerous burrows of the
fluttering shearwater (Puffinus gavia) and, in considerably smaller
numbers, of sooty shearwater (Puffinus griseus). In the same locality,
however, one isolated burrow was found in a small opening in the scrub,
and proved to contain a sitting black-winged petrel and egg. The
entrance was concealed beneath dense sedge, and outwardly was similar
to recently formed burrows of the fluttering shearwater. The bird
uttered a soft piping note repeated several times when first disturbed.

Internally the burrow consisted of a straight shaft of uniform
diameter approximately 24 feet in length, at the end of which was
accumulated a quantity of dry sedge and other nesting material.

Feeding habits of the red-billed gull (Larus novaehollandiae) in
relation to the dispersal of puka.

It was found during the present visit that red-billed gulls were
feeding on cicadas. This observation is significant in view of the con-
clusion reached by Baylis (1948, 1951) that transport of seeds by the
gulls has enabled the puka (Meryta sinclairii) to become established on

142 TURBOTT.

Great Island. The puka was absent, apart from one small sapling seen
on the cliffs, before the destruction of goats on the island in 1946, but
the seedlings appeared shortly afterwards. G. A. Buddle observed the
gulls feeding eagerly on puka fruits on South West Island on 3rd Janu-
ary, 1947; and during the present visit droppings obtained on the inland
rocks near the castaway depot on Great Island were found to contain
seeds of puka and of ngaio (Myoporum laetum). It would evidently’
be possible for the gulls breeding on Great Island to bring seeds back
from South West Island or North East Island, on both of which puka
is plentiful. Or gulls coming from the breeding colonies on South West
Island might bring seed to Great Island.

The transport of seeds of puka into the forest on Great Island
requires some explanation other than voiding by gulls passing over the
island. During the present visit gulls were seen from time to time
hovering over the forest canopy, and it was found on further observa-
tion that they were attracted by the large cicada*, which was everywhere
numerous and very noisy. The insects made somewhat clumsy’ flights
through the trees, and, when they darted out into the open above the
canopy, were caught in flight by the gulls. It also seemed probable that
the gulls attempted to catch the cicadas singing on the higher branches.

Although only a few gulls might be observed hovering at any time,
it was evident that they were regularly taking the cicadas. The observa-
tion was confirmed by the examination of droppings or castings which
were found to contain remains of cicadas; and one of the insects was
found on the ground still alive with the abdomen missing, having
probably been dropped by its captor to the forest floor.

Visits to the Group have not been made regularly enough to provide
observations on the arrival and departure of the gulls, but the colonies
would probably be in occupation from approximately late August or
early September until March, these being the times of arrival and
departure recorded by Fleming (1946) at the Mokohinau Group: on
the Three Kings the peak of nesting is apparently a little later than on
the latter Group.

The cicadas, as is the case on the mainland, probably appears trom
approximately December to April. The song was not heard and the
cicadas had apparently not yet begun to emerge on Great Island during
the visit 30th November-6th December (1945) (fide A. W. B. Powell),
nor were they recorded 13th April-l6th May (1946). They were
abundant during late February (1934) and early January (1951), and
the greatest number would be taken during these months.

As described by Baylis (1951), seedling pukas occur most com-
monly above South East Bay, only an occasional young plant of this
species being found elsewhere on Great Island. Baylis points out that
this localised distribution would appear to be explained only in part by
the nature of the soil and other environmental factors. Although gulls
were seen over the forest at various points, including the upper Tasman
Valley, the possible explanation might be suggested that gulls would
more frequently take cicadas close to the coast than at any considerable
distance inland. The slopes rising above South East Bay would be
especially accessible as compared with other portions of the island.
ee ery

* Melampsalta cingulata (Fabr.).

Birds of Three Kings 143

General Observations

A grey duck (Anas poecilorhyncha) was seen on 11th January in
the lower Tasman Valley, where it rose from one of the pools near the
waterfall at the mouth of the stream. There has been no previous
record of this species, and it is apparently a straggler to the Group.

In addition, the following land birds were recorded during this
visit: spotless crake (Porzana tabuensis), brown quail (Synoicus),
harrier (Circus approximans), morepork (Ninox novaesaclandiae), red-
fronted parakeet (Cyanoramphus novaezelandiae), kingfisher (Halcyon
sanctus), pipit (Anthus novaescelandiae), fantail (Rhipidura fuligi-
nosa), bell bird (Anthornis melanura), and starling (Sturnus vulgaris).
The spotless crake, which had been glimpsed briefly on previous visits,
was observed by Dr. Gilbert Archey, who saw one only momentarily on
the forest floor in Tasman Valley.

REFERENCES.
BAYLIS, G. T. S., 1948. Vegetation of Great Island, Three Kings Group, Rec.
Auck. Inst. Mus., 3, 239-252 (appendix).

BAYLIS, G. T. S., 1951. Incipient Forest Regeneration on Great Island, Three
Kings Group. This issue.

HUDDLE, G. A., 1948. Gannetries North of Auckland, Season 1947-48, Nez
Zealand Bird Notes, 3, 40-42.

BUDDLE, G. A., 1949. Birds of Three Kitigs and Neighbouring Waters, New
Zealand Bird Notes, 3, 147-150.

FLEMING, C. A., 1946. Breeding of Red-billed Gull: a Preliminary Census of
Mokohinau Colony, New Zealand Bird Notes, 2, 27-29,

JOHNSON, M. E., 1950. Landing on West King, Notornis, 4, 21.

TURBOTT, E. G., and BUDDLE, G. A., 1948. Birds of the Three Kings Islands,
Rec. Auck. Inst. Mus., 3, 319-336.

RECORDS

OF THE

AUCKLAND INSTITUTE
AND MUSEUM

VoL. 4, No. 3

Published by Order of the Council:
Gilbert Archey, Director

Edited by: A. W. B. Powell,

Assistant Director

22ND DECEMBER, 1952

CONTENTS

VOL. 4, No. 3

Human Forms in the Art of Melanesia.

By Paul S. Wingert, Columbia University

A Note on the Geology of the Albatross Point District, Kawhia.

By A. P. Mason, Associate Geologist, Auckland Museum

The Genus Khopalimorpha Dallas (Heteroptera, Pentatomidae).

By J. G. Pendergrast, Auckland

Four New Species of New Zealand Land Snails and the Systematic
Position of Gerontia cordelia Hutton.

By A. W. B. Powell, Auckland Museum

New Zealand Molluscan Systematics, with Descriptions of New Species,
ace f:
By A. W. B. Powell; Auckland Museum

Some Stray Tropical and Sub-Tropical Sea Birds in New Zealand.

By E. G. Turbott, Auckland Museum

Page 145

Page 153

Page 159

Page 169

Page 187

145

Human Forms in the Art of Melanesia

By PAUL S. WINGERT,
of the Faculty of Columbia University.*

Even a general familiarity with the art of mankind from his distant
prehistoric beginnings to the present time reveals the common practice
of expressing his ideas, emotions, and imaginings through the rendering
of the human figure. This is equally true of the art of primitive peoples
and that of the great civilizations. The reason for this is readily ascer-
tainable. Since the subject matter of art is often the crystalization in
form, line, and colour of the objective and subjective experiences of
man with man, it is evident that the majority of these experiences can
be recorded or expressed through the human form.

The sculptures carved for the great cathedrals of the Middle Ages,
the painting and sculpture of Michelangelo, and the paintings of Leon-
ardo da Vinci and Rembrandt are classic examples of the importance of
the human form in European art. But in no instance do these artists
represent the human form scientifically. Instead, certain parts, such
as the hands, the head, the bulk of the body, or its verticality, are
emphasized. Although these distortions are often slight, and always
subtle, they nevertheless exist ; and from them this art achieves its ereat-
ness, aesthetically and expressively. Terror, contemplation, physical
tension and relaxation, excitable imaginative moments, and devoutness
can all be presented by the human form. In some instances, facial
features are utilized almost exclusively (cf. the “Mona Lisa”), but in
other cases the entire figure contributes to the desired expression.

Melanesian art differs from that of the great European masters
particularly in the freedom with which the human figure is used. But
this freedom is not a matter of personal choice by the artist; rather, the
particular forms and patterns, as in the art of all primitive peoples, are
largely determined by the cultural tradition of the regions of their origin.
It is because the regions dominated by a cultural tradition are numerous
in Melanesia, and often diverse in culture, that the art of these islands
is so extremely varied in its character. This is particularly true in the
different ways the human form is rendered. Although numerous simi-
larities are apparent, the great diversity in this art must certainly be
considered an outgrowth of the cultures of peoples who were divided
into many small and often unrelated and hostile tribes.

It must be noted, however, that the domination of Melanesian art
by tradition did not reduce the artist to the rdle of a mere technician.
On the contrary, although he could not deviate very far from the art
style of his tribe, these very restrictions relieved the artist of the neces-
sity to invent or devise new forms. He was therefore free to devote his
creative energies to the task of interpreting anew within an established
tradition. That this did not lead to a dead level of mediocrity is amply

* This study was made by Dr. Wingert while he was working with the Auckland
War Memorial Museum collections as the recipient of a grant from the
Wenner-Gren Foundation.

146 WINGERT.

evidenced by the many examples of Melanesian art in the Auckland
War Memorial Museum. But all of these are not of equal quality.
Some stand apart from others as true masterpieces, and make it clear
that primitive artists, just as artists of every era and place, were of
unequal artistic capabilities. The present concern is with a group of
masterpieces of Melanesian sculpture.

“Art may be considered the product of an extremely perceptive and
receptive sensibility. That no two persons see the same thing when
looking at an identical object is a truism. The artist, because of his
sensitivity and training, is particularly observant of many of the details
of life about him which pass unnoticed by other persons. From these
details he selects the characteristic, the expressive, and presents them
in his medium in such a way as to give a revealing interpretation to a
common reality shared by everyone. The sculptor is particularly sensi-
tive to the structure, proportions, postures, and movements of the human
form. Through it he interprets and expresses all moments and aspects
of life as he experiences them. The work he creates is therefore a
personal document, and through it he shares his experiences with others,
whose experiences he in turn enriches. It is his perceptivity to life forms
which enables the Melanesian artist to infuse traditional patterns with
a personal and sensitive interpretation.

Unlike many European artists of the past and present, the
Melanesian sculptor does not work with the human form, nor with
examples of his tribal forms before him as models. He has been so
sufficiently trained in the latter that his visual memory of them is com-
plete; and he has been so observant of his fellow men that he has in
his mind a personal synthesis of these observations. It is for this
reason that the human form, although the natural aspects of it are
substantially retained in the art of various regions of Melanesia, is not
rendered in a scientifically descriptive manner. Rather, its particular
character is presented by simplified but vigorous sculptural forms which
express the component parts of the body and their integration. This
may be called a “naturalistic” rendering.

Sculptures of this type are not, however, common in Melanesia.
They appear, in fact, as characteristic forms in only two areas—the
western Solomon Islands and the Admiralty Islands. In both of these
areas, moreover, the human figure is also rendered in a number of other
different ways. In the Solomon Islands sculptures of the naturalistic
style have compact proportions and heavy, full-volumed forms
(Pl. 29). Each part—arms, legs, body, neck, and head—is given as a
distinct entity; and in this way the structural components of the human
form are forcefully expressed. This is made even more emphatic by
a slight emphasis of the points of articulation, that is the joints, which
serve the dual role of separating the parts and of connecting them.

In his rendering of each part, the Solomon Island sculptor gives a
simplified synthesis of its basic character, its roundness, shape, and
weight. Descriptive modelling or expository details, such as muscles,
would weaken the force of his statement. He has succeeded, moreover,
in giving a remarkable vitality to his figure through the way all of the
parts work organically together and by the slight flexion or relaxation
given to the various joints. Since it is not a personalized figure, the
facial features are not individualized.

Art of Melanesia.

147

This small figure is an excellent example of a purely sculptural
expression. Its aesthetic quality derives from its rhythmic organization

of heavy volumes in space; its smooth surfaces,
which allow an uninterrupted flow of soft light over
the forms ; and its concentration on and slight empha-
sis of essentials of parts. In it the sculptor discloses
his knowledge of the human form, and by it he adds
to the knowledge of others. That this was the style
of a Solomon Island area and not that of an indi-
vidual is demonstrated by other examples with the
same provenience (Fig. 1).

In contrast, the Admiralty Island artist in his
rendering ignores the roundness of component parts
and the significance of their assemblage and, instead,
expresses the verticality of the figure as a whole
without stressing its individual parts (Pl. 30, Fig. 1).
The pose is given a frozen, eternal quality, with not
the least suggestion of movement or elasticity. A
strict bilateral symmetry adds to this expression of
stiffness. The body parts are attenuated and of tight,
slightly squarish columnar shapes. Only the
spheroidal head is a pure geometric shape. And yet,
the quiet upward movement of steady line and the
compact unity of slight forms in space express in
subtle manner the thoughtful, introspective aspect of
man’s nature. While the details of Admiralty Island
style vary considerably (Fig. 2), the more natural-
istic carvings all have this quality of expression.

In these examples of Admiralty and Solomon
Islands sculpture certain characteristics are stressed
over others; but a moderate degree of distortion
of the human form is also typical of Melanesian art.
This treatment has, in fact, a wider distribution than
the naturalistic rendering. For expressive reasons it
is utilized in many different ways, although it is
particularly apparent in the enlargement of the head
and in the elongation or compression of different
parts of the body.

An excellent example of this style is seen in an
elongated female figure from the Solomon Islands
(Pl. 30, Fig. 2). This interpretation of human form
is further removed from nature than that in the
naturalistic styles. The forms of the body and head
are now slightly distorted to agree with geometric
shapes. While the shapes are separated or marked
off so as to reveal with a minimum of description
their fundamental geometric character, they are also
unified by a strong linear pattern into a system of

Fic. 2.

sculptural relationships, which, in turn, are expressive of those of the
human body. A similar treatment is likewise evident in the slightly
enlarged shaping of the head and in the interpretation of facial features.
Another particularly good example of this style appears in a small figure
carved on the handle of a wooden knife from the Banks Islands (Fig. 3).

148 WINGERT.

With few exceptions, a single style of sculpture prevails in each
primitive tribe or area.* The style is not only found in ceremonial
objects of great importance, but also in the decorative carving on utili-
tarian objects, whether or not these were intended for ceremonial use.
A unity of style is therefore recognizable within an area. For example,
the head so magnificently carved on the handle of a New Hebrides adze
(Pl. 31, Fig. 1) conforms to the style of sculpture prevail-
ing in a particular part of these Islands. It, too, is rendered
in a moderately distorted manner. By pointing the top of
the head to agree with a pointed chin a basically diamond
shape is achieved. An inversion of natural form is adopted
by giving this shape a vertical concavity of surface. Only
the rounded plane of the forehead adheres to nature. The
huge nose is an exaggeration of the large, fleshy Mela-
nesian nose ; while the pierced septum is based on the actual
custom, so widespread in Melanesia, of piercing the sep-
tum, through which an ornament of some sort was worn.
The eyes are very large, sharp, protruding ovoid forms and
the mouth a rather restrained simple crescent. Of consid-
erable importance is the flattening out of the facial planes,
since by this expedient the eyes and the nose are given
added expressive and sculptural force. And yet this head
is only moderately distorted, and is clearly derived from
natural form.

An examination of any comprehensive collection of
Melanesian art discloses that in by far the greater number
of examples distortion of the human form is carried to

Fic. 3. extremes. In this kind of sculpture the parts of the body

and the overall appearance of the human form are fre-
quently far removed from reality. The distortions were generally
motivated by the necessity to express a non-physical being, such as a
spirit or a supernatural force. These were experienced by certain
persons in moments of hysteria or great psychological tension, when
their tangible appearance was revealed. The appearance of the being
was thereafter established. But it is important to note that the world
of normal experience was a strong factor in these moments of abnormal
experience, since the distortions are those of familiar normal forms.
This group of styles can certainly be considered dominated by expressive
desires. Aesthetic principles and elements were, on the other hand,
constantly employed to enhance the expressive power of a work, espe-
cially when the sculptor was a great artist.

The distribution of these greatly distorted styles covered almost all
of the Melanesian islands. It was particularly marked in certain regions
of the Solomon Islands, in New Guinea, New Britain, and in New
Ireland. A small canoe prow figure from the central Solomons is a
superb example (Pl. 31, Fig. 2). The carving expresses a powerful
spirit which protected the occupants of headhunting canoes during their
grisly expeditions. Characteristic of these figures, the head, its shape
and size, and the facial features, are gross distortions; while the

——————_——— ee —————————— ESSE

* “Style” may be said to consist ot the system of proportions, the interpretation and
shaping of parts, the rendering of descriptive or decorative detail, and the
use of colour which are adhered to consistently within an area or by a group.

Art of Melanesia. 149

remainder of the carving consists of a pair of arms, with no body but
free space between them, and hands clasped under the chin, the hands
in some examples represented as holding a small human head. The
expression of these sculptural shapes is of dramatic intensity. The taper-
ing elongation of the cranium and the powerful forward thrust of the
facial area contribute greatly to this effect. Although the facial forms,
which consist of distorted lower and upper jaws, nose, mouth, and teeth,
have an animalistic appearance, an examination of a number of these
carvings shows that they are merely exaggerations of the prognathism,
fleshy noses, and full-lipped mouths of the Solomon Islanders. In this
and many other objects from these Islands a very careful shell inlay 1s
used on the black surfaces to define further structural parts and details.

Human figure distortions are in some Melanesian sculpture com-
bined with non-human forms. Many examples from the Sepik River
area of New Guinea have heads with bird features
attached to a distorted human body (Fig. 4).
These were made for use in ancestor rites or as
containers of ancestor spirit power, the concept
being that at death the spirit departed as a bird,
hence the combined representation of the spirit
and physical aspects of an ancestor. But the bird
features are also distorted, so that the resulting
figure is both related to and unlike reality. It is
rather a tangible and aesthetically exciting sculp-
tural expression of a concept.

From the extreme south-eastern part of New
Guinea another style within this group is repre-
sented by a small squatting figure carved as the
handle of a lime spatula* (PI. 32). This carv-
ing, too, has a large animal-like head; but a
wooden comb carved as projecting above the back
of the head establishes it clearly as a distorted
human figure. These distortions have decorative
as well as expressive intent. The expression 1s
that of a squatting figure holding his knees in his
hands in the pose of balancing on the small plat- 7
form at the base of the blade. The arms and legs, Fie. 4.
as slender curvilinear projections from the body,
are interlocked in a rhythmic pattern which conveys the balance and
tension of the pose. The enlargement of the head serves the practical
function of a knob or firm hand-grip at the top of the handle; but in
the ovoid shape of the head are presented the curves so effectively
developed in the forms below. Even the flattened planes of the face
follow the curvature of the surface of the head. The design of this
little carving is a masterpiece of rhythmic relationships.

In still another example from New Guinea, from the Huon Gult
area on the north east coast, human figures are used in a distorted manner
as supports for a neck-rest (Fig. 5). This style, in common with those
from various other parts of Melanesia, consists of non-naturalistic

* Spatulas such as this were used in the betel-nut chewing habit to transfer lime
from a container to the mouth.

150

WINGERT.

figures composed of a unique assemblage of freely interpreted parts of
the human form. It suggests that there was initially a dismember-

Fic. 5.

ment of the human body. The parts
were then geometrically re-shaped and
were re-combined in a new order so as to
effect a sculptural expression of the
dynamic forces of the human figure. In
the Huon Gulf sculptures, the body is a
heavy trapezoidal block; the stiff, piston-
like arms at the sides emphasize the
vigorous downward thrust of this block:
the legs are curving, flexible forms,
resembling the springy, curved legs of a
tubular metal chair, and are well capable
expressively of supporting this thrust,
while the head, shaped as a very high
relief form on the surface of the upper
part of the body (the neck has been dis-
carded), exerts an even greater down-

' ward thrust parallel to that of the body.

There is a powerful compression in these
small carvings and an adroit balancing
and counterpoising of thrusting forces.
In their analysis of the dynamics of the
human figure, they show an amazing per-
ception of reality; but their component

parts are pure sculptural devices for the realization of this expression.

The step beyond the Huon Gulf style is the creation of ‘geometric
or abstract equivalents for the various parts of the human figure. In

a few regions in Melanesia, such a scheme or pattern,
often approaching pure geometrization, is used to
indicate the human form or the human head. This is
strikingly evident in a group of wooden knives from
the Banks Islands (Fig. 6 and cf. Fig 3), where the
body parts are presented as geometric shapes arranged
in an open pattern. Some designs, however, especially
those from the Papuan Gulf area on the south coast
of New Guinea, represent a schematic transcription on
a two-dimensional surface of three-dimensional forms
(PI. 33). When flattened out a rounded form such as
the forehead here becomes a wide crescent, and an open
mouth a deep V-shape. It is as though a frontal view
of a form were bisected and the two halves juxtaposed
on a flat surface. These Papuan Gulf patterns, often
restricted to the human head or face, are rendered in
wide lines or bands usually cut in low relief, and are
combined at times with purely geometric elements
(Fig. 7). The various parts of the design are further
defined by the extensive use of colour, the pigments
being white, black, and various shades of red. Many
of these designs approach an abstract interpretation of
life forms. Such near-abstractions occur in other parts

of Melanesia, particularly in New Britain, New Ireland, and the Sepik

River area of New Guinea.

Fic. 6.

Art of Melanesia. 151

The human form is rendered with very great variety in the sculpture
of Melanesia. This often spectacular and always aesthetically arresting
ast demonstrates conclusively that the expressive possibilities through
the use of the human form are almost limitless. It also reveals that
masterpieces will result in any art, regardless of tradition and motivation,

since capability, perception, and sensitivity are shared in common by
great artists of all times and places.

Fre. 7.

on m Te ae
‘i -) 2 i

rw. ow
F

29,

PLATE

Solomon Island

inches high.

1
2

jed wood.

at
i

ose-gral

figure in hard, cl

ing

tand

i]
nw

at ee oe
ae ate Wo it

Ay
pees 5 gp
oe re
Le

eral
Pm 4
fet,
: ie aie

L
¥

Amide

PLATE 30.

Fig. 1. Ancestor figure.
Fig. 2. Female figure.

52 inches high. Admiralty Islands.
26 inches high. Solomon Islands.

al ae

aren ee _ +t
Bears ree ey

in
ee
eT a=,
Pai)

BE esate
iat

ma A ji we A ee
rT ee i OL ae
ces Ricans ile
Tiee. nee ie ie a
2 was on a cater -
= + eek tar ;
nif. a
ee rah

& Fe} fat

re
pees

D
>

a
i

eer te ts
ir

a
¥
a
my

aaa

hart Paks

eee,

fi be o
Pati Meee
| aa | i ‘

- me

ee
; ips
‘ ny Asai
2

' ee, Pa ¢

7 a rig nl

shel Fay NY a ig

Seer ee
ae.

PLATE 31.

Fig. 1. Adze decorated with carved head. 20 inches long. New Hebrides.
Fig. 2. Canoe ornament with shell inlay. 11 inches high. Solomon Islands.

—e

ie
© ho i? an
mi eon

a
fis o

a ans

a
if

ed ay
mee

fae
Ait Nee

7 At a
- nl a
Laer ain ve

a C3
Laat | By

7 my =~

mae: pot Soe }
eT ee ie ; ‘ Pir. Als a - ;
Picea te teens Sah NS ow arco ab amis ede de tee Cees eae Ae aC Se
pe er oe = ee —s Vs a # 1 = ee hy Tok it) 7. rs =
Pee ehhs mites ay Pe Teen ech Abe OCS eet Mea ea caer Uy age
re mae es Pit, ra © Gay ™ i Le eee ea 9 ae we bor if ewe ci; ~ pee : “- re
ge ne a ae ao Mag bk a ee pores ify eek Yn . Wietaee ace Te

anew
=~ oer
i

“
es 5”

i
ie ul,
“ . _. é le a rag “J
cosy A 2 a a 0 Se Baas ey ne a" i
res) i : : t : ‘ anes ok st
cm 5 an _ > a 5 | ee al Re 7Py
op ne er eo | : o _s : : tty if. aCe
FIRE tg pS at : i
/ ae
C eo Le

a -- : yy

se Na (a ae 4
i i ny
I
so. a ae
ig tae os “fe aa tae
- ia aa cw
oF ee SRNL E te Sanh
apt i cada a eae
i: f Li id Ml

ree ae oo

sok TREN oem
ie

i Tea iu
A ae ee
ee ee tery ee

ea en
4 eee ft
eS ae

set De AY,

a
= err =

Bate

PLATE 32.

Handle of lime spatula. 10 inches long. Massim area, S.E. New
Guinea.

a ay

ir
—s

ah
in
=
—"-
ve Hi
J

‘.
Shi

/
>

i ce :
Riva tis.
rte oe

ca oe ny ak oe et
. ny ry, ae ele se te 4 i = as al
a, - say ea a i

: Pita pee ae

5 ee ot Tee
Thy oily “ate Pes uv
Pai ee of lee fel aa ; F a - f Oe
Lope a mee e = ae A Peet ae et
ees ge ame va! ee 7 “iL ae aa Bint a om in
a ge ie we high 7 ee Pa Ee inl
} eae he nme, i a _ + wy a og oat

— bas
Vics. «=

i
Oh ame he
* Pe

PLATE 33.

Shield from the Gulf of Papua. 30 inches high.

= ey
ay =

=ty!

A
Fi

ein
F elie t

‘

i ope)
vy

“eu,

buf
te hd

me, fl” yp
i fh a!

i eel
ty ae mye

{teres

nubentes = ee at

aye
0.3

ny:

Rs
é

ope
eee

om ak Fe
* ? ae y ; =

153

A Note on the Geology of the Albatross
Point District, Kawhia.

By A. P. MASON, Associate Geologist, Auckland Museum.

Abstract.

Molluscan and plant fossils collected from Albatross Point, Kawhia, indicate
that the Mesozoic rocks of the area, previously regarded as Triassic, are not older
than mid-Jurassic. The presence of a major fault is suggested to explain the
absence, west of Arawi Point, of 10,000 feet of Upper Triassic and Lower Jurass‘c
strata.

INTRODUCTION.

As the Kawhia section provides one of the most nearly complete
sequences of Triassic and Jurassic rocks in New Zealand it has received
considerable attention in geological literature. However, Albatross
Point, on the outer coast of the harbour, has been strangely neglected
in most accounts, evidently because of its isolated situation. McKay
(1884) did not examine the rocks at Albatross Point, but noted the
westerly dip of the beds on the west side of the Orangiwhao andesite
intrusion and considered that they belonged to the Wairoa or a younger
series. Henderson and Grange (1926) failed to collect fossils west ot
Arawi Point, but as the beds closely resembled the Upper Triassic beds
further east they were correlated with them.

In January, 1950, the writer had the opportunity of spending
several days in the Albatross Point area and this account is based on
observations made during that visit. The outer coast is extremely
rugged and it was not found possible to examine the section at Alba-
tross Point itself nor that between Rimurapa and Waioioi. Thus,
although the total thickness of beds exposed in the area is approximately
1,800 feet, 1,200 feet of these were not examined.

STRUCTURE.

The regional structure, from the Orangiwhao andesite mass, west
of Arawi Point, to the outer coast, is an assymetric syncline which
plunges in a direction a few degrees west of south. On the west wing
of the syncline the beds strike consistently at about 340° and dip to the
east at 40°. This structure is given surface expression in the trends of
the outer and inner coasts and in the fine example of a dip slope on the
inner side of the point. On the east wing, however, the disposition of
the beds is less consistent, due to the influence of the Orangiwhao
andesite intrusion, and it is probable that the westerly dip of the beds is
largely due to the upward movement of the intrusion. The geological
map of Albatross Survey District (Henderson and Grange, 1926) shows
two faults between Parapara and Te Angina to explain this irregularity

154 MASON.

PARAPARA
TE ANGINA

Po00204 Conglomerate fy» ara}
2 rs Sandrtone —sS Current bed ee f

“—Hsandstone 20

600° of strata
Nor examined

=e
a

RIMURAPA,

Serr es

i f) i
} ie 4 ron fl ” ©
i Po CUB oge saan pConglomerate
i pogoouke =. Conglomerate : 50’k

Bion5 oi with andy

iC z
4 silenses 207 Ff

lS pn CE eo eR cee ae wmerche

ae SE a er

1 Ct * Gt Fossiliter ous

conglomerate yore
N73/515 1 N73/5t4 f-- ;
engees Pa tk :
30! IM KA 44) 8andstone
et: ’
et IOC
=m » ,
. Vem axe ; a
i . Oma, ‘ 4 ji
\ Sandstones TH eta et r
N73/520 [7.7 3 pene s/s
Ve: Ne ms iviuvi
feagietens — with plants He. " | Sandstone
a oe ae ’ | a
a's} 79 ; Jwith plants
i — ah V.ve Vi ao’
H ar: it
I Gora | Sandetones :
Id occbcookan d Fo not |
iV iM. feonglomerdtes
\- . “4 if
| 25% xposed
4 ;
Catal fl |
H : *- . . s 4
hexposed} loon © FCOdrse
' : NV. Vdsandstone
not

:
: xposed
[ial ype ear se
——— wood tand¢tone i i
= =e oe eam * eur:/ j
é nt

WCurreni~bedded

)>. “4 sandstone

6OO0!fof strata
WAIOIOIL not examined ,

SW
4

Current~bedded Pa ;
N73/516 sandstone 20! me sg p Orargiwhao
N73/518 carbonaceous bangs 15-20 4 WT cintrusive)
N73/519 Sandstone and

conglomeratet

Text fie. 1. Stratigraphic columns for Albatross Point area.

Geology Albatross Point. 155

in disposition of the beds. The succession contains several distinctive
beds (massive conglomerates and fossiliferous sandstones) which allow
correlation of the two arms of the syncline.

STRATIGRAPHY.

The entire sequence is obviously of shallow-water origin, with thick
bands of coarse conglomerate, current-bedded sandstones, and abundant
plant remains (Text-fig. 1). The lowest beds seen were those exposed in
the coastal section from 15 to 50 chains south-south-east of Waioioi.
The coast is here a strike coast, so that no great thickness of strata was
observed. The lowest bed is a gritty sandstone with conglomerate lenses
which is succeeded by 15 or 20 feet of alternating sandstones and car-
bonaceous bands with indistinct plant remains and fragments of wood
up to 2 feet in length. There then follows a massive, current-bedded
sandstone at least 20 feet thick (upper limit not observed).

The inaccessible nature of the coast prevented examination of the
500 or 600 feet of strata exposed between Waioioi and Rimurapa. At
Rimurapa, a hard, coarse, gritty sandstone is overlain by 3 feet of
coarse, soft sandstone that contains wood fragments and occasional
plant remains. This is followed by alternating sandstone and carbon-
aceous bands, and there is then a gap in the sequence of approximately
100 feet due to lack of exposures. The next beds seen are sandstones,
erits, and conglomerates (at least 25 feet thick) with fragments of wood
up to 3 feet in length. Following this are 70 feet of alternating sand-
stones and shales, in bands 1 foot to 2 feet thick, with well-preserved
plant remains. Next is a massive sandstone (80 feet) containing lenses
of fine conglomerate and layers very rich in Gervillea n. sp. This is
overlain at a sharp contact by at least 20 feet of conglomerate (upper
limit not seen) with lenses of sandstone. The pebbles in this conglomer-
ate have a maximum length of 6 inches.

Precipitous cliffs prevented examination of the succeeding 600 feet
of strata and the next beds seen were the gritty sandstones exposed
along the strike coast on the inner side of the point. At Parapara, these
are overlain by 6 feet of coarse conglomerate (pebbles up to 3 inches
long), which in turn are followed by shattered sandstone—the highest
bed seen in the area.

The section on the eastern arm of the syncline, between Parapara
and Tokatapu, is similar to that at Rimurapa. The massive conglomer-
ate, which was the highest bed seen at Rimurapa, is exposed on the west
side of Te Angina. It is approximately 50 feet thick and is succeeded
by alternating bands, 2 to 3 inches deep, of coarse and fine cross-bedded
sandstone. The lowest bed exposed is a coarse, cross-bedded sandstone
with carbonaceous bands and rare wood fragments. It rests directly on
the Orangiwhao andesite mass, about 300 yards south-east of Tokatapu.
Isolated outcrops of sedimentary rock occur further east involved with
the andesite, but their stratigraphic position 1s uncertain.

Marine fossils were collected from the massive sandstone at Rimu-
rapa and from its equivalent east of Te Angina, whilst well-preserved

156 Mason.

plants occur in the underlying sandstones and shales. Fossils were also
found in beach boulders south of Waioioi. The writer is indebted to
Dr. J. Marwick and Mr. R. McQueen for the following identifications:

N73/516
N73/517
N73/518
N73/519
N73/520

Sheet Fossil Number

Gervillea n. sp.
Meleagrinella sp. x
°Astarte sp. x
?Tancredia sp. x
Cladophlebis denticulata (Brong.) | hf

x! N73/514
XxX! N73/515

Klatocladus sp.
Taentopteris spatulata McClelland
Cladophlebts australis (Morris)

x
xX xX

N73/514.—East side of Te Angina (Grid reference 234068).
N73/515 and N73/520.—10 chains east-north-east of Rimurapa (217077).
N73/516.—Beach boulders 40 chains south-south-east of Waioioi (219055).

N73/517 ae N a i boulders 15 chains south-south-east of Waioioi
(217062).

N73/519.—Beach boulder 20 chains south-south-east of Waioioi (218057).

The stratigraphic positions of the fossils are indicated in the
columnar sections in Text-fig. 1. Collectively, they indicate a Middle or
Upper Jurassic age for the Albatross Point beds. Correlation of the
pelecypod fauna with those of other parts of New Zealand suggests.
that the beds belong to the Temaikan stage, but present knowledge of
New Zealand’s Mesozoic flora is too incomplete to allow close correla-
tion of the plant beds.*

DISCUSSION.

The discovery that the rocks of the Albatross Point Block are
Jurassic in age is rather surprising, for at Arawi Point, 1 mile to the
east and on the opposite side of the Orangiwhao andesite intrusion,
Otamitan (Mytilus problematicus) beds occur at the base of the western
arm of the Kawhia syncline. East of Arawi Point, successively younger
beds, dipping consistently eastward, outcrop in regular succession for
a distance of 7 or 8 miles. Between the Otamitan (Carnian) beds at
Arawi Point and the Temaikan (Bajocian-Bathonian) beds at Te Maika
(with which the Albatross Point beds may be correlated) are approxi-
mately 10,000 feet of strata which are not represented west of Arawi
Point.

The Albatross Point Block is nowhere in contact with the main
mass of the Kawhia Mesozoic succession and it is therefore difficult to
interpret the relation between the two. However, it appears highly
probable that the area west of Arawi Point has been down-faulted in
relation to the country further east. South of Kawhia Harbour, the
coastal area between Tongaporutu and Marakopa is characterised by

* Private communications.

Geology Albatross Point. 13%

a series of en echelon faults which trend a few degrees east of north
(Henderson and Ongley, 1923; Marwick, 1946). These constitute the
ereat fracture zone that separates the Herangi Range from the narrow,
depressed coastal strip on its western margin. The northernmost fau:t
(Whakahau Fault) reaches the coast near Marakopa, and north of this
point the depressed coastal strip passes beneath sea level. However, 11
is reasonably certain that the fracture zone (and the Whakahau Fault }
continues north of Marakopa, and it is here suggested that the Albatros:
Point Block is a northern remnant of the depressed coastal strip of Te
Kuiti Subdivision, In Whareorino Survey District are two andesite
masses, Pehimatea and Whareorino*, which are closely similar to that
of Orangiwhao (Henderson and Grange, 1926, p. 66; Williamson, 1932,
p. 8). Both Pehimatea and Whareorino occur along the surface trace
of the Whakahau Fault, and as similar rocks do not occur elsewhere in
Huntly-Kawhia or Te Kuiti Subdivisions, it seems more than coincidence
that the Orangiwhao mass should occur on the northward continuation
of the same line.

An alternative hypothesis which should, perhaps, be considered 1s
the possibility of an unconformity in the Mesozoic succession east ot
Arawi Point. The palaeontological evidence, which, however, 1s incom-
plete, shows no sign of a break, but the sudden decrease in dip (from
75° to 40°) in the vicinity of the conglomerate band, half a mile north-
east of Arataura,, which marks the base of the Jurassic, may possibly
indicate an angular unconformity. It is unfortunate that no detailed
survey, comparable in scope to those made by recent workers in South-
land, is available for the Kawhia section to allow discussion of thir
suggestion.

ACKNOWLEDGMENTS.

The writer desires to thank Miss H. Macdonald and Mrs. R. L.
Oliver for the opportunity of visiting the area. He would also like to
express his appreciation to Dr. J. Marwick and Mr. R. McQueen fo~
their palaeontological determinations and to the Lands and Survey
Department for permission to publish the aerial photographs used m
Plate 34. He is also indebted to Dr. R. N. Brothers for helpful criticism
of the text.

REFERENCES.

HENDERSON, J., and GRANGE, L. L, 1926. The Geology of the Huntly-
Kawhia Subdivision. N.Z. Geol. Surv. Bull. No. 23 (n. s.).

HENDERSON, J., and ONGLEY, M., 1923. The Geology ef the Mokau Sub-
division. N.Z. Geol. Surv. Bull. No, 24 (n. s.).

HUTTON, C. O., 1944. Some Igneous Rocks from the New Plymouth Area.
Trans. Roy. Soc. N.Z., vol. 74, pp. 125-153.

McKAY, A., 1884. On the Geology of the Kawhia District. Rep. Geol. Explor.
during ISS3-84, pp. 140-148.

MARWICK, J., 1946. The Geology of the Te Kuiti Subdivision, N.Z. Geol.
Surv. Bull. No. 41 (n. s.).

WILLIAMSON, J. H., 1932. Te Kuiti Subdivision. 26th Ann. Rep. N.Z. Geol.
Surv., pp. 5-8.

* Hutton (1944, p. 151) classifies the rocks of Pehimatea and Whareorino as dacites.

PLATE 34.

Sa.

ite

he dotted |
a

T
S

ande

whia
1adO

Survey Department. )

Ka
1W

’
al)

trict
Or

dis
the

Lt
iL

int

Po
Vg

SS
ounda

e b

Albatr

e
yx1mat

of th
ppre

erap
s the

to
e

O
licat

| ph

111¢

the

{

permission O

yr
/

and

Land

(Photograph publ

te a F -
i :

z a

i ,

me :
*

wer vie wees eee
rs

Be ae tem hot

Caen
ee

See a nr ore as

eres aes re a

159

The Genus Rhopalimorpha Dallas
(Heteroptera, Pentatomidae.)

By J. G. PENDERGRAST, Auckland.

The original description of the genus Rhopalimorpha Dallas 1s
included in a key in a list of Hemipterous insects in the British Museum
(Dallas, 1851). Because it is inadequate as a generic description it has
been thought advisable to redescribe this genus. Similarly, White's
description (1851) of R. obscura is now of little value and this species
has been redescribed. . lineolaris Pendergrast, the other species
recorded from New Zealand, has been described in a previous paper
(1950) and identified material of both species is in the collection of the
Auckland War Memorial Museum.

The writer wishes to acknowledge his indebtedness to Dr. T. EF.
Woodward, of the Zoology Department, Auckland University College’,
for valuable suggestions and for his generosity in allowing examination
of Acanthosomatinae in his collection. He would also like to express
his gratitude to Dr. W. E. China, of the British Museum (Natural
History), for giving him access to the type material.

FAMILY PENTATOMIDAE.
SUB-FAMILY ACANTHOSOMATINAE.
Genus RHOPALIMORPHA Dallas, 1851.

Rhopalomorpha Mayr, 1866.

Body elongate oval; thorax and anterior half of abdomen ot
uniform depth; angle between scutellum and pronotum surfaces almost
straight, typical “hunched” appearance of sub-famuly lacking. Head
wide, more than half greatest pronotum width; length equalling or
slightly exceeding that of pronotum; head width: length:: about 1.25,
tylus conspicuous, rounded, projecting beyond juga; maxillary plate
region with protuberance in front of antennal base; rostrum never
reaching beyond intermediate coxae. Antennae about half body-length ;
fifth segment slightly longer than fourth, second equalling or slightly
exceeding third; first segment reaching or barely surpassing apex of
head. Pronotum flat, trapeziform, narrow; pronotum width: body
length:: 0.4; lateral margins straight; anterior angles blunt, without
terminal papilla; lateral angles not produced. Mesothoracic carina small,
ridge-like. Hind femur length: body-length:: about 0.27. Abdomen
slender, greatest width not exceeding half body-length ; without median

* Now of the Department of Entomology, University of Queensland.

160 PENDERGRAST.

ventral keel; ventral spine small, reaching scarcely beyond hind coxae;
sixth sternum in female with pair of dark circular setose areas, absent
on the seventh; in male genital chamber opens caudad.

Type: Rhopalimorpha obscura White, in the British Museum
(Natural History).

1. Rhopalimorpha obscura A. White, 1851.
Rhopalimorpha similis Mayr, 1864.
Rhopalomorpha similis Mayr, 1866.
Rhombocoris similis (Mayr) Walker, 1867.
Rhopalimorpha ignota Hutton, 1898.

Length: Female, 7.5-9.5mm. Male, 7.0-8.5 mm.

General Colour: Dorsally usually ochreous, sometimes olivaceous,
henuelytra darker. Ventrally much lighter except as detailed. Male
usually distinctly green.

Head: Dorsally coarsely punctured with black including mid-line ;
clefts bordering tylus sometimes marked with black. Lateral margins
shghtly raised, Ventrally glabrous except for pubescent gula and few
conspicuous punctures on genae. In front of antennal base maxillary
plate with inconspicuous rounded protuberance. Rostrum not reaching
intermediate coxae. Second segment of antenna longer than third
(1.2 : 1.0), greater parts of fourth and fifth segments dark brown.
yes dark red to purple; ocelli red.

Thorax: Pronotum and seutellum densely punctured except on
callus areas. Mid-line sometimes marked by lighter stripe obscured by
punctation ; stripe more obvious on scutellum, especially apically. Scutel-
lum apex acute. Hemielytron narrow, corium dark brown, densely
punctured, membrane buff, veins brown. Scent gland orifice bordered
above by less conspicuous plate than in Jmeolaris. Mesothoracic carina
very small, ridge-hke. Femora without obvious punctation.

Abdomen: Slender; connexivum somewhat developed, without
dark markings, partly projecting laterad of hemielytron, Venter dark
in mid-line; ventral spine small, slender, rarely extending cephalad of
hind coxae.

Female: Sixth sternum with pair of rather inconspicuous dark
circular setose areas. Valviters somewhat triangular, mesial margins
raised. Seventh sternum produced posteriorly into sharp median keel.

Male: Pygophor with strongly convex ventral posterior margin
bearing two patches of long bristles. Claspers partially hooked, blunt.

Type: In the British Museum (Natural History).
2. Rhopalimorpha humeralis Walker, 1867.
Described from Queensland, Australia.

3. Rhopalimorpha lineolaris Pendererast, 1950.

Genus Rhopahmor pha. 161

Labels for the following figures in the writer’s 1950 paper should
read as follows :—

Fig. 2. Rhopalimorpha lineolaris sp. nov. Apex of scutellum.

Fig. 3. Rhopalimorpha lineolaris sp. nov. Male. Ventral view

posterior abdominal segments.

Fig. +. Rhopalimorpha lineolaris sp. nov. Female. Ventral view
posterior abdominal segments.

Rhopalimorpha obscura White. Lateral view thorax aiter removal oi
coxae. C, Mesothoracic carina: PN, Pronotum: SC, Scutellum: V5,
Ventral spine.

Rhopalimorpha obscura White. Lateral view head.

Comparison of outlines of right halves of pronota of: A, Rhopalimorpha
obscura White, and B, Acanthosoma haemorrhoidaie (L.).

Claspers. A, Rhopalimorpha obscura White: B, Rhopalimorpha lneolaris
Pendergrast.

] 62 PENDERGRAST.
REFERENCES.

BA bras, Wy: ae Iso. List Hemipterous Insects in Brit. Mus., Part 1,
pp. 193-197.

HUTTON, F. W., 1898. Trans. N.Z. Inst., vol. 30, p. 159.

MAYR, G., 1864. Verh. Zool. Bot. Ges. Wien, 14, p. 912.
————— 1866. Reise Osterreichischen Fregatte Novara um die Erde, 1857-1859,
Zool. Hemipt., pp. 74-76, Plate 2, fig. 14.

PENDERGRAST, J. G., 1950. Rec. Auck. Inst. Mus., vol. 4, no. 1, pp. 31-34.

WALKER, F., 1867. Cat. Specimens Heteroptera in Collection of Brit. Mus.,
Part 2, pp. 312 and 376.

WHITE, A., 1851. In Dallas; List Hemipterous Insects in Brit. Mus., Part 1,
p. 293.

163

Four New Species of New Zealand Land Snails
and the systematic position of Gerontia
cordelia Hutton.

By A. W. B. POWELL, Auckland Museum.

FAMILY FLAMMULINIDAE
Genus ALLODISCUS Pilsbry, 1892.

Two new species of the group of Allodiscus dimorphus are described
below. Each occurs in an area presumed to have been separated
formerly from the mainland. A third species, cooperi Suter, is related
to fallax, one of the new species, and is known only from the Poor
Knights Islands. The fourth member of the group, dimorphus Pfeiffer,
is widespread over most of the North Island.

The following key provides easy identification of the four species
of the group.

A. Spirals absent or microscopic:

a. Radials on penultimate 45-48... - dimorphus ( Pfeiffer )
b. Radials on penultimate 75-80... a) fallax n. sp.
c. Radials on penultimate 89-90 .. e, cooperit (Suter)

B. Spirals strong, fenestrating radials:
d. Radials on penultimate 100-104 .. - spiritus 1. sp.

Allodiscus fallax n. sp. Text fig. 2.

Shell of similar size and shape to dimorphus (Text fig. 1) but with
a darker and more clear-cut tessellated pattern, as well as more numer-
ous axials, 75 to 80 on the penultimate, compared with 45 to 48 in
dimorphus, Interstices of radials with 8 to 10 secondary radial threads
compared with 10-12 in dimorphus. Very dense and extremely’ fine
spiral threads, only on the latter part of the protoconch, the first post-
nuclear whorl and around the closed umbilicus. Whorls 5, including a
depressed protoconch of almost two whorls, faintly malleated, with
occasional axial growth lines and exceedingly fine dense spirals over
the second whorl.

Diameter, 7.25 mm.; height, 4.75 mm. holotype).

Locality: Oruru Bay near Knuckle Point, Rangiawhia Peninsula,
Northland. under leaf mould in stunted coastal scrub on steep cliff face
near head of bay, 29/1/1950.

The Rangiawhia Peninsula, tied to the mainland by low country
and extensive dunes, was probably formerly an island. The species
dimorphus has not been found in the area.

164 POWELL.

Allodiscus spiritus n. sp. Text fig. 3.

Shell almost as large as dimorphus, a similar but darker and more
clear-cut tessellated pattern, more than twice as many radials, 100-104
on the penultimate and dense distinct spiral threads over all whorls,
forming a regular interstitial reticulation with the secondary radials,
which number 8 to 10 for each interspace. Whorls ‘54, including a
depressed protoconch of 1# whorls, sculptured as in fallax but with the
spirals more distinct. The adult whorls are more rounded and not so
deep as in dimorphus. Imperforate.

Diameter, 7.0 mm.; height, 4.5 mm, (holotype). i

Localities: Waterfall Gully, Kapo Wairua, Spirits Bay, in Astelia,
Jan., 1950 (holotype) ; Unuwhao, 750-900 feet east of Spirits Bay, in
Astelia and under decaying leaves. i

The Cape Maria van Diemen-North Cape Block has a distinctive
land snail fauna obviously developed during former separation of the
area from the Northland Peninsula. The species dimuorphus is unknown
from this area also. :

Dentition: 32 + 1+ 32. Radula similar to that of dimorphus,
Central tooth with the base longer than broad, narrower in front and
with a single prominent, long cusp with a minute denticle on either side.
Laterals similar to the central tooth but with a distinct denticle on the
outer side only. Marginals at first longer than broad but broader than
long towards the extremities, with bidentate cusps and 3 to 6 denticles
on the outer side.

Genus THALASSOHELIX Pilsbry, 1892.
Thalassohelix prousei n.sp. Text figs. 4, 5, 6, 8 and 9.

Shell similar to that of zelandiae but peripheral carina almost
obsolete, spiral sculpture much stronger and axial growth lines weaker,
a wider umbilicus and a characteristic light zone surrounding the
timbilical area, the epidermis being here much thinner, allowing the
white shell to show through. Whorls 5, including a low rounded proto-
conch of 1¢ whorls, bearing subobsolete microscopic spirals over the
last half whorl. Post-nuclear whorls intricately sculptured with dense
distinct spiral threads crossed by numerous somewhat irregular weak
axial growth lines and a very dense surface pattern of minute wrinkle-
striae. There are about twenty spirals on the penultimate. Umbilicus
open and deep; one-seventh major diameter. Spire more than half
height of aperture. Colour of epidermis pale horny.

Diameter, 5.5 mm.; height, 4.75 mm,

Locality: Paturau River, in small patch of bush on the property of
Mr.°H. S. Prouse.

Dentition: (14 +9) +1-+ (9+ 14) (Text figs. 8 and 9). Central
tooth with a narrow rectangular base and a relatively short stout meso-
cone, ectocones obsolete. Laterals with the addition of a short stout
ectocone. Marginals long, oblique, awl-shaped without ectocones or

denticles.

N.Z. Land Snails. 165

ZT NSS He
3 my. -
‘=. Sea oS

wee ol af Abn Ss : ;
MSI
re eto ass Nie

Fig. 1. Allodiscus dimorphus (Pfeiffer), Waitakere Range.
Fig. 2. Allodiscus fallax n.sp. Holotype, 7.25 x 4.75 mm.

Fig. 3. Allodiscus spiritus n.sp. Holotype, 7 x 4.5 mm.

Figs. 4-6. Thalassohelix prousei n. sp. Holotype, 5.5 x 4.75 mm.

FAMILY PARYPHANTIDAE
Genus RHYTIDA Albers, 1860.

Rhytida forsythi n.sp. Text fig. 10.

This is a miniature relative of dunniae with an adult size of less
than half the linear dimensions attained by that species. The egg also is
approximately half the size of that of dunniae. The peripheral carina
is not so sharply keeled. The dentition resembles that of dunniae except
for the central tooth, which is the same size as adjacent laterals, not half

the size.

Whorls 4, including a low rounded smooth protoconch of 14 whorls.
Post-nuclear whorls sculptured, as in dunniae, with dense anastomosing
radial wrinkles, These wrinkles are irregularly thickened at the

166 POWELL.

periphery but interrupted or spaced, not fused into a continuous ridged
keel as in adult dunniae. Umbilicus deep, about one-sixth diameter ot
the base. Colour uniformly fuscous,

Dimenstons of shell:
forsythi; diameter, 13.0mm.; height, 6.5mm, (Holotype)

> 13.5 mm. 3 5, 725mm. (Largest seen, —
Herekino Gorge)
dunniae: . 30.5mm.;__,, 15.0mm. (S. of Kaeo)

, 30.0mm.; _,, 16.0mm. ( Pekerau)
\, 250mm.t , 12.0mm. (Whangarei )
r 24.0mm.; ,, 110mm, (Type)
ay, 23.0mm.;___,, 11.75 mm. (Cornwallis )
Dimensions of egg:
forsythi: length, 2.6mm.; width, 2.15 mm.
dunniae: + 3.5 mm. ; re 275mm, (Cornwallis )
+ 4.5 mim. ; + 3.75mm. (Kaeo)
Dentition: Radula almost identical with that of dunniae. Formula 17 + 1 + 17.
The outer five laterals increase in size from the margin to the sixth, which is large

and massive, then they diminish gradually to the centre. Central tooth as large as
adjacent laterals,

Localities: (forsytht) one mile up Taipa Estuary, south side (A.W.B.P.,
20/1/1950) (Holotype) ; Whatuwhiwhi, Rangiawhia Peninsula, Doubtless Bay (D.
G. Forsyth) ; Oruru Bay, near Knuckle Point, Rangiawhia Peninsula (A.W.B.P.,
29/1/1950); Quarry up valley north side of Taipa-Oruru Road, Mangonui County
(A.W.B.P., 1/2/1950) ; Broadwood, Summit of Mangamuka; Owhata, Herekino
Harbour; Moerewa; Okaihau and Waipoua Forest (N. Gardner).

Localities: (dunniae) Kaitaia to Thames (fide Suter, 1913); Church Road,
ca. 4 miles from Awanui-Mangonui highway; head of Pekerau Valley, ca. 2 miles
S.W. of Lake Ohia; subfossil in dunes, Tokerau, Doubtless Bay; Takahui, Victoria
Valley, Kaitaia; Kaeo; Whangarei; Kauri Mountain, Whangarei Heads; Parua
Bay; Woodcocks; Mangawai Gorge; Kawau Island; Wade River; Cornwallis,
Manukau; Centennial Drive, Waitakere Range, at Waiatarua, and near Titirangi.

The species forsythi has a restricted range extending from Rangi-
awhia Peninsula and Mangonui westward across the Peninsula to Manga-
muka, Herekino and Waipoua Forest. Its distributional pattern inter-
polates with but does not seem to overlap that of the northern extremity
of the dunniae range. I have never found both species at any one
locality, but they do occur in an apparent haphazard proximity, espe-
cially between Mangonui and Kaitaia.

The differences in carination and dentition indicate that forsythi is
not merely a size mutant of dunniae. Otherwise it could be surmised
that lack of lime or some other deficiency’ is responsible for the size
disparity.

Reference to the N.Z. Geological Survey North Island Map, 1947,
shows that dunniae occurs mostly in association with sedimentary rocks
of the Oligocene-Miocene formations 15-18 and fosythi in volcanic
(No. 3) and Senonian (No. 13) areas. However, there are exceptions,
notably in the Waitakere area, where dunniae, not a “dwarfed form,”
occurs in association with a volcanic formation.

Reference to text figures (10 and 11) shows that forsythi is
undoubtedly adult at 13 mm., for it has the characteristic sag of the outer
lip near its junction with the body-whorl. Figure 11] is of an immature

N.Z. Land Snails. 167

dunniae from Pekerau; both to same scale. ‘The fully developed radula,
complete with central tooth and marginals, and the presence of eggs also
show that the species 1s mature.

Fig. 10. Rhytida forsythi n.sp. Holotype, 13 x 6.5mm.
Fig. 11. Rhytida dunniae (Gray) to same scale (juvenile)

I prefer to consider forsythi as specifically not subspecifically dis-
tinct largely on account of its local range and curious mode of occur-
rence, interpolated with dunniae but evidently nowhere inter-breeding.
The inference is that forsythi originated as the result of some previous
isolating factor no longer apparent.

_I associate the name of Mr. D. G. Forsyth with this species, since
he first brought the problem to my notice.

Genus DELOS Hutton, 1904
DELOUAGAPIA New Subgen.
Type: Gerontia cordelia Hutton

The small snail long known as Gerontia cordelia has been considered
2 rarity. The type locality was cited as Titirangi, but to my knowledge
it has not since been collected from the vicinity of Auckland. I know
of it, however, from a number of Northland localities: Maungakaramea,
near Whangarei (A. E. Brookes, 1928) ; the northern headland block,
Whangaruru Harbour (A. C. O’Connor and A.W.B.P., Feb., 1948 ;
Oruru Bay, Rangiawhia Peninsula (A.W.B.P., Jan., 1950) ; Cape Maria
van Diemen (mainland), Kapo Wairua, Spirits Bay and Kerr Point
near North Cape (A.W.B.P., March, 1949).

At Whangaruru and at Kapo Wairua this snail was found to be
arboreal, living in clumps of Astelia, epiphytic on the limbs of puriri
(Vitex lucens). At Kerr Point dead shells were abundant on the ground
under clumps of Astelia in the stunted and rather sparse herbfeld.
Again, at Cape Maria van Diemen, cordelia was associated with Astelia,
erowing on the ground, there being no tall vegetation at either locality.

It may be noted that shells labelled “Delos jeffreysiana, Cape Maria
van Diemen,” in the Suter collection are cordelia.

The dentition and salient features of the animal show that cordelia
is a carnivorous Paryphantid of the genus Delos.

The shell is very similar to that of Delos jeffreysiana except for
the dark reddish-brown marbled and streaked colour pattern which is
more in accord with that of the Melanesian-Polynesian genus Ouagahta,

16& POWELL.

The type of Ouagapia is the New Caledonian raynalit Gassies, a
large shell measuring 33 mm. x 17mm. Its dentition is unknown, but
the several small species ascribed to the genus have more aculeate teeth
than Delos. The formulae range from 12 +-0+ 12-13+0-+ 13 for
the Fijian ratusukuni Cooke, 1942, to (20-23) + 1+ (20-23) for the
Caroline Islands oualanensis Pease, 1866.

The dental formula for both the New Zealand Delos coresia and
jeffreysiana is 9+0 + 9 and that of cordelia 12 +0-+ 12. (Text fig.
7.) The teeth in all three are stout, relatively broad-based and slightly
hooked. In Delos typical, the second tooth from the centre is largest,
after which they regularly diminish. In Delouagapia the fourth tooth
fromthe centre is disproportionately large, then regularly diminishing
from the fifth to the twelfth.

at
9
Fig. 7, Delouagapia cordelia (Hutton).
Figs. 8, 9. Thalassohelix prousei n. sp.

The animal of cordelia is slate-grey except for the sole, which is
white. Surface reticulate veined, scarcely warty. Two deep parallel
grooves on the dorsal area run back from between the superior tentacles,
which are blunt, cylindrical and moderately long. Inferior tentacles
short, genital orifice on right below mantle collar, proboscis capable of
protrusion, foot narrow, transversely wrinkled and with a moderately
long pointed tail. __ |
~ Yoshio Kondo (1943, Occas. Pap. Bernice P. Bishop Mus., vol. 17,
No. 19, p, 247) found close relationship between the anatomy of Delos
and that of the smaller Melanesian-Polynesian species ascribed to
Ouagapia, but since the anatomy of the large-size type species, raynali,
is unknown, admitted the possibility of ultimate subgeneric division.

Although cordelia conforms with Delos .in shell features, the
Ouagapia-like colour pattern and in particular the disproportionately
large fourth lateral tooth would seem to warrant the subgeneric status
proposed above for this species.

The holotypes of the four species described in this paper are in the
Auckland Museum,

169

New Zealand Molluscan Systematics, with
Descriptions of New Species, Part 1.

By A. W. B. POWELL, Auckland Museum.

Family PECTINIDAE
Chlamys (Mimachlamys) taiaroa n. sp. Pl. 35, fig. 1.

Shell resembling dichroa in size, shape and coloration, but more
inflated and with different sculptural detail. In dichroa the radials ‘are
strong, broad and flat-topped with channelled interspaces throughout.
the concentric lamellae closely spaced, relatively strong but not spinose.

In taiaroa the radials are narrowly rounded and channelled only
in the early growth stages. After about 25mm. the radials spread to
broadly triangular with the addition of a weak margining radial on each
side and one or two secondary radials in each intercostal space. The
concentric lamellae are weak, scarcely apparent in the interspaces but
forming weak irregular scales on both primary and secondary radials.
“Camptonectes’”’ striations present but very weak compared with those
in dichroa. The primary radials range between 15-20 in tataroa and
18-21 in dichroa.

Holotype with 19 primary radials on the right valve and 18 on the
left. Anterior lug of right valve with six scaly radiate ribs. Colour
pink to orange-pink, alternating in intensity in the form of broad con-
centric zones corresponding with growth stages. Colour stronger and
zones more clearly defined on the inside.

Height, 43.0mm. Length, 42.0mm. Thickness, (2 valves) 16.0 mm. (holotype )

e 39.5 mm. - 40.0 mm. i i 16.0mm. (paratype)
37.5 mm. f 36.0 mm. ‘, y. 15.0 mm. >

‘ 41.5 mm. . 38.5 mim. i i; 110mm. (dichroa)

é 41.0 mm. * 39.0 mm, ty ‘ 10.0 mm.

i 31.0 mm. ‘3 31.5 mm. * “A 8.75 mm.

Locality: Off eastern coast of Otago, 50-70 fathoms (trawled by Captain J.
Black, Dunedin),

Holotype: Auckland Museum, presented by Mr. J. G. Smith, Dunedin. Para-
type in collection of Mrs. N. Gardner, Auckland.

The Chatham Island shells figured by Finlay (1928, Trans. N.Z.
Inst. 59, p. 269, Pl. 42, figs 45-48) do not represent typical dichroa,
which is quite accurately portrayed in Suter’s Atlas, Pl. 52, fig. 1.
Finlay’s material is finer and more delicately ribbed, but I have two
strongly ribbed typical dichroa from Kaingaroa, Chatham Islands. There
is insufficient material available to determine if there is more than one
species of the dichroa group at the Chathams.

A series of ten topotypes of dichroa from the stomachs of cod taken
at Port Pegasus, Stewart Island, are constant in their strong flat-topped
radials with channelled interspaces and non-inflated valves.

170 PowELt.

Young examples of Chlamys delicatula Hutton, a species that occurs
commonly on the continental shelf of Eastern Otago, somewhat resemble
taiaroa, but are readily distinguished by their shape, higher than broad,
more numerous radials and thicker, stronger shell.

Family MONTACUTIDAE
Genus TAHUNANUIA n. gen.

Type: 7. alata n. sp.

The shell described below is closely allied to a Victorian species,
Saxicava subalata Gatliff and Gabriel, 1910 (Proc. Roy. Soc. Vict. N.S.
23, p. 85). These shells, however, have nothing to do with Savicava
(= Hiatella), nor is Cotton and Godfrey’s (1938) location of subalata
in Eximiothracia any better (The Moll. S. Aust. 1, p. 136).

There is a pallial sinus in both Saricava and Exinuothracia and it
is especially deep in the latter, whereas the “‘swbalata” group has an
uninterrupted pallial line.

Other features of the “subalata”’ group are a single cardinal in each
valve, a long external ligament and. an oblique large and very distinct
resilifer, seated on the nymphs, A character common to both the Vic-
torian and New Zealand species is a slight radial ridge bordering the
inner margin of both adductor scars. The genus seems to be nearest to
Scintillona Finlay, 1926, but the hinge differs in the large oblique well-
formed resilifer, that of Scintillona being weak, long, narrow and lying
almost parallel to the dorsal edge. Further, the cardinal in the left valve
is small, broadly triangular and located close to the dorsal margin, That
of Scintillona is larger and projects below the hinge plate, but is not so
prominent as in Mysella.

It may be noted that T. Soot-Ryen, 1951, Antarctic Pelecypods, Sci.
Res. Norweg. Ant. Expeds., p. 33) has referred Mysella to the Mont-
acutidae and that family location seems to be preferable for all three
above-mentioned genera.

Tahunanuia alata n.sp. Text fig. 1, la.

Shell rather small, thin, dull-white, minutely granulated, ovate-
trapezoidal, somewhat inflated arcuately from beaks to posterior-ventral
extremity. Beaks nearer to anterior end, which is narrowly rounded
and slightly gaping. Posterior end broadly winged with an oblique flat-
tened posterior slope, Hinge-line long and relatively straight, the hinge
plate weak anteriorly, deep over the cardinal area and moderate along
the posterior dorsal slope. Hinge of right valve with a single deep
strongly projecting and forwardly inclined narrow cardinal in front of
the umbo, an obliquely triangular functionless space directly under the
umbo followed by a conspicuous oblique divergent resilifer with clear cut
edges, seated on and occupying most of the nymph, Left valve with a
weak broadly triangular cardinal situated on the upper half of the hinge
plate in front of the umbo. Ligament long, extending from the umbo
to the posterior slope. Muscle scars deeply impressed, narrowly ovate,
the anterior one with a radial ridge margining its inner face and a less
distinct ridge margining the inner edge of the posterior scar.

N.Z. Molluscan Systematics. 171

Pallial line entire.
Height, 5.5mm.; length, 10.0mm.; inflation (both valves), 3.6mm. (holotype).

Localities: Tahunanui Beach, Nelson (type); off White Rocks, Queen Char-
lotte Sound, 25 fathoms (Dominion Museum) ; off Hen and Chickens Islands, 26-30
nati Dominion Museum); off Mayor Island, 45 fathoms, Bay of Plenty (C
Williams ). :

Holotype: Auckland Museum.

Fig. 1 and la. Tahunanuia alata n.sp. Holotype, 5.5mm. x 10.0 mm.
Fig. 2. Tahunanuia trigonia n.sp. Holotype, 3.5mm. x 4.75 mm.

Tahunanuia trigonia n.sp. Text fig. 2.

Shell rather small, thin, dull-white, minutely granulate, trigonal,
narrowly rounded anteriorly and broadly winged posteriorly. Beaks at
about the anterior third. Somewhat inflated posteriorly by a strong
arcuate angulation running from the beaks to the posterior-ventral angle.
Hinge plate long and relatively straight, narrow but deeper over the
cardinal area. Right valve with a single, strongly projecting, narrow,
forwardly inclined cardinal and a conspicuous oblique divergent resiliter,
seated on and occupying most of the nymph. In this species the ligament
is set in a long narrow groove posterior to the beaks. Muscle scars small,
ovate, subequal, the anterior one bounded on the proximal edge by a
slight radial ridge.

Height, 3.5mm.; length, 4.75 mm.
Locality: Perseverance Harbour, Campbell Island, 18 fath. (Capt. J. Bollons).
Holotype: 1 right valve, Powell coll., Auckland.

The species is much shorter and more trigonal than alata and has
the umbonal-ventral ridge stronger and distinctly angulate.

{72 PAS ae WERE ee

Family CUSPIDARITIDAE his (2a
Cuspidaria willetti Fleming, 1948.

The type locality for this species is Chalky Sound, 14 fathoms,
Fiordland. It is now known to me-from the following Aupourian locali-
ties: Two miles off N.W. end of Motiti Island, Bay of Foie from
stomach of a tarakihi taken in 35 fathoms (Mr. Gordon Williams) ; off
ee 20 fathoms (Mr. W. La Roche) ; ‘trawied, ‘Bay of Plenty (Mr.

oss).

Localities for both frailli and fairchildi are:—
trail Hutton, 1873 ee ten l ty ewer

Stewart Island, 14 fathoms eke hee vere Strait, 15 fathoms; 96 fathoms,
14 miles N°N.E. of Mayor Island; off Opotiki, 20 fathoms: off Hen and Chickens
Islands, 25 fathoms; off Little Barrier Island, 25-30 fathoms : + miles off Leigh,
Hauraki Guli, 30 fathoms ; between Spirits Bay and Three ae Islands, 95
metres; off Great Island, Three Kings Islands, 98 fathoms.

fairchildi Suter, 1908

Off Flat Point, 75 fathoms (type); 96 fathoms, 14 miles N.N.E. of Mayor
Island; 84 fathoms, 14 miles N.E. of Mayor Island.

——

Family HALIOTIDAE
Haliotis (Sulculus) virginea crispata Gould.

1847—Haliotis crispata Gould, Proc. Boston Soc. N.H., 2, p. 251.
1890—Haliotis crispata: Tyron and Pilsbry, Man. Conch. 12, p. 109, PI. 16,
figs. 87, 88 (copied, U.S, Expl. Exped. f. 248, 248a).

Localkty: “Australia”? = New Zealand.

For many years I have recognised two forms of mainland virginea:
(1) a large southern one, predominantly greenish, with deep spiral
grooves and weak meandering radials, which become obsolete as the shell
reaches maturity; (2) a small northern one, variously and brightly col-
oured, the dorsal surface often bright orange, or variegated red and
green, with deep spiral grooves crossed throughout by prominent
meandering radial folds.

The difficulty has been to decide which form is true zirginea. Dr,
C. A. Fleming during a recent visit to the Australian Museum, Sydney.
kindly examined for me the original figures of wrginea in Chemnitz
(Conch. Cab. 10, pl. 166, figs. 1607 and 1608). He affirms that these
figures are definitely of the southern form. Mr. Tom Iredale then
suggested that Gould’s Haliotis crispata, described with a query, as from
Australia, really represented the northern New Zealand form of virginea.
This undoubtedly seems to be the case and Gould’s original description
quoted below fits the northern shell exactly.

“Shell small, very thin.and- delicate, of an elongated oval and more
than usually convex form, the surface marked with fine, regular, equal,
revolving threads, and with very delicate, branching, oblique, zigzag
ripples, which are almost equally conspicuous in the interior. The spire
is prominent, of a little less than three whorls, the apex nearly on the

NZ, Molluscan Systematics. 173

median line. The perforations are small, rounded, slightly tubular,
numerous and crowded, six or seven of them open; and external to the
series is a deep canal. The colour is br ight brick-red or red-lead colour,
having between the canal and the margin a few narrow and distant
yellowish-white stripes. The interior is ‘brilliant silver y, and somewhat
iridescent. Length an inch and three-eighths; breadth seven-eighths
of an inch.”

It is desirable to nominate type localities for both subspecies.
Gmelin’s material was from Cook’s voyages and the sources of the col-
lections were Bay of Islands, Queen Charlotte Sound and Dusky Sound.
The United States Exploring Expedition visited Bay of Islands, Akaroa
and Auckland Islands. I therefore nominate Queen Charlotte Sound as
type locality for virginea virginea and ey of Island for vwirginea
crispata.

Localities and dimensions of specimens for both subspecies are as
follows :—

virginea virginea Gmelin, 1790.

South Island of New Zealand, Stewart Island, Wellington and southern coast
ot the North Island. =m |

73mm. x 51mm. near Dunedin.

75mm. x 49mm. Ocean Beach, Stewart Island.

56mm. x 38.5mm, Kartigi Beach, North Otago. Normal Otago adults.
55.5mm. x 36mm. Croixelle Islands, Nelson. clits |
68mm. x 44mm. Lyall Bay, Wellington. (Extra large for Cook Strait.)
42mm. x 28mm. Island Bay, Wellington. (Normal Cook Strait adult.)
Other localities: Portobello, Dunedin; Oamaru; Timaru; 4 miles south of

Clarence River, Marlborough ; Cascade Po: nt, Westland: Kahurangi Point, West
Nelson,

virginea crispata Gould,

North Island, Bay of Plenty to North Cape.

40 mm. x 27 mm. Leigh, Hauraki Gulf.
40mm. x 27.5mm. Great Barrier Island.

Other-localities: Whangarei Heads; Bay of Islands; Doubtless Bay; Waikuku
Beach; Tom Bowling Bay; Cape Maria yan Diemen.

Family SE TOMATUSAD

Venustas punctulata multigemmata n. subsp. Pl. 35, figs. 2, 3.

.. Shell large, thin, tall-spired, pale coloured, sculptured with fine,
very numerous gemmulate spiral cords and intermediate threads. Whorls
eleven, including a low, minute, smooth protoconch of 14 smooth whorls.
First post-nuclear whorl with two spiral cords, second and third whorls
with three cords, fourth with four and increasing to sixteen primary
gemmulate cords and about nine plain intermediate threads on the
penultimate. Spire tall, almost one and two- thirds height of aperture.
Whorl outlines strongly convex, not angled. Colour buff, speckled with
light brown om the cords: between the gemmules. Some examples with
narrow irregular and intermittent light reddish-brown axial streaks,

ig POWELL.

Height, 54mm.; diameter, 50 mm. (holotype).
Height, 48 mm.; diameter, 44mm. (paratype).

Locality: 50 to 70 fathoms off Eastern Otago. Trawled by Mr. J. Black.
Holotype: Auckland Museum.

