m £ ,'W// m ot m X ^ ^nT^y m 'OSocJ

TION^ N0liniliSNl”NVIN0SHimS_S3 I d Vd 9 ll~L I B R A R I E S ^SMITHSONIAN ~ I NSTITUTION ^ NOlinill'

I i mill § ^fai i w i i * °

1^ LI B RAR I ES^SMITHSONIAN INSTITUTION NOIlOlliSNI NVINOSHil WS^S 3 I d V d a 11 2 L I B R A R

— u) <f> — . <j) —

TION NOlinillSNI NVINOSH1IIMS S3ldVd9ll LIBRARIES SMITHSONIAN INSTITUTION NOlinill'
z; r- v z r- z r- 2

co
X)

> |p!
j/gj x)

rn

<J) _ (/j ii </>

BRARIES SMITHSONIAN INSTITUTION NOIlfUllSNI NVIN0SH1MS S3ldVd8ll LIBRAR

%

<

z

o

if)

5

c/}

riON NoiiniiiSNi nvinoshii ws S3iavaan libraries Smithsonian institution Noiiniiis

ian libraries Smithsonian institution NoiiniiiSNi nvinoshiiws saiavaan librar

TION NOlinillSNI NVINOSH1IIMS S3ldVdail LIBRARIES SMITHSONIAN INSTITUTION NOlinill

in

in

o

N I | g rpr 1 --VI

‘2 cfl Z if) * Z

Jail LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVINOSH1IWS S3ldVd9

if) x; o) — if)

1_ LI B RAR
z

o — r'/y/~ O

TION ^ NOlinillSNI ~NVINOSHlllNS "^S 3 I d V d 8 11 LIBRARIES SMITHSONIAN INSTITUTION NOlinill!
z _ r* v z

Nosmms saiavaan libraries Smithsonian institution NouniiiSNi nvinoshuws saia

co 2 ..... <o z u* 2 co

2S*££^ > 4#r 2 ^ 5 i ^S5L2S-^ > '4

2 CO * 2 CO ‘ 2 CO 2

THSONIAN INSTITUTION NOUfUllSNI NVINOSH1IWS S3iavaail LIBRARIES SMITHSONIAN INSTI

5 f?. Z ,.v i

.« W

— \c/

o ^Cy, QCJS _ X(OitxsgX O 'it “ XiOciiSjZ' Q XgNjjOX “ O

MOSHIIWS^SB I avaan^LIBRARI E S^ SMITHSONIAN^ INSTITUTION 2 NOlinillSN I “NVINOSHIIWS^S 3 I U

33
>

33

m x^vos>^ ^ m

— co _ co

THSONIAN INSTITUTION NOlifUllSNI NVINOSH1IWS S3iavaail

CO 2 CO z

.A 2 Ax -I WgAy 2 /^STMA H Z

o \XY% X fMV'VA O siS ^S\ 1 M&SWj O

« s A» 8 £ If. Ji §

1 A//-

2 to 2 co

BRARIES SMITHSONIAN INSTI
w - 2

2

co

NOSHiitMS saiavaan libraries Smithsonian institution NouniiiSNi nvinoshiiws saia

THSONIAN

c ja$i Z JfSS -f/''. ' c

O '^C'vosv'^X ~ O ^ \K-AV,^

2

INSTITUTION NOlifUllSNI NVINOSHlIINS S3iavaail LIBRARIES SMITHSONIAN INSTI

NOSHims S3 tava a n libraries Smithsonian institution NouniiiSNi nvinoshiiws saia

CO 2 . ...CO 2 CO 2 CO

1 ^ | 1 | 1 ^ 1 1

co

VOLUME 8

Palaeontology

1965

PUBLISHED BY THE
PALAEONTOLOGICAL ASSOCIATION
LONDON

Dates of publication of parts in Volume 8

Part 1, pp. 1-198, pis. 1-22 26 February 1965

Part 2, pp. 199-373, pis. 23-47 9 July 1965

Part 3, pp. 375-576, pis. 48-79 28 October 1965

Part 4, pp. 577-767, pis. 80-1 11 15 December 1965

THIS VOLUME EDITED BY N. F. HUGHES, GWYN THOMAS, ISLES STRACHAN, AND M. R. HOUSE

© The Palaeontological Association, 1965

PRINTED IN GREAT BRITAIN

CONTENTS

Part

Adams, T. D., and Haynes, J. Foraminifera in Holocene marsh cycles at Borth, Cardigan-
shire (Wales) 1

Allen, K. C. Lower and Middle Devonian spores of North and Central Vestspitsbergen 4
Allison, R. C. Apical development in turritellid classification with a description of Cristi-
spira pugetensis gen. et sp. nov. 4

Alvin, K. L. A new fertile lycopod from the Lower Carboniferous of Scotland 2

Barker, D. See Kaye, P.

Bate, R. H. Freshwater ostracods from the Bathonian of Oxfordshire 4

Bates, D. E. B. A new Ordovician crinoid from Dolgellau, North Wales 2

Bulman, O. M. B. See Rickards, R. B.

Burgess, I. C. Calcifolium (Codiaceae) from the Upper Visean of Scotland 1

Campbell, K. S. W. An almost complete skull roof and palate of the Dipnoan
Dipnorhynchus sussmilchi (Etheridge) 4

Chaphekar, M. On the genus Pothocites Paterson 1

Clarke, R. F. A. Keuper miospores from Worcestershire, England 2

British Permian saccate and monosulcate miospores 2

Collins, J. S. H. Arcoscalpellum comptum (Withers), a species of cirripede new to the
Gault 4

Cookson, I. C. On a new species of Hoegisporis Cookson 1

Copper, P. Unusual structures in Devonian Atrypidae from England 2

Eggert, D. A. See Hibbert, F. A.

Elliott, G. F. The interrelationships of some Cretaceous Codiaceae (Calcareous Algae) 2
Felix, C. J. Neogene Tasmanites and Leiospheres from Southern Louisiana, U.S.A. 1

Ford, T. D. The Palaeoecology of the Goniatite Bed at Cowlow Nick, Castleton, Derby-
shire 1

Hallam, A. Environmental causes of stunting in living and fossil marine benthonic
invertebrates 1

Haynes, J. See Adams, T. D.

Hibbert, F. A., and Eggert, D. A. A new calaniitalean cone from the Middle Pennsyl-

vanian of Southern Illinois 4

Holdsworth, B. K. The Namurian Goniatite Nuculoceras steUarum (Bisat) 2

James, J. The development of a dicellograptid from the Balclatchie Shales of Laggan Burn 1
Jefferies, R. P. S., and Minton, R. P. The mode of life of two Jurassic species of ‘ Posi -
donia' (Bivalvia) 1

Jull, R. K. Corallum increase in Lithostrotion 2

Kaye, P., and Barker, D. Ostracoda from the Sutterby Marl (U. Aptian) of South
Lincolnshire 3

Kilenyi, T. I. Oertliana, a new ostracod genus from the Upper Jurassic of North-west
Europe 3

McAlester, A. L. Systematics, affinities, and life habits of Babinka, a transitional Ordo-
vician lucinoid bivalve 2

Medd, A. W. Dionella gen. nov. (superfamily Membraniporacea) from the Upper
Cretaceous of Europe 3

Miller, T. G. Time in stratigraphy 1

Minton, R. P. See Jefferies, R. P. S.

Pedder, A. E. H. Some North American species of the Devonian tetracoral Smithi-
phyllum 4

Page

27

687

666

281

749

355

192

634

107

294

322

629

39

358

199

16

186

132

681

226

41

156

204

375

572

231

492

113

618

CONTENTS

Part Page

Rickards, R. B. New Silurian graptolites from the Howgill Fells (Northern England) 2 247

and Bulman, O. M. B. The development of Lasiograptus harknessi (Nicholson

1867) 2 272

Roberts, J. A Lower Carboniferous fauna from Trevallyn, New South Wales 1 54

Ross, J. R. P. Homotrypa and Amplexopora ? from the Caradoc series, Shropshire 1 5

Rudwick, M. J. S. Sensory spines in the Jurassic brachiopod Acanthothiris 4 604

Spinner, E. Westphalian D megaspores from the Forest of Dean Coalfield, England 1 82

Strusz, D. L. Disphyllidae and Phacellophyllidae from the Devonian Garra formation of
New South Wales 3 518

Tavener-Smith, R. A new fenestrate bryozoan from the Lower Carboniferous of County

Fermanagh 3 478

Temple, J. T. The trilobite genus Oedicybele from the Kildare Limestone (Upper Ordo-
vician) of Eire 1 1

Tripp, R. P. Trilobites from the Albany division (Ordovician) of the Girvan district,

Ayrshire 4 577

Turner, J. Upper Jurassic and Lower Cretaceous microfossils from the Hautes-Alpes 3 391

Van Valen, L. Some European Proviverrini (Mammalia, Deltatheridia) 4 638

Walmsley, V. G. Isorthis and Salopina (Brachiopoda) in the Ludlovian of the Welsh
Borderland 3 454

Webby, B. D. Quantoxocrinus, a new Devonian inadunate crinoid from West Somerset,

England 1 11

Wiedmann, J. Origin, limits, and systematic position of Scap/iites 3 397

VOLUME 8 • PART 1

Palaeontology

FEBRUARY 1965

PUBLISHED BY THE
PALAEONTOLOGICAL ASSOCIATION
LONDON

Price £3

THE PALAEONTOLOGICAL ASSOCIATION

The Association was founded in 1957 to further the study of palaeontology. It holds
meetings and demonstrations, and publishes the quarterly journal Palaeontology.
Membership is open to individuals, institutions, libraries, &c., on payment of the
appropriate annual subscription:

Institute membership £5. 5s. (U.S. $15.50)

Ordinary membership . . . . . £3. 3 s. (U.S. $9.50)

Student membership £2. 2s. (U.S. $6.50)

There is no admission fee. Student members will be regarded as persons receiving
full-time instruction at educational institutions recognized by the Council; on first
applying for membership, they should obtain an application form from the Secretary
or the Treasurer. Subscriptions are due each January, and should be sent to the
Assistant Treasurer, Dr. C. Downie, Department of Geology, The University,
Sheffield, England.

Palaeontology is devoted to the publication of papers (preferably illustrated) on all
aspects of palaeontology and stratigraphical palaeontology. Four parts are published
each year and are sent free to all members of the Association. Members who join for
1965 will receive Volume 8, Parts 1 to 4.

All back numbers are still in print and may be ordered from B. H. Blackwell,
Broad Street, Oxford, England, at the prices shown below (post free):

Vol. 1 (for 1957-8) in 4 parts at £2 or U.S. $6.00 per part.

Vol. 2 (for 1959) in 2 parts at £2 or U.S. $6.00 per part.

Vol. 3 (for 1960) in 4 parts at £2 or U.S. $6.00 per part.

Vol. 4 (for 1961) in 4 parts at £2 or U.S. $6.00 per part.

Vol. 5 (for 1962) in 4 parts at £3 or U.S. $9.00 per part.

Vol. 6 (for 1963) in 4 parts at £3 or U.S. $9.00 per part.

Vol. 7 (for 1964) in 4 parts at £3 or U.S. $9.00 per part.

A complete set, Volumes 1-7, consists of 26 parts and costs £64 or U.S. $192.

Manuscripts on all aspects of palaeontology and stratigraphical palaeontology are
invited. They should conform in style to those already published in this journal, and
should be sent to Mr. N. F. Hughes, Sedgwick Museum, Cambridge, England.
A sheet of detailed instructions for authors will be supplied on request.

THE TRILOBITE GENUS OEDICY BELE FROM
THE KILDARE LIMESTONE (UPPER ORDOVICIAN)

OF EIRE

by J. T. TEMPLE

Abstract. A new species Oedicybele kildarensis is described from the Kildare Limestone (Upper Ordovician)
of Eire.

The genus Oedicybele Whittington, originally described from the Upper Ordovician of
Wales, has subsequently been recorded from Bornholm, Sweden, and Poland by Kielan
(1957, p. 170), who rightly considers Jemtella Thorslund a subjective synonym. A speci-
men from the Chair of Kildare Limestone (Upper Ordovician) of Eire in the collections
of the Geological Survey and Museum, London (GSM 87363), represents a new species
of Oedicybele and further extends the geographical range of the genus.

Family staurocephalidae Prantl and Pribyl 1948
Genus oedicybele Whittington 1938

Type species. Oedicybele kingi Whittington 1938, p. 446.

Synonym. Jemtella Thorslund 1940 (type species: J. clava Thorslund 1940, p. 160).

Oedicybele kildarensis sp. nov.

Plate 1, figs. 1-5

Diagnosis. A species of Oedicybele with coarse pitting on the cheeks and without genal
spines.

Holotype. Cranidium (GSM 87363), the only known specimen, from the Chair of Kildare Limestone,
Kildare, Eire. Mr. J. D. D. Smith tells me that nothing is known about its detailed history, but that
the accompanying label reads ‘Caradoc. Chair of Kildare. Staiirocephalns globiceps Portlock’.

Dimensions.

Sagittal length of glabella (including occipital ring) . . 6-3 mm.

Length of glabellar stalk ....... 2-7 mm.

Maximum width of frontal lobe ..... 5-3 mm.

Width of glabella at first glabellar lobes .... 2-3 mm.

Maximum width of cranidium (at genal angles) . . . 11-9 mm.

Description. Test preserved over much of glabella but missing over parts of both cheeks;
occipital ring and furrow incompletely preserved.

Cranidium vaulted, the distal parts inclined steeply down and anterior part of frontal
lobe of glabella overhanging. Maximum width of cranidium at genal angles. In dorsal
view glabella projects about 2 mm. in front of cheeks.

Axial furrows not deeply incised, diverging only slightly in an outwardly concave
curve from basal glabellar lobes to third glabellar furrows; beyond the latter the curve

1 Palaeontology, Vol. 8, Part 1, 1965, pp. 1-4, pi. 1.]

B 6612

B

2 PALAEONTOLOGY, VOLUME 8

of the axial furrows causes them to diverge rapidly around the frontal lobe to a maximum
at about half the length (sag.) of the frontal lobe. Axial furrows almost horizontal
posterior to third glabellar furrows, but in front they curve gradually down, becoming
finally vertical. At about half-way along the sides of the frontal lobe (measured along
the axial furrows from opposite the ends of the third glabellar furrows) the axial furrows
are deepened into shallow anterior pits (fossulae).

Glabellar furrows and lateral parts of occipital furrow deep, with apodemes. (Much
of occipital ring is broken, part of the dorsally convex occipital doublure visible on
left.) Lateral part of occipital furrow transverse; first glabellar furrows short, oblique,
bifurcating inwards and cutting off small, sub-triangular, swollen, first lobes; second
glabellar furrows transverse; third glabellar furrows pit-like and not reaching axial
furrows. Outer ends of glabellar and occipital furrows equally spaced. Glabellar stalk
reaches above level of cheeks and is roundedly triangulate in cross-section. Crest of
glabella is depressed opposite third and second glabellar furrows, the pairs of which
are thus connected by shallow transverse grooves; first glabellar furrows apparently
share a common transverse depression with occipital furrow.

Frontal lobe of glabella prominent and convex, transversely oval in plan, longer and
wider than glabellar stalk but rising only slightly higher, its great apparent convexity
being caused by the long vertical and slightly overhanging drop in front and at the sides.
Anterior margin of frontal lobe defined by facial suture. Postero-laterally the frontal
lobe bears an oblique furrow on each side, lying approximately parallel to and slightly
in front of a line joining the third glabellar furrow to the anterior pit, and with the
front end of the oblique furrow near the mid-point of this line. The left oblique furrow
is slightly concave postero-laterally, the right one sensibly straight. The parts of the
frontal lobe between these oblique furrows and the axial furrows have slight inde-
pendent convexity. A pair of tubercles side by side close together on the third glabellar
ring; on the frontal lobe a pair of faint tubercles about 1-5 mm. apart a little in front of
the highest point of the frontal lobe, in front of these a row of four faint tubercles,
and over half-way from these latter to the anterior margin of the frontal lobe a row of
six tubercles fainter still.

Posterior margin of cheek nearly transverse proximally, curving increasingly back-
wards near the rounded genal angle. Posterior border furrow transverse, sharp and
V-shaped proximally, becoming broader near genal angle where it swings slightly back
and then forward to form lateral border furrow. Posterior border narrow proximally,
widening distally slowly at first and then very rapidly near genal angle. Part of cheek
within posterior and lateral borders consists of a gently convex, more or less horizontal,
inner part, forming in dorsal view a rough equilateral triangle with straight posterior
and convex inner and outer sides, the latter passing rapidly but continuously over into
the steeply inclined outer part of cheek. Granular, distally truncated, blister-like palpe-
bral lobe is situated just postero-distal to anterior corner of left cheek triangle (right
palpebral lobe is missing). From the palpebral lobe the facial suture which forms the

EXPLANATION OF PLATE 1

Figs. 1-5. Oedicybele kildarensis sp. nov. Flolotype, GSM 87363. Chair of Kildare Limestone (Upper
Ordovician), Kildare, Eire. 1, X 10; 2-5, x 6 approx. Specimen whitened with ammonium chloride.
The colour contrast on the glabella in fig. 3 is a whitening effect.

Palaeontology, Vol. 8

PLATE !

TEMPLE, Oedicybele

< &

J. T. TEMPLE: ORDOVICIAN TRILOBITE OEDICYBELE FROM EIRE 3

margin of the specimen (the free cheeks being missing and unknown) runs backwards
in a slightly sinuous oblique line (approximately parallel in side view to the posterior
margin of the cheek), and forwards in a short outwardly concave curve to the end of the
axial furrow and then arches up as it crosses the mid-line of the cephalon. (On the right
cheek, before the test was flaked off during preparation, a faint eye-ridge ran from the
axial furrow just behind the anterior pit obliquely forward to a position corresponding
to the palpebral lobe on the left cheek.)

Cheeks within posterior and lateral borders bear coarse, closely spaced pits, the poly-
gonal ridges between which are finely granular. Frontal lobe of glabella and posterior
and lateral borders of cheeks are finely granular (other parts of test are not well enough
preserved to retain ornament). Right cheek bears four tubercles, the most posterior
being near posterior border furrow at about mid-width of cheek, another close behind
posterior facial suture about mid-way from position of palpebral lobe to border furrow,
two others near region of inflection from inner to outer part of cheek. Left cheek is
worse preserved but traces of corresponding tubercles are visible, that behind the facial
suture being testiferous and less acute than that on the right cheek, which is preserved
as an internal mould.

Discussion. Genal spines are found in both Oedicybele kingi and O. c/ava, and their
absence in the Kildare specimen is the outstanding diagnostic feature of the new species.
O. kildarensis differs from O. kingi also in having coarser pitting on the cheeks and in
lacking a median glabellar tubercle on the second glabellar ring; in the latter character
it resembles O. c/ava, from which it differs, though, in cross-section (contrast PI. 1, fig. 5
with Thorslund’s pi. 10, fig. 13). The general disposition of tubercles on the frontal lobe
(the rows of six, four, and two tubercles) is similar in all three species, and, together
with the propensity to develop tubercles on the second and third glabellar rings and on
the cheeks, is probably characteristic of the genus ; but the extent to which these tubercles
are developed is variable — they are, for instance, stronger in most specimens of O. kingi
than in O. kildarensis. O. kildarensis alone shows anterior pits, but this is probably
because of the excellence of its preservation.

The Chair of Kildare Limestone with O. kildarensis is probably Ashgillian in age.
O. kingi is also an Ashgillian species in Wales, Scandinavia, and Poland. O. c/ava,
though, is a much earlier form, coming from the Middle Ordovician (Lower Chasmops
limestone — about Nemagraptus gracilis age) of Jemtland, Sweden.

The oblique furrows on the frontal lobe of Oedicybele are reminiscent of the oblique
outer parts of the third glabellar furrows of some Phacopinae (for instance, R. and E.
Richter 1926, pi. 9, figs. 51, 52, 61), and although there are no anterior pits in the
Phacopinae, there are pits at the outer ends of the oblique third furrows in some other
Phacopids. The structure of the frontal lobe of Oedicybele may be analogous (if not
homologous) to that of the Phacopids, but there seems no doubt that the genus is
correctly placed by Kielan in the Staurocephalidae.

Acknowledgements. I am very grateful to those who have readily placed at my disposal specimens in
their charge: Dr. F. Brotzen (Swedish Geological Survey), Dr. V. Jaanusson (Paleontological Institu-
tion, Uppsala), Dr. H. Mutvei (Riksmuseum, Stockholm), Mr. J. D. D. Smith (Geological Survey and
Museum, London), and Dr. Isles Strachan (Birmingham University). I have also benefited from dis-
cussions on Oedicybele with Professor Per Thorslund and Dr. Zofia Kielan; while more recently I am
particularly indebted to Mr. R. P. Tripp for some stimulating criticism of the text.

4

PALAEONTOLOGY, VOLUME 8

REFERENCES

kielan, z. 1957. On the trilobite family Staurocephalidae. Acta palaeont. polon. 2, 155-75.
prantl, f. and pribyl, a. 1948. Classification of some Bohemian Cheiruridae(Trilobitae). Shorn. Narod.
Mus. Praze (b), 3, 1-44.

richter, r. and e. 1926. Die Trilobiten des Oberdevons. Abh. preuss. geol. Landesanst. n.f., 99, 1-314.
thorslund, p. 1940. On the Chasmops series of Jemtland and Sodermanland (Tvaren). Sverig. geol.
Unders. Afh. (c), No. 436, 1-191.

Whittington, h. b. 1938. The geology of the district around LlansantfTraid ym Mechain, Mont-
gomeryshire. Quart. J. geol. Soc. Load. 94, 423-55.

J. T. TEMPLE,

Department of Geology,
Birkbeck College,

Malet Street,
London, W.C. 1

Manuscript received 14 January 1964

HOMOTRYPA AND AMPLEXOPORA? FROM
THE CARADOC SERIES, SHROPSHIRE

by JUNE R. PHILLIPS ROSS

Abstract. Additional bryozoan species from the Harknessella subqnacliata horizon of the Hoar Edge Lime-
stone, Evenwood Quarry, include unique inverted cone-shaped colonies of Homotrypa oweni sp. nov. and two
species questionably assigned to Amplexopora, Amplexopora? evenensis sp. nov. and Amplexopora? sp. A.

Material collected in 1963 by Dr. D. Owen, Manchester University Museum, from
the upper part of the Evenwood Quarry, Shropshire, in the Harknessella subquadrata
horizon of the Hoar Edge Limestone, permits observations on bryozoan species of
Homotrypa and Amplexopora? that add to the bryozoan fauna described by Ross (1963).
This additional material in a large slab from Evenwood Quarry includes a great abun-
dance of Homotrypa oweni sp. nov. Associated with it are Phaenopora stubblefieldi Ross
and Amplexopora? evenensis sp. nov., which are common, and fragments of Amplexo-
pora thomasi Ross and Amplexopora? sp. A, which are sparse.

The three newly discovered species, Homotrypa oweni , Amplexopora? evenensis , and
Amplexopora? sp. A, are distinctly different from previously described species of these
genera. The distinctive inverted-cone shape of colonies of Homotrypa oweni permits
ready identification of this species. However, the phylogenetic affinities of this species
are at present not determinable as it appears to have little similarity with previously
described species. Amplexopora? evenensis and Amplexopora? sp. A both display
strongly crenulate zooecial walls which thicken only slightly in the narrow peripheral
regions. Such a marked crenulation of the zooecial walls as is seen in these two species
has not been noted so far in other species of the genus Amplexopora , so that these two
Caradocian species have been assigned with question to Amplexopora. The concentra-
tion of diaphragms in the subperipheral region and the narrowness of the peripheral
region are distinctive features of Amplexopora? evenensis. Amplexopora? sp. A has
isolate cystiphragms on its distal walls, which together with limited diaphragms in the
zooecial tubes and numerous diaphragms in the mesopores, single out this species.
Amplexopora? evenensis has some general similarities in the structure of the peripheral
region and size and shape of acanthopores with certain species of Amplexopora from
the Champlainian and Cincinnatian Series of North America, including such species
as A. winchelli (Ulrich), A. pustulosa Ulrich, A. ampla Ulrich and Bassler, and A. con-
voluta Bassler; however, for reasons noted above it has no strong affinities with any
of these species. Amplexopora? sp. A has little similarity to previously described species.

SYSTEMATIC DESCRIPTIONS

Family monticuliporidae

Genus homotrypa Ulrich

Homotrypa oweni sp. nov.

Plate 2, figs. 1-5; Plate 3, fig. 6; Table 1
[Palaeontology, Vol. 8, Part 1, 1965, pp. 5-10, pi. 2-3.]

6

PALAEONTOLOGY, VOLUME 8

Material. Holotype , LL 2807A; three figured paratypes, LL 2807B-D; 8 unfigured thin-sectioned
paratypes, LL 2807E-L; three unfigured hand-specimen paratypes, LL 2807M-O.

Description. Hollow, inverted-cone shaped colonies encrust calcite fragments and
crinoid columnals (PI. 2, fig. 5). The cones are about 2 to 3 mm. in diameter at the
tapered end and expand distally to 4 mm.

The subpolygonal zooecial openings interlock in a regular arrangement and cysti-
phragms are visible across the zooecial openings (PI. 2, figs. 1, 2, 5). Short zooecial
tubes are reclined for a short distance in the proximal part along the basal lamina;
zooecial tubes are straight for the remainder of their length (PL 2, figs. 3-5; PI. 3,

table 1. Measurements of Homotrypa oweni sp. nov. (in millimetres)

Catalogue no.

LL 2807 B

LL 2807 A

LL 2807C

Diameter of colony .....

Not determined

2-6 to 3-8

2-6 to 4-1

Length of colony .....

Not determined

12-5

9

No. of zooecia per 2 mm. longitudinally .

9 to 10

8 to 10

9 to 11

Diameter of zooecial opening, max. .

0-24x0-24

0-24x0-22

0-18x0-18

min.

0-13x0-16

0-18x0-18

0-15x0-13

Combined thickness of adjacent zooecial
walls in peripheral region .

0-01 to 0-02

0-01 to 0-02

0-02 to 0-03

Diameter of mesopore, max.

0-08x0-06

0-08x0-10

0-08x0-10

min.

0-04x0-04

0-05x0-05

0-03x0-03

Diameter of acanthopore

0-02

0-01 to 0-02

0-02

No. of acanthopores per zooecium .

1 to 3

2 to 3

1 to 4

No. of diaphragms per 1 mm. in zooecial
tube in peripheral region

Not determined

1 to 2

Not determined

fig. 6). Diaphragms are sparse in the zooecial tubes, commonly one diaphragm in the
upper part or reclined portion of zooecial tube and another diaphragm more distally
located; diaphragms more closely spaced in mesopores, 4 to 5 in length of mesopore.
Mesopores develop in distal region of reclined portion of zooecial tubes (PI. 2, fig. 5).
Cystiphragms occur almost always as regular overlapping structures on the distal walls
of zooecial tubes and rarely occur on the proximal zooecial walls; they develop just
above the inclined portion of zooecial tubes and succeeding cystiphragms overlap by
a half to one-third the preceding cystiphragms (PI. 2, figs. 3-5; PI. 3, fig. 6). Acantho-
pores are small and indistinct and are difficult to locate in longitudinal sections; in
tangential sections they occur at the junctions of zooecial and mesopore walls and
appear as small dark spots (PI. 2, fig. 2). Two to four mesopores occur around each
zooecium but do not isolate zooecia. Zooecial walls thin; they display laminate wall
structure in which the laminae are steeply inclined along the inner part of the zoo-

explanation OF PLATE 2

Figs. 1-5. Homotrypa oweni sp. nov. 1, Deep tangential section across more proximal and medial
parts of colony, paratype LL 2807C, X 20. 2, Tangential section showing thin-walled zooecial tubes,
small acanthopores, and cystiphragms across zooecial openings, paratype LL 2807B, X 50. 3, 4,
Parts of longitudinal sections showing basal lamina, reclined proximal portions of zooecial tubes,
and numerous cystiphragms and sparse diaphragms in more distal portions of zooecial tubes, holo-
type LL 2807A, paratype LL 2807D, x 50. 5, Oblique longitudinal section through encrusting
tapering colony, holotype LL 2807A, X 20.

Palaeontology, Vol. 8

PLATE 2

PHILLIPS ROSS, Ordovician Bryozoa

J. R. P. ROSS: CARADOCIAN BRYOZOA FROM SHROPSHIRE 7

ecial tube and curve for a short distance in the outer part of zooecial wall where they
abut against laminae of adjacent zooecial walls.

Remarks. In this species the form of the colony is most distinctive; both the cone shape
and the encrusting habit. This species is not closely comparable to any previously
described species of Homotrypa and in internal features is characterized by regular,
overlapping cystiphragms on the distal zooecial walls and sparse diaphragms in the
zooecial tubes, indistinct acanthopores, and an abundance of mesopores which are
evenly spaced between zooecial openings.

This species is also present in collection RR 2669 in the Geological Survey Museum,
London, from Wylde’s Quarry, 30 chains south-west of Harnage Grange, Shropshire
(Pocock et al. 1938, fig. 28). The species is named after Dr. D. Owen, Director, Man-
chester University Museum.

Family amplexoporidae Miller
Genus amplexopora Ulrich

Amplexopora? evenensis sp. nov.

Plate 3, figs. 1-3; Table 2

Material. Holotype, LL 2823A. Paratypes, LL 2823B figured, LL 2823C-L unfigured.

Description. Robust cylindrical to subglobular branching colonies; stems 4 to 5 mm.
in diameter; some small overgrowths. Polygonal zooecial openings are enclosed by very
slender amalgamate zooecial walls (PL 3, fig. 1). Polygonal mesopores are interspersed
between zooecial tubes and small, though at times indistinct, acanthopores commonly
penetrate the junctions of the zooecial walls and, in some instances, a second set of
acanthopores are located near the junctions of the zooecial walls and may lie near the
inner part of the zooecial walls. Long zooecial tubes with broadly crenulate zooecial
walls slope steeply from the axial region and pass into the very narrow peripheral region
where zooecial walls are closely crenulated ( PL 3, figs. 2, 3). Diaphragms are absent in the
axial region, closely spaced in the subperipheral region, and absent in the peripheral
region. Diaphragms in the subperipheral region may be fiat, curved, or overlapping
cystoidal. Acanthopores, commonly difficult to observe in longitudinal sections, appear
as narrow rods with walls of steeply inclined laminae; some extend from the sub-
peripheral region and others project above the zoarial surface. Deep tangential sections
display only a few acanthopores and expanding mesopores. Zooecial walls are slightly
thickened in the peripheral region, where they display inclined laminae in the inner
part of the walls. These laminae are curved in the outer part of the zooecial walls and
are amalgamate in a dark irregular band with laminae of the adjacent wall. Mesopores
arise in the subperipheral region, extend to the periphery, and have sparse diaphragms
in the subperipheral region. At three to four corresponding levels in the colonies,
there is bifurcation of the zooecial tubes with a marked increase in the diameter of the
zoarium; sometimes associated with this bifurcation of the tubes is a curved band of
one or two diaphragms which extends across the colony.

Remarks. This species is characterized by slender, crenulate zooecial walls, small but

8

PALAEONTOLOGY, VOLUME 8

distinct acanthopores, some of which extend into the subperipheral region, and a small
number of diaphragms in the subperipheral region of the zooecial tubes and mesopores.
Amplexopora? evenensis has little similarity with previously described species, which
generally display thicker walls, larger zooecial tubes, and a greater abundance of dia-
phragms. Amplexopora winchelli (Ulrich) (1893, ‘Middle third of Trenton shales, at St.
Paul’, pi. 27, figs. 1, 4, 6) has polygonal zooecial openings, small acanthopores, dia-
phragms in the subperipheral region, and a lack of diaphragms in the peripheral
region. However, Ulrich’s other illustrations (pi. 27, figs. 2, 3, 5, 7, 8) show thicker-
walled zooecia, more numerous and larger acanthopores, and a great abundance of

table 2 Measurements of Amplexopora? evenensis sp. nov. (in millimetres)

Catalogue no.

LL 2823 B

LL 2823 A

Diameter of colony ........

Not determined

4

No. of zooecia per 2 mm. .......

9 to 11

9 to 11

Diameter of zooecial opening, max. .....

0-27x0-28

0-29

min. .....
Combined thickness of adjacent zooecial walls in peripheral

0-16x0-20

0-17

region ..........

0-02

0 01 to 0-03

Diameter of mesopore, max. ......

0-16x0-05

Not determined

min. ......

0-05x0-02

0-04

Diameter of acanthopore .......

0-01 to 0-03

0-01 to 0-02

No. of acanthopores per zooecium .....
No. of diaphragms per 1 mm. in zooecial tube in subperipheral

2 to 5

Not determined

region ..........

Not determined

3 to 4

diaphragms. A. pustulosa Ulrich (1890, ‘Cincinnati group. Hanover, Clarksville, and
other localities in Ohio ’) likewise, while displaying some similarities in polygonal zooecial
openings and small acanthopores, is considerably more robust in its zooecial and
zoarial structures. A. ampla Ulrich and Bassler (1904, ‘Fairmount beds, Cincinnati,
Ohio’) has larger zooecial tubes and diaphragms in the axial region. Slender zooecial
walls, narrow peripheral region, and small acanthopores suggest similarities with
A.? evenensis.

This new species takes its name from Evenwood Quarry.

EXPLANATION OF PLATE 3

Figs. 1-3. Amplexopora? evenensis sp. nov. 1, Deep tangential section showing polygonal zooecial
openings and mesopores, and small acanthopores at junctions of zooecial walls, paratype LL
2823B, x 50. 2, Peripheral region of part of longitudinal section showing thickened crenulate
zooecial walls, holotype LL 2823A, X 50. 3, Part of longitudinal section showing sparse diaphragms
in axial region and concentration of diaphragms in subperipheral region, holotype LL 2823A, x 20.

Figs. 4, 5, 7, 8. Amplexopora? sp. A. 4, Longitudinal section showing crenulate zooecial walls and
sparse diaphragms in zooecial tubes and mesopores and occasional cystiphragm in zooecial tube,
LL 2824B, X 50. 5, Longitudinal section showing general aspect of zooecia and mesopores, LL
2824B, x 20. 7, Deep tangential section showing round to subpolygonal zooecial openings, poly-
gonal mesopores, and an occasional acanthopore at junction of zooecial wall, LL 2824A, x 50.
8. Longitudinal section showing part of overgrowth with very distinct cystiphragms, LL 2824B, X 50.

Fig. 6. Homotrypa oweni sp. nov. Part of longitudinal section showing laminate structure of zooecial
walls and cystiphragms in zooecial tubes, paratype LL 2807D, X 100.

Palaeontology, Vol. 8

PLATE 3

PHILLIPS ROSS, Ordovician Bryozoa

J. R. P. ROSS: CAR ADOCI AN BRYOZOA FROM SHROPSHIRE

9

Amplexopora? sp. A
Plate 3, figs. 4, 5, 7, 8; Table 3
Material. Figured specimens, LL 2824A, 2824B.

Description. Very slender cylindrical branches with overgrowths (PI. 3, figs. 5, 8). Round
to subpolygonal zooecial openings are enclosed by slender amalgamate zooecial walls
(PI. 3, fig. 7). Mesopores and occasionally acanthopores occur at junctions of zooecial
walls. Long slender zooecial tubes with slightly crenulate walls slope steeply from the
axial region and curve abruptly into subperipheral region (PI. 3, fig. 5). Narrow peri-
pheral region displays limited thickening of crenulate zooecial walls and extension of

Table 3. Measurements of Amplexopora? sp. A (in millimetres)

Catalogue no.

LL 2824 A

LL 2824B

Diameter of colony ........

Not determined

1-6

No. of zooecia per 2 mm. longitudinally ....

10 to 11

11 to 13

Diameter of zooecial opening, max. .....

018x016

0-15

min. .....
Combined thickness of adjacent zooecial walls in peripheral

010x013

0-09

region ..........

0 05 to 0 08

0 01 to 0-02

Diameter of mesopore, max. ......

0-08x0- 10

0 1 1

min. ......

0-02x0-02

0-04

Diameter of acanthopore .......

0-01 to 0-02

0-02

No. of acanthopores per zooecium .....

2 to 3

Not determined

No. of diaphragms in zooecial tube in peripheral region .

Not determined

4 in 0-3 mm.

mesopores from subperipheral region. Occasionally one or two diaphragms are present
in subperipheral region or peripheral region of zooecial tube. Diaphragms are abun-
dant and closely spaced in mesopores (PI. 3, figs. 4, 5). An occasional isolate or over-
lapping set of two or three cystiphragms occur on distal walls of some zooecial tubes
(PI. 3, figs. 4, 5). Acanthopores extending from subperipheral region to the periphery
are almost as wide as the zooecial walls but are not readily observable in longitudinal
section. The slightly thickened zooecial walls of the peripheral region have a wall struc-
ture as in Amplexopora? evenensis.

Remarks. Amplexopora? sp. A has certain similarities to A.? evenensis in the arrange-
ment of the zooecial tubes, crenulate zooecial walls, laminate zooecial wall structure,
and arrangement of the zooecial openings. It differs from A.? evenensis in having more
slender colonies, narrower zooecial tubes, very few diaphragms in the zooecial tubes,
cystiphragms in the zooecial tubes, thicker zooecial walls in the peripheral region, and
an abundance of diaphragms in the mesopores. The species has little similarity to pre-
viously described species of Amplexopora and its assignment to the genus is questioned.

Acknowledgements. I am most grateful to Dr. D. E. Owen, Director, Manchester Museum, for kindly
sending this material to me for examination and study.

Repositorv. All the specimens referred to are in the Manchester Museum, The University, Manchester
13.

10

PALAEONTOLOGY, VOLUME 8

REFERENCES

pocock, r. w., whitehead, T. H., wedd, c. b., and Robertson, T. 1938. Shrewsbury District including
the Hanwood Coalfield. Mem. geol. Surv. U.K.

ross, J. R. p. 1963. Trepostome Bryozoa from Caradoc Series, Shropshire. Palaeontology, 6, 1-11,
pi. 1-3.

ulrich, e. o. 1890. Palaeozoic Bryozoa. Illinois geol. Surv., Geologv and Palaeontology, 8, 283-688,
pi. 29-78.

1893. On Lower Silurian Bryozoa of Minnesota. Minnesota geol. and nat. Hist. Surv., Geology

of Minnesota, 3 (1), Paleontology, 96-332, 28 pi., 1895. (Author's separate, 1893).

and bassler, r. s. 1904. A revision of the Paleozoic Bryozoa. Pt. 2, Trepostomata. Smithson.

misc. Coll. 47, 15-55, pi. 6-14.

JUNE R. PHILLIPS ROSS

Department of Geology,
University of Illinois,
Urbana, Illinois, U.S.A.

Manuscript received 13 January 1964

QU AN TO XO C R IN U S, A NEW DEVONIAN
INADUNATE CRINGED FROM WEST SOMERSET,

ENGLAND

by B. D. WEBBY

Abstract. Quantoxocrinus ussheri gen. et sp. nov. is described from the upper part of the Ilfracombe Beds
(Upper Givetian or Lower Frasnian) in the Quantock Hills, west Somerset, and the palaeoecological signi-
ficance of its occurrence is discussed.

During recent geological studies in the Quantock Hills, numerous external moulds of
a small, delicate crinoid were found in a thin brownish-grey, weathered (originally cal-
careous) siltstone. The crinoid bed, only 6 inches thick, and seemingly underlain and
overlain by unfossiliferous siltstones and sandstones, is exposed beside a forestry track
in Wind Down Plantation (National Grid Ref. ST221339). Much of the material is too
poorly preserved for detailed study, but a limited number of good specimens were
extracted, and form the basis of the present paper.

This new inadunate crinoid occurs stratigraphically towards the middle of the Leigh-
land Beds, in the upper part of the Ilfracombe Beds, and lies near, probably just below,
the Givetian-Frasnian boundary. A similar inadunate, Decadocrinus oaktrovensis, has
previously been described from the lower part of the Ilfracombe Beds of west Somerset,
in the Oaktrow Sandstone of probable Middle Givetian age (Webby 1962, 1964).

The Ilfracombe Beds of west Somerset are a thick, argillaceous and arenaceous suc-
cession, containing several important developments of limestone at different horizons.
These limestones are composed of abundant crinoidal fragments, mostly disarticulated
columnals (it is unusual to find articulated columnals). Evidence from the crinoidal
remains suggests that the bulk of these limestones accumulated in a shallow-water
environment subjected to strong wave or current action. Scattered, disarticulated
columnals are found on bedding planes at irregular intervals throughout the rest of
the Ilfracombe succession, and these occurrences similarly suggest moderate to strong
action. However, at two localities, one in the Oaktrow Sandstone of the Brendon Hills,
and the other in the Leighland Beds of the Quantocks, small, fragile, complete and partly
complete crinoids occur abundantly within a single thin bed of siltstone. In the Oaktrow
occurrence, the crinoids are mainly randomly distributed through the siltstone, with
broken crown and stem fragments associated with the more complete specimens. The
crowns of many of these crinoids have parted from their stems, but have otherwise
remained intact. The crinoid occurrence in the Quantocks bears many similarities to
that of Oaktrow, but there is a slightly stronger alignment of specimens, and a tendency
for crowns and, more especially, stems to lie on bedding planes. Also, a greater propor-
tion of stems to crowns seem to be preserved. These gregarious and delicate crinoids
bear cirri, suggesting that they were anchored in the muddy silt of the sea bed. They
either established themselves on the sea floor during a period of reduced sedimentation
and very gentle current action or, alternatively, they colonized a shallow depression on
the sea floor, presumably protected from the stronger currents (Laudon 1957, p. 963).

fPalaeontoIogy, Vol. 8, Part 1, 1965, pp. 11-15, pi. 4.]

12

PALAEONTOLOGY, VOLUME 8

Apparently, after a comparatively short period of occupation, the crinoids were over-
whelmed suddenly by a stronger, sediment-laden current. The Oaktrow fauna was
probably buried more or less where it was overwhelmed, whilst the Quantock fauna,
with its crude alignment of specimens and disarticulation of stems from crowns, was
perhaps transported a short distance by the current. The greater proportion of stems to
crowns in the Quantock fauna may be the result of the break up of the more fragile
crowns during entombment or, perhaps, the crowns detached themselves from their
stems and floated away in the stronger currents preceding the entombment.

SYSTEMATIC DESCRIPTION

The numbers of specimens in the Geology Department collection. University of
Bristol, have the prefix bu.

Subclass inadunata Wachsmuth and Springer
Order cladoidea Moore and Laudon
Suborder dendrocrinoidea Bather
Genus quantoxocrinus gen. nov.

Type species. Q. ussheri sp. nov.

Diagnosis. Small, slender, ten-armed dendrocrinoid, having prominent infrabasals; three
anal plates in the dorsal cup, the radianal similar in size, or slightly larger than anal X,
and a large anal sac, plicated distally; arms branching isotomously on the fourth or fifth
primibrach, and bearing pinnules; stem, proximally, composed of pentagonal colum-
nals; distally, round columnals with crenulate sutures, and cirri.

Discussion. The genus bears similarities to a number of dendrocrinoid genera, including
Iteacrinus Goldring, Decadocrinus Wachsmuth and Springer, Denariocrinus Schmidt,
and Rhadinocrinus Jaekel. Iteacrinus (type species I. flagellum) exhibits the closest resem-
blance, but differs in having heterotonious branching, with ten main rami bearing long,
slender, unforked ramules at regular intervals alternately on each side (Goldring 1923,
p. 344). Devonian species of Decadocrinus differ in exhibiting isotomous branching on
the second primibrach and, judging from D. oaktrovensis, rounded columnals with
smooth articular faces in the distal part of the stem. Denariocrinus (type species D. ferula)
also shows isotomous branching on the second primibrach, and has four anal plates in
the dorsal cup (Schmidt 1941, p. 163). Denariocrinus was previously suggested as a
synonym of Decadocrinus (Webby 1962, p. 539), but perhaps it should be retained as a
separate genus, characterized by having four anal plates in the dorsal cup, while
Decadocrinus has only two or three. Rhadinocrinus (type species R. rhenanus) differs from
Quantoxocrinus in exhibiting forked ramules at widely spaced intervals, a radianal smaller
than anal X and, proximally, a round stem (Jaekel 1895, p. 87; Schmidt 1941, p. 152).
The problematical species R. minae (Schmidt) differs in the latter two characters, viz.,
a radianal smaller than anal X, and a round stem (Haarmann 1922, p. 29).

Iteacrinus , Decadocrinus, Denariocrinus, and Rhadinocrinus have each been classified
in a different family of the Suborder Dendrocrinoidea, Iteacrinus being grouped in the
Dendrocrinidae by Moore (1962, p. 37), Decadocrinus in the Scytalocrinidae by Moore
and Laudon (1943, p. 59), Denariocrinus in the Poteriocrinitidae by Schmidt (1941,
p. 163), and Rhadinocrinus in the Botryocrinidae by both Schmidt (1941, p. 152), and

B. D. WEBBY: QUANTOXOCRINUS, A DEVONIAN CRINOID FROM SOMERSET 13

Moore and Laudon (1943, p. 54). Decadocrinus and Denariocrinus exhibit such small
differences that they should definitely be referred to the one family, and Quantoxocrinus
should either be assigned to this family or to the Dendrocrinidae.

Quantoxocrinus ussheri sp. nov.

Plate 4; text-fig. 1

Diagnosis. A species of Quantoxocrinus with dorsal cup from as wide as high to a little
wider than high; infrabasals prominent; basals higher than wide, depressed at sutures;
radials convex, as wide as high; brachials usually as wide as high, four to five primi-
brachs, and pinnules, spine-like on each alternate brachial; anal sac large, erect or

D □

text-fig. 1. Diagram showing the arrangement of plates in the dorsal cup, the
proximal part of the anal sac, and the proximal part of an arm of Quantoxocrinus
ussheri gen. et sp. nov. The radianal is shown ruled, and anal X stippled.

recurved, anal plates becoming plicated distally; proximally, the stem is composed of
pentagonal columnals with slightly stellate nodals and low internodals; distally, colum-
nals are round, with articular faces marked by radiating striae on outer edges; cirri
short and tapering, borne on a few high nodals.

Description. Dorsal cup small, conical, sutures depressed, particularly between basals.
Infrabasals moderately large, pentagonal, a little wider than high. Basals hexagonal,
except for heptagonal posterior basal and right posterior basal; convex, higher than
wide. Radials pentagonal, except for trapezoidal right posterior radial; convex, as wide
as high; facets curved, extending the whole width of radial. Primibrachs four or five;
quadrangular, apart from pentagonal axillary; arms long, slender and gradually taper-
ing above the single isotomous branching; twenty-two secundibrachs in an arm length
of 23 mm. above axillary of one specimen (BU 18432); brachials uniserial, smooth,
convexly rounded, pinnulate, varying from as high as wide to slightly higher than wide.
Pinnules on alternate brachials appearing as moderately long, spine-like projections;

14

PALAEONTOLOGY, VOLUME 8

pinnule segments not clearly observed. Posterior interradius with pentagonal radianal
below level of the radials, slightly smaller than right posterior radial; hexagonal anal X,
similar in size to the radianal; slightly smaller, hexagonal, first tube plate. Anal sac
large, erect or recurved; separate rows of tube plates rest on anal X and radianal; the
plates in these rows decrease uniformly in size and become slightly plicated distally;
these two rows form a prominent ridge, on either side of which are depressions and
secondary ridges formed by additional rows of more intensely plicated tube plates; erect
anal sac observed to a height of 27 mm. (BU 18438), and others up to 5 mm. broad at
the level of the main branching of arms. Proximal part of stem (at least 40 mm. in length)
composed of pentagonal columnals, alternating between high, slightly stellate nodals
and low internodals; distally round, with columnals including a few high, cirri-bearing
nodals; articular faces of distal columnals marked by radiating striae restricted to outer
edges, and forming a crenulate suture; axial canal circular; cirri rather short and
sharply tapered, up to 15 mm. long; composed of numerous, thin disc-like segments
similar to distal columnals but smaller.

Dimensions (mm.)

Holotvpe
BU 18431

BU 18432

BU 18433

BU 18434

BU 18435

BU 18436

BU 18437

He

5-5

7-6*

50

4-3

51

4-2

5-9

Hb

2-6

2-2

2-1

21

2-2

21

2-8

Hib

1-4

1 9

1 -5

1-3

1-5

1-3

1-8

Wcr

5-8

5-2

51

5-3

6-2

60

+4-5

Web

2-0

2-0

1-9

1-8

2-0

1-9

1 9

He , height of dorsal cup; Hb, average height of basals; Hib, average height of infrabasals; Wcr, width
of dorsal cup at the level of the radial facets; Web, width of dorsal cup at base of cup; *, distorted
dorsal cup.

EXPLANATION OF PLATE 4

All figures x3-5; from latex (Revultex) casts of external moulds.

Figs. 1-10. Quantoxocrinus ussheri gen. et sp. nov. 1, Holotype, BU 18431, posterior view, showing
arrangement of plates in the posterior interradius, the proximal part of the anal sac, and the iso-
tomous branching of an arm on the fourth primibrach. The anal sac is broad and apparently erect,
composed of several rows of plates, gradually diminishing in size distally. 2, Paratype, BU 18432,
right postero-lateral view, showing proximal part of stem, composed of stellate columnals, elon-
gated (probably tectonically distorted ), poorly preserved dorsal cup, and long, slender arms, branching
isotomously on the fourth primibrach. 3, Paratype, BU 18433, right lateral view, showing compara-
tively large infrabasal, basals, and radials, and isotomous branching on the fifth primibrach. 4, Para-
type, BU 18434, posterior view, showing arrangement of plates in the dorsal cup, with a moderately
large infrabasal, gently convex basals and radials, and a radianal similar in size to anal X; anal sac
recurved and distally plicated. Primibrachs of left posterior arm appear to be slightly distorted; they
exhibit long, slender pinnules. 5, Paratype, BU 1 8435, right postero-lateral view, showing stem with
alternating high, stellate nodals and low internodals, large infrabasals, basals with depressed sutures,
radials, and plates of the posterior interradius. 6, Paratype, BU 18452, view of a portion of a plicated
anal sac, and the distal part of the stem of another specimen. 7, Paratype, BU 18454, view of
moderately long, spine-like pinnules, occurring on alternate secundibrachs. 8, Paratype, BU 18459,
view of distal part of stem with round columnals of irregular height, and crenulated sutures. 9, Para-
type, BU 18453, view of distal part of stem, showing a high nodal which bears a short, tapering
cirrus. 10, Paratype, BU 18451, view of distal stem columnals, showing the articular face of a cirrus-
bearing nodal with radial striae restricted to the outer edge, and a circular axial canal.

Palaeontology, Vol. 8

PLATE 4

i£. u i

Lpi

r k

ft ^

WEBBY, Devonian inadunate crinoid

B. D. WEBBY: QUANTOXOC RINUS, A DEVONIAN CRINOID FROM SOMERSET 15

Holotype. BU 18431, Plate 4, fig. 1. Paratypes. BU 18432-40; 18444-5; 18449-54; 18459; 18461-2.

Derivation of specific name. After W. A. E. Ussher, in recognition of his important contributions to the
knowledge of the geology of west Somerset.

Acknowledgements. This study has been carried out in the University of Bristol with the financial support
of a grant for special researches from the Department of Scientific and Industrial Research.

I am grateful to Professor W. F. Whittard, F.R.S., and to Dr. W. H. C. Ramsbottom, for reading
and criticizing the manuscript; to Dr. R. J. G. Savage for useful discussions on palaeoecology of
crinoids; and to Mr. E. W. Seavill for supplying the photographs.

REFERENCES

goldring, w. 1923. The Devonian crinoids of the State of New York. Mem. N.Y. St. Mas. 16, 1-670,
pi. 1-60.

haarmann, e. 1922. Die Botryocriniden und Lophocriniden des rheinischen Devons. Jb. preuss. geol.
Landesanst. 41 (1), 1-87, pi. 1-6.

jaekel, o. 1895. Beitrage zur Kenntnis der Paliiozoischen Crinoiden Deutschlands. Paldont. Abh. 7 (1),
3-116, pi. 1-10.

laudon, l. r. 1957. Crinoids, In ladd, h. s. Ed., Treatise on Marine Ecology and Paleoecology, 2.
Mem. geol. Soc. Amer. 67, 961-72.

moore, r. c. 1962. Ray structures of some inadunate crinoids. Uni v. Kansas Paleont. Contrib., Echino-
dermata, Art. 5, 1-47, pi. 1-4.

and laudon, l. r. 1943. Evolution and classification of Paleozoic crinoids. Spec. Pap. geol. Soc.

Amer. 46, 1-153.

schmidt, w. e. 1941. Die Crinoideen des Rheinischen Devons. II Teil. Abh. Reichsst. Bodenfi, n.f.,
182, 1-253, pi. 1-26.

webby, b. d. 1962. A Middle Devonian inadunate crinoid from west Somerset, England. Palaeontology,
4, 538-41, pi. 67.

1964. Devonian corals and brachiopods from the Brendon Hills, west Somerset. Ibid. 7, 1-22,

pi. 1.

B. D. WEBBY,

Department of Geology,
The University,
Bristol 8

Manuscript received 13 February 1964

NEOGENE TASMANITES AND LEIOSPHERES FROM
SOUTHERN LOUISIANA, U.S.A.

by CHARLES J. FELIX

Abstract. Tasmanites , Tytthodiscus , and leiospheres assignable to Leiosphaeridia , are present in Neogene shales
of the Gulf Coast area of southern Louisiana, U.S.A. Seven species of Tasmanites , one species of Tytthodiscus,
and two species of Leiosphaeridia are described from sidewall core samples; all except Tytthodiscus are new.

The microfossils considered here are important representatives of the microflora and
fauna of various brackish-marine clays and shales investigated by the Sun Oil Company
palynological research group. All samples are believed to be from upper Miocene sedi-
ments. However, the Plio-Miocene boundary is still subject to debate in the area of
study, and Neogene seems most applicable. The wells utilized in this investigation are
listed below with numerical designations corresponding to their locations on text-fig. 1.

1. Sun Chacahoula L. R. and P. No. 7 (Sec. 66, 15 S., 15 E.) La Fourche Parish, Loui-
siana.

2. Sun Belle Isle No. 1 (Sec. 26, 17 S., 10 E.) St. Mary Parish, Louisiana.

3. Humble No. 1 (Sec. 12, 22 S., 16 E.) Terrebonne Parish, Louisiana.

4. Sun Lake Pelto No. 1 (Sec. 10, 23 S., 18 E.) Terrebonne Parish, Louisiana.

All of the samples used in this study were processed by means of mechanical disaggrega-
tion without the use of acids as described by Felix (1963).

For years there has been debate concerning the microfossils assigned by different in-
vestigators to Tasmanites and to the various leiosphere genera. Schopf, Wilson, and
Bentall (1944) thoroughly covered the nomenclature problem of Tasmanites, and Schopf
(1957) later surveyed contributions dealing with this problematic genus. Winslow (1962)
has made the most systematic survey of Tasmanites in upper Devonian and lower Missis-
sippian beds. Eisenack (1938) identified as Leiosphaera a microfossil resembling Tas-
manites in some respects. Leiosphaera subsequently proved to be invalid and Eisenack
(1958a, 19586) established Leiosphaeridia as a leiosphere genus to include forms not
attributable to Tasmanites. These developments and subsequent contributions to leio-
sphere taxonomy have been reviewed by Staplin (1961), and more recently Downieand
Sarjeant (1963) in a critical analysis of leiosphere taxonomy have brought some measure
of order to the taxonomy. However, there is still general disagreement among current
investigators on both nomenclature and classification of these organisms.

Tasmanites has been most notably associated with the Devonian-Mississippian black
shales by American geologists. It has proven to be of practical value in stratigraphic
studies, and Jodry and Campau (1961) have surveyed the useful potential of the genus.
Even though it is more often associated with Paleozoic sediments, there have been
references to Tasmanites being present in younger sediments. Radforth and Rouse (1956)
noted the projection of the Tasmanites range into Cretaceous and Tertiary strata. Rouse
(pers. communication) also confirmed the presence of Tasmanites in the Onakawana

(Palaeontology, Vol. 8, Part 1, 1965, pp. 16-26, pi. 5-8.]

C. J. FELIX: TASMANITES AND LEIOSPHERES FROM LOUISIANA 17

complex of the Tertiary of Ontario, Canada, while recently Winslow (1962) reported
probable early Cretaceous occurrences in Alaska. It has also been recorded from the
middle Silurian of Illinois in the U.S.A., the lower Carboniferous of England, and the
original identification of Tasmanites was from beds in Tasmania dated as Permian.

For several years the author has found Tasmanites and leiospheres in different Ter-
tiary sediments, and they have been especially numerous in subsurface shales of Neogene
age in the Gulf of Mexico coastal area of southern Louisiana. Study of this unexpected

text-fig. 1. Locality map, showing area of study and wells mentioned in the present investigation.

occurrence was approached with consideration of possible contamination or re-
deposition. The former possibility must be ruled out in view of the use of core samples
and the degree of carefulness employed in sample preparation. Moreover, thousands of
samples were processed and many were reprocessed with the same end results. Redeposi-
tion was carefully considered since these microfossils do have a long history of such
re-occurrence. However, in the final analysis there is no evidence for redeposition of the
specimens considered here. These Tertiary specimens do not fit into any forms known
from the Paleozoic; their association with other microfossils to form characteristic, and
easily recognized, assemblages of specific environments does not remotely resemble
associations of the classical Paleozoic forms ; and the several other geological and paleon-
tological disciplines of this research laboratory collaborating on the same research
problem have not established any evidence for redeposition. Recently the report by
Wall (1962) of similar organisms in the modern seas, and this writer’s find of Tasmanites
and leiosphere-like bodies in the modern Gulf of Mexico sediments lends additional
credence to their existence in the Tertiary.

The confusion concerning the systematics of these fossils renders naming difficult.
On the basis of Schopf, Wilson, and Bentall’s (1944) comprehensive study of Tasmanites
and Winslow’s (1962) emendation of Tasmanites huronensis , the genus seems to be
amply described. Admittedly some microfossil forms do differ markedly from Tasmanites

B 6612

c

18

PALAEONTOLOGY, VOLUME 8

and some of the differences of opinion concerning the taxonomic status of the leiospheres
are probably justified. It appears that punctation of the disseminule wall is the major
feature of Tasmanites. Although degrees of punctation vary in Tasmanites, it is never
a feature of the leiospheres. The writer does not propose to emend existing taxonomy,
nor contribute new genera to the already existing superfluity, since ample classification
niches exist for the purposes of this study. Inasmuch as punctations are a feature of
the wall in Tasmanites , those microfossils with such optical properties will be treated as
Tasmanites, with the one exception being assignable to Tytthodiscus. Those thin-walled
disseminules without wall punctations are considered here as leiospherids. Many
resemble described specimens of Leiosphaeridia, in which Eisenack featured the pylome
as a systematic character, but such a structure has not been observed in any of the
thousands of specimens examined. Hence it would seem that Timofeev’s (1959) Proto-
Ieiosphaeridium would be most applicable. However, Timofeev indicated a maximum
size range of about 50 /x for Protoleiosphaeridium, and the author is in agreement with
its rejection by Downie and Sarjeant (1963) as a synonym of Leiosphaeridia. Their
emended diagnosis of Leiosphaeridia to include specimens with or without pylomes
is also accepted. Numerous leiospheres have been noted in Paleozoic studies in this
laboratory in which the pylome is present, and some of the excellent illustrations of
Eisenack (1958a, 19586) leave little doubt of the pylome’s existence. However, in the
thousands of specimens viewed in this investigation, the pylome was a very questionable
feature and never observed with certainty.

Two of the Neogene species, T. fissura and T. balteus, are characterized in part by
splitting or collapsing of the cell wall. Wall (1962) has discussed a similar splitting as
a possible suture, comparable to the pylome, which Eisenack (1958a, 1962) has con-
sidered to be a germinal opening ‘Schlupflocher’. Eisenack (1962, p. 74) has also
suggested this bursting of the test to be a suture but has never demonstrated this. The
presence of any such germinal feature in Tasmanites has never been reliably shown. Even
considering the questionable presence of a pylome in a few instances, this does not
account for the many examples in the literature which possess neither pylomes or suture-
like devices. The author is of the opinion that splitting of the cell walls in T. fissura and
T. balteus is due to structural faults only. In the study of plant spores several species
have been noted which have a tendency to compress into certain shapes or to develop
uniform splits due to preservational compaction, even though they may possess well
developed trilete germinal apertures. Chaloner (1953) has demonstrated such a selective
orientation under compaction in Sigillarian spores. It does not seem unrealistic to draw
such a comparison between Tasmanites and the trilete plant spores in view of their com-
parable modes of preservation, sizes, and similar morphology in many instances.

The determination of species in Tasmanites and the leiospheres is especially difficult
due to the absence of haptotypic features. Species distinction must necessarily be made
on very small differences, and there are relatively few distinctive emphytic characteristics
available to provide sharply defined dissimilarities. The species defined in this study
have been established with consideration of several points. The dimensions are signi-
ficant, and total diameters are variable. The relatively small sizes of the Neogene species
are noteworthy, for they scarcely attain one-half the size of most Paleozoic forms. Wall
thickness has a wide range, and the outer wall may be conspicuously thick or so thin
as to be nearly indiscernible. In Tasmanites the wall punctae are important in numbers.

C. J. FELIX: TASMANITES AND LEIOSPHERES FROM LOUISIANA 19

spacing, and dimensions. They vary in numbers from rare to a high degree of density.
A noticeable difference exists in the type of punctae orifices on the body surface. The
canals are often a prominent feature and differ in angle of inclination, extent of pene-
tration through the wall, and degree of taper. The manner of folding and the colour
are also considered although some investigators question the usefulness of the latter
feature. Winslow (1962) has noted that both features are indirectly related to wall thick-
ness versus diameter.

TASMANITES GROUP
Genus tasmanites Newton 1875
Tasmanites porosus sp. nov.

Plate 5, figs. 1, 2

Holotype. Slide 91-1, location 21-8 X 107 (Ref. 20-2x 117) (PI. 5, fig. 1). Core, 11,958 ft. Lake Pelto
well No. 1, Terrebonne Parish, Louisiana.

Diagnosis. Spherical, 1 50—165 /x diameter (12 specimens measured) holotype 160 p.
Wall 10-12 p thick, penetrated by distinct canals; canals always appear to pass from
the inner wall surface, with most not reaching the outer surface. The canals are uni-
formly distributed 10-15 p apart; they possess a slight angle of inclination and may pass
through more than one optical plane. They are 1-1-5 p in width, with appreciable taper.
Body surface laevigate, characterized only by scattered pores representing termination
of some canals. At high magnification they are observed to possess a slightly raised rim
some 0-5 p in width, and pores and encircling rim or border are 2-3-5 p in diameter.
Body outline is very regular and never characterized by folding. Colour yellow to orange
in transmitted light.

Comparison. T. porosus is easily distinguished from the other Neogene entities by its
rather large, uniformly spaced wall canals and the conspicuous bordered pores on its
surface. It shows some similarity to T. roxoi Sommer (1953, 1956), but the latter species
has a minimum size of 370 p and possesses much larger and more closely spaced wall
canals.

Tasmanites fissura sp. nov.

Plate 5, fig. 3

Holotype. Slide 475-1, location 42-6x 1 1 2-4 (Ref. 19x 117-6) (PI. 5, fig. 3). Core, 8,536 ft. Belle Isle
well No. 1, St. Mary Parish, Louisiana.

Diagnosis. Oval, 1 70-300 p diameter, average about 230 p (25 specimens measured)
holotype 276 pX 293 p. Wall indistinct, difficult to discern, 6-12 p thick (11 p in holo-
type). Canals rare, scattered; few ill-defined pores on wall surface visible at high
magnification. Wall with occasional minor folding; surface usually has weathered or
corroded appearance. Colour light yellow to yellow-orange in transmitted light.

Remarks. This species is distinguished by the wall usually being ruptured so as to present
an appearance suggestive of a split-open grape hull. It is usually split at only one point,
rarely are two such breaks observed, and the rupture extends a distance of one-third
to one-half the body diameter. Although the canals and pores are difficult to ascertain,
a scattered few are present on every specimen.

20

PALAEONTOLOGY, VOLUME 8

Comparison. This species is among the largest in size of the Neogene entities. T. balteus
has a similar size range and rare wall canals, as well as an affinity for a collapsed or split
wall. However, the possibility of misidentification is remote since T. balteus has a well
defined and much thicker wall and a more regular breaking of the cell wall. The only
published reference of Tasmanites similar to this species appears to be from the Winni-
pegosis formation of the middle Devonian from the Williston Basin of North America,
where Jodry and Campau (1961) figured a representative within the size range of T.
fissura and possessing a similar type of wall rupture. However, their Tasmanites illustra-
tions were not accompanied by formal descriptions.

Tasmanites corrugatus sp. nov.

Plate 5, fig. 4

Holotype. Slide 88-1, location 27 x 116-1 (Ref. 17x 118) (PI. 5, fig. 4). Core, 4,769 ft. Lake Pelto well
No. 1, Terrebonne Parish, Louisiana.

Diagnosis. Spherical, 130-190 p diameter (15 specimens measured) holotype 1 88 /u..
Wall usually distinct about entire body periphery and 5-10 p thick (9 p in holotype).
Wall with narrow, inclined canals, passing from inner wall surface but seldom pene-
trating outer surface; canals scattered, usually about 10 p apart. Body outline irregular
and folded. Colour light yellow in transmitted light.

Comparison. The species bears a general resemblance to Leiosphaeridia plicata but is
not as plicated and crumpled, and the wall canals are always very easily distinguished
in T. corrugatus. The canals resemble those of T.fulgidus but the wall structure, size, and
the degree of folding are distinctly different from those of that species. It also bears a
general resemblance to both T. sinuosus Winslow and to T. mourai Sommers. Winslow
(1962) noted the similarity between T. sinuosus and T. mourai and chose to treat them
as distinct partly on the age differences of sediments and the degree of geographic
separation. However, T. corrugatus is completely outside of the size range of T. mourai ,
being very small in comparison. The dense, well-marked punctae of T. mourai do not
compare with the scattered punctae of T. corrugatus. T. corrugatus is included within
the lower size limits of T. sinuosus, but the punctae of the latter are numerous, tubular,
and straight, while those of T. corrugatus are relatively few, scattered, and possess a
pronounced angle of inclination.

Tasmanites usitatus sp. nov.

Plate 6, fig. 5

Holotype. Slide 485-1, location 33-9X 124-5 (Ref. 19x 118-3) (PI. 6, fig. 5). Core, 9,342 ft. Belle Isle
well No. 1, St. Mary Parish, Louisiana.

EXPLANATION OF PLATE 5

Figs. 1, 2. Tasmanites porosus sp. nov., holotype. 1, Entire specimen, X 500. 2, Wall detail, X750.
Slide 91-1, location 21 -8 X 107 (Ref. 20-2 X 117).

Fig. 3. Tasmanites fissura sp. nov., holotype. X 300. Slide 475-1, location 42-6 x 1 12-4 (Ref. 19x1 17-6).
Fig. 4. Tasmanites corrugatus sp. nov., holotype. X 500. Slide 88-1, location 27x 116-1 (Ref. 17x1 18).
Fig. 5. Leiosphaeridia raiia sp. nov., holotype. X 500. Slide 784-1, location 29 X 117-6 (Ref. 17-3 X 118).

Palaeontology, Vol. 8

PLATE 5

FELIX, Neogene Tasmanites

C. J. FELIX: TASMAN1TES AND LEIOSPHERES FROM LOUISIANA 21

Diagnosis. Oval, 80-120^ diameter (25 specimens measured) holotype 112/xxll3/x.
Wall usually distinct, forming pronounced marginal rim 5-8 p thick (7 p in holotype).
Wall with scattered pores 10-20 p apart, with slight border visible at high magnification.
Canals very rare and difficult to distinguish, slightly inclined and seldom more than
2 or 3 visible in one plane of focus. Body outline regular with only minor folding evident ;
wall plications not a diagnostic feature. Colour light to dark yellow in transmitted light.

Remarks. T. usitatus probably illustrates even better than does T. balteus the tenuous
boundary between Tasmanites and the leiospheres. The wall canals are easily over-
looked in a casual survey, and the pores are not easily distinguished. With its relatively
thin wall, failure to differentiate the canals could possibly result in its assignment to
Leiosphaeridia.

Comparison. In size and general appearance T. usitatus resembles T. medius Eisenack.
Eisenack (1963) does indicate a somewhat thicker wall for his species, as well as more
conspicuous pores and canals. In addition he cites the pylome as a diagnostic feature
and illustrates an undoubted circular aperture in the body wall (1962, pi. 4, figs. 7-8;
1963, fig. 6). T. usitatus is one of the more numerous species in the Neogene sediments,
and scores of specimens have been examined. However, no evidence of a pylome has
been noted, and virtually no variation in wall, pore, or canal features is displayed. At
this time there is insufficient evidence to warrant its assignment to T. medius.

Tasmanites fulgidus sp. nov.

Plate 7, figs. 1-4

Holotype. Slide 802-2, location 27-3 X 112-5 (Ref. 20x 118) (PI. 7, fig. 1). Core, 8,730 ft. Chacahoula
well No. 7, La Fourche Parish, Louisiana.

Diagnosis. Spherical, 180-275 p diameter (25 specimens measured) holotype 270 p.
Wall prominent, 10-18 p thick (15 p in holotype); penetrated by narrow, steeply
inclined canals \0-\6 p apart and some passing from outer to inner surface. Canals
present a bordered appearance where they penetrate the surface, which is laevigate
otherwise. Body outline regular, rarely folded; fig. 1 shows the maximum degree of
folding observed in the species. Colour yellow-orange to red-orange in transmitted light

Remarks. There is a possibility that additional research will warrant a further division.
Two general size groups appear represented, one about 180-200 p and a second 250-
275 p. The majority of specimens do fall within these ranges, but there is a sufficient
gradient between them to render it difficult to draw a line on size only. Occasionally
specimens appear to have canals differing from the majority. Plate 7, fig. 3 illustrates
a specimen with canals more numerous and less inclined than most and ending in a more
pronounced pore than is usually the case. There is relatively little other variation, and
there is not sufficient differentiation to necessitate additional specific division.

Comparison. The relatively large diameter of this species is a distinguishing characteristic,
along with the thick wall and numerous, evenly spaced, steeply inclined punctae. It does
resemble the classic T. huronensis in general appearance, although smaller in all dimen-
sions. In her emendation of T. huronensis , Winslow (1962) described the punctae as
being straight, radially aligned, and flared on the interior. The punctae of T. fulgidus

22 PALAEONTOLOGY, VOLUME 8

possess a steep angle of inclination and show no evidence of flaring. It compares slightly
with T. avelinoi Sommer, which also has steeply inclined canals, but the latter is signi-
ficantly larger in diameter, and the canals are fewer in number and without the pro-
nounced surface pores observed in T. fulgidus.

Tasmanites validus sp. nov.

Plate 8, figs. 1, 2

Holotype. Slide 802-1, location 21-3 x 111-6 (Ref. 161x117-5) (PI. 8, fig. 1). Core, 8,730 ft. Chacahoula
well No. 7, La Fourche Parish, Louisiana.

Diagnosis. Spherical, 1 50-21 5 /x diameter (15 specimens measured) holotype 166 /xX 168^.
Wall distinct, 7—1 5 /x thick (10-11 p in holotype); penetrated by short, narrow canals,
slightly inclined and passing through more than one optical plane. Canals scattered,
10-20 /x apart, appearing to pass from outer to inner wall surface, terminating on the
otherwise laevigate surface in small pores with only slight rim or border development.
Body outline irregular, somewhat undulate, but rarely folded. Colour yellow to orange
in transmitted light.

Comparison. The slightly undulate margin and sturdy wall with its distinct, but widely
spaced, pores distinguish this species. The only similar published example appears to
be a specimen of T. huronensis figured by Eisenack (1963, fig. 4), and the undulate
outline is the only comparative feature. T. validus is markedly different from T. huronen-
sis in size, wall thickness, and character of canals and pores.

Tasmanites balteus sp. nov.

Plate 8, fig. 3

Holotype. Slide 488-1, location 25 1 X 106-9 (Ref. 16-6x 117-2) (PI. 8, fig. 3). Core, 9,644 ft. Belle Isle
well No. 1, St. Mary Parish, Louisiana.

Diagnosis. Oval, 205-285 p diameter (10 specimens measured) holotype 210^x280^.
Wall prominent, 9-15 p thick (13-14 p in holotype); canals very rare and widely spaced,
appearing always to extend from the inner wall surface and not reaching the outer
surface; wall surface laevigate without visible pores. Specimens characteristically
resembling a collapsed sphere as to present two half spheres superimposed one upon
another. Colour red-orange with wall considerably lighter in colour in transmitted light.

Remarks. The rarity of canals in the species adds to the existing taxonomic difficulty
of the group. The canals are difficult to detect and may be easily overlooked. Since wall
canals are a feature of Tasmanites , but not of the leiospheres, failure to detect canals
would necessitate its inclusion in the leiospheres. Should such a form be described in
which the canals could not be distinguished, there would likely not be a depository for

EXPLANATION OF PLATE 6

Figs. 1-4. Tytthodiscus californiensis Norem 1955. 1, X 500. Slide 1556-1, location 31-8x124-1 (Ref.
14-2x118). 2, Entire specimen, x 500. Slide 1501-1, location 23-9x112-2 (Ref. 16x120). 3, Wall
detail. x750. 4, Detail of outer rim. X750.

Fig. 5. Tasmanites usitatus sp. nov., holotype. X 500. Slide 485-1, location 33-9x124-5 (Ref.
19x118-3).

Palaeontology, Vol. 8

PLATE 6

FELIX, Neogene Tytthodiscus

C. J. FELIX: TASMANITES AND LEIOSPHERES FROM LOUISIANA 23

it. Leiosphaeridia is described as possessing a pylome, which is not evident in T. balteus ,
but Downie and Sarjeant (1963) have included thin- walled leiospheres without pylomes
in Leiosphaeridia. However, the thick wall of T. balteus would appear sufficient to
exclude it from Leiosphaeridia. Thus T. balteus seems to be transitional in several fea-
tures between Tasmanites and the various leiosphere genera, and it provides some sup-
port to belief by various investigators that they constitute a single biological group.

Comparison. The collapsed sphere shape and the wide, relatively undifferentiated
marginal ring are diagnostic. In this respect T. battens appears to resemble T. euzebioi
(Sommer 1953) except for the size difference. The latter ranges from 370-520 /x or nearly
twice the dimensions of T. balteus.

Genus tytthodiscus Norem 1955
Tytthodiscus californiensis Norem 1955

Plate 6, figs. 1-4

Figured specimens (Plate 6, figs. 1-4) are from core at 8,040 ft. (Slide 1501-1) and from 8,970 ft.
< Slide 1556-1). Humble well No. 1, Terrebonne Parish, Louisiana.

Diagnosis. Spherical, 150-185 ft (25 specimens measured). Wall distinct, 5-14 /x thick
with average of about 12/x; no apparent thickness pattern, and walls are usually less
than 10 per cent, of diameter. Closely spaced oval canals 1-5-2 p. apart, not inclined;
about 1 p, in diameter, without taper and all completely penetrate through the wall.
Body sturdy with only minor folds viewed where considerable compression occurred.
Colour light yellow in transmitted light.

Remarks. The assignment to Tytthodiscus is based largely upon the feature of numerous
and closely spaced wall canals. It is the author’s belief that Tytthodiscus, Tasmanites,
and many of the leiospherids possess similar biological affinities. Eisenack (1958u)
suggested this by his inclusion of Tytthodiscus within the family Leiosphaeridae.
However, Tytthodiscus is a validly described entity, and admittedly its true biological
relationship is still conjectural. In addition to Norem’s (1955) original report of its
occurrence in marine Tertiary sediments of California, representatives of the genus
have been reported by Waloweek and Norem (1957) from Miocene age rocks of Alaska,
and from Tertiary sediments of Colombia by Sole de Porta (1961).

Comparison. Norem (1955) considered the hexagonal pattern of the wall segments to
be diagnostic for the genus. These hexagonal wall segments characterize the Louisiana
specimens and are quite unlike the scattered canal pattern of Tasmanites. The Neogene
specimens compare with Norem’s description of T. californiensis in general appearance,
size, wall thickness, and character of the hexagonal wall units. Tytthodiscus chondrotus,
the only other described species, is considerably smaller. It also possesses a granulate
surface with the granules arranged in a triangular pattern.

A micro-organism bearing some similarity to Tytthodiscus is Hungarodiscus, described
by Krivan-Hutter (1963) from the Palaeogene of the Dorog Coal Basin of Hungary.
The author distinguished it from Tasmanites and Tytthodiscus primarily on its extremely
thin wall, which is only 1/55 to 1/65 of the total diameter. She further differentiated it

24

PALAEONTOLOGY, VOLUME 8

from Tasmanites on its development of radially oriented tubules opening to outer and
inner surfaces and from Tytthodiscus by the lack of hexagonal wall segments; however,
in her species type description the author does describe the pores as having hexagonal
symmetry. Although Krivan-Hutter’s description of the wall pore symmetry is some-
what unclear, the illustrated pore arrangement is neither the hexagonal type of Tyttho-
discus calif or niensis nor the triangular of T. chondrotus. Actually Hnngarodiscus appears
to have different size pore openings, and there is a difference in pore pattern between
the pylome side and the opposite side of the body. Perhaps its most distinguishing feature
is the conspicuously large pylome, occupying about one-third of the entire diameter.
The very thin wall, the large pylome, and the variable pattern of the wall tubules suffice
to distinguish the genus from Tytthodiscus. Krivan-Hutter has placed it into the family
Leiosphaeridae, and this taxonomic assignment appears acceptable in view of features
similar to Tasmanites and Tytthodiscus.

LEIOSPH AERE GROUP

Genus leiosphaeridia (Eisenack) Downie and Sarjeant 1963
Leiosphaeridia plicata sp. nov.

Plate 8, fig. 4

Holotype. Slide 512-1, location 42T X 107-4 (Ref. 18x 118-6) (PI. 8, fig. 4). Core, 11,150 ft. Bell Isle
well No. 1, St. Mary Parish, Louisiana.

Diagnosis. Spherical, 120-200 /x diameter (50 specimens measured) holotype 145 ft x 154/x.
Wall distinct, thin, 3-7 p thick (5 p in holotype). Canals or pores not present on cell
wall and no evidence of pylome. Body outline irregular, always crumpled and plicated
with the numerous folds being characteristic of the species. Surface laevigate. Colour
light yellow in transmitted light.

Remarks. There is also a suggestion of an intergradational population such as occurs
in Tasmanites fulgidus. A single, excellently preserved specimen of 240 p in diameter,,
and indistinguishable from the species in other respects, was observed. However, this
was the single instance, in a study of hundreds of specimens, in which the 200^ figure
was exceeded, and it does not appear to warrant extension of the size range at present.

EXPLANATION OF PLATE 7

Figs. 1-4. Tasmanites fulgidus sp. nov. 1, holotype. X 300. Slide 802-2, location 27-3x112-5 (Ref.
20x118). 2, holotype. Wall detail, X500. 3, Specimen showing wall detail, x 500. Slide 1403-1,
location 34-6 x 1191 (Ref. 17 x 112-6). 4, Entire specimen, X 300. Slide 495-1, location 32-1 x 127-8
(Ref. 231 X 118).

EXPLANATION OF PLATE 8

Figs. 1, 2. Tasmanites validus sp. nov., holotype. 1, Entire specimen, X 500. 2, Wall detail, x750.

Slide 802-1, location 21-3x 111-6 (Ref. 16-1x117-5).

Fig. 3. Tasmanites balteus sp. nov., holotype. X 300. Slide 488-1, location 25-1x 106-9 (Ref.
16-6x117-2).

Fig. 4. Leiosphaeridia plicata sp. nov., holotype. X 500. Slide 512-1, location 42-1x 107-4 (Ref.
18x118-6).

Palaeontology, Vol. 8

PLATE 7

FELIX, Neogene Tasmanites

Palaeontology, Vol. 8

PLATE 8

FELIX, Neogene Tasmanites, Leiosphaeridia

C. J. FELIX: TASMANITES AND LEIOSPHERES FROM LOUISIANA

25

Comparison. The species bears some similarity to Leiosphaeridia voigti (Eisenack 19586)
in appearance, but the pylome is well defined in L. voigti and is definitely not present
in L. plicata. This species was the most numerous encountered in this study, and hun-
dreds of specimens were examined without any suggestion of a pylome. L. plicata com-
pares also with both Tasmanites mowed Sommer (1953) and T. simtosus Winslow (1962)
in general appearance, but both species of Tasmanites possess numerous punctae.
In size and the characteristic plications it resembles Tasmanites salustianoi , but Sommer
(1953, 1956) does note the presence of a few canals in the latter, although admittedly
difficult to distinguish. L. plicata definitely does not possess canals or pores.

Leiosphaeridia ralla sp. nov.

Plate 5, fig. 5

Holotype. Slide 784-1, location 29 x 117-6 (Ref. 1 7-3 x 1 18) (PI. 5, fig. 5). Core 6,930 ft. Chacahoula
well No. 7, La Fourche Parish, Louisiana.

Diagnosis. Spherical, 87-100 p diameter (15 specimens measured) holotype 99 p. Wall
thickness 1-3 p (3 p in holotype) but indistinct, without well-defined marginal rim.
Canals or pores not present in cell wall and no evidence of a pylome. Body outline
irregular, usually with one to three prominent folds. Surface laevigate. Colour light to
dark yellow in transmitted light.

Remarks. The species is easily differentiated from L. plicata both in size and morphology.
It is never crushed or plicated in the manner of L. plicata but usually with only a few
elongate folds. The very appearance of the specimens bears a suggestion of fragility.

Comparison. L. ralla shows the greatest similarity to Leiosphaeridia voigti. However,
its smaller size is far less than L. voigti, whose lower size range is 190 p. A further distinc-
tion is the definite pylome of L. voigti.

Location of types. The exact field position of specimens is noted in text and plate explanations as
coordinates, in parentheses, followed by a reference point coordinate for each slide. Calibration was
on a Leitz Ortholux microscope mechanical stage to tenths of millimeters, with horizontal (smaller)
reading listed first. Traverse (1958, 1960) and Pierce (1959) have dealt in detail with methods of co-
ordinate conversion used here. Type slides are filed in the Sun Oil Company Paleontological Collec-
tions, Richardson, Texas, U.S.A.

Acknowledgement. The author gratefully acknowledges the permission of the Sun Oil Company to
publish these findings.

REFERENCES

chaloner, w. g. 1953. On the megaspores of Sigi/laria. Ann. Mag. Nat. Hist. Ser. 12, 6, 881-96,
1 pi.

downie, c. and sarjeant, w. a. s. 1963. On the interpretation and status of some hystrichosphere
genera. Palaeontology, 6, 1, 83-96.

eisenack, a. 1938. Hystrichosphaerideen und verwandte formen im baltischen Silur. Z. Geschiebe-
forsch. u. Flachlandsgeol. 14, 1-30, 4 pi.

1958a. Tasmanites Newton 1875 und Leiosphaeridia n.g. als gattungen der Hystrichosphaeridia.

Palaeontographica, A 110, 1-19, 2 pi.

19586. Mikrofossilien aus dem Ordovizium des Baltikums. Senckenbergiana, 39, 389-405, 2 pi.

1962. Mitteilunger fiber Leiospharen und fiber das pylom bei Hystrichospharen. N. Jb. Geol.

Palaont. 114, 58-80, 3 pi.

26 PALAEONTOLOGY, VOLUME 8

eisenack:, A. 1963. Uber einige arten der gattung Tasmcmites Newton 1875. Grana Palynologica, 4,
203-16, 1 pi.

felix, c. j. 1963. Mechanical sample disaggregation in palynology. Micropaleontology , 9, 337-9, 1 pi.
jodry, r. l. and campau, d. e. 1961. Small pseudochitinous and resinous microfossils: new tools for
the subsurface geologist. Bull. Amer. Assoc. Geol. 45, 1378-91, 3 pi.

KRiVAN-HUTTER, e. 1963. Microplankton from the palaeogene of the Dorog Basin. I. Ann. Univ. Sci.
Budapest, Sect. Geol. 6, 71-91, 6 pi.

norem, w. l. 1955. Tytthodiscus , a new microfossil genus from the California Tertiary. J. Paleont. 29,
694-5, 1 pi.

pierce, r. l. 1959. Converting co-ordinates for microscope scales. Micropaleontology, 5, 377-8.
radforth, n. w. and rouse, g. e. 1956. Floral transgressions of major geological time zones. Trans.
Roy. Soc. Canada, 50, 17-26, 1 pi.

schopf, j. m. 1957. ‘Spores’ and problematic plants commonly regarded as marine. Geol. Soc. Amer.
Mem. 67, 709-18.

, wilson, l. r., and bentall, r. 1944. An annotated synopsis of Paleozoic fossil spores and the

definition of generic groups. Kept. Inv. III. State Geol. Surv. 91, pp. 72, 3 pi.
sole de porta, n. 1961. Contribucion al estudio palinologico del terciario en Colombia. Bol. Geol.
7, 55-81, 5 pi.

sommer, f. w. 1953. Os esporomorfos do folhelho de Barreirinha. Brazil Div. Geol. Min., Bol. 140,
pp. 49, 2 pi.

■ 1956. South American Paleozoic sporomorphae without haptotypic structures. Micropaleonto-

logy, 2, 175-81, 2 pi.

staplin, f. l. 1961. Reef-controlled distribution of Devonian microplankton in Alberta. Palaeontology,
4, 3, 392-424, 4 pi.

timofeev, b. v. 1959. The ancient flora of the Pre-Baltic and its stratigraphic significance. Mem.
VNIGRI, 129, pp. 350, 25 pi. [In Russian.]

traverse, a. 1958. Locating plant microfossils on mixed slides. Micropaleontology, 4, 207-8.

1960. Still more on conversion of microscope co-ordinates. Ibid. 6, 424.

wall, d. 1962. Evidence from recent plankton regarding the biological affinities of Tasmanites Newton
1875 and Leiosphaeridia Eisenack 1958. Geol. Mag. 99, 353-62, 1 pi.
waloweek, w. and norem, w. l. 1957. Geographic range of Tytthodiscus extended to Alaska. J.
Palaeont. 31, 674-6.

winslow, m. r. 1962. Plant spores and other microfossils from upper Devonian and lower Mississip-
pian rocks of Ohio. USGS Prof. Paper 364, pp. 93, 22 pi.

CHARLES J. FELIX

The Sun Oil Company
Richardson, Texas
U.S.A.

Manuscript received 3 January 1964

FORAMINIFERA IN HOLOCENE MARSH CYCLES
AT BORTH, CARDIGANSHIRE (WALES)

CONTRIBUTION — CARDIGAN BAY RESEARCH PROJECT

by t. d. adams and john haynes

Abstract. The exposed Holocene deposits at Borth include two well-marked marsh cycles. The dominant
foraminifera in these sediments are marsh and estuarine forms such as Jadammina macrescens , Protelphidium
depressulum , Elphidium orbiculare, Elphidium excavation , and Ammonia beccarii var. batavus. The distribution of
these species reflects the different stages of the cycles and thus the late post-glacial history of Borth Bog.

Borth Bog and the tidal flats and estuary of the Dovey valley are situated in the north-
west of Cardiganshire (text-fig. 1). They form a roughly triangular area of low-lying
marshy land and open sand flats extending from Borth in the south to Aberdovey in the
north, with Glandovey at the eastern apex of the triangle (O.S. 1" Map No. 127, grid
ref. 92/63). The surrounding hills are formed of lower Palaeozoic grits and mudstones
which have been sharply folded and faulted on a NNE. to SSW. strike. It is noteworthy
from the ecologic viewpoint that limestones are absent in the drainage area and that
mineralized veins with lead and zinc occur. The markedly straight northern shore of the
estuary probably coincides with the Llyfnant fault of Jones and Pugh (1935). During
the last Pleistocene glaciation the area was occupied by outflow ice from the Aran
Mawddwy volcanic range and afterwards, during the Holocene, the open valley became
infilled with estuarine and salt marsh sediments and peat.

The full stratigraphic details of this Holocene sequence are unknown, the unconsoli-
dated deposits being for the most part of indeterminate thickness. Nevertheless, two
records of borings given by Yapp, Johns and Jones (1916, 1917) place the solid Palaeo-
zoic basement at 80 to 90 feet below the surface at Glandovey and 1 50 feet below the
surface at Dovey Junction. In this early work may also be found details of the geology
and of the plant communities recognizable on the present marshes and raised bog.

On the foreshore, to the west of Ynyslas and Borth, a submerged fossil forest is seen
at low water, resting on blue silty clays, which are the oldest Holocene sediments
exposed. These Scrobicularia clays are laminated deposits, pale blue to grey in colour,
with occasional brown staining due to bacterial action, a feature which has been noted
in the present marshes by Chapman (1960). There is considerable included silt, as well
as sand at some horizons, and concentrations of plant remains occur at certain sites
immediately beneath the overlying peats. The clays contain a poorly developed mollus-
can fauna, dominated by Scrobicularia piperata Gmelin, as well as extremely rich fora-
miniferal assemblages. These deposits appear to be equivalent to the Downholland silts
(also named Formby and Leasowe Marine Beds) of Lancashire and Cheshire (Neaver-
son 1947).

The clays disappear eastwards beneath the fossil forest and the succeeding peats of
Borth Bog. Godwin (1943) ran a line of borings from north to south across the region,

[Palaeontology, Vol. 8, Part 1, 1965, pp. 27-38. 1

28

PALAEONTOLOGY, VOLUME 8

as well as a line extending from east to west linking the first section to the beach ex-
posures of the Ynyslas foreshore. They revealed that the fossil forest and the succeeding
raised bog rest upon a flat surface of Scrobicularia clay at minus 2 feet o.d. The sequence
of pollen has been studied by Godwin and Newton (1938) and Godwin (1943), and
evidence found for a late marine transgression bringing in a ‘clay intercalation’ from
the Dovey Estuary, extending southwards into Borth Bog below the surface peats. To
the west, the eroded seaward edge of the forest peats has now been over-ridden by
a prominent, north to south, pebble storm beach, which is succeeded at its northern
end by the blown sands of Ynyslas. To the north, the Scrobicularia clays and later peats
are truncated by, or disappear beneath, the fine sands and muddy silts of the estuary and
marine marsh of the river Dovey.

Absolute age determination and pollen analyses give what appears to be a reasonably
accurate date for the formation of these Holocene deposits at Borth. Radiocarbon
dating of the fossil forest peat layer, which immediately overlies the Scrobicularia clays,
places its formation at 4000 b.c. (Godwin and Willis 1961). The pollen content of the
immediately overlying peats shows that the forest layer falls within British Quaternary
florizone VIIa and thus accumulated during the early Atlantic climatic phase. Pollen
dating also shows that the late marine transgression occurred during the early sub-
Atlantic phase, florizone VIII, and is thus comparable to the Romano-British trans-
gression of the Fens (Godwin 1943), which began about 500 b.c. and apparently reached
a maximum before 200 a.d. This represents the last substantial rise in sea level around
the British Isles and initiated a cycle of marsh development in the Dovey area which
proceeded until freshwater fen was again established. This stage was apparently reached
some hundreds of years ago and since then the area has been relatively stable with marsh
growth balancing channel cutting and meander migration. The idea that the formation of
Borth Bog required some 1,500 years is supported by accretion studies (Richards 1934;
Chapman 1960). These suggest that about 550 years are required for the development of
mud flats into high marsh by vertical accretion. Even though sedimentation rates may
have differed in the past a long period (historically) is indicated for the vertical and
lateral accretion that accompanied the latest marsh cycle.

SAMPLE MATERIAL

During the severe storms of October and November 1960, the foreshore at Borth and
Ynyslas was stripped of its superficial cover of unconsolidated sands, to give a maximum
exposure of the fossil forest layer and the underlying Scrobicularia clays. It enabled six
one-inch diameter auger borings, varying in length from 6 to 12 feet, to be made along
a traverse extending from north to south across the outcrop (text-fig. 1). Excepting Bore
A, in which basement rock was reached at minus 10 feet o.d., the base of the clay was
nowhere penetrated. Similar borings were made through the deposits of the latest marsh
cycle, one adjacent to the river Leri, one adjacent to the river Clettwr, and three from
the mudflats of the Dovey Estuary.

Samples were collected at 1 foot intervals, and labelled in progressive order down
each bore. These samples were examined for foraminiferal content. The relative homo-
geneity of the sediments made selection of a standard weight preferable to that of stan-
dard volumes (Smith 1954). Consequently 50 grams of dried material from each sample

T. D. ADAMS AND J. HAYNES: HOLOCENE FORAMINIFERA FROM WALES 29

were weighed, disintegrated by boiling, and then dried once more. The sediment was
then passed through a nest of sieves and the total foraminiferal fauna picked from mesh
sizes 500 microns, 251 microns and 152 microns. Foraminifera remaining in the sedi-
ment on the 74 micron mesh sieve were retrieved by means of simple carbon tetra-
chloride concentration. The species present were identified and the total numbers of
specimens found within each 50-gram sample were plotted stratigraphically (text-figs.
2 and 3). Very large residues were split by the standard quartile technique of Twenhofel
and Tyler (1941).

text-fig. 1. Map of Dovey Estuary and Borth Bog, with geology after Yapp, Johns and Jones (1916,
1917), showing location of boreholes. Land above the 50-foot contour and the open sand flats of the

estuary left blank.

STRATIGRAPHICAL SEQUENCE OF FORAMINIFERA

Only the uppermost part of the Holocene sequence (from florizone Vic onwards) is
exposed. These deposits have been studied in detail. The populations of foraminifera
include:

iii. The present-day association in the Dovey Estuary. (The study of these forms is still
continuing and the results will be published separately.)
ii. The foraminifera in the deposits of the latest marsh cycle,
i. The foraminifera in the uppermost Scrobicularia clays.

These associations are almost identical, with the same species dominant in each.
There are, however, some minor differences.

30

PALAEONTOLOGY, VOLUME 8

Species found only in the Scrobicularia clays include:

Asterigerina sp. Elphidium goesi

Biorbulina bilobata Patellina corrugata

Bolivina malovensis Triloculina brongniartii

Cassidalinoides tenuis

Species found only in the deposits of the latest marsh cycle include:

Acervulina inhaerens
Alveolophragmium jeffreysi
Bulimina mexicana
Dentalina neugeboreni
Guttulina lac tea
Hyalinea baltbica
Lenticulina suborbicularis
Oolina borealis

Protoschista findens
Pyrgo williamsoni
Spirolocu/ina subimpressa
Spirophthalmidium acutimargo
Textularia conica
Triloculina trihedra
Verneuilina media

Apart from this slight qualitative difference, which may be due to collection failure
since all these species are rare, the total number of specimens is also higher in many
of the samples from the deposits of the latest marsh cycle. This is related to distance from
the present estuary.

Within these Holocene deposits as a whole, 25 species of foraminifera occur in
considerable numbers. The dominant species, which form a very abundant population
element at most horizons (often more than 200 specimens per sample and more than
500 in some), are:

Ammonia beccarii var. batavus Elphidium orbiculare

Elphidium excavatum Protelphidium depressulum

Occurring more rarely, but exclusively dominant at certain horizons, are Jadammina
macrescens and Trochammina inflata.

Species which form a subsidiary population element at most horizons (often more than
20 specimens per sample and more than 50 in some) are:

Angulogerina angulosa var.
Buccella frigida
Bulimina aculeata var.

B. affinis
B. gibba

Cibicides lobatulus
Discorbis bradvi

D. williamsoni
Elphidium bartletti

E. crispum

E. discoidale
E. incertum
E. macellum
E. margaritaceum
Heminwayina mamilla
Oolina williamsoni
Protelphidium barleeanum
P. pompilioides
Quincpieloculina seminulum

(A full species list is given at the end of the paper)

ECOLOGICAL SIGNIFICANCE OF THE FAUNA

Certain species are critical in the recognition of brackish water environments:

Trochammina inflata ; see Brady 1870, Hedberg 1934, Phleger and Walton 1950,
Van Voorthuysen 1951.

Jadammina maerescens; see Bartenstein and Brand 1938, Brady 1870, Van Voor-
thuysen 1951.

T. D. ADAMS AND J. HAYNES: HOLOCENE FORAMINIFERA FROM WALES 31

Miliammina fusca; see Hedberg 1934, Phleger and Walton 1950.

Protelphidium depressulum ; see Van Voorthuysen 1947, 1951.

Ammonia beccarii; see Bandy 1953, Hedberg 1934, Le Calvez and Le Calvez
1951, Van Voorthuysen 1947, 1951.

When either one or more of these species are dominant in a fossil assemblage, brackish
waters are considered to have existed. These are the species which live in the present
estuary and which also dominate the sediments at Borth, and reflect, by their distribu-
tion, the gradually changing environmental conditions of the marsh cycles.

DETAILED ANALYSIS OF FAUNA FROM BORES

1. The Foraminifera of the uppermost Scrobicularia clays. The Scrobicularia clays were
penetrated in the north-south line of borings A, B, C, D, E, F shown on the map (text-
fig. 1). The first four were started in the lowest peats of the forest layer and, in the case
of Bore B, penetrated 10 feet into the underlying clays. These bores thus provide an
excellent opportunity for study of the included microfauna in relation to the palaeo-
geographical changes which eventually culminated in the growth of a forest cover with
pines and birches. At the locations of Bores E and F the overlying forest layer has been
eroded off. These bores therefore start lower in the sequence than the others and pene-
trate further into the clays (see text-fig. 2).

In Bore A, to the south, the fauna in the clays below the peat is dominated throughout
by Trochammina inflat a and Jadammina macrescens. In the present Dovey estuary these
species characterize the high marsh zone, Juncetum, of Yapp, Johns and Jones (1917).
Examination of the charts shows that the dominance of T. inflata becomes more marked
upwards prior to the disappearance of the entire fauna in the peats. This certainly
reflects the later stages of growth of a marine marsh and its passage into freshwater
swamp.

The fauna in Bore B is more abundant and the marsh cycle is more completely shown.
In the lowest clays penetrated the fauna is close to that of the present day mud flats and
low marsh (Salicornietum) of the Dovey with Elphidium excavation, Elphidium orbi-
culare, Protelphidium depressulum and Ammonia beccarii var. batavus as dominant
forms. The fauna diminishes in number of individuals and species towards the top of the
bore and six species only, three of them arenaceous, were retrieved from the clays about
2 feet below the peats. These include the high marsh species Jadammina macrescens and
Trochammina infiata.

The charts for Bore C again illustrate the marsh cycle, with gradual change and
diminution of the fauna accompanying the passage into the peats from a maximum at
the 9-foot interval. Below this interval the fauna is again sparse. As in Bore B the
uppermost Scrobicularia clays are unfossiliferous, with intercalated peat, wood frag-
ments, and rootlets. Iron staining in these deposits resembles that seen in the present
high marsh of the Dovey.

Bore D shows similar faunal trends, but here the lowest clays penetrated yield an
abundant fauna dominated by A. beccarii var. batavus and including thirty-four other
species, which suggests strong estuarine to marine influences. Again the fauna tends to
diminish towards the top of the bore, the fauna at 1 foot depth being dominated by

32

PALAEONTOLOGY, VOLUME 8

FORAMINIFERA

BORE A

2toM5fc|71i

Jad. macrescens

Iroch. inf lata

Mil- fusca

Prol. depressulum
Prot. barleeanum

Elph excovatum

Bucc. frigida

Orb universe

Epon. concameratus
Elph. margaritaceum
Quin, suborenarea

Bui. aff inis

Prot. pompitioides

Elph. orbiculore

Amm. beccari f var.

Quin seminulum

Quin patagon ica

Lent, rotulata

Ool. melo

Bol. malovensis

Cass, islondica

Cib. tobatulus

Elph. discoidale

Fiss. lucido

Bol. voriabilis

Cib. f letcheri

Glob bulloides

Elph. bartletti

Lag. laevis

Ool. will iamsoni

Bui, aculeata

Bui. marginota

Quin, lata

Elph. crispum

Mil. subrotunda

Lag. sulcata int.

Bui gibba

Oise. williamsani

Elph. incertum

Fiss. orbignyana

Elph. crispum spin.

Elph. macellum

Tril. tricarinata

Quin, cliarensis

Lag. clavota

Lag. sulcata

Bui. affinis

Bol pseudoplicata

Bol, spothulata

Ang anguloso

Ang. angulosa cor

Pat corrugata

Plan mediterran

Disc, bradyi

Bol simplex

Cib. refulgens

Log. semistriata

Ool lineata

Fiss, margmata

Glob, inflota

Log loevis

Astr. gallowayi

Tril. brongniartii

Tril trigonula

Oum, agglutinata

Mil. cnuckchiensis

Lag. substriata

Log sulcata spir.

Ool hexagona

Cass, tenuis

Dyo biserialis

Hem mamilla

Elph. macellum acul.

Reus, oligocenica

Bol. compacta

Bior. bilobata

Non atlantica

Elph. goesi

Quin, depressa

Bui e leqant issima

Glob gibba

Ang. angulosa Hu.

Ool. globosa

O

1 O

1 l

• l
©
1 •

BORE E

1 1 2 1 3 K I sTeTT

9999999

-OO-O-

-00—09

®9®

OOO

OOO

,0909909

oo

0-0

o

— oo-

-o

- o--

0-0

BORE

F

uMisM

l6bl7*b

-

- ®9 ®

9-0

-9-

—

-990

—

9-

9

- O

9»® *

• 99

91* •

• 09

—

—

• O

- © 9 9

90 9

-990

O- O

-O

9-0

— 9 — —
O-

O-

- -

0909

9-0

-9-0

o--

O — —

_ .

900

—

_

—

- _

— —

—

OO

O — —

o-

— 9 — —

0 -

-O - -

-

_ _

— — —

-O o

o -

-

”

o

•

900

9- -

*

text-fig. 2. Distribution of foraminifera in the Scrobicularia clays.

FREQUENCY SYMBOL . - 1 -- 2 - 5 - - 6-20 0 - 21 - 50 0 - 51 -200 ® = 201 -500 n ->500

T. D. ADAMS AND J. HAYNES: HOLOCENE FORAMINIFERA FROM WALES 33

Protelphidium depressulum and Elphidium excavation, together with Trochammina
infiata, typical marsh species, although a number of other species occur rarely.

The top of Bore E is some 2 or 3 feet below the forest layer, which at this location has
been eroded away, and older sediments are penetrated at the bottom than in the other
borings. Nevertheless, clays with a marsh fauna are preserved at the top and the same
general ecologic evolution can be seen. The fauna in the lowermost samples, 5-7 feet
depth, is very abundant, dominated by Buccella frigida and Ammonia beccarii, with up
to fifty-five other species, indicating strong estuarine to marine influences. This fauna
diminishes rapidly upwards in the boring, above a Scrobicularia shell bank at 4\ feet,
and the uppermost samples show the typical marsh fauna seen in the other cores.

Bore F is similar to Bore E but with more of the uppermost Scrobicularia clays
eroded off, so that less of the marsh phase is seen in the charts. The fauna is less abun-
dant in species but more abundant in numbers of specimens than Bore E but shows a
similar diminution at the top, illustrating the progress of the marsh cycle towards fresh-
water conditions.

Detailed study of the bores therefore supports the idea that the Scrobicularia clays
of the Ynyslas foreshore are the deposits of a cycle of marsh deposition, the faunas
indicating a passage upwards from estuarine deposition through an open intertidal
flat and salt marsh stage to freshwater conditions. The faunal associations are essentially
the same as those found in the present Dovey estuary. The sediments are more or less
contemporaneous in age, although the most northern bores (Eand F), because of removal
by erosion of the highest clays, reached older beds.

Besides the marked vertical change seen in the Scrobicularia clays there is also a lateral
change in the faunas from south to north. Bore A appears to represent the former limit
of clay deposition at the edge of the existing marshes, its fauna being typical throughout
of the present Dovey high marsh. Northwards through bores B, C, and D, the increasing
estuarine and tidal influence of the Dovey is reflected in the gradual increase in number
of foraminifera present, the faunas becoming more and more typical of the present day
open tidal flat zone. Bores E and D, in close proximity to the main Dovey channel, show
the best developed estuarine faunas. Thus palaeogeographic evidence suggests that the
river Dovey has maintained its present location in post-glacial time from at least late
Scrobicularia clay times (end of Boreal florizone VI).

2. The Foraminifera in the deposits of the latest marsh cycle. The deposits of the latest
marsh cycle were penetrated in borings G, H, I, J, and K (text-fig. 1). Three of these
borings (I, J, and K) were made on the present Dovey marshes; Bore H was made near
the river Clettwr; and Bore G near the river Leri at its tidal limit.

The faunal changes which occur through these bores resemble those seen in the bores
from the foreshore (text-fig. 3). As is well seen in Bores K and H, the total number of
specimens and species reaches a maximum approximately half-way up each boring,
then gradually falls. Although this is not so marked in Bores I and J the faunal changes
in all four indicate a full physiographic cycle with a change from a marsh or open flat
environment to estuarine conditions, then a return to high marsh. This is well shown by
Bore K situated at the head of the estuary. The lowest samples yield a marsh fauna with
J. macrescens, together with open tidal flat species such as E. excavation and E. orbicular e.
Such a fauna might be expected at the beginning of the transgression which undoubtedly

B 6612

D

34

PALAEONTOLOGY, VOLUME 8

FORAMINIFERA

BORE G

1 1 2 1 3

Cent, aculeata

Jad macrescens

Troch. inf lata
Mil. fusca

Prot. depressutum

Elph. excavatum

Elph- discoidale

Elph. incertum

Elph. macellum

Quin, seminulum

Bucc. triqida

Ool. wiltiamsoni

Bui. aculeata

Bui. at f inis

But gibba

Bol. psuedopticata

Bol. spathulata

Ang. angulosa

Cib. lobatulus

Disc. bradyi

Disc, wiltiamsoni

Elph. margaritaceum
Elph, orbiculare

Amm. beccarii var.

Tril. trihedra

Tril. tricarinata

Spiro, subimpressa

Quin, lata

Mil. subrotunda

Lag. laevis

Lag. perlucida

Lag semistriata

Lag sulcata

Ool. lineata

Ool. melo

Fiss. lucida

Bui. marg inata

Plan, mediter ran

Glob, inflata

Hem. mamilla

Cib. refulgens

Epon, concameratus
Elph. crispum

Tril. trigonula

Quin, cliarensis

Quin, depressa

Quin, surarenaria

Lag. clavata

Ool hexagona

Fiss. orbignyana

Marg. calva

Cass, islandica

Lag. d itf lugiformis

Alveo. jeff reysi

Text, conica

Vern. media

Lent, rotulata

Bui. etegantissima

Bol. compacta

Prot. barleeanum

Prot. pompilioides

Quin, agglutinata

Fiss. marginata

Proto, findens

Cib fletcheri

Mil. chukchiensis

Spiro, acutimargo

Glob, gibbo

Bol. variabilis

Astr. gallowayi

Elph. Bartlett i

Quin, patagonica

Ool. borealis

Bol. simplex

Hyal. balthica '

— 0

•

BORE K
1 1 2 1 3 1 4 1 5 1 6

Feet

O-

-9 00

0--

-O-

-oo-

text-fig. 3. Distribution of foraminifera in the deposits of the latest marsh cycle.

FREQUENCY SYMBOL •=! --2-5 _ - 6-20 0 = 20- 50 ® = 50 - 200 ©= 200-500 E =>500

T. D. ADAMS AND J. HAYNES: HOLOCENE FORAMINIFERA FROM WALES 35

caused both erosion and migration of the marsh zones landwards, probably with fairly
rapid conversion of much of the former freshwater bog area into open tidal flat. The
fauna at 4 feet shows the peak of the transgression with maximum estuarine to marine
influence. The dominant species in this association is E. orbiculare , a dominant form in
the present-day estuary. The succeeding samples show a marked diminution in the num-
ber of species, the fauna at the top of the boring representing a typical high marsh
assemblage with J. macrescens, Miliammina fusca, and Protoschista findens , the same
fauna as in the present high marsh. The lateral succession of present-day marsh zones
is thus mirrored by the vertical succession in the upper part of the borings; the sharply
defined estuarine, low marsh, and high marsh associations indicating steady accretion
presumably accompanying a relatively stable sea level.

Bore J was located in the lower high marsh (lowest salt marsh sward, Glycerietum)
near the mouth of the Clettwr, so that the full cycle is not seen, but as in Bore K a sparse
marsh fauna with open flat elements is found in the lowest samples penetrated. The
fauna towards the middle of the boring, dominated by E. orbiculare , indicates maximum
estuarine influence, and that in the upper part of the boring shows the gradual passage
to low marsh. The peak of the transgression is indicated by samples about 5 or 6 feet
below the surface, which is near the top of the low marsh. This compares with a maxi-
mum of 2 feet total thickness for the same interval in Bore K. This strikingly confirms
the observations made by Richards (1934), which suggested that the rate of accretion
in the lower estuary was about twice that at the river end.

The charts for Bore I show very similar faunal trends except that fewer species were
recovered. The marsh cycle is also very well shown in Bore H located well within the
present freshwater area, near the east bank of the Clettwr. Here the period of maximum
transgression is indicated by the total faunal maximum in samples between 2\ and 54
feet. This boring begins in low marsh sediments and the full cycle to high marsh is not
seen. Marsh sediments only were found in Bore G. Here the appearance of Centropyxis
indicates the onset of freshwater conditions.

CONCLUSIONS

Fauna and climate. The oldest Holocene deposits studied at Borth are the Scrobicularia
clays. These were laid down in late Boreal times, over 6,000 years ago, while the deposits
above the clays represent almost continuous deposition down to the present day. The
sediments thus span a period of great climatic vicissitudes from the late Boreal, through
Atlantic and sub-Boreal, into the present sub- Atlantic phase. Although the fossil faunas
differ little from those living today there is some evidence of change with climate in
the faunal sequence. Buccel/a frigida, generally considered a cold-water form, is one of
the most abundant species in the Scrobicularia clays but is relatively less significant in the
succeeding assemblages. This trend is paralleled by a simultaneous increase in Bulimina
gibba, both possibly related to the change from Boreal (Continental) to Atlantic
(Oceanic) climatic conditions.

Faunal affinities. The distinctive foraminiferal associations in the Holocene beds at
Borth may be compared with those distinguished in other parts of the British Isles,
particularly by Macfadyen (1938, 1942, 1955). The two main areas are (i) Swansea Bay

36

PALAEONTOLOGY, VOLUME 8

and the Somerset Levels, (ii) the Norfolk Broads and the Fens. Fifteen species of forami-
nifera are common to all areas, seventeen are restricted to the Fens and Broads, twenty-
two to the Swansea area, fifty-six to the Borth beds. It is possible that the high figures
for Borth merely reflect the more detailed work carried out; nevertheless, these figures
provide some evidence of distinct sub-provinces in the British area.

Summary of palaeogeographic events. The lowest sediments studied, the Scrobicularia
clays, are the final deposits of the main, post-glacial (Flandrian) transgression. The
foraminifera in these deposits support the conception of a succession from estuarine
conditions through salt marsh to fen carr and finally to pine and birch forest. This forest
climax was reached by about 4000 b.c. These events were followed by a change in climate
and during the wetter Atlantic phase the forest was overwhelmed by freshwater peats.
At approximately 500 b.c. renewed marine transgression took place, reaching a peak
during Romano-British times. Another cycle of marsh growth was thus initiated, to be
completed some hundreds of years ago. The stages of this physiographic cycle can be
interpreted from the foraminiferal faunas.

Acknowledgement. Acknowledgement is made of generous financial aid given to one author (T. D.
Adams) by the Department of Scientific and Industrial Research.

SPECIES LIST

Full names of species plotted in text-fig. 2 :

Jadammina macrescens (Brady)
Trochammina inflat a (Montagu)
Miliammina fusca (Brady)
Protelphidium depressulum (Walker
and Jacob)

Protelphidium barleeanum (Wil-
liamson)

Elphidium excavation (Terquem)
Buccella frigida (Cushman)
Orbulina universa d’Orbigny
Eponides concameratus Montagu
Elphidium margaritaceum (Cush-
man)

Quinqueloculina subarenaria Cush-
man

Bulimina afflnis d’Orbigny
Protelphidium pompilioides (Fichtel
and Moll)

Elphidium orbiculare (Brady)
Ammonia beccarii (Linne) var.
batavus Hofker

Quinqueloculina seminulum (Linne)
Quinqueloculina patagonica d’Or-
bigny

Lsnticulina rotulata (Lamarck)
Oolina melo d’Orbigny
Bolivina malovensis Heron-Alien
and Earland

Cassiduliita islandica Norvang
Cibicides lobatulus (Walker and
Jacob)

Elphidium discoidale (d’Orbigny)

Fissurina lucida (Williamson)
Bolivina variabilis (Williamson)
Cibicides fletcheri Galloway and
Wissler

Globigerina bulloides d’Orbigny
Elphidium bartletti Cushman
Lagetta laevis (Montagu)

Oolina williamsoni (Alcock)
Bulimina aculeata (d’Orbigny)
Bulimina marginata d’Orbigny
Quinqueloculina lata Terquem
Elphidium crispum (Linne)
Miliolinella subrotunda (Montagu)
Lagena sulcata (Walker and Jacob)
var. interrupta (Williamson)
Bulimina gibba d’Orbigny
Discorbis williamsoni Cushman and
Parr

Elphidium incertum (Williamson)
Fissurina orbignyana (Seguenza)
Elphidium crispum (Linne), spinose
var.

Elphidium macellum (Fichtel and
Moll)

Triloculina tricarinata d’Orbigny
Quinqueloculina cliarensis (Heron-
Alien and Earland)

Lagena clavata (d’Orbigny)

Lagena sulcata (Walker and Jacob)
Bulimina afflnis (d’Orbigny)
Bolivina pseudoplicata Heron-Alien
and Earland

Bolivina spathulata (Williamson)
Angulogerina angulosa (William-
son)

Angulogerina angulosa (William-
son) var. carinata (Cushman)
Patellina corrugata Williamson
Planorbulina mediterreana d’Or-
bigny

Discorbis bradyi Cushman
Bolivina simplex Phleger and
Parker

Cibicides refulgens (Montfort)
Lagena semistriata Williamson
Oolina lineata (Williamson)
Fissurina marginata (Montagu)
Globigerina inflata d’Orbigny
Lagena laevis (Montagu)
Astrononion gallowayi Loeblich
and Tappan

Triloculina brongniartii d’Or-
bigny

Triloculina trigonula (Lamarck)
Quinqueloculina agglutinata Cush-
man

Miliolinella chukchiensis Loeblich
and Tappan

Lagena substriata Williamson
Lagena sulcata (Walker and Jacob)
var. spirata Bandy
Oolina hexagona (Williamson)
Cassidulina tenuis Phleger and
Parker

T. D. ADAMS AND J. HAYNES: HOLOCENE FORAMINIFERA FROM WALES 37

Dvocibicides biserialis Cushman
and Valentine

Heminwayina mamilla (Williamson)
Elphidium macellum (Fichtel and
Moll) var. ac idea turn Silvestri
Reussella cf. R. oligocoenica Cush-
man and Todd

Rare species not plotted in Fig. 3 :

Acervulina inhaerens Schultz .
Dentalina neugeboreni (Schwager) .
Bulimina mexicana Cushman .
Lenticulina suborbicularis Parr

Bolivina compacta (Sidebottom)
Biorbulina bilobata (d'Orbigny)
Nonionella atlantica Cushman
Elphidium goesi Stschedrina
Quinqueloculina depressa d’Or-
bigny

Buliminella elegantissima (d’Or-
bigny)

Globulina gibba (d’Orbigny)
Angulogerina angulosa (William-
son) fluens Todd
Oolina globosa (Montagu)

Textularia conica d'Orbigny
Triloculina irihedra Loeblich and
Tappan

Verneuilina media Hoglund
Thecamoebinae

Centropyxis aculeata (Ehrenberg)

Additional species plotted in text-fig. 3:

Alveophragmium jeffreysi (William-
son)

Hyalinea balthica (Schroeter)
Lagena perlucida (Montagu)
Lagenammina difflugifonnis
(Brady)

Oolina borealis Loeblich and Tap-
pan

Protoschista Andens (Parker)
Spiroloculina subimpressa Parr
Spirophthalmidium acutimargo
(Brady)

Bore J at 4'
Bore H at 1J'
Bore J at 3'
Bore J at 4'

Guttulina lactea (Walker & Jacob) . Bore J at 5'
Oolina globosa (Montagu) . . Bore J at 3'

Pvt go williamsoni (Silvestri) . . Bore J at 5'

REFERENCES

bandy, o. l. 1953. Ecology and paleoecology of some Californian Foraminifera. /. Paleont. 27,
161-82, pi. 21-25.

bartenstein, h. and brand, e. 1938. Die foraminiferen Fauna des Jade-Gebietes. Senckenbergiana ,
20, 381-85.

brady, h. b. 1870. The ostracoda and foraminifera of tidal rivers. Ann. Mag. nat. Hist., Ser. 4, 273—
306, pi. 11, 12.

chapman, v. J. 1960. Salt marshes and Salt Deserts of the World. London.

Godwin, h. 1943. Coastal peat beds of the British Isles and North Sea. Journ. Ecol. 31, 199-247.

and newton, l. 1938. The submerged forest at Borth and Ynyslas, Cardiganshire. New Phytol.

37, 333-44.

and willis, e. h. 1961. Natural radiocarbon measurements, III. Radiocarbon, 3, 60-76.

hedberg, h. d. 1934. Some recent and fossil brackish to freshwater foraminifera. J. Paleont. 8, 469-76.
jones, o. t. and pugh, w. j. 1935. The geology of the districts around Machynlleth and Aberystwyth.
Proc. Geol. Ass. Lond. 46, 247-300.

le calvez, j. and le calvez, i. 1951. Contribution a l’etude des foraminiferes des eaux saumatres.
Vie et Milieu. 2, 237.

macfadyen, w. A. 1938. Post-Glacial foraminifera from the English Fenlands. Geol. Mag. 75, 409-17.

1942. A Post-Glacial microfauna from Swansea docks. Geol. Mag. 79, 133-46.

1955. ‘Foraminifera’ in Godwin, h. Studies on the post-Glacial history of British vegetation,

XIII. The Meare Pool region of the Somerset Levels. Phil. Trans., B, 239, No. 662, 161-90.
neaverson, e. 1947. Coastal changes around Liverpool Bay since the Ice Age. Proc. Lpool Geol. Soc.
19, 3-31.

phleger, f. b. and walton, w. r. 1950. Ecology of marsh and bay Foraminifera, Barnstaple, Massa-
chusetts. Amer. J. Sci. 248, 274-94.

Richards, f. j. 1934. The salt marshes of the Dovey Estuary, IV. The rates of vertical accretion, hori-
zontal extension and scarp erosion. Ann. Bot. 48, 225-59.
smith, m. l. 1954. A method of selecting sample sizes. J. Paleont. 28, 116-17.
twenhofel, w. h. and tyler, s. a. 1941. Methods of study of sediments. New York.
voorthuysen, j. h. van. 1947. Holocene foraminifera from borings in tidal marshes. The Micro-
paleontologist, 1 (2), 5-6.

38 PALAEONTOLOGY, VOLUME 8

voorthuysen, j. h. van. 1951. The quantitative distribution of Elolocene Foraminifera in the
N. O. Polder. Proc. 3rd Lit. Cong. Sediment. Neth.. 267-72.
yapp, r. h., johns, d., and jones, o. t. 1916. The salt marshes of the Dovey Estuary, Pt. 1. /. Ecol.
4, 27-42.

1917. The salt marshes of the Dovey Estuary. Pt. 2. /. Ecol. 5, 65-103.

T. D. ADAMS JOHN HAYNES

Department of Geology,
University College of Wales,

Manuscript received 23 December 1963 Aberystwyth

ON A NEW SPECIES OF HOEGISPORIS COOKSON

by ISABEL C. COOKSON

Abstract. A new species of the microspore genus Hoegisporis is described from Australian Cretaceous deposits
and its stratigraphical significance considered.

When the genus Hoegisporis was first established (Cookson 1961) a clear indication
was given that, in all probability, two distinct forms were being included in the type
species H. lenticulifera Cookson. Since that time more examples of both forms have been
found with the result that it is now possible confidently to define a second species of
Hoegisporis from Australian Cretaceous deposits. The erection of this new species, to
be described below under the name Hoegisporis uniforma, necessitates the removal to
it of the specimen figured earlier as a paratype of H. lenticulifera (Cookson 1961, pi. 76,
fig. 4).

Genus hoegisporis Cookson 1961
Hoegisporis uniforma sp. nov.

Plate 9, figs. 1-5, 7-9

Hoegisporis lenticulifera Cookson 1961, pi. 7, fig. 10, Nat. Mus. Victoria P20511.

Holotype. Plate 9, fig. 2, Univ. W.A. Geol. Dept. no. 51413.

Age and Occurrence. Western Australia (1) Perth Basin, (a) Gingin area. Probably Albian; Moora
Bore between 86 and 170 ft., Regan’s Ford on the Moore River, ‘Wapets’ seismic shot holes L 8 at
240 ft., L 9 at 306 ft., and M2 at 55 ft. ( b ) Perth area. Probably Aptian; Rakich's Bore, Caversham at
350-55 ft., Attadale Bore at 619 ft. Probably Albian; West Guildford Bore at 300 ft., Rakich’s Bore,
Caversham between 120 and 150 ft., Power House Bore at 640 ft., Kenwick School Bore at 90 ft., Mt.
Lyell Bore at 886 ft. (c) Fremantle area. Probably Albian, Jandakot Bore at 450 ft. (2) Carnarvon Basin.
Probably Albian, Lower Gearle Siltstone, ‘Wapets’ Rough Range no. 1 Bore at 2,750 ft.

South Australia. Probably Aptian, ‘Santos’ Ltd’. Oodnadatta Bore between 1,032 and 1,052 ft.;
probably Albian, Oodnadatta Bore at 327, 347, 387 ft.

Description. Microspore flattened, variously crumpled, outline of favourably preserved
specimens approximately circular in polar view and frequently somewhat wavy. Exine
about 1-5 p thick, intectate, finely, densely, and uniformly pilate; pila about 0-9-1 -2 p
long; equatorial thickenings 6-12 in number, their outlines circular in surface view.

Dimensions. Holotype — diameter 15 p. Range — diameter 39-80 p, equatorial exinous
thickening c. 8-18 p in surface view.

Comment. H. uniforma can be readily distinguished from H. lenticulifera by the constant
and complete absence of the clavae which are invariably present, although in varying
numbers, in H. lenticulifera (PI. 9, fig. 6). In three examples of H. uniforma a large open-
ing of irregular outline is present on one of the surfaces (PI. 9, fig. 7). Whether this
represents a natural opening or an artificial break is not known. Both species are of
infrequent occurrence.

(Palaeontology, Vol. 8, Part 1, 1965, pp. 39-40, pi. 9.]

40

PALAEONTOLOGY, VOLUME 8

The stratigraphical distribution of the two species of Hoegisporis is of interest since
there is evidence that the vertical distribution of both is distinctive and restricted. This
is demonstrated by the occurrence of these species in (a) the Osborne Formation in the
Perth and Fremantle areas and (b) equivalent Cretaceous beds in the Gingin area of
the Perth Basin, Western Australia (McWhae et al. 1958, p. 116). In each area H. uni-
forma has been recovered only from the lower beds, probably of Aptian-Albian age, and
H. lenticulifera from the upper portion of each sequence, the age of which approximates
to Upper Albian-Cenomanian.

In addition both species are associated with certain types of Dinophyceae in all the
deposits studied. H. lenticulifera is associated with one or more species of Ascodinium
Cookson and Eisenack 1960, a genus which, on the present knowledge, appears to be
restricted to Mid-Cretaceous deposits (Cookson and Hughes 1964). H. uniforma, in
contrast, is associated with such Australian Lower Cretaceous types as Hystricho-
sphaeridium arundum Eisenack and Cookson 1960 , Apteodinium maculatum Eisenack and
Cookson 1960, Pseudo ceratium turneri Cookson and Eisenack 1958 and occasionally
(Oodnadatta Bore, S. A., between 1,032 and 1,052 ft.) Dingodinium cerviculum Cookson
and Eisenack 1958.

REFERENCES

cookson, i. c. 1961. Hoegisporis, a new Australian Cretaceous Form Genus. Palaeontology, 31, 485-6.

and eisenack, a. 1958. Microplankton from Australian and New Guinea Upper Mesozoic

Sediments. Proc. Roy. Soc. Vic. 70, 19-78

— 1960. Microplankton from Australian Cretaceous sediments. Micropaleontology, 6, 1-18.

and hughes, n. f. 1964. Microplankton from the Cambridge Greensand (Mid-Cretaceous).

Palaeontology, 7, 37-59.

eisenack, a. and cookson, i. c. 1959. Microplankton from Australian Cretaceous deposits. Proc.
Roy. Soc. Vic. 72, 1-11.

mcwhae, j. r. h., playford, p. e., lindner, A. w., glenister, b. f., and balme, b. e. 1958. The Strati-
graphy of Western Australia. Geol. Soc. Australia J. 4, 1-16.

ISABEL C. COOKSON
Department of Botany,
University of Melbourne,

Manuscript received 21 January 1964 Australia

explanation of plate 9

The numbered specimens are housed in the Department of Geology, University of Western Australia.

Figs. 1-5, 7-9. Hoegisporis uniforma sp. nov. 1, Power Elouse Bore, W.A. at 640 ft., X c. 1000; no.
51412. 2, Holotype, Oodnadatta Bore, S.A. at 347 ft., x c. 690; no. 51413. 3, 4, 9, Jandakot Bore,
W.A. at 450 ft., X c. 1000; nos. 51414-6. 5, 7, Moora Bore, W.A. between 86 and 100 ft. 5, details
of exine, X c. 1350. 7, showing ‘hole’ on surface, X c. 730. 8, Regan's Ford, W.A. Seismic shot hole
L 8 at 240 ft., x c. 1000.

Fig. 6. Hoegisporis lenticulifera Cookson. Roberts Street Bore, near Perth, W.A.; portion of exine
showing pila and clavae, x c. 1350; no. 51417.

Palaeontology, Vol. 8

PLATE 9

m

COOKSON, Cretaceous microspores

THE DEVELOPMENT OF A DICELLOG RAPTI D
FROM THE B A LCLATCH I E SHALES OF LAGGAN

BURN

by JUDITH JAMES

Abstract. This paper decribes the growth series, the thecal morphology, and the origin of the thecae of Dicello-
graptus sp. Points of comparison with other dicellograptids are noted.

The material has been obtained from thin bands of nodular limestone occurring in the
Balclatchie Shales of the Ardmillan Series of Laggan Burn near Girvan. The exposure
is referred to in Lapworth’s Girvan Succession (Lapworth 1882, p. 591, 2), in Peach and
Horne (1899, p. 512), and in Bulman (1944-7, pp. i, iii). Alwyn Williams (1963) de-
scribes the stratigraphy and palaeontology of the area.

The normal technique for isolating graptolites from impure limestone was employed.
Treatment with hydrochloric acid removed all the calcium carbonate, whilst con-
centrated hydrofluoric acid disintegrated the well-washed remaining material to a
muddy sludge. The free graptolites were then isolated. Most of the specimens yielded
to bleaching with concentrated nitric acid and potassium chlorate. The time taken for
this process to be completed depended on the individual specimens, but usually varied
between fifteen minutes and one hour. Older specimens failed to bleach readily due to
secondary thickening of the periderm, and prolonged bleaching tended to destroy the
growth lines.

The limestone nodules contained broken stipes (never more than four thecae in length),
a few mature proximal ends, and a number of growth stages of the form described below.
The rest of the fauna comprised Dicranograptus nicholsoni Hopkinson, Diplograptus
leptotheca Bulman, Climacograptus bicornis Hall, Climacograptus brevis Elies and Wood,
Pseudoclimacograptus scharenbergi Lapworth, Orthograptus apiculatus (Elies andWood)
Bulman, Lasiograptus and Cryptograptus in various stages of development.

The originals of the figured specimens have been placed in the Sedgwick Museum and
bear the Museum’s catalogue numbers.

SYSTEMATIC PALAEONTOLOGY

Family dicranograptidae Lapworth
Genus dicellograptus Hopkinson

Dicellograptus sp.

Description. Rhabdosome broken, only a few thecae in length. Axial angle approxi-
mately 260°-320°. Overall sicular length 1 -0-1 -35 mm., usually 1-25 mm. (of which
one-third to one-quarter is the prosicula), commonly incorporated to a varying extent
into the second stipe. The tip of the prosicula prolonged into a short nema. Complete

[Palaeontology, Yol. 8, Part 1, 1965, pp. 41-53.]

42

PALAEONTOLOGY, VOLUME 8

virgella spine exceeds the metasicula length. Sicula width
increases gradually to approximately 0-2 mm. at the aper-
ture. Width of stipe at level of mesial spine on th 21 040-
0-55 mm. Lengths of distal thecae 1-56-1 -90 mm. All
thecae examined bear mesial spines 0-25-0-30 mm.
from ventral lip of aperture; free ventral wall of thecae
inclined slightly outwards to mesial spine. Apertures
introverted and more or less covered by apertural
flanges. Width and length of excavation of th 22 0-12
mm. and 0-33 mm. respectively. Thecae overlap for
approximately half their length.

This species corresponds most closely in general
form, shape of thecae, structure of proximal end, and
the position of the sicula, with Dicellograplus divari-
catus Hall (Elies and Wood 1904, p. 143, pi. XX, figs.
5a, 5b, and Elies 1940, p. 429), but bears apertural
flanges not apparent in Dicellograptus divaricatus and
is therefore not placed in this species. It is felt that no
specimen of the form described below is complete
enough to justify the erection of a new variety, and
the existing specimens of Dicellograptus divaricatus,
described by Elies and Wood 1904 and Elies 1940 are
not well enough preserved to establish identity with
these forms.

GENERAL MORPHOLOGY
The development of the early growth stages

The initial bud arises from a foramen approximately half way down the length of the
sicula, a little above the point of origin of the virgella. It grows down the wall of the
sicula as a narrow tube, until the foramen for th l2 is formed, high up on the initial
bud (text-fig. 1a, b). Early stages of the formation of the th l1 and the foramen of th l2
are absent. Th l1 now continues to grow as a much wider tube down the wall of the
sicula, until just below the sicula aperture, it turns abruptly and in the form of a U is
directed outwards and then upwards (text-fig. 2a, b). The space between the two limbs
of the U will later be filled by th 21.

Th l2 grows at first upwards and inwards, and then, curving round in the form of
a hood is directed downwards close to th l1 and the reverse sicula wall (text-fig. 2a, b).
Th 21 arises from the virgella side of the earliest downwardly directed part of th l2
when th l2 is still in the early stage of growth. Th l1 now appears as a more or less
complete split tube; the dorsal wall is present, but seemingly only until the level of
the mesial spine. Th l2 continues growth downwards, at first more or less parallel to
the initial bud, and later obliquely across the reverse sicula wall, until, just before
reaching the aperture, it turns outwards and then upwards in a similar manner to th l1.
The space between the two arms of the U formed in this way will be filled by th 22
(text-figs. 3a, 4a).

A B

text-fig. 1 . First-theca growth stage
of Dicellograptus sp., x45 approx.
a, Reverse side, A54471. b, Obverse.

J. JAMES: DEVELOPMENT OF A DICELLOGRAPTID FROM SCOTLAND 43

text-fig. 2. Second- and third-theca growth stages of Dicellograptus sp. ;
all X 45 approx, a, Reverse side, A54472. b. Reverse, A54473. c, d, A54474;
c, Reverse; d, Obverse.

Th 21 grows downwards close to th l2 and th l1 until approximately the level of the
mesial spine on th l1. The aperture is directed slightly inwards. A thickened diaphragm
of chitin occurs within the tube and whilst the exact nature of this is uncertain (text-
figs. 2c, d, 3a, b, c, d), it is most probably related to the origin of th 22. An upward-
growing flange unites with the initial portion of th 21 (text-fig. 3 d) producing two
apertures from which develop the later portion of th 21, on that side facing th l1, and
the initial part of th 22. Th 21 grows abruptly upwards filling the space formed by the U
shape of th l1 (text-figs. 3b, c, d). Th 22 grows across the sicula, and obliquely upwards

44

PALAEONTOLOGY, VOLUME 8

text-fig. 3. Second- and third-theca growth stages of Dicellograptus sp., showing flanges
(/) over the aperture of the first theca; all x 45 approx, a. Reverse, A54475. b. Obverse,
A54476. c. Obverse, A54477; d, Reverse.

between the limbs of th l2 (text-figs. 4a, 5a). Th 31 arises from th 21 and th 32 from th 22
(text-fig. 5a).

The development of Dicellograptus sp. may be closely compared with those of
Dicellograptus divaricatus var. sa/opiensis Elies and Wood (Strachan 1959), Dicello-
graptus geniculatus Bulman (Bulman 1932), Leptograptus sp. (Whittington 1955), and

J. JAMES: DEVELOPMENT OF A DICELLOG R APTI D FROM SCOTLAND 45

A

text-fig. 4. Fourth-theca growth stage and developing prothecal base of
fifth theca of Dicellograptus sp. a, Reverse side, b, Obverse side of same
specimen, A54478, x45 approx, pb, prothecal base; i, inner wall of
prothecal base; d, dorsal wall of prothecal base;/, flange.

