st 3 Peeetitatatnsptetatotecenegienrten te pereroctlstgrslestnralanenttatetslriptgtrsaee mob rteenlar st et peatentee ee. rears sebstysa prrearitista nea eteitcnteeste ster peate Spahr
Seer thes 2 . sseecesbaretiteerier tee iepas f ¢s mired

Teeseedwesettemetseirithttes

Seetrin io
eter eure Si arate

BULLETIN OF
THE BRITISH MUSEUM
(NATURAL HISTORY) __

An
\s
<y
ix

GEOLOGY
MOLy Xa
1966—1967

TRUSTEES OF
THE BRITISH MUSEUM (NATURAL HISTORY)
LONDON: 1969

DATES OF PUBLICATION OF THE

No.
No.
No.
No.
No.
No.
No.

1. 22nd November
2. 8th February .
3. 7th April

4.
5
6
7

21st April

. 23rd May
. 13th June
. 21st July

PRINTED IN GREAT BRITAIN
BY ALDEN & MOWBRAY LTD
AT THE ALDEN PRESS, OXFORD

PARTS
1966
1967
1967
1967
1967
1967
1967

CONTENTS
GEOLOGY VOLUME XIV

Some pearl-bearing Ceramoporidae (Polyzoa). K.P. OAKLEY

The Bathonian Upper Estuarine Series of Eastern England. Part 1:
Ostracoda. R. H. BATE

New Cretaceous berycoid fishes from the Lebanon. C. PATTERSON

Stratigraphy and palaeography of the Yorkshire Oolites and their
relationships with the Lincolnshire Limestone. R. H. BATE

Further notes on palaeoniscid fishes with a classification of the
Chondrostei. B.G. GARDINER

Lower Carboniferous trilobites of North Devon and related species
from Northern England. J. E. PRENTICE

Fossil Mammals of Africa. No. 22: Pelorovis oldowayensis Reck, an
extinct bovid from East Africa. A. W. GENTRY

Index to Volume XIV

Tit

243
301

.

OME PEARL-BEARING
MOPORIDAE (POLYZOA

SOME PEARL-BEARING CERAMOPORIDAE | #
(POLYZOA)

BY

KENNETH PAGE OAKLEY, D.Sc.
\

Pp. 1-20; 9g Plates

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)
GEOLOGY Vol. 14 No. 1
LONDON: 1966

THE BULLETIN OF THE BRITISH MUSEUM
(NATURAL HISTORY), instituted im 1949, 1s
issued in five series corresponding to the Departments
of the Museum, and an Historical serves.

Parts will appear at irregular intervals as they become
veady. Volumes will contain about three or four
hundred pages, and will not necessarily be completed
within one calendar year.

In 1965 a separate supplementary series of longer
papers was instituted, numbered serially for each
Department.

This paper is Vol. 14, No. 1 of the Geological (Palae-
ontological) series. The abbreviated titles of periodicals
cited follow those of the World List of Scientific
Periodicals.

© Trustees of the British Muscum (Natural History) 1966

TRUSTEES OF
THE BRITISH MUSEUM (NATURAL HISTORY)

Issued 22 November, 1966 Price Li 5s.

SOME PEARL-BEARING CERAMOPORIDAE
(POLYZOA)
By KENNETH P. OAKLEY

CONTENTS
Page
I. INTRODUCTION a 6 : . 0 c c : c I
II. SYSTEMATIC POSITION AND MORPHOLOGY OF FAVOSITELLA : 4
III. D&EScRIPTION OF PEARL-BEARING SPECIES . : : ¢ : 7
Favositella interpuncta (Quenstedt) ‘ ; : : : a
Favositella squamata (Lonsdale) . : : ‘ : : 14
Favositella anolotichoides sp. nov. : 5 : 5 : 16
Favositella gotlandica sp. nov. F 0 : . : é 19
V. REFERENCES . ; z : : 3 : 3 ; ; 19
SYANOIPSIUS)

Studies of British and Swedish ceramoporid Polyzoa resulted in an emendation of the generic
diagnosis of Favositella, in the recognition of two new species of this genus and of several
“formae’ of the type species. The known pearl-bearing Polyzoa are referred to four species :
F. interpuncta (Quenstedt), F. sgquamata (Lonsdale), F. anolotichoides sp. nov., and F. gotlandica
sp.nov. The first three species are found in the Wenlock and Ludlow Series of Britain, the fourth
in the Gotlandian formation of Sweden. In Britain the pearl-bearing Favositella stock ranges
from the Woolhope Limestone to the Aymestry Limestone. An account is given of various
morphological characters of the Ceramoporidae as exemplified by Favositella. It is suggested
that the extra-macular zooecia were occupied by autozooids, and the mesopores by nanozooids.
Evidence is adduced that maculae were the seat of reproductive processes, and it is suggested
that they were sometimes converted into brood-chambers, the large macular zooecia having
lodged gonozooids. The wide range of morphological facies exhibited by some species of
Favositella, notably F. interypuncta (Quenstedt), is attributed to the varying effects of different
environmental factors on the development of the zoarium. To express differences of this order
a number of formae are established.

I. INTRODUCTION

PRELIMINARY Studies of the Polyzoa of the Wenlock Limestone showed that a number
of Ceramoporidae contained pearl-like phosphatic calculi which had been formed
during life, sometimes around “brown bodies’, and occasionally embalming the
secondary embryo of a polypide (Pl. 1, fig. r). That these phosphatic “ pearls ”’
were formed most commonly in species of the genus Favositella at a time when this
genus was approaching extinction suggested that they represented a physiological
defect to which members of this stock were prone. The structure, mineralogy, and
mode of occurrence of calculi in Favositella have already been described in detail,
and the probable biochemical factors involved in the deposition of phosphate in
coelomic fluid analysed as far as possible (Oakley 1934). There remained the need
to publish a systematic account of the species of Favositella found to contain calculi.

GEOL. 14, I I

4 SOME PEARL-BEARING POLYZOA

Professor N. Spjeldnaes (Copenhagen) informed the author (personal communica-
tion, 1951) that he had found similar but smaller phosphatic calculi in old (gerontic?)
individuals of a few non-ceramoporoid Polyzoa (e.g. in an Ordovician species of the
trepostome Dianulites, and in a Gotlandian species of the cryptostome Ptilodictya).
He confirmed the author’s observations regarding optical properties and chemical
composition of polyzoan calculi and fully agreed that the deposition of phosphatic
matter was probably a sign of degeneration. Professor Spjeldnaes added the
interesting observation that to a vertebrate palaeontologist the phosphatic calculi
of Polyzoa are practically like globular calcified cartilage.

it, SYSTEMATIC POSITION AND MORPHOLOGY (OR FA VOSi i yee EA

Order CYCLOSTOMATA Busk
Suborder CERAMOPOROIDEA Bassler
Family CERAMOPORIDAE Ulrich
Genus FAVOSITELLA Etheridge & Foord 1884: 472
1893 Bythotrypa Ulrich : 324.
Type Species (by monotypy) Favosites interpunctus Quenstedt 1878.

The genus Favositella was established by Etheridge & Foord on the basis of a
determination of the internal characters of Favosites interpunctus Quenstedt from the
Wenlock Limestone of Dudley. Owing to the supposed relationships of this genus
with Favosites it was for long regarded as a tabulate coral, but Bassler’s re-investiga-
tion of topotype material (191I : 100) showed that the type species was actually a
polyzoan belonging to the generic group for which Ulrich had proposed the name
Bythotrypa, a genus erected to include ceramoporoids with loosely constructed
interzooecial tissue formed by irregular mesopores. Bassler showed that it was
possible to apply Ulrich’s definition of Bythotrypa to the type species of Favositella
without modification. It is as follows :

“ Zoaria massive or lamellate. Zooecia forming long continuous tubes,
intersected by thin diaphragms, their walls minutely crenulate and with the
structure characterizing the ceramoporoids. Lunarium well defined, large,
projecting above the rest of the aperture margin. Mesopores numerous, open
at the surface, interiorly forming a species of vesicular tissue unusually loose
and irregular in construction.”

Progress in the study of ceramoporoid Polyzoa has necessitated from time to time
a widening of the bases of distinction of the various genera, and Favositella has
proved no exception to this trend. In the course of the present work further species
of Favositella have been brought to light, and a study of these based on a large
amount of material has indicated the necessity of minor modifications of the generic
diagnosis. Thus the zoarium may be free or encrusting. In thin, laminate forms
the mesopores are so short that there is no more than a tendency for them to form
the loose vesicular tissue which characterises the genus. Relatively large, irregular

SOME PEARL-BEARING POLYZOA 5

pores in the zooecial walls, although not invariably present, are sufficiently common
to serve as a useful guide in recognizing species. The lunarium in one newly-
described species contains acanthopore-like granules.

EMENDED DIAGNOSIS. Zoaria massive or lamellate, encrusting or free. Zooecia
forming tubes of variable length, usually long, intersected by thin diaphragms,
their walls minutely crenulate and structure characteristic of ceramoporoids. Large
mural pores commonly present. Lunarium well-defined, usually large and pro-
jecting above rest of apertural margin. Mesopores numerous, open at the surface,
interiorly tending to production of species of vesicular tissue unusually loose and
irregular in construction.

The characters of the family are well exhibited by Favositella. Externally the
main distinguishing feature of the Ceramoporidae is the elevation of the frontal
margin of the zooecial aperture into a hood-like structure known as the lunarium.
When typically developed this accentuates any obliquity which the zooecial apertures
may have, and gives an imbricate appearance to the celluliferous surface. The
lunarium is usually crescentic in cross-section. Its position on the frontal side of
the zooecium—the side on which the aperture of the zooid is situated in the living
polyzoan—suggests that it may have had a protective function. It is formed of
auxiliary ectocyst and originates as a lining of the frontal notch of the zooecium,
which corresponds with the sinuation characteristic of the aperture of many Cheilo-
stomes. While a reinforcement of the frontal margin of the zooecial aperture is not
an uncommon feature in Polyzoa, the elevation of the thickened portion as a hood-
like structure is a feature mainly characteristic of certain Palaeozoic groups. The
lunarium is found with modifications in both families of the Ceramoporoidea
(Ceramoporidae, Fistuliporidae) and also in one family of Cryptostomata (Cystodicty-
onidae). There is one Jurassic cyclostomatome, Chilopora guernoni Haime, in
which the zooecia have raised zooecial lips indistinguishable from the lunaria of
Palaeozoic ceramporoids (Haime 1854, pl. 10, fig. 50).

The ectocyst in the Ceramoporoidea has a very distinctive character. In highly
magnified thin-sections it exhibits a finely laminated structure and, at the same time,
a clouded or minutely granulated appearance. It is thus easily distinguishable from
the uniformly hyaline primary ectocyst of the Trepostomata. Whereas in the latter
the ectocyst appears to consist of crystalline calcite, in the Ceramoporoidea (and
probably in all Cyclostomata) it is not only minutely porous, but apparently com-
posed of intimately connected, sub-microscopic fibres, perhaps originally embedded
in a corneous ground-mass.

The fracture of a ceramoporoid zoarium is very different from that of a trepostome.
On breaking a trepostome the walls of adjacent zooecia part cleanly down the middle;
the fracture has thus a clean-cut appearance. On the other hand, when a ceramo-
poroid polyzoan, or for that matter any fossil cyclostome, is broken open, the
fracture passes through and across walls without discrimination, and the broken
surface appears dulland amorphous. The two styles of fracture are a reflexion of the
histological differences between the two groups. In trepostomes the ectocyst is
crystalline, Furthermore, the zooecial walls are duplex in origin so that a median

6 SOME PEARL-BEARING POLYZOA

plane of weakness exists between the ectocyst of contiguous zooecia. The amor-
phous style of fracture in ceramoporoids is an outcome of the intimately connected,
perhaps felted, structure of the zooecial walls and the resulting absence of a median
plane of weakness.

In Favositella the primary ectocyst is sometimes invested by auxiliary ectocyst.
The secondary ectocyst lining the frontal end of the zooecial aperture, and con-
stituting the lunarium, is distinguished from the rest of the zooecial wall by its
lighter colour.

In common with those of most Ceramoporoidea the zoaria of Favositella exhibit an
internal polymorphism. That is to say two types of ‘cell’ are present: zooecia,
and interstitial ‘ cells’, or mesopores. On the analogy of Recent Cyclostomata it
may be assumed that the majority of the zooecia were occupied by normal polyzoan
individuals, or antozooids (Borg 1926: 188), and the mesopores by modified indi-
viduals, or heterozooids (i.e. nanozooids or kenozooids). An observation made in
the course of the present work on Favositella tends to suggest that the mesopores
were occupied by nanozooids rather than by kenozooids ; that is by zooids in which
there was a simplified, but nevertheless functional, polypide. I refer to the occasional
presence of dahllite calculi within the mesopores of Favositella. Since these pearls
are believed to be the indirect result of polypide-degeneration, they are unlikely to
have originated in closed ‘cells’ occupied only by kenozooids (the extreme modi-
fication of zooid in which all functional organs have disappeared).

A striking feature of Ceramoporoidea (but one which they share with other Palaeo-
zoic Polyzoa, notably the Trepostomata) is the grouping of the zooecia about regu-
larly spaced centres composed of modified ‘cells’. These centres are termed
maculae. They consist of concentrations of mesopores, usually interspersed with,
or surrounded by, zooecia larger than the average and thus falling into the category
of topomorphs (Lang 1906: 66). The maculae may be level with the general surface
of the zoarium, but they are more commonly either slightly depressed or slightly
raised (in which case they are known as monticules).

It would seem that the maculae were, in the first place, centres of growth. Thus
the simple discoidal zoaria of Ceramopora imbricata Hall are made up of a central
macula from which the zooecia radiate towards the growing edge (Bassler 1906 : 19).
Similarly in more complex zoaria the frontal, or lunariate, ends of all zooecial aper-
tures proximate toa macula. It is probable that in a mature zoarium the mesopore
and enlarged zooecia in the maculae took on functions connected with reproductive
processes. Ulrich has already suggested (1890 : 298) that the larger zooecia in or
surrounding maculae served as the receptacles in which embryos were developed into
the larval stage. He compares them with the genoecia of the Recent genus Cvisia,
which are merely slightly modified zooecia set aside for reproductive functions.
Zooids adapted and transformed for the production of embryos have been termed
gonozooids by Borg (1926: 188). The heterozooids which are believed to have
occupied the mesopores may have served a protective function.

In the course of investigating the Silurian species of Favositella a number of
observations have been made which support the idea that the maculae were, at any

SOME PEARL-BEARING POLYZOA 7

rate at a certain stage in astogeny, the seat of reproductive processes. It has been
noted that openings in the walls of zooecia and mesopores are most frequent in the
maculae, while in a few mature zoaria of Favositella interpuncta the maculae are
found to have been converted into moderately deep, stellate, or circular pits, open
at the surface. These were first described by Quenstedt (1878 : 10, pl. 143, fig. gy ;
“ Sterngruben ’) who seems to have regarded them as either pathological or due to
damage during the life of the organism. It appears, however, that they have origin-
ated through the coalescence of the mesopores and zooecia in the maculae by the
breaking-down, or resorption, of their walls. These pits (Pl. 3, fig. 6; Pl. 9, fig. 6)
are so strikingly similar to the ‘ brood-chambers’ observed in certain Recent
Heteroporidae (Borg 1933 : 267, pl. 2, figs. 1, 4) that they may have served the same
function. It is not improbable, on the analogy of Recent forms, that in the breeding
season the walls of the mesopores and large zooecia in the maculae were partially
resorbed, and that the developing embryos were discharged into the central space so
formed, there to undergo larval development. However, further work on this aspect
of the Ceramoporidae is required before this interpretation can be credited with any
degree of certainty.

III. DESCRIPTION OF PEARL-BEARING SPECIES
Favositella interpuncta (Quenstedt)

(Px, fies. 1-6; Pl. 2, figs. 1-3,6,7; Pl. 3, figs. 4-6; PI. 4, figs. 1-6; Pl. 6, figs. 1=5 ;
Pin7 ics ua. Pi 8 tigsen, 3) 407 Ping) tigs.6; 7)

1851 Chaetetes Bowerbanki (partim) Edwards & Haime: 272.

1855 Monticulipora? Bowerbanki (partim) Edwards & Haime : 268, pl. 63, figs. 1b, ¢ (non
fig. 1a).

1873 Monticulipora sp. 5 (partim) Salter : 109.

1873 Monticulipora sp. 6 (partim) Salter : 109.

1878 Favosites interpunctus Quenstedt : 10, pl. 143, fig. 9.

1879 Favosites fibrosus Sollas : 510 (vide pl. 24, figs. 12, 17-20).

1884 Favositella interpuncta (Quenstedt) Etheridge & Foord : 473, pl. 10, figs. 1, 1a—f.

191i Favositella interypuncta (Quenstedt) ; Bassler : ror, text-fig. 35.

The specimen figured by Edwards & Haime (1855) as an immature form of their
Chaetetes (Monticulipora?) bowerbanki has been found on examination to be identical
with the polyzoan described below as Favositella interpuncta forma texturata. As
this specimen (now preserved in the Sedgwick Museum, Cambridge: A4024) is
one of the syntypes of Chaetetes (Monticulipora?) bowerbanki, that name must be
included in the synonymy of Favositella interpuncta. However, it is clear that
Edwards & Haime mainly based their diagnosis of C. bowerbanki on specimens such
as the other syntype figured by them (1855: pl. 63, fig. ra). Since this specimen
is a coral referable to the genus Favosites, it is convenient to restrict the name
bowerbanki to forms agreeing with the anthozoan syntype, and to reject it as a name
for the polyzoan on the grounds of its being a homonym.

8 SOME PEARL-BEARING POLYZOA

LECTOTYPE (here chosen). Specimen from the Wenlock Limestone of Dudley
figured by Quenstedt (1878 : pl. 143, fig. 9). Preserved in the Museum of the Institute
of Geology and Palaeontology at Tiibingen, Germany. This is almost certainly the
type since it is the only specimen preserved in the Quenstedt Collection.

DEscRIPTION (Pl. 2, figs. 1, 6; Pl. 8, figs. 1, 4). Free, roughly campanulate
zoarium with spreading margins. Total height 31 mm. ; maximum diameter 47 mm.
Actual thickness averages 7 mm., but locally this is exceeded owing to the super-
position of two or more layers of zooecia. The deeply concave inferior surface is
covered by a thin, coarsely wrinkled epitheca or epizoarial membrane ; where this is
worn the nearly prostrate bases of the zooecia tend to show through (PI. 2, fig. 6).
The celluliferous surface is moderately smooth. Over the greater part of the surface
the zooecial tubes have relatively thick walls and in outline are mainly rounded
polygons (Pl. 8, fig. 1). There are 4 to 5 of these zooecial openings in an interval of
2mm. The lunaria are represented by inconspicuous slightly raised thickenings of
the frontal angle. In one area of the zoarial surface (Pl. 8, fig. 4) the zooecia are
thin-walled and more quadrangular in outline, while the lunaria are thin, hood-like
structures occupying the frontal fourth of the zooecial margin.

Open mesopores are moderately frequent ; the majority are small and circular,
and occur mainly at the angles of junction of the zooecia. They are more abundant
in some areas than in others.

Maculae are well developed and arranged roughly quincuncially ; the average
distance separating their centres is 5 mm. They show no tendency to become
monticules. Two types, or phases, can be recognized amongst the maculae. The
first type, which occurs in areas where the zooecial walls are thick and the lunaria
inconspicuous, is characterized by an abundance of small circular mesopores scattered
amongst zooecia which are slightly larger than average. There is a tendency for
the walls of the zooecia and mesopores to break down in the centre of such maculae,
resulting in the formation of the stellate pits noted by Quenstedt, and now regarded
as brood-chambers (p. 7). The second type of macula is associated with the other
surface phase, characterized by thin-walled zooecia and conspicuously developed
hood-like lunaria. Such maculae consist of a ring of large zooecia, with overarching,
lunarial hoods, enclosing an irregularly-shaped area occupied by a cluster of large,
thin-walled mesopores which seem to form a species of open vesicular tissue. The
exposed surfaces of the adjacent lunarial hoods are marked by a series of fine lines
which converge upwards and appear to be continuations of the walls of the mesopores
on which the lunaria impinge.

Internal characters may be deduced from topotypes (see p. 10).

VARIATION IN EXTERNAL CHARACTERS. Examination of a large number of
specimens of I’. interpuncta has shown that the external variation in this species is
very great, and it is found convenient to recognize a number of formae (see p. I3).

Forma typica (PI. 1, figs. 4-6; Pl. 2, figs. 2,3; Pl. 3, fig. 6; Pl. 4, figs. 1,55 PL 6;
figs. 3-5, Pl. 9, fig. 6). Zoaria tumular, with well-defined circular or elongate-oval
outlines. Commonly such zoaria commenced their growth on some bulbous foreign
body, such as a large gastropod shell, but eventually extended beyond it, so that

SOME PEARL-BEARING POLYZOA 9

when fully developed their margins were free epithecated expansions (PI. 1, fig. 4).
As observed by Etheridge & Foord, the form of the shell or other foreign body to
which this species became attached frequently determined the external form assumed
by the zoarium. This is illustrated by a specimen in the Holcroft Collection at
Birmingham University (Pl. 1, figs. 4,5). This polyzoan commenced growth on the
side of a large Loxonema, with the result that the zoarium is broader and higher at
one end than at the other and has the general form of a drumlin. Some campanulate,
or tumular zoaria appear to have commenced growth on a knot of hardened mud on
the sea-floor. In such specimens the base is deeply concave.

While the total height of these tumular zoaria may be as much as 50 mm., the zoarial
thickness rarely exceeds 7 mm., either on account of the space occupied by the
‘host’ or by reason of their actual concavity (see Pl. 2, fig.2). Where the zoarial thick-
ness exceeds about 7 mm. there has usually been reduplication of the zooecial layers.

In forma typica, zoaria range in size from small bun-shaped colonies measuring
30 mm. in diameter and 12 mm. in height (e.g. B.M. N.H., D.33918) to large tumular
masses 90 mm. in length, 60 mm. in width and 30 mm., or more, in height.

The zooecial apertures are relatively thick-walled and present the form of rounded
polygons. Lunaria are inconspicuous, usually little more than blunt processes.
Thin-walled quadrangular zooecia, with prominent cowl-like lunaria, are only present
when there has been new growth. This type of surface is regarded as constituting a
distinct forma (see last paragraph below).

Mesopores are small and scattered, and have the appearance of pin-pricks in the
thickened angles of junction of the zooecial walls. Typically the maculae are
inconspicuous (cf. first type in lectotype), but in some zoaria they are raised into low
monticules (Pl. 1, fig. 6). Occasionally the zooecia and mesopores in the maculae
have become confluent through break-down of the walls, thus giving rise to con-
spicuous pits which in all probability served as brood-chambers (see p. 7 and Pl. 3,
fig. 6; Pl. 9, fig. 6).

Banna byeupora (Pl-2; fig. 7; Pl. 4, figs. 2, 6; Pl. 6, figs. 1, 2; Pl. 9, fig. 7).
This forma is typically exhibited by thin zoaria encrusting the surface of smooth-
shelled brachiopods such as Meristina tumida (Dalman), and more rarely tumid
gastropods such as Poleumita globosa (Schlotheim). The thickness of the zoarium
is normally 2-3 mm.; the surface is remarkably smooth and maculae are incon-
spicuous. The zooecial walls are thickened distally with the result that the apertures
are sub-polygonal or rounded. Small circular mesopores, occurring at the zooecial
angles, are relatively abundant, particularly in the maculae. A local tendency to
linear arrangement of the zooecia is occasionally noticeable.

Bomna textuvata (Pln3; fig) 5); Pl. 4, fig. 4; Pl. 7, fig. 2; Pl. 8, fig. 3). This
facies is also typically expressed when the whole zoarium is a thin encrusting sheet,
as on smooth-shelled brachiopods such as Meristina tumida (Dalman), or Gypidula
galeata (Dalman). Over the whole surface the zooecial apertures are arranged in a
regular lace-like pattern, a dominant feature of which is the occurrence of sharply
defined star-like maculae in quincunx. Each macula consists of an irregular ring of
five or six large zooecia with prominent, usually finely ribbed, lunarial hoods which

GEOL. 14, I. 2

10 SOME PEARL-BEARING POLYZOA

face away from a slightly depressed area of thin-walled, angular mesopores. The
bases of the lunarial hoods of the large zooecia occupy about a third of the whole
circumference of the zooecial aperture.

The zooecia outside the maculae are thin-walled and mainly quadrangular. They
are slightly larger than in the forma typica, there being 3 to 4 in an interval of 2 mm.
They have small but prominent lunaria, the frontal or external surface of which is
rounded and sometimes marked by fine ribs converging upwards. Mesopores are
almost entirely confined to the maculae.

All zooecia are oriented with respect to some macula. The intersection of the
zones of propagation of the several maculae has resulted in the surface being divided
up into variously shaped sectors, within each of which the lunarial hoods have the
same direction, and across the boundaries of which the orientation of the zooecia
changes. Within each sector of constant orientation the zooecia consequently
have a very characteristic appearance (rather like that of a net pulled from various
centres) which is markedly different from that of the forma tyfica. Nevertheless,
the fact that areas of new growth in typical zoaria closely approximate to the
texturata condition (see p. 13, and PI. 8, fig. 4), is proof that this form is largely
transitory.

Forma irregularis (Pl. 4, fig. 3; Pl. 7, fig. 1). Zoaria forming irregular crusts on
the surface of gastropods, corals and brachiopods. The zoarium usually has a very
uneven surface, and varies in thickness from 2mm. to 10mm. The character of
the surface shows an irregular transition from that of forma typica to that of forma
texturata. The two types of surface-character are so intimately mixed that the
zoaria have a distinctive appearance. Maculae are unevenly distributed, and very
variable in structure, although the type associated with the pure ¢exturata condition
predominates. There are on the average 4 zooecial apertures in 2 mm.

Forma intermedia (PI. 3, fig. 4) only differs from zvvegularis in mode of origin of the
zooecial irregularity (see p. 4).

DESCRIPTION OF INTERNAL CHARACTERS. These have been studied in topotype
material.

Forma typica (Pl. 4, figs. 1, 5; Pl. 6, figs. 3-5). In tangential sections (PI. 4,
figs. I, 5) the majority of the zooecia appear as irregularly rounded polygons with
an average diameter of 0-4 mm. Such zooecia have an indistinct lunarium occupying
a notch in the frontal margin. The zooecial walls in the mature zone of a typical
zoarium are relatively thick (civca 0:05 mm.), and are formed of coarsely lamellar,
crypto-fibrous tissue. The lunaria are formed of more pellucid tissue. Sporadic
breaks can be observed in the zooecial walls and indicate the occurrence of large
mural pores. Mesopores appear as rounded or oval spaces 0-Imm. to 0-2 mm.
in diameter, and occur at the angles of junction of the zooecial walls. They are
noticeably more abundant in the maculae.

The occurrence of areas in which the zooecia show linear arrangement, sub-ovate,
quadrangular, or subtriangular outlines, and more distinct, pointed lunaria, indicates
that the section has passed into an immature zone in which there is an approach to the
texturata condition.

SOME PEARL-BEARING POLYZOA II

In deeper transverse sections through the mature region the zooecia and mesopores
have thinner walls and appear more angular in outline ; irregularly pentagonal and
sub-triangular outlines predominate amongst the zooecia. In macular areas the
zooecia show elongate outlines with a longer axis measuring as much as 0-6 mm.
Breaks in the walls are relatively common and lead occasionally to meandrine
zooecial spaces. The lunaria are inconspicuous, each being represented by a small
triangular patch of light-coloured tissue in the frontal angle of the zooecium.

Transverse sections through the sub-ephebic levels show rather quadrangular
zooecia arranged in diagonal rows, recalling the ¢extwrata condition fully described
below.

Vertical sections through typical zoaria (Pl. 1, fig. 3; Pl. 6, figs. 3-5) show one
to three layers of zooecia varying in thickness from 3 mm. to 12 mm. Each layer
has a thin basal lamina. In free expansions the basal lamina of the lowest zooecial
layer is coated by a thin, corrugated epithecal membrane. The zooecial tubes arise
obliquely from the basal lamina (PI. 6, fig. 3), but after a short distance become more
or less vertical. In the mature zone mesopores are locally intercalated between
the zooecia.

The walls of the zooecia are irregularly crenulate ; in sufficiently thin sections it
is possible to detect small, and often rather oblique, mural pores. When a vertical
section cuts a zooecial wall tangentially the ectocyst is seen to have a transversely
laminated structure. The zooecial walls sometimes show oblique, upwardly directed,
spine-like processes jutting into the zooecial cavity (Pl. 6, fig. 3). It is unlikely that
these are comparable with the spines observed in certain trepostomes (Cumings &
Galloway 1915 : 358); they more probably represent incipient mesopores. Thin
diaphragms occur at intervals throughout the zooecial tubes; they are slightly
convexo-concave in the upward sense, and are separated by an interval generally
varying from one to two tube-diameters.

Mesopores tend to be loosely vesicular in character. Their walls contract and
expand in an irregular manner so as to produce a roughly moniliform effect (Pl. 1,
fig. 3; Pl. 6, fig. 5). In some cases the mesopores are cut up into vesicles by oblique
or transverse projections of the wall, similar to those already noted in the zooecia.
Thin diaphragms occur sporadically in the mesopores. The walls of the mesopores
in fully mature zoaria are thickened near the surface.

Forma brevipora (Pl. 4, figs. 2,6; Pl. 6, figs. 1, 2). Tangential sections of zoaria
in this condition do not differ materially from those of forma tyfica. Mesopores are
variable in size and frequency. The zooecia present mainly sub-polygonal, rounded
or oval outlines with an average diameter of 0:4 mm. ; lunaria are indistinct.

Vertical sections show that the zooecia become erect after an abbreviated oblique
phase. The essential differences from forma typica are: (i) the shortness of the
mature zone ; and (ii) the simplicity of the mesopores, which are formed by a bifurca-
tion of the zooecial wall, as in forma texturata.

Forma texturata (Pl. 4, fig. 4; Pl. 7, fig. 2). Tangential sections bring out the
marked linear arrangement of the zooecia within circumscribed sectors. The
zooecia are predominantly quadrangular in outline, although some may show oval or

GEOL LAs: 28

12 SOME PEARL-BEARING POLYZOA

rounded pentagonal cross-sections. They are thinner walled (average thickness,
0-035 mm.) than in forma tyfica, and vary in diameter from 0-4 mm. to 0:5 mm.
The lunaria of all zooecia are sharply defined, and usually sagittate in outline (Pl. 4,
fig. 4). Small rounded mesopores occur, but are almost entirely confined to the
maculae.

Vertical sections (PI. 7, fig. 2) are indistinguishable from those through the proximal
region of forma typica. They bring out the shortness and obliquity of the zooecia
typical of this form. Lunaria are difficult to distinguish in vertical sections, but are
clearly direct prolongations of the frontal walls of the zooecia, lined internally with
auxiliary ectocyst. Mesopores are shallow and thin-walled. Each is formed by the
bifurcation of a ‘septum’ from the upward-facing, or basal surface of a zooecial
wall, usually at a distance of 0-5 mm. to 1:0 mm. above the basal lamina. There is
no epitheca ; the so-called basal lamina has no separate existence but is merely the
common basal wall of the prostrate portions of the zooecia. Diaphragms are usually
present.

The zooecia vary in height from 1 mm. to 3 mm. The superposition of several
layers of zooecia in the texturata condition is rare. Consequently the average thick-
ness of zoaria which are wholly in this condition is 2 mm.

Forma irregularis (Pl. 4, fig. 3; Pl. 7, fig. 1). Tangential sections show areas with
regularly aligned, quadrangular zooecia, with distinct sagittate lunaria, which pass
abruptly, or by mixture, into areas where the zooecia have rounded polygonal outlines
and indistinct lunaria. Vertical sections show that this forma is associated with the
irregular and local reduplication of zooecial layers ; thick layers show the characters
of forma typica, thin layers those of forma texturata (cf. Pl. 7, fig. 1).

STRATIGRAPHICAL DISTRIBUTION. The earliest known example of the species is
a specimen from the Lower Wenlock Shales at Buildwas, preserved in the Foord
Collection at the British Museum (Natural History) (D.36318). It is in the
brevipora condition, and forms an incrustation on a shell of Meristina tumida (Dal-
man). Similar specimens occur in the Upper Wenlock Shales at Walsall.

The species occurs in greatest abundance in the Wenlock Limestone, particularly
in Staffordshire and at Rumney, near Cardiff. In common with other Ceramo-
poridae it is rare in the Wenlock Limestone of Shropshire. There is a single record
of the species in the Wenlock Limestone of the Woolhope area in Herefordshire
(Gardiner 1927 : 323).

Twenty-five per cent of all the Polyzoa recovered from the thin representative of
the Wenlock Limestone exposed in the Ty Mawr Lane at Rumney proved to belong
to this species!. The zoaria from this locality are all small. Most of them seem to
belong to the typical forma, but the intractable ferruginous matrix with which they
are encrusted makes an examination of surface features a matter of difficulty. In
some specimens mesopores are very sparse (PI. 6, fig. 4).

The great majority of the specimens of F’. inteypuncta preserved in old collections

1 Many of the specimens of Favositella interpuncta from Rumney are remarkably rich in pearls. There

can be no doubt that the pearl-bearing specimens erroneously described by Sollas as Favosites fibrosus
actually belong to this species.

SOME PEARL-BEARING POLYZOA 13

appear to have come from the shaly beds of the Wenlock Limestone at Wren’s Nest
and Castle Hill, Dudley, Staffordshire. Recent collecting has shown that the horizon
at which they occur in greatest abundance is in the upper part of the Middle Nodular
Beds. All the described formae appear to be represented at this level. The only
specimen recovered from the Lower Limestone is one in the brevipora condition from
the basement beds at Walsall. One specimen of forma ivregularis from Dudley,
preserved in the Sedgwick Museum (A5943), has a matrix which suggests derivation
from the passage beds between the Upper Limestone and the Lower Ludlow.

A small fragmentary specimen of Favositella from the Aymestry Limestone of
Shucknall Hill, Woolhope, Herefordshire, in the Geological Survey & Museum
(57956) appears to represent a variant of this species.

REMARKS. So great is the diversity of form exhibited by Favositella interpuncta
that some sort of subdivision of the species would seem to be desirable. An examina-
tion of an extensive series of specimens from the Wenlock beds has indicated, how-
ever, that the observed variation is continuous and not of a mutational order. The
diversity of form is partly explained by the fact that different stages of astogeny are
quite distinct in appearance, and partly by the marked effects on the development of
this polyzoan produced by different environments.

It was therefore proposed by the author (1938) that the only way of expressing
the main structural facies exhibited by the species was by applying formal as distinct
from varietal, names to the several forms. The differences which the form-names
indicate may be in one case developmental, in another due to the effect of a special
environment, and so on. However, in the present state of our knowledge of these
Polyzoa it is possible to make only very tentative correlations of this sort.

The forma ¢ypica seems to represent the normal form assumed by a zoarium which
developed to full maturity (?gerontic stage) under constantly favourable conditions.

The forma brevipora is apparently the condition of zoaria which attained maturity
under constant, but unfavourable conditions. In tubular Polyzoa it would seem
that the length of the zooecia is not a reflexion of the rate of deposition of sediment as
it is in corals, but depends mainly on the frequency of polypide-regeneration ; this
in its turn being correlated with such factors as temperature, salinity and food-supply.

The forma texturvata is the one which provides the most interesting problem.
Zoaria showing this forma are so different in aspect from the typical zoaria that they
were for long considered as belonging to a distinct species. In fact, judging by the
evidence of museum labels, they have been generally regarded, even by palaeonto-
logists familiar with typical Favositella interpuncta, as representing a species of
Ceramopora.

The recognition of the zoaria in question as a forma of Favositella interpuncta
resulted from the discovery that they showed features almost identical with those of a
small portion of the celluliferous surface of the type-zoarium of that species. A
close examination of the holotype and other typical specimens of F. interpuncta
has shown, in fact, that wherever a new layer of zooecia has commenced to develop
on the surface of a mature zoarium (i.e., wherever there has been local rejuvenation),
the area of new growth shows the characters which comprise the texturata condition.

14 SOME PEARL-BEARING POLYZOA

In other words this condition is essentially that of the neanic or sub-ephebic phases
of development in Favositella interpuncta. Nevertheless there are reasons for re-
garding this condition as constituting in some cases a distinct forma.

Complete zoaria consisting of a single layer of zooecia in the ¢extwrata condition are
of common occurrence, particularly on smooth-shelled brachiopods. The well-
defined maculae which they exhibit suggest that the zoaria were fertile. The
probable explanation of these forms is that they represent zoaria whose development
was arrested at the sub-ephebic stage owing to certain environmental conditions.

The forma texturata preserves, to some extent, features of more primitive Ceramo-
poroid genera such as Crepipora; for example, the restriction of mesopores to the
maculae.

Zoaria in which there has been a tendency for new layers of zooecia to develop at
irregular intervals, and over small, limited areas, have a very irregular aspect which
contrasts markedly with that of both forma typica and forma texturata. For such
zoaria the term forma ivregularis was proposed in 1938.

The irregular rejuvenation of the zoarial surface which tended to the production of
this forma, was most probably connected with the variable character of the micro-
environment. The extent to which environmental factors may have controlled
growth-form in Favositella, may be gauged by the fact that 12 out of 13 specimens
of F’. interpuncta recovered from a single exposure in the Wenlock Limestone proved
to belong to the same forma. The bed from which they came was largely composed
of current-drifted shells and corals which had been accumulated in an area adjacent
to a coral reef (‘ ball-stone ’).

After further work on Favositella interpuncta the author found it desirable to
recognize another forma, amongst the forms grouped as forma iregularis. The
additional forma, called zntermedia, is illustrated by Pl. 3, fig. 4. It differs from
forma irregularis in so far as the irregularity of the surface is not due to patchy
rejuvenation, but to unequal development of individual zooecia.

Favositella squamata (Lonsdale)
(Plz, fig..4 5 Pl: 33tig.2)5 Pls) ies. 1.8 s Plazes; (RE 8 ipe2 > Pliocene)

1839 Discopora squamata Lonsdale : 679, pl. 15, figs. 23, 23a.
1873 Monticulipora sp. 4 (partim) Salter : 109.
1890 Cyvrepipova squamata (Lonsdale) Ulrich : 471.

LECTOTYPE (here chosen). Specimen marked a on slab of shelly flagstone from
Wenlock Limestone, Sedgeley, figured by Lonsdale (1839, pl. 15, figs. 23, 23a).
Geological Society Coll. 6596, Geological Survey & Museum. It is probable that
Lonsdale’s diagonsis of the species was wholly based on this specimen.

DESCRIPTION (Pl. 9, fig. 5). Fragment of flat laminate zoarium, 1:5 mm. thick;
apparently encrusting the valve of a brachiopod. The fragment is 8 mm. long and
5mm. wide. The surfaceis rather worn. The zooecia are sub-erect and contiguous.
Outside the raised area the zooecial apertures are thin-walled, mainly rhomboidal
and arranged in intersecting, diagonal rows ; 4 to 44 apertures occur in an interval of

SOME PEARL-BEARING POLYZOA 15

2mm. In the subtriangular frontal half of the zooecium the apertural margin is
elevated as a well-defined lunarium. The raised part of the zoarial surface is clearly
a macula ; it is distinguished by the relatively large size of the zooecia and by the
occurrence of numerous irregular mesopores. Mesopores have not been observed
outside the macula.

External characters. Based on study of topotype material (see Pl. 2, fig. 4; Pl. 3,
fig. 2; Pl. 8, fig. 2; Pl. 9, figs. 4,5). Zoaria form thin encrusting sheets, generally
on flat surfaces such as are provided by the valves of strophomenid brachiopods.
The margins of the zoarium sometimes extend beyond the encrusted object. Free
marginal expansions formed in this way have developed a coarsely wrinkled basal
epitheca (Pl. 2, fig. 4). The surface of the zoarium is sometimes irregularly monticu-
lose, but the raised areas do not invariably coincide with maculae.

The zooecial apertures are subtriangular, oval or rhomboidal. In the frontal half
of the zooecium the apertural margin is raised into a prominent and rather angulated
lunarium, the apex of which is often slightly overarching particularly in the maculae.
The external angle of the lunarium may be acute or rounded ; in the latter case it
is marked by several strong rugae or ribs which converge towards the apex. There
are from 4 to 5 zooecial apertures in an interval of 2mm. Maculae are well-marked
and sometimes form low monticules; they are superficially similar to those in
F.. interpuncta forma texturata, but usually show a more confused structure. The
large zooecia in typical maculae have abnormally high, pointed lunarial hoods.

Mesopores are mainly small and rounded, although larger, more angular ones occur
in some of the maculae. Outside the maculae, mesopores are, for the most part,
only of sporadic occurrence. A notable feature of the species is the occurrence of
certain circumscribed areas, distinct from normal maculae, in which minute thick-
walled mesopores are so abundant that they surround the zooecial apertures. In
these areas which are of irregular distribution and usually larger than the maculae,
the zooecial apertures are oval or sub-pyriform, and have low, inconspicuous lunaria
(Pir s; fig. 2).

Internal characters. Based on study of topotype material. Tangential sections
(Pl. 5, fig. 3) show fairly regularly aligned, thin-walled zooecia, varying in outline
from rhomboidal to sub-polygonal or sub-pyriform, and with an average diameter of
04mm. The frontal end of the zooecial aperture is deeply arcuate, and lined with
the lighter coloured tissue of the lunarium, which varies in form from a wide crescent
to a small triangle, depending on the level at which the section has been cut. The
zooecial walls are composed of the finely laminated and minutely granular tissue
characteristic of the genus ; they have an average thickness of 0-028 mm. Rounded
or oval mesopores, ranging in diameter from 0:08 mm. to 0-I mm., are seen in localized
areas.

In transverse sections through the sub-distal or proximal region (Pl. 5, fig. I),
the zooecia present irregularly polygonal outlines. The zooecial walls are very
tenuous and tend to be somewhat flexuous. Small sub-polygonal mesopores occur
mainly in the maculae. In the macular areas frequent breaks can be observed in the
zooecial walls,

16 SOME PEARL-BEARING POLYZOA

Vertical sections (Pl. 7, fig. 5) : in typical zoaria there are from one to four layers
of zooecia. The layers vary in thickness from 1mm. to 4mm. The zooecial tubes
arise from a thin, basal lamina, which, in the case of the lowest zooecial layers in
free-growing parts of a zoarium, is usually covered by a thin epitheca. The zooecial
tubes normally become almost vertical after a short prostrate phase, although in some
specimens they maintain an oblique course. In the peripheral region the zooecial
walls become thickened and sometimes irregularly crenulate ; short and rather thick-
walled mesopores are rarely more than 0:2 mm. deep. In the peripheral zone, the
walls of the mesopores and the zooecia commonly develop a ragged or loosely con-
structed appearance, owing to the occurrence of irregular mural pores. Diaphragms
are infrequent, or absent.

OccURRENCE. The species has been recorded only from the Wenlock Limestone
of Dudley and Sedgeley. It occurs mainly in the upper part of the Middle Nodular
Beds (for example at Wren’s Nest), and is much rarer than FP’. interpuncta.

Remarks. This species is easily confused with F. interpuncta forma texturata.
In both the lunaria are prominent, and in both the zoarial surface tends to be divided
into sharply defined sectors within which the zooecia have a constant orientation with
respect to some macula. However, F. squamata differs from F. imterpuncta forma
texturata in the following respects: (i) the zooecial apertures are on the average
slightly smaller and narrower ; (ii) the lunaria tend to be longer and more arcuate ;
(iii) the maculae show a more confused structure ; (iv) there are circumscribed
areas, apart from normal maculae, in which mesopores are abundant ; (v) the zooecial
tubes are longer and more erect. The character mentioned under (iv) is an example
of topomorphism.

There is no doubt that specimens of F’. sguamata showing the characters described
above are normal, fully mature zoaria ; whereas the superficially similar specimens
of F. interpuncta forma texturata are evidently immature, or stunted zoaria.

Favositella anolotichoides sp. nov.
(Plog; figse2, 3. Piss, fies’2, 5); PIN7s figs) 3 ee lho Mite ha)

1873 Monticulipora sp. 6 (partim) Salter : 109.
1934 Cevamoporella sp., Oakley : 314, pl. 14, figs. 15, 18.

HotoryPe (Pl. 3; fig: 3; Pl. 5)migs’2) 5; Pl: 7;digs. 3,4; Plo; fig. 3) bares:
(Nat. Hist,), D.33926, Wenlock Limestone, upper part of Middle Nodular Beds ;
exposure by lime-kiln on SW side of Wren’s Nest, Dudley, Staffs. The holotype is
an oval, encrusting zoarium, 4—5 mm. thick, originally 40 mm. long and 27 mm.
wide, attached to the surface of an ‘ Ovthoceras’ shell.

PARATYPES. Brit. Mus. (Nat. Hist.), 8 specimens from the type locality (D.36324,
D.36326- 36332), and 5 others (PD.4613, D.36319, D.36325, D.36333-34) ; Sedgwick
Museum, Cambridge, 2 specimens (A5964, A58q9); National Museum of Wales,
I specimen (G.591).

“>

SOME PEARL-BEARING POLYZOA 17

HORIZON AND LOCALITY. ‘The species is common in the upper part of the Middle
Nodular Beds of the Wenlock Limestone at Wren’s Nest, Dudley. It has also been
recorded from the Wenlock Limestone of Ty Mawr Lane, Rumney, near Cardiff;
and from the middle beds of the Wenlock Limestone in the Coate’s Farm Quarry,
Presthope Road, near Much Wenlock, Shropshire.

Description. External characters. Zoaria form moderately thick encrusting
sheets on cephalopod shells, corals, stromatoporoids, and more rarely on brachiopod
shells. Where the zoarium extends beyond the encrusted body, the basal surface
becomes enveloped by a thin, wrinkled epitheca. The surface of the zoarium is
usually somewhat uneven, but there are no definite macular elevations or depressions.
The zooecia are relatively thick-walled and radiate from macular centres spaced at
intervals of about 5mm. Each has a thin, horse-shoe shaped lunarium, the edges of
which bear minute, acanthopore-like granules. The zooecial apertures vary in
outline ; they tend to be roughly elongate-oval or sub-pyriform, but a slight con-
striction of the walls at the ends of the lunarial loop gives the apertures the form of a
key-hole. Locally the apertures become meandrine owing to the coalescence of some
of the zooecia.

Internal characters. In tangential sections (Pl. 5, figs. 2, 5) the majority of the
zooecia present slightly ‘ waisted’, oval outlines, with an average major axis of
0-45 mm. and an average minor axis of0-27mm. The outlines of some zooecia appear
geniculate owing to a slight declination of the axis of the lunarium relative to the
axis of the anterior part of the aperture. As seen in section, the lunarium is a narrow
band of light-coloured tissue, varying in form from a semi-circle to a three-quarter
circle, and constituting the frontal third of the zooecial wall. In sections of well-
preserved specimens it is possible to see g-12 acanthopore-like granules, or tubules,
within the lunarium (PI. 5, fig. 2). These appear as pellucid spots with an average
diameter of 0-02 mm. They are rather irregularly placed and their margins occa-
sionally project into the zooecial cavity. The zooecial walls vary considerably in
thickness ; the frontal wall, formed by the lunarium, has an average thickness of
0-025 mm., but elsewhere the walls may attain a thickness of as much as 0:05 mm.
Sub-polygonal mesopores, varying in diameter from 0-I mm. to 0:3 mm., are abundant
and practically surround the zooecia. Gaps in the walls of both mesopores and
zooecia are common and lead locally to meandrine outlines.

Vertical sections (Pl. 7, figs. 3, 4) usually reveal a single layer of zooecia ; more
rarely there are two superimposed layers. The thickness of a zooecial layer varies
from 2mm. to 5mm. The zooecial tubes arise from a thin basal lamina which is
epithecated in the case of free-growing expansions. The zooecia pass through a
brief prostrate phase and then rise vertically ; they have irregularly crenulate walls
in which occasional breaks may be observed. The walls are thickened throughout
the mature zone by auxiliary ectocyst. Thin diaphragms, concave upwards, occur
fairly frequently, but without any regular spacing, the intervals separating them
varying from one to three tube-diameters. Mesopores are intercalated between the
zooecia at the point where they become vertical, and persist throughout the mature
zone. They are loosely moniliform, or in some cases definitely vesicular in character

18 SOME PEARL-BEARING POLYZOA

(Pl. 7, fig. 4). Diaphragms similar to those in the zooecia are usually seen in some of
the mesopores. The walls of the mesopores, like those of the zooecia, are locally
broken by numerous gaps, a tendency particularly marked in macular centres.

The wall-tissue is less laminar and more granulated than in the two preceding
species.

REMARKS. This species provides an interesting problem in systematics. It
differs from all previously described species of Favositella in the isolation of its
zooecia by mesopore tissue and in the presence of acanthopore-like granules in its
lunaria. In the first of these characters, as in its habit, it recalls species of Ceramo-
porella. On the other hand, the lunarial structures as seen in tangential sections are
reminiscent of the tubules found in the lunaria of the Ordovician genus Anolotichia.

However, in Anolotichia the maximum number of tubules in a lunarium appears to
be seven, whereas in the present form there are often as many as twelve. Further-
more, in vertical section there is no indication that the tubules in this species have
either the extension or the tabulated structure characteristic of those in Anolotichia.
They can scarcely be regarded as tubules at all, and seem to have much more in
common with the granules found in the walls of certain species of Ceramoporella
(e.g. C. granulosa Ulrich 1890: 466). This fact, considered in conjunction with the
abundance of the mesopores, at first suggested that the correct reference of this
polyzoan was to Cevamoporella. More detailed investigation showed, however, that
the zooecial walls are frequently perforated by irregular pores—a feature, which has
never been observed in species of Cevamoporella. Moreover, in many specimens the
mesopores are found to exhibit the loose, vesicular character generally associated with
those of Favositella. These and more general considerations finally left no doubt in
the author’s mind that this polyzoan was, in fact, like the commoner pearl-bearing
forms, a species of Favositella, but one with a superficial resemblance to Anolotichia
on the one hand, and to Ceramoporella on the other.

Although there is no doubt that this polyzoan agrees more closely with Favositella
than with any other known genus of Ceramoporidae, the fact that it differs from all
previously described species of that genus in having acanthopore-like granules in the
frontal wall of the zooecium might have been taken as sufficient reason for regarding
it as the type of a new genus or sub-genus. It is considered, however, that the
wiser course is to include it in Favositella, at any rate until there is more proof that
such acanthopore-like granules are of phylogenetic significance. The sporadic
appearance of these or similar structures in isolated species of distantly related stocks
suggests that they are of no more than specific importance. For example, acantho-
pore-like granules can be detected in the zooecial walls of Crepipora lunatifera Bassler
(1911 : 88, fig. 27d), but have not been observed in other species of that genus.
Again, analogous granules occur in the zooecial walls of some species of Ceramo-
porella (e.g. C. granulosa Ulrich 1890 : 466), but not in others. Tabulated lunarial
tubules fall into a different category, since they are only found in species with a
number of important characters in common, and their presence is therefore justifiably
taken to indicate membership of a single generic group, to which the name Anolotichia
has been given.

SOME PEARL-BEARING POLYZOA 19

Favositella gotlandica sp. nov.
(Ez tiem sae 5) fies 4 On elo ties. 1,2)

HoitotyrPe. Brit. Mus. (Nat. Hist.), D. 33919; and D.33923 (thin-section) ;
Upper Silurian (Gotlandian), Miilde-margelsten (—Lower Ludlow), Miilde Tile-
factory, near Frojel, Gotland. It consists of an encrustation on a shell of Meristina
tumida (Dalman).

DEscrRIPTION. Zoarium thin and encrusting ; surface slightly uneven. Zooecial
apertures regularly aligned, thin-walled and sharply rhomboidal in outline, with
small, sharply elevated lunaria. On the average there are 4 zooeciain 2mm. The
zooecial tubes are sub-erect and contiguous. Small mesopores, having the appearance
of punctations, occur sporadically at wall-intersections. Maculae, although some-
times slightly raised, are relatively inconspicuous ; they are mainly recognizable by
the local abundance of mesopores.

In sections the zooecial walls are remarkable for their uniform thinness throughout.
Their average thickness is 0:02 mm. (compared with 0:05 mm. in the mature zone of
F. interpuncta forma typica ; 0-035 mm. in the forma ¢extuvata of that species ; and
0-028 mm. in F’. squamata).

In tangential section the zooecia show rhomboidal outlines with an average major
axis of 0-6 mm., and an average minor one of 0-35 mm. in length. The frontal end
of the zooecium is narrowly rounded and lined by an extremely narrow, crescentic
lunarium which can only be observed in sections which pass very close to the surface.
Mesopores are initially quadrangular, but their lumen has usually become rounded
through the growth of auxiliary ectocyst.

The vertical section shows a single zooecial layer with an average thickness of 0-7
mm. The zooecial tubes curve obliquely upwards after a brief prostrate stage.
The actual apertures are sub-direct. Mesopores are simple, short, and widen
upwards. Small gaps in the zooecial walls can be observed in parts of the zoarium,
and indicate the presence of mural openings. Diaphragms have not been observed.

RemMARKS. This species is closely related to F’. sqwamata (Lonsdale) from the
Wenlock Limestone, the two resembling one another in the rhomboidal form of the
zooecial apertures and in the comparative thinness of the zooecial walls. The present
species is mainly distinguished from F’. squamata by the smallness of its lunaria, by
the more regular shape of the zooecial apertures, and by the curving obliquity of its
zooecial tubes as seen in vertical section.

The horizon from which the holotype was collected is in Lindstrém’s Division c of
the Gotlandian formation ; that is the Miilde-margelsten, which has been correlated
by Hede (1921 : 87) with the Lower Ludlow beds of this country.

IV. REFERENCES

BassLer, R.S. 1906. The Bryozoan Fauna of the Rochester Shale. Bull. U.S. Geol. Surv.,
Washington, 292 : 1-137, pls. 3-31.
1911. The Early Paleozoic Bryozoa of the Baltic Provinces. Bull. U.S. Nat. Mus.,
Washington, 77 : 1-382, pls. I-13.

20 SOME PEARL-BEARING POLYZOA

BorG, F. 1926. Studies on Recent Cyclostomatous Bryozoa. Zool. Bidr. Upps., 10: 181-

507, pls. 1-14.

1933. A Revision of the Recent Heteroporidae. Zool. Bidr. Upps., 14: 253-394, pls, 1-14.

Cumincs, E. R. & Gattoway, J. J. 1915. Studies of the Morphology and Histology of the
Trepostomata or Monticuliporoids. Bull. Geol. Soc. Amer., New York, 26: 349-374, pls.
IO-15.

Epwarps, H. M. & Hare, J. 1851. Monographie des Polypiers fossiles des Terrains palaeo-

zoiques. Aych. Mus. Hist. nat., Paris, 5 : 1-502.

1855. A Monograph of the British Fossil Corals, P. 5, Corals from the Silurian Formation :

245-299. Palaeontogy. Soc. [Monogy.| London.

ETHERIDGE, R. (JuN.) & Foorp, A. H. 1884. Descriptions of the Palaeozoic Corals in the
Collections of the British Museum (Nat. Hist.). Ann. Mag. Nat. Hist., London (5) 13:
472-476, pl. 17.

GARDINER, C.I. 1927. The Silurian Inlier of Woolhope (Herefordshire). Quart. J. Geol. Soc.
Lond., 83 : 501-529.

Hate, J. 1854. Description des Bryozaires Fossiles de la Formation Jurassique. Mém.
Soc. géol. Fr., Paris (2) 5 : 157-218.

HarMER, S.H. 1913. Recent Work on Polyzoa. Proc. Linn. Soc. Lond., 143: 113-168.

HeEbE, J. E. 1921. Gottlands Silurstratigrafi. Sver. geol. Unders., Stockholm (C) 14: 1-100.

Lanc, W. D. 1906. The Reptant Eleid Polyzoa. Geol. Mag. Lond. (5) 3: 60-69.

Linpstrom, G. 1885. List of the Fossils of the Upper Silurian Formation of Gotland. Stock-
holm: I-20.

LonsDALE, W. 1839. In R. I. Murchison. The Silurian System. P. 2: 675-694, pls. 15-16
bis.

OakLEy, K. P. 1934. Phosphatic Calculi in Silurian Polyzoa. Proc. R. Soc., London (B)

116 : 296-314, pls. 12-14.

1938. Silurian Pearl-bearing Polyzoa (Bryozoa). University of London Ph.D. Thesis,
May 1938.

QUENSTEDT, F. A. 1878-81. Petrefactenkunde Deutschlands., 6: Korallen (Rohren-und-
Sternkovallen). 1093 pp., 42 pls. Leipzig.

SALTER, J. W. 1873. A Catalogue of the Collection of Cambrian and Silurian Fossils in the
Geological Museum of the University of Cambridge. ix—xiviii + 204 pp. Cambridge.
SoLLAs, W. J. 1879. On the Silurian District of Rhymney and Pen-y-lan, Cardiff. Quart. J.

Geol. Soc. Lond., 35 : 475-507, pl. 14.

STEINMANN, G. 1889. Ueber Schalen-und Kalksteinbildun. Ber. Naturf. ges. Freiburg. i. B.,
4 : 288-293.

Urricn, E.O. 1890. Paleontology of Illinois, sect. 6. Geol. Suv. Illinois, 8 : 287-688, pls.

29-78.

1893. On Lower Silurian Bryozoa of Minnesota. In The Geological and Natural History

Survey of Minnesota, 3 : 96-332, pls. 1-28.

ne

PACA Bye

Fic. 1. Section of spherule with nucelus resembling polyzoan ‘secondary embryo’ (cf.
Borg 1926: pl. 14, 90). Taken from section of Favositella interpuncta (Quenstedt). x 420.
Wenlock Limestone; Dudley. Brit. Mus. (N.H.) D.33702.

Fic. 2. Group of polyzoan pearls. 50 (approx.). Extracted from specimen of Favositella
interpuncta (Quenstedt) from Wenlock Limestone ; Rumney. Brit. Mus. (N.H.) D.36473.

Fic. 3. Favositella interpuncta (Quenstedt). Vertical section showing typical position of
pearl-like spherules within a zooecium. x60. Wenlock Limestone; Dudley. Brit. Mus.
(N.H.) D.33702.

Fic. 4. Favositella interpuncta (Quenstedt) forma typica. Inferior aspect of Fig. 5,
showing shell of Loxonema over which the polyzoan has grown.

Fic. 5. Favositella interpuncta (Quenstedt) forma typica. Lateral aspect of ‘ drum-
linoid’ zoarium. x1. Wenlock Limestone; Dudley. MHolcroft Coll. 370, Geol. Dept.,
Birmingham University.

Fic. 6. Favositella interpuncta (Quenstedt) forma typica. Lateral aspect of tumular
zoarium with conical elevation, showing monticulose maculae; originally figured by Etheridge &
Foord (1884, pl. 17, fig. 1). x1. Wenlock Limestone: Dudley (erroneously catalogued as
from Benthall Edge). Brit. Mus. (N.H.) R.1186.

PAGE el

Bull. B.M. (N.H.) Geol. 14, 1

GEOL. 14, 1.

PLATE 2

Fic. 1. Favositella interpuncta (Quenstedt). Lectotype. Lateral aspect. x2. Wen-
lock Limestone; Dudley. Geolog.-Palaontologisches Institut, Tiibingen.

Fic. 2. Favositella interpuncta (Quenstedt) forma typica. Vertical section through
‘“drumlinoid’ zoarium showing hollow base. x1. Wenlock Limestone; Dudley. Brit.
Mus. (N.H.) D.33701.

Fic. 3. Favositella interpuncta (Quenstedt) forma typica. Largetumular zoarium. x1.
Wenlock Limestone; Dudley. Brit. Mus. (N.H.) D.36340.

Fic. 4. Favositella squamata (Lonsdale). Inferior surface of zoarium showing epithecated
marginal expansion. 1. Wenlock Limestone; Dudley. Brit. Mus. (N.H.) D.33711.

Fic. 5. Favositella gotlandica sp. nov. Holotype, zoarium encrusting shell of Meristina
tumida (Dal.). 1. Miilde-margelsten; nr. Frojel, Gotland. Brit. Mus. (N.H.) D.33919.

Fic. 6. Favositella interpuncta (Quenstedt). Lectotype. Inferior surface showing
corrugated epitheca. 2. Wenlock Limestone; Dudley. Geolog.-Palaontologisches Institut,
Tiibingen.

Fic. 7. Favositella interpuncta (Quenstedt) forma brevipora. Zoarium encrusting shell
of Meristina tumida (Dalman). 3/2. Wenlock Limestone; Dudley. Greenough
Coll. 028, Geol. Dept., Univ. College, London.

Bull. B.M. (N.H.) Geol. 14, 1 PLATE 2

GEOL. 14, I. 4§

PLATE 3

Fic. 1. Favositella anolotichoides sp. nov. Paratype 4:5. Wenlock Limestone;
Coate’s Farm Quarry, Much Wenlock. Brit. Mus. (N.H.) D.36325.

Fic. 2. Favositella squamata (Lonsdale). Celluliferous surface of zoarium. 2. Wen-
lock Limestone; Dudley. Brit. Mus. (N.H.) R.2592.

Fic. 3. Favositella anolotichoides sp. nov. Holotype. 2. Wenlock Limestone,
Middle Nodular Beds; Wren’s Nest, Dudley. Brit. Mus. (N.H.) D.33926a.

Fic. 4. Favositella interpuncta (Quenstedt) forma intermedia. Zoarium encrusting
ramose coral. 1:5. Wenlock Limestone, Middle Nodular Beds; Wren’s Nest, Dudley. Brit.
Mus. (N.H.) D.36339.

Fic. 5. Favositella interpuncta (Quenstedt) forma texturata. Zoarium encrusting shell
of Meristina tumida (Dalman.). 1. Wenlock Limestone; Dudley. Brit. Mus. (N.H.)
D.33715-

Fic. 6. Favositella interpuncta (Quenstedt) forma typica. Tumular zoarium with several
hollow maculae. 2. Wenlock Limestone; Dudley. Holcroft Coll. 570, Geol. Dept.,
Birmingham University.

PIL ANID,

Bull. B.M. (N.H.) Geol. 14, 1

PLATE 4

Fic. 1. Favositella interpuncta (Quenstedt) forma typica. Tangential section. x20.
Wenlock Limestone; Dudley. Brit. Mus. (N.H.) D.33697.

Fic. 2. Favositella interpuncta (Quenstedt) forma brevipora. Tangentialsection. x 22.
Wenlock Limestone; Dudley. Greenough Coll. 028, Geol. Dept., Univ. College, London.

Fic. 3. Favositella interpuncta (Quenstedt) forma irregularis. Tangential section show-
ing transition from immature zone, with diagonal rows of quadrangular zooecia showing sagittate
lunaria, to mature zone with rounded-polygonal zooecia of the typical forma. x20. Wenlock
Limestone; Dudley. Brit. Mus. (N.H.) D.33720.

Fic. 4. Favositella interpuncta (Quenstedt) forma texturata. Tangential section showing
sagittate lunaria. x20. Wenlock Limestone; Dudley. Brit. Mus. (N.H.) D.33672.

Fic. 5. Favositella interpuncta (Quenstedt) forma typica. Tangential section passing
through macular area in which numerous mural openings are evident. x22. Wenlock Lime-
stone; Ty Mawr Lane, Rumney. Brit. Mus. (N.H.) D.33686.

Fic. 6. Favositella interpuncta (Quenstedt) forma brevipora. Tangentialsection. X22.

Wenlock Limestone, Basement beds of Lower Limestone ; Daw End railway-cutting, Walsall.
Brit. Mus. (N.H.) D.36337.

Bull. B.M. (N.H.) Geol. 14, 1 PLATE 4

PLATE 5

Fic. 1. Favositella squamata (Lonsdale). Transverse section through sub-ephebic level
of zoarium. X20. Wenlock Limestone; Dudley. Brit. Mus. (N.H.) D.33718.

Fic. 2. Favositella anolotichoides sp. nov. Thin tangential section of holotype, showing
acanthopore-like granules in lunarium. 40. Wenlock Limestone, Middle Nodular Beds;
Wren’s Nest, Dudley. Brit. Mus. (N.H.) D.33926c.

Fic. 3. Favositella squamata (Lonsdale). Tangential section showing arcuate form of
lunaria. x20. Wenlock Limestone; Dudley. Brit. Mus. (N.H.) D.33714.

Fic. 4. Favositella gotlandica sp. nov. Sagittal section of holotype, showing narrow
lunarium (1) and obliquely curving zooecia. X22. Upper Silurian (Gotlandian) ; nr. Frojel,
Gotland. Brit. Mus. (N.H.) D.33923.

Fic. 5. Favositella anolotichoides sp. noy. Tangential section of holotype. x30.
Wenlock Limestone, Middle Nodular Beds; Wren’s Nest, Dudley. Brit. Mus. (N.H.) D.33926.

Fic. 6. Favositella gotlandica sp. nov. Transverse section of holotype, showing dahllite
pearls within the zooecia. 30. Upper Silurian (Gotlandian); nr. Frojel, Gotland. Brit.
Mus. (N.H.) D.33923.

Bull. B.M. (N.H.) Geol. 14, 1 PIL ANID, &

PLATE 6

Fic. 1. Favositella interpuncta (Quenstedt) forma brevipora. Vertical section showing
two superimposed layers of zooecia. 22. Wenlock Limestone, Basement beds of Lower
Limestone ; Daw End railway-cutting, Walsall. Brit. Mus. (N.H.) D.36338.

Fic. 2. Favositella interpuncta (Quenstedt) forma brevipora. Vertical section showing
two layers of zooecia. 22. Wenlock Limestone; Dudley. Greenough Coll. 028, Geol. Dept.,
Univ. College, London.

Fic. 3. Favositella interpuncta (Quenstedt) forma typica. Vertical section cutting
prostrate bases of zooecia longitudinally. x20. Wenlock Limestone; Dudley. Brit. Mus.
(N.H.) D.33689.

Fic. 4. Favositella interpuncta (Quenstedt) forma typica? Vertical section showing
contiguous zooecia with crenulate walls ; a pearl is visible on bottom right of mid-line. The
upper surface of the section is on the left. 22. Wenlock Limestone; Ty Mawr Lane,
Rumney. Brit. Mus. (N.H.) D.33696.

Fic. 5. Favositella interpuncta (Quenstedt) forma typica. Vertical section through
mature zone. 20. Wenlock Limestone; Dudley. Brit. Mus. (N.H.) D.33696.

Bull. B.M. (N.H.) Geol. 14, | leIb a IIs,

PLATE 7

Fic. 1. Favositella interpuncta (Quenstedt) forma irregularis. Vertical section of
zoarium showing local rejuvenation of zoarial surface: layer of zooecia in texturata condition
(cf. Pl. 7, fig. 2) overlying normally matured zooecia. 20. Wenlock Limestone; Dudley.
Sedgwick Museum, Cambridge. A58o1d.

Fic. 2. Favositella interpuncta (Quenstedt) forma texturata. Vertical section. 20.
Wenlock Limestone; Dudley. Brit. Mus. (N.H.) D.33670.

Fic.3. Favositella anolotichoides sp. noy. Vertical section of holotype, showing charac-
teristic mesopores. 30. Wenlock Limestone, Middle Nodular Beds; Wren’s Nest, Dudley.
Brit. Mus. (N.H.) D.33926d.

Fic. 4. Favositella anolotichoidessp.nov. Vertical section of holotype, showing unusually
vesicular mesopores. X22. Brit. Mus. (N.H.) D.33926b.

Fic. 5. Favositella squamata (Lonsdale). Vertical section through double-layered
zoarium; in the lower layer the prostrate portions of the zooecia are cut longitudinally, in the
upper, transversely. x20. Wenlock Limestone; Dudley. Brit. Mus. (N.H.) D.33712.

Bull. B.M. (N.H.) Geol. 14, 1 IPL de EIB: 7)

PLATE 8

Fic. 1. Favositella interpuncta (Quenstedt). Lectotype. Typical portion of celluliferous
surface showing rounded thick-walled zooecial apertures; hollow stellate maculaincentre. XTIo.
Wenlock Limestone ; Dudley. Geolog.-Palaontologisches Institut, Tiibingen.

Fic. 2. Favositella squamata (Lonsdale). Celluliferous surface of a typical zoarium;
area with abundant mesopores and repressed lunaria in top right-hand corner. 8. Wenlock
Limestone; Dudley. Holcroft Coll. 517, Geol. Dept., Birmingham University.

Fic. 3. Favositella interpuncta (Quenstedt) forma texturata. Celluliferous surface. X10.
Wenlock Limestone; Dudley. Brit. Mus. (N.H.) D.33670.

Fic. 4. Favositella interpuncta (Quenstedt). Lectotype. Rejuvenated portion of celluli-
ferous surface showing thin-walled, quadrangular zooecia with thin, hood-like lunaria; a macula
is visible at the centre. X10. Wenlock Limestone; Dudley. Geolog.-Palaontologisches
Institut, Tubingen.

PEALE S

Bull. B.M. (N.H.) Geol. 14, 1

“ » * ¥ cS 73
- + ey & 1 * *
t. gs fd x a les , shy
. eS 7 Hh pet
, . © a! ; Er
a P ‘ my og
+

<! *

i.
a he
hone |
La the

ie

PLATE 9

Fic. 1. Favositella gotlandica sp. nov. Celluliferous surface of holotype. x6. Upper
Silurian (Gotlandian) ; nr. Fréjel, Gotland. Brit. Mus. (N.H.) D. 33919.

Fic. 2. Favositella gotlandica sp. nov. Ditto. x6.

Fic. 3. Favositella anolotichoides sp. nov. Celluliferous surface of holotype. x5.
Wenlock Limestone, Middle Nodular Beds ; Wren’s Nest, Dudley. Brit. Mus. (N.H.) D.33926.

Fic. 4. Favositella squamata (Lonsdale). Celluliferous surface. <6. Wenlock Lime-
stone; Dudley. MHolcroft Coll. 182, Geol. Dept., Birmingham University.

Fic. 5. Favositella squamata (Lonsdale). Celluliferous surface of lectotype. x4.
Wenlock Limestone; Dudley. Geol. Soc. Coll. 6596, Geol. Surv. Mus.

Fic. 6. Favositella interpuncta (Quenstedt) forma typica. Portion of surface of Pl. 3,
fig. 6, enlarged to show hollow maculae (cf. brood-chambers in Neofungella ; Borg 1933; pl. 2
fig. 4). Wenlock Limestone; Dudley. Holcroft Coll. 570, Geol. Dept., Birmingham University.

Fic. 7. Favositella interpuncta (Quenstedt) forma brevipora. Celluliferous surface of
zoarium (on reverse aspect of specimen illustrated in Pl. 2, fig. 7). 4. Wenlock Limestone;
Dudley. Greenough Coll. 028, Geol. Dept., Univ. College, London.

PIL AN IES,

Bull. B.M. (N.H.) Geol. 14, 1

i oes

a aly

Ya
4 wh
“ a

a b J] t : :
: +0 noe, waned am ee sa ae ne
a ~ a _ re 7 ; 2 $ _ ma _
i . ‘ = > 7 ; 7 +

: ha Semin ee
er al

PRINTED IN GREAT
By ADLARD & SON
BARTHOLOMEW PRESS

8 FEB 1967

‘
ees,

a Teer er Oe =
_ ee a ttre eee pe ee
\ bx ,

‘BULLETIN OF
HH MUSEUM (NATURAL HISTORY)

Vol. 14 No. 2
Penna: 1967

8 FEB!
fab BATHONIAN UPPER ESTUARINE SERIES CH wit
OF EASTERN ENGLAND

PAR Sh Os tTRACOD

BY

RAYMOND HOLMES BATE |

Pp. 21-66; 22 Plates; 1 Text-figure

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)
GEOLOGY Vol. 14 No. 2
LONDON : 1967

THE BULLETIN OF THE BRITISH MUSEUM
(NATURAL HISTORY), instituted im 1949, 1s
issued in five series corresponding to the Departments
of the Museum, and an Historical series.

Parts will appear at irregular intervals as they become
veady. Volumes will contain about three or four
hundred pages, and will not necessarily be completed
within one calendar year.

In 10965 a separate supplementary series of longer
papers was instituted, numbered serially for each
Department.

This paper is Vol. 14, No. 2 of the Geological
(Palaeontological) series. The abbreviated titles of
periodicals cited follow those of the World List of
Scientific Periodicals.

World List abbreviation: Bull. Br. Mus. nat.
Hist. Geology.

© Trustees of the British Museum (Natural History) 1967

TRUSTEES OF
THE BRITISH MUSEUM (NATURAL HISTORY)

Issued 8 February, 1967 Price f2-i0s.

EE AT HONIAN, UPPER ESTUARINE SERIES
OR SEAS EERN ENGLAND

PARE Iie OSTRAC ODA
By R. H. BATE

CONTENTS

Page

I. INTRODUCTION AND ACKNOWLEDGMENTS . : : : : ; 26

II. SYSTEMATIC DESCRIPTIONS . : 2 : . : ; 6 AS
Order Popocopipa Miiller ; : : ¢ ‘ : i 2
Suborder Podocopina Sars . : : : 5 : — 2
Superfamily CyPRIDACEA Baird b . : 2 é 2
Family PARACYPRIDIDAE Sars ‘ : : : 4 2
Genus PARACYPRIS . : 2
Paracypris terraefullonica (Jones & Sheeborn) 2

Superfamily DarwinuLacwa Brady & Norman . j > Ae)

Family DARWINULIDEA Brady & Norman 5 F _) 28

Genus DARWINULA Brady & Robertson : : 28

Darwinula incurva sp. nov. 5 , > 28

Superfamily CyTHERACEA Baird ‘ ; 9 2 5 2)

Family LIMNOCYTHERIDAE Klie_. 3 : 29

Subfamily TIMIRIASEVIINAE Mandelstam . 5 BQ)

Genus BISULCOCYPRIS Pinto & Sanguinetti =) 29

Bisulcocypris anglica sp. nov. . ¢ : 30

Bisulcocypris ancastevensis sp.nov... 7 32

Family ByTHOCYTHERIDAE Sars. c 5 : we SS}

Genus MONOCERATINA Roth . : : = 33

Monoceratina scarboroughensis Bate . 4633

Family CyTHERIDEIDAE Sars : ; : : 3: #333

Subfamily CyTHERIDEINAE Sars. q : 2 o 38

Genus FABANELLA Martin. ‘ . : 6 1 835}

Fabanella bathonica (Oertli) : 33

Subfamily GALLIAECYTHERIDEINAE Andreev & Mandelstam. 34

Genus GALLIAECYTHERIDEA Oertli. ; 3, 34

Galliaecytheridea? kingscliffensis sp. nov. 5 34

Genus BELEKOCYTHERIDEA nov. . j - 36

Belekocytheridea punctata sp. nov. : 5 6

Genus PICHOTTIA Oerth . : : ; fl 7

Pichottia magnamuris sp.nov. . : 5, 38

Family SCHULERIDEIDAE Mandelstam . 3 : He $39

Subfamily SCHULERIDEINAE Mandelstam . < 5 BIS)

Genus SCHULERIDEA Swartz & Swain F a3

Subgenus EOSCHULERIDEA nov. . 41

Schuleridea (Eoschuleridea) bathonica sp.

nov. . : : eh

Genus PRAESCHULERIDEA Bate : ‘ 5 Ae

Praeschuleridea quadrata sp. nov. 5 eA 2

GEOL, 14, 2. 5

24 BATHONIAN UPPER ESTUARINE OSTRACODA, I

Family CyTHERURIDAE Miiller

Genus METACYTHEROPTERON Oertli

Metacytheropteron drupacea (Jones)
Family PROGONOCYTHERIDAE Sylvester-Bradley
Subfamily PROGONOCYTHERINAE Sylvester-Bradley

Genus PROGONOCYTHERE Sylvester- ne
Progonocythere levigata sp. nov.
Progonocythere rugosa sp. nov.
Progonocythere triquetva sp. Nov. .

Genus GLY PTOCYTHERE Brand & Malz
Glyptocythere guembeliana (Jones)
Glyptocythere juglandica one)

Genus KLIEANA Martin
Klieana levis Oertli

Genus LOPHOCYTHERE Sylvester- eraaley
Lophocythere ostveata (Jones & Sherborn)
Lophocythere scabra scabra Triebel
Lophocythere septicostata sp. nov.
Lophocythere transversiplicata sp. nov. .

Genus MACRODENTINA Martin .

Subgenus MEDIODENTINA nov. : 5
Macrodentina (Mediodentina) bathonica
sp. nov. :

Genus MARSLATOURELLA Malz
Marslatourella bullata sp. nov.

Genus MICROPNEUMATOCYTHERE Bate .
Micropneumatocythere postrotunda sp. nov.
Micropneumatocythere quadrata sp. nov.
Micropneumatocythere subconcentrica (Jones)

Family TRACHYLEBERIDIDAE Sylvester-Bradley
Subfamily TRACHYLEBERIDINAE Sylvester-Bradley
Genus OLIGOCYTHEREIS Sylvester-Bradley
Oligocythereis sie Oe & ee
Family Uncertain :
Genus PLATYCYTHERE nov.
Platycythere verriculata sp. 1 nov.
III. PALAEOECOLOGY .
IV. REFERENCES

SYNOPSIS

Twenty-nine species of Ostracoda, of which 18 are new, are described from the Bathonian
Upper Estuarine Series of Eastern England. Of this ostracod fauna two genera, Belekocytheridea
and Platycythere, and two subgenera, Eoschuleridea and Mediodentina are new.
ecological study of the ostracod faunas indicates an alternation throughout the succession of
marine and freshwater conditions. Of the two new genera, Belekocytheridea inhabits a brackish
water environment and Platycytheve a marine to brackish water environment. Of the two new
subgenera, Eoschuleridea is marine whilst Mediodentina is euryhaline, ranging from marine

through brackish to almost freshwater conditions.

A palaeo-

BATHONIAN UPPER ESTUARINE OSTRACODA, I

NORTH

York SEA

scale: one inch to

sLincoln solmiles

eAncaster

*Ketton
e °King's Cliffe

PK ettering(C.St.J.)

Dane Hill
Oxford

yy)

Fic. t. Outcrop of Middle and Upper Jurassic Rocks in Eastern and Northeastern
England with localities sampled within the Upper Estuarine Series shown.

26 BATHONIAN UPPER ESTUARINE OSTRACODA, I

I. INTRODUCTION AND ACKNOWLEDGMENTS

STRETCHING from North Lincolnshire to the borders of Oxfordshire is a narrow strip
of marine and freshwater clays, with some thin limestones, showing evidence of
deposition in shallow water. These beds, known as the Upper Estuarine Series,
underlie the rubbly limestones of the Bathonian Great Oolite Limestone and with a
marked unconformity overlie the Bajocian Lincolnshire Limestone. They form
the basal Bathonian sediments in the area.

The absence of ammonites from the Upper Estuarine Series makes the dating of
these sediments rather difficult. However, mapping and the examination of the
invertebrate faunas suggests that they may be equivalent to the Hampen Marly
Beds of further south. Certainly the ostracod faunas would not, at this stage,
disagree with this. If this correlation is to be regarded as correct, the Upper
Estuarine Series would be Middle Bathonian in age and belong to the zone of Tulites
subcontractus (Morris & Lycett).

Although of shallow water origin, there is evidence to suggest that these beds are
not estuarine deposits in the strictest sense. Dr. C. J. Aslin of the University of
East Anglia has for some time now been working on the stratigraphical and sedimen-
tological problems of the Upper Estuarine Series, and it was he who made available
his large collection of ostracods which form the basis of the present paper.

The sections from which the ostracods were obtained occur in large quarries where
the Upper Estuarine Series represents the overburden. The continual cutting back
of these sediments makes it impossible to maintain a permanent section. The
quarries are located as follows (see Text-fig. I):

Kings Cliffe—TL/o12966.

Ancaster (Thompson’s Pit)—SK_/992409.

Ketton (Portland Cement OQuarry)—SK/972057-09.
Cranford St. John (Kettering)—SP/926764.

Dane Hill—SP/465273.

It should be noted that in some cases the quarries are sited a mile or so from the
town whose name they bear.

The morphological terms used in the present paper are taken from Sylvester-
Bradley (1956), Moore (1961) and Bate (1963). All the ostracods described in the
text have been deposited in the collections of ti » Palaeontology Department, British
Museum (Natural History).

I should like to record my grateful thanks to Dr. H. J. Oertli, S.N.P.A., Pau,
France, and to Dr. H. Malz, Senckenberg Museum, Frankfurt, Germany, for the loan
of material which proved invaluable for comparison with the present fauna. Mr.
S. H. Eagar printed the photographs originally taken by myself.

II. SYSTEMATIC DESCRIPTIONS

The beds from which the ostracods were obtained are identified by a letter and
refer to beds described by Aslin (in press).

BATHONIAN UPPER ESTUARINE OSTRACODA, I 27
Order PODOCOPIDA Miiller 1894

. Suborder PODOCOPINA Sars 1866
Superfamily CYPRIDACEA Baird 1845
Family PARACYPRIDIDAE Sars 1923

Genus PARACYPRIS Sars 1866

Paracypris terraefullonica (Jones & Sherborn)

(Pl. 1, figs 1-6)
1888 Macrocypris terrae-fullonicae Jones & Sherborn : 252, pl. 5, figs. 3a-c.
1888 Macrocypris horatiana Jones & Sherborn : 252, pl. 5, figs. 2a—c.

Diacnosis. Paracypris with elongate carapace, posteriorly acuminate. Anterior
rounded. Ventral margin almost straight in the larger left valve, more strongly
concave in right. Dorsal margin arched with antero-dorsal slope tending to be
slightly concave, more noticeably so in right valve. Shell surface smooth. Anterior
and posterior vestibules well developed. Radial pore canals branching.

LectrotyPe. Selected here, 1.1875, left valve, from the Blue Fullers-earth clay;
Midford near Bath, figured Jones & Sherborn 1888, pl. 5, figs. 3a—c.

OTHER MATERIAL. 1.1874, right valve, from the Blue Fullers-earth clay; Midford
near Bath, figured Jones & Sherborn 1888, pl. 5. figs. 2a-c. lo.2250-58, from the
Upper Estuarine Series, beds M, R & S, Kings Cliffe; base of bed H: Ketton and
bed I, Kettering.

DESCRIPTION. Carapace elongate, subreniform, rounded anteriorly, acuminate
posteriorly. Dorsal margin arched, convex in the right valve, almost straight,
sloping posteriorly in the left. Cardinal angles rounded. Antero-dorsal slope tends
to be slightly concave, particularly just behind the anterior margin, a situation more
noticeable in the right valve. Ventral margin almost straight in the left valve,
concave medially in the right. Greatest height of carapace in anterior third in the
left valve, almost median in the right. Greatest length below mid-point; greatest
width median. Shell surface smooth. Left valve larger than the right which it
overlaps along the ventral, postero-dorsal and antero-dorsal slopes. Posteriorly the
left valve slightly over-reaches the right whilst anteriorly the right valve over-reaches
the left. Hinge consists of a simple groove in the left valve into which the dorsal
edge of the right valve fits. Inner margin and line of concrescence do not
coincide terminally, prominent vestibules being produced. Anteriorly the vestibule
is broad whilst posteriorly more narrow, extending along the postero-ventral margin
of the inner part of the valve up to the centre of the median incurvature. Radial
pore canals not clearly observed, but appear to be few in number and antero-
ventrally can be seen to be branching. Three centrally situated oval muscle scars
can be distinguished in the lectotype with a fourth situated behind.

28 BATHONIAN UPPER ESTUARINE OSTRACODA, I

Dimensions. Lectotype. 1.1875, left valve, length 0-60 mm., height 0-28 mm.
Other material. 1.1874, right valve, length 0:58 mm., height 0:26mm. Io.2250,
carapace, length 0-65 mm., height 0:31 mm., width 0:27 mm. I[o.2251, right valve
length 0-65 mm., height 0-31 mm. I[o.2255, right valve, length 0-65 mm., height
031mm. I[o0.2256, carapace, length 0:60 mm., height 0:29 mm., width 0-24 mm.
Io. 2257, right valve, length 0°61 mm., height 0°27 mm.

Remarks. Paracypris terraefullonica was originally described by Jones &
Sherborn (1888 : 252) as Macrocypris terrae-fullonicae. Also described in the same
publication is the ostracod M. horatiana which occupies a position within the publica-
tion previous to that of M. terrae-fullonicae. Both these species are considered to
be synonymous. The decision to select M. terrae-fullonicae as the type was influenced
by the better preservation of the specimen available and the preference in name,
indicating as it does, derivation from the Bathonian Fullers-Earth.

Paracypris terraefullonica is similar to Paracypris? sp. A. Schmidt (1955 : 52) but
differs posteriorly. In P.? sp. A. the posterior margin is close to the posterior cardinal
angle, whilst that part of the valve behind the cardinal angle is more elongate. P.
bajociana Bate (1963 : 186, pl. 2, figs. 1-8) has a much longer and more straight
dorsal margin.

Superfamily DARWINULACEA Brady & Norman 1889
Family DARWINULIDAE Brady & Norman 1889
Genus DARWINULA Brady & Robertson 1885

REMARKS. Darwinula stevensoni, the type species, has the right valve larger than
the left, a feature given as characteristic of the genus in the Treatise on Invertebrate
Paleontology (Moore 1961 : 254). Several species of Darwinula are known, however,
to possess a carapace in which the left valve is the larger. For example, the Recent
D. daps Harding (1962 : 60, figs. 45-53) and the Purbeckian D. leguminella (Forbes
mm Lyell 1855 : 294, text-fig. 334c). Many more species are so described in the
literature. D. incurva sp. nov., described below also possesses a carapace in which
the left valve is the larger.

Darwinula incurva sp. nov.
(Pl. 1, figs. 7-12)
1965 Darwinula sp.A. Bate: 751, pl. 109, figs. 1-4.
Diacnosis. Darwinula of large size with broadly arched dorsum, rounded
anterior and posterior, and strongly incurved ventral margin anterior of valve centre.
Left valve strongly overlaping right along ventral margin and around posterior.

Anteriorly left valve over-reaching right. Shell surface finely punctate. Muscle
scars as for family.

BATHONIAN UPPER ESTUARINE OSTRACODA, I 29

HoLotyPe. l[o.2259, a complete carapace from bed QO Kings Cliffe.
PARATYPES. Io.2260~-74, from bed QO Kings Cliffe, and beds O & R Ancaster.

DESCRIPTION. Carapace oval-elongate, very finely punctate, appearing smooth
in most specimens, although the internal surface of the valves is quite strongly
punctate. Greatest length passes either through or just below mid-point, being
dependent upon the outline of the anterior margin which may be either broadly
rounded or extended forward slightly below mid-length. Posterior broadly rounded
with the greatest width situated in the posterior third. Greatest height median.
Dorsal margin arched. Ventral margin convex in the posterior half, strongly and
characteristically incurved antero-medially. Left valve larger than the right which
it strongly overlaps around the posterior and along the ventral margin. Around the
anterior the overlap is replaced by overreach though along the antero-and postero-
dorsal slopes overlap of the right valve by the left is noticeable. Muscle scars
typical of the genus, consisting of a rosette of scars which in one specimen is composed
of eleven in number. Inner margin and line of concrescence coincide anteriorly
(not seen posteriorly) where a narrow duplicature is present. The surface of the
duplicature in the left valve is serrated at right angles to the inner margin. The
purpose of this is conjecture at the moment. Anterior radial pore canals short,
straight and evenly spaced, about 20 in number. A long, narrow groove extends
along the dorsal margin of the right valve, into which the dorsal margin of the left
valve fits for purposes of articulation.

Dimensions. Holotype. lo.2259, carapace, length 1:03 mm.; height 0-43 mm.;
width 0:37 mm.

Paratypes. lIo.2260, carapace, length 1:°03mm.; height o48mm.; width
o4omm. lo.2261, left valve, length 0-83 mm.; height 0:37 mm.

REMARKS. Darwinula sp. A. Bate 1965 is a juvenile instar of the present species
and is here placed into synonymy. Comparison with other British Mesozoic
darwinulids shows D. incurva to be larger, more elongate and more strongly incurved
than previously described species. D. incurva is similar in outline to D. tubiformis
Ljubimova (1956 : 119, pl. 23, figs. ra—b.) from the Lower Cretaceous of Mongolia
but differs in not being so strongly swollen posteriorly and in being more slender in
dorsal view. D. barabinskensis Mandelstam (as figured in Ljubimova r960 : 28,
pl. 2, fig. 2) from the Lower Cretaceous of the western Siberian Lowlands (Krasnoyarsk
region) is close to D. incurva in dorsal outline but does not taper quite so much
anteriorly neither is it as incurved antero-ventrally.

Superfamily CYTHERACEA Baird 1850
Family LIMNOCYTHERIDAE Klie 1938
Subfamily TIMIRIASEVIINAE Mandelstam 1960
Genus BISULCOCYPRIS Pinto & Sanguinetti 1958

Remarks. As stated elsewhere (Bate 1965) difference of opinion exists relating

30 BATHONIAN UPPER ESTUARINE OSTRACODA, I

to the validity of the two genera Bisulcocypris and Theriosynoecum. Sylvester-
Bradley & Pinto (MS.) are currently of the opinion that within the Timiriaseviinae
there are two distinct genera, one with hollow tubercles (Theriosynoecum), the other
with or without solid tubercles (Bisulcocypris). An opinion with which I am in
full agreement.

The presence of tubercles amongst these fresh to brackish water ostracods would
appear to be either a genotypic or a phenotypic character. In those species which
are normally without such ornamentation but which occasionally show it—it is
conceivable that the presence of tubercles is in some way controlled by the environ-
ment (pH, salinity etc.) and as such would be a phenotypic character. There appears
to be a definite division between those species which show this possible phenotypic
variation and those in which the tubercles are genotypically controlled. The genus
Theriosynoecum is considered to fall into the latter category, whilst in Bisulcocypris
both phenotypic and genotypic tuberculate species occur. A genotypic species is
one in which tubercles are present through all the ontogenetic stages, whilst a pheno-
typic species only shows the development of tubercles or nodes in the adult stage,
and then, as appears to be often the case, in only one dimorph. In B. anglica sp.

nov., for example, the nodes are characteristically developed in the male dimorph
only.

Bisulcocypris anglica sp. nov.
(Pl. 2, figs. 1-1T)
1965 Bisulcocypris sp. A., Bate : 753, pl. 109, figs. 13, 14.

Diacnosis. Bisulcocypris, oval/elongate in side view, tapering slightly to anterior.
Posterior tapered in female dimorph. Greatest height at posterior cardinal angle.
Ventro-lateral margin strongly convex, overhanging ventral surface. Two, short,
crescentic sulci extend down from dorsal margin in anterior half of valve. Shell
surface laterally punctate to reticulate. Ventro-lateral and ventral surfaces with
strong lateral ridges. Slender males may possess small nodes in posterior half;
females without nodes. Female with characteristic narrow posterior, produced by
steeply sloping posterodorsal margin of posterior swelling. Male tending to have
broader (height) posterior. Hinge lophodont. Valves equivalve.

HoLotyPe. lo.2275, female carapace from bed QO, Kings Cliffe.
PARATYPES. Io0.2276-99. Locality as above.

DESCRIPTION. Carapace oval/elongate with the greatest height being in the
posterior half at the posterior cardinal angle in adults but is situated at the anterior
cardinal angle in juveniles. Greatest length through or slightly below mid-point.
Greatest width in the posterior half in female dimorphs, median in the male.
Posterior broadly rounded in the male dimorph whilst in the posteriorly swollen
female the posterior becomes much narrower (height) due to the steeply sloping
dorsal surface of the posterior swelling. Posterior cardinal angle sharply angled in
both sexes, anterior cardinal angle shallow, broadly rounded sweeping down to the

BATHONIAN UPPER ESTUARINE OSTRACODA, I 31

narrowing anterior margin. Dorsal margin medially concave in the female dimorph,
but less noticeably so in the male. Ventro-lateral margin strongly convex, over-
hanging the ventral surface. Shell surface strongly punctate to reticulate. Promi-
nent lateral ridges extend along the ventro-lateral and ventral surfaces. In the male
dimorph a number of small nodes (usually two in number) may occur in the posterior
half of the carapace. One node is positioned just below and behind valve centre
with the second node a short distance below and in front of this. In a single male,
two smaller nodes occur dorsal to these. Two very small nodes are situated at the
extreme posterior and a number around the anterior margin of both sexes. These
are the only ones which occur in the female. Occasionally the median node described
for the male is found also in a juvenile instar. Juveniles tend to be rather square
in outline. Two, shallow, rather concentric sulci extend for a short distance only
from the dorsal margin. Both sulci are situated in the anterior half of the carapace.
Both left and right valves are of equal size. Along the ventral margin there is no
overlap; postero-ventrally the right valve over-reaches the left whilst at the position
of the antero-median incurvature the left over-reaches the right. Along the antero-
dorsal slope the left valve progressively over-reaches the right with the maximum
over-reach being at the anterior cardinal angle. Behind this angle the right valve
strongly over-reaches the left for a short distance. Very slight overlap of the right
by the left occurs at the posterior cardinal angle. Muscle scars are situated low
down on the lateral part of the carapace below the first more medially situated sulcus,
and consist of a backwardly sloping row of four adductor scars. Muscle scars
anterior to these have not been observed. Hinge lophodont, only observed in the
right valve where the median groove is long and rather broad. The anterior tooth
is large and blade-like, being an enlarged, flattened continuation of the selvage.
Posterior tooth not seen. From the dorsal view of a complete carapace it is obvious
that an accommodation groove is present in the posterior half of the left valve.
Around the anterior the inner margin and line of concrescence do not quite
coincide. The separation is not really sufficient, however, for a vestibule to be
developed. Anterior radial pore canals long and straight extending into the broad
anterior flange, approximately 20 in number.

Dimensions. Holotype. Io.2275, female carapace, length 1-00 mm.; height
058 mm.; width 0:58 mm.

Paratypes. lo.2277, male carapace, length 1-04 mm.; height 0-63 mm.; width
o5rmm. Io.2278, male carapace, length o-98mm.; height 056mm.; width
047mm. lo.2279, female carapace, length (broken) 0-98 mm.; height 0-66 mm. ;
width 068mm. Io.2280, juvenile carapace, length o61mm.; height 0-35 mm.;
width 0:28 mm.

ReMARKS. The development of nodes in this species is restricted in adults to
the male dimorphs and to a number of juvenile instars. Possibly those juveniles
which show this feature may have been immature males, although there is no way
of confirming this. Bzsulcocypris anglica is close to B. tenuwimarginata (Oerth
1957: 7605, pl. 23, figs. 15-24), specimens of which were kindly sent to me by
Dr. Oertli, but is a much larger more elongate species. The anterior half of B.

32 BATHONIAN UPPER ESTUARINE OSTRACODA, I

tenuimarginata is quite short and stubby when compared with the elongate anterior
of B. anglica. The French species does not appear to show any tendency towards
the development of lateral nodes. Bisulcocypris sp. A. described from the Bathonian
of Oxfordshire is considered to belong to this species.

Bisulcocypris ancasterensis sp. nov.
(Pl. 3, figs. 1-10)
1965 Bisulcocypris sp. B; Bate: 753, pl. 109, figs. 10-12.

Diacnosis. Bisulcocypris rectangular in outline with height at anterior cardinal
angle almost equal to that at posterior cardinal angle. Strongly bi-sulcate, dimor-
phic. Shell surface punctate/reticulate with three nodes in posterior half and two
nodes in anterior half. Smaller additional nodes occur at extreme posterior and
anterior. Both nodes and surface ornamentation more positively developed in
juvenile instars. Muscle scars as for genus.

HototyPe. lo.2282, female carapace from bed R, Ancaster.
PARATYPES. Io0.2283-5, 2300-1, from bed R, Ancaster.

DESCRIPTION. Carapace rectangular, dimorphic, rounded anteriorly and posteri-
orly. Dorsal and ventral margins in the male dimorph are almost parallel with only
a shallow concavity medially. In the female, identified by the swollen posterior
half, the dorsal margin is quite strongly concave medially. Ventro-lateral margin
in both dimorphs overhangs the ventral surface as viewed laterally. Greatest length
through mid-point with greatest height and width in posterior half in males and
females; greatest height in anterior half in juveniles. Carapace in the anterior half
bi-sulcate, the posterior sulcus being the better developed of the two; this is especially
true in the female dimorph. The adductor muscle scars which consist of an oblique
row of four scars, are situated at the base of this sulcus. Cardinal angles distinct.
Shell surface punctate to reticulate, further ornamented by a number of prominent
nodes. In the posterior half of the carapace three nodes are arranged in a triangular
pattern; a large central node with one of equal size situated in front and below this
and a smaller node situated behind and below. Smaller nodes may occur above
these and on the posterior. In the anterior half two nodes are present; one in front
of the ventro-lateral termination of the posterior sulcus and the second, much
higher up, in front of the termination of the anterior sulcus. Ventral surface
ornamented with low, longitudinal ridges. Left valve larger than the right, which
it overlaps along the ventral margin, especially mid-ventrally, and again at the
anterior and posterior cardinal angles. There is no terminal overlap. Mid-dorsally
the right valve over-reaches the left. Juvenile instars reticulate with well developed
nodes. Posteriorly the two low nodes which occur dorsally above the main group
of three in the adults are here strongly developed and are equal in dominance to the
others. Internal details have not been observed in the present material but have
been described for B. sp. B. (see Bate; 1965 : 754).

BATHONIAN UPPER ESTUARINE OSTRACODA, I 33

Dimensions. Holotype. Io.2282, female carapace, length o-g1 mm.; height
0°55 mm.; width 0-5r mm.

Paratypes. lIo.2284, male carapace, length 0:98 mm.; height 0-51 mm.; width
036mm. Io.2285, juvenile carapace, length 0-61 mm.; height 0:35 mm.; width
0:2I mm.

REMARKS. Bisulcocypris ancasterensis sp. nov., in being strongly nodose differs
at once from B. anglica sp. nov., and B. tenuimarginata (Oertli). Moreover, it may
also be distinguished from the above in outline (having a much deeper anterior than
found in B. anglica and a much more elongate anterior half than that occurring in
B. tenuimarginata) and by the more strongly developed second sulcus, never developed
to such an extent in the other two species. Bzsulcocypris sp. B. described from the
Bathonian of Oxfordshire is almost certainly conspecific with the present species,
despite its larger size.

Family BY THOCYTHERIDAE Sars 1926
Genus MONOCERATINA Roth 1928
Monoceratina scarboroughensis Bate
(Bins ige ih)

1965 Monoceratina scarboroughensis Bate : 99, pl. I, figs. 1-12.

ReMARKS. A single female carapace, of good preservation has been found in bed
B in the Kettering section. The range of this species is thus extended from the
Bajocian (blagdent in part) into the Bathonian, although here again correlation of
the horizons with known ammonite zones has not been possible.

Family CYTHERIDEIDAE Sars 1925
Subfamily CYTHERIDEINAE Sars 1925
Genus FABANELLA Martin 1961

Remarks. Although placed by Martin into the Cytheridae, Fabanella is here
considered to be more closely allied to the Cytherideidae on account of the shape and
structure of the carapace, simple pore canals and muscle scar pattern. Although
the antennal scar is not so strongly U-shaped as for the nominal genus Martin’s
illustration (pl. 1, fig. 50) for the type species indicates that the arrangement is the
same and belongs to Type B (Bate 1963 : 181, figs. 8—ro).

Fabanella bathonica (Oertli)
(Pl. 4, figs. 1-5)

34 BATHONIAN UPPER ESTUARINE OSTRACODA, I

1957 Cyprideis? bathonica Oertli: 758, pl. 21, figs. 12-20, pl. 22, figs. 1-6, 11, 12, (Non. figs.
7, 8, ? figs. 9, Io).
1963 Fabanella bathonica (Oertli) Oertli, pl. 28 (2), fig. m.
MATERIAL. Specimens registered in the collections: Io.2303-15 from bed R,
Kings Cliffe and bed U, Ancaster.

Remarks. LFabanella bathonica does not occur generally throughout the Upper
Estuarine Series, being restricted to only a few horizons. However, when it does
occur it is a common species amongst a marine to brackish-water fauna. Specimens
of this species were kindly sent to me for comparison purposes by Dr. Oertli.

Subfamily GALLIAECYTHERIDEINAE Andreev & Mandelstam 1964

REMARKS. The subfamily Galliaecytherideinae introduced by Andreev &
Mandelstam contains genera which although possessing some external resemblance
to those placed in the Schulerideidae are nevertheless distinguished by their possession
of a type A muscle scar pattern and simple radial pore canals. The genus Gal-
liaecytheidea Oertli (1957 : 654) was placed into the Schulerideidae in error by Bate
(1963 : 207), a situation which is here corrected.

Andreev & Mandelstam include within their new subfamily the following genera:
Galliaecytheridea Oertli, Lyubimovina Neale, Palaeocytheridella Mandelstam, A sciocy-
there Swain, Rubracea Mandelstam and Procytheridea Peterson. In the present paper,
however, Asciocythere, Procytheridea and Rubracea are not considered to belong here.
Belekocytheridea gen.nov. and Pichottia Oertli are considered to belong to this sub-
family. Van Morkhoven (1963 : 307) considers the genus Palaeocytheridella Mandel-
stam to be a junior synonym of Vernoniella Oertli.

Subfamily GALLIAECYTHERIDEINAE Andreev & Mandelstam 1964
Genus GALLIAECYTHERIDEA Oertli 1957
Galliaecytheridea? kingscliffensis sp. nov.

(Pl. 4, figs. 6-12; Pl. 5, figs. 1-8)

Dracnosis. Galliaecytheridea?, dimorphic: females plump, subquadrate with
backwardly sloping dorsal margin and concave postero-dorsal margin, producing
short, upturned posterior. Males elongate. Both sexes with compressed anterior
marginal border. Shell surface very finely punctate. Hinge entomodont. Muscle
scars type A. Radial pore canals, straight, widely spaced; g anteriorly; 4
posteriorly.

HorotyPe. Io.2316, female carapace, bed S, Kings Cliffe.
PARATYPES. I0.2317-27, beds J & S, Kings Cliffe and bed W, Ancaster.

DEscriPTION. Carapace plump and subquadrate in outline with triangular-
shaped posterior, slightly upturned, in the female dimorph. Males more elongate

BATHONIAN UPPER ESTUARINE OSTRACODA, I 35

in outline but otherwise similar. Greatest length of the carapace passes through
mid-point and the greatest height in the anterior half; through the anterior cardinal
angle in the female but just behind this in the male. Greatest width median. In
dorsal view the carapace is noticeably convex, especially in the female, with
compressed anterior and posterior marginal borders. Dorsal margin straight or
very slightly convex, sloping to the posterior, again more noticeably so in the female.
Cardinal angles prominent. Anterior broadly rounded; posterior broadly tri-
angular with a concave postero-dorsal slope producing a slight upturning. Ventral
margin medially incurved, obscured in a whole carapace by the median convexity
of the ventro-lateral margin. Shell surface very finely punctate with small, widely
scattered, circular normal pore canal openings. There is no definite eye swelling,
although a short, diagonal furrow, as found in such genera as Praeschuleridea and
Schuleridea, occurs below the anterior cardinal angle. Here, however the groove
may be seen in both valves. In the previously mentioned genera it is generally a
feature of the right valve only. Left valve larger than the right which it overlaps
along the ventral margin and over-reaches along the antero-, and postero-dorsal
and dorsal margins. Terminally there is neither overlap nor over-reach. Hinge
entomodont: left valve with terminal, elongate, loculate sockets separated by a
strong median bar which is dentate along its length. The denticles of the median
bar are very much enlarged in the anterior section. Above the median element a
broad, deep, accommodation groove is developed. In the right valve there are six
posterior teeth and probably a similar number anteriorly although damaged in the
present material. The median groove which is loculate is expanded considerably
in its anterior section. Muscle scars consist of a vertical row of four oval adductor
scars with a single, round antero-dorsal antennal scar which is situated opposite the
top two adductor scars. Mandibular scar crescent-shaped and antero-ventral in
position. This muscle scar pattern thus falls into type A. Inner margin and
line of concrescence coincide, producing a duplicature of moderate width.
Radial pore canals straight and widely spaced, nine anteriorly and approximately
four posteriorly. In single valves, the incurved part of the ventral margin can be
seen to be obscured ventrally by a flattened “ lip’ convex downwards.

Dimensions. Holotype. Io.2316, female carapace, length 0-62 mm.; height
0-45 mm.; width 0:36 mm.

Paratypes. lo.2317, female carapace, length 0-64 mm.; height 0-44 mm.; width
035mm. Io.2318, male carapace, length o-77Imm.; height 0:43 mm.; width
035mm. I[o.2321, female left valve, length 0-64 mm.; height 0-45 mm. _ Io.2322,
male left valve, length 069 mm.; height 043mm. I[o.2326, female right valve,
length 0:58 mm.; height 0:36 mm.

ReMArRKS. Galliaecytheridea? kingscliffensis sp. nov. is placed into the genus with
a query owing to the fact that the hinge, in being entomodont, is quite unlike that
found in any other species of this genus. It is not proposed to erect a subgenus to
contain this species at this stage. The dimorphism found here is more clearly
apparent than in other members of the genus. In fact the male dimorphs bear some
resemblance to Galliaecytheridea postrotunda Oertli (1957 : 656, pl. 2, figs. 45-55)

36 BATHONIAN UPPER ESTUARINE OSTRACODA, I

although they do not taper so noticeably towards the posterior nor do they possess
a posterior spine. The female dimorphs on the other hand are close to G. dissimilis
Oertli (1957 : 655, pl. I, figs. 32-49, pl. 2, figs. 40-44) from which they may be
distinguished by their greater convexity (in dorsal view), dorsal margin with a
slightly more pronounced slope to the posterior and by the presence of an entomodont
hinge.

Genus BELEKOCYTHERIDEA nov.

TYPE SPECIES. Belekocytheridea punctata sp. nov.

Diacnosis. Galliaecytherideinae with bean-shaped carapace; straight, somewhat
flattened dorsal margin; antero-ventral and postero-ventral margins characteristically
convex, projecting slightly below line of ventral surface. Antero-dorsal slope short,
passing into uniformly rounded anterior. Postero-dorsal slope long, slightly convex.
Cardinal angles prominent. Posterior narrowly rounded with greatest length passing
below mid-point. Left valve larger than right. Hinge antimerodont; accommoda-
tion groove poorly developed. Inner margin and line of concrescence coincide.
Radial pore canals straight, widely spaced, approximately 9g anteriorly and 5
posteriorly. Muscle scars a vertical row of 4 oval adductors with antero-dorsal
antennal scar and antero-ventral mandibular scar (Type A).

RemMARKS. The genus Belekocytheridea may be distinguished from all other
ostracods by its characteristic, rather angular bean shaped carapace. At the
present time only the type species has been definitely placed into the genus although
a small number of specimens of a much larger species have been found within the
Bajocian Upper Lincolnshire Limestone. These will be considered in a separate
publication. The ecology of the type species suggests that the genus may be marine
to brackish water in habit. In this it resembles Lycopterocypris Mandelstam 1956,
a genus having a similar external appearance but internally distinguished by the
presence of an Adont hinge.

Belekocytheridea punctata sp. nov.

(Pl. 5, figs. 9-13, Pl. 6, figs. 1-5)

Dracnosis. Belekocytheridea with ventral and ventro-lateral surfaces ornamented
by fine, longitudinal striae which terminally pass up onto lateral surface of carapace
to parallel anterior and posterior margins. Terminally striae poorly developed,
dying out before reaching dorsal margin. Remainder of lateral surface punctate.
Species dimorphic. Hinge and muscle scars as for genus.

Ho.rotypPe. Io.2328, female carapace, bed H, Ketton.

PARATYPES. lo.2329-57, male and female carapaces and single valves, bed H,
Ketton; bed R, Ancaster and bed G, Kings Cliffe.

DeEscRIPTION. Carapace bean shaped, convex in dorsal view, with prominent

BATHONIAN UPPER ESTUARINE OSTRACODA, I 37

cardinal angles when viewed laterally. Dorsal margin straight, very slightly convex.
Antero-dorsal slope short passing into rounded anterior margin. Postero-dorsal
slope long, slightly convex and passing into narrowly rounded posterior. Ventral
margin incurved medially, overhung laterally by the convex ventro-lateral margin.
Terminally the convex antero-ventral and postero-ventral margins project slightly
below the ventral surface. Sexual dimorphism strongly apparent, the presumed
males being very much more elongate but otherwise morphologically similar to the
shorter female dimorphs. Greatest length of carapace occurs below mid-point whilst
greatest height occurs at the posterior cardinal angle and greatest width through
mid-point. Shell surface punctate laterally; terminally with weak striae paralleling
the margins, the striae extending on to the lateral surface from the ventral surface.
Normal pore canal openings circular, prominently developed and well spaced over
the lateral surface. Left valve larger than the right which it overlaps along the
ventral margin and strongly along the antero -, and postero-dorsal slopes. Terminally
and dorsally there is no overlap. Hinge antimerodont; left valve with elongate
terminal loculate sockets and a denticulate median bar. Accommodation groove
virtually absent. Right valve with approximately 5-6 terminal teeth and an
elongate locellate median bar. Inner margin and line of concrescence coincide,
duplicature being of moderate width. A narrow flange extends around the
anterior margin. Radial pore canals long and straight, widely spaced; 9 anteriorly
and 5 posteriorly. Muscle scars (Type A) with the 4 adductors in a vertical row,
the large rounded antennal scar being antero-dorsal in position. Mandibular scar
antero-ventral.