This is the largest of the punctulata series and the finest and most
delicately sculptured. It seems to occur abundantly at the type locality
and shows little variation. It stands nearest to punctulata ampla Powell,
1939, a stronger shell with fewer and coarser spirals, from shallow
water, Stewart Island.

Family PLANAXIDAE
Hinea braziliana (Lamarck) 1822.

Suter (1913, Man. N.Z. Moll. p. 194) admitted this species to the
New Zealand fauna, citing Bay of Islands but no authority, and Finlay,
in his 1926 commentary (p. 376), rejected the record for lack of definite
evidence of New Zealand occurrences. However, | can name two
records: Great Barrier Island (Rev. W. H. Webster collection, Auck-
land Museum), a dead shell, and Whangaroa Harbour (collected W. E.
La Roche, ca. 1924), one living example, now in my collection, Auck-
land.

The species is common in eastern Australia, Queensland to Victoria,
Tasmania, Lord Howe Island and Kermadec Islands.

Family NATICIDAE

Two extra limital Naticoids and one beach specimen of a New
Zealand upper Pliocene fossil deserve mention but, on the present
evidence, not inclusion in the Recent faunal list.

Conuber conica (Lamarck), 1822.

Two half-grown, well preserved examples collected by Mrs. L.
Worthy at Tauranga Bay, Whangaroa. This is a common East Austra-
lian species. At the same time and place a single well preserved example
of the Mitrid genus Areninitra was obtained. It resembles exasperata
Reeve and arenosa Lamarck but is much more slender. This also should
not be added to the fauna until more are found.

Propesinum umbilicatum (Quoy and Gaimard), 1833.

A single slightly damaged but comparatively fresh example from
Stewart Island, collected by Mrs. W. H. Harrison. This is a common
Tasmanian species. The Stewart Island shell has the typical colour
pattern but a slightly shorter spire than any I have seen,

Eunaticina cincta (Hutton), 1885.
A stained and rather old shell of large size (20mm. x 17 mm.)

picked up in beach drift at Paihia, Bay of Islands, by Mr. L. W. Delph.
The species is otherwise known only from Landguard Bluff and Te Piki,
uppermost Castleclifhan ( Pliocene ) New Zealand. The Paihia specimen
may have come from some raised beach deposit. The rusty brown

staining of the specimen certainly suggests that source,

“I
a

N.Z. Molluscan Systematics. |

Family CYMATIIDAE
Charonia capax Finlay.

1913—Septa rubicunda: Suter (not of Perry), Man. N.Z. Moll, p. 303,
Pl. 43, tf. 1

1926—Charonia capax Finlay, Trans. N.Z. Inst. 57, p, 397, Pl. 20, f. 67.

1926—Charonia capax euclioides Finlay, Trans, N.Z. Inst., 57, p. 398, Pl. 20,
f. 68. .

Type localities: Off Otago Heads, 20 fathoms (capax) and 40 fathoms
(euclioides ) ‘

When Finlay described capax and euclioides he had only the holo-
type of each. Eight additional specimens from the vicinity of the type
locality (60-70 fathoms off Eastern Otago) now before me, would on
Finlay’s criteria separate into seven capax and one euclioides. However,
the slender shape and smaller aperture seem to be entirely resultant
from adventitious whorl! acceleration caused by injury or deflection to
avoid adherent growth. Neither the stronger nodulation nor the
narrower shoulder cords, features claimed by Finlay as characteristic
of euclioides, are, in the present series, restricted to the alleged sub-
species. In fact, they seem to represent but one species.

Northern shells tend to vary considerably, with strong rounded
nodulation associated with a dark reddish-brown pattern for shallow-
water shells. However, those from the deeper. waters (continental
shelf) from Bay of Plenty northward are pale in colour, like the Otago
shells, but mostly strongly nodose.

There is, however, a second species of Charonia in northern waters
which I previously misidentified as capax. This is :—

Charonia rubicunda (Perry).

1811—Septa rubicunda Perry, Conchology, London, PI. 14, f. 4.
1924—Septa rubicunda: Bucknill, Sea Shells of New Zealand, Pl, 4, fd:
1933—Charonia cf. capax: Powell, Trans. N.Z. Inst. 63, p. 162.
1937—Charonia capax: Powell, The Shellfish of New Zealand, Pl. 14, f. 10.

Type locality: New Holland. Probably New South Wales.

A long series of New South Wales rubicunda now enables me to
claim this species as an occasional occurrence in Northern New Zealand
waters. It is broader than capa., has very few nodules on the spire
and the whole of the shell is crossed by closely spaced spiral cords with
deeply incised interspaces. These cords are somewhat variable in width
on the body-whorl, but in all cases they are uninterrupted by nodulation.
The coloration is characteristic, a rich pinkish brown with a bright
reddish-brown maculated pattern. The outer lip is sharply ledged
internally and chequered with clear cut dark reddish-brown rectangular
patches alternating with white.

Height (actual), 139 mm. (estimated) 142 mm.; diameter, 84mm. (1937 figured
specimen).

Localities: Near old stone wharf, Pilot Bay, Tauranga, six living examples
with eggs taken by Dr. C. E. R. Bueknill, July, 1922 (my figured example, 1937,
‘s one of these); Tairua, near Mercury Bay (Mrs. Stocker).

176 POWELL.

Monoplex australasiae Perry.

1811—Monoplex australasiae Perry, Conch. or Nat. Hist. of Shells (London),

* uJ ,

1873—Triton (Simpulum) acclivis Hutton, Cat. Mar. Moll. N.Z., p. 13.

1913—Septa cea ees puter (non Born, 1778), Man. N.Z. Moll. p: 305,

1915—Monoplex parthenopeum: Iredale (non von Salis, 1793), Trans. N.Z.
Inst., vol. 47, p. 459.

1926—Monoplex acclhivis: Finlay, Trans. N.Z, Inst., vol. 57, p. 398.

Finlay (l.c.) advocated the use of Hutton’s acclivis for the New
Zealand shells, stating that they differed from Australian examples in
having a longer canal and a different outer lip. However, after examin-
ing long series of both New. Zealand and East Australian shells I fail
to find any constant points of difference. The name parthenopeum was
given to a Mediterranean shell, but I have not seen specimens. Since
both the South African and Japanese forms of Monoplex show obvious
differences the best course seems to be the adoption of Perry’s name,
given to a New South Wales shell, for the Austro-neozelanic Monoplex.

Particymatium strangei (Adams and Angas).

1864—Triton stranget Adams and Angas, Proc. Zool. Soc., p. 73.
1933—Cabestana? labiosa: Powell, Trans. N.Z. Inst. 63, p. 159, Pl. 23, £. 9.
1936—FParticymatium strange:: Iredale, Rec. Aust. Mus. 19 (5), p. 307.

The New Zealand record is based upon a single well preserved beach
specimen from Takapaukura, Tom Bowling Bay. This specimen has
suffered injury at two points and this has caused whorl acceleration
resulting in an abnormally high spire. The varix on the body-whorl
opposite the aperture is an unusual feature, but can be matched in at
least one instance in a Kermadec series. These abnormalities undoubt-
edly influenced Iredale (1936) in doubting the identity of my New
Zealand record. Iredale (l.c.) stated that he was misled in accepting
the British Museum locality of West Indies for labiosum Wood, the
locality being unknown and the figure very like the Sydney shell. He
then suggested continuing the use of the name strange: for the New
South Wales shells and noted that Kermadec shells showed no differ-
ences. |

Mayena australasia vossi n. subsp. Pl. 35, fig. 4.

Shell of similar size to auwstralasia, fusiform, sharply shouldered,
with two nodulous spiral keels, the uppermost at shoulder the stronger
and the lower one at the suture. The nodules are strong, pointed and
fewer than in aistralasia, 4-6 between varices. Spire shorter than aper-
ture plus canal, but the canal is almost twice as long as in australasia.
Colour pale buff, weakly maculated, mostly on the nodules and as a
narrow peripheral line. Interior of aperture and labial parietal callus,
porcellaneous white. Epidermis buff, delicately and densely reticulated
by axial and spiral threads with microscopic short bristles at all points
of intersection. In australasia the reticulation is more dense, giving a

velvety matted texture.
Height (actual) 81.5 mm. (estimated) 83.0 mm.; diameter 43.5 mm. (holotype )..

Height (actual) 73.5 mm. (estimated) 77.0 mm.; diameter 43,0 mm.-(paratype ).

N.Z. Molluscan Systematics. 177

- Lecality: Eight miles east of Mayor Island, 70 fathoms, Bay of Plenty.
Holotype: Presented to the Auckland Museum by Mr. S. Voss, of Tauranga.

This proposed new subspecies seems to be nearest allied to Mayena
australasta benthicola \redale, 1929 (Rec. Aust. Mus. 17 (4), p. 174,
Pl. 41, f. 4), from the continental shelf, New South Wales. Both have
two rows of very strongly developed but sparsely spaced nodules. In
benthicola the spire is described as being ‘longer than the aperture”
(plus canal) and from the figured holotype there appear to be at least
6-7. nodules. between varices. A deep water South Australian relative,
euclia Cotton, 1945, is a narrower shell with an even higher spire and
a-longer canal. In vossi the spire is considerably shorter than the aper-
ture plus canal, and the nodules are reduced to from 4-6 between varices.

Finlay (1926, Trans. N.Z. Inst. 57, p. 400) described the New
Zealand littoral form as Mayena zelandica, type from Tauranga, said
to differ from the Australian australasia in having a subobsolete lower
keel and many nodules on the peripheral keel (about 9 between varices
in the specimen figured).

In 1933 (Powell, Trans. N.Z. Inst. 63, p. 163) I pointed out that
only a few of the New Zealand shells have as many as nine nodules
between the varices, the average for eight New Zealand specimens taken
at random being 6.875, and for the same number of New South Wales
specimens 6.625; the difference is negligible. Further, the majority of
New South Wales examples examined have the subobsolete lower keel.

There are two colour forms in australasia, one buff to light yellow-
ish-brown indefinitely clouded or marbled with reddish-brown, and the
other dark reddish to purplish-brown with conspicuous white patches
where the main spirals cross the varices. These two colour forms seem
to be distributed irrespective of locality, depth and sex. It may be
noted that the dark-brown form tends to obsolescence of the nodular
peripheral keel over the last whorl.

The new subspecies vossi is a buff form, but another benthic
example from 45 fathoms off Mayor Island (S. Voss) is identical with
the shallow-water dark brown form.

It appears that whereas the larvae of the littoral form is probably
freely transported in surface plankton across the Tasman by means of
the East Australian Current, the benthic occurrences are below the effec-
tive influence of the current and thus we get comparative isolation in the
deeper waters of the continental shelves of the two countries, resulting
in local benthic subspecies.

Family TONNIDAE
Tonna dolium (Linnaeus). Pl. 35, fig. 5.
1758—Buecinum dolium Linn. Syst. Nat. Ed. 10, p. 735.
1952—Tonna dolium: Tinker, Pacific Sea Shells, Honolulu, figures facing p 136.
Tropical Pacific and Indian Oceans.
I now have three New Zealand records of this handsome tropical

shell. (1) Tokerau Beach, Doubtless Bay (Mrs. F. Bloomfield,
2/9/1950) ; (2) Off Whangaroa, in crayfish pots obtained by Mr. Eric

178 POWELL.

Sanderson, specimen now in collection of Mrs. I. Worthy, Patumahoe,
Auckland; (3) Mount Maunganui Beach, obtained during winter of
1950 by a visiting American collector.

| have examined all three specimens, which are in a good state of
preservation, especially the Whangaroa one, which was inhabited by a
hermit crab. This shell has full colour and a high gloss with no signs
of wear and evidently was living at the locality shortly before it was
taken,

It is probable that these odd occurrences represent survivors from
occasional drift from warmer areas of the South Pacific during abnor-
mally prolonged spells of northerly winds, They possibly reach here
in their early post-larval stages with planktonic drift, but occurrences
are evidently too infrequent and too sparse for mating and permanent
establishment of the species in New Zealand waters.

Two of the three records are from sparsely inhabited areas. The
human factor is unlikely to have accounted for the presence of these
shells, so the inference is that the species can reach here and survive
in these waters.

It seems best to admit the species to the fauna, for its claim is
equal to that of other “‘extra-limital’’ Northland inclusions in the faunal
list: eg., Xenophaliuin royanuim Iredale, Cymatilesta bolteniana A,
Adams, Recluzia lutea Bennett, Hydatina physis Linnaeus and a number
of others.

Family CASSIDIDAE
Xenophalium (Xenogalea) matai n. sp. Pl. 35, fig. 6.

A member of the pywruim group but small, much more slender, minus
nodules and with the columellar callus-plate less expanded, not dilated
and restricting the false umbilicus to a very small opening.

Shell small, rather thin, narrowly ovate. Spire about one-third
height of aperture. Whorls 64, including typical protoconch of 24
whorls. Labial varix recurved, smooth and slightly thickened behind.
Pillar with seven distinct but weak oblique plications and the usual ridges
bordering the base of the columellar-callus. Callus-plate thick but rather
narrow, set tightly to the columellar, the free distal edge bluntly rounded
and bridging a small rounded false umbilicus. Surface smooth except
for a few spiral incised lines on the spire and on the base. There are
about ten spiral lines on the early spire-whorls and four or five rather
distantly spaced on the base. The only axial sculpture is in the form of
weak irregular axial folds. Colour pale pinkish buff, with a very weak
pattern of pale purplish-brown blotches, arranged on the body whorl
in five spiral series, the uppermost at the suture.

Height, 45.5 mm.; diameter, 31.0 mm. (holotype).

Locality: Beach Harbour, Breaksea Sound, Fiordland, alive on intertidal mud-
flat (A.W,.B.P., Nov., 1934, on N.Z.G.S. Matai).

Holotype: Auckland Museum.

The species resembles pyritin in colour pattern, lack of labial denti-
cles and in the presence of spiral sculpture, and labiatum in its narrow
shape, small false umbilicus and thickened labial varix, but is quite dis-
tinct from either and cannot be considered hybrid.

N.Z. Molluscan Systematics. 179

Family MURICIDAE

Genus UTTLEYA Marwick, 1934.
Uttleya williamsi n. sp. Pl. 36, figs. 5, 5a.

Shell small, narrowly fusiform. Spire tall, acuminate, a little less
than height of aperture. Protoconch high, narrowly conical, of three
and a half smooth lightly convex whorls, with a small central nucleus.
Post-nuclear whorls three, sculptured with regular, distinct, rounded
spiral cords, with linear interspaces. There are 9 cords on the ante-
penultimate and 11-12 on the penultimate. These cords become obso-
lescent on the upper and median portion of the last half-whorl, but those
on the neck and fasciole, numbering about eight, are strong right to the
outer lip. Aperture narrowly ovate, produced below into a short canal,
with a broad shallow sinus. Fasciole not defined. Inner-lip smooth,
rather wide, well defined and slightly excavated. Outer-lip evenly
arcuate and in profile, sinuous, with a broad shallow concavity over the
upper half. Colour deep buff, body-whorl broadly spirally banded, two
bands with a subequal band of the ground colour occupying the upper
half of the body-whorl followed by a second peripheral hght band and
a final broad band extending to but not over the fasciole.

Height, 10.75mm.; diameter, 3.7 mm. (holotype).

Locality: 25 miles north of Mount Maunganu, 18 tathoms, Bay of Plenty.
Several from the stomach of a fish; Moki (JLatridopsis ciliaris Forster), Mr.
Gordon Williams.

Dentition: Pl. 36, fig. 5a.

The dried animal in the holotype was extracted with the aid of a
wetting agent and a mount of the radula prepared. The dentition 1s
Trophonoid. Central tooth with three long cusps of equal size and length
and two half sized intermediates each placed nearer to the outer cusp
than to the central tooth. The base of the central is wide and shallow
with upcurved extremities.

Operculum horny, very thin, ovate, with subapical nucleus,

The species is nearest to aliiparana Powell, 1927, but that species 1S
more slender with more narrowly convex whorls, weaker sculpture and
more numerous (19-20) spiral cords on the penultimate.

Holotype: Presented to the Auckland Museum by Mr. Gordon Williams.

Uttleya marwicki n. sp. Pl. 36, fig. 6.
1934—Uttleva sp. Marwick. Proc. Malac. Soc. 21, p. 20, Pl. 2, fig. 13.

Shell ovate-fusiform, slender at first but suddenly inflated after the
first post-nuclear whorl. Protoconch high, narrowly conical, of three
and a half smooth, lightly convex whorls, with a small central nucleus.
Protoconch followed by a brephic half-whorl, smooth, except for three
distant thread-like varices. Post-nuclear whorls sculptured with distinct
rounded spiral cords with subequal interspaces, 8 on spire whorls 1n
holotype (increasing to about I4 obsolescent ones on the penultimate ot
the large Wellington example) : colour buff with a broad white zone

180 | PowELL.

encircling the body-whorl at and below the top of the aperture. In the
large Wellington specimen the spirals are obsolete on the body-whorl
except for six linear spaced cords. on the base just above the fasciole.

Height, 8.1 mm.; diameter, 3.6mm, (juvenile Holotype). |
Height, 20.0 mm.; diameter, 8.0 mm, (Hypotype of Marwick’s Uttleya sp. 1934).

Localities: Tahunanui Beach, Tasman Bay, Nelson (type); Wellington Har-
bour, 5-6 fathoms (fragmentary specimen in New Zealand Geological Survey,

Holotype: Presented to the Auckland Museum by the writer.

_ This Cookian Uttleya is apparently very rare. I nominate the juven-
ile Nelson specimen as holotype since it exhibits the sculptural detail
much better than in the large Wellington fragment.

A feature of the species is the sudden inflation after the first post-
nuclear whorl and subsequent wide and strongly convex whorls.

Family COLUMBARIIDAE
Coluzea mariae n. sp. Pl. 35, fig. 8.

Shell fusiform, with broadly conic spire and long, straight, tapered
canal. Whorls 8, including a smooth papillate typical protoconch of 2
whorls. Spire rather squat, less than half the height of the aperture
plus canal; angle 55°, outline strongly keeled just below the middle of
the whorls and excavated below at suture. Sculptured with distant
sharply raised spiral cords, three above and two below the peripheral
carina on spire whorls and about 24 on the base and neck; six of these
are stronger than the rest and occupy the base from level with the top of
the aperture to the commencement of the neck and canal. All post-
nuclear whorls crossed by numerous weak axial folds which crenulate
the peripheral carina, forming 23 to 24 blunt tubercles per whorl. The
axials become rapidly obsolete over the base. Colour uniformly white.

‘Height, 81.0 mm.; diameter, 21.5 mm.

Locality: 60-70 fathoms off Eastern Otago (trawled by Captain J. Black,
Dunedin).

~ ° Holotype: Auckland Museum.
The species is named in honour of Mrs. Black.

Mr. R. K. Dell, of the Dominion Museum, Wellington, informs me
that there is more than one species of the spiralis group in New Zealand
waters. The type of spiralis was neither figured nor adequately
described. No dimensions were given and the location cited was simply
“New Zealand. Mus. Cuming.” However, Hutton 1880 (Man. -N.Z.
Moll., p. 50) synonymised his Fusus pensum, 1873 (Kapiti Island) wit!
‘spiralis and Tryon, 1881 (Man. Conch. 3, pl. 85, fig. 593) figured Adams
and Reeve’s Fusus spectrum as a synonym of spiralis. The point is that
all these references are to a narrow shell with a spire angle ranging
between: 32° for pensum to 42° for Northland “spiralis” = Colum-
barium suteri Smith, 1915. Etim

The Otago shell represents a distinct species with a short broadly
conic spire (55°) and very numerous peripheral crenulations.

N.Z. Molluscan Systematics. 18]

Family COMINELLIDAE
Cominella virgata brookesi n. subsp. PI. 35, fig. 7.

A geographic subspecies with a distinctive colour pattern, distributed
from Whangaroa to Parengarenga Harbour. Suter (1913, Man. N.Z.
Moll., p. 290) gives the range of virgata as Bay of Islands to East Cape,
and the type locality is Bay of Islands.

Typical virgata has weak spiral lirae marked out as thin dark brown
continuous lines, three on the spire-whorls and 6-7, rarely 8, on the body-
whorl and base. In addition there is a weaker under pattern of flexuous
axial flammules. Interspaces of both spirals and axials approximately
equal, which results in a reticulated effect. The new subspecies has a
dense pattern of flexuous, narrow, dark-brown axial lines on an olive-
erey ground. The pattern has frequent meanders and interrupted lines,
caused by damage to the outer lip during growth. This pattern is
descriptive of well preserved examples. When erosion takes place, close
spiral dark lines appear from underneath the surface pattern, but both
are never visible on an uneroded surface. Worn shells exhibit approxi-
mately eight dark brown spirals.on the spire-whorls and sixteen on the
body-whorl and base, twice the number shown in the typical species.
Columella and outer lip bright orange as in the typical species.

At Te Hapua, Parengarenga, a stunted form of virgata brookesi
occurs. It has a tendency to develop stronger and more persistent axial
costae, some extending on to the body-whorl, but this is probably only
an ecological form.

Height, 32.5 mm.; diameter, 17.00 mm. (holotype).

Localities: Whangaroa Harbour (W. H. Webster Coll., Auckland Museum ) ;
Whatuwhiwhi, Rangiawhia Peninsula, Mangonui County (D. Forsyth) (type
locality); Aurere, Doubtless Bay (A. E. Brookes); Te Hapua, Parengarenga
Harbour (A.W.B.P.).

Holotype: Auckland Museum.

In the Auckland Museum collection typical virgata is represented
from the following localities: Bay of Islands ; Whangarei Harbour ; Port
Fitzroy, Great Barrier Island; Little Barrier Island; Alderman Islands ;
Big. Mercury Island; Narrow Neck and Takapuna, Auckland; and
Mount Maunganui, Bay of Plenty.

I am indebted to Mr. A. E. Brookes for first bringing this sub-
species to my notice, to Mr, D. Forsyth for the Whatuwhiwhi material,
and to Mr. V. W. Lindauer for a long series of topotypes of virgata.

Family NASSARITDAE

When Finlay (1926) provided the new name aoteanus for “Arcu-
laria coronata var.” (Smith, 1915) from 11-20 fathoms near North Cape,
New Zealand, he also remarked that ‘Powell and La Roche have col-
lected one or two specimens and these prove to be of the ‘glans’ type, not
like spiratus but close to particeps Hedley.”

One of the specimens, formerly in Mr. La Roche’s collection, is
definitely a particeps, but the other is a typical spiratus. No further
New Zealand examples of particeps are known to me, but I have addi-
tional Northland records of spiratus; listed below.

1&2 POWELL.

Dredging operations on the continental shelf , Northland, have pro-
duced several aoteanus but no spiratus, which seems to occur only in
shallow water. The specimen Suter listed from 38 fathoms near Cuvier
[sland as “Nassa suturalis dunkeri n. n.” (1908, Trans. N.Z. Inst., 40,
p. 350) was thus almost certain to have been an aoteanus. However,
his subsequent description and figure (1913, Man. N.Z. Moll., p. 398
and 1915, Atlas, Pl. 45, £.17) is of particeps, probably an Australian
example, and not the Cuvier Island shell. As pointed out by Finlay (Lc. )
the name diunkeri must be dropped, since it is merely a new name for
Nassa intermedia Dunker, 1866, non F orbes, 1844.

Nassarius particeps (Hedley).

1915—Arcularia particeps Hedley, Proc. Linn. Soc, N.S.W. 39 (4), p. 738.
1926—Nassarius cf. particeps: Finlay, Trans. N.Z. Inst. 57, p. 419.

Localities: Cavalli Islands, Northland (W. La Roche ca. 1924), one exaniple,
Powell coll., Auckland). The type locality is Port Jackson, New South Wales.

The species is characterised by three narrow brown spiral lines on
the spire-whorls and five or six on the body-whorl, in addition to a
maculated pattern.

Nassarius spiratus (A. Adams).
1852—Nassa spirata A. Adams, Proc. Zool. Soc. London for 1851, p. 106.

Localities: Tom Bowling Bay, W hangaroa, Matauri Bay and Cavalli Islands,
Northland; Kaitoke, Great Barrier Island (Powell coll., Auckland). The type is
from Swan River, Western Australia.

The species occurs commonly in New South Wales. Norfolk Island
and at Sunday Island, Kermadee Group. It has a maculated pattern but
lacks the spiral lines of the glans-particeps group.

Nassarius aoteanus Finlay.

1915—Arcularia coronata var., Smith, Brit, Antarctic “Terra Nova’ Exped.
1910, Moll. p. 85, Pl. 1, f. 28.
1926—Nassarius aoteanus Finlay, Trans. N.Z. Inst. 57, p. 419.
Localities: Near N orth Cape, New Zealand, 11-20 fathoms (type) ; half-way
between Outer Chicken Island and Mokohinau Islands, 62 fathoms; off Cuvier

Island ca. 40 fathoms; Tryphena and Kaitoke, Great Barrier Island, from crayfish
pets (Powell coll., Auckland).

The species has a channelled suture, coronated by short stout axials,
and is coloured uniformly pale yellowish-brown,

Family MITRIDAE
Austromitra planateilla Finlay, 1930.

The holotype from off Cuvier Island in 38 fathoms is an immature
shell measuring 10.8 mm. x 4.5mm. Two adult specimens in my collec-
tion measure 13.6mm. x 5.8mm. and 13mm. x 5.5 mm. respectively.
The first is dull white and is from 30 fathoms off Mayor Island, Bay of
Plenty, the second is light pinkish-brown and was found at Whangaroa.
A feature of the species is the pinched or narrowly arched character of
the axials,

N.Z. Molluscan Systematics. 183

Austromitra brunneacincta n.sp. Pl. 36, fig. 4.

Shell of moderate size for the genus, rather broadly fusiform, spire
about equal to height of aperture plus canal, strongly axially costate,
crossed by numerous clearly incised spiral lines. Colour buff, with a
pattern of narrow light reddish-brown spiral bands, two on the spire
whorls and five on the body-whorl. The first is subsutural and the fifth
covers the fasciole. Four of the bands show strongly within the aper-
ture. Whorls 6, including a smooth globular protoconch of 13 whorls.
Axials moderately strong, rounded, regular, vertical, continuous from
suture to suture and rapidly becoming obsolete on the base, 16 on the
penultimate. The whole surface crossed by distinctly incised spiral lines,
about 14 on the spire whorls and about 55 on the body-whorl, including
the fasciole. Columellar plaits four, oblique, uppermost strong.

Height, 10.6 mm.; diameter, 4.9 mm.

Locality: Mayor Island, 4 mile off west side, from stomach of a tarakihi,
Dactylopagrus macropterus (Forster), taken in 18 fathoms (Mr. Gordon Williams,
12/12/1947}.

Holotype: Presented to the Auckland Museum by Mr. Gordon Williams.

The species is nearest to planatella but in that species the shell is
larger, the axials are pinched or narrowly arched, the spiral sculpture
is much weaker and there is no colour pattern.

Austromitra zafra n.sp. Pl. 36, fig. 3.

Shell small, narrowly fusiform, uniformly dark reddish-brown,
sculptured with numerous rather weak narrowly rounded axials crossed
by closely spaced spiral cords with linear interspaces. Whorls 5, includ-
ing a smooth pupoid protoconch of 14 whorls. Axials slender, vertical,
about 25 on the penultimate. Spiral cords evenly developed over all
post-nuclear whorls, including the fasciole, 16 on penultimate. Columel-
lar plaits four, oblique, uppermost strongest.

Height, 5.25mm.; diameter, 2.25 mm.

Locality: Mayor Island, 1 mile off south-west end, from stomach of a tara-
kihi, taken in 50 fathoms (Mr. Gordon Williams, 20/2/1949).

Holotype: Presented to the Auckland Museum by Mr. Gordon Wilhams.

The style of sculpture is reminiscent of that of the Pyrenid genus
Zafra.

Miecrovoluta obeonica n.sp. Pl. 36, fig. 2.

Shel! small ovate-biconic, solid, sculptured with low distant axial
folds, about ten per whorl and faint incised spiral lines. Whorls 45,
including a smooth dome-shaped protoconch ot 14 whorls. Spire less
than height of aperture. Spire whorls with a subsutural fold and furrow
and two median spiral incised lines. Body-whorl with an additional
spiral line proceeding from the suture and four more on the upper part
of the base. A rather closely spaced group of four rounded spiral cords
on the neck separate the body-whorl from the smooth fasciole. Columel-
lar plaits four, upper three strong, lower one weak and very oblique.

184 _ POWELL,

Colour buff, with three broad white spiral zones on the body-whorl, each
with a colour pattern of yellowish-brown arcuate to chevron-shaped
narrow axial lines.

Height, 5mm.; diameter, 2.75 mm. (holotype).

Locahty: Off Spirits Bay, Northland, 30 fathoms (type) ; Discovery II St.
933, off Three Kings Islands, 260 metres.

The species is more ovate and inflated and has much weaker axial
sculpture than either biconica Murdoch and Suter, 1906, or cuvierensis
Finlay, 1930.

: Holotype presented to the Auckland Museum by the writer.

~~" Family MARGINELLIDAE

Very tew of the large number of species ascribed to this family are
known anatomically. ‘This fact, coupled with an almost general lack of
sculptural features in the shell, has induced either a conventional lumping
of species in the type genus or their assignment to a few genera or sub-
genera, often quite inappropriately.

Hahe, 1951, in “Hlustrated Catalogue of Japanese Shells” (No. 16,
pp. 101-107), proposed three new generic names for Japanese groups
and these seem to be represented:in Australian and New Zealand waters.
They are Volvarinella Habe, 1951, type. V. makiyamai Habe, Kogomea
Habe, 1951, type. Marginella novemprovincialis (Yokoyama) and
Microvulina Habe, 1951, type. M. nipponica Habe. Habe also used
Volvarina generically. However, since most of the species are based
solely on shell characters 1t would seem safer for the present to assign
subgeneric rather than generic status to the groups.

‘The acceptance of Habe’s new names would involve the following
changes in the New Zealand Recent faunal list:

Subgenus Volvarinella for amoena Suter, 1908; aoteana Powell,
1932; cairoma Brookes, 1924; fusula Murdoch and Suter, 1906;
hebescens Murdoch and Suter, 1906; lurida Suter, 1908; stewartiana
Suter, 1908; subfusula Powell, 1932 and possibly subamoena Powell,
1937. Subgenus Microvulina for angasi Crosse, 1870. Subgenus
Kogomea for Gibberula ficula (Murdoch and Suter, 1906).

Family CANCELLARITIDAE
Zeadmete barkeri n.sp. Pl. 36, fig. 1.

Shell nearer to the Castlecliffan (Upper Pliocene) pliocenica
Finlay, 1930, than to ¢raidli Hutton, 1873. It resembles pliocenica in
being shouleered, in having the spiral cords of the spire-whorls stronger
and more openly spaced and the pillar plaits much weaker and more
oblique than in trawl. It differs from pliocenica in having only sub-
obsolete axials, which render the upper spiral cords of the spire whorls
weakly crenulate but not gemmate, in the narrower and less distinct
shoulder and in the more broadly- ovate outline to the whole shell,

N.Z. Molluscan Systematics. 185

In trailli the shell is more obese than either pliocenica or barkert,
not shouldered, the spiral cords are closer, linear spaced, the aperture
wider and relatively shorter and the pillar plaits very strong.

Shell small, ovate, dull white, sculptured with revolving series of
flat-topped cords, four primary ones on spire whorls increasing to six
on the penultimate, three more on the upper part of the base followed by
about eleven linear-spaced, rounded cords, on the lower half of the base.
Spire about equal to height of aperture.

Height, 8.8mm; diameter, 4.75 mm. (holotype).
Locality: Off Mayor Island, 35 fathoms.
Holotype: Presented to the Auckland Museum by Mr. G. W. Barker.

Family DIAPHANIDAE
Austrodiaphana maunganuica n.sp. Pl. 36, fig. 7.

Shell small, thin, squat, ovate-cylindrical with flat spire and sharply
carinated shoulder having a weak concavity immediately below it.
Whorls 34, the small, smooth, globular protoconch scarcely visible above
the level of the shoulder. Aperture narrow above and expanded below.
Columella thin, separated from body-whorl by a long, deep, crescentic
umbilicus. Sculptured with about 18 subobsolete spiral lirae and more
distinct but irregular axial growth lines. Colour pale yellowish-brown
with two narrow colourless spiral zones, one immediately below the
shoulder and the other about the middle of the body-whorl. The columella
and umbilicus are colourless also. One specimen has the addition of
a darker-brown line margining the lower side of each clear zone.

Height, 2.1 mm.; diameter, 1.45 mm.

Locality; 24 miles north of Mount Maunganui, Bay of Plenty, from stomach
of a tarakihi, Daciylopagrus macropterus (Forster), taken in 18 fathoms (Mr.
Gordon Williams, April, 1948) (type); off Hen and Chickens Islands, 25 fathoms
(Finlay coll, Auckland Museum).

Holotype: Presented to the Auckland Museum by Mr. Gordon Williams.

The species is smaller and narrower than cole: Fleming, 1948, trom
Fiordland, the only other New Zealand member of the genus so far
described. A conspicuous difference is in the form of the umbilicus,
which is short in colei but long in maunganuica, Also there is a weak
shoulder constriction in maunganuica, but it is not nearly so pronounced
as in the Australian genotype bragieri Angas.

The small paratype of colei is about the size of maunganuica but
the differentiating characteristics are still apparent.

PLATE 35.

t
;

Fig. 1. Chlamys (Mimachlamys) taiaroa n.sp. Holotype 43 x 42mm.

Figs. 2 and 3. Venustas punctulata multigemmata n. subsp. Holotype (Fig.
ZY. 54 xe 50 arin.

Fig. 4. Mayena australasia vossi n. subsp. Holotype, 81.5 x 83mm.
Fig. 5. Tonna dolium Linnaeus, 83 x 73mm.
Fig. 6. NXenophalium (Xenogalea) matai n. sp. Holotype, 45.5 x 31mm.
Fie. 7. Cominella virgata brookesi n.sp. Holotype, 32.5 x 17mm.
o

Fig. 8. Colusea mariae n.sp. Holotype, 81 x 21.5 mm.

Pe chee ee Sai ee Fond ies he Ey Mm eee oe oe ele ee So

“. a - - 7 i
~ rar , os. tae r ‘ a nit 14 roa
he! a - 4 ~ en mn » 7 a -“." -_ : = a . . b - oe 7 a a 1 ae a ae ; _ 7 a ru
a a co ‘Sa ro a md - F * 7: " - Ts re = “Ss alt ‘ee = Las a a
4 - - 7 = if) a ." ial o x an a —_
= ai 7 ; i : a - Ld -| _— oF oo ras a ag ' = | aoe i
d ar = i a = : * Ts a a = i i _ —_
i 7 a i hs C : in } a = ~ Pha at ‘tia aarl ea - Se 7 ” a 2 7 a | na :
Ly : : is ie oe ,_f a. — ee ao = . = a
I tne oy 8 : 7 a - - : ie a a -
a i: a) : ni i a = = Ts 7 : ; a ne es i ba oe ines
= a a = Pe asi =e ss one, a ¥ a oc 22 eee
=) <4 “uP 7 Som | - ‘ i el = = a a i: pe
to o i te i ar hel 7 eat = ll ar = an 4 om) "I : 7 - cae il ee ie aa
‘* = in 2 7 . a * a. a ¥ + 14 2 oe | i= a 7 i ee - a ae i 7
- nm ee igre 7 ; - 7 7 Z lial
a XK r a 7) 4 de | dal : ae ey 7 a “ os es i = ake F ais a : a ae ao == i} "1.
a ee ee =) ais i Weaken es ee Dee Ee a i ke 1S
4 id oo i” —_ - ‘oe ae ie ok = a | tae a Ta fiat y ; . | id A
in ee eats 5 ie Seger i et a koe ee a a
a Lh - : 7 or a 7 oe re 7: 7 ‘ r 7 ., a : J : = ae he rE 7 Ps ee
. Vig ire cg Nee Thy ter ee ae ge ee es a
- ; se 7 Cae oO a eg ris ~ i =e = a oe 2 : = om, PSs! |e ec Caen
rr re sy fee eee ae y 5 a ha oe, 92 ee eee
:* ae ae er a ee ee | “# a tet 42 7 toe i?) in - 1 . Pp el oe tf
r Pd g ie hem Lae = a bi. in eS a i a i , art = i : a" a ae zy. —- =e ul
ee a 7 ee et. ee 7 i a ¥ 2 ea . ee ae a
7 ‘ih. =. iy . 2 oe > F oe re, ‘ bet a ee (a 7 fe Py
Pah bees Sl a en ne eee, ae ae io gee ees
a ae 7 in of a fot eas oy. le ee El nt i ms ea a. Ae ——s L. Bic os ss 5 arf ae 7 a) sae -
er ite! Pos ieee i = be ad Te igh ; : , Po We Ge hp eee ifens Oe
E:: 8 pee Se ee ak a oe - i 7 ee are ae
eo i = - = = “ a a te oe iy ji lh iiags a ‘ el at ee | a
A = . of —— i ae ieclape Ld = * cy Post =" Ld ——— — ae a a - a - a =
bs =o i is s - 7 7 a poe 4 ‘ ony “a SS i z a on o i 7 Coa | Bas eo =
WL Ce re eee Mig fre a Side 2 eles ee te ees ee De renee ag ie oe 5)" See eee
al ris a 7 = os - I | a a 7 i :
ee ie Ral bh ru ca oT " / "Pa _ A 7 — ie i, a tpn cay, = 7 e° a Sa - ne < : a Pars te hd a a ee
oo »-l a oo & a me Par) er eee ar = baie == 7 — 7 i: a = Pe) i ee en, = a a Lilie
p i eee a : < kt a, ee ed 7 re a Ta
se 7 an . Lan Te eae ee a i Se a es ethos
7 — x > =- — . ie 7 7 . ad a 5 i oe = ie os - - - 7
ho att a ; ee th 7 al a ‘ r ee a i=" -, si ee, 1 a ey ae ig ee rif = ist anes
= rhe — Eas oe es is mal = a: oO ne, sas a ee ee Se oe 2 ake 5
oa at amie 7 Sain == a is a. FE Pee ee a : ore haa : — en al * Pe et aes ae
- =f, a a a =i se 7 ‘ 7s a fue Pie a a files a = 7 he a" ~ = oe le Via
a aa [7 = 7 inf c > hs ‘a = 7 r = - a on — —_) ihe = - ra Fj ret oe =] I i —— ——
= i : = } a ee 8 Jie Ef A = a : o: ie oe = oS agpn*tr =: 2 if mee 7 7 alk Py a
— aoe - J = Z p+ eT mi 7 a a * ba co eT ea: had rn * == 5. 2
7 M-s 7 a < cd 7 At ; < ra i
er ao oe
_
a
- af.
= Foi a Py ie
=F in 7 Pe ele one = re - a in i =<"
a _ — as z Z ae ee ee —
i 7 - = Se. 7 neo a. _
ie ae ea eS oes col
ae ee mS iee ia ]
al eat a= a ants m eee ty rs
= 8 2 ee aaa _s ae mv = ea 7 a
a) +

OP hol
"a 7 i im a

F : ay il > =. : i *: ea 7 ‘i f ae 3
ah Pa ie oe o.-e St ee een ee, ae ‘@ : ‘
na oe ee a ' sl 7 fail ~~ fe
renk Gan aero Ree

on Hei ih. <r i a : a] Licavek « 7 —_
= a rt

Ge ies
“sy

is a

To “Se

: i a
| - 3 po:

-
.
wi am ales
oe ies a

= - i: :
= 7 ae a. a ssl a nee i ry. as

Sig
a

= =
: ** _ a
-

Ter or
=e Pe 4 he 73"
: sts feo |
: ey aa
a a ne Peer

o1 oe oe > a —

eee re -_ ¥ a oe oo

: - _ a : ot
1a...

tk i rh hal

PLATE 36.

5 m1

47
Holotype, 5 x 2.75 mm.

Holotype, 5.25 x 2.25 mm.

x.
=>

e, 8.8

loty

Sp.
Sp.

‘1 nl.

RC)

adimete bat

nie

*
‘|
va

at EAs

Ticrovoluta obconic

Sp.

4ustromitra safra n.

acincta Nn. sp.

Austromitra briunne

Holotype, 10.6 x 4.9mm.

Holotype, 10.75 x 3.7mm.

sp. radula.

Sp.

Uttleya williamsi n.

On of)
“te

aaees

a. Uttleyva williamsi n.

.
mw }

x 3.6mm

8.1
Holoty

i n.sp. (Holot

?
\

Uttleya marwick

Fig.
Fig. 7.

Austrodiaphana maunganuica n. sp.

1 x 1.45 mn

pe,

ate!
wii. js *

* Al heel t
ea ,
* a f
; i
ra One, F I
haa i
ed eh
i, iL
n
: }
PE :
| ' any
ian! SR j
;
“)
ne

187

Some Stray Tropical and Sub-Tropical
Sea Birds in New Zealand.

By E. G. TURBOTT, Auckland Museum.

Fregata minor (Gm.) (greater frigate bird).

The only New Zealand specimen in the Auckland Museum 1s one
which was included amongst records of Fregata ariel by Oliver (1930),
and had been incorrectly assigned to that species in the collection, It
has the following data: A.M. 111.1, female, collected at Lake Kimihia
(near Huntly, Waikato district), in 1911, by G. Clinch. The specimen
was mounted in flight. It is in adult plumage with characteristic grey
throat and foreneck. Dimensions are as follows: wing 5&3, tail 413,
tarsus 20, middle toe 67, culmen 98 mm.

Fregata ariel (G. R. Gray) (lesser frigate bird).

One record only of a-specimen of the lesser frigate bird from New
Zealand in the Auckland Museum collections is given by Oliver (1930).
that of an adult female obtained at Pahi, on the Kaipara Harbour, on
30th March, 1907. Cheeseman (1908) gives details of the capture of
this specimen, which was found on the farm of Mr. G. V. New. It was
apparently almost exhausted when captured and died a few days later.
The locality is about half-way between the east and west coasts and the
distance in a direct line from the sea (a little overestimated by Cheese-
man) is 15-20 miles. In this specimen, as noted by Cheeseman, the
collar on the hind neck is white, mottled with dark brown, without trace
of rufous. A few scattered feathers form the only rufous coloration
on the breast. ,

A second specimen was obtained in March, 1918, when it was blown
ashore during a heavy gale. This was a female and the locality recorded
was Auckland. It is in immature plumage, and the only rufous colora-
tion on the white areas is a faint wash on the crown and nape. A
mottling of dark brown appears on the forehead and crown. On the
underparts, an irregular band of greyish brown extends across the bellv
from the flanks.

The third specimen obtained was an adult male, in brilliantly
iridescent plumage, found dead at Panmure, near Auckland, on /th
June, 1922, by N. Freebairn. It was mounted, as in the case of the
two specimens above, being shown with the bright throat sac partly
inflated. |

In March, 1936, the partial remains of another specimen were
found at Te Kuiti, approximately 100 miles to the south of Auckland,
and 25 miles inland. This was a female, and in immature plumage.

L188 TURBOTT.

A recent example, and the fifth from New Zealand in the Auckland
Museum, was received from the caretaker of Little Barrier Island, Mr.
C. H. Parkin, who discovered the bird exhausted on. the ground near
the homestead on 4th March, 1951. Shortly afterwards Mr. Parkin
disturbed a harrier in the act of carrying it off, but was able to save
the specimen. It had apparently died before being attacked and was
still undamaged. The specimen is an adult female. It is in process of
moult, and numerous fresh feathers are to be seen coming out of the
sheaths on neck, mantle and belly. Where the new plumage has already
appeared it is iridescent, especially the lanceolate feathers of the head
and back which have a strong greenish and purplish sheen; and wings
and tail are distinctly iridescent. Compared with this specimen, the
adult female from Pahi previously mentioned is in duller plumage ; it
is faintly iridescent on the lower back, wings, tail and head, and has a
patch of worn brown feathers on the back immediately below the hind
neck. There is a much greater amount of rufous coloration on the
white areas in the present specimen, forming an irregular wash on the
underparts; the collar round the hind neck is mottled white and dark
brown mid-dorsally, and the feathers are chestnut at the tips. The wing
spread of this specimen was approximately 5ft. 9in. (175 cm.).

The following are observations from several informants which
record the movements apparently of the same bird during the previous
day, 3rd March. The bird was seen at noon at Laingholm, on the north
shore of the Manukau Harbour, near Auckland, by Mr. E. D. Willis.
A sketch of the bird in flight made by Mr. Willis shows the forked
tail, the distinctive shape of the bill and white ventral area. It was
seen a second time at 4 p.m. near South Head, Kaipara Harbour, by
Messrs. B. Roy and J. B. Herman, who noted the same distinguishing
characters, and believed that it was a frigate bird. It flew in to rest on
a tree, perching on a broken limb and remaining for three-quarters of
an hour, and appeared to be weak or sick. The wing spread was esti-
mated at 5-6 feet, and the reddish legs were noted. The distance from
Laingholm to South Head, Kaipara Harbour, is approximately 45 miles
in a N.W. direction, and Little Barrier Island, where it was found by
Mr. Parkin, lies about 50 miles E.N.E. of South Head.

This frigate bird was undoubtedly brought to New Zealand by the
particularly severe cyclonic storm experienced in the North Auckland
and Auckland districts frorit 27th February-lst March, 1951. It is of
interest that a similar occurrence was recorded in Australia, a frigate
bird of unknown species being seen near Sydney on 13th March, 1951,
shortly after a period of severe weather (Marshall, 1951). An account
of some other occurrences of storm-driven sea birds in eastern Australia
in relation to cyclones is given by Goddard and Hindwood (1951).

The following are the dimensions of the above specimens of
FP. arvel:—

A.M. 112.1: Pahi; wing 525, tail 315, tarsus 18, toe 57, culmen 92 mm.
A.M. 112.2: Auckland; wing 520, tail 315, tarsus 17, toe 58, culmen 86 mm.
A.M. 112.4: Panmure; wing 545, tail 343, tarsus 16, toe 61, culmen 85.5 mm.
A.M. 112.5: Te Kuiti; culmen 90 mm.

A.M. 112.6: Little Barrier I.; wing 525, tail 335, tarsus 17, toe 61.5, culmen

88.5 mm.

Stray Sea Birds. 189

Observation of Fregata minor. The writer is indebted to Mr.
H.R. McKenzie, of Clevedon, for information ona f rigate bird observed
at Kawakawa Bay, Clevedon, on 18th F ebruary, 1951. The observations
were made by Mr. J. G. Moffat, who saw the bird from a boat offshore
closely enough to give details of its plumage which suggest that it was
an adult female F. minor, The underparts were white, and region of
the throat and foreneck white or grey, but not black as in the adult
female F. ariel. There had been strong winds from the N.I:, for several
days before the bird was observed. It forms an addition to several
recent field records (from Masterton, August, 1949; Tauranga, May,
1950; Whangarei, June, 1950), all apparently having been adult females
of F. minor ( stidolph; Davenport; Turbott: 1950).

Sterna bergii Licht. (crested tern).

A specimen in immature plumage from Sunday Island, Kermadecs,
is listed by Oliver (1930). In addition, the Museum has the dried
remains of an adult picked up by H. R. McKenzie on the beach at Kapo
Wairua, at the eastern extremity of Spirits Bay, on 25th March, 1951
(McKenzie, 1952). Dimensions of the two specimens are :—

A.M. 138.1: Sunday Island; male; wing 320, tail 138, tarsus 27, toe 33,
culmen 57min, (1st April, 1910.)

A.M. 138.9: Spirits Bay; tail (worn) 166 approx., tarsus 27, toe 32, culmen
6L.5mm. (25th March, 1951.)

Portions of the plumage of head, back, tail and the upper coverts
of one wing are still attached to the specimen found at Spirits Bay,
showing that it was in breeding plumage: the crown is black, lower back
and wing coverts uniform dark grey and bill yellow with a dark area
at the base, The flight feathers are missing from both wings. The
coloration of the lower back and wing coverts is of a dark grey shade,
corresponding closely to the only example of Sterna bergii cristata
Stephens* in the Auckland Museum collection (A.M. 138.2: Cronulla
Reef, Sydney’). Unfortunately the wing length is not available, but it
is close to this specimen in other dimensions. ‘The distribution of
cristata is from the Malay Peninsula and the Riu Kiu Islands to Aus-
tralia and the central Pacific.

It is presumed that the immature specimen from Sunday Island also
belongs to this subspecies; the head is dark brown mottled with white on
the crown ; back and median wing coverts pale grey, some feathers faintly
washed with brown; lesser wing coverts darker grey; secondaries and
scapulars, on outer webs, dark brown; outer tail feathers (old) dark
brown, inner tail feathers (new) grey; four outer primaries (old) dark
brown, inner primaries (new) silver-grey; outermost primary short:
neck and underparts white,
Rr
*cf. Peters, “Checklist of Birds of the World,” II (1934); also Whittell and

Serventy, “A Systematic List of the Birds of Western Australia” ( 1948),

who consider that S.b. gwendolenae Mathews (Rockingham, Western
Australia) is of unsatisfactory status.

190 TURBOTT.

Sterna fuscata L. (sooty tern).

A specimen from Waitomo Caves, near Te Kuiti, in the Auckland
Museum is recorded by Oliver (1930). The following data on the
specimen are available: A.M. 87.1, male, collected near the Waitomo
Caves, after the cyclonic storm of March, 1918, by H. F. Smith. It is
in adult plumage, measurements being: wing 298, tail 194, tarsus 23,
toe 27, culmen 42.6 mm.

The three following specimens in immature plumage have also been
obtained :—

A.M. 87.65: Tauranga, approximately 5th February, 1936. Meas-
urements: wing 284, tail 117, tarsus 23, toe 27, culmen 38.7 mm. There
are well marked pale tips on the feathers of the back and wing coverts,
and the tail feathers.

A.M. 87.64: Manurewa, near Auckland, picked up after a northerly
gale, on 26th March, 1936. Measurements: wing 280, tail 119, tarsus
23, toe 25, culmen 38.8mm. Only a few white tips on the upper parts;
tail much worn, new tail feathers and coverts appearing.

A.M. 87.66: New Plymouth, on the beach near the Waiwhakaiho
River, 4th March, 1951. Only the tail and one foot of this specimen
were saved, the remainder being destroyed subsequently by dogs.
Dimensions are as follows: tail 115, tarsus 22, toe 30 mm.

Notes were made by the finder, Mr. M. J. S. Black, and include. the
following details: the whole upper surface blackish brown, with lighter
flecking on mantle and wing coverts ; primary feathers sooty black; fore-
neck and breast brown; abdomen and under tail coverts greyish white;
length of bill approximately 47 mm., wing approximately 295 mm:

Information on another specimen, apparently in transitional plum-
age from juvenal (or post-juvenal) to adult, has been made available
by Dr. C. A. Fleming, who found it on Muriwai Beach, west of Auck-
land, on 7th February, 1936.* The specimen was apparently placed in
the Auckland Museum, but unfortunately cannot now be traced.
Accordingly, the following detailed description is recorded (notes by
C. A. Fleming): S. fuscata; young; much decayed and sex not deter-
mined; wing 290, tail 145, bill 41, tarsus 24.5mm.; forehead white;
head, crown black; hind neck and general upper surface sooty black,

* This specimen, and the specimen from Tauranga (A.M. 87.65), were found: after
the cyclonic storm on Ist-2nd February, 1936. Two additional reports: of
birds blown ashore during the storm were received by R. A. Falla, who
included all the records in nature notes in The New Zealand Herald, 7th
March, 1936: they were a sooty tern, found at Te Awamutu (80 miles
south of Auckland), and a red-tailed tropic bird (Phaethon rubricauda
Bodd.) at Taupo. The storm was the most severe which had been experi-
enced in New Zealand since March, 1918, when the specimens mentioned

above of the lesser frigate bird and sooty tern were obtained.

Stray Sea Birds. 191

with most feathers margined with white; wings sooty black above,
marginal coverts, primaries and secondaries sooty black, grey beneath,
white quills, under coverts white; tail coverts black above, grey beneath,
white quills ; streamers white-quilled above, inner margin whitish becom-
ing sooty towards tips, outer margins grey and sooty, darker at extrem-
ity; bill and feet black. Notes.—Differs from juvenal plumage in that
the white parts of the adult are present; dark parts sooty black like the
adult. Differs from adult (breeding) in having white tips to body
feathers, and grey streamers.

Procelsterna cerulea albivitta Bonap.* (grey noddy).

Buller (1888) in recording the specimen which he obtained in the
early months of 1882 from Cape Maria van Diemen gives notes from the
collector, C. H. Robson. The specimen was found exhausted on shore
after a heavy S.W. gale; and the same informant stated that another
was “observed on the wing, one very calm day.”

It has been observed to the west of New Zealand by Cunningham
(1950). Two birds were seen at sea about 300 miles from North Cape
en route from Sydney to Auckland on 16th October, 1949.

On 16th January, 1951, on the return voyage from the Three Kings
Islands, the writer closely observed a flock of four grey noddies. This
was close inshore off Cape Karikari, to the north of Doubtless Bay,
North Auckland. The birds remained near the vessel for half an hour,
either flying overhead or circling widely at no great distance, once coming
down to rest on the sea. The white under wing coverts were clearly
seen. The weather was calm, with settled conditions for some time
previously.

Some further observations of interest were made to the south of
the Kermadec Islands when the writer accompanied the Danish
“Galathea” Deep Sea Expedition in February, 1952. A single grey
noddy was first seen on 13th February, and two on the following day at
a distance of approximately 350 miles south of Sunday Island, Kerma-
decs, and 340 miles E.N.E. of Auckland. (The position was roughly
35° 20’ S, 178° 50’ W.) A single bird was again seen on both 19th and
20th February, when the “Galathea” was again in nearly the same
position. The weather during this and the preceding period was almost
continuously settled.

There is some indication in these observations that this non-
migratory species wanders tairly widely, on occasions at least, away
from the immediate neighbourhood of the breeding stations, and possibly
also is a fairly regular visitor to northern New Zealand.

* Trinomial follows Peters, “Checklist of Birds of the World,” Il (1934). The
birds observed were identified as “grey” and not “blue-grey” noddies, and
are presumed to belong to this subspecies (distribution, Friendly and
Kermadec Islands, Lord Howe Island, Norfolk Island). It is unlikely
that the Easter [. or St. Ambrose [. (Chile) subspecies would reach this

area.

192 TURBOTT.

REFERENCES.

BULLER, W. L., 1888. A History of the Birds of New Zealand (2nd ed.),
London.

CHEESEMAN, T. F., 1908. Notice of the Occurrence of the Lesser F rigate-bird
10,205." ariel) in the North Auckland District, Trans. N.Z. Inst.,
Q, 265.

CUNNINGHAM, J. M., 1950. Occurrence of Grey Noddy in N.Z. Waters, New
Zealand Bird Notes, 3, 227.

DAVENPORT, J., 1950. Frigate Bird at Tauranga, Notornis, 4, 35.

GODDARD, M. T., and HINDWOOD, K. A., 1951. Sea-birds and Cyclones:
Some Interesting New South Wales Records, Emu, 51, 169.

MARSHALL, ae 1951. Frigate-birds near Sydney, Emu, 51, 80.
McKENZIE, H. R., 1952. Summarised Classified Notes, Notorms, 4, 187.
OLIVER, W. R. B., 1930. New Zealand Birds, Wellington.

STIDOLPH, R. H. D., 1950. Summarised Classified Notes, New Zealand Bird
Notes, 3, 204.

TURBOTT, E. G., 1950. Frigate Bird at Whangarei, Notormnis, 4, 35.

RECORDS

OF THE

AUCKLAND — INSTITUTE
AND MUSEUM

VoL. 4 No. 4

Published by Order of the Council:
Gilbert Archey, Director

Edited by: A. W. B. Powell,

Assistant Director

12TH FEBRUARY, 1954

CONTENTS

VOL. 4, No. 4

The Torehine Beds of Coromandel Peninsula.

By R. N. Brothers, Auckland University College, and A. P. Mason,
Auckland Museum a AY as 24 be Ar a)

Algae of the Three Kings Islands.

By V. J. Chapman, Auckland University College

Pohutukawa X Rata.
By R. C. Cooper, Auckland Museum

A note on the Occurrence of Chelisoches morio (Fabricius) on Pit-
cairn Island, South East Pacific Ocean (Dermaptera: Labiduridae).

By E. T. Giles, Auckland

New Records and Descriptions of Hemiptera—Heteroptera from the
Three Kings Islands.

By T. E. Woodward, University of Queensland

New Zealand Molluscan Systematics, with Descriptions of New
Species, Part 2

By A. W. B. Powell, Auckland Museum

A New Rail from Cave Deposits in the North Island of New Zealand.

By R. A. Falla, Dominion Museum, Wellington

A bird Census and some Recent Observations on Birds on Great Island,
Three Kings Group.

By E. G. Turbott, Auckland Museum, and P. C. Bull, Animal Ecology
Section, D.S.I.R. A “} xe hy .% a i.

Page

Page 2

Page

Page 2

193

199

205

213

235

241

193

The Torehine Beds of Coromandel Peninsula

By Rk. N. BROTHERS, Auckland University College, and
A. P. MASON, Auckland Museum.

Abstract.

Microfaunal evidence indicates that the Torehine Beds of Coromandel Penin-
sula are Waitakian-Duntroonian in age. This determination assists in limiting
the age of the First Period Volcanics of the area, .

INTRODUCTION.

Because of the economic importance of its goldfields, Coromandel
Peninsula was the subject of many geological reports in the period
1870-1910. The scattered outcrops of Tertiary sediments in the north-
ern portion of the peninsula were first examined in detail by McKay
(1886), and his report formed the basis of later.accounts given by Park
(1897) and McKay (1897). Fraser and Adams (1907) gave a detailed
description of the beds, grouping them as the Torehine Series which,
on the basis of earlier accounts by Park (1897) and MacLaren (1900),
they regarded as Lower Eocene in age. Since the publication of Fraser
and Adams’ account the Torehine Beds have received no attention in
geclogical literature.

One of the present authors visited several of the outcrops in 1946
in the company of Dr. B. H. Mason and it was then thought worthwhile
to examine the beds in the light of the modern standard Tertiary sub-
division. A further visit was made by the writers in 1951 and the
opportunity was taken to collect systematically from the several beds.
It is apparent from the accounts given by previous workers that the
beds were formerly far better exposed than they are today and, in
common with most of the later authors, the writers can add little to
McKay’s original (1886) description.

In the course of the present survey the following localities were
visited :—

(a) Torehina (= Torehine of earlier writers) 1 mile north of
Amodeo Bay.

(b) Cutting on main road 300 yards south of Tawhetarangi Creek,
Amodeo Bay.

(c) Valley of Umangawha Stream, 1$ miles east of Torehina
(= “west branch of the Umangawha River” of McKay, 1886,
and “Branch Creek” of Fraser and Adams, 1907).

(d) Coast 14 miles east of Cape Colville.
In addition, samples for microfaunal study were collected from a

cutting on the main road 1 mile east of Torehina.

Rec. Auck. Inst. Mus. Vol. 4, No. 4, pp. 193-198, 12th February, 1954

194 BROTHERS AND MASON.

STRATIGRAPHY.

The basal beds are “coal measures” which are best exposed in a
cutting on the main road 300 yards south of Tawhetarangi Creek. Here
the succession in descending sequence is as follows :—

(6) Dark sulphurous mudstone (at least 10 feet thick).

(5) Coarse sandstone with impure coaly bands (5 feet).

(4+) Light-coloured sandy shales with plant remains (5 feet).
(3) Greywacke conglomerate (5 feet).

(2) Light-coloured shales similar to (4).

(1) Pre-Tertiary basement (Manaia Hill Series).

At this locality the beds are intersected by an andesite dyke and also
by a minor fault.

The most nearly complete section of the beds is that exposed at
Torehina. The succession, as given by Fraser and Adams (1907, p. 55),
is —

Coralline limestone with foraminifera (top).

Calcareous sandstone.

Marly sandstone.

Sandy shales with carbonaceous material.

Conglomerates.

(Readers are referred to Fraser and Adams, Joc. cit., for a more
detailed account of the succession. )

The conglomerates are intraformational and consist mainly of
fragments of the sandy shales, although greywacke pebbles are more
abundant towards the upper limit. They rest unconformably on the
upturned beds of the Manaia Hill Series (Fig. 1), and together with
the sandy shales are equivalent to the beds exposed south of Tawheta-
rangi Creek. The marly sandstone is now poorly exposed, being repre-
sented only by weathered outcrops at the back of Torehina Beach. The
next point north of Torehina consists of a glauconitic shelly limestone
(“‘coralline limestone”) which grades downwards into a fossiliferous

calcareous sandstone.

At Torehina the main road to Colville turns inland and climbs over
the divide into the valley of Umangawha Stream. Towards the summit
of the road a blue-grey medium sandstone which is probably’ equivalent
to the fossiliferous calcareous sandstone at Torehina is exposed in the
road cuttings.

The section in the valley of Umangawha Stream is obscure, but
both the limestone and the calcareous sandstone can be recognised. Some
distance below these is a dark, carbonaceous mudstone which possibly
represents the basal beds of the coastal section.

The outcrop near Cape Colville is restricted to a cliff face approxt-
mately 50 yards in width. Fraser and Adams (1907, p. 58) give the

following succession :—

‘spog oulysioy, jo sdoidyno jo suummyjoo o1ydeisijyetjg "] (8Y—}xoL

195

$31435
IIH DIDUDW

73294 OF B yout | s:ajDIs ajowixouddy

i saypsawoljbuos
EOS/5EN §

saioys Apubs

2uo ySpuDsS

Wvadis fants

LOS/6EN IONVYVLISHMVL

Torehine Beds.

BOs ‘SOS/6EN

9OS/6EN
2u0 ysSpuNns
$noasD31|D5

CVNIHSYOL
JO 1SV3 37IW 1)

QVOd ATNAIOO

Due WsI2aWI}
PutppOsOD

f LOGISWYQV QNV |

'8ssvag 30 Sdad |

KATIVA VNIRSYOL
VHMVYONVYAN

196 BROTHERS AND MaAson.

Sandstones and mudstones (top).

shelly conglomerate (2 feet to 3 feet), marly sandstone (10 feet),
shelly conglomerate (1 foot).

Marly sandstone (15 feet to 20 feet).
Shelly conglomerate (20 feet), (highly fossiliferous).

The authors were unable to distinguish all the beds mentioned by
Fraser and Adams and it appears that the vigorous wave erosion at this
locality has greatly altered the exposure during the last 40 years. There
was no trace of the basal “highly fossiliferous” conglomerate and the
two samples collected for microfaunal study proved barren. The beds
are lithologically dissimilar to those at Torehina.

AGE AND CORRELATION.

Both in 1946 and 1951 samples were collected for foraminiferal
study. The stratigraphic positions of the samples collected are shown
in Text—fig. 1, and the writers are indebted to the late Dr. H. J. Finlay
and Mr. N. de B. Hornibrook, of the New Zealand Geological: Survey,
for the following age determinations:— ‘/~

Sheet Grid
Locality. Sample No. Reference. Age.
Torehina N39/f501 946834 Waiutakian-Duntroonian
N 39/1502 - No fauna
x . N39/4509 . ? Duntroonian
F N39/f510 : Waitakian-Duntroonian
N39/f511 ‘: “/ (a
. N39/{512 946833 ¥ vs
4 et N39/f513 t ,
Tawhetarangi Creek N39/{503 943823 No fauna
Colville Road N39/f505 964832 Waitakian-Duntroonian
Fe i N39/1508 . No fauna
Umangawha Valley N39/£506 976830 Duntroonian
P, ‘ N39/f507 976829 ? Duntroonian

Thus the foraminiferal faunas indicate a Waitakian-Duntroonian
age for the Torehine Beds at and near Torehina. The lithologically
distinct beds at Cape Colville could well be of different age.

The Yorehine Beds may be correlated with the younger members
af the Whangarei Group of North Auckland. Marshall (1916, pp. 89
and 93) remarked on the lithological resemblance between the Whangarei
Limestone and the crystalline limestone of the Torehine Beds.

MINERALOGICAL ASPECTS OF THE TOREHINE BEDS.

A qualitative check was made on the non-opaque, heavy mineral
content of sandstone samples from localities (a), (c), and (d) above.
In all samples the heavy mineral suites were limited in their content, the
few species present being the more stable ones that are typical of the
gereywacke undermass. These are apatite, brown biotite, garnet, titanite,
green tourmaline and zircon. One grain of common augite was located,

Torehine Beds. 197

but a source for this may be found in tuffaceous facies of the ereywacke,
The general absence of ortho- and clino-pyroxene confirms the position
of the Torehine Beds as antedating the main effusion of andesite.

Following a discussion of the evidence available, Fraser and Adams
(1907, p. 89) assign a pre-Jurassic age to the diorite that intrudes the
Moehau Series (pre-Jurassic) on the west flank of Moehau Range. The
lack of hornblende in the Torehine sediments suggests that this diorite
was not exposed to erosion in Waitakian-Duntroonian times but was
still contained by the pre-Tertiary country rock. This view is strength-
ened by the fact that diorite pebbles are absent from the basal con-
glomerate.

DISCUSSION.

In Coromandel Subdivision, the First Period Volcanics rest uncon-
formably on the Torehine Beds, which are now known to be of Landon
age and are, in turn, covered unconformably by volcanic rocks of the
second Period. Plant microfossils from Second Period rocks of Great
Barrier Island indicate an Upper Taranakian or Lower Wanganuian age
(Couper, 1953). The First Period Volcanics, therefore. are either
Pareoran or Southlandian.

ACKNOWLEDGMENTS.

The writers desire to thank Dr. B. H. Mason for permission to use
information obtained during the 1946 visit to the beds. They are also
indebted to the late Dr. H. J. Finlay and Mr. N. de B. Hornibrook, of
the New Zealand Geological Survey, for examination of microfaunal
samples.

REFERENCES,

COUPER, R. A., 1953. Plant Microfossil Dating of Some New Zealand Upper
Cretaceous Volcanic Rocks. N.Z. Journ. Sci. and Tech., vol. 34
(Sec. B), pp. 373-377.

FRASER, C., and ADAMS, J. H., 1907. The Geology of the Coromandel Sub-
division. N.Z. Geol. Surv. Bull. No. 4 (ns.).

McKAY, A., 1886. On the Geology of Cabbage Bay District, Cape Colville
Peninsula, Rep. Geol. Explor. during 1885, pp. 192-202.
1897. Report on the Geology of the Cape Colville Peninsula, Auck-
land. Parl. Paper C.—9, pp. 1-75.

MACLAREN, J. M., 1900. Geology of the Coromandel Goldfields. Parl. Paper
C—9, pp. 1-18.

MARSHALL, P., 1916. The Younger Limestones of New Zealand. Trans. N.Z.
Inst., vol. XLVIII, pp. 87-99.

PARK, J., 1897. The Geology and Veins of the Hauraki Goldfields, New Zealand.
Trans. N.Z. Inst. Min. Eng., pp. 1-105.

PLATE 3

Fig. 1. Angular unconformity between Manaia Hill Series (Jurassic) and
Torehine Beds (mid-Tertiary); coast at Torehina.

(io, 2, Crystalline limestone; valley of Umangawha Stream.

>" —-s

Z a “/ ; 7 tT |
. J a= 8 . yy yA,
E ; 7 - a4 = oe aCe a Als es oe 1 my md i
- i Pe. y : yet, foes! eae bl om =. ae Li ‘ dees ti
_ pi ita —e _* 7 > “a a - = pon ot or ce ts a ta se a at ee 7 a . | Wh meal he:
ee a Co ee a ] a 7 ‘ig ee et eee a | | ele aera es ay Ca ii er fe. om any
al - i r “5 - 7 7 a. ont — 2 o! A or is = iaen tilt —. a maf, a ie _ Ts it a hi or si) |
—- 3 o> ae i 4, ee ee: eee ee ee eee a ae STW Se pat
7 J tg, ot) ail a is fe a a te a . i ed a d 1 im = Na . a Be i iJ
o2ee ek — - tiga, i : =. = eS >=" s ea arth
am » id * :) See , te hy Le a - h a ue tf ~ = ba Lt) 4 "
i) "3 ie fara. ; ‘a? Ae, = ' = a ae a - r oo ts i s ¥ ‘al = ae
ah oul = fie i: ae i 4 ie nes a La el sel _ a
= ~ » Sa ne
eos ‘hs 2 - - a | hd
7 je cas

—) = oe i : Pa saat ie a

—. | é a) ; oe a a ZZ 7 - *
7 a 7 mel Pao a oe ae a < £ aa i ni a

a oe a Pe oe ot Z i: _ 4 ‘ota _ ed = ‘a "2 fx eh ib i ut =f A F al
Bae es ee ha MS et. ee ees eee eee eer “T ;
om _ =} ee Yi Fr ee be - a = he ec a ao te a7 : Z =o
7 =m eae =e a ce i: ‘a 7
T 7 am
i /

+ on) mn
—— a - a

ee a” =u

= i ayy
=

Mala lieiars oe

scat t

199

Algae of The Three Kings Islands,
New Zealand

By V. J. CHAPMAN, Auckland University College.

At the end of 1952 and the beginning of 1953 an expedition from
the Auckland Museum visited the Three Kings group of Islands. Mr.
J. Edwards, a member of the expedition, collected aleae from various
places, and these were handed to the present author for identification,
The marine algae were generally those that are to be found on rocky
coasts elsewhere in the Auckland Province (Chapman, 1950; Chapman
and Beveridge, 1950; Dellow, 1950; Carnahan, 1952). There were,
however, two outstanding items in the collection. The first was a new
species of brown alga allied to Sporochnus with unusual features inter-
mediate between typical Sporochnus species and the genus Bellotia. The
other was a new species of the genus Grateloupia, which will be
described in a separate communication by Miss J. Trevarthen. The
new records suggest that a more detailed study of the north coast of
the North Island should be very profitable.

General collection, S.E. Bay, Great Island, 8th January, 1953:

Chlorophyceae
Ulva sp. (sporelings only).
Cladophoropsis herpestica (Mont.) Kuetz.
Caulerpa sedoides (R. Br.) C. Ag,

Phaeophyceae
Ecklonia radiata (C. Ag.) J. Ag.
Xtphophora chondrophylla (R. Br.) Mont. var, minus J. Ag.
Halopteris hordacea (Harv.) Sauy. |
Perithalia capillaris J. Ag.
Sargassum verruculosum (Mert.) J. Ag.

Sargassuim sp. (material not sufficient for identification. Mr. Lin-
dauer has had the same material from the north but never
in a state to permit of identification—personal communi-
cation ).

Rhodophyceae
Caulacanthus spinellus (Hook. f. et Harv.) Kuetz.
Gelidium caulacantheum J. Ag.
Pterocladia lucida (R. Br.) J. Ag.
Vidalia colensoi (Hook, f. et Harv.).

Rec. Auck, Inst. Mus. Vol. 4, No. 4, pp. 199-204, 12th February, 1954

200 CHAPMAN.

Phacelocarpus labillardicri (Mert.) J. Ag.

Melanthalia abscissa (Turn.) Hook. f. et Hary.

Corallina gracilis Lam. var, lycopodioides Taylor.

Lithothamnion sp. (not fruiting).

Bostrychia arbuscula Hook. f. et Harv.

Ballia callitricha (C. Ag.) Mont.

Laurencia sp. (very young, possibly L. thrysifera).

Lophosiphonia macra (Hook. f. et Harv.) Fkbg.

Polysiphonia sp. (corymbifera?) 12-14 siphons.