Dicranograptus nicholsoni Hopkinson (Bulman 1944). In all these forms the initial up-
ward growth of th l2, its hood-like curvature down, and the downgrowth of the first
formed part of th 21 imparts to the proximal end a close resemblance to that of Glypto-
graptus dentatus and other primitive diplograptids (Bulman 1944). The form described
in this paper resembles more closely Dicranograptus nicholsoni Hopkinson, than the

46

PALAEONTOLOGY, VOLUME 8

text-fig. 5. Fifth- and sixth-theca growth stage of Dicellograptus
sp. ; X 45 approx, a, Reverse side, A54479. b, Obverse side,
A54480. Both show prothecal base of youngest theca; /', inner
wall of prothecal base; d, dorsal wall of prothecal base;/, flange.

other forms referred to above in the compactness of the proximal end. Only a small
portion of th l1 and th l2 show horizontal growth, and the distal portions are directed
upwards and even occasionally inwards (text-fig. 6a). Variation in the growth of the sicula
and the first two thecae are shown by the three superimposed outlines of text-fig. 6a.

J. JAMES: DEVELOPMENT OF A DICELLOG R APTI D FROM SCOTLAND 47

text-fig. 6. Variation in growth of intitial bud and first two thecae
of Dicellograptus sp. shown by three superimposed outlines: a,

Obverse side, solid line, A54481, dashed line, A54482; dot-dash
line, A54483. b, Diagram of growth of proximal thecae. Both
x 45 approx.

Thecal morphology

The adult thecae overlap for half their length, which varies from 1 -56 mm. to 1 -99 mm.

The greatest width of the stipe (at the level of the mesial spine) varies from 0-58 to 0-76 mm.

The more proximal thecae are smaller. The free ventral wall of the thecal segment

48

PALAEONTOLOGY, VOLUME 8

inclines slightly outwards as far as the mesial spine, approximately 0-5 mm. from the
aperture, to which it then inclines inwards (text-fig. 7). All the thecae examined have a
mesial spine, and one theca (text-fig. 7) appears to have two. The breadth of the thecae
is not constant, narrowing considerably above the aperture of the preceding theca until
the mesial spine is reached, after which it quickly expands to the wide apertural lip

text-fig. 7. Fragment of stipe of text-fig. 8. Fragments of thecae of Dicellograptus sp.

Dicellograptus sp. showing adult showing the aperture and the flanges over the aperture,

thecae, A54484; X 22 approx. x45 approx, a, A54484. b, A54485. c, 54486.

(text-fig. 7). Lists are developed longitudinally along the mid-ventral line of each theca
and transversely at the level of the mesial spine. The aperture is oblique and introverted
and lies in a deep excavation. The ventral wall of the lip of the aperture is not straight
but forms a broad, low, mid-ventral lappet. The aperture is more or less covered by two
oblique, downwardly and outwardly directed flanges, one on either side, situated below
the geniculum of the succeeding theca (text-figs. 7, 8).

The flanges are composed of fuselli, similar in thickness and width to those of the
stipe in general. The growth lines of these flanges (of which there are approximately
six), are more or less parallel to the free outer edge. Although there is no projecting
flange from the mid-ventral area, thickening, as well as the convergence of the growth

J. JAMES: DEVELOPMENT OF A DICELLOGRAPTID FROM SCOTLAND 49

text-fig. 9. Stages in the growth of the protheca of Dicellograptus sp. ; all X 45 approx, a, b, Dorsal
and ventral sides of the same specimen, A54487. c, Dorsal, A54488. d. Dorsal, A54489. pb, prothecal
base; /, inner wall of prothecal base; d, dorsal wall of prothecal base;/, flange.

lines in this region, suggests that the flanges are continuous one with the other. This is
confirmed by the development. It is here, where the flanges appear least protective, that
the ventral lip of the aperture rises to form a broad lappet. A closer study of the flanges
in the mature stipe, and also of the stages of their development, suggests that they are a
continuation of the fusellar system of the dorsal wall of the metatheca, rather than a
derivative of the ventral wall of the succeeding protheca. For example, the dorsal wall

B 6612

E

50

PALAEONTOLOGY, VOLUME 8

of a theca in the region of the aperture is continuous with the flange over the aperture,
so that it is impossible to say where one begins and the other ends (text-figs. 3b, c, d;
4; 9d). The growth lines of this wall and those of the flange are seemingly conformable,
converging over the aperture towards the lateral edge. Growth lines of the ventral wall
of the succeeding theca rise upwards and converge towards the dorsal wall of the
aperture of the preceding theca, suggesting the presence of the latter (and the flanges
with which it is continuous (during the growth of the former.

The development of the flanges of the proximal thecae

Th l1 appears to be a more or less complete split tube when th l2 is in an early stage
of growth, and it is not until th l2 has reached the formation of a spine that a wall
across the dorsal part of the aperture becomes distinct. A thick band of chitin is present
at the base of the dorsal wall of the aperture, and is common along the lateral edge of

text-fig. 10. Diagrammatic longitudinal section of fragment of stipe of
Dicellograptus sp., th 3 is shaded. Protheca and metatheca refer to th 3,
X 30 approx, pb, prothecal base; /, inner wall of prothecal base; d, dorsal
wall of prothecal base; i.s., interthecal septum.

the former (text-fig. 3a). Growth now continues and flanges develop over the aperture.
The specimen shown in text-fig. 3c, d has complete dorsal wall and flanges of th l1
when th 21 is at an early stage of development. Th l2 is seemingly as yet a split tube.

The development of the thecal segment

Each thecal segment consists of a protheca, which ‘corresponds approximately at
least with the stolotheca of the Dendroidea' (Bulman 1951) and a metatheca which
corresponds approximately to the autotheca of the Dendroidea. The protheca com-
mences growth at approximately the level of the aperture of the second preceding theca
(text-fig. 10). It has a free dorsal wall, and a ventral wall which is the interthecal septum
lying between it and the metatheca of the preceding theca. The metatheca has a free
ventral wall, and a dorsal wall which is the interthecal septum lying between it and the
protheca of the succeeding theca. It has an aperture distally which communicates with
the exterior. The development of the protheca involves the formation of a structure
here named the prothecal base (text-figs. 10, 4, 11,5, 9). This has the form of a cylinder
flattened dorso-ventrally, and curving inwards such that the plane of its aperture is
parallel to the direction of the stipe (text-figs. 5b, 9b). The dorsal wall of the cylinder
(labelled d ), is continuous with the dorsal wall of the preceding protheca. The ventral
inner wall of the cylinder (labelled /), is continuous with the interthecal septum. Text-
fig. 4a, b shows the prothecal base of th 31 developing from the distal portion of the
protheca of th 21. Th 21 although broken, is almost complete, but it is uncertain whether
the dorsal wall of the metatheca (i.e., the interthecal septum between it and the new
unformed protheca) is formed. The thickened and slightly frayed nature of the ventro-

J. JAMES: DEVELOPMENT OF A D I CEL LOG R A PTI D FROM SCOTLAND 51

lateral edges suggest that it was formed, but not preserved. An inner wall appears to
be present, although the relationship between it and the developing dorsal wall of the
prothecal base and the lateral wall of the preceding protheca is obscure. Although in

B

C

D

text-fig. 11. Part of longitudinal section series of Dicellograptus sp.,
specimen A54491; section interval 10 p, x45 approx, a, aperture;
/, flange; pb, prothecal base, a, b, c, d, section numbers 21, 20, 18, 17

respectively.

text-fig. 9a, b the metatheca of th 3 has only reached the stage of the formation of a
spine, the prothecal base of th 4 is more advanced than the last example. It has more
growth lines and is now curving inwards due to greater deposition of chitin dorsally
than laterally. The aperture of the prothecal base of this specimen can be seen in ventral
view (text-fig. 9b). The rim of the inner wall proximally, the growth relationships between
this, the interthecal septum, the dorsal wall, and the lateral wall, are not clear. Some time
after the completion of the prothecal base, the dorsal and lateral walls, and perhaps the

52

PALAEONTOLOGY, VOLUME 8

interthecal septum of the new protheca, develop. The space between the prothecal base
and the metatheca of the preceding theca is filled by deposition of chitin in a manner
suggested by the growth lines (text-figs. 9c, d). At first deposition is more or less parallel
to the incurling edge of the prothecal base, but later, when the most distal part of the
structure is reached, bands of chitin are laid down over the dorsal part, forming the
new dorsal wall of the developing protheca, as well as continuing laterally. Text-fig. 5b
shows the lateral wall just beginning to extend back over the dorsal wall of the prothecal
base. It is at a slightly later stage of development in text-figs. 9c, d.

Thus the ventral wall of the metatheca is more or less complete (text-figs. 4a, 5a) when
the prothecal base of the succeeding theca is just beginning to form. It is possible that the

text-fig. 12. Proximal part of the rhabdosome
of Dicellogrciptus sp., A54481 ; X 22 approx.

interthecal septum is also present as a thin and easily destroyed membrane. A thickened
band of chitin is common across the base of the dorsal part of the aperture of the
metatheca (as it is in the proximal thecae), and may suggest a discontinuity of growth in
this region. The dorsal wall of the aperture of the metatheca becomes apparent as the
lateral and dorsal walls of the new succeeding protheca are forming, developing in a
similar manner to those of the proximal thecae. The wall extends over the aperture and
forms the flanges (text-figs. 9c, d).

The interthecal septum

The interthecal septum lies between the protheca of one thecal segment and the meta-
theca of the preceding segment. Longitudinal sections of the stipe (text-figs. 10, 11)
show that it is continuous proximally with the inner wall of the prothecal base, and
distally with the dorsal wall of the aperture and consequently the flanges over the aper-
ture. The zooid (or zooids) responsible for the formation of the septum, apertural
flanges, and the inner wall of the prothecal base, is not known, but since the flanges over
the aperture of one theca are complete when the succeeding protheca is at an early stage
of growth (text-figs. 3b, c, d; 9d) it seems likely that the zooid responsible for their
formation was the zooid over whose aperture the flanges lie.

Acknowledgements. I should like to express my grateful thanks to Professor O. M. B. Bulman for
constant help and advice throughout this work, and who made available to me the material from a
collection supplied by Dr. J. Pringle. I am indebted also to the Department of Scientific and Industrial
Research for an award, during tenure of which the work has been carried out.

J. JAMES: DEVELOPMENT OF A D ICELLOGR APTID FROM SCOTLAND 53

REFERENCES

bulman, o. m. b. 1932. On the Graptolites prepared by Elolm 2-5. Ark. f. Zool. 24A, 9, 1-29.

1944-7. Caradoc Graptolites from Laggan Burn. Palaeontogr. Soc. Monogr., London, 1-78.

1951. Thecal Variation in Monograptus. Geok Mag. 78, 316-28.

elles, G. l. 1939. The Stratigraphy and Faunal Succession in the Ordovician Rocks of the Builth-
Llandrindod Inlier, Radnorshire. Quart. J. geok Soc. Lond. 95, 383-442.

and wood, e. m. r. 1901-1918. Monograph of British Graptolites. Palaeontogr. Soc. Monogr.,

London, 1-526.

lapworth, c. 1882. The Girvan Succession. Part I. Stratigraphy. Quart. J. geok Soc. Lond. 38,
537-662.

peach, b. n., and horne, J. 1899. The Silurian Rocks of Britain: 1, Scotland. Mem. geok Surv. U.K.
strachan, I. 1959. Graptolites from the Ludibundus Beds (Middle Ordovician) of Tvaren, Sweden.
Bull, geok Instn. Univ. Uppsala, 38, 47-60.

Whittington, h. b. 1955. Additional new Ordovician Graptolites and a Chitinozoan from Oklahoma.
J. Paleont. 29, 837-51.

williams, A. 1963. The Barr and Lower Ardmillan Series (Caradoc) of the Girvan District, South-West
Ayrshire. Mem. Geok Soc. London. 3, 5-267.

JUDITH JAMES
Department of Geology,
Sedgwick Museum,

Manuscript received 3 February 1964 Cambridge

A LOWER CARBONIFEROUS FAUNA FROM
TREVALFYN, NEW SOUTH WAFES

by JOHN ROBERTS

Abstract. A Lower Carboniferous (Visean II5— 1 1 1^) fauna from Trevallyn near Gresford, N.S.W., is listed
and the age of the fauna briefly discussed. The species considered in detail are: Fenestellci allynensis sp. nov.,
Werriea australis Campbell, Eomarginifera tenuimontis sp. nov., Marginatia patersonensis sp. nov., Cleiothyri-
clina australis Maxwell, Kitakamithyris triseptata (Campbell), Dielasnia picketti sp. nov., ‘ Caiuarotoechia' sp. B,
Stenoscisma laevis sp. nov., Prolecanites sp., ? Girtypecten sp., Pernopecten trevallynensis sp. nov., and Diodontop-
teria delicata sp. nov.

Marine fossils were first collected from Trevallyn during the geological mapping of
the Gresford district (Roberts, 1961). A search of the literature reveals no record of any
previous collection from this locality which is situated immediately east of the Paterson-
Gresford road 2 miles south of the town of Gresford and approximately 20 miles north
of the city of Maitland (text-fig. 1).

Grid references quoted in this paper are from the Dungog One Mile Military Sheet.
All fossil locality numbers refer to the palaeontological register at the University of New
England, Armidale, N.S.W.

STRATIGRAPHY

The stratigraphy of the Gresford district has been described by Roberts (1961). The
Trevallyn fossil localities occur in the Bingleburra Formation, the oldest exposed forma-
tion in the Gresford district, which consists dominantly of mudstones and siltstones
together with interbedded lithic sandstones, conglomerates and oolitic and crinoidal
limestone lenses. Thickness in the type section 7 miles north of Trevallyn is approxi-
mately 3,000 feet, the base being unexposed. Near Gresford the formation has two
different facies; a bank-type environment in the north containing oolitic and crinoidal
limestones grades southwards into a region characterized by conglomerates and coarse
lithic sandstones. The Trevallyn localities occur in this southern part of the Bingleburra
Formation and are situated towards the top of a stratigraphic section in the Trevallyn
fault block (Roberts, 1961, fig. 1) east of Gresford. The fauna is approximately 2,000 feet
stratigraphically above the exposed base of the section at Fewinsbrook where there is a
rich Upper Tournaisian fossil assemblage (Roberts, 1963). Because of faulting neither
the top nor the bottom of the Bingleburra Formation is exposed in this block, but it is
clear from the composition of the fauna that the Trevallyn horizons occur high in the
formation.

Localities. The Trevallyn localities are situated at the following grid references, L. 233 being highest in
the sequence; L. 208 at 573864; L. 270 at 572864; and L. 233 at 570864. The locality bordering the Pater-
son-Gresford road previously referred to L. 207 (Roberts, 1961) is now incorporated in L. 233. Text-
fig. 2 shows the stratigraphic relationships between the three horizons. The L. 233 horizon is exposed
in a gully running north-east from Trevallyn quarry. Most fossils have been collected from hard

[Palaeontology, Vol. 8, Part 1, 1965, pp. 54-81, pi. 10-13 ]

J. ROBERTS: LOWER CARBONIFEROUS FAUNA FROM NEW SOUTH WALES 55

grey siltstone immediately above or below a thin lens of dark fossiliferous limestone containing a
small cephalopod fauna (Brown, Campbell, and Roberts, in press). L. 270 is situated at the top of a
hill east of Trevallyn quarry and occurs in grey siltstones. The siltstone horizon can be traced
approximately half a mile north and south of the L. 270 collecting point. The locality L. 208 is found
on the eastern slope of a hill due east from Trevallyn quarry in a bed of coarse calcareous
feidspathic sandstone. The bed is lenticular and so the horizon cannot be traced for more than
100 yards.

FAUNA

The following is a complete list of all identifiable species collected, those forms
described in this paper being marked by an asterisk.

L. 233: Fistulamina inornata Crockford, *FenestelIa allynensis sp. nov., Gonioclcidia
laxa (de Koninck), Ramipora bifurcata Crockford, Clodochonus tenuicollis M'Coy,
Streptorhynchus spinigerci (M’Coy), Rhipidomella australis (M’Coy), Schizophoria verula-
mensis Cvancara, Leptagonia cf. L. analoga (Phillips), Rugosochonetes kennedyensis

56

PALAEONTOLOGY, VOLUME 8

Maxwell, Rugosochonetes sp., Krotovia sp., Fluctuaria sp., *Eomarginifera tenuimontis
sp. nov., *Marginatia patersonensis sp. nov. Waagenoeonelia delicatula Campbell.
*Cleiothyridina australis Maxwell, * Kitakamithyris triseptata (Campbell), Kitakamithyris
uniplicata (Campbell), Unispirifer striatoconvolutus (Benson & Dun), Acuminothyris

Trevollyn

fault

O

O

siltstone
mudstone
sandstone
limestone
fossil horizon

text-fig. 2. Stratigraphical column of the upper portion
of the Lower Carboniferous Bingleburra Formation at
Trevallyn.

triangularis Roberts, Asyrinxia lata (M’Coy), *Dielasma picketti sp. nov., Camay o-

toechia'' sp. B., * Stenoscisma laevis sp. nov., *Girtypecten sp., *Pernopecten trevallv-
nensis sp. nov., *Diodontopteria delicata sp. nov., Streblopteria sp., *Prolecanites sp.,
Beyrichoceras trevallynense Brown, Campbell, and Roberts, Mooreoceras regulare
Brown, Campbell, and Roberts, Vestinautilus sp., Knightoceras sp., Phillipsia cf. P.
dungogensis Mitchell.

L. 270: Schizophoria verulamensis Cvancara, *Cleiothyridina australis Maxwell.

L. 208: Streptorhynchus spinigera (M’Coy), *Werriea australis Campbell, Scliizo-

J. ROBERTS: LOWER CARBONIFEROUS FAUNA FROM NEW SOUTH WALES 57

phoria verulamensis Cvancara, Leptagonia cf. L. ana/oga (Phillips), Krotovia sp.,
*Eomarginifera tenuimontis sp. nov., *Marginatia patersonensis sp. nov., Waageno-
concha delicatula Campbell, *Cleiothyridina australis Maxwell, Kitakamithyris uniplicata
(Campbell), * Kitakamithyris triseptata (Campbell), Brachythyris elliptica Roberts,
Unispirifer striatoconvolutus (Benson and Dun), Camarotoechia ’ sp. B., Diodontopteria
sp., Platyceras sp., Phillipsia sp.

The Trevallyn assemblage is quite distinct from the Upper Tournaisian fauna from
Lewinsbrook (Roberts, 1963) occurring approximately 2,000 feet stratigraphically lower
in the Bingleburra Formation. The Lewinsbrook fauna is characterized by the following
species: FenesteUa brownei Roberts, F. gresfordensis Roberts, F. wilsoni Roberts,
Bibucia tubiformis Roberts, Productina globosa Roberts, Pustula multispinata Roberts,
? Thomasaria voiseyi Roberts, Streblochondria obsoleta Roberts, and Conophillipsia
brevicaudata Roberts.

The majority of the species from Trevallyn have also been recorded from L. 53 Green-
hills (also known as Hilldale), 1-5 miles south-west of the village of Hilldale. The
composition of the Trevallyn fauna suggests that it is slightly older than that from Green-
hills. Species present at Trevallyn and absent from Greenhills are: FenesteUa allynensis
sp. nov., Brachythyris elliptica Roberts, and Acuminothyris triangularis Roberts. Brachy-
thyris elliptica, however, appears to be longer ranging than suggested here and has been
collected from L. 204 Lewinsbrook Syncline at the base of the Bonnington Formation
in the northern part of the district (Roberts, 1961). L. 233 Trevallyn is stratigraphically
the highest known locality from which Acuminothyris triangularis has been recorded.

AGE OF THE FAUNA

The age of the Trevallyn fauna has been determined as Middle Visean, possibly upper
Ilg or 111,*, by Brown, Campbell, and Roberts (in press). This is based on the occurrence
of Beyrichoceras trevallynense Brown, Campbell and Roberts, which is morphologically
similar to B. submicronotum Bisat, and Prolecanites sp.

Support for the Middle Visean age is provided by Werriea australis Campbell which
is very close to Werriea keokuk (Hall) from the Keokuk Limestone in the Mississippi
Valley, U.S.A. (Campbell, 1957). Collinson et al. (1962) have correlated the Keokuk
Limestone with the II§ zone of Germany.

Additional evidence is provided by the presence of the following distinctive species
in the Middle Visean Babbinboon fauna in the Werrie Basin, N.S.W. ; Werriea australis
Campbell, Waagenoconcha delicatula Campbell, Kitakamithyris triseptata (Campbell),
and K. uniplicata (Campbell).

Type of preservation. Fossils from L. 208 occur as well-preserved external and internal
moulds in the weathered portions of the bed. Shelly material still remains in the un-
weathered rock and moulds must be prepared by leaching with hydrochloric acid. In
weathered parts of L. 233 and L. 270 fossils are preserved as internal and external
moulds. Shell material is present in a limestone lens interbedded with the fossiliferous
siltstones and also occurs in the unweathered portions of the siltstone bed.

58

PALAEONTOLOGY, VOLUME 8

SYSTEMATIC DESCRIPTIONS

Family fenestellidae King 1850
Genus fenestella Lonsdale 1839

Type species. Fenestella subantiqua d'Orbigny, 1852, from the Wenlock, Dudley, England.

Fenestella allynensis sp. nov.

Plate 13, figs. 5-7

Diagnosis. Fine rectangular mesh; branches wide, flat, bearing a broadly rounded carina
with four nodes per fenestrule plus one on dissepiment; fenestrules rectangular; aper-
tures hooded distally; two to three apertures per fenestrule plus one on dissepiment;
reverse surface of branches finely striate.

TABLE 1

Speci-

men

Num-

Branches

Dissepi-

ments

Fenestrules

Apertures

Nodes

No. in 10

No. in

No. per
Fene-

No. per

No. per
Fene-

ber

Width

mm.

Width

Width

Length

10 mm.

Diameter

strule

5 mm.

strule

F. 6904

0-3-0-35

17

(8 in 5 mm.)

0 1-0-15

0-3-0-4

0-75-1

8-9

0-15

2-3 + 1
on

dissep.

14-15

4 + 1 on
dissep.

F. 6914

0-25

(9 in 5 mm.)

0-1-0-15

0-25-0-3

0-6-0-7

10

0-14

2-3+1

on

dissep.

4+ 1 on
dissep.

F. 6913

0-3

20

0 15

0-25-0-3

0-65-0-8

10

0-1-014

3 + 1 on
dissep.

19

4 + 1 on
dissep.

F. 6905

0-3

| 16 at base
( 19 at top

0 15

0-2-0-3

0-75

10

0-14

3 + 1 on
dissep.

21

4+ 1 on
dissep.

F. 6915

0-3

22-28

0-15

0-25

0-75-0-8

0-15

2-4+1

on

dissep.

20

4+ 1 on
dissep.

Description. Colony broadly conical with very regular meshwork; bifurcation distant;
branches moderately wide, 0-25-0-35 mm., with density of 16-20 per 10 mm. in older
portions of colony and 22-28 per 10 mm. in younger portions; reverse surface of branch
ornamented with fine longitudinal striae; obverse surface of branch flat below carina;
carina broadly rounded, 0-1 mm. wide, moderately high, bearing small pointed nodes
having a density of 4 per fenestrule plus one on the dissepiment; apertures sub-circular,
0-1-0-15 mm. in diameter, hooded distally, with narrow peristome and placed well
below carina; two to three (occasionally four) apertures per fenestrule plus one on dis-
sepiment, and approximately 14 and 21 apertures per 5 mm. on an older and younger
portion of the colony, respectively; fenestrules rectangular to sub-rectangular, from
0-25-0-4 mm. wide and from 0-6-1 mm. long, the latter averaging 0-75 mm.; dissepi-
ments short, thin, 0-1-0-15 mm. long and approximately level with both reverse and
obverse sides of branches; zooecia semi-polygonal, closely spaced, with zigzag median
interlocking section. For measurements see Table 1.

Remarks. Two Eastern Australian Carboniferous species may be compared with
Fenestella allynensis. F. cellulosa Crockford (1947) from Barrington House, Barrington

J. ROBERTS: LOWER CARBONIFEROUS FAUNA FROM NEW SOUTH WALES 59

Tops, has a finer meshwork, a longer carina, smaller apertures and a smooth reverse
surface. F. cribriformis Crockford (1947), from Rouchel Brook, has thinner branches,
a more narrow carina and a smooth reverse surface. Of overseas species, F. serratula
Ulrich described by Condra and Elias (1944) and Koenig (1958) from the Warsaw Beds,
Illinois, superficially resembles F. allynensis. The Trevallyn species has one additional
aperture per fenestrule and a slightly coarser meshwork than that of F. serratula.

F. frutex M’Coy, revised by Miller (1961), has morphologically similar apertures,
keel, and zooecia, and also has a comparable regularity in the meshwork. However, it is
distinguished from F. allynensis by its shorter ovoid fenestrules and one or two less
apertures per fenestrule. F. frutex is common in the Lower Carboniferous rocks of
Europe, Turkestan, and North America. These same distinctions apply to F. tenax
Ulrich (1890), also described by Condra and Elias (1944), from the Mississippian
Warsaw and Chester Beds of Illinois.

This species is named after the Allyn River near Gresford, N.S.W. It is known only
from the type locality, L. 233 Trevallyn.

Material. F. 6904--F. 6921. Holotype F. 6904, paratvpes F. 6905-F. 6908.

Family schuchertellidae Stehli 1954
Genus werriea Campbell 1957

Type species. (By original designation) Werriea australis Campbell, 1957, from the Namoi Formation,
Werrie Basin, N.S.W.

Remarks. Campbell (1957) separated Werriea from Orthotetes Fischer on the basis of
the morphology of the spondylium and median septum in the pedicle valve. He obtained
details of Orthotetes from Sokolskaya’s (1954) redescription of the genus and under-
stood that in young specimens a small ‘ primary’ spondylium formed by the unification
of dental plates was prolonged into a median septum. In Werriea , the septum and the
dental lamellae are completely separate in young stages and become fused in later stages
of growth to form a ‘secondary’ spondylium. Lane (1963), however, was hesitant to
accept this difference because Campbell’s interpretation was based on one poor text-
figure in Sokolskaya (1954), and suggested that Werriea may be synonymous with
Orthotetes.

The genus Pseudoorthotetes Sokolskaya (1963) appears to be very close to Werriea
but is not fully understood by the author. It is distinguished from Werriea by the poses-
sion of longer ‘crural plates’ and longer lobes on the cardinal process (Sokolskaya,
p. 98). It is not clear whether the ‘crural plates’ refer to the whole of the socket plates
and their buttressing ridges or to only part of these structures. The type of spondylium
is also obscure. For the present, Pseudoorthotetes is probably best regarded as a junior
synonym of Werriea.

G. A. Thomas (1958, p. 20) noted that Permorthotetes may be synonymous with
Werriea. Both genera possess the same type of secondary spondylium, but Permorthc-
tetes can be distinguished from Werriea by the possession of curving divergent socket
plates which always have small projections on their anterior extremities, and buttress
ridges extending antero-laterally along the floor of the valve from the ends of the
socket plates. The socket plates in Werriea are recurved, sub-parallel with the hinge and
are not supported anteriorly by buttressing ridges.

60

PALAEONTOLOGY, VOLUME 8

The relationship of Permorthotetes with Orthotetes is mainly dependent on the nature
of the spondylium in the latter genus, although the distinctive divergent socket plates
with distal projections and buttressing ridges in Permorthotetes should be sufficient to
distinguish between the two genera.

The present material is placed in Werriea until the relationship between this genus
and Orthotetes is clarified.

Werriea australis Campbell

Plate 10, figs. 10-14

Werriea australis Campbell 1957, p. 45, pi. 11, figs. 1-7.

Remarks. The specimens from L. 208 Trevallyn and L. 53 Greenhills agree with Camp-
bell’s type material from Babbinboon in all features except those listed below. (1) The
number of lirae in the intercostal troughs ranges from 1-5, compared with 1-2 noted by
Campbell. (2) The socket plates are only slightly divergent from the hinge in contrast
with a divergence of approximately 30° in the type material. (3) Adductor scars in mature
brachial valves are moderately impressed and are triangular to flabellate in outline. They
have rounded posterior terminations and straight or obsolete anterior margins. These
scars were apparently obscure in the type material and were not described.

Occurrence. Werriea australis Campbell is known from L. 35 Babbinboon, the type
locality, L. 208 Trevallyn and L. 53 Greenhills.

Material. F. 6789-F. 6800.

Family marginiferidae Stehli 1954
Genus eomarginifera Muir-Wood 1930

Type species. (By original designation) Productus longispinus J. Sowerby, 1814, from the Lower
Carboniferous of Linlithgowshire, Scotland.

Remarks. Minor features of this material which are not in agreement with the revised
diagnosis of the genus (Muir-Wood and Cooper 1960) are as follows: (1) A marginal

EXPLANATION OF PLATE 10

Figs. 1-5. Marginatia patersonensis sp. nov. 1, F. 6828a. Internal mould of pedicle valve, X 1-5. 2,
F. 6815. Internal mould of pedicle valve; paratype, X 1. 3, F. 6813. Rubber cast of pedicle valve
exterior; note the row of spines along the hinge; holotype, X 1. 4, F. 6817. Rubber cast of brachial
valve interior. (L. 53 Greenhills), X 2. 5, F. 6825c. Internal mould of brachial valve, X 1.

Figs. 6-10. Eomarginifera tenuimontis sp. nov. 6a, F. 6701. Rubber cast of brachial valve interior; holo-
type, X 1-6. 6b, F. 6701. Internal mould of the same valve and apex of pedicle valve; holotype, X 1-6.
6c, F. 6701. Rubber cast of brachial valve exterior and apex of pedicle valve; holotype, xl-6.
7, F. 6704. Rubber cast of brachial valve interior, X 1 -6. 8, F. 6825a. Rubber cast of pedicle valve
exterior, X F5. 9a, F. 6708. Internal mould of pedicle valve; paratype, X 1-7. 9b, F. 6708. Lateral
view of the same specimen; paratype, X 1-7. 10, F. 6717. Internal mould of pedicle valve showing
well-developed marginal ridge, X 1-5.

Figs. 11-16. Cleiothyridina australis Maxwell 11, F. 6688. Rubber cast of pedicle valve exterior, x 2.
12, F. 6684. Internal mould of brachial valve, X 2. 13, F. 6687. Rubber cast of pedicle valve exterior,
X 1-6. 14, F. 6686. Internal mould of pedicle valve, X 2. 15, F. 4815. Internal mould of pedicle valve,
X 2. 16, F. 6689. Internal mould of pedicle valve and apical part of brachial valve, X 2.

Palaeontology, Vol. 8

PLATE 10

ROBERTS, Australian Carboniferous fauna

J. ROBERTS: LOWER CARBONIFEROUS FAUNA FROM NEW SOUTH WALES 61

ridge is present on the interior of some, but apparently not all, pedicle valves. (2) In the
region between the adductor scars on the brachial valve the median septum has a
variable strength but is usually stronger than the ‘thread-like breviseptum’ in the type
material. (3) The type material may be more strongly spinose on the interior of the
brachial valve.

Eomarginifera tenuimontis sp. nov.

Plate 10, figs. 6-10

Diagnosis. Shell average size for genus; hinge-line produced into small auricles; fourteen
to fifteen costellae per 10 mm. on median anterior portion of shell; six to eight con-
centric ribs on posterior half of brachial valve; ribs weaker on pedicle valve; occasionally
a marginal ridge in pedicle interior; two pairs of adductor scars in brachial valve;
brachial ridges enclosing kidney-shaped brachial impressions; cardinal ridge 10°-15°
divergent from hinge.

Description. External. Shell of average size for genus, strongly concavo-convex, globular
to rectangular, with a long steep trail; hinge-line straight, approximately equal in width
to widest part of shell, and bearing small flattened auricles on its lateral extremities;
costellae well defined on body of shell but becoming faint towards tip of umbo on pedicle
valve and on steep posterior portions of brachial valve; density of costellae 14-15 per
10 mm., measured laterally on median anterior portion of shell; costellae increase by
intercalation and bifurcation; six to eight concentric ribs on posterior half of brachial
valve more strongly defined than those on same region of pedicle valve; six large
halteroid spines regularly placed on pedicle valve, two on lateral slopes, two on ears,
and two on median anterior portion of valve; brachial valve aspinose.

Pedicle valve with a convex visceral disc, steep flanks and an incurved umbo extending
4 mm. behind hinge-line on a valve 19 mm. wide and 21 mm. long; sinus shallow at
umbo and running as a narrow obsolete furrow towards front of valve.

Brachial valve strongly concave, with deepest portion near the umbo; posterior bor-
ders slightly convex and slope steeply into the concavity; median fold commences at
the umbo and becomes broader but less well defined anteriorly.

Internal. Pedicle valve. Muscle field situated a short distance from beak; adductor
scars elevated, vary from narrow rectangular ridges to broad platforms and in some
cases divided by a median groove; posterior margins faintly dendritic and slightly im-
pressed into thickened umbonal region; diductor scars divergent, triangular, slightly
impressed, with irregular anterior and posterior terminations; they extend a short
distance in front of adductor platform; diductor scars ornamented by radial striae in
some cases continuous with internal ornament of valve; marginal ridge originates near
inner portion of ears, extends around the trail, but may be almost totally obsolete;
internal surface of most valves marked by three types of ornament — sharp radial ridges
separated by narrow grooves around the muscle field; very small pits on a region extend-
ing from front of muscle field to the marginal ridge; obsolete longitudinal ornament
on the trail.

Brachial valve. Median septum arises from the rounded base of the cardinal process
and extends as a thin sometimes spinose ridge to mid-point of valve where it expands
into a high knob-like inflation; strength of posterior portion variable; adductor scars in

62

PALAEONTOLOGY, VOLUME 8

two distinct pairs; posterior scars rectangular, elevated on platforms at posterior end
of median septum and marked by an obsolete dendritic pattern; smaller anterior scars
slightly impressed and situated immediately behind knob-like tip of median septum;
brachial ridges arise from antero-lateral margins of posterior adductor scars and run a
short distance antero-laterally before making hook-like curves, enclosing kidney-shaped
brachial impressions; brachial ridges well defined and high on inner margin of curve
around brachial impressions, but elsewhere narrow; floor of valve smooth between
brachial impressions and anterior adductor scars; cardinal process bilobate internally,
trilobate externally and supported by broad rounded ridges; cardinal ridges diverge at
10°-15° from hinge and run across inner margins of auricles and around anterior margin
of valve; surface of visceral disc ornamented by fine pustules, and trail by occasional
longitudinal ridges.

Measurements (in mm.)

Pedicle valve:

Specimen

Number

Length

Width

Height

Muscle Field

Length

Width

F. 6708

22

20

9

F. 6709

16-5

20

7

F. 6717

19

22

7

F. 67176

20

10

13

F. 6717c

20

20

10

10

15

Brachial valve :

Specimen

Number

Length

Width

Length

of

Median septum

Distance between
external margins of
Brachial Ridges

F. 6701

15

21

7-5

15

F. 6704

15

18

8

14-5

F. 6825 b

18 est.

24

9

17-5

F. 6705

14 est.

18

8

13-5

The width is measured at the mid-point of each valve.

Remarks. Eomarginifera paradoxus (Campbell, 1957) from Babbinboon, N.S.W., is
distinguished from Eomarginifera tenuimontis by the presence of a ginglymus, weaker
brachial ridges and impressions and differences in ornament. Costellae are slightly finer
on E. paradoxus and have a density of 16-20 per 10 mm., compared with 14-15 per 10
mm. on E. tenuimontis, both measurements being made on the anterior portion of the
trail. Concentric ribs are much reduced in E. tenuimontis and only 6-8 are present on the
posterior third of the pedicle valve, compared with 13-17 on the same region of E.
paradoxus.

Eomarginifera Jongispinus (Sowerby) described by Muir-Wood (1928) from the Visean
of Great Britain has a weaker median septum, less well-defined brachial ridges and im-
pressions and a more globose pedicle valve.

Eomarginifera setosa (Phillips) described by Muir-Wood (1928) is characterized by

J. ROBERTS: LOWER CARBONIFEROUS FAUNA FROM NEW SOUTH WALES 63

a weaker posterior portion on the median septum and a stronger marginal ridge in the
brachial valve.

Eomarginifera derbiensis (Muir-Wood, 1928) is more convex than E. tenuimontis , but
because its internal structures are unknown a more detailed comparison cannot be made.
Both E. setosa and E. derbiensis occur in the Visean and Namurian of Great Britain.

The specific name is taken from the Latin tenuis — thin, and montia — ridge, and refers
to the thin or narrow brachial ridges between the brachial discs and the muscle field.

Occurrence. Eomarginifera tenuimontis occurs at L. 208 and L. 233 (the type locality)
Trevallyn and L. 53 Greenhills. Specimens closely resembling this species have been col-
lected from L. 215, L. 216, L. 217, and L. 86 Lewinsbrook (Roberts, 1963).

Material. F. 670 1-F. 6724, F. 6825 a-b. Holotype F. 6701 , paratypes F. 6708, F. 6709.

Family buxtoniidae Muir-Wood and Cooper 1960
Genus marginatia Muir-Wood and Cooper 1960

Type species. (By original designation) Productus fernglenensis Weller 1909, from the Fern Glen
Formation, Missouri.

Remarks. Only three features of this material do not conform with those of the type
species : ( 1 ) A single row of small erect spines occurs along the hinge on the pedicle valve.
(2) The ears are well differentiated from the visceral disc. (3) Adductor scars in the
brachial valve may be divided into two pairs and are not marked dendritically.

The remainder of the characters described below are closely comparable with those
of the type species. The external ornament on the brachial valve is strikingly similar to
that of Marginatia cf. M. burlingtonensis (Hall) and Marginatia sp. figured by Muir-
Wood and Cooper (1960, pi. 99, figs. 5, 13, respectively).

Marginatia patersonensis sp. nov.

Plate 10, figs. 1-5

Diagnosis. Pedicle valve transverse to subquadrate; shallow sinus extending from near
umbo to anterior margin of valve; hinge extended into blunt, flattened auricles; costae
with density of 12-14 per 10 mm. at anterior margin; rugae prominent on ears and up to
16 present on visceral disc; erect row of fine spines along hinge; brachial valve geni-
culate, with fold crossing trail and visceral disc; adductor scars smooth and possibly
divided into two pairs.

Description. External . Pedicle valve transverse to subquadrate; strongly convex on posterior
portion of valve and on trail; hinge widest part of valve and extended into blunt
flattened auricles; umbo pointed, slightly incurved over hinge and well differentiated
from flattened postero-lateral margins; flanks steep; shallow sinus extends from near
umbo to anterior margin of valve; costae twice as wide as separating sulci, and especially
well defined on trail; costae with density of 12-14 per 10 mm. at anterior margin of
valve, increasing mainly by intercalation; rugae pronounced on ears, and up to 16 rugae
ornament the visceral disc, forming a well-defined reticulate pattern ; rugae do not extend
on to trail; spines present in fine erect row along hinge, a group of two or three slightly
larger spines on postero-lateral margins, and a row of larger spines with circular bases

64

PALAEONTOLOGY, VOLUME 8

on anterior portion of trail; other small anteriorly pointing spines, arising from costae,
randomly distributed over visceral disc.

Brachial valve geniculate, with gently concave visceral disc; valve increases in
concavity at trail; postero-lateral margins sharply defined from visceral disc; umbo
depressed below level of hinge; fold crosses both visceral disc and trail; costae have same
density as those on pedicle valve; a variable number of rugae present on visceral disc
give rise to a marked reticulate ornament; rugae obscure costae on umbo and are
absent from trail; several small spine bases present on lateral margins; small pits occur
in positions corresponding with spines on body of opposite valve.

Internal. Pedicle valve. Muscle scars poorly defined and obscured by impressions of
external ornament; ginglymus apparently absent; marginal ridge arises at postero-lateral
extremities of valve and crosses the venter at anterior margin of visceral disc.

Brachial valve. Adductor scars may be divided into two pairs; inner scars elongate,
smooth, pointed posteriorly and broadly rounded anteriorly; second pair more obscure,
may occur on lateral margins of those described above and appears to be longitudinally
striate; median septum extends at least two-thirds length of visceral disc, being broadly
rounded at posterior of valve and becoming narrow and blade-like for anterior three-
quarters of its length; septum bears narrow median furrow on rounded posterior por-
tion and a small antron in front of cardinal process; cardinal process bilobate anteriorly;
posterior surface not observed; lateral ridges well defined and slightly divergent from
hinge-line; brachial ridges arise on antero-lateral margins of lateral adductor scars, but
are only well defined when surrounding brachial discs; brachial discs large smooth
ovoid impressions, their anterior margins being level with tip of median septum;
postero-lateral margins of valve ornamented with fine pustules; remainder of internal
surface more regularly and coarsely spinose in front of muscle scars and brachial im-
pressions; granulated radial ribs present between large spine bases on anterior of shell;
on some specimens internal ornament obscured by impressions of external ornament.

Measurements (in mm.)

Pedicle valve:

Specimen

Width at mid-

Number

point

Length

F. 6813

27

22

F. 8030

25

20

F. 6826a

22

18

Brachial valve :

Specimen

Number

Width at mid-
point

Length

F. 7196

31-5

23

F. 6815

17

15

F. 6817

20

15est.

F. 6825c

28

21-5

Remarks. Differences with the type species have been outlined above.

Marginatia burlingtonensis (Hall), described by Weller (1914) from the Burlington

J. ROBERTS: LOWER CARBONIFEROUS FAUNA FROM NEW SOUTH WALES 65

Limestone of Iowa, Missouri, and Illinois, is larger, more inflated, has a longer trail
on the pedicle valve, and lacks a row of spines along the hinge of the pedicle valve.
M. bwlingtonensis has a similar style of ornament, and one notable similarity is its
possession of well-defined auricles on the pedicle valve; the pedicle valve has a sulcus
comparable with that on M. patersonensis. Marginalia cf. burlingtonensis (Hall) from
Lake Valley, New Mexico (Muir-Wood and Cooper, 1960), is distinguished from this
species by a coarse costate ornament on the pedicle valve, a more globose shape, the
absence of a sulcus and well-defined auricles on the pedicle valve, and a weaker posterior
portion on the median septum. Marginalia sp., illustrated by Muir-Wood and Cooper
(1960), also from New Mexico, is larger than this species, has a finer and more reticulate
ornament, dendritic adductor scars in the brachial valve, and lacks ears on the pedicle
valve.

This species is named after the Paterson River, near Gresford, N.S.W.

Occurrence. Marginalia patersonensis has been collected from L. 86, L. 215, and L. 217
Lewinsbrook, L. 50 Gresford Quarry, L. 53 Greenhills, L. 204 and L. 206 Lewinsbrook
Syncline (Roberts, 1961), and L. 208 and L. 233 Trevallyn (the type locality).

Material. F. 6812-F. 6826. F. 7196-F. 7200. F. 8030-F. 8032. Holotype F. 6813, paratype F. 6815.

Family athyridae Davidson 1884
Genus cleiothyridina Buckman 1906

Type species (By original designation) Spirifer deroysii L’Eveille 1833, from Tournai, Belgium.

Remarks. From recent unpublished work mentioned by Sanders (1958) it appears that
the type species of Cleiothyridina requires revision to clarify the taxonomic status of the
genus. At present there is confusion between Cleiothyridina and Actinoconchus M’Coy.
Actinoconchus, as recently interpreted by Sarycheva (1960) after Davidson (1857), has
its concentric lamellose ornament crossed by radial striae. According to Hall and Clarke
(1894) the striae were formed by fine tubular spines, but this cannot be confirmed until
the type species, A. paradoxus M’Coy, has been revised.

The anterior portions of Cleiothyridina australis Maxwell, described below, have the
appearance of being radially costate, a state of affairs which could possibly suggest an
affinity with Actinoconchus M’Coy. However, a close examination of the external orna-
ment reveals the posterior half of the shell to be ornamented with concentric lamellae
bearing vqry small flattened spines and the anterior half to possess long flat radially
arranged overlapping spines. The overlapping of these spines first occurs in the region
where growth lamellae become crowded. When the spines are broken their bases pro-
duce a pseudocostate surface ornament.

Cleiothyridina australis Maxwell
Plate 10, figs. 11-16; text-figs. 3-4
Cleiothyridina australis Maxwell 1954, p. 43, pi. 5, figs. 5-6.

Description. External. Shell sub-equally biconvex, moderately transverse, two-thirds wider
than long, and elliptical to sub-elliptical; greatest width at mid-line; hinge-line slightly
curved and one-half to two-thirds width of shell; ornament of regular imbricating

66

PALAEONTOLOGY, VOLUME 8

concentric lamellae, anterior margins of which produced into an overlapping fringe
of flat spines; spines become coarser and more elongate anteriorly and project from
anterior and antero-lateral margins; on anterior half of shell spines regularly produced
into radial rows and when broken give the shell a costate appearance; elongate spines
generally confined to anterior half of pedicle valve and may extend a short distance
further posteriorly on brachial valve; twenty to thirty concentric lamellae and twelve to
fifteen spine bases per 5 mm. on mid-anterior portion of brachial valve.

A

text-fig. 3. Cleiothyridina australis Maxwell. Inter-
nal view of the pedicle valve showing pallial mark-
ings. Vascula genitalia black, vascula media ruled
(a. F. 6686, b. F. 4815, both x2).

text-fig. 4. Cleiothyridina australis Maxwell.
Showing the trunks of vascula genitalia in the
brachial valve (F. 6684, x 2).

Pedicle valve. Beak moderately incurved and containing a sub-circular foramen;
umbo slightly elevated and shoulders slope evenly to lateral margins; median sinus faint
near umbo, widens into a broader but more shallow depression towards front of valve
and may become obsolete anteriorly; delthyrium broad, triangular, having lateral mar-
gins convex outwards; delthyrium filled by umbo of brachial valve.

Brachial valve slightly less convex than pedicle valve, with weaker incurved umbo;
fold faintly developed; lateral margins flattened.

Internal. Pedicle valve. Teeth supported by blunt peg-like dental lamellae emerging
from near antero-lateral margins of pedicle cavity; dental lamellae run a short distance
laterally and do not extend around muscle field; pedicle cavity rounded to V-shaped,
expands anteriorly and narrows posteriorly; floor of cavity divided by three longitudinal
grooves; shell thickened around front of pedicle cavity and posterior portions of muscle

J. ROBERTS: LOWER CARBONIFEROUS FAUNA FROM NEW SOUTH WALES 67

field; muscle field deeply impressed posteriorly and divided by an expanding myophragm
which usually runs length of muscle field; myophragm completely obsolete in some
specimens; adductor scars triangular to rectangular, occurring at posterior end of muscle
field; diductor scars large, generally less well impressed, and rectangular to triangular
in shape; pointed posterior margins of diductor scars situated slightly behind mid-point
of adductor scars; their broadly rounded to almost straight anterior margins usually
well in front of adductor scars, up to eight radiating blade-like trunks of vascula geni-
talia diverge from thickened area on postero-lateral margins of muscle field and become
obsolete in mid-portions of valve; two trunks of vascula media draped from lateral
margins of adductor scars over diductor scars and then run towards anterior margin of
valve; faint genital pits present on thickened areas on either side of muscle field; the
pattern of pallial markings is shown in text-fig. 3.

Brachial valve. Hinge-plate apically perforated, triangular, and linking the crural
plates; crurae widely divergent, well rounded, concave outwards, and highest at their
anterior extremities; sockets trough-like, rounded, and widely divergent; median sep-
tum commences from beneath hinge-plate as a thick broadly rounded ridge bearing a
median carina, weakens abruptly a short distance behind its mid-length and continues
as a sharp ridge to mid-point of valve; adductor scars elongate, deeply impressed and
bluntly rounded posteriorly, broaden slightly anteriorly and terminate obscurely near
end of median septum; each scar divided by a divergent longitudinal line, the inner
portion of scar being more deeply impressed than the outermost division; trunks of
vascula genitalia diverge from postero-lateral margins of muscle field in a faint radial
pattern and rapidly become obsolete towards margins of valve (text-fig. 4).

Measurements (in mm.)

Pedicle valve :

Muscle Field

Specimen

Width at

Number

Length

Width

Length

Anterior End

F. 4815

16

23

9

7-5

F. 6686

21

28-5

11-5

8

F. 6688

21

30

F. 6698

13

19

7

7

F. 6687

21

27-5

Brachial valve:

Specimen

Length Median

Muscle Field

Number

Length

Width

Septum

Length

Width

F. 6684

19

26

12-5

8 approx.

4

F. 6689

21

26

Remarks. This material differs slightly from the Mt. Morgan specimens described by
Maxwell (1954) in its finer concentric lamellose ornament, a more transverse shape,
more deeply impressed muscle scars in the pedicle valve and a stronger median septum
in the brachial valve.

Maxwell (1954) discusses the relationship of C. australis with overseas species.

68

PALAEONTOLOGY, VOLUME 8

Occurrence. Cleiothyridina australis Maxwell is known from the Schizophoria Zone of the
Neils Creek Clastics, Mt. Morgan district, Queensland, L. 53 Greenhills and L. 204
Lewinsbrook Syncline (Roberts, 1961), and L. 233, L. 270, L. 208 Trevallyn.

Material. F. 4813-F. 4818. F. 6684-F. 6700.

Family spiriferidae King 1846
Genus kitakamithyris Minato 1951

Type species. (By original designation) Kitakamithyris tyoanjiensis (Minato), 1951, from the Hikoroiti
series, North East Flonsyu, Japan.

Remarks. On present knowledge it is difficult to distinguish between the genera Kitaka-
mithyris and Phricodothyris George. This problem has been dealt with at length by
Maxwell (1961) and has been briefly mentioned by Campbell (1961).

Maxwell considered that Kitakamithyris Minato could be distinguished from Phrico-
dothyris George by the possession of larger radially arranged spine bases, a lack of inter-
spinous pustules and the presence of dental lamellae and a long median septum in the
pedicle valve. However, he doubted (p. 98) that minor details of the surface ornament
could be successfully used to distinguish between phricodothyroid genera, because of
the results of George’s (1932) work on the British phricodothyroids. His interpretation
is therefore based on the fact that all of the species so far described from Eastern
Australia and Japan possess well-defined dental lamellae and a median septum in the
pedicle valve. In contrast, the British species of Phricodothyris show great variability of
internal structures.

Campbell (1955, 1961) placed species from N.S.W. in the genus Phricodothyris, but
(1961, p. 437) noted that ‘the invariable development of strong dental lamellae, ventral
adminicula, and median septa, may yet require the removal of the New South Wales
species from Phricodothyris and their inclusion in Kitakamithyris ’. He also maintained
that revision of the European phricodothyroids is required before the two genera can
be successfully distinguished.

Until conclusive evidence is forthcoming the author accepts Maxwell’s interpretation
and assigns this material to Kitakamithyris.

The spines on the specimens of Kitakamithyris triseptata (Campbell) described below
consist of a pair of divided tubes which are incompletely closed anteriorly. The dividing
split between the two tubes is clearly visible on the surface facing the shell. Transverse
barbs are also present. With the exception of the last mentioned feature the spines are
identical with those described by Campbell (1961) on Phricodothyris immensa Campbell.
This type of spine is different from the completely closed tube-like structure described
by George (1932) for the British species of Phricodothyris, and from those of the type
species of Kitakamithyris figured by Minato (1952, pi. 9, figs. 4c, 4 d) which also appear
to be completely closed.

Kitakamithyris triseptata (Campbell)

Plate 11, figs. 1-3

Phricodothyris triseptata Campbell, 1955, pp. 379-380, pi. 18, figs. 10-15.

Description. Campbell’s original description can be supplemented by the following
observations:

J. ROBERTS: LOWER CARBONIFEROUS FAUNA FROM NEW SOUTH WALES 69

External. Pedicle valve. Ornament of six to seven concentric lamellae per 10 mm. and
five to six biramous spine bases per 5 mm., measured on median anterior portion of
valve; faint radial lirae in line with spine bases; spines barbed, tubular and biramous,
divided down their mid-length by a narrow furrow and incompletely closed anteriorly;
base of spine broad; spines sharply pointed distally, up to 5 mm. long at anterior margin
of valve; interarea strongly concave, 2-5 mm. high on a shell 23-5 mm. wide and 16 mm.
long, and ornamented with fine vertical striae; delthyrial angle approximately 60°.

Internal: Pedical valve. Diductor scars smooth, commence at tip of umbo, well
defined only where surrounded by adminicula and become poorly impressed and
obsolete anteriorly; adductor scars narrow and not extending posteriorly as far as the
diductor scars.

Brachial valve. Adductor scars shallowly impressed, rectangular to quadrangular,
expanding slightly anteriorly and extending almost the entire length of median ridge;
cardinal process approximately 15 small vertical lamellar platelets.

Remarks. Differences between the Trevallyn specimens and those from Babbinboon,
N.S.W., described by Campbell (1955), are of a minor nature. The median ridge in the
brachial valve is longer and the angle of divergence of the adminicula smaller in the
Trevallyn specimens.

Occurrence. Kitakamithyris triseptata (Campbell) has been collected from L. 35 Babbin-
boon (the type locality), L. 53 Greenhills and L. 204 Lewinsbrook Syncline (Roberts,
1961) and L. 208 and L. 233 Trevallyn.

Material. F. 6613-F. 6626.

Family dielasmatidae Schuchert and Le Vene 1929
Genus dielasma King 1859

Type species. (By original designation) Terebratulites elongatus Schlotheim, 1816, from Possneck,
Thuringia, Germany.

Dielasma picketti sp. nov,

Plate 12, figs. 1-3

Diagnosis. Shell slightly longer than wide; commissure gently sinuate; cardinal margin
terebratulid; pedicle valve with sub-erect umbo; foramen mesothyrid to permesothyrid;
small disjunct deltidial plates at apex of delthyrium; palintrope gently concave; pedicle
interior with large curving teeth; hinge-plate extending half the length of brachial valve.

Description. External. Shell equally biconvex, slightly longer than wide, terebratuliform
in shape, lacking a fold and sinus and with gently sinuate commissure; ornament re-
stricted to irregular concentric growth lines; punctae evenly distributed, having a density
of approximately 300 per square mm. on internal surface of shell.

Pedicle valve with sub-erect umbo perforated by a mesothyrid to permesothyrid
foramen; pedicle collar well defined; cardinal margin terebratulid; small disjunct
deltidial plates restricted to apex of delthyrium; palintropes gently convex, extending
some distance laterally and separated from body of shell by strong beak ridges.

Brachial valve. Area of greatest convexity towards posterior of valve; umbo pointed
but not incurved.

70

PALAEONTOLOGY, VOLUME 8

Internal. Pedicle valve. Teeth large, curved, antero-laterally directed and supported
on strong dental lamellae; dental lamellae broadest at floor of valve and extending a
short distance anteriorly as small ridges; muscle scars not observed.

Brachial valve. Hinge-plate resting medially on floor, forming a shallow spoon-like
structure, and extending almost to mid-point of valve; hinge-plate joins with socket
plates and bears triangular divergent crural bases mid-way between socket plates and
floor of the valve; adductor scars in two pairs; inner scars elongate, rectangular and
slightly divergent anteriorly; anterior margins of adductor scars grade into trunks of
vascula myaria; posterior margins not observed; outer adductor scars rectangular but
less well defined; thick divergent vascula myaria trunks arise from anterior margins of
inner adductor scars and extend towards the anterior margins of the valve; cardinal
process not observed.

Measurements (in mm.)

Pedicle valve:

Specimen

Number

Length

Width

F. 6669

18

15

F. 6670

17-5

14

Brachial valve :

Specimen

Number

Length

Width

F. 6669

16

15

F. 6673

15-5

13

F. 6674

16-5

13

Remarks. The species Die/asma saccuhtm var. hastata (Sowerby) described by Dun
(1902) from Wallaroo Hill, Clarencetown, N.S.W., is of doubtful validity. Insufficient
detail is available for a comparison of this species with D. picketti. The only comparable
overseas form appears to be D. avellana (de Koninck) from the Visean of the Moscow
Basin (Sarycheva and Sokolskaya, 1952). It has a similar external shape but cannot be

EXPLANATION OF PLATE 11

Figs. 1-3. Kitakamithyris triseptatci (Campbell) la, F. 6613. Rubber cast of pedicle valve exterior
showing spinose ornament, X 1-5. 1 b. F. 6613. Internal mould of pedicle valve, x2. 2, F. 6617.
Internal mould of brachial valve, X 2. 3, F. 6620a. Rubber cast of pedicle valve exterior, x 1-5.

Figs. 4-9. ‘ Camarotoechia' sp. B, 4, F. 6892a. Internal mould of brachial valve, X 2. 5, F. 6898. Rubber
cast of brachial valve exterior, X 2. 6, F. 6884. Internal mould of pedicle valve, X 2. 7, F. 6894.
Internal mould of brachial valve and apical portion of pedicle valve, X 2. 8, F. 6888. Internal mould
of pedicle valve, x 2. 9, F. 6896. Rubber cast of brachial valve exterior, X 2.

Figs. 10-14. Werriea australis Campbell 10, F. 6796. Internal mould of apical portion of pedicle valve;
note the spondylium and septum. (L. 53 Greenhills), X 2-5. 11 a, F. 6793. Rubber cast of cardinal
process and cardinal area, X 1. 1 lb, F. 6793. Rubber cast of brachial valve interior showing the
muscle field and cardinal process, X 1. 12, F. 6789. Internal mould of pedicle valve; note the short
septum and spondylium, x 1. 13, F. 6792. Rubber cast of brachial valve exterior, X 1. 14, F. 6797.
Internal mould of brachial valve. (L. 53 Greenhills), X 1.

Palaeontology, Vol. 8

PLATE 11

ROBERTS, Australian Carboniferous fauna

J. ROBERTS: LOWER CARBONIFEROUS FAUNA FROM NEW SOUTH WALES 71

compared in detail because its internal structures are unknown. D. picketti is named after
Dr. J. W. Pickett.

Occurrence. This species has been collected from L. 215 Lewinsbrook, L. 208 and L. 233
(the type locality) Trevallyn and L. 53 Greenhills.

Material. F. 6669-F. 6683. Holotype F. 6669, paratype F. 6670.

Family camarotoechiidae Schuchert and Le Vene 1929
Genus camarotoechia Hall and Clarke 1893

Type species. (By original designation) Atrypa congregata Conrad, 1841, from the Skaneateles Forma-
tion of New York.

Remarks. As a result of the work of Sartenaer (1961) the concept of Camarotoechia has
been restricted and the genus is now confined to a small number of Devonian forms.
It should not strictly be applied to the species described below but until the Carboni-
ferous rhynchonelloid genera have been revised this material is best referred to ‘ Camaro-
toechia’.

‘ Camarotoechia ’ sp. B
Plate 1 1 , figs. 4-9

Rhynchonella pleurodon (Phillips), Dun, p. 82, pi. 23, figs. 10-1 1.

Description. External. Shell small to moderate size for the genus, globular to sub-
triangular and subequally biconvex; slightly wider than long, with well-rounded anterior
and antero-lateral margins ; five acutely angular plications on each lateral slope increasing
in size away from the umbo; commissure serrated and uniplicate.

Pedicle valve noticeably triangular, with narrow pointed umbo extending a short
distance behind the hinge; floor of sinus ornamented by four to five plicae; towards the
anterior margin the sinus slopes steeply to commissure, extending as a tongue-like pro-
jection into the fold of the brachial valve; no details of foramen or delthyrium observed.

Brachial valve globular to rounded and more inflated than pedicle valve; median fold
weak, commences near mid-point of valve and ornamented by usually four or sometimes
five plicae; lateral slopes moderately convex and extending evenly to lateral margins;
beak incurved.

Internal. Pedicle valve. Teeth broad, well defined and supported by short divergent
dental lamellae; muscle scars not observed.

Brachial valve. Median septum slender, extends for approximately one-third the length
of the valve and supports a small V-shaped septalium; apical hinge-plate divided, tri-
angular, and bounds inner surfaces of sockets with its concave postero-lateral margins;
sockets broad, divergent and strongly serrated.

Remarks. ‘'Camarotoechia' sp. A described by Roberts (1963) from Lewinsbrook,
N.S.W., is smaller, less transverse and has more broadly rounded plicae.

I nsufficient material is available for a detailed comparison to be made with the Western
Australian Carboniferous ‘ Camarotoechia ' species described by Veevers (1959).

Occurrence. This species has been collected from L. 208 and L. 233 Trevallyn, L. 53
Greenhills, L. 204 Lewinsbrook Syncline and also occurs at Clarencetown, N.S.W.

Material. F. 688 5-F. 6899.

72

PALAEONTOLOGY, VOLUME 8

Family stenoscismatidae Muir-Wood 1955
Genus stenoscisma Conrad 1839

Type species. (By original designation) Terebratula schlotheimii von Buch, 1834, from the Permian of
Germany.

Remarks. Carboniferous species belonging to this genus have previously been
assigned by many workers to Camarophoria King, a junior objective synonym of
Stenoscisma.

Sarycheva and Sokolskaya (1952) divided Camarophoria into two sub-genera,
Camarophoria sensu-stricto and Levicamera Grabau, but Cooper (1956) showed that
Levicamera Grabau was invalid because the designated type species remains undescribed.
Cooper erected a new genus Psilocamara to embrace forms previously referred to
Levicamera.

This material is distinguished from Psilocamara Cooper (1956) by the possession of
a plicate sinus on the pedicle valve.

Camarophorinella Licharew (1936), another sub-genus of the junior synonym Camaro-
phoria, is similarly invalid because the type species has not been described. Camaro-
phorinella was defined on the absence of intercamarophorial plates and the nature of the
hinge-plate. The intercamarophorial plates are independent of the median septum, rest
against the base of the tornydium and wedge into the hinge-plate. They have only been
observed in some Russian species, for example S. ( Camarophoria ) superstes (Verneuil),
and have not been described from the type species of either Camarophoria or Stenoscisma.
The absence of these plates, therefore, cannot be used as a diagnostic feature in the
designation of a new sub-genus.

The term ‘tornydium’, proposed by Cooper (1956) for the brachial plate in Steno-
scisma previously described as a cruralium, appears to be synonymous with the term
‘camarophorium’ suggested by Kozlowski (1929) for the boat-shaped plate in the
brachial valve of Camarophoria. The use of the term tornydium is preferred to that of
camarophorium because the latter suggests an affinity with the genus Camarophoria.

EXPLANATION OF PLATE 12

Figs. 1-3. Dielasma pickett i sp. nov. 1, F. 6670. Internal mould of pedicle valve, X 2. 2a, F. 6669. Rubber
cast of brachial valve exterior and cardinal region of the pedicle valve; holotype, x2. 2b, F. 6669.
Internal mould of the same shell ; note the hinge plate and crural bases in the brachial valve, X 2.
3, F. 6673. Internal mould of brachial valve. (L. 53 Greenhills), X 2.

Figs. 4-6. Diodontopteria delicata sp. nov. 4a, F. 6805. Rubber cast of exterior of right valve; holotype,
x2. 4b, F. 6805. Internal mould of right valve showing the anterior tooth; holotype, X2. 4c,
F. 6805. Rubber cast of interior of same valve; holotype, X 2. 5, F. 6801b. Internal mould of left
valve. (L. 53 Greenhills), x2. 6, F. 6801a. Internal mould of left valve. (L. 53 Greenhills), X 2.

Figs. 7-8. Prolecanites sp. 1, F. 5942. Internal mould of part of one whorl, X 1 . 8, F. 5943. Internal mould
of the conch, x 1.

Figs. 9-10. ? Girtypecten sp. 9a, F. 481 1. Rubber cast of left valve exterior, x 2. 9b, F. 481 1. Internal
mould of the same valve, X2. 10, F. 4810. Rubber cast of left valve exterior, x2.

Figs. 11-13. Pernopecten trevallynensis sp. nov. 11, F. 6784. Rubber cast of left valve exterior, X 2.
12, F. 6786. Rubber cast of right valve exterior; paratype, X 2. 13a, F. 6783. Internal mould of left
valve; holotype, X 2. 13Z>, F. 6783. Rubber cast of left valve exterior; holotype, X 2.

Palaeontology , Vol. 8

PLATE 12

ROBERTS, Australian Carboniferous fauna

J. ROBERTS: LOWER CARBONIFEROUS FAUNA FROM NEW SOUTH WALES 73

Stenoscisma laevis sp. nov.

Plate 13, figs. 1-4; text-fig. 5

Diagnosis. Pedicle valve rostrate to triangular; median sinus quadrangular, outlined by
two low plicae and situated mainly on anterior of valve; floor of median sinus orna-
mented by from one to three weaker plicae; lateral slopes free of radial ornament but
marked by growth lamellae; spondylium narrow, trough-like, extending one-third length
of valve; vascula media branching in angular dichotomous style.

Description. External. Pedicle valve rostrate to tri-
angular, with rounded lateral margins; greatest
convexity a short distance in front of umbo; lateral
shoulders slope evenly from umbo and become
marginally flattened; umbo narrow and extending
well behind hinge-line; delthyrium open, triangular,
and set obliquely to hinge; both the hinge and the
palintropes curved; commissure regularly unipli-
cate; median sinus quadrangular in profile, extended
anteriorly into a curved lingual projection; sinus
outlined by two broadly rounded plicae and its floor
ornamented by from one to three weaker plicae;
sinus commences in front of umbo in positions
ranging from one-third to one-half length of valve; lateral slopes free from radial
ornament but marked by well-defined concentric growth lamellae.

Internal. Pedicle valve. Spondylium narrow, trough-like, linked posteriorly with den-
tal lamellae, extending one-third the length of valve and supported anteriorly by the
median septum; floor of spondylium concave posteriorly, slightly convex anteriorly;
median septum thickened at its base, tapers upwards towards the spondylium and pro-
trudes a short distance in front of the latter structure; teeth broad, bluntly rounded and
situated at base of delthyrium; vascula media in two sets on raised ridge-like thickenings
arising from the anterior extremity of median septum; vascula media branches in typical
angular dichotomous rhynchonelloid style (text-fig. 5), the faint terminal channels
reaching the margins of valve; small circular genital pits ornament floor of valve.

Measurements (in mm.)

Pedicle valve :

Specimen

Number

Length

Width

Spondylium

Length

F. 6900

1 1 est.

15

4

F. 6901 a

10

10

3-5

F. 69016

1 1

12

F. 6902

10

13

F. 6903

10 est.

14

Remarks. The brachial valve of Stenoscisma laevis has unfortunately not been found.
However, the widespread occurrence of this form warrants the designation of a new

text-fig. 5. Vascula media trunks in the
pedicle valve of Stenoscisma laevis sp.
nov. (F. 6900, X 3-5).

74

PALAEONTOLOGY, VOLUME 8

species. Stenoscisma donica (Rotai, 1931) resembles S. laevis in the shape of the pedicle
valve and morphology of the median sinus. S. donica occurs in the Dx or Beshevo Lime-
stone of the Donetz Basin which has been correlated with the D3 (Visean) of Great
Britain. S. laevis is readily distinguished from the British Carboniferous species 5.
? isorhynchia (M’Coy), S. crumena (Martin), and S. ? proava (Phillips), figured by
Davidson ( 1 860), by its smooth lateral slopes and extremely weakly plicate median sinus.
S. bisinuata (Rowley), as described by Weller (1914) from the Fern Glen Formation and
Burlington Limestone of the Mississippi Valley, U.S.A., has smooth lateral margins
comparable with those of 5. laevis. The latter species is distinguished by its more trans-
verse shape and stronger concentric ornament.

The specific name is taken from the Latin laevis — smooth, and refers to the smooth
lateral slopes of the shell.

Occurrence. Stenoscisina laevis has been collected from L. 233 Trevallyn (the type
locality), L. 86 Lewinsbrook, L. 53 Greenhills, and L. 234 Wiragulla.

Material. F. 6900-F. 6903. Holotype F. 6900.

Family prolecanitidae Hyatt 1884
Genus prolecanites Mojsisovics 1882

Type species. Goniatites mixolobus Mojsisovics, 1882 (under Article 70 b of the Code of Zoological
Nomenclature, non P. mixolobus Phillips, 1836) = P. mojsisovicsi Miller, 1938, Lower Carboniferous
of Germany.

Prolecanites sp.

Plate 12, figs 7-8; text-fig. 6

Description. Conch platyconic, moderately thin, discoidal and evolute; umbilicus wide,
exposes all inner whorls and measures 19 mm. in diameter on a conch 45 mm. in dia-

text-fig. 6. Suture line of Prolecanites sp. from Trevallyn. x4.

meter; whorls number four to five; whorls slightly overlapping, with outer whorl a little
more embracing than inner whorls; volutions compressed and flattened laterally; whorl
profile more or less rectangular, with a gently rounded ventral surface and a slight dorsal
expansion; umbilical shoulders moderately rounded; whorl flanks almost flat away from
dorsal and ventral margins; greatest whorl width occurs immediately above dorsal
margin; whorl height increases in outer volutions, the outer whorl being four times
higher than wide; where the latter is 15 mm. high the adjacent whorl is 6 mm. high;

J. ROBERTS: LOWER CARBONIFEROUS FAUNA FROM NEW SOUTH WALES 75

aperture not observed, but body chamber at least one-third the length of the last
volution.

Conch ornamented by faint biconvex ribs, the great majority being restricted to dorsal
half of the volution; some stronger ribs run almost to the ventral margin.

At the adult stage the suture contains three sub-equal lanceolate lateral lobes and a
smaller curved fourth lateral lobe (see text-fig. 6); ventral lobe obscure in specimens
examined, but ventral portion of first lateral saddle runs as though ventral lobe con-
stricted; second lateral lobe largest, slightly asymmetrical, having a ventrally inflected
point; third lateral lobe similarly shaped but with no ventral inflection; fourth lateral
lobe narrow, sharply pointed, curved slightly dorsally; first lateral saddle high and very
slightly asymmetrical; third lateral saddle broad and regularly rounded; fourth lateral
saddle relatively large and U-shaped.

Remarks. Compared with Prolecanites sp. most European species have a more rounded
or blunt fourth lateral lobe and less lanceolate lateral lobes — viz. P. postappkmatus
Kullmann (1963) from the Cantabrian Mountains of Northern Spain. The American
species P. americanus Miller and Garner (1953) and P. monroensis (Worthen) have
sutures with pointed fourth lateral lobes. P. americanus , from the Salem or Ste Gene-
vieve Limestone of Illinois, has a similarly shaped fourth lateral lobe; the remaining
three lateral lobes have comparable shapes but are unequal in size. P. americanus has
a more rounded and much wider whorl profile.

P. monroensis , described by Collinson (1955) from the Ste Genevieve Limestone, has
a suture with three sub-equal lanceolate lateral lobes and a more or less pointed fourth
lateral lobe. It is distinguished from this species by its more rounded and sub-elliptical
whorl profile and smaller ratio of whorl height to whorl width.

Occurrence. Prolecanites sp. has been collected from L. 233 Trevallyn.

Material. F. 5942-F. 5943.

Larnily aviculopectinidae Etheridge Jr. emend. Newell 1937
Genus girtypecten Newell 1937

Type species. (By original designation) Aviculopecten sublaqueatus Girty, 1908, from the Permian of
West Texas.

Remarks. The species described below possesses a style of ornament which suggests that
it be referred to Girtypecten. However, when compared with the type species these speci-
mens have less well-defined secondary costae on the body of the valve and a less pro-
nounced and more irregularly developed concentric ornament. In addition, the marginal
spines originating from the costae are hollow in contrast with the solid rod-like spines
occurring on the type species.

J. M. Dickins (personal communication) suggests that this material is similar to
Pseudaviculopecten exactus (Hall) from the Hamilton Group, Lake Canandaigua, New
York. He also notes that if the Trevallyn species has a chevron ligament area, which is
characteristic of P. exactus, these two species could then possibly be included in a sepa-
rate genus. The lack of specimens prevents any further research into this problem.

Morphologically similar forms have been placed in the genus Pterinopecten by Hind
(1903) and Branson (1938). Pterinopecten, however, is characterized by a much denser
external ornament of radiating ribs.

76

PALAEONTOLOGY, VOLUME 8

? Girtypecten sp.

Plate 12, figs. 9-10

Aviculopecten tesselatus (Phillips), p. 21, pi. 15, figs. 8-9.

Description. Left valve. Shell acline to very slightly opisthocline, small, subquadrate,
unequally auriculate, with long straight hinge-line; ornament consists of narrow costae
and concentric growth lamellae; costae number 24-27; costae narrow and sharp at the
umbo, expand ventrally into broader well-rounded ridges approximately half as wide as
the separating sulci; small hollow spines, 1-1-5 mm. long, on ends of costae produce a
scalloped margin on the shell; occasional faint intercalating secondary costae present
between primary ribs; concentric ornament on the umbo consisting of distinct filae
with a density of 6 per 3 mm., measured from tip of umbo; ventrally the filae become
lameilose and more widely spaced, forming small spinose projections where they cross
the costae; anterior auricle small, convex, rectangular, and well defined by shallow non-
costate auricular sulcus; anterior auricle ornamented with concentric growth lamellae
and 2-3 strong secondary costae; posterior auricle flattened laterally and towards the
dorsal margin, and poorly differentiated from body of valve; postero-lateral margin of
posterior auricle concave and ornamented with growth lamellae and 5-6 secondary
costae; muscle scars and ligament pits not observed; right valve not collected.

Measurements (in mm.)

Specimen

Number

Length

Width

Length of
Hinge

F. 4811

20

16-3

18

F. 4810

16-5

13

17

Remarks. This species differs from Aviculopecten tesselatus Phillips (1853), from the
Carboniferous Limestone of Yorkshire, Derbyshire, and Ireland, in the possession of a
greater number of costae on the body of the shell, a less pronounced concentric orna-
ment, a shorter and less acute posterior auricle, a less convex anterior auricle and a
stronger marginal ridge on the dorsal portion of the posterior auricle. The morphology
of A. tesselatus was clarified by Hind (1903), who restudied and refigured the holotype
(pi. 9, fig. 10).

Aviculopecten nobilis de Koninck, described by Hind (1903) from the Carboniferous
Limestone of Yorkshire, Northumberland, Derbyshire, and Ireland, is distinguished from
this species by its more numerous concentric ribs, weaker posterior auricle and the lack
of spines around the ventral margin of the shell.

EXPLANATION OF PLATE 13

Figs. 1-4. Stenoscisma laevis sp. nov., 1, F. 6901a. Internal mould of pedicle valve, x2. 2, F. 6901b.
Internal mould of pedicle valve, X 2. 3, F. 6902. Rubber cast of pedicle valve exterior, X 2. 4a,
F. 6900. Internal mould of pedicle valve; note the spondylium and pallial trunks; holotype, x2.
4b, F. 6900. Rubber cast of exterior of same valve; holotype, X 2.

Figs. 5-7. Fenestella allynensis sp. nov. 5, F. 6907. Rubber cast of reverse side of colony ; paratype, X 6.
6 a, F. 6904. Rubber cast of obverse side of colony ; holotype, X 4. 6b, F. 6904. Rubber cast of obverse
side of colony; holotype, X 10. 7, F. 6905. Rubber cast of obverse side of colony; branches flattened;
paratype, X 12.

Palaeontology, Vol. 8

PLATE 13

ROBERTS, Australian Carboniferous fauna

J. ROBERTS: LOWER CARBONIFEROUS FAUNA FROM NEW SOUTH WALES 77

Pterinopecten sedaliensis Miller, described by Branson (1938), from the Chouteau
Limestone, Missouri, has a more elongate posterior auricle and a denser costate
ornament.

Occurrence. ? Girtypecten sp. occurs at L. 233 Trevallyn, L. 53 Greenhills, L. 86 Lewins-
brook and has also been recorded by Etheridge and Dun (1906) from a locality at or
near L. 210 Toryburn.

Material. F. 4810-F. 4811.

Genus pernopecten Winchell 1865

Type species. (By original designation) Aviculopecten limaeformis White and Whitfield, 1862, from the
Burlington Formation, Iowa.

Pernopecten trevcillynensis sp. nov.

Plate 12, figs. 11-13

Diagnosis. Shell suborbicular; small triangular auricles of left valve only slightly above
hinge; ornament of broad obsolete radial folds crossed by closely spaced concentric
growth lamellae; ornament weaker on right valve.

Description. External. Shell higher than wide, suborbicular and acline to slightly proso-
cline.

Left valve. Auricles triangular, with sharp tips, and separated from body of valve by
narrow linear sulci; auricles project only a short distance above hinge-line; a flattened
area on either side of umbo differentiated from body of valve by two obsolete sinuses,
the posterior sinus being the most strongly developed; umbo bluntly pointed, with an
umbonal angle of 90°-95°; ventral and anterior margins, including the portion below
the anterior auricle, convex and well rounded; slightly concave posterior margin present
immediately below posterior auricle; ornament of closely spaced concentric growth
lamellae crossed by obsolete radial folds in the shell; twenty to thirty concentric lamellae
per 3 mm. near ventral margin of shell.

Right valve. Auricles subequal, with straight dorsal margin, separated from the body
by linear sinuses, depressed below level of the valve, and ornamented with growth
lamellae; younger portion of anterior auricle with well-developed byssal notch, and
curved lamellae near the umbo which become straight in later stages of growth; con-
centric lamellae weak to absent on body of the valve; valve flattened on either side of
umbo and characteristically smaller than left valve.

Internal. Left valve. Two rounded ridges bearing small tooth-like projections diverge
from umbo and run ventrally to anterior and posterior margins; obsolete radial orna-
ment on internal surface; no central resilifer observed.

Measurements (in mm.)

Specimen

Number

Length

Width

F. 6783

Left valve

24

26

F. 6786

Right valve

12

13-5

78

PALAEONTOLOGY, VOLUME 8

Remarks. Pernopecten trevallynensis differs noticeably from all species of Pernopecten
described by Newell (1937) in having smaller auricles on the left valve. This species is
named after the property ‘Trevallyn’.

Occurrence. P. trevallynensis has been collected from L. 233 Trevallyn (the type locality)
and L. 53 Greenhills.

Material. F. 6783-F. 6788. Holotype F. 6783, paratype F. 6786.

Genus diodontopteria La Rocque 1950

Type species. (By original designation) Diodontopteria ehlersi La Rocque, 1950, from the Detroit
River Group, Michigan.

Remarks. La Rocque (1950) noted that Diodontopteria was distinguished from Lepto-
desma Hall by the possession of anterior teeth; one on the right valve and two on the
left. Externally the two genera were indistinguishable.

He has not, however, considered the relationship between this genus and Leiopteria
Hall from which Hall (1884) described an anterior septum or tooth-like structure. The
reason for this was probably his acceptance of a statement by Williams and Breger
(1916), who suggested that the ridge observed by Hall was the posterior boundary of a
small anterior adductor scar. Until the type specimens of Leiopteria are studied in
detail the question of whether this is a tooth or a muscle scar will remain unanswered.
Williams and Breger (1916) and Caster (1930) have distinguished Leptodesma from
Leiopteria by its greater obliquity, more acute cardinal angle and the possession of a
rounded rather than an auriculate anterior auricle. These same distinctions apply to
Diodontopteria which is externally indistinguishable from Leptodesma.

Diodontopteria has previously been recorded from the Lower Devonian of North
America. To the author's knowledge this is the first record of the genus from rocks of
Carboniferous age.

Diodontopteria delicata sp. nov.

Plate 12, figs. 4-6

Diagnosis. Antero-dorsal extremity subalate; posterior auricle extends backwards for
two-thirds length of shell, having spinose dorsal margin; concentric ornament well
defined; large anterior tooth in right valve; two smaller teeth in left valve.

Description. External. Shell pterioid, moderately convex, having a large flattened well-
differentiated posterior auricle; both valves of almost equal convexity; posterior auricle
approximately two-thirds length of shell with a sinuous posterior border and a dorsal
margin extended into a long spine-like process; antero-dorsally shell produced into a
small rounded subalate extremity separated from the umbo by a shallow sulcus; umbo
pointed, prosogyral, projects a short distance over hinge-line and raised high above
posterior auricle; body of shell expands postero-ventrally, ending in a well-rounded
ventral margin; gently sloping posterior margin runs into the posterior auricle; steep
convex anterior margin ventral to the anterior auricle; a narrow amphidetic ligament
area present along hinge-line; fine concentric ridges ornamenting shell become more
widely spaced ventrally and tend to crowd as they swing in a curve from body of shell
on to posterior auricle.

J. ROBERTS: LOWER CARBONIFEROUS FAUNA FROM NEW SOUTH WALES 79

Internal. Right valve. Anterior tooth large, divergent from hinge-line and flanking
two shallow sockets; posterior socket well defined, pointed towards umbo and becomes
deeper posteriorly.

Left valve. Anterior teeth small, enclosing a single socket; posterior tooth long,
slender, slightly divergent from hinge and extending to mid-length of posterior auricle;
adductor scars not observed in either valve.

Measurements (in mm.)

Specimen

Number

Length

Height

Angle between hinge
and elongation

F. 6809

Left valve

18

10

25°

F. 6801a

Left valve

24

18

30°

F. 68017)

Left valve

22

16 est.

30°

F. 6805

Right valve

22

10

35°

Remarks. Diodontopteria delicata is distinguished from the type species, D. ehlersi La
Rocque, from the Lower Devonian (pre-Traverse) of Michigan, L.S.A., by the posses-
sion of a more acutely spinose posterior margin and a less marked obliquity. The specific
name is taken from the Latin delicata — delicate, and refers to the external ornament of
the shell.

Occurrence. This species has been collected from L, 233 Trevallyn (the type locality),
L. 53 Greenhills, and L. 235 Dungog.

Material. F. 6801-F. 6811. Holotype F. 6805 , paratvpe F. 6809.

Acknowledgements. The author wishes to thank the following workers for helpful advice received during
the preparation of this paper; Dr. K. S. W. Campbell, Australian National University, Canberra,
A.C.T. ; Dr. P. J. Coleman, University of Western Australia, Nedlands, W.A.; Dr. C. Collinson, State
Geological Survey of Illinois, Urbana, Illinois, U.S.A.; and Dr. J. M. Dickins, Bureau of Mineral
Resources, Canberra, A.C.T.

REFERENCES

branson, e. b. 1938. Stratigraphy and palaeontology of the Lower Mississippian of Missouri. Pt. I.
Univ. Mo. Stud. 13, No. 3, 1-205, pi. 1-20.

brown, d. a., Campbell, k. s. w., and Roberts, j. 1965. A Visean cephalopod fauna from New South
Wales. Palaeontology 7, 682-94.

Campbell, k. s. w., 1955. Phricodot hyris in New South Wales. Geol. Mag. 92, 374-84, pi. 18.

1957. A Lower Carboniferous brachiopod-coral fauna from New South Wales. J. Paleont. 31, 34-

98, pi. 11-17.

■ 1961. Carboniferous fossils from the Kuttung rocks of New South Wales. Palaeontology , 4,

428-74, pi. 53-63.

caster, k. e. 1930. Higher fossil faunas of the Upper Allegheny. Bull. Amer. Palaeont. 15, No. 58,
1-174, pi. 1-59.

collinson, c. w. 1955. Mississippian prolecanitid goniatites from Illinois and adjacent states. J.
Paleont. 29, 433-8, pi. 45.

• scott, a. j., and rexroad, c. b. 1962. Six charts showing biostratigraphic zones and correlations

based on conodonts from the Devonian and Mississippian rocks of the upper Mississippi Valley.
Circ. III. Geol. Surv., 328, 1-32.

condra, g. e. and elias, m. k. 1944. Study and revision of Archimedes (Hall). Spec. Pap. Geol. Soc.
Amer. 53, 1-243, pi. 1-41.

80 PALAEONTOLOGY, VOLUME 8

cooper, g. a. 1944. In Shinier, H. W., and Shrock, R. R., Index Fossils of North America. John Wiley
and Sons, New York.

1956. New Pennsylvanian brachiopods. J. Paleont. 30, 521-30, pi. 61.

crockford, J. 1947. Bryozoa from the Carboniferous of New South Wales and Queensland. Proc.
Linn. Soc. N.S.W. 72, 1-48, pi. 1-6.

davidson, t. 1857-1863. A monograph of the British Carboniferous Brachiopoda. Palaeontogr. Soc.
( Monogr .) part 5, 1-280, pi. 1-55.

diener, c. 1899. Anthracolithic fossils of Kashmir and Spiti. Palaeont. Indica. Ser. 15, 1, part 2,
1-95, pi. 1-8.

— 1915. The Anthracolithic faunae of Kashmir, Kanaur and Spiti. Palaeont. Indica. N.S. 5, 1-135,
pi. 1-11.

dun, w. s. 1902. Notes on some Carboniferous brachiopods from Clarencetown. Rec. Geol. Surv.
N.S.W. 7, 72-88, pi. 21-23.

etheridge, r. and dun, w. s. 1906. Carboniferous and Permian Invertebrata of New South Wales.
Vol. II. Pelecypoda. Part I. The Palaeopectens. Mem. Geol. Surv. N.S.W. Palaeontology 5, 1-39
pi. 1-16.

foord, a. H. and crick, g. c. 1897. Catalogue of the fossil Cephalopoda in the British Museum
(Natural History). Part 3. Containing the Bactritidae and part of the suborder Ammonoidea. 1-303.
British Museum (Nat. Hist.), London.

george, t. n. 1932. The British Carboniferous reticulate Spiriferidae. Quart. J. Geol. Soc. Loud. 88,
516-75, pi. 31-35.

hall, j. 1884. Natural History of New York, Palaeontology, 5, part I, Lamellibranchiata I. xviii-
268, pi. 1-92.

— and clarke, j. m. 1894. Natural History of New York. Palaeontology, 8. An introduction to the
study of the genera of Palaeozoic Brachiopoda, Part 2, 1-394, pi. 21-84.

hind, w. 1901-1905. British Carboniferous Lamellibranchiata. Palaeontogr. Soc. {Monogr.) 2, 1-223,
pi. 1-25.

International Code of Zoological Nomenclature adopted by the XV International Congress of Zoology.

1-176. International Trust for Zoological Nomenclature, London. 1961.
koenig, J. w. 1958. Fenestrate Bryozoa in the Chouteau Group of Central Missouri. J. Paleont. 32,
126-43, pi. 21-22.

kozlowski, r. 1929. Les brachiopodes gothlandiens de la Podolie Polonaise. Palaeont. Polon. 1, 1-254,
pi. 1-12.

kullman, j. 1963. Die goniatiten des Unterkarbons im Kantabrischen Gebirge (Nordspanien). II.
Palaontologie der U.O. Prolecanitina Miller & Furnish. Die altersstellung der faunen. N. Jb. Geol.
Paldont. Abh. 116, 269-324, pi. 17-20.

lane, n. g. 1963. A silicified Morrowan brachiopod faunule from the Bird Spring Formation, southern
Nevada. J. Paleont. 37, 379-92, pi. 43-45.

la rocque, a. 1950. Pre -Traverse Devonian pelecypods of Michigan. Contr. Mas. Paleont. Univ.
Mich. 7, 271-366, pi. 1-19.

licharew, b. 1936. The interior structure of Camarophoria King. Amer. J. Sci. 32, 55-69.
maxwell, w. g. h. 1954. Upper Palaeozoic formations in the Mt. Morgan district — faunas. Pap. Dep.
Geol. Univ. Queensland , 3, No. 5, 1-69, pi. 1-6.

1961. Lower Carboniferous brachiopod faunas from Old Cannindah, Queensland. J. Paleont.

35, 82-103, pi. 19-20.

miller, a. k. and garner, h. f. 1953. The goniatite genus Prolecanites in America. J. Paleont. 27,
814-16, pi. 86.

miller, t. g. 1961. Type specimens of the genus Fenestella from the Lower Carboniferous of Great
Britain. Palaeontology, 4, 221-42, pi. 24-27.

minato, m. 1951. Ont he Lower Carboniferous fossils of the Kitakami Massif, northeast Honsyu,
Japan. J. Fac. Sci. Hokkaido Univ. Ser. 4, 7, 355-82, pi. 1-5.

1952. A further note on the Lower Carboniferous fossils of the Kitakami Mountainland, north-
east Japan. Ibid., Ser. 4, 8, 136-74, pi. 1-11.

mojsisovics, e. 1882. DieCephalopodendermediterranenTriasprovinz. Kaiserl-Kdnigl. Geol. Reichsanst.
Abh. 10, 1-322, pi. 1-94.

J. ROBERTS: LOWER CARBONIFEROUS FAUNA FROM NEW SOUTH WALES 81

muir-wood, h. m. 1928. The British Carboniferous Producti, II. Productus (sensu stricto); semireti-
culatus and longispinus groups. Mem. Geol. Surv. U.K. Palaeontology, 3, 1-217, pi. 1-12.

— and cooper, g. a. I960. Morphology, classification and life habits of the Productoidea (Brachio-
poda). Mem. Geol. Soc. Amer. 81, 1-447, pi. 1-135.

Newell, n. D. 1937. Late Palaeozoic pelecypods: Pectinacea. Geol. Surv. Kans. 10, 1-123, pi. 1-20.
paeckelmann, w. 1931. Die Fauna des deutschen Unterkarbons. II. Die brachiopoden des deutschen
Unterkarbons. 2. Die Productinae und Productus-nah I ichen Chonetinae. Preuss. Geol. Landesanst.
Abh., n.f. 136, 1-440, pi. 1-41.

Roberts, J. 1961. The geology of the Gresford district, N.S.W. J. Proc. Roy. Soc. N.S.W., 95, 77-91.

1963. A Lower Carboniferous fauna from Lewinsbrook, New South Wales. Ibid. ,97, 1-31, pi. 1-6.

rotai, a. 1931. Brachiopods and stratigraphy of the Lower Carboniferous of the Donetz Basin.

Trans. Centr. Geol. Prosp. Inst. U.R.S.S., 73, 35-144, pi. 1-10.
sandberger, g. and sandberger, f. 1850-1856. Die Versteinerungen des rheinischen Schichten-
systems in Nassau. 1-564, pi. 1-39. Kreidel and Niedner, Wiesbaden.
sanders, j. h. 1958. Brachiopoda and Pelecypoda in Easton, W. H., et cd. Mississippian fauna in
north-western Sonora Mexico. Smithson. Misc. Coll. 119, No. 3, 1-87, pi. 1-9.
sartenaer, p. 1961 . Etude nouvelle, en deux parties, du genere Camarotoechia Hall and Clarke, 1893.
Premiere partie: Atrypa congregata Conrad, espece-type (1). Bull. Inst. Sci. Nat. Belg. 37, No. 22,
1-11, pi. 1.

sarycheva, t. g. (editor) 1960. Osnovi pa/eontologii. Manual for palaeontologists and geologists
of U.S.S.R. Bryozoa, Brachiopoda. 1-343. Acad. Sci. U.S.S.R., Moscow.

and sokolskaya, a. n. 1952. Index of Palaeozoic brachiopods of the Moscow Basin. Trud.

Akad. Palaeont. inst. S.S.S.R. 38, 1-307, pi. 1-71. (French translation.)
sokolskaya, a. n. 1954. Strophomenidae of the Russian platform. Trud. Akad. nauk. Palaeont. inst.
S.S.S.R. 51, 1-187, pi. 1-18.

1963. In Sarycheva, T. G., Sokolskaya, A. N., Beznosova, G. A., and Maksimova, S. V. Brachio-
pods and Palaeogeography of the Kuznets Basin. Trud. Akad. nauk. Palaeont. inst. S.S.S.R. 95,
1-547, pi. 1-64.

thomas, g. a. 1958. The Permian Orthotetacea of Western Australia. Bull. Bur. Miner. Resour. Aust.
No. 39, 1-115, pi. 1-22.

ulrich, e. o. 1890. Palaeozoic Bryozoa. Geol. Surv. 111. 8, 283-668, pi. 29-78.

veevers, j. j. 1959. Devonian and Carboniferous brachiopods from north-western Australia. Bull.
Bur. Miner. Resour. Aust. No. 55, 1-42, pi. 1-4.

weller, s. 1914. The Mississippian brachiopods of the Mississippi Valley Basin. Monogr. Geol Surv.
III. 1, 1-506, pi. 1-83.

williams, H. s. and breger, c. l. 1916. The fauna of the Chapman Sandstone of Maine, including
descriptions of some related species from the Moose River Sandstone. Prof. Pap. U.S. Geol. Surv.
89, 1-347, pi. 1-27.

JOHN ROBERTS
Department of Geology,
University of Western Australia,
Nedlands, W.A.

Present Address:

Bureau of Mineral Resources,
Geology and Geophysics,
Childers St., Turner,

Canberra City, A.C.T.

Manuscript received 4 February 1964

WESTPHALIAN D MEGASPORES FROM THE
FOREST OF DEAN COALFIELD, ENGLAND

by EDWIN SPINNER

Abstract. A first description is given of Westphalian D dispersed megaspores in Britain. Among the seventeen
species recorded five are new: Setosisporites pilarus, Lcigenicula verrurugosa, L. perverrucala, L. ? verrucata, L.
irregularis. The species are assigned to nine genera as defined by Potonie and Kremp (1954). The genus Zonale-
sporites (Ibrahim) Potonie and Kremp is redefined to include Superbisporites , Rotatisporites, and Racliatisporites
as defined by Potonie and Kremp (1954). A list of additional macro-plant species from the Coal Measures con-
cerned is also given.

During the last thirty years several studies have been carried out on Carboniferous
megaspores in Europe and North America. European workers, particularly Zerndt
(1930-38) and Dijkstra (1946-56), have demonstrated the value of megaspores in the
broad zonation and correlation of coal basins and in the correlation of individual seams
within a basin. Similar studies on the Mississippian and Pennsylvanian of North
America were carried out by Schopf (1938), Arnold (1950), and Winslow (1959). These
studies have resulted in the recognition and description of a large number of megaspore
species from rocks ranging in age from Dinantian to Stephanian.

In Britain the study of dispersed Carboniferous megaspores has been largely neglected
since the early work of Bennie and Kidston (1886) on megaspores from the Carboni-
ferous of Scotland. A number of papers by Slater, Evans, and Eddy (1930-32) described
megaspores in thin sections from some coal seams from the Yorkshire coalfield and
indicated the use of these fossils in correlation between the Yorkshire and Lancashire
coalfields. Since this work was based on thin sections only, much of the detail of shape
and ornament of the spores could not be observed.

Techniques devised by Schulze (1855) and Zetzsche and Kalin (1932) for the extraction
of spores from unweathered coals were applied by Zerndt and Dijkstra and rapidly led
to greater possibilities for the taxonomic study of megaspores and their application to
stratigraphical correlation.

The present investigation has followed these techniques and constitutes a systematic
account of the megaspores and of their stratigraphic distribution in a British coalfield.
Thirteen coal seams and associated shales have been examined from the Forest of Dean.
All these samples yielded well-preserved megaspores. Approximately 1,000 megaspores
were examined in detail. Amongst these occur twelve previously known species, all of
which had not been recorded before from the Westphalian D of Britain. Eight new
species have also been recognized, five of which are described in the present paper. The
remaining new species will be described in the future, when more complete information
becomes available. The descriptions are based on an examination by means of trans-
mitted and reflected light.

GENERAL GEOLOGY OF THE COALFIELD

The Forest of Dean coalfield is a small, broadly triangular shaped area situated in
Gloucestershire between the Severn and Wye valleys. The coalfield is some thirty square

I Palaeontology, Vol. 8, Part 1, 1965, pp. 82-106, pi. 14-17.]

E. SPINNER: WESTPHALIAN MEGASPORES FROM THE FOREST OF DEAN 83

miles in extent and forms an outlier of high country surrounded by rocks of Lower
Carboniferous and Devonian ages. To the west and south lie the larger coalfields of
South Wales and Bristol/Somerset.

Within the Forest of Dean the Coal Measures are restricted to Upper Westphalian

WOORGREEN

a

O

o

or

o

CROW DELF
TWENTY INCH
LOWERY
STARKEY
ROCKEY
CHURCHWAY
NO COAL
BRAZILLY

z

<

z

I

<
a
a
3
i n

<

a

z>

O

a

O

YORKLEY

WHITTINGTON

>z

<

z

z

uJ

a

COLEFORD HIGH
DELF

TRENCHARD

F-

UJ

Z

O

M

E?

z>

2

UJ

y-

<

I-

r>

<

z

O

0
<
a

1

t-

z

<

text-fig. 1. Coal Measures succession in the Forest of Dean
(after L. R. Moore 1954, p. 127, fig. 7).

age and rest with marked unconformity upon Lower Carboniferous and Old Red
Sandstone. The Coal Measures are completely exposed and are approximately 2,000 to
2,300 feet in total thickness. The succession is generally subdivided into three lithological
formations (text-fig. 1). Since these formations have already been described in detail
by Trotter (1942) and Moore (1954), only a summary is given here. The lower or Tren-
chard formation, between 50 and 400 feet thick (Trotter, 1942, p. 28), consists of con-
glomerates, grits, sandstones, and shales. It contains a single coal seam, the Trenchard

84

PALAEONTOLOGY, VOLUME 8

seam, which is well developed in the south-western part of the coalfield, but deteriorates
northwards and splits into two leats, known as the Lower and Upper Trenchard seams.
These coals have not yet been proved in the eastern part of the coalfield. The succeeding
Pennant formation extends from the Coleford High Delf seam to the base of the Brazilly
seam and consists of 800 feet of massive felspathic sandstones with intervening thin
shale horizons containing coal seams. The highest subdivision or Supra-Pennant forma-
tion is some 1,100 feet thick and contains most of the workable coal seams found in the
Forest of Dean. Within this formation two further subdivisions have been recognized,
viz., a lower division from the Brazilly seam to the top of the Crow Delf seam, consisting
mainly of shales and thin sandstones, and an upper division containing massive sand-
stones and thin coals (i.e. including the Woorgreen coals).

Structurally, the Forest of Dean coalfield is represented by a north-south elongated
basin formed by small anticlinal and synclinal folds (Trotter 1942, pp. 3-8).

MACROPALAEONTOLOGICAL EVIDENCE FOR AGE OF COAL

MEASURES

The lowest recorded evidence is that from the roof shales of the Coleford High Delf
seam (Trotter 1942, p. 38) which included a number of non-marine lamellibranchs form-
ing an Anthraconauta tenuis-phillipsi assemblage. After a re-examination of the fauna,
Calver (in Welch, Trotter, et ah 1961, p. 90) considered that the horizon should be
placed in the Anthraconauta tenuis zone of the upper Coal Measures (as redefined by
Stubblefield and Trotter 1957, p. 3).

The first detailed examination of the plant macrofossils of the coalfield was carried
out by Arber (1912), who concluded that the assemblages obtained from the different
searns were practically the same. He assigned them to the ‘Upper Coal Measures’ which
may be broadly equivalent to high Westphalian. However, Crookall (1930, p. 225) after
a re-examination of Arber's material regarded only the Woorgreen coals as belonging
to the Upper Coal Measures, the remainder of the sequence being referred to the
Staffordian. In a later paper by Crookall (1955, table A, p. 2) the Upper Coal Measures
and the Staffordian were approximately equated with Westphalian D and upper West-
phalian C respectively.

Moore ( 1947, p. 291) pointed out that the floras of some of the coal seams in the Forest
of Dean were indicative of floral zone H of Dix (1934), i.e. Westphalian D. A list of plant
species collected from the Forest of Dean and considered to be diagnostic of West-
phalian D is given by Welch, Trotter, et al. (1961, p. 90).

More plant fossils were found during the collection of material for the present
investigation, in company with R. H. Wagner, who has kindly identified the species
quoted below. The fossils are grouped according to the colliery tips on which they
were found.

Northern United colliery tip (Coleford High Delf seam): Neuropteris ovata Hoff-
mann, Lobatopteris vestita (Lesquereux) Wagner, Pecopteris dentata Brongniart,
SphenopliyUum cf. emarginatum Brongniart, Asterophvllites equisetifonnis (Schlo-
theim), Lepidophloios laricinus Sternberg, Lepidophyllum sp.

Steam Mills colliery tip (Brazilly seam): Neuropteris flexuosa Sternberg, N. scheu-
chzeri Hoffmann, Odontopteris lindleyana Sternberg, Sphenopteris neuropteroides

E. SPINNER: WESTPHALIAN MEGASPORES FROM THE FOREST OF DEAN 85

(Boulay) ZeiWer, Lobatopteris vestita (Lesquereux) Wagner, Sphenophyllum sp. Lepido-
dendron sp. Lightmoor colliery tip (composite tip of seams from No Coal to Twenty
Inch): Neuropteris ovatci Hoffmann, Polymorphopteris polymorpha (Brongniart)
Wagner, Pecopteris unit a (Brongniart), P. hemitelioides Brongniart, Sphenophyllum
cf. cuneifolium Sternberg.

New Fancy Colliery tip (composite tip of seams from Churchway to Crow Delf):
Neuropteris flexuosci Sternberg, N. cf. flexuosa Sternberg, N. scheuchzeri Hoffmann,
Alethopteris ambigua Lesquereux pars D. White emend, Pseudomariopteris ribeyroni
(Zeiller) Danze-Corsin, Sphenopteris neuropteroides (Boulay), Lobatopteris vestita
(Lesquereux) Wagner, Pecopteris cisti Brongniart, Sphenophyllum emarginatum Brong-
niart, Annularia stellata (Schlotheim), Aster ophyllites equisetiformis (Schlotheim),
Catamites suckowi Brongniart, Lepidodendron cf. wortheni Lesquereux.

Wagner considers the evidence from the New Fancy colliery tip to be particularly
significant, since it contains elements not previously mentioned from the Forest of
Dean. The most important is Pseudomariopteris ribeyroni, a well-known Stephanian
element in Western European floras. However, Bell (1938) recorded this species from
strata of approximate Westphalian D age in Nova Scotia, Canada. Judging from the
present assemblage it also appears to occur in high Westphalian D rocks of the British
Isles. Alethopteris ambigua was originally figured from the New Fancy colliery as
Alethopteris davreuxi? Brongniart by Arber (1912, pi. 11, fig. 8). Lobatopteris vestita
has been usually recorded from Britain under the name of Pecopteris miltoni (Artis).
Although well known in North America, L. vestita has only been encountered spora-
dically in Western Europe, apart from the British Isles. Alethopteris ambigua Lesquereux
(= Alethopteris friedeli P. Bertrand) seems equally well represented in Europe and
North America.

The sum total of plants encountered in the Forest of Dean strongly suggest a West-
phalian D age for the measures from the Coleford High Delf seam upwards. Both
Neuropteris ovata and Lobatopteris vestita are considered to indicate at least West-
phalian D.

No macrofossils have yet been found in the Trenchard formation, the age of which
is uncertain. On the unpublished micropalaeontological evidence of Williams (1956) as
quoted by Butterworth and Millott (1960, p. 159), it appears that the Trenchard coal
could be referred to either the highest Westphalian C or the Westphalian D.

TECHNIQUES OF STUDY OF MICROFOSSILS

Most of the material used was obtained from channel samples cut in the coal seams
worked from small adits. Associated shales were also collected. Where seams were
sampled at outcrop, the identification was based on the Geological Survey 6 inch to
1 mile maps of the area. Supplementary samples were also taken from tips, if definite
knowledge of the seams worked was available at the Gaveller’s office. Samples from the
Howie Hill outlier (see Trotter 1942, p. 3, fig. 1) were also examined. The microfossils
obtained from these samples are referred to under ‘Occurrence’ as ? Trenchard (Howie
Hill).

Approximately 10 grams of coal were taken at a time and broken into small pieces

86

PALAEONTOLOGY, VOLUME 8

approximately 5 mm. in diameter. The coal was then placed in a glass flask and treated
with Schulze's solution for a period of time which varied between twelve and forty-eight
hours. After decantation of the Schulze's solution and washing with distilled water, a weak
solution (5 per cent.) of potassium hydroxide was added. The samples were repeatedly
washed with water in a sieve (mesh size 180 /x) until the water ran clear. The time neces-
sary for acid and alkali treatment varied for each sample and was only obtained by
experimentation and careful observation of the state of the sample during the process.
The mineral matter in shale samples was removed by treatment with hydrocloric and
hydrofluoric acids before the oxidation of the remaining organic material.

The residue obtained by sieving was immersed in water in a small sorting tray and
examined under a stereoscopic microscope (magnifications x35 and x70) and the
megaspores picked oft' with a fine brush and steel needle and stored temporarily in
distilled water in corked glass tubes. A little acid was added to prevent any mould
developing.

Specimens were examined with both transmitted and reflected light. In general the
greatest detail was obtained by using transmitted light, especially on the thinner walled
lageniculate forms. The large thick walled forms were studied more successfully by
reflected light, since the body colour of some specimens was too dark for transmitted
light study. The transparency of the spore coat was frequently improved by further
treatment with concentrated nitric acid or a sodium hypochlorite solution (30-50 per
cent.). The latter was found to be the quicker method, but careful attention was necessary
in order to prevent specimens from being destroyed. Specimens placed in such a solution
soon lost their dark colour and if the solution was not neutralized by the addition of
sodium sulphide the spores disintegrated after a short time. The large thick walled spores,
e.g. Laevigatisporites, were successfully bleached by this method, but became very
fragile and were difficult to mount after such treatment. Consequently, most of the large
spores were placed on thin pieces of glass in cardboard single cell slides with cellulose
covers and allowed to dry, before examination by reflected light (magnification x70
and x200). Some thin-walled forms were also treated in the same manner, but the
majority w'ere mounted in glycerine jelly on glass slides, the coverslips being sealed with
beeswax. These were examined under transmitted light (magnification x 100 and x450).
Photographs were taken with a Zeiss 35 mm. attachment camera using transmitted and
oblique reflected light. All the illustrated specimens are lodged in the permanent collec-
tions of the Micropalaeontology Laboratory, Department of Geology, University of
Sheffield.

Classification. Because of the dispersed nature and the uncertain botanical affinities of
many spores, the artificial classification based on spore morphology as proposed by
Potonie and Kremp (1954) is used. Slight modifications on this classification are pro-
posed, where this is considered advisable in the light of the present investigation. Four
new species of Lagenicula (Zerndt) Potonie and Kremp and one new species of Setosi-
sporites (Ibrahim) Potonie and Kremp are described, and the genus Zonalesporites
(Ibrahim) Potonie and Kremp is redefined. The descriptive terms are mainly used in
accordance with the recommendations made by the Commission Internationale de
Microflore du Paleozoique (C.I.M.P., 1961, Krefeld).

E. SPINNER: WESTPHALIAN MEGASPORES FROM THE FOREST OF DEAN 87

SYSTEMATIC DESCRIPTIONS

Anteturma sporites H. Potonie 1893
Turma triletes (Reinsch) Potonie and Kremp 1954
Subturma azonotriletes Luber 1935
Infraturma laevigati (Bennie and Kidston) Potonie 1956
Genus laevigatisporites (Ibrahim) Potonie and Kremp 1954

Type species. Laevigatisporites primus (Wicher)

Laevigatisporites glabratus (Zerndt) Potonie and Kremp sensu Dijkstra

Plate 14, figs. 1, 2

1930 Triletes glabratus Zerndt, pp. 43-45, pi. 1, figs. 1-3.

1933 Laevigatisporites reinschi Ibrahim, p. 18, pi. 4, fig. 28.

1934 Sporites primus Wicher, p. 169.

1946 Triletes glabratus Dijkstra, pp. 26-28, pi. 1, figs. 28-40.

1955 Laevigatisporites glabratus Potonie and Kremp, p. 53, pi. 1, figs. 4-8.

1955 Laevigatisporites reinschi Potonie and Kremp, p. 55, pi. 2, figs. 9-10.

1955 Laevigatisporites primus Potonie and Kremp, p. 55, pi. 1, fig. 2.

1958 Laevigatisporites glabratus Pierart, p. 40, pi. 1, figs. 1-2.

1959 Triletes glabratus Winslow, p. 28, pi. 6, figs. 7-10.

1959 Laevigatisporites glabratus Danze and Vigreux, p. 132.

Remarks. Considerable variation in size, haptotypic structures, and thickness of the
spore wall occurs within this species. Two main types seem to be present, (1) a large
relatively thin-walled spore, circular in outline with almost suppressed curvaturae and
laesurae, (2) a smaller thicker-walled form, rounded triangular in outline with prominent
contact faces, curvaturae, and triradiate ridges extending from the laesurae. However,
many specimens were found in the same assemblage linking both types. Some specimens
could be assigned to L. primus, L. reinschi, or L. glabratus as described by Potonie and
Kremp (1955, pp. 51-56), but the majority have characteristics common to more than
one of the above species. In view of this wide variation in size range (max. diam. 800-
2400 p) and limited morphological features the group has been referred here to one
‘broad’ species, i.e. sensu Dijkstra (Schopf 1949, p. 509). Pierart (1958, p. 34) has been
followed in retaining the generic name Laevigatisporites.

Affinities. Bochenski (1936, p. 225; 1939, p. 5, pi. 5, figs. 30-32, pi. 7, figs. 42-45) demonstrated similar
variation in size of this type of megaspore in the cone species Sigillariostrobus czarnockii. Laevigati-
sporites type megaspores have also been recorded from Mazocarpon oedipternum Schopf (1941)
and Sigillariostrobus gothani Bode (1928), as stated by Chaloner (1953^, p. 887).

Stratigraphic range. Namurian-Stephanian (Dijkstra 1946).

Occurrence. ?Trenchard (Howie Hill), Lowery (loc. 26), Twenty Inch (loc. 28, 29) seams.

Infraturma apiculati (Bennie and Kidston) Potonie and Kremp 1956
Genus tuberculatisporites (Ibrahim) Potonie and Kremp 1954

Tuber culatisporites brevispieulus (Schopf) Potonie and Kremp 1955

Plate 14, figs. 3-5

88 PALAEONTOLOGY, VOLUME 8

1938 Triletes brevispiculus Schopf, p. 26, pi. 1, figs. 13 a-r, pi. 2, fig. 6, pi. 3, figs. 1-4.

1946 Triletes mamillarius Dijkstra pars, non Bartlett, p. 28 (synonymy).

1955 Tuberculatisporites brevispiculus Potonie and Kremp, p. 90.

1957 Tuberculatisporites brevispiculus Bhardwaj, p. 91, pi. 24, figs. 14-16.

1959 Triletes mamillarius Winslow pars, non Bartlett, p. 29.
non 1955 Tuberculatisporites brevispiculus in Horst, p. 163.

Remarks. Dijkstra (1946, p. 28) considered this species as a synonym of ‘ Triletes ’
mamillarius Bartlett and maintained that the variation in size of ornament was so great
that only one species could be satisfactorily defined. Evidence obtained by Bochenski
(1939, p. 21, pi. 4, figs. 16-26) and Chaloner (19536, p. 882, pi. 22, figs. 1-3) of Tuber-
culatisporites type spores in Sigillarian cones partly supports Dijkstra’s interpretation.
However, neither author found spores with sculptural elements of the small size found
in T. brevispiculus. Moreover, Arnold (1961, p. 250) has re-examined Bartlett’s type
material of T. mamillarius and concluded that it is distinct from T. brevispiculus. All the
specimens from the Forest of Dean agree closely with Schopf’s original description.
Horst (1955, p. 163) described specimens characterized by sculptural elements 15 to
30 p high, 40 to 65 p in diameter as T. brevispiculus. None of these were figured; but
judging from the dimensions of the cone-shaped elements, it is doubtful that they belong
to T. brevispiculus.

Affinities. Sigillariaceae.

Stratigraphic distribution. U.S.A.: Herrin (No. 6) Coal, Carbondale series, Pennsylvanian, Illinois
(Schopf 1938). Europe: Saar and Ruhr coalfields, lower Westphalian D (Bhardwaj 19576).

Occurrence. Trenchard (loc. 1-3), Coleford High Delf (loc. 6), Yorkley (loc. 15), Woorgreen (loc. 31)
seams.

Subturma lagenotriletes Potonie and Kremp (1954) emend. Bhardwaj 1957
Infraturma gulati Bhardwaj 1957
Genus setosisporites (Ibrahim) Potonie and Kremp 1956

Type species. Setosisporites hirsutus (Loose) Ibrahim 1933

Setosisporites pilatus sp. nov.

Plate 14, fig. 6; Plate 15, figs. 1, 2

Holotype. Slide FD/12, Crow Delf coal, Collingwood level. Plate 15, fig. 1.

Diagnosis. Trilete megaspores, circular to oval outline, maximum diameter 450 to 700 p.
Laesurae expanded at proximal pole to form small apical prominence, 60 to 100 p
diameter. Small club-shaped pilae and baculae 10 to 15^ long, 2 to 6 p wide, cover the
distal surface. Contact faces are distinct, laevigate, occupying up to three-quarters of the
proximal surface of the compressed spore. Spore wall is approximately 20 p thick.

Description

Size and Shape. Small trilete megaspores, circular to oval in outline, maximum diameter
varies between 450 and 700 p, mean 530 p (twelve specimens measured in glycerine
jelly). Oblique polar compressions are most common, lateral compressions rare. The
spore body was originally more or less spherical in shape.

E. SPINNER: WESTPHALIAN MEGASPORES FROM THE FOREST OF DEAN 89

Haptotypic structures. Tetrad mark is represented by laesurae, equal to three-quarters
of the spore radius in length, 20 to 25 p wide. Near the proximal pole, the laesurae are
abruptly expanded to form a blunt pyramidal prominence, varying between 60 and 100 p
in maximum diameter. There is no appreciable expansion of the contact faces involved
in the formation of this structure. The contact faces are distinct, laevigate, locally
thickened near the proximal pole, and occupy half to three-quarters of the proximal
surface of the compressed spore. The abrupt change of ornament from the laevigate
contact faces to the pilate elements of the distal suface and the relatively thinner exine
of the contact faces determines the position of the curvaturae.

Exine Structure and Sculpture. The spore body, excluding the contact faces, is covered
with a combination of pilate and baculate elements, densely developed in places (approxi-
mately 5 p apart), but elsewhere more widely dispersed ( 15 to 25 p). The elements vary
between 10 and 15 p in overall length, 3 to 6 p in width. They are pilate in the sense that

text-fig. 2. Diagram to illustrate variation in ornament of Setosisporites pilatus sp. nov., X 2,000;

drawn from holotype slide FD/12.

some are composed of a distinct ‘head’ and ‘stalk’ whilst in others a smooth transition
exists between the smaller basal diameter and the apical diameter, thus forming what
might be described as ‘club-shaped’ baculae. The ‘stalks’ are approximately 1 p long
up to 5 p wide, with straight or slightly concave sides bearing spherical heads, 3 to 6 p
in diameter. The heads are terminated by small cone-like structures, approximately
2 p high. Some of the baculae have typically straight, parallel sides. There is no notice-
able segregation of the two types of ornament on the spore body except in the region
of the curvaturae, where a dense development of small baculae predominates. The spore
wall is approximately 20 p thick, as measured in optical section, and appears to have a
rather complex infrastructure. Under reflected light the spore appears yellow-brown in
colour. In places the exine of the contact faces appears darker (i.e. thickened), but these
areas are not prominently elevated above the spore wall. The pilate-baculate ornamenta-
tion gives the sporea a matted or hairy appearance.

Comparison. S. pseudotenuispinosus Pierart (1958) is similar in size to S. pilatus but
has a larger apical prominence (80 to 200 p wide, 80 to 150 ju, high) and a different
type of distal ornamentation consisting of small granules up to 5 p in diameter. S.
hirsutus var. brevispinosus (Zerndt) Potonie and Kremp (1955) also has smaller sculp-
tural elements (tubercles, 6 p long) than S. pilatus. The contact faces in S. pilatus are
also laevigate, not ornamented as in S. hirsutus var. brevispinosus, and although the con-
tact faces are locally thickened, no prominent radial folds are developed as in S. hirsutus
var. brevispinosus. S. globosus (Arnold) Potonie and Kremp (1955) and the varieties of
S. globosus described by Winslow (1959), can all be distinguished from S. pilatus by the
larger sculptural elements on the distal surface and the small elements on the contact
faces. Also, S. globosus var. B Winslow has an equatorial flange structure and a larger

90

PALAEONTOLOGY, VOLUME 8

apical prominence than S. pilatus. S. hirsutus (Loose) Ibrahim (1933) is slightly larger
than S. pilatus and has longer sculptural elements, up to 200 p, long. S. praetextus
(Zerndt) Potonie and Kremp is larger (600 to 1,800 pi) than S. pilatus and has longer
(200 to 300 p) bifurcating sculptural elements usually restricted to a zone around the
equator. The apical prominence in S. praetextus is also much larger than in S. pilatus.

Remarks. This species is placed in the genus Setosisporites on the basis of the spherical
shape of the spore body, the small apical prominence formed by the proximal part of the
laesurae, the smooth contact faces, and the ‘hairy’ appearance of the spore coat.

Due to the breaking of some of the sculptural elements the spore coat may some-
times appear verrucose.

Affinities. S. pilatus type spores have not been described from any known cone species. According to
Potonie (1962), Setosisporites type spores have been found in cones belonging to the Bothrodendraceae.

Occurrence. Crow Delf Coal (loc. 30).

Genus lagenoisporites Potonie and Kremp 1954
Type species. Lagenoisporites rugosus (Loose) Potonie and Kremp 1954

Remarks. The differences between Lagenoisporites and Lagenicula are not very clear.
According to Potonie and Kremp (1954, p. 151), Lagenoisporites has a more or less
smooth exine and, in any case, does not show a distinct ornamentation as in Lagenicula
and Setosisporites. However, the distinctiveness of an ornamentation can be affected by
the size of the sculptural elements, the state of preservation, and laboratory treatment,
insufficient oxidation may cause the ornament to be obscured, while over-oxidation can
cause breakage or removal of part of the ornament. The method of examination is also
important, since a small ornament may be less distinct under reflected light than it is
under transmitted light. Moreover, Chaloner (1953c, p. 284, text-fig. 20) has found
both smooth and ornamented lageniculate megaspores in the same fructifications (i.e.
Lepidostrobus rusellianus Binney, L. olryi Zeiller). It therefore seems that the difference
between the two genera as at present described is inadequate and, in fact, Bhardwaj and
Kremp (1955, p. 43, pi. 4, figs. 2, 3), have assigned species with a distinct ornamentation

EXPLANATION OF PLATE 14

All figures x 50, under reflected light, unless otherwise stated.

Figs. 1-2. Laevigatisporites glabratus (Zernit) Potonie and Kremp sensu Dijkstra. Lowery seam (loc.
26), FD/1, FD/2.

Figs. 3-5. Tuberculatisporites brevispiculus (Schopf) Potonie and Kremp. 3, mature form, Trenchard
seam (loc. 2), FD/3. 4, sculptural elements x400, transmitted light, Trenchard seam (loc. 2),
FD/4. 5, abortive form, Woorgreen seam (loc. 31), FD/5.

Fig. 6. Setosisporites pilatus sp. nov. Crow Delf seam (loc. 30), transmitted light, FD/6.

Fig. 7. Lagenoisporites rugosus (Loose) Potonie and Kremp; lateral compression, transmitted light;
Trenchard seam (loc. 2), FD/7.

Figs. 8-9. Lagenicula verrurugosa sp. nov. 8, holotype X 80, transmitted light, Trenchard seam (loc. 2),
FD/8. 9, Exine ornamentation x 500, transmitted light, FD/8.

Figs. 10-12. Triangulatisporites regalis (Ibrahim) Potonie and Kremp. 10, central body after removal of
outer reticulate layer, exposing commissures and triangular inner membrane, trasmitted light,
x 60, Whittington seam (loc. 12), FD/9. 1 1, proximal surface, polar view, Coleford High Delf seam
(loc. 6), FD/10. 12, lateral compression, transmitted light, Crow Delf seam (loc. 30), FD/ll.

Palaeontology, Vol. 8

PLATE 14

\ t-M*1 .

SPINNER, Westphalian D megaspores

E. SPINNER: WESTPHALIAN MEGASPORES FROM THE FOREST OF DEAN 91

to Lagenoisporites. Lagenoisporites is regarded here as a generic group for lageniculate
spores with an apical prominence formed by the expansion of laesurae and parts of the
contact faces, and a laevigate spore exine.

Lagenoisporites rugosus (Loose) Potonie and Kremp 1954
Plate 14, fig. 7

1932 Sporonites rugosus Loose, p. 452, fig. 59.

1955 Lagenoisporites rugosus Potonie and Kremp, p. 122, pi. 4, fig. 22.

1958 Lagenoisporites rugosus Pierart, p. 40, pi. 3, fig. 1 1 ; pi. 10, figs. 1-4.

1959 Triletes rugosus Winslow pars, pp. 22-24, pi. 3, figs. 4-6.

Remarks. Dijkstra (1946, p. 48; 19556, p. 10; 19566, p. 259) described a small, thick-
walled ornamented ‘immature ’form within this species. However, the work by Chaloner
(1953c, pp. 272-86) and Felix (1954, pp. 357-60) on L. rugosus type spores from
Lepidostrobus cones does not support this interpretation.

Stratigraphic distribution. Europe: Westphalian A-Stephanian (Dijkstra 1955 a, b, 1956 b). U.S.A.:
Pennsylvanian, Michigan (Arnold 1950), Illinois (Schopf 1938; Winslow 1959).

Occurrence. Trenchard to Woorgreen (No. 2) seams.

Genus lagenicula (Bennie and Kidston) Potonie and Kremp 1954
Type species. Lagenicula horrida Zerndt 1934.

Remarks. Bharadwaj and Venkatachala (1962, p. 25) proposed a new genus Rostra ti-
spora , based on megaspores obtained from the Lower Carboniferous of Spitsbergen.
This genus is, according to its authors, distinguished from Lagenicula by its verrucose
ornament and small apical prominence. However, Potonie and Kremp ( 1954, p. 151) in
their emendation of Lagenicula made some provision for lageniculate forms with a verru-
cose ornament by describing the exine as being closely covered with warts (verrucae) on
which strong, pointed spines or hairs could stand or occur. The other main difference
between Rostratispora and Lagenicula would appear to be in the ratio of size (height) of
apical prominence to spore body. The only species of Rostratispora described (R.
iucundus Bharadwaj and Venkatachala 1962) shows a preferred lateral direction of com-
pression, which indicates a larger polar axis, as occurs in Lagenicula. The two genera
are therefore very similar and it seems that Rostratispora falls within the range of
variability admitted for Lagenicula as defined by Potonie and Kremp 1954.

Lagenicula verrurugosa sp. nov.

Plate 14, figs. 8, 9; Plate 15, fig. 3
? 1955 Triletes rugosus Dijkstra pars, p. 10, pi. 2, figs. 19, 20.

Holotype. Slide FD/8, Trenchard seam, Mapleford Colliery. The species name refers to the characteristic
ornament on the distal surface. Plate 14, fig. 8.

Diagnosis. Flask-shaped trilete megaspores, 500 to 900 p maximum diameter, including
apical prominence. Apical prominence varies from 160 to 280 p in height, 200 to 320 p
in width. Contact faces distinct, occupying half the proximal surface of compressed

92

PALAEONTOLOGY, VOLUME 8

spore, arcuate ridges low, 30 p wide. Distal surface covered with verrucae and rugulae,
up to 30 p in diameter, height less than half diameter. Spore wall ?foveolate, 25 p thick.

Description

Size and Shape. Medium-sized trilete megaspores varying between 500 and 900 p in
maximum diameter (including apical prominence), mean 655 p (32 specimens measured
in glycerine jelly). Polar compressions are circular to oval in outline, lateral compressions
flask-shaped, the apical prominence forming the ‘neck’ of the flask. Spore body was
originally spherical in shape.

Haptotypic structures. Tetrad mark is represented by an apical prominence formed by
the laesurae and the apical parts of the contact faces. Laesurae are equal in length to
one-third to one-half of the radius of the spore body; they are 30 p to 40 p high and wide
at the junction with the curvaturae and gradually increase in height (up to 100 p) at
the proximal pole. Near the proximal pole the laesurae are often contorted. The apical
prominence is 160 to 280 p high (measured from curvaturae to apex on lateral compres-
sions) and 200 to 320 p wide. Contact faces are distinct; they occupy approximately
half the proximal surface of the compressed spore and are delimited by low arcuate
ridges, up to 30 p wide.

Exine Structure and Sculpture. Small densely placed verrucae, circular in outline, up to
10 /x in diameter, occur on the apical prominence. The arcuate ridges are usually orna-
mented with larger verrucae, 20 to 40 p in diameter, up to 20 p high. A combination of
densely developed verrucae and rugulae covers the distal surface. These elements are
elongate, irregular to circular in outline, up to 30 p in diameter, and up to 10 ft high.
Spore wall is approximately 25 p thick, ?foveolate (fovea 1 p in diameter). Under
reflected light the spore appears yellow-brown in colour, glossy, and ‘granulose’.

Comparison. Lagenoisporites rugosus (Loose) Potonie and Kremp is larger (300 to
1,170 ft) than L. verrurugosa and lacks the ornament on the distal surface. L. irregularis
sp. nov. is larger than L. verrurugosa in overall size (700 to 1,210 ft), and has the distal
surface covered with large irregular thickenings 35 to 70 p wide, 10 p high, and a thinner
spore wall (10 to 15 ft). L. verrucata sp. nov. is similar in size to L. verrurugosa but has
large verrucae, 40 to 70 p diameter, 20 to 30 p high, developed on the curvaturae and
distal surface. The apical prominence is smaller (up to 150 ft high). L. perverrucata sp.
nov. is larger (300 to 1,100 ft) than L. verrurugosa and has a coarser ornamentation, with
verrucae 20 to 35 p high. Rostratispora iucundus Bharadwaj and Venkatachala 1962 is
similar in shape and ornamentation to L. verrurugosa, but is much smaller (300 to 500 p)
in spore size and in the size of the apical prominence (50 to 80 ft).

Remarks. Dijkstra (1946, p. 48; 1955A p. 259) described some specimens as ‘immature’
forms oCTri/etes rugosus ’ which are similar to L. verrurugosa. His description and illus-
trations are based on reflected light only and it is difficult to compare with the greater
detail shown by transmitted light. Three slides containing specimens of L. verrurugosa
after being examined by transmitted light were dismantled, washed and re-examined
by reflected light. They appeared very similar to the illustrations of Dijkstra (1955/?,
pi. 2, figs. 19, 20). However, Dijkstra’s evidence for immaturity is not very convincing
(see remarks in Lagenoisporites rugosus above).

E. SPINNER: WESTPHALIAN MEGAPSORES FROM THE FOREST OF DEAN 93

Stratigraphic Distribution. ? Sfaia, Egypt, Upper Westphalian C, Dijkstra (19556).

Occurrence. Trenchard (loc. 2, 3), Colefield High Delf (loc. 6, 8, 9, 11), Brazilly (loc. 17, 18), Rockey
<loc. 23), Starkey (loc. 24, 25), Lowery (loc. 26, 27), Twenty Inch (loc. 28), and Woorgreen (No. 2)
(loc. 31) coal seams.

Lagenicula perverrucata sp. nov.

Plate 15, figs. 4, 5

Holotype. Slide no. FD/16, ? Trenchard seam (Howie Hill); Plate 15, fig. 4.

Diagnosis. Flask-shaped trilete megaspores, varying between 500 and 1 , 1 00 p i n maximum
diameter. The contact faces and laesurae form a large apical prominence equal in height
to approximately one-third polar axis, slightly constricted at the base, often ruptured
to form spoon-shaped segments. Large verrucae, 20 to 70p in diameter, height equal to
half diameter, cover spore body and lower part of the apical prominence. Verrucae are
polygonal in outline and dome-shaped in lateral view. Spore wall is 20 to 30 p thick.

Description

Size and shape. Medium to large size trilete megaspores, which are flask-shaped due to
the development of a large apical prominence. The maximum diameter of the com-
pressed spore varies between 500 and 1,100 p (mean 750 p, as measured on nineteen
specimens in glycerine jelly). Lateral compressions are most common; the polar axis
being longer than the equatorial axis.

Haptotvpic Structures. The tetrad mark is represented by the large apical prominence
formed by the expansion of the greater part of the contact faces and the laesurae. In
lateral compressions, the apical prominence is oval in outline and varies between 175
and 400 p in height and width. The height of the apical prominence is approximately
equal to one-third of the length of the polar axis and the maximum width occurs just
above the slight basal constriction. The apical prominence is commonly found ruptured,
forming spoon-shaped flaps. Originally, the apical prominence was more or less bluntly
pyramidal in shape. Most of the details of the laesurae, contact faces, and curvaturae are
obscured by the dense verrucose ornament of the exine. On some specimens the low
arcuate thickenings of the curvaturae can be distinguished (under transmitted light) just
below the constriction at the base of the apical prominence.

Exine Structure and Sculpture. Large, closely-spaced verrucae cover the spore body and
lower part of the apical prominence. The verrucae are polygonal at the base, appearing
dome-shaped where they project from the spore margin. They are up to 30 p high and
vary between 20 and 70 p in diameter, generally more than 30 p. The height usually
equals approximately half the diameter. Narrow grooves (c. 1 p wide) run radially from
the centres of the verrucae towards the margins (as visible under oil immersion X 1 ,000).
The distance between two adjoining verrucae is usually less than the diameter of an
individual verruca. On the basal part of the apical prominence the verrucae are smaller
than elsewhere (approximately 30 p). The upper part of the apical prominence is smooth,
but scattered verrucae may occur. The spore wall ( ? foveolate) varies between 20 and
30 p in thickness.

Comparison. Lagenicula perverrucata differs from Lagenicula angulata Zerndt (1937) in

94

PALAEONTOLOGY, VOLUME 8

lacking the horn-shaped protuberances formed at the junction of the laesurae and curva-
turae. The ornament of L. perverrucata is larger (30 to 70 p) than in L. angulata (25 p
length, 10 to 15 /x diameter). Lagenicula agnina Zerndt (1937a) has an ornamentation
similar to that of L. perverrucata , but possesses smooth contact faces and a thicker spore
wall (80 to 120 p). 'Triletes' furcatus Dijkstra (1956a) is similar in size and thickness of
the spore wall to L. perverrucata , but the sculptural elements of T. furcatus are more
conical, and there is a marked difference in size between those on the distal surface (30
to 70 p long, 20 p diameter) and on the contact faces (3 to 5 p diameter, 7 p long).
Lagenicula splendida Zerndt (1937a) is larger and has more conical elements than those
of L. perverrucata. L. verrucata sp. nov. is smaller in overall size (370 to 585 p) and in
the size of the apical prominence (150/x) than L. perverrucata and has prominently
ornamented curvaturae and distinct contact faces. L. irregularis sp. nov. is similar in size
and shape to L. perverrucata , but the shape of the ornament is different, with verrucae
35 to 70 p in diameter, but only 10 p high. In L. perverrucata the height of the verrucae
is approximately half the diameter.

Remarks. L. perverrucata is somewhat atypical of the genus in that the contact faces
are not very distinct from the remainder of the spore body.

Occurrence. ? Trenchard (Howie Hill) (loc. 4), and Woorgreen (No. 2) (loc. 31) seams.

Lagenicula? verrucata sp. nov.

Plate 15, figs. 8, 9

Holotvpe. Slide FD/20, Trenchard seam, Mapleford Colliery; Plate 15, fig. 8.

Diagnosis. Small trilete megaspores varying between 370 and 585 p in maximum dia-
meter. Prominent curvaturae are marked by strongly developed verrucae (40 to 70 p
diameter), which also characterize the distal surface. A small apical prominence may
be formed by the laesurae and apical parts of the contact faces. Spore wall is approxi-
mately 20 p thick.

Description

Size and Shape. Small trilete megaspores varying between 370 and 585 p in maximum
diameter (mean 500 p, nine specimens measured in glycerine jelly), more or less circular
in equatorial outline. Lateral compressions are slightly flask-shaped. Most specimens are

EXPLANATION OF PLATE 1 5

All specimens with transmitted light.

Figs. 1-2. Setosisporites pilatus sp. nov. 1, Holotype, X 100; Crow Delf seam (loc. 30), FD/12. 2, Apical
prominence of FD/12, X 250.

Figs. 3-9. Lagenicula spp. 3, L. verrurugosa sp. nov.; lateral compression X 50, Trenchard seam (loc.
2), FD/13. 4-5. L. perverrucata sp. nov. 4, Holotype, x50; lateral compression, Trenchard seam
(loc. 3), FD/14. 5, Sculptural elements of FD/14, X 500. 6-7. L. irregularis sp. nov. 6, Holotype,
X 50; lateral compression, Trenchard seam (loc. 2), FD/15. 7, Sculptural elements of FD/15, X 500
8-9. L.? verrucata sp. nov. 8, Holotype, X 100; Trenchard seam (loc. 2), FD/16. 9, Tetrad scar of
FD/16, X 250.

Fig. 10. Triangularisporites regalis (Ibrahim) Potonie and Kremp, distal reticulation, X 500; Coleford
High Delf seam (loc. 8), FD/17.

Palaeontology, Vol. 8

PLATE 15

SPINNER, Westphalian D megaspores

E. SPINNER: WESTPHALIAN MEGASPORES FROM THE FOREST OF DEAN 95

compressed slightly oblique to the polar axis. Originally, the spore was spherical to
slightly flask-shaped, the polar axis being only slightly larger than the equatorial axis.

Haptotypic Structure. The tetrad mark is represented by laesurae equalling approxi-
mately half the radius of the spore body in length. At the junction with the curvaturae,
laesurae are 30 to 50 p high, increasing gradually up to 80 p at the proximal pole. The
contact faces are distinct, occupying up to half the proximal surface of the compressed
spore. The apical parts of the contact faces are slightly thickened and, together with the
laesurae, form a small bluntly pyramidal prominence, 100 to 150 p high. The curvaturae
form low but prominent arcuate ridges, ornamented with large verrucae and/or pila.