Dimensions. Holotype. Io.2328, female carapace, length 0:57 mm.; height
0-31 mm.; width 0-29 mm.

Paratypes. lIo.2329, male carapace, length 0-66 mm.; height 0:33 mm.; width
o-29mm. Io.2331, female carapace, length 0:54mm.; height 0-30 mm.; width
027mm. Io.2356, male left valve, length 0-68 mm.; height 0:38 mm.

Remarks. Belekocytheridea punctata sp. nov., is quite unlike any previously
described species and has, therefore, been placed into a new genus. The specimens,
already referred to, as occurring in the Upper Lincolnshire Limestone are quite
smooth whilst the present species has a partial ornamentation of longitudinal striae.

Genus PICHOTTIA Oertli 1959

Remarks. In the original diagnosis of Pichottia by Oertli (1959 : 115), size was
considered to be a diagnostic character. This was based upon the very small size
of the type species where the length of the female was in the order of 0:37—0-4I mm. ;
and that of the male 0:43-0:47 mm. In Pichottia magnamuris sp. nov., described
below, the length of the female dimorph is in the region of 0:61-0:69 mm.; and that
of the male 0-68-0-76 mm. A small size should not, therefore, be considered to be
diagnostic of this genus.

GEOL. 14, 2. 6

38 BATHONIAN UPPER ESTUARINE OSTRACODA, I
Pichottia magnamuris sp. nov.
(Pl. 6, figs. 6-14, Pl. 7, figs. 1-6)

Diacnosis. Pichottia, plump, strongly convex in dorsal view, oval in lateral
view. Dimorphic. Shell surface very finely punctate. Females of length o-61—
0-69 mm.; males of length 0-68—0-76 mm.

HorotyrPe. Io.2358, female carapace, bed S, Kings Cliffe.

PARATYPES. Io.2359-68, male and female carapaces and single valves from bed S,
Kings Cliffe and Dane Hill.

DESCRIPTION. Carapace plump, strongly convex in dorsal view, the valves
parting slightly at extreme anterior. Ovoid in lateral view tapering to the narrowly
rounded posterior. Male dimorphs more elongate than the females, more strongly
tapering posteriorly and not so convex in dorsal view. Anterior rounded with
long, convex antero-dorsal slope; short convex dorsal margin and long postero-
dorsal slope, strongly convex in the female, long and steeply sloping in the male.
Posterior narrowly rounded. Ventro-lateral margin convex, overhanging the ventral
surface in side view. Ventral margin antero-medially incurved. Cardinal angles
broadly rounded. Greatest length of carapace passes below mid-point. The
position of greatest height occurs at the posterior cardinal angle although the height
at the anterior cardinal angle is only very slightly less. Greatest width behind
mid-point. Left valve slightly larger than the right, which it overlaps along the
ventral margin, and along the antero-dorsal and postero-dorsal slopes. Shell surface
punctate with small, round, normal pore canal openings evenly but widely spaced
over the carapace. Hinge antimerodont: left valve with elongate, strongly loculate
terminal sockets, separated by a very short but strongly dentate median bar, above
which there is a small accommodation groove. Right valve with 6—7 terminal teeth
and a short, loculate median groove. Inner margin and line of concrescence
coincide, duplicature of moderate width. Radial pore canals short, straight and
few in number; 5 posteriorly, up to 13 anteriorly. Muscle scars of type A (Bate,
1963 : 180). Four oval adductor scars are situated in a vertical row with a large
antennal scar situated in front of the uppermost two adductor scars. A much
smaller mandibular scar is situated antero-ventrally to the adductors.

Dimensions. Holotype. lo.2358, female carapace, length 062 mm.; height
037 mm.; width 0:37 mm.

Paratypes. lo.2360, male carapace, length 0-68 mm.; height 0-37 mm.; width
034mm. Io.2362, female left valve, length 0-60 mm.; height 036mm. Io0.2363,
female right valve, length 065 mm.; height 037mm. Io.2366, male left valve,
length 0-75 mm.; height 041mm. Io.2367, male right valve, length 0-76 mm.;
height 0:38 mm.

REMARKS. Pichottia magnamuris sp. nov. is very close to P. murs Oertli
(1959 : 116, pl. 1, figs. I-10), specimens of which were kindly sent to me by Dr.
Oertli, hut differs in being very much larger, generally not so acuminate posteriorly
and in being more noticeably convex dorsally and ventro-laterally.

BATHONIAN UPPER ESTUARINE OSTRACODA, I 39

Family SCHULERIDEIDAE Mandelstam 1959
Subfamily SCHULERIDEINAE Mandelstam 1959
Genus SCHULERIDEA Swartz & Swain 1946

EMENDED DIAGNOSIS. Carapace subovoid with greatest length subventral.
Dorsal margin convex, umbonate in larger left valve. Posterior narrowly rounded.
Eye swelling either restricted to right valve or present as prominent node on both
valves. Hinge paleomerodont. Muscle scars, type C. Anterior radial pore canals
splayed fan-like, varying in number from 10 to approximately roo. Inner margin
and line of concrescence coincide. Shell surface without strong ornamentation.

REMARKS. The genus Schuleridea was erected by Swartz & Swain (1946) with
the type species being described from the Upper Jurassic, Schuler formation,
Louisiana. In the following year (1947) Mandelstam erected the genus Aequacy-
theridea with the Tertiary ostracod, Cytheridea perforata (Roemer) as type. Both
Aequacytheridea and Schuleridea possess a paleomerodont hinge, type C muscle scar
pattern (Bate 1963 : 18r) and anterior radial pore canals arranged fan-like and
curving outwards away from a line drawn through mid-point. Subsequently
(Chernysheva 1960 : 370 and Moore 1961 : O 284) these two genera were considered
to be synonymous, Aequacytheridea being the junior synonym.

Malz (1958 : 120) noted the evolutionary increase in the number of radial pore
canals and considered Aequacytheridea to be a subgenus of Schuleridea. Kollmann
(1960 : 185) recognized 3 subgenera for Schuleridea: Schuleridea (Schuleridea)
Swartz & Swain 1946, Schuleridea (Aequacytheridea) Mandelstam 1947 and Schuleridea
(n. subgen?)

The three subgenera were not diagnosed nor was the third subgenus named.

The Bathonian species of Schuleridea described in the present paper possesses
some 18—20 anterior radial pore canals, a number much lower than for previously
described species. The number of anterior radial pore canals is considered here to
indicate sub-generic identification. Accordingly three subgenera are here recognized :
Schuleridea, Aequacytheridea and a new subgenus, Eoschuleridea. All three form an
evolutionary sequence with Eoschuleridea probably evolving from the genus
Praeschuleridea Bate (1963 : 207) in early Bathonian or late Bajocian times.

The broad stratigraphic position of the three subgenera and the evolution from
Praeschuleridea is shown in Table I below. For convenience a numerical range of
anterior radial pore canals is given for each subgenus. Overlap at the boundaries
is to be expected in any continuous sequence.

The increase in number of the anterior radial pore canals is shown in Table I to be
without any break between the three subgenera. This is, at the moment, a purely
artificial grouping with each of the subgenera concerned tending to fall somewhere
about the centre of the number of pore canals suggested. When the majority of
the species for all the subgenera have been described, a more exact range in the
number of anterior pore canals may be possible.

40 BATHONIAN UPPER ESTUARINE OSTRACODA, I

TERTIARY

(AEQUACY THERIDEA)

(Aequacytheridea) is the Tertiary
representative of the genus having a
sub-ovate carapace and an increased
number of anterior radial pore canals
(60-100). An eye swelling is generally
prominent on both valves. Dies out in the
Miocene. Type species Cytheridea
perforata (Roemer).

(Schuleridea) is an Upper Jurassic to
Cretaceous subgenus, again sub-ovate in
outline with paleomerodont hinge; 30-60
anterior radial pore canals; eye swelling
slightly more pronounced. Type species

Schuleridea acuminata Swartz and Swain,

Range of (Eoschuleridea) probably
restricted to Bathonian. Sub-ovate in
outline with paleomerodont hinge; 18-30
anterior radial pore canals; eye swelling
onR.V. Type species Schuleridea

(Eoschuleridea) bathonica sp. nov.

Praeschuleridea first recorded from the
Up. Toarcian of Germany, common in L.

Bajocian of England. Oval in outline with

paleohemimerodont hinge; 11-30 anterior
radial pore canals; eye swelling on R.V.
Probably becomes extinct in the Upper
Jurassic. Type species Cytheridea

subtrigona Jones & Sherborn.

CRETACEOUS
(SCHULERIDEA)
UPPER
J URASSIC
BATHONIAN (EOSCHULERIDEA)
BAJOCIAN PRAESCHULERIDEA
LIAS
TABLE I.

Evolutionary Series of Praeschuleridea and Schuleridea.

BATHONIAN UPPER ESTUARINE OSTRACODA, I 41

It is interesting to note that just as the number of anterior radial pore canals
increases throughout the evolution of the genus Schuleridea so also is it the case for
Praeschuleridea which in the Bajocian (and Aalenian) never appears to possess more
than 16 anterior radial pore canals, whereas within the Bathonian the number may
be as high as 30.

Subgenus EOSCHULERIDEA nov.

Dracnosis. Schuleridea having reduced number (18-30) anterior radial pore
canals. Eye swelling poorly developed. Other details as for genus.

TYPE SPECIES. Schuleridea (Eoschuleridea) bathonica sp. nov.

REMARKS. As previously mentioned, Eoschuleridea is both morphologically and
geologically situated between Schuleridea s.s. and Praeschuleridea from which it was
most probably derived. The gradual increase in the number of anterior radial pore
canals and the development of an eye node produces the evolutionary series indicated
above.

Schuleridea (Eoschuleridea) bathonica sp. nov.
(Pl. 7, figs. 7-13, Pl. 8, figs. 1-11)

Diacnosis. LEoschuleridea sub-ovate in side view, posteriorly acuminate in
female dimorph, oval elongate in male dimorph. Dorsal margin arched, left valve
more strongly umbonate in male dimorph. Shell surface punctate with equally
spaced normal pore canal openings. Eye swellings indistinct. Eighteen to twenty
anterior radial pore canals, 9 posterior canals. Larger left valve overlaping right
valve on all sides. Hinge and muscle scars as for genus. Duplicature broad.

HototyPe. lo.2369, female carapace, bed S, Kings Cliffe.

PARATYPES. lo.2370—-94, male and female valves and carapaces from beds F & S,
Kings Cliffe and bed H, Ketton.

DESCRIPTION. Carapace sub-ovate in outline, acuminate posteriorly in the
female dimorph where the line of greatest length passes below mid-point. In the
more elongate, oval, male dimorph, greatest length passes through mid-point.
Greatest height just behind anterior cardinal angle in both dimorphs. Greatest
width median. Anterior broadly rounded in the female, whilst the posterior is more
narrowly rounded. In the male, both anterior and posterior are equally rounded.
Ventro-lateral margin medially convex, ventral margin antero-medially incurved.
Dorsal margin convex, especially in the left valve and particularly in the male where
the left valve is decidedly ““ umbonate’”’. Cardinal angles rounded with the anterior
angle right valve male and the posterior angle right valve female being the most
sharply defined. Postero-dorsal slopes in the female larger and more nearly straight
than in the male. Larger left valve overlaps the smaller nght valve on all sides.
Shell surface smooth or punctate, depending upon preservation with well developed
normal pore canal openings evenly scattered over the valve. Hinge paleomerodont.

42 BATHONIAN UPPER ESTUARINE OSTRACODA, I

Left valve with elongate strongly loculate terminal sockets, anteriorly with 6 grooves
and posteriorly with 8. Median, connecting groove short, and very finely locellate.
Accommodation groove shallow and rather poorly developed. Right valve with
terminal elongate dentate ridges bearing 6 anterior and 8 posterior teeth connected
by a short, very finely denticulate ridge. The median element in the male dimorph
is longer than that present in the female, the number of terminal teeth, is, however,
the same. Inner margin and line of concrescence coincide the duplicature
being quite broad, especially in the postero-ventral region. Radial pore canals
long and curved. Anteriorly there appears to be generally 19 in number though
18 and 20 have been observed, arranged fan-like. Posteriorly there are 9 canals, of
which 8 are situated below a line through mid-point. Muscle scars of type C.
The four oval adductor scars form a concentric row around the antero-median oval
antennal scar. The mandibular scar is antero-ventral in position.

Dimensions. Holotype. Io.2369, female carapace, length 0-74 mm.; height
0-43 mm.; width 0:36 mm.

Paratypes. lo.2370, male carapace, length 0-80 mm., height 0:47 mm., width
038mm. Io.2373, female right valve, length 0-65 mm.; heighto:339 mm. Io0.2375,
female left valve, length 0-70 mm.; height 0-46mm. _ Io.2376, female right valve,
length 0-67 mm.; height 040mm. lIo.2378, male right valve, length 0-77 mm.;
height 0.441 mm. _ Io0.2379, male left valve, length 0-74 mm.; height 0:43 mm.

REMARKS. Schuleridea (Eoschuleridea) bathonica sp. nov., may be distinguished
from described species of Schuleridea not only on variations in the outline but in the
number of radial pore canals present. The male dimorph of this species, externally
resembles the male dimorph of Galliaecytheridea? kingscliffensis but may be
distinguished by its more oval outline, strongly umbonate left valve and internally
by the hinge, muscle scars and radial pore canals.

Genus PRAESCHULERIDEA Bate 1963
Praeschuleridea quadrata sp. nov.
(Pl. 9, figs. 1-12)

DiaGnosis. Praeschuleridea with oval carapace in male dimorph, dorsally arched
in female. Both dimorphs with distinctly angled posterior cardinal angle. Shell
surface punctate. Hinge not strongly paleohemimerodont. Radial pore canals,
varying from 20-30 anteriorly.

HoLotyPe. 10.2395, female carapace, bed R, Kings Cliffe.

PARATYPES. I[o.2396-—2413, male and female valves and carapaces, beds R & S,
Kings Cliffe.

DESCRIPTION. Carapace sub-trigonal in the female dimorph, arched dorsally,
with greatest height passing through mid-point. In the male dimorph the carapace
is oval-elongate with greatest height being just behind the anterior cardinal angle.

BATHONIAN UPPER ESTUARINE OSTRACODA, I 43

Characteristically in both dimorphs the posterior cardinal angle is sharply angled
and in the female set high up away from the posterior. Dorsal margin long and
gently convex in the male, umbonate and strongly convex in the female. Anterior
cardinal angle in both cases, broadly rounded. Anterior and posterior broadly
rounded in the male, posterior more narrowly rounded in the female. Ventral
margin convex, incurved slightly antero-medially. Postero-dorsal slope longer in
the female than in the male. Greatest length through mid-point in the male, slightly
below this in the female. Greatest width median in both dimorphs. Shell surface
punctate with rather large, circular normal pore canal openings. Left valve
larger than the right which it overlaps on all sides other than dorsally, in which
region the left valve over-reaches the right. Hinge paleohemimerodont. Left
valve with terminal loculate sockets connected by a short locellate groove which
extends over a low median bar. In male dimorphs the median groove is virtually
impossible to distinguish, whilst it is extremely well developed in the female. In
some material the dorsal edge of the groove projects noticeably above the groove.
Above the median element there is an elongate/triangular accommodation groove.
In the right valve the dentate terminal elements consist of anteriorly 6 teeth and
posteriorly 7 teeth corresponding with the terminal sockets in the left valve. Median
ridge low and rather poorly defined. Inner margin and line of concrescence
coincide producing a duplicature of moderate width. Anterior radial pore canals
splayed fan-like, curving outwards from a line drawn through mid-point. The
number present varies within individuals from 20-30. Posteriorly the pore canals
also curving away from a line drawn through mid-point are situated largely below
this line and vary from approximately 9-18. Muscle scars, type CG: Adductor
scars form a slightly crescentic row in front of which is an oval antennal scar, the
long axis of which is vertical.

Dimensions. Holotype. Io.2395, female carapace, length 0-63 mm.; height
0-43 mm.; width 0:34 mm.

Paratypes. lo.2397, male carapace, length 0:82 mm., height 0-48 mm.; width
037mm. Io.2308, female left valve, length 064mm.; height 0-45mm. _ lo.2399,
female right valve, length 0:56mm.; height 035mm. Io.2400, male left valve
length 0-78 mm.; height 0-46 mm.

REMARKS. The hinge present in Praeschuleridea quadrata sp. nov., although
paleohemimerodont, does not possess such a positive median element in the left
valve as occurs in all the previously described species of the genus. A second feature
of this species is that the number of radial pore canals has increased and appears
to vary within the population. It is interesting to note also that those specimens
which possess the larger number of anterior pore canals also possess the larger number
of posterior canals.

The posterior cardinal angle in the present species is situated much higher up on
the carapace than is the case in P. subtrigona (Jones & Sherborn), Bate (1963 : 207,
pl. 12, figs. 12-16; pl. 13, figs. 1-9).

Schuleridea pentagonalis Swartz & Swain (1946 : 368, pl. 53, figs. 1-8) is externally
similar to Praeschuleridea quadrata but may be distinguished by the outline of the

44 BATHONIAN UPPER ESTUARINE OSTRACODA, I

dorsal margin which is strongly umbonate in the left valve of P. guadrata but only
broadly convex in S. pentagonalis. Internal details for the latter are not known.

Family CYTHERURIDAE Miiller 1894

Genus METACYTHEROPTERON Oertli 1957
Metacytheropteron drupacea (Jones)
(Pl. Lo, figs. 1-9)

1884 Cythere drupacea Jones : 772, pl. 34, fig. 30.
1957a Metacytheropteron sp. 50 Oertli: table tr.

Diacnosis. Metacytheropteron with elongate/sub-ovate carapace. Greatest
height at anterior cardinal angle, posterior acuminate. Dorsal margin broadly
convex, especially in left valve. Shell surface strongly ornamented with triangular
arrangement of longitudinal and obliquely transverse ridges.

HototypPe. IN.43498, female carapace from Richmond boring at 1205 ft.,
figured Jones 1884, pl. 34, fig. 30.

OTHER MATERIAL. Io.2414—18, single valves (male and female dimorphs) from
bed M, Ketton.

DESCRIPTION. Carapace sub-ovate tapering strongly to the posterior in the
female dimorph, rather more elongate in the male. Dorsal margin arched in the left
valve, medially convex in the right. Anterior broadly rounded; posterior acuminate,
with convex postero-ventral slope and concave postero-dorsal slope (convex in the
male dimorph). Ventro-lateral margin convex, particularly in the female. Ventral
margin with median incurvature. Greatest length of carapace passes slightly below
mid-point; greatest height at the anterior cardinal angle in the left valve, median in
the right; greatest width median. Shell surface strongly ornamented with a
triangular arrangement of longitudinal and obliquely transverse ridges. The dorsal
apex of the triangle (equilateral) meets the dorsal margin above valve mid-point.
The centre of the triangle breaks down into a reticulate ornamentation. Ventral
surface with 4—5 longitudinal ridges. Numerous normal pore canal openings are
clearly seen only on the ventral and ventro-lateral surfaces. An oval-elongate eye
swelling is only really clearly seen in the male dimorph just below the anterior
cardinal angle. The deep, oblique groove present behind the eye swelling is,
however, well developed in both dimorphs. Left valve larger than the right which
it overlaps along the ventral margin and along the antero-dorsal slope, and probably
also along the postero-dorsal slope. Anteriorly the left valve appears to merely
extend beyond the right. Elsewhere the valve relationships are not clearly seen
in the present material. Hinge antimerodont. Left valve with elongate loculate
sockets and a denticulate median bar, above which there is only a very poorly
developed accommodation groove. Right valve with approximately 6 posterior
teeth and 8 anterior teeth, median groove locellate. Inner margin and line of
concrescence coincide, although perhaps not exactly so around the anterior.

BATHONIAN UPPER ESTUARINE OSTRACODA, I 45

Radial pore canals straight, rather strongly developed and widely spaced, 8
anteriorly at least 3 posteriorly but probably no more than 4. Muscle scars not
observed.

Dimensions. Holotype. IN.43498, female carapace, length 0:50 mm.; height
0:30 mm.; width 0:29 mm.

Other material. lo.2414, female left valve, length 0:54 mm.; height 0:33 mm.
Io.2415, male left valve, length 0-67 mm.; height 0:35 mm. lIo.2416, male right
valve, length o60mm.; height 030mm. Io.2417, male right valve, length
0-61 mm.; height 0-30 mm.

REMARKS. The holotype is the only complete carapace available at the present
time and this has suffered a certain amount of crushing in the postero-dorsal region.

Dr. H. J. Oertli kindly loaned me material of the type species, Metacytheropteron
elegans Oertli (1957 : 664, pl. 4, figs. 116-124) to compare with the species described
here. Although very close to M. elegans, M. drupacea may be distinguished by the
more strongly arched dorsal margin in the left valve and is, therefore, not so narrow
and elongate in lateral view. Ornamentally M. drupacea has the more definitely
triangular arrangement of ridges without such a large central reticulate development.

Family PROGONOCYTHERIDAE Sylvester-Bradley 1948
Subfamily PROGONOCYTHERINAE Sylvester-Bradley 1948
Genus PROGONOCYTHERE Sylvester-Bradley 1948

Progonocythere levigata sp. nov.
(Pl. xo, figs. ro-14, Pl. 11, figs. 1-9)

Diacnosis. Progonocythere with subquadrate/elongate punctate carapace over
which large, circular, normal pore canal openings are evenly spaced. Dorsal outline
of left valve, female dimorph, characteristically umbonate, anterior cardinal angle
being set well back only just in front of valve centre. Small marginal denticles may
occur anteriorly and posteriorly in right valve. Ventro-lateral margin broadly
convex. Anterior and posterior borders compressed. Internal details as for genus.

HorotyPe. Io.2419, female left valve, bed_H, Ketton.

PARATYPES. Io.2420-33, carapaces and single valve (both dimorphs), bed H,
Ketton.

DESCRIPTION. Carapace subquadrate with high, arched dorsal outline in the
female dimorph, elongate in the male. Anterior broadly rounded with, in two
right valves, a small group of denticles situated at mid-height. Posterior narrowly
rounded situated high on the male carapace; at mid-height in the female left valve,
subventral in the female right valve where a small group of denticles may be found
on the postero-ventral slope. The postero-dorsal slope in both female valves is very
steeply angled. Dorsal margin broadly convex in the male, steeply angled in the

46 BATHONIAN UPPER ESTUARINE OSTRACODA, I

female, especially within the left valve where the anterior cardinal angle is situated
just anterior of valve middle, producing an umbonate outline. Ventro-lateral
margin strongly convex. Ventral margin with antero-median incurvature. Greatest
length above mid-point in the male left valve; through mid-point in the female
left valve, below mid-point in the right. Male right valve damaged but greatest
length probably passes through mid-point. Greatest height through the anterior
cardinal angle in the female, median in the male. Greatest width median. Anterior
and posterior marginal borders compressed. Shell surface finely punctate with
prominent, evenly spaced, normal pore canal openings. Left valve larger than
the right which it overlaps along the ventral margin and slightly along the antero-
dorsal slope. Elsewhere the left valve over-reaches the right. Hinge entomodont.
Left valve with strongly loculate terminal sockets and a dentate median bar, the
anterior portion of which is even more strongly dentate. A long, narrow accom-
modation groove is present above the median element. Right valve with 7 posterior
teeth and an indeterminate number of anterior teeth (probably 8-9), and a loculate
median groove, the anterior part of which is greatly expanded. Inner margin and
line of concrescence coincide; duplicature only of moderate width. Radial
pore canals not distinguished. Muscle scars of type A. The 4 adductor scars
are in a subvertical row with the round antennal scar situated opposite and close
to the most dorsal scar in the male. In the female theantennal scar is slightly
lower in position and occurs a good deal further away from the adductors.

Dimensions. Holotype. lIo.2419, female left valve, length 0-81 mm.; height
0°54 mm.

Paratypes. lo.2420, female right valve, length 0:82mm.; height 0-49 mm.
Io.2421, male left valve, length 0:85 mm.; height 050mm. lIo.2422, male right
valve, length (broken) 0-93 mm.; height 051 mm. Io.2423, female carapace, length
0-71 mm.; height 0:46 mm.; width 0:35 mm.

REMARKS. Progonocythere levigata approaches P. cristata Bate (1963 : 191, pl. 4,
figs. 5-15, pl. 5, figs. 1-6) in outline but does not develop the ventro-lateral overhang
characteristic of the latter. The male dimorph of P. levigata also is a much more
oval ostracod in side view. Pyvrogonocythere stilla Sylvester-Bradley (1948 : Igo,
pl. 12, figs. 1, 2, pl. 13, figs. I, 2) is easily distinguishable by its more truncated
posterior.

Progonocythere rugosa sp. nov.

(Pl. xa; figs. 10-14, Pl x2; fies. 1-o} RL 13 ite. 25)

DiaGnosis. Pyrogonocythere with coarse ornamentation of pits, grooves and ridges
aligned parallel to ventro-lateral and terminal margins. Carapace subdivided by
deep median, vertical sulcus. Ventro-lateral margin extended keel-like below
ventral surface. Internal details as for genus.

HoLotyPe. Io.2434, female carapace, bed S, Kings Cliffe.

PARATYPES. Io0.2435-52, both dimorphs, carapaces and single valves from beds
R & S, Kings Cliffe, Dane Hill, and bed S, Ketton.

BATHONIAN UPPER ESTUARINE OSTRACODA, I 47

DESCRIPTION. Carapace subquadrate, more elongate, however, in the male
dimorph. Ventro-lateral margin strongly overhangs the ventral surface and tapers
to form a keel-like structure, particularly in the postero-ventral region. Greatest
length of carapace tends to pass slightly above mid-point in both dimorphs, the
narrowly rounded posterior being set high up on the carapace. Anterior broadly
rounded. Dorsal margin with slight concavity behind anterior cardinal angle,
through which the line of greatest height passes. This angle is set further back in
the female left valve than in the male; a condition which to some extent also applies
in the right valve. Antero-dorsal slope tends to be rather long and convex, whilst
the postero-dorsal slope is concave and much shorter. Greatest width just behind
the median sulcus. Anterior and posterior marginal borders compressed. Ventral
margin distinctly incurved antero-medially. Shell surface strongly ornamented.
Ventrally the V-shaped ventral surface, bounded on either side by the ventro-lateral
overhang, is ornamented by 3 prominent longitudinal ridges per valve, not counting
the ridge which produces the keel of the ventro-lateral margin. Lateral surface
coarsely ornamented by a series of pits, grooves and complimentary ridges aligned
parallel to the ventro-lateral and terminal margins. Particularly in the female,
there is a deep vertical sulcus developed about valve centre. Normal pore canals
few in number, large and widely spaced. Left valve larger than the right, which it
overlaps noticeably along the ventral margin, and both antero- and postero-dorsally.
Anteriorly the left valve projects shghtly beyond the right and strongly over-reaches
dorsally. Hinge entomodont. Left valve with coarsely loculate terminal sockets
(8 loculi anteriorly and 7 posteriorly), median bar dentate, more coarsely so in
anterior half. Accommodation groove broad and rather shallow. Right valve with
8 anterior teeth and 7 posterior teeth; median groove loculate, expanded anteriorly.
Muscle scars of type A. The antero-dorsal antennal scar is quite large in the
single specimen showing it, and appears to be formed by a fusion of two smaller,
round, scars. Antero-ventral scar round, also larger than the four round adductor
scars which are arranged in a slightly crescentic row. Inner margin and line of
concrescence coincide, the duplicature being rather narrow. As a result the
widely spaced, straight, radial pore canals are rather short. Anteriorly there are
8 whilst posteriorly only 3.

Dimensions. Holotype. Io.2434, female carapace, length 0-72 mm.; height
0-49 mm.; width 0-42 mm.

Paratypes. lIo.2435, male carapace, length 0:80 mm.; height 0-49 mm.; width
039mm. lo.2436, female right valve, length 0-63 mm.; heighto-4o mm. _ Io.2437,
male left valve, length 0-74 mm.; height 0-46mm. Io.2438, female right valve,
length 0-65 mm.; height 0:40 mm.

REMARKS. Progonocythere rugosa sp. nov. is distinguishable from other species
of the genus by the type and coarseness of the ornamentation which in many ways
is similar to that often found in species of Glyptocythere. However, as with other
species of Progonocythere, the absence of a dorsomedian projection in the right valve
prevents this species being a Glyptocythere. It is, in all other details, typical of the
genus in which it has been placed.

48 BATHONIAN UPPER ESTUARINE OSTRACODA, I

Progonocythere triquetra sp. nov.
(Pl. 12, figs. ro-13, Pl. 13, figs. 1, 3-9)

Diacnosis. Pyrogonocythere with high domed, sub-triangular carapace in female;
elongate-subrectangular carapace in male. Both dimorphs in dorsal view have
steeply angled lateral surfaces only slightly convex. Ventro-lateral margin extended
below ventral surface. Ventral surface ornamented by longitudinal ridges; lateral
surface punctate, over which are widely scattered a number of large, circular, normal
pore canal openings. Internal details as for genus.

HoLotyPe. Io.2453, female carapace, bed J, Kings Cliffe.

PARATYPES. I[o.2454-63, male and female single valves and female carapace
from beds H & J, Kings Cliffe and bed H, Ketton.

DESCRIPTION. Carapace sub-triangular in the female dimorph having a high
domed dorsal outline (left valve) with steeply angled dorsal margins. The male
dimorph is much more rectangular in outline and does not possess the high dorsal
outline of the female. Greatest length of carapace passes through mid-point with
the greatest height at the anterior cardinal angle and greatest width median.
Anterior broadly rounded, posterior triangular with concave postero-dorsal slope
and convex postero-ventral slope. Dorsal margin very slightly convex, only gently
sloping to the posterior in all valves other than the female left valve where it is
steeply angled. Antero-dorsal slope convex and short, again in all valves other
than in the female left valve where it is long, due to the situation of the anterior
cardinal angle only just anterior of valve centre. Ventro-lateral margin strongly
convex and extended below the ventral surface. Ventral margin incurved antero-
medially, only very shallow in the male left valve and with a ventral “ lip” in the
right valve. Ventral surface of each valve ornamented by 4-5 longitudinal ridges.
Lateral surface punctate. Normal pore canal openings widely scattered, large,
circular and comparatively few in number. Left valve larger than the right which
it overlaps along the ventral surface, in the region of the cardinal angles but strongly
over-reaches mid-dorsally. Hinge entomodont. Left valve with terminal loculate
sockets and a dentate median bar, the dentations being much coarser in the anterior
half, above which is an elongate, deep, accommodation groove. Right valve with
6 anterior teeth and 7 posterior teeth in both males and females. All teeth are
dorsally bifid. Median groove strongly loculate and expanded anteriorly. Inner
margin and line of concrescence coincide. Duplicature narrow; selvage
prominent. A narrow flange extends around the anterior and posterior margins
and along the ventral margin in which region it forms the ventral “ lip ’’ below the
median incurvature. The flange is only clearly seen in the right valve of both
dimorphs. Radial pore canals short, straight and widely spaced, 8 anteriorly.
Muscle scars, type A. Adductor scars oval, aligned in a slightly crescentic row
with a round antero-dorsal antennal scar and a round antero-ventral mandibular
scar.

DimEnsIons. Holotype Io.2453, female carapace, length 0-75 mm.; height
0°54mm.; width 0-40 mm.

BATHONIAN UPPER ESTUARINE OSTRACODA, I 49

Paratypes. lo.2454, female left valve, length 062mm.; height 0-45 mm.
Io.2455, female right valve, length 066mm.; height 040mm. Io.2456, male
left valve, length 0'86mm.; height 050mm. lIo.2457, male right valve, length
o-70mm.; height 037mm. Io.2458, female carapace, length 0-79 mm.; height
055 mm.; width 0:38 mm.

REMARKS. The subtriangular carapace in the female and the steeply sloping
lateral sides when viewed dorsally readily distinguish this species from others of the
genus.

Genus GLYPTOCYTHERE Brand & Malz 1962

REMARKS. The genus first appeared in print in Brand & Malz 1962a when 7
species of Glyptocythere were described. The diagnosis of the genus and the erection
of a type species was not published until a short while later in Brand & Malz 19620.

Glyptocythere guembeliana (Jones)
(Pl. 13, figs. 10-16, Pl. 14, figs. 1-8)

1884 Cythere guembeliana Jones : 772, pl. 34, figs. 32, 33 [non fig. 31).
1888 Cytheridea pulvinay Jones & Sherborn : 266, pl. 3, fig. 2a—c.

Diacnosis. Glyptocythere with subquadrate carapace, elongate in male dimorph.
Lateral surface with transverse ridges extending down from dorsal margin although
generally poorly developed in most specimens. Marginal borders compressed.
Ventro-lateral margin evenly convex in female, sharply directed upwards posteriorly
in male right valve; may possess deep longitudinal groove in either dimorph or may
possess short blade-like extension. Hinge weakly entomodont.

LeEcToTYPE. Selected here, IN .43493, male right valve, Great Oolite, Richmond
boring at a depth of 1205 ft., figured Jones 1884, pl. 34, fig. 33.

PARALECTOTYPE. Io.3338, male carapace from Richmond boring, depth 1205 ft.

JONES’S COLLECTION. 1.1858, lectotype of Cytheridea pulvinar from the Blue
Fullers Earth Clay; Midford, Bath. A female right valve. Figured Jones &
Sherborn 1888, pl. 3, fig. 2a—c; Io.2464, male right valve fragment from the Blue
Fullers Earth Clay; Midford Bath.

OTHER MATERIAL. Io.2465-2512, male and female valves and carapaces from
beds F, O & R, Kings Cliffe; beds H & L, Ketton and bed U, Ancaster.

DESCRIPTION. Carapace somewhat variable in size, subquadrate in the female,
more elongate in the male. Greatest length of carapace passes through mid-point
whilst the greatest height median or at the anterior cardinal angle. Greatest width
behind mid-point. Anterior broadly rounded in the left valve, tending to develop
a short, slightly concave antero-dorsal slope in the right valve. Posterior triangular
with a strongly concave, short, postero-dorsal slope and a longer, convex, postero-
ventral slope. Ventro-lateral margin broadly convex, obliquely angled posteriorly

50 BATHONIAN UPPER ESTUARINE OSTRACODA, I

in the male right valve. Ventral margin with a median incurvature. Dorsal margin
strongly convex medially in the right valve, less obviously so in the left. Cardinal
angles prominent. Shell surface punctate and generally poorly ornamented though
some specimens may develop quite prominent transverse ridges which extend
downwards to die out at a line drawn through mid-point. A broad, shallow
depression is most noticeably developed in the right valve just behind the anterior
cardinal angle. A short blade-like crest may be developed in some specimens on
the ventro-lateral margin, whilst others may develop a short but deep, longitudinal
groove, just above the ventro-lateral margin. Normal pore canals with large,
circular openings evenly spaced over the carapace. Left valve larger than the right
which it overlaps along the ventral margin and slightly along the antero-dorsal slope.
Elsewhere, along the dorsal margin the left valve over-reaches the right. Hinge
very poorly entomodont. Left valve with terminal, elongate, loculate sockets
between which is a strong, coarsely dentate median bar, the teeth of which show a
tendency in the anterior portion to become partially united in pairs. Accommoda-
tion groove poorly represented. Right valve with 6 anterior and 7 posterior, dorsally
bifid, teeth. Median groove coarsely loculate, expanded very slightly in the anterior
portion. Inner margin and line of concrescence coincide. Duplicature of
moderate width with anteriorly 8, straight and widely spaced radial pore canals
and 3 posteriorly, 2 of which are situated together at the point of the posterior margin.
Muscle scars of type A with the antero-dorsal antennal scar being much larger
than any of the adductors.

Dimensions. Lectotype. IN.43493, male right valve, length 0-96 mm.; height
049mm. Paralectotype. Io.3338, male carapace, length o-7r1mm.; height
0:36 mm.; width 0:32 mm.

Other material. 1.1858, female right valve, length 0:82 mm.; height 0-47 mm.
Io.2465, male left valve, length 0-71 mm.; height 037mm. Io.2466, male right
valve, length 083mm.; height 042mm. Io.2467, female left valve, length
0-770 mm.; height 045mm. Io.2468, female right valve, length 0-72 mm.; height
042mm. Io.2469, female right valve, length 0-66 mm.; heighto-39 mm. I[o.2471,
male carapace, length 0773mm.; height o4omm.; width 036mm. Io.2472,
female right valve, length 0-78 mm.; height 0-45mm. Io.2473, female carapace,
length 0-73 mm.; height 0:46 mm.; width 0:39 mm.

REMARKS. Apart from the paralectotype, all specimens from the Jones collection
are larger than those of the same sex present within the Upper Estuarine Series.
That a male dimorph (the paralectotype) of small size occurs with larger specimens
suggests that the adult size of this species is somewhat variable. Also variable
within a population is the tendency in some specimens to develop a deep groove
near to the ventro-lateral margin and in others for the ventro-lateral margin to be
thickened or extended into a blade-like keel. The ornamental features of this species
readily identify it from existing species of the genus.

The two species placed in synonymy are simply male and female dimorphs;
Jone’s Cythere guembeliana being the male and Cytheridea pulvinar described by
Jones & Sherborn, the female.

BATHONIAN UPPER ESTUARINE OSTRACODA, I 51
Glyptocythere juglandica (Jones)

(Bit: fis. a)

1884 Cythere juglandica Jones ; 766, 768, pl. 34, figs. 36, 37.
1888 Cythere juglandica var. majoy Jones & Sherborn : 255, pl. 4, fig. 2a—b.
1948 Pyvogonocythere juglandica (Jones) Sylvester-Bradley : 193, pl. 12, figs. 5, 6, pl. 13, fig. 8.
1963 Progonocythere juglandica juglandica (Jones) ; Grekoff : 1731, pl. 3, fig. 55.
1963 Pyvogonocythere? juglandica (Jones) ; Oerth, pls. 28, 29, 30.

MATERIAL. I[o.2513-17, male and female carapaces and a female left valve from
beds N, O & S, Kings Cliffe.

REMARKS. The strong ornamentation and characteristic mid-dorsal extension of
the right valve places this species within the genus Glyptocythere. The doubtful
position of juglandica within Progonocythere has been noted by Oertli (1963).

Genus KLIEANA Martin 1940
Klieana levis Oertli

(Pl. 14, figs. 10-13, Pl. 15, figs. 1-5)
1957 Klieana levis Oertli: 760, pl. 22, figs. 13-19.
MATERIAL. [o.2518—-36, male and female carapaces and single valves from bed
O, Kings Cliffe; bed U, Ancaster and bed L, Kettering.

REMARKS. Due to the kindness of Dr. Oertli, I have been able to examine some
of the material originally described, and to compare it with the present material.
The specimens of Klieana levis, which are found, often in fairly large numbers,
within the Upper Estuarine Series tend to be slightly larger than those described by
Oertli, which have an average length of 0:50-0:60 mm., and possess a punctate rather
than a smooth carapace. All the specimens figured by Oertli are female dimorphs
when compared with the present material and as he states (p. 760) that dimorphism
is not recognized with certainty it is probable that the male dimorphs, which are
never very common, were not found in the French material. The male dimorph
is a very noticeable form much more elongate than the female and does not have
the strongly arched dorsal margin to the left valve. The muscle scars, not observed
in Oertli’s material, are of type A. Radial pore canals straight and widely spaced,
8 anteriorly and 4 posteriorly. Hinge antimerodont.

Dimensions. l[o.2518, female carapace, length o61mm.; height o-40mm.;
width 032mm. lIo0.2520, female right valve, length 0-63 mm.; height 0:39 mm.
Io.2521, female left valve, length 066mm.; height 043mm. Io.2522, male
carapace, length 0-74 mm.; height 0-36 mm.; width 0:35 mm.

Genus LOPHOCYTHERE Sylvester-Bradley 1948
Lophocythere ostreata (Jones & Sherborn)

1888 Cytheridea ostreata Jones & Sherborn : 271, pl. 4, fig. 6.
1948 Lophocythere ostreata (Jones & Sherborn) Sylvester-Bradley : 195, pl. 14, figs. 1-4, pl. 15,
figs. 1, 2.

52 BATHONIAN UPPER ESTUARINE OSTRACODA, I

REMARKS. A single, juvenile carapace (lo.2540) found in bed S, Kings Cliffe.
Not figured.

Lophocythere scabra scabra Triebel
(Pl. 15, fig. 6)

1951 Lophocythere scabra Triebel : 95, pl. 46, figs. 26-30, pl. 47, figs. 31-34.
1960 Lophocythere scabra scabya Triebel ; Lutze : 429, pl. 37, fig. I.

1962 Lophocythere scabva scabya Triebel; Brand & Fahrion : 147, pl. 21, fig. 32.
1963 Lophocythere scabra Triebel ; Oertli: 43, pls. 26, 28-30.

MaTERIAL. [o.2537-39 and Io.2541, male and female valves and carapaces
from beds F & R, Kings Cliffe.

REMARKS. Like Glyptocythere juglandica, Lophocythere scabra scabra is a typical
Bathonian ostracod and is common in the majority of the marine sediments of the
Upper Estuarine Series.

Lophocythere septicostata sp. nov.
(Pl. 15, figs. 7-13, Pl. 16, figs. 1-4)
1888 Cytheridea byadiana Jones & Sherborn : 272, pl. 4, figs. 11a—c.

Diacnosis. Lophocythere having seven longitudinal ridges on lateral surface and
two on ventral surface. A vertical ridge extends downwards from oval eye swelling.
Interspaces between ridges, punctate, with large, circular, normal pore canal openings.
Anterior and posterior marginal borders compressed.

HoLotype. Io.2542, female carapace from bed B, Kettering.

PARATYPES. Io0.2543-49, male and female carapaces and male left valve from bed
B, Kettering; bed V, Ancaster and bed N, Kings Cliffe. 1.1843, female right valve,
described by Jones & Sherborn as Cytheridea bradiana figd., pl. 4, figs. Ira—c, from
the Blue Fullers Earth Clay; Midford, Bath. Eight male and female specimens
(Lo.3600-07) of the Mockler collection from the Fullers Earth; Midford nr. Bath.

DESCRIPTION. Carapace subquadrate in the female dimorph, elongate in the
male. Greatest length passes through mid-point whilst greatest height occurs at
the anterior cardinal angle in the female, posteriorly in the male. Greatest width
in the posterior third. The carapace possesses prominent, well rounded cardinal
angles with the dorsal margin between tending to be medially concave in the left
valve, gently convex in the right. The ventral margin is incurved medially whilst
the ventral surface is overhung by the convex ventro-lateral margins. Anterior
high, broadly rounded; posterior triangular with concave postero-dorsal slope and
convex postero-ventral slope. Anterior and posterior marginal borders compressed
in the female, only posteriorly so in the male. Shell surface laterally ornamented
by a series of 7 oblique, longitudinal ridges which in the anterior half bend downwards
and fuse together so that only 2 ridges extend as far as the antero-ventral margin.

BATHONIAN UPPER ESTUARINE OSTRACODA, I 53

The median ridges tend to be short but one of these is long and extends to the antero-
ventral margin, although it is not always the same ridge in each specimen which
does this. An oval eye swelling, situated below the anterior cardinal angle, has a
prominent ridge extending from it as far as the uppermost of the two ridges. Surface
of carapace between the ridges is punctate and in some specimens also seen to possess
large, circular normal pore canal openings. Ventral surface possesses two further
longitudinal ridges. Left valve larger than the right which it overlaps along the
ventral margin and slightly at the cardinal angles. Elsewhere dorsally, the left
overreaches the right. Hinge entomodont, left valve not clearly seen but in the
right there are 5 anterior teeth and 6 posterior teeth, the coarsely loculate median
groove being noticeably expanded in its anterior portion. Inner margin and line
of concrescence coincide, no radial pore canals observed. A narrow flange
extends around the anterior margin and continues much reduced outside the ventral
margin. Muscle scars not observed.

Dimensions. Holotype. Io.2542, female carapace, length 0-72 mm.; height
0-40 mm.; width 0:34 mm.

Paratypes. 1.1843, female right valve (Jones & Sherborn colln.) length 0:64 mm. ;
height 033mm. Io.2546, female carapace, length 0-72 mm.; height 0-43 mm.;
width 035mm. Io.2547, male carapace, length 0-77mm.; height 0:36 mm.;
width 0:35 mm.

REMARKS. Jones (1884 : 772) described a new ostracod species named Cythere
bradiana, subsequently placed into Lophocythere by Sylvester-Bradley (1948 : 196).
Jones & Sherborn (1888) described material from the Fullers Earth near Bath and
identified some specimens as belonging to Cythere bradiana although they changed
the generic name to Cytheridea without comment. This material was not, however,
conspecific with their earlier Cythere bradiana although like that species, belongs in
the genus Lophocythere. Jones & Sherborn’s material is included here in the new
species Lophocythere septicostata. L. septicostata differs from L. bradiana in the
number of lateral ridges and absence of reticulation between (although a single male
carapace, lo.2547, shows some evidence of reticulation), characters which also serve
to distinguish the present species from L. multicostata Oertli (1957 : 667). In
addition, the compressed anterior marginal border present in the female dimorph
of L. septicostata distinguishes this species in dorsal view from the other two species.
The possession of an eye swelling identifies this species as belonging to the genus
Lophocythere rather than to the related genus Terquemula Blaszyk & Malz 1965.

Lophocythere transversiplicata sp. nov.
(Pl. 16, figs. 5-15)

Diacnosis. Lophocythere, with three primary longitudinal ridges; secondary
transverse ridges radiating down from ridge to produce broad reticulation between
longitudinal ridges. All three longitudinal ridges converging towards antero-ventral

GEOL. 14, 2. 7

54 BATHONIAN UPPER ESTUARINE OSTRACODA, I

margin. Minor transverse ridge extending down from region of anterior cardinal
angle. Ventral surface also ornamented with longitudinal ridges.

Ho.totypPer. Io.2625, female carapace, bed J, Ketton.

PARATYPES. Io.2626-—29, 3087-89, male and female carapaces and single valves,
bed J, Ketton.

DESCRIPTION. Carapace subquadrate to subrectangular, the more elongate
specimens being the males. Anterior high and broadly rounded; posterior triangular
with concave postero-dorsal slope and convex postero-ventral slope. Anterior
cardinal angle, through which passes the line of greatest height, prominent; the
right valve has a low eye swelling situated below. Posterior cardinal angle not so
prominent as the anterior angle. Dorsal margin of left valve overreached strongly
by an extension of the valve. Ventral margin incurved antero-medially. Greatest
length of carapace passes through mid-point whilst the greatest width les in the
posterior third. Ventral surface ornamented by four prominent longitudinal ridges
per valve. Lateral surface with three longitudinal ridges which converge towards
the antero-ventral margin. The dorsal ridge also bends down posteriorly to join
the median ridge below. A series of minor transverse ridges extend down from the
dorsal ridge and produce a coarse reticulation between the longitudinal ridges. In
addition a transverse ridge extends down from the region of the anterior cardinal
angle to fuse with the three longitudinal ridges. Left valve is larger than the right
which it overlaps along the ventral margin and slightly at the cardinal angles.
Dorsally the left over-reaches the right. Hinge entomodont. Left valve with
terminal loculate sockets, virtually no accommodation groove and a dentate median
bar. Because of damage to this bar, the typical entomodont character cannot be
seen. Right valve with 5 anterior and 6 posterior teeth. Median groove loculate,
expanded in the anterior half. Inner margin and line of concrescence coincide.
Radial pore canals straight, 9 anteriorly and 4 posteriorly. Other internal details
not observed.

Dimensions. Holotype. Io.2625, female carapace, length 0:-56mm.; height
0:34mm.; width 0:31 mm.

Paratypes. Io.2626, male carapace, length 0-70 mm.; height 0-34 mm.; width
031mm. Jo.2627, female right valve, length 0-53 mm.; heighto:28 mm. Io.2628,
male right valve, length 068mm.; height 027mm. Io.2629, male left valve,
length 0-62 mm., height 0:32 mm.

Remarks. Lophocythere transversiplicata sp. nov. bears some resemblances to L.
flexicosta Triebel (1951 : 97, pl. 48, figs. 46-48) and L. plena Triebel (1951 : 100,
pl. 49, figs. 60-63) but differs from both in not having such a neat reticulate ornamen-
tation over the lateral surface. The reticulation that does occur being produced
simply by branching and anastomosing transverse ridges restricted to the interspaces
between the longitudinal ridges and not generally present over the entire lateral
surface. It is distinguished from species of Tevquemula Blaszyk & Malz 1965 on
ornamentation and possession of an eye swelling.

BATHONIAN UPPER ESTUARINE OSTRACODA, I 55
Genus MACRODENTINA Martin 1940

REMARKS. Malz (1958) erected three subgenera for the genus, namely :—
Macrodentina (Dictyocythere)
Macrodentina (Macrodentina)
Macrodentina (Polydentina)

Each subgenus is identified by the type of amphidont hinge present, in all cases
the median element being smooth. The first appearance of the genus according to
the present literature occurs towards the base of the Upper Jurassic (Kimmeridge).
Within the Upper Estuarine Series, however, there is found, often in large numbers,
a species of ostracod which undoubtedly belongs to the genus Macrodentina. Here
the hinge is a much more primitive type with a dentate/loculate median element.
This form extends the range of the genus down into the Middle Jurassic and a new
subgenus, Mediodentina, is erected to contain it.

Subgenus MEDIODENTINA nov.

DiacGnosis. Macrodentina with entomodont hinge. Other details as for genus.
Type SPECIES. Macrodentina (Mediodentina) bathonica sp. nov.

Macrodentina (Mediodentina) bathonica sp. nov.
(Pl. 17, figs. 1-12, Pl. 18, figs. 1-4)

Diacnosis. Mediodentina having sub-quadrate to sub-rectangular carapace with
prominent cardinal angles. Tendency to develop small marginal denticles antero-
and postero-ventrally. Dimorphic. Shell surface strongly punctate with large,
circular, normal pore canal openings. Punctation may be so well developed as to
produce almost reticulate type ornamentation.

Hototyre. lo.2550, female carapace, bed Q, Kings Cliffe.

PARATYPES. I1o.2551~72, male and female carapaces and single valves from beds
J, Q & R, Kings Cliffe and bed L, Ketton.

DEscRIPTION. Carapace subquadrate in the female dimorph, sub-rectangular
in the more elongate male dimorph. Greatest length in both sexes passes through
mid-point with the greatest height through the anterior cardinal angle and the greatest
width in the posterior third. Anterior high, broadly rounded; posterior quite broad
and also rounded, although the short, almost vertical postero-dorsal slope gives a
concavity just below the posterior cardinal angle. Both cardinal angles are
prominent but the posterior one is especially so. Dorsal margin slightly convex in
the female and with a very shallow concavity behind the anterior cardinal angle.
In the male these features are more strongly exaggerated. Ventro-lateral margin

56 BATHONIAN UPPER ESTUARINE OSTRACODA, I

convex, overhanging the ventral surface. Ventral margin antero-medially incurved.
Left valve larger than the right which it overlaps along the ventral margin. In the
region of the antero- and postero-dorsal slopes the valve relationship is one of over-
reach rather than overlap. Shell surface strongly punctate, the punctation in a
single specimen almost giving rise to reticulation—the punctae being enlarged almost
to size of pits. Normal pore canal openings are large and circular and both
evenly and widely spaced over the carapace. Small marginal denticles are commonly
developed in the postero-ventral and antero-ventral regions. Hinge strongly
entomodont. Left valve with oval, coarsely loculate terminal sockets and a coarsely
dentate median bar, the denticles being greatly enlarged in the anterior portion.
Accommodation groove shelf-like, though rather narrow, particularly in the male
dimorph. Right valve with 6 anterior and 6 posterior, dorsally bifid, teeth. Median
groove coarsely loculate and greatly expanded in its anterior half. Inner margin
and line of concrescence coincide, duplicature being of moderate width. Anterior
radial pore canals 8 in number, straight and widely spaced; only 2-4 posteriorly.
Right valve has a prominent “ lip’ developed just below the median incurvature.
Muscle scars of type A. The four adductor scars are rather small as is the round
antennal scar which is antero-dorsal in position.

Dimensions. Holotype. Io.2550, female carapace, length 0:86 mm.; height
o-5I1mm.; width 0:42 mm.

Paratypes. lo.2551, male carapace, length 0:85 mm.; height 0-45 mm.; width
037mm. Io.2552, male left valve, length 0°85 mm.; height 0-46mm. I[o0.2553,
male left valve, length 0-95 mm.; height 051mm. Io.2554, female left valve,
length 0-84mm.; height 054mm. Io.2555, immature female left valve, length
0-65 mm.; height 040mm. Io0.2556, female right valve, length 0-68 mm.; height
043mm. I[o.2558, juvenile left valve, length 0-67 mm.; heighto-4r mm. _ Io.2559,
juvenile right valve, length 0-61 mm.; height 0:38 mm.

REMARKS. In some ways this species resembles the ostracod Indet. genus sp. A.
Oertli (1957 : 676) but may be distinguished by the possession of more prominent
cardinal angles, especially posteriorly, and the convexity of the ventro-lateral
margin. None of these features is particularly well developed in Oertli’s material.

Genus MARSLATOURELLA Malz 1959
Marslatourella bullata sp. nov.

(Pl. 18, figs. 5-14, Pl. ro, figs. 1, 2)

Diacnosis. Marslatourella with subquadrate carapace; prominent eye node
with short vertival ridge extending ventrally from it. Ventro-lateral margin with
two knob-like extensions. Shell surface punctate with anterior denticles. A small
“lip” of right valve overlaps left mid-ventrally. Carapace equivalve.

HoLotyPe. Io.2573, right valve from bed H, Ketton.

PARATYPES. Io.2574~-81, single valves and complete carapaces from beds H & N,
Ketton; bed R, Kings Cliffe and bed J, Kettering.

BATHONIAN UPPER ESTUARINE OSTRACODA,I 57

DESCRIPTION. Carapace subquadrate with well rounded anterior; rounded-
triangular posterior; straight dorsal margin with acute cardinal angles and strongly
concave (medially) ventral margin. Male dimorph slightly more elongate than the
female for a given height. Greatest length passes through mid-point with the
greatest height situated at the anterior cardinal angle. Greatest width median.
The carapace here is equivalve with the left and right valves fitting side by side.
The left valve very slightly projects beyond the right terminally whilst mid-ventrally,
a small“ lip ”’ of the right valve overlaps the left. A narrow flange extends around
the free margin of each valve and especially around the anterior is broken up into
a number of small denticles. Shell surface punctate. Ventro-lateral margin
extended into 2 prominent knob-like processes which terminally tend to flatten out
and become blade-like. Eye node prominent at the anterior cardinal angle with a
short, straight ridge extending ventrally from it. Normal pore canal openings
small and rather numerous. Hinge delicately antimerodont. Left valve with a
long, finely denticulate median bar and small terminal sockets which appear to be
ventrally open to the interior of the valve. Right valve with a long narrow groove,
presumably locellate and finely dentate terminal ridges, the exact number of teeth
involved not determined. Inner margin and line of concrescence coincide,
radial pore canals straight, rather thin and well spaced, approximately 16 anteriorly
and 6 posteriorly. Muscle scars as seen through the carapace appear to be of
type A, the antero-dorsal antennal scar being round.

Dimensions. Holotype. lo.2573, female right valve, length 0-68 mm.; height
0:39 mm.

Paratypes. lIo.2575, male carapace, length 0-78 mm.; height 0-41 mm.; width
034mm. Io.2578, female left valve, length 069 mm.; height 0-40 mm. [0.2579,
female carapace, length 0-77 mm.; height 0-43 mm.; width 0:37 mm.

Remarks. WM. bullata sp. nov. is larger than the type species (MM. exposita Malz
1959 : 20) of which Dr. Malz kindly sent me material for comparison purposes.
M. exposita has an average length of 0-51-0559 mm. WM. bullata further differs from
the type species in the form of the ventro-lateral outgrowths which are knob-like
rather than alate and in its more rounded posterior. The denticulate flange,
especially around the anterior margin is a further distinguishing feature of this species.

Genus MICROPNEUMATOCYTHERE Bate 1963
Micropneumatocythere postrotunda sp. nov.
(Pl. ro, figs. 3-10, 13-16)

Diacnosis. Micropneumatocythere having ovoid rather tumid carapace; arched
dorsal outline; broadly rounded posterior; more narrowly rounded anterior. Shell
surface punctate with 3 longitudinal ridges extending along ventral surface of each
valve. Normal pore canal openings fairly large. Low swelling occurs just below
anterior cardinal angle. Internal details as for genus.

58 BADHONIAN UPPER VE sm ULAR TNE Ost RAC OD Aved

PARATYPES. lo.2583-Q1, single valves and carapaces from beds J, P & R, Kings
Cliffe.
HoLoTyPeE. Io.2582, carapace, bed P, Kings Cliffe.

DESCRIPTION. Carapace tumid, ovoid in outline with terminally flattened
broadly rounded posterior and a more narrowly rounded blunted anterior. Dorsal
margin arched with rounded cardinal angles. Ventro-lateral margin medially
swollen, overhanging the ventral surface in lateral view. Ventral margin incurved
antero-medially. Greatest length of carapace passes through mid-point with both
the greatest length and width median. Shell surface very finely punctate with
widely spaced normal pore canals prominent. Ventral surface of carapace weakly
ridged. A low swelling below the anterior cardinal angle may represent an eye
swelling. Left valve overlaps the right along the ventral margin but diverges away
from the right as it approaches the anterior where there is no overlap. Posteriorly,
however, the overlap by the left valve continues round to the posterior cardinal
angle. Dorsally the left valve over-reaches the right. Hinge antimerodont. Left
valve with elongate loculate terminal sockets and a long, denticulate median bar,
above which there is a shallow accommodation groove. Right valve with 7 posterior
teeth and 6 anterior teeth. All the teeth are dorsally bifid. Median groove locellate.
Inner margin and line of concrescence coincide, the duplicature being rather
narrow. Radial pore canals short and thick, widely spaced; 8 anteriorly, 4
posteriorly. Muscle scars of type A, situated fairly low down in the anterior part
of the carapace. The 4 adductor scars are situated in a subvertical row with the
most dorsal scar being slightly offset to the anterior. Antennal scar large, round
and antero-dorsal in position. The round mandibular scar has an antero-ventral
position.

Dimensions. Holotype. lIo.2582, carapace, length 0-60 mm.; height 0-4 mm. ;
width 0:33 mm.

Paratypes. Io.2583, carapace, length 064mm.; height o44mm.; width
034mm. Io.2584, left valve, length 064mm.; height 043mm. I[0.2585, left
valve, length 0:56mm.; height 038mm. Io.2586, right valve, length 0-49 mm.;
height 0°32 mm.

REMARKS. Muicropneumatocythere postrotunda sp. nov. is easily distinguishable
from all other species of the genus by its well rounded somewhat flattened posterior
and blunt, narrower anterior. Dimorphism has not been observed for this species.

Micropneumatocythere quadrata sp. nov.
(Piro; figsw ii 12 pole ZO ese 12)

DiacGnosis. Muicropneumatocythere, with sub-quadrate rather deep carapace
(elongate in male dimorph); cardinal angles prominent, especially posterior angle.
Postero-dorsal slope steeply angled; postero-ventral slope characteristically broad,

BATHONIAN UPPER ESTUARINE OSTRACODA, I 59

deep, strongly convex. Shell surface punctate with widely spaced, large, normal
pore canals. Internal details as for genus.

HoLtotype. lo.2592, female carapace, bed P, Kings Cliffe.

PARATYPES. Ilo.2593-2606, male and female carapaces and single valves from
beds F, O, P, R & S, Kings Cliffe.

DESCRIPTION. Carapace subquadrate with high, arched dorsal outline in the
female dimorph, elongate in the male. Cardinal angles especially the posterior angle
sharply defined, the anterior angle in the left valve is, however, broadly rounded.
Anterior broadly rounded, posterior triangular and slightly upturned at extremity.
Dorsal margin more strongly convex in the left valve than in the right. Antero-
dorsal slope broadly convex; postero-dorsal slope steep, slightly concave in the left
valve but more strongly so in the right. Postero-ventral slope broadly convex.
Ventral margin antero medially incurved. Ventro-lateral margin typically swollen
and overhanging the ventral surface in lateral view. Posterior border compressed
at extremity. Greatest length of carapace passes through mid-point, whilst both
greatest height and width are median. Shell surface finely punctate, over which
large normal pore canal openings are scattered. Ventral surface with 4-5
longitudinal ridges per valve. Left valve larger than the right which it overlaps
along the ventral margin except medially where a small section of the right valve
slightly overlaps the left. Around the anterior and along the antero-, and postero-
dorsal slopes the left valve slightly over-reaches the right. Mid-dorsally, however,
the over-reach of the right valve by the left is very strong, the elongate, triangular
accommodation groove of the left valve being completely visible. Hinge anti-
merodont. Left valve with elongate, loculate terminal sockets, a strong, dentate
median bar and an elongate triangular accommodation groove. Right valve with
6 (dorsally bifid) anterior teeth and 7, also bifid, posterior teeth. Median groove
loculate. Inner margin and line of concrescence coincide; duplicature of
moderate width. Radial pore canals few in number, widely spaced and tend to
widen towards the inner margin. 7-8 anteriorly; approximately 3 posteriorly.
Muscle scars not observed.

Dimensions. Holotype. lo.2592, female carapace, length 0:56 mm.; height
039 mm.; width 0:35 mm.

Paratypes. Io.2593, male carapace, length 068 mm.; height 0-41 mm.; width
037mm. Io.2594, female right valve, length o-61 mm.; heighto-38 mm. [0.2595,
female left valve, length 0:55 mm.; height o-40omm. Io.2596, female carapace,
length 0-56 mm.; height 0-41 mm.; width 0:37 mm.

Remarks. Mucropneumatocythere quadrata sp. nov. is distinguished from all
other species of the genus by the sharply angled posterior cardinal angle coupled
with the broad, deep, postero-ventral slope. Although tending to vary with
individuals, the left valve rather more strongly overreaches the right valve than is
usual for the genus.

60 BATHONIAN UPPER ESTUARINE OSTRACODA, I
Micropneumatocythere subconcentrica (Jones)
(Pl. 21, figs. r-13)
1884 Cythere subconcentrica Jones : 768, pl. 34, figs. 28, 29.

Diacnosis. Micropneumatocythere with oval carapace, tapering to posterior.
Ventro-lateral margin swollen. Shell surface punctate laterally. Ventral surface
with longitudinal ridges extending onto ventro-lateral margin and turning upwards
anteriorly and posteriorly. Internal details as for genus.

LECTOTYPE. Selected here. In.43505, left valve, Great Oolite; Richmond
boring at depth of 1,151 ft. 6 in. figured Jones 1884, pl. 34, fig. 28. Although Jones
refers to his fig. 28 as being of a right valve, the illustration appears to be a left valve.
Specimen In. 43505 is considered to be the one illustrated.

OTHER MATERIAL. I[o.2606—12, carapace and single valves from bed F, Kings
Cliffe and bed H, Ketton.

DESCRIPTION. Carapace ovoid with arched dorsal outline and rounded cardinal
angles. Posterior tapering, slightly rounded with concave postero-dorsal slope and
convex postero-ventral slope. Ventro-lateral margin swollen projecting below
ventral surface in side view. Anterior rounded. Ventral margin antero-medially
incurved. Line of greatest length passes through mid-point, greatest height and
width median. Shell surface laterally punctate, ventrally ridged, the ridges passing
onto the ventro-lateral margin and sub-concentrically turn upwards anteriorly and
posteriorly. This rather weak ornamentation rapidly dies out upwards. Normal
pore canals are widely spaced over the carapace, their openings being rather large
and circular. Left valve larger than the right which it overlaps along the ventral
margin except for a small median section where the right valve overlaps the
left. Dorsally the left valve over-reaches the right. Hinge antimerodont. Left
valve with elongate, loculate, terminal sockets, a denticulate median bar and a
deepish accommodation groove. Right valve with approximately 5-6 terminal
teeth and a long, delicate, locellate groove. Muscle scars of type A, the small,
round, antennal scar having an antero-dorsal position. Inner margin coincides
with the line of concrescence apart from antero-medially where a very narrow
vestibule may be present, but not in all cases. Radial pore canals few in number,
straight and widely spread, 7 anteriorly, 3 posteriorly. A narrow flange may be
preserved around the anterior margin but as this is a very delicate structure it is
usually lost.

Dimensions. Lectotype. IN.43505, left valve, length o56mm.; height
038mm. Other material. Io.2607, carapace, length 0:57 mm.; height 0°37 mm. ;
width 032mm. _ Io.2608, left valve, length 0:51 mm.; height 034mm. Io.2609,
right valve, length o-49 mm.; heighto-30 mm. Io. 2610, left valve, length 0-50 mm.;
height 0:34 mm.

REMARKS. Muicropneumatocythere subconcentrica (Jones) resembles M. globosa
Bate (1964 : 12) in outline but differs in being a much larger species and in having

BATHONIAN UPPER ESTUARINE OSTRACODA, I OL

a more prominent ornamentation of ridges on the ventro-lateral margin. Dimorphism
has not been observed for M. subconcentrica although it is well developed in M.
globosa. The close similarity of the two species suggests, however, that the present
species is a descendant of M. globosa.

Family TRACHYLEBERIDIDAE Sylvester-Bradley 1948
Subfamily TRACHYLEBERIDINAE Sylvester-Bradley 1948
Genus OLIGOCYTHEREIS Sylvester-Bradley 1948

Oligocythereis fullonica (Jones & Sherborn)

(Pl/2x, figs. 14, 15)
1888 Cytheris fullonica Jones & Sherborn : 256, pl. 4, fig. 13a—c.
1948 Cythereis cf. fullonica Jones & Sherborn; Sylvester-Bradley : 186, pl. 12, figs. 7-10;
pl. 13, figs. 3, 9.

1948a Oligocythereis fullonica (Jones & Sherborn) Sylvester-Bradley : 796, pl. 122, figs. 1-6.
1962 Oligocythereis cf. fullonica (Jones & Sherborn) ; Brand & Fahrion : 150, pl. 21, fig. 27.
1963 Oligocythereis fullonica (Jones & Sherborn) ; Oertli: 41, pl. 25 fig. a, pl. 26, fig. a.

REMARKS. Only two specimens of this species have been found within the
Estuarine Series and these occur in the Ancaster bed W (possibly Gt. Oolite),
and Dane Hill sections. Both are ornamentally different. Sylvester-Bradley
(1948 : 187) records three varieties within the species but did not have sufficient
material to determine whether this was of systematic significance. The same is
true here.

Family Uncertain.
Genus PLATYCYTHERE nov.

DiaGnosis. Cytheracea with subquadrate to subrectangular carapace, laterally
flattened with thickened overhanging dorsal and ventral areas. Dorsal and ventral
surfaces flattened. In dorsal view parallel sided or slightly diverging to venter.
Ornamentation strongly to weakly reticulate. Left valve very slightly larger than
right. Hinge entomodont, muscle scars type A. Inner margin and line of concresc-
ence coincide. Radial pore canals straight, few, widely spaced; 8 anteriorly, 3
posteriorly. Normal pore canals large, few, widely and irregularly scattered over
carapace.

TYPE SPECIES. Platycythere verriculata sp. nov.

RemArRKs. Although possessing characters which identify the genus as belonging
to the Cytheracea it is not possible at this stage to place it in any known family.
The unusually flattened carapace, provides only a narrow internal space for the
animal when closed, whilst the swollen dorsal and ventral parts of the carapace
might have had a functional use such as acting as stabilizers whilst crawling over
bottom sediment.

62 BATHONIAN UPPER ESTUARINE OSTRACODA, I

Platycythere verriculata sp. nov.

(Pl. 22, figs. 1-13)

Diacnosis. Platycythere with subquadrate/subrectangular carapace flattened
laterally in dorsal view. Ventro-lateral and dorso-lateral margins swollen, projecting
beyond lateral surface. Dorsal and ventral surfaces also flattened. Shell surface
coarsely reticulate with smooth marginal border in male dimorph, tending to be
smooth entirely in juvenile instars and in female dimorph, although some ornamen-
tation usually observed in region of dorsal and ventral swellings. Ventral surface
ornamented with longitudinal ridges in both dimorphs. Internal details as for
genus.

HoLotyre. Io.2613, male carapace bed J, Kettering.

PARATYPES. lo.2614—22, female and juvenile carapaces and male valves, beds
A, G & J, Kettering and bed N, Kings Cliffe.

DESCRIPTION. Carapace subquadrate in the female dimorph, elongate, subrect-
angular in the male. Carapace typically flattened in dorsal view, the sides being
parallel in the male but tending to slope outwards towards the venter in the female.
Ventro-lateral and dorso-lateral margins swollen, projecting beyond the lateral
surface when viewed dorsally. Dorsal and ventral surfaces flattened. The female
carapace tapers towards the posterior, whilst the male tends to have dorsal and
ventral margins almost parallel. Anterior broadly rounded, posterior rounded, but
more narrowly so. The ventral margin has only a very shallow antero-median
incurvature; dorsal margin almost straight sloping gently or strongly towards the
posterior depending upon the dimorph. Greatest length of carapace extends
through mid-point whilst greatest height is approximately median in the female and
either at the anterior cardinal angle or just behind middle in the male. Greatest
width possibly through middle of carapace in the female, whilst in the male the
parallel lateral margins give no point at which the carapace is widest. Shell surface
coarsely reticulate in the male dimorph, the reticulations tending to die out towards
the anterior and posterior so that an unornamented border is apparent. In the
female dimorph and in juvenile instars there is almost no trace of the reticulate
ornamentation save in the region of the ventral and dorsal swellings. Ventral
surface ornamented with 5-6 longitudinal ridges per valve in both dimorphs.
Normal pore canals very widely and irregularly scattered over the carapace, few
in number and rather large, circular or almost hexagonal in outline. Left valve
slightly overlaps the right at the cardinal angles and, particularly in the female
slightly projects beyond the right around the anterior and posterior. Ventrally the
two valves have neither overlap nor overreach. Hinge entomodont. Left valve
with elongate terminal sockets, presumably loculate but infilled with matrix, and
a long denticulate median bar, the anterior portion of which is very coarsely dentate.
No accommodation groove. Right valve with 6 strong anterior teeth and 5 much
smaller and weaker posterior teeth. Median groove loculate, expanded and more
coarsely loculate in its anterior portion. Muscle scars of type A, the four smali

BATHONIAN UPPER ESTUARINE OSTRACODA, I 63

adductor scars arranged in a slightly crescentic row with the small, round, antennal
scar positioned antero-dorsally. Antero-ventrally the mandibular scar is also
rounded and is very much larger than those previously described. Inner margin
and line of concrescence coincide; duplicature of moderate width, radial pore
canals straight, few in number and widely spaced, 8 anteriorly and 3 posteriorly.

DimEnSIoNS. Holotype. lo.2613, male carapace, length 0:67mm.; _ height
037 mm.; width 0:30 mm.

Paratypes. Io.2614, male right valve, length 066mm.; height 037 mm.
Io.2615, male right valve, length 054 mm.; height 031mm. [o-2616, male left
valve, length 066mm.; height 037mm. I[o.2621, female carapace, length
060 mm.; height 0:37mm.; width 0:26mm. Io.2622, juvenile carapace, length
0-47 mm.; height 0:30 mm.; width 0-19 mm.

Remarks. Platycythere verriculata sp. nov. is quite unlike any previously described
species and belongs to a distinct genus. The variation present within the species
with regard to the ornamentation could suggest the presence of two distinct species.
However, the outline of the smooth species suggests that the most probable explana-
tion is that this is the female dimorph, the smaller juvenile instars being similarly
without ornamentation. Only in the more elongate male dimorph is the reticulate
ornamentation strongly developed over the lateral surface.

The width measurements given above are for the carapace excluding the swollen
dorsal and ventral parts and are taken at valve centre.

Ill. PALAEOECOLOGY

The work undertaken by Aslin has shown there to be a variable and alternating
succession of both marine and freshwater sediments exposed along the entire outcrop
of the Upper Estuarine Series. The study of the ostracod faunas is, therefore,
doubly interesting in that the stratigraphy of the succession has been examined in
some detail.

The base of the Upper Estuarine Series is dominated by purplish grey clays with
plant remains representing a continental period of deposition. The absence of
ostracods from these beds and from the freshwater horizons which return from time
to time throughout the succession is almost certainly due to decalcification. The
marine horizons are more strongly calcified and contain a good macro- and micro-
fauna, the evidence of which suggests deposition close to land and influenced by
river effluent. As such both brackish water and marine ostracods are encountered.
Completely freshwater faunas are probably not encountered for the reasons mentioned
above, although faunas of low salinity are recorded and represent ostracod populations
living some distance upstream from a river mouth or in a more static back swamp
body of water. These low salinity populations are represented by the following
species: Darwinula incurva, Bisulcocypris ancasterensis, Belekocytheridea punctata
and Macrodentina (Mediodentina) bathonica at Ancaster, and at Kings Cliffe by
Darwinula incurva, Bisulcocypris anglica, Klieana levis, Macrodentina (Medtodentina)
bathonica and Muicropneumatocythere postrotunda. Both populations are found

64 BATHONIAN UPPER ESTUARINE OSTRACODA, I

associated with freshwater-brackish-water charophytes. The two _ species of
Bisulcocypris together with Darwinula incurva are considered to represent freshwater-
oligohaline ostracods whilst the species of Klieana, Macrodentina, Belekocytheridea
and Micropneumatocythere are considered to be euryhaline and found more typically
in brackish water assemblages.

The evidence in support of the freshwater-oligohaline habit of the species mentioned
above is as follows:

Firstly, the sediments in which the ostracods were found occur in that part of
the rythmic sequence which is subject to the greatest freshwater influence.

Secondly, the ostracod population is restricted in the number of species present
and lacks the normal marine ostracods (e.g. Lophocythere and Progonocythere).

Thirdly, the reported occurrence elsewhere of the ostracod genera concerned by
other authors substantiates the findings here. For example Klieana levis Oertli
(1957 : 760) was originally described from lacustrine sediments from Poitou in
France whilst Van Morkhoven (1963 : 268) records the environmental range of the
genus as being from fresh to brackish water. Darwinula is an essentially freshwater
genus with some species ranging into brackish water. The type species, Darwinula
stephensont (Brady & Robertson 1870) is living at the present time in rivers and
lakes in East Anglia from which area it was originally described. Van Morkhoven
(1963 : 29) and Moore (r1g6r : 0254) give the ecological range of the genus as fresh-
brackish water. The genus Bisulcocypris is regarded by Pinto & Sanguinetti
(1962 : 75) as being typically freshwater. Malz in his excellent work on the genus
Macrodentina gives the ecological habit of each of the subgenera comprising the
genus as being either marine or marine to brackish. The new subgenus Medtodentina
is considered to have a wide range from marine through brackish to almost freshwater
conditions.

In at least two horizons in the Kettering section mixed assemblages of truly marine
ostracods such as Lophocythere scabra; Pichottia magnamuris, Marslatourella bullata
and Schuleridea (Eoschuleridea) bathonica are found associated with the euryhaline
species of Klieana and Macrodentina and the oligohaline species of Bisulcocypris and
Darwinula, This is not a typical assemblage and results either through the action
of rivers bringing into the area ostracods which normally inhabit waters of lower
salinity or as is more likely in this case, transgression of marine conditions results
in the mixing of faunas, the low salinity ostracods being killed but preserved in the
sediments deposited.

The euryhaline ostracods present in the Upper Estuarine Series have been mentioned
already but in addition to these a number of normally marine species may also be
found under brackish conditions. These are: Glyptocythere guembeliana; Platycythere
verriculata and the two remaining species of Micropneumatocythere.

As has been suggested above regions of deposition where periods of marine trans-
gression are common are liable to produce mixed assemblages. Blurring of bound-
aries between freshwater, brackish water and marine conditions is normal and results
in mixing of faunas to some extent. There are, however, four ways in which mixing
of ostracod faunas may result.

BATHONIAN UPPER ESTUARINE OSTRACODA, I 65

1. Normal mixing at the boundaries of salinity zones.

2. Freshwater species brought into brackish or marine conditions by rivers.

3. Transgression of the sea flooding bodies of freshwater and bringing marine
ostracods into the area.

4. Reworking of sediments whether marine or freshwater and the redeposition of
the ostracods present. Thus part of the ostracod fauna would be alloch-
thonous and part autochthonous.

It is more than likely that all four variables have been operative during the deposi-

tion of the Upper Estuarine Series.

The ostracod species which appear to be most truly marine throughout are:
Paracypris terraefullonica; Monoceratina scarboroughensis; Galliaecytheridea? kings-
cliffensis; Pichottia magnamuris; Schuleridea (Eoschuleridea) bathonica; Praeschuleri-
dea quadrata; Metacytheropteron drupacea; Progonocythere levigata; P. rugosa and
P. triquetra; Glyptocythere juglandica; Lophocythere scabra; L. septicostata and L.
transverstplicata; Marslatourella bullata and Oligocythereis fullonica. All the ostra-
cods mentioned as occurring in brackish waters also are found associated with the
more restricted marine species listed above. Only Macrodentina (Medtiodentina)
bathonica, Belekocytheridea punctata and Klieana levis are to be found in all associa-
tions.

IV. REFERENCES

ANDREEV, J. N. & Manverstam, M. I. 1964. On the systematic position of the genus
Ljubimovina. Pal. Zhurn., Moscow, 2 : 152-154.

Astin, C. J. The Upper Estuarine Series of Eastern England Pt. Il. Bull. By. Mus. nat. Hist.
Geology, London. [in press].

Bate, R. H. 1963. Middle Jurassic Ostracoda from North Lincolnshire. Bull. Br. Mus.

nat. Hist. Geology, London, 8 : 173-219, pls. I-15.

1964. Middle Jurassic Ostracoda from the Millepore Series, Yorkshire. Bull. By. Mus.

nat. Hist. Geology, London, 10 : 1-33, pls. 1-14.

1965. Freshwater Ostracods from the Bathonian of Oxfordshire. Palaeontology, London,

8, 4 : 749-759, pls. Iog—11IT.

Buraszyk, J. & Marz, H. J. 1965. Terquemula n.g., eine neue Ostracoden-Gattung aus dem
Ober-Bathonien. Senck. leth., Frankfurt a. M., 46 : 443-451, pl. 36.

Branp, E. & Fanrion, H. 1962. Dogger NW-Deutschlands : 123-158, in Lettfossilien der
Mikropaldontologie. 432 pp., 59 pls. Berlin.

BRAND, E. & Matiz, H. J. 1962a. In BRAND, E. & FauHrion, H. Dogger NW-Deutschlands :

123-158. In Leitfossihen der Mikvopaldontologie. 432 pp., 59 pls., Berlin.

19626. Ostracoden-Studien im Dogger, 5: Glyptocythere. Senck. leth., Frankfurt a. M.,
43 : 433-435, 2 figs.

CHERNYSHEVA, N. E. (editor) 1960. Osnovy paleontologu ; Chlenistonogie trilobitoobraznie i
vakoobrazmie. 515 pp., 18 pls. Moscow.

GrEKoFF, N. 1963. Contribution a l’étude des Ostracodes du Mésozoique Moyen (Bathonien-
Valanginien) du Bassin de Majunga, Madagascar. ev. Inst. frang. Pétrole, Paris, 18 : 1709—
1762, pls. I-10.

HarpincG, J. P. 1962. Mungava munda and four other new species of ostracod crustaceans
from fish stomachs. The Natural History of Rennel Island, British Solomon Islands,
4:51-62. Copenhagen.

Jones, T. R. 1884. Notes on the Foraminifera and Ostracoda from the deep boring at
Richmond. Quart. J. Geol. Soc. Lond., 11 : 765-777, pl. 34.

66 BATHONIAN UPPER ESTUARINE OSTRACODA, I

Jones, T. R. & SHERBORN, C. 1888. On some Ostracoda from the Fullers-earth Oolite and
Bradford Clay. Pyvoc. Bath nat. Hist. Fld. Cl., 6 : 249-278, pls. 1-5.

IKKOLLMANN, K. 1960. Cytherideinae und Schulerideinae n. subfam. (Ostracoda) aus dem
Neogen des 6stl. Oesterreich. Mutt. Geol. Ges. Wien, 51 : 89-195, 21 pls.

Lyusrmova, P. S. 1956. Ostracods from the Cretaceous deposits of the eastern part of the
Mongolian National Republic and their significance for stratigraphy. Tvud. vses. nef.-
nauch issled. geol. Inst. (VNIGRI), Leningrad (N.S.) 93 : 1-174, pls. 1-25 [in Russian].

—, Kazmina, T. A. & RESHETNIKOVA, M.A. 1960. Ostracods of the Mesozoic and Caenozoic
deposits of the Western Siberian Lowlands. Tvud. uses. ned.-nauch issled. geol. Inst
(VNIGRI), Leningrad (N.S.) 160 : 1-427, pls. 1-24 [in Russian].

Lutze, G. F. 1960. Zur Stratigraphie und Palaontologie des Callovien und Oxfordien in
Nordwest-Deutschland. Geol. Jb., Hannover, 77 : 391-532, pls. 26-42.

LYELL, C. 1855. A Manual of Elementary Geology. 5thed. 655 pp., 717 figs. London.

Matiz, H. J. 1958. Die Gattung Macrodentina und einige andere Ostracoden-Arten aus dem

Oberen Jura von Nordwestdeutschland, England und Frankreich. Abh. senckenb. naturf.

Ges., Frankfurt a.M., 497 : 1-67, 11 pls.

1959. Ostracoden-Studien im Dogger, 1: Marslatourella n.g. Senck. leth., Frankfurt

a. M., 40: 19-23, figs. 1-14.

Manpve stam, M. I. 1947. Ostracoda from Middle Jurassic Deposits of the Mangislaka
Peninsula (Eastern side of the Caspian). Jn Microfauna, Petroleum Occurrence, Caucasus
Emba and central Asia. WVNIGRI, Leningrad-Moscow : 239-259, 2 pls. [in Russian].

Martin, G. P. R. 1961. Die Gattung Fabanella n.g. (Ostracoda) im NW-deutschen Malm und
Wealden. Senck. leth., Frankfurt a. M., 42 : 181-195, pl. 1.

Moore, R. C. (editor) 1961. Tveatise on Invertebrate Paleontology, Pt. Q., Arthropoda, 3.
Xxiiit+ 442 pp., 334 figs. Kansas.

OERTLI, H. J. 1957. Ostracodes du Jurassique Supérieur du Bassin de Paris (Sondage Vernon

1). Mev. Inst. frang Pétrole, Paris, 12 : 647—695, pls. 1-7.

1957a. Ostrakoden als Salzgehalts-Indikatoren im obern Bathonien des Boulonnais.

Ecol. Geol. Helv., Basel, 50 : 279-283, 3 figs.

— In BERNARD, F., Bizon, J. J. & OERTLI, H. J. 1957. Ostracodes Lacustres du Bathonien

du Poitou (Bassin de Paris). Bull. Soc. géol. France, Paris, 6 : 753-770, pls. 21-23.

1959. Les Ostracodes du Bathonien du Boulonnais, 1. Les ‘“‘ Micro-ostracodes ’’.

Rev. Micropaléont., Paris, 2 : 115-126, pls. 1-3.

1963. aunes d’ostrvacodes du Mésozoique de France. 57 pp.,9o0pls. Leiden.

Pinto, I. D. & Sancurinett1, Y. T. 1962. A complete revision of the genera Bisulcocypris
and Theriosynoecum (Ostracoda) with the world geographical and stratigraphical distribu-
tion (including Metacypris, Elpidium, Gomphocythere and Cytheridella). Esc. Geol. P.
Alegre, 4: 1-165, 17 pls.

ScumiptT, G. 1955. Stratigraphie und Mikrofauna des mittleren Malm im nordwest-deutschen
Bergland. Abh. Senckenb. naturf. Ges., Frankfurt a. M., 491 : 1-76, pls. 1-18.

Swartz, F. M. & Swain, F. M. 1946. Ostracoda from the Upper Jurassic Cotton Valley
Group of Louisiana and Arkansas. J. Paleont., Tulsa, 20 : 362-373, pls. 52, 53.

SYLVESTER-BRADLEY, P. C. 1948. Bathonian Ostracods from the Boueti Bed of Langton

Herring, Dorset. Geol. Mag., London, 85: 185-204, pls. 12-15.

1948a. The ostracode genus Cythereis. J. Paleont., Tulsa, 22 : 792-797, pl. 122.

1956. The structure, evolution and nomenclature of the ostracod hinge. Bull. Br.

Mus. nat. Hist. Geology, London, 3 : 1-21, pls. 1-4.

—— & Pinto, I. D. A Lower Bathonian freshwater ostracod of the genus Theriosynoecum
from the Viviparus Marl of Oxfordshire. (MS).

TRIEBEL, E. 1951. Einige stratigraphisch wertvolle Ostracoden aus dem hodheren Dogger
Deutschlands. <Abh. senckenb. naturf. Ges., Frankfurt a. M., 485 : 87-101, pls. 44-49.

VAN MorkKHOvEN, F. P. C. M. 1963. Post-Palaeozoic Ostracoda, 2. 478 pp., 763 figs.
Amsterdam.

IRIE AID IT, 3

Paracypris terraefullonica (Jones & Sherborn) Pp. 27

Fic. 1. Internal view, right valve. Jo.2251. X85.

Fic. 2. External view, left valve. Lectotype, 1.1875. x85.

Fic. 3. External view, right valve. 1.1874. X85.

Fics. 4-6. Right, dorsal and ventral views, complete carapace. Io.2250. X85.

Darwinula incurva sp. nov. p. 28

Fic. 7. External view of crushed specimen, left side, showing muscle scars. Paratype,
10.2274. X70.

Fic. 8. Muscle scars. Paratype, lo.2274. 180.

Fic. 9. Internal view of left valve showing denticulate ventral margin. Paratype, Io.226r.
x 70.

Fics. to-12. Ventral, left and right views, complete carapace. Holotype, Io.2259. 7o.

Bull. Br. Mus. nat. Hist. Geology, 14, 2 PAE

GEOL. 14, 2. 8

PLATE 2

Bisulcocypris anglica sp. nov. p. 30

Fics. 1-4. Right, left, ventral and dorsal views, male paratype. Io.2277. x70.
Figs. 5-7. Dorsal, right and left views, female holotype. Io.2275. x70.
Figs. 8-11. Left, right, dorsal and ventral views, juvenile paratype. Io.2280. 70.

Bull. By. Mus. nat. Hist. Geology, 14, 2

PLATE 3

Bisulcocypris ancasterensis sp. nov. p. 32

Fics. 1-3. Left, right and dorsal views, female holotype. Io.2282. x7o.
Fies. 4-6. Left, right and ventral views, male paratype. Io.2284. x70.

Fics. 7-10. Right, left, dorsal and ventral views, juvenile paratype. Io.2285.

Monoceratina scarboroughensis Bate Pp. 33
Fic. 11. Right view, female carapace. Io.2302. X85.

x 70.

PLATE 3

Bull. Br. Mus. nat. Hist. Geology, 14, 2

PLATES

Fabanella bathonica (Oertli) P. 33

Fic. 1. Internal view, female left valve to show radial pore canals. Io.2305. 70.
Fics. 2, 3. External and internal views, male right valve. JIo.2303. X7o.
Fics. 4, 5. Internal and external views, female left valve. Io.2304. x7o.

Galliaecytheridea? kingscliffensis sp. nov. Pp. 34

Fics. 6, 7. Right and left views, female carapace. Holotype, Io.2316. x 8o.

Fic. 8. Dorsal view of hinge, male left valve. Paratype, lo.2322. 100.

Fic. 9. Internal view, female left valve. Paratype, lo.2321. x 8o.

Fic. 10. Dorsal view showing median bar of hinge, female left valve. Paratype, Io.2321.
x 100.

Fic. 11. Internal view, female right valve. Paratype, Io.2326. x85.

Fic. 12. Muscle scars, female right valve. Paratype, lo.2326. 250.

Bull. Br. Mus. nat. Hist. Geology, 14, 2 PIL INARID, 7

PLATE 5
Galliaecytheridea? kingscliffensis sp. nov. Pp. 34

Fic. 1. Internal view to show radial pore canals, female left valve. Paratype, Io.2321.
x 85.

Fics. 2, 3. Dorsal and ventral views, female carapace. Holotype, Io.2316. x 8o.

Fic. 4. External view, female left valve. Paratype, Io.2321. 85.

Fics. 5-8. Left, right, dorsal and ventral views, male carapace. Paratype, Io.2318. x 8o.

Belekocytheridea punctata gen. et sp. nov. Pp. 36
Fics. 9-12. Left, right, dorsal and ventral views female carapace. Holotype, Io.2328.
x 85. .
Fic. 13. Right side showing muscle scars, female carapace. Paratype, Ilo.2331. X85.

PLATE 5

Bull. Br. Mus. nat. Hist. Geology, 14, 2

°° =":

oe

aan?

~= .
ong

PLATE 6

Belekocytheridea punctata gen. et sp. nov. p. 36
Fics. 1-4. Left, right, dorsal and ventral views, male carapace. Paratype, lo.2329.
x85.
Fig. 5. Internal view, female left valve. Paratype, Io.2357. X85.

Pichottia magnamuris sp. nov. p. 38

Fias. 6-9. Left, right, dorsal and ventral views, female carapace. Holotype, Io.2358.
x 85.

Fic. 10. Dorsal view to show median hinge bar, male left valve. Paratype, Io.2366.
x 100.

Fic. 11. Dorsal view of hinge, female left valve. Paratype, lo.2362. 100.

Fic. 12. Dorsal view of hinge, female right valve. Paratype, Io.2363. 100.

Fic. 13. Internal view to show radial pore canals, male right valve. Paratype, lo.2367.
x 100.

Fic. 14. Muscle scars, male left valve. Paratype, lo.2368. 180.

PLATE 6

Bull. Br. Mus. nat. Hist. Geology, 14, 2

te
“=F

Ef “omnes "

PLATE 7

Pichottia magnamuris sp. nov. p. 38

Fics. 1-4. Left, right, ventral and dorsal views, male carapace.
x 85.

Fic. 5. Internal view, female right valve. Paratype, Io.2363. x85.

Fic. 6. Internal view, female left valve. Paratype, Io.2362. x85.

Schuleridea (Eoschuleridea) bathonica subgen. et sp. nov.

Fics. 7-10. Left, right, dorsal and ventral views, female carapace.
x 85.
Fic. 11. Dorsal view of hinge, female right valve. Paratype, Ilo.2376.

Paratype, Io.2360.

p. 41
Holotype, Io.2369.

x 100.

Fics. 12, 13. Female right valve showing radial pore canals. Fig. 12 160, fig. 13 x85.

Paratype, Io.2373.

PLATE 7

Bull. Br. Mus. nat. Hist. Geology, 14, 2

PLATE 8

Schuleridea (Eoschuleridea) bathonica subgen. et sp. nov. p. 41

Fic. 1. Internal view to show hinge, female right valve. Paratype, lo.2373. XIoo.

Fics. 2, 3. External and internal views, male left valve. Paratype, lo.2379. x85.

Figs. 4, 5. Internal and external views, female left valve. Paratype, lo.2375. 85.
Fics. 6, 7. External and internal views, female right valve. Paratype, Ilo.2376. x85.
Figs. 8-10. Dorsal, right and ventral views, male carapace. Paratype, Io.2370. x85.
Fic. 11. Muscle scars, note large size of antennal scar, male left valve. Paratype, Io.2378.

X 300.

Bull. By. Mus. nat. Hist. Geology, 14, 2

ioe ee ea >
HPIRFLTS?

>

.

PLATE 9

Praeschuleridea quadrata sp. nov. p. 42

Fies. 1-4. Right, left, dorsal and ventral views, female carapace. Holotype, Io.2395. x85.
Figs. 5-7. Dorsal view ( x 100), external and internal views, female right valve. Paratype,
Io.2399. X85.
Fics. 8,9. Internal views to show radial pore canals and hinge, female left valve. Paratype,
To.2398. x85.
Fics. 10, 11. Left and dorsal views, male carapace. Paratype, Ilo.2397. x85.
Fic. 12. Muscle scars, male left valve. Paratype, Io.2400. 300.

Bull. By. Mus. nat. Hist. Geology, 14, 2 PLATE 9

GEOL. 14, 2. 9

PLATE to

Metacytheropteron drupacea (Jones) Pp. 44

Fics. 1, 2. External view and internal view showing radial pore canals, male left valve.
MOpArtiee, — S<{sI55.

Fics. 3, 4. Left and right views, female carapace. Jones’ original specimen, holotype,
IN.43498. x85.

Fies. 5, 7. External and internal views, female left valve. Ilo.2414. x85.

Fics. 6, 8, 9. Internal and external views, 85, and dorsal view, x 100, male right valve.
Io.2416.

Pp: 45

Progonocythere levigata sp. nov.
Fic. 10. Dorsal view of hinge to show terminal teeth and plate-like extension of the anterior

portion of the median groove. Female right valve. Paratype, lo.2420. ™X1Ioo.
Figs. 11, 12. External and internal views, female left valve. Holotype, lo.2419. %7o.

Fic. 13. Muscle scars of holotype, lo.2419. 170.
Fic. 14. Internal view, female right valve. Paratype, Io.2420. x70.

Bull. By. Mus. nat. Hist. Geology, 14. 2 PLATE to

PLATE 11

Progonocythere levigata sp. nov. Pp. 45

Fic. 1. Internal view, male right valve. Paratype, Io.2422. X7o.

Fics. 2-4. External and internal views, x70, and muscle scars, x 310, male left valve.
Paratype, Io.2421.

Fics. 5-8. Dorsal, ventral, left and right views, female carapace. Paratype, Io.2423.
x70.

Fic. 9. External view, female right valve. Paratype, Io.2420. X70.

Progonocythere rugosa sp. nov. p. 46

Fics. 10-13. Right, left, dorsal and ventral views, female carapace. Holotype, Io.2434.
X 70.
Fic. 14. Dorsal view, female right valve. Paratype, Io.2438. X1Ioo.

Bull. Br. Mus. nat. Hist. Geology, 14, 2

IPDIL INI SD

Progonocythere rugosa sp. nov. p. 46

Fics. 1-4. Left, right, dorsal and ventral views, male carapace. Paratype, Ilo.2435, x70.

Fics. 5, 7, 9. Internal views to show radial pore canals and hinge, x70, and muscle scars,
x 350, male left valve. Paratype, lo.2437. x70.

Figs. 6, 8. External and internal views, female right valve. Paratype, Io.2438. x7o.

Progonocythere triquetra sp. nov. p. 48

Figs. 10-13. Left, right, dorsal and ventral views, female carapace. Holotype, Io.2453.
X70.

Bull. Br. Mus. nat. Hist. Geology, 14, 2 PLATE 12

PEALE 13

Progonocythere rugosa sp. nov. p. 46
Fic. 2. Dorsal view, hinge, male left valve. Paratype, lo.2437. Xx Ioo.
Progonocythere triquetra sp. nov. p. 48

Fics. I, 5, 6. Dorsal view, X1Ioo, and external and internal views, x70, male left valve.
Paratype, Io.2456.

Fics. 3, 9. Dorsal view, x 100, and external view, x70, female left valve. Paratype,
1o.2454.

Fics. 4, 7, 8. Dorsal view, x 100, and external and internal views, x70, female right valve.

Paratype, 10.2455.
Glyptocythere guembeliana (Jones) Pp. 49

Fic. 10. External view, male right valve. Lectotype, IN.43493. x70.

Fic. 11. External view, male right valve. Io.2466. X70.
External view, female right valve, Jones & Sherborn’s. Cytheridea pulvinar. I. 1858

FIG. 12.
X70.
Fic. 13. Dorsal view of median hinge bar, male left valve. Io0.2465.
Fics. 14, 16. External and internal views, female left valve. Io.2467. X7o.

Fic. 15. Internal view, female right valve. Io.2469. x70.

Bull. Br. Mus. nat. Hist. Geology, 14, 2 PLATE 13

PLATE 14

Glyptocythere guembeliana (Jones) Pp. 49

Fics. 1, 5. External view showing ventral alate extension, and internal view, female right
valve. 10.2472. X70.

Fic. 2. External view of female right valve without alate extension. Io.2469. X7o.

Fics. 3, 4. Dorsal and ventral views, female carapace. 10.2473. X70.

Fics. 6, 7. Internal view, female right valve showing radial pore canals. Fig. 6 150,
fig. 7 x70. I0.2468.

Fic. 8. Muscle scars x 360, female right valve. Io.2468.

Glyptocythere juglandica (Jones) p. 51
Fic. 9. Right side, male carapace. JIo.2516. X70.

Klieana levis Oertli p. 51
Fics. 10, 11, 13. Left, right and ventral views, male carapace. 10.2522. x85.
Fic. 12. Internal view showing hinge, female left valve. JIo.2521. x85.

Bull. Br. Mus. nat. Hist. Geology, 14, 2 PLATE 14

IPIL ANID IS, 2
Klieana levis Oertli p. 51

Internal view, female right valve, to show radial pore canals. Io.2520. X85.

Geis
Fic. 2. Internal view, hinge, female right valve. Io.2520. too.
Fics. 3, 4. Left and right views, female carapace. Io0.2518. X85.

Fic. 5. Muscle scars, female right valve. Io.2520. 240.

Lophocythere scabra scabra Triebel p. 52

Fic. 6. External view, female right valve. Io.2539. x85.

Lophocythere septicostata sp. nov. Pp. 52

Left, right, dorsal and ventral views, female carapace. Holotype, Io.2542.

Fics. 7-10.
x 85.
Fics. 11-13. Internal and external views, x 85, and dorsal view, x 100, female right valve.

Paratype 1.1843. This specimen was originally figured by Jones & Sherborn 1888 as Cytheridea

bradiana (Jones).

Bull. Br. Mus. nat. Hist. Geology, 14, 2 IPIL JAAD, ng

PLATE 16

Lophocythere septicostata sp. nov. Pp. 52
Fics. 1-4. Left, right, dorsaland ventral views, male carapace. Paratype, Io.2547. x85.

Lophocythere transversiplicata sp. nov. P- 53

Fics. 5-7. Left, right and dorsal views, male carapace. Paratype, lo.2626. x85.

Fics. 8-10. Left, dorsal and ventral views, female carapace. Holotype, lo.2625. x85.

Fics. 11, 12. Internal view, x85, and dorsal view, x100, female right valve. Paratype,
Io. 2627.

Fic. 13. External view, male right valve. Paratype, lo.2628. x85.

Fic. 14. External view, male left valve. Paratype, lo.2629. 85.

Fic. 15. Internal view showing radial pore canals, female right valve. Paratype, Io.2627.
x 85.

RAE 6

Bull. Br. Mus. nat. Hist. Geology, 14, 2

PLADE, 17

Macrodentina (Mediodentina) bathonica subgen. et sp. nov. Pp. 55

Fics. 1-3. Left, right and ventral views, female carapace. Holotype, lo.2550. %7o.

Fics. 4, 7,12. External view (x70), internal and dorsal views of hinge ( x 100), female right
valve. Note plate-like extension of the lower margin of the anterior part of the median groove
in dorsal view. Paratype, lo.2556.

Fies. 5, 6. External and internal views, female left valve. Paratype, lo.2554. 70.

Fic. 8. Internal view showing radial pore canals, female right valve. Paratype, lo.2557.
x70.

Fic. 9. Internal view showing radial pore canals, male left valve. Paratype, lo.2552. X70.

Fic. 10. Dorsal view showing median hinge bar, male left valve. Paratype, Io.2552.
x 100.

Fic. 11. Dorsal view showing enlarged antero-median part of median bar, female left valve.
Paratype, 10.2555. 100.

Bull. By. Mus. nat. Hist. Geology, 14, 2 PIAL E17

GEOL. 14, 2. 10

PLATE 18

Macrodentina (Mediodentina) bathonica subgen. et sp. nov. Pp. 55

Fics. 1, 4. External and internal views, male left valve. Paratype, Io.2552. x7o.

Fic. 2. External view to show strength of surface ornamentation in this specimen, masking
the normal pore canals. Male left valve. Paratype, lo.2553. X70.

Fic. 3. Muscle scars, female right valve. The dark ringed circles, including that showing
a diagonal cross-bar, are normal pore canals. The mandibular scar is not shown in this illustra-
tion. Paratype, 1lo.2557. 250.

Marslatourella bullata sp. nov. Pp. 56

Fics. 5-8. Right, left, dorsal and ventral views, male carapace. Paratype, Io 2575 x 70.
Fics.9-12. Right, left, dorsal and ventral views, female carapace. Paratype, 10.2579. X70.
Fic. 13. External view, female right valve. Holotype, lo.2573. x7o.

Fic. 14. Internal view of hinge, female left valve. Paratype, lo.2578. X1Io0o.

Bull. Br. Mus. nat. Hist. Geology, 14, 2 PLATE 18

PLATE 19

Marslatourella bullata sp. nov. p. 56

Figs. 1, 2. Internal view showing radial pore canals and external view. Female left valve.
Paratype, Io.2578. 7o.

Micropneumatocythere postrotunda sp. nov. Pp. 57

Fics. 3-6. Left, right, dorsal and ventral views of carapace. Holotype, Io.2582. 85.

Fics. 7, 8, 13. Internal and external views (x85) and dorsal view (x100), left valve.
Paratype, Io.2584.

Fics. 9, 10, 14. External and internal views (x85) and dorsal view (x 100), right valve.
Paratype, Io.2586.

Fic. 15. Anterior radial pore canals, left valve. Paratype, Io.2585. 160.

Fic. 16. Muscle scars, paratype, Io.2583. 270.

Micropneumatocythere quadrata sp. nov. Pp. 58
Figs. 11, 12. External and internal views, female left valve. Paratype, Io.2595. 85.

Bull. Br. Mus. nat. Hist. Geology, 14, 2 PLATE 19

PLATE 20

Micropneumatocythere quadrata sp. nov. Pp. 58
Fics. 1-4. Left, right, dorsal and ventral views, female carapace. Holotype, Io.2592. x85.
Fics. 5-8. Left, dorsal, right and ventral views, male carapace. Paratype, Io.2593. x85.

Fics. 9, 10, 12. Internal and external views, x85, dorsal view, x100, female right valve.
Paratype, Io.2594.

Fic. 11. Dorsal view, female left valve. Paratype, Io.2595. x Ioo.

PLATE 20

Bull. By. Mus. nat. Hist. Geology, 14, 2

PLATE 21

Micropneumatocythere subconcentrica (Jones) p. 60

—4. Right, left, dorsal and ventral views, carapace. Io.2607. X85.
13. Dorsal view, 100, and internal view to show radial pore canals, x85. Left

To. 2608.
. 6, 11. Dorsal view, 100, and external view, x85, right valve. Io.2609.
Ti

, 8. External and internal views, left valve. Lectotype, IN.43505. x85.

.9, 10. External and internal views, left valve. Jo.2610. X85.

Fic. 12. Muscle scars, left valve. Io.2608. 340. 4

Oligocythereis fullonica (Jones & Sherborn) p. 61

Fic. 14. External view, left valve. Io.2624. x85.
Fic. 15. External view, right valve. Io.2623. x85.

Bull. Br. Mus. nat. Hist. Geology, 14, 2 PLATE 21

PLATE 22

Platycythere verriculata gen. et sp. nov. p. 62

Fics. 1-4. Right, left, dorsal and ventral views, male carapace. Holotype, Io.2613. x85.

Fias. 5-8. Right, left, ventral and dorsal views, female carapace. Paratype, Io.2621t. x85.

Fic. 9. Internal view, male right valve. Paratype, Io.2615, x85.

Fic. 10. Dorsal view showing enlarged anterior portion of median hinge bar. Male left
valve. Paratype, Io.2616. x Ioo.

Fic. 11. Left side of juvenile carapace. Paratype, Io.2622, x85.

Fics. 12, 13. Muscle scars, 220, anterior radial pore canals, 200, male right valve.
Paratype, Io.2614.

PEATE, 22

Bull. Br. Mus. nat. Hist. Geology, 14, 2

mY 7 , 7 OV | i
r ad IDA om , =) 7 ' /
71 ,
‘ ri 7

co aes i i /
y Be, A BOE 2 ve ore ae ‘ei $h2 Mange Ae he
i i ’ re _ “ , ’ 7 ‘ ;
/ . fea! vy"

sp U

/

bo

AT
ESS

ee

IN G
&

_BARTHOLOMEW PR

ADLARD

a
I
B
Z
4
%
Ay

BY

i

§ APR 19

#, ‘
cs A
Cay KS

-

“NEW CRETACEOUS BERYCOID
_ FISHES FROM THE LEBANON

C. PATTERSON

_ BULLETIN OF | ey is

, Vol. 14 NG. 3 is
LONDON: 1967 sae

y
“4
t
4
ne

Pi
i %
>)

NEW CRETACEOUS BERYCOID FISHES \e

FROM. THE LEBANON

BY

COLIN PATTERSON, Ph.D.
SS See

Pp. 67-109 ; 4 Plates ; 11 Text-figures

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)
GEOLOGY Vol. 14 No. 3
LONDON : 1967

THE BULLETIN OF THE BRITISH MUSEUM
(NATURAL HISTORY), istituted im 1949, 1s
issued in five series corresponding to the Departments
of the Museum, and an Historical series.

Parts will appear at irregular intervals as they become
veady. Volumes will contain about three or four
hundred pages, and will not necessarily be completed
within one calendar year.

In 1965 @ separate supplementary series of longer
papers was instituted, numbered serially for each
Department.

This paper is Vol. 14, No. 3 of the Geological
(Palaeontological) series. The abbreviated titles of
periodicals cited follow those of the World List of
Scientific Periodicals.

World List abbreviation :
Bull. Br. Mus. nat. Hist. (Geol.).

© Trustees of the British Museum (Natural History) 1967

PTRUSDE ES Om
THE BRITISH MUSEUM (NATURAL HISTORY)

Issued 7 April, 1967 Pyice fie ss:

NEW GRETAGEOUS BERYCOID” FISHES
FROM THE LEBANON

By COLIN PATTERSON

CONTENTS
Page
I. INTRODUCTION ‘ , ‘ : - d ; : , 69
II. LocALiriEs AND CONDITIONS OF DEPOSITION : : : ‘ 70
Ill. THE AGE OF THE CENOMANIAN FISH BEDS . : ; : : 71
IV. SYSTEMATIC DESCRIPTIONS ; é : : ; ‘ : 73
Order BERYCIFORMES
Suborder Berycoidei

Family TRACHICHTHYIDAE
Lissoberyx day (Smith Woodward) . : : : 73
Gnathoberyx stigmosus gen. et sp. nov. 5 : : 82

Family HOLOCENTRIDAE
Stichocentrus livatus gen. et sp. nov. . : : : 88
Caproberyx pharsus sp. nov. : 2 . ; ‘ 97
V. Discussion . : : ‘ : : . ¢ ; ; 103
VI. REFERENCES . 5 ; : : : : : : : 108
SYNOPSIS

This paper contains descriptions of four berycoids, three from the Cenomanian fish beds of
Hakel and Hajula, Lissoberyx gen. nov. dayi (Smith Woodward) (Trachichthyidae), Stichocentrus
livatus gen. et sp. nov. and Caproberyx pharsus sp. nov. (both Holocentridae), and one from the
Upper Senonian beds of Sahel Alma, Gnathoberyx stigmosus gen. et sp. nov. (Trachichthyidae).
Lissoberyx dayi is the most primitive berycoid yet known, and lies near the origin of the sub-
order. Acvogastey anceps Arambourg is probably a second species of Lissoberyx. Gnathoberyx
is the only acanthopterygian yet discovered in which a toothed maxilla is the dominant bone in
the upper jaw. A new subfamily Caproberycinae 1s made for the Cretaceous Holocentridae,
which do not appear to be ancestral to the Tertiary and living subfamilies. These and other
points are considered in a discussion of the origin and early evolution of the Berycoidei.
Preliminary observations on the fish beds at Hakel suggest that they were deposited in a sub-
marine canyon or gully, the abundance of fishes being due to mass mortalities caused by
“waterbloom’’. It is suggested that the fish beds at Hakel and Hajula are of Middle Ceno-
manian age rather than Upper Cenomanian.

I. INTRODUCTION

In the spring of 1964 I spent a month in the Lebanon, collecting Upper Cretaceous
fishes and examining the very large collection of these fishes in the Geology Depart-
ment of the American University, Beirut. This paper is the first of a series dealing
with new material resulting from this trip. It contains preliminary discussions of the
palaeoecology and age of the Cenomanian fish beds of the Lebanon, and descriptions
of four berycoid fishes, three of them representing new genera.

It is a pleasure to acknowledge here all the help I received in the Lebanon, in
particular from Prof. Theodore Raven, Geology Department, American University,

GEOL. 14, 3. II

70 CRETACEOUS FISHES FROM THE LEBANON

Beirut, who gave me every facility and allowed me to borrow specimens from the
collection in his care. My thanks are also due to Mrs. Raven, to Mr. Bud Young of
the Geology Department and Dr. Karl George of the Zoology Department in the
American University, and especially to my friends Mr. & Mrs. I. N. H. Seymour of
Beirut. I am also most grateful to Dr. D. D. Bayliss for his comments on the
microfauna of the Lebanon fish beds, to Mr. C. I. Macadie for his help in preparing
the specimens, and to Mr. N. Tanti, who photographed the fishes.

The material described here is in the collections of the British Museum (Natural
History), identified by the prefix “ P.”’ before the registered number, and the Geology
Department of the American University, Beirut, referred to by ‘“ AUB ”’ before the
number.