Our knowledge of the New Zealand species of Polysiphonia is not

sufficient for accurate identification. P. corymbifera C. Ag. is the only
species recorded for New Zealand with this number of siphons.

General collection, N.W. Bay, Great Island, 8th January, 1953:

Chlorophyceae
Caulerpa sedoides (R. Br.) C. Ag.

Phaeophyceae

Perisporochnus regalis n. sp.

Carpophyllum plumosuim (A. Rich.) J. Ag.

Carpophylluin maschalocarpum (Turn.) Grev.

Xiphophora chondrophylla (R. Br.) Mont. var. minus J. Ag.
Sargassum sinclair Hook. f. et Harv.

Sargassum undulatum J. Ag.

Sargassum verruculosum (Mert.) J. Ag.

Landsburgia quercifolia Hook f. et Harv.

Rhodophyceae

Porphyra coluimbina Mont.

Nemastoma oligarthra (J. Ag.) Kylin.
Grateloupia sp. nov*

Grateloupia fastigiata J. Ag.

Corallina gracilis Lamour.

Calophyllis hombroniana (Mont.) Kuetz.
Gigartina circumcincta J. Ag.
Gymnogongrus nodiferus (C. Ag.) J. Ag.
Laurencia sp. (young).

South East Bay, Great Island, 7th January, 1953. (Two collec-
tions.) :

Chlorophyceae
Ulva sp. (sporelings ).

* To be described separately.

Three K ings Algae. 201

Rhodophyceae
Gelidium pusillum (Stackl.) Le Jol.
Gelidium pseudointricatum Skottsb. et Leur.
Catenella fusiformis J. Ag.
Laurencia thrysifera? J. Ag. (denuded).
Bostrychia arbuscula Hook. f. et Harv.

North West Bay, Great Island, Corallina zone, 5th January, 1953:

Chlorophyceae
Ulva sp. (sporelings).

Rhodophyceae
Coraliina gracilis Lamour.
Nemastoma oligarthra J. Ag.
Grateloupia sp. nov.
Laurencia thrysifera J. Ag. (?) denuded.

South East Bay, Great Island, 7th January, 1953:

Myxophyceae
Calothrix confervicola (Roth.) Ag.

Chlorophyceae
Rama longtarticulata (J. Ag.) Chapman (see Chapman, 1951).

Rhodophyceae

Gelidium pusilluin (Stackl.) Le Jol.
Catenella nipae Zanard.
Bostrychia arbuscula Hook f. et Harv.

Tasman Stream (fresh water), Great Island. Collections 30th
December, 1952, and 5th January, 1953:

*Qedogonium sp. (not fruiting).
Spirogyra sp. (not fruiting).
*Vaucheria sp. (not fruiting).
Ulothrix tenerrima Kuetz.
* Microspora sp.
*Rhigoclonium riparium Harv.
Scytonema cincinnatum (Thur.).
The species marked * were also collected from a pool at the edge
of the Tasman Stream.

Perisporochnus n. gen.

Plantis ex haustorio coniformi orientibus, axe primario ramis et
ramulis vestito, ramis ramulos spiraliter ferentibus; axe et ramis uni-
formiter constantibus e parvis cellulis, cum singula serie epidermal

202 CHAPMAN,

cellularum magnarum; sporangiis unilocularibus lateraliter natis in
paranematis dichotomose ramosis.

Plants arising from a conical holdfast, main axis clothed with
branches and branchlets, branches bearing branchlets in whorls; axis
and branches uniformly composed of small cells with a single epidermal

layer of large cells; unilocular sporangia borne laterally on dichotom-
ously branched paranemata.

Type: Perisoprochnus regalis n. sp.

Perisporochnus regalis n. sp. Pl. 1; figs 1, 2.

Plantis ad 24 cm longis orientibus ex haustorio coniformi rhizoidi-
bus septatis operto, axe primario per totam longitudinem multis ramis
et ramulis dense vestito; axe et ramis constantibus e parvis cellulis
cum singula serie magnarum cellularum epidermalium, parietibus crassa-
tis sed per foveas tenuiparietales distinctis; ramis ramulos spiraliter
ferentibus; ramulis plerumque simplicibus, raro cum laterali brevi, per
receptaculum tumescens et cristam capillorum terminatis; pedicello
2-5-ies longiore quam crista apicali ; capillis simplicibus, 40-70u diametro
in regione basali meristematica, apud apicem ad 140u expansioribus;
sporangis unilocularibus 30-34u longis, 11-11.5u latis, lateraliter natis
in paranematis dichotomose ramosis.

Plant up to 24cm. long, arising from a conical holdfast covered
with septate rhizoids, main axis densely clothed throughout its length
with numerous branches and branchlets; axis and branches composed
of small cells with one row of large epidermal cells, walls thickened but
studded with thin walled pits; branches bearing branchlets in whorls;
branchlets usually simple, rarely with a short lateral, terminating in
swollen receptacle and tuft of hairs; pedicel 2-5 times longer than the
apical tuft; hairs simple, 40-70 u diameter in basal meristematic region,
expanding to 140u at apex; unilocular sporangia 30-34u long x
11-11.5 u wide, borne laterally on dichotomously branched paranemata.

Type specimen in herb. Auckland Inst. and Museum.

The whole plant is much coarser than are species of Sporochnus
and it also differs from Sporochnus in the whorled arrangement of
branchlets on the branches. In this latter respect it is more like
Rellotia, but it differs from that genus in having the sporangia con-
centrated terminally instead of medially, and the paranemata are typically
Sporochnalean in being dichotomously branched. Anatomically it differs
from Sporochnus and Encyothalia in that there is no evidence of any
larger central cells, the entire main axis and branch axes being composed
of a mass of uniform thick-walled small cells, perforated here and
there, especially towards the periphery, with thin-walled pores, though
these are not true sieve plates. The outermost layer is composed of
large cells also with thin-walled pores, restricted to the inner wall and
the inner half of the lateral walls. Anatomically it appears to have
affinities with Perithalia, which also has a similar structure and lacks
the characteristic large central cells. The new genus therefore lies inter-
mediately between Sporochnus, Bellotia and Perithalia.

I wish to express my thanks to Mr. W. A. Crawley, of Auckland
University College, for the latin diagnoses. Dr. Drouet of the Chicago
Natural History Museum, identified the Scytonema cincmmnatum.

Three Kings Algae. 203

Fig. 1. Perisporochnus regalis, n. sp. (a) Branchlet with swollen receptacle;
(b) branchlet with branch; (c) main branch showing whorled lateral
branchlets; (d) single paranemata (a-b x 2; ¢ x Le}

Fig, 2. Perisporochnus regalis n. sp. (a) Unilocular sporangia on branched para-
nemata: (b) central cells of main axis; (c) t.s. main axis showing row of
large epidermal cells and small cells within; (d) Ls. epidermal cells
enlarged to show pits; (e) Ls. central cells showing pits.

204 CHAPMAN.

REFERENCES.

CARNAHAN, J. A., 1952. Inter-tidal Zonation at Rangitoto Island, New Zea-
land. (Studies in Inter-tidal Zonation, 4.), Pac. Sen 0 (14, Oo

CHAPMAN, V. J., 1950. The Marine Algal Communities of Stanmore Bay,
New Zealand. (Studies in Inter-tidal Zonation, 1.), Pac. Sct., 4 C1),
63.

1951. New Entities in the Chlorophyceae of New Zealand, Trans.
Roy. Soc. N.Z. 80 (1), 47.

and BEVERIDGE, W. A., 1950. The Zonation of Marine Algae at
Piha, New Zealand, in relation to the Tidal Factor. (Studies in Inter-
tidal Zonation, 2.), Pac. Sci., 4 (3), 188.

DELLOW, U. V., 1950. Inter-tidal Ecology at Narrow Neck Reef, New Zealand.
(Studies in Inter-tidal Zonation, 3.), Pac. Sci. 4 (4).

» PLATE

Perisporochnus regalis n. sp. type specimen (x

5 )

no
edd),

205

POHUTUKAWA x RATA

Variation in Metrosideros (MYRTACEAE)
on Rangitoto Island, New Zealand

By R. C. COOPER, Auckland Museum.

Abstract.

Attention is drawn to certain recent papers describing new techniques for the
study of variation in plants. One of these techniques, the p‘ctorialized’ scatter
diagram is illustrated, using specimens of IV/etrosideros spp. from a known hybrid
swarm on Rangitoto Island, and the results from an ordinary herbarium collection
and two mass collections are contrasted.

The method by which a pictorialized scatter diagram is developed
is explained in Anderson (1949 and 1952) and Anderson and Gage
(1952). Anderson (1949 and in press) has also explained the ecological,
genetical and mathematical criteria upon which scatter diagrams are
based. Stebbins (1953) pointed out that the method is “by far the
best yet devised for making the observer aware of a pattern of varia-
tion in respect of three or more characters which are varying simul-
taneously.” He added, “to be sure the selection of characters to study
and of their arrangement on the diagram are based on just as strictly
subjective judgment as are the descriptions of the traditional systematist.
But once the method is learned it is, like every other scientific method,
a valuable tool for making complex natural phenomena relatively clear
to the human mind without seriously distorting them.” The same author
also remarked that the method “has the advantage of being repeatable.”

The figures listed in the three tables below are the measurements
of certain characters of specimens of Metrosideros spp. (Myrtaceae)
gathered on Rangitoto Island, a circular volcanic cone situated in Hau-
raki Gulf at the entrance to Auckland Harbour. The measurements in
the first table were made on specimens collected by various botanists
over the last eighty years and preserved in the Cheeseman Herbarium
at the Auckland Museum. The measurements in the second and third
tables were made on specimens gathered as “mass collections” by the
writer immediately prior to the preparation of this paper. These collec-
tions are also preserved at the Auckland Museum.

Measurements of leaf width and stamen length were chosen as
abscissa and ordinate respectively because they varied consistently and
could be measured accurately. The other three characters, leaf length,
internode length and calyx length, were indicated by rays from each
dot on the scatter diagram. Limits of the three grades of each of these
characters were chosen so that extremes associated with higher values
for leaf width and stamen length are indicated by long rays and extremes
associated with lower values for leaf width and stamen length are repre-
sented by no rays.

Rec. Auck. Inst. Mus. Vol. 4, No. 4, pp. 205-212, 12th February, 1954

206

an - oo -ma)
& "ygsueyT eqn vépe@

Oe SE se) “mo

Sf “4yaSueT epoureiuy)

6°S - T°S “Wo

@ ‘yyfuetT jJrey

(eMmeyt

po}yeoIpul o1e

SIajoeIeYD Idy}O 9aIY} {YypSsus, Uses ‘STIX [BOTJIOA SY}PIM Jeo] “STXB [B}JUOZTIO FT

‘apIs UIO}JSed 94} WoOAT UOIJDITTOD ssey “9
‘PURISI SY} IO dpPIS UTOYINOS IY} Wioay UOTDIaT]OO sseyy “q

‘unasnyy puepypony oy} url suouioeds umtieqioy AreuIpigQ “e

‘purleazy MON ‘puUeIS]T OJO}SUeY UO (e}eY) Vsugo4 “Jy pue

WyNyOg) VsjaIxvI SoAIpISOAJIPY VooMyod Waems pligdy e jo sapduies saIy} JO SUIVISeIP 19}}BOS PoIZITeL4IOPI YT

Oh: STE eee A

Pohutukawa x Rata. 207

At one extreme, the upper right-hand corner of each diagram, are
the specimens with leaves which are markedly long and broad, and with
stamens, internodes and calyx tubes which are markedly long. On the
qiagram of herbarium specimens (diagram A) the two dots enclosed
by a line in the upper right-hand corner are specimens belonging to the
species formerly known as Metrosideros tomentosa Rich. but now known
as lM, excelsa Sol. ex Gaertn. and commonly called pohutukawa or
Christmas tree. One of these specimens was collected by Kirk on the
Waitemata in December, 1874, and the other was gathered by the
writer in a remnant of coastal scrub on his property at Blockhouse Bay
this year. Both are similar to specimens collected on the southern slope
of Kangitoto Island and shown in the extreme upper right-hand corner
of diagram B.

At the other extreme, the lower left-hand corner of each diagram,
are the individuals with leaves which are markedly short and narrow,
and with stamens, internodes and caylx tubes which are markedly short.
On the diagram of herbarium specimens (diagram A) the dot enclosed
by a line in the lower left-hand corner is a specimen belonging to the
species known as M. robusta A. Cunn. and commonly called rata. This
specimen was collected by Kirk at Mahurangi about 1870 and is similar
to specimens collected on the eastern slope of Rangitoto Island and
shown in the extreme lower left-hand corner of diagram C,

Metrosideros excelsa and M. robusta are distinguished readily hy
the leaves of the former being tomentose beneath and the leaves of the
latter being glabrous, by the stamens of the former being crimson and
the stamens of the latter being dark scarlet, and so on. The characters
used in the diagrams, however, were those which could be measured
most easily in the field.

Individuals which connect M, excelsa and M. robusta were described
by Kirk (1899) as M, robusta var. intermedia from specimens collected
at Rangitoto Island. Carse (1927) described plants intermediate be-
tween M7. excelsa and M. robusta from Lake Taupo, Bank of Whau and
Titirangi as x M. sub-tomentosa nov. hyb., and Oliver (1928) adopted
this name instead of intermedia “under the authority of a rule which
states that the first name used in a specific sense must stand.” In terms
ef Article H. 1 of Appendix II of the 1952 Code of Nomenclature the
intermediates will be known in future as Mctrosideros x sub-tomentosa
Carse (= Metrosideros excelsa x M. robusta), if a name is required
for them.

The significance of the three scatter diagrams is clear.

1. They show a high degree of correlation among the characters
employed, and support the existing taxonomic arrangement.

2, They show the superiority of mass collections over ordinary
herbarium collections. On this subject I cannot do better than quote
Anderson (1941). “The information derived from a study of mass
collections is useful in two ways. It will, in the first place, aid the
systematist in cataloguing the various entities involved, species, varieties,
forms, etc. While it may raise more new questions than it may solve
old ones, it will aid in the production of monographs whose categories
are more accurately adjusted to the variation patterns of their particular
genera. Mass collections have for some time been customary in avian

208 COOPER.

taxonomy (see, for instance, Mayr’), and Kinsey, in a series of brilliant
monographs’, has shown their superiority in insect systematics. If
taxonomy were to be nothing more than cataloguing, and if taxonomists
were to confine themselves to the problems raised by their herbaria,
mass collections would still be a useful adjunct to herbarium technique
and in many critical groups would provide more efficient working
material, even when their special difficulties of collecting and filing are
considered.

There is no reason, however, why taxonomy should be content to
cultivate such a narrow field. If collectors and herbarium adminis-
trators could be persuaded to encourage mass collections, critically made
and carefully assembled, a second kind of problem could be investigated
in herbarium material. The description and analysis of geographical
trends in variation, the delimitation and interpretation of centres of
variation, the establishment and analysis of variation patterns in differ-
ent genera and families, are only a few of the problems that might well
be investigated. It is already possible to correlate information from the
field of taxonomy with that from cytogenetics. The time is not far
distant when the biochemist of the germplasm will also turn to the
taxonomist for morphological evidence derived from studying the pro-
ducts of the germplasms. To speak with authority on such questions
taxonomists will need to refine their biological as well as their biblio-
graphical techniques.”’

In this very minor example of the technique, diagrams B and C
show that the prevalence of individuals closely resembling Kirk's speci-
men of M. excelsa from the Waitemata and my collection from Block-
house Bay is greater in the sample from the southern side of Rangitoto
Island (diagram B) than it is in the sample from the eastern side
(diagram C). In the sample from the eastern side of the island none
of the specimens is identical with Kirk’s specimen of M. excelsa from
the Waitemata or my collection from Blockhouse Bay, but there are
several individuals which closely resemble Kirk’s specimen of M. robusta
from Mahurangi. Further fieldwork, using ecological techniques, would
be required to verify these distribution patterns.

General: Cockayne and Allan (1927) recognized ten classes and
grades of individual polymorphy and seven classes and grades of poly-
morphy in groups of related individuals. The two authors considered
hybridism to be the most prolific source of diversity, however, and
emphasized that in the majority of cases there occur amongst hybridis-
ing species not a few intermediate individuals but a multitude of forms
producing a motley “swarm.” In 1934 they published an annotated list
of 491 groups of wild hybrids in the New Zealand flora and expressed
the opinion that of these 396 were established beyond reasonable doubt.
They recognized the hybrids as such by field studies, supplemented by
observations of cultivated specimens, and in several instances Allan bred
plants similar to suspected wild hybrids by controlled crossings of the
putative parents.

’Mayr, Ernst. Speciation phenomena in birds. Amer. Nat. 74: 249-278. 1940.

‘Kinsey, Alfred C. The gall wasp genus Cynips. A study in the origin of species.
Indiana Univ. Studies, 84-86: 1-577, 1930; The origin of higher categories
in Cynips. Indiana Uniy, Publ. Sci. Ser, 4: 1-334, 1936.

Pohutukawa x Rata. 209

Evidence relating to only some of the 491 groups has been pub-
lished, however, and in herbaria other than that of the Botany Division,
D.S.1.R., Wellington, where most of the collections of Cockayne and
Allan are preserved, there are few representative collections of the
hybrid swarms. In other words, much work remains to be done, on
the foundations provided by Cockayne and Allan, especially as Ander-
son (1949) has emphasized that the almost imperceptible introgression
of one species into another, by repeated back crossing of the hybrids to
one or both parents, may be of far greater biological significance than
hybridisation which leads to “bizarre hybrid swarms, apparent even to
the casual passer-by.” Valuable new techniques for this work, the
study of introgression and the further analysis of variation in the New
Zealand flora, are mass collecting (Anderson, 1941 and 1943), inclusive
herbarium sheets (Anderson, 1951 and 1952), and pictorialized scatter
diagrams and similar methods of polygraphic analysis (Anderson, 1949
and 1952). The value of these methods in the study of cultivated plants
(Anderson, 1952; Stebbins, 1953) and the passible use of the pictorial-
ized scatter diagram in other fields of research (e.g., archaeology) must
also be mentioned,

ACKNOWLEDGMENTS,

I am indebted to Dr. Edgar Anderson, Assistant Director, Missouri
Botamical Garden and Engelmann Professor of Botany, Washington
(University, St. Louis, Missouri, and to Dr. Norton II, Nickerson, of
the University of Massachusetts, for much stimulating advice regarding
introgression and the techniques for its study. Mr. H. Grimson, Assist-
ant Education Officer of the Auckland Museum, gave much appreciated
lielp with the mass collections on Rangitoto Island.

REFERENCES.

ANDERSON, E., 1941. The technique and use of mass collections. nn. Mo. _

Bot, Gard. 28: 2874292.
ANDERSON, E., 1943. Mass collections. Chron. Bot, 7: 378-380.
ANDERSON, E., 1949. Introgressive hybridization. Wiley, New York, 109p.
ANDERSON, E., 1951. Inclusive herbaria. Jind. J. Genet. 11: 1-3.
ANDERSON, E., 1952. Plants, man and life. Little, Brown & Co., Boston, 245 Dp.

ANDERSON, E., and AMY GAGE, 1952. Introgressive hybridization in Phlox
bifida. Amer. J. Bot. 39: 399-404,

CARSE, H., 1927, Botanical notes, with descriptions of new species. Trans. N.Z.
Inst. 57: 89-93,

COCKAYNE, L., and H. H, ALLAN, 1927. The bearing of ecological studies in
New Zealand on botanical taxonomic conceptions and procedure.

J. Ecol, 15: 234-277.

COCKAYNE, L., and H. H. ALLAN, 1934. An annotated list of groups of wild
hybrids in the New Zealand flora. Ann. Bot. 48: 1-55,

KIRK, T. [1899]. The students’ flora of New Zealand and the outlying islands.
Gov. Print., Wellington, 408 p.

OLIVER, W. R, B., 1928. The New Zealand species of Metrosideros with a note
on Metrosideros collina (Forst.) Gray. Trans. N.Z Inst. 59: 419-423.

STEBBINS, L., 1953. The evolution of cultivated plants and weeds, Evolution
6: 445-448,

210 COOPER.

TABLE

Specimens in the Cheeseman Herbarium, Auckland Museum.

Leaf Leaf Internode Stamen Calyx

No. length. width. length. length. length.
Kirk, 1874 7.7 2.8 no 2.6 1.1
Kirk, c. 1870 3.5 1.3 0.8 1.0 0.7
Kirk, 5593 4.2 1.6 0.6 15 0.7
Cheeseman, 5574 3.4 1.6 1.8 1.3 0.7
Cheeseman, 5575 3.2 1.5 2.4 1.2 0.8
Cheeseman, 5576 3.4 1.4 0.7 1.2 0.7
Cranwell, a 5.2 1.8 1.5 1.3 ).7
Cranwell, b 3.6 1.7 1.2 1.6 ().7
Cranwell, c 5.2 2.4 2.4 1.8 0.6
Cranwell, d 4.5 1.7 1.3 1.7 ().7
Cooper, 36092 8.3 3.2 1.8 2.8 1.1

TABLE 2.

Mass collection from the southern slope of Rangitoto Island between
reference points 377666 and 375674 on the N.Z. Lands and Survey
Motutapu map of 1943 (1: 25,000 series).

Leaf Leaf Internode Stamen Calyx
length. width. length. length. length.

] 6.7 2.2 1.1 2.7 1.0

2 6.5 2.7 bed 2.4 1.1

| be 2.0 1.6 3.0 0.9

4 6.1 2.0 1.0 2.6 1.0

5 6.7 2.1 1.3 2.7 0.9

6 6.4 2.3 Re 2.4 0.9

7 6.8 2.3 1.4 2.6 1.1

8 6.0 1.9 1.1 24 1.0

Y Ag 1.4 0.4 1.8 0.7
10 So es 1.2 2.7 1.0
ll 6.2 2.1 1.2 2.8 1.0
12 ape 1.7 1.1 2.9 10)
13 8.1 2.3 1.2 3.2 1.3
14 5.9 2.0 1.1 2.9 1.1
15 7.5 on 1.4 27 1.0
16 9] 2.8 Ss 3.1 1.1
17 7.7 2.7 1.2 2.7 1.0
18 7.4 2.8 1.3 26 1.0
19 7.0 2.6 1.6 2.8 1.2
20 6.3 2.6 1.2 29 10
21 7.4 2.6 1.0 2.3 1.0
22 7.0 1.9 0.9 29 O.8
23 6.2 2.0 L.1 29 1.1
24 7.7 3.0 1.3 2.5 1.0
25 6.9 2.5 1.1 2.7 1.0
26 7.5 3.0 1.1 2.7 1.1
27 6.9 2.4 0.9 2.6 1.0
28 6.1 2.1 0.9 2.2 0.9
29 5.8 2.1 1.1 2.3 1.0
30 5.6 aa 1.3 yA, 0.9
31 7.0 2.7 1.5 2.7 1.0
32 4.4 2.0 0.6 1.9 0.9
33 7.3 2.4 1.2 a0 1.0
34 4.9 1.7 1.0 Le 1.0
35 rx 2.8 15 2.6 1.1

Pohutukawa «x Rata. 211
TABLA. S:

Mass collection from the eastern slope of Rangitoto Island between
reference points 382688 and 394692 on the N.Z. Lands and Survey
Motutapu map of 1943 (1: 25,000 series).

Leat Leaf Internode Stamen Calyx
length. width. length. length. length.
41 6.7 an 1.2 2.8 1.0
42 vf 1.5 0.6 0.9 0.5
43 6.8 1.8 0.9 2.7 1.1
44 4.5 1.7 0.8 1.3 0.6
45 5.9 2.6 0.8 1.9 0.8
46 5.7 2.0 0.8 2.9 0.9
47 7.6 2.5 ().9 2.6 0.9
48 7.3 1.9 1.3 2.4 0.9
49 1.6 0).7 1.5 0.8
50 6.4 2.7 1.0 2.3 0.9
51 6.1 2.1 0.9 2.4 0.7
52 7.2 2.2 1.1 2.6 1.0
53 6.6 2.3 0.8 Fe 1.0
54 79 2.8 1.9 3.0 0.9
55 7.0) 2.4 1.2 2.9 0.8
56 3.4 1.6 0.6 1.0 0.6
57 4.9 2.5 1.6 2.6 0.9
58 6.3 Ae 1.3 2.4 0.9
59 8.2 2.6 1.1 2.9 0.9
60 o4 Zo 1.1 2.8 1.0
61 5.5 2.7 1.2 2.6 0.9
62 4.7 19 1.4 1.7 0.7
63 4.8 1.6 0.7 1.9 0.5
64 53 22 1.1 2.5 1.1
65 8.6 3.0 1.3 2.9 1.0
66 6.9 2.0 1.1 3.0 0.9
67 7.5 2.4 0.9 2.9 1.0
68 74) 2.1 i] 3.0 0.9
69 7.2 ou is 2.6 1.0
70 7.6 gs 1.0 2.7 0.8
71 6.2 22 0.9 3.0 1.0
72 4.0 1.8 1.0 2.7 ().7
73 6.3 1.9 0.8 2.6 1.0
74 6.3 25 tA 2.4 0.8
75 5.3 1.9 1.3 2.1 0.8

Notes—
1. All measurements are in cm.
2, Nos. 36-40 were not used in the numbering of mass collections.

3. “Leaf length” and “leaf width” were measured on leaves at the second node
beneath the apex of a mature woody branchlet bearing the flowering shoot.

4. “Internode length” refers to the third internode from the apex of the mature
woody branchlet bearing the flowering shoot.

5. “Calyx tube length” refers to the calyx of a fully-open flower in the centre
of a cyme.

213

A Note on the Occurrence of
Chelisoches morio (Fabricius) on Pitcairn
Island, South East Pacific Ocean

(Dermaptera : Labiduridae.)
By E. T. GILES, Auckland.

Abstract.

__ An adult 9 Chelisoches morio (Fabricius) is recorded from Pitcairn Island.
9.E. Pacific Ocean. This is a new locality for the species and the first earwig
record from the island.

Through the courtesy of Dr. G. Archey, Director, and Mr. E. G.
Turbott, Ornithologist and Entomologist, Auckland Institute and
Museum, this earwig came to the writer’s attention. Data supplied with
it read: “Found in bath, alive, within abdomen of second. Pitcairn
Island. Coll: A. W. Moverley, 26. viii. 1952.”

The specimen is an adult 9 Chelisoches morio (Fabricius) with
its head firmly held inside the posterior six segments of another earwig,
almost certainly of the same species. A summary of the carnivorous
habits of the species is given by Hincks (1948). It could easily develop
cannibalistic traits and doubtless the one was feeding on the remains
of the other.

Both are of a dark reddish-brown colour, but usually this earwig
is black. Specimens of a similar light colour are rare in the Burr
Collection in the British Museum (Natural History), which was
examined through the courtesy of Mr. D. R. Ragge. It would be inter-
esting to find the proportion of light-coloured individuals in the popula-
tion of the species on Pitcairn Island. The effect of “drift” (Wright,
1940) could offer an explanation if it were found to be high.

This is the first record of a dermapteran from Pitcairn Island;
Hincks (1938) gives none. C. morio is of world-wide distribution and
it is not surprising that it should turn up on the island, although it is
isolated and of volcanic origin. Specimens of such a hardy, active insect
could be easily transported there on rafts of vegetation carried by ocean
currents or among the plants and possessions of early native voyagers.
Instances of this species being introduced are given by Burr (1910)—
to the East Coast of Africa; by Lucas (1920)—to Kew Gardens,
London; and by Hebard (1933)—to the coast of California. In Auck-
land on 7.111.1950 the writer took a nymph among bananas shipped from
Western Samoa.

Rec, Auck. Inst. Mus. Vol. 4, No. 4, pp. 213-214, 12th February, 1954

214 GILES.

REFERENCES.

BURR, M., 1910. Dermaptera: in The Fauna of British India. (London. )

HEBARD, M., 1933. The Dermaptera and Orthoptera of the Marquesas Islands.
Bull. Bishop Mus., Honolulu, 114: 105-140.

HINCKS, W. D., 1938. The Dermaptera of Oceania. J.F.M.S. Mus., 18 (2):
299-318.

1948. Preliminary notes on Mauritian earwigs (Dermaptera). An.
Mag. nat. Hist. (11), 14: 517-540,

LUCAS, W. J., 1920. A Monograph of the British Orthoptera. (The Ray
Society, London. )

WRIGHT, S., 1940. The Statistical Consequences of Mendelian Heredity in

Relation to Speciation, in The New Systematics, ed. J. S. Huxley,
pp. 161-183. (Oxford.) ,

215

New Records and Descriptions of
Hemiptera-Heteroptera
from the Three Kings Islands.

By T. E. WOODWARD,

Department of Entomology, University of Queensland.

Abstract.

Twenty-five species of Heteroptera are now known to occur on the Three
Kings. Species found also both on the mainland of New Zealand and overseas are
—Pentatomidae: Cuspicona simplex Walker; Cydnidae: Philapodemus australis
(Erichson) ; Lygaeidae: Pachybrachius miyriceps (Dallas) ; Reduviidae: Empicoris
rubromaculatus (Blackburn); Nabidae: Nabis capsiformis Germar; Miridae:
Megaloceroea reuteriana Buch. White, Euryvsiylus australis Poppius, Coridromins
variegatus (Montrouzier) ; Veliudae: Microvelta halei Esaki. Species known only
from New Zealand (Maorian sub-region)—Lygaeidae: Nysins huttont Buch.
White, Rhypodes clavicornis (F.) Se some local variation), Jargarema
staals Buch. White, Taphropeltus putont (Buch, White), Camus novaeselandiae sp.
nov.: Nabidae: Nabis maoricus Walker ; Miridae: Calocoris laticinctus (Walker) ;
Sthenarus myersi Woodward. Forms not recorded outside the Three Kings—
Pentatomidae: Cermatulus nasalis turbotti Woodward; Lygaeidae: Tomocoris
insularis Woodward. In addition, one species of Anthocorid and one Aradid have
been sent to specialists for determination, and there have been noted one apparently
endemic species of Sthenarus and three species of Lygus, of which one is not known
from the mainland.

Cymus novaesclandiae sp. nov. and the nymphis of Cermatulus nasalis turbotti
are described. Information is given on local variation in rostral length of Rhypodes
clavicornis,

ACKNOWLEDGMENTS.

I am much indebted to the authorities of the Auckland Institute and
Museum, and particularly Dr. G. Archey and Mr. E. G, Turbott, for
the opportunity of visiting the Three Kings on one of the expeditions,
and of examining much material, to Mr. Turbott for specimens of
Cerimatulus, to Mr. J. S. Edwards for the collection of a large series of
Hemipterous material and for locality and habitat information, and to
Dr. W. E. China for kindly checking some of the species against British
Museum material and for the determinations acknowledged in the text.
The author’s expenses in participating in the 1951 expedition were
defrayed from a Hutton Research Grant of the Royal Society of New
Zealand,

INTRODUCTION.

The types of Cermatulits nasalis turbotti were collected by Mr. E.
G. Turbott during the autumn, 1946, expedition of the Wild Life Branch
of the Department of Internal Affairs and described by the author in

Rec, Auck. Inst. Mus. Vol. 4, No. 4, pp. 215-233, 12th February, 1954

216 WoopWARD.

1950. The rest of the material was collected by Mr. Turbott and the
author on the Auckland Museum expedition of summer, 1951, and by
Mr. J. S. Edwards on the Museum’s expedition of 1952-53.

While a survey so restricted in time and season cannot, of course,
be regarded as all-inclusive, nevertheless collecting has been fairly intens-
ive and carried out at probably the optimum season for insect activity.
Many mainland species which were especially sought for in appropriate
localities were not found in these islands. The time available for col-
lecting on the smaller islands of the group has been extremely limited,
so that our present conception of the Three Kings Heteropteran fauna
is based almost entirely on that of Great Island, which has suffered the
most floristic alteration in recent times (Baylis, 1948; Turbott, 1948).

The general conclusion must be that the Heteropteran fauna of the
Three Kings is a restricted sample of the mainland fauna, together with
a few interesting endemic species and subspecies which have apparently
developed there in isolation. The total number of discovered species of
Heteroptera in the group is twenty-five. Three species, Tomocoris
insularis Woodward, Lygus sp., and Sthenarus sp., and one subspecies,
Cermatulus nasalis turbotti Woodward, have not been recorded from
the mainland or seen by the author in mainland collections, and are
apparently not represented elsewhere. The first is undoubtedly a relict
species of a genus once more widely distributed in New Zealand (Wood-
ward, 1953, p. 213). In addition, the Great Island population of
Rhypodes clavicornis (Fab.) shows a constant structural difference from
all mainland specimens examined in the strikingly longer rostrum (p.
222), and thus might be regarded as constituting an insular subspecies
or incipient subspecies. The S.W. Island population of this species 1s
intermediate as regards this character between the Great Island and the
mainland forms.

It is hoped later to present a more detailed comparison of the
Hemipteran faunas of the mainland and the Three Kings when the
Homoptera of the latter have been worked out.

The relative sparsity of the present bug fauna seems attributable
to four main causes:

(1) Non-representation of mainland species in the Three Kings
area when this was connected to the mainland (due to restrictions of
seography, small area, and less variety of plant cover).

(2) The emergence of species and subspecies on the mainland since
the separation of the Three Kings. This probably applies to many of
the mainland forms with restricted distribution and of apparently recent
origin, such as the brachypterous Rhyparochrominae inhabiting leatf-
mould. The converse of this is the development of species and sub-
species on the Three Kings.

(3) The introduction of species into the mainland but not to the
Three Kings, due to the relative inaccessibility and the much smaller
area of the latter. Such recently introduced Pentatomids as Negara
viridula (Linn.) and Antestia orbona Kirkaldy are examples. The
former, although apparently it has not yet invaded the Three Kings, has
recently spread to the far northern coast of the North Island of New

Hemiptera-Heteroptera. 217

Zealand. On the other hand, Cuspicona simplex Walker, which has
recently been introduced into the mainland, has also found its way to at
least one island of the Three Kings group.

(4) The elimination of host plants and habitat niches, firstly by
Maori occupation and secondly by goats since their introduction last
century until their destruction in 1946 (Baylis, 1948; Turbott, 1948).
The effects would presumably have been greatest on tree and forest
dwelling insects. On the other hand, survival of species frequenting
kanuka (Leptospermum ericoides), at least some grasses and sedges,
and the more inaccessible cliff and rock plants, has been relatively fav-
oured. This is seen by the collection data below and accords with the
effects of Maori occupancy and of goats as detailed by Baylis and Tur-
bott. It is of interest that both on the Three Kings and on the mainland
kanuka has been found by the writer to suport a much richer bug fauna
than manuka (L. scoparium), and, but for the survival of the former
as a dominant element of the flora, probably the Hemiptera would have
been even more depleted. On the Three Kings, only three Heteropteran
specimens were taken from manuka, a fifth instar nymph of the pre-
dacious Pentatomid Cermatulus nasalis turbotti, and one male and one
nymph of the Lygaeid Pachybrachius nigriceps (Dallas).

It is, of course, at present difficult to distinguish the effects of (1)
and (4), but additional information would no doubt be obtained by
intensive collecting from South-West Island and the other smaller islands
of the group, which probably support a virtually primitive plant com-
munity, though more restricted than that formerly covering Great Island
(Baylis, 1948).

The positions of the quadrats and other localities noted in the fol-
lowing records are shown by Turbott (1948; pl. 41), Baylis (1948; p.
242), and Battey (1951).

The bulk of the material recorded is to be deposited in the Auckland
Museum.

FAMILY PENTATOMIDAE.
SUB-FAMILY PENTATOMINAE.

Cuspicona simplex Walker.

1867—Cuspicona simpler F. Walker, Cat. Spec. Het. Hem. Coll. Brit. Mus..
pt. 2: 388.

One 2, S.W. Island, 13/1/51, sweeping a mixed growth of Sola-
num nigrum L. and Solanum aviculare Forst. (poroporo). After the
capture of this specimen, stands of Solanum on both S.W. and Great
Island were vigorously swept, but no further specimens were found.
This Australian species was recorded from the mainland of New Zea-
land by Spiller and Turbott (1944; pp. 79-80), where it is apparently a
recent introduction. Its incursion into the Three Kings is also appar-
ently recent, and perhaps it has not yet had time to become fully estab-
lished or to spread to Great Island. Spiller and Turbott record it from
North Auckland (Hokianga), Auckland (widely distributed, including
Waiheke Island), and Taranaki (New Plymouth). Besides Auckland
itself, the writer has taken it in various parts of North Auckland (Paihia

218 WOODWARD.

(13/2/49) and Russell (14/2/49), Bay of Islands, and Mangamuka
Gorge (17/2/51) ), Little Barrier Island (11/12/50), and Wellington
(Otaki R. (S. of Levin) and Point Howard (31/1/51) ). All captures

were on solanaceous plants.

SUB-FAMILY ASOPINAE (AMYOTINAE).

Cermatulus nasalis (Westwood) subsp. turbotti Woodward.

1950—Cermatulus turbotti Woodward, Rec. Auck. Inst. Mus. 4 (1): pp. 24-30;
figs. 1-2.

1953—Cermatiulus nasalis turbotti Woodward, Trans. R. Soc. N.Z. 80 (3 and
4): pp. 299, 308-310, 312, 318; figs 9, 34.

Type ¢ @ collected by E. G. Turbott, Great I.: 1 @, Tasman
Valley, 6/5/46; 2 @@, near depot, 5 and 10/5/46; all on kanuka
(Leptospermum ericoides A. Rich.).

On the 1951 expedition, the following specimens were collected,
all from Great Island: 2 adults from near depot (1 ¢, 12/1/51, E. G.
Turbott: 1 ¢, 14/1/51, T.E.W.) ; 4 last (fifth) instar nymphs (2 from
the eastern slopes of Tasman Valley, 12/1/51; 2 from the East Point,
14 and 15/1/51, T.E.W.); 3 young nymphs from eastern slopes of
Tasman Valley, 12/1/51 (1 second instar, 2 third instar, T.E.W.). All
except one of the specimens were taken on kanuka (L. ericoides) ; one
of the fifth instar nymphs from East Point was on L. scoparium Forst.
(manuka).

The two last instar nymphs caught on 12/1/51 emerged in captivity,
cne ( 6) on the day of capture, while the other ( 2 ) was found to have
emerged by the rettirn to Auckland (18/1/51). The former was seen,
at emergence, to be bright red, but by the next day had darkened to the
normal adult coloration, the hemelytra being the last to change colour.

Adult Male

The males of this subspecies show the same tendency as the femaies
to a greater body size than in nasalis, Length (3 6 6), 12.0-12.5 mm.
Width across pronotal shoulders, 6.25-6.5mm. Head as in 92, except
that apical margin of tylus is more broadly convex and dise of juga has
dark ochreous mottlings. As in 2, eye only 4 as wide as interocular
space. Head above, pronotum and scutellum in all three specimens with
metallic bronzy and greenish reflections. Antennal segment III 4-7 as
long as II; IV 1/ 5¢h_3 TIT, and V subequal to or rather shorter than LV.
Pronotum across shoulders 2.20 times median length. Membrane of
hemelytra infuscated, particularly near anal angle of corium, appearing
hronzy-brown in closed position; veins darker brown. Hind wings
with green reflections. Black sternal spots of abdomen smaller than in
o. Other features as described for 2 (Woodward, 1950, of. cit.).

Nymphs
Fifth (last) Instar:

Ovoid, about 4 as long again as wide across abdomen (9.5 mm. :
G6mm.). Width of head across eyes, 2.26imm. Dorsal surface of head,
pronotum, scutellum, and wing pads shining bronzy-black with metallic

Hemiptera-Heteroptera. 219

green reflections, except for the eyes, which are brownish black, and a
broad, orange-brown marginal band on each side of pronotum, narrow-
ing at posterior shoulders, and along anterior half of each wing pad.

Dorsal surface of abdomen with five median bands of similar metallic

colour to above, the first transversely linear, very short antero-pos-
teriorly, the second wider and several times longer, and flanked on each
side by a small patch of similar colour, the third the widest and more
than twice as long as second, the fourth the longest, the fifth the narrow-
est and nearly rectangular. Dorsum with five large lateral patches of
similar metallic colour on either side, each invaded at margin by an
orange-brown semicircular area. Connexivum below with simular,
though smaller, dark metallic patches, the invading semicircles propor-
tionately larger. A sixth, smaller, anterior patch on each side of
dorsum may be reduced to a transverse bar. Last three (reduced)

terga, including that of anal tube, completely dark metallic, except for
small, lateral, orange-brown patch on each side of antepenultimate
(eighth). All metallic areas rugose and, especially on thorax and wing
pads, strongly punctate ; spaces between them on abdomen orange-brown
mottled with purplish brown, without rugae and not or scarcely punctate ;
an unmottled yellowish brown area inside mesial margin of each wing
pad. Antennae orange-brown; first sep ment with external dorso-lateral
aspect black; second black on apical 4; third and fourth black except at
base. Rostrum reaching hind coxae; infuscated ochreous, last segment
blackish brown. Ventral surface of body yellowish to pinkish ochreous,
with dark spots and bands. Ventro-lateral regions of head and thorax
shining dark metallic. Legs yellowish ochreous with reddish and black
mottlings; femora at apex, tibae at base and apex, and whole of tarsi
shining black. Ventral black spine of fore tibiae pronounced. Wing
pads reaching half-way along third abdominal tergum (level with pos-
terior margin of second median metallic bar). Scutellum well defined
as a bluntly rounded triangle between them.

A general darkening of the pale areas may occur after death in
dried specimens.

Sides of tylus parallel throughout; shape of tylus and juga as
described for adult ¢. Eye only 0.29 times as wide as interocular
space (7,5: 26). First antennal segment very short, not reaching to
apex of head; proportions of segments IT-IV, 45: 23: 23. Pronotum
subtrapeziform ; anterior margin widely and rather shallowly concave ;
posterior margin nearly straight, with the backward triangular projec-
tions smaller and more rounded than in adult, base outside them slightly
sinuate ; sides gently sinuate, slightly concave just behind middle; median
length nearly equal to length of head, an about 4 posterior width and
3 anterior width of pronotum (32: 354: 90: 42), Pronotum, and
scutellum except at apex, with a well hind impressed median line
(ecdysial cleavage line). Mesosternum on each side of low carina trans-
versely striate, but ochreous, not metallic as in adult. Anterior abdom-
inal spine rounded: very short, scarcely reaching anterior margin of

hind trochanters.

Ecdysis: At the last ecdysis the nota of the @? split in the median
longitudinal line from the anterior margin of the pronotum through
the “lengt h of the scutellum and metanotum, which is very short in ie
mid-line. Cleavage thus occurs along and for a short distance behind

220 WooDWARD.

the median line visible in the fifth instar nymph. In this respect the
last ecdysis resembles that described by Myers (1926; p. 497) for
Glaucias amyoti (A. White) rather than in Cermatulus nasalis nasalis
as described by the same author (p. 493). However, the process 1s
subject to a certain amount of individual variation. The last ecdysis
in the latter subspecies was observed by Myers in only one specimen.
In the only other instance so far observed in turbotti the two halves
of the thoracic plates have closed back, one over the other, so firmly
that it is impossible to delimit the ecdysial line in this region. In both
specimens of furbotti a transverse split occurred along the whole dorsal
width of the base of the head, in the membranous posterior border ot
the occipital region which in life is covered by the anterior part of the
pronotum. From this, on each side, a split extended forward along
the inner dorsal margin of the eye to its anterior end. Thus the entire
sclerotised dorsal plate of the head between and before the eyes
remained intact and was levered up from behind and pushed forward
while the adult head and its appendages were being withdrawn; the
rest of the body was drawn forward through the median gap in the
thoracic plates.

The fifth instar nymph of wnasalis differs from that of turbotti
mainly in the evenly convex margins of the pronotum, which lack any
concavity or sinuation, and in the absence of the marginal orange-brown
areas in the dark connexival patches. Other colour differences may well
be subject to individual variation in both species. For a description and
Wustration of the fifth instar nymph of C. nasalis nasalis, see Myers,
1926, pp. 491-492, fig. 12, where these differences from turbotti are
well shown.

Third Instar:

Ovoid. Length about 4mm. and width about 2.75 mm., but the
comparative softness of the integument makes for variation in size in
life and for considerable shrinkage and distortion after death, Width
of head across eyes, 1.27-1.32mm. (Head capsule width is a better
criterion for differentiating instars.) Distribution of light and dark
areas as for fifth instar, except in the following respects: dorsal surfac
of head and thorax shining black with greenish reflections, but without
the pronounced metallic bronzy sheen; sides of pronotum and, more
obscurely, of mesonotum, yellow-brown; lateral black areas of abdomen
with the marginal orange-brown invasions less clearly defined; a pair
of conspicuous yellow triangular areas, one on each side of dorsum of
abdomen at its anterior end; a pair of smaller yellow spots towards
posterior end of dorsum. To the naked eye, apart from these patches,
the dorsum appears completely black, the median black patches forming
a more or less continuous block, especially after death. Venter of
abdomen reddish brown. Legs, rostrum (last segment paler), and
ventral surface of head and thorax shining blackish brown; tars! black.
First and second segments of antennae blackish brown, third and
fourth black, with base of third orange-brown. Dorsal black areas only
very obscurely rugulose and shallowly and sparsely punctate.

Wing pads not evident. Scutellum indicated only by the obtuse
hackward angulation of the mesonotum, not covering the metanotum.
Fore tibiae with ventral spine small and easily overlooked. Rostrum

Hemiptera-Heteroptera. 221

well surpassing hind coxae. Abdominal spine, ventral thoracic carina
and striate areas not developed. Thoracic cleavage line as in last instar.
Pronotum proportionately much shorter than in last instar; anterior
margin concave, posterior convex, the two nearly parallel; sides nearly
straight; median length just over 4 length of head and about 2/7%
posterior width and rather less than 4 anterior width of pronotum (12:
20: 41: 26). Eye 0.3 times as wide as interocular space (4.5: 15).
Proportions of antennal segments II-[V, 23: 13: 16.

Second Instar:

Ovoid. Length about 3mm. Width about 2.25 mm. Width of
head across eyes, 1.05mm. Colour pattern as for third instar, except
that yellow-brown on margins of pronotum is less pronounced, and is
absent from mesonotum; legs, rostrum, ventral surface of head and
thorax, antennae except for extreme base of third segment, all black ;
first median black bar of abdominal dorsum scarcely apparent, second
bar without the lateral spots; lateral black patches of dorsum without
any signs of marginal orange-brown invasions.

Rostrum passing hind coxae, Proportions of antennal segments
I-IV, 14: 9:12. Pronotum shaped as in third instar; median length
4, length of head and about 4 posterior width and just over 4 anterior
width of pronotum (7.5: 15: 29: 21). Eye 3 as wide as interocular

space (3.75: 11.5).

(Proportionate measurements to the scale 18 units = 1 mm.)

The paucity of Pentatomids as represented in these collections is
outstanding, and particularly the absence of the common and widely
distributed mainland species Dictyotus caenosus (Westwood). Its habits
as a frequenter of the ground and of low-growing plants may tend to
restrict its dispersal, though it also occurs in Australia; it is, however,
a form that could easily have been transported by the early ships. It
is quite possible, of course, that there occur on the Three Kings some
species of this family, particularly the less common ones, which have been
nussed in collecting. However, there is undoubtedly a strong element
of chance as to which species happen to reach such an area, and the
results of these collections tend to emphasise this.

FAMILY CYDNIDAE.
Philapodemus australis (Erichson).
1842—Cydnus australis Erichson, Arch. fiir Naturg. 8: 275, 276.

One nymph, East Point, Great I., 15/1/51, under Disphyma aus-
trale (A. Cunn.) ; at cliff-top (T.E.W.).

Occurs in the Oriental and Australian regions, including New
Zealand.

FAMILY LYGAEIDAE.
SUB-FAMILY LYGAEINAE.

Nysius huttoni Buch. White.
1878—Nysius huttoni Buch, White, Ent. mon. Mag. 153» 32,

Collected by author; 1 @, Tasman Valley, Great I., 12/1/51, sweep-
ing sedges, grasses and rushes; 1 ¢, Great I., foot of cliff below depot,

222 WoopDWaARD,

14/1/51, on prostrate ngaio (Myoporum laetum Forst. ye tole
2 2 9, Great I., below cliffs near depot, 14/1/51, Chenopodium triand-
rum Forst.; 3 646, 2 2 ¢, East Point, Great I., 15/1/51, under
Disphyma australe (A. Cunn.) Black.

Collected by J. S. Edwards: 1 4, Quadrat II, Great I., 1/1/53.
on puka (Meryta sinclairti (Hook.) Sum.; 1 4, cliff vegetation, Great
I., 4/1/53, on flowering kanuka (Leptospermum ericoldes); 3. 8 é,
cliff edge above Castaway Valley, Great I., 4/1/53, dry grass.

Restricted to New Zealand.

Rhypodes clavicornis (Fab.).
1794—Lygaeus clavicormwis Fabricius, Ent. Syst. 4: 169,

Collected by author: 2 4 6, 6 2 9, 4 nymphs, Tasman Valley,
Great I., 11-12/1/51, sweeping sedges, grasses and rushes; 1 ?, Great
I., depot, 12/1/51, kanuka (Leptospermum ericoides); 4 4 2, 39 9,
4 nymphs, Great I., foot of cliffs below depot, on prostrate ngaio
(Myoporum laetuin); 7 8 &, 4 29, East Point, Great I.. on and
under Poa anceps Forst., on cliff slopes; 1 @, 2 nymphs, Great I.,
below cliff near depot, 14/1/51, on Hebe insularis (Cheeseman) Ckne
and Allan; 3 ¢ ¢, 2 nymphs, Great I., below cliffs near depot, 14/1/51,
Chenopodium triandruin; 2 9 9, East Point, Great I., 15/1/51, on
L, ericoides; 3 6 6,5 2 9, East Point, 15/1/51, sweeping sedges and
rushes; 14 ¢ 6,6 9 9, S.W. Island, 13/1/51, sweeping grassy slopes.

Collected by J. S. Edwards: 1 @, Tasman Valley, Great I.
31/12/52, sweeping from Leptospermum; 2 @ 6,1 9, cliffs below
Tasman Valley, Great I[., 31/12/52, swept from Haloragis erecta
(Murr.) Schindler; 1 ¢, 19, 1 nymph, cliffs at end Tasman Valley,
Great I., 31/12/52, sweeping grasses and Scirpus; 1 2, Saddle, Great
I., 2/1/53, tussock (Scirpus and Carex); 1 2, cliff edge above Cast-
away Valley, Great I., 4/1/53, dry grass; 1 @?, cliffs near Castaway
Valley, Great I., 4/1/53, on Tetragonia.,

The Great Island population of FR. clazvicornis differs from all
specimens examined from a wide range of localities in both North and
South Islands and smaller islands near the mainland coast, in the con-
siderably longer rostrum. No other constant differences are apparent
and I have not separated this form taxonomically, although in view of
the wide use of rostral length as a systematic character in the Heterop-
tera, one might perhaps feel justified in regarding this form as a sub-
species. However, it seems sufficient, at least at present, to draw atten-
tion to it as an example of early intraspecific divergence of a population
in isolation. It is a geographic form distinguished by only one visible
structure from the general population, and as such is perhaps best
considered an incipient subspecies; with continued isolation it might
be expected eventually to develop other correlated differences. The
real interest of the form, whatever it be called, is as an example of evolu-
tion in progress. In most cases the difference in rostral length from
that of mainland specimens is so considerable as to be immediately
obvious, while there 1s no overlap in the ranges of relative rostral length
in the material examined of the two populations. The proportionate
rostral length of the S.W. Island material is intermediate between that

Henuptera-H eteroptera. 223

of the Three Kings and mainland specimens, and the range overlaps
that of both. The ratio, length of rostrum: width of pronotum (across
humeral angles) is as follows :—

Mainland. S.W. Island, Great Island.
3 1.18-1.40 1.35-1.48 1.45-1.53
2 1,07-1.32 1.30-1.41 1.40-1.49

In the Great I. specimens the rostrum reaches as far as the middle
of to just beyond the posterior coxae, often reaching or extending on to
the first visible abdominal sternum; segment I usually surpassing base
of head, II surpassing fore coxae and extending on to mesosternum,
ITI reaching middle or hind end of middle coxae. In the mainland
specimens the rostrum usually reaches middle coxae, but does not extend
as far as their middle, never reaching first visible abdominal sternum:
segment I not or barely passing base of head, IT not or barely passing
fore coxae, III not or barely reaching middle coxae, not reaching to
their middle. The positions given apply when, as is usually the case,
the head is retracted to or near the maximum extent within the pro-
thorax, so that the eyes touch the latter.

As Usinger (1942; pp. 42, 45-46) has pointed out, Rhypodes
clavicornis varies considerably in the degree of development of the
sublateral processes of the posterior pronotal margin; in some specimens
they are well developed triangular projections, in others they are absent
or scarcely discernible. Usinger found the absence of these lobes com-
moner in the South Island specimens. They are well developed in all
the Three Kings specimens. An examination of mainland material has
indicated a probable tendency to local geographic variation in respect of
body colour and of size of the pronotal processes, but as yet insufficient
material has been available to give a clear picture of the distribution of
the forms involved and to determine whether any of them warrant
taxonomic separation. This is a problem which should be considered
together with the Three Kings material, and it is hoped later to continue
these studies in greater detail, as they provide examples of the ways in
which speciation might begin.

SUB-FAMILY RHYPAROCHROMINAE.

Targarema staali Buch. White.
1878—Targarema staali Buch. White, Ent. mon. Mag. 15 (1): 34.

Collected by author: 12 ¢ 6, 13 2 2, Tasman Valley, Great I.,
11-12/1/51, kanuka (Leptospermum ericoides); 1 &, 1 9, Tasman
Valley, Great I., sweeping sedges, grasses and rushes; 1 ¢, 3 9? 9,
depot, Great I., 12/1/51, on L. ertcoides; 2 9 2, Bald Hill, Great L.,
12/1/51, sweeping grasses and rushes; 2 ¢ 8, East Point, Great L,,
14/1/51, under Poa anceps Forst. on cliff slopes; East Point, Great L.,
15/1/51, 3 ¢¢, 2 22, on L. ertcoides, 1 2 sweeping sedges and
rushes.

Collected by J. S. Edwards: 6 6 ¢, 4 92, Castaway Valley,
Great I., 30/12/52, kanuka about camp site (L. ericotdes); 7 ¢@ ¢,
3 99, Tasman Valley, Great I., 31/12/52, sweeping from Leptoas-
permum,; 1 6, 1 9, Eastern Arm, Great [., 1/1/53, sweepings from

224 W oopWARD.

eround vegetation (Carex, mosses, etc); 2 ¢ 6, kanuka canopy near

Quadrat II, Great L, 2/1/53 (L. ericoides); 15 66, 2 22, The

Saddle, Great I., 2/1/53, L. ericotdes; 1 8, near summit on N.W., cliffs,

Great I., 3/1/53, kanuka canopy (L. ericoides); 1 ¢, 1 @, cliff vege-

tation, Great I., 4/1/53, on flowering kanuka (L. ericoides); 1 4,

wh Tasman Valley, Great 1., 5/1/53, Colensoa physaloides (A. Cunn.)
ook.

Restricted to New Zealand.

Pachybrachius nigriceps (Dallas).
1852—Rhyparochromus nigriceps Dallas, Cat. Hem. Brit. Mus. 2: 577.
Collected by author: 6 6 6, 9 9 ¢@, 1/7 nymphs, Tasman Valley,

Great I, 11-12/1/51, sweeping grasses, sedges and rushes; 6 2 9, ll
nymphs, Bald Hill, Great L., 12/1/51, sweeping grasses and rushes ;

2 4 4, East Point, Great I., 15/1/51, sweeping sedges and rushes.
Collected by J. S. Edwards: 2 nymphs, cliffs at end of Tasman
Valley, Great I., 31/12/52, sweeping grasses and Scirpus; 1 6, Eastern

Arm, Great I., 1/1/53, sweepings from eround vegetation (Carex,
mosses, etc.) ; 8 nymphs, N.W. cliffs, Great I., 4/1/53, sweeping Carex
and Scirpus; 1 ¢, 1 nymph, Bare Saddle, S.E. Bay, Great I., 5/1/55,
low manuka (Leptosperimum scoparium);1 4, The Saddle, Great I.,
2/1/53, tussock (Scirpus and Carex); 38 6 6,42 22, 2 nymphs,
cliff brow, S.W. Cove, Great I., 5/1 /53, on Scirpus nodosus Rottb.

Widespread in the Pacific area, including New Zealand.

Taphropeltus putoni (Buch. White).
1878—Scolopostethus putoni Buch. White, Ent. mon. Mag. 15: 75.

1 @, Tasman Valley, Great I. 11/1/51, sweeping sedges, grasses,
and rushes (T.E.W.).

Known only from New Zealand.

Tomocoris insularis Woodward.
1953—Tomocoris insularis Woodward, Kec. Cant. Mus. 6 (3): 218.

1 ¢, Castaway Valley, Great Ty, 0571/54,-e¢ teat mould under
sedges beside stream (TW. 3):

Known only from the Three Kings.

SUB-FAMILY CYMINAE.
Genus CYMUS Hahn, 1832.
Cymus novaezelandiae sp. nov., figs.1-2.

This species occurs in two forms, macropterous and sub-brachypter-
ous, of which the latter appears to be by far the more common.

Diagnosis: Colour brown; stramineous or ochreous, sometimes
‘nfuscated or ferruginous. Second antennal segment short, subequal
in length to first and about 2/ 5th as long as third; fourth segment and

Hemiptera-H eteroptera. 225

dilated apex of third black; fourth segment very strongly swollen, in
middle about 4 as wide as long and rather thicker than first segment,
about 2 as long as third segment. Head rather wider across eyes than
long. Under-surface of head and thorax with golden, deciduous, scale-
like hairs. Bucculae half or very nearly half as long as first rostral
segment, which does not quite reach base of head. Pronotum rather
longer than head and 2-3 as long as wide at basal angles, length equal
to anterior width. Pronotum raised between calli but with median
carina obsolete; scutellum convexly raised behind the depressed base,
without true carina. Metathoracic scent-gland spouts very small,
ventrally truncate. Costal margins of coria strongly convex, sinuately
incurved at about 1/5" from base, whence nearly straightly converging
to base; in macropterous, but not in sub-brachypterous form, also sinu-
ately incurved towards apex. In sub-brachypterous form, membrane
short and narrow, 2 as long as and narrower than corium, and extending
for only about $ of its length beyond apex of coria; in macropterous
form, slightly to considerably wider than corium and { or more as long,
extending for + or more of its length beyond apex of coria. Length:
sub-brachypterous 2.4-3.2 mm.; macropterous about 3.5mm. Width:
sub-brachypterous 1.0-1.3 mm.; macropterous (two specimens), ¢ 1.15
mm.; @ 1.4mm,

Sub-brachypterous form, fig. 1.

Colour: Ochreous; head ferruginous, usually darker; eyes brown-
ish black; ocelli reddish; antennae with fourth segment and usually
extreme apex of third black; fourth rostral segment, the claws and the
apex of third segment of tarsi black; venter of thorax black in middle
between coxae, the rest ochreous to dark brown; pronotum and scutel-
ium ochreous, usually darker than hemelytra and more or less infuscated ;
calli brown; hemelytra pale stramineous to ochreous, sometimes more
or less infuscated in streaks and around punctures; inner margin of
corium, bordering membrane, narrowly infuscated, more conspicuously
black at base behind apex of clavus and at apex; whole dorsal sutface
occasionally with reddish tinge; membrane colourless and transparent
towards apex, infuscated towards base; venter of abdomen ochreous or
ferruginous, sometimes nearly black towards hase.

Head: From above about 4 as wide again across eyes as long
(52:38). Each eye about 4 as wide as interocular space (9: 34) ;
distance between ocelli twice that between each ocellus and the level of
the innermost margin of the eye (taken transversely) (15: 7). Head
above with close punctation, giving a granularly roughened appearance,
punctures more remote on tylus and antenniferous tubercles; below
finely and more remotely punctate; above and below with a covering of
deciduous, golden, scale-like hairs, most of those of upper surface
usually rubbed off, those of under surface usually concealing the pune-
tures.

Antennae 1/6%-4 as long again as head and pronotum together ;
segment I short, stout, equal or subequal in length to distance between
ocelli, subcylindrical, narrowing slightly towards apex; segment
TI short, equal or subequal in length to I and only 3-g as long as ITT,
slightly and gradually thickened on apical half ; [Il the longest and most
slender. swollen at extreme apex; IV fusiform, subacute apically,

226 W oopw arb.

eB“ @¢oe*sam
8 eh Tar enaceeroe®
ore ca-*? ew. 2 7 @e y

@*
=

®_ =

2
2
oe

x
22
e*

C=
2
Ce
a

2
Pe
32

‘LU UU o:¢

&
a»

« 8
[7
@ 243 ®
@
€
2

"ul OZ

Figs, 1-2. Cymus novacselandiae sp. nov. 1, sub-brachypterous ¢ ; 2, macropter-
ous ¢, outline to show form of hemelytra.

strongly swollen, width near middle about 4 length and rather wider than
I; relative length of segments I-IV, 15: 15: 42: 30 (III and IV may
vary even between the antennae of the one individual) ; relative width,
I 7: II 2.5 at base, 4 at apex; III 3 at base, 4.5 at apex; IV 9.5 in
middle ; all segments with minute, setiferous tubercles, the hairs of seg-
ments I-IiI very short, pale, inconspicuous, semi-recumbent, rather
sparse; IV with a close covering of very much longer black hairs.
Rostrum reaching to posterior end of middle coxae, sometimes as far
as anterior margin of hind coxae; segment I not quite reaching base ot
head; IL reaching or almost reaching front coxae; III passing front
coxae but not reaching middle coxae; segment I the longest, then in
erder, IV, II, II], but II sometimes equal to III, and sometimes
nearly as long as IV; relative length of segments I-IV, 24: 16: 16: 21.
Bucculae half as long as first rostral segment or nearly so, pale ochreous,
with three or four punctures.

Thorax: Pronotum trapeziform; anterior and posterior margin
nearly straight, the latter very broadly and shallowly emarginate; sides
nearly straight, very feebly sinuate; rather longer than head (43: 38 )
and 1/6-1 as wide again across posterior angles as head across eyes
(62: 52); median length 3-} posterior width (43: 62) and equal to or
only slightly greater than anterior width; anterior collar 7 as long as
whole pronotum, demarcated laterally, in front of calh, by more or less
distinct transverse impressions; central part of disc, between calli, rather
convexly raised but only obsoletely carinate; posterior half of dise
scarcely raised, nearly flat, not markedly declivous at basal margin ;

Hemiptera-Heteroptera. 227

whole surface of pronotum, except the well defined calli, with large
punctures mostly less than one puncture-width apart. Under-side of
all thoracic segments, to bases of legs, similarly though rather more
obscurely punctate, with a covering of deciduous hairs similar to those
of the head (pronotum apparently originally with similar hairs, one to
each puncture, but these haye been worn down to their extreme bases,
except where protected in the depressions between anterior collar and
calli). Median part of thoracic venter, between bases of legs, with fine
punctures several puncture-widths apart. Metathoracic scent-gland
spout very small, tubercle-like, ventrally truncate. Scutellum with fine,
rather obscure punctures, disc convexly raised towards median line, but
without a distinct carina; base strongly depressed; 1¢ to twice as wide
at base as long (31: 17). Legs short, with a covering of very short,
pale, inconspicuous hairs, longest on hind tibiae and tarsi; relative length,
femur: tibia: tarsus (excluding claws), as 32: 33: 18 (front leg) ;

50: 47: 20 (hind leg).

Hemelytra: Costal margins of corium strongly and evenly convex
until about 1/5 from base, where sinuately incurved, thence nearly
straightly converging to base; costal border above only narrowly flat-
tened: corium twice to 2} times as long as outer margin and 3-4 times
inner margin (behind scutellum) of clavus (116: 55:35). Clavus with
three rows of deep punctures larger than those of pronotum, those of
each row nearly contiguous: a row closely parallel to claval suture, a
V-shaped row parallelling the inner and the scutellar margins, and a
straight row between them; in addition, usually one or two punctures
hetween the two last-mentioned rows near the inner end; whole_of
corium covered with similar punctures. Membranes relatively short
and narrow, each about as long as corium (75: 116) and with greatest
width less than (0.55-0.88) greatest width of corium (25:40) ; extend-
ing beyond apex of coria for only 1/ 10%-1/6 of their own length
(12: 75) and for a distance equal to only | /17*-1/9 of the corium
length.

Abdomen with venter very finely pubescent at base, except at sides,
remainder shining, nearly glabrous, with extremely short, sparse, pale
hairs.

The actual measurements are those of the holotype ¢ ; the ranges
of proportions are from the whole available series of both sexes. There
are no constant sexual differences in any of the proportionate dimen-
sions, except as detailed below.

Length: ¢ 24mm-2.7mm.; 2? 2.9 mm-3.2 mm.
Width: ¢ 1.0mm.-1.13mm.; 2 1.20 mm.-1.27 mm.
Ratio of length to width: 3 2.2-2.5; 9 2.42.5.

Macropterous form, fig. 2.

This differs from the sub-brachypterous in the following respects:
Hemelytra longer, with costal margin of corium sinuately incurved at
apex as well as base. Membrane much longer and wider, 1m the two
specimens to hand 0.84 ( ¢ ) and 0.75 (¢) times as long as corium
wider than corium (¢@ 60; 42; 9 54: 50); extending beyond apex of
corium for about 4 of its own length in @ (43: 112) and about 4 in

228 Woopwarp,

9 (31: 112) and for nearly 4 of corium length in 6 (43: 133) and
about 1/5 in @ (31: 150). Pronotum proportionately rather larger,
about 4 as long again as head (51:41 in ¢,50:44in @ ). Size greater
and form more elongate. Length: ¢, @ 3.5mm. Width: 3 1.15 mm. :
9 14mm. Ratio of length to width: ¢ 3.0; @ 2.5.

Both these macropterous specimens are dark; the pronotum, clavi
and coria similarly infuscated ochreous; calli brownish black in 4 ;
center of abdomen ferruginous, infuscated on basal half, where nearly
black in 2.

Sub-brachypterous form:

Types: Holotype ¢, Paiaka, Manawatu, North I. 9/1/50, T.E.W.
Allotype ?, Otautu area, Cape Colville, Coromandel Pen., North I.,
16/1/52, sweeping grasses, T.E.W. In Auckland Museum.

Other specimens examined (including paratypes). Mainland and
nearby islands: 1 6,192, Kaitaia, N. Auckland, N.I., 13/5/23, J. G.
Myers (Dom. Wins. ae <a, a Rs Auckland, 20/3/49, T.E.W.; 13,

Auckland, 8/49, T.E.W.; 19, Paiaka, Manawatu, 2/2/51, T.E.W.;
1 ¢, Kawau I., Auckland, 5/1/51, T.E.W.

Three Kings Is.: 16, N.W. cliffs, Great I., 4/1/53, sweeping
Carex and Scirpus, J. S. Edwards; 1 ¢@, cliff brow, 5.W. Cove, Great L.,
5/1/53, on Scirpus nodosus, J. 5. Edwards.

Macropterous form:

One @, Dyer’s Pass, Christchurch, 5.L., 9/1/25, C. E. Clarke, coll.
(Auck. Mus.); 1 ¢, Remuera, Auckland, 1/2/52, E. T. Giles.

This is in all probability the species referred to by Myers (1926;
pp. 457, 462, 485) as Cymodema sp., of which he notes: “Adults have
been taken in August, December, and February in the sweeping-net, and
ir, May and July in winter quarters at the bases of rush-clumps, where
they were most abundant in the later month.” Of host plants, he says,
‘“Meadow-grasses, especially Bromus unioloides K.B.K., are favoured.
In winter the bases of Juncus effusus L. afford them shelter. a
He records the species as having been collected in North Auckland
(Kaitaia), Wanganui, Levin (Weraroa), and Waikanae, all in the
North Island.

Abnormalities:

In one @ (Otautu) the punctation of the mght clavus is much
reduced. The inner row has only 5 punctures behind scutellar apex
(instead of about 8) and 2 near base; the outer row has only 2 punc-
tures in apical half and 3 near base. The middle row is lacking. The
basal half of the clavus is thus impunctate except at the extreme base.
Those punctures present are of normal form, The punctation of the left
clavus is normal.

Two specimens show antennal oligomery, a common malformation
in the Lygaeidae. In 1 6 (Auckland, 8/49), the left antenna shows
no division between segments IIT and IV, and the resultant fused seg-
ment is considerably shortened (ratio to III + IV on right antenna,

Hemiptera-Heteroptera. 229

40: 64) and apically dilated as a black club, smaller and blunter at base
and apex than the normal segment IV. The appearance 1s of regenera-
tion following loss of the two apical segments in a nymphal instar.
In 1 @ (Kaitaia), the left antenna shows no obvious segmentation after
segment I and the unsegmented region is shortened as in the é (ratio to
1] + Ill + IV of right antenna, 56: 87). There is also a curious
rmalformation of the unsegmented region, which is clavately swollen and
black at about half-way, then twisted, the apical half coming off from
one side of the club, and narrowed; at the extreme apex the antenna
is swollen again as an ovoid, stouter, and much shorter black club, The
impression given is of an initial break at the extreme base of segment
II of the nymphal antenna, followed by a second, apical break and/or
damage after the regeneration of the first club.

(Ali the proportionate measurements given for the species are to
the scale 75 units — 1 mm.)

Dr. W. E. China has determined the present species as belonging
to Cymus. The short second antennal segment, subequal to the first,
is not usual in the genus, but there is another instance in the North
American species Cymus breviceps Stal.

FAMILY REDUVIIDAE.
SUB-FAMILY EMESINAE.

Empicoris rubromaculatus (Blackburn).

1889—Ploiariodes rubromaculatus Blackburn, Proc. Linn. Soe. N.S.W. 3
(1): 349.

One 2,1 9, Tasman Valley, Great [., 31/12/52, sweeping from
Leptospermum, J. S. Edwards.

A widely distributed species, occurring in the Americas, the Pacific,
Australia and New Zealand.

FAMILY NABIDAE.

Nabis maoricus F. Walker.
1873—Nabis maoricus F. Walker, Cat. Hem, Het. Brit. Mus. 7: 145.
1878—Nabis saundersi Buch. White, Ent. mon. Mag. 15: 159.
The Saddle, Great I., 2/1/53, 1 2 on kanuka (Leptospermuim

cricoides), 1 2 on ngaio (Myoporum laetum), J. 5. Edwards. 1 9, E.
block, Great I., 15/1/51, T. E. Woodward.

Endemic to New Zealand, where common and widespread. Dr.
W. E. China has compared New Zealand specimens of both this spectes
and the following with material in the British Museum, and gives the
above new synonymy as probably correct. Previously, NV. saunders!
has been synonymuised by some authors with N. capsiforiits.

Nabis capsiformis Germar.

1837—Nabis capsiformis Germar, Silbermann Rev. Ent. 5: 132,

One @, 2 nymphs, Castaway Valley, Great L., 12/7 by Si Ee Bs,
Woodward.

230 WoopwarbD.

This species is widespread in the Pacific and is now nearly cosmo-
politan.

FAMILY ANTHOCORIDAE,

Specimens of apparently a single species of Anthocorid were taken
by J. S. Edwards and the author from Great I. (on Chenopodium
triandrum and Leptospermum and from leaf mould, and by the author
from S.W. Island (sweeping Myoporum lactum and Muehlenbeckia
complexa). As the Anthocoridae of the Australian region are in pro-
cess of revision by Mr. G. F. Gross, of the South Australian Museum,
specimens have been forwarded to him, and it has seemed desirable to
await his determination on the basis of this wider study.