Exine Structure and Sculpture. Large verrucae are present on the distal surface, curva-
turae and sometimes the laesurae. They are most densely developed on the curvaturae,
being more scattered on the distal surface; diameters vary between 40 and 90 p, height
up to 30 p. The height of an individual element is less than half the diameter. Circular
to slightly irregular in outline, as seen from above, the verrucae have a low dome-shaped
outline with a slightly wrinkled surface in lateral view. Pila may occur intermingled with
the verrucae. The spherical heads of the pila vary between 20 and 35 p in diameter. They
are borne by a short stalk, 8 to 10 high, approximately 5 p wide. Smaller verrucae
(5 to 20 p diameter, up to 10 p high) may occur between the larger elements and on the
thickened apical parts of the contact faces. The spore wall (? foveolate) is approximately
20 p thick, as measured in optical section.

Comparison. Lagenicula? verruccita sp. nov. can be distinguished by its small size and the
large verrucate thickenings on the curvaturae and distal surface. These characteristics
clearly set it apart from all other species of Lagenicula that have so far been described.

Remarks. This species combines some characteristics of both Lagenicula and Rostrati-
spora Bharadwaj and Venkatachala (1962). The raised laesurae are often folded at the
proximal pole and the thickened apical parts of the contact faces ( PI. 1 5, fig. 9) indicate
the presence of a low pyramidal prominence. The species is atypical of Lagenicula in
the small size of this structure. Oblique compressions are most common, the polar axis
being only slightly longer than the equatorial axis. In Lagenicula , lateral compressions
are common as a result of the longer polar axis of the spore.

This species resembles Rostratispora in its verrucose ornament and small apical
prominence. However, in view of the close similarity between Lagenicula and Rostrati-
spora as defined at present, this species has been tentatively assigned to the older genus
Lagenicula until more evidence is available.

Occurrence. Trenchard seam (loc. 2).

Lagenicula irregularis sp. nov.

Plate 15, figs. 6, 7

Holotype. Slide FD/18, Trenchard seam, Mapleford colliery. Plate 15, fig. 6.

Diagnosis. Flask-shaped trilete megaspores, size range 700 to 1,210 p. Apical prominence
formed by the laesurae and apical parts of the contact faces, height 180 to 300 p.
Contact faces occupy less than half the proximal surface of the compressed spore, and

96

PALAEONTOLOGY, VOLUME 8

are ornamented with verrucae 20 p in diameter, up to 15 ft high. Arcuate ridges are not
developed. Distal surface covered with large lens-shaped thickenings (verrucae),
irregular in outline. Spore wall is 10 to 15 ft thick. It has a complex structure.

Description

Size and Shape. Medium-sized trilete megaspores, flask-shaped, varying between 7C0
and 1,210 p in maximum diameter, including apical prominence, mean 966 p (after 15
specimens in glycerine jelly). Lateral compressions are the most common. Originally the
spore body was more or less spherical in shape; the apical prominence forming the
‘neck’ of the flask.

Haptotypic Structures. The tetrad mark is represented bylaesurae less than half the spore
radius in length. The laesurae and most of the contact faces are expanded to form an
apical prominence, elliptical in outline in lateral compressions, 180 to 300 ft high (as
measured from curvaturae to apex). Small verrucae, up to 20 p in diameter and height,
may occur on the apical prominence. The contact faces are not very distinct and occupy
less than half the proximal surface of the compressed spore. No arcuate ridges are
developed, but the exine may be thickened opposite the laesurae, i.e. curvaturae im-
perfectae. The contact faces are distinguished by the absence of the large verrucate orna-
ment which characterizes the distal surface.

Exine Structure and Sculpture. The distal surface is covered by large verrucae, 35 to
70 p in diameter, rarely more than 10 p high (5 to 15 p). In polar view, the verrucae are
irregular to circular in outline; narrow lens-shaped where they project from the spore
margin. Often the verrucae are joined to form rugulae or large, irregular, thickened areas.
The spore wall is 10 to 15 ft thick, as measured in optical section, with a complex
structure.

Comparison. L. irregularis can be distinguished from L. perverrucata sp. nov. by its
irregularly shaped verrucae, since these show a ratio of height to diameter which is
greater (1:4 to 1:7) than that of L. perverrucata (1 :2). The spore wall in L. perverrucata
is thicker than in L. irregularis (30 to 33 ft). Lagenoisporites rugosus (Loose) Potonie and
Kremp can be distinguished from L. irregularis by its smooth distal surface. L.? verru-
cata sp. nov. is smaller in size (370 to 585 ft) than L. irregularis and has distinctly orna-
mented curvaturae.

Occurrence. Trenchard seam (loc. 2), Brazilly seam (Iocs. 17, 18).

Turma zonales (Bennie and Kidston) Potonie 1956
Subturma auritotriletes Potonie and Kremp 1954
Infraturma auriculati (Schopf) Potonie and Kremp 1956
Genus valvisisporites (Ibrahim) Potonie and Kremp 1954

Valvisisporites auritus (Zerndt) Potonie and Kremp 1956 sensu Bhardwaj 1957

Plate 17, figs. 6, 7

1930 Triletes auritus Zerndt, p. 46, pi. 1, figs. 4-5.

1957 Valvisisporites auritus Bhardwaj, p. 98, pi. 26, figs. 10-13.

1959 Valvisisporites auritus Kalibova, p. 431, pi. 11, figs. 1-4.

E. SPINNER: WESTPHALIAN MEGASPORES FROM THE FOREST OF DEAN 97
Stratigraphic Distribution. Westphalian D-Stephanian (Bhardwaj 1957a).

Occurrence. Coleford High Delf (loc. 6, 7, 8, 10, 1 1), Lowery (loc. 26, 27), Twenty Inch (loc. 28), Crow
Delf (loc. 30) seams.

Valvisisporiles nigrozonales (Stach and Zerndt) Potonie and Kremp 1956

Plate 17, figs. 8, 9

1931 Triletes nigrozonales Stach and Zerndt, p. 1123, pi. 2, figs. 26-27.

1956 Valvisisporites nigrozonales Potonie and Kremp, p. 96, pi. 5, figs. 35-37 (part I).

1957 Valvisisporites nigrozonales Bhardwaj, p. 98.

1958 Valvisisporites nigrozonales Pierart, p. 50, pi. 2, figs. 1-4.

1959 Valvisisporites nigrozonales Danze and Vigreux, p. 134.

Remarks. The megaspores described here as V. nigrozonales are similar in size (560 to
900 p) to those described by Potonie and Kremp, but have no clearly granular sculpture.
However, this might be due to preservation. Under transmitted light the spore wall
appears to have a complex structure (? infrareticulate). A thin folded inner membrane
may also be seen in some specimens.

Affinities. The megaspore illustrated by Chaloner (1958, p. 31, fig. 5) from Polysporia mirabilis New-
berry is similar to V. nigrozonales. Chaloner compared with V. auritus sensu lato.

Stratigraphic Distribution. Lower Westphalian B-Westphalian D (Potonie and Kremp).

Occurrence. Trenchard (loc. 2, 3), Coleford High Delf (loc. 8, II), Woorgreen No. 2 seams.

Valvisisporiles sofiaense (Schopf, Wilson, and Bentall) Bhardwaj 1957

Plate 16, fig. 4

1932 Triletes auritus III Maslankiewiczowa, p. 161, figs. 37-38.

1944 Triletes sofiaense Schopf, Wilson, and Bentall, p. 25.

1946 Triletes auritus Dijkstra pars, p. 31 (synonomy).

1951 Triletes auritus Kalibova pars, p. 14.

1957 Valvisisporites sofiaensis Bhardwaj, p. 101.

Description. Spores similar in size (1,100 to 1,300 p) to V. auritus (Zerndt) Bhardwaj.
The laesurae are straight to slightly sinuous, up to 100 p high, 75 to 120 p wide at the
proximal pole, gradually decreasing in height and width towards the auriculae. They
extend on to the base of the auriculae. Distal surface, contact faces, and laesurae are
covered with a coarse reticulum of muri (12 to 25 p wide), surrounding lumina, which
are up to 50 p in diameter, circular to polygonal in outline.

Remarks. Bhardwaj (1957a, p. 101) and Kalibova (1959, p. 434) referred to this species as
"Valvisisporites sofiaensis (S. W. and B.) POT. and KR.’ However, no direct reference
to this species seems to occur in Potonie and Kremp 1954, 1955, 1956, which were the
papers quoted by both Bhardwaj and Kalibova. Both Potonie and Bharadwaj (pers.
communications, November 1963) have agreed that the combination in Bhardwaj
1957a, p. 101 was in error. The combination has therefore been credited here to Bhard-
waj, not Potonie and Kremp.

Stratigraphic Distribution. Maslankiewiczowa (1932, p. 161) first described this species from the
Laziska series in Poland, upper Westphalian B (Schopf, Wilson, and Bentall 1944, p. 25). Bhardwaj
(1957a, p. 101) stated that ‘ V. sofiaensis is reported from the Stephanian' but did not mention a locality.

Occurrence. Twenty Inch (loc. 28) and Crow Delf (loc. 30) seams.

B 0612

H

98

PALAEONTOLOGY, VOLUME 8

Subturma zonotriletes Waltz 1935
Infraturma zonati Potonie and Kremp 1954
Genus triangulatisporites Potonie and Kremp 1954

Triangulatisporites regalis (Ibrahim) Potonie and Kremp 1956

Plate 14, figs. 10-12; Plate 15, fig. 10

Description

Size and Shape. Trilete megaspores with a distinct spherical central body, surrounded
by an equatorial flange or zona, which is widest opposite the laesurae; hence the trian-
gular outline of the spore. Maximum diameter of the spore (including zona), measured
from one radial extremity to the opposite margin of the zona, varies between 600 and
1,040 p (mean 752 p, 35 specimens measured in glycerine jelly). The central body varies
between 430 p and 650 p in diameter (mean 525 p). Both lateral and polar compressions
occur, polar being the most common.

Haptotypic Structures. Tetrad mark on complete specimens represented by prominent
triradiate ridges, wavy in outline, 50 to 70 p high, extending to the margin of the zona.
On denuded specimens (i.e. where the outer reticulate layer of exine has been removed),
the laesurae are up to two-thirds the spore body radius in length. The commissures are
distinct, with low marginal thickenings. On the contact faces low narrow irregular ridges
occur. Some of these appear to coalesce with the triradiate ridges. Towards the spore
body margin, outside the contact faces, these ridges form a reticulum (lumina 20 to 30 p
diameter, muri 5 to 12 p wide). No arcuate ridges are present, but the contact faces are
more or less delimited by the reticulum occurring on the remainder of the spore body.
The lumina become wider on the distal surface of the spore.

Exine Structure and Sculpture. The outer reticulate layer of the exine is extended to
form a zona, 100 to 215 p wide radially, 80 to 120 p interradially. The meshes are most
distinct on the distal surface and consist of irregular polygonal to circular lumina, 40 to
80 p in diameter (mainly 50 to 60 p), surrounded by narrow muri, 5 to 15 p high. Near
the spore body margin the lumina are radially elongate, with the muri extending on to
the zona. The latter may appear pleated or reticulate. The central body is smooth to
finely granular on removal of the reticulate outer layer. The wall of the central body is
15 to 30 p thick. Within the body a thin folded membrane, rounded triangular in outline,
appears to be attached to the laesurae.

EXPLANATION OF PLATE 16

All figures X 50, reflected light illumination unless otherwise stated.

Figs. 1-3. Zonalesporites brasserti (Stach and Zerndt) Potonie and Kremp. 1, Oblique lateral compres-
sion; roof-shale Yorkley seam (loc. 15), FD/18. 2, Detail of zona, transmitted light X 100; roof-shale
Yorkley seam (loc. 15), FD/19. 3, Proximal view, polar compression; roof-shale Yorkley seam
(loc. 15), FD/18.

Fig. 4. Valvisisporites sofiaense (Schopf, Wilson and Bentall) Bhardwaj 1957, lateral compression;
Twenty Inch seam (loc. 28), FD/20.

Figs. 5-9. Cystosporites spp. 5-6, C. giganteus (Zerndt) Schopf. 5, Part of mature specimen with two
adhering abortive spores, transmitted light; Lowery seam (loc. 26), FD/21. 6, Abortive specimen,
transmitted light: Trenchard seam (loc. 1), FD/22. 7-9. C. varius (Wicher) Dijkstra, ? abortive
specimens; Trenchard seam (loc. 1), FD/23.

Palaeontology , Vol. 8

PLATE 16

SPINNER, Westphalian D megaspores

E. SPINNER: WESTPHALIAN MEGASPORES FROM THE FOREST OF DEAN 99

Comparison. T. triangulatus (Zerndt) Potonie and Kremp is smaller in overall size than
T. regalis. Also the meshes of its distal reticulum are smaller, with the lumina up to 60 p
in diameter, separated by muri 1 5 to 20 p wide. T. zonatus (Ibrahim) Potonie and Kremp
differs from T. regalis in its smaller size (400 to 500 p, Potonie and Kremp 1956, p. 131)
and the presence of smaller meshes forming the distal reticulum. Zerndtisporites laxo-
marginalis (Zerndt) Bhardwaj is similar in shape to T. regalis , but the tri radiate ridges
do not extend on to the zona and there is no distal reticulum.

Remarks. The size range of the specimens described here is greater than that of the
specimens described by Potonie and Kremp (1956, p. 129). This may be due to differ-
ences in mounting media used (see Winslow 1959, p. 14; Pant and Srivastava 1962), and
is not considered to be of any taxonomic significance. No papillate ornament as men-
tioned by Winslow (1959, p. 38) was observed on the folded membrane within the cen-
tral body. The outer reticulate layer is characterized by narrow muri (5 to 15 p wide),
surrounding large circular to polygonal lumina (40 to 80 p diameter).

Megaspores from Selaginellites crassicinctus Hoskins and Abbott 1956 (assigned by
those authors to Triletes triangulatus Zerndt) are similar to T. regalis in overall size
(710 to 935 /a), diameter of central body (380 to 450 p), and size of distal reticulations
(30 to 80 /a). However, its equatorial flange is stated to be 100 p thick and, although the
megaspores are not compressed (petrifaction), it is questionable whether a compressed
flange would become thin and membranous as in Triangulatisporites. Megaspores from
Selaginellites suissei Zeiller, as described by Chaloner (1954r/, p. 82), are smaller (450
to 520 p) and agree more closely with T. triangulatus (Zerndt) Potonie and Kremp.

Stratigraphic Distribution. As many workers include this species within Triletes triangulatus Zerndt,
its stratigraphic range is uncertain. According to Potonie and Kremp (1956, p. 129), it occurs in
Upper Westphalian B of the Ruhr; and in Herrin (No. 6) Coal, Illinois, Lower Westphalian D (after
Schopf, 1938).

Occurrence. Trenchard (loc. 1, 2), Coleford High Dell' (loc. 6-11), Whittington (loc. 12), Lowery
(loc. 26), Twenty Inch (loc. 28), and Woorgreen (No. 2) (loc. 31) seams.

Genus zonalesporites (Ibrahim) Potonie and Kremp emend.

Type species. Zonalesporites brasserti (Stach and Zerndt) Potonie and Kremp 1956.

1954 Radiatisporites Potonie and Kremp, 1954, p. 163, pi. 14, fig. 13.

1954 Rotatisporites Potonie and Kremp, 1954, p. 163, pi. 15, fig. 65.

1954 Superbisporites Potonie and Kremp, 1954, p. 164, pi. 15, fig. 64.

Emended diagnosis. Trilete megaspores, circular or oval to rounded triangular in equa-
torial outline and consisting of a distinct spore body with a surrounding equatorial to
subequatorial structure. Laesurae are distinct, wavy or straight in outline and extend to
the spore body margin. Generally, the laesurae are higher at the radial extremities than
they are at the proximal pole. Contact faces occupy most of the proximal surface of the
spore body. They are either laevigate or ornamented with small verrucae, spinae, or
baculae. The equatorial structure consists of numerous fimbria which are either single
or bifurcating and coalescing, and sometimes separated by fovea of varying size. The
distal surface of the spore body may be laevigate or covered with elements similar to those
forming the equatorial structure.

100

PALAEONTOLOGY, VOLUME 8

Comparison. Triangulatisporites Potonie and Kremp 1954 is similar to Zonalesporites in
that an equatorial structure is present. However, this structure or zona is smaller, more
coherent and continuous than in Zonalesporites. No individual elements can be distin-
guished within the zona which appears to be formed by a radial extension of the
reticulate outer layer of the spore wall.

Remarks. The genera Zonalesporites (Ibrahim), Superbisporites, Rotatisporites, and
Radiatisporites, as described by Potonie and Kremp (1954), have been distinguished
mainly on differences in the number and degree of coalescence and bifurcation of the
elements forming the equatorial structure. These differences are small and do not seem
to warrant the retention of four generic names. A similar recommendation was made by
a subcommittee of the Commission Internationale de Microflore du Paleozoique (at
Krefeld 1961), but no formal emendation was proposed.

Zonalesporites brasserti (Stach and Zerndt) Potonie and Kremp

Plate 16, figs. 1-3

1931 Triletes brasserti Stach and Zerndt, p. 1 123, figs. 16, 18-31.

1955 Triletes brasserti forma 2. Dijkstra, p. 334, pi. 38, fig. 15; pi. 39, fig. 23; pi. 41, fig. 27.

1956 Zonalesporites brasserti Potonie and Kremp, p. 122, pi. 7, figs. 52-56.

1958 Zonalesporites brasserti Pierart, p. 57, pi. 10, figs. 17a, b; pi. 11, figs. 1, 2.

1959 Triletes brasserti Winslow, p. 35, pi. 9, figs. 3-10.

Remarks. Under transmitted light the equatorial structure can be seen to consist of
several (? 6) layers of fimbria, varying in length, densely placed, and fused laterally
except for their distal extremities. The outermost layer form the ‘bar-like’ processes
(Winslow, p. 36) projecting at the margin. Small oval dissections (c. 30 p diameter) may
occur near the margin of the equatorial structure, which thickens towards the spore
body and has a fine structure. In some broken specimens a thin membrane occurs within
the spore body.

Stratigraphic Distribution. Namurian B-Westphalian C (Potonie and Kremp 1956, p. 122).
Occurrence. Roof shale of Yorkley coal seam (loc. 14, 15).

Zonalesporites dentatus (Zerndt) comb. nov.

Plate 17, figs. 1, 2

1938 Triletes dentatus Zerndt, pp. 22-27 .

1952 Triletes dentatus Dijkstra, p. 166, pi. 5, figs. 1, 2, 6.

1956 Superbisporites dentatus Potonie and Kremp, p. 135.

1958 Superbisporites dentatus Pierart, p. 59, pi. 4, figs. 1-7; pi. 5, figs. 1-6.

1959 Superbisporites dentatus Danze and Vigreux, p. 137.

Diagnosis. See Dijkstra 1952c, p. 166.

Remarks. Specimens were found with characters common to both ‘ Triletes dentatus ’
Zerndt and ‘ Triletes ramosus ’ Arnold, which partly supports Dijkstra’s view (1952c,
p. 167) that the two species are essentially the same. Maximum diameter of spore body
varies between 750 and 2,000 p (mean 1,150 p, after twenty-five specimens measured in
water). Laesurae are 125 to 250 p high, 60 p broad at the base. They become mem-
braneous towards the vertex. The details of the equatorial structure are best seen under

E. SPINNER: WESTPHALIAN MEGASPORES FROM THE FOREST OF DEAN 101

transmitted light. The fimbria forming the equatorial structure are 400 p or more long,
10 to 25 p wide, coalescing and bifurcating to form a continuous marginal rim. Small
dissections occurring between fimbria give a fenestrate appearance. The rim of the
flange is characterized by small protrusions 8 to 25 p long, 12 to 25 ^ wide with small
cone-like terminations, 2 to 6 ft in diameter. The size of the spore body, the height of
the laesurae and the width of the flange are similar to those of the specimens de-
scribed as ‘ Triletes ramosus ’ by Winslow (p. 33) and do not coincide entirely with those
of ‘ Triletes dentatus ’ Dijkstra (1952c, p. 166). This may be due in part to the state of
the specimens when measured. Dijkstra’s measurements are based on dry specimens,
Winslow’s on specimens mounted in Canada Balsam. Winslow (p. 144) recorded a
19 per cent, difference in size between specimens that were measured in water, allowed
to dry, and then remeasured. However, the processes on the rim of the equatorial struc-
ture are more similar to those of Superbisporites dentatus (Pierart 1958, p. 61) than
those described by Arnold (1950, p. 72, pi. 3, fig. 2) and Winslow (p. 33, pi. 8, figs. 3-6)
for ‘ Triletes ramosus The age of the specimens described here (Trenchard Seam,
Westphalian C-D) lies within the known stratigraphic range of Z. dentatus in Europe.
Occurrence. Small coal seam in Trenchard measures, Cinderford (loc. 5).

Zonalesporites rotatus (Bartlett) comb. nov.

Plate 17, figs. 3-5

1929 Triletes rotatus Bartlett, p. 21, pis. 9-12.

1954 Rotatisporites rotatus Potonie and Kremp, p. 163.

1956 Rotatisporites rotatus Potonie and Kremp, p. 135.

1959 Triletes rotatus Winslow, p. 32, pi. 8, figs. 1, 2.

1961 Triletes rotatus Arnold, p. 249.

Description. See Bartlett, 1929a, p. 21; Dijkstra 1952c, p. 166.

Remarks. This species has not previously been recorded in rocks younger than West-
phalian B (Dijkstra 1952c, p. 166). It seems more generally characteristic of Namurian
strata. The marginal rim of the equatorial structure of specimens recorded here as Z.
rotatus differs from the descriptions given by Zerndt (1937a, p. 9) and Dijkstra (1952c,
p. 166) for specimens from the Namurian. The width of the rim is 30 to 60 p, approxi-
mately twice that of the Namurian forms ( 1 5 to 30 p) ( Dijkstra, 1 952c) and the projections
from the rim are smaller, 10 to 20 ft high, and more scattered. Zerndt (p. 9) described
the processes as cylindrical and dentate, 12 to 60 ft high (usually 30 ft). These differences
may in future warrant specific or varietal differentiation, but an insufficient number of
specimens with well preserved equatorial structure has been found at the present time
to justify this move.

Stratigraphic Distribution. Lower Pennsylvanian, U.S.A. (Bartlett 1929, Cross 1947, Winslow 1959),
? Dinantian-Namurian C, Europe (Dijkstra 1952c), Westphalian B, Limburg, Belgium-Netherlands
(Dijkstra 1955, Pierart 1955).

Occurrence. Roof shale of Yorkley coal seam (loc. 14, 15).

Turma cystites Potonie and Kremp 1954
Genus cystosporites Schopf 1938

Cystosporites giganteus (Zerndt) Schopf 1938

Plate 16, figs. 5, 6

102

PALAEONTOLOGY, VOLUME 8

1930 Triletes giganteus Zerndt, pp. 71-79, pis. 9-11.

1950 Cystosporites giganteus Arnold, p. 87, figs. 157-160.

1951 Cystosporites giganteus Kalibova, p. 27, pi. 4, figs. 4-5.

1955 Cystosporites giganteus Dijkstra, p. 339, pars, pi. 44, figs. 48-50.

1956 Cystosporites giganteus Potonie and Kremp, p. 150, pi. 10, figs. 76-79.

1958 Cystosporites giganteus Pierart, p. 62, pi. 3, fig. 7, pi. 10, figs. 14-15.

1959 Cystosporites giganteus Winslow, p. 52, pi. 11, fig. 9, pi. 12, figs. 1-4.

Remarks. Of the fertile form, only five specimens were found intact. These varied
between 3,000 and 4,000 p in maximum diameter. Dijkstra (1946, p. 57) gave a range in
maximum size of 3,500 to 1 1,000 p. No apical prominence, as described by Chaloner
(1954 b, p. 30), was observed, the laesurae being relatively short in length (up to 200 p),
usually concealed by adhering abortive forms. The proximal part of the spore wall is
thicker (20 p) than the remainder and appears finely vermiculate to granulate under
transmitted light. The medial and distal portions of the spore wall are thinner, more
translucent, and consist of fibrils, 6 p wide, which coalesce and bifurcate to form the
meshwork that characterizes this genus.

A ffinities. Lepidocarpaceae (Potonie 1962, p. 132).

Stratigraphic Distribution. Europe, ? Dinantian-Westphalian D (Dijkstra 1946); U.S.A., Mississip-
pian, Michigan, Indiana, Pennsylvania, (Chaloner 1954); Illinois (Winslow 1959); Pennsylvanian,
Michigan (Arnold 1950), Illinois (Winslow 1959).

Occurrence. Trenchard (loc. 1-4) and Woorgreen (No. 2) (loc. 31) seams.

Cystosporites varius (Wicher) Dijkstra 1946

Plate 16, figs. 7-9

1934 Sporites varius Wicher, pp. 1973, 1974, pi. 8, figs. 3-4.

1950 Cystosporites varius Arnold, pp. 88-91, pis. 16-17.

1951 Cystosporites varius Kalibova, p. 28, pi. 2, figs. 1-6.

1956 Cystosporites varius Potonie and Kremp, p. 152, pi. 10, figs. 80-84.

1958 Cystosporites varius Pierart, p. 61, pi. 3, fig. 8.

1959 Cystosporites varius Winslow, p. 51, pi. 12, figs. 5-8.

1959 Cystosporites varius Danze and Vigreux, p. 137.

Remarks. Specimens found varied between 700 to 1,500 p in maximum diameter. They
are intermediate in size between the fertile and abortive forms as described by Dijkstra

EXPLANATION OF PLATE 17

All specimens X 50 under reflected light.

Figs. 1-5. Zonalesporites spp. 1-2, Z. dentatus (Zerndt) comb. nov. 1, Proximal surface; Trenchard
seam (loc. 5), FD/24. 2, Distal surface; Trenchard seam (loc. 5), FD/25. 3-5. Z. rotatus (Bartlett)
comb. nov. 3, Proximal view of spore body with flange removed; roof-shale Yorkley seam (loc. 15),
FD/26. 4, Proximal view of entire spore, oblique compression; roof-shale Yorkley seam (loc. 15),
FD/26. 5, Distal view of spore with almost entire flange; roof-shale Yorkley seam (loc. 15), FD/26.

Figs. 6-9. Valvisisporites spp. 6-7, V. auritus (Zerndt) Bhardwaj. 6, Proximal view; Coleford High
Delf seam (loc. 7), FD/27. 7, Lateral compression; Coleford High Delf seam (loc. 8), FD/27.
8-9. V. nigrozonales (Stach and Zerndt) Potonie and Kremp. 8, Proximal surface showing narrow
laesurae and small auriculae; Woorgreen seam (loc. 31), FD/28. 9, Distal surface; Woorgreen seam
(loc. 31), FD/28.

Palaeontology, Vol. 8

PLATE 17

SPINNER, Westphalian D megaspores

E. SPINNER: WESTPHALIAN MEGASPORES FROM THE FOREST OF DEAN 103

( 1946, p. 58) and Potonie and Kremp (1956, p. 152). The spore wall is smooth to finely
granulose, no reticulate ornamentation being present as in C. breretonensis Schopf
(Winslow 1959, pp. 50-51).

Affinities. Lepidocarpaceae. C. varius type spores have been obtained from Lepidocarpon bohdano-
wiczii (Bochenski) Potonie 1962.

Stratigraphic Range. Europe, Westphalian A-Stephanian (Dijkstra 1946, 19554); N. America, Lower
Pennsylvanian, Michigan (Arnold 1950); Illinois (Winslow 1959).

Occurrence. Trenchard to Woorgreen (no. 2) seams.

MEGASPORE EVIDENCE FOR AGE OF COAL MEASURES

The occurrence of Vcilvisisporites (Zerndt) Potonie and Kremp 1956 sensu Bhardwaj
1957 in the Coleford High Delf and seams above it confirms the Westphalian D age of
the measures as suggested by non-marine lamellibranch and macrofloral evidence.
Sen ( 1964, p. 99) has recently recorded Triletes auritus Zerndt from the Westphalian A-B
of the Ayrshire Coalfield. No description, remarks, or illustrations of the species were
given by this author, who followed Dijkstra’s (1946) ‘broad’ interpretation of megaspore
species. Dijkstra (1946, pp. 31-33) distinguished Triletes auritus Zerndt and Triletes
auritus Zerndt var. grandis Zerndt. Since V. auritus (Zerndt) Potonie and Kremp sensu
Bhardwaj is equated to T. auritus Zerndt var. grandis Zerndt it seems unlikely that the
species T. auritus referred to by Sen is the same species as recorded here.

An abundance of Tubercidatisporites brevispiculus (Schopf) Potonie and Kremp, and
Zonalesporites dentatus (Zerndt) comb. nov. suggests a lower Westphalian D or possibly
highest Westphalian C age for the Trenchard measures. Such an age is also supported
by the absence of Setosisporites praetextus (Zerndt) Potonie and Kremp, S. hirsutus
(Loose) Ibrahim, Lageniculci horrida Zerndt, and L. subpilosa (Ibrahim) Potonie and
Kremp, species which are common in Westphalian assemblages up to middle West-
phalian C (Dijkstra 19556, p. 6).

Acknowledgements. The author gratefully acknowledges the encouragement and advice of Professor
L. R. Moore, who first suggested the topic. He is also indebted to Mr. R. H. Wagner for supervision
and for the identification of plant macrofossils quoted in the present paper. Drs. H. J. Sullivan and
R. Neves kindly made available a first suite of samples and critically read the manuscript. Dr. W. G.
Chaloner offered most helpful advice on the redefinition of Zonalesporites. Further sampling was
greatly facilitated by indications from Messrs. A. Wood and A. Howells of the Crown Offices at
Coleford, Gloucestershire. Mr. G. S. Bryant kindly helped in the production of photographs. This
work was carried out while being in receipt of a D.S.I.R. research studentship.

LIST OF LOCALITIES

The map references given below are based upon the 2| inch to 1 mile topographical maps of the

Ordnance Survey of Great Britain. The localities refer to coal samples unless otherwise stated.

Trenchard seam. 1, Edenwall, Coleford, opencast site, SO. 51/582103. 2, Mapleford colliery, SO. 50/
595078. 3, Old Ten Acre and Norchard levels, tips, SO. 60/634050. 4, Howie Hill near Ross-on-Wye,
outcrop and tips, S062/200611. 5, Morse Lane, Drybrook, near Cinderford, SO. 61/641 176.

Coleford High Delf seam. 6, Berry Hill, Coleford, opencast site, SO. 5 1/5731 18. 7, Edenwall, Coleford,
opencast site, SO. 5 1/585 104. 8, Wells level, Coleford, SO. 5 1/568 134. 9, Hillersland colliery, near
Coleford, SO.51/569140. 10, Bream's cove colliery. Bream, SO. 60/61 1047. 11, Northern United
colliery, Cinderford, SO.61/637150.

104 PALAEONTOLOGY, VOLUME 8

Whittington seam. 12, Parking Hill No. 3 Level, Bream, SO. 60/616047. 13, Pillowell, opencast site.
SO. 60/631062.

Yorkley seam. 14, Worcester Lodge, near Coleford, coal and roof-shales, SO. 51/5961 15. 15, New road
level, Cannop, coal and roof-shales, SO. 61/6031 15. 16, Hamiltons level, Cannop, SO. 61/607125.
Brazil/y seam. 17, Steam mills, Cinderford, tips, coal, and shales, SO.61/607125. 18, Outcrop near Swan
Inn, Brierley, SO.61/626151.

No Coal seam. 19, Outcrop near Cannop colliery, SO. 61/61 1 128. 20, Outcrop near Cannop ponds,
coal and shales, SO. 61/61 1 107. 21, Nine wells near Blakeney, SO. 60/61 1082.

Churchway seam. 22, Outcrop near weir, Cannop ponds, SO. 61/612107.

Rockey seam. 23, Outcrop near Cannop colliery, SO. 61/61 1128.

Starkey seam. 24, Oaken hill railway cutting, near Parkend, coal and shale, SO. 60/63 1079. 25, Outcrop
near Cannop colliery, SO. 61/61 1 129.

Lowery seam. 26, Adit near Lightnroor colliery, coal and roof-shales, SO. 61/641 109. 27, Outcrop near
Cannop colliery, SO. 61/61 1 129.

Twenty Inch seam. 28, Morgan's adit near Swan Inn, Brierley, coal and roof-shales, SO.61/627152.

29, Delves inclosure, Brierley, SO. 6 1/642 124.

Crow De/f seam. 30, Collingwood level, Nelson green, tip, SO. 61/152632.

Woorgreen seam. 31, Adit, near Dilke hospital, Cinderford, SO.61/644155.

REFERENCES

Arnold, c. a. 1950. Megaspores from the Michigan Coal Basin. Contr. Mas. Pcilaeont. Univ. Mich. 8*
59-111, 18 pi.

1961. Re-examination of Triletes sitperbus, T. rotatus and T. mamillarius of Bartlett. Brittonia,

13, 245-52, figs. 1-5.

arber, e. a. n. 1912. On the Fossil Flora of the Forest of Dean Coalfield (Gloucestershire) and the
relationships of the Coalfields of the West of England and South Wales. Phil. Trans. Roy. Soc.
London , B 202, 233-81, pi. 1 1-13.

bartlett, H. 1929. Fossils of the Carboniferous coal pebbles of the glacial drift of Ann Arbor. Pap.
Mich. Acad. Sci. 9, 11-28, 25 pi.

bell, w. a. 1938. Fossil flora of Sydney coalfield, Nova Scotia. Mem. geol. Surv. Can. 215, 1-334,
pi. 1-107.

bennie, j. and kidston, R. 1886. On the occurrence of spores in the Carboniferous formation of Scot-
land. Proc. roy. Phys. Soc. Edinb. 9, 82-117, pi. 3-6.
bharadwaj, d. c. and venkata chala, b. s. 1962. Spore assemblage out of a lower Carboniferous shale
from Spitsbergen. The Palaeobotanist, 10, 18-47, pi. 1-10.
bhardwaj, d. 1955a. The spore genera from the Upper Carboniferous coals of the Saar and their
value in Stratigraphical Studies. The Palaeobotanist, 4, 118-49, 3 pi.

and kremp, g. 1955. Die Sporenfiihrung der Velener Schichten des Ruhrkarbons. Geol. Jb.

71, 51-68, pi. 4.

1957a. The Palynological investigations of the Saar Coals (Part 1 — Morphography of

Sporae dispersae). Palaeontographica, 101 B, 73-125, pi. 22-31.

1957b. The spore flora of Velener Schichten (Lower Westphalian D) in the Ruhr Coal

Measures. Ibid, 102, 1 10-38, pi. 23-26.

bochenski, T. 1936. Ober Sporophyllstande (Bliiten) einiger Lepidophyten aus dem produktiven
Karbon Polens. Ann. Soc. geol. Pol. 12, p. 193-240, pi. 2-7.

1939. On the structure of Sigillarian Cones and the Mode of their Association with their Stems.

Publ. Silesiennes Acad. Pol. Sci. Lett. Trav. geol. 7, 1-28, 11 pi.
butterworth, m. a. and millott, j. o’n. 1960. Microspore distribution in the Coalfields of Britain.
Proc. hit. Commit. Coal Petr. 3, 157-63.

chaloner, w. g. 1953a. A new species of Lepidostrobus containing unusual spores. Geol. Mag. 90,
97-100, pi. 2.

1953b. On the Megaspores of Sigillaria. Ann. Mag. Nat. Hist. 12, 6, 881-97, pi. 22.

1953c. On the Megaspores of four species of Lepidostrobus. Ann. Bot. (n.s.) 17, 263-93, figs. 1-23.

E. SPINNER. WESTPHALIAN MEGASPORES FROM THE FOREST OF DEAN 105

chaloner, w. g. 1954 a. Notes on the spores of two British Carboniferous Lycopods. Ann. Mag.
Nat. Hist. 12, 7, 81-91, figs. 1-7.

1 954/>. Mississippian megaspores from Michigan and adjacent States. Contr. Mas. Pa/aeont.

Uuiv. Mich. 12, 3, 23-35, 2 pi.

1958. Polysporia mirabilis Newberry, a fossil Lycopod cone. J. Palaeont. 32, 199-209, pi. 31-32.

crookall, r. 1930. Flora of the Forest of Dean Coalfield. Proc. Cotteswold Nat. Fid. Club, 23,
226-43.

1955. Fossil Plants of the Carboniferous of Great Britain. Mem. geol. Surv. G.B. (Palaeont.), 4,

1-84, pi. 1-24.

cross, a. t. 1947. Spore floras of the Pennsylvanian of West Virginia and Kentucky. J. Geol. 55,
285-308, 5 pi.

danze, .1. and vigreux, s. 1959. Distribution verticale des megaspores de l’assise de Bruay a Bruay.
Ball. Soc. bot. Fr. 12, 130-9.

dijkstra, s. 1946. Eine monographische Bearbeitung der karbonischen Megasporen. Meded. Geol.
Stickling , ser. C-III-I, 1, 1-101, 16 pi.

1952u. New Carboniferous niegaspores from Turkey. Ann. Mag. Nat. Hist. 12, 5, 102-4, 2 pi.

1952 b. Megaspores of the Turkish Carboniferous and their stratigraphical value. Rep. Int. Geol.

Congr. 18th sess. Great Britain (1948), 10, 11-17.

1952c. The stratigraphical value of megaspores. C.R. Congr. Av. Etud. Strat. carb. Heerlen (1951),

1, 163-8, pi. 5-7.

1955u. Megaspores carboniferas espanolas y su empleo en la correlation estratigrafica. Estad.

geol. Inst. Mallada, 11, 27-28, 277-354, pi. 35-45.

19556. The megaspores of the Westphalian D and C. Meded. Geol. Stichting, n.s. 8, 5-11, pi. 1-2.

1956u. Some Brazilian megaspores, Lower Permian in age, and their comparison with Lower

Gondwana spores from India. Ibid. 9, 5-10, pi. 1-4.

19566. Megasporas carboniferas de La Camocha (Gijon). Estad. geol. Inst. Mallacla, 12, 245-55

(Engl, summary 256—62), pi. 48-57.

dix, e. 1934. The sequence of floras in the Upper Carboniferous, with special reference to South Wales.
Trans, rov. Soc. Edinb. 57, 789.

felix, c. 1954. Some American arborescent lycopod fructifications. Ann. Mo. bot. Gdn. 41, 351-94,
pi. 13-18.

horst, u. 1955. Die Sporae dispersae des Namurs von Westoberschlesien und Mahrisch-Ostrau.
Palaeontographica, 98 B, 137-236, pi. 17-25.

hoskins, J. and abbott, m. 1956. Selaginellites crassicinctns, a new species from the Desmoinesian
series of Kansas. Amer. J. Bot. 43, 36-46, 27 figs.

kalibova, m. 1951. Megaspores of the Radnice Coal Measure Zone of the Kladno-Rakovnik Coal
Basin. S. Geol. Surv. Czechoslovakia, 18 (Palaeont.), 21-63, pi. 5-8.

1959. Rod Valvisisporites (Ibrahim 1933) Pot. & Kr. 1954 (Triletes auritus Zerndt 1930, ty 11

Zerndt) a jeho druhy v ceskem permocarbon. Vestnik U.U.G. 34, 429-36, 4 pi.
KOWALEWSKA-MASLANKiEWicz.ovvA, zofja 1932. Megasporen aus dem Floz ‘Elzbieta’ in Siersza.
Acta Soc. Bot. Pol. 9, 155-74.

moore, l. r. 1947. The sequence and structure of the southern portion of the East Crop of the South
Wales Coalfield. Q.J.G.S. 103, 261-300.

1954. The Forest of Dean Coalfield in Trueman, The Coalfields of Great Britain, Arnold, London,

126-33.

nowak, J. and zerndt, j. 1936. Zur Tektonik des ostlichsten Teils des Polnischen Steinkohlenbeckens.
Ball. Acad. Pol. Sci. Letts. A, 56-73.

pant, d. and srivastava, g. k. 1962. Structural studies of L. Gondwana Megaspores part I. Palaeonto-
graphica 1 1 1 B, 96-1 11.

pierart, p. 1955. Les megaspores contenues dans quelques couches de houille du Westphalien B et C
aux charbonnages Limbourg Meuse. Pabl. Ass. Etud. Paleont. Houill. 21, hors, ser., 125-40, pi. B-F.

1958. Palynologie et Stratigraphie de la zone de Neeroeteren (Westphalian C superieur) en Cam-

pine beige. Pabl. Ass. Etud. Paleont. 30, 23-102, pi. 1-18.
potonie, r. 1956. Synopsis der Gattungen der Sporae dispersae I Teil. Sporites. Geol. Jb. 23, 1-103.
1962. Synopsis der Sporae in situ. Ibid. 52, 1-204.

106

PALAEONTOLOGY, VOLUME 8

potonie, r., ibrahim, a., and loose, f. 1932. Sporenformen aus den Flozen Agir und Bismarck des
Ruhrgebietes. Neues Jb. Miner. Mb. 57, 438-54, 7 pi.

- and kremp, g. 1954. Die Gattungen der palaozoischen Sporae dispersae und ihre Stratigraphie.
Geol. Jb. 69, 1 1 1-94, pi. 4-20.

1955. Die Sporae dispersae des Ruhrkarbons, ihre Morphographie und Stratigraphie mit

ausblicken auf Arten anderer Gebiete und Zeitabschnitte, Teil 1. Palaeontograpliica, 98 B, 1-136,
pi. 1-16.

1956. Ibid. Teil 2. Ibid. 99 B, 86-191. pi. 17-22.

schopf, j. m. 1938. Spores from the Herrin (No. 6) Coal Bed in Illinois. Rep. Invest. III. geol. Surv. 50,
1-73, 8 pi.

1949. Research in coal palaeobotany since 1943. Econ. Geol. 44, 492-513.

, wilson, l. R., and bentall, r. 1944. An annotated synopsis of Palaeozoic fossil spores and the

definition of generic groups. Rep. Invest. III. geol. Surv. 91, 1-72, 3 pi.
schulze, f. 1855. Uber das Vorkommen wohlerhaltener Cellulose in Braunkohle und Steinkohle.
Ber. K. Acad. W iss. Berlin, 676-8.

sen, j. 1964. The megaspores of the Ayrshire Coalfield and their stratigraphic value. Micropalaeonto-
logy, 10, 97-104.

slater, l. and eddy, g. 1932. The significance of spores in the correlation of coal seams, Part II. The
Barnsley seam, south Yorkshire area. Part III, The Silkstone seam, south Yorkshire area. Pap.
Dep. sci. industr. Res. 23, 1—21.

, evans, m., and eddy, g. 1930. The significance of spores in the correlation of coal seams, Part I.

The Parkgate seam, south Yorkshire area. Ibid. 17, 1-28.
stach, e. and zerndt, t. 1931. Die Sporen in denn Flamm — Gasflamm — und Gaskohlen des Ruhr-
karbons. Gliickauf 67, 1118-24.

Stubblefield, c. j. and trotter, f. 1957. Divisions of the Coal Measures on Geological Survey maps
of England and Wales. Bull. geol. Surv. G.B. 13, 1-5.
trotter, f. 1942. Forest of Dean Coal and Iron Ore Field. Mem. geol. Surv. G.B. 1-95.
welch, f., trotter, f., et al. 1961. Geology of the Country around Monmouth and Chepstow. Ibid.
233 and 250, 1-164.

wicher, c. 1934. Sporenformen der Flammkohle des Ruhrgebietes. Arb. Inst. Paldobot. Petrog.
Brennst. 4, 165-212.

williams, r. 1956. The Sequence of Microfloras in the Coalfields of Southern Britain. Thesis, Univer-
sity of London (unpublished).

winslow, m. 1959. Upper Mississippian and Pennsylvanian megaspores and other plant microfossils
from Illinois. Bull. III. geol. Surv. 86, 7-102, 16 pi.

zerndt, j. 1930a. Megasporen aus einem Floz in Libiaz (Stephanien). Bull. Acad. Pol. Sci. Lett. B,
39-70.

19306. Triletes giganteus n. sp., eine riesige Megaspore aus dem Karbon. Ibid. B, 71-79, 3 pi.

- 1934. Les Megaspores du Bassin Houiller Polonais, partie I. Acad. Pol. Sci. Lett. Trav. Geol. 1,
1-55, 32 pi.

1937a. Ibid, partie II. Ibid. 3, 1-78, 241-78.

- 19376. Megasporen aus dem Westfal und Stefan in Bohmen. Bull. Acad. Pol. Sci. Lett. A,
583-99, 6 pi.

1938. Vertikale Reichweite von Megasporentypen im Karbon des Bassin du Nord. Ann. Soc. geol.
Pologne (Krakow), 13, 21-30, 13 pi.

zetzsche, f. and kalin, o. 1932. Eine Methode zur Isolierung des Polymerbitumens (Sporenmem-
branen, Kutikulen, u.s.w.) aus Kohlen. Braunkohle, 31, 345-63.

EDWIN SPINNER
Department of Geology,
University of Sheffield,

St. George's Square,
Sheffield 1

Manuscript received 21 February 1964

ON THE GENUS POTHOCITES PATERSON

by M. CHAPHEKAR

Abstract. Specimens of cones identified as Potliocites grantonii Paterson from compressed material are
described. In the light of the present findings, reasons are given for regarding Potliocites as the legitimate name
for all known cones of Archaeocalamites (Asterocalamites), including those from petrified material, which have
been previously described under the name Protocalamostachys Walton.

The cones of Archaeocalamites from compressed material have been noted and
described by most earlier investigators under the name of Potliocites grantonii Paterson.
Two species of petrified cones believed to have been borne on Archaeocalamites stems
have been described under the names Protocalamostachys arrctnensis (Walton 1949) and
Protocalamostachys petty cur ensis (Chaphekar 1963).

Walton (1949) compared Protocalamostachys arranensis with Potliocites grantonii and
concluded that the petrified and the compressed cones were probably similar in general
structure and size. However, due largely to the compressed nature of the fossils to which
the name Potliocites is given, little is known about their internal structure. Further
information concerning the structure has been obtained from the specimens of Potho-
cites described below.

Material. The following specimens from the Calciferous Sandstone Series of the Lower
Carboniferous were used in the present investigation :

(i) Three specimens from Loch Humphrey Burn, Kilpatrick Hills, Dumbartonshire,
Scotland, Kidston Collection (Geological Survey Museum) numbers 5373-5375.

(ii) Two specimens from Glencartholm, Eskdale, Dumfriesshire, Scotland, British
Museum (Natural History) Collection, numbers V195 and V758.

SYSTEMATIC DESCRIPTION
Genus pothocites Paterson 1841
1949 Protocalamostachys Walton

Emended diagnosis. Strobilus or fertile axis bearing non-alternating whorls of sporangio-
phores. Sporangiophore consisting of a main shaft bearing at the distal end four curved
pedicels each with a single terminal oblong sporangium orientated parallel to the sporan-
giophore shaft. Axis of strobilus containing a ring of mesarch xylem strands which do
not anastomose. Sporangium wall cells with fold or peg-like thickenings on the anti-
clinal walls. Spores of the Calamospora type.

Type species. Pothocites grantonii Paterson.

Other species. Pothocites arranensis (Walton) comb. nov.

Pothocites pettycurensis (Chaphekar) comb. nov.

[Palaeontology, Vol. 8, Part 1, 1965, pp. 107-12, pi. 18. |

108

PALAEONTOLOGY, VOLUME 8

Pothocites grantonii Paterson
Plate 18, figs. 1-6; text-fig. 1

Synonymy. See Kidston (1883).

Material. The identification of the cones described here as Pothocites grantonii is based
on their morphological similarity to Paterson’s (1841) original specimens as judged from
his illustrations and description and from the work of Kidston (1883) who re-examined
Paterson’s material. Unfortunately I have been unable to trace the type specimens at the
Royal Botanic Garden, Edinburgh (Kidston 1883). Although all fossil plants from the
Garden are believed to have been transferred to the Royal Scottish Museum, the Potho-
cites grantonii type material is not in the collections of that institution and is presumed
lost.

The specimens from the two localities are closely similar in general structure and size
(Table 1). The two specimens from Glencartholm, although poorly preserved, are as
long as 8-5 cm. and 1 1 cm. respectively, and show the characteristic segmentation of the
cone illustrated originally by Paterson (1841) and again more convincingly by Kidston
(1883). Kidston illustrated one specimen which showed evidence of whorls of leaves
or bracts at the constrictions.

The specimens from Loch Humphrey Burn are much better preserved than those from
Eskdale and have yielded good spores as well as certain information concerning the
internal structure.

Description. Specimens 5373 (PI. 18, fig. 1) and 5375 are longitudinally split and show
the axis of the cone with attached sporangiophores. Specimen 5374 (PL 18, fig. 2) shows
part of the outer surface of the cone on which the ends of sporangiophores with sporangia
arranged in crosswise manner can be seen. By measuring the diameter of the cone
and the distance between the ends of two adjacent sporangiophores in a whorl, it was
possible to calculate the probable number of sporangiophores occupying the circum-
ference of the cone (i.e. the number of sporangiophores per whorl); for cone specimen
5374, the number was twelve. Text-fig. 1 a shows part of the cone axis in which the coaly
material has split so as to expose what is probably one side of the vascular cylinder. If
the four ridges are assumed to represent four adjacent vascular strands, then the total
number of vascular strands in the axis may be calculated in a similar way to the number
of sporangiophores per whorl; this number is eight.

If the two specimens upon which these calculations are based are similar, and both
had eight vascular strands and twelve longitudinal series of sporangiophores, then eight
of the sporangiophores in each whorl were probably attached in pairs opposite vascular
strands and four attached singly. The sporangiophores are 2-5 mm. long and 0-25 mm.

EXPLANATION OF PLATE 18

Figs. 1-6. Pothocites grantonii Paterson. 1, Longitudinally split cone showing axis with whorls of
sporangiophores bearing sporangia; specimen 5373; approx. X 6. 2, Cone in surface view, showing
whorls of sporangiophores with sporangia arranged in crosswise manner; specimen 5374; approx.
X 5. 3, Part of the macerated sporangium showing group of spores adhered together (the inner
body in some spores has dropped out); slide 1, X 120. 4, Intact spore showing distinct trilete mark and
faint granulations of the inner layer; slide 2, x480. 5, Spore with outer layer only, showing trilete
mark; slide 3, X 480. 6, Freed inner bodies showing granulations very clearly; slide 1, X480.

Palaeontology, Vol. 8

PLATE 18

■ i

»«»v pirn's

JPlEjF'- vv<

HyT ■•3J

i

w2[V, ■ jyE «

%{'■ IB,

JNk- V-

J ~:f

CHAPHEKAR, Lower Carboniferous Pothocites

M. CHAPHEKAR: ON THE GENUS POTHOCITES

109

in diameter. These measurements were obtained from specimens 5373 and 5375. At the
distal end of the sporangiophore there are attached four sporangial pedicels which
terminate in sporangia. Text-figs, lc and Id, drawn from specimen 5374, show sporan-
giophores in surface (end) view. Portions of the pedicels can be seen attached to some of
the sporangia.

A number of sporangia were separated by immersing a small fragment of the cone
in hydrofluoric acid for one day. The acid was decanted off and the fragments washed
thoroughly in water. Some of the isolated sporangia also showed part of the attached

Table 1

Dinmptpr

Distance

between

Sporangiophore

whorls

Sporangia

Specimens

of cone

length

width

Spores

Polhocites grantonii
^,Z^).Loch Humphrey)
5375 j Burn'

8 mm.
6-5 mm.
7 mm.

1 -6-1 -8 mm.
1-5 mm.
1-7 mm.

2-6 mm.
2 mm.

0-8 mm.
0-7 mm.
0-8 mm.

82 p
82 p

j Glencartholm. j

7-5-9 0 mm.
8 mm.

15-1-8 mm.
1 -3—1 -5 mm.

2 mm.

0-75 mm.

104 p

Protoca/amostachys
ananensis
P. pettycuresis .

7-9 mm.
4 mm.

0-8 mm ( ?)
1 mm.

1 -2 mm.
0-9 mm.

0-65 mm.
0-4 mm.

66 p

38 p

pedicel (p) (text-fig. 1 g). As in Protoca/amostachys pettycurensis the pedicels are attached
to the sporangia on the side away from the main shaft of the sporangiophore. This is
seen well in a portion of specimen 5373 illustrated in text-fig. 1b. Further evidence of
this is seen in some of the isolated sporangia where a portion of the sporangiophore
stalk can be seen adhering to the sporangium wall on the side opposite the remnant of
the pedicel (text-figs. 1e, f). Small cellulose acetate ‘pulls’ were prepared from one
specimen in a region where some sporangia were exposed. The surface was first flooded
with acetone and then a cellulose acetate film was gently lowered on to it. The ‘pulls’,
which were removed after drying for about thirty minutes, pulled away fragments of
sporangial walls. The sporangial wall cells have ridges or thickenings on their anticlinal
walls (text-fig. 1h) similar to those found in Protoca/amostachys.

The spores were extracted by immersing some isolated sporangia in Schulze’s macer-
ating fluid for different periods ranging from 6 to 24 hours. The fluid was then removed
by thorough washing and the sporangia treated with dilute ammonia. Macerated
sporangia were mounted in dilute glycerine. The spores generally remained firmly in con-
tact with one another even after complete maceration. Some separation was obtained
by squashing or teasing with a needle. The spores vary in size from 72 p to 95 p.
The average of thirty-five individually measured spores from two separate cones was
82 p in each case.

The spore coat appears to consist of two layers, an outer brownish layer bearing a
small though distinct trilete mark and an inner yellowish layer or body with a faint

110

PALAEONTOLOGY, VOLUME 8

C

text-fig. 1. Pothocites grantonii Paterson, a. Part of the longitudinally split specimen 5373, to show
four vascular strands (vs), cortex (c), and sporangia (s); x 14. b, One sporangiophore ( sp ) from the
specimen in a, to show attachment of one of its four sporangia; p, pedicel; x 14. c-d, Surface view of
the distal end of the sporangiophore to show the crosswise arrangement of sporangia with parts of
attached pedicels (p); sp, sporangiophore; specimen 5374, x21. e-f, Abaxial and adaxial view of an
isolated sporangium to show the attachment of pedicel (p) in f, on the opposite side of the sporangio-
phore (sp) ; x9. G, Isolated sporangium with attached pedicel (p); x9. H, Surface section through

sporangial wall, x210. i, Spore, x 520.

M. CHAPHEKAR: ON THE GENUS POTHOCITES

111

trilete mark. The spores adhere together in the macerated sporangia (PI. 18, fig. 3).
The intact spore (PI. 18, fig. 4) shows faint granulations. These markings belong to the
inner layer, for in spores from which the inner body has dropped out, the wall is seen
to be quite without granulations (PI. 18, figs. 3, 5) and freed inner bodies show the
granulations very clearly (PI. 18, fig. 6). The two layers, at least after maceration, appear
to be free from one another and the inner structure appreciably smaller than the outer.
Thus when the outer layer ruptures the inner layer tends to fall out. The inner layer does
not withstand severe maceration as well as the outer. In samples which had been macer-
ated for twenty-four hours, the inner layer was very distorted and no trilete mark could
be discerned.

The interpretation of the apparently two-layered wall of these spores is difficult. It is
possible that it is an artifact due to the preservation or macerating treatment. Perhaps
the inner layer is comparable to the ‘inner body’ reported in over-macerated micro-
spores of the lycopod cone Porostrobus zeilleri (also a compressed cone) by Bharadwaj
(1959). The trilete mark as seen on the outer layer of the spore (PI. 18, fig. 4 and text-
fig. li) has rays about 18 p in length. It is smaller and less pronounced than in Proto-
calamostachys petty cur ensis.

The spores of these Pothoeites specimens can be regarded as being of the general
Calamospora kind only if the inner layer of the wall is neglected. It is, however, of interest
that Calamospora microrugosa (Ibrahim) Schopf (Playford 1962) has been reported as
sometimes having a minutely granulate exine; it is also of the same size as the Pothoeites
spores described here (62-104 p,; mean 82 p). The spores from specimen V195 were
somewhat larger in size, varying from 90 p to 115^. The average size of nineteen indi-
vidually measured spores was 104 p. This larger size could have been due to the swelling
of the very poorly preserved spores. The trilete mark showed only very indistinctly.

Discussion. The present investigations have shown that while in its general structure
Pothoeites grantonii is essentially similar to the two known species of Protocalamostachys,
it cannot be regarded as identical to either of these species. There is, however, no known
morphological distinction to separate it generically.

Only one morphological character observed in certain specimens of Pothoeites gran-
tonii has not so far been demonstrated in the petrified cones; this is the more or less
regular constrictions along the length of the cone at which whorls of leaves or bracts
were probably borne. The fact that no such constrictions have yet been reported in the
petrified cones seems likely to be due only to the fragmentary nature of these specimens.

Cones of the Pothoeites type have been found attached to leafy shoots of Archaeo-
ealamites radiatus (Bgt.) Stur [= A. scrobiculatus (Schloth.)] (Stur 1875) and cones of
Protocalamostachys pettycurensis have been reported in attachment to slender stems of
A. goeppertii (Chaphekar 1963).

There is thus good reason for classifying all these cones showing a similar morpho-
logical organization and belonging to Archaeocalamites under the name having priority,
i.e. Pothoeites Paterson, regardless of mode of preservation.

The cone originally named Bornia radiata by Renault (1893-6) is almost certainly
to be regarded sis a Pothoeites, in spite of Renault’s interpretation of the sporangiophores
as being truly peltate. From the apparent number of sporangiophores in the whorl, it
may be identical to Pothoeites arranensis.

112

PALAEONTOLOGY, VOLUME 8

The status of Nathorst’s (1914) Pothocitopsis bertilli is uncertain. Enquiries made of the
Swedish Museum of Natural History and of the Palaeontological Museum, Oslo, failed
to bring to light the original specimen. However, there is nothing in Nathorst’s descrip-
tion to provide any evidence that the cone was anything other than a poorly preserved
specimen of Pothocites. From its dimensions, it would probably be classifiable as
Pothocites grantonii.

Acknowledgements. I am indebted to the Trustees of British Museum (Natural History) and Geological
Survey Museum for the loan of material. 1 am grateful for the inspiration of Dr. K. L. Alvin, under
whose guidance this work was carried out. I should also like to thank Mr. A. Horne for the photo-
graphic work.

Location of Collection. The slides are preserved at the Geological Survey Museum, London.

REFERENCES

bharadwaj, d. c. 1959. On Porostrobus zeilleri Nathorst and its spores with remarks on the systematic
position of P. Bennhaldi Bode and the phylogeny of Densosporites Berry. The Palaeobotanist , 7, 67.
chaphekar, m. 1963. Some Calamitean plants from the Lower Carboniferous of Scotland. Palaeonto-
logy, 6, 408.

kidston, r. 1883a. On the affinities of the genus Pothocites, & c. Trans, bot. Soc. Edinb. 16, 28.

- 1 883/?. On the affinities of the genus Pothocites, &c. Ann. Mag. Nat. Hist. Ser. 5, 11, 297.
nathorst, a. g. 1914. Zur foss. Flora der Polarlander I: 4. Nachtr. Z. Palaeoz. Flora Spitzbergens, 77.
Stockholm.

paterson, r. 1841. Description of Pothocites grantonii, a new fossil vegetable, &c. Trans, bot. Soc.
Edinb. 1, 45.

playford, g. 1962. Lower Carboniferous microfloras of Spitsbergen. Palaeontology, 5, 550.
Renault, b. 1893-6. Bassin Houiller et Permien d’Epinac. Etudes des Gites Min. de la France, Paris.
stur, d. 1875-7. Die Culm-Flora. Abli. Kais. Konigl. Geol. Reichanst. 8.

walton, j. 1949. On some Lower Carboniferous Equisetineae from the Clyde area. Trans, roy. Soc.
Edinb. 61.

M. CHAPHEKAR

Department of Botany,
Imperial College,
London, S.W. 7

Manuscript received 14 January 1964

TIME IN STRATIGRAPHY

by T. G. MILLER

Abstract. Recent publication of several formalized systems of stratigraphic classification and nomenclature
provides an opportunity for the re-assessment of certain stratigraphic concepts.

The scope and categories of stratigraphic studies are examined and related to the nature of the chronological
record available in extant rocks. Lithostratigraphic data are identified as the matrix of a body of biostratigraphic
evidence which serves as the basis for construction of a model time-scale using biochronological divisions. The
notion of a separate ‘time-stratigraphy’ with ‘time-stratigraphic’ units is considered to be invalid.

The nature of stratigraphic boundaries in relation to diastrophic processes, the general nature of stratigraphic
divisions, and the special status of the biochronologic zone are discussed, together w ith problems of correlation.

The papers presented at a recent Symposium on Harmony in Stratigraphic Classifica-
tion (American Journal of Science , 1959) formed an important exposition of current
North American ideas and practice. It has been followed by an equally important
summary of the corresponding Russian views (Rotay 1960), which has been criticized
by Hedberg (1961). A more complete and formal statement of the American position
is contained in the Code of Stratigraphic Nomenclature (referred to below as the Code )
presented by the American Commission on Stratigraphic Nomenclature (1961) and
further developed in the Statement of Principles of the International Subcommission on
Stratigraphic Terminology (1961).

There is, in addition, a formidable literature of ‘theoretical stratigraphy’, the develop-
ment and elaboration of which has rested, for the greater part of this century, mainly
in the hands of American workers. This is hardly surprising in view of the enormous pro-
gramme of successful stratigraphic investigation carried out during this period, and the
great range and diversity of the problems involved. At the same time a programme
of similar scope has been undertaken by Russian stratigraphers, working on similar
problems and in areas approximately comparable with those of their American counter-
parts. Because of language and other difficulties of intercommunication much of the
Russian work seems to have proceeded in isolation. Jeletzky (1956) has already drawn
attention to some currently accepted over-refinements and complications in strati-
graphic theory, and to some defects and fallacies. Certain aspects of the subject have also
been briefly dealt with by Henson (1944).

It is possible that confusion in the interchange of ideas has arisen through a tendency
to allow the formulation of generalized concepts to be too closely influenced by practical,
and especially economically important, requirements. In the discussion that follows,
reference is made, wherever possible, to examples of contemporary usage.

THE STRATIGRAPHIC RECORD

It will perhaps be generally agreed that in all kinds of stratigraphic study the rocks,
whether already formed or in the course of formation, are the only source of evidence.
This body of evidence, and indeed stratigraphy as a scientific discipline, may be appre-
hended in either a chronological or a physico-geometrical sense. The first of these cate-
gories is concerned with the establishment of a true time-order of the actual events and

[Palaeontology, Vol. 8, Part 1, 1965, pp. 113-31.]

B 6612

I

114

PALAEONTOLOGY, VOLUME 8

processes leading to the formation of rock masses, and requires a consideration of time
both as age and duration of events and causal processes. The second category requires
the empirical description of rock masses in purely material, i.e. physical or geometrical,
essentially non-temporal, terms. It is a paradox that while the chronological aspect of
stratigraphy is interpretative or derivative, and hence strictly secondary to the visible
physico-geometrical aspect which is the actual matrix of the evidence, it must neverthe-
less be considered aetiologically primary, since the processes involved are causally
anterior to their material product.

If this dual nature of stratigraphic studies is accepted, it becomes necessary to con-
sider how ‘time-evidence’, whether of age (date), i.e. temporal location, or of duration,
i.e. elapsed time, is contained in or represented by the available documentary records,
namely the rocks.

The Code states (Article 2): ‘Categories of stratigraphic units are multiple. According
to different concepts and criteria, they comprise mutually overlapping but distinct types
of stratigraphic units. This Code provides regulations and recommendations relating to
(i) rock-stratigraphic units, (ii) soil-stratigraphic units, (iii) biostratigraphic units, and
(iv) time-stratigraphic units. The Code also treats two categories of units that are not in
themselves stratigraphic units but are closely related. These are (v) geologic-time units,
which are fundamentally related in concept to time-stratigraphic units, and (vi) geologic-
climate units, which are based on Quaternary stratigraphic units.’

The categories with which the present discussion is concerned are (i), (iii), (iv), and (v\
It is immediately clear that among these a difference of kind exists between, on the one
hand, notions of rock- and bio-stratigraphy — (i) and (iii) — dealing with material objects;
and, on the other, time-stratigraphy and geologic time — (iv) and (v). Time itself cannot
be said to accumulate, and can only be said, rather naively, to ‘exist’ as an instantaneous
dimensionless present, which ‘is’ a continuous transition from the observer’s past to his
future. However, since time as a phenomenon, is, so far as stratigraphy is concerned,
‘given’, omnipresent, continuous, rectilinear, and irreversible, so that the metaphor ‘the
passage of time’ is continually in use, there can be no advantage in trying to codify
‘geologic time’ differently from ‘ordinary’ time, with its artificial minute-hour-day-year
organization based on the behaviour of the earth as a planet. As more radiometric evi-
dence of age is collected, so the importance of ‘ordinary time’ fixed points in strati-
graphic analysis will increase. Neither does there seem to be any justification for setting
up a scale of ‘time-stratigraphic’ units purporting to be independent of others. The
Code , indeed, states (Article 26a) that the time-stratigraphic units proposed ‘. . . are
usually made to coincide with . . . some other kind of stratigraphic unit . . . which thus
serves as an objective reference’. It may be asked in this connexion whether any ‘time-
stratigraphic’ unit — a system or a stage, for example, could be recognized with any
certainty (except by analogy) in the absence of fossil — i.e. biostratigraphic — evidence?
Or whether, supposing such a unit to have been recognized either by analogy or on
radiometric grounds, its boundaries could be satisfactorily determined in the absence
of fossils?

Time and geological events

In a ‘normal’ sedimentation-denudation situation, various states or conditions of the
earth’s surface, whether subaerially exposed or concealed by water, in the instant of

T. G. MILLER: TIME IN STRATIGRAPHY

115

observation are isochronous or synchronous, considered stratigraphically. Records of
the surface form and composition made at arbitrarily chosen intervals would catalogue
and describe successive surfaces of identical age marking a temporal progression ana-
logous to the successive frames of a cinematograph film. However, in the denudation
sector there is systematic retrogressive destruction of surfaces that were once continuous
with preserved surfaces in the sedimentation sector, taking this to be undisturbed and
‘normal’. The record of isochronous surfaces is, therefore, potentially complete only
in the sedimentation sector. In that such a preserved pile of separate, complete, iso-
chronous surfaces would in effect ‘represent’ or ‘record’ the passage of time,* it may be
taken as an imaginary model of potentially complete elapsed time. This stratal pile,
or ideal succession, is the primary model or reference-frame in geochronological
stratigraphy. If the ocean basins are considered to be permanent features of the earth’s
surface, it is conceivable that such a complete succession might actually exist. But if it
does exist it remains at present inaccessible, while the visible successions are demon-
strably incomplete. However, since it is currently thought unlikely that the ocean basins
are permanent, it follows that a complete stratal record is also unlikely to exist.

In this connexion it should be noted that some kinds of igneous rock may become
effectively part of the sedimentational or accumulative record, while others will not.
The emergence at the earth's surface (whether subaerial or subaqueous) of extensive
masses of lava or pyroclastic material causes immediate modification of the surface.
Such newly formed igneous or para-igneous rocks then take their place as parts of
successive isochronous surfaces. On the other hand, intrusive bodies emplaced below
the surface cannot at the same time form part of that surface, although they may deform
it. Such bodies are, nevertheless, the result of finite and, theoretically, identifiable event-
sequences. It is a paradox that these bodies, whose position in a stratigraphic-age scale
is often difficult to fix closely, are commonly used radiometrically to provide ages-in-
years (the so-called ‘absolute ages’ of many stratigraphers) for their emplacement,
and hence an ‘age-bracket’ for their sedimentary envelope and cover. Similar
limitations on precision in stratigraphic-age fixing apply also to metamorphic and other
deformation-processes or event-sequences, although, as in many igneous complexes,
internal relative histories may often be derived from physico-geometric considerations
such as metamorphic grade, crystal condition, petrofabric, tectonic style, and so on.

Thus it emerges that only a limited part of the total geological record, namely the
sedimentary part sensu Into, can be used in attempting to provide a step-ladder or scale
having a general use as the main fabric of a stratigraphic model of elapsed time. The
primary task of the stratigrapher who wishes to establish a valid geochronology is there-
fore to isolate from the multiplicity of the sedimentary record the significant base-data
for a time-scale. Chronologically significant details must be extracted from the mass
of chronologically irrelevant material, much or even all of which may, nevertheless,
be utilizable in physico-geometrical analysis, or even, in some cases, in providing
support for a relative local ‘time-order’. Thus electric-log character, inorganic
lithology, mineralogy, and other purely physical attributes may provide circumstantial
evidence of local temporal order, but cannot refer directly to a real age-scale or general
chronology.

* The problems connected with the notion of the ‘passage of time’ are not discussed here, and the
notion is accepted uncritically for the present purpose as a convenient fiction (cf. Smart 1956).

116

PALAEONTOLOGY, VOLUME 8

Lithostratigraphic evidence as ‘ primary matrix ’

It is clear that among the many characteristics of the sedimentary rocks, those having
special relevance to time-order are the ones which retain a link with, or are a direct
manifestation of, the original causal processes of deposition and erosion. However,
deposition in the sense of simple inorganic accumulation in itself alone cannot provide
a systematically arranged series of identifiable steps capable of being used as an age-
scale model, since there is no way at present either of ‘signalling’ a unique age or of
telling how much elapsed time is ‘represented by’ a particular body of sediment or rock,
or what segments of elapsed time are not represented at all. Refined long-range correla-
tional analysis of the inorganic accumulative lithostratigraphic record is therefore diffi-
cult or even impossible. How could the Karroo succession, for example, be fitted to the
British stratigraphic column without fossil evidence? Nevertheless virtually all pre-
Phanerozoic and many other successions have to be analysed, faute de mieux, and
tentatively correlated, by reference to some non-systematic or historically ‘coarse’
scheme based on the pure lithostratal record, i.e. on (1) simple before-and-after relations
(‘simple’ here in the sense of primitive: the considerations actually involved, for
example, in the elucidation of a polyphase metamorphic complex, may be extremely
sophisticated); and (2) radiometric age-in-years determinations of contained minerals.
An example of this state of affairs is the gradual unravelling of the relations of the
Moinian and Dalradian rocks of Scotland and their tentative and partial correlation
with the Torridonian and certain Lower Palaeozoic formations.

Biostraligraphic evidence

Since the inorganic element in the available evidence is not yet able to supply geo-
chronological data adequate for refined and extended analysis, we are left with the
originally organic or biostratigraphic element as the main chronologically significant
body of evidence contained within the primary lithostratigraphic matrix.

The relevant principle in this connexion has been concisely set out by Teichert (1958,
p. 99): ‘The only natural processes that are of a unidirectional and non-reversible nature
and that leave universally occurring testimony in the rocks are radioactive decay and
organic evolution. The study of past life-forms and their distribution in the rocks
(palaeontology and biostratigraphy) provides a reference-system for determining the
order of succession of geological events from the Cambrian onwards. Radioactive decay
supplies an approximate (skeleton) reference-grid of dates in absolute [sic] terms.’

This admirably unequivocal proposition can be taken as the foundation upon which
to build a systematic frame of reference related to the passage of time and divisible into
more or less discrete parts, i.e. a true but still internally relative (and, at present, rather
loose) geochronology (cf. Jeletzky 1956). It is explicitly supported in the Code, Article
19/: ‘ . . . Commonly, biostratigraphic evidence is the most useful means for determining
time-stratigraphic boundaries, but criteria for defining biostratigraphic and time-
stratigraphic units differ fundamentally.’ Hedberg (1959, p. 680) also takes this position,
but with an important addition, when he writes: ‘The evolutionary sequence of fossils
may always be superior to any other means for geochronological dating of fossiliferous
sediments, but we already know that other means can contribute greatly to dating and
to time-stratigraphic correlation even in fossiliferous rocks ’ Earlier in the same paper

T. G. MILLER: TIME IN STRATIGRAPHY

117

Hedberg states (p. 676): . some workers have proposed that the same set of units

should be used for biostratigraphic divisions as for time-stratigraphic divisions, inferring
that fossil zones are the only time-stratigraphic units. . .

It will be noticed that in these extracts — as in the Code — reference is made to a cate-
gory of time-stratigraphic divisions, although this is not a feature of Teichert’s funda-
mental proposition, which deals only in biostratigraphic and radiometric terms. The
question therefore arises as to the place of this ‘time-stratigraphy’ in geochronological
analysis, and its relation (if any) to biostratigraphy.

THE ‘TIME-STRATIGRAPHY’ CONCEPT

The Code states, in Article 26: ‘A time-stratigraphic [chronostratigraphic] unit is a
subdivision of rocks considered solely as the record of a specific interval of geologic
time.’ This is somewhat extended in Remark (a): ‘. . . They are material units. Each is
the record of an interval of time that extended from the beginning to the ending of its
deposition. ... In actual practice, the scope of a time-stratigraphic unit in its type-
section or type-area is usually made to coincide with that of some other kind of strati-
graphic unit, such as a biostratigraphic or a rock-stratigraphic unit, which thus serves
as an objective reference. As time-stratigraphic units depend for definition on actual
sections of rock, care should be taken to define geologic-time units in terms of time-
stratigraphic units and not vice-versa.’

Time-stratigraphy has been discussed at some length by Wheeler and Beasley (1948),
Hedberg (1948), and Wheeler (19586); lithostratigraphy by Wheeler and Mallory (1956);
and biostratigraphy by Wheeler (1958<7) and Young (1960). Reservations have been
expressed by Rodgers (1959) and Story and Patterson (1959). It will be convenient to
take specific examples from some of these accounts:

(i) Hedberg (1948, p. 456); ‘The time-value of stratigraphic units based on fossils
will fluctuate from place to place with faunal-facies variation in much the same
manner as the time values of a lithologic formation may vary.’

(ii) Hedberg (1959, pp. 681-2): ‘The boundaries of biostratigraphic zones may cut
across time-horizons, across formation-units, and across the boundaries of any
other kind of stratigraphic unit.’

(iii) Wheeler (19586, p. 1048): ‘Analysis of time-stratigraphy . . . has led to the
observation that not all time-stratigraphic units are entities of constant temporal
value. Some regarded as most useful for regional synthesis and thus as bases for
historical interpretation occur as space-time variables.’

(iv) Wheeler (1959, p. 700): ‘. . . the criteria involved in delineating biostratigraphical
units . . . may not serve directly to delineate time-stratigraphical units.’

These examples exhibit certain inconsistencies in the understanding of the regions of
applicability and validity of the various stratigraphic categories under discussion, in
particular the ability of the undifferentiated stratigraphic record to signal identifiable
temporal location and specific duration.

One of the fundamental aspects used by Wheeler in developing the idea of ‘time-
stratigraphy’ as a discrete concept is the ‘complexly variable three-dimensional relation-
ship’ encountered in litho- and bio-stratigraphy, and the difficulty that only two of these

PALAEONTOLOGY, VOLUME 8

can be represented on a two-dimensional surface (Wheeler 1958h, p. 1047; cf. also Bell
et al. 1961). Time, in the sense of comparative duration, must in Wheeler’s view be shared
with, and to a greater or lesser extent correspond to or be involved in, stratal thickness,
as a ‘vertical’ dimension. Now it is clear that thickness in sedimentary successions does
not systematically represent temporal duration, although its origin in the familiar ‘law’
of superposition is fundamentally valid in a highly generalized way. It is also true that
there generally seems to be a relation between thickness and elapsed time, since the
only visible sign of duration is stratal thickness. This is the case, for example, in a regu-
larly and delicately laminated mudstone, where the observer feels intuitively that sedi-
mentary thickness, signifying accumulation at a definite rate, is a direct function of
duration. In the same way complete cyclothemic sequences may convey an even stronger
impression of ‘pure’ — i.e. continuous — recorded duration. But the presence of well-
marked bedding-planes, erosion surfaces, nodule beds, condensed deposits, even
abnormally well-sorted accumulations, must all indicate subtractions from the stratal
record. The need to account for these gaps in the record has led to the introduction of a
whole series of technical terms, e.g. lacuna, hiatus, erosional vacuity, &c. (cf. Bell et al.
1961, Sanders 1957). However, none of these inorganic phenomena is able to contribute
anything more than strictly local and circumstantial, and non-comparative, chrono-
logic information. The establishment of a chronologic scale depends upon the avail-
ability for investigation and analysis of progressive and irreversible change in relatively
easily identifiable entities either independent of the processes under direct examination
— as in the case of ‘ordinary’ time, with numerically ordered points and intervals
counted from a conventionally agreed initial datum point — or only secondarily asso-
ciated with them. The temporal arrangement of such associated or secondary pheno-
mena is capable of furnishing a relative time-scale whenever direct methods fail, or
whenever the primary events are transient (as they usually are), and incapable of pro-
viding permanent evidence. Recourse to this ‘coarse’ assessment of the passage of time
is the familiar standard method of the archaeologist and pre-historian, now, like non-
human stratigraphy, supplemented by radiometric techniques which can provide
scattered ‘age-in-years’ reference points.

The important point for emphasis here, in relation to stratigraphy, is the clear distinc-
tion that needs to be made between (u) the age of events, ( b ) time as duration of con-
ditions or processes, and (c) the recording of the passage of time. A time-scale is an
artefact for representing systematically and rationally (i.e. ordering) a succession of
events (themselves, it should be noted, also no more than hypothetical constructs) within
a temporal reference frame. Ideally such a time-scale would be independent of events,
in the way that our everyday conventional time-scale seems to us to be, but of course is
not, independent. A way out of this confusion is to regard events as simultaneous not
because they appear to occupy the same ‘moment of time’, but simply because they
happen together. ‘They correlate themselves because they co-exist. . . . [The] moment is
not a temporal entity existent in its own right, it is simply the class of co-existent events
themselves. We derive time from events , not vice-versa’ (Gunn 1929, p. 323, quoted in
Whitrow 1961, p. 36: my italics). Thus thestratigrapher’s time-scale has to be constructed
from a selection of traces of progressive and irreversible events or processes (secondary
organic) contained within a matrix of products of events which are effectively non-
progressive and potentially reversible, but whose temporal location it is intended to

T. G. MILLER: TIME IN STRATIGRAPHY

119

systematize. The primary events are the preserved relics of the ‘transfer process’ of
Wilson (1959) — removal+transport+deposition — now represented by the whole body
of lithostratigraphic evidence, and they are without strict chronological significance.
The secondary events whose traces are retained within the primary matrix constitute the
body of biostratigraphic evidence, and these, since they are the results of identifiable pro-
gressive change, have positive chronological significance. There is an additional body of
evidence within the primary matrix, namely the products of diagenetic effects. These are
in their turn capable of conveying some chronological significance, but this will nor-
mally be crude (simple before/after) and local; and, which is more important, it will
refer to a portion of elapsed time subsequent by an unknown amount to the original
depositional events.

The only meaningful time signals are therefore contained in the biostratigraphic body
of evidence. However, because of limitations in the discriminatory power of palaeonto-
logical analysis, not all of the organic content of a rock mass can be put to a chrono-
logical use. For example, the much-quoted ‘repetition’ of facies-controlled organic
assemblages is a repetition only in the sense that the evolutionary changes involved are
below the level of palaeontological detection. The effect of this limitation is important
in that it necessitates the separation of two kinds of biostratigraphic evidence according
to chronological significance. A zonal assemblage which ‘repeats’ in a facies-controlled
situation is of little or no chronological value. The assemblage which invariably appears
in the same sequential relation to precedent and following assemblages is the only
source of chronologically significant ‘signals’. The strata containing it should be distin-
guished as a biochronologic zone.

Thus it appears that the only presently available rational geochronological indices
are biostratigraphically based — i.e. biochronologic. However, if these biochrono-
logic indices are to be realized as rock-divisions, this realization must be in terms of
identifiable parts of their lithostratigraphic matrix, in order to make description and
comparison possible. The question therefore arises whether, and to what extent,
biochronologic divisions can be said to have meaningful and determinable lithostrati-
graphic expression.

THE NATURE OF STRATIGRAPHIC BOUNDARIES

It is true that very occasionally a stratal surface, or vertically limited band, can be
safely taken to represent an original sedimentation surface of short-term duration for
a comparatively large fraction of the total former sedimentation area. In other words,
it is a real ‘horizon’, and is really synchronous or isochronous. Examples are some lava
flows, bentonite seams (cf. Adams and Rogers 1961), possibly tonstein seams (cf.
Scheere 1954), thin, widespread evaporite sequences, and thin unique fossil bands like,
for example, the Saccocoma and Uintacrinus bands in the upper Jurassic and upper
Cretaceous respectively. Widespread glacigene ‘horizons’ may fall into the same
category, and, in non-fossiliferous or pre-Cambrian successions, these, together with
such other distinctive divisions as thin limestone bands of wide lateral extent, may be
taken as effectively synchronous within a rather ‘coarse’ geochronological context.

Such stratigraphically ‘two-dimensional’ horizons may with some confidence be re-
garded as chronologically ‘ instantaneous ’ surfaces, and may be represented in tabulations

120

PALAEONTOLOGY, VOLUME 8

or on maps as lines. They probably represent the highest level of geochronological pre-
cision at present available to the stratigrapher. However, most of the boundary lines
placed in stratigraphic tabulations, or drawn on geological maps, are of a much lower
level of precision, in that they are not virtually two-dimensional, but represent a more
or less extended transition from one lithological condition to another, i.e. involve a tem-
poral duration, or interval.

Inter-formational intervals can either be positive or negative in terms of sedimentation.
If sedimentation was continuous through the interval (the positive case), a separation
into neighbouring formations is seen only if a change of regime occurred such as to pro-
duce a change in lithology which is usually gradual. There is thus no dividing-line (sur-
face) between the two, but a transition which ‘represents’ at a given locality a certain
amount of elapsed time. It cannot legitimately be represented graphically by a line,
although of course it normally is. If, on the other hand, sedimentation was interrupted,
the interval will have a negative significance in terms of the accumulative potential, and
will normally be expressed visibly as a surface of non-depositional or erosional signi-
ficance, i.e. as the familiar ‘break in the sequence’, lacuna, hiatus, &c. Such surfaces may
legitimately be taken to be more or less satisfactory approximately two-dimensional
physico-geometric entities, and may be represented graphically by a line.

Nevertheless, an interval of this negative kind must still resemble its positive counter-
part in having a durational significance. The erosional interval has in reality two ‘sides’,
one the preserved residual upper limiting surface of the originally older sedimentary
regime; the other the initial depositional surface of the younger sedimentary regime.
Thus, while in the positive case the interval represents duration extending from the
initiation to the termination of change from lithology A to lithology Bin continuous sedi-
mentary sequence, in the negative case the interval represents the duration of the ero-
sional (or non-depositional) episode sandwiched between the sedimentation-states now
represented by lithologies A and B. Furthermore, in both cases, if a considerable area
is involved, the interval must be interpreted as the expression of more or less deep-seated
diastrophic processes, and thus may have varied systematically in age from place to
place, so that the actual durations involved may have been of similar length but different
mean age, or ages of starting and finishing. The geochronological significance of such
intervals thus requires investigation, since it is the ‘temporal origin’ of many, indeed
most, graphical boundary lines. It is clearly of the first importance to establish the precise
nature of these intervals, since practical utility and the necessity for communication
both demand the continual creation and recognition of boundary lines as reference
signs or markers in stratigraphic analysis realized as maps or tabulations, and especially
wherever economic considerations are involved.

Breaks in the succession

‘Breaks’ in the sedimentary stratal sequence indicate subtraction from the steady
supply of stratigraphic data, whether produced by the neutral condition of non-
deposition, or the true negative of erosion. At the present time the only mechanism
known to be capable of producing virtually simultaneous (synchronous) sedimentation
changes in widely separated parts of the world is a eustatic sea-level alteration caused by
relatively rapid variation in oceanic water volume or distribution. For this the controlling
factor is the expansion or contraction of the polar ice-caps. This would account for only

T. G. MILLER: TIME IN STRATIGRAPHY

121

a comparatively small change in sea-level, and probably, in the long run, only a small
interruption of major sedimentary regimes. Furthermore, since glacial events on a large
scale, although well known, appear to be rather rare in the stratigraphic record, these
glacigene eustatic changes must be considered of minor importance as break-producing
mechanisms.

However, a far more powerful mechanism, but capable of producing only non-
synchronous, or temporally progressive, sedimentation-changes, is available in the pro-
cesses of diastrophism or crustal deformation. Such deformation is expressed at the
earth's surface as major or minor form alterations (depth) of the ocean basins and their
margins, and similar form alterations (relief) of continental interiors and their margins.
These form alterations involve resultant radial changes of position of the surface relative
to the geoid, but are unlikely, except in the special cases of rifts, diapirs, cauldron sub-
sidences, &c., to be truly radial, and in consequence must be expressed as tilts or warps,
or tangential translations on various scales (cf. Hallam 1963). Tilts or warps are unlikely
to be produced instantaneously on a large scale, and their formation may therefore be
expected to have secular duration. Consequently, there will follow changes in the sedi-
mentary regime of the regions affected, and these may be severe enough to cause shifts
through the neutral state of non-deposition to the negative state of erosion, or vice versa.
But whatever may be the actual end-state of tilt-induced regiminal changes, inter-
regional temporal variation of their initiation, duration, and termination will normally
be present, and must lead ultimately to an expression as rock surfaces (whether de-
positional or erosional) having systematically variable age. Thus both sedimentation-
limiting changes, and non-depositional or erosional situations, will come to have variable
ages. This geologically familiar state of affairs (‘facies crossing time-lines ' auctt.) has for
long been known to British stratigraphers as diachronism (after Wright 1926), and con-
tinuous age-variable surface, or even whole lithostratigraphic units, formed in diastro-
phically controlled circumstances are called diachronous. Since diastrophic control is
ubiquitous, it seems likely that most positive lithostratal transitions, and negative sur-
faces or intervals, both of which may be used in one way or another as stratigraphic
boundaries, must be to some extent diachronous.

However, it is fortunate that normal stratigraphic successions do contain these inter-
vals or transitions, even though diachronous, since they allow fragmentation, albeit
quite arbitrary, of the local section into more or less compact lithostratigraphic divisions
(members, formations, groups, &c.). Moreover, where visible erosion surfaces, un-
conformities, re-work levels, condensed deposits, and so on are numerous, it is clear that
considerable subtraction from the stratigraphic record within divisions has also occurred.
But since diastrophic processes as the root cause of sedimentation changes must, as
we have seen, have temporal duration, and since variations in local and regional surface
configuration will produce varying reaction-times to structural changes, there is no
a priori reason to expect detailed sequential similarities at widely separated localities.
On the contrary, it is the observed detailed non-coincidence of such episodes that allows
the gradual filling in of stratigraphic gaps in successions by the collection of new
evidence not present in primary or type areas, but discovered by extension of investiga-
tion to regions of complementarity in the alternation of deposition and erosion. For
example: ‘. . . Possibly the explanation of the occurrence, in Central Arabia, of Ermo-
ceras and Magherina without Strenoceras, Garantiana, and Parkinsonia, infallible indices

122

PALAEONTOLOGY, VOLUME 8

of the Bajocian stage all over the world, is that the Ermoceras fauna lived in the interval
between the Middle and Upper Bajocian, which in many parts of Europe was a time of
earth-movement and erosion, and is always indicated by a break (non-sequence). This
was the period of the Bajocian denudation of Buckman’ (Arkell 1952, p. 296). Thus the
lithostratigraphic divisions of the stratigraphic column in one part of the world need
not, and in general will not, correspond to those in another, distant, part.

By contrast, general organic event-sequences, or the pattern of evolutionary change
as displayed in fossil assemblages, when expressed in terms of the available geochrono-
logical scale, in which the smallest operational division, the biochronological zone,
represents a duration in the order of 500,000 years, may reasonably be expected to show
a high level of coincidence in widely separated areas. When observed on this scale many
local and regional variations in rate of change of assemblages will be smoothed out and
cease to be significant as anomalies. Such a smoothed record accounts for the observed
high level of coincidence in the sequential order of geochronological divisions based on
globally distributed fossil groups such as many graptolites, ammonoids, fusulinids, and
agnostids. It is this coincidence which makes world-wide stratal correlation a practical
possibility. However, for refined regional analysis, which may use benthonic, even
sessile, rather than pelagic forms, contained within the limits of a single stage or sub-
stage, considerations of homotaxial rather than synchronous correspondence, migration
rates, arrival and extinction levels (cf. Bancroft 1945), and, pre-eminently, facies control,
will amost certainly be involved. Nevertheless, strictly correlative rock-units are still
only identifiable in biostratigraphic terms, however much circumstantial or supporting
evidence may be supplied by the lithologic matrix. It is probable that the tendency un-
critically to identify as ‘time-lines’ geochronologically non-significant physico-geo-
metrical horizons or surfaces has been a cause of confusion in stratigraphic thinking.
It must be kept clearly in mind that the only valid ‘time-lines’ are the real transitions or
intervals between divisions of the biochronologic scale, together with the rare syn-
chronous horizons already mentioned (supra, p. 120).

PERIODS, ZONES, AND STAGES

The view expressed by Bell and others (1961, p. 668) in a ‘Note for General Considera-
tion’ relating to the Code, that the period is the fundamental geochronologic unit, is open
to criticism, on the grounds that not only are they (the periods) no more than conceptual
units, but that they are based on lithostratal records (systems, &c.) which, being termi-
nated in most cases by physical breaks, are, ex hypothesi, incomplete. Moreover, the
boundaries between the periods are normally without real chronologic significance, just
as the ‘boundary’ between the eighteenth and nineteenth centuries is without real
significance in human history.

Historically, stratigraphic studies have gone forward by a progressive refinement, in
the form of multiple subdivision, of the whole stratal reference system or stratigraphic
column. Lyell, in the definitive third edition of his Principles of Geology (1834), shows
(vol. iv, appendix, pp. 305-14) the once-familiar division of the stratigraphic column
into Primary, Secondary, and Tertiary periods. These, in turn, are broken down into,
for example, Jura limestone group, Lias group, New red sandstone group, Carboniferous

T. G. MILLER: TIME IN STRATIGRAPHY

123

group (including Old red sandstone), and Greywacke group; and these again are shown
as consisting of ‘ Principal members’ (p. 309) at a variety of named localities.

By 1886, Jukes-Browne was able to put forward, in his Students' Handbook of His-
torical Geology, almost the modern sequence of systems, but continued to show, now
as ‘stages’, Lyell's ‘formations’, although of course in a greatly increased number. But
Jukes-Browne was already making the fallacious assumption that systems could be
subdivided into stages or groups, these into sub-groups, and the sub-groups into zones.
That this was, in fact, a fallacious procedure could have been deduced from the ex-
perience of Sedgwick and Murchison, who, in the middle quarters of the nineteenth
century, had been building up, respectively, the Cambrian and Silurian system, from the
central overlapping portion of which Lapworth (1879) eventually carved the Ordo-
vician.

Further, the non-primacy of the classical systems, and hence of the periods, in actual
practice, can be seen in the large number of cases of disagreement on the assignment of
certain formations, or parts of formations, or stages, to adjoining systems. Thus the
Tremadocian stage is, in Britain, conventionally placed in the Cambrian, because its upper
limit is commonly a surface of unconformity, while in Scandinavia it is placed in the
Ordovician (Stubblefield 19586, p. 4); the British Rhaetic (so-called, it may be noted, in
the absence of ammonite evidence) is placed in the Jurassic, whereas in southern Europe
it is placed in the Triassic; the British Downtonian was for long regarded as the top divi-
sion of the Silurian but is now made the lowest of the Devonian, and so on. In fact, as
stratigraphy developed, two different methods of chronological ‘labelling’ seem to have
become super-imposed: (i) an older, rather crude arrangement based partly on local
formational limits, frequently coincident with important local structural breaks, and
partly on a general assessment and understanding of the organic content; and (ii) a
newer, more refined, appreciation of the biostratigraphic sequence ideally independent
of structural interruptions, and concerned to produce a succession of biochronologic
zones, independent of lithology, complete for any given region, and by interdigitation,
overlap, or lateral replacement, eventually envisaged as complete, or potentially com-
plete, for the whole earth. For convenience of everyday reference, and more particularly
for inter-regional correlation, which may be forced to use fairly broad divisions, since
only these are recognizable at long range, the old period and system names have been
retained, even although they are capable of causing positive confusion by the disordering
of stratigraphic, and particularly biostratigraphic, observations (cf. Eames et al. 1962).
But as soon as more refined analyses and correlations are undertaken, the value, in terms
of both accuracy and precision, of the old labels decreases, and first the zone and then
the stage become the more useful and significant.

Status of the ‘ biostratigraphic ’ zone

The notion of the geochronologically significant biostratigraphic zone is fairly gener-
ally established in the minds of stratigraphic palaeontologists, especially in Europe,
almost to the extent of uncritical acceptance as something ‘given’. The definition
of Marr (1898), or a variant of it, is often quoted: ‘Zones are belts of strata, each of
which is characterized by an assemblage of organic remains, of which one abundant
and characteristic form is chosen as index.’ However, this definition is not sufficiently
close, and its potential ambiguity may be illustrated by reference to a paper already

124

PALAEONTOLOGY, VOLUME 8

mentioned (Young 1959, p. 755): . to me, if the fauna is recurrent, the zone is recur-

rent.’ The possibility of a properly defined (i.e. unique) bio chronologic zone recurring
in a vertical succession of rocks is directly opposed to the basic principles of correct
geochronological analysis.

Article 23 of the Code defines a concurrent range zone as ‘ . . . the overlapping ranges of
specified taxa, from one or more of which it takes its name’. Remark (6) states: ‘This
. . . is the zone generally recognized by stratigraphers when they use fossils in attempting
time-correlation of strata. Such zones are formal zones. Historically this usage derives
from Oppel.’ It is interesting to compare the Russian position as it is given by Rotay
(1960, p. 47): ‘The compass and boundaries of the zone are defined by the limits of the
extent of a definite grouping of widely distributed and preferably rapidly changing
organisms, constituting the zonal faunal (or floral) assemblage, which is not repeated
either in the overlying or underlying deposits. Into the content of each zonal assemblage
ought to enter as far as possible all the stratigraphically most important groups of fauna
(flora) represented in the given deposits. The extent of a zone embraces generally a whole
biogeographic region or province, less often a significant part of the latter; sometimes
a zone can be extended also through two or even several regions or provinces. To a zone,
distinguished in the deposits of this or that facial content, may be added also deposits
contemporaneous with it but of different facies, which are interbedded with palaeonto-
logically characterized deposits of the zone or directly replace it in its range.

‘. . . When necessary, one may introduce into the name of the zone the two or three
most typical species, distinctive in the whole region of the extent of the zone or character-
istic in various combinations in the different parts of that region. . . . For deposits that
are more or less contemporaneous but belong to sharply different biogeographic pro-
vinces (especially if these are developed in geographically widely separated regions) or
for successions that are sharply different in facies and stratigraphically not directly con-
nected (especially for subdivisions of synchronous marine and continental deposits),
separate schemes of zonal division can be applied.’

This Russian exposition is worthy of note in that it refers clearly and without am-
biguity to the ‘mutually exclusive’ situations in which possibly equivalent zonal schemes
contain few or no shared indicator elements in their organic assemblages.

Kinds of biochronologic zone

It is a convention to take sections in rocks of marine origin as major standards for
systematization and reference in the Phanerozoic Eon. In these standard successions there
are considerable advantages in making a distinction between those zones that are defined
in terms of benthonic assemblages and those defined in terms of pelagic assemblages.
The former may be expected to have a more restricted lateral geographic application,
being more closely facies-controlled, than the latter. The two kinds may be distinguished
as 6-zones and /7-zones respectively. In all stratigraphic analyses of inter-regional scope,
and ideally in all such analyses, the chronological status of the /7-zones will be higher in
terms of inter-regional utility, whereas the 6-zones will usually allow greater local refine-
ment. An example of this contrast is provided by a comparison of the standard grapto-
lite-based /7-zones of the British Ordovician Caradoc Series with the approximately
corresponding 6-zones based on brachiopod-trilobite assemblages proposed by Ban-
croft (1945). Two, and part of a third, /7-zones are considered by Dean (1958, 1960) to

T. G. MILLER: TIME IN STRATIGRAPHY

125

be equivalent to fourteen b- zones. Similarly, in the British Jurassic Cornbrash formation
two ammonite /7-zones correspond to four brachiopod 6-zones (Arkell 1956), and the thin
(less than 35 ft.) formation contains an inter-stage boundary.

Delimitation of zones

Stratigraphers have little direct evidence of the ways in which faunal assemblages
actually change and replace each other over long periods of time. In fossil assemblages
it is generally difficult to distinguish cases of real evolutionary acceleration from cases
of simple non-representation in the lithostratigraphic record. In a seemingly fairly
uniform and homogeneous formation like the English Chalk, morphological change in
some animal groups, such as the echinoids, may appear as a smooth sequence, while in
others it may be jerky and show discontinuities. Brinkmann’s (1929) work on the
(mainly) Callovian Oxford Clay, and its fauna of kosmoceratid ammonites, revealed in
an apparently unbroken clay succession a series of discontinuities, some of which
Cammonitenschlachtfelder’) were marked by concentrations of ammonite shells. The
time ‘represented by’ these diastems appears to exceed the ‘rock-recorded’ time.
Despite the occurrence of such diastems, and in view of the difficulty of interpreting
their ‘time-significance’, it is often impossible to decide exactly when one zone has
‘finished’ and another has ‘begun’, in lithologically homogeneous successions. Con-
versely, the presence of strongly developed and almost certainly diachronous dis-
continuities makes it impossible in many cases to be sure of the geochronological
significance of the physico-geometrical boundary actually observable.

An example of the difficulties commonly encountered is provided by Young (1959,
p. 753): ‘. . . the remaining zones are transitional from one to the next. The top of the
Monopleura-Toucasia zone is drawn at that horizon at which Caprinuloidea becomes a
more dominant fossil than the combination of Monopleura and Toucasia. This is a sub-
jective choice and the margin of error is probably plus or minus one and a half feet.’

The correct approach to the general difficulties of the manipulation and ordering of
observational data used in determining zones and stages has been set out by McLaren
(1959, p. 736): ‘. . . the presence of one or two species among the whole fauna indicates
that the palaeontologist is dealing with a zone previously recognized some distance away.
His knowledge of the evolution or time-range of these species may lead him to suggest
their age-equivalence to the previously-recognized zone. This procedure involves the
application of hypothesis (an “act of faith”) . . . and . . . is . . . dominantly subjective.

. . . Finally ... he may state that in his opinion the rocks are of Palaeozoic, or Devonian,
or late Cretaceous, or early late Norian age. This is purely subjective and depends on
application of a series of hypotheses. . . . Palaeontologic correlation is a difficult and
skilled discipline, requiring a high order of experience and judgment. . . .’

It cannot be too strongly emphasized that biochronologic zones are not entities
capable of determination by means of real boundary surfaces to which real ‘time value’
can be ascribed. Nor can zonal boundaries be said to exist, except in odd and very
exceptional surfaces, by themselves. Nevertheless, the orderly procession of zones has
real significance in geochronological analysis. The criterion for the validity of a zone
must be its regular and universal position in space above a precedent zone and below
a following one in a ‘sandwich’ pattern. The construction of the refined stratigraphic
column as we now understand and use it has depended on a leap-frog advance from one

126

PALAEONTOLOGY, VOLUME 8

sandwich relationship to another, progressively, from zones of the Lower Cambrian
to those of the Pliocene and Pleistocene. The recognition of the presence of a zone by
means of the fossils contained in rock strata precedes conceptually, and is more im-
portant than the pragmatic need to specify zonal boundary surfaces intended to delimit
the zone for purposes of reference or reproduction. The fossil evidence can only signal
the presence of ‘ rock-record ’ indicating a genera l age, i.e. a non-limited sector of elapsed
time, in the way that the sound of a human voice indicates the presence of a person
without being capable of indicating the shape or position of that person. It therefore
follows that the zonal sequence should almost always be in a state of incipient refine-
ment, as unsuspected gaps are filled in by the discovery of more complete successions.
It must also be borne in mind that the more completely the rock record seems to ‘account
for’ elapsed time the more difficult it must become to draw inter-zonal transitions. It
remains true of most stratal sequences that physico-geometrical surfaces are without
geochronological significance. Only where extremely sharp biochronological breaks
occur, or where several zonal assemblages follow each other in a much reduced thickness
of strata, can physico-geometrical surfaces be said to approximate fairly closely to the
condition of chronological significance, and even this must be regarded as strictly local.

The important principle to establish is the conceptual precedence and validity of the
biochronologic zone as an entity without rigidly determinable boundaries, as against
the artificial pragmatic requirement of dividing-lines on a geochronologic scale on maps
and descriptive tabulations.

Status of the Stage

In most Phanerozoic regional stratigraphic analyses the stage is now the basic division
for correlation and interpretation. It is necessary, therefore, to investigate the current
status and usage of the stage as a concept. Some recent examples of such usage are set
out below.

(i) Young (1960, p. 347): ‘Albert Oppel . . . gave the zone its present stratigraphic
concept [sic] by making it a subdivision of a stage.’

(ii) Storey and Patterson (1959):

(a) p. 709: ‘D’Orbigny’s stratigraphic zone is more familiar from its adoption
and refinement by Oppel as a subdivision of a stage.’

( b ) p. 719: ‘It is serious violation of stratigraphic principles to consider that a
regional unconformity occurs within a stage.’

(c) p. 715: ‘Zone, stage, series, system and group are not palaeontological units
and should not be confused with geological time units. The latter are bounded
arbitrarily by abstract time-planes which should not be confused with real
stratigraphic boundaries.’

(iii) Wheeler (1959, pp. 697—8) : ‘. . . boundaries between units of constant time-value
must be established arbitrarily, and those between time-variable units occurring
within maximum time-limits (such as system, series, and stage) are in part
arbitrary and in part natural. Configuration of all other space-time units is con-
trolled by natural phenomena such as deposition, non-deposition and erosion.’

From these examples it is immediately clear that the biochronologic zone is con-
sidered to be a subdivision of a stage. However, this was not the intention of Oppel,

T. G. MILLER: TIME IN STRATIGRAPHY

127

rightly regarded as the architect of the modern systematic zonal scheme, following, as
Jeletzky points out ( 1 956, p. 703), the pioneer work of D’Orbigny ( 1 842—9). Oppel ( 1856-8),
tabulating the zonal sequence in the European Jurassic rocks, set out three columns,
headed Formationsabteilungen, Etagen oder Zonengruppen , Zonen. The use of the word
etage is derived from D’Orbigny, who had recognized in the Jurassic rocks of France a
regular succession of ten distinct faunal assemblages. Oppel’s work, in effect, resulted
in an increase in the number of faunal assemblages that could be distinguished in the
Jurassic rocks, from ten to thirty-three. At the same time Oppel rearranged D’Orbigny's
terminology, so that the stages now became groups of zones, the zone being the primary unit.

It is from this historical base that Arkell (1956, p. 7) derived his formulation of the
relation between stage and zone: ‘Just as it is convenient to group together formations
into series,* so it is convenient to group like zones together and reduce the numbers
for practical purposes, and above all to have a grouping which enables several zones
to be correlated in a general way over long distances when the zones individually are
too precise. Such groupings of zones are stages.’ Later (p. 9) Arkell elaborates this defini-
tion: ‘As units of the single world scale of classification Stages must be based on zones.
As now used they are essentially groupings of zones, but they transcend zones both
vertically and horizontally. . . .’ Again (p. 11): ‘The possibility of describing and analys-
ing a geological system as a whole, all over the world, depends primarily on availability
of a single universal language for use in classification. This language the stages provide.

. . Finally (p. 9): ‘When a new fauna is found . . . not present or not detected at the
type locality, it falls readily into place if it comes between two zones already in the stage,
but if it falls at the boundary between two stages it has to be classed according to its
nearest palaeontological affinities.’

We may thus conclude that although the stage must be regarded as the most con-
venient unit for long-range correlation and, therefore, as the basic unit in inter-regional
stratigraphic analysis, it is not itself the primary biochronologic unit. This primary
quality clearly resides, as we have seen, in the biochronologic zone. It is possible that
confusion has arisen in the minds of many stratigraphers from the appearance, in the
usual setting-out of hierarchical geochronological organizational systems, of the stage
as ‘superior’ to the zone. But in fact this ‘superiority’ does not imply primacy. Just as,
in military terms, a battle-group or division may be (erroneously) considered as divided
into brigades, battalions, and squadrons, in fact such a group is made up of the subor-
dinate units, whose existence is anterior to the larger category, itself more arbitrary and
abstract than its components. Similarly, the biochronologic zone is anterior to the stage,
but, in Arkell’s words, is transcended by the more abstract conception of the stage.
‘Whereas the individual zone cannot be recognized beyond the area of occurrence of its
index species or typical fauna, a stage can be followed all over the world by a series of
overlapping correlations, and by the general grade of evolution of its critical fauna’
(Arkell 1956, p. 7).

* This use of series is, of course, wrong. The lithostratigraphic term used to unite a number of
formations is group.

CONCLUSIONS

From the several aspects of stratigraphic analysis discussed above it emerges that
geochronologically significant conclusions can only be reached by means of radiometric

128

PALAEONTOLOGY, VOLUME 8

or biological data. Physico-geometrical data (apart from radiometric) can do no more
than provide a crude local relative chronology, or circumstantial evidence in support
of a biochronologic framework. The historical development of stratigraphic studies
has led to the adoption of a set of time-indices which have acquired an appearance

I if ho. A

TOTAL

RECORD

partial record
at A

bio.A x

/

/

\ \
\

\

\

\i

partial record
atB

R

Q
P

e
d

c

bio.B

O

✓ ' /

. ' /b

/ /

/ /
/ /
/ /

/ /

/

/

/ /
/ /

/ /

/

B2

B,

text-fig. 1. Diagrammatic model to illustrate the geological time problem.

I if ho. B

of independence, but which are, in reality, biochronological. The notion of theoretically
identifiable ‘time-stratigraphic’ indices is considered invalid and is rejected.

In order to provide a visible illustration, and if possible a clarification of the geological
time problem, a diagrammatic model is set out in text-fig. 1, of which the centre column
represents an ‘ideal’ of separate, individually synchronous surfaces, in a sedimentation-
situation of unbroken, steady-rate accumulation. The fossils contained within such an

T. G. MILLER: TIME IN STRATIGRAPHY

129

uninterrupted succession are assumed to be reducible to zonal assemblages grading into
each other. Index letters are attached to successively distinguishable assemblages, and
it is further assumed that an upper and lower grouping on biological grounds, stemming
from a real developmental progression, permits the recognition of two stages, one with
zones b, c, d , and e; the other with zones P, Q , and R.

The two narrow columns immediately flanking the central one represent actual
sedimentation records at two separate localities A and B. The vertically ruled sectors
show positive episodes of preserved sedimentation, while the intervening blanks record
neutral (non-depositional) or negative (deposition-with-erosion) intervals. In each case
the appropriate fossils are assumed present in the preserved sectors, which are correctly
placed in relation to the central complete record.

The two outermost columns on each side show the actual lithostratigraphic and bio-
stratigraphic successions available to observers at A and B, i.e. they reproduce the inner
records, but closed up into a contiguous sequence, with diastems numbered Ax — , Bj
respectively, and lithologies indicated by conventional ornament. In terms of bio
chronologic zones and diastems the two successions are

at A at B

R R

A4-

Q Q

P (P missing) B3

A3 (e missing) e

d Bo

A2 d

d

c ( c missing) Bx

a4-

b b

It will be noticed that the two lithostratal successions are superficially similar, but in
terms of the unknown ideal succession of the centre column, only at the extreme top
and bottom (i.e. in the R and b zones) was sedimentation in fact proceeding at the same
time at the two places. Moreover, only three cases of ‘real’ inter-zonal transitions are
to be found, namely, c/d and P[Q at A, and Q/R at B. No boundary line can, of course,
be drawn for these. Where physical breaks (diastems) occur, and appear to coincide
with inter-zonal boundaries, as at b/c and 0/R at A; and at b/d , d/e, and e/Q at B, the
missing sectors cannot be estimated unless sub-zonal discrimination is possible. It is
therefore not permissible to regard such boundaries, identified at different localities, as
isochronous, unless the amount of interval is precisely measurable in biochronologic
terms. It will also be noticed that the inter-stage boundary (at A3 and B3) is clearly
identifiable at the two localities, but has different duration-significance in the two cases.

Considered in terms of the crude lithological successions, apart from the misleading
similarity of the two columns, there would be a strong temptation to draw an important
boundary line at the base of the uppermost lithological type, i.e. at diastems A4 and B3.

B 0612

K

130

PALAEONTOLOGY, VOLUME 8

While at B this ‘contains’ a true inter-stage boundary, at A it is chronologically con-
siderably above the base of the upper stage.

This model brings out the primacy of the biological evidence in geochronologi-
cal analysis, and shows that sequential division must be in terms of intervals , which
represent more or less temporal duration, rather than synchronous surfaces, which,
although imaginable, and ideally present, cannot at present be detected in sufficient
number to be practically useful.

Acknowledgements. I am indebted for valuable discussion of some of the problems of stratigraphic
analysis to Dr. B. K. Holdsworth, Dr. J. M. L. Lambert, and Mr. J. E. Thomas.

REFERENCES

adams, j. a. s. and Rogers, j. j. w. 1961. Bentonites as absolute time-scale calibration-points. Amu
N. Y. Acad. Sci. 91 , 390-6.

American commission on stratigraphic nomenclature. 1961. Code of Stratigraphic Nomenclature.
Bull. Amer. Assoc. Petr. Geol. 45, 645-65.

arkell, w. j. 1952. Jurassic ammonites from Jebel Tuwaig, Central Arabia. Phil. Trans. Roy. Soc.
B 633, 236, 241-313.

— 1956. Jurassic Geology of the World. Edinburgh: Oliver & Boyd, xiv+806.

Bancroft, b. b. 1945. The brachiopod zonal indices of the stages Costonian to Onnian in Britain.
J. Palaeont. 19, 181-252.

bell, w. c., kay, m., Murray, g. e., wheeler, h. e., and wilson, j. a. 1961. Report of Sub-Committee
of American Commission on Stratigraphic Nomenclature: Note 25 — Geochronological and Chrono-
stratigraphic units. Bull. Amer. Assoc. Petr. Geol. 45, 666-73.
brinkmann, r. 1 929. Statistische-biostratigraphische Untersuchungen an mittel-jurassischen Ammoniten
fiber ArtbegrifF und Stammesentwicklung. Abh. Gesell. Wissensch. zu Gottingen, M.-Ph. Klasse N. F.
13, 1-250.

dean, w. t. 1958. The faunal succession in the Caradoc Series of South Shropshire. Bull. Brit. Mus .
(Nat. Hist.) Geology, 3, 193-231.

■ 1960. The Ordovician trilobite faunas of South Shropshire. I. Ibid. 4, 73-143.

d'orbigny, a. 1 849-52. Coins elementaire de paleontologie et de geologie stratigraphique, 2. Paris.
fames, F. E., banner, F. T., blow, w. H., and clarke, w. J. 1962. Fundamentals of Mid-Tertiary Strati-
graphic Correlation. Cambridge: University Press, viii+151.
hallam, a. 1963. Major epeirogenic and eustatic changes since the Cretaceous, and their possible rela-
tion to crustal structure. Amer. J. Sci. 261, 397-423.
hedberg, h. d. 1948. Time-stratigraphic classification of sedimentary rocks. Bull. Geol. Soc. Amer .
59, 447-62.

1958. Stratigraphic classification and terminology. Bull. Amer. Assoc. Petr. Geol. 42, 1881-96.

1959. Towards harmony in stratigraphic classification. Amer. J. Sci. 257 , 674-83.

• 1961. The stratigraphic panorama. Bull. Geol. Soc. Amer. 72, 499-518.

henson, f. r. s. 1944. Stratigraphic classification and nomenclature. Geol. Mag. 81, 166-9.
international subcommission on stratigraphic terminology. 1961. Statement of Principles of
Stratigraphic Classification and Terminology (H. D. Hedberg, ed.) Rep. XXI Int. Geol. Cong. pt. 25,
1-38.

jeletzky, j. a. 1956. Palaeontology the basis of practical geochronology. Bull. Amer. Assoc. Petr .
Geol. 40, 679-706.

jukes-browne, a. j. 1886. Handbook of Historical Geology. London: Bell, xi+597.
lapworth, c. 1879. On the Ordovician system. Geol. Mag. 16, 13-14.
mclaren, d. j. 1959. Role of fossils in dating rock units. Amer. J. Sci. 257, 734-51.
marr, j. e. 1898. Principles of Stratigraphical Geology. Cambridge: University Press, 304.
oppel, a. 1856-8. Die Juraformation Englands, Frankreichs, und des siidwestlichen Deutschlands .
Stuttgart.

T. G. MILLER: TIME IN STRATIGRAPHY

131

rodgers, j. 1959. The meaning of correlation. Amer. J. Sci. 257, 684-91.

rotay, a. p. ( ed .) 1960. Stratigraphic Classification and Terminology (Second edition). Nat. Comm.

Geol. U.S.S.R. Moscow: State Publishing Office, 59.

Sanders, j. e. 1957. Discontinuities in the stratal record. Trans. N.Y. Acad. Sci. Ser. II, 19, 287-97.
scheere, j. 1955. Contribution a l’etude des tonstein du terrain houiller beige. Pnbl. Assoc, pour 1' etude
de la paleontologie et de la stratigraphie houillere , 19,
smart, j. j. c. 1956. In flew, a. g. n. (ed.) Essays in Conceptual Analysis. London: Macmillan, xi+265.
storey, t. p. and patterson, 3. R. 1959. Stratigraphy— traditional and modern concepts. Amer. J.
Sci. 257, 707-21.

Stubblefield, c. J. 1958. In wheeler, h. e. 1958 (discussion). Colloques internationaux du Centre
national de la recherche scientifique, 76, 15-23.
teichert, c. 1958. Biostratigraphic concepts. Bull. Geol. Soc. Amer. 69, 99-1 19.
wheeler, h. e. 1958tf. Primary factors in biostratigraphy. Bull. Amer. Assoc. Petr. Geol. 42, 640-55.
19586. Time-stratigraphy. Ibid. 42, 1047-63.

1959. Stratigraphic units in space and time. Amer. J. Sci. 257, 692-705.

and beasley, e. m. 1948. Critique of the time-stratigraphic concept. Bull. Geol. Soc. Amer. 59,

75-86.

and mallory, v. s. 1956. Factors in lithostratigraphy. Bull. Amer. Assoc. Petr. Geol. 40, 271 1-23.

whitrow, g. j. 1961. The Natural Philosophy of Time. Edinburgh: Nelson, xi+324.
wilson, j. A. 1959. Transfer, a synthesis of stratigraphic processes. Bull. Amer. Assoc. Petr. Geol.
43, 2861-2.

wright, w. b. 1926. Stratigraphic diachronism in the Millstone Grit of Yorkshire. Rep. Brit. Assoc.
Adv. Sci. 354.

young, k. 1959. Technique of mollusc zonation in Texas Cretaceous. Amer. J. Sci. 257 , 752-69.

• 1960. Biostratigraphy and the new palaeontology. J. Pa/aeont. 34, 347-58.

T. G. MILLER

Department of Geology,
Keele University,

Manuscript received 10 February 1964 Staffordshire

ENVIRONMENTAL CAUSES OF STUNTING IN
LIVING AND FOSSIL MARINE BENTHONIC
INVERTEBRATES

by A. HALLAM

Abstract. The various environmental causes of stunting or dwarfing in marine benthonic invertebrates are
reviewed for living species and an attempt made to apply the results to cited instances among fossils. Particular
attention is devoted to possible hazards of interpretation, and the criteria for distinguishing between stunted
adults and juveniles among fossils is outlined. It is argued that the principal factors involved apart from food
supply are the salinity, oxygen content, turbidity, agitation, and temperature of the sea water, together with
population density. Palaeontological and sedimentological criteria for the distinction of these factors are pro-
posed, but it is concluded that information on the primary factor, food supply, will continue to remain elusive
to palaeoecologists.

The stunting of growth among species of marine benthonic invertebrates by adverse
environmental influences is potentially a most valuable tool in palaeoecological research,
since it can provide clues to the physical environment which might otherwise be over-
looked and give information on the tolerances of extinct forms. It has furthermore, an
important bearing on matters of taxonomy. Unfortunately few subjects in palaeoecology
are so beset with difficulties and confusion. Although there is an extensive literature,
little critical analysis has been undertaken. Many authors have neglected to provide
even simple measurements of fossils or criteria of maturity, thereby failing to exclude
the possibility that the supposed stunted adults were actually juveniles. The sedimentary
matrix has rarely been described in detail and the many different interpretations of the
cause of stunting inadequately supported by reference to living animals. This should
not always be held to reflect adversely on the palaeontologists concerned. Evidently in
many instances stunting was of incidental interest and some worked at a time when less
rigour was demanded and when the subjects of marine ecology and sedimentology were
in their infancy.

Ager (1963) has preferred the term ‘stunting’ to the more widely used ‘dwarfing’ for
species whose growth had been arrested or hindered by environmental influences, since
the latter seems to imply a genetic control. Hence ‘dwarfed organism’ could best be
applied to a small minority of monstrosities in an otherwise normal population, whereas
environmental stunting may affect whole populations. It could be argued that the
meaning of the term has in fact usually been made clear in context as signifying environ-
mental control, but the author agrees with Ager that ‘stunting’ is preferable, since
dwarfing suggests an extreme condition, the size being much below the normal for the
species. It is certainly debatable in the present state of knowledge whether organisms
of this kind are at all common in nature, but stunting in the sense of a perceptible
reduction in size below the species optimum is widespread at the present day and should
be a common condition among fossils. We have to bear in mind, however, that the litera-
ture on fossils is concerned primarily with forms whose size is considerably less than
normal since these are the more striking.

[Palaeontology, Vol. 8, Part 1, 1965, pp. 132-55.]

A. HALLAM: ENVIRONMENTAL CAUSES OF STUNTING

133

Certain other terms have sometimes been used. Such words as dwarf, diminutive, and
micromorphic merely signify small size and should be abandoned in the context under
discussion, while the term depauperate is obscure and unfamiliar, as Ager pointed out.

The observational recognition of sexual maturity should not be difficult for most
living organisms but is not directly possible for fossils and one must here proceed by
reasonable inference. The distinction of adults and juveniles is a necessary first step in
the proper study of such faunas. If stunted fossils are to be used for the elucidation of
conditions in the past, it is necessary to assume a correlation between size and environ-
ment, and that the stunting is not directly inherited (although in the long term mutants
resulting in dwarfs might be favoured by natural selection). This seems more reasonable
than the Lamarckian alternative on grounds of the relationship of size to food consump-
tion, to be discussed later, and is supported by the experiments of Coe (1942) on the
gastropod Crepidula. He found that if the objects for settlement were small the organisms
remained stunted throughout life but if their offspring were transferred to experimental
bottles they quickly grew to a size twenty times or more that of their parents. In the
dwarfs the cells and eggs were smaller in number than average but of normal size. It is
true that Ford (1928) recognized genetic control of growth in the amphipod Gammarus
but this served merely to reduce the growth-rate and not the size at maturity.

Experiments performed by Crabb (1929) on the pond snail Lymnaea stagnatis appressa
demonstrated clearly that while extreme crowding markedly retarded growth the indivi-
duals rapidly reached normal size after being transferred to standard conditions.

The influence of particular environmental factors on the size of marine invertebrates
is not always easy to access because very few controlled experiments have been performed.
One has to rely largely on inferences based on observations in a complex multivariate
system, not often well documented quantitatively. It may be difficult in consequence
to estimate the relative importance of different variables or isolate the significant factor.
Thus Moore (1958) has pointed out that the small size of organisms in stagnant water
might be due either to oxygen deficiency or to an absence of currents supplying food.
Furthermore, not all variables are independent. For example, low temperature can
reduce the salinity tolerance of estuarine species (Moore 1958). Because of such con-
siderations as these, and because each species has its own environmental optimum, it is
pointless as yet to seek any precise quantitative relationships from the data available.
Nevertheless one can perceive a number of qualitative relationships which are likely to
be of general validity, for the past as for the present.

Though this review is concerned essentially with marine organisms a few examples
relating to freshwater forms are cited where the data appear to be relevant. It has also
been found convenient to include ammonoids among the benthos.

FACTORS AFFECTING THE SIZE OF LIVING ORGANISMS IN

THE SEA

Food supply. An obvious relationship exists between body growth and the intake of food
so that it is hardly surprising that marine ecologists lay stress on the importance of this
factor. If food supply is low but adequate for survival then growth will be stunted, as is
apparent enough in our own species. It is desirable to make a distinction between the

134

PALAEONTOLOGY, VOLUME 8

availability of food and the capacity of the organism to consume it. Food availability
as a factor is most clearly observed among animals with an intertidal distribution.
Kristensen (1957), Savage (1956), and Hancock and Simpson (1961) have all recorded
a direct relationship between the growth of the suspension feeders Cardium edule and
Mytihis edidis and the period of immersion, that is, the time available for feeding.

It is not clear at present, however, whether marked variation in food supply is a
particularly important factor below low tide mark, since Fox’s (1957) calculations sug-
gest that generally speaking there is more than enough food available for the larger
organisms in the sea, and Kristensen (1957) has estimated that current velocities of only
a few centimetres per minute should be sufficient to ensure enough food for suspension
feeders. On the other hand. Dr. H. B. Moore has called the author’s attention to un-
published work demonstrating much slower growth of Echinocardium below than at low
water, and he himself has found much smaller Echinus in deeper than in shallower
water. In both cases he is inclined to believe that deficiency in food supply is the signi-
ficant factor.

A gradual decrease in the size of benthonic organisms with depth of sea may possibly
result from the lesser availability of food in deep water (Yonge 1961), which is probably
a consequence of the progressive destruction of dead plankton on its long slow descent
to the sea bottom (Emery and Rittenberg 1952), and the greater distance from river
mouths. This is hardly relevant, however, to the rich populations of shallow seas, which
have had the most attention and are geologically the most significant. Food supply in
this case might diminish within areas of restricted circulation, where the current activity
is inadequate for replenishment. Also important is the availability of suitable nutriment
for organisms with specialized feeding habits. Thus Moore (19366) found that the gastro-
pod Nucella (= Purpura) lapillus attains a greater size at maturity on a diet of Mvtilus
than on Balanus.

Stunting may often result from a reduced capacity to feed because of an abnormal
environment. For example, oysters and mussels may remain closed and cease feeding
when conditions are adverse. Mytihis edidis can survive for several weeks in the absence
of oxygen but does not feed (Moore 1958). The actual consumption of food is more
important than its availability and is obviously the prime factor controlling size
variations.

Salinity. Numerous reports in the literature suggest that abnormal salinity is one of the
most potent factors inducing appreciable size reduction in marine species. Though most
of this literature relates to molluscs it is apparent that a wide variety of invertebrate
groups is affected, including foraminifera (Pack 1919, le Calvez 1951), ostracods (Barker
1963), amphipods (Spooner 1947), coelenterates (Rawlinson 1934, Segerstrale 1957) and
echinoderms (Segerstrale 1957), so there is good reason to regard the phenomenon as
general. Gunter (1947n, b) considered that salinity is by far the most important factor
in producing size gradients in animals over a small area.

The most fully studied areas where the salinity differs appreciably from that of normal
marine waters are estuaries such as the Tamar Estuary in south-west England (Milne
1940) and marine gulfs or inland seas where freshwater inflow from rivers and precipita-
tion exceed evaporation. Examples include the Baltic Sea (Goldring 1922, Segerstrale
1957, Sorgenfrei 1958), the Black and Caspian Seas (Goldring 1922), the Texas coast

A. HALLAM: ENVIRONMENTAL CAUSES OF STUNTING

135

bays (Gunter 1950, Ladd et ah 1957) and the Gulf of Pechili, off the Yellow Sea (Grabau
1931). All these areas are characterized by brackish water, of comparatively stable
salinity in the partly isolated seas and gulfs but fluctuating considerably in salinity with
tidal action in the estuaries. The marine animals that can tolerate these low salinities,
dominantly species of lamellibranchs and gastropods, are frequently stunted, while
stenohaline groups such as coelenterates, brachiopods, echinoderms, and cephalopods
are absent or occur only sporadically. As opposed to this common restriction in size
and variety, individuals are often numerous. It should be remembered that certain ani-
mals such as Mya and Scrobicularia are well adapted to brackish water and may attain
their greatest size there.

The interesting experiments of Bradshaw (1957) on the foraminifer Ammonia beccarii
tepida have yielded an apparently anomalous result, in that specimens reared in water
of abnormally low or high salinity grew to slightly greater size than in normal salinity,
as a result of delayed maturation.

Whilst a direct relationship between size and reduced salinity is apparent, quantitative
data are sparse and difficult to interpret in the absence of more detailed descriptions of
the environment. The difficulties are well illustrated by the work of Boettger (1950) on
molluscs in a low-salinity inlet of the Baltic. He found no straightforward relationship
between size and salinity because of the intervention of other variables such as the
oxygen content of the water. These were not, however, fully evaluated. Sorgenfrei (1958)
pointed out the contrast between an upper brackish and a deeper more saline zone in the
Baltic waters, suggesting that the size of benthonic animals might vary to some extent
with the depth of the sea floor. Moore (1936u) observed that the barnacle Balanus
balaiioides benefits so greatly from the increased quantity of food suspended in estuarine
water and from currents bringing the food within its reach that it actually grows faster
and larger in some estuaries than in the open sea. Barnacles can be used to illustrate
yet further complications. Balanus improvisus in England was a characteristic species of
brackish water, but in the Florida region it is confined to the sea, never penetrating
estuaries, and B. eburneus takes its place in brackish waters (H. B. Moore, personal
communication).

In the great majority of cases ecologists studying such low-salinity regions have been
confident that the organisms they were dealing with were genuinely stunted adults. This
is most obvious with sessile or semi-sessile forms but with actively vagrant animals the
situation may be different. Gunter (1950) struck a cautionary note in his description of
certain swimming crustaceans in the Gulf of Mexico. Many species grow up in the low
salinity bays in the summer months and migrate into more open waters during the follow-
ing autumn and winter. The young forms seem in fact more tolerant of low salinity than
the adults. A palaeoecologist could perhaps misinterpret such regional size variations
though in this particular instance the crustaceans would probably not be fossilized.

Though most of the literature refers to brackish waters there is evidence that stunting
can also be induced by high salinity. Andrews (1940) observed that where evaporation
raised the salinity of ponds isolated from the sea, populations of the gastropod Neritina
virginea increased in numbers but diminished in size. Experiments by Pack (1919) on
ciliates demonstrated that growth was retarded in more saline water. Many years ago
Bateson ( 1 889) made a careful study of depressions such as the Shumish Kul which were
formerly connected with the Aral Sea. The depressions dried up progressively during the

136

PALAEONTOLOGY, VOLUME 8

Quaternary. Successive stages of evaporation were correlated with size reduction in
Cardium edu/e, associated with thinning of the shell and minor changes in shape.

As opposed to this evidence, it was found during the Cambridge expedition to the
Suez Canal in 1924 that only the foraminifera were stunted in the hypersaline Bitter
Lakes and that several species of lamellibranchs and crabs attained the same size as in
the sea (Fox 1929). It is to be observed, however, that the lamellibranch species cited are
oysters and mussels, both of which have many euryhaline representatives and can pre-
sumably adapt as well to hypersaline as to brackish-water conditions. Of especial interest
was the observation that a species of holothurian actually grew larger in the Bitter Lakes
than in the sea, thus contradicting the general rule that echinoderms are stenohaline.

The actual cause of stunting in water of abnormal salinity appears to have received
little attention. It could perhaps be a consequence of the physiological difficulties asso-
ciated with osmoregulation, demanding a higher proportion of the organisms’ energy
and so leaving less for the intake and absorption of food. Alternatively organisms may
cease feeding for long periods.

Temperature. The relationship of water temperature to the size of benthonic inverte-
brates is complex and frequently difficult to evaluate (Boni 1942, Tasch 1953, Moore
1958). In a review by Wimpenny (1941) it was pointed out that there is a general tendency
for the size of both terrestrial and marine animals to increase with latitude, that is, with
decreasing temperature, but the marine examples cited are not benthonic. Significantly,
important exceptions include molluscs with calcareous shells, as exemplified by the giant
gastropods and lamellibranchs of tropical seas. This may relate partly to the greater
ease of precipitation of calcium carbonate in warm water. On the other hand, Coe and
Fox (1944) stated that, generally speaking, lamellibranch species grow to greater sizes
in higher latitudes. The same may be true of many foraminifers, and Bradshaw’s (1957,
1961) experiments have thrown new light on this subject. It seems doubtful if there is any
general rule for benthonic animals and each species should be considered on its merits and
its optimum environmental conditions determined (Moore 1958).

The complexity is probably due largely to the interaction of opposing factors.
Metabolic rates normally increase with temperature, for example in the Mytilidae, as
measured by the rate of oxygen consumption (Thorson 1957), and in Balanus , as indicated
by the rate of beat of the cirri (Moore 1958). Growth will be more rapid in warmer water,
as is evident enough from the pronounced seasonal variations in the growth-rate of such
well-known lamellibranchs as Cardium edu/e and Mytilus edu/is. Coe and Fox (1944)
found that the annual increment in length of certain lamellibranchs tends to be greater
in lower latitudes because of the longer season available for rapid growth. As opposed
to this, maturation is often later in higher latitudes so that the time available for growth
during the whole life of the organism is longer (Wimpenny 1941, Coe and Fox 1944; see
also Phleger 1960 and Bradshaw 1961).

Regional variations of size in relation to temperature are generally slight and often
not readily recognizable (Gunter 1947 b, Kristensen 1957). Temperature does not there-
fore seem to be a particularly significant factor in stunting, at least at the species level.

Oxygen content. Aquatic invertebrates differ considerably in their oxygen requirements,
and experiments by Fox and Taylor (1955) on species of molluscs, worms, and crusta-
ceans showed that some actually lived longer, and the young grew bigger, in water that

A. HALLAM: ENVIRONMENTAL CAUSES OF STUNTING 137

was only one-fifth aerated than in fully aerated water. Pure oxygen can actually have
a toxic effect. Fully aerated water is probably exceptional, however, in natural conditions
on the sea bottom and it would perhaps have been more illuminating if the authors had
made more realistic comparisons and discussed more fully the natural environment of
the organisms under study. A pronounced deficiency of oxygen in the environment is
known in fact to inhibit growth considerably, due allowance being made for certain species
which are seemingly well adapted to such conditions. Stunting as a result of this factor
has been recognized in foraminifera (Miller 1953. Said 1953), gastropods (Crabb 1929,
Humphrey and Macey 1930), and lamellibranchs (Kindle 1930). Twenhofel (1915)
described a bottom fauna in poorly aerated black muds in sheltered bays off the Estonian
coast which was stunted to a fifth or a quarter of the normal size. He did not, however,
evaluate the influence of low salinity.

Alternative causes of stunting in this case can be suggested. Body metabolism would
be expected to diminish in conditions of low oxygen availability, and hence feeding and
growth would be retarded. This is to some extent supported by the observations on
Mytilus referred to earlier. On the other hand, oxygen deficiency is characteristic of
bodies of stagnant water so that the amount of food brought in by currents would be
little. This could result in the starvation of suspension feeders but could hardly be rele-
vant to carnivors or deposit feeders since the amount of organic matter available in the
bottom deposits of such stagnant bodies is normally considerable.

Turbidity. Kristensen (1957) found that both floating plant detritus, temporarily de-
posited on the bottom, and high silt content of the water inhibit the growth of Cardium
edule, which apparently tends to grow faster on sandy than on muddy bottoms. Other
things being equal, this differential growth will result in the attainment of larger size
at maturity. Rapid growth in regions of strong currents is related to the removal of
inhibiting suspended matter. The growth of Mytilus eduiis is also, according to Kristen-
sen, inhibited by very turbid water. According to de Lapparent (1906, p. 132) the eastern
part of the Mediterranean has more stunted animals than the western; this is attributed
to the influence of sediment from the Nile but it is not made clear whether or not sedi-
ment in suspension is the significant factor. Eagar (1948) quoted the results of experi-
ments showing that silty sedimentation can kill mussels. Turbidity was in this case
excluded as a factor because of the survival of forms suspended above the bottom in
crates. All that appears to have been demonstrated here is the destruction of juveniles,
not the inhibition of growth.

Turbidity probably inhibits the growth of suspension feeders by disturbing normal
feeding activities. Thus Loosanoff and Tommers (1948) clearly showed in their experi-
ments on oysters that when the silt content of the water was raised, the quantity of water
pumped through the gills (and hence the quantity of food consumed) fell sharply. They
agreed with the earlier observations of Nelson that oysters can feed in turbid water but
wished to emphasize that an increase in turbidity usually causes a decrease in feeding
rate. This seems to be a significant observation, since any organism which flourished on
a muddy bottom must inevitably be adopted to a certain amount of turbidity.

Exposure to waves. While it has been seen that rapid water movements may aid growth
of suspension feeders by removing fine sediment, extreme agitation can have an in-
hibiting effect.

138

PALAEONTOLOGY, VOLUME 8

It is relevant to mention here the results of Brown and others (1938) who studied the
variation with environment of eight freshwater lamellibranch species. They found that
size was greatest in almost every case in a sheltered locality and smallest in a very much
exposed one, and concluded from this that wave action had a stunting effect upon growth.
Gibson (1956) observed a similar relationship among scallops and suggested that exces-
sive particle bombardment interfered with feeding. Kauffman (in press) discusses this
effect of agitated water on shallow-water epifauna in general. Coe and Fox (1944)
thought that storms could inhibit the feeding activity and hence the growth of mussels.