Il. LOCALITIES AND CONDITIONS OF DEPOSITION

The fishes described here are from the three classic fish localities in the Lebanon,
Hakel, Hajula (both Cenomanian in age) and Sahel Alma (Upper Senonian). I was
unable to visit Sahel Alma and have nothing to add to published information on the
locality (Roger, 1946; Dubertret, 1963: 119; Patterson, 1964 : 365 and references
cited there). At Hakel and Hajula, two villages about 4 km. apart, at about 700 m.,
respectively 1r and 9g km. inland from Byblos (Jebail), the fish beds are very similar
in age and fauna (Roger, 1946; Dubertret, 1963:57; Patterson, 1964: 362 and
references cited there). Although much has been published on the fauna of Hakel
and Hajula, knowledge of the conditions under which the beds were deposited rests
on Roger’s monograph (1946) which dealt with the invertebrates of all three localities
and gave particular attention to the palaeoecology. Roger had not visited the
Lebanon and described the fauna of Hakel as being from two localities, Hakel and
Maifouk (a neighbouring village), when only one exposure of fish beds exists there
(Dubertret, 1963 :58). Roger concluded (p. 83) that the Cenomanian fish beds
were laid down as soft, fine mud on a deoxygenated bottom in channels between
shallows supporting reefs of rudists, well clear of coasts. He visualized the rocks as
being deposited in several hundred metres of water, basing this estimate on the
crustacean fauna and the similarity of the rocks to Globigerina ooze.

Further fieldwork is necessary before a detailed account of the palaeoecology of the
Cenomanian fish beds can be given, but some preliminary observations are worth
mentioning here. At Hajula the exposure is in the centre of the village and building,
cultivation and the inhabitants prevented me from forming an accurate estimate of
the extent and thickness of the fish beds, but at Hakel the beds are well exposed in a
narrow valley, in dip on the southern side and in strike on the northern, the beds
dipping to the north at about 30°. The exposure is about 250 m. long (east-west)
and about 200 m. broad (north-south). On the south side of the valley blasting was
carried out in the centre of the exposure about six years ago. The fish beds are at
least 20m. thick and their base has not been seen: above they pass into flaggy,
unfossiliferous limestones. The fish beds consist of thin-bedded, siliceous limestones
alternating irregularly with more massive limestones. Occasional nodules and lenses
of impure chert occur throughout the beds. The rock is normally pale buff in colour,
but bluish and grey beds occur occasionally, the former mainly in the massive

CRETACEOUS FISHES FROM THE LEBANON 71

limestones, the latter in the thin-bedded. Under the hammer the rock often gives
off a bituminous smell, as Roger (1946: 76) noted. From museum collections one
gains the impression that the bedding planes are flat and smooth, but in fact there is
a great deal of complex small scale folding in the thinner beds, probably due to
slumping in the unconsolidated sediment. This suggests that the fish beds were
deposited on a slope. Apart from these contorted beds, the bedding planes are flat
and I have seen no ripple marks nor any signs of current bedding or graded bedding.

At each end of the exposure the fish beds terminate abruptly against massive,
structureless and unfossiliferous limestones. These junctions are not faulted as
earlier authors have thought, for higher in the valley wall beds pass across the line of
junction without interruption. At the junction the fish beds are slightly contorted
and shattered, tending to bend upwards, but there is no sign of any breccia or
conglomerate. The line of junction is steeply inclined outwards (away from the
centre of the exposure) at both ends of the fish beds, but where the junction is
best exposed, in the stream bed at the eastern end of the exposure, the fish beds
appear to undercut the massive limestone. Two possible explanations of this
contact present themselves : the first that the massive limestones are a reef, evidently
entirely recrystallized since no fossils or structures are preserved ; the second that the
limestones are the walls of a submarine canyon or gully. The following points, each
of which is difficult to reconcile with deposition close to a reef (indicating shallow,
well oxygenated water), suggest that the second of these interpretations is preferable :

(i) the high organic content and bituminous smell of the fish beds, together with
the complete absence of sessile benthos, indicating a deoxygenated bottom.

(ii) the absence of any reef breccia at the contact between the massive limestones
and the fish beds.

(ui) the microfauna of the fish beds consists of abundant radiolarians, moderately
abundant pelagic Foraminifera (Hedbergella) and rare benthonic Foraminifera
(textulariids) : my colleague Dr. D. D. Bayliss, who kindly examined thin sections
of the fish beds and identified the Foraminifera, considers the rock to be a foramini-
feran/radiolarian ooze, suggesting near bathyal depths.

As a preliminary hypothesis, it is therefore suggested that the Cenomanian fish
beds of Hakel were laid down in a submarine canyon or gully which in some way
served as a trap for fishes and invertebrates which were preserved in large numbers
because of the deoxygenated bottom and a supply of fine sediment. Further, the
extreme abundance of well-preserved fishes and crustaceans through many metres of
rock, some of the bedding planes being completely covered by fishes, is clear evidence
of the occasional occurrence of mass mortalities. As a further hypothesis I would
propose that these mass mortalities were caused, like the majority of known examples
(Brongersma—Sanders, 1957), by toxins released in “‘ waterbloom ” conditions, and
that the high percentage of silica in the rocks (21% at Hakel, not detrital but
colloidal, Roger, 1946 : 77) results from the solution of diatom frustules.

Ill. THE AGE OF THE CENOMANIAN FISH BEDS

The age of the beds at Hakel and Hajula, though known to be Cenomanian, is not
yet precisely fixed stratigraphically but estimated from the fish faunas. d’Erasmo

72 CRETACEOUS FISHES FROM THE LEBANON

72

(1946 : 134) and Arambourg (1954 : 163) both concluded that an Upper Cenomanian
age is most probable, and in an earlier paper (Patterson, 1964 : 362) I accepted this
estimate. d’Erasmo’s conclusion was based on a comparison with the fauna of
Comen, near Trieste: he thought that both faunas were Upper Cenomanian in age.
Arambourg, describing the fauna of Jebel Tselfat, Morocco, found that it is closest
to that of Comen, and that since Comen and Jebel Tselfat share certain archaic
genera (Belonostomus, Thrissops, Clupavus) which are absent at Hakel and Hajula
they are probably Lower Cenomanian in age while the Lebanese localities are Upper
Cenomanian. Some new evidence suggests that a re-examination of these con-
clusions is necessary. The microfauna of Hakel (Hedbergella present, no Globo-
truncana or Rotalipora) suggests an age low in the Cenomanian according to Dr. D. D.
Bayliss (personal commn.), and Dubertret (1963 : 57) speaks of the fish beds at Hakel
and Hajula as lying low in the Cenomanian. The more advanced teleosts suggest
close relationship between the Lebanese fauna and that of Jebel Tselfat : Protobrama
(Hajula) is close to Tselfatia (Jebel Tselfat) (Patterson, 1967 : 230), Caproberyx is
present at both Hakel and Jebel Tselfat (p. 98) and Lissoberyx dayi (Hakel &
Hajula) resembles Acrogaster anceps (Jebel Tselfat) (p. 79). Further, one cannot
attach much importance to the presence at Comen and Jebel Tselfat of Belonostomus,
Thrissops and Clupavus, which Arambourg holds to be archaic forms absent at Hakel
and Hajula. Belonostomus occurs rarely at Hakel (P.4029, P.8676) and in any case
is of little value as an indicator of age since it ranges upwards to the Maestrichtian.
Although Thrissops is apparently absent at Hakel and Hajula, Eubiodectes (Hay,
1903 : 415) is very similar and possible synonymous (Bardack, 1965 : 35). As I hope
to show in a forthcoming paper, Clupavus or a related genus is abundant at Hakel
and Hajula, although not previously recognized: the species Clupea gaudryi Pictet
& Humbert (1866: 60, pl. 5, figs. 2-5) is not a Scombroclupea as Smith Woodward
(rg0r : 138), Kramberger (1895 : 37), d’Erasmo (1922:72; 1946: 70) and other
authors have supposed since the type material lacks ventral scutes and finlets behind
the anal fin. The specimens described as Scombroclupea gaudry1 by Smith Woodward,
d’Erasmo and Kramberger are true Scombroclupea but are not conspecific with the
type material, which is close to Clupavus.

But in spite of this evidence of relationship with the faunas of Comen and Jebel
Tselfat, Hakel and Hajula contain some advanced groups which are absent in the
Moroccan and Dalmatian localities. The most important of these are the primitive
eels (Uvenchelys, Anguillavus and possibly Enchelion) and the Ctenothrissiformes
(Ctenothrissa and Pateroperca). These two groups occur otherwise only in the
English Chalk (with the exception of a species of Anguillavus in the Kansas Chalk
(Martin, 1920 : 95, pl. 6)), whose lower zones are definitely of Upper Cenomanian age.
But the Lebanese fauna is almost certainly older than that of the Upper Cenomanian
zones of the English Chalk, for it lacks groups such as the specialized ichthyodectids
(Ichthyodectes, Xiphactinus) which are present there, and contains primitive forms
(Clupavidae, Diplomystus brevissimus) absent in the Chalk.

In summary, the fauna of Hakel and Hajula is closely related to that of Comen
and Jebel Tselfat, Lower Cenomanian in age, but is probably younger than these
(eels and ctenothrissoids present). It is also closely related to the fauna of the

CRETACEOUS FISHES FROM THE LEBANON 73

Upper Cenomanian zones of the English Chalk, but is probably older than this
(clupavids and Dziplomystus brevissimus present, no advanced ichthyodectids).
Pending more precise stratigraphic work in the Lebanon, a Middle Cenomanian age
seems most likely for Hakel and Hajula.

IV. SYSTEMATIC DESCRIPTIONS
Order BERYCIFORMES
DiAGNosis. See Tate Regan (1911: 2).

Suborder BERYCOIDEI
DiAGnosis. See Patterson (1964 : 433).

Family TRACHICHTHYIDAE Bleeker 1895
Diacnosis. See Patterson (1964 : 305).

Genus LISSOBERYX nov.

D1AGnosis. Small Cretaceous Trachichthyidae with the bones of the head without
ornament except for weak serrations on the edges of the infraorbitals, preopercular,
interopercular and subopercular ; skull roof broad and flat, without ornament, crests
or mucus cavities ; supraoccipital crest short and high, supratemporal fossa ending
above posterior edge of orbit ; post-temporal fossa partially roofed; toothless
maxilla expanded posteriorly, two supramaxillae, ectopterygoid toothed ; opercular
covered by scales; 23 vertebrae including one ural centrum; dorsal fin with 5
spines, anal with 4, less than ro soft rays in each; scales thin and ctenoid, none
enlarged, no ventral ridge scales.

TYPE SPECIES. Acrogastey dayi Smith Woodward, 1942.

Lissoberyx dayi (Smith Woodward)
(Plat, fig. x, Pl. 4, fig. 1; Text-figs. 1-3)

1942 Acrogastey dayi Smith Woodward: 540, pl. 4, fig. 2.
1964 Acrogastey dayi Smith Woodward ; Patterson: 410.

Diacnosis. As for genus, only species: reaching about 4 cm. in standard length ;
PEG wWALV, 8 » Prehr05 Vil, 6.

Hototyre. AUB 108930, Day colln., American University, Beirut.

MATERIAL. In addition to the holotype, four specimens, AUB ro1997, 107578,
108926 and 109129, Day colln., American University, Beirut.

HORIZON AND LocatitiEs. Middle Cenomanian ; Hajula (4 specimens) and Hakel
(I specimen), Lebanon.

DESCRIPTION. Smith Woodward’s original description (1942: 540) mentions
nothing except the proportions of the trunk and the composition of the dorsal and
anal fins. The description which follows is based mainly on AUB 107578 (PI. 4,

74 CRETACEOUS FISHES FROM THE LEBANON

fig. 1) and 108926 (PI. I, fig. 1) which have been prepared with acid after embedding
in resin. Text-figures I-3 are composite restorations, based on four specimens.

Measurements and proportions. The dimensions of the five specimens are
shown in Table r.

TABLE 1.—Dimensions (in mm.) of specimens of
Lissoberyx dayt (Smith Woodward)

Total Standard Maximum Length Predorsal Preanal

Specimen length length depth of head length length
AUB 108930* 48 41 22 16 24 30
101997 32 26 14 12 13 23
107578 -- — 13 10 13 20
108926 40 34 16 15 17 28
», Iogt29 44 39 20 15 21 30

mean % standard
length 118%, 100% 49% 42% 53% 80%

* Corrected measurements from Smith Woodward’s figure (1942, pl. 4, fig. 2) of the holotype, which is
not natural size as stated but x c. 1:2.

Lissoberyx dayt was a small, deep-bodied fish (Text-fig. 3) reaching about 40 mm.
in standard length, 50 mm. in total length. The maximum depth of the trunk is
almost exactly half the standard length, the length of the head about 42% of the
standard length.

Neurocranium. No details of the basicranium are visible in any specimen but the
skull roof (Text-fig. 1) is moderately well exposed in the two acid prepared specimens,
although few of the sutures can be seen. The skull roof is short and very broad.
It is unlike that of any living berycoid in being quite smooth, without the strong
ornament characteristic of the holocentrids or the crests and mucus cavities of the
trachichthyids and their relatives. There is a high, triangular supraoccipital crest
(soc.), thickened anteriorly, which rises from a short, broad supratemporal fossa
(st. f.). The supratemporal fossa is limited anteriorly and laterally by alow, smooth
crest, formed by the frontals anteriorly and the parietals (pa.) laterally. The dorsal
limb of the extrascapular (ext.) articulated with the hind end of the parietal crest
and a short groove on the medial face of the crest carried the terminal part of the
supratemporal commissural sensory canal, as in many fossil and living berycoids.
The limits of the parietals, epiotics and supraoccipital within the supratemporal fossa
cannot be distinguished. Lateral to the supratemporal fossa there is a high, narrow
post-temporal fossa (pt. f.). The post-temporal fossa is partially roofed by the parietal
and pterotic (pto.), which make contact in the lateral wall of the fossa (108926), as in
other Cretaceous berycoids (Hoplopteryx, Tvrachichthyoides, Caproberyx pharsus :
Patterson 1964 : 360, text-figs. 47,55: p.98). This contact between the parietal
and pterotic is a relic of the complete roof of the post-temporal fossa in more primitive
teleosts. Lateral to the post-temporal fossa the pterotic projects in a broad, smooth
wing, continued anteriorly by the sphenotic (spfo.). On the lower margin of the

CRETACEOUS FISHES FROM THE LEBANON 75

Fic. 1. Lissoberyx dayi (Smith Woodward). Restoration of the skull roof in dorsal view,
x8 approx. /fy., frontal; mes., mesethmoid; na., nasal; pa., parietal; pt.f., post-
temporal fossa ; pio., pterotic ; soc., supraoccipital ; spo., sphenotic ; st.f., supratemporal
fossa. The broken line on the left side of the figure indicates the course of the sensory
canals.

exposed area of the pterotic there is a flange projecting dorsally : this flange carried
the main cephalic sensory canal on its dorsal surface, the infraorbital and preopercular
canals passing ventrally before and behind it respectively. Below the flange there
is a moderately deep dilatator fossa. An exactly similar flange is present in the
living holocentrid Myripristis.

The major part of the skull roof is formed by the frontals (/7.), the two bones
together covering an area about as broad as it is long. Each frontal has a broad,
smooth supraorbital flange passing back to its junction with the sphenotic. The
supraorbital sensory canal, passing on to the frontal from the sphenotic, entered the
bone at the level of the front of the supratemporal fossa through an elongated
opening covered by a projecting flange of bone. A small pore just in front of this
opening transmitted a short branch of the canal on to the supraorbital flange, and a
postero-medial branch opened through a larger pore, again covered by a projecting
flange, at the foot of the crest bounding the supratemporal fossa. The canal passed
to the anterior end of the frontal in a wide tube which opens through two elongated

76 CRETACEOUS FISHES FROM THE LEBANON

pores above the anterior third of the orbit, one leading laterally on to the supra-
orbital flange, the other opening into a long, shallow median depression between the
frontals. This simple, largely enclosed canal system resembles that in holocentrids
like the living Holocentrus and Myripristis and the Cretaceous Stichocentrus and
Caproberyx (Text-figs. 7, 10) ; it is quite unlike the system of open mucus cavities
found in the trachichthyids and their allies. From a wide opening in the anterior
edge of the frontal the supraorbital canal passed forwards to the nasal (na.). The
nasals are moderately large, smooth bones, tapering forwards, with the lateral and
medial edges rolled upwards to form a partially enclosed channel for the sensory
canal. The nasals were loosely attached to the frontals, not sutured on as they are in
most trachichthyids.

Enclosed between the anterior ends of the frontals is the “ V ”’ shaped upper edge
of the mesethmoid (mes.). The lateral ethmoids (/.e.) are deep and very broad.
The parasphenoid is deep, broad and without teeth. The vomer is not visible in any
specimen, nor can an orbitosphenoid be seen.

pmx.

ang.

pg

Itc. 2. Lissoberyx dayi (Smith Woodward). Restoration of the skull and pectoral and
pelvic girdles in left lateral view, x5 approx. ang., angular; ayrt., articular; cl.,
cleithrum ; cory., coracoid; den., dentary; d.pc., dorsal postcleithrum; evxt., extra-
scapular; /fy., frontal; iop., interopercular; Ja., lachrymal; J.e., lateral ethmoid ;
mx., maxilla; na., nasal; op., opercular; pa., parietal; p.g., pelvic girdle; pmvy.,
premaxilla; pop., preopercular; p.sm., posterior supramaxilla; pto., pterotic; qu.,
quadrate; scl., supracleithrum; soc., supraoccipital; sop., subopercular; spo.,
sphenotic ; ssc., suprascapular ; v.pc., ventral postcleithrum,

CRETACEOUS FISHES FROM THE LEBANON 77

Circumorbital bones. The infraorbital series contains the normal complement of
four infraorbitals and a lachrymal. All the infraorbitals are deep (Text-fig. 2), with
a deep flange overhanging the groove for the sensory canal. There are a few weak
serrations on the edge of this flange on the first and second infraorbitals and two
shallow grooves marking branches of the canal on the second infraorbital : the bones
are otherwise smooth. There is a broad subocular shelf on the second infraorbital
only. The lachrymal (/a.) is no deeper than the infraorbitals ; dorsally it articulates
with the lateral ethmoid by a strong facet. The groove for the sensory canal on the
lachrymal is bridged by a bar of bone where it turns upwards anteriorly, and there
are a few weak serrations on the ventral edge of the bone. There is no trace of an
antorbital but the apparent absence of this small bone could be due to imperfect
preservation.

Palate and jaws. The hyomandibular lies almost vertically and is slender, with an
undivided head. The symplectic inclines forwards slightly and the condyle of the
quadrate (qu.) lies below the hind edge of the orbit. The endopterygoid is not
toothed. The ectopterygoid is toothed, and although the palatine is not visible in
any specimen it will almost certainly have borne teeth, since I know of no acantho-
pterygian in which a toothed ectopterygoid occurs with a toothless palatine.

The ascending process of the premaxilla (pmvx.) is about half as long as the toothed
alveolar process and is well marked off from the articular process. The maxilla (mx.)
is about half as long again as the premaxilla and strongly expanded posteriorly, but
there are no teeth on the posterior expansion as there are in Myripristis and the
Cretaceous Hoplopteryx macracanthus (Patterson, 1964 : 334) and Gnathoberyx (p. 83).
There are two supramaxillae, the posterior (pf. sm.) with a slender process overlying
the anterior. The mandible is long and deep, both the dentary (den.) and the
articular (avt.) forming a high coronoid process, that of the dentary toothed through-
out its length. At the symphysis the band of teeth on the dentary becomes broader
but does not extend on to the lateral face of the bone as it does in some berycoids.
At the back of the mandible there is a very small angular (ang.). The mandibular
sensory canal ran in a closed tube opening by three pores in the dentary and a pore
at the suture between dentary and articular. All the bones of the jaws are without
ornament.

Opercular bones. The preopercular (pop.) is long, broad, inclined backwards a
little and weakly angulate. There are weak serrations on the posterior edge of the
lower half of the vertical limb and a few serrations above the angle in the bone on the
edge of the flange overhanging the groove for the sensory canal. The opercular (of.)
is smooth, but has two weak spines on its posterior edge. The opercular is covered
by scales: two very large cycloid scales cover the anterior part of the bone with
three or four smaller scales on the posterior part (101997, 107578). The subopercular
(sop.) has the normal spike overlapping the ventral corner of the opercular, and three
ridges, each ending in a serration, on its ventral edge. The interopercular (cop.)
bears a number of weak ridges ending in feeble serrations along its ventral edge.

Hyoid arch and branchiostegals. The ceratohyal is ossified in the usual two pieces ;
the distal ossification is very deep and is perforated by a large oval fenestra. There
are seven branchiostegals, three articulating with the medial face and two with the

78 CRETACEOUS FISHES FROM THE LEBANON

lateral face of the distal ceratohyal, and two articulating the lateral face of the
proximal ceratohyal.

Vertebral column. ‘There are 23 vertebrae, 9 abdominal and 14 caudal, including
one free ural centrum (the second). Ribs are present on all the abdominal vertebrae
except the first, and are borne on transverse processes on the last four (five in 101997).
There are no epineurals or epipleurals. The caudal skeleton will not be described in
detail here since I am dealing with the caudal skeletons of all Cretaceous Beryciformes
in a forthcoming paper : suffice it to say that the caudal of Lissoberyx agrees with
that of the living trachichthyid Hoplostethus (Gosline, 1961 : 14) in having a neural
crest on the second pre-ural centrum and a free second ural centrum.

Pectoral and pelvic girdles and fins. The pectoral girdle contains the normal
complement of bones, a curved extrascapular (ext.) articulating with the parietal

4

SES

Fic. 3. Lissoberyx dayi (Smith Woodward). Restoration of the skeleton,
scales omitted, 2-5 approx.

above and the pterotic below, a forked suprascapular (ssc.) articulating with the
epiotic and intercalar, a blade-shaped supracleithrum (scl.), a large, sigmoid cleithrum
(cl.), two post-cleithra, the upper (d. pc.) scale-like, the lower (v. pc.) rod-like, and the
scapula and coracoid. All the dermal bones of the girdle are smooth and un-
ornamented. The anterior process of the coracoid (cor.) is moderately stout and
long, arching forwards to join the cleithrum just above its tip. The pectoral fin
contains about ten rays.

The pelvic fins are thoracic, the fin originating well in front of the tip of the ventral
post-cleithrum and the girdle (f. g.) ending between the anterior processes of the
coracoids. The pelvic fin contains a stout, smooth spine and six soft rays.

Median fins. The dorsal fin originates above the pelvic and occupies little more
than half the back of the fish ; it contains five smooth spines, increasing in length
from front to rear, and g soft rays. The longest spine is equal in length to the

CRETACEOUS FISHES FROM THE LEBANON 79

distance between the base of the fin and the vertebral column. The first radial of
the fin is not enlarged, and lies between the neural spines of the second and third
vertebrae. There are two predorsals (Smith & Bailey, 1961) in front of the fin.
The anal fin arises below the hind end of the dorsal ; it contains four spines, increasing
in length, and 8 soft rays. The fourth anal spine is equal in length to the fifth dorsal
spine. The first anal radial is very small and the second is only a little enlarged.

The forked caudal fin contains 19 principal rays with 17 branched, 9 in the upper
lobe and 8 in the lower. Preceding the upper lobe there are four small spines and
one segmented ray, preceding the lower three spines and a segmented ray.

Squamation. The scales are large, thin and coarsely ctenoid. There are 13 scales
in a transverse series on the trunk, the lateral line passing through the fifth from the
top. The lateral line scales are not enlarged. Scales cover the cheek and the
opercular, but the scales do not extend over the bases of the fins. There are no
ventral ridge scales.

AFFINITIES. Smith Woodward (1942 : 540), in his original brief description of
Lissoberyx dayt, placed the species in the trachichthyid genus Acrogaster. Clearly
the species cannot be included in this genus: in Acrogaster the skull roof bears a
normal trachichthyid system of large mucus cavities separated by high, serrated
crests, the alveolar process of the premaxilla is almost as long as the maxilla and the
maxilla is not expanded behind it, the preopercular bears a number of large spines
at the angle, there are no scales on the operculum, there are eight branchiostegals and
the scales are cycloid (Patterson, 1964: 410). In all these characters A. dayi differs
from the Senonian species of the genus, and since it shows no signs of close relation-
ship with any other Cretaceous genus it is necessary to make a new genus to receive it.
The name Lissoberyx refers to the smooth skull roof and general absence of ornament
on the head. Acyrogaster anceps Arambourg (1954 : 153), known by a single specimen
from the Lower Cenomanian of Morocco, is probably another species of Lissoberyx :
it agrees with L. day: and differs from the Senonian species of Acrogastey in having
the maxilla expanded posteriorly, in the small number of vertebrae (21) and in the
form of the dorsal and anal fins. Through the kindness of Prof. J. P. Lehman and
Mlle. J. Signeux I was able to examine this specimen and could find no trenchant
differences from L. dayz. Although Arambourg found traces of two or three spines
on the angle of the preopercular he spoke of these as “ assez fruste to me it
appears that these traces are not spines but ridges on the surface of the bone marking
branches of the sensory canal. Arambourg described the pelvic as containing 7 or 8
soft rays, a difference from L. day: and all trachichthyids, but in my opinion the
pelvic fins are not sufficiently well preserved for an accurate count to be made.

That Lissoberyx dayi is a member of the Beryciformes is shown by the presence of
fin spines, pelvics with more than five soft rays, two supramaxillae and a nineteen-
rayed tail. Within the Beryciformes Lissoberyx falls in the sub-order Berycoidei
(Patterson 1964: 433) because of the toothless endopterygoid, the absence of
epineurals, the pelvic spine and the nineteen-rayed caudal. All known Cretaceous
Berycoidei are members either of the Trachichthyidae or the Holocentridae, and
there is normally no difficulty in deciding to which of these two families a genus

”

80 CRETACEOUS FISHES FROM THE LEBANON

belongs. However, as the Trachichthyidae and Holocentridae are traced back to the
Cenomanian they begin to converge, holocentrid characters such as maxillary teeth
and an antorbital appearing in the trachichthyid Hoplopteryx and trachichthyid
characters such as a small number of dorsal fin spines, a large supraoccipital crest
and a supratemporal fossa appearing in the holocentrid Caproberyx (Patterson,
1964 : 341, 359; p. 102). In Lissoberyx this trend goes further : the fish is a mosaic
of trachichthyid and holocentrid features.

First, there are several characters of Lissoberyx which are common to both
trachichthyids and holocentrids in the Cretaceous, though not always to living
members of these families. These include the partial roof over the post-temporal
fossa (as in Hoplopteryx, Trachichthyoides and Caproberyx pharsus, p. 98), the
presence of ectopterygoid teeth, two supramaxillae (only one in living trachichthyids),
the mandibular sensory canal running in a canal opening by three or four pores (as in
Cretaceous but not living holocentrids), the preopercular without a spine at the angle,
the deep, perforate ceratohyal (as in Cretaceous but not in living holocentrids), the
presence of predorsals in front of the dorsal fin and the insertion of the first dorsal
radial between the second and third neural spines (as in Hoplostethus and living
holocentrids), the four anal fin spines and the unmodified anal radials (the third anal
spine is enlarged in most holocentrids, but not in Caproberyx) and the unspecialized
ctenoid scales.

Secondly, there is a group of characters in which Lissoberyx resembles the trachich-
thyids and differs from the holocentrids. These include the high supraoccipital crest
and moderately large supratemporal fossa (a similar supraoccipital crest occurs in the
holocentrids Caproberyx and Stichocentrus, Text-figs. 8, 10, but there the supra-
temporal fossa is much smaller), the deep infraorbitals with the lachrymal no deeper,
the subocular shelf on the whole of the second infraorbital (as in Hoplopteryx but not
in living tracbichthyids), the 23 vertebrae (the vertebral number never seems to fall
below 26 in holocentrids), the six soft rays in the pelvic fin (7 in holocentrids except
in Caproberyx pharsus, p. 102), and the form of the dorsal fin, with five spines and
nine soft rays (no holocentrid is known to have less than six spines and eleven
soft rays).

Thirdly, characters in which Lissoberyx resembles the holocentrids and ditfers from
the trachichthyids include the absence of large mucus cavities on the skull roof and
the pattern of the supraorbital sensory canal (as in Myripristis), the form of the
pterotic (again as in Myripristis), the nasals not sutured to the frontals or to each
other, the unreduced mesethmoid, the short alveolar process of the premaxilla and
the strongly expanded maxilla, the presence of scales on the operculum (though in
Cretaceous holocentrids only the antero-dorsal corner of the opercular is scaly), and
the rather long and stout anterior process of the coracoid.

Finally, there are a few characters in which Lissoberyx differs from both the
trachichthyids and the holocentrids. These are the smoothness of the skull roof
and the almost complete absence of ornament on the bones of the head, the seven
branchiostegal rays (eight in both trachichthyids and holocentrids) and the complete
scaly covering of the opercular.

Evaluation of this complex of characters is difficult : the characters in each group

CRETACEOUS FISHES FROM THE LEBANON 81

cannot simply be counted because they obviously differ in significance. On balance,
I feel sure that Lissoberyx lies nearer to the trachichthyids than to the holocentrids,
particularly significant associations with the trachichthyids being the supratemporal
fossa, the form of the infraorbitals, the number of vertebrae and the dorsal and pelvic
fins. But the most important features in which Lissoberyx resembles the holo-
centrids, the characters of the skull roof and the upper jaw, are almost certainly
primitive, since the deep mucus cavities of the trachichthyid skull roof can hardly be
anything but specialized and since a short premaxilla and an expanded maxilla are
more likely to be primitive than a premaxilla extending to the tip of the maxilla
(Patterson, 1964: 439). Of the characters in which Lissoberyx differs from both
holocentrids and trachichthyids, the smooth skull roof and absence of ornament on
the head must be primitive while the seven branchiostegals are advanced. The
scaly opercular is probably primitive for the berycoids (see below p. 107). It may
be significant that in the seven branchiostegals and the scaly opercular (as also in the
lack of ornament and the unreduced mesethmoid) Lissoberyx resembles the Berycidae
(particularly Beryx), a group probably derived from the trachichthyids.

The presence of ribs on all but the first abdominal centrum in Lissoberyx is a
peculiar feature which is otherwise known only in Stichocentrus (p. 93) among
Beryciformes. In acanthopterygians the first two vertebrae normally bear only
slender bones which appear to be in series with the epipleurals rather than the ribs
(Starks, 1904: 616; Gosline, 1963: 28). It may be that in Lissoberyx an anterior
centrum has become incorporated in the neurocranium, but the fact that the first
dorsal radial lies between the second and third neural spines in Lissoberyx, Hoplo-
stethus, Stichocentrus and living holocentrids suggests that the first free vertebra is
homologous in all these forms. It is also possible that the slender bones borne on the
first two vertebrae of some acanthopterygians are not epipleurals but reduced ribs :
the skeletons of living Beryciformes that I have examined suggest that this may be
true of trachichthyids and some holocentrids. In any event, Rosen (1964 : 242)
shows that differences in the point of origin of the first rib cannot be interpreted in
the simplest terms in teleosts.

In summary, Lissoberyx is a trachichthyid, but it shows more resemblance to the
holocentrids than any other trachichthyid and must lie very close to the common
stock of these two families and of the sub-order Berycoidei. As the most generalized
member of this sub-order, Lissoberyx throws some light on the origin of the group and
is discussed further on p. 106 below.

Genus GNATHOBERYX nov.

Diacnosis. Small Cretaceous Trachichthyidae in which the maxilla is toothed
and forms more than half the gape ; superficial bones of the head and scales spiny ;
long, slender teeth in the upper jaw, supramaxillae reduced or absent ; 27 vertebrae,
including a free second ural centrum ; dorsal and anal fin each with four spines and
less than ten soft rays; scales ctenoid and spiny, lateral line scales enlarged and
thickened, ventral ridge scales present.

TYPE SPECIES. Gnathoberyx stigmosus sp. nov.

82 CRETACEOUS FISHES FROM THE LEBANON

Gnathoberyx stigmosus sp. nov.
(Pl. 1, fig. 2, Pl. 2; Text-figs. 4, 5)

Diacnosis. As for genus, only species : reaching about 4 cm. in standard length,
depth of trunk about 45% of standard length, length of head about 35°% of standard
lengthy sD) LVR oiseAn VGN Se

HototyPe. AUB No. 100402 (PI. 1, fig. 2, Pl. 2, fig. 2).

MATERIAL. In addition to the holotype, AUB No. 103838 (PI. 2, fig. I).

HORIZON AND LOCALITY. Upper Senonian ; Sahel Alma, Lebanon.

DescripTION. Acid preparation has not been used on this species because there
are only two specimens and material from Sahel Alma often gives poor results in
transfer preparations. For this reason much of the cranial anatomy remains
unknown at present.

Measurements and proportions. The dimensions of the two specimens are as
follows (in mm.), the first figure in each case being for the holotype, the second for
103838 : total length: 50-52, standard length : 40-c. 42, maximum depth of trunk :
c. 20-18, length of head: 16-15, predorsal length : 21-23, preanal length: c. 28-27.

Although both specimens are distorted, 103838 by oblique crushing, the holotype
by displacement of the anal fin and deepening of the trunk, they are clearly closely
comparable in size. G. stigmosus was a small, deep-bodied fish, about 40 mm. in
standard length, with the maximum depth of the trunk about 46% of the standard
length and the length of the head about 35°% of the standard length.

Neurocranium. The skull roof is not well exposed in either specimen. The
supraoccipital crest was low, not projecting above the skull roof, and apparently

Tic. 4. Gnathoberyx stigmosus gen. et sp. nov. Jestoration of the skeleton,
scales omitted, x 2:5 approx.

CRETACEOUS FISHES FROM THE LEBANON 83

rather short, arising at the level of the hyomandibular facet. The skull roof bore a
pattern of high crests bounding mucus cavities, as is normal in trachichthyids, but
the details of the pattern cannot be made out. The lateral margin of the frontal
above the orbit bears a row of large, pointed spines and there is a more medial crest
bearing similar spines which is continued back by the parietal. The postero-lateral
part of the skull roof is smooth and without ornament. The nasals are large, thick,
scroll-like bones, probably sutured to the frontals, with coarse spines on their
margins. The nasals extend antero-ventrally almost to the vomer, indicating that in
this species, as in Hoplopteryx and living trachichthyids, the mesethmoid was much
reduced. These facts suggest that the skull roof of Gnathoberyx was of the same
type as in Hoplopteryx and Hoplostethus (Patterson, 1964, text-figs. 46, 54, 65), and
that the supratemporal fossa was short, as in Trachichthyidae, not extended forwards
as it is in Berycidae.

An orbitosphenoid is not visible in either specimen. The parasphenoid is straight
and slender, and there is a long basisphenoid pedicel articulating with it at the hind
edge of the orbit. No details of the otic region are visible.

Infraorbital bones. The infraorbitals are very deep, as deep as or deeper than the
lachrymal. Dorsally they bear a thickened, serrated flange projecting over the
groove for the sensory canal and the ventral edges are also coarsely serrated. The
subocular shelf is confined to the second infraorbital. All these features are typical
of trachichthyids. There is no trace of an antorbital.

Palate and jaws. The hyomandibular is inclined posteriorly a little but the
quadrate and the elongated symplectic are inclined forwards so that the articular
condyle of the quadrate lies only just behind the centre of the orbit, and the gape is
shorter than in most trachichthyids. The endopterygoid is toothless, as usual in
Berycoidei. There is a long patch of teeth on the border of the anterior part of the
palate, but whether these teeth extend on to the ectopterygoid or are all on the
palatine cannot be seen. There is a normal maxillary process on the tip of the
palatine.

The upper jaw (Text-fig. 5, Pl. 2, fig. 2) is remarkable in having a long, toothed
maxilla which makes up more than half of the gape. The premaxilla is not well
preserved in either specimen and the details of the head of the bone cannot be made
out, but the ascending process was clearly very small, no higher than and probably
hardly distinct from the articular process. The alveolar process, extending back
below the maxilla, bears minute clustered teeth on its ventro-medial surface and a
single irregular row of six or seven long, slender teeth along its outer margin. On
the dorsal surface of the alveolar process there is a low, rounded postmaxillary process
lying medial to the maxilla. The total length of the premaxilla is about 40% of the
length of the maxilla. As in the premaxilla, the articular head of the maxilla is not
well preserved. Above the alveolar process of the premaxilla the maxilla extends
back as a stout rod which deepens abruptly at the hind end of this process ; along
the margin of this deep posterior part of the maxilla there is a single irregular series
of about a dozen long, slender teeth, the anterior ones curved forwards. The
anterior end of the toothed border of the maxilla projects forwards medial to the
tip of the premaxilla in exactly the same way as in the ctenothrissiform Avwlolepis

GEOL. 14, 3. 12

84 CRETACEOUS FISHES FROM THE LEBANON

Fic. 5. Gnathoberyx stigmosus gen. et sp. nov. Left upper jaw. Below, in lateral view ;
above, the anterior part of the jaw in medial view. X12 approx.

(Text-fig. 6B). On the outer face of the toothed part of the maxilla there is a groove,
overhung by a flange anteriorly, which closely resembles the groove in Ctenothrissa
which I (Patterson, 1964: 232; Text-fig. 6A) interpreted as housing a ligament
attaching the upper jaw to the mandible. The hind end of the maxilla is not
completely preserved in either specimen: the bone evidently became much thinner
and was without teeth. The dorsal edge of the toothed part of the maxilla is thin and
smooth so far as it is preserved, and in neither specimen is there any trace of supra-
maxillae : if these were present they must have been small and flimsy, not like the
large supramaxillae of Ctenothrissiformes and most Berycoidei. In contrast to
most of the superficial bones of the head, the upper jaw is smooth and without
ornament.

The head of the quadrate lies almost below the centre of the orbit, and the mandible
is rather short. The ventral edge of the dentary and the flange overhanging the
groove for the sensory canal on both the dentary and articular are produced into
large, closely packed, recurved spines. In size and shape these spines resemble the
large marginal teeth of the upper jaw, but differ from them in having no pulp cavity.
The oral border of the dentary bears small, clustered teeth, apparently with a single
large procumbent tooth at the symphysis. Whether the mandibular sensory canal
ran in a closed tube or in an open groove cannot be seen.

Opercular bones. The preopercular is typically trachichthyid, but because of the
forward position of the suspensorium the ventral limb is about half as long as the
dorsal and the angle in the bone is acute, almost 90°. The bone is broad throughout
its length and, as usual in Cretaceous trachichthyids, without a spine at the angle.

CRETACEOUS FISHES FROM THE LEBANON 85

At and below the angle in the bone there are three stout ridges ending in marginal
spines, with similar spines spaced out between them, and above the angle there are
several shorter radiating ridges.

The opercular is without scales (except perhaps at its antero-dorsal corner, which is
not visible) and has a series of stout, spiny ridges radiating from its point of suspen-
sion and ending in spines on the edge of the bone. The subopercular bears similar
spiny ridges radiating from its antero-dorsal corner, and on the elongated inter-
opercular there are rows of spines radiating from the centre of the bone.

Hyoid arch and branchiostegals. No details of the hyoid skeleton are visible in
either specimen. There are eight branchiostegal rays, the first three with coarse
spines along their ventral edges.

Vertebral column. There are eleven abdominal vertebrae and sixteen caudals,
including a free second ural centrum. The ribs are small and intermuscular-like on
the first three abdominal vertebrae and are borne on transverse processes on the last
three. There are no epineurals and no epipleurals are visible. The caudal skeleton
is preserved only in the holotype ; although it is compressed and distorted it appears
to agree with that of the living trachichthyid Hoflostethus (Gosline, 1961 : 14) in
having no neural spine on the second pre-ural centrum, a slender urodermal “‘ wedged
into ”’ the first pre-ural centrum and a separate second ural centrum, which appears
larger than that of Hoplostethus.

Pectoral and pelvic girdles and fins. There is a broad extrascapular of the usual
triradiate form and a forked suprascapular with a long dorsal limb and a few spines
near its postero-dorsal corner. The supracleithrum is not clearly visible in either
specimen. The posterior plate of the cleithrum bears a few spiny ridges and small
spines near its posterior edge. Both the endoskeletal pectoral girdle and the pectoral
fin are poorly preserved in the holotype, and in 103838 they are missing. The
anterior process of the coracoid appears to have been short and slender, meeting the
cleithrum well above its tip. The ventral postcleithrum reaches the ventral border
of the trunk just behind the origin of the pelvic fin and the pelvic girdle probably made
contact with the cleithrum. The pelvic girdle and fin are missing in 103838 and in
the holotype only the spine of the pelvic fin is visible; it is ridged and equal in
length to about one-third of the depth of the trunk.

Median fins. The dorsal fin contains four stout spines, strongly ridged and
increasing in length from front to rear, and nine soft rays. The fourth dorsal spine
is equal in length to about 55% of the maximum depth of the trunk. The first
dorsal radial lies between the third and fourth neural spines and is preceded by three
predorsals, one in front of each of the first three neural spines. Except for the first
two and last three dorsal radials (two to each vertebra) there is a one-to-one relation-
ship between the fin supports and the vertebrae.

The anal fin contains four stout, ridged spines and eight soft rays. The fourth
anal spine is equal in length to about half the depth of the trunk. The first anal
radial is enlarged but not hooked forwards distally. The last anal radial lies in front
of the fifth haemal spine.

The caudal fin is almost entirely missing in 103838. In the holotype only the
upper lobe of the fin is preserved. This contains three small spines, one short

86 CRETACEOUS FISHES FROM THE LEBANON

segmented ray and ten principal rays, the outermost unbranched. The presence
of nine branched rays in the upper lobe is normal in caudal fins with seventeen
branched rays.

Squamation. The body scales are thin but coarsely ctenoid and spiny, with their
exposed surfaces covered with small spines. The lateral line scales are enlarged and
thickened, each with a raised, ornamented bridge covering the sensory canal. There
is a series of enlarged, thickened and coarsely ornamented ridge scales along the
ventral border of the trunk in front of the anal fin, with a pair of enlarged, thickened
and ornamented axillary scales at the origin of the pelvic fins. The dorsal ridge
scales in front of the dorsal fin also appear to be slightly thickened. The scales do
not cover the median fins but form a sheath along their bases, as in Tvachichthys.
The scales cannot be counted exactly, but they were clearly larger below the enlarged
lateral line scales. There were about thirteen scales in each transverse series, with
about six below and above the lateral line. Scales cover the cheek.

AFFINITIES. Gnathoberyx is shown to bea member of the sub-order Berycoidei
by the presence of fin spines, a pelvic spine, eight branchiostegal rays, a caudal skeleton
with a free second ural centrum and a caudal fin with seventeen branched rays.
Within the Berycoidei the genus falls in the Trachichthyidae because of the form of
the skull roof, the deep infraorbitals with a subocular shelf only on the second, the
broad, cavernous preopercular, the short dorsal and anal fins, and the enlarged
lateral line scales and ventral ridge scales.

The species shows various resemblances to other trachichthyids, living and fossil,
such as the spiny head bones and scales (cf. Hoplopteryx spinulosus, also from Sahel
Alma), the enlarged lateral line scales (cf. Tubantia from the Campanian of West-
phalia and the living genera Gephyroberyx and Hoplostethus), and the short dorsal and
anal fins, with four spines and less than ten soft rays (cf. Tubantia and Lissoberyx
among Cretaceous forms). But the structure of the upper jaw, with a long, toothed
maxilla apparently without supramaxillae, clearly sets the species apart from all
other berycoids and makes a new genus necessary to receive it. The upper jaw of
Gnathoberyx, improbable as it is in a genuine acanthopterygian, is not entirely out of
place in the Berycoidei, for maxillary teeth are already known in the living Myrip-
vistis and the Cretaceous Hoplopteryx macracanthus (Patterson, 1964: 439; Text-fig.
6E, F). In these forms, however, the maxilla hardly enters the gape, forming less
than a quarter of the margin of the jaw. In Gnathoberyx the maxilla is the dominant
bone in the upper jaw, in typical clupeiform or protacanthopterygian (Greenwood,
Rosen, Weitzman & Myers, 1966) fashion. In particular there are striking resem-
blances to the upper jaw of the Ctenothrissiformes Awlolepis and Ctenothrissa
(Text-fig. 6A, B), and also to the upper jaw in the living Macristium (Text-fig. 6c)
which Marshall (1961) has suggested is a living ctenothrissoid, a hypothesis which I
earlier (Patterson, 1964: 243) felt to be unproven. The similarities between the
upper jaws of Gnathoberyx and Macristium extend to the form of the premaxillary
teeth and the apparent absence of supramaxillae. Without more material of
Gnathoberyx it is impossible to study the articular heads of the premaxilla and
maxilla to discover whether these resemblances are more than superficial, but in view

CRETACEOUS FISHES FROM THE LEBANON 87

Fic. 6. The left upper jaw in lateral view of Ctenothrissiformes (left) and Berycoidei
(right). A. Ctenothrissa vadians (Agassiz), Ctenothrissidae, Upper Cenomanian, S.E.
England, x15. B. Aulolepis typus Agassiz, Aulolepidae, Upper Cenomanian, S.E.
England, x2. C. Macristium chavesi Tate Regan, Macristiidae (? Ctenothrissiformes),
Extant, Atlantic, x27. After Marshall, 1961. D. Gnathoberyx stigmosus gen. et sp.
nov., Trachichthyidae, Upper Senonian, Sahel Alma, Lebanon, x8. E. Hoplopteryx
macracanthus Patterson, Trachichthyidae, Senonian, S.E. England, x1-5. F. Myri-
pristis murdjan Forskael, Holocentridae, Extant, Red Sea, x1°6.

of the evidence that the Ctenothrissiformes lie closer than any other group to the
ancestry of the Beryciformes (Patterson, 1964 : 463-466 ; Greenwood et al., 1966 :
369) they must be carefully considered. There can be little doubt that the maxillary
dentition of Guathoberyx is primitive, not secondary. In Hoplopteryx macracanthus
and Myripristis there is every reason to believe that this is true (Patterson, 1964 :
440), and the argument receives added force from the well-toothed maxilla of
Gnathoberyx. But it is possible that the upper jaw of Gnathoberyx is specialized, for
whatever purpose, by reduction in the ascending process of the premaxilla and
elongation of the toothed part of the maxilla, secondarily producing a ctenothrissiform
type of jaw from a more typical berycoid one. Some support for this hypothesis may
be found in the relatively late age of Gnathoberyx (U. Senonian compared with the
first trachichthyids in the L. Cenomanian) and in the generally specialized aspect of
the fish—abundant spiny ornament, enlarged lateral line and ventral ridge scales,
the short gape, reduced or lost supramaxillae and large marginal teeth in the upper
jaw. And in one respect, the presence of a well developed postmaxillary process on
the premaxilla, the upper jaw of Guathoberyx is clearly advanced over those of the
Ctenothrissiformes and resembles normal acanthopterygians, These facts suggest

88 CRETACEOUS FISHES FROM THE LEBANON

that the resemblance between the upper jaws of Gnathoberyx and the Ctenothrissi-
formes is not necessarily evidence of close relationship, but this resemblance certainly
adds to the evidence supporting the hypothesis that the Beryciformes evolved from
near the Ctenothrissiformes rather than from the myctophoids, in which the maxilla
is never toothed.

Family HOLOCENTRIDAE Richardson 1846
DiaGNosis. See Patterson (1964 : 341).

Subfamily CAPROBERYCINAE nov.
DIAGNOSIS. See p. 97.

Genus STICHOCENTRUS nov.

DiAGNnosis. Cretaceous Holocentridae with a moderately high supraoccipital crest
projecting above the skull roof, frontals covering the anterior part of the parietals,
small mucus cavities on the skull roof; nasals small and tubular; no antorbital,
infraorbitals moderately deep, lachrymal large and deep, not extending back below
the first infraorbital, no subocular shelf on first infraorbital ; superficial bones of the
head (except the maxilla) ornamented with ridges, spines and tubercles ; head of
hyomandibular broad and single, suspensorium inclined backwards a little, ecto-
pterygoid toothed, no maxillary teeth ; no spine on the preopercular ; 26 vertebrae,
no free ural centrum ; nine dorsal spines, the last four decreasing in length a little
but dorsal fin not divided, five anal spines, the fourth thicker than the fifth but a
little shorter ; scales large, rough and ctenoid, abdominal ridge scales present.

TYPE SPECIES. Stichocentrus livatus sp. nov.

Stichocentrus liratus sp. nov.
(Pl. 3; Text-figs. 7-9)

Diacnosis. As for genus, only species. Reaching about 7cm. in standard
SG DO GON AY SG) ee Lee tee SWE LI 78

Ho.otyPe. B.M. (N.H.) No. P.47835 (Pl. 3).

MATERIAL. In addition to the holotype, seven specimens in the Day Colln.,
American University, Beirut, AUB 105736, 105987, 106809, 108923-24, 108927,
108929.

Horizon AND Locality. Middle Cenomanian ; Hajula, Lebanon.

DESCRIPTION. The description and figures are based mainly on the holotype and
AUB 108924 which have been prepared with acid after embedding in resin.

Measurements and proportions. The dimensions of the more complete
specimens are given in Table 2,

CRETACEOUS FISHES FROM THE LEBANON 89

TABLE 2.—Dimensions (in mm.) of the five most complete specimens of
Stichocentrus livatus gen. et sp. nov.

Total Standard Maximum Length Predorsal Preanal

Specimen length length depth of head length length
P.47835 76 64 32 30 29 54
AUB 105736 56 44 _— 22 20 37

105987 50 42 19 19 19 —
106809 — C.74 _— C.35 — C.57
108929 C.41 C.34 C.17 C.15 — C.25
mean % standard
length 121% 100% 48% 465% 45% 78%

Stichocentrus liratus was a small, moderately deep-bodied fish of normal berycoid
form (Text-fig. 9). The largest specimen, AUB 106809, is very incomplete but must
have had a standard length of about 74 mm.: this would give a total length of about
gomm. The length of the head and the maximum depth of the trunk are approxi-
mately equal, about 47° of the standard length.

Neurocranium. The neurocranium is typically holocentrid in shape, broad and
deep posteriorly and tapering forwards. The skull roof (Text-fig. 7) is well exposed
in the holotype and 108924. There was a moderately large supraoccipital crest
(soc.) with a thickened anterior edge ; the crest is higher and longer than it is in
living holocentrids, but not so large as it is in Caproberyx (Patterson, 1964, text-fig.
67; Text-fig. 10). The supraoccipital crest arises from a short, broad, shallow
supratemporal fossa (sé. f.), limited anteriorly by the frontals and parietals. Again,
the fossa is larger than it is in living holocentrids but smaller than in Caproberyx.
The parietals (pa.) are separated by the supraoccipital and bear a raised area,
ornamented with ridges and tubercles; the parietal branch of the supraorbital
sensory canal opened through a pore in the frontal immediately in front of this area,
and passed into a depression on the parietal from which two short grooves, the
medial one bifurcated, lead on to the ornamented area. The parietals bear very
similar grooves and depressions in the living holocentrid Holotrachys. The supra-
temporal articulated with the hind edge of this ornamented area, transmitting the
terminal part of the supratemporal commissural sensory canal, which ran in a short
groove behind the ornamented area. It is possible that the parietal met the pterotic
in the wall of the post-temporal fossa (ft. f.), but the area where the two bones would
have made contact is covered by a posterior extension of the frontal. The ventral
limb of the supratemporal articulated with the pterotic (pto.), and the main cephalic
sensory canal passed forwards in a groove covered laterally by a smooth raised
flange on this bone. The preopercular sensory canal passed ventrally through a
notch at the hind end of this flange. The sphenotic (sfo.) has the dermal and
cartilage components completely fused. The infraorbital sensory canal passed
ventrally between two raised, ornamented flanges on the sphenotic.

The frontals (fr.) are very large, extending posteriorly to cover much of the
parietals and pterotics, a characteristic holocentrid feature. In Stichocentrus this
posterior extension of the frontals is not so marked as it is in living holocentrids but

90 CRETACEOUS FISHES FROM THE LEBANON

Pto.

Fic. 7. Stichocentrus livatus gen. et sp. nov. Restoration of the skull roof in dorsal view,
x5 approx. epo., epiotic; fy., frontal; mes., mesethmoid; na., nasal; pa., parietal ;
pt.f., post-temporal fossa; pto., pterotic; soc., supraoccipital; spo., sphenotic; stf.,
supratemporal fossa. The broken line on the right side of the key diagram shows the
course of the sensory canals.

greater than it is in Caproberyx. Most of the surface of the frontal, particularly the
posterior part and the supraorbital flange, is strongly ornamented with small
tubercles and sinuous ridges: this ornament resembles that in Caproberyx superbus
rather than the large straight ridges on the frontals of living holocentrids. The
supraorbital sensory canal passed on to the frontal from the sphenotic, gave off a
short branch into a depression on the posterior part of the supraorbital flange, and
then entered the bone through an elongated pore. Within the bone the canal gave
off two posterior branches which passed back in tubes to the hind end of the frontal,
the more medial of these branches leading into the depression and grooves on the
parietal. The sensory canal gave off a medial branch opening into a narrow median
channel between the frontals, similar in size and position to the median mucus cavity
in Caproberyx superbus, but partially roofed by raised flanges of bone. Passing
forwards within the frontal, the sensory canal gave off a lateral branch through a
much elongated pore, covered above by a flange of bone, leading to a groove above
the anterior part of the orbit, From a pore in the anterior end of the frontal the

CRETACEOUS FISHES FROM THE LEBANON QI

canal passed into the nasal. The nasals (na.), which were attached to the frontals by
connective tissue, not by suture, are small, tubular bones, ornamented with ridges
and tubercles. Between the nasals the large “‘ V ’’-shaped dorsal part of the meseth-
moid (mes.) is exposed. There is no trace of an antorbital in any specimen: the
bone was probably absent, as it is in Caproberyx. In the roof of the orbit of the
holotype and 105736 an orbitosphenoid (ors., Text-fig. 8) is visible, similar in size and
shape to that of Caproberyx superbus but apparently without the lateral fenestra
present in the latter ; in living holocentrids the orbitosphenoid is much reduced in
comparison with these Cretaceous forms. As is usual in these small, crushed fishes
from the Lebanon, no details of the basicranium are visible in any specimen. The
parasphenoid is straight, with lateral flanges articulating with the endopterygoids.
The vomer is not visible in any specimen. The lateral ethmoids (/. e.) are deep and
well ossified.

Infraorbital serves. The infraorbital series consists of the usual five bones. The
two posterior infraorbitals are small and slender with coarsely serrated posterior
edges and a smooth flange covering the groove for the sensory canal. The first and
second infraorbitals are longer and a little deeper, with the edge of the flange over the
sensory canal serrated and coarse ridges on the ventral part of the bone. There is a

Fic. 8. Stichocentrus livatus gen. et sp. nov. Restoration of the skull in left lateral view,
X3°5 approx. ang., angular; avt., articular; den., dentary; enp., endopterygoid ;
esc., extrascapular ; fm., hyomandibular ; /.e., lateral ethmoid ; mpt., metapterygoid ;
mx., maxilla; mx.p., maxillary process of palatine ; ovs., orbitosphenoid ; pa., parietal ;
pt.f., post-temporal fossa; qgu., quadrate; so.s., subocular shelf; ssc., suprascapular.

92 CRETACEOUS FISHES FROM THE LEBANON

subocular shelf (so. s.) on the second, third and fourth infraorbitals, but not appar-
ently on the first, although the shelf on the second infraorbital extends forwards some
way along the medial face of the first. The lachrymal is deeper than the infra-
orbitals but does not extend back below the first infraorbital as it does in Caproberyx
(Text-fig. 10). The groove for the sensory canal on the lachrymal is covered for
most of its length by a broad bridge of ornamented bone, with two large pores in its
ventral part. Dorsally the lachrymal articulated with the lateral ethmoid by a
broad facet. As noted above, there is no antorbital.

Palate and jaws. The hyomandibular (hm.) has a broad, undivided head, and is
inclined backwards a little. The symplectic inclines forwards at about 35° to the
hyomandibular so that the condyle of the quadrate (qu.) lies below the posterior part
of the orbit. The ectopterygoid and palatine are toothed but the endopterygoid
(enp.) is toothless. The large maxillary process of the palatine (mx. p.) fits in a
broad groove on the dorsal surface of the maxilla.

The premaxilla has a low ascending process, which is only a little over a quarter of
the length of the toothed alveolar process and is hardly longer than the articular
process of the bone. The toothed border of the premaxilla becomes broader and
projects a little anteriorly, but the anterior teeth are not enlarged as they are in
Holocentrus and Myripristis. The maxilla (mx.) is more than two-thirds as long
again as the premaxilla and is expanded behind the latter, but is without teeth.
The posterior expansion of the maxilla has a few weak ridges near the ventral margin.
There are two supramaxillae, the posterior with a process overlapping the anterior.
The posterior supramaxilla is strongly ornamented with longitudinal ridges. The
anterior supramaxilla is rather large, extending forwards well beyond the tip of the
process on the posterior bone, and is ornamented with a few very weak ridges.

The mandible is long, and moderately deep. The coronoid process of the dentary
bears teeth to its tip; the band of teeth does not become much broader at the
symphysis and the teeth are not enlarged there. The ventral parts of the dentary
(den.) and articular (art.) are ornamented with strong longitudinal ridges. There is
a small angular (ang.) at the back of the jaw. The mandibular sensory canal ran ina
tube, closed in the articular but with five pores in the dentary, two at the symphysis,
two equally spaced along the bone, and one at the suture with the articular.

Opercular bones. The preopercular is long and strongly angulated, with its vertical
limb inclined backwards a little. The posterior edge of the bone bears ridges ending
in serrations which grow stronger towards the angle, where one is enlarged into a
small spine. The edge of the flange covering the groove for the sensory canal is
smooth except near the angle, where there are a few weak, blunt serrations and a
single narrow bridge over the groove.

The opercular is large and strongly ornamented. Radiating from the point of
suspension of the bone there are sinuous, bifurcating ridges and on the ventral
two-thirds of the bone there is also a series of strong, parallel, spiny ridges, each
ending in a coarse serration on the edge of the bone. Opposite the point of suspen-
sion there are five or six spines on the edge of the bone, with weaker serrations above
them. The ornament of the opercular is very like that in Hoplopteryx simus, a
trachichthyid from the English Chalk (Smith Woodward, 1902, pl. 8, fig. 2). On

CRETACEOUS FISHES FROM THE LEBANON 93

the antero-dorsal corner of the opercular there are two or three small scales. The
ornament of the subopercular is similar, with parallel ridges ending in serrations,
interspersed with weaker, sinuous ridges. The large interopercular has coarse,
sinuous ridges on the ventral two-thirds of its surface, and serrations on its ventral
edge. In the centre of the edge of the interopercular there is an excavation: a
similar excavation occurs in the living Holocentrus, but there it is covered by the
much enlarged preopercular spine.

Hyoid arch and branchiostegals. The ceratohyal is very deep and is perforated by
a large, oval fenestra : it resembles that of Caproberyx (Patterson, 1964, text-fig. 73)
rather than the shallow, unperforated ceratohyals of living holocentrids. There are
eight branchiostegals, the four anterior rays articulating with the medial face of the
ceratohyal, the four posterior ones with the lateral face of the bone. There is a
large, rectangular urohyal.

Vertebral column. There are 26 vertebrae, eleven abdominal and fifteen caudal.
All the abdominal vertebrae except the first bear ribs, which are inserted on trans-
verse processes on the last three. There appear to be no epipleurals. The caudal
skeleton will be described in detail in a forthcoming work, but it agrees with that of
Holocentrus (Gosline, 1961: 14) in having a neural crest on the second pre-ural
centrum and the second ural centrum fused with the fused first pre-ural and ural
centra, although the line of fusion is clearly visible and the caudal skeleton appears
more “ upturned ”’ than in living holocentrids. The neural and haemal spines of the
first three pre-ural vertebrae are expanded, as they are in living holocentrids.

Pectoral and pelvic girdles and fins. The pectoral girdle contains the normal
berycoid complement of bones. The extrascapular (esc.) has serrations on its

Fic. 9. Stichocentrus livatus gen. et sp. nov. Restoration of the skeleton,
scales omitted, xX 1-7 approx,

04 CRETACEOUS FISHES FROM THE LEBANON

posterior edge, ridges ending in serrations on its ventral edge, and a ridge over the
groove for the sensory canal in the centre of the bone. On the suprascapular (ssc.)
the sensory canal ran forwards in a groove covered by a vertical flange of bone ;
this flange is ornamented with ridges, and on the edges of the posterior half of the
bone there are ridges ending in serrations. The supracleithrum bears similar ridges
and serrations. The posterior plate of the cleithrum bears only a few sinuous ridges.
The anterior process of the coracoid is long and stout, arching forwards to the tip
of the cleithrum. The four pectoral radials are the usual small, hourglass-shaped
bones, increasing in size downwards. The large pectoral fin contains ten rays, the
third of which is the longest.

The pelvic fins are thoracic, the pelvic bones making contact with the cleithra and
the fins being inserted well in front of the tip of the ventral postcleithrum. The
pelvic fin contains a very stout, ridged spine, equal in length to the longest anal
spine, and seven soft rays.

Median fins. The dorsal fin arises close behind the head, in advance of the
pectorals, and occupies most of the back of the fish. The fin contains nine spines and
eleven soft rays ; the spines are stout, gently curved, and almost smooth. The first
four spines increase in length, the fifth and sixth are equal in length and the last three
decrease in length very slightly. The spines are alternately inclined to right and
left, as they are in living holocentrids, so that in the fossils they appear to be
alternately thick and thin (Pl. 3). In the holotype (64 mm. standard length), on
which Text-fig. 9 is mainly based, the longest dorsal spines are equal in length to a
little less than the distance between the base of the spines and the vertebral column,
but in some of the smaller specimens, particularly 105987 (42 mm. standard length),
the spines are longer, equal in length to about half the maximum depth of the trunk.
The available material suggests that this difference is not taxonomically significant,
but that the length of the dorsal spines was variable in the species and that during
ontogeny the depth of the trunk increased in proportion to the length of the dorsal
spines.

The first radial of the dorsal fin is not enlarged, and is inserted between the second
and third neural spines. The fin is preceded by two predorsals lying between the
first and second neural spines. The spine-bearing dorsal radials are expanded,
meeting each other in slightly dentate sutures.

The anal fin originates below the middle of the soft dorsal and contains five spines,
the first of which is extremely small, and nine soft rays. The spines increase in
length from front to rear and are weakly ridged and very stout. The fifth spine,
although longer than the fourth, is more slender, as in Myripristis (where there are
only four spines). The longest anal spine is rather variable in length: it is longer
than the longest dorsal spine in the holotype, about equal to the longest dorsal spine
in 105736 and 108929, and shorter than the longest dorsal spine in 108923. The
first anal radial is not much enlarged.

The forked caudal fin contains nineteen principal rays with seventeen branched
(nine in the upper lobe, eight in the lower). The principal rays are preceded by four
small spines and one segmented ray above and below.

Squamation. The scales are thick, large and coarsely ctenoid. The exposed area

CRETACEOUS FISHES FROM THE LEBANON 95

of each scale is ornamented with weak ridges and tubercles passing back to the
ctenii on the hind edge. There are twelve scales in a transverse series on the trunk,
with the lateral line passing through the fifth scale from the top, and about twenty-
six scales along the lateral line. Scales cover the cheek and there are one or two
scales on the antero-dorsal corner of the opercular. The scales overlap the bases of
the dorsal and anal fins, both soft and spinous portions, as they do in living holo-
centrids. Between the pelvic and anal fins there is a median series of ridge scales,
not enlarged or much thickened, as in Caproberyx.

AFFINITIES. As has been emphasized in the comparisons made in the description
above, Stichocentrus is a holocentrid berycoid. Falling in the suborder Berycoidei
because of its toothless endopterygoid, pelvic spine and nineteen-rayed caudal, it is
allied with the Holocentridae rather than the Trachichthyidae by the nine dorsal
spines (no more than eight in any trachichthyid), the seven soft rays in the pelvics,
the absence of crests and large mucus cavities on the skull roof, the posterior exten-
sion of the frontals and the small supratemporal fossa, the large mesethmoid, the
rather shallow infraorbitals with a subocular shelf on the last three, and the enlarged
penultimate anal spine.

Three other genera of Cretaceous holocentrids are known: Caproberyx (Lower
Cenomanian of Morocco, Middle Cenomanian of Hakel (p. 98) and Turonian of
England), Tvachichthyoides (Upper Cenomanian of England) and Kansius (Lower
Senonian of Kansas). Tvachichthyoides is known only by a single head (Smith
Woodward, 1902, pl. 8, fig. 5 ; Patterson, 1964: 359). It differs from Stichocentrus
in having the mucus cavities on the skull roof larger and the ornamented areas and
supraoccipital crest smaller, in the deeper, more strongly ornamented infraorbitals,
with the lachrymal extending back below the first infraorbital, and in the shorter
jaws, with the suspensorium inclined forwards. But the general features of the two
skull roofs and the form and ornament of the opercular bones are very similar and in
Trvachichthyoides there is no subocular shelf on the first infraorbital, just as in
Stichocentrus.

From Kansius, in which only the trunk is known (Hussakof, 1929), Stichocentrus
differs in having one or two fewer dorsal fin spines, with a less marked decrease in
length in the posterior spines, and in having the last anal spine longer than the
penultimate. But in size, in the proportions of the trunk and the size and position
of the fins the two genera are very similar.

In Caproberyx (Patterson, 1964 : 341, 416; p. 98) the skull roof resembles that of
Stichocentrus more closely than any other known form, but in Stichocentrus the
supraoccipital crest and supratemporal fossa are smaller and the frontals extend
back farther. Caproberyx also differs from Stzchocentrus in having only six or seven
dorsal fin spines, with the last the longest, in having more soft rays in the anal fin, in
the posterior extension of the lachrymal below the first infraorbital, the subocular
shelf on the first infraorbital, the shorter, more strongly expanded maxilla, etc., but
again there are many similarities between the two genera.

The five anal spines of Stichocentrus are a peculiar feature. No living holocentrid
has more than four, and although five spines have been described in Kansius

96 CRETACEOUS FISHES FROM THE LEBANON

(Hussakof, 1929 : 3) Conrad (1941: 17) has given reasons for doubting that there
were more than four, an interpretation which seems to be confirmed by Hussakof’s
description of the enlarged third spine as having two spines behind it, a feature
known in no other holocentrid. In Caproberyx superbus, which has been described as
having only four spines (Smith Woodward, 1902:12; Patterson, 1964: 357) a
complete series of spines is not well preserved in any specimen, but some individuals,
particularly in P.g153, appear to have had a very small anterior spine making a total
of five, as in Stichocentrus.

Recent workers (Conrad, 1941 ; Nelson, 1955; Dunkle & Olsen, 1959) recognize
two subfamilies in the Tertiary and recent holocentrids, the Holocentrinae, contain-
ing among living genera Holocentrus and Plectrypops according to Conrad (1941),
Holocentrus alone according to Nelson (1955), and the Myripristinae, containing
Myripristis, Ostichthys, Holotrachys, Plectrypops and Corniger according to Nelson
(1955). These subfamilies are differentiated by characters such as the form of the
otic bulla, the spine on the preopercular, the elongated nasals and premaxillary
ascending processes, short jaws and deep spiny lachrymal of Holocentrus. Among
Cretaceous holocentrids, Tvachichthyoides is allied to the myripristine line in
characters of the skull roof, lachrymal and preopercular, while Caproberyx resembles
the holocentrines in the skull roof and the deep lachrymal. Kansius is insufficiently
known to be confidently placed in either subfamily. Stichocentrus resembles
Caproberyx and the holocentrines in its skull roof and infraorbitals, but is closer than
Caproberyx to the living holocentrids in the nine dorsal fin spines, alternately inclined
to left and right and with the last four decreasing in length, and in the enlarged
penultimate anal spine. Thus among the four known genera of Cretaceous holo-
centrids one can see the gradual acquisition of such Recent holocentrid features as the
form of the dorsal and anal fins (in the sequence Caproberyx—Stichocentrus—Kanstus),
the posterior extension of the frontals and the reduction of the supratemporal fossa
and supraoccipital crest (in the sequence Caproberyx—Stichocentrus), the development
of a subocular shelf on all the infraorbitals (in the sequence Tvachichthyoides—
Stichocentrus—Caproberyx). But all these Cretaceous genera (so far as they are
known) differ from the Tertiary and living holocentrids in a number of characters
such as the absence of an antorbital, the simple, tubular nasals, the large pores along
the course of the mandibular sensory canal (in living holocentrids the mandibular
sensory canal is almost entirely enclosed), the presence of scales only on the antero-
dorsal corner of the opercular (in living holocentrids scales cover the whole anterior
part of the opercular), the deep, perforate ceratohyal and the occasional presence of
five anal spines. Some of these characters are certainly primitive, but others such
as the absence of an antorbital and maxillary teeth and the five anal spines are
specialized and seem to exclude the known Cretaceous holocentrids from the direct
ancestry of later members of the family. Certainly the Cretaceous holocentrids are
more closely related to each other than they are to the Tertiary and living genera,
and they cannot reasonably be included in either of the subfamilies recognized among
living forms (these subfamilies seem merely to emphasize the specializations of
Holocentrus itself). At present the evidence suggests that the Holocentridae under-
went two radiations, one in the Cretaceous and one in the Tertiary, the Tertiary forms

CRETACEOUS FISHES FROM THE LEBANON 97

being derived from ancestors as yet unknown. In the Cretaceous radiation the fins
and skull roof gradually acquired features approaching those of modern forms, but
there remain considerable differences between the two groups. I propose that the
Cretaceous forms be included in a new subfamily Caproberycinae, defined as follows :
Holocentridae in which the dorsal fin is undivided, with 6-11 spines, anal with 4-5
spines, the penultimate sometimes enlarged, nasals simple and tubular, no antorbital,
no maxillary teeth, large pores along the course of the mandibular sensory canal,
otic bulla as in Myripristis where known, no preopercular spine, only two or three
scales on the antero-dorsal corner of the opercular, ceratohyal deep and perforate.

One other species must be mentioned in discussing Stichocentrus. This is Hoplo-
pteryx lewisi (Davis), under which name the specimens of Stichocentrus in Beirut were
catalogued by Smith Woodward (in ms.). Stichocentrus livatus and Hoplopteryx
lewist are distinguished by the occurrence of the first only at Hajula, the second only
at Hakel, and by the presence in H. lewisi of only 22-23 vertebrae (not 22-24 as I
stated: Patterson, 1964: 406), a long premaxilla, an unexpanded maxilla, a large
supratemporal fossa extending well forward over the orbit, the penultimate anal
spine not enlarged, etc. But in size, shape and proportions of the trunk, number and
structure of the fin spines and the shape and ornamentation of the superficial bones
of the head the two species are strikingly similar. Further, in P.8689 and P.10709
(H. lewist) there are traces of at least one large scale on the antero-dorsal part of the
opercular, a difference from all other trachichthyids except Lissoberyx (p. 77) and a
resemblance to Stichocentrus. This tends to confirm that the scaleless opercular of
trachichthyids is secondarily derived from a scaly opercular (p. 107). The strong
similarity between Stichocentrus livatus and Hoplopteryx lewisit provides further
evidence of the close relationship between the Holocentridae and Trachichthyidae in
the Cretaceous: in the English Chalk a similar comparison may be made between
Hoplopteryx lewesiensis and Caproberyx superbus, but here we know that the simi-
larities extend to habitat as well as habitus, for in B.M. (N.H.) 33486 two large
individuals, one of C. superbus and one of H. lewesiensis, are preserved side by side
in the same block of Chalk.

Genus CAPROBERYX Tate Regan, torr : 8

Dracnosis. See Patterson (1964: 341), but add “ lachrymal extending back
below first infraorbital, dorsal fin with six or seven spines, dorsal spines not alternately
inclined to right and left, scales moderately large, about 12-15 in each transverse
series ”’.

TYPE SPECIES. Beryx superbus Dixon (1850) (= Berycopsis major Smith Wood-
ward) from the Turonian zones of the English Chalk.

Caproberyx pharsus sp. nov.
(Pl. 4, fig. 2; Text-fig. 10)
Diacnosis. A Caproberyx known only by a specimen lacking the posterior half
of the trunk ; probably about 6-25 cm. in standard length ; length of head slightly

98 CRETACEOUS FISHES FROM THE LEBANON

less (93%) than maximum depth of trunk, the latter probably equal to about 45%
of the standard length ; skull roof without ornament except for a few ridges on the
parietal, the supraorbital flange and near the median mucus cavity, which is partially
roofed, as in Stichocentrus ; infraorbitals with serrated edges, otherwise smooth,
posterior infraorbitals not tubular; hind margins of preopercular and opercular
coarsely serrated, operculum with weak radiating ridges ; dorsal fin with six spines,
pelvic apparently with only six soft rays; scales thick on antero-ventral part of
trunk, thin elsewhere, 15 or 16 scales in a transverse series.

Horotype. B.M. (N.H.) P.47836 (PI. 4, fig. 2), a fish lacking the posterior half of
the trunk, part of the head in counterpart. The only specimen.

HoRIZON AND LOCALITY. Middle Cenomanian; Hakel, Lebanon.

DESCRIPTION. Since this species is known only by a single incomplete and rather
poorly preserved specimen the description which follows is incomplete, and the
assignment to Caproberyx must be regarded as provisional. The specimen has been
prepared by transfer in a resin block.

Measurements and proportions. The length of the preserved part of the fish
(to the second caudal vertebra) is 40 mm. ; the standard length is estimated to have
been about 6:25 cm. The length of the head is 26 mm., the maximum depth of the
trunk 28mm. The species was thus rather deep-bodied, the depth of the trunk
being greater than the length of the head and equal to about 45°% of the standard
length, compared with about 40% in the other known species of Caproberyx.

Neurocramvum. The skull roof is shown in Text-fig. 10. There is a high, triangular
supraoccipital crest (soc.), thickened anteriorly, which is larger than that of Sticho-
centrus (Text-fig. 7) but smaller than in C. superbus (Patterson, 1964, text-fig. 67).
The supratemporal fossa was clearly short and broad, as in Stichocentrus and C.
superbus, but detailed comparisons are not possible. The parietals (pa.) are poorly
preserved, but near the mid-line they bore an area of fine ridged and tubercular
ornament. Laterally, the parietal extends ventrally in the wall of the post-temporal
fossa ; while it is impossible to be certain how much of this wall was formed by the
parietal it appears that there was a broad contact, not covered by the frontal,
between the parietal and the pterotic (fto.) in the wall of the fossa. This is a primi-
tive feature which does not occur in Caproberyx superbus or Stichocentrus, but is
present in Tvachichthyoides and Lissoberyx (Text-fig. 2) and is indicated in Aram-
bourg’s figure (1954, text-fig. 68) of the skull roof of Caproberyx polydesmus. The
frontal appears to extend as far postero-laterally as it does in C. superbus but not so
far as in Stichocentrus and in general the posterior part of the skull roof seems closest
to that of Caproberyx but with a more complete roof to the post-temporal fossa. The
frontals (J. fr., 7. fy.) are much less strongly ornamented than they are in Caproberyx
and Stichocentrus, the only ornament consisting of a few weak ridges on the flange
which roofs the median mucus cavity (see below) and a series of short ridges on the
supraorbital flange of the bone.

The main cephalic sensory canal, passing on to the pterotic from the extrascapular,
ran forwards in an open groove, covered laterally by a raised flange, with the pre-
opercular sensory canal passing ventrally behind this flange in the usual way.

CRETACEOUS FISHES FROM THE LEBANON 99

Fic. to. Capyoberyx pharsus sp.nov. The head of the holotype as preserved, some bones
partially restored from the counterpart, x5 approx. Inset at top right is a diagram of
the left frontal showing the course of the supraorbital sensory canal. a.sm., anterior
supramaxilla ; eo., exoccipital ; esc., extrascapular ; f.hm., articular facet for hyomandi-
bular; hm., hyomandibular; 7.1, 7.2, 2.3, 1.4, infraorbitals 1-4; 1op., interopercular ;
la., lachrymal; /.fy., left frontal; /.m.c., roof of median mucus cavity on left frontal ;
l.mx., left maxilla; /.pm., fragment of head and alveolar process of left premaxilla ;
L.py., left prootic; myo., myodome; na., left nasal; op., opercular; pa., parietal ;
pl., left pleurosphenoid; pop., preopercular; ps., parasphenoid; p.sm., posterior
supramaxilla ; pto., pterotic ; qu., quadrate ; y.fy., right frontal, crushed and displaced ;
y.m.c., roof of median mucus cavity on right frontal; y.mzx., head of right maxilla ;
y.pm., right premaxilla; yv.py., right prootic; so.s., subocular shelf on second infra-
orbital ; soc., supraoccipital crest; sop., subopercular; spo., left sphenotic, its orbital
margin bent upwards; ssc., suprascapular ; sym., symplectic ; vo., vomer.

Passing over the surface of the sphenotic (where there was presumably an anastomosis
with the infraorbital canal), the supraorbital canal ran antero-dorsally on to the
frontal and entered a horizontal tube in the bone through a much elongated foramen

(inset, Text-fig. 10).

There is no sign in C. pharsus of the depression on the supra-

orbital flange of the frontal which in C. superbus and Stichocentrus received a branch
given off from the canal before it entered the bone : the great elongation in C. pharsus
GEOL. 14, 3. 13

100 CRETACEOUS FISHES FROM THE LEBANON

of the foramen through which the canal entered the bone suggests that this branch
was given off within the opening, approaching the condition in living holocentrids
where the branch is given off after the canal has entered the tube in the bone. In
the tube in the frontal the canal gave off the usual posterior branch, which in C.
pharsus was subdivided into four (compared with two in C. superbus and Sticho-
centvus), one large branch passing back in a tube to the hind end of the frontal, a
smaller and shorter branch in a tube above this and two still smaller and shorter
ones below it (inset, Text-fig. 10). The medial branch of the supraorbital canal
passed into an elongated median mucus cavity, similar in size and position to that of
C. superbus but partially roofed by a raised, ornamented flange of bone (l.m.c., 7.m.c.),
as in Stichocentrus (Text-fig. 7). After giving off the median branch, the supraorbital
canal passed forwards within the frontal to its anterior end, giving off a lateral
branch through an elongated pore above the anterior half of the orbit.

Only the left nasal (na.) is preserved, a moderately large, trapezoid bone, without
ornament and apparently tubular, enclosing a broad passage for the sensory canal,
as in Stichocentrus.

The vomer (vo.) is preserved but displaced, and it is impossible to see whether it
bore teeth. The parasphenoid (fs.) is straight and toothless, as is normal in holo-
centrids. The anterior parts of the prootics are visible (J. pr., 7. pr.), showing nothing
to distinguish them from the prootics of C. superbus. A wide myodome (myo.) opens
between the prootics. The elongated facet for the hyomandibular (f. Am.) is formed
by the prootic, sphenotic and pterotic, in the usual way, and there is a large dilatator
fossa in the sphenotic and pterotic above this facet. No basisphenoid is preserved.
The pleurosphenoid (f/.) appears to be larger than in C. superbus and shows the usual
groove for the superficial ophthalmic nerves. Traces of an orbitosphenoid are
preserved, but the shape of the bone cannot be made out.

Infraorbital series. The infraorbital series contains the usual five bones. The
lachrymal (/a.) resembles that of Caproberyx superbus (Patterson, 1964, text-fig. 70)
in depth and in the long postero-ventral process extending back below the first
infraorbital to touch the second ; as in C. superbus there is a broad bridge over the
groove for the sensory canal, with large pores near the ventral margin, but the surface
of the bone is almost smooth. The ventral edge of the lachrymal is serrated. The
first infraorbital (7. r) is shallow and unornamented except for a few ridges on the
posterior part of the flange over the sensory canal. The second, third and fourth
infraorbitals (7. 2-4) are all deeper than the corresponding bones in C. superbus and
much deeper than those of Stichocentrus ; they have a smooth flange over the groove
for the sensory canal and serrated ventral edges. The second infraorbital, in-
complete ventrally, has a broad groove in its centre marking a branch of the sensory
canal. There is a very broad subocular shelf (so. s.) on the second infraorbital, but
whether the shelf was present on all the infraorbitals as in C. superbus and living
holocentrids cannot be seen.

Palate and jaws. The hyomandibular (m.) has a broad, undivided head and a
very broad proximal part, as in C. superbus. The hyomandibular is inclined back-
wards a little, with the symplectic (sym.) and quadrate (qu.) inclined forwards so that
the quadrate condyle lies below the posterior edge of the orbit. The endopterygoid

CRETACEOUS FISHES FROM THE LEBANON 101

is toothless but the anterior part of the ectopterygoid is toothed. The palatine is not
visible.

The premaxilla (7. pm.) has a rather high ascending process, probably equal in
length to about one-third of the toothed alveolar process (although the latter is not
completely preserved). As in Stichocentrus, the tooth patch on the premaxilla
becomes broader anteriorly, but there are no enlarged teeth. The maxilla (/. mx.)
is about 40% longer than the premaxilla, as in C. superbus, proportionately much
shorter than in Stichocentrus. The maxilla is expanded behind the premaxilla and
is without ornament or teeth. There are two supramaxillae, the posterior one
(p. sm.) ornamented with ridges radiating from a longitudinal ridge in the centre of
the bone, the anterior (a. sm.) almost smooth. The posterior supramaxilla bears
the usual process overlying the anterior.

Of the mandible almost nothing is preserved. The bones were clearly only
feebly ornamented.

Opercular series and branchiostegals. The preopercular (popf.) is bent through
about 50°, compared with 60° in C. superbus and Stichocentrus. The dorsal limb of
the preopercular is inclined backwards a little. The broad flange covering the
groove for the sensory canal is smooth, as in C. swperbus, and does not form a bridge
at the angle in the bone as it does in Stichocentrus. The hind edge of the preopercular
is strongly and coarsely serrated but there is no enlarged spine at the angle in the bone.

The posterior edge of the opercular (op.) is produced into spines, with a large one
opposite the point of suspension and a larger one just below, as in C. polydesmus
(Arambourg) but not in C. superbus. The ornament of the opercular, weak ridges
radiating from the point of suspension, resembles that in C. superbus and is quite
unlike that in Stichocentrus. There are two or three scales on the antero-dorsal
corner of the opercular, as is usual in Cretaceous holocentrids. The subopercular
(sop.) and interopercular (iop.) are poorly preserved but appear to have been of
normal form, the interopercular with ridges near its ventral edge, the subopercular
with a few ridges, possibly ending in weak serrations, at its antero-ventral margin.

The hyoid arch and branchiostegals are very imperfectly preserved : most of the
branchiostegals are missing and their number cannot be estimated.

Vertebral column. Only the first thirteen vertebrae are preserved. There are
eleven abdominal vertebrae, the last four with transverse processes on which the
ribs are inserted. Whether there are ribs on the second centrum, as there are in
Stichocentrus and Lissoberyx, cannot be seen. Epipleural bones are visible articula-
ting with the centra of the last three abdominal vertebrae.

Pectoral and pelvic girdles and fins. The extrascapular (esc.) is similar in shape to
that of C. swperbus, with a short, broad ventral limb and a longer, more slender dorsal
limb, but has a smooth posterior edge and only a few weak ridges on the ventral
limb. The suprascapular (ssc.) has a moderately long dorsal limb and a very broad
posterior plate, with a coarsely serrated hind margin. The supracleithrum appears
to be less broad than that of C. superbus and has coarse serrations on the upper
part of its posterior edge. The posterior plate of the cleithrum is smooth. Of the
endoskeletal pectoral girdle nothing can be seen. The ventral postcleithrum reaches
the ventral border of the trunk well behind the origin of the pelvic fin and the pelvic

GEOL. 14, 3. 13§

102 CRETACEOUS FISHES FROM THE LEBANON

girdle was in contact with the cleithra. The pectoral fin contains about eleven rays
and is rather small, its length being only about one-quarter of the maximum depth
of the trunk. The pelvic fin contains a stout, weakly ribbed spine, equal in length
to a little more than one-third of the maximum depth of the trunk, and apparently
only six soft rays, a difference from C. superbus and all other holocentrids which is
possibly due only to imperfect preservation, although the fin appears to be complete.

Median fins. Of the median fins, only the anterior part of the dorsal fin is pre-
served. This contains six smooth spines, increasing in length from front to rear, and
nine soft rays: there were probably two or three more soft rays. The spines are not
alternately inclined to left and right as they are in Stichocentrus and living holo-
centrids. The longest spine, the sixth, is equal in length to about 40% of the
maximum depth of the trunk. The first dorsal radial, supporting the first and
second spines, is inserted between the third and fourth neural spines. Preceding the
dorsal fin there are three predorsals, one lying in front of each of the first three
neural spines.

Squamation. The scales on the antero-ventral part of the trunk are thick and
coarsely ctenoid ; on the rest of the trunk the scales are much thinner and though
their hind margins are nowhere perfectly preserved they appear to be feebly ctenoid.
On the trunk there were fifteen or sixteen scales in each transverse series, with the
lateral line passing through the tenth or eleventh scale above the ventral border.
It is not possible to see whether ventral ridge scales were present. The lateral line
scales are not enlarged or thickened. Scales cover the cheek, the postero-lateral
parts of the skull roof and the antero-dorsal corner of the opercular.

AFFINITIES. Although this species is known only by a single poorly preserved
and very incomplete specimen it shows some points ofinterest. The specimen is shown
to be a member of the Berycoidei by the combination of an orbitosphenoid, two
supramaxillae, a pelvic with a spine and six soft rays, and dorsal fin spines. Within
the Berycoidei the form of the skull roof, with a small supratemporal fossa, a largely
enclosed supraorbital sensory canal with only a small median mucus cavity, and
various other resemblances to the Cretaceous holocentrids Caproberyx and Sticho-
centrus, show it to be a member of the Holocentridae. The specimen differs from all
other Holocentridae in having only six dorsal fin spines, with the sixth the longest,
but in this it is close to Caproberyx, in both species of which (C. superbus (Dixon) and
C. polydesmus (Arambourg)) there are seven dorsal spines. Among other resem-
blances to Caproberyx the most striking is the form of the infraorbitals, with the
lachrymal sending back a long process along the entire length of the first infraorbital
(the lachrymal is similar in Tvachichthyoides). Less important resemblances to
Caproberyx include the form of the posterior part of the skull roof (see p. 98), the
proportions of the maxilla and premaxilla, the opercular ornament and the presence
of epipleurals. All these characters are in contrast with Stichocentrus, but there are
others in which the specimen resembles Stichocentrus and differs from Caproberyx.
These include the roofed median mucus cavity and the simple, non-tubular posterior
infraorbitals. Apart from these characters shared with either Caproberyx or
Stichocentrus, the skull roof and superficial bones of the skull in general are less

CRETACEOUS FISHES FROM THE LEBANON 103

strongly ornamented than they are in any other holocentrid, living or fossil, the
posterior branch of the infraorbital sensory canal is more complex than it is in
Caproberyx and Stichocentrus, there is an extensive contact between the parietal and
pterotic in the wall of the post-temporal fossa, a character otherwise known in
holocentrids only in Tvachichthyoides, and there are only six soft rays in the pelvic
fin, a difference from all other holocentrids and a resemblance to Trachichthyidae.
Reviewing these various similarities and differences, I think it possible that the
species represents a new genus, resembling Caproberyx in many characters, Sticho-
centrus in a few, and being more primitive than either in others. But I am unwilling
to erect a new genus on such fragmentary material and pending the discovery of more
complete specimens refer the species provisionally to Caproberyx.

V. DISCUSSION
There are described in this paper four berycoids, three from Cenomanian beds
(Hakel & Hajula) and one from Senonian (Sahel Alma), representing three new
genera and possibly four. Apart from increasing the number of known Cretaceous
berycoid genera by 50% these new forms show certain features bearing on the
origin and evolution of the Berycoidei.

(1) Distinction between families in Cretaceous Berycoidet

All known Cretaceous Berycoidei can be placed in either the Trachichthyidae
(Hoplopteryx, Acrogaster, Tubantia, Lissoberyx, Gnathoberyx) or the Holocentridae
(Caproberyx, Trachichthyoides, Kansius, Stichocentrus). But as has been shown
above (pp. 80, 97) separation of these families becomes increasingly difficult as more
Cretaceous forms are described. This is borne out by a list of the characters in which
the two families are held to differ in the most recent diagnoses (Patterson, 1964 :
304, 341) :

(i) In Trachichthyidae there are fewer dorsal spines. But in the holocentrid
Caproberyx the number of dorsal spines falls to 6-7, less than in many living and
some fossil trachichthyids ; and in Caproberyx the spines are not inclined alternately
to left and right as they are in most holocentrids.

(u) The pelvic fins have six soft rays in Trachichthyidae, seven in Holocentridae.
But in the holocentrid Caproberyx pharsus (p. 102) there appear to be only six soft
rays (this conclusion is based only on one specimen, but in it the pelvic fin is well
preserved).

(ii) In Trachichthyidae the skull roof bears high crests separating large mucus
cavities, in Holocentridae broad ridges separating small mucus cavities. But in the
trachichthyid Lissoberyx the skull roof is without crests or mucus cavities and has a
sensory canal pattern very like that in the holocentrid Myripristis.

(iv) In Trachichthyidae the supratemporal fossa is large, in Holocentridae the
frontals extend posteriorly and the fossa is small. But in the trachichthyid
Lissoberyx the supratemporal fossa is of moderate size, intermediate between those
of Hoplopteryx (Trachichthyidae) and Caproberyx (Holocentridae).

(v) In Trachichthyidae the mesethmoid is reduced. But in Lissoberyx the
mesethmoid is not much reduced and approaches those of holocentrids in size,

104 CRETACEOUS FISHES FROM THE LEBANON

(vi) The infraorbitals are deep in Trachichthyidae, shallow in Holocentridae.
This difference still holds good for the Cretaceous genera, but in the living holo-
centrids Plectrypops and Holotrachys the infraorbitals are deeper than the lachrymal,
just as they are in most trachichthyids.

(vii) The subocular shelf extends along all the infraorbitals in Holocentridae but
is confined to the second infraorbital in Trachichthyidae. This difference still seems
to hold good, although there is no subocular shelf on the first infraorbital in the
holocentrids Stichocentrus and Trachichthyovdes.

(viii) In Trachichthyidae the gape is large and the maxilla is not much expanded
posteriorly, in Holocentridae the gape is smaller and the maxilla is expanded. But
in the trachichthyids Lissoberyx and, in particular, Gnathoberyx there is no significant
difference from holocentrids.

(ix) In Holocentridae there are scales on the anterior part of the opercular, in
Trachichthyidae the opercular is without scales. But in the trachichthyid Hoplo-
pteryx lewisi there is at least one large scale on the antero-dorsal part of the opercular,
and in Lissoberyx the opercular is almost completely scaled.

Some of the features which differentiate living members of the Trachichthyidae
and Holocentridae are already known to fail in Cretaceous genera (two supramaxillae
in Holocentridae, one in Trachichthyidae; antorbital present in Holocentridae,
absent in Trachichthyidae ; ventral ridge scales present in Trachichthyidae, absent
in Holocentridae : see Patterson, 1964), but the new forms described here show that
almost all the differences between the two families no longer hold good. From a
purely taxonomic point of view the exceptions listed above make it very difficult to
provide adequate diagnoses of the two families, but since there is rarely any difficulty
in deciding to which of the families any particular form belongs (see p. 80) there is
little point in modifying the familial diagnoses by listing exceptions to every
character. From a more general standpoint the Cretaceous Berycoidei give an
excellent and well documented picture of the early evolution of a group. Today the
Trachichthyidae and Holocentridae are well separated, moderately successful groups,
the first containing bathypelagic fishes of wide distribution, the second neritic fishes
of the tropics and sub-tropics. The trachichthyids also seem to be the stem group of
a number of bathypelagic (Melamphaeidae, Stephanoberycidae, Gibberichthyidae,
Korsogasteridae, Berycidae, Anoplogasteridae, Diretmidae) and neritic (Mono-
centridae, Anomalopidae) families, some of which were already differentiated in the
Eocene (Berycidae, Monocentridae ; Casier, 1966), others in the Miocene (Melam-
phaeidae, Ebeling, 1962), but most of which are without fossil record. The
Trachichthyidae and Holocentridae can both be traced back to the Cenomanian,
but in the Upper Cretaceous the two families converge strongly, with a blurring of the
distinctions between them. This shows that in the Upper Cretaceous the “ trachich-
thyid’’ and “ holocentrid’’ facies had not yet become fully established, the Cre-
taceous fauna consisting of forms showing “ experimental’’ combinations of
characters. Simpson (1953 : 340-349) gives an excellent account of similar cases,
mainly in mammals, and rightly comments on the difficulties of attempts to cram the
early, diverging members of phyletic lines into higher categories (in this case families)
based on what these lines are later to become,

105

CRETACEOUS FISHES FROM THE LEBANON

@DUILJUBIOjOY, n
apuiystidisAy eT P |

' snisuDy | 1 |
y :
| |
| | | smiyuadOUIHS
| Y ' sapio !
SIS
] faeg clieze) IsnuusapAjod 1.
xAsaqoidn>
f l i | smsapyd'> !
j ; | Ui) Giga osaessdocUider aaasceen
: F sdaoup ~
' | ce eral , xAiagoss!]
“249 ‘anpioAiag i 7 1 oO. ‘
i snuju>o190w' 14! \ IsIMa]
4 H SISUaISaMa|"} uses ;
: sanbyuo"} | snyjouBosdydab-y | i
‘Isnsojnuids"}y | snus” 4
‘)sn2puAS"y 1 |
(dds y)
Jajsob013y
! xAsaqoujous
DyUuDgn,

“D249 ‘appiAujusy>0o4)

“248 ‘appienydwipjayw

@pDpljua.ouow

AXVILNSL

NVINON&AS

NVINOYNL

a1Gg|Iw

NVINVWON4)

NVIG1V

A diagram suggesting the probable interrelationships of the known Cretaceous

Trachichthyidae (left) and Holocentridae (right) and the living families of Berycoidei.

PIG. II.

Distance from the central, vertical line indicates degree of divergence from the common

ancestor of the Berycoidei.

106 CRETACEOUS FISHES FROM THE LEBANON

My present opinion of the interrelationships of the known Cretaceous Berycoide1
is summarized in Text-fig. rr. General features worth noting in the early history of
this suborder are the mosaic nature of the character combinations in the early forms
and the fact that no genus seems to provide an entirely adequate ancestor for any
later genus.

(71) The Origin of the Berycoider

We have seen that in the Cenomanian the Berycoidei were a group of recent origin
in which much variation occurred. These early berycoids also exhibit a number of
primitive characters which are absent or rare in Tertiary and living berycoids,
including a toothed maxilla (Gnathoberyx, Hoplopteryx macracanthus, Myripristis) a
partially roofed post-temporal fossa (Hoplopteryx, Lissoberyx, Trachichthyoudes,
Caproberyx pharsus) and an antorbital (Hoplopteryx, living holocentrids). In
particular, Lissoberyx is a form which must lie very close to the origin of the suborder,
sharing many characters of Trachichthyidae and Holocentridae. The primitive
features of Lissoberyx have some bearing on the origin of the order Beryciformes as a
whole, a question which I have discussed at some length (Patterson, 1964 : 459).
I thought then that the Beryciformes might be diphyletic, the Berycoidei having an
origin independent of the other two suborders, because I could see no convergence
towards a common type between the two main suborders, Polymixioidei and
Berycoidei, as they were traced back towards their first appearance in the Ceno-
manian, although there is good evidence of convergence towards a basal type within
each suborder. While this is still true of the major characters separating the two
suborders (caudal formula and presence or absence of a pelvic spine, epineurals and
endopterygoid teeth), Lissoberyx is a berycoid which tends towards the polymixioids
in the absence of ornament on the bones of the head, the smooth skull roof without
mucus cavities, the high supraoccipital crest (all characters which I have used in
separating the two suborders: Patterson, 1964 : 433), and in the completely scaled
opercular. In all these characters Lissoberyx also resembles the four monotypic
Cretaceous families placed in the suborder Dinopterygoidei (Patterson, 1964 : 434).
The structure of Lissoberyx therefore increases the possibility that the Beryciformes
is a monophyletic order and suggests that the ancestral form would have been a small
fish with a smooth skull roof, a high supraoccipital crest arising from a moderately
large supratemporal fossa, no ornament on the bones of the head, a reduced antorbital,
a toothed maxilla, a scaly operculum and a partial roof to the post-temporal fossa.
Only the last of these points excludes the Ctenothrissiformes from the direct ancestry
of the group.

One or two other points in the anatomy of the Cretaceous Berycoidei are worth
discussing briefly. All known Cretaceous Beryciformes, both polymixioids and
berycoids, have a very constant sensory canal pattern on the skull roof: the main
features of this pattern are a well developed parietal branch of the supraorbital canal
extending back to open at the hind end of the frontal, usually into a depression on the
parietal, this branch often being subdivided (into five branches in Berycopsis and
Homonotichthys, four in Caproberyx pharsus, etc.), a medial branch meeting its fellow
in a median depression over the orbit, and a lateral branch over the anterior part of

CRETACEOUS FISHES FROM THE LEBANON 107

the orbit. This pattern of the supraorbital canal seems to be very widely distributed
among primitive teleosts, although Gosline, Marshall & Mead (1966 : 3) find that the
medial branches fail to meet in a cross-commissure in “‘ isospondylous”’ fishes. An
exactly similar pattern occurs in the Ctenothrissiformes, except that in Awlolepis
the parietal branch is reduced and does not reach the parietal. A well developed
parietal branch of the supraorbital canal, terminating in or near the parietal, is
generally held to be a relic of the posterior part of the supraorbital canal and the
anterior pit-line of halecostomes and more primitive actinopterygians. Arambourg
(1950; 1954: 34, 72) considers that such parietal branches occur only in the most
primitive teleosts (Clupavus, Thrissopater) but my own observations suggest that
they are more widely distributed, although it is doubtful whether the parietal branches
are strictly primary or even homologous in all teleosts.

The subocular shelf has been shown by Smith & Bailey (1962) to be an important
taxonomic feature. Among living teleosts a subocular shelf occurs only in the
Acanthopterygii (sensu Greenwood et al., 1966) and in the osteoglossiform family
Notopteridae (Greenwood ef al., 1966 : 363), where it is undoubtedly developed
independently. There is also a subocular shelf in all the Cretaceous Beryciformes.
Otherwise, a subocular shelf is only known to occur in the Cretaceous ctenothrissiform
Ctenothrissa (Patterson, 1964 : 229), where there is a narrow shelf on the first and
anterior part of the second infraorbitals. This is powerful additional evidence for
close relationship between the Ctenothrissiformes and the ancestral Beryciformes.

The presence of scales on the opercular is undoubtedly an advanced character in
actinopterygians since scales could not develop over the dermal opercular until it had
lost its covering of enameloid tissue, which can develop only in contact with the
ectoderm. The opercular is without scales in all elopomorph, clupeomorph and
osteoglossomorph teleosts, and in protacanthopterygian teleosts opercular scales are
only present below the myctophoid level in esocoids and alepocephaloids. In
Beryciformes the occurrence of a completely scaled opercular in Cretaceous and living
polymixioids, in dinopterygoids and in Lissoberyx strongly suggests that this feature
is primitive for the order, and presumably for all acanthopterygians. This hypo-
thesis receives support from the completely scaled operculars of the Ctenothrissi-
formes Aulolepis, Pateroperca and the Lebanese species of Ctenothrissa. In Bery-
coidei scales are absent on the opercular in all Trachichthyidae except Lissoberyx and
Hoplopteryx lewist while in Holocentridae the opercular is only scaled anteriorly.
These are clearly cases of secondary reduction from a complete scale covering,
evidently associated with the development of spiny ornament on the opercular. An
analogous situation occurs in the Ctenothrissiformes, where Ctenothrissa microcephala
and C. radians show progressive reduction in the scaling of the opercular associated
with increased ornamentation.

A common ancestor of the three suborders of Beryciformes, all of which are
present and clearly distinct in the Cenomanian, must have lived in the Albian or
earlier. It is worth briefly reviewing the few records of supposed Beryciformes in
pre-Cenomanian beds. Weiler (1947) referred to the Berycomorphi isolated scales
and a pelvic fin from the Upper Aptian of Armenkovy Island, S. Georgia. These
specimens have not been figured and the scales, which were very briefly described,

108 CRETACEOUS FISHES FROM THE LEBANON

are evidently without clearly diagnostic features. The pelvic fin is described as
containing a short, slender spine, one unbranched soft ray and six branched rays.
These specimens seem doubtful evidence of Beryciformes in the Aptian. Much
earlier in time, beryciform otoliths have been described from the Tithonian and
Wealden of Germany (Martin & Weiler, 1954, 1957). These Jurassic and Wealden
otoliths must be treated with some reserve: since otoliths can at present only be
identified by comparison with living forms the accuracy of otolith determinations
must decrease with increasing age.

In the Gault of S.E. England (at Small Dole, near Henfield, Sussex, Ford Place,
Wrotham, Kent, and King’s Lynn, Norfolk) there occur fragmentary remains of two
small teleosts, one with ctenoid scales bearing parallel rows of small, flat spines which
resemble those of Ctenothrissa and some berycoids, the other with scales in which the
exposed circuli are broken up into very small tubercles. As yet, I have seen no
fin spines or other structures diagnostic of Beryciformes with these fragments, but
various skull bones are very suggestive of the Ctenothrissiformes and Beryciformes.
The first of these small teleosts appears to have a toothed maxilla and two very small
patches of teeth on the endopterygoid: the second has a broad, cavernous pre-
maxilla of typical trachichthyid type. Since fragments of these small fishes are not
uncommon at certain horizons in the Middle Gault it may eventually prove possible
to give an account of Beryciformes in the Middle Albian.

VI. REFERENCES

ARAMBOURG, C. 1950. Nouvelles observations sur les Halécostomes et l’origine des Clupeidae.
C.R. Acad. Sci., Paris, 231 : 416-418, 1 fig.

1954. Les poissons crétacés du Jebel Tselfat (Maroc). Notes Mém. Serv. géol. Maroc,
Rabat, 118 : 1-188, 20 pls.

BaARDACK, D. 1965. Anatomy and evolution of chirocentrid fishes. Paleont. Contry. Univ.
Kans., Topeka, Vertebrata 10 : 1-88, pls. 1, 2.

BRONGERSMA-SANDERS, M. 1957. Mass Mortality in the Sea. Mem. Geol. Soc. Amer.,
Washington, 67, 1 : 941-1,010, 7 figs.

CaSIER, E. 1966. Faune ichthyologique du London Clay. xiv + 496 pp., 68 pls. Brit. Mus.
(Nat. Hist.), London.

ConrapD, G. M. 1941. A fossil squirrel-fish from the Upper Eocene of Florida. Geol. Bull.
Dept. Conservation, Florida, Tallahassee, 22 : 5~25, 3 pls.

DuNKLE, D. H. & OtsEn, S. J. 1959. Description of a beryciform fish from the Oligocene of
Florida. Spec. Publs Fla geol. Suvv., Tallahassee, 2 : 1-20, 4 figs.

DUBERTRET, L. 1963. Liban et Syrie. Lewxique Strat. Int., Paris, 3, tocl : 7-155, 3 pls.
EBELING, A. W. 1962. Scopelogadus (?) capistraensis, a new fossil melamphaid (Pisces,

Teleostei) from Capistrano Beach, California. -Postilla, New Haven, 71: 1-6, 1 fig.
b’Erasmo, G. 1922. Catalogo dei Pesci fossili della Tre Venezie. Memorie Ist. geol. miner.

Umi. Padova, 6: 1-181, 6 pls.

1946. L’ittiofauna cretacea di Comeno nel Carso triestino. Atti Accad. Sci. fis. mat.
Napoli (3a) 2, 8 : 1-136, 1 pl.

GOSLINE, W. A. 1961. Some osteological features of modern lower teleostean fishes.
Smithson, Misc. Coll., Washington, 142, 3 : 1-42, 8 figs.

1963. Considerations regarding the relationships of the percopsiform, cyprinodontiform,
and gadiform fishes. Occ. Pap. Mus. Zool. Univ. Mich., Ann Arbor, 629 : 1-38, 11 figs.
GosLinE, W. A., MARSHALL, N. B. & Meap, G. W. 1966. Iniomi. Characters and Synopsis

of Families. Mem. Seavs Fdn. mar. Res., New Haven, 1, 5: 1-18, figs. 1-6.

CRETACEOUS FISHES FROM THE LEBANON 109

GREENWOOD, P. H., Rosen, D. E., Weitzman, S. H. & Myers, G.S. 1966. Phyletic studies
of Teleostean Fishes, with a provisional Classification of Living Forms. Bull. Amer. Mus.
Nat. Hist., New York, 131 : 339-456, pls. 21-23, 32 charts.

Hay, O.P. 1903. Onacollection of Upper Cretaceous fishes from Mount Lebanon, Syria, with
descriptions of four new genera and nineteen new species. Bull. Amer. Mus. Nat. Hist.,
New York, 19 : 395-452, 14 pls.

Hussakor, L. t929. A new teleostean fish from the Niobrara of Kansas. Amer. Mus. Novit.,
New York, 357 : 1-4, 2 figs.

KRAMBERGER, C. G-. 1895. De Piscibus fossilibus Comeni, Mrzleci, Lesinae et M. Libanonis.
Djela jugosl. Akad. Znan. Unyetn., Zagreb, 16 : 1-67, 12 pls.

MarsHatt, N. B. 1961. A young Macristium and the Ctenothrissid fishes, Bull. Br. Mus.
nat. Hist. (Zool.), London, 7 : 353-370, 4 figs.

Martin, G.H. 1920. Anguillavus hackberryensis, a new species and a new genus of fish from
the Niobrara Cretaceous of Kansas. Kans. Univ. Sci. Bull., Lawrence, 13: 95-97, pl. 6.

Martin, G. P. R. & WEILER, W. 1954. Fisch-Otolithen aus dem deutschen Mesozoikum

(Dogger bis Wealden). Senck. leth., Frankfurt a.M., 35 : 119-192, pls. 1-4.

1957. Das Aldorfer Otolithen-“‘ Pflaster’’ und seine Fauna. Senck. leth., Frankfurt a.M.,

38 : 211-250, pls. 1-3.

Netson, E. M. 1955. The Morphology of the Swim Bladder and Auditory Bulla in the
Holocentridae. ieldiana, Zool., Chicago, 37 : 121-130, 3 pls.

PATTERSON, C. 1964. A review of Mesozoic acanthopterygian fishes, with special reference to

those of the English Chalk. Phil. Tvans., London (B) 247 : 213-482, pls. 2-5.

1967. A second specimen of the Cretaceous teleost Pyrotobyvama and the relationships of the

sub-order Tselfatioidei. Avk. Zool., Stockholm, 19: 215-234, 8 figs.

Pictet, F. J. & HumBert, A. 1866. Nouvelles vécherches sur les poissons fossiles du Mont
Liban. vii+ 114 pp., 19 pls. Geneve.

ReGAN, C. TaTE. 1911. The anatomy and classification of the Teleostean fishes of the orders
Berycomorphi and Xenoberyces. Ann. Mag. Nat. Hist., London (8) 7: 1-9, 1 pl.

RoceEr, J. 1946. Les invertébrés des couches a poissons du Crétacé supérieur du Liban.
Mém. Soc. géol. Fr., Paris, 23, 2 : 1-92, to pls.

Rosen, D. E. 1964. The relationships and taxonomic position of the halfbeaks, killifishes,
silversides, and their relatives. Bull. Amer. Mus. Nat. Hist., New York, 127: 217-268,
pls. 14, I5.

Smmpson, G. G. 1953. The Major Features of Evolution. xx + 434 pp., 52 figs. Columbia
University, New York.

SmitH, C. L. & Batrey, R. M. 1961. Evolution of the dorsal-fin supports of percoid fishes.
Pap. Mich. Acad. Sci., Ann Arbor, 46 : 345-364, 1 pl.

1962. The Subocular Shelf of Fishes. /. Morph., Philadelphia, 110: 1-17, pls. 1-3.

StTarRKS, E.C. 1904. The osteology of some Berycoid fishes. Pvoc. U.S. Nat. Mus., Washing-
ton, 27 : 601-6109, 9 figs.

WEILER, W. 1947. Fische, pp. 17-18 7m Wilckens, O. Paldontologische und geologische
Ergebnisse der Reise von Kohl-Larsen (1928-29) nach Sud-Georgien. Abh. senckenb.
naturforsch. Ges., Frankfurt a.M., 474 : 1-75, 7 pls.

Woopwarp, A. SmitH. 1901. Catalogue of the fossil fishes in the British Museum (Natural
History), 4. xxxviii + 636 pp., 19 pls. Brit. Mus. (Nat. Hist.), London.

1902. The fossil fishes of the English Chalk. Part I. Mon. Palaeontogy. Soc., London,

1902 : 1-56, pls. 1-13.

—— 1942. Some new and little-known Upper Cretaceous fishes from Mount Lebanon. Ayn.
Mag. Nat. Hist., London (11) 9: 537-568, 5 pls.

PLATE 1
Fic. 1. Lissoberyx dayi (Smith Woodward). AUB 108926, Hajula, Lebanon, prepared
by transfer in a resin block. x 3°5.
Fic. 2. Gnathoberyx stigmosus gen. et sp. nov. The holotype, AUB 100402, Sahel
Alma, Lebanon, before preparation of the upper jaw. x3.

PLALE x

Bull. Br. Mus. nat. Hist. (Geol.) 14, 3

14

GEOL. 14, 3.