Mr. Gross has identified these specimens as belonging to a new
species of Lasiochilus Reuter, which he ts describing, and which is also
widespread on the mainland of New Zealand.

FAMILY ARADIDAE.

A single specimen was extracted from leaf mould collected by the
author under ngaio (Myoporum laetum) on S.W. Island, 13/1/51.
Together with other New Zealand material, this has been forwarded
to Prof. R. L. Usinger, of the University of California, who is engaged
in a revision of the Aradidae of the world.

FAMILY MIRIDAE.
SUB-FAMILY MIRINAE.

Megaloceroea reuteriana Buch. White.
Sih iA Soo (Megaloceroea) reuteriana Buch. White, Ent. mon. Mag.
; 18v.

Collected by author: 4 6 4,10 2 2,5 nymphs, Tasman Valley (east
side), Great I., 12/1/51, sweeping grasses and sedges; 2 6 6,3 9 @,
7 nymphs, near depot, Great I., 12/1/51, sweeping grasses and sedges;
East Point, Great I.,17 ¢ ¢, 5 @ 2, 5 nymphs, 14/1/51, on and under
Poa anceps Forst. on cliff slopes, 1 ¢,1 &, 15/1/51, sweeping sedges
and rushes.

Collected by J, S. Edwards: 1 9, Tasman Valley, Great L.,
30/12/52, on Ipomoea; 5 4 6, 1 nymph, cliffs at end Tasman Valley,
Great I., 31/12/52, sweeping grasses and Scirpus; 2 2 9, Eastern arm,
Great I., 1/1/53, sweepings from ground vegetation (Carex, mosses,
etc.); 1 nymph, kanuka canopy near Quadrat II, Great lL. 2/1/53
( Leptospermum ericoides); 2 8 &, the Saddle, Great I., 2/1/53, on
ugaio (Myoporum laetuim Forst.) ; 1 nymph, N.W. cliffs, Great L.,
4/1/53, 480ft., sweeping Carex and Scirpus; 1 4,1 @, cliff brow, 5.W.
Cove, Great I., 5/1/53, on Scirpus nodosus Rottb.

This species is widespread in Australia and New Zealand and is
common on grasses, sedges and rushes.

Calocoris laticinctus (Walker).
1873—Capsus laticinctus FP. Walker, Cat. Hem.-Het. Brit. Mus. 6: 128.

Tasman Valley, Great I.;2 6 6,1 9, 11/1/51, sweeping Blechnuim ;
1 4,1 9, 12/1/51, sweeping grasses and sedges, T.E.W.

Heimiptera-Heteroptera, 231

This species, which has long been synonymised with Capsus
ustulatus EF. Walker, loc. cit., was placed by Distant in the genus
Calocoris, but, as Dr. China has pointed out (im Iitt.), it differs from
other species of Calocoris in having the head completely transversely
carinate behind. Eventually it will most probably have to be removed
{rom this genus. The species is known only from New Zealand.

EKurystylus australis Poppius.
1911—Eurystylus australis Poppius, Ofvers, Finsk. Vet. Soc. 53 A (4): 15.

One 9, cliffs below Tasman Valley, Great I., 31/12/52, swept
from Haloragis erecta (Murr.) Schind.; coll, J. S. Edwards.

This species occurs in both Australia and New Zealand, having
apparently been introduced from the former to the latter.

Genus LYGUS Hahn, 1833.

Three species of Lygus were taken, but as this genus is a very large
and cosmopolitan one, and most of the New Zealand and Australasian
species are either undescribed or inadequately described, the naming
of isolated new species is highly undesirable. The Lygus fauna of these
regions needs revision as a whole and the author hopes to attempt this
before long.

Sp. 1. Numerous specimens (several hundred) were collected by
J. S. Edwards and the author from Leptospermum ericoides (kanuka)
on Great I. The same species is also extremely abundant on kanuka on
the mainland of New Zealand.

Sp. 2. Specimens were collected by J. S. Edwards and the author
from Myoporum laetum (ngaio) on Great 1. This species also occurs
on the mainland.

Sp. 3. Taken by the author from shrubs on both Great I. and
S.W. Island. Not known from the mainland.

SUB-FAMILY ORTHOTYLINAE (CYLLECORINAE).
Coridromius variegatus (Montrouzier).
1861—Ocypus varicgatus Montrouzier, Ann. Soc. ent. Fr. (sér. 4) 1: 67.

Eight ¢ ¢,9 2 2,21 nymphs, $.W. Island, 13/1/51, on and under
Salicornia australis Forst.; 14 6 6, 17 2 9, 28 nymphs, East Point,
Great I., 14/1/51, Chenopodium triandrum Forst., on rocks at toot of
cliffs: 1 ¢, 3 22, near depot below cliffs, Great I., Chenopodium
triandrum,; coll. T. E. Woodward,

This species was kindly determined by Dr. W. E. China, British
Museum (Nat. Hist.). This is the first record of the species from
New Zealand, and I have subsequently taken it from the Wellington
district, North I. (1/2/51, Day’s Bay, on Lepidium oleraceum Forst.,
and Titahi Bay, on and beneath Salicorma australis, Lepidium oleraceum
and Chenopodium triandrum). All the plants on which it has been
found in New Zealand are coastal succulents. This small, mottled

232 W COoDWARD.

species occurs also in Australia and New Caledonia. The hind femora
are incrassate and the insect is capable of jumping some distance into
the air when disturbed.

SUB-FAMILY PHYLINAE (PLAGIOGNATHINAE).
Sthenarus myersi Woodward.

1950—Sthenarus myersi Woodward, Rec. Auck. Inst. Mus. 4 (1); 22-23;
figs, 12-15.

Collected by author: 22 ¢ ¢, 18 @ 9, 2 nymphs, Tasman Valley,
Great I., 11-12/1/51; 2 9 @, 1 nymph, near depot, Great I., 12/1/51;
3 64,7 22,2 nymphs, East Point, Great I., 15/1/51:

Collected by J. S. Edwards: 1 2, near depot, Great I., 30/12/52;
1 ¢, Tasman Valley, Great I., 31/12/52; 1 @, on cliffs, Great L.,
4/1/53; 1 ?, Bare Saddle, S.E. Bay, Great I., 5/1/53.

All the specimens were collected on kanuka (Leptospermuim eri-
coides A. Rich.).

This species is known only from New Zealand, where it has a wide
range at least in the North Island.

With the larger series now available, additional information can be
given on variations in colour. The males tend to be darker than the
females, but both sexes may be more or less rufescent, either above or
both above and below, the cuneus sometimes entirely red or reddish
brown, and some males, though in smaller proportion than the females,
have the basal two-thirds of the second antennal segment ochreous. All
specimens are of the larger type, as described by the author (1950)
from Manawatu. |

Sthenarus sp.

This species, which so far as can be determined at present 1s new,
and is not known to the author from the mainland, was taken on pohutu-
kawa (Metrosideros excelsa Gaertn.), 12-15/1/51 (T.E.W.). Untor-
tunately, the specimens are damaged, and description has been deferred
until a better series can be obtained.

FAMILY VELIIDAE.
SUB-FAMILY MICROVELIINAE.

Microvelia halei Esaki.

1926—Microvelia oceanica Hale, Rec. S. Austr. Mus. 3: 208; nec Distant
1914, in Sarasin and Roux, Nova Caledonia, Zool. 1: 383.

1928—Microvelia halci Esaki, Ins. of Samoa 2 (3): 69; new name for M,
oceanica Hale,

Thirty-six apterous ¢ 6, 33 apterous 9 2,1 macropterous 9, 40
nymphs, Tasman Stream, Great I., 11/1/51, T.E.W.

Occurs in Australia, New Zealand, and Lord Howe I.

Hemuptera-Heteroptera. 233

REFERENCES.

BATTEY, M. H., 1951. Notes to Accompany a Topographical Map and a Pro-
visional Geological Map of Great Island, Three Kings Group. ec.
Auck. Inst. Mus. 4 (2), 93-97, pls. 8-9.

BAYLIS, G. T. S., 1948. Vegetation of Great Island, Three Kings Group. Mec.
Auck. Inst. Mus. 3 (4 and 5), 239-252.

MYERS, J. G., 1926. Biological Notes on New Zealand Heteroptera. Tvrans.
N.Z. Inst. 56, 449-511.

MYERS, J. G, and CHINA, W. E., 1928. A list of New Zealand Heteroptera
with the description of a remarkable green Aradid representing a
New Genus. Ann. Mag. Nat. Hist. (10) 1, 377-394.

SPILLER, D., and TURBOTT, E. G., 1944. The occurrence of some Australian
Insects and a Spider in New Zealand. Rec. Auck. Inst. Mus. 3 (1),
79-83.

-TURBOTT, E. G., 1948. Effect of Goats on Great Island, Three Kings, with
descriptions of Vegetation Quadrats. Rec. Auck. Inst. Mus. 3 (4
and 5), 253-272.

USINGER, R. L., 1942. The Orsillini of New Zealand. Trans. R. Soc. N.Z.
72 (1), 41-52.

WOODWARD, T. E., 1953a. The Heteroptera of New Zealand. Part I—Intro-
duction; Cydnidae; Pentatomidae. Trans. R. Soc. N.Z. 80 (3 and
4), 299-321.

WOODWARD, T. E., 1953b. New Genera and Species of Rhyparochrominae
from New Zealand (Heteroptera; Lygaeidae). Rec. Cant. Mus. 6
(3) > 191-218.

New Zealand Molluscan Systematics,
with Descriptions of
New Species, Part 2.

By A. W. B. POWELL, Auckland Museum.

Genus SEPTIFER Recluz, 1848.
Septifer cf. bilocularis Linn.
1758—M ytilus bilocularis Linn. Syst, Nat. 10th ed., p. 705 Indian Ocean.

A juvenile valve in fresh condition was obtained by Mr. Gordon
Williams from the stomach contents of fish taken in 45 fathoms off
Mayor Island, Bay of Plenty. It is 3.5mm. in length and 3.2 mm. in
height. The fish was a tarakihi (Dactylopagrus macropterus Forster).

The occurrence of this widespread Indo-West Pacific coral dwell-
ing mussel in New Zealand waters is surprising, but small valves similar
to the New Zealand example are not uncommon in a dredging from
10-30 metres off Sunday Island, Kermadecs.

Although the New Zealand specimen is very small it shows the
unmistakable characteristics of the genus; i.e., variegated blue, green
and reddish brown coloration, crenulated internal edge of the shell,
short hinge on an internal ledge, muscle shelf behind the hinge and
external sculpture of radiating closely-packed flattened ridges.

The muscle shelf is only half developed in this juvenile shell, It
extends as a triangular plate from the ventral margin, but in some
shightly larger Kermadec specimens the shelf extends two-thirds of the
way across.

The excellent condition of the Mayor Island valve indicates that
it could not have been long in the fish’s stomach and must have been
taken in the vicinity, thus ruling out the possibility of the fish having
migrated from warmer seas with the shell in its stomach.

Proxiuber hulmei n. sp.

Shell of similar size to australis but proportionately broader, not
so globose, with a distinctive colour pattern of two broad spiral zones
of rectangular axial dark reddish-brown markings on a white ground
covered by a very thin buff epidermis. Umbilicus a more convex cres-
cent than in maorta, widely open, although half filled by the funicle
and columellar callus. The two colour zones are both broad, the first
subsutural and the second peripheral. They are separated by a clear
space equal to half the width of a colour zone and the lower part of
the base, almost equal to the width of a colour zone, is clear also. The
surface 1s smooth and glossy, with faint, closely-spaced axial growth
lines. The operculum is calcareous, smooth, white, paucispiral with
two faint grooves margining the outer edge.

Rec. Auck, Inst. Mus. Vol. 4, No. 4, pp, 235-240, 12th February, 1954

236 POWELL.

Height, 4.2mm.; diameter, 4.5mm, (Holotype).

Locality: Obtained from an Auckland trawler, exact locality and station not
known.

Holotype: Presented to the Auckland Museum by Master S. G. Hulme.

Proxiuber hulmei n. sp. 4.2mm. x 4.5 mm.

Genus MYSTICONCHA Allan, 1936.

Mysticoncha harrisonae Powell.
1946—Mysticoncha harrisonae Powell. Rec. Auck. Inst. Mus. 3 (2) p. 144.

Holotype: Lowrie’s Beach, The Neck, Stewart Island. Powell collection,
Auckland.

Two further examples of this species are now know. (1) 60
fathoms east of Stewart Island in stomach of blue cod (Mr. T. E.
Jensen), (2) Pohara Beach, near Takaka, Nelson (the late Mr. W. V.
Hadfield).

Cabestana (Cymatilesta) otagoensis n. sp. Pl. 39, fig. 1.

Shell large for the waterhousei group, slender, solid. Protoconch
missing. Early whorls with two bifid, beaded prominent spiral keels,
uppermost medial and four secondary beaded spiral cords, two above
the medial keel, one between the keels and one below. On the ante-
penultimate the secondary spirals increase to four above the medial keel
and two below the lower keel. On the body-whorl there are about 13
primary bifid spirals and a single cord in each interspace. Varices
prominent, flange-like, spaced at two-thirds of a whorl intervals. Inter-
variceal axials weak, six to seven in number, unequally spaced, always
bunched over two-thirds of the distance, leaving a clear space betore
the next varix commences. Height of spire about four-fifths height
of aperture plus canal. Colour buff, intercostal spaces on the varices
banded with pale reddish brown. Operculum ovate-pyriform with a
terminal nucleus. Epidermis completely worn off.

Height (actual), 93.0mm. (estimated), 94.0mm.; diameter, 46.0 mm.

Locality: 10 miles north of Moeraki, Otago, 45 fathoms.
Holotype: Auckland Museum, Presented by Captain J. Black.

N.Z. Molluscan Systematics. 237

This species belongs to the debilior line rather than to the water-
housei line. The characteristic features are (a) the varices curving
upwards and clasping the preceding whorl almost to the height of the
lower keel and (b) the curious bunching of the axials in the inter-
variceal spaces.

_ In the Castlecliffian debilior Finlay, 1930, there are four bunched
axials per intervariceal space compared with six or seven in otagoensts.

A shell washed up at Hokeo Beach, Levin, has the characteristic
four bunched axials of debilior. Shells of a number of species judged
to be derived from Pliocene beds are not uncommon on the Manawatu
beaches. Cabestana manawatuna Fleming, 1943, from Locality 2492,
Tahoraiti (S.W.) S.D., Lower Nukumaruan, seems to be ancestral to
debilior rather than to waterhousel.

Mayena australasia blacki n. subsp. Pl. 39, fig. 2.

Shell larger than the typical species, prominently shouldered.
Whorls angled medially, scarcely keeled and bearing bluntly rounded
nodules, seven or eight between varices. There is no second or sutural
angulation. Surface sculpture of dense spiral lirations and weak low
primary spirals, two or three on the shoulder and about five on the body-
whorl and base. Spire taller than aperture plus canal. Aperture
strengthened with a heavy varix, strongly dentate within and with a
massive parietal tubercle. Canal short, slightly oblique but little
recurved. Colour pinkish buff obscurely maculated with pale reddish-
brown. Varices banded with light purplish brown, and white where
crossed by the weak primary spirals. Labial callus and interior of
aperture porcellanous white. Epidermis yellowish-brown, densely axially
lamellated, the lamellae crowded with short bristles.

Height (actual), 121.0mm. (estimated), 124mm.; diameter, 62.5mm. (holo-
type).

Height (actual), 118.5mm.; diameter, 59.5 mm.

Locality; Off Eastern Otago in 60-70 fathoms.

Holotype: Auckland Museum. Presented by Captain J. Black.

Two large examples of a Mayena from Bluff oyster boats, taken in
10 to 15 fathoms Foveaux Strait, resemble Northland shallow water
australasia except for a relatively taller spire and weak numerous
nodules (10-14 between varices) confined to the single peripheral angu-
‘ation. When more material is available the Foveaux Strait form may
be separable from typical australasia.

The relationship of blacki appears to be with this shallow-water
Forsterian form of australasia rather than with subspecies voss: Powell,
1952. from 70 fathoms off Mayor Island, Bay of Plenty.

Other interesting records of warm-water Cymatidae in the
Forsterian are :—
Ranella multinodosa (Bucknill), off Eastern Otago, 60-70 fathoms.

Cabestana (Cymatilesta) spengleri (Perry), Foveaux Strait, oyster
dredge (Mrs. R. H. Harrison) and (Mrs. E. Smith).

Cabestana (Cymatilesta) otagoensis n,sp., 45 fathoms, 10 miles
north of Moeraki.

238 POWELL.

Genus EMOZAMIA Iredale, 1929.
Type (0.d.): Murex licinus Hedley and Petterd, 1906.
Emozamia licina (Hedley and Petterd). Pl. 39, figs. 5-7.

1906—Murex lhcinus Hedley and Petterd, Rec. Austr. Mus. 6 (3), p. 219,
Pl. 37, fig. 6.

1929—Emozamia licina: Iredale, Rec. Austr. Mus. 17 (4), p. 185.
Holotype: Off Sydney, New South Wales, 250 fathoms.

The New Zealand specimen described below and figured (Plate 39,
fig. 7) is probably identical with the New South Wales species. It is
a rare shell in New South Wales collections, but through the courtesy
of Miss Gertrude Thornley and Mr. C. F. Laseron, both of Sydney, |
have two specimens on loan for comparison with the New Zealand shell.
One of these is from Bateman’s Bay, 60 fathoms, and the other Port
Stephens, 30-40 fathoms.

The secondary or intermediate spirals are absent in Hedley's draw-
ing of the holotype, weak in the Bateman’s Bay and New Zealand
specimens and strongly developed in the Port Stephens example. There
is insufficient material to evaluate these differences, so, for the present,
only one species is admitted. Following is a description of the New
Zealand shell :—

Shell small, lightly built, broadly rounded, with a low spire and
sculptured with vertical, low, rather narrow fluted varices, eight per
whorl. On the spire-whorls there is a single rounded spiral cord and
on the body-whorl the addition of five similar cords. There is a weak
secondary spiral thread between the suture and the peripheral cord and
one between each pair on the body-whorl. The cords do not cross the
varices but form into a weak recurved spinose process at the crest ot
each varix. The surface of the shell is otherwise smooth and the colour
uniformly pale pinkish-buff. Aperture relatively large, broadly-ovate.
Anterior canal, partially closed, broad at the base but rapidly tapered,
recurved and about one-third the height of the aperture in length.
Fasciole narrowly arched, strongly imbricated and with a narrow false
umbilicus partially bridged by the almost free parietal callus.

Height, 16.0mm.; diameter, 12.0mm. .

Locality: 46-82 fathoms N.N.E. of Mayor Island, Bay of Plenty (S. M.
Hovell); obtained from an Auckland trawler, exact locality and station not
known (S. G. Hulme).

Genus IREDALINA Finlay, 1926.

The holotype of [redalina mirabilis Finlay is a large, slender shell
with a narrowly angled attenuated spire. It remained unique for over
twenty-five years, but dead encrusted shells of /redalina have been taken
recently in some quantity by Captain J. Black, of the Dunedin trawler
*“Taiaroa.”

In 1950 (Rec. Auck. Inst. Mus. 4 (1), p. 81) I recorded four
specimens from the above source and noted that these had a shorter
spire and a more inflated body-whorl. Dell (1951, Rec. Cant. Mus.
6 (1), p. 57) also recorded a similar squat inflated example from 80
fathoms off Banks Peninsula.

N.Z. Molluscan- Systematics. 239

The considerable number of Eastern Otago shells I have now
examined show that there are two forms or species of Jredalina on the
eastern Otago shelf and that the genotype, mirabilis is much the rarer
of the two.

At first | was inclined to consider the holotype to be an abnormality
exhibiting: elongation resultant from some early injury, but further
examples of the slender typical mirabilis show that there is a constant
re ie between the slender and squat shells that is not related to
shape.

All five typical mirabilis I have examined are lacking in sculpture
on the pillar and all of the thirty or more examples of the squat species
have about ten slightly raised spiral threads running around the pillar
and terminating at the broad shallow oblique anterior notch. This
pillar sculpture was noted by Dell (1.c.) also in his Banks Peninsula shell.

Iredalina mirabilis Finlay. Pl. 39, fig. 3.
1926—Jredalina mirabilis Finlay, Proc. Mal. Soc. 17, pp. 59-62.

Localities: 40 fathoms off Otago Heads (Holotype); 58 fathoms off Waitaki
tiver (Captain J. Black).

Holotype: Finlay collection, Auckland Museum.

Shell large, always slender with a spire angle of 32-33° and no
spiral threads on the pillar. The colour is indicated as uniformly
salmon-pink without colour pattern and with a high glaze in life. No
living examples have been taken.

Height. Diameter Ht. + Diam. Ht. + Ht. Aperture. Spire Angle.

140.0 mm. 48.0 mm. 2.91 1.97 ion tt
147.0mm. 48.0 mm. 3.06 1.98 33°
120.0mm. 40.5 mm. 2.96 1.93 32°
112.0mm. 39,5 mm. 2.83 1.89 32°

* — Holotype.

Iredalina aurantia n.sp. Pl. 39, fig. 4.
1950—/redalina mirabilis: Powell, Rec. Atcck. Inst. Mus. 4 (1), p. &l.
1951—J redalina mirabilis: Dell, Rec. Cant. Mus. 6 (1), p. 57.

Localities: 50-60 fathoms off Cape Saunders; off eastern Otago, 60-70 fathoms
(Holotype) (Captain J. Black) ; 80 fathoms off Banks Peninsula.

Shell smaller than mirabilis, more inflated with a spire angle of
44-52° and about ten slightly raised spiral threads running round the
pillar. The colour, indicated by one very well preserved example from
55 fathoms off Cape Saunders, is uniformly salmon-pink with a high
glaze. It is a very similar colour to that of the well known Fijian orange
cowry, Cypraea aurantium Linn.

The station of /redalina is still problematical. They must occur
adjacent to the 60-70 fathoms hard bottom where the shells are trawled.
The 55 fathoms Cape Saunders shell is the best preserved one so far
taken, and this shell was not long dead, having full colour, original
glaze, and was neither worm eaten nor encrusted.

240 POWELL.

Height. Diameter Ht. ~ Diam. Ht. + Ht. Aperture. Spire Angle.

77.0mm. 35.0 mm, 2.17 1.68 52°(A)

87.0mm. 38.5 mm. 2.26 1.58 50°(B)
109.0mm. 49.0 mm. B22 1.60 St Bae
116.0mm. 46.0 mm. 2.52 1.75 50° (B)
116.0mm. 50.0 mm. 2.30 1.61 46°(B)
116.0mm. 52.0 mm. 2.23 1.61 51°(B)
117.0mm. 50.0 mm. 2.34 1.54 58° (B)
119.0 mm. 49.0 mm. 2.42 1.70 44°(C)

A = 80 fathoms off Banks Peninsula, B = 60-70 fathoms off Eastern Otago,
C = 50-60 fathoms off Cape Saunders. * = Holotype.

Pachymelon (Palomelon) smithi Powell.

1950—Pachymelon (Palomelon) smith Powell. Rec. Auck. Inst. Mus. 4 (1),
p. Sl.
Holotype: Off Eastern Otago ca. 70 fathoms. Auckland Museum.

A living specimen was trawled by Captain Black in 55-60 fathoms
south of Timaru. The ground colour is pale orange with a conspicuous
stain of bright purple around the fasciole and the suture at the termina-~
tion of the last whorl. Most shells are devoid of other markings, but
some have a sparse pattern of narrow angular streaks in dark reddish-
brown arranged in three bands, one subsutural, one medial, and the third
just above the fasciole.

Height. Diameter.
118.0 mm. 45.5 mm. (Holotype)
113.0 mm. 44.0 mm.

73.0 mm. 29.0 mm.

The third specimen is one of several that have a well callused outer
lip and are obviously adult. The axial sculpture is more pronounced and
extends on to the body-whorl but otherwise there are no differences.
The phenomenon of nanism is not uncommon among the deep water
volutes of both New Zealand and Australia.

PLATE 39.

Fie. 1. Cabestana (Cymatilesta) otagoensis n. sp. Holotype 93 x 46mm.

Fig. Z. Mayena australasia blacki n. subsp. Holotype 121 x 62.5 mm.

Fig. 3. Iredalina mirabilis Finlay. Holotype 140 x 48 mm.

Fig. 4. [redalina aurantia n. sp. Holotype 109 x 49mm.

Fig, 5. Emozamia licina (Hedley and Petterd), Bateman’s Bay, 60 fathoms, New

South Wales; Fig. 6. Port Stephens, 30-40 fathoms, New South Wales
Fig. 7. 46/82 fathoms N.N.E. of Mayor Island, Bay of Plenty, New
Zealand, 16 x 12mm. (All three figures uniform magnification),

A New Rail From Cave Deposits in the
North Island of New Zealand.

By R. A, FALLA, Dominion Museum, Wellington.

Bird bones from a limestone cave about 13 miles from Hamilton,
North Island, N.Z., discovered in 1948 by a party of the New Zealand
Speleological Society led by Mr. H. G. Lambert and presented by the
Society to the Auckland Museum, have been sent to me for examination,
and I am grateful to the Director of the Auckland Museum, Dr. G.
Archey, and to Mr. E. G. Turbott, ornithologist, for the opportunity to
describe them.

The circumstances of the discovery have been described in a narra-
tive account by P. Chester (Nezsview, Auckland, May, 1953) and from
it the following extracts are quoted: “High spot of the society’s activity
is the Karamu Cave, the most extensive one so far discovered in New
Zealand. . . . Passages run through it for eight miles . . .” The
article continues that: “The Bird Cave got its name from the skeleton
of an extinct bird which Lambert and a party found on their second
survey of the cave,’ and gives Lambert’s own description as follows:
“For four hours we'd lugged photographic equipment through the cave,
hoping to find something worth photographing. Eventually we came
to a spot below a gallery which we decided to explore. It took us
another hour to find a way up. Then we spied the bird skeleton on
the floor.”

When finally extracted from the limestone matrix of the cave floor
the bones were in fragmentary condition, but fortunately an excellent
flashlight photograph (Plate 40) was taken by Mr. Lambert before they
were disturbed, and this has saved much conjecture as to the probable
proportions of the bird. It is a small rail, resembling in general pro-
portions Cabalus modestus Hutton, but larger. Pectoral girdle, sternum,
and wing-bones are missing, so that its flight potential remains unknown
and can only be guessed at. From the fact that the legs have the pro-
portional stoutness found in known flightless rails (Gallirallus, Cabalus )
4 similar condition might be inferred. The bones available may from
their position in situ be referred to one individual. They are :—

Skull: Right half cranium with occipital foramen intact ; premaxilla and
mandibles complete but now in fragments tending to crumble
because of their original light cancellated structure.

Vertebrae: Axis, atlas, ten other cervicals, and the last three free dorsals.
Pelvis: Almost complete. |

Hind limbs: Right femur complete ; left tibia, proximal end only ; right
tibia, distal end only; right tarsometatarsus complete.

Dimensions of these bones are included in the following description.

Rec. Auck. Inst. Mus. Vol. 4, No. 4, pb. 241-244, 12th February, 1954

242 FALLA.

Capellirallus, new genus.

Generic characters: Bill long (ratio of premaxilla to length of rest
of skull, 2:1) decurved, tapering to blunt, rather flattened tip; possibly
flexible and sensitive in life. Pelvis narrow (width 38% of total length
compared with 45% in Rallus (philippensis)). Tarsometatarsus com-
paratively short, and, as in Gallirallus, lacking prominent outer ‘‘ridge”
of Rallus. Type: Capellirallus karamu n. sp.

Capellirallus karamu n. sp.

Characters as given above for the genus. Available bones have
the following dimensions:

Tarsometatarsus.—Lenegth, 39.0; prox. w., 6.7; distal w., 7.0; med. w., 3.8 mm.

Tibia (parts of 2 bones used). Probable length, 65 (more or less); prox, w.,
10.3; dist. w., 6.0; med. w., 3.5 mm.

Femur.—Lenegth, 44.3; prox. w., 7.6; dist. w., 8.0; med. w., 3.8 mm.

Pelvis.—Length, 38; greatest width, 20; narrowest, 8.6 mm.

Skull—Diameter of occipital foramen, 6; interorbital width, 5.7; total skull
length (tip of bill to occiput), 88; premaxilla, 56 mm.

Holotype: An incomplete skeleton, Auckland Museum (No. 901.1).

Distribution

It seems likely that this rail had a wide distribution, at least
in the North Island. From a large collection of bird bones obtained
in limestone caves at Coonoor, near Dannevirke, in 1914, and sent
to the Dominion Museum, Mr. J. C. Yaldwyn has lately sorted out
a number of bones not referable to any described species. Some of
them are clearly referable to C. karamu. They are in a better state of
preservation than is the type material and consist of:

A complete cranium (C.130).

Right femur (C.132).

Matching pair of tarsometatarsi (C.129).
Right tarsometatarsis (C.129).

Complete pelvis (C.131).

I am also indebted to Mr. Yaldwyn for drawing my attention to
an additional record from a collection of bones made by Mr. A. M.
Hall in a limestone cave at Waitanguru, Waitomo, in 1949. These are
some bones of an individual skeleton (133) consisting of fragments of
pelvis, portions of both femora, pair of tarsometatarsi, and both tibiae.
One of the tibiae is complete. Dimensions of the additional material,
which show little variation from those given for the type, are here
given for comparison.

A. Coonoor

“4

Cranium.—Length, paroccipital to fronto-facial suture, 33; width (parietal),
20.5 mm.

Right femur.—L., 43.5; dist. w., 7.5; prox., 8.5; mid., 3.3 mm.

Tarsometatarsus (2 individuals)—(a) L., 39; prox., 6.5; mid,, 4; dist., 7.5 mm.
(b) L., 37; prox., 6.5; mid., 3.7; dist, 7mm.

Pelvis—Length, 36; greatest width (post-acetabular), 78; least width (mid-
iliac), 9 mm.

A New Rail.

B. Waitanguru
Femur.—Mid., 3.5; dist., 8 mm.
Tibia.—L., 65; prox., 9; mid., 3.2; dist., 6 mm.

Tarsometatarsus.—L., 39; prox., 7; mid., 4; dist., 7.2 mm.

The type material, although in general less well preserved, is more
completely representative of an individual bird, possessing the only
known remains to date of the highly distinctive beak.

Relationships

The absence of any identifiable sternum and of all wing bones
makes it difficult to offer any useful comparison between the rail and
related forms. One respect in which comparison can be made is in
jength of hind limb bones and proportions of hind limb bones one to
another. If the tibia is taken as 100 in all cases, femur and metatarsus

measurements in several species can be expressed as follows:

Femur.
Hypotaenidia philippensis > 71
Hypotaentdia sp. .. ¥, i SO
Nesolimnas dieffenbachi .. a 76
Cabalus silvestris .. vA e- 72
Ocydromus australis Ss = 70
Ocydromus greyi .. .: + 68

Rallus karamu 7) ha na 69

Metatarsus.

63
64
Si
58
56
58
62

ers

ah
. Ls
“held ut bat
a '

ee
ined Pah,
Ae

eee
iy

(ah! 5 oom J
‘
ia
hele 14
I j ‘

eV ALF aie

fa ref
oa! ao
fied PW diey,
A j ¥ : le rT : e0 \ ‘ : , iat ae ; , f ae we

Rpt at Een ‘Fags pai : oes { ¥ ike Sis i ; Neat ook 1A an eva J ‘ i i et: ae : Sia ea) if

F anne a , Pad

eat
— or
aby = jo ae")
fey: Me i
teas

au er
Sits yeu
oe
Pa

Aches ©
-
i
4
‘
‘ii
Pa, On

PLATE 40.

Photograph of type specimen wm sil.

Photo: H. G. Lambert.

Pe ig

:
a
,

7, ee
2
ft 7

PLATE 41.

Capellirallus karamu (type). 1. Cranium (from above). 2. Cranium (side view).
3. Pelvis (side view). 4 Pelvis (from above)

Photos: C., Hale, Dominion Museum,

PLATE 42.

Capellirallus karamu (type). 1. Right tarsometatarsus (from front). 2. Right
femur (from above). 3. Left tibia, prox. end (from behind). 4. Right tibia,
dist. end (from front).

Photos: C. Hale, Dominion Museum.

Ok pet uel

tres AM
aoa
AY bn

A Bird Census and Some Recent
Observations on Birds on Great Island,
Three Kings Group

By E. G. TURBOTT, Auckland Museum, and P. C. BULL,
Animal Ecology Section, D.S.I.R.

Abstract.

Thirteen counts of land-birds on a ten-acre square quadrat on Great Island,
Vhree Kings group, between January 2nd and 8th, 1953, gave an average of 62 birds
per count (58% were bellbirds). Different weather conditions, observer-error,
time of day and bird movements caused the counts to vary between 39 and 90 birds,
the larger counts being characterised by a higher proportion of sight records,
Although bird density varied in different parts of the quadrat, the species retained
their relative order of abundance which was similar to that in other parts of the
island. The breeding population was assessed at 28 pairs (11 species).

Part II contains miscellaneous observations on the breeding and feeding habits
of 22 species of land and sea birds, with special reference to the effect of vegetation
changes which followed the removal of wild goats in 1946, The increased ground
cover provides more quail food, but reduces the area inhabited by pipits and
restricts the extent of gull colonies. Conditions are improved for burrowing petrels,
except where the vegetation is especially dense.

PART I. CENSUS.

A visit was arranged by the Auckland Museum to Great Island, of
the Three Kings group, between 29th December, 1952, and 9th Janu-
ary, 1953. The main purpose of this paper is to record some quantitative
observations made on the land birds during this visit. With the present
data as a basis, it 1s hoped that future observations will determine the
nature and extent of any changes in the abundance of birds. Such
changes are to be expected because the removal of all the goats from
the island in 1946 has already resulted in obvious changes in the vegeta-
tion (Baylis, 1948, 1951; Turbott, 1948; Holdsworth, 1951). For an
account of the history and other features of the island at the time of
the destruction of the goats see papers in Records of the Auckland

Institute and Museum, Vol. 3, Nos. 4 & 5 (1948).

Modification of the vegetation was brought about by early Maori
occupation, followed by the influence of goats. The forest consisted,
in 1946, of a fairly uniform canopy of kanuka (Leptospermuim ert-
coides), and as a result of the activities of goats the open forest floor
was clothed for the most part only in sparse sedges and turf. In some
localities, a few climax forest remnants gave variety to the otherwise
uniform vegetation. Obvious changes since 1946 have been the appear-
ance in many places of an undergrowth of young forest trees (Fig. 1)
and of the vigorous herbaceous plant Colensoa physaloides, and the
development of the ground layer in nearly all parts of the forest into a
dense cover of matted sedges, grasses and herbs up to two feet in

Rec. Auck, Inst. Mus. Vol. 4, No. 4, pp. 245-262, 12th February, 1954

246 TuRBOTT AND BULL.

height (Fig. 2). This ground cover is so dense that it has checked
the re-establishment of seedlings of kanuka, which even in the presence
of goats had been able to grow beneath openings in the canopy (Baylis,

1951),

Unfortunately, census work was not possible in 1946 when the
goats were destroyed nor during the period when the earliest stages of
regeneration were taking place. The discussion of the status of land
birds in relation to the effects of the goat population (Turbott, 1948)
was thus based on rough estimates made by P. C. Bull (November-
December, 1945) and E. G. Turbott (April-May, 1946), respectively.
The bellbird* was the most abundant species, the remainder being
represented only by small populations. ‘The pipit was placed next in
order of abundance, although considerably below the bellbird in num-
bers. Pipits and brown quail were observed feeding on the open forest
floor.

In some parts of the island the new undergrowth of young trees
and shrubs has provided additional cover for forest birds, and the much
denser ground cover may be altering the status of at least two species.
The pipit is now apparently restricted to the more open parts of the
island—rocky outcrops in the forest, grassland and coastal rocks—and
may thus have become less abundant since 1945-46. The brown quail
was apparently not reduced in numbers, but it evidently now had difh-
culty in rising from the ground vegetation when flushed, sometimes
even flying off partly entangled in the long seed heads of sedge (Carex
testacea).

Although these observations suggest that some slight changes had
occurred in the bird community during the period 1946-53, it is believed
that the present study is early enough to provide a useiul reference
point from which to measure the larger changes which may occur in the
future. Various additional observations made on the birds of the
island during the present visit are included in Part II, but for a full
account of the birds reference should be made to the earlier papers hy
Turbott and Buddle (1948) and Turbott (1951).

Methods.

Counts were made of the birds seen while walking over a quadrat
measuring 220 yards square, ie., ten acres in area. The south-east
corner of this quadrat was also the south-east corner of the permanent
Vegetation Quadrat I set up by Turbott in 1946, and full directions for
locating this point have already been published (Turbott, 1948). The
second permanent Vegetation Quadrat (No. II) is located towards the
centre of the census quadrat. Additional topographical detail is ayail-
able on the map of Great Island prepared by Battey (1951), The
larger, new census quadrat slopes towards the south-east and its lower
part includes two small valleys containing remnants of the original
climax forest, but except for these and two rocky outcrops, the area 1s
uniformly covered by kanuka forest (Fig. 2), which becomes increas-
ingly stunted towards the higher north-western portion of the quadrat
where it is more exposed to the wind. Six parallel counting lines, each

* Scientific names of birds are listed in Part ILI.

Three Kings Bird Census. 247

40 yards apart, were marked with light cord before the counts began,
and each of these was divided into upper and lower halves with respect to
the slope of the quadrat. The counts were made by the observer moving
at a slow walking pace along the lines and recording all birds seen or
heard at an estimated distance of up to 20 yards on either side of lines
2 to 6, but on only one side of line 1 (in order to cover the correct area).

In all, 13 counts were made over a period of seven days, and were
carried out at different times of the day and under various weather
conditions. On two counts the two observers walked together, counting
separately, but for the remaining counts each worked alone. Each count

ccupied about two hours, and a separate record was made of the num-
bers of each species seen or heard in each half of the six counting lines:
These data have been summarised in the present paper. Full details of
each count are deposited at the Auckland Museum.

A limited number of traverses and observation periods carried out
over other parts of the island helped to indicate the extent to which the
relative abundance of species on the census quadrat was typical of the
island as a whole. The traverses consisted merely of a record of the
number of individuals of each species seen or heard while walking over
the island in connection with other work. The observation periods were
five-minute intervals, during which the observer tried to see or hear as
Inany species as possible from a fixed point, each species being recorded
only once per period. The frequency with which a species was recorded
in a series of observation periods gives some measure of its abundance
and conspicuousness. It has an advantage over traverses of the above
type in that the observer gives his full attention to watching and listening
to birds, so that the conspicuousness factor is somewhat reduced except
for song. |

The quadrat census, supplemented by figures for relative abundance
(traverses and observation periods) on other parts of the island, has
the advantage of being readily repeated, and a future census could be
carried out on a comparatively brief visit, especially as the quadrat is
situated near the usual landing place on the island. As two of the
permanent vegetation quadrats are located within the census quadrat,
fairly detailed correlation of the status of the land birds with changes
in the vegetation may also be possible on the area.

Results.

The results of the 13 counts are summarised in Table 1, and contain
a total of 805 records or an average of 62 individual birds per count.
Of the 11 species recorded, bellbirds accounted for some 58% of the
total records, followed by parakeets and quail (about 9% each) and
then blackbirds at some 5%. ‘There is considerable variation between
the figures obtained during the different counts and the causes of this
must now be examined. |

Counts 1 and 2 were made with the observers walking together but
counting independently, and the results show fairly close agreement. A
similar result was obtained from counts 9 and 10, which were done in
the same way. Since each pair of these counts was carried out simul-
taneously, the slight differences obtained are the result of observer-error.
Decisions involving whether or not a given bird has been counted already

TuRBOTT AND BULL.

248

‘yerpenb jo yey toddn ul papsodas sparqyjeq “ya Jag : (q)

‘jYySIs AQ popsIOIII SpsIq{[eq *}Usd Jog : (#)

—IjJOoN

rite Mk Se wf See 2 = 2 — pp ‘ds yoes JOF S}UNOD jsatsIy XISs Jo URI
bbe as OG oh eZ. EST IE EZ €T-8 S}uNOD JO [RIO L,
cor 9 6 IE SZ SI tl £Z Le Lp fe (Ez) (09) 192 Q-[ S}UNOd JO [eI],
CO ye Eee a ore 860 gs eee Sp4O991 [¥}0} JO % se satoods Yoey
cos 8 If SI rE 92 IZ OL Se ZZ ve (HZ) (OS) 8% sp1ooo4 [230
Se Se er Pe TE Ae ee OR eS re ead | 9z994q “W'S “OH | 0021-0001 Bese e]
06 — + — @ tT | — 2 € Z£ (EE) (ez) OF | “TO'd | 2290Iq “YS ‘Auuns 3 UTE | OOZT-OOST | £S/T/Z | ZI
eg ae ar ge ee as te a oe Sehr Keane cased 9Zz991q “A'S ive O08T-O09T Be IT
fae oe A I el he ES a RE NE Sa ert | aza9Iq “A'S SU0NS | 0060-0040 | £$/1/9 | OL
cS hy ee TE i TE Re ae ree | azaeiq “YS 80S | 0060-0040 | €S/T/9 | 6
oe OR OR a RE OE SS PORT: h RIREIAD Seed A SES. pe OUZT OUST ec/T/S | 8
ye oe ae Eee ores. WN ae | PUIM SUISII puke IIZZLIq | OO6I-OOZI | ¢S/I/r | Z
pos, = ot SO ae ee sh oe ie re | usem pue wey | 0060°00Z0 | £S/T/r | 9
oo £ £ € * E€ L LTS GS Coy esd Fe | Lot | Surddorp ozseaq “sunt phat | OO8T-009T | ¢S/T/E | s
Sh = eS Ze be Ete | 970214 “N ‘js€91294Q | OOET-OOIT .| €S/T/E |
me eG EO SE oe hee )s Copy Sa ey as oz091q "N ‘3880594 | OOZI-OOOT | ES/T/E |
US, cee SP eh te Re ale CT” ote) RAI wae I azaaiq "N “Is¥2194Q | OO6T-OO9T | ES/T/Z | Z
oS gy a es a es Oe) es OES EG azaa1q “N “3sB194Q | QO6T-OO9T | €S/T/zZ | T
ON “ON ON ‘ON ON ‘ON “ON ON ON ‘ON ‘ON (9) (®) ON | | dibilitas | is ela — | LE
SB a @ £F $F ® te & EB Bw = 6

ae > ee oa By

oe =~ 2 «a & 5 a. 6

wa a o ‘ -

= :

og es ig ES ae

Three Kimgs Bird Census. 249

are probably more important in causing observer-error than are acute-
ness of sight and hearing. The two present observers compared notes
after the counts and also had worked together previously; thus the
results obtained by a completely independent observer might show greater
variation.

The frequent presence of observers in the quadrat might be expected
to disturb the birds and cause some of them to leave. To check this
point, counts 3 and 4 were carried out with one observer walking one
hour behind the other, but in this instance the later observer actually
vecorded more birds than the first one—a result probably attributable
to observer-error. If birds were being disturbed one might expect to
record fewer birds on the later counts. The number of birds recorded
during the first six counts was compared with that during the last six
(Table 1), and it was found that for every species except morepork
owls, the later counts were very much smaller than the earlier ones.
However, this may be largely due to less favourable weather conditions.
(The lower records for moreporks during the earlier counts were due
to the fact that their regular roosting places were not all known at this
time, but once found they were more likely to be recorded on subsequent
counts.) Weather conditions, especially wind, undoubtedly played an
important part in causing variations in the different counts, The quad-
rat is completely sheltered from northerly winds, which prevailed during
counts 1 to 5, which were all fairly large. Count 6, the largest, was
carried out under conditions of calm and warmth. The remaining
counts were low and were carried out in drizzle (7 and &) or with a
south-east wind (9-13)—the quadrat being particularly exposed to this
direction. The total for count 6 (calm) is over twice that for count
10 (strong S.E. wind), although both counts were done by the same
observer at the same time of day. For the remaining counts it is difficult
to separate the effect of weather from that of time of day, but the latter
is generally considered to be important (Colquhoun, 1940a).

However, the present counts probably provide a fair indication of
the amount of variation to be expected in counts of the bird population
which existed on the quadrat at that time; the last four counts did not
increase the variation established by the previous nine.

When birds were heard an attempt was made to sight them if this
could be done fairly quickly and without causing too much disturbance.
Otherwise a single bird was recorded unless there was reason for sus-
pecting more, but counting from sound is difficult in a species such as
the bellbird, in which the song sometimes has a ventriloquial effect. The
two observers varied in the proportion of birds they recorded by’ sight
and sound respectively. For bellbirds (the only species with numbers
allowing individual analysis), 53% of one observer’s total records were
by sight, but for the other this figure was only 42%, and statistical tests
‘adicate that this difference is no greater than could be accounted for
by chance (p = 0.15). The percentage of sight records of bellbirds
on each count varied between 23% and 71% for one observer, and
28% and 60% for the other. When both observers did their counts
at the same time the proportion of bellbirds seen by each was simular.
There is evidence of a direct relationship between the proportion of
sight records and the total number of bellbirds recorded on each count.
For instance, count 6 produced both the largest number of bellbird

250 TURBOTT AND BULL.

records (58) and the highest proportion of sight records (71%), while
count 9 (carried out at the same time of day) produced the hep
number of bellbirds (22) and a low proportion of sight records (32% ).
The results of other counts showed a similar trend. Thus, in Badge
to the observer effect, the census showed that weather and time of day
are important in determining the relative frequency of records by sight
and sound respectively. A high proportion of sight records suggests
a high degree of bird activ ity, and thus conspicuousness, leading to a
high total count. Unfavourable weather conditions such as wind and
rain result in a small total count, less activity, reduced conspicuousness
and a small proportion of sight records. Unfavourable weather con-
ditions, especially wind, reduce vocal activity in blackbirds (Colquhoun,
1939), but the present work suggests that for bellbirds flight activity is
affected first, then song, and call notes last of all.

There is evidence that birds were not evenly distributed over the
census quadrat. When the total number of bellbirds recorded in the
upper halves of the 6 counting lines was compared with that in the
lower ones (Table 1) it was found that the former produced only 24%
of the total bellbird records. For the individual counts this figure
varied between 6% and 33%. If birds were evenly distributed over
the quadrat the numbers found in each half should be approximately
the same, but statistical tests of the present figures show that in 12 of
the 13 counts the differences found were greater than could be accounted
for by chance (p = 0.03). The concentration of bellbirds in the lower
part of the quadrat could be due to its more sheltered position, to the
taller trees or to the more diverse vegetation, and thus food, provided
by the remnants of the original forest.

Some idea of the relative frequency of species on other parts of
the island was obtained from the traverses and observation periods, and

TABLE 2.
Traverses Fach Species as % of:

| T1. | T2.| T3.| T1-3. | Records | Census Records | tion Periods.
Bellbird | 14 | 24 | 38 76 42 Ds ee a ie ht ea
Parakeet Pall) tell wee 4] 22 9 55
Quail ae 4} 10 23 13 9 36
Blackbird | 4 7 11 6 5 23
Kingfisher Li sy 4p ORB 3 2 23
Starling esa: 3 3 9
Chaffinch | <2 | | +) 6 3 3 9
Fantail ) 2 | | 1 | 3 | 2 | 4 5
Pipit Sees 1 4 2 2 5
Thrush PTs) | 2 = Je
Harrier Cea. 1 = fin
Long-tailed cuckoo i: l 1 — s.
Morepork i= | = = | 4 - ai

Banded Rail =f. - l ‘ Ot
- Total ae! 43, } | 40) | 100 | 183 wh 100 | 100 | =

Note——Traverses Nos. 1 and 2 were done by E.G.T. and No. 3 by P.C.B.
Traverse No. 3 occupied over twice the time and distance of the others.

Table 2 compares these results with those obtained from the quadrat
counts. It is evident that the first four species retain their order of
abundance irrespective of the method of assessment. Since a census

Three Kings Bird Census. 251

count is only a very careful traverse count over a known area the two
methods should give the same order of abundance if carried out in the
one locality under similar conditions of season and weather. The
results obtained (Table 2) are similar even though carried out in differ-
ent localities, so it seems reasonable to assume that the land bird com-
munity of the quadrat was typical of the island as a whole, at least as
regards the relative abundance of the various species. Some land birds
present on Great Island were not recorded on the quadrat, but the
missing species are numerically unimportant.

Discussion.

The present work was designed primarily to determine the amount
of variation which might be expected in a series of counts of an approxi-
mately stable bird population. Thus a future observer. using the present
inethods on the same quadrat and at the same time of the year, will have
some basis for judging whether or not his counts reflect a real change
in the size or composition of the bird community. As a secondary con-
sideration it was hoped to obtain a reasonable estimate of the ‘actual
number of individuals of each species living on the ten acres so that
comparisons could be made with work in other areas. To this end it
is now necessary to consider the meaning of the various bird counts
made on the quadrat.

The several counts yielded rather different values, and this is
accounted for by observer-error (double counting and under counting),
and probably by movements of birds out of the area (human disturb-
ance) or into it (e.g., to mob moreporks), and by changes in the con-
spicuousness of the birds themselves (largely influenced by time of day
and weather conditions). Observer-error and movements of birds can
either increase or reduce the count, and may act together or against
each other. Their total effect cannot be accurately assessed from the
data available, and they are perhaps best allowed for by taking the mean
value of all the counts. Changes in conspicuousness, however, can only
make the count too low, and from this point of view the highest count
should be the most accurate. In view of these considerations, and from
the experience gained while working in the quadrat, it was felt that
the mean values for the six highest counts (Table 1) provided the best
indication of the abundance of various species.

The importance of relative conspicuousness in census work on bird
communities has been jstressed by Colquhoun (1940b), and this must
also be considered in the present work. For instance, bellbirds are
highly conspicuous by virtue of their song and noisy flight, while rails are
seldom heard by day and less often seen, since they live under the dense
cover of the tall sedges and rarely fly even if disturbed. Parakeets
with their bright colours in full sunlight, and harsh flight calls, are on
occasions perhaps even more conspicuous than bellbirds, although some-
times silent and easily overlooked. While bellbirds could usually be
seen at distances up to 20 yards on either side of the counting line, quail
would flush only when the observer passed within a few feet of them,
although their call notes sometimes allowed identification at much greater
distances. Thus if it be agreed that the mean values obtained for the
six highest counts provide a reasonable indication of the abundance of

252 TURBOTT AND BULL.

such birds as bellbirds and parakeets, then the comparable values for
banded rail and quail are almost certainly too low. How much too low
ig at present unknown,

The birds counted during the census included both adults and young
birds of the year. It was not possible to count these two groups sepa-
rately because this would have necessitated obtaining close views of
every individual and such an attempt would cause further disturbance
and add greatly to the time occupied by the count. However, for
purposes of comparison with other work it is necessary to convert the
present figures into terms of breeding pairs. The available information
does not allow this to be done accurately, but if an approximation is
to be made it is best done by the observers who did the counts, In doing
so they have been guided by their records of young birds actually seen
and by general impressions obtained while working in the quadrat.
With this in mind, the mean values for the six highest counts (Table 1)
are now discussed species by species.

Bellbird. It is considered that approximately half the birds seen
were young of the year. The mean value of 44 birds thus comes to
represent some 11 pairs of breeding adults.

Parakeet. No young birds were seen on the island, and the only
nest found contained eggs, There was no evidence that any parakeets
lived exclusively on the quadrat, which provided few suitable nesting
places. The birds seen feeding there probably had nests in the cliff
face below. An average of four pairs of parakeets were recorded
feeding on the quadrat, but the numerous birds flying about overhead
made counting difficult, and three pairs is thought to be a more accurate
figure.

Quail. No young were seen but a nest of eggs was found near
the camp, The average figure for the counts suggests that four pairs
lived on the quadrat.

Blackbird. This species was frequently heard but rarely seen.
Comparison may be made with a population of blackbirds studied in
Lower Hutt, where, by December, 50% consisted of young birds of the
year (Bull, 1953). On this basis the Great Island quadrat would have
supported one pair, but from past experience of the retiring habits of
this species when living far from human influence it is thought that
two pairs 1s a more likely figure,

Kingfisher. Two pairs of adults were seen, but as both were on
the boundaries of the quadrat the population is assessed at one pair.

Starling. At the time when the census was made most starlings
were in flocks, including a few young birds of the year. Single birds
and small parties were also frequently seen—perhaps being temporarily
detached from the main flock or not yet having joined it. These odd
birds were the ones recorded during the census counts. The counts
suggest a little less than one pair of adult birds per ten acres.

Chaffinch. Two pairs situated as for kingfishers, so that the area
probably supported only one pair.

Fantail. Several of the birds seen were young of the year, and
one pair is believed to be the breeding population.

Three Kings Bird Census. 253

Pipit. The breeding population is estimated at one pair.

Morepork. ‘Two pairs of adult birds were found—one in each
valley. One pair had flying young.

Banded Rail. At least one pair was present.

The above estimates give a total of 28 pairs of breeding adults
and include 11 different species, Converting this to a 100-acre unit
the figure becomes 280 pairs or 560 individuals. Ina census on Taranga
(Hen Island) Turbott (1940) found a population of 728 breeding birds
(17 species) per 100 acres of forest habitat and 467 birds (8 species )
on a previously-burned area then in mixed thicket and shrubs. The
quadrat on Great Island cannot be taken as typical of the island as a
whole, since it contains a disproportionately large amount of remnant
forest and sheltered valley habitat. The upper half of the quadrat is
more typical, and considering this portion alone the bird density was
only 440 per hundred acres. The relatively low density of birds on
Great Island as compared with Hen Island is probably related to the
modified vegetation, which, lacking variety, may not provide adequate
food throughout the year.

REFERENCES.

BATTEY, M. H., 1951. Notes to Accompany a Topographical Map and a
Provisional Geological Map of Great Island, Three Kings Group,
Rec. Auck. Inst. Mus., 4, 93-97.

BAYLIS, G. T. 5., 1948. Vegetation of Great Island, Three Kings Group,
Rec. Auck, Inst. Mus., 3, 239-252.

BAYLIS, G. T. S., 1951. Incipient Forest Regeneration on Great Island, Three
Kings Group, Rec. Auck. Inst. Mus., 4, 103-109.

BULL, P. C., 1953. Observations on a Marked Population of Blackbirds at
Lower Hutt, Notornis, 5, 149-156. —

COLOQUHOUN, M. K., 1939. The Vocal Activities of Blackbirds at a Winter
Roost, British Birds, 33, 44-47.

COLQUHOUN, M. k., 1940a. The Density of Woodland Birds Determined by
the Sample Count Method, J. Anim. Ecol,, 9, 53-67.

COLQUHOUN, M. &., 1940b. Visual and Auditory Conspicuousness in a Wood-
land Bird Community: A Quantitative Analysis, Proc. Zool. Soc.
Lond., Ser. A, 110, 129-148.

HOLDSWORTH, M., 1951. Effect of Goats on Great Island, Three Kings:
The Permanent Quadrats Resurveyed, Rec. Auck. Inst. Mus., 4,
113-121.

TURBOTT, E. G., 1940. A Bird Census on Taranga (the Hen), Emu, 40, 158-161.

TURBOTT, FE. G., 1948. Effect of Goats on Great Island, Three Kings, with
Descriptions of Vegetation Quadrats, Rec. Auck. Inst. Mus., 3,
253-272.

TURBOTT, E. G., 1951. Notes on the Birds of the Three Kings Islands, Rec.
Auck. Inst. Mus., 4, 141-143.

TURBOTT, E. G., and BUDDLE, G. A., 1948. Birds of the Three Kings Islands,
Rec. Auck. Inst. Mus., 3, 319-336.

254 TurRBotr AND BULL.

PART If. GENERAL OBSERVATIONS.

Diving petrel. Pelecanoides wrinatrix (Gm.). All the burrows
exainined were empty, although a number of dried-up remains of adults
and fully fledged young were found on the steep slopes above South
Kast Bay, and the south-western cove.

The slopes above South East Bay, formerly free of scrub, and
kept relatively bare by goats, now have a dense ground vegetation which
may have an effect upon this and possibly other petrels. Some burrows
are now covered by dense sedges, grasses and herbaceous plants, which
probably make them difficult of access. A twined mass of the trailing
stems of native cucumber (Sicyos angulala) covers much of this area,
and several dead diving petrels, including a fledgling with shreds of
down still attached, were found entangled in the cucumber stems. This
type of vegetation may later be replaced by mixed scrub, which would
bring about a reduction in the ground cover and would be more favour-
able for petrels.

It should be noted that the effect of the goat population is_ still
reflected in the distribution of burrows of petrels of various species.
These birds were formerly unable to breed on most of the inland portions
of the island where the hard soil, resulting from the lack of plant cover
and trampling by goats, was unsuitable for burrowing (Turbott, 1948).
The influence of the later stages of forest regeneration upon the distri-
bution of petrels will thus be of considerable interest.

Fluttering shearwater. Puffinus gavia (Forst.). The young
varied in age in different burrows from downy chicks with the first
feathers showing to fully fledged young with a few wisps of down on
the lower breast. The burrows are thickly scattered over the seaward
slopes wherever the soil is comparatively loose, as well as under low
scrub along the cliff-tops. In some places, e.g., on the saddle between
North West and South East Bays, tunnels were found penetrating dense
sedge as though the birds had been trying to become established there
recently, but unless the soil was loose no proper burrows had been
formed. Burrows were abundant on the slopes above South East Bay,
this species apparently being less affected by the dense ground vegetation
in this area than the diving petrel. It seemed probable, on the basis of
our previous observations, that there had been some increase since
1945-46 in the numbers of burrows of this species. This was observed
mainly on seaward slopes which had originally been much trampled by
goats. Apart from the attempts already mentioned to form burrows
amongst dense sedges on flatter areas adjacent to the seaward slopes,
burrows are situated inland only in a few places in cavities among rocks.
However, a burrow containing a well-grown chick was noted near the
camp in Castaway Valley in a flat situation in fairly soft ground under
the kanuka forest (the same burrow was in occupation during the pre-
ceding visit by an Auckland Museum party, 10th-15th January, 1951).
This burrow ran underground only a few inches below the surface,
beneath a deposit of kanuka twigs and camp debris; it is not known
whether it had been formed in 1945-46, but it is possible that more
burrows will be found in situations of this type now that there is no
further trampling by goats.

Three Kings Bird Census. 255

The first fluttering shearwaters were heard at night shortly before
9 o'clock, but the time of arrival and the numbers heard in the air varied
considerably under different weather conditions. On clear nights with
moonlight the calls began later, and few birds were heard, but there was
much calling on misty or cloudy nights. The chirping note of the chicks
in burrows was heard as early as 8.45 on a cloudy night with some
rain, The adults of the burrow near the camp were ringed*, No.
11302 (sex unknown) was ringed on arrival at the burrow at 10.30 p.m,
on 2nd January, and it returned to the burrow at almost the same time
on the following night (3rd). On 4th January its mate visited the
burrow, also at about 10.30 p.m., and was ringed, No. 11303. On the
last evening (8th) both adults were observed sitting together outside
the burrow, the first to arrive being No. 11302, at 9.50 p.m. When one
of the adults was being handled on arrival, it disgorged a quantity of
small shrimp-lke crustaceans which have been identified by Mr, R. K.
Dell, of Dominion Museum, Wellington, as the common Australian and
New Zealand Euphausiid Nyctiphanes australis G.O. Sars.¢ (See also
identification of the same food organism cast up by young red-billed
ells. )

Grey-faced petrel. Pterodroma macroptera (Smith). This early
breeding petrel was apparently absent during the visit. It was com-
monly observed in April-May, 1946 (see Turbott and Buddle, 1948).

Pterodroma sp. The “ti-ti” or “kek-kek” call characteristic of
certain gadfly petrels (P. cooki, P. inexpectata, P. pycrofti) was heard
at about 9 p.m. above Castaway Valley and from the cliff-tops to the
west of the depot during this visit. The same call was heard by P.C.B.
in November-December, 1945, at about the same point (not to the east
of the depot as stated by Turbott and Buddle, 1948). A search for
burrows was made in likely areas without success, but this as yet uniden-
tified Pterodroma evidently breeds on Great Island in small numbers.

White-faced storm petrel. Pelagodroma marina (Lath.). A
second call was noted several times in the same locality as the above, and
was identified tentatively as that of the present species. Previous evi-
dence suggesting that it may breed on the island was obtained in 1945,
when remains were found in a morepork’s nest.

* For records of ringing on this visit see Bull, P. C., 1953: “Ringing Operations.
Summary for the Year Ended 31 March, 1953.” Notornis, 5, 138-141.
Both of the above P. gavia were ringed on the left leg.

+ We are indebted to Mr. Dell for his identification of the material and for reter-
| ring us to the following literature on N. australis, It was taken, often in
very large numbers, in plankton obtained during July-September, 1911, by
the British Antarctic (“Terra Nova’) Expedition off the Three Kings
Islands (see Tattersall, W. M., 1924: “Euphausiacea,” Brit, Ant. (“Terra
Nova’) Exp., 1910, Nat. Hist. Rep, Zool, VIII; Crustacea, Pt. VIII,
1-36). Together with other species which swarm during their breed-
ing period, it is probably an important source of food for schooling
fish and sea birds, according to Sheard (Sheard, K., 1953; “Taxonomy,
Distribution and Development of the Euphausiacea (Crustacea), B.A.N.Z.
Ant. Res. Exp., 1929-1931, Rep., B, VII, 1-72). N. australis is confined
to neritic waters of Australia and New Zealand of the cooler sub-tropical

zone, Most of the specimens sent to Mr, Dell were adult.

256 TURBOTT AND BULL.

Black shag. Phalacrocorax carbo (L.). A single bird almost
certainly of this species, which has not been recorded previously from
the Three Kings group, was seen at some distance in North West Bay,
where it was fishing close inshore. .

Red-billed gull. Larus novachollandiae Steph, Immediately after
the goats were destroyed the large colonies above South East Bay appear
to have expanded considerably on the upper portion of the slopes (Tur-
bott, 1948: footnote), possibly because disturbance by goats had pre-
viously restricted the colonies to the lower slopes. The colonies in
South East Bay are now reduced to a few isolated groups on bare,
tocky outcrops not far above sea level, and this may be due to the
dense ground vegetation mentioned above.

There is also a large colony, on slopes previously accessible to goats,
tbove the S.W. cove, but in this area the ground vegetation is much less
rank. The slopes were formerly to a large extent bare, and the area
is more exposed to constant strong winds than above South East Bav.
However, a considerable amount of regeneration of ground vegetation
has taken place, especially on the lower portion occupied by the gulls;
this bears lush mats of Disphyma australe, Dichondra repens, Chenopo-
dium triandrum and Salicornia australis, amongst which grow grasses
(Polypogon monspeliensis and Poa anceps), and herbs (Senecio lautus
and Guaphalium luteo-album), There is also a sprinkling of low bushes
of ngaio (Myoporuim laetuin), taupata (Coprosma repens), Hebe insu-
laris, and tall sedge, etc. (Cyperus ustulatus, Arundo kakaho and
Phormium), As the ground vegetation in this area is probably increas-
ing, it will be interesting to compare the extent of the colony as shown
by photographs on later visits, It was also noted that the presence ot
rank vegetation in the colonies may endanger nestlings. Two young
chicks were found beside one nest entangled in the soft, drooping grass
Poa anceps; one chick was already dead and the other was firmly held
by one leg, which had evidently been dislocated by the chick's struggles
to escape.

Further large colonies which were also examined cover the steep
slopes above Tasman Bay, and extend into the low scrub of taupata
(Coprosma repens) at some points on the cliff-tops. The slopes to the
south of the Tasman Stream, which falls into the head of the Bay, are
clothed in dense iceplant (Disphyma australe), but the greater part of
the nesting birds are concentrated on the rocky ledges near sea-level or
on the steeper portions of the cliffs, only a few nesting on the slopes in
hollows in the iceplant.

From observations during the present visit on the above colonies
it thus seems probable that, with the increase of ground vegetation on
the less steep seaward slopes of the island, colonies of this species are
likely to be restricted to cliff ledges and the more rocky portions of the
slopes. It seems probable that there was at first a tendency for the
colonies to expand in certain areas on the seaward slopes after the
destruction of the goats. However, this was only temporary owing to
encroachment by the vigorously regenerating vegetation.

Most of the eggs were just hatching at the end of December, but
there were a few older chicks and fledglings, and at least one almost
ready to fly. The following observations were made on clutch size
and hatching dates in various colonies on different days :—

Three Kings Bird Census. 257

SIV. cove, 30th December.—Of 134 nests with contents 76 had
one egg or chick and 58 had two eggs, two chicks or an egg and a chick.

Of 135 nests, 57 had one egg, 27 had two eggs, nine had one chick,
20 had two chicks and 22 had an egg and a chick. (There were also a
few empty nests and these were not counted. One-egg clutches may
include some in which one egg had hatched and the chick left. Also,
when only one chick was present, another might have been hiding

nearby.) (P.C.B.).

Lasman Bay, 30th December.—Of 119 nests on the north side of
the stream outflow 48 had one egg, 33 had two eggs, 17 had one chick,
11 had two chicks and 10 had one chick and one egg.

Of seven nests to the south of the stream outflow five contained one
egg’, one two eggs and one a chick. (P.C.B.)

fasman Bay, 7th January—Of 101 nests on the lower slopes to
the south of the stream 23 had one egg, 10 had two eggs, 11 had one
chick, 14 had two chicks, two had three chicks. There were six with a
chick and an egg and another with two chicks and one egg; 34 nests were
empty. (Breeding was more advanced here than in the colony above the
5.W. cove, a proportion of the chicks which had left the nest showing
scapulars and wing quills through the down, and several being at an
advanced stage of fledging. The most advanced was fully fledged, with
only wisps of down showing; wings and tail were not yet fully grown,
but this chick was not far from the flying stage.) (E.G.T.)

South East Bay, 3rd January—Of 33 nests, just above sea level,
18 were empty, seven contained one egg, five had two eggs, one had one
young and two had two young.

Of 43 nests at the W. end of the Bay (below Castaway Stream)
13 were empty, eight contained one egg, six had two eggs, five had one
young, 10 had two young and one had one chick and an egg.

(Many of the empty nests had probably been deserted hy the
young soon after hatching.) (P.C.B.)

Ringing was carried out at the Tasman Bay colony on 8th January
by E.G.T. The 49 young birds ringed were fledelings, and_ this
figure was probably nearly the total of young which were of suitable
age in the colony. Two adults were also caught in the colony and
ringed.

An additional 15 adult birds were caught and ringed at the camp in
Castaway Valley where they came for scraps. (See ringing records,
quoted above: Bull, 1953.)

These birds, which came about the camp when scraps were put out
towards the end of the visit, seemed to stay about constantly, and were
thought to be “unemployed” birds. Other such groups of apparently
unoccupied birds were present on the shore, e.g., at the head of North
West Bay.

Red-billed gulls feed in characteristic “swirls” of birds in the
waters surrounding the Three Kings group, generally associating with
fluttering shearwaters (Puffinus gavia). A constant passage of birds

258 TURBOTT AND BULL.

may be observed from the breeding colonies to these feeding groups at
varying distances offshore, and the groups frequently change position
over changing concentrations of food organisms.*

Some of the recently-fed chicks regurgitated pink shrimp-like crus-
taceans identified by Mr, R. K. Dell as Nyctiphanes australis, the
same species as that taken from an adult fluttering shearwater. The
chicks appear to be fed mainly on plankton, and there is further evidence
of this in the mainly reddish, paste-like droppings of both adults and

young.

Notes on gulls feeding on insects over forest on Great Island —
Observations have appeared in a previous paper (Turbott, 1951) on
the capture of cicadas by red-billed gulls, and it was suggested that as
gulls would otherwise only pass occasionally over the forest, this habit
would explain visits by gulls to all parts of the island. This made it
possible to account for the transport of seeds of puka (Meryta sin-
clairit) which had been eaten by the gulls, into various parts of the
forest. On the present visit further observations were made, and
some additions and corrections to the above discussion are now
required. Observation of gulls passing across the island, especially in
the area between Tasman Bay and the S.W cove, indicated that there
was a considerable passage of gulls between these points, which might
have included gulls passing across the island to join feeding “swirls” at
various points offshore, and probably also gulls from the south-western
coast attracted to the fresh-water pool above Tasman Falls, where large
numbers bathed regularly. The gulls frequently followed an erratic
course in order to take advantage of variations in the air currents, and
the routes taken thus varied considerably, so that an individual bird
might pass over a fairly wide area. A similar regular passage of gulls
was observed between North West and South East Bays, the birds
crossing the narrow saddle at a low altitude and gliding down to sea
level.

The present visit (29th December-9th January) was made approxi-
mately a fortnight earlier than the visit in 1951 (10th-15th January),
and, possibly for this reason, few cicadas} were heard at first, although
the numbers later increased considerably. Gulls were again observed
swooping down to capture cicadas above the canopy.

* Sheard in a discussion of the schooling aggregation and behaviour of certain Aus-
tralian fishes based upon stomach contents states that schooling obviously
depended “at first or second remove, on the swarming of crustacean species
for breeding purposes, as well as on swarms of their larvae, either as they
became concentrated in local rips and eddies, or as they were released in
floods during brief, mass hatching periods.” The feeding behaviour of sea
birds is evidently similarly influenced by surface concentrations of the food
organisms, (See Sheard, K,, 1953: B.A.N.Z. Ant. Res, Exp., 1929-1931,
Rep., B, VIII, 1-72.) Mr. R. K. Dell has kindly commented further
(im litt.) on the feeding grounds for sea birds off the Three Kings Islands,
and suggests that, in addition to upwelling on quite a large scale off the
islands, there is probably minor upwelling, due to sudden changes in the
bottom topography. Such areas of minor upwelling, e.g., off promontories,
are known to encourage concentrations of feeding sea birds because the
food organisms are carried to the surface.

+ Melampsalta cingulata.

Three Kings Bird Census. 259

On this visit, observations were also made on the interesting rela-
tionship between gulls, hovering over the canopy to obtain cicadas, and
one or more large flocks of starlings. The latter could be seen moving
fairly rapidly through the open kanuka forest, and at the same time a
flock of 30-40 gulls hovered over the canopy following the course taken
by the starlings. Close observation by E.G.T. of a flock of starlings in
Tasman Valley showed that they were feeding actively at all levels from
the ground layer to the canopy, and it was noted that cicadas were
several times disturbed by the flock. Another member of the party,
Mr, J. 5. Edwards, watched the flock of starlings at close quarters, and
saw the birds feeding on top of the canopy, Mr. Edwards suggested
that the starlings were probably taking large numbers of the abundant
bronze beetle (Eucolaspis brunneus). He saw a cicada caught by'a gull
in mid-air, and in other cases gulls chased cicadas which had been dis-
—turbed but failed to catch them. While the gulls seemed to be interested
chiefly in the relatively large cicadas, the bronze beetle was also possibly
taken.

Also of interest in relation to this habit of the gulls is the observa-
tion that a small black beet common on the heavily flowering kanuka
was eaten in large numbers. This was first suggested by the discovery
of gull castings containing the remains of bees on a rock above Castaway
Valley. Gulls were occasionally seen apparently attempting to capture
the bees above the kanuka canopy, and it was found later that bees were
captured at the burrow entrance on the cliff-tops at the head of North
West Bay, where the bees form a large colony in the bare clay. The
gulls were often seen here running about and catching the bees as they
emerged from the burrows, and fresh castings consisting entirely of
remains of the bees were found nearby.