Though these various observations may have quite a general validity, it must be borne
in mind that certain specialized animals are well adapted to strong wave action. Balanus
balanoides, for example, is largest and most common in the areas of greatest wave
exposure (Moore 1935). Doochim and Smith (1951), however, showed in the case of
three species of Balanus on the Florida coast that strong currents resulted in lower
growth rates and higher mortality.

Population density. It could be argued a priori that a high population density on the sea
bottom would result in a smaller mean size of the individuals composing the population
because of the relatively intense competition for food. Alternatively it could be argued
that a high density of successfully settled organisms largely represented a response to
an abundant food supply. Yonge (1961) made an important distinction in this respect
between suspension and deposit feeders. The former can live crowded closely together
if the water currents bring in abundant food whereas deposit feeders such as the Telli-
naceae are more or less uniformly spaced out, each with its feeding area.

Such empirical observations as are available suggest that there may in fact be an
inverse relationship between individual size and population density, both in deposit
feeders such as Tellina (Moore 1958, p. 388) and suspension feeders such as Balanus
(Moore 1935) and Teredo (Isham et al. 1951). Kristensen (1957) gave some quantitative
data on this inverse relationship for Cardium. Crabb (1929) performed experiments
which appeared to demonstrate this effect for certain gastropods.

Bottom sediment. Whereas the character of the bottom sediment is extremely important
in controlling the distribution of benthonic communities (Jones 1950, Yonge 1961) it is
in itself of limited significance in influencing growth within individual species adapted
to a particular type of environment. The importance of the type of sediment lies largely
in the indication it gives of the strength of the water movements and it is to this more
fundamental factor that palaeoecologists should normally refer their interpretations.
Among deposit feeders, however, the character of the sediment may play a more signi-
ficant role, since it is the direct source of food. Thus McNulty et al. (1962) have found a
high correlation between the size of the dominant organism and particle size among a
deposit-feeding infauna in Biscayne Bay, Florida.

It is possible that a soft bottom could influence size, in that animals larger (and heavier)
than a certain critical value would sink into the sediment and be suffocated, and fine
mud could clog the feeding mechanisms of burrowers and hence inhibit growth, but
I know of no recorded instance of stunting as a result of these factors.

Other factors. In this section brief consideration will be given to a number of factors
which have been proposed at one time or another, mainly by palaeontologists, to account
for stunting, but which appear to be irrelevant or of very limited application.

A. HALLAM: ENVIRONMENTAL CAUSES OF STUNTING

139

That depth of sea as such is unimportant as an environmental factor is demonstrated
by the well-known fact that certain members of level bottom communities in shallow
water in the Arctic Ocean are found at depths of 1,000 to 2,000 metres on similar bot-
toms in warm temperate and subtropical seas, indicating that the controlling factor is
temperature (Thorson 1957).

Though Shimer (1908) suggested that the reduction in size of animals at depth might
be related in part to the small amount of light penetrating to the bottom, no supporting
evidence was given. The amount of light is probably important only in the special case
of reef corals, since Goreau (1959) found that their growth was significantly diminished
by its exclusion. He concluded that the effect of light on growth was mediated in part
through the zooxanthellae.

The possible influence of hydrogen ion concentration in sea water has been referred
to by several workers. Humphrey and Macey (1930), in their research on Littorina in
tidal pools, noted a correlation between size and pH, though it is not clear how rigorously
they excluded other factors such as oxygen content and salinity. Lalicker (1948) also
stated that the size of the ciliate Paramecium decreased with pH, but remarked that the
low pH values were often associated with abundant decomposing organic matter. This
introduces the possibility that oxygen deficiency rather than pH might have been the
significant factor involved in the size reduction. Scott (1948) suggested that low pH was
a prime factor in stunting, quoting the work of Humphrey and Macey. He also referred
to some experimental observations indicating that acid-toxic waters lower the absorptive
capacity of the gut epithelium in certain lamellibranchs, so that stunting would be a
natural consequence. Be this as it may, such observations have little relevance to sea
water, which has a very constant, mildly alkaline pH owing to the buffering action of the
calcium carbonate-bicarbonate-carbon dioxide system.

The chemical composition of sea water is normally very constant and is unlikely to
have differed appreciably in the comparatively recent geological past, since the time that
highly organized invertebrates first appeared. Hence various suggestions by palaeonto-
logists that abnormal concentrations of certain ions might have caused stunting in fossils
have an air of implausibility. Much has been made, for instance, of the observations of
Sarasin (1913, 1917) in support of the idea that a high concentration of iron can have
an adverse effect upon growth. Sarasin discovered that certain ponds with a rock sub-
stratum rich in iron contained minute gastropods, crabs, and fish that were supposedly
stunted. No chemical analyses of the water were given, however, nor the possible in-
fluence of other factors considered. There remains, moreover, the considerable doubt
whether the restricted and perhaps rather special environment of these ponds has much
relevance to conditions in the sea.

Similar doubts surround the experiment cited by Loomis (1903) in which small tad-
poles, fish, and snails were kept in an aquarium in which the water was saturated with
an iron compound; all the individuals showed stunting effects after several months. One
would like to have known more about the degree of oxygenation and the availability
of food compared with the natural environment before accepting that the iron was the
principal cause of stunting. Again, such experiments as this are irrelevant to conditions
in the sea, where the water is never saturated with iron compounds. Moreover, the signi-
ficance of iron concentration is doubtful, since it is its availability in chelated form and
not its total quantity that matters (H. B. Moore, personal communication).

140

PALAEONTOLOGY, VOLUME 8

It may be true that the presence or abundance of certain cations such as ‘Cu’ retard
or stop growth (Tasch 1953), though this is denied for crustaceans by Scott (1948), but
again the claim is open to the charge of irrelevance. Tasch proposed, furthermore, that
the water polymer trihydrol, produced by melting ice, might have an important in-
fluence, based on experiments on the growth of diatoms. No evidence was adduced that
this is true also for benthonic invertebrates and no marine ecologist has, to my know-
ledge, ever suggested this. Similar doubts surround Tasch’s suggestion of the possible
importance of sulphoxide and sulphhydrol.

Grabau (1931 ) thought that the high magnesium chloride and sulphate concentrations
in the Gulf of Pechili might have played a role in the stunting of a number of inverte-
brates but such stunting could in this case be merely the consequence of reduced salinity
(see above).

Another suggestion favoured by a number of palaeontologists relates to the existence
of ‘algal meadows’ in very shallow waters. Dr. Fergusson Wood has kindly offered the
following statement. ‘So-called “algal meadows” consist of either true algal meadows
or meadows of sea-grasses. The former grow attached to a rock substrate or on a sedi-
ment in the littoral or supra-littoral. In the first case they consist of brown algae such
as Fucus, Hormosira (in the Southern Hemisphere), Sargassum, or as algal “forests” of
Ecklenia , Laminaria or Macrocystis, or of coralline red algae, particularly in warmer
waters. The latter are usually filamentous green algae such as Enteromorpha , Chaeto-
morpha, Halimeda &c. or blue-greens such as Oscillatoria or Lyngbya. This community
frequently has a reducing layer below and the algae are possibly capable of photo-
reduction of C02, getting the hydrogen from H2S rather than H20. The sea-grass
meadows usually have an oxidized layer at the surface of the sediment and a reduced
layer or region below. The most important association, however, is frequently the
epiphytic community which consists of diatoms, blue-greens, and red algae such as
Ceramium, Polysiphonia and Lawrencia which may outweigh the sea grass.’

‘Algal meadows’ provide a multitude of micro-environments harbouring a rich and
highly diverse animal life dominated by the epifauna. The fauna may be small as in-
dividuals, as apparently in Messina Harbour (Fuchs 1871). Fuchs maintained that this
was the consequence of the concentration of juveniles in the meadows as a result of the
physical exclusion by the algae of larger forms. If this interpretation is correct it
obviously has nothing to do with true stunting. Evidence of stunting of mature indi-
viduals in any algal meadow appears indeed to be lacking at present.

INTERPRETATION OF STUNTING IN FOSSILS

Criteria for establishing maturity. The palaeontologist is naturally handicapped in the
recognition of stunting because he cannot prove conclusively that given fossils represent
sexually mature organisms. The most he can hope to do is make a reasonable case by
establishing (a) that the suspected stunted fossils differ only in their smaller size from
other forms of the same geological age, suggesting that they may belong to the same
species, and ( b ) that the characters of the fossil shells correspond with those of sexually
mature living organisms. Unless this is done the sceptically minded can maintain either
that the small fossils belong to naturally small species or that they represent juveniles.

A. HALLAM: ENVIRONMENTAL CAUSES OF STUNTING 141

Fortunately in many invertebrates there are a number of criteria for establishing
maturity with a considerable degree of confidence. Seasonally produced growth rings
can be used in two ways. At the onset of maturity the growth of invertebrates generally
tends to slow down rather than cease and this can lead to the crowding of growth rings.
Vogel (19596) has made use of this fact to demonstrate by graphical means stunting in
living lamellibranchs and brachiopods.

Secondly, the spacing of growth rings can give useful information. In the course of
a study of a Jurassic ironstone I came across a thin band of rock containing abnormally
small specimens of two common lamellibranch species. A graphical study of the spacing
of the growth rings showed that those of the small individuals were appreciably closer
together than in larger forms elsewhere in the ironstone, suggesting that they were in
fact truly stunted (Hallam 1963).

The suture lines or the septa of ammonoids can be treated in a similar way. It has often
been remarked that adults exhibit a crowding together of the last few suture lines. It is
probably more reliable in this case, however, for reasons to be discussed later, to use the
technique proposed by Vogel (1959n), who plotted the amount of septal separation
against whorl diameter for some Cretaceous ammonites and was able by this means to
demonstrate a probable case of stunting.

While growth-ring studies are most applicable to lamellibranchs and brachiopods they
may prove useful in some other groups as well. Annual growth rings have, for instance,
been recognized in corals and in the genital plates of echinoids.

Adults may possess distinctive morphological features, such as the thickening of the
apertural margin in some gastropods. Callomon (1957) gave several morphological
criteria for the recognition of mature ammonites. Provided comparison can be made
between suspected stunted and normal forms apparently of the same species in different
facies of the same age, the number and arrangement of, for instance, septa in corals,
chambers in foraminifers and whorls in gastropods may be significant, though this is a
field requiring much further study.

In the description of suspected stunted fossils it is important at least to give both the
range and mean of some measure of size. Tasch (1953, 1957) would go further and advo-
cated plotting graphically the complete size-frequency distribution for given species.
While this seems desirable where material is abundant and easily collected, the inter-
pretation of such distributions can be more complicated than Tasch evidently appreciated
in his study of a Pennsylvanian shale. For example, in the case of the lamellibranch
Nuculana bellistriata, Tasch found that the size distribution fell into two distinct groups.
The main group was unimodal but there were a few much larger specimens. These two
groups were interpreted as juveniles and adults. It could equally well be, however, that
a small percentage of the total could have experienced pathological gigantism (cf.
Boettger 1952) and a large proportion could have been adults.

Again, the brachiopod Crurithyris planoconvexa showed a unimodal distribution.
Tasch contrasted this with the multimodal distribution of a living brachiopod species
described by Percival (1944) and suggested that therefore there were only one or possibly
two juvenile stages represented in the Crurithyris sample. It is inadmissable, however,
to compare fossils accumulated through a large quantity of sediment in perhaps a con-
siderable time with a living brachiopod sample, only a few years old, collected from
a few square metres of ground. Size-frequency distributions are, moreover, the result

142 PALAEONTOLOGY, VOLUME 8

of the interaction of a number of variables which are not always easy to disentangle
(Craig and Hallam 1963).

It must be admitted that if the various claims of stunting in fossils are subjected to the
sort of tests outlined above, few could be considered proven conclusively. Indeed, Tasch
(1953) went so far as to dismiss virtually all the cases he reviewed on grounds of in-
sufficient evidence. While one can only endorse the request for fuller documentation, it
is felt that Tasch has been unduly sceptical.

In the remainder of this article an attempt will be made to evaluate some of the claims
of stunting in the light of our present knowledge of marine ecology and sedimentology„
and make judgements on their relative plausibility. It would be tedious and rather point-
less to consider each example in detail since so many authors have omitted critical data,,
even when one suspects that they could have made a convincing case. Interpretation at
present is necessarily tentative and a whole series of detailed reinvestigations is called
for. It should be stressed, however, that a too-ready dismissal of stunting for fossils of
abnormally small size for the species may create more problems than it solves.

Abnormal salinity. It should not be difficult to recognize stunting in fossils that is the
result of low salinity because brackish-water faunas have a distinctive character. They
are normally restricted in variety, stenohaline groups being rare or absent. The euryhaline
forms that dominate are largely composed of a few species of lamellibranchs, gastro-
pods, and ostracods which may be numerous as individuals. Stunting of the molluscs
together with other forms should occur, though it must be borne in mind that certain
species may have been well-adapted to brackish water and attained their greatest size
there. The sediments containing the fossils may give some indication of the proximity
of rivers, such as deltaic sands with drifted plant remains (Schmidt 1951).

In the case of late Mesozoic and younger rocks it may be possible to utilize genera or
even species, the salinity tolerance of whose modern relatives is well known. Goldring
(1922) compared the Pleistocene fauna of the Champlain and St. Lawrence Valleys with
that of the Baltic Sea and argued convincingly that the changing character of the former
was the result of a southward decrease in salinity in the seas of that time. Gekker (1957)
has reported salinity control of the fauna of the Palaeogene Ferghana Gulf in central Asia,
brackish-water conditions being marked by the absence of nummulites, brachiopods,
and cephalopods, and the rarity and small size of echinoids and corals. The lamelli-
branchs and gastropods were large in number but few in species and exhibited stunting.
A group of Cretaceous molluscs in Maryland gave indications of low salinity conditions
by an association of stunted marine forms with accepted brackish-water forms (Vokes
1948).

Barker (1963) has recently made a study of size variations among ostracods of the
genus Fabanella in the Portland and Purbeck Beds of Buckinghamshire. He was able
to discount the factor of mechanical sorting and related a decrease of maximum size up
the succession to a reduction in salinity, by taking into account the character of the
associated fauna.

Dunbar (1941) quoted a relevant example from the Permian of the Soviet Union. As
the Kazanian strata are traced eastwards across the Russian Platform richly fossiliferous
marine limestones pass into continental red beds via a transition zone characterized by
reduced variety and apparent stunting of the fauna. Although this is attributed to

A. HALLAM: ENVIRONMENTAL CAUSES OF STUNTING 143

brackish-water conditions one wonders whether it might rather be a case of increased
salinity, bearing in mind the rich evaporate deposits near the Urals.

There is some doubt concerning Grabau’s (1931) attribution of the Permian Jisu
Honguer Limestone of Mongolia to deposition in brackish water. Although widespread
stunting was claimed on the basis of comparison with similar species of the same age
elsewhere, the fauna is composed largely of brachiopods together with a few corals and
bryozoans, groups which at the present day are notably stenohaline.

Hypersaline faunas should have characteristics similar to those of brackish watei but
should be distinguishable by the occurrence of evaporites in the associated sediments.
A possible example of such a fauna was described from the Muschelkalk of Selva Nera
by Hohenstein (1913). It is composed mainly of lamellibranchs and gastropods together
with a few brachiopods and cephalopods ; corals, bryozoans, and echinoderms are absent.
Hohenstein interpreted the small size of the fossils (1 to 45 mm.) as the result of stunting
in conditions of high salinity, as suggested by the presence of gypsum and anhydrite.

Trechmann (1913) described a likely case from the Permian Magnesian Limestone
of Durham. In the so-called Shell-Limestone Reef an impoverishment of the fauna is
discernible up the succession. Brachiopods die out, leaving a fauna considerable in
numbers but restricted in variety, composed of lamellibranchs and gastropods; some of
these are apparently dwarfed. Thus species of Pleurophorus are only about one-third of
their normal size. Taking into consideration the lithological evidence, Trechmann related
this dwarfing to an increasing amount of sulphates in the sedimentation.

Arkhangel’skaya and Grigor’yev (1961) briefly allude to fossil stunting in deposits
of the Lower Cambrian evaporites of Siberia.

Temperature. From what has been said earlier it seems unlikely that temperature changes
can have effected appreciable variations in the size of fossil species, at least in adjacent
rock formations. They should nevertheless have had a discernible effect if large enough
regions are considered. An example (not strictly at the level of species) is the well-known
case of the Cretaceous rudist lamellibranchs, which grew to considerable size in the
Mediterranean province, in the region of the old Tethyan seaway. Associated with large,
thick-shelled gastropods and large foraminifera, their growth can reasonably be related
to high water-temperatures, but northwards in Europe (into presumably higher latitudes)
their size diminishes appreciably (Dacque 1915, pp. 423-4).

It may prove possible to detect regional changes in size which are the result of
temperature differences by correlating them with gradients of diversity (Fischer 1961,
Stehli and Halsey 1963) since it is well known that organisms become progressively
richer in variety towards the equator. Fischer notes that this is more marked in the
epifauna than in the infauna. No confusion should arise between gradients of this sort
and those that are the result of abnormal salinity. Unlike the latter, temperature gradients
will probably only be detectable by the consideration of whole continents, and steno-
haline animals will persist even in the regions of least diversity. However, it should be
borne in mind that significant changes in size might only be recognizable at the level of
genera.

Size gradients with temperature among more than a few elements in the fauna are
unlikely to be common, as the optimum conditions for different species are likely to
have varied considerably. This situation contrasts markedly with that concerning, for

144

PALAEONTOLOGY, VOLUME 8

instance, abnormal salinity, since the optimum here will correspond with normal marine
conditions for the bulk of the fauna.

Finally, if suitable shell material is forthcoming, it may be possible to determine their
temperature of formation from the oxygen isotope ratios, and thence compare size and
temperature directly.

Oxygen deficiency. Stunting as a result of oxygen deficiency is to be expected in seas
where the water circulation is severely restricted and hydrogen sulphide generated by
sulphate-reducing bacteria. The bottom sediments associated with this type of environ-
ment are characteristically fine-grained because of deposition in quiet water and rich in
unoxidized organic matter and iron sulphide, produced by the reaction of hydrogen
sulphide with iron salts in solution. An association of bituminous and/or pyritic shales
with small-sized, apparently stunted fossils has indeed been repeatedly recognized
(Grabau 1917, Price 1920, Scott 1924, Marr 1925, Chao 1927, Ruedemann 1935, Williams
1937). Broadhurst (1959), in a careful study of some non-marine lamellibranchs from
the Upper Carboniferous of Lancashire, found an inverse correlation between shell
size and the quantity of pyrite in the sediment. He interpreted this as stunting in con-
ditions of oxygen deficiency and perhaps reduced food supply.

Though this association is impressive and probably significant, documentation is
unfortunately slight and criteria of stunting rarely adduced. Fisher (1951 ) is a welcome
exception in that he studied the beak features of brachiopods and the growth lines of
lamellibranchs to deduce stunting in a rich fauna collected from iron sulphide con-
cretions. While, moreover, most authors have attributed the stunting to oxygen defi-
ciency, Grabau (1917) related it without good evidence to estuarine conditions for the
dark Genesee Shale, and Scott’s (1940) interpretation seems somewhat confused. Discus-
sing small pyritic ammonites from the Cretaceous of Texas, which he had earlier (1924)
interpreted as stunted forms, he remarked (p. 311) that many specimens were probably
preserved inner whorls of large specimens, but on the next page he stated that ‘many of
them are obviously dwarfed’.

In cases where the preservation of pyritic micromorphs precludes the sure recognition
of maturity it should not necessarily be assumed that they represent juveniles. On the
other hand, a danger of misinterpretation arises from the possibility that small shells
may have been preferentially pyritized. If larger shells have not been well preserved the
pyritic specimens alone may have found their way into museum collections and give a
misleading impression of the composition of the original fauna. Associated moulds, im-
pressions or calcitic fossils should therefore always be sought. Kummel (1948), for in-
stance, discussing ammonites from Texas, observed that pyritic micromorphs occur in
the same beds as larger forms preserved in calcite, and Callomon (1957) pointed out that
the so-called micromorphic ammonites in the Oxford Clay of Buckinghamshire are
actually the pyritized nuclei of immature shells with non-pyritized body chambers.

Kummel (1948) also expressed scepticism about a claim by Jaworski (1923) of stunting
in small cephalopods collected from Triassic bituminous cherty limestones in Peru.
According to Kummel they appear to be of the normal size for the species and genera.

Besides more data on the fossils a proper appreciation of the evidence of the rocks is
required. That not all black shales were laid down in a poorly-aerated environment is
indicated, for example, by the rich benthonic fauna, including burrowers, of the

A. HALLAM: ENVIRONMENTAL CAUSES OF STUNTING

145

Devonian Hunsriickschiefer (Richter 1931). Pyrite usually forms diagenetically within
fine-grained sediments and is not necessarily relevant to conditions in the bottom waters,
though it is probably true in most cases that an abundance of pyrite signifies a poorly
aerated sediment surface. The surest indication of anaerobic or near-anaerobic condi-
tions is a fine lamination produced by alternations of organic matter with clay and cal-
cite, probably annual in origin, as found for instance in the deeper parts of the Black
Sea (Archanguelsky 1927). Laminated shales or limestones of this type are frequently
barren of benthonic fossils except for extremely thin layers of nektonic animals or surface-
livers, which can sometimes be shown to be stunted (Hallam 1960); alternating phases of
anaerobic and poorly aerated water are implied.

One of the most detailed studies of a small-sized fauna in a formerly pyritic shale was
undertaken by Tasch (1953, 1957). The Pennsylvanian Dry Shale of Kansas now contains
limonite rather than pyrite because of subsequent oxidation and Tasch considered that
the original environment was in fact poorly aerated. He nevertheless discounted the
possibility of stunting in all but a few instances on the basis of his analysis. Since his
work has important implications it needs scrutinizing in some detail.

Tasch collected some 4,000 fossils, mostly goniatites, brachiopods, and lamellibranchs.
Almost the whole fauna was of pebble grade, mostly ranging between 4 and 8 mm. For
the goniatites Tasch used the crowding of the last few suture lines as an index of maturity.
Among 205 usable specimens of Imitoceras grahomense 30 were considered mature but,
peculiarly enough, the size range of these approximated to that of the supposed juveniles.
This paradox could be resolved by accepting that the bulk of the sample consisted in fact
of mature specimens. From the data given it is far from easy to assess maturity on the
basis of suture-line crowding since this is highly variable and can be repeated during the
growth of a single individual, as the author has found also in certain Jurassic ammonites.
Similar considerations apply to the other goniatite studied in detail, Gonioloboceras
goniolobum, interpreted as consisting entirely of juveniles though not many suture lines
were visible. It would be better to make sagittal sections and use the ingenious technique
of Vogel (1959a) of plotting septal spacings against size for normal and suspected
stunted ammonoids. Doubts concerning the use of size-frequency distributions in the
brachiopods and lamellibranchs have already been alluded to. One can nevertheless
follow Tasch's interpretation and see where it led him.

Tasch was faced with the problem of accounting for an assemblage consisting largely
of juveniles. This he did by invoking mass catastrophe, resulting in high juvenile mortal-
ity, and selective size-sorting by currents. The stunted Imitoceras required a further
explanation based on the supposed toxic action of ionic iron in the sea water. All these
explanations are open to objection.

The sort of catastrophic destructions of benthonic animals known to marine ecologists
are normally intermittent, short-lived events. If such events leave any clear record at all
in sedimentary successions, it is more likely to be as thin bands of fossils of all sizes,
both juveniles and adults (cf. Hallam 1961) than as juveniles scattered through a whole
rock unit such as the Dry Shale. The latter would imply a series of catastrophes con-
tinued over a long period of time, with recurrent repopulation from another region.
It is dubiously justifiable, furthermore, to invoke size-sorting by currents without
independent evidence. The work of Craig and Hallam (1963) on modern shells suggests
that this process cannot be assumed without evidence even for high energy environments

B 0012

L

146

PALAEONTOLOGY, VOLUME 8

and is much less likely to have been important in the low energy environment signified
by the formation of shale. One would also expect the larger shells to remain behind and
the smaller to be carried away, yet Tasch, in his analysis of Crurithyris, made a direct
comparison with Percival’s (1944) work on living brachiopods fixed by pedicles in
attempting to prove the juvenile composition of the fauna.

Finally, from what has been said in an earlier section, it is improbable that ionic iron
can attain toxic concentrations in sea water. Why, moreover, should only a small fraction
of the fauna be affected?

While the presence of some juvenile stages is not denied, it is surely more reasonable
to explore further the possibility that the Dry Shale fauna was stunted as a result of
oxygen deficiency, since Tasch himself did not deny the existence of a poorly aerated
environment. This interpretation has at least the advantage of being economical.

Turbidity. Many palaeontologists are familiar with the fact that fossil species are fre-
quently larger in rocks deposited in a relatively high energy environment, such as sand-
stones and calcarenites, than in shales. Examples can be cited from the Lias in Britain.
The Frodingham Ironstone, a limonite oolite, contains lamellibranch species that are
appreciably larger than in shales of the same age (Hallam 1963). The same is true for
the Hettangian Sutton Stone in Glamorgan, which is a calcarenite deposited close to an
old shoreline, above surf base (Hallam 1 960). A possible example among the foraminifera
is given by Lalicker (1948). Heterostegina texana, of the Oligocene of Texas, attains a
greater size in calcareous sandstone than in shale, but in this case the factor of current
sorting has to be considered.

These facts suggest the possibility of a control on growth by the turbidity of the bot-
tom water, in cases where oxygen deficiency can be ruled out by the presence in the shales
of a rich benthonic fauna including burrowers and by the absence of lamination or
bituminous matter.

Species may also be larger in slowly deposited fine-grained rocks such as certain
calcilutites as compared with thicker, more argillaceous successions. A high rate of
sedimentation has been invoked as a cause of stunting both by Bradley (1921) and Eagar
(1951) but this is probably only important in that it may signify increased turbidity.

It should be remembered that the turbidity effect has only been demonstrated as yet
for suspension feeders.

Strong wave agitation. The author knows of only one case where stunting in fossils was
related to extreme water disturbance. Croneis and Grubbs (1939) described the fauna
collected from ellipsoidal calcite nodules in a fine-grained Silurian dolomite in Illinois.
They provided what appears to be good evidence for stunting. Thus brachiopods of one-
third normal size have mature pedicle structures; crinoids with calices not greater than
2 mm. have a complete cyclic arrangement of basals and radials with well-developed arm
sockets, just like the mature forms of larger specimens elsewhere; gastropods, ranging
in size up to 12 mm., have the same number of whorls as their more characteristic but
larger relatives.

Their interpretation, however, that the nodules were formed directly by storm action,
is highly questionable. From the illustration and description given, the nodules appear
to be diagenetic rather than syngenetic in origin. The evidence cited for the latter, the

A. HALLAM: ENVIRONMENTAL CAUSES OF STUNTING 147

bending of shaly laminae around the nodules, can easily be accounted for by differential
compaction. Though storms may periodically stir up sediment in environments of
shallow-water carbonate mud deposits such as the area west of Andros Island on the
Bahama Bank (Black 1933), there is no evidence to suggest that they result in the
formation of balls of calcareous mud. It is furthermore extremely unlikely that such
balls would preserve their identity for long.

A more likely case of stunting in agitated water is given by Kauffman (in press) in his
description of some Cretaceous oysters.

Exposure to waves may conceivably have some bearing on possible cases of stunting
in reef limestones, such as described for a rich Middle Triassic fauna by Jekelius (1935).
The author has also seen a peculiar micromorphic brachiopod assemblage in the Upper
Jurassic sponge reefs of southern Germany, and the so-called Upper Coral Bed of the
Middle Jurassic of the Cotswolds contains a rich fauna of minute brachiopods which may
be stunted (Richardson 1907).

In contrast to the normal faunas of most rocks deposited in high energy environments,
certain oolites have been found to contain abnormally small-sized assemblages and it is
perhaps relevant to discuss these here.

The classic example is that of the fauna of the Mississippian Salem Limestone of
Indiana described by Cumings et a/. (1906). Stunting of this rich fauna, which includes
blastoids, brachiopods, bryozoans, corals, gastropods, lamellibranchs, and trilobites,
was claimed on the basis of the minute size of the fossils compared with similar forms in
the adjacent rocks. The Salem Limestone has traditionally been considered an oolite
but Patton (1953) has pointed out that true ooliths are rarely found and it should rather
be described as a bioclastic limestone exhibiting a high degree of size sorting. Patton
considered that the fossil content has been determined largely by particle size rather
than by the character of the organic life. This interpretation seems to be borne out by
the fact that both the foraminifera (of the genera Endothyra and Plectogyra) (Lalicker
1948) and the ostracods (Scott 1948) are of the normal size for their species. That wave
and/or current sorting rather than dwarfing is responsible for the small size of the fossils
seems plausible enough for this type of rock, but Cloud (1948) has remarked that some
at least of the minute brachiopods have adult beak characters. More work on this
interesting fauna is clearly desirable.

Loeblich and Tappin (1964) have recently queried the interpretation of the Salem
endothyrids as normal forms because of their large size, citing the work of Bradshaw
(1957) to suggest that abnormally large foraminifers may signify unfavourable conditions.
Though this is conceivable, it should be borne in mind that Bradshaw’s experiments,
illuminating as they were, dealt only with one species, and the size differences perceived
in a wide range of salinities did not exceed 5 per cent, of the maximum.

In the case of another Mississippian oolite, foraminifera of the genus Plectogyra do
appear to be genuinely stunted (Lalicker 1948), while Stauffer (1937) has described a
fauna of small gastropods and cephalopods in a silicified dolomite oolite in the Ordo-
vician of Minnesota but no evidence supporting stunting was put forward.

A further example of an oolite containing a minute-sized fauna (lamellibranchs and
gastropods) was cited from the Middle Jurassic of the Cotswolds by Arkell (1933,
p. 272). This fauna was interpreted, however, as consisting of juveniles concentrated by
the sorting action of currents.

148

PALAEONTOLOGY, VOLUME 8

High population density. Stunting as a result of high population density would be very
difficult to substantiate for fossil assemblages, though it has been proposed for different
faunas by Cumings et a/. (1906), Kuhn (1936), and H. and G. Termier (1951). This is
principally because fossils are only rarely preserved in their positions of life, having in
most cases been repeatedly disturbed by water movements and the activity of burrowers
prior to ultimate burial. Local variations of shell density in fossiliferous rocks are nor-
mally held to reflect differences in rate of sedimentation and hence have little bearing
on the original space conditions of the living organisms.

Abnormal ionic concentrations. Although appreciable changes in the chemical composi-
tion of sea water can effectively be discounted, the accumulation of iron and phosphate
ions in toxic quantities has been proposed on several occasions to account for particular
examples of fossil stunting.

Loomis (1903) made a classic study of stunting in the fauna of the so-called pyrite
layer of the Tully Limestone in the Devonian of New York State. Among over fifty
species of varied animals stated as including adults, few exceeded 2-0 mm. in size.
Stunting was in part attributed to a high concentration of iron in the sea water, as evi-
denced by the abundance of pyrite. However, the amount of iron in pyrite in modern
sediments bears little relationship to the amount dissolved in the adjacent sea water,
in fact may vastly exceed it, coming perhaps mostly from the solution of diatoms within
the sediment (Emery and Rittenberg 1952). Accepting the stunting as genuine (and no
one has yet disproved it) another explanation must be sought.

That high sulphide concentration and hence oxygen deficiency is in fact responsible
is suggested both by the facts cited earlier and by the special case of the Snap Band in
the Frodingham Ironstone. This pyrite-rich layer contains small lamellibranchs that
appear to be stunted compared with the large shells in the normal ironstone, in which
limonite (an alteration product of chamosite) is the dominant mineral (Hallam 1963).
Similar probable misinterpretations were made by Marr (1925) and Tasch, whose work
has already been discussed, while Ager (1956) related brachiopod stunting in the Marl-
stone Rock-bed ironstone to an enrichment of iron in the sea water.

An association of apparently stunted faunas with phosphatic deposits has led Ladd
(1925), Branson (1930), and Ball (1935) to relate the stunting to high phosphate con-
centrations in the sea water, but the existence of phosphate-rich deposits by no means
implies greatly enriched phosphate in the bottom waters (Bushinski 1964). Almost
all dissolved phosphorus brought by rivers into the sea is immediately assimilated by the
plankton and subsequently deposited on the bottom after death of the organisms.
Solutions within muds deposited in the Bering Sea contain thirty to ninety times more
phosphorus than the bottom waters and Bushinski attributed the formation of phos-
phorite layers to processes of diagenesis. In the case of the Maquoketa Shale of Iowa
described by Ladd (1925), in which the rich fauna has an upper size limit of 6 mm., an
abundance of pyrite was also noted. Here, as in the case of the Tully Limestone, one
can see that an alternative explanation of the stunting is available.

‘ Algal meadows . ’ It has been observed earlier that while algal (or sea grass) meadows
may perhaps account for a concentration of juvenile organisms in certain cases there is
no evidence as yet that they are responsible for extensive stunting. Nevertheless, Kutassy
(1930) proposed this as an explanation of stunting in a fauna from lignite-bearing

A. HALLAM: ENVIRONMENTAL CAUSES OF STUNTING 149

sediments in the Miocene of Hungary. Stunting was recognized by comparison with
normal Miocene species elsewhere, but as the fauna is characterized by numerous lamelli-
branchs and gastropods of normal marine types together with a very small number of
species of echinoderms, bryozoans, and corals, the alternative of low salinity conditions
should have been fully explored, especially as this is just the sort of environment sug-
gested by lignitic beds.

Fuchs (1871) thought that the diminutive character of the rich Alpine Triassic fauna
of St. Cassian could be the result of selective exclusion of adults as in Messina Harbour,
but Boni’s review (1942) suggests that some species at least are genuinely stunted, though
the evidence presented is not entirely satisfactory.

Kuhn (1936) attributed the small size of a Middle Miocene fauna to stunting because
of competition for food and living space in algal meadows, but the paucity of data makes
it impossible to assess whether or not adult forms were present. The richness in epifauna
(corals and bryozoans) does perhaps in this case support the possibility of algal meadows.

A provocative hypothesis was put forward by H. and G. Termier ( 1951 ) to account for
the smaller size in shales than in limestones of many goniatites. As they thought that the
shell characters of the goniatites in carbonaceous shales signified immaturity they pro-
posed an environment of algal meadows which could have served as developing grounds
for the young (cf. Bauer 1929). This is conceivable because goniatites were probably
fairly active swimmers, but the Termiers felt obliged to interpret the small lamellibranchs
and gastropods in the same beds as adults stunted because of isolation and crowding.
Unfortunately little evidence was provided for distinguishing adults from juveniles and
an application of the techniques proposed by Vogel (1959u, 19596) could here prove
very rewarding.

No writer on the subject of algal meadows has paid much attention to the problem
of inferring their existence from the sedimentary record, which is clearly desirable if a
convincing case is to be presented. Bauer (1929), describing Posidonia meadows in the
Mediterranean, stated that they only occur where large stones are available in sand as
attachment surfaces. Over the course of time black sulphurous mud rich in organic
matter accumulates in the sheltered environment created by the seaweeds. A sediment
of mixed stones, sand, and mud should therefore result and the associated facies should
indicate the proximity of a shoreline because seaweeds are confined by their light require-
ments essentially to the littoral or shallow sublittoral zones. It is by no means certain
that such sediments will have retained much of their original organic matter produced
by decaying vegetation, pace the Termiers, since only slight changes of environment
could result in strong wave or current action and strongly oxidizing conditions in such
a regime as this.

The complication of evolutionary size changes. Increase or decrease in size during the
course of evolution is familiar in fossil vertebrates and is no less true for many inverte-
brate (Newell 1949). That this may have operated even intraspecifically is suggested by
detailed work on the Blue Lias in southern Britain (Hallam 1960). As much as a four-
fold regional increase in size up the succession (which is probably valid for much of
Europe as well) was discernible in a number of molluscan species. This was not evi-
dently related to sedimentary facies and environmental control over such a large area
could reasonably be discounted. When, as in this case, appreciable changes take place

150

PALAEONTOLOGY, VOLUME 8

0) C3 •

O d) G

Is!

£ = E
ftg o
. E(J

>un
3 •

c 2

^.sS

o X co

X c*-

-£ > o

> ^ n

,2 o

2 ^

'£. >-

p =ts «j

£ C 3

<5 « 3
7 E g
«

3 d> • —

"£ CO

o _ o
X"G to
tr <u »-

X to

O d>

3 '—

O d>

io

d> C
r » H

^8^

G d> C
3 _ n • — :

o t!

Q. to 03

S §.s-

C to O'.

T3 3
d) O

n O

0> w

£ c

o £
o o
-O a
- £
(L> • —
£ >»
4_, X
o G

z

^ c
<L> g E
- -g £
t— > o

O^o

^ oo ^
3 3 d)

o a ^

<D co

X C
3 55

X - 03
OX X

I CO O CO

E'~ »;

-K +-» 1) U.

g 3 C M d>
<JX) _ C-Q
t- . 5 *X 4>
£ J .

l'cl^-2

Is S?.£S

n.H-- 3 a

m “t]2 »l

° § C - 3

fig -o go

3X a>x

IP 11
ISI.il

■" w o

>, Ouu

•~ >>o w .

>73 u-x

d> -X > 3 a3
X 3 ^ '4“ X
t- *-■ o
>,« Ou u.

03 a c o x

S g So

O d>
^ tH X g*
O O a>

c u
X Ui d>
c aM-=

3 _ _C c/3

4> ■

<u

c

- _, a>
c •- >

y >,£-2,

c 5 x o
o ° u c

S rf >.,2

^ G
X co

Cl'4-

,Eo

_ U

sg

_ 03

11
g.H
•a x

d>

>. 3
X 03
03 O
C <D

3 (33

x h ^

^ to c

co C
0) D c/>
•- N „v

3-0 >>

3‘c «

■a E^
r <dx
.> o

Co 3

X W

|r 0*0

£ c ^

Ot. ^
O.^ 3

E E.c

-o <l> c
to d)

tj-o

<2 o
&«
2-2

75 <D

5 00

c CJ

X O
w G
M G
G _g
o =
JS'S

W

iJ

a

<

H

.2 o y ■
> — .£•
3 nf — -
tc 0.2
°Q

C*_ tt- ■

° °

V ■o ■

E'i

■8 c .Si

T3 O,

c o£ ,

2 -2

E2-.

C o

£ ;

go^J

3 .£ 3 i

2£ g,

CJ-— ~ ■

Q

Q. c X
in ° U

2 a
> u c
u3 O

»P
■2 | g.

£ 3

o o >

££ X <d

i.£ c c
: 3.S °

1s.sl

o ^ *o _r

X 0/5 C C

.2 2 g.2

2 | « E
2£ E£

“■d c
O Q-> — i ■— .
to 3

CO b o c
3 _ X O
O ^ x
ft; y 3 «

= 3: g

mm 3o^

3 u o

d) <C 3 w
*- "3 . X x.

O d> co c O

£ w.S2 § „
, 5 o

T3 9 c

« c«

.£ «* « In
Sg£

to 3

:.oS

! co

, •— co co

l-tt-S

c ^ <l>

d>

"O co O

_, GO
^ 3 3

«■£ g

.&|
o ^

w - to

X.fJ G

x *-• c
3^3
a> i- >>
73

3 c ^

£03

«88

CQ

m

X 3

X .JL, T3
.y T3 £

r- d> X

>

^ o
C G X
.-^O d)
co

d> o x

d> £

3 G ^

X 3 ..

X d»X

•Sis .

3 to G C

P'S 0.2

8 r o g
t. t-ss

S'm ^ 2

g S £ro
2

G3 “ C —

g £

O M

>, co co co

Ccftt
d> o c CO
>••=00
3 to

G-rj d)^>

U r-G

^ofc

aga-

8 G 60.2

x G ci ^

3 £2 co

c .£ O C

0*2 3 co. 2
X 9 o 3.—
S c"5 “’"S
u o c e

S-SfES

° cS 3 3 2

U.-0 £ £ o

3 d> X u

G £ y MU)

u° Sec

3 d> 3‘3‘3
D. 3 Q, C C

Q to 3 CO CO

C 3 CJ
Xc£

X ^ X 3

3| §
u ? a&
-3 1)G
X O u
3 G X co
O W w £

U. W QT1

a£ ^ =
£ o

XX
.2? O

j=‘£

^ 9

'00

d>

d> . 33 X
O >, q 0

§« g-f

§ >0 u p
•2 - 2 m g
^ .E x> 2o
Ox.

— _ a aj

3o«-°t
■=■■= « s 0

pS S a£

G ^ 0 d> 3
3 co w CJ u.

■Xi3 c
^2 x_o
^ X X
C< 3 O
MT3

- toS
2.2 >

a2

2'P-cp.

Si

os’3'
“ p a-
•n E <u

T3 M -- . 1

c fi, 7$ x 1
3

a>

3

o c

3 05

3 toE
■° c n

CO Qd_

— 3

G G'X

O G 3

o ^_g
.T3 £

_ - d) u.

£ o >>3!*"'
O S-£ & o

C d>

x

g 5 J2 co c
x G C 3 d>

3

O d)'

£ c

O c to--

! ^ C '

GX qj
> d) >
C >

O u-
T3 d> 3
aj ooX
o G
G.t

—•’OX

a> c «*

^3 g co

— 3 B
G w 3 3
05 XX £

to

a> a>
c ° -
3 X G
^ £ d>

§ 2oc

X 3 .
G O co

«JDX

C.tS co £
d) -> •— • d>
co >X >

S £ g-.SP

■a g «

o o _

3 X co _>^
— O G d,

T> X ^. >

t0G g G

.e£ E £

ja o.

£ 0

> £ us

CT3 £ bo Qd ®

J= £

o >."o

" » o « o
■0 a u p a
c 3 S- o
2 o w c £ p

= m^-2 3.2
S c 5 2 ®S
-21 2-EE

c — <L>
P P t?
rt'S5—

E o §

o i E

d> 3

O O w
S- x -

o CO .

. d) co
co tO <D
X £ G c

2-2 o P

d)

• - o CO >
— 3 tO S •

i_ c c ;

O 3 G 3"

m^'5 I ^
-5 3 x ;

rjG O Q,”

d) ^ 0 O

g^j p.E

G d) 0 d>

r 2 P

mP »I

SSEgSi
= ^ g S S s

o 1/5 ~-a

J-. G O 'O d>
O 'G d> - G>
CcX* ^ n C G
G O-D

G to u- £ 3 E

! <D

3

> ex
37

JO 1/3
co

G 3 1 w co
O d> X tO 3

3J8S3.S

1-^.SES

t. — e o

p -I -o s: 2-3
0 5 e y^ c
tj.S 13 ‘-•a'”,
J2 cs w u m"

3 »— r 3 3

3 £ _2*s

,a>

• 3 dJ ' — - >

3 fi— u, 13 d> <D
rv^ TD 1— >*

— 3 a> a> i> .3
O •- ucoxx o

■— — <u

co

„,X
G £

H g

co

°6S°

CO G
d) d) ■ — 1
-o u d) 1>
! G£;X

Zi ^ .

3 !- ,

! co

5«.g|

co a> -n »-

cox O
O c*-
c— — O
x CD d)

IgsS

C X X G
3 co co X

uT

d) o
3 i_
^ 'd>
u- X
d> ■4“'
GO
SX

C

d) 3 a

E G ^

X' G
X'""
3 X
X 3

£ E

a-

CO

— 3

S «

to

X

A. HALLAM: ENVIRONMENTAL CAUSES OF STUNTING

151

within the confines of a single formation there is a serious danger of misinterpretation
unless the factor of evolution is taken into account.

It is quite conceivable that many of the small Triassic fossils discussed by Boni (1942)
are of the normal size for the stage of evolution to which they belong (cf. Kummel 1948).

CONCLUDING REMARKS

If the interpretations in this review are hedged with qualifications it is not merely
because the many variables are sometimes difficult to disentangle and the data often
inadequate. It is also the result of the diversity of organic adaptation. If there is any
really general rule in this respect, it is that if an environment exists that is tolerable to
life, even though generally regarded as unfavourable, there will be some organisms that
flourish in that environment. Thus although most marine invertebrates react adversely
to brackish water there will always have been some that have achieved a good adaptation
and hence will not exhibit a stunting of growth, just as others can live apparently little-
affected under conditions of low oxygen tension. Therefore in this field one cannot be
hopeful that further research will always produce high correlations or clear-cut results;
it is more likely that the exercise of judgement and discretion will continue to play a
major role in interpretation. In particular, knowledge of variation of the primary factor
in stunting, food supply, is likely to remain elusive to palaeoecologists.

Nevertheless, it is clear that a considerable research field exists that has hitherto been
scarcely touched upon. More critical work is required on both recent and fossil faunas.
There is a strong need for further experiments on living species and quantitative data on
both the size of organisms and their environment. In the case of fossils criteria of
maturity should be sought where possible and the rock from which they were collected
fully described. The author’s interpretation of the principal environmental causes of
stunting, and the main criteria by which they may be distinguished, is summarized in
Table 1.

The study of suspected stunting in fossils can hardly fail to be rewarding, since even
a rejection of stunting in favour of immaturity, if based on sound evidence, will provoke
fresh problems, the solution of which may prove equally illuminating.

Acknowledgements. The writer is greatly obliged to Dr. G. Y. Craig and Dr. H. B. Moore for reading
the first draft of this article and making a number of helpful suggestions.

REFERENCES

ager, d. v. 1956. The geographical distribution of brachiopoda in the British Middle Lias. Quart. J.
geol. Soc. Loud. 112, 157-87.

■ 1963. Principles of paleoecology. New York (McGraw-Hill).

Andrews, e. a. 1940. The snail, Neritina virginea L., in a changing salt pond. Ecology, 21, 335-46.

archanguelsky, a. d. 1927. On the Black Sea sediments and their importance for the study of sedi-
mentary rocks. Bull. Soc. not. Moscow, 35, 264-89.

arkangel’skaya, n. a. and grigor’yev, v. n. 1961. Formation of halogenic zones in marine basins
illustrated by the example of the Lower Cambrian evaporite basin of the Siberian platform. Izvestiya
Acad. Sci. U.S.S.R. geol. ser., 1960, no. 4 (A.G.I. translation 1961).

arkell, w. j., 1933. The Jurassic system in Great Britain. Oxford.

ball, j. r., 1935. Dwarfed gastropods in the basal Guttenberg, S.W. Wisconsin, Proc. geol. Soc.
Amer. 384.

152 PALAEONTOLOGY, VOLUME 8

barker, d., 1963. Size in relation to salinity in fossil and recent euryhaline ostracods. J. mar.
biol. Ass. U.K. 43, 785-95.

bateson, w. 1889. On some variations of Cardium edule apparently correlated to the conditions of life.
Phil. Traits, roy. Soc. B 180, 297.

bauer, v. 1929. Ueber das Tierleben auf den Seegraswiesen des Mittelmeeres. Zool. Jb. 56, 1-42.
black, m. 1933. The precipitation of calcium carbonate on the Great Bahama Bank. Geol. Mag.
70, 455-66.

boettger, c. r. 1950. Ein Beitrag zur Frage des Ertragens von Brackwasser durch Molluskenpopula-
tionen. Hydrobiologia, 2, 360.

- 1952. Grossenwachstum und Geschlechtsreife bei Schnecken und pathologischer Riesenwuchs
als Folge einer gestorten Wechselwirkung beider Factoren. Verb. Dtsch. Zool. Ges. Freiburg, 468.

boni, a. 1942. Faune pigmee triassiche; contributo alia paleobiologia del nanismo di faune marine.
Boll. Soc. geol. It at. B 61, 228-69.

bradley, j. h. 1921. The Brachiopoda of the Maquoketa of Iowa. Bull. Mus. comp. Zool. Harv.
64, 503-25.

bradshaw, j. s. 1957. Laboratory studies on the rate of growth of the foraminifer ‘ Streblus beccarii
(Linne)' var. tepida (Cushman). J. Paleont. 31, 1138-47.

1961. Laboratory experiments on the ecology of the foraminifera. Coutr. Cushman Lab. foram.

Res. 12, 87-106.

branson, c. c. 1930. Paleontology and stratigraphy of the Phosphoria formation. Missouri Univ.
Studies, 5, no. 2, 1-61.

broadhurst, f. m. 1959. Anthraconaia pulchella sp. nov. and a study of palaeoecology in the Coal
Measures of the Oldham area of Lancashiie. Quart. J. geol. Soc. Loud. 114, 523-41.
brown, c. j., clark, c., and gleissner, b. 1938. The size of certain naiades from western Lake Erie
in relation to shoal exposure. Amer. Midi Nat. 19, 682.
bushinski, G. i. 1964. On shallow water origin of phosphorite sediments. In Deltaic and shadow
marine deposits, Ed. L. M. J. V. van Straaten. Elsevier.
callomon, j. h. 1957. Field meeting in the Oxford Clay of Calvert and Woodham Brick Pits, Bucking-
hamshire. Proc. Geol. Ass. Loud. 68, 61-64.

chao, y. t. 1927. Fauna of the Taiyan formation of North China — Pelecypoda. China geol. Surv.
ser. B, 9, 1-48.

cloud, p. e. 1948. Assemblages of diminutive brachiopods and their paleoecological significance.
J. sediment. Petrol. 18, 56-60.

coe, w. r. 1942. Influence of natural and experimental conditions in determining shape of shell and
rate of growth in gastropods of the genus Crepidula. J. Morphol. 71, 35-47.

and fox, d. l. 1944. Biology of the California sea-mussel ( Mytilus calif ornianus). III. Environ-
mental conditions and rate of growth. Biol. Bud. Woods Hole , 87, 59-72.
crabb, e. d. 1929. Growth of a pond snail Lymnaea stagnalis appressa as indicated by increase of
shell-size. Biol. Bud. Woods Hole 56, 41-63.

craig, g. y. and hallam, a. 1963. Size-frequency and growth-ring studies of Mytilus edulis and Car-
dium edule, and their palaeoecological significance. Palaeontology, 6, 731-50.
croneis, c. and grubbs, d. m. 1939. Silurian sea-balls. J. Geol. 47, 598-612.

cumings, e. r., beede, j. w., branson, e. b., and smith, e. a. 1906. Fauna of the Salem Limestone.

Indiana Dept. Geol. & Nat. Res., 30th Ann. Rep. 1189-1394.
dacque, e. 1915. Grundlagen und Methoden der Paldogeographie. Jena.
de lapparent, a. 1906. Trade de geologie. Vol. II. Paris.

doochin, h. and smith, f. g. w. 1951. Marine boring and fouling in relation to velocity of water cur-
rents. Biol. mar. Sci. Gulf, and Caribb. 1, 196-208.
eager, r. m. c. 1948. Variation in shape of shell with respect to ecological station. A review dealing
with recent Unionidae and certain species of the anthracosiidae in Upper Carboniferous times.
Proc. roy. Soc. Edinb. B 63, 130-48.

- 1952. Growth and variation in the non-marine lamellibranch fauna above the Sandrock Mine of
the Lancashire Millstone Grit. Quart. J. Geol. Soc. Lond. 107, 339-73.

emery, k. o. and rittenberg, s. c. 1952. Early diagenesis of California basin sediments in relation to
origin of oil. Bud. Amer. Ass. Petrol. Geol. 36, 735-806.

A. HALLAM: ENVIRONMENTAL CAUSES OF STUNTING

153

fisher, a. g. 1961. Latitudinal variations in organic diversity. Amer. Scientist 49, 50-74.
fisher, d. w. 1951. Marcasite fauna in the Ludlowville formation of western New York. J. Paleont.
25, 365-71.

ford, e. b. 1928. The inheritance of dwarfing in Gammarus chevreuxi. J. Genetics , 23, 93-102.
fox, d. l. 1957. Particulate organic detritus. Geol. Soc. Amer. Mem. 67, vol. 1, 383-9.
fox, h. m. 1929. Cambridge expedition to the Suez Canal, 1924. XXXIX. Summary of results.
Trans zool. Soc. Land. 22, 843-63.

■ and taylor, a. e. r. 1955. The tolerance of oxygen by aquatic invertebrates. Proc. roy. Soc. Loncl.

B 143, 214-25.

Fuchs, t. 1871 . Ueber die locale Anhaufung kleiner Organismen und insbesondere fiber die Fauna von
St. Cassian. Verb. K. u. K. Geol. Reichsanst. Wien , 12, 204-206.
gekker, r. f. 1957. Introduction to Paleoecology. Moscow. (In Russian. French translation, Bases de
la paleoecologie, by 3. Roger, 1960. Paris. Bur. Recherches Geol. Min.)
gibson, f. a. 1956. Escallops (Pecten maximus L.) in Irish waters. Sci. Proc. R. Dublin Soc. 27, 253.
goldring, w. The Champlain Sea. Bull. N.Y. St. Mas. nos. 239-40, 153-94.

goreau, t. f. 1959. The physiology of skeleton formation in corals. I. A method for measuring the rate
of calcium deposition by corals under different conditions. Biol. Bull. Woods Hole , 116, 59-75.
grabau, a. w. 1917. Stratigraphic relationships of the Tully Limestone and Genesee Shale in eastern
North America. Bull. geol. Soc. Amer. 28, 945-58.

1931. The Permian of Mongolia. In The Natural History of Central Asia. Amer. Mus. not. Hist.

4, 1-665.

gdnter, g. 1947o. Paleoecological import of certain relationships of marine animals to salinity.
J. Paleont. 21, 77-79.

19476. Extended remarks on relationships of marine animals to salinity. Ibid. 21, 498.

1950. Seasonal population changes and distributions, as related to salinity, of certain inverte-
brates of the Texas coast, including the commercial shrimp. Publ. Inst. mar. Sci. Univ. Texas, 1,
7-51.

hallam, a. I960. A sedimentary and faunal study of the Blue Lias of Dorset and Glamorgan. Phil.
Trans, roy. Soc. B 243, 1-44.

1961. Brachiopod life assemblages from the Marlstone Rock-bed of Leicestershire. Palaeontology,

4, 653-9.

1963. Observations on the palaeoecology and ammonite sequence of the Frodingham Ironstone

(Lower Jurassic). Ibid. 6, 554-74.

Hancock, d. a. and simpson, a. c. 1961. Parameters of marine invertebrate populations. In The ex-
ploitation of natural animal populations, editors E. D. le Cren and M. W. Holdgate. Oxford.
hohenstein, v. 1913. Beitrage zur Kenntnis des mittleren Muschelkalks und des unteren Trochiten-
kalks am ostlichen Schwarzwaldrand. Geol. Palciont. Abh. 16, 175-272.

Humphrey, r. r. and macey, r. w. 1930. Observations on some of the probable factors controlling the
size of certain tidal pool snails. Publ. Puget Sd. Mar. (biol.) Stat. 7, 205-8.
isham, l. b., moore, h. b., and smith, f. g. w. 1951. Growth rate measurement of shipworms. Bull,
mar. Sci. Gulf, and Caribb. 1, 136-47.

jaworskj, e. 1923. Die marine Trias in Sfidamerika. Neues Jb. Min. Geol. Palciont. 47, 93-200.
jekelius, e. 1935. Der weisse Triaskalk von Brasov und seine Fauna. Ann. Inst. geol. Roman. 17, 1-107.
jones, n. s. 1950. Marine bottom communities. Biol. Rev., 25, 283-313.

kauffman, e. g. (in press). Middle and late Turonian oysters of the Lopha lugubris group. Smithsonian
Misc. Contr.

kindle, e. m. 1930. The intertidal zone of the Wash, England. Repr. nat. Res. Conn. Wash. no. 92, 5.
kristensen, i. 1957. Differences in density and growth in a cockle population in the Dutch Wadden
Sea. Arch, neerl. Zool. 12, 351.

kuhn, o. 1936. Eine mittelmiocane Zwergfauna aus Kreta und die Entstehung mariner Zwergfaunen.
Zbl. Miner. Geol. Palciont. Abt. B, 255-68.

kummel, b. 1948. Environmental significance of dwarfed cephalopods. J. sediment. Petrol. 18,
61-64.

kutassy, a. 1930. Eine mittelmiocane Zwergfauna aus Ungarn und ihre Entstehungsbedingungen.
Zbl. Miner. Geol. Palciont., Abt. B, 194-204.

154 PALAEONTOLOGY, VOLUME 8

ladd, h. s. 1925. The stratigraphy and fauna of the Maquoketa Shale of Iowa. Doctoral cliss., Iowa
State Univ. 1-214.

hedgpeth, j. w., and post, r. 1957. Environments and facies of existing bays on the central Texas

coast. Geol. Soc. Amer. Mem. 67, vol. 2, 599-639.
lalicker, c. g. 1948. Dwarfed protozoan faunas. J. sediment. Petrol. 18, 51-55.
le calvez, j. and y. 1951. Contribution a l’etude des Foraminiferes des eaux saumatres. 1. Etangs de
Canet et de Salses. Vie et Milieu 2, 237-54.

loeblich, a. r. and tappan, h. 1964. Foraminiferal facts, fallacies and frontiers. Bull. geol. soc Amer.
75, 367-92.

loomis, f. g. 1903. The dwarf fauna of the pyrite layer at the horizon of the Tully Limestone in western
New York. Bull. N.Y. St. Mus. no. 69, 892-900.

loosanoff, v. l. and tommers, F. d. 1948. Effect of suspended silt and other substances on rate of
feeding of oysters. Science, 107, 69-70.

marr, j. e. 1925. Conditions of deposition of the Stockdale Shales of the Lake District. Quart. J. geol.
Soc. Loud. 81, 112-33.

mcnulty, j. k., work, r. c., and moore, h. b. 1962. Level sea bottom communities in Biscayne Bay
and neighbouring areas. Bull. mar. Sci. Gulf and Caribb. 12, 204-33.
miller, d. n. 1953. Ecological study of the foraminifera of Mason Inlet, North Carolina. Contr. Cush-
man Lab. for am. Res. 4, 41-63.

milne, a. 1940. The ecology of the Tamar estuary. IV. J. mar. biol. Ass. U.K. 24, 69-87.
moore, h. b. 1935. The biology of Balanus balanoides. IV. Relation to environmental factors. Ibid. 20,
279-307.

1936a. The biology of Balanus balanoides. V. Distribution in the Plymouth area. Ibid. 701-16.

1936 b. The biology of Purpura lapillus. I. Shell variation in relation to environment. Ibid. 21,

61-89.

1937. A comparison of the biology of Echinus escu/entus in different habitats. Ibid. 711-20.

1958. Marine ecology. New York.

Newell, N. d. 1949. Phyletic size increase — an important trend illustrated by fossil invertebrates.
Evolution, 3, 103-23.

pack, d. a. 1919. Two ciliata of Great Salt Lake. Biol. Bull. Woods Hole, 36, 273-6.
patton, j. b. 1953. Petrology of the Salem Limestone (Indiana building stone). Econ. Geol. 48, 331.
percival, e. 1944. A contribution to the life history of the brachiopod Terebratella inconspicua Sowerby.
Trans, roy. Soc. N.Z. 74, 1-23.

phleger, f. 1960. Ecology and distribution of recent foraminifera. O.U.P.

price, w. a. 1920. Maximum size of West Virginia Derbyas as influenced by sedimentation. W. Va.
geol. Surv. (Co. Rep.) 545-50.

rawlinson, R. 1934. A comparative study of Metridium senile (l.) var. dianthus (Ellis) and a dwarf
variety of this species occurring in the River Mersey, with a discussion of the systematic position of
the genus Metridium. J. mar. biol. Ass. U.K. 19, 901-19.

Richardson, L. 1907. The Inferior Colite and contiguous deposits of the Bath-Doulting district.
Quart. J. geol. Soc. Loud. 63, 383-423.

richter, r. 1931. Tierwelt und Umwelt im Hunriickschiefer. Zur Entstehungeinesschwarzen Schlamm-
steins. Senckenbergiana, 13, 299-342.

ruedemann, R. 1935. Ecology of black mud shales of eastern New York. J. Paleont. 9, 79-91.
said, r. 1953. Foraminifera of Great Pond, East Falmouth, Massachusetts. Contr. Cushman Lab.
For am. Res. 4, 7-14.

sarasin, f. 1913. Neu-Caledonien. Z. Ges. Erdk. Berk no. 8, 585,

1917. Neu-Caledonien und die Loyalty-Inseln Reise — Erinnerungen eines Naturforschers. Basel.

savage, r. f. 1956. The great spatfall of mussels (Mytilsu edulis L.) in the River Conway estuary in
spring 1940. Fish. Invest. (2), 20, 7. H.M.S.O.
schmidt, h. 1951. Erkennbarkeit fossiler Brackwasserabsatze. Z. Dtsch. Geol. Ges. 103, 9-16.
scott, g. 1924. Some gerontic ammonites of the Duck Creek formations. Texas christ. Univ. Quart.
1, 5-31.

1940. Paleoecological factors controlling the distribution and mode of life of Cretaceous ammo-
noids in the Texas area. J. Paleont. 14, 299-323.

A. HALLAM: ENVIRONMENTAL CAUSES OF STUNTING

155

scott, h. w. 1948. Significance of crustaceans in dwarfed faunas. /. sediment. Petrol. 18, 65-70.
segerstrale, s. G. 1957. Baltic Sea. Geol. Soc. Amer. Mem. 67, vol. 1, 751-800.
shimer, h. w. 1908. Dwarf faunas. Amer. Midi. Nat. 42, 472-8.

sorgenfrei, t. 1958. Molluscan assemblages from the marine Middle Miocene of south Jutland, and
their environments. Damn. geol. Unders. II. Raekke. 79, 1-503.
spooner, g. m. 1947. The distribution of Gammarus species in estuaries. Part 1. J. mar. biol. Hss. U.K.
27, 1-52.

stauffer, c. r. 1937. A diminutive fauna from the Shakopee dolomite (Ordovician) at Cannon Falls,
Minnesota. J. Paleont. 11, 55-60.

stehli, f. g. and halsey, c. E. 1963. Paleontologic technique for defining ancient pole positions.
Science, 142, 1057-9.

tasch, p. 1953. Causes and paleoecological significance of dwarfed fossil marine invertebrates. J.
Paleont. 27, 356-444.

1957. Fauna and paleoecology of the Pennsylvanian Dry Shale of Kansas. Geol. Soc. Amer. Mem.

67, vol. 2, 365-406.

termier, h. and termier, g. 1951 . Les herbiers marins et la signification des faunes pyriteuses. Extraits
de la Revue scientifique, no. 3309, 16-48.

thorson, g. 1957. Bottom communities (sublittoral or shallow shelf). Geol. Soc. Amer. Mem. 67, vol.
I, 461-534.

trechmann, c. t. 1913. On a mass of anhydrite in the Magnesian Limestone of Hartlepool, and on
the Permian of South-eastern Durham. Quart. J. Geol. Soc. Loud. 69, 184-218.
twenhofel, w. h. 1915. Notes on black shale in the making. Amer. J. Sci. 40, 272-80.
vogel, k. 1959c. Zwergwuchs bei Polyptychiten ( Ammonoidea). Geol. Jb. 76, 469-531.

19596. Wachstumsunterbrechungen bei Lamellibranchiaten und Brachiopoden. Neues Jb. Miner.
Geol. Paldont. Abb. 109, 109-129.

vokes, h. e. 1948. Cretaceous mollusca from depths of 4,875 to 4,885 feet in the Maryland Esso well.

In Cretaceous and Tertiary subsurface geology. Maryland Dept. Geol. Bull. 2, 126-51.
williams, j. 1937. Pennsylvanian invertebrate faunas of southeastern Kansas. Bull. Kans. geol. Surv.
29,92-122.

wimpenny, r. s. 1941. Organic polarity: some ecological and physiological aspects. Quart. Rev. Biol.
16, 389-425.

yonge, c. m. 1961. Life and environment on the bed of the sea. Adv. Sci. 18, 383.

A. HALLAM

Grant Institute of Geology,
University of Edinburgh,
Edinburgh 9

Manuscript received 16 February 1964

THE MODE OF LIFE OF TWO JURASSIC SPECIES
OF ‘POSIDONIA' (BIVALVIA)

by R. p. s. Jefferies and p. minton

Abstract. On facies grounds Bositra buchi (Romer) was almost certainly not benthonic, but probably nekto-
planktonic rather than pseudoplanktonic, since other possible pseudoplankton is rare in its typical facies, there
is no sign of attachment at any stage of life, and there are morphological signs of swimming ability (anterior
and posterior gapes, thin shell, and probable wide angle of opening in life). Experiment indicates that this mode
of life is feasible. The hinge of B. buchi is described for the first time and the species is transferred from Posidonia
to Bositra de Gregorio 1886.

' Posidonia' radiate Goldfuss was also almost certainly not benthonic on facies grounds. It was probably not
pseudoplanktonic since it never occurs attached to wood like indubitable pseudoplankton in the same beds, and is
much more abundant than such forms. Morphological evidence confirms that it was probably nektoplanktonic.

The method employed in the feasibility experiments should be widely applicable in studies of the functional
morphology of small organisms.

This paper deals with the mode of life of Bositra buchi (Romer) (= Posidonia aipina
and P. ornati of authors) and ‘ Posidonia ’ radiata Goldfuss ( Posidonia (Steinmannia)
bronni magna of authors). These bivalves are of interest because they belong to a group
widely regarded as pseudoplanktonic.

The most important sources of evidence concerning the mode of life of fossil animals
in general are facies, analogy with living relatives, and functional morphology. In the
present case facies is the most important of these. Direct analogy with living relatives
is unreliable but functional morphology is useful, particularly since many structures are
analogous to those of living relatives.

B. buchi and ‘ P radiata belong to a group of thin-shelled Pectinacea which character-
istically occur in black shales, together with ammonites or goniatites, in rocks of
Devonian to Jurassic age (Newell et ah 1953). The group includes Bositra, Posidonia ,
Steinmannia, Daonel/a, Halobia, Monotis, Buchio/a, and Dunbarel/a, and three main
views have been expressed about its ecology. A benthonic mode of life was assumed by
Pompeckj (1901, p. 178), Haug (1907, p. 152), Vadasz (1910, p. 41), Krumbeck (1921,
p. 68; 1924, p. 128), Craig (1954, p. 108), and Zangerl and Richardson (1963, p. 135).
A pseudoplanktonic mode of life, attached to floating seaweed, was suggested by Clarke
(1904, pp. 199, 215), Pompeckj (1914, p. 458), Hundt (1939), Hudson and Cotton (1943,
p. 149), Schwarzacher (1948, p. 38), Newell et al. (1953, p. 14), Newell (1955, pp. 13,
22), Nalivkin (1956, p. 188), Allan (1956, p. 369), Ichikawa (1958, p. 183), and Sadykov
(1962, p. 86). Finally, the view that members of the group were pelagic and free-swim-
ming was put forward by Molengraaf (1917, p. 255, not seen but see Krumbeck 1921,
p. 68) — who compared Halobia and Daonella with pteropods — Guillaume (1928, p. 227),
and Prokovlev (1959, p. 122). The views of Weigelt (1922, 1927) and Paul (1939) on the
mode of life of Posidonia becheri and of Hauff (1921) on ‘ Posidonia ’ radiata and Stein-
mannia bronni are discussed in more detail below. Recent works on Mediterranean
microfacies refer to B. buchi, S. bronni and ‘TV radiata as pelagic without discussing
whether they were plankton or pseudoplankton (Peyre 1959, agip 1959).

[Palaeontology, Vol. 8, Part 1, 1965, pp. 156-85, pi. 19.]

JEFFERIES AND MINTON: MODE OF LIFE OF JURASSIC ‘ POSIDONIA' 157

Nearly all authors who have recently considered the matter in detail (e.g. Newell
1955, Ichikawa 1958, and Allan 1956) have strongly favoured a non-benthonic mode of
life for this group of Pectinacea on facies grounds, regarding the animals as attached
to seaweed. Schmidt (1935) was the first to give wide currency to this view. On the other
hand, Craig (1954) believed that the Carboniferous Posidonia corrugata was benthonic,
but this was based merely on comparison with living Pectinidae and is contradicted by
his admirably detailed facies evidence.

THE MODE OF LIFE OF BOSITRA BUCHI (ROMER)

Bositra buchi (Romer) (Posidonomya alpina and Posidonia ornati of authors) is a pecti-
nacean of Toarcian to Oxfordian age. The shell is very thin (60-70 /x at a length of 1 cm.)
with a prismatic, calcitic, outer layer, and a laminar, nacreous, inner layer; muscle scars
are unknown; the outline is almost circular or elliptical, with a height/length ratio of
0-75 to 0-90, and is slightly oblique with prosogyrous, somewhat anterior umbones;
rounded, concentric folds of variable spacing, which do not affect the shell thickness, are
separated by angular furrows; weak, radial striae are sometimes present; wide anterior
and posterior gapes are visible in uncrushed specimens; there is no pteriiform stage in the
life history; the valve margins are often brownish.

The cardinal area (text-fig. 1, PI. 19, fig. 9), which
is somewhat thicker-shelled than the rest of the
shell, has been seen in material collected by Mr.

H. G. Owen from the Oxford Clay of Elstow, Beds.

(data with PI. 19, fig. 9). A shallow, triangular,
median pit marked with strong growth lines
occupies 45 per cent, of the hinge length; it is the
site of the internal ligament, of which traces some-
times remain in the Elstow material. The anterior
and posterior portions of the hinge, each repre-
senting about 27 per cent, of its length, are the
site of the external ligament. Steinmann’s descrip-
tion of the cardinal area (in Wanner 1907, p. 204)
is incorrect.

Bositra buchi differs from the type species of Posidonia (P. becheri Bronn) in two im-
portant respects (cf. Weigelt 1922). The hinge is of pteriid rather than arcid type, since it
has no chevrons for the attachment of external ligaments, and the life history is much
simpler (see below, p. 170). These differences demand a separate genus and the name
Bositra de Gregorio (1886, p. 11) is available. Posidonia ornati Quenstedt has been
selected as type species of Bositra (Cox 1964, p. 47), and P. ornati , as explained below,
is a subjective junior synonym of Posidonia buchi Romer.

The family relationships of B. buchi are somewhat obscure at present. The Pterino-
pactinidae, in which Newell (1937, p. 37) placed Posidonia , have an arcid ligament and
cannot include Bositra. Nevertheless, B. buchi may be descended from Posidonia becheri
or one of its relatives, since it has in common the very thin shell consisting of prismatic
and laminar layers, similar outline and ornament, a long straight hinge line, and occur-
rence in the same very unusual black shale facies. The pteriid hinge of Bositra resembles

I mm

text-fig. 1. Bositra buchi (Romer)
showing the cardinal area (LL17400).
Data as Pi. 19, fig. 9.

158

PALAEONTOLOGY, VOLUME 8

that of an aviculopectinid, but may be an independent development by neoteny from
an arcid hinge since Bernard (1896) has shown that modern Arcacea have a pteriid
hinge when very young. The process is paralleled in the Arcacea by the evolution of
Limopsis from Glycitmris.

With regard to the specific name, Guillaume (1928) ignored priority by favouring
Posidonomya alpina Gras 1852 over its synonyms Posidonia ornati Quenstedt 1851 and
Posidonia buchii Romer 1836. The form described by Romer (1835-6, p. 8, pi. 4, fig. 8)
from the Lower Bathonian ‘ Walkererde’ of North Germany is certainly identical to the
other two, since the ornament and position of the umbo are the same, the specimen is in
the characteristic valves-open position, and is described as having a straight hinge line
and a very thin shell. The height/length ratio is lower in the type figure of P. buchii than
in those of P. alpina and P. ornati , but this probably represents intraspecific variation
since the tall form occurs in the Bathonian (e.g. LL 17414 from the Fuller’s Earth of
Normandy).

Facies Relationships

Pelagic organisms are independent of the nature of the sea bottom and are widespread
geographically. Their fossil remains are similarly widespread, and occur in a great
variety of deposits. They are often abundant in deeper-water deposits, associated almost
exclusively with plankton and nekton, even where the bottom waters seem to have been
foul and uninhabitable. They may, however, also occur in shallow-water deposits with
abundant benthos. The graptolites are the classic illustration of this. The conclusion
that an extinct animal was pelagic is thus primarily based on facies, but must not be
contradicted by its morphology.

The Typical Bositra buchi Facies. The typical facies in which B. buchi occurs is, as Flaug
(1907, p. 152) pointed out, very like the graptolitic facies. It consists mainly of shale,
often bituminous, with subordinate limestone and radiolarian chert. Apart from B. buchi ,
which is often exceedingly abundant, almost the only macrofossils are ammonites, often
including Phylloceras and Lytoceras , together with aptychi and cephalopod ( ? ammonite)
beaks ( Rhynchoteuthis ), though tracks, including Zoophycus, may be present. This facies
is widespread in Middle Jurassic rocks, and according to Arkell (1956) occupies large
areas in south-east France, the eastern Alps, the Carpathians, Lombardy, Peninsular
Italy, Greece and Albania, the Betic Cordillera, Morocco and Algeria, Anatolia, the
Crimea, the Caucasus, Azerbaijan, Persia and Iraq, Kenya and Tanganyika, and the
Southern Andes. Boehm (1912, p. 130) recorded it in the Sula Islands of Indonesia.
Slight signs of posthumous drifting may be present in the typical B. buchi facies, but there
is no possibility that such drifting explains the regular occurrence of the species, for the
individual areas where the facies is developed are often hundreds of miles across.

Comparison of figures of geographically separate specimens of B. buchi from, for
example, Peru (Steinmann 1881, pi. 10, figs. 3, 4), Indonesia (Boehm 1912, pi. 32, fig. 2),
and the Crimea (Stremoouchow 1895, pi. 10, figs. 1-8), with those given here (PI. 19)
from France and Iraq shows a striking resemblance in form.

Few detailed studies of the typical B. buchi facies have been made. Haug (1892), in
a classic work, recognized an area of deep-water sediments of Bajocian to Callovian
age in the French Alps. In this ‘facies dauphinois’ B. buchi and ammonites were by

JEFFERIES AND MINTON: MODE OF LIFE OF JURASSIC ‘ POSIDONIA ’ 159

far the commonest fossils. Indeed Hang (p. 59) said that ‘Les cephalopodes [mainly
ammonites] et les posidonomyes [B. buchi] sont, pour ainsi dire, les seuls fossiles que
Ton rencontre dans le Bajocien et dans le Bathonien de nos chaines subalpines.’ It is
true that this somewhat overstates the position since tracks, which Haug called Cancel-
lophycus {Zoophycus), are sometimes abundant, though in places absent, and Inoceramus
and belemnites occur at one horizon (Lower Bajocian, op. cit., p. 62), but the broad
impression conveyed is accurate. In the Bajocian the ‘facies dauphinois’ extends con-
tinuously some 290 km. from north to south within France and 130 km. from east to
west (op. cit., fig. 14).

One of the authors (R.P.S.J.) has examined a small quarry in rocks of Bathonian or
Callovian age in the ‘facies dauphinois’ ( 0 - 6 km. north of the Church at St. Barthelemy-
de-Vif, Isere). The rocks exposed were strongly folded, brown, bituminous shales with
subordinate bituminous calcarenitic limestones. In order of decreasing abundance the
fossils found during five days’ collecting consisted of B. buchi , which thickly covered
many of the bedding planes, tracks (not Zoophycus) mainly associated with the lime-
stones, ammonite conchs (indeterminate pachydiscids and a phylloceratid), ammonite
beaks (Rhynchoteuthis), aptychi. Pinna (one specimen only), and two spines of a diade-
matoid sea urchin. A count of all the macrofossils encountered in 60 cm. of rock, of
which 10 cm. was a band of limestone and the rest shale, revealed 1 19 single valves and
numerous fragments of B. buchi , 2 complete ammonites and some fragments, and nothing
else. All the fossils came from the shale except for one valve of B. buchi in the limestone.
The comparative rarity of fossils in the limestones suggests that the latter were deposited
quickly, possibly as turbidites.

Contemporary shallow-water rocks exist to the east and south of the Dauphine
facies (Briangon and Provence facies) and have abundant benthos, including many
bivalves, brachiopods, echinoderms, &c., but B. buchi is not recorded.

There is no doubt, therefore, that in south-east France B. buchi had a different facies
distribution to other bivalves. It is rare where benthonic fossils are abundant, and
abundant where benthonic fossils (except for tracks) are rare. Its common associates,
the ammonites, were most probably nektonic (cf. Diener 1912). It is true that benthos is
not quite absent in the Dauphine facies since, apart from the single Pinna and echinoid
fragments noted above, tracks also occur. Tracks, however, particularly grazing tracks
like Zoophycus, are necessarily much more abundant than the causative organism.

In Iraq a close analogue to the Dauphine facies is provided by the thin-bedded,
highly bituminous shales and limestones of the mainly Bajocian-Bathonian Sargelu
Formation (PI. 19, fig. 1) (Dunnington in van Bellen et a/. I960, p. 250). In its type sec-
tion in Kurdistan this formation is 115 m. thick; the top 12 m. contains innumerable
specimens of Bositra buchi, together with ammonites and fragments of wood; the middle
21 m. contains poor ammonite impressions; and the lowest 82 m. contains rhyncho-
nellids, Gryphaea cf. balli, ‘ Posidonia cf. opalina ’ and ammonites. The top of the Sargelu
Formation therefore presents the B. buchi facies in very typical form. In thin section the
formation is readily recognized by abundant specimens of ‘ Paleotrix', which Peyre
(1959) showed to be transverse sections of B. buchi. ‘Ostracods’ were also recorded, but
these are probably juvenile specimens of B. buchi, which are very common in the rock
and scarcely distinguishable from ostracods in thin section. Other fossils include radio-
laria and ‘minute frondiculariids and nodosariids’ (Foraminifera). The Iraqi analogue

160

PALAEONTOLOGY, VOLUME 8

to the Provence or Briangon facies is the Bathonian Muhaiwir Formation, equivalent
to the upper part of the Sargelu. This consists of well-bedded limestones and marly
limestones, sandstones, and oolites, with abundant rhynchonellids, terebratulids,
echinoids and gastropods together with bivalves including Mactromya, Ceratomya,
Homomya, Pho/adomya, Mytilus, and Eligmus; B. buchi is unknown, and therefore
shows much the same facies relationships in Iraq as in Provence. Indeed Dunnington
(loc. cit.) believed, presumably with particular reference to the upper part of the forma-
tion, that the Sargelu was euxinic, since benthos is rare and the rock highly bituminous;
this being the case, B. buchi could not have been benthonic. The typical development of
the Sargelu Formation is very extensive, being found from Kurdistan to Kuwait over a
distance of 600 km. It is continuous with the B. buchi facies of Azerbaijan, the Caucasus,
and Persia.

In the Bajocian of Argentina Grober (1918) described a very similar contrast of facies.
The area of typical B. buchi facies is here 300 km. from north to south and 200 km. from
east to west.

In the Middle Jurassic of Greece and Albania the peculiar nature of the facies in
which B. buchi occurs was emphasized by Renz (1927, p. 487). The rocks consist mainly
of yellow, grey, or black shales and limestones with abundant examples of B. buchi
(quoted as P. alpina), aptychi and beaks of ammonites, and, in one bed Pseudomonotis
cf. substriata Ziethen. No other fossils were found. The presence of ammonite aptychi
and beaks but not their conchs can be explained by differential solution; ammonite
conchs, being aragonitic, are more soluble than the calcite of aptychi or the outer layers
of B. buchi. This phenomenon seems to have led Arkell (1956, p. 585) to describe the
B. buchi facies as ‘inhospitable’ for ammonites, although elsewhere in the same book he
quoted long lists of ammonites from the facies. At St. Barthelemy the ammonites are
preserved as shell-less moulds while aptychi and B. buchi retain their shells. Differential
solution has operated here, therefore, and slightly different conditions might have
destroyed the ammonites completely.

One of the most thorough thin-section studies of the B. buchi facies was made by
Colom (1955, 1957) in the Betic Cordillera. The sections of B. buchi in his slides were
rather confusingly referred to as ‘ HalobicT, though he said (1955, p. 112) that they were
sections of P. alpina (B. buchi). The associated microfauna is probably entirely plank-
tonic for it consists largely of globigerinids, while beds with abundant specimens of
B. buchi tend to alternate with beds with abundant radiolaria. The macrofauna consists
of B. buchi and ammonites.

Bositra buchi outside its Typical Facies. If B. buchi were found only in the typical B. buchi
facies, scarcely ever had indisputable benthonic associates, and never occurred in
shallow- water rocks, one might suspect that it was a deep-water form, insensitive to lack
of oxygen. But such is not the case; B. buchi also occurs in rocks like those of its typical
facies but with an abundant benthos of a few species, and in shallow-water deposits with
abundant and varied benthos.

The Callovian rocks of la Voulte-s. -Rhone illustrate the first group. Hess (1960)
described a sequence of calcareous marls and clays from this locality containing abun-
dant examples of B. buchi, the ophiuroid Ophiopinna e/egans (Heller), benthonic
foraminifera, and a few specimens of a probably benthonic cumacean crab. Hess noted

JEFFERIES AND MINTON: MODE OF LIFE OF JURASSIC POSIDONIA ’ 161

(p. 377) that some beds contained B. buchi alone, often abundantly, but that beds with

O. elegans and foraminifera always contained B. buchi as well. This suggests that the
sea floor was sometimes uninhabitable and that B. buchi was independent of the
bottom.

A rather similar case, but with a more varied benthos, occurs in the 40 feet of brown
shale at the base of the Oxford Clay at Elstow. B. buchi is by far the commonest fossil
here but is associated with Me/eagrinella, Nuculana, Procerithium, an aporrhaid, and
scolecodonts, as well as ammonites {Kosmoceras jasoni).

Cases where B. buchi occurs in shallow-water rocks, of very different lithology to its
typical facies and containing an abundant and varied benthos, include:

1. Sandstone; ‘gres a Murchisonae’ of Alsace (Guillaume 1928, p. 227).

2. Shell fragment limestone; calcareous beds in Fuller’s Earth Clay near Sherborne
(thirteen more or less complete valves and many fragments, L84734-8); Inferior Oolite,
schloenbachi Zone near Bradford Abbas (twelve more or less complete valves and some
fragments, L84320-3).

3. Iron-shot oolite; ‘oolite ferrugineuse de Bayeux’ (Guillaume 1928, p. 227).

The numerous specimens in the British Museum collections just quoted show that
B. buchi can be quite abundant when it occurs at all in shallow-water rocks. These occur-
rences are therefore not fortuitous, but need explanation. There is no difficulty in explain-
ing them if B. buchi was pelagic. The sporadic nature of shallow-water occurrences is
probably due to the thin shell (60-70 p, see below, p. 164). In shallow-water rocks such
a shell would usually break into unrecognizable fragments before burial; indeed it often
fragmented before burial even in the Dauphine facies of St. Barthelemy. It can be com-
pared with a living pelagic mollusc : the thecosomatous pteropod Spirate/la ( = Limacina)
is said to be the most numerous living mollusc but its fragile shell is a rarity in Recent,
marine, shallow-water deposits.

The facies evidence for a pelagic mode of life for B. buchi may be summarized as
follows: the species is extremely widespread geographically, is unlike contemporary
bivalves or other indisputable benthos in its typical mode of occurrence, and resembles
in its facies relationships the ammonites, radiolaria, and globigerinids, which were
probably planktonic or nektonic. Its abundance in conditions that were probably
euxinic and its sporadic occurrence, in considerable numbers, in rocks of shallow-water
origin confirm that it was pelagic.

Nothing in the facies evidence is contradicted by the morphology of the shell. This
was thin, as might be expected in a planktonic organism, the prodissoconch develops
quite gradually into the adult shell without sign of settlement, and there are morpho-
logical signs of swimming ability (see below, p. 162).

Facies evidence bearing on the pseudoplanktonic hypothesis. If B. buchi was not benthonic
it could have been either planktonic or pseudoplanktonic. Facies considerations do not
decisively favour one of these possibilities over the other. However, if B. buchi was
attached to flotsam it is surprising that other animals with a similar mode of life are so
rare in the typical B. buchi facies. None were found at St. Barthelemy. It is true that
Pinna , the only bivalve discovered other than B. buchi , was byssiferous, but like Recent
members of the genus which it closely resembles, this form almost certainly lived um-
bones down in the sea floor, with the byssus spread out in the mud.

B 6012

M

162

PALAEONTOLOGY, VOLUME 8

A comparison may be made with the fauna associated with gulfweed in the Sargasso
Sea. Four species of epizoa which would be expected to occur fossil are found in every
sample of this weed, viz. Membranipora sp., Spirorbis sp ., Lepas pectinata and Lepas sp.
(Timmermann 1932, p. 306), while others, such as crabs and gastropods, are found
sporadically. It cannot be excluded that Jurassic gulfweed harboured only one common
species, any others being exceedingly rare, but this would be surprising for there were
many attached forms of life in the Jurassic. A hypothesis that does not demand abundant
flotsam is therefore more acceptable, on facies grounds, than one that does.

Both at Elstow and St. Barthelemy the shells of B. buchi sometimes occur in clusters.
These superficially resemble those described in Posidonia becheri by Paul (1939), who
rightly used them as evidence of a partly pseudoplanktonic mode of life in that species.
In B. buchi, however, some of the clusters (e.g. PI. 19, fig. 4) consist almost entirely of
single-valved specimens, as can be proved by dissection. Such clusters must have arisen
by posthumous drifting together, and clusters containing mainly double valves could
easily have arisen in the same way, since slight currents would separate double- from
single-valved specimens. Zangerl and Richardson (1963, caption to pi. 24a, p. 131)
have noted a similar case involving Pteria and Dimbareila in a Pennsylvanian black shale.

On facies grounds, therefore, B. buchi was more probably planktonic than pseudo-
planktonic.

Functional Morphology

Swimming in relatives o/Bositra and Posidonia. No living bivalves are nektoplanktonic
but benthonic swimmers occur in two related superfamilies, the Pectinacea and Limacea.

The mode of swimming is very similar in
both groups. The inner lobe of the mantle
forms a pallial curtain or velum round the
edge of each valve. In swimming the free
edges of the vela are turned in and touch
each other (text-fig. 2), except just anterior
and posterior to the hinge, where the vela
are exceptionally flabby.

Before swimming begins the valves are
opened as wide as possible to increase the
volume of water between them. When the
adductor muscle contracts and the valves
come together the vela form a ‘valve’ (in
the engineer’s sense) so that water can only
escape at the flabby places near the hinge,
which function like the nozzle of a bellows.
Water easily enters between the vela when
the adductor relaxes and the valves open.
By opening and shutting its valves the animal
thus pumps out water near the hinge and
moves through the water with the ventral margin foremost.

Though the mode of swimming is basically the same in both Limidae and Pectinidae
there are differences in detail between the two groups. Thus the Limidae usually swim

Side View

Top View

— > Water jet
WLm

text-fig. 2. Swimming position in a limid and a
pectinid (diagrammatic).

JEFFERIES AND MINTON: MODE OF LIFE OF JURASSIC ‘ POSIDONIA' 163

with the commissure vertical except when tired (Studnitz 1931, p. 305) and swimming is
assisted by the rowing action of the tentacles (Gilmour 1963, p. 85) round the edge of the
mantle. Pectinidae, on the other hand, do not row with the tentacles and usually swim
with the commissure roughly horizontal but raised at the ventral end; in addition to
normal swimming, Pectinids can also propel themselves hinge first, in an ‘escape
reaction’, away from a noxious stimulus. It is said (Haas 1941, p. 506) that equivalve
Pectinidae are more likely to swim with the commissure vertical than inequivalve species.

The fact that Recent Limacea and Pentinacea swim in basically the same way suggests
that they may have inherited swimming ability from a common ancestor. The earliest
known Limacea are Carboniferous in age (Cox 1943, p. 153), while the earliest Pectinacea
are probably Devonian. Their common ancestor would therefore probably be Devonian
or Silurian in age and would be a pterinopectinid (in Newell’s sense (1937)) and therefore
close to the ancestor of B. buchi and ‘ Posidonia ’ radiata. These two forms could easily
have inherited swimming ability from the same source.

Several additional facts are relevant here. Firstly, some Lower Carboniferous Pecti-
nacea (e.g. Pernopecten sowerbvi (McCoy)) have smooth thin shells with an accurate
plane of symmetry perpendicular to the commissure. Such forms could probably swim
since they much resemble both the Recent Cyclopecten groenlandicus, which is a good
swimmer (S. Jensen in A. S. Jensen 1905, p. 332), and Amusium, which Yonge (1938,
p. 81) thought to be probably the best swimmer among bivalves as judged by the sym-
metry, thinness, and smoothness of its shell. Secondly, the musculature of Pennsyl-
vanian Pectinacea, as judged by the muscle scars (Newell 1937, p. 23), was basically like
that of Recent Pectinidae. Thirdly, Newell (1937, p. 20) thought that early Pectinacea
were very active because they are seldom encrusted with epizoa. Fourthly, the Recent
Lima excavata can swim (T. H. J. Gilmour, personal communication) and is identical
in shell characters to P/agiostoma , which first occurs in the Trias.

It is therefore very probable that the common ancestor of the Limacea and Pectinacea
in general, including that of Bositra , Posidonia, Steinmannia, Dunbarella, Halobia,
Daonel/a, and Monotis, was able to swim.

Shell structure and thickness. The thickness of the shell of B. buchi has been determined
in the following instances:

(a) Bathonian; Fuller’s Earth Clay. Cliff End, near West Bay, Dorset. LL17415; prismatic layer 47 /x ;
laminar layer 10 /x; total 57 yu. ; length 1 cm. LL17416; prismatic layer 48 g,; laminar layer 16 /x; total
64 n; length 1 cm. Associated ammonites and nuculids were perfectly preserved, indicating that
laminar aragonite like that of the inner layer of B. buchi was not dissolved. Associated heterodonts,
consisting of crossed-lamellar aragonite, had partly lost the shell.

(b) Lower Bajocian. Blue clay, Kidugallo borehole, Tanganyika. L88714; thickness of shell 73 jtx at
length 1 cm., 32 yu. at length 1-2 mm. Laminar and prismatic layers not distinguishable.

( c ) Bajocian-Bathonian; Sargelu Formation. Sirwan Gorge, Kurdistan, Iraq. LL17402. Thickness of
shell in several individuals 16-32 /x but difficult to measure because of recrystallization. Laminar layer
brownish, about as thick as prismatic layer.

There are some relevant data in the literature. The sections of ‘ Posidonia alpina ’ in
Peyre (1959, pi. 1, fig. 4) show both layers of the shell and are about 50 n thick. Roth-
pletz (1892, p. 93) described the shell of Triassic specimens of Halobia from Indonesia
as consisting of laminar and prismatic layers and varying from 6 to 60 /x in thickness,
and Krumbeck (1924, p. 126) quoted the shell thickness of Halobia as 15-25 /x.

164

PALAEONTOLOGY, VOLUME 8

The thinness of the shells of B. buehi has led many authors to identify them as fila-
mentous algae, as discussed by Peyre (op. cit.). The ‘prodissoconchs of lamellibranchs’
to which Peyre ascribed some of the sections he studied are probably juvenile stages of
B. buehi ; one of them (op. cit., pi. 1, fig. 2) is in the valves-open position (see below,

p. 166).

The thickness of the shell of B. buehi therefore seems to have been about 60-70 p. at
a length of about 1 cm. The thickness is very uniform all over the valve, since the con-
centric folds are corrugations affecting the whole shell and do not represent increased
thickness. Only the cardinal area is appreciably thicker than the shell in general. This
would strengthen the shell where specially needed without much increase of weight. The
style of ornament closely resembles that of the Recent thecosomatous pteropod
Balantium.

An interesting feature of the valves of B. buehi is that they are often brownish round
the edges. This is seen for example at la Voulte-s. -Rhone (PI. 19, fig. 7) on the top surface
of Bed 9e and adjacent bedding planes (Hess 1960), where the margins are often the
colour of haematite. Hess has shown that in ophiuroids from this locality pyritization
is connected with original high organic content (op. cit., p. 343). The brown shell
margins of B. buehi are probably due to oxidation of pyrite, and therefore also indicate
a high organic content. They were noticed in valves of length (in mm.) 0-7, 1-0, 1-7, 2-8,
4-2, 4-3, 5-0, 7-8, 8-5, and 12-5. They are probably due, therefore, to new shell material
not being completely calcified as soon as it was formed.

Shells as thin as that of B. buehi occur in both benthonic and planktonic living
molluscs. This thinness, therefore, does not prove that B. buehi was planktonic — a con-
clusion which is based on facies. It is more relevant to whether B. buehi was planktonic
or pseudoplanktonic, since a thin shell is a swimming adaptation in the Pectinacea
(Yonge 1938, p. 81) and since the analogy of the pteropod Cavolinia (shell 40 p thick at
length 1 cm.) shows that a pelagic animal with a shell so thin would not require attach-
ment. The analogy of Cyclopecten groen/andieus (right valve 80 p thick, left valve 50 p
thick at length 2-3 cm.) is particularly suggestive, because this species, like B. buehi, is
a Pectinacean and, though benthonic, is a considerable swimmer.

C. groen/andieus is also like B. buehi in having a high organic content round the edges
of the valves. Indeed the whole shell is springy to the touch as if somewhat horny, and
the shell margins sometimes flex outwards when the animal shuts its valves at death
(Collin in A. S. Jensen 1905, p. 332). The shell of B. buehi probably had similar proper-
ties, for otherwise it would have been extremely brittle.

The presence of gapes. Examination of apparently uncrushed specimens of single valves
of B. buehi from various localities suggested that the commissure was not plane. It is
difficult to establish this feature, however, when the fossil is buried in the rock, and the
shell is too thin to be excavated. Plaster casts were therefore made, using thick rubber
moulds to avoid distortion, and trimmed away along a growth line. The model thus
produced was mounted on a cover slip with the hinge line touching the cover slip
throughout its length. Under these conditions, in the six cases tried, the only other places
where the model touched the slip was a point on the ventral margin (PI. 19, fig. 2). For
growth stages much less than 1 cm. long the model was trimmed back to the relevant
growth line and a plane cut parallel to the hinge and a point tangent to the ventral margin.

JEFFERIES AND MINTON: MODE OF LIFE OF JURASSIC ‘ POSIDONIA' 165

This produced a model of the growth stage on top of a short pillar of plaster-of-Paris,
which was mounted on a cover slip. By this means it was shown that the commissure
was not plane at a length of 3-3-5 mm. (PI. 19, fig. 3).

The analogy of Cyclopecten groenlandicus suggests that the valves of B. buchi would
probably have distorted very easily in life. Nevertheless, the curvature of the commissure

Anterior

Wzntral

Posterior

Observed

Reconstructed

'■ S3 Gapes

text-fig. 3. Gapes in Bositra buchi. Based on plaster models
treated as explained in text, a, LL 17402a (For data see
PI. 19, fig. 2); b, L84735 (Data as PI. 19, fig. 3); c, L84734c
(Data as b).

is so characteristic in shape, and occurs so often when the matrix is hard enough to
prevent crushing, that it must correspond to the usual shape of the original shell. This
indicates, since B. buchi was equivalve, that big gapes would normally be present at
front and back as shown in text-fig. 3.

In Recent Pectinidae and Limidae anterior and posterior gapes serve to release the
swimming jets (Jackson 1890, p. 339; Verrill 1897, p. 44) and their size is related to
swimming ability (text-fig. 4). Thus Chlamys opercularis is a good swimmer and seldom
byssally attached (Rees 1957, p. 27) and has large anterior and posterior gapes, whereas
C. varius, which lives byssally attached to rocks (Dalmon 1935, p. 273) has very small
gapes. Lima hians has large gapes and in speed and endurance is as good a swimmer as

166

PALAEONTOLOGY, VOLUME 8

C. opercularis (T. H. J. Gilmour, personal communication) although Studnitz (1931),
perhaps through working with tired individuals, reported otherwise. Large gapes also
occur in Amusium, e.g. A. pleuronectes, and Cyclopecten groenlandicus.

By analogy with living relatives, therefore, the gapes of B. buchi suggest swimming
ability. They are not to be confused with the gapes of burrowers such as Mya through
which the foot protrudes anteriorly and the siphons posteriorly. Such burrowers are not
related to B. buchi , for which a benthonic mode of life is most unlikely on facies grounds,
and siphons could hardly have been present in B. buchi, since they were absent in

text-fig. 4. Gapes in living relatives of B. buchi. a, Amusium pleuronectes,
posterior, b, Chlamys opercularis, posterior, c, Chlamys varius, posterior, cl, Lima
hians, posterior, e, the same, anterior. The anterior gapes of a-c are like the

posterior gapes.

Palaeozoic Pectinacea (cf. Newell 1937) and are lacking in all living representatives of
the order Pteroconchida (Cox 1960) in which the Pectinacea and Limacea are included.

If the valves were elastic, B. buchi may have been able to eliminate the gapes by pulling
the valves tightly shut, just as Cyclopecten groenlandicus can. It is the normal presence of
gapes which suggests swimming ability, however, as the analogy of C. groenlandicus
shows. The disposition of the gapes on either side of the hinge in B. buchi suggests that
water was forced out at these points and that swimming resembled that of living Limacea
and Pectinacea. This supports the suggestion that the common ancestor of these super-
families could probably swim.

The valves-open position. When Bositra buchi is found with both valves together they are
nearly always open and spread out on the bedding plane. This is well shown, for instance,
in Plate 19, figs. 1 and 7, and was emphasized by Guillaume (1928, p. 225) and Boehm
(1912, p. 131). Wanner (1907, p. 204) described the same feature in Daonella indica and
examination of the literature shows that it is usual in Triassic and Jurassic examples of
Halobia, Daonella, ‘ Posidonia ’, Bositra, and Steinmannia.

In B. buchi the valves-open position seems to be more often concave-up than concave-
down. Thus in a specimen from the Oxford Clay at Elstow twelve specimens were
concave-up and one concave-down. In specimens from the Sargelu Formation of

JEFFERIES AND MINTON: MODE OF LIFE OF JURASSIC ‘ POSIDONIA' 167

Northern Iraq in which the weathered side was assumed uppermost, the corresponding
numbers were 21:6, 30:2, 2:10, 5:3, 8:2, 0:4, and 11:2, or 77:29 in total. This is
probably because the concave-down position is unstable in bivalves with rounded ventral
margins that are wafted about by slight currents after death with the ligament intact
(Schafer 1962, p. 187). The valves-open position is not fortuitous in B. buchi ; it is very
characteristic of this species and other Triassic and Jurassic bivalves whose facies suggests
they were pelagic, and is not present as a regular feature in other Pectinacea or Limacea.

Wanner (loc. cit.) held that the regular occurrence of the valves-open position indi-
cated swimming ability. This proposition is probably true for a pectinacean and can be
re-argued as follows: for the valves regularly to have fallen open, whether concave-up or
concave-down, suggests that before burial they opened to a very wide angle. The angle
in question would be determined by the ligament after the soft parts had rotted. In
Pectinacea (though not in many other bivalves) this angle is equal to the maximum
angle of opening during life, which is that used in swimming. A wide angle of opening
is advantageous to a swimmer and, therefore, the regular occurrence of the valves-open
position in a pectinacean suggests swimming ability.

The stages in this argument can be elaborated. Firstly, experiments using an alumi-
nium model of B. buchi with valves of the right convexity and a ligament of glue, showed
that when the ligament rotted in water with the valves concave-up on hard bottom, the
valves fell open when the initial angle of opening exceeded about 60°. This would prob-
ably be roughly correct for the sort of bottom over which B. buchi could be wafted by
currents. The conclusion that the angle of opening was very wide, however, is based
primarily on the rarity of the valves-open position in normal Pectinacea and Limacea.
Secondly, in Pectinacea the maximum angle of opening in life is very precise and is
determined by the ligament alone. It therefore equals the angle of opening when the
soft parts are removed. In two newly dead individuals of Chlamys opercularis and Pecten
maximus the means of this angle were respectively 31° and 34°; this is about equal to the
maximum angle of opening in life observed during swimming. Many bivalves differ from
Pectinacea in this respect, including some that often occur in the valves-open position
on beaches. In these the angle of opening when the soft parts are removed is rather im-
precise and greatly exceeds the maximum angle in life, which is determined by ventral
muscular connexions between the valves. The tellinaceans Arcopagia crassa and Donax
vittatus, and Cardium edide, are examples of this condition. B. buchi was a pectinacean
and therefore almost certainly lacked ventral muscular connexions, for these are absent
in all modern related forms included in the Pectinacea, Anomiacea, Limacea, Pteriacea,
and Ostreacea. Further, as already mentioned. Upper Palaeozoic Pectinacea had muscu-
lature much like that of living ones. In addition, the ligament of B. buchi resembled that
of modern Limidae and Pectinidae. It therefore probably controlled the maximum angle
of opening in life, which may have been about 60°, as compared with 30° for modern
Pectinidae. Lastly, a wide angle of opening is advantageous to a swimmer because it
increases the volume of water expelled at each contraction of the adductor.

In summary, therefore, Tellinacea and other bivalves are often found in the valves-
open position on beaches, but this has no biological significance. The regular occurrence
of the valves-open position in B. buchi, on the other hand, suggests swimming ability.

The vcdves-closed position. Specimens of B. buchi with the valves closed are uncommon.

168

PALAEONTOLOGY, VOLUME 8

Apart from examples where the valves were not properly articulated, two cases were
encountered in the present work. The first was in a hard limestone from the Sargelu
Formation (LL17402). In this, the microcoprolite Aggregatella pseudohieroglyphicus
Elliott (1962, p. 40) is abundant in the matrix but absent within the closed valves. The
specimens had evidently closed their valves before they died and did not open again
before burial. Their gapes must have been small enough to prevent the coprolites from
entering, perhaps because the valves were slightly elastic, and pulled hard together by
the adductor; four single valves from the same block showed the characteristic curvature
of the commissure (e.g. PI. 19, fig. 2). The second instance was described by Hess (1960)
from the upper surface of Bed 9c in the Callovian of la Youlte-s. -Rhone, where speci-
mens of B. buchi up to 1-2 mm. long are sometimes preserved with the valves closed and

text-fig. 5. Three living bivalves with the byssus (black) protruding from a concavity in the

lateral outline (diagrammatic).

filled with pyrite (PI. 19, fig. 8). The commissures are roughly perpendicular to the
bedding, with the anterior or posterior end (one cannot say which), or the umbones,
pointing stratigraphically downwards (my observation, cf. Hess, p. 377). These speci-
mens probably fell into the superficial, almost liquid layer of bottom mud, closed their
valves and died.

B. buchi could, therefore, close its valves in life. This fact is important in estimating
its hydrostatic properties (see p. 174).

Adult outline and convexity. The adult outline of Bositra buchi is at every point convex
outwards; there is no byssal notch, no ears, and the shell is exactly equivalve.

If B. buchi lived attached to a floating support it could either have rested on one valve,
or held the commissure perpendicular to the substratum. In the first case, which is typical
in Pectinacea, one would expect the valves to be unequal in convexity, probably with
the right less convex than the left as in young individuals of Anomia (Odhner 1914) or
young Posidonia becheri, and with a byssal notch. It is most unlikely, therefore, that B.
buchi rested on one valve.

If the commissure were perpendicular to the substratum the antero-ventral margin
would probably be flat or concave where it touched the substratum and the byssus
emerged, as in Mytilus edulis, Lima excavata and the independently evolved Dreyssena
polymorpha (text-fig. 5). This gives a firmer grip on the substratum. It is true that some
byssally attached shells are almost circular in outline, e.g. Kellia suborbicularis. This

JEFFERIES AND MINTON: MODE OF LIFE OF JURASSIC ‘ POS1DONIA ' 169

species, however, like other Erycinacea, lives in very sheltered positions under shells or
holdfasts (Step 1951, p. 101), which is not comparable to a life fixed to floating seaweed.

The adult outline and convexity of B. buchi , therefore, suggests that the adult was
almost certainly not attached, so that it lay habitually on one valve, with the commissure
perpendicular to the substratum. The adult outline and convexity, in fact, provide no
evidence whatever of attachment.

text-fig. 6. Ontogeny of the outline of B. buclii, LL 17402a. For data
see PI. 19, fig. 2.

Ontogeny. The ontogeny of the outline of Bositra buchi has been worked out by drawing
with a camera lucida the troughs between successive concentric ribs (text-figs. 6, 7).
Only uncrushed specimens with perfect umbones can be used for this work. The onto-
geny of the outline is remarkably simple. A form with roughly median umbones at a
length of 0-5 mm. changes gradually into a form with somewhat anterior umbones at a
length of about 1 cm. The outline may become somewhat more elongate.

Double-valved specimens never have the valves unequal in outline or convexity at
any stage of growth. This is shown by specimens from la Youlte with the valves closed
up to a height of IT mm. (PI. 19, fig. 8, estimated length 1-2 mm.) and by specimens
with the valves open, which are somewhat more difficult to interpret in this respect, from
la Voulte (Hess Collection B. 52, length T2 mm.; B. 56, length 2-5 mm.), St. Barthelemy

170 PALAEONTOLOGY, VOLUME 8

(LL17406, length 2-6 mm. and others of similar size) and the Kidugallo borehole
(L88713, length 1-6 mm). Double-valved specimens of greater size are also exactly
equivalve (e.g. PI. 19, figs. 1, 7). The valves-open position occurs at a length of 0-6 mm.
(at la Voulte) and at all larger sizes.

Plots of length against frequency for single bedding planes (text-fig. 8) very often
show a peak at a nominal length of 1 mm. This peak is not due to some form of post-
humous drifting, since it occurs in nine out of twelve randomly selected bedding planes

o

o

O

Figsl-4 , lmm

Figs 5-6 . Imm .

text-fig. 7. Ontogeny of the outline of B. buchi , L84734c.

For data see PI. 19, fig. 3.

from France, Tanganyika, and Iraq. It must represent high natural mortality (i.e. that
not due to predation) at this length.

The thickness of the shell, on the basis of material from Kidugallo, increases from
about 30 /x at a length of T2 mm., to about 70 /j. at 1 cm.

The ontogeny of B. buchi is much simpler than that of Posidonia becheri as described
by Weigelt (1922, cf. text-fig. 9). At a length of 0-5 mm. this species is equilateral and
equivalve ; at a slightly greater size the umbones become somewhat anterior, the right
valve becomes flatter than the left, and a byssal sulcus develops in its antero-ventral
margin; later the shell becomes still more pteriiform, remaining so to a length of about
1 cm., when the adult form begins to be attained.

Weigelt (1922, p. 74; 1927) thought that the earliest stage was the prodissoconch of
the veliger larva, that subsequent changes were due to settlement (right valve downwards
as usual in Pectinacea) on flotsam, and that the change in outline at a length of about
1 cm. was due to attachment on the sea floor with the commissure vertical. The changes

JEFFERIES AND MINTON: MODE OF LIFE OF JURASSIC ‘ POSIDONIA ’ 171

at a length of about 0-5 mm. must indeed be due to settlement, since the sequence is almost
exactly as in Pecten irradians (Jackson 1890, p. 342) and Anomia spp. (Odhner 1914;
Miyazaki 1935) except that settlement took place in P. becheri at a larger size. It is also

text-fig. 8. Plot of number of valves against length of B. buchi
for various bedding planes. 1-4, Bathonian or Callovian of St.
Barthelemy, respectively LL17405, 4, 6, 3. 5-6, Bajocian, Sargelu
Formation; Ruwanduz, Northern Iraq (LL17407). 7, Bathonian,
Sargelu Formation; Ser Amadia, Northern Iraq (LL 17409). 8,
Bathonian, Sargelu Formation; Barsarin. Ruwanduz, Northern
Iraq, LL17408. 9-11, ? Bajocian, Kidugallo borehole, Tanganyika,
LL88715. 12, ditto, L.L88713.

probable that settlement usually took place on flotsam, since Hind (1893, p. 541)
recorded small specimens of Posidonia attached to Catamites fragments and Paul (1939)
recorded medium-sized and large examples of P. becheri attached to wood and to each
other. Weigelt (1927, p. 75) suggested that the forms attached to wood figured by Hind
(1896-1901, p. 94, pi. 6, fig. 24) as Posidoniella laevis were intermediate stages of P.
becheri.

172

PALAEONTOLOGY, VOLUME 8

Weigelt’s suggestion that at a length of about 1 cm. P. becheri settled on the sea floor
is unlikely, however, in view of the facies in which the species usually occurs. Schmidt
(1935, pp. 138, 142) thought that the adults were attached to flotsam like the inter-
mediate stages, and Paul’s evidence shows that the adult could attach itself. Neverthe-
less, some change in habit between the two stages seems necessary to explain the change
in shape, and it may be that the adult was a habitual swimmer that attached itself
occasionally.

In any case, the absence in B. buchi of any abrupt change in shape at a length of 0-5
to 1-5 mm., and the fact that the valves are never unequal or a byssal notch present,
together with the gradualness with which the adult form is acquired with only a small
increase in shell thickness and little change in outline, suggests that this species (unlike
its relatives Posidonia becheri, Pecten irradians, and Anomia) was never attached, and
indeed never settled on a substratum, and that the adult, in accordance with facies con-
siderations, was pelagic like the larva.

This implies that swimming by clapping the valves developed in ontogeny as soon as
swimming by cilia was abandoned. This is not improbable since gapes already existed
at a length of 3 mm. The 1 mm. mortality peak may indicate the change from ciliary
to muscular swimming. It need not be ascribed to unsuccessful settlement, of which
there is no other sign in the ontogeny. Thorson (1946, p. 467) noted that benthonic
opisthobranch gastropods sometimes metamorphose in mid-water, without touching a
substratum. In B. buchi, as in thecosomatous pteropods which are descended from
benthonic opisthobranchs, this had probably become a regular part of the life cycle.
The ontogeny of B. buchi is therefore analogous to that of the pteropod Spiratella
(Lemche 1948, p. 24), in which the shell of the pelagic larva enlarges gradually into that
of the pelagic adult.

EXPLANATION OF PLATE 19

Figs. 1-4, 6-9, Bositrci buchi (Romer). Fig. 5, 1 Posidonia ’ radiata Goldfuss.

Fig. 1. Rock specimen (bituminous limestone) of the typical facies. Note the valves-open position of
many individuals. Bathonian, Sargelu Formation; Scr Amadia, Northern Iraq. LL17409, x0-66.

Fig. 2. Plaster cast of LL1 7402a. Bathonian, Sargelu Formation; Sulemanya, Sirwan Gorge, Northern
Iraq. Right valve. Model trimmed along growth line and mounted on a flat surface to show gapes.
a. Anterior oblique; b , posterior oblique; c, ventral oblique; d , lateral aspects. x3-5.

Fig. 3. Plaster cast of L84734c. Left valve. Bathonian, Lower Fuller’s Earth Clay; Silverslake Farm,
near Sherborne, Dorset. Model trimmed to length of 3-5 mm. a, lateral; b, ventral aspect. Note
curvature of commissure in ventral aspect. X 10.

Fig. 4. Posthumous cluster of single valves. LL1741 1. Bathonian or Callovian; St. Barthelemy-de-Vif,
Isere. x 2-6.

Fig. 5. L93741. Upper Lias, base of acutum Zone; Bracebridge Brick Pit, 2 miles south of Lincoln.
X 0-66.

Fig. 6. Aluminium model with tentacles used in feasibility experiments. x0-66.

Fig. 7. Two specimens showing brown shell margins. Fless Collection, Basel University, B.50. Bed 9c,
Callovian; la Voulte-s.-Rhone, Ardeche. X 3-3.

Fig. 8. Individual IT mm. in height with valves closed and embedded with hinge line perpendicular
to bedding. Hess Collection, Basel University, B.55. Top surface of Bed 9c. Callovian, la Voulte-s.-
Rhone, Ardeche. X 10.

Fig. 9. Specimen showing internal ligament tissue, outlined in black. LL17412. Kosmoceras jasoni
Zone, Callovian, base of Oxford Clay, 20 in. above oyster bed at base of quarry; Elstow, Bedfordshire
(Nat. Grid Ref. TL043456). x33.

Palaeontology, Vol. 8

PLATE 19

JEFFERIES and MINTON, Posidonia’

JEFFERIES AND MINTON: MODE OF LIFE OF JURASSIC 'POSIDONIA' 173

It is true that Lima, which is perhaps as closely related to B. buchi as the Pecten
irradians studied by Jackson, has no inequivalve stage in development, though it settles
in the usual manner and the adult is benthonic (Odhner 1914; Lebour 1937). This is
because on settlement the post-larva, like an adult Lima, crawls with the commissure
vertical instead of lying on one valve. As already mentioned, however, if this was the

5

R

2

3

4

Figs 1-4 , Q51™).

Figs 5-8 . Icfn ■

text-fig. 9. Ontogeny of Posidonia becheri Bronn.

mode of life of B. buchi one would expect to see a straight anterior margin, and this does
not exist.

Conclusions. Assuming, on facies evidence, that B. buchi was not benthonic, it was either
attached to flotsam or a planktonic swimmer. The pseudoplanktonic hypothesis meets
with three main difficulties. Firstly, where B. buchi is abundant other possible pseudo-
plankton is usually very rare. Secondly, there is no evidence of attachment in the shell
outline. Thirdly, by contrast with living and fossil relatives, there is no sign of an attached
stage in the life history. Arguments in favour of the pseudoplanktonic hypothesis are
weak; thus the occurrence of clusters of B. buchi is probably due to posthumous drifting
and does not indicate attachment; and a mortality peak at a length of 1 mm. can be
explained by metamorphosis rather than unsuccessful attachment.

174

PALAEONTOLOGY, VOLUME 8

Directly in favour of the nektoplanktonic hypothesis are the presence of gapes, the
thinness of the shell, and the regular occurrence of the valves-open position. All these
indicate swimming ability by analogy with living relatives. A nektoplanktonic mode of
life is therefore much more likely than a pseudoplanktonic one.

The feasibility of a planktonic mode of life in Bositra buchi

So far it has been shown that a planktonic mode of life for B. buchi is most likely.
Experimental evidence that the animal could reasonably have stayed up in the sea by
using structures present in Recent Limidae and Pectinidae, i.e. by swimming, without
the aid of a pneumotophore, is now examined.

Swimming plankton may remain at a roughly constant level by moving horizontally,
like the pteropod Cavolinia (Schiemenz 1905, p. 6) or by swimming up and sinking down
like the pteropods Spiratella (= Limacina ) (Morton 1954, p. 298) or Creseis (Kornicker
1959), or even by swimming up and swimming down as the copepod Calanus finmar-
chicus sometimes does (Hardy and Bainbridge 1954, p. 425). Only passive sinking can be
investigated in the present instance so we have assumed that B. buchi remained at one
level by swimming up and sinking down. This would probably be less efficient than
horizontal swimming, which would be aided by hydrodynamic lift, so that if ‘hopping
and sinking" is feasible, horizontal swimming is also feasible.

We have assumed that B. buchi sank umbones downwards in the observed position of
stability for models when tentacles are present or when the valves are open to more than
45°. This is reasonable since in ‘hopping and sinking’ the animal sinks in the position
of passive stability (e.g. Spiratella , Creseis and Calcmus). It is consistent with the equi-
valve character of Bositra , since bilaterally symmetrical swimmers like Lima, Chlamys,
Pecten, plaice, herring (and submarines) swim with the median plane of symmetry
vertical.

We have also considered the effect of a fringe of tentacles from the median lobe of the
mantle margin, which exists in many Recent relatives of B. buchi, including Spondylus
gaderopus, Anomia epluppium, Ostrea edulis, and Pteria hirundo, and is very well de-
veloped in the Pectinidae (e.g. Chlamys opercularis and Pecten maximus) and Limidae
(e.g. Lima hians). L. hians rows with the tentacles when swimming (Gilmour 1963, p. 85),
which shows that they are stiff enough to be hydrodynamically effective.

The object of the feasibility studies, therefore, was to determine settling velocity for
B. buchi of different lengths, corresponding to different phases of the life history,
different specific gravities of protoplasm, and with and without tentacles.

The hydrostatic properties of B. buchi of different lengths were estimated by com-
parison with a model 2 cm. long, having a shell of aluminium of specific gravity 2-7,
shell thickness 49-4 p and weight 96 mg. in air, and with an internal volume when closed
of 1-4 cm.3 The internal volume at different lengths was calculated assuming geometrical
similarity, and the weight of the shell by assuming geometric similarity for the surface
area, a specific gravity of 2-7 (calcite), and a thickness deduced from text-fig. 10. The
volume of protoplasm was assumed to equal the internal volume of the shell, which
might be true when all the soft parts were retracted inside, as seems to have been possible
(see p. 168). This gives a maximum value for the buoyancy for a given specific gravity
of protoplasm.

It was also assumed that the specific gravity of the soft parts was not less than 0-9,

JEFFERIES AND MINTON: MODE OF LIFE OF JURASSIC ‘ POS1DONIA" 175

when the whole internal volume would be filled with fat. The possibility of a gas-filled
float was neglected, as already explained. If the soft parts were extended by being
inflated with sea water, their buoyancy would be unaffected. The tentacle array (PI. 19,
fig. 5) was assumed to be roughly like that of Lima hians. The valves were assumed to be
open at 45°.

The ambient sea-water was assumed to be at 20° C. and 35%0 salinity with a specific
gravity of 1-025 (Harvey 1955, p. 128) and viscosity 1-07 centipoises (Miyake and Koi-
zumi 1948, p. 65).

At this point it is necessary to consider the hydrodynamics of a passively sinking

Length=cm.

text-fig. 10. Plot of shell thickness against length in B. buchi as
used in feasibility study. Circles indicate values obtained from
thin section of LL88714, ? Bajocian, borehole 5 miles N. of
Central Railway, Kidugallo, Tanganyika.

B. buchi. The slow settling of bodies through a stationary fluid has been the subject of
many investigations, from the classical work of Stokes to, for instance, White (1946).
When a body settles through a uniform fluid at a steady velocity, i.e. with no acceleration,
and therefore with no resultant vertical force acting upon it, the weight acting downwards
equals the sum of the buoyancy force and drag force acting upwards.

The weight (W) of B. buchi can be estimated from the information given above and
depends on the specific gravity of the protoplasm Sp and the weight of the shell. The
buoyancy force ( B ), due to the weight of liquid displaced, depends on the volume (V)
and the density (pw) of the surrounding water, while the drag ( D ) depends, among other
variables, on the velocity of fall.

Now, as already stated, W = B+D (1)

or VsPsg+VpPpg = (^sJr^/p)pwSJrL),

where Vs = volume of shell, ps = density of the shell, Vp = volume of the protoplasm,
Pp = density of the protoplasm, and g = acceleration due to gravity.

If Ws = immersed weight of shell

= Vs Ps g—Vs Pw g

176

PALAEONTOLOGY, VOLUME 8
p = density of fresh water, then

Ws = gVp(pw— pp)-\-D

and

= PgVp(b025-Sp)+D

assuming ambient sea water of specific gravity 1-025.

The drag ( D ) may be estimated from experiments with models. The drag of B. buchi

where p = viscosity of fluid, v the settling velocity, / the length and a a dimensionless
coefficient of resistance. The value of a depends on the shape of the falling body and the
pattern of flow round it. Geometrically similar bodies have geometrically similar flow

are the same. Consequently a can be determined experimentally with models of B. buchi
of any length (/), by varying viscosity (p) to give the required Reynolds’ numbers. In
studying the effect of change in size it is more convenient to use one model in liquids
of varying viscosity, than models of different sizes, sometimes microscopic, in a liquid
of constant viscosity.

The experiments were carried out in a cylindrical glass jar 19 cm. in internal diameter
and 36 cm. deep, at about 20° C. Four bands of paper about 8 cm. apart were fixed round
the jar. Preliminary experiments with tap water, in which models were timed with a stop
clock, falling from first to third and from second to fourth bands, gave very similar
velocities for both distances. The time taken to sink from second to fourth bands,
16-3 cm. apart, was therefore assumed to give terminal velocity. These experiments also
showed that the presence or absence of concentric ribs on models carefully shaped to
the convexity of B. buchi did not affect settling velocity, at least at length 1-2 cm. Models
open at 60° and without tentacles fell at roughly 75 per cent, of the velocity of models
open at 45°. Models open to 45° or more always fell umbones downwards. At a smaller
angle they tended to fall one valve underneath. With tentacles of specific gravity equal
to water the models always fell umbones downwards.

The final experiments were carried out in a mixture of glycerine and water, the mix-
ture being varied to give viscosities from about 8 poises to 1 centipoise. Viscosities were
measured with a series of U-tube viscometers. Temperature was constant at 20°=b0-5° C.
and the specific gravity, measured with a hydrometer, varied from 1 -257 to 1-000.

Two models of length 2 cm. were constructed of aluminium foil of thickness 100 /x
shaped to resemble Bositra buchi in outline and convexity, with the valves open to 45°.
One of the models was provided with a fringe of flat tentacles of aluminium foil (PI. 19,
fig. 5) and weighed 950 mg., while the other was without tentacles and weighed 190 mg.

From the experimental results the drag (D) of the models was tabulated for various
settling velocities (v) in solutions of viscosity (p) and density (p).

where a R is a function only of /?, the Reynolds’ number. The experimental results were
reduced to this form and plotted logarithmically (text-fig. 1 1) over a range of Reynolds’

can be expressed as

D = OLpvl ,

(2)

JEFFERIES AND MINTON: MODE OF LIFE OF JURASSIC ‘ POSIDONIA' 177

numbers from 10“ 2 to 103 (Note: B. buchi of / = 1 mm. and settling velocity 1 mm. /sec
has a value of R = 1). Values of <xR for spheres from Goldstein (1938, p. 16) were plotted
for comparison on the same axes and produce a very similar curve.

To present these results in a biologically useful form it was decided to compute a graph

text-fig. 11. Coefficient of resistance (a) X Reynolds’ number against
Reynolds’ number for B. buchi with and without tentacles, and for spheres.

of settling velocity against length (text-fig. 12). Values of specific gravity of protoplasm
(Sp) were chosen, the buoyancy ( B ) and weight (W) calculated for various lengths (/)
and the drag (D) found from equation (1). R was calculated from equation (2) assuming
sea water of the characteristics already mentioned. Reference to the appropriate line on
text-fig. 12 led to the Reynolds' number from which the settling velocity was calculated.

In considering the biological significance of these results an analogy can be drawn
between B. buchi, living as here postulated, and the Recent thecosomatous pteropods
such as Spiratella and Creseis which likewise are shelled, microphagous planktonic

B 0612

N

178

PALAEONTOLOGY, VOLUME 8

mollusca that stay up by ‘hop and sink’, are roughly the same size as B. buchi, and are
able to withdraw into the shell. There is little information concerning passive settling
velocities in this group, but in SpirateUa trochiformis the mean of three descents was
1 -3 cm. /sec. varying from 1 -2 to 1-5 cm. /sec. (J. A. McGowan, personal communication).

text-fig. 12. Calculated passive settling velocities of B. buchi
with and without tentacles, for various sizes and specific
gravities of tissue.

and in Creseis acicula the mean of 24 descents was IT 5 cm. /sec. ranging from 0-72 to
T80 cm. /sec. (Kornicker 1959, p. 334).

The graph suggests that B. buchi provided with tentacles could have sunk passively
at a comparable velocity of less than 2 cm. /sec., if the specific gravity was about equal
to sea water or less. Without tentacles even extremely buoyant tissues (specific gravity
= 0-9) would not have kept the settling velocity below 3 cm./sec. at one stage in the life
history. It therefore seems that a planktonic mode of life for B. buchi is feasible if it was
provided with a long fringe of tentacles. It should be noted that cylindrical tentacles
would probably be somewhat, but not much, less effective than the flat ones in the model.

During the preliminary experiments it was noted that an empty shell of the Recent

JEFFERIES AND MINTON: MODE OF LIFE OF JURASSIC ‘ POSIDONIA ’ 179

pteropod Cavolinia globosa of length 0-8 cm. sank at 6-3 cm. /sec. in tap water. This is
the same as the settling velocity of a model of the shell of B. buchi of length 1 cm.,
angle of opening 60° and shell thickness 50 p. This experiment, in itself, establishes the
feasibility of a nektoplanktonic mode of life for B. buchi.

The method of sinking large models in viscous solutions to give low Reynolds' num-
bers should be widely applicable in studies of the functional morphology of small
biological objects such as microplankton and pollen grains.

THE MODE OF LIFE OF ‘ POSIDONIA ' RADIATA GOLDFUSS
Synonymy.

Posidonia radiata Goldfuss 1836, p. 119, pi. 1 14, fig. 4. Vadasz 1910, p. 41, fig. 8.

Posidonia hronni Zieten var. magna Quenstedt 1858, p. 260, pi. 37, fig. 8. (n.b. Posidonia bronnii
Voltz 1830 = Posidonia bronni parva Quenstedt 1858.)

Posidonomya {? Daonella) radiata (Goldfuss); Guillaume 1928, p. 219, pi. 10, fig. 1.

The two forms of ‘ Posidonia ’ of the German Posidonienschiefer are commonly called
Posidonomya bronni Voltz var. magna Quenstedt and Posidonomya bronni Voltz var.
parva Quenstedt (e.g. Hauff 1921, p. 19). They are probably distinct species since magna
reaches a length of about 5 cm. and parva only about 1 cm., and the detailed stratigraphical
distributions are different. Magna is of rock-forming abundance in the lowest beds of the
Posidonienschiefer, but scarcely occurs higher, where "parva' is very abundant.

Posidonomya bronni magna is identical with Posidonia radiata Goldfuss as judged by
Goldfuss’s figure. Goldfuss’s material came from the Lias of Boll, a famous locality for
Posidonienschiefer 7 km. east of Holzmaden. The ‘faint, scarcely visible, radiating
lines’ which gave it its name can be seen on material that can certainly be identified as
P. bronni magna from Holzmaden. The generic allocation of Posidonia radiata is un-
certain since the hinge is unknown. It is expedient to leave it in the genus Posidonia
Bronn 1828 (= Posidonomya Bronn 1837).

‘ Posidonia ’ radiata is larger and taller \~ = 0-95 j than B. buchi, with more median

umbones. It is mainly Toarcian in age. It is figured in PI. 19, fig. 5.

Facies. " Posidonia ' radiata occurs in much the same facies as Bositra buchi. Thus Renz
(1927, p. 487) notes its abundance as Posidonia bronni var. magna in the black, Upper
Liassic ‘Posidonienschiefer’ of Albania and Western Greece — which otherwise contains
only "P. bronni parva', aptychi, and in some beds a bivalve compared with Pseudo-
monotis substriata.

The best investigated occurrence, however, is in the Upper Lias of Holzmaden in
Swabia (Hauff 1921, p. 20). Hauff examined the quarries there for many years and pre-
sented his results in semi-quantitative form (op. cit., pi. 7-13).

The beds in which Hauff records "P.' radiata in extreme abundance are the somewhat
bituminous Koblenzer IQ (mainly at the top), Hainzen 1I2 and Fleins II3, representing
40 cm. of rock. Associated fossils recorded by Hauff include: ‘ Pentacrinus' briareus
(rare), "P'. briaroides (rare), "P.' subangularis (rare), "P. ' hiemeri (rare), "P.' colligatus
(rare), Ostrea sp. (common), Inoceramus dubius (common); four species of ammonite
(one very common, one common and two rare), and aptychi; three species of belemnites
(one common and two rare); four species of squid (all rare); two species of eryonid crab

180

PALAEONTOLOGY, VOLUME 8

(both rare); ten species of fish (all rare); two species of ichthyosaur (both rare); and
coprolites. The abundances are deduced from Hauff's distribution charts.

At first sight, as Haufif remarked (op. cit., p. 38) of the Posidonienschiefer in general,
there seem to be many benthonic species. However, the ‘ Pent acr inns’ spp. occur attached
to fossil wood and must be regarded as having floated in; the oysters ( Ostrea sp .) are
found attached to wood or ammonite shells, or in little nests where the ammonite shell
to which they were attached has apparently disappeared. Inoceramus dubius is also
commonly attached to wood. The only remaining possible benthos (leaving aside
‘/V radiata) are the crabs; no conclusions can be based on these, however, for they are
exceedingly rare; moreover, modern eryonids may have a certain degree of swimming
ability, for almost adult juveniles have been taken in mid-water (Dr. I. Gordon, personal
communication).

Among forms which die out at the base of the Koblenzer, where iP. ’ radiata becomes
common, are the burrow Fucoides granulatus, Rhynchonella amalthei and Plicatula
spinosa , which must have been benthonic. Belemnites become much rarer than lower in
the sequence. Thus B. paxillosus, which occurs abundantly (‘ Massenvorkommen’)
beneath the Koblenzer, is merely constantly present (‘durchgehendes Vorkommen’) in
the Koblenzer, Hainzen, and Fleins. The heavy skeleton of the belemnites suggests that
they were habitually benthonic, but, being swimmers, they might be expected occasion-
ally where the bottom was uninhabitable.

Thus there was probably no benthos in the beds with abundant ‘P. ’ radiata. The
apparent cases can be explained as pseudoplankton, or need not always have been
strictly benthonic, or are too rare to be used in evidence. In view of this, and the bitumen
in the rock, these 40 cm. of beds were probably deposited under stagnant conditions
with poisonous anaerobic bottom water. Hauff (op. cit., p. 42), following Pompeckj
(1901), compared the Posidonienschiefer with the bottom deposits of the Black Sea.

The view that the Posidonienschiefer are euxinic deposits is generally accepted in
Germany (discussion in Hauff, Jr. 1960, p. 36). Objections were raised by Beurlen ( 1925),
who held that the bivalves and crabs must have been benthonic. His views were reiterated
by Kuhn (1952, 1953, 1955) as regards Franconia, where crabs are commoner than at
Holzmaden and a conglomerate is present. Faber (1931 ) pointed out that in Luxembourg
the bivalves were often single valves only, indicating water movement. Finally, Einsele
and Mosebach (1955, p. 340) favoured gyttja rather than sapropel conditions for the
Holzmaden area, because of the presence of burrows at the base and at one level within
the Posidonienschiefer, the ubiquitous abundance of the tiny gastropod Coelodiscus
minutus, which had not been observed by Hauff, and the presence of eryonid crabs.

These objections, however, are lacking in force. In the first place, at Holzmaden
crabs are exceedingly rare, and not common even in Franconia. Limited swimming
ability is enough to explain their occurrence. The conglomerate suggests that conditions
in Franconia were not exactly as in Swabia. Secondly, the water movement required to
shift valves of Steinmannia bronni, ‘ Posidonia 1 radiata , or Inoceramus dubius would be
extremely slight, since these are very thin-shelled. Indeed an oscillation rather than a
current would be sufficient. Since deposition was probably slow such movements need
only be infrequent, and disarticulation of the valves is therefore compatible with
euxinic conditions. Thirdly, burrows are a sensitive indicator of benthos, in the sense
that a few benthonic animals can cause a great number of tracks. The absence of burrows,

JEFFERIES AND MINTON: MODE OF LIFE OF JURASSIC ‘ POSIDONIA ' 181

except at the base of the Posidonienscheifer and one horizon within them, is therefore
a strong argument for euxinic conditions in most of the formation. Burrows are absent
in the beds where ‘P. ’ radiata is abundant. Fourthly, the tiny thin-shelled Coelodiscus
minutus, which is also abundant in the equivalent Jet Rock of Yorkshire, was probably
pelagic. It much resembles the Recent and Tertiary pelagic gastropod Spiratella.
Hauff’s conclusion that the Posidonienschiefer in general, and the Koblenzer, Hainzen
and Fleins in particular, were deposited in anaerobic bottom water therefore remains
extremely probable.

Pieces of wood are common in the Holzmaden Posidonienschiefer, including the
Fleins. ‘ Pentacrinus’, Inoceramus dubius and Ostrea sp. are often attached to them, and,
assuming euxinic conditions, must have settled and grown on the wood when it was
floating. However, neither '‘Pd radiata nor S. bronni has ever been found attached to
wood. A particularly striking specimen is preserved in the Hauff Museum (Hauff 1960,
pi. 63). This is a slab of Fleins, 112x68 cm. in size, with a piece of driftwood near one
edge to which are attached Inoceramus dubius and ‘ Pentacrinus ’ subangularis. Distributed
over the rest of the surface of the block are innumerable valves of ‘P.’ radiata and there
must be several thousand more of these buried in the block. Those on the surface have
been smoothed off during preparation, so that on Hauff’s photograph they merely
appear as a pepper-and-salt effect. They can readily be seen, however, on the block itself.
A very similar specimen (E25388) from Boll is displayed in the British Museum (Natural
History). This shows a mass of ‘ Pentacrinus ’ subangularis embedded in Fleins with
numerous specimens of ‘P.’ radiata scattered over the rock. I. dubius and pieces of wood
do not occur in this specimen but it was probably formed in much the same way as the
specimen in the Hauff Museum.

These specimens show that ‘P.’ radiata had a different mode of life from, for instance,
Inoceramus dubius. Since conditions were euxinic, it must have been pelagic. It is
improbable that it was pseudoplanktonic, since other pseudoplankton was dependent
on lumps of wood or ammonites to remain afloat. A nektoplanktonic mode of life is
strongly indicated.

Functioned morphology. At Holzmaden ‘P. ’ radiata is always so compressed that studies
of functional morphology are very difficult. The anterior margin, however, appears to
be circular with no sign of an anterior ear or any concave portion that might suggest
attachment. It can be contrasted with Inoceramus dubius, which is mytiliform.

Specimens in the Fleins sometimes occur with the valves open like B. buchi. In a quarry
on the south side of the road from Zell to Ohmden near Holzmaden, 33 such specimens
were concave-up and 5 concave-down. The same conclusions apply as with B. buchi.
Guillaume (1928, p. 10, fig. 1) figured specimens of "P. ’ radiata , including one in the
valves-open position, which seem less crushed than usual. The figures confirm that the
anterior margin was circular while the posterior dorsal part of each valve was convex
outwards in a way that strongly suggests a posterior gape. As in B. buchi this would
indicate swimming ability. Guillaume’s figured specimens were not available to the
present authors.