PLATE 2
Gnathoberyx stigmosus gen. et sp. nov., Sahel Alma, Lebanon.
Fic. 1. AUB 103838, x 3°25.
Fic. 2. The head of the holotype, AUB 100402, after preparation to show the toothed
maxilla and premaxilla. x 10.

PLATE z

Bull. By. Mus. nat. Hist. (Geol.) 14, 3

PLATE 3

Stichocentrus liratus gen. et sp.nov. The holotype, B.M. (N.H.) P.47835, Hajula, Lebanon
prepared by transfer in a resin block. x 3.

PLATE 3

Bull. Br. Mus. nat. Hist. (Geol.) 14, 3

PLATE 4
Fic. 1. Lissoberyx dayi (Smith Woodward). AUB 107578, Hakel, Lebanon, prepared by
transfer in a resin block. 5:5.
Fic. 2. Caproberyx pharsussp.noy. The holotype, B.M. (N.H.) P.47836. Hakel, Lebanon.
prepared by transfer in a resin block. x 2°85.

PLATE 4

Bull. Br. Mus. nat. Hist. (Geol.) 14, 3

TV:
OF es

+ + ‘ ghee

ew eee

— t , ; : - r
7 < ee rs pubben 2 e: 7 ' i oe
ee

a a. Pe east

PRINTE

Aa

s STRATIGRAPHY AND
_ PALAEOGEOGRAPHY OF THE
_ YORKSHIRE OOLITES AND THEIR
__- RELATIONSHIPS WITH THE
LINCOLNSHIRE LIMESTONE

R. H. BATE

a BULLETIN OF

_ THE BRITISH MUSEUM (NATURAL HISTORY)
_ GEOLOGY Vol. 14 No. 4
ee ! LONDON : 1967 .

0 APR IS
STRATIGRAPHY AND PALAEOGEOGRAPHY & a
OF THE YORKSHIRE OOLITES AND —_
THEIR RELATIONSHIPS WITH THE
LINCOLNSHIRE LIMESTONE

BY

RAYMOND HOLMES BATE _

Pp. 111-141 ; 5 Text-figures ; 4 Tables

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)

GEOLOGY Vol. 14 No. 4
LONDON : 1967

THE BULLETIN OF THE BRITISH MUSEUM
(NATURAL HISTORY), imstituted im 1949, ts
issued in five series corresponding to the Departments
of the Museum, and an Historical series.

Parts will appear at irregular intervals as they become
veady. Volumes will contain about three or four
hundred pages, and will not necessarily be completed
within one calendar year.

In 1965 a separate supplementary series of longer
papers was instituted, numbered serially for each
Department.

This paper is Vol. 14, No. 4 of the Geological
(Palaeontological) series. The abbreviated titles of
periodicals cited follow those of the World List of
Scientific Periodicals.

World List abbreviation
Bull. Br. Mus. nat. Hist. (Geol.).

© Trustees of the British Museum (Natural History) 1967

TRUSTEES OF
THE BRITISH MUSEUM (NATURAL HISTORY)

Issued 21 April, 1967 Price Sixteen Shillings

SERATIGRALEY AND PALAEOGEOGRAPEHY
OF TTaE YORKSHIRE OOLTES<AND
THER RELATIONSHIPS WITH VITHE
LINCOLNSHIRE LIMESTONE

By RAYMOND HOLMES BATE

CONTENTS
Page
I. INTRODUCTION c : : : j : i ‘ ‘ II3
II. STRATIGRAPHY ; : ; " : : : é é 115
III. STRATIGRAPHICAL SECTIONS : ; : ; : : : 120
TV. CORRELATION . ; : : 5 : 5 5 ; ‘ 132
V. PALAEOGEOGRAPHY . : 5 6 : : : . : 134
VI. REFERENCES . : : : : : j : : ; 140
SYNOPSIS

The Lincolnshire Limestone is correlated stratigraphically and on its ostracod fauna with
the Hydraulic Limestone/Eller Beck Bed horizon and the Cave, Whitwell and Millepore Oolites
and associated Upper Limestone and Yons Nab Beds of Yorkshire. The stratigraphy of these
beds is discussed. The Hydraulic Limestone and the Eller Beck Bed are identified as facies
variants of the same marine transgression correlated with the Lower Lincolnshire Limestone.
The Cave, Whitwell and Millepore Oolites, the Upper Limestone and the Yons Nab Beds are
correlated with the Upper Lincolnshire Limestone. The highest beds of the Upper Lincolnshire
Limestone are older than the Grey Limestone Series of Yorkshire which is not represented
in Lincolnshire. Eleven geological sections are described in detail and the palaeogeography
of the marine horizons is discussed.

I INTRODUCTION
In previous publications (Bate 1963, 1963a, 1964) I have dealt with the ostracod
faunas of the Lower Lincolnshire Limestone, the Cave Oolite and underlying marl
and of the Whitwell and Millepore Oolites of Yorkshire. This study involved a
stratigraphical investigation of the beds in question and it is now possible to show
how the oolites, geographically isolated one from the other, were laid down con-
temporaneously in a shallow sea which covered north-eastern England at that time.

The dating of the Yorkshire and Lincolnshire Oolites is rather tenuous at the
present time, and all that can be satisfactorily stated is that they are Bajocian, with
part of the period of deposition taking place during Hyperlioceras discites times.

The purpose of the present paper is to give detailed sections, complete with the
ostracod faunas found therein, and to use these faunas to correlate the Lincolnshire
Limestone and the Yorkshire Oolites and at the same time to suggest the probable
palaeogeography at that time.

In order to retain uniformity throughout related publications on the Middle
Jurassic of north-eastern England I have retained the stratigraphical names used

GEOL. 14, 4. 15

114 YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE

WHITBY

RAVENSCAR

CLOUGHTON
G WYKE

a
ee?
Ge

ae

NORTH

SEA

Mee SCARBOROUGH
KN
OSGODBY NAB
SKIPTON YONS NAB
HILL -
J
FORDON NO.
+
KIRKBY
UNSED 2
YORK
GREAT
S GIVENDALE®) iL LINGTON
MARKET
WEIGHTON -+
SOUTH
CAVE
t
BRANTINGHAM
N
SCUNTHORPE REE
01234 8miles
———d

KIRTON 23
LINDSEY ~

i ~[omiles

1
LINCOLN

Fic. 1. Map showing localities, sections and outcrop of the Lincolnshire Limestone south
of the Humber, and of the correlated marine horizons in Yorkshire. Isolated outcrops
in the north-central part of the map belong to the Eller Beck Bed.

YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE 115

previously. By international agreement the term Series is to be restricted to the
primary division of a System. The units bearing this term here should not be so
regarded.

The localities mentioned throughout the text are indicated in Text-fig. 1, whilst
those which have supplied detailed geological sections are listed below with full map
references.

1. Lincoln—Greetwell Quarry. Lower Lincolnshire Limestone consisting of the
Blue and Silver Beds and the Kirton Cementstone Series, completely exposed.
Map reference TF /002725.

2. Kirton Lindsey—Kirton Cement Quarry. Base of the Upper Lincolnshire
Limestone, the Kirton Shale (Acanthothiris crosst Bed) completely exposed. Map
reference SE/942011.

3. South Cave—Eastfield Quarry. Complete exposure of the Cave Oolite and,
in 1947, of the underlying marl. Map reference SE/913323.

4. Kirkham—road cutting on the Firby Road. Hydraulic Limestone and under-
lying marl. Map reference SE/738658.

5. Whitwell—Seamer Lime and Stone Co. Quarry. Almost complete exposure of
the Whitwell Oolite and sandy base of the Upper Limestone. Map reference
SE/734672.

6. Crambeck—Stonecliff Wood. Top of Whitwell Oolite and almost complete
exposure of the Upper Limestone. Map reference SE/737675.

7. Yons Nab headland—Cayton Bay. Complete foreshore section through the
Millepore Oolite and Yons Nab Beds. Map reference TA/084844.

8. Cloughton. Complete foreshore section through the Millepore Oolite and Yons
Nab Beds. Map reference TA/021958.

g. Hayburn. Complete foreshore section through the Eller Beck Bed. Map
reference TA /017964.

10. Ravenscar. Complete cliff section through the Millepore Bed. Map refer-
ence NZ/9810109.

11. Eller Beck. Almost complete stream section through the Eller Beck Bed.
Map reference NZ/833023.

Acknowledgments. The contents of this paper have been based upon part of a
Ph.D. thesis submitted to the University of Sheffield in 196r.

II STRATIGRAPHY
North Lincolnshire
The Lincolnshire Limestone

The Lincolnshire Limestone forms a thick lens of limestone striking north-south
through Lincolnshire. The facies is variable and in parts (e.g. Ancaster) the lime-
stone is quarried as a valuable building stone. Elsewhere (Kirton Lindsey) it is used
in the manufacture of cement.

Although it is intended to make some reference to the Lincolnshire Limestone in
the south of the county our main concern is with the outcrop in the north, around
Lincoln and Kirton Lindsey. Evans (1952) has written an excellent account of the

116 YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE

Blisworth Clay
BATHONIAN Great Oolite Limestone

Upper Estuarine Series

Upper Lincolnshire Hibaldstow Oolite
Limestone Acanthothiris crossi Bed
Lower Lincolnshire Kirton Cementstone Series
Limestone

BAJOCIAN

Blue & Silver Beds
Lower Estuarine Series
Northampton Sand

Upper Lias

TABLE I. Succession of Middle Jurassic strata in North Lincolnshire.

Lincolnshire Limestone of this region and this work should be consulted for more
detailed information.

The succession in the Lincoln District is as follows (Table 1) :

The base of the Bathonian, the Upper Estuarine Series, overlies the Upper Lincoln-
shire Limestone unconformably, and so far as the ostracod faunas are concerned bears
no relationship to the limestone beneath. A description of the Upper Estuarine
Series ostracods is now in the press.

The Lincolnshire Limestone is divisible into Upper and Lower with further sub-
divisions (in the north) into the Hibaldstow Oolite, the Acanthothiris crosst Bed, the
Kirton Cementstone Series and the Blue and Silver Beds. Towards the south of
Lincolnshire the name Hibaldstow Oolite is changed to the Ancaster Beds. The
Lincolnshire Limestone itself lies unconformably on the Lower Estuarine Series
beneath. This Series tends to be composed of unfossiliferous sands and clays with
occasional plant remains.

According to Swinnerton & Kent (1949) towards the centre of the Lincolnshire
Basin the Lincolnshire Limestone is most complete, the highest beds being restricted
to the Great Ponton—Ropsley area. The Great Ponton Terebratula Beds are the
youngest of all. Apart from the A. cross? Bed at the base, the Upper Lincolnshire
Limestone has generally yielded only a poor ostracod fauna and no detailed sections
are given. The ostracods obtained from samples of the uppermost Upper Lincoln-
shire Limestone are identical with those present lower down. For example the
fauna obtained from a large overgrown quarry at Braceby (map reference TF /009351)
included the following: Praeschuleridea subtrigona subtrigona (Jones & Sherborn),
Aulacocythere punctata Bate, Eocytheridea carinata Bate, Bairdia hilda (Jones),
Ektyphocythere triangula (Brand), Fuhrbergiella (Praefuhrbergiella) arens Bate,
Systenocythere exilofasciata Bate, Cytherelloidea catenulata (Jones & Sherborn)
Pleurocythere kirtonensis Bate, Dolocythere maculosa Bate, Muicropneumatocythere
globosa Bate, Southcavea reticulata Bate, Monoceratina vulsa (Jones & Sherborn),
Paracypris bajociana Bate, and Glyptocythere sp. Of this fauna only Glyptocythere

YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE 117

sp. is not present also in the Lower Lincolnshire Limestone, being confined to the
highest beds of the Upper Lincolnshire Limestone. This is possibly the lowest
occurrence (stratigraphically) of the genus, more typically developed in Yorkshire
within the Grey Limestone Series.

The base of the Upper Lincolnshire Limestone is represented by the A. crossi Bed
which forms a prominent marker across Lincolnshire. In the north at Kirton
Lindsey the 14 feet of strata known as the Kirton Shale contain the brachiopod
Acanthothiris crossi and are regarded here as the lateral equivalent of a much harder
limestone bed further south.

The Kirton Cementstone Series of the Lower Lincolnshire Limestone consists of
poorly oolitic, rather chalky limestones with marl bands interbedded. The Blue and
Silver Beds, on the other hand, tend to be rather more massively bedded and in some
cases strongly oolitic. The Lincolnshire Limestone as a whole extends northwards
as far as the Humber and is represented on the north bank by the Cave Oolite.
Arkell (1933 : 215) correlates this with the Hibaldstow Oolite of Lincolnshire and
points out that in a boring at Brantingham the beds below the Cave Oolite are of
similar facies to the Kirton Beds. A stratigraphical correlation is therefore possible
across the Humber. A few miles further north at South Cave the marine marls
below the Cave Oolite were exposed in 1947 in a sump put down in Eastfield Quarry,
but as will be seen later do not contain an ostracod fauna completely identical with the
Kirton Beds south of the Humber, a number of locally restricted species being
present.

South Yorkshire

The Cave Oolite is a shelly oolitic limestone with sand intercalations overlain by
approximately 12 feet of sand known as the Upper Estuarine Series. It is doubtful,
however, if these sands are of estuarine origin. At the base of the Cave Oolite the
brachiopod Acanthothiris broughensis Muir-Wood (1952:123) is recorded and
together with the record of Acanthothiris crossi s.l. from the Whitwell Oolite further
north has been used to correlate the Whitwell, Cave and Hibaldstow Oolites (Kent,
1955 : 208).

Below the Cave Oolite the marl sequence developed in the region of South Cave
forms part of the Basement Beds (Fox-Strangways, 1892 : 176), a name now restricted
to the marl beds alone (Neale, 1958:164). These marine marls, from 4-6 ft. in
thickness, are no longer exposed. The Basement Beds overlie a thin porcellanous
limestone known as the Hydraulic Limestone, and Fox-Strangways (1892 : 176)
recorded the thickness of this bed in the Market Weighton District as 2 ft. 6 ins.
At the present time it does not appear in section, only as fragments on the surface.
At Ellerker (map reference SE/927297) a small ostracod fauna comprising the
following has been obtained from this bed: Pavacypris bajociana Bate, Progono-
cythere cristata Bate and Ektyphocythere triangula (Brand).

The Lower Estuarine Series, although no longer exposed around Market Weighton
(not sampled), is reported to consist of clays and yellow sandy shales with plant
remains, shells and crustaceans (Neale 1958 : 164).

118 YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE

The Cave Oolite and the underlying beds extend from the Humber to Market
Weighton where they pass beneath the overstepping Chalk.

Much has been written concerning the geological implications of the break in
outcrop which occurs at Market Weighton, admirably reviewed by Kent (1955).
The ideas put forward have varied from Fox-Strangways’ interpretation (1892) that
the Market Weighton area was a land barrier extending from the west, to that of
Kendall’s (1905) that it was an anticlinal feature, a continuation of the Wharfe Axis.
Kent’s interpretation (1955) was that the Market Weighton area was more in the
nature of a broad non-subsiding rigid block.

Jurassic sediments from Rhaetic times onwards were affected by the structure and
thin appreciably towards it from both north and south. During the period of deposi-
tion with which we are concerned in this paper, it is doubtful whether there was
continuous deposition over the structure north-south but rather continuation was
effected around the structure to the east. The interpretation that the Market
Weighton area was a stable region during Middle Jurassic times, acting as a partial
land barrier, is accepted here.

North-East Yorkshire
Hydraulic Limestone/Eller Beck Bed
54+ miles north of Market Weighton the Middle Jurassic sediments reappear at
Millington and again at Great Givendale, but are only continuous at outcrop from
Kirkby Underdale, 34 miles farther north.
The succession of beds in this part of the Jurassic outcrop is indicated in Table 2.
At the base there is a marine sandstone known as the Dogger. This bed does not

N.E. INLAND N.E, COASTAL

Upper Deltaic Series Upper Deltaic Series

rey Limestone Series rey Limestone Series

Middle Deltaic Middle Deltaic
Series (Upper) Series (Upper)

STAGE N.E CENTRAL

CALLOVIAN Cornbrash

Upper

BATHONIAN Upper Deltaic Series

Estuarine

Grey Limestone Series

Series
Middle

Deltaic
BAJOCIAN Series

Basement Beds} Middle Deltaic Middle Deltaic
Series (Lower) Series (Lower)
Hydraulic Eller Beck Bed / Eller Beck Bed /
Limestone Hydraulic Limestone Hydraulic Limestone
Lower Lower Deltaic Series Lower Deltaic Series

Estuarine

TABLE 2, Succession of Middle Jurassic strata in Yorkshire,

Eller Beck Bed

Lower Deltaic Series

Dogger

TOARCIAN

SOUTH

UPPER
LINCOLNSHIRE

LIMESTONE

LOWER 25

LINCOLNSHIRE

LIMESTONE 19

au

SOUTH

NORTH

UPPER
UPPER
LINCOLNSHIRE / LIMESTONE
LIMESTONE

2. KIRTON
LINDSEY

WHITWELL
CAVE

OOLITE
OOLITE

LOWER.

LINCOLNSHIRE

LIMESTONE Ww

BASEMENT
BEDS

3. SOUTH
CAVE

5. WHITWELL

1 LINCOLN 4. KIRKHAM

Fic. 2 Sections 1-6,

YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE 119

appear to be developed south of Acklam, but as it is of no particular importance here
will not be dealt with further.

Overlying the Dogger is the first evidence of the northern delta, the Lower Deltaic
Series. This deltaic sequence is the first of three major deltaic episodes which
interfinger with marine incursions entering from the east and south-east. The source
of the deltaic sediments lies to the north of the Yorkshire Basin. The Lower Deltaic
Series varies in thickness from about 50 ft. at Acklam in the south-west to as much
as 160 ft. in the north.

Above the Lower Deltaic Series comes the first marine episode, represented by the
Hydraulic Limestone, in lithology exactly the same as in the South Cave area.
Outcrops are rare and for the most part the course of the limestone is traced only on a
break in land slope and by fragments which commonly occur at the surface. The
only outcrop encountered here was at Kirkham (section No. 4), although loose debris
from an early working on the hillside behind Castle Howard Station, map reference
SE/737668, provided a good source.

The Hydraulic Limestone is found at the surface only along the western flanks of
the Yorkshire Basin (the centre being obscured by Alluvium of the Vale of Pickering).
In the east it probably occurs out to sea but has been found in the Fordon borehole
(Falcon & Kent 1960 : 27, where it was incorrectly named the Ellerbeck Bed) asa
4{t. bed of cementstone. In the west the Hydraulic Limestone facies is recognizable
as far north as Skipton Hill on the western flanks of the Hambleton Hills. To the
north of this region a marine horizon is recognizable, represented by a thick, ripple-
marked sandstone with associated fossiliferous ironstone bands, known under the
single name of Eller Beck Bed, after the type locality on the Eller Beck (section
No. 11). This facies is found over the whole of the north-central outcrop of the
Middle Jurassic and in the coastal exposures to the east, where a good section was
measured close to Hayburn Wyke (section No. 9). The Eller Beck Bed facies
outcrops to the south of Hayburn but does not extend as far south as Cloughton, the
outcrop striking out to sea. The Hydraulic Limestone facies correlated with the
Eller Beck Bed by Fox-Strangways (1892: 194, pl. 4) almost certainly comes in
again (out to sea) a few miles south of Cloughton.

The Millepore|W hitwell Oolite

One of the more important marine horizons in the Yorkshire Basin is that repre-
sented by oolitic sediments known locally in the east as the Millepore Oolite and in
the west as the Whitwell Oolite. Although geographically isolated these beds have
long been correlated (Wright 1860:32; Fox-Strangways 1892:206). The
Millepore/Whitwell Oolite horizon is separated from the Hydraulic Limestone/Eller
Beck Bed horizon below by deltaic sediments of the Lower Middle Deltaic Series and
overlain by sediments of the Upper Middle Deltaic Series. As the oolitic horizon
does not, in fact, extend all over the northern part of the Basin, in that part the
Upper and Lower Middle Deltaic Series are not identifiable as separate units.

In the east the Millepore Oolite first appears at outcrop in Gristhorpe Bay with the
best section exposed at Yons Nab headland (section No. 7). The Millepore Oolite is,
however, present at depth in the Fordon borehole (Falcon & Kent 1960), where

GEOL, 14, 4. 15§

120 YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE

57 {t. of oolite and sandstone is recorded. At Yons Nab the Millepore Oolite is
represented by 15 ft. of false-bedded oolite and overlain by 25 ft. of marine sand-
stone and shale known as the Yons Nab Beds, a name first introduced by Sylvester-
Bradley (1953 : 37). These contain a rich shallow water lamellibranch fauna as well
as plentiful ostracods and plant remains. The Millepore Oolite is well developed at
Osgodby Nab, the northern headland of Cayton Bay, but is there overlain by 20-30 ft.
of deltaic sandstone and not by the Yons Nab Beds. This sandstone probably
represents a contemporaneous deltaic distributary (Bate 1959 : 163). At Cloughton
Wyke, seven miles to the north, a reduced thickness of the Millepore Oolite (xo ft.
I in.) is overlain by 7 ft. 6 ins. of arenaceous shale and sandstone containing marine
fossils. This upper marine horizon, equivalent to the Yons Nab Beds, can be seen
to pass laterally into a typical deltaic sandstone only a few hundred yards from the
section. The Millepore Oolite similarly passes laterally northwards into a pure
sandstone facies which at Ravenscar (section No. 10) still contains casts of marine
fossils, e.g. ““ Tvigonia’’ sp. Farther north it is impossible to distinguish this horizon
from the deltaic sandstones of the region.

Along the western outcrop of the Middle Jurassic the oolitic facies is known as the
Whitwell Oolite after its development at Whitwell (section No. 5). The rock is a
coarse-grained oolite here often massively bedded and developing some clay beds
towards the top. Some 33 ft. gins. of sediment belonging to the Whitwell Oolite
can be seen, together with 11 ft. 9 ins. of sand and sandy oolite which occur above.
At Crambeck (section No. 6), $ mile to the east, only the upper part of the Whitwell
Oolite is exposed together with a more complete development of the overlying beds
which consist of 12 ft. of sand and sandstone, a 1 ft. limestone bed and 11-12 ft.
of flaggy oolite, the latter termed the Upper Limestone by Hudleston (1873 : 327)
and retained here to incorporate the limestone and the underlying sands. The Upper
Limestone is a localized upper division of the Whitwell Oolite comparable to the
Yons Nab Beds in the east, and may be traced northwards for a distance of about
ro miles and southwards for about 2 miles from the Whitwell Quarries. From
Burythorpe the Whitwell Oolite may be traced to Kirkby Underdale as a continuous
outcrop and as an isolated outlier at Great Givendale. Although the Muillepore
Oolite in the east thickens to the south (see Fordon borehole) the Whitwell Oolite
appears to thin onto the Market Weighton structure.

Like the Millepore Oolite the Whitwell Oolite, when traced northwards, becomes
progressively more sandy until at Kirkby Knowle it is nothing more than a false-
bedded coarse-grained sandstone identified more on its position in relation to other
beds than on its fossil content.

The stratigraphy and palaeogeography of the other important marine horizon of
the Yorkshire Middle Jurassic, the Grey Limestone Series, have been dealt with
elsewhere (Bate 1965).

III STRATIGRAPHICAL SECTIONS
All the ostracods listed below have been described by Bate (1963, 1963a, 1964) and
are represented in the collections of the Department of Palaeontology, British
Museum (Natural History).

YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE 121

SECTION No. 1. Lower Lincolnshire Limestone (Text-fig. 2), complete section in
Greetwell Quarry, Lincoln, of the Blue and Silver Beds and of the Kirton Cement-
stone Series.

Kirton Cementstone Series

fil. ™m.
33. Soft grey marl passing upwards into subsoil. Ostracods abundant :
Bairdia hilda, Progonocythere cristata, Ektyphocythere triangula,
Cytherella fullonica and Praeschuleridea subtrigona subtrigona

seento I to)

32. Massive grey oolitic limestone . I 3

31. Coarse white oolite with abundant shells . , : , toa 2
30. Yellow marl with ooliths. Ostracods: Aulacocythere punctata,

Cytheromorpha ? greetwellensis, Progonocythere cristata, Systeno-

cythere exilofasciata, Dolocythere maculosa, Praeschuleridea sub-

trigona subtrigona, Fuhrbergiella (Praefuhrbergiella) arens and

Ektyphocythere triangula : : : ; : , eo i
29. Coarse white oolite. Ostracods: Aulacocythere punctata, Systeno-

cythere extlofasciata, Pneumatocythere carinata, Dolocythere macu-

losa, Bairdia lulda and Praeschuleridea subtrigona subtrigona . 1 )
28. Coarse rubbly oolite. Ostracods: Fuhrbergiella (P.) arens,

Systenocythere exilofasciata, Pneumatocythere carinata, Mucro-

pneumatocythere convexa, Southcavea reticulata, Asciocythere

lacunosa, Dolocythere maculosa and Praeschuleridea subtrigona

subtrigona . : ; : : ; : einai 6
27. Shelly blue-hearted oolite. Corals and horizontal burrows . “ay 9
26. Grey oolitic shale with wedges of marl. Ostracods: Dolocythere

maculosa, Pneumatocythere bajociana, Acanthocythere (Proto-

acanthocythere) faveolata, Micropneumatocythere convexa, Systeno-

cythere exilofasciata, Aulacocythere punctata and Praeschuleridea

subtrigona subtrigona . : : ie) De S=TO
25. Coarse shelly oolite with fraements of id als von “Ostracods -

Ektyphocythere triangula, Asciocythere lacunosa, Monoceratina

vulsa, Eocytheridea faveolata, E. reticulata & E. lacunosa, Micro-

pneumatocythere globosa & M. convexa, Aulacocythere punctata,

Dolocytheremaculosa, Systenocythere exilofasciata, Pneumatocythere

bajociana, Camptocythere lincolnensis, Cytheropterina conuca &

C. gravis, Praeschuleridea subtrigona subtrigona and P. ventriosa. 0 8
24. White marlstone. Ostracods rare : Pneumatocythere ee and

Praeschuleridea subtrigona subtrigona a | 8
23. Chocolate-brown marl, weathering grey. Ostracods rare: ‘Pnew-

matocythere bajociana and Micropneumatocythere globosa_ . SONG 2

22. White marlstone. Ostracods rare: Praeschuleridea subtrigona
subtrigona + indet. species . : : : é ; be BE 2

21.

20.

19.

18.

efi

16.

i.

T4.

ce

IZ.
aI Cr

YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE

Chocolate-brown, weathering grey marl. Ostracods rare: Pneu-
matocythere bajociana and Dolocythere maculosa :
Rubbly white marl with scattered ooliths. Ostracods common :
Praeschuleridea subtrigona subtrigona, Pneumatocythere bajociana,
Micropneumatocythere globosa, Dolocythere maculosa, Systeno-
cythere exilofasciata, Cytheropterina comica, Ektyphocythere
triangula, Cytherelloidea catenulata, Paracypris bajociana,
Monoceratina vulsa and Acanthocythere (P.) faveolata
Blue-hearted fine-grained limestone weathering cream. Nevined
and lamellibranch shells common. Scattered ooliths. Ostra-
cods rare: Muicropneumatocythere convexa and Cytherelloidea
catenulata ; :
Grey marl with scattered ooliths and maretone fraemente) Ostra-
cods : Dolocythere maculosa, Pneumatocythere bajociana, Ektypho-
cythere triangula and Praeschuleridea subtrigona subtrigona .
White marlstone with abundant ostracods: Praeschuleridea sub-
trigona subtrigona & P. ventriosa, Ektyphocythere triangula,
Cytheropterina comica & C. gravis, Pneumatocythere bajociana,
Acanthocythere (P.) faveolata, Systenocythere exilofasciata, Dolo-
cythere maculosa, Micropneumatocythere globosa and Cytherelloidea
eastfieldensis : : é ‘ : : : ;
Oolitic marl. Ostracods: Pneumatocythere bajociana, Praeschu-
leridea subtrigona subtrigona, Dolocythere maculosa, Systenocythere
extlofasciata, Cytheropterina comica and Micropneumatocythere
globosa 3 : : :
White marlstone sth scattered eolithe: Ostracods: Platella
gurassica, Cytheropterina gravis, Acanthocythere (P.) faveolata,
Eocytheridea faveolata, Micropneumatocythere globosa, Ektypho-
cythere triangula, Cytherelloidea catenulata, Systenocythere exalo-
fasciata and Praeschuleridea subtrigona subtrigona
Shaly oolitic marl. Ostracods: Systenocythere exilofasciata,.
Pneumatocythere bajociana, Camptocythere lincolnensis, Cythero-
pterina gravis, Micropneumatocythere globosa and Dolocythere
maculosa

Blue and Silver Beds

Cream limestone with scattered ooliths, passing laterally into more
coarsely oolitic limestone. In part also a shelly limestone.
Ostracods not common : Micropneumatocythere globosa, Cytherel-
loidea catenulata and Praeschuleridea subtrigona subtrigona.

Fine grained limestone with scattered ooliths

Sandy shale—no microfauna

ft.

in.

Io.

Ob

Te

YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE

Cream marlstone with shells and scattered ooliths. Ostracods :
Cytheropterina comica & C. gravis, Systenocythere exilofasciata,
Praeschuleridea subtrigona subtrigona, Cytherella fullonica, Campto-
cythere lincolnensis, Pneumatocythere Oak Fuhrbergiella
(P.) arens and Bairdia hilda . : ; 5 :

. Brown clay with ooliths. Ostracods rare: Systenocythere extlo-
fasciata, Camptocythere lincolnensis, Pneumatocythere bajociana,
Fuhrbergiella (P.) arens, Cytherella fullonica, Cytherelloidea
catenulata, Cytheropterina comica, ee triangula and
Dolocythere maculosa F : : :

. Cream marly bed with scattered Solis, Ostracods: Praeschu-
leridea subtrigona subtrigona, Dolocythere maculosa and Systeno-
cythere exilofasciata :

. Coarse, cream oolite with reddish Colon ‘besa nid srerstiall Pipes
Ostracods : Cytheropterina gravis and Praeschuleridea subtrigona
subtrigona . :

. Blue hearted fine grained limestone with cream weathering suitinas.
Ooliths and shell fragments present throughout. Ostracods :
Systenocythere es lr ET aaa gravis and indet.
species

. Sandy limestone eee coarse cole”

. Fine grained, ochre coloured sandstone

. Ferruginous oolite .

Lower Estuarine Series
. Grey-green oolitic clay

Northampton Sand Ironstone
Oolitic ironstone with Boxstones and concretions. : seen to
SECTION No. 2. Complete section through the Kirton Shale, Kirton

Cement Quarry, Kirton Lindsey (Text-fig. 2).

3
2.

1fe

Base of Hibaldstow Oolite.

Kirton Shale. Black clay, becoming brownish-buff in upper 3 feet.
Ostracods extremely abundant: Acanthocythere (P.) faveolata,
Aulacocythere punctata & A. reticulata, Dolocythere maculosa,
Bardia lulda, Monoceratina vulsa & M. cf. scrobiculata, Ektypho-
cythere triangula, Praeschuleridea subtrigona subtrigona, Para-
cypris bajociana, Progonocythere cristata, Kuirtonella plicata,
Cytherella fullonica, Cytheromorpha ? greetwellensis, Cytherelloidea
catenulata, Platella jurassica, Pleurocythere kirtonensis & P.
nodosa and Fuhrbergiella (P.) arens

Top of Kirton Cementstone Series,

ft.

fal ©) fal

14

123

mM.

Cour COD

Io

124

17,

16.

15.

14.

30),
12)

IOC

YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE

SECTION No. 3. Cave Oolite, Eastfield Quarry, South Cave, York-

shire. Almost complete section through the Cave Oolite and
(in 1947) partial section through the underlying marls (Text-
fig. 2).

Cave Oolite

. Flaggy oolite, rather coarse grained and compact. Shell frag-

ments, no ostracods . : 4 : : : seen to

. Sandy marl
. Cream, sandy, eolitic limestone, aeey th ‘shails and plane

remains. Ostracods present but Pe preserved and indeter-
minate

. Yellow-brown sandy eval aah cream ooltre manletone: fan

Ostracods present in both lithologies: Ektyphocythere triangula,
Aulacocythere punctata, Micropneumatocythere globosa, Systeno-
cythere exilofasciata, Fuhrbergiella (P.) arens, Eocytheridea
elongata, Praeschuleridea subtrigona subtrigona, Southcavea reti-
culata and S. bajociana .

Fine grained creamy limestone, politic ‘

Shell sand with purple marly parting at top. Osmaceds.: M 1cro-
pneumatocythere convexa, M. globosa, Aulacocythere punctata,
Praeschuleridea subtrigona subtrigona, Systenocythere exilofasciata,
Southcavea reticulata and S. grandis .

Shelly oolite. Ostracods: Eocytheridea ? astricta, Praeeierden
subtrigona subtrigona, Dolocythere maculosa, Fuhrbergiella (P.)
arens, Pleurocythere nodosa, Micropneumatocythere convexa, Aula-
cocythere punctata, Southcavea bajociana, S. reticulata, Ascio-
cythere lacunosa and A. acuminata

Shell sand with purple marly partings at top " Ostracods :
Aulacocythere punctata, Southcavea reticulata, Micropneumato-
cythere convexa, Praeschuleridea subtrigona subtrigona, Pneumato-
cythere carinata, Ektyphocythere triangula, Fuhrbergiella (P.)
arens and Dolocythere maculosa ‘ é : : :

Blue hearted creamy limestone. Shelly and in part oolitic .

Ochre coloured shell sand with Pentacrinus ossicles, echinoid spines
and shell fragments. Ostracods: Pnewmatocythere carinata,
?Pneumatocythere bajociana, Praeschuleridea subtrigona magna,
Aulacocythere punctata, Fuhrbergiella (P.) arens, Southcavea
reticulata, S. grandis, Eocytheridea carinata, E. elongata, Ascio-
cythere lacunosa, Dolocythere maculosa, Systenocythere exilo-
fasciata, Ektyphocythere triangula and A canthocythere (P.) faveolata

Blue hearted, coarsely oolitic limestone. Ostracods: Dolocythere
maculosa and Praeschuleridea subtrigona magna

tn ae
3 0)
0 3
2 3
0 7
2 3
0) 3)
(0) II
7

I 9
0) 8
I 8

YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE

10. Creamy-yellow, oolitic limestone, very fossiliferous and crowded
with Pentacrinus ossicles. Ostracods: Asciocythere lacunosa,
Praeschuleridea subtrigona magna, Southcavea bajociana, S.
grandis, S. reticulata, Cytherelloidea catenulata, Pleurocythere
nodosa, Ektyphocythere triangula, Paracypris bajociana, Systeno-
cythere extlofasciata, Bairdia hilda, Micropneumatocythere convexa,
Dolocythere maculosa, Fuhrbergiella (P.) arens, Eocytheridea ?
astricta, E. elongata and E. carinata : : : :

g. Coarsely oolitic cream oolite. Ostracods: Monoceratina vulsa,
Praeschuleridea subtrigona subtrigona, Paracypris bajociana and
indeterminate ostracods : : : : : :

8. Shelly, coarsely oolitic limestone. Ostracods: Southcavea bajo-
cana & S. reticulata, Eocytheridea carinata & E. faveolata,
Aulacocythere punctata, Micropneumatocythere globosa, M. con-
vexa, Paracypris bajociana, Ektyphocythere triangula, Fuhrber-
giella (P.) arens, Asciocythere acuminata, Praeschuleridea sub-
trigona subtrigona, Systenocythere exilofasciata, Dolocythere macu-
losa, Acanthocythere (P.) faveolata and Pneumatocythere carinata

Base of Cave Oolite section as seen in Eastfield Quarry. Section
continued in sump excavated in 1947. Details of section from
Professor P. C. Sylvester-Bradley.

Basement Beds

7. Rubbly pellety limestone with lamellibranchs . . seen to

6. Ferruginous limestone full of shells :

5. Ferruginous marl full of lamellibranchs and crinoid stems. This
marl has been dumped by the side of the sump and has provided
the following ostracod fauna: Cytherelloidea eastfieldensis,
Paracypris bajociana, Progonocythere reticulata, Monoceratina
vulsa, Acanthocythere (P.) faveolata, Aulacocythere punctata,
Micropneumatocythere convexa, Pneumatocythere bajociana, Pleuro-
cythere kirtonensis, Pleurocythere sp., Dolocythere maculosa,
Homocytheridea cylindrica, Tetracytheridea punctata, Asciocythere
lacunosa, Eocytheridea elongata, E. lacunosa, E. ? astricta, E. ?
erugata, Paraschuleridea ornata, Paraschuleridea sp., Praeschu-
leridea ventriosa, Cytheropterina conuica, C. gravis, Ektyphocythere
triangula, Southcavea bajociana and Systenocythere exilofasciata .

4. Shale, shelly at base, marlstone impersistant at top

3. Nodular limestone with lamellibranchs :

2. Grey limestone full of gastropods and crushed lamellibranchs

I. Shelly shale. Ostracods: Progonocythere reticulata, Eocytheridea
lacunosa, Praeschuleridea ventriosa, Asciocythere lacunosa, A.
acuminata, Cytheropterina conuca, Micropneumatocythere convexa,

ft.

Ist ©) Tal Isl

125

mn.

Io
It

126 YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE

ft.

Pneumatocythere bajociana, Dolocythere maculosa and Ektypho-
cythere triangula . : : : ; : : seen to

SECTION No.4. Hydraulic Limestone and associated marls exposed
in a road cutting at Kirkham. The Firby Road section (Text-fig. 2).
These sediments come below the marine succession exposed in the
Eastfield Quarry, South Cave, and occur to the north of the Market
Weighton structure.

4. Hydraulic Limestone—occurring below the level of the subsoil as
isolated boulders of greyish-white porcellaneous limestone.
Ostracods: Ektyphocythere triangula, Praeschuleridea subtrigona
subtrigona, Cytheropterina conuca, Micropneumatocythere globosa,
Kirtonella plicata and Progonocythere cristata.

3. Grey clay. Ostracods: Kuirtonella plicata, Asciocythere lacunosa,
Tetracytheridea punctata and Eocytheridea lacunosa . ; :

2. Ironstone-mudstone. Lamellibranchs present throughout. Ostra-
cods: Progonocythere cf. reticulata, Cytheropterina gravis,
Praeschuleridea subtrigona subtrigona, P. ventriosa, Asciocythere
lacunosa and Micropneumatocythere globosa : : :

1. Grey calcareous sandstone with shells. Ostracods : Progonocythere
reticulata, Cytheropterina gravis, Praeschuleridea subtrigona
subtrigona, P. ventriosa, Asciocythere lacunosa, Eocytheridea
lacunosa, E. ? erugata, Paracypris bajociana and Pneumatocythere
bajociana . : 2 ; : : : : seen to

SECTION No.5. Whitwell Oolite, Seamer Lime and Stone Co.
Quarry, Whitwell. Almost complete section through the Whitwell
Oolite and the base of the Upper Limestone (Text-fig. 2).

Upper Limestone

1g. Yellow sand .

18. Yellow, flaggy soadsione ith interbedded cand. ‘ ;

17. Sandy limestone—almost a calcareous sandstone. Ostracod :
Praeschuleridea subtrigona magna .

16. Yellow, false-bedded sand with thin iomeeliae of clay at hacen

15. Clay ‘ 3 : ; 3

14. False-bedded, yellow cand : : : , : :

13. Lens of flaggy sandy limestone. Ostracods: Southcavea reticulata,
Progonocythere cristata and Praeschuleridea subtrigona magna

12. False-bedded sand . :

rr. Alternating bands of white sand and chocolate- brown clay |

to. Alternating bands of yellow sand and brown clay

i=l tal

OoOBRH

HH HH

mM.

ONOwWw NR

H

nm of

>

SOUTH

NORTH

10.RAVENSCAR

8.CLOUGHTON

MILLEPORE

OOLITE

7 YONS NAB

9. HAYBURN VLELLER BECK

Fic. 3. Sections 7-11.

YORKSHIRE OOLITES AND LINCOLNSMIRE LIMESTONE

ft.
Whitwell Oolite
g. Flaggy, sandy oolite. Ostracods: Progonocythere cristata, Prae-
schuleridea subtrigona magna, Eocytheridea lacunosa and Micro-
pneumatocythere globosa : 3 : : , : a O
8. Purple oolitic clay . : : : : ; ©

7. Grey oolitic clay. Osperodls: ENG esiiuis Fle subtrigona magna,
Eocytheridea elongata, E. ? erugata, E. ? astricta, Micropneumato-
cythere globosa and Pneumatocythere bajociana : 2

6. Flaggy oolite, ironstained in parts. Surface showing ripple snark
ings. Ostracods: Eocytheridea faveolata and Praeschuleridea
subtrigona magna . ; : ; ; ; : : ae eto)

5. White-weathering, soft, coarsely oolitic limestone. Ostracods :
Praeschuleridea subtrigona magna, Dolocythere maculosa, Systeno-
cythere exilofasciata, Kirtonella reticulata, Eocytheridea ? erugata,

E. faveolata, E. carinata, Fuhrbergiella (P.) minima, Mazcro-
pneumatocythere globosa, Monoceratina vulsa and Pleurocythere
kirtonensis . : d : e'6

4. Cream weathering, blue shearted polite: Ostracods: Kirtonella
reticulata, Pyraeschuleridea subtrigona magna, Eocytheridea ?
evugata, E. ? astricta, E. carinata, Micropneumatocythere convexa,

M. globosa, Systenocythere exilofasciata, Monoceratina vulsa,
Fuhrbergiella (P.) minima, Paracypris bajociana, Aulacocythere
punctata, Cytherelloidea catenulata, Ektyphocythere triangula and

Dolocythere maculosa . : : : 4 : 3 J, ALE
3. Yellow-brown sandstone. Ostracods: Pvraeschuleridea subtrigona

magna and Micropneumatocythere sti ; ; : ar O
2. Yellow sand with shells at top ; , ; : ee)

1. False-bedded, coarsely oolitic limestone. Ostracods: Progono-
cythere cristata, Praeschuleridea subtrigona magna, Eocytheridea ?
astricta, E. carinata, Fuhrbergiella (P.) minima, Micropneumato-
cythere globosa and Paracypris bajociana : : seen to 6

SECTION No. 6. Upper Limestone and Whitwell Oolite exposed at
Crambeck. The section through the Upper Limestone is almost
complete (Text-fig. 2.).

Upper Limestone

10. Thin, flaggy, oolitic limestone. Ostracods: Eocytheridea ? erugata,
Micropneumatocythere globosa, Praeschuleridea subtrigona magna,
Cytheropterina plana and incertae sedis . : : .about 10-12

g. Yellow sandstone . ; : : F 3 é : Peer i:
8. Soft yellow sand. : ; ; ung

127

10

on

0
2
0

128

YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE

Flaggy sandy oolite. Ostracods: Eocytheridea ? acuta, Muicro-
pneumatocythere globosa, Dolocythere maculosa, Praeschuleridea
subtrigona magna, Kirtonella reticulata and Pneumatocythere
carinata 5

Yellow sandstone and neconeondened sain

Whitwell Oolite

. Hard, crystalline limestone : : ‘ ; ; :
. Sandy oolite. Ostracods: Mucropneumatocythere globosa, Prae-

schuleridea subtrigona magna and Systenocythere exilofasciata

. Sandy limestone
. Soft yellow sand : ‘ $ ; : : : ;
. Coarse creamy oolite. Ostracods: Focytheridea carinata, E. ?

evugata, Micropneumatocythere globosa, Praeschuleridea subtrigona
magna, Dolocythere maculosa, Pneumatocythere carinata and
Paracypris bajociana . 5 : i : : seen to

SECTION No.7. Complete section through the Millepore Oolite and
the overlying Yons Nab Beds exposed along the foreshore at Yons Nab
headland (Text-fig. 3). For the list of macrofossils obtained from this
section see Bate (1959 : 158-9).

20.

IQ.

18.

17
16.
15.
I4.
13.

IZ.

IIe,

Grey shale of the Upper Middle Deltaic Series

Yons Nab Beds

Medium-grained grey sandstone, ironstained in the upper 12 inches
with ironstone nodules and fossil casts; worm burrows in the
lower 34 feet

Grey sandy shale

Ironstone band

Grey sandy shale

Ironstone band

Sandy shale

Argillaceous sandstone ath fhoasicne modtuiles af tye ae.
Southwards this bed grades laterally into a sandy limestone.
Ostracods: Pyrogonocythere cristata, Pnewmatocythere bajociana,
Eocytheridea lacunosa, E. faveolata, E. ? astricta, E. ? acuta,
Muicropneumatocythere convexa, Systenocythere exilofasciata, Dolo-
cythere maculosa and Praeschuleridea subtrigona magna.

Highly fossiliferous grey shale with white shells. Ostraceds-
Progonocythere cristata, Pneumatocythere bajociana, Paracypris
bajociana, Micropneumatocythere globosa, Kirtonella plicata and
Praeschuleridea subtrigona magna .

Sandy, micaceous, grey shale

ft

COO OW -

mn.

ao Ww

onsuaw oun

YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE

to. Yellow micaceous sandstone with plant remains and ripple markings

g. Grey, sandy shale with plant remains

8. Ironstone band

7. Grey shale

6. Mudstone band : : : : : ; :

5. Grey shale. Ostracods common: Cytheropterina plana, Pneuma-
tocythere bajociana, Muicropneumatocythere globosa, Paracypris
bajociana, Systenocythere exilofasciata, Ektyphocythere triangula,
Asciocythere lacunosa, Kirtonella reticulata, Eocytheridea lacunosa.
E. ? erugata, E. ? astricta, E. carinata, Dolocythere maculosa,
Homocytheridea cylindrica, Praeschuleridea ventriosa and P.
subtrigona magna.

Millepore Oolite

4. Coarse, false-bedded, shelly oolite. Ostracods: Pvrogonocythere
cristata, Pneumatocythere carinata & P. bajociana, Micropneuma-
tocythere globosa, Monoceratina cf. vulsa, Eocytheridea faveolata,
E. ? erugata, E. lacunosa, E. ? astricta, Dolocythere maculosa and
Praeschuleridea subtrigona magna . ; : ; i ;

3. Grey, shelly oolite. Ostracods: Mucropneumatocythere globosa,
Paracypris bajociana, Eocytheridea carinata and Praeschuleridea
subtrigona magna. : 5 ‘ é 5 : : ;

2. Fine-grained limestone. Ostracods: Systenocythere exilofasciata,
Eocytheridea lacunosa and Praeschuleridea subtrigona magna

Lower Middle Deltaic Series

1. Yellow sandstone, the top 6 inches containing crinoid ossicles
seen to

SECTION No. 8. Complete section through the Millepore Oolite and
the overlying Yons Nab Beds. Section exposed at Cloughton Wyke,
along the foreshore and at the base of the low cliff (Text-fig. 3).

Upper Middle Deltaic Series

16. Sandy shale . , ; seen to
15. Yellow, false-bedded cdetone with plant nots

Yons Nab Beds

14. Sandy shale, indet. ostracods : : ;

13. Grey-black, ironstained, fossiliferous Sree with oak Lamelli-
branch casts abundant. Ostracods mainly represented by
internal costs: Praeschuleridea subtrigona magna

Soo cam os

Io

129

mM.

QO N OC H

130

OQ

I.

YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE

. Thinly-bedded sandstone with fossil casts and indeterminate

ostracods

. Grey-black, oneramed shalom very Jiosallfioncrs. Lamellibranch

and ostracod casts

. Sandy shale with thin pncerone iberala Tosi casts and plant

remains. Ostracods indeterminate

. Flaggy sandstone with plant remains

Millepore Oolite

. Yellow, false-bedded sandstone with fossil casts at the base .
. Sandy limestone with ironstone nodules in the upper part. Ostra-

cods : Eocytheridea lacunosa, FE. ? erugata, E. ? acuta, E. ? astricta,
E. carinata, Micropneumatocythere globosa, Praeschuleridea sub-
trigona magna, Dolocythere maculosa, Cytheropterina plana,
Kirtonella reticulata, Southcavea reticulata, Ektyphocythere tn-
angula and Paracypris bajociana

. Fossiliferous mudstone. Ostracods: Eocytheridea carinata, E. ?

astricta, Micropneumatocythere convexa, M. globosa, Praeschu-
leridea subtrigona magna, Fuhrbergiella (P.) minima and Kirton-
ella reticulata

. Fossiliferous limestone. Ostracods: Eocytheridea ? erugata, E.

carinata, E. ? astricta, Micropneumatocythere globosa, Praeschu-
leridea subtrigona magna, Dolocythere maculosa, Fuhrbergiella
(P.) mimma, Cytheropterina plana, Kuirtonella reticulata and
?Homocytheridea cylindrica

. Yellow sandstone with fossil casts ; E :
. Fine-grained, fossiliferous, calcareous mudstones Ostracods :

Micropneumatocythere globosa, Eocytheridea ? astricta, Prae-
schuleridea subtrigona magna and ?Monoceratina vulsa

. Dark-grey shale. Ostracods as internal casts

Lower Middle Deltaic Series
Light grey shale.

SECTION No. 9. Hayburn, complete section through the Eller Beck
Bed (Text-fig. 3).

15.

14.
13:
IZ.
IEIE.

Grey shale of the Lower Middle Deltaic Series.

Eller Beck Bed

Massive sandstone with ripple markings along bedding planes
Ironstone band : :

Alternating bands of shale and sandstone 3

Ironstone band

ft.

oO

oo OW

mM.

10

Io

No

It

Io.

iS)

I.

wb uo

YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE

Sandstone

. Ironstone band

Flaggy, micaceous sandstone

. Fossiliferous ironstone band

Sandstone

. Fossiliferous fronstone band

. Dark grey shale : : ;

. Ironstone band from which worm pamors of ine same eonnpastitiod
pass down into the bed below

. Grey shale, with impersistent ironstone at the hase

Lower Deltaic Series

Grey shale. : ; : ; ; ‘ : seen to

SECTION No. 10. Complete section through the Millepore Bed as
exposed at the top of the high cliff at Ravenscar (Text-fig. 3).

Upper Middle Deltaic Series

. Grey-black shale. ; : : ; ; . seen to
. Coaly shale

. Grey shale with plant none scuendine oo ined 13

. Sandy shale with plant roots

. Sandy shale :

. Ironstained shale :

. Yellow sandstone with plant debris :

. Grey-black sandy shale, carbonaceous at base

Millepore Bed

. Massive sandstone with upper 2 ft. 6 ins. silver ganister containing
vertical plant roots. Lower 11 ins. with lamellibranch casts

. Flaggy, false-bedded sandstone

. Yellow sandstone with fossil casts

. Flaggy, yellow, micaceous sandstone

Lower Middle Deltaic Series

. Band of doggers, weathering hollow . : ‘ ; ;
. White, flaggy sandstone . ‘ ; : : f seen to

fi

IS G2 G2 © WS C JF

TF Bao NyNoaosd 2

N OO HW

Il

Now no dou

132 YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE

ft. an:

SECTION No.11. The type section of the Eller Beck Bed, a complete

section exposed in the banks of the Eller Beck (Text-fig. 3).
Eller Beck Bed

6. Massive, buff-yellow sandstone with ripple markings along some
bedding planes. : . 13-14 0
5. Sandy shale and thin, flaggy sandstone showing false bedding Really” Z
4. Dark-grey ironstained shale ‘ Lt eal )
3. Fossiliferous ironstone band 0) 5
2. Dark-grey ironstained shale. ‘ , ; ; ; ya 2
1. Fossiliferous ironstone. i ; : ; 5 seen to I 0)

IV CORRELATION

The only definite dating of the Oolites at the present time is the recorded occurrence
of ammonites of discites age in the Lower Lincolnshire Limestone. Arkell (1933:
214), however, mentions that Tvigonia hemisphaerica has been found in the Kirton
Marl of Kirton Lindsey and that this lamellibranch is to be found in the Cotswolds
only in the Lower Trigonia Grit, a bed also of discites age. By inference this would
suggest that the Lower Lincolnshire Limestone and the base of the Upper Lincoln-
shire Limestone belong to the single ammonite zone of Hyperlioceras discites. Within
the Upper Lincolnshire Limestone the only recorded ammonite is stated to be one of
the “‘ humphriesianus-group’’ by Cross (1875: 121) from the Scunthorpe district.
No precise locality was given. Kent (personal communication) states that he has
been unable to locate this ammonite and the record could be erroneous. The Grey
Limestone Series of Yorkshire which is of humphriesianum age, certainly in part, has
a very different ostracod fauna from that of the Lincolnshire Limestone and hence
no part of the latter can be correlated with beds of that age.

Table 4 indicates a general uniformity throughout the Lincolnshire Limestone
and the equivalent beds in Yorkshire. It also shows the restriction of Cytheropterina
comica and C. gravis to the Lower Lincolnshire Limestone and the Hydraulic Lime-
stone—Basement Beds horizon in Yorkshire, and the replacement of these species by
C. plana in the Millepore and Whitwell Oolites. C. plana has not yet been found in
the Upper Lincolnshire Limestone. Tetracytheridea punctata is restricted to the
Basement Beds and the horizon below the Hydraulic Limestone but has not yet been
recorded from the Lower Lincolnshire Limestone.

Fuhrbergiella (Praefuhrbergiella) minima is restricted to the Millepore and Whitwell
Oolites. Pvaeschuleridea subtrigona subtrigona is now known to be restricted to the
Lincolnshire Limestone and a few beds within the Cave Oolite, and is replaced to the
north of Market Weighton by the geographic subspecies P. subtrigona magna.
Glyptocythere, abundant within the Grey Limestone Series, is virtually absent from the
Lincolnshire Limestone but occurs within the highest beds of the Upper Lincolnshire
Limestone. The species there is similar to G. costata Bate (1965 : 106) but is much
larger. No specific name can yet be given to this ostracod because of indifferent

YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE 133

Stage North of Market Weighton

Upper Deltaic Series
Grey Limestone Series

Upper Middle Deltaic Series

Yons Nab Beds/Upper Limestone

South of Market Weighton North Lincolnshire

BATHONIAN Upper Estuarine Series

Upper Estuarine Series Unconformity - beds of

equivalent age missing

Hibaldstow Oolite

Cave Oolite

Millepore Oolite/Whitwell Oolite Kirton Shale (Acanthothiris

crossi Bed)

BAJOCIAN

Lower Middle Deltaic Series

Hydraulic Limestone/Eller Beck
Bed

Lower Deltaic Series Lower Estuarine Series

Dogger No equivalent,
Dogger missing

TABLE 3. Correlation of Bajocian sediments of Yorkshire and Lincolnshire. It is possible
the highest beds of the Upper Lincolnshire Limestone in South Lincolnshire are of the
same age as the Upper Middle Deltaic Series. The Grey Limestone Series is not, however,
represented in Lincolnshire.

Basement Beds Kirton Cementstone Series

Blue & Silver Beds
Hydraulic Limestone

Lower Estuarine Series

Northampton Sand

preservation. From the evidence of the ostracod faunas it is possible to suggest a
broad correlation of the Yorkshire and Lincolnshire sediments as indicated in Table 3.

Although the ostracod faunas in the Lower Lincolnshire Limestone, Hydraulic
Limestone and associated marls and sandstones, and the Basement Beds are broadly
uniform with beds higher in the succession the common occurrence of C. gravis and
C. comica and their restriction to this horizon supports previous stratigraphical
correlations. A number of ostracod species peculiar to the Basement Beds are
considered to be of local significance only. Correlation of the Hydraulic Limestone
with the Eller Beck Bed farther north must be considered solely on stratigraphical
grounds as there are no ostracods in the latter marine horizon. This northward
facies change from a limestone into a sandstone is to be expected on approaching a
delta and will be dealt with in the next section.

The Millepore and Whitwell Oolites have Fuhrbergiella (P.) minima as a common
ostracod but otherwise the fauna is basically uniform with that in the Upper and
Lower Lincolnshire Limestones. Cytherelloidea eastfieldensis, found in the Lower
Lincolnshire Limestone and in the Basement Beds is neither represented in the
Millepore/Whitwell Oolites nor in the Upper Lincolnshire Limestone. The Yorkshire
sediments also appear to have a dominant Eocytheridea fauna which may be geo-
graphically controlled. One species E. ?acuta is entirely restricted to Yorkshire.
The Yons Nab Beds and Upper Limestone have the same ostracods as those appearing
in the underlying Oolites and must be part of the same marine transgression.

By inference the major part of the Lincolnshire Limestone probably belongs to the

134 YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE

Hyperlioceras discites Zone, although the highest beds of the Upper Lincolnshire
Limestone are possibly slightly younger than this. By correlation the two marine
horizons in Yorkshire must also be of discites age—this is particularly true for the
Hydraulic Limestone/Eller Beck Bed horizon, the marine sediments below, the
Basement Beds above and the Lower Middle Deltaic Series. It is less certain how
much (if not all) of the Cave, Whitwell and Millepore Oolites may be similarly dated.
Certainly there is no evidence in the ostracod fauna to suggest any considerable
range in time.

In conclusion, the ostracods, although not giving a very fine correlation between
the Yorkshire and Lincolnshire Oolites, substantiate previous stratigraphical
correlations. The reason why the ostracods are so uniform throughout the succession
is considered to be due to the fact that the Yorkshire and Lincolnshire sediments
were deposited during a very short period of time, too short for much evolutionary
change to take place. The only apparent evolutionary change appears in the genus
Cytheropterina. Perhaps one of the more important aspects of the interpretation of
the ostracod faunas is to place the topmost beds of the Upper Lincolnshire Limestone
below the Grey Limestone Series. Certainly it is doubtful whether the Lincolnshire
Limestone as a whole ranges any higher than the Sonninia sowerbyi Zone and
probably is contained completely within that zone.

APPENDIX

A number of minor sections have been examined which do not appear in the
“ Stratigraphical Sections ’’ because they are either incomplete or not essential to
the stratigraphy of the area. However, a number of ostracod species occur in these
minor exposures but not at the same horizon elsewhere. The distribution table
(Table 4) would, therefore, be erroneous if they were omitted. A complete faunal
list for each of the localities mentioned below is not given, and only those records
which extend the stratigraphical range of a species are listed.

1. Upper Limestone—Stonecliff Wood, close to section No. 6, map reference
SE/736675 : Asctocythere acununata, Eocytheridea ? astricta, Eocytheridea faveolata,
Systenocythere exilofasciata and Systenocythere ? sp.

2. Millepore Oolite—Osgodby Nab, map reference TA/065855: Avulacocythere
punctata, Eocytheridea ? acuta and Eocytheridea reticulata.

3. Millepore Oolite—Cloughton, map reference TA/021958 : Cytheropterina plana.

4. Whitwell Oolite—Bulmer, map reference SE/704678 : Southcavea grandis.

5. Kirton Cementstone Series—Kirton Lindsey, map reference SE/g42011 :
Pleurocythere kirtonensis.

V PALAEOGEOGRAPHY
Examination of the ostracod faunas has shown the Lincolnshire Limestone to be of
equivalent age to the Yorkshire Oolites and the Hydraulic Limestone/Eller Beck Bed
horizon below. Lithologies in some instances remain constant though eventually all
become strongly arenaceous and finally completely so when traced northwards.
The environment of deposition of the Lincolnshire Limestone may be treated as a
whole. In the Lincoln area the Upper Lincolnshire Limestone is strongly bedded,

OSTRACOD
SEECIES

Glyptocythere sp.

Grey Limesto
Series

polita scitu

Zone Zon

Glyptocythere costata

Basement Beds

Kirton Cementstone
Series

Blue and Silver Beds

Glyptocythere polita

Glyptocythere scitula

Malzia bicarinata

Malzia unicarinata

Monoceratina scarboroughensis
Caytonidea faveolata

peice favenloraie eS

“Cloughtonella rugosa

Fuhrbergiella (Pracfuhrbergiella) horrida
jorrida

Progonocythere acuminata
vr

0; jonocythere _yonsnabensis
Pleurocythere sp.

Paracytheridea ? caytonensis

Praeschuleridea subtrigona intermedia

Ljubimovella piriformis

Mesocytheridea howardianensis
Southcavea microcellulosa

Systenocythere ovata

Eocytheropteron f sp '.

Vernoniella Bajociana

Vernoniella ? caytonensis

FPISPIRIRRIRIRISIIL BIE

Paracypris bajociana
Cytherella fullonica

Cytherelloidea catenulata

Platella jurassica
Bairdia hilda
Monoceratina vulsa

Monoceratina sp. cf. M. scrobiculata

Progonocythere cristata
Aulacocythere punctata
Aulacocythere reticulata

Fuhrbergiella (Praefuhrbergiella) minima
Fuhrbergiella (Pracfuhrbergiella) arens
Micropneumatocythere convexa
Micropneumatocythere globosa

Pneumatocythere bajociana

+

Pneumatocythere carinata

Pleurocythere Rirtonensis
Pleurocythere nodosa

Pleurocythere sp.

Dolocythere maculosa

Asciocythere acuminata
Asciocythere Tacunosa

Eocytheridea ? acuta

Focytheridea ? astricta
ocytheridea Carinata

FS

ic

Eocytheridea elongata
Eocytheridea ? erugata

Eocytheridea faveolata

peemitirideaTocunosa

Focytheridea reticulata
Praeschuleridea subtrigona subtrigona

Praeschuleridea subtrigona magna

Praeschuleridea ventriosa
Cytheropterina comica

Cytheropterina gravis

Cytheropterina plana

Rimeelle slices plicata
Kirtonella reticulata
Ektyphocythere triangula

Southcavea reticulata

Southcavea bajociana

Southcavea grandis
Systenocythere exilofasciata
Sys hi

ystenocyt ere Sp.
amptocythere lincolnensis

Gytheromorpha (?) greetwellensis

Cytherelloidea eastfieldensis
Frogonoc) there reticulata

Homocytheridea cylindrica

Tetracytheridea punctata

Paraschuleridea ornata

Paraschuleridea sp.

Grey Limestone Upper

Series Lincolnshire | Yons Nab | Upper Limestone | Millepore Oolit rae ae a
TRACOD polita sctula | Limestone | Beds Pore Oolite | Whitwell olite || Cave Oolite || Kirton Shale

Zone Zone South

Hydraulic Limestone

and Basem
Recor tted eds ent Beds | Kirton Cementstone | Blue and Silver Beds

:
IE

Traairaing
Tepnndes [ovesks

roel
eure ERergelay Perrida
Fraga IOS

Pa ronal

iene
Teateidea 2 captoOO
Frieden mburiqooa interme
[parle inform

eseterides Reverses

Gahanna

| ° 5 5 : a
= = : : i =
Fenwana spies Mi xrobielexa ° A : - : |
s o . | o . =;
5 5 = S
i | O ° ° = 5
. 7 *

Tocmanyiore

Fremaraythere
Texaythee fi

Fae = a
Frayer 5p ===
Saran ° . I ° . o ni i
° == |
z a ° ° e =I .
. iL = =]
= 5
2 : - =
2 . = =
: = is = =
rs a °
z ° °
de =
= _—— ° ° ° 5
y ° 5 =
fox s a - ee
= . .
= . 5 == IE
z A * °
Saks E ° ° . .
Sexhenes bijacioa a ;
Sako grands :

°
.
elelelele

pt

Gmesthecriljeina

reenter ap,
jc mumamncas
(Gleneerpi 7 grea
edad rea
Trager raters
ee indica, 5
ae eis porare E
Tawar oe IF =e x

fraser des 5

el

Taste 4. Distribution Chart of Ostracoda within the Bajocian of N.E. England.

YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE 135

often coarsely oolitic. The base of this division, the A. crossi Bed, when traced only
a few miles to the north becomes a clay-marl, rich in ostracods. The Lower Lincoln-
shire Limestone on the other hand is fairly stable in this part of its development as a
chalky, only poorly oolitic rock with chalky marl bands in the upper part (Cement-
stone Series) but becomes more massively-bedded and in part coarsely oolitic towards
the base (Blue and Silver Beds). Deposition appears to have taken place in warm
waters rich in calcium carbonate probably similar to the type of deposition prevalent
at the present time in the Bahamas. Current action is not much in evidence at some
horizons, particularly during the deposition of the Kirton Shale and the marl bands
of the Cementstone Series, and here it is not unusual to find all growth stages in the
development of the ostracod retained in the sediment. Current action almost
invariably winnows out the smaller instars. Horizons exhibiting oolith formation
have experienced some current action, and rolled gastropods, often coated with
calcium carbonate, are not uncommon.

All the lithological units of the Lincolnshire Limestone point to deposition in a
warm, shallow sea in which chemical precipitation of calcium carbonate was probably
high and current action, when present, resulted in the formation of ooliths, again very
much like the present-day Bahamas.

To the north of Lincolnshire a fairly large delta, situated in north-east Yorkshire,
was discharging into the sea, but had little effect upon the Lincolnshire Limestone as
such. This was almost certainly due to the presence in the region of Market Weighton
of a stable land barrier (occasionally covered by shallow water) which effectively cut
off the Yorkshire Basin from marine deposition to the south. Correlation of the
Lincolnshire Limestone with the marine horizons in Yorkshire suggests that if the
sea did not transgress over the Market Weighton stable area, then it must have
passed around it to the east, the main body of the sea being situated in a similar
position to the present-day North Sea.

The Lincolnshire Limestone is represented by two marine horizons in Yorkshire,
and it is proposed to deal with these separately.

Hydraulic Limestone/Eller Beck Bed

From Lincoln northwards there is an increase in the marl facies within the Lower
Lincolnshire Limestone so that to the north of the Humber marked changes occur.
Here, the Cave Oolite is still an oolitic limestone but the beds beneath are pre-
dominantly marls (the Basement Beds) with some rubbly limestone and a thin
(2 ft. 6 ins.) grey, porcellaneous limestone termed the Hydraulic Limestone. The
Lower Estuarine Series beneath may be marine but is no longer exposed. The
Hydraulic Limestone and associated marine beds are the northern equivalent of the
Lower Lincolnshire Limestone as correlated on the ostracod faunas. The ostracod
population here is complete in all stages from juvenile instars to adult carapaces,
indicative that deposition during Basement Beds time proceeded in relatively quiet
waters with little or no current action.

The Hydraulic Limestone does not appear to have passed over the land barrier
north of Market Weighton although this area might well have been submerged under

130 YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE

shallow-water, almost littoral conditions at that time. Sandy beds below the
Hydraulic Limestone at Acklam to the north provide additional evidence that land
was close by.

Marine beds continue to underlie the limestone (section No. 4) until about 2-4
miles north of Whitwell where deltaic sediments of the Lower Deltaic Series appear.
This is the southernmost extension of the delta front in the west at this time. Farther
north the Hydraulic Limestone (generally 2 ft. 6 ins.—3 ft. 0 ins. in thickness) trans-
gresses over an increased thickness of deltaic sediments. As previously mentioned
the limestone is not exposed in the east but has been identified in the Fordon borehole
as 4 ft. of cementstone. In the west the Hydraulic Limestone can be traced as far
north as Skipton Hill after which it is replaced by the Eller Beck Bed facies of ripple

EASTERN LAND
AREA

: 22’ 5° “SEINE

oO Osmer net ly S
: . Shicllow Water ‘

5 . "Eller Beck Bed Facies TE ‘

Hydraulic Limestone Facies
Pickering

Malton

Whitwell, —— —
— — “limestone 34"
marine shale below
sandy facies below
Raa limestone at Acklam

GZ
Y Bridlington

Area of probable ~ \
shallow water

PENNINE LAND A

AREA

fa
4

«Market Weighton

7

Z

Hydraulic limestone not
separable from the
Lincolnshire Limestone

ie} 5 10 15 20 miles

Fic. 4. Palaeogeographic map of the Hydraulic Limestone/Eller
Beck Bed marine transgression,

YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE 137

marked sandstone and fossiliferous ironstone bands. This marked change in facies
indicates the closeness of the delta front and the shallowness of the water is further
indicated by the presence of ripple markings and worm burrows. The considerable
increase in the thickness of this marine bed from the Hydraulic Limestone is due to
the vast quantity of marine deposited, arenaceous material brought into the area by
the northern delta. The absence of a marine fauna in the sandstone is probably due
to rapid deposition, always an unsuitable environment for benthonic organisms.
Text-fig. 4 indicates the probable palaeogeographic conditions under which the
Hydraulic Limestone and the Eller Beck Bed were deposited, open sea being to the
south-east.

Deltaic Series

In discussing the marine horizons which transgressed the deltaic sediments it is
impossible to get a clear idea as to palaeogeography without also considering the
deltaic sediments themselves. I do not propose to go into details of the subdivision
of the Series but rather to give general views on their deposition and source.

Prior to 1924 this vast thickness of false bedded sandstones, shales with thin coal
seams and important plant beds had been considered estuarine in origin, being
so-called by Fox-Strangways (1880). Kendall & Wroot (1924 : 309) were the first
to recognize their true deltaic origin, later confirmed by Black (1928, 1929, 1934).

The features which confirm the deltaic nature of these beds are: false bedding
(here mainly topset), washouts (especially common in Upper Deltaic Series in Cayton
Bay and postulated at Osgodby Nab in the Middle Deltaic Series), beds of Equisetites
and coal seams. The burnt bed in the Lower Deltaic Series represents an old forest
fire. The Equisetites grew in swamps behind the advancing delta front whilst the
more substantial trees grew further back.

Deposition of these deltaic sediments took place in a gently subsiding basin with a
Pennine land mass to the west, a partial land barrier to the south at Market Weighton
and possibly land to the north and north-east. Open sea lay to the south-east.
There is some difference of opinion as to whether land was present to the north and
whether the source of the deltaic sediments lay in that direction or to the east.

Black (1934 : 279) considered the Yorkshire basin to be in a south-western angle
of a gulf opening to the north with derivation of sediment from the east. He did not
believe that the Pennine land supplied much material, a conclusion with which I
agree. Black also stated that the emergent land in the Market Weighton area
formed a barrier and that the mineral suites differ so remarkably to the north and
south that they must have been derived from different sources. Smithson (1942)
came to a similar conclusion with regard to the mineral suite but did not follow
Black in deriving the sediments from the east and considered the main source to have
been from the north-north-west. This was confirmed by a grain size study, especially
of zircon, which gave this same constant direction of incoming material for both
deltaic and marine sediments. Sorby (im Fox-Strangways, 1892 : 391) was also of
the opinion that the drift-bedding and ripple marks were produced by north-north-
westerly currents. The findings of Sorby and Smithson are accepted here.

138 YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE

Correlation of the Lincolnshire Limestone with the marine horizons in Yorkshire
indicates that the marine transgression came from the south-east and skirted around
the Market Weighton barrier. The marine beds deposited during the transgression
die out northwards against the delta, in which direction land obviously lay. Smithson
also considered land to be present to the east. The increase in arenaceous sediments
to the north-east in particular would suggest that land certainly lay in that direction,
as indicated in the palaeogeographic interpretations of the Hydraulic Limestone and
of the Millepore and Whitwell Oolites. Derivation of the deltaic sediments in
Yorkshire is thus considered to have been mainly from the north-west with sub-
sidiary material from the north-east. The Pennine land to the west and the Market
Weighton land to the south represent the coastline of the bay. The land-derived
material in the Lincolnshire basin is not dominant in the limestone facies and was
possibly derived from the Pennine land to the west and/or the London—Ardennes
island to the south.

Millepore and Whitwell Oolites

For a time the Yorkshire basin was filled by deltaic sediments of the Middle
Deltaic Series until further subsidence produced another marine transgression, much
more prominent than that which formed the thin Hydraulic Limestone and the
poorly-fossiliferous Eller Beck Bed. The Whitwell Oolite in the west is perhaps the
better developed and, apart from the rather sandy beds overlying it, does not
contain a large proportion of arenaceous material until traced northwards. Although
a Pennine land mass must have been present to the west it did not supply much
material compared with conditions farther north. In the west and south-west the
present-day outcrop represents a shallow water deposit close to a land mass which
did not have large rivers bringing in material, the sand being more of a shore-line
deposit. Shallow-water conditions may have covered the land barrier in the Market
Weighton area though there was little connection between the Yorkshire sediments
and those just north of the Humber. The reasons for this are based to some extent
on faunal evidence. Pvraeschuleridea subtrigona magna was virtually restricted to the
Yorkshire basin and is regarded as a geographic subspecies ; only a few specimens
have been recorded from the Cave Oolite and none from the Lincolnshire Limestone.
The marine Bairdia hilda does not occur north of Market Weighton where the effect
of the delta might have lowered the salinity to some extent, although Pavacypris
bajociana, also regarded as a good marine ostracod, is common. Within the York-
shire basin there is also a preponderance of species of Eocytheridea, some of which
have not been recorded from Lincolnshire. This may also be due to ecological
conditions in the vicinity of the delta.

Farther east, away from the western land mass, the Millepore Oolite as
exposed in Cayton Bay exhibits strong false-bedding and current action must have
been fairly strong in that region. When traced northwards both the Whitwell and
the Millepore Oolites become more arenaceous until they are finally indistinguishable
from a typical deltaic sandstone. The delta front has been reached. The uppermost
beds of the Whitwell Oolite are known as the Upper Limestone and those of the

YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE 139

@ Middiesborougn

EASTERN LAND AREA

Osmotherly
Deltaic area not covered

by Millepore Sea

~NGim,;
Sale "Nt
“SOF Millepore

Gea
‘ons Na? Fordon |
Clear water

Limit of Y

strong currents

eMalton
Whitwell
25/9"
30°11
fi
nee 7-9 of yellow sand
~ + above Whitwell Oolite

lf

Bridlington

PENNINE LAND

AREA
1. area of probable upwayp
GP “-and shallow water
York Schaar
i aMarket Weighton
N \ ‘: “sand interbedded with
> and overlying the

"Cave Oolite

Lincolnshire
Limestone
Facies

(0) 5 10 15 20miles
LN ed

Fic. 5. Palaeogeographic map of the Millepore/Whitwell Oolite marine transgression.

Millepore Oolite as the Yons Nab Beds. The former is a siliceous limestone whilst
the latter tends to be a shale/sandstone sequence. These higher beds represent the
final stages in the marine transgression and do not extend as far north as the main
marine transgression represented by the oolites beneath. The Yons Nab Beds are
close to the delta front throughout their exposed development. The palaeo-
geographic interpretation of this marine episode illustrated in Text-fig. 5 is closely

comparable to Smithson’s (1942, fig. 15).

Conclusions
The land barrier in the vicinity of Market Weighton represented a stable area
between subsiding basins to the south and the north.

140 YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE

The southern basin was infilled with calcareous sediments (the Lincolnshire
Limestone) deposited in lime rich waters. The Pennine land to the west and the
London-Ardennes Island to the south did not supply a large quantity of terrigenous
material.

The northern basin, on the other hand, received a considerable amount of terri-
genous material from its northern boundaries in the form of deltaic sediments. The
Yorkshire delta was quite substantial and had a considerable effect upon sedimenta-
tion although this was limited to Yorkshire. Open sea lay to the south-east and it
was from this direction that the sea entered to transgress the Yorkshire delta on two
separate occasions during the continuous deposition of the Lincolnshire Limestone
farther south.

The first transgression gave rise to the Hydraulic Limestone and the Eller Beck
Bed and took place in Lower Lincolnshire Limestone times. A period of deltaic
sedimentation separates this transgression from the second marine incursion during
which the Whitwell Oolite and associated Upper Limestone and the Millepore Oolite
and Yons Nab Beds were deposited. This second transgression occurred in Upper
Lincolnshire Limestone times. Both incursions came from the south-east.

During Middle Jurassic times a third marine transgression covered the Yorkshire
delta and was responsible for the Grey Limestone Series (Bate 1965). The sea did
not enter from the south-east, but came directly from the east, indications being that
land lay to the south, probably in Lincolnshire where no marine deposition would
have taken place at that time.

VI REFERENCES

ARKELL, W. J. 1933. The Jurassic System in Great Britain. xii + 681 pp. Oxford.

Bate, R.H. 1959. The Yons Nab Beds of the Middle Jurassic of the Yorkshire Coast. Pvoc.

Yorks. Geol. Soc., Leeds, 32 : 153-164, pl. 3.

1963. Middle Jurassic Ostracoda from North Lincolnshire. Bull. By. Mus. nat. Hist.

(Geol.), London, 8 : 173—219, pls. 1-15.

1963a. Middle Jurassic Ostracoda from South Yorkshire. Bull. Br. Mus. nat. Hist.

(Geol.), London, 9 : 19-46, pls. 1-13.

1964. Middle Jurassic Ostracoda from the Millepore Series, Yorkshire. Bull. By. Mus.

nat. Hist. (Geol.), London, 10: 1-34, pls. 1-14.

1965. Middle Jurassic Ostracoda from the Grey Limestone Series, Yorkshire. Bull. By.

Mus. nat. Hist. (Geol.), London, 11 : 73-134, pls. 1-21.

Brack, M. 1928. ‘“‘ Washouts”’ in the Estuarine Series of Yorkshire. Geol. Mag. Lond.,

65 : 301-307.

1929. Drifted Plant Beds of the Upper Estuarine Series of Yorkshire. Quart. J. Geol.

Soc. Lond., 85 : 389-437.

1934. Sedimentation of the Aalenian rocks of Yorkshire. Pyvoc. Yorks. Geol. Soc., Leeds,

22 : 265-279.

Cross, J. E. 1875. The Geology of N.W. Lincolnshire. Quart. J. Geol. Soc. Lond., 31 : 115-130.

Ievans, W. D. 1952. The furrassic Rocks of the Lincoln District. Pvoc. Geol. Ass. Lond.,
63 : 316-335.

Fatcon, N. L. & Kent, P. E. 1960. Geological Results of Petroleum Exploration in Britain
1945-1957. Mem. Geol. Soc. Lond., 2 : 1-56.

l’ox-STRANGWAYS, C. 1880. The Geology of the Oolitic and Cretaceous Rocks south of
Scarborough. Mem. Geol. Surv. Lond., 41 pp.

-—— 1892. The Jurassic Rocks of Britain, 1. Yorkshire. Mem. Geol. Surv. Lond., 551 pp.

YORKSHIRE OOLITES AND LINCOLNSHIRE LIMESTONE 141

Hupreston, W.H. 1873. The Yorkshire Oolites. Pt.1. Pvoc. Geol. Ass. Lond., 3 : 283-333.

KENDALL, P. F. 1905. Sub-report on the concealed portion of the Coalfield of Yorkshire,
Derbyshire and Nottinghamshire—Final report of the Royal Commission on Coal Supplies.
Pt. 9. Appendix 3 : 18-35.

—— & Wroot, H. E. 1924. Geology of Yorkshive,1. xxii + 660 pp. Vienna.

Kent, P.E. 1955. The Market Weighton Structure. Pvoc. Yorks. Geol. Soc., Leeds, 30: 197—
22T7e

Murr-Woop, H. M. 1952. Some Jurassic Brachiopoda from the Lincolnshire Limestone and
Upper Estuarine Series of Rutland and Lincolnshire. Pyvoc. Geol. Ass. Lond., 63 : 113-142.

NEALE, J. W. 1958. In Boer G. de, Neale, J. W. & Penny, L. F. A guide to the geology of
the area between Market Weighton and the Humber. Pyvoc. Yorks. Geol. Soc., Leeds,
31 : 157-209.

SmiTHson, F. 1942. The Middle Jurassic Rocks of Yorkshire: A Petrological and Palaeo-
geographical Study. Quart. J. Geol. Soc. Lond., 98 : 27-59.

SWINNERTON, H. H. & Kent, P.E. 1949. The Geology of Lincolnshire. Lincs. Nats. Union,
Lincoln, 1. 126 pp.

SYLVESTER-BRADLEY, P. C. 1953. A Stratigraphical Guide to the Fossil Localities of the
Scarborough District. In The Natural History of the Scarborough District, 1: 19-48.
Scarborough.

Wricut, T. 1860. On the subdivisions of the Inferior Oolite in the South of England compared
with equivalent beds of that formation on the Yorkshire Coast. Quart. J. Geol. Soc. Lond.,
16 : 1-48.

"Ae
uke Meats

i
a.

4 a4
’
P
i
: ’
¥ a
+
ie

“pRINTBD IN GREAT BRI

BY ADLARD & SON LI
BARTHOLOMEW PRESS, |

Qe) Vee
=

site MAY pe

S
OA

aay wey

: FURTHER NOTES ON

: PALAEONISCOID- FISHES WITH A

_> CLASSIFICATION OF THE
CHONDROSTEI

-B. G. GARDINER

a BULLETIN OF
. HE BRITISH MUSEUM (NATURAL HISTORY)
‘ Vol. 14 No. 5 _
ee 1967

PURTHER NOTES ON PALAEONISCOID FISHES

WITH A CLASSIFICATION OF THE
CHONDROSTEI

BY

BRIAN GEORGE GARDINER, aw

(Queen Elizabeth College, University of London) )\

bar
J

Pp. 143-206 ; 3 Plates; 24 Text-figures

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)
GEOLOGY Vol. 14 No. 5

LONDON: 1967

a

\3
NAL 1

THE BULLETIN OF THE BRITISH MUSEUM
(NATURAL HISTORY), instituted im 1949, 1S
issued in five series corresponding to the Departments
of the Museum, and an Historical sertes.

Parts will appear at irregular intervals as they become
veady. Volumes will contain about three or four
hundred pages, and will not necessarily be completed
within one calendar year.

In 1965 a separate supplementary series of longer
papers was instituted, numbered serially for each
Department.

This paper is Vol. 14 No. 5 of the Geological
(Palaeontological) series. The abbreviated titles of
periodicals cited follow those of the World List of
Scientific Periodicals.

World List abbreviation
Bull. Br. Mus. nat. Hist. (Geol.).

© Trustees of the British Museum (Natural History) 1967

TRUSTEES OF
THE BRITISH MUSEUM (NATURAL HISTORY)

Issued 23 May, 1967 Price fiers:

PORTH ER INOTES ON PALAEONISCOID FISHES
WE a CLASSIFICATION” OF WHE
CHONDROSTEI

By B. G. GARDINER

CONTENTS
Page
1. INTRODUCTION é é j : : : é : 6 140
II. SYSTEMATIC DESCRIPTIONS c 6 : g : 6 p 146
Order PALAEONISCIFORMES
Family GONATODIDAE Nov. é 5 5 : . 146
Genus Gonatodus Traquair . : : : : 146
Gonatodus punctatus (Agassiz) . : : : 147
Genus Pseudogonatodus nov. : : : : 153
Pseudogonatodus parvidens (Traquair) : : 153
Pseudogonatodus macrolepis (Traquair) : j 157
Genus Drydenius Traquair . é : : : 159
Drydenius insignis Traquair : : : ; 159
Drydenius molyneuxi (Traquair) : : : 161
Family TRISSOLEPIDIDAE Fri€. 0 : : : 163
Genus Sphaevolepis Frié : : ; : ; 164
Sphaerolepis kounoviensis Fri€ . : ; : 164.
Genus Sceletophorus Fri€ . : ; : : 168
Sceletophorus biserialis Frié 3 : : ‘ 169
Family PHANERORHYNCHIDAE Stensio . : é é 172
Genus Phanerorhynchus Gill . : . : ‘ 172
Phanerorhynchus aymatus Gill. : : : 172
Family OSORIOICHTHYIDAE nov. . 4 é é 175
Family GYROLEPIDOTIDAE nov. . : : : 175
Family AMPHICENTRIDAE Moy- nes : : : 176
Family ATHERSTONIIDAE nov. . : : : : 176
Family LAWNIIDAE nov. . 3 : 5 : j 176
Family COSMOLEPIDIDAE nov. . é 3 : : 177
Family BRACHYDEGMIDAE nov. . : ° ; : 177
Family BOREOSOMIDAE nov. f : 5 : : 177
Order PELTOPLEURIFORMES
Family HABROICHTHYIDAE nov. . : : 0 0 178
III. Discussion . : : 2 é é 178
IV. CLASSIFICATION OF THE Cuompnoenea Fi , A 3 : 196
V. SUMMARY é 5 : 6 : 6 , ; ; 201
Vi. ACKNOWLEDGMENTS : ¢ : . 5 . é 6 202
VII. L&rTTERING USED IN TEXT FIGURES . : ; : 3 : 202
VIII. REFERENCES : : ; 4 6 : ‘ : ‘ 202
SHAN OSHS

A redescription is given of the type species of the genera Gonatodus Traquair, Drydenius
Traquair, Sphaerolepis Frié, Sceletophorus Frié and Phanerorhynchus Gill. Pseudogonatodus
gen. nov. is proposed for two species previously placed in Gonatodus. Eight new palaeoniscoid
families are erected, the Gonatodidae, Osorioichthyidae, Gyrolepidotidae, Atherstoniidae,
Lawniidae, Cosmolepididae, Brachydegmidae and Boreosomidae. One other new chondrostean
family is proposed, the Habroichthyidae which belongs to the order Peltopleuriformes. The
evolution of the Chondrostei is discussed and a classification of the Subclass given.

GEOL. 14, 5. 16

146 PALAEONISCOID FISHES AND THE CHONDROSTEI
Te eUN ROD UiC der OA)

TRAQUAIR (1877-1914), in his monograph on “ The Ganoid fishes of the British
Carboniferous formations ’’, described the majority of the palaeoniscoids occurring
in our Carboniferous strata. Subsequently Moy-Thomas and Dyne (1938) re-
described and supplemented that portion of the fauna which occurred in the Lower
Carboniferous rocks of Glencartholm. In an earlier paper (Gardiner 1963) I began
the task of revising all the genera described by Traquair (1877-1914) except those
adequately dealt with by Moy-Thomas & Dyne (1938) and this paper is intended to
be a continuation of that work. In addition to genera described by Traquair I have
redefined the Upper Carboniferous genus Phanerorhynchus Gill and redescribed
two genera from the Upper Carboniferous of Czechoslovakia, Sphaerolepis Frié and
Sceletophorus Fri¢.

Having examined almost all the palaeoniscoids, I have attempted a classification
of the Palaeonisciformes, and, treated the much bigger problem of the evolution of
the Chondrostei and its classification.

Later I hope to redescribe Cryphiolepis Traquair, Acrolepis Agassiz, Myriolepis
Egerton and Styvacopterus Traquair and then, in order to complete the survey, all
the many species which were dealt with by Traquair will be re-examined in the light
of my emended generic diagnoses.

Ie SY SDEMARECED E SiGinemhO Nis
Order PALAEONISCIFORMES

Family GONATODIDAE nov.

Dracnosis. Trunk deeply fusiform ; dorsal fin situated behind the middle of the
back, more posterior in position than in Elonichthys Giebel. The remaining diagnosis
as for the genus Gonatodus.

REMARKS. From my description of Gonatodus it is clear that this genus does not
belong in the Family Elonichthyidae where both Romer (1945 : 579) and Obruchev
(1964 : 352) placed it; further it does not fit into any other previously described
family. A new family, Gonatodidae, is therefore erected to include this genus and
the related genera Drydenius Traquair and Pseudogonatodus nov. The Gonatodidae
later gave rise to both the Commentryidae Gardiner (1963 : 290) and the Amblyp-
teridae Romer (Gardiner 1963 : 290).

Genus GONATODUS Traquair, 1877
1835 Amblypterus Agassiz (partim) 2, 1 : 109.

DraAGnosis (emended). Trunk deeply fusiform ; dorsal fin situated behind middle
of back, partly in advance of, partly opposing, anal, both being large, triangular and
approximately the same size. Dorsal contour arched in advance of dorsal fin.
Paired fins large, pelvic pair midway between pectorals and anal. All fins with
minute fulcra anteriorly and with rays closely articulated, so as to impart scale-like
appearance to individual joints; all rays distally bifurcated. Skull with suspen-
sorium somewhat inclined, not so near vertical as in Amblypterus, moderately over-

PALAEONISCOID FISHES AND THE CHONDROSTEI 147

hanging rostrum and relatively stout sclerotic ring. Opercular more than twice as
deep as subopercular; suborbital series and dermohyal present. Branchiostegal
rays numerous, skull roofing bones coarsely striated. Teeth closely set, of moderate
size and in one series. Scales large with distinct peg and socket articulation, and
ornamented with fine, oblique striae.

TYPE SPECIES. Amblypterus punctatus Agassiz.

REMARKS. Agassiz (1835, 2, 1 : 10g) in describing [the new species] Amblypterus
punctatus used three specimens (R.S.M. 1878.18.4, R.S.M. 1878.18.6 and one in the
Oxford University Museum) all of which are from the Calciferous Sandstone Series
(Lower Carboniferous) of Wardie. Traquair (1877 : 265; 1877b: 555; 1877c: 60)
realized that the species Amblypterus punctatus Agassiz was founded not only on two
distinct species but that each of the two species belonged to different genera. For
one of Agassiz’s original syntypes (R.S.M. 1878.18.4) Traquair (1877) retained the
specific name “‘ punctatus ”’ and used it to form the type of the new genus Gonatodus.
To the other two syntypes of Agassiz (1835, 2, 1, pl. 4c, figs. 3, 5) Traquair gave the
new specific name Elonichthys intermedius. Traquair (1901:67) decided that
Elonichthys intermedius was merely a variation of Elonichthys robisoni (Hibbert),
and called it Elonichthys robisoni1 Hibbert var. intermedius Traquair. From an
examination of the type material it is clear that Traquair was justified in separating
the syntypes of Agassiz’s Amblypterus punctatus into the two distinct species,
Gonatodus punctatus (Agassiz) and Elonichthys robisoni (Hibbert).

The genus contains but the type species, the other two species referred to it by
Traquair (1907), Gonatodus parvidens Traquair and Gonatodus macrolepis Traquair,
are placed in a new genus, Pseudogonatodus.

Two other species that have in the past been included in this genus, Gonatodus
brainerdi (Newberry 1873: 346; 1890:125) and Gonatodus ? totlliezi (Koninck
1878 : 11), clearly belong elsewhere as already indicated by Traquair (1907 : 93).
Gonatodus brainerdi from the Berea Grit of Ohio (Lower Carboniferous) is from its
large size possibly a Nematoptychius ; while Gonatodus ? totlliexi from the Lower
Carboniferous of Viesville, Belgium, has large fulcra and seems closer to Canobius
than to Gonatodus.

Gonatodus punctatus (Agassiz)

(Text-figs. 1-4)
1835 Amblypterus punctatus Agassiz, 2, I : 109, pl. 4c, fig. 4 (non figs. 3, 5-8).
1872 <Amblypterus anconoaechmodus Walker : 119, pl. 1.
1877 Gonatodus punctatus (Agassiz) Traquair : 265.
1877b Gonatodus punctatus (Agassiz) ; Traquair : 555.
1877c¢ Gonatodus punctatus (Agassiz) ; Traquair : 16, 60, pl. 2, figs. 4, 5
1882 Gonatodus punctatus (Agassiz) ; Traquair : 546.
1890 Gonatodus punctatus (Agassiz) ; Traquair : 391.
1891 Gonatodus punctatus (Agassiz) ; Woodward : 434.
1903 Gonatodus punctatus (Agassiz) ; Traquair : 690, 700, 701.
1907 Gonatodus punctatus (Agassiz) ; Traquair : 93, pl. 19, text-figs. 2, 3a.
1907b Gonatodus punctatus (Agassiz) ; Traquair : 106, 114, 115, pl. 2, figs. 1, 2.
1925 Gonatodus punctatus (Agassiz) ; Watson : 859, text-fig. 27.
1954 Gonatodus punctatus (Agassiz) ; Waterston : 58.

148 PALAEONISCOID FISHES AND THE CHONDROSTEI

D1aGnosis (emended). Fishes not exceeding 18-5 cm. in total length, body deeply
fusiform, length of head contained slightly more than three and a half times and
depth of body about two and three-quarter times in total body length (measured to
bifurcation of caudal fin). Head short with bluntly rounded snout. Skull with two
dermohyals, and teeth on jaws arranged in single closely set row. Scales of moderate
size deeper than broad on flank and ornamented with distinct concentric ridges of
enamel imparting delicate serration to posterior margin. Many ridges on dorso-
posterior region of scales short and terminating in points before posterior margin is
reached (Text-fig. 3).

HoLtotyre. R.S.M. 1878.18.4, head and anterior half of fish wanting fins, from
the Calciferous Sandstone Series (Lower Carboniferous) of Wardie, Edinburgh.

MATERIAL. In addition to the holotype, five complete fish, three tolerably com-
plete fish, five skulls, isolated maxillae and scales in the British Museum (Natural
History) and the Royal Scottish Museum, Edinburgh.

REMARKS. The specific name functatus is rather an unfortunate one since the
normal scale ornamentation consists of quite distinct, characteristic ridges of enamel.
Only the occasional specimen has scales which show coarse punctures over the exter-

Clay Ang

Fic. 1. Gonatodus punctatus (Agassiz). Reconstruction of skull in lateral view.

PALAEONISCOID FISHES AND THE CHONDROSTEIL 149

nal surface, and in these cases it is due to the resorption of most of the enamel layer,
so that one is merely looking at the deeper layers of the scale, the punctures repre-
senting old blood vessel pathways.

I have already discussed the loss of ornamentation in the genus Namaichthys
(Gardiner 1962:19) but without really emphasizing this process of resorption.
Resorption has been noted in the Agnatha, the Dipnoi and the Crossopterygii, in
skull roofing elements as well as scales, but has not so far been noticed in the Actinop-
terygii. Resorption of the scale ornamentation is most marked in the genus
Elonichthys (E. robisont (Hibbert), E. semistrviatus Traquair) but I have also observed
it in Rhadinichthys, Drydenius and Gonatodus. Normally the ornamentation of the
scales in the posterior region is the first to undergo resorption, thus often the anterior
part of the fish shows scales with a complex ornamentation of ridges and striae of
enamel while the posterior scales are smooth apart from a distinct puncturing. In
the past when the erection of species on scale ornamentation was a common practice
this led to the naming of species which can often now be shown to be merely differ-
ences in resorption of an original scale type.

The scales from Wardie figured by Traquair (1901, pl. 9, figs. 6, 7) as Elonichthys
vobisont (Hibbert) var. intermedius Traquair are identical with those of Gonatodus
punctatus (Agassiz). Traquair believed that the sculpturing exhibited by these
scales was due to the loss of the ganoine (enamel) layer. However, it is now clear
that these figures of Traquair’s (pl. 9, figs. 6, 7) represent the normal, ornamented,
external scale surface of Gonatodus punctatus.

DescripTion. Theskull. The general shape of the skull can be seen from Text-fig.
1. Although the snout has a distinct rostrum, it is not nearly so pronounced as in
the genera Elonichthys Giebel and Nematoptychius Traquair. The most characteristic
features are the dermohyals and the arrangement and shape of the teeth.

The skull roofing bones are all ornamented with stout ridges of enamel, which in
places give way to tubercles. On the parietals, frontals, dermopterotics and postro-
stral bones the ridges are broken up into elongated tubercles which run from the
centre of ossification outwards. The ridges of enamel on the suprascapulars and
extrascapulars are in the form of short striae which run concentrically. A similar
ornamentation is seen on the suborbitals. The maxilla is covered by long ridges of
enamel, partially broken up into segments, which sweep up and round the bone,
running parallel with its posterior and superior borders. Ventrally the maxilla has
more tubercles. On the subopercular the ridges are short and run more or less
parallel horizontally, while on the opercular they are also short but run diagonally
across the bone from the antero-dorsal corner.

The mandible is covered by long straight ridges of enamel which run along its
entire length, a similar condition is seen on the branchiostegal rays, except that the
ridges are much fewer in number.

The skull is relatively short and the suspensorium not too far off the vertical.
The suprascapulars are broad and meet in the midline and the extrascapular series
consists of the normal narrow, single pair of bones. The paired parietals are rec-
tangular while the frontals are the longest bones in the skull roof and anteriorly

150 PALAEONISCOID FISHES AND THE CHONDROSTEI

meet both the postrostral and nasals. Laterally the frontal is bordered by the
dermopterotic, supraorbital and nasal.

The lateral roof of the skull is comprised of two bones, the dermopterotic and the
supraorbital. The dermopterotic is the larger and its dorsal margin is strongly
emarginated at two points. Anteriorly the dermopterotic joins the shorter supra-
orbital, which separates it from the nasal. The nasal is short and broad and its
lateral margin forms together with the dorsal margin of the antorbital a distinct
notch, which indicates the position of the posterior nostril. The anterior nostril lies
between the nasal and the postrostral. Posteriorly the nasal joins the frontal and
supraorbital. The postrostral is another stout, broad bone, moderately convex
anteriorly, which meets ventrally the paired, toothed, rostro-premaxillaries.
Posterior to the rostro-premaxillary is a stout antorbital, which does not enter into
the jaw margin.

There are two members of the infraorbital series while the dermosphenotic is long
and forms a considerable portion of the upper posterior margin of the orbit. The
suborbital series consists of two rounded bones, the upper of which is the larger.

The maxilla is of the normal palaeoniscoid pattern and curves slightly upwards
before it meets the rostro-premaxillary. The ventral margin bears teeth along its
entire length ; the teeth are nearly uniform, long and stout and arranged in a single
closely set row. Each tooth has a distinct terminal cap. The preopercular is a high
bone with its posterior margin much nearer to the vertical than that of Elonichthys.
Behind the preopercular and filling the gap between it and the opercular lie two
wedge-shaped dermohyals. However in one specimen, R.S.M. 1926.57.16, at
least three dermohyal elements are present. The opercular is rectangular in outline
and more than twice as high as the subopercular. Preceding the subopercular are
twelve to thirteen broad branchiostegal rays and a large median gular. The orbit
is supported by a stout sclerotic ring composed of four elements.

The lower jaw. Posteriorly the lower jaw is much deepened due to the upward
extension of the dentary and to the presence of a surangular, much as in Amblypterus.
The outer surface of the jaw is formed by a stout angular posteriorly while the
remainder is made up of the large dentary. The dentary also forms part of the inner
surface anteriorly. The remainder of the inner surface is covered by a large pre-
articular. The upper border of the dentary supports a single series of stout, tall,
rather closely set teeth. There is a coronoid series which overlaps the dorsal,
internal portion of the dentary. Posteriorly the articular portion of Meckel’s
cartilage is ossified.

The palate. The palate has already been figured by Watson (1925, text-fig. 27).
The quadrate is stout and the ectopterygoid small in comparison with the much
longer entopterygoid. The metapterygoid is well ossified as is the remainder of the
suprapterygoid series which forms a single, large ossification. Anteriorly the suprap-
terygoid is notched for the passage of the maxillary and mandibular divisions of the
fifth nerve. The palatine is a mere sliver of bone, but both it and the ectopterygoid
bear a single row of large teeth. Anteriorly there is a pair of toothed vomers.

The neurocranium. Apart from the parasphenoid, little can be made out. The

PALAEONISCOID FISHES AND THE CHONDROSTEI I51

fH] i} gala
Hf Hee

ZL
SSS SLLALILLL-”
SST

Fic. 2. Gonatodus punctatus (Agassiz). Restoration of whole fish.

parasphenoid is short and broad as in Pteronisculus White, with a well marked
basipterygoid process (see Traquair 1877c, pl. 2, fig. 5).

The paired fins and their girdles. The pectoral girdle consists of a supracleithrum
which stretches down to the junction between the opercular and subopercular, an
elongate cleithrum and a stout clavicle.

The ornamentation on the supracleithrum is delicate, with ridges of enamel which
branch to a limited degree and run across the bone rather than following its length
as they do on the cleithrum.

The pectoral fin is large with between thirty and thirty-two lepidotrichia. The
rays are smooth and the articulations tolerably close, especially in the finer rays of the
posterior part of the fin. However, in the most anterior fin rays the articulations are
quite far apart. The base of the fin is covered by a series of small scales.

The pelvic fin is somewhat smaller with between eighteen and twenty rays, all
closely articulated.

Ane,

Fic. 3. Gonatodus punctatus (Agassiz). Anterior flank scale.

152 PALAEONISCOID FISHES AND THE CHONDROSTEL

The unpaired fins. Both dorsal and anal fins are triangular and of similar size
with between forty and forty-five rays. The articulations of the rays in both fins
are so close that the joints of the base of the fins look most scale-like. The fin rays are
smooth apart from a single longitudinal furrow.

The caudal fin is heterocercal, inequilobate and deeply cleft.

The squamation. The scales of the flank are slightly higher than broad and at
the deepest point of the body number around twenty-two rows. They possess a
distinct peg and socket articulation and in proportion to body size the scales are
large. The ornamentation is most distinct consisting of prominent ridges of enamel
(Text-fig. 3). The ridges follow the postero-ventral margin and also the dorsal
margin. There is a triangular portion posteriorly where short, pointed, overlapping
tubercles pass back towards the hind margin which is pectinated.

There are three large ridge scales in front of the dorsal fin, and the ridge scales of
the axial lobe start immediately behind the dorsal fin.

Scale structure. The general structure of the scale can be seen from Text-fig. 4.
Aldinger (1937 : 212) has already pointed out the similarity of the scale structure to
that seen in Elonichthys. The canal plexus of the dentine layer is composed of
horizontal and radial canals (vascular). The horizontal canals bear tree-like dentine
tubules. The radial canals penetrate obliquely through the bony lamellae from the
scale margins, sending feeder branches upwards to connect with the horizontal
canals. Thus the radial canals are for the most part at a much deeper level than the
horizontal canals. A few canals (canals of Williamson) penetrate upwards from the
centre of the base of the scale to connect with both the radial canals and the horizontal

h.v.can

nenrcan
a) CH AVAA Y W. WNZw aM
LE BU Aa : SNS Pa

eer) :
; ee ncan

Swe = = = 4
= S )
See Ve h.la

Fic. 4. Gonatodus punctatus (Agassiz). Dorso-ventral cross section through mid-lateral
scale. 45. From B.M.N.H. P.11704.

canals. The horizontal canals often connect with one another by means of radial
cross connections. Occasionally ascending branches of the dentine canal plexus
penetrate the external enamel layer (as in Elonichthys). Bone cells are present in the
bony lamellae but no fibres of Sharpey were observed.

In overall structure the scale of Gonatodus punctatus is close to that of Elonichthys.

PSEA ONES COLD hls EH ES AND! DHE CHOON DROS TET 153

LOCALITY AND HORIZON. Apparently confined to the Calciferous Sandstone Series
(Lower and Middle Visean) at the base of the Scottish Lower Carboniferous. It is
recorded from ironstone nodules at Wardie, Edinburgh, from Gullane, East Lothian
and from Pitchorthy, Fifeshire. One specimen, also in an ironstone nodule, is
recorded from Collinton, Edinburgh (B.M.N.H. P.11704).

Genus PSEUDOGONATODUS nov.

1877 Gonatodus Traquair (partim) : 271.
1882 Gonatodus Traquair (partim) : 546.

Diacnosis. Body fusiform; dorsal fin placed rather far back (nearer to tail
than in Gonatodus Traquair), partly in advance of, partly opposing the anal but
giving appearance of being almost opposite it. Both fins of moderate size, triangular,
smaller than in Gonatodus. Paired fins large, with stout rays, pelvics placed nearer
to anal than to pectorals ; caudal fin deeply cleft and inequilobate. All fins with
prominent fulcra anteriorly and rays closely articulated and distally bifurcated.
Skull with almost vertical suspensorium (as in Amblypterus) and rostrum not pro-
nounced. Opercular less than one and three-quarter times as deep as subopercular.
Suborbital and dermohyal series present. Branchiostegal rays numerous, skull
roofing bones and cheek bones ornamented with coarse, stout ridges of enamel ;
teeth of variable size but in single series. Scales rhomboid and large (larger than in
Gonatodus), ornamented with fine striae and distinct punctations.

TYPE SPECIES. Gonatodus parvidens Traquair.

REMARKS. Traquair (1877; 1882) placed three species in the genus Gonatodus,
G. punctatus (Agassiz), G. macrolepis Traquair and G. parvidens Traquair. It is now
clear that G. parvidens and G. macrolepis are not congeneric with G. punctatus and
the new genus Pseudogonatodus is here proposed for them. For the type species I
have selected G. parvidens since this is by far the better known of the two species.

Pseudogonatodus parvidens (Traquair)
(Text-figs. 5, 6)

1881 Gonatodus sp., Traquair : 315 (name only)

1882 Gonatodus parvidens Traquair : 546.

1890 Gonatodus parvidens Traquair ; Traquair : 392.

1891 Gonatodus parvidens Traquair ; Woodward : 435, pl. 16, fig. 7.
1903 Gonatodus parvidens Traquair ; Traquair : 695, 696, 700, 701.
1907 Gonatodus parvidens Traquair ; Traquair : 99, pl. 21, text-fig. 3c.
1937 Gonatodus parvidens Traquair ; Aldinger : 212, text-fig. 55.

1954 Gonatodus parvidens Traquair ; Waterston : 57.

D1AGnosis (emended). Fishes not exceeding 22 cm. in total length, body fusiform,
length of head contained little over five times and greatest depth of body four times
in total body length. Suspensorium nearly vertical, head short and with bluntly
rounded snout. Maxilla of distinctive shape, high posteriorly. Single large sub-
orbital present. Teeth on jaws very small, arranged in single closely-set row.

154 PALAEONISCOID FISHES AND THE CHONDROSTEI

Scales large in proportion to body size, deeper than broad on the flank, feebly
ornamented with few striae and exhibiting coarse punctations. Posterior borders
of scales delicately serrated.

LECTOTYPE, here chosen. R.S.M. 1926.57.19, a tolerably entire fish from the
Borough Lee Ironstone, Edge Coal series, Loanhead, near Edinburgh.

MATERIAL. In addition to the lectotype, about twelve almost complete fish,
eight bodies wanting skull, two skulls, isolated maxillae, scales and jaws in the
British Museum (Natural History) and the Royal Scottish Museum, Edinburgh.

REMARKS. Pseudogonatodus parvidens (Traquair) was first recognized by Traquair
(r882 : 546) as a distinct species using detached maxillary bones from the Borough
Lee Ironstone supplemented by more or less entire fish from the same locality. The

Pa Exsc
Dpt
Fr
Ds
& Ssc
Scl
Pop
nay
na
2 Op
Inf
Sop

Br Cl

Ang
Clav

10mm

Fic. 5. Pseudogonatodus parvidens (Traquair). Reconstruction of skull in lateral view.

PALAEONISCOID FISHES AND THE CHONDROSTEI 155

maxilla is perhaps still the most characteristic feature by which the species can be
recognized.

The ornamentation of the scales, with few striae and many punctations, is mis-
leading and is once more the result of resorption of superficial enamel. It would seem
that if younger, better preserved specimens were available a more extensive ornamen-
tation comprising ridges of enamel would be apparent.

Description. The skull. The bones of the skull and course of the sensory canals
are shown in Text-fig. 5. The more distinctive features are the almost upright
suspensorium and the shape of the maxilla.

The skull roofing bones are all ornamented with stout, long ridges of enamel, more
pronounced than in G. punctatus. The ridges more or less follow the length of the
bones with a few interspersed tubercles on the frontals and the anterior end of the
postrostral. The ridges on the maxilla are likewise prominent, running up and round
the bone parallel with the posterior and superior borders. Even on the anterior end
of the maxilla short ridges run upwards from the ventral surface. On the sub-
opercular the ridges run parallel from front to back while on the opercular they run
diagonally across from the antero-dorsal corner. On the mandible the ridges of
enamel are again stout and run along its entire length. Similarly on the gular plate
and branchiostegal rays stout long striae run along the long axis of the bones.

The skull is short and the suspensorium almost upright (nearer the vertical than in
Gonatodus). The suprascapulars are of normal size and are preceded by a narrow
pair of extrascapulars. The parietals are square in outline and the frontals are very
large. Anteriorly the frontal meets both postrostral and nasal. Laterally the parie-
tals and frontals are bordered by the dermopterotic and dermosphenotic. The
dermopterotic is roughly rectangular but anteriorly is produced into a point. The
dermosphenotic is larger than the dermopterotic and anteriorly meets the nasal.
Both the nasal and the postrostral are long, stout bones of somewhat the same pro-
portions as in Amblypterus. The position of the anterior nostril is clearly marked
between the postrostral and nasal while that of the posterior is indicated by a distinct
notch in the nasal. Both nostrils are borne rather high up on the snout as in Amblyp-
terus. The most anterior portion of the snout is not clearly preserved in any of the
specimens I have examined.

There are two members of the infraorbital series, the second of which is much
expanded. The infraorbital sensory canal on its passage through this second infra-
orbital gives off many short branches posteriorly. The suborbital series consists of
one bone only which is much higher than broad.

The maxilla is of unusual shape, high posteriorly and with an almost vertical
posterior border. The ventral margin bears very small teeth along its entire length,
arranged in a single closely set series. Although small, the teeth are of the same shape
and arrangement as in G. punctatus. The preopercular is a high bone with a gap
between it and the opercular. Although it cannot be seen with certainty, a dermo-
hyal element appears to fill this gap. The opercular is almost rectangular, less than
one and three-quarter times as high as the subopercular. Below the subopercular
are ten branchiostegal rays and a large median gular. A sclerotic ring probably
supported the eye.

156 PALAEONISCOID FISHES AND THE CHONDROSTEI

The lower jaw. Posteriorly the lower jaw is much expanded, giving a high pro-
minence, much as in Amblypterus and Gonatodus. Anteriorly the jaw is much
shallower. The angular bone does not extend very far along the ventral jaw surface
and the articular region is ossified. The dentary bears a single series of small closely
set teeth. Anteriorly the dentary forms part of the inner jaw surface while the
remainder of the inner surface is covered by the prearticular and a coronoid series.