It seems clear from these observations that the insects mentioned
above form a fairly important supplementary food, and that this habit
serves to draw numbers of gulls to the forest canopy. On the eastern
portion of the island, gulls were also observed in association with
starlings in the neighbourhood of the census quadrat.

It was noted also that the gulls called while hunting in flocks over

the canopy, just as in flocks feeding at sea on plankton, the behaviour
in both situations forming a fairly close parallel.

Banded rail. Rallus philippensis L. Observed on the census area.
One was also seen in Tasman Valley (Edwards).

Brown quail. Synoicus sp. A nest containing six eggs was found
near the depot, Castaway Valley. Specimens were collected under
permit to assist in determining the identity of this form. The follow-
ing analyses of the stomach contents of the four specimens obtained,
and of three obtained in 1946, was kindly carried out for us by Miss
Ruth Mason, Botany Division, D,S.I.R.:

(a) Collected in May, 1946:
(1) seeds of: Carex testacea (sedge).
(2) seeds of: Carex testacea, mostly; Carex virgata, a few.
(3) no identifiable vegetable material; much animal material.

+ Specimens of the bee obtained on the island have not yet been identified.

260 TuURBOTT AND BULL.

(b) Collected in January, 1953:

(1) seeds of: Carex testacca, the most abundant: Solanuin
nigrum (nightshade); Dianella intermedium (blueberry) ;
Haloragis procumbens, | only; Melicytus ramiflorus (mahoe),
2-3; Myoporum laetum (ngaio), 1 only. |

(2) seeds of: Carex testacca, a good deal; Carex lucida, a few:
Solanum nigrum, a good deal; Deyeuvia sp. (grass), a few.

(3) seeds of: Carex testacea; Bromus mollis (grass), a few.

(4) seeds of : Stcyos angulata (native cucumber) ; Solanum nigrum,
a few; Coprosma rhamnoides, 2 only; Melicytus ramiflorus,
a few. Some anitnal matter.

The analyses show that Carex testacca, still the most abundant
element in the ground layer under the forest, continues to provide a
large proportion of the food. Most of the other seeds found in the recent
specimens could have been obtained, although only in a relatively few
places, in 1946. The exceptions, representing additional sources of
food supply since goats were removed, are Dianella intermediui,
Solanum nigrum and Sicyos angulata, all extremely rare in 1946, As
the specimens were obtained in different seasons, this may also be
reflected in the analyses. However, the analyses suggest that the quail
are able to feed on a much greater variety of vegetable food, and it
seems probable that a more abundant total food supply is available than

in 1946,

Harrier. Circus appoximans Peale. A nest above Tasman Bay
was occupied by a young bird which was fully fledged, and flew a short
distance when approached closely. Remains of red-billed gulls and
fluttering shearwaters were found at the nest, but no land birds, It is
possible that the gulls and petrels had been picked up dead. A day
later two adults and a young bird were seen flying about together over
North West Bay. <A pair, or single birds, were also seen on other
occasions, one being observed flying slowly beneath the kanuka forest
canopy in Tasman Valley.

Morepork. Nino+« novaeseelandiae (Gm.). This species was
observed on a number of occasions in different parts of the island, and
several flying young still with a little down were seen.

In the course of field work a number of patches of feathers were
noted on the ground, the total number recorded (P.C.B.) being 15,
consisting of eight parakeets, four quail, two bellbirds and a fantail.
All the birds appeared to have been completely plucked, and there was
nothing left except the feathers. Considering the small area which
could be examined, the total number of birds killed must have been
large; two were found in the census quadrat during the work there.
The birds were apparently killed at night, as several patches of feathers
were found in the morning on tracks passed over on the previous day.
Since there are no mammals on the island, it appeared that these birds
had been killed by moreporks, and it seems that predation on small birds
may be intensified during the period when young are being fed. Remains
of bellbirds were found near the nests of moreporks during the visit in
1945 (Turbott and Buddle, 1948). The abundant lizards, as well as
insects, also provide an important food supply, although they may be
insufficient at this season.

Three Kings Bird Census. 261

Red-fronted parakeet. Cyanoramphus novaezelandiae (Sparrm.).
A nest containing eight eggs was found on 30th December ina hollow,
apparently formed by several old kingfishers’ burrows, in the soft earth
of a bank of the Tasman Stream. The eggs rested on a small amount
of nesting material, mainly parakeet’s feathers and a few twigs. ‘The
height of the nesting chamber was about nine inches, and the situation
was about six feet above the stream bed. Adults which had been incu-
bating, as shown by their frayed tail feathers, were also observed. Two
parakeets observed at close quarters on the census quadrat were seen
chewing off grass seed heads (Danthonia semiannularis) while perching
on a low branch.

Long-tailed cuckoo. Eudynamis taitensis (Sparrm.). A single
bird was seen twice in Tasman Valley.

Kingfisher. Halcyon sancta V. and H. In addition to those
observed on the census area, pairs were noted near the camp and halt
way up the slopes of North West Bay. The latter were seen carrying
food, once a cicada and on a second occasion a lizard. A nest containing
five eggs was found on 30th December in the bank of the Tasman
Stream.

Pipit. Anthus novaeseelandiae (Gm.). A nest with two eggs was
found in the lower Tasman Valley on 7th January. There was also
probably a nest near the depot, where a pipit was seen carrying food.
On the bare cliff-top at the saddle above North West Bay this pair was
observed feeding on smal' bees (see red-billed gull). Each bee was
caught with the tip of the bill, beaten on the ground and then swallowed.
Pipits up to eight in number were observed several times on the grass-
land and prostrate scrub of the south side of Tasman Valley, and else-
where this species was seen mainly on bare patches or rocky outcrops
in the forest, and on coastal rocks (see introduction, Part I).

Fantail. Rhipidura fuliginosa (Sparrm.). Adults and well-grown
young, still keeping more or less to family parties, were seen in Cast-
away Valley, Tasman Valley and in the census area. The distinctive
nature of the call note of the fantail on the island has been noted (Tur-
bott and Buddle, 1948). The song was heard several times on the
census quadrat, and the distinctive pitch and more eccentric rhythm
were noted, as compared with the mainland form.

Bellbird. Anthornis melanura obscura Falla. Many young were
observed, and a female was seen feeding two young birds on the census
area, Although most of the adults were moulting, there was much song.
The bellbird appeared to be feeding largely on insects, and one watched
for some time was industriously searching in the head of a cabbage tree
(Cordyline australis), apparently obtaining small caterpillars concealed
in the leaf bases. A bellbird was seen catching a cicada during one of
the census counts.

Chaffinch. Fringilla coelebs L. The full song was frequently
heard, and a young bird apparently not long out of the nest was regu-
larly heard calling in answer to the parents (see census). One nest
(old) was found.

House sparrow. Passer domesticus (L.). Seen on the south side
of the lower Tasman Valley, but nowhere else.

262 TURBOTT AND BULL.

Song thrush. Turdus ericetorum Turt. Seen once.

Blackbird. Turdus merula L. Blackbirds were frequently heard
singing strongly in the late afternoon. An old nest was found on the
ground at the base of a low kanuka in Tasman Valley, concealed by
overhanging sedges. Droppings found on a rock above Castaway Valley
containing seeds of nightshade (Solanum nigrum) were probably those
of this species; and one was seen holding a struggling cicada. An
albino blackbird was seen in Tasman Valley. |

Starling. Sturnus vulgaris L. The largest flock was estimated
at 200 birds, and there was another flock of perhaps 50 in the Castaway
Valley and quadrat areas (see under red-billed gull above). Single
birds were seen feeding in the kanuka forest from time to time, moving
quietly amongst the branches in contrast with the marked activity of
birds forming the flocks. A young bird of the year in brown plumage
was seen amongst the larger flock, but this consisted mainly of birds in
adult plumage.

ACKNOWLEDGMENT.

The party was transported to Great Island in the yacht ‘“Tara,”’
and thanks are expressed to the yacht’s owner and master, Mr. C. H.
Wild, for his assistance throughout the trip.

PLATE 45.

Fie. 1. Regeneration of mixed coastal forest, of which young trees at present
form an undergrowth. In foreground puka (JVeryta sinclairu )
about 9 feet high. Lower portion of census quadrat, January, 1953.

Fie. 2. Kanuka forest showing dense ground cover, mainly sedges. Upper
portion of census quadrat, January, 1953.
Photos: E, G, Turbott,

=o ae al 7 =
ay Uaee
A Bi eel! ee tees

i> a |
rs eel oe

ean 2

ray

Try
-

tae

~ bi

RECORDS

OF THE

AUCKLAND — INSTITUTE
AND MUSEUM

VoL. 4, No.5
Published by Order of the Council:

A. W. B. Powell, Acting Director

Edited by: A. W. B. Powell

20TH DECEMBER, 1954

CONTENTS

VOL. 4, No. 5

The Occurrence of Babingtonite in Spilite from Three Kings Islands.
By M. H. Battey, Auckland Museum

Melolonthinae (Coleoptera) from the Three Kings Islands.

By B. B. Given, Entomological Research Station, Nelson

The Molluscan Land Operculate Genus Liarea.
By A. W. B. Powell, Auckland Museum

Variation in Hebe (Scrophulariaceae) at Huia and Blockhouse Bay,
New Zealand.

By R. C. Cooper, Auckland Museum

Page

263

295

263

The Occurrence of Babingtonite in Spilite
from Three Kings Islands

By M. H. BATTEY, Auckland Museum.

Abstract

_The paper records the occurrence and optical properties of the rare mineral
babingtonite in spilite. Comparison is made with other parageneses of this mineral.

In his account of spilite and keratophyre from Three Kings Islands,
northern New Zealand, the late Professor J. A. Bartrum described and
figured an unidentified pyroxene-like mineral in the spilite (Bartrum,
1936, p. 420 and Pl. XI, 4). He noted its striking pleochroism, prob-
able positive optical character and large optic axial angle, and inferred
from its occurrence in short veinlets, its xenomorphism, and its rela-
tively coarse crystallization, that it crystallized from a liquid enriched
in volatiles. In a footnote (p.420) he records that Dr. N. L. Bowen,
to whom he showed the slide, suggested that the mineral might belong
to the epidote group. A chemical analysis is given of the rock in
which it occurs.

This mineral now proves to be the rare species babingtonite.
In the course of a more extended study of the lavas of Great Island,
Three Kings Group, Professor Bartrum’s material has been re-exam-
ined and another occurrence of the mineral in spilite has been found.

The original occurrence was in spilitic pillow lava that “outcrops
near sea-level in a band 40 feet thick . . . traceable in the sea-cliffs
for about 200 yards” (Bartrum, 1936, p. 415). The outcrop is about
a quarter of a mile west of the landing-place in North West Bay, Great
Island, and is accessible only from a small boat (Bartrum, personal
communication). In the original slide (No. 559 in the Auckland Uni-
versity College Collection), as recorded by Professor Bartrum, the
mineral occurs in little veinlets and nests, in the form of anhedral
crystals up to 0.25 mm. across, as well as in rare smaller crystals scat-
tered through the rock. In the veinlets it is accompanied by quartz.
In some cases, where veinlets cut the acicular augites characteristic of
the rock, babingtonite replaces the augite prisms (fig. 1).

Amongst Bartrum’s hand specimens are some with veins up to
l-em. wide of white prehnite which enclose, parallel with their walls,
strings of dark granules that prove also to be babingtonite.

Babingtonite: also occurs in spilite outcropping in the prominent
crag on the spur leading to Hapuka Point on Great Island (specimen
6203 in the Auckland Museum Collection). In this rock it forms small
subhedral crystals, embedded in intersertal ferriferous chlorite (bruns-
vigite ).

Rec. Auck. Inst. Mus. Vol. 4, No. 5, pp. 263-266, 20th December, 1954

264 BATTEY.

e 7 . e
chlor ite quartz ~babingtonite
: vein

; . | Pa BOA, 0,
babing tonite 7% LOD, aa 2 a4 |
is > ¥ a f
| a an, \ \s ; * AN
quartz-~ m jj aN & Shy
baking tonite vein ~ babingtonite

9-25 mm.

Fig. 1. Babingtonite in spilite (Slide No. 559, Auckland University College
Collection ).

The following optical properties were determined for the Great
Island habingtonite:
wm X 1.713 = 0.003 X_ bright green.
nY 1.724 = 0003 Y pale purple-brown.
nZ 1.744 = 0.005 Z strong purple-brown.
De Poy gabe,
Dispersion r > v_ distinct.
Extinction angle Z: cleavage (001) in sections normal to Y = 47°.
()ne good cleavage is shown in sections normal to Y.
Comparative optical data are:

Arendal and Arendal'! Baveno! Yakuku? Woburn
Somerville! Mass.?

A et Weg 2 IAL 1.713 Ato 1.720
wz Y 1.730 DY As Lites 1.725 1.731
we Viv od 1.746 1.746 1.740 1.753

2V,, 75° (calc.) 91 aha A
| r > v strong
(1) Washington and Merwin, 1923.
(2) Watanabe, 1922.
(3) Richmond, 1937.

Babingtonite in S pilite. 265

Babingtonite is a mineral conspicuous for the constancy of its
properties. It apparently does not admit extensive atomic substitu-
tions, for a number of good analyses are published, all of which agree
well. One, given by Palache and Fraprie (1902), which deviated
somewhat from others, was later shown by Palache and Gonyer (1932)
to be incorrect. The last-quoted authors assign to the mineral the
formula Caz Fe: Fe Si; O,, (OH). The ratio of ferrous to ferric
iron is near 1:1. Richmond (1937) shows that there are two of these
formula units to the unit cell,

Various views on its relationships have been advanced and in the
most recent study (Richmond, 1937) ‘ts relationship to rhodonite has
been re-emphasized. It remains, however, an isolated mineral species,

Palache and Gonyer (1932) give a list of its parageneses in which
they record four occurrences in veins or cavities in diabase, three in
pegmatitic granites, a contact iron skarn, a granodiorite-slate and
limestone contact and two veins in gneiss. From the present point of
view the occurrences in diabase are of particular interest. In these
cases it is found associated with prehnite, quartz, epidote, albite,
chlorite, zeolites and calcite.

Secondary minerals formed in cavities, and partly from interstitial
glass, in pillow lavas of the Watchung basalts of New Jersey were
described by Fenner (1910). Amongst them was a mineral subse-
quently (Fenner, 1914) identified as babingtonite. Fenner made a
careful analysis of the order of formation of the secondary minerals,
and in his paragenetic scheme babingtonite enters with albite, quartz.
actinolitic amphibole and garnet, immediately after the cessation of
precipitation of labradorite, diopside and magnetite. The low tempera-
ture limit of its stability range is uncertain, since it was not separately
identified in his first paper, but it clearly extends through that of
prehnite, upon which it is sedentary in the New Zealand rock, while it
was presumably a coprecipitate with ferriferous chlorite in the rock
from Hapuka Point crag. This chlorite is uniformly distributed in
intersertal position throughout the rock and there 1s no evidence that
it has formed at the expense of any pre-existing mineral phase.

Since babingtonite 1s a mitieral precipitating in the temperature
stability field of pegmatites, its presence as a disseminated constituent
in some spilites accords with the belief that crystallization of these
rocks continued to temperatures much below those normal to extrusive
rocks. It is not a mineral that would be produced by post-consolidation
metamorphism of the rock, while the texture of specimen 6203 with its
delicate ceryicorn (and quite unaltered) augites is clearly magmatic.
The mineralogy of the rock probably indicates that magmatic crystalliza-
tion continued to the low temperatures at which babingtonite is stable.
Final separation of residual fluids led to the formation of babingtonite-
bearing prehnite veins,

266 BATTEY.

REFERENCES.

BARTRUM, J. A., 1936. Spilitic rocks in New Zealand. Geol. Mag., 73, 414-423.

rENNER, C. N., 1910. The Watchung basalt and the paragenesis of the zeolites
and other secondary minerals. Ann. N.Y. Acad. Sct., 20, 93-187.

—_——————., 1914. Babingtonite from Passaic County, New Jersey. J. Wash.
Acad. Sct., 4, 552-558.

PALACHE, C., and FRAPRIE, C. R., 1902. Babingtonite from Somerville, Mass.
Proc. Am. Acad. Arts & Sci., 38, 383-393.

PALACHE, C., and GONYER, F. A., 1932. On babingtonite. Am. Mineral., 17,
295-303.

RICHMOND, W. E. Jr., 1937. On babingtonite. Am. Alineral., 22, 630-642.

WASHINGTON, H. S., and MERWIN, H. E., 1923. On babingtonite. <A.

Mineral., 8, 215-223.
WATANABE, M., 1922. On the babingtonite from the contact metamorphic

deposits of the Yakuki Mine, Province Iwaki, Japan. Am. Journ.
Sct., 5th ser., 54, 159-164.

26/7

Melolonthinae (Coleoptera)

from the _
Three Kings Islands

By B. B. GIVEN, Entomological Research Station, Nelson.

Introduction

The five specimens of melolonthine coleoptera dealt with in this
paper were collected by Mr. E. G. Turbott and Mr. J. S. Edwards
during May, 1946, and January, 1953, respectively, while visiting the
islands to study regeneration of vegetation following the eradication
of goats which had previously caused much havoc in the area.

In correspondence, Mr. Turbott writes: “. . . I mentioned the
heetle (as an Odontria) in my paper (p. 262) because I thought it
might be characteristic of Great Island during the days of the goat-
induced park-like vegetation (kanuka canopy, with sedges, etc., form-
ing a ground layer). It seemed possible that this situation would
change with regeneration of a mixed forest. . . . I should add, how-
ever, that there were quite a few coming to the light in May, 1946,
which were not captured.”” The conclusion embodied in this statement
by Mr. Turbott is quite likely to be correct, and it will be very inter-
esting to determine whether or not this species of Odontria becomes
less plentiful as forest regeneration proceeds.

It is less likely that the second species (Xylostygnus piceus Broun)
will be affected by the vegetational change.

Collections of insects from the outlying islands about New Zealand
are of great value in assessment of inter-specific relationships. In the
case of the Odontria species herein described, this is amply demon-
strated, and even the Xylostygnus species has characters which, should
further evidence be forthcoming from other islands, may affect the
status of X. brookest. It is therefore of the utmost importance that
our outlying and coastal islands should be carefully studied and collected.

The opportunity of studying this small but interesting collection
has been accepted with gratitude, and it is with pleasure that due
acknowledgment is made to the Director and the Entomologist of the
Auckland Institute and Museum for the loan of the specimens.

Xylostygnus piceus Broun. Figs. 10, 11.
1881—X ylostygnus piceus Broun. Man. N.Z. Coleopt., [V, p. 956.

The single male specimen of this species in the collection was taken
by Mr. J. S. Edwards amongst rocks on the north-west cliffs of Great
Island. Description of the species can be found in Broun (1881), or
Given (1952, p. 100).

Rec. Auck. Inst. Mus. Vol. 4, No. 5, pp. 267-270, 20th December, 1954

(CSIVEN.

268

wee Tbe", Mat maa
Mekaes eda MTT? rp ,
Li

LMipyypues

A;
A

“aun AiCUNininys

> AIM QA

_->

: ©

N
the,
MOUMUU MALIN IAM mn

SAV AAAAAA RRA SAS SEAN ALARA CANVAS

. hehe eas

Three Kings Coleoptera. 269

EXPLANATION OF FIGURES.

igure
1. Odontria carinata n. sp. male antenna,
2. . 1. , female antenna.
3. $4 +4 ,, male fore tibia.
4. “3 ’, . temale fore tibia.
5. ‘ : . IMmale hind tibia, anterior aspect.
6. 54 2 male hind tibia, ventral aspect.
A e ; . temale hind tibia, anterior aspect (spines omitted).
8. - . » male parameres, lateral. |
9. o _ » male parameres, postero-dorsal.
10. Aylostyguus piceus Broun male antenna.
11. PA 53 a parameres and basal shield, lateral.

ec ee ee

The specimen under discussion is similar to the type, except for
the pronotal outline which is similar to that of X. brookesi Broun (see
Given, 1952, plate 21). This is a genus which is apparently confined
to the north-east coast of New Zealand and the nearby coastal islands.
Of material so far collected, specimens from the mainland (near Auck-
land and Whangarei) show strongest relationship to X. brookesi,
diverging in the direction of picews in northern specimens. Specimens
from the islands (Tiritiri, Mokohinau, Three Kings) are of piceus
affinity, diverging towards brookesi in the north (Three Kings). It
is possible that if the genus is found on other islands and further north
on the mainland, the species brookesi may collapse. Recent work at
the British Museum (Natural History) indicates that through the
species Ocnodus unidentatus Lea, the genus Xylostygnus Broun may
ultimately collapse under Ocnodus Burm.

The figures (10, 11) illustrate the antenna and male genitalia of
the Three Kings specimen.

Odontria carinata n.sp. Figs. 1-9.

Male: Colour rich deep red brown with faint alternate interstitial
mottling on elytra, head piceus or dark brown, ventral surface somewhat
lighter and more reddish, legs red brown. Pronotum and elytra, with
a fine moderately dense brassy pilosity, sternal elements with puilosity
rather longer.

Head large, coarsely, uniformly and moderately closely punctate
on clypeus and frons; eyes prominent; antennae (fig. 1) quadrilamellate,
with an internal process on joint 4. Fore tibia (fig. 3) tridentate; hind
tibia highly distinctive, being strongly and somewhat sinuously carinate
ventrally (figs. 5, 6).

Genital parameres (claspers) elongate, slender, and almost bilater-
ally symmetrical (figs. 8, 9).

Length, 17.5 mm.

Female: As male except for antenna (fig. 2) and hind tibia
(fig. 7). The tact that the fore tibia illustrated in fig. 4 1s less acute
apically than that of the male is probably due to wear.

Types

Holotype male, collected by E. G. Turbott, Great Island, at light,
May 5, 1946. <Allotype female, collected by J. S. Edwards, Great
Island, amongst rocks, north-west cliffs, January 5, 1953. Paratype

270 GIVEN.

male, collected by E. G. Turbott, Great Island, May 5, 1946, from
kanuka (Leptospermum ericoides A. Rich.). Paratype female, col-
lected by J. S. Edwards, Great Island, amongst rocks, north-west cliffs,
January 5, 1953.

All type material is in the collection of the Auckland Institute
and Museum.

Remarks

Apart. perhaps, from O. magnum Given, this is the largest known
member of the genus Odontria. In colour and vestiture it is very
similar to O. ranthosticta White, but in genitalia is closer to marmorata
Broun or perhaps striata White or varicolorata Given. In antennal
characters, closest affinity appears to be with sylvatica Broun, velutinum
Given, xanthosticta White, and varicolorata Given. The carinate hind
tibiae are very distinctive, and perhaps unique in the genus.

REFERENCES.

BROUN, T., 1881. Man. N.Z. Coleopt., IV, p. 956.
BROUN, T., 1921. N.Z. Inst. Bull., I, VI, p. 534.
GIVEN, B. B., 1952. D.SJ.R. Bull., 102, pt. 1.
TURBOTT, E. G., 1948. Rec. Auck. Inst. Mus., 3, p. 262.

271

The Molluscan Land Operculate

Genus Liarea

By A. W. B. POWELL, Auckland Museum.

Abstract.

Over 3,000 specimens from 150 localities in the North Island of New Zealand
were used in this taxonomic and distributional survey. There are two groups of
species in Liarea, one with a simple suture and the other with a margined suttre.

A feature of the simple sutured group is the hydrophanous epidermal pattern,
which varies with the age of the individual and is best developed under conditions
of maximum alternation of wetting and drying. Two new species and five new
subspecies are described.

INTRODUCTION.

The genus Liarea is restricted to the North Island of New Zealand
with its maximum development in the Northland Peninsula, only spor-
adic occurrence south of there and complete absence from the Central
Plateau and East Coast south of the Bay of Plenty.

According to Thiele (1929) the genus belongs to the Pupimidae,
which he makes a subfamily of the Cyclophoridae. His associated
genera are all of Indo-Malayan-Australian occurrence and the assump-
tion is that Liarea is a relatively recent acquisition from that region.

There are no known fossil occurrences for Liarea other than those
from post-Pleistocene consolidated dunes near Cape Maria van Diemen,
so there is no positive indication of the length of time these snails have
been in New Zealand. The fact that Liarea is mainly of northern North
Island occurrence may be due to climatic rather than to time considera-
tions, but on the other hand the comparatively large number of not very
clearly differentiated local forms suggests segregational trends during
a late stage of topographical development.

Even at the assumed specific level there are no strongly marked
divergences and there is evidence in the species ornata that two groups,
(A) hochstetteri with a margined suture and (B) egea-turriculata with
a simple suture, have been closely enough allied in the not very distant
past to enable the production of hybrid stock. Even now there are no
very marked differences in the dentition of the several species repre-
sentative of the two groups so far examined. All the group (A) hoch-
stetteri series have a widely expanded labial flange and all but one, ornata,
are without an epidermal pattern. In the group (B) egea-turriculata
series the labial flange ranges from subobsolete to wide, and most of
the species and subspecies develop an elaborate epidermal pattern in
the adult.

The position is complicated by the fact that most of the group (B)
species and subspecies assume several transitional forms in the develop-
ment of the epidermal pattern, due to wear and age. There is an imtial

Rec. Auck. Inst. Mus. Vol. 4, No. 5, pp. 271-293, 20th December, 1954

“vc POWELL.

more or less unicoloured delicately axially costate stage followed by loss
of part, or of all of the ribbing, accompanied by, or followed by a
differential loosening of the adhesion of the epidermis, which results
in varied and often quite complicated patterns. With increasing age
and exposure the epidermis lifts more and more until the appearance
is uniformly pale, a condition found in many empty shells.

The form of the pattern seems to be rather haphazard, having no
significance other than a tendency towards predominence of axial
streaks in turriculata and egea and zigzags, chevrons and tessellations
in fessellata and aupouria.

The epidermal patterns in Liarea seem to be parallelled in the
Philippine genera of the helicoid group centred around Helicostyla and
in cyclophorids from the same region.

Concerning these Philippine shells, Pilsbry (1894), described the
surface as “covered with a thin transparent cuticle, often porous, when
it becomes white and opaque, producing the ‘hydrophanous’ pattern
which ornaments most species.”

Why the epidermal patterns should assume such complicated and
regular designs is not clearly understood, but it may he related to
differential porosity in the formation of the shell.

The group (A) carinella-hochstetteri series and the group (B)
turriculata-egea series frequently occur at the same locality, but they
invariably occupy different ecological stations, the former tending
towards a hydrophile and the latter towards a xerophile.

Group (A) favours the damp gullies and recesses of the forest,
especially amongst rotting masses of nikau palm debris, but Group (B)
occupies the better drained slopes and ridges of the forest floor, and is
found most frequently under fallen leaves and around clumps of Carex.

That the “hydrophanous” pattern develops or is accentuated by
alternations of wetting and drying is suggested by the fact that forms
of group (B) from the perhumid areas such as Waipoua Forest, Broad-
wood, and elevated areas of Great Barrier Island show the absence of or
only a slight tendency towards patterning. That climate has an influence
upon shell size is indicated by the general tendency towards the distri-
bution of small sized species at and south of Auckland and increasing
size north of there, with the maximum sized form in the extreme north.

The cgea series in particular exhibits a spectacular cline or prob-
ably more explicitly a geocline of the usage of Huxley (1939). That
is, quantitative gradation based upon topographic or spatial separation.
In text figure 3 the interruptions to the obvious cline are represented
by two size breaks, one coincident with the long sandy isthmus, probably
long devoid of suitable forest cover, which joins Awanui with the Cape
Maria van Diemen-North Cape block, and the other, now less effective,
coincident with the Auckland Isthmus and north of there.

In this latter instance former extensive volcanism may have caused a
lengthy period of unsuitable conditions, temporarily segregating south-
ern typical egea from its now larger northern counterpart.

The distributional map for the egea series (Text fig. LA) also
suggests that typical egea reached Great Barrier, Little Barrier and

Molluscan Genus Liarea,

WIIAaNIuy> IwaisaisHs0H CE

viwnyso @)

WLAV 1USLLALSHDOH &

IMALIAISHDOH IVALLSLSHDOH @

€]

WINLUNd VIWINDIUMAL QO
adA1LO93 GYOUE 3BOBYT WLYVINCIUY AL &
VNOd IM WLW INIIBUNL Q

va0a vz0a @
viwnassai va0a &
‘yuv. vinnodny (@)

wiunodny vidnosny (D

Distributional patterns in (A) egea, (B) turriculata, and (C) hochstettert

series.

Fig. 1.

274 POWELL.

Chicken Islands by the Coromandel route, not necessarily a continuous
land connection, whilst the assumed Auckland Isthmus volcanic phase
allowed the development in temporary isolation of a larger sized ccunter-
part of egea north of Auckland.

In turriculata there seems to be no genuine subspeciation, but a
tendency towards larger size in localities where optimum conditions of
a more luxuriant leaf mould prevail, i.e., slopes of Manaia, Waro and
Whangarei. West of Whangarei a large more capacious whorled form
occurs in or adjacent to elevated perhumid areas. These may be more
correctly evaluated as ecotypes.

It was noted that the small typical form of turriculata, 7 mm. to
9mm. in height, was invariably’ found where there was little accumula-
tion of leaf mould.

In general, land snails develop more heavily calcified shells wher-
ever there is an abundant source of lime, but in the case of Liarea
turriculata, the distributional area for which is spread over a number
of rock formations, including limestone, there is no apparent difference
in size or weight of shell that could be correlated with the nature of the
soil.

The carinella-hochstetteri series exhibit a clinal pattern also, with
the small sized carinella distributed at and south of Auckland and the
larger hochstetteri to the north of Auckland.

This series does not reach the far northern block and its largest
member is not found in the far north but in and adjacent to the per-
humid areas west and south-west of Whangarei. Ecotypes follow the
pattern of an exaggerated spread of the flanged peristome in perhumid
areas irrespective of geographic considerations.

Garnier (1950, 1951) has given very useful summaries of the
New Zealand climate, but the paucity of stations relevant to Northland
does not permit a close correlation between the snail occurrences and
the climatic factors. From his map of North Island moisture types
(1951, p. 89) two types only are involved in the Liarea distribution,
the perhumid and the humid.

There is an indication, however, that the scale of winter tempera-
tures may be a limiting factor in the distributional patterns. The
summer temperatures for the whole area involved, apart from high
altitude perhumid areas, are more uniform and are less likely to present
a critical distributional factor.

The furthest south occurrence of the genus is in lepida Suter,
considered by its author to be a subspecies of turriculata. Its relation-
ship, however, is nearer to egea, but it is distinctive enough to warrant
specific status.

This species occupies a relatively small distributional area extend-
ing from the vicinity of Masterton to Ormondville and Mauriceville,
and thence westward through the Manawatu Gorge to the Horowhenua
coastal plain.

It is separated from the distributional areas of the northern mein-
bers of the genus by the central plateau and apparently the country
extending westward to the vicinity of the Awakino River, Taranaki.

Moiluscan Genus Liarea. 275

The Taupo pumice showers and other voleanic disturbances asso-
ciated with the Central Plateau may be one of the causes of discontinuity
in the Liarea distributiona! patterns in the southern part of the range.
Climatic influences would count also, since the evidently critical winter
low temperatures associated with the central higher altitude perhumid
areas would be a decidedly limiting factor in respect to the reoccupation
of areas that were formerly devastated by the pumice showers.

The genus Liarea is comparable with certain New Zealand fresh-
water molluscan genera such as Potamopyrgus, Isidora and Hyridella,
all of which are evidently plastic and are undergoing rather rapid
responses to a series of changing environmental influences.

The close similarity in the dentition of the species of Liarea so
far examined shows that the external differences do not as yet reflect
very significant morphological changes. The several forms which I
evaluate subspecifically undoubtedly represent species in the making
that only time will resolve.

Much of our former almost continuous forest cover is now broken
up into innumerable small isolated patches, and so it is probable that
this artificially induced isolating factor may accelerate subspeciation and
speciation in the future.

The gathering of material for this paper has been spread over the
past twenty years, and apart from the results of my own field work I
have had the use of extensive material collected by Mr. Norman
Gardner and the late Mr. A. C. O'Connor.

Although the material examined consists of 150 locality lots and
is represented by over 3,000 specimens, the present survey must be
considered merely a provisional outline.

The topography of Northland is so broken that further localised
subspecies will almost certainly be discovered when a more evenly dis-
tributed series of stations is achieved and in particular when more ot
the high country is investigated.

ACKNOWLEDGMENTS.

The writer is greatly indebted to the late Mr. A. C. O’Connor, to
Mr. Norman Gardner, and to Messrs. R. A. and H. S. Prouse for
transport and for their participation in the field work. Also to Dr. J.
Marwick, of the New Zealand Geological Survey, and later to Mr.
R. K. Dell, of the Dominion Museum, for the loan of the Liarea
material in the Suter collection, and to Mr. G. L. Wilkins, British
Museum (Natural History), for photographs of the holotype of Liarea
turriculata Pfeiffer.

This paper was read in part in a Symposium on “The Species
Concept” at the Eighth New Zealand Science Congress, Auckland,

May, 1954.

276 POWELL.

SYSTEMATIC KEY.

A. SUTURE MARGINED

(a) Labial flange wide
Epidermal pattern present

Epidermal pattern absent

Spire 15-1 2/3 height of aperture

Adult size 6.3 mm.-8.7 mm.
Adult size 7.5 mm.-10.4 mm.
Spire 2-24 height of aperture

Adult size 8.6 mm.-12.9 mm.

i OU dete! or eke

(a) Labial flange subobsolete
Epidermal pattern present
Spire 2 height of aperture
Adult size 9.1 mm.-10.0 mm.

(b) Labial flange narrow
Epidermal pattern present

Spire 14 to 12/3 height of aperture

Adult size 5.8 mm.-8.6 mm.

Epidermal pattern bold axial streaks

Spire 2 to 2} height of aperture
Adult size 7.2 mm.-17.7 mm.

Epidermal pattern bold axial streaks

Spire narrowly tapered

ornata

h. cartnella

h. hochstettert

h. aita

a. fara

e. egea

t. turrriculata

Epidermal pattern zigzags and tessellations

Spire bluntly conical
Adult size 9.8 mm.-13.20 mm.
Epidermal pattern streaks and chevrons

Spire broadly and bluntly conical

Epidermal pattern absent
Spire 2 height of aperture

Adult size 7.6 mm.-10.30 mm.

(c) Labial flange wide
Epidermal pattern present
Spire 2 height of aperture
Adult size 6.3 mm.-7.9 mm.
Epidermal pattern absent

Spire 14 to 11/3 height of aperture

Adult size 7.0 mm.-8.2 mm.

e. tessellata

a. aupourw

t, walpoua

lepida

t. partula

Molluscan Genus Liarea.

LOCALITY KEY.

277

A. Cape Maria van Diemen-North Cape block ad. aupourta
Kerr Point near North Cape only a. tara
ik. Northland Peninsula (south of Awanu1)

Mainly east coast

West coast perhumid areas - A
West and central areas south to Woodcocks
West to east in north mainly east to Auckland
West and south, vicinity of Whangarei
Wellsford-Leigh-Omaha

South of Warkworth

t. turriculata
ft. qwarpoud

c. tessellata
h. hochstetteri
h. alta

ornatla

t. partula

(. East Coast Islands

Chickens and Little Barrier Islands - fs, a ¢. egea

Great Barrier Island yy. +f u yr ¥c e, egea and
t. turriculata

Kawau Island ei on 3 es hs >. h. cartnella

Waiheke Island J: i A , x 4! c. cgea

Db. Auckland Isthmus
Mainly east coast... i bk: Je ns Ay é. egea
Mainly west coast .. 3s 7 - i fo h. carinella

EF, South of Auckland

Mainly western to Awakino ay. ar i h. carineli:
Mainly central and eastern to Te Puke and Rotorua .. ce. egea

4. Wellington
Wairarapa- Manawatu-Horowhenua i ie 14 lepida

See text figure 1 for maps showing detailed plotting of stations !
A. egea, B. turriculata, C. hochstettert series.

A. THE HOCHSTETTERI SERIES.

The distinguishing features of the hocistettert series are the mar-
gined suture, the more or less uniform coloration without “hydrophan-
ous” patterns (except in ornata n. sp.), the strongly carinated body-whorl
and the flat, broadly expanded, concentrically striated flange-like lip,
encircling the inner, narrow, slightly raised peristome.

The known geographic range of the hochstettert series is from
Awanui in Northland to Awakino in Northern Taranaki.

The hochstetteri, alta and carinella subspecies favour the more or
less constantly moist locations in the forest, and are most frequently
found clinging to rotting fallen leaf sheaths of the nikau palm. Since
these snails are thus not subject to extremes of alternate wetting and
drying, as is the case with the turriculata-egea series, which favour drier
situations, this may be the factor that determines the preseice or absence,
or the degree of development, of a hydrophanous pattern.

It is significant that the habitat of ornata, the only member of the
hochstetteri series to develop the hydrophanous pattern, is in relatively
dry, well drained situations, and conversely that the turriculata ecotypes
from perhumid or verging upon perhumid locations show only slight
tendencies towards epidermal patterning.

278 POWELL.

Another variation in the hochstetteri series that may be considered
ecotypic rather than subspecific is in the relative expanse of the outer-
lip flange, which is narrow in populations from the relatively dry coastal
areas ot the vicinity of Mangonui and Bay of Islands and very wide in
the perhumid western areas. The maximum development of this feature
is shown in examples from Herekino Gorge.

The physical function of the expanded outer-lip flange may be to
assist the animal’s ability to cling to the soft, slimy surface of decaying
vegetation by operating as a suction disc, and it may also serve as a
copulatory aid.

The only sinistral example of the genus known to me is a specimen
of hochstetteri typical, in my collection, from Oretere Bush, near Kaeo.

oO - 1:0
> qd vv: a o
a2 wo ~

aus aq 29095 t¢>-cke seuF Sy 5 of

po 0) qczZz wodw w Za
ig rae ae we 7 po iL & Meare
o1,09 C25byryFt5,-ui4zrr3 3 Fu>isa wa e < 0
actor Pex~x~*yeIGyn4 52 f/m | -@ %¢CC O
$250 WZEzowweFTe Gel 2s ales 2 12, _cwW gx%Z03%
C? ae sorte eee TOY IsaS 3 wr,t-c2 w7+ Jox5 tcOcyg
SPER 2wr2co sere reeves si2 FeozeSg 3¢u sistase ris:
efasr aw ss cSICslawoy ee Ger Sebhse ge-
qd ce Ey 2% ES OSES EES u Z2003¢c¢f er <2 CCE KG
<ofnovalisSzbidyvragoshau OlIdaak agi “viZsidzazo3se

Fig. 2. Histogram of populations for Liarea hochstetteri series. Columns represent
an average of at least ten adult specimens for each locality (see left-hand
scale in mm.). Top of column represents height averages and bottom of
column width averages. Stepped line below plots the spire height index
(see right-hand scale in mm.). A = hochstetteri hochstetteri. B = hoch-
stetteri alta. C = ornata. D = hochstettert carinella.

Molluscan Genus Liarea. 279

Liarea hochstetteri hochstetteri (L. Pfeiffer). Pl. 44, fig. 6;
Pl. 45, figs. 9-12.

1861—Realia hochstetieri Pieiffer, Malak. Bl. 8, p. 149.

1865—Realia hochsitcticri: Pieiffer, Monog. Pneumonopomorum Viventium,
2nd suppl., p. 170.

1913—Realia hochstettert: Suter, Manual N.Z. Mollusca, p, 196, Pl. 38, fig. 32.

Shell of moderate size, 7.5mm. to 10.4 mm. in height with broadly
conical spire and a strongly angled to keeled bedy-whorl. Whorls 7
to 74, including a depressed globose protoconch of two smooth whorls.
Spire-whorls, lightly convex, rapidly and regularly increasing. Suture
supra-margined with a narrow rounded sharply raised cord. Spire one
and a-half to one and two-third height of aperture. Spire height index
(1.e., body-whorl width into spire height), 1.12 to 1.56 with an average
of 1.36. Sculpture consisting of closely spaced retractive narrow axial
plications. Aperture ovate-rotund, sligntly oblique, subangled above.
Peristome with a slightly raised inner rim, surrounded by a broadly
expanded concentrically striated thin flange. Perforation open, rather
broadly crescentric, margined by a rounded cord. Colour ranging from
uniformly dark horny to sepia.

_ Type: Bay of Islands (Hochstetter) K. K. Hofmuseum, Vienna.
“Long. 9, diam. 4 mill.”’

The dimensions given by Pfeiffer indicate either an abnormally
narrow shell or more likely a basis of measurement different from the
one adopted in this paper. My method gives the diameter as the
maximum distance from the point of greatest diameter of the body-
whorl on the left to the point of greatest convexity of the outer lip
flange on the right, measured at right angles to the vertical axis of
the shell.

Typical hochstetteri occurs only in Northland, where it occupies a
block extending from coast to coast south of Awanui to Hokianga on
the West Coast and Bay of Islands on the east coast. It extends south-
wards through the inland country between Whangarei and towards the
west coast, where it is represented by the taller-spired hochstctteri alta
n. subsp. (described following) and reaches its apparently southern
limit, again approximating the typical subspecies, in the east coast area
from Wellsford to Waiwera.

The latter present a rather different appearance by having the
plications standing out as white threads on a shining dark-horny to
sepia ground colour.

Further inland collecting stations between Kawakawa and south
of Whangarei are required to determine if there is a continuity between
the northern and southern typical hochstetteri populations, irrespective
of the tall-spired hochstetteri alta, which is sandwiched between, but
more to the westward. Suter recorded hochstetteri from Whangaret
Heads, but I have not been able to substantiate this record in the field.

Locality Smallest* Largest Average Spire Ht.
Index
Kaingaroa 7.10 x 4.80 7.80 x 5.20 Poe x S.0 1.25
Oruru Valley, Mangonui 8.10 x 5.10 9.90 x 5.40 90 x 52 1.53
One mile up Taipa 7.80 x 4,90 8.50 x 5.20 8.1 x 4.9 1.26
Two miles West of Tupou 7.10 x 4.70 8.20 x 5.00 7.7x 48 1,22
Okahumoko Bay 8.25 x 4.80 8.60 x 4.90 8.4 x 4.8 1.43

280 POWELL.

Locality Smallest* Largest Average Spire Ht.

Index
Kaeo 8.20 x 4.80 8.60 x 5.30 8.3 x 5.1 1.28
Pupuke 7.20 x 4.90 7.70 x 4.90 7.4 x 4.9 1.25
Three miles S. of Moerewa — 9.70 x 5.50 — 1.49
Maxwell’s, Omahuta 10.10 x 6.00 10.80 x 5.80 10.4 x 5.7 1.47
Near Awanui 8.50 x 5.70 9.20 x 5.50 8.8 x 5.5 1.12
Larner’s Road, Kaitaia 8.20 x 4.80 9.10 x 5,10 8.6 x 5.0 132
North entr. Herekino Gorge 8.10 x 5.10 9.80 x 5.80 8.9 x 5.4 1.28
Top of Herekino Gorge 9.20 x 5.80 9.90 x 5.50 95 x 5.6 1.25
Lower Mangamuka Gorge 8.60 x 5.30 9.80 x 5.20 OS % 52 1.56
35 m. W. of Mangamuka 9.10 x 5.10 10.00 x 5.50 9.6 x 5.5 1.32
Kohukohu, Hokianga 9.30 x 5.10 9.20 x 5.00 91 x 5.0 1.38
Flokianga, South side 8.60 x 5.50 9.10 x 5.10 8.9 x 53 1.46
Three miles S. of Wellsford 8.50 x 4.90 8.90 x 5.00 8.7 x 4.9 1.42
Omaha Valley 7.60 x 4.30 8.10 x 4.70 7.8 x 4.5 1.39
Four miles North of Puhoi 920 x 5.50 10.60 x 6.00 100 x 5.8 1.42
? mile South of Puhoi 8.80 x 5.00 9.80 x 5.00 93 x 4.9 1.54
Puhoi-Waiwera 969 x 5.20 10.00 x 5.00 9.9 x 5.1 1.50
South side of Waiwera 8.80 x 5.00 10.10 x 5.30 02 x 5.1 1.32
Average of combined averages of above lots: Sie ok yl 1.36

* Smallest sized example with adult labial features.

Dentition: Pl. 48, fig. 39. Kaeo, Northland.

Localities: Kaingaroa, between Awanui and Mangonui (A. E. Brookes, 1917) :
Oruru Val'ey, near Mangonui (A.W.B.P. coll.) ; one mile up south side of Taipa
[stuary (A.W.B.P., 20/1/1950) ; Uruaiti Bush, Northland (N. Gardner, August,
1950) ; two miles west of Tupou Bay, east of Mangonui (Map N. 7, Ref. 144884)
(A.W.B.P., 20/6/1947) ; Okahumoko Bay, Whangaroa (Mrs. I. Worthy, 1948) :
Oreter1 Bush, near Kaeo (Mrs. I. Worthy); Pupuke, near Whangaroa (Mrs. I.
Worthy); Bay of Islands (type) (Hochstetter): three miles south of Moerewa
(N. Gardner) ; Maxwell’s Farm, Omahuta, five miles south of Mangamuka Bridge
(A, Hancox, 1948); near Awanui (W. La Roche); Quarry, Larner’s Road, Kai-
tala (N. Gardner, December, 1947); northern entrance to Herekino Gorge
(A.W.B.P., 31/1/1950) ; Manukau North, near Herekino (N. Gardner, 2/1/1950) :
Top of Herekino Gorge (Mrs. I. Worthy); Mangamuka Gorge, south entrance
(A.W.B.P., January, 1948) ; Mangamuka Bridge to Tutekehua, Lower Mangamuka
Gorge (A. Hancox, 1948) ; 35 miles west of Mangamuka Gorge (Mrs. I. W orthv ) :
Hick’s Bush, near Kohukohu-Broadwood Road Junction (A. E. Brookes) ; Kohu-
kohu, Hokianga; Hokianga River, south side (W. La Roche): six miles south
of Warkworth (K. Hipkins); three miles. south of Wellsford (N. Gardner.
27/3/1948) ; Omaha Valley, near Matakana (N. Gardner, 25/3/1948): four miles
north of Puhoi (N. Gardner, 28/3/1946) ; three-quarter of a mile south of Puhoi
(N. Gardner, 29/3/1948) ; half a mile south of Waiwera (N. Gardner, October,
1947); rorth side of Waiwera-Puhoi hill (Map N. 38, 202990, ca. 300ft.)
(A.W.B.P., April, 1947).

Liarea hochstetteri alta n. subsp. Pl. 45, figs. 13 and 14.

Shell relatively large with narrowly conical spire and weakly carin-
ated body-whorl. Whorls 9, including a depressed globose protoconch
of Z smooth whorls, Spire whorls lightly convex, gradually increasing
to the eighth, the ninth very little wider. Suture supra-margined with
a narrow rounded sharply raised cord, Spire twice to two and a half
times height of aperture. Spire height index (i.e., body-whorl width
into spire height) 1.58 to 2.14 with an average of 1.77. Sculpture con-
sisting of closely-spaced retractive narrow axial plications. Aperture
ovate-rotund, slightly oblijue, subangled above. Peristome with a
slightly raised narrow inner rim, surrounded by a broadly expanded
concentrically striated flange. Perforation narrow, crescentic, margined
by a rounded cord. Colour dark sepia.

Molluscan Genus Liarea. 281

Holotype: Between Tauraroa and Wa‘otira. Auckland Museum. Height
12.75 mm.; diameter 5,60 mm.

Locality Smallest Largest Average Spire Ht.

Index
Onetea, Northern Wairoa 9.00 x 5.00 910 x 4.90 90 x 49 1.58
Houto Mountain (pale) 920 x 5.00 11.20x 510 102 x 5.1 1.67
East slope, Houto Mountain 860 x 4.90 11.20 x 5.20 10.0 x 5.1 1.82
Ruahuia Viaduct 980 x 5.60 11.40 x 5.70 10.7 x 57 1.73
Three miles S. of Parakao 10.90 x 6.00 11.20 x 5.80 111 x 59 1.63
Parakao-Kirikopuni 9.70 x 5.20 11.60 x 5.50 10.6 x 5.3 1.83
Tauraroa-Waiotira 12.3 x 5.60 12.90 x 5.90 12.6 x 5.7 2.14
Average of combined averages of above lots: 10.6 x 5.4 1.77

The subspecies alta is larger than typical hochstettert and has a
relatively mucli taller spire. The peristome is broadly expanded as in
the ecotype from MHerekino and other western perhumid areas.
Examples from Onetea, Northern Wairoa, and the western slopes ot
Houto Mountain are pale to dark horny.

Localities: Between Tauraroa and Waiotira, west side of railway track
(A.W.B.F., 27/3/1949) (Map N. 24, Ret. 725750, 250-300ft.) (Holotype) ; Rua-
huia Viaduct, Parakao-Kirikopuni Road, Mangakahia District (Map N. 19, Ref.
477941, ca. 500ft.) (A.W.B.P., 28/10/1947) ; three miles south of Parakao, Manga-
icahia District (N. Gardner, October, 1947) ; Parakao-Kirikopuni Road, Manga-
icahia District. foot of western slope of Houto Mountain (Map, Dargaville, N. 23.
Ref. 480897, 200-250ft.) (A.W.B.P., 28/10/1947) ; Onetea, Northern Wairoa.

Liarea hochstetteri carinella (L. Pfeiffer). Pl. 44, fig. 7; Pl. 45,
fig. 15.

1861--Realia carinella Pfeiffer, Malak. Bl. 8, p. 150.

1865—-Realia curinella: Pfeiffer, Monog. Pneumonopomorum Viventium, 2nd
suppl. p. 170.

1913—Realia carinella: Suter, Manual N.Z. Mollusca, p. 195, Pl. 38, fig. 30.

Shell small, 6.3mm. to 8.7 mm, in height, with narrowly conical
spire and a very sharply angled and keeled body-whorl. Whorls 7,
including a depressed globose protoconch of 2 smooth whorls. Spire
whorls lightly convex, regularly increasing, the whole outline of the
spire straight, to slightly concave between the protoconch and the second
to third pest-nuclear whorls. Spire one and a-half to one and two-
thirds height of aperture. Spire height index (i.e., body-whorl width
into spire height) 1.16 to 1.42, with an average of 1.29, Suture supra-
margined with a narrow rounded sharply raised cord. Sculpture con-
sisting of closely spaced retractive narrow weak axial threads which
hecome subobsolete over the base. Aperture vertical, ovate-rotund,
subangled above. Peristome with a slightly raised inner rim, surrounded
by a broadly expanded concentrically striated thin flange. Perforation
open, narrowly crescentic, margined by a rounded cord. Colour light
brown, or warm dark-brown.

Type: Drury and Taupiri (Hochstetter ). K. K. Hofmuseum, Vienna. “Long.
7, diam. 34 mill.”

Again the dimensions given by Pfeiffer indicate that the width is
understated. A sight measurement based upon the greatest width, taken
at right angles to the axis of the shell, shows 4.10 mm. for a shell 7 mm.
in height.

Although the extremes appear recognisably distinct, the differences
between hochstetteri. hochstetteri and hochstetteri carinella are not very

282 POWELL.

narked when large series from numerous localities are considered. The
former ranges the larger (8.8mm. x 5.1 mm. average) with capacious
whorls and the latter the smaller (7.1 mm. x 4.2 mm. average) with a
narrower spire and less capacious whorls. Also, the axials tend towards
obsolescence on the base only in carimella, but not invariably so. The
spire height index shows a similar range of variation in both subspecies.

The histogram (Text fig. 2) shows clearly the constantly smaller
size range of the Auckland and south of Auckland carinella subspecies.
See also the map (Text fig. 1C.), which shows the carinelia distributional
area from the northern end of the Waitakere Range to Awakino; all
western drainage except fur an isolated eastern location at Kawau

Island.

Locality Smallest Largest Average Spire Ht.
Index
Waitakere Range 7.00 x 4.70 8.70 x 5.50 8.2 x 5.1 1.30
Muriwai 6.80 x 4,00 7.00 x 4.10 6.9 x 4.1] 1.21
Kawau Island 7.60 x 4,20 7.90 x 4.40 78 x 43 1.42
Mauku 5.90 x 3.70 6.80 x 4.20 6.3 x 3.9 1.26
Four miles South of Waiuku 6.00 x 4.00 6.80 x 4.00 6.4 x 3.9 1.23
Near Tuakau 7.00 x 4.00 7.80 x- 4.20 7.4 x 44 1.32
Port Waikato 6.50 x 4.10 7.90 x 4.20 Zl x 42 1.32
N garuawahia 6.70 x 4.20 6.90 x 4.10 6.8 x 4.1 1.16
Chaupo 7.10 x 4.00 7.50 x 4.20 74x 4.1 1.32
Waitomo Caves 6.10 x 3.80 7.50 x 4.40 7.0 x 4.2 1.29
Awakino Gorge 6.40 x 3.90 7.50 x 4.50 6.9 x 42 1.36
Average of combined averages of above lots: 7.1 x 42 1.29

Localities: Abcve School House Bay, Kawau Island (A.W.B.P., July, 1949) ;
Muriwai, West Coast, Auckland (A.W.B.P.); Swanson, Waitakere Range (H.
Suter coll.) ; Pukematakeo, Waitakere Range (Map N. 41, Ref. 063566, 1104.t.)
(N. Gardner, 15/1/1948) ; Huia, Manukau (N. Gardner, 1947); Titirangi, Auck-
land (N. Gardner, October, 1946); Mt. Wellington lava fields, Auckland (A.
Suter); four miles south of Waiuku (A.W.B.P., 1927); Mauku, Pukekohe (N.
Gardner, October, 1947); Maketu, Hunua Range (H. Suter coll.) ; near Tuakau
(A.W.B.P., March, 1946); Port Waikato (W. La Roche); Hill behind Ngarua-
wahia (N. Gardner, 7/1/1949) ; Ohaupo, Waikato (H. Suter Coll.) ; Mt. Pirongia
(A. E. Brookes); Tarukenga, Rotorua (H. Suter coll.) ; Entrance to Waitomo
Caves (A.W.B.P., February, 1949); Awakino Gorge (A.W.B.P., 1926),

Liarea ornata n.sp. Pl. 48, fig. 37.

Shell small, 8.4 to 8.7mm. in height with conical, straightsided
whorls and subangled to keeled body-whorl. Whorls 63 to 7, including
a depressed globose protoconch of two smooth whorls. Spire-whorls
almost flat in outline, regularly increasing, Suture prominently supra-
margined with a rounded sharply raised cord, Spire one and a-half
times height of aperture. Spire height index (i.e., body-whorl width
into spire height) 1.28 to 1.36 with an average of 1.31. Surface smooth
and polished, with subobsolete closely to irregularly spaced axial growth
lines. Base rounded, smooth, with a deep broadly crescentic umbilicus
margined by a rounded cord. Aperture ovate-rotund, oblique and sub-
angled above. Peristome with a slightly raised narrow inner rim, sur-
rounded by a moderately wide concentrically striated thin flange. Colour
dark reddish-brown with a conspicuous hydrophanous pattern in pale
buff in the form of elaborate chevrons. The base is uniformly dark
brown, the umbilical cavity and outer lip flange pale brown.

Molluscan Genus Liarea. 283

Holotype: Three miles south of Wellsford. Auckland Museum. Height,
8.4mm; diameter 5.0 mm.

Locality Smallest Largest Average Spire Ht.
Index
Three miles S. of Wellsford 7.90 x 4.40 8.40 x 5.00 8.1 x 48 1.28
Omaha Valley 7.40 x 4.40 8.50 x 4.50 7.9 x 44 1.36
Leigh 7.80 x 4.40 8.70 x 5.00 8.3 x 4.7 1.30
Average of combined averages of above lots: 8.1 x 4.6 1.31

This species is the only one of the hochstetteri series to exhibit the
hydrophanous pattern so characteristic of the egea-turriculata series.
That it belongs to the hochstetteri series and not to the latter is clearly
shown by the margined suture, rim-margined umbilicus and expanded
outer lip flange.

__It occupies a compact block of territory extending from the vicinity
of Warkworth to Leigh and Omaha.

Typical hochstetteri occurs in association with ornata at one locality,
Warkworth, but no intergradation is evident.

An original hybrid origin is suspected for ornata, which now seems
to have acquired genetical isolation, and if this is so hybridism must have
occurred in the past when the hochstetteri and the egea-turriculata series
were less strongly differentiated than they are at present. Certainly
under present circumstances elsewhere, in locations where both series
occur more or less together, there is ne sign of hybrid influences.

Localities: Three miles south of Wellsford in reserve near bridge, main high-
way (Map N. 33, Ref. 072233, 200ft.) (N. Gardner and A.W.B.P., January, 1952)
(Holotype); Omaha Valley, Matakana (N. Gardner, 28/3/1948); Leigh (Map
N, 34, Ref. 290322, 30-50ft.) (A.W.B.P., 26/2/1948).

B. THE EGEA-TURRICULATA SERIES.

The distinguishing features of the egea-turriculata series are the
simple unmargined suture, the well developed hydrophanous epidermal
pattern, the rounded body-whorl, subangled at most, the relatively narrow
labial flange and the small to vestigial umbilical cavity.

The known geographical range of the egea-turriculata series is from
the northernmost tip of Northland to the Horowhenua Plain north of
Wellington. South of Auckland the series extends to the Waikato and
eastwards to Rotorua, but is absent from the remaining southern and
eastern areas of the North Island, apart from Jepida which occupies a
compact area from the Wairarapa, the Manawatu and the Horowhenua
Plain.

The species /urriculata is restricted to the Northland Peninsula to
as far north as Kaingaroa, near Awapuni. Its preference is for the
warmer and drier eastern areas, but a presumed subspecies of slightly
broader proportions with an obsolete or less prominently developed
hydrophanous pattern, occupies sporadically, perhumid and verging upon
perhumid locations, mostly in central areas, to the westward or in high
country.

The egea series exhibit a spectacular geocline with an increasing
size range from south to north, as already explained in the introduction,
and is graphically shown in the histogram (Text fig. 3).

284 POWELL.

Liarea egea egea (Gray). Pl. 44, fig. 4; Pl. 47, figs. 27-32.

1850—Realia egea Gray, Proc, Zool, Soc. (Lond.) for 1849, p. 167.
1852—Realia eyea: Pteiffer, Monog. Pneumopomorum Viventium, p. 305.
1913—Realia egea: Suter, Manual N.Z. Mollusca, p. 196, Pl. 38, fig. 31.

Shell small, 5.8 mm. to 8.6 mm. in height with broadly conical spire,
whorl outlines strongly convex with body-whorl rounded to weakly sub-
angled. Whorls 6 to 63, including a small papillate protoconch of two
smooth whorls. Suture simple. Spire one and a-half to one and two-
thirds height of aperture. Spire height index (1.e., body-whorl width
into spire height) 1.13 to 1.58 with an average of 1.355. Umbilicus
small to moderate, crescentic and often margined with a weak cord.
Aperture ovate to oblique D-shaped. Peristome narrow, slightly raised
and surrounded by a moderate to relatively narrow thin flange.
Sculpture in well preserved examples consists of numerous narrow
membranous axials which rapidly wear down and leave faint narrow
axial thread-like folds on an otherwise smooth surface. ‘The initial
colour pattern is a broad basal spiral band on a unttorm ground colour
of light to dark brown. After the wearing down of the axials the
hydrophanous epidermal pattern develops on the spire and 1s usually a
bold design of broad axial streaks, often flexuous but seldom chevroned
to any extent.

I3-

-20

O - ~ Tr eee *, 7 ~10
=e Oo > ns “ 4 =
ava on “xn Sehe Ww rT pose apes <" «
ee 9s gS OS 4 WeWet o Clee BO) pkey 4 “Sy me
aqtatyu = 3x%-"00 zesstte o aztssa_e ¥575_ fur
c= Oo -uzrato SA Ie> a Ret 7Oe rw -aoept*we
Teaatz z = Wey Wo or fe aqa*7x 70a0¢ “”
os- oO f£os xu Wow4On-f£Oyxantaeid -2-
Zazxael tw? SalZete-d¢utoae= Jee owas,
eo oes & VTS TES Uma A sus eS USP 2b arse ct Togo
= Om yet et Se enh ate, a are a er p=
zqhiz fF Ze0540 Tene tbe ze rs FEC ESS ep tuttwao
QJaxuz x XROWAS Uik-msyOLOUFLOSTS Srmuoarrexerra
= ° - bs r - ‘4 _ = a. ~ — y 7, f , } ———
Fig. 3. Histogram of populations tor Liarea egea series. A = egea egea. B
egea tessellata. C = aupouria tara. D = aupouria aupouria. Note the

geocline with size gaps coincident with (a) the far northern sandy
isthmus and (b) the east coast to the north of Auckland.

Molluscan Genus Liarea. 285

Type: Auckland (Greenwood), British Museum (Natural History). “Length
24 lines” (Gray).

Locality Smallest Largest Average Spire Ht.

Index
(Chicken Island 6.75 x 3.80 8.00 x 4.25 74x 4.1 1.58
Little Barrier Island 6.90 x 4.20 7.90 x 4.60 7.3 X 4.3 1.38
Tryphena, Great Barrier 6.20 x 3.90 7.00 x 4.00 6.7 x 3.9 1.42
Three miles S. of Wellsford 7.00 x 4.00 8.60 x 4.20 79x 41 1.40
\Vaiwera 6.60 x 3.90 8.40 x 4.20 7.6 x 3.9 1.58
Greenhithe 6.70 x 4.10 8.00 x 4.30 Fea AZ 1.40
Orakei, Auckland 5.90 x 4.00 6.90 x 4.00 6.3 x 3.9 1.18
Mt. Rd., Mt. Eden, Auck. 6.10 x 3.30 8.00 x 4.40 71x 4.1 127
Oneroa, Waiheke Island 5.80 x 3.40 7.40 x 3.90 6.6 x 3.6 1.56
Cowes Bay, Waiheke 6.70 x 3.70 7.00 x 3.70 ime eA 1.48
Whitford 6.80 x 3.50 7.30 x 4.00 7.2 x 3.8 1.32
Two miles W. of Miranda 6.10 x 3.70 7.00 x 3.80 6.4 x 3.7 1.42
tunua 6.00 x 3.25 7.00 x 3.80 62 22 3:5 1.38
Bombay Hill 6.30 x 3.70 7.90 x 4.00 7.4 x 4.0 1.45
Te Aroha 5.60 x 3.50 6.00 x 3.50 So Ke oo 1.28
McLaren's, Kaimai 5.60 x 3.30 6.40 x 3.50 HZ) xe 36 1.26
Whitianga 5.80 x 3.60 7.10 x 4.10 6.6 x 3.8 1.44
Hongi’s Track 5.70 x 3.30 6.60 x 3.80 6.2 x 3.5 1.42
Roto Ma, Rotorua 5.70 x 3.30 6.60 x 4.00 6.0 x 3.6 1.29
Te Puke 6.20 x 3.50 6.60 x 3.80 6.3 x 3.5 1.26
Mill Bay, Manukau 5.70 x 4.00 6.20 x 4.00 6.0 x 4.0 1.43
Titirangi Beach 7.30 x 4.50 7.90 x 4.10 20 3 42 1.32
Parua Bay, Manukau 6.90 x 3.90 7.40 x 4.10 ol sad Ets
Manukau South Head 6.00 x 3.50 7.30 x 4.00 io Kos 1.33
Waiuku 6.40 x 4.00 710 x 4.20 67x 4.1 1,32
Port Waikato 5.70 x 3.00 6.50 x 4.00 5.9 x 3.6 1.29
Average of combined averages of above lots: 6.4 -& 39 1.35

Dentitien: Pl. 48, fig. 38. Whitford, Auckland.

Localities: Chicken Islands (2nd island from east) (A.W.B.P., “Will Watch”
Exped., February, 1934) ; Little Barrier Island (A. E. Brookes) ; Tryphena, Great
Barrier Island (N. Gardner, January, 1951); Three miles south of Wellsford (N.
Gardner); Waiwera (T. F. Cheeseman); Greenhithe (N. Gardner); Hillyer’s
Creek, Auckland (A.W.B.P., 25/4/1927) ; Orakei Bush, Auckland (A.W.B.P.) ;
Mountain Road, Mt. Eden, Auckland (N. Gardner); Mt. Wellington lava fields,
Auckland (W. La Roche); Between Oneroa and Palm Beach, Waiheke Island
(A.W.B.P., January, 1933) ; Onetangi, Waiheke Island (N. Gardner, 21/2/ 1948) ;
Cowes Bay, Waiheke Island (W. La Roche); Four miles south of Howick, Auck-
land (A.W.B.P., 4/6/1927) ; Whitford (N. Gardner, 5/6/1948) ; Hunua Falls (N.
Gardner, August, 1947); Hunua Range (H. Suter coll.) ; Summit of Bombay Hill
(N. Gardner, 7/1/1949) ; Two miles west of Miranda (Raines Rd.) (N. Gardner,
26/12/1948); Te Aroha (W. H. Webster) ; Whitianga (K. Hipkins, 1948) ;
McLaren’s Falls, Lower Kaimai (N. Gardner, 27/12/1948) ; Upper Kaituna River,
Te Puke (N. Gardner, 27/12/1948) ; Two miles east of Roto Ma Lake, Roterua
(N. Gardner, 5/1/1949) ; Hongi’s Track, Rotorua (N. Gardner, 5/1/1949); Titi-
rangi Beach (N. Gardner, November, 1947): Mill Bay, Manukau Harbour (N.
Gardner, December, 1947); French Bay, Manukau Harbour, in Manuka scrub
(N. Gardner) ; Waikowhai Bush, Manukau Harbour (D. H. Graham) ; Cornwallis,
Manukau (A.W.B.P.); Manukau South Head (W. La Roche); Parua Bay,
Manukau (N. Gardner, December, 1947); Mauku, near Patumahoe (N. Gardner,
October, 1947) ; Waiuku (W. H. Webster) ; Tuakau (H. Suter coll.) ; Port Wai-
kato (W. La Roche); Mt. Kakepuka, near Te Awamutu (N. Gardner, 7/4/1947),

The typical subspecies is found around Auckland, south of there
to the Waikato and eastwards to Rotorua. Northwards it extends spor-
adically to Wellsford and to the islands of Little Barrier, Great Barrier,
and the Chickens. North of Wellsford to Hokianga and Herekino a
larger subspecies (described following) is distributed, and this exhibits
a complex chevroned to tessellated hydrophanous pattern.

286 POWELL.

Liarea egea tessellata n. subsp. Pl. 44, fig. 5; Pl. 47, figs. 24-26.

Shell of moderate size, 8.6mm. to 11.0mm. in height with tall
conical spire. Whorl outlines strongly convex, with body-whorl rounded
to weakly subangled. Whorls 63 to 7, including a small papillate proto-
conch of two smooth whorls. Suture simple, deeply impressed. Spire
twice height of aperture. Spire height index (i.e., body-whorl width
into spire height) 1.40 to 1.64 with an average of 1.54. Umbilicus
moderate, broadly crescentic, margined by a weak cord. Aperture ovate-
rotund, only slightly subangled above. Peristome narrow, slightly
raised and surrounded by a narrow thin flange. Sculpture consisting of
numerous weak axial threads which rapidly wear off, leaving the surface
smooth. Hydrophanous epidermal pattern a striking complicated alter-
nation of dark reddish-brown and buff in the form of axial streaks,
zigzags and chevrons, often resulting in tessellation.. Base uniformly
dark or with a broad dark band on the upper part of the base only.

Holotype: Opononi, Hokianga. Height 8.7 mm.; diameter 4.5mm. Auckland
Museum.

Locality Smallest Largest Average Spire Ht.
Index.
34+ miles W. of Mangamuka 7.90 x 4.20 9.00 x 4.50 8.5 x 4.4 1.57

Top of Herekino 920 x 4.50 11.00 x 5.30 10.1 x 4. 1.54
Cpononi 8.00 x 4.70 9.40 x 4.90 8.7 x 4.6 1.54
Owhatu, Herekino 8.20 x 4.70 10.30 x 4.70 9.5 x 4.6 1.48
Two miles W. Tangowahine 8.60 x 4.20 10.60 x 5.00 9.6 x 4.7

Ruahuia Viaduct 9.50 x 4.80 10.00 x 4.90 98 x 4.9 1.57
14 miles East of Parakao 9.30 x 4.90 10.10 x 5.00 98 x 49 1.64
Three miles S. of Wellsford 7.00 x 4.00 8.60 x 4.20 8.0 x 4.1 1.40
W oodcocks 8.00 x 4.50 10.20 x 4.80 92x 47 1.52
Average of combined averages of above lots: 9.2 x 4.6 1.53

The subspecies tessellata is not only larger than egea typical with
half to one more whorl, but it has a relatively taller spire and almost
invariably a more complicated zigzag to tessellated pattern.

Its range is from Woodcocks to Herekino, Northland, over central
and western areas of the peninsula, and it favours the warmer and drier,
more open outskirts of the forest.

Localities: Top of Herekino Gorge (Mrs. I. Worthy); 345 miles west ot
Mangamuka Gorge, southern entrance (Mrs. I. Worthy); Opononi, Hokianga
(W. La Roche) (type); Two miles west of Tangowahine, Dargaville District
(Map No. 23, Ref. 427786) (A.W.B.P., 28/10//1947) ; Ruahuia Viaduct, Parakao-
Kirikopuni Road, Mangakahia District (Map N. 19, Ref. 477941) (A.W.B.P.,
28/10/1947) ; 14 miles east of Parakao, north side of road, Mangakahia District
(Map N. 19, Ref. 504986) (A.W.B.P., 25/10/1947) ; Three miles south of Wells-
ford (N. Gardner): Between Tauraroa and Waiotira, west side of railway track
(A.W.B.P., 27/3/1949) (Map N. 24, Ref. 725750, 250-300ft.) ; Woodcocks (Map
N. 33, Ref. 093086, ca. 550ft.) (A.W.B.P., 11/2/1948).

Liarea aupouria aupcouria n.sp. Pl. 44, fig. 2; Pl. 48, figs. 33 and
33a.

Shell largest for the genus, 9.80 mm, to 13.20 mm. in height. Tall-
spired but broad in proportion. Whorl outlines moderate convex with
body-whorl rounded and only occasionally weakly subangled. Whorls

4, including a blunt dome-shaped protoconch of two whorls followed
by a half-whorl of closely spaced brephic axial threads. Suture simple,
deeply impressed. Spire twice to two and a-quarter times height of
aperture. Spire height index (1.e., bodywhorl width into spire height)

Molluscan Genus Liarea. 287

1.59 to 1.82 with an average of 1.67. Umbilicus small, crescentic, not
margined. Aperture relatively large, oblique ovate to D-shaped, strongly
subangled above. Peristome narrow, slightly raised and surrounded by
a relatively narrow thin flange. Sculpture of rather closely spaced weak
axial threads which rapidly wear off, leaving a smooth surface. Colour
reddish-brown to dark sepia with a complex hydrophanous epidermal!
pattern in buff to pale yellowish brown. The pattern ranges from simple
broad irregular axial streaks to complicated zigzags and chevrons. The
base 1s mostly light brown with a broad diffused spiral band above, or
the entire base may be dark brown.

Holotype: Unuwhao, 850-900ft. Height 13.20 mm.; diameter 6.0mm. Auck-
land Museum.

Locality Smallest Largest Average Spire Ht.