Because of the lack of well-preserved specimens, the shell thickness of B. buchi cannot
be accurately estimated but it was much thinner-shelled than Inoceramus dubius in the
same bed.

182

PALAEONTOLOGY, VOLUME 8

Feasibility of a planktonic mode of life. Experiments show that, with the same shell
thickness, 5 cm. models sink slower than 2 cm. models. For example, with a shell thick-
ness of 50 /x, angle of opening 60° and no tentacles, settling velocity was respectively
5-3 cm. /sec. and 6-7 cm./sec. This decrease in velocity with size may be due to the hydro-
dynamic effect of the sides of the jar in which the experiment was carried out, but there
is no reason to doubt that, in an infinite medium, the settling velocities would be similar
to each other. In addition, the available tissue buoyancy at 5 cm. could be greater than
at smaller sizes. A nektoplanktonic mode of life is therefore feasible for ‘P.’ radiata as
for B. buchi.

Conclusions. The evidence concerning ‘P. ’ radiata may be summarized as follows:
firstly, on facies grounds as with B. buchi , the species was almost certainly not benthonic,
and its failure to occur attached to wood, like obvious pseudoplankton in the same beds,
indicates that it was not pseudoplanktonic. Secondly, the circular anterior margin con-
firms that it was not attached. Thirdly, the thin shell, the probable presence of a posterior
gape, and the occurrence of the valves-open position suggest swimming ability. 'P. '
radiata was therefore probably nektoplanktonic.

The same conclusions may apply, at least in the adult, to the relatives of B. buchi and
‘P. ’ radiata occurring in the same characteristic facies, including Steinmannia, Halobia ,
Daonella , Dunbar ella, Mono t is , Buchiola and Palaeozoic and Triassic occurrences of
Posidonia. Each species needs to be examined ecologically from this point of view using
the twin approaches of facies and functional morphology.

Acknowledgements. The authors wish to thank numerous colleagues with whom they have discussed
the subject of this paper, and in particular Dr. L. R. Cox for help with systematics, Prof. C. M. White
for discussion of the hydrodynamics, and also Mr. G. F. Elliott, Dr. C. G. Adams, Dr. W. T. Dean,
Dr. A. Smith, and Prof. R. G. S. Hudson. Dr. T. H. J. Gilmour kindly supplied information on the
habits of Recent Limidae, and Dr. K. W. Ockelmann drew attention to Cyclopecten groenlandicus.

For help in the field, response to inquiries, or permission to work at the institutes under their
charge they wish to thank Dr. B. Hauflf, Jr. (Hauff Museum, Holzmaden), Dr. F. S. Russell (Plymouth
Marine Biology Laboratory), and Dr. J. P. Thieuloy (Department of Geology, University of Grenoble).
Dr. H. Hess kindly supplied specimens from la Voulte-s.-Rhone. Mr. A. F.. Rixon made many helpful
technical suggestions. Mr. H. G. Owen helped with the experiments, took some of the photographs
and discovered the material of B. buchi showing the hinge.

This paper is published by permission of the Trustees of the British Museum (Natural History).
Specimen numbers refer to specimens in the Museum collections, unless otherwise stated.

REFERENCES

agip mineraria 1959. Microfacies italiane. Milan, 35 pp., 145 pi.

allan, r. s. 1956. Report of the standing committee on datum planes in the geological history of the
Pacific region. Proc. Pan-Pacif Sci. Congr. (1953) 2, 325-423.
arkei.l, w. j. 1956. Jurassic geology of the world. London.

bellen, r. c. van, dunnington, H. v., Wetzel, r., and morton, D. M. 1960. Lexique stratigraphique
Internationale. 3. Asie. Fascicule 1 0a. Iraq. Paris.

Bernard, f. 1 896. Deuxieme note sur le developpement et la morphologie de la coquille chez les lamelli-
branches (Taxodontes). Bull. Soc. geol. Fr. 24, 54-82.
beurlen, k. 1925. Einige Bemerkungen zur Sedimentation in dem Posidonienschiefer Holzmadens.
Jber. oberrhein. geol. Ver. 14, 298-302.

boehm, g. 1912. Beitrage zur Geologie von Niederlandisch-Indien. 1. Abteilung. Die Sudkilsten der

JEFFERIES AND MINTON: MODE OF LIFE OF JURASSIC ‘ POSIDONIA' 183

Sula Inseln Taliabu und Mangoli. 4. Abschnitt Unteres Callovien. Palaeontographica, Suppl. 4, 1
(4), 123-79, pi. 32-44.

clarke, J. m. 1904. The Naples fauna. 2. Mem. N. Y. St. Mus. 6, 199-454, pi. 1-20.
colom, g. 1955. Jurassic-Cretaceous pelagic sediments of the Western Mediterranean zone and the
Atlantic area. Micropaleontology , 1, 109-23, pi. 1-5.

1957. La sedimentation des geosynclinaux mesozo'iques. Bull. Soc. geol. Fr. 7, 1 167-88.

cox, l. R. 1943. The English Upper Lias and Inferior Oolite species of Lima. Proc. malac. Soc. Lond.
25, 151-87, pi. 6-29.

1960. Thoughts on the classification of the Bivalvia. Ibid. 34, 60-88.

1964. Notes concerning the taxonomy and nomenclature of fossil Bivalvia (mainly Mesozoic).

Ibid. 36, 39-48, pi. 1.

craig, g. y. 1954. The palaeoecology of the Top Hosie Shale (Lower Carboniferous) at a locality near
Kilsyth. Quart. J. geol. Soc. Lond. 110, 103-19.

dalmon, J. 1935. Note sur la biologie du Petoncle ( Chlamys varius). Rev. Trav. Off. Peches marit. 8,
268-81.

diener, c. 1912. Lebensweise und Verbreitung der Ammoniten. Neues Jb. Miner. 2, 67-89.
finsele, g. and mosebacei, r. 1955. Zur Petrographie, Fossilerhaltung und Entstehung der Gesteine
des Posidonienschiefers im Schwabischen Jura. Neues. Jb. Geol. 101, 319-430, pi. 36-40.
elliott, G. f. 1962. More problematica from the Middle East. Micropaleontology, 8, 29-44, pi. 1-6.
faber, g. 1931. Beitrag zur Kenntnis des Posidonienmeeres. Bull. Soc. Nat. luxemb. (n.s.) 41, 1 12-13.
gilmour, t. h. j. 1963. A note on the tentacles of Lima hians (Gmelin), Bivalvia. Proc. malac. Soc.
Lond. 35, 83-85.

c.oldfuss, A. 1826-1844. Petrifacta germaniae, etc. Diisseldorf, 3 vols.
goldstein, s. 1938. Modern developments in fluid dynamics. Oxford, 2 vols.

gregorio, a. de. 1886. Monographic des fossiles de Valpone (Mont Grappa) du sous-horizon Grappin
de Greg. Annls Geol. Paleont. 2, 1-20, pi. 1-6.

grober, p. 1918. Estratigrafia del Doger en la Republica Argentina. Boln. Dir. Minas Geol. B. Aires,
18, 1-81, pi. 1-5.

guillaume, l. 1928. Revision des Posidonomyes jurassiques. Bull. Soc. geol. Fr. 27, 217-34, pi. 10.
haas, f. 1935-56. In bronn, h. g. Klassen und Ordnungen des Tierreichs. 3, Abt. 2: Bivalvia. Leipzig,
Teil 1, 2.

hardy, a. c. and bainbridge, r. 1954. Experimental observations on the vertical migrations of plank-
ton animals. J. Mar. biol. ,455. U.K. 33, 409-48.
harvey, H. w. 1955. The chemistry and fertility of sea waters. Cambridge.

hauff. b. 1921. Untersuchung der Fossilfundstatten von Holzmaden im Posidonienschiefer des oberen
Lias Wtirttembergs. Palaeontographica, 64, 1-42, pi. 1-21.
hauff, b. (Jun.). I960. Das Holzmadenbuch. Ohringen.

haug, e. 1892. Les chaines subalpines entre Gap at Digne. Bull. Carte geol. Fr. 3 (21), 1-197, pi. 1-4.
1907. Trait e de geologic, 1. Paris.

hess, h. 1960. Neubeschreibung von Geocoma elegans (Ophiuroidea) aus dem unteren Callovien von
La Voulte-sur-Rhone (Ardeche). Eclog. geol. Helv. 53, 335-85.
hind, w. 1893. Description of a slab from the shale above the Kinder Scout Grit, Rabchester, Lanca-
shire. Geol. Mag. (3) 10, 540-1.

1896-1905. A monograph of the British Carboniferous Lamellibranchiata. Palaeontogr. Soc.

(. Monogr .) 1, 1-486, pi. 1-54; 2, 1-222, pi. 1-25.

Hudson, R. g. s. and cotton, g. 1943. The Namurian of Alport Dale, Derbyshire. Proc. Yorks, geol.
Soc. 25, 142-73.

hundt, R. 1939. Zur Lebensweise von Pterinaea. Zbl. Miner. Geol. Palaont. (B), 3, 105-7.
ichikawa, k. 1958. Zur Taxionomie und Phylogenie der triadischen ‘Pteriidae’ (Lamellibranch.) mit
besonderer Beriicksichtigung der Gattungen Claraia, Eumorphotis, Oxytoma und Monotis. Palaeon-
tographica, (A), 111 (5), 131-212, pi. 21-24.

jackson, r. t. 1890. Phylogeny of the Pelecypoda. The Aviculidae and their allies. Mem. Boston Soc.
nut. Hist. 4, 277-400,' pi. 23-30.

jensen, a. s. 1905. On the mollusca of East Greenland, 1. Lamellibranchiata. Medd. Gronland, 29,
299-362.

184 PALAEONTOLOGY, VOLUME 8

kornicker, l. s. 1959. Observations on the behavior of the pteropod Creseis acicula Rang. Bull. mar.
Sci. Gulf Corrib. 9,331-6.

krumbf.ck, c. 1921-4. Die Brachiopoden, Lamellibranchiaten und Gastropoden der Trias von Timor,
1. 1, Stratigraphischer Teil; 2, Palaontologischer Teil. Paldontologie von Timor, 10 (17) 1-142;
13 (22) 1-271, pi. 1-20.

kuhn, o. 1952. Neue Crustacea Decapoda und Insecta aus dem untersten Lias e von Nordfranken.
Palaeontographica, (A), 101, 153-66, pi. 13-16.

1953. Zur Frage der Durchliiftung des frankischen Lias e Meeres. Geol. Bl. Nordost-Bavern, 3,

43-45.

1955. War das siiddeutsche Posidonienschiefermeer lebensfeindlich? Neues Jb. Geol., Mb., 504-6.

lebour, m. v. 1937. Larval and post-larval Lima from Plymouth. J. Mar. biol. Ass. U.K. (n.s.) 21,
705-10.

iemche, h. 1946. Northern and arctic tectibranch gastropods, i, The larval shells, ii, A revision of the
cephalaspid species. K. Danske Vidensk. Selsk. (Biol.), 5 (3), 1-136.
miyake, y. and KOizuMr, m. 1948. Measurement of the viscosity coefficient of sea water. J. Mar. Res. 7,
63-66.

miyazaki, i. 1935. On the development of some marine bivalves with special reference to the shelled
larvae. ./. Fish. Inst. Tokyo, 31. 1-10, pi. 1-3.

molengraaf, g. a. f. von. 1917. De Timorexpeditie en hare palaeontologische resultaten. Hand. ned.

nat. en geneesk. Congr. (16, ’s Gravenhage), [near p. 250 (not seen)].
morton, j. e. 1954. The biology of Limacina retroversa. J. Mar. biol. Ass. U.K. 33, 297-312.
nalivkin, b. v. 1956. Pseudoplanktonic pelecypods of Domanik. C. R. Acad. Sci. U.R.S.S. Ill, 188.
[In Russian.]

Newell, n. d. 1937. Late Palaeozoic pelecypods. Pectinacea. Rep. Kans. geol. Survey, 10, 1-123, pi.
1-20.

1955. Permian pelecypods of East Greenland. Medd. Gronland, 110, 1-36, pi. 1-5.

rigby, j. k., fischer, a. g., Whiteman, a. j., hickox, j. e., and bradlf.y, j. s. 1953. The Permian

reef complex of the Guadalupe Mountains region, Texas and New Mexico. San Francisco.
odhner, n. j. 1914. Notizen tiber die Fauna der Adria bei Rovigno. Zool. Anz. 44, 156-70.
paul, h. 1939. Eine festgeheftete Muschel des Kulm-Meeres, Posidonia becheri. Natur u. Volk. 69,
402-6.

peyre, y. 1959. Etude sur les organismes du jurassique presentant en section taillee l'aspect de fila-
ments. Rev. Micropaleont. 2, 80-87, pi. 1-2.

pompeckj, j. F. 1901. Die Juraablagerungen zwischen Regensburg und Regenstauf. Geogn. Jh. 14,
139-220.

1914. Das Meer des Kupferschiefers. In Branca Festschrift, 445-94. Leipzig.

prokovlev, v. a. 1959. On the systematic constitution and stratigraphical significance of the Upper
Devonian pelecypods of the central part of the Volga- Ural area. Bull. Acad. Sci. U.R.S.S. (Ser.
geol.) 2, 119-22. [In Russian.]
quenstedt, f. a. von. 1856-8. Der Jura. Tubingen.

rees, w. j. in cox, i. (Ed.) 1957. The living scallop. In The Scallop, 17-31. London.
renz, c. 1927. Die Verbreitung und Entwicklung des Oberlias und Doggers im adriatisch-ionischen
Faziesgebiet von Hellas und Albanien. Verb. Naturf. Ges. Basel, 38, 482-515, pi. 5.
romer, f. a. 1835-9. Die Versteinerungen des norddeutschen Oolithengebirges. Hannover.
rothpletz, a. 1892. Die Perm, Trias- und Juraformation auf Timor und Rotti. Palaeontographica,
39, 57-106, pi. 9-14.

sadykov, a. m. 1962. Middle palaeozoic bivalve mollusca of Atasa. Alma-Ata. [In Russian.]
schafer, w. 1962. Aktuo-paldontologie nach Studien in der Nordsee. Frankfurt am Main (Sencken-
berg. Ges.).

schiemenz, p. 1905. Pteropoden. Erg. Atlant. Planktonexped. (Zool), 2. Kiel and Leipzig.
schmidt, h. 1935. Die bionomische Einteilung der fossilen Meeresboden. Fortschr. Geol. 12 (38),
1-154.

schwarzacher, w. 1948. Sedimentpetrographische Untersuchungen kalkalpiner Gesteine. Halstatter-
kalke von Halstatt und Ischl. Jb. Geol. Reichsanst. Bundesanst., Wien, 91 (1946), 1-48.
step, e. 1951. Shell Life. London.

JEFFERIES AND MINTON: MODE OF LIFE OF JURASSIC ‘ POSIDONIA' 185

steinmann, G. 1881. Zur Kenntnis der Jura- und Kreideformation von Caracoles (Bolivia). Neues Jb.
Min. Geol. Paldont. 1, 256-60, pi. 10.

stremoouchow, m. d. 1895. Note sur la Posidonomyci buchi Roeiner des schistes de Balaclava en
Crimee. Bull. Soc. nat. Moscou, 391-5, pi. 10.

studnitz, G. von. 1931. Die Morphologie und Anatomie von Lima inflat a, der Feilenmuschel, nebst
biologischen Untersuchungen an Lima hians Grnel. ZooL Jb. Anat. Ont. 53, 199-316.
thorson, g. 1946. Reproduction and development of Danish marine bottom invertebrates. Medcl.

Komm. Havundersog., Kbh., Ser. Plankton , 4, 1-523.
timmermann, Ci. 1932. Biogeographische Untersuchungen fiber die Lebensgemeinschaft des treibenden
Golfkrautes. Z. Morph. Okol. Tiere, 25, 288-385.

vadasz, e. 1910. Die Juraschichten des siidlichen Bakony. Result. Wiss. Ergebn. Balatonsees, 1 (1),
Pal. Anh. 3 (9), 1-89, pi. 1-2.

verrill, A. e. 1897. A study of the family Pectinidae, with a revision of the genera and subgenera.
Trans. Conn. Acad. Arts Sci. 10, 41-96, pi. 16-21.

wanner, j. 1907. Triaspectrefakten der Molukken und des Timor-Archipels. Neues Jb. Miner. Geol.
Paldont. 24, 161-220, pi. 7-12.

weigei.t, j. 1922. Die Bedeutung der Jugendformen karbonischer Posidonomyen fur ihre Systematik.
Palaeontographica, 64, 43-130, pi. 22-31.

1927. Das Posidonomyenproblem im Obercarbon Westfalens. Jb. Hallesch. Verb. Erf. Mit-

teldtsch. Bodensch. (n.f.), 6, 72-82.

white, c. m. 1946. The drag of cylinders in fluids at low speeds. Proc. roy. Soc. (A), 186, 472-9.
yonge, c. m. 1938. The evolution of the swimming habit in the lamellibranchia. Mem. Mas. Hist. nat.
Belg. 3, 77-100.

zangerl, r. and Richardson, e. s. 1963. The paleoecological history of two Pennsylvanian black
shales. Fieldiana, Geol. Mem. 4, 1-352, pi. 1-55.

R. P. S. JEFFERIES

Dept, of Palaeontology,

British Museum (Natural History),
London, S.W. 7

P. MINTON

Dept, of Civil Engineering,

Imperial College of Science and Technology,
London, S.W. 7

Manuscript received 28 February 1964

THE PALAEOECOLOGY OF THE GONIATITE BED
AT COWLOW NICK, CASTLETON, DERBYSHIRE

by TREVOR DAVID FORD

Abstract. The Cowlow Nick Goniatite Bed is shown to be an accumulation of randomly oriented, hollow, or
spar-filled goniatite shells of several species in a matrix of calcilutite. Of very limited dimensions, the bed is sur-
rounded by algal limestones of the fore-reef facies of Upper B» age (Lower Carboniferous). Suggestions as to
the mode of accumulation are discussed and it is concluded that the bed represents a specialized drifted assem-
blage of floating shells which were washed gently into an inactive surge channel or submarine cave. Some com-
parable occurrences elsewhere in Derbyshire are noted.

The small bed of limestone crowded with goniatites occurring in Cowlow Nick, west of
Castleton in Derbyshire (Nat. Grid Ref. SK/ 142824) is well-known to workers on the
Carboniferous Limestone. It was first recorded by Bisat (1934) who described species
found there by G. B. Alexander. Subsequently it was included in the Castleton Lime-
stones (of reef facies) by Shirley and Horsfield (1940) who included these in the B2 sub-
zone, resting unconformably against ‘cliffs’ of the massif-facies D zone limestones.
Parkinson (1947) agreed in placing the Cowlow Nick beds in the B2 sub-zone, but thought
that this reef facies was equivalent in age to the massif-facies limestones to the south
which are of Dx age. Both Hudson and Cotton (1945) and Parkinson (1947) discussed
further subdivision of the B2 sub-zone, but as there is some discrepancy between the
results of these authors and with the indications of the fauna listed herein, there seems
to be some revision needed.

None of the above workers discussed the palaeoecology of the Goniatite Bed except
as a very broad interpretation of all the pockets of fossils in the reef limestones as
drifted assemblages. Wolfenden (1958) described the palaeoecology of the whole reef
complex of B2/D1 age by analysing the distribution of the various elements of the fauna
in terms of fore-reef, reef, back-reef, and shelf biotopes. The existence of a basinal
biotope was noted, but not analysed. Goniatites were noted as being almost confined to
the fore-reef facies (and basin by inference). Wolfenden challenged the drifted-assem-
blage concept of Hudson and Cotton (1945) and said that as the goniatites occurred
throughout the fore-reef in all sizes and were well preserved, and were not often
transported out of that biotope, they were probably preserved where they lived — a life
assemblage. Wolfenden, however, did not discuss the Cowlow Nick Goniatite Bed nor
did he mention localized accumulation of goniatites to the almost total exclusion of
other members of the biota.

It is the purpose of this short communication to analyse the composition of the fauna
in the Cowlow Nick Goniatite Bed, to examine its relationship with surrounding
deposits and to attempt to draw conclusions about the mode of accumulation.

THE FAUNA

The material collected by Alexander and described by Bisat (1934) included: Beyricho-
ceras submicronotum Bisat, B. vesciculiferum (de Koninck), Goniatites maximus Bisat,

[Palaeontology, Vol. 8, Part 1, 1965, pp. 186-91, pi. 20.]

T. D. FORD: PAL AEOECOLOG Y OF THE COWLOW NICK GONIATITE BED 187

Dimorphoceras gilbertsoni (Phillips), Sagittoceras acutum Hind (= discus Roemer),
Pronorites cyclolobus (Phillips), Prolecanites discoides Foord and Crick.

Shirley and Horsfield (1940), Hudson and Cotton (1945), and Parkinson (1947) failed
to add to this, though other species were found in the Castleton Limestones at the nearby
localities of Cavedale and Treak Cliff. Wolfenden’s list (1958) noted facies distribution,
but not individual localities.

Further collecting by the writer and by Dr. J. R. L. Allen has yielded: B. rectangu-
larum Bisat, G. cf. hudsoni Bisat, P. serpentinus Phillips, P. cf. P. discoides Foord and
Crick, ? Discitoceras sp., ‘ Orthoceras 1 spp.

A sample of the bed was broken up and the following abundances of the various
species were noted:

B. submicronotum 29

B. vesciculiferum 2

B. rectangularwn 8

G. maximus 29

G. cf. hudsoni 1

D. gilbertsoni 1

P. serpentinus 1

P. cf. P. discoides 3

‘ Orthoceras ’ spp. 6

? Discitoceras sp. 1

Total 81

Other samples produced much the same proportions, showing a slight preponderance
of Beyrichoceras spp. over Goniatites spp. Owing to the difficulties of extraction of the
goniatites in an undamaged condition (particularly of young forms) no reliable size
frequency distributions could be recorded. However, both polished and thin sections
show all sizes to be present, although adults are predominant with only a few young.
The sizes of the adults in the different species groups differ considerably. In the Beyri-
choceratids the diameters range from 15 mm. to 20 mm. whilst in the Goniatites spp.
the range is from 18 mm. to 25 mm. Orthoceratids are scattered throughout and vary
up to 100 mm. in length and 20 mm. diameter.

The Goniatite Bed has only a few other macrofossils, except for a small lens of young
brachiopods near the roof. This is seen only for about 30 cm. in length and up to 10 cm.
thick, and is composed of young Dielasma hastata, all much the same size, about 7-8
mm. in length. A few productids are scattered through the bed, and also a few gastro-
pods, some of which show traces of original colouring. The microfossils present are
bryozoan fragments, and very occasional young brachiopods, foraminifera, and
ostracods.

STRATIGRAPHICAL RELATIONSHIPS

The Cowlow Nick Goniatite Bed reaches a maximum of nearly 1 metre thickness in
the mouth of an old lead-vein working, now partly collapsed. As far as can be judged
from the irregular bedding it has a dip of some 15° NE. which is rather less than the hill
slope and considerably less than the dips in the neighbouring outcrops of reef-complex

188

PALAEONTOLOGY, VOLUME 8

limestones, which range between 20° and 40° within a few hundred metres. As a result of
these relationships, the Goniatite Bed disappears into the hillside on the up-dip side,
whilst down dip its base is lost by the steep hill-slope. Along the strike exposures are
discontinuous and the bed is lost beneath the turf only 3 m. to either side of the lead
vein. The vein is, however, oblique to the strike and in the north-west wall the bed
appears to wedge out in both directions, with about 3-5 m. between the extremities. The
nearest exposures in all directions are calcite-mudstones of the reef-complex frequently
showing algal banding, and with pockets of reef-brachiopods. Lenses of crinoidal
debris appear lower down the hill slope.

Thus the detailed stratigraphical relationships of the bed are obscure, but in the general
sense it may be said that it occurs in steeply dipping reef-complex limestones, which by
projection south-eastwards of Wolfenden’s (1958) mapping in the Winnats Pass are
within the fore-reef facies, close to the reef wall.

Fortunately within the lead-vein opening parts of the roof and floor of the bed are
visible, though only for a short distance laterally. The roof is composed of calcite-
mudstone with algal banding. Immediately beneath this is the lens of young brachiopods
in a matrix of calcite-mudstone. A metre below is the floor of the Goniatite Bed, which
is composed of either a fragmental limestone with small crinoid ossicles, shell-fragments
and calcite-mudstone fragments or of calcite-mudstone with algal banding. These are
seen only for a length of about 0-5 m. and to a similar depth below the goniatite bed, and
within this grain-size is variable up to about 8 mm.

PETROGRAPHY

The goniatites in the Cowlow Nick bed are of varied sizes as discussed above. They
are randomly oriented in a fine-grained calcite-mudstone matrix. The shells are almost
undamaged and the ornament is not worn. Body chambers can be seen in sections, but
can rarely be extracted undamaged. No evidence of size sorting has been obtained.

Diagenetic alterations obscure much of the detail and the following features have
been observed: (u) frequent veinlets of calcite and occasional fluorite cutting both
goniatites and matrix: (b) patches of calcite spar, some with carbonaceous impurities,
sometimes surrounded by algal reef ‘ tufa ’ ; (c) fine calcite crystalline mosaic in the matrix
obscuring details of original depositional textures; ( d ) recrystallization of parts of the
shells to form either coarse calcite spar or fine crystal mosaic, and ( e ) scarce authigenic
quartz crystals and patches of limonite staining. In places brecciation appears to be
associated with the mineral vein.

The goniatite shells are mostly filled with coarse calcite spar. A few are still hollow or
with only some chambers hollow and lined with calcite crystals. In a few cases there are
crystals of fluorite or blebs of bitumen in either the spar filling or in the hollow shells.
Some shells, which appear to be broken on a polished section, are seen in thin section

EXPLANATION OF PLATE 20

Polished vertical section of the Cowlow Nick Goniatite Bed. Approximately twice natural size. The
section shows the random orientation of goniatite shells of various sizes in a calcilutite matrix. Both
primarily damaged and diagenetically recrystallized shells are present. Cavity fillings of calcite, fluorite,
and carbonaceous matter are also present.

Palaeontology, Vol. 8

PLATE 20

FORD, Cowlow Nick goniatite bed

T. D. FORD: PALAEOECOLOG Y OF THE COWLOW NICK GONIATITE BED 189

to pass into the fine crystal mosaic of the matrix. No evidence has been found of original
sedimentary filling of the shells other than in body chambers.

The fine grained matrix appears to be largely due to diagenetic recrystallization, but
in places traces of original textures remain, chiefly irregular algal banding.

PALAEOECOLOGY

From the above it may be surmised that the Goniatite Bed is an assemblage of several
species of goniatites, of varying sizes, preserved virtually undamaged and randomly
oriented in a lenticular bed of restricted dimensions in the reef-complex close to the
probable position of the algal reef-wall. In many of its characters the Bed compares
with the well-known reef-brachiopod beds of the reef-complex on Treak Cliff, &c. Three
possible explanations may now be examined:

A. The Goniatites were gregarious sessile benthonic animals, not, as usually supposed,
free-swimmers, and they were preserved as a ‘life-assemblage’, possibly by some
catastrophe.

B. The Goniatites were, as usually supposed, free-swimming reef-grazers, which at
times of storm, &c., retired to a sheltered hollow in the reef, where they were over-
whelmed by a catastrophe.

C. The Goniatites are an unusual drifted assemblage, for which some mechanism
must be found to account for a concentration of floating shells in the midst of
largely algal limestones.

The first of these is difficult to uphold as it is completely in contrast to most opinion
of the habits of ammonoids. Also one would expect such sessile gregarious colonies to
have been preserved much more frequently. The same species are found in shale facies
elsewhere associated with planktonic and epi-planktonic faunas, and they are also found
as scattered individuals in association with reef brachiopods elsewhere in the reef com-
plex. Finally, gregarious sessile benthonic associations are usually formed of a single
species.

The possibility B seems more likely at first as it does not require any unusual means
for concentrating shells if they retired to this shelter of their own accord. However, one
would again expect to find such occurrences more often. If the goniatites had retired to
a sheltered hollow during some time of stress only to be trapped there by some catas-
trophe such as the collapse of flanking deposits, it would be expected that the shells
would be crushed together, not dispersed at random through the Bed. Also such a shel-
tered hollow might be expected to be populated with a sessile fauna before being occu-
pied by the goniatites.

The third possibility C, that of a special type of drifted assemblage, must now be
examined, and a perusal of literature on modern reefs has given a suggestion. In the
Bikini Atoll Reports, Emery and others (1954, part A pi. 51, and part C figs. 92 and 93)
have shown diagrams of surge channels in the active reef edge, and they have described
there the gradual closure and abandonment of such surge channels by algal growth over
the top and outwards around the mouths of the channels, leaving submarine caves. It is
thus possible that the Cowlow Nick Goniatite Bed represents a concentration of
goniatite shells in a surge channel, in a Lower Carboniferous reef complex. The fact that

190

PALAEONTOLOGY, VOLUME 8

the shells are not usually broken indicates that it could not have been an active surge
channel with the repeated pressure of heavy waves. Shells recently dead and floating at
or near the sea surface could be washed into an inactive channel, which algal growth or
fore-reef accumulation had left as a submarine cave. Only gentle currents would be
required to wash in floating goniatites. Such inactive surge channels could be potential
homes for sessile brachiopods, bryozoans, &c., but the exclusion of all but a few
scattered individuals may be explained by supposing that the Cowlow channel was open
for a relatively short time and that as fast as the shells were washed in they were en-
veloped in algal deposits and precipitated calcite mud, now both represented by the
recrystallized calcilutite matrix. It is also possible that B above is partly correct and that
this particular surge channel was used as a shelter by a few goniatites which devoured
the incoming spat of any sessile organisms.

In suggesting that the Cowlow Nick Goniatite Bed is a concentrate of shells in an
inactive surge channel, the concept of it being a drifted assemblage as suggested by
Hudson and Cotton (1945) is maintained, with the addition of details of the mechanism.

COMPARATIVE EVIDENCE FROM OTHER LOCALITIES

Despite a search of the literature on cephalopods, no description of a comparable
accumulation of cephalopod shells has been found. Nor has any comparable accumula-
tion of modern Nautilus shells been recorded. Wolfenden (1958) mentioned 'pockets of
goniatites’ in Upper Dovedale but gave no localities or descriptions.

There are, however, some other comparable occurrences known to the writer in
Derbyshire. A block of calcite mudstone containing fifteen Goniatites maximus group in
about 2 kgm. of matrix was found loose in the old quarries above Treak Cliff Cavern
(SIC/1 3583 1), but no parent rock could be found and it may have been quarried away.
High on the summit ridge of Treak Cliff a small pocket in the middle of an algal lime-
stone mass (Wolfenden’s upper reef wall) contained six goniatites in what must have
been a hollow some 20 cm. across between adjacent algal domes — perhaps not a
surge channel, but at least a temporary hollow in the reef surface. Near Brassington
(SK/224545) a mass of calcite mudstone little more than a cubic metre has been found
crowded with Prolecanites cf. discoides close to an algal limestone mass. Here again
detailed relationships are obscured by a small mineral vein and by extensive dolomitiza-
tion all round the outcrop, and the shells have been impossible to extract as most are
still hollow.

It seems from these few occurrences, and there are doubtless others not seen by the
writer, that the conditions at Cowlow Nick were not unique, though they were appa-
rently best developed there.

CONCLUSION

From the evidence of a variety of species and sizes, preserved undamaged in a re-
stricted lenticular deposit in the midst of algal fore-reef limestones, it is concluded that
the goniatites accumulated as a result of gentle current sorting and washing into a hollow
such as an inactive surge channel or submarine cave.

Acknowledgements. The writer would like to thank Dr. J. R. L. Allen for his manuscript notes and for
discussion of the problem. Thanks are also due to Drs. F. M. Broadhurst and H. Sadler for critical
reading of the manuscript.

T. D. FORD: PALAEOECOLOG Y OF THE COWLOW NICK GONIATITE BED 191

REFERENCES

bisat, w. a. 1934. The Goniatites of the Beyrichoceras Zone in Northern England. Proc. Yorks. Geo!.
Soc. 22, 280-309.

emery, k. o., tracey, j. i., and ladd, h. s. 1954. Geology of Bikini and nearby Atolls. U.S. Geol.

Snrv. Prof. Paper, 260, Part A, 265 pp.. Part C, 275-80.

Hudson, r. G. s. and cotton, g. 1954. The Lower Carboniferous in a boring at Alport, Derbyshire.
Proc. Yorks. Geol. Soc. 25, 254-330.

parkinson, d. 1947. The Lower Carboniferous of the Castleton District, Derbyshire. Ibid. 27,
99-125.

shirley, j. and horsfield, E. l. 1940. The Carboniferous Limestone of the Castleton-Bradwell Area,
Derbyshire. Quart. J. Geol. Soc. Loud. 96, 271-99.

wolfenden, e. b. 1958. Paleoecology of the Carboniferous Reef Complex and Shelf Limestones in
Northwest Derbyshire. Bull. Geol. Soc. Amer. 69, 871-98.

TREVOR D. FORD

Department of Geology,
The University,

Manuscript received 19 March 1964 Leicester

CALCIFOLIUM (CODIACEAE) FROM THE UPPER
VISEAN OF SCOTLAND

by IAIN C. BURGESS

Abstract. Two species of the algal genus Calcifolium Shvetzov and Birina occur in two Upper Visean lime-
stones in Ayrshire and Lanarkshire. At the lower horizon, that of the Hurlet Limestone, the characteristic form
is C. punctatum Maslov; in the overlying Blackhall Limestone, the species is replaced by C. okense Shvetzov and
Birina. Both species are easily recognized in thin sections, and the genus appears to be an accurate marker for
the Pi~E1 stratigraphical interval.

The codiacean algal genus Calcifolium was first described from limestones of the Okian
Series (Lower Carboniferous) of the Moscow Basin (Shvetzov and Birina 1935). It was
identified in thin sections as \ . . narrow (OT mm.) curved strips of different lengths
(0-3 to 2 mm.) with a fairly regular arrangement of round holes (diam. 0-016 mm.) close
to one edge. . . .’ (op. cit., quoted in Maslov 1956, p. 47). One species, C. okense, was
erected to include such fragments and also a few which were noted as having several
rows of perforations. The genus was redescribed by Maslov (1956), who assigned the
Tnany-rowed ’ fragments to a second species, C. punctatum, and also, from a complex
branch of C. okense, made a reconstruction of the probable shape of the thallus (Maslov
1956, pp. 49, 50, fig. 9; Johnson 1958, fig. 2). The genus was first recognized from the
Carboniferous of Great Britain in the Great Limestone (Ef) of the Alston Block and
Northumberland trough (Johnson 1958). The specimens resembled C. okense in having
only one row of perforations, but as they were of larger size, and differed in several
details, they were assigned to a new species, C. bruntonense.

During a revision of the Upper Visean strata in the western part of the Midland
Valley of Scotland, the occurrence was noted in two limestones of algal fragments refer-
able to Calcifolium. At the lower horizon, the Hurlet Limestone, C. punctatum is the
characteristic form; in the overlying Blackhall limestone, C. okense is found.

SYSTEMATIC DESCRIPTION
Genus calcifolium Shvetzov and Birina 1935
Type species. Calcifolium okense Shvetzov and Birina.

Diagnosis. The thallus consists of a tubular calcareous stem, at first rather large, then
branching dichotomously into smaller tubes of constant diameter. At intervals, the stem
gives rise to thin, flat, or slightly curved lateral plates, which may themselves branch to
form other plates. A dichotomously branching canal system is longitudinally arranged
in the stem wall, radially divergent in the lateral plates.

Remarks. Calcifolium occurs in limestone, usually where the lithology is biomicrite or
biomicrosparite, in association with other algae, foraminifera, and brachiopod, crinoid,

[Palaeontology, Vol. 8, Part 1, 1965, pp. 192-8, pi. 21-22.]

I. C. BURGESS: CALCIFOLIUM FROM THE VISfiAN OF SCOTLAND 193

bryozoan, and coral debris. The alga is not normally visible in hand specimen, except
where selective pyritization of the fragments has taken place. The calcite forming the
thallus is very fine-grained (less than 1ft) and appears dark in thin section. It is very
resistant to recrystallization, and the alga is frequently preserved when almost all the
other detritus in the rock has been destroyed.

Calcifolium okense Shvetzov and Birina
Plate 21, figs. 1-5; text-fig. la-c.

Calcifolium okense (pars) Shvetzov and Birina 1935, pi. 4, figs. 11,12, 14, 1 5 (fide. Maslov 1956).
Calcifolium okense Shvetzov and Birina; Maslov 1956, pi. 8, figs. 1, 3-7; pi. 9, figs. 2-5; pi. 10,
figs. 1, 3-5; text-fig. la, b.

Diagnosis. Calcifolium , in which a single series of canals lies close to the inner surface
of the tubular stem, and close to one side of each lateral plate. Dimensions as in
Table 1.

table 1 . Dimensions of Calcifolium spp. from the West of Scotland, the Moscow Basin,

and Northern England.

This paper

Maslov 1956

Johnson 1958

okense

punctatum

okense

punctatum

bruntonense

Diameter of stem-tube

300-400 p

500-1,200 p

300-400 p

1,000 p

Thickness of stem wall

100-1 80 /x

100-250 p

70-1 00 p

1 00-200 /x

Diameter of stem wall canals

20-25 p

10-25 p

40 /x

Thickness of lateral plates .
Maximum length of individual

60-1 70 ii

80-140 /x

70-100 jit.

100 p

100-150 /x

plates .....

3 mm.

4 mm.

2 mm.

Diameter of canals in plates

1 5—30 /x

1 0-25 p

16 /x

20 p

20-50 p

Angle of branching of canals

20-40°

20-30°

10-25°

10-45°

15°

Distance between canals .

10-30 p

10-30 p

0-40 /x

0-40 p

1 5-20 p

Description. The stem wall is pierced by a single series of longitudinal canals, which are
close to the inner surface of the tube, except where they branch and cross over into the
bases of the lateral plates. These plates have an irregularly oblong triangular shape, or
are semi-discoidal or oval. Within the plates the canals diverge radially, branching
dichotomously, usually at an acute angle. They are always arranged close to one side of
the plate and branch approximately in one plane. In transverse sections, the plates appear
as dark, almost opaque strips with nearly circular, light-coloured specks — sections across
canals — close to one edge of the strip. In oblique sections of the plates, the canals have
the appearance of oval or oblong light-coloured strips. The plates often form lamellar
branches in which the canals are arranged on the same side as those in the initial plate.
As a result of repeated branching, very complex branches may be formed.

Remarks. The dimensions of Calcifolium from the Blackhall Limestone are given in
Table 1, where they are compared with those of other described specimens. Although
they cover a wider size range than that given by Maslov for the typical C. okense, it is
believed that all the Scottish material is conspecific, since the extreme variants occur in

194

PALAEONTOLOGY, VOLUME 8

the same microsection. Also, several of Maslov's figured specimens (Maslov 1956;
pi. 8, figs. 4, 5) exceed the sizes which he gives in the text of his paper. Apart from the
central stem, the size range of the Scottish specimens almost includes that of C. brunto-
nense Johnson. The diameter of the central stem is, however, very close to that of the
type material of C. okense, and the average size of the fragments is closer to that of C.
okense than C. bruntonense. For this reason, they are assigned to the former species.
It is of interest to note that C. bruntonense occurs abundantly in the Great Limestone

text-fig. 1 . Diagrammatic representation of sections of Calcifolium okense ( a , b, c) and C.
panel at um (d, e, f) approximately x25; a , d, stem cross-sections; b, e, plate tangential

sections; c,f plate cross-sections.

of N. England ( Johnson 1958) and this limestone is usually equated with the Top Hosie
Limestone of the Scottish succession — a higher horizon than the Blackhall Limestone.
It may be that records of Calcifolium with a single series of canals from lower limestones
in N. England (Short 1954) in fact indicate the presence of the smaller form C. okense.

The probable form of the thallus of C. okense is discussed in some detail by Maslov
(op. cit.). The Scottish specimens are generally too broken up to give any further
information. The structure must have been of some rigidity, for several plates have
attached foraminifera, usually growing on the side farthest from the perforations.

Calcifolium punctatum Maslov
Plate 22, figs. 1-5; text -fig. 1 d-f.

Calcifolium okense Shvetzov and Birina 1935 pars.

Calcifolium punctatum Maslov 1956, plate 8, fig. 2; plate 9, fig. 1; plate 10, fig. 2; text-fig. 8.

EXPLANATION OF PLATE 21

Figs. 1-5. Calcifolium okense x28. 1. Cross-section of complex branch; Hawthorn Lst., Bankend
quarry (text-fig. 2, locality 13). (Slide no. PS2998), Geological Survey Collection, Edinburgh.)

2. Cross-section of stem, with lateral plates; Hawthorn (Wee) Lst., River Nethan (loc. 11). (PS2996.)

3, 4. Cross- and tangential-sections of lateral branches; Blackhall Lst., Avon Water (loc. 6). (PS3001.)
5. Cross-section of plate, with attached foraminifer (top right); Blackhall Lst., Avon Water (loc. 7).

(PS3002.)

Palaeontology, Vol. 8

PLATE 21

BURGESS, Calcifolium

I. C. BURGESS: CALCIFOLIUM FROM THE VISfiAN OF SCOTLAND 195

Diagnosis. Calcifolium, in which the canal system takes the form of a network which may
completely fill the thickness of the stem wall and the lateral plates. Dimensions are as
in Table 1.

text-fig. 2. Sketch-map of the western part of the Midland Valley of Scotland, showing areal distri-
bution of Upper Visean strata and fossil localities: 1, Corrieburn; 2, Baldow Glen; 3, Bridge of Weir;
4, Thorntonhall; 5, Calderwood Glen; 6, 7, Avon Water; 8, Birkwood Burn; 9, 10, 11, 12, River
Nethan; 13, Bankend; 14, Ponfeigh Burn; 15, Craig Burn; 16, Glenbuck; 17, Limefield; 18, Wedder
Law; 19, Kennox Water; 20, Blairmill; 21, Ardrossan; 22, Nettlehirst; 23, Hessilhead; 24, Broadstone;

25, Roebank Glen.

Description. The stem wall is pierced by a network of longitudinal canals, which in larger
specimens is confined to the inner half of the wall, but which in smaller specimens fills
the wall completely. Within the lateral plates the canals diverge radially, and branch
dichotomously at an acute angle, forming a network throughout the thickness of the
plate. In transverse sections, the plates appear as dark, often ragged strips, with the

196

PALAEONTOLOGY, VOLUME 8

canals arranged in two or three rows, or distributed irregularly throughout the thickness
of the plate. Repeated branching of the plates frequently takes place, forming complex
structures.

Remarks. The specimens from the Hurlet Limestone closely resemble those described by
Maslov in size and form (Table 1). The thallus generally appears to be more irregular
in shape than that of C. okense. The canals also lack the regularity of those in C. okense,
and in some specimens give the appearance of forming an anastomosis within the lateral
plates. In one specimen (Plate 22, fig. 1) C. punctatum grew attached to a corallite of
Lithostrotiom junceum, a feature never observed in C. okense.

STRATIGRAPHICAL DISTRIBUTION

The Upper Visean strata are a series of alternating marine and non-marine beds,
which rest unconformably and with internal overlap on Clyde Plateau Lavas and older
rocks. Their stratigraphy has in the past been the subject of many investigations (sum-
marized by Macgregor 1930) due partly to their former economic importance and partly
to the wealth of fossils which are found in many of the limestones.

The stratigraphical sequences in North Ayrshire and South Lanarkshire (text-fig. 3)
are comparable both in lithology and thickness. The limestones are generally thick, and
yield a prolific macrofauna. In the Central Coalfield basin, the sequence is much thicker,
but individual limestones tend to be thinner and less fossiliferous. The change in facies
coincides closely with the gap in exposures caused by the ridge of older rocks separating
the Central Coalfield from the basins to the south (text-fig. 2). Thus while the correlation
of the marine horizons between North Ayrshire and South Lanarkshire is well established
(Richey 1946) that between these areas and the Central Coalfield is more difficult, and
has in the past been a controversial subject (Carruthers and Richey 1915; Macnair
1915).

In the Central Coalfield basin, Calcifolium punctatum occurs abundantly in the Hurlet
Limestone in most exposures, except those around Paisley and Howwood, in which
district the limestone is crinoidal. In the region between Lesmahagow and Carluke, it is
commonly found in association with Saccamminopsis fusulinaformis (M’Coy). In the
Coalburn-Douglas district, C. punctatum and S. fusulinaformis are again found together
in the Productus giganteus Limestone. Traced south-westwards into the Muirkirk syn-
cline, this limestone becomes first shaly, then sandy. C. punctatum is not found there,
being confined to those exposures where the lithology of the limestone is biomicrite or

EXPLANATION OF PLATE 22

Figs. 1-5. Calcifolium punctatum X 28. 1. Cross-section of alga, initially encircling a corallite of Litho-
strotion junceum then giving rise to a complex branch; Productus giganteus (Big) Lst., River Nethan
(text-fig. 2, locality 12). (Slide no. PS2995, Geological Survey Collection, Edinburgh.)

2. Tangential section of lateral plate with canals infilled with pyrites; Productus giganteus Lst., Coal
Burn, Glenbuck (loc. 16). (PS2994.)

3. Cross-section of lateral branch; Broadstone Lst., Broadstone quarry (loc. 24). (PS2999.)

4. Cross-section of stem, with remnant of lateral plate (top left); Productus giganteus (Big) Lst.,
River Nethan (loc. 9). (PS2997.)

5. Cross-section of part of stem (bottom left) with longitudinal sections of lateral branches; Broad-
stone Lst., Nettlehirst quarry (loc. 22). (PS3000.)

Palaeontology , Vol. 8

PLATE 22

BURGESS, Calcifolium

I. C. BURGESS: CALCIFOLIUM FROM THE VISEAN OF SCOTLAND 197

biomicrosparite. The Broadstone Limestone of North Ayrshire also contains abundant
C. punctatum at many localities.

Calcifolium okense is found over a more restricted area. In the Central Coalfield basin,
it occurs in a 1-foot band just above the base of the Blackhall Limestone, in the region
around Strathaven and Birkwood, frequently in association with a small encrusting form
of Girvanella , with tubes about 3-4 ^ in internal diameter. Elsewhere, the lower part of
the limestone is shaly, or isasiltyostracod limestone. Calcifolium has not been recorded in

CENTRAL

COALFIELD

text-fig. 3. Generalized stratigraphical sequences in the Central Coalfield, North Ayrshire and South
Lanarkshire, approximately to scale, showing main limestones and proposed correlations.

these facies. In the Coalburn-Douglas district C. okense occurs near the base of the Haw-
thorn Limestone at Bankend (again in association with small Girvanella) and in a 1-foot
band above the base of the equivalent ‘Wee’ Limestone at several localities in the River
Nethan. In the Muirkirk syncline the lower part of the Hawthorn is shaly in the Glen-
buck region and becomes sandy to the south-west. C. okense has not been found at any
exposures in this area. In North Ayrshire, the equivalent Dockra Limestone (Richey
1946) has been sampled at many exposures, but Calcifolium has not been found.

None of the other limestones in the west of Scotland has been proved to contain
Calcifolium. This probably reflects more an absence of the specific environmental con-
ditions required by these algae than a genuine stratigraphical control on their distribu-
tion, as elsewhere the two species have a much wider range (Short, 1954). Within this
region, however, each species is confined to its own horizon, and on this basis the corre-
lations shown in text-fig. 3 are proposed. Both limestone horizons are of P2 age (Currie
1954). In the north of England, the genus was first recorded by Johnson, as ‘Organism a’,

198

PALAEONTOLOGY, VOLUME 8

in 1949 (Johnson 1958), from the Great Limestone (£)) (which material he later assigned
to a new species, C. bruntonense Johnson 1958). Calcifolium punctatum has not been
recorded as such from the North of England. However, Short in an unpublished thesis
(1954) records this species as ‘ Algae /S', from the Yoredales of the Alston Block. He
gives the range of ‘Algae /T as Jew Limestone to Five Yard Limestone, with a maximum
in the Tynebottom Limestone. He also records a form with a single row of punctations
which he assigns to ‘ Algae a’ Johnson, ranging from Five Yard to Great Limestone.

C. punctatum thus ranges from high P1 or basal P2 (there is still some disagreement as
to the base of P2 in the Yoredales) up into P2, to be replaced by C. okense and C.
bruntonense which extend up into basal E±.

In Russia (Maslov 1956), the genus Calcifolium is common in the upper part of the
Okian series ( P2 ) and extends into the base of the Serpukhovian series (Ex, Ramsbottom
1957), only C. punctatum being present in the lowest beds and only C. okense in the
Serpukhovian.

Acknowledgements. The author’s thanks are due to Dr. W. H. C. Ramsbottom for criticism of the paper
in manuscript, and especially to Dr. K. C. Short, for permission to quote unpublished information
from his thesis. The research was carried out at the Department of Geology, University of Glasgow,
under the supervision of Professor T. Neville George, while the author was in receipt of a Research
Studentship from the Department of Scientific and Industrial Research.

REFERENCES

carruthers, r. g. and richey, J. E. 1915. The Lower Limestones of Renfrewshire and North Ayrshire:
a comparison and correlation. Trans, geol. Soc. Glasg. 15, 249-64.
currie, e. d. 1954. Scottish Carboniferous goniatites. Trans, roy. Soc. Edinb. 62, 527-99.
johnson, G. A. l. 1958. Biostromes in the Namurian Great Limestone of Northern England. Palaeon-
tology, 1, 147-57.

macgregor, m. 1930. Scottish Carboniferous stratigraphy: an introduction to the study of the
Carboniferous rocks of Scotland. Trans, geol. Soc. Glasg. 18, 442-558.
macnair, p. 1915. The Hurlet sequence in North Ayrshire. Trans, geol. Soc. Glasg. 15, 200-48.
maslov, b. p. 1956. Calcareous algae of the U.S.S.R. Trans, inst. geol. Science, 160, 1-301.
ramsbottom, w. h. c. 1957. Namurian goniatites in the U.S.S.R. Geol. Mag., 94, 305-11.
richey, j. e. 1946. The Lower Limestone Group near Drumclog, Lanarkshire. Trans, geol. Soc.
Glasg., 21, 49-60.

shvetzov, M. s. and birina, l. m., 1935. On the petrology and origin of the Oka Limestones. Trans.
Moscow Geol. Trust, 10.

short, k. c. 1954. The Geology of the Pennine Escarpment from Croglinwater to Ardale. Ph.D. thesis,
University of Nottingham.

IAIN c. burgess
Geological Survey Office,
Ring Road Halton, Leeds 15

Manuscript received 23 April 1964

THE PALAEONTOLOGICAL ASSOCIATION

COUNCIL 1964-5

President

Dr. L. R. Cox, British Museum (Natural History), London
Vice-Presidents

Professor T. Neville George, The University, Glasgow, W. 2
Dr. W. S McKerrow, University Museum, Oxford
Professor F. H. T. Rhodes, University College, Swansea

Treasurer

Dr. T. D. Ford, Department of Geology, The University, Leicester
Assistant Treasurer

Dr. C. Downie, Department of Geology, The University, Sheffield

Secretary

Dr. C. H. Holland, Department of Geology, Bedford College, London, N.W. 1

Editors

Mr. N. F. Hughes, Sedgwick Museum, Cambridge
Dr. Gwyn Thomas, Department of Geology, Imperial College of Science, London, S.W. 7
Dr. I. Strachan, Department of Geology, The University, Birmingham, 15
Dr. M. R. House, University Museum, Oxford

Other members of Council

Dr. C. G. Adams, British Museum (Natural History), London
Dr. F. M. Broadhurst, The University, Manchester
Professor O. M. B. Bulman, Sedgwick Museum, Cambridge
Professor P. M. Butler, Royal Holloway College, Surrey
Dr. W. J. Clarke, British Petroleum Company, Sunbury on Thames
Dr. G. Y. Craig, The University, Edinburgh
Dr. W. T. Dean, British Museum (Natural History), London
Dr. F. A. Middlemiss, Queen Mary College, London
Mr. M. Mitchell, Geological Survey and Museum, London
Dr. A. J. Rowell, The University, Nottingham
Dr. M. J. S. Rudwick, Sedgwick Museum, Cambridge
Dr. R. J. G. Savage, The University, Bristol
Professor Scott Simpson, The University, Exeter

Overseas Representatives

Australia-. Professor Dorothy Hill, Department of Geology, University of Queensland, Brisbane
Canada-. Dr. D. J. McLaren, Geological Survey of Canada, Department of Mines and Technical
Surveys, Ottawa

India: Professor M. R. Sahni, Department of Geology, Panjab University, Chandigarh

New Zealand : Dr. C. A. Fleming, New Zealand Geological Survey, P.O. Box 368, Lower Hutt

West Indies and Central America: Dr. L. J. Chubb, Geological Survey Department, Kingston, Jamaica

Eastern U.S.A. : Professor H. B. Whittington, Museum of Comparative Zoology, Harvard Univer-
sity, Cambridge 38, Mass.

Western U.S.A. : Professor J. Wyatt Durham, Department of Paleontology, University of California,
Berkeley 4, Calif.

PALAEONTOLOGY

VOLUME 8 • PART 1

CONTENTS

The trilobite genus Oedicybele from the Kildare Limestone (Upper Ordovician)
of Eire. By j. t. temple 1

Homotrypa and Amplexopora? from the Caradoc Series, Shropshire. By
J. R. P. ROSS 5

Quantoxocrinus, a new Devonian inadunate crinoid from west Somerset.

By B. D. WEBBY 1 1

Neogene Tasmanites and leiospheres from southern Louisiana, U.S.A. By
C. J. FELIX 16

Foraminifera in Holocene marsh cycles at Borth, Cardiganshire (Wales).

By T. D. adams and j. haynes 27

On a new species of Hoegisporis Cookson. By i. c. cookson 39

The development of a dicellograptid from the Balclatchie Shales of Laggan Bum.

By J. james 41

A Lower Carboniferous fauna from Trevallyn, New South Wales. By j. Roberts 54

Westphalian D megaspores from the Forest of Dean Coalfield, England. By
e. spinner 82

On the genus Pothocites Paterson. By m. chaphekar 107

Time in stratigraphy. By t. g. miller 113

Environmental causes of stunting in living and fossil marine benthonic inverte-
brates. By a. hallam 132

The mode of life of two Jurassic species of ‘Posidonic? (Bivalvia). By r. p. s.
JEFFERIES and P. MINTON 156

The palaeoecology of the Goniatite Bed at Cowlow Nick, Castleton, Derby-
shire. By t. D. ford 186

Calcifolium (Codiaceae) from the Upper Visean of Scotland. By i. c. burgess 192

PRINTED IN GREAT BRITAIN AT THE UNIVERSITY PRESS, OXFORD
BY VIVIAN RIDLER, PRINTER TO THE UNIVERSITY

VOLUME 8 • PART 2

Palaeontology

JULY 1965

PUBLISHED BY THE
PALAEONTOLOGICAL ASSOCIATION
LONDON

Price £3

THE PALAEONTOLOGICAL ASSOCIATION

The Association was founded in 1957 to further the study of palaeontology. It holds
meetings and demonstrations, and publishes the quarterly journal Palaeontology.
Membership is open to individuals, institutions, libraries, &c., on payment of the
appropriate annual subscription :

Institute membership . . . . £5. 5s. (U.S. $15.50)

Ordinary membership . . . £3. 3s. (U.S. $9.50)

Student membership . . . £2. 2s. (U.S. $6.50)

There is no admission fee. Student members will be regarded as persons receiving
full-time instruction at educational institutions recognized by the Council; on first
applying for membership, they should obtain an application form from the Secretary
or the Treasurer. Subscriptions are due each January, and should be sent to the
Treasurer, Dr. C. Downie, Department of Geology, The University, Mappin Street,
St. George’s Square, Sheffield I, England.

Palaeontology is devoted to the publication of papers (preferably illustrated) on all
aspects of palaeontology and stratigraphical palaeontology. Four parts are published
each year and are sent free to all members of the Association. Members who join for
1965 will receive Volume 8, Parts 1 to 4.

All back numbers are still in print and may be ordered from B. H. Blackwell,
Broad Street, Oxford, England, at the prices shown below (post free):

Vol. 1 (for 1957-8) in 4 parts at £2 or U.S. $6.00 per part.

Vol. 2 (for 1959) in 2 parts at £2 or U.S. $6.00 per part.

Vol. 3 (for 1960) in 4 parts at £2 or U.S. $6.00 per part.

Vol. 4 (for 1961) in 4 parts at £2 or U.S. $6.00 per part.

Vol. 5 (for 1962) in 4 parts at £3 or U.S. $9.00 per part.

Vol. 6 (for 1963) in 4 parts at £3 or U.S. $9.00 per part.

Vol. 7 (for 1964) in 4 parts at £3 or U.S. $9.00 per part.

A complete set, Volumes 1-7, consists of 26 parts and costs £64 or U.S. $192.

Manuscripts on all aspects of palaeontology and stratigraphical palaeontology are
invited. They should conform in style to those already published in this journal, and
should be sent to Mr. N. F. Hughes, Department of Geology, Sedgwick Museum,
Downing Street, Cambridge, England. A sheet of detailed instructions for authors
will be supplied on request.

© The Palaeontological Association, 1965

THE INTERRELATIONSHIPS OF SOME
CRETACEOUS CODIACEAE
(CALCAREOUS ALGAE)

by GRAHAM F. ELLIOTT

Abstract. The anatomy and evolution of fossil codiacean calcareous algae referred to Arabicodium and Boueina
(Jurassic-Cretaceous) and Halimeda (Upper Cretaceous-Recent) are discussed. It is considered that all are
referable to one botanical genus Halimeda ; differentiation of (he earlier species-groups ( Boueina and Arabicodium)
was followed by hybridization and selection in the Upper Cretaceous, leading to Halimeda s.str., a group of
species of more vigorous growth and showing a combination of characters from the older species-groups.

The genus Halimeda (Chlorophyta, family Codiaceae) is one of the most successful of
modern warm-water marine algae. It abounds in clear shallow lagoonal and coastal
waters throughout the tropical and subtropical areas of the Indo-Pacific, and the
Atlantic with its adjacent Caribbean and Mediterranean. Since the segments of inter-
nodes of the Halimeda fronds are usually heavily calcified and growth of the plants is
rapid, it is an important sediment-former, as directly evidenced by borings on atolls
(Hinde 1904, Johnson 1961). Throughout the Tethyan area remains of Halimeda are
common in the appropriate facies of Tertiary limestones, and segments of similar forms
to those of some of the Recent species are known from rocks as old as the Eocene (e.g.
H. eocaenica and H. praemonilis; Morellet and Morellet 1941). The origin of Halimeda
is clearly pre-Tertiary.

SOME CRETACEOUS CODIACEAE

Segmented codiaceae in the Cretaceous are represented by assemblages of dissociated
segments which have been referred to three genera: Arabicodium , Boueina , and Hali-
meda. Although typical segments for each differ in size and shape, and in internal
thread-detail, they all agree in possessing longitudinal medullary threads and a zone of
more or less radial cortical threads, and it seems reasonable to conclude that they came
from plants of similar general morphology to that of the living Halimeda, and of close
relationship to it. This was suggested for Boueina and Halimeda by Steinmann (1899)
and Pia (1926, 1927). Boueina and Arabicodium have Lower Cretaceous type species,
and are represented earlier in the Jurassic. Halimeda does not appear before the Upper
Cretaceous. There are also numerous Cretaceous species from Aptian to Senonian,
which are difficult to assign conclusively to any of the three genera since they show inter-
mediate characters.

Boueina (type species: B. hochstetteri Toula 1884 from the Lower Cretaceous of
Serbia) shows elongate cylindrical segments, circular in cross-section, and not flattened
or branched as are those of most Tertiary and Recent Halimeda. In thin-section they
show a medullary zone of coarse tangled threads (PI. 23, fig. 1), and a cortex of approxi-
mately radial finer branching threads, whereas in Halimeda the central coarse threads
are mostly longitudinally directed, and the cortical threads show constrictions and

[Palaeontology, Vol. 8, Part 2, 1965, pp. 199-203, pi. 23, 24.]

C 3009 P

200

PALAEONTOLOGY, VOLUME 8

swellings (PI. 23, fig. 2). Steinmann (1899), in his comparison of the two, indicated
a frequent triple branching of the swollen cortical threads of Halimeda as compared with
a corresponding double branching of the slimmer Boueina- threads.

Arabicodium (type species : A. aegagrapiloides Elliott 1957, from the Lower Cretaceous
of Arabia) occurs as slim cylindrical segments, occasionally showing incipient terminal
branching. In thin-section they show a medullary zone of fine slightly wavy longitudinal
threads, with cortical zones of very fine radially directed threads (PI. 23, figs. 3, 4).

These two genera, well differentiated, both appear in the Jurassic and extend into the
Lower Cretaceous. B. hochstetteri liasica was described from the Middle Lias of Morocco
(Le Maitre 1937), and recorded from about the same level in Iraqi Kurdistan (Elliott
1960, and PI. 23, fig. 1). In the Lower Cretaceous, besides the type-occurrence, there are
Italian records of the species (Zuffardi-Comerci 1937) and of a variety B. hochstetteri
moncharmonti (de Castro 1960). Arabicodium is recorded from the Middle Jurassic of
Prance (Dufaure 1958), is known from the Upper Jurassic of Borneo, and has been seen
from Tethyan Lower Cretaceous localities other than the type-locality.

In the Upper Cretaceous codiaceans are common: they have been referred variously
to Boueina or Halimeda, but sometimes also resemble the later-described Arabicodium.
Pfender (1940) listed various circum-Mediterranean Upper Cretaceous records and
referred to the difficulty of allocating them to Boueina or Halimeda, though she recorded
the definite occurrence of the latter from the Syrian Turonian. Pia (1936) described
Boueina pygmaea from the Cenomanian-Turonian of Libya: it is a much smaller species
than the type, and without the distinctive medullary threads. It has been recorded from
the Cenomanian of Prance and Spain (Pfender 1940) and the Turonian of the Trucial
coast, Arabia (Elliott 1960). In describing Halimeda nana from the Moroccan Danian
Pia made the generic allocation on the occurrence of a branching segment and not on the
internal structure, which was not well preserved (Pia, Pfender, and Termier 1932).
Records of this species from the Palaeocene of the Middle East (Elliott 1955, 1960) are
based on similarly small segments which, however, show a more markedly halimediform
thread-structure than may be distinguished in the type-figures.

EXPLANATION OF PLATE 23

Fig. 1 . Boueina hochstetteri Toula var. liasica Le Maitre, longitudinal thin-section, X 28 ; coarse irregu-
larly directed medullary threads, cortical threads mostly ill-preserved. Lower Jurassic, Sekhaniyan
Formation (Middle/Upper Lias); Sekhaniyan, Surdash, Sulemania Liwa, NE. Iraq. Brit. Mus.
(Nat. Hist.), Dept. Palaeont., V 51381.

Fig. 2. Halimeda praemonilis Morellet, longitudinal thin-section, x28; coarse longitudinally directed
medullary threads and characteristic cortical layer. Palaeocene, Sinjar Formation; Rowanduz,
Erbil Liwa, NE. Iraq. V 41615.

Figs. 3, 4. Arabicodium aegagrapiloides Elliott, longitudinal thin-sections, X 28. 3, Outline and fine
threads of cortical layer. 4, Distinctive fine longitudinally directed medullary threads. Basal Lower
Cretaceous, Hugf area, Oman, Arabia. V 41625.

Fig. 5. Halimeda sp., random thin-section of damaged segment, x40. Upper Cretaceous, Tanjero
Formation (Maestrichtian); Balambo, Sulemania Liwa, NE. Iraq. V 51380.

EXPLANATION OF PLATE 24

Figs. 1, 3-5. Boueina spp., thin-sections, x40. 1, 5, Upper Cretaceous (Cenomanian), Djebel Meketsi,
Algeria. V 51383, V 51384. 3, 4, Lower Cretaceous (Upper Aptian), Ain Keiriane, Algeria. V 51382.

Figs. 2, 6. Boueina cf. pygmaea Pia, thin-sections, x40. Upper Cretaceous (Cenomanian), Bou Saada,
Algeria. V 51385.

Palaeontology, Vol. 8

PLATE 23

ELLIOTT, Mesozoic and Cretaceous Codiaceae

Palaeontology, Vol. 8

PLATE 24

ELLIOTT, Cretaceous Codiaceae

GRAHAM F. ELLIOTT: SOME CRETACEOUS CODIACEAE

201

Emberger (1961), in fossil-lists for successive Cretaceous stages in Algeria, listed
codiacean remains as Boueina in the Neocomian, and as Boueina-Halimeda for Albian,
Cenomanian, and Senonian. A selection of Algerian codiacean material is now figured,
from Upper Aptian and Cenomanian (PI. 24). All these segments are relatively small,
when compared with those of type Boueina or Arabicodium, or with most Tertiary
Halimeda- segments. Those from the Aptian, all considered referable to one species, are
most similar in medullary threads to Boueina , though not always sharply distinct in this
character from Arabicodium. Cenomanian examples are varied: two species appear to
be represented, with differing medullary thread-structures, typical neither of Arabicodium
nor of Boueina.

In the Middle East the Cretaceous codiacean remains listed by Elliott (1960), which
include Boueina pygmaea, are usually extremely fragmentary. Segments definitely of
Halimeda, however, occur in the Maestrichtian of Iraqi Kurdistan (PL 23, fig. 5).
Andrusov’s record (1939) of a Senonian Halimeda from the Carpathians, although of
a true Halimeda, is now known to be of a Palaeocene occurrence (Misik and Zelman
1959).

MESOZOIC CODIACEAN EVOLUTION

A comparison of these fossil remains with what is known of living algae suggests that
the evolution which occasioned the distribution outlined above may better be expressed
by a taxonomic simplification. In Recent algae, not merely may two species of the same
genus show strikingly different morphology (e.g. Codium), but even the sexual and
asexual forms of the one species may be morphologically very different. Svedelius (1953)
studied this in Galaxaura, and his work was applied by Elliott (1961) to the study of
Cretaceous Per mo calculus, where the fossil remains would not otherwise have been
recognized as all belonging to the same species. Without invoking this degree of close-
ness of relationship, it seems that the Mesozoic codiaceans had by the Jurassic differen-
tiated into two species-groups or variable long-range species, Boueina and Arabicodium,
possibly as a result of geographical isolation. They are very similar in structure and
morphology, and show little evolution through the Jurassic and Lower Cretaceous when
compared with members of the other contemporary chlorophyte family, the Dasy-
cladaceae. Their generic ranking is the result of initial studies of isolated distinctive
assemblages, and has been difficult to apply as more and more material has been
examined. By the beginning of the Upper Cretaceous hybridization occurred, and
a plexus of intermediate forms resulted, typical neither of Boueina, Arabicodium , nor yet
of the later Halimeda. Halimeda itself, appearing but not common at the end of the
period, is a form in which the thickness of the tangled medullary threads of Boueina is
combined with the longitudinal orientation of the thin medullary threads of Arabi-
codium. In the very detailed revision of Recent Halimeda by Hillis (1959), these coarse
medullary threads are seen to be classifiable into several patterns according to the degree
and nature of the fusion between them at the nodal points connecting adjacent segments:
this character is connected with the growth of the new successive segments.

Once evolved, Halimeda rapidly replaced the older segmented codiaceans, which are
not known after the Cretaceous. Records of Palaeocene Boueina (Rao and Vimal 1955,
Elliott 1960) refer to Halimeda sp., and to poorly preserved dasyclads, respectively.
Halimeda seems to have undergone rapid differentiation into some of the species-groups

202

PALAEONTOLOGY, VOLUME 8

represented in the living flora early in the Tertiary, along with an increase in segment-
size, and probably size of the plants also, in many species. The occasional origin of new
and vigorous species by hybridization is a well-known botanical phenomenon; the
measure of the success with which Ha/imeda has fitted its warm-water niche is indicated
by the extent of its contribution as a sediment-builder during the Tertiary and at the
present day (Chapman and Mawson 1906, Johnson 1957).

All the fossil codiaceans discussed here are considered referable to a single genus: of
available names, Halimeda Lamouroux 1812 has priority. Boueina and Arabicodium are
available for use as subgenera for fossils of the appropriate species. Halimeda has already
been used for some Cretaceous species interpreted here as hybrids (e.g. Pfender 1938).

REFERENCES

andrusov, d. 1939. Role des thallophytes dans la constitution des roches sedimentaires des Carpathes
tchechoslovaques. Mem. Soc. Sci. Boheme, ann. 1938, 11.
chapman, f. and mawson, d. 1906. Halimeda- limestones of the New Elebrides. Quart. J. geol. Soc.
Loud. 62, 702-11.

de castro, p. 1963. Nuove osservazioni sul livello ad Orbitolina in Campania (Nota preliminare).
Boll. Soc. Nat. Napoli , 71, 103-35.

dufaure, p. 1958. Contribution a l'etude stratigraphique et micropaleontologique du Jurassique et
du Neocomien, de l’Aquitaine a la Provence. Rev. Micropaleont. 1, 87-115.
elliott, g. F. 1955. Fossil calcareous algae from the Middle East. Micropaleontology, 1, 125-31.

1957. New calcareous algae from the Arabian Peninsula. Ibid. 3, 227-30.

■ — — 1960. Fossil calcareous algal floras of the Middle East, with a note on a Cretaceous problemati-
cum Hensonella cylindrica gen. et sp. nov. Quart. J. geol. Soc. Lond. 115, 217-32.

■ 1961. The sexual organization of Cretaceous Permocalculus (Calcareous Algae). Palaeontology,

4, 82-84.

emberger, J. 1961. Esquisse geologique de la partie orientale des Monts des Oulad Nail (Atlas saharien,
Algerie). Bull. Carte geol. Alger. 27 [dated I960].

hillis, l. w. 1959. A revision of the genus Halimeda (order Siphonales). Inst. Marine Sci. Unix. Texas,
6, 321-403.

hinde, G. J. 1904. Report on the materials from the borings at the Funafuti Atoll. In The Atoll of
Funafuti. Roy. Soc. Lond.

Johnson, j. h. 1957. Geology of Saipan, Mariana Islands: Calcareous Algae. Prof. Pap. U.S. geol.
Surv. 280-E.

1961. Fossil algae from Eniwetok, Funafuti and Kita-Daito-Jima. Ibid. 260-Z.

le maitre, d. 1937. Etudes paleontologiques sur le Lias au Maroc: nouvelles recherches sur les spongio-
morphides et les algues du Lias et de l’Oolithe inferieur. Notes Serv. Min. Maroc, 43.
misik, m. and zelman, j. 1959. Ober die Zugehorigkeit der Algen-Korallenriffe des Hiigellandes von
Myjava (Brezouske Pohorie) zum Palaogen. Geol. Sbornik (Bratislava), 10, 301-8.
morellet, l. and morellet, j. 1941. Etudes sur les algues calcaires de FEocene du Cotentin. Bull. Soc.
geol. Fr. (5) 10, 201-6.

pfender, j. 1940. Les algues du Nummulitique egyptien et des terrains Cretaces-Eocenes de quelques
regions mesogeennes. Bull. Inst. Egypte, 22, 225-50.

• 1938. Etude micrographique des calcaires Cretaces et Eocenes de lTmini, de Tamdakht et de

Skoura. In L. Moret, Contribution a la paleontologie des couches Cretacees et Eocenes du versant
sud de l’Atlas de Marrakech. Notes Serv. Min. Maroc, 49.
pi a, j. 1926. Pflanzen a/s Gesteinsbildner. Berlin.

1927. Thallophyta. In M. Hirmer, Handbuch der Paldobotanik. Munich-Berlin.

— — 1936. Calcareous green algae from the Upper Cretaceous of Tripoli (North Africa). J. Pa/eont.
10, 3-13.

• pfender, j. and termier, h. 1932. Etudes geologiques sur les calcaires de Bekrit et de Timhadit

(Moyen Atlas). Notes Min. Carte Maroc, 20.

GRAHAM F. ELLIOTT: SOME CRETACEOUS CODIACEAE

203

rao, s. r. n. and vimal, k. p. 1955. Fossil algae from Sind, Pakistan. Micropaleontology, 1, 91-92.
steinmann, g. 1899. Ueber Boueina, eine fossile Alge aus der Familie der Codiaceen. Ber. Naturf Ges.
Freiburg i. B. 11, 62-72.

svedelius, n. 1953. Critical studies on some species of Galaxaura from Hawaii. Nova Akta Soc. Sci.
Upsal, (4) 15, (9).

toula, f. 1 884. Geologische Untersuchungen im westlichen Theile des Balkan und in den angrenzenden
Gebieten. X. Von Pirot nach Sofia, auf dem Vitos, liber Pernik nach Trn und liber Stol nach Pirot.
S.B. Akad. Wiss. Wien , 88 (1), 1279-1348.

zuffardi-comerci, r. 1937. Sui generi Chaetetes Fischer, Pseudochaetetes Haug, e Solenopora Dybow-
ski. Boll. Ujf. geol. Ital. 62, (2).

GRAHAM F. ELLIOTT

Iraq Petroleum Co. Ltd.,
33 Cavendish Square,

Manuscript received 13 March 1964 London, W. 1

CORALLUM INCREASE IN LITHO STROTION

by R. K. JULL

Abstract. Serial section examination of increase in ten species of Lithostrotion from Australia and Great Britain
has shown that increase is a valuable specific character, and has confirmed earlier reports that in Lithostrotion it
is mainly lateral; axial increase has already been reported in Lithostrotion from Russia and peripheral increase is
herein described in a specimen from Australia.

Four types of lateral increase in Lithostrotion are recognized: (1) increase in L.junceum (Fleming) in which the
daughter corallite has an early aseptate stage after which septa are inserted independently of the septa of the
parent; (2) increase in fasciculate species with a narrow dissepimentarium, in which the septa of the daughter are
independently inserted; (3) increase in fasciculate species with a wide dissepimentarium, in which the initial
septa of the daughter corallite are inherited from the parent; (4) increase in cerioid species, which is related to
that in fasciculate species with a wide dissepimentarium.

The genus Lithostrotion Fleming comprises a group of colonial corals possessing a re-
markably variable morphology, the variation commonly being expressed as morpho-
logic ‘trends’ within the genus. Much disagreement exists as to the limits of the generic
characters. Although the details of reproduction are quite variable both between and
within species and occasionally even within the same corallum, basic patterns of increase
do exist within the group, and these can be categorized. Lithostrotion lends itself well to
ontogenetic studies since adult corallites of most species have an obvious bilateral
symmetry, marked by prominently developed axial septa. Septal insertion during the
development of new corallites can thus be readily traced.

Reproduction in ten species and one subspecies of Lithostrotion was examined in
detail by the use of closely spaced acetate peel sections. Six of the species are fasciculate
— L. junceum (Fleming), L. arundineum Etheridge, L. arundineum subsp. n., L. cf.
pauciradiale (McCoy), L. stanvellense Etheridge, L. sp. nov. 7, and L. sp. ; one is semi-
fasciculate — L. cf. martini Edwards and Flaime; and three are cerioid — L. cf. portlocki
( Bronn), L. minus (McCoy), and L. columnar e Etheridge. These species, which collectively
represent most of the basic morphologic types of Lithostrotion , were deliberately chosen
for study with a view to establishing the general pattern of reproductive variability
within the group. An appreciation of the minor factors involved in variation was gained
by the examination of two or more examples of increase in all species except L. cf.
pauciradiale.

Examination was made of only hysterocorallite development. Studies of protocorallite
ontogeny have not been made since the preservation of a colony with its protocorallite
still intact is extremely rare.

References to increase in Lithostrotion have been made by many writers, but details
were not discussed until recently when Dobrolyubova (1958) commented on increase in
some species from the Russian Platform. Her observations, which are based on thin-
section examination, are generally borne out by the present study, but differences of
interpretation do exist, these being commented on below.

Matthai (1926) concluded from his study of reproduction in living members of the
scleractinian family ‘Astraeidae’ (Faviida) that an imperfect and often misleading under-

[Palaeontology, Vo], 8, Part 2, 1965, pp. 204-25.]

R. K. JULL: CORALLUM INCREASE IN LITHO STROTIO N

205

standing of coral reproduction is derived from the study of corallite development alone.
As this would probably apply equally well to the Rugosa, the following descriptions
deal only with the mechanics of corallite development. Remarks on polyp formation,
apart from a few conclusions, are left until additional knowledge is gained on colony
formation in the Rugosa.

All the specimens prepared for this study are stored in the Department of Geology
and Mineralogy, University of Queensland.

Acknowledgements. I am very grateful to Professor D. Hill of the University of Queensland for her
interest and advice given during this study. Messrs. M. R. Banks of the University of Tasmania, and
J. S. Jell of the University of Queensland also made many helpful suggestions; information on the
age of a number of the British specimens was kindly given by Mr. J. Selwyn Turner of the University
of Leeds, and Dr. C. D. Waterston of the Royal Scottish Museum.

TERMINOLOGY

Most of the definitions outlined by Hill (1935, 1956) are adopted herein. The following
discussion, as well as elaborating on some of the terms pertinent to reproduction in the
Rugosa, includes a brief resume of the results of the present study.

TYPES OF INCREASE

Three main types of increase, namely axial, peripheral, and lateral increase, occur in
the Rugosa. A few other types are known, but as these do not occur in Lithostrotion ,
they are not discussed below.

Axial increase, a parricidal condition which involves the splitting of the parent corallite
through the axis into two to four daughter corallites, is the least common of the three
main types of increase in the Rugosa. Dobrolyubova (1958) reported three species of
Lithostrotion from the Lower Carboniferous of the Russian Platform which increase by
both axial and lateral means. This is the only report of axial increase in the genus.

Peripheral increase, which is usually, but not invariably, parricidal in nature and in-
volves the incorporation of part of the parent corallite into the morphology of the
daughter corallite, is known to occur in many fasciculate genera and sporadically in
predominantly solitary forms, such as the Devonian Heliophyllum Hall and the Visean
Symplectophyllum Hill. Peripheral increase is known in only a single specimen of
Lithostrotion, this being described below. De Groot (1963) referred to peripheral increase
in some Spanish Lower Carboniferous lithostrotionids and lonsdaleiids but she ap-
parently used the term for lateral increase in cerioid coralla as described in the present
study.

Lateral increase, the commonest type of increase in the Rugosa, is non-parricidal and
involves the formation of one or occasionally two or more daughter corallites in the
peripheral parts of the calice. Four types of lateral increase are recognized below. The
first three listed pertain to fasciculate species, and the last characterizes cerioid species.

(i) Lateral increase in L. junceum (Fleming) auctt. is characterized by the daughter
corallite forming almost entirely external to the normal edge of the parent calice and
having an early aseptate stage (text-fig. 1 : 1 a-m). The Visean Diphyphyllum simplex
(Thomson), a diphymorph of L. junceum, increases by this means (Dobrolyubova, 1958,

206

PALAEONTOLOGY, VOLUME 8

pis. 29-30), and some individuals of Sudetia lateseptata Rozkowska (1960, pp. 35-43)
of Upper Frasnian age from Poland, have a somewhat similar type of increase.

(ii) Lateral increase in species having a narrow dissepimentarium, is characteristic of
those species of Lithostrotion with one or two rows of dissepiments (text-fig. 2: 2a-m).
The new corallite arises almost at the periphery of the calice, and typically does not
inherit septa from the parent. Rozkowska (1960) described a somewhat similar type of
increase in Disphyllum geinitzi Lang and Smith from the Polish Upper Givetian.

(iii) Lateral increase in species having a wide dissepimentarium, is found in those
species of Lithostrotion with three or more rows of dissepiments (text-fig. 5: 1 a-m). In
this type, the parent corallite may or may not insert new septa preliminary to the ap-
pearance of the daughter corallite, and the daughter corallite inherits its initial septa
from the ends of parent septa, and may develop from well within the parent calice. Both
the Ludlovian Weissennelia lindstromi (Smith and Tremberth) from Gotland and the
Lower Carboniferous D2 Corwenia rugosa (McCoy) from Wales possess lateral increase
which may be of this type.

(iv) Lateral increase in cerioid coralla, earlier called intermural increase by Hill (1935,
p. 491), is often similar to lateral increase in fasciculate species with a wide dissepi-
mentarium except that the daughter corallite does not grow laterally free of the parent,
and the wall dividing the daughter from the parent is of a different nature (text-fig. 6:

1 a-m). It is the only type of increase known in cerioid species of Lithostrotion.

It is suggested that the term ‘intermuraT be restricted to increase in which the daughter
appears to develop between corallite walls without any particular parent corallite.
Dobrolyubova (1958) reported this type of increase in Lonsdaleia rossica rossica Stucken-
berg of Upper Visean to Lower Namurian range on the Russian Platform, and I have
observed it in a partially diphymorphic corallum of L. floriformis floriformis (Martin)
from Upper Visean D2 beds at the Wrekin, Shropshire, England.

DEVELOPMENTAL STAGES

In the present study, the septa in hysterocorallites which are equivalent to the proto-
septa of protocorallites are referred to as primary septa. Only four septa are regarded as
primary septa in Lithostrotion , these being the counter and cardinal septa (axial septa)
and the alar septa. Counter-lateral septa are treated as metasepta (septa formed after the
primary septa) since they are neither distinct in character, nor do they particularly
precede metaseptal development.

The hystero-ontogenetic terms of Smith and Ryder (1926), modified by Smith (1945),
are applied with some modification in the present study. In L. junceum and fasciculate
species of Lithostrotion having a narrow dissepimentarium, the hystero-brephic stage is
characterized by independently inserted primary septa, and is followed by or overlaps
the neanic stage, in which metasepta commence to appear, often before all four primary
septa are inserted. In cerioid species and in fasciculate species having a wide dissepi-
mentarium, at least in the examples studied, the hystero-brephic stage occurs when the
daughter is intimately associated with the parent, and since this condition persists for
some time after the appearance of the metasepta, the following stage is referred to as
hystero-neanic.

The late neanic stage commences when the daughter achieves a bilateral symmetry
and assumes a form typical of the ephebic stage except for a smaller corallite size and

R. K. JULL: CORALLUM INCREASE IN L1THO STROTIO N

207

fewer septa and dissepiments. The rate of corallite development decreases at this time.
In lateral increase in fasciculate species, the late neanic stage is generally initiated before
the daughter becomes discontiguous from the parent corallite.

WALL

In lateral increase, contiguous daughter and parent corallites share a common outer
wall, and are separated from each other by a wall which is formed in various ways. In
cerioid species, this separating wall is identical with the normal wall separating adult
corallites. It is secreted by both polyps and often consists of three obvious layers, a median
thin, dark layer bounded on either side by fibrous layers, the nature of these fibres being
similar to that of the septal fibres.

In fasciculate species of Lithostrotion , the wall separating daughter from parent,
henceforth referred to as the partition, differs from the equivalent wall in cerioid species.
It is formed in the region of increase by the dilatation of the ends of both parent and
daughter septa forming a thick, generally solid, fibrous wall in which the median dark
layer is absent. The partition is confined to the region of attachment of the daughter to
the parent. When the daughter becomes discontiguous from the parent, a normal outer
wall is developed entirely around each corallite.

Partitions do not invariably separate daughter from parent corallites in fasciculate
species. Lithostrotion cf. martini from the Lower Carboniferous of the British Isles forms
a wall analogous to that which divides daughter from parent in cerioid species. Roz-
kowska (1960, text-figs. 5-7) figured a similar three-layered dividing wall in Disphyllum
geinitzi Lang and Smith from the Upper Givetian of Poland, but she has subsequently
stated {in litteris) that the wall is actually two-layered, with the thin dark inner zone
being a bituminous residium rather than a distinct wall layer.

DESCRIPTIONS OF INCREASE IN LITHOSTROTION
LATERAL INCREASE

Daughter corallites generally appear one at a time at the periphery of the parent and
in no preferred position relative to the axial plane of the parent. Occasionally two or
rarely three corallites may appear on the parent, either simultaneously or shortly after
one another. Pairs of new corallites typically lie adjacent to each other (text-fig. 2:
2 a-m).

Lateral increase always occurs peripheral to the tabularium in the parent corallite.
Since the partition (or wall) dividing the daughter from the parent may be formed as far
into the calice as the axial edge of the dissepimentarium, daughter corallites arising from
parents with a wide dissepimentarium may be positioned well within the parent calice.
In fasciculate species, they later grow laterally discontiguous from the parent. In species
with a narrow dissepimentarium, the partition forms near the edge of the parent calice,
so that in most of its development the daughter corallite overhangs the edge of the
calice. In some cases, before the appearance of the daughter corallite, the dissepiments of
the parent are locally enlarged, so that the partition, when it forms, is positioned further
within the calice than the normal edge of the dissepimentarium of the parent. The effect
is thus that the daughter corallite appears to lie partly within the tabularium. This effect

208

PALAEONTOLOGY, VOLUME 8

may account for Dobrolyubova’s opinion (1958, p. 152) that increase in L. rossicum
(Stuckenberg) was ‘sometimes only in the zone of the dissepimentarium’.

Septal insertion is in the four positions typical of the Rugosa (text-fig. 3). In the early
stages, insertion in the cardinal quadrants is accelerated. Septal insertion is commonly
somewhat irregular, especially during hystero-brephic and early neanic development.
The axial plane of the daughter corallite is radially disposed with respect to the axis
of the parent, with the counter septum attached to the partition or, less commonly,
to the wall immediately adjacent to the partition, and the cardinal septum attached
to the wall opposite the partition. Dissepiments of the parent are not carried forward
into the daughter corallite.

I have examined lateral increase in a number of species of Aphrophyllum Smith, from
Visean beds in Queensland and New South Wales, and find that the above remarks also
apply to this genus.

Type (i): Lateral increase in L. junceum ( Fleming ) auctt.

Text-fig. 1 : 1 a-m, 2-5; text-fig. 4: 2

Three specimens from Great Britain were examined for increase. Their age and
localities are: (a) Upper Dj or D2 zone, west of Black Rigg Quarry, near Settle, York-
shire; ( b ) D2-D3 zones, near Ticknall, Derbyshire; (c) dL4 zone (MacGregor, 1960 =
Coral Zone 3), near Barness, Edinburgh. Dobrolyubova (1958), pp. 145, 183-6) re-
ported a similar type of increase in L. junceum junceum (Fleming) and Diphyphyllum
simplex (Thomson) from the Lower Carboniferous of the Russian Platform. The latter
species is a diphymorph of L. junceum.

Lithostrotion junceum is one of the most distinctive species of the genus. Besides
lacking dissepiments and possessing distinctive types of increase (both this type and
axial increase which is later discussed), it often lacks minor septa, corallite diameter is
very small, and corallites often throw out tubules connective to their neighbours. The
specimen from North Wales figured by Easton (1957, pi. 71, figs. 5, 6) as L. junceum, is
probably L. pauciradiale.

Hystero-brephic stage (text-fig. 1; 1 c-f, 3). The daughter corallite is developed al-
most entirely external to the normal circumference of the parent calice, this being
achieved by a local outswelling of the wall of the parent. The partition is very thick and
is formed by dilatation of the peripheral ends of four, or more typically five major and
minor septa at the position originally occupied by the wall of the parent corallite. Al-
though minor septa are absent in many adult corallites, they are apparently invariably

text-fig. 1. Figs, la-5, x8 approx.; 6-8, x4 approx. Numbers below figs. 1 c-m are cumulative
distances of distal growth in mm. C = cardinal septum; K = counter septum; A = alar septum.
Figs. 1 ci-m, 2-5. Lateral increase in L. junceum (Fleming). 1 a-m, Hystero-ontogenetic sequence,
figs. 1 a-b are hypothetical drawings based upon other examples of increase in the specimen, F2636,
Upper Dx or D2 zone, west of Black Rigg Quarry, near Settle, Yorkshire. 2, 3, Nearly aseptate
daughter corallites and primary septal swellings, F2636/2, 10, same corallum as above. 4, Elongated
cardinal septum, F30206/12, dL4 zone (= Coral Zone 3), Barness, near Edinburgh, Scotland. 5,
Elongated counter septum, F29972/10, D2-D3 zone, Ticknall, Derbyshire.

Figs. 6-8. Lateral increase in L. arundineum subsp. n. 6, Primary septa, F28888a/ 15, Visean, Baywulla
Formation, Yarrol Basin, near Monto, Queensland. 7, Primary septa, F30431a/3, same locality.
8, Fossulae in the later part of the early neanic stage, F30431a/17, same locality.

TEXT-FIG. 1.

210

PALAEONTOLOGY, VOLUME 8

present in the zone of increase at the time of formation of the partition. Commonly these
minor septa are suppressed after the partition is formed.

Shortly after the partition is formed, small swellings commonly appear on the wall of
the daughter corallite in the positions of the axial septa and, less commonly, alar
septa (text-fig. 1 : 3), but the daughter may then grow distally as much as 1 mm. before
typical septa are developed. The order of insertion is variable, but most often the cardi-
nal septum is the first to appear. The counter septum is commonly developed after
metasepta are inserted in the cardinal quadrants. Alar septa may or may not precede
the metasepta.

Neanic stage (text-fig. 1 : 1 g-k, 2, 4, 5). This stage overlaps the hystero-brephic stage if
counter or alar septa are slow in developing. Metasepta develop rapidly, appearing to
emerge from the wall. Insertion is commonly accelerated in the cardinal quadrants.
Axial septa may or may not be lengthened more than the metasepta and when such
lengthening occurs, it may be of either the cardinal septum (text-fig. 1 : 4) or the counter
septum (text-fig. 1: 5), but not of both simultaneously. Dobrolyubova (1958, p. 145,
pi. 18, figs. 3b, 3/3) reported a similar situation in her specimens. Alar septa may be
lengthened to the axial region but metasepta typically are not. Minor septa, apart from
occasional small swellings on the wall, are absent in daughter corallites which are
contiguous with their parents, whether or not the parent possesses them. Columellae
are absent or marked by only a slight dilatation of the axial septa if they are united.

In the Yorkshire specimen, the daughter corallites become discontiguous from the
parent some 2 to 2-5 mm. distally from the point where increase commenced. The
average diameter of the daughter corallite at its point of discontiguity is l-4x 1-6 mm.,
as compared with an average adult corallite diameter of 2-2 mm. The number of major
septa is fourteen, the same as in most adult corallites, and septal spacing is thus closer in
young corallites. In one instance, two young corallites become discontiguous from the
same parent before septa had emerged beyond a few short projections (fig. 2).

The relation of the diameter and septal number of daughter corallites to those of their
adult corallites in the other two specimens is similar to that in the Yorkshire specimen.

After it has become discontiguous, the young corallite gradually grows to the adult
stage (text-fig. 1 : 1/-/7?), and a normal columella may develop, depending upon whether
or not the corallum has diphymorphic tendencies.

An interesting analogy is found in Sudetia lateseptata Rozkowska from the Polish
Upper Devonian. Rozkowska (1960, pp. 36-40) found that new corallites in this species
are mainly aseptate during the early part of increase, and also that adult corallites throw
out connective processes similar to those in L. junceum. Rozkowska referred to this type
of increase in Sudetia as ‘syringoporoid’, and considered that similar increase occurs in
Peneckiel/a jevlanensis Bulvanker and Donia russiensis Soshkina, both Upper Frasnian
corals from the Russian Platform. Similarities between L. junceum and Sudetia may

text-fig. 2. All figures X4 approx. Numbers below figs. 2 a-m are cumulative distances of distal
growth in mm. C = cardinal septum; K = counter septum; A = alar septum.

Figs. 1 , 2a-m. Lateral increase in species with a narrow dissepimentarium. 1 , Diphymorphic corallites
of L. arundineum Etheridge, F44268/20, Visean, Baywulla Formation, Yarrol Basin, near Monto,
Queensland. 2 a-m, Flystero-ontogenetic sequence in L. arundineum subsp. n., F28888b, from the
same locality.

TEXT-FIG. 2.

212

PALAEONTOLOGY, VOLUME 8

point to a relationship between the two. It is quite possible that L.junceum has a different
ancestor from that of the other lithostrotionids.

Type (ii): Lateral increase in species with a narrow dissepimentarium
Text-fig. 1: 6-8; text-fig. 2: 2 a-m

This type of increase is typical of those species of Lithostrotion examined which possess
one or two rows of dissepiments. Specimens examined in detail were of L. arundineum
Etheridge, and L. arundineum subsp. n., both of Visean age from the Baywulla Forma-
tion in the central-western part of the Yarrol Basin, near Monto, Queensland, and L. cf.
pauciradiale (McCoy) of S2 age from the Scaleber Quarry, near Settle, Yorkshire. This
last-named species is conspecific with ‘L. cf. martini ’ figured by Hudson (1930, pi. 1,
fig. 2) from the same locality, and differs from L. pauciradiale ( sensu Hill, 1 940, pp. 1 69-70,
text-fig. C) by having a slightly larger corallite diameter and septal number, longer
septa (not a diagnostic feature), and tabulae which are more strongly arched up to the
columella.

Preliminary to the appearance of the daughter corallite, the region of increase in the
parent corallite is modified in the following ways: septa may become wavy and dis-
continuous from the wall, some minor septa may be suppressed, often sclerenchyme is
deposited, and in some corallites, the dissepimentarium is widened a short distance into
the tabularium.

Hystero-brephic stage (text-fig. 1 : 6-7; text-fig. 2: 2 b-d). The partition is formed in a
similar manner to that observed in L.junceum, but it differs in being thinner, and in some
cases, not solidly formed during the period of contiguous development of the daughter
corallite. Sclerenchyme usually lines the outer wall, and to a lesser extent, the partition
of the daughter corallite; occasionally it is so abundant during the early stages of in-
crease that it masks the initial insertion of septa.

Septa are inserted almost immediately after the formation of the partition, or even
before the septa of the parent are withdrawn from the region of increase. Primary septa
commonly appear almost simultaneously (text-fig. 1 : 6-7), but often their insertion is
irregular and they may appear in almost any order and after the insertion of metasepta
(text-fig. 2: 2 c-e). In one example of increase in L. arundineum, axial septa appear to
have formed from the prolonged end of a parent septum. Dobrolyubova (1958, pp. 1 52-5)
concluded from her examination of increase in L. rossicum Stuckenberg and L.
volkovae Dobrolyubova, both of which species have one or two rows of dissepiments,
that the daughter inherited septa from the peripheral ends of parent septa. One of her
figures (1958, text-fig. 27) seems to indicate this, but possibly the study of increase in
these species by use of closely spaced serial sections might show that septa are in-
dependently inserted. If her conclusions are correct, however, independent insertion of
all daughter septa cannot be taken as a consistent feature of this type of increase.

Neanic stage (text-fig. 1:8; text-fig. 2: 1 e-m). This stage commences with the ap-
pearance of metasepta, and continues beyond the point of discontiguity of the daughter
from the parent corallite. If one or more of the primary septa are inserted late, the
hystero-brephic stage overlaps this stage.

As a result of asymmetrical development of the daughter corallite in the early part of
this stage, septal insertion is generally not equal between cardinal quadrants or counter
quadrants. Eventually, corresponding quadrants are of equal size and septal number.

R. K. JULL: CORALLUM INCREASE IN LITHO STROTIO N

213

Septal insertion differs in cardinal and counter quadrants, particularly during the
early neanic stage. This is illustrated in text-fig. 3, a graphic representation of septal
insertion in the hystero-ontogenetic sequence figured in text-fig. 2: 2 a-m. It is typical of
insertion in daughter corallites arising by this type of lateral increase. Since serial sections
were taken at regularly spaced intervals, the order of insertion is in doubt for those septa

Pig.

Distance

Cardinal Quadrants

Counter Quadrants

No.

(mm. )

Left

Right

Left

Right

2a

0

2b

.05

.10

2c

.15

1

2d

.20

2 1

121

2e

.25

i

3 2 1

1 2

21 3

2f

.30

12 3

3 2 1

1 2 3

2 14 3

2g

.40

111

111

I 13

2 14 1

2b

.50

1 4 2 5 1 6

4 111

14 2 3

2 14 1 5

.60

1 4 2 5 ] 6

ill!

14 2 3

2 14 1 5

.70

1 4 2 5 1 6

i i l i

7 6 5 1 4 2 3

2 14 1 5 6

2i

.80

1 4 2 5 1 6

i 1 2 1

76581423

2 14 1 5 6

2j

.90

14253.768

84716251

7 6 5 8 1 / 2 3

2 14 17 5 6

1.00

14251768

84716251

9 76581 23

2 14 17 5 6

1.10

9 Z/5581 23

21417586

1.20

9 I 5 8 1 2 3

21417586

1.30

10 2. I 581 23

21417586 9

1.40

10 9 7 5.81 23

21417586 9

1.50

1.60

1.70

10 9 7 581 23

21417586 9

1.80

10 2. 11 7 581 23

21417586 10 9

1.90

10 2. 11 76581423

21417586 10 9

2.00

10 9 11 76581423

2 14 17586 10 9 11

2.10

10 9 11 16581423

21417586 10 2. 11

21

2.20

14251768

8i7l6251

10 9 11 76581423

21417586 10 9 11

Point of

discontiguity

text-fig. 3. Septal insertion in L. arundineum subsp. n., figured in text-fig. 2 : 2a-m. Septa are numbered
according to their order of insertion. Underlined numbers (e.g. 1) — major septa; numbers not
underlined = minor septa or initially short major septa; stroked numbers (e.g. 6) = suppressed septa;
C, K = point of insertion of cardinal and counter septa; number 1 septum of cardinal quadrants

alar septa.

which are represented in this text-figure as appearing simultaneously. Thus of the septa
numbered 4, 5, and 6 which appear at the 0-50 mm. level in the left cardinal quadrant,
septum number 6, the major septum, should theoretically have been the first inserted of
the three, but this could not be proven. For consistency, these three septa, as well as
other such cases of apparently simultaneous insertion, are numbered progressively
towards the fossula.

In the cardinal quadrants of the species examined, septal insertion is at first slightly
accelerated, and is of the zaphrentid mode with three or four major septa appearing
consecutively before any minor septa are inserted. When minor septa appear in these
quadrants, they do so two or three at a time between the major septa and remain short,
buried in the sclerenchyme which lines the wall. Septal activity then becomes very slow

214

PALAEONTOLOGY, VOLUME 8

and major and minor septa subsequently appear alternatively, as is the case with septal
insertion in the counter quadrants.

In the counter quadrants, positioned on the attached side of the daughter corallite,
insertion is typically in the cyathaxonid mode, with major septa alternating in order of
development with minor septa. A short septum which is inserted in the alar fossula
subsequently lengthens into a major septum, and during or following this lengthening,
two other short septa are inserted on either side of it. The one on the counter side, out-
side the fossula, is generally the first to appear, and remains short as a minor septum.
The short septum on the cardinal side, lying in the alar fossula, lengthens into another
major septum which in turn has septa inserted on either side, with the cycle then re-
peating. Coope (1956) described a similar order of insertion in the later growth stages of
Clisiophyllum and Aulophyllum.

This scheme of septal insertion, which I have also found to occur in AphrophyUum
hallense Smith, a cerioid species in Visean beds in New South Wales, commonly varies
in minor details. A particularly noteworthy variation is the quite common suppression
of newly inserted minor septa, or rarely major septa, for varying lengths of time in the
counter quadrants.

With continuing development, the septal number in the counter quadrants surpasses
that in the cardinal quadrants until in the ephebic stage about two-thirds of the total
number of septa are present in the counter quadrants. Cardinal and alar fossulae com-
monly are marked by one or more short, contratingent septa which are newly inserted
(text-fig. 1 : 8). As septa lengthen, they cease to be contratingent, but if septal insertion is
fairly rapid, newly formed contratingent septa maintain the fossular opening. In the late
neanic stage, fossulae are generally not obvious. Cardinal fossulae may later reappear in
the ephebic stage, this time marked by a shortened cardinal septum.

In non-diphymorphic coralla, the counter and cardinal septa are united from the time
of their development; similarly most of the other major septa are united at the axis soon
after they are inserted. Conjoined axial septa may or may not be dilatated to form
a columella in the early neanic stage. When a columella is first formed, it is irregular in
shape and does not assume the typical lithostrotionid form until just before the corallite
becomes discontiguous from the parent, at the start of the late neanic stage.

Dissepimental development could not be established in the cardinal quadrants of the
species examined since these plates are masked by sclerenchymal deposits at the wall.
Not until the late neanic stage do they become obvious. In the counter quadrants, dis-
sepiments develop later than the minor septa, with one complete row present during the
late neanic stage. Dobrolyubova (1958, pp. 153-4) noted that while dissepiments are
absent during early hystero-ontogeny in L. rossicum , a single row is developed at an
early stage in the cardinal quadrants in L. volkovae.

At the point of discontiguity in all specimens examined, the daughter corallite pos-
sesses the diagnostic characters of the species, except that corallite diameter and septal
number are smaller than in the ephebic stage. At this point in four corallites of L.
arundineum subsp. n., septal numbers range from 20 to 22 of each order, and corallite
diameters are approximately 5 mm., as compared with adult septal numbers of 26 to
29, and corallite diameters of 7 to 8 mm. The diameter and septal number of corallites at
the point of discontiguity in L. arundineum and L. cf. pauciradiale are similarly smaller
than those of their adult stages.

R. K. JULL: CORALLUM INCREASE IN LITHO STROTION

215

The amount of distal growth between the commencement of increase and the point of
discontiguity varies considerably. A range of 2 to 4 mm. is found in L. arundineum
subsp. n., 1 to 2-5 mm. in L. arundineum (a species with small corallites), and 5-5 mm. in
the single example of increase studied in L. cf. pauciradiale. Septal spacing is slightly
narrower in detaching corallites than in the adult.

Between the point of discontiguity and the achievement of the ephebic stage, the rate
of corallite expansion and insertion of septa in the daughter is slow.

An interesting variation in hystero-ontogeny was noted in a diphymorphic corallum
of L. arundineum subsp. nov., in which two examples of increase were studied (text-fig.
2: 1). Both daughter corallites arise from consistently diphymorphic, non-columellate
parents, and their septa are not united during early hystero-ontogeny. Throughout
development in one daughter, axial and other major septa are never united at the axis
and a columella does not develop. However, during the late neanic stage of the other,
before it becomes discontiguous from the parent, axial septa are lengthened and become
united. After it becomes discontiguous, a small, thin columella appears, in spite of the
parent being non-columellate.

Type (iii): Lateral increase in species with a wide dissepimentarium

Text-fig. 4: lc; 5

Increase of this type was studied in three species which typically have three or more
rows of dissepiments. These species are: L. cf. martini Edwards and Haime from an
unspecified locality in the British Isles, L. stanvellense Etheridge from the Visean Bay-
wulla Formation in the west-central part of the Yarrol Basin, near Monto, Queensland,
and L. sp. nov. 1 from the same locality as L. stanvellense. Differences exist in their man-
ner of increase, but in all the daughter corallite inherits its initial septa from its parent.

Litliostrotion cf. martini (text-fig. 4: la-c; 5), unlike the other two species studied,
possesses an unusual morphology which is transitional in nature between the fasciculate
and cerioid forms. This is manifested by both the semi-compacted nature of corallites
and the characters of increase. Whereas increase is fasciculate in nature by the daughter
growing laterally free of the parent, rather than into an elongated corner as occurs in
cerioid forms, it is cerioid in nature by the daughter being divided from the parent by
a wall, rather than a partition. Moreover, the daughter becomes independent of the
parent by simply growing to the side of it and forming a straight-sided contact with it,
as is typical in cerioid lateral increase. Were this transitional nature known in L. martini
(sensu Edwards and Haime, 1852, pp. 197-8, pi. 40, figs. 2, 2 a-g), I would have no hesita-
tion in regarding my specimen as conspecific with it, although it differs in the minor
details of having a slightly larger columella and more strongly arched tabulae.

The following description is based mainly on increase in L. cf. martini , in which two
examples were studied, but it applies in general detail to that found in L. stanvellense.
Increase in this latter species is quite irregular and the progressive development of the
daughter corallite could not be studied in as much detail as was done in L. cf. martini.
Early development in L. sp. nov. 1, which differs from the other two species, is men-
tioned later.

Hystero-brephic stage (text-fig. 5: 1 b-g, 2 b-e). Increase commences with the inser-
tion of a variable number of short septa in the region of increase in the parent corallite

Q

C 3009

216

PALAEONTOLOGY, VOLUME 8

(text-fig. 5: 1 b-c, 2 b-c). A variable number is formed, with only one appearing in each
loculus. The peripheral ends of these newly inserted septa are located first at the outermost
row of dissepiments, but are subsequently lengthened both peripherally to meet the wall
and axially a variable distance. Concurrent with their insertion and lengthening is a pro-
gressive failure of the dissepiments and a shortening of the minor septa in the region
of increase.

Three minor septa of the parent are destined to become the cardinal and two alar
septa of the daughter corallite. The counter septum is formed later when the wall (or

text-fig. 4. Figs. 1 a-c, X 2; 2, X 4. 1 a-b, transverse and longitudinal sections of L. cf. martini Edwards
and Elaime, F30508B, C, Lower Carboniferous, unspecified locality, British Isles, lc, as fig. 1 a, showing
daughter corallite (D) lying against its parent (P) and encroaching into a neighbouring corallite. 2, L.
junceum (Fleming), showing connective tubules F2636/18, Upper Dj or D, zone, west of Black Rigg

Quarry, near Settle, Yorkshire, England.

partition) is constructed between the daughter and the parent corallite. When four
minor septa are involved in the region of increase, one of them is suppressed (text-fig. 5:
1 e-f, 2 d-f). Some of the newly inserted septa, which number more than four, take the
place of these minor septa, and the rest are suppressed, either in the late hystero-brephic,
or early hystero-neanic stage.

Hystero-neanic stage (text-fig. 5: 1/i-w, 2 f-k). During the early part of this stage, the
daughter corallite is still intimately associated with the parent and sclerenchymal deposits
are common, particularly lining the outer wall of the daughter.

This stage commences with the retreat of major and newly inserted parent septa from
the wall in the region of increase. United to the peripheral ends of these septa, which are

text-fig. 5. Figs. \a-m, x4 approx.; 2a-k, not to scale. Numbers below figs. 1 a-m are cumulative
distances of distal growth in mm. C = cardinal septum; K = counter septum; A = alar septum.
Figs. 1 a-m, 2 a-k. Lateral increase in L. cf. martini Edwards and Elaime. 1 a-m, Elystero-ontogenetic
sequence, F30508b, unspecified locality, British Isles. 2a-k, Semi-diagrammatic sketches of increase
in the above sequence. White septa = parent septa; black septa = septa which will be, or are,
those of the daughter.

TEXT-FIG. 5.

218

PALAEONTOLOGY, VOLUME 8

their thickest parts, are thin septa. These thin septa are the first metasepta of the daugh-
ter corallite, and three form in each cardinal quadrant. As the parent septa retreat
axially, the daughter septa lengthen, maintaining their attachment to the wall and to the
ends of the parent septa. Variations in metaseptal development occur, but these are of
minor significance.

As the metasepta of the daughter lengthen, minor septa are rapidly inserted between
them (text-fig. 5: 1 j, 2 i). When the septa of the parent have completed their retreat, the
daughter is divided from the parent corallite either by a wall, as in L. cf. martini (fig. 1 1),
which progressively forms across the zone of contact in line with the edge of the dis-
sepimentarium, or by a partition, as in L. stanvellense, which forms in a manner as
previously described.

The major septa in the cardinal quadrants of the daughter corallite are attached to the
dividing wall when it is first formed, but they almost immediately retreat. The first septa
formed in the counter quadrants, which are usually majors and include the counter
septum, appear on the dividing wall in the positions previously occupied by the ends of
the septa of the cardinal quadrants (text-fig. 5: 1/). Minor septa soon appear between
them, often in pairs.

The two axial septa are initially separate, but soon unite in the axial region and form
an irregularly shaped columella. The columella does not become typically fusiform until
the late neanic stage. Other major septa may be united at the axis in the early hystero-
neanic stage (not shown in text-fig. 5), but later they usually do not reach the axial
region. A single row of dissepiments in the cardinal quadrants is first obvious just before
the wall separating the daughter from the parent corallite is formed and a second row is
progressively developed before the daughter grows outside of the parent corallite (text-
fig. 5: 1 m). Dissepiments are formed later in the counter quadrants.

The maximum diameter of the daughter corallite of L. cf. martini when it reaches the
point of lying outside of the parent corallite is 5 mm., and its septal number is 20 septa
of each order, as compared with a typical adult diameter of 7 mm. and septal number of
24. The relative size and septal number of daughter to parent at the point of dis-
contiguity in L. stanvellense is similar to that in L. cf. martini. Approximately 3 to 4 mm.
of distal growth occurs in L. cf. martini between the start of increase and the separation
of the daughter from the parent, and 6 to 7 mm. occurs in L. stanvellense.

Newly formed corallites were not traced up to the ephebic stage. However, supporting
observations indicate that subsequent development was slow, and in L. cf. martini septa
are inserted with their peripheral ends resting against the outermost row of dissepiments,
as is the case of insertion during the initial stages of increase in this species. Septa appear
to be inserted from the wall in L. stanvellense , but this was not definitely established.

A variation in the above pattern of increase was noted inL. sp. nov. 1. In this species,
septa are not inserted in the parent corallite at the start of increase, and the cardinal and
alar septa, as well as the first metasepta in the cardinal quadrants of the daughter coral-
lite are formed at the peripheral ends of parent septa.

Type (iv): Lateral increase in cerioid species
Text-fig. 6

Sequences of increase were examined in three cerioid species. Two of them, L. cf.
portlocki (Bronn) from the D2 zone, near Victoria Bridge, Menai, Anglesey, Wales, and

R. K. JULL: CORALLUM INCREASE IN LITHO STROTION

219

L. minus (McCoy) from the S2 zone at Crummackdale, near Austwick, Yorkshire,
exhibit typical lateral increase, but the specimen of L. columnare Etheridge, which has
lonsdaleoid dissepiments and is from the Upper Visean Lion Creek Limestone in the
northern part of the Yarrol Basin, Stanwell, near Rockhampton, Queensland, has
a somewhat different type of increase.

The following description is based on the two British specimens, particularly L. cf.
portlocki, in which five complete sequences of increase were observed. Increase in L.
columnare is discussed later.

Hystero-brephic stage (text-fig. 6: 1 b-c). Increase during this stage is similar to that in
L. cf. martini. The dissepimentarium is suppressed in the critical area, either simply by
a progressive decrease in the number of dissepiments, as in L. minus, or by firstly
a modification of the dissepiments to a herringbone pattern, followed by their failure, as
in L. cf. portlocki (text-fig. 6: 1 b-e). New septa in the parent are inserted simultaneously
with the suppression of dissepiments, in the five examples of increase observed in L. cf.
portlocki, from one to five septa are inserted, and their peripheral ends are located axial
to the wall. Four septa are inserted in L. minus , but the spacing of serial sections did not
permit the establishment of their place of insertion with respect to the wall.

Hystero-neanic stage (text-fig. 6: 1 d-m). The daughter corallite is still intimately as-
sociated with the parent during the early part of this stage which commences with the
retreat of the parent septa from the wall in the region of increase.

With this retreat, the first septa in the cardinal quadrants of the daughter corallite
appear. They may either be formed independently, as occurs in some corallites of L.
cf. portlocki (text-fig. 6: 1 d-f), or at the ends of parent septa and apparently derived
from them. Minor septa are inserted between these majors before the dividing wall is
formed.

Some of the original septa of the parent are suppressed before and during the forma-
tion of the dividing wall, and these are replaced by the newly inserted septa. The dividing
wall is formed progressively across the zone of contact just peripheral to where the axial
edge of the dissepimentarium was located. It cuts oft' the peripheral ends of the parent
septa (text-fig. 6: g-i) and these ends form the initial septa in the counter quadrants of
the daughter corallite. Minor septa in these quadrants may be developed at this time as
well as majors. Septa remain short until the wall is entirely constructed, at which time
they are lengthened and united at the axis (text-fig. 6: 1/).

In L. minus, the columella appears as a swelling at the end of the counter septum as
soon as this septum is formed, but in L. cf. portlocki the columella does not appear until
the late neanic stage, after the axial septa are united at the axis. The first dissepiments in
the cardinal quadrants are deposited before the wall is completed (text-fig. 6: 1 g-h), one
row developing at a time.

The late neanic stage is reached shortly after the wall dividing the daughter from the
parent is completed (text-fig. 6: 1/). Septa at this time are arranged symmetrically and
united at the axis, and a narrow dissepimentarium and a typical columella are present.
This stage of development is approximately equivalent to the point of discontiguity of
daughter corallites in fasciculate species. The rate of corallite development up to this
point varies considerably between individuals of L. cf. portlocki, ranging from T2 to
2-9 mm. between the start of increase in the parent to the attainment of the late neanic
stage. In L. minus, about 3-8 mm. is involved. The ratio of septal number of daughter to

220

PALAEONTOLOGY, VOLUME 8

adult corallites at this stage in both species is approximately 4/5, somewhat similar to
that found in lateral increase in fasciculate species.

Following the attainment of the late neanic stage, corallite development becomes very
slow. Septa are inserted in the typical rugosan manner in fossulae which are not pro-
minently marked.

Two examples of increase were examined in a lonsdaleoid corallum of L. columnare
(text-fig. 6: 2 a-f). Both daughter corallites are formed entirely within lonsdaleoid areas
of the parent calices, and in this respect, their early development more closely resembles
that which occurs in some species of Lonsdaleia (see Smith, 1916; Dobrolyubova, 1958)
than in the cerioid species of Litliostrotion which are described above.

The corallum in question is only weakly lonsdaleoid, with most corallites either
lacking lonsdaleoid dissepiments or having only small developments of them, mainly
in their elongated corners. The two corallites which exhibit increase, however, have
notably wide lonsdaleoid areas. Whether this is simply a coincidence, or whether it
suggests some relationship between increase and lonsdaleoid dissepiments, is not known
at present.

The daughter corallite arises at the edge of the parent calice (text-fig. 6: 2d). Although
septa of the parent are expressed as short developments on the upper surfaces of the
lonsdaleoid dissepiments, none extend continuously across the lonsdaleoid dissepi-
mentarium to the region of increase during early hystero-ontogeny; hence all daughter
septa seem independently inserted. Sclerenchymal deposits are initially present in the
daughter corallite, but these soon disappear.

During the hystero-brephic and early hystero-neanic stages of development, the
daughter corallite is semi-circular in shape and is divided from the parent by a thin wall
(text-fig. 6 : 2 a-e). Corallite expansion occurs entirely within the parent calice. In neither
of the two developing corallites examined were the axial septa the first to be inserted,
owing to the early asymmetrical development of these corallites. Primary septa and
some metasepta are united at the axis soon after they are inserted, and interseptal disse-
piments are formed at an early stage of development.

Upon reaching the late neanic stage (text-fig. 6: 2/), the daughter corallite gradually
assumes a sub-hexagonal or similar shape, it occupies the entire width of the lonsdaleoid
area of the parent, and later growth is only in a distal direction. The wall dividing the
daughter from the parent is thickened and becomes a typical intercorallite wall. A single
row of interseptal dissepiments is incompletely developed around the corallite, and lons-
daleoid dissepiments are absent, except for a few which occur between some minor
septa and the wall.

This type of increase differs considerably from lateral increase in non-lonsdaleoid
cerioid coralla. Further studies may recognize it formally as a separate type, perhaps
characteristic of lonsdaleoid cerioid species, and epitomized by increase in Lonsdaleia.

text-fig. 6. Figs. 1 a-m, x6 approx.; 2 a-f, x4 approx. Numbers below figs. 1 a-m are cumulative
distances of distal growth in mm. C = cardinal septum; K = counter septum.

Figs. 1 a-m. Lateral increase in L. cf. portlocki (Bronn), Hystero-ontogenetic sequence, F31047, D2zone,
near Victoria Bridge, Menai Strait, Anglesey, Wales.

Figs. 2 a-f. Lateral increase in L. columnare Etheridge, Hystero-ontogenetic sequence, F44601, Upper
Visean, Lion Creek Limestone, Yarrol Basin, Stanwell, near Rockhampton, Queensland.

c

222

PALAEONTOLOGY, VOLUME 8

PERIPHERAL INCREASE
Text-fig. 7: 1 a-b, 2a -m

Peripheral increase is known in only a single specimen of Lithostrotion, this being L.
sp., a koninckophylloid form from the Upper Visean Riverleigh Limestone in the
southern part of the Yarrol Basin, near Mundubbera, Queensland. Hill (1934, pi. 10,
fig. 31) illustrated one of the three examples of peripheral increase present in this
specimen. A single example of lateral increase is also present. This was figured by Hill
(1934, pi. 10, fig. 31) and is apparently an aberrant type in which the partition dividing
daughter from parent is suppressed at an early stage of increase and the two corallites
remain contiguous until after the daughter has achieved the ephebic stage of develop-
ment.

The following description is based on examination of serial sections of one example of
peripheral increase, and thin and polished sections of the other two examples.

From four to eight corallites are formed at different levels in the peripheral parts of
the parent calice. Generally, the nearer to the calical rim that a corallite is formed, the
smaller is its initial size. The newly formed corallites develop at varying rates.

Corallites are not necessarily formed in the peripheral regions of the calice. In the two
other examples of peripheral increase, one of which is illustrated herein (text-fig. 7:
1 a-b) a corallite appears well within the periphery, and is one of the first to be formed.
The development of the corallite in this position was not traced in detail, but it is similar
in appearance to the daughter corallites formed at the periphery.

Before increase, septa in the calice are shortened in a normal manner until that point
when part of the periphery is modified to accommodate the lowermost of the daughter
corallites formed (text-fig. 7 : 2b). A number of thin walls are formed at the axial ends of
major septa and at varying levels in the calice, ranging from near the base to near the
top. These walls are of differing lengths, and are eventually resolved into the inner walls
of from one to three corallites. They are sub-symmetrically arranged on either side of
the axial plane of the parent but are not as equally ranged around the periphery of the
calice as are those in the Wenlockian Entelophylhun articulation (Wahlenberg) or the
Carboniferous D2 Nemistium edmondsi Smith (see Smith and Tremberth, 1929; Smith,
1928).

When septa are enclosed by a wall, they either cease to shorten distally, or more com-
monly, continue to shorten but at a slower rate than the unenclosed septa. Enclosed
septa are usually incorporated into the morphology of the daughter corallite, but some
continue shortening until they disappear, later to be replaced by newly inserted septa.
Almost as soon as the enclosing wall is formed, it commences to migrate towards the
axis of the parent and becomes disunited from the major septa which were attached to it

text-fig. 7. Figs. 1 a-b, 2 a-m, X 2 approx.; 3, X 5 approx. Numbers below figs. 2 a-m are cumulative
distances of distal growth in mm. C = cardinal septum ; K = counter septum.

Figs. 1 a-b, 2 a-m. Peripheral increase in Lithostrotion sp. \a-b. Two consecutive stages of development,
dashed line indicates the approximate original position of the parent calice, F2965h, j. Upper
Visean, Riverleigh Limestone, Yarrol Basin, near Mundubbera, Queensland. 2 a-m, Hystero-
ontogenetic sequence of another example from the same specimen as figs. 1 a-b.

Fig. 3. Axial increase in L. junceum junceum (Fleming), C{g\ Russian Platform. Drawing is based on
Dobrolyubova, 1958, pi. 18, fig. li>.

TEXT-FIG. 7.

224

PALAEONTOLOGY, VOLUME 8

(text-fig. 7 : 2d). Short major septa appear on this wall at many points where peripheral
septa were attached. Additional major septa soon appear on the dividing wall, followed
shortly by minor septa. The septa on the dividing wall lengthen to a varying extent and
the wall is modified in such a way that the one or more daughter corallites which it
encloses gradually assume a sub-circular shape. A columella is not developed until one
or both axial septa are lengthened to the axis. In the largest corallite formed, this occurs
almost immediately, but in some of the others, a columella had still not formed by the
last observed stage of their development, up to 3 mm. above their initial development.
The axial plane of the daughter corallite is not orientated radially with respect to the
axis of the parent, as it is in lateral increase, but approximately at right angles to this
direction (text-fig. 7: 1 b, 21).

From one to three rows of parent dissepiments are incorporated into the morphology
of those daughter corallites which also inherited parent septa. A single row of dissepi-
ments is formed between septa attached to the dividing wall, shortly after these septa are
inserted. A second and third row of dissepiments are progressively formed around the
corallite as it increases in size and septa become longer.

The late neanic stage of development (fig. lb) is achieved when corallites grow free of
the parent calice. Development thenceforth resembles that in corallites formed by lateral
increase.

AXIAL INCREASE

Dobrolyubova (1958) reported axial increase in L. junceum junceum (Fleming), L.
caespitosum (Martin), and L. scoticum Hill, and also in three species of Diphyphyllum ,
all of which are from Lower Carboniferous deposits on the Russian Platform. In all but
one of the species of Diphyphyllum , lateral increase occurs as well. Dobrolyubova (1958)
called axial increase ^eAemie’ which I have translated as ‘division’ (rather than the
dictionary alternative, ‘fission’).

Her description and illustration of axial increase in L. junceum junceum (1958, p. 145,
pi. 18, fig. It; herein refigured in text-fig. 7: 3) show it to be particularly unusual. The
parent corallite diameter and septal number are nearly doubled, and twin columellae
appear; then follows the development of a wall delimiting two new corallites which
occupy the entire calice of the parent.

Another type of increase in L. junceum junceum, called ‘connective tubular’ increase, was
named and described by Dobrolyubova (1958, p. 145, pi. 18, fig. 1/3). She considered that
a young corallite developed in a connective tubule which the parent had thrown out to a
neighbour. However, the example which she illustrates appears to be lateral increase in
which the contiguous daughter corallite is elongated to form a tubule connective to a
neighbouring corallite. This elongation is common in laterally arising immature corallites
of L. junceum auctt. (see text-fig. 1 : 1 m, 2; text-fig. 4: 2) and should not be regarded as
a particular type of corallum increase.

CONCLUSIONS

1. The mode of increase in Lithostrotion is of such variable character that it has
limited generic value. This is also known in other rugosan genera and Matthai (1926,
p. 351) noted a similar situation in scleractinian corals.

2. The differences in the type of lateral increase between fasciculate and cerioid forms
of Lithostrotion may justify their generic separation.

R. K. JULL: CORALLUM INCREASE IN LITHO STROTION

225

3. Hystero-ontogenetic development in Lithostrotion seems to confirm the suggestion
of Smith and Lang (1930, p. 179) that Diphyphyllum arose from Lithostrotion, and not
the reverse, as was suggested by Minato (1955, text-fig. 20).

4. Axial and peripheral increase are without equivalents in the Scleractinia, but lateral
increase in which the daughter corallite inherits septa from the parent resembles the
distomodaeal condition of intratentacular budding.

5. Laterally arising hysterocorallites in which the septa are all independently inserted,
appear to undergo almost completely independent development, and may mirror the
ontogenetic development of their protocorallites.

REFERENCES

coope, g. r. 1956. The insertion of septa in the later growth stages of clisiophyllid corals. Geol. Mag.
93, 233-41, pi. 9.

Dobrolyubova, t. a. 1958. Lower Carboniferous colonial tetracorals from the Russian platform.

Trav. Inst, paleont. Acad. sci. U.R.S.S. 70, 1-224, 38 pi. [in Russian],
easton, w. h. 1957. On the tetracoral Lithostrotion harmodites Milne-Edwards and Haime. J. Paleont.
31, 616-22, pi. 71.

Edwards, h. m. and haime, j. 1852. A monograph of the British fossil corals. Part 3. Corals from the
Permian formation and the Mountain limestone. Palaeontogr. Soc. [Monogr.], 147-210, pi. 31-46.
groot, g. e. de, 1963. Rugose corals from the Carboniferous of Northern Palencia (Spain). Leid.
geol. Mec/ed. 29, 1-123, 26 pi.

hill, d. 1934. The Lower Carboniferous corals of Australia. Proc. roy. Soc. Qd. 45, 63-115, pi. 7-1 1.

1935. British terminology for rugose corals. Geol. Mag. 72, 481-519.

— 1938-41. A monograph on the Carboniferous rugose corals of Scotland. Palaeontogr. Soc.
[Monogr.], 213 pp., 11 pi.

1956. Rugosa ; In Treatise on Invertebrate Paleontology (R. C. Moore, Ed.). Part F : Coe/enterata,

pp. 233-324. Univ. Kansas Press and Geol. Soc. Amer.

Hudson, r. g. s. 1930. The age of the ‘ Lithostrotion arachnoideum ’ fauna of the Craven lowlands.

Proc. Leeds phil. Soc. (Sci. Sect.), 2(2), 95-101, pi. 1.
macgregor, a. G. 1960. Divisions of the Carboniferous on Geological Survey Scottish maps. Bull,
geol. Surv. G.B. 16, 127-30.

matthai, g. 1926. Colony-formation in astraeid corals. Phil. Trans, roy. Soc. Load. (Ser. B), 214,
313-67, pi. 24-28.

minato, m. 1955. Japanese Carboniferous and Permian corals. J. Fac. Sci. Hokkaido Univ. (Ser. 4),
9, 1-202, 43 pi.

rozkowska, m. 1960. Blastogeny and individual variations in tetracoral colonies from the Devonian
of Poland. Acta, palaeont. polon. 5, 3-64.

smith, s. 1916. The genus Lonsdaleia and Dibunophylhan rugosum (McCoy). Quart. J. geol. Soc. Load.
11, (for 1915), 218-72, pi. 17-21.

1928. The Carboniferous coral Nemistium edmondsi, gen. et sp. n. Ann. Mag. not. Hist. (Ser. 10),

1, 112-20, pi. 5.

1945. Upper Devonian corals of the Mackenzie river region, Canada. Spec. Pap. geol. Soc. Amer.

59, viii+126 pp., 35 pi.

and lang, w. d. 1930. Descriptions of the type-specimens of some Carboniferous corals of the

genera ‘ Diphyphyllum ’, ‘ Sty last raea ’, Aulophyllum, and Chaetetes. Ann. Mag. nat. Hist. (Ser. 10),
5, 177-94, pi. 7-8.

and ryder, t. a. 1926. The genus Corwenia, gen. nov. Ibid. (Ser. 9), 17, 149-59, pi. 5-6.

and tremberth, r. 1929. On the Silurian corals Madreporites articulatus Wahlenberg, and

Madrepora truncata Linnaeus. Ibid. (Ser. 10), 3, 361-76, pi. 7-8.

R. K. JULL

Department of Geology and Mineralogy,
University of Queensland,
Brisbane, Queensland

Manuscript received 29 February 1964

THE NAMURI AN GONIATITE NUCULOCERAS
STELLARUM (BISAT)

by B. K. HOLDSWORTH

Abstract. The goniatite previously known as ‘ Cravenoceratoides stellarwn is spirally ornamented and usually
possesses a small umbilicus. The early Homoceratina are best classified at generic level in terms of ornament and
stellarwn should be included in the genus Nuculoceras. English material is described and comparison made with
foreign descriptions of the species. A revision of zonal classification in the Arnsbergian Stage (E2) of the
Namurian is proposed.

Nuculoceras stellarwn, originally described by Bisat (1932) from Gill Beck, Cowling,
Yorks, occurs at a single horizon in the English shale-sandstone development of the
Namurian Series. The band containing the goniatite is thus a valuable marker horizon in
stratigraphical studies. Unfortunately the original description of N. stellarwn is some-
what inadequate and during recent work it has become apparent that neither Bisat’s
description nor the amplified description given by Hudson (1946) brings out the most
important diagnostic features of the species or allows a clear distinction to be made
between N. stellarwn and the rather closely allied species Cravenoceratoides nititoides
(Bisat) and N. nuculum Bisat. As the horizon of Ct. nititoides is immediately below N.
stellarwn and three horizons with N. nuculum (Ramsbottom et al. 1962, p. 130) im-
mediately succeed the N. stellarwn band there is some difficulty at present in distinguish-
ing between three important levels in the Namurian succession. Ct. nititoides has been
redescribed and figured (Yates 1962, p. 391, pi. 57, figs. 4, 5) and it is desirable that N.
stellarwn should also be redescribed.

The present description is based upon three collections :

Collection 1. Small suite of specimens completely crushed in hard limestone. Upper
Dove Valley, south-west Derbyshire, Grid Ref. SK 08666631 (Locality 326 — Holds-
worth 1963).

Collection 2. Suite of specimens crushed and partially crushed in decalcified silty
limestone lying 13 feet above the horizon of Eumorphoceras rostratum Yates and 25
feet above Cravenoceras hohnesi Bisat, Oakenclough Brook, north-east Staffordshire,
Grid. Ref. SK 05046368 (Locality 206c — Holdsworth 1963).

Collection 3. Topotypes of N. stellarwn, Geological Survey of Great Britain collec-
tion, Leeds Office, Nos. Da 1626-60.

SYSTEMATIC DESCRIPTION

Order ammonoidea Zittel 1884
Suborder goniatitina Hyatt 1884
Superfamily goniatitacea de Haan 1825
Lamily goniatitidea de Haan
Subfamily homoceratina Spath
Genus nuculoceras Bisat 1924

Type species. Nuculoceras nuculum Bisat 1924.

[Palaeontology, Vol. 8, Part 2, 1965, pp. 226-30, pi. 25.]

B. K. HOLDSWORTH: THE NAMURIAN GONIATITE

227

Diagnosis. Early Homoceratina with ventral lobe of suture narrower than in Homoceras.
Conch involute, subglobose with small or very small umbilicus. Shell surface bears
bifurcating transverse ribs and subsidiary spiral ornament.

Nuculoceras stellarum (Bisat)

Plate 25, figs. 1-6

1927 Homoceras cf nitidum (Phillips), Bray, p. 55.

1932 Cravenoceras stellarum Bisat, pp. 33-34, pi. 2, fig. 1.

1934 Cravenoceras stellarum Bisat; Schmidt, p. 450, fig. 46.

1934 ? Cravenoceras nititoides Bisat; Schmidt, p. 450, fig. 47.

1941 Cravenoceras stellarum Bisat; Demanet, p. 144, pi. 6, figs. 9, 10.

1941 Cravenoceratoides stellarum (Bisat); Hudson, p. 282, footnote.

1946 Cravenoceratoides stellarum (Bisat); Hudson, p. 380, pi. 21, fig. 9.

The true original shape of the shell is impossible to determine in the available material,
but is probably subglobose (Hudson 1946, p. 380). The ratio of umbilicus diameter to
diameter of the crushed shell (u/d) is variable but, except in the very early growth stage,
usually exceeds 4-0 — i.e. the umbilicus is relatively small.

Specimen

Shell

diameter

Umbilicus

diameter

uld

Collection 1

326.2

> 1 TO mm.

T5 mm.

> 7-33

326.3

> 190

40

>4-75

326.1

> 25 0

50

> 5 00

Collection 2

206c. 1

40

TO

4-00

206c.4

90

T5

600

206c.2

> 160

< 20

>800

206c. 3

>28 0

60

>4-70

Collection 3

Da. 1631

60

ca. TO

600

Da. 1639

100

20

5 00

Da. 1632

ca. 1 TO

2-0

5-50

Da. 1628

> 160

2-5

>6-40

Da. 1642

> 20 0

30

> 6-60

Da. 1637

> 25 0

6-5

> 3-84

Da. 1629

ca. 26 0

50

5-20

At 4-0 mm. diameter the shell is evolute. At 9-0 mm. the typical narrow umbilicus has
been assumed. Ribs with spacing c. 4 per mm. at the venter have a very slight forward
tendency at the umbilicus (PI. 25, fig. 6). Neither umbilical rim nor spiral ornament
appear to be developed, and at this growth stage there is a similarity with the adult Cl.
nititoides (cf. Yates 1962, pi. 57, figs. 4, 5). At diameter slightly greater than 16-0 mm.
the "nititoides aspect’ is lost. Rib direction now appears essentially truly radial with
spacing c. 5 per 2 mm. at the venter. Ribs are symmetrical (tented) in elevation and the
external mould clearly displays spiral corrugation of inter-rib areas. Though hardly ever
detectable on shell surfaces, this spiral ornament can frequently be seen on good external
moulds of the adult flank (PI. 25, fig. 3) and venter (fig. 4). Almost invariably ribs appear
non-crenulate on shell surfaces and crenulation is not detectable on external moulds.
Only on a few shell sectors of Specimen 326.1 is a faint, rather broad crenulation of ribs

228

PALAEONTOLOGY, VOLUME 8

detectable. The umbilical margin in the adult is very frequently raised into a rounded
rim across which the ribs pass without weakening. The feature is particularly clear on
Specimens Da 1629 and Da 1637 (PI. 25, fig. 2) and the rim imprint is commonly seen
on external moulds (PI. 25, fig. 5).

Discussion. The holotype of N. stellarum has u/d ratio 3-6 at diameter 18 mm. and in
view of the measurements made on topotypes (Collection 3) appears to be untypical of
the species at the Gill Beck locality. There is no reason to believe, therefore, that the
‘similar form, but with a smaller umbilicus’ recorded by Bisat (1932, p. 34) from Glutton
Bridge, Derbyshire — possibly the locality of Collection 1 — is ‘a late form of the species’.