The palate. Although little could be made out, both the ectopterygoid and pala-
tine bear numerous small teeth.

The unpaired fins and their girdles. The supracleithrum extends ventrally beyond
the junction of the opercular with the subopercular, the cleithrum is elongate and
the clavicle stout. The ornamentation of both supracleithrum and cleithrum is
distinct and consists of elongate striae of enamel which follow the long axis of the
bone.

The pectoral fin is large with seventeen to nineteen lepidotrichia.

The pelvic fin is smaller with only twelve to thirteen fin rays.

The unpaired fins. Both dorsal and anal fins are triangular with the dorsal fin
placed well back. The number of rays cannot be determined with complete accuracy
but is not many more than twenty in either fin. The fulcra are distinct, larger than
in Gonatodus.

The caudal fin is heterocercal, inequilobate and deeply cleft.

All the fin rays are closely articulated and distally bifurcated.

The squamation. The scales are higher than broad on the flank and at the deepest
part of the body number around ten to twelve rows. Thus in proportion to body
size the scales are large (larger than in Gonatodus). The ornamentation can be seen
from Text-fig. 6 and consists of two or three striae following the antero-ventral border
and of serrations (fine denticulations) posteriorly. The main portion of the scale is
smooth apart from numerous punctations.

Scale structure. The scale structure of Pseudogonatodus parvidens (Traquair)
has already been partly dealt with by Aldinger (1937, fig. 55). In general it re-
sembles that of Elonichthys. For a more complete description of the scale structure
of Pseudogonatodus see under Pseudogonatodus macrolepis (Traquair).

LOCALITY AND HORIZON. Carboniferous of Scotland, Namurian to Ammanian
(Westphalian A). From the Edge Coal series (Borough Lee Ironstone) at Borough
Lee, Loanhead, Wallyford, Possil and Lochgelly. Also from the South Parrot coal
seam, Niddrie.

I'ic. 6. Pseudogonatodus parvidens (Traquair). Anterior flank scale.

PALAEONISCOID FISHES AND THE CHONDROSTEI 157
Pseudogonatodus macrolepis (Traquair)
(Text-figs. 7-9)

1877 Gonatodus macrolepis Traquair : 271.

1877b Gonatodus macrolepis Traquair : 556.

1881 Gonatodus macrolepis Traquair ; Traquair : 35 (name only).

1882 Gonatodus macrolepis Traquair ; Traquair : 546.

1890 Gonatodus macrolepis Traquair ; Traquair : 391.

1891 Gonatodus macrolepis Traquair ; Woodward : 435, pl. 16, fig. 8.

1903 Gonatodus macrolepis Traquair ; Traquair : 692, 694, 700, 701.

1907 Gonatodus macrolepis Traquair ; Traquair: 97, pl. 20, figs. 9-14, text-fig. 3b.
1937 Gonatodus macrolepis Traquair ; Aldinger : 212.

1954 Gonatodus macrolepis Traquair ; Waterston : 57.

D1acGnosis (emended). Fishes not exceeding 18 cm. in total length, body fusiform,
length of head contained five times and greatest depth of body little more than four
times in total body length. Maxilla of distinctive shape, postero-ventral margin
being sharply bent downward, so much so that main blade of maxilla is much higher
than broad (see Text-fig. 7). Teeth on jaws large, arranged in single row but of
same shape and disposition as in Pseudogonatodus parvidens and Gonatodus punctatus.

mm,

Fic. 7. Maxilla of Pseudogonatodus macrolepis (Traquair).

Scales large in proportion to body size, numbering around ten to twelve rows at
deepest part of body. Scales deeper than broad on flank, rhomboidal, feebly orna-
mented with few traces of striae and with finely serrated posterior border. Remain-
der of scale surface smooth apart from delicate punctures (Text-fig. 8). Again
resorption seems to have occurred. Large ridge scales preceding caudal fin.

LecToTYPE. R.S.M. 1926.57.20 and counterpart B.M.N.H. P.11648, an im-
perfect fish from the Gilmerton Ironstone, Lower Carboniferous Limestone series,
Venturefair Pit, Gilmerton, near Edinburgh. Designated as the “ type specimen ”’
by Traquair (1907, pl. 20, fig. 9).

MATERIAL. In addition to the lectotype, six tolerably complete fish, four bodies

158 PALAEONISCOID FISHES AND THE CHONDROSTEI

without heads, three isolated maxillae and one dentary in the British Museum
(Natural History) and the Royal Scottish Museum, Edinburgh.

Remarks. Apart from the shape of the maxilla and the large teeth, this species
is very similar to the type species.

a

3mm
|

Fic. 8. Pseudogonatodus macrolepis (Traquair). Anterior flank scale.

Scale structure. The fine structure of the scale can be seen in Text-fig.9. In most
respects it is similar to that already described in Gonatodus punctatus (Agassiz).
The arrangement of the canals in both the dentine and the bony lamellae is identical
with those in Gonatodus punctatus. However, in the scale of Pseudogonatodus
macrolepis there is a great concentration of fibres of Sharpey which was not observed

PPAR

=_
_

Fic. 9. Pseudogonatodus macrolepis (Traquair).

Dorso-ventral cross section through
mid-lateral scale. 100. From B.M.N.H. P.11648.

PALAEONISCOID FISHES AND THE CHONDROSTEI 159

in Gonatodus punctatus, while there appears to be a greater number of cell spaces in
the bony lamellae of Pseudogonatodus macrolepis.

The structure of the scale of Pseudogonatodus is close to that of both Gonatodus
and Amblypterus.

LOCALITY AND HORIZON. Lower Carboniferous (Middle to late Visean). A few
specimens from the Calciferous Sandstone of Straiton, the remainder from the Gilmer-
ton Ironstone, Venturefair Pit, Gilmerton.

Genus DRYDENIUS Traquair 1890
1877¢ Microconodus Traqtair 112, 33 (name only, but see also Traquair 1907 : 103).
1888 Gonatodus Traquair (partim) : 252.

Dracnosis (emended). Body fusiform, dorsal fin placed opposite space between
pelvics and anal, both dorsal and anal fins triangular, anal smaller. Caudal fin
heterocercal, inequilobate. Pelvics and pectorals with less than ten rays. All fins
have small number of rays articulated and distally bifurcated ; small fulcra fringe
leading edges. Suspensorium appears to have been almost vertical. Teeth on both
maxilla and dentary in single series, stout and closely set. Coronoid series also
bearing large teeth set in more than one row. Scales very large in proportion to
body size, deeper than broad on flank and ornamented with occasional striae.

TYPE SPECIES. Drydenius insignis Traquair.

REMARKS. This genus contains but two species. D. insignis Traquair and D.
molyneuxt (Traquair), both of which were placed in it by Traquair (1890 : 392 ;
1907: 102). The large scales, stout but few fin rays and the presence of coronoid
teeth make the genus readily identifiable. An interesting feature of Drydenius is
that the maxilla has begun to grow down around the peg-like teeth and partially
encases them proximally. In the later genus Pavamblypterus Sauvage which was
derived from the Gonatodidae the maxilla completely encases the teeth apart from the
distal tips.

Drydenius insignis Traquair
(Text-figs. 10, II)
1890 Dyrydenius insignis Traquair : 392, 399.
1891 Drydenius insignis Traquair ; Woodward : 437.
1903 Drydenius insignis Traquair ; Traquair : 695, 700, 701.
1907 Drydenius insignis Traquair ; Traquair : 101, pl. 22, figs. 5-9.
1954 Drydenius insignis Traquair ; Waterston : 59.

Dracnosis (emended). Fishes not exceeding 13 cm. in total length, body fusiform,
dorsal fin opposite space between pelvics and anal. Caudal fin heterocercal and
deeply cleft. Length of head contained little more than five times and greatest
depth of body four and a half times in total body length. Scales denticulated pos-
teriorly, surface smooth with no ornamentation apart from fine punctations (see
Text-fig. 11).

LECTOTYPE. Selected by Waterston (1954:59). R.S.M. 1950.38.85, a splenial
bone from the Borough Lee Ironstone, Edge Coal Series ; Loanhead near Edinburgh.

GEOL. 14, 5. 17

160 PALAEONISCOID FISHES AND THE CHONDROSTEI

3mm

Fic. 10. Maxilla of Drydenius insignis Traquair.

MATERIAL. In addition to the Lectotype, two complete fish, one body wanting
head, two isolated maxillae, and two splenials in the British Museum (Natural
History) and the Royal Scottish Museum.

DeEscrIPTION. The external skull bones, including the jaws, are all ornamented
with long, relatively stout striae of enamel which more or less follow the length
of the bones. The suspensorium appears to be almost upright.

3mm

Fic. 11. Dvrydenius insignis Traquair. Anterior flank scale.

On the coronoid series is a single row of very large, stout teeth, and internal to
this there are often numerous small granular teeth. The palatal bones also appear
to be clothed with small granular teeth. The large teeth on the coronoid are larger
than the teeth on the dentary or maxilla.

The maxilla is of unusual shape and quite diagnostic. The main blade is small in
comparison with the normal palaeoniscoid maxilla, and not as high as in Pseudo-
gonatodus macrolepis. The teeth on the maxilla are large and stout and in a single
series along the anterior two-thirds of the bone (Text-fig. ro).

The pectoral fin (R.S.M. 1890.78.18) has from eight to ten lepidotrichia, and is
short-based. The pelvic fin has some eight or nine stout rays while the dorsal fin
is the largest with from twenty to twenty-two rays. The anal fin has fifteen rays.

The scales are large in proportion to body size, at the deepest part of the body
numbering between nine and ten rows. Scales ornamented with striae and puncta-
tions ; hinder margin denticulated.

LOCALITY AND HORIZON. Carboniferous of Scotland, early Namurian (E,—E,).
From the Borough Lee Ironstone at Borough Lee and Loanhead. (The Borough Lee
Ironstone is a member of the Edge Coal or Middle Carboniferous Limestone series.)

PALAEONISCOID FISHES AND THE CHONDROSTEI 161

Drydenius molyneuxi (Traquair)
(Text-figs. 12-15)

1877¢ Microconodus molyneuxt Traquair : 33 (name only).

1888 Gonatodus molyneuxi (Traquair) Traquair : 252.

1890 Gonatodus molyneuxi (Traquair) ; Ward : 178, pl. 6, fig. 11.
1891 Gonatodus molyneuxi (Traquair) ; Woodward : 436, 437.

1905 Gonatodus molyneuxi (Traquair) ; Ward: p. 302, pl. 6, fig. 2.
1907. Drydenius molyneuxi (Traquair) Traquair : 102, pl. 20, figs. 6-8.
1919 Drydenius molyneux (Traquair) ; Pruvost : 425, 426, pl. 29, figs. 6-17.
1930 Drydenius molyneuxt (Traquair) ; Pruvost : 130, pl. 1, fig. 5.
1943 Dyrydenius molyneuxi (Traquair) ; Heide : 39, pl. 3, fig. 5a-f.
1954 Drydenius molyneuxi (Traquair) ; Waterston : 59.

1958 ?Drydenius sp. Vangerone : 472, pl. 23, fig. 3.

Ptr

3mm
——— ny

Fic. 12. Dyrydenius molyneuxi (Traquair). Dorsal view of skull roofing bones. From
B.M.N.H. P.7973 & counterpart.

162 PALAEONISCOID FISHES AND THE CHONDROSTEI

D1AGnosis (emended). Fishes not exceeding 7 cm. in total length, body fusiform.
Length of head contained four a half times, and greatest depth of body a little over
four times, in total body length. Blade of maxilla rounded dorsally. Scales very
finely denticulated posteriorly (far more denticulations than in D. imsignis), with
fine striations running up diagonally from these denticulations. These fine striae
only on posterior third of scale; anterior portion with few punctures and an occa-
sional stria following anteroventral margin (Text-fig. 14).

LECTOTYPE, here chosen. B.M.N.H. P.7973 and counterpart P.7976, a tolerably
complete fish from the Deep Mine Ironstone (Westphalian C), Longton, Staffordshire.

MATERIAL. In addition to the lectotype, five comparatively complete fish, four
bodies without skulls, and isolated lower jaws in the British Museum (Natural
History) and the Royal Scottish Museum.

DESCRIPTION. The external skull bones including the jaws are ornamented with
stout ridges of enamel, which are more or less contorted and not in straight lines as in
D. insignis, Pseudogonatodus and Gonatodus. The orbit is large and the snout
rounded. The suspensorium is upright and the lower jaw short and stout. The
sutures between the frontals and between the frontals and parietals are digitate
(Text-fig. 12). The maxilla is of characteristic shape (Text-fig. 13) with a larger

3mm

Fic. 13. Maxilla of Drydenius molyneuxi (Traquair).

blade than in D. insignis. The coronoid bears large stout teeth as in D. insignis.
The second infraorbital is expanded as in Pseudogonatodus parvidens and the infra-
orbital canal much branched posteriorly as it passes through this bone. There are
eleven branchiostegal rays and a median gular.

(\

umm

Fic. 14. Drydenius molyneuxi (Traquair). Anterior flank scale.

PALAEONISCOID FISHES AND THE CHONDROSTEI 163

The fins are always poorly preserved. No specimen I have examined shows a
pectoral fin, while all that can be said about the anal is that it only possessed a few
rays. The pelvics have some seven or eight stout rays and the dorsal about fifteen.

The scales are very large in porportion to body size, at the deepest part of the body
numbering between eight and nine rows.

Scale structure. The structure of the scale (Text-fig. 15) is almost identical with
that of Pseudogonatodus macrolepis (Traquair) even down to the large number of
fibres of Sharpey. The bone cells are however larger and considerably fewer in
Drydenius molyneuxt.

d
dt

h.v.can

h.la

can.W

Fic. 15. Dyrydenius molyneuxi (Traquair). Dorso-ventral cross section through mid-lateral
scale. x90. From B.M.N.H. P.7974.

LOCALITY AND HORIZON. The lectotype came from the North Staffordshire Coal
Measures (Westphalian C), in the Deep Mine Ironstone at Longton, Staffordshire.
It is also recorded from Broadsfield, Fenton. On the Continent it has been recorded
from the Upper Carboniferous of France (Pruvost 1919: 425), Belgium (Pruvost
1930: 130; Heide 1943: 39) and, doubtfully, from Aachen, Germany (Vangerone
1958 : 472).

Family TRISSOLEPIDIDAE Fri¢ 1893
1936 Gymnoniscidae Berg : 345.
Diacnosis. Trunk fusiform ; dorsal fin in front of anal fin. Fin rays of all fins
relatively few, articulated but not distally bifurcating. Suprascapular large ;
suspensorium upright or nearly so; snout rounded; teeth prominent on both jaws

164 PALAEONISCOID FISHES AND THE CHONDROSTEI

and branchiostegal rays numerous. Fins with few stout basal fulcra (Woodward
1942) anteriorly (much as in Phanerorhynchidae) but true series of fulcral scales
absent.

REMARKS. Frié (1893) erected the Family Trissolepidae to include Tyissolepis
Fri¢é. Unfortunately Tvissolepis is a synonym of Sphaerolepis Fri¢. This family
contains Sphaerolepis Frié and Sceletophorus Frié from the Upper Carboniferous. The
Trissolepididae were probably derived from the earlier Holuridae and as such show
some similarity to the Phanerorhynchidae and to the Teleopterinidae which likewise
had their origins in the Holuridae. The Family Atherstoniidae nov. (which includes
Atherstonia Woodward) are also related to the Trissolepididae.

Genus SPHAEROLEPIS Frié 1877
1893 Tvissolepis Friێ : 76.

Diacnosis (emended). Trunk fusiform, tail very long ; dorsal fin in front of anal
fin, opposite gap between anal and pelvic fins. Dorsal contour arched in advance of
dorsal fin ; pectoral fins small, pelvics larger than pectorals. Apart from pelvics
all fins have stout basal fulcra (Woodward 1942) forming their leading edges, but
true series of fulcral scales absent. All lepidotrichia stout, articulated and not
distally bifurcated. Suspensorium upright ; opercular more than three times as
deep as subopercular. Dorso-posterior blade of maxilla rounded as in Pseudo-
gonatodus. Fossa present in skull roof at junction of parietal, frontal and dermop-
terotic. Teeth numerous, of moderate size and in one series. Scales cycloidal, those
in anterior trunk region being pectinated posteriorly.

TYPE SPECIES. Sphaerolepis kounoviensis Fri¢. The only species.

REMARKS. Fri¢ (1875: 76) in the first mention of the species Kounoviensis
merely stated that it belonged to a new genus without erecting a generic name for it.
Later in 1877 he placed it in the new genus Sphaerolepis. Two years later in his
“ Fauna der Gaskohle ” (1879, 1 : 31) he referred to it as Sphaerolepis kounoviensis,
but later in the same work (1893, 3, 2 : 76) changed the name to T7ssolepis kouno-
viensis and put it in a new family, the Trissolepidae. Clearly since his original
descriptions (1875, 1877) give enough information to recognize the species then the
generic name Sphaerolepis must stand and Tvissolepis must be treated asa synonym.

Sphaerolepis kounoviensis Fri¢é
(Plate x; Text-figs. 16-18)

1875 (Nov. gen.) Kounoviensis Frié : 76.

1877 Sphaerolepis kounoviensis Frié : 46.

1879 Sphaerolepis kounoviensis Frié; Frié, 1: 31.

1891 Sphaerolepis kounoviensis Frié ; Woodward : 523.

1893 Trissolepis kounoviensis (Fric) Fri¢, 3, 2: 76, pls. 109-112, text-figs. 277, 278.
1907. Sphaevolepis kounoviensis Frié ; Traquair : 106.

1909 Sphaevolepis kounoviensis Frié ; Traquair : 107.

1944 Trissolepis kounoviensis (Fri€) ; Westoll : 65.

PALAEONISCOID FISHES AND THE CHONDROSTEI 165

DraGnosis. Fishes not exceeding rr cm. in total length ; body fusiform, length
of head contained almost five times and greatest depth of body just over four times in
total body length. Dorsal fin shorter based than anal fin ; two ridge scales in front
of dorsal. Fossa present in skull roof at junction of parietal, frontal and dermop-
terotic.

SYNTYPES. Four specimens in the Narodni Museum, Prague, from the Upper
Carboniferous of Kounova, Czechoslovakia.

MATERIAL. Photographs and casts of syntypes, and no. 47491 in the British
Museum (Natural History).

Description. The skull. The bones of the skull and course of the sensory canals
are shown in Text-fig. 16. The skull is short and high with an almost vertical
suspensorium while the snout is bluntly rounded. The suprascapulars are large,
stout and meet in the midline anteriorly. The extrascapular series, consisting of the

Exsce
Ssc Pa

Dpt

VVVYY

~ Tv V
ory y WY
7
1
AVA AA AAAW AA AAA AAA

Cl y
Sop LaF Gu

Br Smm

Fic. 16. Sphaerolepis kounoviensis Fri¢é. Reconstruction of skull in lateral veiw.

166 PALAEONISCOID FISHES AND THE CHONDROSTEI

normal single pair of bones, is somewhat larger than in the average palaeoniscoid.
The postero-dorsal surface of each extrascapular is peculiarly scalloped. The
parietals are almost square while the lateral wall of the skull roof is composed of a
short, broad, dermopterotic, but the position and extent of the dermosphenotic is
not clear on any of the specimens. _ Fri¢ (1893, pls. 109-112) in his interpretation of
the skull roof indicated a true fossa in the region of the junction between the parietal,
frontal and dermopterotic. From a careful study of that region in the specimen
concerned I agree with Fri¢ that such a fossa did exist in life and was not caused
after death by an otolith as suggested by Westoll (1944: 65). In this respect
Sphaerolepis shows some similarity to Pyritocephalus Fri¢ (Westoll 1944).

The frontals are large and the nasals and postrostral long, but the remaining bones
constituting the snout region could not be distinguished with any degree of certainty
although the position of the nares could be seen.

The maxilla is of unusual shape being rounded dorsally as in Pseudogonatodus nov.
The teeth on the maxilla are of moderate size (somewhat larger than in the normal
palaeoniscoid) and arranged in a single series. The preopercular is almost vertical
and much narrower than in most palaeoniscoids and in this respect it also resembles
Pseudogonatodus nov. The opercular is more than three times deeper than broad.
Beneath the subopercular are some ten branchiostegal rays with a median gular
anteriorly,

Lower jaw. The angular only occupies a small portion of the jaw surface while
the dentary is considerably thickened ventrally and forms the major portion of the
outer surface of the jaw. A single row of stout teeth is present on the upper margin
of the dentary.

The palate. The parasphenoid is short, narrow anteriorly, but expands posteriorly,
and in general size and shape is similar to that of Pteronisculus White (Nielson 1942).
Both the vomers and palatines are large and bear prominent teeth. Those on the
palatine are stout, rounded, arranged in a cluster about three deep.

The paired fins and their girdles. The supracleithrum extends about half way
down the opercular and the cleithrum is elongated and comparatively stout. A

_5Smm_,

Fic. 17. Sphaerolepis kounoviensis Frié. Restoration of whole fish.

PALAEONISCOID FISHES AND THE CHONDROSTEI 167

distinct ornamentation is visible on both the supracleithrum and the cleithrum,
consisting of lines of sharply pointed tubercles that follow the length of the bone.
Anteriorly a short clavicle can be seen.

The pectoral fin is small with eleven to thirteen lepidotrichia. Apart from the ray
forming the leading edge all the lepidotrichia have distinct articulations.

The pelvic fin is larger with fourteen to sixteen rays, all of them articulated.

The unpaired fins. The dorsal fin is composed of about fifteen stout, articulated
lepidotrichia and is preceded by two ridge scales. The anal fin is longer based than
the dorsal and is made up of more rays, between twenty-five and twenty-eight.
As in the dorsal fin there are two or three basal fulcra (Woodward 1942) in the anal
forming a stout leading edge to the fin.

Imm

see
Fic. 18. Sphaerolepis kounoviensis Frié. Anterior flank scales.

168 PALAEONISCOID FISHES AND THE CHONDROSTEI

The caudal fin is heterocercal but not cleft as in most palaeoniscoids, and in this
respect Sphaerolepis resembles Phanerorhynchus Gill, Palaeoniscinotus Rohon,
Holurus Traquair and Holuropsis Berg. The rays are all articulated apart from the
two stouter ones (basal fulcra) which form the leading edge.

The squamation. The scales, apart from those on the tail, are cycloidal. In this
feature Sphaerolepis shows some similarity to Cryphiolepis Traquair and Browneich-
thys Woodward (Griffith 1958). The exposed surface of the majority of these
cycloidal scales is covered with closely-set, fine, concentric ridges of enamel while
those in the anterior flank region have a much stronger ornamentation. Thus the
first three to four rows of scales behind the pectoral girdle have several rows of
backwardly pointing tubercles on the posterior surface and the posterior edge
pectinated (Text-fig. 18).

The scales on the caudal body-prolongation are rhombic on the sides and on the
dorsal surface form a median row of imbricating, V-shaped scales.

LOCALITY AND HORIZON. All the known specimens came from the “ cannel ”’
coal (“ gas-coal’’) of Kounova, Zabor, Hiedl and Knézoves, which is uppermost
Westphalian extending into the lowermost Stephanian (Floral zones H—I).

Genus SCELETOPHORUS Frié 1894
1894 Phanerosteon Traquair; Fric : 92 (errore).
1936 Gymnoniscus Berg : 345.

DiaGnosis (emended). Trunk fusiform, tail short ; dorsal fin situated in front
of anal fin, opposite gap between anal and pelvic fins. Dorsal contour more or less
flat ; pectoral and pelvic fins of about same size, both relatively small. All fins
with stout basal fulcra (or spines) forming their leading edges, but true series of
fulcral scales absent. All fin rays stout, articulated, but not distally bifurcated.
Opercular not quite twice as deep as subopercular. Dorso-posterior blade of maxilla
rounded as in Sphaerolepis. Teeth numerous, of moderate size, arranged in single
series. Scales rhomboidal and, for most part, ornamented with fine, concentric
ridges of enamel. Ridge scales prominent along dorsal contour.

TYPE SPECIES. Sceletophorus biserialis Fri¢. The only species.

REMARKS. Frié (1894: 92) also described the new species Phanerosteon pauper
from the same locality as Sceletophorus biserialis. Phanerosteon pauper is a small
naked form which Frié (1894 : 92) put into Traquair’s (1881 : 39) genus Phanerosteon
even though he thought it possibly the young of Sceletophorus biserialis. Later
Berg (1936: 345) erected the new genus Gymnoniscus for its reception. Westoll
(1944 : 66) believed Frit¢ (1894 : 92) to be right in his suggestion that Phanerosteon
pauper was probably the young of Sceletophorus biserialis, and after a study of the
specimen concerned I have no doubt that that is so.

Wade (1935) has already shown in Brookvalia that scale development starts behind
the pectoral girdle, along the lateral line canal and on the axial lobe of the tail, and
this is precisely where scales are to be found on the juvenile form of Sceletophorus
biserialis Fri¢. In Phanerosteon Traquair and to a certain extent in Carboveles
White (1927) the development of scales has been arrested at this point and thus they
both provide examples of paedomorphosis.

¢

PALAEONISCOID FISHES AND THE CHONDROSTEI 169

Sceletophorus biserialis Fri¢
(Plates 2,3; Text-figs. 19-21)

1894 Sceletophorus biserialis Frié, 3, 3 : 88, pls. 116, 117.

1894 Phanerosteon pauper Fri€, 3, 3: 92, pl. 117.

1912 Phanerosteon pauper Fri€ ; Traquair : 168.

1936 Gymnoniscus pauper (Fri¢) Berg : 345.

1944 Sceletophorus biserialis Fri€ ; Westoll : 65.

1944 Phanerosteon pauper Frié ; Westoll : 66.

Dracnosis. Fishes not exceeding 6 cm. in total length, body fusiform, length of
head contained just over four times and greatest depth of body just under four times

Sse Exse

Pa

Scl
Fr

Op
Pop

Sop K

Cl
eal
Den
Br
Mx Clav Ang

Fic. 19. Sceletophorus biserialis Fri¢. Reconstruction of skull in lateral view.

170 PALAEONISCOID FISHES AND THE CHONDROSTEI

in total body length. Four ridge scales in front of dorsal fin with paired basal fulcra.
Series of large ridge scales running from immediately behind dorsal fin to tip of tail.
Single ridge scale in front of anal fin.

SyNTYPES. Both the two syntypes of Scletophorus biserialis Frié and the two
syntypes of Phanerosteon pauper Frié are in the Narodni Museum, Prague. All are
from the Upper Carboniferous of Tremosna, Czechoslovakia.

MATERIAL. Photographs and casts of syntypes.

Description. The skull. The skull is large in comparison to total body length.
The suspensorium is vertical and the snout bluntly rounded. The suprascapulars
are large and the extrascapulars narrow. In front of the extrascapulars are a pair of
large, rectangular parietals which are followed by equally stout frontals. The post-
rostral and nasals are long and narrow, but the bones making up the lateral roof of
the skull could not be distinguished with any degree of accuracy. However there
is no evidence of a fossa in the skull roof as in Sphaerolepis Fric.

The maxilla is rounded postero-dorsally much as in Sphaerolepis and Pseudo-
gonatodus. The teeth on the maxilla are of moderate size, arranged in a single row
and similar to those in Sphaerolepis. The preopercular is nearly vertical and shaped
much as in Sphaerolepis. However the opercular is less than twice as deep as the
subopercular and considerably deeper than it is broad. Beneath the subopercular
are about ten branchiostegal rays with a median gular anteriorly.

The lower jaw. Again very similar to Sphaerolepis with the angular only occupying
a small portion of the jaw surface and the dentary thickened ventrally. A single
row of stout teeth is borne on the upper margin of the dentary.

The axial skeleton. The neural and haemal arches are seen as distinct ossifications
(see Pl. 2). The ossified vertebral centra (Wirbelréhre) figured by Frié in Scleto-
phorus biserialis (1894, fig. 286) and also in Phanerosteon pauper (1894, fig. 287)
do not exist : what Frié saw was the lateral line canal of the opposite side of the body,
which shows through the body scaling as a segmental, tubular structure, in such a
position (due to post mortem twisting of the body) as to be aligned in the gap
between the anterior neural and haemal arches (PI. 2).

FSSss
=SSSs

=
=

WM
RY
ao

UES |

Lei

x

SAT
Lif
Jif

Fic. 20. Sceletophorus biserialis Frié. Restoration of whole fish.

PALAEONISCOID FISHES AND THE CHONDROSTEI 171

The paired fins and their girdles. The supracleithrum is longer than in Sphaero-
lepis and extends over two-thirds of the way down the opercular. The cleithrum is
broad and anteriorly united with a short clavicle.

The pectoral fin is longer than the pelvic and composed of only eight rays. An-
teriorly the leading edge is made up of three, short, stout unarticulated basal fulcra
(Woodward 1942). Frié (1894, pls. 116, 117) in his reconstruction of Sceletophorus
biserialis figures some sixteen lepidotrichia in the pectoral fin, but again he was
misled by the impression of the pectoral fin of the opposite side which shows through
the body scaling (Pl. 2).

The pelvic fin contains about thirteen lepidotrichia all of which are stout and arti-
culated. At the leading edge of the fin there are three short, unarticulated basal
fulcra.

Fic. 21. Sceletophorus biserialis Frié. Anterior flank scale.

The unpaired fins. The dorsal fin has between fifteen and seventeen articulated
fin rays with the leading edge reinforced by three pairs of unarticulated basal fulcra.

The anal fin is longer with some twenty robust articulated rays. The leading
edge is again strengthened by three unarticulated basal fulcra which unlike those in
front of the dorsal fin are unpaired.

The caudal fin is heterocercal and only partially cleft, shorter than in Sphaerolepis.
The rays are stout and articulated and again there are three or four strong, unarticu-
lated basal fulcra preceding it.

The squamation. The scales are rhomboidal and ornamented with fine concentric
ridges of enamel, but the scales immediately around the pectoral fin are almost
cycloidal while those in the anterior flank region have in addition to the fine con-
centric ridges of enamel two or three small backwardly directed spines in the mid-
scale region.

172 PALAEONISCOID FISHES AND THE CHONDROSTEI

LOCALITY AND HORIZON. All the known specimens came from the “ cannel”’ coal
of Tremosna, which is uppermost Westphalian extending into the lowermost Steph-
anian (Floral zones H-I).

Family PHANERORHYNCHIDAE Stensi6 1932
Diaenosis. As for the genus Phanerorhynchus.

REMARKS. This family includes the single genus Phanerorhynchus Gill. The
peculiar nature of its fins suggests a relationship to the Haplolepiformes while the
skull could easily have been derived from that of a primitive haplolepid-like form
by elongation of the snout region. Like the Haplolepiformes the Phanerorhyn-
chidae were probably derived from the earlier Holuridae. Superficially the Phanero-
rhynchidae show considerable similarity to the later Chondrosteiformes and Acipen-
seriformes.

Genus PHANERORHYNCAHUS Gill 1923

Dracnosis. Body fusiform; dorsal fin almost opposite anal fin, both fins
triangular, anal smaller of two, form of caudal fin unknown. Pectoral and pelvic
fins small ; all fins with stout, unjointed rays which never bifurcate. Snout drawn
out as distinct rostrum projecting well beyond anterior limit of lower jaw. Sus-
pensorium almost vertical, opercular more ovoid than rectangular. Orbit small,
dentition feeble or absent. Scales large, dorsal and ventral ridge scales prominent,
ridge scales graduating into very stout basal fulcra (Woodward 1942) in front of
fins (true fulcral scales absent).

TYPE SPECIES. Phanerorhynchus armatus Gill, the only known member of the genus.
Phanerorhynchus armatus Gill

(Text-figs. 22,23)
1915 Imperfectly preserved small fish, Woodward : 73.
1923 Phanerorhynchus aymatus Gill: 465, text-fig. I.
1932 Phanevorhynchus avmatus Gill; Stensio : 78, text-fig. 29.
1936 Phanerorhynchus Gill; Berg : 345.
1939 Phanerorhynchus avmatus Gill; Moy-Thomas: 118, text-fig. 31a.

Diacnosis. Fishes not exceeding 4 cm. in total length, body fusiform, length of
head contained slightly more than three times and greatest depth of body three
times in total body length. Lepidotrichia stout, smooth, unjointed and not bi-
furcated. Scales large, median flank row containing lateral line about three times
deeper than broad. Only six obvious scale rows. Dorsal ridge scales with stout
spines graduating into basal fulcra in front of dorsal fin. Ventral ridge scales
prominent but with less obvious spines.

Hototyre. Manchester Museum, L.8585, an almost complete specimen, minus
tail, in a clay ironstone nodule; the only known specimen of this species. From
the Middle Coal Measures of Sparth, Lancashire.

Description. The skull. The general shape of the skull can be seen from Text-
fig. 22. The anterior end is drawn out into a very distinct rostrum. The supra-
scapulars are large and are joined to a pair of extrascapulars anteriorly. The parietals

PALAEONISCOID FISHES AND THE CHONDROSTEI 173
Ssc

Exsc

Fic. 22. Phanerorhynchus armatus Gill. Reconstruction of skull in lateral view.

are relatively small and square in outline, while the frontals are long with their
anterior margins produced into a point thus giving the postrostral a V-shaped in-
sertion between the two frontals. On either side of the single median postrostral is a
stout, long, nasalelement. The nasalis broad and its lateral margin has an emargina-
tion (naz) which indicates the position of the posterior nostril. The anterior nostril
(na,) lies between the nasal, postrostral and rostro-premaxillo-antorbital. Pos-
teriorly the nasal joins the frontal and the dermosphenotic, anteriorly the rostro-
premaxillo-antorbital. The postrostral is only slightly convex transversely in its
posterior half, but the anterior half of the bone is strongly curved in a transverse as
well as longitudinal direction, giving the head a distinct rostrum. Ventrally the
postrostral meets the paired rostro-premaxillo-antorbitals. The rostro-premaxillo-
antorbital is notched anterio-dorsally (naz) while posteriorly it is ornamented with
several backwardly pointing short spines. Dorsally it articulates with the nasal and
posteriorly with both the maxilla and infraorbital.

The lateral wall of the skull roof comprises two bones, the dermopterotic and the
dermosphenotic. The dermopterotic is broad posteriorly but tapers almost to a point
anteriorly where it fits between the dermosphenotic and the frontal. The dermo-
sphenotic is larger than the dermopterotic and anteriorly meets the nasal. Ventrally

174 PALAEONISCOID FISHES AND THE CHONDROSTEI

the dermosphenotic abuts against the apparently single, long, curved infraorbital
bone. Behind the infraorbital and below and behind the dermosphenotic are two
suborbitals, of which the dorsal is by far the larger.

The maxilla is of the normal palaeoniscoid outline, but its anterior prolongation is
somewhat short ; it is ornamented with concentric striae of enamel. The pre-
opercular is again of the normal palaeoniscoid form and from the inclination of its
posterior border the suspensorium can be seen to be almost upright. The opercular
is large, ovoid in shape and ornamented with sparse longitudinal ridges of enamel.
The subopercular is much smaller and_scarcely larger than the succeeding branchio-
stegalray. At least four branchiostegal rays can be seen, but whether or not a gular
plate is present could not be determined. As already mentioned by Watson (in
Gill 1923) there appears to be a small triangular plate between the top of the pre-
opercular and the dermopterotic.

The lower jaw. The lower jaw is ornamented with longitudinal striae of enamel,
but apart from the mandibular sensory canal with its three lateral tubules running

\ NS a
iS S\ ee Lae
ae NSS wih

\

(SESS
Lie

Smm

Fic. 23. Phanerorhynchus armatus Gill. Restoration of whole fish.

out obliquely towards the surface of the jaw little could be made out. Whether or
not teeth were present could not be determined: if present they must have been
very small.

The paired fins and thew girdles. Of the pectoral girdle, the supracleithrum and
cleithrum are ornamented with concentric ridges of enamel which run parallel to
the margins, nearer the centre of the bones these ridges give way to tubercles. The
supracleithrum and cleithrum are of normal palaeoniscoid proportions and ventrally
there is a pair of stout clavicles. Of the pectoral fin, only the base and one fulcral
scale are to be seen on the specimen, but probably there were four to eight fin rays.
The pelvic fin has a very short base and is composed of four to six rays.

The unpaired fins. Both the dorsal and anal fins are small, situated far back and
almost opposite one another, the dorsal being slightly the larger. Of the dorsal fin,
some six rays can be seen with certainty but the total number was probably nearer
twelve. On the anal fin about nine rays can be counted, but again there may have
been one or two more. Only the base of the slender tail shaft is preserved, with two
very stout fulcra ventrally, followed by three stout fin rays.

PALAEONISCOID FISHES AND THE CHONDROSTEI 175

The squamation. The body scales have been somewhat distorted but the true
arrangement is given in Text-fig. 23. On each of the largest, middle flank scales
there are two distinct tubercles, one near the dorsal edge and the other two-thirds
of the way down: the dorsal one represents the passage of the lateral line. The first
row of scales beneath the dorsal ridge scales also has a large tubercle near the base of
each scale. All these tubercles are long and point posteriorly, often extending beyond
the posterior margin of the scale in which they arise.

The dorsal row of ridge scales is composed of single scales, but the ventral row
consists of paired scales up to the pelvic fin. After the anal fin the ridge scales are
single and unpaired.

LocaLITy AND HORIZON. Sparth, near Rochdale, Lancashire. Middle Coal
Measures, Upper Carboniferous, Ammanian (Anthracoceras Zone A).

Family OSORIOICHTHYIDAE nov.

Diacnosis. Trunk fusiform; mandibular suspension moderately oblique,
pectoral fins with large number of articulated and distally bifurcating rays. Rostrum
well developed, teeth arranged in single series. Opercular small, with accessory
opercular separating it completely from subopercular, all three bones approximately
same size. Branchiostegal rays very numerous, gular large; scales small and
rhomboidal.

Remarks. This family has been erected to include the single genus Osorioichthys
Casier from the Upper Devonian of Belgium. The condition of the opercular appara-
tus separates it from all other described genera with the exception of Rhabdolepis
Troschel (Gardiner 1963). It differs from Rhabdolepis in the more primitive arrange-
ment of the skull roof and in the large size of the accessory opercular in relation to
the opercular. The Osorioichthyidae represents a specialized side line from the
main palaeoniscoid stock.

Family GYROLEPIDOTIDAE nov.

DracGnosis. Body fusiform; dorsal fin arising just anterior to anal fin; both
fins of moderate size; caudal fin powerful, deeply cleft, equilobate. Paired fins
moderate to large. All fins with large, articulated and distally bifurcating rays, and
stout fulcra anteriorly. Skull rounded anteriorly without well developed rostrum ;
suspensorium only slightly inclined, opercular high and subopercular somewhat
reduced ; dermohyal large, but no accessory opercular present. Dermopterotic long,
suborbital series well defined, teeth stout and in single series. Large dorsal ridge
scales running from occiput to dorsal fin and from dorsal fin to caudal extremity ;
large ventral scales extending from anal fin to base of caudal.

Remarks. This family includes Gyrolepidotus Rohon and Palaeobergia Matveeva
from the Lower Carboniferous of Russia. It belongs to a closely related complex of
families including the Carbovelidae, Gonatodidae nov, Cosmoptychiidae, Acro-
lepididae, Elonichthyidae, Rhadinichthyidae, Canobiidae, Pygopteridae, Rhab-
dolepididae, Styracopteridae and Cryphiolepididae. All possess many features in

GEOL, 14, 5. 18

176 PALAEONISCOID FISHES AND THE CHONDROSTEI

common and were derived from the same stock. The Gyrolepidotidae however can
be separated from any one of these families by the stout nature of the fin rays, the
very large fulcral scales and the large size of the dermohyal.

Family AMPHICENTRIDAE Moy-Thomas 1939 : II5

DraGnosis (emended). Body very deeply fusiform ; dorsal and anal fins triangu-
lar and long based, caudal fin heterocercal, inequilobate and deeply cleft : lepido-
trichia of all fins jointed and distally bifurcated ; fulcral scales large. Suspensorium
almost vertical, orbit small, opercular smaller than subopercular, maxilla triangular
and massive. Gular narrow, flank scales deeper than broad; dentition modified
for crushing. Ectopterygoids and coronoids with robust teeth.

RemArRKS. This family includes the following genera: C/ivodus M’Coy, Cheiro-
dopsis Traquair, Eurynothus Agassiz, Eurynotoides Berg, Paraeurynotus Chabakov,
Globulodus Munster, Protoeurynotus Moy-Thomas & Dyne and possibly Tompaichthys
Obruchev. Moy-Thomas (1939 : 115) first recognized that the platysomids could be
separated into two distinct families and within the Amphicentridae he included
Amphicentrum Young, Protoeurynotus Moy-Thomas & Dyne, Eurynotus Agassiz
and Cheirodopsis Traquair. However Amp/licentrum Young 1866 is synonymous
with Chirodus M’Coy 1848 (Dyne 1939 : 195).

Family ATHERSTONIIDAE nov.

Diacnostis. Body fusiform ; dorsal fin arises well in front of anal fin ; both fins
triangular. Pectoral and pelvic fins large, anterior rays of pectoral unarticulated
proximally. All fins with numerous small fulcra anteriorly and rays closely arti-
culated but not distally bifurcated. Skull with moderately developed rostrum and
oblique suspensorium. Opercular much larger than subopercular. Scales rhom-
boidal with pronounced ridges of enamel; large ridge scales running from occiput
to dorsal fin.

Remarks. The family includes the single genus Atherstonia Woodward which
ranges from the Upper Permian to the Upper Triassic. The Atherstoniidae are close
to the Trissolepididae, particularly in the make-up of the fins in which the rays are
not bifurcated distally. They differ from the Trissolepididae in the presence of
fulcral scales, in the absence of basal fulcra and in the more oblique suspensorium.

Family LAWNIIDAE nov.

DiaGnosis. Body fusiform ; dorsal fin large, arising anterior to smaller anal fin ;
caudal fin powerful, deeply cleft and inequilobate. Paired fins stout, of moderate
size. Lepidotrichia of all fins closely articulated and distally bifurcated ; small
fulcral scales present along leading edges of fins. Skull without well-developed
rostrum, suspensorium only moderately oblique, opercular and subopercular of
approximately same size. Preopercular high, with small dermohyal between it and
opercular. Posterior nasal aperture (nostril) completely enclosed posteriorly by
antorbital (a specialization not normally seen below holostean grade). Teeth
stout and in single series. Ridge scales present in front of unpaired fins.

PALAEONISCOID FISHES AND THE CHONDROSTEI 177

REMARKS. The Family Lawniidae is used to include a single fresh water genus
Lawma Wilson (1953), from the Permian of Texas. In body shape, disposition of
fins and scale structure this family resembles both the Amblypteridae and the earlier
Gonatodidae, but the specialized arrangement of the snout in the Lawniidae clearly
separates it from either of these families.

Family COSMOLEPIDIDAE nov.

Diacnosis. Trunk elegantly fusiform ; dorsal and anal fins triangular, with pos-
terior rays very short. Dorsal fin opposed to space between pelvics and anal, caudal
fin deeply forked, inequilobate and with upper lobe much attenuated. All fins of
moderate or small size with broad, articulated rays, distally bifurcating. Rays of
pectoral fins unarticulated proximally, minute fulcra on all fins. Mandibular sus-
pensorium oblique, teeth consisting of inner series of stout conical laniaries and outer
series of more numerous smaller teeth. Supraorbital sensory canal unites with
infraorbital canal. Scales thick and small.

REMARKS. The Cosmolepididae is another monogeneric family, containing the
Lower Liassic Cosmolepis Egerton. This family resembles the earlier Palaeoniscidae
in general body form, but can be distinguished from it by the greater length of the
jaws, the more oblique suspensorium and the structure of the scales. A diagnostic
character is that the supraorbital sensory canal unites with the infraorbital canal, a
condition seen in only one other chondrostean (Brookvalia Wade).

Family BRACHYDEGMIDAE nov.

Diacnosis. Skull without pronounced rostrum, suspensorium almost upright,
orbit small ; opercular and preopercular of comparable size, preopercular high and
with several dermohyals between it and opercular series. Branchiostegal rays
numerous and suborbital series present. Maxilla stout, with very large teeth on
anterior half. Large teeth in single series on dentary and rostro-premaxillary.
Form of body and fins unknown.

REMARKS. This family is used to include the single genus Brachydegma Dunkle
(1939), from the Texas Permian. The skull is stout and quite unlike that of any
other known palaeoniscoid. On the evidence of scale structure and the shape of the
preopercular the Brachydegmidae appear to be nearer the Amblypteridae than any
other described family.

Family BOREOSOMIDAE nov.

Diacnosis. Body fusiform; dorsal fin arises in advance of pelvics; anal fin
small, remote ; caudal fin heterocercal, moderately to deeply cleft. Lepidotrichia
of fins articulated, distally bifurcated ; fulcra present on all fins. Skull with pro-
nounced rostrum, almost upright suspensorium, high preopercular, and both dermo-
hyals and suborbitals present.

178 PALAEONISCOID FISHES AND THE CHONDROSTEI

ReMARKS. The family is erected to include the wide-ranging genus Boveosomus
Stensid, while Mesembroniscus Wade from the Trias of Australia also possibly
belongs here. The Boreosomidae show some affinities to the Palaeoniscidae but are
distinguishable from the Palaeoniscidae by the more vertical suspensorium, the high
preopercular, the shape of the maxilla, the shorter jaws and the anterior postion of
the dorsal fin.

Order PELTOPLEURIFORMES
Family HABROICHTHYIDAE nov.

Diacnosis. Body fusiform ; dorsal fin arising in front of anal fin, both fins small
and triangular. Caudal fin strongly forked and superficially homocercal ; paired
fins small. All fins comprise few, stout rays bifurcated distally and (except in
caudal) unjointed proximally ; fulcral scales absent. Skull large, suspensorium
upright, opercular at least twice as high as subopercular ; maxilla palaeoniscoid in
shape, preopercular large, with vertical posterior margin. Dentition feeble, scaling
reduced to single row of greatly deepened scales on flank. Posteriorly this scaling
terminates in enlarged, symmetrical, semicircular scale.

Remarks. This family is erected to include the single genus Habroichthys Brough,
which is readily distinguished from the Peltopleuridae by the presence of the single
row of greatly deepened scales on the flank.

III. DISCUSSION
Subclass CHONDROSTEI

Although the first undoubted members of the class Actinopterygii are found in the
Middle Devonian, it seems probable that certain isolated scales recorded from the
Lower Devonian (personal observation) may belong to this class. Comparatively
rare at first, the actinopterygians flourished, until today they are the most numerous
of the vertebrates. The actinopterygians, like the crossopterygians and dipnoans,
probably had their origin in salt water (White 1958). However, by the Upper
Devonian the actinopterygians were to be found in both freshwater and marine
environments and it seems likely that the primitive actinopterygians were tolerant to
both salt and freshwater conditions. Only by this assumption can the world-wide
distribution of the palaeoniscoids in the Upper Devonian and Lower Carboniferous
be explained. By the beginning of the Mesozoic Era the actinopterygians were
almost entirely marine and this has been the main centre of their evolution ever since.
Today the vast majority of teleosts are marine and of the freshwater forms many
have returned secondarily to this environment from salt water (Romer 1945).

For the sake of convenience, the actinopterygians are divided into four groups
which are given Subclass status, in ascending order: the Chondrostei, the Holostei,
the Halecostomi and the Teleostei. These groups represent grades of evolutionary
development rather than natural subdivisions. During the Palaeozoic the evolution
of the Actinopterygii consisted of modifications of the original basic chondrostean

PALAEONISCOID FISHES AND THE CHONDROSTEI 179

design and resulted mainly in abortive attempts to reach the holostean level of
organization. In the past it has been customary to group all these [abortive]
attempts, together with the more successful ones, into the Subclass (or order)
“ Subholostei ”’ (Brough 1936, 1939), but this has been abandoned since it not only
cuts completely across phyletic lines but was based on adaptive characters. The
same criticism can be levelled at the Holostei (which include two distinct lineages)
but this group, although not a monophyletic one, includes members which at least
have all attained the same grade of structural organization.

The Subclass Chondrostei contains all the Palaeozoic actinopterygians (with one
late Permian exception) and although still well represented in the early Mesozoic, its
members dwindled rapidly in the late Triassic (as they were replaced by the better
adapted holosteans to which they had given rise) until today there are some eight
degenerate survivors (Acifensey Linnaeus, Huso Brandt, Scaphirhynchus Heckel,
Kessleria Bogdanon, Polyodon Schneider, Psephurus Guenther, Erpetoichthys Smith
and Polypterus Lacépede.) from two orders (Acipenseriformes, Polypteriformes).

The Holostei are the characteristic actinopterygians of the Mesozoic ; the earliest
member is recorded from the late Permian (Gill 1923a) while today there are but two
survivors from two distinct orders, the Semionotiformes (Lepzsosteus) and the
Amiiformes (Ama).

The Halecostomi enter the fossil record in the early Mesozoic (Rayner 1937 ;
Gardiner 1960) and evolved contemporaneously with the Holostei: they disappeared
at the end of the Mesozoic, having given rise to the teleosts.

The Teleostei appeared in the Upper Jurassic: the majority of the present day
orders can be distinguished in the Upper Cretaceous, while the majority of modern
families seem to have arisen in the Eocene or later.

The early actinopterygians are separated from the crossopterygians and dipnoans
by their possession of so-called “‘ ganoid scales’’. However, these have undergone
considerable modification and reduction in the later members, leaving a relatively
thin, simple scale. The “ ganoid scale’ consists of the same three basic units as in
the “ cosmoid scale’’, a basal portion of bone arranged in parallel layers and an
upper portion of dentine (cosmine) capped with layers of enamel (ganoine) : separat-
ing the two is a spongy layer representing blood vessel plexuses. The central unit
of the “ ganoid scale’ is equivalent to the basic unit of the “‘ cosmoid scale ’’ and
the remainder of the scale arises by the deposition in onion-like fashion of concentric
layers of all three materials (bone, dentine and enamel) around this central unit. In
the light of the “ lepidomorial theory ”’ (Jarvik 1960 ; Stensid 1961) it would appear
that the “ ganoid scale ”’ is more complex than the “‘ cosmoid ”’ in that units, instead
of being added to the edges, arise in complete rings. In other words, the primordia
which go to make up the scale have reached a much higher level of fusion in the
“ ganoid scale ”’.

In most Mesozoic forms the scales lose the middle dentine layer and the enamel
covering becomes thin in many holosteans and halecostomes, disappearing completely
in the teleosts, while in some cases the entire scaly covering may be reduced or lost
(pycnodonts and some teleosts). Most of these changes are possibly linked with a
need of greater flexibility for more efficient swimming.

180 PALAEONISCOID FISHES AND THE CHONDROSTEI

The “ ganoid scale ’’ of the early actinopterygian Cheirolepis is very close in struc-
ture to the scales of acanthodians (Gross 1947) and it would seem that the actinop-
terygians and acanthodians were derived from the same ancestral stock.

The actinopterygians, by virtue of several evolutionary bursts, are a large and
diverse group and it is not easy to give a comprehensive definition of such a varied
assemblage. However they differ from the majority of crossopterygians and dip-
noans (with the exception of the coelacanths) in the absence of internal nostrils.
Both nostrils are borne high on the face in primitive forms and the two nares of either
side are always separated by the supraorbital sensory canal, although in subsequent
evolution the two nostrils become confluent (Gardiner 1963). They differ further
from the crossopterygians and dipnoans in the arrangement of the sensory canals
of the head, the preopercular canal rarely if ever uniting with the postorbital portion
of the infraorbital canal as it does in all crossopterygians. Other points of contrast
are to be seen in the nature of the paired and unpaired fins. In the actinopterygians
the internal skeleton of the unpaired fins is never concentrated into basal plates but
normally consists of separate radials. The paired fins primarily do not have the large
fleshy lobes seen in the crossopterygians and dipnoans, but instead the entire fin
web is supported by long-based, flexible lepidotrichia which do not have the radials
concentrated at their bases. In the more advanced teleosts these fins become much
narrower at the point of insertion and consequently much more flexible in their
movements. There is normally a single dorsal fin in contrast to the two seen in
crossopterygians.

The arrangement of the dermal bones of the head in the early actinopterygians is
very similar to that seen in the crossopterygians.

Order PALAEONISCIFORMES

The earliest actinopterygians belong to the Order Palaeonisciformes. This order
had its beginnings in the Lower Devonian and the first undoubted members are
recorded from Middle Devonian freshwater and marine deposits (Guppy, Lindner,
Rattigan & Casey 1958). The Palaeonisciformes attained their maximum develop-
ment during the Carboniferous and Permian when they were the commonest fresh-
water fishes, while many had also invaded the seas. They were world-wide in
distribution during the Palaeozoic Era, being recorded from Great Britain, many
parts of Europe, Greenland, Antarctica, Northern Asia, South and East Africa,
South and North America and Australia. Of the thirty-nine families so far recog-
nized, twelve survived into the early Mesozoic, but by the Lower Jurassic only three
families remained. However, these three families retained the basic palaeoniscoid
condition practically unaltered to the last. One of these three is the Family Cocco-
lepididae, containing the single genus Coccolepis, which ranges upwards from the
Lower Lias (Coccolepis liassica) to the Wealden (Coccolepis macroptera) and includes
the last members of the Palaeonisciformes. The Palaeonisciformes as a group
declined rapidly at the close of the Permian, and by the Middle Triassic they had
ceased to be a major constituent of the fish fauna. This very rapid decline can be
correlated with the advent of more highly evolved groups of chondrosteans (Perleidi-
formes, Saurichthyiformes, Pholidopleuriformes, etc.) in the early Triassic.

PALAEONISCOID FISHES AND THE CHONDROSTEI 181

The Palaeonisciformes represent the basal stock of the actinopterygians and from
them all the other actinopterygians evolved.

Basically the palaeoniscoids form a closely-knit group, with most of the members
very similar in make up. Owing to inaccurate descriptions it has not hitherto been
possible to group the palaeoniscoids effectively in families. However, in the light
of more recent researches one may at least attempt to group most of the genera with
some assurance. Thus the Devonian palaeoniscoids fall into four distinct families—
the Cheirolepididae (Chetrolepis), the Stegotrachelidae (Stegotvachelus, Moythomasia,
etc.), the Tegeolepididae (Tegeolepis) and the Osorioichthyidae nov. (Osorioichthys).
However, although all these families are fairly generalized, none of them gave rise
to any of the other palaeoniscoid families. The most typical palaeoniscoids belong
to the families Elonichthyidae, Acrolepididae, Palaeoniscidae and Coccolepididae,
and these families belong to the central stem of palaeoniscoid evolution, one which
led to ultimate extinction in the Cretaceous. These families illustrate the funda-
mental palaeoniscoid plan. There are, however, many variations on this theme,
and some of these were to give rise to better adapted chondrostean orders in the late
Palaeozoic and early Mesozoic. From some of these more advanced chondrostean
orders the later holosteans and halecostomes were derived.

After the Lower Trias the occurrence of palaeoniscoids in marine beds is rare and
it appears that about the beginning of their decline the palaeoniscoids forsook the
seas and entered fresh waters where they had few if any competitors (many of the
more advanced chondrosteans such as the redfieldiids also returned to fresh water
during the Lower Triassic). By the Middle Trias they are more conspicuous in
fresh-water than in marine deposits. In the fresh-water Middle Triassic beds of
Australia there are some seven or eight recorded palaeoniscoid genera (Wade 1935),
while in similar beds in South Africa and North America several new genera occur
(Woodward 1889 ; Broom 1913 ; Gardiner 1966).

The Cheirolepididae, containing but the single genus Cheivolepis Agassiz, is a
fresh-water family (Middle-Upper Devonian), not far removed from the more normal
palaeoniscoid condition except in the possession of minute scales which do not overlap
and which closely resemble those of the acanthodians. Another such monogeneric
family is the Osorioichthyidae nov. from the Devonian of Belgium (Casier 1952, 1954),
characterized by a large accessory opercular (in this respect it resembles the later
Rhabdolepididae). The Tegeolepididae, which range from the Upper Devonian of
North America (Newberry 1888) to the Triassic of Australia (Gardiner 1963), have
no fulcral scales, unarticulated rays to the pectoral fins, and small thin scales. The
remaining Devonian family, the Stegotrachelidae, is known from marine and fresh-
water deposits of Middle and Upper Devonian age and continues into the Lower
Carboniferous (Gardiner 1963). It is very close to the main palaeoniscoid stem
which was to give rise to the majority of the Carboniferous forms but is exceptional in
possessing a pineal foramen. All these four families which arose in the Devonian are
end lines independently derived from some hitherto as yet unknown ancestor(s).
On the other hand the bulk of the Lower Carboniferous palaeoniscoids form a closely
related complex consisting of some twelve families. These are the Carbovelidae,
Gonatodidae nov., Gyrolepidotidae nov., Cosmoptychiidae, Acrolepididae, Elonich-

182 PALAEONISCOID FISHES AND THE CHONDROSTEI

thyidae, Rhadinichthyidae, Canobiidae, Pygopteridae, Rhabdolepididae, Styracop-
teridae, and Cryphiolepididae. All possess many features in common and were
clearly derived from the same basal stock (one not far removed from the Stego-
trachelidae). They all closely adhere to the typical palaeoniscoid condition. From
this complex a further thirteen families were subsequently derived—the Brachy-
degmidae nov, Coccocephalichthyidae, Boreolepididae, Boreosomidae nov, Palaeonis-
cidae, Centrolepididae, Coccolepididae, Cosmolepididae nov, Amblypteridae, Com-
mentryidae, Dicellopygidae, Aeduellidae and Lawniidae nov.

The Brachydegmidae nov, Boreolepididae and Coccocephalichthyidae are all too
poorly known to establish relationships.

The Palaeoniscidae, Boreosomidae nov, Cosmolepididae nov, and Centrolepididae
all stemmed from an ancestor not too far removed from the Acrolepididae while the
Coccolepididae was probably derived from the Palaeoniscidae.

The Amblypteridae, Aedullidae, Commentryidae, Dicellopygidae and Lawniidae
nov are all related to the Gonatodidae.

The Family Palaeoniscidae includes some ten genera, ranging in time from the
Upper Permian to the Upper Trias, and all of which retain the basic palaeoniscoid
condition. They are typically fusiform fishes with strongly heterocercal tails,
rhomboidal, enamel-covered scales and an oblique suspensorium. The earliest genus
is Palaeoniscum Blainville (Westoll 7m Aldinger 1937) which ranges from the Upper
Permian to the Lower Trias. Pteronisculus White (Nielsen 1942) from the Lower
Trias of Madagascar, Spitsbergen and Greenland is another typical member, while
some of the freshwater palaeoniscoids recorded from the Middle Trias of Australia,
such as Agecephalichthys Wade, Myriolepis Egerton and Belichthys Wade (Wade
1935), are also included tentatively in this family. The Palaeoniscidae are close to
both the Elonichthyidae and the Acrolepididae, and represent a very conservative
stock. The body form, fins and scaling are essentially similar in all three families.
The Palaeoniscidae, however, differ from the Elonichthyidae in the development of
the cranial roofing bones, in the dorso-anterior development of the preopercular, the
enlargement of the suborbital series, the lengthening of the jaws and in the structure
of the scales (Aldinger 1937). The Palaeoniscidae differ from the Acrolepididae
in the development of the cranial roofing bones, in the more ventral position of the
nares, in the greater fragmentation of the infraorbital and suborbital series and in the
structure of the scales (Aldinger 1937). Of the two families Elonichthyidae and
Acrolepididae, probably the Acrolepididae gave rise to the Palaeoniscidae.

The Centrolepididae contains a single genus Centrolepis Egerton which is confined
to the marine Lower Lias (Gardiner 1960). It is a fusiform fish with an elongate
body, and the fins are of the normal palaeoniscoid pattern with the dorsal opposed
to the space between the pelvics and anal. The scales are thick, enamelled and
highly ornamented, the suspensorium very oblique and the rostrum pronounced.
Apart from the absence of a dermohyal the skull is not dissimilar to that of Rhadinich-
thys or Palaeoniscum. In most respects this family has retained the conservative
palaeoniscoid structure to the last. It arose from the same stock that gave rise to
the Palaeoniscidae but itself gave rise to no other forms.

The Cosmolepididae nov is another Lower Liassic marine family, with but one

PALABONISCOID FISHES AND THE CHONDROSTELI 183

? +Polypteriformes

ee Saurichthyiformes

ae a Chondrosteiformes ——————— Acipenseriformes

/-Phanerorhynchidae é
Birgeriidae

Yi Haplolepiformes

Urosthenidae

Cheirolepididae hs A rissolepid idae
6 as Atherstoniidae

/ _-— Tarrasiiformes fo a Sconile pid dae
J Aeduellidae ; :
Holuridae = Parasemionotiformes

—-Lawniidae
see +Dicellopygidae

Tegeolepididae VE
nee Pholidopleuriformes
Cryphiolepididai / /Z-Commentryidae Pp
PypnSepaiade la y Redfieldiiformes

Carbovelidae YZ
(sie a Sreternes:
y

Gonatodidae —————- Amblypteridae ________<__—_-+-Perleidiformes——Platysiagiformes
//asmortycidce << Peltopleuriformes
Y_-_.-Acrolepididae ; Cephaloxeniformes
+Elonichthyidae ——__ Palaeoniscidae - Coccolepididae
SX Rhadinichthyidae hi = ~Cosmolepididae
\\\SosCanobiidae pape Centrolepididae
\Pygopteridae = _—S¥——————-Boreosomidae —Ptycholepiformes
Rhabdolepididae ~ ———Boreolepididae

sty racopteridae ?— Coccocephalichthyidae
latysomidae ? —__. Brachydegmidae

Amphicentridae :
E ee Dorypteriformes

*Cornuboniscidae

Bobasatraniiformes

Osorioichthyidae
Stegotrachelidae

Fic. 24. Phylogeny of the Chondrostei.

representative, Cosmolepis Egerton. One of the characteristics of Cosmolepis is that
the supraorbital sensory canal posteriorly does not enter the parietal but unites with
the infraorbital canal (Watson 1925) (a similar condition is seen in Brookvalia Wade,
Acentrophorus Gill and teleosts). The body is elongate, the dorsal and anal fins
long-based and the scales small. It differs from the Palaeoniscidae in the greater
length of its jaws and suprascapular, the more oblique suspensorium and the scale
structure (Gardiner 1960). In general this family is close to Palaeoniscum and, like
the Centrolepididae, shared a common ancestor with the Palaeoniscidae.

Finally the Family Coccolepididae (Berg 1940) which ranges from the Lower Lias
to the Wealden comprises a single genus Coccolepis. The coccolepids are slender
fusiform fishes with heterocercal tails, deeply forked in most forms but inequilobate.
The suspensorium is very oblique, the scales are thin and deeply imbricating but the
overhanging rostrum has been reduced and lost and there is no suborbital series
present (Gardiner 1960). Coccolepis is very similar to the earlier Palaeoniscidae
but differs in the absence of a suborbital series and in the more rounded outline of the
opercular series. It was directly derived from the Palaeoniscidae.

These last three families all represent end lines which managed to survive com-
paratively unchanged from the Palaeozoic and which eventually gave way to the
far better adapted holosteans.

The Family Boreosomidae nov ranges from the Permian to the Middle Trias
(Gardiner 1966a). The suspensorium is almost vertical but there is still a pronounced
rostrum (Lehman 1952). The tail is strongly forked and the dorsal fin is situated well
forward with its anterior border in front of the pelvics ; the orbit is large. Boreo-
somus Stensio itself was an active midwater to surface swimmer, presumably plankton

184 PALAEONISCOID FISHES AND THE CHONDROSTEI

feeding. The affinities of this family are not at all clear, although it seems possible
that it is distantly related to the Palaeoniscidae and probably had its origin in the
Acrolepididae. However, what is more certain is what the family gave rise to ;
Brough (1939) has clearly demonstrated the affinities of the family to the Ptycho-
lepiformes, and there can be little doubt that this later chondrostean order was
derived from the Boreosomidae.

The Amblypteridae contains four genera ranging from the Upper Carboniferous to
the Lower Permian. Its members are characterized by the large paired fins, upright
suspensorium and the single series of teeth in the jaws (Gardiner 1963). The family
is derived from the earlier Gonatodidae nov. The Amblypteridae have many features
in common with the earlier Gonatodidae, in particular the height of the preopercular,
the straightening of the suspensorium and the single series of teeth in the jaws. The
Amblypteridae differ from the Gonatodidae in that the suspensorium is nearer the
vertical, the branchiostegal rays are fewer in number and the head is somewhat
deeper. Within the Gonatodidae the genera Pseudogonatodus nov. and Drydenius
Traquair show how the maxilla has grown down around the peg-like teeth, and so lead
on to the condition seen in the later genus Paramblypterus Sauvage (Amblypteridae)
where the maxilla almost encases the teeth. A similar sheathing of the teeth is to be
seen in Aeduella Westoll (Blot & Heyler 1963) which adds weight to the conclusion
that the Aeduellidae were also derived from the Gonatodidae.

The Family Aeduellidae is made up of but two genera, Aeduella and Westollia
White & Moy-Thomas, both from the marine Permian. Both are characterized by
an upright, narrow preopercular and a maxilla which has no posterior expanded
blade (Westoll 1937). The Aeduellidae resemble the earlier Gonatodidae in body
shape, disposition and make-up of the fins, and in the possession of a high preoper-
cular. The Aeduellidae differs from the Gonatodidae in that the suspensorium is
inclined forward, in the absence of a suborbital series (although anamestic frag-
mentation of the preopercular is occurring) and in the reduction of the posterior
blade of the maxilla.

The Commentryidae (Gardiner 1963) with its single genus has a more typical
palaeoniscoid dentition with a double tooth series including an outer row of larger
teeth (Blot 1963). There are several parietal ossifications present and the dorsal fin
is situated far back with the anal fin more anteriorly placed. This family, like the
Amblypteridae, stemmed from the Gonatodidae. The Commentryidae differ from
the Gonatodidae in the more posterior insertion of the dorsal fin, in the fragmentation
of the parietals and extrascapulars, and in the possession of a double tooth series.

Close to the Commentryidae is another monogeneric family, the Dicellopygidae,
from the fresh-water Lower Trias of Bekker’s Kraal, South Africa (Brough 1931).
The suspensorium is nearer the vertical and the caudal fin strongly forked and almost
equilobate, but the general make-up of the skull is comparable to that of the Com-
mentryidae. The scales are relatively large but the maxilla and cheeks are typically
palaeoniscoid. This family represents a fresh-water side line.

From fresh-water deposits of Permian age in North America comes yet another
monogeneric family, the Lawniidae nov (Wilson 1953). In general Lawnia resembles
the Amblypteridae, apart from the rather specialized arrangement of its antorbital in

PALAEONISCOID FISHES AND THE CHONDROSTEI 185

relation to the narial opening and in body shape (which is fusiform). However there
can be little doubt that this family has been independently derived from the earlier
Gonatodidae.

Returning now to the closely related complex of twelve families in the Lower
Carboniferous, there are two other families, the Amphicentridae and the Platysomidae
(Moy-Thomas 1939) which are not so very far removed from this complex and which
unquestionably shared a common ancestry with it. These two families include all the
deep bodied palaeoniscoids, from the Lower Carboniferous up to the Triassic.

In all essential features the Amphicentridae and Platysomidae are almost identical
with the palaeoniscoids making up this related complex of twelve families, but differ
mainly in their body shape. The body is laterally compressed and dorso-ventrally
deepened. The differences between these two families and the other palaeoniscoids
are merely a question of modification as a result of a change in body shape. The
most obvious differences are the shape of the maxilla which is almost triangular, the
jaw suspension which is vertical (as in the families Canobiidae, Amblypteridae,
Aeduellidae, etc.), the short, deep skull and the elongation of the body scales in the
dorso-ventral plane. With the change in slope of the face, the nostrils are borne
much higher up than in most palaeoniscoids. The dorsal fin is often considerably
elongated, while the heterocercal tail is equilobate in many forms, superficially
resembling the homocercal condition seen in more advanced actinopterygians.
These two families undoubtedly arose from the same parental palaeoniscoid stock
and, although their members flourished throughout the Carboniferous and Permian,
only one or maybe two genera survived into the Mesozoic (Caurichthys Broom,
Platysomus Agassiz). The two families are separated mainly on the form of the
dentition. The Platysomidae have small, conical, often pointed teeth while the
Amphicentridae have a more powerful crushing dentition with tooth plates developed
on the coronoids and endopterygoids (Dyne 1939). The Amphicentridae did not
survive the Palaeozoic, but the Platysomidae are represented in the Lower Trias by
Caurichthys and also possibly by the long ranging Platysomus. Both families were
adapted for browsing among the lagoonal coral reefs and finally succumbed to the
rather more highly evolved bobasatranids. It seems probable that the bobasatranids,
with their more powerful crushing dentition, came from the same basal stock as the
Amphicentridae, while both the Amphicentridae and the Platysomidae were derived
from the same ancestral stock which gave rise to the related complex of twelve
families.

An important Carboniferous family which has been independently derived from
the ancestral palaeoniscoid stock is the Holuridae (Moy-Thomas 1939). In this
family the bones of the skull conform to the normal palaeoniscoid condition, but it is
in the make-up of the fins that the Holuridae differ markedly from all the other
palaeoniscoid lineages. The lepidotrichia of all the fins are articulated but not
distally bifurcating and there are vo fulcra present. Further the caudal fin is not
cleft. The Holuridae gave rise to the Phanerorhynchidae, Birgeriidae, Trisso-
lepididae, Atherstoniidae nov, and possibly the Urosthenidae and Scanilepididae,
while the later chondrostean orders, Haplolepiformes, Saurichthyiformes and Chon-
drosteiformes also came of this lineage.

186 PALAEONISCOID FISHES AND THE CHONDROSTEI

The Phanerorhynchidae are known from a single genus, Phanerorhynchus Gill
from the Upper Carboniferous (Gill 1923). The skull is much as in the more typical
palaeoniscoids, apart from the snout which is drawn out into a distinct rostrum
projecting beyond the anterior limits of the lower jaw. The actual arrangement of
the snout bones is almost identical to that seen in the Haplolepiformes (Westoll
1944). The body scaling is peculiar and consists of very large scales with prominent
ridge scales along the dorsal contour, as in the Holuridae. The fin rays are unarticu-
lated and never bifurcated and there are no fulcral scales (as in the Holuridae).
The tail was probably not cleft. This family is not far removed from either the
Haplolepiformes or the Chondrosteiformes.

The Family Birgeriidae also contains only one recorded genus, Birgeria Stensi6
which occurs throughout the marine Trias. It is characterized by the large number
of suborbitals in the cheek region (Nielsen 1949). The opercular is triangular, the
subopercular splint-like and similar to the succeeding branchiostegal rays. The
suspensorium is very oblique. The anterior fin rays are not bifurcated and fulcral
scales are absent. Like the Phanerorhynchidae it was derived from the holurid
lineage.

The Trissolepididae is composed of two genera—Sphaerolepis Frié and Sceleto-
phorus Frié, both from the Upper Carboniferous of Czechoslovakia. The suspen-
sorium is upright and the snout rounded. The fin rays of all the fins are relatively
few in number, articulated but not bifurcating distally. Fulcral scales are absent
and the caudal fin isnot truly cleft. The Trissolepididae in all these points resemble
the earlier Holuridae from which they were derived.

The Family Atherstoniidae nov has been erected to include the genus Atherstonia
Woodward (Upper Permian—Upper Trias). The fin rays in this family are very
numerous, articulated but not distally bifurcating. The family is close to the
Trissolepididae.

The Scanilepididae are a Triassic family, characterized by a very long dorsal fin
containing seventy rays or more (Aldinger 1937). The skull is typically palaeonis-
coid, the suspensorium oblique and it may be related to the earlier Holuridae.

Finally the remaining family of this lineage, the Urosthenidae (Woodward 1931),
containing the single genus Uvosthenes Dana, comes from the Permian of New South
Wales. The fins have no fulcral scales but it would appear that some of the rays are
bifurcated. The absence of dorsal ridge scales on the tail and the peculiar lobed
nature of the unpaired fins make the systematic position of this family exceedingly
doubtful.

There is one final palaeoniscoid family which has not as yet been dealt with, the
Cornuboniscidae (White 1939), containing the single genus Cornuboniscus White from
the Upper Carboniferous of Cornwall. The genus is characterized by the large maxil-
lae which meet anteriorly and by the reduced opercular and preopercular. Since it
does not appear to be related to any of the families described above, an independent
derivation from the ancestral palaeoniscoid stock must be postulated.

Leaving the Palaeonisciformes we move on to more advanced Chondrostean orders
which have been derived from them.

PALAEONISCOID FISHES AND THE CHONDROSTEI 187
Order TARRASITFORMES

The Order Tarrasiiformes includes a single family, the Tarrasiidae, which has only
two genera, Tarrasius Traquair from the Lower Carboniferous of Scotland and
Palaeophichthys Eastman from the Upper Carboniferous of Illinois.

The skull in Tarrasius is identical with that seen in some of the more primitive
Palaeonisciformes (Moy-Thomas & Dyne 1938), while the scales (which are confined
to the posterior region) are similar to those of Cheirolepis Agassiz (Moy-Thomas &
Dyne 1938). However the body is elongated and resembles that of the recent
Polypterus Lacépéde as do the rounded, fleshy lobed, pectoral fins. The dorsal and
anal fins are continuous with the caudal which is diphycercal. The fins have no
fulcral scales and the rays are articulated but not distally bifurcated; in these
respects the Tarrasiiformes resemble the Holuridae. The Tarrasiiformes are related
to the palaeonisciform family Holuridae and must have shared a common ancestry
with that family.

Order HAPLOLEPIFORMES

Again this order is composed of a single family, the Teleopterinidae (Berg 1936,
Westoll 1944), from the Upper Carboniferous of Europe and North America.

The order is characterized by the structure of the fins in which the lepidotrichia are
stout, few in number and not distally bifurcated. Rather peculiar, large, fulcral
scales fringe the fins anteriorly, the cleithrum is considerably expanded ventrally, the
opercular apparatus is small, the branchiostegal rays are reduced, and the gulars
much expanded. The head is broad and short, and anamestic fragmentation of the
preopercular is taking place. From a comparison of the snout and fins it seems
clear that the Haplolepiformes are fairly close to the Palaeoniscoid family Phanero-
rhynchidae and as such were derived from the earlier Holuridae.

Order SAURICHTHYIFORMES

This order was founded (Lehman 7m Grassé 1958) on the single family Belonorhyn-
chidae which contains two genera, Saurichthys Agassiz, in which all Triassic forms
are placed, and Saurorhynchus Reis which contains only two species, both from the
Lias.

Saurichthys is a widely occurring genus found in the marine Lower Trias of Spits-
bergen, Greenland, Madagascar, Europe and North America and in the fresh water
Middle Triassic deposits of Australia. Saurorhynchus occurs in the marine Lower
and Upper Lias of Europe.

These fishes range in size from a few inches to several feet and are elongate, slender
forms with a much produced rostrum (Stensid 1925; Gardiner 1960). The tail is
abbreviate-diphycercal and the dorsal fin is situated far back, above the anal. The
lepidotrichia exceed the endoskeletal supports in number and long slender ribs are
present. Fulcral scales are minute or absent. The squamation is not continuous,
usually only four rows of scales are present, one dorsal, one ventral and one lateral
on either side supporting the lateral line, otherwise the body is naked [Saurorhynchus
brevirostris (Woodward)]. The maxilla is typically palaeoniscoid, firmly attached

188 PALAEONISCOID FISHES AND THE CHONDROSTEI

to the preopercular and quadratojugal, and from within to the ectopterygoid and
dermopalatine. The suspensorium is almost upright, and the opercular apparatus
consists of a single large opercular, the branchiostegal rays being reduced to one or
completely wanting. The dentition consists of well-spaced, large, conical teeth,
with numerous intervening smaller teeth. These well-armed, extremely long jaws
mark the saurichthyids as among the most predaceous of the Triassic actinoptery-
gians. The neurocranium is completely ossified but lacks a basipterygoid proces
(Gardiner 1960). The sensory canal system is essentially palaeoniscoid and the nasal
bone contains two nasal orifices between which the supraorbital sensory canal passes.

The Saurichthyiformes form a degenerating series, closely related both to the
Palaeonisciformes and to the Chondrosteiformes, and like the Chondrosteiformes they
are not far removed from the earlier Phanerorhynchidae. Phanerorhynchus with its
small fins, reduced number of scale rows and pronounced rostrum shows the way by
which the Saurichthyiformes could have been derived from the Palaeonisciformes.
The Saurichthyiformes although successful in the Triassic, never gave rise to any
further forms.

Order CHONDROSTEIFORMES

The Chondrosteiformes, like the Saurichthyiformes, appeared in the Lower Trias
[Evrolichthys (Lehman 1952)] and died out in the Upper Jurassic (Liu & Zhou 1964).
However, they appear to have been restricted to a purely marine habitat. The
Chondrosteiformes show reduction in both body scaling and skull bones. The scaling
is rudimentary and the pectoral fin is devoid of fulcral scales and without articulations.
The rostrum is moderately to well developed. The maxilla and opercular bones
are reduced, the suspensorium still somewhat backwardly inclined and the supra-
scapular is much elongated. The tail is heterocercal with a well developed scaly lobe.
The unpaired fins are typically palaeoniscoid, the rays more numerous than their
supports. The best-known member is Chondrosteus Egerton from the Lias of
Europe. In this form the mouth is withdrawn behind the projecting rostrum and
was probably suctorial as in the Recent sturgeons.

The Chondrosteiformes, like the Saurichthyiformes, were probably derived from
the earlier Holuridae (possibly via the Phanerorhynchidae) but unlike the Saurich-
thyiformes they went on to give rise to more recent groups. The Acipenseriformes
were derived from the Chondrosteiformes.

Order ACIPENSERIFORMES

The Order Acipenseriformes includes two distinct families, the Acipenseridae and
the Polyodontidae. Both these families first occurred in the Upper Cretaceous and
are represented today by several genera.

The Family Acipenseridae (sturgeons) is widespread today in both salt and fresh
water. The best-known genus is Acipensey Linnaeus which is first found in marine,
Upper Cretaceous deposits. Reduction in ossification has continued from the
condition seen in the Chondrosteiformes until little or no ossification of the internal
skeleton remains. The scales, as in the Saurichthyiformes, have been reduced to a

PALAEONISCOID FISHES AND THE CHONDROSTEI 189

few rows of large bony scutes, but the fins are still essentially palaeoniscoid in struc-
ture and the tail is heterocercal. Acipenser is a bottom feeding scavenger, picking up
molluscs, crustaceans, etc.

The Family Polyodontidae or paddle fishes first occurred in the Upper Cretaceous of
Montana [Palaeopsephurus MacAlpin (1947)]._ Today members are found in river
systems in North America and China (Polyodon Schneider and Psephurus Guenther
respectively). In this family the rostrum has become very elongated and tactile.
The eyes are small and above the anterior end of the upper jaw. The primary jaws
are very large and the hyomandibular oblique. The opercular has been lost and the
subopercular, still large in Palaeopsephurus, is much reduced in the living Polyodon.
The tail is heterocercal and the unpaired fins much as in the Acipenseridae. The
skull bones and body scaling are much reduced. As already stated, this order was
derived from the earlier Chondrosteiformes.

Order POLYPTERIFORMES

This order is only known from two Recent genera, Polypterus Lacépede and
Erpetoichthys Smith which are confined to the rivers and swamps of tropical Africa.

It has been fashionable of recent years (Lehman im Grassé 1958) to separate the
Polypteriformes from the Chondrostei and place them in their own Subclass, the
Brachiopterygii. However, they retain so many obviously palaeoniscoid features
that I do not consider this justified. The scales are typically “ ganoid ”’, the sensory
canal system is much as in the higher chondrosteans, and in some respects as in the
holosteans. The preopercular is still very much palaeoniscoid in make-up, although
its intimate connection with the maxilla has been lost. The snout is primitive and
the large number of individual bones above the preopercular is yet another chondo-
strean character. Polypterus shows many discrepancies from the normal chondro-
stean pattern, especially in the nature of the pectoral fins, the dorsal fins and the
tail, but the chondrostean Tarvasius from the Lower Carboniferous possesses a
continuous dorsal fin and a diphycercal tail. Polypterus is merely a much modified
chondrostean survivor for which, unfortunately, the connecting links are as yet
missing.

Order PERLEIDIFORMES

The Perleidiformes represent a big step forward in chondrostean evolution, and
in many respects resemble the contemporaneous holosteans. They were a very
successful order, containing some twenty-two or more genera from three families.
The earliest members are recorded from the Lower Triassic and after a brief but
interesting history the order died out at the end of the Trias due to increasingly
unsuccessful competition with the more advanced holosteans.

The origins of this order are somewhat uncertain although it would seem that it was
derived from the Amblypteridae. Like the Amblypteridae the Perleidiformes have a
high preopercular, which in both Paramblypterus and Perleidus (Stensi6 1921) shows
anemestic fragmentation dorsally. The Perleidiformes still possess typically
“ ganoid ”’ scales but the general structure of the more advanced members is much

190 PALAEONISCOID FISHES AND THE CHONDROSTEI

closer to the holostean than to the palaeoniscoid condition. The heterocercal tail is
modified to a hemiheterocercal condition while the lepidotrichia of the unpaired
fins come to equal their endoskeletal supports in number.

The Family Colobodontidae (Stensi6 1916) was essentially a marine one, but with
a few fresh-water members. In this family the lepidotrichia of the unpaired fins
equal the endoskeletal supports in number and are only articulated distally, the
proximal portion being entire. The rays are stiff and bifurcated distally. The
fins closely resemble those of holosteans. The suspensorium is upright and the
preopercular high and large. Colobodus Agassiz from the Middle to Upper Trias
was a large, marine form, reaching a length of between two and three feet, and
probably a bottom dweller like the present day Cod. Its dentition consists of partly
pointed and partly crushing teeth, and this type of dentition marks it off sharply
from the Redfielduformes. Perleidus Deecke ranges throughout the marine Trias
and is a more typical member of this family, while Dollopterus Abel, Thoracopterus
Broom, Gigantopterus Abel and Albertonia Gardiner (1966) on the other hand, all
possess enlarged pectoral fins, suggestive of the modern teleostean flying fishes.

The Family Aetheodontidae (Brough 1939) shows some similarity to the Colobo-
dontidae, especially in the form of the dentition, the skull roof, opercular apparatus
and preopercular, and it would appear logical to include it here in the Perleidiformes.
There is a single genus Aetheodontus Brough which occurs in the marine Middle to
Upper Trias (Brough 1939). The suspensorium is vertical, the dorsal and pelvic fins
remote and the rays few in number. The tail is hemiheterocercal with a short
scaly lobe, and the scales are small, stout and numerous.

Finally the Family Cleithrolepididae which is both marine and fresh-water and
confined to the Trias. Members are recorded from the Lower Trias of Australia, the
Middle Trias of South Africa, the Middle and Upper Trias of Germany and the Upper
Trias of England. The fins are much as in the colobodontids except that in the
anterior portion of the dorsal fin in Cleithrolepidina Berg the lepidotrichia outnumber
their supports (Brough 1931). The chief characteristics of this family are the very
much deepened form of the trunk and the weak mandible which bears minute teeth.
The opercular is smaller than the subopercular, the suspensorium is upright and the
preopercular is high, similar to that of some colobodontids (Mezduchthys Brough).
The family retains some palaeoniscoid characters in common with the Colobodon-
tidae and in the present state of knowledge can most usefully be associated with the
Colobodontidae and the Aetheodontidae in the order Perleidiformes.

Order PELTOPLEURIFORMES

This order shows some points of similarity to the Colobodontidae (cf. Meridensia
Stensi6 (r921)) and clearly came from the same ancestral stock. Both Peltopleurus
Kner and Meridensia have a similar body shape, almost identical fins and comparable
skulls. Peltoplewrus differs in the great elongation of its flank scales and in the
make up of the snout (Brough 1939). The order contains two families, both of which
are confined to the marine Upper Trias. Its members are small fishes with an up-
right suspensorium, large orbits and a hemiheterocercal tail. The tail is strongly

PALAEONISCOID FISHES AND THE CHONDROSTEI 191

forked, almost symmetrical externally, and with a reduced scaly lobe. The unpaired
fins are small with relatively few rays and the rays are unjoined proximally (apart
from the caudal) but distally bifurcated. The bones of the head are essentially
palaeoniscoid, particularly the maxilla, but the suspensorium is upright. The
opercular is large and the preopercular is high and shows anamestic fragmentation,
as in the Colobodontidae. The dentition is weak, the scales few in number and those
of the flank greatly elongated dorso-ventrally as in the Cephaloxeniformes.

The Family Peltopleuridae (Brough 1939) contains two genera, Peltopleurus
and Placopleurus Brough from the Upper Trias of Besano. These were probably
plankton feeders, since their teeth are minute.

The Family Habriochthyidae nov. has only one genus Habroichthys Brough (1939)
and this also comes from the Upper Trias of Besano. Habriochthys differs from the
Peltopleuridae in having but a single row of greatly deepened scales on the flank,
while the tail looks completely homocercal. Further the scaling posteriorly finishes
in an enlarged, symmetrical, semicircular scale.

Order CEPHALOXENIFORMES

This order contains one family, the Cephaloxenidae with but a single genus,
Cephaloxenus Brough, which ranges from the marine Middle to Upper Trias (Brough
1939).

Cephaloxenus is a small fish of deeply fusiform shape and with massive, thick skull
bones _ The fins have relatively few, stiff rays, which are large and unarticulated but
distally bifurcated. The suspensorium is slightly inclined backwards, the opercular
large and the subopercular small. The tail is hemiheterocercal and almost sym-
metrical externally, with the scaly lobe reduced. The scales are stout, few in number,
those of the flank being greatly elongated dorso-ventrally. The preopercular,
although not large, is still high, but the maxilla is rounded posteriorly and relatively
broad anteriorly and the orbit is small. The deepened body, large fins, heavily
armoured head and crushing dentition shows Cephaloxenus to have been a bottom
dweller, presumably feeding on molluscs, crustaceans, etc. The affinities of this
order remain obscure, but in general body form and elongation of the flank scales there
is some similarity to the previous order.

Order LUGANOIITFORMES

The Order Luganoiiformes includes the single Family Luganoiidae which is repre-
sented by two genera from the marine Middle and Upper Trias (Brough 1939).
The members are small, very advanced chondrosteans with fusiform bodies and
somewhat dorso-ventrally compressed heads. The skull is characterized by a certain
amount of fusion of the roofing bones. Thus in Luganoia Brough the parietals,
dermopterotics and dermosphenotics have fused into a posterior bony plate and the
frontals have also fused into an anterior plate which is distinctly narrow anteriorly.
In Besania Brough all these elements have fused into a single bony sheet. The
opercular and subopercular are of approximately the same size and form a semi-

GEOL. 14, 5. 19

192 PALAEONISCOID FISHES AND THE CHONDROSTEI

circular opercular cover, with a straight anterior border. The suspensorium is
inclined forwards as in the semionotids (holosteans), but the preopercular is still
large and high as in the Colobodontidae and the Peltopleuriformes. The maxilla
however is reduced in size and has lost its intimate connection with the other cheek
bones, particularly the preopercular, and has migrated anteriorly to resemble closely
that of the more advanced semionotids. The lepidotrichia of the unpaired fins are
few in number and presumably, equalled their endoskeletal supports. The individual
rays are stiff and unarticulated proximally. The hemiheterocercal tail is markedly
rounded and its scale lobe very short. The lower jaw has the beginnings of a coronoid
process. The scales are thick, enamelled and those of the anterior flank region
elongated dorso-ventrally. These fishes stand on the threshold of the holostean
grade of evolution but still retain sufficient chondrostean characters to be regarded
as among the most advanced chondrosteans, just falling short of the holostean grade.
The obvious chondrostean characters include the lack of an interopercular and the
retention of a large plate-like preopercular with a series of triangular bones above its
dorsal extremity. This order provides another example of parallel evolution.

Both Luganoia and Besama have pointed teeth along the jaw margins and although
the gape is somewhat restricted they were probably quite voracious, surface to mid-
water feeders. The Luganoiiformes have several features in common with the
Peltopleuriformes, in particular the greatly elongated flank scales and the high,
large preopercular, which is dorsally fragmented. This order shared a common
origin with the Perleidiformes, Peltopleuriformes and Cephaloxeniform es.

Order PLATYSIAGIFORMES

Another order composed of a single family, the Platysiagidae, containing only one
genus, Platysiagum Egerton. Platysiagum extends from the Middle Trias to the
Lower Lias (Gardiner 1960, 19662) and is essentially a marine form. It is of elongate
fusiform shape with a deeply forked, equilobate tail (hemiheterocercal). The paired
fins are holostean in structure, as in the previous order, and are of moderate size.
All the fins have numerous small fulcra, and the lepidotrichia are stout, bifurcating
and only distally articulated. The mandible has a broad coronoid process and the
dentition consists of a series of large, conical, pointed teeth, interspersed with
numerous, irregularly arranged smaller teeth, suggestive of a predaceous habit.
The suspensorium is vertical, the maxilla palaeoniscoid in shape and the preopercular
high, broad and dorsally fragmented, similar to that of the Colobodontidae. The
Platysiagiformes have basically the same skull structure as the Colobodontidae but
differ in possessing an incipient interopercular and in the absence of a suborbital
series. This order was probably derived from the Colobodontidae (Perleidiformes)
and represents another chondrostean order which has moved independently towards
the holostean grade of structure.

Order REDFIELDIITFORMES

The Order Redfieldiiformes includes the single family Dictyopygidae, which for a
long time has been associated with the Colobodontidae in the Order Perleidiformes.

PALAEONISCOID FISHES AND THE CHONDROSTEL 193

The Redfieldiiformes, however, can be distinguished from the Perleidiformes by the
excess of rays over radials in both the dorsal and anal fins and in having often only
one modified branchiostegal ray.

The origins of this order are somewhat obscure although the contemporaneous
Dicellopygidae appears to have been derived from the same palaeonisciform stock
(Gonatodidae—Commentryidae).

The Family Dictyopygidae is a fresh-water group, well represented in the Lower
and Middle Triassic fresh-water beds of South Africa, Australia and North America
(Brough 1931). The most primitive members of the Dictyopygidae exhibit the more
conservative palaeoniscoid condition apart from the tail which is now hemihetero-
cercal. In the main the redfieldiids differ from the perleidids in that the dermal rays
of the fin are more numerous than the endoskeletal supports and are also completely
articulated. The snout still has the prominent palaeoniscoid rostrum and the rostro-
premaxillo-antorbital is still a single bone in the primitive members. The sus-
pensorium is oblique. However, Brough (1931) has shown that within the redfieldiids
structural changes were taking place, so that the more advanced forms had come to
resemble the perleidids much more closely. The suspensorium straightens and the
lepidotrichia of the pectoral fin are undivided proximally, the caudal fin becomes
less and less heterocercal and the lepidotrichia of the paired fins come to almost equal
the endoskeletal supports. Of the sixteen or more described genera, eight are
recorded from Australia and another five from South Africa. Redfieldius Hay is
well known from the Upper Trias of South Africa and North America. The early
redfieldiids are thus very similar to the predaceous palaeoniscoids, both in body
shape, fin structure and dentitition, and presumably had very similar habits.

Order PHOLIDOPLEURIFORMES

This order ranges from the Lower to Upper Trias. It contains one family, the
Pholidopleuridae, three members of which are marine and the fourth fresh-water
(Macroaethes Wade). The pholidopleurids are small to moderately long, slender
fishes. The dorsal and anal fins are far back, the origin of the anal being anterior
to that of the dorsal. The lepidotrichia of the unpaired fins are more numerous than
the endoskeletal supports and all are articulated and distally bifurcated. The tail
is hemiheterocercal, deeply cleft and with a reduced scaly lobe. The suspensorium
is almost vertical to moderately oblique, the orbit large and the preopercular high
and perleidid in make up. The frontals are very large and the rostrum is blunt.
The parietals are considerably subdivided into a series of elements. The maxilla has
the normal palaeoniscoid proportions, the teeth are small and pointed and the scales
thin. <Australosomus Piveteau from the Lower Triassic, marine deposits of Mada-
gascar, Greenland and Spitsbergen was a wide-ranging species and it is also recorded
from the Lower Trias of Tanzania, while Macroaethes Wade is confined to Middle
Triassic, fresh-water deposits of Australia. This again is another chondrostean
order whose affinities remain obscure, but which possibly originated from the same
palaeoniscoid stock as the Redfieldiiformes.

104 PALAEONISCOID FISHES AND THE CHONDROSTEI

Order PARASEMIONOTIFORMES

This order is probably the most important of the more advanced chondrostean
orders since it provides the almost ideal stepping-stone to the more advanced holo-
steans and halecostomes. The order contains two families, the Parasemionotidae
and the Tungusichthyidae, which are confined to the marine Lower Trias.

The Family Tungusichthyidae, containing the single genus Tungusichthys Berg
(Berg 1941) from the Lower Trias of the Tunguska Coal Basin, Siberia, is not well
known. The caudal fin still has a pronounced scaly lobe but superficially tends
towards the homocercal condition, being only weakly cleft. The fins are distinctly
holostean in make-up with few rays and these are bifurcated but only distally
articulated. The suspensorium is vertical and the preopercular narrow as in the
caturids; in this respect it differs from the Parasemionotidae. The maxilla is
thin and reduced, very much holostean in appearance and with a supramaxillary in
articulation ; the interopercular is small.

The Family Parasemionotidae contains eight genera from the Lower Trias of
Madagascar and Greenland. They represent a fairly uniform group but with con-
siderable variation in the preopercular and snout regions (Lehman 1952).

Apart from the skull, the Parasemionotidae are completely holostean, although the
scaly lobe of the hemiheterocercal tail still extends almost halfway along the dorsal
lobe. The body shape, size, scaling and nature of the paired and unpaired fins are
all distinctly holostean and halecostome. In the skull the roofing bones correspond
in basic structure and arrangement to those of the early holosteans and halecostomes.
In the cheek region the maxilla is freed from the preopercular while the opercular
series includes a true interopercular. In both these features the Parasemionotidae
closely approach the holosteans. Only in the size and shape of the preopercular
and absence of a true suborbital series do the Parasemionotidae fall short of the
holostean and halecostome grades. The preopercular is still large and broad medially
and would need to become more curved antero-ventrally and broadened in that
region to approach the condition found in the halecostomes. However, Lehman
(1952) has shown that within the Parasemionotidae suborbitals are being formed by
anamestic fragmentation of the anterior part of the preopercular. Therefore, he
concludes that the Parasemionotidae are probably ancestral to the caturids.
Further, Gardiner (1960) has shown the remarkable similarity that exists between
the Parasemionotidae on the one hand and the halecostomes on the other.

The actual origin of the Parasemionotiformes is less clear. Of the known chondo-
strean families the Aeduellidae (Palaeonisciformes) look the most likely ancestors.
The family includes the two genera Aeduella Westoll and Westollia White & Moy-
Thomas from the Lower Permian of Autun and Thuringia respectively. These are
characterized by the upright nature of the suspensorium, the reduction of the pos-
terior expanded portion of the maxilla, and the shape and size of the preopercular.
The body is of the right shape and proportions to have given rise to the Parasemiono-
tiformes although the fins are distinctly palaeoniscoid. There is no interopercular.
Thus it seems that the Aeduellidae and the Parasemionotiformes shared a common
ancestry in the upper Carboniferous.

PALAEONISCOID FISHES AND THE CHONDROSTEI 195
Order PTYCHOLEPIFORMES

Members of this order are first encountered in the Middle Trias and the order sur-
vived into the Upper Lias. The order is represented by a single family, the Ptycho-
lepididae, containing the single genus Ptycholepis Agassiz.

Ptycholepis is a marine form and ranges in size from small to large (some 60 cm.).
It is elegantly fusiform with an acutely pointed snout (Gardiner 1960). The caudal
fin is deeply forked and hemiheterocercal. The pectoral and pelvic fins are well
developed, the dorsal and anal fins are triangular, the former opposed to the pelvics,
the latter smaller and remote. The lepidotrichia of the unpaired fins are few in
number and completely articulated and distally bifurcated, nearly equalling their
endoskeletal supports in number. The suspensorium is almost vertical, the gape
wide and the orbit large. The dentition consists of two series of small, close-set
teeth and these fishes were presumably mid-water, plankton feeders. The frontals
are large and elongate, making up the major portion of the skull roof. The skull
roofing bones are distinctly ornamented with high ridges of enamel. The snout is
produced into a blunt rostrum with a prominent postrostral present. The opercular
is large and quadrangular, the scales thick, longer than wide and much elongated in
the ventral region.

Because of the holostean-like character of the body and fins, this genus has in the
past been grouped in the Holostei, but the absence of an interopercular, the possession
of a palaeoniscoid-type maxilla and preopercular (covered with suborbitals in later
species) and other obvious chondrostean characters of the skull, such as the snout,
show that Ptycholepis is a representative of yet another independent line from the
palaeoniscoids, which has not yet quite reached the holostean level. Brough (1939)
has shown that the Ptycholepiformes can be directly derived from the earlier Boreo-
somidae.

Order BOBASATRANITFORMES

The bobastraniids form a compact little group found only in the marine Lower
Trias. The single Family Bobasatraniidae has representatives from Spitsbergen,
Madagascar, Greenland and North America.

The bobasatraniids resemble the later pycnodonts in many respects but are in no
way related to them. The bobasatraniids are an offshoot from the same palaeonis-
coid stock which gave rise to the amphicentrids, and they died out without giving
rise to any other group. In general body-form the bobasatraniids resemble the
earlier Amphicentridae, particularly in the make-up of the shoulder girdle and the
unpaired fins and in the much deepened, laterally compressed body. The lepido-
trichia of the fins are slightly more numerous than the endoskeletal supports and the
dorsal and anal fins are long as in some species of Platysomus Agassiz. However the
median fins are holostean in form while the opercular apparatus is peculiar. The
opercular is small with a much expanded preopercular plate below it and with the
branchiostegal rays completely reduced. The clavicle has been lost.

Bobasatramia White (1932) has a modified crushing dentition reminiscent of that
seen in Clivodus, The suspensorium is upright, the gape small, the pectoral fins

196 PALAEONISCOID FISHES AND THE CHONDROSTEI

long and the pelvics wanting. The dorso-ventrally deepened body, strongly forked
heterocercal tail, long dorsal and anal fins and crushing dentition suggest that this
was a browsing form probably feeding close inshore amongst corals.

Order DORYPTERIFORMES

The Order Dorypteriformes includes the family Dorypteridae which is represented
by the single genus Dorypterus Germar from the marine Upper Permian (Gill 1925 ;
Liu & Tseng 1964). Dorypterus shows many points of similarity to the Bobasa-
traniiformes and, like this order, was probably derived from the earlier Amphi-
centridae. There are close relationships between the axial skeleton and fin skeleton
of Dorypterus and Bobasatrania White ; the expanded sinuous axonosts are in contact
with one another, while the development of the body axis of the caudal fin is more or
less identical in both. The bones of the upper jaw are similarly developed in Doryp-
terus and Bobasatramia and the pectoral girdles show many likenesses. However,
despite these few similarities there are many divergent features ; no known bobasa-
traniid shows the extreme modifications of the skull found in Dorypterus and likewise
the body of Dorypterus is not completely covered with thick scales (as in bobasa-
traniids). The scaling is reduced to the anterior portion of the trunk in Dorypterus.
It would appear that the dorypterids and bobasatraniids must have come from the
same early amphicentrid-like stock but both lines soon died out.

IV. CLASSIFICATION OF THE CHONDROSTEI
Class ACTINOPTERYGII
Subclass CHONDROSTEI

Order PALAEONISCIFORMES

Family CHEIROLEPIDIDAE Pander 1860
Cheirolepis Agassiz 1835
Family STEGOTRACHELIDAE Gardiner 1963
Stegotrachelus Woodward & White 1926, Moythomasia Gross 1950, Orvikuina
Gross 1953, Kentuckia Rayner 1951
Family OSORIOICHTHYIDAE nov.
Osorioichthys Casier 1954 (Stereolepis Casier 1952, Stereolepidella Whitley
1954)
Family TEGEOLEPIDIDAE Romer 1945
Tegeolepis Miller 1892 (Actinophorus Newberry 1888), ?Apateolepis Wood-
ward 1890, ?Megapteriscus Wade 1935, ?Elpisopholis Woodward 1908
Family CARBOVELIDAE Romer 1945
Carboveles White 1927, Phanerosteon Traquair 1881
Family GONATODIDAE nov.
Gonatodus Traquair 1877, Drydenius Traquair 1890, Pseudogonatodus nov.
Family GYROLEPIDOTIDAE noy.
Gyrolepidotus Rohon 1889, Palaeobergia Matveeva 1958

PALAEONISCOID FISHES AND THE CHONDROSTEI 197

Family COSMOPTYCHIIDAE Gardiner 1963
Watsonichthys Aldinger 1937, Cosmoptychius Traquair 1877

Family ACROLEPIDIDAE Aldinger 1937
Acrolepis Agassiz 1833, Acropholis Aldinger 1935, Acrorhabdus Stensi6 1921,
Hyllingea Aldinger 1935, Plegmolepis Aldinger 1937, Reticulolepis Aldinger
1937, Mesonichthys Gardiner 1963

Family ELONICHTHYIDAE Aldinger 1937
Elonichthys Giebel 1848 (Ganacrodus Owen 1867, Propalaeoniscus Pomel
1853), Namaichthys Giirich 1923

Family RHADINICHTHYIDAE Romer 1945
Rhadinichthys Traquair 1877, Cycloptychius Young 1866, Rhadinoniscus
White 1937, Aetheretmon White 1927, Stvepheoschema White 1927, Ment-
zichthys Jubb 1965, Eurylepidoides Case 1935, Ganolepis Woodward 1893

Family CANOBIIDAE Aldinger 1937
Canobius Traquair 1881, Mesopoma Traquair 1890, Wiiteichthys Moy-
Thomas 1942, ?Aldingeria Moy-Thomas 1942

Family PyGoPpTERIDAE Aldinger 1937
Nematoptychius Traquair 1875, Pygopterus Agassiz 1833

Family RHABDOLEPIDIDAE Gardiner 1963
Rhabdolepis Troschel 1857

Family STYRACOPTERIDAE Moy-Thomas 1939
Styracopterus Traquair 1890 (Fouldenia White 1927), Benedenius Traquair
1878 (Benedenichthys Traquair 1890)

Family CRYPHIOLEPIDIDAE Moy-Thomas 1939
Cryphiolepis Traquair 1881

Family AMPHICENTRIDAE Moy-Thomas 1939
Chirodus M’Coy 1848 (Amphicentrum Young 1866, Cheivodus M’Coy 1855,
Hemucladodus Davis 1884), Cheirodopsis Traquair 1881, Eurynothus Agassiz
1834 (Eurinotus Agassiz 1836, Euronotus Agassiz 1835, Eurunotus Pander
1860, Eurynotus Agassiz 1834, Notaemon Gist] 1848, Plectrolepis Egerton
1850), Eurynotoides Berg 1940, Paraeurynotus Chabakov 1927, Globulodus
Minster 1842 (Eurysomus Young 1866, Lekanichthys Brough 1934),
Proteurynotus Moy-Thomas & Dyne 1938, ?Tompoichthys Obruchev 1964.

Family PLATYSOMIDAE Young 1866
Platysomus Agassiz 1833 (Stvomateus Blainville 1818), Mesolepis Young
1866 (Pododus Agassiz 1844), Paramesolepis Moy-Thomas & Dyne 1938,
Wardichthys Traquair 1875, Caruichthys Broom 1913

Family HoLurIDAE Moy-Thomas 1939
Holurus Traquair 1881, Holuropsis Berg 1947, Palaeoniscinotus Rohon 1890,
Peleichthys Broom 1913, Disichthys Broom 1913

Family CORNUBONISCIDAE White 1939
Cornuboniscus White 1939

Family PHANERORHYNCHIDAE Stensi6 1932
Phanerorhynchus Gill 1923

198

PALAEONISCOID FISHES AND THE CHONDROSTEI

Family TRISSOLEPIDIDAE Fri¢ 1893
Sphaerolepis Frié 1877 (Trissolepis Frié 1893), Sceletophorus Frié 1894
(Phanerosteon Frié 1894, Gymnoniscus Berg 1936)

Family BIRGERIIDAE Aldinger 1937
Birgeria Stensi6 1919 (Xenesthes Jordan 1907), Ohmdemia Hauff 1953

Family ATHERSTONIIDAE nov.
Atherstonia Woodward 1889 (Hypterus Owen 1876, Broometta Chabakov
1928)

Family UROSTHENIDAE Woodward 1931
Urosthenes Dana 1848

Family SCANILEPIDIDAE Romer 1945
Scanilepis Aldinger 1935, Evenkia Berg 1941

Family AMBLYPTERIDAE Romer 1945
Amblypterus Agassiz 1833 (Aedua Sauvage 1890, Archaeonichthys Whitley
1940, Archaeoniscus Sauvage 1890, Leiolepis Goldenburg 1873), Param-
blypterus Sauvage 1888 (Amblypterops Sauvage 1889, Cosmopoma Sauvage
1889, Dipteroma Sauvage 1889, Geomichthys Sauvage 1889), Amblypterina
Berg 1940, Tholonotus Dunkle & Schaeffer 1956

Family COMMENTRYIDAE Gardiner 1963
Commentrya Sauvage 1888 (Elaveria Sauvage 1888)

Family LAWNIIDAE nov.
Lawnia Wilson 1953

Family DICELLOPYGIDAE Romer 1945
Dicellopyge Brough 1931, ?Aneurolepis White & Moy-Thomas 1941 (Urolepis
Bellotti 1857)

Family AEDUELLIDAE Romer 1945
Aeduella Westoll 1937, Westollia White & Moy-Thomas 1940 (Lepidopterus
Pholig 1892)

Family BOREOLEPIDIDAE Aldinger 1937
Boreolepis Aldinger 1937

Family PALAEONISCIDAE Vogt 1852
Palaeoniscum Blainville 1818 (Eupalaeoniscus Rzchak 1881, Palaeoniscus
Agassiz 1833, Palaeothrissum Blainville 1818, Palaeomuzon Weigelt 1930),
Pteronisculus White 1933 (Glaucolepis Stensid 1921), Agecephalichthys
Wade 1935, Myriolepis Egerton 1864, Tvachelacanthus Fischer de Waldehim
1850, Gyrolepis Agassiz 1833, ?Gyrolepidoides Cabrera 1944, Turseodus
Leidy 1857, Belichthys Wade 1935, ?Progyrolepis Fri¢é 1894, ?Challaia
Rusconi 1946, Leptogenichthys Wade 1935

Family COSMOLEPIDIDAE nov.
Cosmolepis Egerton 1854 (Oxygnathus Egerton 1854, Thrissonotus Agassiz
1844)

Family CENTROLEPIDIDAE Gardiner 1960
Centrolepis Egerton 1858

PALAEONISCOID FISHES AND THE CHONDROSTEI 199

Family COCCOLEPIDIDAE Berg 1940
Coccolepis Agassiz 1844, Browneichthys Woodward 1889, Sunolepis Liu
1957, Pteroniscus Chekker 1848
Family CocCOCEPHALICHTHYIDAE Romer 1945
Coccocephalichthys Whitley 1940 (Coccocephalus Watson 1925, Cocconiscus
White & Moy-Thomas 1940)
Family BRACHYDEGMIDAE nov.
Brachydegma Dunkle 1939
Family BOREOSOMIDAE nov.
Boreosomus Stensi6 1921 (Diaphorognathus Brough 1933), Mesembroniscus
Wade 1935
Order TARRASIIFORMES
Family TARRASIIDAE Traquair 1881
Tarrasius Traquair 1881, Palaeophichthys Eastman 1907.
Order HAPLOLEPIFORMES
Family TELEOPTERINIDAE Berg 1936
Haplolepis Miller 1892 (Eurylepis Newberry 1857, Mecolepis Newberry
1856, Mekolepis Newberry 1857), Pyritocephalus Fri¢ 1894 (Teleopterina
Berg 1936)
Order SAURICHTHYIFORMES
Family BELONORHYNCHIDAE Woodward 1888
Saurichthys Agassiz 1834 (Belonorhynchus Bronn 1858, Giffonus Costa 1862,
Ichthyorhynchus Bellotti 1857, Stylorhynchus Martin 1873), Saurorhynchus
Reis 1892 (Belonostomus Agassiz 1844, Belonorhynchus Bronn 1858,
Acidorhynchus Stensid 1925, Gymnosaurichthys Berg 1940)
Order CHONDROSTEIFORMES
Family CHONDROSTEIDAE Traquair 1877
Chondrosteus Egerton 1858, Gyrosteus Morris 1854, Stichopterus Reis 1909,
Strongylosteus Jaekel 1929
Family ERROLICHTHYIDAE Lehman 1952
Errolichthys Lehman 1952, Psilichthys Hall 1900
Family PEIPIAOSTEIDAE Liu & Zhou 1965
Peipiaosteus Liu & Zhou 1965
Order ACIPENSERIFORMES
Family ACIPENSERIDAE Bonaparte 1831
Acipenser Linnaeus 1758, Huso Brandt 1833, Kessleria Bogdanon 1882
(Hemiscaphirhynchus Berg 1911, Pseudoscaphirhynchus Nicolsky 1900),
Protoscaphirhynchus Wilimovsky 1956, Scaphirhynchus Heckel 1835
Family PoLYDONTIDAE Bonaparte 1838
Polyodon Schneider 1801, Palaeopsephurus MacAlpin 1947, Pholidurus
Woodward 1889, Psephurus Guenther 1873, Crossopholis Cope 1883
Order POLYPTERIFORMES
Family PoLYPTERIDAE
Polypterus Sainte-Hilaire 1802 (Lacépéde 1803), Erpetoichthys Smith 1865
(Calamoichthys Smith 1866)

200 PALAEONISCOID FISHES AND THE CHONDROSTEI

Order PERLEIDIFORMES
Family COLOBODONTIDAE Stensi6 rg16
Colobodus Agassiz 1844, Crenolepis Carus 1888 (Crenilepis Dames 1888,
Crenilepoides Strand 1929), Meridensia Stensid 1916, Perleidus Alessandri
1910, ?Thoracopterus Bronn 1858, ?Gigantopterus Abel 1906, Mezdiichthys
Brough 1931, Mendocinichthys Whitley 1953 (Mendocinia Bordas 1944),
Tripelta Wade 1940, Chrotichthys Wade 1940, Zeuchthiscus Wade 1940,
Pristisomus Woodward 1890, Manlietta Wade 1935, Procheirichthys Wade
1935, Dimorpholepis Teixeira 1947, Engycolobodus Oertle 1927, Dollopterus
Abel 1906, Albertonia Gardiner 1966
Family AETHEODONTIDAE Brough 1939
Aetheodontus Brough 1939
Family CLEITHROLEPIDIDAE Wade 1935
Cleithrolepis Egerton 1864, Cleithrolepidina Berg 1955, Hydropessum
Broom 1909, Dipteronotus Egerton 1854
Order LUGANOIIFORMES
Family LUGANolDAE Brough 1939
Luganoia Brough 1939, Besania Brough 1939
Order PELTOPLEURIFORMES
Family PELTOPLEURIDAE Brough 1939
Peltopleurus Kner 1866, Placopleurus Brough 1939
Family HABROICHTHYIDAE nov.
Habroichthys Brough 1939
Order CEPHALOXENIFORMES
Family CEPHALOXENIDAE Brough 1939
Cephaloxenus Brough 1939
Order PLATYSIAGIFORMES
Family PLATYSIAGIDAE Brough 1939
Platysiagum Egerton 1872
Order REDFIELDIIFORMES
Family DicTyopyGIDAE Hay 1889
Redfieldia Hay 1899 (Catopterus Redfield 1837), Dictyopyge Egerton 1847,
Daedalichthys Brough 1931, Sakamenchthys Nauche 1959, Helichthys
Broom 1909, Atopocephala Brough 1934, Brookvalia Wade 1935, Beconia
Wade 1935, Dictyopleurichthys Wade 1935, Geitonmichthys Wade 1935,
Molybdichthys Wade 1935, Philyctaenichthys Wade 1935, Schizurichthys
Wade 1935, Ischnolepis Haughton 1934, Sinkiangichthys Liu 1958, Pseudo-
becoma Bordas 1944, ?Rushlandia Bock 1959
Order PHOLIDOPLEURIFORMES
Family PHOLIDOPLEURIDAE Wade 1932
Australosomus Piveteau 1930, Pholidopleurus Bronn 1858, Macroaethes
Wade 1932, Arctosomus Berg 1941 (Neavichthys Whitley 1951)
Order PryCHOLEPIFORMES
Family PTyCHOLEPIDIDAE Brough 1939
Ptycholepis Agassiz 1833

PALAEONISCOID FISHES AND THE CHONDROSTEI 201

Order DoRYPTERIFORMES
Family DoRYPTERIDAE Gill 1925
Dorypterus Germar 1842

Order BOBASATRANIIFORMES
Family BOBASATRANIIDAE Stensid 1932
Bobasatrania White 1932 (Lambeichthys Lehman 1956), Ecrinesomus
Woodward 1910

Order PARASEMIONOTIFORMES
Family PARASEMIONOTIDAE Stensi6 1932
Parasemionotus Piveteau 1929, Stensionotus Lehman 1952, Watsonulus
Brough 1939, Jacobulus Lehman 1952, Thomasinotus Lehman 1952, Ospia
Stensi6 1932, Broughia Stensi6 1932, Helmolepis Stensid 1932
Family TUNGUSICHTHYIDAE Berg 1941
Tungusichthys Berg 1941

Chondrostei incertae sedis

Anaglyphus Rzehak 1881, Anatoia Rusconi 1946, Caminchaia Rusconi 1946, Cenchrodus Meyer
1847, Cenechoia Rusconi 1946, Cephaliscus Whitley 1940 (Cephalacanthus Beyrich 1848),
Echentaia Rusconi 1946, Guaymayenia Rusconi 1946, Hemilopas Meyer 1847, Neochallaia
Rusconi 1949, Nephrotus Meyer 1851, Omphalodus Meyer 1847, Oxypteriscus Matveeva 1958,
Pasambaya Rusconi 1946, Schigospondylus Frié & Bayer 1902.