Index
Unuwhao 12.10 x 5.60 13.20 x 600 128 x 59 1.74
Pandora 10.80 x 5.60 12.80 x 6.00 11.9 x 58 1.59
Kapowairua 10.00 x 5.00 11.10 x 5.00 11,0 x 4.9 1,59
Cape Maria (fossil) 980 x 5.00 12.00 x 5.60 106 x 5.1 1.63
Near North Cape 11.00 x 500 11.70 x 5.10 11.0 x 5.1 1.82
Average of combined averages of above lots: 11.4 x 5.3 1.67

Localities: Cape Maria van Diemen (mainland) consolidated dunes, site of
type locality for Placostylus ambagiosus priscus Powell (A.W.B.P.); Kahuronaki
(Kahuroa on suryey maps), between Te Paki and Kapo Wairua Road, ca. 700-
c00it. (A.W.B.P., February, 1944); S.E. slope of hill behind Pandora, Spirits Bay
(A.W.B.P., February, 1944) ; Hill behind lagoon, Spirits Bay (A.W.B.P., January,
1952) ; Waterfall gully at Kapo Wairua, Spirits Bay (N. Gardner, March, 1949);
Unuwhao, between Spirits Bay and Tom Bowling Bay, 800-900ft. (A.W.B.P.) ;
Coastal cliff, half mile south of North Cape (N. Gardner, January, 1952),

The species is characterised by its large size, broad whorls and
narrow apertural flange. It 1s restricted to the far northern Cape Maria
van Diemen-North Cape block, and apart from a local subspecies,
described following, is the only Liarea found in that area.

Other land snails restricted to this far Northland block, i.e., Pary-
phanta watt Powell, the subfossil Rhytida duplicata Suter, its Recent
descendant duplicata vivens and a number of subspecies of Placostylus
ambagiosus, all point to former insular isolation of this block from the
rest of the Northland Peninsula, to which it is now joined by a long,
sandy isthmus lacking in suitable forest covering.

Liarea aupouria tara n. subsp. PI. 48, fig. 34.

Shell of moderate size, 9.10 mm. to 10.0mm. in height. Tall-
spired but of narrow proportions. Whorl outlines moderately convex,
body-whorl rounded without angulation. Suture simple, deeply
impressed. Whorls 74, including a blunt dome-shaped protoconch of
two whorls. Spire twice to two and a quarter times height of aperture.
Spire height index (i.e., body-whorl width into spire height) 1.68 to
1.89 with an average of 1.77. Umbilicus small, narrowly crescentic.
Aperture ovate-rotund, subangled above. Peristome narrow, slightly
raised, reinforced around the outer-lip section only by a slight thicken-
ing, scarcely a flange. Sculpture in the form of irregular weak axial
threads. Colour pattern dark olive brown with a variable hydrophanous
epidermal pattern in buff to straw colour. The pattern varies from
simple rather diffused axial streaks to intricate zigzags and chevrons,
Base uniformly dark or pale with a broad upper spiral band of dark
brown,

288 POWELL.

Holotype: Kerr Port herbheld, North Cape block, among decaying leaves
under stunted clumps of Hebe speciosa brevifolia Cheesem. on steep northern cliff
face. Height 9.1mm.; diameter 4.0mm. Auckland Museum.

This subspecies was found in abundance during the Auckland
Museum Three Kings Expedition of January, 1953, in Mr. Colin Wild’s
yacht “Tara.”

Although the subspecies bears superficial resemblance to turriculata
it is at once distinguished by the greatly reduced, almost non-existent,
labial flange. Closer inspection reveals the relationship with aupouria,
the only other Liarea from the far northern block, from which it differs
in its much smaller size, more slender proportions and more diffused
epidermal pattern.

Undoubtedly the lecal rigorous conditions of the habitat, which
afford little shade and is subject to periods of extreme dryness in
summer, have induced the development of this distinctive subspecies.

Locality Smallest Largest Average Spire Ht.
Index
Kerr Point 910 x 4.10 10.0 x 430 95 x 42 iz?

Liarea turriculata (L. Pfeiffer). Pl. 44, fig. 1; Pl. 46, figs. 16-23.

{855—Realia turriculata Pfeiffer Proc. Zool. Soc. (Lond.) for 1854, p. 304.
1865—Kealia turriculata. Pfeiffer Monog. Pneumopomorum V iventium, 2nd
Suppl., p. 170.

Shell of moderate size, 7.2mm. to 11.7mm. in height with tall
slender spire and a rounded body-whorl. Whorls moderately convex, 7
to 7%, including a small smooth protoconch of two globose whoris.
Suture impressed, simple. Spire twice to two and a half times height
of aperture. Spire height index (i.e., body-whorl width into spire
height) 1.50 to 2.30 with an average of 1.87. Umbilicus a narrow
crescent without a margining rib. Aperture ovate, subangled above.
Peristome a narrow raised rim surrounded by a moderate to relatively
narrow thin flange. Sculpture in well preserved examples of closely
spaced regular narrow membranous axials often bearing short minute
bristles. Initial coloration yellowish-brown with a broad spiral zone of
.dark brown on the upper part of the base. After the wearing down of
the axials the hydrophanous epidermal pattern develops on the spire.
This is usually in the form of bold, somewhat irregular axial streaks.
The ground colour of the shell deepens to a dark brown and the porous
ov lifted areas of the epidermis form the interstices to the pattern. In
occasional specimens a tendency towards a zigzag pattern forms on the
penultimate and body whorls. In some examples the whole of the base
is dark. The apertural flange is brown to dark-brown and the rim of
the peristome light brown.

Type: New Zealand. British Museum (Natural History). “9 x 32/3 mills.”
Pfeiffer (loc. cit.) later gave the locality for turriculata as Kakepuku (Hoch-
stetter ).

Mr. N, Gardner collected a large number of Liarea from Mt. Kake-
puku, near Te Awamutu, but all are typical egea egea.

A drawing made from a photograph of the type specimen kindly
supplied by Mr. Wilkins of the British Museum (PI. 46, fig. 21) clearly
shows that the name applies to the form of Liarea common around

Molluscan Genus Liarca. 289

—25
-20
l =~
i , ~~ ie
zs §$ > k ot z a = oO
ae if >) Se <x z Zz > uv 4 ~ oO WG Sl
“uf Sze 23 o¢ eh es ie eo Se
a « we a I>o%=-w i WW “ z ‘
£ oe ra 4 =
z wo = = x x= ° -eoao
Tere IHF ea a le a>Ze 72 ie wea $6
xsa2tzFa We yOs XYquaed af, aXe > ZV 59 -
© Ivz t52¢4 nate az7x rd x
tf£s52¢f oad re | ey ws z —_ C ai 2 o>
YegsecOSzszcs2@eSxFarrziigusuarze aS FSo
axa ea eae eS erie te zest eet cae lew 72 Sszx
ruerecOtet2icaqi get i Sisaweaesatse the Cf OE
UaosFRYxX3SEaGaLaxars i siamsre002 0¢CGxr Za xrdé

hig. 4. Histogram of populations for Liarea turriculata and subspecies. A =

turriculata turriculata. B = turriculata waipoua. C = turriculata turri-
cilata (broad ecotype or subspecies?). Aa = small typical ecotype.
Ab = large slender ecotype from broadleaf substrata.

Whangarei and extending up the East Coast of Northland to at least
Whangaroa. Mair Park, Whangarei, is here nominated as the type
locality for turriculata, since Kakepuku is obviously incorrect.

Subspeciation or incipient subspeciation is apparent in the turricu-
lata assemblage in the western high country perhumid areas, in the
vicinity of Houto Mountain west of Whangarei and again south of
Warkworth. The first and last of these merit subspecific nomination,
but the second is not named at this stage since there is insufficient
material from the vicinity to properly evaluate its relationship to the
smaller anda proportionately narrower typical subspecies.

It is also noted that in material from near Auckland there is a
distinct trend towards a less slender shell as the egea egea distributional

area is approached or entered, 1.e., Albany and Kauri Gully, Northcote.

Locality Ssimallest Largest Average Spire Ht.
Index
Church Rd., E. of Awapuni 8.50 x 4.00 9.30 x 4.40 8.6 x 4.2 1.72
Pekerau Valley, Mangonuti 6.80 x 3.60 8.20 x 3.90 F fe as aly PE 1.50
Oruru Bay, Mangonui 8.10 x 3.70 910 x 4.00 8.5 x 3.8 1.75
Whatuwhiwhi 7.80 x 3.70 8.00 x 3.70 7.9 x 3.7 1.75
\Wainui Bay, Whangaroa 7.00 x 3.30 7.90 x 3.90 Fig A 3 1.66

290 POWELL.

Locality Smallest Largest Average Spire Ht.
index
Totara North 7.50 x 3.70 9.00 x 3.90 Sl x 3.7 2,00
‘tauranga Bay, Whangaroa 8.00 x 4.20 9.00 x 4.00 8.6 x 4] 1.89
Kaeo 7.20 x 3.50 9.10 x 4.00 yo ee a 1.75
Kaikohe | 960 x 4.20 11.30 x 4.00 10.6 x 4.1 2.14
Home Po:nt, Bland Bay 7.80 x 3.70 8.50 x 3.90 8.1 x 3.7 2.00
Whangaruru 7.40 x 3.50 8.40 x 3.80 8.0 x 3.7 1.93
Tauranga-Kawau Point 7.10 x 3.40 8.20 x 3.70 the a 1.9]
Helena Bay 7.70 x 3.70 8.20 x 3.70 7.9 x 3.6 1.96
Matapouri Bay 7.60 x 3.20 8.20 x 3.90 8.0 x 3.6 1.90
Kiripaka 8.00 x 3.80 10.30 x 4.10 91 x 39 1,90
Pataua 8.00 x 3.90 8.50 x 4.00 Ro % 39 1.76
Manaia ILO x 320 12.90% 5.30 11.7 x 5.2 2.30
Parua Bay 92ZU x 400 10.10 x 4.30 9.6 x 4.1 1.74
Manganese Point 7.90 x 3.40 8.90 x 3.50 G0) 4c 3:6) 2.10
Waikaraka, Onerahi 8.G0 x 3.80 10.00 x 3.90 9.1 x 3.9 1.96
Waro, Hikurangi 980 x 4.00 10.80 x 480 10.2 x 43 2.30
Mair Park, Whangarei 9.00 x 4.00 10.40 x 4.40 99 x 4.1 2.03
RKuahuia Viaduct 900 x 390 10.50 x 4.50 97 x 42 2.08
Brynderwyn Hill 8.00 x 4.0 8.50 x 4.00 8.2 x 3.9 1.83
12 miles S. of Pukapuka Rd. 8.50 x 4.10 9.40 x 4.10 8.9 x 41 1,72
South side Waiwera Hill 7.00 x 3.60 8.25 x 4.00 7.9 x 3.9 1.80
Hatheld’s Beach 8.20 x 4.10 9.30 x 4.40 8.7 x 42 1.79
Orewa 8.00 x 3.80 10.00 x 4.20 7.8 x 4.0 1,90
White Hillis, Silverdale 8.40 x 4.00 9.50 x 4.25 8.9 x 4.1 1.89
Wade Estuary 8.00 x 3.90 8.50 x 4.00 8.3 x 3.9 1.72
Okura 8.30 x 4.00 9.10 x 4.10 88 x 4.0 1,90
Aibany 8.70 x 4.10 9.10 x 4.20 $.9 x 4.1 1.67
{sreenhithe Road 7.00 x 3.90 8.00 x 3.90 tfc ee 1,83
Kauri Gully, Northcote 7.90 x 4.10 8.40 x 3.90 8.1 x 4.0 1.69
Average of combined averages oi above lois: S7/x 3.9 1.87

Dentition: Pl. 48, fig. 40. Houto Mountain, west of Whangarei.

Localities; Church Road, near Kaingaroa, east of Awanui (A.W.B.P.); head
of Pekerau Valley, near Lake Ohia, Mangonui (A.W.B.P., January, 1948); Oruru
Bay, Rangiawhia Peninsula, Doubtless Bay (A.W.B.P., 21/1/1950) ; Whatuwhiwhi,
Rangiawhia Peninsula (D. Forsyth); Wainui Bay, Whangaroa (R. K. Dell) ;
Totara North (W. La Roche); Tauranga Bay, near Whangaroa (N. Gardner,
28/12/1949); Kaeo (Mrs. I. Worthy); Pupuke, near Whangaroa (Mrs. I.
Worthy) ; Kaikohe (R. Cumber) ; Home Point, Bland Bay (A.W.B.P., 7/2/1948) ;
Whangaruru, northern headland (A.W.B.P., 9/2/1948); Tauranga Kawau Point,
north of Whananaki (N. Gardner, October, 1947, Map N. 16, Ref. 949285) : Helena
Bay (N. Gardner, October, 1949); Matapouri Bay (A.W.B.P.); Kiripaka Reserve,
near Neunguru (A.W.B.P.); Pataua, half mile back from beach, near Whangarei
Tleads (A.W.B.P., January, 1948); Manaia, Whangarei Heads (R. K. Dell);
Manganese Point, Parua Bay, Whangarei (A.W.B.P., 26/10/1947); Waikaraka,
Onerahi-Parua Bay Road, Whangarei (N. Gardner, 26/10/1947, Map N. 20, Ref.
900927) ; Waro, Hikurangi (A.W.B.P., 18/11/1950) ; Mair Park, Whangarei (N.
Gardner, 1947); Houto Mountain, west of Whangarei (EK. Fairburn); East slope
af Houto Mountain, near Houto Post Office, Mangakahia District (A.W.B.P.,
25/10/1947, Map N. 19, Ref. 507916); Maungatapere highway, 1{ miles east of
Parakao, north side of road (A.W.B.P., 25/10/1947, Map. N. 19, Ref. 504986) ;
Three miles south of Parakao (N. Gardner, October, 1947); Five miles west of
Titoki (N. Gardner, October, 1947); Ruahuia Viaduct, Parakao-Kirikopuni Road,
Mangawahia District (A.W.B.P., 28/10/1947, Map N. 19, Ref. 477941); Kaihu
(H. Suter coll.) ; Maungaturoto (A.W.B.P., 28/11/1927); Brynderwyn Hill, near
Kaiwaka (A.W.B.P., 1/4/1947) ; 14 miles south of Pukapuka Road, south of Wark-
worth (N. Gardner, 27/3/1943) ; South side of Waiwera Hill (N. Gardner, Sep-
tember, 1947) ; north end of Hatfield’s Beach, near Waiwera (N. Gardner, October,
1947); Eare’s Bush, Orewa (A. E. Brookes, 1944); White Hills, near Silverdale
(N. Gardner, 11/10/1947) ; Wade (Weite) Estuary, ridge, south side near mouth,
ca. 305ft. (A.W.B.P., 26/12/1949); Okura, near Wade River (A.W.B.P.,
24/1/1949) ; Hobson Road, near Albany (N. Gardner, 11/10/1947) ; half mile along
Greenhithe Road from main highway (N. Gardner, August, 947); Kauri Gully,
Northcote (N. Gardner, October, 1947).

Molluscan Genus Liarea. 291

Liarea turriculata waipoua n. subsp. Pl. 44, fig. 3; Pl. 48, fig. 36.

Shell of moderate size, 7.6 mm. to 10.30 mm. in height, tall-spired
with a proportionately wide body-whorl and a rapidly tapered spire, with
rather straight outlines. Whorls only slightly convex. Whorls 7 to 73,
including a small semi-globose protoconch of two smooth whorls. Suture
impressed, simple. Spire twice height of the aperture. Spire height
index (1.e., body-whorl width into spire height) 1.48 to 1.76 with an
average of 1.66. Umbilicus a narrow crescent without a margining rib.
Aperture ovate-rotund, subangled above. Peristome a narrow, slightly
raised rim surrounded by a moderate to relatively narrow thin flange.
Sculpture of closely spaced, regular, narrow, membranous axials bearing
short minute bristles. Colour uniform brown to dark reddish brown
with a spiral zone of darker brown upon the upper part of the base.
In some examples (Broadwood) there is an indistinct paler zone imme-
diately above the basal band on the body-whorl and also below the suture
on the spire whorls.

No examples so far taken exhibit a hydrophanous pattern, but in
aged examples the whole epidermis lifts and becomes pale yellowish-
brown.

Holotype: Waipoua Forest, Northland (N. Gardner). Auckland Museum.
Height 10.6mm.; diameter 5.1 mm.

Locality Smallest Largest Average Spire Ht.
Index
North side Mangamuka 7.60 x 3.90 9.40 x 4.60 8.3 x 4.1 1.76
Broadwood 10.10 x 4.90 10.30 x 4.90 10.2 x 4.9 1.75
Waipoua Forest 8.70 x 5.00 10.60 x 5.10 96x 5.0 1.48
Average of combined averages of above lots: 92x 4.5 1.66

Localities: Waipoua Forcst, Northland (N. Gardner, 2/1/1950) (type): two
miles north of Broadwood (N. Gardner, 2/1/1950),

Liarea turriculata partula n. subsp. Pl. 48, fig. 35.

Shell small, 7.0 mm. to 8.2 mm. in height with broadly conical spire
of slightly bulging outline Whorls 6 to 64, including a blunt dome-
shaped protoconch of two whorls. Whorls moderately convex, body-
whorl rounded. Suture simple, impressed. Spire one and a-half to
one and one-third times height of aperture. Spire height index (i.e.,
body-whorl width into spire height) 1.17 to 1.41 with an average of
1.30. Umbilicus a relatively large crescent with a weak margining rib.
Aperture oblique ovate-rotund, subangled above. Peristome a narrow,
raised rim surrounded by a wide thin flange. Surface polished, sculp-
tured with distant subobsolete oblique narrow membranous axials.
Colour pale brown to light reddish-brown with a dark reddish-brown
spiral band at the top of the base. Variations range from uniform pale
yellowish-brown to reddish-brown with a dark red-brown base. Hydro-
phanous pattern absent.

Holotype: 14 miles south of Pukapuka Road near main highway south of
Warkworth (N. Gardner, 28/3/1948). Auckland Museum. Height 8.2 mm.;
diameter 4.6 mm.

Locality Smallest Largest Average Spire Ht.
Index
14 miles S. Pukapuka Road 7.00 x 4.00 8.20 x 4.70 7.4 xX 4.3 1.30

292 POWELL.

‘This subspecies has the simple suture and rounded body-whorl of
turriculata coupled with the wide labial flange, open rim-maregined
umbilicus and lack of hydrophanous epidermal pattern, features char-
acteristic of hochstetteri. With it occurs typical turriculata but not
hochstetter!. The subspecies may well have had a hybrid origin, but it
seems now to be stabilized, for there appear to be no intermediate forms
between it and turriculata.

Deciding its taxonomic position in harmony with an adimuttedly
arbitrary nomenclatural system is difficult, and in aligning the subspecies
with turriculata I have presumed a turriculata dominance on the evidence
of the simple suture, rounded body-whorl and lack of strong axial
sculpture.

The subspecies is known only from the type locality, but a large
block of surrounding country remains to be investigated.

Liarea lepida (Suter). Pl. 44, fig. 8.

1904—Realta turriculata lepida Suter, Proc, Malac. Sac. 6, p. 157.
1913—Realia turriculata lepida: Suter Man. N.Z. Mollusca, p. 197.

Shell small, 6.3 mm. to 7.9mm in height with narrowly conical
straight sided spire. Whorl outlines convex, body-whorl rounded.
Suture simple, deeply impressed. Spire a little more than twice height
of aperture. Spire height index (i.e., body-whorl width into spire
height) 1.41 to 1.65, with an average of 1.55. Umbilicus crescentic,
narrow but deep. Aperture ovate-rotund, subangled above. Peristome
a narrow slightly raised rim margined externally but not across the
parietal wall with a relatively wide thin flange. Sculpture of closely
spaced regular very oblique narrow membranous axials, becoming obso-
lete over the body-whorl and absent from the base. Colour pale olive,
with a hydrophanous pattern of irregular pale buff maculations, or simi-
lar patterning in darker-brown. The base is uniformly olive to dark
brown without markings or zones.

Holo:ype: Forty-mile Bush, near Mauriceville (H. Suter), Dominion
Museum, Wellington.

Locality Smallest Largest Average Spire Ht,
Index
Poison Point, Masterton 7.00 x 3.90 7.40 x 3.70 12x 38 1.65
Mauriceville 6.25 x 3.40 7.75 x 4,00 rant Saat 1.54
Hastwell 6.30 x 3.40 6.80 x 4.00 6.5 x 3.6 1.41
Manawatu Gorge 7.80 x 3.90 7.90 x 4.00 7.8 x 3.9 1.62
Florida Road, Levin 7.50 x 4.00 7.60 x 4.00 7.5 x 40 1.44
Average of combined averages of above lots: Lax 38 1.55

Suter made his lepida a subspecies of turriculata, but it is better
evaluated as a distinct species characterized by its many slowly increas-
ing whorls and straight spire outlines. Its distributional area is far
removed from that of the Northland turriculata and it is much more
likely to have had a common ancestry with the egea group. From egea
it is readily distinguished by the same differentiating characters cited
in reference to turriculata,

It is the most southern Liarea known and has a compact area of

distribution ranging from Northern Wairarapa through the Manawatu
Gorge and down the Horowhenua coastal plain.

Molluscan Genus Liarea. 293

Localities: Forty-mile Bush, near Mauriceville (type); Hastwell; Seventy-
mile Bush, near Ormondville (H. Suter); Poison Point, Masterton (Powell coll.
Auckland) ; Manawatu Gorge (A. E. Brookes) ; Florida Road, Levin (N. Gardner,
28/12/1952).

REFERENCES.

GARNIER, B. J., 1950. New Zealand Weather and Climate. Miscellaneous Ser.
No. 1, N.Z. Geographical Soc., pp. 1-154.

GARNIER, B. J., 1951. Thornthwaite’s New System of Climatic Classification
in its Application to New Zealand. .Trans. Roy. Soc. N.Z. 79 (1),
pp. 87-103.

HUXLEY, J. S., 1939. Clines: An Auxiliary Method in Taxonomy. Bijdr.
Dierk. 27.

MORTON, J. E., 1952. A Preliminary Study of the Land Operculate Murdochia
pallidum (Cyclophoridae, Mesogastropoda). Trans. Roy. Soc. N.Z.
80 (1), pp. 69-79.

PILSBRY, H. A., 1894. Manual of Conchology (n.s.) vol. 9, p. 216.

Y fia a
ial = = re
me ry

bri
iS.

Fig.

PLATE 44.

Liarea turriculata turriculata (Pfeiffer) Mair Park, Whangarei
(10.6 mm. x 4.1 mm.).

Liarea aupouria aupouria n. sp. Unuwhao, 850-900 feet, Northland (Holo-
type, 13.2mm. x 6.00 mm.).

Liarea turriculata wa‘poua n. subsp. Waipoua Forest, 8.9mm. x 4.5mm.).
Liarea egea egea (Gray). Orakei Bush, Auckland (6.75 mm. x 4.00 mm.).
Liarea cgea tessellata n, subsp. Opononi, Hokianga (Holotype, 8.7 mm
x 4.5 mm.).

Liarea hochstetteri hochstctteri (Pfeiffer), near Awanui (ecotype with
extra large labial flange. 8.5mm. x 5.5 mm.).

Liarea hochstetteri carinella (Gray). Muriwai, Auckland, West Coast
(7.7mm. x 4.6 mm.). |

Liarea Iepida (Suter). Forty-mile Bush, Mauriceville (Topotype,

7.0mm. x 3.7mm.).
(Figures 1-8 to uniform scale)

PLATE 45.

AQ
A
A\\\

A
(UY

\\\ :

Fig. 9. Liarca hochstettert hochstettert (Pfeiffer). Two miles West of Tupou
3av, Mangonut (7.9mm. x 5.0 mm.). |

Fig. 16. Liarea hochstetteri hochstetteri (Pfeiffer). North side of Waiwera-
Puhoi Hill (10.25 mm. x 5.75 mm.).

Fig. 11. /iarca hochstetteri hochstetteri (Pfeiffer). Maxwell’s Farm, Omahuta
(11.0 mm. x 6.0 mm.), .

Fig. 12. Liarea hochstettert hochstettert (Pfeiffer). Near Awanui (Wide labial
flange ecotype) (9.00 mm. x 5.75 mm.).

Pig. 13. Liarea hochstetteri alta n.subsp. Between Tauraroa and Waiotira
(Holotype, 12.75mm. x 5.60 mm.).

Fig. 14. Liarca hochstetteri alta n. subsp. (pale coloured ecotype). Houto Moun-
tain, west of Whangarei (10.5 mm. x 5.0mm.). .

Fig. 15. Liarea hochstetteri carinella (Pfeiffer). Muriwai, West Coast, Auckland
(7.0mm x 4.0mm.). |

‘ (Figures 9-37 to uniform scale)

PLATE 46.

Liarca turriculata (P-eiffer) (Initial sculpture and pattern). Woaikaraka,
Onerahi.

Liarea turriculata (Pfeiffer) (Worn sculpture plus hydrophanous
pattern). Waikaraka, Onerahi. ;

Liarea turriculata (Pfeiffer). Whangaruru (8.6mm. x 4.0 mm.).
Liarea turriculata (Pfeiffer). Mair Park, Whangarei (11.0mm. x
4.0 mm.).

Liarea turriculata (Pfeiffer), Waro, Hikurangi (10.0mm. x 4.0 mm.).
Liarca turriculata (Pfeiffer). From photograph of holotype.

Liarca turriculata (Pfeiffer). Between Houto Post Office and Titoki
(10.5 mm. x 4.5 mm.).

Liarea turriculata (Pfeiffer) (Broad ecotype or subspecies?) Eastern
slope of Houto Mountain (11.3 mm. x 5.0mm.). |

PLATE 47.

ALN

Fie. 24. Liarea egea tessellata n. subsp. Opononi, Hokianga (Holotype) 8.7 min.
x 4.5 mm.).

Iie. 25. Liarea egcea tessellata n. subsp. Ruahuia Viaduct (10.0 mm. x 4.9 mm.).

lig. 26. Liarea cgea tessellata n. subsp. Woodcocks (9.1 mm. x 4.75 mm.).

Fig. 27. Liarea egea egea (Gray). Te Puke (6.6mm. x 3.8mm.).

Fig. 28. Liarea egea egea (Gray). Mt. Kakepuka (6.35 mm. x 3.60 mm.).

Fig. 29. Liarea egea egea (Gray). Chickens Islands (7.1mm. x 4.1 mm.).

Fig. 30. Lrarca cgea egea (Gray). Greenhithe, auckland (7.6mm. x 4.6 mm.)
(Initial sculpture and pattern).

Mig. 31. Liarea egea egea (Gray). Mountain Road, Mt. Eden, Auckland.
(7.6mm. x 4.9mm.)

Fig. 32. Liarea egea egea (Gray). Orakei Bush, Auckland. (6.9mm. x 4.0 mm.)

Fig.
Fig.

Fig

38.
39,
. AG.

PLATE 48.

ay 4 VAN \
sake ;

Ar'\ \

Liarea aupouria aupouria 0. sp. Unuwhao, Northland. (Holotype,
1320mm x 6.0mm.) 33a. Extreme zigzag pattern.

Liarea aupouria tara n.subsp. Kerr Point, Northland. (Holotype,
9.1mm. x 4.0 mm.)

Liarca turriculata partula n. subsp. Pukapuka Road, south of Warkworth.
(Holotype, 8.2mm. x 46 mm.)

Liarca turriculata waipoua n. subsp. Waipoua Forest, Northland (Holo-

type, 10.6 mm. x 5.1 mm.).
Tiarea ornata n.sp. Three miles south of Wellsford ( Holotype,
84mm. x 5.0 mm.).

DENTITION.

Liarea cyea eyea (Gray). Whittord, Auckland.

Liarea hochstctteri hochstetteri (Pfeiffer). MKaeo, Northland.

Liarea turriculata turriculata (Pfeiffer). Houto Mountain, west of
Whangarei.

295

Variation in Hebe (GSCROPHULARIACEAE)
at Huia and Blockhouse Bay, New Zealand

By k. C. COOPER, Auckland Museum.

Abstract.

Measurements of specimens from two populations of Hebe have been plotted
in pictorialized scatter diagrams. The diagrams support the existing taxonomic
arrangement. The value of such diagrams for the recognition of taxa, for the
iumination of relationships, and for the illustration of variation is stressed.

Last December I wrote a short paper (Cooper, 1954) to draw the
attention of New Zealand botanists to the value of the techniques devised
by Dr. Edgar Anderson (Anderson, 1949) for the study of hybridiza-
tion in wild populations. In this paper I have used the techniques to
illustrate the complex pattern of variation in vegetative and floral char-
acters of Hebe from two localities near Auckland.

Huta is a bay on the Manukau Harbour and is 23 miles west of
Auckland City. The bay is at the southern end of the Waitakere Ranges
and the steep hills about it are clad in coastal scrub and second growth
forest. Specimens of Hebe were gathered on the east and west sides of
the bay, along the road to Whatipu between Little Huia and Mt. Donald
McLean, and along a track from this hill to the Karamatura Stream.
1 am indebted to Mrs. K. Wood for most of the collections from Huia.

Duck Creek is a small stream entering the Manukau Harbour in
Blockhouse Bay, and the stream valley contains remnants of coastal
forest and scrub. The Blockhouse Bay specimens were gathered on
the margins of the forest and in the scrub. Samples of both popula-
tions were collected at random, but specimens were chosen deliberately
from approximately the same position on each plant.

The following characters were measured for the first four dia-
grams:
1. Length of one of the pair of leaves immediately below the
inflorescence.

2. Width of the leaf.

3. Length of the second internode below the inflorescence.

4+. Length of the pedicel of a fully opened flower in the raceme.
5. lL.ength of the fully opened flower.

Internode length and leaf width were chosen as the vertical and
horizontal axis respectively and the measurements of these characters
were plotted as scatter diagrams. Leaf, pedicel and flower lengths
are represented in the diagrams as rays from the dots. The measure-
ments of 35 plants from Huia and of the same number of plants from

Rec. Auck. Inst. Mus. Vol. 4, No. 5, pp. 295-308, 20th December, 1954

296 Cooper.

Blockhouse Bay were grouped in three equal categories to determine
the values to be given for no rays, short rays and long rays.

The first diagram shows the pattern of variation in a collection
from Huia. In the lower left-hand corner of the diagram plants with
small internodes and narrow leaves have short leaves, pedicels and
Howers. In the upper right-hand corner of the diagram a plant with
long internodes and broad leaves has long leaves, pedicels and flowers.
Specimens near these two extremes show to a lesser degree the char-
acters of the extremes.

The second diagram shows the pattern of variation in one of the
collections made at Duck Creek, Blockhouse Bay. Again the characters
are correlated, but most of the specimens are larger than those from
Huia.

In the third diagram the measurements of herbarium specimens
collected in the district between Auckland City and the West Coast are
plotted. In the upper right-hand corner of the diagram the four symbols
with long rays represent herbarium collections of Hebe macrocarpa
(Vahl) Ckn. et Allan, identified as this species by Cheeseman. The
number alongside each symbol refers to the list of species and hybrids
given later. In the lower left-hand corner there are four dots without
rays representing the type collections of H. obtusata (Cheesem.) Ckn.
et Allan, and immediately above them are two dots without rays repre-
senting type collections of Veronica x bishopiana Petrie, a suspected
hybrid between H. obtusata and H. salicifolia. The intermediates repre-
sented by dots with a single short or long ray are H. salicifolia (Forst.
t.) Pennell var. stricta (Hook. f.) Ckn. et Allan and var. longiracemosa
(Ckn.) Ckn, et Allan. The intermediate represented by a symbol with
two long rays in the centre of the diagram is a specimen of H. x macro-
sala Ckn. et Allan, a putative hybrid between H. macrocarpa and
Ff. salicifolia. Only a few specimens of each species and hybrid are
shown on the diagram as the herbarium collections of Hebe from
the Auckland district are small, and frequently the material in a species
folder is so uniform that there can be little doubt that it all came from
a single plant. Other specimens have been added to the herbarium
because they are unusual in one or more characters, and sometimes
these abnormal specimens outnumber the typical specimens.

An analysis of the descriptions in Cheeseman’s Manual of the New
Zealand Flora, ed. 2, 1925, and the published notes of Cockayne and
Allan, indicates that the critical characters used for the separation of
the species are plant habit and size, leaf shape and size, flower colour
and size, and capsule shape and size. On the five vegetative and floral
characters used in these diagrams the Huia collection comprises:

H. obtusata.—Dots without rays in the lower left-hand corner.

Petrie’s x bishopiana—bDots without rays slightly above and to the
left of Hl. obtusata.

H., salicifolia var. stricta—Intermediate symbols with short rays.

H,. x macrosala—Dots with some long rays towards the upper
right-hand corner.
FT, macrocarpa.—Dot with three long rays in the upper right-hand
corner,

Variation in Hebe. 297

The Blockhouse Bay collection comprises :

FA. salicifolia var. stricta—Dots with short rays
HT. x macrosala—Dots with some long rays.

FT. inacrocar pa. —Dots with three long rays.

In using these names 2 am following Cockayne and Allan (1926),
who transferred the wild species of V’ eronica to the genus Hebe and
recognized the following Seaeres and hybrid swarms as native to the
Waitakere Ranges and suburbs of Auckland:

l. H. obtusata (Cheesem.) Ckn. et Allan. This species was
described by Cheeseman from plants collected on the sea cliffs at Kare-
kare and Muriwai.

2. H. salicifolia (Forst. {.) Pennell var. stricta ( Hook, i.) Ckn.

| Allan. The type material cited by J. D. Hooker in Flora Novae

Zelandiae 1: 191, 1853, was collected by Banks and Solander. This

collection was not made in the Auckland district and may not be identi-
cal with Auckland plants.

3. Hf. salicifola (Forst.f.) Penneil var. longiracemosa (Ckn.)
Ckn. et Allan. Cockayne described this variety in Trans. N.Z. Inst. 49:
61, 1917, and gave the distribution of it as Egmont-W anganui hotanical
district. Cockayne and Allan (1926) mentioned that the variety occurs
without evidence of polymorphy throughout that district. Cheeseman
in the Manual, ed. 2, 791, 1925, recorded the variety from the Volcanic
Plateau, East Cape and South Auckland districts, and in his herbarium
there is a specimen which he collectea at the Waitakere Falls.

+, Hf. macrocarpa (Vahl) Ckn. et Allan. Vahl’s paper and the type
are not available. and Cockayne and Allan (1926) considered the species
to be a linneon which required extended study in the field. Hebe macro-
carpa (Vahl) Ckn. et Allan var. latisepala (Kirk) Ckn. et Allan has not
been included in the scatter diagram of herbarium specimens as it has
not been reported from the vicinity of Auckland City or the Waitakere
Ranges,

5. H.x macrosala Ckn. et Allan. (H. macrocarpa x salictfolia).

6. H. x affints (Cheesem.) Ckn. et Allan. The type locality of
Cheeseman's var. affinis was “headlands in the Waitemata and Manu-
kau Harbours.’ Cockayne and Allan (1926) considered the variety to
he part of the hybrid swarm between H. macrocarpa and H. salicifolia,
which they named f7. x macrosala.

7. Petrie (1926) described Veronica bishopiana as a_ hybrid
hetween H. salicifolia and H. obtusata from plants collected on rocky
knobs between Huia Hill and Little Huia. Cockayne and Allan (1926)
could not determine the status of the plant “since his [Petrie’s| cee
tion might well apply to an “invariable” species, and there is only one
specimen in his herbarium.”

Dr. Edgar Anderson in a paper on recombination in species crosses
(Anderson, 1939) pointed out that many generations of deliberate
breeding would be required to break all the linkages between multiple
factor characters and that from this there followed two obvious criteria
of hybridization under natural conditions:

1. The intermediacy of separate characters will be correlated.
Hybrids intermechate in one character will tend to be intermediate in

298 COopPER.

others. Hybrids which are most like either parent in any one character
will tend to resemble that parent in all other characters.

2. Variation between individuals will lessen as parental character
combinations are approached.

On the Huia diagram specimens matching the type collection of
Petrie’s x bishopiana are intermediate between specimens resembling
Cheeseman’s H. obtusa‘a and others which are recognized as H. salici-
fola. Again, in the Blockhouse Bay diagram specimens matching
Cockayne and Allan’s H. x macrosala are intermediate between speci-
mens resembling the reputed parents H. macrocarpa and H. salicifolia.
Hebe macrocarpa flowers mainly in August and H. salicifolia flowers
in June and July, but the flowering times of the two species overlap and
the difference in flowering times is not a barrier to hybridization.

It seems then that the scatter diagrams of internode, leaf and
doral characters support the existing taxonomic arrangement {o some
extent, but the plants obviously need further study. It is remarkable
in view of the number of species recognized previously that none of the
herbarium material matches the specimens in the extreme left-hand
corner of the Huia diagram, These may be depauperated specimens of
HT. obtusata and H. salicifolia var. stricta. ~

Another hypothesis to account for the variation is that the two
extremes, represented by rayless dots on the Huia diagram and long
rayed dots on the Blockhouse Bay diagram, are “species” while all the
intermediate forms are part of a hybrid population between them.
Genetic analysis of the populations would be necessary to provide some
supporting evidence for this suggestion and that study is outside the
scope of this paper, the purpose of which is to stress the potential value
of mass collections and scatter diagrams in formal taxonomy. It is
obvious, however, that the diagrams illustrate the range of variants in
each population remarkably well and would be a useful guide to a
geneticist in planning the analysis of the populations.

It may be doubted whether a collection of 35 plants is an adequate
sample of a population. Two subsequent collections from the Block-
house Bay area show, however, a similar pattern of variation to that of
the first collection. It may also be doubted whether the variation in
the Huia and Blockhouse Bay collections should be explained on
genetical grounds. The genus Hebe is notoriously plastic and the varia-
tion may be the result of ecological factors. In making the collections,
however, small areas were chosen which appeared to be uniform in soil
and climatic conditions. On an exposed clay bank at Blockhouse Bay,
eleven flowering plants were found which were obviously dwarfed,
being 30 cms. or less in height. The measurements of five characters
of these plants are plotted on daigram 4. The dots are all at the extreme
lower left-hand corner of the diagram, as the plants have very short
internodes and narrow leaves, but the pedicels and flowers of ten of the
eleven specimens are represented by rays. Probably the flowers are
less plastic than the stems and leaves and indicate that the plants are
dwarfed members of the intermediate group.

As a check on the diagrams further collections were made from
fruiting plants and the following characters of each specimen were
measured :

1. Length of one of the pair of leaves immediately below the

inflorescence, as before.

NO
er,
\ ~

Variation in Hebe.

Width of same.
Length of the second internode, as before.
Width of a ripe capsule.

um kw N

Length of same.

In diagrams numbered 5, 6, 7 and 8, the measurements of. these
additional collections from Huia, Blockhouse Bay, the Cheeseman
Herbarium, and the clay bank near Duck Creek, have been plotted.
The arrangement of the symbols in the diagrams of fruiting specimens
is very similar to the pattern of variation illustrated in the first four
diagrams of flowering plants.

General: From pictorialized scatter diagrams such as those of the
Huia and Blockhouse Bay populations of Hebe, information can be
obtained regarding:

1. The grouping of characters ;
2. The relationships of taxa; and
3. The variation within taxa.

1. Character groupings: Robson (1928) remarked “ill defined as
they may be and of varying dimensions, a certain tendency to character
eroupings of a certain stability is fairly recognizable [in biological
material]. The designation of such groupings as “‘species” or “variety”
presents difficulty, however. The system is arbitrary, but only in this
respect—the character groupings themselves have reality. The discern-
ment of morphological similarities and differences is intuitive through
contemplation of the form of plant structures (cf. Agnes Arber, pp.
121-126, 1954), and Woodson (personal communication), has described
the process as “the unconscious application of the frequency curve
technique.” In a pictorialized scatter diagram a number of frequency
curves may be studied together and the diagram, which was devised
originally for the study of hybrid populations, should prove to be of
great value for the recognition of plant taxa.

2. Relationships: Robson (1928) wrote: “. . . if the systematist
were to adopt some method of expressing character groupings and
combinations as an adjunct to his traditional method, it would illustrate
the structural relationships of allied forms in a very useful manner.”
Ina study of the Australian and New Zealand species of Pittosporum
(unpublished). 1 interpreted the distribution of life-forms, the various
kinds of inflorescence, the various leaf types and capsule types as due
to evolution by reduction, possibly under the influence of aridity, and
used a ccatter diagram of the average measurements of five characters
for each species in support of my hypothesis.

3. Variation within taxa: Clausen (1951) emphasized that the
local population is the basic unit in plant evolution and that there is
considerable individual variation within each local population, even in
yopulations of apomictic species that propagate as clones. He used
photographs, diagrams, histograms, graphs and tables to illustrate this
variation, and his illustrations convey a much clearer impression than
the subspecific or varietal epithet. The scatter diagram should prove
as valuable as the other illustrations mentioned to provide an accurate
pictorial image of the variation ina local population or larger taxonomic

category.

300 COOPER.

Table 1.

Mass collection of flowering specimens from Huia, mainly between
reference points 088420 and 066397 on the N.Z. Lands and Survey
Waitakere map of 1943 (1: 63360 series). a |

Ne. Internode ’ Leat Leat Pedicel Flower

length. width. length. length. — length.
i 15 16 56 ! 4 9
2 15 1] 36 3 7
3 14 15 45 2 5
4 16 18 64 3 8
5 b2 10 51 2 7
0 16 22 RU 2 S
7 19 Za 91 3 10
8 11 19 65 fs 9
9 8 11 38 2 7
10 14 12 50 3 -
11 10 8 27 ae 5
12 8 16 37 3 ¢
13 6 15 50 3 6
14 7 13 42 Ms 6
15 5 11 20 rs 5
16 14 17 67 3 m3
17 11 10 35 3 6
18 15 13 36 a 6
19 11 14 37 Z §
20 8 8 28 2 7
21 7 11 47 2 6
22 7 10 46 2 6
23 24 21 94 4 8
24 il 9 42 2 5
25 17 13 AO 3 7
26 9 10 44 2 5
27 15 11 50 Z 5
28 11 13 43 3 7.3
29 BS he Sf 2 9
30 8 11 35 2.5 6
31 6 “2 30 Z 4
32 6 7 21 2 4
33 15 14 54 2 6
34 9 18 58 ws 6
35 15 iy 65 3 7
Table 2.

Mass collection from Duck Creek, Blockhouse Bay, between reter-
ence points 237517 and 232515 on the N.Z. Lands and Survey Titirangi
map of 1944 (1: 25,000 series).

No. Internode Leaf Leaf Pedicel Flower

length. width. length. length. length.
81 27 20 67 4 8
82 foe 21 73 2 rs)
83 24 23 78 4 8
84 23 20 73 4 7
85 16 20 66 3 6
86 23 19 74 4 7
&7 17 18 71 4 7
88 17 13 52 3 6
89 19 11 36 4 7
O1) 29 26 82 4 8

Variation in Hebe. 301

No. Internode Leaf Leaf Pedicel Flower
length. width. length. length. length.
9] met 20) (84 4 65-
92 pee. 19 54 4 6.5
93 25 Zh 101 4 a
94 14 16 59 3 7
95 15 15 52 3 if
96 19 17 70 3 8
97 | 21 | 21 64 4 9
98 24 19 9] ee 9
99 15 21 75 4 8
100 3 15 63 3 7.5
101 19 18 70 3 7
102 17 13 45 3 id
103 17 19 61 3 7
104 25 17 68 3 9
105 18 17 69 3 8.5
106 16 19 60 3 7
107 19 16 64 4 8
108 18 19 54 4 7
109 13 16 52 3 7
110 18 19 89 2 8
11] 20 22 109 4 7
112 14 17 57 3 6
113 12 19 67 3 7
114 18 21 77 4 8
115 von =i 25 92 5 8
Table 3.

Flowering Specimens in the Cheeseman Herbarium, Auckland
Museum.

Internode Leaf Leaf Pedicel Flower
Identity, Locality and No. length. width. length. length. length.

ffebe obtusata

Muriwai, 7670 10 15.5 35 2 6

Muriwai, 7671 10 14 24 RS 6

Muriwai, 7672 12 14 22 2 5

Kare Kare, 7673 8 16 31 2 6
leronica bishopiana

Huia, 2160 12 14 45 2 5

Huia, 7674 14 13 40) 2 5
Flebe salicifolia var. stricta

Woodhill, 7763.1. 13 15 59 2 6

Woodhill, 7763.2. 22 16 68 2 5.5

New Lynn, 7764 14 12 63 2 5
var. /Jongiracemosa

Waitakere Falls, 7775 19 20) 79 2 6
fLebe macrocarpa

Anawhata, n.n. 18 22 86 4 8

Nihotupu, 7713.1. 29 21 87 4 3

Nihotupu, 7713.2. 23 | 23 83 4 8

Nihotupu, 7714 20 24 107 4 8
Ivebe x macrosala

Northcote, 7730 (as H. x | |

affints ) 15 16 67 ys 8

302 CooPER.

Table 4.

| Collection of flowering specimens from an exposed clay bank above
Duck Creek, Blockhouse Bay, at reference point 235516 on the N.Z.
Lands and Survey Titirangi map of 1944 (1: 25,000 series).

No. Internode Leaf Leaf Pedicel Flower
length. width. length. length. leneth.
40) 5 7 32 2.5 6
71 7 10 33 3 7
V2 4 8 16 3 6.5
73 3 9 23 2 5
74 6 13 38 3 7
75 5 12 38 3 6.5
76 BS 10 37 3 7
77 9 14 42 3 7
78 6 9 27 3 7
79 5 13 43 3 7
80) 9.5 7 27 S 5
Table 5.

Mass collection of fruiting specimens from Huia.

No. Internode Leaf Leaf Capsule Capsule
length. width. length. width. length.
wy 11 10 35 2 3
18 15 13 36 2 Ate
19 11 14 37 a 3
25 17 13 60 2 3
27 15 11 50 Z 45
28 11 13 43 5 7
30 8 11 30 Z 3
32 6 7 ea Zi 3
34 9) 18 58 a0) S
327 18 19 re 5 6
329 13 14 60 4.5 Pe
331 20 18 85 5 6
341 9 y) 42 2 3
344 je l2 43 2 3
346 7 11 45 “3 3
347 14 10 4] 2 Soe
348 19 19 ~ LNSS 4 6
349 17 17 70 3.5 me
354 16 17 69 ) es,
351 18 24 83 4 -
Sys 9 17 67 os 3
353 18 23 79 4 7
354 7 * =H 45 2 3
obs 8 13 63 2 oe)
356 8 14 62 2 4
357 ) 15 47 Zea Pe
358 17 19 76 5 8.5
359 8 9 37 2.5 3
360 13 ae 62 4 6
361 9 13 60 2 3:5
362 5 15 44 2 (Re
363 10 18 ‘Gi 4 5
364 12 15 81 Zz 4
365 6 15 44 2 3
366 10 6 32 2 3

Variation in Hebe. 303

Table 6.

Mass collection of fruiting specimens from Duck Creek, Blockhouse
Bay.

No. Internode Leaf Leaf Capsule Capsule
length. width. length. width. length.
143 14 12 49 3 4
145 17 15 68 4 a Pe
150 ieee 17 79 4 6
151 30 21 87 4 6
156 10 12 4? te 3
158 13 14 43 fae 4
159 19 19 64 4 5
160 25 18 88 4 6
163 17 : 18 51 30 55
176 19 11 50 2 3
177 18 15 S1 oe 5
181 20 23 90 4.5 6.5
186 15 9 38 Z 3
187 18 9 44 2 3
189 14 13 At) oe Fes) oe)
190 10 10 50 | Re 3
194 13 14 64 3 5
197 17 16 75 35 0
199 Le 19 67 4 6
200 21 rz 64 3.5 6
201 19 22 103 2 3
203 17 19 80 + 55
205 13 r 54 4 5.5
206 2} 15 75 3 6
207 19 18 62 4 6.5
208 21 15 73 BS 6.5
210 22 18 67 4 6.5
211 14 20 66 4 5.5
214 24 20 82 4 6.5
215 Ze 15 70) 3.5 5
216 30 23 109 5 7
224 20 23 O0) 3 6
225 30 17 89 4 ‘i
226 15 18 59 4 6
233 se 22 82 3.5 6
Table 7.

Fruiting specimens in the Cheeseman Herbarium, Auckland
Museum.

Internode Leat Leaf Capsule Capsule
Identity, Locality and No. length. width. length. width. length.
Hebe obtusata |
Muriwai, 7670 10 13 29 20 30D
Muriwai, 7671 11 17 36 2.5 4
Muriwai, 7672 11 16 28 3 5
Kare Kare, 7673 8 16 31 2.5 4
Anawhata, n.n. 7 15 21 re 3
Veronica bishopiana
Huia, n.n. 9 13 44 2 3
Hebe salicifolia var stricta
Woodhill, 7763.1 13 15 59 2 3
New Lynn, 7764 14 12 63 2 3

304 CoopPeER.

f Internode Leat Leai Capsule Capsule
Identity, Locality and No. iength. width, length. width. length.
var. longiracemosa
Waitakere Falls, 7775 19 2) 79 Z 3.5
[debe macrecarpa
Nihotupu, 7713.2 23 23 $3 4 6
fi. x macrosala
Northcote, 7730 (as I7. »
affinis ) 15 16 67 3.5 6

Table &.

Collection of fruiting specimens from an exposed clay bank above

Duck Creek.

No. Internode Leat Leat Capsule Capsule
length. width. length. width. length.
72 4 8 16 3.5 5
74 6 13 38 3,5 6
75 5 12 38 3 5
76 5.5 10 37 7 5
7/ 9 14 42 4 6
79 5 13 43 3.5 5
80 9.5 7 27 3 4.5

N otes—

1. All measurements are in mm.

2. “Internode length” refers to the second internode beneath the lowermost
pair of racemes on a mature woody branchlet.

3. “Leaf width” and “leaf length” were measured on one of the pair of leaves

subtenditige the lowermost racemes.

4. “Pedicel length” refers to the pedicel of a fullytopen flower near the hase
of one of the racemes.

5. “Flower length” is the length of the calyx and corolla of the fully-open
flower.

6. “Capsule width” and “length” refer to a mature capsule near the base at
a raceme.

REFERENCES.

ANDERSON, E., 1949. Introgressive hybridization. Wiley, New York, 109 pp.

ANDERSON, E.. 1939, Recombination in species crosses. Genetics 24: 668-098.

ARBER, AGNES, 1954. The mind and the eve. A study of the biologist’s stand-
point. Cambridge Univ. Press, 146 pp.

CHEESEMAN, T. F., 1925. Manual of the New Zealand flora. ed. 2, 1163 pp.

CLAUSEN, J., 1951. Stages in the evolution of plant species. Cornell Uniy.
Press, 2C6 pp.

COCKAYNE, L., 1929. New combinations in the genus Hebe. Trans. NZ. Inst.
60: 465-472.

COCKAYNE, L., and H. H. ALLAN, 1926. The present taxonomic status of the
New Zealand species of Hebe. Trans. N.Z. Inst. 57: 11-47.

COOPER, R. C., 1954. Pohutukawa x Rata. Variation in Metrosideros
(Myrtaceae) on Rangitoto Island, New Zealand. Rec. Auck. Inst.

Mus. 4 (4): 205-211.

COOPER, R. C., 1953. The Australian and New Zealand species of Pittosporumt.
Ph.D. thesis (unpub.), Auckland Mus. library.

PETRIE, D., 1926. Descriptions of new native plants. Trans. N.Z. Inst. 56: 6-16,
ROBSON, G. C., 1928. The species problem, Oliver and Boyd, London, 283 pp.

Variation in Hebe. 305

Diagrams of Measurements in Tables 1-4 to show Variation in flowering
specimens of Hebe,

ws
|

fyi

Fig. 1. specimens from Huia, collected July-August, 1954,

>
~
-

Cy.

Fig. 2. 35 specimens from Blockhouse Bay, collected July-August, 1954.

Pig. 3. 13 specimens in the Cheeseman Herbarium. The numbers alongside the
symbols refer to the species and hybrids: 7

lL. Hebe obtucata.

2. H. salicifol.a var. stricta.

3. HH. salicifolia var. longiracemosa.
4+. HH. macrocarpa.

5. HT. x macrosala.

/. Veronica x bishopiana.

}

Fig. 4. 11 dwaried specimens from a clay bank above Duck Creek, Blockhouse
Day, collected August, 1954.

Hori.ontal axis, leaf width; vertical axis, internode length.

Three other cnaracters diagrammed by rays:

Leaf leneth: 20 - 50 @ 51 ~ 65 66+

Pedicel length 1.5 - 2 e& 205 =~ 345 Lt

R+

Flower length 4-6 @ 6.5-7.5

All measurements are in tm, ,

COOPER.

306

Variation in Hebe. 307

Diagrams of Measurements in Tables 5-8 to show Variation in fruiting
specimens of Hebe,

Fig. 5. 35 specimens from Huia collected July-November, 1954.
ig. 6. 35 specimens from Blockhouse Bay collected September-October, 1954.

—_~-e

Fig. 7. 11 specimens in the Cheeseman Herbarium. The numbers alongside the
eymbols refer to the species and hybrids:

Hebe obtusata.

H. salicifolia var. stricta.

H. salicifolia var. longiracemosa.
fl. macrocarpa.

ff. x macrosala.

eronica x bishopiana.

ale gee = fw bor

lig. 8. 7 dwarfed specimens from a clay bank above Duck Creek, Blockhouse
Bay, collected August, 1954.

Horizontal axis, leaf width; vertical axis, internode length.

Leaf length 20 - 50 gS 51 - 65

Capsule. Lenceh 2.5m 5 DBD 4 = 5.5

X\ 66+
Capsule width 15-2 @ 2.5-3.5 4 +
’ 6+

Al] measurements are in mm.

COOPER.

308

RECORDS

OF THE

AUCKLAND INSTITUTE
AND MUSEUM

VoL. 4 NO. 6

Published by Order of the Council:
Gilbert Archey, Director

Edited by: A. W. B. Powell,

Assistant Director

25TH OCTOBER, 1956

CONTENTS

VOL, 4, No. 6

Observations on the Structure of Far Northern New Zealand.

By M. H. Battey, Auckland Museum

Some Coleoptera from the Noises Islands, Hauraki Gulf.

By J. C. Watt, Papatoetoe

Aquatic Insects of Little Barrier Island.

By K. A. J. Wise, Plant Diseases Division, Auckland

Spiders from the Three Kings Islands.

By B. J. Marples, University of Otago

Notes on the Plumages and Breeding Cycle of the Spotted Shag,
Phalacrocorax (Stictocarbo) punctatus punctatus (Sparrman, 1786).

By E. G. Turbott, Auckland Museum

Tauihu: The Maori Canoe Prow.
By Gilbert Archey, Director

The Three Kings Cabbage Tree.
By W. R. B. Oliver, Wellington

Page

Page :

Page

309

aig

343

365

381

309

Observations on the Structure of Far Northern
New Zealand

By M. H. BATTEY, Auckland Museum.

Abstract.

Evidence is presented for the occurrence of two epochs of compression 1n
northern North Auckland Peninsula since the Middle Tertiary, in the first of
which the pressure was directed from north and south, while in the second it was
{rom east and west.

East-west fold axes due to north-south compression are important in the
region north of the Bay of Islands. <A later system of tear faults extends from
North Cape to the west head of Whangaroa Bay, coinciding with the belt of
intrusion and mineralization recognised long ago by Hector, and is thought to be
related to compressive force from east and west.

Such successive foldings at right angles have already been indicated by Lillie
in other parts of New Zealand, particularly as an alternative to Macpherson’s
supposed swinging strike in the Watapu district.

The idea that fold axes strike north-west in the Far North is not confirmed,
except in so far as the whole New Zealand Ridge may represent a geanticline,

INTRODUCTION.

The structural interpretation of the North Auckland Peninsula
has long been a matter of uncertainty. Two main ideas have been pro-
posed. The first, which was, in general, held by the Old Geological
Survey, is that the trend of the Peninsula is not that of the fold axes
but was determined by later fractures. This view was adopted by
Benson in 1924, The other proposal was that the trend of the Peninsula
does reflect that of the fold axes, and was held tentatively by Ferrac
(1927) on the basis of a few observations of strike in the older rocks,
Bartrum and Turner (1928) supported this view, being apparently
largely influenced by the north-west trend of “foliation” in the gabbro
at North Cape. While this second hypothesis may be correct in so far
as the New Zealand Ridge as a whole represents a geanticline, it is felt
that to draw a group of north-west-trending anticlines and synclines
throughout the length of the Peninsula, as has been done by Macpherson
(1946), may be misleading.

DIRECTIONS OF FOLD AXES.

Study of the available information on the Far North (north of the
Bay of Islands) suggests that we have here fold axes running between
west-south-west and west-north-west, oblique to the length of the North
Auckland Peninsula,

Of this large region the North Cape area is perhaps the best known.
A map, compiled from the work of McKay (1894) and Bartrum and
Turner (1928), is presented to show the available information on the
dip and strike and distribution of the beds there (fig. 1). Two points
about the map call for comment: (1) McKay showed andesitic con-
elomerate extending over the area marked'as Cretaceous, between the

Rec. Auck. Inst. Mus. Vol. 4, No. 6, pp. 309-315, 25th October, 1956

310 BATTEY.

two question marks and east-south-east of them. Bartrum and Turner
found Cretaceous lavas in this area, but patches of andesitic conglom-
erate are shown flanking the Cretaceous, for it is very probable that
McKay had some basis for his opinion, while Bartrum and Turner
examined only exposures adjacent to the track from Te Paki to Te
Hapua. (2) The beds around the northern and western shores of
Parengarenga Harbour were placed below the andesitic conglomerate
by McKay. Bartrum and Turner, however, place them above the con-
glomerate, for conglomerate emerges from beneath them in low cliffs
on the north shore of the harbour, and this later conclusion is here
adopted.

The recorded dips and the distribution of the beds suggest very
strongly that there is a system of folds trending about west-north-west
in the North Cape area. Data on the attitude of the Cretaceous lavas
and sediments are scanty, but the strike observed at Pandora suggests
that the long straight coast between Spirits Bay and Te Reinga may be
a strike coast. The youngest beds recorded as affected by this folding
are the sandstones, mudstones and grits (t3) which immediately overlie
andesitic conglomerates dated as Miocene (Altonian) (Couper, 1952).

A second map (Pl. 49) covers the area between North Cape and
Whangaroa Harbour on the east and Herekino River on the west.
In the south-eastern part of this region Bell and Clarke (1909) infer
the existence of two anticlinoria of late Palaeozoic to Triassic beds, the
more southerly, inland one running west-south-west and the other,
further north, on the coast east of Whangaroa Harbour, running west-
north-west. Westwards, the geology is not well known and we are
dependent mainly upon McKay’s map of 1894, There appear to be two
massifs of older rocks, of unknown age, lying in an east-west line
south of Kaitaia. They comprise much igneous material, partly intrusive
and probably partly ancient lava flows. They may in part be Cretaceous
in age and comparable with the Cretaceous lavas of the North Cape
area. ‘This is believed to be true of lavas, pillowy in part, around
Mangonui, in the south-east corner of Doubtless Bay, for these lavas,
on the coast at Taipa, appear to be conformable with the Cretaceous beds
in the lower part of Taipa River; but no boundary can yet be drawn
between these supposedly Cretaceous lavas and any older rocks that
ay be present.”

The 1948 Geological Map of New Zealand, published by the Geo-
legical Survey, subdivides McKay’s Cretaceo-Tertiary formation in a
broad way, into beds of Cretaceous (Mata) and Tertiary (Landon)
age, with a strip of rocks of Arnold age east of Kaitaia and some
Wanganui beds in Victoria Valley. These distinctions are very valuable
and serve to show the general trend, between west-south-west and due
west, of the axes of folding in the area between Reef Point and Whanga-
roa Harbour. In Taipa River (south shore of Doubtless Bay) the
strike of the Cretaceous beds is east and west, the dip south at moderate
to high angles, for a mile across the strike.

In Rangiawhia Peninsula (Battey, 1950) two main formations are
present, breccia, conglomerate and grits standing vertical, striking west

*It is well known, on definite evidence from other parts of North Auckland, that
contemporaneous lavas occur in both Permian and Cretaceous parts of
the marine sedimentary sequenice.

311

Geology of Northland.

‘SIN 8 + € zt oO

22 /4OID 9 Od gGqIeH

“7 00sELI

cs!
<7

Tp ay 53 LV
eg 77/58

“

"syspues A4LBU0I2312u03
VL0/ ‘S@(EYS "SMO/4-PASY DISEG SNOBIE}OA*)
j Sage4auMo0jbu0d '$44N7
E%7 J ‘Saugjspriu ‘sauozspueg

43] 7e42W0/ buds oigisapuy
S2t4B
[783] Sauogspnus ‘seuozspues
[~ ] wmanyegseunp pues AsPusazeny
VIdAV AdVI ALYON
IHL 10 dVW TVOID0TOAD

A4e1743]

reek SEES : a rae 5° " Lr AQ
0 Ag 9 OS AGEN Ys BNR Yj eee (Ce 42
. é 5 ps a oN: - Sate S32 | Ue ey 7
ee 3 Uys e 23° we he . 5 ay
Ay Wp at 7 gs S
Fae Ls Myjweee z
wee i Upped " F eSurey ades

ee pine
Pa, 13 bse

Fig, 1.

312 BATTEY.

and younging south along the south-east coast, which rest unconformably
upon a group of basic pillow lavas and keratophyres, in which a roughly
westerly strike is inferred from correlation of pillow lava bands and
from topographic expression of the main keratophyre horizon. Neither
group 1s satisfactorily dated, but there are some grounds for referring
the extremely heavy conglomerates (with included graphic granite
houlders 9ft. across) to the same transgression as the conglomerates at
the base of the Cretaceous succession at Whangaroa (also with granitic
pebbles) recorded by Bell and Clarke (1909). |

In Mount Camel (fig. 2), to the west-north-west, both basic pillow
lavas and keratophyres similar to those of Rangiawhia occur, and from
the distribution of characteristic types of keratophyre a strike about
N85°W is inferred. For this whole group of pillow lavas and kera-
tophyres the name Mt. Camel formation, used by Bell and Clarke
(1910) is convenient, In passing it may be noted that a very similar
group of rocks builds the western part of Great Island in the Three
Kings Group. It seems reasonable to regard the Mt. Camel formation
at Rangiawhia and Mt. Camel as representing an anticlinal axis.

To sum up, we seem to have fold axes striking west-south-west
forming an eastward-pointing V with west-north-west folds in the
south, an east-west group in the angle of the V, and a dominance of the
west-north-west trend in the north. Broadly speaking, we can regard
these folds as due to compression from north and south, or north-north-
cast and south-south west.

TRANSCURRENT FAULTING.

This fold system is cut across obliquely by a fault system running
between N30°W and 40°W along the north-east coast, with which is
associated a large dyke-like intrusion of gabbro and diorite with sub-
sidiary andesite dykes. This zone was recognised long ago hy Hector
(1891) as extending from North Cape through Rangiawhia Peninsula
and Stephenson’s Island to Cape Brett. He pointed it out in connection
with the mineralization that has taken place at points along its course.
In 1894 Hector described the antimony prospects in the Cape Brett
area and records a general trend of N40°W for the stibnite lodes in
that south-eastern extension of the zone.

Mapping at Rangiawhia has shown that there the faults of this
system are tear faults (transcurrent faults) with sinistral displacement
(i.e., north-westward movement on the north-east sides of the planes)
with an aggregate horizontal shift of at least 24 miles, distributed over
a belt of country 34 miles wide (Battey, 1950, with maps),

This faulting is later than the folding of the upper Cretaceous rocks
(with Aucellina) which are intruded by the associated igneous rocks
and sheared by the movements at Pa Island on the west shore of
Whangaroa Bay (see also the Survey 1 inch map of 1909).

It is to the stress system associated with this transcurrent faulting
and multiple dyke-injection that the north-westerly foliation in the
North Cape gabbro must be ascribed. The new data thus necessitate
a revision of the view of Bartrum and Turner (1928) that this miner-
alogical banding is due to folding about north-westerly axes, and
removes one of the principal criteria on which this interpretation of the
structure was based.

313

asAYdoJessy 21981
-IQIEIEYI 4O PUET

paddizs 714g

Geology of Northland,

ov Sz Meee y
bt tw dip PUB BIAS . << NP,
BAL] Moria D12/014BA DSeg ~\. LC, ee
aurAydazes3ay were <e

QUOISPUES Pappaq pazeanpuy és s s
a SS )

TIWVD LNNOW K-10) |
zr \

10 yh, ° ) 2 “aoe \ Sook my
HILINS TWII9OT0IO eth: SB

AYU NIN TI ad y (! 2 Ss ) <
t

314 BATTEY.

MECHANICS OF FAULTING.

The directions of the sub-vertical dykes associated with this move-
ment, and regarded as occupying fissures produced by it, may be
expected to give information about the nature of the forces involved.
Measurements of these directions and of the directions of very steeply-
dipping joints in the intrusive rock have been made, chiefly at Rangi-
awhia. Both kinds of measurement show the same direction-frequencies
and the two sets combined have been plotted as a direction-frequency
diagram (PI. 49). This diagram shows equal maximum concentrations
in the azimuth-groups N40-45°W and N25-30°W. Two other equal
concentrations lie between N&85°W and W and N5°E and N. A pair
of much less marked concentrations occurs at N50-55°E and N60-65°F,
that is at 95° and 90° to the N40-45°W and N25-30°W groups respect-
ively,

These directions can, on the whole, be explained remarkably well
in terms of the stress theory of the rupture of homogeneous bodies in
compression, as outlined by Wilson (1947) and applied (to a problem
somewhat similar to the present one) by Blyth (1950), but there remain
some uncertainties in interpretation, If compression acted from north
and south the northerly fractures could be explained as tension cracks,
hut the westerly ones remain unexplained. If compression acted from
east and west, the westerly cracks can be regarded as tensional but the
northerly ones cannot be explained. In the field both kinds are filled
by dykes in the country rocks and in the main body of the intrusion.
Those of the westerly group are perhaps more marked and it is note-
worthy that east-west dykes occur over a rather wide area. There is
a relatively big one at Taipa River mouth on the south shore of Doubt-
less Bay, which is quarried. It is 100 feet wide and can be followed
eastwards for three-quarters of a mile. At Pa Island (west shore of
Whangaroa Bay) there is another conspicuous one, and another at
Tupou Bay to the west-north-west.

If the stress ellipsoid is orientated with maximum pressure from
a shade east of north and west of south, the main north-west shear
directions fit slightly better the theoretical locus of stress shear planes,
than if maximum pressure from just north of east and south of west
be assumed.

What may decide the issue is the sinistral sense of the movement
on the faults along the Rangiawhia coast. Sinistral movement on a
north-west-striking transcurrent fault apparently implies east-west com-
pression. If the pressure were from north and south the movement
would be dextral. This seems to be a rather rigid requirement of the
theory. For this reason we are apparently compelled to assume east-
west compression, If we could postulate more or less north-south
compression we should be able to explain the folds already described,
and the subsequent tear faulting, as due to the one system of compressive
forces. As it is, it seems necessary to assume first a north-north-east—
south-south-west compression to produce the folds, followed by com-
pression from east and west to explain the tear faults.

This demand is not a new one, for Lillie (1951, pp. 236-8) has
recently postulated a similar abrupt change in the directions of com-
pression to explain the way in which folds in the Tertiary strata run

Geology of Northland. 315

almost at right angles to those in the Cretaceous rocks in the Waiapu
area, and has found evidence of similar happenings in the south of the

South Island. His hyphothesis may prove to have rather wide applica-
tion.

It may be remarked in connexion with the Rangiawhia fault pat-
tern that a fault striking about N55°E runs along the south-east coast
of the Peninsula between tide marks. It is older than the north-west
faults and is shifted by them. Nothing definite is known of the type
of displacement along it, but its direction may suggest that it is a con-

jugate fracture of the same stress system as produced the north-west-
striking tear faults.

REFERENCES.

BARTRUM, J. A., and TURNER, F. J., 1928. Pillow-lavas, peridotites and
associated rocks of northernmost New Zealand. TJvrans. N.Z. Jnst.,
59, 98-138.

BATTEY, M. H., 1950. The geology of Rangiawhia Peninsula, Doubtless Bay,
North Auckland. Rec. Auck. Inst. and Mus., 4, (1), 35-59,

BELL, J. M., and CLARKE, E. deC., 1909. The geology of the Whangaroa Sub-
division, Hokianga Division. N.Z. Geol. Surv. Bull., No. 12 (n.s.).
, 1910. A geological reconnaissance of northernmost New Zealand.
Trans. N.Z. Inst., 42, 613-624.

BENSON, W. N., 1924. The structural features of the margin of Australasia.
Trans. N.Z. Inst., 55, 99-137.

BLYTH, F. G. H., 1949. The sheared porphyrite dykes of South Galloway.
O.J.G.S., 105, 393-423.

COUPER, R. A., 1952. The spore and pollen flora of the Cocos-bearing beds,
Mangonui, North Auckland. Trans. Roy. Soc. N.Z., 79, 340-348.

FERRAR, H. T., and others, 1925. The geology of the Whangarei-Bay of Islands
Subdivision, Kaipara Division. N.Z. Geol. Surv. Bull. No. 27 (us.).

HECTOR, J., 1891. Progress Report. epts. Geol. Explor. during 1890-91
(No. 21), pp. Ixxx-lxxxil.

-—_—___-_———., 1894. Progress Report. MRepts. Geol. Explor. during 1892-93.
(No. 22), p. xxiii.

LILLIE, A. R., 1951. Notes on the geological structure of New Zealand. 7 rans.
Roy. Soc. N.Z., 79, 218-259.

McKAY, A., 1894. On the geology of Hokianga and Mongonui Counties, North-
ern Auckland. Repts. Geol. Explor. during 1892-93 (No. 22), 70-90.

MACPHERSON, E. O., 1946. An outline of late Cretaceous and Tertiary dias-
trophism in New Zealand. N.Z. Dept. Sci. and Industr. Res., Geologi-
cal Memtor No, 6.

WILSON, GILBERT, 1947. The relationship of slaty cleavage and kindred
structures to tectonics. Proc. Geol. Assn., 57, 263-302.

PLATE 49

y
a
~
g

Te Reinga North Cape

34°30S
!
TENSIONI
Seana abate
' Directions of dykes & <—Pmax
' sg Joints in intrusion at
\ angiawhia (112 meas -
NORTHERN urements)
— NORTH AUCKLAND —
SKETCH OF STRUCTURE
Explanation
[ Recent & Hawera :also
Wanganui in Victoria Valley
= a Ultrabasic to intermediate
intrusives ‘
Tertiary - Upper Coal Point 5 Big
Beds at North Cape a as 7 Stephenson's Is
picts rad Tertiary = anh ingaht® cong- ? Pe, ee) Whangarea Bay » :
ee 2 omerates ys “ Se) WW a SS NE es <s Og eG __ gS 008
ats ; Z, ay 2] : i Mr? Ro mers
Tertiary - Landon 7% if lp Yy Ri inst

Tertiary - Arnold

\
\

Upper Cretaceous - Mata
Permian to Triassic Reef Pt.

Mt Camel formation —-
keratophyres &c.

Unclassified formations
?mostly pre-Cretaceous

—+— fold axes

5 10

L160

Miles

Some Coleoptera from The Noises Islands,

Hauraki Gulf
By J. C. WATT, Papatoetoe.