N. stellarum is distinguished from Ct. nititoides and all true Cravenoceratoides (see
below) by its weak, rather coarse spiral ornament. In collections of moderate preserva-
tion, traces of this ornament and the presence of the raised umbilical rim are the most
useful criteria for distinguishing stellarum from nititoides. In small specimens of stel-
larum neither feature seems to be developed and distinction between stellarum and
nititoides cannot be made with certainty. N. nuculum is a smaller species than stellarum,
‘the adult being apparently not more than 18 mm.’ in diameter (Bisat 1924, p. 100).
Nuculum lacks the raised umbilical rim of stellarum at the shell surface, though a rim is
sometimes apparent on the solid internal cast. In shale, mudstone, and fissile limestone
the spiral ornament of nuculum , in contrast to that of stellarum , is seldom detectable on
either shell surfaces or external moulds. Conversely, the crenulation of transverse ribs —
so very rarely visible on stellarum — can usually be detected on nuculum specimens in
a similar state of preservation.

The generic assignment of stellarum

The genus Nuculoceras was founded by Bisat (1924, p. 100) with nuculum as type
species and sole member. Spiral ornament and globose conch are the two features men-
tioned in the generic diagnosis. Bisat (1928, p. 132) erected the genus Cravenoceras to
include ‘early Homoceras-Uke forms’ having a suture with ventral lobe narrower than
in Homoceras proper, two of which forms — malhamense and nitidum — had earlier (Bisat
1924) been included in Homoceras. Hudson (1914, p. 282, footnote) restricted the name

EXPLANATION OF PLATE 25

Fig. 1. Nuculoceras stellarum (Bisat), external mould of specimen crushed in hard limestone with some
small areas of shell-surface, slightly displaced. Specimen shows small umbilicus, essentially truly
radial ribs, weak imprint of umbilical rim and traces of spiral ornament. Upper Dove Valley,
Derbyshire. 326.1, x4-4.

Fig. 2. Nuculoceras stellarum (Bisat), shell-surface of topotype crushed in mudstone and showing
raised umbilical rim; Gill Beck, Yorkshire. Da 1637, X 3-2.

Fig. 3. Nuculoceras stellarum (Bisat), detail of external mould showing spiral ornament. 326.1, X60.

Fig. 4. Nuculoceras stellarum (Bisat), external mould of venter with shell patina, showing spiral
ornament; Oakenclough Brook, Staffordshire. 206c.7, x6 0.

Fig. 5. Nuculoceras stellarum (Bisat), external mould of umbilical fragment, showing imprint of
umbilical rim; Oakenclough Brook, Staffordshire. 206c.8, x 5-0.

Fig. 6. Nuculoceras stellarum (Bisat), external mould of small adolescent showing ‘ nititoides-aspect' ;
Oakenclough Brook, Staffordshire. 206c.4, x 7-8. (Owing to a common optical illusion the ribs
appear, in the photograph, to be preserved in relief. In fact the specimen is a mould and the true
appearance can be obtained by inverting the figure.)

Palaeontology, Vol. &

PLATE 25

HOLDSWORTH, Namurian goniatite Nuculoceras

B. K. HOLDSWORTH: THE NAMURIAN GONIATITE

229

Cravenoceras to early Homoceratina with non-bifurcating transverse ribs, proposing the
new genus Cravenoceratoides to include forms such as nitidum in which the ribs bifurcate.

As regards width of ventral lobe there is no marked difference between nuculum (cf.
McCaleb 1963, fig. 3a) and species included in Cravenoceras. Material of species as-
signed to Cravenoceratoides seldom displays sutures, but the ventral lobe and general
form of suture in nitidum (cf. Bisat 1924, p. 107) and Cravenoceratoides eda/ense (Bisat)
(cf. Schmidt 1934, fig. 30) are essentially similar to the basic Cravenoceras type. Thus the
early Homoceratina, including nuculum , appear to constitute a rather closely related
group, best classified at generic level in terms of ornament.

The distinction between Nuculoceras and Cravenoceratoides is largely blurred by the
fact that Delepine (1941) defined Nuculoceras as a genus of globose forms with extremely
reduced umbilici, making no reference to the important spiral ornament of the type
species. He included in Nuculoceras his new species djeradae, lacking spiral ornament,
and a second new species, agaiae, insecurely founded on a single specimen with no orna-
ment preserved.

The original conch shape of both stellarum and nititoides is probably subglobose and
similar to that of nuculum. The umbilicus, though variable in both stellarum and niti-
toides, may be very small. Thus, though the two forms are currently included in Craveno-
ceratoides they also fall within Delepine’s definition of Nuculoceras. Classification might
be rationalized by either (a) suppressing the name Cravenoceratoides and including all
early Homoceratina with bifurcating ribs in the earlier genus Nuculoceras or ( b ) by re-
turning to Bisat’s original concept of Nuculoceras and including here only forms with
both bifurcating ribs and spiral ornament. The latter alternative is the more satisfactory
as it involves comparatively little revision of present nomenclature and also serves to
separate stratigraphically earlier ( Cravenoceratoides ) and later ( Nuculoceras ) members
of the pr e-Homoceras group of goniatites seen in European successions.

Here, therefore, stellarum, together with nuculum and tenuispirale Demanet ( 1941 ), are
assigned to Nuculoceras and djeradae to Cravenoceratoides. In the absence of preserved
ornament the genus of agaiae is uncertain.

Zonal allocation of N. stellarum

Currently four zones are recognized as constituting the Arnsbergian Stage (E2) of
the Namurian, stellarum being taken as index of the third zone, E2c, and nuculum as
index of the fourth, E2d. With respect to faunal richness, E2c is hardly comparable
with the other three zones. It contains only a single fauna — that of stellarum — developed
in a single thin band. A tripartite zonal division of the Arnsbergian seems more justi-
fiable, the Nuculoceras-con\2L\nmg strata being grouped together in the Nuculoceras
nuculum Zone E2c, with stellarum as index of a lower subzone, E2c.l, and nuculum
index of an upper subzone, E2c.2.

N. stellarum at other localities

Poor material from 294-5 feet in the Alport Boring (Geological Survey Collection
Zh. 933-42), recorded by Hudson and Cotton (1943, p. 161) as Ct. stellarum and Ct. cf.
nititoides, includes a few specimens showing the relatively small umbilicus with raised
rim typical of stellarum. The remaining specimens are not identifiable with certainty, but
there appears to be no form referable to nititoides. Material from the Crowburn Brook,

230

PALAEONTOLOGY, VOLUME 8

Edale (Geological Survey Collection RS. 2873-82) recorded as stellarum includes a single
external mould showing spiral ornament.

Schmidt (1934, p. 450) identified as stellarum a spirally ornamented form with u/d
ratio about 4-0 at 20-0 mm. diameter (fig. 46) and at the same horizon — ‘Schicht mit
Cravenoceras stellarum'1 — found a spirally ornamented form with u/d ratio about 8-0 at
comparable diameter (op. cit, supra , fig. 47) to be slightly more common. This latter
form he attributed to nititoides, but in view of the re-examination of the features of
stellarum it seems probable that both widely and narrowly umbilicate forms should be
referred to this species. There is a suggestion of an umbilical rim in Schmidt’s fig. 46.

The form from the Asisse de Chokier identified by Demanet (1941, p. 144) as stellarum
is similar to the English specimens in that spiral ornament is apparently visible only on
external moulds. It is not clear whether Demanet’s specimens possess an umbilical rim.
The superposition of slightly displaced portions of crushed shell, giving a reticulate
appearance, which Demanet described is seen in specimens of Collection 1 (cf. PI.
25, fig. 1).

Acknowledgements. The writer is grateful to the Director of the Geological Survey of Great Britain
for permission to examine Survey collections and borrow specimens and to Dr. W. H. C. Ramsbottom
for valuable discussion concerning the contents of the paper and critical reading of the manuscript.

REFERENCES

bisat, w. s. 1924. The Carboniferous goniatites of the north of England and their zones. Proc. Yorks.
Geol. Soc. 20, 40-124, pi. 1-10.

1928. The Carboniferous goniatite zones of England and their continental equivalents. Congr.

Avanc. Et Stratigr. carbonif Heerlen, 1927, 117-33, pi. 6, 6a.

1932. On some Lower Sabdenian goniatites. Trans. Leeds geol. Ass. 5, 27-37, pi. 1, 2.

bray, a. 1927. The Carboniferous sequence between Lothersdale and Cowling (Colne). J. Manchr.
geol. Ass. 1, 44-57.

delepine, g. 1941. Les Goniatites de Carbonifere du Maroc et des Confins Algero-Marocains du sud
(Dinantien-Westphalien). Notes Serv. Min. Maroc. 56, 1-108, pi. 1-8.
demanet, f. 1941. Faune et stratigraphie de l’etage namurien de la Belgique. Mem. Mas. Hist. nat.
Belg. 84.

holdsworth, b. k. 1963. Unpublished Ph.D. Thesis, University of Manchester.

Hudson, r. g. s. 1941. The Mirk Fell Beds (Namurian, E2) of Tan Hill, Yorkshire. Proc. Yorks, geol.
Soc., 24, 259-89.

1946. The Namurian goniatites Cravenoceratoides bisati Hudson and Cravenoceratoides lirifer

n.sp. Proc. Yorks, geol. Soc. 25, 375-86, pi. 21, 2la.

and cotton, g. 1943. The Namurian of Alport Dale, Derbyshire. Ibid. 25, 142-73.

mccaleb, j. a. 1963. The goniatite fauna from the Pennsylvanian Winslow Formation of north-west
Arkansas. J. Paleont. 31, 867-88, pi. 111-15.

ramsbottom, w. h. c., rhys, g. h. and smith, e. g. 1962. Boreholes in the Carboniferous rocks of the
Ashover district, Derbyshire. Bull. Geol. Survey Gt. Br. 19, 75-168.
yates, p. J. 1962. The palaeontology of the Namurian rocks of Slieve Anierin, Co. Leitrim, Eire.
Palaeontology, 5, 355-443, pi. 51-62.

B. K. HOLDSWORTH

Department of Geology,
The University,

Keele,

Manuscript received 5 June 1964 Staffordshire

SYSTEMATICS, AFFINITIES, AND FIFE HABITS
OF BABINKA, A TRANS IT IONAF ORDOVICIAN
FUCINOID BIVAFVE

by A. LEE MCALESTER

Abstract. The rare bivalve genus Babinka from lowest Middle Ordovician rocks of the Bohemian Basin shows
multiple muscle scars which have led several palaeontologists to suggest a relationship to some metameric mol-
luscan ancestor. A systematic and morphologic revision reveals that Babinka is a typical bivalve in all features
except the pedal and gill muscle-scar patterns. These scars are not like those of other bivalves, but are almost
identical to the pattern found in the recent monoplacophoran Neopilina, and in some early Palaeozoic Mono-
placophora. This close similarity confirms the suggestion that the muscle pattern in Babinka is an inheritance
from a monoplacophora-like ancestral mollusc.

Babinka is among the first bivalves to appear in the fossil record. The genus is both chronologically and
morphologically an ideal ancestor for the earliest lucinoid bivalves which appear abruptly in Middle Silurian
deposits. Morphological features of Babinka which are strongly suggestive of lucinoid affinities are : the anteriorly-
expanded shell shape; the elongate anterior adductor muscle with associated ‘elongate impression'; the non-
sinuate pallial line; and the typical lucinoid hinge, dentition, and ligament. Babinka provides the first direct
evidence of an evolutionary transition between the Bivalvia and more primitive molluscs.

Functional comparison with recent lucinoid bivalves suggests that Babinka was a mobile, infaunal suspension
feeder that received nutrient-laden water into the mantle cavity through an anterior inhalent opening main-
tained by extrusion of the foot.

Several students of molluscan phytogeny have recently called attention to the curious
early bivalve genus Babinka (Barrande, 1881) from Ordovician rocks of Czechoslovakia.
This rare monotypic genus is known only from the Bohemian Basin where it occurs in
a formation (Sarka beds) that is probably of Llanvirn (lowest Middle Ordovician) age.
Bivalve molluscs are extremely rare in Llanvirn or pre-Llanvirn rocks, and for this
reason alone Babinka is of particular interest as one of the first known representatives of
the Bivalvia. Further interest attaches to the genus because internal moulds preserve
clear impressions of multiple pairs of muscle scars. These multiple muscle scars have led
to the suggestion that Babinka is a primitive transitional form between the Bivalvia and
some metameric molluscan ancestor (Vokes, 1954; Cox, 1959, 1960; Ruzicka and Prantl,
1960; Horny, 1960; Vogel, 1962; Merklin, 1962).

Although Babinka has been the source of much speculation regarding the early history
of the Bivalvia, the genus has not been critically restudied since its first cursory descrip-
tion by Barrande almost a century ago. While preparing a review of the phylogeny and
adaptations of Palaeozoic lucinoid bivalves, I have noted many characters of Babinka
that suggest a relationship to the first lucinoid forms which appear abruptly in Middle
Silurian deposits. The present study was prompted by this possibility of lucinoid affinities,
and by the often suggested transitional evolutionary position of the genus. This paper has
been prepared in order to: (1) review the systematics and morphology of Babinka ;
(2) further examine the functional and phylogenetic significance of the muscle scar
pattern and other morphologic features of the genus; (3) suggest that Babinka is
an ancestral lucinoid bivalve, and; (4) attempt to interpret the life habits of Babinka
by analogy with recent lucinoid forms. More general phylogenetic and systematic

[Palaeontology, Vol. 8, Part 2, 1965, pp. 231-46, pi. 26-28.]

C 3009 R

232 PALAEONTOLOGY, VOLUME 8

conclusions which have resulted from the study will be treated in a separate paper
(McAlester, 1965).

BABINKA AS A PRIMITIVE BIVALVE

Vokes (1954) appears to have been the first to call attention to the possible phylo-
genetic significance of the muscle scars in Babinka. He noted that the multiple scars of
the genus, and the dissimilar multiple muscle scars of several Ordovician nuculoid
species, are suggestive of the series of pedal muscle scars seen in fossil monoplaco-
phorans, and he concluded (p. 236) : ‘ . the muscle scars shown by these Ordovician
pelecypods can be shown to be close to those exhibited by primitive gastropods . . .
they therefore may be interpreted as reflecting the musculature present in the ancestral
stock from which the Pelecypoda were derived. Further, they suggest that the adductor
muscles of the Pelecypoda are derived from discrete pairs of the ancestral musculature,
rather than from the union of multiple pairs. ’

Vokes suggestion was discussed by Cox (1959, 1960), who agreed in concluding (1959,
p. 204) that Babinka ‘could well have been newly evolved from the ancestral mollusc’.
Cox further noted (1960, p. 71) that ‘ Babinka appears to have approximated to the
theoretical concept of the newly evolved bivalve mollusc. Little can be said about the
role it played in bivalve phylogeny until it is better known.’

The idea that Babinka might be closely related to a monoplucophora-like ancestor
was repeated by Ruzicka and Prantl (1960), and greatly expanded by Horny (1960), who
regarded the genus as ‘the phylogenetically initial form of the pelecypods’ (p. 479).
Because of this proposed phylogenetic position. Horny erected a new family (Babinki-
dae) and order (Diplacophora) for the genus. The latest discussions are those of Vogel
(1962) and Merklin (1962), both of whom agree with Horny’s conclusions.

Barrande’s original descriptions, and all later discussions of Babinka , have not clearly
established the nature and number of the multiple muscle scars which have aroused so
much interest. In an attempt to clarify the generic morphology, I have restudied all
specimens of Babinka in North American museums, and in the Narodni Muzeum at
Prague (see the section on Systematics, p. 241). This revision has provided several natural
internal moulds which reveal for the first time the fine details of the muscle impressions.
Of particular importance in showing the precise pattern is one extraordinarily clear
internal mould of a right valve from the collections of the Narodni Muzeum which was
generously made available for study by Dr. Horny (PI. 28, figs. 9-11). These internal
moulds show that Babinka has normal adductor muscle scars (text-fig. 1, aam, pam)
and, in addition, a series of eight smaller scars above and between the adductor impres-
sions. To avoid functional connotations, these eight scars will be temporarily referred
to as the ‘intermediate’ muscle-scar impressions (text-fig. 1, im). Below some of these
intermediate impressions is a series of about twenty-five still smaller scars. These will be
temporarily called the ‘small’ muscle-scar impressions (text-fig. 1, sm). Finally, a large
but obscure and faintly bounded ‘elongate impression’ (text-fig. 1, ei) extends ventrally
from the anterior adductor scar, and a faint non-sinuate pallial line of mantle muscle
attachment connects the adductors in the usual position (text-fig. 1, pi).

The first problem in interpreting the muscle scars of Babinka concerns the cross-
sectional shape of the muscles which attached to the adductor, ‘intermediate’, and
‘small’ scars. On well-preserved internal moulds these three groups of scars are strongest

233

A. LEE MCALESTER : SYSTEMATICS, ETC., OF BAB1NKA

in sharply raised areas at their ventral extremities (the raised areas on the internal moulds
represent strong depressions on the original shell interior). Extending dorsally from the
raised extremities are more faintly raised ‘tails’, which converge toward the umbonal
region (text-fig. 1). These ‘tails’ are the traces of the position of the muscle scars at
earlier stages of growth. In most bivalves the earlier muscle attachment sites are com-
pletely obscured by later deposition of inner shell material, but in Babinka this later
deposition was not thick enough to cover completely the earlier trace of the muscle
scars. Some workers have assumed that muscles attached along the entire elongate im-
pression, but it is now apparent that the functional muscle at any one time occupied

text-fig. 1. Muscle-scar pattern in Babinka, aam, anterior adductor
muscle scar; ei, ‘elongate impression’; im, ‘intermediate’ (pedal)
muscle scars; pam, posterior adductor muscle scar; pi, pallial line;
sm, ‘small’ (gill) muscle scars.

only the ventral extremity of each impression. The shape of the strongly raised extremi-
ties show that only the anterior adductor muscle was somewhat elongate in life. The
muscles which attached to the posterior adductor scar and to all of the ‘intermediate’
and ‘small’ scars were approximately round in cross-section.

The two largest muscle scars in Babinka occupy anterior and posterior marginal
positions along the line of pallial attachment, as do the adductor muscles in all iso-
myarian bivalves, and there is no reason to doubt that they are the attachment sites of
typical isomyarian adductor muscles. The function of the muscles which attached to the
‘intermediate’ and ‘small’ scars is more problematical.

The intermediate scars were considered by Barrande, Voices, and Horny to represent
the attachment sites of the pedal musculature. All recent isomyarian bivalves have
paired pedal muscles, with one muscle of each pair attaching to each valve (text-fig. 2).
These muscles commonly leave distinct shell impressions. In many recent isomyarian
bivalves the foot is anchored by only two strong pairs which attach immediately above
the adductor muscles (text-fig. 2; Crassatella, Codakia, Mercenaria, Tellina). Other
groups have additional strong pedal muscle pairs which attach and leave scars in the
central dorsal region. Living Nuculana and related protobranch forms commonly have
five or six pairs of pedal muscles (text-fig. 2, Nuculana), some living Cardiacea have
three strong pairs (text-fig. 2, Cardium), and some Mactracea have as many as five pairs

234

PALAEONTOLOGY, VOLUME 8

(text-fig. 2, Mesodesma). Recent isomyarian bivalves thus show considerable variation in
the number of pedal muscle pairs. Normally only two strong pairs attach above the
adductor muscles, but in several unrelated groups there are from one to four additional
pairs between the two principal pairs.

The ‘intermediate’ muscle scars of Babinka are similar in size and position to the
pedal scars of recent isomyarian bivalves. As in recent forms, two pairs of pedal muscles
attach directly above the adductors. The six additional pairs of muscles between the
adductors in Babinka are almost certainly analogous to the additional strong pedal

text-fig. 2. Pedal musculature of genera representing eight superfamilies of recent isomyarian bivalves.
Pedal-attachment sites shown for right valves only. The pattern is repeated in the left valves making
symmetrical right-left pairs of pedal muscles. Note the presence of 3 to 5 pedal attachment sites in
Nuculana, Cardium, and Mesodesma. Data from Allen, 1958; Heath, 1937; Pelseneer, 1891, 1911;

Yonge, 1939, 1949.

muscle pairs found in several unrelated recent superfamilies. These similarities strongly
suggest that the ‘intermediate’ muscle scars of Babinka do in fact represent the attach-
ment sites of pedal muscle pairs.

The pedal muscles of bivalves have completely different functions than do the adduc-
tor muscles, and it is most probable that the two kinds of muscles had separate evolu-
tionary origins. Because both the pedal and adductor impressions in Babinka show
a similar elongate shape, there has been a tendency to assume that the large adductors
represent two additional pairs of pedal muscles which have become hypertrophied. It is
much more likely, however, that the adductor muscles of the Bivalvia did not originate
from modification of the pedal musculature, but instead arose independently by cross-
fusion of the pallial attachment muscles, as has been convincingly stressed in the writings
of Yonge (1953, 1957). If this reasonable conclusion is correct, then the adductor muscles
in Babinka cannot be considered to represent additional pairs of modified pedal muscles.

The ‘small’ muscle scars and the faint ‘elongate impression’ have not been previously
recognized in Babinka, and it is the ‘intermediate’ muscle scars (which will hereafter be

235

A. LEE MCALESTER : SYSTEMATICS, ETC, OF BABINKA

termed the ‘pedal muscle scars’) which have led to the repeated suggestion that Babinka
is related to some metameric ancestral mollusc. Similar multiple pairs of pedal muscles
occur in several unrelated and divergently specialized groups of recent bivalves, and it is
therefore evident that the mere presence of such muscles does not indicate a primitive
condition. A strong suggestion of primitiveness is seen, however, from comparing the
pattern of pedal and ‘small’ muscle scars in Babinka with the muscle attachment pattern

ANTERIOR POSTERIOR

text-fig. 3. Comparison of muscle-scar patterns in Babinka and
the recent monoplacophoran Neopilina. a-h, Eight pairs of prin-
cipal pedal muscles. The small dots below the pedal muscles show
the position of gill attachment muscles in Neopilina, and the posi-
tion of the ‘small’ muscle scars in Babinka. Data on Neopilina
from Lemche and Wingstrand, 1959.

of Neopilina, the only recent representative of the primitive molluscan Class Mono-
placophora.

Lemche and Wingstrand have provided detailed descriptions of the pattern of muscle
attachment to the shell of Neopilina galatheae; in that species the foot and visceral mass
attach by eight strong pairs of pedal muscles (Lemche and Wingstrand, 1959, figs. 120,
121, 130). Associated with these eight pedal muscle pairs are a series of much smaller
muscles having various functions, including pallial, ctenidial, radular, and visceral
muscles. Among the strongest of these small muscles are the ctenidial muscles, which
serve to attach the gills to the shell. N. galatheae has five pairs of gills which attach
to the shell by many small muscles situated around the third through seventh pairs of
larger pedal muscles (text-fig. 3).

Lemche and Wingstrand (1959, fig. 133) note that the muscle pattern in N.
galatheae is closely analogous to the strong muscle-scar pattern of the Silurian mono-
placophoran genus Pilina. Like Neopilina , this early fossil probably had eight strong pairs
of pedal muscles and associated smaller ctenidial, radular, pallial, and visceral muscles.

236

PALAEONTOLOGY, VOLUME 8

This eight-paired pattern is not universal in the Monoplacophora, for some other early
Palaeozoic genera show fewer than eight pedal muscle pairs. It may be significant,
however, that eight appears to be the maximum number of pedal muscle pairs found in
any monoplacophoran, and in some species the smaller number of pedal scars appear to
have resulted from fusion of originally more numerous pairs.

The pattern of pedal and ‘small' muscle scars in Babinka shows an amazing similarity
to the pattern of pedal and ctenidial muscle attachment in Neopilina (text-fig. 3). As in
Neopilina, Babinka has eight pairs of pedal muscle scars. Even more strikingly, the
‘small' muscle scars of Babinka occur under the third to seventh pairs of larger pedal
muscle scars in the same position as the ctenidial attachment muscles in N. galatbeae.
Although it is tempting to draw immediate phylogenetic conclusions from these
similarities, several facts suggest that the relationships may not be as simple as they
first appear.

First, the detailed pattern of muscle impressions in Babinka is clearly visible on only
one unusually well-preserved internal mould of a right valve from the collections of the
National Museum at Prague (Plate 28, figs. 9-11). This is the only specimen which
shows the ‘small’ muscle scars and all eight pedal muscle scars. The central pairs of
pedal scars are preserved on many specimens, and these have been the source of the
previous speculation regarding the muscle-scar pattern of the genus. A few specimens
also preserve either the anterior or posterior pedal scar above the adductors, but only
the single Prague specimen clearly preserves all eight pairs. It is therefore impossible to
fully evaluate the variability in number and position of the pedal and ‘small’ muscle
scars. Composite evidence from many specimens suggests a reasonably constant pedal
muscle pattern, but the variability of the ‘small’ muscles is completely unknown.

Further difficulties are raised by the presence of fewer than eight pedal muscle pairs in
many fossil monoplacophorans, and also by the occurrence of six instead of five gill
pairs in a second recent species of Neopilina , N. ewingi Clarke and Menzies (1959). The
anatomical details of this species have not yet been described, but it is probable that it
has a somewhat different pattern of gill muscle attachment than does N. galatbeae.
In spite of these cautions and qualifications, I feel that the muscle patterns in Neopilina ,
Babinka , and some early fossil monoplacophorans are too similar to be entirely the
result of chance, and I believe it is reasonable to infer that the pedal and ‘small’ muscle
scars in Babinka do in fact represent an inheritance from some monoplacophora-like
ancestor. It will be stressed later that in all features except the pedal and ‘small’ muscles,
Babinka is a typical representative of the Class Bivalvia.

Implicit in the above comparisons is the suggestion that the ‘small’ muscle scars in
Babinka represent the site of attachment of the gills. This possibility is raised not only
by the similar position of these scars and the gill muscles of Neopilina, but also by the
observation that no other large organs are likely to have been attached to the shell in the
position of the ‘small’ scars. Direct gill attachment to the shell by many small muscles
has no obvious analogue in recent bivalves, but the position of the scars in Babinka is
geometrically correct to have supported a ctenidial structure in the mantle cavity. In
addition, the many separate pedal scars of Babinka suggest that the animal still lacked
the united pedal-visceral muscle system which supports the gills in most recent bivalves.
A strong direct attachment of the gills to the shell may therefore have still been neces-
sary. It is most probable that the ‘small’ scars were the sites of gill attachment, and they

A. LEE MCALESTER : SYSTEM ATICS, ETC., OF BABINKA 237

will henceforth be referred to as the ‘gill muscle scars’. The many small muscles which
attached to these scars may have supported a single large gill or, less probably, they
might represent the attachment sites of several small gills.

The two final internal scars preserved in Babinka are the non-sinuate pallial line, and
the faint ‘elongate impression’ below the anterior adductor. Both of these features sug-
gest a relationship between Babinka and the bivalve Superfamily Lucinacea, and they
will be considered in detail in the next section.

BABINKA AS AN ANCESTRAL LUCINOID BIVALVE

The oldest undoubted lucinoid bivalves are found in Middle and Upper Silurian lime-
stones on the island of Gotland, Sweden. Two lucinoid species are found in abundance
in the Gotland deposits (Hede, 1921; Munthe et a/., 1925; Haffer, 1959). One species,
for which the correct name is probably Paracyclas hisingeri (Murchison and Verneuil),
is a small, rounded, inflated form which is similar in shape to recent species of the
lucinoid family Diplodontidae. The other Gotland species, Ilionia prisca (Hisinger), is
a much larger, compressed form which closely resembles some recent species of the
family Lucinidae. The internal morphology of P. hisingeri is poorly known, but the larger
species, I. prisca , is found principally as internal moulds which preserve some muscle
scar impressions. All of the morphologic features of I. prisca are strongly characteristic
of recent Lucinacea (Allen, 1958). Among the similarities are: an extremely elongate
anterior adductor muscle; an unusual anteriorly-expanded shell shape; and a unique
radial shell groove near the dorsal valve margin which corresponds to the internal line of
attachment of the gill to the visceral mass. The presence of these distinctive lucinoid
characteristics in I. prisca makes it extremely probable that the species is closely related
to recent Lucinacea. This superfamily was therefore fully established in mid-Silurian
time. The group has a scattered but continuous fossil record after the Silurian, and is
represented in modern oceans by about two dozen genera which are usually assigned to
three families. This abrupt appearance of fully developed and essentially modern lucinoid
bivalves in Middle Silurian deposits indicates that the group must have had a con-
siderable evolutionary history before the Silurian, but as yet no possible ancestral or
related fossil forms have been recognized in older deposits. Many morphologic features
of Babinka strongly suggest lucinoid affinities and these, coupled with its occurrence in
Middle Ordovician rocks, make it both morphologically and stratigraphically an ideal
ancestor for such Silurian lucinoids as Ilionia. The morphologic features of Babinka
which are strongly suggestive of lucinoid affinities are: (1) the characteristic anteriorly
expanded shell shape, (2) the elongate anterior adductor muscle scar and associated
‘elongate impression’, (3) the simple, non-sinuate pallial line, and (4) the typical lucinoid
hinge, dentition, and ligament. In short, the only features of Babinka which are not
typically lucinoid are the primitive patterns of pedal and gill muscle scars.

Comprehensive studies of living Lucinacea (Allen, 1958, 1960) have shown that the
characteristic anteriorly-extended shape and elongate anterior adductor muscle are
related to an unusual mode of life found in all recent representatives of the group.
Instead of drawing respiratory and feeding currents into the mantle cavity through
posterior siphons, as do most deeply buried infaunal bivalves, the Lucinacea construct
a unique mucous-lined, anterior inhalent tube in the surrounding sediment by means of

238

PALAEONTOLOGY, VOLUME 8

the elongate, cylindrical foot. The characteristic elongate anterior adductor muscle is
directly related to this habit for the solid outer face of the muscle is ciliate and serves
as a preliminary sorting area for food particles brought in by the anterior inhalent
current (text-fig. 4). This unusual habit is universal in living lucinoids, and was almost
certainly shared by Silurian and Devonian lucinoids which show the characteristic
elongate anterior adductor scar.

The anteriorly expanded, flattened shell of Babinka is similar in shape to the Silurian

text-fig. 4. Life position of recent lucinacean bivalves (modified from Allen, 1958). Nutrient-laden
water is brought into the mantle cavity through a mucous-lined anterior inhalent tube constructed by
the foot. In some genera the posterior exhalent current discharges directly into the sediment, in others
it is channelled to the sediment surface through a retractable posterior siphon. The anterior face of
the elongate anterior adductor muscle is covered with cilia and acts as a preliminary sorting area for

incoming food particles.

species I. prisca (text-fig. 5; PI. 27, figs, 6, 7). Even more suggestive is the pattern of
the anterior adductor muscle and associated ‘elongate impression’ in Babinka. The
anterior adductor scar of Babinka is considerably more elongate in a radial direction
than is the posterior scar, although it does not yet show the extreme ventral elongation
seen in Ilionia and most younger lucinoids. The radial elongation of this muscle in
Babinka does, however, suggest the beginning of a trend toward increasing the surface
area of the anterior adductor. Furthermore, several internal moulds of Babinka show
a faint ‘elongate impression’ marking the site of an obscurely bounded depression ex-
tending ventrally from the anterior adductor on the original shell interior. This ‘elongate
impression’ has exactly the same shape and position as the elongate anterior adductor
muscle scar in Ilionia and most other lucinoids (text-fig. 5; PI. 28, figs. 10, 12). The im-
pression is too faint to represent an expansion of the actual adductor muscle, but it does
indicate that there was some differentiation and specialization of the mantle in the
region below the anterior adductor in Babinka. The ‘elongate impression’ might

A. LEE MCALESTER : SYSTEMATICS, ETC., OF BABINKA

239

B a b i n k a

Ancestral mollusc

text-fig. 5. Proposed evolutionary relations of Babinka. Note the progressive
reduction of the pedal muscles and the expansion of the anterior adductor
muscle between Babinka and Ilionia.

reasonably represent the attachment surface of some kind of specialized ciliary sorting
area which was similar in position and function to the elongate adductor muscle surface
in later lucinoids. This sorting area would probably not have formed a connected parti-
tion between the valves and would have been less efficient than the sorting tube formed
by the solid face of an elongate adductor muscle. It is not difficult, therefore, to visualize

240

PALAEONTOLOGY, VOLUME 8

an evolutionary progression between Babinka and Ilionia involving an expansion of the
adductor into the position of the ‘elongage impression’. It is also worth noting that
Ilionia shows a rounded posterior adductor scar with an elongate trace of the earlier
growth position which is almost identical to that of Babinka (text-fig. 5). Regrettably, all
of the internal moulds of Ilionia available for comparison in the Yale Peabody Museum
collections are too poorly preserved to show the details of the pedal muscle scars, but
the genus most probably had the typical lucinoid pattern with one pair of strong pedal
muscles above each adductor.

Because of the anterior inhalent tube and consequent absence of a posterior inhalent
siphon, the Lucinacea are unusual among deeply buried infaunal bivalves in lacking
a pallial sinus for siphon retraction. Some recent lucinoids do have a small posterior
exhalent siphon, but this single siphon is retracted by a unique inside-out inversion
which does not require an indentation in the line of pallial muscle attachment. True
lucinoid bivalves all lack a pallial sinus, and it is suggestive that Babinka also shows
a non-sinuate line of pallial muscle attachment.

In many fossil and recent lucinoid species the hinge teeth are poorly developed or
absent, but when present the dentition consists of a large, commonly lobed, cardinal
tooth in the right valve fitting between two smaller teeth in the left valve (PI. 28, figs.
5-8; Allen, 1960; Chavan, 1937-8, 1962). In some genera lateral teeth and an
additional small cardinal tooth in the right valve are added to this basic pattern.
Several internal moulds of Babinka preserve impressions of the dentition, and latex casts
of these impressions clearly show the original dental pattern of the genus to have been
identical to the basic dentition of the Lucinacea (Plate 28, figs. 1-8). As in recent
lucinoids, Babinka has a large, lobed tooth in the right valve which fits between two
smaller teeth in the left valve.

The ligament in the Lucinacea is opisthodetic; the principal ligament elements nor-
mally occupy an obscure groove in the hinge plate posterior to the cardinal dentition.
In addition, the dorsal-hinge region posterior to the umbones normally shows a slight
gape where elements of the ligament were exposed on the surface. Anterior to the um-
bones, recent lucinoids commonly show a well-developed lunule. The hinge region and
ligament attachment area in Babinka show this same pattern. As in recent lucinoids,
Babinka has a faint ligament groove and ligament gape posterior to the cardinal denti-
tion, and the genus also shows the characteristic anterior lunule (PI. 27, figs. 1, 2). In all
features of the hinge and ligament, Babinka is a typical lucinoid bivalve.

LIFE HABITS

Silurian and younger fossil lucinoid bivalves almost certainly shared the adaptations
for deeply buried suspension feeding seen in all recent Lucinacea (Allen, 1958) because
the fossils show characteristic morphologic features, such as the elongate anterior adduc-
tor scar, which are directly related to that mode of life. Although the evidence is less
conclusive for Babinka, it seems likely that it had similar habits.

The strong multiple pedal attachment muscles in Babinka suggest a large active foot
which had probably only begun to develop the extremely extensible, cylindrical form of
later Lucinacea. If this were the case, then Babinka would have been a rather shallow
burrower, for the depth of burial in the Lucinacea is controlled by the degree of ex-
tensibility of the foot. Babinka may not have had the ability to form a distinct, mucous-

A. LEE MCALESTER: SYSTEMATICS, ETC., OF BABINKA

241

lined inhalent tube, but instead might have merely used repeated extrusions of the foot
to maintain a crude anterior opening to the surface through a relatively thin cover of
overlying sediment (text-fig. 6). Such habits would be a likely early stage in the develop-
ment of the typical lucinoid anterior inhalent tube.

As discussed earlier, the ‘elongate impression’ below the anterior adductor in Babinka
may represent a specialized area of the mantle which functioned as a preliminary ciliary
sorting area for food particles brought in by the anterior inhalent current. A specialized
sorting area on the mantle below the adductor is a likely preliminary step in the adaptive

text-fig. 6. Inferred life position of Babinka. Compare with text-fig! 4.

trend leading to dorsal extension of the adductor muscle in the same position. The gills
of Babinka were probably already functioning as food-gathering organs which filtered
particles directly from the incoming water, and which also received food from ciliary
tracts on the surface of the mantle and visceral mass.

In summary, Babinka probably was a buried suspension feeder which lived just below
the surface of the sediment, drawing in nutrient-laden water through an anterior depres-
sion in the sediment surface maintained by extrusion of the foot. The foot was probably
strong and active, enabling the animal to burrow and move through the substrate with
ease. The animal probably fed by ciliary trapping and sorting of small food particles on
the surface of the mantle, visceral mass, and gills.

SYSTEMATIC DESCRIPTIONS
GENUS BABINKA BARRANDE 1881

Type species, by monotypy and subsequent designation of Ruzicka andPrantl 1960 (p. 48), Babinka prima
Barrande 1881, pi. 266, figs, vi, 1-16.

Discussion. The genus is known only from the type species found in the lowest Middle
Ordovician of the Bohemian Basin, Czechoslovakia. The alternative generic name
Anuscula was simultaneously proposed by Barrande and should be treated as a junior
objective synonym of Babinka. The name Babinka is undoubtedly valid because no
additional generic or specific names have been proposed for the type species, and there
are no other named species which are likely to be subjectively synonymous with Babinka
prima.

242

PALAEONTOLOGY, VOLUME 8

Babinka prima Barrande

Plate 26, figs. 3-12; Plate 27, figs. 3-5; Plate 28, figs. 1-4, 9-14

Babinka prima Barrande, 1881, pi. 266, figs, vi, 1-16. Vokes, 1954, p. 234, fig. 1. Cox, 1959,
p. 204, fig. 2. Ruzicka and Prantl, 1960, p. 48. Horny, 1960, p. 480, pi. 1, figs. 1, 2. Vogel,
1962, p. 235, pi. 5, figs. 5-6. [not?] Thoral, 1935, p. 162, pi. 13, figs. 4-5.

Revised description. Medium-sized, anteriorly extended, compressed, equivalved bivalves
showing considerable shape variability (probably exaggerated by post-depositional dis-
tortion in many specimens). Surface sculpture of fine, concentric ridges (PL 26, figs.
3-6). One articulated specimen preserving the original shell material shows prominent
lunule (PI. 27, fig. 2). Dentition consisting of one large, triangular, cardinal tooth with

EXPLANATION OF PLATE 26

All specimens of Babinka prima are from concretions from the Sarka Beds (Middle Ordovician) at
Osek, near Rokycany, Czechoslovakia.

Figs. 1, 2. Codakia ( Ctena ) sp. Recent, Bikini Island, X 1-5, showing characteristic anteriorly-extended
shape of the Lucinacea. 1, Right valve, YPM 23869. 2, Left valve, YPM 23868.

Figs. 3-6. Babinka prima Barrande. A series of latex casts of natural external moulds, showing the
shape and sculpture of the original valve exteriors. 3, Left valve, YPM 23858, X 2 (see also PI. 28,
fig. 13). 4, Right valve, NMP CD228a (cast shown in photograph deposited in YPM), X 1-5. 5, Left
valve, NMP CD228b (cast shown in photograph deposited in YPM), X 1 -5. 6, Left valve, unnumbered
NMP paratype (cast shown in photograph deposited in YPM), originally figured by Barrande, 1881,
as fig. vi, 16 of plate 226, X 1-5 (see also fig. 11 below).

Figs. 7-12. Babinka prima Barrande. A series of internal moulds of right valves (7-9) and left valves
(10-12), x2. 7, Unnumbered MCZ specimen (see also PI. 28, fig. 12). 8, Unnumbered NMP para-
type, originally figured by Barrande, 1881, as figs, vi, 10-12 of plate 266. 9, Unnumbered NMP
paratype, originally figured by Barrande, 1881, as figs, vi, 13-15 of plate 266. 10. Lectotype, NMP
CD229a, originally figured by Barrande, 1881, as figs, vi, 7-9 of plate 266, and refigured by Horny,
1960, as fig. 2 of plate 1.11, Unnumbered NMP paratype, the external mould of this specimen was
originally figured by Barrande, 1881, as fig. vi, 16 of plate 266 and is reillustrated here as fig. 6
above. 12, Paratype, NMP CD229b, originally figured by Barrande, 1881, as figs, vi, 4-6 of plate 266.

MCZ = Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts, U.S.A.

NMP = Narodni muzeum v Praze, Prague, Czechoslovakia. YPM = Peabody Museum of Natural

History, Yale University, New Haven, Connecticut, U.S.A.

EXPLANATION OF PLATE 27

All specimens of Babinka prima are from concretions from the Sarka Beds (Middle Ordovician) at
Osek, near Rokycany, Czechoslovakia. Abbreviations as on explanation of Plate 26.

Fig. 1. Codakia orbicularis (Linne). Recent, Barbados, West Indies, YPM 23865, dorsal view showing
posterior ligament gape (dark area to left of umbones) and lunule (to right of unbones), X2.

Figs. 2-5. Babinka prima Barrande. 2, Dorsal view of unnumbered NMP specimen preserving original
shell material, showing posterior ligament gape (dark area to left of umbones; internal matrix
darkened on photograph for contrast) and lunule (to right of umbones), X 3. 3, Right view of
specimen shown in fig. 2, X 3. 4, Unnumbered NMP specimen, a well-preserved internal mould of
a left valve (photo, by Prof. N. D. Newell), X 4. This specimen was figured by Horny, 1960, as fig. 1
of plate 1 . 5, Unnumbered NMP specimen, a well-preserved internal mould of a right valve showing
the complete pattern of muscle impressions, X 3. Other photographs of this specimen are included
on Plate 28 (figs. 4, 9-11).

Figs. 6, 7. Ilionia prisca (Hisinger). Silurian, Gotland, Sweden. 6, YPM 23870, right view of articulated
internal mould showing general shape and elongate anterior adductor muscle scar, X 1. 7, YPM
23871, left view of articulated internal mould showing general shape and adductor muscle scars, X 1.

Palaeontology , Vol. 8

PLATE 26

McALESTER, Ordovician lucinoid Babinka

Palaeontology, Vol. 8

PLATE 27

McALESTER, Ordovician lucinoid Babinka

A. LEE MCALESTER : SYSTEMATICS, ETC., OF BABINKA 243

raised median ridge in right valve, two smaller wedge-shaped cardinal teeth in left
valve; lateral surfaces of teeth covered with fine, widely spaced parallel ridges (PI. 28,
figs. 1-4). Short, weak ligament groove along dorsal margin posterior to cardinal teeth
(PI. 28, figs. 1-4); slight ligamental gape along dorsal margin behind umbones (PI. 27,
fig. 2). Prominent anterior and posterior adductor muscle scars; posterior scar strongest
in rounded area at ventral extremity, anterior scar strongest in elongate-oval area at
ventral extremity (PI. 28, figs. 9, 10, 12, 14). Faint, elongate, crescent-shaped depression
(‘elongate impression’) extending ventrally from anterior adductor scar (PI. 28, figs.
10, 11; text-fig. 1). Irregularly rounded pedal muscle scars adjoin dorsal side of adductor
scars. Six additional pairs of pedal muscle scars between the adductors, becoming much
stronger in rounded area at ventral extremity (PI. 28, figs. 9, 10; text-fig. 1). Adductor
and some pedal scars show faint concentric growth increments parallel to the advancing
margin (PI. 28, figs. 9, 10). Five posterior pairs of six central pedal scars bounded
ventrally by about twenty-five smaller and more closely spaced, rounded muscle scars
which probably represent the site of gill attachment (PI. 28, figs. 9-11; text-fig. 1).
Pallial line non-sinuate (PI. 28, fig. 14). Original shell material preserved on a single
articulated specimen showing fine, radial structural elements (PI. 27, fig. 3); shell
material and structure otherwise unknown.

Types. Lectotype, by designation of Ruzicka and Prantl (1960, p. 48), the left valve
shown by Barrande, 1881, as figures vi, 7-9 of plate 266; No. CD 229a in the Barran-
deum collections of the Narodni muzeum v Praze [National Museum], Prague, Czecho-
slovakia. Type locality : Vosek [= Wosek, Osek], Czechoslovakia. Stratigraphic position :
dj [Sarka beds], Middle Ordovician [approximately Llanvirn]. Paratypes: one left valve
and two right valves glued to the same plaque and bearing the same number as the
lectotype, and two additional unnumbered right valves and one unnumbered left valve
in the same collection. These seven specimens were figured by Barrande, 1881, as figures
vi, 1-6 and 10-16 of plate 266. All were collected at the same horizon and locality as the
lectotype.

Material. This revised description was based on personal examination of about 100
specimens. These include the lectotype, paratypes, and about ninety additional speci-
mens in the following collections: Narodni Muzeum, Prague, Czechoslovakia (about
eighty specimens); Yale University Peabody Museum, New Haven, Connecticut, U.S.A.
(six specimens); Museum of Comparative Zoology, Harvard University, Cambridge,
Massachusetts, U.S.A. (one specimen). Dr. Radvan Horny of Prague informs me that
the siliceous concretions in which Babinka occurs can rarely be collected today from the
Sarka beds. The 100 studied specimens represent the bulk of the previously collected
examples of B. prima, and it is unlikely that the species will ever be known from many
additional Bohemian specimens.

Occurrence and preservation. The species is known with certainty only from the Sarka
beds (approximately Llanvirn or lowest Middle Ordovician, see Horny, 1963) in the
vicinity of Prague, Czechoslovakia. An occurrence of the species in Lower Ordovician
rocks of southern France has also been reported (Thoral, 1935), but the illustrated
specimens in Thoral’s report cannot be positively recognized as B. prima , and the
occurrence will require additional study before it can be accepted without question.

244

PALAEONTOLOGY, VOLUME 8

In addition to Babinka, the fauna of the Sarka beds includes the bivalve genera ‘Leda\
‘ Nucula and Redonia , as well as species of monoplacophorans, gastropods, hyolithids,
ostracods, and trilobites (Barrande, 1881; Horny, 1963). The fauna occurs in small,
dark, rounded, siliceous concretions which must be broken open to reveal the fossils.
Most specimens in these concretions are preserved as sharp internal and external moulds
separated by a space representing the site of the original shell material. All the examined
specimens of Babinka are preserved in this way except for a single articulated specimen
which preserves somewhat abraded and recrystallized original shell material (PI. 27,
figs. 2, 3). In most specimens of Babinka the concave external mould was discarded when
the concretions were originally broken, and only the convex internal moulds, showing
impressions of the internal shell features, are found in the collections. Fortunately, a few
of the external moulds were not discarded, and latex casts of these provide additional
knowledge of the external shape and sculpture of the valves. Most specimens of Babinka
are disarticulated but otherwise well-preserved single valves; right and left valves occur
in about equal abundance. Approximately 10 per cent, of the specimens preserve both
valves in close articulation.

CONCLUSIONS

1 . A reconsideration of the complex muscle-scar pattern of the problematic Middle
Ordovician bivalve Babinka (Barrande, 1881) indicates that the adductor and pallial
muscles of the genus were similar to those found in most other isomyarian bivalves.

EXPLANATION OF PLATE 28

All specimens of Babinka prima are from concretions from the Sarka Beds (Middle Ordovician) at
Osek, near Rokycany, Czechoslovakia. Abbreviations as on explanation of Plate 26.

Figs. 1-4. Babinka prima Barrande. A series of latex casts of the hinge regions of internal moulds,
showing the nature of the original dentition. 1, Left valve, YPM 23862, x 5. 2, Right valve, YPM
23861, X 5. 3, Left valve, paratype NMP CD229b (cast shown in photograph deposited in YPM;
sockets darkened on photograph for contrast; see also Plate 26, fig. 12), X 5. 4, Right valve, un-
numbered NMP specimen (cast shown in photograph deposited in YPM). Other photographs of the
same specimen are figs. 9-11 of this plate, and fig. 5 of Plate 27.

Figs. 5, 6. Diplodonta punctata Say. Recent, Kingston Flarbour, Jamaica, dentition, x 8, showing
similarity in tooth pattern between Babinka and the lucinoid Family Diplodontidae. 5, Left valve,
YPM 23863. 6, Right valve, YPM 23864.

Figs. 7, 8. Divaricella sp. Recent, New Zealand, dentition, x4, showing similarity in tooth pattern
between Babinka and the cardinal dentition of the Family Lucinidae. 7, Left valve, YPM 23866.
8, Right valve, YPM 23867.

Figs. 9-14. Babinka prima Barrande. 9-11, Three views of unnumbered NMP internal mould of
a right valve showing the complete pattern of muscle-scar impressions. Other photographs of this
same specimen are fig. 4 of this plate and fig. 5 of Plate 27. Fig. 9 shows the posterior adductor-
pedal scar and the central pedal and gill scars, X 5. Fig. 10 repeats the central scars and shows the
anterior adductor-pedal scar and ‘elongate impression’, x 5. Fig. 1 1 is an enlargement of the fourth
pedal scar and associated gill scars, X 10. 12, Unnumbered MCZ specimen, X 5, internal mould of
right valve showing anterior adductor muscle scar and ‘elongate impression’ (see also PI. 26, fig. 7).
13, YPM 23858, latex cast of dorsal region of articulated internal mould, showing nature of dental
articulation as viewed from the interior side of closed valves. The base of the single large right valve
tooth projects downward between the two smaller teeth of the left valve, X 5 (a cast of the external
mould of this specimen is shown in fig. 3 of PI. 26). 14, Unnumbered NMP internal mould of left
valve, showing posterior pedal muscle scars, posterior-adductor pedal muscle scar, and non-sinuate
posterior pallial line, X 3.

Palaeontology , Vol. 8

PLATE 28

McALESTER, Ordovician lucinoid Babinka

245

A. LEE MCALESTER : SYSTEMATICS, ETC., OF BABINKA

Babinka differs from other bivalves in having eight well-developed pairs of regularly
spaced pedal muscles above and between the adductors instead of the more usual pattern
of two to five pedal muscle pairs. In addition, Babinka shows a unique linear series of
much smaller scars below some of the pedal muscle pairs. These scars probably mark the
site of gill muscle attachment.

2. The multiple pedal and gill musculature of Babinka may reasonably be considered
to represent an inheritance from some monoplacophora-like ancestor as has been re-
peatedly suggested, for the pattern of these muscles is almost identical to that found in
recent Neopilina and some early fossil Monoplacophora. In all morphologic features
other than the pedal and gill muscles, Babinka is a typical isomyarian bivalve.

3. Both stratigraphically and morphologically, Babinka is an ideal ancestral form for
the first lucinoid bivalves which appear abruptly in Middle Silurian deposits. Morpho-
logical features of Babinka which are strongly suggestive of lucinoid affinities are:
(1) the characteristic anteriorly expanded shell shape, (2) the elongate anterior adductor
muscle and associated ‘elongate impression’, (3) the simple, non-sinuate pallial line,
and (4) the typical lucinoid hinge, dentition, and ligament. Babinka differs from other
lucinoid bivalves only in the pattern of pedal and gill muscle scars.

4. Recent lucinoid bivalves all share adaptations for life as deeply buried suspension
feeders receiving nutrient-laden water into the mantle cavity through a unique, mucous-
lined, anterior inhalent tube constructed by the foot. The morphology of Babinka sug-
gests that it was also an infaunal suspension feeder with an anterior inhalent current.
The large foot probably served not only for locomotion and digging, but also to main-
tain an anterior inhalent opening to the surface of the sediment.

5. Babinka is one of the first bivalves to appear in the fossil record. The genus probably
represents a transitional evolutionary link between the successful Silurian to recent
bivalve Superfamily Lucinacea and some monoplacophora-like molluscan ancestor
(text-fig. 5). The transitional evolutionary position of Babinka indicates that lucinoid
bivalves arose independently from a non-bivalved ancestor, and raises fundamental
questions regarding the evolutionary history and classification of the Bivalvia. These
more general topics will be the subject of a separate paper (McAlester, 1965).

Acknowledgements. I am indebted to Drs. N. D. Newell, E. L. Yochelson, and H. B. Whittington for
the loan of specimens, casts, and photographs, and to Dr. J. A. Allen for stimulating discussions
regarding the possible lucinoid affinities of Babinka. I am also extremely grateful to Drs. Radvan Horny
and Vlastislav Zazvorka for their generous co-operation and many kindnesses during my visit to
Prague in June 1963. Drs. Copeland MacClintock, N. D. Newell, and J. H. Ostrom read the manu-
script and offered many helpful suggestions. The text-figures and photographs were prepared by Mr.
D. W. Harvey, Mr. D. M. Keith, Mrs. Martha Erickson, and Miss Martha Dimock. This study was
supported in part by grant no. G19961 from the National Science Foundation, and in part by the
Charles Schuchert Fund of the Peabody Museum, Yale Cffiiversity.

Note added in proof. After the section above on ‘Occurrence and preservation’ was prepared, I have
had an opportunity to visit the University of Montpellier, France, where Professor J. Avias kindly
made available the specimens from southern France described as Babinka prima by Thoral (1935).
Thoral’s collections contain about three dozen small internal moulds which are almost certainly
conspecific with the Bohemian specimens of B. prima. The specimens are all poorly preserved and
contribute no new morphologic information, but they do extend both the geographic and strati-
graphic range of the species. Most of Thoral’ s material was collected in the ‘neighbourhood of St.
Chinian’ (Montagne Noire region) from rocks which he considered to be of upper Tremadoc age.
These specimens may therefore be slightly older than the Bohemian occurrence of B. prima.

246

PALAEONTOLOGY, VOLUME 8

REFERENCES

allen, J. A. 1958. On the basic form and adaptations to habitat in the Lucinacea (Eulamellibranchia).
Phil. Trans. 241(B), 421-84.

1960. The ligament of the Lucinacea (Eulamellibranchia). Quart. J. micr. Sci. 101, 25-36.

barrande, j. 1881. Systeme silurien (lit centre de la Boheme, volume 6, classe des Mollusques, ordre des
Acephales. Prague and Paris [in 4 vols.].

chavan, a. 1937-8. Essai critique de classification des Lucines. J. Conchyliol. 81, 133-53, 198-216,
237-82; 82, 59-97, 105-30, 215-43.

1962. Essai critique de classification des Ungulinidae. Bull. Inst. roy. Sci. nat. Belgique, 38, no. 23.

clarke, A. h. and menzies, r. j. 1959. Neopilina [Verna) ewingi, a second living species of the Paleozoic
Class Monoplacophora. Science, 129, 1026-7.

cox, l. r. 1959. The geological history of the Protobranchia and the dual origin of taxodont Lamelli-
branchia. Proc. malac. Soc. London, 33, 200-9.

1960. Thoughts on the classification of the Bivalvia. Ibid. 34, 60-88.

haffer, j. 1959. Der Schlossbau fruh-heterodonter Lamellibranchiaten aus dem rheinischen Devon.
Palaeontographica, 112(A), 133-92.

heath, h. 1937. The anatomy of some protobranch mollusks. Mem. Mus. Hist. nat. Belgique, 10, (2).
hede, j. e. 1921. Gottlands silurstratigrafi. Sverig. geol. Unders. Afh. 305(C) (Arsb. 14, 1920, No. 7).
horny, r. 1960. On the phylogeny of the earliest pelecypods (Mollusca). Vestn. geol. Ust. csl. 35,479-82.

1963. Lower Paleozoic Monoplacophora and patellid Gastropoda (Mollusca) of Bohemia.

Shorn. Ustredniho Ust. geol., oddil paleontologicky, 28.
lemche, h. and wingstrand, k. g. 1959. The anatomy of Neopilina galatheae Lemche, 1957 (Mollusca
Tryblidiacea). Galathea Report, Copenhagen, 3, 1-71, pi. 1-56.
mcalester, a. l. 1965. Evolutionary and systematic implications of a transitional Paleozoic lucinoid
bivalve (in preparation).

merklin, r. l. 1962. Ob odnoy novoy sisteme dvustvorchatykh mollyuskov. Bull. Soc. nat. Moscou,
otdel geologicheskiy, 37(3), 136.

munthe, h., hede, j. e. and post, l. von. 1925. Gottlands geologi, en oversikt. Sverig. geol. Unders. Afh.
331(C), (Arsb. 18, 1924, No. 3).

pelseneer, p. 1891. Contribution a l’etude des lamellibranches. Arch. Biol., Paris, 11, 147-312, pi. 6-23.

191 1. Les lamellibranches de l’expedition du Siboga. Partie anatomique. Siboga-Expeditie Mon.,

Leiden, 53a.

ruzicka, b. and prantl, F. 1960. Types of some Barrande ‘s pelecypods (Barrandian). Zvldstni Otisk
Casopisu Ncirodniho Mus., oddil prirodovedny, 1, 48-55 [in Czech with English summary].
thoral, m. 1935. Contribution a l’etude paleontologique de l’Ordovicien inferieur de la Montagne
Noire et revision sommaire de la faune cambrienne de la Montagne Noire. Thes. Fac. Sci. Univ.
Paris, 1541(A), 2nd these.

vogel, k. 1962. Muscheln mit Schlosszahnen aus dem spanischen Kambrium und ihre Bedeutung fur
die Evolution der Lamellibranchiaten. Abh. Math.-nat. KI. Akad. Wiss. Mainz, Jahrg. 1962, 4.
vokes, h. e. 1954. Some primitive fossil pelecypods and their possible significance. /. Washington Acad.
Sci. 44, 233-6.

yonge, c. m. 1939. The protobranchiate mollusca; a functional interpretation of their structure and
evolution. Phil. Trans. 230(B), 79-147.

1949. On the structure and adaptations of the Tellinacea, deposit feeding Eulamellibranchia. Phil.

Trans. 234(B), 29-76.

■ 1953. The monomyarian condition in the Lamellibranchia. Trans, roy. Soc. Edinburgh, 62,

443-78.

1957. Mantle fusion in the Lamellibranchia. Pubbl. Staz. zool. Napoli, 29, 151-71.

A. LEE MCALESTER
Department of Geology,
Yale University,
New Haven,
Connecticut, U.S.A.

Manuscript received 15 June 1964

NEW SILURIAN GRAPTOLITES FROM THE
HOWGILL FELLS (NORTHERN ENGLAND)

by R. B. RICKARDS

Abstract. Recent work on the graptolite faunas of the Silurian strata of north-west Yorkshire and Westmorland
has unearthed, from the Wenlock and Ludlow Series, species previously recorded from the Continent, and in
addition several new species and subspecies. The following new forms are described : Monoclimacis griestonensis
nicoli subsp. nov., M. shottoni sp. nov., M. flumendosae kingi subsp. nov., Pristiograptus watneyae sp. nov.,
P. welchae sp. nov., P. dubius pseudolatus subsp. nov., P. auctus sp. nov., Monograptus finnus sedberghensis
subsp. nov., M. radotinensis inclinatus subsp. nov., M. minimus cautleyensis subsp. nov., M. danbyi sp. nov., M.
simulatus sp. nov.

The nineteen graptolites described in this paper form part of a large graptolite fauna
obtained from the Wenlock and Ludlow Series during a revision of the Silurian strati-
graphy of the Howgill Fells. Text-fig. 1 gives the ranges of the described species against
the zones recognized in the region, and the correlation of these zones with those estab-
lished by Elies (1900) and Wood (1900) in the Welsh Borders.

Eight of the described forms have previously been recorded only from the continental
countries where they are found at similar horizons. Monoclimacis flumendosae (Gortani)
occurs earlier in the Howgill Fells than in Czechoslovakia, where it first appears in the
rigidus Zone, but in both areas it ranges into the higher Wenlock strata. However,
whilst the species remains unchanged throughout its range in Czechoslovakia, the
typical form is replaced by the subspecies flumendosae kingi in the lundgreni Zone of the
Howgill Fells.

Monograptus minimus cautleyensis and M. finnus sedberghensis occur at approximately
the same horizons as the type subspecies in Czechoslovakia and may represent genuine
cases of geographical subspeciation. M. minimus s.s. and M. finnus s.s. have not been
recorded from the British Silurian.

Monograptus radotinensis inclinatus is best regarded as a chronological subspecies
since it is found in the zone above that from which Boucek’s radotinensis s.s. was re-
corded in Czechoslovakia.

Monoclimacis? haupti has previously been recorded only from erratic boulders (Kuhne
1955, Germany; Urbanek 1958, Poland) where it is associated with a nilssoni-scanicus
Zone fauna. The associated nilssoni-scanicus fauna in the Howgill, Barbon, and Middle-
ton Fells of northern England confirms the horizon given by the above authors.

Monoclimacis griestonensis nicoli and Pristiograptus dubius pseudolatus have probably
evolved from their respective type subspecies and reflect a tendency to increased robust-
ness of the rhabdosome in some monoclimacids and pristiograptids.

The new species described also have considerable significance particularly as regards
local stratigraphy. Monograptus danbyi, M. shottoni, and their associates, for example,
allow a correlation, not only throughout the Cautley district of the Howgill Fells, but
with the Cross Fell area to the north east of the main Lake District Silurian outcrop.
This association comprises the centrifugus Zone assemblage.

tPalaeontology, Yol. 8, Part 2, 1965, pp. 247-71, pi. 29-31.]

C 3009 s

248

PALAEONTOLOGY, VOLUME 8

In the systematic descriptions below distinction is always made between specimens
preserved in relief and flattened specimens. The term ‘flattened’ is used to describe
specimens which have been reduced by diagenetic processes to a filmy deposit on the
bedding plane, whilst the term ‘compressed’ is used to describe rhabdosomes (either
flattened or in relief) which have suffered tectonic deformation. Crustal shortening can

&

ELLES (1900)
WOOD (1900)

HOWGILL

FELLS

LUDLOW

leintwardinensis

leintwardinensis

L

tumescens

incipiens

scanicus

nilssoni

nilssoni-scanicus

. . ■

vulgaris

o

WENLOCK

lundgreni

lundgreni

__ n

ellesi

ellesi

linnarssoni

linnarssoni

rigidus

belophorus

T

1

riccartonensis

antennularius

riccartonensis

1 1 ' '

murchisoni

murchisoni

centrifugus

II 1 1 1 ■ 1 !

ZONES

SPECIES

gnicoli

linnarssoni

shottoni

danbyi

simulatus

m.cautleyensis

sp. A

watneyi

r. inclinatus

f. sedberghensis

flumendosae

d.pseudolatus

menegh ini

f. belophorus

pseudodubius

f. kingi

auctus

haupti

welchi

text-fig. 1. Ranges of the species described, plotted against the zones recognized in the Howgill Fells;
and the correlation of these zones with those established in the Welsh Borders. The Wenlock Zones of
Elies (1900) and the Ludlow Zones of Wood (1900) are shown in one column. For wutneyi and welchi

read watneyae and welchae respectively.

usually be detected by the presence on the bedding planes of distinct lineations, which
are, in fact, minute folds.

Several morphological terms used in the descriptions below require definition:

Width is the total rhabdosomal width inclusive of thecal hooks; length of theca! tube is measured
along a line midway between the ventral and dorsal walls, as seen in profile view; thecal spacing is
measured for a small number of thecae (1-3), and is then translated to ‘thecae per cm. ’, since it is con-
sidered that this gives a better idea of the change in cell size along the rhabdosome. Packham (1962)
seems to have used a similar technique.

The Howgill Fells form a most distinct topographical feature of high, rounded hills
extending from north-west Yorkshire into Westmorland, and may be contrasted with the
Carboniferous country to the east and north. They occupy a broadly triangular area
with the town of Sedbergh at the southern apex. Tebay forms the north-west limit of the

R. B. RICKARDS: NEW SILURIAN GR APTOLITES

249

fells and Ravenstonedale village the north-east. The village of Cautley, from the
vicinity of which many of the graptolites were obtained, is situated between Sedbergh
and Ravenstonedale on the eastern flanks of the fells. The area is covered by the fol-
lowing Ordnance Survey 6" sheets : SD69 NW,NE,SW, SE; SD79 NW, SW; NY60 NE,
NW, SW, SE; NY70 SW.

The appendix gives four-figure grid references to all the localities mentioned in the
systematic descriptions. All the type specimens are deposited in the Department of
Geology, Hull University. Abbreviations associated with catalogue numbers are as
follows: B.U. Birmingham University; HUR. Hull University Rickards’ Collection.

Acknowledgements. I should particularly like to thank Dr. J. W. Neale and Dr. I. Strachan for their
many helpful suggestions and whole-hearted support at all stages of this work.

SYSTEMATIC DESCRIPTIONS

Class graptolithina Bronn 1 846
Order graptoloidea Lapworth 1875
Suborder monograptina Lapworth 1 880
Family monograptidae Lapworth 1875
Genus monoclimacis Freeh 1897

Type species. Graptolites vomerinus Nicholson 1872, emend. Lapworth.

Diagnosis. Rhabdosome often long and more or less straight, though slight curvature
common proximally and rarer distally; ventral wall of each theca subsequent to th. 1,
usually with distinct excavation ; the infragenicular wall overhangs the apertural region
of the preceding theca; apertural region often appears to be ’hooked’, but in some
representatives, at least, this is a monofusellar structure growing from the geniculum of
the succeeding theca.

Monoclimacis griestonensis nicoli subsp. nov.

Plate 30, fig. 4

Holotype. HUR./8P/19, a proximal fragment in full relief with sicula preserved.

Other material. Five specimens preserved in relief.

Horizon and localities. Zone of C. centrifugus, Wenlock Series; Pickering Gill (8P), near Cautley.
Derivation of name. After J. Nicol, author of M. griestonensis.

Diagnosis. Rhabdosome known from fragments only, but probably quite short. Maxi-
mum breadth 0-3 mm. Thecae long, narrow tubes closely adpressed to the axis, num-
bering 9-9b in 10 mm.

Description. The subspecies is known only from short fragments, up to 1 cm. long,
which do not exceed 0-3 mm. in width. This width is achieved at a distance of 5 mm. from
the base of the sicula and thereafter the rhabdosome is parallel-sided.

The sicula is prominent, 1-5 mm. long, and its apex reaches almost to the level of the
aperture of th. 1. The sicular aperture measures 0-2 mm. across and is furnished with
a short, slim virgella. Th. 1 arises 0-4 mm. above the base of the sicula and is 1T7 mm.
long.

250

PALAEONTOLOGY, VOLUME 8

Thecal overlap is approximately one quarter. Each thecal tube grows almost parallel
to the axis for a distance of 0-5 mm. and then takes a slight bend towards the ventral side
which results in a shallow excavation. The thecae then grow once again at a low angle
of inclination.

Remarks. The diagnostic features considered above fall within the ranges of variation
given by Elies and Wood (1910) for griestonensis s.s. However, the specimen figured by
these authoresses as text-fig. 219a (B.U. 1556) has the thecae more closely spaced (13 in
1 0 mm.) than is indicated by their description ; and their fig. 6a. plate 41 , which occurs on
Nicol’s type slab, has 12-14 thecae per centimetre. (The only other fossils on the type
slab are two of Monograptus spiralis s.l. and another very poor specimen of M. gries-
tonensis.) Other specimens from the type locality of Grieston Quarry examined by the
writer also show a close spacing of the thecae in proximal region. M. griestonensis
nicoli differs, therefore, from the type subspecies in having its thecae more widely
spaced. In addition the rhabdosome is initially more robust and the excavation of the
ventral margin of the thecal wall is less. The sicula in nicoli is very prominent and almost
twice the size of that in griestonensis s.s.

M. g. nicoli has a similar thecal spacing to M. g. kettneri Boucek 19316, and M. g.
minuta Pribyl 1940a, and a rhabdosomal width akin to the latter. From both subspecies
nicoli differs, however, in its relatively robust initial portion, large sicula, and less con-
spicuous excavation. M. g. nicoli also differs from kettneri in its lack of dorsal curvature
in the proximal region.

Monoclimacis linnarssoni (Tullberg)

Plate 30, fig. 5; text-fig. 2a, b

1883 Monograptus linnarssoni Tullberg, p. 20, pi. 2, figs. 5-9.

Material. A single distal fragment in low relief, some 54 cm. long; two proximal ends in full relief
with sicula preserved, and five other fragments.

Horizon and localities. Zone of C. centrifugus ; Middle Gill (4M), Pickering Gill (lOP), Cautley.

Description. The distal fragment of the rhabdosome is quite straight and fully 5J cm.
long. At the most proximal point seen the rhabdosomal width is 0-9 1 mm. The rhabdosome
is almost parallel-sided and at the distal extremity is still only 1T7 mm. wide.

The thecae are sigmoidally curved tubes inclined to the axis at a low angle (approxi-

text-fig. 2. a, Monoclimacis linnarssoni (Tullberg), HUR./4M/72, long distal fragment in low relief,
centrifugus Zone, b, M. linnarssoni (Tullberg), EIUR./T OP/41, proximal end in relief with sicula pre-
served, centrifugus Zone, c, M? haupti (Ktihne), HUR./2W/35, slightly distorted specimen in full
relief, nilssoni-scanicus Zone, d, M. f. flumendosae (Gortani), HUR./17N/232, flattened proximal end
with rather short sicula, linnarssoni Zone, e, M. shottoni sp. nov., holotype, HUR./28W/76, complete
specimen in relief, centrifugus Zone, f Pristiograptus welchae sp. nov. holotype, HUR./lAb/22, com-
plete but flattened specimen, leintwardinensis Zone, g, P. watneyae sp. nov., holotype, HUR./37W/17,
proximal part of a long specimen in full relief, centrifugus Zone, h, P. auctus sp. nov., holotype,
HUR./7W/46, expanded virgella only poorly preserved, nilssoni-scanicus Zone, i, P. pseudodubius
(Boucek), HUR./26N/11, lundgreni Zone, j, P. meneghini (Gortani) HUR./17N/46, linnarssoni Zone.
k, P. dubius pseudolatus sp. nov., holotype, HUR./18N/53a, long flattened specimen, belophorus Zone.

I, P. d. dubius (Suess), HUR./18N/6, flattened specimen, belophorus Zone. All figures X 5.

J

252

PALAEONTOLOGY, VOLUME 8

mately 20°), and number 9 in 10 mm. throughout the whole 5J cm. Excavations of the
ventral margin of each theca are conspicuous and deep, occupying almost half the width
of the rhabdosome. The length of the excavation is 0-4 mm.

The proximal ends have thecae whose general characters are identical to those shown
on the distal fragment, but they are smaller and number 10 in 10 mm. The sicula is long
and slender, measuring slightly over 2 mm., and has its apex situated midway between
the apertures of th. 1 and th. 2.

Remarks. The thecal characters distinguish this rare species from all others occurring at
Cautley. It closely resembles Tullberg’s original figures, particularly in the size and
position of the sicula; the only difference is that the Cautley specimens have slightly
longer thecal excavations. Tullberg (1883, p. 20) gives a thecal count of 7-8 in 10 mm.
but Pribyl (1940u) gives a range of 10-8 in 10 mm. Pribyl, however, includes Elies and
Wood’s Monograptus cf. griestonensis in his synonymy of linnarssoni. This in the
writer’s opinion is a dubious step since the former shows major differences from lin-
narssoni particularly as regards thecal spacing. M. cf. griestonensis has a thecal spacing
of 9-12 in 10 mm. and a sicula, only 1-5 mm. long, whose apex is midway between the
apertures of th. 1 and th. 2. Furthermore, M. cf. griestonensis has ‘hooked’ apertures to
the first few thecae. This does not seem to be the case with M. linnarssoni Tullberg.

Monoclimacis flumendosae flumendosae Gortani
Plate 29, figs. 1-3 ; text-figs. 2d, 4 f

71911 Monograptus vomerinus var. j9 Elies; Watney and Welch, text and tables (pars).

1922 Monograptus linnarssoni v. flumendosae Gortani.

1931a Monograptus lejskoviensis Boucek, pp. 9-10, text-fig. 3 a-c.

1940a Monoclimacis flumendosae (Gortani 1922); Pribyl, p. 6, pi. 2, figs. 14-16.

Material. Several hundred specimens, all flattened, many well preserved.

Horizons and localities. Zone of Monograptus autennularius to Zone of Cyrtograptus ellesi', Middle
Gill (16M, 18M, 19M, 20M, 23M, 26M, 27M, 28M, 29M, 30M), Near Gill (16N-23N), Wandale Hill
(43-46W), R. Rawthey (9-1 IRa), mouth of Wandale Beck (67-69W).

Diagnosis. Full length unknown but probably greater than 30 cm. ; distal width 2 mm.
(flattened), initial width 0-3 to 0-5 mm.; slight, but characteristic, dorso-ventral curva-
ture; thecae number 8-10 in 10 mm.

Description. The rhabdosome has a very characteristic appearance, widening from
a slender and graceful proximal end, which almost invariably shows distinct dorsal

DESCRIPTION OF PLATE 29

Figs. 1-3, Monoclimacis flumendosae flumendosae (Gortani), linnarssoni Zone. 1, HUR./17N/244,
well-preserved specimen showing characteristic double curvature. 2, HUR./17N/216, distal part of
rhabdosome. 3, HUR./17N/232, proximal end with sicula. Figs. 4-6, Monograptus simulatus sp.
nov., centrifugus Zone. 4, HUR./4M/62, two thecae showing thecal hook, X 10. 5, HUR./4M/62,
paratype. 6, Holotype, HUR/28W/25, specimen in relief. Fig. 7, Pristiograptus meneghini (Gortani),
HUR./17N/46, linnarssoni Zone. Figs. 8-9, Monograptus flexilis belophorus, respectively HUR./
13M/74 and HUR./16M/20, riccartonensis and belophorus Zones, proximal and distal regions,
flattened, showing typical curvature. Fig. 10, Pristiograptus pseudodubius (Boucek), HUR./26N/11,
flattened specimen with sicula and nema, lundgreni Zone. All figs. X 5 except where otherwise
stated, unretouched.

Palaeontology, Vol. 8

PLATE 29

RICKARDS, Silurian graptolites

R. B. RICKARDS: NEW SILURIAN GRAPTOLITES

253

curvature (PI. 29, figs. 1, 3) to a long and variously curved distal region. Whilst many
specimens are almost straight distally, equally as many show a gentle ventral or dorsal
flexure.

For the first 3-5 mm. the proximal end is sharply recurved and the sicula is prominent.
Specimens in relief must be very slender and graceful in this region.

The sicula is 2 mm. long and its apex is above the level of the aperture of th. 1.
Occasional specimens show a slightly shorter sicula whose apparent apex barely reaches
the level of the aperture of th. 1 ; but in these specimens the true apex may be hidden.
The sicula is rarely curved.

Th. 1 originates fully 04 mm. above the base of the sicula which is, therefore, con-
spicuous with its short virgella. At the proximal end the thecal spacing is very constant at
10 thecae in 10 mm. The change distally is gradual and even the most distal fragments
obtained do not show less than 8 thecae in 10 mm.

Thecal excavation is well marked throughout the rhabdosome but distally it increases
in length. As a rule it occupies about one-third the total width of the rhabdosome. The
excavation is, however, rather less prominent than in some earlier monoclimacids. The
distal thecae occasionally show a pristioform view akin to that sometimes seen in distal
thecae of Monoclimacis vomer ina basilica (Lapworth 1880).

Remarks. M.f. flumendosae differs from the subspecies M.f. kingi subsp. nov. (described
below) in being more robust throughout the length of the rhabdosome and in the less
closely spaced proximal thecae. Otherwise the two are very closely allied, and kingi,
found in the lundgreni Zone, evolved from the type subspecies which is common in the
pr e-lundgreni zones of antennularius to ellesi.

The only other form with which M. f. flumendosae might be confused is M. vomerina
basilica. The latter form is, however, much broader distally and lacks the curvature and
large sicula typical of the proximal end of flumendosae.

The Cautley specimens agree closely with those described and figured by Gortani
( 1 922) as Monograptus linnarssoni var. flumendosae from the Silurian of Sardinia, and
with Boucek’s (1931u) Monograptus lejskoviensis from Czechoslovakia. Pfibyl ( 1 940a)
rightly regards flumendosae and lejskoviensis as the same species. The only difference
between the writer's material and that described by the above authors is that flumendosae
from the Cautley district has a more slender proximal end, which is 0-3 to 0-5 mm. wide
at the level of th. 1, and may be contrasted with the figure of 0-6 mm. given by Pfibyl
(1940tf).

M. f. flumendosae has not previously been recorded from the Cautley area but it is
possible that the form described by Watney and Welch (1911) as Monograptus vomerinus
var. (= gracilis ; Elies and Wood 1910, p. 44) should, in part at least, be referred to M.
flumendosae. M. vomerina gracilis (Elies and Wood) has not been found by the present
writer from the Silurian of Cautley.

Monoclimacis flumendosae kingi subsp. nov.
Text-fig. 4e

Holotype. HUR./28N/4, a proximal end with sicula, flattened.

Other material. About thirty specimens, all flattened but well preserved.

254

PALAEONTOLOGY, VOLUME 8

Horizon and localities. Lower half of the lundgreni Zone, Wenlock Series; Near Gill (25N-29N),
Crosshaw Beck (lCr, 2Cr); Cautley, near Sedbergh.

Derivation of name. In honour of the late Professor W. B. R. King, contributor for many years to the
problems of Lower Palaeozoic stratigraphy.

Diagnosis. Rhabdosome similar to type subspecies but more slender throughout; initial
width 0-3 mm., distally less than 2 mm.; thecal spacing 1 1-8 in 10 mm.

Description. The whole general form of the rhabdosome is very close to that of the type
subspecies (described above) but is invariably more slender throughout its whole length.
At the level of th. 1 the rhabdosomal width (flattened) is 0-3 mm.

The thecae are more closely spaced and counts of 11 in 10 mm. are typical over the
first few centimetres. Thereafter the thecae become gradually larger and the count falls
to 10 and 9 in 10 mm. Extreme distal fragments, which are less than 2 mm. in width,
have a thecal spacing of as low as 8 in 10 mm.

Monoclimacis shottoni sp. nov.

Plate 30, fig. 3; text-fig. 2e

?1900 Monograptus vomer inus var. y Elies, p. 405, text-fig. 17.

1935 Monograptus vomerinus var. crenulatus (Tornquist); Shotton, p. 661.

Holotype. HUR./28W/76 a complete specimen in full relief.

Material. Numerous specimens in full relief, all well preserved.

Horizon and localities. Zone of C. centrifugus ; Wandale Hill (28 W), mouth of Wandale Beck (49W,
51W), Pickering Gill (3P, 5P, 6P, 10P), River Rawthey (8Ra); Cautley, near Sedbergh; Cross Fell.

Derivation of name. In honour of Professor F. W. Shotton.

Diagnosis. Rhabdosome short, with slight ventral curvature, narrowing distally; sicula
prominent; thecae with distinct sigmoidal curvature numbering 12-13 in 10 mm.

Description. No specimens over 7 mm. long have been obtained and since these narrow
towards their distal extremities it is thought that they are full grown. The maximum
width is reached at th. 4 and rarely exceeds 0-71 mm. At th. 7 the width has decreased
to 0-58 mm.

The sicula is 2 mm. long and its apex invariably reaches the level of the second thecal
aperture. It is 0-29 mm. wide at the base and shows a faint ventral curvature. Th. 1
arises 0-20 mm. above the sicular aperture. Thecal lengths are as follows: th. 1, 0-9 mm.;
th. 2, 0-9 mm.; th. 3, 1T7 mm.; th. 4, T23 mm.; th. 5, T3 mm.

DESCRIPTION OF PLATE 30

Fig. 1, Monograptus aff minimus cautleyensis subsp. nov., HUR./1M/1 16, specimen in relief with thecal
hooks apparently less enrolled, centrifugus Zone. Fig. 2, Monograptus danbyi sp. nov., holotype,
HUR./8P/1, specimen in full relief with sicula, centrifugus Zone. Fig. 3, Monoclimacis shottoni sp.
nov., holotype, HUR./28W/76, complete specimen in relief, centrifugus Zone. Fig. 4, M. griestonensis
nicoli subsp. nov., holotype, HUR./8P/19, specimen in relief, centrifugus Zone. Fig. 5, M.
Iinnarssoni (Tullberg), HUR./10P/47, centrifugus Zone. Fig. 6, Prist iograpt us welchae sp. nov.,
holotype, HUR./l Ab/22, flattened specimen, leintwardinensis Zone. Figs. 7-8, P. watneyae sp. nov.,
distal and proximal parts of holotype, HUR./37W/17, specimen in relief, centrifugus Zone. All
figs, x 1 0, unretouched.

Palaeontology, Vol. 8

PLATE 30

RICKARDS, Silurian graptolites

4 — e

R. B. RICKARDS: NEW SILURIAN GRAPTOL1TES

255

Growth-lines are usually visible on the thecae but not always on the sicula. As in the
case of M. haupti (Kuhne) described below, there is an increase in width of the growth-
bands from the proximal to the distal thecae. In M. shottoni there are over twenty
growth-bands in th. 1 (7-8 in the metathecal portion) but each succeeding theca has only
14-15 such bands (6-7 in the metathecal portion). There is no diminution in width of the
bands in the distal region where the rhabdosome begins to narrow (th. 5-7).

The apertural margins of the thecae are not ‘hooked’ but appear to be slightly
everted. The ventral excavation of the thecal tube is pronounced.

Remarks. M. shottoni closely resembles those specimens figured by Elies (1900) as M.
vomerinus var. y and included by Elies and Wood (1910) in their synonomy of M. crenu-
latus Tornquist. These, which are figured natural size, are rather broader than the
Cautley specimens, however, but have the same size sicula (2 mm.), the same thecal
count (12-13 thecae in 10 mm.), similar general size, and are recorded from the mur-
chisoni Zone. (Reference to their table 10, p. 406, suggests that the figure may be more
than natural size.) No other figured specimens of M. crenulata (Tornquist) resembles the
Cautley material.

M. shottoni was first recorded by Professor Shotton (1935) from his locality ‘g\
Swindale Beck, as M. vomerinus var. crenulatus (Tornquist). Some of these specimens
are now contained in the Sedgwick Museum, Cambridge, and are identical with the
Cautley form.

Monoclimacis? haupti (Kuhne)

Text-fig. 2c

1955 Monograptus haupti Kuhne, pp. 365-8, fig. 3a-f.

1958 Monoclimacis haupti (Kuhne); Urbanek, pp. 89-92, text-figs. 59-65, pi. 4, fig. 5.
Holotype. Specimen figured by Kuhne (1955) fig. 3a-f.

Material. A single specimen from the Howgill Fells, north of Sedbergh, and other less well-preserved
specimens from the Barbon Fells, south of Sedbergh.

Horizon and localities. Near the base of the nilssoni-scanicus Zone, Ludlow Series; Wandale Hill
(2W), Barbon Fells.

Diagnosis. Rhabdosome short, almost straight, sicula conspicuous and ventrally curved ;
thecae with distinct sigmoidal curvature numbering 12-14 in 10 mm.

Description. The sicula has a minimum length of 1-43 mm. Its apex is hidden in the
specimens seen but probably extends to the level of the second thecal aperture. In spite of
the fact that the best specimen has been displaced at the level of th. 3 (giving an apparent
dorsal curvature) the sicula is clearly curved ventrally. Three distinct rings are present
on the upper half of the sicula, which are thought to be equivalent to the ‘peridermal
rings’ described by Urbanek (1958, p. 58). At the base the sicula is 0-32 mm. across, and
between this point and the lowest peridermal ring there are about thirty-five growth
bands. This figure agrees closely with Urbanek’s fig. 61c (1958, p. 90). The middle
peridermal ring is 0-26 mm. above the lower, and the top one 0T3 mm. above the
middle.

Th. 1 originates 0T mm. above the base of the sicula and has a length of 0-7 mm.
Succeeding thecae increase slowly in length up to th. 5 (the last measurable theca on the

256 PALAEONTOLOGY, VOLUME 8

specimens available) which is 1-3 mm. long. At this point the width of the rhabdosome
is 0-71 mm.

Each prothecal portion has approximately 14-15 growth-bands (counted in profile
view) and the metathecae 8-9 bands. There is a distinct and gradual increase in the width
of the growth-bands (measured in a sense parallel to the length of the rhabdosome) from
the sicula, where they are narrow, to th. 5 where they are 0-50 mm. wide.

Remarks. Urbanek (1958) considers that Kiihne’s species is in fact a monoclimacid in
view of the strong sigmoidal curvature of the thecae. Furthermore, he suggested (p. 90)
that M. haupti may be conspecific with Monograptus praeultimus Munch 1942. M.
praeultimus has a thecal spacing of 12-14 in 10 mm. Its maximum width is 1 T mm. M.
ultimas Perner 1899 is also considered by Urbanek to be referable to the genus Mono-
climacis Freeh.

The specimens described by Kiihne (1955) and Urbanek (1958) were obtained from
erratic boulders.

Genus pristiograptus Jaekel 1889
Type species. P. frequens Jaekel 1889.

Diagnosis. Rhabdosome of very variable length and curvature but commonly almost
straight; thecae are straight, simple tubes throughout length of rhabdosome, and have
varying degrees of overlap and inclination; sicula small to very large.

Pristiograptus watney ae sp. nov.

Plate 30, figs. 7, 8; text-fig. 2 g
71900 Monograptus hisingeri Carr, var., Elies, tables.

71911 Monograptus hisingeri Carr, var., Watney and Welch, p. 219 and tables (pars.).
Holotype. E1UR./37W/19 and counterpart /1 7, specimen in full relief with a length of 14 cm.

Horizon of holotype. Zone of C. centrifugus.

Derivation of name. After Miss G. R. Watney, joint author, with Miss E. G. Welch (1911), of a valuable
paper on the Salopian stratigraphy of the Cautley area.

Material. Two well-preserved specimens, the holotype and a mesial fragment, and other doubtful
specimens.

Horizon and localities. Zones of C. centrifugus and M. riccartonensis, doubtful above the latter; Wandale
Hill (37W), Hobdale Beck (lHd).

Diagnosis. Rhabdosome long and straight, distally 2-3 mm. wide; thecae simple over-
lapping tubes numbering 7-13 in 10 mm.; angle of inclination a maximum of 40°.

Description. (Drawn mainly from the holotype.) The rhabdosome is long and mostly
straight but with a very slight dorsal curvature at the proximal end, the whole being
very similar to P. regularis (Tornquist 1899).

The sicula has a length of T43 mm. and is quite inconspicuous. Its apex reaches 0T5
mm. above the level of the aperture of th. 1 . Over the first three thecae the thecal count is
13 in 10 mm. and these early thecae are inclined to the axis at a very low angle (5-10°).
Both the thecal spacing and angle of inclination of the thecae increases rapidly along the
rhabdosome so that at 1 cm. from the base of the sicula they number 9 in 10 mm. and
are inclined at 20° to the axis. The overlap at this point is rather less than one-half.

R. B. RICKARDS: NEW SILURIAN GRAPTOLITES

257

There is no change in the thecal characters distally but the thecae are less closely
spaced (7 in 10 mm.) whilst the angle of inclination may be as high as 40°. The thecal
overlap is rather more than half and the tubes themselves have a length of 2-5 mm.

Remarks. The species recorded as M. hisingeri Carr. var. by Watney and Welch (1911)
was probably in part P. watney ae for the latter certainly belongs to the nudus group of
pristiograptids. On the other hand, Watney and Welch record their variety throughout
the Wenlock Series, stating that it is rare in their lowest and topmost zones. The writer
feels that they have confused pristioform views of Monoelimacis flumendosae Gortani
1922 with P. watneyae and lumped the two together as M. hisingeri Carr. var. M.
flumendosae is common in their zones of riccartonensis and rigidus but less common in
their highest zone of lundgreni (as they record for M. hisingeri Carr. var.). M. flumendosae
is absent in their zone of murchisoni, where P. watneyae occurs rarely.

The two species P. watneyae and M. flumendosae, though superficially resembling
each other, particularly as regards overall size, are in fact quite distinct. The latter, quite
apart from the ventral excavation of the thecal tube, has a characteristic dorsal curvature
at the extreme proximal end, a prominent sicula, is more slender, and has a different
thecal spacing.

From members of the dubius group of pristiograptids P. watneyae differs in its
slender and slowly widening proximal region. The closest species in the nudus group of
pristiograptids is P. nudus (Lapworth). From this species it differs in the following ways:
(a) the distal thecae are long rather than broad; ( b ) the rhabdosome is broader distally
and more slender proximally; (c) the thecal spacing is different; (d) the sicula is longer
and its apex positioned differently; (e) the distal angle of inclination (maximum 40°) is
less than in nudus.

P. watneyae is more robust than P. r. regularis (Tornquist 1899) and P. r. solidus
(Pfibyl 19406).

Pristograptus dubius pseudolatus subsp. nov.

Plate 31, fig. 5; text-fig. 2k

Holotype. HUR./18N/53a, specimen well preserved, flattened, proximal end preserved, about 4 cm.
long.

Horizon of holotype. Zone of M. flexilis belophorus.

Derivation of name. L, to indicate that it is distinct from P. dubius latus Boucek 1932.

Material. Thirty-one specimens, proximal and distal fragments, all flattened.

Horizon and localities. Zone of M. flexilis belophorus to the basal beds of the Zone of C. linnarssoni;
Near Gill (16N, 18N, 19N).

Diagnosis. Rhabdosome several centimetres long, proximally with dubius-Yike curvature,
but distally quite straight; maximum width 2-7 mm. ; thecal tubes simple, inclined to the
axis at 20°-30°, numbering 7-10 in 10 mm.

Description. This subspecies is superficially very similar to the type subspecies, but the
rhabdosome, whilst showing the same ventral curvature at the proximal end, is distally
straight, and broader. A width of 2-0 mm. is reached at only T5 cm. from the proximal
end, which, at the level of th. 1, is already 0-75 mm. in width. At 3J cm. the width has
increased to 2-34-2-47 mm. and in the most distal fragments seen reaches 2-7 mm.

258

PALAEONTOLOGY, VOLUME 8

Initially P. d. pseudolatus has the same thecal spacing as P. d. dubius (10 in 10 mm.)
but at 1-|- cm. the count has already fallen to 7 in 10 mm. This latter value is then main-
tained to the distal extremity.

The angle of inclination ranges between 20° and 30° being usually nearer the lower
angle. The sicula is over 2 mm. long, usually 2-3 mm., and its apex reaches to the level
of the second thecal aperture.

Remarks. Many of the bedding planes at the locality of the holotype show no signs of
compression and this subspecies is clearly not a broad form of dubius s.s. resulting from
tectonic deformation of the rock. In any case if a specimen of dubius s.s. is compressed
in such a manner so as to increase the width of the rhabdosome, then the angle of inclina-
tion of the thecae and the thecal counts are both increased. In dubius pseudolatus the
angle of inclination is lower than in dubius s.s. and the thecal spacing shows a similar
range but is distally less. P. d. pseudolatus bears some resemblance to P. d. latus Boucek
but in the latter the thecae are inclined at a higher angle than dubius s.s. (see, for example,
Pribyl 1944, pi. 1, fig. 7).

From P. meneghini giganteus (Gortani) the Cautley form differs in having a higher
thecal count; and from P. s. sardous (Gortani) in being rather more slender and in
having a lower angle of inclination.

Pristiograptus meneghini meneghini (Gortani)

Plate 29, fig. 7; text-fig. 2 j

1857 Graptolithus (Monograpsus) colonus Barr?; Meneghini (pars.), p. 164.

1922 Monograptus meneghini Gortani, p. 47, pi. 8 (1), figs. 3-8, pi. 12 (5), fig. 6 d, pi. 13 (6),
figs. 2c, 4 a.

1922 Monograptus meneghini Gortani, p. 99, pi. 17 (3), fig. 10.

1936 Monograptus paradubius Haberfelner, pp. 89-90, figs. 2 a-b.

1944 Pristiograptus meneghini meneghini (Gortani 1922); Pribyl, pp. 11-13, text-fig. 2, figs.
El-3, pi. 1, figs. 1-2.

1952 Pristiograptus ( Pristiograptus ) meneghini meneghini (Gortani); Munch, p. 86, pi. 18,
fig. 9.

1952 Pristiograptus ( Pristiograptus ) meneghini meneghini (Gortani); Pribyl, pp. 26-27, pi. 1,
figs. 4, 15.

71958 Pristiograptus ( Pristiograptus ) cf. meneghini meneghini (Gortani 1922); Pribyl, pp. 117-
18, pi. 1, fig. 9.

71962 Pristiograptus meneghini meneghini (Gortani); Romariz, p. 283, pi. 13, figs. 13, 16, 17.
Lectotype. Specimen figured by Gortani (1922), pi. 8 (1), fig. 4.

Material. About eighty specimens, all flattened but well preserved, several proximal ends but distal
fragments more common.

Horizon and localities. Zones of antennularius, flexilis belophorus, and linnarssoni; Near Gill (16N,
17N, 20-23N), Middle Gill (16M, 21 M, 22M, 25M, 30M), Wandale Hill (68W).

DESCRIPTION OF PLATE 31

Fig. 1, Monograptus firmus sedberghensis subsp. nov., holotype, HUR./40W/1 riccartonensis Zone.
Figs. 2-4, Pristiograptus auctus sp. nov. nilssoni-scanicus Zone. 2, Paratype, HUR./7W/43. 3, Para-
type, HUR./7W/34. 4, Holotype, HUR./7W/46. Fig. 5, Pristiograptus dubius pseudolatus subsp.
nov., holotype, HUR./18N/53a, belophorus Zone. Fig. 6, P. d. dubius (Suess), HUR./18N/6, for
comparison with pseudolatus subsp. nov. All figs. X 5, unretouched.

Palaeontology, Vol. 8

PLATE 31

RICKARDS, Silurian graptolites

R. B. RICKARDS: NEW SILURIAN GRAPTOLITES

259

Diagnosis. Rhabdosome similar in appearance to, but broader than, P. pseudodubius
(described below); proximal regions ventrally curved, but distal fragments straight;
maximum width 1 -6—1-7 mm.; thecal spacing 7-10 in 10 mm.

Description. The rhabdosome is longer than that of P. pseudodubius and may reach 5
cm., whilst a width of 1-6-1 -7 mm. is reached distally.

The sicula is fully 2 mm. long and its apex extends to about the level of the second
thecal aperture. At the level of th. 1 the rhabdosome has a width of 0-65 mm. which
increases distally to 1-43-1-50 mm. in most specimens. Some specimens, however, from
the zone of M. antennularius, are rather broader and reach 1-6-1 -7 mm.

The thecal spacing of those specimens from the antennularius zone is 7-9 in 10 mm.
whilst those from the higher beds have 8-10 in 10 mm. In each case the closer spacing is
at the proximal end.

The angle of inclination is approximately 30°.

Remarks. The slight shift in the range of variation of some biocharacters, from the
lower to the higher strata, is regarded as a change which would, if continued, give rise to
P. pseudodubius (Boucek). Those specimens from the zone of antennularius are identical
with Gortani’s original specimens whilst those from the zones of flexilis belophorus and
linnarssoni are slightly narrower and have more closely spaced thecae. P. pseudodubius
from the lundgreni Zone is narrower still and has even more closely spaced thecae.

Pristiograptus pseudodubius (Boucek)

Plate 29, fig. 10; text-fig. 2 i

1932 Monograptus pseudodubius Boucek, pp. 1-2, pi. 2 e-f.

1944 Pristiograptus pseudodubius (Boucek 1932); Pribyl, pp. 8-9, pi. 1, fig. 8, text-fig. I, 3.

1945 Monograptus pseudodubius Boucek; Waterlot, pi. 26, fig. 288.

71962 Pristiograptus pseudodubius (Boucek); Romariz, pi. 16, fig. 3.

Lectotype. Specimen figured by Boucek as fig. 2e and refigured by Pribyl, text-fig. I, 3.

Material. About thirty specimens, invariably flattened, usually poorly preserved.

Horizon and localities. Zone of C. lundgreni ; Near Gill (25N-28N), Hobdale Beck (3Bd), River
Rawthey (2Ra).

Diagnosis. Rhabdosome short, narrow, with gentle ventral curvature throughout; maxi-
mum width 1 mm., thecae number 9|-11 in 10 mm.

Description. The rhabdosome is typically short and slender, appearing gently arched
with the thecae on the concave side. Specimens over 2 cm. long have not been observed.
Compressed specimens at right angles to the lineation on the bedding plane may reach
1-17 mm. in width; specimens distorted in the opposite sense rather less than 1 mm.

The conspicuous sicula is 1-5 mm. long, with its apex reaching almost to the level of
th. 2. A short, stout virgella is present.

Proximally the thecal spacing is rather closer (10-11 in 10 mm.) falling to 9|-10 in
10 mm. mesially and distally. The thecal tubes have a maximum length of 2 mm. and
overlap for one-half their length. They are inclined to the axis at 20-30°.

Remarks. In all their diagnostic features the Cautley specimens agree with P. pseudo-
dubius (Boucek). The Cautley material is, however, closer in general appearance to other

2C0

PALAEONTOLOGY, VOLUME 8

material figured by Pribyl (1944, pi. 1, fig. 8) than to the original figured by Boucek
(1932, fig. 2e) and Pribyl (1944, text-fig. I, 3) which appears to be a rather broad variant.
Pribyl (op. cit., p. 9) does mention that broad forms occur.

P. pseudodubius has a superficial resemblance to P. m. meneghini (Gortani). It differs
in being even narrower and in having its thecae more closely spaced. It is considered that
pseudodubius may have evolved from meneghini through forms intermediate in all
measurable features but possessing the same general appearance of P. meneghini. These
species maintain the same order of appearance in Bohemia (Pribyl 1944, p. 44) as in the
north of England, but in the latter area meneghini first appears at a lower level.

Pristiograptus auctus sp. nov.

Plate 31, figs. 1-3; text-fig. 2 h

Holotype. HUR./7W/46, almost complete, flattened specimen.

Horizon of holotype. Ludlow Series, nilssoni Zone.

Derivation of name. Auctus, L. ‘increase’, ‘growth’.

Material. About forty specimens, all flattened.

Horizon and localities. Low in the nilssoni Zone; Wandale Hill (7W, 8W).

Diagnosis. Rhabdosome long, broad, and stiff; proximal end with slight ventral curva-
ture, distal parts usually straight; thecae long, simple tubes numbering 1 1-18 in 10 mm.;
sicula long, virgella short and transversely (?) expanded into a disc.

Description. The rhabdosome is about 4 cm. long and usually straight distally, though
some specimens show a gentle dorsal curvature. The proximal end is invariably ventrally
curved to the extent that six thecae are involved. A maximum width of rather less than
2 mm. is achieved within 2 cm. of the sicula but specimens at right angles to the bedding
plane lineation (where present) are often slightly over 2 mm. At the level of th. 1 the
rhabdosome is 0-70-0-75 mm. wide.

The sicula is not conspicuous but has a length of 2-3 mm. Its apex reaches to the level
of the aperture of th. 3. Thecal spacing over the first few millimetres of the rhabdosome
is very close and varies from 17-20 in 10 mm. depending upon the direction of compres-
sion. A value of 18 in 10 mm. in the most constant. At a distance of 4-7 mm. from the
base of the sicula the thecal count has fallen to 13-18 in 10 mm., 15 being the usual
figure, whilst distally 10-14 is the total range encountered.

The thecae are simple tubes which reach a maximum length of 2-5 mm. in the distal
region. Here the overlap has increased to three-quarters from two-thirds proximally;
and the thecal tubes are inclined to the axis at angles up to 45°.

One of the most striking features of this species is the presence of a short virgella
(0-6 mm.) which swells into a bulb-like shape, and has the appearance of a droplet
hanging from the proximal end of the rhabdosome (PL 31, fig. 3). This swelling is 04-
0-5 mm. in diameter. Thickening of the virgella in this manner is invariably present,
but one specimen, less expanded than the others, suggests the possibility that the virgella
is transversely expanded, and that only upon flattening of the rhabdosome does it rotate
to the bedding plane. If this is the case, however, rather more specimens with the swelling
half buried would be expected.

R. B. RICKARDS: NEW SILURIAN GRAPTOLITES

261

Remarks. P. auctus is clearly close to such species as P. vulgaris (Wood 1900) and P.
tumescens (Wood 1900). The degree of curvature is intermediate between these two latter
species, whilst the breadth of the rhabdosome is nearer tumescens. The combination of
characters described above serves to distinguish auctus from both these species.

Elies and Wood (1910, p. 380) described forms of tumescens from the Lake District
which are shorter, broader, and have a higher thecal count (12-13 in 10 mm.) than those
from the type area. It is possible that these are related to the species here described,
although P. auctus is longer than tumescens, and Watney and Welch (1911) make no
mention of any such forms.

Pristiograptus welchae sp. nov.

Plate 30, fig. 6; text-fig. 2/

Holotype. Specimen HUR./lAb/22, and counterpart 22a.

Horizon of holotype. Zone of M. leintwardinensis.

Derivation of name. After Miss E. G. Welch.

Horizon and localities. Zones of nilssoni-scanicus and leintwardinensis ; Adamthwaite Bank (1 Ab) and
Weasdale (lWe).

Material. About twenty specimens, all flattened.

Diagnosis. Very small, narrow rhabdosome with only a few thecae developed; length
approximately 5 mm.; width (flattened) 0-65 to 0-70 mm.; proximal extremity with
slight ventral curvature; sicula T2 mm. in length; thecal spacing 14 in 10 mm.

Description. The tiny rhabdosome is most characteristic and unflattened specimens must
be of the order of 0-5 mm. wide. Some specimens slightly exceed 5 mm. in length, the
holotype being 5-33 mm. The maximum rhabdosomal width is 0-7 mm. and is achieved
by the fourth or fifth theca.

The sicula is not prominent but has a length of T2 mm. Its apex reaches to the level
of the aperture of th. 1. About 6-8 thecae may be present on the rhabdosome and these
are all of simple pristiograptid type with a maximum length of T3-T4 mm. and a width
of 0-20-0-22 mm. Distally the thecae overlap for one half their length, but rather less
than this proximally. The thecae number 14 in 10 mm. throughout the rhabdosome and
are inclined to the axis at a low angle — about 20° distally and less proximally.

Remarks. The only species approaching P. welchae in dimensions is P. praeultimus
Munch from the nilssoni-scanicus Zone of Thuringia. This is of similar length but
rather broader ( 1-0-1 -1 mm.) whilst the thecal spacing is similar. The thecal tubes of
praeultimus are, however, of ultimus type and appear to have a ventral excavation of the
thecal margin suggesting that they are not of simple pristiograptid form.

This is a rare species at Cautley and it was not recorded by Watney and Welch (1911)
in their work on the Salopian rocks.

Genus monograptus Geinitz 1852

Type species. Lomatoceras priodon Bronn 1835; subsequently designated Bassler 1915.

Diagnosis. Emended here only to exclude Rastrites Barrande 1850, Pristiograptus
Jaekel 1889 and Monoclimacis Freeh 1897; form of thecae highly variable; many species
biform; curvature of rhabdosome highly variable.

262

PALAEONTOLOGY, VOLUME 8

Monograptus firmus sedberghensis subsp. nov.

Plate 31, fig. 1; text-fig. 3 a
Holotype. HUR./40W/1, flattened specimen.

Material. Two well-preserved specimens, other fragmentary specimens.

Horizon and locality. Top of the M. riccartonensis Zone; Wandale Hill (40W).

Derivation of name. After the nearby town of Sedbergh.

Diagnosis. Rhabdosome with gentle dorso-ventral curvature, widening from 0-6 mm.
to almost 2 mm. at 4 cm. from the sicula; thecae with small hooks numbering 14-15 in
10 mm. over the first few mm. and 12 in 10 mm. distally; overlap one-half, increasing
distally.

Description. The rhabdosome is not robust but reaches a maximum width of 2 mm. at
about 4 cm. from the proximal end. The total length of the rhabdosome is clearly much
longer since almost straight distal fragments are associated with the curved proximal
regions. At the proximal end the rhabdosome is characterized by a very striking ventral
curvature for a length of about 1 cm. when a change to gentle dorsal curvature occurs
which is maintained throughout much of the rhabdosome.

The sicula is small and not prominent. Its length is T2 mm. and its apex reaches to
about the level of the hook of th. 2. The dorsal margin of the sicula is continuous with
the dorsal margin of the rhabdosome, an arrangement which is the prime cause of the
slight, but striking, proximal ventral curvature.

The proximal thecae number 14-15 in 10 mm. but become more widely spaced after
a few millimetres and remain at 12 in 10 mm. throughout the rest of the rhabdosome.

Thecal overlap increases slightly to rather more than half in the distal portion but
the thecae themselves are uniform throughout. There is no apparent change in the
nature of the thecal hook which always involves only the top of the thecal tube and
closely resembles that of M. riccartonensis Lapworth. The distal thecae have a maximum
length of approximately 2-5 mm. Throughout the rhabdosome the thecae are inclined
to the axis at 30°.

Remarks. The original proximal end of M. firmus figured by Boucek (1931b, fig. 5b)
seems to be a specimen preserved in subdorsal view. Nevertheless the slight ventral
curvature of the extreme proximal end can be ascertained, and is sufficient to distinguish
the species from M. riccartonensis Lapworth and M.fiemingi Salter. M.f. sedberghensis
differs from M. f. firmus in having a less robust proximal end, a more flexuous rhabdo-

text-fig. 3. a, Monograptus firmus sedberghensis subsp. nov., holotype, HUR./40W/1, flattened speci-
men, riccartonensis Zone, b, M. simulatus sp. nov., paratype HUR./4M/62, preserved in relief, centri-
fugus Zone, c, d, M. minimus cautleyensis subsp. nov., respectively paratype HUR./1M/117 and holo-
type HUR./1M/50, both in relief, the holotype partly as external mould, centrifugus Zone, e, M.
radotinensis inclinatus subsp. nov., proximal end is part of holotype, HUR./39W/3, distal fragment is
an adjacent paratype on same slab, riccartonensis Zone, f g, M. flexilis belophorus (Meneghini),
HUR./16M/20 and HUR./13M/74, fairly well-preserved specimens showing typical curvature,
respectively from riccartonensis and belophorus Zones, h, M. danbyi sp. nov. holotype, HUR./8P/1,
specimen in full relief, centrifugus Zone, i, M. simulatus sp. nov., holotype, HUR./28W/25, preserved

in relief, centrifugus Zone. All figures X 5.

264

PALAEONTOLOGY, VOLUME 8

some, and more closely spaced thecae throughout the rhabdosome. M. f. sedberghensis
seems to occur at a slightly higher horizon than M.f. firmus which is recorded from strata
underlying the riccartonensis Zone of Bohemia (Boucek 1933).

M. tariccoi Gortani is a species showing a close resemblance to M. firmus. M. f.
sedberghensis differs from Gortani’s species in being more flexed, more slender, and in
the thecal spacing, whilst M.f. firmus is shorter, has a more robust proximal region, and
has more closely spaced thecae. M. tariccoi is a Wenlock species but occurs at a higher
horizon than M.f. sedberghensis and is associated with C. rigidus Tullberg.

Monograptus radotinensis inclinatus subsp. nov.

Text-fig. 3e

Holotype. HUR./39W/3, a flattened specimen showing proximal and mesial regions.

Material. Six specimens on a single slab ; all flattened but moderately well preserved.

Horizon and locality. Top of zone of M. riccartonensis ; Wandale Hill (39W).

Derivation of name. Inclinatus, L. ‘inclined towards’.

Diagnosis. Distal part of rhabdosome like M. riccartonensis but more slender; proximal
region flexuously curved, dorsal curvature at extreme proximal end; thecae like those
of M. r. radotinensis ; maximum rhabdosomal width 1-2-1 -3 mm.; thecae number 9|-
11 in 10 mm.

Description. The rhabdosome shows dorsal curvature over the first 2 cm. followed by
ventral curvature for a similar length. Distal fragments are more or less straight, fully
8 cm. long, and it seems likely that the total length of the rhabdosome must have
exceeded 10 cm. In the specimens obtained the amount of curvature has been lessened
by compression. The rhabdosomes are probably close to the original width since although
the specimens are compressed in a manner tending to decrease the width, they are also
flattened which increases the width.

The sicula is almost 1-7 mm. long and 0-3 mm. wide at the base. Its apex reaches
slightly above the level of the hook of th. 1. The proximal thecae number 11 in 10 mm.
and are very similar to those of the type subspecies and to those of M. irfonensis Elies.
They are approximately 1-5 mm. long with a small beak-like hook in the apertural
region. Overlap of the thecal tubes is rather less than half.

The distal thecae are more widely spaced (9+ in 10 mm.) and the amount of overlap
is about one-half. There is also a slight change in the angle of inclination of the thecae
from 10°-15° proximally to 20° distally. The nature of the thecal hook does not appear
to change.

Remarks. M. r. inclinatus is very similar to the type described by Boucek (19316). It
differs from this form, however, in the thecal spacing (9^-11 cf. 9-10), the rhabdosomal
width (T2-T3 cf. TO mm.), the length (3 cm. cf. 10 cm.) and the curvature. M. r.
radotinensis has pronounced dorsal curvature at the proximal end whereas M. r.
inclinatus is flexuously curved, at first dorsally and then ventrally, for a distance which
itself exceeds the full length of the rhabdosome of the type subspecies.

M. r. radotinensis was recorded by Boucek (1933) from the zones of C. murchisoni
and C. insectus whilst M. r. inclinatus occurs at the top of the riccartonensis Zone at
Cautley. It seems quite likely, therefore, that the latter has evolved from the type sub-

R. B. RICKARDS: NEW SILURIAN GRAPTOLITES

265

species by overall increase in rhabdosome size, flexuosity, and spread of the proximal
thecal characters further along the rhabdosome. The proximal thecae of both sub-
species closely resemble those of M. irfonensis Elies (Elies 1900, fig. 19, and Elies and
Wood 1912, text-fig. 292) which may have evolved from this stock. If this is the case
then irfonensis has resulted from further spread of the proximal characters along the
rhabdosome to the extent that the riccartonensis-\ike distal parts of M. radotinensis
are lost.

M. lovisatoi Gortani 1922 has a superficial resemblance to M. radotinensis but differs
in being more robust and in having widely spaced thecae. The nature of the thecal hook
in lovisatoi is also rather obscure.

Monograptus simulatus sp. nov.

Plate 29, figs. 4-6; text-figs. 3b, i

Holotype. HUR./28W/25, long specimen in relief, preserved mainly as an external mould.

Material. Two well-preserved specimens in relief.

Horizon and localities. Zone of C. centrifuges; Wandale Hill (28 W) and Middle Gill (4M).

Derivation of name. Simulatus, L. ‘feigned’.

Diagnosis. Rhabdosome with dorso-ventral curvature, maximum width 0-3 mm.,
thecae long, narrow, with apertural hook, numbering 5-6 in 10 mm.; overlap nil;
sicula unknown.

Description. This rare fossil has a highly characteristic rhabdosome. In the most proxi-
mal part known it shows dorsal curvature. More distally the curvature becomes ventral
and then once again dorsal. It widens almost imperceptibly from 0-26 mm. to 0-30 mm.

The thecae are widely spaced numbering 6 in 10 mm. proximally and 5 in 10 mm.
distally. Throughout most of their length the thecae are closely adpressed to the axis
but at their extreme distal end the aperture is involved in a small but prominent hook.
As far as can be ascertained the hook is formed quite simply by retroversion of the
dorsal lip. The hook occupies about one-third to one-half of the width of the rhabdo-
some. There is no overlap. The prothecal portion arises as a slender tube approximately
0-07 to 0-09 mm. in diameter, and at this point, because of its delicate nature, often
shows a slight crumpling. The protheca widens steadily throughout its length to a maxi-
mum of 0T9 mm. immediately prior to the hook itself. Thus the whole prothecal por-
tion takes the form in profile of an axially elongated triangle.

Remarks. The form of the thecae and rhabdosome is so distinctive as to enable separa-
tion immediately from other slender monograptids such as M. iStreptograptus) Yin 1937
and M. ( Mediograptus ) Boucek and Pribyl 1948.

A species similar in general form and thecal size is M. eapillaris (Carruthers) but in
this species the hook is more prominent, the rhabdosome wider, and the thecae more
closely spaced.

Another similar species is M. crinitus which Wood (1900) recorded from the Ludlow
Series ( nilssoni Zone). M. crinitus has a similar thecal spacing, thecal hook, and general
size, but is rather more robust distally and the protheca lacks the distinctive shape of
that of simulatus. Nevertheless, the similarity of general form is quite remarkable.

266

PALAEONTOLOGY, VOLUME 8

M. saccu/iferus Boucek 1931 6, which occurs at a similar horizon, is a broader
species and has its thecae more closely set.

Monograptus minimus cautleyensis subsp. nov.

Plate 30, fig. 1 ; text-figs. 3c, d

Holotype. HUR./1M/50, a specimen in relief, almost complete, but lacking the sicula; preserved partly
as a mould.

Material. About twenty specimens; not rare but apt to be overlooked because of its diminutive
appearance.

Derivation of name. After the district of Cautley, near Sedbergh.

Horizon and localities. Zone of C. centrifugus ; Near Gill (8N), Middle Gill (2M, 3M, 4M), Wandale
Hill (28W, 29W, 37W), R. Rawthey (49W).

Diagnosis. Rhabdosome slender, dorsal curvature throughout; widens very slowly
indeed from a small sicula; thecae inconspicuous, apertural regions coiled and closely
sessile on the rhabdosome, numbering 9 in 10 mm.; maximum width 0-40-0-50 mm.

Description. The proximal end shows pronounced dorsal curvature but the distal parts
are only gently flexed. A gradual widening of the rhabdosome takes place from 0- 19—
0-20 mm. at the level of the lobe of th. 1 to a maximum breadth of 0-40-0-50 mm.
achieved after 2 cm.

The sicula is small (0-8 mm.) but prominent and its apex reaches almost to the level
of the first thecal lobe.

The thecae are long and have their apertural regions coiled into a lobe which is
adpressed to the rhabdosome, occupying only one-quarter to one-fifth of the total
width. Details of the aperture cannot be seen and in many specimens it is impossible to
detect the presence of a coiled apertural region.

The presence of slight overlap can be detected in the better preserved specimens.
Throughout most of the rhabdosome the thecae number 9 in 10 mm. but over the first
few millimetres is as high as 12 in 10 mm.

Remarks. The closest species are M. kolihai Boucek 19316 and M. kodymi Boucek
19316. From M. k. kolihai and M. k. minor Boucek 19316, M. minimus cautleyensis
differs in having more closely spaced thecae and in its less prominent lobation. M.
kodymi is a much more robust species. The Cautley form is, however, extremely close to
M. ( Mediograptus ) minimus Boucek and Pribyl 1952 and is best regarded as a sub-
species of this form. It differs from the Bohemian species in the wider spacing of the
thecae and the more robust rhabdosome. M. ( Mediograptus ) minimus Boucek and Pribyl
is recorded from their zone of C. murchisoni whereas M. minimus cautleyensis is found
in the preceding zone of C. centrifugus.

Monograptus danbyi sp. nov.

Plate 30, fig. 2; text-fig. 3 li

Holotype. HUR./8P/4, and counterpart /l, well-preserved specimen in relief with proximal region and
sicula intact.

Material. Three well-preserved specimens in relief.

Horizon and locality. Zone of C. centrifugus', Pickering Gill (8P).

Derivation of name. After the writer’s colleague, the late Mr. C. M. Danby.

R. B. RICKARDS: NEW SILURIAN GRAPTOLITES

267

Diagnosis. Rhabdosome short, slender, with pronounced dorsal curvature particularly
at the proximal end; thecae with very prominent lobes and characteristic prothecal
portions, numbering about llj-9f in 10 mm.; sicula small.

Description. The dorsal curvature of the rhabdosome is particularly strong in the initial
few millimetres, where the sicula and th. 1 and 2 lie in a line at right angles to the
general trend of the more distal parts of the rhabdosome. A maximum width of 0-52
mm. is reached by th. 7 or th. 8 and is maintained to the distal extremity.

The sicula is prominent, measures 09-1 T mm. in length and has a width at the base
of OT3-OT5 mm. Its apex reaches well above the apertural lobe of th. 1. At the proximal
end the thecae are more closely spaced and number 10-1 If in 10 mm. Distally this value
falls to 9J-10 in 10 mm.

The species has uniform thecae, the lobes of which occupy approximately half the
width of the rhabdosome. A distinctive feature of the thecal tubes is the relatively broad
initial part of each protheca which is 0-26 mm. wide. This narrows conspicuously as the
lobe is approached and at its distal extremity the protheca is only 0T9 mm. wide. A
rapid narrowing then follows and the early part of the metatheca is about 0-06 mm. wide
(in profile view). However, it is clear that at this point the tube is transversely expanded
and may measure 0-20 mm. in this direction. Thus whilst the initial part of each pro-
theca is expanded in the dorsoventral plane the initial part of each metatheca is trans-
versely expanded.

The metatheca is then coiled into a tight lobe, the exact nature of which cannot be
determined. The tube is certainly coiled through at least 700°, and there seems to be
a certain amount of torsion of the thecal axis. This latter feature, however, may be a
result of partial flattening of this particular part of the thecal tube which, because of
its transverse expansion, must be relatively unstable in the face of diagenetic processes.

Remarks. The thecal characters described above ally this species with those mono-
graptids grouped by Yin (1937) into Monograptus ( Streptograptus ), and dealt with in
more detail by Boucek and Pribyl (1942). M. danbyi differs from streptograptids of the
exiguus group (Pribyl 1941, Boucek and Pribyl 1942) in its strong dorsal curvature. On
the other hand, the thecal lobe is certainly of the nodifer type.

In its proximal dorsal curvature M. danbyi resembles some members of the antennu-
larius group of streptograptids but the species flexuosus Tullberg 1883, retrover sus
Pribyl 1941, antenmdarius Meneghini 1851, floridus Gortani 1922, retroflexus Tullberg
1883, and extenuatus Boucek and Pribyl 1942, all seem to have a prothecal portion
with the ventral margin more or less parallel to the axis of the rhabdosome, in addition
to being, with the exception of extenuatus , rather more robust. The thecae in danbyi are
also more closely spaced than in most of the above species.

The form of the prothecal tube of danbyi is very similar to M. runcinatus Lapworth
1876 and the thecae are similarly spaced, but the rhabdosome of runcinatus is stiff
and robust.

The closest species to danbyi seem to be the Llandovery forms M. nodifer Tornquist
1881 and M. runcinatus Lapworth if the thecal type alone is considered. If, however,
the strong dorsal curvature of the rhabdosome is taken into account then danbyi may
bear some relationship to forms such as flexuosus Tullberg.

268 PALAEONTOLOGY, VOLUME 8

Monograptus flexilis belophorus (Meneghini)

Plate 29, figs. 8, 9; text-figs. 3/, g

1857 Graptolithus ( Monograpsus ) belophorus Meneghini (pars) p. 166, tab. B, fig. 4 b; II, 4, 4a.
1857 Graptolithus (Monograpsus) Gonii Meneghini (pars) p. 172, tab. B, pi. II, 6 a.

1857 Graptolithus ( Monograpsus ) priodon Meneghini (pars) p. 178, tab. B, pi. II, 9, 9a.

1922 Monograptus belophorus Mgh. em. Gortani p. 17, (57), pi. 10, (3), figs. 9-15, pi. 12 (5),
figs. 3b, 14, pi. 13 (6), fig. 1.

1922 Monograptus belophorus var. laxus Gortani, p. 10, (94), pi. 16, (2), figs. 7-8, pi. 18 (4),
?figs. 12a, pi. 19 (5), fig. 4.

1922 Monograptus ballaesus Gortani, pp. 10-11 (94-95), pi. 16 (2), figs. 12-18 (4), figs. 11a,
pi. 19 (5), figs. 2a, 3, 6c.

1942 Monograptus flexilis belophorus (Meneghini 1857, em. Gortani 1922); Pfibyl, pp. 6-7,
text-fig. 1, figs. 6-7, pi. 2, fig. 1.

1945 Monograptus belophorus Meneghini; Waterlot, pi. 35, fig. 361.

Lectotype. Specimen figured by Gortani (1922), pi. 10 (3), fig. 9.

Material. Many fragments and some fairly well-preserved proximal ends; all flattened.

Horizon and localities. Zones of riccartonensis and flexilis belophorus ; Middle Gill (13M, 16M), Near
Gill (13N, 14N, 18N, 19N).

Diagnosis. Rhabdosome with dorsal curvature proximally and ventral distally; maxi-
mum width 2-5 mm.; thecae number 7^—1 0 in 10 mm.; sicula about 1 -5-2-0 mm.

Description. The rhabdosome exhibits the typical S-shaped curvature of the flexilis
group and widens gradually from a width of 0-6-0-9 mm. to a maximum of about
2-5 mm. The proximal dorsal curvature is stiff.

The broad sicula is 1-5-2-0 mm. long and its apex reaches to approximately the level
of the hook of th. 1. The thecae are hooked in a manner intermediate between M.
riccartonensis Lapworth and M. f. flexilis Elies, and have, therefore, rather a beak-like
apertural region. The thecae are uniform and number 7-|— 10 in 10 mm. Lower and
higher values have been obtained but these seem to be the result of compression.

Remarks. The number of badly preserved fragments probably referable to this species,
which have been obtained from the same level on other sections, suggests that the species
is quite common at this horizon. Fossils are, however, particularly difficult to extract
wherever the zone crops out.

The Cautley specimens compare well with previously figured and described material.
The thecal spacing of 7J-10 is slightly different to that given by Pfibyl (1942) in his
review of the flexilis group (5-9 in 10 mm.). Well-preserved distal fragments have not,
however, been obtained and it is possible that the thecae distally are more widely
spaced than 7| in 10 mm. The specimen figured by Pfibyl (op. cit., text-fig. 1, no. 7)
appears to have more than the 9 in 10 mm. given in the description (op. cit., p. 7) for
the proximal region.

M. f. belophorus is smaller in all dimensions than M. flexilis falcatus (Meneghini
1857) and differs also in being less recurved proximally. From M.f. flexilis it is equally
distinct, having a stiffer proximal curvature, a more slender proximal end, and a sicula
with an apex reaching only to the level of the hook of th. 1.

R. B. RICKARDS: NEW SILURIAN GRAPTOLITES

269

Monograpius sp. A.

Text-fig. 4 a

Material. Three specimens in relief, two showing the sicula and thecae 1 and 2, the third showing two
proximal thecae.

Horizon and localities. Low in the centrifugus Zone; Pickering Gill (5P), Five Gills (9Fi).

Description. The prominently hooked thecae of this species ally it with monograptids
of the priodon type (= Monograptus s.s. of some authors). Whilst it resembles M.
priodon Bronn in some respects the thecal hooks are distinct (text-fig. 4a) and the
proximal extremity is quite straight.

text-fig. 4. a, Monograptus sp. A, proximal end in full relief, centrifugus Zone, b-d, respectively M.
galaensis Lapworth, M. riccartonensis Lapworth and M. priodon (Bronn) (after Elies and Wood 1901—
18) for purpose of comparison with Monograptus sp. A. e, Monoclimacis flumendosae kingi subsp. nov.,
holotype, HUR./28N/4, moderately well preserved, but flattened proximal end, lundgreni Zone. /,
Monoclimacis f. flumendosae (Gortani), HUR./17N/232, flattened proximal end for comparison with

M. f kingi. All figures x 10.

The sicula is prominent, at least 1-43 mm. long, but merges into the nema in such
a manner that exact measurement is not possible. The sicular aperture is only 0T9 mm.
across, but the sicula is widest about 0-4 mm. from the aperture, at the point of origin
of th. 1 . The apex of the sicula is approximately midway between the hooks of th. 1
and th. 2.

Th. 1 has a length of T3 mm., of which 0-4 mm. are involved in the hook. Th. 2 is of
similar length and has the same proportion involved in the hook. The thecal count is
\2\ in 10 mm. At the aperture of th. 2 the total width of the rhabdosome is 0-52 mm.
The hook occupies between one-third and one-quarter of the total width. Thecal
overlap is slight.

Remarks. The species is perhaps closest to M. riccartonensis Lapworth but is narrower,
has less thecal overlap, and the thecae themselves are smaller. The hook is very similar.
Monograptus sp. A is even more distinct from the other species of monograptid found at
Cautley and it is not merely the proximal end of one of these.

270

PALAEONTOLOGY, VOLUME 8

APPENDIX. FOSSILIFEROUS LOCALITIES

Howgill Fells. Adamthwaite Bank lAb (7090, 0059); Crosshaw Beck lCr (6928, 9396), 2Cr (6927,
9394); Five Gills 9Fi (7249, 9991); Hobdale Beck 1 Bd (6815, 9417), 3Bd (6804, 9453); Middle Gill 2M,
3M, 4M (7073, 9719), 13M (7070, 9721), 16M (7068, 9723), 18M, 19M, 20M (7065, 9725), 21M
(7064, 9725), 22M (7063, 9727), 23M, 25M, 26M, 27M (7060, 9730), 28M, 29M (7063, 9730), 30M
(7056, 9731); Near Gill 8N (7058, 9703), 13N, 14N (7050, 9706), 16N, 19N (7045, 9708), 17N, 20N
(7044, 9708), 18N (7047, 9707), 21N (7043, 9708), 22N, 23N (7042, 9708), 25N, 26N (7042, 9710),
27N (7041, 9711), 28N (7040, 9712), 29N (7041, 9711); Pickering Gill 3P (6888, 9753), 5P, 6P, 8P, 10P
(6888, 9654); River Rawthey 2Ra (6916, 9537), 8Ra (6936, 9540), 9Ra (6929, 9538), lORa (6928, 9538),
1 IRa (6924, 9539); Wandale Hill 2W (7002, 9820), 7W (7022, 9899), 8W (7026, 9923), 22W, 29 W (7043,
9797), 37W, 46W (7057, 9842), 39W (7053, 9844), 40W (7054, 9844), 43W, 44W, 45W (7049, 9843),
49W (7071, 9777), 51W (7069, 9778), 67W, 68W, 69W (7061, 9770); Weasdale lWe (6919, 0091).

REFERENCES

barrande, j. 1850. Graptolites de Boheme. Prague.

bassler, R. s. 1915. Bibliographic Index of American Ordovician and Silurian Fossils. Bull. U.S. Nat.
Mas. 92.

boucek, B. 1931fl. Sur la presence de la Zone a Cyrtograptus rigidus Tullb. et d’autres Zones dans le
Gothlandien de la Boheme. Vest. geol. IJst. csl. r. 7, c. 1, 1-14.

• 19316. Communication preliminaire sur quelques nouvelles especes de graptolites provenant du

Gothlandien de la Boheme. Ibid. r. 7, c. 3, 1-21.

■ 1932. Preliminary report on some new species of Graptolites from the Gothlandien of Bohemia.

Ibid. r. 8, c. 3, 1-5.

1933. Monographic der obersilurischen Graptoliten aus der Familie Cyrtograptidae. Place geol.

pal. ust. Karlovy Univ. 1, 1-84, pi. 1-7.

and pribyl, a. 1942. (jber bohmische Monograpten aus der Untergattung Streptograptus Yin.

Rozpr. ceske Akad. 52, 1-23, pi. 1-3.

— 1948 (see Pribyl, A. 1948).

— — 1952. On some slender species of the Genus Monograptus Geinitz, especially of the sub-
genera Mediograptus and Globosograptus. Rozpr. ceske Akad. 62, 1-32, pi. 1-3.
bronn, H. g. 1835. Lethaea Geognostica, 1, Stuttgart.

1846. Index Palaeontologicus, B. Enumerator, Stuttgart.

elles, G. L. 1900. The Zonal Classification of the Wenlock Shales of the Welsh Borderland. Quart.
J. geol. Soc. Loud. 56, 370-414.

■ and wood, e. m. r. 1901-18. Monograph of British Graptolites. Palaeontogr. Soc. [Monogr.].

frech, F. 1897. Lethaea Geognostica, 1, Graptolithen. Stuttgart.

geinitz, H. b. 1852. Die Versteinerungen der Grauwacken formation (Die Graptolithen), Leipzig.
gortani, m. 1922. Fauna paleozoic della Sardegna. Graptoliti di Goni. 1. Graptoliti della Sardegna
orientale. 2. Paleontogr. Ital. 28.

haberfelner, e. 1936. Neue Graptolithen aus dem Gotlandium Bohems, Bulgariens und der Karni-
schen Alpen. Geol. balk. 2.

jaekel, o. 1889. Uber das Alter des sogenannten Graptolithengesteins. Z. dtsch. geol. Ges. 41, 653-90,
pi. 28, 29.

kdhne, w. g. 1955. Unter Ludlow-Graptolithen aus Berliner Gescheiben. Neues Jb. Geol. Palaont.
Abh. 100, 3, 350-401. Stuttgart.

lapworth, c. 1875. In hopkinson and lapworth. The graptolites of the Arenig and Llandeilo rocks
of St. Davids. Quart. J. geol. Soc. Lond. 31, 631-72, pi. 33-37.

1876. On Scottish Monograptidae. Geol. Mag. (2), 3, 308-21, 350-60, 499-507, 544-52, pi. 10,

11, 12, 13, 20.

1880. On the geological distribution of the Rhabdophora. pt. 3, Results, and pt. 4, Conclusions.

Ann. Mag. nat. Hist. (5), 6, 16-29, 185-207.
meneghini, g. 1857. Paleontologie de Pile de Sardaigne, Turin.

R. B. RICKARDS: NEW SILURIAN GRAPTOLITES 271

munch, a. 1942. Die Graptolithenfauna des unteren Ludlow von Ronneburg und Ungebung. Beitr.
Geol. Thuring. 6, 241-66.

1952. Die Graptolithen aus dem anstehenden Gotlandium Deutschlands und der Tschecho-

slowakei. Geologica, Berl. 7.

nicholson, h. a. 1872. Monograph of British Graptolites, Edinburgh and London.
nicol, j. 1850. On the Silurian strata of the S.E. of Scotland. Quart. J. geol. Soc. Load. 6, 53-65.
packham, g. H. 1962. Some diplograptids from the British Lower Silurian. Palaeontology , Load. 5,
498-526, pi. 71, 72.

perner, j. 1894-9. Etudes sur les Graptolithes de Boheme. Prague.

pribyl, a. 1940m Revision der bohmischen Vertreter der Monograptidengattung Monoclimacis Freeh.
Rozpr. ceske Akad. 50, 1-16, pi. 1-3.

1940 b. O ceskych za stupcich monograptidu ze skupiny Pristiograptus nudus. Ibid. 50, 1-11, pi.

1-2.

■ 1941. O nekolika novych druzich graptolitu z ceskeho siluru. Ibid. 51, 1-10, pi. 1-2.

- — — 1942. Prispevek k poznani monograptidu zeskupiny druhu Monograptus flexilis. Ibid. 52, 1-12, pi.
1-2.

1944. Revision aller Vertreter der G a 1 1 u n g Pristiograptus aus der Gruppe P. dubius und P. vulgaris

aus dem bohmischen und auslandischen Silur. Ibid. 53, 1-49, pi. 1-4.

1948. Bibliographic Index of Bohemian Silurian Graptolites. Knihovna Geol. List. csl. 22, 1-96.

1952. Contribution to the knowledge of the Silurian graptolites of Bulgaria. Rozpr. ceske Akad.

53, 1-40, pi. 1-2.

1958. Ein neuer Beitrag zur Kenntnis der Bulgarischen Graptolithen. Bull. Inst. geol. Acad. Scien.

Bulg. 6, 107-37, pi. 1-3.

romariz, c. 1962. Graptolitos do Silurico Portugues. Rev. Fac. Sci. Lisboa, (2), C, 10, 115-312, pi.
1-22.

shotton, F. w. 1935. The stratigraphy and tectonics of the Cross Fell Inlier. Quart. J. geol. Soc. Loud.
91, 639-704.

tornquist, s. L. 1881. Om nagra Graptolitarter fran Dalarne. Geol. Foren. Stockh. Fbrh. 5, 434-45,
pi. 17.

1899. Researches into the Monograptidae of the Scanian Rastrites Beds. Acta Univ. land. 35,

1-25, pi. 1-4.

tullberg, s. a. 1883. Skanes Graptoliter, 2. Sverig. geol. Unders., C, 55, 1-43, pi. 1-4.
urbanek, a. 1958. Monograptidae from erratic boulders of Poland. Palaeont. polon. 9, 1-105, pi. 1-5.
waterlot, g. 1945. Les Graptolithes du Maroc; premiere partie: generalities sur les Graptolithes.
Notes. Serv. geol. Maroc. 63.

watney, g. r. and welch, e. g. 1911. The zonal classification of the Salopian Rocks of Cautley and
Ravenstonedale. Quart. J. geol. Soc. Lornl. 67, 215-37.
wood, e. m. r. 1900. The Lower Ludlow Formation and its graptolite fauna. Quart. J. geol. Soc. Loud.
56, 415-90, pi. 25.

yin, t. h. 1937. Brief description of the Ordovician and Silurian fossils from Shihtien. Bull. geol. Soc.
China, 16, 281-97.

R. B. RICKARDS

Sedgwick Museum,

Manuscript received 16 June 1964 Cambridge

THE DEVELOPMENT OF LASIOGRAPTUS
HARKNESSI (NICHOLSON 1867)

by r. b. rickards and o. m. b. bulman

Abstract. The development of Lasiograptus harknessi (Nicholson) is described, and its significance discussed.
A lectotype for the species is proposed. A new monotypic genus, Dicaulogrdptus, is erected with the type species
Lasiograptus hystrix Bulman 1932.

Within the family Lasiograptidae Lapworth 1879 little is known of the detailed
morphology of Hallograptus Lapworth, Neurograptus Elies and Wood, and Nympho-
graptus Elies and Wood. Of the remaining genera, Gymnograptus Bulman and
Lasiograptus Lapworth, at least fifteen species have been described, but the proximal
development is known in only four: G. linnarssoni (Moberg), G. retioloides (Wiman)
(Urbanek 1959 and Jaanusson 1960), L. hystrix Bulman 1932, and L. harknessi
(Nicholson), the development of which is described below.