V. SUMMARY

This paper is the second of a series intended to form the basis for a revision of
the palaeoniscoid fauna of the British Carboniferous. The type species of three
genera from this fauna are redescribed and for comparative purposes two other type
species from the Upper Carboniferous of Czechoslovakia.

Eight new Palaeonisciform families are erected. They are the Osorioichthyidae
which appears to be an independently derived side line from the ancestral stock ;
the Gonatodidae, a family close to the Acrolepididae and to the Amblypteridae ;
the Gyrolepidotidae allied to the Acrolepididae—Elonichthyidae complex; the
Atherstoniidae, a family close to the Trissolepididae ; the Lawniidae which was
probably derived from the Gonatodidae ; the Cosmolepididae, Brachydeymidae and
the Boreosomidae. One other new chondrostean family is proposed, the Habroich-
thyidae which belongs to the Peltopleuriformes.

A new genus Pseudogonatodus is erected for Gonatodus parvidens Traquair (1892)
and it is also used to include Gonatodus macrolepis Traquair (1877).

The classification and evolution of the Palaeonisciformes is discussed and since
all the chondrostean orders stemmed from the Palaeonisciformes, the subsequent
evolution of the Subclass Chondrostei is outlined and a complete classification of the
Subclass is attempted.

202 PALAEONISCOID FISHES AND THE CHONDROSTEI
VI. ACKNOWLEDGMENTS

I wish to thank Dr. E. I. White, F.R.S., of the British Museum (Natural History)
and Dr. C. D. Waterston of the Royal Scottish Museum for the hospitality received
in their respective departments ; Dr. V. Zazvorka of the Narodni Museum, Prague,
for the photographs of Sceletophorus and Sphaerolepis which I have used in my plates,
and Dr. C. Patterson and Mr. H. A. Toombs for their generous assistance, advice and
kindness to me.

VII. LETTERING USED IN TEXT BIGURES

Ang angular mp median pit line
Ant antorbital Mx maxilla
ap anterior pit line
Art articular Na nasal

na, anterior nasal aperture (nostril)
Br branchiostegal ray nag posterior nasal aperture (nostril)
can.W canals of Williamson ° pore
Cl cleithrum Op opercular
Clav clavicle orp postmaxillary sensory line
Cor coronoid
c.sp cell spaces Pa parietal

Pop preopercular
d dentine pp posterior pit line
Den dentary Ptr postrostral
Dpt dermopterotic
Dsp dermosphenotic I ridges of enamel
d.t dential tubules r.can radial canal
Dyh dermohyal r.cr.con radial cross connection

R.pmx rostro-premaxillary
e enamel R.pmx.ant rostro-premaxillo-antorbital
Exsc extrascapular

Sbo suborbital
fe fenestra Scl supracleithrum
Fr frontal SC.T. sclerotic ring

Sh.f Sharpey’s fibres
Gu gular plate So supraorbital

soc supraorbital sensory canal
he supramaxillary sensory line Sop subopercular
hla horizontal bone lamellae Ssc suprascapular
h.v.can horizontal network of canals
Hy hyomandibular B.M.N.H. British Museum (Natural His-

tory), London
Inf infraorbital R.S.M. Royal Scottish Museum, Edin-
infc infraorbital sensory canal burgh
VIII. REFERENCES

Aeassiz, L. 1833-44. Recherches sur les Poissons fossiles. 5 vols. 1429 pp., 396 pls., with

supplement. Neuchatel. [For dates of publication of various sections of text and plates
see Woodward & Sherborn 1890].
ALDINGER, H. 1937. Permische Ganoidfische aus Ostgronland. Meddy. Gronland, Kjoben-
havn, 102, 3: xliv + 392 pp., 44 pls.
ANDERSSON, E. 1916. Uber einige Trias-Fische aus der Cava Trefontane, Tessin. Bull.
Geol. Inst. Univ. Uppsala, 15 : 13-24.

PALAEONISCOID FISHES AND THE CHONDROSTEI 203

Bere, L. S. 1936. Teleopterina n.g., a highly organized acanthopterygian fish from the Car-
boniferous of North America. Dokl. Akad. Nauk SSSR for. Lang. Edn., Moscow, 4, 7 : 345-
347, fig. 1. [In the Russian edition 4, 7 : 333-335 for Acanthopterygian read Actinop-
terygian. ]

1940. Classification of Fishes, both Recent and Fossil. Tyvudy zool. Inst., Leningr., 5,

2. 517 pp., 190 figs.

— 1941. Lower Triassic fishes of the Tunguska Coal basin, Siberia. Izv. Akad. Nauk.
SSSR, Moskva, 3 : 458-474, 24 figs.

Brot, J. 1963. Etude des Palaeoniscidés du Bassin Houiller de Commentry (Allier). Thesis.

Biot, J. & Heyer, D. 1963. Sur une particularité anatomique de certains Poissons du
Permo-Carbonifere des bassins de Commentry et Autun. Bull. Soc. géol. Fr., Paris, (7)
5 : 64-69, pls. 7-9.

Broom, R. 1913. Onsome fossil fishes from the diamond bearing pipes of Kimberley. Tvans.
R. soc. S. Afy., Cape Town, 3 : 399-402.

Broucu, J. 1931. On Fossil Fishes from the Karroo System and some general considerations
on the Bony Fishes of the Triassic Period. Proc. zool. Soc. Lond., 1931 : 235-295.

1936. On the evolution of bony fishes during the Triassic period. Bzol. Rev., Cambridge,
M1 : 385-405.

1939. The Triassic Fishes of Besano, Lombardy. ix + 117 + xiv pp., 7 pls. Brit. Mus.
(Nat. Hist.), London.

CasiER, E. 1952. Contributions a l’étude des Poissons fossiles de la Belgique, X. Un
Palaeoniscoide du Famennien inférieur de la Fagne: Steveolepis marginis n. gen., n. sp.
Bull. Inst. r. Sci. nat. Belg., Bruxelles, 28, 47 : 1-17, pls. 1, 2.

1954. Contributions a l’étude des Poissons fossiles de la Belgique, XI. Note additionelle
relative a ‘‘ Steveolepis’’ (=Osorioichthys nov. nom.) et a Vorigine de l’interoperculaire.
Bull. Inst. r. Sci. nat. Belg., Bruxelles, 30, 2 : 1-12, 3 figs.

DunkLe, D. H. 1939. A new palaeoniscid fish from the Texas Permian. Am. J. Sci.,
New Haven, 237 : 262-274, 1 pl., 2 figs.

Dyne, M. B. 1939. The skull of Amphicentrum granulosum. Proc. zool. Soc. Lond. (B) 109:
195-210, pls. I, 2.

Frié, A. J. 1875. Uber die Fauna der Gaskohle des Pilsner und Rakonitzer Beckens. Sber. K.

bohm. Ges. Wiss., Prag., 1875 : 70-79, 1 table.

1877. Zur Fauna der Gaskohle von Zabot bei Schlan Krouéova bei René und TiemoSna
bei Pilsen, sowie iiber die Sphaerosideritkugeln von Zilov. Sber. K. béhm. Ges. Wiss., Prag.,
1877 : 45-52, 1 table.

1879-1901. Fauna dey Gaskohl und der Kalksteine der Peymformation Bohmens. 4 vols.
illust. Vol. 1, 1879-83: Vol. 2, pt. 1, 1885, pt. 2, 1885, pt. 3, 1888, pt. 4, 1889: Vol. 3,
pt. I, 1890, pt. 2, 1893, pt. 3, 1893-94, pt. 4, 1893: Vol.4, pt. 1, 1899, pt. 2, 1899, pt. 3,
Igol. Prag.

GarRDINER, B. G. 1960. A revison of certain Actinopterygian and Coelacanth fishes, chiefly

from the Lower Lias. Bull. Br. Mus. nat. Hist. Geology, London, 4 : 239-384, 8 pls., 81 figs.

1962. Namaichthys schvoederi Giirich and other palaeozoic fishes from South Africa.
Palaeontology, Lond., 5 : 9-21, pl. 6.

1963. Certain Palaeoniscoid fishes and the evolution of the snout in actinopterygians.

Bull. Br. Mus. nat. Hist. Geology, London, 8 : 255-325, 2 pls., 20 figs.

1966. A Catalogue of Canadian Fossil Fishes. R. Ont. Mus. Univ. Toronto, 68: 1-154.

1966a. In press. Chondrostei, Holostei and Halecostomi in A review of the fossil record.

Gitt, E.L. 1923. Anundescribed fish from the Coal Measures of Lancashire. Ann. Mag. Nat.
Hist., London, [9] 11 : 465-471, 1 fig. [Note on the preceding Paper by D. M. S. Watson,
ibid. : 471-472.]

1923a. The Permian fishes of the genus Acentrophorus. Proc. zool. Soc. Lond., 1923 : 19-

40.
1925. The Permian fish Dorypterus. Trans. R. Soc. Edinb., 53 : 643-661, 6 figs.

204 PALAEONISCOID FISHES AND THE CHONDROSTEI

GrassE, P. P. 1958. Tvaité de Zoologie ... Anatomie, Systématique Biologie : 13, 3 : Agnathes
et Poissons ... anatomie, éthologie, systématique ... pp. 1813-2758, figs. 1308-1889. Paris.

GRIFFITH, J. 1958. The systematic position of Browneichthys ovnatus Woodward (Pisces,
Actinoptergii). Proc. zool. Soc. Lond., 131, 2 : 157-161, pl. I.

Gross, W. 1947. Die Agnathen und Acanthodier des obersilurischen Beyrichienkalks.
Palaeontographica, Stuttgart, 96A : 91-158, pls. 20-28.

Guppy, D. J., LINDNER, A. W., Ratrican, J. H. & CasEy, J. N. 1958. The Geology of the
Fitzroy Basin, Western Australia. Bull. Bur. Miner. Resour. Geol. Geophys. Aust., Can-
berra, 36: 116 pp., 5 pls. + maps.

HEIDE,S.v.d. 1943. Lefaune ichthyologique du Carbonifere supérieur des Pays-Bas. Meded.
geol. Sticht., Maastricht, (C) 4, 3, 2 : 1-65.

Jarvix, E. 10960. Théovies de l’Evolution des Vertébrés. 104 pp., 30 figs. Paris.

Konincxk, L.G. 1878. Faune du Calcaire Carbonifére de la Belgique. Anns. Mus. y. Hist. nat.
Belg., Brussels, pt. 1, 2: 152 pp., 31 pls.

Leuman, J. P. 1952. Etude complémentaire des Poissons de l’Eotrias de Madagascar.
K. svenska VetenskAkad. Handl., Uppsala & Stockholm (4) 2, 6: 1-201, 48 pls.

Liu, H. T. & Tsene, H. Y. 1964. Note ona Permian Dorypterus of Hsinhua, Central Hunan.
Vertebr. palasiat., Peking, 9, 3 : 318-320, 1 pl.

Liu, H. T. & Zuou, J. J. 1964. A new sturgeon from the Upper Jurassic of Liaoning, North
China. Vertebrl palasiat., Peking, 9, 3 : 237-248, 3 pls.

MacAtpin, A. 1947. Palaeopsephurus wilsoni, a new polyodontid fish from the Upper Cre-
taceous of Montana, with a discussion of allied fish, living and fossil. Conty. Mus. Paleont.
Univ. Michigan, 6, 8 : 167-234.

Moy-TuHomas, J. A. 1939. Palaeozoic Fishes. ix + 149 pp., 33 figs. London.

Moy-Tuomas, J. A. & DynE, M. B. 1938. Actinopterygian fishes from the Lower Carboni-
ferous of Glencartholm, Eskdale, Dumfriesshire. Tvans. R. Soc. Edinb., 59 : 437-480, 2 pls.

NEWBERRY, J. S. 1873. Descriptions of fossil fishes. Rep. geol. Surv. Ohio, Columbus, 1873,

1, 2 : 245-355, 17 pls.

1888. On the fossil fishes of the Erie Shale of Ohio. Tvans. N.Y. Acad. Sci., New York,

7: 178-180.

1890. The Palaeozoic fishes of North America. Monogr. U.S. geol. Surv., Washington,

1889, 16: iv + 340 pp., 53 pls.

NIELSEN, E. 1942. Studies on Triassic fishes from East-Greenland. I. Glaucolepis and

Bovreosomus. Meddr. Gronland, Kjobenhavn, 138: 403 + 31 pp., pls. I-30.

1949. Studies on Triassic fishes from East Greenland. II. Austvalosomus and Birgeria.

Meddy. Gronland, Kjobenhavn, 146 : 309 pp., 20 pls.

OBRUCHEV, D. V. 1964. Osnovy paleontologit : 2, Beschelyustnye, ryby. 522 pp., 55 pls.
Izd. Nanka, Maskova.

Pruvost, P. 1919. La Faune Continentale du Terrain Houiller du Nord dela France. Mém.

Carte géol. dét. Fy., Paris: xxxii + 587 pp., 29 pls.

1930. La Faune Continentale du Terrain de la Belgique. Mém. Mus. rv. Hist. nat. Belg.,

Bruxelles, 44 : 105-282, 14 pls.

Rayner, D. M. 1937. On Leptolepis bronni Agassiz. Ann. Mag. nat. Hist., London (10)

19 : 46-74, 14 figs.

1948. The structure of certain Jurassic Holostean fishes, with special reference to their

neurocrania. Phil. Trans. R. Soc., London (B) 233 : 287-345, 4 pls.

Romer, A. S. 1945. Vertebrate Palaeontology. 2nd ed., ix + 687 pp., 377 figs. Chicago.

StTENSIO, E. A. 1916. See Andersson, 1916.

1921. Tviassic fishes from Spitzbergen, Pt. I. xxviii + 307 pp., 34 pls., 87 figs. Vienna.

1925. ‘Triassic Fishes from Spitzbergen, Part II. K. svenska VetenskAkad. Handl.,

Uppsala & Stockholm (3) 2, 1: 261 pp., 33 pls.

1932. ‘Triassic Fishes from East Greenland collected by the Danish Expeditions in 1929—

1931. Meddr. Gronland, Kjobenhavyn, 83, 3 : 345 pp., 39 pls., 94 figs.

PALAEONISCOID FISHES AND THE CHONDROSTEI 205

STENSIO, E. A. 1947. The sensory lines and dermal bones of the cheek in fishes and amphibians.
K. svenska VetenskAkad. Handl., Uppsala & Stockholm [3], 24, 3 : 1-195, 38 figs.

1961. Permian Vertebrates IN Geology of the Arctic. 1: 231-247, 2 text-figs., Univ.
Toronto.

Traquair, R. H. 1877. On new and little known fossil fishes from the Edinburgh district,
No. 1. Proc. R. Soc. Edinb., 9 : 262-272.
1877b. On the Agassizian genera Amblypterus, Palaeoniscus, Gyrolepis and Pygopterus.
Q. Jl. geol. Soc. Lond., 33 : 548-578.
1877c. The ganoid fishes of the British carboniferous formations. Part I. Palaeonis-
cidae. Palaeontogy. Soc. (Monogr.), London, 31 : 1-60, pls. 1-7.
1881. Notice of New Fish Remains from the Blackband Ironstone of Borough Lee
near Edinburgh. Geol. Mag., London (2) 8 : 34-37.
1882. Notice of new fish remains from the Blackband Ironstone of Borough Lee, near
Edinburgh, No. 3. Geol. Mag., London (2) 9 : 540-546.
1888. New Palaeoniscidae from the English Coal Measures No. 2. Geol. Mag., London
(3) 5: 251-254.
1890. List of the fossil Dipnoi and Ganoidei of Fife and the Lothians. Proc. R. soc.
Edinb., 17 : 385-400.
1go1. The ganoid fishes of the British Carboniferous formation. Part I. Palaeoniscidae.
Palaeontogy. Soc. (Monogr.), London, 31: 61-87, pls. 8-18.
1903. On the distribution of fossil fish remains in the Carboniferous rocks of the Edinburgh
district. Tvans. R. Soc. Edinb., 40 : 687-707, 2 pls.
1907. ‘The ganoid fishes of the British Carboniferous formations. Part I. Palaeoniscidae.
Palaeontogr. Soc. (Monogr.), London, 31 : 88-106, pls. 19-23.
19076. Report on Fossil Fishes collected by the Geological Survey of Scotland from Shales
exposed on the Shore near Gullane, East Lothian. Tyvans. R. Soc. Edinb., 46 : 103-117,
pls, 2.
1g09. The ganoid fishes of the British Carboniferous formations. PartI. Palaeoniscidae.
Palaeontogr. Soc. (Monogy.), London, 31 : 107-122, pls. 24-30.
tg12. The ganoid fishes of the British Carboniferous formations. Part I. Palaeonis-
cidae. Palaeontogr. Soc. (Monogy.), London, 31 : 159-180, pls. 36-40.
VANGERONE, C.F. 1958. Mikropalaontologische untersuchungen in dem Kohlscheider Schich-
ten im Wurmrevier bei Aachen. Geol. Jb., Hannover, 73 : 457-506, 4 pls.

Wave, R. T. 1935. The Triassic fishes of Brookvale, New South Wales. 110 pp., Ito pls.
Brit. Mus. (Nat. Hist.), London.

WALKER, R. 1872. Ona New Species of Amblyptevus, and other Fossil Fish remains from
Pitchorthie, Fife. Tvans. Edinb. geol. Soc., 2, 1 : 119-124, 1 pl.

Warp, J. 1890. The Geological features of the North Staffordshire Coalfields, their organic
remains, their range and distribution, with a Catalogue of the Fossils of the Carboniferous
System of North Staffordshire. Tvans. N. Staffs. Instn Min. Mech. Engrs., Newcastle,
10: iv + 189 pp., 9 pls.

1905. In Gibson, W. Palaeontology of the Pottery coalfield. Mem. geol. Suw. U.K.,
Geol. N. Staffs. Coalfields : 285-357, pls. 5-8.

Waterston, C. D. 1954. Catalogue of Type and Figured Specimens of Fossil Fishes and
Amphibians in the Royal Scottish Museum, Edinburgh. Tvans. Edinb. geol. Soc., 16: x +
QI pp.

Watson, D. M.S. 1925. The structure of certain Palaeoniscids and the relationships of that

group with other bony fish. Pyoc. zool. Soc. Lond., 1925 : 815-870, 2 pls., 30 figs.

1928. On some points in the structure of Palaeoniscid and other allied fish. Proc. zool.
Soc. Lond., 1928 : 49-70.

WEsTOLL, T. S. 1937. On a remarkable fish from the Lower Permian of Autun, France.
Ann. Mag. nat. Hist., London (10) 19 : 553-578, 7 figs.

1944. The Haplolepidae, a new family of Late Carboniferous bony fishes. A study in

taxonomy and evolution. Bull. Am. Mus. nat. Hist., New York, 83: 1-122, pls. I-10.

206 PALAEONISCOID FISHES AND THE CHONDROSTEI

Waiter, E. I. 1927. The Fish-Fauna of the Cementstones of Foulden, Berwickshire. Tvans.
R. Soc. Edinb., 55 : 257-287, 46 figs.

1932. On a new Triassic fish from North-East Madagascar. Ann. Mag. nat. Hist.,
London (10) 10 : 80-83, 1 fig.

1939. A New Type of Palaeoniscoid Fish, with Remarks on the Evolution of the Actinop-
terygian Pectoral Fins. Pvoc. zool. Soc. Lond. (B) 109 : 41-61, 1 pl.

1958. Original Environment of the Craniates In Studies on Fossil Vertebrates : 212-234,
8 text-figs., Univ. London, Bristol.
Witson, J. A. 1953. Permian Vertebrates from Taylor County, Texas. J. Paleont., Tulsa,

27, 3: 450-470, I5 figs.

Woopwarp, A.S. 1889. On Atherstonia, a new Genus of Palaeoniscid Fishes from the Karoo
Formation of South Africa. Ann. Mag. nat. Hist., London (6) 4: 239-242, pl. 14.

1891. Catalogue of the Fossil Fishes in the British Museum (Natural History), London, 2.
xliv + 597 pp., 16 pls. Brit. Mus. (Nat. Hist.), London.

1915. The Anniversary Address of the President 1914-1915. Proc. geol. Soc., London,
1915 : lii—lxxiv, 2 figs.

1931. On Urosthenes, a Fossil Fish from the Upper Coal Measures of Lithgow, New South
Wales. Ann. Mag. nat. Hist., London (10) 8 : 365-367, pl. 14.

1942. The beginning of the Teleostean fishes. Ann. Mag. nat. Hist., London (11) 9 : 902-
912, 7 figs.
Woopwarp, A. S. & SHERBORN, C. D. 1890. A Catalogue of British Fossil Vertebrates.

xxxv + 396 pp., London.

20

S\

PLATE tr

Sphaerolepis kounoviensis Frié.
Photograph of cast of one of the syntypes. %2°8.

Bull. Br. Mus. nat. Hist. (Geol.) 14, 5 IDL ANAS,

21

GEOL. 14, 5.

PLATE 2

Sceletophorus biserialis Fri¢.
Photograph of cast of one of the syntypes. x 3°5.

Bull. Br. Mus. nat. Hist. (Geol.) 14, 5 PLATE 2

PLATE 3

Sceletophorus biserialis Fric¢.
Photograph of cast of one of the syntypes. 3'1.

Bull. Br. Mus. nat. Hist. (Geol.) 14, 5 IPL JIE IS 33

R= tlimnaeillialinneter

y i ‘
/ ~

wv

1

tay

AS ey]
og Se Coe, |
: !
mes:
vay }
bey
th ft
' 3
r
F 7
Pa
i
i
,
‘
‘I b
Py '
ain
y
id,
4
, 4
x
ay rr
q
¥ WA
i

era eho apenas SUP eceemmann yt) bs

Se Nee

____ LOWER CARBONIFEROUS
4 -TRILOBITES OF NORTH DEVON
_ AND RELATED SPECIES FROM
BS. NORTHERN ENGLAND

J. E. PRENTICE

eS BULLETIN OF
| BRITISH MUSEUM (NATURAL HISTORY)

Vol. 14 No. 6
LONDON : 1967

EOWER 'CARBONIFEROUS TRILOBITES OF
NOW DEVON AND RELATED: SPECIES
FROM NORTHERN ENGLAND

BY

JOHN EDWARD PRENTICE, Ph.D.

King’s College, University of London

Pp. 207-241; 7 Plates; 3 Text-figures

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)
GEOLOGY Vol. 14 No. 6
LONDON : 1967

THE BULLETIN OF THE BRITISH MUSEUM
(NATURAL HISTORY), tmstituted im 1949, 1s
issued in five series corresponding to the Departments
of the Museum, and an Historical series.

Parts will appear at irregular intervals as they become
veady. Volumes will contain about three or four
hundred pages, and will not necessarily be completed
within one calendar year.

In 10965 a separate supplementary series of longer
papers was instituted, numbered serially for each
Department.

This paper is Vol. 14, No. 6 of the Geological
(Palaeontological) series. The abbreviated titles of
periodicals cited follow those of the World List of
Scientific Periodicals.

World List abbreviation
Bull. Br. Mus. nat. Hist. (Geol.).

© Trustees of the British Museum (Natural History) 1967

TRUSTEES OF
THE BRITISH MUSEUM (NATURAL HISTORY)

Issued 13 June, 1967 Price £1 15s.

row ER CARBONIFEROUS TRILOBITES OF
NOREH DEVON AND RELATED SPECIES
FROM NORTHERN ENGLAND

By JOHN EDWARD PRENTICE

CONTENTS
Page
I. INTRODUCTION ‘ F 3 2 6 : : 209
II. SySTEMATIC DESCRIPTIONS . : : > - 5 PALI
Family PHILLIPSIIDAE . : c : , : EsTgT
Genus PHILLIPSIA . : : c : : 5 Bin
Family PROETIDAE : 6 : : : : 216
Subfamily CYRTOSYMBOLINAE : : : : . 216
Genus Cyrtosymbole . : : : : : 216
Subgenus Macrobole ; . : ‘ : S206
Subgenus Waribole : 5 ; i : 258
Genus Phillibole : 2 : 3 é : 5 PRO)
Genus Liobole . é : : ; F : e227
Genus Spatulina ‘ : : . : : 289
Genus Typhloproetus : : c : : 5 §6@aG
Genus Diacoryphe . : : 0 . ¢ 0 —36)
III. StTRATIGRAPHICAL CONCLUSIONS 5 5 : é 235
IV. REFERENCES : ; c 0 . : : ez3o
SYNOPSIS

A systematic description is given of the known Carboniferous trilobites from Devon.
Phillipsia leet, Phillibole coddonensis and Spatulina spatulata, originally described by Woodward,
are redescribed, together with Phillibole polleni and Diacorphye ? vandergrachtii from northern
England. Macrobole cf. brevispina, M. cf. laticampa, Phillibole aprathensis and Liobole glabra
previously described from Poland and Germany, are recognized among the specimens from
Devon and described. Phillibole culmica from Germany is identified with P. coddonensis.
Three new species, Waribole chudleighensis, Spatulina longispina and Typhloproetus cephalispina,
are described.

Five successive trilobite faunas are tentatively recognized in the Devon Carboniferous, and
their ages relative to the goniatite time-scale discussed. They are characterized by (i) Macrobole
aff. laticampa (low zone II), (ii) Spatulina (II—IIIa), (iii) Phillibole coddonensis (IIIa) (iv) P. apra-
thensis (IIIa) and (v) Phillipsia leei (III f-IITy).

I INTRODUCTION
Tue Carboniferous trilobites of the British Isles can be divided into two groups,
whose distribution appears to be closely linked with lithological facies. On the one
hand are the large, robust, thick-shelled and heavily ornamented species, often with
large eyes, which are found in the limestones of “ reef” and “ massif” facies of the
Carboniferous limestone. On the other hand, in the radiolarian cherts and shales
which characterize the Lower Carboniferous of south-west England is found a
different group, which occasionally penetrated northwards in a shale facies to
northern England. This latter group comprises generally small forms, their tests
GEOL. 14, 6. 22

210 LOWER CARBONIFEROUS TRILOBITES

thin and fragile, mostly smooth, and their eyes small or absent. It is reasonable to
assume that these represent a hemi-pelagic fauna of the open sea, whilst the former
group is part of a shallow water benthos. This account sets out to describe the
hemi-pelagic group in detail, and to examine their stratigraphical distribution. In
contrast to the “ limestone ”’ fauna, which is dominated by the Family Phillipsiidae,
the trilobites of south-west England consist largely of members of the Cyrto-
symbolinae ; the latter sub-family, however, is by no means confined to this facies,
and includes many robust and large eyed forms of both Devonian and Carboniferous
age.

A reappraisal of the classification of some of these trilobites has been made by
Hahn (1965). In his scheme the following species described herein would be classified
as follows:

Cyrtosymbole (Macrobole) aff. laticampa = ? Carbonocoryphe

Cyrtosymbole (Macrobole) cf. brevispina = Archegonus (Phillibole) brevispina
group

Cyrtosymbole (Waribole) chudleighensis = Archegonus (Waribole)

Phillibole aprathensis

Phillibole pollent = Archegonus (Phillibole) aprathensis group.

Phaillibole coddonensis

The first detailed descriptions of these trilobites were by Woodward (1884), who
erected three new species of Phillipsia in an appendix to his monograph of Carboni-
ferous trilobites, thus tacitly commenting upon their separateness. The collections
from the radiolarian cherts of Devon by Hinde & Fox (1895) enabled Woodward to
indicate the presence of several new forms, and later (1902), using the more extensive
collections of the local amateurs J. G. Hamling and A. Coomaraswamy, he was able
to give more complete descriptions of these species. In 1909 I. Thomas described
comparable faunas from South Devon, but since that time little attention has been
paid to these fossils in this country. In the course of major revisions of Upper
Palaeozoic trilobites in Germany, R. & E. Richter (1937-1951) described many new
genera and species, making comparisons with the specimens described by Woodward.
Similar work has been carried out in Czechoslovakia by Pribyl (1950) and Chlupac
(1961), and in Poland by Osmdlska (1962). The stratigraphical revision of the North
Devon succession by J. M. Thomas, E. E. Swarbrick and the writer has established
the horizons of the described trilobites more firmly, and has added much new
material (see p. 238). In the course of similar studies in the underlying Pilton Beds,
Goldring (1955) has drawn stratigraphical conclusions from the comparative work of
Richter. The northern representatives of this fauna were described first by
Woodward (1894), but since then they have been mentioned only briefly in strati-
graphical descriptions in Geological Survey Memoirs (e.g. Calver & Ramsbottom
1962).

When preserved in chert, or in cherty shale, the test of these trilobites is commonly
replaced by finely crystalline silica ; in consequence the outer lamina of the test
adheres firmly to the external mould, and the inner lamina to the internal mould.
Thus the external view of the outer lamina is rarely seen. In contrast, specimens
preserved in shales are found either as normal internal or external moulds, or with

LOWER CARBONIFEROUS TRILOBITES 211

their dorsal exterior exposed. Thus the mode of preservation can fundamentally
alter the appearance of the fossil, and this has given rise to much confusion in the
past. Thus the “ brim”’ of the cephalon or pygidium as described by Woodward
often proves to be the mould of the ventral doublure. Moreover, there are sometimes
structural differences between the outer and inner lamina. In Phillibole polleni
for example, the outer surface of the glabella is smooth and unfurrowed, while the
inner lamina bears strong furrows; in Phillipsia lee: the interpleural furrow sof
the pygidium are visible on the internal but not on the external lamina. Thus the
greatest care is necessary in the examination and description of this material, and
comparison with earlier descriptions is by no means easy.

II SYSTEMATIC DESCRIPTIONS
Conventions and abbreviations

The conventions adopted in the systematic descriptions are those used in the
Treatise of Invertebrate Palaeontology (Harrington 7m Moore 1959). For descrip-
tion of facial sutures the a—/—y—d-e-w system is adopted from R. & E. Richter
(1949: 68). Lateral glabellar lobes and furrows are numbered from posterior to
anterior, Ip being the first anterior to the occipital furrow. The directions sagittal
(abbreviated sag.), exsagittal (abbreviated exsag.) and transversal (abbreviated #7.)
are indicated where doubt exists of the meaning of width (W.) or length (L.).

Described material is deposited in museum collections ; these are indicated by the
following abbreviations.

BMNH—British Museum (Natural History)
GSM—Geological Survey & Museum
NDA—North Devon Athenaeum, Barnstaple
KCL—King’s College, London

Family PHILLIPSIIDAE Oehlert 1886 emend. Hupé 1953
Genus PHILLIPSIA Portlock 1843 emend. Weller 1936
Type SPECIES. Phuillipsia kellii Portlock by subsequent designation, Vogdes 1890.

Phillipsia leei Woodward
(Pl. 1, figs. 1-6 ; Text-fig. 1)

1884 Phillipsia leet Woodward : 66-68, pl. 10, figs. 1-4.

1884 Phillipsia minor Woodward : 68, pl. 10, figs. 5, 6a, b, 7, 8a.

1884 Phillipsia clifford: Woodward : 69, pl. 10, figs. 8b, 9-12.

1884 Phillipsia articulosa Woodward : 70, pl. 10, figs. 6c, d, 13.

1909 Phillipsia minor Woodward ; Thomas: 201, pl. 7, fig. 11.

1909 Phillipsia cf. minor Woodward ; Thomas: 201-2, pl. 8, figs. 12, 13.
1909 Phillipsia sp., Thomas: 202, pl. 7, fig. 14.

DracGnosis. Small Phillipsia with internal punctation but no external granula-
tion to carapace; glabellar furrows other than Ip very indistinct or absent ;
pygidium with clearly defined border ; pygidial and cephalic borders with longi-
tudinal striations on underside.

212 LOWER CARBONIFEROUS TRILOBITES

MATERIAL. The specimens upon which Woodward based his original description
was obtained from Waddon Barton Lane, near Chudleigh, Devon by Mr. J. E.
Lee. Some of these were presented by Woodward to the British Museum (Natural
History) at the time, while others remained in Lee’s private collection at Torquay.
On Lee’s death, his collection was transferred to the British Museum.

Specimen BMNH. In. 58281, figured by Woodward (1884, pl. 10, fig. 2) is here
selected as lectotype ; this is one of the few specimens which displays an external
view of the external shell-lamina, and includes some attached thoracic segments
(Pl. 1, fig. 1). Specimen BMNH. In. 5280, one of the syntypes figured by Woodward
(pl. 10, fig. 3) shows additional cephalic features and the internal view of the outer
shell-lamina is here illustrated (Pl. 1, fig. 2). Specimen BMNH. In. 58283, figured as
P. cliffordi by Woodward (1884, pl. 10, fig. 10), here considered to be a synonym of
P. leet, shows the pygidial characters of the species. P. articulosa BMNH. I. 1861
(Woodward, pl. ro, fig. 13) and P. minor BMNH. I. 1860 (Woodward, pl. 10, fig. 7c)
are also considered to be synonyms of P. Jeez. The Waddon Barton material con-
tains, in addition to the type specimens, 6 nearly complete cephala, 7 cranidia, 6 free
cheeks, 10 fragments of thorax, and 46 pygidia. In addition two specimens from
Waddon Barton are in the Museum of the Geological Survey, where also, those
figured by I. Thomas (1909) are preserved. Additional material from North Devon,
has been collected by theauthor and by J. M. Thomas and is preserved at King’s
College. A further 10 specimens have been found by M. R. House in excavations
in Ugbrooke Park, Chudleigh. A single headshield was collected from a comparable
horizon in the north-east Rhenish Schiefergebirge by G. Warrington.

HORIZON AND LOCALITIES.

I. Red-stained black cherty shales, associated with Neoglyphioceras spirale and
Mesoglyphioceras aff. granosus i.e. Pyg—P, (Butcher & Hodson 1960) at Waddon
Barton Lane, Chudleigh, Devon, in sides of lane leading south from Waddon, one
mile east of Chudleigh. Nat. Grid Ref. SX 885793. Also similar shales in various
trenches in and around Ugbrooke Park, Chudleigh.

2. White chert: Horizon unknown at Hestow Farm, Ideford, Devon. One
mile due south of Ideford, near Chudleigh, Devon. Nat. Grid Ref. SX 888761.

3. Upper Shales and limestones associated with N. spfivale at Whipcott Quarry,
Westleigh, N. Devon. Nat. Grid Ref. ST 075188 (J. M. Thomas collection).

4. Upper Shales and limestones, associated with Posidonia bechert. Hole Lake
Farm Quarry, Staple Cross, Huntsham, N. Devon. Nat. Grid Ref. ST 022208
(J. M. Thomas collection).

5. Rhenaer Kalk (Bed 23, Meischner 1962) associated with G. spirale. Aartal-
strasse Adorf-Flechtdorf, north-east Rhenish Schiefergebirge.

Description. Cephalon. Outline nearly semicircular, slightly shorter (sag.)
than width (tr.) ; posterior border almost straight. Glabella does not reach anterior
margin: outline rounded anteriorly, almost parallel sided but slightly constricted
medianly ; maximum width less than half length. Anterior of glabella high and
globose, lower posteriorly. Occipital segment of glabella clearly demarcated by deep

LOWER CARBONIFEROUS TRILOBITES 213

occipital furrow which is slightly bent forward centrally ; smoothly rounded, narrow
(sag.). Ip furrows very deep, backwardly directed to join occipital furrow at points
1/3 along its length ; cut off high triangular glabellar lobes. 2p furrows very faint
or absent. Axial furrow sharp, separating glabella from almost flat fixed cheek.
Glabella reaches forward to touch narrow, smooth, raised anterior rim. Facial
suture cuts anterior margin at a point (w) immediately in line (exsag.) with the
forward projection of the sides of glabella at its minimum width, swings outwards
sharply to point # which lies in line (tr.) with the maximum width of the glabella,
gently curves round to subparallel, slightly convergent with axial furrow to eye
region, which lies in line or immediately posterior of rp furrows. From posterior of

js a ee a |
One Centimetre

Fic. 1. Phillipsia leei Woodward. Reconstruction.

eye (i.e. immediately in front of occipital furrow) swings sharply outwards to cut
posterior margin of headshield (w) a short distance inside angle of genal spine.
Thus fixed cheek has sharply pointed triangular extension in occipital region. Eye
moderately small, less than } length of headshield, reniform, multilensed, situated
on a raised platform surrounded by a shallow depression. Free cheek gently in-
flated, with shallow border furrow and deep pleuroccipital furrow meeting at high
angle in genal region. Brim flattened with rounded external margin ; on underside
carries fine subparallel longitudinal striae, six or seven laterally, diminishing to 4

214 LOWER CARBONIFEROUS TRILOBITES

anteriorly : 3 or 4 striae continuous with striae on underside of pre-glabellar brim
and of genal spine. Spine in direct continuity with cephalic brim ; broadly based
tapering evenly to a sharp point ; length variable from less than half to nearly equal
to sagittal length of cephalon ; striae on underside externally continuous with those
on cephalic brim ; inner striae bend sharply round at genal angle to parallel occipital
margin. Whole of interior of cephalon except brim ornamented with a fine indistinct
punctation, showing no definite arrangement on cheeks, but a vague alignment into
transverse rows on the underside of the glabella. Hypostome has spatulate outline,
with strongly elevated median body and flat wings: anterior border convex for-
ward, antero-lateral angles a little more than 90°; posterior border a smoothly
rounded parabola. Median body high, subrectangular outline, touches anterior
border.

Thorax. Consists of 9 segments: axis clearly defined, distinctly higher than
pleural region; axis approximately 1/3 width (tr.) of thorax. Axial rings with
raised posterior border, sloping forward to a deep straight transverse furrow which
divides it from articulating half-ring, which is narrow (sag.) and triangular with a
strong rounded ridge anteriorly. Pleural regions moderately flat, pleural segments
nearly straight with a broad triangular articulating flange at anterior and distinct
posterior ridge. Pleural grooves strong and deep, subcentral, slightly oblique.
Pleural extremity truncated, with a short backwardly directed spine.

Pygidium. Outline hemielliptical, somewhat wider (tr.) than long (sag.). Anterior
margin nearly straight, inflected forward into a broad triangular flange in the pleural
region, deflected slightly forwards in axial region ; antero-lateral angles smoothly
rounded. Axis 1/3 or more of width (tr.) anteriorly, tapering rapidly to a smoothly
blunt extremity in contact with posterior brim. Axial rings Io to 14, separated by
straight furrows which are quite distinct except at extreme posterior. Axial furrow
distinct. Pleural regions flatly arched, with 8—1o clear pleural grooves ; anterior
groove is nearly parallel with anterior margin, posterior grooves become increasingly
oblique. Pleural ridges strongly asymmetrical, with steep slope facing backwards.
Pleural grooves become somewhat fainter posteriorly, but are never effaced ; each
groove ends abruptly at the pygidial brim. Interpleural furrows sometimes visible
as a faint ridge on the underside of the pygidium, at most over whole length (tr.) of
pleural segment anteriorly, but mostly only at extremities of first two or three
segments ; at the brim they are sharply deflected backwards. On the internal
mould these interpleural furrows make the pleural grooves seem bifid. Pygidium
with a narrow brim, without clearly defined furrow on dorsal side. Ventral doublure
on underside flat, narrow anteriorly widening rapidly posteriorly, but without trace
of mucronation. Underside of brim carries fine slightly irregular subparallel longi-
tudinal striae, 4 or 5 external striae continuous around pygidium; additional
striae, up to 9 in all, inserted at inner side as brim broadens to posterior. Whole
underside of pygidium except brim finely punctate. Axial rings have up to Io
moderately large widely spaced punctae along their posterior margins; in the
posterior rings these punctae are anterioposteriorly elongate and occupy the whole
axial ring length (sag.). In each pleural segment there are three transverse rows of
punctae ; the coarsest are immediately anterior of the intersegmental boundary,

LOWER CARBONIFEROUS TRILOBITES 215

the other two rows are finer, none occurring anterior of the pleural groove. The
punctae are fainter in more posterior segments. Appendage muscle scars strongly
developed on one specimen (BMNH. In. 55935, Pl. 1, fig. 7) as shallow circular
depressions on sides of front five axial rings.

DIMENSIONS (in mm.)

Cephalon Pygidium
a SSS — SF Axis
BMNH L. (sag.) W. (tr.) L. (sag.) W. (tr. ant.) W. axis (tr. ant.)
In. 58281 6:0 II-2
I. 864 2°4 4°8
I. 1064 4°5
I, 1090 5:0 555 18
I. 1092 5:0
In. 55902 4:8 3:0
In. 55903 48 7:2 2°I
In. 55914 9:0 39
In. 55917 4:0 7:2 2°5
In. 55919 5°3 12°4 3°2
In. 55932 103 16-4 9:0 12:5
In. 55935 8-9 9:9
In. 55938 4°9 6-0
GSM
23434 4:0 4°0

REMARKS. Woodward (1884) identified four species from among the Waddon
Barton material. In view of the continuity of variation displayed by this
assemblage, however, I would regard them as constituting one species. Phillipsia
clifordi was distinguished solely on pygidial characters. Woodward appears to
have been misled by the different aspects afforded by the internal moulds, and to
have given the name clifford: to those moulds which show an apparently broader
brim, and in which the traces of the intersegmental boundary give the appearance of
bifurcation to the pleural grooves. P. minor appears to have been distinguished by
the characters of the head-shield, but the differences were not specified. The curva-
ture of the genal spine in pl. 10, fig. 6a is, as is apparent from the counterpart
(BMNH. I. 1864), due to the superimposition of a thoracic segment upon the genal
angle. The original of pl. ro, fig. 7c (BMNH. I. 1860) shows that this same feature is
a misrepresentation. P. articulosa was described as having 17 coalesced segments
in the pygidial axis. No specimens in the Lee collection have this number, and
indeed the two specimens in figure 6, whose counterpart has been preserved, show
the normal 12—13 (as in fact Woodward's figure shows) while the original of figure 13
(BMNH. I. 1861) shows that 3 thoracic segments have been drawn by the artist
as part of the pygidium.

I. Thomas (1909) followed Woodward’s definitions but suggested that one of his own
specimens might be separable as it possesses a strong second pair of glabellar furrows.
Examination of the specimen (GSM. 23436) leaves no doubt that these furrows were
produced by the crushing of the carapace into the first pair of furrows.

216 LOWER CARBONIFEROUS TRILOBITES

The generic and family affiliations of this species are difficult to determine. This is
due partly to the different definitions of the possible groups given by various authors.
It would fit within the family Phillipsidae as defined in the Treatise of Invertebrate
Palaeontology (Weller 7m Moore 1959 : 399) except for the absence of more than one
pair of glabellar furrows. However, this feature is not regarded as diagnostic by
Hupé (1953) whose description in an amended form is purported to have been used
in the “ Treatise’’. Alternatively Hupé does insist that a granular test is charac-
teristic of the family, which the “ Treatise’ definition does not make essential.
Despite these minor points, however, the form of the glabella, the relatively narrow
pre-glabellar region, and the multisegmented pygidium, place the species clearly in
this family.

The species fits the definitions of the genus Phillipsia as emended by Weller
(1936 : 704; 1959: 399) except in having a well defined border or flange to the
pygidium. In fact, the holotype of Phillipsia kellia Portlock 1843 (GSM. 63045)
the type species of Phillipsia by subsequent designation, does have a quite clearly
defined, though narrow, pygidial border. It does, however, possess 3 clearly
defined glabellar furrows and a very well-marked surface granulation. Since neither
of these characteristics is regarded by Weller (1936, 1959) as diagnostic of the genus,
the most satisfactory course seems to be to retain the species Jeez in the genus
Phillipsia. The present species differs in many ways from the very closely defined
generic diagnosis of Reed (19420), who based his definition upon Asaphus gemmuliferus
of Phillips. Reed’s emendation is framed in such detail as to exclude all but the
type species and I prefer to follow the less rigidly defined diagnosis of Weller (1959).
The two genera Paladin Weller 1936 and Kaskia Weller 1936 possess similarities to
the present species, and have a more clearly defined pygidial border than Phillipsia
kellit, but both have very large eyes occupying a large area of the free cheek, are
strongly granulose externally and have an anterior expansion of the glabella. The
species has some characters in common with Weberides Reed 1942), notably in the
internally striated border to the cephalon and pygidium ; but again the small size of
the eyes, and the absence of anterior glabella expansion exclude it from this genus.

Family PROETIDAE Salter 1864
Subfamily CYRTOSYMBOLINAE Hupé 1953
Genus CYRTOSYMBOLE R&. Richter 1913
Subgenus MACROBOLE R. & E. Richter 1951

TYPE SPECIES. Cyrtosymbole (Macrobole) drewerensis R. & E. Richter 1951.
Macrobole, first proposed by R. & E. Richter (1951), has subsequently been given a
fuller diagnosis (¢n Moore 1959). Many species of this genus have been described from
Germany, Poland and Czechoslovakia, but it is difficult to refer the few specimens
from Devon to any of those species which in any case are not mutually comparable.

LOWER CARBONIFEROUS TRILOBITES 217

In view of the paucity of material it has been thought best not to erect new species,
but to describe briefly individual specimens and note their affiliations.

The majority of the species belonging to this subgenus are found in Zone I of the
Carboniferous, although some range into Zone II. Only if Phillibole culmica R. & E.
Richter is accepted as a Macrobole could the subgenus be said to range into Zone III
(see p. 238). In fact it is not easy to draw a line between Phillibole and
Macrobole on cephalic characters, the former seemingly having evolved from the
latter : but the strongly segmented tail of Macrobole contrasts markedly with that of
Phillibole.

Cyrtosymbole (Macrobole) cf. brevispina Osmdlska
Plv2) figse1, 2:

1960 Cyrtosymbole (Waribole) cf. aequalis (Meyer) ; Prentice: 271, pl. 12, fig. 5.
1962 Cyrtosymbole (Macrobole) cf. brevispina Osmdlska: 146, pl. 13, fig. I.

Diacnosis. Macrobole with clearly segmented glabella, moderately wide (long.)
preglabellar field, pointed anterior margin to cranidium, short pointed genal spine.

MATERIAL. The specimen (KCL t18z2) previously figured by me (Prentice 1960)
is presumed to have come from Bed X (see p. 238) at Park Gate Quarry,
Tawstock, and consists of a crushed cranidium. Another cranidium from the same
quarry (BMNH. I. 3223) contains on the same surface a free cheek presumed to
belong to the same individual, together with two free cheeks and a pygidium of
Spatulina spatulata.

Horizon. Osmdlska’s specimens came from the upper beds of the Pericyclus
Zone (i.e. the German zone II) ; while the Tawstock horizon has been correlated by
the writer with the boundary of zones II and III «.

REMARKS. The broad-based, tapering glabella, with clearly defined occipital
furrow and distinct backwardly directed Ip, 2p and 3p glabellar furrows are features
which place the North Devon specimens clearly in the Macrobole group. The facial
suture, expanded widely in front, lying close and parallel to the axial furrow in the
palpebral region, and then extending rapidly outwards posteriorly, is also charac-
teristic of the genus. The genera Waribole and Archegonus (to which the species
aequalis is now referred) have facial sutures which diverge less abruptly from the
axial furrow anteriorly and posteriorly, the eyes and palpebral lobes of Waribole
are much larger, and the occipital lobe of the glabella is not of equal width throughout.
Osmdlska noted the presence of an anterior point to the cranidium in larval
specimens of M. brevispina, but this is not evident in specimens of comparable size
to the Devon examples. She does, however, refer to 7 cranidia as “ cf. brevispina”’
which attain the length of 2 mm. without losing the anterior spine. The Devon
specimens resemble these in their wide preglabellar field, and broadly rounded
glabella front.

218 LOWER CARBONIFEROUS TRILOBITES

Cyrtosymbole (Macrobole) aff. laticampa Osmolska
Bin2 aies. 3) 5.

1962 Cyrtosymbole (Macrobole) laticampa Osmolska: 139, pl. 11, figs. 1-5.
1962 Cyrtosymbole (Macrobole) ? laticampa Osmdlska: 141, pl. 11, fig. 7.

REMARKS: The specimens found in Devon consist of one cranidium preserved as
an internal mould with the internal layer attached (PI. 2, fig. 3), and its counterpart
an external mould with outer layer attached, together with a portion of a cephalon
(Pl. 2, fig. 5) which doubtfully belongs to the same species. The cranidium closely
resembles M. laticampa, especially in having a wide (sag.) preglabellar field ; but
this field carries a broad, deep excavation across its anterior which is unknown in
M. laticampa. The part cephalon is larger than the cranidium, but its deeply lobed
glabella and occipital segment of equal width (sag.) place it in Macrobole; it has
however a short librigenal spine, whilst that of M. laticampa is long.

Both specimens were found in a thin bed of white chert in an old quarry in Claypit
Coverts, West Buckland (SS 662291) associated with goniatites of “ pericyclid”’
type. The horizon is estimated to be a few feet above the base of the Chert Forma-
tion. M. laticampa in Poland is restricted to the Gattendorfia Zone (i.e. German
zone I), but there is no other evidence to suggest that the Chert Formation begins in
this zone: an attribution to zone II is therefore the most likely for this horizon.

Genus CYRTOSYMBOLE R. Richter 1913
Subgenus WARIBOLE R. & E. Richter 1926
TYPE SPECIES. Cyrtosymbole (Waribole) warsteinensis R. & E. Richter 1926.
Cyrtosymbole (Waribole) chudleighensis sp. n.
PIP 2; fig. 7-

DiaGnosis. Minute Cyrtosymboliinids with smooth inflated glabella and widely
diverging anterior branches to facial sutures.

HOLOTYPE and MATERIAL. The species is known from one block of specimens
only, M. R. House collection 1100, BMNH. It. 1433, from Pit L (of House & Butcher
1962), at the eastern end of the field immediately north-east of Mount Pleasant,
Chudleigh, Devon. The block contains abundant fragments of Posidonia, and
is assigned to the “ P’”’ shales of the succession. Four cranidia, with attached and
detached thoracic segments, 3 fragments of free cheeks and one pygidium are
present all lying within 1 sq.cm. The cranidium figured in PI. 2, fig. 7a is taken as
holotype. The two larger free cheeks probably belong to Phillipsia leei (see p. 211)
which also occurs at this horizon.

DESCRIPTION. Cephalon. Very small size, length (sag.) 1-3, I-4 and r-6 mm.
Glabella strongly inflated, parallel-sided, with a steep, rounded, blunt anterior
extremity. Occipital furrow distinct, strongly curved forward in centre, so that
occipital segment is much wider (sag.) in centre than at sides. Occipital segment

LOWER CARBONIFEROUS TRILOBITES 219

same width (tr.) as glabella. Anterior of glabella smoothly rounded and without
furrows, Ip furrows shallow, extend diagonally from sides of glabella to join occipital
furrow, thus cutting off small triangular lobe. Pre-glabellar field strongly concave,
with narrow but distinct enrolled anterior rim. Facial suture cuts anterior border
at some distance from the mid-point, then takes an almost semicircular course
laterally and posteriorly, swinging inwards to the anterior of palpebral lobe (y) at
a point half way along length (exsag.) of glabella, and close to the axial furrow.
From the posterior of the palpebral lobe (e) the suture swings outwards almost
parallel with the posterior margin of the cephalon ; i.e. almost along occipital furrow ;
then gently back to cut posterior margin at a shallow angle distally. Fixed cheek
is thus broad and elliptical in front, very narrow in palpebral region, and produced
to a long point occipitally. The concavity of the preglabeller field extends on to
the anterior of the fixed cheek, which rises along the sides of the glabella. Palpebral
lobe high, crescentic, with two distinct palpebral ridges extending from inner side
on to the rising flanks of the glabella immediately in front of the 1p furrow. Free
cheek with strong marginal concavity and thin narrow rim. Genal angle probably
with short spine.

Thorax. Number of segments unknown. Each segment has high arched axis,
slightly less than 4 of the total width of the thorax. Axial furrow deep. Axial
ring has deep transverse furrow and an anterior ridge ; articulating half-ring narrow
(sag.). Pleural regions shallowly arched, pleural furrows broad and deep. Pleural
extremities obscure, perhaps truncate.

Pygidium. Short and subquadrate. Anterior margin strongly curved, with
rounded lateral angles. Posterior margin a flattened curve, with a broad concave
brim which reduces in width rapidly laterally and anteriorly. Axis short, conical,
sharply tapering, with seven axial rings, each with a deep axial furrow. Pleural
furrows six, deep and distinct, asymmetrical with steeper slope facing anterior.
Anterior furrows reach margin, posterior ones become diagonal and are effaced
laterally. Four most anterior interpleural furrows clearly defined.

Discussion. The generic affiliation of this species is problematical. It belongs
with the Cyrtosymbolinae rather than the Phillipsiidae because of the presence
of a pre-glabellar field, and the small number of segments in the pygidium. It is
tentatively placed with the subgenus Waribole because it resembles the type-
species W. warsteinensis in the following features: the possession of a strongly
diagonal pair of rp glabellar furrows, the position of the palpebral lobes, the lateral
narrowing of the occipital ring, the short pygidial axis with few segments, and the
broad pygidial border. If differs from this species mainly in having a parallel-sided
glabella, and in the strongly diverging anterior branches of the suture line. In this
latter feature it resembles Carbonocoryphe R. & E. Richter, the cranidium figured by
R.&E. Richter (1950, pl. 1, fig. 12) as Carbonocoryphe? ferruginia showing a somewhat
similar wide fixed cheek ; but the pygidial characters of this genus are quite unlike
this species. A cranidium described under the name of Cyrotsymbole librovitchi var.
latilimbata by Weber (1937 : 30, text-fig. 15) is very similar, but the associated
pygidium has no widened posterior border ; that of Cyrtosymbole librovitchi var.

220 LOWER CARBONIFEROUS TRILOBITES

euryaxis (Weber 1937 : 30, text-fig. 14) has the same shape as the present species,
but the axis does not taper so strongly. Phillipsia krasnopolskit Weber 1937 has
some similarities, notably in the brim of the pygidium and the shape of the glabella,
but has no pre-glabellar field and 10-11 axial segments in the pygidium.

Genus PHILLIBOLE R. & E. Richter 1937 : 108
Phillibole aprathensis R. & E. Richter
Pl. 7, figs. 6-9.

1882 Phillipsia aequalis Meyer; Kayser: 68, pl. 3, figs. 7, 8.

1882 Phillipsia aff. Eichwaldi Fischer ; Kayser: 73, pl. 3, fig. 6.

1895 Phillipsia cliffordt Woodward ; Woodward: 646, pl. 28, fig. 3a only.

1902 Phillipsia polleni (?) Woodward : 482, pl. 20, fig. 2, 13 only.

1932 Cylindraspis aprathensis R. & E. Richter ; Haubold: 216, 220, 223, 240 (nomen nudum).
1932 Cylindvaspis aprathensis R. & E. Richter ; Kobold: 484, 508, (nomen nudum).

1937 Phillibole aprathensis R. & E. Richter: 109, text-figs. I, 2.

Diacnosis. Phillibole with rounded triangulate headshield: tapering glabella
with slight but distinct median constriction ; long, distinctly segmented pygidium
with long tapering axis reaching nearly to posterior extremity.

MATERIAL. Two cranidia (BMNH. In. 228912, In. 18415) and one pygidium
(BMNH In. 18420) from Coddon Hill (SS 5729): cranidia (BMNH. I. 4559, NDA. 807
(Woodward 1902, fig. 13), and NDA 808), free cheek (BMNH. I. 4563) and pygidium
(BMNH. In. 18413) from Hannaford Quarry (SS 6029); free cheek (KCL t174) from
Templeton Quarry, Tawstock (SS 542297) ; cranidium and counterpart from a loose
block above Warrenshill Copse, Bampton (SS 978222) (BMNH It. 1434 a, 3) ;
pygidium from Upper half of Chert Beds, Trench Q, Ugbrooke Park, (House coll.
429); cranidium and part of thorax (BMNH. I. 1435) Kersdown Beds, Little
Holwell Quarry, Bampton (SS 964232).

Horizon. The holotype is from the Posidonien-schiefer (horizon IIIf) of Aprath
(R. & E. Richter 1937). Haubold (1932) records the species in association with
Gomiatites intermedius (IIla/P) and with G. falcatus and G. waddingtom (IIIf) ; he
also records it doubtfully with a IIIy, fauna. Kobold (1932) records the species
from IIIy, at Ecksberge in the Harz. However, the associated fauna includes
goniatite species subsequently renamed G. striatus mucronatus and G. striatus
koboldi (Pickel 1937; Ruprecht 1937), as well as G. granosum spirale; a fauna
which was attributed to the IIIf, subzone by Pickel and which compares closely
with that of P,, in the British Isles (cf. Hudson & Cotton 1945 : 275). A range from
the top of III~ (= upper P,,) to the base of IIIy (=P,4—P,,) is thus indicated for
this species. This lies within the known stratigraphical range of the Chert Formation
in Devon, from which all the specimens are derived. Ph. aprathensis has not been
found in close association with Ph. coddonensis and there is a little evidence to suggest
that it may occupy a higher stratigraphical horizon than the latter.

DESCRIPTION OF DEVONSHIRE MATERIAL. Glabella long, with broad base, evenly
attenuated with slight median constriction; anterior rounded slightly pointed.

LOWER CARBONIFEROUS TRILOBITES 221

Anterior of glabella higher than posterior. Occipital segment wide (sag.), slightly
wider at sides than in centre ; occipital furrow almost straight, distinct. Trace of
occipital spine or pustule on centre of occipital segment. Glabellar furrows faint ;
1p curved backwards to cross occipital furrow at less than a third the transverse
distance from the sides, then projected straight back across the occipital segment to
the posterior margin; 2p strongly bent back to join Ip, 3p slightly less strongly
curved backwards. Axial furrow clear and deep. Preglabellar field narrow and
concave, rising gently to an enrolled brim. Fixed cheeks rise from the anterior
concavity along sides of glabella, flat behind. Facial suture begins (~) within the
projection forward of the axial furrow at the mid-point (exsag.) of the glabella,
swings sharply outwards to /, which is rounded and lies behind the line of the
glabella anterior; then swings back inwards and from midline of glabella length lies
parallel and close to the axial furrow. After crossing the occipital furrow it swings
sharply outwards to cut the posterior margin some distance from the axial furrow.
Eye lobes very long and very slightly curved, almost impossible to distinguish.

Free cheek gently inflated, with shallow marginal concavity rising to an ill-defined
narrow enrolled rim. Pleuroccipital furrow straight, poorly defined; pleuroccipital
segment broad, gently inflated, of equal width throughout. Genal angle with small,
short, sharply pointed spine. Eye surface comparatively large, inner margin nearly
straight, outer margin strongly curved ; surface flat, finely but irregularly reticulate.
Ventral doublure broad, curved, with four irregularly spaced parallel striae on the
inside.

Thoracic segments with broad axis, clear axial furrow. Pleural sectors with
strong median pleural furrow.

Pygidium parabolic in outline ; axis high and distinct, broad at anterior, nearly
equal to one-third of the width of the pygidium, rapidly tapering to a moderately
blunt extremity a little short of the posterior border. 7 to 10 axial rings, transverse
furrows moderately distinct to end of axis. Traces of pygidial appendage muscle
scars on the three anterior rings. Pleural regions gently arched, pleural furrows
4 to 6, becoming effaced posteriorly ; strongly oblique, terminating at edge of pleural
field. Trace of interpleural furrows on two anterior segments. Border furrow
moderately sharp, rim enrolled with roof-shaped cross-section ; faintly longitudinally
striate. Post-axial field flat, no post-axial ridge.

Remarks. Phuillibole aprathensis is a larger trilobite than Ph. coddenensis ; it
most nearly resembles Ph. polleni, from which it may be distinguished by the tri-
angulate form of the headshield, the faint median constriction of the glabella, and the
longer pygidial axis. A closely similar form described by Stubblefield (7m Calver &
Ramsbottom 1962) as Phillobole aff. aprathensis from B, horizons in the north of
England, differs in having a slightly wider fixed cheek in the post-ocular region.

Phillibole polleni (Woodward)
Pl. 4, figs. I-4.

1894 Phillipsia polleni Woodward: 487, pl. 14, figs. 7-12.
1962 Phillibole polleni (Woodward) Calver & Ramsbottom: 180.

222 LOWER CARBONIFEROUS TRILOBITES

Dracnosis. Phillibole with nearly semicircular headshield; gently tapering
glabella without median constriction. Pygidial axis with a blunt, rounded extremity
which lies some distance in front of posterior border.

LectotyPE. BMNH It. 3714 (figured Woodward, 1894, pl. 14, fig. 9) here selected
as lectotype.

MATERIAL. The lectotype and all specimens figured by Woodward are preserved
in the British Museum (Natural History), under registered numbers It. 369-374.
Allare preserved as slightly flattened carapaces in shale, with the original structure
of the outer lamina preserved, or as external moulds of this. The structure of the
inner lamina can be rendered visible through the test by wetting the surface.

LOCALITY AND HORIZON: The species is only known from the type locality ;
the banks of the River Hodder near Stonyhurst, Lancashire. According to Calver &
Ramsbottom (1962: 180) this horizon lies above the B. hodderense Beds, in beds
equivalent in age to the Pendleside Limestone. The upper part of this limestone
belongs to the B, zone, the lower probably to B,, and Phillibole cf. polleni has been
recorded in association with Bollandoceras hodderense (Calver & Ramsbottom 1962 :
311-312) at some Pendleside Limestone localities. It may thus be assumed that the
stratigraphical horizon is at or just below the B,/B, zonal boundary (see p. 238).

DESCRIPTION. Carapace moderately large, flatly arched, with a broad, rather
prominent axis. Test moderately thin.

DIMENSIONS (in mm.) :

It. 371 It. 369 It. 370
L.cephalon . , 3 9°5 — 7:8
W.cephalon . 6 : 16:0 — I0'8
L. glabella : : : 6-0 _
L. pygidium . : ; 7:8 51 6:2
W. pygidium . 6 : T4:0 9°8 12:0
W. pyg. axis. 0 : 6-0

Cephalon nearly semicircular in outline. Glabella broad at base, tapering for-
ward, front bluntly rounded with axial furrows straight or slightly convex. Occipital
segment equal in length sagitally and exsagitally. Occipital furrow nearly straight
but composed of three equal, forwardly convex, curves. Anterior of glabella smooth,
lateral furrows not visible externally. Internally there are three short, backwardly
directed furrows. Ip begins at axial furrow a greater distance in front of occipital
segment than the length (sag.) of the latter ; gently curved backwards at first then
sharply inclined towards occipital furrow. 2p not so far in front, parallel with
lateral part of Ip; short. 3p closer still, and shorter. Preglabellar field narrow,
deeply concave rising to a high, strongly enrolled rim, which bears 3 or 4 subparallel
longitudinal striations on its upper surface. Fixed cheeks flat along sides of glabella.
Facial suture begins (a) in line with, or slightly wide of, the forward projection of the
axial furrows at the base of the glabella ; swings sharply outwards to /, so that a-/ is
very short ; then back in a smooth outwardly convex curve to very near axial furrow
at y. y-—d-e is a smooth outwardly convex curve along side of glabella reaching

LOWER CARBONIFEROUS TRILOBITES 223

nearest point to axial furrow just ahead of the pleuroccipital furrow. From here
bends sharply outwards across pleuroccipital furrow to cut the posterior margin (w) at
internal angle of genal spine. Fixed cheeks thus widely expanded in front, narrow
along sides of glabella, with palpebral lobe situated almost exactly half way along
length of cranidium. Pleuroccipital furrow sharp. Posterior margin straight
adaxially, laterally bent back into a posteriorly projecting curve.

Free cheeks broad, only slightly inflated, with a broad marginal concavity which
rises laterally to a narrow, strongly enrolled rim; this is broader anteriorly and
narrower laterally and carries 3-4 parallel striations. Genal spine very short,
sharply pointed ; outer edge in direct continuation of the cephalic curve ; inner edge
concave so that posterior margin of the free cheek is deeply notched. Eye platform
large, flat, kidney-shaped; eye with numerous lenses, arranged in two intersecting
series concentric upon the anterior point and a point on the inner margin one third
eye-length from posterior. Ventral doublure broad, bearing two strong subcentral
and several irregular lateral subparallel longitudinal striae, all of which continue on
to genal spine. Pleuroccipital furrow clear, dying out laterally ; pleuroccipital
segment slightly raised.

Thorax. Consists of 9g segments. Axis broad, considerably broader than pleural
fields, gently arched. Axialfurrows sharp. Axial ring equal width (sag. and exsag.)
across axis; a deep transverse groove is situated immediately under the posterior
margin of the preceding ring ; from this the ring rises flatly to the posterior margin,
which is sharply angulate and carries a row of small pustules.

Pleural fields arched a little less than the axis; pleurae inflected dorsally and
posteriorly at the mid-length. Pleural segments of equal width throughout, with a
sharp posteriorly-directed point at their extremity. Pleural furrows moderately
deep and broad, sub-central ; otherwise the surface of the pleurae are flat, inclined
forward beneath the posterior of the preceding segment.

Pygidium. Anterior margin straight in centre, slightly bent back laterally from
mid-point of pleural fields ; antero-lateral corners rounded. Posterior outline short
shield-shape. Axis broad, rapidly tapering to a raised blunt extremity situated well
forward of posterior margin ; axis high, slightly carinate. 10-12 axial furrows on
internal lamina, crowded together at posterior: less than this visible externally.
Pleural fields gently arched along mid-line. Anterior border has a low but distinct
ridge, parallel with the anterior margin adaxially ; but distally from the point of
inflection of the border it diverges and diminishes, thus separating off a narrow
triangular area antero-laterally. 3-7 pleural furrows visible, becoming increasingly
oblique posteriorly ; do not reach border. 3-4 interpleural furrows faintly visible
on internal laminae. Rim furrow a gentle concavity, rising to a wide slightly
raised, carinate brim.

Remarks. Phillibole pollen: differs from P. aprathensis in having a more nearly
semicircular headshield ; in having a more smoothly attenuated (i.e. not medianly
constricted) glabellar outline; and in having a shorter pygidial axis. From
Ph. coddonensis it differs in its larger size, in the shape of the headshield and in the
larger relative size of the pygidium.

GEOL. 14, 6. 23

224 LOWER CARBONIFEROUS TRILOBITES

Phillibole coddonensis (Woodward)
Pl. 3, figs. I-5.

1895 Proetus sp. a & b Woodward: 648, pl. 28, figs. 7, 8.
1895 Phillipsia leei Woodward; Woodward: 646, pl. 28, figs. 1, 1a.
1902 Proetus coddonensis Woodward : 483, pl. 20, figs. 5, 7-11.

Diacnosis. Small Piullibole with broad axis; some glabellar segmentation ;
thorax with eight segments ; very short pygidium.

LectotyPE. BMNH.I. 4560 (figured Woodward 1902, pl. 20, fig. 8), here
selected as lectotype.

MATERIAL. The specimens upon which this species was founded were collected
from North Devon by Hamling & Coomaraswamy, and were subsequently presented
to the British Museum (Natural History). Fourteen others from the same collec-
tions are to be found in the North Devon Athenaeum, Barnstaple (NDA 812-816).
Additional material has been collected from north-east Devon by J. M. Thomas and
from north-west Devon by the present author. The total collection examined com-
prises 3 complete shields, 1r headshields, 1r cranidia, 8 parts of thorax and 15
pygidia.

LOCALITIES AND HORIZON.

I. Coombe (or Combe) Quarry and Wood, Barnstaple. There are a number of
old quarries in the neighbourhood of Coombe Wood, which is situated at the eastern
end of Codden Hill, 3 miles south-east of Barnstaple, North Devon (Nat. Grid
ref. SS 592295). Several of these quarries expose the white chert in which the
museum specimens are preserved, but intensive search has failed to yield any more
specimens. This group of white cherts occupies a median position in the Chert
Formation of the Codden Hill region.

2. Templeton Quarry, Tawstock (Nat. Grid ref. 55541296). Bed Y, Chert
Formation (Prentice 1960).

3. Warrenshill Quarry, Bampton. Bed 16 and 18, Bampton Chert Formation
(see p. 238).

4. Phillibole culmica regarded as nearly related, was described by R. & E. Richter
from the Posidonienschiefer of Aprath, Germany, at an horizon referred to zone IIT/.

DESCRIPTION. Cephalon. Outline semicircular, length (sag.) being almost
exactly twice width (tr.). Glabellar outline sometimes almost parallel-sided, but
mostly tapered anteriorly ; base is broad and parallel-sided, attenuation begins
abruptly about mid-length of glabella, then becomes parallel sided again anteriorly.
Anterior end is bluntly rounded ; anterior portion high, spherical, posterior somewhat
lower and flatter. Axial furrow distinct. Occipital furrow nearly straight, or
forwardly bowed in centre ; occipital segment narrow (sag.) and strongly inflated.
1p furrow very faint to moderately distinct ; begins well forward and is sharply
directed backward: in some specimens joins occipital furrow } to 4 along its length:
in a very few specimens crosses the occipital furrow and curves outwards across
occipital segment to reach axial furrow again near posterior border. 2p furrows
sometimes absent, sometimes distinct ; begin at the point of glabella attenuation,

LOWER CARBONIFEROUS TRILOBITES 225

and run backwards across glabella more or less parallel to Ip. 3p furrows rarely
visible, faint only at sides, directed straight across glabella or slightly forward.
Fixed cheek in front of glabella is deeply concave, rising sharply forward to a high
enrolled rim; from this concavity the fixed cheeks rise gently on each side of the
glabella, reaching maximum height in the eye region: then fall sharply to occipital
furrow. Suture-line begins (~) at point in line with the forward projection of the
axial furrows along the sides of the glabella, and swings sharply outwards to /,
which lies in the rim furrow in line (exsag.) with or just behind the anterior point of
the glabella. Then converges gently with sides of glabella, becoming nearly parallel
with these in neighbourhood of 2p furrow. Inflection in eye region very slight.
Near to the occipital furrow it swings sharply outwards to cut posterior margin at a
point just inside the angle at the base of the spine. Eye small or absent, situated
posteriorly of mid-line. Free cheek gently inflated in centre, crossed by strong
straight pleuroccipital furrow; anterior slope of furrow steeper than posterior.
Rim furrow broad and deep, rising to flat brim whose edge is sharply enrolled.
Doublure broad and flat, carrying 3 or more rows of faint longitudinal striae on
underside. Genal spine short, sharply pointed, outer margin deflected slightly
outwards.

Thorax. Eight segments only. Axis substantially broader than pleural fields,
strongly raised with deep axial furrow. Axial segments W-shaped, especially
anteriorly : strongly inflated with a deep transverse furrow at anterior and posterior.
Articulating half-ring concealed. Pleural segments broad (exsag.) and short (tr.),
sinuous, with rounded truncate extremities. Pleural furrows a deep median ex-
cavation, bounded on anterior by strongly diagonal pleural ridge.

Pygidium. Small, narrower than thorax (tr.) at anterior ; short subtriangular in
outline. Anterior margin bent forward at axis, but backwards in pleural region.
Anterolateral corners rounded. Axis broad anteriorly, broader than pleural region,
narrowing rapidly posteriorly to a rounded extremity a short distance anterior of the
margin. Faint, low, narrow post-axial ridge rarely developed. Axial rings 5-8,
separated by slightly flexuous transverse furrows. First anterior furrow deeper and
wider than all others; posterior furrows diminishing in intensity until posterior
extremity is almost smooth. Pleural fields gently raised, indistinct brim furrow and
narrow brim with enrolled edge. Pleural furrows 5 to 6, reaching nearly to margin,
becoming increasingly oblique and fainter posteriorly. First anterior pleural furrow
deeper than others. Interpleural furrows faintly developed laterally only on first
2 to 3 segments. Ventral doublure broad and flat to posterior, narrowing rapidly
anteriorly.

Meraspid stage. One specimen and its counterpart (BMNH. In. 22892), collected
by Wheelton Hind from Codden Hill, has 5 thoracic segments and clearly represents
the meraspid stage. The proportions of the headshield are similar to those of the
adult, but the genal spines show a slight outward deflection. There is a slight in-
dication of a pointed anterior margin to the pre-glabellar region. The suture-line
has a similar course to the adult, but is proportionately further away from the
axial furrow in the median region, and the posterior divergence begins somewhat in
front of the occipital furrow. The five thoracic segments are of adult form. The

GEOL. 14, 6. 23§

226 LOWER CARBONIFEROUS TRILOBITES

pygidium is small, and shows a posterior bent downwards. Segmentation of the
axis is strong ; four axial rings are clearly defined by deep furrows, which gradually
decrease in depth towards the rear. Four pleural furrows, and two interpleural,
are visible.

REMARKS: The similarity between P. coddonensis and P. culmica was acknowl-
edged by R. & E. Richter (1939 : 110), but they maintained the identity of the latter
on the grounds that P. coddonensts has a less-diminishing glabella, and a tail with a
well developed brim. The former feature is somewhat variable in P. coddonensis,
but in the material examined there are many specimens which can be matched in
this feature with Richter’s illustrations of P. culmica. Photographs of the holotype
of P. culmuica (kindly supplied by Dr. G. Hahn) show however that these specimens
have a longer and more elliptical headshield, a more strongly attenuated glabella,
and a narrower (tr.) posterior section of the fixed cheek than P. coddonensis. The
appearance of a brim to the tail in Woodward’s illustrations is seemingly due to the
preservation as a mould, which leave a space between the inner and outer moulds of
the edge of the doublure. In reality P. coddonensis has no more brim than
P. culmica.

Osmolska (1962: 75) has referred P. culmica to the subgenus Macrobole. The
present species resembles M. dvewerensis (the type species) in the lobation of the
glabella, but the strong segmentation of the pygidium in Macrvobole is a major point
of distinction. The pygidium of P. coddonensis, while unlike that of almost any other
contemporary trilobite in form, does have the accentuated anterior furrow and the
rather indistinct pleural furrows of Phillibole. The outline of the glabella and the
course of the suture line are similar to those of Phillibole aprathensis (the type
species), but coddonensis is distinguished by the presence of glabellar segmentation.
It should be borne in mind, however, that in some species of Phillibole (e.g. P. pollent
see p. 221) the glabella is smooth on the outer, and segmented on the inner lamina.
Since the type material of P. aprathensis is preserved in shale, the holotype presum-
ably preserves the outer lamina, while all the material of P. coddonensis consists of
moulds of the inner. The presence of only eight thoracic segments, the breadth of
the axis and the very small pygidium distinguish this species from all other species
of Philhbole.

DIMENSIONS (in mm.)

Cephalon :
L. (sag.) W. (tr.) Lg. (sag.) Weg. (tr.) LL. (genal spine)
BMNH.
I. 4560 3:8 5°8 3:0 2:0 I‘O
I. 4561 4°9 8-1 4°2 35 0-9
I. 4564 4:2 I1'5 2°8 — —=
I. 4565 4:0 9:0 35 3°1 =
I. 4571 4:0 58 — 2:0 I-o
KC;
t. 170 36 5:0 — — 0'9

t. 191 48 ~- 3°4 2:2 0:9

LOWER CARBONIFEROUS TRILOBITES 227

Thorax :
L. (sag.) W. (tr.) W. ax. (tr.)
BMNH. I. 4560 3:0 4°8 2-1 ant. 1-6 post.
Pygidium :
L. (sag.) W. (tr.) L. ax. (sag.) W. ax. (tr. ant.)
BMNH.
I. 4560 18 38 1‘6 1-2
I. 4564 23) 6-1 18 2°7
In. 18409 2-1 5:0 I°5 2:0
KCL.
t. 188 373 6-1 —- —
t. 208 3:6 4°4 — =

Genus LIOBOLE R. & E. Richter 1949

The genus Liobole was proposed as a subgenus of Phillibole by R. & E. Richter
(1949: 71) with Phillipsia glabra Holzapfel 1889 as type species. The name was
later raised to generic rank by R. & E. Richter (e.g. 1951 : 231), Struve (7m Moore
1959) and Osmolska (1962). The Lzobole group seems more closely linked with the
genus Liobolina (see Osmdlska 1962: 76, 92) than with Phuillibole s.s., so that
its generic rank seems to be substantiated.

Liobole glabra (Holzapfel)
Pl. 3, figs. 1-5.

1889 Phillipsia glabra Holzapfel : 73, pl. 8, fig. 23.

1895 Griffithides acanthiceps Woodward : 674, pl. 28, figs. 5, 5a.

1902 Phillipsia sp. Woodward: 482.

1902 Proetus sp. Woodward : 484, pl. 20, fig. I.

1949 Phillibole (Liobole) glabra glabra (Holzapfel) ; R. & E. Richter: 72, pl. 1, figs. 1-3.
1949 Phillibole (Liobole) glabra hiemalis R. & E. Richter: 74, pl. 1, figs. 4-11, (subsp. n.).
1960 Phillibole (Liobole) glabra glabra (Holzapfel) ; Prentice: 271, pl. 12, fig. I.

Diacnosis. As for R. & E. Richter 1949: 72.

MATERIAL. Three cranidia (BMNH. It. 381, It. 1436, It. 14374, 6, KCL t175-6)
one with thoracic segments and a fragment of free cheek attached, both somewhat
crushed. One specimen (NDA 822 and counterpart BMNH. In. 33775) with
thoracic segments and slightly displaced pygidium : three separate pygidia (BMNH.
I. 1438, I. 4562, In. 18411). Two separate free cheeks and counterparts (BMNH. I.
3020, I. 3216).

LOCALITIES AND HORIZON.

i. Park Gate Quarry, Tawstock. Nat. Grid ref. SS 555297. Probably from
Bed X, correlated (Prentice 1960 : 273) with zone II/III«, at a low horizon within the
Codden Hill facies of the Chert Formation.

228 LOWER CARBONIFEROUS TRILOBITES

2. Codden Hill; Hannaford Quarry ; Templeton Quarry, Tawstock. All these
localities lie within the Codden Hill Cherts, but the last-named locality is doubtful,
as all other fossils collected from this quarry belong to a higher horizon.

3. Warrenshill Quarry, east of Bampton, North Devon (Nat. Grid ref. SS 978222),
Bed 15, Bampton Chert beds (see p. 238), J. M. Thomas collection.

DESCRIPTION OF NORTH DEVON MATERIAL.

Cephalon. Outline probably semicircular, shorter (sag.) than wide (tr.) Glabellar
outline broad, attenuated forward, anterior end almost semicircular in plan.
Posterior margin of glabella convex forward, so that occipital segment is very
narrow (sag.) in centre; this margin has a sharp raised rim. Occipital glabellar
furrow distinct and deep in central third, branches laterally ; posterior branch fades
away rapidly, anterior branch (1p) runs forward and nearly reaches axial furrow.
These two branches enclose a lobe which is pear-shaped in outline, moderately
raised laterally. At the posterior margin of the occipital segment is a deep furrow
bounded posteriorly by a very narrow posterior rim; this furrow meets the axial
furrow and the posterior fixigenal furrow at a deep triangular apodemal pit. Second
glabellar furrow (2p) starts at or near the axial furrow, runs gently backwards and
deepens, then swings sharply backwards to join rp and occipital furrows at their
point of divergence, becoming fainter as it does so. Third furrow (3p) faint, nearly
parallel with 2p but confined to sides of glabella and shows no backward swing.
Axial furrow clear but not deep. Preglabellar field narrow, concave, rising steeply
to a sharply enrolled rim higher than the glabella. Anterior doublure broad, carries
four parallel longitudinal striae on inner side. Anterior of fixed cheek faintly concave
in continuity with concavity of preglabellar field, then rises until nearly flat along
sides of glabella: rises again slightly just in front of pleuroccipital furrow. Pleur-
occipital furrow broad and deep, widens laterally ; anterior slope steeper than pos-
terior ; course straight. Occipital segment of fixed cheek very narrow, inflated,
bounded posteriorly by very deep furrow. Facial suture begins («) close to mid line,
cuts very obliquely across border in an almost straight line to £, which lies only just
behind the line of the anterior of the glabella. From /, which is gently rounded,
suture curves inwards to within a short distance of the axial furrow. There is a
shallow palpebral lobe (y—e) situated well forward ; from ¢ suture diverges in a
gentle outwardly concave curve to cross occipital furrow and reach w wide of the
axial region. Posterior part of the fixed cheek is thus broadly triangular in shape.

Free cheek gently, flatly inflated. Margin broad, gently raised, with a very in-
distinct border furrow; genal angle rounded, produced into short, spatulate ex-
tension. Pleuroccipital furrow sharp near axis, dies out abruptly at border furrow.
Eye visible on inner test lamina only, no corresponding impression on inside of outer
laminae, nor presumably on dorsal exterior. Eye platform very slightly raised,
situated well forward near the border furrow. Inner side of eye practically straight,
outer side strongly convex, so that eye is nearly semi-circular; surface finely
reticulate, composed of two series of intersecting concentric ridges, centred at points
1 and # along inner border of eye ; each interspace so formed bears a central circular
depression. Doublure of the free cheeks broad, strongly enrolled, diminishing in

LOWER CARBONIFEROUS TRILOBITES 229

width posteriorly. Inside of doublure carries fine sub-parallel striations, which are
not present on the underside of the dorsal margin.

Thorax. Axis broad, equal in width or somewhat wider than the pleural regions ;
somewhat more raised than pleurae. Axial ring shows astrong transverse concavity ;
anterior is a low ridge higher along median line than laterally, where it is largely
covered by the posterior of the preceding ring. The concavity is steeper on its
anterior side ; behind, it rises gently to a high transverse ridge, which terminates at
each end in a boss at the axial furrow. Beneath this posterior ridge in the median
line the carapace is deeply infolded ; laterally this fold opens out so that the ring
shows a posterior sulcus bounded behind by a low triangular flange. The posterior
margin of this flange is continuous with the posterior margin of the pleural segments.
Pleural segments are flatly arched, nearly straight with blunt rounded extremities ;
each has a transverse concavity bounded by a high posterior ridge ; the concavity
shallows laterally and is crossed by an oblique ridge which commences anteriorly
near the axial furrow and terminates near the posterior side at the pleural ex-
tremities. This ridge is steeper on its posterior side, and slightly lower laterally
than near the axis.

The maximum number of segments found articulated is eight.

Pygidium. Shield-shaped in outline, somewhat wider than long. Anterior
margin bowed forward in axial region, gently bowed backwards in pleural ; antero-
lateral angles rounded. Axis moderately broad, slightly wider than pleural region
at pygidial anterior; axial furrows straight, so that axis is regularly tapering ;
axial furrows and identity of axis become less pronounced posteriorly. Axis
terminates abruptly at a rounded extremity before reaching posterior margin, con-
tinues as a very low postaxialridge. Anterior of axis bears a deep, forwardly convex,
transverse furrow, which separates off a high, narrow, anterior axial ring. Axial
rings faint, up to 12 sometimes faintly visible. Paired pygidial muscle apodemes
visible on internal mould at ends on first 6 transverse furrows. Pleural regions gently
inflated, nearly smooth. Anterior has a deep transverse furrow, bounded by a high,
narrow, anterior ridge ; this furrow lies parallel with the anterior border from axis
to midpleural (tr.) region, then swings sharply backwards, so that antero-lateral
corner is a narrow triangular flange. Posterior pleural furrows very faint (9+),
separated by equally faint interpleural furrows ; both sets are just visible laterally
as far as the border furrow. Border furrow faint, accentuated by crushing, separat-
ing a moderately broad flat brim. Ventral doublure broad, strongly enrolled,
slightly wider at posterior extremity ; bears faint traces of internal longitudinal
striation.

REMARKS. The features which distinguish the subspecies L. glabra glabra and
L. glabra Miemalis (R. & E. Richter 1949) are unfortunately most susceptible to
modification by crushing and distortion. It has been thought advisable therefore
to refer the North Devon specimens only to the species without attempting further
subdivision. They are distinguished (a) from L. coalescens R. & E. Richter by the
much clearer development of “ occipital-solution’’ and the more clearly defined
axial region, (0) from L. glabroides R. & E. Richter by the broader and less attenuat-

230 LOWER CARBONIFEROUS TRILOBITES

ing glabella, and by the broadly triangulate free cheeks, (c) from L. subaequalis
(Holzapfel) by the furrowed glabella and (d) from L. zarembiensis Osmolska by the
narrower free cheeks.

Thoracic segments have rarely been described in the Cyrtosymbolids, and never
previously for a species of Liobole. There is a tendency for the ventral anterior
doublure to adhere to the free cheek when this is separated from the cranidium ;
this gives the appearance of a short spatulate “ genal spine’”’, and resulted in Wood-
ward (1895, text-fig. 5) illustrating a specimen in an inverted position. The present
account shows that L. glabra has a rounded, slightly extended genal angle similar to
other members of the genus (e.g. L. subaequalis Holzapfel).

Genus SPATULINA Osmolska 1962

TYPE SPECIES. Phillipsia spatulata Woodward 1902.

The genus Spatulina with S. spatulata as type species, includes also Phillibole?
(Cystispina) nasifrons R. & E. Richter 1949 and Phillibole (? Cystispina) sp. nov.
Prentice 1960 (Osmolska 1962 : 74, 90, 180-181).

EMENDED DIAGNOSIS. Cyrtosymbolids with stout, hollow, genal spines. Cylin-
drical glabella with moderately distinct or faint dorsal furrows. Occipital glabellar
segment of equal length sagitally and exsagitally, occipital furrow complete and un-
branched. Fixigenae very narrow, with an abrupt forward slope anteriorly (based on
diagnosis by Osmolska 1962 : 180).

Osmolska suggests that the resemblances of this group of Cystispina, in which
S. spatulata was formerly placed by R. & E. Richter, are merely superficial, and that
the cranidial structure marks them out distinctively from that genus. Certainly
the three known species of Spatulina have very similar cephala, the only points of
distinction being the shape of the librigenal spines, and slight differences in the shape
of the glabella. The thorax is known only in S. longispina sp. nov., in which it
consists of g segments. The pygidium is known from all three species, and is,
characteristically, nearly smooth, with a high tapering axis terminating some
distance in front of the posterior extremity, which falls away rapidly from the tip
of the axis. The pleural fields have only a very narrow indistinct rim.

This small, closely knit group, has a very restricted range geographically and
stratigraphically. It is known only from the Chert Formation of North Devon, and
from the Erdbach & Belecke limestones in the Rhineland ; Pericyclus Zone (II).

Spatulina spatulata (Woodward)
Pl. 6, figs. 5, 6, Text-fig. 2.

1902 Phillipsia spatulata Woodward : 482, pl. 20, figs. 3, 4.

1939 Phillibole ? (Cystispina) spatulata (Woodward) R. & E. Richter: 107, text-figs. 17, 18.
1960 Phillibole ? (Cystispina) spatulata (Woodward) ; Prentice: 271, pl. 12, fig. 9.

1962 Cystispina spatulata (Woodward) ; Osmdlska: 180.

Diacnosis. Spatulina with short, bluntly rounded, hollow spines. Glabella
parallel-sided.

LOWER CARBONIFEROUS TRILOBITES 231

LectotypE. BMNH. I. 3215 selected by R. & E. Richter (1939) from among the
syntype material cited but not figured by Woodward (1902).

MATERIAL. In addition to the lectotype three specimens described by Woodward
are preserved in the British Museum (Natural History): I. 4572 (Woodward 1902,
pl. 20, fig. 4), I. 3219 and I. 3223. Other specimens from the Woodward collection
are in the North Devon Athenaeum: NDA 802, the original of pl. 20, fig. 3 and its
counterpart NDA 803, NDA 804-805. Additional material collected by the writer
is now in the British Museum (Natural History) nos. BNMH. It. 14392, b.

Locality AND Horizon. The species is only known from the Chert Formation of
Codden Hill, Combe Wood and Tawstock, Barnstaple, North Devon. Specimens
BMNH. It. 1439a, b and I. 3223 came from Park Gate Quarry, Tawstock, where Bed X,
of Zone II/IIIla age is exposed (see p. 238); on the latter specimen also occurs
Macrobole cf. brevispina (see p. 238) indicating a Zone II horizon.

DescripTIon. A full description was given by R. & E. Richter (1939 : 108) ; the
following are additional features seen on the new material:

Occipital segment, which is nearly parallel sided transversely, sometimes carries a
small median tubercle; occipital furrow a little fainter medianly than at sides.
Lateral glabella furrows distinct on internal mould ; pre-occipital lobe long (exs.),
triangular. Preoccipital (Ip) furrow begins at axial furrow, curves gently back,
then more sharply to terminate abruptly close in front of occipital furrow. 2p
lobe narrower than Ip, 2p furrow parallel to rp ; 3p furrow faint, parallel to 2p but
shorter. 2p furrow begins at axial furrow at point almost halfway along (exs.) the
glabella length.

Pregabellar border has 2 or more fine longitudinal striae on inner side. Fixed
cheeks have a gentle anterior concavity, rising along the sides of the glabella to the
highest point near the 2p furrow ; posterior falls again to pleuroccipital furrow.

Free cheek gently inflated in centre ; an obscure concavity in front deepens to
become a moderately deep rim furrow posteriorly. Pleuroccipital furrow distinct,
meets rim-furrow at 90°, and the two are produced for a short distance along the
dorsal face of the genal spine. Rim of free cheek slightly raised, margin sharply
geniculate, so that exterior of free cheek is flattened flange, which continues on to
outside of genal spine. Doublure broad, strongly enrolled, with 7-8 subparallel
longitudinal striations on the inside of the lower fold.

Genal spine short, approximately half length of cephalon (sag.) ; outer lateral
side of spine flattened, in continuity with flattening of the side of the cephalon.
Spine at genal angle nearly circular in cross-section ; medianly is slightly inflated and
terminates sharply in a slightly mucronate chisel-shaped end, whose edge lies nearly
vertically to the plane of the dorsal shield. Dorsal surface of spine gently rounded,
impressed for a short distance by the continuation of the coalesced marginal and
pleuroccipital furrow ; falls slightly at point. Interior of spine bears fine longi-
tudinal striae in continuity with those of the interior of the doublure.

Thoracic segments unknown.

Pygidium shorter than wide, shield-shaped. Anterior margin has a narrow raised
rim; antero-lateral angles rounded. Axis strongly raised, rapidly tapering,

LOWER CARBONIFEROUS TRILOBITES

nN
Ww
N

One Centimetre

Fic. 2. Spatulina spatulata (Woodward). Reconstruction.

terminating in a high, blunt, rounded end. Axial furrow distinct. 8 or more
faintly defined axial rings, bearing faint traces of interior of paired pygidial muscle
impressions. Pleural regions gently inflated, fall away rapidly behind; narrow,
gently raised rim with an indistinct border furrow. Pleural furrows very faint.

REMARKS. As indicated by Osmdlska (1962: 180) the differences between
S. spatulata and Cystispina cystispina are sufficient to warrant their generic separa-
tion. These differences are in the shortness of the preglabellar field, the non-
tapering glabella, and the narrow fixigenae in S. spatulata. On the other hand, the
distinctness of the glabellar furrows is not a point of difference, as in both genera
these are clear on the internal mould and indistinct on the interior (and presumably
on the dorsal exterior) of the carapace. The pygidia of the two genera are also very
similar. From S. longispina sp. nov. (see below) S. spatulata differs in the shape of
its spines and its parallel-sided glabella ; from S. nasifrons (R. & E. Richter 1949)
it differs in that the latter species has a more prominent, slightly constricted glabella
and ( fide Osmolska 1962 : 180, text-fig. 5f) a very broad spine.

Spatulina longispina sp. nov.
Pl. 6, figs. 1-4; Text-fig. 3.

1895 Griffithides longispinus Portlock ; Woodward: 647, pl. 28, fig. 6.
1960 Phillibole (? Cystispina) sp. nov. Prentice: 271, pl. 12, fig. 2.

Diacnosis. Moderately large Spatwlina with long, hollow, pointed librigenal
spines. Glabella slightly tapering.

LOWER CARBONIFEROUS TRILOBITES 233

HoLotyPe. One nearly complete specimen, a somewhat crushed internal mould
(BMNH. It. 1440a@) from Park Gate Quarry, Tawstock, North Devon.

MATERIAL. Holotype and counterpart (BMNH. It. 1440d): two separated free
cheeks, one pygidium and other fragments on a single slab (BMNH. It. 1441): a
nearly complete specimen with crushed cranidium and displaced free cheeks
(BMNH. I. 45734) and partial counterpart (BMNH. I. 45730): parts of cephala
(NDA. 809, NDA. 825).

LocaLiry AND Horizon. The specimens (KCL 170-178, BMNH. It. 1440-41)
collected by the author are from Bed X in Park Gate Quarry, Tawstock (see p. 238),
of IJ-IIIa zone age. Specimen BMNH. I. 4573 came from the Chert Formation,
Coddon Hill, and those in the North Devon Athenaeum are from Combe Wood
and Hannaford from similar horizons.

DESCRIPTION. Cephalon. Outline, when restored, nearly semicircular ; length
(sag.) of holotype 11-1 cms., width at occipital region 15:2cms. Glabella tapers
slightly forward, slightly constricted medianly, with oval rounded anterior, moder-
ately high. Occipital furrow clear and nearly straight, continuous across cephalon,
tripartite, occipital segment same width (sag.) throughout. No median occipital
spine. Axial furrow distinct, glabella raised above fixed cheeks. Preoccipital lobe
only slightly longer than occipital segment. Ip glabellar furrows very vague,
directed backwards to the mid-point of the occipital furrow. Pre-glabellar field
very narrow, upwardly inclined then bent over to form a very narrow brim. This
brim continues along the anterior margin of the fixed cheeks, where a deep concavity
is continuous with the concavity of the preglabellar field. Facial sutures begin (a)
at a point within the projection of the axial furrow at the glabellar sides then swing
outwards very obliquely across the brim to /, thence curve inwards to y which lies
close to the axial furrow about half way along (exsag.) the glabella. Palpebral
lobes not evident ; suture line follows axial furrow for short distance, then swings
outwards gently to w, which lies near to the internal angle of the genal spine.

Free cheeks nearly flat except for narrow brim ; occipital furrow clear becoming
fainter laterally. Genal angle with long spine (7-5 mm. in holotype) circular in cross
section for most of its length, then narrowing rapidly to a slightly mucronate point.
Anteriorly the dorsal face of the spine bears a deep triangulate concavity, the con-
tinuation of the occipital furrow. The spine is hollow with faint internal longitu-
dinal striations. Doublure broad, hollow, in continuity with genal spine ; carries
4 to 5 longitudinal striations on its underside.

Thorax. The remains of only 4 segments are preserved in the holotype, but the
specimen figured by Woodward (BMNH. I. 4573a) is complete and shows 9 segments.
The axis is wide (4-5 mm. in holotype) slightly wider than original to width of the
pleurae, moderately raised. Axial ring has deep transverse furrow slightly anterior
of the median (sag.) line; anteriorly a narrow, posteriorly a broad, raised rim.
Pleural regions less raised than axis, with broad shallow pleural furrow, and bluntly
rounded pleural extremities.

Pygidium. Elongate, hemi-elliptical in outline, with strongly raised, tapering,
axis separated from somewhat flatter pleural fields by a sharp axial furrow. Length

234 LOWER CARBONIFEROUS TRILOBITES

Fic. 3. Spatulina longispina sp.n. Reconstruction.

One Centimetre

(holotype) 9:9 mm. (sag.), width at anterior border 11-9 mm. (tr.), width of axis
4.5mm. Axis rapidly tapers to a sharply rounded, strongly elevated point, which
lies a short distance (2 mm. in holotype) in front of the posterior of the pygidium.
Axial rings 9-11, clearly defined at anterior becoming effaced behind. Pleural
fields nearly flat, with traces of four oblique pleural furrows and three interpleural
furrows anteriorly. Margin gently raised, no border furrow. Doublure sharply
infolded, 1 mm. wide, with faint longitudinal striae internally.

REMARKS: The species may be distinguished from S. spatulina by its long,
cylindrical, pointed spines, by its slightly tapering glabella with very faint lateral
furrows, and by the more distinct segmentation of the pygidium. These same
characters distinguish it from S. nasifrons (R. & E. Richter).

Specimen BMNH. I. 4573 was assigned by Woodward to Griffithides longispinus
Portlock. However, the large eyes and the absence of a pre-glabellar field in
Portlock’s species preclude the inclusion of the present specimens. The trivial name
longispinus is, nevertheless, apt, and is employed here.

LOWER CARBONIFEROUS TRILOBITES 235
Genus TYPHLOPROETUS R. Richter 1913

Type spEcIES. Jyphloproetus microdiscus R. Richter, 193T.

This genus includes 14 species ranging in age from the Middle Famennian to the
Lower Visean. They have in common a tapering glabella, anteriorly divergent
often ankylosed sutures, vestigial eyes and a characteristically laterally narrowed
occipital glabellar segment.

Typhloproetus cephalispina sp. nov.
PI32; figs 24,6.

Diacnosis. Typhloproetids with nearly semicircular cephalon, bearing a short
spine at anterior extremity. Glabella with one pair of lateral furrows; eyes
vestigial. Suture line cuts anterior border near midline, anterior part widely
divergent.

HoLtotyrPe. Cephalon, BMNH. It. 1442 (J. M. Thomas coll.), Bed 18, Chert
Formation, Warrenshill Quarry, Bampton, North Devon (see p. 238).

MATERIAL. In addition to the holotype, one other cephalon with counterpart
showing three thoracic segments attached (BMNH. It. 1443a, 6) from same horizon
and locality.

Horizon. Upper Visean, Zone III/.

DESCRIPTION. Cephalon semicircular in outline ; holotype length (sag.) 2-0 mm.,
width 4.0mm. Glabella broad and flat posteriorly, moderately high and rounded
anteriorly ; width (holotype) at occipital segment 1-6 mm. ; length (sag.) I-g mm.
Axial furrow clear anteriorly, faint posteriorly. Glabellar outline gently tapering,
with strongly rounded anterior; constricted medianly. Occipital furrow deep,
the anterior slope being especially steep, curved forward in centre. Occipital
segment narrows laterally ; strongly inflated with a steep posterior rim. Ip lateral
furrows distinct, begin at axial furrow far forward, and are strongly inclined back-
wards towards, though not reaching, the mid-point of the occipital furrow. Pre-
occipital lobes are thus long and triangular. In front of rp furrows glabella is
strongly inflated, pyriform. Narrow preglabellar field, strongly concave, rising to a
high anterior rim, in centre of which is a short, bluntly rounded cephalic spine,
directed upwards and forwards. Fixed cheeks rise from their anterior concavity to
faint palpebral lobes situated parallel with mid-line (sag.) of glabella. Posterior of
fixed cheek deeply incised by occipital furrow.

Suture line ill-defined, probably ankylosed. Cuts anterior margin (a) very near
to centre of cephalic anterior, swings outwards very obliquely to £ which lies in the
border furrow; then curves in broad curve to near axial furrow near Ip lateral
furrow ; subparallel to axial furrow for a short distance posteriorly, then diverges
in pleuroccipital region to cut posterior margin of headshield at a point (w) midway
between axial furrow and lateral margin of headshield.

Free cheek gently inflated centrally, falling to a broad marginal concavity
anteriorly and laterally and steeply to a straight pleuroccipital furrow behind. Eye

236 LOWER CARBONIFEROUS TRILOBITES

platform, small, raised, elongate, situated a little in front of the mid-line (sag.).
Rim of cephalon sharply upturned, folded over to a wide doublure which carries 3
subparallel longitudinal striations on the underside. Small pointed genal spine.

Thorax with broad axis, more than a third of the width (tr.) of the carapace. Axial
ring high, strongly curved sagitally with a deep anterior transverse furrow separating
the articulating half-ring. Pleural regions gently inflated, with their lateral ex-
tremities bent sharply ventrally. Pleural furrows deep and wide, extending
subcentrally the whole length of the pleurae.

Pygidium unknown.

REMARKS. The species is included in Tyfhloproetus on the basis of the form and
nature of the facial sutures, the vestigial eyes, and the shape of the occipital glabella
segment. It differs from the genotype, 7. microdiscus in its more prominent glabella,
its more widely divergent facial sutures, and the shorter cephalic outline. From
T. dietzi R. & E. Richter it differs in having a less tapering glabella, and a narrower
anterior rim. Neither of these species has a procranidial spine. T. koslowsku
Osmolska, and ? 7. angustigenalis Osmolska (which shows some indication of an
anterior point) have more elongate glabellae than the present species.

Genus DIACORYPHE R. & E. Richter 1951

The genus Diacoryphe with type-species D. pfeifferi includes also D. gloriola
Richter: both are from the Zone I horizon of Germany (R. & E. Richter 1951 : 252).
Osmolska (1962 : 177) included a new species stvenuispina in this genus on the basis
of cranidial similarities, although the thick librigenal spine and strong anterior
border would exlude it from the generic diagnosis of R. & E. Richter. The British
species vandergrachttur has many similarities to strenuispina, but differs even more
from the type species of Diacoryphe. It is therefore, only doubtfully assigned to
this genus.

Diacoryphe? vandergrachtii (Woodward)
Pl. 7, figs. 1-5 ; Text-fig. 4.
1894 Phillipsia van-der-Grachtit Woodward: 485, pl. 14, figs. 1-6.

DIAGNOsIS. Small trilobites possessing a pyriform glabella with effaced axial
furrow at posterior; strong anterior border-ridge; long inflated genal spines ;
thorax with 8 segments ; and strongly segmented pygidium.

MATERIAL. The six specimens described and figured by Woodward (1894) are
preserved in the British Museum (Natural History). The specimen figured by
Woodward (1894, pl. 14, fig. 3) is here selected as lectotype. The specimens are
preserved in shale, either as crushed carapaces showing the external view of the
dorsal side, or as external moulds of the same. The specimens were collected from
the same locality as those of Piuillibole polleni, whose stratigraphical horizon (see
p. 238) is at or just below the B,—B, zonal boundary.

LOWER CARBONIFEROUS TRILOBITES 237

DeEscripTION. Outline of carapace elliptical, nearly twice as long as wide.
Cephalic outline parabolic, widest at posterior margin, wider than long. Glabella
sharply demarcated from fixed cheek anteriorly, but posteriorly the axial furrow is
poorly defined. Anterior of glabella rounded, widest (tr.) part 4 of length (sag.)
from anterior; behind this gently narrowing to a minimum width (tr.) at the
occipital furrow. Occipital lobe of glabella widens again to attain width of the
anterior. Occipital furrow faint but continuous across glabella, bent forward in
centre ; no trace of other furrows. Preglabellar field narrow, rising rapidly to a
broad enrolled rim, which carries three longitudinal subparallel ridges. Facial
suture cuts anterior margin at a point («) just inside the forward projection of the
widest part of the glabella ; from thence it swings sharply outwards to # which lies
almost within the marginal furrow. Swings inwards more sharply to y, which is
broadly rounded and lies very close to the axial furrow at the point of maximum
glabellar width. From y the suture is gently convex outwards swinging inward and
near to the axial furrow at e, which lies just in front of the occipital furrow. On
crossing the occipital furrow the suture line runs sharply outwards in a straight line,
to cut the posterior margin of the headshield (w) just inside the genal spine. Fixed
cheek is thus very wide in front and very narrow behind ; in profile almost flat except
for the raised rim. Free cheeks smooth, gently raised in centre, falling to a broad
deep furrow inside araised rim. This rim is broader anteriorly, where it is continuous
with the rim of the fixed cheek, becomes sharper and narrower posteriorly, and at
genal angle passes into the librigenal spine. This spine is as long as the headshield,
reaching to the posterior thoracic segment ; broad and inflated, terminating in a
blunt rounded extremity. The interior angle with the posterior margin of the head-
shield is rounded. Pleuroccipital furrow of free cheek faint and broad.

Thorax of eight segments; axial region strongly raised, separated from pleural
regions by a sharp axial furrow. Axis one third width (tr.) of carapace at anterior,
narrowing slightly posteriorly. Axial rings nearly parallel-sided (tr.) slightly
embayed, posteriorly and bent forward anteriorly ; articulating half-ring not visible.
Pleural segments parallel-sided, slightly flexed backwards from a point close to axial
furrow ; extremities bluntly rounded. Pleural furrows faint ; in median (ensag.)
position near axial furrow, then directed obliquely backwards across segment,
becoming effaced halfway along (tr.).

Pygidium sub-triangular, 13 times broader (tr.) than long (sag.): anterior margin
strongly and evenly curved forward. Axis sharply raised, separated by a sharp
axial furrow from the more moderately upcurved pleural regions. Axis tapers
rapidly, terminating in a triangular point a short distance in front of the posterior of
the pygidium. 8 or more axial rings, separated by strong, straight furrows. Pleural
fields crossed by 6 furrows, becoming progressively weaker and more oblique towards
the posterior. All pleural furrows die out at rim, which is moderately broad and
slightly flattened. Postaxial region bent ventrally, no postaxial ridge.

RemMArKS. The genus Diacoryphe is almost unique among Carboniferous trilobites
in showing posterior effacement of the glabella, and it is for this reason that the
species vandergrachtii is consigned to this genus. It differs from the type species,
D. pfeifferi, in the less widely expanded anterior part of the fixed cheek, on the long

238 LOWER CARBONIFEROUS TRILOBITES

palpebral lobes, and in the strong anterior border. It resembles D. strenuispina
Osmolska in having a strong, inflated genal spine. The pygidium, with its strong
segmentation, is more like that of a Phillipsiid, than that of D. pfeifferi. It seems
likely that a new genus is needed for the reception of the British species.