During late August, 1954, a limited amount of collecting was done,
chiefly during a fortnight spent on Otata Island, the inhabited island
of the Noises Group. The David Rocks (or Four Brothers), islets to
the south-east of Otata, and the other main island, Motuhurakia, to the
north of Otata, were also visited. One of the David Rocks was revisited
on 3rd March, 1956.

The Noises group lies 15 miles to the north-east of Auckland in
the Hauraki Gulf. Otata, a few acres in extent, is covered chiefly by
low regenerating Leptospermum scrub and tussock dominated by
Phormium tenax, a large portion of the island having been fairly
recently burnt; the remainder is covered by typical Auckland coastal
forest. Motuhurakia, slightly smaller, is completely covered by coastal
forest in a fairly advanced stage of regeneration, while the remainder
of the Noises group are little more than rocks, bearing a scanty, wind-
swept vegetation of hardy coastal shrubs.

Family CARABIDAE.
Ctenognathus sp.

One specimen was recorded from Motuhurakia but the specimen
was later unfortunately mislaid.

Family TROGOSITIDAE.

Leperina brounii Pasc.

Two specimens under bark of pohutukawa (Metrosideros excelsa),
Motuhurakia.

Phycosecis discoidea Pasc.
Common on sandy beach above H.W.M., Otata.

Family TENEBRIONIDAE.

Cilibe humeralis Bates.

Very common on the David Rocks under stones; also found on
Otata and Motuhurakia. More active on second visit to the David
Rocks, probably due to the season.

Lorelus sp.
Beaten from a shrub, Motuhurakia, two specimens,
Leiopeplus expolitus Br.

Very common in rotten wood on Otata and Motuhurakia and under
fallen branches and stones on the David Rocks.

Rec, Auck. Inst. Mus. Vol. 4, No. 6, pp. 317-319, 25th October, 1956

318 WATT.

Family CERAMBYCIDAE.

Navomorpha sulecatum Fabr.

One specimen beaten from flowering shrub, Motuhurakia.

Xyloteles lynceus Fabr.

Two specimens beaten from flowering shrubs, Motuhurakia.

Xyloteles griseus Fabr.
One specimen as above.

Xyloteles nanus Bates.

Four specimens as above; also two specimens from the David
Rocks, March, 1956.

Xyloteles sp.
One specimen as above.

Family CURCULIONIDAE
Sub-family EUGNOMINAE
? Hoplocneme sp.

One specimen beaten from flowering shrub, Motuhurakia.

DISCUSSION.

It will be noted that two species, Cilibe humeralis and Leiopeplus
expolitus, are very common. Hudson (“New Zealand Beetles and Their
Larvae,” 1934) states that C. humeralts is “common under stones above
high water mark on all the beaches around Wellington.” My records
indicate that it is also common on or near the sea shore in the vicinity
of Auckland, especially on islands. L. expolitus also appears to be a
coastal insect but not as exclusively as C. humeralis, Situations for this
species according to my observations, besides those above, are under
bark of dead karaka (Corynocarpus lacvigata) and live puriri (Vitex
lucens) trees, so it is obviously a more versatile insect than C. humeralis,
which only occurs under stones. Both species are probably scavengers,
feeding on decaying animal and vegetable matter. It would be interest-
ing to study the feeding habits of the adults on the David Rocks, where
vegetation is scanty and little soil is present and much competition for
food appears probable. Phycosecis discoidea is common on most sandy
beaches, feeding on decaying animal matter. The several species ot
Xyloteles are probably quite common on the Noises; they are phyto-
phagous insects.

No doubt intensive collecting would yield many more species, but
the majority of the common ones that are normally adults in early spring
are probably contained in the species list above.

ACKNOWLEDGEMENTS.

I have greatly appreciated the opportunity to examine material of
the above species in the Auckland Museum. A series of duplicates has
been placed in the Museum collection.

Coleoptera from Noises. | 319

APPENDIX.

_Horuhoru, a small island just to the north of the eastern end of
Waiheke Island, and E.S.E. of the David Rocks, was visited on 3rd
March, 1956. This islet is only a little larger than the western islet of
the David Rocks, but the vegetation has been modified by a large gannet
colony, and consists mainly of stunted taupata (Coprosma repens).
The following species were taken:

Family HISTERIDAE.
Abraeus sp.

Four specimens beaten from taupata.

Family TROGOSITIDAE.
Leperina brounii Pasc.

One specimen cut from dead taupata.

Family CRYPTOPHAGIDAE.
? Cryptophagus sp.
Eight specimens beaten from taupata. There are eleven specimens
of this species in the C, E. Clarke Collection, Auckland Museum, from
various localities near Auckland.

Family TENEBRIONIDAE.

Cilibe humeralis Bates.

Common under rocks. This species is often found with Anisolabis
littorea (White), the large coastal earwig which is common on islands
in the Hauraki Gulf.

Family CURCULIONIDAE.
Sub-family CRYPTORRHY NCHINAE.

Two species belonging to Acalles or a related genus beaten trom
taupata, one species very common.

321

Aquatic Insects of Little Barrier Island

By K. A. J, WISE, Plant Diseases Division, Department of
Scientific and Industrial Research, Auckland.

In November, 1954, a collection of aquatic insects was made on
Little Barrier Island. This island lies at the entrance to the Hauraki
Gulf, 14 miles from the mainland. The island, which is approximately
7,000 acres in area, is more or less circular and has a central mountain
group with a consequent radiating topography, It is a sanctuary and
is covered with native forest.

The aquatic fauna of the island is of a restricted ty pe, as the streams
are ephemeral. Watersheds are steep and the run-off is rapid. Hamil-
ton (1935) stated, “Except after heavy rains, many of the streams carry
little or no run of water during the drier seasons of the year.” None
of the streams investigated was flowing; pools and stream beds yielded
the specimens recorded below.

Collecting was confined to the south-western sector of the island.
The Te Wairere stream bed was the most western investigated and
there nymphs of Ameletopsis perscitus (Eaton) and Atalophlebia den-
fata (Heaton) were found amongst damp fine gravel and vegetable debris
under stones as well as in pools. In the Neamanauraru stream bed,
hanging above Ngamanauraru Bay, only one pool was examined. Wat-
pawa, Turner’s (opening out on to Marae Roa), Te Waikohare, and
Awaroa stream beds were also investigated, In the last-named a larva
of Archichauliodes diversus (Walker) was seen amongst stones in the
dry bed. Specimens taken at light were collected on the “the flat”

(Marae Roa).

In addition to specimens collected on this expedition a specimen in
the Auckland War Memorial Museum collection is included in these
records. Some duplicates from the material collected are lodged in the
Museum collection, the rest are in the Plant Diseases Division collection.
Unless otherwise stated, all specimens were collected by the author.

PLECOPTERA
Family Gripopterygidae

Nesoperla trivacuata Tillyard
1923—Nesoperla trivacuata Tillyard, Trans. N.Z. Inst., 54 + 211.

19. Running on stone in rain, Awaroa Stream bed, 25/11/1954,

EPHEMEROPTERA
Family Siphlonuridae

Ameletopsis perscitus (Katon)
1899—Ameletus perscitus Eaton, Trans. Ent. Soc. Lond., 47 ; 291,
2 Nymphs. ex pools, Te Wairere Stream bed, 24/11/1954.

Rec. Auck. Inst. Mus. Vol. 4, No. 6, pp. 321-327, 25th October, 1956

322 WISE.

Family Leptophlebiidae
Atalophlebia dentata (Eaton)
1871—Leptophlebia dentata Eaton, Trans. Ent. Soc. Lond., 19: 80, Pl. 4, fig. 18.

1 Imago. ex Tirikakawa Stream bed, 20/11/1947 (J. Dingley).
(Auckland Museum collection).

1 Imago. On surface of pool, Te Wairere Stream bed, 24/11/1954.
11 Nymphs, ex pools, Te Wairere Stream bed, 24/11/1954 (5,
Auckland Museum collection).
1 Nymph, ex pool, Waipawa Stream bed, 28/11/1954.
ODONATA
Anisoptera
Family Corduliidae
Procordulia smithii (White)
1845—Cordulia smithit White, Zool. Erebus and Terror, Insects, P1, 6, fig. 2.
1 Nymph. ex pool, Te Wairere Stream bed, 24/11/1954.

This nymph fits the description of Procordulia smithit by Hudson
(1904) but it could possibly be Somatochlora braueri (de Selys) the
nymph of which is as yet undescribed.

ZAygoptera
Family Coenagriidae
Xanthocnemis zealandica (McLachlan)
1873—Telebasis sealandica McLachlan, 4nn, Mag. Nat. Hist. (4), 12 : 35.
Ig,1¢. Flying above pool, Turner's Stream bed, 29/11/1954.

eggs in leaf tissue. ex pool, Turner’s Stream bed, 29/11/1954
(3, Auckland Museum collection).

The egg has not previously been described. A description is given
below.

Length: .84 mim. Width: 18mm, Elongate-oval, pedicel pointed.
Cream (in alcohol), pedicel brown. Chorion thin, colourless.

eggs were inserted, at random, into the soft tissues of half-rotten
plant debris just below the surface of the water.
HEMIPTERA
Heteroptera
Family Veliidae
Sub-family Microveliinae
Micrevelia sp.

l apterous 2 ;7 Nymphs. On surface of pool, Te Wairere Stream
bed, 24/11/1954.

Insects from Little Barrier. 323
1 apterous ¢; 1 apterous? ; 2 Nymphs. On surface of pool,
Ngamanauraru Stream bed, 24/11/1954.

_ Dr. T. E. Woodward has advised that these specimens are not
Microvelia halei Esaki but probably M. macgregori Kirkaldy, although
they differ somewhat from the description of that species,

NEUROPTERA
Megaloptera
Family Corydalidae
Sub-family Chauliodinae
Archichauliodes diversus (Walker)
1853—Hermes diversus Walker, List Specimens, Neur. Ins, Brit. Mus., 2: 206.

1 Pupa. Under stone, Awaroa Stream bed, 25/11/1954 Chak
Salmon),

TRICHOPTERA
Inaequipalpia
Family Sericostomatidae

Oeconesus maori McLachlan

1862—Oeconesus maori McLachlan, Trans. Ent. Soc. Lond. (3), 1 : 303.

1 2. ex Awaroa Stream bed, 28/11/1954.
Olinga feredayi (McLachlan)

1868—Olinx feredayi McLachlan, Journ. Linn. Soc. Lond. Zool., 10 : 198.

| Pupa in case, 2 Larvae in cases. ex pool, Te Wairere Stream bed,
24/11/1954.

The larval case is figured in Plate 50.

Helicopsyche sp.
1 Larva in case. ex pool, Te Wairere Stream bed, 24/11/1954.
The helicoid case is figured in Plate 50,

? Pycnocentria sp.
1 Larval case. ex pool, Waipawa Stream bed, 28/11/1954.

This case (Plate 50) is similar to that of Pycnocentria evecta
McLachlan described and figured by Hudson (1904). Probably belongs
to a species of Pycnocentria or an allied genus. Family determination
was made from larval and pupal skins which it contained.

The case is 6mm. in length, formed of sand grains on a horny
base. Slightly tapered and curved. The anterior sieve membrane
(text-fig. 1) is 1 mm, in diameter. It is quite substantial, made entirely
of a secretion, and shaped like a pill-box lid. The single opening is a
slightly curved slit.

324 WISE.

Text-fieure 1. 2? Pycnocentria sp. Anterior sieve membrane.

Aequipalpia
Family Philanisidae
Philanisus plebeius Walker
1852—Philanisus plebeius Walker, List Specimens Neur. Ins, Bru. Mus., 1: 116.
1 @. Swept ex boulder beach, 24/11/1954 (R. A. Harrison).
1 9. Swept ex boulder beach, 24/11/1954.
1 @. At light, 26/11/1954.

344,19. At light, 27/11/1954 (2 6 8,19, Auckland Museum
collection). |

Family Leptoceridae
Sub-family Triplectidinae
Triplectides obsoleta (McLachlan)
1262—-Pseudonema obsoleta McLachlan, Trans. Ent. Soc. Lond, (3), 1 : 305.
2 Larval cases. ex pool, Te Wairere Stream bed, 24/11/1954.

Each case (Plate 50) is a hollowed out piece of twig. One end of
the tube is blocked by small stones.

Family Polycentropodidae
Polyplectropus sp.
1 ¢,19. At light, 26/11/1954.
1 ¢@. At light, 28/11/1954.
2 Larvae. ex pool, Waipawa Stream bed, 28/11/1954.
2 Larvae. On debris in pool, Turner’s Stream bed, 29/11/1954.
1 Pupal case. ex pool, Te Wairere Stream bed, 24/11/1954.
1 Pupal case. ex pool, Waipawa Stream bed, 28/11/1954.

Insects from Little Barrier. 325

Adult specimens belong to an undescribed species of this genus,
but, as specimens representing at least two undescribed species are
known in other collections, description of a new species is deferred.

Larvae and imagines cannot definitely be assigned to the same
species.

_ The pupal cases are made of small stones tied together loosely with
silk (Plate 50).

Pupal cases have been associated with the larvae by means of cast
larval skins remaining in the cases.
Family Philopotamidae
Hydrobiosella stenocerea Tillyard °
1924—Hydrobiosella stenocerca Villyard, Trans. N.Z. Inst., 55 + 289.
1 9. On surface of pool, Te. Wairere Stream hed, 24/11/1954.
DIPTERA
Nematocera
Family Culicidae
Sub-family Culicinae
Aédes antipodeus (Edwards)
1920—Ochlerotatus, antipodeus Edwards, Bull. Ent. Res., 10 : 132.

1. Swept at bush margin, Te Titoki Point, 25/11/1954 (R. A.
Harrison). |

2292. ex Waipawa Stream bed, 25/11/1954 (R. A. Harrison).
1 6. ex Waipawa Stream bed, 28/11/1954 (R. A. Harrison).
1?. ex Turner’s Stream bed, 29/11/1954 (R. A. Harrison).
Culex fatigans Wiedemann
1828—Culesx fatigans Wiedemann, Assereur. sweifl. Ins., 1 : 10.
1 Pupa; 3 Larvae. ex pool, Ngamanauraru Stream bed, 24/11/1954.
1 ¢ ; 16 Larvae. ex pool, Te Waikohare Stream bed, 26/11/1954.
Family Dixidae
Dixa (Paradixa) sp.
1 Larva, ex pool, Te Wairere Stream bed, 24/11/1954.

Description of Larva

Length: 10mm. Body colour (in alcohol) white with darker seg-
mental patches on dorsum. Abdominal segments without dorsal crown
of setae. Ambulacral combs on segments 5, 6, 7. Structure of end of
abdomen as shown in text-figure 2. Gut somewhat extruded from anus.
Sloping anterior wall of saucer-shaped spiracular depression bears small
bifid chitinised plate. Lip of wall above plate bears two pairs of many
branched setae, outer pair double-tufted. Caudal appendage pubescent ;

Insects from Little Barrier. 327

ACKNOWLEDGEMENTS

The author is grateful to Mr. E. G. Turbott, of the Auckland
Museum, for the loan of a specimen from the museum collection, Dr.
T. E. Woodward, of The University of Queensland, Brisbane, Australia,

has kindly given his opinion on the identification of the Microveliids in
this collection.

REFERENCES.
HAMILTON, W. M., 1935. The Little Barrier Island. N.Z. J. Sci. and Tech.,
I7; 465-494.
HUDSON, G. V., 1904. New Zealand Neuroptera. West, Newman. London.
pp. 102.

TONNOIR, A. L. 1924. New Zealand Dixidae (Dipt.). Rec. Cant. Mus. 2 (4) :
221-33,

326 WISER.

long caudal setae inconspicuously plumose, two most dorsal and one of
nuddle pair missing. Lobe of lateral plate with outer strongly chitinised
opaque ring and inner transparent portion with light and dark areas,
dark area filling basal half and narrowing distally. At peak of dark
area a short tooth-like process arises from ventral surface of lobe and
projects posteriorly. Side of lateral plate setose on posterior margin,
postero-ventral angle bears three heavily chitinised teeth, one long and
finely produced. two short and stubby.

A B
Text-figure 2. Dixa (Paradixa) sp. Larva.
End of abdomen, A. Dorsal B. Lateral.

This specimen belongs to the subgenus Paradixa which was erected
by Tonnoir (1924) for two New Zealand species, D. neozelandica Tonn.
and D. fuscimervis Tonn. Tonnoir described the larvae of both species.
They can be separated by the form of the basal part of the lateral plates.
Two teeth present at the postero-ventral angle in D. neogzelandica are
absent in D, fuscinervis. The larva from Little Barrier Island bears,
at that point, three teeth, and there are other differences from Tonnoir’s
species in the structures at the end of the abdomen. It seems, therefore,
that this larva represents a third, and as yet undescribed, species of the
subgenus Paradixa.

;
Lv ew
Fs .

Caddis cases from Little Barrier Island.

Top: Polyplectropus sp. Pupal case.
Centre: ¢° Pycnocentria sp. Larval case.
Bottom leit: Olinga feredayt. Larval case.

Bottom middle: Helicopsyche sp. Larval case.

Pottom reht: Triplectides obsoleta. Larval case.

PLATE 30.

eo”

329

Spiders from the Three Kings Islands
By B, J. MARPLES, University of Otago.

The Three Kings Islands form a group lying about 35 miles to the
north-west of Cape Maria van Diemen, the extreme northern tip of
New Zealand. J am indebted to Mr. E. G. Turbott, of the Auckland
Museum, and to Dr. G. Archey, its Director, for the opportunity of
examining this collection of spiders. Most of the specimens were
collected by Mr. Turbott on Great Island, but a few are from South
West Island, and one from Stella Rock, Great Island is irregular in
shape with greatest diameters of about 14 miles and rising to a height
of some 1,000ft.; South West Island is oval, about a 4-mile long and
S00 ft. high.

In attempting to describe any general collection of New Zealand
spiders at the present time, great difficulties are immediately encountered.
Although over 300 species have been described, the great majority of
the descriptions are worthless, and sometimes to identify even a common
spider would necessitate a revision of the group to which it belongs.
Accordingly, some of the following identifications are tentative, and in
some cases identification has not been attempted. Little is known of
the distribution of spiders in New Zealand, and more tends to be known
of the launae of small adjacent islands where special collecting has been
done than of the main islands themeslves,

The measurements were made by means of a micrometer eyepiece
and are given in millimetres. The sizes and distances apart of the eyes
are given in direct scale readings and so are comparative only. The
lifferent legs are denoted by Roman numerals and the leg indices are
obtained by dividing the length of the leg by the length of the carapace.
The tibial index, which gives a measure of the stoutness of the leg, 1s
obtained by dividing the combined lengths of the patella and tibia by
the diameter of the proximal end of the patella.

Two new genera and five new species are described, and the males
of two already known species are described for the first time. Twenty-
six species are represented in the collection, but identification 1s not
attempted in five cases either because of immaturity or because of the
difficulties already mentioned. The following is the list of species:

DIPLEURIDAE DYSDERIDAE
-lparita bipectinata Todd Ariadna bellatoria Dalmas
MIGIDAEF DRASSIDAE
Migas paradoxus LL. Koch Scotophocus pretiosus (1. Koch)

CLUBIONIDAL

’SECHRIDAE yy ari
PSECHRIDAI Clubtona pecnharis L. Koch

Matachia ramultcola Dalmas cm. ;
Chiracanthium insulare n. sp.
DICTYNIDAE THOMISIDAE

? Epiitectnus sp. Ihiaca albolimbata L. Koch

Rec. Auck. Inst. Mus. Vol, 4, No. 6, pp. 329-342, 25th October, 1956

330 MARPLES.

SALTICIDAE THERIDIIDAE
Trite auricoma (Urquhart) Armigera turbotti gen. et sp. noy.
[rite btmaculata (Urquhart) Moneta conifera (Urquhart)
Two other species Lithyphantes regius n. sp.

OXYOPIDAE Theridion veruculatum Urquhart
Oxyopes gregarius (Urquhart) Theridion longicrure n. sp.

eS gi TETRAGNATHIDAE
Lycosa hilaris L. Koch Tetragnatha flavida Urquhart
Lycosa sp.

AGELENIDAE EPEIRIDAE
Cambridgea antipodiana (White) Argiope protensa L. Koch
Gasparia nebulosa gen, et sp. nov. Epeira sp.

DIPLEURIDAE
Aparua bipectinata Todd

1g, 119. Recorded from the camp and depot area, Tasman
Valley, Castaway Valley, N.W. Cliffs, and in Maori Cave, important in
the soil.

This species was described from Wanganui. The average length
of the carapace of 12 females, the type and paratypes, is 4.7 mm., the
limits being 3.6 and 6.0mm. The specimens from Great Island are con-
siderably larger, the average of 10 specimens being 6.7 mm., with limits
6.0 and 8.3mm. In my own collection are four specimens from Cam-
bridge, average length 6.1 mm. and one from Houhora, in the North
Auckland peninsula, length 6.1mm. The numbers are too small for
certainty, but the suggestion of an increase in size on passing towards
the north is interesting.

MIGIDAE
Migas paradoxus L. Koch

3 2. In litter in the camp area. The species has been recorded
from Auckland, New Plymouth and Wellington.

PSECHRIDAE
Matachia ramulicola Dalmas. Text fig. 1.

4¢ and 1imm., 192 and 1imm. Collected in Maori Cave, in the
tent at night and by beating kanuka. The species was described from
specimens from Nelson and seems to be widespread, but the male has
not previously been described.

Male.—Length, 6.72 mm, Carapace pale yellowish brown, slightly
darker between the eyes and the thoracic groove, dark brown along the
anterior margin and with faint radiating pale streaks, Appendages and
sternum pale yellowish brown. Abdomen pale with brown markings.
Mid-dorsally are two parallel streaks, and posterior to these three large
and three very small chevrons with apices directed anteriorly. Antero-
laterally are spots which become streaks along the sides and merge into
a dark area on each side close to the spinnerets. Under side very lightly
spotted. All specimens similarly coloured,

Spiders from Three Kings. 331

Carapace: Length 3.04 mm, breadth 1.96mm. Low and smooth,
truncated abruptly in front. Thoracic groove longitudinal.

Eyes: 8, all pale. From above the anterior row is very slightly
recurved and the posterior row very slightly procurved. From in front
the anterior row is straight, the posterior row procurved. Ratio of the
sizes of the eyes and their distances apart: AM, 56; AL, 63; PM, 64:
PL, 68; AM-AM, 40; AM-AL, 101; AM-PM, 62; PM-PM, 94:
PM-PL, 108; L-L, 19; clypeus, 66.

Chelicerae: With boss. Long and tapering, the fang long and the
groove very oblique. Two minute teeth on the retromargin of the
groove and 4 on the promargin. Of these the next-but-one to the
proximal end is about three times the size of the others. The eTroove
is very slightly developed. A row of bristles parallel to each row of
teeth, the prolateral being much the larger. Anterior surface of cheli-
cerae with few or no bristles.

Maxillae: Long, with parallel sides, rounded anteriorly with the
median corner truncated obliquely,

Lip: Rectangular, the anterior border slightly concave. A little
more than half the length of the maxillae,

Sternum: Length 1.52 mm., breadth 1.16 mm. Rebordered, straight
anteriorly and ending posteriorly in a point between the hind coxae.
Lateral margins slightly indented opposite the coxae.

Palp: As in figure (text fig. 1). Tibia with bifurcated retrolateral
apophysis. One distal dorsal spine on the patella, 2 on the femur, also
a single one near the middle, Four trichobothria on the tibia,

I II IV IT] Palp
Legs. —§ ——_ Mm ——
4.95 4.01 2.78 2.59 1.38
Femur. Pat. & Tib. Metatarsus. Tarsus. Total.
Palp 1.62 0.96 —— 1.62 4.20
I 3.36 5.32 4.69 1.71 15.08
II 3.16 4.33 3.44 1.26 12.19
III 2.33 2.75 2.26 0.53 7.87
IV 2.42 3.11 2.04 0,90 8.47

Tibial Index I 7.5. Tibial Index IV 9.5.

Three claws, the paired ones with 10 pectinations, the median one
with 2 curved ones. The majority of the bristles on the legs and on the
body also are clothed with fine setules throughout their length. Tarsi
with spurious articulations. Very small tarsal organ distally situated.
Trichobothria: 6 in a row decreasing in size proximally on tarsi I and
II, 2 on IJ, 4 on IV. Similar rows occur on the metatarsi and tibiae.
Spines: all legs similar. Metatarsus, 2 pairs and 1 ventral at the distal
end, proximal half with 5 pro and 4 retro. Tibia, 1 distal dorsal, 2 pro
and 4 retro, 1 ventral. Femur, 3 distal, 2 dorsal, The tibia and femur
have rows of long hooked hairs. No calamistrum, only about half a
dozen straggly hairs.

Abdomen; Length 3,84 mm., breadth 1.79 mm. Numbers of hooked
hairs similar to those on the legs. Anterior median spinnerets large and
triangular, touching at the base. Posterior spinnerets largest, end joint

REVS MARPLES.

conical. Median spinnerets and anal tubercle small. Cribellum undt-
vided.

DICTYNIDAE

One immature Dictynid was collected under stones at the summit
of South West Island. From consideration of its cheliceral teeth it
clearly does not belong to the genus /veuticus, which is found throughout
New Zealand, but possibly to Epimecinus, which oceurs in Australia
and New Caledonia. A revision of the Dictynids of New Zealand at
present being carried out has already shown that some similar species
occur in the Auckland district, so no description of the present speci-
men is given here,

DYSDERIDAE
Ariadna bellatoria Dalmas

Z 2 and 1limm. From under stones and bark. This species was
described by Dalmas from a female from Taumarunui. Its clearest
distinction from A, barbigera Simon is the greater number of spines on
the anterior legs. The immature specimen, which may be a male, has
fewer spines, but this is insufficient evidence for the presence of both
species, especially as the male of neither has been described.

DRASSIDAE
Scotophoeus pretiosus (L. Koch)

1g and 1imm.,2¢@. In dry cave, Stella Rock and at the camp
area, Great Island.

CLUBIONIDAE
Clubiona peculiaris L. Koch

Ig, 3 2 and 4 imm. In camp area and collected by beating
kanuka in lower Tasman Valley.

Chiracanthium insulare n. sp. Text fig. 2.
1 é. Great Island.

Male.—Length 10.57 mm. Chelicerae, maxillae and lip chestnut
brown, Carapace chestnut brown anteriorly shading to paler brown
posteriorly. Sternum pale brown with chestnut margin, legs pale brown.
Abdomen pale greyish brown with two pale patches side by side
anteriorly, followed by 5 chevrons. Anterior end and posterior dorsal
patch pale brown similar to carapace.

Carapace: Length 4.49 mm., breadth 3.58mm. Smoothly domed
above with longitudinal groove, sides curved and an indentation above
the waist.

fives: 8, all pale. From above anterior row recurved, posterior
row straight. Width of eyegroup 1.57 mm, Ratio of eyes and their
distances apart: AM, 114; AL, 97; PM, 92; PL, 106; AM-AM, 70:
AM-AL, 58; AM-PM, 94; PM-PM, 162; PM-PL, 145; L-L, 71;
clypeus, 54,

Spiders from Three Kings. 333

Chelicerae: Stout and inclined anteriorly, the anterior surface some-
what geniculate and with bristles on the antero-median aspect. Groove
very oblique with 2 promarginal teeth at the proximal end and 2 retro-
marginal teeth, one proximal and the other near the base of the fang.
A. very pronounced swelling on the prolateral side close to the base of

the fang, from which a row of bristles extends along the promargin of
the groove.

Mazillae: Long, constricted in the middle, median anterior corner
truncated obliquely,

Lip: Long with sides converging. Truncated anteriorly with a
concave margin,

Sternum: Length 2.5limm., breadth 1.78mm. Oval and flat.
Margin with points opposite and between the coxae.

Palp: As in figure (text fig. 2). Tarsus not swollen, no back-
wardly directed process, tibial apophysis small.

Legs; Leg TV is missing on both sides of this specimen.

I II III IV Palp

3.24 2.73 Bide 0.93
Femur. Pat. & Tib. Metatarsus. Tarsus. Total.
Palp 1,69 1.55 —- 1.01 4.52
] 3.96 5.18 3.28 2.83 15.25
II 3.41 5.00 2.85 1.25 12.51
lI] 3.56 3.84 2.56 1.04 11.00

lV Lindo ee
Tibial Index I 10.0,

Two claws apparently not pectinated. Claw tuft. Narrow scopula
on tarsi and metatarsi I and II and on tarsus III. Spines: I and II,
metatarsus, 2 proximal ventral ; tibia 1 median ventro-prolateral; femur
1 distal prolateral, 1 median dorsal, III, metatarsus 3 pairs ventral;
tibia 3 pairs ventral, 2 retrolateral; femur 2 dorsal, 2 dorsal prolateral,
| dorso-retrolateral.

Abdomen; Length 5.60 mm., breadth 2.57 mm. Cylindrical, anterior
end chitinised and with long bristles, double row of bristles down the
dorsal side and others ventro-lateral. Six spinnerets, long and cylindri-
cal, terminal joints of the dorsal ones very short. Tracheal spiracle
appears to be close to the spinnerets.

——— a

This species is doubtfully put into the genus Chiracanthium. It
does not have the usual backwardly directed process of the cymbium
and it has a thoracic groove, but there are other members of the genus
exceptional in these respects. It differs from Chiracanthium stratioti-
cum 1. Koch in the absence of the process of the cymbium.

THOMISIDAE
Diaea albolimbata L. Koch

4 9 and2imm. Great Island, beaten from kanuka at the Saddle,
Tasman Valley and the east end. These specimens all show the pattern
of reddish markings lateral to white ones, as figured by Dalmas but
not by Koch.

334 MARPLES.

SALTICIDAE
Trite auricoma (Urquhart)
1 @. Maori Burial Cave area, Great Island.
Trite bimaculata (Urquhart)
2 4. Great Island.

The taxonomy of this family is in a most unsatisfactory condition,
some 47 species having been described. It seems undesirable to add to
this st until it has been revised. T’wo other species seem to be repre-
sented, each having 2 promarginal and 1 retromarginal teeth. One is
represented by 19 from the camp on Great Island, the other by 3¢,
22 and 1 imm. collected by beating kanuka on Great Island. This
resembles Jotus ravus (Urquhart) in general and in having a large lobe
on the bulb of the male palp which extends proximally for about the
length of the tibia. It differs, however, in the absence of a dorsal
scutum on the abdomen mentioned by Bryant (1935, p. 67) but not by
Urquhart (1892, p. 186), and in the presence of dark scales on the
ventral side of the tibia and patella of leg I.

OX YOPIDAE
Oxyopes gregarius (Urquhart)

1 ¢. Beaten from kanuka in lower Tasman Valley, Great Island.

LYCOSIDAE

Lyecosa hilaris L. Koch

1@ and 1imm. Great Island.
Lycosa sp.

limm, Great Island. This clearly does not belong to the previous
spce es, but 1s very immature.

AGELENIDAE

Cambridgea antipodiana (White)

12. Great Is!and, from under stone on Quadrat 1. In this speci-
men the median and lateral dark stripes on the carapace and the annula-
tions on the legs are very well marked, while the abdomen shows only
vague markings.

Gasparia gen. nov.

Size small. Chelicerae with teeth on both margins of the groove.
None of the spinnerets enlarged.

Gasparia nebulosa n. sp. Text fig. 3.
19. Under stone under litter. Quadrat I, Great Island.

Female.—Length 3.84mm. Carapace pale brown with dark
lateral bands about one-third the distance from the edge to the centre
formed of 4 or 5 coalescing marks. Dark round the eyes and a streak
at the fovea. Appendages brown, the legs with dark bands on the

Spiders from Three Kings. 335

femora, tibiae and metatarsi, most marked on IV scarcely visible on I.
sternum pale brown. Abdomen pale with dark pattern. Mid-dorsally
and anterior band followed by two small spots, followed by three
chevrons. A dark band on each side expanding posteriorly into a
mottled area reaching the chevrons. Ventral side pale with no marks.

Carapace: Length 149 mm., breadth 1.09mm. Low and smooth,
anterior and posterior margins straight.

Eyes: 8, AME dark. From above, anterior row slightly recurved,
posterior row straight. Ratio of the sizes of the eyes and their dis-
tances apart: AM, 76; AL, 114; PM, 107; PL, 123; AM-AM, 58:
AM-AL, 27; AM-PM, 116; PM-PM, 82; PM-PL, 72; L-L, 48:
clypeus, 65. Breadth of eyegroup 0.44 mm.

Chelicerae: Groove oblique, 1 tooth on the promargin, 3 on the
retromargin, the one nearest the fang being the largest.

Maxulae: More or less parallel, anterior median corner truncated.

Lip: Free, slightly longer than broad. Anterior margin slightly
concave.

Sternum,: Length 0.84 mm., breadth 0.71 mm. Smoothly rounded,
with blunt projection between coxae IV.

Palp: Claw without pectinations. On the tibia 1 dorsal and 3
dorso-rect-olateral trichobothria.

I lV II Ill Palp
Legs: Oo
25a 2.50 2.29 2.00 0.94

Femur, Pat. & Tib. Metatarsus. Tarsus. Total.
Palp 0.49 0.50 0.41 1.40
] 1.09 1.37 0.83 0.49 3,78
Il 1.01 1.20 0.69 0.51 3.41
III 0.81 1.04 0.60 0.51 2.96
[V 1.10 1.31 0.80 0.51 3.72

Tibial Index I 7.9. Tibial Index IV 8.5,

Three claws, the dorsal with 8 pectinations, the ventral with 2 which
are long, slender and curved. Trichobothria;: tarsus with a row of 4
increasing in size distally ; metatarsi I and II with 2, IJf and IV with 3;
tibiae I and II with 3, II] and IV with 5. Spines or stout bristles ; meta-
tarsus and tibia I and II each with 3 pairs, III and IV irregularly
arranged; 1 dorsal on each femur. Tarsal organ small, on I 15% of
the length of the tarsus from the distal end.

Abdomen: Length 2.24 mm., breadth 1.52 mm. Spinnerets normal,
posterior slightly the largest. Anterior separated by less than their
diameter. Tuft of bristles in place of colulus, Anal tubercle small.
Epigynum as in figure, a slight elevation in the centre and the brown
spermathecae showing through the surface.

THERIDIIDAE

Armigera gen. nov.

Male with dorsal and epigastric sclerites, and a less hardened one
surrounding the spinnerets. Stridulating organ at the waist. Cheliceral

336 MARPLES.

groove with teeth on both margins. Colulus present. One pair of book
lungs, and an unbranched pair of abdominal tracheae opening together
posteriorly. Female not known.

Armigera turbottin. sp. Text fig. 4.
2é. Under stones, Quadrat I, Great Island.

Male.—Length 1.56mm. Dark chestnut brown, appendages
slightly lighter.

Carapace: Length 0.87 mm., breadth 0.67 mm. High, with vertical
clypeus. Slightly concave posterior surface where overhung by the
abdomen. Carapace, sternum and the hard sclerities on the abdomen
with a uniform granular surface.

Eyes: 8, AME dark. From above, anterior row strongly recurved,
posterior row straight. Ratio of the diameters of the eyes and their
distances apart: AM, 105; AL, 94: PM, 105; PL, 103; AM-AM, 57;
AM-AL, 32; AM-PM, 70; PM-PM, 108; PM-PL, 66; L-L, 0: clypeus,
225. Breadth of eyegroup 0.37 mm.

Chehicerae: Small and vertical.

Maxillae: Broad, truncated anteriorly, with black serrula. Outer
margins parallel, inner margin converging but not meeting.

Lip: Free, more or less semicircular.

Sternum: Length 0.41 mm., breadth 0.43 mm, Heart-shaped with a
blunt posterior end widely separating coxae ]1V and joined to the cara-
pace behind them.

Palp: Slender, with large palpal organ, as in figure. The embolus
is long. On the prolateral side it is bent into a circle then turns back on
itself, passes between the bulb and the cymbium and turns again as in
the figure.

] I] IV III Palp
Legs: =§=-.-—-____—_
3.02 2.28 1.84 1.40 1.05
Femur. Pat.& Tib. Metatarsus. Tarsus. Total.
Palp 0.34 0.18 —— 0.34 0.95
I 0.89 0.9] 0.53 0.33 2.66
I] 0,62 0.68 0.38 0.29 1.97
III 0.39 0.36 ().25 0.21 1.21
IV 0.51 0.47 0.34 0.28 1.60

Tibial Index I 9.9. ‘Tibial Index IV 13.4.

Three claws with few pectinations. Few pectinated bristles, no
inore on IV, No spines, Trichobothria: 1 on metatarsi, 2 on tibiae.
Tarsal organ present, on [| 63% of the length of the tarsus from the
distal end.

Abdomen: Length 1.12 mm., breadth 0.90mm, The whole dorsal
surface covered by a single smoothly-domed sclerite. An epigastric
sclerite covers the anterior two-thirds of the ventral surface, lateral to
it on each side is a very small sclerite, and an annular one surrounds the

Spiders from Three Kings. 337

spinnerets. In the cuticle covering the remainder are small thickenings
forming longitudinal ridges, three of which pass dorsal to the spinnerets.

The skin of the other specimen was prepared by boiling in potash,
when further details were visible. The promargin of the cheliceral
groove has 2 teeth at the ventral end, the retromargin 3 smaller ones
about the middle. There is a well-developed stridulating organ, con-
sisting of the posterior slope of the carapace where it is overhung by
the abdomen. This is covered with fine transverse striations. Rubbing
upon it is a pair of small projections on the abdomen, each provided
with a dorsal bristle, arising on the part of the epigastric sclerite which
passes dorsal to the waist. The respiratory system consists of a pair
of book-lungs and a pair of unbranched tracheae arising close to the
spinnerets and confined to the abdomen. A small colulus is present
bearing two bristles. The sclerite surrounding the spinnerets, which
in the intact animal resembles the others, in the cleared skin is seen to
be much thinner.

It seems that this species may belong to the sub-family Pholcom-
matinae of the Theridtidae. Several genera formerly placed here have
been removed because they were found not to possess lungs, but more
study seems necessary to clear up the relationships. In the meantime
the present species 1s placed in a new genus.

Lithyphantes regius n. sp. Text fig. 5.
1¢. Great Island.

Female.—Length 3.90mm. Carapace, chelicerae and sternum
chestnut brown, palps and legs brown. Abdomen reddish brown with
irregular more or less interrupted mottled white bands round the edge
of the dorsal surface and in the mid-dorsal line. The mid-dorsal band
is continuous with the marginal one anteriorly and posteriorly.

Carapace: Length 1.66mm., breadth 1.30mm. Heart-shaped.
Thoracic groove, shallow, transverse, recurved.

Eyes: 8, all pale. From above anterior row strongly recurved,
posterior row straight, from in front anterior row straight, posterior
row procurved. Ratio of eyes and their distances apart: AM, 98;
AL, 137; PM, 117; PL, 119; AM-AM, 96; AM-AL, 107; AM-PM,
128: PM-PM, 115; PM-PL, 105; L-L, 32; clypeus, 200.

Chelicerae: Vertical. Fang stout. No groove, but one blunt tooth
prolaterally placed.

Mawyillae: Converging but not meeting over the lip.
Lip: Free, rounded anteriorly, broader than long.

Sternum: Length 0.82 mm., breadth 0.79mm. Heart-shaped with
finely grooved surface.

Palp: Claw with 5 pectinations. Pectinated bristles present and
1 trichobothrium on the tibia.

338 MARPLES.

I IV II III Palp
Legs .——§_ A
2.94 2.92 2.49 2.15 0.89
Femur. Pat.& Tib. Metatarsus. ‘Tarsus. Total.
Palp 0.51 0.47 —— 0.49 1.47
if 1.39 1,71 1.06 0.71 4.87
II 1,22 1.43 0.91 0.57 4.13
Ill 1.06 1,21 0.76 0.53 3.56
iV 1,39 1.75 1.03 0.69 4.86

Tibial Index I 8.3. Tibial Index IV 8.1.

Three claws, the paired ones with 10 pectinations on I and 5 on IV.
Tarsal organ small, situated 29% of the length from the distal end in I.
Pectinated bristles along the length of tarsus IV, a few only on the
other legs. No spines. Trichobothria: tibiae 3, metatarsi 1.

Abdomen: Length 3.16 mm., breadth 2.42 mm. Oval, overhanging
the carapace to about its middle. Spinnerets small. Colulus slender,
about half the length of the anterior spinnerets. Epigynum anterior to
the furrow. It has a pale transparent projection arising from a pale
ridged area through which was visible the dark spermathecae, The
ridges encircle the projection and its base, run transversely anterior and
posterior to it and form a whorl on each side.

This species differs from Lithyphantes lepidus in its smaller size
and general reddish instead of blackish colour, though the white mark-
ings on the abdomen are similar. The epigynum of L. lepidus, figured
by Dalmas, has a smaller projection, ridges arranged concentrically and
a median V-shaped chitinised structure. L. lepidus has only been
recorded from the South Island.

Moneta conifera (Urquhart). Text fic. 6.

1é and 1 imm., 79. Great Island, collected by beating kanuka.
Described from Waiwera, Te Karaka, Auckland Province, this species
is widespread. The male has not previously been described.

Male.—Length 3.30mm. Carapace, sternum and appendages
brown, abdomen pale mottled above and on the sides, grey below. Some
reddish brown streaks on the sides, especially posteriorly.

Carapace: Length 1.18mm., breadth 0.96 mm. Heart-shaped,
depressed above. In side view the outline is highest posteriorly and
slightly concave, rising again towards the eyes, Clypeus projecting
forward.

Eyes: 8, all pale, situated on reddish tubercles. From above
anterior row strongly recurved, posterior row straight. Ratio of eyes
and their distances apart: AM, 100; AL, 105; PM, 95; PL, 102:
AM-AM, 94; AM-AL, 59; AM-PM, 63; PM-PM, 103; PM-PL, 73:
L-L, 0; clypeus, 200.

Chelicerae: Small, apparently no teeth on the margins of the groove,

Lip: As long as broad.

Maxillae: Strongly curved inwards and almost meeting above the
lip.

Sternum: Length 0.81 mm., breadth 0.50mm. Flat, slightly rough

surface. Extends between the bases of the legs, and broadly between
coxae [V.

Spiders from Three Kings. 339

Palp: As in figure (text fig. 6). 1 trichobothrium on tibia.

- IV Il Ill Palp
Legs: = =—.
5,59 4.35 2.88 1.65 1.81
Femur. Pat.& Tib. Metatarsus. Tarsus. Total.
Palp 0.71 0.51 0.93 2.15
I 2.22 2.10 2.03 0.29 6.64
II 1.13 Peze 0.85 0.21 3.41
III 0.54 6.70 0.48 0.23 1.95
IV 1.71 1.53 1.62 0.63 5.49

Tibial Index I 13.3. Tibial Index IV 11.3.

Three claws on an onychium, dorsal claws with 2 or 3 pectinations,
apparently 1 small one on the ventral claw. ‘Tarsal organ one-third the
length of the tarsus from the proximal end. Tarsus IV with pectinated
bristles along its whole length, the distal ones on the metatarsus also
pectinated. Trichobothria: none on the tarsi, metatarsi I, Il and Itt
with a very large one, almost as long as the tarsus, at the distal end,
tibiae I, II and III with 3, tibia [TV with 4.

Abdomen: Length 2.16mm., breadth 0.95mm. Bluntly pointed
hehind, indented above the waist. No dorsal protuberance in this
specimen. 6 spinnerets subterminal, together with the anal tubercle
_ forming a rounded group. No colulus.

Theridion veruculatum Urquhart

34 and 2 imm., 32 and 4 imm. Beaten from kanuka on Great
Island. Said by Dalmas to be common in both North and South Islands.

Theridion longicrure n. sp. Text fig. 7.
14. Great Island.

Male.—Length 2.78mm. Pale greyish brown, area between and
behind the eyes as far as the fovea, brown, margin of carapace, grey,
Abdomen thickly mottled with white.

Carapace: Length 1.29mm., breadth 1.06mm. Low, rounded in
outline but slightly constricted behind the bases of the chelicerae.

Eyes: 8, AME dark. From above, anterior row strongly recurved,
posterior row slightly procurved. Ratios of the diameters of the eyes
and their distances apart: AM, 108; AL, 97; PM, 100; PL, 113;
AM-AM, 131: AM-AL, 85; AM-PM, 107; PM-PM, 108; PM-PL,
140: L-L, 0; clypeus, 170. Breadth of eyegroup, 0.48 mm.

Chelicerae: Promargin of the groove with a large tooth having a
small one at its base.

Mavillae: Twice as long as lip, margins more or less straight so
that the anterior angles are sharp. Wider distally, converging but not
meeting,

Lip: As wide as long.

Sternum: Length 0.69mm., breadth 0.76mm. Truncated where
it meets the lip with a well-marked indentation on each side at the base

of each maxilla. Extends between coxae IV to touch the carapace,
whose edge passes ventral to the waist.

340 MARPLES.

_ Palp: Slender, with small palpal organ, as in figure (text fig. 7).
[wo trichobothria on tibia.

i II I\ II] Palp
Legs: = =——————
5.72 3.75 3.43 2.34 (0).92

Femur. Pat,& Tib. Metatarsus. Tarsus. Total.
Palp 0,48 0.38 = 0.33 1,19
I 2.30 2.34 2.03 0.71 7,38
[I 1.5] 1.57 1.24 0.52 4.84
III 0.96 0.92 0.84 0.36 3.08
IV 1.49 1.40 1.16 0.38 4.43

Tibial Index | 7.6. Tibial Index IV 9.9,

Three claws. On I the proclaw has a single large pectination near
the tip, the retroclaw has six large pectinations and the median claw
has one small one near its centre. Below the base of the median claw
is a stout, blunt, upcurved projection. Few pectinated bristles except on
TV, where they are very lightly pectinated and extend along the tarsus
and on to the distal end of the metatarsus. No spines, but some stout
bristles on tibiae and patellae. Two trichobothria on tibiae I and II.
3 on tibiae II] and IV, On I the tarsal organ is 53% of the length of
the tarsus from the distal end.

Abdomen: Length 1.78 mm., breadth 1.9mm, Ovoid with the
spinnerets postero-ventral. These are conical, the anterior being the
stoutest. A colulus appears to be present but very small. A stridulating
organ is present at the waist, consisting of a striated area on the posterior
surface of the carapace and several small projections on the anterior end
of the abdomen.

A large number of species of Theridion were described by Urqu-
hart, the majority without figures or differential characters. The present
specimen does not seem to correspond with any of the descriptions. It
is provisionally placed in the genus Theridion, though it does seem to
have a minute colulus. It is notable for the disproportionate length ot
the first pair of legs.

TETRAGNATHIDAE

Tetragnatha flavida Urquhart

1¢,4¢@,. Great Island. Beaten from kanuka. One female is
much larger than the rest, but the details of the chelicerae resemble this
species and Dalmas says that its size is very variable.

EPEIRIDAE

Argiope protensa L. Koch
1¢,19. Great Island, beaten from kanuka.

Epeira sp. Text fig. 8.

Some 53 species of Epeira have been described from New Zealand,
the great majority on the basis of coloration, which is very’ variable,
but the actual number of species is clearly much less than this. Many
of the species described are unrecognisable, and, with some exceptions,
it is impossible to identify members of this genus until a revision has
been undertaken.

Spiders from Three Kings. 341

Fig. 1. Matachia ramulicola, (a) Ventral view, and (b) retrolateral view of

the right palp of the male.

Fig. 2. Chiracanthium insulare. (a) Ventral view of the left palp of the male.

(b) Retrolateral view of the right palp of the male.
Fig. 3. Gasparia nebulosa. Ventral view of epigynum.

Fig. 4. Armigera turbotti. (a) View of the right side of the male with legs
removed, the sclerites shown stippled. (b) Retrolateral view of the right
palp of the male.

Fig. 5. Lithyphantes regius. Ventral view of epigynum.
Fig. 6. Moneta conifera. Retrolateral view of left palp of male.
Fig. 7. Theridion longicrure, Retrolateral view ot right palp of the male.

Fig. 8. Epeira sp. (a) Ventral view of epigynum. (b) View at right angles to
“a” with the whole epigynum turned forward. The posterior end of the

scape is seen above.

342 MARPLES.

1¢,1992 and 6imm. Great Island. This species appears to be
the common if not the only Epeirid on the island. The total length is
about 8 mm. and the colour pattern variable, but mostly grey and brown,
The epigynum as seen in the figure consists of a smooth rounded
structure with a short spoon-shaped scape, which may be turned
freely forward. If the whole hard structure is turned more forcibly
forward it is seen to be somewhat cylindrical and to appear as in the
figure. This view is at right angles to the previous one and the posterior
end of the scape appears at the top of the figure. The central part con-
sists of soft white membrane, on each side of which is a tough sclerite
with dark wrinkled edges. This epigynum resembles Dalmas’ figure
(fig. 53, p. 387), which he tentatively attributes to E. venustulus Urqu-
hart, though his figure does not very closely resemble that of Urquhart
himself (1890, plate 21, fig. 12). Dalmas states, however, that in EF.
venustulus the antero-lateral tubercles of the abdomen are very striking,
while in the present specimens they are only slightly developed.

Three immature specimens from South West Island differ trom
the previous ones in having much more clearly annulated legs. ‘Their
abdominal colour pattern is, however, not unlike and they are probably
of the same species.

SUMMARY.

The collection consists of 26 species of spiders from the Three
Kings Islands, which lie about 35 miles north-west of the extreme
northern tip of New Zealand. Two new genera and five new species
are proposed as follows: Clubionidae, Chiracanthinm insulare ; Ageleni-
dae, Gasparia nebulosa; Theridiidae, Armigera turbotti; Lithyphantes
regius, Theridion longicrure. The males of the following species are
also described: Meatachia ramulicola Dalmas; Moneta conifera (Urqu-
hart).

REFERENCES.

BRYANT, E. B., 1935. Notes on some of Urquhart’s species of spiders. Rec.
Cant. Mus. [V., 53-70.

DALMAS, Comte de., 1917. Araignées de Nouvelle-Zélande. Ant. Soc. ent.
France., LXXXVTI., 317-430.

KOCH, L., 1871. Die Arachniden Australiens.

URQUHART, A. T., 1890. On some new species of Araneae. Jams. N.Z. Inst.,
XXIIT,, 128-189. |

URQUHART, A. T., 1892. Descriptions of new species of Araneae. Trans. N.Z,
| Inst., XXV., 165-190.

343

Notes on the Plumages and Breeding Cycle
of the Spotted Shag, Phalacrocorax (Stictocarbo)
punctatus punctatus (Sparrman, 1786).

By E. G. TURBOTT, Auckland Museum,

The spotted shag, Phalacrocorax (Stictocarbo) punctatus punctatus
(Sparrman, 1786), breeds in the Auckland area (Hauraki Gulf and
west coast), Cook Strait, south-west Nelson, Banks Peninsula and Otago
Peninsula. It is probably this form which breeds on islands off the
coast of South Westland, although the identity of these birds is not
yet certain. There were colonies formerly to the north of Auckland
(Bay of Islands) and at Cape Kidnappers.

The representative form in the Stewart Island area, and apparently
also in Foveaux Strait, is P, punctatus steadi (Oliver, 1930) (blue
shag), and at the Chatham group P. punctatus featherstoni Buller, 1873
(Pitt Island shag). Oliver (1955) regards Stictocarbo as worthy of
generic rank, and at the same time allots full specific rank to P. puncta-
tus featherstoni, because of the absence of the white line on the side ot
the head and neck found in the other two. However, the relationship
of P. punctatus featherstom to these geographically separated forms 1s
evidently so close that the present arrangement into three subspecies
is a more satisfactory means of expressing their relationship.

P. punctatus 1s considered to be related elsewhere only to the red-
footed cormorant, P, gaimardi: (Lesson, 1828), of South America
(Oliver, 1930a; Falla, 1932; Murphy, 1936). Oliver (1955) differs
from this view, as he considers that P. punctatus and P, gaunardi show
certain well-marked differences and are more likely to be derived
separately from the pan-antarctic subgenus Leucocarbo. Thus, if
Oliver’s view is correct, similarities between P, punciatus and P.
gamardi are due to convergence.

All three forms keep entirely to coastal waters, and the breeding
colonies are characteristically situated either on coastal cliffs providing
ledges, often with overhanging rock faces, or in the interior of caves.

Information on the breeding and plumage sequences of the spotted
shag is given by Falla (1932), based mainly upon material in the Auck-
land Museum collection from the Hauraki Gulf. It is now possible to
modify Falla’s descriptions in some particulars on the basis of additional
field information and material added to the collections from the Auck-
land area. Stead (1932) also includes in his general account observa-
tions on plumages made at colonies on Banks Peninsula.

An earlier and valuable description is that of Potts (1873), whose
detailed observations on breeding habits and plumages were made at
Banks Peninsula colonies. A general account by Oliver (1930a; 1955,
2nd ed.) includes further field notes. In the present account, no attempt
has been made to include information from the colonies other than those

Rec. Auck. Inst. Mus. Vol. 4, No. 6, pp. 343-363, 25th Ocrober, 1956

344 ‘TURBOTT.

Heur ek:
aoa
oy 9e Qtiritiré |.
MOSES 18.
Marrs
Beach
OAIA 1.4
BE THELLS
Pihe
Kerekers X

Text Fig. 1.
Location of breeding stations of the spotted shag.

The Spotted Shag. 345

in the Auckland area, except for a specimen from Banks Peninsula in
the Dominion Museum, and one from Queen Charlotte Sound in the
Auckland Museum collection, mentioned below.

I wish to express my thanks to Mr. O. Petersen and Mr. P. A. S.
Stein for permission to incorporate their invaluable field records which
form the basis of much of this paper. The fine series of photographs of
several different pairs at successive stages taken by Mr. Petersen has
enabled the plumage sequences to be very greatly clarified, and I am
particularly indebted to him for permission to include them here.

Others who have kindly provided field notes are Miss N. Macdon-
ald, Dr. R. A. Falla, Dr. W. R. B. Oliver, Mr. A. N. Breckon, Mr. J. C.
Davenport and Mr. R. Moynihan.

NOTES ON COLONIES IN THE AUCKLAND AREA.
a. East Coast Colonies.

In 1910, according to Falla (1932), the spotted shag had an
extensive distribution in the Hauraki Gulf, colonies being present at
the following points: islands off Tiritiri, the Noises, Rakino Island,
Waiheke Island, Shag Rock, and off Coromandel. After this date the
colonies suffered severely from shooting, until in 1932 Falla had evidence
of only one colony remaining on the western side of the Gulf. Buddle
(1951) refers to the destruction of spotted shags which he saw practised
some forty years earlier by shooting parties in the Hauraki Gulf. Ina
further comment, Falla (1940) states that “for many years the species
suffered constant persecution in common with all other species.” He
adds: “If anything, it suffered more, for its habits are such that it -
showed no ability to move from an area of persecution and establish
itself elsewhere.’’ Measures for halting the process of extermination
were taken in January, 1931, when an Order-in-Council giving total
protection to the spotted shag came into force.

Falla (1932) considered that by 1932 only one colony remained.
This colony, situated in a cave passing through the western shoulder of
one of the islets known as the David Rocks, in the Noises group, is a
well-established one at present. It has probably increased in numbers
since the above date, and nests now extend on to the rocky slopes of
the islet round the cave entrance.

A further colony can now be recorded on the Noises Islands. At
least within recent years, spotted shags have also bred on cliffs on the
north-eastern coast of the main island of the group, Otata Island, An
attempt to breed here is believed to have been made in the 1934 season,
but the birds suffered from shooting in spite of protection (pers. comm.
R. A, Falla). Sibson (1948) reported nesting on Ist December, 1946,
and found the colony “firmly established’? on 21st December, 1948
(Sibson, 1950). According to Davenport (1951), on 2nd April, 1950,
the colony consisted of approximately 40 birds, some on nests. It was
occupied on 4th April, 1954, and in late August, 1954, according to notes
from N. Macdonald and R. Moynihan respectively. Whether this
colony has been regularly occupied since 1934 is thus uncertain, but it is
of interest that none were present on 24th March, 1956, although the
birds had roosted in the area fairly recently (Turbott).

346 ‘TURRBOTT.

Records kept by Stein since about 1919, show that Falla was incor-
rect in his belief that only the David Rocks colony remained at the
height of persecution in 1931. Throughout the period of Stein’s
observations portion of the present colony situated near the eastern
point of Waiheke Island was recorded. This colony has at different
times consisted of three distinct groups close together on the same
rocky headland, although only two of these sites are at present being
used. The oldest portion occupies a cave at the northern end, and has
been observed since c. 1919 by Stein. In a second cave to the south
breeding was observed each season in the period 1924-1927, but not
afterwards until 1951 and 1954, when attempts at nesting were unsuc-
cessful. Situated roughly between these two, a third group breeds in
a relatively conspicuous position on a cliff face (the “Terraces” in
Stein's records). According to information from W. R. B, Oliver it
was in occupation in 1916, and a photograph published by Oliver
(1930a) was taken there. Stein records that in 1927, after the southern
cave was abandoned, a few nests were built on the “Terraces.” This
group gradually increased, and has been present every year, with the
exception of 1952; the greatest number recorded has been 90 birds.
References to recent observations at this colony are given below in the
section on breeding cycle in the Auckland area. It is located on the
rocky point between Hooks Bay and Anita Bay.

Apparently the largest of the early colonies was on Shag Rock
(Tarakihi Island), which lies a little over two miles to the south-east of
the east point of Waiheke Island. According to Stein’s observations it
consisted in August, 1914, of four separate groups on the north-eastern
cliffs of the island, and one additional group on the north-west corner.
Stein’s information continues that the birds suffered an attack in 1919,
resulting in almost complete annihilation. Small numbers continued to
breed until 1930, but the colony was then entirely deserted. In Decem-
her, 1923, breeding was first observed some three miles to the south at
the north-eastern end of Ponui Island, and it was Stein’s impression
that the stage of near extermination on Shag Rock was followed by a
period when breeding was being attempted at both localities. The
colony on Ponui Island appears to have increased gradually. Stein
counted 90 in nuptial plumage on a visit in August, 1949.

On Shag Rock attempts were made at nesting in the mid-1940s*,
and again at a new site in 1952, but all attempts were doubtfully success-
ful until December, 1955, when Stein observed about 40 nests on three
of the original cliff-sites. On a visit on 10th January, 1956, well-grown
chicks were seen.

Observations by Cox (1946) are of interest in showing continued
disturbance of the spotted shag in recent years in spite of protection
since 1931. Some twenty newly constructed nests were observed on the
north-east coast of Shag Rock on 11th September, 1945, but a month
later the nests contained egg fragments, and no birds were seen. Buddle
(1951) refers to the shooting of spotted shags at a colony in the Haurakt
Gulf as recently as 1950.

A habitat group of spotted shags in the Auckland Museum was
constructed by L, T. Griffin, using specimens obtained at Rakino Island

* Roberts (1943, ee) reported breeding 1942-44, and Cox (1946) in the 1945
season, as mentioned later,

The Spotted Shag. 347

and Shag Rock in 1914: the background by the well-known landscape
artist Kennett Watkins represents the early colony on Shag Rock.

On the eastern side of the Hauraki Gulf, there has apparently been
a colony continuously during the ahove period on Motutakapu Island,
one of the islands off Coromandel Peninsula to the south of Colville
(Motukawao group). A. N. Breckon observed the colony on this island
during the period 1920-30, and Stein has made regular observations
since 1952. During recent years at least, this has been a small colony,
according to Stein never containing more than 30 birds. Chicks were
seen in the nests on 9th April, 1955. The colony is situated near the
south-western point of the island, which is approximately 12 miles
distant from the nearest colonies in the western Gulf on Shag Rock and
Waiheke Island.

b. West Coast Colonies.

Three colonies are known, in an area which is somewhat different
from the Hauraki Gulf ecologically, the coast being exposed, with
generally a considerable swell and much heavy surf from the Tasman
Sea.

The largest colony is that on Thumoana cliff at Bethells (or Te
Henga), 25 miles by road west of Auckland. The colony was first
mentioned by Falla (1932), and has recently been observed regularly
by Petersen as described in the present paper.

Judged by telescope and binocular views from the mainland coast,
it has seemed probable that a colony existed on Oaia Island, which lies
nearly a mile offshore at the south end of Muriwai Beach, and about
34 miles north of Bethells. G. and A. T. Wightman have made landings
during the past five years, and according to their observations (Wight-
man, 1953, and Wightman, 1956) the colony, situated low down on the
north side of the island, contained nests with eggs on 11th November,
1951, and on 29th November, 1953, a nest with well-grown chicks.
However, there is still some doubt as to the numbers contained in the
breeding colony. The island is generally encircled by surf, often
extending to the mainland, and landing earlier in the breeding season
is likely to be difficult.

Spotted shags which are washed up fairly often on the west coast
beaches from Muriwai to Karekare are evidently from either the Bethells
or Oaia colonies (cf. Sibson, 1946).

The third colony on the west coast is about 50 miles to the south,
and is included here as it is the only other colony on the west coast of
the North Island. This is at Girdwood Point, a short distance to the
south of Kaawa Creek and about nine miles south of Waikato Heads.
The colony is situated both on the mainland cliff and on a tall basalt
stack (Cylinder Rock) separated from the mainland by a narrow cleft.
E. S. Richardson, the first to describe this colony, observed small naked
chicks on 20th April, 1946 (Turbott, 1947). Sibson (1952) gave an
account of a visit on 19th October, 1951, when the colony contained at
least 175 pairs of breeding birds, most of which had nests with eggs.

348 TURBOTT.

BREEDING CYCLE AND PLUMAGE SEQUENCES.

The spotted shag has a relatively long breeding season, and most
of the birds probably remain in the neighbourhood and return to the
colony to roost for a great part of the year. However, it has been
observed, or collected, at a considerable distance from any of the breed-
ing stations, and movements from the breeding areas are at present
mainly unknown. According to Stead (1932), parties of adults and
young go on journeys up or down the coast after the breeding season,
and they reach Motunau Island, some 50 miles north of the Banks
Peninsula colonies,

There may also be a considerable lag in breeding at any one colony,
with differences of some weeks between laying, and possibly more than
one breeding peak.

The plumage sequences include the seasonal assumption of crests,
decorative plumes and a distinctive head pattern. The terminology used
is that adopted by Murphy (1936), whose discussion has greatly clari-
fied the plumage changes in other southern hemisphere shags—
Phalacrocorax atriceps, Phalacrocorax albiventer, Phalacrocorax magel-
lanicus—in the South American region. According to Murphy, in P.
atriceps the pre-nuptial moult occurs in the late southern summer, so
that “breeding” plumage is at its height by the end of June. At this
time (June-July) the gonads also begin to enlarge. With actual mating
and egg-laying, a prominent feature of this plumage, the crest, is lost
and the birds enter upon a distinctive stage, nuptial plumage proper.
In P. atriceps a feature of the latter plumage is the appearance of a
white alar bar. Murphy (p. 885) states that “the height of plumage
should not be called a ‘breeding’ garb, but rather a pre-nuptial plumage,
at its best during the early part of the rather lengthy courtship which
precedes the nesting season,” while the later stage, when the eggs are
laid, is more appropriately termed nuptial plumage.

The next phase, according to Murphy, comes at the annual (or post-
nuptial) moult, which results in the replacement of both the quills and
body plumage, accompanied in P. atriceps by the appearance of a dis-
tinctive patch of white feathers on the back. This stage should be
termed the post-nuptial, for as Murphy points out (p. 884) it “begins
early during the nesting period as with most other cormorants”: it is
thus misleading to use the term “winter plumage” for this stage, which
appears in the late spring and summer. The term “non-breeding plum-
age” must also be avoided, as the plumage is assumed while the nesting
season is still in progress.

The use of the terms pre-nuptial, nuptial and post-nuptial sug-
gested by Murphy for the seasonal plumages of the adult is most helpful,
and has been adopted in the following account. Table 1 shows the
phases of the breeding cycle and adult plumages, derived mainly from
Petersen's notes and photographs of the spotted shag at Bethells. The
information obtained by this observer is the most complete so far avail-
able, but differences in the annual cycle at other breeding stations are
discussed later in this account.

The Spotted Shag. 349

Table 1. Spotted Shag: Breeding Cycle and Adult Plumage at Bethells.

White head and
neck line

Plumage is
termed:

Decorative
plumes

Season Crests

Pre-nuptial

(begins April-May) present maximum present PRE-NUPTIAL
Egg-laying re. ‘abraded and Py, we
(begins August) Busear sche moulted PiSEent NUPTIAI,
Reari ;
pees Se aia 3 absent lost obscured | POST-NUPTIAL.

(begins September)

The above sequence is similar to that described by Murphy (p. 884)
for P. atriceps, which is given in Table 2 for comparison (plumage
terms have been added).

Table 2. Phalacrocorax atriceps: Breeding Cycle and Adult Plumage,
according to Murphy

Plumage is

Season : Crest Alar bar Dorsal patch bate ad
Pre-nuptial gage Be absent absent PRE-NUPTIAL.
Egg-laying ohne =i present bung hte - NUPTIAL rc
Rearing of young lost A er. ya maximum - POS ENUPT ON

Note.—Of the two other South American species referred to
earlier, P. albiventer is characterised in nuptial plumage by an alar bar,
but in post-nuptial plumage this species has no white dorsal patch. In
P. magellanicus the whole of the chin, throat and foreneck is black in
pre-nuptial and nuptial plumage, but becomes white in post-nuptial, The
crest is lost in both species before the post-nuptial plumage is assumed.

The following is an outline of the details of the breeding cycle, and
corresponding plumages, observed at the Bethells colony by Petersen
during 1952 to 1955. According to these observations, pre-nuptial
plumage is assumed as early as the last week in April, and many birds
have been observed in full plumage in May. There may evidently be
some lag in breeding activity, as birds wearing their decorative plumes,
and apparently mated, have been seen as late as 1st September, In the
same part of the colony nests may contain eggs when adjacent nests
have young almost ready to leave, and the last young may not be ready
to fly until the end of February.*

The chief characteristics of the pre-nuptial plumage (Fig. 1) are
the two fully-developed crests, the black throat and, in contrast, the
wide line of white on either side of the head and neck. In addition, an
often profuse decoration of narrow white plumes is scattered over the
nape, back, rump and flanks.

In July, nest-building generally begins in earnest: the birds are
seen carrying long sections of seaweed, or green cliff herbage, and the
colony generally contains well-built but empty nests.f Until this time
they sit about the colony with a certain amount of courtship.

* The seaward (western) portion of the colony is the earliest to start, according
to Petersen’s records, and is from three weeks to a month more advanced
than the more inland (eastern) portions, occupying the inner cliffs.

+ Carrying of nest material was observed in the first week in June at Bethells
(Turbott, 1946).

350 TURBOTT.

The transition from pre-nuptial into nuptial plumage is character-
ised by the loss of the decorative plumes and the reduction of the crests
by wear and moult. The period during which the plumes is retained ts
very variable, according to Petersen’s observations, and apparently they
are not developed at all in some birds. In many birds they are lost after
only a few weeks, but they may remain for much longer and even persist
very exceptionally until the eggs have been laid. The crests generally
become much reduced before egg-laying, and disappear soon afterwards.
There is thus a period of perhaps two months (June-July) when most
birds have lost the decorative plumes, but are characterised by the black
throat, white stripe on head and neck, and distinct but diminishing
crests (Figs. 2-4) : the nuptial plumage.

The earliest eggs observed were found in the second week in
August, but laying may evidently begin even earlier, as a chick was
found just hatched on Ist September. Incubation takes over four, and
less than five, weeks, according to Petersen’s observations. Both
parents incubate, and the change over was observed on several occasions.

With the laying of their eggs the birds enter upon the post-nuptial
moult, according to Petersen’s observations. By the time of hatching
both parents have changed into post-nuptial plumage. As mentioned
above, the crests are soon lost, and brown or greyish-brown flecks
appear on the white line as early as egg-laying, gradually increasing
until hatching. The black throat becomes mottled with grey and white,
or may even become a featureless pale grey or almost white (see below
under “adult plumages”). The dark line down the mid-line of crown
and hind-neck also becomes greyish brown, The effect of these changes
is to reduce the tone of the striking head and neck colouration by replac-
ing it with more or less mottled grey and white, but the rate of change
varies greatly, and the degree of pattern on the head and neck varies in
post-nuptial plumage (Figs. 6-10).

The fledging of several chicks took about nine weeks, at which
stage the young were observed to fly. The earliest record for a newly-
hatched chick was Ist September, and the latest was the third week in
December. On the more open slopes, the fledglings tend to clamber
about away from the nests when only six to seven weeks old, but on
narrow ledges cut off from the rest of the colony the entire fledging
period of individual chicks could be observed. At this stage the most
notable feature of the colony is the resemblance superficially between
the young in their juvenal plumage and the post-nuptial adults, especially
those in the “greyer’ type of plumage. The juvenals can be distin-
guished by the brownish lower back and flanks, white under tail-coverts
and, at close range, by the pinkish facial skin. The spots on the back
are much less pronounced than in the adult.

The following are notes on material in the Auckland Museum,
chiefly from the Hauraki Gulf, where collections were made some years
ago, and from west coast beaches, where more recently a series of
storm-killed birds has been obtained. In addition to the record of plum-
age sequences by Petersen, information is included from my own field
notes.

Juvenile Plumages: Nestling

The naked chick has a dark lead grey skin; in a chick examined at
the David Rocks, Hauraki Gulf, on 11th May, 1935, the first down was
just appearing as a sparse coat, dark above and whitish below (Turbott).

The Spotted Shag. 351

The first nestling down, which is described by Oliver (1955) as
“dark brown above, whitish below,” is shown in Petersen’s photographs
of the younger chicks (Fig. 9). There are several specimens, all appar-
ently of about the same age, in the Auckland Museum collection (Nos.
AV. 96.6, included in exhibition group; 96.106 and 96.107). A note-
worthy feature is the brownish face and under-parts. However, in
AV. 96.107 the face is white, and the head pattern in this down may well
prove to vary considerably. It seems probable that the under surface
gradually becomes whiter as the second down is acquired.

The second down is shown by a fledgling in the collection, AV.
96.101, found on Bethells Beach, 8th December, 1951. In this there is
a very distinct demarcation between the ashy-brown of crown, hind-
neck and dorsal surface, and the under surface, which is white including
the face to the level of the eye and the fore-neck. Only a trace of brown
appears on the breast and abdomen. The quills and scapulars are well
grown, Small filoplumes are scattered through the down on the hind-
neck, The bare facial skin and gular pouch, which are pinkish in life
in both fledglings and juvenals (Turbott), were a faint greenish-blue in
the fresh specimen. The feet appear flesh-coloured in life, but in the
fresh specimen had faded to a dull yellowish grey; the tarsus, outer toe
and webs with brown shading. Fig. 10 shows nestlings at this stage.

Juvenal

As already mentioned, this can be recognised fairly readily by
several characters in the field, the most satisfactory being the whitish
under tail-coverts only faintly shaded with brown. The under tail-
coverts and lower half of the abdomen in later plumages are dark grey
to deep greenish black.

A juvenal, AV. 96.102 (Bethells Beach, 8th December, 1951) has
the upper parts and thighs grey brown, faintly glossed with green on the
lower back, upper tail-coverts and thighs; and the wing and tail quills
dark brown. The chin is white, merging with the pale brown of the
sides of the head, fore-neck and upper breast. The abdomen and under
tail-coverts are white lightly shaded with brown. The feathers of the
mantle, scapulars and wing-coverts have greenish-brown tips, the spots
so formed being relatively faint. Filoplumes are present on the hind-
neck and thighs. In this specimen the coloration of the facial skin and
feet was similar to the fledgling described above (AV. 96.101), but there
was more brown shading on the feet. Petersen’s photographs also show
this plumage (Fig. 11).