Each of these four species has a different type of diplograptid development (text-
fig. 1). Thus in G. linnarssoni (text-fig. 1 a) th. I2 grows directly across and upwards;
th. 22 is the dicalycal theca, and the rhabdosome is cryptoseptate (Urbanek 1959).
G. retioloides (text-fig. 1 b), on the other hand, has th. I2 of the ‘basic’ diplograptid type;
th. 22 is the dicalycal theca; the origin of th. 21 and th. 22 is of distinctly primitive
appearance, and the rhabdosome is septate. L. hystrix has a ‘dentatus’ stage develop-
ment, with th. 21 the dicalycal theca (text-fig. lc).

The diplograptid development of L. harknessi, described herein, further complicates
the picture. In this species th. I2 grows directly across and downward; th. 21 appears
to be the dicalycal theca, and the rhabdosome is at first cryptoseptate and finally,
beyond the sicular apex, is septate.

Because of the obvious difficulty in assessing the taxonomic rank of the type of
proximal development, workers have used thecal form, nature of clathria, and presence
of lacinia as a means of classification (Urbanek 1959, Jaanusson 1960, Lee 1963). On
this basis Jaanusson considers that the species L. hystrix has gymnograptid thecae
throughout the length of the rhabdosome; and he doubtfully refers it to the genus
Hallograptus on the supposition that upon flattening of the rhabdosome L. hystrix
would acquire a general hallograptid appearance. The placing of L. hystrix, even doubt-
fully, in the genus Hallograptus is here considered incorrect. The published figures of
hallograptid species, and the specimens examined by the writers in the Sedgwick Museum
collections, whilst invariably of flattened material, do seem to have thecae which vary
from lasiograptid or gymnograptid proximally, to gymnograptid or orthograptid
distally. The thecae of L. hystrix, however, are unique in both the degree and nature of
the introversion of the apertural margin, and the nature of the thecal spines. Indeed,
the thecal form is more reminiscent of the Dicranograptidae than the Lasiograptidae.
It seems to the writers that hystrix should be referred to a new, monotypic genus, and
the name Dicaulograptus gen. nov. is here proposed.

[Palaeontology, Vol. 8, Part 2, 1965, pp. 272-80.]

R. B. RICKARDS AND O. M. B. BULMAN: LASIOGRAPTUS HARKNESSI 273

L. harknessi (Nicholson) is therefore the only species, with known proximal develop-
ment, which can be assigned to the genus Lasiograptus.

Material. A single specimen of L. harknessi was isolated by Bulman (1947) from the
Laggan Burn limestones. This specimen is now deposited in the Geological Survey
Museum (no. 74259). The proximal end is opaque and the details of development can-
not be ascertained.

At a later date about a dozen specimens were isolated, cleared, and mounted by Miss
J. James, and of these, four are well preserved, showing the thecal relationships except
at the extreme proximal end. A further specimen, isolated by Miss James, has been

text-fig. 1. Mode of development of a, Gymnograptus linnarssoni (after Urbanek 1959); b, G. retio-
loides (after Urbanek 1959); c, Dicaulograptus hystrix (after Bulman 1932); d, L. harknessi. In each

case the dicalycal thecae is shown in black.

cleared and mounted by the writers and this specimen shows the development of the
extreme proximal end very clearly.

Specimens of L. harknessi from the Laggan Burn section are extremely difficult to
mount, owing to a tendency to collapse under their own weight, if the inside of the
rhabdosome is not filled with liquid during the whole of the clearing and mounting
process. The specimen was therefore transferred in a drop of alcohol to a slide upon
which the centre of a large blob of euparal gum had been thinned considerably with
euparal essence. The alcohol, and specimen, sank into the euparal. (At this stage, if
the euparal is not thinned, the alcohol merely runs off leaving the specimen to collapse.)
A thin sliver of blotting paper was then inserted into one of the thecal tubes, and as the
alcohol was drawn out of one end of the rhabdosome the euparal entered at the other,
thus supporting the rhabdosome continuously. All the alcohol was removed in this
manner.

Of the six isolated specimens, five are deposited in the Sedgwick Museum, Cambridge
(nos. A54,975-A54,979), and the sixth in the Geological Survey Museum (no. 74259).
The lectotype is deposited in the British Museum (Nat. Hist), specimen no. Q53c.

Acknowledgements. We are indebted to Dr. H. W. Ball of the British Museum (Nat. Hist.), Dr. D. E.
White of the Geological Survey Museum, and Mr. A. G. Brighton of the Sedgwick Museum, for the
loan of material. Our thanks are also due to Dr. I. Strachan for many helpful suggestions, and to Mr.
E. A. Tait of the Department of Geology, University of Aberdeen, for his search for Nicholson’s type
specimen.

274

PALAEONTOLOGY, VOLUME 8

SYSTEMATIC DESCRIPTION

Order graptoloidea Lapworth 1875
Suborder diplograptina Lapworth 1880 emend Bulman 1963
Family lasiograptidae Lapworth 1879
Genus lasiograptus Lapworth 1873

Lasiograptus harknessi (Nicholson 1867)

Text-figs. 1-3

1867 Diplograpsus Harknessi Nicholson, p. 262, pi. 11, figs. 6, 7.

1908 Lasiograptus ( Thysanograptus ) Harknessi (Nicholson); Elies and Wood, pp. 325-6,
pi. 34, figs. 1 a-d, text-fig. 214.

1947 Lasiograptus harknessi (Nicholson); Bulman, pp. 71-72, pi. 8, figs. 11, 12.

1949 Lasiograptus harknessi (Nicholson); Sherrard, p. 75, pi. 2, fig. 14, text-fig. 21a-c.

1963 Lasiograptus harknessi (Nicholson); Geh Mei-yu, pp. 254-5, pi. 5, (?) fig. 17, fig. 18;
text-fig. 13 a.

Lectotype. The specimen figured by Elies and Wood 1908, p. 326, text-fig. 2146, and now contained in
the British Museum (Nat. Hist.) as specimen no. Q53c.

The counterpart of this specimen was figured as pi. 34, fig. 1 a by Elies and Wood (op. cit) and was
doubtfully referred to the type specimen. Unfortunately the slab from which fig. 1 a was drawn is
now missing. However, the counterpart (Q53) of this slab contains, in addition to Q53c, two proximal
ends of L. harknessi one of which was figured by Elies and Wood as text-fig. 214 a. Q53c very closely
resembles one of Nicholson’s original figures (pi. 11, fig. 6) and it is significant that in his description
(p. 262) he records that this slab contained one adult and ‘two or three germs’. There can be little
doubt that slab Q53 is the counterpart of the slab from which Nicholson drew his description and
figures, and from which Elies and Wood figured pi. 34, fig. 1 a. In view of this the adult specimen
(Q53c) is here proposed as lectotype of the species L. harknessi (Nicholson).

Material. Six proximal ends in relief; isolated, cleared, or partially cleared, and mounted on glass
slides. Numerous specimens, including the lectotype, preserved as flattened films.

Horizon and localities. Corynoides Band, Caradoc Series, Laggan Burn, Ayrshire; Hartfell Shales,
Hartfell, Dumfries.

Revised diagnosis. Rhabdosome short; usually 2 mm. or less in width (excluding pro-
cesses); thecal spacing variable, 10-16 in 10 mm., the highest values being at the proxi-
mal end; thecae of lasiograptid type; excavations deep, inclined proximally in dorsal
parts of excavation, apertural and pleural lists well developed, parietal lists weak, con-
necting apertural region to strong aboral lists; development diplograptid, rhabdosome
cryptoseptate after th. 21 and septate after 32; lacinia present (but not preserved in the
isolated material) ; nema prominent, undulating in profile view.

Development. The broad-based sicula is a prominent feature of the proximal part of the
rhabdosome. It is exposed on the obverse side for about 1 mm. of its length, and on the
reverse side for about one-tenth of a millimetre (text-fig. 3 f g). The sicular aperture
possesses a robust virgella and, opposite this, a pair of equally robust spines. The apex
of the sicula lies at the level of the supragenicular wall of th. 31.

The prosicula is preserved in two specimens and in the one measurable instance
(no. A54,975) is 0-45 mm. in length. No structural details can be ascertained other than
a faint longitudinal pattern and a thickening near the apex.

R. B. RICKARDS AND O. M. B. BULMAN: LASIOGRAPTUS HARKNESSI 275

The metasicula is about 1-33 mm. long and widens from 0T3 mm. initially to 040
mm. at its aperture. Close to the junction with the prosicula the growth-lines are
approximately 0-013 mm. apart, but at the sicular aperture are about 0-03 mm. apart.

Th. I1 originates at a point 0-65-0-75 mm. from the base of the sicula, and grows
downwards, adpressed to the sicula, for 0-75-0-88 mm. before turning upwards and
outwards in a characteristic trumpet shape. There is, of course, no geniculum on th. I1
but a single spine arises from a homologous position. The ventral wall above the spine
varies from nearly vertical to slightly inclined towards the axis. The apertural margin
is almost horizontal in some specimens and undulating in others. In the latter cases the
margin is excavated at the dorsal edges of both lateral walls to a depth of up to 0-07
mm. (see text-fig. 2c) and two growth-bands may be incompletely developed. On th. I1
the growth-lines are initially more closely spaced than on the adjacent part of the sicula,
but near the thecal aperture are about 0-03 mm. apart.

The initial bud of th. I2 begins to form on th. I1 when only 0-08-0-13 mm. of the latter
have been laid down, and is not complete until th. I1 is 0-3-04 mm. long.

Th. I2 grows immediately downwards from its broad initial bud and extends almost
to the apertural region of the sicula. It begins to turn upwards when it has reached
a length of 0-25 mm. and gives rise to a narrow initial bud of th. 21 (text-fig. 3c). The
distal part of th. I2 is not dissimilar to that of th. I1 in that no geniculum is developed
but a spine arises from a homologous position. The ventral wall above the spine is
almost vertical and, as in the case of th. I1, the apertural region is less undulating than
in the more distal thecae. Near the aperture the growth-lines are approximately 0-03
mm. apart.

Viewed from the reverse side the bud of th. 21 is on the level of the sharp bend in the
dorsal wall of th. I1 (text-fig. 3c), and a transverse spine is positioned on its dorsal side.
The separation of th. 22 from th. 21 is complete after half a millimetre (text-fig. 3d).

The thecae beyond th. I2 are more typically lasiograptid. The excavations are deep,
and in the dorsal region are inclined proximally. The supragenicular wall becomes
increasingly inclined towards the axis as successive thecae are developed. Paired spines
are developed from the genicula. Successive thecae have similar numbers of growth
segments (14-18 on the free ventral wall, 4-6 being in a supragenicular position) but
the spacing of the growth-lines increases to about 0-07 mm. at the point where the
maximum thecal size is reached (about th. 31 or th. 32). Thereafter no change can be
detected.

Occasionally the proximal extension of the thecal excavation is filled by secondary
growth-bands (text-fig. 2a) in a manner similar to that described by Urbanek (1959) in
the case of Gymnograptus linnarssoni.

The apertural and pleural lists are well developed, but the parietal lists are weak.
Towards the point where they connect with the apertural lists, the pleural lists become
somewhat weaker and in this region the passage of the growth bands across the lists
can be seen (text-fig. 2c).

The probable origin of th. 22 from th. 2l has been considered above, but th. 21 also
possesses a conspicuous foramen flanked by a robust aboral list (text-fig. 2c). Th. 21
is, on this interpretation, the dicalycal theca and gives rise to both th. 22 and th. 31.
However, there is no trace of a median septum until the apex of the sicula is reached.
At the apex of the sicula is a strong transverse rod, and in the angle this makes with the

276

PALAEONTOLOGY, VOLUME 8

virgula a portion of median septum is preserved (text-fig. 2b). The absence of the
median septum prior to the sicular apex could be explained as a preservational feature,
but this is unlikely in view of the presence of other equally delicate structures, and,
indeed, of the median septum itself beyond the sicula.

text-fig. 2. a, Part of specimen no. S.M. A54,976, obverse view, showing relationships of thecae near
sicular apex, X 40 approx, b, Same specimen, portion of median septum preserved between nema and
transverse list, reverse view, Xll5 approx, c, Simplified camera-lucida drawing of specimen no.
S.M. A54,975, reverse view, showing dicalycal nature of th. 21, and origins of th. I1, l2, 21, and 22;
heavy stipple indicates badly preserved areas. x45 approx, n = nema, c = cortical tissue, i =
median septum, 1 = transverse list, s = spine, a = aboral list, ap = apertural list, pi = pleural list.

If the arrangement of growth-lines on the lateral walls of the rhabdosome is examined
(text-figs. 3 f g ) the proximal end might, at first inspection, be deduced as being aseptate

text-fig. 3. a-d, Development of L. harknessi: taken from specimen no. S.M. A54,975. Figures are
not growth stages, a, Origin of th. I1, obverse view, b. Reverse view showing initial bud of th. I2,
c. Further development of th. I2 and initial bud of th. 21. d, Further development of th. 21 and origin
of th. 22. e, Transparent reconstruction of rhabdosome. f-g, Reverse and obverse views respectively,
of reconstructed rhabdosome showing growth-lines as they appear on lateral walls. All figures X 25
approx., p = prosicula, m = metasicula, v = virgella, n = nema, i = median septum, 1 = transverse
list, s = spine, a = aboral list, p = parietal list, ap = apertural list, pi = pleural list, f = foramen,

se = interthecal septum.

TEXT-FIG. 3.

278

PALAEONTOLOGY, VOLUME 8

up to 32. But the most proximal growth segments of th. 31 appear to be intimately re-
lated to th. 21 (text-figs. 2c, 3/ g ), which implies that two separate series are developed
from th. 23.

In view of the above evidence it is considered that a cryptoseptate arrangement exists
between th. I2 and th. 41. This interpretation means that the thecae 23, 22, 31, and 32 are
interconnected.

A puzzling feature which, for preservational reasons, cannot be fully explained, is
the presence of transverse rods at certain points on the rhabdosome. These occur at
the proximal extremities of the thecal foramina of the second series, and in some cases
at least project beyond the lateral walls of the rhabdosome as spines. The first such rod
is found about half-way along the sicula near the proximal extremity of the foramen of
th. 22. It projects from both sides of the sicula (but does not pass through it) and one
specimen (S.M. no. A54,976) shows that it extends on the obverse side as a spine. The
growth-lines at the proximal end of th. 31 appear to be connected with the rod in some
way, but the rod itself is so heavily pigmented that the exact relationship remains
obscure.

The second transverse rod is at the apex of the sicula, which point coincides with the
proximal extremity of the foramen of th. 32. The growth-lines at the proximal end of
th. 41 appear to be directly connected to the rod (text-fig. la).

Beyond the second rod a true septum is present, but it cannot be ascertained whether
any further rods exist. It is not impossible that the rods described are in fact homo-
logous with septal lists.

Remarks. The Laggan Burn specimens described above are assigned to Nicholson’s
species L. harknessi mainly on the grounds of rhabdosomal width. Thus whilst our
largest specimen (Geol. Surv. no. 74259) is of similar width at the level of th. 4 to the
type specimen of Lapworth’s species L. costatus, it is thereafter parallel-sided and at
the level of th. S2^1 is only T5 mm. wide. There is no indication of any greater widening
distally.

INCERTAE SEDIS

Genus dicaulograptus gen. nov.

Type species. Lasiograptus hystrix Bulman 1932.

Diagnosis. Rhabdosome minute, slender, slightly less than 1 mm. wide at origin, IT
mm. wide at level of second pair of thecae and thereafter parallel-sided. Thecae about
20 in 10 mm., scarcely overlapping, provided with long mesial spines. Th. I1 and th. I2
provided with pair of slender lateral spines in region of aperture; succeeding thecae
with pair of slender lateral spines at proximal end of free ventral wall. Apertural margin
of all thecae strongly introverted; in th. 21 and later thecae the introverted lip becomes
fused with lateral portion of apertural margin resulting in a pair of lateral foramina.

Derivation of name, cauta, L. opening, hole.

CONCLUSIONS

It is clear from the above that each of the four species, in which the proximal develop-
ment is known, has a different mode of diplograptid development. Until more species

R. B. RICKARDS AND O. M. B. BULMAN: LASIOGRAPTUS HARKNESSI 279

are known in comparable detail the mode of development cannot shed much light on
the classificatory or evolutionary problems within the Lasiograptidae. Thecal form
remains the only basis for classification; that it may also be a guide to evolution has
been suggested by Urbanek (1959, pp. 327-9).

Urbanek considers that Lasiograptus may have developed from an ancestor with
amplexograptid thecae and, like Jaanusson (1960, p. 335), implies that Lasiograptus
has uniform thecae. That this is not the case with L. harknessi has been demonstrated
above. The presence of ‘angular fuselli’ in harknessi also places in doubtful standing
another of Urbanek’s distinctions between Gymnograptus and Lasiograptus. The main
distinctions between the two genera would seem to be the length and inclination of the
supragenicular wall, the presence or absence of a zigzag list on the lateral walls of the
rhabdosome, and, to a lesser extent, the nature of the excavation.

The presence or absence of a lacinia is of more doubtful value in taxonomy, if only
for the reason that it is rarely preserved, and it is here considered that the erection of the
genus Prolasiograptus by Lee (1963), for those early Lasiograptus species apparently
lacking lacinia, is impractical.

The recognition of a cryptoseptate development, in part, for L. harknessi, is in
accordance with the work of Urbanek and Jaanusson, and it is perhaps to be expected
in genera in which the periderm is being reduced. A further point of significance is that
in those species in which the internal structure is not known, but in which the growth-
lines of the lateral rhabdosomal walls suggest an aseptate development, the presence
of an early dicalycal theca, and hence two series, may be obscured. Finally, in those
species having a partial median septum, rhabdosomal control may still be effected by
an early dicalycal theca. It is not impossible that a species may have a septate develop-
ment on one side of the rhabdosome, and a cryptoseptate development on the other.
On the cryptoseptate side of the rhabdosome the growth-lines might suggest an aseptate
arrangement of thecae.

REFERENCES

bulman, o. m. b. 1932. On the graptolites prepared by Holm. 1. Certain ‘Diprionidian’ graptolites
and their development. Ark. Zool. 24A, no. 8, 1-46, pi. 1-9.

■ 1947. A monograph of the Caradoc (Balclatchie) graptolites from limestones in Laggan Burn,

Ayrshire. Palaeontogr. Soc. [Monogr.].

1963. The evolution and classification of the Graptoloidea. Quart. J. geol. Soc. Loud. 119,

401-18.

elles, g. L. and wood, e. m. r. 1908. A monograph of British graptolites, pt. 7. Palaeontogr. Soc.
[Monogr.].

geh mei-yu. 1963. Graptolites from the Miaopo Shale (Middle Ordovician), W. Hubei (11). Acta
palaeont. sin. 11, 240-61, pi. 3-5.

jaanusson, v. 1960. Graptoloids from the Ontikan and Viruan (Ordov.) limestones of Estonia and
Sweden. Bull. geol. Inst. Univ. Uppsala, 38, 289-366, pi. 1-5.
lapworth, c. 1875. In hopkinson, j. and lapworth, c. Descriptions of the graptolites of the Arenig
and Llandeilo rocks of St. Davids. Quart. J. geol. Soc. Loud. 31, 631-72, pi. 33-37.

1879. On the geological distribution of the Rhabdophora. pt. 1, Historical. Ann. Mag. nat. Hist.

(5), 3, 245-57, 449-55.

1880. On the geological distribution of the Rhabdophora. pt. 3, Results and pt. 4, Conclusions.

Ibid. (5), 5, 273-85, 358-69; 6, 16-29, 185-207.

lee, c. k. 1963, Some Middle Ordovician graptolites from Gueizhou. Acta palaeont. sin. 11, 554-78,
pi. 1.

C 3009

u

280 PALAEONTOLOGY, VOLUME 8

nicholson, H. A. 1867. On a new genus of graptolites, with notes on reproductive bodies. Geol. Mag.
4, 256-63, pi. 11.

sherrard, k. 1949. Graptolites from Tallong and the Shoalhaven Gorge, New South Wales. Proc.
Linn. Soc. N.S. W. 74, pts. 1-2, 62-82, pi. 1-2.

urbanek, A. 1959. Studies on graptolites 11. On the development and structure of graptolite genus
Gymnograptus Bulman. Acta palaeont. polon. 4, 279-338, pi. 1-2.

R. B. RICKARDS and O. M. B. BULMAN
Sedgwick Museum,

Manuscript received 13 June 1964 Cambridge

A NEW FERTILE LYCOPOD FROM THE
LOWER CARBONIFEROUS OF SCOTLAND

by K. L. ALVIN

Abstract. The morphology and anatomy of the leafy stems and cone are described and attributed to Oxroadia
gracilis gen. et sp. nov. The leaves were recurved and show no evidence of abscission. The plant was probably
herbaceous. Comparison is made with other lycopods based on petrifactions and also with forms based on com-
pressions. Megaspores closely associated with the cone are described under Triletes subpalaeocristatus sp. nov.

The material upon which this study is based consisted of a single limestone block
containing the petrified remains of several vegetative stems and a fertile axis belonging
to a hitherto unknown lycopod. The block was collected from the shore at Oxroad
Bay, East Lothian, Scotland. The matrix of the block resembles that of similar blocks
collected from the volcanic ash cliffs at the same locality (Gordon 1938, Barnard 1959).
There is no reason to doubt that the block came originally from the cliffs. The outcrop
belongs to the Cementstone Group of the Calciferous Sandstone Series.

The preservation is generally rather poor. Much crystalization of the matrix has
occurred and iron pyrites has been precipitated round many of the plant remains. How-
ever, the xylem is usually very well preserved, and so too very often is the epidermis,
especially the cutinized outer walls; small portions of certain other tissues are occasion-
ally fairly well preserved.

The material has been studied mainly by close serial peel sections prepared by the
now well-known method of Joy, Willis, and Lacey (1956).

SYSTEMATIC DESCRIPTION
Genus oxroadia gen. nov.

Diagnosis. Lycopod with dichotomously branched stems without distinct leaf cushions
but with more or less decurrent leaf bases. Stem protostelic, exarch; leaf traces mesarch.
No secondary thickening. Leaves spirally arranged (but longitudinal ranks also well
marked), eligulate, dorsi-ventrally flattened, with single median vein; attachment to
stem narrow, round; no leaf abscission; no parichnos. Stomata in two bands on the
lower surface. Sporophylls in a terminal strobilus, spirally arranged, eligulate. Sporan-
gium elongated, attached to the horizontal basal part of the sporophyll by an elongated
attachment; sub-archesporial sterile pad present.

Oxroadia gracilis sp. nov.

Plates 32, 33, 34, figs. 1-5; text-figs. 1, 2, 3a-d

Diagnosis. Stem (2-5— )5-0(— 6-5) mm. in diameter. Leaves arranged in 2/11 spiral. Leaf
bases somewhat decurrent giving the transverse section of the stem an approximately
hexagonal outline. Stem protostele solid, typically with eleven protoxylem points. Leaf

[Palaeontology, Vol. 8, Part 2, 1965, pp. 281-93, pi. 32-34.]

282

PALAEONTOLOGY, VOLUME 8

traces arising successively from every alternate protoxylem point. Tracheids of meta-
xylem scalariform with reticulate fibrils between the bars. Outer part of cortex con-
sisting of compactly arranged, non-seriated cells, round or polygonal in transverse
section, rectangular in longitudinal section. Epidermal cells (surface view) 70-100 y.
long and 20-30^ broad, walls 3-5/x thick, generally with asymmetrical, bluntly pointed
ends. Leaf recurved, ovate-lanceolate, 6-8 mm. long, expanding to about 3 mm. broad
at the widest part from about 1 mm. at its attachment; margin smooth. Mesophyll
without a palisade. Xylem strand surrounded by a zone of transfusion tracheids.
Stomata arranged in irregular longitudinal rows in the stomatal bands. Stoma sur-
rounded by (5— )6(— 7) polygonal cells not different from the other epidermal cells within
the stomatal bands. Epidermal cells of the midrib region more elongated and lying in
longitudinal rows. Cells of upper epidermis polygonal.

Strobilus at least 4 cm. long, about 1 cm. in diameter. Axis 1-1-5 mm. in diameter,
with solid or hollow exarch protostele. Sporophyll with horizontal basal portion 4 mm.
long with narrow attachment to the axis but expanding to about 2 mm. broad and
thickening to about 1 mm. at the ‘heel’; upturned part about 3 mm. long. Sporangium
borne on an elliptical attachment about 2-0 X 0-7 mm. Xylem strand becoming lost in
numerous transfusion tracheids towards the end of the horizontal part of the sporophyll.

Locality and horizon. Oxroad Bay, East Lothian, Scotland. Cementstone Group (Upper Tournaisian),
Calciferous Sandstone Series, Lower Carboniferous.

Holotype Material. British Museum (Natural History), Palaeontology Department, V 51512-13.
Detailed description

Morphology of the vegetative shoot. Something of the external features of the shoot
can be seen from the specimen exposed on a broken surface of the block and illustrated
in Plate 32, fig. 1. One leaf is seen attached on the left and the stem surface shows
a number of leaf bases indicating quite well the leaf arrangement. The surface here
appears quite smooth, but this is not the genuine external surface of the stem. The
epidermis is situated in the outermost dark layer just beyond the thin light layer seen in
the cross-section at the bottom of the figure. Judging from peel sections (PI. 32, fig. 2),
however, the external surface of the stem was in fact smooth except for about six obtuse
angles representing the decurrent leaf bases; these were not separated by furrows.

The phyllotaxis is of interest. From series of transverse sections it is evident that leaf
traces are given off from the protoxylem points of the stem stele successively from every
alternate one. As there are eleven points which persist, it follows, provided that the
points run vertically, that the phyllotaxis is a 2/11 spiral. Evidence that this is so can

EXPLANATION OF PLATE 32

Figs. 1-7. Oxroadia gracilis gen. et sp. nov. 1, Shoot exposed on broken surface of the block showing
leaf bases and one attached leaf. The surface of the stem does not represent the genuine external
surface but a layer several cells below the epidermis. Specimen V51512a. X 5. 2, Transverse section
of stem. Slide V51513a. X 14. 3, The stem stele with protoxylem points and leaf traces and part of
a band of preserved cortical tissue. Slide V5 1513b. X 60. 4, Portion of the protostele enlarged showing
the departure of a leaf trace. Slide V51513b. X240. 5, Single leaf trace from stem cortex. Slide
V51513b. X240. 6, Portion of a tracheid in longitudinal section showing the reticulate fibrils be-
tween the scalariform bars. Slide V51513c. X 1200. 7, Cuticle (top), epidermis and hypodermis
(bottom) of the stem in surface section. Slide V51513r. x 120.

Palaeontology, Vol. 8

PLATE 32

ALVIN, Lower Carboniferous lycopod

K. L. ALVIN: A NEW FERTILE LYCOPOD FROM SCOTLAND

283

also be obtained from isolated transverse section (text-fig. 1a). Moreover, that there are
about eleven ranks of leaves can also be determined by dividing the stem circumference
by the measured distance apart of adjacent ranks in the specimen in Plate 32, fig. 1.
The diagram in text-fig. 1b is based on this same specimen.

The 2/1 1 spiral itself is remarkable in that it is not one of the Fibonacci fractions.
Apart from this, the phyllotaxis is of interest because it gives not only very apparent
longitudinal rows of leaves, but also alternating pseudowhorls formed by the turns of
the rather flat spiral (text-fig. 1b). It would appear that this sort of leaf arrangement was

text-fig. 1. Oxroadia gracilis gen. et sp. nov. A, Camera-lucida
drawing of stele and leaf traces in transverse section; numbers
indicate the order in which the traces have arisen from the pro-
toxylem points. Slide V51513q. X 60. b, Diagrammatic recon-
struction of part of the stem showing the leaf arrangement;
dotted lines indicate the turns of the spiral (= pseudowhorls);
broken lines, the vertical ranks or orthostichies.

frequent amongst early lycopods. It also occurs in Lycopodium. This is discussed further
below.

The leaves are strikingly recurved so that the extreme leaf tip is approximately
parallel to the stem (PI. 33, fig. 1). The base of the leaf is about TO mm. wide and round
or oval in section, but the leaf expands into a flattened lamina 3 mm. broad before
tapering to a point. The base of the leaf just beyond its attachment is generally very
badly preserved (PI. 33, fig. 1). However, careful examination of many leaf bases has
revealed no evidence of a ligule.

Anatomy of the stem. The most striking feature of the transverse section is the central
protostele (PI. 32, figs. 2, 3). The solid xylem strand possesses typically eleven proto-
xylem points from which the mesarch leaf traces are given off (PI. 32, fig. 4).

The metaxylem consists of scalariform tracheids mainly 30-50/1. in diameter. The
protoxylem elements also appear to be scalariform, or possibly more or less reticulate.
Between the bars of thickening of probably all tracheids, including transfusion tracheids,
there is a network of fibrils (PI. 32, fig. 6; text-fig. 2g) resembling basically that found
in many fossil lycopods. As reported by Smith (1962) in certain arborescent forms,
the fibril reticulum together with a narrow strip of material along the edges of the

284

PALAEONTOLOGY, VOLUME 8

scalariform bars appears to consist of a different, more translucent material from that
of the bars themselves. An indication of this is seen in the photograph (PI. 32, fig. 6);
no attempt has been made to show it in the drawing (text-fig. 2g).

text-fig. 2. Oxroadia gracilis gen. et sp. nov. a, Stem epidermis in surface view. Slide V51513r. X 280.
b. Distribution of stomata in stomatal band. Slide V5 1 5 1 3s. x65. c, d. Stoma. Slide V51513v. X465
(solid black represent mineral), e, Lower epidermis of leaf in midrib region. Slide V5 1 5 1 3t. x465.
F, Upper epidermis of leaf. Slide V51513u. X465. g. Portion of tracheid showing reticulate fibrils
between the scalariform bars. Slide V51513d. X 1100.

Outside the xylem in the best-preserved specimens there is a narrow zone containing
the remains of a delicate tissue which may reasonably be assumed to represent the
phloem (PI. 33, fig. 4). This is enclosed by a ring of larger-celled, better-preserved tissue
which is interpreted as part of the inner cortex. In most specimens the phloem is entirely
absent and except for a narrow belt of inner cortex (PI. 32, fig. 3) there is nothing

K. L. ALVIN: A NEW FERTILE LYCOPOD FROM SCOTLAND 285

between the xylem and the outer part of the cortex (PI. 32, fig. 2). The outer cortex itself
is nearly always conspicuous as a dark zone, although its preservation, except for
a narrow belt of cells on the inside (bordering the space) and some of the superficial
cells, is poor. These alternating zones of preserved and unpreserved tissues are a pecu-
liar feature of most of the stems in the material. Occasionally, however, the preserva-
tion is different and the tissues are more continuous (PI. 33, fig. 4). I believe that the
cortex was originally differentiated into two zones, an outer consisting of compact
parenchyma, possibly rather thick walled, and an inner zone of delicate parenchyma.
Even in the best-preserved specimens the inner cortex is never well preserved and some-
times there is a suggestion that this region may have been aerenchymatous (PI. 33,
fig. 4). However, no definite aerenchyma has been observed. The alternation of pre-
served and unpreserved regions probably resulted from the early breakdown of the
softest tissues, i.e. phloem and inner cortex, and the penetration of these spaces by
petrifying mineral. This mineral preserved the immediately adjacent tissues, i.e. the
xylem and the narrow strip of cortical tissue close to the stele and just inside the dark
outside zone; the epidermis and hypodermis were presumably penetrated and preserved
by mineral external to the stem. The intervening outer cortical tissue were then destroyed
to give the dark residual material seen in most specimens.

There is no indication of any radial seriation of cells in the cortex, and thus no sug-
gestion of secondary activity.

The epidermis is usually remarkably well preserved (PI. 32, fig. 7; text-fig. 2a). The
outer walls seen in surface section have a striking yellowish-brown colour due pre-
sumably to the presence of a fairly thick cuticle.

The leaf traces pass through the cortex rather steeply so that as one trace departs
from a protoxylem point the last one in the same orthostichy has not yet entered its
leaf. Indeed, the trace departs at a level corresponding to about two leaves below the
one into which it eventually passes, a distance of about 11 mm. The trace remains
mesarch with rather more centrifugal than centripetal metaxylem. Nothing is preserved
of the trace except the xylem of which there are generally some twenty tracheids. As
the trace passes through the preserved regions of the cortex it is accompanied by a small
space which presumably represents the phloem (PI. 32, fig. 5).

Branching. Three specimens have been observed showing evidence of branching.
The best one was traced through the block for a distance of some 7 cm. during which
it dichotomized once (PI. 33, figs. 2, 3). The distance between these two sections is
approximately 1-5 cm. The dichotomy is evidently equal. The protostele of the parent
axis divides into two equal portions by a median suture (PI. 33, fig. 3). Five or six new
protoxylem points arise along the metaxylem adjacent to the suture in each of the
daughter steles. The specimen illustrated in Plate 33, fig. 4, contains two steles of some-
what different appearance suggesting that this may represent a kind of unequal dicho-
tomy. However, in a distance of some 4 mm. this specimen shows no perceptible
change and there is no proof that the two steles have arisen by division. One side of
this specimen (top of photograph) was broken and the inner cortex was open to the
outside.

Anatomy of the leaf. The leaves are generally very poorly preserved except for the
vascular strand and the epidermis (PI. 33, fig. 1).

At the base of the leaf the vascular trace is small and compact resembling the leaf

286 PALAEONTOLOGY, VOLUME 8

trace in the stem cortex, but as the leaf expands the strand acquires a sheath of trans-
fusion tracheids.

The mesophyll is very badly preserved except for small patches close to the epidermis
or at the extremities of the leaf (PI. 33, fig. 5). It was apparently more or less uniform in
structure.

The essential features of the epidermis including the stomata are given in the specific
diagnosis and illustrated in text-fig. 2b-f.

The strobilus. It is believed on the evidence of anatomy as well as association that the
single badly distorted fertile axis found in the same block as the vegetative shoots
belongs to the same plant.

The fertile axis is more slender than the smallest vegetative stem; its diameter de-
creases from T5 mm. at the proximal end to only TO mm. at the distal end. It has been
traced through the block for a total distance of about 4 cm. The fact that the axis is
small and decreases upwards suggests that it represents a terminally borne strobilus
rather than an intercalary fertile region.

At the thicker end of the axis a transverse section shows a solid protostele exactly
similar to that of a vegetative stem. Through the greater part of the axis, however, the
protostele is hollow (PI. 34, fig. 2). There is no evidence of parenchyma at the centre :
possibly the central tracheids failed to mature. There are about eleven protoxylem
points, and the sporophyll traces arise from them in a similar manner to leaf traces in
the stem. The traces themselves are similar (PI. 34, fig. 3). The pitting of the tracheids is
identical to that in the stem. The cortex, though much less massive than in the stem, is
probably similar in structure ; there is always a space inside the preserved outer region
(PI. 34, fig. 2).

Sporophylls are rather infrequently found actually attached to the axis; they mostly
lie in a rather distorted mass around it. The form of the sporophyll has been elucidated
by tracing individual examples through serial peels. It consists of a lower, horizontal
or possibly somewhat descending portion about 4 mm. long carrying the sporangium
on its upper side, and an upturned distal portion which is always very badly preserved
so that its exact size and shape is uncertain. The sporangium-bearing portion is slender
at the base but gradually expands both in breadth and thickness into a massive ‘heel’
beyond the sporangium (text-fig. 3a-d). As the expansion occurs, the slender vascular
bundle becomes surrounded by a mass of transfusion tracheids until eventually in the
‘heel’ the strand itself seems to become entirely lost (text-fig. 3c). Only a few transfusion
tracheids appear to enter the upturned portion. Beneath the sporangial attachment,
some transfusion tracheids pass towards the sporangium but do not enter it (PI. 34,
fig. 5; text-fig. 3b).

All the sporangia have dehisced. Generally the remains of the sporangial wall have

EXPLANATION OF PLATE 33

Figs. 1-5. Oxroadia gracilis gen. et sp. nov. 1, Longitudinal section of stem with attached leaves. Slide
V51513d. X 5. 2, Transverse section of stem at point of dichotomy. Slide V51513e. x9. 3, Section
of the stele of the same specimen as in fig. 2 below point of dichotomy. Slide V51513f. x45. 4,
Transverse section of a stem with a better preserved but broken cortex and two steles. (For further
explanation see text.) Slide V51513g. X40. 5, Longitudinal section through the base of a leaf show-
ing upper epidermis, apparently homogeneous mesophyll and the vascular strand. Slide V51513h.
x 60.

Palaeontology, Vol. 8

PLATE 33

ALVIN, Lower Carboniferous lycopod

K. L. ALVIN: A NEW FERTILE LYCOPOD FROM SCOTLAND

287

text-fig. 3. a-d, Oxroadia gracilis gen. et sp. nov. E, F, Triletes subpalaeocristatus sp. nov. a, Nearly
radial longitudinal section through a sporophyll. Slide V51513w. x20. b, Vertical section through
a sporophyll passing through sporangial attachment. Slide V51513o. x20. c. Somewhat oblique
vertical section distal to sporangial attachment near the sporophyll ‘heel’ showing distribution of
transfusion tracheids. Slide V51513x. x20. d. Longitudinal section through horizontal portion of
sporophyll in dorsi-ventral plane showing part of the vascular system and portions of the wall of the
dehisced sporangium. Slide V51513y. x20. E, Reconstruction of Triletes subpalaeocristatus sp. nov.
(proximal view), f, Hairs from the trilete ridge. Slide V51513m, n. x215. p, subarchesporial pad;

5, wall of sporangium; t, transfusion tracheids.

recurved round the basal portion of the sporophyll (text-fig. 3b-d). I estimate that the
sporangium was about 4x1 mm. The attachment, which in text-fig. 3 appears short, is
in fact elongated radially. This is because the section is not perfectly radial and does
not pass through the whole length of the attachment. The elongated form of the attach-
ment has been demonstrated by following through serial peels.

A striking feature of the sporangium is the massive pad of parenchyma forming
a short, wide columella extending up into the sporangium from the sporophyll. This
tissue also spreads along the sporangial wall where it resembles inner layers of wall cells.

288

PALAEONTOLOGY, VOLUME 8

The tissue appears homogeneous and consists of large, thin-walled cells. The outermost
layer of the sporangial wall is always conspicuous. It is about 25 p thick, palisade-like
in cross-section, and in section parallel to the surface it resembles collenchyma, the
polygonal cells having thickenings in the corners. Such a wall structure is common
amongst fossil lycopods. Indeed, in no structural feature, except for the apparent absence
of a ligule, is the strobilus significantly different from Lepidostrobus.

There is no evidence concerning the microspores that the cone may have borne.
A sample of the block containing a portion of the cone was macerated and numerous
microspores were obtained. There were, however, some twelve different kinds of spores
present, and no preponderance of any one kind. Most abundantly represented were
spores belonging to the genera Punctatisporites , Convolutispora, and Apiculatisporites.
There is evidence of association that the megaspore described below may have been
borne by the cone.

Genus triletes Bennie and Kidston ex Zerndt
Triletes subpalaeocristatus sp. nov.

Plate 34, figs. 6-7; text-fig. 3e-f

Diagnosis. Megaspore spherical; mean diameter 1,355/x (range 1 ,250-1, 460 /u, for three
spores measured). Wall about 35-45 p thick, undecorated except near the trilete ridge;
arms of trilete ridge extending the whole radius of the spore; ridge up to about 150/m
high. Lips of trilete ridge bearing a crest of simple and variously branched hair-like
appendages up to about 1 50/x long and (3-5— )6-5(— 1 3 0) ^ thick; hairs often with small
ball-shaped tips; crest of hairs about 250ft in total breadth.

Holotype. Sectioned spore in slides V51513m-p, British Museum (Natural History), Palaeontology
Department (PI. 34, figs. 6, 7).

Locality. Oxroad Bay, East Lothian, Scotland.

Further description. Altogether six megaspores have been seen in the block: four were
in the immediate vicinity of the cone, i.e. more or less within the region of sporophylls
(PI. 34, fig. 1), another was within a distance of 2 cm. and the sixth was elsewhere in the
block, remote from the cone. This last spore showed evidence of being a different kind
of spore from the other five; its wall showed a membranous outermost layer not
apparent in the others which were probably all similar to one another although their
preservation varied. This close association is impressive but is the only evidence that
this new spore was borne by Oxroadia.

EXPLANATION OF PLATE 34

Figs. 1-5. Oxroadia gracilis gen. et sp. nov. 1, Cross-section of strobilus showing the axis (top left of
centre), portions of sporophylls and dehisced sporangia and four associated megaspores (m). Slide
V51513i. X 13. 2, Strobilus axis. Slide V 5 1 5 1 3 j . x45. 3, Part of protostele of strobilus axis and two
sporophyll traces. Slide V5 1 5 1 3k. x 240. 4, Portion of wall of a sporangium. Slide V51513k. X 120.
5, Sporophyll in oblique vertical section through the horizontal part showing dehisced sporangium
with its subarchesporial pad (top) and transfusion tracheids in the ‘heel’ region. Slide V515131. X 27.

Figs. 6, 7. Triletes subpalaeocristatus sp. nov. 6, Section of the holotype passing through the summit of
the trilete ridge. Slide V5 1513m. x 27. 7, Section of the trilete ridge (near the summit) showing hairs.
Slide V51513n. X 120.

Palaeontology, Vol. 8

PLATE 34

ALVIN, Lower Carboniferous lycopod

K. L. ALVIN: A NEW FERTILE LYCOPOD FROM SCOTLAND 289

Of the five spores seen only one was complete, the others all having lost portions in
saw cuts. This whole spore (the holotype) was also fortunately the best preserved, and
is represented in a series of peel sections. The holotype was the largest of the three
measurable spores. The reconstruction in text-fig. 3e is based on this spore.

The spore shows no indication of an equatorial flange or of arcuate ridges. The wall
is uniformly thick except at the lips of the trilete ridge where it thickens to about 50ft.
The arms of the ridge are nearly 900ft long, so that in a view of the proximal face they
would be seen to extend the whole radius of the spore. The hair-like appendages are
often quite elaborately branched (PI. 34, fig. 7; text-fig. 3f). The whole crest of hairs is
probably about 150 ft high (excluding the height of the ridge itself ).

Discussion. This spore is strikingly similar to one described by Chaloner (1954) under
the name Triletes cristatus (later changed to T. palaeocristatus Chaloner (1956) to avoid
synonymy) from the Beaver Bend Limestone of the Indiana Mississippian. This American
spore differs from the new species only in its somewhat larger size (1,700-2,200 ft) and
rather thicker wall (50ft). It is, however, younger, its horizon being correlated with the
Lower Chesterian in the American succession or Upper Dinantian of the European.

It is convenient at present to follow Winslow (1959, 1962) in using the name Triletes ,
as this spore does not fall readily into any of the more narrowly defined genera into which
fossil megaspores are now usually classified. Potonie (1958) classified T. palaeocristatus
in his Turma Barbates, a group which comprises mainly a number of Mesozoic spores
with hairs or processes along the trilete ridge. The hair-like appendages of the new
spore, in being frequently branched, are similar to those in Carboniferous megaspores
of the Setosisporites kind, such as T. globosus Arnold and T. praetextus Zerndt ; this is
in contrast with the generally simple appendages in spores of the section Lagenicula ,
such as T. subpi/osus Schopf, Wilson, and Bentall and T. horridus Schopf, Wilson, and
Bentall. In both of these groups, however, the distribution of the appendages is quite
different from that in T. palaeocristatus and T. subpa/aeocristatus; other important
differences lie in the presence of apical prominences and generally also arcuate ridges.

GENERAL DISCUSSION

The view that Oxroadia gracilis represents a herbaceous lycopod rather than the
ultimate branchlets of a woody arborescent form is based mainly on the absence of
secondary activity in either the vascular system or the cortex. The absence from the
block, and indeed from the Oxroad Bay flora as far as it is at present known, of any
woody lycopod axis to which these shoots could have been attached lends a little support
to this view. The absence of any indication of rooting organs, on the other hand,
suggests that the plant was not at any rate a creeping herb. However, although Zimmer-
mann (1930) reported roots in the Upper Carboniferous Eleutherophyllum mirabile,
roots are rare amongst the remains of early Palaeozoic herbaceous lycopods.

Oxroadia differs from most fossil lycopods known from petrifactions in the absence
both of distinct leaf cushions and of a ligule. One of the few with which it shares these
characters is Levicaulis arranensis Beck (1958). However, Levicaulis has a protostele
with an almost smooth surface due to the very small and probably much more numerous
protoxylem points. It has a very similar, though usually slightly more complex, system
of reticulate fibrils between the scalariform bars. Although secondary activity in the

290

PALAEONTOLOGY, VOLUME 8

cortex appears to begin at an early stage, only one specimen was reported in which there
was secondary wood. There were fibrous strands in the cortex.

Of interest in the comparison between Oxroadia and Levicaulis is the general simi-
larity between the strobilus of Oxroadia and the fructification Lycostachys protostelicus
Pant and Walton (1961) which, on the evidence of both structure and association, was
tentatively referred to Levicaulis. The sporophylls of the two fructifications are very
similar in structure, differing only in certain details such as the absence of any tracheids
in the subarchesporial tissue and of any indication of trabeculae in Oxroadia. The
attachment of the sporangium to the sporophyll, which Pant and Walton claim was
short in the radial direction, was, I believe, probably elongated as it is in Oxroadia or,
for that matter, in Lepidostrobus. If the sporophyll illustrated in text-fig. 2f in Pant and
Walton’s paper is interpreted as curved so that the longitudinal section passes through
the attachment twice, this attachment was at least as long as the distance apart of these
‘two attachments’. However, Lycostachys is considerably larger than the strobilus of
Oxroadia ; moreover, the axis contains a solid protostele (throughout its length?) with
smaller and more numerous protoxylem points. If the strobilus of Oxroadia were known
only as an isolated fructification, it would probably be classifiable in the genus Lyco-
stachys. It should be pointed out, however, that both Lycostachys and the Oxroadia
strobilus appear to differ from Lepidostrobus only in the absence of a ligule. If the
megaspores associated with the strobilus of Oxroadia are accepted as belonging to it,
this would constitute a very important difference from any species of Lepidostrobus in
which the megaspores are known, and probably also from Lycostachys. No megaspores
were found inside the sporangia of Lycostachys , but those found in close association
had a well marked equatorial flange and processes on the distal face.

With Paurodendron Fry (1954), a genus of herbaceous lycopods represented in both
the Lower and Upper Carboniferous, Oxroadia shares a number of features. The
cortex has a similar two-layered structure, though the outer layer is probably more
fibrous, and the stele is a solid exarch protostele with the number of points varying with
the species. However, important differences from Oxroadia include the presence of
a ligule, the unequal dichotomy and the usually non-anastomosing nature of the fibrils
between the scalariform bars.

As it has been possible to determine the general external morphology of Oxroadia,
it is worth attempting a comparison with certain fossil lycopods represented by com-
pressions.

According to the classification of Krausel and Weyland (1949), Oxroadia would fall
into the Protolepidophytales and might fit in either the Protolepidodendraceae or the
Eleutherophyllaceae. It is clearly distinct from the genera Protolepidodendron and
Colpodexylon on account of the simple leaves, although in passing, it is worth noting
the general anatomical similarity to Protolepidodendron gilboense Grierson and Banks
(1963). It is much more comparable with certain species of Lepidodendropsis Lutz
which Krausel and Weyland also classify in the Protolepidodendraceae. This genus has
recently been redefined by Danze-Corsin (1958a and b) who place it in a separate sub-
family (Lepidodendropsidaceae) of their Ulodendraceae within the Lepidophytales. It
is generally typified by its leaf cushions which are arranged in ‘verticilles alternes’
(= pseudowhorls), also by the lack of a distinct leaf scar, parichnos and ligule, and the
attachment of the leaf at or near the top of the cushion. The conception of the genus

K. L. ALVIN: A NEW FERTILE LYCOPOD FROM SCOTLAND 291

is, however, still wide and it includes species which were relatively massive and
perhaps with secondary thickening, e.g. those described by Jongmans, Gothan, and
Darrah (1937), Jongmans (1939), and Jongmans and van der Heide (1955), as well as
some which were relatively slender and perhaps herbaceous, e.g. L. recurvifolia Lacey
(1962). The resemblance between the Lower Carboniferous L. recurvifolia and Oxroadia
is in fact considerable. Both are of similar size with similarly recurved leaves, though the
leaves are a little larger in L. recurvifolia. The phyllotaxis, judging from Lacey’s figures,
is probably the same. The one important difference lies in the presence in L. recurvifolia
of distinct leaf cushions. These cushions, however, do not taper at the bottom, but
represent the decurrent leaf bases running down the stem to the pair of leaves next
below. The surface of the cushion has the same epidermal structure as have the narrow
strips of stem between adjacent cushions, suggesting that they were simply ridges on the
stem. This is in contrast to a larger stemmed species, L. jonesi Lacey, where the leaf
cushion and stem surface are quite different in epidermal structure. The stem epidermis
in L. recurvifolia consists of elongated cells of similar width but rather longer than those
in Oxroadia-, in L. jonesi these cells are irregular and thick walled and quite different.
Although L. recurvifolia and Oxroadia have much in common and may have been
closely related, they are certainly not identical; apart from small differences in size and
proportions, the decurrent leaf base was evidently much more strongly developed in
L. recurvifolia.

Comparison may also be made with the genera Eleutherophyllum and Zimmermannia,
with both of which, incidentally, Lacey compares his Lepidodendropsis recurvifolia.
Grierson and Banks (1963) give the bifid leaf as being a characteristic feature of Eleu-
therophyllum, although the Namurian E. drepanophyciforme Remy and Remy (1960) is
described as having a simple lanceolate leaf. The leaf arrangement too in this species
appears to be very similar to that in Oxroadia, but a difference lies in the extension of
the leaf base up the stem above where the free part of the leaf is attached. A similar
type of leaf base, rather resembling the base of a rose thorn, is also characteristic of the
early lycopods Drepanophycus, Protolepidodendron, Colpodexylon, and Archaeosigillaria
(Banks 1960, Grierson and Banks 1963). Another important point of difference lies in
the reproductive parts of Eleutherophyllum drepanophyciforme-, there were no distinct
cones and the sporangia were attached by a non-elongated attachment to the upper
surface of the fertile leaves. E. waldenburgense (Stur) Zimmermann has a larger number
of leaves in the pseudowhorl than Oxroadia. The Upper Devonian Zimmermannia
eleutherophylloides Gothan and Zimmermann (1932) agrees with Oxroadia in size as
well as in the recurved leaves, but the leaves appear to be arranged in true whorls.

The herbaceous lycopod described by Krausel and Weyland (1937) as Lycopodites
oosensis from the Upper Devonian bears some resemblance to Oxroadia. The stem sur-
face and leaf base were probably very similar and the authors remark that the leaves,
although spirally arranged, often have the appearance of being in alternating whorls
and, where more widely separated, of being in longitudinal rows. But it was a smaller
plant than Oxroadia and had upturned falcate leaves. Its sporophylls were borne on
fertile branchlets (presumably strobili), but although microspores were found, there
was no evidence of heterospory.

The anomalous phyllotaxis of Oxroadia in comparison with most living plants has
already been referred to. From published accounts of various other fossil lycopods,

292

PALAEONTOLOGY, VOLUME 8

however, it appears highly likely that many of them had a comparable phyllotaxis.
Thus, the leaves of such genera as Lepidodendropsis, Eleutherophyllum , Protolepido-
dendron, Arcliaeosigillaria , Co/podexy/on, and Drepanophycus would appear to be
arranged in a spiral 2/x, where x is a relatively high, odd number. In all of these the
leaf arrangement can generally be interpreted in terms of spirals, pseudowhorls, or
longitudinal rows. Similarly, even many of the larger lepidodendrids such as Ompha-
lophloios, Sigillaria , Lepidopldoios spp., and Lepidodendron spp., where the leaf cushions
are arranged in more or less obvious vertical ranks, may have a fundamentally similar
phyllotaxis. In the living Lycopodium I found a 2/15 spiral in a specimen of L. clavatum ,
although elsewhere on the same plant there was no traceable spiral but rather irregular
alternating whorls. Similar alternating whorls were found in L. annotinum , L. se/ago ,
L. lucidulum, and L. obscurum. In a specimen of L. alpinum , the erect shoots had
opposite decussate leaves, but the main creeping shoot, a very irregular spiral; in no
instance was a regular spiral of the Fibonacci series observed.

Acknowledgements. I would like to express thanks to Dr. W. S. Lacey and Mr. F. A. Hibbert for kindly
examining and reporting on preparations of microspores. My thanks are also due to Miss D. Vince
for preparing peel sections and to Mr. A. Horne for photographic assistance.

REFERENCES

banks, h. p. 1960. Notes on Devonian lycopods. Senckenbergiana 41, 59-88.

barnard, p. d. w. 1959. On Eosperma oxroadense gen. et sp. nov. : a new Lower Carboniferous seed
from East Lothian. Ann. Bot., Lond., n.s., 23, 285-96.
beck, c. b. 1958. Levicaulis arranensis gen. et sp. nov., a lycopsid axis from the Lower Carboniferous
of Scotland. Trans, roy. Soc. Edinb. 63, 445-56.

chaloner, w. g. 1954. Mississippian megaspores from Michigan and adjacent states. Contr. Mas.
Paleont. Univ. Mich. 12 (3), 23-35.

1956. Triletes palaeocristatus Chaloner, a new name. Micropaleontology , 2, 298.

danze-corsin, p. 1958a. Observations sur les formes lepidodendroides du Devonien superieur et du
Dinantien (Culm). Bull. Soc. bot. Nord. France, 11, 39-54.

19586. Nouvelle classification des lepidophytes du Primaire. C.R. Acad. Sci. Paris, 247, 1226-9.

fry, w. l. 1954. A study of the Carboniferous lycopsid Paurodendron gen. nov. Amer. J. Bot. 41,
415-28.

Gordon, w. t. 1938. On Tetrastichia bupatides : a Carboniferous pteridosperm from East Lothian.
Trans, roy. Soc. Edinb. 59, 351-70.

gothan, w. and zimmermann, f. 1932. Die Oberdevonfiora von Liebichau und Bogendorf. Arb. Inst.
Palaobot., Berl. 2, 103-30.

Grierson, j. d. and banks, h. p. 1963. Lycopods of the Devonian of New York State. Palaeontographica
Amer. 4, 217-95.

jongmans, w. j. 1939. Die Kohlenbecken des Karbons und Perms in U.S.S.R. und Ost-Asien. Jversl.
geol. Bur. Heerlen, 1934-7, pp. 15-192.

gothan, w. and darrah, w. c. 1937. Beitrage zur Kenntnis der Flora der Pocono-Schischten

aus Pennsylvania und Virginia. C.R. 2nd Congr. Strut. Carb. Heerlen, 1, 423-44.

and van der heide, s. 1955. Flore et faune du Carbonifere inferieur de l’Egypte. Meded geol.

Sticht., n.s., 8, 59-75.

joy, k. w., willis, a. j., and lacey, w. s. 1956. A rapid cellulose peel technique in palaeobotany. Ann.
Bot., Lond., n.s., 20, 635-7.

krausel, r. and weyland, h. 1937. Pflanzenreste aus dem Devon. X. Zwei Pflanzenfunde im Ober-
devon der Eifel. Senckenbergiana, 19, 338-55.

1949. Pflanzenreste aus dem Devon. XIV. Gilboaphyton und die Protolepidophytales.

Senckenbergiana, 30, 129-52.

K. L. ALVIN: A NEW FERTILE LYCOPOD FROM SCOTLAND 293

lacey, w. s. 1962. Welsh Lower Carboniferous plants. I. The flora of the Lower Brown Limestone in
the Vale of Clwyd, North Wales. Palaeontographica , Bill, 125-60.

pant, d. d. and walton, j. 1961. Lycostaehys protostelicus gen. et sp. nov. and some associated mega-
spores from the Lower Carboniferous of Scotland. Ibid. 108, 1-10.

potonie, r. 1958. Synopsis der Gattungen der Sporae dispersae. II. Beih. geol. Jb. 31, 1-1 14.

remy, r. and remy, w. 1960. Eleutherophyllum drepanophyciforme n. sp. aus dem Namur A von
Niederschlesien. Senckenbergiana , 41, 89-100.

smith, d. l. 1962. Three fructifications from the Scottish Lower Carboniferous. Palaeontology, 5,
225-37.

winslow, M. R. 1959. Upper Mississippian and Pennsylvanian megaspores and other plant micro-
fossils from Illinois. Bull. III. geol. Surv. 86, 1-135.

1962. Plant spores and other microfossils from Upper Devonian and Lower Mississippian rocks

of Ohio. Prof. Pap. U.S. geol. Surv. 364.

zimmermann, f. 1930. Zur Kenntnis von Eleutherophyllum mirabile (Stbg.) Stur. Arb. Inst. Paldobot.,
Berl. 2, 83-102.

K. L. ALVIN

Department of Botany,
Imperial College,
London, S.W. 7

Manuscript received 20 June 1964

KEUPER MIOSPORES FROM WORCESTERSHIRE,

ENGLAND

by R. F. A. CLARKE

Abstract. Twenty-six species assigned to eighteen genera are recorded from the Upper and Lower Keuper of
Worcestershire. One new infraturma (Striatapiti), one new genus (Brodispora), and five new species are described.
The Zechstein, Lower and Upper Keuper spore assemblages are compared and the macrofloral changes within
this period discussed. The present assemblages are compared with previously published Triassic microfloras and
a distribution chart for twenty-two genera is given.

This paper gives an account of British Keuper miospores which are compared with
previously described assemblages, and with the British Zechstein microfloras studied by
the author. While plant remains have been known for many years to occur in British
Keuper sediments (Murchison and Strickland 1837, Brodie 1856, 1865, Arber 1909,
Wills 1910) no account of the microflora has been published. The only previous records
of Keuper spores in Britain are Wills (1910) who figures some spores incidentally in the
course of a study of the macrofossils, and Chaloner (1962).

Location and geological horizon of samples

1. Samples BH 1-BH 6 are from the Lower Keuper Sandstone (Waterstones of some
authors) exposed in a quarry in the grounds of Bromsgrove Hospital (text-fig. 1). This
is the largest of four quarries located on Rock Hill (Murchison and Strickland 1837,
Wills 1910), all of which are now disused and in the process of being filled in. The
quarry in the hospital grounds exposes the typical false-bedded, coarse-grained sand-
stone in which are conspicuous lenses of marl (‘lifts’ of Wills 1910). These are variable
in extent both laterally and vertically and probably represent the sites of old water
courses or temporary lakes. It is from such lifts that the majority of the best plant
remains have been collected. A most conspicuous lens is observed on the north-west
face of the quarry and situated about half-way up the sequence where extensive collect-
ing has created a considerable overhang. The spore-bearing samples were collected
from these lenses. It is from these quarries on Rock Hill that Arber (1909) recorded
Yaccites vogesiacus and subsequently Wills (1910) recorded the same species together
with Schizoneura paradoxa, Equisitites arenaceus, and ‘'Vo/tzia heterophylla’ . This latter,
a supposed male conifer cone, is now reassigned as Masculostrobus willsi Townrow
(Townrow 19626).

2. Sample EL 1 is from the Lower Keuper Sandstone of Elmley Lovett, 4 miles west
of Bromsgrove (text-fig. 1). This old exposure mentioned by Murchison and Strickland
(1837) consists of a track cutting and a small quarry which is very much overgrown, but
one side of the track again exposes yellow, coarse-grained, false-bedded sandstone con-
taining numerous plant remains. Sample EL 1 was collected from the marl lens at the
base of the sequence where the section is ‘shored-up’ by bricks.

3. Samples HA 1 and HA 2 are from the Lower Keuper of Hadley Mill (National
Grid Ref. 865642) approximately 2 miles west of Droitwich (text-fig. 1). The quarry

[Palaeontology, Vol. 8, Part 2, 1965, pp. 294-321, pi. 35-39.]

R. F. A. CLARKE: KEUPER MIOSPORES FROM WORCESTERSHIRE 295

is disused and somewhat overgrown but consists of massive false-bedded sandstone
which, as at Elmley Lovett, contains plant remains surrounded by ‘oxidation rings’.
The samples were taken from the more marly lenses as in the previous localities.

4. Sample L 2 is from the Arden Sandstone (Upper Keuper) of a small exposure on
the west side of the yard at Rectory Farm, Longdon (Grid Ref. 836354), 3 miles south

text-fig. 1 . Sketch-maps of the localities from which samples have been collected for the present study.

of Upton-upon-Severn. Here medium grained, white weathering sandstone with
carbonaceous ‘flecks’ is seen alternating with more marly layers (Richardson 1905).

5. Sample BR 1 was obtained from the Geological Survey Museum, collected by
Brodie. The location is given as Longdon but it is not known if it comes from Rectory
Farm.

Sample lithologies. BH 6, Reddish-brown, slightly micaceous marl; plant remains. BH 5, Reddish-
brown, non-micaceous mudstone; plant remains. BH 2 and BH 1, Reddish-brown slightly calcareous
marl. HA 2, Fine-grained, slightly calcareous, micaceous sandstone. HA 1, Reddish-brown mudstone.
EL 1, Reddish-brown marly sandstone. BR 1, Fine-grained laminated marly sandstone; plant remains.
L 2, Green and red mudstone.

C 3009

X

296

PALAEONTOLOGY, VOLUME 8

Maceration technique. (1) Twenty grams of sediment is crushed to less than one millimetre particle
size. (2) If calcareous, sample is allowed to stand in dilute 20 per cent, hydrochloric acid, and then
brought to the boil. (3) Residue is placed in 40 per cent, cold commercial hydrofluoric acid until no
further reduction in bulk takes place (time, 24 hours to 4 days). (4) The ‘silica gel’ is rejected by
adding 20 per cent, hydrochloric acid and centrifuging (5 minutes at 2,500 r.p.m.). (5) If much carbon-
aceous matter is present, sample is oxidized; 12-15 hours in concentrated nitric acid. (6) Humic acids
are neutralized by addition of sodium carbonate solution. (7) Sample is centrifuged, washed, and
slightly acidified; a small drop of phenol is added. Where necessary bromoform, diluted with one-fifth
acetone per volume, is used to concentrate the spores by floatation, after the sample has been thoroughly
dehydrated with acetone.

The slide collection. The majority of spores illustrated are from single spore preparations mounted in
glycerine jelly, unstained except where indicated on the slide, and made permanent with a candle-wax
surround. Reference to a particular spore in an assemblage slide is made by a circle on the back of the
slide. All the preparations are housed in the Geological Survey and Museum, London.

Acknowledgements. I should like to thank Professors S. E. Hollingworth and T. Barnard for the use
of laboratory facilities at University College London, and the Geological Survey and Museum for
making available Upper Keuper material from Longdon. The present paper formed part of a doctoral
thesis of the University of London under the supervision of Dr. W. G. Chaloner to whom the author
is greatly indebted for constant advice, helpful discussion, and the critical reading of the manuscript.
Finally I wish to express my gratitude for a grant from the Department of Scientific and Industrial
Research.

SYSTEMATIC SECTION

Anteturma sporites H. Potonie 1893
Turma triletes (Reinsch 1881) emend. Potonie and Kremp 1954

Subturma azonotriletes Luber 1935
Infraturma apiculati (Bennie and Kidston 1886) emend. Potonie 1956
Subinfraturma verrucati Dybova and Jachowicz 1957
Genus verrucosisporites (Ibrahim) emend. Potonie and Kremp 1954

Discussion. The more inclusive emendation of Potonie and Kremp (1954) is used here
in preference to the restrictive use advocated by Bhardwaj (1956, p. 125) which I find
difficult to apply in the present material.

Verrucosisporites morulae Klaus 1960

Plate 35, figs. 4-5

Discussion. Although the size of the distal verrucae bases varies between 2-5 p the
height remains constant (also commented upon by Klaus 1960, p. 130). In the present
specimens the sculpture of the contact faces is less coarse than that covering the re-
mainder of the spore surface. Such a feature is not apparent from the photograph of
the holotype and does not appear in the specific diagnosis. The British specimens are
here assigned to V. morulae on the basis of their size range, size of the verrucae, and
exine thickness, together with stratigraphic and geographic considerations. V. morulae
differs from V. tumulosus Leschik 1955 in the higher verrucae; other than this the two
species are very similar. V. morulae has previously been recorded from the Carnian of
the Eastern Alps (Klaus 1960).

R. F. A. CLARKE: KEUPER MIOSPORES FROM WORCESTERSHIRE

297

text-fig. 2. Verrucosisporites contactus sp. nov. Diagrammatic
reconstructions. A, Polar aspect, proximal on the right-hand side
showing the differentiated contact areas and the form of the distal
verrucae bases on the left-hand side, b, Polar section, x 1,000.

Verrucosisporites contactus sp. nov.

Plate 35, figs. 1-3; text-fig. 2

Holotype. Plate 35, fig. 3. Slide PF2392. Sample BH 6, Bromsgrove Hospital Quarry; Lower Keuper.

Diagnosis. Triradiate, verrucate miospore, 60-102^ in diameter (mean 75^; forty
measured specimens). Amb circular; verrucae low with irregular bases; differentiated
contact faces.

Description. The spore exine is 3-5 ju. thick and the proximal face bears a well-developed

298

PALAEONTOLOGY, VOLUME 8

triradiate mark which extends one-half to one-third the spore radius. The commissures
are hair-like, unaccompanied by any obvious thickening or elevations (lips), and each
arm frequently bifurcates at its extremity (in more than 90 per cent, of the specimens
observed). The contact faces are differentiated by their smaller sculptural elements and
their limits are often depicted by their darker colour (PI. 35, figs. 1, 3); no curvaturae
(arcuate ridges) are present. The distal surface and that part of the proximal face not
occupied by the contact faces possess a sculpture made up of low, flat or round-topped
verrucae 1-2 p high and 2-6 /x wide at the base; in polar view the bases of the verrucae
are rounded, irregular or polygonal. Forty to sixty verrucae may be seen in profile
around the equator.

Comparison. V. contactus sp. nov. differs from V. morulae Klaus in the smaller, more
irregular-sized verrucae, the better-defined contact areas, and the bifid terminations of
the triradiate mark. These latter two features also serve to differentiate the new species
from V. tumulosus Leschik 1955.

Subinfraturma granulati Dybova and Jachowicz 1957
Genus cyclogranisporites Potonie and Kremp 1954

Cyclogranisporites congestus Leschik 1955
Plate 35, figs. 7-9

Discussion. This species has only previously been recorded from the Middle Keuper of
Switzerland (Leschik 1955). The British specimens agree very well with the description
given by Leschik except in the larger size of the present specimens, but it is not clear
if the measurements given by Leschik represent mean values, or dimensions of the holo-
type. C. congestus is very similar to a form described as Conosniundasporites othmari
by Klaus (1960) and from the description of the latter it seems possible that C. othmari
is a badly preserved specimen of C. congestus.

Cyclogranisporites oppressus Leschik 1955
Plate 35, fig. 6

Discussion. This locally abundant Lower Keuper species has been previously recorded
from the Swiss Keuper (Leschik 1955). It is a very small form between 20-30 p (mean
26 p, measured on nine specimens), and is further differentiated from C. congestus by
the absence of lips.

EXPLANATION OF PLATE 35

All magnifications X 750.

Figs. 1-5. Verrucosisporites spp. 1-3, V. contactus sp. nov. Oblique polar views showing the dif-
ferentiated contact areas and the bifid terminations of the tetrad scar. 1, PF2393. 2, PF2395, 3,
PF2392. Holotype. 4-5, V. morulae Klaus. 4, PF2387. 5, PF2388,

Figs. 6-9. Cyclogranisporites spp. 6, C. oppressus Leschik; PF2386.

7-9. C. congestus Leschik. 7, PF2382. 8, PF2396. 9, PF2383.

Figs. 10-11. Succinctisporites radialis Leschik. 10, PF2359. 11, PF2458.

Fig. 12. Alisporites minutisaccus sp. nov. Oblique polar view of holotype; PF2424.

Localities of figs. 1-8, 10-11, Lower Keuper, Bromsgrove. Figs. 9, 12, Lower Keuper, Hadley.

Palaeontology, Vol. 8

PLATE 35

CLARKE, Upper Triassic miospores

R. F. A. CLARKE: KEUPER MIOSPORES FROM WORCESTERSHIRE

299

Subinfraturma nodati Dybova and Jachowicz 1957
Genus osmundacidites Couper 1953

Type species. O. wellmanii Couper 1953, pi. 1, fig. 5. Jurassic, New Zealand.

Discussion. The genus was originally described as triradiate although such a feature is
often not clearly displayed (e.g. Couper 1958, pi. 16, fig. 4). Forms thought to be
assignable to this genus occurring in the British Triassic lack a discernible triradiate
mark; such is the case also in the material studied by Klaus (1960). This contrasts
curiously with forms described by Balme (1963) from the Australian Trias which are
clearly triradiate.

Cyclogranisporites is differentiated from Osmundacidites by the presence of closely
spaced sculptural elements which can only be described as granae while Osmundacidites
may also, and commonly does, develop cones, papillae, and sub-baculate processes.
On general morphological grounds, however, the two genera would appear to be very
similar.

Osmundacidites alpinus Klaus 1960
Plate 37, figs. 13-14

Discussion. The original description of Klaus (1960) constitutes the sole previous record
of this species which is differentiated from the type species by being smaller and having
smaller sculptural elements.

Subinfraturma baculati Dybova and Jachowicz 1957
Genus conbaculatisporites Klaus 1960

Type species. C. mesozoicus Klaus 1960, pi. 29, fig. 15. Keuper, Eastern Alps.

Discussion. The cardinal characteristics of this genus are the triangular outline in polar
view, the length of the triradiate mark being approximately two-thirds the spore
radius and the possession of baculate processes. A miospore form found in the British
Upper Keuper satisfies two of the above requirements but has a very small triradiate
mark. Rather than create a new monotypic genus the present forms are assigned to
Conbaculatisporites Klaus (also monotypic). The present genus differs from Baculatis-
porites Thomson and Pflug 1953 only in the triangular polar contour.

Conbaculatisporites longdonensis sp. nov.

Plate 36, figs. 1-5; text-fig. 3

Holotvpe. Plate 36, fig. 1. Slide PF2475. Sample L 2, Rectory Farm, Longdon; Upper Keuper.

Diagnosis. Triangular baculate miospore. Triradiate mark small; proximal baculae
smaller than those borne distally; baculae variable in shape and size; discrete. Overall
size 49-66fi (mean 59/x, based on nine specimens).

Description. The exine is about 2g thick and the small triradiate mark, which on many
specimens is not easily seen, is unaccompanied by any form of thickening. The size of
this feature is variable but seldom exceeds one-third the spore radius. No contact areas
are delimited but the proximal face bears a sculpture of small cones and baculae which
increase in size towards the equator and which show their greatest expression distally

300

PALAEONTOLOGY, VOLUME 8

where they may be up to 9 /j.. Although the shape of these processes may vary (see text-
fig. 3) the cross-section is circular, being 2-3^ at the base. These baculae are flat-topped,
round-topped, occasionally pointed, never bifid at the tips (as in Raistrickia) and well
separated (4-5 g apart). C. longdonensis sp. nov. differs from C. mesozoicus Klaus in the
smaller triradiate mark and the larger, more variable sculptural elements.

Turma aletes Ibrahim 1933

Subturma azonaletes (Luber 1935) emend. Potonie and Kremp 1954
Infraturma striatapiti infraturma nov.

Diagnosis. This new infraturma is proposed to include all alete miospores showing
striations concentrated in the (presumed) equatorial region.

Genus brodispora gen. nov.

Type species. B. striata sp. nov.

Diagnosis. Oval striate body. Striations localized in a median zone; remainder of body
laevigate.

Discussion. These alete striate bodies are fairly common in the British Upper Keuper.
They are presumed to be miospores on the grounds that the wall is resistant to oxida-
tion, and behaves like that of the miospores with which they are associated; the possi-
bility that they might be Acritarchs is lessened by the fact that other planktonic bodies
are absent from the samples containing Brodispora gen. nov. As there is no tetrad scar
and as the grains are always found singly it is impossible to give an indisputable basis
for their orientation. The most plausible orientation, which is used here in describing
the spore, is set out in text-fig. 4.

Comparison. The genera Chomotriletes (Naum.) ex Naumova 1953 from the Upper
Devonian and Circulisporites de Jersey 1962 from the Trias are striate alete spores,
differing only in the incomplete striae of Chomotriletes as opposed to the continuous
striations of Circulisporites. They both differ from Brodispora in being of circular out-
line and developing the striations in such a way that the whole of at least one spore face
is covered rather than their being concentrated in the equatorial zone.

Brodispora striata sp. nov.

Plate 36, figs. 6-9; text-fig. 4

Holotvpe. Plate 36, fig. 9. Slide PF2478. Sample L 2. Rectory Farm, Longdon; Upper Keuper.

explanation of plate 36
Magnification X 750, unless otherwise stated.

Figs. 1-5. Conbaculatisporites loagclonensis sp. nov. 1, Holotype, PF2475, showing the distal sculpture.
2-4, Oblique polar aspects of other specimens. 5, Part of 4 showing the small triradiate mark, X
2,000. 2, PF25 16/760271. 3, PF2474. 4-5, PF2476.

Figs. 6-9. Brodispora striata gen. et sp. nov. 7, x 1,500. 6-7, PF25 16/850209. 8, PF25 16/878267.
9, Holotype, PF2478.

Figs. 10-11. Camerosporites secatus Leschik. 10, x 1,500. PF2516/780283. 11, PF2447.

All specimens from the Upper Keuper of Longdon.

Palaeontology, Vol. S

PLATE 36

CLARKE, Upper Triassic miospores

R. F. A. CLARKE: KEUPER MIOSPORES FROM WORCESTERSHIRE 301

Diagnosis. Exine thin, outline oval, striae thin and polar areas unsculptured. Size
30-40^ x 20-34 p (mean of seventeen specimens, 35x28 /a).

Description. The outline in polar view is oval with broadly rounded ends and is smooth

text-fig. 3. Conbaculatisporites longdonensis sp. nov.
Diagrammatic reconstructions, a, Proximal polar aspect
(above) and distal aspect (rectangle), b, Showing variation
of the sculptural appendages, c, Polar section. X 1,000.

text-fig. 4. Brodispora striata gen. et
sp. nov. Diagrammatic reconstruc-
tions. a, Equatorial view, b. Polar
section, c, Polar view. X 1,000.

except in polar section where the striae are seen in optical section. The striae con-
centrated in the equatorial area are narrow, only 1-2^ wide and separated by areas of
smooth exine 2-6 p wide which in equatorial view appear to widen somewhat terminally.
The number of striae is difficult to determine but appears to vary between seven and
fourteen; sometimes small transverse striae may connect the major ones.

332

PALAEONTOLOGY, VOLUME 8

Anteturma pollenites R. Potonie 1931
Turma saccites Erdtman 1947

Subturma monosaccites (Chitaley 1951) emend. Potonie and Klaus 1954
Infraturma aletesacciti Leschik 1955
Genus enzonalasporites Leschik 1955

1955 Vallasporites Leschik, pi. 6, figs. 6-8, 10.

Type species. E. vigens Leschik 1955, pi. 5, fig. 24; Keuper, Switzerland.

Discussion. Although it was considered by both Leschik (1955) and Klaus (1960) as
a saccate genus I am doubtful whether this is correct. Enzonalasporites seen as a flattened
object can reasonably be described as having an inner central area surrounded by an
outer equatorial feature, but this structure is probably not a saccus. It appears rather,
that the exoexine of the proximal face is a series of sinuous ridges which become better
developed at the equator without the wall being cavate. However, until the nature of
the wall is elucidated the genus is left as originally classified. Enzonalasporites differs
from Zonalasporites Leschik 1955 in its smaller size and the less distinct separation of
the equatorial and central areas.

Enzonalasporites vigens Leschik 1955
Plate 37, figs. 8-10

1955 Enzonalasporites obliquus Leschik, pi. 5, figs. 23, 25.

Discussion. This species is known from the Keuper of Switzerland (Leschik 1955). E .
tenuis Leschik is very similar to the present species and may be synonymous with it.
Klaus (1960) records E. tenuis from the Carnian of the Eastern Alps.

Genus patinasporites (Leschik 1955) emend. Klaus 1960
Type species. P. densus Leschik 1955, pi. 16, fig. 1 1 ; Keuper, Switzerland.

Discussion. In the emended sense Patinasporites differs from Enzonalasporites in the
greater width of the surrounding equatorial feature, the better development of the
sinuous exoexinal ridges (muri), and the generally larger size. Zonalasporites is similar
to the present genus but differs in the smaller exoexinal muri.

Patinasporites cf. densus Leschik 1955
Plate 37, figs. 11-12

Comparison. Leschik’s (1955) species is based primarily on size, and on this basis the

EXPLANATION OF PLATE 37
Magnification X 750, unless otherwise stated.

Figs. 1-2. Accinctisporites lignatus Leschik. 1, PF2480. 2, PF2497.

Fig. 3. Succinctisporites grandior Leschik, PF2526.

Fig. 4. Diploxylonoid grain. Platysaccus sp., PF2441.

Figs. 5-7. Klausipollenites devolvens comb. nov. 5, PF2429. 6, PF2428. 7, PF2426.

Figs. 8-10. Enzonalasporites vigens Leschik. 8-9, PF2434. 9, X 2,500. 10, PF2435.

Figs. 11-12. Patinasporites cf. densus Klaus. 11-12, PF2440. 12, X 2,500.

Figs. 13-14. Osmundacidites alpinus Klaus. 13-14, PF2457. 14, X 2,000.

Localities of figs. 1-3, Lower Keuper, Bromsgrove. Figs. 4-14, Upper Keuper, Longdon.

Palaeontology, Vol. <Y

PLATE 37

CLARKE, Upper Triassic miospores

R. F. A. CLARKE: KEUPER MIOSPORES FROM WORCESTERSHIRE 303

British specimens are compared with the type species. This differs from P. iustus Klaus
1960 which is larger and has a more clearly defined central area.

Genus ellipsovelatisporites Klaus 1960

Type species. E. plicatus Klaus 1960, pi. 36, fig. 65; Carnian, Eastern Alps.

Discussion. The diagnostic characteristics of this genus are the elliptical outline and the
presence of a coarsely wrinkled velum (saccus?) which completely surrounds the spore
body. The conspicuous series of sinuous ridges may be localized on the proximal face
of the spore body or be present also as part of the velum sculpture; they frequently
show a micropunctation (PI. 39, fig. 1). Ellipsovelatisporites is most similar to Vesieaspora
Schemel 1951 from which it differs in the presence of the sinuous muri.

Ellipsovelatisporites plicatus Klaus 1960
Plate 39, figs. 1-2

This species found in the British Upper Keuper has been previously recorded from
the Carnian of the Eastern Alps (Klaus 1960).

Genus succinctisporites Leschik 1955

Type species. S. grandior Leschik 1955, pi. 7, fig. 12; Keuper, Switzerland.

Discussion. Many of the species originally assigned to this genus are unacceptable
(Jansonius 1962, p. 62). The type species and a few other species may be regarded as
conforming to the original diagnosis, and the genus, in this sense, is present in the British
Trias. Leschik (1955) does not discuss the attachment of the saccus to the spore body
except to remark that it is obscure. While this is so it appears from the present material
that a thinner exinal area exists over the (presumed distal) polar region. As there is no
tetrad scar it is not possible to give a definite orientation but it is considered here that
the exine becomes cavate near to the equator and the saccus is regarded as attached
to the central body on the distal face (text-fig. 5). Succinctisporites differs from Accincti-
sporites Leschik 1955 in having a saccus showing a greater width terminally than
laterally, when viewed down the polar axis.

Succinctisporites grandior Leschik 1955
Plate 37, fig. 3; text-fig. 5

This locally abundant Lower Keuper form has been recorded from the Keuper of
Switzerland.

Succinctisporites radialis Leschik 1955
Plate 35, figs. 10-11

Comparison. This species differs from S. grandior in the radial alignment of the saccus
reticulum, and like that species is known from the Swiss Keuper. In some samples it
becomes difficult to differentiate the two species and the specific diagnoses appear to
represent extremes of a more or less continuous series.

304

PALAEONTOLOGY, VOLUME 8

Genus accinctisporites Leschik 1955

Type species. A. lignatus Leschik 1955, pi. 6, fig. 17; Keuper, Switzerland.

Discussion. The morphology of this genus is similar to that of Succinctisporites Leschik,
differing only in the possession of a circular spore body surrounded by a saccus of
uniform width.

text-fig. 5. Succinctisporites grandior Leschik. Diagrammatic
reconstructions. A, Proximal face (right), distal face (left), b, Polar
section, x 1 ,000.

Accinctisporites lignatus Leschik 1955
Plate 37, figs. 1-2

Discussion. As in Succinctisporites the saccus attachment is often obscure but the
relationship between the saccus and the spore body is presumed to be similar to that
already indicated for Succinctisporites. The terse circumscription of Leschik’s species
makes comparison difficult, but it seems that A. angustus Leschik has a wider saccus
and that A. sinuosus Leschik is separated from the type species by the presence of
endoexinal swellings. Both A. exundatus Leschik and A. nexus Leschik are much larger

R. F. A. CLARKE: KEUPER MIOSPORES FROM WORCESTERSHIRE 305

forms (90-105 p). A. lignatus has previously been recorded only from the Middle Keuper
of Switzerland.

Subturma disaccites Cookson 1947
Infraturma striatiti Pant 1954

Genus lueckisporites Potonie and Klaus 1954 emend. Klaus 1963

Lueckisporites triassicus sp. nov.

Plate 38, figs. 7-11 ; text-fig. 6

Holotype. Plate 38, fig. 7. Slide PF2408. Sample BH 6, Bromsgrove Hospital Quarry; Lower Keuper.

Diagnosis. Diploxylonoid, spore body circular, proximal cap split by longitudinal
laesure into two halves. Proximal cap smooth or micropunctate, overlapping the spore

text-fig. 6. Lueckisporites triassicus sp. nov. Diagrammatic reconstructions.
a, Polar aspect of proximal face showing the divided ‘cap’ (right), and the
distal face (left), b, Equatorial view, c, Terminal polar section. X 1,000.

body profile in polar view. Sacci larger than the spore body and more or less semi-
circular in outline. Sacci distally attached near to the pole leaving only a narrow area of
thinner exine between the attachments.

Description. This species is common in samples examined from the Lower Keuper,
above which it is absent in the area studied. The spore body dimensions are 30x28|U
(means of fifteen measured specimens) while the overall length varies from 48-77/x. The
proximal cap is divided by a laesure passing through the proximal pole but the two
halves are never widely separated and are not greatly thickened. The sacci are large,
without a lateral exoexinal connexion, and the thin saccus exine possesses a micro-
reticulate sculpture. The attachments distally are generally straight, extend to the
equator and are accompanied by crescent shaped folds (thickenings?); between these

306

PALAEONTOLOGY, VOLUME 8

folds is a leptoma but no distinct colpus is developed. The saccus offlap is greater than
the overlap on to the spore body.

Comparison. L. triassicus sp. nov. differs from the type species in the circular spore body,
the finely reticulate saccus sculpture and the presence of folds associated with the distal
saccus attachments, and from L. junior Klaus and L. tattooensis Jansonius in the circular
spore body and the greater saccus offlap.

Genus chord asporites Klaus 1960

Type species. C. singulichorda Klaus 1960, pi. 33, fig. 45; Keuper, Eastern Alps.

Discussion. This genus is characterized by the presence of an exinal strand (chorda)
developed parallel to the long axis of the grain and passing through the pole (presumed
proximal); the chorda is present on that face opposite the convergence of the sacci
which is taken to be distal. The shape of the chorda and the irregularity of its develop-
ment suggests that it is a fold produced by compression. The position of the fold is
probably a result of the compression of the cap which is thickest in a line passing
through the proximal pole; such a line of thickening may be accompanied on either side
by narrow lines of thinner exine. Chordasporites differs from Lueckisporites s. str. in
the presence of a chorda and a generally smoother proximal cap.

Chordasporites singulichorda Klaus 1960
Plate 38, figs. 1-3

Comment. The genus contains only two very similar species. The type species is known
only from the present record and that of Klaus (1960). C. australensis is an Australian
form which differs from C. singulichorda in having a thinner spore body, absence of
a proximal cap and thinner exinal areas adjacent to the chorda (de Jersey 1962).

Genus ovalipollis Krutzsch 1955 emend. Klaus 1960
1955 Unatextisporites Leschik.

Type species. O. ovalis Krutzsch 1955, pi. 1, fig. 2; Lias, Germany.

Discussion. The interesting point about this genus is the interpretation of the position
and function of a furrow which is present on one of the spore surfaces and disposed
parallel to the long axis of the grain. Krutzsch in the original description makes no
reference to proximal and distal faces but Klaus (1960) states that the furrow is on the
side opposite the sacci convergence. If Ovalipollis has a saccus arrangement comparable

EXPLANATION OF PLATE 38
Magnification X 750, unless otherwise stated.

Figs. 1-3. Chordasporites singulichorda Klaus. 1, PF2402. 2, PF2401. 3, PF2399.

Figs. 4-6. Alisporites toralis comb. nov. 4, PF2418. 5, PF2416. 6, PF2420.

Figs. 7-11. Lueckisporites triassicus sp. nov. 7, Polar aspect of holotype, PF2408. 8-9, PF2410. 9,
X 2,000. 10, PF2409. 11, PF2411.

Figs. 12-13. Camerosporites secatus Leschik. 12, PF2446. 13, PF2448.

Localities of figs. 1-3, 7-11, Lower Keuper, Bromsgrove. Figs. 4-6, Lower Keuper, Hadley, Figs. 12-
13, Upper Keuper, Longdon.

Palaeontology, Vol. 8

PLATE 38

CLARKE, Upper Triassic miospores

R. F. A. CLARKE: KEUPER MIOSPORES FROM WORCESTERSHIRE 307

to that in other bisaccate grains then the furrow will be on the proximal face (cf. Jan-
sonius 1962). The function of this furrow remains problematical. In bisaccate grains of
extant plant groups germination is distal from a point between the sacci attachments.
Many Palaeozoic bisaccate grains apart from having apparent distal germination also
show a tetrad scar on the opposite face. Perhaps the ‘furrow’ of Ovalipollis is a monolete
tetrad scar, corresponding with a tetragonal tetrad arrangement. On this basis then
Ovalipollis may be considered comparable in its anatomy and orientation to other
bisaccate grains in the Permo-Triassic. The ‘furrow’ is, however, often open at the
ends, giving it an elongated ‘hour-glass’ shape. Either the genus is a monosulcate
pollen with a distal germinal furrow, or it has a proximal monolete aperture. Until
Ovalipollis is found in a tetrad the question of proximal and distal faces will remain
unsolved, but for the present purpose it is regarded as a bisaccate pollen with a long
monolete mark on the proximal face and distally inclined sacci.

Ovalipollis breviformis Krutzsch 1955
Plate 39, figs. 11-12

1955 Unatextisporites mohrensis Leschik, pi. 8, figs. 7-8.

1960 Ovalipollis grebeae Klaus, pi. 35, figs. 52, 55.

Discussion. This species is known from the Rhaeto-Lias of Germany (Krutzsch 1955),
the Middle Keuper of Switzerland (Leschik 1955), the Keuper of the Eastern Alps
(Klaus 1960), and the Lower Triassic of Western Canada (Jansonius 1962). It differs
from O. ovalis Krutzsch in the smaller size and more oval outline, and is less fusiform
than O. longiformis Krutzsch.

Ovalipollis ovalis Krutzsch 1955
Plate 39, figs. 9-10

1955 Unatextisporites mohri Leschik, pi. 8, fig. 9.

1960 Ovalipollis lunzensis Klaus, pi. 34, figs. 46-49; pi. 37, fig. 67.

Discussion. This species, usually associated with O. breviformis, has been previously
recorded from the Rhaeto-Lias of Germany (Krutzsch 1955), the Middle Keuper of
Switzerland (Leschik 1955), the Keuper of Poland (Pautsch 1958), the Keuper of the
Eastern Alps (Klaus 1960), and the Lower Triassic of Western Canada (Jansonius
1962).

Infraturma disaccimonoletes Klaus 1963
Genus labiisporites Leschik 1956 emend. Klaus 1963

Type species. L. granulatus Leschik 1956, pi. 22, fig. 11 ; Zechstein, Neuhof.

The type species, based on Permian material, persists in small amounts from the
British Zechstein through the Keuper (PI. 39, fig. 5).

Infraturma pinosacciti Erdtman 1956 emend. Potonie 1958
Genus alisporites Daugherty 1941

1955 Scopi/lisporites Leschik.

Discussion. The most pertinent comparison of this genus is with Pityosporites Seward
1914 emend. Manum 1960. Despite several emendations to both genera there still

308

PALAEONTOLOGY, VOLUME 8

exists no satisfactory basis for their separation. To arrive at a solution to this problem
would seem almost impossible as the diagnosis of Pityosporites is based on a specimen
in lateral (equatorial) view while the type of Alisporites is orientated in the equatorial
plane (polar view). It is therefore not known what Pityosporites looks like in polar view.
To base a distinction on the degree of saccus inclination distally is difficult unless the
attachments both proximally and distally are distinct and the degree of saccus inclina-
tion seen in lateral view is often a function of the amount of collapse which has taken
place between the saccus attachments. Generic separation based on the presence or
absence of a colpus (sulcus) is subjective in that the presence may be a function of
maturity and its retention a matter of preservation. This seeming lack of a good basis
for their separation confuses the stratigraphic use of the two genera and it may be better
to combine them. If this be the case Pityosporites has priority as a name. However,
Alisporites is based upon a specimen in polar view which is well preserved while Pityo-
sporites is in lateral view and badly preserved. Because of these latter two shortcomings
and the scepticism of some authors as to its type material being pollen at all (Walton
1925, Edwards 1928) Pityosporites may be considered a ‘confused genus’ and abandoned.

Alisporites toralis (Leschik) comb. nov.

Plate 38, figs. 4—6; text-fig. 7
1955 Scopulisporites toralis Leschik.

Discussion. In the majority of specimens of this species the sacci are connected laterally
by a narrow exinal strip which is never more than a few microns wide in polar view.
The sacci attachments both proximally and distally are ill defined and a distinct
colpus is seldom developed. This species is known from the Swiss Keuper and is a
common form in the British Lower Keuper. The species differs from the type species
in the shape of the spore body and the less well-defined saccus attachments, and from
A. microreticulatus Reinhardt 1964 in the shape of the spore body and the less- well-
defined ‘Keimfurche’.

EXPLANATION OF PLATE 39
Magnification X 750, unless otherwise stated.

Figs. 1-2. Ellipsovelatisporites plicatus Klaus. 1, X 2,500, showing the proximal spore body sculpture.
1-2, PF2524.

Figs. 3-4. Alisporites cf. parvus de Jersey. 3, PF2466. 4, PF2516/780161.

Fig. 5. Labiisporites granulatus Leschik. 5, PF2516/885272.

Figs. 6-8. Alisporites circulicorpus sp. nov. 6, PF2461. 7, PF2465. 8, Slightly oblique polar view of
holotype, PF2460.

Figs. 9-10. Ovalipollis ovalis Krutzsch. 9, PF25 15/735 172. 10, PF2442.

Figs. 11-12. O. breviformis. 11, PF25 15/665300. 12, PF2515/584210.

Fig. 13. Trisaccate grain, PF2462.

Figs. 14-17. Cycadopites spp. 14-15, C. acerrimus comb. nov. Distal polar view. 14, PF25 16/740332.
15, PF2453. 16-17. C. subgranulosus comb. nov. 16-17, PF2449. 17, x 2,000, showing the granulose
sculpture.

Location of figs. 1, 2, 4, 5, 9-12, 14, Upper Keuper, Longdon. Figs. 3, 6, 7, 13, 15, Lower Keuper,
Bromsgrove. Figs. 8, 16, 17, Lower Keuper, Hadley.

Palaeontology , Vol. 8

PLATE 39

CLARKE, Upper Triassic miospores

R. F. A. CLARKE: KEUPER MIOSPORES FROM WORCESTERSHIRE

309

Alisporites circulicorpus sp. nov.

Plate 39, figs. 6-8; text-fig. 8

Holotype. Plate 39, fig. 8. Slide PF2460. Sample HA 1, Hadley; Lower Keuper.

Diagnosis. Haploxylonoid. Spore body circular or oval. No tetrad scar visible. Sacci
small. Offlap small and crescent shaped.

text-fig. 7. Alisporites tora/is comb. nov. Diagrammatic
reconstructions, a, Polar view, b, Polar section. X 1,000.

Description. Spore-body dark coloured and large (51 x50fi; means of fifteen measured
specimens) compared with the sacci which in polar view show a small offlap terminally.
Laterally the sacci may be connected by a thin exoexinal strand, but this is not common.
The attachments of the sacci both proximally and distally remain obscure and thus it
is not clear at what point the exine becomes cavate. The saccus sculpture is reticulate
and a colpus has not been observed between the saccus attachments distally. The over-
all length is 46-70/x (mean 61/x; measured on fifteen specimens).

Comparison. A. circulicorpus sp. nov. differs from A. opii Daugherty and A. tora/is
comb. nov. in its spore body/saccus ratio. The general sculpture of A. circulicorpus
sp. nov. is similar to that of A. toralis comb. nov. and in some respects to Succincti-
sporites grandior Leschik.

310

PALAEONTOLOGY, VOLUME 8

Alisporites cf. parvus de Jersey 1962
Plate 39, figs. 3-4

Discussion. The distinctive features of this species are the small size (overall length
42-53/x, measured on eight specimens) and the comparatively small sacci. Specimens

text-fig. 8. Alisporites circulicorpus sp. nov. Diagrammatic recon-
structions. a, Proximal polar aspect, b, Polar section. X 1,000.

present in the British Keuper are thus compared with those described as A. parvus by
de Jersey (1962) from the Australian Triassic Ipswich Coalfield.

Alisporites minutisaccus sp. nov.

Plate 35, fig. 12; text-fig. 9

Hoiotype. Plate 35, fig. 12. Slide PF 2424. Sample HA 1, Hadley; Lower Keuper.

R. F. A. CLARKE: KEUPER MIOSPORES FROM WORCESTERSHIRE

311

Diagnosis. Spore small, body circular; sacci smaller than spore body. Saccus width
smaller than spore body width. Saccus attach-
ments terminal; no distinct leptoma.

Description. Spore-body circular, only occasion-
ally oval 31x32/x (means of ten measured
specimens). The spore body proximal face bears
a microgranular sculpture and is slightly more
densely coloured than the sacci. These latter are
small, semicircular in polar outline and their
width is considerably less than that of the spore
body (i.e. diploxylonoid in the sense of Hart
1960, p. 5). The sacci are discrete and attached
terminally although in many instances the
attachments are not clearly defined. The saccus
sculpture is punctate to microreticulate.

Comparison. A. minutisaccus sp. nov. differs
from the type species in the much smaller size,
and the saccus/spore body ratio. A. toralis
comb. nov. is a much larger and more coarsely
reticulate form and A. cf. parvus has tapering text-fig. 9. Alisporites minutisaccus sp. nov.
sacci which are attached distally near to the A> Oblique polar viewjbased on^specimen).
distal pole.

b, Polar profile. X 1,000.

Genus klausipollenites Jansonius 1962

Type species. K. (al. Pityosporites ) schaubergeri Potonie and Klaus 1954, pi. 10, fig. 7; Zechstein,
Alpine area.

Klausipollenites devolvens (Leschik) comb. nov.

Plate 37, figs. 5-7

1955 Pityosporites devolvens Leschik.

Comparison. This species is recorded from the Middle Keuper of Switzerland. It is
similar to K. schaubergeri (which is observed in small numbers in the present material,
see text-fig. 11), but differs in the more prolate form of the spore body, and the less
tapering sacci.

Infraturma podocarpoiditi Potonie, Thomson, and Pflug 1950
Genus platysaccus (Naumova) ex Potonie and Klaus 1954

Geno/ectotype. P. papilionis Potonie and Klaus 1954, pi. 10, fig. 12; Zechstein, Alpine area.

Platysaccus sp.

Plate 37, fig. 4

Description. Non-striate diploxylonoid grains form a rare constituent of the British
Upper Keuper assemblages. Such grains have been placed in Platysaccus. On three
measured specimens the overall length is 57-68 /x and the spore body 33x27 p. The

C 3009

Y

312

PALAEONTOLOGY, VOLUME 8

sacci are large compared with the spore body but are not so markedly diploxylonoid as
those observed in the Zechstein.

Turma plicates (Naumova 1937, 1939) emend. Potonie 1960
Subturma monocolpates Iversen and Troels-Smith 1950
Infraturma intortes (Naumova 1937) emend. Potonie 1958
Genus cycadopites (Wodehouse 1933) ex Wilson and Webster 1946

1938 Azonaletes Luber, p. 154, figs. 10-11.

1939 Subsacculifer Luber, pi. A, fig. 1.

1953 Ginkgocycadophytus Samoilovich.

1954 Entylissa (Naumova) ex Potonie and Kremp.

1955 Cycadoletes Luber, figs. 170-1.

1955 Gynkaletes Luber, figs. 168-9.

1960 Lagenella (Malawkina 1949) Klaus (pars).

Type species. C. foUicularis Wilson and Webster 1946, pi. 1, fig. 7; Tertiary, Montana.

Discussion. Malawkina (1949) erected the genus Lagenella without designating a type.
Klaus (1960) validated the genus and selected L. cincta Malawkina as the ‘genotype’.
This species is a non-striate form and cannot be separated from Cycadopites as used
here. However, Klaus includes within Lagenella monosulcate striate miospores pre-
viously assigned to Decussatisporites Leschik 1955. This latter genus is validly established
and I prefer to rate the presence or absence of striations as a generic character. Decus-
satisporites is thus used in the original sense of Leschik (1955) and the non-striate forms
of Lagenella, sensu Klaus, are placed in synonymy with Cycadopites.

Cycadopites subgranulosus (Couper) comb. nov.

Plate 39, figs. 16-17

1958 Monosidcites subgrcimdosus Couper.

This species, based on British Liassic material (Couper 1958), is found in small
numbers in Upper and Lower Keuper samples. It differs from those species described
by Jansonius (1962) in the nature of the exinal sculpture.

Cycadopites acerrimus (Leschik) comb. nov.

Plate 39, figs. 14-15

1955 Moiiocolpopollenites acerrimus Leschik.

Discussion. This common British Keuper species is also known from the Swiss Keuper
and similar forms are present in the Canadian Trias ( Cycadopites sp. R, Jansonius 1962).
C. acerrimus differs from C. subgranulosus comb. nov. in having a smooth exine, from
C. dijkstrae Jansonius 1962 in the absence of lips and from C. hartii Jansonius in the
lack of ‘drawn out cones’ at the ends of the long axis.

Genus camerosporites Leschik emend.

Type species. C. secatus Leschik 1955, pi. 5, fig. 11 ; Keuper, Switzerland.

Emended diagnosis. Amb elongate-oval, bilaterally symmetrical. On one face is a thin

R. F. A. CLARKE: KEUPER MIOSPORES FROM WORCESTERSHIRE

313

elongated exinal area (? sulcus) surrounded equatorially by large verrucose sculptural
elements which may extend on to the opposite face.

Discussion. The genus is rather summarily described by Leschik: . . Nur eine Sym-

metricebene vorhanden. Kammern verschieden gross.’ The sulcus in Camerosporites is
often irregular and not clearly defined (unlike the sulcus in Cycadopites). Nevertheless,
there is an elongated thin area which can reasonably be in- ,:W.v„5v_

terpreted as a sulcus and for this reason the genus is emended ,

and placed in the Turma (Abteilung) Monocolpates. 'fc A ’• :-b

Comparison. Camerosporites here emended differs from yCyi .% \

Thymospora Wilson and Venkatachala 1963 (syn. Verruco- , \[C.. > y.'.yC/d- /V'

sosporites (Knox) ex Potonie and Kremp 1954) in the more C ' ' fj

fusiform outline and being monosulcate as opposed to f-
monolete. Hoegisporis Cookson 1961 from the Australian

Cretaceous is similar to Camerosporites in the possession of rV; ; UyfiVrqk.'' A: .'■/
large verrucate processes but differs in the circular outline w-:;

and the smaller number of verrucae which, in Cookson's , Cf

spore, are restricted to the equator. The distinctive sculp- ''••YU CCkXU

ture of Camerosporites differentiates it from all other A
monosulcate grains.

Camerosporites secatus Leschik 1955
Plate 36, figs. 10-11; Plate 38, figs. 12, 13; text-fig. 10

Description. The outline is fusiform 44x31/x (means of
nineteen measured specimens) with broadly rounded ends.

On one surface (presumed distal) the exine is thin forming
an elongated sulcus of which the boundaries are not pre-
cisely defined. This sulcus is fairly wide but towards the
equator small verrucae appear which rapidly become large
at the equator where they appear as flat or rounded-topped text-fig. 10. Camerosporites
protuberances 4-7 /x high and up to 1 3 /x wide at the base. secatl,s Leschik. Diagrammatic
The large sculptural elements may be confined to this reconstructlons- A> Distal polar
equatorial zone or be present on the other (proximal) lace (sulcus?), b. Polar section.

(text-fig. 106).

x 1,000.

COMPARISON OF ASSEMBLAGES WITH THOSE OF THE
BRITISH ZECHSTEIN (UPPER PERMIAN)

From the range chart (text-fig. 13) it will be seen that very few miospores persist
from the older assemblages into the Triassic. Changes in the groups above generic rank
can be followed. Text-fig. 11 gives the constituent percentages of supra-generic groups
for the Upper Keuper, Lower Keuper, and the Upper Permian. The Permian information
has been compiled by averaging all the frequencies observed by the author in samples
from Hilton, Westmorland, and Kimberley, Nottinghamshire. (A fuller account of
British Upper Permian miospores is given in the next paper in this volume.) It can be

314 PALAEONTOLOGY, VOLUME 8

seen that for the bisaccate Striatiti a marked decrease occurs from the Zechstein to the
Lower Keuper and that this trend is maintained in the Upper Keuper. Of some six
bisaccate striate genera present in the Upper Permian, two are present in the Lower

A.

SAMPLE

AGE

NUMBER

«r a

L 2

CL O-

a =>
3

B R 1

BH 6

BH 5

a
a uj

BH 2

i a

BH 1

o =>

HA 2

*

HA 1

EL 1

Q : — UJ

u UJ u- X X

IIIU

til CL
U UJ
oe h-

70

72

72

76

70

75

73
75
75

BISACCATE STRIATE
BISACCATE NON-STRIATE
MONOSACCATE
MONO SULCATE
TRI RADIATE

text-fig. 11. The percentages (based upon counts of 200 grains) of the various species (A) and selected
supra-generic groups (B) in some British Permo-Trias deposits.

Keuper (one of which is restricted to this horizon), while only one genus ( Ovalipollis )
is present in the Upper Keuper. The reverse of this is seen in the sharp increase in
triradiate miospores (i.e. non-cingulate and non-zonate types) in the Lower Keuper and
which is also maintained in the Upper Keuper. The sudden increase in the number
and species of bisaccate non-striate forms in the Upper Permian (such types being
almost absent in the Carboniferous) is sustained in the Trias although many species are

R. F. A. CLARKE: KEUPER MIOSPORES FROM WORCESTERSHIRE

315

text-fig. 12. Distribution Chart for twenty-two genera present in the Trias. 1, Present Record. 2,
Klaus (1959, 1960). 3, Pautsch (1958), 4, Leschik (1955). 5, Taugourdeau-Lantz and de Jekhowsky
(1959), Krutzsch (1955). 6, Balme (1963). 7, de Jersey (1949, 1962). 8, Jansonius (1962). 9, Daugherty
(1941). 10, Scott (1960). 11, Jux (1961). For Camerozonosporites read Camerosporites , and omit
record 2; for Striatites read Protohaploxypinus.

316

PALAEONTOLOGY, VOLUME 8

CALAMOSPORA NATHORSTII
CYCADOPITES ACERRIMUS

LABIISPORITES GRANULATUS
PLATYSACCUS

KLAUSIPOLL. SCHAU8ERGERI
POTONIEI SPORITES
NUSKOISPORITES
SIMPLICESPORITES

VESTIGISPORITES
LUECKISPORITES VIRKKIAE
TAENIAESPOR1TES
STR1ATITES
STRIATOAB1ETITES
STRIATOPODOCARPITES
CRUSTAE SPORITES
VITTATINA

ALISPORITES NUTHALLENSIS

falcisporites

CYCADOPITES RARUS
C. SUBGRANULOSUS
VERRUCOSISPORITES MORULAE
CYCLOGRANISP. CONGESTUS
C. OPPRESSUS
SUCCINCTI SPORI TES
ACCINCTISPORITES
CHORDASPORITES
LUECKISP. TRIASSICUS
ALISPORITES MINUTISACCUS
A. TORALIS
A. CIRCUL1CORPUS
VERRUCOSISP. CONTACTUS
OSMUNDACIDITES ALPINUS
CON B ACULAT l SP. LONDONENSIS
ENZONALASPORITES VIGENS
PATINA SPORITES cf. DENSUS
ELL1PSOVELATISP. PLICATUS
OVALIPOLLIS

KLAUSIPOLL. DEVOLVENS i

ALISPORITES cf. PARVUS
BRODISPORA STRIATA

2 —

2

_ 2

text-fig. 13. Chart of the ranges
For Striatites

of some British Permo-Triassic
read Protohaploxypinus.

nnospores.

R. F. A. CLARKE: KEUPER MIOSPORES FROM WORCESTERSHIRE 317

different, while monosaccate forms are also more generally abundant than in the Upper
Permian. Monosulcate grains, which first appear in Britain in the Permian, also show
greater representation both specifically and numerically in the Keuper. Monolete forms
(e.g. Laevigatosporites), often present in considerable numbers in the Carboniferous,
have not been observed in the Permo-Trias of the area studied.

Owing to a lack of knowledge of the natural botanical affinity of many of these types,
the changes outlined above, expressed in terms of changes in the macroflora, must be
conjectural. It is not clear, for instance, which group or groups of plants is represented
by the decline and virtual extinction in Trias times of the bisaccate Striatiti. Bisaccate
striate pollen ( Lueckisporites s.str.) have been found in the fructification of the conifer
Ullmannia frumentaria Goeppert, while others ( Protohaploxypinus s.str.) are closely
associated with G/ossopteris, a presumed Pteridosperm (Potonie and Schweitzer 1960,
Pant and Nautiyal 1960). Further Coniferous groups (as well as possible Cordaites) are
represented by monosaccate grains, while diploxylonoid pollen of the Platysaccus habit
may have Podocarp affinities. Alisporites- like pollen suggests the presence of Pterido-
sperms (or possible Conifers) in the assemblages. The absence of cingulate and zonate
triradiate types ( Densosporites , Cristatisporites ) in the Permo-Trias attests to the decline
of some Pteridophyte group (probably the Lycopsida), but the reappearance of triradiate
non-zonate spores in the Trias may represent the re-emergence of other Pteridophyte
groups (most probably the Filicales). The steady increase in the type and number of
monosulcate pollen is taken to be indicative of the rise of the Cycadophytes, although
this type of pollen may be also Pteridospermous (Townrow 1960).

The decrease in the triradiate non-zonate miospores in the Permian and the great
increase in saccate forms is seen as a response to a climatic change towards aridity, to
which the seed habit of the Gymnosperms is better adapted than the ‘water dependent’
life cycles of most Pteridophyte groups. The presence or return of such forms in the
late Trias is probably correlated with a return to more humid climatic conditions.

COMPARISON WITH PREVIOUSLY DESCRIBED TRIASSIC

ASSEMBLAGES

European Trias

The most comprehensive works on Keuper microfloras are those of Leschik (1955)
and Klaus (1960). The majority of the species found in the present study can be identified
with reference to these two works and many forms are common to all three areas. The
most important of these are the presence of Verrucosisporites, Camerosporites, Enzonala-
sporites, Alisporites , and Ovalipollis. This latter genus appears for the first time generally
in the Keuper (but see Taugourdeau-Lantz 1962), and is present in all European Keuper
assemblages examined (see also Pautsch 1958, Taugourdeau-Lantz and Jekhowsky
1959, Reinhardt 1964). However, several forms recorded by Leschik (1955) and Klaus
(1960) are not apparently represented in the present samples, viz. Zebrasporites,
Kraeuselisporites, Decussatisporites , and Aratrisporites.

Australian Trias

Knowledge of Triassic microfloras in Australia is due mainly to de Jersey (1962) on
the Ipswich Coalfield (pre-Middle Trias) and Balme (1963) on the Lower Triassic

318

PALAEONTOLOGY, VOLUME 8

Kockatea Shale of Western Australia. Papers by de Jersey (1949) and Taylor (1953) are
less useful because of the numerical system of nomenclature employed and the absence
of photographs. The Kockatea assemblage has little in common with the British
assemblages. In some respects (the presence of Striatites, Taeniae sporites, Crustae-
sporites) a Permian flavour is present although Kraeuselisporites , Vitreisporites, Osmunda-
cidites, and Lycopodiacidites emphasize its Mesozoic character. Ova l ipo l /is, which appears
to be a Northern Hemisphere genus, is absent, as are Cycadopites and Alisporites- like
bisaccate grains. This latter genus, however, is present in some quantity in the Ipswich
Coalfield, associated with Cycadopites (= Ginkgocycadophytus of de Jersey), Calamo-
spora , and Osmundacidites. The most interesting record, however, is that of undisputed
Chordasporites previously known only from the Alpine Keuper (Klaus 1960) and now
recovered from the British Lower Keuper. The presence of this genus and the absence
of Ovalipollis would suggest a Lower Keuper age for the Ipswich deposits.

North America

Jansonius (1962) describes a rich microflora from the Lower Triassic Toad/Grayling
Formation of Canada. This differs from the British Keuper assemblages essentially in
the presence of a variety of bisaccate striate grains which perhaps, but not necessarily
(Leschik 1955), suggests its Early Triassic age. The appearance of Ovalipollis, however,
tends to discredit such an assumption and this Canadian assemblage is perhaps most
similar to that described by Balme (1963).

REFERENCES

arber, E. a. n. 1909. On the affinities of the Triassic plant Yuccites vogesiacus Schimper and Mougeot
Geol. Mag. 6, 11-14.

balme, b. e. 1963. Plant microfossils from the Lower Triassic of Western Australia. Palaeontology, 6,
12-41.

bhardwaj, d. c. 1956. The spora genera from the Upper Carboniferous coal of the Saar and their
value in stratigraphic studies. The Palaeobotanist, 4, 119-49.

and singh, h. p. 1957. Asterotheca meriani (Brongn.) Stur and its spores from the Upper Triassic

of Lunz (Austria). Ibid. 5, 51-55.

bolkhovitina, n. a. 1959. Spore-pollen complexes of Mesozoic deposits and their stratigraphic sig-
nificance. Akad. nauk. SSSR Trudy Geol. inst. 24. [In Russian.]
brodie, p. b. 1856. On the Upper Keuper Sandstone (included in the New Red Marl) of Warwickshire.
Quart. J. geol. Soc. Loud. 12, 374-6.

— — 1865. On the fossiliferous beds in the New Red Sandstone (Upper and Lower Keuper) in War-
wickshire. Geol. Mag. 2, 567.

chaloner, w. g. 1962. British Rhaetic and Triassic spores. (Resume.) Pollen et Spores, 4, 339.
cookson, i. c. 1961. Hoegisporis, a new Australian Cretaceous form genus. Palaeontology, 3, 485-6.
couper, r. a. 1953. Upper Mesozoic and Cainozoic spores and pollen grains from New Zealand. N.Z.
Geol. Surv. Pal. Bull. 22, 77p.

1958. British Mesozoic microspores and pollen grains. Palaeontographica , B 103, 75-179.

daugherty, l. h. 1941. The Upper Triassic Flora of Arizona. Carnegie Inst. Wash. Publ., 526, 1-108.
edwards, w. n. 1928. The occurrence of Glossopteris in the Beacon Sandstone of Ferrar Glacier, South
Victoria Land. Geol. Mag. 65, 323.

florin, r. 1936. On the structure of pollen grains in the Cordaitales. Svensk. Bot. Tidskr. 30, 624-51.
hart, g. f. 1960. Microfloral investigation of the Lower Coal Measures (K2): Ketewaka-Mchuchuma
Coalfield, Tanganyika. Geol. Surv. Tanganyika. Bull. 30, 1-18.
hennelly, j. p. f. 1959. Spores and pollen from a Permian-Triassic transition, New South Wales.
Proc. Linn. Soc. N.S. W. 83, 363-9.

R. F. A. CLARKE: KEUPER MIOSPORES FROM WORCESTERSHIRE

319

jacob, k. 1950. Microflora of the Leigh Creek Coalfield, Australia. Jour. Ind. Bot. Soc. 29, 19.
jansonius, j. 1962. Palynology of Permian and Triassic sediments, Peace River area, Western Canada.
Palaeontographica, B 1 10, 35-98.

jekhowsky, b. de, sah, s. c. d. and letullier, a. 1960. Reconnaissance palynologique du Permien,
Trias et Jurassique des sondages effectues par la Societe des petroles de Madagascar dans le bassin
de Morandava. C. r. sommaire des seances de la Soc. geol. frangais. 7, 166.
jersey, n. j. de. 1949. Principal microspore types in the Ipswich coals. Pap. Dept. Geol. Univ. Queens-
land, 9, 1-8.

1962. Triassic spores and pollen grains from the Ipswich coalfield. Geol. Surv. Queensland

Pub!. 307.

jux, u. 1961. The palynologic age of diapiric and bedded Salt in the Gulf Coast Province. Dept. Cons.
Louisiana Geol. Surv. 38, 1-46.

kara-mourza, e. n. 1962. Triassic spores and pollen complexes and their significance for the strati-
graphy and correlation of marine and continental and volcanic deposits in the Asiatic part of the
Soviet arctic region. (Resume.) Pollen et Spores, 4, 356.
klaus, w. 1959. Sporenfunde in der karnischen Stufe der alpinen Trias. Verb. Geol. Bundesanst. 2,
160-3.

■ — — 1960. Sporen der karnischen Stufe der ostalpinen Trias. Jb. Geol. B. A. 5, 107-83.
krutzsch, w. 1955. Uber einige liassische ‘angiospermide’ Sporomorphen. Z. Geologie, 4, 67-76.
lanjouw, J. et al. 1961. International Code of Botanical Nomenclature. Utrecht.
leschik, g. 1955. Die Keuperflora von Neuewelt bei Basel 2, Die Iso- und Milcrosporen. Schweizer.
Paldont. 72, 5-70.

1958. Sporenstratigraphie im Permien und in der Trias. Z. Deutsch. Geol. Gesellsch. 110,

13-14.

luber, a. a. 1938. Spores and pollen from the Permian of the U.S.S.R. Probl. Soviet. Geol. 152-61.
(English summary.)

1939. The correlation by means of spores of coal bearing Upper Palaeozoic deposits of the

Kuznetsk and Minussinsk basins. Bull. Acad. Sci. U.S.S.R. 6, 88-104. (English summary.)

1955. Atlas of spores and pollen from the Palaeozoic deposits of Kazakhstan. Ak. Nauk. Kazakh.

SSSR Alma-Ata. 1-126. [In Russian.]

lundblad, b. 1959. On Riccisporites tuber culatus and its occurrence in certain strata of the ‘Hollviken
1 1 ' boring in S.W. Scania. Grana Palynologica, 2, 1-10.
malawkina, v. s. 1949. Determination of spores and pollen of the Jurassic and Lower Cretaceous.
Trudy VNIGRI (All Union Sci. Petrol. Res. Inst.), 33, 1-137. [In Russian.]

1953. Spores and pollen from the Upper Trias and Lower and Middle Jurassic from the east and

west Pre-Urals. Ibid. 75, 93-147. [In Russian.]

manum, s. I960. On the genus Pityosporites Seward 1914, with a new description of Pityosporites
antarcticus Seward. Nytt. Mag. Bot. 8, 11-15.

markova, l. g. 1962. Spore and pollen complexes of Mesozoic deposits of the west-Siberian Lowland.
(Resume.) Pollen et Spores, 4, 362.

matley, c. a. 1912. The Keuper (or Arden) Sandstone group and associated rocks of Warwickshire.
Quart. J. geol. Soc. Loud. 68, 252-80.

Murchison, r. i. and Strickland, h. e. 1837. On the Upper Formations of the New Red Sand-
stone system in Gloucestershire, Worcestershire and Warwickshire, etc. Trans. Geol. Soc. Loud. 5,
331-48.

naumova, s. n. 1937. Spores and pollen of the coals of the U.S.S.R. Rept. 17th internat. geol. Congr.,
Moscow, 1, 353-64.

1953. Sporo-pollen complexes of the Upper Devonian of the Russian Platform and their strati-

graphical value. Tr. Inst. geol. nauk. Akad. SSSR 143, 1-204. [In Russian.]
nilsson, t. 1958. Uber das Vorkommen eines mesozoischen Sapropelgesteins in Schonen. Lunds
Universitets arsskrift, 2, 1-111.

pant, d. d. 1949. Triassic plant remains from the Salt Range in the Punjab. Nature, 163, 914.

and nautiyal, d. d. 1960. Some seeds and sporangia of the G/ossopteris flora from the Raniganj

Coalfield, India. Palaeontographica, B 107, 41-64.
pautsch, m. 1958. Keuper sporomorphs from Poland. Micropalaeontology, 4, 321-5.

320 PALAEONTOLOGY, VOLUME 8

potonie, R. 1956. Synopsis der Gattungen der Sporae Dispersae. I Teil: Sporites. Beih. Geol. Jahrb.
23, 1-103.

1958. Idem. II Teil: Sporites (Nachtrage), Saccites, Aletes, Praecolpates, Polyplicates, Mono-

colpates. Ibid. 31, 1-114.

1960. Idem. Ill Teil: Nachtrage Sporites, Fortsetzung Pollenites. Mit Generalregister zu Teil

1-111. Ibid. 39, 1-189.

1962. Synopsis der Sporae in situ. Ibid. 52, 1-204.

and kremp, g. o. w. 1954. Die Gattungen der palaeozoischen Sporae Dispersae und ihre Strati-

graphie. Geol. Jb. 69, 1 1 1-94.

1955. Die Sporae Dispersae des Rhurkarbons, ihre Morphographie und Stratigraphie mit

Ausblicken auf Arten anderer Gebiete und Zeitabschnitte. Teil I. Palaeontographica B 98, 1-136.
1956u. Idem. Teil II. Ibid. 99, 85-191. 1956 b. Idem. III. Ibid. 100, 65-121.

and Schweitzer, h. j. 1960. Der Pollen von Vllmannia frumentaria. Palaont. Z. 34, 27-39.

reinhardt, p. 1964. Uber die Sporae Dispersae der Thiiringer Trias. Sonderdruck aus Monatsberichte
Deutscb. Akad. \ Viss. 6, 46-56.

richardson, l. 1905. On the occurrence of Rhaetic rocks at Berrow Hill near Tewkesbury. Quart.
J. geol. Soc. Loud. 61, 425-30.

romanovskaja, g. m. 1962. Triassic, Lower and Middle Jurassic spore and pollen complexes of
Western Kazakhstan. (Resume.) Pollen et Spores, 4, 373-4.

roselt, g. 1955. Eine neue mannliche Gymnospermenfruktifikation aus dem Unteren Keuper
von Thtiringen und ihre Beziehungen zu anderen Gymnospermen. \ Viss. Z. Univ. Jena. -2,
75-118.

1958. Neue Koniferen aus dem Unteren Keuper und ihre Beziehungen zu verwandten Fossilen

und Rezenten. Ibid. 4-5, 387-409.

samoilovich, s. r. 1953. Pollen and Spores from the Permian deposits of the Cherdinsk and Aktju-
binsk Ural regions. Trudy VNIGRI, n.s., 75, 1-56. [In Russian.] Translation by M. K. ellas, Okla.
Geol. Surv. Circ. 56.

schultz, E. von, and krutzsch, w. 1961. Echinitosporites iliacoides nov. fgen. et fsp., eine neue
Sporenform aus dem Keuper der Niederlausitz. Geologie, 10, 122-7.

scott, r. a. 1960. Pollen of Ephedra from the Chinle Formation (Upper Trias) and the genus Equise-
tosporites. Micropaleontology, 6, 271-6.

seward, a. c. 1914. Antarctic fossil plants. British Antarctic Expedition 1910. Nat. Elist. Rept., London,
1914.

1933. An Antarctic pollen-grain; fact or fancy? The New Phytologist, 32, 311-13.

sherlock, r. L. 1926. A correlation of Permo-Triassic rocks. Proc. Geol. Assoc. 37, 1-72. 1928, Idem.
Ibid. 39, 48-95.

sitholey, r. v. 1951. On the occurrence of two winged pollen in the Triassic rocks of the Salt Range,
Punjab. Current Science , 20, 266.

taugourdeau-lantz, j. 1962. Contribution a la connaisance de la microflore du Trias. (Resume.)
Pollen et Spores, 4, 360.

1963. Note preliminaire a une etude sur la microflore du Trias franqais. Mem. B.R.G.M. frangais,

15, 570-5.

and jekhowsky, b. de. 1959. Spores et Pollen du Keuper, Jurassique, et Cretace inferieur d’Aqui-

taine. C. r. de la Soc. geol. frangais, 167.

taylor, J. h. 1953. The spore content of the Leigh Creek Coal. S. Austr. Dept. Mines. Min. Rev. 99,
155-70.

townrow, J. A. 1960. The Peltaspermaceae, a Pteridosperm family of Permian and Triassic age.
Palaeontology, 3, 333-61.

1962<7. On Pteruchus, a microsporophyll of the Corystospermaceae. Bull. Brit. Museum (Nat.

Hist.) Geol. 6, 289-320.

• 19626. On some Disaccate pollen grains of Permian age to Middle Jurassic age. Grana Palyno-

logica, 3, 13-14.

vernon, r. d. 1910. On the occurrence of Schizoneura paradoxa Schimper and Mougeot, in the Bunter
of Nottingham. Proc. Cantab. Phil. Soc. 15, 401-5.

Walton, J. 1925. On some South African fossil woods. Ann. S. Afr. Museum, 22, 1.

R. F. A. CLARKE: KEUPER MIOSPORES FROM WORCESTERSHIRE 321

wills, l. j. 1910. On the fossiliferous Lower Keuper rocks of Worcestershire. Proc. Geol. Assoc. 21,
249-331.

wilson, l. r. and venkatachala, b. s. 1963a. Thymospora, a new name for Verrucososporites. Okla.
Geol. Notes, 23, 75-79.

19636. Morphological variation of Thymospora pseudothiessenii (Kosanke) Wilson and

Venkatachala 1963. Ibid. 23, 125-32.

— — - and webster, r. m. 1946. Plant microfossils from a Fort Union coal of Montana. Amer. J. Bot.
33, 271-8.

R. F. A. CLARKE

Bataafse Internationale Petroleum Maatschappij N.V.,
Carel van Bylandlaan 30,

The Hague,

Manuscript received 27 June 1964 Netherlands

BRITISH PERMIAN SACCATE AND
MONOSULCATE MIOSPORES

by R. F. A. CLARKE

Abstract. Thirty-three species belonging to seventeen genera are recorded and described from the British Upper
Permian (Zechstein). Six species are considered to be new. Three variants are described for Lueckisporites
virkkiae Potonie and Klaus 1954 and the diagnosis is emended. The sample localities, brief lithological descrip-
tions, and some spore frequencies are given together with a comparison of the present assemblages with those
previously described from other parts of the world. It is concluded that a uniform flora existed throughout the
Upper Permian in Great Britain and that this differed little from the Permian vegetation of Western Europe in
general.

Previous work on the miospores of the British Permian is limited to a short publica-
tion by Chaloner and Clarke (1962) and the observations of Jansonius (1962). The aim
of the present paper therefore is to describe the British Permian microfloral assemblages
and to compare these with those previously described from other parts of the world.

Slide collection and maceration technique. The majority of the specimens illustrated
in this paper are from single spore mounts. The method of making these and the pre-
paration of the residues from the samples has already been described (Clarke 1965,
this volume). The slide collection is housed in the Geological Survey and Museum,
London.

Classification and terminology. The system used here to group spores into supra-generic
categories is the ‘Morphographic Classification’ outlined by Potonie and Kremp (1954,
1955, 1956) coupled with subsequent additions (Potonie, 1956, 1958, 1960) and modifica-
tions set up by later authors. Some of the less obvious terms employed to describe the
morphology of the different spore types are illustrated in text-fig. 1.

Sample localities

1. Hilton, Westmorland. The Hilton Plant Bed is exposed in Hilton Beck, 3 miles east-north-east of
Appleby, Westmorland. The junction between the Penrith Sandstone and the Plant Beds is seen in the
river bank on the north side at the western end of Ash Bank Wood. Most of the Plant Beds are exposed
in the river bluff on the south side of the beck and consist of a series of well-bedded, alternating
sandstones and thin pale-grey or olive-green shales (text-fig. 2).

2. Kimberley, Nottinghamshire. A 20-foot section of Lower Permian Marl (= Marl Slate of some
authors) is exposed on the south side of the more northerly railway cutting (disused) some 500 yards
west of the tunnel at Kimberley, west of Nottingham. At this point a 4-foot-thick Permian breccia
overlies, unconformably, Carboniferous Coal Measure Sandstone. This is overlain by the alternating
sandstones and shales of the Lower Permian Marl, containing ill-preserved plant remains. The top of
the cutting exposes the Lower Magnesian Limestone, which here is arenaceous and gritty, obscuring
a clear demarcation between this and the Lower Permian Marl (text-fig. 3).

3. Haughton Hall Boring. Located 4 miles south of East Retford, Nottinghamshire. All Permian
samples examined between 987-1,1 17 feet contain spores.

Acknowledgements. I would like to express my thanks to Professors S. E. Hollingworth and T. Barnard
for the use of laboratory facilities at University College, London. The help of Mr. M. A. Calver and
the Geological Survey of Great Britain who kindly made available the Haughton Hall Borehole

[Palaeontology, Vol. 8, Part 2, 1965, pp. 322-54, pi. 40-44.]

R. F. A. CLARKE: PERMIAN SACCATE AND MONOSULCATE MIOSPORES 323

material and gave permission for the data to be published, and Mrs. H. H. Stoneley who supplied
information on the Kimberley section is acknowledged with gratitude. To Dr. J. Jansonius I owe
special thanks for allowing me complete freedom with the study of the Hilton Plant Beds after he had
already started work on this section himself. The present paper formed part of a larger study of
British Permo-Triassic spores, for a doctoral thesis, under the supervision of Dr. W. G. Chaloner to
whom I am greatly indebted for immeasurable help and the critical reading of this manuscript. Finally
I should like to thank the Department of Scientific and Industrial Research from whom I have been
in receipt of a grant.

o - SPORE BODY WIDTH,
b - SACCUS WIDTH,
c - SPORE BODY LENGTH
d - SACCUS LENGTH
e- SACCUS OFF-LAP.
f - SACCUS OVER LAP.

LA TERAL

text-fig. 1 . A generalized bisaccate miospore, in polar view, illustrating various terms used in this paper.

SYSTEMATIC SECTION

Anteturma pollenites R. Potonie 1931
Turma saccites Erdtman 1947

Subturma monosaccites (Chitaley 1951) Potonie and Klaus 1954
Infraturma vesiculomonoradites (Pant) Bhardwaj 1955
Genus potonieisporites Bhardwaj 1954

1962 Hoffmeisterites Wilson, pi. 3, fig. 4.

Type species. P. novicus Bhardwaj 1954.

Discussion. P. novicus and other species since assigned to the genus all show a series of
folds which, broadly speaking, may be resolved into two separate sets. The first is

PALAEONTOLOGY, VOLUME

O,

y

(/>

on

<

or

I-

o

y HLSI -
/ HL50 -

3 00

/HL49 -

A

Magnesian

Limestone

yHL48 -
• HL47-

/HL46 -
• HL45-

n

•HL44-

>HL43-

• HL42 -

• HL4| -
•HL40-

c c
ft) D
CL U)

I oo'

>HL39 -

y HL38 -
yHL 37
y HL 36 -

HLI2-HL35

HLl-HLil-

_Bimoda / Sand stone
Red Shale

Mognes /on L imestone
( maybe 20 feet)

Magnesian Limestone
Grey sandy Shale
-Gap

Purple and red Shales sometimes
calcareous

_Gap

JPurptc Sandstone and grey Shale

Gap

Purple sandy shales
wilh thin Sandstone
bands

Red t grey micaceous Sands to
Sandstone

_ Purple Shale
_Grey Shale
_Purple sandy shale

Gap

Purple sandy shale
Gap .

Purple sandy shales with thin sandstone
Gap

_ Thin sandstone
G op

Sandstone with thin
shale partlnqs

Red micaceous sandstone

Soft red grey micaceous
sandstone and thin shale

Alternations of soft and hard Sandstones
and grey mar!

Plant Remains

Gap

Grey Marls and yellow Sandstones
Plant Remains

• Spores present
y Spores absent

text-fig. 2. Stratigraphic section of the Upper Permian at Hilton
Beck, Westmorland, showing the sample numbers and positions.

R. F. A. CLARKE: PERMIAN SACCATE AND MONOSULCATE MIOSPORES 325

Coarse dork red highly col core ous
mi coceous Sondstone and Grits
No Plant Remains

Medium grained non laminated calc. Si Itstone . Plant Remains
Yellow , red } la mmoted , micaceous ,

Sandstone with thinner more

argillaceous bonds . Abundant Plant Remains.

Red laminated calc. Sandstone . Plant Remains
Yellow non-lominoted Sandstone ■ Plant Remains.

Red t yellow argi l/oceous J laminated
Sandstones , with thin Marl bonds.

Yellow calcareous laminated

Sit t stone . Abundant Plant Remains

Red , yellow ^medium groinedt
lominat ed , co/coreous
Sandstone with thinf
light grey Sho/e bonds.

Ill preserved Plant Remains

Blue grey calcareous non-
lamina t ed SH tst one

SECTION OBSCURED BY
SLUMPED MATERIAL

Coarse Breccia

• Spores present
/ Spores absent

text-fig. 3. Stratigraphic section of the Upper Permian (Lower Permian Marl) at
Kimberley, Nottinghamshire, showing the position and number of samples.

a set disposed along the inner margin of the spore body while the second set is situated
closer to the polar region and orientated perpendicular to the long axis of the grain.
It is the interpretation of the position and function of these folds which has led to
certain differences of opinion with respect to the reconstruction of the spore. Bhardwaj
(1956) interprets both sets of folds as compression features and regards the saccus as
being free (unattached) in the distal region (Bhardwaj 1956, text-fig. 11). A somewhat

326

PALAEONTOLOGY, VOLUME 8

different view is taken by Potonie and Lele (1961) who suggest that both series of folds
are situated distally and form part of a single set which more or less delimits the germinal
area. From the present study I accept the latter view and believe that the saccus is
attached in the region of these folds (text-fig. 4 b).

Comparison. Vestigisporites Hart 1960 is a monosaccate, monolete genus which, in
polar view, can appear similar to Potonieisporites (e.g. V. methoris Hart 1960). Vestigi-
sporites differs, however, in the less well-developed saccus swelling laterally, which is

text-fig. 4. Diagrammatic reconstruction of Potonieisporites
Bhardwaj, in lateral (equatorial view), a. The reconstruction
of Bhardwaj ( 1956), showing the saccus unattached (free) over
the distal surface, b. The present interpretation where the
saccus is attached to the spore body in this region.

never as large as that terminally. Florinites Schopf, Wilson, and Bentall 1944 has
a distal attachment of the saccus, which is free proximally, and also has a triradiate
mark. The lateral constriction of the saccus and the characteristically thickened ‘sulcus’
is sufficient to distinguish Sahnisporites Bhardwaj 1954 from Potonieisporites.

Potonieisporites novicus Bhardwaj 1954
Plate 40, fig. 6, Plate 44, fig. 13; text-fig. 4

Comparison. P. neglectus Potonie and Lele 1961 differs from this species in the poly-
gonal or trapezoid shape of the spore body and the smaller width of the saccus laterally.
P. simplex Wilson 1962 is a smaller form in which lips are developed around the laesura
(commissura of Wilson), while P. bhardwaji Remy and Remy 1961 appears to differ

R. F. A. CLARKE: PERMIAN SACCATE AND MONOSULCATE MIOSPORES 327

in its larger size, absence of a rim separating the saccus edge and the spore body, and
a longer monolete mark.

Genus vestigisporites Balme and Hennelly 1955 emend. Hart 1960
Type species. V. rudis Balme and Hennelly 1955.

Comparison. Vestigisporites differs from lilinites (Kosanke) Potonie and Klaus in being
an essentially monosaccate genus with the bisaccate construction as the exception, and
lacks the variation in the shape of the tetrad scar exhibited by lilinites.

text-fig. 5. Vestigisporites minutus sp. nov. a, X 1,000 (based on holotype).

Polar view showing the monolete mark (in rectangle), the proximal surface
on the right-hand side, and the distal surface with ‘sulcus’ on the left.
b. Lateral polar section, c. Terminal polar section.

Vestigisporites minutus sp. nov.

Plate 40, figs. 7-9; text-fig. 5

Holotype. Plate 40, fig. 7. Slide PF2196. Sample H 12, Hilton Beck, Near Appleby, Westmorland;
Upper Permian.

Diagnosis. Small, bilateral, monosaccate miospores. Haploxylonoid. Spore body cir-
cular, rarely oval. Exine thin. Monolete. Terminal saccus swelling small compared with
spore body; connected laterally by exoexinal strip. Saccus finely infra-reticulate. The
saccus attachment distally leaves an elongated oval area free where the exine is thin.