III STRATIGRAPHICAL CONCLUSIONS

The trilobites described here occur mostly in a region of great structural and
stratigraphical complexity, which has as yet been very incompletely unravelled.
The diagram (Table 1) attempts to record the succession in various areas of North
Devon, in so far as it has been resolved. With few exceptions the trilobites occur in
a group of beds of distinctive pale grey or white colour, which occupies a median
position in the Chert Formation. The Chert Formation is underlain by the Basement
Beds, and these by the Pilton Beds. Goldring (1955), on goniatite and trilobite
evidence has classified the top of the Pilton Beds with the Protocanites Zone (1)
The occurrence of Protocanites cf. lyont, probably in the Basement Formation
(Prentice & Thomas 1965) indicates that the Protocanites Zone extends above the
Pilton Beds. The occurrence of Pericyclus aff. homoceratoides at Swimbridge, also in
beds which are probably Basement Formation, indicates that this formation also
includes the base of zone II (Butcher & Hodson 1960). The top of the Chert
Formation is in many places defined by the “G. spivalis”’ band, of P,—P, age
(Prentice 1960), indicating that this Formation extends throughout most of zones IT
and III of the German sequence.

Within this succession it seems possible to recognize five successive trilobite
faunas. The lowest occurs near the base of the pale chert group, probably close
above the Basement Formation, and, so far, has been found in only one place.
It yields Macrobole aff. laticampa, which occurs in Zone I in Poland. The second is
more widespread: it is the fauna described by the author (Prentice 1960) from
Bed X in Park Gate Quarry, Tawstock. This bed is characterized by the occur-
rence of the genus Spatulina, whose only representative outside Devon occurs in
Zone II of Germany (Richter, R. & E. 1949). Associated with this is Liobole glabra
and Macrobole cf. brevispina. Liobole glabra also occurs at Warrenshill Quarry,
Bampton, in white cherts referred by Thomas to the Kersdown Beds (Webby &
Thomas 1965). These cherts lie immediately below similar beds with Phillibole
coddonensis, which forms the chief element of the third trilobite fauna. This has
been found in place also in Templeton Quarry, Tawstock (Bed Y, Prentice 1960).
The occurrence of both the Spfatulina fauna and the P. coddonensis fauna at various
localities (Coombe Wood, Hannaford, Codden Hill) indicated by museum specimens,
suggests that these two faunas are widespread and stratigraphically are closely
superposed. Phillibole aprathensis is also recorded from these localities, but there is
some evidence to suggest that it may occur in and characterise a higher horizon.
For example, it occurs in a loose block above the beds exposed at Warrenshill Quarry,
Bampton ; and at a high horizon in cherts in the Chudleigh region. The highest
fauna is that represented by Phillipsia leet, a species constantly found in the “ spirale
band ”’ which marks a very widespread depositional change in south-west England.

Appendix : 48).
Schmidt.

amas, J. M. (1965 : 34)-
i Meek & Worthen.

Warsbolz) cf. aequalis).

ssimus (de Koninck).

onensis Woodward

tus cephalisprina sp. nov.

Thomas, J. M. (1965) and the

EXPLANATION OF TABLE I

Pilton Beds.
Tutshill farmyard.

Well near Mt. Sandford,
Barnstaple.

Bydown Quarry, Swimbridge.
Park Gate Quarry, Tawstock.

lata Woodward (as Ph. ? (Cy.) spatulata).
‘as Ph. ? (Cy.) sp. nov.)
b ci. brevispina Osmélska (as

Templeton Quarry, Tawstock.

Railway Cutting, Swimbridge

Kents Hill Quarry, Dulverton.

Bed 15, Warrenshill Quarry,
Bampton.

Beds 16 & 18, Warrenshill
Quarry, Bampton.

Bed 18, Warrenshill Quarry,
Bampton.

to, Webby, B. & Thomas, J. M. 1965.
Bollandites sp.
11-14. Butcher, N. E.:& Hodson, F. (1960 : 77).
11. Goniatites sphaericostriatus Bisat
12. Girtyoceras burhennei (Schmidt)
13. Goniatites falcatus Roemer

14. Gontatites of the concentricus/striatus group

15-19. Prentice, J. E. & Thomas, J. M. (1960 : 6, 7).
15. Goniatites hudsoni—antiquatus group

16. Goniatites crenistria Phillips
17. Goniatites falcatus Roemer

18. Hibernicoceras carraunense Moore & Hodson

Sudeticeras crenistriatus Bisat
19. Neoglyphioceras spirale (Phillips)
Sudeticeras splendens (Bisat)
Prentice, J. E. (1960 : 275).
Neoglyphioceras spivale (Phillips)
cf. Mesoglyphioceras granosum

20 & 21.

22. This paper (p. 212).
Phillipsia leet Woodward.

23. Prentice, J. E, & Thomas, J. M. (1960: 7).
Neoglyphioceras spirale (Phillips).
cf, Mesoglyphioceras granosum (Portlock)
Dimorphoceras kathleenit Moore
Sudeticeras splendens (Bisat)

24. This paper (p.212).
Neoglyphioceras spirale Phillips
Phillipsia leei Woodward

Numbers in circles represent goniatite faunas, those in squares contain trilobites. The following list indicates only the latest reference to the fossils.

Kersdown Quarry, Bampton.

Fremington.
Fremington.
Hele Quarries, Dulverton.
Hele Quarries, Dulverton

Westleigh Quarries.
Westleigh Quarries.
Westleigh Quarries.
Westleigh Quarries.

Westleigh Quarries.

Fremington Pill, and Tawstock,
Fremington Pill, and Tawstock.

Hole Lake Farm Quarry,
Bampton.

Above Bampton Limestones.

Whipcott Quarry, Westleigh.
Whipcott Quarry, Westleigh.

z)
FREMINGTON BARNSTAPLE SWIMBRIDGE| SOUTH MOLTON | BRUSHFORD BAMPTON WESTLEIGH |< |
o 5S
£ = y| (Prentice, JE 19604Prentice,JE. 1960, | (Unpublished) | (Unpublished) (Swarbrick,£E. ‘6 2\(Swarbrick, EE, '62, |(Thomas.JM’65) E'a|
faq & unpublished ) | Thomas,JM '65) =
=f 5 \°2
BON!) Limekiln Black Shale Black Shale | Black Shale Black Shale Giffords €
|) &
é Beds Formation Formation Formation Formation Beds |
|
2a |020 e2i 023 v24 |
Pig [el2P E ‘ Baileys Beds 19 a|
id Hearson Bestridge ‘© Hele Manor (-U. Bampton |, Upper |
aa @ o fon |
Re ol i Hill Limestone “ue Chert B Limestone JM Nes |
3 Mudstone S g-Bampton 4,7 Limestone —}
Bp Chert Beds & Whitehill 32 ¢ Limestone ES) |
=| z Smalldon = a
=x a
a. || (Fremington = Limes vone E E F 216 |
Je facies) § Combe WnSEND @ Kents Hill | a Xersdown iat nes
| | a ! = |
F | G Wood 5 Chert 5 (-M Bampton
L 4 Chert 2 10% Chert, UMT.
tt le E Lo
| o E ig £ wer
| v5 oO High re 2 -Sampton
| x pees ve Westleigh
© Down = ———_| §& Chert, EES)
a, | Sl £Hulverton | Limestone |
| 5 Bydown Chert Hill Chert 2 \a
| 03 “Chert |
= 06 = 5 Hayne Beech I
oraey
8 ke Basement Basement Basement © 2 peds (MT)
| Pilton ** Formation | Formation Formation | % £ “oaddiscambel |
if mu Beds (MT) 1
1
e) Beds Bitten Pilton Pilton Pilton
Beds Beds Beds Beds

20 - Goniatite horizon

v4 - Trilobite horizon

P - Posidonia

LOWER CARBONIFEROUS TRILOBITES 239

That the species may have entered the area prior to this is indicated by its occurrence
in white chert at Ideford, Devon, and in the upper part of the Bailey’s Beds
(Thomas 1962) at Bampton.

The correlation of these faunas with their contemporaries in the rest of the
British Isles and in northern Europe presents some problems. The M. aff. laticampa
locality is within 2 kilometres of the railway-cutting from which Pericyclus aff.
homoceratoides is recorded by Butcher & Hodson (1960 : 76) and the latter horizon
may not be very far below the base of the Chert Formation. Nevertheless this
must mean that the Chert Formation begins in zone II, and a solid, but decayed,
pericyclid found with M. aff. laticampa supports this correlation. The Spatulina
fauna has been equated (Prentice 1960) with the top of zone II, a correlation con-
firmed by the occurrence of M. cf. brevisbina. There is no doubt that this is
succeeded immediately by the Phillibole coddonensis fauna. P. coddonensis is
closely related to Phillibole culmica R. & E. Richter, and with Ph. aprathensis is
known only from the Zone IIIf of Germany. It seems almost certain that in the
British Isles these two species extend down into the equivalents of Zone IIIa.
At Warrenshill Quarry, Bampton P. coddonensis is found, less than a foot above a
bed with Liobole glabra, with no sign of a stratigraphical break between. In the
Brushford area, the pale cherts (Kents Hill Chert) are succeeded by beds containing
P,, (i.e. Zone IIe) goniatites (Butcher & Hodson 1960). In the north of England
a form related to P. aprathensis characterized the B, zone (Calver & Ramsbottom
1962), which certainly lies below the P,, (IIIa) zone, and which includes in its fauna
Entogonites grimmeri Kittle, whose type material comes from Zone II in Sauerland
(Schmidt 1942: 49). The occurrence of Bollandites cf. castletonense, in the Kersdown
Beds (Thomas 1962) a typical B, zone form in the north of England, further links
these Piullibole faunas with zones B, and IIIa.

The age of the Phillipsia leer fauna is less equivocal, as it is associated with the
“ spiralis band ”’ lying at the P,—P, junction, or in the sub-zone of III ,, (see above
and Prentice 1960). The finding of a specimen of this species at a precisely similar
horizon in Sauerland suggests that the fauna may be of zonal significance.

IV REFERENCES

ButcHeEr, N. E. & Hopson, F. 1960. A review of the Carboniferous goniatite zones in Devon
and Cornwall. Palaeontology, London, 3 : 75-81, pls. 17-19.

Carver, M. A. & Ramsgpottom, W. H. C. 1962. Jn Earp, J. et al. The Geology of the
country around Clitheroe and Nelson. Mem. Geol. Surv. U.K., London. .

CuHLuPAG, I. 1958. Nova spodnokarbozaka fauna od Hranic na Moravé. Sb. usted. Ust.
geol., Praha, 24 : 279-312, pls. 1-4.

1961. New Lower Carboniferous trilobites from the Moravian karst. Vest. ustred. Ust.
geol., Praha, 36 : 229-234, pls. I, 2.

Gotprine, R. 1955. The Upper Devonian and Lower Carboniferous trilobites of the Pilton
Beds in N. Devon. Senckenbergiana, Frankfurt a.M., 36 : 27-48, 2 pls.

Haun, C. 1965. Revision der Gattung Avdhegonus Burmeister 1843. Senckenbergiana,
Frankfurt a.M., 46 : 229-262.

Havugpotp, W. 1932. Ueber das Unterkarbon auf Blatt Goddelscheim am Ostrande der
Rheinischen Schiefergebirge. Jb. preuss. geol. Landesanst., Berlin, 53 : 208-246, pl. 18.

240 LOWER CARBONIFEROUS TRILOBITES

Hinpe, G. J. & Fox, H. 1895. Ona well-marked horizon of radiolarian rocks in the Lower
Culm Measures of Devon, Cornwall and west Somerset. Q. J. Geol. Soc. Lond., 51 : 609-668,
pls. 23-28.

HorzapreL, G. H. 1889. Die Cephalopoden-fuhrenden Kalke des Unteren Karbon yon
Erdbach-Breitscheid bei Herborn. Paldont. Abh., Berlin, 5 : 1-74, 8 pls.

House, M. R. & ButcHEer, N. E. 1962. Excavations in the Devonian and Carboniferous
rocks of the Chudleigh area, south Devon. Pvoc. Ussher Soc., Camborne, 1 : 28-29.

Hupson, R. G. S. & Corron, G. 1945. The Lower Carboniferous in a boring at Alport,
Derbyshire. Pvoc. Yorks. Geol. Soc., Leeds, 25 : 254-311.

Hupk, P. 1953. Classe des Trilobites. In Tvaité de Paléontologie, 3, Paris.

Kayser, E. 1882. Beitrage zur Kenntnis von Oberdevon und Culm am Nordrande des
theinischen Schiefergebirges. Jb. preuss. geol. Landesanst., Berlin, 2 : 51-91, pls. 1-3.
Kogpoip, A. 1932. Die Gliederung des Oberharzer Kulms nach Goniatiten. Jb. preuss.

geol. Landesanst., Berlin, 53 : 450-515, pls. 22-23.

MEIScHNER, K. D. 1962. Rhenaer Kalk und Posidonienkalk im Kulm des nordéstlichen
Rheinischen Schiefergebirge und der Kohlenkalk von Schreufa (Eder). Abh. hess.
Landesamt. Bodenforsch., Wiesbaden, 39 : 1-47, pls. 1-7.

Moore, R.C. 1959. Tveatise on Invertebrate Palaeontology. ParvtO, Arvthropoda,1. (Kansas.

OsmoétsKA, H. 1962. Famennian and Lower Carboniferous Cyrtosymbolinae (Trilobites)
from the Holy Cross Mountains, Poland. Acta palaeont. polon., Warszawa, 7: I-53,
piss ISK

PIcKEL, W. 1937. Stratigraphie und Sediment-analyse des Kulms an der Edertalsperre.
Z. d. geol. Ges., Berlin, 89 : 233-280.

Porttock, J. E. 1843. Report on the geology of the county of Londonderry, and of parts of
Tyrone and Fermanagh. xxxi + 784 pp., 38 pls. Dublin & London.

PRENTICE, J. E. 1960. The Dinantian, Namurian and Lower Westphalian rocks of the
region south west of Barnstaple, North Devon. Q. J. Geol. Soc. Lond., 115 : 261-280,
pls. 12-13.

PRENTICE, J. E. & THomas, J. M. 1965. Prolecanitina from the Carboniferous rocks of
North Devon. Proc. Yorks. Geol. Soc., Leeds, 35 : 33-46, pls. 1, 2.

PrRipyit, A. 1953. On the Carboniferous Trilobites of Moravia—Silesia. Bull. int. Acad.
tchéque Sci., Prague, 51 : 209-230, pls. I, 2.

Reep, F. R.C. 1942a. Notes on the Upper Devonian trilobites in the Whidborne Collection

in the Sedgwick Museum. Geol. Mag., London, 81 : 121-126.

1942b. Some new Carboniferous trilobites. Ann. Mag. nat. Hist., London (11) 9: 649-

672, pls. 8-11.

RicHter, R. 1931. Beitrage zur Kenntnis devonischer Trilobiten II. Oberderonische
Proetiden. Abh. senckenb. naturforsch. Ges., Frankfurt a.M., 31 : 341-423, pls. 22, 23.
RicuHter, R. & E. 1926. Die Trilobiten des Oberdevons. <Abh. preuss. geol. Landesanst.,

Berlin, 99 : 1-314, pls. 1-12.

1937. Kulm-Trilobiten von Aprath und Herborn. Senckenbergiana, Frankfurt a.M.,

19 : 108-115, 6 figs.

1939. Proetidae von oberdevonischer Tracht im deutschen, englischen und mittelmeer-
ischen Unter-Karbon. Senckenbergiana, Frankfurt a.M., 21 : 82-112, 31 figs.

1949. Die Trilobiten der Erdbach-Zone (Kulm) im Rheinischen Schiefergebirge und
im Harz. 1. Die Gattung Phillibole. Senckenbergiana, Frankfurt a.M., 30: 63-94,
5 pls.

1950. Tropidocoryphinae im Karbon. Senckenbergiana, Frankfurt a.M., 31: 277-284,
elle amy
1951. Der Beginn des Karbons in Wechsel der Trilobiten. Senckenbergiana, Frankfurt

a.M., 32 : 219-266, 5 pls.

Ruprecut, L. 1937. Die Biostratigraphie des Obersten Kulm im Sauerlande. Jb. preuss.
geol. Landesanst., Berlin, 57 : 238-283.

LOWER CARBONIFEROUS TRILOBITES 24T

SWARBRICK, E. E. 1962. Facies changes in the Chert formation of the Lower Culm Series
of the Bampton area, North. Devon. Proc. Ussher Soc., Camborne, 1 : 30-31.

Tuomas, I. 1909. Notes on the trilobite fauna of Devon and Cornwall. Geol. Mag., London,
46 : 191-204, pl. 7.

Tuomas, J. M. 1962. The Culm Measures in north-east Devon. voc. Ussher Soc., Cam-
borne, 1 : 29-30.

Voapes, A. W. 1890. A bibliography of Palaeozoic crustasea. Bull. U.S. geol. Surv.,
Washington D.C., 63: 177 pp.

WEBBy, B. & Tuomas, J. M. 1965. Whitsun Field Meeting: Devonian of West Somerset
and Carboniferous of north-east Devon. Proc. Geol. Ass., London, 76 : 179-193.

WEBER, V. 1933. Trilobites of the Donetz Basin. Tvudy vses geol.-vazv. Ob” ed NKTP.,

Leningrad, 255 : 1-95, 3 pls.

1937. Trilobites of the Carboniferous and Permian System of the U.S.S.R., Paleontologiya
S.S.S.R., Moscow.

WELLER, J. M. 1936. Carboniferous trilobite genera. J. Paleont., Tulsa, Okla., 10: 704—
714, pl. 95.

Woopwarp, H. 1883-4. A monograph of the British Carboniferous trilobites. Palaeontogr.

Soc. (Monogr.], London. 86 pp., ro pls.

1894. Note on a collection of Carboniferous trilobites from the banks of the Hodder,
near Stoneyhurst, Lancashire. Geol. Mag., London, 31 : 484-4809, pl. 14.

1895. Trilobites. Jn Hinde, G. J. & Fox, H. Ona well-marked horizon of radiolarian
rocks in the Lower Culm Measures of Devon, Cornwall and west Somerest. Q. J. Geol.
Soc., Lond., 51 : 609-668, pl. 18.

T902. On a collection of trilobites from the Coddon Hill Beds, Lower Culm Measures,
near Barnstaple, north Devon, and one from Glamorganshire. Geol. Mag., London, 39:
481-487, pl. 20.

Fie.
1ey
Fic.
Fic.
Fic.
FIG.
x 4°8.

Fia. 7.

muscles.

SU UES CaS

PLATE 1
Phillipsia leei (Woodward)

All specimens from Lee Collection, British Museum (Nat. Hist.).
Posidonia shales, Waddon Barton Lane, nr. Chudleigh, N. Devon.

Lectotype.
Paratype.
Paratype.
In. 58283.
In. 58283.
Paratype.

Tn. 55935.
X 4°7.

In. 58281. Dorsal view of outer shell-lamina. X 4:8.

In. 58280. Ventral view of inside of outer shell lamina. x 4:8.
In. 58280. Counterpart of above. 48.

Mould of ventral side of free-cheek and interior of ventral doublure. x 5-2.
Counterpart of above: ventral view of outer shell-lamina. X 5:2.

In. 58283. Internal mould of pygidium and ventral surface of doublure.

Internal mould of pygidium, showing attachment scars of pygidial

Bull. By. Mus. nat. Hist. (Geol.) 14, 6 PLATE 1

GEOL. 14, 6. 24

PLATE 2

Fic. 1. Cyrtosymbole (Macrobole) cf. brevispina Osmdlska. BMNH. I. 3223. Chert
Formation, Park Gate Quarry, Tawstock, N. Devon. 6:1.

Fic. 2. Cyrtosymbole (Macrobole) cf. brevispina Osmdlska. BMNH. It. 444. Chert
Formation, Park Gate Quarry, Tawstock, N. Devon. xX 6:2.

Fic. 3. Cyrtosymbole (Macrobole) aff. laticampa Osmdlska. BMNH. It. 1445a, b.
Chert Formation, Claypit Coverts, West Buckland, N. Devon. X5'1.

Fic. 4. Typhloproetus cephalispina sp. nov. Holotype, BMNH. It. 1432. Bed 18,
Chert Formation, Warrenshill Quarry, Bampton, N. Devon. x 3°8.

Fic. 5. Cyrtosymbole (Macrobole) aff. laticampa Osmdlska. BMNH. It. 1432. Chert
Formation, Claypit Coverts, West Buckland, N. Devon. X2°8.

Fic. 6. Typhloproetus cephalispina sp.nov. BMNH. It. 1443d. Bed 18, Chert Forma-
tion, Warrenshill Quarry, Bampton, N. Devon. x8.

Fic. 7. Cyrtosymbole (Waribole)? chudleighensis sp. nov. M.R. House collection

1100 BMNH. It. 1433. Pit L, “ Posidonia’’ shales, N.E. of Mount Pleasant, Chudleigh, S.
Devon. X13.

PLATE 2

Bull. Br. Mus. nat. Hist. (Geol.) 14, 6

PLATE 3
Phillibole coddenensis (Woodward)

Fic. 1. Meraspid stage. BMNH. In. 22892. Dorsal view of inner shell lamina. Wheelton
Hind collection, Codden Hill. X 6'5.

Fic. 2. Counterpart of above ; ventral view of outer shell lamina. x 6°5.

Fic. 3. Lectotype, Woodward (1902, pl. 20, fig. 8). BMNH. I. 4560. Dorsal view of inner
shell lamina. Codden Hill Beds, Combe Quarry, Barnstaple. x 6:5.

Fic. 4. BMNH.I. 4565. A. Coomaraswamy collection. Dorsal view of inner shell
laminae. Low Culm Measures, Coombe Wood, nr. Barnstaple. x 6:5.

Fic. 5. Syntype, figured Woodward (1902, pl. 20, fig. 9). BMNH.1I. 4561. Dorsal view of
inner shell lamina. x 6:2.

Specimens uncoated.

PLATE 3

Bull. Br. Mus. nat. Hist. (Geol.) 14, 6

PLATE 4
Phillibole polleni (Woodward)
All from banks of the River Hodder near Stoneyhurst, Lancs.

Fic. 1. Lectotype, BMNH. It. 371a. External view of dorsal shield. Figured Woodward
(1894, pl. 14, fig.9). x 3°3.

Fie. 2. Counterpart of above. BMNH. It. 371b. External mould of dorsal shield. x 3°1.

Fic. 3. BMNH. It. 372. External view of dorsal shield. Figured Woodward (1894,
pl. 14, fig. 10). 3-6.

Fic. 4. BMNH. It. 373. Internal view of outer shell lamina of free cheek. Figured
Woodward (1894, pl. 14, fig. 11). X62.

Bull. Br. Mus. nat. Hist. (Geol.) 14, 6 PLATE 4

oe Aeon es ;
ES age EE A= wheat Ted

Y 4

PLATE 5
Liobole glabra (Holzapfel)

Fic. 1. BMNH. It. 1436. Chert Formation, Park Gate Quarry, Tawstock, N. Devon. x BO,

Fic. 2. BMNH. It. 1438. Chert Formation, Bed 1 5, Warrenshill Quarry, Bampton, N.
Devon. X 2:3.

Fic. 3. BMNH. It. 1437. Chert Formation, Park Gate Quarry, Tawstock, N. Devon. x 3:1.

Fic. 4. N.D.A. 822. Chert Formation, Templeton Quarry, Tawstock, N. Devon. x I'5.

Fic. 5. BMNH. It. 381. Chert Formation, Codden Hill, N. Devon. x 5:3.

PLATE 5

Bull. Br. Mus. nat. Hist. (Geol.) 14, 6

PLATE 6
Spatulina longispina sp. nov.

Fic. 1. Holotype. BMNH. It. 1440a. Internal mould of carapace with inner lamina
adhering. Bed X, Chert Formation, Park Gate Quarry, Tawstock. X 3:1.

Fic. 2. Counterpart of holotype. BMNH. It. 1440b. External mould of carapace with
outer lamina adhering. Bed X, Chert Formation, Park Gate Quarry, Tawstock. X93:I.

Fic. 3. Internal mould of pygidium with inner lamina adhering. BMNH.It. 1441. Bed X,
Chert Formation, Park Gate Quarry, Tawstock. 2:7.

Fic. 4. Internal mould of free cheek. BMNH. It. 1441. Bed X, Chert Formation, Park
Gate Quarry, Tawstock. 4-4.

Spatulina spatulata (Woodward)

Fic. 5. Internal mould of cephalon, with inner lamina adhering. (KCL. t. 173). Chert
Formation, Park Gate Quarry, Tawstock. x 4:7.

Fic. 6. External mould of cephalon, with outer lamina adhering. (KCL.t.172). Chert
Formation, Park Gate Quarry, Tawstock. X 4:3.

PLATE 6

Bull. Br. Mus. nat. Hist. (Geol.) 14, 6

*

*

eae

“hs
xt
|

PLATE 7
Diacoryphe? vandergrachtii (Woodward)

Fic. 1. Lectotype. BMNH. It. 1446. Banks of R. Hodder, near Stoneyhurst, Lancashire.
a 2. Syntype. BMNH.It.1447. Banks of R. Hodder, near Stoneyhurst, Lancashire.
mE 3. Syntype. BMNH. It. 1448. Banks of R. Hodder, near Stoneyhurst, Lancashire.
he. 4. Same as figs. 2,3. BMNH.It. 1449. X1Io.

Fic. 5. Sameas figs.2,3. BMNH.It. 1450. X1Io.

Phillibole aprathensis R. & E. Richter

Fic. 6. BMNH. It. 1434. Chert Formation, loose block above Warrenshill Copse, Bampton,
N. Devon. X3.

Fics. 7, 8. BMNH. It. 1435. Bampton Limestone Group, Little Holwell Quarry, N. of
Bampton, N. Devon. X6°5.

Fic. 9. BMNH. I. 4563. Chert Formation, Hannaford Quarry, Barnstaple, N. Devon. X 4:7.

PLATE 7

Bull. By. Mus. nat. Hist. (Geol.) 14, 6

e

Re Were
ee

AS TN |
12 JUN 1967

N
SS

"\

Sh

PRINTED IN GREAT BRITAIN —
BY ADLARD & SON LIMITED |
BARTHOLOMEW PRESS,

: vit
: et
*] v
i iy Oe #!
ui
le
‘

ew

Y 0 JUL v6

FOSSIL MAMMALS OF AFRICA
Naw 22

ELOROVIS OLDOWAYENSIS RECK,
AN EXTINCT BOVID FROM
EAST AFRICA

-

Aare Wie GENTRY

me BULLETIN OF
4 THE BRITISH MUSEUM (NATURAL HISTORY)

_ GEOLOGY Vol. 14 No. 7
| LONDON : 1967

tT

i
*

ee

=

FOSSIL MAMMALS OF AFRICA ae
INO 22 \

PELOROVIS OLDOWAYENSIS RECK,
AN EXTINCT BOVID FROM EAST AFRICA

BY
ALAN WILLIAM GENTRY, D.Phil.

(Centre for Prehistory and Palaeontology, ‘Nairébi)

Pp. 243-299; 6 Plates; 40 Text-figures

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)

GEOLOGY Vol. 14 No. 7
LONDON : 1967

THE BULLETIN OF THE BRITISH MUSEUM
(NATURAL HISTORY), imstituted in 1949, 1s
issued in five series corresponding to the Departments
of the Museum, and an Historical series.

Parts will appear at irregular intervals as they become
veady. Volumes will contain about three or four
hundred pages, and will not necessarily be completed
within one calendar year.

In 1965 a separate supplementary series of longer
papers was instituted, numbered serially for each
Department.

This paper is Vol. 14, No. 7 of the Geological
(Palaeontological) series. The abbreviated titles of the
periodicals cited follow those of the World List of
Scientific Periodicals.

World List abbreviation
Bull. Br. Mus. nat. Hist. (Geol.).

© Trustees of the British Museum (Natural History) 1967

TRUSTEES OF
THE BRITISH MUSEUM (NATURAL HISTORY)

Issued 21 July, 1967 Price {1 16s.

FOSSIL MAMMALS OF AFRICA
INjo: 22

PELOROVIS OLDOWAYENSIS YRECK,
mee LINCT, BOVID FROM, EAST. APRICA

By ALAN WILLIAM GENTRY

SYNOPSIS
New material of Pelorovis oldowayensis Reck from the upper part of Bed II at Olduvai shows
that this animal is not a member of the Caprinae, but belongs instead to the Bovini and is related
to the African buffalo Syncerus Hodgson. Material used to describe Bularchus arok Hopwood
in 1936 is found to be assignable to Pelorovis oldowayensis. There is possible evidence for a
second Bovine species in Bed II.

CONTENTS

Page
I. INTRODUCTION . c é : - . : 5 é 0 §6PAGR

II. DESCRIPTION OF NEW MATERIAL OF PELOROVIS OLDOWAYENSIS
RECK . : : : : ‘ : : : : : 249
III. THE SySTEMATIC POSITION OF PELOROVIS : ; : ¢ 265
IV. A Note on BULARCHUS AROK Hopwoop ‘ : . : 290
V. PHYLOGENETIC AND FUNCTIONAL CONSIDERATIONS P ‘ : 294
VI. ACKNOWLEDGMENTS . : ; ; i ; ‘ : ‘ 297
VII. Summary ‘ : : . : F 4 , : ; 297
VIII. REFERENCES . 5 ; : ‘ : ; F j 3 298

I INTRODUCTION

In 1952 and in several subsequent years Dr. & Mrs. L. S. B. Leakey excavated at site
BK II in Olduvai Gorge, a site which they had first located in 1935 and which is
now known to lie near the top of Bed II. It consisted of a former land surface
covered with sands and with extensive clay deposits above the sand, the land surface
being littered with large numbers of splintered bones, as well as with stone tools and
waste flakes of an evolved Oldowan culture. Adjoining the former land surface
was a clay-filled gully in which more completely preserved animal remains including
some articulated limb bones were found. Leakey (1954) believes that the men who
used the land surface drove their prey into the mire of the gully and then dragged
them out to be eaten, but that occasionally the larger animals had to be left behind.

From BK II were excavated a complete skull and many skull fragments of the
large Bovid Pelorovis, also many limb bones and vertebrae among which the best
preserved set of limb bones was known to have come from the same individual as
the complete skull.

Remains of Pelorovis had first been recovered from Olduvai by the German expedi-
tion of 1913, and described as the single species P. oldowayensis by Reck (1925, 1928).

GEOL. 14, 7. 25

246 PELOROVIS OLDOWAYENSIS RECK

The holotype in Berlin is somewhat weathered and consists of the back part of a
skull with its horn cores ; since the front of the skull is missing no teeth were avail-
able for Reck to work on. In his 1928 article, kindly translated for me by Mr. &
Mrs. G. J. Warren, Reck noted that possible similarities to previously described
fossils were few, but on the basis of horn core characters he made comparisons with
various Caprines, and only briefly considered the possibility of relationship with
other Bovidae such as the African buffalo Syncerus, or the Alcelaphini. He finally
chose the sheep as being least far phylogenetically from Pelorovis, but was aware
that the East African animal could have evolved its horn shape in isolation from
other known living or fossil sheep. In fact we can now be certain, with more complete
material and particularly from features of the teeth, that Pelorovis is not any kind
of Caprine.

In the following diagnosis and expanded definition the skull is imagined as having
its tooth row horizontal.

Genus PELOROVIS Reck

TypE SPECIES. Pelorovis oldowayensis Reck, 1928.
GENERIC CHARACTERS. As for the species.

Pelorovis oldowayensis Reck
1928 Pelovovis oldowayensis Reck: 57, pls. 1, 2, text-fig. 1.
1936 Bularchus arok Hopwood: 639.
Ho.otyre. Back part of a skull with horn cores, in the Institut fiir Palaontologie
und Museum der Math.-Naturwissenschaftlichen Fakultat der Humboldt-Universitat,
Berlin.

REFERRED MATERIAL. A complete skull numbered Pel 1; horn cores and skull
pieces numbered Pel 2 to Pel 23; an unnumbered frontlet and a left horn core ;
teeth, limb bones and vertebrae as listed at various places in the text. This material
is in the National Museum of Tanzania, Dar es Salaam, and a cast of the complete
skull is in the British Museum (Natural History), London. Also in the British
Museum (Natural History) is a frontlet with horn cores M.14947, a frontlet with left
horn core M.14948, a horn core tip M.14949, and paired mandibular rami M.15856 ;
all this material had been assigned to Bularchus arok. Also in London are teeth
numbered M.25676—81, M.25688 and M.25692.

Horizon. The holotype is of unknown provenance within Olduvai Gorge ;
most of the remains in Dar es Salaam are from site BK II, but there are also some
from SHK and other sites in the upper part of Bed II. M.14947—48 were originally
supposed to have come from Bed IV, but they are now believed to have come from
Bed II (see p. 290). M.15856, M.25676-81, M.25688 and M.25692 are from Kanjera,
a site which has usually been taken to be equivalent to Bed IV at Olduvai.

AGE. Middle Pleistocene.

Dracnosis. A large Bovid with long curved horn cores without keels, horn cores
inserted close together and so far posteriorly on the skull that they overhang the
occipital surface. The face is long, the tooth row is placed anteriorly, and the median

PELOROVIS OLDOWAYENSIS RECK 247

indentation at the back of the palate has its anterior edge further forwards than the
lateral ones. Teeth moderately hypsodont, but have only small basal pillars, and
their occlusal pattern is not advanced. The anterior part of P, has a medial wall.

DEFINITION. The preceding diagnosis selected the most noticeable characters of
the skull; in this expanded definition are listed all those characters which I have
found useful in comparing P. oldowayensis with some other Bovidae.

A large Bovid with a long and low skull. Its massive horn cores are slightly
compressed dorso-ventrally (with the skull in a horizontal position) and without
keels or transverse ridges; they arise close together, well behind the orbits and
partly behind the occipital surface as well, then they pass successively backwards
and outwards, then outwards and perhaps slightly downwards, then forwards and
slightly or markedly upwards as well. When it is at all evident, the spiralization
is therefore clockwise in the right horn from the base upwards. The horn cores
taper gently from base to tip, sometimes a deep longitudinal groove runs along their
back outside curve. Females have shorter and more sharply curved horns than the
males. The horn cores and frontals are hollowed.

The skull is about as wide at the orbits as across the occipital surface ; the frontals
curve slightly downwards in lateral view as they pass forwards over the orbits. The
orbits are small (probably an allometric effect) and the orbital rims project only
slightly. Supraorbital foramina are found above the back part of the orbits or a
short distance behind them in longitudinally extended supraorbital pits. Behind
the orbits is a long temporal fossa to house the coronoid process of the lower jaw
and its attached musculature. The nasals are rather domed transversely, and they
are widest just behind the point where their front part loses contact with the maxillae ;
their central anterior flanges are pronounced and there are no lateral anterior flanges.
The back of the nasals probably lies above the front half of the orbits. There is
no ethmoidal fissure nor a localized preorbital fossa. The anterior part of the lower
edge of the zygomatic arch passes forwards well below the orbits. The infraorbital
foramen lies above or in front of P?. The palatal fissures of the premaxillae appear
to have been rather small; the premaxillae become narrower as they rise, and have
little or no contact with the nasals. The tooth row is set rather anteriorly, and the
median indentation at the back of the palate passes further forwards than the lateral
ones. The vomer is not fused with the back of the palate. At the top of the
occipital surface is a deep inverted triangular depression, the mastoid exposure of
the periotic is entirely on the occipital surface, and the occipital condyles are wide.
The top edge of the foramen magnum is not so posterior relative to the condyles
as in Ovyx. The anterior tuberosities of the basioccipital are sufficiently close
together to give the bone a triangular shape ; they are large and have poor longitu-
dinal ridges extending behind them ; there is a poor longitudinal groove along the
centre of the basioccipital ; and the bone is not transversely constricted across its
middle. The foramina ovales are small and situated just in front of the level of
the anterior tuberosities. There is an indentation in the sides of the squamosal
shelf immediately in front of the mastoid. The auditory bulla is a little compressed
from side to side and appears to be but little inflated. The tips of the paraoccipital
processes are not noticeably turned forwards.

248 PELOROVIS OLDOWAYENSIS RECK

The teeth are moderately hypsodont ; their enamel surfaces, having many tiny
longitudinal striations, may be described as rugose. The upper molars have wide
and quadrate occlusal surfaces ; basal pillars are present but not strongly developed ;
the walls of the central cavities of the teeth do not have a complex outline; the
styles are not strongly developed ; the outward bowings of the walls between the
styles are neither localized nor strongly developed ; and cement is present. The
mandible is deep below the teeth ; the anterior edge of the coronoid process is only
slightly curved backwards (this is correlated with the relatively anterior position
of the upper tooth row) ; and the anterior part of P, has a medial wall, thereby closing
off a central valley on the tooth.

The great trochanter of the femur is tall with a slanted antero-dorsal edge ; there
is a slight indentation between the great trochanter and the articular head in anterior
view and the top edge of the articular head is a little slanted in anterior view ; the
articular head is not very narrowed in its lateral part in dorsal view ; distally the
lateral condyle is not sharply pointed anteriorly; and the patellar fossa is wide. On
the tibia there is no middle patellar groove at the top of the cnemial crest ; the
lateral facet on the top articular surface is without an upturned lateral edge; and
distally the medial malleolus exceeds the central anterior flange in length. On the
astragalus there is a deeply incised facet at the back of the medial side for the
naviculo-cuboid ; the ridge for the astragalo-metatarsal ligament on the medial side
is present ; this ridge is at the same level or slightly lower than the ridge for articula-
tion with the medial malleolus of the tibia ; the top of the back of the medial side
of the astragalus does not project behind or away from the main mass of the bone.
The naviculo-cuboid is not deep; and the back edge of its medial wall is fairly
straight. The metapodials are not antero-posteriorly compressed ; and they possess
distal anterior and posterior foramina. The articular facets at the top of the meta-
tarsal are almost flat.

On the scapula the tuber scapulae is set above the anterior edge of the glenoid
facet ; the lateral side of the glenoid facet in ventral view is slightly indented ; and
the area for the origin of the teres minor extends forwards to the base of the spine.
The lateral tuberosity of the humerus is low; its posterior eminence is small and
not antero-posteriorly long ; the infraspinatus insertion is longer than it is deep ;
the front of the infraspinatus insertion is level with the front edge of the lateral side
of the bone ; the bicipital groove is wide; distally there is a coronoid fossa; and
the medial condyle is tall. The radius is rather long relative to the humerus for a
Bovid the size of Pelorovis. On the ulna the area for the origin of the flexor carpi
ulnaris extends near to the top of the bone, i.e. the roughened area at the top of the
olecranon is smaller. On the radius the medial edge of the medial facet on the top
articular surface does not project as a rim; the postero-medial part of the medial
facet is not greatly expanded; the lateral facet is antero-posteriorly long; the
lateral tubercle is moderately sized; the axis is not swollen at its distal end in
lateral view ; and the ridge between the posterior surfaces of the scaphoid and lunate
facets is slanted and not sharply marked. The carpal bones are antero-posteriorly
long and not tall. The tubercle towards the front of the upper facet of the scaphoid
is little pronounced ; and the lower edge of the scaphoid is little indented in medial

PELOROVIS OLDOWAYENSIS RECK 249

view. Because of the length of the lunate its ventral projection where it passes
behind the unciform is rather forwardly sited; the back edge is not very pointed
in medial view ; and the more posterior lower parts of the lateral side project little
in dorsal view.

The cervical vertebrae are transversely wide and antero-posteriorly short ; with
neural spines not slanted forwards. The hollows on the ventral surface of the atlas
are deep. The vertebrarterial foramina on the axis are small. The anterior and
posterior openings of the foramen transversarium on the third cervical are close to
the front and back ends of the centrum ; and the transverse processes of this vertebra
are already separated from their ventral flanges.

REMARKS. It is clear that Pelorovis cannot belong to the Caprinae, as can be seen
from the skull characters alone. Its skull is relatively lower in side view than in
Caprinae, its horn cores are set far behind the orbits and there is a temporal fossa,
the premaxillae are not reduced as in some Caprine tribes, the anterior tuberosities
of the basioccipital are not set very widely apart, its upper molar teeth are wide,
they have basal pillars, they have no tendency towards exaggerated styles on the
lateral walls, and there is not a pronounced reduction in length of the premolar row.
The purpose of this paper is to suggest an alternative relationship for Pelorovis.

II DESCRIPTION OF NEW MATERIAL OF PELOROVIS

Skulls and horn cores
The complete skull

The complete skull of Pelorovis oldowayensis (Pls. I, 2, 6) has been distorted in
various ways—the tips of the premaxillae have been thrust upwards close to the
nasals, the bones on the left side of the face have become separated, and the right
P? and most of the left premolar row are missing. Enough of the skull has been
preserved to suggest that drawings of how it might have appeared in life would be
feasible, and the three resulting reconstructions are shown as Text-figs. I—3.

Most of the characters mentioned in the definition can be seen on the complete
skull. Its horn cores are not large in relation to the size of the skull; the length of
the right one along its back edge is only 116 cm. which may be compared with 153 cm.
for the two 1957 horn cores numbered Pel 7 and Pel 8. The short length and the
sharp radius of curvature which reduces the span of these horn cores suggest that
the animal was a female. The sharpness of the curvature of the horn cores precludes
the possibility of them later growing to a size comparable with that of Pel 7, and the
teeth show that the animal was already fully adult. Another pair of horn cores
numbered Pel 18 and Pel 19 are a little smaller than those of the complete skull,
otherwise the known horn cores are intermediate in size between those of the complete
skull and Pel 7, without sign of a sharp distinction of males from females. In so far
as the curvature of the horn cores in three planes involves spiralization, the direction
of the spiralization is clockwise from the base of the right horn core. Nothing of
interest concerning the horn cores can be added to Reck’s (1928) detailed description
of the Berlin specimen.

250 PELOROVIS OLDOWAYENSIS RECK

Fics. 1-3. Reconstructions of skull of Pelovovis oldowayensis. The reconstructions are
based on the complete skull Pel 1 for the pre-orbital regions, and on specimen Pel 2 in
conjunction with Pel 1 for the post-orbital regions. 1, ventral view; 2, lateral view
showing sutures of lachrymal and jugal ; 3, dorsal view.

The supraorbital foramina cannot easily be seen on the skull; they lie above the
level of the back edge of the orbits or very slightly behind them, and are not widely
separated from one another. On the left side there is a very small foramen just
in front of the position of the main supraorbital foramen. A narrow and shallow
longitudinal furrow runs along the bone surface in front of the supraorbital foramina
on both sides almost as far as the level of the back of the nasals. The anterior parts
of the nasals are transversely domed, and the downflanged lateral parts of the nasals
anteriorly are fused with the maxilla as often happens in large Bovidae. Although
there is no localized preorbital fossa, the whole region of the face in front of the orbits
and on either side of the nasals has a hollowed surface. Towards the back of the left
nasal a piece of what is presumably the lachrymal rises up and shows the beginning
of the posterior narrowing of the nasal; the back of the nasals is not otherwise
indicated. On the left side of the face part of the suture marking the maxilla-
lachrymal boundary can be seen, and lower down a more doubtful indication which,
if it is a suture at all, marks part of the maxilla-jugal boundary ; there is nothing
worthy of comment in the position of these sutures, and they are shown in Text-
fig. 2.

PELOROVIS OLDOWAYENSIS RECK 251

The lower part of the occipital surface is at an angle of 90° to the plane of the top
of the skull, and two low rounded ridges run up the occipital to reach the sides of
the inverted triangular depression. The paraoccipital processes have been broken
off. The basioccipital has both its anterior tuberosities complete. A small basal
pillar can be seen on the left M? and somewhat larger ones on both Ms ; a good deal
of cement remains around the basal pillars.

The lower jaws of this animal (Pl. 6), which were found in position on the under-
lying skull, are undistorted. All teeth are present except the right P,, but the left
P, has been worn very low. The basal pillars on M, and M, have been nearly worn
away, but that on Mgisstill present. The front cavity on M, has almost disappeared.

Remains of other skulls and horn cores

All but one of the other specimens of Pelorovis are of posterior parts only. The
most useful of these is numbered Pel 2, and possesses the proximal parts of both
horn cores with a complete but fractured occipital surface between them (PI. 3).
The left side of the top of the skull is missing, presumably because the men broke
into it to get out the brain ; this had not happened to the complete skull which was
found in the clay filled gully. A feature to be noted is the deep groove running
along the back outside curve of the horn cores. On the right upper side of the skull
is a lengthened supraorbital pit containing two foramina, the front one of which lies
over the back of the orbit. The posterior upper part of the right orbit has been
preserved although the rim is nowhere complete. The top of the occipital surface
is evenly curved and marked by a small but sharp ridge even along those parts of
its course which are overhung by the horn cores’ bases ; this ridge has a more constant
radius of curvature from the foramen magnum than it does in Syncerus. An inverted
triangular depression is again seen at the top of the occipital surface. The basi-
occipital is present but without its anterior tuberosities ; one foramen ovale and
the larger foramen rotundum can be seen, also the deep and long temporal fossae
beneath the bases of the horn cores. Parts of the squamosal shelf for articulation
with the condyle of the mandible are present on both sides, and the thickened base
of the right paraoccipital process is present.

An unnumbered complete pair of horn cores with part of the frontals and still
less of the occipital surface is shown in Pl. 2. The length of the outwardly directed
part of the horn cores is greater than in the complete skull and their upward curvature
is less marked.

Pel 5 is a distorted pair of horn cores with the occipital, the condyles and the
basioccipital still surviving. The radius of curvature of the horn cores is quite small,

-and both of them have a moderately sized shallow longitudinal groove running along
their back curve. The anterior tuberosities of the basioccipital are missing.

Pel 3 is a less complete skull piece than Pel 2, but it has an almost complete and
undistorted temporal fossa and part of the orbital rim on the left side. The specimen
thus shows well the transverse constriction of the skull behind the orbits, a feature
which recalls Bos (in which the orbital rims are more projecting and the horn cores
more widely set than in Pelorovis). The gap between the bases of the horn cores

252 PELOROVIS OLDOWAYENSIS RECK

along the top of the frontals is narrow and appears as a deep longitudinal incision ;
the inference from this is that the horns of Pel 3 would have been larger and longer
than those of the two skulls described above. The lack of the horn cores above their
bases and of most of the right frontal allows the hollowing in the bone above the
brain to be seen. The horn core cavity extends to within 4 cm. of the level of the
back of the orbits, and the frontals are hollowed in front of this point, but without
connection to the horn core cavity. The level to which the skull is preserved along
the frontals is about the same as that to which it is preserved in Pel 2 and in the
type specimen in Berlin. On the ventral surface the anterior tuberosities of the
basioccipital and the paraoccipital processes are again missing. The foramina
ovales are in the horizontal plane and slightly in front of the anterior tuberosities.

The skull Pel 4 has retained part of its right horn core, the occipital surface, a
small part of its ventral surface and part of the temporal fossa; the anterior tubero-
sities of the basioccipital are missing. The triangular depression at the top of the
occipital surface is less pronounced than in Pel 2.

A number of Pelorovis horn cores have been taken from BK and other sites in Bed
II at Olduvai. One of the largest is Pel 7, a left horn core excavated in 1957 from
BK, which is 153 cm. long. About 12 cm. from its base there begins a deep groove
which runs along the back edge as far as about 32 cm. from the tip. A small
part of the occipital surface is preserved at the horn core’s base. Towards its
tip the horn core is turned well upwards, and thus comes to possess detectable
spiralization.

Pel 8 is a right horn core from BK II which is again 153 cm. long and with about
the same degree of spiralization as Pel 7; however it may belong to a different
individual from Pel 7 because of its lack of a deep longitudinal groove and because
the patterning of the bony surface of the frontals at the base of the horn core seems
to be different. Part of the temporal fossa and a small part of the brain cavity are
also preserved.

Pel g is a right horn core with the excavation number 1957, BK II, 1344. Spiral-
ization is slightly less than in Pel 7 and 8, 1.e. the tip is not quite so upwardly turned
and remains directed mainly forwards. A rather wide, shallow, longitudinal groove
exists in its more distal parts in a position slightly dorsal to that occupied by the
much longer and deeper groove mentioned in Pel 7.

A left horn core, Pel 10, of which only about 40 cm. exists, may be the partner of
the complete horn core Pelg. A further piece of this left horn core may be preserved
as the tip numbered Pel 11 which has a similar shallow longitudinal groove.

Pel 12 is a right horn core which is a little smaller than those mentioned previously
and is less upwardly curved towards its tip. There is no longitudinal groove along
the back surface. A very small part of the frontal and of the occipital surface has
survived.

Pel 13 is a left horn core without its tip; it is more sharply curved than Pel 12,
and a wide shallow, longitudinal groove exists for a short distance in its more distal
parts.

Pel 14 is an incomplete right horn core, rather little compressed and with some
degree of spiralization. Since the core is broken off about 60 cm. above its base,

PELOROVIS OLDOWAYENSIS RECK 253

the degree of spiralization is only reliably indicated by the course of the longitudinal
grooving and ridging on the lower part of the core. A wide and marked but rather
shallow longitudinal groove begins on the back outside curve of the horn core about
15 cm. from the base. A fragmentary right horn core, Pel 17, could come from the
same individual as Pel 14.

Pel 15 is a left horn core with part of the frontal, occipital and basicranial surfaces
attached, and the brain cavity itself has survived practically complete. The anterior
tuberosities of the basioccipital are missing and the left foramen rotundum can be
seen. The base of the right horn core is also present. The left horn core overhangs
the top of the occipital as in other Pelorvovis specimens.

Pel 16 is a much damaged fragment of a left horn core with part of the frontal
remaining at the horn base.

Pel 22 is a right horn core which exists in two unconnected pieces, and shows
much spiralization. Its span would have been about 69 cm., and compares with
about 41 cm. for the female horn core Pel 18 (see below) and about 93 cm. for the
large Pel 7.

Pel 6 is a very crushed and incomplete right horn core about 70cm. long. It
comes from site SHK II about 26 feet lower than BK II.

Except for Pel 1, all the horn cores and skull remains hitherto considered may be
regarded as being either of males or of indeterminate sex. Two further horn cores
agree with Pel i in appearing to belong to female animals, these have been numbered
Pel 18 and Pel 19. They are evenly tapered along their entire length, but consider-
ably shorter and more sharply curved than the other horn cores ; the tips are more
upwardly curved than in males of comparable size. The right horn core, Pel 18,
has attached to it a part of the frontal, the back of the orbit and the roof of the
temporal fossa. The temporal fossa is less deeply excavated than in either of the
skull pieces numbered Pel 2 and Pel 3, and this is presumably linked with the
smallness of its horns. Pel 19 is very probably the same individual as Pel 18.

There are also many other fragments of distal ends or the more distal parts of
horn cores assignable to Pelorovis ; from these broken pieces we can see that the
horn cores had thickened outer shells, inside which was a central cavity becoming
narrower towards the tips. At any one level of cross section the thickness of the
outer shell varies, but I did not find any of the internal struts which Reck (1928)
refers to and illustrates.

A well preserved basicranial piece, Pel 20, shows several interesting features.
The occipital condyles, being complete and undistorted, are wide as in other Pelorovis
remains; the basioccipital has its anterior tuberosities with slight longitudinal
ridges behind them ; the bases of the auditory bullae show them to have been rather
compressed and presumably poorly inflated. Both external auditory meati can be
seen. This specimen is the only one to retain complete paraoccipital processes, they
are short and squat and their tips are not at all forwardly directed. Parts of the
occipital surface and of both temporal fossae are also preserved, but they give no
information not already available from other specimens.

There is an occipital fragment, Pel 21, possessing both condyles and part of the
right temporal fossa.

254 PELOROVIS OLDOWAYENSIS RECK

Tooth remains

From a badly crushed skull, Pel 23, there have been preserved parts of both
maxillae and premaxillae (Pl. 3) and of the tooth rows. The premaxilla is wide at
its base but narrows rapidly as it rises; at its top it joins the maxilla in a sutural
contact with the nasals, as often happens in large Bovids. Its teeth are a little
smaller than others assigned to Pelorovis.

A good number of fossil maxillary and mandibular pieces and isolated teeth belong-
ing to large Bovids are present in the collections from Olduvai. Many of them have
morphological characters appropriate for very large Alcelaphini, but the largest
among them are clearly not Alcelaphine and must belong to Pelorovis. Examples
of its upper teeth are distinguished from those of Alcelaphini by the fact that the
whole tooth row is more or less straight and not set in an arc, by the presence of
rather weakly developed basal pillars, and the relatively simple course of the enamel
walls of the central cavities. They appear to be less hypsodont than the Alcelaphine
teeth, although this is an awkward feature to assess except with unworn examples
of both groups for comparison. Finally, the upper teeth of Pelovovis are wider than
those of Alcelaphini. Many examples of both upper and lower Pelorovis teeth show
rugosity of the enamel, a characteristic which is perhaps of use in binding the sur-
rounding cement to the tooth. The largest lower teeth of Bovidae in the Olduvai
collections differ from those of Alcelaphini by the presence of basal pillars. It is
apparent from looking at the mandibles and lower teeth that the size of a basal pillar
will change with the stage of development of its tooth. In little worn teeth the
level of the top of the basal pillar may not have been reached, later the level of
maximum cross section is reached, finally in much worn teeth the level of the bottom
of the basal pillar is passed, and it becomes joined to the body of the tooth.

1957, SHK II, 232 (Pl. 3) is an adult right maxilla with P* to M®._ Its teeth are
comparatively little worn, and they show the relatively simple occlusal surfaces of
the teeth of even younger Pelorovis.

A number of isolated upper molars are mostly smaller than the M?s in the complete
skull ; they all have traces of infoldings into the central cavities but these are never
very pronounced. The outline of the occlusal surfaces becomes more squared medially
as the teeth are worn down.

A few pieces of large Bovid mandibular ascending rami belonging to Pelorovis
are known: 1953, BK II, area C, g1?, can be identified by the relative lack of curva-
ture on the anterior edge of its coronoid process. Its mandibular foramen is situated
about half way between the front and the back edges. Two other condyle fragments
are a right and left pair with the excavation numbers 1955, BK II, 5 and 6. The
top of a left ascending ramus, 1952, BK II, 123, differs from the other mandibular
remains of Pelorovis by the lowness of its coronoid process above the condyle ; it
could perhaps have come from a younger and smaller animal.

A number of specimens of lower teeth give information additional to that available
from the more worn teeth of the mandible belonging to the complete skull.

1952, BK II, 117 (Pl. 3) is a left mandibular fragment containing all of the teeth
from P, backwards except that M, is without its posterior lobe. The top of the

PELOROVIS OLDOWAYENSIS RECK 255

jaw is rather more concave than in older specimens. The P, has not been long
erupted and the anterior part of its medial wall is only just closed at the top. The
tops of the medial edges of the lobes on the molars rise high above the level of the
tops of the intervening styles. The basal pillar is small on M,, was probably bigger
on Mg, but is not yet visible on Mj. The anterior end of the comparatively little
worn M, has a sharply medially turned flange, but this would have become less
pronounced with increasing age. Each molar has little contact with the molar
immediately in front.

1952, BK II, 118 is another fragment of a left mandible, but here M, is the only
tooth to have survived, P, to M, being broken off at their necks. M, has been about
as much worn down as in the mandible of the complete skull. A great amount of
cement has been preserved between the first and second lobes of M3, and the first
lobe has been excessively worn down on its medial side.

1952, BK II, 119 (Pl. 3) is another left mandible fragment, this time preserving
at least parts of all its teeth. The adult premolar row is newly erupted, P, has a
closed anterior part of its medial wall, there is only narrow contact between M, and
M,, and there is again a strong medially turned flange at the front of Ms. The front
of M, has been pushed into the back of M,, and since the underlying part of the jaw
bone is complete and undistorted this appears to have occurred in the living indi-
vidual. In addition the plane of the occlusal surface of the front of M, shows that
IIg is a second example of an animal with a tooth deformity. This jaw comes from
a younger animal than 117 and since the incomplete but little worn M, is exposed
to its roots, one can get an impression of the hypsodonty of the species. The ramus
looks as if it is less deep than the one belonging to the complete skull. The medial
sides of the lobes do not rise so high in comparison with the styles as in 117, but this
probably has no taxonomic significance. The level of the top of the basal pillar has
not been reached by the occlusal surface of Mg, the pillar is quite large on M,, and
its condition in M, cannot be clearly seen.

1957, BK II, 1032 is a left mandibular fragment which has M, and M, in an early
state of wear, hence they are not very wide at their occlusal surfaces. The last lobe
of M, has not been worn at all, and the central cavities of its first and second lobes
are still open medially at their posterior ends. Basal pillars are not visible on either
tooth, perhaps because the pillar of M, is hidden under cement while that of M,
has not yet been reached.

1952, BK II, 127 is a left mandibular fragment with M, to M3, the central cavity
of the front lobe of M, having been worn away. An unnumbered specimen consists
mainly of plaster around a left M, to M, in which much cement is present and the
front central cavity of M, is again missing. 1953, BK II Extension, 79 is a right
mandibular fragment with M, and most of M, in about the same state of wear as the
last two pieces. An unnumbered fragment from MRC II is a piece of a left mandible
with M3, M, and the roots of M,.

There are a good number of isolated lower teeth among which some show points
of interest. Thus 1957, SHK II, 165 is a little worn right M, which shows anteriorly
a medial flange becoming less prominent towards the base of the tooth; a basal
pillar is present all the way down. 1952, BK II, 129 (Text-fig. 4) is a little worn right

256 PELOROVIS OLDOWAYENSIS RECK

M, or M, which shows well the height of these animals’ teeth and in addition preserves
one of its roots complete.

A lower left molar, 1953, BK II Extension, 355 (Text-fig. 5) is likely to have
belonged to a Pelorovis yet differs from other lower molars in having less pronounced
and more localized outbowings of the medial walls between the styles.

———————1

0
1 CM

ee |

4 1952 BKII 129 > 1953 BK Il EXT. 355

Fics. 4, 5. Medial view of right lower molar 1952, BK II, 129, and occlusal view of
left lower molar 1953, BK II Extension, 355. The arrows point anteriorly.

Measurements of skulls and teeth

On the complete skull—Pel 1—the following measurements, all expressed in
centimetres, were taken:

Skull length from the front of the premaxilla to the back of the occipital condyles,

estimated before distortion at . c é 6 c 5 : : 68:8
Skull width across the posterior side of ing onsites 0 c ¢ : 27°4
Distance from the front of the premaxilla to the nearest aoe oi on the pcbital rim . : 42°4
Length of horn core along its back edge : 5 336)
Dorso-ventral diameter of horn core at a distance “Srenm its pce alone the back edge

equal to half the maximum diameter at its base . , A ‘ : 5 ‘ 9:7
Diameter of the horn core at 90° to above measurement . . 12°4
Width between the supraorbital foramina, taken between the cone moins of fag

lateral walls. : : . 4 : 5 ; : : 0 116
Length of nasals, estimated ah : : : : : : : : 31°8
Width of nasals, taken across level a in Text fig. I. : 6:8
Distance from the front of premaxilla to rearmost point of (occlusal surface of MS,

estimated before distortion at . é 34:9
Distance from the rearmost point of occlusal sums of NG to pace of occipital somdavies

estimated before distortion at . : , 34°0
Occipital height from the top of foratien meee to top a pecipirall crest) c : II‘
Skull width across mastoids immediately behind external auditory meati . : A 27°3
Width across anterior tuberosities of basioccipital . 6 : ; 9 . 5 4°9
Width across posterior tuberosities of basioccipital . : . : é : : 8:5
Occlusal length M! to M?. c ‘ ; : : 0 : A F 6 II'l
Occlusal length of M? . : 0 : : : 5 6 c 0 . : 3°75
Occlusal width of M? . : : j : 5 : F : ; p 5 3°3
Occlusal length M, toM, . : : : j : ‘ ; : é : 11'8
Occlusal length M, A , j 5 j j : i i ; : é 3°5
Occlusal width M, 5 5 c é é ; : : : : 6 : 2:1
Occlusal length P, tOrPAR : : 5 : : : : : 6:0

Occlusal length as the feawily worn P, 4 : 5 6 ; . : : 15

PELOROVIS OLDOWAYENSIS RECK 257

Measurements on other skull pieces of Pelorovis were:

Pel 2 Pel 3 Pel 4 Pel 20

Skull width at orbits . : _- 28:2 = —_
Width between ey

foramina . : : 13°4 15'0 —— -—
Occipital height . : ; 9:9 9°7 13'2 _
Skull width at mastoids : 26:8 273 23'8 26:7
Width across ant. tubs. of

basioccipital . : 4:2 4°55 4:1 5:0
Width across post. anit. of
basioccipital F ; F 78 T7 8-3 8-7

Measurements of width of anterior and posterior tuberosities on other basioccipitals were:
Pel 5, 4:3 and 8-3; Pel 15, 4-1 and 7:25.

Horn core dimensions, taken as indicated above, were:

Dorso-ventral Diameter at 90°
diameter to the last Length
Pel 2 9:2 13°1 —
eli; 11-6 13°8 153*
Pel 9 10°5 14:0 142*
Reler2 10:6 13°6 149
Pel 18 9:6 12:6 102
Pel 5 1074 I4'I —
Pel 6 II‘ 13°3 125
Pel 4 113 13'9 —
Pel 14 I1'7 1453 os
Pel 15 II‘o 15°4 —
Reliz2 THTECHE 14:0 —

* The tips of these horn cores are of plaster.

The span between the tips of Pel 6 horn cores was 184, and of Pel 7 and Pel 8 together 205.

Pel 23 had the following tooth measurements: length M1—M$ 9-95 ; length M? 3-1; width M?

2-7; length P?-P4 6-05 ; length P? 1-9.

The right maxilla 1957 SHK II 232 has length M1—M$ 10-2 ; length M? 3-5 ; width M? 2-7.

The mandible 1952 BK II 117 has an M, 3-9 long x 1:8 wide.

The mandible 1952 BK II 119 has length P,—P, 6-75 ; length P, 1-8; height of M, from its neck

to the valley on the medial side between the first and second lobes 5:9.

1957 BK II no number has an M, 3:7 long.

The unnumbered left mandible with much plaster has an M, measuring 3:6 long x 2:0 wide.

A left mandible fragment 1952 BK II 127 has length M,—M, 11-1 and M, 3-4 long x 2-0 wide.
A list of isolated upper molars follows, for most of which length and breadth are given :

1941 surface of Bed I F.107 3-4 x 2:45; 1952 BK II 124 3-9 long; 1952 BK II 140 3:7 x 2:4;

1952 BK II 141 is probably the same individual as 140; 1952 BK II 132?, 292 and 293 are

heavily worn ; 1953 BK II Extension 57 3:6 x 2:1; 1953 BK II Ext. 90; 1953 BK II Ext. 107

3°3 X 3:1; 1953 BK II Ext. 299 3:65 x 2:4; 1953 BK II Ext. 302 is probably from the same

individual as 299; 1953 BK II Ext. 352 3°55 & 2:1; 1955 BK II 97 3:8 long; 1955 BK II

116 3:7 long; 1955 BK II 224 two left upper molars measuring 3:7 2:7 and 3:3 X 2:6; 1955

BK II 322 3:5 x 2:6; 1957 BK II 604 a fragmentary upper molar; 1957 BK II 880 3:4 x 2:6;

1957 BK II illegible number is probably the same individual as 880; 1957 SHK II 179; no

number 3:6 X 2°5.

258 PELOROVIS OLDOWAYENSIS RECK

Isolated lower molars are: 1952 SHK II 668 fragmentary ; 1953 BK II area C no number M,
or M, 3:3 X 1:9; 1953 BK II Extension 58 M, or M, 3-6 X 1:3; 1957 BK II 974 a much worn
M, or M,; 1957 SHK I1 1129 M, or M,; 1953 BK II Ext. 79M 34:9 x 1:8; 1955 BK II 282 M, ;
1957 SHK II 165 M; 4:5 long. 1952 BK II 129 (Text-fig. 4) is a right M, or M, measuring
3°75 X 1:5; since it is scarcely worn its height may be measured as was that of M, on the
mandible 1952 BK II 119—the distance is 5-3. The left M, or M, 1953 BK II Extension 355
is 3°6 long.

Many isolated upper premolars are assigned to Pelorovis: 1952 BK II 134, 142, 143, 144, 145
and 146; 1953 BK II Extension 86, 300 and 301 (probably from the same individual as molar
299), and 361; 1957 BK II 653, 692, 977, 1077, 1361? and 1461. Isolated lower premolars are
1953 BK II Extension 87; 1955 BK II 143 and 144.

Finally there are numbers of large incisors which probably belong to Pelorovis,
and an unnumbered fragment of a mandible with a premolar which is not sufficiently
brachyodont for a giraffid, and which is small in comparison with the size of the
mandible.

Limb bones

There is from site BK II an almost complete limb bone skeleton belonging to the
same individual as the complete skull of Pelorovis ; this skeleton can be used as a
basis for description of the characters and proportions, while other fragmentary
limb bones sometimes supplement this information. The largest Bovid limb bones
from BK II and other Bed II sites are taken to be from Pelorovis because they
showed no substantial morphological differences from the associated Pelorovis
skeleton. The characters of the limb bones have already been listed in the definition
of P. oldowayensis, and there follow here points of lesser interest. Occasional com-
parisons are made with the African buffalo Syncerus caffer because this is the largest
living Bovid in Africa and because a skeleton of it was available in Nairobi.

Femur

The paired Pelorovis femora, 1952, BK II, 267 and 268 (Pl. 4; Text-figs. 12-14)
from the associated skeleton are very slightly longer than the buffalo, and are at
least as thin. Other femoral fossils are: 1952, BK II, 187—a right proximal end,
and 1953, BK II Extension, no number—a right proximal end which is slightly
smaller than those from the complete skeleton.

Tibia
The tibiae of Pelorovis are longer and proportionately more slender than in the
buffalo. Tibial remains assigned to Pelorovis are:

1952, BK II, 269, right complete, from the associated skeleton
(Pl. 4)

1952, BK II, 270, left complete, from the associated skeleton

1952, BK II, 189, left distal end

1952, BK II, 190, right distal end

1952, BK II, 881, right distal end

1953, BK II Extension, 179, right distal end

1953, BK II Extension, 421, left distal end and part of shaft

PELOROVIS OLDOWAYENSIS RECK 259

Calcaneum

Only two calcanea, 1952, BK II, 195 (right) and an unnumbered left one, are
referable to Pelorovis. They are longer bones than in the buffalo, as is shown in
measurements. The height of the top of the astragalus facet above the base of the
bone was 5:7 cm. in both the fossil 195 and the Recent buffalo specimen, but the
length of the buffalo calcaneum was 11-1 and of the fossil 11-9 cm.

Astragalus

The astragali of Pelorovis are slightly taller and less squat than in the buffalo.
Assigned specimens are:

1952, BK II, ror, left

1952, BK II, 196, right ; this is the largest one.
1957, BK II, 625, right

1957, BK II, 1413, left (Text-figs. 16-18)

1953, BK II Extension, 324, left

1953, BK II Extension, 325, left

1957, SHK II, 1164, right

1959, KK II, 211, right

An unnumbered astragalus is about the same size as these but is more squat.
It is illustrated in Text-fig. r9. I think thzt its proportions are too different for
it to belong to P. oldowayensis. Measurements (in cm.) of the length and breadth
of the astragali in the above list are: 191, 8-5 xX 5:6; 196, 9:2 x 5:8; 625,
eye 5-0, L413, 7-9 X 5°25 324, 83. X 5:6; 211, 7°4 X 5:1 cem. The mean of the
ratios of breadth as a percentage of length is 66-2, while the squat astragalus measures
7-9 X 6:2 and has a ratio of 78-5. Two astragali numbered M.12802 are in the
Felix Oswald Collection at the British Museum (Natural History) ; they come from
Homa Mountain and agree with Pelorovis in size and proportions. One would need
other bones for a positive identification.

Naviculo-cuboid
Naviculo-cuboids belonging to Pelorovis are :

1952, BK II, 192, right

1952, BK II, 209, left

1955, BK II, 282, left

1952, BK II, no number fragmentary right

No numbers two left (Text-figs. 22, 23)

The smaller of the two unnumbered naviculo-cuboids (Text-fig. 23) has a nearly
vertical back edge of the medial wall.

Metatarsal

The metatarsal of Pelorovis is considerably longer than in the buffalo and propor-
tionately more slender. A number of distal ends exist and can be identified as left
GEOL. I4, 7. 26

260 PELOROVIS OLDOWAYENSIS RECK

or right by the slight asymmetry of the condyles—the medial side is a little smaller
than the lateral side and its lowest point a little higher than the corresponding point
on the lateral half. Also the lateral tendon insertion at the distal end is usually
noticeably deeper than the medial one. Bones assigned to Pelorovis are :

1952, BK II, 275, left complete, from the associated skeleton
(Pl. 5; Text-figs. 24, 26)
1952, BK II, 197, right proximal end
1953, BK II Extension, 415, right proximal end
1957, BK II, 630, left proximal end
1952, BK II, 220, left distal end
1953, BK II Extension, 316, right distal end
1953, BK II Extension, 413, right distal end
1953, BK II Extension, 414, left distal end
MRC IT, no number, right distal end
Scapula

The two bones assigned to Pelorovis are :

1952, BK II, 183, a left stem (Text-figs. 27, 28)
1953, BK II Extension, 122, another left stem

122 is larger and has a more pronounced muscle insertion at the back of the medial
side of the stem, a character which is seen in some large Bovinae.

Humerus

Humerus specimens assigned to Pelorovis are :

1952, BK II, 271, left complete, from the associated skeleton
(Pl. 4)

1952, BK II, 272, right complete, from the associated skeleton

1952, BK II, 184A, right proximal end

1952, BK II, 185, right proximal end

1957, BK II, 1847, left articular head

1955, BK II, no number, left shaft

1941, S.2, F.929, left distal end

1952, BK II, 186, right distal end

1952, BK II, 352, left distal end

1953, BK II Extension, 44, right distal end

1953, BK II Extension, 317, right distal end

1953, BK II, no number, right distal end

The proximal end 184A has the front of its articular head more ventrally sited
than in the other specimens, but this is probably not an indicator of specific
difference.

PELOROVIS OLDOWAYENSIS RECK 261

Radius
Radi assigned to Pelorovis are :
1952, BK II, 273, left complete, from the associated skeleton
(Pl. 4; Text-figs. 31, 32)
1952, BK II, 274, right complete, from the associated skeleton
1955, BK II, 294 left, almost complete (Text-figs. 33, 35)
1952, BK II, 353, right proximal end
1953, BK II Extension, 425, right proximal end (Text-fig. 34)
1952, BK II, 188, right distal end
1955, BK II, 242, left distal end
1953, BK II Area C, no number, right distal end, water rolled
SHK II, 25, right distal epiphysis

SHK II, 25, for which the year of excavation is unknown, and a fragment of an
axis vertebra 1957, BK II, 889 are the only known juvenile pieces of Pelorovis.
1955, BK II, 294 is smaller than the other radii and more swollen at its distal end
in side view ; it could be a different species from Pelorovis oldowayensis.

Carpal bones

Only one poorly preserved right unnumbered scaphoid of Pelorovis is available.
It is less tall and slightly longer than in the African buffalo ; such proportions are
seen in all the carpal bones. There are two lunates of Pelorovis, one left and one
right, and both of them unnumbered. A left and a right cuneiform are both
unnumbered. A left and right magnum-trapezoid are both unnumbered; the
better preserved left one is antero-posteriorly longer and also broader than in the
buffalo, the other one is smaller and has proportions more like Syncerus. There
are an unnumbered left unciform and a right one numbered 1957, BK II, 1255.

Metacarpal
Metacarpals belonging to Pelorovis are :

1952, BK II, 277, left complete, from the associated skeleton
(Pl. 5)

1957, BK II, 1037, left proximal and distal ends of the same bone
(Text-fig. 36)

1955, BK II, 295, left, almost complete

1955, BK II, 82, right proximal end

No number, poorly preserved right proximal and distal
ends

1952, BK II, 199, distal end

1952, BK II, 354, distal end

1953, BK II, no number, distal end

1953, BK II Extension, 52, distal end

1953, BK II Extension, 138, distal end

1953, BK II Extension, 170, distal end

1955, BK II, 147, distal end

1957, BK II, 19, distal end

1957, BK II, 863, distal end

1957, BK II, 1188, distal end

1957, FC S, 367 distal end

262 PELOROVIS OLDOWAYENSIS RECK

It is not possible with metacarpal distal ends to distinguish left from right as was
done with the distal ends of metatarsals. The complete metacarpal is longer than
in the buffalo and rather thinner. 1957, BK II, 1037 is a larger bone than the
complete metacarpal, but it cannot belong to the giraffid Sivatherium because the
medial facet is not sufficiently flat and because of the existence of a small ridge
between the medial and lateral facets. 1955, BK II, 295 is rather small and presum-
ably came from the same individual as the radius 1955, BK II, 294.

Phalanges

Five fossil first phalanges are without numbers, one of them being from MRC II,
others are numbered 1952, BK II, 202, 203 and 208, and 1941, S.1, F.837. Five
second phalanges are without numbers, one of them again being from MRC II, the
others are 1952, BK II, 201 and 204, 1955, BK II, 231 and 1941, S.1, F. 326. 10952,
BK II, 239 and 321 are smaller bones but probably within the range of variation of
the species. The length of combined first and second phalanges was greater in
Pelorovis than in the buffalo ; for short (front) phalanges Pelorovis measured I1-6 cm.
and two buffaloes 11-0 and 11-1, while for longer (back) ones Pelorovis measured 12-0
against 11-5 and 11-6 for the buffaloes.

There are six third phalanges without excavation numbers, plus 1952, BK II,
200 and 205, 1953, BK II Extension, 402, and a somewhat smaller bone 1952,
BK II, 238 evidently matching the second phalanx 239.

Measurements of limb bones

Measurements of length (in cm.) were taken on the long limb bones of the associated
skeleton of Pelorovis :

Length of femur from the lateral end of the articular head to the ventralmost level of

the medial condyle . : 43°3
Length of tibia from the ventralmost pone of the top nat facet to the tip of the one
behind the medial malleolus. 43°1

Length of metatarsal from the highest mete of the pare iyi the wishes aes of the
ectocuneiform facet to the articular surface on the medial side of the most projecting

part of the medial condyle distally . 27°5
Length of humerus from the top of the lateral aiikenosiey to the “eaten pata of the

medial side distally . ; 35°1
Length of radius from the anette ios? note Of the medal fees to the most eis point

of the ridge bounding the scaphoid facet medially 0 : 355

Length of metacarpal from the anterior edge of the articular facet at ine exaiwnsert came
radialis insertion to the articular surface on the median side of the most projecting
part of the medial condyle distally . : : : 5 é : : : 23°8

Least thicknesses of these bones were: femur, 5°23 ; tibia, 5°72 ; metatarsal, 3:90; humerus,

5°01; radius 5°44; metacarpal, 4°88. The left radius 1955, BK II, 294 was c.32°7 long and
had a least thickness of 4°81.

Vertebrae

A great many vertebrae belonging to large Bovidae are known from BK II.

PELOROVIS OLDOWAYENSIS RECK 263

Atlas

The best preserved atlas (Text-fig. 6) belongs to a complete row of cervical vertebrae
which may have belonged to the same individual as the skull to be described on
p. 291. It is a wide and very short bone with large transverse processes which do
not extend very far posteriorly. The sides of the bone are straight or very slightly
convex.

6 7

Fics. 6-8. Three vertebrae from a complete set of cervicals. 6, atlas in ventral view ;
7, axis, in lateral view ; 8, fifth cervical in anterior view. a = vertebrarterial foramen.

A second and less complete atlas is similar to the first but more indented at the
front of the dorsal side.

A third atlas has its transverse processes projecting less backwards, and more
prominent ridges behind the ventral hollows, which may be correlated characters.
Its front articular facets are less wide than in the first two atlases mentioned, and its
dorsal side slightly more hollowed.

A fourth atlas has still more pronounced ridges at the back of the ventral surface
than in the last one, and its front articular facets are as narrow as in that specimen.
The dorsal hollowing is intermediate between that of the first two and the third
atlases mentioned above. The indent at the front of the dorsal surface is as deep
as in the second atlas, and the openings for the vertebrarterial and alar foramina
are well separated in an elongated common fossa. It is conceivable that this atlas
belongs to a different species from the others.

The articular facets of the first and second atlases fit the skull to be described on
page 291 better than they do the complete skull of Pelorovis oldowayensis on account
of the former’s smaller occipital condyles. The first atlas fits Pel 5 and Pel 21, but
does not fit Pel 2, Pel 4 or Pel 24 any better than the complete skull. If the first
three atlases are conspecific with the skull remains of P. oldowayensis, the problem
exists of why the largest one fits only two out of the six available occipital condyles,
and not even these two very well.

264 PELOROVIS OLDOWAYENSIS RECK

Axis

The most complete axis (Text-fig. 7) belongs to the same set of cervical vertebrae
as the first atlas described above, and has preserved its transverse processes. A
larger fossil axis has lost its transverse processes and most of its neural spine. The
base of the neural spine is longer and the postzygapophyses more slanted in posterior
view than in the first axis. This second axis is too large to articulate with any of
the atlases but seems to belong with a second row of very large cervical vertebrae.
A ventral fragment of an axis vertebra is numbered 1957 BK II 889, and by the
condition of the rear of its centrum is from a young animal.

The position so far is that we have one atlas and one axis from the same complete
set of cervicals, possibly belonging to a skull to be described later, and an axis from
a larger set of cervicals complete except for the atlas and fifth cervical. There are
three other atlases, one of which may come from another species, and an axis frag-
ment, possibly juvenile.

Cervical vertebrae

The third cervical vertebra from the complete set is short. It has a tall neural
spine and a deep posterior indentation in the neural arch—this is correlated with the
shortness of the bone which brings the neural spines of adjacent vertebrae close
together. The third cervical from the larger and less complete set is a longer bone
and has a proportionately less deep indent posteriorly.

Measurements of the lengths of the two cervical rows were taken while they were
supported in approximately their natural positions in a sandbath. The length of
the complete set was 46-1 cm. from the mid-ventral point of the front of the atlas
to the mid-ventral point of the back of the centrum of the 7th cervical. The length
of the less complete row from the mid-ventral point of the articular rim behind the
odontoid facet of the axis to the 7th cervical, allowing for the missing 5th, is 44-6 cm.
So far as could be seen from mounted specimens in the British Museum (Natural
History), these lengths are about the same as in Bovini.

Two other cervicals from BK II probably belong to Pelorovis, so too an unnumbered
7th cervical from HWK II.

Thoracic vertebrae

A well preserved but unnumbered series of thoracic vertebrae runs from about
number four to number eleven. There are also parts of eighteen other thoracic
vertebrae, fifteen unnumbered and the other three a series—1953, BK II Extension,
25,26 and 27. 25-27 and an unnumbered pair are transitional to lumbars, i.e. from
the back of the thoracic series. Two fragmentary first thoracics are recognizable
by their very widely spaced prezygapophyses, and one of them may articulate with
the seventh of the most completely preserved set of cervicals. The prezygapophyses
on the first thoracic vertebra are concave. On this vertebra the foramen behind the
transverse process is not closed posteriorly, but in two other anterior thoracics it is.
In the most complete and best preserved set of thoracic vertebrae this foramen
becomes progressively larger in the more posterior parts of the row, as is the general

PELOROVIS OLDOWAYENSIS RECK 205

rule in Bovidae. There are fairly deep median indentations between the prezy-
gapophyses. The postzygapophyses are nearer to the bases of the neural spines
than in smaller Bovidae which is a difference which is better seen in anterior members
of the series. The neural spines of the most complete set are tall, and the length of
the tallest from the mid-dorsal point of the anterior end of the centrum to the top
of the spine is 43cm. This seems to be about usual for Bovini. The best preserved
set of thoracics is not certainly from the same individual as the most complete set
of cervicals.

Lumbar vertebrae

No long series of lumbars are present, but there are remains of six, all without
numbers. The foramina at the side of the neural arches are small, and towards the
end of the lumbar series their posterior edges are no longer closed.

Sacrum and tail

There is a fragment of a sacrum which is unnumbered, and twelve caudal vertebrae
numbered 1952, BK II, 255 to 266.

Ill THE SYSTEMATIC POSITION OF PELOROVIS

It has already been noted that Pelorovis does not belong to the Caprinae. A far
better choice for relationship is the Boselaphine-Bovine stock, but there are also a
number of resemblances to Ovyx Blainville in the tribe Hippotragini. These include
the low wide skull, the horn insertions being in the same plane as the face, the absence
of projecting orbital rims, the long nasals, the anterior position of the tooth rows
and the characters of the teeth. The dorsal view of the Pelorvovis skull (Pl. 1, fig. 2 ;
Text-fig. 3) reminds me of Oryx, and the occlusal views of the molar teeth (Text-fig.
II) are similar. I shall conclude that such resemblances exist because Oryx has
some similarities to the Bovini as a whole and not because Pelorovis belongs to the
Hippotragini. None the less it seems desirable to include the Hippotragini with the
Bovini in a detailed character comparison with Pelorovis. There are two living
Hippotragine genera other than Oryx—Hippotragus Sundevall and Addax Rafinesque;
the latter is similar to Oryx in most skull characters and need not be considered
separately.

In this part of the paper bones of the living species of Oryx and Hippotragus
and the Bovine genera Bubalus H. Smith, Bos Linnaeus, and Syncerus Hodgson will
be compared to see which of them most resembles Pelorovis. The genus Bos is taken
to include Bibos Hodgson (with the Asian species Bos javanicus D’Alton and B.
gaurus H. Smith), Bison H. Smith, and Poéphagus Gray, as well as oxen in the narrow
sense. The word Bubalus will refer only to Bubalus bubalis (Linnaeus) ; the small
anoa, B. depressicornis (H. Smith), will be referred to by its full Latin name or as
the anoa. The word Syncerus will refer only to Syncerus caffer caffer (Sparrman)
and not to the West African bush buffalo, S. c. manus (Boddaert), which will be
specified separately. The English word “ buffalo ”’ will also refer only to S. c. caffer.
Measurements, and the ratios derived from them, will be used where they are helpful ;

266 PELOROVIS OLDOWAYENSIS RECK

the methods of taking measurements have already been given. Skull measurements
were taken on adult male animals, or on animals of unknown sex which could not
be reliably distinguished from males, in the collections of the British Museum (Natural
History). Measurements on Syncerus were confined to animals from East and Central
Africa, these being larger than West African bush buffaloes; measurements on
Hippotragini were separated according to species.

Fossil genera are considered wherever they are a source of relevant information.
Among Hippotragini, fossils of Hippotragus are known from the Tatrot and Pinjor
(Villafranchian) of India, and from South Africa (Cooke 1947). A number of early
species, probably ancestral to later Oryx and Hippotragus, are known from Pikermi
and other Lower Pliocene sites of Europe and Asia. Also of Lower Pliocene age are
the Chinese genera Prosinotvagus and Sinotvagus (Bohlin 1935), which represent a
group not related to later African species. Aeotvagus garussi is a name given by
Dietrich (1950) to fossil Hippotragine teeth from the Laetolil area of East Africa.

Among Bovine fossils are the extinct large and long-horned buffaloes of the African
Upper Pleistocene, recognized by Bate (1951) to be closer to Syncerus than to Bubalus,
and given by her the generic name Homovtoceras.1 I have seen the cast of the
Sudanese H. singae Bate in London; a cranium with partial horn cores, another
horn core and a few limb bones of the North African H. antiquus (Duvernoy) in
Paris; and the skeleton of H. nilssoni (Lonnberg) from Malewa near Naivasha,
Kenya and now in Stockholm. I agree with Miss Bate about the resemblance of
these animals to Syncerus, indeed there is at least a possibility that they might
include the ancestors of the living species which can be regarded as specialized in
its skull proportions and small horn cores.

Simatherium kohllarseni is a name given by Dietrich (1942) to a buffalo-sized
Bovine from the beds in the Laetolil area, which are thought to be slightly older
than Bed I at Olduvai.

Leptobos Riitimeyer is the Villafranchian ox of Europe and India, differing from
the closely related Bos by having less posteriorly inserted horn cores, and by other
characters associated with this. The earliest examples of Bos are contemporary
with Leptobos ; they are the large and long-horned B. acutifrons Lydekker from the
Indian Pinjor, and B. sivalensis (Lydekker) and B. palaeosinensis (Teilhard de
Chardin & Piveteau) which are two bison from the Pinjor and from Nihowan in
China. Later in the Pleistocene comes the great extinct Palaearctic ox Bos primi-
genius Bojanus, with its Indian subspecies B. primigenius namadicus (Falconer)
which had somewhat longer horns with a posterior keel towards their base.

Hemibos Falconer is an extinct Villafranchian genus from India and Palestine
(Pilgrim 1941) related to Bubalus ; it differs from Lepftobos by having keeled horn
cores, longitudinally convex frontals between the horn bases, and a braincase more
bent down on the facial axis.

Proamphibos Pilgrim from the Dhok Pathan and Tatrot (Pliocene and earliest
Pleistocene) of India is smaller than Hemibos, and could be its ancestor. A con-
temporary genus from Europe is Pavabos Arambourg and Piveteau. A cranium of
P. cordieri (Christol) from Montpellier and a skull and cranium of P. boodon (Gervais)

1The name Homoioceras actually dates from 1949 (see Bate 1951).

PELOROVIS OLDOWAYENSIS RECK 267

from Perpignan are available in Paris, together with more fragmentary pieces. The
teeth of these animals are slightly less advanced than in Proamphibos. It is not
necessary to continue the list of fossils ; all that need be said is that there is little
doubt that later Pliocene Europe and India were inhabited by a variety of Boselaphine
or Bovine species, not all of which need have phyletic connections with later Bovini.
Teeth in the palaeontological collections at Uppsala show that similar animals
inhabited later Pliocene China. It is an as yet unanswered question whether the
ancestry of Syncerus, like that of Bubalus and Bos, lies among these animals, or
whether its progenitors were by then already living in Africa, having separated from
the Eurasian stock at an earlier date.

I. It may be noted at the outset that the large size of Pelorovis is a good reason
for thinking it to be Bovine ; Hippotragini are medium to large Bovidae, and although
there are small Bovini such as the anoa, the progress of evolution through Boselaphini
to Bovini has usually been accompanied by increasing size. According to measure-
ments of skull length and length of upper molar row (Text-fig. 9) Pelorovis is large
among Bovini; in orbital width as a measure of size it is easily exceeded by Bos
primigenius ; while its femur length is less than in Homotoceras nilssoni, Bubalus,
or Bos primigenius. The largeness of Bovini hinders comparison with other Bovid

SKULL LENGTH ; ORBITAL WIDTH
+ © t v at a2 at
Hx —4 B =e
-——— x ——— G xX 4
a U ——x —S
-— x —VH4 Y -—— x ————4

454.50 55 60 65 70 20 25 30 35
LENGTH MI! TO M3 é FEMUR LENGTH
a o++4+t cs + CY)
re re | B xx
el G XX) 4x
-—_ xX —_ U x
-— x —H Y x x x
aes ate Lae 4 Ga oes eee
7 8 9 10 I 12 35 40 45 50
= COMPLETE PELOROVIS SKELETON
+ OTHER REMAINS OF PELOROVIS B BOS PRIMIGENIUS
+ 2ND COMPLETE SKULL FROM BK I G BOS GAURUS
e HOMOIOCERAS NILSSONI
’ um ANTIQUUS E BUBALUS
4 us SINGAE Y SYNCERUS

Fic. 9. Four diagrams to show the size of Pelorovis oldowayensis in relation to some Bovini.
Individual readings are shown for African fossils and for the femur lengths of all animals ;
in other cases the mean value and observed range are shown. Measurements in cms.

268 PELOROVIS OLDOWAYENSIS RECK

tribes because I am so often unsure about the extent of allometric influence on the
characters.

2. A comparison of general skull proportions shows that Syncerus has a wide
skull with a shortened facial region ; that Bubalus has a narrower skull with a longer
face ; and that Bos is very wide across the frontals, with oxen having a long face
and bison a shortened face. In Hippotragus the skull is high and narrow and although
Oryx does have a broader and lower skull than Hippotragus, this is not carried so
far as in the larger Bovine genera, for which allometry must be largely the deciding
factor. Pelorovis is closest to Bubalus in these proportions, but if the supposed
allometry is taken into account it could equally be considered to be like Oryx.

The two ratios of length from the premaxilla tip to the front of the orbit as a
percentage of skull length and orbital width as a percentage of skull length give
quantitative indications of skull shape. Readings for the first ratio in Bovini were :

Bos Bos
Syncerus Bubalus gaurus primigenius
Number measured . 18 a 9 4
Mean value 6 : 53°53 599 56°5 63°4
Range. . . 49°7-50°4 57°4-61°9 54°3-58°2 62°4-63°7

The face of Syncerus is shorter than in the two Bos species or Bubalus, although
this effect might be less pronounced in females and the smaller West African bush
buffaloes which have not such large basal bosses to their horn cores. The reading
for Homotoceras nilssont was 52:8—close to that for Syncerus, and for the complete
skull of Pelorovis 61-6—close to Bos primigenius. Hippotragini have values from
59°6 (mean of three Addax) to 63-8 (mean of five Oryx betsa), i.e. their faces are as
long as in Bubalus, Bos primigenius and Pelorovis. However in Hippotragus niger
the component of length of face in this ratio has been increased by the bending of
the braincase on the facialaxis. Hippotragus equinus has a slightly longer braincase ;
its appearance of having a very long braincase is caused partly by the height and
narrowness of its skull and partly by the uprightness of its horn insertions.

Readings for the ratio orbital width as a percentage of skull length were :

Bos Bos
Syncerus Bubalus gaurus primigenius
Number measured . 18 7 9 4.
Mean value : 3 48°7 41:2 52°3 55'1
Range. . . 44°8-53°7 39°0-44°0 48-0-56:0 47°5-64'5

Bos is very wide at the orbits, due to expansion of the frontals, while Bubalus is
rather narrow, and Syncerus is intermediate. The value for the complete skull of
Pelorovis is 39:8, which is narrower than the mean for Bubalus, but in male animals
with bigger horns the orbital width would probably have been greater. In Homoto-
ceras nilssomi the value was c. 36:3, the orbits of this specimen being probably closer
together than in either the H. antiquus or H. singae specimens. Hippotragini have

PELOROVIS OLDOWAYENSIS RECK 269

values from 38-1 (mean of six Hippotragus niger) to 44:2 (mean of three Addax).
Length of face also affects the dimensions of the nasals and the position of the tooth
row, both of which will be considered later.

3. Among living forms horn cores are long in Hippotragini, long in Bubalus,
intermediate or short in most Bos, and short in Syncerus. The African Homotoceras
had very long horns, and so too did the Indian Bos acutifrons which is older than
Pelorovis. This character is obviously not useful in determining the affinity of
Pelorovis.

4. The horn cores are straighter in Oryx, except in O. algazel, than in the other
genera considered here, and Oryx is also unlike Pelorovis. The torsion of the horn
cores of Pelorovis, in so far as it exists, is clockwise from the base upwards in the right
horn core, which is the same as in Leptobos, Bos and Syncerus. The difference
between weakly clockwise and weakly anticlockwise torsion is slight, so even were
Pelorovis different from the Bovines, it would count for little. However the actual
course of the curvature as opposed to the torsion of the horns of Pelorovis, as seen
in Pl. 1, fig. 2, consists first of a backwards swing, then of one to the front, whereas
in Homotoceras the swing is at first forwards then backwards; i.e. the horns of
Homotoceras are on a different arc of a clockwise circle. The curvature of Pelorovis
horns is like that of long-horned species of Bos.

5. At the present time dorso-ventral compression of the horn cores is more
pronounced in Syncerus than in Bos, with Bubalus being intermediate. Measure-

g g y
ORY X x XoXe % KX
PELOROVIS x x 0B
LEPTOBOS FALCONERI x x x
BOS PRIMIGENIUS x x Xx
BOS GAURUS xe ox x XX
HEMIBOS TRIQUETRICORNIS & ACUTICORNIS x XIE XGNEX x x x x
BUBALUS PRX
s a n
SYNCERUS x x x x x
a —i = it ———
40 50 60 70 80 90 100 110

Fic. 10. Histogram of dorso-ventral diameter of horn cores as a percentage of transverse
diameter. g = Oryx gazella, y =O. algazel; the remaining three Ovyx belong to O.
beisa. v =the second complete skull from Olduvai BK II, s = Homoioceras singae,
a = H. antiquus, n = H. nilssont. One specimen of Hemibos had a ratio of 126.8.

270 PELOROVIES OLD OW AWE IN SES) RoE GS

ments are difficult to take, but it does seem (Text-fig. 10) that the horn cores of
Leptobos are less compressed than in Bos, Hemibos than in Bubalus, and Homotioceras
possibly less than in Syncerus. Therefore it is not possible to say that Pelorovis,
with its slight compression, resembles one Bovine group rather than the others.
Oryx has almost rounded horn cores (although there are probably valid differences
among the species for this), but Hippotragus has transverse compression.

6. Keels on the horn cores are absent in Peloyovis, in living and most fossil
Hippotragini, in Bos except for the Indian B. acutifrons and B. primigenius namadicus,
and in Leptobos. They are present in Bubalus, Hemibos and Proampiibos, and in
West African bush buffaloes. Those in larger Syncerus and in Homotoceras antiquus
appear to be a result of the compression of the horn cores, and they are less marked
in the less compressed horn cores of H. milssoni. All that emerges with certainty
from this character is that Pelorvovis is unlike the Bubalus group.

7. Among Bovini the earlier species have more nearly uprightly inserted horn
cores in lateral view, but later in evolution their basal parts come to be more nearly
in the plane of the face. The horn cores of Hippotrvagus are more uprightly inserted
than in any other genus considered here ; however, those of Ovyx are in the plane
of the face as in later Bovini, and this state was acquired earlier in the Ovyx stock
than in Bovini as witnessed by the Lower Pliocene Palaeoryx. This character is
therefore of little use in helping to decide the tribal affinity of Pelorovis, since the
Bovini lie between the two Hippotragine extremes.

8. The horn cores of Bovini arise behind the orbits. In larger Syncerus their
front edges are more anterior than in comparably sized Bubalus. Some Bos species
have very posteriorly-sited horn cores, so much so that they may overhang the
occipital surface of the skull, and for this character Bos most resembles Pelorovis.
Horn cores are inserted behind the orbits in Ovyx but not in Hippotragus, the
difference here being linked with the angle of insertion of the horns.

g. In larger Syncerus the great basal bosses bring the horn cores closer together
than in the larger Bubalus, while Bos has horn cores set far apart on its wide frontals.
Hippotragini have closer insertions than Bovini, and within the Hippotragini
Hippotragus has closer insertions than Ovyx. The question of how widely apart
horn cores are inserted is not always easy to decide since increasing size of horn cores
in evolution may bring their bases close together, and also because backward migra-
tion of the horn cores must sometimes allow them to become more widely inserted ;
this last factor is perhaps responsible for their greater separation in Ovyx than in
Hippotragus. In Pelorovis there is no doubt that the horn cores are set close together.
This may be appreciated by comparing the Indian Bos acutifrons (Lydekker 1878,
pls. 12, 13), which appears to have horn cores of about the same size and with a
similar curvature to Pelorovis, with Pl. 1, fig. 2 and Pl. 2, fig. 3 in this paper. The
contrast between Pelorovis and any species of Bos for this character suggests that
both genera have evolved their posterior horn insertions independently.

10. Both sexes have horns in even the earliest known Hippotragini; in the
Bovini all Recent species except the anoa have horned females, but the fossil genera
Hemubos, Proamplibos, Leptobos and the skull which Pilgrim (1913, 1939) called
Proleptobos all have, or are, hornless females. The fact that I interpret Pelorovis

PELOROVIS OLDOWAYENSIS RECK 271

to have horned females is of no consequence, since it occurs late enough to have
evolved from ancestors with or without horned females.

tz. In Boyini the parieto-frontals’ suture has a forwardly pointed indentation
between the horn bases, although it is usually invisible in adults. In living Hippo-
tragini it is transversely directed without anterior indentation between the horns,
but early Chinese and European Hippotragini bad such an indentation like Bovini.
One might expect some correlation between shape of the suture and antero-posterior
position of the horn cores; however in such earlier Bovine genera as Leptobos,
Hemibos and Proampiibos the suture does lie behind the level of the horn cores,
but is none the less anteriorly indented in the centre of its course across the top of
the skull. This character is not useful in relation to Pelorovis, since its horn cores
are so posterior that the suture would be bound to be indented, even were it visible
in a single specimen. The suture is straight in living Boselaphus Blainville, but this
animal is probably remote from Bovine lineages.

12. Temporal fossae are present behind the orbits in Peloyovis and in Bovini,
but their presence is at least partly correlated with the massiveness of the horn bases,
and thus with the size of the animal, and also with the generally posterior position
of the horn insertions in Bovini. They are absent in Hippotragini.

The character of temporal crests behind the horns ought to be mentioned here.
Their presence or absence is said by Pilgrim (1939 : 148-149) to be correlated with
the degree of bending down of the facial on the basicranial axes. The axes are at
an angle to one another in Hippotragini, and temporal crests are not found.
Originally the axes were little bent on one another in Bovini and their ancestors,
hence temporal crests are found in Proamphibos and Leptobos. However Hemibos
has developed a greater angle between the axes (the longitudinal convexity of the
frontals between the horn bases is probably linked with this), and Bos has moved
its horn cores to a more posterior position, thereby doing away with the need or
possibility of temporal crests. The horns are so large in Bubalus and have such large
bosses in Syncerus that there is no room for temporal crests, but it is not easy to
say whether the facial axis is more bent on the basicranial axis than in Leptobos
and Proamphibos, and whether the convexity of the frontals in the small bush
buffaloes and small examples of Bubalus bubalis are comparable with that in Hemibos.
It is apparent that Pelorovis, having horn insertions as far back as in Bos, could
not have temporal crests.

13. The supraorbital pits are relatively closer together in Hippotragini than in
Bovini. Readings for the distance between the supraorbital pits as a percentage
of orbital width in Bovini were :

Bos Bos
Syncerus Bubalus gaurus primigenius
Number measured. 21 8 Io 4
Mean value : < 51°1 58-1 56-6 57:6
Range . : : 42°2-59'5 55°I-61°9 43°2-72:2 55°2-62°2

The closeness of the supraorbital pits in Syncerus could be interpreted by supposing
that the growth in size of their basal horn bosses, particularly in males, must have
caused an increase in orbital width. Bubalus and Bos have supraorbital pits which

272 PELOROVIS OLDOWAYENSIS RECK

are wider apart than in Syncerus. The values for Homovtoceras singae, H. antiquus
and H. nilssont were 50-4, c. 49°3 and c. 65-2, the last value probably reflecting that
specimen’s narrowness across its orbits. The value for the complete Pelorovis skull
was 42°3 and for Pel 3 53:2. In Hippotragini the values for this ratio extend from
42°3 (a single Oryx leucoryx) to 47-3 (the mean of five O. beisa), thus supraorbital
pits are closer together than in Bovini.

A point to be remembered is that the antero-posterior position of the supraorbital
pits may influence their transverse separation from one another. They are more
anteriorly sited in Syncerus (and some bison) than in Bubalus or Bos (other than
bison), so that they have less space in which to be wide apart. This factor would,
however, have little effect on the difference between Bos (other than bison) and
Syncerus compared with the effects of the wider frontals of Bos. This factor does
not seem to apply to Pelorovis, which has supraorbital pits set more posteriorly than
in Syncerus or Bubalus, but none the less close together.

14. The orbital rims are strongly projecting in Bos (especially in bison species),
less so in Bubalus, while in Syncerus such tubularity occurs only in few individuals.
The orbits usually protrude little in Hippotragini except in Hippotragus equinus.
The small bush buffaloes of West Africa have horn cores which are close to the orbits,
and, being directed diagonally backwards at their bases, prohibit the development
of tubularity in the orbits, but this explanation does not apply to the absence of
tubularity in most larger Syncerus. The orbits of Pelorovis do not project very
strongly, thus making it most like Hippotragini or Syncerus.

15. The nasals of Oryx are flatter across their upper surface than in the other
genera considered here.

16. The nasals are wide in Bos, slightly less so in Syncerus, and narrow in Bubalus.
This character can be assessed quantitatively by taking nasals’ breadth as a percent-
age of nasals’ length, and nasals’ length as a percentage of skulllength. The readings
for the first ratio in the Bovini were :

Bos Bos
Syncerus Bubalus gaurus primigenius
Number measured. 14 5 9 2
Mean value , : 33°2 26:2 30-6 35'5
Range . ; : 25°7-40°1 22:0—30°6 31°3-39'°8 35°2 & c. 35°8

Width of nasals is a variable dimension.
Readings for nasals’ length as a percentage of skull length were :

Bos Bos
Syncerus Bubalus gaurus primigenius
Number measured . 16 5 8 2
Mean value : : 39°8 42°4 43°8 430
Range . : : 34°1-43°8 40°3-45'8 39:9-48:0 42°4 & C. 43°6

Homotoceras singae had a value of c. 37-6 for the first ratio, but H. nilssoni could
not be measured.

Despite its very long and narrow nasals according to the first ratio, Bubalus has
nasals no longer relative to skull length than in the Bos species. The complete skull

PELOROVIS OLDOWAYENSIS RECK 273

of Pelorovis has a value of c. 21-4 for the first ratio and c. 46:2 for the second ; these
could be taken as showing great length of the nasals (presumably linked with the
long face) rather than narrowness, especially as the nasals are wider in their front
half than in their back half (Text-fig. 3). Hippotragine readings for the first ratio
ranged from 18-75 (mean of six Hippotragus niger) through 21-3 (mean of nine
H. equinus) to 30-9 (mean of five Oryx beisa), and for the second from 36-2 (one
O. leucoryx) to 42-5 (mean of three O. gazella). Thus Pelorovis is nearest to Bubalus,
Oryx and possibly Hippotragus equinus for the first ratio, while for the second it is
near the extreme for either tribe.

17. Lateral flanges at the anterior end of the nasals are pronounced in Bubalus,
Bos javanicus and some B. gaurus, and practically absent in other Bos species,
Syncerus and Homotoceras. They are present in Hemibos if the skull 23109 at the
British Museum (Natural History) is indeed of Hemibos, although Pilgrim (1937, fig.
47c) has drawn the genus without them. They are absent in Hippotragini. In the
complete skull of Pelorovis they are clearly absent.

18. A transverse expansion at the back of the nasals is present in Bubalus and
Hemibos and in some Syncerus, and absent in Bos, Hippotragini and Pelorovis.

19. The posterior end of the nasals is in front of or barely level with the front of
the orbits in Hippotragini, but lies slightly more posteriorly in Bovini, except that
some individuals of Bos may overlap Hippotragus individuals. This character is
unlikely to be caused by allometry because in the anoa the back of the nasals is level
with the front half of the orbits and in the large Caprine Ovibos Blainville they start
in front of the orbits. The position of the suture is not certain in Pelorovis, but
it appears to be level with the front half of the orbits.

20. The nasals of Syncerus are often more nearly parallel to the tooth row in
profile than in Bubalus ; Bos appears to be nearer to Bubalus than to Syncerus, and
the Hippotragini are like Bubalus. This is not a useful character in Pelorovis. The
reconstruction of its skull in lateral view (Text-fig. 2) does show the nasals in a
position similar to Syncerus, and the drawing was made before I was aware of the
Syncerus character, but one would need the preservation of an undistorted skull
before accepting this character.

21. The absence of an ethmoidal fissure in the complete skull of Pelorovis makes
it unlike Ovyx and also unlike Hippotragus in which the existing fissure has become
blocked by underlying bone. It is less unlike early Hippotragini in which the fissure
is often small or difficult to see. Among Bovini it is more like Syncerus and Bubalus
than Leptobos or Bos in which the fissure can often be seen at least until the early
part of adult life.

22. The preorbital fossa is absent or only faintly indicated in Bovini, and is
shallow in early Hippotragini (but deep in Szmotragus) and absent in later ones ;
thus its absence in Pelorovis gives no indication of possible tribal relationship.

23. The deep zygomatic bar of Pelorovis is a character seen in the Alcelaphines
Damaliscus and Alcelaphus, and also in the skull of Homotoceras nilsson. I think
that its appearance is connected with the mechanics of supporting the horn cores.

24. The position of the infraorbital foramen is a distinguishing character between
Hippotragini and Bovini. It is found in Oryx relatively high above the tooth row

274 PELOROVIS OLDOWAYENSIS RECK

and relatively posteriorly sited, whereas in Bubalus and Syncerus it is lower and
further forwards, perhaps in front of P?. Bos and Hippotragus may not always be
distinguishable, but their foramina are often near to the positions occupied in their
respective tribes. The foramen of Pelorovis is undoubtedly in a Bovine position.

25. Thesize of the palatal fissures between maxillae and premaxillae varies among
the genera considered here. They are large, chiefly noticeable in their great length,
in Hippotragus, but much smaller in Oryx. In Bovini they are smaller in Bubalus
and Syncerus than in Bos but not as small as in Oryx. In Pelorovis only the front
parts of the fissures are present, but I would judge them to lie between Ovyx and
Syncerus in size.

26. In the Hippotragini the premaxillae make or just avoid making a short
contact with the nasals; this character is less constant within Bovine genera, but
Syncerus appears to be nearest to the Hippotragine condition. In Bos the premaxilla
often falls well short of a contact with the nasals, but in Bubalus it invariably has a
long contact. The premaxilla of Pelorovis is closest to the condition of Syncerus
and Hippotragini.

27. The tooth row is set more anteriorly in some Bos, e.g. B. grunmens (the yak)
and B. primigenius, than in other Bovini, and in Oryx and Hippotragus niger than in
H. equinus. This character can be linked with the more posteriorly sited orbits of
the Bos species, Oryx, and Hippotragus mger, and can be quantitatively expressed
by the length from the premaxilla tip to M? expressed as a percentage of the length
from the premaxilla to the nearest point of the orbit :

Bos Bos
Syncerus Bubalus gaurus primigenius
Numbermeasured . ne) 8 Ke) 3
Mean value j ; 108-8 IOI‘O 108°8 89-1
Rane euae é 9 102°8-114°'8 97°3-103°5 106:3-111°5 88-1—90-1

Readings for Hippotragini pass from 91-1 (mean of five Oryx beisa) to 98-8 (mean
of ten Hippotragus equinus). The reading for the complete skull of Pelorovis is 82:3,
which is most nearly approached by Bos primigenius in the above table. In
Homovoceras nilssont the reading is 97-4 ; this is most interesting because this animal
has as short a face as in Syncerus, as was seen on page 268 above. It is thus less
remote from Pelorovis for this proportion than are the Syncerus in this sample.

This is perhaps a good place to sum up the differences in skull shape revealed by
the last ratio and those under characters 2 and 13 above. Syncerus has a very short
face as shown by the anterior position of its orbits, although this effect is likely to be
exaggerated in males. In Homotoceras nilssoni too the face is short, but the skull
is narrower across the orbits, and the tooth row is situated more anteriorly. Bubalus
has a longer face and a less posteriorly placed tooth row, presumably correlated
features. It is narrow across the orbits but less narrow than in Homotoceras nilssont,
and its supraorbital pits are not so anterior or so close as in Syncerus and the three
Homotoceras species. Bos is characterized by a wide frontals region, and agrees with
Bubalus in having supraorbital pits set widely apart and posteriorly. Some Bos
may have long faces and tooth rows set far forwards as do B. primigenius and B.

PELOROVIS OLDOWAYENSIS RECK 275

grunmens. Pelorovis has a long face, is narrow across the orbits, and has a markedly
anterior tooth row. It differs from Syncerus in all these features, and from Homoto-
ceras nilssont in its long face and to a lesser extent in the position of its tooth row.
It differs from Bubalus in its more anterior tooth row, and from Bos in its narrow
width across the orbits and very slightly in the position of its tooth row. The
Hippotragini have longer faces than some Bovini, are narrow across the orbits, and
their supraorbital pits are more anterior and closer together.

28. The level of the median indentation at the back of the palate is a good
distinguishing character between Hippotragini and Bovini. In Hippotragini it is
level with the lateral ones or slightly anterior or posterior. Bubalus is the Bovine
most remote from the Hippotragini, by having a long backwards extension of the
median part of the palate taking the indentation well posterior to the lateral ones.
This also occurred in Hemibos and Proamphibos. Many individuals of Syncerus
are less remote from the Hippotragine condition, while Bos lies between Syncerus or
Bubalus. In Pelorovis the median indentation is anterior to the lateral ones, more
so than in many Hippotragini, and if Pelovovis is a Bovine this character is no longer
distinctive for the two tribes.

29. The vomer is fused with the back of the palate in Bubalus bubalis, B. depres-
sicorms and Hemibos, but not in Pelorovis or any other genus considered here.
Pilgrim (1939 : 274) states that fusion probably occurred in Proamphibos.