Oliver (1955) gives a detailed description of the juvenal plumage
under the heading “immature,” but states that there is a “stripe on side
of neck mottled ashy brown and white,” a characteristic of sub-adult or
adult post-nuptial plumage.

Sub-adult

Information at present remains inconclusive on the later immature
plumages and little can be added without larger collections, or further
field work based upon ringing. It is doubtful whether classification as
immature on labels, based on the condition of the gonads, is reliable.
The labels have thus not been followed except when confirmed by plum-
age characters in the following material.

352 TURBOTT.

According to Stead (1932) the spotted shag breeds at the end of
its first year, a conclusion based on the presence of a group of birds in
the colony moulting into “mating plumage” some two months later than
the main colony. These birds, which occupied an area near the colony,
were considered by Stead to be the young of last year’s breeding season.
Some individuals were still in full juvenal plumage, and according to
Stead were no doubt “late-hatched ones of the previous year.”

The presence of a similar group of presumably sub-adult birds
towards the outskirts of the colony, moulting into pre-nuptial plumage
at a late stage in the nesting season, was noted by Petersen at Bethells.

Falla (1932) stated that the juvenal plumage is followed at the
next moult by a stage which, according to his description, corresponds
in most characters to the adult post-nuptial plumage. The description
includes the statement: “The stripe down the side of the neck remains
mottled ashy brown and white until the assumption of the nuptial orna-
ments consisting of frontal and occipital crests, and extensive series of
eloneated white feathers of fine texture.” However, the two specimens
(AV. 96.3 and 96.72) upon which the description is based are both of
the type with a fairly definite pattern on the head and throat, and show
no <haracters to distinguish them from the strongly-patterned type of
adult in post-nuptial plumage. Falla also examined two specimens
moulting out of juvenal plumage (AV. 96.5, 96.71).

The series in moult described below shows that the post-juvenal
moult, when the birds probably pass into a “post-juvenal’ plumage,
eccurs not long after the breeding season. The most advanced of the
specimens (AV. 96.84, 96.90, 96.108) have almost completely changed
into a plumage which is indistinguishable from the “greyer” type of
post-nuptial adult. However, the material is insufficient to indicate
whether this coloration is constant at the post-juvenal stage, so that at
present no distinction can be drawn between this stage and the corre-
sponding adult plumage (adult post-nuptial). Stead’s observations sug-
gest that this is followed by a “first pre-nuptial plumage” which is
apparently indistinguishable from adult pre-nuptial plumage.

Specimens in the Auckland Museum collection at some stage of
moult out of juvenal plumage are as follows:

No. AV. 96.5 Bay of Islands; 1896. Throat and fore-neck
lightly shaded with grey. New spotted plumage on mantle and
scapulars, mixed with some faded brown feathers. Black upper
tail-coverts appearing at base of tail. Flanks brown. Under tail-
coverts still whitish, but a group of dark grey feathers at the vent.

AV.96.71. Female; Thames; 29th June, 1882. New spotted
plumage on mantle and scapulars, with some faded brown feathers.
Black over base of tail, but otherwise in juvenal plumage.

AV. 96.84. “Juv. male”; Muriwait Beach; 12th May, 1933.
Faint grey shading on throat and fore-neck. Mantle and scapulars
almost completely replaced; greenish-black feathers scattered over
hind-neck, lower back and rump, above base of tail, and a few on
the flanks. Under tail-coverts white, with scattered dark grey
feathers. Two new black middle tail quills; remaining quills much
faded.

The Spotted Shag. 353

AV. 96.90. Immature female; Muriwat Beach; 27th August,
1935. As adult post-nuptial plumage, except for the much-faded
wing-coverts where replacement is beginning, some remaining
brown feathers on the flanks, and two faded tail quills. The throat
and fore-neck are lightly shaded with grey, as in AV. 96.84. The
lower abdomen and under tail-coverts are dark grey, darkest on the
under tail-coverts. The whole remaining under-parts are suffused
with pale ashy grey. It will be noted that, but for juvenal char-
acters recorded above, this specimen might be regarded as a post-
nuptial adult of the palest type quite frequently observed amongst
breeding birds.

Av. 96.103, Female; Muriwai Beach; 18th May, 1951. Much
like AV. 96.5 and 96.84, but fewer feathers replaced on the mantle
and scapulars; less advanced than AV. 96.84 on lower back and
flanks.

AV. 96.108. Muriwai Beach; 21st May, 1934. Like AV.
96.84, but all the upper-parts replaced except the faded wing-coverts
(replacement beginning). A good deal of brown remaining on the
flanks. Lower abdomen and under tail-coverts almost replaced by
dark grey, but scattered white feathers remaining in front, In
addition to the two middle tail quills, two side quills are being
replaced. The under-parts suffused with pale ashy grey as in

AV. 96.90.

AV.96.110. Piha Beach (West Coast) ; 12th February, 1956.
This specimen was decomposed and only a wing was kept. It was
at an early stage of the post-juvenal moult, having new black-tipped
feathers under the juvenal feathers on the mantle and scapulars.
Wing coverts and scapulars faded to pale greyish brown.

All the above specimens show, in addition to the characters
described, a faded region on the wing coverts, and, in the less advanced
specimens, on the scapulars. This marking is prominent in the field and
is referred to by Stead (1932) as “pale sandy yellow.” When the
juvenal plumage is first acquired the wing coverts have dark-brown
spotted tips as noted above, but the soft feathers of this plumage tend to
wear rapidly at the tips. This is accompanied by fading, especially on
the scapulars and wing coverts.

In two of the above specimens (AV. 96.90 and 96.108) the lower
abdomen and under tail-coverts have changed to dark grey. This colora-
tion may be an immature character, but it should be noted that a breed-
ing bird photographed by Petersen (Fig. 3) shows dark grey between
the thighs. There is also a variable amount of dark grey, often mottled
with black, on the anterior portion of this region in specimens which are
otherwise in adult plumage, and in some as little as a trace of dark grey
on the black, The latter include both sexes in nuptial and post-nuptial
plumage.

The following specimen (the only one in the collection from this
locality ) is included tentatively under this heading:

AV. 96.78. Immature; Queen Charlotte Sound; 26th August,
Plumage like the adult post-nuptial, having a white stripe on head
and neck, flecked with brown; throat dark grey faintly mottled

354 TURBOTT.

with pale grey and white; greenish-black feathers appearing on the
crown -and hind-neck, and well developed on the anterior mantle.
Crests sprouting beneath the feathers. Filoplumes present on hind-
neck and back, and abundantly on the flanks. .

This is possibly a sub-adult, as the greater wing-coverts are faded,
although the remaining wing-coverts and scapulars have all been
replaced. In addition, it is evidently in the process of changing into
pre-nuptial plumage. By this season (26th August) at Bethells, only
young of the previous year would be at this stage; and at the Noises
Islands, Hauraki Gulf, the pre-nuptial moult would be mainly earlier
in the year. It might, however, be a post-nuptial adult from a later
breeding colony (such as that on eastern Waiheke Island). Perhaps
some fading occurs during the later stages of the adult post-nuptial
moult, |

Observations were also made on a visit to the David Rocks (Noises
Islands) colony on 3th March, 1956. A bird changing out of juvenal
plumage observed closely showed the much abraded juvenal feathers on
the mantle and scapulars, amongst which a dozen fresh grey feathers
with black tips were conspicuous. There were fresh dark feathers above
the base of the tail, but otherwise the back and flanks were apparently
still brown; the tail quills were partly replaced. The under tail-coverts
were whitish, and the head and neck coloration like the juvenal. The
bare facial skin in this sub-adult bird was chrome yellow, whereas in
post-nuptial individuals observed at the same time, and presumed to be
adults, the skin was yellow distinctly tinged with olive-green. The feet
were mainly dark brown, slightly pinkish. (Turbott.)

Adult Plumages: Pre-nuptial and Nuptial

There is a series in various stages of pre-nuptial and nuptial plum-
age in the collection. The following is a brief description of the plumage
at these two stages: The side of the head, throat and fore-neck are dull
black, with a faint green gloss; above this a line of pure white extends
down the side of the head and neck, and continues but becomes narrower
between the mantle and fore-neck, until it reaches the wing. The crests,
crown and a narrow line on the hind-neck are black, with a brilliant
green gloss; glossy greenish black extends on to a wider zone on the
anterior portion of the mantle, and there merges with the spotted dorsal
plumage. The posterior mantle, scapulars and wing-coverts are ashy
brown, each feather bearing a greenish black terminal spot; the wing
quills are brown with a grey bloom. The lower back, rump and upper
tail-coverts, continuing below to the flanks, under tail-coverts and lower
abdomen, are glossy greenish black. This meets the abdomen at a dis-
tinct line across the body, linking the thighs*. The breast and upper
abdomen are pale silver-grey. In pre-nuptial plumage, white decorative
plumes may almost completely hide the dark mid-line of the neck, and
sometimes occur on the mantle, back and rump, and on the thighs.

Notes on a specimen in nuptial plumage (AV. 96.105) state: “feet
vellow, soles (under toes) dark brown; bill horn colour, brown on top
of culmén” (Turbott).

—_ . ~ ee en, - es Se

* Variations in the coloration of this region are mentioned under “sub-adult.”

The Spotted Shag. 355

The bare facial skin and gular pouch in pre-nuptial plumage are
generally described as dark blue (“rich royal blue,” according to Stead)
and the ring of beads round the eye greenish blue (for the general
appearance of the facial characters see Fig. 1). As already mentioned,
birds in post-nuptial plumage, probably adults, observed on the David
Rocks on 3rd March, 1956, had olive-green to yellow skin on the face
(Turbott). Field notes are recorded by Falla (1932) for a post-nuptial
in which the skin in this region was ‘“‘viridine green.” It thus seems
probable that the bare skin changes to bluish green, and finally to green-
ish yellow during the transition into post-nuptial plumage, but confirma-
tion could be obtained by field observation in the breeding season.

__ The following specimens in pre-nuptial and nuptial plumages are
included in the series in the Auckland Museum:

No. AV. 96.1. Male; Rakino Island; 1886. Nuptial.
AV. 96.4. Rakino Island, 1887. Pre-nuptial.

AV. 96.83. Female; Muriwai Beach, 12th June, 1933. Moult
is still in progress on the head and mantle, but this specimen is
almost in full pre-nuptial plumage.

AV.96.105. Female; Muriwai Beach; 15th July, 1957.
Apparently in full pre-nuptial, although there are only a few scat-
tered white plumes on hind-neck and thighs. Two side quills
sprouting in the tail (cf. Murphy, ibid., with regard to the pro-

tracted period required for the replacement of the quills in P. atri-
ceps, moult still showing in some specimens as late as April).

“The observations made at Bethells by Petersen show evidence of
the change out of nuptial plumage soon after egg-laying. The crests
have been lost and the white lateral line obscured by the time the eggs
hatch. However, Petersen’s photographs taken at a considerably later
stage, some two months after laying, provide some evidence that the
period taken to moult the black throat may be variable. By this time,
the moult of these black feathers is complete, or almost complete, in
some individuals (Figs. 7 and. 9). However, in a number apparently
mainly with the more pronounced type of head pattern (Figs. 6 and 8),
there are still plentiful scattered black feathers on the throat. These
appear to be wholly lost before the next pre-nuptial moult, as specimens
of this strongly-patterned type changing again into pre-nuptial plumage
are included in the collection, and.show fresh black feathers on the
throat, which is otherwise mottled dark grey and white.

It 1s not possible at present to distinguish between the sexes in
the field. -Petersen found that, of the three pairs which he photographed,
one bird appeared to be smaller and was more confident at the nest. In
each pair, the smaller was the duller in nuptial, and of the “greyer” type
in post-nuptial, and was considered to be the female. However, the
specimens in post-nuptial plumage mentioned below include both males
and females with both types of pattern.

The principal changes marking the transition to post-nuptial plum-
age are: (a) The appearance of brown or greyish brown feathers as
flecks in the white line. These occur along the whole length of the
white line, and may even be so numerous that the line is practically
obscured (Figs. 6 and 8). They are present in both sexes in the collec-
tion, and persist in specimens which are moulting again into pre-nuptial

356 ‘TURBOTT.

plumage. In the “greyer” type of post-nuptial individual the line is
indistinct, as the greyish brown on crown and hind-neck becomes merged
into almost the same shade on the side of the head and neck (Figs. 7 and
9). (b) Mottling of the throat region, again with considerable variation
ranging from individuals with pronounced mottling to those showing
the extreme “greyer” coloration. In the former, the throat is uniform
dark grey, or mottled dark grey and white, in the post-nuptial plumage ;
black feathers may be present, but, as mentioned above, it is uncertain
how long these are retained. The “greyer” type is a featureless pale
grey with a white chin in extreme cases, but the chin and throat are
quite commonly a darker shade of grey, mottled more or less with white.
As shown by a specimen (AV. 96.89) included below, replacement of
the black throat starts on the chin round the base of the bill, and Oliver
(1955) has a photograph (p. 231) of a bird, wrongly stated to be
“immature,” which is evidently at about this stage.

The following specimens illustrate several stages of transition to
post-nuptial plumage:

AV.96.89, Female, Muriwai Beach; 7th September, 1936.
In nuptial plumage, but with a few scattered brown flecks on the
dorsal half of the white line, and dark grey feathers appearing on
the black chin at the base of the bill.

AV. 96.3. Female; Whangaparaoa; 5th September, 1896.
Brown flecks present on the white line. Grey, white and black
feathers on the throat. On the crown, hind-neck and mantle there
are still greenish black feathers in the new grey-brown plumage.

AV. 96.72, Waiheke Island. Throat dark grey, mottled with
white. Crown and hind-neck greyish brown, white line indistinct
and narrow. Some dark greenish black feathers on the lower hind-
neck and mantle.

AV. 96.77. Male; Noises Islands; 30th May, 1932, Similar
to AV. 96.72, but the throat is dark grey only faintly mottled with
pale grey and white. Filoplumes are numerous on the head, and
there are a few short decorative plumes on the hind-neck. Although
collected at such an early date, this bird appears to have performed
the post-nuptial moult, but breeding begins considerably earlier at
the Noises Islands (David Rocks) than at Bethells (see below).
It may be coming into pre-nuptial moult, although apart from the
dark feathers on the hind-neck and mantle no signs of this are yet
present.

Through the courtesy of Dr. R. A. Falla and Mr. C. McCann a
further specimen has been received on loan from the Dominion Museum,
Wellington (DM. 1837). It is a female collected off Lyttelton, 28th
January, 1926, by R. H, Beck, and is of special interest as it is in the
palest “greyer” type of post-nuptial plumage, and shows a sprinkling of
black feathers in the grey plumage on either side of head and fore-neck.
There 1s no evidence of moult, and, as suggested above, these black
feathers have apparently persisted until a late stage, but would probably
be lost before the next pre-nuptial moult.

Post-nuptial

The collection includes two specimens in the “greyer” type of post-
nuptial plumage. The throat in both specimens is mottled with dark grey

The Spotted Shag. 357

and white, forming an indistinct grey shadow. The crown and hind-
neck are greyish brown, with a faint green gloss, darkening only slightly
on the mantle. These specimens are:

AV. 96.79. “Immature” male; Karekare Beach (West Coast) ;
26th February, 1953. (Labelled “immature” but there is no basis
for this on plumage characters. )

AV. 96.82. Female; Karekare Beach; 4th April, 1933.

In addition, the following specimens are evidently in the process
of changing into pre-nuptial plumage:

AV. 96.80. Female; Noises Islands; 28th March, 1933. Crests
well-developed, and perhaps still growing. New dark feathers
growing and already abundant on crown, hind-neck and mantle.
Brown feathers still sprinkled in the already distinct white line.
Filoplumes on head and neck, including throat; some sho't decora-
tive white plumes on the hind-neck and flanks. The chin and throat
still show some grey and white feathers, but many new black
feathers have appeared, giving a dark grey effect generally, A side
tail quill sprouting and the two central quills only half grown.

AV.96.87. Female; Hauraki Gulf; 30th April, 1935, Like
96.80, but more black feathers have appeared on the chin and throat,
which has a generally darker appearance. The crests are not so
strongly developed.

AV. 96.93. “Immature”; Muriwai Beach; 12th April, 1942.
The white line is much less evident than in the two specimens above,
being still almost obscured by brown feathers; new dark feathers
are less abundant on the crown, hind-neck and upper mantle. The
crests are short, but sprouting. There is only a sprinkling of black
feathers on the chin and throat mixed with the grey and white. A
few filoplumes on the crown and hind-neck, and scattered short
white decorative plumes on the lower back and flanks. Some faded
brown quills in the tail, but mainly replaced. (Labelled “immature”’
but see AV. 96.79. )

AV.96.94. Female; Muriwat Beach; 12th April, 1942.
Resembles AV, 96.93 closely, but decorative plumes almost absent
-on the lower back ; the white line is less heavily flecked with brown.

BREEDING CYCLE IN THE AUCKLAND AREA

As Petersen’s observations at Bethells show that the colony there
follows a fairly regular annual cycle, in the present section this cycle is
compared with the information available from other colonies in the
Auckland area.

Unfortunately, there are only somewhat scattered observations on
colonies at the Noises Islands, and the large amount of field work by
Stein refers mainly to the eastern Waiheke Island colony, where breed-
ing is evidently exceptionally irregular. The opportunity is taken to
include in this section some additional notes on the Bethells colony.

At the Noises Islands, the colony situated for so long 1n and out-
side the entrance to a cave at the David Rocks seems to have a generally

358 "TURBOTT.

earlier breeding season than the one at Bethells. Falla (1932) found
that “‘the cavern-dwelling colony that alone now represents the species
on the western side of the Hauraki Gulf has for some years past been
breeding in mid-winter.”” He says: ‘‘Full nuptial plumage is present in
most of the birds in May, by the end of which month in 1932, about
twenty new nests of green Mesembryanthemuin australe had been built
up, but no eggs laid.”’ However, the season may be prolonged until
October, eggs having been observed in this month, and the season thus
extends almost as late as the Bethells colony.

Some 200 birds were counted when the colony was visited on 3rd
March, 1956, but examination of the cave showed that nesting had not
yet started. In addition to the juvenals mentioned earlier, several adults
were seen with the white stripe already distinct on head and neck, and
the position of incipient crests could be seen in one bird observed at
close quarters (Turbott). It is worth noting that this is approximately
one month before the first birds assume pre-nuptial plumage at Bethells
according to Petersen’s observations. Ona visit to the latter colony only
six days earlier, on 27th February, none had been seen changing plum-
age (Turbott).

The earliest breeding record at the David Rocks was made on 11th
May, 1935, nearly three years after the account by Falla (1932). The
colony was then low in numbers, only about 54 birds being present. Of
these, a number were in pre-nuptial or nuptial plumage and one nest
contained three young chicks in sparse down. There were also a number
of empty but completed nests, substantially built of seaweed and ice-
plant, Mesembryanthemum australe (now Disphyma australe). Some
birds were in non-breeding plumage, and were apparently either the
earliest adults in post-nuptial, or those not yet changed into pre-nuptial
plumage. Further, the colonies at this season also contain sub-adults,
still at the post-juvenal stage, which, as mentioned above, it 1s not yet
possible to distinguish from the adults. (Turbott.)

The colony was examined by Stein on 31st March, 1956, but only
one bird showed signs of pre-nuptial plumage: the throat was almost
black and dark feathers were present on the crown. There were also
incipient crests, still almost hidden.

From these observations, it is evident that the onset of breeding 1s
irregular in this colony, or has changed since the period of Falla’s
observations. Further, the period between the pre-nuptial moult and
egg-laying—a full three months at Bethells—inay prove to be com-
paratively short at the David Rocks, as the earliest hatching recorded
is in early May (laying approximately a month earlier), but the pre-
nuptial moult was evidently just beginning on visits on 3rd and 3lst
March. A connected series of several years’ observations are now needed
on the breeding cycle in this colony.

Evidence is also given by Cox (1946) of much later breeding at
this colony, as nests were being reconditioned and built on a visit on
31st July, 1946. One nest contained two eggs, apparently well incubated.
Fleming (1940) visited the colony on 6th August, 1938, and found
“nest building in progress, four well-grown young and three clutches
of fresh-looking eggs.” Fleming also reports that in October, 1939,
there were “all stages of young and eggs.”

The Spotted Shag. 359

Two specimens, included in the section above, may be mentioned,
as they give additional information in relation to the David Rocks colony.
In AV. 96.77, as already mentioned, it is likely that the post-nuptial
moult is almost complete, although alternatively the specimen may be
just entering into the pre-nuptial moult. In the former case, the date
of collecting (30th May) would correspond to the earliest hatching at
the colony. In AV. 96.80 pre-nuptial plumage is being assumed on 28th
March, and laying and hatching would probably in this case be consider-
ably later. Unfortunately, full data are not given with several earlier
specimens from the Hauraki Gulf.

On Otata Island, the largest island of the Noises group, the colony
may not be regularly occupied, as mentioned earlier. The few observa-
tions available on this colony are of considerable interest, as the com-
mencement of breeding apparently corresponds to the earliest dates for
the David Rocks colony. On 2nd April, 1950, Davenport (1951) found
that there were some birds on nests; and on this visit (pers. comm.) saw
two nests each containing a clean egg, which had apparently only recently
been laid. Sibson (1948) reported approximately 10 nests in this
colony, but only four birds were seen during his visit on Ist December,
1946. By this date, the breeding season was probably nearly over.

The colony on eastern Waiheke Island, near Anita Bay, has been
observed more systematically by Stein. Two groups included in this
colony are relatively inaccessible, as they are cave-dwelling, and Stein's
detailed observations on breeding have been made mainly on the cliff
nesting portion of the colony (the “Terraces”). In addition to regular
visits by Stein, the following are notes on the colony made by other
observers: (a) According to Oliver (1930a), eggs were found in Janu-
ary, and a photograph is shown (taken on Ist January, 1916, pers.
comm., W. R. B. Oliver). (b) Cox (1946) examined the colony which
was “nesting in a small cave, some ten or twelve nests being located on
ledges within two or three feet of the cave roof.” This visit was on 23rd
October, 1945, and the nests that could be examined contained eggs.
(c) On 29th September, 1946, according to Buddle, Sibson and Fleming
(1947), there were “about 40 birds and 17-20 nests; few adults still
crested; of 11 nests, three new and empty, one with three eggs, rest
with young up to nine inches high.” (d) McKenzie (1948) states that
the colony consisted of 55 occupied nests containing eggs or young, when
visited by T. M. Roberts on 28th December, 1946. About 55 young,
some newly-hatched, were counted on a later visit by McKenzie on 25th
January, 1947; and the nests were empty on 23rd February, 1947,
although there were still many young in the colony. Finally (e) photo-
graphs taken by G. A. Buddle in 1946 of the “Terraces” portion of the
colony are shown by Buddle (1951) and Oliver (1955).

A brief summary of Stein’s observations gives the following infor-
mation on the breeding cycle on the “Terraces.” A large proportion are
in nuptial plumage in August. In addition, beginning in 1949, it has
been found that a number assume nuptial plumage by the second and
third weeks of February. At three separate periods eggs are observed:
in late August, in December and (observed from 1950) in March. The
young chicks are present in early October, January and April. As
shown by Stein’s observations, these dates represent separate breeding

360 TURRBOTT.

peaks and were not due to disturbance tollowed by attempts at re-nest-
ing. The peaks have been observed during successive years. It would
be of special interest to follow up the breeding cycle in this colony by
colour ringing. Although the August and March laying periods evi-
dently correspond to the peaks of nuptial plumage in August and Feb-
ruary respectively, there were no observations of nuptial plumage pre-
ceding the egg-laying period in December. It will be noticed further
that the pre-nuptial stage is not included in the observations, but this
would occur, at least before the first peak of breeding, while no visits
were being made in mid-winter. Birds in pre-nuptial plumage were not
observed before the later breeding peaks.

A. few additional notes may be given on the Bethells colony, as these
suggest some modification of the annual cycle based on Petersen's
records. Fleming (1940) on 11th June, 1939, recorded “incubating,
nest material carried,” but his field notes (pers. comm.) show that the
use of the term “incubating’’ was not justified as no eggs were seen.
Birds sitting on well built or partly-finished nests would suggest that
breeding had started some weeks earlier than found by Petersen.
Macdonald (1951) notes that there were no young to be seen on 23rd
September, 1950, when the landward portion of the colony was exam-
ined: it would appear from Petersen’s record that young are generally
present by this date. Observations on visits on 24th November and 8th
December, 1951, conformed fairly closely to Petersen's: fledglings were
seen in the nests on 24th November and a number were in juvenal plum-
age on 8th December, although on this date the least advanced were still
downy chicks in the nest (Turbott)*.

On 31st March, 1956, 10 birds were observed fairly closely on the
landward portion of the colony, eight clearly showing newly sprouting
crests and greyish-black or black throats, as would be expected accord-
ing to the cycle observed by Petersen. There were several with an
almost pure white line on head and neck, but 1n others numerous brown
flecks still present on the line were visible. Two appeared to be in post-
nuptial plumage with no signs of moult. (Turbott.)

The only observations from other west coast colonies suggest that
the breeding cycle is approximately the same as at the Bethells col-
ony: well-grown chicks in late November at Oaia Island and eggs in
mid-October at Girdwood Point, However, it is significant that small
naked chicks were observed at Girdwood Point on 20th April, 1946, by
E. S. Richardson, and there may be a particularly extended season at
this colony, or there may be more than one breeding peak. In addition,
there is some doubt regarding the breeding season at Oaia Island, as
Fleming (1940) reports finding a “downy chick.” probably from this
island, washed up on Muriwai Beach on 17th June, 1939. This record
suggests that breeding may also be extended here, with the beginning of
egg-laying early in the year.

In Table 3 the information available on the David Rocks and eastern
Waiheke Island colonies is summarised and compared with the breeding
cycle at Bethells.

* Two specimens, a fledgling (AV. 96.101) and a juvenal (AV. 96.102) found on
the beach at Bethells on this date are included in the material described earlier.

The Spotted Shag. 361

Table 3. Breeding Cycle of the Spotted Shag at Bethells and in the Hauraki Gulf.

Roosting on Courtship, MTL MEGSGhtctcin. ta. | ee Sage Meee ei
site of nest-building, Eggs laid. hatch. teaetst yc
colony. (Pre-nuptial (Nuptial ( Post-nup- (Post-nuptial
( Post-nup- and nuptial plumage) tial plum- Bis
tial plumage) plumage) age) plumag
R begins begins begins
eh Bobet Ceri eer le May August obtaue td November
[Napili uptial plum- bite Ang. early Oct.
age Aug.
(pre-nuptial December January
Eastern and nuptial
Waiheke 1. suppressed?) no records
ec ag March April
age Feb.

(pre-nuptial
__ suppresse essed ?)

cat. July, Ras: Aug,,

Aug., Oct. October
pest Rate April — (scattered (scattered no records
. records records
ed) | ee ee a ee ee ee only)

A major factor, which may have affected breeding times and
resulted in the differences in breeding season between the Hauraki Gulf
and the west coast, was the early uncontrolled destruction of the colonies
in the Hauraki Gulf. However, as noted above, there is some eviderre
that Bethells differs from the other west coast colonies (Oaia Island and
Girdwood Point), and it is not known whether there was disturbance
of any of the west coast colonies. When fuller information has been
obtained, a comparison of the breeding cycle at the colonies in the Auck-
land area may suggest other factors resulting in differences in breeding
season from colony to colony.

SUMMARY

1. The past and present distribution of the spotted shag in the
Auckland area is outlined. In addition to a colony previously recorded,
another colony is known to have survived in the eastern Hauraki Gulf
during the period of greatest destruction, 1910-1931.

2. Terminology for stages of plumage as suggested by Murphy
for certain South American shags is adopted, as it conforms to data on
the colony at Bethells. Field notes on the breeding cycle and correspond-
ing plumage changes at Bethells are given, and material ; in the Auckland
Museum discussed with reference to the field observations.

3. It was not possible to find a satisfactory means of distinguishing
between the post-juvenal stage (i.e., the first plumage following the
juvenal) and the corresponding adult plumage (adult post-nuptial).
An investigation based on ringing would indicate whether this plumage
differs from the adult. The fitst pre-nuptial plumage apparently does
not differ from the adult pre-nuptial.

4. The observations on breeding cycle at Bethells are compared
with the records available from colonies on the David Rocks and eastern
Waiheke Island (Hauraki Gulf), and differences in breeding seasons
are briefly discussed.

362 TURBOTT.

ACKNOWLEDGMENTS

I am grateful to Mr. E. EK. Owen, who has kindly drawn the map
showing distribution, and to Dr. R. A, Falla and Dr. C. A. Fleming for
their comments on the manuscript.

I am also greatly indebted to the Marine Department and to Mr.
E. W. Gilliver, District Inspector of Fisheries, Auckland, for trips to
the Noises Islands and Waiheke Island in the patrol vessel M.V. “Ocean
Star.”

BIBLIOGRAPHY
BUDDLE, G. A., 1951. Bird Secrets. Wellington.

BUDDLE, G. A., SIBSON, R. B., and FLEMING, C. A., 1947. Summarised
Classified Notes, New Zealand Bird Notes, 2 (3), 40.

BULLER, W. L., 1888. A History of the Birds of New Zealand, 2nd Edition.
London.

BULLER, W. L., 1905. Supplement to the Birds of New Zealand. London.
COX, T. W., 1946. Spotted Shag Near Auckland, New Zealand Bird Notes,
Pate), ak |

‘

DAVENPORT, Jj. C., 1951. Summarised Classified Notes, Notornis, 4 (3), 41.

FALLA, R. A., 1932. New Zealand Cormorants in the collection of the Auckland
Museum, with notes on field observations, Rec. Ack. Inst. Mus.,
1 (3), 139-154.

FALLA, R. A., 1940. New Zealand Sea and Shore Birds. Wellington.

FLEMING, C. A., 1940. Summarised Reports, Ann. Rep. N.Z. Ornith. Soc.,
1939-40, 7. (Reprint, 1953, 9.)

FLEMING, C. A. (et al.), 1953. Checklist of New Zealand Birds, Ornitho-
logical Society of New Zealand, Checklist Committee. Wellington.

MACDONALD, N., 1951. Summarised Classified Notes, Notornis, 4 (3), 41.
MACDONALD, N., 1953. Summarised Classified Notes, Notornis, 5 (3), 88.

McKENZIE, H. R., 1948. Summarised Classified Notes, New Zealand Btrd
Notes, 2 (7), 158.

MURPHY, R. C., 1936. Oceanic Birds of South America. New York.

OGILVIE-GRANT, W. R., 1898. Catalogue of the Birds in the British Museum,
Vol. 26. London.

OLIVER, W. R. B., 1930a. New Zealand Birds, lst Edition. Wellington.

OLIVER, W. R. B., 1930b. The New Zealand Double-crested Shags; With
Description of a New Species, Trans. N.Z. Inst., 61, 138.

OLIVER, W. R. B., 1955. New Zealand Birds, 2nd Edition. Wellington.
POTTS, T. H., 1873. On the Birds of New Zealand, Trans. N.Z. Inst., 5, 171.
ROBERTS, T. M., 1943. Summarised Classified Notes, V.Z. Bird Notes, 1 (3), 19.
ROBERTS, T. M., 1944. Summarised Classified Notes, N.Z. Bird Notes, 1 (7), 68.

ROBERTS, T. M., 1946. Spotted Shag Near Auckland, New Zealand Bird
Notes, 2 (2), 31.

ROBERTS, T. M., and McKENZIE, H. R., 1942. Summarised Classified Reports,
Bull. Ornith. Soc. N.Z., 3, 11. (Reprint, 1953, 83.) |

ROBERTS, T. M., and McKENZIE, H. R., 1946. Summarised Classified Notes,
N.Z. Bird Notes, 1 (11), 124.

The Spotted Shag. 363

SIBSON, R. B., 1946. Spotted Shag Near Auckland, New Zealand Bird N otes,
rte 31.

SIBSON, R. B., 1948. Summarised Classified Notes, New Zealand Bird Notes,
2 (7); 158.

SIBSON, R. B., 1950. Summarised Classified Notes, New Zealand Bird Notes,
3 (8), 204.

SIBSON, R. B., 1952. A North Island Colony of Spotted Shags, Notornis,
4 (8), 214.

STEAD, E. F., 1932. The Life Histories of New Zealand Birds. London.
TURBOTT, E. G., 1946. Spotted Shag Near Auckland, New Zealand Bird N otes,
PCY? Sly

TURBOTT, E. G., 1947. Summarised Classified Notes, New Zealand Bird N otes,
2 (3), 40.

WIGHTMAN, G., 1953. Summarised Classified Notes, Notornis, 5 (3), 88.
WIGHTMAN, A. T., 1956. In press, Nofornis, 7 (2).

PLATE 51.

tig. 1. Spotted shag ‘n pre-nuptial plumage showing the characteristic decorative

plumes on the nape, back and rump. The transition to nuptial plumage
is marked by the loss of the decorative plumes and reduction of the crests
by wear and moult. Bethells colony.

Photo: QO. Petersen.

—
aon

—_—t «

~
“

PLATE 52.

Nuptial plumage, characterised by the black throat, white lateral line on
head and neck, and distinct but diminishing crests. Adult close to nest,
st B, early August, 1953. Bethells. This bird is shown 1 t- ial
nest Bb, early August, lyoo. Dbethells. us bird 1s shown 1n post-nuptiat
plumage in Fig. 6.
Nuptial plumage: mate of the above nest-building, nest B, early August,
1953: shown in post-nuptial plumage in Fig. 7. In this and the bird above
a few decorative plumes have not yet been shed.

Photos: O. Petersen.

BLA EE So.

Fig. 4. Nuptial plumage, showing transition to post-nuptial during incubation.
| The crests have been lost and brown flecks are appearing in the white
line on head and neck. Nest C, September, 1954, Bethells.

Fig. 5. Newly hatched chick, nest A, October, 1952, Bethells.

Photos: O. Petersen.

PLATE 54.

Post-nuptial plumage: both adults at nest B, October, 1953, Bethells.
The bird feeding a chick shows the more pronounced type of head pattern,
and there are still a number of black feathers amongst the new grey and
white on the throat (shown in nuptial plumage in Fig. fig Ny

Post-nuptial plumage: “greyer” type of adult with chicks approximately
four weeks old, nest B, October, 1953. (Shown in nuptial plumage in
Fig, 3.)

Photos: O. Petersen.

PAA TESS.

Fig. 8. Post-nuptial plumage: pair at nest A, November, 1952, Bethells. The
single chick is partly hidden by the upper bird.

tig. 9. Post-nuptial plumage: the lower bird in Fig. 8 (see above), showing the
“oreyer” type of plumage. The head pattern is indistinct and a few black
feathers remain on the throat. Chick three to four weeks old showing
first down partly replaced by the second. November, 1952.

Photos: O. Petersen.

tig. 10.

Fig. 11.

Peele se ita,

Pair in post-nuptial plumage (left and right) and three fledgl1
(centre), still showing the second down on head and neck. The fledglit
are between five and six weeks old. Nest C, November, 1954, Bethe

ls.

ae ee a)
JQ TS

Juvenal plumage: (right) two young from nest C approximately seven
weeks old on ledge near the nest, and (left) two juvenals just over nine
weeks old. The young fly at nine weeks. December, 1954, Bethells

Photos: O. Petersen.

re -
i a
See
aa
ayer wrx
ca

365

Tauihu: The Maori Canoe Prow
By GILBERT ARCHEY,

The purpose of this paper is two-fold. In the first place it records
the types of canoe-prow made by the Maori in New Zealand and illus-
trates the various decorative designs that enhance their structure and
form, Secondly it provides instances and details to amplify the observa-
tion that these several types, at first sight so different, possess important
features in common, and that the differences themselves are no more
than modes in which these common characteristics are presented or
developed.

The photographs and drawings which follow will also reveal what-
ever aesthetic quality tawihu possess ; we hope they will be found worthy
of attention for this as well.

River-Canoe Prow

The plainest type of prow belonged to the fishing canoe, the small,
broad dug-out with wash-strakes, used for everyday coastal work. It
will be seen that this simple prow (Fig. 1), known as tete, is essentially
a bow-cover with a transverse wash-board behind and a carved head in
front. Its almost purely functional structure and its very general use
in New Zealand fairly mark it as the prototype.

The bow-cover portion is fitted and lashed to the dug-out below
and to the wash-strakes behind; the head is borne at the end of a neck
of varying length. Apart from the typical mask details of the face the
river-canoe prow was undecorated.

The stern-post of this work-a-day canoe was likewise unorna-
mented; it was little more than the necessary rest or fulcrum for the
large steering paddle to bear against. Nevertheless, it swept upward
gracefully as a continuation of the curve, or sheer, of the after end ot
the vessel, as, at the other end, the neck of the prow carried the corre-
sponding curve upward and forward to the figure-head.

Our next example is a prow from Doubtless Bay illustrated in text
figure 3 and Plate 57, fig. 2. Although unfamiliar in general appearance,
it maintains the functional structure of a bow-cover typical of the river-
canoe type. The head, with its small attendant creature behind, is
unusual in appearance, and both it and the vertical neck-pillar are
studded with thorn-like projections. Similar spurs or spikes project
from human figures and heads carved on a slab recovered from the
Awanui swamp only twelve miles distant (Archey, 1933, p. 209).

The long, projecting mouth of this figure-head would suggest a
bird motive, were it not for the large conical teeth (matched in human
head carvings from this district) and the essentially similar though not
so extreme projection of the mouth in other river-canoe prows (Text
fig. 2). Indeed, the three prows here illustrated (text figures 1-3) pro-
vide a typical example of extension or decorative elaboration of an

Rec. Auck. Inst. Mus. Vol. 4, No. 6, pp. 365-379, 25th October, 1956

366 ARCHEY.

anatomical feature, in this case the mouth, that is so common a feature
of Maori wood-carving.

Another unusual prow is that outlined in Figure 4 and Plate 5/7,
fig. 1. More so than any other it is a practical bow-cover. We do not
know what the canoe it belonged to looked like, but in our sketch we
suggest something long, narrow and shallow, feeling that the gentle
upward sweep of the prow would have been an expression or an exten-
sion of similar lines of the bow of the canoe, The prow itself is clearly
another variant of the tete. In a way its upward and forward sweep
foreshadows the outline of the leading edge of the highly decorated
prow of the large war-canoe, waka taua, to which, as the main subject
of this paper, we now turn,

The War Canoe Prow

Structurally the war-canoe prow, tauihu (Fig. 5), is but an elabora-
tion of the prow of the river-canoe. It comprises the same bow-cover
or lid with a transverse wash-board at the after end; but the simple
projecting neck and head of the tete have now become a full human
figure vigorously postured. The upper level of the bow-cover, instead
of curving downward as a neck, continues horizontally forward beyond
the transverse wash-board to reach and merge with the curved body ot
the leading figure. A final modification is that instead of the whole of
the wood between the wash-board and the head having been cut away,
there has been left a mid-line vertical panel connecting them, a panel
that vies with the leading figure itself for our interest and attention.

In the first place, this panel has an obvious structural or strengthen-
ing function; it is also an escutcheon for a striking decorative design.
Although the general composition of its decoration is the same for all
tauthu in its group, it is saved from being stereotyped by an intriguing
variety in its details and in the proportion of its parts. None the less, it
is standardized in another way, for although it is unmistakably the
pattern peculiar to ftauihu, it also comes within an even wider convention
characteristic of the greater part of Maori wood carving.

This convention I have described elsewhere (Archey, 1955, p. 12)
as an alternation of tiki (human figures standing fullface) and manaia
(human figures in lively attitude and with profile face), a theme that
has become further developed into an alternation of figures (tiki or

manaia) and double or interlocking spirals.

If the reader will turn to Plate 58 he will see, in an exceptionally
fine tauihu from the Bay of Plenty, a clear presentation of this alterna-
tion. The elements comprise in succession: (a) the leading figure;
(b) a part manaia facing aft; (c) a large double-spiral (pitaw) ; (d) a
stylized full-face figure in openwork or pierced carving; (e) another
large pitau; (f£) a forward-looking manaia elongated and somewhat
cramped to fit the available space; and (g) a figure with its back to the
wash-strake looking aft into the canoe.

The theme of alternate figures and spirals appears regularly in
door-lintels and in many other carvings. The tawihw panel version
acquires its special characteristics from the carver having taken advan-
tage of the proportions of the panel to emphasize and expand the double-
spirals so that they become the dominant element in the design. The

Maori Canoe Prow. 367

@ r=)

Fig. 1. River-canoe prow, tete. Coromandel. Auckland Museum; presented
Miss Lucy Cranwell.

Co
aH © ee cas

Fig. 4. Prow in Taranaki Museum.

368 ARCHEY.

Maori name for these elements stresses this special interest: pitau
denotes the double spiral pattern itself: it also stands for this type of
prow and as well for a canoe that bears it.

The base of the prow also has a standard composition or content
in its decoration but with rather more variation. Typically the hori-
zontal upper surface of the bow-cover hears a4 full-face human figure
looking upward (Plate 59). It is, of course, divided into two halves
by the vertical panel.

On the vertical sides of the base (Fig. 5) we again find a succession
of human forms. At the after end is a human figure, full-face though

Fig. 5. Tauihw: carved prow of war-canoe.

in sideways stance; in front of this is a large forward-looking profile
face with upper lip only, from under which projects a large curved
tongue. This enloops anteriorly with another element, apparently a
tongue, or a lip maybe, or even a body grasped around, as it often is,
by a hand,

The prow itself is wide behind where, therefore, the two sides of
the base stand separate, each abutting against the canoe wash-strake of
its side. Anteriorly the base narrows, whereby the anterior portions

Fig. 6. Design on the vertical sides of base of fanithu; shown as if the sides had
been splayed out horizontally,

Maort Canoe Prow. 369

of the lips of the large profile face meet medianly and, with a now single
tongue and anterior loop, form a median basal support for the narrower
forward portion of the prow. If one were able to slice off the horizontal
flat bow-cover, and splay outwardly the vertical walls, the design on the

base would appear something like the pattern outlined in figure 6.

Figs. 7, Bay of Plenty, cf. Pl. 58: 8, Wanganui Museum, cf. Pl. 61, 1.

The different lengths of the base among tauihu offered opportunities
lor varying the design of the sides, either by including an extra figure
or telescoping the elements together. The prow of Te Toki a Tapiri,
the 82-ft. waka tava in this museum, is exceptionally long ; the composi-
tion of its base pattern (Plate 62, hg. 2) includes no less than three
figures, 1.e., a full-face figure aft, a manaia looking forward, and next a
human body with its neck joining the top and back of the customary
large head profile. The interloop motive by which the design terminates
is composed of limbs or bodies. In a much shortened model-canoe prow
(Plate 65, fig. 3) it is the upper lip itself of the profile face that pro-
vides the first part of the interloop. Extra room for this relatively
large interloop was obtained by omitting the forward horizontal part of
the bow-cover: after all, it was hardly necessary in a model. Neverthe-
less, the same omission of bow-cover to allow for a more ample inter-
loop, or incipient double-spiral, is a feature of a fine tauihu from the
Wanganui district in the Dominion Museum (Plate 65).

Coming now to the design which separates the large double spirals
of the central panel, we find one of the most interesting of Maori carv-
ing patterns. Its theme is simple enough, a standing figure, usually full
face; but it is handled in all degrees of intricacy of open-work or tracery.
Two examples, from the Royal Scottish (fig. 11) and from the Auck-
land Museum (fig. 7) show it in fairly simple outline; some of the
ensuing elaboration is illustrated in the accompanying text figures (8 to
13) and others can be followed in the photographic illustrations. Two
faces, one upside down and each with fingers in the mouth, comprise the
pattern in the Ngatiawa prow of Plate 63, fig. 1; and even more intri-
cate details of face profiles appear in Plate 63, fig. 3, where the lower
portion of the pattern is a medley of face and figure profiles,

370 ARCHEY.

The rear-facing figure with its back to the wash-board is usually
naturalistic, but even this may be patterned. The most elaborate
included here is in figure 3 of plate 64; could it be that the complexity
in this case ensued by way of compensation for the carver, who had
somehow rendered the central panel figure more than usually naturally ?

A final detail remaining for mention is the keel or band borne by
the leading figure; pitaw and manaia are the usual elements but often
reduced or cramped together.

The foregoing description may have been somewhat tedious in its
detail, though it will have served to reveal the ingenious complexity the
Maori carver often indulged in. Greater interest, however, attaches to
the strong sense of design appearing in parts of tauihu composition and
to the presentation in one carved object of three or more stages of the
handling of subject matter in decorative art.

Thus, referring to Plate 63, fig. 1, the vigorous leading figure and
the small figure aft are hardly so far modified i in the direction of applied
sculpture as to remove them from the representative or realistic, and the
same can be said of the relief figure that gazes steadily upward from
the broad basal portion or bow-cover of the prow. Stylization appears
in the large profile face of the base, and is well advanced in the intricate
pierced figures between the spirals. The spirals are, of course, abstract
forms of high quality, and the piftau, as they are named, have become
an accepted form frequently used in composition, though still freely
employed in all stages of stylized interlocking lips.

It is, however, in the openwork central figure of the panel that we
find not merely versatility within a convention but also an originality that
can fairly be rated as creative design. In figures 8 and 10, for example,
we see how the features above the stylized mouth surrender their
natural form to become abstract decorative detail. The limbs are
handled to the same purpose even more successfully; obviously their
shape as limbs was of little concern to the craftsman intent on winning
a design from them.

Appreciation of the Maori carver’s possession of this conscious
sense both of design, and of abstraction as a means of achieving it,
is of prime significance for our understanding of Maori art either
aesthetically or historically. It enables us to see the carver as someone
positively aware of the design possibilities of the natural forms he is
using, and capable of taking hold of them and bending them to his
purpose. This interest in pure pattern can hardly be seen better than in
figure 14, where two bodies are first drawn out into curved parallel
bands aligned with the sweep of the double spirals between which they
stand and then recurved as scroll-rendered manaia faces to fill the upper
and lower areas

An abstract design so neatly achieved is not only aesthetically
acceptable; it speaks of creative art and of individual purpose as well
as feeling as its source. And reverting to the natural forms that inspired
it, it is not without interest to compare these two elongated abstractions
with the slender undulating manaia that form the primary motive in the.
next form of tauihu we introduce—the trapezoid prow.

Muort Canoe Prow. 37 |

Figs. 9, Auckland Museum, cf. Pl. 64, 3; 10, Waitara, Bishop Museum, 1424;
11, Royal Scottish Museum, cf. Pl. 65, 1; 12, East Coast, cf. Pl. 63, 3;
13, Okehu, Wanganui, cf. Pl. 65, 2; 14, Hamilton, Maori Art, p. 46, Pl. 11.

372 ARCHEY.

The Trapezoid Prow

We turn then to the form of prow illustrated in figure 15. A name
applied to it was tolere. At first sight it seems to stand completely apart
from the fauihu we have been describing. It is undoubtedly different,
but not entirely so, either structurally or in its decorative design.

Considering it first structurally, we observe that a panel (toverc
stands vertically above a flat bow-cover (tauimatua, ie., support) and
backs against a transverse wash-board (paretai), The toiere thereby
occupies the same position as the mid-line panel of the pitau-decorated
war canoe prow; it is its homologue.

Observing it next as decoration, we soon recognize the unusual
elements comprising it as no more than forms with which we are familiar
handled in a different manner.

Fig. 15. The trapezoid prow; British Museum.

Fundamentally, the composition or content of the panel decoration
is the same in both types of prow, 1.e., an alternation of human figures
or manaia with interloop (or double-spiral) tracery. It is only in the
relative size of the spirals and in the treatment of the human figure that
they differ. In the one we see openwork full-face figures as already
described ; in the other the figures are extremely elongated manaia of the
type found in other carvings from the Northland area. What we are
looking at 1s an art preference, wherein the fine spiral rhythm of one
school stands in contrast to the rhythm of undulating figures of the
other. And in the latter case the whole of the human figures, not only
the enlooped mouths, have become stylized to produce the desired pat-
tern. They are still recognizable, however, as figures in profile, not
having been carried forward beyond stylization to the degree of abstract
design of figure 14 discussed above.

Maort Canoe Prow. Bfa

The basal portion and the transverse wash-board were also orna-
mented, No satisfactory illustration is available for inclusion here, but
Plate X of Hamilton’s Maori Art shows naturalistic hgures on the wash-
board and a pattern of stylized figures on the bow-cover, An additional!
feature was a carved head with tattooed face (Pl. 67) carried right
forward on the hull itself.

Trapezoid prows have, from time to time, been referred to as
“northern”; but this is by no means a reliable allocation. One such
prow is from the Waikato River ; the two illustrated in Maori Art, p53,
Plate V, are localized “Auckland,” but if the city is intended they may
have reached it from almost anywhere. The finest of this type, in the
British Museum (Plate 66, fig. 1), is unlocalized. While, therefore, the
attribution of these to “Northland” may be tentatively made on the basis
of the carving style, it should be remembered that this is only con-
jectural.

A Connecting Link

The last prow to be mentioned is particularly interesting, not merely
because it is old stone-tool work, but also for its clearly intermediate
position between the two types of prow we have been considering. Like
each of them, it comprises (Fig. 16) a bow-cover base and a transverse

2#o>see
ay... am

Fie, 16. North Taranaki prow.

wash-strake, though a very low one, an a median panel. The latter,
though somewhat crudely carved, bears the simplest possible rendering
of the alternate human figure and double spiral (or loops in this case)
characteristic of the median panel in both of the others. The panel
itself, moreover, is of the same form and proportion as the standard
type, and its upward sweeping leading edge would require only the
slightest treatment to make a man of it. Instead, the panel bears a
manaia, niuch reduced, at the extremity, a figure that would only have
to be lengthened to make the long manata of, say, the superb British
Museum prow.

The three prows comprising Plate 66 have but to be compared to
enable us to realize that all three are related in functional structure, in
basic form, and in the content of their decorative design. Tauihu thus

Maori Canoe Prow. 375

Fig. 18 a and b. Tahiti: Hornell.
Fig. 19. Nukutavake: Hornell.

Fig. 20. Rarotonga: Auckland Museum.

Cate

—— >
AA Wi WN cS ay
Zi

18)

Fig. 21. Atiu: Auckland Museum.

In Rarotonga the bow-cover (Fig. 20) was perfectly plain and
projected only slightly beyond the hull; but in an Atiu canoe in this
museum (Fig. 21) there is a long flat fore-deck covering the anterior
one-fifth of the hull, next in front is a short bow-cover and terminally a
small upward projection fitted between bow-cover and hull.

374 ARCHEY.

stand, with pare or door lintels, as examples of the manner in which the
Maori carver used his tiki and manaia in repetition and alternation with
spirals to produce patterns basically the same but diverging in method
of treatment. These styles comprise what might be called the schools
of Maori art, but not of schools precisely defined either geographically or
stylistically, because we already see from the relatively few examples
available how varied in manner they are and how frequently and strongly
the ideas and feelings of individual artists find expression in them.

Canoe Prows in Polynesia

On comparing the Maori canoe prow with those of Polynesia we
again quickly realize how similar they all are, at least in basic form.
Structurally, or practically, each is a bow-cover which extends the sheer
of the hull upward and forward; symbolically or commemoratively each
nearly always carries in front a head or a human figure. The prows in
the islands exhibit this structural arrangement in varying manner, but
simply and without elaboration except in the Marquesas, where addi-
tional human figures, incipiently stylized, appear. The accompanying
sketches, for the most part copied from Hornell (1936), show the styles
characteristic in each group.

Fig. 17aandb. Marquesas: Hornell.

The basic relationship between hull, wash-strakes and bow-cover 1s
seen in the Marquesan prow illustrated in text-figure 17a. The slender
curved forward reach, and its termination, look like a bird neck and head
in profile view, but the upward-gazing face is distinctly human. The
close similarity of this prow to the Taranaki Museum example outlined
in figure 4 (p. 367) is readily apparent, as is its general resemblance to
the standard Maori river canoe prow of figure 1, except in the style of
the face or head.

In Tahiti one type of prow (Fig. 18a) is a plain plank-like projec-
tion narrowing slightly forward; another is an upwardly curved exten-
sion of the bow with a small human figure looking forward (Fig 180).
A second figure on this canoe looks behind from the stern, which differs
from the prow only in being higher, Hornell (p. 124) thought, how-
ever, that this canoe might be Tuamotuan, or from Rurutu. A definitely
localized Tuamotuan canoe described by Hornell is from Nukutavake ;
its prow (Fig. 19) is a solid “long and gracefully tapered blunt-ended
projection.”

376 ARCHEY.

In Tikopia (Mig. 23) the prow is carved from the dug-out hull
itself, as it is in Samoa (Fig. 22) ; the dental decoration is also similar,
though more extensive in Tikopia. In the latter a separately fitted bow-
cover is lashed on above the prow.

All these prows are very simple; the one attempt at elaboration,
from the Marquesas (Fig. 17D), has stylized human figures in mid-line
between the transverse wash-board and the terminal carved face. Simple
though it be, its basic form invites comparison with that of the Maori
war canoe, each comprising a transverse wash-board, a horizontal base
plate, a terminal face and a vertical mid-panel, IXxcept, however. for
this very tentative approach, the island canoes have no part in the exten-
sion and elaboration of structure and ornament that so strikingly char-
acterises the tauihu of Aotearoa.

, AAD
LDN AAN ANEW AS ALN

Fig. 22, Samoa: Auckland Museum.

Fig. 23. Tikopia: Auckland Museum.

a) lp

TTA TANG
24

Fig. 24. Hawaii: Hornell.

Maori Canoe Prow. 377

Nevertheless, the fundamental structural design appears in all of
them: 1.e., in the bow-cover and terminal head or figure, Here, then,
is the essential relationship among them. From these basic elements
the Maori, and only the Maori, has developed further structural elements
and decorative complexity. Not only tauthu, but taurapa also (Archey,
1938), exhibits the development or evolution from a plain practical form
towards extension and elaboration, and in practically every phase of
Maori wood-carving we find told a similar story of local development
not only of structure but also of decorative design.

Indeed, throughout Polynesia the basic content of wood-carving 1s
the same; the development is different in each area, though with rela-
tionships between the island arts of the Oceanic region where a rec-
tilinear fashion prevailed, Even the simple spirals of the Marquesas,
developed from insect legs and antennae, have experienced the rectilinear
restriction, the outcome, I suggest, of the difficulty of carving in really
hard wood.

Only the Maori, favoured with the soft wood of the totara and
with sharp greenstone chisels, had launched into the complexity of free
flowing patterns, with what success the tawihu patterns are by no means
the only examples.

When, therefore, we see, in the Polynesian region from which the
Maori traditionally came and to which he is culturally related, the basic
structural elements alone of the tauwihu without any decorative elabora-
tion, and when we see in the remote and isolated colony of Aotearoa
every degree of departure from them and every stage in evolution of
structural and ornamental complexity, the history of the art of the Maori
and of its design elements ought to be readily apparent.

The view that art motives in Maori carving have a local origin 1s,
I submit, abundantly attested in the body of the art itself ; their develop-
ment accords with the principle stated by Duff (1950, p. 2), that “per-
sistent and continuous change . . . is self-motivated or spontaneous”’
and “independent of . . . influx of foreign populations,’ though
hardly, I would say, free from the effect of environment, an environ-
ment that in this case not only provided occasion in the needs of house
building and transportation, and supplied means in suitable wood and
effective tools, but also furnished inspiration in a stimulating climate and
noble natural surroundings.

More immediately and technically, or psychologically if you prefer
it, the inspiration that has developed Maori carving patterns has been
the individual craftsman’s direct and positive interest in form as such,
and his awareness of the possibilities inherent in these forms for the
creation of harmonious and well-balanced design.

Discussion: A Native Art?

Two aspects of enquiry have been appearing alternately in the
foregoing: the active practice of Maori art and its manner or place of
origin. They are, in my opinion, inter-related, for there is, in the range
of expression of an art and the versatility and competence of the hand-
ling of all its aspects, a significance for its origin equal to that which
might be sought in apparent similarities in the forms appearing in differ-
ent places.

ARCHEY.

w
NI
oO

To return to the primary subject of this paper, the canoe-prow:
at the time of Cook’s discovery, tawihu were being carved in every stage
of structural and decorative extension or elaboration. So were taurapa;
so were house carvings. Carvers were producing simple, practical
articles and plain naturalistic sculpture; either or both of these
might be stylized or elaborated or wrought into patterns. The patterns
themselves were won from whole figures or from any part—face, body
or limbs; the patterns would trend in the one direction of involved
curves or spirals or in the other direction towards simple, restrained
abstraction. Moreover, all these details and the trends appeared even
in one small composition, the central panel device we have been dis-
cussing. To repeat: the head appears full-face or profile with the
fingers varyingly introduced to enhance the complexity ; in most of them
the limbs are stylized and set at studied angles almost in the “contem-
porary” manner, while in one illustrated (Fig. 14) the natural form is
drawn out into a graceful, evenly flowing design that stands in accord
with the spiralling of the bordering pitau.

We have seen elsewhere (Archey, 1933; 1955) how the pitau
itself is almost invariably a pattern of interlocking lips in varying
degrees of extension or expansion; occasionally by way of further
versatility, or creative enjoyment, whole bodies or limbs are so en-
whorled. In another school, Taranaki, an entwining of undulating
bodies forms the pattern, while in the Kaitaia carving a simplification of
limbs produces an abstraction of strangely moving power. Few arts
can present so many styles. These parallels of pattern evolution are
themselves evidence of local development, unless, of course, there should
be, as there are not, art motives elsewhere of these several kinds from
which we could fairly derive them.

Coming then to the question of origins, I am constrained to add a
comment on the supposed introduction or borrowing of the New Zealand
double spiral from an outside art. Does it not, in the face of such clear
design competence as Maori art displays, appear altogether unnecessary,
or even trivial, to introduce one such borrowed element when, within
the art, not only this one but so many others are so freely created and
used ?

If there were real evidence of spiral forms in Central Polynesia we
should, of course, have to accept the possibility or even the probability
of their having been brought here; but where are they? It is precisely
here that the theory of introduced spirals encounters its main ethnologt-
cal difficulty, 1.e.: in the need to find a satisfactory or convincing place
of origin and route to New Zealand. Skinner (1924, 233) recognized
this need and postulated a curvilinear art style formerly dominating
Polynesia and later lost in the centre through a “strong new rectilinear
fashion” from which Maori art and to a less extent Marquesan were
“preserved by isolation.”

Barrow has recently (1955, 17) dismissed this argument as “‘lacking
evidence,’ and in even briefer terms; I myself have never found it
acceptable, nor indeed more than an unsupported supposition. It 1s
quoted approvingly, however, by Duff (1950, p. 5), who sees in it sup-
-port for his own theory that marginal distribution of an item of culture
-is evidence of its former existence at the centre of the area. It should,
however, be commented here that the evidence for Duff’s theory is the

Maori Canoe Prow. 379

existence of identical adzes at no less than ten marginal localities,
whereas there are only two by no means similar arts for consideration,
Marquesan and Maori, in which, moreover, the single pair of supposedly

related spiral elements are obviously different both in their origins and
their form.

In each of these two arts the spiral is an end product derived from
a natural form, but a different form in each case. Marquesan art, like
Maori, also stylizes face masks, but in a manner as near to Haida Indian
as to Maori. Interestingly, one Marquesan prow (Fig, 17&) has®
features in common with tauihu, but the relationship is in basic essen-
tials and not in the elaborations that comprise the full decorative vigour
of the Maori achievement.

Not only in its unmatched variety and creative vitality, but also by
the continuing existence within it of all phases of its development, Maori
art 1s marked as a local achievement. On the other hand, the absence
from areas in which one would expect to find it of evidence of outside
relationship, except in simple basic components, indicates its derivation
from a central art in which those as yet undifferentiated elements, mostly
naturalistic human forms, were common to all.

The closest parallel to Maori carving, in both its component ele-
ments and its art form, is in the moderately stylized human figures set
alternately full face and sideways in the staff gods of Rarotonga. The
Cook Islands, moreover, are quite a likely area in which to find a parallel
to the basic patterning of Maori art.

All this has, however, taken us away from fauihu, to which we
return only to recall that it is in the basic structural features that it and
the canoe-prows of Polynesia closely and clearly resemble one another.
Except in Aotearoa the Polynesian canoe prow has remained in the
unspecialized form; only the Maori has developed it. He has done so
structurally, in the composition of its decorative theme, in the richness
and diversity of its patterning, and most notably in his conception and
achievement of design.

REFERENCES,
ARCHEY, G., 1933. Wood carving in the North Auckland area. Rec. Auck. Inst.
and Mus. Vol. 1, No. 4, pp. 209-218.

ARCHEY, G., 1938. Tau rapa: the Maori canoe stern-post. Rec. Auck. Inst. and
Mus. Vol. 2, No. 3, pp. 171-175.

ARCHEY, G., 1955. Sculpture and Design: an outline of Maori Art. Handbook
of the Auckland War Memorial Museum.

BARROW, T. T., 1955. An introductory essay on Maori Art. Part 2 Te Ika a
Maui, by Padovan and T. T. Barrow. Wellington, October, 1955.

DUFF, ROGER, 1950. The Moa-hunter Period of Maori Culture. Department
of Internal Affairs.

HAMILTON, A., 1896. The Art Workmanship of the Maori Race of New Zea-
land. New Zealand Institute. Commonly referred to as Hamilton’s
Maori Art.

HORNELL, JAMES, 1936. Canoes of Oceania, Vol. 1. Bernice P. Bishop
Museum Special Publication 27.

SKINNER, H. D., 1924. The Place and Relationships of Maori Material Culture
and Decorative Art. Journal of the Polynesian Society, Vol. 33,
pp. 229-243.

7a,

gst,
ed

eae

PLATE. A

For cultural reasons, these images have been removed.
Please contact Auckland Museum for more information.

Fig. 1. Taranaki Museum.

a
Fig. 2. Doubtless Bay. Auckland Museum, 3654.

ate

es te

ee hee ae

pen = ¥

ws A

Tae oa :
xs)

Ta "iy

Pp es. es

For cultural reasons, this image has been removed.
Please contact Auckland Museum for more information.

War-canoe prow, tauihu; Bay of Plenty. Auckland Museum, 171. (a) Leading
figure; (b) part-manaia facing aft: (c) double-spiral, pitaw; (d) stylized
human figure; (e) pitau; (f) elongated manaia looking forward; (¢)
human figure looking aft.

| ai : vf
a ae
= al

:
an

;

a
rs

-
=
a
-
=

wy as
od te ee
a .

a v7 rs a el abe?

tag ea Wel clam 3°

For cultural reasons, this image has been removed.
malstz lois ere) alt= lem ANC lol d=] arom iV iUrcxo10 laa ce)manleacmialielaaarciiielar

Basal portion (bow-cover) of tauihu, seen from above. Auckland Museum, 29722.

ae.

7 aes

PLATE 60.

For cultural reasons, these images have been removed.
malstsloiomere) alt= lem ANU (el .¢t-] arom iV iercrol0 lam ie)mmanleacmialielaaarciiielar

Fig. 1. University of Pennsylvania Museum.

Figs. 2 and 3. Loc. Kapiti. Canterbury Museum, E. 141.787. (Wash-boards
renewed. )

= .
yh Ley

S luqy Es.
tae is ‘

i
abe ak

Le
a
-

PLATE 61.

For cultural reasons, these images have been removed.
Please contact Auckland Museum for more information.

Locality and place of deposition unknown. Photo. Dominion Museum.

Captured by Negaitai of Whakatane from raiding Ngapuhi. Locality
probably Bay of Islands. Auckland Museum, 197.

Bay of Plenty. Auckland Museum, 171. Photo. H. Powell.

gh a sa 6

For cultural reasons, these images have been removed.
ma (ate lsiom exe) al F=lerm ANU (er .dt- [arom lV Ukcxoel anim e)mancelacmiale)aaat-lilelap

Fies. 1 and 3. Waikanae; but “probably carved by east coast natives” (Hamilton,
Maori Art, p. 46). Dominion Museum. Photo. Charles Hale.

Fig. 2. Prow of Te Toki a Tapiri, built about 1836 on East Coast (Ngati Kahu-

—_-e

neuneu tribe). Auckland Museum, 150. Photo. H. Powell.

Beir gis
7 cas sae

he
PRs
Ps

For cultural reasons, these images have been removed.
Please contact Auckland Museum for more information.

Carved by the Ngati-awa chief Wiremu Kingi. Auckland Museum. 7375.

Ngatiawa: a relic of Te Rauparaha’s raid to Queen Charlotte Sound.

University of Pennsylvania Museum.

East Coast of North Island” (Hamilton, Maori Art, p. 44). Present
location unknown. Photo. Dominion Museum.

5
as

a oat a
Fe = we oa 7
“eee
5, -

Pa be 64

For cultural reasons, these images have been removed.
masts ojo exe) al F=(OLm ANU (ot df= aomm \V[Ukcxo1el anim ie)mancelasmiale)aaatslilelap

Peabody Museum of Salem.

Wanganui Museum.

Purchased in England by the donor, Mr. T. H. Hopkins. Auckland
Museum, 29722.

Pigkt E65,

For cultural reasons, these images have been removed.
Please contact Auckland Museum for more information

Royal Scottish Museum.
Okehu, Wanganui. Dominion Museum.
On model canoe purchased in England.

. “a

ime bp
os
3

No record. Auckland Museum.
Photo. H. Powell.

Rs,
«s

7 La

PA faa
~ ae nal 1
‘le ees) ifs

ree

ae ype ice ae
Vote = aoe

SR wl eS ee
Silene eS 2H. |
‘ Pe ay i *
a bl Ce te bi ihe -

”

=
Se ie |
va =
at

a
= =
ik

Ss es. mH, od

Ree ures oe
coe oP ey he 6 me ra -
eee ned 4 bal Ach

F wa
it |

PLATE. 66,

For cultural reasons, these images have been removed.
ma Cate lois exe) al F=leim@ ANG (el dt= [arom lV Ukcxol0 lane) mancelacmliale)aaatslilelap

s. 1. British Museum. Locality unknown.
ie. Z. Mokau. Auckland Museum, 5676.
iv. 3. Locality uncertain. Canterbury Museum, 141./788*

Qo fee a a ee ee eee eee ee eee
*“Tocality . . . probably a little to the north of East Cape’: Maori Art, p. 44.
Dr. Duff comments: “I think Hamilton’s reference to this as from Fast
Cape was a guess based on style. On grounds of style and the likelihood
of provenance which our records can establish, | would assign this to Cook
Strait.” (North Taranaki influence. )

For cultural reasons, these images have been removed.
malstsoloexe) a] t= (el AVC(e1 df=] alomiViersrol0 [aan ce)mmaglelasmlayie)aaarciiiela

Fig. 1. Auckland Museum, 2711. Locality unknown.

Fie. 2, Canoe figure-head. Thames district. Auckland Museum, 5998.

381

The Three Kings Cabbage Tree
By W. R. B. OLIVER,

The species of Cordyline found in the Three Kings group has up to
the present been assigned to C. australis. Recently I had the opportunity
of examining a living plant in the grounds of the Plant Diseases Division
of the Department of Scientific and Industrial Research, Auckland, and
of specimens taken from this tree, as well as specimens gathered from
another living tree growing in the garden of Mr. A. W. B. Powell,
Auckland. Together with these I have examined specimens collected
on the Three Kings islands and now in the herbaria of the Auckland
Museum, the Dominion Museum and the Botany Division of the Depart-
ment of Scientific and Industrial Research, Wellington. Comparing
these specimens with those of C. australis, important differences are
evident warranting the naming of the Three Kings form as a distinct
species. I am indebted to Miss Joan Dingley for specimens from the
tree in the grounds of the Plant Diseases Division, and to Mr, A. W. B.
Powell for specimens from his living tree.

The Three Kings cabbage tree was first recorded by Cheeseman
(1888) from Great Island. Three years late (1891) he recorded it
from Southwest Island and from Northeast Island. The trees on
Southwest Island were found above a colony of gannets and were
described as short-stemmed, luxurient plants growing in sheltered places
mixed with Meryta sinclairn. Mr. W. M. Fraser visited the Three
Kings in December, 1928, and reported on the cabbage trees as follows
(1929): “Cabbage trees growing to large dimensions, with many
branches bearing heads of very long leaves, and flowering profusely,
were found near running streams facing the east, and while resembling
both the Cordyline australis and C. banksti the writer is of opinion that
the Three Kings variety differs from all those found on the mainland.”
Collections made in 1934 and subsequently are in the Auckland and
Dominion Museums. These are recorded in a paper by myself (1948).

Cordyline kaspar n. sp.

Affinis C. australis sed differt foliis latioribus et brevioribus ; brac-
teis latioribus et brevoribus, basi paniculae lobatis; floribus longioribus.

A. small, widely branching tree with large terminal leaf clusters
and panicles. New branches arise from the base of the terminal clusters.
Leaves ensiform, widest above the middle, gradually tapering to an
acuminate tip and to a contracted base above an expanded sheath ; midrib
obscure above, more prominent below, widened towards the base, ribbed
on both surfaces; laminae with fine parallel ribs diverging at an angle
of about 7° from the midrib; above the sheath the leaf contracts to
about half its maximum width; length 60-65 cm., width 55-70 min..,
width above sheath 25-35 mm. Panicles terminating the branches, up
to 80cm. or more long, compound, the secondary axes branching once
or, occasionally, twice. Bracts of the rachis broadly lanceolate, leaf-like,

Rec. Auck. Inst. Mus. Vol. 4, No. 6, pp. 381-382, 25th October, 1956

382 OLIVER.

but the lower ones usually with a lobe on one or both sides, up to 26 cm.
long and 25 mm. wide, the lobes usually short but may be 70 mm. long;
upper bracts progressively shorter and proportionately broader, becom-
ing oblong with truncate bases and acute tips; bracts subtending the
tertiary axes ovate, acuminate, membranous with dark veins; there are
two small bracts in each axil of the branches. Flowers rather closely
placed on the tertiary branches and terminal part of the secondary.
Bracteoles 3, broadly ovate, acute, hyaline with dark central line, less
than half the length of the perianth segments which are narrow oblong,
obtuse; margins white, centre pale yellow with 3 dark yellow ribs.
Berry (unripe) 3-lobed, with 1-2 curved shining black seeds in each
cell, ‘Fruit white” (E. G. Turbott).

Type specimen in Botany Division, Department S.I.R., No. 87645.

The specific name is that of one of the Three Kings, Kaspar,
Melchior and Balthazar, after whom the group was named by Tasman,
who discovered it on Twelfth Night eve, 1643.

Cordyline kaspar differs from C. australis in the shorter and
broader leaves with veins arising at a wider angle from the midrib; in
the shorter and broader bracts with one or two lobes on the lower ones ;
and in the longer flowers.

Distribution: Three Kings Islands: Great, Southwest and North-
east Islands.

REFERENCES.
CHEESEMAN, T. F., 1888. Trans. N.Z. Inst., 20, p. 150.
CHEESEMAN, T. F., 1891. Trans. N.Z. Inst., 23, pp. 412, 419.
FRASER, W. M., 1929. N.Z. Jour. Sci. Tech., 11, p. 152.
OLIVER, W. R. B., 1948. Rec. Auck. Mus., 3, p. 219.

PLATE. 68

Three Kings Cabbage Tree growing in grounds of Plant Diseases Division,
Auckland. Stick is 2m. long.

Photo. J. W. Endt.

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Records of the Auckland Institute and Museum"

See other formats