Description. The grains are fossilized giving a preferred orientation flattened in the
equatorial plane. The circular spore body is dark coloured and the monolete mark is
short, often indistinct and sometimes open. In polar view the saccus completely sur-
rounds the spore body but is not as wide laterally as terminally. One edge of the saccus
appears to be attached equatorially and the other edge attached distally. The saccus
offlap terminally (text-fig. 1) is slightly greater than the overlap on to the spore body.

Dimensions. (Thirty specimens.) Spore-body length 24(29)31 p, spore-body width
28(32)35 p, overall length 42(48)54^, overall width 32(35)40 p.

Comparison. V. minutus sp. nov. is most similar to V. ItenneUyi Hart 1960 but differs in
its very much smaller size and darker spore body. A reduction of the lateral saccus

z

C 3009

328

PALAEONTOLOGY, VOLUME 8

extension to give a bisaccate condition has not been observed for V. minutus sp. nov. ;
such a condition exists for V. hennellyi (Hart 1960, p. 15).

Infraturma triletesacciti Leschik 1955
Genus perisaccus Naumova 1953 emend. Klaus 1963

1955 Simplicesporites Leschik.

Type species. P. verruculatus Naumova 1953.

Discussion. The genus Perisaccus is based on Russian Upper Devonian material and
first appears as a nomen nudum in Naumova (1937); the genus is validated by Naumova

text-fig. 6. Diagrammatic reconstructions of
Perisaccus granulosus comb, nov., X 1,000.
a. Polar view, b, Polar section, showing the
relationship of the saccus to the spore body.

(1953). The emendation of Potonie (1958)
describes the spore as monosaccate, without
a Y-mark and with an infra-reticulate saccus.
A different aspect is given by the subsequent
emendation of Klaus (1963) who ascribes to
the genus a small Y-mark and a granular
saccus sculpture.

Leschik (1955, 1956) describes and figures
spores assigned to Simplicesporites Leschik
which are almost certainly the same generi-
cally as those described by Klaus (1963) as
Perisaccus. Leschik describes Simplicesporites
as a ‘zonate’ genus and makes no reference
to a Y-mark. However, the British material
shows a development of secondary folds
which cross the spore body margin without
displacing it, and two independent super-
imposed sets of secondary folds can be seen
within the area of the zona. These observa-
tions are consistent with a saccate rather than
a zonate structure. Such features are clearly
seen in Leschik’s 1955 pi. 5, fig. 3. The
inequality of the zona width described by
Leschik for S. virgatus, &c., appears to be
merely due to the difference in orientation of
the grain when fossilized. The triradiate
mark is always small and appears often as
a triangular tear in the exine. Although not
clearly demonstrable it appears that the
saccus is attached proximally and is separated
from the endexine distally (text-fig. 6).

Comparison. Perisaccus differs from Florinites
in the position of the saccus attachment,
which is distal in the latter genus. Endo-
sporites Wilson and Coe possesses a larger

R. F. A. CLARKE: PERMIAN SACCATE AND MONOSULCATE MIOSPORES 329

Y-mark and a limbus which is not present in Perisaccus. Remysporites Butterworth and
Williams 1958 differs also in the larger Y-mark, and its larger overall size. Nuskoisporites
Potonie and Klaus has the saccus attached both proximally and distally. While Leschik's
genera Accinctisporites, Patinasporites, Zonalasporites, and Succinctisporites are mono-
saccate, the attachment of the saccus is not discussed by that author and all are con-
sidered to be alete.

Perisaccus granulosus (Leschik 1955) comb. nov.

Plate 42, figs. 10-11; text-fig. 6

1955 Simplicesporites granulosus Leschik.

1963 Perisaccus granulatus Klaus, pi. 4, fig. 12.

Description. The central body, as seen in polar view, is circular or oval and darker
coloured than the saccus. The spore body is smooth, the exine Ip thick and there is
a small Y-mark, often in the form of a triradiate tear. The Y-mark is less than one-
third the spore radius and often open. The saccus is attached to the spore body proxi-
mally; the distal surface remaining free. The saccus offlap (width) is less than half the
spore body radius, and sculptured with closely packed isodiametric granules Ip in
diameter.

Dimensions. (Twenty specimens.) Spore-body diameter 35(44)51 p, saccus width 6(9) 13^,
overall diameter 54(62)72 p.

Comparison. P. granulosus comb. nov. differs from P. pendens (Leschik 1955) comb,
nov., the latter having a triangular outline, although the spore body is circular. Other
than this the two species are closely similar.

Perisaccus laciniatus (Leschik 1955) comb. nov.

Plate 42, fig. 12

Comparison. P. laciniatus comb. nov. differs from P. virgatus comb. nov. in the lack of
spines attached to the granules, and from P. granulosus comb. nov. in the larger saccus.

Genus nuskoisporites Potonie and Klaus 1954
1944 Hymenozonotriletes Mehta, pi. 1, fig. 1.

1951-2 Hymenozonotriletes Mehta; Goswami, pi. 13, fig. 11 (not fig. 10 as stated).

1951-2 (?) Florinites sp. Schopf, Wilson, and Bentall; Goswami, pi. 12, fig. 4.

Type species. N. dulhuntyi Potonie and Klaus.

Discussion. The diagnosis of Potonie and Klaus (1954) is written broadly around the
type species; the genus at that time being monotypic. Some authors regard the presence
of a limbus as an essential feature of the genus (Balme and Hennelly 19566, Pierart
1959), while others (e.g. Potonie and Lele 1961) take a broader generic concept.

The saccus in Nuskoisporites is attached both proximally and distally leaving a non-
cavate area over the proximal and distal poles where the exine (exoexine plus endexine)
is thinner. The actual site of saccus attachment is seldom clearly shown and appears to
be variable. It thus seems that Nuskoisporites may have functioned as a pollen with
distal germination thus differing from a monosaccate microspore like Endosporites.

330

PALAEONTOLOGY, VOLUME 8

Separation of species is based primarily upon the length and form of the Y-mark,
the saccus reticulum and the width of the saccus in relation to the spore body radius.
Many authors (Virkki 1945, Potonie and Lele 1961, Hoeg and Bose 1960) have remarked
on the considerable variation within the species recognized.

Nuskoisporites dulhuntyi Potonie and Klaus 1954
Plate 40, figs. 1-2

Discussion. The cardinal characteristics of this species are the short Y-mark, the uni-
form width of this feature, and the presence of a limbus. N. rotatus Balme and Hennelly
differs from the present species in the larger spore body, while N. triangularis (Mehta)
Potonie and Lele 1961 possesses a larger Y-mark and lacks a limbus. N. crenulatus
Wilson 1962 is a smaller form, without a limbus, and with a crenulate contact edge
where the saccus overlaps the spore body. N. radiatus Hennelly 1958 lacks a limbus and
has a finely infra-reticulate saccus.

Nuskoisporites cf. rotatus Balme and Hennelly 19566
Plate 40, fig. 3

Discussion. The present specimens lack a limbus. Balme and Hennelly (19566, p. 245)
state that such a feature is only sometimes present and no mention of the limbus is
made by Hoeg and Bose (1960). N. rotatus is characterized by its small saccus width,
long Y-mark, and the general absence of a limbus.

Infraturma striasacciti Bhardwaj 1962 (= striatornati Jansonius 1962)

Genus crustaesporites Leschik 1956 emend. Jansonius 1962

1955 Lueckisporites Potonie and Klaus; Balme and Hennelly, pi. 4, fig. 44.

1955 Lueckisporites Potonie and Klaus; Klaus, pi. 33, fig. 5.

1962 ‘Multistriate, monosaccate grain’, Jizba, pi. 122, fig. 24.

Type species. C. globosus Leschik 1956.

Discussion. Leschik (1955) describes Crustaesporites as a trisaccate genus. Jansonius
(1962), however, considers the trisaccate appearance to be the result of irregular con-
striction of a monosaccate miospore. The British material shows considerable variation

explanation of plate 40
Magnification x 750 unless otherwise stated.

Figs. 1-3. Nuskoisporites spp. 1-2, N. dulhuntyi Potonie and Klaus. 1, PF2189. 2, PF2190. 3. N. cf.
rotatus Balme and Hennelly, PF2191.

Figs. 4-5. Crustaesporites globosus Leschik. 4, Trisaccate condition, PF2193. 5, Showing the develop-
ment of four sacci; PF2194.

Fig. 6. Potonieisporites novicus Bhardwaj, X 500, PF2202.

Figs. 7-9. Vestigisporites minutus sp. nov. 7, Holotype, PF2196. 8, PF2197. 9, PF2198.

Figs. 10-11. Falcisporites zapfei Leschik. 10, PF2318. 11, PF2274.

Fig. 12. Illinites ktausi sp. nov., holotype, PF2290.

Localities of figs. 1-2, 4-6, 12, Lower Permian Marl, Kimberley. Figs. 7-11, Hilton Plant Bed. Fig. 3,
Haughton Hall Boring, Lower Permian Marl, depth 1,095 feet.

Palaeontology , Vol. 8

PLATE 40

w y-

CLARKE, Permian miospores

/>>

R. F. A. CLARKE: PERMIAN SACCATE AND MONOSULCATE MIOSPORES 331

between a trisaccate condition and an irregular monosaccate structure (text-fig. 7,
PI. 40, figs. 4-5). While it is perhaps reasonable to consider the original structure as
monosaccate, such irregularities of saccus structure have been demonstrated to occur
as ‘aberrants’ of living bisaccate pollen (Van Campo-Duplan 1947, 1950, Van Campo-
Duplan and Gaussen 1948, Martin 1961).

The frequency of occurrence of Crustaesporites in the present material is less than
1 in 2,000 and it is always associated with bisaccate striate genera. This association is
maintained in all other localities from which Crustaesporites has been recorded ( Europe,

text-fig. 7. Polar views of Crustaesporites globosus Leschik (drawn from specimens) illustrating
variation in the outline of the saccus structure, a. More or less trisaccate condition, b and c. Irregular

saccus.

Canada, Australia). In number of taeniae and saccus sculpture Crustaesporites re-
sembles the bisaccate genus Protohaploxypinus most closely and probably represents
aberrant spores of this genus. Nevertheless, I consider their recognition as a distinct
genus an inevitable consequence of a morphographic treatment.

Crustaesporites globosus Leschik 1956
Plate 40, figs. 4-5

The variation in the shape of the saccus structure, encountered in the present material,
is shown in text-fig. 7.

Subturma disaccites Cookson 1947
Infraturma striatiti Pant 1954

Genus lueckisporites Potonie and Klaus 1954 emend. Klaus 1963
Type species. L. virkkiae Potonie and Klaus 1954.

Lueckisporites virkkiae (Potonie and Klaus 1954) emend.

Plate 43, figs. 3, 6-11 ; text-fig. 8
1960 Lueckisporites nyakapendensis Hart, pi. 1, fig. 12.

Emended diagnosis. Bilateral, bisaccate pollen grains. Sometimes haploxylonoid but
typically diploxylonoid in overall outline. Spore body circular or oval where the length
exceeds the width. Proximal face possesses a variable thickening (Kalotte of Potonie

332

PALAEONTOLOGY, VOLUME 8

and Klaus) which is split by a laesura parallel to the long axis of the spore into two,
more or less equal, halves. Sculpture is infrapunctate or infrabaculate. A monolete mark
may be present. Sacci semicircular or more in outline, well developed and discrete.
Sacci offlap may be considerable or non-existent. One saccus edge is attached at the
equator; the attachment of the other being variable. Infra-sculpture of anastomosing
muri forming a microreticulum or punctation; both types may show a radial pattern
developed from the distal saccus roots. The exine is thin.

text-fig. 8. Lueckisporites virkkiae Potonie and Klaus emend. Diagrammatic recon-
structions to show the differences between the variants A, B, and C. a-c. Variant A.
d-f, Variant B. g-i. Variant C. a, d, g. Proximal polar views, b, e, h. Lateral views.
c, f, i, Terminal polar sections.

Discussion. The British Permian has yielded many specimens referable generically to
Lueckisporites s.str. There is a great variety of extreme forms connected by inter-

, , . _.C(, EXPLANATION OF PLATE 41

Magnification X/50.

Figs. 1-3. Protohaploxypinus spp. 1-2, P. jacobii Hart. 1, PF2216. 2, PF2217. 3, P. microcorpus comb,
nov., PF2218.

Figs. 4, 8. Taeniaesporites spp. 4, T. nubilus comb, nov., PF2226. 8, T. bilobus sp. nov., holotype,
PF2221.

Fig. 5. Labiisporites granulatus Leschik, PF2272.

Figs. 6-7. Illinites tectus comb. nov. 6, PF2287. 7, PF2289.

Figs. 9-10. Striatopodocarpites fusus Potonie. 9, PF2222. 10, PF2223.

Localities of figs. 1, 2, 4, 8, 10, Lower Permian Marl, Kimberley. Figs. 3, 6, 7, Haughton Hall Boring,
Lower Permian Marl, depth 1,095 feet. Figs. 5, 9, Hilton Plant Bed.

Palaeontology, Vol. 8

PLATE 41

CLARKE, Permian miospores

~fTr*s

R. F. A. CLARKE: PERMIAN SACCATE AND MONOSULCATE MIOSPORES 333

mediates, making the separation of several species impossible. Potonie, who has seen
the material here under discussion, agrees that the entire range can be included in L.
virkkiae (personal communication 1962). For this reason it has been thought desirable
to emend the specific diagnosis to include all these forms. Within the species three main
extremes can be recognized and these are here referred to as variants A, B, and C (text-
fig. 8). At no Permian horizon, so far studied, is one of the variants absent although
there may be an assemblage shift in any of these directions. Variant A is characterized
by the well-developed proximal thickenings, their distinct separation and well-developed
sacci (PI. 43, figs. 3, 8, 9). This variant is most similar to the holotype and L. microgranu-
latus Klaus 1963. Variant B differs in the less well-developed sacci and the negligible
amount of oflflap (PI. 43, figs. 10-11). Variant B is similar to L. parvus Klaus 1963.
Variant C is recognized by a weakly developed proximal cap and its incomplete separa-
tion into two halves and the generally smaller overall size (PI. 43, figs. 6-7). L. micro-
granulatus (kleinere variante) Klaus 1963 is most similar to variant C.

Genus taeniaesporites Leschik emend. Klaus 1963

1954 Lueckisporites Potonie and Klaus (pars).

1955 Lunatisporites Leschik, pi. 7, figs. 21-24.

1955 Succinctisporites Leschik, pi. 7, figs. 4-5.

1956 Jugasporites Leschik (pars).

1958 Pollenites Pautsch, pi. 1, fig. 8.

1962 Lueckisporites Potonie and Klaus; Grebe and Schweitzer (pars).

1963 Striatites Pant; Schaarschmidt (pars).

1963 Striatites Pant; Klaus (pars).

Type species. T. kraeuseli Leschik 1955.

Discussion. While I accept the arguments of Klaus (1963) concerning previous uses and
emendations of the genus and follow his emendation, I feel that the genus should be
broadened to include all bisaccate miospores with four primary taeniae and which are
haploxylonoid or diploxylonoid in outline, and not to restrict the genus to those forms
showing a wide separation of the central taeniae. This is better considered a specific
character.

Comparison. Taeniaesporites differs from Lueckisporites s.str. in possessing more than
two taeniae, from Protohaploxypinus emend. Flart 1964 in having less than six primary
taeniae, from Striatopodocarpites emend. Hart 1964 in the spore body to saccus ratio,
and from Striatoabietites emend. Hart 1964 in the presence of fewer taeniae.

Taeniaesporites noviaulensis Leschik 1956
Plate 42, figs 6-7

1962 Taeniaesporites novimundi Jansonius (pars), pi. 13, fig. 25 only.

1962 Lueckisporites noviaulensis Grebe and Schweitzer, pi. 5, fig. 7.

1963 Striatites noviaulensis Schaarschmidt, pi. 15, figs. 5-7, 9.

1963 Taeniaesporites ortisei Klaus, pi. 14, figs. 67-70.

Comparison. T. noviaulensis differs from T. novimundi Jansonius in the shape of the
spore body, the larger sacci and the coarser infra-reticulum; and from T. kraeuseli in
the shape of the spore body, form of the taeniae, and the shape and sculpture of the sacci.

334 PALAEONTOLOGY, VOLUME 8

Taeniaesporites novimundi Jansonius 1962
Plate 44, figs. 1-2

The British specimens agree well with those described by Jansonius (1962) from the
Permo-Triassic of Canada.

Taeniaesporites angulistriatus (Klaus 1963) comb. nov.

Plate 44, figs. 11-12

1963 (May) Striatites angulistriatus Klaus, pi. 17, fig. 83.

1963 (August) Striatites ovaiis Schaarschmidt, pi. 15, figs. 1-4.

Discussion. The most distinctive features of this species are the small size, the narrow
distal area between the sacci attachments and the scabrate sculpture of both the taeniae
and the sacci. T. angulistriatus comb. nov. differs from T. kraeuseli Leschik in the shape
and sculpture of the sacci, and from other species in its taeniae and saccus sculpture.

Taeniaesporites albertae Jansonius 1962
Plate 44, fig. 5

Discussion. This species is distinguished by its broad, slightly thickened taeniae, and the
lack of radial alignment of the saccus sculpture. This species differs from T. tioviaulensis
Leschik and T. novimundi Jansonius in the broader taeniae, small sacci and the finer
reticulum. T. kraeuseli differs in the more embracing sacci giving a narrow distal area.

Taeniaesporites labdacus Klaus 1963
Plate 44, figs. 6-10; text-fig. 9

1954 Lueckisporites sp. Potonie and Klaus, pi. 10, fig. 2.

1962 Lueckisporites tioviaulensis Grebe and Schweitzer, pi. 5, fig. 8 ( non Leschik).

Remarks. This species is characterized by the presence of four taeniae of which the
central pair are better developed than the lateral ones, and which may join terminally
to form an elevated rectangular area around the proximal pole. A monolete mark is
usually present, and thick muri form a coarse infra- reticulum on the sacci.

EXPLANATION OF PLATE 42
Magnifications x 750 unless otherwise stated.

Figs. 1-2. Striatopodocarpites antiquus Potonie. 1, PF2438. 2, PF2225.

Figs. 3-5. Protohapioxypinus chaloneri sp. nov. 3, PF2211. 4, Holotype, PF2210. 5, PF2212.

Figs. 6-7. Taeniaesporites tioviaulensis Leschik. 6, PF2230. 7, PF2229.

Figs. 8-9. Protohapioxypinus cf. samoiiovichii Hart. 8, PF2208. 9, PF2209.

Figs. 10-12. Perisaccus spp. 10-11, P. granulosus comb. nov. 10, PF2204. 11, PF2206. 12, P. iaciniatus
comb, nov., PF2483.

Fig. 13. Taeniaesporites novimundi Jansonius showing taeniae sculpture, X 1,500, PF2232.

Localities of figs. 1, 3, 4, 5, 10, 12, Hilton Plant Bed. Remainder from the Lower Permian Marl,
Kimberley.

Palaeontology, Vol. 8

PLATE 42

CLARKE, Permian miospores

R. F. A. CLARKE: PERMIAN SACCATE AND MONOSULCATE MIOSPORES 335

The variation shown by the present material makes it difficult to maintain the dif-
ference between this species and T. alatus Klaus 1963. The present forms are assigned
to T. labdacus as this is the first of these two species described by Klaus (1963).

text-fig. 9. Taeniaesporites labdacus Klaus, a, Polar view
showing the proximal face with the taeniae and monolete
mark, on the right, and the distal face on the left (drawn from
specimen), x 1,000. b and c. Diagrammatic reconstructions.
b. Lateral view, c. Terminal polar section.

Taeniaesporites bilobus sp. nov.

Plate 41, fig. 8; text-fig. 10

Holotype. Plate 41, fig. 8. Slide PF2221. Sample K 14, Kimberley, Nottinghamshire; Upper Permian
(Lower Permian Marl).

Diagnosis. Spore body small, circular, dark coloured, and bearing up to five taeniae.
Sacci relatively large and not connected laterally. The saccus sculpture is a medium
infra-reticulum with a radial alignment from the saccus roots.

Description. The spore body proximal face is covered by taeniae, more or less parallel-
sided, about 5p wide, and separated by smooth striae. The taeniae are micropunctate
and are frequently divided by transverse splits. A monolete mark is sometimes present.
The saccus offlap greatly exceeds that of the overlap and the distal saccus attachment is
indistinct; the saccus muri are 1-2 p thick.

Dimensions. (Seven specimens.) Spore-body length 27(29)32^, spore-body width
28(30)33 p,, overall length 60(68)79^.

336

PALAEONTOLOGY, VOLUME 8

Comparison. This species differs from other species of the genus in the comparatively
larger sacci and the circular spore body.

text-fig. 10. Taeniaesporites bilobus sp. nov. a, Polar view
showing the arrangement of the proximal taeniae, on the
right, and the distal surface on the left (based on the holo-
type), X 1 ,000. b, and c. Diagrammatic reconstructions.
b, Lateral view, c. Terminal polar section.

Taeniaesporites nubilus (Leschik 1956) comb. nov.
Plate 41, fig. 4

1956 Jugasporites nubilus Leschik, pi. 21, fig. 14.

1962 Striatites? nubilus Jansonius, pi. 14, fig. 20.

1963 Striatites rarostriatus Schaarschmidt, pi. 14, fig. 8.

Discussion. This species is characterized by its dumbell shape, thick spore body exine,
and the coarse saccus infra-reticulum.

No mention of striations is made in the specific description of Jugasporites nubilus
by Leschik (1956) but the illustration of the holotype gives the impression that this
specimen is striate (see also Jansonius 1962). I accordingly assign it to Taeniaesporites.

Comparison. T. nubilus comb. nov. differs from T. bilobus sp. nov. in the more oval
spore body and the thicker exine, and from the type species in the larger sacci.

Genus protohaploxypinus Samoilovich emend. Hart 1964

1963 Striatites Pant emend. Klaus, pi. 17, figs. 79-82.

1963 Striatites Pant; Schaarschmidt, pi. 14, figs. 3-7, pi. 15, figs. 8a-8b.

Type species. P. (al. Pemphygaletes ) latissimus Luber and Waltz 1941.

R. F. A. CLARKE: PERMIAN SACCATE AND MONOSULCATE MIOSPORES 337

Comparison. Protohaploxypinus emend. Hart (1964, p. 1171) differs from Striatopodo-
carpites emend. Hart in being more haploxylonoid and in the more variable form of the
spore body. Lueckisporites s.str. and Taeniaesporites s.str. have fewer taeniae while
Complexisporites Jizba differs from Protohaploxypinus in the presence of a groove sur-
rounding a ‘fissured’ area on the proximal face. Striatosaccites Jizba differs from the
present genus in having transverse sexinal strips distally while Striapollenites Bhardwaj
1962 has taeniae parallel with the transverse axis (i.e. perpendicular to all the above
genera). Striatoabietites emend. Hart 1964 differs from Protohaploxypinus in showing
a distinct angle where the saccus joins the spore body, as seen in polar view.

Protohaploxypinus jacobii Jansonius emend. Hart 1964
Plate 41, figs. 1-2

Remarks. This species is similar to P. sewardi (Virkki) Hart 1964, differing only in its
larger size, while P. amplus (Balme and Hennelly) Hart 1964 has a larger saccus offlap.

text-fig. 11. Protohaploxypinus chaloneri sp. nov. a. Showing the form of the
proximal taeniae, on the right, and the distal surface on the left-hand side (based
on the holotype), X 1,000. b and c. Diagrammatic reconstructions, b, Lateral
view, c. Terminal polar section.

Protohaploxypinus cf. samoilovichii (Jansonius) Hart 1964
Plate 42, figs. 8-9

1963 Striatites samoilovichii Jansonius; Schaarschmidt, pi. 14, figs. 3-5.

Remarks. The present forms are identical, except for their smaller size, with those spores
described by Jansonius (1962) as Striatites samoilovichii.

Dimensions. (Twenty specimens.) Spore-body length 30(36)42^, spore-body width
36(41)48 ft, overall length 60(66)72fi.

Protohaploxypinus chaloneri sp. nov.

Plate 42, figs. 3-5; text-fig. 11

Holotype. Plate 42, fig. 4. Slide PF2210. Sample H 5, Hilton Plant Bed, Westmorland; Upper Permian.

Diagnosis. Spore-body amb circular or subcircular; proximal taeniae 10-12 in number,
sometimes anastomosing or interrupted. Sacci well developed and semicircular in out-
line. Saccus infra-sculpture is a well-defined microreticulum without a radial pattern

338

PALAEONTOLOGY, VOLUME 8

being developed. Sacci attachments distally show a crescent-shaped thickening which
rarely extends to the equator.

Description. In the fossil state the pollen grains are invariably flattened in the equatorial
plane and are therefore reconstructed from polar views. The spore body exine is 1-2^.
thick and the taeniae are separated by narrow striae. A monolete mark is not evident
although it is occasionally suggested either by the gaping of the more centrally placed
taeniae or by an elongated secondary fold in this region. The sacci are discrete and the
offlap is equal to the overlap, or somewhat greater. One edge of a saccus is equatorially
attached while the other is attached distally, typically midway between the equator and
the distal pole.

Dimensions. (Twenty-five specimens.) Spore-body length 30(36)42/l, spore-body width
30(37)41 /x, overall length 48(57)63 ju.

Comparison. P. minor (Klaus 1963) comb. nov. is most similar to the present species,
but differs in the shape of the sacci and the absence of the distal attachment thickenings.

Protohaploxypinus microcorpus (Schaarschmidt 1963) comb. nov.

Plate 41, fig. 3

1 963 Striatites jacobii Jansonius ; Klaus, pi. 1 7, fig. 79.

1963 Striatites microcorpus Schaarschmidt, pi. 14, figs. 6-7.

Remarks. This species, previously known from the Upper Permian of Germany and
Austria, shows a large number of taeniae on the proximal face. The number can only
be estimated at between ten and twenty, an accurate determination being precluded
owing to their crowding and interruption.

Genus striatopodocarpites Sedova emend. Hart 1964
1962 Striatites Pant; Jizba, pi. 122, figs. 25-30.

Type species. S. tojmensis Sedova 1956.

Discussion. The characteristic features of this genus are the distinctly diploxylonoid
outline and the circular spore body with more than five proximal taeniae. The dark
colour of the spore body observed in many species of Striatopodocarpites tends to

EXPLANATION OF PLATE 43
Magnification x 750 unless otherwise stated.

Figs. 1, 15. Alisporites nuthallensis sp. nov. 1, PF2232. 15, Flolotype, PF2277.

Fig. 2. Striatopodocarpites cancellatus comb, nov., PF2252.

Figs. 3, 8, 9. Lueckisporites virkkiae Potonie and Klaus emend., Variant A. 3, Part of ‘Kalotte’,
X 1,500, PF2256. 8, PF2319. 9, PF2257.

Figs. 4-5. Striatoabietites richteri Hart. 4, PF249 1/720220. 5, PF2487.

Figs. 6-7, 10-11. L. virkkiae emend. 6-7, Variant C. 6, PF2259. 7, PF2260. 10-11, Variant B. 10,
PF2263. 11, PF2262.

Figs. 12-13. Platysaccus radialis comb. nov. 12, PF2486. 13, PF2485.

Fig. 14. IUinites klausi sp. nov., PF2291.

Figs. 16-17. Klausipollenites schaubergeri Jansonius. 16, PF2266. 17, PF2264.

Localities of figs. 1, 2, 5, 6, 7, 10, 11, 14-17, Lower Permian Marl, Kimberley. Figs. 3, 8, 9, 13, Hilton
Plant Bed. Figs. 4, 12, Haughton Hall Boring, Lower Permian Marl, depth 1,095 feet.

Palaeontology, Vol. 8

PLATE 43

15

16

17

CLARKE, Permian miospores

R. F. A. CLARKE: PERMIAN SACCATE AND MONOSULCATE MIOSPORES 339

obscure the proximal surface features and it is felt that some species of this genus may
have been wrongly assigned to Platysaccus. It appears that neither Naumova (1937)
nor Potonie and Klaus (1954) consider Platysaccus as a striate genus, and on this
account alone a separation from Striatopodocarpites seems meaningful. However,
Jansonius (1962) assigns to Platysaccus cf. papilionis Potonie and Klaus 1954 forms said
to have from two to four striations on the proximal face (p. 54 and pi. 12, fig. 19).
I prefer to rate the presence or absence of striations as a generic character and to exclude
all striate forms from Platysaccus.

Comparison. Striatopodocarpites differs from Lueckisporites s.str. and Taeniaesporites
s.str. in the greater number of taeniae and the more distinctly diploxylonoid outline.
Rhizomaspora Wilson 1962 is characterized by having radiating or diverging ribs on the
spore body, and Striatopodocarpites differs from Striatoabietites emend. Hart 1964 in
the more circular spore body and the comparatively larger sacci.

Striatopodocarpites fusus (Balme and Hennelly) Potonie 1958
Plate 41, figs. 9-10

Discussion. The cardinal characteristics of this species are the very large sacci, the
multistriate spore body and the radially aligned reticulum of the sacci.

The taeniae, six to ten in number, are sometimes divided into irregular ‘blocks’ by
small splits perpendicular to the long axis of the grain. The fine saccus infra-reticulum
may be made to appear more coarse by corrosion (PI. 41, fig. 10).

Striatopodocarpites antiquus (Leschik) Potonie 1958
Plate 42, figs. 1-2

Remarks. A common feature of this species is the small lateral union of the sacci making
the pollen monosaccate. This also appears to be the case in the holotype. S. antiquus
differs from the type species in the greater irregularity of the taeniae and from S.
phaleratus (Balme and Hennelly) Hart 1964 in the absence of a distal groove bordered
by lips. S. balmei Sukh Dev 1959 is distinguished by its wider taeniae and a coarser
saccus reticulum.

Striatopodocarpites cancellatus (Balme and Hennelly) comb. nov.

Plate 43, fig. 2

1955 Lueckisporites cancellatus Balme and Hennelly, pi. 2, figs. 12-15.

1960 Striatites cancellatus (Balme and Hennelly) Hart, pi. 7, fig. 10.

Remarks. This species is characterized by the dark spore body, the presence of usually
six taeniae and the fine saccus infra-reticulum.

Genus striatoabietites Sedova emend. Hart 1964

1962 Illinites Kosanke; Orlowska-Zwolinska (pars).

1963 Striatites Pant; Schaarschmidt (pars).

1963 Strotersporites Wilson; Klaus (pars).

Type species. S. bricki Sedova 1956.

340

PALAEONTOLOGY, VOLUME 8

Discussion. Although valid since 1956 this genus has not been widely used in Western
palynological literature due mainly to the limited circulation of Sedova’s (1956) publica-
tion. This 1956 publication must be regarded as effective although of limited circulation.

Spores assignable to Striatoabietites have been described by several authors under
Lueckisporites and Striatites. Striatoabietites is characterized by the well-developed sacci
which are generally equal in width, although smaller in length, than the striate spore
body. The sacci form a distinct angle where they join the spore body, as seen in polar
view.

Comparison. The present genus differs from Striatopodocarpites in the saccus to spore-
body ratio and in being less distinctly diploxylonoid, while Lueckisporites s.str. and
Taeniaesporites s.str. have fewer taeniae (i.e. less than six).

Striatoabietites richteri (Klaus) Hart 1964
Plate 43, figs. 4-5

1955 Lueckisporites richteri Klaus, pi. 33, figs. 1-3.

1956 Taeniaesporites richteri Leschik, pi. 22, fig. 8.

1958 Striatites richteri Potonie, p. 51.

1962 Il/inites striatus Orlowska-Zwolinska, pi. 3, fig. 3.

1963 Striatites richteri Potonie; Schaarschmidt, pi. 13, figs. 21-22; pi. 14, figs. 1-2.

1963 Strotersporites jansonii Klaus, pi. 15, figs. 74-75; pi. 16, fig. 78.

1963 Strotersporites richteri Klaus, pi. 15, figs. 76-77.

This species, recorded from many Western European and North American localities,
is clearly circumscribed in the original description of Klaus (1955).

Genus vittatina Luber 1940? ex Potonie 1958
1963 Striatoluberae Hart.

Type species. V. subsaccata Samoilovich 1953.

Discussion. This genus is attributed to Luber (1940) by Samoilovich (1953) who gives
neither a type species nor a generic diagnosis but describes and figures three species of
Vittatina. The relevant literature (i.e. Luber 1940) is not quoted in Samoilovich’s
bibliography. I have not seen this work of Luber, nor does it appear to have been seen
by other authors subsequent to Samoilovich. The genus was thus invalid until Potonie
(1958) gave a short diagnosis and named one species ( Vittatina subsaccata Samoilovich
1953). Although this was not formally designated the type species by Potonie it may be

EXPLANATION OF PLATE 44
Magnification X 750 unless otherwise stated.

Figs. 1-2, 5-12. Taeniaesporites spp. 1-2, T. novimundi Jansonius. 1, PF2234. 2, PF2235. 5, T. albertae
Jansonius, PF2243. 6-10. T. labdacus Klaus. 6, PF2240. 7, PF2244. 8, PF2238. 9, PF2237. 10, PF2236.
11-12. T. angulistriatus comb. nov. 11, PF2245. 12, PF2247.

Figs. 3-4. Illinites delasaucei Grebe and Schweitzer. 3, PF2279. 4, PF2283.

Fig. 13. Potonieisporites novicus Bhardwaj, X 500, PF2203.

Fig. 14. Labiisporites granulatus Leschik, PF2270.

Figs. 15-16. Cycadopites rams sp. nov. 15, PF2292. 16, Distal view of holotype, PF2293.

Localities of figs. 5, 11, 12, Hilton Plant Bed. Remainder, Lower Permian Marl, Kimberley.

Palaeontology, Vol. 8

PLATE 44

CLARKE, Permian miospores

R. F. A. CLARKE: PERMIAN SACCATE AND MONOSULCATE MIOSPORES 341

considered as such since this was the first species used in a valid combination, and the
genus becomes valid from that date.

The type species is characterized by the possession of rudimentary sacci. Recently,
however, there has been some difference of opinion concerning the limits of the genus.
Wilson (1962, p. 24) emends the genus and while accepting V. subsaccata as the ‘geno-
type’ proposes (p. 25) that the genus be restricted to forms without sacci. Jansonius
(1962) makes a similar proposal independently, arguing that the terminal structures,
which may be considered as small sacci, might equally well be terminal extensions of
the equatorial rim. Vittatina is used here for bilateral, multistriate pollen grains which
may possess rudimentary sacci and where the distal (? germinal) furrow is at right
angles to the proximal taeniae.

Comparison. Vittatina differs from Protosaccu/ina emend. Jansonius 1962 in the sacci,
where present, being smaller and from Aumancisporites emend. Jansonius 1962 in the
absence of a transverse distal furrow with thickened margins. Alpern’s original descrip-
tion of Aumancisporites made no mention of a distal furrow, but on re-examination of
Alpern’s material Jansonius (1962) emended Aumancisporites to include forms which,
in other respects, were Vittatina-Mke, and which he considered to have a distal trans-
verse furrow bordered by lips. I think this feature is probably a fold or buckle produced
by squashing, and reject his emendation.

Hamiapollenites Wilson (February) 1962 (= Hamipollenites of Jansonius (April)
1962) differs from Vittatina in the larger size and different shape of the sacci. Vittatina
differs from the type of pollen previously assigned to Ephedripites and Welwitschiapites
primarily in the absence of a colpus or colpi disposed parallel to the striations.

Vittatina hiltonensis Chaloner and Clarke 1962

See Chaloner and Clarke 1962.

Infraturma disaccitrileti Leschik 1955
Genus illinites (Kosanke) emend. Potonie and Kremp 1954

1963 Limitisporites Leschik emend. Schaarschmidt.

1963 Jugasporites Leschik emend. Klaus.

1963 Limitisporites Leschik; Klaus.

Type species. I. unicus Kosanke.

Discussion. The emendation of Potonie and Kremp (1954) emphasizes the inequality
of the length of the arms of the tetrad scar, not evident from the original diagnosis of
Kosanke (1950). In some Permian bisaccate miospores there exists a gradation from
forms possessing a monolete mark ( Limitisporites Leschik) through forms showing
a ‘roof-shaped’ (= dachformig) split ( Jugasporites Leschik) to spores possessing
a triradiate mark, often with one arm developed to a greater or lesser extent than the
other two ( Illinites Kosanke). Such a gradation of the tetrad scar prompts Grebe (1957)
to consider the roof-shaped split as a retrograde or vestigial Y-mark. Such an argument
finds a good deal of support from the work of Potonie and Schweitzer (1960) in their
study of the pollen of Ullmannia frumentaria. Grebe and Schweitzer (1962) accordingly
include all the forms discussed above with variable arrangement of the tetrad scar in
Illinites Kosanke emend. Potonie and Kremp, and this procedure is followed here (see
also Klaus 1963, p. 270 footnote).

342

PALAEONTOLOGY, VOLUME 8

While it is certain that Illinites represents, at least in part, the pollen of Ullmannia
frumentaria it cannot be assumed that this is the only source plant of Illinites, and other
closely related or more distantly allied plants or plant groups may have produced spores
with a similar variety of the tetrad scar.

Illinites delasaucei (Potonie and Klaus) Grebe and Schweitzer 1962

Plate 44, figs. 3-4

1963 Limitisporites delasaucei Schaarschmidt, pi. 11, figs. 14-17.

1963 Jugasporites delasaucei Leschik; Klaus, pi. 6, fig. 19.

Remarks. The present forms are well covered by the clear descriptions given by Potonie
and Klaus (1954), Klaus (1955), Grebe (1957), Grebe and Schweitzer (1962), and
Schaarschmidt (1963). Leschik (1956, p. 132) observes that certain of his specimens
assigned to J. delasaucei have the sacci connected laterally by an exoexinal strip up
to 9/u. wide. The British specimens show this feature to be not greater than 5 /x wide and
in the majority of specimens such a feature is not seen.

Illinites tectus (Leschik 1956) comb. nov.

Plate 41, figs. 6-7; text-fig. 12

Remarks. This species shows a less variable tetrad scar than I. delasaucei, and further
differs in the presence of two ‘roughened areas’ (thickenings) on either side of the
tetrad scar (text-fig. 12).

Illinites klausi sp. nov.

Plate 40, fig. 12, Plate 43, fig. 14; text-fig. 13

Holotype. Plate 40, fig. 12. Slide PF2290. Sample K 14, Kimberley, Nottinghamshire; Upper Permian
(Lower Permian Marl = Marl Slate).

Diagnosis. Spore body circular, bearing a small triradiate mark. Sacci small, offlap
crescent shaped in polar view.

Description. The spore body is large and dark coloured with exine 2g thick. There exists
on the proximal face a large triangular area where the exine is thin and within which is
a Y-mark (text-fig. 13). The sacci are semicircular or less in outline and the offlap equals
the overlap; the greatest width of a saccus, as seen in polar view, is measured along the
distal attachment. The saccus infra-sculpture is a fine reticulum.

Dimensions. (Sixteen specimens.) Spore-body length 22(27)33 g, spore-body width
27(32)39 ju,, overall length 41(47)50^.

Comparison. The present species differs from I. parvus Klaus 1963 in the structure of the
proximal face of the spore body, and from other species of the genus in the shape of
the saccus offlap.

Infraturma disaccimonoletes Klaus 1963
Genus labiisporites Leschik emend. Klaus 1963

Type species L. granulatus Leschik 1956.

Comparison. Labiisporites differs from Illinites in the absence of a Y-mark and the

R. F. A. CLARKE: PERMIAN SACCATE AND MONOSULCATE MIOSPORES 343

a.

text-fig. 12. Il/inites tectus comb. nov. a. Polar view showing
proximal face, on the right, and distal face on the left
(drawn from specimen), x 1,000. b and c. Diagrammatic
reconstructions, b. Lateral polar section, c. Terminal polar
section.

text-fig. 13. Illinites klausi sp. nov. a, Polar view of proximal
face showing the small triradiate mark surrounded by a larger
triangular area (based on holotype), x 1,000. b and c. Diagram-
matic reconstructions, b. Lateral polar section, c. Terminal polar

section.

A a

C 3009

344

PALAEONTOLOGY, VOLUME 8

presence of a distal sulcus. Alisporites emend. Nilsson 1958 differs from Labiisporites
in having a distinct spore body, coarser saccus infra-reticulum and overall larger size.

Labiisporites granulatus Leschik 1956

Plate 41, fig. 5, Plate 44, fig. 14; text-fig. 14

Remarks. This species constitutes a small percentage of the microfloral assemblage in
most of the Permian samples studied. There is, however, less variation than observed
by Klaus (1963).

a.

c.

text-fig. 14. Labiisporites granulatus Leschik. a, Polar view
illustrating proximal face with an indistinct monolete mark,
on the right, and the distal face on the left (drawn from
specimen), X 1,000. b and c. Diagrammatic reconstructions.
b, Lateral polar section, c. Terminal polar section.

Infraturma disacciatrileti (Leschik 1955) Potonie 1958
Genus klausipollenites Jansonius 1962

1963 Falcisporites Leschik emend. Schaarschmidt (pars).

Type species. K. (al. Pityosporites) schaubergeri Potonie and Klaus 1954.

Discussion. Manum’s (1960) emendation of Pityosporites Seward 1914 excluded many
species assigned to this genus by Potonie and Klaus (1954). His suggestion that these
forms be included in Jugasporites Leschik 1956 found little favour and Jansonius
(1962) erected the genus Klausipollenites to accommodate certain of these miospores
excluded from Pityosporites by Manum’s emendation.

Although Klausipollenites is bisaccate there are specimens, which on other grounds
cannot be properly excluded from the genus, which show a more or less monosaccate
condition. In this case it is difficult to separate the genus from Vesicaspora Schemel
1951 (see Jizba 1962).

Comparison. Klausipollenites differs from Falcisporites emend. Klaus 1963 in the
elongated oval outline and the lack of a distal furrow and from Alisporites emend.

R. F. A. CLARKE: PERMIAN SACCATE AND MONOSULCATE MIOSPORES 345

Nilsson 1958 in the indistinct spore body of oblate outline and the absence of clear
saccus attachments.

Klaasipollenites schaubergeri (Potonie and Klaus) Jansonius 1962

Plate 43, figs. 16-17

1963 Falcisporites schaubergeri Schaarschmidt, pi. 15, figs. 10-17.

Remarks. This ubiquitous Zechstein form is adequately described in the original
description of Potonie and Klaus (1954).

Genus falcisporites Leschik emend. Klaus 1963
Type species. F. (al. Pityosporites) zapfei Potonie and Klaus 1954.

Discussion. Falcisporites emend. Klaus is characterized by a circular spore body,
a distal furrow, and sacci which are not connected laterally. As such the genus is clearly
differentiated from KlausipoUeniles but it becomes more difficult to separate Falcisporites
and Paravesicaspora Klaus 1963. This latter genus is said to be characterized by the
rhombic-shaped spore body, sacci which embrace laterally, and a distal furrow which
is constricted over the distal pole. In the present material and Upper Permian material
from Germany (kindly made available by Dr. H. Grebe), which I have examined, I find
the distinction between these genera hard to apply. The generally indistinct spore body
and the characteristic reticulate sculpture are common to both genera. The overall out-
line is always oval but the attachment of the sacci may vary between being terminal,
leaving a wide zone between the sacci attachments distally, or being more closely
attached, leaving a narrow distal zone within which a sulcus may be present. The
presence or absence of this last feature may depend on the maturity and preservation
of the grain, and the spore-body outline is often obscured by the sculpture. Unless
there is stratigraphical significance it may be better not to maintain the two genera.
Similarly the spores described by Balme and Hennelly (1955) as Florinites ovatus and
those by Leschik (1956) as Sulcatisporites may be better assigned to Falcisporites.

Falcisporites zapfei (Potonie and Klaus) Leschik 1956
Plate 40, figs. 10-1 1

1954 Pityosporites zapfei Potonie and Klaus, pi. 10, figs. 9-10.

Remarks. The variability of this species has already been remarked on (Grebe and
Schweitzer 1962, p. 15), and the species is clearly circumscribed in Potonie and Klaus
(1954).

Infraturma pinosacciti (Erdtman 1945) Potonie 1958
Genus alisporites Daugherty emend. Nilsson 1958

Type species. A. opii Daugherty 1941.

Discussion. Most workers agree that the original diagnosis of Daugherty is too broad.
Potonie and Kremp (1956a) restrict the genus. Nilsson’s (1958) formal emendation
(which seems to have been neglected) is not in conflict with this and the genus has been
widely accepted in this sense and is so used here, in preference to the much wider
concept adopted by Rouse (1959).

346 PALAEONTOLOGY, VOLUME 8

Alisporites differs from Falcisporites emend. Klaus in the more distinct prolate spore
body.

Alisporites nuthallensis sp. nov.

Plate 43, figs. 1,15; text-fig. 15

Holotype. Plate 43, fig. 15. Slide PF2277. Sample K 14, Kimberley, Nottinghamshire; Upper Permian
(Lower Permian Marl).

Diagnosis. Spore body elliptical where the width exceeds the length, dark coloured,
reticulate sculpture. Sacci well developed, joined laterally by a very thin exoexinal strip.
Sculpture of sacci medium infra-reticulate.

a.

text-fig. 15. Alisporites nuthallensis sp. nov. a. Polar view
of proximal face, on the right, and distal face on the left
(based on holotype), X 1,000. b and c, Diagrammatic
reconstructions, b. Lateral polar section, c. Terminal polar
section.

Description. The outline in polar view is generally oval where the greatest width of the
saccus feature is a line drawn over the distal pole at right angles to the long axis of the
grain. Occasionally the saccus width is greatest away from the spore body (PI. 43,
fig. 15); in this case the saccus shows a slight indentation at the spore body margin.
Distally there is a narrow leptoma delimited by straight or slightly convex saccus
attachments. The saccus infra-sculpture is a coarser variation of that seen on the
proximal spore body face. The saccus offlap is greater than the overlap.

Dimensions. (Eight specimens.) Spore-body length 30(32)35 g., spore-body width
36(42)44 p, overall length 65(70)74^.

R. F. A. CLARKE: PERMIAN SACCATE AND MONOSULCATE MIOSPORES 347

Discussion. Spore types 74 and 75 of Virkki (1945) appear from illustration and descrip-
tion to be very similar to A. nuthallensis sp. nov. As pointed out by Virkki (p. 134) there
is close resemblance between these miospores and some pollen of Caytonanthus described
by Harris (1937).

Infraturma podocarpoiditi Potonie, Thomson, and Thiergart 1950
Genus platysaccus (Naumova) ex Potonie and Klaus 1954

1955 Lueckisporites Potonie and Klaus; Balme and Hennelly, pi. 1, figs. 6-9.

1955 Cuneatisporites Leschik.

1962 Cuneatisporites Leschik; Bhardwaj, pi. 13, fig. 185.

Genolectotype. P. papilionis Potonie and Klaus 1954.

Discussion. This genus is distinguished by its distinct diploxylonoid outline and non-
striate spore body. Leschik (1955) erects the genus Cuneatisporites for diploxylonoid
forms with a ‘Keimspalte’ (germinal furrow). I cannot see this feature in the illustration
of the holotype, which appears to show merely the area of thinner exine between the
sacci attachments distally, a feature observed in many bisaccate miospores. Leschik
does not compare Cuneatisporites with Platysaccus and a separation upon the presence
of a germinal furrow does not seem justified. Fimbriaesporites Leschik 1959 differs from
Platysaccus in the presence of an exoexinal fringe around the spore body and a trans-
verse split (Leschik 1959, p. 72).

Platysaccus radialis (Leschik 1955) comb. nov.

Plate 43, figs. 12-13
1955 Cuneatisporites radialis Leschik, pi. 10, fig. 6.

Description. The spore body is oval where the width exceeds the length. The proximal
face is granular or scabrate. The sacci are discrete, three-quarters of a circle in outline,
and the offlap is greater than the overlap. One edge of a saccus is attached equatorially
while the other is attached near to the distal pole. The distal attachments are accom-
panied by crescent shaped thickenings (? folds) which usually extend to the equator.
A leptoma is present but no colpus is seen within this area. The saccus sculpture is
medium reticulate and a radial pattern is developed from the saccus roots.

Dimensions. (Seven specimens.) Spore-body length 20(25)31 ft, spore-body width
32(36)40 p, saccus width 42(45)50 ju,, overall length 56(64)77 /x.

Comparison. P. radialis comb. nov. differs from P. papilionis in the shape of the spore
body and comparatively smaller sacci, while P. umbrosus Leschik 1956 possesses an
irregular enveloping saccus. P. leschiki Hart 1960 is much larger than the present species.

Turma monocolpates Iversen and Troels-Smith 1950
Subturma intortes (Naumova) Potonie 1958
Genus cycadopites ( Wodehouse) ex Wilson and Webster 1946
Type species. C. follicularis Wilson and Webster 1946.

Comparison. Cycadopites differs from Monosulcites (Erdtman 1947, Cookson 1947)
ex Couper 1953 in the more fusiform outline of the latter, and the dumb-bell shape of
the furrow in the former.

348 PALAEONTOLOGY, VOLUME 8

Cycadopites ranis sp. nov.

Plate 44, figs. 15-16

Holotype. Plate 44, fig. 16. Slide PF2293. Sample K 14, Kimberley, Nottinghamshire; Upper Permian.

Diagnosis. Outline elliptical; exine thin; large well-defined irregular sulcus extending
to the equatorial margin; sculpture finely granular.

Description. The sharp ends of the elliptical outline are ‘squared off’ where the sulcus
is open at this point. The sharply pointed termination (PI. 44, fig. 15) is formed by the
overlap of one side of the sulcus. The sulcus is somewhat irregular and frequently does
not show a good dumb-bell shape although in grains where the sulcus is open a closure
over the distal polar region is present. Thin lips surround the sulcus which, in normally
orientated grains, is approximately one-third the width of the grain at its widest part.
Fine closely spaced granules form the sculptural pattern which is uniform over the
whole grain.

Dimensions. (Five specimens.) Overall length 66(82)108 p, overall width 28(32)40 p.

Comparison. C. rams sp. nov. is appreciably larger than any of the monosulcate species
described by Jansonius (1962). The closest comparison is made with Ginkgocycadophytus
sp. Samoilovich 1953 which is very similar.

AN EVALUATION OF THE MICROFLORAL ASSEMBLAGES

Text-fig. 16 shows the distribution and frequencies of the miospores in the two
principal sections studied. In all cases the counts are based on 300 spores in each
sample, and in each case the percentage determined is given. Those excluded from the
total are badly folded, broken, or corroded grains, detached air sacs together with some
alete bodies and a few triradiate spores. Of this latter category most are Carboniferous
in character and may be reworked; they are not abundant.

The most striking feature of the assemblages is their complete domination by bisaccate
forms of which the bisaccate Striatiti constitute the largest part. Lueckisporites virkkiae
is the dominant species in all the assemblages studied while Taeniaesporites noviaulensis
and T. labdacus are always present. Of the non-striate forms the most common species
are KlausipoIIenites schaubergeri, FaJcisporites zapfei, Labiisporites granulatus, and
lUinites delasaucei. Monosaccate species are represented by Perisaccus granulosus , P.
laciniatus, Potonieisporites novicus, Vestigisporites minutus, and Nuskoisporites dulhuntyi.
None of these monosaccate species is common but V. minutus occurs frequently in the
Hilton section but is curiously absent from the Kimberley section. Monosulcate forms,
represented by the single genus Cycadopites, constitute a small percentage in most of
the samples.

This similarity of the microfloral assemblages is taken to be the product of a uniform
vegetation existing during Upper Permian times, and although little can be deduced
from the spores concerning the type of vegetational cover, the apparent lack of local
masking effects of particular miospore species in individual samples or sections, suggests
that the pollen rain had become well mixed. This together with the general saccate
character of the grains and the uniformity of the assemblages, despite the varied

R. F. A. CLARKE: PERMIAN SACCATE AND MONOSULCATE MIOSPORES 349

PERCENT

DETERM-

INED.

67

82

77

76

77

63

z

78

UJ

b-

in

72

X

u

78

UJ

N

72

80

75

82

80

KIMBERLEY SECTION —

text-fig. 16. Miospore distribution in the Hilton and Kimberley sections.

lithology, suggests that the source vegetation was situated some distance from the area
of deposition.

The great increase of saccate forms and the small number of triradiate types in the
Permian, as opposed to the Carboniferous, expressed in terms of macrofloral changes,
has already been discussed (Clarke 1965a). The probable botanical affinities of other

350

PALAEONTOLOGY, VOLUME 8

spore types is also discussed in that paper. Vittatina is considered by some authors to
represent the Gnetales. There are, however, both botanical and stratigraphical objec-
tions to this and Vittatina is probably the pollen of some other Gymnosperm group.

COMPARISON OF THE MICROFLORA WITH OTHER PERMIAN

ASSEMBLAGES

The microflora of the Western European Upper Permian is well known. Potonie and
Klaus (1954), Klaus (1955, 1963), Leschik (1956), Grebe (1957), Orlowska-Zwolinska
(1962), Grebe and Schweitzer (1962), and Schaarschmidt (1963) all describe Upper
Permian assemblages. One striking feature of these assemblages is their similarity to
each other and to the present assemblages. The ubiquity of L. virkkiae, K. schaubergeri ,
I. delasaucei, F. zapfei, N. dulhuntyi, and S. richteri is maintained in all the areas, and
the frequencies, where plotted (Grebe 1957, Grebe and Schweitzer 1962, Schaarschmidt
1963, Klaus 1963) present a constant pattern suggesting a very uniform composition of
the vegetation existing in this region during Upper Permian (Zechstein) times.

Schaarschmidt (1963) separates the Zechstein from the Rotliegende of Germany by
the presence, in the former, of L. virkkiae, K. schaubergeri, Striatites sp., and S. richteri.
This distinctness of the European Upper Permian microflora, compared with Lower
Permian assemblages, is seen with reference to the Autunian of Germany (Remy and
Remy 1961) and France (Alpern 1958, Doubinger 1960, 1962). In Western Europe no
striate form with well-developed air bladders is recorded before the Upper Permian.
There is a singular lack of this pollen in the Autunian which contains a distinctive
Carboniferous element associated with an increase in the saccate genera Florinites,
Wilsonites, Potonieisporites, Nuskoisporites, Guthorlisporites, and Alisporites. The only
striate pollen occurring at this stratigraphic level are Vittatina-Wke forms ( Aumanci -
sporites, sensu Alpern).

Jansonius (1962) describes a microfloral assemblage of Upper Permian to Lower
Triassic age from Canada. As with the European Upper Permian the Canadian assem-
blages are dominated by bisaccate forms of which the bisaccate Striatiti again constitute
the most conspicuous part. Similarities extend to the specific level: S. richteri, T.
novimundi, T. noviaulensis, T. albertae, T. nubilus, and P. jacobii are common in Britain
and Canada. There are, however, important differences; L. virkkiae, K. schaubergeri
(although K. staplinii is similar), F. zapfei, I. delasaucei, and N. dulhuntyi are not recorded
from Western Canada. Further differences include the absence of Haniiapollenites
(= Hamipollenites) in the British material, and the association, in Canada, of the
above forms with Ovalipollis, a genus not recorded before the Upper Keuper in Britain.

Permian assemblages are described by Wilson (1962) from the Flowerpot Formation
(Guadalupean) and by Jizba (1962) from the American Mid-Continent area (Late
Pennsylvanian to Middle Permian in age). Although stratigraphically somewhat older,
these American assemblages show certain similarities with those of Britain. Wilson
(1962, p. 5) gives the relative percentages of L. virkkiae, Vittatina, Potonieisporites, and
Alisporites, all of which are present in the British material. However, Illinites, Taeniae-
sporites, Protohaploxypinus, and Klausipollenites are absent in the Flowerpot Formation.
The miospore assemblages described by Jizba (1962) from Kansas, Texas, and Okla-
homa have, like that from the Flowerpot Formation, in part a European flavour and
partly a distinctive element.

R. F. A. CLARKE: PERMIAN SACCATE AND MONOSULCATE MIOSPORES 351

The Lower Gondwana (Permo-Carboniferous) of Australia contains a variety of
bisaccate striate forms but these are associated with many Carboniferous-like trilete
and monolete spores (Dulhunty 1945, Balme and Hennelly 1955, 1956).

The Talchir stage assemblage in India (Virkki 1945, Potonie and Lele 1961) is very
similar to that of Australia but in the younger Raniganj stage (Bhardwaj 1962) con-
siderable diversity appears among the bisaccate Striatiti; Lueckisporites s.str. is, how-
ever, absent.

The Permian assemblages of Tanganyika (Hart 1960) and the Congo (Pierart 1959,
Hoeg and Bose 1960) are like the Australian Permian microflora.

A comparison with the Antarctic Permian material described by Schopf (1962) is
difficult as the preservation of his material is poor. Most of the specimens are assigned
to Accinctisporites although there is a record of Striatites ( Prolohaploxypinus ) (Schopf
1962, pi. 2, figs. 5, 12 a-b).

The Russian Permian contains a variety of bisaccate striate forms referable to the
genera Protohaploxypinus, Striatopodocarpites, Striatoabietites, and Striatosaccites.
Forms assignable to Vittatina are common and are associated with Florinites, Cyea-
dopites , bisaccate non-striate and monosaccate types. Trilete spores are present in the
Lower Permian.

Thus although bisaccate striate pollen are present in the Permian of all the areas
discussed above and although some genera ( Nuskoisporites , Cycadopites) occur in most
regions, the British Permian assemblages are most similar to those previously described
from Western Europe.

REFERENCES

alpern, b. 1958. Description de quelques microspores du Permo-Carbonifere Francais. Rev. Micro-
paleont. 1, 75-84.

balme, b. e. 1952. The principal microspores of the Permian coals of Collie. Geol. Surv. Bull. West.
Austr. 105, 164-201.

1959. Some palynological evidence bearing on the development of the Glossopteris flora. The

Evolution of Living Organisms., Symp. Roy. Soc. Victoria, pp. 269-80.

1963. Plant microfossils from the Lower Triassic of Western Australia. Palaeontology, 6, 12-40.

and hennelly, j. p. f. 1955. Bisaccate sporomorphs from Australian Permian coals. Austr. J.

Bot. 3, 89-98.

1956a. Monolete, Monocolpate and Alete sporomorphs from Australian Permian sediments.

Ibid. 4, 54-67.

19566. Trilete sporomorphs from Australian Permian sediments. Ibid. 4, 240-60.

bhardwaj, d. c. 1954. Einige neue Sporengattungen des Saarkarbons. Neues Jahrb. geol. Palaeont.
11, 512-25.

1962. The miospore genera in the coals of the Raniganj State (Upper Permian), India. The

Palaeobotanist, 9, 68-106.

and tiwari, r. s. 1964. The correlation of coal seams in Korba coalfield, Lower Gondwanas.

India. C.R. 5th Congr. Internat. Carbonif. 3, 1131-42.

and venkatachala, b. s. 1958. Microfloristic evidence on the boundary between the Carboni-
ferous and the Permian systems in Pfalz (West Germany). The Palaeobotanist, 6, 1-11.
chaloner, w. g. and clarke, r. f. a. 1962. A new British Permian spore. Palaeontology, 4, 648-52.
clarke, r. f. a. 1965a. Keuper Miospores from Worcestershire, England. Ibid. 8, 294-321.
doubinger, j. 1960. Associations sporologiques (microspores) de quelques gisements stephaniens et
autuniens. Proc. Comm. Coal. Petrol. 3, 165-71.

1962. Caracteres palynologiques du stephanien et de 1’Autunien. C.R. Acad. Sci. Paris, 1-3.

and rakotoarivelo, h. 1960. Etude palynologique de quelques echantillons de houille du bassin

de la Sakoa (Madagasca). Ibid. 1-3.

352

PALAEONTOLOGY, VOLUME 8

ghosh, a. k. and sen, j. 1948. A study of the microfossils and the correlation of some productive coal
seams of the Raniganj coalfield, Bengal, India. Trans. Min. geol. Metall. Inst. 43, 67-95.
goswami, s. K. 1951-2. Microfossils from coals from the South Rewa Gondwana basin. J. Sci. Res.
Banaras, Hindu Univ. 2, 189-99.

grebe, h. 1957. Zur Mikroflora des niederrheinischen Zechsteins. Geol. Jahrb. 73, 51-74.

and Schweitzer, h. j. 1962. Die Sporae Dispersae des niederrheinischen Zechsteins. Fortschr.

Geol. Rheinld. u. Westf. 1-24.

hart, G. F. 1960. Microfloral investigation of the Lower Coal Measures (K2); Ketewaka-Mchuchuma
coalfield, Tanganyika. Bull. Geol. Surv. Tanganyika, 30, 1-18.

1963. A probable pr e-Glossopteris microfloral assemblage from the Lower Karoo sediments.

5. Afr. J. Sci. 59, 135-46.

1964. A review of the classification and distribution of the Permian Miospore; Disaccate Striatiti.

C.R. 5 th Congr. Internat. Carbonif. 3, 1171-99.

in press. The microfloral succession of the coal bearing Karoo rocks (Lower Permian) from the

Mchuchuma River Valley, Ketewaka-Mchuchuma coalfield, Tanganyika. Bull. Geol. Surv. Tanga-
nyika.

hennelly, j. p. F. 1958. Spores and pollen from a Permian-Triassic transition, New South Wales.
Proc. Linn. Soc. N.S.W. 83, 363-9.

hoeg, o. A. and bose, m. n. 1960. The Glossopteris flora of the Belgian Congo, with a note on some
fossil plants from the Zambesi Basin (Mosambique). Ann. Mus. Roy. Congo Beige. 32, 1-106.
imgrund, r. 1960. Sporae Dispersae des Kaipingbeckens. Geol. Jahrb. 77, 143-204.
inosova, K. i. and Nesterenko, l. p. 1955. Spores et pollens du bassin du Donetz. Dokl. Akad. Nauk.
S.S.S.R. 100, 779-82. [In Russian.]

jansonius, j. 1962. Palynology of Permian and Triassic sediments; Peace River area, Western Canada.
Palaeontographica, B 110, 35—98.

jersey, n. j. de. 1946. Microspore types in some Queensland Permian coals. Univ. Qld. Dep. Geol. Pap.
3, 1-12.

jizba, k. M. m. 1962. Late Palaeozoic bisaccate pollen from the United States mid-continent area.
J. Paleont. 36, 871-87.

klaus, w. 1953«. Mikrosporen-stratigraphie der ost-alpinen Salzberge. Verb. geol. Bundesanstalt. 3,
161-75.

19536. Alpine Salzmikropalaontologie (Sporendiagnose). Palaont. Z. 21, 52-56.

1953 c. Alpine Salz-sporendiagnose. Z. deutsch. geol. Ges. 105, 234-6.

— 1955. Uber die Sporendiagnose des deutschen Zechsteinsalzes und des alpinen Salzgebirges.
Ibid. 105, 776-8.

1963. Sporen aus dem siidalpinen Perm. Jb. geol. B.A. 106, 229-363.

kosanke, r. m. 1950. Pennsylvanian spores of Illinois and their use in correlation. Illinois. State, geol.
Surv. Bull. 74, 1-128.

1959. Late Palaeozoic small spore floras of the United States. C.R. 9th Congr. Internat. Bot. 2,

200-1.

leschik, g. 1955. Die Keuperflora von Neuewelt bei Basel. 2; Die Iso und Mikrosporen. Schweizer.
Palaont. Abh. 72, 1-70.

1956. Sporen aus dem Salzton des Zechsteins von Neuhof (bei Fulda). Palaeontographica, B 100,

122-42.

1958. Sporenstratigraphie im Perm und der Trias. Z. deutsch. geol. Ges. 110, 13-14.

1959. Sporen aus den ‘Karru-Sandsteinen’ von Norronaub (Sudwest-Africa). Senck. leth. 40,

51-94.

luber, a. a. 1938. Spores and pollen from Permian coals of the U.S.S.R. Probl. Sowjet. geol. 8, 152-61.
[English summary.]

1939. The correlation by means of spores of coal bearing Upper Palaeozoic deposits of the

Kuznetsk and Minussinsk basins. Bull. Akad. Sci. U.S.S.R. 7, 88-104. [English summary.]

1941. Atlas of microspores and pollen grains of the Palaeozoic of the U.S.S.R. Tr. All-Union

geol. Sci. Res. Inst. ( V.S.E.G.E.I. ), 139, 1-107. [In Russian.]

1955. Atlas of spores and pollen of the Palaeozoic deposits of Kazakhstan. Tr. Akad. Nauk.

Kazakh. S.S.R., Alma-Ata, 1-126. [In Russian.]

R. F. A. CLARKE: PERMIAN SACCATE AND MONOSULCATE MIOSPORES 353

manum, s. 1960. On the genus Pityosporites Seward 1914. With a new description of Pityosporites
antarcticus Seward. Nytt. Magasin for Botanik. 8, 11-15.
martin, a. r. h. 1959. South African Palynological Studies 1. Grana Palynologica , 2, 40-68.
mhhta, k. r. 1944. Microfossils from a carbonaceous shale from the Pali Beds of the South Rewa
Gondwana Basin. Nat. Akad. Sci. India, 14, 125-41.
nilsson, t. 1958. Uber das Vorkommen eines mesozoischen Sapropelgesteins in Schonen. Lunds
Universitets Arsskrift., n.f., 2, 5-112.

orlowska-zwolinska, o. 1962. (A first finding of Zechstein sporomorphs in Poland.) Kwart. geol.
Polska, 6, 283-97. [In Polish.]

pant, d. d. 1949. On the occurrence of Pityosporites Seward in a Lower Gondwana Tillite from
Australia and its possible relationship with Glossopteris. Proc. 36th Ind. Sci. Congr. 4, 10-11.

- 1955. On two Disaccate spores from the Bacchus Marsh Tillite, Victoria (Australia) Ann. Mag.
Nat. Hist. London, 8, 757-64.

— and nautiyal, d. d. 1960. Some seeds and sporangia of the Glossopteris flora from the Raniganj
coalfield, India. Pa/aeontographica, B 107, 41-64.

I’ierart, p. 1959. Contribution a 1’etude des spores et pollen de la flore a Glossopteris contenue dans
les charbons de la Luena (Katanga). Mem. Acad. Roy. Sci. Coloniales, 8, 1-58.
potonie, r. 1956. Synopsis der Gattungen der Sporae Dispersae I Teil; Sporites. Beih. Geol. Jahrb.

23, 1-103. 1958. Idem. II Teil. Ibid. 31, 1-114. 1960. Idem. Ill Teil. Ibid. 39, 1-189.

1962. Synopsis der Sporae in situ. Ibid. 52, 1-204.

and klaus, w. 1954. Einige Sporengattungen des alpinen Salzgebirges. Geol. Jahrb. 68, 517-46.

and kremp, g. o. w. 1954. Die Gattungen der palaozoischen Sporae Dispersae und ihre Strati-

graphie. Geol. Jahrb. 69, 1 1 1-94.

1955. Die Sporae Dispersae des Ruhrkarbons, ihre Morphographie und Stratigraphie mit

Ausblicken auf Arten anderer Gebiete und Zeitabschnitte; Teil I. Palaeontographica, B 98, 1-136.
1956u. Idem. Teil II. Ibid. 99, 85-199. 19566. Idem. Teil III. Ibid. 100, 65-121.

• and lele, k. m. 1961. Studies in the Talchir flora in India. 1. Sporae Dispersae from the Talchir

beds of the South Rewa Gondwana basin. The Palaeobotanist, 8, 22-36.

and Schweitzer, H. j. 1960. Der Pollen von Ullmannia frumentaria. Palaont. Z. 34, 27-39.

remy, r. and remy, w. 1961. Beitrage zur Flora des Autunien IV. Monat. deutsch. Akad. Wiss. Berlin,
3, 489-502.

remy, w. 1961. Sporae Dispersae aus dem Stephanien der Halleschen Mulde, 1. Ibid. 3, 408-17.
rilett, m. h. p. 1954. Plant microfossils from the coal seams near Dannhauser, Natal. Trans, geol.
Surv. S. Af. 57, 27-38.

rouse, g. E. 1959. Plant microfossils from Kootenay coal measures strata of British Columbia. Micro-
paleontology, 5, 303-24.

samoilovich, s. r. 1953. Pollen and spores from the Permian deposits of the Cherdyn’ and Aktyubinsk
areas, Cis-Urals. Trav. Inst. Petrol-geol. S.S.S.R. 75, 5-57. (English translation Okla. geol. Surv.
Circ. 56, 1-103.)

schaarschmidt, f. 1 963a. Uber den Wandel der Sporenflora im deutschen Perm. Natur und Museum.
93, 231-6.

19636. Sporen und Hystrichosphaerideen aus dem Zechstein von Budingen in der Wetterau.

Palaeontographica, B 113, 38-91.

schemel, m. p. 1951. Small spores of the Mystic Coal of Iowa. Amer. Midi. Nat. 46, 743-50.
schopf, j. M. 1962. A preliminary report on plant remains and coal of the sedimentary section in the
Central Range of the Horlick Mountains, Antarctica. Inst. Polar Studies, 2, 1-61.
sedova, m. a. 1956. Four new genera and type species. In Materials for Palaeontology. New families
and genera. V.S.E.G.E.I. 12, 246-51. [In Russian.]

stoneley, h. m. m. 1957. Hiltonia, a new plant genus from the Upper Permian of England. Ann. Mag.
Nat. Hist. Lond. 9, 713-20.

1958. The Upper Permian flora of England. Bull. Brit. Museum (Nat. Hist.) geol. 3, 293-337.

sukh dev. 1961. The fossil flora of the Jabalpur Series — 3. Spores and pollen. The Palaeobotanist, 8,
43-56.

surange, k. R. 1958. Studies in the Glossopteris flora of India — 9. A male fructification bearing mono-
lete spores from the Lower Gondwana of India. Ibid. 6, 47-48.

354

PALAEONTOLOGY, VOLUME 8

surange, k. r., srivastava, p. n. and singh, p. 1953. Microfloral analysis of some Lower Gondwana
coal seams of West Bokaro, Bihar. Bull. Nat. Inst. Sci. India, 2, 111-27.
tchigouriaeva, a. a. 1949. Structure du pollen des Gnetales. Akad. Nauk. S.S.S.R. Dokl. 65, 555-7.
(French translation Grana Palvnologica, 1, 95-98.)

townrow, j. a. 1962. On some Disaccate pollen grains of Permian to Middle Jurassic age. Grana
Palynologica, 3, 1 3-44.

tschudy, r. h. and kosanke, r. m. 1959. Gnetaloid and vesiculate Permian sporomorphs. C.R. 9th
Congr. Internat. Bot. 2, 403-4.

van campo-duplan, m. 1947. Notes sur la disparition des Ballonnets des Abietinees. Trav. du Lab.
forest, de Toulouse. 1, 4, Art. 17.

1950. Recherches sur la phylogenie des Abietinees d’apres leurs grains de pollen. Ibid., 2, 4,

Art. 1.

— — - and gaussen, h. 1948. Sur quatre hybrides de genres chez les Abietinees. Ibid. 1, 4, Art. 24.
virkki, c. 1937. On the occurrence of winged spores in the Lower Gondwana rocks of India and
Australia. Proc. Nat. Acad. Sci. India, 6, 428-31.

1939. On the occurrence of similar spores in a Lower Gondwana tillite from Australia and Lower

Gondwana shales, India. Ibid. 9, 7-12.

1945. Spores from the Lower Gondwana of India and Australia. Ibid. 15, 93-176.

wilson, l. r. 1959a. Geological history of the Gnetales. Okla. geol. Notes, 19, 35-40.

19596. Plant microfossils from the Flowerpot shale (Permian) of Oklahoma. C.R. 9th Congr.

Internat. Bot. 2, 432.

— — 1962. Permian plant microfossils from the Flowerpot Formation, Greer County, Oklahoma.
Okla. geol. Surv. Circ. 49, 1-50.

and venkatachala, b. s. 1963. A morphologic study and emendation of Vesicaspora Schemel

1951. Okla. geol. Notes, 23, 142-9.

and webster, r. m. 1946. Plant microfossils from a Fort Union coal of Montana. Amer. J. Bot.

33, 271-8.

zauer, v. v. 1960. On Late Permian Floras from Solikamsk. Palaeontolog. J. 4, 114-24 [In Russian.]
zoritscheva, a. i. and sedova, m. a. 1954. Spore-pollen complexes from the Upper Permian deposits
from some localities in North European Russia. Rept. geol. Res. Inst., Moscow. 40 pp. [In Russian.]

R. F. A. CLARKE

Bataafse Internationale Petroleum Maatschappij N.V.,
Carel van Bylandlaan 30,

Manuscript received 11 March 1964 The Hague, Holland

APPENDIX

Numbers and lithologies of the better spore-bearing horizons.

Hilton Plant Bed. H44 Red/grey shaley sandstone. H41 Grey non-laminated, non-calcareous shale.
H40 Purple sandy non-calcareous shale. H23 Grey sandy, micaceous shale. Plant remains. H21 Grey
sandy, micaceous shale. Plant remains. H19 Sandy micaceous shale. HI 5 Sandy micaceous shale.
H13 Green/grey calcareous shale. Plant remains. H12 Green/grey calcareous shale. Plant remains. HI 1
Green/grey sandy shale. H9 Grey micaceous sandy shale. Plant remains. H5 Grey sandy shale. Plant
remains. H3 Grey/green shale. Plant remains. H2 Grey/green sandy shale. Lowest plant horizon.

Kimberley. K14 Calcareous, micaceous, medium-grained shaley sandstone. Plant remains. K13
Micaceous calcareous shale. Plant remains. K10, Kll Laminated calcareous shale. Plant remains.
K7 Yellowish/grey calcareous siltstone. Plant remains. K6, K4 Medium-grained calcareous shaley
sandstone. Plant remains. K3, K2 Light-grey, calcareous shale.

A NEW ORDOVICIAN CRINOID FROM
DOLGELLAU, NORTH WALES

by D. E. B. BATES

Abstract, locrinus brithdirensis sp. nov. from the Llanvirnian of Dolgellau (Merioneth) is described.

A fauna of trilobites, crinoids, and molluscs is listed in the Directory of British Fos-
siliferous Localities as occurring at Brithdir, Dolgellau. The reference given (Cox and
Lewis 1945, p. 80) mentions only Ogyginus corndensis , Conularia, and didymograptids
of the bifidus group. The horizon is thus lower Llanvirnian, and the succession is of
calcareous flags and ashy mudstones, resting on agglomeratic mudstones. Fourteen
specimens of a crinoid have recently been found.

Acknowledgements. The author wishes to thank Professor A. Wood and Dr. J. Haynes for reading
the manuscript. The type specimens have been presented to the British Museum (Natural History).

Class CRiNOiDEA J. S. Miller 1821
Subclass inadunata Wachsmuth and Springer 1881
Order disparata Moore and Laudon 1943
Family iocrinidae Moore and Laudon 1943
Genus iocrinus Flail 1866

locrinus brithdirensis sp. nov.

Plate 45

Diagnosis. A species of Iocrinus with a smooth conical calyx, the plates without inde-
pendent convexity, over twice as wide at the top as at the base and slightly wider than
high; basals as wide as high and three-quarters the height of the radials; radials slightly
wider than high, with an evenly concave facet occupying the full width of the upper
margin; primibrachs five to eight, just over twice as wide as high; secundibrachs ten to
eleven, tertibrachs eight to twelve, the more distal plates becoming as high as wide;
anal plate and proximal plates of the anal tube just higher than wide, more distal plates
becoming relatively higher and developing a sharp fold; stem pentagonal at the base
of the calyx, but becoming round distally, formed of columnals of varying height.

Holotype. BMNH E51710a-Z>, counterpart moulds; width of calyx 7-7 mm.; height of calyx 5T mm.
Paratypes. BMNH E51711o-6, counterpart moulds; width of calyx 7-5 mm.; height of calyx, 5-7 mm.
BMNH E51712a-g, counterpart slabs with several specimens.

Horizon and locality. Ogyginus corndensis beds, Llanvirnian, in banks of Nant Helygog, 1,660 yards
east by south from Gorwyr farm, 90 yards upstream from the sheep washery. Nat. Grid. Ref. SH/
795183.

Description. Cup just wider than high, though flattened in all specimens, cone-shaped,
the sides faintly convex in side view, the diameter at the top being just over twice that
at the junction with the stem; plates smooth and without independent convexity. Basals

[Palaeontology, Vol. 8, Part 2, 1965, pp. 355-7, pi. 45.]

356

PALAEONTOLOGY, VOLUME 8

three-quarters the height of the radials, pentagonal, as high as wide. Radials slightly
wider than high, widening slightly upwards, facet evenly concave and occupying the
full width of the upper margin.

Apparently five to eight primibrachs, just over twice as wide as high, subcircular in
cross-section, bearing a sharp-edged food groove containing small ambulacral plates.
Ten to eleven secundibrachs, eight to twelve tertibrachs. Arms branch five to six times,
the distal portions narrow with a sharp ridge running up the lower (dorsal) side, more
than fourteen times the length of the calyx.

Superradianal plate pentagonal, wider than high, bearing on its left side the anal
tube, on its right the right posterior arm, with five to six primibrachs. Proximal plates
of the anal tube (including the anal plate) just wider than high, becoming higher than
wide and developing a sharp ridge on both their anterior and posterior sides. The more
distal side plates of the tube are sharply folded transversely, concave with raised sutures
between them, two for every central plate. Entire anal tube over ten times the height
of the calyx.

Tegmen not seen.

Stem formed of columnals of varying height, sharply pentagonal in cross-section at
the base of the calyx, becoming round and decreasing in diameter away from the calyx.

Discussion. The new species differs from /. shelvensis Ramsbottom in having the radials
as wide as high, and without a sigmoidal shaped re-entrant on the facet; in having the
basals three-quarters the height of the radials; and probably in having more secundi-
brachs. I. whitteryi Ramsbottom has the brachials, even in the distal parts of the arms,
substantially wider than high, and has only four secundibrachs (in the one specimen
seen by Ramsbottom).

About fourteen fairly complete specimens were collected, enabling a few statistical
estimates of proportions to be made (Table I). The calyces are variably flattened in most
specimens, so that, although the height and width of the calyx are highly significantly
correlated, there is no guarantee that the proportions would be the same in unflattened
specimens. Similarly the relation between the width of the calyx and the width of the
stem immediately beneath it may be altered.

Crinoids are not common members of the British Ordovician faunas: previous to
RamsbottonTs monograph (1961) only five species had been described, to which he
added a further seventeen. This is due to several factors. As Ramsbottom pointed out
(op. cit., p. 31) the absence of crinoid workers is probably a major factor, but crinoids,
especially the diagnostic calyces, are undoubtedly rare. This is in part due to preserva-
tion, but also to lack of favourable facies. Palaeoecological observations on these

EXPLANATION OF PLATE 45

locrinus brithdirensis sp. nov. from the Llanvirnian of Brithdir, Dolgellau, North Wales. All photo-
graphs are of latex casts, whitened with ammonium chloride. None has been retouched.

Fig. 1. Anterior view of holotype, BMNH E51710u, x9.

Fig. 2. Posterior view of holotype, BMNH E517106, X 8.

Fig. 3. Detail of the food grooves of the holotype, X 4-6.

Fig. 4. Posterior view of the anal tube of the holotype, X 1-7.

Fig. 5. Posterior view of paratype, BMNH E51711n, X 1-2.

Fig. 6. Anterior view of the calyx of the holotype, X 2-2.

Palaeontology , Vol. 8

PLATE 45

BATES, Ordovician crinoid

D. E. B. BATES: A NEW ORDOVICIAN CRINOID FROM WALES

357

TABLE 1

a. Statistics of width (w) and height (h) in thirteen specimens of locrinus brithdirensis sp. nov.

w mm. (var. w) 6-62 (0.479) a (var. a) 0-911 (0.0328)

hmm. (var. h) 5-23 (0.397) b (var. b) 0-70 (1-454)

r 0-725

b. Statistics of width of calyx (w) and width of stem (s) at the base of the calyx in eleven specimens
of I. brithdirensis.

w mm. (var w) 6-71 (0-483)

s mm. (var. s) 2-51 (0-122)

r 0-601

c. Statistics of width (w) and height (h) of radials in fourteen specimens of I. brithdirensis.

w mm. (var. w) 3-20 (0-191) a (var. a) 0-422 (0-00565)

hmm. (var. h) 3-09 (0-034) b (var. b) 1-738 (0 05904)

r 0-766

crinoids are few, and indeed little is known in general about the palaeoecology of the
group (Laudon 1957, p. 961).

The Brithdir crinoids occur in an ashy micaceous mudstone, confined to only one
bedding plane in the 20 feet of strata exposed at the locality. On the bedding plane they
are abundant, and mainly complete, except for the extremities of some of the arms and
the lower parts of the stems. They are not in the position of life, but most specimens are
laid on their sides, with the arms and anal tube spread in a fan enclosing about 45°
to 70°. In a few cases the arms have been spread out in a 360° circle. The disposition
of the arms in some of the slabs suggests current orientation.

Comparison of the species of locrinus suggests that there are two distinct geographical
groups within the genus. The American species, I. crassus , I. similis, I. subcrassus, and
/. torontoensis all have a relatively wide cup (in which the basals are less than half the
height of the radials), both sets of plates have vertical ridges on them, and the stem at
its junction with the calyx is half the diameter of the latter. In contrast the British species,
I. shelvensis, I. whiteryi, and I. brithdirensis, have smooth basal and radial plates, with-
out strongly marked ridges, the calyx is cone shaped, with the stem at its base less than
half the diameter of the calyx, and the stem is altogether more slender. I. cambriensis
does not correspond closely to either of the species groups, and indeed its generic
position is in doubt (Ramsbottom 1961, p. 6).

REFERENCES

cox, a. h. and lewis, h. p. 1945. Summer field meeting 1944. The Dolgelley district. Proc. Geol. Ass.
Lond. 56, 59-81.

laudon, l. r. 1957. Crinoids. In Treatise on Marine Ecology and Paleoecologv. 2, Paleoecology, ed.

H. S. Ladd, 961-72. Mem. Geol. Soc. Amer. 67.

Palaeontographical Society. 1954. Directory of British Fossiliferous Localities. London.
ramsbottom, w. h. c. 1961. A monograph of British Ordovician Crinoidea. Palaeontogr. Soc.
( Monogr .), 1-37, pi. 1-8.

D. E. B. BATES

Department of Geology,
University College of Wales,
Alexandra Road,

Manuscript received 14 August 1964 Aberystwyth

UNUSUAL STRUCTURES IN DEVONIAN
ATRYPIDAE FROM ENGLAND

by PAUL COPPER

Abstract. A distinctive group of Middle Devonian atrypoid brachiopods (Subfamily Palaferellinae Spriesters-
bach 1942), with cemented muscle platforms and remarkable apical structures and crura, is described. This
group, forming part of the newly established genus Mimatrypa Struve 1964, is common in the thick stromato-
poroid-crinoid reefs of Givetian age in north-western Europe. Type material of J. de C. Sowerby, Phillips,
Davidson, and Whidborne is re-examined and revised. The affinity of Mimatrypa desquamata (Sowerby), from
the Devonian of south-western England, to the genus Desquamatia Alekseeva 1960 is refuted. The Eifelian
Gruenewaldtia latilinguis (Schnur) is sectioned in detail. ‘ Karpinskia ’ rhenana Leidhold is removed from the
Atrypida.

The atrypoid brachiopod collections from Devon, on which J. de C. Sowerby (1840),
Phillips (1841), Davidson (1864-84), and Whidborne (1893) based their early descrip-
tions, were re-examined and found to be in need of thorough revision. This paper forms
a small portion of the work completed in this revision, and was based on the material
housed in the British Museum (Natural History), London, the Sedgwick Museum,
Cambridge, and the Geological Survey Museum, London, as well as on the writer’s
own collections from Devon. Through the courtesy of Dr. Hermann Jaeger, comparative
German material was examined in the Museum fur Palaontologie, Humboldt Universi-
tat, Berlin.

Three common ‘related’ Middle Devonian brachiopods are treated. One of these,
? Lycophoria rhenana (Leidhold), frequently described as an atrypid, is removed from the
order Atrypida and placed in the Pentamerida. Its true affinities still remain obscure.
Most of the emphasis in this study is placed on 'Atrypa' desquamata Sowerby (now in the
genus Mimatrypa Struve 1964), which was investigated in detail in view of the paper
by Alekseeva (1960). In her paper, Alekseeva established the subgenus Desquamatia
(type species Atrypa ( Desquamatia ) khavae Alekseeva 1960) for a group of atrypids said
to be similar to ‘ Atrypa ’ desquamata Sowerby. A careful examination of Sowerby’s
species revealed, however, that it had internal structures quite distinct from those
present in the Russian form, and which linked it to the atrypoid subfamily Palaferellinae
Spriestersbach 1942 (emend. Struve 1955, p. 211). Struve (1964) recently established the
genus Mimatrypa , in which the most important internal structures are cemented
muscle platforms, very similar to those present in ‘ Atrypa ’ desquamata Sowerby.
The third form to be discussed is a species of Gruenewaldtia Chernyshev 1885, for which
detailed serial sections (based on acetate peels) are given for the first time.

Three major structures, hitherto not well known in the atrypoid group of brachiopods,
are described and illustrated. These include cemented muscle platforms, medially
fused (not contiguous) deltidial plates continuing into pedicle collars, and minutely
fibrous crura, which are almost vestigial and differ structurally from those present in
normal examples of Atrypa Dalman 1828. A curious type of inner protrusion of deltidial
plates and an anomalous type of jugal process are figured.

[Palaeontology, Vol. 8, Part 2, 1965, pp. 358-73, pi. 46-47.]

PAUL COPPER: UNUSUAL STRUCTURES IN DEVONIAN ATRYPIDAE 359

The abbreviations used in the text and explanations to plates and figures are as
follows: BM, British Museum (Natural History); GSM, Geological Survey Museum;
Geol. Soc. Coll., Geological Society collections (housed in the GSM); SM, Sedgwick
Museum. The morphological terms common to the atrypoid brachiopods are described
and figured in Siehl (1962) and Struve (1955).

text-fig. 1. Idealized stratigraphic table for Middle Devonian sections of south-west England. Ply-
mouth section from Taylor 1951 ; Chercombe Bridge section from Middleton 1954; Lummaton section
from Jukes-Brown 1906; Wolborough and Barton sections were estimated and have not been measured;

Eifel succession from Hotz et al. 1955 and Struve 1961.

Stratigraphy. A generalized Devonian succession for Devon, based chiefly on published
research, is given in text-fig. 1. The correlations agree substantially with those of Elliott
( 1961) and House (1963). It is thought that some of the sub-stages defined for the Give-
tian of the Eifel region in Germany (but not yet internationally accepted) can be recog-
nized in the sections at Plymouth and Chercombe Bridge. The Givetian is arbitrarily
divided into lower, middle, and upper divisions based on the succession in the Eifel
(Hotz et al. 1955; Struve 1961). In the Eifel, the upper Givetian-middle Givetian boun-
dary is doubtfully placed below the Kerpen beds: the base of the Bolsdorf beds, which
are characterized by thick stromatoporoid reefs and detrital limestones in the Eifel,
would probably form a more natural lithological boundary. Further work is necessary
to make such divisions satisfactory, and to revise them.

The material examined comes mainly from three fossiliferous localities in south
Devon. These are Lummaton Hill and Barton (in Torquay) and Wolborough (south-

C 3009 B b

360

PALAEONTOLOGY, VOLUME 8

west of Newton Abbot). Sowerby’s material was stated to have come from Plymouth,
and was probably collected in part at Mount Wise, where fossiliferous upper Givetian
rocks are known to occur (Taylor 1951, p. 154). Whidborne’s atrypids were mainly from
the Lummaton quarries, while Davidson appears to have based his descriptions largely
on Wolborough material collected by W. Vicary.

The Lummaton quarries were described in detail by Jukes-Browne (1906), who was
able to establish a fairly complete succession. This was reviewed by Elliott (1961,
pp. 256-8), who confirmed the age of the Lummaton shell bed as Givetian, a view with
which the writer concurs. In the shell bed Elliott found several specimens of Spinatrypa
trigoneUa (Davidson), a very similar form of which also occurs at Buchel (near Herren-
strunden, Bergisches Land, Germany) in beds known to be of Givetian age. Strati-
graphically directly below the shell bed large specimens of Mimatrypa desquamcita
(Sowerby), now known to be an upper Givetian fossil, are commonly found.

The Barton quarry is not well known. It is thought that only a small portion of the
Givetian, mainly upper, is present. Typical examples of Mimatrypa desquamata
(Sowerby) are not uncommon, but the globose, flattish, and coarsely ribbed varieties
are apparently absent.

Most of the well-preserved atrypids come from the Wolborough quarry, which is
unfortunately no longer accessible. Collections from this quarry contain a large number
of Mimatrypa desquamata (Sowerby), small specimens of which tend to resemble M.
insquamosa (Schnur) of the Eifel. There are also a few coarsely ribbed specimens which
are similar to, but far from identical, with M. flabellata (C. F. Roemer). M. insquamosa
and M. flabellata are of lower Givetian age, and the presence of intermediate forms in
the Wolborough quarry would suggest that more than merely the upper Givetian is
present there.

SYSTEMATIC DESCRIPTIONS

Order Atrypida
Superfamily Atrypacea
Family Atrypidae

Subfamily Palaferellinae Spriestersbach 1942

The classification of the atrypoid brachiopods is a controversial matter, and several
differing systems have recently been proposed. The elevation of the atrypids to the order
level, as separately adopted by Alekseeva (1962), Ivanova (1962), and Tiasheva (1962),
is supported. There is strong evidence that the atrypoids as a group are more closely
related to rhynchonellid than spiriferid brachiopods. It is rather striking that in some
forms of atrypids (e.g. Mimatrypa flabellata (C. F. Roemer) and M. insquamosa (Schnur))
detailed examination of hundreds of specimens has failed to reveal the presence inter-
nally of calcified spiral cones. A similar phenomenon is apparently present in more
primitive atrypoid genera such as Eocoelia, Anabaia, Leptocoelia, and Australocoelia
(Boucot et al. 1964, p. 807). The presence of a spiral brachidium, at present the only
ground for classifying atrypoids in the order Spiriferida, does not necessarily indicate
the same descent.

The palaferellinid brachiopods are given the more conservative subfamily status of
Struve (1955, p. 211) chiefly, however, because their origins are unknown and our

PAUL COPPER: UNUSUAL STRUCTURES IN DEVONIAN ATRYPIDAE 361

knowledge of the group is scant. The direct relationship between synchronous Atrypa
and Mimatrypa is probably purely superficial and, phylogenetically, the discovery of
Mimatrypa in no way strengthens the relationship between the divergent stocks of the
Atrypinae and Palaferellinae (as argued in Struve 1964, pp. 433-4). If Mimatrypa were
to form an evolutionary intermediate between Atrypa and Gruenewaldtia it would have
had to occur in Lower Devonian rocks rather than upper Middle Devonian rocks. At
present it appears more likely that Mimatrypa was an end-member of the divergent
palaferellinid stock.

Rzhonsnitskaya (19606), in the Russian brachiopod ‘Treatise’ (ed. T. G. Sarycheva)
divided the Atrypidae into seven subfamilies, four of which are the Atrypinae, Pala-
ferellinae, Carinatininae, and Karpinskiinae. Boucot et al. (1964) proposed six sub-
families, including the above four, and regrouped some of the genera. At almost the
same time Struve (1964), mainly on the basis of similarity in external sculpture, reduced
the total number of subfamilies to four by including the genera Carinatina Nalivkin 1 930
and Karpinskia Chernyshev 1885 (type genera of two subfamilies) in the Palaferellinae.
This last classification appears to be the most satisfactory. Studies have hitherto been
hampered mainly by the lack of knowledge of shell structure (particularly umbonal
structures which are valuable in atrypoid taxonomy) and it is clear that the classification
of the family Atrypidae is still in an unsettled state.

The palaferellinids, typified by muscle platforms in both valves, include Gruenewaldtia
Chernyshev 1885 (= Palaferella Spriestersbach 1942) and Mimatrypa Struve 1964
(= Desatrypa Copper 1964). Other genera are at present only doubtfully included:
Vagrania Alekseeva 1959 (? ancestral to Carinatina Nalivkin 1930), Karpinskia Cherny-
shev 1885 (? alternative ancestor to Mimatrypa Struve), Carinatina Nalivkin 1930,
and Anatrypa Nalivkin 1941 (possibly derived from Mimatrypa desquama ta (Sowerby)).
Not enough is known about Pseudogruenewaldtia Rzhonsnitskaya 1960a and Na/ivkinia
Bublichenko 1928 either to include or exclude them from the Palaferellinae. Ivanova
(1962) favours their inclusion.

Genus Mimatrypa Struve 1964

Type species. Terebratula prisca var. flabellata C. F. Roemer 1844, p. 66, pi. 5, figs. 4 a, b.

Diagnosis. Biconvex atrypids with even, continuous palmate ribs, few incipient marginal
growth lamellae, a well-exposed inter-area and delthyrium, and ‘cemented’ muscle
platforms in both valves.

Comparison. The genus is distinguished from Gruenewaldtia Chernyshev by its cemented
rather than septally elevated muscle platforms, its vestigial crura, and in part also by
its medially fused deltidial plates and lack of lateral cavities. Vagrania Alekseeva, which
has a similar development of muscle platforms, is otherwise distinct in having a different
rib structure. Adequate comparisons with Anatrypa Nalivkin cannot be made until
inter-specific relationships in this genus are clarified. For example, the type of rib struc-
ture figured for Anatrypa micans (Buch) in Rzhonsnitskaya (19606, pi. 55, figs, la-d)
does not coincide with that figured for A. kadzielniae (Giirich) and A. timanica Mar-
kovsky in Alekseeva (1962, pi. 8, figs. 6, 7). Mimatrypa in part bears some resemblance
to certain species of the genus Desquamatia Alekseeva; internally these two genera are
quite different. Mimatrypa Struve 1964 (3. 8. 1964) is thought to be synonymous with

362

PALAEONTOLOGY, VOLUME 8

Desatrypa Copper 1964 (24. 9. 1964). Externally, specimens of the lower Givetian
Mimatrypa insquamosa (Schnur) are almost indistinguishable from small specimens of
the upper Givetian M. desquamata (Sowerby). Internally, both possess structures which
are not present or are rare in other atrypid genera : medially fused, not contiguous, deltidial
plates; massive dental plates without lateral cavities; thick cardinal blocks and hinge
plates; and vestigial, minutely fibrous, short crura. The type species of Mimatrypa ( M .
flabellata (C. F. Roemer)) differs in having a slightly better developed muscle platform
in the brachial valve.

(b)

text-fig. 2. Sketch diagrams showing the muscle platforms in Mimatrypa desquamata (Sowerby).

Natural size.

Description. The typical features are a biconvex to dorsi-biconvex shell with large
interarea, and prominent delthyrium containing two partially or wholly fused deltidial
plates and apical foramen. Ribs are deep-troughed and round-crested (i.e. palmate),
and are interrupted marginally by a few incipient growth lamellae. Spines are absent
and frills are not likely to have been developed. A fine microscopic concentric ornament
is preserved in rib troughs.

Internally, muscle platforms are weakly developed in the late Givetian species and
somewhat more elevated in early Givetian species. A single unified muscle platform in
the pedicle valve and, in the brachial valve, a platform divided by a bifurcating median
septum, are illustrated in text-fig. 2.

A pedicle collar is well developed in at least one species ( M . insquamosa (Schnur)),
and, where present, forms a unified structure with the deltidial plates (compare text-
figs. 3 and 8). This feature does not seem to have been previously described: pedicle
collars are common to many atrypids, but in other genera they appear always to have
been separated from deltidial plates. In Mimatrypa lacking pedicle collars, the deltidial
plates are turned inwards to form what may be either an incipient or vestigial pedicle

PAUL COPPER: UNUSUAL STRUCTURES IN DEVONIAN ATRYPIDAE 363

collar, an anomaly also common to Gruenewaldtia (text-figs. 7, 8). Deltidial plates are
wholly or partially fused medially, i.e. no suture-line is visible.

Teeth are massive, usually bilobed; dental plates are straight or zigzag in transverse
section. No dental ridges or lateral cavities are known.

The cardinal process is weakly developed and overlaps the hinge plates. A thick
cardinal block and massive hinge plates are typical.

text-fig. 3. Transverse serial sections of Mimatrypa insquamosa (Schnur), lower Givetian, Baarley,
Gerolstein syncline, Germany. Fused deltidial plates are typical of the genus, a, Pedicle collar freed
from the shell wall; b, pedicle collar relatively short and not as well developed. x3-6.

Crura are extremely weak, probably strongly degenerated structures; their curious
development and mode of attachment to the hinge plates is not known in other atrypids
at present (see text-fig. 5). In Mimatrypa insquamosa (Schnur) the normal connecting
spiralia are absent. The crura of Mimatrypa are comparable with the rather short crura
of some forms of Gruenewaldtia.

Primary lamellae, spiralia, and jugal processes may be completely absent in Mimatrypa
insquamosa (Schnur) and M. flabellata (C. F. Roemer), since an exhaustive search in
hundreds of specimens has failed to reveal them. Quenstedt apparently also encountered
similar difficulties for he states (1885, p. 702): ‘Ich habe zwar die Spirallamellen an ihr
nicht finden konnen, doch werden sie wohl nicht fehlen. ’ The apparent absence of spi-
ralia may be related to the presence of vestigial crura.

Distribution. At present Mimatrypa seems to be largely restricted to western Europe.
Breivel (1959, pp. 54, 55) appears to have found it in the Ural Mountains of the U.S.S.R.

364

PALAEONTOLOGY, VOLUME 8

cardinal process

text-fig. 4. Transverse serial sections of hinge-plates and cardinal processes in a, Mimatrypa insquamosa
(Schnur), Baarley, Gerolstein, and b, M. desqucimata (Sowerby), Lummaton, Devon. x9.

b.

text-fig. 5. Transverse serial sections of crural development in a, Mimatrypa insquamosa (Schnur),

Baarley, Gerolstein, and b, M. desquamata (Sowerby), Lummaton, Devon. X 9.

Range. Middle Devonian: upper Eifelian to upper Givetian.

Member species. Terebratula prisca var. flabellata Roemer 1844, p. 66, pi. 5, figs. 4a, b ;
Terebratula insquamosa Schnur 1853, p. 182, pi. 24, figs. 5a, b; Atrypa julii Gortani 1911,
pp. 159-60, pi. 17, figs. 9-11.

Mimatrypa desquamata (Sowerby)

Plate 46, figs. 1-11; Plate 47, figs. 7-10, 13-16; text-fig. 7

1840 Atrypa desquamata Sowerby, explan, to pi. 56, figs. 19-22.

1841 Terebratula ( Atrypa ) desquamata Sowerby; Phillips, p. 82, pi. 33, figs. 146a-//.

PAUL COPPER: UNUSUAL STRUCTURES IN DEVONIAN ATRYP1DAE 365

1 864-65 Atrypa desquamata Sowerby ; Davidson, pp. 58-59, pi. 10, figs. 9-12, pi. 11, figs. 1 , 3-6.
1882-4 Atrypa desquamata Sowerby; Davidson, pp. 39-40, pi. 1, figs. 15, 156.

1885 Atrypa reticularis var. desquamata Linne; Maurer, pp. 181-2, pi. 7, fig. 33.

1893 Atrypa desquamata Sowerby; Whidborne, p. 117, pi. 13, fig. 13.

1928 Atrypa desquamata Sowerby; Leidhold, pp. 96-97.

Atrypa circularis Leidhold, pp. 97-98, pi. 6, fig. 5; pi. 7, fig. 2.

Atrypa globosa Leidhold, pp. 94-95, pi. 6, fig. 4.

1934 Atrypa desquamata Sowerby; Torley, p. 123, pi. 9, fig. 73.

Atrypa circularis Leidhold; Torley, p. 124, pi. 9, fig. 74.

Atrypa globosa Leidhold; Torley, p. 124.

Type locality. This is designated as Plymouth by Sowerby. The stratigraphic horizon and precise locality
are unknown. The species occurs most abundantly at the north-west end of the Lummaton Hill
quarries, Torquay, Devon.

Material. Ninety-seven specimens from Wolborough, Plymouth, Barton, and Lummaton in Devon.
Range. Upper Givetian, Middle Devonian. The species is probably a good index fossil of the upper
Givetian.

Source sediment. This is characteristically a massive to thickly bedded, light grey to
whitish, stromatoporoid-crinoid detrital limestone with abundant large brachiopods,
many of which occur in ‘nests’. A turbulent, high-energy biohermal environment is
indicated. The low argillaceous content of the limestones suggests clear, well-aerated
waters.

Associated fauna. Stringocephalus cf. burtini DeFrance, Uncites gryphus (Schlotheim),
Hypothyridina cuboides (J. de C. Sowerby). Most brachiopods occur in patches. Horn
corals are rare; tabulate corals and colonial corals common. There is abundant stroma-
toporoid and crinoid debris. The fauna is probably largely indigenous. Mimatrypa
desquamata (Sowerby) has not yet been found in argillaceous sediments of the same
age.

Diagnosis. Large, broad, biconvex, somewhat planar, subcircular palaferellinids with a
prominent, blunt, protruding beak and moderately sized foramen. Internally, the main
pallial sinuses are unforked. Spiralia are present. Pedicle collars not observed.

External morphology (terminology after Siehl (1962) and Struve (1955)). The species
ranks among the largest known atrypids: the maximum size is about 65 mm., and the
average mature size is between 40 and 50 mm. (text-fig. 6). The subcircular shell outline
is truncated posteriorly by an obtusely angled hinge-line. Maximum width and maximum
depth occur about mid-length. A broad weak fold is developed on the anterior com-
missure; globose specimens have a more prominent fold (as, for example, in Davidson
1864-5, pi. 11, fig. 1).

The pedicle valve is less convex than the brachial: it is somewhat planar, with a weakly
and broadly convex mid-area (transl. German ‘Mittelfeld’) and flattened margins. The
beak is blunt, orthocline (never incurved); it projects 3-4 mm. over the dorsal apex.
The shoulder angle (‘Schulterwinkel’) is about 140 degrees but may be more acute.
The shoulder line (‘ Schulterlinie ’) is only weakly indented. Hinge corners (‘ Schlossecke ’)
are well rounded. The edge of the interarea (‘ Areakante’) is sharp-edged apically but
rounded and poorly defined laterally. The delthyrium is rather narrow and acutely
angled (70 degrees at the apex). The apical hypothyrid foramen, rarely expanded into
the ventral umbo, is 1 to 2 mm. across. Deltidial plates are crenulated.

366

PALAEONTOLOGY, VOLUME 8

The brachial valve is slightly more convex, and is well rounded. The dorsal apex is
not covered by the deltidial plates of the opposing valve. A weak median furrow occurs
apically in a few specimens.

Internal shell markings. The shape and size of the muscle platforms can be seen in text-
fig. 2 and Plate 46, figs. 1,2,4. Muscle platforms are striated longitudinally. In the pedicle
valve two long, but rather weak, pallial sinuses flank the muscle platform and extend
roughly parallel towards the anterior margin, forming a shallow trough anterior to the

text-fig. 6. Scatter diagrams showing the size relationships in Mimatrypa desquamata (Sowerby).
Data combined from localities at Lummaton, Barton, and Wolborough.

platform. Tributary pallial arteries are extremely weak (PI. 46, fig. 3). The small neph-
ritic adductor scars common to other atrypid genera are not present. Internal markings
are well illustrated by Leidhold (1928, pi. 6, figs. 4-5; pi. 7, fig. 2). It is surprising that
Davidson did not mention muscle platforms in his descriptions. It is suspected that
Davidson’s fig. 9 (1864-5, pi. 11) was a reconstruction drawn from other brachiopods,
some possibly of Silurian age.

Interior structure. Enlarged transverse serial sections are shown in text-fig. 7. The in-
ternal description is based on more than 350 acetate peels taken from four serially sec-
tioned specimens. Numerous specimens had a lining of dolomite crystals.

EXPLANATION OF PLATE 46

All figures natural size.

Figs. 1-11. Mimatrypa desquamata (Sowerby). 1, 2, Ventral and dorsal views of the pedicle valve figured
in Sowerby 1840, pi. 56, fig. 19, paratype SM H3836, Plymouth (from preservation more likely
Wolborough). 3, Ventral view of a decorticated shell showing pallial arteries, SM H3842, Plymouth.
4, Dorsal view of a broken portion of ventral valve (complete specimen probably figured in Davidson
1864-5, pi. 11, fig. 5), BM 22119, Wolborough. 5, 6, Poorly preserved flatfish specimen figured in
Sowerby 1840, pi. 56, fig. 21, paratype Geol. Soc. Coll. 6322, Plymouth. 7, 8. Ventral and posterior
views of well-preserved specimen, rather globose and with angular beak, GSM 50855, Wolborough.
9-11, Dorsal, ventral and lateral views of lectotype figured in Sowerby 1840, pi. 56, fig. 20, Geol.
Soc. Coll. 6342, Plymouth; note rib shape and pattern of bifurcation.

Palaeontology, Vol. 8

PLATE 46

COPPER, Middle Devonian atrypoids

PAUL COPPER: UNUSUAL STRUCTURES IN DEVONIAN ATRYPIDAE 367

A pedicle collar was not observed, an effect possibly due to poor preservation. The
inner margins of the deltidial plates show inward protrusions (text-fig. 7, 0-7-3-0 mm.).
Deltidial plates are corrugated, thin, but not hollow, intricately interlocked medially.

text-fig. 7. Transverse serial sections of Mimatrypa desquamata (Sowerby), Lummaton, Devon.
Intervals in millimetres from the ventral apex. X 2.

and fused near the dorsal umbo. Teeth are massive, squared, with a strong inner lobe
and weak, stubby outer lobe. The inner margins of the dental plates are zigzag; no
lateral cavities appear, nor are there any circular growth lines in the dental plates (as,
for example, in Atrypa and Desquamatia).

In the brachial valve a weak, usually amorphous, cardinal process occupies the noto-
thyrial cavity (text-fig. 4b); in one specimen three to five longitudinal striations were

368

PALAEONTOLOGY, VOLUME 8

observed. Hinge plates are thick. Crura (text-fig. 56) are composed of minute, micro-
scopic calcite fibres oriented normal to the hinge plate and commonly arched over the
hinge plate to leave a small gap (text-fig. 5b, 4-5 mm.). The crural fibres radiate ventro-
laterally, becoming increasingly more diffuse and tenuous towards the sides of the shell
cavity. Jugal processes are not fully known: they appear to be disjunct. In text-fig. 7
(at 9-0 mm.) an anomalous structure, part of the right-hand jugal process, appears to be
strikingly similar to one found in Gruenewaldtia latilinguis (Schnur) (text-fig. 8). David-
son’s conclusions about the jugal processes are contradictory: in Davidson 1864-5,
pi. 11, figs. 7-8, a disjunct set of processes is drawn, but in Davidson 1882-4, pi. 1,
figs. 15, 156, a fused jugum is shown. The spiralia have up to fifteen whorls.

Ribs. The wavelength of the ribs is roughly similar to that of the finely ribbed Mima-
trypa insqucimosci (Schnur) but not as coarse as in M. flabeUata (Roemer). There is a
predominance of the finer type of ribbing. In the mid-area, from eight to thirteen ribs
occupy 10 mm. of arc; the average is about eleven.

Ribs typically are round and wide-crested and rather narrow-troughed (PI. 46,
fig. 10). Some specimens, particularly ones with angular beaks (PI. 46, figs. 7, 8), appear
to have narrower and more thinly crested ribs. The full variation cannot be determined
because of the lack of well-preserved specimens. In specimens with decorticated shells
it is common to see ribs reflected on the shell margins as short furrows (e.g. Whidborne
1893, pi. 13, fig. 13; and this paper PI. 46, fig. 3; PI. 47, fig. 13).

Growth lamellae. These are very weakly developed at the margins. No frills are present.
Presumably Davidson incorporated this feature in the specific name ‘ desquamata'
(Latin de, without or loss of; squamata, covered with squamae or scales).

Growth and variation. A large variability in shape, ranging from extremely flat forms to
rather globose forms, is common, and appears to be typical of reef-inhabiting organisms.
A separation of forms does not appear to be feasible. Sowerby distinguished a variety
which he called compressa (PI. 46, figs. 5, 6; PI. 47, fig. 10), a form which seems to be
more common at Plymouth than elsewhere. At present this is treated as a minor varia-
tion. Rare globose specimens are found at Wolborough.

Three specimens of the variety compressa Sowerby were figured by Whidborne (1893,
pi. 13, figs. 13-15), and they have been re-examined. Of these, only one (fig. 13) is in fact
a small specimen of Mimatrypa desquamata (Sowerby); the other two belong to the
genus Carinatina Nalivkin.

Davidson has clearly demonstrated the variety of forms present in Mimatrypa des-
quamata (Sowerby). His drawings are, unfortunately, almost wholly figurative and have
obviously been restored (occasionally they have been drawn as mirror images, e.g. in
his 1864-5, pi. 11, fig. 2). The narrow forms (Davidson 1864-5, pi. 10, figs. 12, 13) were
not found in collections and appear to be a scarce variety. Extremely wide forms are
not uncommon (op. cit., pi. 11, figs. 4, 5). The shape and outline most common to
M. desquamata are those shown in pi. 10, figs. 9, 10.

Leidhold described two species, Atrypa circularis and A. globosa (1928, pi. 6, figs. 4,
5; pi. 7, fig. 2), from a quarry near Iserlohn, Germany. The type specimens of these two
species, as well as other topotypes, were re-examined. Nearly identical specimens are
present in the collections from Devon. Their rib structure is the same and very little
difference, except in shape, could be found between Leidhold’s own species. Further-

PAUL COPPER: UNUSUAL STRUCTURES IN DEVONIAN ATRYPIDAE 369

more, as support for their synonymy with Mimatrypa desquamata (Sowerby), is their
synchronous upper Givetian occurrence.

Remarks. The species is distinguished from M. insquamosa (Schnur), to which it is most
closely related, by its consistently greater size (almost ‘gigantesque’), somewhat finer
ribs, and internally by the presence of spiralia and lack of pedicle collar.

It differs more strongly from M. flabellata (C. F. Roemer) in its biconvexity and fine
bifurcating ribs. Internally, the latter has a more strongly elevated dorsal muscle plat-
form.

Mimatrypa desquamata (Sowerby) has been reported from all over the world, but it
is likely that the resemblance to most other atrypids is only superficial. In England and
Germany it is common in upper Givetian rocks and appears to have particularly
favoured a biohermal environment. It is known in Germany at Iserlohn (Bilveringsen
quarry), Waldgirmes and Balve. At present it is not known to occur in the Torringer
beds and Plattenkalke of the Bergisch-Gladbach area near Koln, which also are of
Givetian age.

Alekseeva (1960, pp. 421-4) described a new subgenus Atrypa ( Desquamatia ) based on
the type species Atrypa khavae Alekseeva, and which, at that time, appeared to be related
to the ubiquitous ‘ Atrypa ’ desquamata described by Sowerby. A detailed examination
of Sowerby’s types revealed that this was not the case (Copper 1964). Biernat (1964, p.
282) has retained Alekseeva’s interpretation that the large ‘ Atrypa ’ desquamata Sowerby
of the Rhineland, apparently also present in Poland, belongs to Desquamatia (and
hence the subfamily Atrypinae).

Genus Gruenewaldtia Chernyshev 1885
Type species. Terebratula latilinguis Schnur 1851.

Diagnosis. An adequate diagnosis is given by Struve (1955, p. 211). As Struve (1955,
1964) has pointed out, the genus Paiaferella Spriestersbach 1942 is a junior subjective
synonym of Gruenewaldtia Chernyshev 1885.

Gruenewaldtia latilinguis (Schnur)

Plate 47, figs. 1-3; text-fig. 8

Material. Eighteen specimens from Chercombe Bridge shales.

Remarks. This species, which is common to abundant in middle Eifelian sediments of the
Eifel region in Germany, has not been described in detail. A short diagnosis is given by
Struve (1955, p. 212) and a further revision is being undertaken by him. The specimens
from Chercombe Bridge, Devon, are indistinguishable from the German forms. For this
reason, a close correlation is drawn between these two areas.

Davidson described an ‘ Atrypa ' latilinguis from Chercombe Bridge in 1882-4 (p. 41,
pi. 2, figs. 9, 9a) and assumed correctly that it was closely related to the Eifel species
of Schnur. In the Eifel, the genus is wholly confined to sediments of Eifelian age, al-
though Biernat (1964, pp. 280, 333) dated the Skaly beds of the Holy Cross Mountains
in Poland, containing Gruenewaldtia latilinguis (Schnur), as lower Givetian. Ivanova
(1962) has shown that Gruenewaldtia also occurs in the Givetian of Russia. It is possible
to come to the rather contradictory conclusion that the same fossils can have differing,

370

PALAEONTOLOGY, VOLUME 8

text-fig. 8. Transverse serial sections of Gruenewaldtia Iatilinguis (Schnur), Chercombe Bridge, Devon.
The jugal process (at 6T mm.) should be compared with that in text-fig. 7 (at 9 0 mm.). X2.

EXPLANATION OF PLATE 47

All figures natural size.

Figs. 1-3. Gruenewaldtia Iatilinguis (Schnur). Ventral, lateral, and posterior views of specimen serially
sectioned in text-fig. 8, BM B22077, Chercombe Bridge.

Figs. 4-6. Mimatrypa cf. insquamosa (Schnur). Ventral, dorsal, and posterior views of well-preserved
specimen, GSM 50859, East Ogwell.

Figs. 7-9, 14, 16. Mimatrypa cf. desquamata (Sowerby). 7-9, Ventral, dorsal, and posterior views of
finely ribbed specimen, BM B22105, Wolborough. 14, Posterior view of finely ribbed form, BM
B22098, Wolborough. 16, Ventral view of same variety, BM B49921, Wolborough.

Figs. 10, 13, 15. Mimatrypa desquamata (Sowerby). 10, Dorsal view of paratype figured in Sowerby
1840, pi. 56, fig. 22, Geol. Soc. Coll. 6323, Plymouth. 13, Dorsal view of small specimen figured in
Whidborne 1893, pi. 13, fig. 13, SM H4290, Lummaton (?shell bed). 15, Dorsal view of large typical
form, Geol. Soc. Coll. 6911, Wolborough.

Figs. 1 1-12. ILycoplioria rhenana (Leidhold). Lateral and dorsal views of specimen figured in Davidson
1864-5, pi. 11, fig. 12, Geol. Soc. Coll. 6913, Wolborough.

Figs. 17-19. Mimatrypa aff. insquamosa (Schnur). Posterior, dorsal, and lateral views of well-preserved
coarsely ribbed form, BM B44375, Lummaton (?).

Palaeontology, Vol. 8

PLATE 47

COPPER, Middle Devonian atrypoids

PAUL COPPER: UNUSUAL STRUCTURES IN DEVONIAN ATRYPIDAE 371

restricted chronostratigraphic distribution in widely separated areas. Ivanova (1962)
has advanced two highly plausible explanations for such contradictions: one of salinity
controls of brachiopod distribution, and the second of ‘facies’ control.

Gruenewaldtia has so far been found in Britain only at Chercombe Bridge, in shales
believed to be middle Eifelian in age (see text-fig. 1). This is the first time these forms
have been examined for internal structure. Only a single transverse section of the pedicle
valve of Gruenewaldtia latilinguis (Schnur) is available for comparison in the literature
(Struve 1955, fig. 8a). Biernat (1964) did not describe topotype material from the Eifel.
In sectioning topotype Eifel material I have not come across major internal differences
between the Polish, Eifel, and Devon forms, although a few small, but important,
structural details were previously omitted.

A distinctive internal structure, common to most examples of Gruenewaldtia Cherny-
shev and Mimatrypa Struve, and consisting of small internal projections of the deltidial
plates, has also been found in specimens from Devon (text-fig. 8, 2-2-24 mm.). This
structure is also present in Gruenewaldtia matutina Struve and G. cf. rhenana (Spriesters-
bach) from the Eifel. Similarly, it appears to be present in G. sibirica Ivanova (1962,
fig. 36), where it seems partially to join a pedicle collar. In G. latilinguis helenae Rzhons-
nitskaya (1960, pi. 2, figs. 1, 2) the protrusions extend from the pedicle collars rather
than the deltidial plates. This feature may be useful in classification.

In the pedicle valve of specimens of G. latilinguis from Devon, two major septa sup-
port the muscle plate, which consists of a 0-5 to 1 mm. thick, monocrystalline sheet (septa
and muscle plate together form a muscle platform); also, it is significant that one or
two remnant septa are still found apically (text-fig. 8, 1 -2—4-2 mm.). This is also true of
the same species in the Eifel. The dorsal plate is supported by two lateral septa and
a weak median septum.

G. sibirica Ivanova and the Devon form possess a pedicle collar. Lateral cavities are
absent.

Order Pentamerida
Superfamily Porambonitacea
Genus ILycophoria Lahusen 1885
ILycophoria rhenana (Leidhold) 1928

Plate 47, figs. 11-12; text-fig. 9

1864-5 Atrypa flabetlata (C. F. Roemer); Davidson ( partim ), pp. 59-60, pi. 11, figs. 10-12.

1928 Karpinskia rhenana Leidhold, pp. 98-99, pi. 7, figs. 4 a-d, 5.

Remarks. An internal examination by means of serial peels of a representative specimen
of a form previously grouped under ‘ Atrypa' flabeiiata revealed a shell wall structure
and internal morphology which are not found in atrypid brachiopods. However, the
exact affinities of this form are still unknown and it is tentatively placed in the Penta-
merida. Even Whidborne, who examined this form (1893, p. 120), was doubtful about its
affinities and mentioned a resemblance to ? Mutationella guerangeri (Verneuil). There
is also a crude resemblance to ?Nanothyris primaeva (Gosselet) (1920, p. 102, pi. 14,
figs. 19-26). The form may also have affinities with the terebratulid genus Notothyris
Waagen. A more detailed description must be left until better and more abundant
material can be examined, and until its true affinities are confirmed.

372

PALAEONTOLOGY, VOLUME 8

text-fig. 9. Transverse serial sections of ILycophoria rhenana (Leidhold), Wolborough, Devon. The
?deltidial plates (2-5 to 2-8 mm.) and crude septalium (2-3 to 3T mm.) are of interest. X 2.

Conclusions

In the Givetian sediments of north-western Europe, the atrypids grouped under
Mimatrypa Struve appear to have replaced the atrypid genera Gruenewaldtia and Des-
quamatia, which are abundant in sediments of Eifelian age. Mimatrypa lived in a bio-
hermal environment almost exclusively, and appears to have been absent in synchronous
argillaceous sediments.

Mimatrypa is believed to have been derived from Gruenewaldtia. A tentative lineage
sketched by Copper (1964) is based on a number of internal morphological features
which these genera have in common. Mimatrypa shows a strong tendency towards
complete disintegration of structures connecting the spiralia to the hinge plates. Mima-
trypa appears to have become extinct by the end of Middle Devonian time : it possibly
gave rise to Anatrypa Nalivkin.

Acknowledgements. The writer wishes to thank Mr. J. Ferguson (British Museum (Natural History)),
Mr. M. Mitchell (Geological Survey Museum), and Mr. A. G. Brighton (Sedgwick Museum) for
their assistance in the loan of specimens. Dr. Hermann Jaeger (Museum fur Palaontologie, Humboldt
Universitat, Berlin) encouraged me to examine valuable German material. The financial assistance of
the Royal Commission for the Exhibition of 1851 and the National Research Council of Canada is
gratefully acknowledged.

REFERENCES

alekseeva, R. e. 1960. A new subgenus Atrypa ( Desquamatia ), family Atrypidae Gill. Dokl. Akad.
Nauk SSSR 131 (2), 421-4, figs. 1-3. [In Russian.]

1962. Devonian Atrypida from the Kutznetsk and Minussinsk Basins and the eastern part of the

northern Urals. Izd. Akad. Nauk SSSR, 1-196, 12 pi. [In Russian.]
biernat, G. 1964. Middle Devonian Atrypacea (Brachiopoda) from the Holy Cross Mountains,
Poland. Acta palaeont. pot. 9 (3), 277-340, 15 pi.

boucot, a. j. et at. 1964. On some atrypoid, retzioid and athyridoid Brachiopoda. J. Paleont. 38,
805-22, pi. 125-8.

breivel, i. a. in khodalevich, A. n. et al. 1959. Brachiopods and corals of the Eifelian bauxite depo-
sits of the eastern slopes of the central and northern Urals. Minister, geol. okhrany nedr. SSSR,
Ural. geol. uprav. 1-159, 61 pi. [In Russian.]

bublichenko, n. L. 1928. Brachiopod fauna of the lower Paleozoic of Sarachumisk (Kutznetsk)
Basin. Izv. Geol. kom. 46 (8), 996-1008, pi. 49, 50. [In Russian, German summ.]
copper, p. 1964. European Mid-Devonian correlations. Nature, Lond. 204 (4956), 363-4, figs. 1, 2.
(in press). A new Middle Devonian atrypid brachiopod from the Eifel, Germany. Senck. leth.

PAUL COPPER: UNUSUAL STRUCTURES IN DEVONIAN ATRYPIDAE 373

davidson, t. 1864-5. A monograph of the British Devonian brachiopods. Palaeontogr. Soc.
[Monogr.l 3 (6), 1-131, 20 pi.

1882-4. Supplement to the British Devonian Brachiopoda. Ibid., 5 (1), 1-62, pi. 1—3.

elliott, g. f. 1961. A new British Devonian Alga, Palaeoporella lummatonensis, and the brachiopod
evidence of the age of the Lummaton shell bed. Proc. Geol. Ass. 72, 251-60.
gortani, m. 1911. Contribuzioni alio studio del paleozoico carnico. 4, La fauna meso-devonica di
Monumenz. Palaeontogr. ital. 17, 114-228, pi. 16-20.
gosselet, J. 1920. Description de la faune siluro-devonienne de Lievin. Mein. Soc. geol. N. 6 (2),
1-161, 17 pi.

hotz, e. et al. 1955. Zur Geologie der Eifelkalkmulden. Beih. Geol. Jb. 17, 45-204.
house, m. r. 1963. Devonian ammonoid successions and facies in Devon and Cornwall. Quart.
J. geol. Soc. Load. 119, 1-27, pi. 1-4.

ivanova, e. a. 1962. Ecology and development of Silurian and Devonian brachiopods of the Kutz-
netsk, Minussinsk and Tuvinsk Basins. Akad. Nauk SSSR, Trudy Paleontol. inst. 83, 1-150, 20 pi.
[In Russian.]

jukes-browne, a. J. 1906. The Devonian limestones of Lummaton Hill, near Torquay. Proc. Geol.
Ass. 19, 291-302.

leidhold, c. 1928. Beitrag zur Kenntnis der Fauna des rheinischen Stringocephalenkalkes, ins-
besondere seiner Brachiopodenfauna. Abh. Preuss. geol. Landesanst. 109, 1-99, 7 pi.
maurer, F. 1885. Die Fauna der Kalke von Waldgirmes bei Giessen. Abb. Hess. geol. Landesanst.
12, 1-277, 1 1 pi.

middleton, g. v. 1954. The Middle and Upper Devonian of a selected area of south Devon. Ph.D.
thesis (unpubl.), University of London, England.

Phillips, J. 1841. Figures and descriptions of Palaeozoic fossils of Cornwall, Devon and West Somerset.
1-231, 60 pi.

quenstedt, F. A. 1885. Handbuch der Petrefactenkunde. 1-1239, atlas 100 pi.
roemer, c. F. 1844. Das rheinische Uebergangsgebirge. 1-96, pi. 1-6.

rzhonsnitskaya, M. A. 1960a. The genus Gruenewaldtia Chernyshev from the Devonian of the SSSR.
Shorn. Stat. paleontol. biostrat., Nauch.-iss/ed. inst. geol. arkt. 20, 45-50, pi. 1-2. [In Russian.]

19606 in Osnovy paleontologii (ed. sarycheva, g. t.). Izd. Akad. Nauk SSSR 257-64, pi. 53-56.

schnur, j. 1853. Zusammenstellung und Beschreibung sammtlicher im Uebergangsgebirge der Eifel
vorkommenden Brachiopoden nebst Abbildungen derselben. Palaeontograpbica, 3, 169-247,
pi. 22-29, 31-32, 33-45.

siehl, a. 1962. Der Greifensteinerkalk (Eiflium, Rheinisches Schiefergebirge) und seine Brachio-
podenfauna, 1. Geologie; Atrypacea und Rostrospiracea. Ibid. 119 (5/6), 173-221, pi. 23-40.
sowerby, j. de c. 1840, in Sedgwick, a. and Murchison, r. i. On the physical structure of Devonshire
and on the sub-division and geological relation of its stratified deposits. Trans, geol. Soc. Lond.
5 (3), 633-703, pi. 52-58.

struve, w. 1955. Gruenewaldtia aus dem Schonecker Richtschnitt (Brachiopoda, Mittel-Devon der
Eifel). Senck. leth. 36 (3/4), 205-34, 4 pi.

1961. Zur Stratigraphie der siidlichen Eifler Kalkmulden. Ibid. 42 (3/4), 291-345, 3 pi.

1964. Mimatrypa n.g. (Atrypidae/Palaferellinae). Ibid. 45 (5), 433-40, figs. 1-5.

taylor, p. w. 1951. The Plymouth Limestone. Trans. R. geol. Soc. Cornwall, 18, 146-214.
tiasheva, a. p. in tiasheva, a. p. et al. 1962. Brachiopods, ostracods and spores of the Middle and
Upper Devonian of Bashkir. Izd. Akad. Nauk SSSR, Bashkir, filial, Gorno-geol. inst. 5-138, pi. 1-27.
[In Russian.]

torley, k. 1934. Die Brachiopoden des Massenkalkes der oberen Givet-Stufe von Bilveringsen bei
Iserlohn. Abh. Senckenb. Naturforsch. Ges. 43 (3), 67-148, pi. 1-9.
whidborne, g. f. 1893. A monograph of the Devonian fauna of the south of England. Palaeontogr.
Soc. [Monogr.], 2, 89-160, pi. 11-17.

PAUL copper
Department of Geology,
Imperial College,
London, S.W. 7

Manuscript received 2 July 1964

THE PALAEONTOLOGICAL ASSOCIATION

COUNCIL 1964-5

President

Dr. L. R. Cox, British Museum (Natural History), London
Vice-Presidents

Dr. W. S. McKerrow, University Museum, Oxford
Professor F. H. T. Rhodes, University College, Swansea

Treasurer

Dr. T. D. Ford, Department of Geology, The University, Leicester
Assistant Treasurer

Dr. C. Downie, Department of Geology, The University, Sheffield

Secretary

Dr. C. H. Holland, Department of Geology, Bedford College, London, N.W.l

Editors

Mr. N. F. Hughes, Sedgwick Museum, Cambridge
Dr. Gwyn Thomas, Department of Geology, Imperial College of Science, London, S.W.7
Dr. I. Strachan, Department of Geology, The University, Birmingham, 15
Dr. M. R. House, University Museum, Oxford

Other members of Council

Dr. C. G. Adams, British Museum (Natural History), London
Professor P. M. Butler, Royal Holloway College, Surrey
Dr. W. J. Clarke, British Petroleum Company, Sunbury on Thames
Dr. G. Y. Craig, The University, Edinburgh
Dr. T. D. Ford, The University, Leicester
Dr. B. M. Funnell, Sedgwick Museum, Cambridge
Dr. J. M. Hancock, King’s College, London
Dr. G. A. L. Johnson, The University, Durham
Dr. F. A. Middlemiss, Queen Mary College, London
Mr. M.- Mitchell, Geological Survey and Museum, London
Dr. A. J. Rowell, The University, Nottingham
Professor Scott Simpson, The University, Exeter
Dr. L. B. Tarlo, The University, Reading
Dr. H. Dighton Thomas, British Museum (Natural History), London

Overseas Representatives

Australia : Professor Dorothy Hill, Department of Geology, University of Queensland, Brisbane
Canada: Dr. D. J. McLaren, Geological Survey of Canada, Department of Mines and Technical
Surveys, Ottawa

India : Professor M. R. Sahni, Department of Geology, Panjab University, Chandigarh
New Zealand: Dr. C. A. Fleming, New Zealand Geological Survey, P.O. Box 368, Lower Hutt
Eastern U.S.A.: Professor H. B. Whittington, Museum of Comparative Zoology, Harvard Univer-
sity, Cambridge 38 .Mass.

Western U.S.A. -.Professor J. Wyatt Durham, Department of Paleontology, University of California,
Berkeley 4, Calif.

PALAEONTOLOGY

VOLUME 8 * PART 2

CONTENTS

The interrelationships of some Cretaceous Codiaceae (Calcareous Algae).

By G. F. ELLIOTT 199

Corallum increase in Lithostrotion. By r. k. jull 204

The Namurian goniatite Nuculoceras stellarum (Bisat). By b. k. holdsworth 226

Systematics, affinities, and life habits of Babinka , a transitional Ordovician
lucinoid bivalve. By A. l. mcalester 231

New Silurian graptolites from the Howgill Fells (Northern England). By r. b.

RICKARDS 247

The development of Lasiograptus harknessi (Nicholson 1867). By r. b.

RICKARDS and O. M. B. BULMAN 272

A new fertile lycopod from the Lower Carboniferous of Scotland. By K. l.
alvin 281

Keuper miospores from Worcestershire, England. By r. f. a. clarke 294

British Permian saccate and monosulcate miospores. By R. f. a. clarke 322

A new Ordovician crinoid from Dolgellau, North Wales. By d. e. b. bates 355

Unusual structures in Devonian Atrypidae from England. By p. copper 358

PRINTED IN GREAT BRITAIN AT THE UNIVERSITY PRESS, OXFORD
BY VIVIAN RIDLER, PRINTER TO THE UNIVERSITY

30
>

30

^ ” -eg*- <o Ngfoilsgx m m

(/J \ z CO ± CO

!IES SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHIIWS SBIdVaail LI

BRARIES SMITHSONI

v' . CO

2

1SN I NVINOSH1IWS S3IHVaail LIBRARIES SMITHSONIAN INSTITUTION NOliDlllSNI NVINOSHil

^ m VhW"7 “ “ W' o o

R I E S~ SMITHSONIAN- INSTITUTION NOIlfUllSNI NVIN0SH1IINS SBIBVaail LI B RAR I E S^ SMITHSONI
z □ z , 2 i- 2

O ^ m O /<5a '^\ O

m ^ X^vas^

co — to

BRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHil

- <5 ^ _ 2 CO 2

S /SvfiroX < A 2 <

— A ^BK'OA —

A 2

4 wy.u/ -i— xxv Wvvo* w "

t- s\V$vv z , /•’■

CO

RIES SMITHSONIAN INSTITUTION NOlinillSNI NVIN0SH1IINS S3IBVaail LIBRARIES SMITHSON

1SNI NVIN0SH1IWS S3IUVUan LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHil
2 r- 2 r~ 2

DO

>

30
-

m Vou iis2>

co tr co _ co

RIES SMITHSONIAN INSTITUTION NOlinillSNI NVINOSH1IIAIS S3IHVdan LIBRARIES SMITHSON

CO 2 CO 2 CO 2 CO

2E JSk s < 1 ± X -

io'kmSv Zi ../ a-Sy /^WA -i ^ z: -h /S5

\ — ,

x.:

> X^iusA-

* 2 — Tr> x 2 co 2 co

I1SNI NVIN0SH1IIAIS SBIBVBan LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVIN0SH1

— CO 333, CO —

CO

CO

CD
30
>

. _______ oy ;□

m — X>ASH»£

co _ co —

I1SN! NVIN0SH1IINS S3IMVaail LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVIN0SH1

2 co 2 , CO 2 V CO 2

&, * * #k s s 5 m. s

KkWfeVX 2
O
co
*/ X

2 /w - w" = t

__ > ^jr 5 > \*uii^>

dVaan^LIBRARIES^SMITHSONIAN INSTITUTION NOliniliSNI NVINOSHIIIAIS^S 3 I a V a a IT 2 L I B R
— co co ~ ... co

A^V/H~X CO

CO

5 o 5 ^ O

2 _j 2 -i 2 _J 2

riTUTiON NoiiniusNi nvinoshiiiais S3i a va an libraries smithsonian^institution noiic

I 2 I /0^h " /4fr^\ I ~

2

v*r, ■ “ - ■ —

m 2 m x^osxjy X; m

CO — CO ± CO

BRARIES SMITHSONIAN INSTITUTION NOliniliSNI NVIN0SH1IINS SBIBVaaiT LI

co 2 , CO 2 co 2

A § %% s ifm § s #4..

2 xS| | | br

>v 5-

N

V

ITUTION NOliniliSNI NVINOSHIIIAIS S3iavaai1 LIBRARIES SMITHSONIAN INSTITUTION NOIlf

avaan libraries Smithsonian institution NoiiniusNi nvinoshiiiais S3iavaan libr

r~ z *_ _ . 2 r* ... 2 r~

„ 2cc*PwX O . m XoAliSi'Jx O ~ .

avaan libraries Smithsonian institution NoiiniusNi nvinoshiiiais saiavaan libf

“ co ~ co — co —

vy m CD

2 _l 2 _J 2 _

fitution NoiiniusNi nvinoshiiiais saiavaan libraries Smithsonian institution noik

^ov -

SI m ^ m x^os*^2 'll m

co — co _ co 2

a va a i \j>s? aries Smithsonian institution NoiiniusNi nvinoshiiiais saiavaan libr

_ _ CO 2 co 2

< -a. S xST'noX < <- < ^<7Tr^ S . <