30. The occipital surface of the skull is low and wide in Bos gaurus and Syncerus,
but higher in Bubalus and Bos primigenius ; it is also lower in Oryx than in Hippo-
tvagus especially H. niger. Readings for the height of the occiput as a percentage of
its width were :

Bos Bos
Syncerus Bubalus gaurus primigentius
Number measured . 18 5 9 4
Mean value ¢ . 31°9 37°5 32'0 36°3
Range . : ; 28-8—-36°5 34°4-39°8 29°4-35°3 32°7-40°1

The reading for Homotoceras singae was 31-7, for H. antiquus 32-3, and for H.
nilssont 31-9, which agree very well with Syncerus. The readings for Hippotragini
range from 35-6 (one Oryx leucoryx) to 52:0 (mean of six Hippotragus niger), readings
which are rather higher than in Bovini. Three values for Pelorovis are 40-7, 36-9
and 35:5, making it unlike Syncerus and Homotoceras, but more like Bubalus, Bos
primigenius or Oryx. The difference between the size of horns in male and female
cattle was found by Howard zm Mourant & Zeuner (1963 : 96) to affect this ratio,
male animals having a lower occiput.

31. The mastoid exposure of the periotic is entirely on the occipital surface in
Bovini, including the anoa, and in Oryx, but in Hippotragus it is visible in lateral
view as well. In Pelorovis it is on the occipital surface.

32. The top edge of the foramen magnum is more posterior relative to the occipital
condyles in Oryx than in Hippotragus in which the paraoccipital processes are markedly
anterior relative to the condyles. Pelorovis, Bovine genera and Lower Pliocene
Hippotragini are intermediate for this character.

GEOL. 14, 7. 27

276 PELOROVIS OLDOWAYENSIS RECK

33. The distance across the anterior tuberosities of the basioccipital is a variable
dimension ; it is narrow in Syncerus—perhaps because the main mass of the horn
cores is concentrated in their basal bosses, wider in Bos and Bubalus, and wider still
in Hippotragini. Often in Syncerus there is a transverse narrowing immediately
in front of the posterior tuberosity which gives the bone a more quadrangular
appearance than would be expected from its measurements. Readings for the width
across the anterior tuberosities as a percentage of the width across the posterior
ones were :

Bos Bos
Syncerus Bubalus gaurus primigenius
Numbermeasured . 19 7 9 3
Mean value : : 45°9 53°7 55°2 49°3
Range. . . 36°6-55°9 47°7-59°4 46:2-65°0 46°2-53°4

The reading for Homoioceras singae was 50:0, for H. antiquus 45-0, and for H.
nilssont 54:7; probably these animals had wider tuberosities than living Syncerus.
The three available readings for Pelorovis were 53:8, 57-6 and 58-4—all rather wide
compared with the readings in the above table. But there can be no doubt that for
this character Pelorovis shows Bovine affinities ; the living Hippotragini range from
73°3 (mean of three Oryx gazella) to 107-7 (one O. leucoryx). Early fossil Hippotragini
are scarcely less different, as the four following readings on fossils from Pikermi in
the British Museum (Natural History) show :

M.10832, Palaeoryx woodwardt 80-0
M.11415, M.108309, Protoryx carolinae 71-8 and 72°5
M.10833, Microtragus parvidens 83°7

Hippotragus has anterior tuberosities which have grown large and have expanded
to some extent laterally. In the course of expansion a longitudinal ridge has tended
to build up behind each tuberosity, leaving a deep groove running along the centre
ofthe bone. In Oryx the anterior tuberosities are smaller and more localized, without
the development of longitudinal ridges. In Bovini the anterior tuberosities are not
so localized as in Oryx, nor are they so expanded as in Hippotragus. Pelorovis could
be said to be like Oryx or Bovini in this.

34. The foramina ovales are large in living Hippotragini, especially in Oryx, but
even in Hippotragus they are larger than is usual in Bovini, although there is a good
deal of overlap. They are smaller in some lower Pliocene Hippotragini. The
foramina are sufficiently small in Pelorovis for it to resemble the Bovini (or early
Hippotragini) for this character.

35. An indentation in the squamosal shelf immediately in front of the mastoid
occurs in Oryx, Bovini and Pelorovis, and can be seen in ventral view (PI. 2).

36. The size of the auditory bulla is probably not a good distinguishing character
unless one is dealing with undoubtedly closely related forms, so it is here of doubtful
validity. It is less inflated in Bos and Bubalus than in Syncerus and Hippotragini.
In Pelorovis it is within the range of Syncerus and Hippotragini.

PELOROVIS OLDOWAYENSIS RECK 277

37. The upper molar teeth are wide in Hippotragus and Bovini, but not in Oryx.
Within any one genus the width of the molars is a variable dimension, but during their
evolution the Bovini have undoubtedly widened their molars. The widening is
partly connected with increasing size, but large Bovidae such as Taurotragus Wagner
and Ovibos do not have noticeably wider upper molars than their smaller relatives.
Pelorovis has molars as wide as in Hippotragus and Bovini. This character, and
others of the teeth, are illustrated in Text-fig. 11.

9 ES) BY

ORYX PELOROVIS LEPTOBOS PROAMPHIBOS
39
HIPPOTRAGUS HOMOIOCERAS BOS PRIMIGENIUS HEMIBOS
1M 40
[bo |
NS Wy
PARABOS SYNCERUS BOS GAURUS BUBALUS

Fic. 11. Occlusal views of right M? in some Bovid genera. The anterior side is towards

the right side of the page. The numbers are those by which characters are referred to
in the text.

38. Basal pillars on the molar teeth are small in Ovyx, but large and complicated
in Hippotragus ; they are also large and complicated in Bovini. Within the Bovini
this character is affected by allometry since Syncerus and smaller Bos have smaller
basal pillars than Homovtoceras, Bubalus and larger Bos. However those of Pelorovis
are small despite the large size of its teeth.

39. The medial lobes of the upper molars are less narrowed in Ovyx than in
Hippotragus. Neither Pelorovis nor the Bovine genera being considered here have
lobes as narrow as in Hippotragus.

40. Localized outbowings on the lateral walls of the molars are well marked in
Bubalus and some Bos, but less so in Syncerus and smaller Bos ; this is again likely

278 PELOROVIS OLDOWAYENSIS RECK

to be influenced by allometry. They are better marked in Hippotragus than in
Oryx, as can be seen in Text-fig. 11. They are hardly present in Pelorovis.

41. Syncerus and some Bos have a slightly less complicated course of the enamel
borders of the central cavities of the upper molars than in Homovtoceras, Bubalus
or larger Bos, which is probably allometric although Oryx certainly shows less
complication than the equally sized Hippotragus. Pelorovis agrees with Oryx.

42. The front edge of the coronoid process of the lower jaw is strongly curved to
fit the temporal fossa in Bovini other than Homovoceras nilssont and some Bos such
as B. grunmens and B. primigenius. In Pelorovis (Pl. 6) and Oryx there is little
curvature as in the Homovzoceras and two Bos species, while in Hippotragus the curva-
ture is not much stronger. It may be noted that the lack of strong curvature in
Pelorovis, Homotoceras milssom and some Bos is connected with the anterior position
of the tooth row, in Hippotragus the absence of a temporal fossa has an effect, and
in Ovyx both factors may have acted.

43. The horizontal ramus of the lower jaw is deep in the Hippotragini and in
Pelorovis (PI. 6).

44. The lower molars of Oryx have weaker goat folds than in Hippotvagus. They
are absent in the Bovini and in Pelorovis. Goat folds are transverse expansions of
the most anterior part of the lower molars separate from the main mass of the
anterior lobes.

45. The central cavity of the anterior part of P, is open medially to the outside
of the tooth in Hippotragini, Bubalus and Bos. It is enclosed by a medial wall in
Syncerus (except that in 62.219 at the British Museum (Natural History) it is open
on the left and closed on the right). It is also open in Homovoceras nilssoni, but the
medial wall has only just failed to close. It is closed or almost closed in all Pelorovis
(Pl. 3, fig. 2; Pl. 6, fig. 3) except in the lower jaws (M.15856) from Kanjera. This
character is useful for differentiating African from Eurasian Bovini.

Colbert & Hooijer (1953 : 119-20) found that Bubalus never has a small accessory
column between the hypoconid and talonid as is seen in some Bos gaurus, and that
it has a greater tendency to have the posterior medial valley of older P,s and P,s
forming an isolated fossette.

46. The length of the premolar row may be quantitatively expressed as a per-
centage of the length of the molar row, but the results are not very useful because
only small numbers of animals could be measured. In Pel 23, an old animal, the
ratio for the upper dentition is 60-8. This is low compared with the Bovine range
from 60-9 for one Syncerus to 70-6 which was the mean for two Bos gaurus. Eleven
Syncerus had a mean value of 65:5, Homotoceras milssom had a reading of c. 63°5,
four Bubalus a mean of 68-8, and one Bos primigenius a reading of 64:4. In the
Hippotragini Hippotragus equinus has a very long premolar row with a mean value
of 74:8 and a range from 70-7 to 79:4 for ten specimens, H. niger has a shorter row
with a mean of 67-4 and a range from 65:8 to 69:2 for five specimens, and Oryx species
have readings mostly below 60 except that the mean for five Oryx beisa rises to 60:9.
These differences presumably indicate differences in Hippotragine feeding habits.

In lower jaws two values for Pelorovis at 50-9 and 58-5 have a mean less noticeably
different from Bovini than was the value for the upper jaw. The mean of five

PEEOROVIS OLDOWAYBENSIS RECK 279

Syncerus was 58-7, of two Bubalus 59:0, and seven Bos gaurus 57:5. It is obvious
from discrepancies between upper and lower dentitions that larger samples are
needed to clarify the facts, but one may tentatively conclude that Pelorovis had rather
shorter premolar rows than in living Bovini.

47. For the ratio of length of P? or P, in relation to the length of the total
premolar row, Pelorovis seems to lie at about the centre of both Bovine and Hippo-
tragine ranges. It is not possible to find trustworthy differences without more
specimens to measure.

To sum up the tooth characters, while the teeth of Ovyx are not greatly advanced
other than in their hypsodonty, the course of morphological evolution has some
similarities in Bovini and AHippotragus, i.e. increasing width of the upper molars,
increasing size of the basal pillars, stronger and more localized outbowings of the
lateral walls of the upper molars, and a more complicated outline of the enamel
borders to the central cavities. However Hippotragus differs from Bovini in other
details: basal pillars are often Y-shaped in section and are extremely large for the
size of the teeth, lateral outbowings of the upper molars are smaller, borders of the
central cavities are less complicated, and there are goat folds on its lower molars.
Pelorovis therefore resembles Ovyx in possessing a moderate development of
hypsodonty without a very advanced pattern of the occlusal surfaces. It is also
like Boselaphini and early Bovini in its occlusal pattern, differing from a form like
Parabos only in being more hypsodont and having wider upper molars, so that if
Pelorovis is a Bovine genus, then it is one with primitive teeth. The influence of
body size on tooth characters is great, and it is possible that a Parabos, growing to
a size equivalent to that of Pelorovis, would acquire wider and more hypsodont
molars. It is also true that while the living Bovini have more advanced teeth than
Pelorovis, it is the larger species which differ more. Syncerus has less advanced
teeth than Bubalus in several characters mentioned above, but Syncerus is a smaller
animal; the upper molars of the closely related, larger Homotoceras nilssoni
(Lonnberg 1933, pl. 1, fig. 4) are no less advanced than those of Bubalus. Within
Bos too it is the larger animals which have the closest resemblance to Bubalus.

The comparisons of limb bones and vertebrae were based on material in the
Osteology Room at the British Museum (Natural History): five skeletons of
Chillingham and Chartley cattle, three Bos gaurus, one Bos grunmens, one Bubalus
bubalis, three Syncerus caffer, three Hippotragus niger, two H. equinus and two
Oryx beisa. The Bubalus skeleton and two of the three Syncerus were mounted so
that not all characters were accessible. In addition to these specimens, some Bos
primigenius limb bones from Ilford were seen, one Syncerus had previously been
available in Nairobi for direct comparison with the bones of Pelorovis, and a mounted
Syncerus skeleton was seen in Stockholm. The limb bones of the mounted skeleton
of Homoioceras nilssoni were also seen in Stockholm ; characters 48, 51, 55, 62, 63,
69, 70, 77, 78, 82, 83 and 84 were not visible, and I have assumed that the animal
resembles other Bovini in characters 48, 51, 63 and 78.

48. In Hippotragus (but to a lesser extent in Oryx) the great trochanter of the
femur has a horizontal top edge and a tendency to antero-posterior lengthening,

280 PELOROVIS OLDOWAYENSIS RECK

which can be seen in lateral view. In Bovini the trochanter appears taller, and is
without a tendency to antero-posterior lengthening ; it either has a slanted antero-
dorsal edge descending to an anterior point or its whole front edge is evenly rounded.
Pelorovis resembles Bovini (see Text-fig. 13).

5 CMS

Fics. 12-15. 12, dorsal view of top of left femur 1952, BK II, 268 with anterior side
towards top of the page; 13, lateral view of great trochanter of same femur ; 14, lateral
view of distal end of same femur; 15, lateral view of great trochanter of a left femur of
Hippotvagus. a = high back of great trochanter, a—b = its slanted antero-dorsal edge,
c = gluteus accessorius crest, d = vastus lateralis crest ; all these points are mentioned
in character 48. e—f = wide lateral parts of articular head mentioned in character 50,
g = front edge of lateral condyle mentioned in character 51.

In Hippotragini the ridge for the insertion of the gluteus accessorius is often at
an angle to the lower part of the crest for the vastus lateralis which may continue
horizontally behind it. In Bovini the gluteus crest often appears to continue the
line of the vastus crest, and Pelorovis resembles them (Text-fig. 13). The condition
in the Hippotragini is linked with the antero-posterior lengthening of the trochanter
mentioned in the last paragraph.

49. A deeper hollowing is present in anterior view between the articular head and
the great trochanter in Hippotragini, Syncerus, the one Bubalus and probably
Homotoceras mlssom than in Bos. Linked with this character is the fact that Bos
tends to have a steeper slope on the top edge of the articular head in anterior view

PELOROVIS OLDOWAYENSIS RECK 281

than the other genera, although this feature is not distinguishable in every bone.
Pelorovis is unlike Bos, and like the other genera for this character (PI. 4).

50. The articular head of the femur is less narrowed in its lateral parts in dorsal
view in Hippotragini than in Bos and the single Bubalus. Syncerus and Homoio-
ceras nilssont have a condition intermediate between Hippotragini and the other
Bovini, and Pelorovis agrees with Syncerus (Text-fig. 12).

51. The distal lateral condyle of the femur is more sharply pointed anteriorly
in Oryx than in Hippotragus, Bovini or Pelorovis (Text-fig. 14).

52. The patellar fossa on the femur is wider in Hippotragini than in Bovini ;
within the Bovini some individuals of Bos may have it still narrower. It is wide
in Pelorovis as in Hippotragini (PI. 4).

Mm
a 2 CMS
\
16 wa 18 19
i
n. fe}
\
DZ S—,
y, Li
20 re) 22 ae

Fics. 16-23. 16-18, lateral, medial and anterior views of left astragalus 1957, BK II, 1413 ;
Ig, anterior view of a fossil astragalus with different proportions from those of P.
oldowayensis astragali; 20, medial view of astragalus of Bos gawrus ; 21, medial view of
astragalus of Syncevus. 22, medial view of a naviculo-cuboid of P. oldowayensis ;
23, medial view of a naviculo-cuboid rather smaller than others belonging to P. oldoway-
ensis. % = incision for naviculo-cuboid as mentioned in character 56, 7 = ridge for
astragalo-metatarsal ligament mentioned in character 57, k = ridge for medial malleolus
of tibia mentioned in character 58, m = position of projection mentioned in character 59.
n = back edge of the naviculo-cuboid, 0 = localized projection at its base, h = top edge
separate from back edge ; all these points are mentioned in character 60.

282 PELOROVIS OLDOWAYENSIS RECK

53. A middle patellar groove is present at the top of the cnemial crest of the
tibia in Hippotragini, but less markedly or not at all in Bovini or Pelorovis (Pl. 4).

54. The lateral articular facet at the top of the tibia has an upturned lateral
edge in Oryx but not in the other genera being considered or in Pelorovis (PI. 4).

55. The medial malleolus of the tibia is shorter in Syncerus than in the other
genera or Pelorovis (Pl. 4).

56. There is a deep incision for the naviculo-cuboid at the back edge of the
medial side of the astragalus in Bovini and Pelorovis but not in the Hippotragini
(Text-fig. 17).

57. The ridge for the astragalo-metatarsal ligament on the medial side of the
astragalus is weaker in Syncerus than in the other genera or Pelorovis (Text-fig. 17).

58. This ridge is level with the ridge for the base of the medial malleolus of the
tibia in Hippotragini. In Syncerus and the Bubalus individual it may be lower, but
in Bos the ridge for the malleolus is also low, and so the two ridges are at the same
level. Pelorovis is most like Hippotragini, Syncerus and the Bubalus (Text-fig. 17).
This character must be linked with 55 above, and one can say that the tibial-
astragalus articulation of Syncerus differs from that of Pelorovis by having a
shorter medial malleolus of the tibia, and Bos differs from Pelorovis by having a lower
astragalar facet for the tibia’s medial malleolus.

59. The back part of the top of the medial side of the astragalus projects slightly
away from the main mass of the bone in Oryx in anterior view. This does not occur
in any other genus or in Pelorovis (Text-fig. 18). The astragalus of Homotoceras
nilssont has a backwardly-directed extension in this position.

60. The naviculo-cuboid is a relatively deeper bone in Hippotragini than in
Bovini or Pelorovis. This is most noticeable in lateral view.

61. The back edge of the medial side of the naviculo-cuboid is slightly or much
curved in Hippotragini; it tends to be straighter in Bovini and sometimes has a
localized backwards projection at its lower end, and a top edge separate from its
back edge. Pelorovis agrees with the Bovini (Text-figs. 22, 23).

62. Both front and back metapodials of Syncerus and the metacarpals of Bos
primigenius are more antero-posteriorly compressed than in the other genera con-
sidered here or Pelorovis.

63. The back part of the naviculo-cuboid facet on the metatarsal rises high above
the level of the immediately adjacent ectocuneiform facet in Hippotragini, and
therefore has a strongly curved outline in medial view. This is not seen in Bovini or
Pelorovis (Text-figs. 24, 25).

64. In medial view the ectocuneiform facet of Hippotragini is upcurved ; this is
not seen in Bovini or Pelorovis (Text-figs. 24, 25).

65. In Hippotragus the foramen at the top of the posterior surface is situated
on the medial side of the midline. This is less obvious in Oryx while in Bovini and
Pelorovis the foramen is situated centrally (Text-fig. 26).

66. The distal anterior and posterior foramina of the metapodials are absent in
Syncerus, or disappear earlier in adult life than in the other genera and Pelorovis
(BLS);

67. The tuber scapulae is low and strongly projecting in Hippotragini, i.e. its

PELOROVIS OLDOWAYENSIS RECK 283

Fics. 24-30. 24, medial view of top of left metatarsal 1952, BK II, 275; 25, medial view
of top of left metatarsal of Hippotvagus ; 26, posterior view of top of left metatarsal 1952,
BK II, 275; 27, lateral view of left scapula stem 1952, BK II, 183; 28, ventral view of
glenoid facet of 1952, BK II, 183; 209, lateral view of left scapula stem of Bos gaurus ;
30, lateral view of proximal end of left humerus of Hippotvagus. » = back of naviculo-
cuboid facet mentioned in character 63, q = edge of ectocuneiform facet mentioned in
character 64, y = foramen at top of posterior side of metatarsal mentioned in character 65.
¢t = tuber scapulae, s = height of base of tuber scapulae above rim of glenoid facet
mentioned in character 67, w = indent in lateral side of glenoid facet mentioned in
character 68, v = base of spine of scapula, w = front of teres minor hollow mentioned in
character 69. %# = posterior eminence of lateral tuberosity of humerus mentioned in
character 71, y = infraspinatus insertion mentioned in character 72.

ventral edge in lateral view rises from the rim of the glenoid facet and passes far
in front of the facet. The tuber is higher in Bovini, but generally projects as far
or almost as far as in the Hippotragini. In Bos gaurus it has an excavation on its
antero-lateral side. Peloyovis most resembles Bovini other than B. gaurus. (Text-
figs. 27, 20).

68. The lateral side of the glenoid facet on the scapula is indented in Bovini and
Pelorovis, but not in Hippotragini (Text-fig. 28).

69. In Bovini there is a tendency for the teres minor cavity on the scapula not
to pass so far forwards as the base of the spine ; this is less clear in Syncerus than in
Bos and the single Bubalus, although it is not possible to tell Syncerus from every
specimen of Bos. Pelorovis is like the Hippotragini or Syncerus (Text-figs. 27, 20).

284 PELOROVIS OLDOWAYENSIS RECK

70. The lateral tuberosity of the humerus is low in Hippotragini, i.e. it does not
rise far above the bicipital groove in anterior view. It is higher in Bovini, except
that two out of the five available Syncerus were intermediate between other Bovini
and Hippotragini. In Pelorovis the tuberosity is low (PI. 4).

71. Syncerus and Homoioceras nilssoni have a smaller posterior eminence behind
the lateral tuberosity of the humerus than in Bos or the Bubalus individual. In the
Hippotragini the eminence is perhaps still smaller but is antero-posteriorly lengthened
and sometimes approaches having a posteriorly directed point. Pelorovis resembles
Syncerus (Pl. 4; Text-fig. 30).

72. The front of the infraspinatus insertion is level with the front edge of the
humerus in lateral view in Bos, the Bubalus individual and Homotoceras nilsson,
but lies slightly behind the front edge in Syncerus and Hippotragini. Pelorovis
agrees with Bos, Bubalus and the Homotoceras. (Pl. 4; Text-fig. 30).

73. The bicipital groove at the top of the humerus is more frequently narrowed
in anterior view in Bos than in the other genera or Pelorovis (Pl. 4).

74. The coronoid fossa at the distal end of the humerus is more shallow in
Hippotragini than in Bovini or Pelorovis (Pl. 4).

75. The medial condyle at the distal end of the humerus is tall in Hippotragus
and perhaps in Homotoceras nilssont, but lower in other Bovini and Oryx. Pelorovis
is like Hippotragus (Pl. 4).

76. The medial side of the medial articular facet at the top of the radius projects
as a rim in anterior view in Bos and Homoztoceras uilssom but not in the other genera
or Pelorovis (Pl. 4).

77. The postero-medial part ot the medial facet has a greater area in Bos than
in Hippotragini. Two out of four Syncerus, the one Bubalus and Pelorovis are like
Hippotragini.

78. The lateral facet at the top of the radius is antero-posteriorly long in Oryx
and some Hippotragus, but often short in the other genera being considered.
Pelorovis (Text-fig. 34) agrees with Hippotragini, but is not convincingly different
trom Bovini.

79. The lateral tubercle at the top of the radius is larger in Oryx than in Hippo-
tvagus, and intermediate in the Bovini and Pelorovis (Text-fig. 34).

80. The distal ends of radii of Bovini and Hippotragus appear swollen in lateral
view, but this is not so for Pelorovis except in specimen 1955, BK II, 294 which is
smaller than the other fossils and may therefore be a separate species (Pl. 4;
Text-fig. 33).

81. The ridge between the posterior surfaces of the scaphoid and lunate facets
at the distal end of the radius tends to be more slanted or less marked in Syncerus,
the single Bubalus and Homoioceras nilssoni than in Bos or Hippotragini. Pelorovis
is like the former group (Text-fig. 32).

82. The tubercle towards the front of the dorsal facet of the scaphoid is more
sharply marked in Bos than in other genera or Pelorovis.

83. The bottom edge of the medial side of the scaphoid is less indented in
Syncerus and Hippotragus than in other genera or Pelorovis. Correlated with this

PELOROVIS OLDOWAYENSIS RECK

Fics. 31-36. 31, medial view of left olecranon 1952, BK II, 273 with its front edge towards
top of page ; 32, posterior view of distal end of same radius ; 33, lateral view of distal end
of left radius 1955, BK II, 294; 34, dorsal view of top articular surface of right radius
1953, BK II Extension, 425 with the anterior edge towards bottom of page ; 35, dorsal
view of left radius 1955, BK II, 294; 36, dorsal view of top articular surface of left
metacarpal 1957, BK II, 1037 with anterior edge towards top of page. a = high origin
of flexor carpi ulnaris muscle which is a peculiarity of Pelovovis, b = lateral facet men-
tioned in character 78, c = lateral tubercle mentioned in character 79, e = scaphoid facet
and d = ridge between scaphoid and lunate facets mentioned in character 81, g =
swollen distal end of radius mentioned in character 80, f = edge of unciform facet of
metacarpal mentioned in character 85.

is the absence of a projection towards the posterior end of the upper surface of the
magnum-trapezoid.

84. The projecting back edge of the lunate is narrowly pointed in Hippotragus
in medial view, but not in the other genera or Pelorovis.

85. The unciform facet at the top of the metacarpal is angled in Hippotragus
but not in Pelorovis, Oryx or Bovini (Text-fig. 36).

Among the limb bone characters of Pelorovis were three which might be valid as
specific or generic characters, but which were not seen in Bovini or Hippotragini.
These were the small sustentaculum on the calcaneum, the antero-posteriorly long
infraspinatus insertion on the humerus (PI. 4), and the flexor carpi ulnaris origin on
the medial side of the olecranon which extended almost to the top of the bone
(Text-fig. 31).

86. Bovini have wider cervical vertebrae than Hippotragini. This may be
owing to allometry at least in part, but the long cervical vertebrae of a large form

286 PELOROVIS OLDOWAYENSIS RECK

like Taurotragus show that allometry need not be the only determining factor.
Pelorovis agrees with Bovini.

87. Bovini and Pelorovis have deep ventral hollows on the atlas for the para-
occipital processes.

88. The vertebrarterial foramina on the axis (Text-fig. 7) are smaller in Bubalus,
Syncerus, Homotoceras nilsson and Hippotragini than in some Bos. Pelorovis agrees
with the former group.

89. The openings of the foramina transversa are nearer to the front and back
edges of the side of the centrum on the third cervical in Bovini and Pelorovis than
in Hippotragini. This difference is not visible on the more posterior cervical verte-
brae.

go. The transverse processes are separate from their ventral flanges as far
forwards as on the third cervical in Bovini and Pelovovis. In Hippotragini the
separation occurs only in the more posterior cervical vertebrae.

gt. The neural spines of the 4th, 5th and 6th cervical vertebrae are not
so forwardly slanted in Bovini or Pelorovis as in Hippotragini.

The results of comparing length measurements of Pelorovis limb bones with those
of two Syncerus caffer and three Bos gaurus are shown in Text-fig. 37. These
Bovines were chosen for comparison because they were available and were of a
similar massiveness to Pelorovis. One mounted Bubalus could only be measured
for its femur (Text-fig. 9), humerus and radius, and has not been shown. It is
seen that Pelorovis has a relatively long radius and short humerus, and also that
the metapodials of Syncerus are noticeably short. Measurements of least thickness

METATARSAL METACARPAL

G x x x x G

i x % Y

B x x

ee el |

15 20 25 30 15 20 25 30

DIUS

G i TIBIA RADIU f 2

if x x x x Y

P x x

—————— eee eee eee

30 35 40 45 25 30 35 40

FEMUR HUMERUS

G xx xX x G

Y x x x x Y

P x x P

| | ee

35 40 45 50 30 35 40 45
G= BOS GAURUS Y=SYNCERUS = PELOROVIS

Fic. 37. Lengths of long limb bones. Measurements in cm.

PELOROVIS OLDOWAYENSIS RECK 287

of each limb bone in relation to its length failed to show reliable differences between
these genera.

Colbert & Hooijer (1953 : 120) found fossil Bubalus bubalis to have shorter meta-
podials than Bos gaurus.

It is possible to assess numerically the differences of the other genera from
Pelorovis in the gt characters listed above. Character 2 was taken as the two charac-
ters face length and skull width ; character 16 was taken as the relative width of
the nasals irrespective of skull length; characters 63 and 64 were taken as one ;
three characters were added to the total to represent the possibly valid limb bone
characters of Pelorovis which were not seen in other forms; characters 10, 22, 47
and 79 were omitted as not being sufficiently different among the forms compared ;
and characters 11 and 20 were omitted as being not applicable to Pelorovis. This
procedure left 43 skull characters and 45 limb bone and vertebrae characters for
comparison, a combined total of 88. For Bubalus characters 62 and 82 were omitted,
and this genus was assumed to be like other Bovini in characters 63, 78 and 85.
Characters 82, 83 and 84 were omitted for Bos primigenius. Characters 16, 19, 25,
26, 28, 29, 36, 55, 62, 69, 70, 77, 82, 83 and 84 were omitted for Homotioceras nilssont,
it was assumed to resemble other Bovini in characters 48, 51, 63 and 78, and it was
not checked for the three unique limb bone characters of Pelorovis.

In the following table is listed the number of differences which each form shows
from Pelorovis ; the figures in brackets show the number of characters used in each
comparison ; and the last column shows the number of differences which would be
expected if 100 characters had been used in the comparisons, i.e. it is the readings of
the third column converted to percentages of the corresponding figures in brackets.

Limb bone All
and characters
Skull vertebrae All out of

characters characters characters 100

Syncerus . , : 19 (43) 13 (45) 32 (88) 36
Bubalus . : d 21 (43) II (43) 32 (86) 37
Bos gaurus c c 24 (43) 20 (45) 44 (88) 50
Bos primigenius : 18 (43) 19 (42) 37 (85) 44
Homoioceras nilsson . 10 (36) 5 (34) 15 (70) 21
Oryx . . . 16 (43) 24 (45) S (88) 45
Hippotragus . . —-26 (43) 25 (45) x (88) 58

The points to be noted from this table are :

I. That the Hippotragine Ovyx appears as so little different from Pelorovis in its
skull characters ; this is principally owing to the tooth similarities of the two genera.

2. That Bubalus appears as hardly more different from Pelorovis than Syncerus ;
yet my opinion is that Bubalus is certainly much further phyletically from Pelorovis
than is Syncerus.

3. That Homotoceras appears as the animal phenetically closest to Pelorovis, and
as considerably closer than the living African buffalo. Even if the three unique
characters of Pelorovis were added to the total for Homotoceras as differences, its
final reading would still be only 25.

288 PELOROVIS OLDOWAYENSIS RECK

Conclusions

From the examination of skull characters it appears that the question of whether
a fossil is assignable to the Hippotragini or Bovini may best be decided by reference
to the following characters: the position and shape of the parieto-frontals suture,
the level of the back of the nasals, whether the ethmoidal fissure is absent or present,
the position of the infraorbital foramen, the level of the median indentation at the
back of the palate, whether the basioccipital is triangular or more nearly quadrangular,
and the size of the foramina ovales. The first of these characters is not applicable
to Pelorovis, but it agrees with the Bovini in the remainder except for the level of
the palatal indentation. In addition its large size and the posterior insertion of the
horn cores tell for Bovine affinities. In most characters which Pelorovis shares with
either of the Hippotragine genera—usually Ovyx, there is also a resemblance to a
Bovine genus; the level of the palatal indent is the only firm resemblance to
Hippotragini.

For the sake of completeness, three further points must be made. Firstly, the
Lower Pliocene Hippotragini of Europe and China show a forwardly-indented
parieto-frontals suture whenever this feature is visible, their ethmoidal fissure may
be absent or very reduced, their foramina ovales are small, and Sinotragus has a
well marked preorbital fossa.

Secondly, keels on the horn cores are present in many Bovini but not in Hippo-
tragini apart from the early Prosinotragus and Sinotragus and possibly some others
in which they may appear in connection with the transverse compression.

Thirdly, the earlier Bovine or Boselaphine genera Selenoportax Pilgrim, Parabos,
Proamphibos, Hemibos and Leptobos all show (or all contain species showing) horn
cores more divergent than in Hippotragini, and if the horn cores have any curvature
in anterior or dorsal view it involves a reapproach of the tips or at least a lessening
of the angle of divergence. Such a curvature is not seen in Hippotragini.

Within the Bovini, the Hemzbos-Bubalus group has (or has retained) keels on its
horn cores, and Hemubos already has the basicranial axis quite strongly angled on
that of the face. The specialized vomer at the back of the palate may be connected
with aquatic habits, and other distinctive characters such as the posterior expansion
of the nasals and the lateral flanges at their anterior ends could also be connected
with this.

Leptobos has preserved a basicranial axis nearly in line with the facial axis, but
the closely related Bos has developed wide frontals and posteriorly inserted horns,
and in so doing has presumably increased the bending of the basicranial axis on the
face. Some Bos species have such posterior horn cores that they overhang the
occipital surface at their bases. Keels are absent on the horn cores, or have been
lost earlier than in Hemubos and Bubalus, on the other hand the ethmoidal fissure
is not always absent in this stock. One learns from Bos that facial length is not a
reliable character for use in classification above the species level, since there are
striking differences of skull proportion between cattle and bison.

Homovoceras and Syncerus have neither wide frontals nor horn cores placed so far
posteriorly as in some Bos species ; it seems possible that they too have evolved
a bending of the basicranial on the facial axis, as in Hemibos and perhaps Bubalus

PELOROVIS OLDOWAYENSIS RECK 289

and Bos. They have a tendency to flattening of the horn cores and consequently
may possess irregular back and front keels. P, has a closed anterior part of its
medial wall.

It is the absence of wide frontals, of keels on the horn cores, of a posterior contact
of the vomer with the palate, and of an ethmoidal fissure, as well as the possession
of an anterior medial wall of P,, which allow Pelovovis to be considered as related to
Syncerus despite the markedly posterior insertion of its horn cores. The closeness
of the supraorbital pits and the extent of the premaxillary contact on the nasal are
also like Syncerus, but are less convincing resemblances. Pelovovis is unlike
Syncerus in the dimensions of its occipital surface, and in the width of the anterior
tuberosities of its basioccipital. The latter character is likely to be directly related
to the size of the horn cores, since Homozoceras and Bubalus both have tuberosities
wider than in living Syncerus with its short horns.

The limb bones and vertebrae of Pelorovis are like Bovini rather than Hippotragini
in the following characters: the shape of the great trochanter at the top of the
femur ; no anteriorly directed point on the lateral condyle at the distal end of the
femur ; no upturned lateral edge to the lateral facet at the top of the tibia; no
middle patellar groove on the tibia; the deep incision at the back of the medial
side of the astragalus ; no projection at the top of the medial wall of the astragalus ;
naviculo-cuboid not deep ; no markedly curved outlines of the naviculo-cuboid and
ectocuneiform facets at the top of the metatarsal; the central position of the top
posterior foramen of the metatarsal; the shape and position of the tuber scapulae ;
the indented lateral edge of the glenoid facet on the scapula ; the shape of the posterior
eminence of the lateral tuberosity on the humerus; the level of the front of the
infraspinatus insertion ; the depth of the coronoid fossa on the humerus ; the size
of the lateral tubercle on the radius ; the lack of a narrow projection at the back of
the lunate ; the outline of the unciform facet at the top of the metacarpal ; the wide
cervical vertebrae ; the deep ventral hollows on the atlas ; the position of the foramina
transversa on the third cervical; the separation of the transverse processes from
their ventral flanges as far forwards as on the third cervical ; and the neural spines
of the posterior cervicals not being very forwardly slanted.

In a number of other characters Pelorovis is like either or both genera of Hippo-
tragini, but is also like or approached by Syncerus: the indentation between the
articular head and great trochanter of the femur in anterior view, the wide lateral
parts of the articular head in dorsal view, the teres minor cavity on the scapula
reaching as far forwards as the base of the spine, the low lateral tuberosity of the
humerus, the wide bicipital groove, the smallness of the posterior eminence of the
lateral tuberosity, the absence of a rim on the medial side of the medial facet of the
radius, the small area of the postero-medial part of the medial facet at the top of
the radius, the more slanted or less marked ridge between scaphoid and lunate facets
on the back of the radius, the poorly marked tubercle at the front of the dorsal facet
of the scaphoid, the little indented bottom edge of the medial side of the scaphoid,
and the small vertebrarterial foramina on the axis vertebra.

Pelorovis differs from Syncerus in not having the short medial malleolus of the
tibia, not such a weak ridge for the astragalo-metatarsal ligament on the medial side

290 PELOROVIS OLDOWAYENSIS RECK

of the astragalus, not such short or such antero-posteriorly compressed metapodials,
and in having better marked foramina on the metapodials. The only resemblances
of Pelorovis to Hippotragine limb bones but not to Bovini are the wide patellar
fossa of the femur, the length of the lateral facet at the top of the radius, and the
absence of a swollen distal end of the radius in lateral view.

From these results I believe that Pelorovis is a genus of Bovini, and since it shows
slightly more phenetic similarity to Syncerus than to other Bovine genera it may
well be phyletically related to this genus. The posteriorly inserted horn cores, long
face, and the primitive occlusal pattern of the teeth make it very unlikely to have
been the ancestor of Syncerus. It may be noted that so far as skull characters go,
Homotoceras is more like Pelorovis than is Syncerus in its larger size, long horn cores
with less emphasis on basal bosses, possibly having less compressed horn cores, the
strengthened front part of the zygomatic bar beneath the orbit in H. nilsson, a
more anteriorly sited tooth row, and the wider anterior tuberosities of the basi-
occipital. Whether Pelorovis separated from the Homotoceras-Syncerus lineage
before or after they became Bovini is not known. I take it to be Bovine rather than
Boselaphine because of its large size, large posteriorly inserted horn cores, and the
fairly hypsodont teeth, thus choosing to regard these characters as outweighing
other more primitive tooth characters. If Pelovovis is accepted as a Bovine, then a
qualification must be added to the use of the level of the median indentation at the
back of the palate as one of the distinguishing characters between Hippotragini and
Bovini.

IV A NOTE ON BULARCHUS AROK HOPWOOD

This species was described by Hopwood (1936 : 639-40) referring to material
collected by the East African Archaeological Expeditions of 1931-32 and 1934-35.
His generic diagnosis reads: “ Bovidae of large size with massive horn cores, com-
pressed from back to front, oval in cross section, closely approximated at their bases,
curving crescent-wise upwards and downwards. So far as is known, the horns are
in the same plane as the face and are not spirally twisted’”’. The specific diagnosis
for B. avok is: “a Bularchus in which the span of the horn cores measures two
metres, or more, from tip to tip”’. Dr. G. E. Pilgrim is quoted as suggesting that
Bularchus may be a very advanced member of the bubaline group in which the
anterior keel on the horn cores has been suppressed. Two of the three specimens,
M.14947—48, were supposed to have come from Bed IV at Olduvai and M.14949
from Bed II, but Mrs. M. D. Leakey has said (personal communication) that the
matrix on the holotype, M.14947, shows that it is more likely to have come from the
very top of Bed II, probably at the site now known as PLK (see also Leakey,
1965 : 106).

The holotype is a frontlet with horn cores ; it is poorly preserved compared with
the finds of Pelorovis dating from 1952 and later years, but there can be little doubt
that it is a specimen of Reck’s P. oldowayensis. The compression of the horn cores
is to be seen only at their bases where there has been some crushing, more distally
the section becomes more nearly rounded as in other Pelorovis. There was possibly
a longitudinal groove along the right horn core at least. So far as can be seen, the

PELOROVIS OLDOWAYENSIS RECK 291

insertion of the horn cores overhung the occipital surface, and the angle of divergence
of their bases was as in other Pelovovis. The ventral surface cannot be seen on the
specimen as at present mounted.

The paratype, M.14948, is a frontlet with part of the left horn core. The core
appears to be compressed, but part of its thickness may be missing. Taking the
diameters of the horn core by the procedures described earlier gave readings of
10-0 X 15:2cm., but an estimate for the complete horn core was c. 10-7 * 15:2 cm.
In the former case the index of compression would be 65-8 and in the latter 70-4—
a value easily possible for P. oldowayensis.

M.14949 is a horn core tip which is certainly not distinguishable from P. oldo-
wayensis. Also in the British Museum (Natural History) is a pair of mandibular
rami from Kanjera (M.15856) referred to Bularchus arok (see Kent 1942). The
outbowings of the medial walls of the molars are not very localized and the ramus
is quite deep under M,. Both these characters are like Pelorovis, but P, is different
from the Olduvai remains by having an open valley on the anterior part of its medial
wall; I do not think that this difference is sufficient by itself to separate M.15856
from P. oldowayensis.

Measurements on the specimen were :

occlusal length Mj—Ms, nie 2 Ciaak.
occlusal length Me, Beis (Cina.
occlusal width Mz, I'9 cm.
occlusal length Pe, Tai7e COO):

All teeth except P, are present on the right side but M, is damaged, on the left side
P, to M, are present but M, is damaged. Also from Kanjera are a left upper molar,
M.25676, measuring 3:35cm. long by 2:35cm. wide, incomplete upper molars,
M.25677 and M.25688, an incomplete M3, M.25692, a lower premolar, M.25678,
and three tooth fragments, M.25679-81 ; all are inseparable from Pelorovis. There
are also some unregistered fragments of Bovine cervical vertebrae with loose centrum
ends. Unfortunately it is not possible to assign these remains, nor those of some
Bovine limb bones from Kanjera, to Pelorovis, because there is definite evidence
from an upper molar fragment, M.25715, of another Bovine at Kanjera with smaller
and more advanced teeth.

In the National Mcseum of Tanzania, Dar es Salaam, is a large Bovid skull with
long horn cores which was found with the herd of Pelorovis in 1952 (Pls. 5, 6). It
shows some differences from the other skulls and had provisionally been taken
to belong to Bularchus (Leakey 1965: 45). Mandibular rami were found with the
skull. It has been distorted complexly but not too severely. The back of the skull
is twisted on the front part, and there has also been some transverse compression
causing the orbits to be slightly too close. In addition the right orbital rim lies more
anteriorly than its own internal parts and the base of the right horn core has obliter-
ated the right temporal fossa. The left temporal fossa and horn core base are missing,
perhaps because the contemporary men broke into the skull for the brain ; on the
other hand the skull was found in the clay of the gully and not on the land surface
where the men have left their tools. The tooth rows are not in their natural positions

GEOL. 14, 7. 28

292 PELOROVIS OLDOWAYENSTIS RECK

and all the upper premolars are missing. Only the left anterior tuberosity is
preserved on the basioccipital.

The skull is from a slightly younger individual than the complete skull of Pelorovis
oldowayensis, providing that the less worn condition of the lower premolars and M,
indicates age and not that the animal’s food plants were less abrasive. It was
probably four or five centimetres shorter than the complete P. oldowayensis skull,
and there are no signs of sutures on the face bones. Measurements (in cm.) on this
skull were :

Skull length . : : : ; : ; h : : : : 62°6
Skull width across the outrts : : z : : : 238
Distance from premaxilla tip to nearest point c on the orbit ; : : ‘ 413
Length of horn core : : : ‘ : : : é ; . 160
Span of horn cores, tip to tip : : 3 ; : 5 : : 20) Bus
Dorso-ventral diameter of horn core : ; ; : : 5 : ; 9°3
Diameter at 90° to preceding diameter : : ; : : : : 153
Width between supraorbital foramina ; 4 : ; : 5 : 1372)
Length from premaxilla tip to M? . ; : ‘ ; ; 36:0
Length from M® to occipital condyle ; : : ; : 26°7
Occipital height ; : : : : , : ‘ . : ; 9°6
Skull width across mastoids . : : : ; ; s 28°2
Width across anterior tuberosities of hasiocerpiral : : ; : : 2 Ghee
Width across posterior tuberosities of basioccipital. : : ; : : 81
Occlusal length M!—M°8 : : : ; , ; : : 10°4
Occlusal length M? 5 : : : ; : : ; 5 36
Occlusal width M? ; : : j : , 2°8
Occlusal length Mi—M3 : ; . : ; ' - : ; ; 11°75
Occlusal length Me . { : ; : : : : : : : 35
Occlusal width Mz . : ; ; i : 3 : ; : : ; I'9
Occlusal length P2-P, : ; : : : k d : , : 69
Occlusal length P2 . ‘ : : ; : 5 : : 18

The main points of difference between this skull and the other skull remains with
which it was found are :

1. Its slightly smaller size. Despite its long horn cores, the skull appears to
be slightly smaller than the complete skull of Pelorovis which I have taken to be
from a female animal.

2. The horn cores are not so posteriorly inserted and do not overhang the occipital
surface of the skull. There is no triangular depression at the top of the occipital
surface.

3. The horn cores are more dorso-ventrally compressed than in P. oldowayensis,
with an index of 60-8 compared with a range of 70-2 to 84:1 in the other skulls
(Text-figs. 10, 40). They are also more nearly in a single plane than most of the
horn cores of P. oldowayensts, 1.e. there is a little less spiralization.

4. The width between the supraorbital pits expressed as a percentage of orbital
width is 55:9 which compares with values of 42-3 and 53-2 for Pelorovis oldowayensis.

5. The zygomatic arch, continuing forwards under the orbits, is still deeper than
in P. oldowayensis (Pl. 6), and more strongly concave on its ventral surface. This
could be correlated with the mechanically less favourable position of the horn cores.

6. The tooth row is sited less anteriorly than in P. oldowayensis, as may be seen

PELOROVIS OLDOWAYENSIS RECK 293

20
2 CMS 60
100
120
140
58 39 40

Fics. 38-40. Horn core sections at 20cm. above base. 38, in the complete skull of
Pelorovis ; 39, in Pel 7; 40, in the second complete skull from BK II. Sections of Pel 7
are also shown at distances of 60, r00, 120 and 140 cm. above the base. If the horn cores
pictured here are related to skulls with tooth rows in a horizontal position, then the
ventral sides are on the right, the dorsal on the left and the anterior towards the foot
of the page.

by comparing the positions of M? and orbit on PI. 2, fig. 1 and PI. 6, fig. 1 (the orbit
being as far back in this skull as in P. oldowayensis). The ratio of the length from
the premaxilla tip to M? as a percentage of that from the premaxilla tip to the nearest
point on the orbit was c. 87-2 compared with 82-3 in P. oldowayensis.

7. The occipital surface of the skull has a reading of 34:0 for height as a percentage
of width, which compares with readings of 40-7, 36:9 and 35:5 for P. oldowayensis.

8. The reading for the width between the anterior tuberosities of the basioccipital
compared with the posterior ones was 64:2, which is different from the P. oldowayensis
readings of 53°8, 57-6 and 58-4. This could be connected with the very great span
of the horn cores as well as with their insertion in front of the level of the occipital
surface.

g. The coronoid process of the lower jaw is slightly more sharply curved back
than in P. oldowayensis (Pl. 6). This is correlated with character (6) above.

to. The horizontal ramus of the lower jaw is slightly more shallow beneath the
molars.

tr. P, on both sides has an open anterior part of its medial wall.

12. The anterior edge of each molar tooth in the lower jaw has a longer contact
with the tooth in front than has P. oldowayensis.

The length of the lower premolar row as a percentage of the molar row is 58-7,
which compares with values of 50-9 and 58-5 in P. oldowayensis.

In characters 2, 3, 6, 7, 9 and 10 this skull is more like that of Syncerus than is
P. oldowayensis, but in 8 and 11 it is less similar, It is also nearer to Homovoceras
in 2, 3, 6, 7 and 10,

294 PELOROVIS OLDOWAYENSIS RECK

One would not necessarily look for a greater amount of morphological difference
between two Bovine species than exists between this skull and P. oldowayensis ;
on the other hand it is known for the Bovini to be very variable in their skull
morphology (see Pilgrim 1939: 263 on variability in Hemibos triquetricorms
Riitimeyer), and this Olduvai skull is not at all beyond a possible range for P.
oldowayensis. Because I cannot be certain, even with a whole skull, that this animal
is specifically distinct from P. oldowayensis, I shall not name it. More Bovine
remains diminishing the morphological gap between Pelorovis and buffaloes may well
be found as further excavations add to faunal knowledge of the African Pleistocene.

V PHYLOGENETIC AND FUNCTIONAL CONSIDERATIONS

Some of the skull characters of Pelorvovis may be distinguished as primitive,
advanced or specialized. Primitive characters are those in which Pelorovis does
not differ from Boselaphini or early Bovine genera such as Pachyportax, Parabos and
Proamphibos ; advanced ones are those in which it does differ from such early forms
and which it shares with all or many of the later Bovine genera Homovtoceras, Syncerus,
Hemubos, Bubalus, Leptobos and Bos ; specialized ones are those for which Pelorovis
is distinctive although not invariably unique. This sort of assessment is not possible
with all skull characters, nor with those of the limb bones and vertebrae.

Primitive characters, principally to be seen on the teeth, are: the poorly developed
basal pillars, the absence of localized outbowings of the lateral walls of the upper
molars, and the not very complex enamel walls of the central cavities.

The advanced characters of Pelorovis are: its large size and very large horns, the
lack of an ethmoidal fissure, the lack of a preorbital fossa, and the fairly hypsodont
teeth.

Its specialized characters are: the posterior insertion of the horn cores, anteriggly
sited tooth row, and characters linked with these two.

One can suppose that with increasing size in any stock of Bovidae, the support of
larger horn cores becomes a problem; this may be appreciated by comparing the
post-orbital part of a horse’s skull with the more massive construction of a larger
Bovid. Within the Bovini, cattle have the horn cores set widely apart and far
back. Thus the weight of the horn cores is supported above instead of in front of
the occipital condyles, and the strain of support is widely spread across the back
of the skull. In Bubalus and Homoioceras the horn cores are large, and set closer
together in front of the level of the skull’s occipital surface. Hence there is noticeable
enlargement of the frontals for support, and, in Homovtoceras nilssont, strengthening
of the zygomatic arch beneath the orbits as well. The curvature of the long horns
is such that these skulls are most stable in an almost horizontal position, In
Pelorovis the horn cores are set as far back as in cattle, although not so widely apart ;
their curvature carries the greater part of their weight so far behind the level of
the occipital condyles that in life the animal must have counteracted by holding its
head more nearly in the vertical position than do other Bovini. This is probably
the reason for the strengthening of the zygomatic arch in an animal with its horn
cores set so posteriorly. The complete skull from BK II at Olduvai, the identifica-
tion of which is doubtful, has less posteriorly inserted horn cores than in Pelorovis,

PELOROVIS OLDOWAYENSIS RECK 295

and their curvature takes them less behind the level of the occipital condyles. This
might mean that in life the head was held a little less nearly vertically than in
P. oldowayensis, except that in this skull the strengthening of the zygomatic arch
is immense. Unless the skull of this animal was held nearly in the vertical plane,
the curvature of its horn cores would have given less stability than there is in Bubalus
and Homovoceras skulls.

The skull of Pelorovis shows a number of parallels with Alcelaphus, a genus which
also holds its head in a more nearly vertical plane than other Bovidae. In both,
the horn cores are inserted far back and close together, the face is long, there is
sub-orbital strengthening of the zygomatic arch, an anteriorly placed tooth row,
the median indent at the back of the palate passes well forwards, and there are
small palatal fissures between premaxillae and maxillae. The 1952 find of Pelorovis
was of a herd, and it might be thought that these animals could have been plains-
dwelling grazers like Alcelabhus. However this may be carrying the comparison
too far, and it is awkward to accommodate with the primitive occlusal pattern of
the molars. The only ecological possibility likely for a Bovid which I would
eliminate for Pelorovis would be eye-level browsing.

There are also a number of limb bone similarities between Alcelaphini and Peloro-
vis: the hollow between the great trochanter and articular head of the femur, the
wide lateral part of the articular head, the wide patellar fossa on the femur; the
small posterior eminence and low lateral tuberosity of the humerus, the wide bicipital
groove and the high medial condyle distally ; and possibly the non-swollen distal
end of the radius in side view.

(It is none the less quite clear that Pelorovis is not an Alcelaphine because it has
not such distinctive characters of one or more Alcelaphine genera «. an abrupt altera-
tion in the course of the horn cores, transverse ridges on the horn cores, excessively
narrow nasals, a long contact of the premaxillae on the nasals, the premaxillae
narrowing very little as they rise towards the nasals, the foramina towards the back
of the palate being wide apart, the curved arcade of the tooth row, a strongly developed
hypsodonty, the absence of basal pillars, or a complicated course of the enamel walls
of the central cavities of the molar teeth.)

It is possible that while Africa is today inhabited by only one Bovine species which
varies in morphology and ecology, it has been simultaneously inhabited at various
times in the past by two or more species with more restricted ecologies.

The skull described in the last section of this paper is a possible example. There
may also be mentioned two left upper molars, 1953, BK II Extension, 84 and 1957,
BK II, 532 which are surely Bovine but which are smaller than the molars of
Pelorovis, have more localized outbowings of their lateral walls, and central cavities
which have perhaps too complicated enamel walls. The first one is 3-65 cm. long
and 2:5 wide, the second 3°8 x 2:6. M.25715 in the British Museum (Natural
History) is half a Bovine upper molar from Kanjera which is certainly not from
Pelorovis, since it is smaller and definitely more advanced ; other specimens from
Kanjera which could belong to the same species are M.25683, M.25697 and M.25705.
In the Natron beds (see Isaac 1965), contemporary with the upper part of Bed II
at Olduvai, is a Bovine left mandible fragment, WN 64,256, MP 1, which is perhaps

296 PELOROVIS OLDOWAYENSIS RECK

a little smaller than P. oldowayensis and has more advanced teeth. It has retained
M, and M3, and they have large basal pillars, localized outbowings of the medial
walls and noticeably narrowed lateral parts of the front and back lobes.

1957, SHK II, 671 is the tip of a horn core which appears to have been much
shorter than Pelorovis horn cores and with less curvature. From its size it is likely
to have belonged to a Bovine animal.

On the question of the ancestry of Pelovovis, Simatherium kohllarseni (Dietrich
1942 : 119-20, pl. 20, figs. 161, 163, 165) may be mentioned. It is the rear part of a
damaged skull from the Bird River Region, Laetolil, found in deposits thought to
be a little older than Bed I at Olduvai. So far as can be seen from the description
and three photographs this animal has characters which fit it to be ancestral to
Pelorovis oldowayensis. It is apparently about the size of a buffalo, and hence a
little smaller than P. oldowayensis ; the horn cores arise behind the orbits but not
so far back as in P. oldowayensis, and are larger than in its supposed contemporaries
Henubos and Leptobos in Eurasia although they look smaller than in P. oldowayensis.
The horn cores are stated to have some degree of flattening, they either have no
keel or have only a slight one in the position found in Bos primigenius namadicus
and B. acutifrons, there is a wide shallow groove on the less damaged left horn core,
and this horn core also shows a curvature comparable to that in P. oldowayensis. The
frontals are raised between the horn bases; in this the animal resembles Hemibos
rather than Lepftobos. Excavations have yet to reveal the ancestor of Simatherium.

The linking of Pelorovis with Syncerus does not illuminate the evolutionary origins
of the Bovini. Previous work on Asian fossils has shown that the living Bubalus
belongs to a group containing the extinct Hemibos and probably Proampiubos as
well, and Pyroamphibos may be accepted as phyletically close to its Boselaphine
contemporaries Parabos and Pachyportax. The Bos-Leptobos group represents
another stock of Bovini in which the horn cores have more or less evenly rounded
cross sections, and the central question is the history of keels on the horn cores.
Have Leptobos and Bos lost keels so that their ancestors could be Parabos, Pachyportax
or similar forms, or have they never had keels so that ancestors without keeled horn
cores must be looked for?

Pilgrim (1939: 141) supposed that the Leptobos-Bos ancestors did have keels,
his evidence being that the early Bos acutifrons of the Pinjor stage had what could
be the remains of a single keel, so too did Bos primigenius namadicus and at least
one specimen of Leptobos falconerr (American Museum of Natural History, no.
19816, see Pilgrim 1937: 817). It may also be noted that the Pleistocene bison of
East Asia, discussed by Matsumoto (1918), appear from his diagrams to have had
traces of keels. Both Merla (1949 : 147) and Riitimeyer (1878 : 117, 160, 167) were
convinced of the Boselaphine relationship of Leptobos, and since all known genera
of Boselaphini have keeled horn cores, there is a supposition that the ancestors of
Leptobos also did.

The evolution of Bovine horn cores would thus be from the original more or less
circular cross section of the earliest Bovidae such as the Miocene Eotragus Pilgrim,
perhaps through some degree of lateral compression, then through a genus like
Protragocerus Depéret of the European Upper Miocene (according to Thenius 1956

Pia On Owais SOLE DIOMAAYGE NES IES: RUE CK 297

this genus should now include Strepsifortax Pilgrim from the Indian Chinji) which
has at least the beginnings of keels, then an increase in size to Pachyportax and
Parabos or their relatives. This would have been the latest stage at which the
Leptobos-Bos group could have separated from the Hemibos-Bubalus group, since
Proamphibos seems to be already approaching the vomer specializations of the latter
group.

How can Syncerus and Homovoceras be fitted into such a scheme, assuming that
they do have a Boselaphine ancestry? If my opinion of a relationship of Pelorovis
to Syncerus is accepted, then Pelovovis reinforces the views of those who have held
that Asiatic and African buffaloes are not very closely related, and provides no basis
for connecting Syncerus with the Leptobos-Bos group instead. There is no diminution
of the phenetic distance between the African Bovini and those of Eurasia. It is
not clear how far Bovine evolution has consisted of a few long-independent lineages
advancing gradually and often in parallel, or how far there have been repeated
radiations of similar adaptive types at successive levels of overall advance.

There are many interesting chronological and ecological questions still to be
answered on Bovine evolution in Africa alone: did Pelovovis overlap Homotoceras
in time? There are very slender indications (e.g. the Natron Bovine mandible) that
it could have done. Was there a difference in ecological requirements between the
long faced Pelorvovis and the short faced Homovoceras in any way comparable to
that between the long faced Bos primugenius and the short faced bison? What are
the ecological differences between cattle and bison anyway? How did Bos primi-
genius coexist with Homozoceras antiquus in North Africa? What was the other
Bovine at Kanjera, of which we have half a molar tooth?

VI ACKNOWLEDGMENTS

This work was supported by the National Geographic Society of Washington ;
I have also received help and facilities from Dr. L. S. B. Leakey and the staff of the
National Museum and Centre for Prehistory and Palaeontology, Nairobi, the Institut
de Paléontologie, Paris, the Naturhistoriska Riksmuseum, Stockholm, the Paleonto-
logiska Institutionen, Uppsala, and the British Museum (Natural History), London.
All the photographs except PI. 1, fig. r and PI. 3, fig. 1 were taken by Mr. R. J. Clarke,
Pl. 3, fig. r was taken by Mr. R. Campbell.

VII SUMMARY

Much Bovid material assignable to Pelorovis oldowayensis Reck has been excavated
at Olduvai Gorge since 1952. This large species has horn cores inserted close together
and so far behind the orbits that they overhang the occipital surface. The median
indentation at the back of the palate comes well forwards, and the occlusal pattern
of the teeth is not very advanced.

It is clear that Pelovovis is not a Caprine but is a member of the Bovini, and is
phenetically nearest to the African genera Syncerus and Homotoceras. The reasons
for this opinion are given in the conclusions to section III above. Pelorovis is very
unlikely to have been an ancestor of Syncerus and Homovzoceras, but published

298 PELOROVIS OL DOWAYEN SES RECK

photographs (Dietrich, 1942) give no indication that the Laetolil species Simatherium
kohllarsent could not have been an ancestor of Pelorovis.

Material in the British Museum (Natural History) used to define Bularchus arok
Hopwood (1936) is thought to be assignable to Pelorovis oldowayensis.

An almost complete Bovine skull excavated from site BK II at Olduvai in 1952
differs from P. oldowayensis in having horn cores less posteriorly inserted and its
tooth row less anteriorly placed, rather more compressed horn cores, a wider occipital
surface, wider anterior tuberosities of the basioccipital, and a shallower jaw ramus.
It has a greater overall likeness to Homotoceras and Syncerus, and may be a separate
species from P. oldowayensis.

VIII REFERENCES

Bate, D. M. A. 1951. The mammals from Singa and Abu Hugar. In Fossil Mammals Afr.,
2: 1-28, to figs. British Museum (Natural History), London.

Boutin, B. 1935. Cavicornier der Hipparion-Fauna Nord-Chinas. Palaeont. sin., Pekin (C)
9) 4: I=166, pls. 1-22.

BouLKEN, H. 1958. Vergleichende Untersuchungen an Wildrindern. Zool. Jb., Zool. und
Physiol. dey Tiere, Jena, 68 : 113-202, 37 figs.

CoLBErT, E. H. & Hoorer, D. A. 1953. Pleistocene mammals from the limestone fissures of
Szechwan, China. Bull. Amey. Mus. nat. Hist., New York, 102 : 7-134, pls. 1-40.

Cooke, H. B.S. 1947. Some fossil Hippotragine antelopes from South Africa. S. Afr. J. Sct.,
Cape Town, 43 : 226-231, 2 figs.

Dretricu, W. O. 1942. Altestquartare Saugetiere aus der siidlichen Serengeti, Deutsch-

Ostafrika. Palaeontographica, Stuttgart, 94, A : 43-133, pls. 1-21.

1950. Fossile Antilopen und Rinder Aquatorial-afrikas. Palaeontographica, Stuttgart,

99, A: 1-62, pls. 1-6.
Hopwoop, A. T. 1936. New and little-known Fossil Mammals from the Pleistocene of Kenya
Colony and Tanganyika Territory. Ann. Mag. nat. Hist., London (10) 17: 636-641.
Isaac, G. Ly. 1965. The stratigraphy of the Peninj Beds and the provenance of the Natron
Australopithecine mandible. Quaternaria, Rome, 7 : 101-130.

Kent, P. E. 1942. The Pleistocene beds of Kanam and Kanjera, Kavirondo, Kenya. Geol.
Mag., London, 79 : 117-132.

Luakey, L. S. B. 1954. The giant animals of prehistoric Tanganyika, and the hunting-

grounds of Chellean man. New discoveries in the Olduvai Gorge. Jllustvated London News,

19th June : 1047-1051.

1965. Olduvai Gorge 1951-61, I. Fauna and Background. 118 pp., 97 pls. Cambridge.

LONNBERG, E. 1933. Description of a fossil Buffalo from East Africa. Ark. Zool., Uppsala,
25A, 17 : I-32, pls. 1-3.

LypDEKKER, R. 1878. Crania of Ruminants from the Indian Tertiaries. Palaeont. indica,
Calcutta (10) 1 : 88-181, pls. 11-28.

Matsumoto, H. 1918. On some fossil Bisontines of Eastern Asia. Sci. Rep. Tohoku Univ.
(2, Geol.) 3, 1 : 83-102, pls. 24-34.

Mera, G. 1949. I Leptobos Riitim. Italiani. Palaeontogr. ital., Pisa, 46 : 41-155, pls. 5-11.

Mourant, A. E. & ZEUNER, F. E. 1963. Man and Cattle. Occ. Pap. R. anthrop. Inst.,
London, 18 : 1-166.

Pircrim, G. E. 1913. The correlation of the Siwaliks with the Mammal horizons of Europe.

Rec. geol. Surv. India, Calcutta, 43 : 264-326, pls. 26-28.

1937. Siwalik antelopes and oxen in the American Museum of Natural History. Bull.

Amer. Mus. nat. Hist., New York, 72 : 729-874, 81 figs.

—— 1939. The fossil Bovidae of India. Palaeont. indica, Calcutta (n.s.) 26, 1 : 1-356, pls. 1-8.

—— 1941. A Fossil Skull of Hemibos from Palestine. Ann. Mag. nat. Hist., London (11) 7:
347-360, pl. 4.

PELOROVIS OLDOWAYENSIS RECK 299

PomEL, A. 1893. Bubalus antiquus. Carte géol. Algérie Paléont. Monogr., Algiers, 1 : 1-94,
pls. I—Io.

Reck, H. 1925. Aus der Vorzeit des innerafrikanischen Wildes. Leipzig illustr. Zitg., 164:

451. (Not seen).

1928. Pelovovis oldowayensis n.g.n.sp. Wiss. Evgebn. Oldoway-Exped. 1913, Leipzig

(N.F.) 3 : 57-67, pls. I, 2.
RUTIMEYER, L. 1877-78. Die Rinder der Tertiar-Epoche nebst Vorstudien zu einer nattir-
lichen Geschichte der Antilopen. <Abh. schweiz. paldont. Ges., Zurich, 4-5 : 1-208, 7 pls.
Stmpson, G. G. 1945. The principles of classification and a classification of mammals. Bull.
Amer. Mus. nat. Hist., New York, 85 : 1-350.

TEILHARD DE CHARDIN, P. & PIVETEAU, J. 1930. Les mammiferes fossiles de Nihowan
(Chine). Ann. Paléont., Paris, 19 : 1-134, pls. 1-23.

THENIUS, E. 1956. Zur Entwicklung des Knochenzapfens von Pyrotvagocerus Depéret
(Bovidae, Mammalia) aus dem Miozan. Geologie, Berlin, 5 : 308-318, 5 figs.

GEOL. 14, 7. 29

IPL NADIE, i
Pelorovis oldowayensis Reck
(Scales represent Io cm.)

Fic. 1. The complete skull, Pel 1, during its excavation in 1952.
Fic. 2. Dorsal view of complete skull Pel r.
Fic. 3. Palatal view of same skull.

I

PEATE

Bull, Br. Mus. nat. Hist. (Geol.) 14, 7

GEOL. 14, 7.

PLATE 2
Pelorovis oldowayensis Reck
Fic. 1. Lateral view of complete skull Pel r. Scale marked in cm.
Fic. 2. Ventral view of back part of same skull. The arrow points to the indentation
mentioned in character 35, and the scale represents 5 cm.
Fic. 3. The unnumbered frontlet. Scale represents 10 cm.

Bull. Br. Mus. nat. Hist. (Geol.) 14, 7

Fic.
Fia.
Fic.
Fic.
Fic. 5.
premaxilla

eS Coe SI

PLATE 3
Pelorovis oldowayensis Reck
(Scales represent 5 cm.)

Rear view of skull Pel 2 showing the complete occipital surface.

Occlusal view of mandible 1952, BK II, 117.

Lateral view of mandible 1952, BK II, 1109.

Occlusal view of upper teeth 1957, SHK II, 232.

Lateral view of the bones at the front of the face in Pel 23.
and maxilla.

x = suture between

Bull. By. Mus. nat. Hist. (Geol.) 14, 7 PAVE

PLATE 4
Pelorovis oldowayensis Reck
(Scales represent 5 cm.)

Top row from left : anterior view of right femur 1952, BK II, 267 ; anterior view of right
tibia 1952, BK II, 269 ; anterior view of left humerus 1952, BK II, 271 ; anterior view of left
humerus of Syncerus caffer (Sparrman) for comparison. The buffalo humerus has a tall
lateral tuberosity (character 70).

Bottom row from left : anterior view of left radius 1952, BK II, 273 ; lateral view of same
radius ; lateral view of left humerus 1952, BK II, 271 ; lateral view of left humerus of Syncerus
caffer for comparison.

The numbers refer to the characters described in section III of the text.

PLATE 4

Bull. By. Mus. nat. Hist. (Geol.) 14, 7

PLATE 5

Fic. 1. Anterior view of metatarsals of Syncerus caffer (Sparrman) on the left and
Pelorovis oldowayensis Reck on the right.

Fic. 2. Anterior view of metacarpals of Syncerus caffer on the left and Pelorovis
oldowayensis on the right.

The scale for Figs. 1 and 2 represents 5 cm.

Fic. 3. Anterior view of second complete skull from BK II mentioned in section IV of text.
This skull may not belong to Pelorovis oldowayensis. Scale represents 10 cm.

Fic. 4. Palatal view of same skull. Scale represents 5 cm.

PLATE 5

Bull. Br. Mus. nat. Hist. (Geol.) 14, 7

PLATE 6

Fic. 1. Lateral view of second complete skull from BK II to show the strength of the front
of the zygomatic bar beneath the orbit. Scale in cm.

Fic. 2. Medial view of mandible from second complete skull from BK II (above) compared
with lateral view of mandible of Pelorovis oldowayensis Reck from skull Pel 1 (below).
Scale represents 5 cm.

Fic. 3. Occlusal views of lower teeth of second complete skull from BK II (above) compared
with those of Pelorovis oldowayensis (below). Scale represents 5 cm.

XC

Bull. Br. Mus. nat. Hist. (Geol.) 14, 7 PLATE 6

ROLL
a
| 2.9 JUL 1967 }
z

Lea PS, |

INDEX, LO VOEUME Xa

New taxonomic names and the page numbers of the principal references are printed in Bold type.
An asterisk (*) denotes a figure.

Acanthocythere (Protoancanthocythere) faveo-
lata 121-125
acanthodians 180, 181
acanthopterygians 81, 86—7
Acanthothiris broughensis 117
Acanthothiris crossi Bed 115-117, 133, 135
Acentrophorus 183
Acidorhynchus 199
Acipenser 179, 188-9, 199
Acipenseridae 188-9, 199
Acipenseriformes 179, 183, 188, 189, 199
Acrogaster 79, 103, 105
anceps 72, 79
dayi 73, 79
Acrolepididae 175, 181-84, 197, 201
Acrolepis 146, 197
Acropholis 197
Acrorhabdus 197
Actinophorus 196
actinopterygians 178-81, 188
Actinopterygii 178-186, 196
Addax 265, 268-9
Aedua 198
Aeduella 184, 194, 198
Aeduellidae 182-85, 194, 198
Aeotragus garussi 266
Aequacytheridea 39, 40
Aetheodontidae 190, 200
Aetheodontus 190, 200
Aetheretmon 197
AFRICA
Mammalia 243-299
Agecephalichthys 183, 198
Albertonia 190, 200
Alcelaphini 254, 273, 295
Alcelaphus 273, 295
? Aldingeria 197
Amblypteridae 146, 177, 182-5, 189, 198, 201
Amblypterina 198
Amblypterops 198
Amblypterus 146, 153, 155-6, 159, 198
anconoaechmodus 147
punctatus 147
Amia 179
Amiuiformes 179

amphicentrids 195-6
Amphicentridae 176, 183, 185, 195-7
Amphicentrum 176, 197
Anaglyphus 201
Anatoia 201
Ancaster Beds 116
Aneurolepis 198
Anguillavus 72
anoa 265, 267, 270, 273, 275
Anolotichia 18
? Apateolepis 196
Archaeonichthys 198
Archaeoniscus 198
Archegonus 217
(Phillibole) aprathensis Group 210
brevispina Group 210
Archegonus (Waribole) 210
Arctosomus 200
Asaphus gemmuliferus 216
Asciocythere 34
acuminata 125, 134
lacunosa 121, 125-6, 129
ASIA
Mammalia, Bovini 266-7, 271, 288, 296-7
Atherstonia 164, 176, 186, 198
Atherstoniidae 164, 176, 183, 185-6, 198,
201
Atopocephala 200
Aulacocythere punctata 116, I2I—5, 127, 134
reticulata 123
Aulolepis 86-7, 107
typis 87*, 107
Australosomus 193, 200

Bairdia hilda 116, 121, 123, 125, 138
Basement Beds 117, 118, 125, 132, 133, 135
Bate, R. H. 21-66; 111-141
Beconia 200
Belekocytheridea 34, 36, 37, 64
punctata 36, 37, 63, 65, Pl. 6, figs. 1-5, Pl. 5,
figs. 9-13
Belenorhynchidae 187, 199
Belenorhynchus 199
Belenostomus 72, 199
Belichthys 182, 198

302 INDEX

Benedenichthys 197
Benedenius 197
Berycidae 83, 105
Beryciformes 79, 81, 87-8, 106-108
Berycoidei 74, 79, 81, 87*, 95, 102-3, 107
characteristic features 86
Cretaceous 74, 79, 102-6
distinction between Cret. families 103-4
Origin and evolution of 81, 87, 103-105*, 106,
107-8
Recent 104
Berycopis major 97
Beryx superbus 97
Besania 1gI—2, 200
Bibos 265
Birgeria 186, 198
Birgeriidae 183, 185-6, 198
Bison 265, 272
Bisulcocypris 29, 30, 64
ancasterensis 32, 33, 63, Pl. 3, figs. 1-10
anglica 30, 31-33, 63, Pl. 2, figs. 1-11
tenuimarginata 31-33
sp. A 30, 32
sp. B 32, 33
Blue and Silver Beds 115-17, 122-3, 133
Bobosatrania 195-6, 201
bobasatranids 185, 195
Bobasatraniidae 195, 201
Bobasatraniiformes 183, 195, 196, 201
Bollandites cf. castletonense 239
Boreolepididae 182-3, 198
Boreosomidae 177, 178, 182-4, 199, 201
Boreosomus 178, 183, 199
Bos 265-84, 286, 288-9, 294, 296
acutifrons 266, 269-70, 295-6
gaurus 265, 267-69, 271-79, 281*, 283*, 286*,
287
grunniens 274-5, 278-9
javanicus 265, 273
palaeosinensis 266
primigenius 267-69, 271-2, 274-79, 282, 287, 297
primigenius namadicus 266, 270, 296
sivalensis 266
Bos—Leptobos Group 296-7
Boselaphine—Bovine stock 265, 267, 271, 279,
288, 290, 294, 296-7
Boselaphini 267, 279, 288, 290, 294, 296, 297
Boselaphus 271
Bovidae 246-7, 267
Bovine 267, 294
Bovini see under Mammalia
Brachiopterygii 189
Brachydegma 177, 199
Brachydegmidae 177, 182-3, 199, 201
Brometta 198
Brookvalia 168, 177, 183, 200
Broughia 201
Browneichthys 199
Bubalus 265-84, 286-8, 294-6

bubalis 265, 271, 275, 279, 287

depressicornis 265, 275
‘buffalo’ 265

African 243-299

Asian 266-7, 271, 288, 296-7

European 271, 278, 288, 296
Bularchus 290

arok 246, 290-3
Bythotrypa 4

Calamoichthys 199
Caminchaia 201
Camptocythere lincolnensis 121-3
Canobiidae 175, 182, 183, 185, 197
Canobius 147, 197
Caprinae 249, 265
Caprines 243, 273, 297
Caproberycinae 97
Caproberyx 72, 80, 89-93, 95-96, 97, 98, 102-3
pharsus 74, 80, 97, 98, 99*, 100, IoI, 103,
105-6, Pl. 4, fig. 2.
polydesmus 98, 101-2, 105
superbus 90, 91, 96-102, 105
CARBONIFEROUS
Czechoslovakia 146, 165, 168, 170, 172
England 162-3, 238-9
Germany 212, 238-9
Scotland 148, 153-60
Pisces 148, 153-163, 165-89
Trilobita 209-10, 217, 238-9
trilobite-goniatite associations 212, 218, 220,
222, 238-9
Carbonocoryphe 219
? ferruginea 219
? Carbonocoryphe 210
Carboveles 168, 196
Carbovelidae 175, 181, 183, 196
Catopterus 200
Caurichthys 185, 197
Cave Oolite 115, 117, 124, 133-35
Caytonidea faveolata 135
Cenchrodus 201
Cenechoia 201
Centrolepididae 182, 183, 198
Centrolepis 182, 198
Cephalacanthus 201
Cephaliscus 201
Cephaloxenidae 191, 200
Cephaloxeniformes 183, 191, 200
Cephaloxenus 191, 200
Ceramopora imbricata 6
Ceramoporella 18
Ceramoporella sp. 16
Ceramporoidea 5-7
Challaia 198
Cheirodopsis 178, 197
Cheirodus 197
Cheirolepididae 181, 183, 196
Cheirolepis 181, 187, 196

Chilopora guernoni 5
Chirodus 176, 195, 197
chondrosteans 180, 201

Chondrostei 178, 179-182, 183*-189, 190-201

Chondrosteidae 199

Chondrosteiformes 183, 185, 188, 189, 199

Chondrosteus 188, 199
Chrotichthys 200
Cleithrolepididae 190, 200
Cleithrolepidinae 190, 200
Cleithrolepis 200
Cloughtonella rugosa 134
Clupavidae 72
Clupavus 72, 107
Clupea gaudryi 72
Coccocephalichthyidae 182-3, 199
Coccocephalichthys 199
Coccocephalus 199
Coccolepididae 180-3, 199
Coccolepis 180, 183, 198

liassica 180

macroptera 180
Cocconiscus 199
Colobodontidae 190, 192, 200
colobodontids 190
Colobodus 190, 200
Commentrya 198
Commentryidae 146, 182-4, 193, 198
Corniger 96
Cornuboniscidae 183, 186, 197
Cornuboniscus 186, 197

Cosmolepididae 177, 182-3, 198, 201

Cosmolepis 177, 183, 198
Cosmopoma 198
Cosmoptychiidae 175, 181, 183, 197
Cosmoptycius 197
Crenilepis 200
Crenolepis 200
Crenilepoides 200
Crepipora lunatifera 18
squamata 14
CRETACEOUS
Pisces 70-109, 179, 181, 188-9
Cenomanian 71-2
Senonian 79, 82, 87
Turonian 97
Albian 108
Aptian 107-8
Lebanon 70-72, 97
Crisia 6
Crossopholis 199
crossopterygians 178-80
Cryphiolepididae 175, 182-3, 197
Cryphiolepis 146, 168, 197
Ctenothrissa 72, 84, 86, 107-8
microcephala 107
radians 87*, 107
Ctenithrissiformes 86, 87*, 88, 106-7
ctenothrissoid 86

INDEX 303

Cycloptychius 197
Cylindraspis aprathensis 220
Cyrtosymbole 216, 218
librovitchi var. euryaxis 220
librovitchi var. latilimbata 219
Cyrtosymbole (Macrobole) 216
cf. brevispina 210, 217, Pl. 2, figs. 1 & 2
drewerensis 216, 226
laticampa 218
aff. laticampa 210, 218, Pl. 2, figs. 3 & 5
Cyrtosymbole (Waribole) cf. aequalis 217
chudleighensis 210, 218, 219, Pl. 2, fig. 7
warsteinensis 218, 219
Cyrtosymbolinae 210, 216, 219
Cystispina 230
cystispina 232
spatulata 230
Cystodictyonidae (Cryptostomata) 5
Cythere drupacea 44
guembeliana 49
juglandica 51
juglandica var. major 51
subconcentrica 60
Cytherella fullonica 121, 123
Cytherelloidea catenulata 116, 122-23, 125, 127
eastfieldensis 122, 125, 133
Cytheridea bradiana 52, 53
ostreata 51
perforata 39
pulvinar 49
Cytheris fullonica 61
Cytheromorpha ? greetwellensis I2I, 123, 134
Cytheropterina 134
comica I2I-123, 126, 132-3
gravis I2I—-23, 126, 132, 133
plana 127, 129-30, 134
CZECHOSLOVAKIA 146, 165, 168, 170, 172

Daedalichthys 200
Damaliscus 273
Darwinula 28, 64
barabinskensis 29
daps 28
incurva 28, 29, 63-4, Pl. 1, figs. 7-12
leguminella 28
stephensoni 64
stevensoni 28
tubiformis 29
sp. A 28, 29
Darwinulacea 28
Darwinulidae 28
Darwinulidea 23
Deltaic Series, see under Lower, Middle & Upper
DEVONIAN
Pisces 178, 180-81
Diacoryphe 236, 237
gloriola 236
pfeifferi 236-8
strenuispina 236-8

304

vandergrachtii 236

? vandergrachtii 236, 237, Pl. 7, figs. 1-5
Dianulites 4
Diaphorognathus 199
Dicellopyge 198
Dicellopygidae 182-4, 193, 198
Dictyopleuchrithys 200
Dictyopyge 200
Dictyopygidae 193, 200
Dimorpholepis 200
Dinopterygoidei 106
Diplomystus brevissimus 72-3
Dipnoans 178-80
Dipteroma 198
Dipteronotus 200
Discopora squamata 14
Disichthys 197
Dolocythere maculosa 116, 121-30
Dollopterus 190, 200
Dorypteridae 196, 201
Dorypteriformes 183, 196, 201
Dorypterus 196, 201
Drydenius 146, 149, 159, 184, 196

insignis 159, 160*

molneuxi 161, 162*, 163*

Echentaia 201
Ecrinesomus 201

Ektyphocythere triangulare 116, 117, 121-6,

129-30
Elaveria 198
Eller Beck Bed 115, 118-19, 130, 132-38, 140
Elonichthyidae 146, 175, 181-3, 197, 201
Elonichthys 146, 149, 152, 156, 197
intermedius 147
robisoni 147, 149
robisoni var. intermedius 147, 149
semistriatus 149
? Elpisopholis 196
Enchelion 72
ENGLAND
Carboniferous (Pisces) 162-3
Cretaceous, Gault, (Pisces) 107-8
Devon, Carboniferous (Trilobita) 238-9
Eastern England, (Ostracoda) 21-66
Jurassic, Upper Estuarine Series 21-66
Middle 111-141
Lincolnshire, M. Jurassic 111-141
Northern England, (Trilobita) 221-2, 239
Yorkshire, M. Jurassic 117—19, 133, 135-40
Engycolobodus 200
Entogonites grimmeri 239
Eocytheridea 133, 138
carinata 116, 124-5, 127-30
elongata 124-5, 127
faveolata 121-2, 127, 134
lacunosa I2I, 125-30
reticulata 121, 134
? acuta 128, 130, 133-34

INDEX

? astricta 124-5, 127-30, 134
? erugata 125-30

Eoschuleridea 39, 40, 41
Eotragus 296

Equisites 137
Erpetoichthys 179, 189, 199
Errolichthyidae 199
Errolichthys 188, 199
Eubiodectes 72
Eupalaeoniscus 198
EUROPE

Mammalia 271, 278, 288, 296

Estuarine Series, Lower 118, 123, 133, 135
Eurylepidoides 197

Eurylepis 199

Eurynothus 176, 197

Eurynotoides 176, 197

Eurynotus 176, 197

Eurysomus 197

Evenkia 198

Evolution

Mammalia, Bovini 296-98
Ostracoda 39-40

Pisces 178-96

Trilobita 217

Fabonella 33

bathonica 33, 34, Pl. 4, figs. 1-5

Favositella 3, 4, 5, 6, 18

anolotichoides 16, 17-18, Pl. 3, figs. 1, 3;
Pl. 5, figs. 2 & 5; Pl. 7, figs. 3 & 4; Pl. 9,
fig. 3.

gotlandica 19, Pl. 2, fig. 5; Pl. 5, figs. 4 & 6;
Pl. 9 figs. 1 & 2.

interpuncta 7, 8-9, 13-14, Pl. 1, figs. 1-6;
Pl. 2, figs. 1-3, 6, 7; Pl. 3, figs. 4-6; Pl. 4,
figs. 1-6; Pl. 6, figs. 1-5; Pl. 7, figs. 1 & 2;
Pl. 8, figs. 1, 3 & 4; Pl. 9, figs. 6 & 7.

interpuncta forma brevipora 9, 11, 13, Pl. 2,
fig. 7; Pl. 4, figs. 2, 6; Pl. 6, figs. 1, 2; Pl. 9,
fig. 7.

interpuncta forma intermedia 10, 14, Pl. 3,
fig. 4.

interpuncta forma ivregularis 10, 12, Pl. 4, fig.
3}, JEM, a tthe Be

interpuncta forma texuvata 9-16, 19, Pl. 3,
fign 55 Pl 4) figs 4;) Plo 7 etigaesebl ns,
fig= 3.

interpuncta forma typica 8-13, 19, Pl. 1, figs.
4-6; Pl. 2, figs. 2, 3; Pl. 3, fig. 6; Pl. 4, figs.
1 & 5; Pl. 6, figs. 3-5; Pl. 9, fig. 6.

squamata 14, 15, 16, 19, Pl. 2, fig. 4; Pl. 3,
fie, 25 Plo 5) figs, md 3); Ply 7g Sms
fig. 2; Pl. 9, figs. 4 & 5.

Favosites fibvosus 7, 12

interpunctus 4, 7

Fistuliporidae 5
Foraminifera 71-2
Fouldenia 197

INDEX 395

Fuhrbergiella (Praefuhrbergiella) arens 116, 121,
123-25
minima 127, 130, 132-33

Galliaecytheridea 34
Galliaecytheridea dissimilis 36
? kingscliffensis 34, 35-36, 42, 65, Pl. 4, figs.
6-12; Pl. 5, figs. 1-8
postrotunda 35
Galliaecytherideinae 34
Ganacrodus 197
Ganolepis 197
GARDINER, B. G. 143-206
Geitonichthys 200
GentTRY, A. W. 243-299
Geomichthys 198
Geophyroberyx 86
GERMANY
Carboniferous 212, 238-9
Giffonus 199
Gigantopterus 190, 200
Glaucolepis 198
Globulodus 176, 197
Globotruncana 72
Glyptocythere 47, 49
guembeliana 49, 50, 64, Pl. 13, figs. 10-16;
Pl. 14, figs. 1-8
juglandica 51, 65, Pl. 14, fig. 9.
sp. I16, 132
Gnathoberyx 77, 81, 83, 87-8, 103-5
stigmosus 81, 82, 83, 84*, 85-87*, Pl. 1, fig.
erable 2.
Gonatodidae 146, 147, 159, 175, 177, 181-5, 193,
196, 201
Gonatodus 146, 147, 149, 153, 156, 159, 196
brainerdi 147
macrolepis 147, 153, 157, 201
molyneuxi 161
parvidens 147, 153, 201
punctatus 147, 148*, 149—-151*, 152*, 153, 155,
158-9
? tolliezi 147
Grey Limestone 117-18, 120, 132-34, 140
Griffithides acanthiceps 227
longispinus 232, 234
Gymnoniscus 168, 197
pauper 169
Gymnosaurichthys 199
Gyrolepidotidae 175, 176, 181, 183, 201
Gyrolepidotus 175, 196
Gyrolepis 198
Gyrosteus 199
Guaymayenia 201

Habroichthyidae 178, 191, 200-1
Habroichthys 178, 191, 200
halecostomes 194

Halecostomi 178, 181
Haplolepiformes 183, 185, 186, 187

Haplolepis 199
Hedbergella 71, 72
Helichthys 200
Helmolepis 201
Hemibos 266, 269-71, 273, 275, 277; 288, 294, 296
acuticornis 269
triquetricornis 269, 294
Hemibos—Bubalus Group 288, 297
Hemicladodus 197
Hemilopas 201
Hemiscaphirhynchus 199
Hibaldstow Oolite 116-17, 123
Hippotragini 265-78, 280-82, 283*, 284, 286-89
Hippotragus 265-66, 268, 270, 273-74, 276-79,
280*, 281-83*
equinus 268, 272-74, 278-79
niger 268-69, 273-75, 278-79
holocentrid 80-81, 89-91, 95-6, 102-104, 106
Holocentridae 79-80, 95-97, 102-104, 106-7
Holocentrinae 96, 105
Holocentrus 92, 97
holosteans 189-90, 192, 194
Holostei 178-79, 181, 194
Holotrachys 89, 97, 104
Holuridae 164, 183, 185-87, 197
Holuropsis 168, 197
Holurus 168, 197
Homoioceras 266, 269-70, 277-78, 284, 288, 290,
293-95, 297, 298
antiquus 266-68, 270, 272, 275-6, 297
nilssoni 266, 267*, 268, 270, 272-76, 278-80,
281*, 282, 284, 286-7, 294
singae 266-68, 272, 275-6
Homocytheridea cylindrica 125, 126, 129
Hoplopteryx 74, 77, 80, 83, 86, 103, 105-6
antiquus 105
gephyrognathus 105
lewesiensis 105
lewisi 97, 105, 107
macracanthus 86, 87*, 105, 106
simus 92, 105
spinulosus 86, 105
syriacus 105
Hoplostethus 80-1, 83, 85-6
Huso 179, 199
Hydraulic Limestone 117-19, 126, 132-38, I40
Hydropessum 200
Hylingea 197
Hypterus 198

Ichthyodectes 72
Ichthyorhynchus 199
Ischnolepis 200

Jacobulus 201

JURASSIC
Bathonian, Upper Estuarine Series 21-24, 25*,
26*, 27-66

Middle Jurassic 113, 116, 118, 132-34, 140

306 INDEX

Ostracoda 21-66, 116, 133-34
Pisces 179-80, 182-3, 187-8, 192, 195

Kansius 95-6, 103, 105

Kaskia 216

Kentuckia 196

Kessleria 179, 199

Kirton Cementstone Series 115-16, I2I, 123,
133-34, 135

Kirton Shale 115, 123, 132-3, 135

Kirtonella plicata 123, 128
reticulata 127-30

Klieana 51, 64
levis 51, 63-5, Pl. 14, figs. 10-13, Pl. 15, figs.

I-5

Lambeichthys 201
Lawnia 177, 184, 198
Lawniidae 176, 177, 182-84, 198, 201
LEBANON
Upper Cretaceous localities 70-72, 97
Upper Cretaceous, (Pisces) 67—109
Leiolepis 198
Lekanichthys 197
Lepidopterus 198
Lepisosteus 179
Leptobos 266, 269-71, 273, 277, 288, 294, 296
falconeri 269, 296
Leptobos—Bos Group 296-7
Leptogenichthys 198
Lincolnshire Limestone 113, 114*, 115, 116-17,
132, 134, 138, 140
Liobole 227, 230
coalescens 229
glabra 227, 228-30, 238-39, Pl. 3, figs. 1-5
glabra glabra 229
glabra hiemalis 229
glabroides 229
subaequalis 230
zarembiensis 230
Liobolina 227
Lissoberyx 73, 79-81, 86, 98, 103-4, 106-7
anceps I05
dayi 72, 73, 74, 75*, 76*, 77, 78*, 79, 105, PI.
Tenigs Teele 4s feet
Lophocythere 51-3
bradiana 53
flexicosta 54
multicostata 53
ostreata 51
plena 54
scabra 52, 64-5
scabra scabra 52, Pl. 15, fig. 6
septicostata 52, 53, 65, Pl. 15, figs. 7-13; Pl.
16, figs. 1-4.
transversiplicata 53, 54, 65, Pl. 16, figs. 5-15
Lower Deltaic Series 118-19, 131, 133, 137
Luganoia 191-2, 200
Luganoiidae 191, 200

Luganoiiformes 183, 191, 192, 200
Lycopterocypris 36
Lyubinovina 34
Macroaethes 193, 200
Macrobole 216-18, 226
aff. laticampa 238-9
brevispina 217
cf. brevispina 231, 238-9
Macrocypris horatiana 27, 28
terrae-fullonicae 27, 28
Macrodentina 55, 64
Macrodentina (Dictyocythere) 55
Macrodentina (Macrodentina) 55
Macrodentina (Mediodentina) 55
bathonica 55, 56, 63, 65, Pl. 17, figs. 1-12;
Pl. 18, figs. 1-4
Macrodentina (Polydentina) 55
Macristium 86
chavesi 87
Manlietta 200
Marslatovrella 56
MAMMALIA
African 243-299
Asiatic 266-7, 271, 288, 296-7
European 271, 278, 288, 296
Bovini 243-299
Evolution of 296-8
Living genera 265, 271, 276
Characters of 276, 282-3, 285, 287-8, 290
Pleistocene 243-299
Pliocene 243-299
Horns 269-71, 275, 288, 294-8
Jaws 278-9, 297
Limb bones 279, 280*, 281*, 282, 283*, 284,
285*, 286*, 287, 289-90, 295
Palate 274-5
Sexual differences (horns) 270-1
Size 267-8
Skull 268, 270-73, 275, 287-8, 294-5, 298
Teeth 274, 275, 277*, 279, 295, 297
Vertebrae 285-6, 289-90
Hippotragini 243-299
Market Weighton 117-18, 135, 137-9
Marslatourella bullata 56, 57, 64-5, Pl. 18, figs.
5-14; Pl. 19, figs. 1-2.
exposita 57
Mecolepis 199
Mediodentina see Macrodentina (Mediodentina)
? Megapteriscus 196
Meidiichthys 190-200
Mekolepis 199
Melamphaeidae 105
Mendocinia 200
Mendocinichthys 200
Mentzichthys 197
Meridensia 190, 200
Mesembroniscus 178, 199
Mesolepis 197
Menonichthys 197

INDEX 307

Mesopoma 197
Metacytheropteron 44
drupacea 44, 45, 65, Pl. 10, figs. 1-9
elegans 45
Microconodus 159
molyneuxi 161
Micropneumatocythere 57, 60, 64
convexa 121-2, 124-5, 127-8, 130
globosa 60, 61, 116, 122, 124-30
postrotunda 57, 58, 63, Pl. 19, figs. 3-10, 13-16
quadrata 58, 59, Pl. 19, figs. 11, 12; Pl. 20,
figs. I-12.
subconcentrica 60, 61, Pl. 21, figs. 1-13
Microtragus parvidens 276
Middle Deltaic Series
Lower 118-19, 125, 129-31, 133-4, 138
Upper 118-19, 128-29, 131, 133
Millepore Bed 131
Millepore Oolite 115, 118, 119, 120, 129-34,
138-9 140
Molybdichthys 200
Monocentridae 105
Monoceratina 33
scarboroughensis 33, 65, Pl. 3, fig. 11
cf. scrobiculata 123
vulsa 117, I2I-3, 125, 127, 129-30
Monticulipora sp. 7
Moythomasia 181, 196
myctophoids 88
Myriolepis 146, 182, 198
Myripristinae 96, 105
Myripristis 75-7, 86-7, 92, 94, 96-7, 103, 106
murdjan 87

Namaichthys 149, 197
Neavichthys 200
Nematoptychius 147, 149, 197
Neochallaia 201

Nephrotus 201

Notaemon 197

OAKLEY, K. P. 1-20
Ohmdenia 198
Oligocythereis 61
fullonica 61, 65; Pl. 21, figs. 14, 15.
Omphalodus 201
Orvikuina 196
Oryx 265-6, 268-70, 272-79, 281-2, 284, 287-8
algazel 269
beisa 268-9, 272, 274, 278-9
gazella 269, 273, 276
leucoryx 272-3, 275-6
Ostichthys 96
Osorioichthyidae 175, 181, 183, 196, 201
Osorioichththys 175, 181, 196
Ospia 201
OSTRACODA 21-66, II—141
evolutionary series 39, 40*

Middle Jurassic distribution 134
* », faunas 116, 133-4

ornament controls 30
palaeoecology 63-65
sexual dimorphism 50

Ovibos 273, 277

Oxygnathus 198

Oxypteriscus 201

Pachyportax 294, 296
Paladin 216
Palaeobergiia 175, 196
Palaeocytheridella 34
PALAEOECOLOGY
Ostracoda 14, 30, 63-65
Pisces 70-71, 178, 180-81, 184-88, 190-95
Trilobita 209-10
PALAEOGEOGRAPHY 134-35, 136*, 137-39*, 140
Palaeomuzon 198
Palaeoniscidae 178, 181-4. 198
Palaeonisciformes 146, 180, 181-6, 188, 194, 196,
201
Palaeoniscinotus 168, 197
palaeoniscoids 178, 180-2, 185-6, 189-90, 193,
195, 201
Palaeoniscum 182-3, 198
Palaeoniscus 198
Palaeophichthys 187, 199
Palaeopsephurus 189, 199
Palaeoryx 270
woodwardi 276
Palaeothrissum 198
Parabos 266, 277, 279, 288, 294, 296
boodon 266
cordieri 266
Paracypris 27
bajociana 28, 116-7, 122-3, 125-30, 138
terraefullonica 27*, 28, 65. Pl. 1, figs. 1-6
? sp A 28
Paraeurynotus 176, 197
Paramblypterus 159, 184, 189, 198
Paramesolepis 197
Parasemionotidae 194, 201
Parasemionotiformes 183, 194, 201
Parasemionotus 201
Pasambaya 201
Patoperca 72, 107
PATTERSON, C. 67-109
Peipiaosteus 199
Peipiasteidae 199
Peleichthys 197
Pelorovis 245-6, 248, 249, 254, 257-01, 265-67,
290-91, 294-5, Pls. 1-6.
Distinguishing characters 287-8, 289-90, 294-5
Functional morphology 294-5
Phylogeny 265-70, 287, 290, 294-98

Horns 270-1, 293*—296
Jaws & Palate 273-4, 295

308 INDEX

Limb bones 279, 280*, 281*, 282, 283*, 284,
285*, 286*, 287, 289, 290, 295
Nasals 273, 295
Skull 271-76, 288-9, 294-5, 298
Vertebrae 286, 287, 289
Pelorovis oldowayensis 245, 246, 247—50*, 265,
267, 290-92, 295-98, Pls. 1-6
Horns 249-53, 256, 266-68, 293-95
Jaws & Teeth 254—-56*, 207, 293, 295, Pl. 3
Limb bones 258-62, 267, 281*, Pl. 4; Pl. 5
Skull 249-53, 256, 266-68, 293-95, Pls. 1, 2, 5
& 6.
Vertebrae 262, 263*—265
Size 267
Peltopleuridae 178, 200
Peltopleuriformes 178, 183, 190-92, 200-1
Peltopleurus 190-1, 200
Pericyclus aff. homoceratoides 239
Perleidiformes 180, 183, 189, 192-3, 200
Perleidus 189-90, 200
PERMIAN
Pisces 176-80, 182-84, 186, 196
Phanerorhynchidae 164, 172, 183, 185-6, 197
Phanerorhynchus 146, 168, 172, 186, 188, 197
armatus 172, 173*, 174-5
Phanerosteon 168, 196, 198
pauper 168-9
Phillibole 220, 222, 224
aequalis 220
aff. aprathensis 221
aprathensis 210, 220, 221, 223, 226, 238, 239,
Pl. 7, figs, 6-9
aprathensis fauna 238, 239
coddonensis 210, 220-1, 223, 224, 225-6, 238,
Pl. 3, figs. 1-5
coddonensis fauna 238-9
culmica 217, 224, 226, 239
polleni 210-11, 221, 222, 223, 226, 236, Pl. 4,
figs. 1-4
? (Cystispina) nasifrons 230
spatulata 230
sp. Nov. 230, 232
(Liobole) glabra glabra 227
glabra hiemalis 227
Phillipsia 211, 216
articulosa 211, 212, 215
cliffordi 211, 215, 220
aff. eichwaldi 220
glabra 227
kellii 211, 216
krasnopolskil 220
leei 211, 212, 213*, 214-6, 218, 224, 238, 239,
Pl. 1, fig. 16
leei fauna 238-9
minor 211, 212, 215
polleni (?) 220
spatulata 230
vandergrachtii 236

Sp. 227

Phillipsidae 210-11, 216, 219
Phlyctaenichthys 200
Pholidopleuridae 193, 220
Pholidopleuriformes 181, 183, 193, 200
Pholidopleurus 200
Pholidurus 199
Pichottia 34, 37
magnamiuris 37, 38, 64, 65, Pl. 6, figs. 6-14;
Pl. 7, figs. 1-6
muris 38
PISCES
Actinopterygii, origin of 178-186
characters, advanced 81, 107
holocentrid 80-1, 89, 104
primitive 81, 87, 106
trachichthyid 80-1, 104
Chondrostei, classification of 196-201
Environment, freshwater 178, 180-1, 184, 188,
190, 193
marine 178, 180, 181, 188, 190, 192-3,
195
Evolution of Chondrostei 178-183*, 184-196
Palaeonisciformes 180—6
Feeding 185, 189-93, 195-6
Morphology
Fins 78-79, 85, 93*, 94, 102, 151-2, 160, 163,
166-8, 170, 174, 180, 186
Jaws 77, 81, 83, 84*, 86, 87*, 91*, 92, 99*,
IOO-I, 150, 156, 160, 166, 170, 174, 182
Maxilla 157*, 158, 160*, 162*, 170, 184-5
Neurocranium 74-6, 80-2, 89, 9o0*, 91*, 98,
99*, 106, 150
Nostrils 180, 185
Ornament 149, 151
Palate 77, 81, 83, 84*, 86, 87*, 91, 92, 99*,
IOO-I, 150, 156, 166
Pectoral and Pelvic Girdles 76*, 78-9, 85,
QI, 93, IOI, I51, 156, 166, 170, 174
Scales 149, 151*, 152*, 156*, 158*, 160, 162*,
163*, 167*, I71I, 179, 183, 186
Sensory canals 180, 183
Skeleton 78, 82, 93, 170
Skull 148*, 149-50, 154*, 155, 160, 161*,
165*, 166, 169*, 170, 172, 173* 180-82,
186
Squamation 79, 86, 94-5, 102, 152, 156, 168,
170, 175
Tail 182, 183, 186
Vertebral column 78, 85, 93, IOI
Stratigraphical
Recent 81, 85, 90-1, 96, 104, 107-8, 189
Eocene 179
Cretaceous 72-73, 179, 181, 188-9
Berycoidei 74, 79, 95, 102-7
Holocentridae 80, 95-6, 102—5*
Trachichthyidae 80-1, 84, 92, 104, 105*
Jurassic 179-80, 182-3, 187-8, 192, 195
Trias 176-95
Permian 176-80, 182-4, 186, 196

INDEX 309

Carboniferous 146-8, 153, 156, 157, 159-60,
163, 165, 168, 170, 172, 175, 178, 180-81,
184-87, 189

Devonian 178, 180, 181

Placopleurus 191, 200
Platella jurassica 122, 123
Platocythere 61
verriculata 61, 62, 63-4, Pl. 22, figs. 1-13
Platysiagidae 192, 200
Platysiagiformes 183, 192, 200
Platysiagum 192, 200
Platysomidae 185, 197
Platysomus 185, 195, 197
Plectrolepis 197
Plectrypops 96, 104
Plegmolepis 197
Pleistocene (Mammalia) 243-299
Pleurocythere kirtonensis 116, 123, 125, 127, 134
nodosa 123-5
Pliocene (Mammalia) 243-299
Pneumatocythere bajociana 121-9
carinata 121, 124, 128-9
Podocopida 27
Podocopina 27
Pododus 197
Poéphagus 265
Polymixioidei 106
Polyodon 179, 189, 199
Polyodontidae 188, 189, 199
Polypteridae 199
Polypteriformes 179, 183, 189, 199
Polypterus 179, 187, 189, 199
PoLyzoa I-20
attachment, influence on growth 9g, Pl. 1,
figs. 4 & 5
cell functions 6
form diversity 13-14
generic extinction 3
generic distinction 4
internal characters 10-11
internal polymorphism 6
maculae 6-8, 14-15
morphological characters 4—6
‘pear-like’ calculi 3-4
pits; 7, El) 3; fig. 6; Pl. g; fig. 6:
polypide degeneration 3, 4, 6
reproduction 6-7
“Sterngruben”’ 7
Praeschuleridea 39, 40*, 41-2
quadrata 42, 43-4, 65, Pl. 9, figs. 1-12
subtrigona 43
subtrigona magna 125-30, 132, 138
subtrigona subtrigona 116, 121-6, 132
ventriosa I2I—2, 125-6, 129
PRENTICE 207-241
Pristisomus 200
Proamphobos 266, 270-1, 275, 277, 288, 294,
296-7
Procheirichthys 200

Procytheridea 34
Proetidae 216
Proetus coddonensis 224
Sp. 227
Sp. a. 224
sp. b. 224
Progonocythere 45-48
cristata 46, II17, I2I, 123, 126-9
juglandica 51
juglandica juglandica 51
levigata 45, 46, 65, Pl. ro, figs. ro—14; Pl. 11,
figs. I-9
reticulata 125
cf. reticulata 126
rugosa 46, 47, 65, Pl. 11, figs. 10-14; Pl. 12,
figs. 1-9; Pl. 13, fig. 2
stilla 46
triquetra 48, 49, 65, Pl. 12, figs. ro-13; Pl. 13,
figs. I, 3
Progyrolepis 198
Proleptobos 270
Propalaeoniscus 197
Prosinotragus 266
Protobrama 72
Protocanites cf. lyoni 238
Protoeurynotus 176, 197
Protorux carolinae 276
Protoscaphirhynchus 199
Protragocerus 296
Psephurus 179, 189, 199
Pseudobeconia 200
Pseudogonatodus 146-7, 153, 159, 170, 184,
196, 201
macrolepis 147, 156, 157*, 158*, 159, 163
parvidens 147, 153, 154*, 155, 156*, 157
Pseudoscaphirhynchus 199
Psilichthys 199
Pteronisculus 183, 198
Pteroniscus 199
Ptilodictya 4
Ptcholepididae 195, 200
Ptycholepiformes 183-4, 195, 200
Ptycholepis 195, 200
pycnodonts 179
Pygopteridae 175, 182-3, 197
Pygopterus 197
Pyritocephalus 199

Quenstedt collection 8

RECENT
Pisces 81, 85, 90-1, 96, 104, 107-8, 189
Mammalia, Bovini 265, 271, 276
Redfieldia 200
tedfieldiids 181, 193
Redfieldiiformes 183, 190, I92—3, 200
Redfieldius 193
Reticulolepis 197
Rhabdolepididae 175, 181-3, 197

310 INDEX

Rhabdolepis 175, 197
Rhaedinichthyidae 175, 182-3, 197
Rhadinichthys 149, 182, 197
Rhadinoniscus 197

Rotalipora 72

Rubracea 34

Rushlandia 200

Sakamenichthys 200
Saurichthys 187
Saurichthyiformes 180, 183, 185, 187, 188, 199
Saurorhynchus 187, 199
brevirostris 187
Scanilepididae 183, 185, 186, 198
Scanilepis 198
Scaphirhynchus 179, 199
Sceletophorus 146, 164, 168, 186, 198
biserialis 168, 169*, 170*, 171*, Pls. 2 & 3
Schigospondylus 201
Schizurichthys 200
Schuleridea 39, 40*, 41-2
pentagonalis 43-4
Schuleridea (Aequacytheridea) 39, 40
Schuleridea (Eoschuleridea) 39-40
bathonica 41, 42, 64-5, Pl. 7, figs. 7-13; Pl. 8,
figs, I-11
Schuleridea (Schuleridea) 39
Schulerideidae 34
Schulerideinae 23
Scombroclupea 72
gaudri 72
SCOTLAND
Carboniferous (Pisces) 148, 153-4, 156-7,
159-60
Selenoportax 288
Semionotids 192
Semionotiformes 179
Sexual Dimorphism (Mammalia) 270-71
(Ostracoda) 50
SILURIAN
Wenlock Limestone, Polyzoa 12-13
Simatherium 296
kohllarseni 266, 296, 298
Sinkiangichthys 200
Sinotragus 266, 273, 288
Southcavea bajociana 124-5
grandis 124-5, 134
reticulata 116, 121, 124-6, 130
Spatulina 230, 238
Spatulina fauna 238-9
Spatulina longispina 230, 232, 233-34*, Pl. 6,
figs. I-4
nasifrons 232, 234
spatulata 217, 230, 231, 232*, Pl. 6, figs, 5 & 6
{spatulina] 234
Sphaerolepis 146, 164, 168, 170, 171, 186, 198
kounoviensis 164, 165*, 166*, 167*, Pl. r
Spjeldnaes, N. 4
Stegotrachelidae 181-3, 196

Stegotrachelus 181, 196

Stenaionotus 201

Stereolepidella 196

Stereolepis 196

Stichocentrus 81, 88, 89, 95-100, 102-5

Stichocentrus liratus 88, 89, 90*, g1*, 93*, 94,
95, 97, Pl. 3.

Stichopterus 199

Strepheoschema 197

Strepsiportax 297

Stromateus 197

Strongylosteus 199

Stylorhynchus 199

Styracopteridae 175, 182, 183, 197

Styracopterus 146, 197

Subholostei 179

Sunolepis 199

Syncerus 246, 265-80, 281*, 282-4, 286%,
287-90, 293-4, 296-8
caffer caffer 265, 279, 286, Pl. 4; Pl. 5.
caffer nanus 265, 270, 272

Systenocythere exilofasciata 116, I2I-5, 127-9,
134

Tarrasiidae 187, 199
Tarrasiiformes 183, 187, 199
Tarrasius 187, 189, 199
Taurotragus 277, 286
Tegeolepididae 181, 183, 196
Tegeolepis 181, 196
Teleopterina 199
Teleopterinidae 164, 187
Teleostei 178
teleosts 107, 108, 178-9, 183, 190
Terquemula 53-4
Tetracytheridea punctata 125-6, 132
Theriosynoecum 30
Tholonotus 198
Thomasinotus 201
Thoracopterus 190, 200
Thrissonotus 198
Thrissopater 107
Thrissops 72
Tompaichthys 176, 197
Trachelacanthus 198
Trachichthyidae 79-81, 86, 97, 103-7
trachichthyids 80-1, 83, 86-7, 97, 104
Trachichthyoides 74, 80, 95-6, 98, 102-3, 105-6
Trepostomata 56
Trias (Pisces) 176, 178-95
TRILOBITA
Carboniferous 209-10, 238-9
Carboniferous faunas 238-9
Ecology 209-10
Evolution 217
Preservation 210-11
Species variation 215
Tripelta 200
Trissolepidae 164

2?)

INDEX 311

Trissolepididae 163, 164, 176, 183, 185-6, 201 Urosthenes 186, 198
Trissolepis 164, 198 Urosthenidae 183, 185-6, 198
kounoviensis 164
Tselfatia 72 Vernoniella 34
Tubantia 86, 103, 105
Tungusichthys 194, 201 Wardichthys 197
Tungusichthyidae 194, 201 Waribole 217-9, see Cyrtosymbole
Turseodus 198 Watsonichthys 197
Typhloproetus 235, 236 Watsonulus 201
? Typhloproetus angustigenalis 236 Weberides 216
cephalispina 235, 236, Pl. 2, figs. 4, 6 Westollia 184, 194, 198
dietzi 236 Whiteichthys 197
koslowskii 236 Whitwell Oolite 115, 117, 119, 120, 127, 128,
microdiscus 235—6 132-4, 138-40
Upper Deltaic Series 133-7, 137 Xenesthes 198
Upper Estuarine Series 118, 133 Xiphactinus 72
Upper Limestone (M. Jurassic, Yorks.), 120,
126-7, 133-4, 138, 140 Yons Nab Beds 115, 118, 128-9, 133, 139-40
Urenchelys 72
Urolepis 198 Zeuchthiscus 200
a\bi\ if U/

ot
ats

“PRINTED IN GREAT
BY ADLARD &

re

eres
wegises